Research Report 2016 Image Captions Research Report 2016 (Covers the Period 2014-2016) Cover: Are Actually Induced Pluripotent Stem Cells in a Pluripotency Test

Research Report 2016 Image Captions Research Report 2016 (covers the period 2014-2016) Cover: are actually induced pluripotent stem cells in a pluripotency test. The cells are Max Delbrück Center for What looks like coral reefs and brightly glowing fishes in this picture Molecular Medicine in the Credit: Gizem Inak (Lab of Erich Wanker, Group of Alessandro Prigione) Helmholtz Association (MDC) stained with a fluorescent dye (Lin28-DAPI). Communications Department Cardiovascular and Metabolic Disease: Robert-Rössle-Str.10 13125 Berlin and lipid droplets white. These fireworks and balloons are fatty Credit:cells. Insulin Vanessa receptors Schmidt (Lab appear of Thomas red andWillnow) early endosomes green. Nuclei are blue Phone: +49-30-9406-0 Email: [email protected] Cancer Research: The colorful confetti of tumor cells which has been stained This Research Report Credit: Matthias Leisegang (formerly Lab of Wolfgang Uckert) is also available online: with different fluorescent dyes. www.mdc-berlin.de

Diseases of the Nervous System: Under the microscope the cross-section of Editors mouse skeletal muscle looks has a sponge-like structure. Muscle stem cells stained Vera Glaßer (Editor-in-chief) by an antibody against Pax7 (blue) are located under the basal lamina (green). Jana Schlütter Credit: Dominique Bröhl (formerly Lab of Carmen Birchmeier) Stephanie Sturm

BIMSB: Reminiscent of faraway galaxies: cancer cells found in malignant brain Editorial Board tumors produce circular RNA (circRNA), a type of molecule we know relatively Carmen Birchmeier little about. CircRNAs in yellow next to the pink cell bodies with blue nuclei. Norbert Hübner Credit: Christin Stottmeister (Lab of Nikolaus Rajewsky) Friedrich C. Luft Nikolaus Rajewsky ECRC: This image depicts human myoblasts and myotubes derived from a biopsy. Claus Scheidereit Myosin heavy chain (MyHC) protein of fused muscle cells is stained with a green Book Design Credit: Eric Metzler, PhD student (Lab of Simone Spuler) unicom-berlin.de fluorescent dye; nuclei appear in blue. Technology Platforms: The 3-D-mouse embryo is a reconstruction made up of Printing 360 individual images, the developing nervous system appears in orange. Laserline, Berlin 2017 Credit: Julian Heuberger (Lab of Walter Birchmeier), Anje Sporbert (Advanced Light Microscopy (ALM) technology platform) Research Report 2016 Table of Contents

II Table of Contents VIII Introduction by the Scientific Director Prof. Dr. Martin Lohse XI The MDC 1 Cardiovascular and Metabolic Disease Coordinator: Norbert Hübner

4 Molecular Biology of Peptide Hormones Michael Bader 8 Mathematical Cell Physiology Martin Falcke 10 Integrative Vascular Biology Laboratory Holger Gerhardt 14 Neuromuscular and Cardiovascular Cell Biology Michael Gotthardt 18 Genetics and Genomics of Cardiovascular Diseases Norbert Hübner 22 Mobile DNA Zsuzsanna Izsvák 26 Anchored Signalling Enno Klußmann 28 Molecular genetics of chronic inflammation and allergic disease Young-Ae Lee 32 Receptor signalling Martin Lohse 36 Hypertension Genetics, Regulators, and Environment Friedrich Luft 40 Molecular Muscle Physiology Ingo Morano 44 Ultrahigh Field Magnetic Resonance (UHF-MR) Thoralf Niendorf 48 Electrochemical signaling in development and disease Daniela Panáková 50 Molecular Epidemiology Tobias Pischon 54 microRNA and Molecular Mechanisms of Metabolic Diseases Matthew Poy 56 Molecular and Translational Kidney Research Kai Schmidt-Ott 58 Proteome Dynamics Matthias Selbach 62 Molecular and Cellular Basis of Embryonic Development Francesca M. Spagnoli 64 Genetics of Metabolic and Reproductive Disorders Mathias Treier 68 Molecular Cardiovascular Research Thomas Willnow

II MDC Research Report 2016 TABLE OF CONTENTS

Table of Contents

73 Cancer Research Coordinator: Claus Scheidereit

76 Signal Transduction in Development and Cancer Walter Birchmeier 80 Molecular Immunology and Gene Therapy Thomas Blankenstein 84 Structure, function and mechanism of membrane-associated GTPases Oliver Daumke 88 DNA Repair and Maintenance of Genome Stability Michela Di Virgilio 90 Molecular Oncology Gaetano Gargiulo 92 Macromolecular Structure and Interaction Udo Heinemann 96 Microenvironmental Regulation in Autoimmunity and Cancer Uta E. Höpken 98 IPS cell based disease modelling Ralf Kühn 100 Cell Differentiation and Tumorigenesis Achim Leutz 104 Molecular Immunotherapy Antonio Pezzutto 106 Immune Regulation and Cancer Klaus Rajewsky 110 Translational Tumorimmunology Armin Rehm 112 Spatio-temporal control of Rho GTPase signaling Oliver Rocks 114 Signal Transduction in Tumor Cells and Development Claus Scheidereit 118 Cancer Genetics and Cellular Stress Responses in pathogenesis and treatment of lymphatic malignancies Clemens A. Schmitt 120 Stem cell and macrophage biology Michael Sieweke 122 Intracellular Proteolysis Thomas Sommer 126 Molecular Cell Biology and Gene Therapy Wolfgang Uckert 128 Mathematical Modelling of Cellular Processes Jana Wolf

MDC Research Report 2016 III

TABLE OF CONTENTS

131 Diseases of the Nervous System Coordinator: Carmen Birchmeier-Kohler

134 Developmental Biology / Signal Transduction Carmen Birchmeier-Kohler 138 Research Focus Mina Gouti 140 Molecular Pathways in Cortical Development Annette Hammes 142 Physiology and Pathology of Ion Transport Thomas Jentsch 146 Cellular Neurosciences Helmut Kettenmann 150 Molecular Physiology of Somatic Sensation Gary Lewin 154 Neural Circuits and Behaviour James Poulet 156 Mitochondria and cell fate reprogramming Alessandro Prigione 158 Developmental Neurobiology Fritz G. Rathjen 162 Role of GPRIN1 on kainate-receptors function Dietmar Schmitz 164 Proteomics and Molecular Mechanisms of Neurodegenerative Diseases Erich Wanker 168 Development and Function of Neural Circuits Niccolò Zampieri 171 Berlin Institute for Medical Systems Biology Nikolaus Rajewsky

174 BIMSB Bioinformatics / Mathematical Modelling Altuna Akalin 176 Non-coding RNAs and mechanisms of cytoplasmic gene regulation Marina Chekulaeva 178 Quantitative developmental biology Jan Philipp Junker 180 Dynamics and functions of the enhancer landscape during zebrafish development and cardiogenesis Scott Lacadie 182 RNA Biology and Posttranscriptional Regulation Markus Landthaler 184 RNA Structure and Transcriptome Regulation Irmtraud M. Meyer 188 Evolutionary Modeling and Inference in Computational Systems Biology Benedikt Obermayer

IV MDC Research Report 2016 TABLE OF CONTENTS

190 Computational Regulatory Genomics Uwe Ohler 194 Epigenetic regulation and chromatin architecture Ana Pombo 198 Microscopy, Image Analysis & Modeling of Developing Organisms Stephan Preibisch 200 Systems Biology of Gene Regulatory Elements Nikolaus Rajewsky 204 Evolutionary and Cancer Genomics Roland Schwarz 206 Gene regulation and cell fate decision in C. elegans Baris Tursun 208 Systems biology of neural tissue differentiation Robert P. Zinzen 211 The Experimental and Clinical Research Center (ECRC) Friedrich C. Luft

216 Hypertension-induced Target-Organ Damage Dominik N. Müller Ralf Dechend 218 Blood Vessel Function and Target-Organ Damage / Registry Diabetic Nephropathy Maik Gollasch 220 Nephrology and Inflammatory Vascular Diseases Ralph Kettritz 222 Biology of Malignant Lymphomas Stephan Mathas Martin Janz 224 Outpatient Clinic for Neuroimmunology Friedemann Paul 226 Cardiac MRI Jeanette Schulz-Menger 228 Muscle Research Unit and Out-patient Clinic for Muscle Diseases Simone Spuler 230 Translational Oncology of Solid Tumors Ulrike Stein

MDC Research Report 2016 V

TABLE OF CONTENTS

233 Scientific Infrastructures & Technology Platforms Jutta Steinkötter

236 Advanced Light Microscopy Anje Sporbert 238 Biobank Tobias Pischon 240 Bioinformatics Platform, BIMSB Altuna Akalin 242 Chemical Biology Platform Jens Peter von Kries Marc Nazaré 244 Electron Microscopy Bettina Purfürst 246 Genomics, BIMSB Sascha Sauer 248 Magnetic Resonance (B.U.F.F.) Thoralf Niendorf 250 Metabolomics, BIH Platform at MDC Jennifer Kirwan 252 Light Microscopy Platform, BIMSB Andrew Woehler 254 Pathophysiology Arnd Heuser 256 Pluripotent Stem Cells Sebastian Diecke 258 Preparative Flowcytometry Hans-Peter Rahn 260 Proteomics Currently N.N. 262 Proteomics / Metabolomics, BIMSB Stefan Kempa 264 Transgenic Core Facility Ralf Kühn

VI MDC Research Report 2016 TABLE OF CONTENTS

267 Annex

268 Appointments and Recruitments 272 European Research Council (ERC) Grants 274 Awards 276 Selected Highlight Publications 278 Scientific Meetings and Conferences 280 Selected Seminars 2014 – 2016 284 Facts and Figures 286 Organizational Structure 290 Index

MDC Research Report 2016 VII

Introduction by the Scientific Director

Prof. Dr. Martin Lohse

Life in our labs deals with fundamental Within this program, we will concentrate questions of health and disease, while at - the same time focusing on small details – tives on system-wide diseases, namely: such as the intricate workings of an organ- •on cardiovascular five topics that disease reflect anddifferent metabolism, perspec ism, a cell and even an ion channel or recep- which inherently affect the whole orga

seem abstract. However, medicine cannot • developmental biology, which includes advancetor. At first without sight, uncovering such basic theresearch basic promay- anism; better understanding of inherited dis- cesses that underlie life. orders and how diseases may have their roots early in life, and furthers regenera- Exploring these mechanisms and transla • clinical applications has been our mission aretive keymedicine; players in the pathogenesis of fromting our the findings very beginning. as quickly As as wepossible celebrate into manyimmunology major diseases and inflammation, and can be which har- our 25th anniversary in 2017, we are re- - • gene regulation including systems bio ing a vision for the future. logynessed approaches, to develop which novel oftentherapies; form the flecting on our achievements and formulat Health research has traditionally been ap- • homeostasis and cell communication, whichfundamental explore basis how of cells diseases; and organs and regu- line with this, we have been active in cardio- late and maintain their functions, and vascularproached and in neuroscience an organ-specific research manner. as well In how they go wrong in disease. as cancer research. However, most severe diseases cannot be understood or treated MDC researchers cover the entire range of if we consider just the organ that seems to methods and levels of analysis – from ba- be primarily affected. A particularly good sic to translational science – in each topic. example is chronic heart failure, where we They collaborate closely with each other to develop diagnostics and preventive and in blood vessels and hormonal systems, in therapeutic approaches that embrace the kidneysfind not onlyand braina weak and heart, even but in alsothe immunechanges - system. And as a consequence, heart failure terdisciplinary work has long been a hall- therapies work only if we consider the or- markcomplexity of MDC of science. diseases. Indeed, highly in ganism as a whole and address all of these complex and interrelated mechanisms. At the frontiers of science, we must constantly adopt new technologies and develop At the Max Delbrück Center for Molecu- them ourselves. Two types of new develop- lar Medicine in the Helmholtz Association ments will be of particular importance for (MDC), we bring together a wide range of expertise and technology. To fully exploit imaging methods. Recent advances in light andthe MDC.electron The microscopy first example provide involves entirely new our activities in one common research pro- new perspectives on health and disease gramour scientific with a focus potential, on system-wide we aim to integrate and car- processes, by visualizing structures and diovascular disease mechanisms. processes that have so far been beyond our reach. Therefore, we aim to establish an Op-

logists and other experts can further this tical Imaging Center, where physicists, bio

field as innovators and expand the MDC’s scientific profile.

VIII MDC Research Report 2016 INTRODUCTION

The second example is single-cell biology, -

wella field as made disease possible mechanisms by recent to the technologi identity, functioncal advances. and structure It carries of human individual biology cells. as The MDC aims to be a major player in advancing these methods and in contributing to the Human Cell Atlas that would give us Photo: Bettina Flitner Bettina Flitner Photo: worldwide project that will rival the Hu- mana unique Genome ID card Project for allin scale.cell types. This is a

Biomedical research commonly crosses disciplinary and institutional boundaries. The scouting for ideas and projects, and sup- MDC has embraced these developments portsOur technology the process transfer of turning office these is actively ideas into reality. During the last two years, two forming new collaborations such as within new drugs have reached the market in Eu- by readily incorporating new fields and rope and the United States that represent strengthens our longstanding relationship entirely new therapeutic principles based withthe Berlin the Charité Institute – Universitätsmedizin of Health (BIH). ThisBer- on research at the MDC: the cancer drug Blincyto and a new treatment for bleed- for the next years will be to recruit outstan dinglin. In translational the context ofresearchers the BIH, our in mainstem goalcell of these recent successes and constantly research, regenerative medicine and per- striveing disorders, for new opportunities. VONVENDI. We are proud sonalized medicine. This research report showcases our work All these developments would not be pos- during the past three years. We hope that it sible without attracting the brightest minds makes you realize how our research relates to the MDC – graduate students, postdoc- fundamental principles to clinical prob- toral fellows as well as established re- lems, and how our strategies for the future searchers. We aim to support these scien- - tists in their individual development and to sues. We also hope that it conveys some of aid them at their respective levels on their thelink fascination past knowledge and excitement with open scientificthat charac is- - terize our work. lence. paths to scientific independence and excel MDC and the Campus Berlin-Buch offer ideal conditions for translation. These include close cooperation with clinical researchers at the Experimental and Clinical Research Center (ECRC), a common endeavor with Prof. Martin Lohse Leibniz-Forschungsinstitut für Molekulare Scientific Director of the MDC Pharmakologiethe Charité; close (FMP), collaboration including our with large the

options available at the BiotechPark Berlin- Buchcommon to start Screening new companies. Unit; as well as all the

MDC Research Report 2016 IX

12 MDC Research Report 2016 Bohm / MDC Katharina Photo: The MDC

MDC Research Report 2016 XI The MDC

Breaking down barriers in molecular medicine

Research at the MDC

ithout the knowledge of cells, it Delbrück, a native Berliner, often visited is impossible to know anything the biologist Nikolay Timofeev-Ressovsky W about the biology of life. The and his collaborator Karl Günther Zimmer. simplest creatures are single-celled organ- The researchers were producing mutations isms, and an adult human contains 100 tril- lion of these tiny units. All cells can produce their laboratory in Berlin-Buch. They dis- energy, are capable of dividing, and perform coveredin fruit flies that byincreasing exposing the them radiation to X-rays dose in

DNA data set that contains all the instruc- perspective on these observations, Del- tionsa very they specific need task.in order They to also function. each have a brückled to helpedmore mutations. make the conceptTaking a of physicist’s the gene

For more than 150 years now, researchers atoms arranged linearly along the chromo- in Berlin have been trying to understand tangible. If genes were complex clusters of points and cause a break that would split a noted that all cells come from cells. Some gene.somes, The then three the researchersX-rays could published strike random their 70how years cells interact.later, theoretical In 1858, Rudolfphysicists Virchow such as Niels Bohr became interested in discov- the molecular level, it signaled the start of modernstudy in genetics.1936. As the first work to consider

strangeering the when secret you of think life. Bohr’s about enthusiasmit: Humans When the Max Delbrück Center for Mo- produceproved contagious humans, cats for Maxproduce Delbrück: cats, “It’sand lecular Medicine in the Helmholtz Asso- ciation (MDC) was founded in 1992, it was be how physics or chemistry works. Atoms easy to decide whom to name it after. The corn produces corn. This doesn’t seem to German-American biophysicist and Nobel laureate stands for the courage to pursue don’t make the same atoms.” the ability to think beyond the borders of individualone’s own disciplines,ideas (no matter and the how open-mind unusual),- edness of an international perspective.

roughly 1,600 employees and guests – is one Twenty-five years later, the MDC – home to - entof the countries world’s and leading work biomedical together in research around 70institutes. research Its groups. scientists Their come aim from is to shed60 differ light on the complex causes of widespread diseases, such as cardiovascular disease, metabolic disease, cancer, and diseases of the nervous system. Their approaches include methods used in medical systems biology. Most of the researchers are biologists or chemists, while others are bioinformaticians, mathemati-

Photo: Katharina Bohm / MDC Katharina Photo: cians, physicians, or physicists.

XII MDC Research Report 2016 THE MDC

Diversity at the MDC: 47 percent of our scientists come from abroad

USA UK POLAND RUSSIA 2 1 2 3 TURKEY 2 GERMANY AMERICA other countries 2 1 FRANCE GREECE 5 CHINA

2 SPAIN % 3 53 EUROPE INDIA 2 other countries 2 PORTUGAL 4 5 ASIA other countries ITALY 8 BRASIL AFRICA AUSTRALIA 2 1 1

For these experts, molecular medicine is about combining the different perspec- be seen in just a few selected highlights. 16The scien extenttists ofwho the won scientific highly successsought-after can of categorizing patient groups according to ERC grants from the European Research theirtives symptoms, of physicians the andresearchers scientists. start Instead from Council are currently working at the MDC. the molecular biological processes occur- Four ERC grants were awarded in 2016 ring in the cells. Their methods range from alone. The same year saw the number of structural biology, to omics technologies publications rise to 463 (from 416 in 2015), and animal models – in which they can use - new gene-editing tools to simulate diseases ing journals with an impact factor above ten. even more realistically. Their experiments Somewith 88 60 (2015: percent 82) of appearingall studies inwere outstand co-au- show that a mechanism can have complete- thored by researchers from other countries. ly different effects on the body and, conversely, that seemingly identical symptoms approved that are based on work done at can have very different causes. theIn the MDC. past MDC two researchers years, two drugs are also have playing been a leading role in clinical studies and spin- This provides the researchers with funda- offs related to innovative T-cell therapies – mental insights into what happens in the an immunotherapy used to treat cancer. body when it is healthy and when something goes wrong (see the chapter “Discov- - in the Helmholtz Association, the MDC is As one of five health research centers with- eries” for several highlights from 2014– zation. The MDC offers excellent conditions They2016). should Above improve all, though, diagnoses these and findings thera- forpart both of Germany’s established largest and scientific up-and-coming organi pies,should and benefit help healthypatients people as soon avoid as possible. getting researchers. They can use state-of-the-art biomedical technology platforms (see the the very outset. - ill. This has been the MDC’s mission from chapter “Scientific Infrastructures & Tech MDC Research Report 2016 XIII THE MDC

- Genetic blueprints are written in the same tunity to develop their own technologies. language for every living thing. They even Researchersnology Platforms”) have access and to have strong the clinical oppor have many matching text blocks – the match - rate between humans and mice is 97.5 per- stitute of Health or the German Center for cent. However, the differences between Cardiovascularpartnerships, such Research as within (see the the Berlin chapter In species, between individuals, and between “From bedside to bench and back again), as health and illness do not just arise from well as a network of international collaborations. in which genes are switched on and off in anour organism, genes; they and are by also how defined the DNA by thestrands way - are folded. You can think of DNA as a kind search centers for the life sciences, is just of library. Different areas of the body are EU-LIFE,one example an allianceof many. of The 13 renowned spectrum reof activated to perform certain tasks, in much - cludes joint research groups such as the textbook off the shelf if you wanted to learn Helmholtz-Chinathe MDC’s international group collaborationson optogenetic in aboutthe same that way subject. as you Andwould every take nowa specific and evaluation of central cardiovascular con- again, mistakes can occur in the process.

group on stem cell and macrophage biology The methodological search for the control (2015–2018) and the MDC-INSERM nections is now leading researchers away have the opportunity to participate in in- from focusing on one gene and one protein. ternational(2012–2017). training In addition, programs PhD (see students chapter

disciplinesAt the Berlin are Institute investigating for Medical the interplay Systems “Promoting young researchers”). ofBiology genes (BIMSB), and networks scientists in fromcells, differentorgans, neighbors on campus such as the FMP and In Berlin, the MDC works closely with its- - Cure Cluster of Excellence, the MDC was searchand organisms. program. The MDC BIMSB has was been founded receiv- involvedthe city’s inuniversities. setting up Throughthe Einstein the NeuroCenter in 2008 as an extension of the MDC’s re- ing groups – which now number 14 – since in establishing the Einstein Center Digital 2012.ing fixed The institutional more details funding they discover for its workusing Futurefor Neurosciences. and the Einstein It has Centeralso played for Regen a role- high-throughput omics methods, the more - complex the picture becomes. We are still far from solving the great puzzle of life. erative Therapies. For the Integrative Re Humboldt-Universitätsearch Institute for the zu Life Berlin Sciences, (HU) which drew is part of the Institutional Strategy that - Change at the wasup under established the German at HU government’sand the MDC. ExcelAll of thislence helps Initiative, attract a world-class bridging professorship researchers MDC’s helm to the German capital – and as they come together here and engage with each other Since March 1, 2014, Dr. Heike Wolke has – creates an environment where new ideas served as Administrative Director of the can emerge. Max Delbrück Center, after previously

- Helmholtz Centre for Polar and Marine Re- nally presented the structure of DNA in working at the Alfred Wegener Institute,- 1953,When Max James Delbrück Watson was and disappointed. Francis Crick The fi burg pharmacologist Professor Martin idea that our genetic blueprint was a hered- searchLohse took (AWI) over in on Bremerhaven. April 1, 2016 The as WürzChair itary molecule of just four letters combined - to form a twisted rope ladder – the double rector of the MDC. He succeeded Professor helix – seemed a little dull to him. The com- Walterof the Board Rosenthal, of Directors who had and become Scientific Presi Di- plexity only gradually became clear as new dent of the Friedrich Schiller University technologies provided ever more detailed Jena on October 15, 2014. insights into cell interiors.

XIV MDC Research Report 2016 THE MDC

16 scientists currently working at the MDC have won highly sought-after ERC grants.

MDC Research Report 2016 XV With the newbuilding, XVI THE MDC will ariseonthe Buch Under construction: a newBIHlocation campus. know what people really need,” says Luft. The ECRC Luft. actsasaninterface between the says need,” really people what know here are participatingintrials.“We want to tient clinics.Many ofthepatientsthatcome outpa- university Charité’s the of twelve to long ago,butthefacilities are now home The originalhospitalwards may have closed in molecularmedicine. the only way to conducteffective research to-bench andbench-to-bedside approach is ToLuft. thisday, hebelieves thisbedside- ing. “This was a visionary concept,” explains ing itdirectly next to their laboratory build- Vogt setuptheclinicin1920s,position- terwoven. Neuroscientists Oskar andCécile research andpatientcare were closely in- its inception,thiswas aplacewhere basic The ECRC buildingbreathes history. From cal Research Center (ECRC). of what isnow theExperimentalandClini - es’ station,” explains Friedrich Luft, director nurs - the of part formed office myown and hind uswere oncefour-bed hospitalrooms, tine ofhospitallife. “Thelaboratories be- rouforget,the about for - moment, a brief could relax underthecolorfulawnings and Clinical translation attheMDC From bedsidetobenchandbackagain A Photo: kleyer.koblitz.letzel.freivogel.architekten Gesellschaft von Architekten mbH, Berlin once being a place whereplace a being once patients balconies, itiseasy to imaginethis s thesunshinesdown onthelong at justtheright time. the correct cell therapy orimmunotherapy administeringmeans rightthe ormolecule ceives specially tailored treatment. This re - patient each effects, side unbeneficial havingcases some in thus and to withdeal treating allpatientswiththesamemethod, cine –anapproach whereby instead of One mainfocal area ispersonalized medi- lated into patientcare. molecular biologyto bemore swiftly trans- allowsThis issues. cine discoveries inmade with physicians to address systems medi- searchers from theMDC are now working re - basic moreSignificantly box). (see MDC the partnershipbetween theCharité and strengthen considerably to served has BIH tive projects andjointappointments,the funding programs, large-scale collabora- its With Health. of Institute Berlin of tion in 2015thanks to theLaw ontheFounda- autonomous corporation under publiclaw The BIH was founded in 2013, and made an time for theBerlin Institute ofHealth(BIH). lition andreconstruction work began –this demo- additional 2017, May In added. was most recently, anewstem celllaboratory were converted into research facilities and, Charité andMDC.Theformer hospitalrooms MDC Research Report2016 THE MDC

Berlin Institute of Health (BIH)

- a number of translational activities, rang- ticularly important for the MDC: Holger The founding of the BIH in 2013 triggered GerhardtTwo BIH has appointments since 2014 havebeen a been research par

2015,ing from changes them at BIH the Twinningtop ushered Research in the as Professor of Cardiovascular Research at nextGrants phase. to the Erwin Clinical Bo ttinger Research from Unit. New In thegroup Charité. leader He at is the also MDC integrated and BIH into as wellthe activities of the German Center for Cardio- Sinai took up the post of Chief Executive Of- vascular Research (DZHK). The angiogen- York’s Icahn School of Medicine at Mount- esis specialist is investigating new ways to tor of the Helmholtz Association, assumed stop pathological blood vessel growth. Ralf responsibilityficer. Rolf Zettl, for until business then Managing and administra Direc- Kühn, an expert in genome engineering and tive duties. Axel Radlach Pries, Dean of the stem cell biology, came to Berlin as a guest Charité, und Martin Lohse, Chairman of the researcher at the MDC, where he leads the

the MDC Transgenic Core Facility. MDC’s Board of Directors, also joined the BIH working group on iPS cells and heads Executive Board of the BIH.

One example from oncology demonstrates they have secured funding of 3.1 million just how urgently this approach is required. euros until 2019. When a young child develops a neuroblastoma of the autonomic nervous system, This project is just one example of many. physicians are currently unable to tell par- Beyond the Charité outpatient clinics, - there are plans to set up additional pa- dren, spontaneous regression of the tumor tient cohorts aimed at gathering samples ents exactly what to expect. In some chil from a large number of patients who suf- which usually develops in the medulla (infer from the same illness. Scientists will neroccurs. part) In of others, the adrenal this malignant gland or next growth to the – analyze these samples with the help of spinal cord – must be surgically removed. high-throughput technologies, looking for Then there are those who require treat- individual alterations in genetic material, ment with a combination of aggressive che- enzymes, proteins or metabolites and com- motherapy, surgery and radiation, or per- paring these with samples from healthy haps even a bone marrow transplant. For subjects. This will allow researchers to all parents, a year of hope and fear follows as they wait to see if the cancer has been (biomarkers) that suggest the occurrence truly defeated. Only one in ten children sur- offilter abnormal out all processes the molecular in the determinants body. These vives a relapse. biomarkers not only identify where in the

Charité oncologist Angelika Eggert and can also provide information on the course MDC proteomics expert Matthias Selbach oforganism the disease something or indicate has gone where wrong; new theytar- have set out to analyze the different guises geted medication could be used. of this disease and to characterize each tumor by integrating all available data: These samples are stored in biobanks – one data on the genome, transcriptome, proof which was established at the Berlin-Buch teome and metabolome. The researchers campus in 2016. The centerpiece of the building is a room that houses two cooling tanks, both containing 2.5 million liquid developare hoping targeted that suchtherapy molecular strategies. profiling What samples stored in nitrogen. Blood and urine makeswill enable the tumor them toaggressive? define risk When groups do tinyand samples, for example, sit here in minus 160 fragments of the tumor tend to remain, in degrees Celsius, and are continually supple- spite of all the medication? Can a liquid mented over the course of long-term studies. This room has become a treasure chamber for researchers. soonbiopsy be be able used to tobetter predict combat metastases? the tumor. In Currently,addition, genetically27 scientists modified are involved T-cells may in One of these long-term studies is the Ger- this interdisciplinary research, working in man National Cohort, led by the Helmholtz - university and non-university partners. eight subprojects. Thanks to the BIH Col Association’s health centers together with laborative Research Grant “Terminate NB,” MDC Research Report 2016 XVII THE MDC

The nationwide study began its main phase - in 2014, and is documenting the health and vides a better approach for investigating lifestyle of 200,000 adults. For 20 years widespreadIt is clear that diseases, cooperative including research cardio pro- these individuals will undergo regular vascular diseases and their risk factors. medical examinations and answer ques- With the founding of the German Cen- tions on their living conditions and habits, ter for Cardiovascular Research (DZHK), in order to help scientists better research the emergence of widespread diseases. to better connecting this area of research MDC has made a significant contribution Six thousand of the participants will re- working together at seven locations in an ceive full-body magnetic resonance imag- effortin Germany. to improve A total the ofprevention, 28 institutions diagnosis are - and treatment of cardiovascular diseases. cians to see, for example, if there are fatty An intermediary review in 2014 attested depositsing (MRI) in scans. the liver, These which images can allow increase physi a that the DZHK was off to an impressive - start, displaying fantastic development - and excellent preclinical research. Review- ordinatingperson’s risk the of Berlin-Brandenburg diabetes. MDC is responsi Cluster ers were very impressed with the central andble for the all decentralized these MRI scans, storage as well of biological as for co infrastructure for clinical studies that the samples. Another 10,000 participants are DZHK had set up in such a short period of time. study center. to be recruited at the MDC’s Berlin-Nord Photo: Horst Krüger / MDC Horst Krüger Photo:

XVIII MDC Research Report 2016 THE MDC

The MDC is not only represented on the -

involvedDZHK’s Board in the of DZHK. Directors These by Thomasscientists Som are responsiblemer; there are for altogether organizing 58 projectMDC scientists groups such as “Genetics of Congenital Heart Dis-

can exchange knowledge and ideas and makeease” theirand “Nutrition,” expertise – particularlyin which specialists in terms of technology and infrastructure – avail-

example, offers expertise on the generation andable phenotypingto DZHK’s partners. of transgenic Michael rats, Bader, while for - ics and proteomics. Norbert Hübner’s specialist field is genom Structures like the DZHK offer a framework to facilitate cooperative research. However, some projects require a great deal of pa- tience on the side of the researcher before the molecular cause of a disease is found. One example of such a case is the unusual story of a Turkish family who lived from a village by the Black Sea. Whenever a child Photo: Maj-Britt Hansen Photo: was born into this family, the first thing the Okan Toka, a former member of Fred Luft's team at the MDC, normallyrelatives wouldand the do washands examine remained the child’ssmall, speaks to one of his Turkish patients. thefingers. child If would they appeared be expected to notto die be young.growing

family was causing hereditary hypertension,It emerged with recorded that a values single of gene up to within 270/160 the severalmodified other the families same gene with in similar animals symptoms to see if mmHg. Researchers led by Friedrich Luft camethey tooforward. develop The hypertension. scientists decoded In addition, their hoped they could identify this suspicious genetic material, too. And rather than only gene. Little did they know this family would looking at the gene sequence, they also want be the focus of a study that was to last over to work together with structural biologists - to work out the three-dimensional form of searchers announced in Nature Genetics the protein and uncover the mechanism that that20 years. they Ithad was found only the in 2015guilty that suspect: the re a leads to the problems. mutated form of the enzyme PDE3A. This enzyme not only regulates blood pressure, “Sometimes science is like looking for a black but also bone growth.

Luft believed at this point that the genetic bitcat byin abit, dark he room,” says, that reflects we are Luft. starting “And often to real we- sequence had been cracked and the prob- izedon’t the even many know unknowns if the cat that is there.”exist in It scienceis only lem was solved. “But projects like these that have so far been completely overlooked.

don’t go away,” he says. His group has now

MDC Research Report 2016 XIX THE MDC

Discoveries

Selected Research Highlights from MDC Labs

Pressure relief valve discovered in Why the heart fails the cell membrane Defects in splicing are implicated in a What causes cells to shrink? This question number of serious genetic diseases, like has vexed biologists for decades. But now, the heart disease dilated cardiomyopathy after four years of research, a team of scien (DCM). This condition is marked by an entists led by Thomas Jentsch have found the larged, weakened heart and is responsible answer. Together with the Screening Unit, for about a third of deaths from congestive which is jointly run by the FMP and the heart failure. MDC scientists have shown that mutations in the protein RBM20 (RNA unknown gene as an essential constitu- binding motif protein 20) probably contrib- entMDC, of thethe scientistsvolume-regulated identified anion a previously channel ute to DCM by affecting the splicing of cru- (VRAC). The gene is called leucine-rich re- cial molecules in the heart. They provided a detailed account of the operation of RBM20, by-one approach in a large-scale cell cul- which is preferentially present in the heart peat-containing 8A (LRRC8A). Using a one- and orchestrates a large number of target silenced approximately 20,000 human ture experiment, Jentsch’s team transiently revealed that it was located in introns, from the researchers also investigated which of wheremolecules. it tells Studying the spliceosome RBM20’s target to remove pattern a thegenes. genes In anare automated responsible screening for the swelling- process nearby exon, thus shortening the molecule. - But the defective RBM20 version fails to brane. The approximately 130,000 time-de- remove the exon and instead creates RNA pendentactivated measurementschloride flux across were the statistically cell mem with extra segments, resulting in proteins analyzed with help from the Bioinformatics - Group led by Miguel Andrade (until April sential heart protein, this creates a molecu- larthat spring are too that long. is too In theslack case and of muscles titin, an that es

2014 at the MDC; now at the University of heart is forced to work harder and becomes Mainz). Jentsch’s team went on to show that enlarged.don’t contract Further efficiently. study will Ultimately, show if these the VossLRRC8A FK, Ullrich requires F, M u othernch J, Lazarow members K, ofLutter the D,LRRC8 Mah geneN, Andrade-Navarro family to form VRAC. MA, von Kries Maatz H, Jens M, Liss M, Schafer S, Heinig M, JP, Stauber T, Jentsch TJ. Science 344 (6184): Kirchnerfindings areM, clinicallyAdami E, relevant.Rintisch C, Dauksaite 634–638 (9 May 2014). V, Radke MH, Selbach M, Barton PJR, Cook SA, Rajewsky N, Gotthardt M, Landthaler M, Hubner N. Journal of Clinical Investigation 124 (8): 3419–3430 (1 August 2014).

XX MDC Research Report 2016 THE MDC

DNA transcription: Start region Making genome editing more shown to be unidirectional efficient To enable enzymes to grab ahold and start With the help of molecular tricks, research- transcribing a gene at the right place, the DNA contains recognition sequences, so- - called promoters, which are located imme- ers at BIH and the MDC have managed to diately upstream. Ever since researchers significantly improve the efficiency of ge began using high-throughput sequencing allowsnome editingresearchers using to CRISPR-Cas9create DNA double- “gene technology to precisely investigate gene ex- scissors.” The CRISPR-Cas9 technology pression patterns, they have believed that a the genome. These are repaired in cells by large percentage of promoters are not uni- usingstrand one breaks of two at naturally specific locationsoccurring withinmech- directional. But researchers at the MDC and anisms – either by homologous recombi- the University of California in San Diego have nation, which allows researchers to make - very precise changes to genome sequences, ies show that a central part of promoters, or by non-homologous end joining, which thenow core discovered promoter, that is this intrinsically isn’t so. Their unidirec stud- tional, and that transcripts of the opposing frequently deletes DNA sequences. The DNA strand arise from their own core pro- is more efficient but less precise, since it moters. Using high-throughput experiments and various analytical procedures, the re- non-homologousBIH and MDC researchers end joining, have which now makes used searchers determined that in fact about 40 nature’s bag of tricks to suppress cellular- percent of the genes have two opposite core mologous recombination and thus of pre- promoters at variable distances. Through it possible to increase the efficiency of ho the copy of the opposing strand a long non- Cas9 by up to eightfold. coding RNA arises (lncRNA), i.e. a tran- Chucise VT, genetic Weber modifications T, Wefers B, Wurst using W, CRISPR- Sander script, which is not translated into a protein S, Rajewsky K, Kühn R. Nat Biotechnol (24 and whose function has yet to be elucidated. Mar 2015). Since reverse-directed core promoters are common but not universal, the researchers presume that these at least partially help to regulate gene transcription. Duttke SH, Lacadie SA, Ibrahim MM, Glass CK, Corcoran DL, Benner C, Heinz S, Kadona- ga JT, Ohler U. Molecular Cell (Jan 2015). Photo: Horst Krüger / MDC Horst Krüger Photo:

MDC Research Report 2016 XXI THE MDC

What hypertension and Kann M, Schuster H, Chitayat D, Bialer MG, brachydactyly have in common Wienker TF, Ott J, Rittscher K, Liehr T, Jordan After more than 20 years of research, a team J, Plessis G, Tank J, Mai K, Naraghi R, Hodge R, of scientists led by Friedrich Luft at the Ex- Hopp M, Hattenbach LO, Busjahn A, Rauch A, perimental and Clinical Research Center Vandeput F, Gong M, Rüschendorf F, Hübner (ECRC), a joint cooperation between the N, Haller H, Mundlos S, Bilginturan N, Movse- MDC and the Charité, have found the cause sian MA, Klussmann E, Toka O, Bähring S. of a rare disease suffered by a Turkish fam- Nat Genet 47(6): 647–53 (Jun 2015). ily. The affected individuals have heredi- Fighting tumors with T cells and are small in stature. They typically die The consortium “Targeting somatic muta- tary hypertension, unusually short fingers tions in human cancer by T cell receptor

before the age of fifty if their hypertension the development of cancer immunotherapy. eachgoes other,untreated. causes This six disease, different which point afflicts muta- Thegene researcherstherapy” has havemade successfullya major advance modi in- tionsthis and in the five PDE3A other familiesgene. These not mutations related to lead to high blood pressure and shortened bones of the extremities, particularly the werefied immune able to analyze cells (T thecells) antigens to recognize and clear and- metacarpal and metatarsal bones. The sci- specifically target cancer cells in mice. They - cell targets by using a humanized mouse delian hypertension form not based on salt model.ly distinguish This animal between model “good” can and be used“bad” toT reabsorptionentists have thus in the discovered kidneys, thebut firstinstead Men is test the therapeutic suitability of T cell re- more directly related to the structure and ceptors and antigens, which is an important function of the vascular wall. prerequisite for the development of clinical Maass PG, Aydin A, Luft FC, Schächterle C, applications. Weise A, Stricker S, Lindschau C, Vaegler Targeting human melanoma neoantigens by T M, Qadri F, Toka HR, Schulz H, Krawitz PM, cell receptor gene therapy. Leisegang M, Kam- Parkhomchuk D, Hecht J, Hollfinger I, We- mertöns T, Uckert W, Blankenstein T. Journal feld-Neuenfeld Y, Bartels-Klein E, Mühl A, of Clinical Investigation (25 Jan 2016).

Publications 2014-2016

Total > Impact Factor 10 Ø Impact Factor % in Tier 10*

2016 490 77 6,39

2015 432 6.32 38 %

2014 402 6867 6.54 33 %

40 %

* Tier 10: Number of publications which appeared in the top-tier journals of their respective fields (top 10 %).

XXII MDC Research Report 2016 THE MDC

Focus on international collaboration: ISRAEL

What happens within a cell when it has to process information and make decisions? is a powerhouse in competitive biomedi- This question is at the core of Nir Fried- calIsrael research. might notMany be principal a large country, investigators but it - at the MDC have forged collaborations with tems biologist was granted a Humboldt Researchman’s research. Award In that 2015, enabled the him Israeli to con sys- visits, they are actively contributing to the tinue his work in Berlin. Nir Friedman (He- Helmholtz-IsraeliIsraeli scientists. ApartCooperation from short-term in Per- brew University in Jerusalem) and Nikolaus sonalized Medicine through two projects: - • “Natural RNA circles: An emerging group tigating new regulatory signals at the RNA of biomarkers for personalized cancer levelRajewsky – a (BIMSBcollaboration / MDC) fostered are jointly through inves - the SignGene Research School. Moreover, sky (MDC) with Yosef Yarden (Weizmann Prof. Yehudit Bergman, also from the He- medicine” (2015-2017), Nikolaus Rajew brew University in Jerusalem, received in • - 2015 the Helmholtz International Fellow Institute of Science) Award, an honor aimed at fostering her col- (2015-2017),“Network-based Erich identification Wanker (MDC) of person with laboration with MDC scientists and within Rodedalized targetsSharan for(Tel Huntington’s Aviv University) disease” the SignGene program. Two other projects have been funded by the SignGene – the German-Israeli Helm- German-Israeli Foundation: holtz Research School “Frontiers in Cell • - Signaling & Gene Regulation” – is a joint tion of L-type Ca channels by protein endeavor of the MDC, Humboldt University “Mechanisms of isoform-specific regula and Charité in Berlin, the Technion in Haifa Sharon Weiss (Tel Aviv University) with and the Hebrew University in Jerusalem. Hannelorekinases” (2013-2016), Haase and Enno Nathan Klussmann Dascal and Since 2013, 37 principal investigators and (MDC) 20 PhD students have been involved, re- • “Molecular dissection of the epithelial- sulting in 21 papers so far. For their thesis mesenchymal cross-talk in establish- projects, students have a main supervisor at their home institute and a co-mentor Francesca Spagnoli (MDC) with Limor Landsmaning final pancreas (Tel Aviv shape” University) (2015-2018), last up to six months. Joint events such as Jan Philipp Junker of the MDC and Erez symposia,abroad. Visits winter in schools the co-mentor’s and PhD retreats lab can are intended to offer a space for the two mapsLevanon of changes of Bar-Ilan in certain University gene in transcripts. Israel are each other. Among SignGene faculty mem- Theco-developing ARCHES Prize the first-ever, conferred high-resolution in 2016 and bersscientific is Nobel cultures laureate to meet Aaron and benefitCiechanover from endowed with a total of 200,000 euros over a from the Technion. - bine their expertise and tackle this challeng- ingperiod task. of five years, will enable them to com

MDC Research Report 2016 XXIII THE MDC

Development aid

Technology transfer at the MDC

veryone knows about omega-3 fatty - ogy transfer, but not the only one. MDC re- E as salmon and are believed to protect searchersThis is the made MDC’s a key latest contribution coup in technol to two againstacids; heart they disease. are found But innot oily all fishpatients such drugs that came on the market in 2015:

towill be benefit converted from into a change active ofmetabolites diet. This in is recombinantVONVENDI (Baxalta treatment Inc., for now: adults Shire) living and because the omega-3 fatty acids first have withBlincyto von (Amgen).Willebrand VONVENDI disease (VWD), is the an only in- at the Max Delbrück Center for Molecular herited bleeding disorder. Patients lack a Medicinethe body, (MDC) as Wolf-Hagen has discovered. Schunck’s team protein – known as von Willebrand factor – that is crucial for normal blood clotting. The biochemist and his colleagues are now - fore helps control bleeding episodes. VONVENDI provides a substitute and there todeveloping bypass the a substance intermediate called step OMT-28 and act at Blincyto, on the other hand, is an immuno- directlyOMEICOS, on athe MDC receptors spin-off. of It heart is supposed muscle therapy against a very aggressive form of cells. The researchers hope that this will blood cancer (B-cell acute lymphoblastic leukemia). The bone marrow of these pa- risk of heart failure and strokes. The sub- tients produces too many immature white stanceprevent has atrial already fibrillation proved andeffective reduce in ani the- blood cells. Rather than maturing into func- mal models. A phase 1 study on healthy hu- tional cells, these lymphoblasts rapidly re- man volunteers began in March 2017. produce, suppressing normal blood forma-

to destroy this cancer. tion. The drug enlists the body’s own T-cells The royalties from these successes are now helping to translate other basic research

competitive funding formats, as well as expandideas into advisory applications services and for find start-ups new pre- in

transferconsultation is a corewith missionBerlin Institute of the Center. of Health At present,(BIH). The 20 MDC research law states groups that are technology working on new medical therapies or diagnostics intended for commercialization. Each project receives comprehensive support.

early stage—without intellectual property rights,The first nothing step is happens to secure in biomedicalthe patent atmar an-

technologyketing. In the transfer last three company years alone, based 24 on new the Buchpatents campus have beenwhich filed. supports Ascenion numerous and a institutions in the life sciences sector, plays a key role in this process. The technology - ine the market potential of the discovery andtransfer bring office in external and Ascenion consultants jointly to take exam the project to the next step, or explore opportunities for cooperation with other institutes and industry. They determine whether a spin-off would be appropriate or whether

Photo: blümchen36 / photocase.com blümchen36 Photo: the drug should be out-licensed soon.

XXIV MDC Research Report 2016 THE MDC Photo: Monique Wuestenhagen Monique Wuestenhagen Photo:

Captain T Cell, a MDC spin-off, won against 400 international start-ups. With the MDC’s internal funding programs can give- 100,000 pounds prize money, the team will sibleprojects by providing a boost. Since 150,000 2008, euros for example,annually continue to develop a cancer immunothera- for“PreGo equipment Bio” has and been personnel making foradvances up to three pos py that involves a platform to produce per- years. Project teams can also receive coach- ing and advice from industry experts. To complement these opportunities, since the sonalized, cancer-specific T-cell receptors.- cer,With and the systematically help of these receptors, destroy it. the body’s has been helping early-stage projects – own T-cells should be able to find the can whosespring ofpotential 2016 the still funding needs program to be evaluated “Boost” Center Cell Engineering Lab (MD-CEL), a provides access to valuable expertise, for ex- platformIn the spring for transposon-based of 2017, the Max cell Delbrück engi- ample,– get off from the ground. clinicians In addition,at the Experimental the campus neering sponsored by the Helmholtz Asso- and Clinical Research Center (see chapter ciation, began to work closely with science on ECRC) or from the Screening Unit operated jointly with the Leibniz-Forschungsin- stitut für Molekulare Pharmakologie (FMP) teamsand industry. at the MDC. It has its roots in Wolfgang and the Chemical Biology Platform. Uckert’s and Zsuzsanna Izsvak’s research

MDC researchers have no shortage of ideas. the basic research to market approval is Berlin Cures, founded as a spin-off of the The road through the “valley of death” from MDC and the Charité in 2014, is develop- - ing a drug called Aptamer BC007 to treat ratory,long and a substancestony. In the that case could of a be promising used to - discovery in Wolf-Hagen Schunck’s labo rently being tested on healthy volunteers. ten years. And the risk of failure is still 90 the causes of chronic heart failure. It is cur percent.treat atrial Schunck fibrillation, is well thisaware process of this. took But right now he is especially proud to be start-

start-upAlso in 2016,competition for the in first the time life a sciences. German team won at Onestart – the world’s largest ing tests on human volunteers: “That’s a dream come true.” MDC Research Report 2016 XXV THE MDC

Promoting young researchers

Training at the MDC

or Douaa AS Mugahid, the road to Over 360 PhD students and around 230 Boston went through Berlin. Between postdocs are currently engaged in research F 2011 and 2015, the Egyptian research- at the MDC. More than half of them come er worked as a PhD student in Michael Gott from outside of Germany, with 42 different countries represented on campus. The stu- of interests, open-mindedness, and helpful dents are equally varied in terms of subject: andhardt’s diplomatic lab at the MDC.nature, With she her immediately wide range Their backgrounds cover everything from caught the attention of her colleagues. They molecular biology, biochemistry, and phys- found her to be just as capable of discussing ics, to computer science, and medicine. problems as she was of talking about scien- With the Helmholtz Graduate School in student representative and an active mem- Molecular Cell Biology, doctoral students bertific oftheories. the Helmholtz Although Juniors she was network, both a PhDshe still found time to help refugees. that the MDC set up in collaboration with Humboldt-Universitätbenefit from a structured zu PhDBerlin, program Freie Universität Berlin, and Charité – Univer- that much different from the challenges every“I’m graduate not sure studentif the challenges faces: getting I faced the were ex- lab-based projects, students can acquaint periments to work, dealing with large data themselvessitätsmedizin with Berlin. new technologies In addition in to work their- sets and making sense of them within the shops, attend seminars and lectures, and network with each other at events such as - retreats. ingcontext a few of different the biological lines of questions investigation I was in addressing,” says Mugahid. “Also, pursu thematically focused programs – such as parallel meant that I had to learn the art- theSome Helmholtz students alsoResearch participate Schools in the Transla MDC’s- hidof multitasking.” began a postdoctoral Her next fellowship step proves in justthe tional Cardiovascular and Metabolic Medi- Departmenthow well she of succeeded:Systems Biology In 2015, at Harvard Muga cine (TransCard), Frontiers in Cell Signal- Medical School. ing and Gene Regulation (SignGene) – or

Molecular biology in schools

Biomedical knowledge is growing ex- and by the Brandenburg Ministry of Edu- ponentially. Neither textbooks nor care- cation, Youth and Sport. The Lab Meets fully developed curriculum frameworks Teacher courses also produce teaching can keep pace with the latest develop- materials that are freely available as open ments and methods. This is why the MDC educational resources. launched the Lab Meets Teacher training program (Labor trifft Lehrer, LTL) in 2011. The MDC has also spent the past 15 years The program brings teachers into the lab working directly with young people in and allows them to spend a day immersing schools to get them excited about science. themselves in the world of research. They conduct experiments, listen to talks, and Science Learning Lab, a learning center discuss the latest topics with researchers. thatIts efforts gives over here 12,000 include school running pupils the an Life in- All courses are recognized as advanced teacher training by the Berlin Senate De- also sends MDC lecturers into schools to partment for Education, Youth and Science, givesight talks, into biomedicaland does much research more each besides. year. It

XXVI MDC Research Report 2016 THE MDC Photo: Harry Schnitger / MDC HarryPhoto: Schnitger

in the MDC-NYU PhD Exchange Program in Systems Biology, an international pro- from research, academia, or industry, and a - varietyIn the of program, workshops an help experienced postdocs to mentor identify and develop their professional goals gram run by the Berlin Institute for Medi- and personal skills. They can also apply for dentscal Systems and its Biology. Clinician The ScientistBerlin Institute Program of various Helmholtz programs. The MDC is a helpHealth’s lay PhDthe foundationsscholarships for for collaboration medical stu beyond the borders of basic research and network designed to support women on the medicine. The schemes give physicians the pathpartner to leadership in the LIBRA positions project, in science. an EU-wide space they need to do research during their training. Since 2016, the MDC Postdoctoral Associa- tion has been holding Networking Lunch- time Seminars, which are informal and in- together many disciplines and gives young teractive. During the events, young group researchers“The MDC is real a wonderful opportunities center. toIt bringswork leaders and former postdocs who now - work in industry talk about their career paths. The annual MDC Career Day allows workingindependently,” in a center says of Martin this size Lohse, means scien stu- PhD students and postdocs to make initial tific director of the MDC. However, he adds, contact with companies. And in the Career with it and then quickly work out what they wantdents to first achieve need during to acquaint their time themselves here. This time at the MDC and on what came after- applies to postdocs in particular. wards.Pathway lectures, alumni reflect on their

- dates go on to earn the coveted title of pro- Internationally, just 4.4 percent of candi switching“I never wantedto industry, to be but an then academic she real for- prepare themselves for the leap into aca- izedlife,” that says science Mugahid. still She offered often many considering oppor- demicfessor. independence.It is therefore However,crucial that this students is not a tunities to grow professionally. Her advice: leap everyone wants to take. The MDC Men- “Learn to trust yourself and your instincts. toring Program is designed to help post- Many breakthroughs are made by young docs make a well-informed decision about people who have not yet become sceptics their career.

by listening too much to others in the field.” MDC Research Report 2016 XXVII EDITORIAL REMARKS

XII MDC Research Report 2014 Cardiovascular and Metabolic Disease

Coordinator: Norbert Hübner

MDC Research Report 2016 1 Cardiovascular and Metabolic Disease

Coordinator: Norbert Hübner

ardiovascular and metabolic dis- pies to target cardiovascular and metabolic eases are the most common cause of disorders. The Cardiovascular and Meta- C death worldwide, increasing burden bolic Diseases program has revealed some on society. Most cases of cardiovascular dis- considerable key molecules and pathways eases are likely the result of contributions involved in major processes of the cardio- from multiple genes, and apparently also vascular and metabolic systems. For exam- complex epigenetic mechanisms may lead ple, the group of Michael Gotthardt studies to cardiovascular disease. Getting an exper- the role of titin in the regulation of con- imental handle on these topics and improv- tractility in cardiac, skeletal, and smooth ing therapeutics still is the major challenge muscle. Using their animal models they for the next decade. An increasing number - of animal models – particularly strains of mere assembly, to link titin to adrenergic rats and mice – are available and neces- signalwere able transduction, to elucidate lateral titin’s growth role in ofsarco the - sarcomere, and the hypertrophy response. Their work emphasized the role of post- issary hampered to convert by findingsspecial problems into relevant including clini transcriptional regulation in the etiology cal tools. Research in this field sometimes of myocardial pathology. RBM20 interacts extrapolating from a very small number of not only with titin but potentially in a net- cases,finding and patients understanding with a particular complex profile, inter- work of 30 genes previously linked to car- actions within cells and cell types. We are diomyophathy, ion-homeostasis, and sarco- dealing with these challenges while con- mere biology. The group of Thomas Jentsch stantly developing new types of modeling has a long standing interest in investigat- and through a thorough integration of bio- ing the volume-regulated anion channel informatics with clinical and experimental (VRAC) channel and the relevant subunits approaches. The molecular analysis of car- ranging from biophysics, structure function, diovascular and metabolic functions at the to cell biology, mouse models and medicine cellular level is a prerequisite for a mecha- with implications for cardiovascular out- nistic understanding and forms a basis for the rational implementation of novel thera- heteromers as an essential component of VRAC.comes. MetabolicRecently, theydisturbances could identify are emergLRRC8- ing as major risk factors not only of cardiovascular but also neurodegenerative and oncological diseases. However, the molecular mechanisms that link the control of metabolism to other biological systems remain poorly understood. The group of Thomas Willnow is studying endocytic receptors to uncover their roles in normal physiological processes and in human disease. They demonstrated that LRP2, a member of the LDL receptor gene family, balances signaling of Endothelial axial polarity measured by the position of the golgi to the the morphogen sonic hedgehog in the de- nucleus of each endothelial cells (red, left panel) points against the veloping retina, and that lack of this activity direction of blood flow (compare to arrows signifying flow velocity, results in severe overgrowth of the eyes in middle panel). The axial polarity against flow is a response of endothe- mouse models and patients. Organogenesis lial cells to wall shear stress (right panel). Black arrowhead points to involves multiple sequential events such as region of low flow, low wall shear and consequently divergent polarity leading to migration of cells out of the low flow and morphogenesis, which ultimately re- segment and therefore branch regression. sultfate in specification, the development growth, of a differentiation,functional ma- Image courtesy of Claudio Franco and Miguel Bernabeu (Franco et al, 2015 PLoS Biology) / ture organ. This step-wise process relies on AG Holger Gerhardt a coordinated and highly complex interplay

2 MDC Research Report 2015 CARDIOVASCULAR AND METABOLIC DISEASE

of signaling events and transcriptional net- - course. The group of Michael Bader has also tal processes or a loss of their coordination loci influencing the march’s unfortunate- areworks. common Importantly, causes ofaberrant congenital developmen defects of ology and stem cell research. So far, they targeteda strong emphasisgenes for inthe the insulin field ofreceptor, rat embry the cardiac biology Ca2+ ions play pivotal roles (pro)renin receptor, apolipoprotein E, and thein a cardiovasculature, multitude of cellular liver, processes, or pancreas. rang In- ing from excitation-contraction coupling has also been established and successfully to the regulation of hormone secretion and usedthe LDL to ablate receptor. genes CRISPR/Cas in rats. Understanding technology gene expression. The research of Daniela diseases at the molecular and cellular level Pánáková aims to uncover the molecular is an important step in elucidating mecha- mechanisms that mediate the crosstalk be- nisms that ultimately have their effects at tween calcium signaling and other physi- the level of organs and organisms. Follow- ological stimuli with major morphogenetic ing the idea of translational medicine the pathways such as Wnt signaling during car- MDC cooperates with the Charité through diovascular development as well as in dis- the Experimental and Clinical Research Center (ECRC) on several topics. The group diabetes and various types of cardiovascu- of Thoralf Niendorf concentrates on the larease lesions states will in zebrafish.require a deeper Learning knowledge to treat development of MR technology enabling - new ways of mapping/probing morphol- creatic progenitor cells. Deciphering the ogy, function, physiology and metabolism of the genetic profile of myocardial or pan including cardiac and metabolic imaging. particular cell types of the heart, liver, and We further strengthened our collaborations pancreas,molecular willsignatures allow us of to “stem isolate cell-ness” such cells of within the German Center for Cardiovascu- for therapeutic purposes and possibly to in- - struct stem cells or progenitor cells to activate appropriate differentiation programs. notlar Researchonly by joint (DZHK) recruitments and the Berlin(like Holger Insti Using latest NGS technology, Francesca Gerhardt)tute of Health but also (BIH) by whichthe growing is highlighted number Spagnoli aims at elucidating how distinct cell types like liver and pancreas can arise to understand the regulatory genetic net- of scientific projects. The challenge remains organogenesis fate will contribute knowl- metabolic disease risk, to link molecular edgefrom commonnecessary progenitors. to close the Her gap findings between on networksworks that with influence physiological cardiovascular data, and and to basic model research and applied clinics. identify key regulatory nodes that can be The vascular biologist Holger Gerhardt, targeted therapeutically to reduce the burden of cardiovascular and metabolic dis- how developmental vessel regression is eases in patients. drivenwho joined by cell the migration, program inwhere 2015, cells identified are at- tracted to vessel segments that have higher Scientific Management: Kathrin Saar

individualsblood flow, activilyplays an moving important out of role segments in the developmentwith poor flow. and Genetic progression variability of cardiovas among- cular disease. Genetic and genomic studies in human populations are investigated by Young-Ae Lee to understand the etiology of

by other forms of common human cardio- vascularchronic inflammationand metabolic that diseases. is often She shared con-

ducted the first GWAS for the atopic march and showed that there are specific genetic MDC Research Report 2015 3

CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Michael Bader

Molecular Biology of Peptide Hormones

The group focuses on the molecular biology teracts the classical RAS effects in cardio- and function of hormone systems involved vascular and muscular diseases and in fat homeostasis (Fig. 1). Furthermore, these in cardiovascular regulation. The physiologi- animals showed that angiotensin(1-7) and cal functions of these systems are analyzed Mas are important for insulin sensitiv- by the production and analysis of transgenic ity and the pathogenesis of metabolic syn- and gene-targeted animal models. Among rat model with inducible diabetes mellitus the hormones studied, serotonin is of spe- typedrome. 2 a In novel a newly oral establishedformulation transgenic of angio- cial interest, since it is not only involved in tensin(1-7) was shown to exert potent an- vascular homeostasis and other peripheral tidiabetic actions revealing the therapeutic - functions, but also serves as potent and dition, ACE2 was found to be pivotal for se- multifunctional neurotransmitter in the brain. rotoninpotential generation of the protective in brain RAS and axis. periphery In ad In addition, the group is interested in em- by its function as part of the uptake system bryology and stem cell research, in particu- for tryptophan in the gut. This may also be the reason for the blunted exercise-induced lar applying these fields to the rat. hippocampal neurogenesis which we ob-

servedKallikrein-kinin in ACE2-deficient system mice. Renin-angiotensin system and chemokines The renin-angiotensin system (RAS) is of The kallikrein-kinin system (KKS) is an im- central importance in blood pressure regu- portant hormone system for cardiovascular lation and in the initiation of target organ regulation also mostly counteracting the effects of the RAS. As models for the func- generating systems in tissues such as brain, tional analysis of the KKS in intact animals, heart,damage. vessels, In particular, and kidney local are angiotensin-II involved in transgenic rats were generated expressing these processes. Therefore, transgenic rats different components of the system, such with local up- or downregulation of RAS as tissue kallikrein, the kinin B1 or the B2 components in these organs, e.g. by the local expression of antisense-RNA or of a in cardiovascular organs. These animals peptide-liberating protein, were produced supportedreceptor either the protective ubiquitously role or of specifically the KKS in and analyzed to clarify the local functions kidney and heart against ischemic, diabetic, and hypertrophic injury. Knockout mice mouse and rat models for non-classical RAS for the kinin B1 receptor and mice lacking componentsof angiotensin such II. Otheras neurolysin, genetically ACE2, altered the both kinin receptors were generated and (pro)renin receptor, angiotensin(1-7) and revealed important functions of these pro- its receptor Mas, have elucidated the physiological function of these molecules. More- Moreover, kinin receptors turned out to be over, a novel peptide of the RAS, alamandine, involvedteins in pain in arteriogenesis,perception and sepsis,inflammation. stroke, with its receptor, MrgD, was discovered and multiple sclerosis and high-fat diet induced functionally analyzed. Together with trans- obesity. genic rats overexpressing ACE2 or angio- Moreover, downstream mediators of kinins tensin(1-7), ACE2- and Mas-knockout mice - characterized the ACE2/angiotensin(1-7)/ Mas system as a protective axis that coun- such as the chemokine CXCL5 and its re ceptor CXCR1 as well as other chemokines, such as CXCL12, are functionally analyzed 4 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1: Mice deficient for angiotensin converting enzyme 2 (ACE2) develop liver steatosis with age. Liver tissue sections of ACE-knockout (lower panels) and wildtype (upper panels) mice at 3 (left) and 6 months (right) of age were stained for adipophilin, which labels lipid droplets in the cells.

in newly developed knockout mouse mod- altered autonomic control leading to impairment of sleep, respiration, cardiovas- cardiomyocytes and transgenic rats over- cular and metabolic parameters. We also expressingels. Mice with the a chemokinelocal deletion in thisof CXCL12 cell type in revealed that exercise-induced adult neu- revealed an unexpected deteriorating role of this chemokine for heart function after addition, TPH2-deletion is correlated with increasedrogenesis aggression,is abolished maternal in these neglect, animals. and In as major determinant of macrophage foam reduced aversive and anxiety-like behavior. cellischemic formation damage. in atherosclerosis. CXCL5 was characterized Moreover, lack of TPH2 evoked increase in sexual activity of female mice. Since fur- Serotonin system ther behavior analysis of this model was Serotonin is a monoamine which func- hampered by poor ability of the mouse for tions as an important neurotransmitter in complex social interactions and learning the central nervous system and as a major capacity, we now perform these studies peripheral mediator produced by entero-

and released by platelets in the circulation. 2)in and TPH2-deficient in rats with ratsinducible generated knockdown via zinc- of Wechromaffin discovered cells thatof the vertebrates gut and transported have two TPH2.finger-nuclease First data basedindicated gene-deletion communication (Fig. tryptophan hydroxylases, the rate limiting - enzymes in serotonin synthesis, TPH1 and Todeficit study and role impairment of serotonin in inultrasonic embryonic vocal de- responsible for the synthesis of serotonin velopmentization in juvenile we have serotonin-deficient additionally generated rats. inTPH2. the Micegut, showeddeficient thatin TPH1, peripheral the isoform sero- TPH1/2 double knockout mice. tonin is involved in thrombosis, pulmonary Since serotonin is a major determinant in hypertension, remodelling of mammary the pathogenesis of several diseases, such glands, tumor angiogenesis, liver regen- as pulmonary hypertension, we developed eration, and hepatitis, but not in bone me- and patented modulators of TPH activity as tabolism as previously suggested. Mice novel therapeutic drugs together with the high-throughput screening platform and for the synthesis of serotonin in the brain, the medicinal chemistry department of the weredeficient surprisingly in TPH2, viable the isoform and fertile, responsible despite - a near complete lack of serotonin in the cology (FMP) and the Helmholtz Protein brain, and showed growth retardation and SampleResearch Production Institute Facility for Molecular (PSPF). Pharma

MDC Research Report 2016 5 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 2: Rats deficient for tryptophan hydroxylase 2 (TPH2) lack serotonin in the brain. Hindbrain sections of wild-type (TPH2+/+, left) and TPH2-knockout rats (TPH2-/-, right) were stained with anti-serotonin antibodies.

Figure 3: Xgal staining in importin α1 (KPNA2) deficient embryos. Female KPNA2 deficient mice are infertile. To assess KPNA2 expression in early embryos, females heterozygous for a lacZ genetrap in the KPNA2 locus were crossed to wildtype males. Heterozygous embryos show Xgal staining from 2cell stage on, wildtype embryos don’t. The weak staining in all zygotes is caused by maternal KPNA2 mRNA. Thus, KPNA2 is a maternal effect gene and a marker of zygotic genome activation.

6 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Patents / Patent applications

primary hemostasis, US7049336 (B2) (23.05.2006) Method for diagnosing neuronal diseases and for the treatment of deficient (23.05.2012) Neuronally expressed tryptophan hydroxylase and its use,EP1521823 (B1) Method for modulating, treating and/or preventing metabolic syndrome using Androgen receptor The group generated and characterized ani- MAS-G-protein-coupled receptor agonists, US8586054 (B2) (19.11.2013) mal models with altered androgen receptor New Benzimidazolyl-xanthines as Tryptophan-Hydroxylase (TPH) Inhibitors, SelectedEP15156418.4 Publications (24.2.2015) expression in distinct cell types. Transgenic Mühlstedt, S., Ghadge, S.K., Duchene, J., Qadri, F., Jarve, A., Vilianovich, L., Popova, rats overexpressing the androgen receptor E., Pohlmann, A., Niendorf, T., Boye, P., Özcelik, C., Bader, M. Cardiomyocyte- in cardiomyocytes exhibited marked alterations in cardiac rhythmicity probably due myocardial infarction. J. Mol. Med. (Berl) 2016, 94, 1005-1014 to dysregulation of potassium channels and Järve.A,derived Mühlstedt.S,CXCL12 is not Qadri.F, involved Nickl.B, in cardiogenesis Schulz.H, Hübner.N, but plays Özcelik.C, a crucial roleBader.M. in mice lacking the receptor in macrophages Adverse left ventricular remodeling by glycoprotein nonmetastatic melanoma revealed its function in salt homeostasis Huegel, S., Depping, R., Dittmar, G., Rother, F., Cabot, R., Sury, M.D., Hartmann, E., and blood pressure control. protein B in myocardial infarction. FASEB J 2017;31:556-68.

Bader, M. Identification of importin alpha 7 specific transport cargoes using a Importins Rousselle, A., Qadri, F., Leukel, L., Fontaine, J.-F., Rüstem, Y., Sihn, G., Bader, M.,proteomic Ahluwalia, screening A., Duchene, approach. J. Cxcl5 Mol. limits Cell. macrophageProteomics. 2014, foam cell13, 1286-1298formation in

machinery that transports proteins into the Klempin, F., Beis, D., Mosienko, V., Kempermann, G., Bader, M., Alenina, N. Sero- Importins are essential components of the- toninatherosclerosis. is required J. for Clin. exercise-induced Invest. 2013, 123, adult 1343-1347 hippocampal neurogenesis. tive approach with the University of Lübeck tonucleus study ofthe eukaryotic physiological cells. functions In a collabora of al- J. Neurosci. 2013, 33, 8270-8275 pha importins we have generated knockout Group Leader Mikhail Liskovykh* Prof. Dr. Michael Bader Sara Maghsodi* phenotype was discovered in mice lacking Fernanda Malinsky* mice for five paralogs. The most obvious Scientists Daisy Motta* the females are infertile. The female infer- Dr. Natalia Alenina Bernadette Nickl* Dr. Naozumi Araragi tilityimportin of importin alpha1 andalpha7 alpha7: knockout In both mice lines is Katrin Nitz Dr. Megha Bangalore* Aline Oliveira* based on its essential function during zy- Dr. Saleh Bashammakh* Polina Peeva gotic genome activation of developing em- Dr. Mike Becker* Markus Petermann* bryos. For importin alpha1 the molecular Dr. Fernando Botoni* Ashish Ranjan* basis is still under investigation (Fig. 3), as Dr. Laura Chama Anthony Rousselle is the cause for the male infertility which is Dr. Johan Duchene* Valeria de Souza* additionally observed in importin alpha7 Dr. Lorena Gomez* Frederick Wasinski* knockout mice. Furthermore, importin al- Dr. Stefanie Hügel* - Dr. Anne Järve Graduates and Dr. Tina Kienitz Undergraduates Dr. Friederike Klempin analysespha7 is involved were performed in Influenza identifying and Ebola thevi Hannes Daher* Dr. Alexander Krivokharchenko* Benjamin Grimmer* rus infection of cells. In parallel, proteomic - Lisa Johann* portin paralogs. Dr. Na Liu* Louise Kaldjob* specific import cargos of single alpha im Dr. IrinaSusann Lapidous* Matthes Albrecht Keil* Transgenic and stem cell Dr. Valentina Mosienko* Meike Latz* technology Dr. Elena Popova Andreas Magg* The group has also a strong emphasis in Dr. Fatimunnisa Qadri Phillip Schiele* Dr. Luiza Rabelo* Karolina Wszolek-Meißner* research. The rat is the preferred species Dr. Franziska Rother the field of rat embryology and stem cell Prof. Dr. Robson Santos* Technical Assistants Dr. Gabin Sihn order to obtain rat pluripotent stem cells Adelheid Boettger* Dr. Olga Tamyra-Lesniak* Susanne da Costa Goncalves in physiological and behavioural studies. In two methodologies were applied in our Prof. Dr. Mihail Todiras Cathrin Gerhard group: isolation of ES cells from rat pre- Dr. Magdalena Zaniewska* Sabine Grueger* implantation embryos and generation of Dr. Luyu Zhou* Anne Hahmann* induced pluripotent stem (iPS) cells from Reika Langanki PhDs Lisa Mallis carrying pluripotency genes. Furthermore, Daniele Alves* Keven Mallow transgenicfibroblasts uponrats haveinfection been with produced lentiviruses car- Rogerio Oliveira Batista* Charlene Memler* rying constructs, which express small in- Daniel Beis Andrea Mueller terference RNAs suited to downregulate Paula Cavalcante* Ralph Plehm Aline Costa* Thorsten Riepenhausen* Priyavathi Dhandapani* Madeleine Skorna-Nussbeck the insulin receptor, the (pro)renin recep- Janine Wolff* tor,specific apolipoprotein genes. The firstE, and target the genesLDL recep were- Veniamin Fishman* * part of GabrielYasmine Rufino Graf* Estrela* Secretariat the period established and successfully used to ablate Cornelia Hainer reported genestor. CRISPR/Cas in rats. technology has also been Iris Apostel-Krause

MDC Research Report 2016 7 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Martin Falcke

Mathematical Cell Physiology

The group develops mathematical models - for cell biology from single molecules, via onic kidney 293 cells and rat primary hepatocytes,stochastic we component.also found that In humanthe average embry in- protein complexes to cell level. Modelling terspike interval varied between individual serves to test and formulate biological hy- cells (1). To evaluate how individual cells potheses in a quantitative way as well as to reliably encoded stimuli when Ca2+ spikes exhibited such unpredictability, we com- predict the behavior of complex pathways bined Ca2+ imaging of single cells with math- and networks. Current projects comprise ematical analyses of the Ca2+ spikes evoked 2+ analysis and modelling of IP3 induced Ca by receptors that stimulate formation of spiking and excitation contraction coupling 3). This analysis revealed that signal-to-noise ratios in cardiac myocytes. wereinositol improved 1,4,5-trisphosphate by slow recovery (IP from feedback inhibition of Ca2+ spiking operating at Ca2+ is a ubiquitous intracellular messenger the whole-cell level, and they were robust that regulates diverse cellular activities. Ex- against perturbations of the signaling path- tracellular stimuli often evoke sequences of way. Despite variability in the frequency of intracellular Ca2+ spikes, and spike frequen- Ca2+ spikes between cells, steps in stimu- cy may encode stimulus intensity. However, lus intensity caused the stochastic period the timing of spikes within a cell is random of the interspike interval to change by the because each interspike interval has a large same factor in all cells. These fold changes

Fold changes determine a universal exponential concentration-response relation for the average interspike interval of Ca2+ spikes evoked by stimulation of GPCR with the concentrations given in the figures. The lines are single exponential functions fit to the experimental data.(A) HEK cells stimulated with CCh, (B) hepatocytes stimulated with phenylephrine, (C) hepatocytes stimulated with vasopressin, (D) salivary gland stimulated with 5-HT.

8 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Three dimensional visualisation of spatially resolved [Ca2+] at 70.0 ms after depolarisation of the membrane potential. (A) Isosurfaces show 2+ 2+ [Ca ]i in green for [Ca ]i = 0.6 µM 2+ and red for [Ca ]i = 2.4 µM . (B) The yellow isosurface shows free sarcoplasmic [Ca2+] = 430 µM. There are 320 CRU per z-disc, amounting to 5160 CRUs in total, with an average of 50 RyR and 12.5 LCC per Ca2+ release unit (diadic cleft).

reliably encoded changes in stimulus in- ing a quasistatic approximation within the tensity, and they resulted in an exponential - dependence of average interspike interval tegrating the partial differential equations. on stimulation strength. We conclude that Wedyad simulate (3) and finitewhole element cell Ca2+ methods concentration for in Ca2+ spikes enable reliable signaling in a cell population despite randomness and cell-to- RyR-channel Markov schemes. cell variability, because global feedback re- profiles using three previously published duces noise, and changes in stimulus intensity are represented by fold changes in the stochastic period of the interspike interval. Selected Publications

Mathematical modelling of excitation-con- 2+ Spikes. Sci. Signal. 7, ra59 (June 24, 2014, 2014). traction coupling (ECC) in ventricular cardi- J.K. Vierheller, Thurley et W. al., Neubert, Reliable M.Encoding Falcke, S.of H.Stimulus Gilbert, Intensities N. Chamakuri, Within A multiscale Random Sequences computa- ac myocytes is a multiscale problem, and it tionalof Intracellular model of Caspatially resolved calcium cycling in cardiac myocytes: from detailed 6, 255 - (2015). tailed simulation tools. ECC involves gradi- T.cleft Schendel, dynamics R. Thul,to the J. whole Sneyd, cell M. concentrationFalcke, How does profiles. the ryanodine Frontiers receptor in Physiology in the entsis therefore on the length difficult scale to develop of 100 nm spatially in dyadic de ventricular myocyte wake up: by a single or by multiple open L-type Ca2+ channels? European Biophysics Journal 41, 27 (2012). 100 µm of the whole cell, as well as the sub- J. Zimmermann, M. Falcke, Formation of Transient Lamellipodia. PLoS one 9 (2014). millisecondspaces and concentration time scale of profiles local concentra along the- G. Dupont, M. Falcke, V. Kirk, J. Sneyd, Models of Calcium Signalling. S. S. Antmann,, e87638 L. tion changes and the change of lumenal Ca2+ content within tens of seconds. Our concept pp. 436. for a multiscale mathematical model of Greengard, P. J. Holmes, Eds., Interdisciplinary Applied Mathematics (Springer, 2016), Ca2+-induced Ca2+ whole cardiomyocyte electrophysiology in- Leader corporates stochastic release simulation (2) (CICR) of individ and- Prof. Dr. Martin Falcke ual LC- and RyR-channels, spatially detailed concentration dynamics in dyadic clefts, Scientists Behnam Amiri rabbit membrane potential dynamics and Nagalah Chamakuri Nicolai Friedhoff a system of partial differential equations Dr. Stephen Gilbert Daniel Schwabe for myoplasmic and lumenal free Ca2+ and Ca2+ -binding molecules in the bulk of the PhD students Graduate and Undergraduate cell. We developed a novel computational Setareh Dolati students approach to resolve the concentration gra- Janine Vierheller Wilhelm Neubert dients from dyadic space to cell level by us-

MDC Research Report 2016 9 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Holger Gerhardt

Integrative Vascular Biology Laboratory

The Integrative Vascular Biology Laboratory aims to unravel the fundamental principles years at the MDC following relocation from The years 2015 and 2016 mark our first full and molecular/genetic regulation of func- projects in our team currently converge on tional vascular network formation in devel- investigatingthe Francis Crick how Institute endothelial in London. cells collec Many- opment and disease. The ultimate goal of tively coordinate functional vessel remod- our research activities is to advance our eling, i.e. the transition from a primitive and poorly perfused vessel plexus to the understanding of mechanisms and mol- fully remodeled network with hierarchical ecules controlling formation and pattern- diameter control. We are also beginning to ing of a hierarchically branched vascular address the mechanisms regulating main- network such that we can inform, innovate tenance and plasticity of vascular networks in adult physiology, and principles of func- and implement therapeutic approaches to tional adaptation, as well as maladaptation mitigate cardiovascular complications, and in vascular disease. establish or restore tissue homeostasis in The present report however will focus on compromised patients. We develop and use and novel mechanism of lumen formation genetic tools to image and manipulate mol- inthe angiogenesis identification and of its a implications highly unexpected for ves- ecules, complexes and cells in spatiotem- sel plasticity. poral controlled manner in vivo in mouse - and zebrafish models. We further use giogenesis and in 3D-cultured endothelial iteration between predictive computational Previous studies in zebrafish models of an- modeling and experimentation to develop tion and coalescence as a mechanism of cells identified intracellular vacuole forma hypothesis and systems level understanding lumen formation during blood vessel de- - of the processes at play. mens that form in the embryo before heart- beatvelopment. commences, Indeed, as the well earliest as lumens vascular that lu

can form de novo through this mechanism. Mostform vascularin cell culture networks in the in vertebratesabsence of howflow,- ever form and extend after cardiac activity

process known as sprouting angiogenesis, and bloodlumen flow formation are established, mechanisms following in these a conditions are not fully understood. We

apical membrane (i.e. lining the lumen) of endothelialestablished cells fluorescent to allow reportersdetailed dynamic for the imaging of lumen formation in the inter-

pushing the limits of spatial and temporal resolutionsegmental vesselsusing state-of-the of zebrafish artembryos. spinning By disc microscopy, we found no evidence for

10 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Model of lumen formation by inverse membrane blebbing during sprouting angiogenesis.

de novo vacuole formation during sprout- to expand lumens or show dilated, multi- - branched lumens. Together, these results men expands through a mechanism that we unravel a novel mechanism of lumen foring angiogenesis. Instead we found that lu mation in blood vessels, where a tight bal- haemodynamic forces applied on the api- ance between the pushing forces exerted by caltermed membrane ‘inverse of membraneendothelial blebbing’.cells as blood The the blood on the membrane of endothelial pushes into new vessels during sprouting cells and the endothelial cells own contrac- induce spherical deformations of the apical tile responses is required for shaping new membrane reminiscent of the membrane vascular tubes (Gebala et al, 2016). Future blebs observed during cell migration or work will need to establish how this exquisite elasticity of the apical membrane and However, unlike classical blebs, the apical its underlying actin cortex contractility is membranecell division deformations (Charras and display Paluch, an invert 2008).- regulated during the angiogenic process, as ed polarity, with the membrane protruding vessels mature and are subjected to higher and changing levels of blood pressure. dynamics of the actomyosin cytoskeleton duringinto the this cell. process, Interestingly, we found by following that lumen the to be highly instable as cessation of blood expansion by inverse blebbing is tightly Interestingly, the nascent lumen appears regulated through local and transient re- lead to lumen collapse. This is in marked cruitment and contraction of actomyosin contrastflow or increasedto mature endothelial vascular contractility networks, at the surface of the blebs, thus allowing which experience changing pressure and controlled growth of new lumen branches. Using inducible single-cell loss-of-function - somehowflow regimes be stabilized, in the course yet theof physiological mechanisms lial cells with impaired contractility fail areflow unclear. regulation. Thus mature lumens must approaches, we identified that endothe

MDC Research Report 2016 11 CARDIOVASCULAR AND METABOLIC DISEASE

Single endothelial tip cell (cyan, cell membrane) undergoing lumen formation (magenta, apical/luminal membrane) at the sprouting front of a mouse retina vasculature at post-natal day 6 (P6).

A second project in our team on endothelial cells lacking Fmnl3, or exposed to chemical actin regulation uncovered one important fomin inhibitors, maintained isolated ring- mechanism of lumen stabilization – endo- like junctions. Recent work by the team of thelial rearrangement and junctional actin Markus Affolter suggest that the c-terminal nucleation. Whereas nascent lumens can binding of actin to VE-cadherin is critical for form within single endothelial cells or be- junctional force transmission that drives en- tween two cells, the mature endothelial dothelial cell shape changes and rearrangements during vessel elongation and lumen two endothelial cells aligning so that paral- stabilization (Sauteur et al 2014). configuration in capillaries shows generally How Fmnl3 activity and localization is

throughlel junctional dynamic profiles cell runrearrangements along the length that involved in its recruitment to the junction requireof each vessel.regulated The configurationjunctional VE-cadherin is achieved remainsregulated, to and be shown.whether Our blood future flow work itself will is turnover (Bentley et al, 2014) and junc- need to unravel the complexity and dy- tional actin nucleation through the formin namics of the junctional complexes at play Fmnl3 (Phng et al, 2015). Live imaging in during these morphogenic processes in development. From a clinical perspective, localizes to endothelial junctions, and in- we will seek to establish whether and how duciblezebrafish endothelial embryos expression illustrated of that dominant Fmnl3 endothelial reactivation in adult physiology negative Fmnl3 causes lumen collapse and lumens and therefore contribute to small of forming parallel junctions, endothelial vesselor inflammation disease. might destabilize vascular failed endothelial rearrangement. Instead

12 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Perfused blood vessels in the trunk of a transgenic zebrafish embryo expressing the Selected Publications actin marker Lifeact specifically in endothelial (Aspalter, Gordon et al., 2015, Franco, Jones et al., 2016, Gebala, Collins et al., 2016, cells. Note the distinct actin accumulation Phng, Gebala et al., 2015, Ubezio, Blanco et al., 2016) along endothelial junctions visible as thin lines running across and along the vessels. AspalterH (2015) IM, Alk1 Gordon and Alk5 E, Dubrac inhibition A, Ragab by Nrp1 A, Narloch controls J, vascular Vizan P, Geudenssprouting I, downstreamCollins RT, Image courtesy of Li-Kun Phng Francoof Notch. CA, Nature Abrahams communications CL, Thurston 6: G, 7264 Fruttiger M, Rosewell I, Eichmann A, Gerhardt Franco CA, Jones ML, Bernabeu MO, Vion AC, Barbacena P, Fan J, Mathivet T, Fonseca CG, Ragab A, Yamaguchi TP, Coveney PV, Lang RA, Gerhardt H (2016) Non-canonical

remodelling. eLife 5: e07727 Wnt signalling modulates the endothelial shear stress flow sensor in vascular formation by inverse membrane blebbing during angiogenesis in vivo. Nature cell References Gebala V, Collins R, Geudens I, Phng LK, Gerhardt H (2016) Blood flow drives lumen Bentley K, Franco CA, Philippides A, Blanco Phng LK, Gebala V, Bentley K, Philippides A, Wacker A, Mathivet T, Sauteur L, Stanchi biology 18: 443-50 R, Dierkes M, Gebala V, Stanchi F, Jones M, F, Belting HG, Affolter M, Gerhardt H (2015) Formin-mediated actin polymerization at endothelial junctions is required for vessel lumen formation and stabilization. Developmental cell 32: 123-32 Welsh L, Vestweber D, Gerhardt H (2014) TheAspalter role ofIM, differential Cagna G, Westrom VE-cadherin S, Claesson- dynam- K, Gerhardt H (2016) Synchronization of endothelial Dll4-Notch dynamics switch ics in cell rearrangement during angiogen- Ubezioblood vessels B, Blanco from RA, branching Geudens toI, Stanchiexpansion. F, Mathivet eLife 5 T, Jones ML, Ragab A, Bentley esis. Nature cell biology 16: 309-21

way: how to migrate without lamellipodia. Start of the Group: September 2014 NatureCharras reviews G, Paluch 9: 730-6E (2008) Blebs lead the Group Leader Véronique Gebala Prof. Dr. Holger Gerhardt Filipa Neto formationGebala V, Collinsby inverse R, Geudensmembrane I, Phngblebbing LK, (BIH Professorship, DZHK Profes- duringGerhardt angiogenesis H (2016) Blood in vivo. flow Nature drives cell lumen bi- sorship) GraduateInes Uhlenhut and Undergraduate Phng LK, Gebala V, Bentley K, Philippides A, Scientists students Wackerology 18: A, 443-50 Mathivet T, Sauteur L, Stanchi F, Dr. Elena Cano Rincon Belting HG, Affolter M, Gerhardt H (2015) Dr. Russell Thomas Collins Formin-mediated actin polymerization at Dr. Baptiste Claude Coxam TechnicalIris Unterweger Assistants endothelial junctions is required for vessel Dr. Alexandra Klaus-Bergmann Eireen Bartels-Klein lumen formation and stabilization. Devel- Dr. Andre Sousa Rosa opmental cell 32: 123-32 Dr. Anna Bozena Szymborska-Mell Katja Meier Sauteur L, Krudewig A, Herwig L, Ehren- Dr. Anne-Clemence Vion Irene Hollfinger feuchter N, Lenard A, Affolter M, Belting HG Secretariat (2014) Cdh5/VE-cadherin promotes endo- PhD students Regina Philipp thelial cell interface elongation via cortical Marta Bastos de Oliveira actin polymerization during angiogenic Santiago Patricio Andrade Cabrera sprouting. Cell reports 9: 504-13

MDC Research Report 2016 13 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Michael Gotthardt

Neuromuscular and Cardiovascular Cell Biology

The long-term goal of the Gotthardt-lab is to Titin in sarcomere assembly and develop a basic understanding of the transi- heart disease Titin is a giant protein found in vertebrate tion from cardiovascular health to disease at muscle, where it extends over the half-sar- the molecular, cellular, organ and systems comere and integrates into the Z-disc and level and use this information to improve patient care. Specifically, we are interested titin gene expresses differentially spliced mRNAsM-band ofto upform to 100a continuous kb that encode filament. for proThe- in understanding how increased filling of the teins of up to 33,000 amino acids. Titin con- cardiac ventricle leads to improved contraction (Frank-Starling mechanism of the heart), which provide elastic diversity through alternativetains a PEVK splicing segment, and N2B-, posttranslational and Ig regions, how alternative splicing relates to diastolic heart failure, and how the electrical properties serine/threonine kinase domain, which has of the heart and mechanoelectrical coupling beenmodifications. implicated Its in C-terminusmechanotransduction. contains a are regulated. The group focuses on titin, Using our animal models, we were able to the largest protein in the human body, and and to link titin to adrenergic signal trans- the multifunctional coxsackie- and adenovi- duction,elucidate lateraltitin’s rolegrowth in sarcomere of the sarcomere, assembly rus receptor (CAR). We utilize animal models and the hypertrophy response. We study and differentiated stem cells combined with the role of titin in cardiac adaptation and how it converts mechanical input into a cell biological, biochemical, genetic, pharma- molecular signal. cological, and omics tools to establish titin as a mechanosensor and study sarcomere The Coxsackie- and Adenovirus Receptor (CAR) dynamics. With a loss of function approach, Both adenovirus and coxsackievirus are we have demonstrated that CAR is crucial common pathogens associated with myo- for embryonic development, determines the carditis and dilated cardiomyopathy (DCM). electrical properties of the heart, and can They infect cardiomyocytes using CAR and serve as a therapeutic target to improve When we started the CAR conditional cardiac remodeling. Currently our main trans- knockoutits co-receptors project, integrin the analysis αvß3 undof CAR αvß5. in lational focus is on harnessing alternative muscle and heart disease had been restrict- splicing to improve cardiac function using a ed to basic expression and localization analysis, which indicated an important role in pharmacological screen with custom splice remodeling after myocardial infarction. Ac- reporter assays. This work aims to identify cordingly, CAR expression is low in the adult and characterize patients with splice related but CAR is upregulated in areas surrounding myocardial infarction. Our conditional heart disease towards improved personalized knockout provided novel insights into the diagnostics and therapy. An overview of our role of CAR in arrhythmia and established scientific approach is depicted in Figure 1. its role in myocarditis and myocardial infarction. On the cellular level, we study how

14 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1: Experimental approach. We use cell- and animal models to understand the molecular and cellular basis of heart disease. Focusing on the sarcomeric protein titin and cardiac splicing, we integrate a candidate- with a systems biology approach to derive splice regulatory networks and therapeutic targets. With high throughput screening to identify splice regulators and validation in cell and animal models, we aim to develop novel therapeutic strategies for cardiac disease.

CAR effects the cytoskeleton, endocytosis, and signal transduction using inducible - cytes, and neurons (Figure 2). We translate CAR deficient ES and iPS-cells, cardiomyo- terfere with CAR function and improve the outcomeour findings of mice and withdevelop myocardial biologicals infarction to in by improving the response to ischemia.

Figure 2: CAR is a multifunctional protein. CAR (light green) is involved in cell contact formation, actin filament assembly, ion homeostasis (gap junctions), endocytosis, and several signal transduction pathways. It interacts with multiple extra- and intracellular proteins.

MDC Research Report 2016 15 CARDIOVASCULAR AND METABOLIC DISEASE

Alternative splicing and regulation of titin isoform expression in health and disease Alternative splicing has emerged as a major source of proteome diversity and as a central regulator of protein isoform expression - tive approach, we have used rodent and humanin development genetics, and cardiac disease. physiology, In an integra and functional genomics to identify RBM20 as a novel trans- acting factor regulating the differential splicing across a broad range of transcripts during normal development (Figure 3). Notably, this is one of the few examples of a human disease mechanism involving an effect in trans. Our extended splicing analysis suggests that RBM20 targets a subset of cardiac genes that concert- edly affect biomechanics, electrical activity, and signal transduction. This analysis of cardiac isoform expression does not only offer insights into the coordination of splicing events, but provides novel diagnostic and therapeutic targets for cardiac disease. Figure 3: Regulation of cardiac function via - alternative splicing. RBM20 is a cardiac splice peutic target, we have established a dual lu- factor that determines the elastic properties of In addition to evaluating RBM20 as a thera the sarcomeric protein titin. Its substrate spec- inhibit RBM20 activity to interfere with trum is now well characterized and RBM20 titinciferase splicing assay and and thus identified the diastolic molecules proper that- has been linked to human cardiomyopathy ties of the heart. Using genetic engineering with fibrosis, arrhythmia, and sudden death. in vivo

which are and currently the CRISPR/Cas9 evaluated for system, their sub we- have identified novel titin splice factors,

strate profile and regulation.

Figure 4: Visualizing sarcomere dynamics. A) Generation of titin knock-in mice expressing RFP-Titin and Titin-GFP fusion proteins. B) Titin fusion proteins are properly integrated into the sarcomere to produce the characteristic periodic Z-disc (Red) and M-band staining (green).

16 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Advancing technology and model systems To lay the foundation for the in vivo analy-

adaptor domains, we have generated vari- sis of titin’s multiple signaling, elastic, and- gical and mechanical protocols to induce atrophyous titin anddeficient hypertrophy. mice and Recently, established we have sur added a naturally occurring animal model to our portfolio that exhibits little muscle atrophy despite extended periods of immobilization – namely the hibernating griz- zly bear. Using an omics and mathematical Patents / Patent applications modeling approach, we derived novel ther- Means for treating heart disease. Max-Delbrück-Centrum für Molekulare Medizin apeutic targets to help prevent muscle atro- Berlin-Buch Apr, 17 2013: EP2692354 A1 phy in aging, immobilization, and inherited Polynucleotides for use in medicine. Max-Delbrück-Centrum für Molekulare Medizin muscle disease. Berlin-Buch Dec, 29 2010: WO 2010/149332 - Centrum für Molekulare Medizin Berlin-Buch Oct, 21 2004: WO 2004/090131 neered mice to express a combination of Inducible site-directed mutagenesis through conditional gene rescue. Max-Delbrück- In a gain of function approach we engi Texas, 2001 enable the analysis of titin mobility, sar- LDL Receptor Signaling Pathways – US Patent Number 6,428,967. The University of comerefluorescently assembly, labeled remodeling, titin molecules and turn that- Selected Publications over under different mechanical loads, in Hinze F, Dieterich C, Radke MH, Granzier H, Gotthardt M. (2016) Reducing RBM20 embryonic versus postnatal development, activity improves diastolic dysfunction and cardiac atrophy. J Mol Med 94(12):1349- and in cardiac and skeletal muscle disease Freiberg F, Sauter M, Pinkert S, Govindarajan T, Kaldrack J, Thakkar M, Fechner H, (Figure 4). To evaluate compartmentalized 1358. mechanosignaling and follow changes in Klingel K, Gotthardt M. function of the coxsackievirus and adenovirus receptor (CAR). J Virol protein composition along the sarcomere 56. (2014) Interspecies differences in virus uptake versus cardiac in health and disease, we have now estab- Maatz H, Jens M, Liss M, Schafer S, Heinig M, Kirchner M, Adami E, Rintisch 1;88(13):7345- C, lished localization proteomics in vivo us- Dauksaite V, Radke MH, Selbach M, Barton PJ, Cook SA, Rajewsky N, Gotthardt M, Landthaler M, Hubner N. (2014) RNA-binding protein RBM20 represses splicing to orchestrate cardiac pre-mRNA processing. J Clin Invest already provided us with insights into sub- Guo W, Schafer S, Greaser ML, Radke MH, Liss M, Govindarajan T, Maatz H, Schulz H, nuclearing the BioIDcompartmentalization system. This approach of splicing, has Li S, Parrish AM, Dauksaite V, Vakeel P, Klaassen S, Gerull124(8):3419-30. B, Thierfelder L, Regitz- - Zagrosek V, Hacker TA, Saupe KW, Dec GW, Ellinor PT, MacRae CA, Spallek B, Fischer R, Perrot A, Özcelik C, Saar K, Hubner N, Gotthardt M. (2012) RBM20, a gene for heredi- tein phosphorylation and -complex forma- tary cardiomyopathy, regulates titin splicing. Nature Med. tionand nowat subsarcomeric enables the quantificationresolution. This of work pro da Silva Lopes K, Pietas A, Radke MH, and Gotthardt M. (2011) Titin visualization in complements our live imaging analysis of real time reveals an unexpected level of mobility within and 18(5):766-773. between sarcomeres. sarcomere assembly in titin GFP and RFP J Cell Biol knockin mice towards understanding the 193, 785-798. Group Leader We will build on our expertise in generating Prof. Dr. Michael Gotthardt Joanna Kaldrack andplasticity analyzing of myofilaments. animal models to study the Anne Margarete Merks* molecular basis of cardiovascular, skeletal Scentists Franziska Rudolph muscle and neurological disease, identify Dr. Vita Dauksaite Meghna Hiren Thakkar therapeutic targets, and to evaluate novel Dr. Thirupugal Govindarajan therapeutic strategies. To advance patient Dr. Martin Liss Technical Assistants care, we will improve our interaction with Dr. Michael Radke Janine Fröhlich clinicians at the Charité and the DZHK and Beate Goldbrich-Hannig scientists at the FMP. This includes the E:Bio PhD students Carmen-Anisia Judis initiative of the BMBF for personalized di- Olya Bakina agnostics and therapy of splice related car- Claudia Fink Secretariat diac disease as well as the FMP-screening Wiebke Hamann Feraye Kocaoglu and drug development infrastructure and Florian Hinze outsourcing of the production of biologicals Judith Hüttemeister * jointly with Daniela Panakova under GMP conditions.

MDC Research Report 2016 17 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Norbert Hübner

Genetics and Genomics of Cardiovascular Diseases

A major challenge today is to understand Genetic regulation on multiple systems perturbations that are relevant to layers complex diseases in human populations and genes and pathways that predict aberrant that are due to subtle changes induced by physiologyWe are interested in complex in the disease identification or meta of- naturally occurring DNA variation or environ- bolic control. The spontaneously hyperten- mental conditions. Often such perturbations sive rat (SHR) is the most widely studied of biological networks, induced by complete animal model of human hypertension and it displays a large number of other physio- knockout or over-expression of individual logical and pathophysiological phenotypes, genes, have been used to help understand- including stroke, cardiac hypertrophy and ing the molecular basis of human diseases. failure, and features of the human meta- We are applying novel large scale technolo- bolic syndrome. We study the SHR and SHR derived substrains using genetic link- gies and advanced computation in multidisciplinary approaches to expand the systems of genetic control points for gene expres- concept and to accelerate discovery of gene sionage and– or previouslyexpression quantitative identified thousands trait loci (eQTLs) – across the SHR genome. We ex- networks underlying complex cardiovascu- tended our studies on the transcriptional lar and metabolic disease phenotypes in the rat model, and we translate the findings to the molecular pathology of related human strainslevel and bred included from 7SHR tissues and from BN thefounders BXH/ forHXB comprehensive panel of recombinant expression inbredstudies. (RI) We disorders. Our particular interests are inflam- constantly seek to improve our strategies matory and metabolic genetic risk factors and methods using deep re-sequencing and that lead to cardio- and cerebrovascular - target organ damage. Any layer of regulation of key genetic regulators in the animal model.statistical The approaches next step for for us thewas identificato extend tion can affect expression of a gene, and our studies to the regulation on the trans- we are more and more interested to look lational level. Several genetic disorders are not only at the genetic variation regulation of caused by mutations that affect protein translation, but so far genome-wide pro- gene expression but also at effects in post- tein synthesis rates have not been studied transcriptional and epigenetic processes of in complex disease models. Therefore, we complex disease. We are interested in all levels of regulation in the rat and human. We (Ribo-seq), and performed genome-wide adapted a new method, ribosome profiling are using integrated approaches involving in heart and liver tissues to investigate methods like RNA-seq, ChIPseq and ribo- RNA sequencing and ribosome profiling some profiling, and we have already shown the spontaneously hypertensive rat. With strain-specific translational regulation in that translational regulation is extensive in this technique, only the parts of mRNA molecules that are occupied by translating the left ventricles and livers of the BN-Lx and SHR rat strains. our lab, we optimized this method to make ribosomes are captured and sequenced. In

18 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1 The HxB/BxH recombinant inbred (RI) panel and riboQTL mapping for identification of genetic loci responsible for translational differences. Scheme of the RI panel showing alleles from the BN rat (red) and the SHR strain (grey). RNA-seq and Ribo-seq data from left ventricular and liver tissue was integrated with genotyping information for eQTL and riboQTL mapping. QTLs are genetically variable loci that strongly link to transcription and translation differences.

it applicable to mammalian tissues, includ- the SHR rat, including the cardiac disease ing the left ventricle. We found hundreds genes Myh6 and Fads1 and genes regulating metabolite levels (e.g. Acadl and Tat). almost doubling the number of differen- Moreover, several human GWAS candidates tiallyof strain-specific expressed genes. differences in translation, were found to be primarily translationally We extend this approach to liver and regulated in the rat model. These data pro- left-ventricular tissue of all 30 animals of vide insight in the wide variety of genetic mechanisms that introduce variation in panel-wide genotype information to per- gene expression levels, possibly contribut- formthe RI quantitative panel. We integrate trait loci (QTL)these datamapping. with ing to complex disease. Recently, we have The integration of genetic, transcriptional adapted this technique to human left- and translational data sets reveals distinct ventricular biopsies of dilated cardiomy- signatures in 3′UTR variation, RNA-binding opathy (DCM) patients. This resulted in an protein motifs and miRNA expression asso- unprecedented assessment of translated ciated with translational regulation of gene sequences in the human diseased heart, in- expression. Using both transcriptional and cluding evidence that ribosomes translate translational information we can identify truncated titin (TTN) isoforms up to the local (cis) and distant (trans) effectors of truncating mutation, and occasionally even both levels of gene expression regulation and discriminate genetic loci regulating comparing DCM patients and controls on transcript abundance (eQTLs) from genetic beyond (~5-10% read-through). When loci acting purely at the translational level differentially expressed genes including - masterthe transcriptional regulators of level,cardiac we transcription find many ferences on the transcriptional level tend to(riboQTLs). be equally We apparent find that on strain-specific the translational dif 437 differentially expressed long noncod- level. Remarkably, Ribo-seq data also ex- such as TBX20. Interestingly, we also find - appear to be translated. The latter could be tion, which are invisible at the transcrip- aning indicationRNAs (lncRNAs), of a regulatory of which 257function (~59%) for pose strain-specific differences in transla these lncRNAs during translation or the protein abundance, and almost double the differential translation of so far unknown numbertome level. of Theydifferentially do influence expressed final cellular genes micropeptides potentially playing a role in between the strains. Translational regu- DCM. Combined, these data suggest a variety of potential novel disease mechanisms many genes with potential relevance to driven by transcriptional and translational thelation cardio-metabolic influenced the disease expression phenotype levels of regulation in the human heart.

MDC Research Report 2016 19 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 2 Identifying local and distant riboQTLs as cis or trans regulators of differential translation. (A) In cis, a genetic variant may change specific transcript properties that affect translational efficiency. (B) In trans, a distant variant may locally change the expression of ‘gene A’, which regulates the translation of ‘gene B’ at a distant locus.

Figure 3 (A) The Endog gene is differentially transcribed and differentially translated. A clear difference in expression is indicative of a strong cis eQTL (left) and riboQTL (right). (B) An example of a gene (Nop16) with transcriptional regulation in cis, this effect is absent at the translational level (translational buffering). (C) The Myzap gene has no local regulator of transcription, but does have a local genetic variant strongly linked to a translational difference.

20 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Integrated genomic approaches H3K4me3 marks were under distant genetic for identification of gene networks control and one major H3K4me3 regulating - ing proteins that contribute to pathological hundreds of H3K4me3 sites. As the exam- posttranscriptionalWe have identified mutationsregulation in of RNA an entirebind inedgenomic histone region marks was were identified known predictors impacting network of target genes that lead to cardio- of gene expression states, data on genetic - variation underlying histone marks could tion with Michael Gotthardt (MDC) we used genome-widemyopathy and approaches heart failure. in Ina collaborarat strain Hence, we could show that genetic variation significantly increase the number of eQTLs. of-function mutation in RBM20 as the un- in rat tissue that might further help to elu- derlyingdeficient causein titin for splicing the pathological and found isoform a loss- cidatehas a great so far impact unknown on histone genotype-phenotype modifications expression of titin. Mutations in RBM20 relationships in the heart. in humans have previously been shown to disrupt isoform transitions of a network of genes with essential cardiac functions and Selected Publications cause dilated cardiomyopathy. We showed Schafer, S., de Marvao, A., Adami, E., Fiedler, L. R., Ng, B., Khin, E., Rackham, O. J., van Heesch, S., Pua, C. J., Kui, M., Walsh, R., Tayal, U., Prasad, S. K., Dawes, T. J., Ko, N. S., human disease phenotype. Together with Sim, D., Chan, L. L., Chin, C. W., Mazzarotto, F., Barton, P. J., Kreuchwig, F., de Kleijn, D. Markusthat RBM20 Landthaler deficient and ratsNikolaus resembled Rajewsky, the Hub- we investigated the role of RBM20 as a car- ner,P., Totman, N., and T., Cook, Biffi, S. C., A., Tee, Titin-Truncating N., Rueckert, D., Variants Schneider, Affect V., Heart Faber, Function A., Regitz-Zagrosek, in Disease V., Seidman,Cohorts and J. G., the Seidman, General C. Population. E., Linke, W. Nat A., Genet Kovalik, 49, J. no. P., 1O’Regan, (2017): D., 46-53. Ware, J. S., directly RBM20 regulated transcripts by ap- Yusoff, P., Choksi, P. K., Ko, N. S., Singh, M. K., Epstein, D., Guan, Y., Houstek, J., Mracek, plyingdiac specific state-of-the-art splicing factor.RNA-protein We identified cross- Wang,T., Nuskova, M., Sips, H., P.,Mikell, Khin, B., E., Tan, Rotival, J., Pesce, M., Sun, F., Kolar, X., Ahmed, F., Bottolo, R., Widjaja, L., Mancini, A. A., M.,Schafer, Hubner, S., linking and immunoprecipitation coupled N. Angiogenesis. Nat Commun 7 (2016): 12061. , Pravenec, M., Petretto, E., MacRae, C., and Cook, S. A., Wars2 Is a Determinant of - Yang, L., Schafer, S., Sheng, C. C., Haghighi, A., Homsy, J., Hubner, N., Church, G., Cook, rectwith targets high-throughput revealed that sequencing RBM20 acts (CLIP- as a Hinson,S. A., Linke, J. T., W. Chopra, A., Chen, A., C.Nafissi, S., Seidman, N., Polacheck, J. G., and W. Seidman, J., Benson, C. C.E., C., Heart Swist, Disease. S., Gorham, Titin J., seq) experiments. The identification of di -

based proteomics we demonstrated direct Mutations in Ips Cells Define Sarcomere Insufficiency as a Cause of Dilated Cardio splicing repressor in the heart. Using SILAC Roberts, A. M., Ware, J. S., Herman, D. S., Schafer, S., Baksi, J., Bick, A. G., Buchan, R. J., interaction with important spliceosomal Walsh,myopathy. R., John, Science S., Wilkinson,349, no. 6251 S., Mazzarotto, (2015): 982-6. F., Felkin, L. E., Gong, S., MacArthur, J. A., components and showed how mutations Cunningham, F., Flannick, J., Gabriel, S. B., Altshuler, D. M., Macdonald, P. S., Heinig, within RBM20 may disrupt its repressor de Marvao, A., Dawes, T. J., Gulati, A., Birks, E. J., Yacoub, M. H., Radke, M., Gotthardt, function. We further revealed that not only M., Keogh, A. M., Hayward, C. S., Banner, N. R., Pennell, D. J., O’Regan,Hubner, D. P., San, N. T., R., genetic variation, but also the modulation of RBM20 gene expression levels is opera- M.,and Wilson, Phenomic J. G., Dissection O’Donnell, of C. the J., Prasad,Cardiac S. Effects K., Barton, of Titin P. J., Truncations Fatkin, D., in Health and Seidman,Disease. Sci J. G., Transl Seidman, Med 7,C. no.E., and270 Cook, (2015): S. A., 270ra6. Integrated Allelic, Transcriptional, tive in the regulation of its network and may Schafer, S., Adami, E., Heinig, M., Rodrigues, K. E., Kreuchwig, F., Silhavy, J., van be an attractive target for therapeutic inter- Heesch, S., Simaite, D., Rajewsky, N., Cuppen, E., Pravenec, M., Vingron, M., Cook, S. vention. These binding data will be used to A., and Hubner, N., Translational Regulation Shapes the Molecular Landscape of extend our studies determining the impact Complex Disease Phenotypes. Nat Commun 6 (2015): 7200. Maatz, H., Jens, M., Liss, M., Schafer, S., Heinig, M., Kirchner, M., Adami, E., Rintisch, of genomic variability on posttranscrip- C., Dauksaite, V., Radke, M. H., Selbach, M., Barton, P. J., Cook, S. A., Rajewsky, N., Got- thardt, M., Landthaler, M., and Hubner, N., Rna-Binding Protein Rbm20 Represses emphasize the importance of posttranscrip- 3419-30. tional modificationsregulation in incardiac humans. function Our studies and Splicing to Orchestrate Cardiac Pre-Mrna Processing. J Clin Invest 124, no. 8 (2014): provide novel mechanistic insights into the pathogenesis of human heart failure. Group Leader Franziska Kreuchwig Prof. Dr. Norbert Hübner Christin Mieth Epigenetic control Valentin Schneider Scientists of gene expression traits under genetic con- Allison Faber Graduate and Undergraduate trol,In addition we recently to the identificationdescribed the of genetic thousands im- Katharina Grunz students Matthias Heinig Michael Benedikt Mücke Several histone marks were associated to Oliver Hummel functionalpact on histone genomic modifications regions or genein rat expres tissue.- Henrike Maatz Technical Assistants Carola Rintisch Susanne Blachut distribution of H3K4me3 and H3K27me3, Franz Rüschendorf Mathias Gerhard twosion states.methylation We quantified marks theassociated genome-wide with Sebastian Schäfer Anita Müller actively transcribed promoters and silent Sebastiaan van Heesch Giannino Patone Sabine Schmidt PhD students mappedgenes, respectively. H3K4me3 marks In rat tissue,were under we could ge- Eleonora Adami Secretariat show that 14 % (liver) -21 % (heart) of Maria Bikou Maren Stauch

netic influence and segregating within the RI panel. A high proportion of 33 to 49 % of MDC Research Report 2016 21 CARDIOVASCULAR AND METABOLIC DISEASE Photo: Steffen Weigelt/MDC Steffen Photo:

Zsuzsanna Izsvák

Mobile DNA

Transposable elements (TEs) are discrete The impact of domesticated segments of DNA that have the distinctive TE-derived sequences J. Wang, M. Singh, A. Szvetnik, T. Rasko, ability to move from one genetic location C. Sun, V. Raghunathan to another in a genome. Although, large fractions of genomes can be composed of Occasionally transposons are co-opted for TE-derived sequences (~55% of the human cellular functions. A high throughput pro- genome), the vast majority of these ele- interacting partners of the Sleeping Beauty ments are not essential to the host cell, and andtein piggyBac interaction transposases. screen identified Our data cellular sug- often referred as ’junk DNA’ As an attempt - ence of domesticated piggyBac-like ele- to turn “junk” into treasure, the molecu- mentsgest an in important the human ramification genome. for the pres lar reconstruction of ancient transposable Recent studies indicate that some families elements, including Sleeping Beauty (SB) of ancient endogenous retroviruses (ERVs), represents a milestone in applying trans- such as HERVH has a role in the acquisition and maintenance of pluripotency in pri- position-mediated gene delivery in verte- mates. Using human pluripotent stem cells brate species. Our focus is to understand as models, we aim at deciphering the po- the basic mechanism of transposition and tential biological functions of TEs in human the interplay between TEs and host cells pluripotency (see press release: https:// www.mdc-berlin.de/43776186/). in mammals. The Mobile DNA team inte- Recently, TFCP2l1 (LBP9) has been recog- grates basic knowledge with technologies nized as a potential transcription factor that that are revolutionizing genomic manipula- in conjunction with HERVH is involved in human pluripotent network. LBP9 belongs tions in vertebrate species, including gene to the CP2 family of transcription factors. therapy, transgenesis, cancer research and We study how the CP2 family is involved in functional genomics. Our studies also focus the self-renewal and differentiation poten- on evolutionary processes that seem to tial in human pluripotent cells. We investigate the impact of conserved TEs “recycle” inactive TEs and co-opt them for during primate evolution (collab. with L. D. novel cellular function, and argue against Hurst, U. Bath, UK). Transcriptional expres- the dogma that views them exclusively as sion of TEs is tightly controlled during de- parasites or “junk” DNA. velopment and seems to be deregulated in certain diseases. TEs have likely played a key role in distrib- uting non-coding, regulatory elements in the vertebrate genome. We investigate the potential impact of the evolutionary conserved Medium Reiterated frequency repeats (MERs).

22 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

The image shows typical human embryonic stem cell (hESC) colony. The heterogenous colony consist of primed (grey) and naïve-like cells (green). The naive-like cells are expressing HERVH–driven GFP reporter. The reporter-based strategy can be used to derive homogenous, naïve-like colonies from conventional human pluripotent cultures. https://insights.mdc-berlin.de/en/2016/01/how-to-detect-and-preserve-human-stem-cells-in-the-lab/ Shortener Version: https://goo.gl/J1T7XF

The impact of TE-derived Transposon-host interaction activities on disease mechanisms S. Narayanvari, A. Szvetnik, G. Yong, J. Rugor, M. Singh, H. Cai, N. Fuchs*, R. Anwar A.Schmitt*, A. Devaraj*, D. Grzela*,

Global epigenetic changes during early TEs are best described as molecular para- development resets the epigenetic status sites with the potential to give rise to a resulting in transcriptional activation of variety of genetic alterations. Such alterations include mutational damage to gene various HERV families. We investigate the function, but can also provide useful ge- certain TE families, including LINE-1 and netic variability in host genomes. Thus, in upregulation of human endogenous TEs in contrast to most viruses, TEs and the host cancer.risk of genetic destabilization inflicted by have coevolved in a way that permits prop- agation of the transposon, but minimizes noise, generated by various TEs is relatively damage to the host. TEs make several inter- In comparison to other tissues, the genomic actions to host cellular machineries, piggy- aim is to investigate the epigenetic regula- back and modify certain basic mechanisms tionhigh inand placenta. imprinting In collaboration abnormalities with in Ourthe of the host organism. Our studies indicate pathogenesis of preeclampsia, a pregnancy that transposition is sensitive to stress, and disorder associated with cardiovascular stress-response involves a complex, inter- diseases of the mother (collab. R. Dechend, active regulatory platform involving evolu- Charité and L. D. Hurst , U. Bath, UK). tionary conserved cellular mechanisms.

MDC Research Report 2016 23 CARDIOVASCULAR AND METABOLIC DISEASE

We aim at validating that transcription fac- transposon system (Molecule of the Year, - 2009, Press Release MDC https://goo. nalling pathway) targets SB transposition gl/0oCvLQ) has become a popular tool for tor, HMGXB4 (a component of the Wnt-sig non-viral, therapeutic transgene delivery. The SB transposition seems to piggyback theto a heatwell-defined shock response developmental pathway phase. that normally reactivates heat-aggregated, nuclear The SB100X transposon has a favourable proteins. Thus, transposons might exist in systemsafety profile provides as comparedlong-term toexpression widely used in variousretro/lentiviral primary approaches.and stem cells. The As SB100X an im- to sense developmental and environmental portant issue regarding the implementa- changesa “latent” and form manipulate in the genome stress and signalling. are able tion of clinical trials, transposon vectors can be maintained and propagated as plas- Translating experience accumulat- mid DNA, making them simple and inex- ed in TE research to cutting-edge pensive to manufacture (e.g. GMP vector technologies production). H. Cai, H. Escobar*, A. García Pérez, I. Bilic, E. Nagy, SS. Chilkunda,Z. Zhou, J. Menyhert*, vector development and clinical trials we I. Klinge (1)In order try the to fillSB thesystem gap betweenin disease the models recent that were already on clinical trials using TE-based, non-viral integrating vector sys- retroviral vectors, but where safety was a tems represent a novel technology that - opens up new possibilities for gene thera- ease models that were already on clinical trialsserious using issue; retroviral (2) try thevectors, SB system but feasibil in dis-

py. The plasmid-based hyperactive SB100X 24 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

ity was an issue – engineered T cells for Selected Publications - easepersonalized models that cancer were immunotherapy; already on clinical BIH - trials,consortium; using non-viral (3) try the approaches, SB system but in disef- Wang Y.; Pryputniewicz-Dobrinska D.; Nagy E.E.; Kaufman C.D.; Singh M.; Yant S.; Wang J.; Dalda A.; Kay M.A.;Ivics Z.; Izsvak Z. Regulated complex assembly safe Macular Degeneration (AMD), TargetAMD guards the fidelity of Sleeping Beauty transposition. Nucleic Acids Research. (2017- 01-09) Nucleic Acids Res. 2017 Jan 9;45(1):311-326. doi: 10.1093/nar/gkw1164. ficacy was a limiting factor – Age-related - consortium [collab. G. Thumann (U. Hospi- tionPMID: and 27913727 cultivation of naive-like human pluripotent stem cells based on HERVH Wang J.; Singh M.; Sun C.; Besser D.; Prigione A.; Ivics Z.; Hurst L.D.; Izsvak Z. Isola TRANSPOSMART, ERARE tal Geneva, Switzerland)]; Von Willebrand expression Nature Protocols 2016 Feb;11(2):327-46. doi: 10.1038/nprot.2016.016. Epub 2016 Jan 21. PMID: 26797457 - diseasetransposon-based Type III, regenerative technol- Wang J, Xie G, Singh M, Ghanbarian A T, Raskó T, Szvetnik A, Cai H, Besser D, Prigione ogyconsortium; has a potential (4) include – cell-based models pacemakerwhere the A,stem Fuchs cells, N, Nature Schumann 2014 G, 516(7531):405-9 Chen W, Lorincz M, Ivics Z, Hurst L, Izsvák Z. 2014. Tran [collaboration M. Morad (U. of South Caro- scription driven by primate-specific endogenous retrovirus HERVH defines naïve lina)], muscle stem cells, dysferlinopathy - Ivicsgenesis Z, Mátés in rodents L, Yau by TY, pronuclear Landa V, Zidek microinjection V, Bashir S,of Hoffmann Sleeping Beauty OI, Hiripi transposons. L, Garrels W, [collab. S. Spuler (ECRC) (see press release: Kues WA, Bösze Z, Geurts A, Pravenec M, Rülicke T, Izsvák Z. 2014. Germline trans

Z.,Nature and Hamra,Prot. (tripple F.K. 2010. back-to-back) Generating 9(4):773-827 Knockout Rats by Sleeping Beauty Transposon integrationhttps://www.mdc-berlin.de/43612981), vectors with the cutting-edge Izsvák,Mutagenesis Z., Fröhlich, in Spermatogonial J., Grabundzija, Stem I., Shirley,Cells. Nature J.R., Powell, Meth. H.M.,7(6):443-5 Chapman, K.M., Ivics, Mátés, L., Chuah, M.K., Belay, E., Jerchow, B., Manoj, N., Acosta-Sanchez, A., Grzela, MyoGrad)]; (5) combine the plasmid-based D.P., Schmitt,A., Becker, K., Matrai, J., Ma, L., Samara-Kuko, E., Gysemans, C., Pryput-

WeDNA develop delivery the strategies; SB as a novel tool for chro- Molecular evolution ofa novel hyperactive Sleeping Beauty transposase enables niewicz,robust stable D., Miskey,C., gene transfer Fletcher, invertebrates. B., VandenDriessche, Nature Genet. T., Ivics, 41(6):753-61. Z. and Izsvák, Z. 2009 matin 3D mapping. This chromatin mapping approach is based on the ‘local hop- WangEukaryotic Y, Wang Cells. J, Devaraj 2014 PLoS A, Singh Genet. M, 10(3):e1004103. Jimenez Orgaz A, Chen JX, Selbach M, Ivics Z, SB-based method would be that chromatin Izsvák Z. Suicidal Autointegration of Sleeping Beauty and piggybac Transposons in regionsping’ feature are signed of SB. underA great physiological advantage of con the- - Group Leader Angelica Garcia-Perez tations of cross-linking. Dr. Zsuzsanna Izsvák Felix Lundberg Theditions SB avoidingtransposon chemical is suitable fixation for and somatic limi Guo Yong mutagenesis and emerged as a powerful Scientists Vaishnavi Raghunathan tool in cancer research. We utilize SB in a Dr. Enikö Eva Nagy Manvendra Singh rat model to decipher gene regulatory net- Dr. Christine Römer Chuanbo Sun works cooperating in hormone- dependent Dr. Julia Zadora Zhimin Zhou breast cancer development, progression Dr. Suneel Narayanavari and metastasis (collab. M. Bader (MDC), Dr. Tamas Rasko Graduate and Undergraduate D. Largaespada (U. Minnesota), J. Shull (U. Dr. Attila Szvetnik students Wisconsin). Dr. Jichang Wang Transposons can be harnessed as vehicles Natalia Lopez for introducing mutations into genes. By PhD students KathrinIsabel Klinge Radscheit combining transposon mutagenesis with Rabia Anwar the innate potency of spermatogonia, com- Katarina Stevanovic Technical Assistants prehensive libraries of gene knockouts are Aleksandra Kondrashkina Malgorzata Anna Dalda used to study fertility and behaviour (col- Sandra Neuendorf lab. K. Hamra, UTSW, USA). Huiqiang Cai Martina Wasser We use hiPSCs and genome engineering in ShreevathsaIlija Bilic Srinivasan a Kcnq1 model of type 2 diabetes aiming Chilkunda Secretariat drug screening. Anantharam Devaraj Beate Valeske * former students

MDC Research Report 2016 25 CARDIOVASCULAR AND METABOLIC DISEASE Photo: Christiane Trabert Christiane Photo:

Enno Klußmann

Anchored Signalling

Cyclic adenosine monophosphate (cAMP) is a ubiquitous second messenger that di- itsdefined substrates. cellular Phosphatases compartments dephosphory and thereby- rects a plethora of cellular processes. In the limitlate target the kinase’s proteins, access counteracting to only a PKA subset activ of- cardiovascular system, cAMP, for example ity. Compartmentalized PKA in concert with modulates contractility of cardiac myocytes PDEs and phosphatases facilitates coordinated phosphorylation of target proteins in and participates in the control of blood pressure through vascular smooth muscle cells. The Anchored Signalling group aims domainsresponse for to stimulationbinding PKA, of aPKA specific substrates, GPCR. to elucidate molecular mechanisms confer- PDEs,AKAPs phosphatases possess specific and proteinfurther interactionsignalling proteins such as additional kinases and may ring specificity to cAMP signalling. constitute the platforms for signalling hubs

AKAPs and PDEs define cAMP that facilitate specific cellular responses to signalling compartments Functionsspecific extracellular of phosphodiesterase cues (figure 1). 3 Stimulation of various G protein-coupled re- in cardiac myocytes and vascular ceptors (GPCRs) induces adenylyl cyclases smooth muscle cells (AC) to synthesise cAMP. cAMP could theoret- Two PDE3 subfamilies, PDE3A and PDE3B, ically diffuse freely throughout the cell. How- have been described, which both hydrolyse ever, this does not occur due to its degrada- cAMP and cGMP competitively. PDE3A ex- tion by cyclic nucleotide phosphodiesterases pression is higher in cardiac myocytes, platelets, vascular smooth muscle cells (VSMC), are the only means for terminating cAMP sig- and oocytes. PDE3A exists in three isoforms, nalling.(PDEs). TherePDEs arehydrolyse 21 genes cAMP encoding to 5’AMP 11 fami and- PDE3A1, PDE3A2 and PDE3A3, which are lies of PDEs, PDE1–11. The genes give rise located in different cellular compartments. to >100 mRNAs and thus to >100 isozymes. PDE3A1 is a membrane protein, while - PDE3A2 and PDE3A3 are both cytosolic and lies hydrolyse cAMP, those of the PDE5, PDE6 membrane-associated in human myocardium. andThe PDE9PDEs offamilies the PDE4, hydrolyse PDE7 cGMP,and PDE8 while fami the PDE3A2 is the dominant isoform in human members the PDE1, PDE2, PDE3, PDE10 and VSMC. PDE3A1, A2 and A3 all contain the PDE11 families hydrolyse both cAMP and same catalytic region and are similar concern- cGMP. PDEs are constitutively active enzymes ing catalytic activity and inhibitor sensitivity. that are located in compartments, and thus Ca2+ release from the sarcoplasmic reticu- establish gradients of cyclic nucleotides. lum (SR) into the cytosol of cardiac myocytes The cAMP gradients are sensed by cAMP stimulates contraction, while its removal from effectors, amongst them protein kinase A the cytosol causes relaxation of the myocyte. (PKA). PKA holoenzyme consists of a dimer of regulatory R subunits and two catalytic (C) observed that it participates in the control of subunits. cAMP activates the kinase, which CaWe2+ hadreuptake identified into the AKAP18δ SR of cardiac in the myocytes rat and then phosphorylates nearby substrates. PKA during diastole through direct interactions directly interacts with A-kinase anchoring - proteins (AKAPs), a family of around 50 scaf- laboration with the groups of V. Magianello, folding proteins, that tether the enzyme to with PKA and phospholamban (PLN). In col

NIH, USA, and M. Movsesian, University of 26 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

The interplay of cAMP synthesis and degradation and compartmentalisation of relevant enzymes confers specificity to cAMP signalling. A. Activation of cAMP signalling through G protein-coupled receptors (GP- CRs). B. Termination of cAMP signalling through phosphodiesterases (PDEs) and phosphatases. AKAPs bind directly PKA, PKA substrates, PDEs, phosphatases and further signalling proteins and coordinate cAMP signalling (fig. adapted from Dema et al. Cell Signal. 272474-87).

Utah, USA, we recently showed that in human erated a mouse model where expression of cardiac myocytes PDE3A1 phosphorylation- the Gskip gene was abolished through a gene dependently interacts with a multi-protein - - ment,targeting and approach. is associated The losswith of the GSKIP inability leads to controlscomplex, Ca which2+ reuptake contains into AKAP18γ, the SR. the hu aberrant regulation of GSK3β during develop man ortholog of AKAP18δ, and SERCA2A and expression in mice causes a cleft palate re- semblingclose the Hemifacial palatal shelves. microsomia The loss (Goldenhar of GSKIP ECRC, we analysed six missense mutations in syndrome) in humans. theIn collaboration gene encoding with PDE3A F. Luft from and six S. unrelated Bähring, families with hypertension and brachydactyly type E (HTNB). The mutations are adjacent Patents / Patent applications to each other and responsible for amino acid Patent No.: US 9,102,667 B2 substitutions in a conserved region that is N-Arylaminomethylene benzothiophenones for treatment of Cardiovascular Disease present in PDE3A1 and PDE3A2 and to which a function has not been ascribed, but which is SelectedInventors: E. Publications Klussmann, W. Rosenthal, J. Milic, M. Bergmann in close proximity to the catalytic domain. We Deák, A. V., Skroblin, P., Dittmayer, C., Knobeloch, K.-P., Bachmann, S., Klussmann, E. observed that the mutations lead to increased

phosphorylation of PDE3A1 and PDE3A2 by Maass,The A-kinase P. G., Aydin, anchoring A. Luft, protein F. C., Schächterle, GSKIP regulates C.*, Weise, GSK3β A., activity Stricker, and S., controls Lindschau, palatal C., PKA, which in turn renders the enzymes hy- Vaegler,shelf fusion M., Qadri,in mice. F., (2016)Toka, H. J. R.,Biol. Schulz, Chem. H., 291, Krawitz, 681-689 P. M., Parkhomchuk, D., Hecht,

H., Chitayat, D., Bialer, M. G., Wienker, T. F., Rittscher, K., Liehr, T., Jordan, J., Plessis, G., PDE3A is associated with increased prolifera- J.,Tank, Hollfinger, J., Mai, K.,I., Wefeld-Neuenfeld,Naraghi, R., Hodge, Y., R., Bartels-Klein, Hopp, M., Hattenbach, E., Mühl, A.,L. O.,Kann, Busjahn, M., Schuster, A., Rauch, tion.peractive. Together In VSMC with the increased presence constriction of hyperactive this A., Vandeput, F., Gong, M., Rüschendorf, F., Hübner, N., Haller, H., Mundlos, S., Bilgin- is likely causative for the increased peripheral turan, N.,, Movsesian, M.M., Klussmann, E., Toka, O., Bähring. S. (2015) PDE3A mutations cause autosomal-dominant hypertension with brachydactyly. Nature Genetics. vascular resistance in the patients and explains the hypertensive phenotype. Eccles, R. L., Czajkowski, M. T., Barth, C., Müller, P. M., McShane, E., Grunwald, S., Be- audette,47, 647-653. P., Mecklenburg, *C. Schächterle, N., Volkmer, shared first R., Zühlke, authorship K., Dittmar, from Klussmann G., Selbach, laboratory M., Hammes, A., Daumke, O., Klussmann, E., Urbe, S., and Rocks, O. (2016) Bimodal antagonism of The new AKAP GSKIP facilitates PKA signalling by ARHGAP36. Nature Commun. 7, 12963 crosstalk with GSK3β and controls Ahmad, F., Shen, W., Vandeput, F., Szabo-Fresnais, N., Krall, J., Degerman, E, Goetz, palatal bone fusion F., Klussmann, E., Movsesian, M., Manganiello, V. (2015) Regulation of SERCA2 AKAPs may interact not only with PKA, PKA activity by PDE3A in human myocardium: Phosphorylation-dependent interaction of PDE3A1 with SERCA2. J Biol Chem. 290, 6763-76 substrates or PDEs but also with further Dema, A., Schröter, M.F., Perets, E., Skroblin, P., Moutty, M.C., Deàk, V.A., Birchmeier, signalling proteins and thereby facilitate W., Klussmann, E. The A-Kinase Anchoring Protein (AKAP) Glycogen Synthase crosstalk with other signalling systems. We Kinase19630 3β Interaction Protein (GSKIP) Regulates β-Catenin through Its Interactions with Both Protein Kinase A (PKA) and GSK3β. (2016) J. Biol. Chem. 291, 19618- ithave binds identified PKA and GSK3β a second interaction kinase, glycogen protein (GSKIP) as a new AKAP, and have shown that- Group Leader tously expressed, a component of multiple PD Dr. Enno Klußmann Maike Schulz signalingsynthase systems kinase (GSK)3β.like canonical GSK3β Wnt, is insulin, ubiqui (Permanent Fellow)

is involved in the regulation of different bio- Scientist GraduateTanja Vukićević and Undergraduate logicalHedgehog, processes, Notch, including and TGFβ embryonic signaling, andde- Dr. Carolin Schächterle Eileen Maren Hallscheidt velopment, cell cycle progression, glycogen metabolism and immune regulation. Deregu- PhD student Technical Assistants Veronika Deak Andrea Geelhaar such as cancer or cardiac hypertrophy. Our in Alessandro Dema Dr. Kerstin Zühlke lation of GSK3β is associated with diseases- Mohamed Hassan Mostafa tates the phosphorylation and thus inhibition Ekaterina Perets Secretariat vitro analyses had indicated that GSKIP facili Katharina Schrade - of GSK3β by PKA. In order to gain insight into Iris Apostel-Krause the physiological function of GSKIP we gen MDC Research Report 2016 27 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Young-Ae Lee

Molecular genetics of chronic inflammation and allergic disease

The allergic disorders, eczema, asthma, vide strong evidence for a steady rise in and hay fever, are the most common chron- the incidence of diverse types of chronic ic diseases in childhood. The prevalence of countries over the past decades, indicating allergic diseases has increased to epidemic thatinflammatory genetic variants disorders that in have industrialized evolved to dimensions over the past decades. In the arm our immune system against multiple industrialized countries, 25-30% of the infectious/environmental exposures now population are affected. The goal of our group is to use genetic and hygienicinfluence environment immune response of modern thresholds civiliza to- genomic approaches in large study groups tion.favor chronic inflammatory disease in the

of affected patients and families to identify Our recent studies highlight the genetic genes and genetic variants that predispose overlap between risk loci for allergic dis- to eczema, asthma, and chronic inflam- - mation. We complement clinical pheno- matory conditions that have previously beenease withthought those to be for unrelated. other chronic Recently, inflam the types with precise molecular phenotypes largest genome-wide association study on to characterize underlying functional path- eczema comprising over 21,000 cases and ways. Enhanced understanding of how the interaction between environmental factors, bringing the total number of eczema sus- 95,000 controls identified 10 novel loci genetics, and the immune system results in the eczema loci also confer risk for other inflammation will enable the development of ceptibility loci to 31 (1). Notably, 80% of novel diagnostic and predictive clinical tools, ankylosingchronic inflammatory spondylitis, diseasesmultiple includingsclerosis, and of targets for innovative therapies. rheumatoidinflammatory arthritis, bowel or disease, type 1 psoriasis,diabetes, pointing to common mechanisms involved in eczema and other immune-mediated traits. Molecular pathways involved in innate immune signaling and T-cell polar-

the importance of the skin barrier defect the core paradigms of chronic human dis- andization of werean exaggerated significantly immune enriched. response While easeChronic underlying inflammation numerous has becomeconditions one in of- to environmental allergens in eczema is cluding allergic and autoimmune disorders, well established, all new susceptibility loci cardiovascular, and neurological disease. were related to (auto-) immune regulation, - in particular innate signaling and T-cell

toRegardless environmental, of the microbialmechanism or of self inflamma antigens our results suggest that targeted immune contributetion, inappropriate to dysfunction inflammatory and destruction responses modulationactivation and could specification. be another promising In summary, ap- of target organs. Epidemiologic data pro- proach for the treatment of eczema.

28 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Eczema usually develops in early infancy and is characterized by inflammatory and itchy lesions at typical sites of the body. Children with eczema have an increased risk to develop allergic asthma later in life in a disease course referred to as the “atopic march”. We analyzed the genetic determinants underlying disease progression from eczema to asthma.

Clinical and molecular the young infant and continues with the subphenotyping development of respiratory airways disease later in childhood and adulthood. We To better understand the clinical diversity have recently initiated a worldwide col- and temporal trajectory of allergic disease laboration of research groups to perform we evaluate patients in our research outpatient clinic (Hochschulambulanz) for pediatric allergy at the Experimental and withthe first longitudinal genome-wide data associationwho strictly study met our for Clinical Research Center (ECRC of Charité the “atopic march”. Focussing on children- and MDC). Over the years, we have recruit- lowed by asthma, we studied 2,400 cases phenotype definition of early eczema fol- many cases, we have obtained prospective gions in the human genome which were as- clinicaled several data thousand of the of allergicstudy participants. disease course In sociatedand 17,000 with controls the atopic and march identified at genome- 7 re and sensitization patterns. ). While four of -8 Recently, we used the longterm clinical wideAP5B1/OVOL1 significance (11q13.1), (P < 10 and C11orf30/ LRRC32them, FLG (11q13.5), (1q21.3), were IL4/KIF3A known (5q31.1), suscep- - tibility factors for eczema and one region data to perform the first genome-wide conditionanalysis of that allergic often disease heralds prognosis. a lifetime Inof we discovered two new regions (EFHC1 at fantile eczema is considered a “gateway”- 6p12.3harbored and an SLC6A15 asthma locusat 12q21.3) (IKZF3, that 17q21), were - not previously known to be involved in any ceptibleallergic disease.child commonly In an age-dependent passes a charac pat- teristictern known sequence as the of transient ”atopic march”, or persistent a sus One of the novel loci contained the EF-hand disease stages that begins with eczema in domainallergic/chronic containing inflammatory protein 1 gene condition. (EFHC1)

MDC Research Report 2016 29 CARDIOVASCULAR AND METABOLIC DISEASE

Loss-of-function mutations in the filaggrin gene (FLG) cause skin barrier deficiency and strongly predispose to eczema. We showed that having a FLG carrier mother increased the eczema risk of the child even when the mutations were not transmitted. The FLG maternal effect was observed only when mothers had allergic sensitization, suggesting that FLG mutation-induced systemic immune responses in the mother may influence eczema risk in the child.

which has been involved in ciliary dysfunc- Parent-of-origin effects tion in brain and lung providing a potential link to asthma. The second novel locus in- Epidemiological studies on allergic dis- cluded the solute carrier family 6, member eases have shown that maternal allergy is a 15 gene (SLC6A15). Selective inhibition of stronger risk factor for the child than pater- SLC6A15 by the histamine H1 receptor an- nal allergy. The molecular basis of this pref- tagonist loratadine which is clinically used erential maternal transmission of allergy for the treatment of allergic disease, was risk is currently unknown but it can poten- recently reported. tially occur through two different biological mechanisms: genomic imprinting and

risk for the atopic march without any effects imprinting, either the maternally or the pa- onBoth eczema novel alone loci specifically or asthma increasedalone. Of the ternallydirect maternal inherited genotype allele iseffect. expressed In genomic and the other allele is silenced. Thus, the effect more prominent than asthma loci. Support- of an allele depends on parental origin re- five known risk loci, eczema loci were much sulting in phenotypic differences between eczema loci were found to be associated reciprocal heterozygotes (parent-of origin ing these results, almost 90% of all known effects). Alternatively, maternal genotype effects occur when the maternal genotype Ourwith resultsthe atopic suggest march, that whereas genes triggeringonly 25% eczemaof the known are the asthma main genetic loci were determinants identified. genotypedirectly influences and occurs the through child’s the phenotype. maternal- the role of the skin as the interface between lyThis provided effect is independentenvironment ofduring the child’s prenatal own hostof the and atopic environment march. This in findingshaping highlights not only development. local, but also systemic immune responses We have now investigated parent-of-origin affect distant organs in the host. A causal effects for the loss-of-function mutations relationshipthat induce chronicbetween inflammation eczema and thisand asth may- ma subtype has important implications for strongest genetic risk factors for eczema interventional strategies. Since no cure for in the filaggrin gene (FLG) which are the- asthma exists, the modulation of skin im- - mune homeostasis early in infancy could ciencyknown to to eczema date and (3). provided We analyzed the first the four mo be an effective strategy to prevent not only mostlecular prevalent evidence European linking skinFLG mutations barrier defi in eczema but also the disease progression 759 complete nuclear families with eczema along the atopic march. from Central Europe. For the analysis, we

30 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

a multinomial and maximum-likelihood ap- were even more enriched in these regions applied the PREMIM/EMIM tool which uses suggestingand in regions that closeindels to are UTRs. more Indel likely eQTLs to be origin effects. functional because they have a more del- proach for flexible modelling of parent-of- eterious effect on regulatory elements and Our study replicated the known effect of - portion of SNPs previously found in GWAS a relative risk (RR) of 3.1 for heterozygous onsplice disease sites. traitsInterestingly, were correlated a substantial with pro in- andthe children’s of 10.5 for genotype homozygous on eczema, FLG mutation yielding dels with larger effects on eQTLs, support- carriers. By including the genotype of the ing the hypothesis that indels are more - likely to be causal.

formother children in our having model, a FLG we identifiedmutation carry a sig- ingnificantly mother increased which was eczema independent risk (RR, of 1.55) the - ternal effect was replicated in another 450 Europeanchildren’s families genotype. and Importantly, it was not observed this ma Selected Publications for the paternal genotype. Our study re- The EArly Genetics and Lifecourse Epidemiology (EAGLE) Eczema Consortium [MDC

F, Bauerfeind A, Hübner N, Lee YA] Multi-ancestry genome-wide association study forvealed, a common for the disease. first time, The a maternal direct influence FLG ef- contribution: Esparza-Gordillo J, Marenholz I, Kalb B, Matanovic A, Rüschendorf of a maternal mutation on the child’s risk Nature Genetics epub ahead of print. 2015. fect was only found when the mothers were of 21,000 cases and 95,000 controls identifies new risk loci for atopic dermatitis. allergic, and was absent in families of non- BD, Baurecht H, Margaritte-Jeannin P, Saaf A, Kerkhof M, Ege M, Baltic S, Matheson MarenholzMC, Li J, Michel I, Esparza-Gordillo S, Ang WQ, McArdle J, Ruschendorf W, Arnold F, A, Bauerfeind Homuth G, A, Demenais Strachan F, DP, Bouzigon Spycher FLG-induced changes in the maternal im- E, Soderhall C, Pershagen G, de Jongste JC, Postma DS, Braun-Fahrlander C, Horak allergic mothers. Our finding suggests that E, Ogorodova LM, Puzyrev VP, Bragina EY, Hudson TJ, Morin C, Duffy DL, Marks GB, Robertson CF, Montgomery GW, Musk B, Thompson PJ, Martin NG, James A, Sleiman system during pregnancy and increase the P, Toskala E, Rodriguez E, Folster-Holst R, Franke A, Lieb W, Gieger C, Heinzmann A, mune response shape the child’s immune Rietschel E, Keil T, Cichon S, Nothen MM, Pennell CE, Sly PD, Schmidt CO, Mata- that maternal FLG mutations act as strong novic A, Schneider V, Heinig M, Hubner N, Holt PG, Lau S, Kabesch M, Weidinger S, Hakonarson H, Ferreira MA, Laprise C, Freidin MB, Genuneit J, Koppelman GH, Melen environmentalchild’s risk for eczema.risk factors Our study for theindicates child and highlights the potential of family-based loci involved in the atopic march. Nat Commun E, Dizier MH, Henderson AJ, and Lee YA. Meta-analysis identifies seven susceptibility studies in uncovering novel disease mecha- 6, 8804. 2015. nisms in medical genetics. Lee-Kirsch MA, Nordenskjold M, Winge MC, Keil T, Kruger R, Lau S, Beyer K, Kalb Esparza-Gordillo J, Matanovic A, Marenholz I, Bauerfeind A, Rohde K, Nemat K, mutations increase the risk of atopic dermatitis in children: an effect independent of Functional analysis of disease B,mutation Niggemann inheritance. B, Hubner PLoS N, CordellGenet 11[3],HJ, Bradley e1005076. M, and 2015. Lee YA. Maternal filaggrin genes Gunther C, Kind B, Reijns MA, Berndt N, Martinez-Bueno M, Wolf C, Tungler V, Chara O, Lee YA, Hubner N, Bicknell L, Blum S, Krug C, Schmidt F, Kretschmer S, Koss S, A major goal is to characterize the effects Astell KR, Ramantani G, Bauerfeind A, Morris DL, Cunninghame Graham DS, Bubeck of the susceptibility genes for allergic and D,Alexopoulou Leitch A, Ralston D, Conrad SH, K,Blackburn Dahl A, Roers EA, Gahr A, Alarcon-Riquelme M, Witte T, Vyse TJ, ME, Melchers Jackson I, AP, Mangold and to better understand the earliest events in E,Lee-Kirsch Nothen MM, MA. Aringer Defective M, removal Kuhn A, ofLuthke ribonucleotides K, Unger L, fromBley DNAA, Lorenzi promotes A, Isaacs systemic JD, autoimmunity. J Clin Invest 125[1], 413-424. 2015. humanchronic physiologyinflammatory that diseases lead to (4)this in type order of Huang J, Chen J, Esparza-Gordillo J, Ding J, Elder JT, Abecasis GR, Lee YA, Mark LG,

expression data from relevant target tis- Nat Commun suesinflammatory to characterize response. the Wemolecular integrate mecha gene-- Moffatt MF, Cookson WO, and Liang L. eQTL mapping identifies insertion- and deletion-specific eQTLs in multiple tissues. 6, 6821. 2015. order to identify the functional effect of Group Leader PhD students smallnisms structural underlying variants disease (short associations. insertions In Prof. Dr. Young-Ae Lee Sarah Grosche or deletions, indels) within the human ge- Anja Matanovic nome, the association of these variants with Till-Yong Mohr gene expression levels (expression quanti- Scientists tative trait loci, eQTLs) from multiple tis- Dr. Jorge Esparza Gordillo Technical Assistants sues was analyzed. Previous studies have Dr. med. Birgit Kalb Christina Flachmeier demonstrated that SNPs with an effect on Theresa Thuß gene expression were enriched in introns Dr. Ingo Marenholz

MDC Research Report 2016 31 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Martin Lohse

Receptor signalling

Our group focuses on the mechanisms of Current Projects activation and signalling by of G-protein Spatio-temporal Patterns of coupled receptors (GPCR), and subsequent GPCR Activation downstream events which lead to alterations in cell function. GPCR activation is triggered by binding of G protein-coupled receptors (GPCRs) are a a ligand (e.g. adrenaline, glutamate) to a large family of approximately 800 members, by a conformational change in the recep- which include receptors for light, taste and tor,specific which site enables of a receptor. it to bind This to and is followed activate smell, but also many receptors for transmit- G-proteins. We have developed methods to - ters and hormones. They are also the most rescence microscopy, and we aim to study important group of targets for drugs, for wheremonitor and these when signalling in a cell these events signals via fluo oc- example beta blockers, antihistamines and cur. We use techniques to trigger receptor opiates. All GPCRs share common struc- activation by light to monitor signalling with sub-millisecond resolution. tural features, presumably also common We are also working on techniques to adapt activation mechanisms, and they all signal such measurements to high throughput via heterotrimeric G proteins (Gαβγ) and – at screening in order to search for compounds with unusual signalling properties, which later points upon activation, via β-arrestins. might make them new classes of drugs with We use a variety of biochemical and phar- only a subset of effects compared to con- macological methods as well as new micro- ventional drugs. scopic techniques to study the mechanisms of their function and their potential to serve Spatial Patterns of cAMP Signalling as targets for innovative drug therapies. cAMP is a second messenger downstream of GPCR activation, which plays an important role in many physiological processes, ranging form regulation of heartbeat and force to synaptic plasticity. Although cAMP appears to be a freely diffusible substance, recent evidence suggests that it may act in cells in a highly localized manner. We are at- tempting to visualize such local effects and are searching for mechanisms that might lead to the postulated nano-compartments. To do so, we use models ranging from the -

mousedesign andmodels, expression where ofwe specific attempt cAMP to image sen localsor proteins cAMP signals to studies (Figure in transgenic 1). fly and

32 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1 Imaging of cAMP in Drosophila larvae. Shown is the end of a motor neuron from a transgenic fly expressing a cAMP sensor. cAMP levels are indicated as a graded color response derived from fluorescence resonance energy transfer (FRET). Adapted from Maiellaro et al. Cell Rep. 2016.

Signalling by β-arrestins Dynamics of GPCR Oligomerization

Ever since their discovery more than two GPCRs can form supramolecular complexes (i.e. di-/oligomers) on the cell surface. How- with new functions and properties. Origi- ever, their size, nature and dynamics as well nallydecades thought ago, β-arrestins to represent have just surprised inhibitors us as their physiological and pharmacological of receptor signalling, they have now been implications are still largely unknown. We focus on investigating the dimerization of own right. When they bind to receptors, receptors (especially adrenergic, opiate theyidentified terminate as signalling receptor signalling proteins to in G theirpro- and chemokine) on the surface of intact teins and at the same time trigger a number cells as well as evaluating the role of dimer- of subsequent steps: receptor internaliza- ization on receptor activation and down- tion and activation of downstream protein stream signalling. We do so by a variety of kinases. Recent data indicate that during biochemical and microscopic techniques, - notably brightness analyses and single par- tion/deactivation cycle (Figure 2). ticle microscopy. this process β-arrestins undergo an activa

MDC Research Report 2016 33 CARDIOVASCULAR AND METABOLIC DISEASE

Mechanisms of heart failure Outlook

GPCRs are important for a plethora of Our group is in the process of translocat- physiological functions, one of them being from the University of Würzburg to the ing the regulation of cardiac frequency MDC, and will focus initially on the spatio- and contractility. We are interested in the temporal aspects of GPCR signalling and biochemical mechanisms that underlie the their analysis by advanced microscopy. Se- development of heart failure, and in poten- nior members of the team are leaving to as- tial strategies to combat this sequence of sume leading positions elsewhere: Kristina events. We do so by studying heart failure Lorenz has become director at the Leibniz in mouse models and search for transgenic - as well as pharmacological strategies for its mund, while Carsten Hoffmann and Davide treatment and prevention (Figure 3). CalebiroInstitute will of assume Analytical chairs Sciences at the universi in Dort- ties of Jena and Birmingham.

Figure 2 An activation/deactivation cycle of β-arrestin. Upon binding to an active phosphorylated receptor (1) β-arrestin adopts an activated conformation (2) that might facilitate fitting to the activated, phosphorylated receptor surface. Parts of C- and/or N-domain undergo further movements (3) to bring β-arrestin into a receptor-specific activated conformation. Upon agonist removal, β-arrestin dissociates from the receptor (4) and remains active for some time (5) before its active state is reversed (6). Adapted from Nuber et al. Nature 2016.

34 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 3 RKIP is a protein that protects the heart. Shown is the survival of several mouse lines after cardiac stress (induced by a high blood pressure). Mice transgenic for RKIP in the heart (RKIP-tg) survive better, and mice lacking RKIP (RKIP-/-) survive less than controls. Adapted from Schmid et al. Nature Medicine, 2015.

Selected Recent Publications

Cell Reports Maiellaro I, Lohse MJ, Kittel RJ, Calebiro D. cAMP Signals in Drosophila Motor Neurons dependentAre Confined activation/deactivation to Single Synaptic Boutons. cycle. Nature . 2016;17:1238-1246. Nuber S, Zabel U, Lorenz K, et al. β-Arrestin biosensors reveal a rapid, receptor- Nat.. 2016;531:661-664. Medicine LohseSchmid MJ, E, HoffmannNeef S, Berlin KP. SpatialC, et al. and(2015) temporal Cardiac aspects RKIP inducesof signaling a beneficial by G-protein- coupledβ-adrenoceptor-dependent receptors. Mol Pharmacol positive inotropy. . 2015;21:1298-1306 Calebiro D, Hannawacker A, Lyga S, et al. Mutations of PKA catalytic subunit in adrenocortical Cushing´s adenomas. 2015;88(3):572-578. impair association with the regulatory subunit. Nature Comm

. 2014;5:5680. Start of the Group: Januar 2017 PhD Student Selma Anton Group Leader Yevgenii Grushevskyi Prof. Dr. Martin Lohse Jan Möller Scientists HannesAli Isbilir Schihada Dr. Paolo Annibale Dr. Andreas Bock Technical Assistant Ruth Pareja Alcaraz Dr. Jean-Yves Tano Victoria Prochnov Dr. Isabella Maiellaro

MDC Research Report 2016 35 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Friedrich Luft

Hypertension Genetics, Regulators, and Environment

Gene laboratory Our goal is to reduce cardiovascular mor- analyses of mesenchymal stem cell-derived bidity and mortality. Hypertension is the vascularfore age 50smooth years muscle when untreated.cells (VSMCs) In vitro and most important risk factor. We investigate chondrocytes provided insights into molec- genetic, mechanistic, regulatory, and envi- ular pathogenesis. The mutations increased protein kinase A-mediated PDE3A phos- pursued a Mendelian form of hypertension phorylation and resulted in gain of func- thatronmental cosegregates influences. with Since brachydactyly 1994, we havetype tion, with increased cAMP-hydrolytic activ- E (HTNB). We recently reported six mis- ity and enhanced cell proliferation. Levels sense mutations in PDE3A (encoding phos- of phosphorylated VASP (vasodilator-stim- phodiesterase 3A) in six unrelated families ulated phosphoprotein) were diminished, with HTNB. The syndrome features brachy- and PTHrP (parathyroid hormone-related dactyly type E (BDE), severe salt-indepen- peptide) levels were dysregulated. We sug- dent but age-dependent hypertension, an - - tions cause the syndrome. VSMC-expressed cular contact at the rostral-ventrolateral PDE3Agested deserves that the scrutiny identified as PDE3Aa therapeutic muta increased fibroblast growth rate, neurovas- target for the treatment of hypertension. sure regulation, and death from stroke be- (Figure 1) medulla, altered baroreflex blood pres

Fig. 1: (a) Peptide sequence encoded by PDE3A exon 4 that harbors the mutations (which alter amino acids 445, 447, 449) and phosphorylation sites Ser428 and Ser438 near the altered residues. Mutations cause a gain-of-function effect.(b) Mild BDE of VI/9 compared tomore severe BDE of cousin VI/1.(c) Haplotype analysis identified a new recombination event.(d) Chondrogenic Pde3a expression in developing mice. PTHLH expression exhibits a downregulation responsible for BDE.

36 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

We also conducted clinical studies of vas- Eicosanoid regulators cular function, cardiac functional imaging, Cytochrome P450 (CYP)-dependent me- platelet function in affected and nonaf- tabolites of arachidonic acid (AA) con- fected persons, and performed cell-based tribute to the regulation of cardiovascular assays. Large-vessel and cardiac functions function. CYP enzymes also accept eicosa- seemed to be preserved in HTNB. The pentaenoic acid (EPA) and docosahexae- platelet studies showed normal platelet noic acid (DHA) to yield more potent va- function. Cell-based studies demonstrated sodilatory and potentially anti-arrhythmic that available phosphodiesterase 3A inhibi- metabolites, suggesting that the endog- tors suppressed the mutant isoforms, but - less effective. However, increasing cGMP vorably shifted by dietary omega-3 fatty to indirectly inhibit the enzyme seemed acids.enous To CYP-eicosanoid test this hypothesis, profile we can treated be fa to have particular utility. Our results shed 20 healthy volunteers with an EPA/DHA more light on phosphodiesterase 3A activa- supplement and analyzed for concomi- tion and could be relevant to the treatment tant changes in the circulatory and uri- of severe hypertension in the general popu- nary levels of AA-, EPA-, and DHA-derived lation. metabolites produced by the cyclooxy-

Fig. 2: A synthetic analog of omega-3 fatty acid-derived epoxyeicosanoids protects against cardiac arrhythmia in a rat model of myocardial infarction (MI). The animals (n= 9-10 per group) received the analog or its vehicle two hours before inducing MI by coronary artery ligation. The compound reduced both the duration and severity of MI-elicited spontaneous ventricular arrhythmias as monitored by continuous ECG recording.

MDC Research Report 2016 37 CARDIOVASCULAR AND METABOLIC DISEASE

collection for sodium excretion (UNaV) dependent pathways. Raising the Omega-3 genase-, lipoxygenase (LOX)-, and CYP-- ±25 mmol deviation from the average dif- ily resulted in a large increase of EPA-de- ferenceshowed infradianbetween rhythmicity.recorded sodium We defined intake a rivedIndex fromCYP-dependent about four epoxy-metabolites to eight primar and UNaV as the prediction interval to ac- followed by increases of EPA- and DHA- curately classify a 3-g difference in salt intake. Because of the biological variability in metabolites including the precursors of UNaV, only every other daily urine sample derived LOX-dependent monohydroxy- themselves were not detected. The me- - tabolite/precursorthe resolvin E and fatty D families; acid ratios resolvins indi- tionscorrectly to 3 classified consecutive a 3-g 24-hour difference collections in salt cated that CYP epoxygenases metabolized intake (49%). By increasing the observa-

24-hourand sodium urines intakes, and sodium classification intake samples accu AA.EPA Effectswith an on 8.6-fold leukotriene, higher efficiencyprostaglandin and racy improved to 75%. Collecting seven E,DHA prostacyclin, with a 2.2-fold and higherthromboxane efficiency forma than- We conclude that single 24-hour urine col- tion remained rather weak. We propose lectionsimproved at classificationintakes ranging accuracy from 6 toto 92%.12 g that CYP-dependent epoxy-metabolites of salt per day were not suitable to detect a EPA and DHA may function as mediators 3-g difference in individual salt intake. Re- of the vasodilatory and cardioprotective peated measurements of 24-hour UNaV im- effects of omega-3 fatty acids and could prove precision. This knowledge could be serve as biomarkers in clinical studies in- relevant to patient care and the conduct of vestigating the cardiovascular effects of intervention trials. EPA/DHA supplementation. We collaborate with John R. Falck (UT Southwestern, Dallas) in developing CYP-eicosanoid- derived pharmacological approaches to

Therapeutics is a spin-off company of the MDCprevent on cardiovascularthe BioTech-Campus disease. Berlin-Buch OMEICOS (http://www.omeicos.com) that resulted from this work. (Figure 2)

Environment/salt intake Dietary salt intake is said to be responsible for salt-sensitive hypertension. We found earlier that sodium follows a three-compartment, rather than a two-compartment model in mammals. The ion is bound to negatively charged proteoglycans that are widely distributed, but largely reside in

9, and 12 g/day salt intakes, we found that sodiumskin. In twois excreted very long according balance studiesto circasep at 6,- tan rhythms regulated by aldosterone. Our 2013 report in Cell Metab is in strong run-

ning for “cover of the decade”. of recorded intake, indicating long-term steady-stateUrinary recovery sodium of dietarybalance saltin both was stud 92%-

ies. Even at fixed salt intake, 24-hour urine 38 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

We believe there are teleologically sound reasons why sodium might be stored in skin. The microdomain interface between stored sodium bound to proteoglycans is hypertonic

regulate a hypertonic microenvironment in to plasma and interstitial fluid. Immune cells of increased skin sodium concentrations is unknown.the skin; however, We found the that biological sodium advantage accumu- lated at the site of bacterial skin infections in humans and in mice. We used the protozoan parasite Leishmania major as a model of skin-prone macrophage infection to test the Selected Publications hypothesis that skin-sodium storage facilitates antimicrobial host defense. Activation Maass PG, Aydin A, Luft FC, Schächterle C, Weise A, Stricker S, Lindschau C, Vaegler of macrophages in the presence of high NaCl M,Bialer Qadri MG, F, WienkerToka HR, TF,Schulz Ott J,H, Rittscher Krawitz K,PM, Liehr Parkhomchuk T, Jordan J, D, Plessis Hecht G, J, TankHollfinger J, Mai I,K, Wefeld-Neuenfeld Naraghi R, Hodge Y,R, Bartels-KleinHopp M, Hattenbach E, Mühl LO,A, Kann Busjahn M, Schuster A, Rauch H, A, Chitayat Vandeput D, F, - Gong M, Rüschendorf F, Hübner N, Haller H, Mundlos S, Bilginturan N, Movsesian concentrations modified epigenetic markers MA, Klussmann E, Toka O, Bähring S. PDE3A mutations cause autosomal dominant factorand enhanced of activated p38 T mitogen-activatedcells 5 (NFAT5) activa pro- tion.tein kinase This high-salt (p38/MAPK)-dependent response resulted nuclear in el- hypertension with brachydactyly. Nat Genet 2015;47:647-53. evated type-2 nitric oxide synthase (Nos2)- TokaA, Gnoth O, Tank S, Mühl J, Schächterle A, Toka HR, C, Wefeld-Neuenfeld Aydin A, Maass PG, Y, Elitok Utz W, S, Töpper Bartels-Klein A, Jordan E, HollfingerJ, dependent nitric oxide (NO) production and I,Schulz-Menger Lindschau C, MaiJ, Klussmann K, Boschmann E, Bähring M, Rahn S, Luft G, Movsesian FC. Clinical MA, effects Müller of phosphodiesT, Doescher - improved Leishmania major control. Finally, terase 3A mutations in inherited hypertension with brachydactyly. Hypertension. we found that increasing sodium content in Fischer R, Konkel A, Mehling H, Blossey K, Gapelyuk A, Wessel N, von Schacky C, the skin by a high-salt diet boosted activa- Dechend2015;66:800-8. R, Muller DN, Rothe M, Luft FC, Weylandt K, Schunck WH. Dietary omega-3 tion of macrophages in an Nfat5-dependent - manner and promoted cutaneous antimi- Lerchlfatty acids K, Rakova modulate N, Dahlmann the eicosanoid A, Rauh profile M, Goller in man U, primarilyBasner M, via Dinges the CYP-epoxygen DF, Beck L, crobial defense. We suggested that the hy- ase pathway. J Lipid Res. 2014;55:1150-1164. - pertonic microenvironment could serve as a ken J, Kirsch K, Johannes B, Krannich A, Luft FC, Titze J. Agreement between 24-hour barrier to infection. Agureevsalt ingestion A, Larina and sodiumI, Baranov excretion V, Morukov in a controlledB, Eckardt environment.KU, Vassilieva Hypertension. G, Wabel P, Vien

- Perspectives born2015;66:850-7. D, Linz P, Binger KJ, Gebhardt M, Heinig M, Neubert P, Fischer F, Teufel S, David We will reproduce HTNB in mouse and rat Jantsch J, Schatz V, Friedrich D, Schröder A, Kopp C, Siegert I, Maronna A, Wendel models and will pursue cell-based studies Schuler G, Uder M, Bogdan C, Luft FC, Titze J. Cutaneous Na+ storage strengthens JP,the Neufert antimicrobial C, Cavallaro barrier A, functionRakova N, of Küper the skin C, Beckand boosts FX, Neuhofer macrophage-driven W, Muller DN, host to elucidate the function of PDE3A further. This work will be performed in cooperation with Enno Klussmann and his laboratory defense. Cell Metab. 2015;21:493-501. Group Leader Eicosanoids - Prof. Dr. Friedrich Luft Dr. Wolf-Hagen Schunck (Head searchat the MDC.group Wolf-Hagenthat had the Schunck’s courage groupto es- tablishis the firstan independent (and only) company. clinically This based com re- Gene Lab pany is now developing synthetic analogs PD Dr. Sylvia Bähring (Director Dr.eicosanoid Christina laboratory; Westphal Board of of omega-3 fatty acid-derived epoxyeico- gene laboratory) MaximilianOMEICOS Therapeutics) Blum sanoids for the prevention and treatment Dr. Atakan Aydin Ramona Zummach of cardiovascular diseases including atrial Dr. Philipp Maaß (Currently Christel Andrée

at Vanderbilt University, where Friedrich C. Yvette Wefeld-Neuenfeld Doctoral students Luftfibrillation. also has In an conjunction adjunct appointment, with Jens Titze we AstridBroad InstituteMühl Boston) will pursue novel avenues of sodium me- Dr. Birgit Meußer, PhD Mirjam Karber (BIH fellowship) sodium-based effects on metabolism. tabolism including new findings regarding Rosana Molé Illas

MDC Research Report 2016 39 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Ingo Morano

Molecular Muscle Physiology

We are studying the functional roles of sub- with mouse VLC-1 (mVLC-1) prepared from units of key proteins of Ca2+-handling and C57BL/6. Additionally, we could show that max. left ventricular pressure development force generation in muscles. Further, our lab of isolated perfused hearts from TgMhVLC-1 aims to understand the effects and mechanisms of sex and estrogen in the heart in TgME56G were significantly higher than hearts from more detail. Particularly, we are screening that ELCs decreased myosin stiffness, in vitro motility, or C57BL/6. and thereby These cardiacfindings contrac showed- new protein binding factors of estrogen tion. receptor and determine their effects on estrogen receptor actions in the heart. The sticky N-terminus of ELC binds to a cluster of acidic residues at the C-terminus of actin. We could show that although both Structural and functional roles of hALC-1 and hVLC-1 have similar secondary the myosin essential light chain - (ELC) in the cardiac contraction Two myosin heavy chains (MyHC) and four structure, but the actin affinity of the N-ter non-covalently linked myosin light chains, provideminus of a hALC-1lower mechanical was significantly load placed weaker on two essential myosin light chains (ELC) thethan cross-bridges actin affinity than of hVLC-1,the hVLC-1, which thus may al- and two regulatory light chains (RLC), form lowing a higher maximal shortening velocity. Hence, differential expression of ELC drives muscle contraction (Figure 1). The isoforms could modulate muscle contrac- primarythe native structure Type II myosinof cardiac molecule, ELC isoforms which [ALC-1 (atrial ELC) and VLC-1 (ventricular Re-expression of the ALC-1 in the hyper- ELC)] presents with an elongated N-termi- trophiedtile activity human via distinct ventricle, α-actin therefore, interactions. seems nal, and a dumbbell-like C-terminal domain to be an auto-regulatory mechanism of the ELCs bind with the utmost lysine-rich N- human heart to adapt to an increased work terminus to actin and with their C-terminal demand. domain to the myosin lever arm, RLC and myosin motor domain (Figure 1). ELCs are Human ALC-1 as a co-repressor involved in modulation of force generation of ERα-mediated transcription in of myosin motors and cardiac contraction, human heart while its mechanism of action remained We showed that E2 attenuated hALC-1 ex- elusive. To assess whether ELC modulates pression in human atrial tissues of both myosin stiffness, and thus its force produc- sexes and in AC16 cells (a human ventricu- tion and subsequently the cardiac contrac- lar cell line). E2 induced the nuclear trans- tion, we generated heterologous transgenic mouse (TgM) strains with cardiomyocyte- and hALC-1 in AC16 cells, where they co- operativelylocation of estrogenregulate receptor the transcriptional alpha (ERα) VLC-1 (TgMhVLC-1) or E56G-mutated hVLC- activity of hALC-1 promoter. E2-activated 1specific (hVLC-1 expressionE56G E56G of). ELCOur data with showed human that stiffness and actin sliding velocity of of hALC-1 promoter. This inhibitory effect myosin from ; TgM TgMhVLC-1 ERα reduces the transcriptional activity- higher than myosin from TgME56G or myosin ence of hALC-1, and thus, hALC-1 acts as were significantly was significantly potentiated in the pres

40 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1: Schematic interaction interfaces of myosin essential light chain (ELC). The interaction domains of ELC (green) with the MyHC, actin, and regulatory light chain (RLC) (red double arrows).

- Effects of sex and 17β-Estradiol on nuclear translocation of the interactsa co-repressor physically of ERα-mediated with hALC-1 transcrip in the cardiac L-type calcium channel presencetion. Additionally, of E2. As wea further showed novel that effect, ERα C-terminus we showed that chronic E2-treatment of The cardiac L-type Ca2+ channel (Cav1.2) adult mouse cardiomyocytes overexpress- constitutes the main entrance gate for Ca2+ ing hALC-1 resulted in reduced cell-short- that triggers cardiac contraction. The distal ening amplitude and twitching kinetics of these cells independent of Ca2+ activation - shuttlesC-terminus between fragment the of plasmaCav1.2 α1Cmembrane subunit ment counteracts the hALC-1-mediated and(α1C-dCT) the nucleus, is proteolytically which is regulated cleaved both andde- improvementlevels. Our findings of contractile show that anfunction E2 treat in velopmentally and by Ca2+. We could show mouse cardiomyocytes. This could be of importance for the compensation process in overloaded hearts and preservation of infor athe sex-dependent first time that manner. the nuclear The shuttlingratio of cardiac function. of α1C-dCT in cardiomyocytes is regulated

nuclear to cytosolic intensity of α1C-dCT MDC Research Report 2016 41 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 2: Pronounced mitochondrial remodelling in female WT mice after VCR. A) Representative electron micrographs of mitochondria; X11.700 magnification; scale bar: 2µm. B) Mitochondrial number per unit area is significantly increased after VCR in both sexes. (C) Both sexes show a significant increase in small mitochondria, but (D) only in female mice, large mitochondria are significantly reduced. n = 4–6 / group and sex. 2-way ANOVA: *P ≤ 0.05; ***P ≤ 0.001. VCR: voluntary cage wheel running; sed: sedentary. Small mitochondria: ≤0.5 µm2, large mitochondria: ≥1 µm2.

nuc cyt Effects of sex on cardiac female cardiomyocytes compared to male mitochondrial remodeling cardiomyocytes.(I /I ) was significantly Furthermore, higher we in isolatedfound a - - ences exist in exercise-induced physiologi- calIn a myocardialrecent study, hypertrophy we showed that(MH). sex Female differ cardiomyocytessignificant decrease upon in nuclearserum withdrawal.staining in mice showed a higher increase in cardiac tensity of α1C-dCT in both female and male mass compared with males, independent of forced or voluntary exercise character. Interestingly, subsequent E2 treatment for- During exercise-induced physiological MH, cytes,8h and but 45min not normalizedin female thecardiomyocytes. intracellular mitochondrial adaptations are needed to distribution of α1C-dCT in male cardiomyo cope with the heart‘s increased energy dean ER-inhibitor, indicating the involve- mand, which can be achieved by increased mentThis effect of ER of in E2 this was signaling revised bypathway. ICI 182780, This mitochondrial biogenesis or mitochondrial remodelling. We hypothesized that sex dif- explain, at least partly, the observed sex differences in mitochondrial adaptation could ferencesidentified in novel the regulation signaling of mechanism cardiac Cav1.2 may contribute to a sexual dimorphism in physi- channel activity.

ological MH. Indeed, we showed that the 42 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

expression level of mitochondrial key modulators, such as NRF-1, NRF-2, Mef2a and

female WT mice after voluntary cage wheel runningAtp5k are (VCR) significantly (Figure 2). increased Furthermore, only our in data showed that although the number of Selected Publications 2) in both sexes increased, the number of large mito- Duft K, Schanz M, Pham H, Abdelwahab A, Schriever C, Kararigas G, Dworatzek E, Davidson MM, Regitz-Zagrosek V, Morano I, Mahmoodzadeh S smaller mitochondria (≤0.5 µm - . 17β-Estradiol- chondria (≥1 µm2), however, decreased sig inducedEpub 2016 interaction Nov 11. of estrogen receptor α and human atrial essential myosin light chain modulates cardiac contractile function. Basic Res Cardiol. 2017; 112(1):1. nificantly only in female WT mice after VCR, Mahmoodzadeh S, Haase H, Sporbert A, Rharass T, Panáková D, Morano I. Nuclear reduction in Mfn-2 protein only in females translocation of the cardiac L-type calcium channel C-terminus is regulated by sex (Figureand this 2). was These in linedata with indicate the significantthat exercise increases mitochondrial number and Mahmoodzadeh S*, Dworatzek E*, Jaisser F, Messaoudi S, Morano I, remodelling in both sexes, with a shift from and 17β-estradiol. J Mol Cell Cardiol. 2016; 97:226-34. Schuster I*, large towards smaller mitochondria only in Regitz-Zagrosek V. Cardiomyocyte-specific overexpression of oestrogen receptor female VCR WT mice. Ongoing studies will β improves survival and cardiac function after myocardialWoischwill infarction in C female, Bornstein and clarify in which extent E2/ER are involved male mice.Morano Clin Sci I (Lond). 2016; 130(5):365-76. in sexually dimorphic exercise-induced car- Lamounier-Zepter V, Look C, Schunck WH, Schlottmann I, SR, Xu A, . Interaction of epoxyeicosatrienoic acids and adipocyte fatty diac mitochondrial remodeling. acid-binding protein in the modulationMahmoodzadeh of cardiomyocyte S, Neef contractility. K, Albrecht Int J, JBurkert Obes K, Oliverio(Lond). 2015; M, Nguemo 39(5):755-61. F, Choi YH, Neiss WF, Morano I, Hescheler J, Stamm C. Suscep- The functional role of the ahnak Brodaractibility of murineA, Šarić inducedT, Oberwallner pluripotent B, stem cell-derived cardiomyocytes to hypoxia protein family in skeletal muscle Kusch A, Schmidt M, Gürgen D, Postpieszala D, Catar R, Hegner B, Davidson MM, fibers Mahmoodzadehand nutrient deprivation. S, Dragun Stem D. 17ß-Estradiol Cell Res Ther. regulates2015; 23;6:83. mTORC2 sensitivity to ra- - colemma and T-tubuli of cardiomyocytes Petrov G, Dworatzek E, Schulze TM, Dandel M, Kararigas G, Mahmoodzadeh S, pamycin in adaptive cardiac remodeling. PLoS One. 2015; 10(4):e0123385. indirectlyIn the heart, associated ahnak1 withis located the voltage- at the sarde- Knosalla C, Hetzer R, Regitz-Zagrosek V. Maladaptive remodeling is associated with impaired survival in women but not in men after aortic valve replacement. JACC pendent L-type Ca2+ channel (L-VDCC) via

L-VDCC current. AHNAK function and sub- Cardiovasc Imaging. 2014; 7(11):1073-80. cellularits β2-subunit localization (ß2), in where skeletal it muscle stimulates are Group Leader unclear. Co-localization of ahnak1 and Prof. Dr. Ingo Morano ahnak2 with vinculin clearly demonstrates that both proteins are components of the Scientists Laura van Vuuren Dr. Shokoufeh Mahmoodzadeh Clara Ortegon expression was detected in the T-tubule Dr. Frank Suhr Oriane Larcheveque system.costameric A laser network. wounding In contrast, assay withno ahnak ahn- Dr. Christiane Woischwill Monica Akram Kamal Ghobrial Dr. Ahmed Kolade Oloyo Mariam Gamal Mohamed El Said is not involved in membrane repair. Using Dr. Hannelore Haase atomicak1-deficient force fibersmicroscopy suggests (AFM), that we ahnak1 ob- Technical Assistants PhD student Petra Domaing Karolin Duft Carolin Gärtner suggestserved anovel significantly functions higherof ahnak transverse proteins Cindy Schriever Karin Karczewski stiffness of ahnak1-/- fibers. These findings Anna-Maria Ströhl-Sattler muscle biopsy specimens obtained from pa- Graduate and Steffen Lutter tientsin skeletal with limb muscle. girdle Interestingly, muscular dystrophy human Undergraduate student showed that ahnak1 lost its sarcolemmal Katharina Sarah Koch Secretariat localization and appeared in the extracel- Ahmed Abdelwahab Manuela Kaada lular space in both muscular dystrophies.

MDC Research Report 2016 43 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Thoralf Niendorf

Ultrahigh Field Magnetic Resonance (UHF-MR)

Our research concentrates on the develop- imaging (BIOQIC) and the European Ultra- ment of MR technology enabling new ways high-Field Imaging Network in Neurodegen- of mapping/probing morphology, function, erative Diseases (EUFIND). Our initiatives physiology and metabolism together with resulted in 65 peer-reviewed publications explorations into the benefits of ultrahigh during 2014-16, 70 abstract contributions field (UHF) MR. These efforts are designed to the 2014/16 meetings of the International to characterize (patho)physiological and bio- Society of Magnetic Resonance in Medicine physical processes to harmonize basic re- plus one US patent. Thoralf Niendorf recent- search with clinical science. To meet this goal ly received an ERC advanced grant by the small animal (9.4 T) and whole-body human European Research Council for his project (7.0T & 3.0T) MR are advanced. Our research ThermalMR: A New Instrument to Define the is built on collaborations with the Charité, Role of Temperature in Biological Systems the PTB, Siemens, Bruker Biospin and other and Disease for Diagnosis and Therapy (inter)national institutions. The group forms which eloquently speaks about the excel- a key/integral part of major research initia- lence of our science and the translational tives including the Helmholtz Alliance for and cross-domain nature of our collabora- Imaging and Curing Environmental Diseases tions. (ICEMED), the population imaging program of the German National Cohort (GNC) (1), the Technology Focus: DFG research group FOR 1368 on Hemody- Enabling Methodology for UHF-MR namics of Acute Kidney Injury, the imaging The sensitivity gain afforded by UHF-MR is program of the German Center for Cardio- the driving force behind our technological developments. We pioneered multi-channel ra- vascular Research (DZHK), the individualized diofrequency (RF) coil technology including Medicine (iMED) initiative of the Helmholtz the development and clinical evaluation of 6-, Association, the BMBF project on targeted tailored for clinical cardiovascular, brain and radiofrequency heating (THERAHEAT), the 12-, 16- and 32 channel(2) at RF7.0 coil T (Figureconfigurations 1). To BMBF funded validation project on the diag- stay at the forefront of research we developed nosis of renal diseases (renalMROXY), the aophthalmic modular 16 MRI channel bow tie antenna RF coil German Ultrahigh Field Imaging (GUFI) net- array (3) (Figure 1). Going beyond conventional proton (1 work, the EU COST Action PARENCHIMA on renal imaging, the DFG funded project on better insight intoH) MRmetabolic we are and exploring (nano)mo MRI- Fluorine MR technologies to study cellular lecularof sodium, processes. fluorine, 23 Na,and 19 otherF RF coil nuclei arrays to gain (Fig a- therapies, the Marie Sklodowska Curie project ure 1) designed for (pre)clinical applications were developed to form the basis for transla- SODIUMMRI-4-EU (EU project 752489), the tional research. We were able to demonstrate DFG funded graduate school on quantitative the feasibility of sodium MR imaging of the

44 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 1: Enabling Methodology for UHF-MR RF coil technology including (a) 16-channel, (b) 6-channel and (c) 8-channel transceiver arrays designed, and clinically evaluated by our group which were put to clinical use for (d) high fidelity imaging of the heart, (e) for ophthalmic imaging and (f) for in vivo 19F MRT of fluorinated drugs.

Figure 2: Next Generation of MR Technology and Applications Our group explores innovative MR applications that are not clinical mainstream yet. We developed a novel instrument for Thermal Magnetic Resonance which supports targeted RF heating at 7.0 T (a,b). We extended our simulations to a 20 T class MR system (1 GHz) (c,d). Assessment of MR safety of passive conductive implants including stents constitutes another research focus of our group (e-h).

human heart and showed proof-of-principle strument provides a revolutionary method for high spatial resolution in vivo sodium im- for focal, in vivo temperature manipulations aging of the human skin. Our group was the (Figure 2). To master the physics of even 19 - the in vivo study of pharmacokinetics of 19F- first that facilitated human F MRI at 7.0 T for higher magnetic fields we devoted our re containing drugs (Figure 1). (EMF) simulations up to B0=23 T (1GHz) (4). searchMR safety to is numerical also one of electromagnetic our focuses, to fieldgain The group also focuses on developing novel a better insight into RF heating of stents and MR applications. These efforts include our other medical devices, which is of paramount push towards a Thermal Magnetic Resonance relevance for driving UHF-MR into the clinic (ThermoMR) instrument (4) for the study of (Figure 2). the role of temperature in biological systems and disease. For this project, Thoralf Niendorf Clinical Focus: Explorations into was rewarded with an ERC advanced grant Cardio- and Neurovascular Diseas- by the European Research Council. Our nu- es Using UHF-MRI merical simulations showed how to adapt an UHF-MR instrument by installing high-density arrays of RF antennae that will permit us applications-orientedThe fields of cardio- andresearch. neurovascular En route MR to to generate heat in highly focused regions of novelare in imaging-guided the spotlight of diagnostic our group’s and clinicalthera- tissue along with temperature mapping and peutic approaches we strive towards the non- diagnostic imaging (Figure 2). This new in- invasive assessment and early diagnosis of

MDC Research Report 2016 45 CARDIOVASCULAR AND METABOLIC DISEASE

Figure 3: UHF-MRI of Cardio- and Neurovascular Diseases TOP: En route to novel diagnostic approaches we strive for high spatial resolution and microstructural imaging of the heart including 1H a) anatomical MRI, b) parametric imaging using MR biomarker and c) sodium MRI. BOTTOM: Examples of clinical studies of neurovascular diseases including d) diagnostic MRI of stroke e) differential diagnosis of neuroinflammatory pathologies and f) early diagnosis of ocular masses.

Figure 4: Quantitative Mapping and (Nano)Molecular Probing TOP: a) Microstructural mapping of mice heart, b) functional MRI of mice brain, c) monitoring of brain tumour growth and d) tracking of nanoparticlelabeled cell migration.BOTTOM: Understanding the mechanisms of acute kidney injury (AKI) is our motivation for changing renal diagnostics with e) anatomic MRI and with f) blood oxygenation level sensitive MRI to (g) monitor interventions and (h) to assess ischemia.

cardiac and neurovascular diseases through Experimental Focus: Quantitative MR-biomarker based tissue characterization, Mapping and (Nano)Molecular quantitative parametric mapping, high spa- Probing tial resolution and microstructural imaging The ultimate aim of the group is to har- of the heart, brain and other target anatomy monize research carried out in the area (Figure 3). We have been able to conduct of preclinical animal imaging with clinical imaging. We are recognized leaders in at 7.0, which showed an increase in the MR in vivo and ex vivo the world’s first cardiac MR patient study relaxation parameter T2* for patients suffer- anatomical, functional and microstructural ing from hypertrophic cardiomyopathy. Our imagingthe field ofof thehigh heart fidelity using cryo-cooled RF research results warrant an early diagnosis coil technology (Figure 4). With this boon of pathologies (e.g. hypertrophic cardiomy- we developed a setup for normothermic opathy), advances diagnostic imaging (e.g. of stroke and small vessel disease, Figure focal thermostimulation for probing noci- 3), secures a differential diagnosis of pathol- ceptionfunctional (Figure brain 4)MRI. Our in achievementsmice during acute sup- ogies (e.g. multiple sclerosis versus orphan ported in vivo and non-invasive monitoring (Figure 3) of tumor expansion (5) (Figure 4). Our and enables emerging clinical applications achievements supported in vivo and non- neuroinflammatory(e.g. diagnosis of ocular diseases masses, Figure 3). invasive monitoring of tumor expansion

46 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

(5) (Figure 4). We made major progress to - genesis of autoimmune encephalomyelitis study brain inflammation during the19 F) patho and physicochemical characteristics of nanopar- (EAE). Our know-how on fluorine ( - tory processes and cell migration in real time withticles greateris being sensitivity leveraged toin followvivo by inflamma boosting

(Figure 4). Gainingcellular a uptake better ofunderstanding fluorinated nanoparticlesof the mecha- Patents / Patent applications

the center of our experimental research radar nisms behind acute kidney injury (AKI) (Fig is in- SelectedUS patent: US Publications 8,842,897 B2 ure 4) and quantitative physiological tech- (01) Bamberg F, Kauczor HU, Weckbach S, Schlett CL, Forsting M, Ladd SC, Greiser screen.niques to We monitor successfully renal hemodynamics combined MRI in an KH, Weber MA, Schulz-Menger J, Niendorf T, Pischon T, Caspers S, Amunts K, Berger K, Bulow R, Hosten N, Hegenscheid K, Kroncke T, Linseisen J, Gunther M, Hirsch JG, now taken to the next level by incorporating Kohn A, Hendel T, Wichmann HE, Schmidt B, Jockel KH, Hoffmann W, Kaaks R, Reiser innovative hybrid setup (MR-PHYSIOL). This is Design, and Technical Background. Radiology 2015:142242. (02)MF, Volzke Paul K, H. Graessl Whole-Body A, Rieger MR J, Imaging Lysiak D, in Huelnhagen the German T,National Winter Cohort:L, Heidemann Rationale, R, goalnear to infrared translate spectroscopy the diagnostic (NIRS) capabilities as part of Lindner T, Hadlich S, Zimpfer A, Pohlmann A, Endemann B, Kruger PC, Langner S, Stachs O, Niendorf T. Diffusion-sensitized ophthalmic magnetic resonance imaging our translational project renalMROXY with the free of geometric distortion at 3.0 and 7.0 T: a feasibility study in healthy subjects clinical imaging. and patients with intraocular masses. Invest Radiol parametric MRI from experimental research to (03) Winter L, Oezerdem C, Hoffmann W, van de Lindt T, Periquito J, Ji Y, Ghadjar 2015;50(5):309-321. Future Directions: P, Budach V, Wust P, Niendorf T. Thermal magnetic resonance: physics consider- Radiat Oncol Our strategic research will continue to be formed around MR physics to drive/support (04)ations Huelnhagen and electromagnetic T, Hezel F, fieldSerradas simulations Duarte T,up Pohlmann to 23.5 Tesla A, Oezerdem (1GHz). C, Flemming translational explorations into cardio- and B,2015;10:201. Seeliger E, Prothmann M, Schulz-Menger J, Niendorf T. Myocardial effective trans-

neurovascular diseases. We are dedicated to study. Magn Reson Med 2016, (epub ahead of print). throw further weight behind the solution of (05)verse Feutlinske relaxation F, time Browarski T2* Correlates M, Ku MC, with Trnka left P,ventricular Waiczies S,wall Niendorf thickness: T, Stallcup A 7.0 T WB, MRI the remaining underserved clinical needs. To Glass R, Krause E, Maritzen T. Stonin1 mediates endocytosis of the proteoglycan NG2 meet this goal MR technology remains to be and regulates focal adhesion dynamics and cell motility. Nat Commun developed and perfected. The development of 2015;6:8535. next generation RF coil technology and imaging techniques will be tailored to demonstrate Group Leader Moritz Raphael the clinical value of UHF-MR. On the MR phys- Prof. Dr. Thoralf Niendorf Teresa Silva Serradas Duarte ics side we strive to extend our activities and Oliver Stangl master MR electrodynamics by progressing Scientists our numerical simulations up to 70.0 T to Dr. Min-Chi Ku Technical Assistants Dr. Katharina Paul Antje Els For this reason we got heavily engaged in Dr. Andreas Pohlmann Sabrina Klix thesupport Helmholtz explorations application into for extreme a 14 T field and MR. 20 Dr. Sonia Waiczies Stefanie Kox T class whole body MR system submitted to Dr. Lukas Winter the National Roadmap for Research Infra- Secretariat structure initiative of the Federal Ministry PhD students Rosita Knispel of Education and Research. Our ThermalMR Till Hülnhagen approach for thermal interventions will elimi- Yiyi Ji Others nate the main barriers to the study and use of Eva Oberacker Dr. Petr Dusek temperature - a critical dimension of life that is Henning Reimann Dr. Adrian Gonzales Lopez of intense clinical interest, but very poorly un- Leili Riazy Misha Pivovarov derstood so far. Our approach opens an entire Oliver Weinberger Dr. Helmar Waiczies thermal phenotyping as Daniel Wenz Dr. Jens Würfel a link to individualised medicine and is in full Joao dos Santos Periquito alignmentnew research with field the of mission and need to setup Celal Özerdem NAKO Team a precision medicine ecosystem at the Campus Yvonne Bahlke Berlin-Buch. To meet this goal our research is Graduates Dr. Beate Erdmann aligned along the mantra of the great physicist Hendrik Paysen Dr. Andrea Hasselbach Freeman Dyson who stated that “New direc- Michael Pham Lisa Krenz tions in science are launched by new tools Feryal Ramezani Michael Rohloff

much more often than by new concepts.” MDC Research Report 2016 47 CARDIOVASCULAR AND METABOLIC DISEASE Photo: Peter Himsel/MDC Peter Photo:

Daniela Panáková

Electrochemical signaling in development and disease

Ca2+ homeostasis is central to signaling Molecular bases of interactions pathways that control cell fate, differentia- between Wnt non-canonical pathway and L-type Ca2+ channel tion, and organogenesis. In cardiac biology The L-Type Ca2+-channel (LTCC) has an im- Ca2+ ions play pivotal roles in a multitude of cellular processes, ranging from excitation- to a family of voltage-gated Ca2+ channels that 2+ contraction coupling to the regulation of areportant the major function point in cardiac of entry biology. for Ca It ions belongs into - hormone secretion and gene expression. ductance is regulated by membrane depo- Our research aims to uncover the molecu- the cytoplasm of cardiomyocytes. Its con lar mechanisms that mediate the crosstalk A (PKA) pathway. The dysregulation of the channellarization results or by in β-adrenergic/Protein arrhythmias as well cardiac kinase between calcium signaling and other physi- hypertrophy, underlying pathologies leading ological stimuli with major morphogenetic to sudden cardiac death. We reported that the pathways such as Wnt signaling during car- Wnt11 non-canonical pathway regulates the diovascular development as well as in dis- intercellular electrical coupling in the devel- ease states in zebrafish. We have recently conductance (Nature 2010). However, the shown that the calcium-mediated reactive molecularoping zebrafish mechanisms heart by byattenuating which Wnt11 the LTCC sig- oxygen species metabolism is essential for naling might modify LTCC, and thus regulates robust activation of Wnt/ß-catenin signal its activity are unclear. Our results suggest that Wnt11 neither alters the LTCC at the tran- transduction ensuring more efficient and scriptional level, nor exerts any effect on the sustained signaling required for cell differ- localization or abundance of the main pore- entiation (JBC 2014). Currently, we pursue we found that Wnt11 signaling regulates the three lines of research. First, we investigate generationforming subunit of two ofC-terminal the LTCC. isoforms Interestingly, of the the molecular mechanisms that lead to the attenuation of calcium fluxes through the isoform (dCt), and a novel proximal C-termi- L-type Ca2+ channel by non-canonical Wnt nalα1C isoform channel (pCt). subunit: We observed the distal that C-terminal Wnt11 signaling affects the essential interaction be- signals during heart development. Next, we tween A-kinase anchoring protein (AKAP) study the role of calcium fluxes through L- and PKA, required for the formation of two C- type Ca2+ channel in angiogenesis. Last, we terminal isoforms. Our data indicate that the investigate the mechanisms that govern the Wnt11 signaling pathway regulates the LTCC conductance by preventing the generation of development of the functional cardiac syn- the distinct C-termini. However, how exactly cytium. Our work expands the understand- and which AKAP has a key role in this process ing of fundamental molecular processes has to be determined.

that lead to proper formation of functional Role of L-type calcium channel cardiovascular system, and their relevance in during angiogenesis disease. Several calcium channels are responsible for the intracellular calcium oscillations in endothelial cells. Although, the increase in

48 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Epithelial remodeling guides early cardiogenesis. (a-c) Confocal images of wild type hearts at 26 hpf (a), 54hpf (b) and 72 hpf (c) stained for Alcam as membrane marker with transition states in red dots. (d) Graph showing the reduction of transition states over time. 9.1 transition states per 100 cells are at 26 hpf (n=11), 5.2 at 54 hpf (n=10), 4.5 per 100 cells at 72 hpf (n=6). Scale bar = 10 µm. *** = p<0.001.

intracellular Ca2+ concentrations is essential chamber architecture through changes in ac- for physiological function of arteries, the role of Ca2+ - changes in the localization of the phosphory- tomyosin organization. We observed specific voltage gated fluxes L-Type in regulating Ca2+ channel angiogenic encoded pro by chain that are regulated by the PCP pathway cessesthe cacna1c is unclear. gene is We expressed first showed in endothelial that the corelated components. non-muscle myosinMoreover, II regulatorythis process light is cells. To better understand what processes accompanied by impaired SRF signaling. Tak- are regulated by the Ca2+ en together, our data indicate that the Wnt/ during angiogenic sprouting, we downregu- PCP pathway regulates cardiac chamber re- fluxes through LTCC modeling by modulating actomyosin activity and monitored the outgrowth of intersomitic that is coupled to mechanosensitive signaling lated or stimulated the LTCC in the zebrafish controlling cardiomyocyte differentiation. LTCC modulate the angiogenic sprouting of vessels (ISVs). We were able to show that TRPC1 (transient receptor potential type C) Selected Publications channelsISVs. Furthermore, interact geneticallywe showed indicatingthat LTCC andthe Arumughan, A., Roske, Y., Barth, C., Forero, L.L., Bravo-Rodriguez, K., Redel, A., Kos- redundancy of different plasma membrane tova, S., McShane, E., Opitz, R., Faelber, K., Rau, K., Mielke, T., Daumke, O., Selbach, M., Sanchez-Garcia, E., Rocks, O., Panáková, D., Heinemann, U., and Wanker, E.E. (2016). calcium channels in regulating vessel growth. Our data also suggest that Ca2+ - disrupts functional p97 hexamers. Nat Commun 7: 13407. eral are implicated in modulating the endo- Mahmoodzadeh,Quantitative interaction S., Haase, mapping H., Sporbert, reveals A., an Rharass, extended T., UBXPanáková, domain D. in, and ASPL Morano, that thelial cell motility, and that the distinctfluxes in intra gen- regulated by sex 17beta-estradiol. J Mol Cell Cardiol 97: 226-234. cellular calcium concentrations are crucial for Mosimann,I. (2016). Nuclear C.*, Panáková, translocation D.*, Werdich, of the cardiac A.A., Musso, L-type G.,calcium Burger, channel A., Lawson, C-terminus K.L., Carr, is L.A., Nevis, K.R., Sabeh, M.K., Zhou, Y., Davidson, A.J., DiBiase, A., Burns, C.E., Burns, - thePCP proper pathway ISV development. drives cardiac tional partitioning of the heart. Nat Commun Rharass,C.G., MacRae, T., Lemcke, C.A., and H., Zon, Lantow, L.I. (2015). M., Kuznetsov, Chamber S.A., identity Weiss, programs D.G., and Panáková,drive early funcD. remodeling (2014). Ca2+-mediated mitochondrial ROS metabolism 6: 8146. augments Wnt/beta-catenin The profound morphogenetic changes that pathway activation to facilitate cell differentiation. J Biol Chem lead to the looped and chambered vertebrate Panáková, D.*, Werdich, A.A.*, and MacRae, C.A. (2010). Wnt11 patterns a myocar- 289, 27937-27951 heart occur concomitantly with the establish- dial electrical gradient through regulation of the L-type Ca(2+) channel. Nature 466, - * contributed equally nication between cardiomyocytes. However, 874-878. thement molecular and refinement and cellular of intercellular mechanisms commu that underlie the patterning of functional myocar- Group Leader Technical Assistants dial syncytium during early embryonic devel- Dr. Daniela Panáková Alexander Meyer opment are not thoroughly explored. During Jana Richter* vertebrate cardiogenesis the two-chambered Scientist Christina Schulz* (Azubi) looped heart is formed from a simple linear Dr. Mariana Guedes Simões Violeta Coralluzzo* (Azubi) tube. Changes in tissue architecture are con- Dr. Tareck Rharass* trolled by junctional remodeling, cell inter- Secretariat calations, and collective cell migration. Wnt/ PhD students Petra Haink Planar cell polarity (PCP) signaling plays a Laura Bartolini crucial role in guiding the tissue remodeling. Anne Margarete Merks# Graduates and Undergraduates We showed that the non-canonical ligands Wnt11 and Wnt5b and the PCP core compo- Nicola Victoria Müller§ Carlos Bilbao* nents Fzd7, Vangl2, Dvl2, and Pk1 are involved MarieKitti Dóra Swinarski* Csályi Ines Fechner* in the regulation of cell re-arrangements dur- Theodor Lorenzen§* * part of the period reported ing cardiac chamber formation. Downstream # jointly with Prof. Dr. Michael Gotthardt effectors of the PCP pathway target cell adhe- § jointly with Prof. Dr. Reinhardt Kreutz, sion, cytoskeleton, and migration. We showed Charité

that deficient PCP signaling affects the cardiac MDC Research Report 2016 49 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Tobias Pischon

Molecular Epidemiology

The Molecular Epidemiology Group studies cretes a variety of cytokines and hormones. at the molecular level the relationship be- Within the large European Prospective Investigation into Cancer and Nutrition tween lifestyle, diet, genetic, metabolic, and (EPIC), we evaluated in collaboration with environmental factors with risk and outcome of chronic diseases in human populations. and other partners the extent to which 11 The focus is on biomarkers for accurate and biomarkersthe German may Institute mediate of Humanthe association Nutrition of abdominal adiposity with colon cancer risk precise assessment of exposure, intermediary effects, and early disease development. positive association was mostly accounted Our aims are to contribute to the under- for(Aleksandrova by 3 biomarkers, et al, Int HDL-cholesterol, J Cancer 2014). non- The HMW adiponectin and soluble leptin recep- standing of disease etiology, to allow more tor, suggesting that alterations in levels of precise disease prediction, and to improve these biomarkers may represent a mecha- the identification of high risk individuals as nism of action. well as the quantification of the effect of Visceral adipose tissue derived cytokines interventions, thus envisioning a reduction of may induce hepatic secretion of acute chronic disease risk through targeted pre- phase proteins, including C-reactive pro- vention. tein (CRP). We previously found higher CRP levels associated with higher CRC risk (Aleksandrova et al, Am J Epidemiol 2010). Mendelian Randomization approach we observed CRP genetic variability lead- Metabolic dysfunction and risk of ingIn a to raised CRP concentrations associated chronic disease Abdominal obesity is associated with ab- J Cancer 2015), supporting the hypothesis normal glucose metabolism, elevated blood thatwith circulating higher risk CRP of CRC may (Nimptsch play a causal et al, role. Int pressure, dyslipidemia, and type 2 diabetes The liver is central in human metabolism, mellitus (T2D), which cluster within the and obesity is associated with non-alcohol- Metabolic Syndrome (MetS). Components ic steatosis, steatohepatitis, liver cirrhosis, of MetS may not only be risk factors for and liver cancer. Fetuin-A is a liver protein cardiovascular disease (CVD), but also for that inhibits insulin actions, and studies certain cancers, particularly colorectal and found fetuin-A levels related to risk of T2D and CVD. Hyperinsulinemia is a possible risk factor for CRC, but the role of fetuin-A liver (Nimptsch & Pischon, Horm Mol Biol was unclear. We found higher circulating investigateClin Invest 2015).associations As part of of metabolicthe Helmholtz fac- fetuin-A levels associated with a higher torsPortfolio with topicchronic “Metabolic disease risk. Dysfunction” For colorec we- tal cancer (CRC) we found the association 2015). Although fetuin-A genetic variation with MetS largely accounted for by ab- explainedrisk of CRC a large (Nimptsch proportion et al, of Intfetuin-A J Cancer lev- dominal obesity and abnormal glucose me- els, genetically determined higher fetuin-A tabolism (Aleksandrova et al, Cancer Prev was not associated with CRC, suggesting Res 2011). Abdominal obesity is mostly that fetuin-A may not be causally related to accounted for by visceral fat, which se- CRC.

50 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

We also studied prospectively the asso- surgery among persons within the general - elderly population. POCD can also be con- markers with risks of liver and bilary tract cancersciation of (Aleksandrovainflammatory andet al,metabolic Hepatology bio analysis,sidered a wemodel found to studyonly a the few influences studies that on C-peptide, and non-HMW adiponectin were havecognitive investigated impairment the associationin general. Inof aobesity meta- associated2014). Higher with concentrations higher risk of of hepatocel CRP, IL-6,- with risk of POCD and the results provide lular carcinoma (HCC), suggesting that el- only limited support that obesity increases risk (Feinkohl et al, Diabetes Metab Res Rev and hyperinsulinemia are associated with a 2016). As part of the project “BioCog – Bio- higherevated levelsrisk of of HCC, biomarkers independent of inflammation of obesity marker Development for Postoperative and established liver cancer risk factors, and Cognitive Impairment in the Elderly”, may be able to improve risk assessment. supported within FP7 by the European Union we investigate associations of obe- The insulin-like growth factor (IGF) axis sity and metabolic dysfunction with risk of mainly regulates cellular proliferation, dif- POCD in detail. ferentiation and apoptosis, but has also been linked to type 2 diabetes mellitus Methodological studies Abdominal obesity, as assessed by waist circumference, is closely associated with met- investigated(T2D). IGF binding together protein with 3 (IGFBP-3)partners from may abolic dysfunction. Manual measurements increase T2D risk via binding of IGF-1. We have so far been the method of choice for body circumference assessment. 3D body theirthe German ratio in relation Institute to of T2D Human incidence Nutrition with- surface scanners facilitate computation inserum EPIC-Potsdam concentrations (Drogan of IGF-1, et al, IGFBP-3 Am J Epide andof body measures, but limited information is available on validity and reliability. We found strong correlations between auto- miol 2016). Our findings did not confirm an mated measurements with the bodyscanner association between IGF-1 and risk of T2D, whereas higher IGFBP-3 levels may increase (Jaeschke et al, PLOS ONE 2015). Reliability MetabolomicsT2D risk independent is the ofsimultaneous IGF-1 levels. study wasVitussmartXXL high and correlations and manual with measurements parameters of numerous low-molecular weight com- of MetS were similar between automated - and manual measurements. Supported by tabolites independently associated with the Federal Ministry for Economic Affairs riskpounds. of T2D, We previouslyincluding sugar identified metabolites, 14 me and Energy (BMWi), we investigate in the amino acids, and choline-containing phos- MetSScan Study to what extent automated pholipids (Floegel et al, Diabetes 2013). We measures of abdominal volume may im- currently investigate to what extent me- prove detection of MetS. tabolites are associated with risk of cardiovascular disease. Analysis of complex data from omics studies requires appropriate statisti- Obesity has also been linked to cognitive cal approaches. We propose a biological impairment, although mechanisms are framework and a statistical model incor- unclear. Post-operative cognitive dysfunc- porating multi-level biological measures, tion (POCD) is a frequent condition after which leads to more powerful genetic asso-

MDC Research Report 2016 51 CARDIOVASCULAR AND METABOLIC DISEASE

ciation analyses (Konigorski et al, BMC Proc may explain the variance in AEE. Further, 2016). We investigated the effect of single variability in PA, observational time needed nucleotide variants on blood pressure and to estimate habitual PA, and reliability are gene expression, while considering the - non-directional dependence between both, ance of AEE explained by accelerometer as- using copula functions. Our results show unknown. In a systematic review, the vari that such tests have smaller p-values than - univariate single or multi marker tests. sessed PA was 4-80% (Jeran et al, Int J Obes predicted2016). Inclusion AEE based of other on accelerometrypredictors signifi as- Diet, Nutrition, Physical Activity sessedcantly increased PA need to it be to interpreted13-41%. Thus, cautiously. data on Circulating fetuin-A, a marker for hepatic As part of the ActivE Study, we aim to de- fat accumulation, has been related to a rive more precise prediction models. We as- higher risk of T2D and CVD. Little is known sessed 24-h-PA in participants wearing ac- about dietary or nutritional determi- celerometers over two weeks (Jaeschke et nants of fetuin-A concentrations. Together al) and found individual PA highly variable with partners from the Helmholtz Center between days, but the day of assessment or Munich, we found in the Bavarian Food the day of the week explained only small Consumption Survey II (BVSII) higher parts of this variance. Our data indicate that - one week of assessment is necessary for re- ated with higher fetuin-A concentrations liable estimation of habitual PA. (Nimptschenergy intake et al, Br nonsignificantly J Nutr 2015). There associ was no clear association for energy-providing Assessment of the association between nutrients, but higher alcohol intake and diet, nutrition, and metabolic factors on a milk and dairy products were associated population level can best be achieved with with lower fetuin-A. Our data indicate that the use of well-designed epidemiologic among dietary factors particularly alcohol studies. We participate in the European consumption and potentially dairy intake Nutritional Phenotype Assessment and may be related to fetuin-A concentrations. Data Sharing Initiative (ENPADASI), sup- Paraoxonase (PON1) and arylesterase ported by the Federal Ministry of Food and (AE) are functions of the enzyme paraox- Agriculture (BMEL) within the Joint Pro- onase, synthesized by the liver and circulating in plasma to protect lipoproteins isgramming to deliver Initiative a research (JPI) infrastructure “A Healthy with Diet alcohol consumption may reduce serum datafor a from Healthy a variety Life” (HDHL).of nutritional The objective studies. PON1against and oxidative AE activities modification. but less Excessiveis known This will foster the analyses of diet, nutrition, and metabolic factors in different pop- found no strong association between alco- ulations. holfor and habitual PON1 alcohol and AE intake. (Schwedhelm In BVSII et weal, Br J Nutr 2016). PON1 activity was lowest Although diet and physical activity are in non-drinkers and highest with medium among the main risk factors for chronic alcohol consumption. AE activity increased diseases, their determinants are less clear. with alcohol consumption. Associations Within the Knowledge Hub Determinants were attenuated after adjustment for HDL of Diet and Physical Activity Choice (DE- concentrations. These results do not sup- DIPAC) port the hypothesis that habitual alcohol HDHL we investigate determinants of diet consumption is related to important altera- and physical, which activity, is part supported of the European by the Fed JPI- tions in PON1 and AE activities. eral Ministry of Education and Research (BMBF) (Condello et al., BMC Public Health Physical activity (PA) is an important fac- 2016). tor in the etiology of chronic diseases, but less is known for activity related energy The National Cohort (NAKO) expenditure (AEE). Traditionally, AEE has With the aim to develop new strategies been estimated from questionnaires. Accel- for risk assessment, early detection, and erometers might help to objectively mea- prevention of major chronic diseases, a sure PA but it is unclear to what extent this network of German research institutions,

52 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Selected Publications

Nimptsch K, Aleksandrova K, Boeing H, Janke J, Lee YA, Jenab M, Kong SY, Tsilidis KK, including the MDC, has started a large pro- Weiderpass E, Bueno-De-Mesquita HB, Siersema PD, Jansen EH, Trichopoulou A, Tjon- spective cohort, supported by the Federal neland A, Olsen A, Wu C, Overvad K, Boutron-Ruault MC, Racine A, Freisling H, Katzke V, Kaaks R, Lagiou P, Trichopoulos D, Severi G, Naccarati A, Mattiello A, Palli D, Grioni Ministry of Education and Research (BMBF), the Helmholtz Association, and the States of Bonet CB, Chirlaque MD, Dorronsoro M, Quiros JR, Travis RC, Khaw KT, Wareham N, Germany. The National Cohort will include S,Riboli Tumino E, Gunter R, Peeters MJ, Pischon PH, Ljuslinder T. Plasma I, Nystrom fetuin-A H,concentration, Brandstedt J, genetic Sanchez variation MJ, Gurrea in the AB, 200,000 participants from the general Nimptsch K, Aleksandrova K, Boeing H, Janke J, Lee YA, Jenab M, Bueno-de-Mesquita population of Germany, recruited through HB,AHSG Jansen gene EH,and Tsilidisrisk of colorectal KK, Trichopoulou cancer. Int A, JWeiderpass Cancer 2015;137:911-20 E, Wu C, Overvad K, Tjon- neland A, Boutron-Ruault MC, Dossus L, Racine A, Kaaks R, Canzian F, Lagiou P, Trichopoulos D, Palli D, Agnoli C, Tumino R, Vineis P, Panico S, Johansson A, Van (GNC) Consortium, Eur J Epidemiol 2014). Guelpen B, Khaw KT, Wareham N, Peeters PH, Quiros JR, Vencesla Garcia A, Molina- All18 studyparticipants centers undergo (German baseline National examina Cohort- Montes E, Dorronsoro M, Chirlaque MD, Barricarte Gurrea A, Key TJ, Duarte-Salles T, tion and 5-year re-examination. Every 2-3 Stepien M, Gunter MJ, Riboli E, Pischon T. Association of CRP genetic variants with years questionnaires on lifestyle changes and new major diseases will be collected. Jaeschkeblood concentrations L, Steinbrecher of C-reactive A, Pischon protein T. Measurement and colorectal of waist cancer and risk.hip circumference Int J Cancer Mortality follow-up and systematic linkage with2015;136:1181-92 a body surface scanner: feasibility, validity, reliability, and correlations with with disease registries will be performed. Aleksandrova K, Boeing H, Nothlings U, Jenab M, Fedirko V, Kaaks R, Lukanova Biomaterials are stored in a centralized A,markers Trichopoulou of the metabolic A, Trichopoulos syndrome. D, Boffetta PLoS One P, Trepo 2015;10:e0119430 E, Westhpal S, Duarte-Salles biobank and in decentralized storages. The T, Stepien M, Overvad K, Tjonneland A, Halkjaer J, Boutron-Ruault MC, Dossus L, Molecular Epidemiology Group coordinates Racine A, Lagiou P, Bamia C, Benetou V, Agnoli C, Palli D, Panico S, Tumino R, Vineis P, the Cluster Berlin-Brandenburg, including A, Sanchez MJ, Gavrila D, Barricarte A, Dorronsoro M, Ohlsson B, Lindkvist B, Johans- the MDC, the Charité-Universitätsmedizin Bueno-de-Mesquita B, Peeters PH, Gram IT, Lund E, Weiderpass E, Quiros JR, Agudo and metabolic biomarkers and risk of liver and biliary tract cancer. Hepatology son A, Sund M, Khaw KT, Wareham N, Travis RC, Riboli E, Pischon T. Inflammatory Nutrition (Steinbrecher et al, Berliner Ärzte Aleksandrova K, Drogan D, Boeing H, Jenab M, Bas Bueno-de-Mesquita H, Jansen 2015).Berlin, andThe theCluster German includes Institute 3 study of Human cen- 2014;60:858-71 ters, recruiting 30,000 participants into the Romaguera D, Westhpal S, Overvad K, Tjonneland A, Halkjaer J, Boutron-Ruault MC, E,Clavel-Chapelon van Duijnhoven F, FJ,Lukanova Rinaldi A,S, FedirkoTrichopoulou V, Romieu A, Trichopoulos I, Kaaks R, Riboli D, Vidalis E, Gunter P, Panico MJ, cohort. Samples from these participants are S, Agnoli C, Palli D, Tumino R, Vineis P, Buckland G, Sanchez-Cruz JJ, Dorronsoro M, stored in a biobank Diaz MJ, Barricarte A, Ramon Quiros J, Peeters PH, May AM, Hallmans G, Palmqvist Berlin Ultrahigh Field Facility (B.U.F.F.), the R, Crowe FL, Khaw KT, Wareham N, Pischon T. Adiposity, mediating biomarkers and risk of colon cancer in the European prospective investigation into cancer and nutri- Molecular Epidemiologyat the Group, MDC. the At Experi MDC’s- mental Ultrahigh-Field MR group, and the MRI measure- tion study. Int J Cancer 2014;134:612-21. ments of 6.000 study participants (Bam- Group Leader Cardiacberg et al, MRI Radiology group conduct 2015). Prof. Dr. Tobias Pischon Ben Jacoby Stephanie Krenzin With support from the German Center for Scientists Katrin Krüger Cardiovascular Research (DZHK) we inves- Carolin Adler Karin Teichmann (since 15.5.2016) tigate the association of lifestyle and meta- Anette Veauthier bolic factors with risk of secondary events Dr. Jürgen Janke (biobank) Maike Wesselmann (since 15.10.2016) in patients with prior myocardial infarc- SabineDr. Insa Mall Feinkohl tion, which may form the basis for second- Dr. Katharina Nimptsch Technical Assistants ary prevention measures. Dr. Mariona Pinart Gilberga Franzy Hohnstädter Dr. Astrid Steinbrecher Marie-Kristin Kusnierz Christina Lauschke PhD students Steffen Lutter Stephanie Jeran Sarah Moreno Garcia Lina Jaeschke Annelie Richter Stefan Konigorski Manuela Stendal

Staff Secretariat Claudia Bartelt Marit Kanter Dr. Claus Derz (biobank) Barbara Eichhorn MPH/MSE Students Julia Glöde Dr. Daniela Hödke (since 01.09.2016) Kathleen Nitschmann Tanja Hinz Carolina Schwedhelm Ines Haase

MDC Research Report 2016 53 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Matthew Poy

microRNA and Molecular Mechanisms of Metabolic Diseases

Type II diabetes has finally become recog- showing subtle phenotypes under steady nized as a major challenge to global health. state conditions, a hypothesis has gradually emerged suggesting the miRNA pathway It is imperative to improve our understand- contributes to cellular stress responses. ing of the molecular mechanisms behind this Moreover, these observations are further disorder and develop new drug therapies. supported by our recent study focusing on The pathophysiology of diabetes is undoubt- the role of Argonaute2 (Ago2) in the compensatory proliferation of the pancreatic edly complex, typically characterized by - hyperglycemia resulting from varying states of insulin resistance and impaired b-cell β-cell during insulin resistance. These re miRNAcent findings now established culminate over among 10 theyears highest since function. Oftentimes, the failure to regulate expressedour initial studysequences which in identified this cell type. miR-375, a circulating blood glucose levels is a consequence of the inability to produce sufficient amounts of insulin by the pancreatic β-cells. studies revealed this miRNA as a nega- tiveThe regulator very first of loss insulin and gain-of-functionrelease via the In our group, we focus on fundamental direct targeting of the gene myotrophin. pathways regulating glucose metabolism The role of miR-375 in insulin release was and how altered pancreatic islet physiology contributes to metabolic disorders such as mouse knockout (375KO). Furthermore, further confirmed in vivo using the total type 2 diabetes. several additional targets of miR-375 in- cludingthis model Elavl4/HuD, facilitated theCadm1, identification Gphn, and of To date, numerous counter-regulatory Rasd1. Moreover, 375KO mice exhibited mechanisms have been characterized in the cell illustrating the dynamic nature of and these effects were further exacerbated the intracellular environment. Among the afterhyperglycemia crossing the and knockout decreased onto β-cell the leptin- mass, more recently discovered is the microRNA - (miRNA) pathway and as of this writing, dicated that miR-375 plays an essential role - deficient ob/ob background. This result in- notated in the miRBase database which now duced by insulin-resistance in ob/ob mice. catalogs2578 mature 206 differenthuman miRNAs plant and have animal been spe an- Whilein the compensatorythe precise causes β-cell proliferationare not under in- stood, insulin resistance is one of the most in C.elegans, a canonical pathway has now common metabolic stress conditions and cies. Importantly, since their identification a consequence of chronic over-nutrition. processed and incorporated into Argonaute- Moreover, our recent observations show- containingbeen identified complexes to describe to maintain how miRNAs their areef- ing loss of Ago2 also blocked proliferation fect on gene expression. Many insights into during insulin resistance further supports the functional role of miRNAs have been a role for the miRNA pathway in mediating recently made, improving our understanding of how these small RNAs integrate into showed that Ago2 is up-regulated in the the already complex landscape of regulating pancreaticcellular stress islets responses. of both genetic Importantly, and diet- we gene expression. Most notably, based on a induced models of insulin resistance and number of miRNA knockout mouse models obesity as a result of the silencing of miR-

54 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

known whether the same miRNAs perform in184. islets Furthermore, of human thesubjects inverse further relationship under- anma identical membrane. regulatory Interestingly, function it in is different still not between Ago2 and miR-184 was confirmed- cell types. As Ago2-associated sequences tion in these mouse models. Of note, miR- such as miR-375 are expressed in several lining the relevance of studying β-cell func unique tissues such as the pituitary and a mitochondrial glutamate transporter, in- adrenal gland, it is unclear how this miRNA dicating184 was miRNAsshown to may target regulate the gene many Slc25a22, stages may regulate growth and secretion in these - other neuroendocrine cell types. ing to insulin secretion. of the canonical pathway in the β-cell lead - tifying components of the RNAi machinery regulate both growth and secretion of the inIn summary,the nucleus in lightincluding of recent Ago2, findings Dicer, iden and While miR-184 has been shown to potently changes in insulin sensitivity as shown in before new mechanistic insights into the theβ-cell, models it remains of obesity to may be determinedcontribute to how the roleGW182, of this it appearspathway onlyare discovered. a matter of While time direct regulation and function of all miR- NAs in this cell type. Meanwhile, it is un- remainsthe identification to be studied of direct in greater targets ofdetail, the may also change in response to alterations themiRNAs role of and small their RNAs biological in the nucleus significance also withinclear whether its metabolic miRNA environment function in the such β-cell as remains to be described. The story of non- levels of extracellular nutrients (glucose, coding RNAs and their functional role in the amino acids or fatty acids), signaling hormones, neurotransmitters, or perturba- exciting time of rapidly evolving technolo- tions in cell-to-cell contact with neighbor- gies,pancreatic the next β-cell 10 isyears far from are certain complete. to bringIn an ing endocrine, endothelial, mesenchymal, clarity to this expanding narrative. or neuronal cells. Moreover, key transcrip-

well as RNA-binding proteins that may spe- Selected Publications tional regulators of miRNAs in the β-cell as- Tattikota, SG., Rathjen, T., Hausser, J., Pandey, AK., Wessels, HH., Esguerra, JLS.,

cifically bind miR-184 and target it for deg Zavolan, M., and Poy, MN. Argonaute2 mediates compensatory expansion of the mayradation coordinately remain unidentified. target genes It to is promote unclear McAnulty, SJ., Musahl, A., You, X., Chen, W., Herrera, P., O’Carroll, D., Eliasson, L., insulinhow the release additional as well abundant as facilitate β-cell compen miRNAs- pancreaticTattikota SG, β-cell. Rathjen (2014) T, Hausser Cell Metabolism. J, Khedkar 19(1):122-34. A, Kabra UD, Pandey V, Sury M, Wessels

HH, Mollet IG, Eliasson L, Selbach M, Zinzen RP, Zavolan M, Kadener S, Tschöp MH, satory expansion of the β-cell as metabolic Jastroch M, Friedländer MR, Poy MN. J Biol Chem. 2015 Aug 14;290(33):20284-94 isletsdemand of increases.insulin-resistant While miR-184mice, additional was the miRNAsmost significantly have been regulatedshown to miRNAparticipate in the in Group Leader Technical Assistant Dr. Matthew Poy Christopher Bishop - Anton Gauert merousthe β-cell miRNAs growth act using in a human concerted islet fashion cells8. Scientists toIt willregulate presumably multiple be targetsdetermined that thatmediate nu Secretariat cell growth and the continual recruitment Dr. Zhen Wang Feraye Kocaoglu of insulin-containing granules to the plas- Dr. Xin Yan

MDC Research Report 2016 55 CARDIOVASCULAR AND METABOLIC DISEASE Photo: Feriet Tunç/BIH Feriet Photo:

Kai Schmidt-Ott

Molecular and Translational Kidney Research

Kidney disease affects millions of people and perturbations of kidney development result causes substantial morbidity and mortality. in urogenital malformations, cystic kidney diseases and other types of congenital kidney Our research group studies the molecular and disease. Furthermore, adult kidney epithelia cellular mechanisms that govern kidney de- preserve the ability to reactivate molecular velopment in the embryo and kidney disease pathways from earlier developmental stages later in life. We strive to discover novel diag- in certain disease states, including acute kid- nostic markers and therapeutic targets, and to better understand the molecular networks Geneney injury, Regulatory kidney fibrosis, Networks and kidney in tumors. that govern tissue differentiation, homeostasis, Kidney Epithelial Development and and regeneration. We conduct clinical studies Homeostasis Gene expression control governs develop- and utilize model organisms employing a wide ment, homeostasis and adaptive responses. spectrum of cell and molecular biology tech- We discover and characterize novel molecular niques, as well as systems biology approach- pathways that regulate epithelial morphogenesis and differentiation during kidney de- es. Our central goal is to improve the diagnosis velopment and their response to renal injury. and therapy of kidney disease. Our past and ongoing studies include the WNT,

Background The kidney is a central organ in cardiovascu- GrainyheadBMP, NF-κB typepathways transcription and their factors, downstream which wetranscriptional found to govern targets. critical In addition, aspects of we epithe study- pressure and solute homeostasis, and produc- lial morphogenesis and differentiation, but eslar hormones.diseases. It High excretes blood toxins, pressure regulates (hyperten blood- which have also been implicated in epithelial sion) and chronic kidney disease rank among - the top risk factors for cardiovascular end or- orative network which includes groups from gan damage. The kidney is composed of struc- Columbiainjury. The University projects benefit(Jonathan from Barasch) our collab and tural units called nephrons, which consist of MDC (Carmen Birchmeier, Thomas Willnow, more than 20 different types of epithelial cells that facilitate transport between the urinary (Hemodynamic mechanisms of acute kidney - injury)Michael we Bader). study the Within role researchof the BMP unit and 1368 NF- bryo, kidney development is initiated during kB pathways in kidney regeneration (with mid-embryogenesis,compartment and the when interstitium. the ureteric In the bud,em Dominik Müller, ECRC, and Ruth Schmidt-Ull- an epithelial tubule extending from the neph- rich, MDC). ric duct, interacts with an adjacent cell population of committed stem cells in the metaneph- Systems biology of kidney diseases ric mesenchyme. The ureteric bud undergoes Kidney injury and phenotpyical transforma- branching morphogenesis to give rise to the tions of kidney epithelia in congenital and ureter, renal pelvis and collecting duct system, neoplastic kidney diseases are associated with while the metanephric mesenchyme progeni- widespread alterations of the genome and its tors generate many additional cell types of the expression. We are combining next generation nephron. The molecular and cellular events sequencing based techniques and bioinfor- underlying kidney development are intricately matics to map alterations of the genome, epig- linked to kidney disease. Exogenous or genetic enome, and transcription factor binding sites

56 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

in normal and diseased kidney cells. For this by the Deutsche Forschungsgemeinschaft purpose we are collaborating with the Depart- ment of Urology at Charité (Kurt Miller), with the group of Walter Birchmeier (MDC) and with within Research Unit 1368 (Hemodynamic Acknowledgementsmechanisms of AKI). Our research is generously supported by the thatsystems a detailed biology elucidation groups at the of theBerlin genome-wide Institute of Deutsche Forschungsgemeinschaft and by alterationsMedical Systems in diseased Biology renal (BIMSB). cells will We yield believe an the Stiftung Urologische Forschung Berlin. unprecedented opportunity to understand in For additional information please visit: detail the molecular mechanisms of renal dis- www.mdc-berlin.de/schmidt-ott ease. These studies are supported by the Uro- logical Research Foundation. Selected patents /patent applications Molecular Diagnosis of Kidney Injury Determining the temporal phase of acute kidney injury involves obtain- We aim to identify molecular signatures or ing test sample from subject, and determining the expression level of at least one biomarker e.g. cation transport regulator-like protein 1. Patent - sivelymarkers studied that are neutrophil linked with gelatinase-associated specific aspects of Number:Delbrueck WO2011157828-A1; Center) EP2582840-A1; US2013165338-A1. Publ. the kidney’s response to injury. We have exten DiagnosingDate: 12/22/2011. acute kidney Inventors: injury Kai in aM. subject Schmidt-Ott, comprises Anne obtaining Wuebken a (viaurine Max sample - and determining amount of neutrophil gelatinase-associated lipocalin- and pressedlipocalin in (NGAL) the kidney and in calprotectin response to (S100A8/structural tissueS100A9), tissue two injury proteins and thatthat quickly are specifically get released ex kidneyM. Schmidt-Ott injury molecule-1 (via Columbia protein; University) Patent Number(s): WO2012068545-A1. into the urine. Our studies indicate that NGAL Publ. Date: 05/24/2012. Inventors: Jonathan Barasch, Thomas L. Nickolas, Kai and calprotectin are valuable non-invasive bio- Selected Publications

Aue A*, Hinze C*, Walentin K, Ruffert J, Yurtdas ZY, Werth M, Chen W, Rabien A, Jonathan Barasch (Columbia University, New Kilic E, Schulzke JD, Schumann M, Schmidt-Ott KM. A grainyhead-like 2/ovo- York),markers Timm of kidney Westhoff injury. (University In collaboration of Bochum), with like 2 pathway regulates renal epithelial barrier function and lumen expansion. Journal of the American Society of Nephrology. 2015, 26:2704-15.* Equal Klemens Budde (Department of Nephrology, contribution. Charité) and Friedrich C. Luft (MDC, ECRC), Walentin K, Hinze C, Werth M, Haase N, Varma S, Morell R, Aue A, Pötschke we are conducting basic and clinical studies E, Warburton D, Qiu A, Barasch J, Purfürst B, Dieterich C, Popova E, Bader M, to study the biology and clinical applications Dechend R, Staff AC, Yurtdas ZY, Kilic E, Schmidt-Ott KM. A Grhl2-dependent gene network controls trophoblast branching morphogenesis. Development. of NGAL and calprotectin in kidney injury. We 2015, 142:1125-36. develop diagnostic algorithms that utilize these Paragas N*, Kulkarni R*, Werth M*, Schmidt-Ott KM*, Forster C, Deng R, Zhang markers to predict renal injury and to differen- Q, Singer E, Klose AD, Shen TH, Francis KP, Ray S, Vijayakumar S, Seward S, Bo- tiate renal injury from related clinical entities. We also aim to identify additional novel and vino ME, Xu K, Takabe Y, Amaral FE, Mohan S, Wax R, Corbin K, Sanna-Cherchi- S,tion. Mori Journal K, Johnson of Clinical L, Nickolas Investigation. T, D’Agati V,2014, Lin CS, 124:2963-76. Qiu A, Al-Awqati *Equal Q, contribuRatner AJ,- compare them to the available diagnostic meth- tion.Barasch J. α-Intercalated cells defend the urinary system from bacterial infec # # odsphase-specific in nephrology. biomarkers These studies of kidney are injurysupported and Nickolas TL* , Schmidt-Ott KM* , Canetta P, Forster C, Singer E, Sise M, Elger A, Guddati A, Polland A, Rahman W, Elitok S, Malik N, Giglio J, El-Sayegh S, Devara- Developing mouse kidneys are analyzed by whole- Maaroufjan P, Hebbar O, Sola-Del S, Saggi Valle SJ, Hahn DA, O’Rourke B, Kettritz M, R, Sherman Luft FC, BaraschE, Lee P, J.Geara Diagnostic A, Imus and P, - mount in situ hybridization (A) or immunofluores- ers of Nephron Damage - A Multicenter Prospective Cohort Study. Journal of cent staining (B). (C) shows an example of an ex- thePrognostic American Stratification College of in Cardiology. the Emergency 2012, Department 59:235-44. Using *Equal Urinary contribution. Biomark # periment to detect genome-wide DNA modifications Corresponding authors. by next generation sequencing. Reads and coverage Marko L*, Vigolo E*, Hinze C, Park JK, Roel G, Balogh A, Choi M, Wuebken A, depict histone H3K4 trimethylation in collecting duct cells following knockdown of the transcription JournalCording J,of Blasig the American IE, Luft FC, Society Scheidereit of Nephrology. C, Schmidt-Ott KM**, Schmidt-Ullrich*Co-first, **EquallyR**, Muller contributing DN**. Tubular principal Epithelial investigators. NF-kB Activity Regulates Ischemic AKI. factor Grhl2 (Grhl2-Knockdown) and in control 2016, 27:2658-69. cells. Note differential H3K4 trimethylation at the promoter of the gene encoding Prom2. Group Leader Janett Ruffert Prof. Dr. Kai Schmidt-Ott (Urological Research Foundation (Hans-Schildbach Stiftungsprofessur) Scholarship) Jutta Swolinsky Scientists Emilia Vigolo Dr. Christian Hinze Ricarda Merle Viol Zeliha Yesim Yurtdas (Urological Research Foundation (BIHDr. Eva Charité Vanessa Clinician Schrezenmeier Scientist Program) Scholarship)

(BIHDr. Katharina Charité Junior Walentin Clinician Scientist Program) Technical Assistants PhD/MD thesis students Tatjana Luganskaja Paula Halbeisen Gabriel Simon Kirchgraber Fatma A. A. Mansour (Stipend from Cairo University) Administrative assistant Carolin Elfriede Markus Petra Haink

MDC Research Report 2016 57 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Matthias Selbach

Proteome Dynamics

Understanding how the genomic information Introduction is interpreted to yield a specific phenotype is - teomics is now a well-established technol- perhaps the most important question in the ogyMass that spectrometry-based can identify essentially (“shotgun”) all proteins pro post-genomic era. Proteins are the central players in this process: First, they are different methods have been developed to the final product of most genes. Therefore, quantifyexpressed proteins in a mammalian by mass cell. spectrometry. In addition, The combination of mass spectrometry understanding gene expression control re- - quires analysis of protein synthesis and deg- tain precise functional information and to radation. Second, function and dysfunction monitorwith quantification temporal changes makes itin possible the proteome. to ob of proteins is directly responsible for cellular isotope labelling with amino acids in cell phenotypes. Studying protein function can culture),In one approach, cells are nameddifferentially SILAC (forlabelled stable- by therefore provide novel insights into biologi- cultivating them in the presence of either cal processes in health and disease. We are normal or a heavy isotope–substituted amino acids. Due to their mass difference, using quantitative mass spectrometry-based pairs of chemically identical peptides of dif- proteomics as our central technology to ferent stable-isotope composition can be study proteome dynamics on a global scale. distinguished in a mass spectrometer. The The lab is interested in two major questions. ratio of intensities for such peptide pairs First, how is the genomic information pro- - cessed to yield a specific proteome? To an- LACaccurately can be reflects used to the quantify abundance differences ratio for in swer this question we are studying protein the corresponding proteins. Classical SI synthesis and degradation. Second, how - LAC)steady-state as a novel protein variant levels to measure (Fig. 1, left). differ In- do proteins that are expressed at a specific encesaddition, in proteinwe developed synthesis pulsed (Fig. SILAC1, middle). (pSI cellular condition affect the phenotype? We are investigating this by analysing protein- - Finally, dynamic SILAC can reveal protein- protein interactions and posttranslational videturnover information (Fig. 1, right). about SILACdifferent is thus aspects a ver of modifications. cellularsatile quantification proteome dynamics. method that can pro

Gene expression control Many researchers use the term “gene ex- - els. However, it is important to keep in pression” when they refer to mRNA lev of most genes. Therefore, understanding genemind expression that proteins control are therequires final productsinforma- tion about protein synthesis and degra-

gene expression cascade can be regulated. dation. It is now clear that all steps in the

58 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Fig. 1: Different variants of SILAC for the analysis of cellular proteome dynamics. Cells are meta bolically labelled with light (L), medium-heavy (M) or heavy (H) stable isotope-encoded amino acids. Differentially labelled peptides can be distinguished by mass spectrometry. Standard SILAC reveals relative differences in protein levels. pSILAC quantifies changes in protein production. Dynamic SILAC provides information about protein turnover.

However, little is known about how the od enabled us to quantify the impact of mi- combined effect of all regulatory events croRNAs on protein synthesis (Selbach et shapes gene expression. The fundamental - question of how genomic information is sky lab). We also used metabolic labeling of - newlyal., 2008; synthesized collaboration mRNAs with (using the 4-thiouriN. Rajew- teome is therefore still largely unresolved. processed to obtain a specific cellular pro We are using metabolic pulse labelling ap- genedine) expression and proteins control (“dynamic at all SILAC”)major lev to- proaches to comprehensively quantify obtain the first global view of mammalian- gene expression at the protein level. For ration with W. Chen and J. Wolf labs). Our resultsels (Schwanhäusser indicated that et gene al., 2011;expression collabo in quantify protein translation on a proteome- - wideexample, scale we (Schwanhäusser developed pSILAC et toal., directly 2009, trolled at the level of translation. We also - foundmouse that fibroblasts translation is predominantlyis actively regulated con

doi:10.1002/pmic.200800275). This meth MDC Research Report 2016 59 CARDIOVASCULAR AND METABOLIC DISEASE

during Schwann cell development in vivo provides direct insights into protein function C. Birchmeier lab). Most recently, we have andIdentifying can reveal interaction disease mechanisms. partners therefore We are used(Sheean pulse et al.,chase 2014; experiments collaboration to study with the using quantitative interaction proteomics as kinetics of cellular protein degradation a very versatile technology to study protein- (McShane et al., 2016). The data revealed protein interactions (Meyer and Selbach, that many proteins are less stable in the This approach has two unique advantages. with age. This so-called non-exponential 2015; doi:10.3389/fgene.2015.00237). degradationfirst few hours (NED) of theiris common, life and conserved stabilize and has important consequences for pro- fromFirst, background accurate quantification contaminants allows with usvery to tein complex formation and regulation of distinguish specific interaction partners- protein abundance. veals how interactions change in response tohigh perturbation. confidence. Therefore, Second, quantification this method can re Protein-protein interactions assess how disease-associated mutations Proteins typically interact with other pro- and cell signaling events affect protein-protein interactions.

teins to exert a specific cellular function.

Fig. 2: Mapping interactions of neurodegenerative disease proteins. (A) Stable isotope-labeled cells are transfected with plasmids encoding tagged wild-type or mutant bait proteins. Quantitative proteomics identifies differential binders. (B) Comparison of preferential interactors for wild-type (blue) or disease-associated variants (red) of ataxin-1 (ATXN1), huntingtin (HTT), presenilin-1 (PSEN1), amyloid precursor protein (APP) and parkin-2 (PARK2). From Hosp et al., 2015.

60 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

We recently used quantitative interaction proteomics to identify binding partners of neurodegenerative disease proteins (Hosp et al., 2015). Functional follow-ups showed

linked to disease phenotypes in animal modelsthat identified and in human interaction patients. partners Moreover, are the direct comparison of wild-type proteins and disease-associated variants allowed us to identify binders involved in pathogenesis, which highlights the power of differential interactome mapping (Fig. 2). Systemat- ic mapping of protein-protein interactions in diseases can thus help bridging the gap - ed variants and disease phenotypes. We are thereforebetween identificationcurrently upscaling of disease-associat this approach and applying it to different diseases. More- over, we are also using quantitative interaction proteomics to explore the function of

uncharacterized human proteins. influenza A virus proteins and previously In vivo quantitative proteomics Cell culture-based experiments cannot re- Selected Publications capitulate all of the complex interactions among different cell types and tissues that occur in vivo. Small animal models such as Selbach M, Schwanhäusser B, Thierfelder N, Fang Z, Khanin R, Rajewsky N. (2008). - Schwanhausser,Widespread changes B., Busse, in protein D., Li, synthesisN., Dittmar, induced G., Schuchhardt, by microRNAs. J., Wolf, Nature J., Chen, 455, W., 58-63. and doi:10.1038/nature07228 natives that are extensively used in many worms and fruit flies are attractive alter areas of biomedical research, especially in Sheean,Selbach, M.E.,M. (2011). McShane, Global E., Cheret,quantification C., ..., Selbach, of mammalian M., and Birchmeier,gene expression C. (2014). control. genetics and development. We have extend- ActivationNature 473, of 337-342. MAPK overrides doi:10.1038/nature10098 the termination of myelin growth and replaces Nrg1/ - lanogaster (Sury et al., 2010, doi:10.1074/ 290-303. doi:10.1101/gad.230045.113 HospErbB3 F, signals Vossfeldt during H, Heinig Schwann M, Vasiljevic cell development D, …, Voigt and A, myelination.Selbach M. (2015) Genes Quantita Dev 28, - mcp.M110.000323)ed the SILAC technology and to DrosophilaCaenorhabditis me tive interaction proteomics of neurodegenerative disease proteins. Cell Reports 19, elegans (Grun et al., 2014, doi:10.1016/j. 1134-1146. doi:10.1016/j.celrep.2015.04.030 celrep.2014.01.001, collaboration with N. Rajewsky lab). Moreover, we have used Z, Marsh JA, Valleriani A, Selbach M. (2016) Kinetic analysis of protein stability re- McShane E, Sin C, Zauber H, Wells JN, Donnelly N, Wang X, Hou J, Chen W, Storchova these model systems to study protein-protein interactions in vivo (Chen et al., 2016, veals age-dependent degradation. Cell 167, 803-815. doi:10.1016/j.cell.2016.09.015 - Group Leader Tommaso Mari ration with B. Tursun and K. Gunsalus). Prof. Dr. Matthias Selbach Katrina Meyer doi:10.1074/mcp.M115.053975; collabo Joao Miguel Parente Fernandes Scientists Djordje Vasiljevic Carlos Viera Dr. Michal Nadler-Holly Dr. DamirKoshi Imami Omerbasic Technical Assistants Dr. Henrik Zauber Martha Hergeselle Dr. Matthias Ziehm Christian Sommer

PhD students Secretariat Boris Bogdanow Sabine Froese Kamila Kutz Petra Haink

MDC Research Report 2016 61 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Francesca M. Spagnoli

Molecular and Cellular Basis of Embryonic Development

Understanding the establishment of cellular capacity for spontaneous regeneration and, identity is a major goal in stem cell biology and therefore, represents an ideal target for lineage reprogramming approaches (Heinrich regenerative medicine. The endoderm germ et al. 2015). The liver is an attractive tissue layer gives rise to a number of vital organs, including the lungs, liver, pancreas and intes- due to its close developmental origin with tine. This remarkable diversity derives from thesource pancreas, for generating regenerative new pancreatic capacity and β-cells, accessibility. Thus, a successful reprogramming a homogenous population of multipotent strategy of liver to pancreas potentially offers cells. The aim of our research is to elucidate a solution to the problem, providing a renew- the fundamental principles that establish and - betes. Lineage reprogramming has its funda- maintain pancreatic cell identity. One of the mentsable source in developmental of β-cells for biology cell-therapy and is of based dia challenges in my laboratory is to understand on the knowledge of transcriptional networks how distinct cell types, such as liver and pan- underlying the establishment of cellular iden- creas, arise from common endoderm progen- identity of liver and pancreas progenitor cells itors in the embryo and acquire specialized isolatedtity. We usedfrom RNASeq the mouse to define embryo the at molecular the time shape to form functional organs. This fasci- of their lineage divergence (Fig. 1). By inte- nating question in developmental biology has grating temporal and spatial gene expression direct clinical relevance. In the long run, our - naling signatures in hepatic and pancreatic investigations will have implications in devel- progenitorsprofiles, we (Rodriguez-Seguel defined mutually exclusiveet al., 2013). sig oping novel strategies to generate pancreatic Our results have provided us with guidelines ß-cells either from programming of stem cells for generating hepatocytes and pancreatic cells from ES and iPS cells and also for for- or re-programming of hepatic cells, closing mulating liver-to-pancreas reprogramming the gap between studies of basic processes in model systems and clinical research. non-canonical Wnt signaling signature in the emergencestrategies. Notably,of pancreas we versus identified liver afrom unique en-

Comprehensive analysis of the he- transcriptional co-regulators, such as the patic and pancreatic lineage diver- doderm progenitors. In addition, we found- gence ferentially expressed between pancreatic and The pancreas consists of two organs in one: TALE homeoproteins TGIF and Pbx, being dif- the exocrine pancreas, which produces diges- tors (e.g. Setd7 histone methyltransferase) tive enzymes, and the endocrine compart- andhepatic lncRNAs, progenitor as potential cells; epigeneticdynamic regulators modula ment, which houses the insulin-secreting of cell identity. Finally, our results provided an - invaluable resource for lineage tracing analy- - sis to pinpoint the exact origin of the hepato- ulators,β-cells andand istheir critical loss foror dysfunction blood glucose results ho pancreatic lineage and isolate progenitor cell inmeostasis. diabetes, The a still β-cells incurable are key disease. metabolic Finding reg populations. Ongoing genetic lineage tracing experiments in the mouse aim at establishing with diabetes is a challenge of top clinical pri- the exact contribution of distinct progenitor ority.ways toThe restore adult pancreaticpancreas has β-cells a very in patients limited domains to liver and pancreas.

62 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Tissue-architecture in the developing pancreas During organogenesis, tissue-architecture de-

- sequentfines particular differentiation. local microenvironments, We, and others, showed which thatin turn cells impact at the progenitor branched specificationtips of the pancreatic and sub epithelium undergo self-renewal ensuring the expansion of the progenitor reservoir, while cells at the trunk become restricted toward the

et al. 2013). Thus, it is conceivable that different accessendocrine to thelineage, surroundings giving rise underliesto β-cells (Petzoldthe fate differences between tip and trunk pancreatic cell populations. To date, how specialized “mes-

within the pancreas epithelium is largely unknown.enchymal Novel niches” lines influenceof research lineage in my laboratory allocation are aimed at i. investigating how intrinsic epi-

its surrounding mesenchyme and ii. addressing thethelium requirement modulators of influencemesenchymal the crosstalk cells in withepi-

Ultimately, better comprehension of these con- ceptsthelial will morphogenesis have practical and implications β cell differentiation. in future regenerative therapies of diabetes.

Figure 1. (A) Schematic of hepatic and pancreatic cell lineage specification (1) and conversion (2). (B-D) RNASeq profiling of foregut, Patents / Patent applications pancreatic and hepatic progenitors isolated from Tg(Prox1-EGFP)

mouse embryos by FACS at distinct developmental stages (E8.5- E10.5 interval). Abbreviations: BP, bipotent progenitor; fg, foregut; TGIF2-INDUCED REPROGRAMMING OF HEPATIC CELLS TO PANCREATIC hf, head fold; lv, liver; pa, pancreas; dp, dorsal pancreas; vp, ventral PROGENITOR CELLS AND MEDICAL USES THEREOF pancreas. Adapted from Rodriguez-Seguel et al. 2013. Lead Inventor: Dr. Francesca M. Spagnoli. SelectedEuropean Patent Publications EP14177287.1, filed on July 2014. Lineage reprogramming to generate functional pancreatic cells N, Andrade-Navarro MA, Bonifacio E, Spagnoli FM. Step-wise reprogramming of liver Our previous studies laid the basis for formu- Cerdá-Estebanto pancreas progenitor N, Naumann state N, by Ruzzittu, a single transcriptionalS, Cozzitorto, C., regulator. Pongrac IM,Nature Hommel Communi A, Mah- cations lating reprogramming strategies between liver Kofent J, Zhang J, Spagnoli FM. Setd7 is a regulator of pancreatic cell identity. and pancreatic cells. Targeted functional stud- Develop. 2017ment 8:14127 doi:10.1038/ncomms14127. ies of developmental regulators of the liver Heinrich C, Spagnoli FM* and Berninger B.* and pancreas fate decision and their lineage Nature Cell Biology.. 2016 Oct 1;143(19):3573-3581. reprogramming activity are ongoing in my Simaite D, Kofent J, Gong M, Ruschendorf F, Jia In S, vivo Arn reprogramming P, Bentler K, Ellaway for tissue C, Kühnen repair. P, Hoffmann GK, Blau N, 2015 Spagnoli Mar;17(3):204-11. FM, Hübner N, *co-correspondingRaile K. Recessive mutations authors in PCBD1 gene cause a new type of early-onset diabetes. Diabetes. - creasgroup. versus We identified liver fate the decision TALE inhomeoprotein the embryo. Wang Y, Chen W, Andrade-Navarro MA, Spagnoli FM. Mutually2014 Oct;exclusive 63:3557-64. signaling Consistently,TGIF2 as an forced intrinsic expression regulator of ofTgif2 the is pansuf- Rodríguez-Seguel E, Mah N, Naumann H, Pongrac IM, Cerdá-Esteban N, Fontaine JF, Genes & Development Petzoldsignatures KM, define Naumann the hepatic H, Spagnoli and pancreatic FM. Rho signalling progenitor restriction cell lineage by divergence. the RhoGAP pancreatic progenitor state both in vivo and ex Stard13 integrates growth. 2013 and Sep morphogenesis 1;27:1932-46. in the pancreas. Development. 2013 vivoficient (Cerda-Esteban to convert differentiated et al. 2017). liverThis cellsstepwise to a Jan 1;140(1):126-35. a novel strategy for controlled generation of Group Leader pancreaticTGIF2-dependent progenitors fate conversionfrom liver and represents a start- Dr. Francesca Spagnoli

Scientist Silvia Ruzittu idealing point platform for production for dissecting of β-cells the intermediate (European Dr. Anne Sophie Escot David Willnow stepsPatent of EP14177287.1).liver-to-pancreas Moreover, interconversion this is and an Juan Zhang understanding the molecular basis of cellular PhD students - Corinna Cozzitorto Technical Assistants tive, we will build on this knowledge i. to im- Julia Kofent* Heike Naumann identity and plasticity. In a long-term perspec (part of the period reported) Laura Mueller Secretariat andprove ii. the efficiency of reprogramming and- (part of the period reported) cytematuration models of for pancreatic-reprogrammed studying fate conversion. cells; to apply our findings to human hepato Iris Apostel-Krause

MDC Research Report 2016 63 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Mathias Treier

Genetics of Metabolic and Reproductive Disorders

Our research is focused on the regulation malian organism despite the fact that until of mammalian physiology by genetic recently genome engineering in mice has been very time consuming. The discovery variation. Transcriptional regulators - constitute the cornerstones of our research mentation for mammalian genome editing activities. In particular, we try to understand hasof the turned CRISPR/Cas9 out to be system a game and changer its imple with how metabolic changes modulate the respect to speed and sophistication in gen- transcriptional response of organs in a quest for maintaining body homeostasis systemerating has DNA reduced modifications the time infor thegenerating mouse during adaptation to novel environmental germgenome. line Employment mutations down of the to CRISPR/Cas9 the genera- conditions. We utilize a plethora of mouse tion time of the animal species investigated. With this technology it is now possible to models for human diseases that we create study polygenetic traits by simultaneously ourselves that allow us to dissect even introducing multiple genetic polymor- complicated physiological questions at the phisms in a single step into the murine ge- - organismal level. print behind. This application that we term FunctionalGenomeWideAssociationStudiesnome without leaving a molecular finger (FGWAS), will allow us in the future to directly evaluate the contribution of various FGWAS using CRISPR/Cas9 combinations of genetic polymorphisms transgenic technology in an isogenic background with respect to Most cases of noncommunicable or chronic development and progression of chronic diseases including cardiovascular disease diseases. and diabetes are the result of mainly unknown combinations of polygenetic varia- Genetic dissection of the tions in the genome of the affected individ- renocardiac syndrome uals rather than the consequence of a single Cardiovascular disease is still the major gene disruption which is the hallmark of noncommunicable disease type in west- rare diseases. Although GenomeWideAssoci- ern societies. The CardioRenal Syndrome ationStudies (GWAS) - (CRS) is within the cardiovascular disease ora of genetic polymorphisms associated complex a subclass of disease entities that with the development have of noncommunicable identified a pleth - diseases their predictive value for devel- cal disorder of the heart and the kidneys opment of the disease later in life remains inis definedwhich acute as a or complex chronic pathophysiologi dysfunction in limited. This is mainly due to the fact that it one organ affects the function of the other. was not possible to experimentally validate which combination of predisposing genetic one of which is the renocardiac syndrome polymorphisms are most deleterious for whereCRS can heart be subdivided damage isin indirectlyfive major elicitedclasses disease development. through acute or chronic kidney disease. Modifying the genome of mice has become The underlying molecular mechanisms of a routine procedure which has made the this organ crosstalk being still incomplete mouse the most genetically studied mam- understood.

64 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Fig. 1: Disease modeling of the renocardiac syndrome in mice. The degeneration of the kidney overtime is mirrored by increased hyperplasia of the heart (heart top row, kidney below at different time points of disease progression).

Based on our discovery that mutations in mutations whose lethality is dependent on - tor, causes nephronophthisis, a progressive chronicGLIS2, akidney zinc finger disease transcriptional (CKD), in mice regula and toloss acceleration of GLIS2 function. of CKD progressionWe have successfully which al- lowsidentified us now novel to modelgenetic the modifiers chronic thatrenocar lead- genes of the renocardiac disease spectrum basedmen, we on have the conceptstarted toof screensynthetic for modifierlethality. Employing FGWAS - Synthetic lethality describes the observation genicdiac syndrometechnology, in we mice are now efficiently able to (Fig.system 1).- that the combination of mutations in at least atically dissect any CRSand subtypeCRISPR/Cas9 on a genetic trans two genes is lethal whereas each mutation on its own is said to be viable. We are in par- novel targets for pharmacological interven- ticular interested to identify novel genetic tionlevel in that cardiovascular should lead todisease. the identification of

MDC Research Report 2016 65 CARDIOVASCULAR AND METABOLIC DISEASE

Fig. 2: Murine blastocysts stained with antibodies marking the three cell lineages of the pre-implantation embryo: epiblast (Nanog/green), primitive endoderm (Gata6/red), trophoectoderm (Cdx2/yellow).

Fig. 3: Identification of neuronal enhancer elements employing BAC (Bacterial Artifical Chromosome) mouse transgenesis. Shown is the endogenous expression pattern of BSX in the mouse hypothalamus visualized by Histone2B- GFP expression with superimposed expression of BAC reporters (Histone2B-RFP) harboring the BSX locus. Note, the enhancer element for the DMHc (compact region of the Dorsomedial hypothalamus) is missing on the BAC-2 reporter (ARC = arcuate nucleus).

66 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Stem cells, plasticity and vous system (CNS) that are involved in the regeneration (dys)regulation of physical activity and The great interest in stem cell biology is sleep, major factors that can contribute to mainly due to the intrinsic ability of stem - tential to differentiate into all cell types of wethe combine development light sheet of obesity microscopy in humans. with tis In- thecells body. to indefinitely The study self-renew of early mouse and their devel po- sueorder clearing to deconstruct and enhancer the BSX mapping CNS network using opment has revealed a lot of basic insights - into the molecular mechanisms underlying ogy to delineate a structure-function corre- stem cell self-renewal and plasticity (Fig. lationCRISPR/Cas9 with respect and BACto neuronal transgenic populations technol 2). These insights are now applied in adult that modulate physical activity and sleep stem cell biology and utilized in regenera- patterns (Fig. 3). We functionally probe the tive medicine for the extrinsic activation of endogenous organ repair processes. DREADD technology (Designer Receptors Along these lines we have been among the Exclusivelyidentified neuronalActivated subpopulationsby Designer Drugs using). DREADDs are engineered G-protein cou- reprogrammed in vivo through the dem- pled receptors which are activated by oth- onstrationfirst showing of somaticthat cells sex can reversal functionally in adult be erwise inert drug-like small molecules. This technology allows us to activate or silence with SALL4 the central transcriptional reg- neuronal networks in free running animals ulatorfemale of mice. stem Furthermore, cell pluripotency. we identifiedUtilizing thereby directly testing their involvement in a proteomic approach we have delineated the regulation of physical activity and sleep. a SALL4 centered protein network in stem cells and currently investigate the molecular function of so far uncharacterized pro- Selected Publications teins in this network. This work has led to Heude É, Bellessort B, Fontaine A, Hamazaki M, Treier AC, Treier M, Levi G, Narboux-Nême N. (2015) Etiology of craniofacial malformations in mouse with extended developmental potential, models of blepharophimosis, ptosis and epicanthus inversus syndrome. Hum Mol whichthe identification now allows us of to embryonic study in greater stem cells de- Bellessort B, Bachelot A, Heude É, Alfama G, Fontaine A, Le Cardinal M, Treier M,Genet.24(6):1670-81 Levi G. (2015) Role of Foxl2 in uterine maturation and function. Hum Mol Genet. embryonic stem cell plasticity which may 24(11):3092-103. havetail the important epigenetic implications modifications for underlying applied Fortin J, Boehm U, Deng CX, Treier M, Bernard DJ. (2014) Follicle-stimulating stem cell therapy in regenerative medicine. 410. Attanasiohormone synthesis M, Uhlenhaut and fertility NH, Sousa depend V, O’Toole on SMAD4 JF, Ottoand FOXL2.FASEB E, Anlag K, Klugmann J.28(8):3396- C, Voluntary physical activity and Treier AC, Helou J, Sayer JA, Seelow D, Nürnberg G, Becker C, Chudley AE, Nürn- sleep disorder berg P, Hildebrandt F & Treier M Obesity is the major risk factor for develop- 1024. (2007) Loss of GLIS2 causes nephronophthisis in ment of cardiovascular disease and diabe- Sakkouhumans M,and Wiedmer mice by increased P, Anlag K,apoptosis Hamm A,and Seuntjens fibrosis. Nature E, Ettwiller Genetics L,Tschop 39:1018- M & tes. Physical activity is a key component of Treier M energy expenditure and therefore of energy Hyperphagia and Locomotory Behavior. Cell Metabolism 5:450-463 balance. Sports-like exercise has a strong (2007). A Role for Brain-Specific Homeobox Factor Bsx in the Control of obesity protective effect in humans. Howev- er, why some people voluntary exercise and Group Leader PhD students others do not remains an open question. We Prof. Dr. Mathias Treier Duygu Akcay have proposed in our initial work on the Mehrshad Aschmann analysis of BSX (Brain-S Homeobox) Scientific Coordinator Jenny Rätzer gene function that differences in voluntary Dr. Carsten Kallfaß Fabio Enrique Rojas Rusak physical activity may havepecific an underlying Yu-Ting Tseng genetic cause. BSX was therefore named Scientists by others the “fidgeting BSX expres- Dr. Anna-Corina Treier Technical Assistants sion marks a neuronal network within the Franziska Block hypothalamus of the brain” gene. that orchestrates Christian Klasen Antje Susan Brouwer-Lehmitz eating behavior, sleep and higher cogni- Dr. Xiushan Yin Madlen Ebert tive functions. Currently, we are interested Janine Teller in the neuronal circuits of the central ner-

MDC Research Report 2016 67 CARDIOVASCULAR AND METABOLIC DISEASE Photo: David Ausserhofer/MDC Photo:

Thomas Willnow

Molecular Cardiovascular Research

We study endocytic receptors to uncover Whereas studies on the LDL receptor gene their roles in normal physiological processes - and in human disease. We demonstrated tionfamily of yielded another insights group intoof receptors, the significance called that LRP2, a member of the LDL receptor VPS10Pof endocytosis domain for receptors, cell signaling, directed identifica our at- gene family, balances signaling of the mor- tention to intracellular protein transport phogen sonic hedgehog in the developing as a cause of human disease. VPS10P domain receptors share structural similarity retina, and that lack of this activity results to sorting receptors in yeast, suggesting in severe overgrowth of the eyes in mouse their involvement in intracellular protein models and patients. We also studied transport in mammalian cell types. While SORLA, an endocytic receptor expressed in studies, including our own work, the many neurons. We showed that this receptor con- functionsthis hypothesis of VPS10P was confirmeddomain receptors by earlier in trols distinct steps in the catabolism of Aβ, human (patho)physiology still remain to be the causative agent in Alzheimer’s disease, and that SORLA dysfunction is a cause com- LRP2identified. is an auxiliary SHH receptor mon to sporadic and familial forms of this essential for eye development disease. Annabel Christ, Julia Klippert, Anna Christa

LRP2 is a receptor expressed in the neural Introduction tube, the embryonic tissue that gives rise Receptor-mediated endocytosis is the main to the CNS. Early on, we recognized the im- mechanism for selective transport of mac- portance of LRP2 for brain development, as

has been made in elucidating the steps of holoprosencephaly, a defect in separation endocytosisromolecules atinto the cells. cellular Significant level. However,progress ofmice the deficient forebrain for hemispheres this receptor and suffer the mostfrom the physiological relevance of many endo- common forebrain anomaly in humans. The cytic pathways for organ (dys)function remains elusive. The main class of endocytic receptors is defectssignificance as the of cause this receptor of Donnai-Barrow in humans synwas- the LDL receptor gene family that mediates drome,confirmed a monogenic by others, disorder demonstrating characterized LRP2 uptake of lipoproteins in many cell types (Fig. 1). Surprisingly, previous studies un- - covered additional functions performed by lecularby forebrain mechanism malformations. of LRP2 action In our in earlier brain these receptors, changing our perception developmentwork, Annabel by Christdemonstrating identified that the it acts mo of lipoprotein receptors from mere cargo as an auxiliary receptor for the morphogen transporters to key regulators of many LRP2 facilitates signaling of SHH through - thesonic receptor hedgehog Patched (SHH). 1, Ina step the essentialneural tube, for onicphysiological development processes. is noteworthy. In particular, Yet, stud the- proper forebrain patterning (Christ et al., iessignificance so far provided of these just receptors a glimpse for at embry their Dev Cell 2012). manifold contributions to developmental Now, Annabel and her colleagues showed processes, with many details still to be un- that the role of LRP2 in SHH signaling ex- covered. tends to eye development (Christ et al., Dev

68 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Fig. 1: Structural organization of mammalian members of the LDL receptor and VPS10 domain receptor (Sortilins) families.

- the retinal margin as cause of the large eye veloping retina, LRP2 mediates endocytic phenotype in patients with Donnai-Barrow clearanceCell 2015). of TheySHH toidentified antagonize that the in action the de of syndrome. this morphogen. LRP2-mediated clearance prevents spread of SHH from the central SORLA controls production of neu- retina into the retinal margin, and protects rotoxic Aβ peptides quiescent progenitor cells in this ciliary Safak Caglayan, Vanessa Schmidt marginal zone (CMZ) from mitogenic stimuli. Loss of LRP2 in mice increases the sen- - sitivity of the CMZ for SHH, causing aber- teolytic processing of the amyloid precur- rant expansion of the retinal progenitor cell sorAβ is protein a neurotoxic (APP). peptide Excessive produced production by pro of pool and hyperproliferation of the retinal - margin. As a consequence, the size of the tion and memory impairment in patients eye globe increases, a phenotype referred Aβ is responsible for the synaptic dysfunc - - ment the ability of LRP2 to act, in a context sentwith Alzheimer’spromising targets disease for(AD). prevention Accordingly, of dependentto as buphthalmos. manner, as These an activator findings or inhib docu- thismechanisms devastating that disease. reduce Previously, Aβ levels Vanes repre- itor of the SHH pathway. Furthermore, they sa Schmidt in the lab documented the abili- uncovered LRP2 activity as the molecular ty of SORLA to act as an intracellular sorting mechanism imposing quiescence of the receptor for APP in neurons that prevents retinal margin in the mammalian eye, and accumulation of APP in endosomes where they suggest SHH-induced proliferation of

processing into Aβ occurs. These findings MDC Research Report 2016 69 CARDIOVASCULAR AND METABOLIC DISEASE

suggested low SORLA levels seen in individuals with AD as the underlying cause

EMBO J 2012). Toof enhancedprovide proofAβ production of concept (Schmidt that overex et al.,-

vivo and represents a therapeutic target forpression AD, Safak of SORLACaglayan reduces generated Aβ mice levels that in

hypothesis, increasing levels of SORLA in overexpress this receptor. In line with our in these animals. Exploring the underlying molecularthe brain profoundlymechanism, reducedSafak, Vanessa, Aβ burden and their colleagues made the surprising observation that SORLA not only acts as sorting receptor for APP but also moves newly

degradation, further reducing the amount ofsynthesized neurotoxic Aβ peptides molecules released to lysosomes by neurons for (Caglayan et al., Science Trans Med, 2014). They also documented that the ability of - tant expressed in a familial form of AD. Tak- SORLA to sort Aβ is lost in a receptor mu SORLA as a potential drug target in AD and en together, these findings substantiated function as a possible cause of early-onset familialidentified AD. loss of the receptor’s Aβ sorting

SORLA-mediated trafficking is a common cause of sporadic and familial forms of AD Sonya Dumanis, Tilman Burgert, Safak Caglayan, Anne-Sophie Carlo

Fig. 2: Expansion of retinal progenitor cell pool and Given the importance of SORLA for amy- eye overgrowth in LRP2-deficient mice. (A) Increased loidogenic processes, Sonya Dumanis, Til- size of the eye globe in adult mice lacking LRP2 (Lrp2-/-) man Burgert, and Safak Caglayan set out to compared to wild type animals (Lrp2+/+). (B, C) Increase elucidate the molecular mechanisms con- in progenitor cell number and proliferative capacity in the trolling SORLA-dependent sorting of APP ciliary marginal zone (encircled) of Lrp2-/- compared to Lrp2+/+ mouse embryos (embryonic day 16.5) as shown by were based on our earlier observations in BrdU incorporation (B) and staining for marker VSX2 (C). celland lines Aβ in that the the brain interaction in vivo. of Their SORLA studies with Scale bars: 50 µm. two cytosolic adaptors, termed retromer and GGA, is required for receptor sorting to and from the trans-Golgi network (TGN). The TGN is a central hub for protein sorting. However, the relevance of anterograde or

SORLA activity in vivo and for progression ofretrograde AD remained trafficking unexplored. at this Now, organelle Sonya and for

70 MDC Research Report 2016 CARDIOVASCULAR AND METABOLIC DISEASE

Fig. 3: SORLA trafficking pathways in Alzheimer’s disease. GGAs mediate anterograde sorting of SORLA and promote Aβ catabolism in lysosomes. A mutation in SORL1, disrupting Aβ binding, causes a familial form of AD (FAD). Retromer mediates retrograde sorting of SORLA/APP complexes to the trans-Golgi network (TGN). Disrupting this pathway enhances APP processing in endosomes. Low levels of SORLA or retromer components are risk factors for late-onset AD (LOAD).

her collaborators generated novel mouse models expressing SORLA variants lacking binding sites for GGA or retromer to query this concept. Disruption of retromer bind- Selected Publications ing resulted in a retrograde sorting defect Christ, A., Christa, A., Klippert, J., Eule, J.C., Bachmann, S., Wallace, V.A., Hammes, with accumulation of SORLA in endosomes A. and T.E. Willnow. (2015). LRP2 Acts as SHH Clearance Receptor to Protect the and depletion from the TGN, and in an over- Retinal Margin from Mitogenic Stimuli. Developmental Cell 35: 1-13. - Dumanis, S.B., Burgert, T., Caglayan, S., Füchtbauer, A., Füchtbauer, E.-M., Schmidt, V. and T.E. Willnow. (2015). Distinct Functions for Anterograde and Retrograde Sorting trast, disruption of the GGA interaction did of SORLA in Amyloidogenic Processes in the Brain. J Neurosci 35(37): 12703-12713. all enhanced APP processing rate. In con not impact APP processing but caused in- Glerup, S., Olsen, D., Vaegter, C.B., Gustafsen, C., Sjoegaard, S.S., Hermey, G., Kjolby, - M., Molgaard, S., Ulrichsen, M., Boggild, S., Skeldal, S., Fjorback, A.N., Nyengaard, uted to a defect in anterograde lysosomal J.R., Jacobsen, J., Bender, D., Bjarkam, C.R., Sørens, E.S., Füchtbauer, E.M., Eichele, G., creased brain Aβ levels, a mechanism attrib- Madsen, P., Willnow, T.E., Petersen, C.M. and A. Nykjaer. (2014). SorCS2 Regulates

sortingtargeting for of SORLA Aβ. Our activities new findings in vivo, substanand they Caglayan,Dopaminergic S., Takagi-Niidome, Wiring and Is Processed S., Liao, F., into Carlo, an A.-S., Apoptotic Schmidt, Two-Chain V., Burgert, Receptor T., Kitago, in Y., Peripheral Glia. Neuron 82(5): 1074-87. uncoveredtiated the significance that anterograde of adaptor-mediated and retrograde Füchtbauer, E.M., Füchtbauer, A., Holtzman, D.M., Takagi, J. and T.E. Willnow. (2014). sorting paths serve discrete receptor functions in amyloidogenic processes. Lysosomal Sorting of Amyloid-β by the SORLA Receptor Is Impaired by a Familial M.N.,Alzheimer’s Olivecrona, Disease G. and Mutation. T.E.Willnow. Sci Transl (2016). Med SORLA 6(223): facilitates 223ra20 insulin receptor Outlook Schmidt, V., Schulz, N., Yan, X., Schürmann, A., Kempa, S., Kern, M., Blüher, M., Poy, Dysregulation of protein uptake and intra- signaling in adipocytes and exacerbates obesity. J Clin Invest 126(7): 2706-20. cellular transport emerges as a molecular mechanism of major importance in many Group Leader human disease processes. Among other Prof. Dr. Thomas Willnow Aygul Subkhangulova functions, faulty protein transport impacts Ina-Maria Rudolph cellular signal reception (as for SHH in Scientists Technical Assistants brain and eye) or results in abnormal pro- Dr. Anne-Sophie Carlo-Spiewok Kristin Kampf Dr. Annabel Christ Christine Kruse the brain). Our data have shown that endo- Dr. Hyoungseok Ju Melanie Liekweg cytictein homeostasisreceptors play (as key for roles Aβ catabolismin these path in- Dr. Anna Malik Tatjana Pantzlaff ological processes. Our future work aims at Dr. Vanessa Schmidt Secretariat PhD students Verona Kuhle expressionrefining the of molecular target molecules, details of and how how these re- Antonino Asaro ceptorreceptors dysfunction affect trafficking contributes and to functional the most Lena Bunatyan Animal Attendant devastating defects in brain development Julia Klippert Maria Schmeißer

and function inflicting humans. MDC Research Report 2016 71 72 MDC Research Report 2016 Cancer Research

Coordinator: Claus Scheidereit

MDC Research Report 2016 73 Cancer Research

Claus Scheidereit

ancer is a serious health burden seminal contributions to our understand- with nearly half a million cases diag- ing of chemokine receptors in immune re- C nosed in Germany each year, making action, lymphoid organogenesis and metas- advancements in diagnosis, treatment and tasis. prevention of cancer a major challenge. Our mission is to understand the principles of cancer development and progression to Selected scientific highlights provide the basis for improved diagnosis and treatment. One aim is to discover and As an exciting new genome editing tool, characterize genes and cellular components that are important in the development and previous gene targeting methods. None- treatment of cancer and that are promising theless,CRISPR-Cas9 most ismutagenic faster and events cheaper are small than as novel targets for therapies. Another aim deletions, while sequence replacements, is to understand and therapeutically exploit required for a precise creation or correc- interactions of the immune system with tu- tion of e.g. cancer related mutations, occur mors. at lower frequency. The groups of Klaus Rajewsky and Ralf Kühn found a way of Cancer Research at the MDC covers the favoring the homology-directed introduc- topics (i) signaling pathways, cell biology - and cancer, (ii) structural and functional pression of the competing NHEJ DNA repair genomics and (iii) tumor immunology and pathwaytion of precise which DNAcauses modifications sequence deletions by sup combines a spectrum of interdisciplinary (Chu et al., Nature Biotechnol., 2015). basic research laboratories and clinically oriented groups. The group of Walter Birchmeier found that impairment of the tyrosine phospha- Michela Di Vir- tase Shp2 in PyMT mammary glands of gilio and Gaetano Gargiulo were recruited mice induces senescence, which blocks tu- asIn theHelmholtz reporting junior period, research group lead- mor formation. Src, Fak and Mapk inhibit ers from The Rockefeller University in senescence by activating expression of - Skp2, Aurka, and Delta-like 1, which block stitute in Amsterdam, respectively. Both p27 and p53. Expression of Shp2 and of alsoNew received York and European the Netherlands Research Cancer Council In target genes predicts human breast cancer (ERC) starting grants. Michela di Virgilio outcome. Therapies that rely on senescence has made important contributions in the induction may block breast cancer (Lan et al., EMBO J, 2015). mechanisms. Her research is relevant to the understandingfield of DNA double of the pathogeneses strand break of repair im- The RNA destabilizing protein RC3H1 (also - - giulo established new procedures to char- nity. The groups of Udo Heinemann and acterizemunodeficiencies cancer genes and incancer. gliomagenesis Gaetano Garand Markusknown as Landthaler ROQUIN) repressespresented autoimmu the crys- contributed to understanding the role of tal structure of a Roquin domain, gain- tumor cell identity in lung cancer. He now ing insight into the mode of RNA binding develops new approaches to combat glio- a collaboration with several MDC groups, CNRS-associated immunologist Michael they(Schuetz demonstrated et al., Nature that Commun., RC3H1 modulates 2014). In Siewekeblastoma. hasIn 2014,been electedthe MDC- member and Inserm/ of the European Molecular Biology Organization contributes post-transcriptional regulation (EMBO). Martin Lipp received the German ofthe the activity DNA damageof the IKK/NF-κB response (Murakawa pathway and et Cancer Award (2017) in recognition of his al., Nature Commun., 2015).

74 MDC Research Report 2016 CANCER RESEARCH

Processing of the precursor proteins p100 Macrophages can repopulate themselves by and p105 is fundamental to activation of self-renewing, but the underlying genetic Claus Scheidereit, programs have remained unknown. Mi- Gunnar Dittmar, Jana Wolf could dem- chael Sieweke and colleagues could dem- onstrateNF-κB. The that groups p100 acts of upstream of p105, onstrate that the transcription factors MafB resulting in concurrent stimulus-induced and c-Maf repress the enhancers of genes production of p50 and p52. Both precur- regulating self-renewal. When macro- sors bind to segregase (p97/VCP), which phages need to proliferate, they transiently promotes proteasomal processing. The decrease expression of Maf transcription - factors. A parallel pathway also operates to etry-based quantitative mathematical mod- control the self-renewal of embryonic stem elsfindings (Yilmaz are et supported al., Cell Reports, by mass 2014). spectrom cells (Soucie et al., Science, 2016).

At the endoplasmic reticulum (ER), aber- Adoptive T cell therapy is a promising al- rantly folded proteins are recognized and exported for degradation by proteasomes. recognize malignant cells via abnormal The group of Thomas Sommer has shown surfaceternative epitopes in the fight generated against by cancer. proteasomal T cells how substrates are targeted for destruc- degradation of mutant proteins. Thomas tion. A key ER membrane component is Blankenstein and colleagues found that Der1 which threads misfolded proteins - from substrate receptors in the ER lumen pendent trimming. Tumor cells with im- through the membrane to ubiquitylation certain epitopes require interferon-γ de- enzymes in the cytoplasm (Mehnert et al. duce such epitopes and thereby escape Nat Cell Biol., 2014). T-cellpaired recognition. interferon-γ Thus, response future doT cell not thera pro- py applications should utilize epitopes that The mechano-chemical GTPase dynamin oligomerizes around the neck of clathrin- trimming (Textor et al., J Exp Med, 2016). coated vesicles and catalyzes membrane are independent of interferon-γ mediated scission in a GTPase-dependent fashion. The group of Clemens Schmitt The group of Oliver Daumke has explored small molecules that promote re-expres- structure and function of the dynamin tet- sion of B-cell phenotype markers identified that are ramer, providing detailed insights into its assembly mode and regulation. The study When applied in a combination treatment, may explain how dynamin mutations can cHLlost incells classical became Hodgkin’s susceptible lymphoma to CD20 (cHL). anti- cause Centro-Nuclear Myopathy or Char- body-mediated apoptosis (Du, et al., Blood, cot-Marie-Tooth Disease (Reubold et al., 2016). Nature, 2015). The group of Oliver Rocks Dendritic cells (DCs) have pro- and anti-tu- novel PKA antagonist, the Rho GTPase ac- morigenic functions. Using a Myc-driven B identified a cell lymphoma model, Uta Höpken, Armin PKA catalytic subunit as a pseudosubstrate Rehm and coworkers showed that DCs andtivating targets protein it for ARHGAP36. rapid ubiquitin-mediated It inhibits the upregulate immunomodulatory cytokines, endolysosomal degradation. PKA inhibi- growth factors and the CCAAT/enhancer tion by ARHGAP36 promotes derepression of Hedgehog signaling, thereby providing a simple rationale for the upregulation of AR- essentialbinding proteinfor the βcreation (C/EBPβ). of a They lymphoma found HGAP36 in medulloblastoma (Eccles et al., growththat C/EBPβ-controlled promoting and immunosuppressive DC functions are Nat Commun, 2016). niche (Rehm et al., Nature Commun., 2014).

MDC Research Report 2016 75 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Walter Birchmeier

Signal Transduction in Development and Cancer

Our research is primarily focused on two - signaling systems: Wnt/beta-catenin and lished the following investigations: In the present report period, we have pub receptor tyrosine kinase signaling. Com- Shp2/Mapk signaling controls ponents of both signaling systems are goblet/paneth cell fate decisions frequently mutated or deregulated in a in the intestine variety of developmental disorders and J. Heuberger, F. Kosel, J. Qi, K. Grossmann, K. Rajewsky, and W. Birchmeier cancers in humans. In the previous years,

the laboratory has studied adhesion and - signaling of E-cadherin/Wnt/beta-catenin testine, Notch and Wnt/beta-catenin signals by biochemical means. We have shown controlIn the developmentstem cell maintenance of the mammalian and their dif in- ferentiation into absorptive and secretory that beta-catenin binds to the transcription cells. Mechanisms that regulate differentia- factors Lef1/Tcf, and that this translocates tion of progenitors into the three secretory beta-catenin to the nucleus. Beta-catenin is lineages, goblet, paneth or enteroendocrine recruited to its degradation complex con- cells, are not fully understood. Using conditional mutagenesis in mice, we observed that taining Axin2/Conductin and APC. We have Shp2-mediated Mapk signaling determines also investigated the role of scatter factor/ the choice between paneth and goblet cell hepatocyte growth factor (SF/HGF) and its fates, and also affects Lgr5+ stem cells. Ab- lation of the tyrosine phosphatase Shp2 in receptor, the Met tyrosine kinase, in the the intestinal epithelium reduced Mapk sig- morphogenesis of epithelial cells. Signals naling and led to a reduction of goblet cells of Met are transmitted by the cytoplasmic while promoting paneth cell development. multi-adapter Gab1 and the tyrosine phos- Conversely, conditional Mek1 activation rescued the Shp2 phenotype, promoted goblet phatase Shp2. Conventional ablations of cell and inhibited paneth cell generation. The beta-catenin and Gab1 in mice result in gas- Shp2 mutation also expanded Lgr5+ stem cell trulation defects and embryonic organ fail- niches, which could be restricted by activated ures, respectively. In recent years, we have Mek1 signaling. Changes of Lgr5+ stem cell quantities were accompanied by alterations examined the role of Wnt/beta-catenin and of paneth cells, indicating that Shp2/Mapk Met/Gab1/Shp2 by conditional mutagenesis signaling might affect stem cell niches direct- in mice. Beta-catenin regulates precursor ly or via paneth cells. Remarkably, inhibition and stem cells in the nervous system, the of Mapk signaling in intestinal organoids and cultured cells changed the relative abundance hair and the heart; Gab1 and Shp2 control of Tcf4 isoforms and by this, promoted Wnt/ precursor and stem cells in skin, limbs, liver beta-catenin activity. The data thus show that and kidney. Met regulates wound healing Shp2-mediated Mapk signaling controls the in the skin. Activation of beta-catenin and choice between goblet and paneth cell fates by regulating Wnt/beta-catenin activity (pub- HGF/Met in adult mouse tissues induces lished in Proc. Natl Acad. Sci. USA 111, 3472- tumors and cancer stem cells. 3477, 2014).

76 MDC Research Report 2016 CANCER RESEARCH

A. PyMT mammary gland tumors undergo senescence on Shp2 ablation. B,C. SA-beta-gal staining of spheres by control and Shp2 knock-out cells and quantification of senescence. D. Growth of tumors in control and GS493-treated mice. E. Kaplan-Meier survival analysis of human breast cancer patients based on cell cycle genes regulated by Shp2. F. Scheme of Shp2-dependent signaling systems and target genes involved in senescence and cell cycle programs (EMBO J. 2015).

Shp2 signaling suppresses senes- Gab1 and Mapk signaling are es- cence in PyMT-induced mammary sential in the hair cycle and hair gland cancer in mice follicle stem cell quiescence L. Lan, J.D. Holland, J. Qi, S. Grosskopf, J. Ö. Akilli Öztürk, H. Pakula, J. Chmielowiec, J. Rademann, R. Vogel, B. Györffy, A. Wulf-Gold- Qi, S. Stein, L. Lan, Y. Sasaki, K. Rajewsky, and enberg, and W. Birchmeier W. Birchmeier

Gab1 is a scaffold protein that acts down- from cell biology, biochemistry, and genet- stream of receptor tyrosine kinases. Here, icsIn thisto investigate study, we the have role used of the techniques tyrosine we produced conditional Gab1 mutant phosphatase Shp2 in tumor cells of MMTV- mice (by K14- and Krox20-cre) and show PyMT mouse mammary glands. Genetic that Gab1 mediates crucial signals in the ablation or pharmacological inhibition of control of both the hair cycle and the self- Shp2 induces senescence, as determined renewal of hair follicle stem cells. Remark- by the activation of senescence-associated ably, mutant hair follicles do not enter beta-gal (SA-beta-gal), cyclin-dependent catagen, the destructive phase of the hair kinase inhibitor 1B (p27), p53, and histone 3 trimethylated lysine 9 (H3K9me3). Se- quiescence and become exhausted, and nescence induction leads to the inhibition thuscycle. no Instead, stem cell hair niches follicle are stem established cells lose in of self-renewal of tumor cells and blockage the bulges. Moreover, conditional sustained of tumor formation and growth. A signal- activation of Mapk signaling by expression - of a gain-of-function Mek1(DD) allele (by stream of Shp2 to counter senescence: Src, focaling cascade adhesion was kinase, identified and thatMap acts kinase down in- restores quiescence of the stem cells. Our hibit senescence by activating the expres- dataKrox20-cre) thus demonstrate rescues hair ancycle essential deficits roleand sion of S-phase kinase-associated protein of Gab1 downstream of receptor tyrosine 2 (Skp2), Aurora kinase A (Aurka), and kinases and upstream of Shp2 and Mapk the Notch ligand Delta-like 1 (Dll1), which in the regulation of the hair cycle and the block p27 and p53. Remarkably, the expres- self-renewal of hair follicle stem cells (pub- sion of Shp2 and of selected target genes lished in Cell Reports 13, 561-572, 2015). predicts human breast cancer outcome. We conclude that therapies, which rely on senescence induction by inhibiting Shp2 or controlling its target gene products, may be useful in blocking breast cancer (published

in EMBO J. 34, 1493-1508, 2015). MDC Research Report 2016 77 CANCER RESEARCH

A,B. The beta-catenin/TCF4 interaction inhibitor LF3 was identified by AlphaScreen-based HTS. C. LF3 reduced the association of beta-catenin to the Axin2 and c-Myc promoters. D. SW480 cells contain Wntlow and Wnthigh populations. E. LF3 reduced sphere formation only from Wnthigh cells. F. Tumor-free survival of mice injected with different cell populations. G,H. LF3 reduced tumor growth from Wnthigh cells and induced differentiation shown by IF (Cancer Res. 2015).

A small-molecule antagonist of the signal. We performed High Throughput the beta-catenin/TCF4 interaction Screening (HTS) using AlphaScreen and blocks the self-renewal of cancer - stem cells and suppresses tumori- ecules that disrupt this interaction. Com- genesis. poundELISA techniquesLF3, a 4-thioureido-benzenesulfon to identify small mol- L. Fang, Q. Zhu, M. Neuenschwander, E. amide derivative, was found as a robust Specker, A. Wulf-Goldenberg, W. I. Weis, J. P. inhibitor of beta-catenin/TCF4 interaction von Kries, and W. Birchmeier and thus a good starting point for drug

Wnt/beta-catenin signaling is essential for embryogenesis and tissue homeostasis, essentialdiscovery, core as suggested structure byand Lipinski’s residues “rulethat which has been highly conserved through of five”. Medicinal chemistry identified an evolution. Deregulation of Wnt/beta- LF3 inhibited Wnt/beta-catenin signals in catenin signals can initiate and promote cellscould with not exogenous be modified reporters in LF3. and Compound in colon cancer cells with endogenously high Wnt head and neck. As such, Wnt/beta-catenin activity. Compound LF3 inhibited features signalinghuman cancers, represents specifically an attractive of colon target and for of cancer cells related to Wnt signaling, incancer therapy. The interaction between cluding high cell motility, cell cycle progres- beta-catenin and the transcription factor sion, and the overexpression of Wnt target TCF4 is necessary in the transduction of genes. LF3 does not cause cell death or

78 MDC Research Report 2016 CANCER RESEARCH

interfere with cadherin-mediated cell-cell eration with the Department of Urology of adhesion. Remarkably, the self-renewal ca- the Charité Berlin, we characterise human pacity of cancer stem cells was blocked by kidney cancer stem cells and identify new LF3 in concentration-dependent manners, treatment strategies (Annika Fendler, to- as examined by sphere formation of colon gether with Klaus Jung and Jonas Busch). and head and neck cancer stem cells in non- attached conditions. LF3 reduced tumor role in Wnt-dependent cancer stem cells in growth and induced differentiation in colon headWe also and identify neck and histone colorectal modifiers cancers in their (in tumors in mouse xenografts. LF3 is thus a cooperation with Francis Stewart, Sylvie - Robine and Daniel Louvard). ing with the potential for further develop- mentspecific for inhibitor preclinical of and canonical clinical Wnt studies signal in cancer treatment (published in Cancer Res.

Cancer76, 891-901, stem 2016). cells regulate cancer-associated fibroblasts via activation of Hedgehog signaling Selected Publications in mammary gland tumors Giovanni Valenti, Hazel M. Quinn, Guus MAPK signaling controls goblet/paneth cell fate decisions in the intestine. Proc. Natl. J.J.E. Heynen, Linxiang Lan, Jane D. Holland, HeubergerAcad. Sci. USA, J, Kosel 111: F; 3472-3477. Qi J, Grossmann K, Rajewsky K, and Birchmeier W. 2014. Shp2/ Regina Vogel, Annika Wulf-Goldenberg, and Akilli Öztürk Ö, Pakula H, Chmielowiec J, Qi J, Stein S, Lan L, Sasaki,Y, Rajewsky K, Walter Birchmeier and Birchmeier W. 2015. Gab1 and Mapk signaling are essential in the hair cycle and hair follicle stem cell quiescence. Cell Reports 13: 561-572. Lan L, Holland JD, Qi J, Grosskopf S, Rademann J, Vogel R, Györffy B, Wulf-Goldenberg Many tumors display intracellular hetero- A, and Birchmeier W. 2015. Shp2 signaling suppresses senescence in PyMT-induced geneity, with subsets of cancer stem cells (CSC) that sustain tumor growth, recur- Qi J, Yu Y, Akilli Öztürk Ö, Holland JD, Besser D, Fritzmann J, Wulf-Goldenberg A, rence, and therapy resistance. Cancer asso- mammary gland cancer in mice. EMBO J. 34:1493-1508. promote experimental metastasis and migration of colorectal cancer cells through Eckertdifferent K, signals.Fichtner Gut, I, and 65, Birchmeier 1690-701, 2016.W. 2015. New Wnt/beta-catenin target genes support and regulate CSC function. Here we investigatedciated fibroblasts the interactions (CAF) have beenbetween shown CSCs to Kries JP, and Birchmeier W. 2016. A small-molecule antagonist of the beta-catenin/ and CAFs in mammary gland tumors driven FangTCF4 L,interaction Zhu Q, Neuenschwander blocks the self-renewal M, Specker of cancerE, Wulf-Goldenberg stem cells and A, suppresses Weis WI, von tu-

and Hgf/Met signaling in mouse mammary morigenesis. Cancer Res. 76, 891-901, 2016. by combined activation of Wnt/β-catenin Group Leader PhD students the hedgehog ligand SHH, which regulated Prof. Dr. Walter Birchmeier Johanna Grinat* CAFsepithelial via paracrinecells. In this activation setting, CSCs of Hedgehog secreted Adam Myszczyszyn* signaling. CAFs subsequently secreted fac- Scientists Jane Joy Thomas* (SignGene) tors that promoted expansion and self-re- Dr. Özlem Akilli-Öztürk* Hazel Quinn* Dr. Liang Fang with the Hedgehog inhibitor vismodegib Dr. Annika Fendler Master students reducednewal of CAFCSCs. and In vivoCSC treatmentexpansion, of resulting tumors Dr. Tamara Grigoryan* Daniel Bauer in an overall delay of tumor formation. Our Dr. Julian Heuberger Eveline Van der Laan results identify a novel intracellular signal- Dr. Gustaaf Heynen* ing module that synergistically regulates Dr. Jane Holland* Technical Assistants CAFs and CSCs. Targeting CAFs with Hedge- Dr. Sarah Kammerer* Simone Jung hog inhibitors may offer a novel therapeutic Dr. Alexandra Klaus-Bergmann* Frauke Kosel strategy against breast cancer (published in Dr. Linxiang Lan Marion Müller Cancer Research 2017 Feb 15). Dr. Hubert Pakula* Regina Vogel Dr. Jingjing Qi* New projects of the laboratory: in coopera- Dr. Giovanni Valenti* Secretariat Dr. Qionghua Zhu* Pharmacology in Berlin-Buch, we develop Annika Seifert* furthertion with small the Leibnizmolecule Institute inhibitors of Molecular of Shp2 Irmgard Wiznerowicz* *part of the time reported -

(Maria Pascual and Marc Nazare). In coop MDC Research Report 2016 79 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Thomas Blankenstein

Molecular Immunology and Gene Therapy

We focus on three areas in cancer immu- ulatory signal, which could not be substitut- nology: 1. Developing novel cancer models rejection was associated with sustained in- that better reflect human disease. 2. Em- ed by higher CAR affinity. Successful tumor ploying the mechanisms of rejection of es- The absence of tumor recurrence raised the tablished tumors by adoptively transferred questionflux and proliferation of whether stromaof transferred targeting T cells. was T cells with specific emphasis of the tumor involved in tumor rejection. Tumor rejection by CAR-Ts was independent of NK cell stroma as target. 3. Employing T cell thera- contribution and required the expression of py of cancer in the clinic. Therefore, we use - transgenic mice that allow the isolation of gether, our results indicate that CAR-Ts can eradicateIFNγ receptor large on established tumor stroma. solid Taken tumors to human TCRs from mice to redirect T cells by antigen-independent direct stroma de- towards superior antitumor activity. struction, whereby a cancer-driving antigen, such as HER-2, is targeted.

IFNγ targeting of tumor stroma Mis-initiation of intrathymic MART- is required for rejection of large 1 transcription and biased TCR established tumors by chimeric usage explain the high frequency antigen receptor-redirected T cells of MART-1 specific T cells T cells can be redirected with new anti- The immune response against melanoma-associated tumor antigens has been therapy (ATT) by introducing either a T cell intensely studied in part owing to the im- receptorgen specificity (TCR), and or chimericused for adoptiveantigen recep T cell- munogenicity of melanomas, the molecu- tor (CARs). The CAR consists of an antigen (Ag)-binding single chain variable fragment associated tumor antigens, and the ease to (scFv) antibody domain and a signaling lar identification of numerous melanoma- such antigen is the Melanoma Antigen Rec- Although adoptive T cell therapy using chi- access tumor infiltrating lymphocytes. One domain, most often the CD3ζ endodomain 26-35ognized in by the T-cells context 1 (MART1). of HLA-A*0201 Intriguingly, are lymphomas,meric antigen complete receptor eradication (CAR) modified of large T aboutCD8 T 100cells times specific more for abundantthe MART-1 compared epitope establishedcells (CAR-Ts) solid showed tumors efficacy has not in yet treating been - reported. Destruction of stroma cells crossed antigens. Moreover, these antigen-spe- presenting the tumor antigen of interest is to T cells specific for other tumor-associat essential for rejection of large established of the Va-region gene TRAV12-2. Here, we - providecific CD8 independent T cells show asupport highly biasedfor this usage no- ed fashion. However, CAR-Ts recognize the tion, by showing that the combinatorial antigentumors byin Tan cells MHC-independent in a MHC class I-restrict manner

exclusively on the cancer, but not stroma, derived from HLA-A2/MART-126-35 pairingT cell clonesof different is unusually TCRα and permissive TCRβ chains in CAR-Ts eradicated large established HER-2+ specific- cells. Here, we report that HER-2 specific- conferring this particular specificity, pro tumors when provided with a CD28 costim vided the CDR1α TRAV12-2 region is used. 80 MDC Research Report 2016 CANCER RESEARCH

Whether this TCR bias by itself accounts for ingly, TCRs isolated from HLA-transgenic the unusual abundance of these T cells has remained conjectural. Here, we provide an required in vitro mutagenesis to achieve ef- alternative explanation: mis-iniated tran- mice had low functional affinity and usually scription of the MART-1 gene resulting in question whether the repertoire of antigen- truncated mRNA isoforms leads to lack of negativeficient cancer mice cellactually recognition, yields better raising TCRs. the promiscuous transcription of the MART- To overcome these limitations, we gener-

126-35 epitope in human medullary thymic epithelial cells and consequently evasion of central self-tolerance towards this epitope. diverseated human T cell TCR-αβ repertoire gene to loci enable and chimericisolation The co-incidence of these two features – bi- ofHLA-A2 human transgenic TCRs against (ABabDII) human mice TAAs with from a ased TCR usage and leaky central tolerance antigen-negative hosts. – might act in an independent and additive manner resulting in the high frequency of Antigen choice is also critical in TCR gene

MART-126-35 therapy. Cancer testis (CT) antigens, like MAGE-A1, are expressed in a variety of Optimal-affinity specific CD8human T cells. T cell re- tumors, but, with the exception of testis, ceptors against the cancer-testis were not detected in healthy adult tissues. antigens MAGE-A1 and NY-ESO Tolerance to CT antigens remains a matter from antigen-negative hosts of debate. On one hand, T cells recognizing CT antigens have been found among

forTCR which engineering TCRs are can available. redirect TCRs patients’ should T blood of tumor patients and healthy indi- cells to specifically target tumor antigens,- viduals.tumor-infiltrating On the other lymphocytes hand, low and levels in the of mRNA encoding various CT antigens, in- rejectionbe of sufficient but are functional rarely found affinity against (for brev tu- cluding MAGE-A1, were detected in human mor-associatedity “affinity” in (self-) the following) antigens (TAAs), for tumor be- medullary thymic epithelial cells that play cause tolerance mechanisms select against a role in deletional tolerance. Here, we describe the generation and characterization

high-affinity TCRs. Technologies to obtain negative mice with a diverse human TCR maturedhigh-affinity TCRs TCRs bears oftenthe risk have of losing pitfalls. speci In- repertoireof MAGE-A1-specific restricted to TCRs HLA-A2. from By antigen- com- vitro mutagenesis and selection of affinity-- parison, human-derived TCRs are likely

ficity or, if TCR-affinity is too high, T cell ef HLA-mismatchedficacy in vivo, raising donor the questionT cells can what yield the skewed towards low-affinity, compatible optimal TCR-affinity is. In vitro priming of NY-ESO.with tolerance. The humanized We also identified mice provide optimal- a must be clearly distinguished from general affinity TCRs specific for the CT antigen allo-reactivehigh-affinity TCRsTCRs. against HLA-transgenic TAAs, but thesemice might allow isolation of suitable TCRs due inpowerful humans. tool to generate optimal-affinity to missing tolerance for many human TAAs, TCRs, which are difficult, if at all, accessible yet immunogenicity of mouse TCR variable regions can impede T cell therapy. Surpris-

MDC Research Report 2016 81 CANCER RESEARCH

Targeting cancer-specific in vitro and were analyzed as therapeutic mutations by T cell receptor gene - therapy neic HLA-A2-transgenic model of large es- The ease of sequencing the cancer genome, tablishedtargets for tumors, TCR gene we found therapy. that In both a synge mu- identifying all somatic mutations and graft- tations differed dramatically as targets for

genes into T cells for adoptive transfer al- ing mutation-specific T cell receptor (TCR) responseTCR-modified to R24L, T cells R24C in vivo.failed While to induce T cells an effectiveexpanded anti-tumor efficiently andresponse. produced Such IFN-γ differ in- TCRlow, forgene the therapy first time, might a truly achieve tumor-specific optimal ences in neoantigen quality might explain and effective therapy. Mutation-specific why cancer immunotherapy induces tumor regression in some individuals, while oth- evidenceefficacy with from least experimental possible toxicity. cancer Recent mod- ers do not respond, despite similar muta- elsclinical that dataantigens confirm encoded the by long-standing the tumor- in vivo model by showing that the MART- the best targets for adoptive T cell therapy. tional load. We confirmed the validity of the Openspecific questions somatic are, mutations how many are somatic potentially mu- tumors expressing analog but not native tations create suitable epitopes, whether MART-11-specific epitopes. TCR DMF5 The inducesdescribed rejection model al of-

somatic mutations qualify as suitable epit- in human cancer that serve as suitable T only individual-specific or also recurrent celllows targets identification and may of help those to predict (neo)antigens clinical - opes and how neoantigen-specific TCRs are willmost be efficiently important obtained. to identify Tumor immunogenic heteroge efficacy.Preventing tumor escape by driverneity needs mutations to be thatconsidered; occurred therefore, early, are it targeting a post-proteasomal essential for cancer cell survival and pres- trimming independent epitope ent in all cancer cells. First TCRs against Adoptive T cell therapy (ATT) can achieve recurrent somatic mutations, e.g. RAC1 fre- regression of large tumors in mice and hu- quently mutated in melanoma, have been mans, however tumors frequently recur. isolated and are currently been analyzed. - - Targeting human melanoma ityHigh are target thought peptide-MHC-I critical to prevent (pMHC) relapse. affin neoantigens by T cell receptor Hereity and we T show cell receptorthat targeting (TCR)-pMHC two epitopes affin gene therapy of the same antigen in the same cancer cells T cell responses in successful cancer im- - munotherapy seem directed towards neoantigens created by somatic mutations, but inby eradication monospecific of Tlarge cells, established which have tumors simi whenlar pMHC targeting and pMHC-TCRthe sub-dominant affinity, but results not cells caused regression of established can- cerdirect is lacking. evidence We that generated neoantigen-specific T cells express T- proteasomalthe dominant epitope.trimming, Only which the “escape” was regu but- receptor (TCR) to target different immuno- not the “rejection epitope” required post- genicing a mutations mutation-specific in cyclin-dependent transgenic kinase T cell cancer variants to evade. The data describe 4 (CDK4) that naturally occur in human alated novel by IFNγ,immune allowing escape IFNγ-unresponsive mechanism and melanoma. Two CDK4 mutations (R24C, R24L) similarly stimulated T cell responses ATT. better define suitable target epitopes for

82 MDC Research Report 2016 CANCER RESEARCH Figure 1

T cell

TCR α β Mutant MHC pep,de

Mutant an,gen

Cancer cell Selected Publications Textor, A., Listopad, J., Wührmann, L., Perez, C., Kruschinski, A., Chmielewski, M., Ab-

Cancer Res. ken, H., Blankenstein, T. and Charo, J. (2014). Efficacy of CAR T cell therapy in large tumors relies upon stromal targeting by IFNγ. 74: 6796-6805. Obenaus, M., Leitao, C., Leisegang, M., Chen, X., Gavvovidis, I., van derNat. Bruggen, Biotechnol. P., Soma,c muta,on Uckert,33: 402-407. W., Schendel D. and Blankenstein T. (2015). Optimal-affinity T cell receptors Blankenstein,against cancer T., antigens Leisegang, can M,be Uckert,identified W. inand antigen-negative Schreiber, H. (2015). hosts. Targeting Curr. Opin. Immunol. 33: 112-119. cancer-specific mutations by T cell receptor gene therapy. Figure 1: Mutation-specific T cells. Somatic Leisegang, M., Kammertoens, T., Uckert, W*. and Blankenstein, T* (2016). Target- ing human melanoma neoantigens by T cell receptor gene therapy. J. Clin. Invest. mutations in cancer may create a neoepitope that can be recognized by T cells. Textor, A., Schmidt, K., Kloetzel, P., Weißbrich, B., Perez, C., Charo, J., Anders, K., Sidney,126:854-858. J., Sette, *equal A., Schumacher, contribution T., Keller, C., Busch, D., Seifert, U. and Blankenstein, T. (2016). Preventing tumor escape by targeting a post-proteasomal trimming-independent epitope. J. Exp. Med.

213: 2333-2348. Group Leader Prof. Dr. Thomas Blankenstein Christian Schön* Katerina Thiede Scientists Lara Lee Wührmann Dr. Kathleen Anders Technical Assistants Dr. Thomas Kammertöns Dr. ElisaIoannis Kieback Gavvovidis Christian Friese Dr. Ana Milojkovic AngelikaIsabell Becker Gärtner Dr. Georgios Papafotiou Lisa Kimberly Henning* Dr. Karin Schmidt* Evelyn Hesse Dr. Vivian Scheuplein* Sabrina Horn Dr. Ana Textor Genevieve Knurr* Prof. Dr. Gerald Willimsky Melanie Manske* Kristin Retzlaff PhD students Marion Rösch Christel Westen Arunraj Dhamodaran Stefanie Petzold* DanaXiaojing Hoser Chen Katrin Hönig Secretariat Meng-Tung Hsu Karin Karamatskos Matthias Obenaus* Sylvia Klahn Christian Pircher Lucia Poncette Michael Rothe * part of the time reported

MDC Research Report 2016 83 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Oliver Daumke

Structure, function and mechanism of membrane-associated GTPases

GTPases of the dynamin and septin sembles into ring-like oligomers and how superfamilies are molecular machines these rings mediate the antiviral activity (Fig. 1a). that assemble at the surface of cellular membranes. Some of these proteins act as We also determined the structure of the dynamic scaffolds that remodel the underlying membrane in a GTPase-dependent bundle signalling element (BSE), a three- helixfull-length bundle, MxA which molecule. bridges We GTPase identified (G) dothe- fashion; others orchestrate the recruitment main and stalk. Based on functional experi- of interaction partner in a temporally and ments, we suggested a molecular pathway spatially defined manner. We areinterested how nucleotide-driven rearrangements in to understand principles of assembly, the MxA oligomer are transmitted from the GTPase domain to the stalk. More recently, function and regulation of these GTPases. we determined crystal structures of puta- To this end, structural studies are com- tive targets of MxA, the nucleoproteins of bined with biochemical and cell-based Toscana and Hantavirus, and studied their RNA binding mode and assembly. approaches. Structural insights into dynamin- mediated endocytosis Dynamin is the founding member of the dy- Structure and antiviral function of namin superfamily. The multi-domain pro- MxA GTPases tein oligomerizes around the neck of clath- The interferon-induced Myxovirus-resis- rin-coated vesicles and induces membrane tance (Mx) GTPases are central players of scission in a GTPase-dependent fashion. My innate immunity. They mediate resistance group is interested to understand the mo- against a wide range of pathogens, includ- lecular mechanisms how dynamin mediates - membrane scission. To this end, we deter- ative-strand viruses. As typical members ofing the influenza dynamin viruses superfamily, and many they other can oligo neg- dynamin and, more recently, the structure merize in ring-like structures around tubu- ofmined the dynaminX-ray structures tetramer. of Thenearly protein full-length com- lar membranes. prises a GTPase domain, a BSE, a stalk and a lipid-binding pleckstrin homology (PH) domain. The interaction site between the the virus binding domain, the stalk, of hu- stalk and the PH domain is often mutated in Using X-ray crystallography, we found that patients suffering from centronuclear my- the crystals, the stalks assembled via three opathy, a congenital disease leading to pro- interfacesman MxA hasin a azigzag four-helix fashion. bundle We showed fold. In gressive muscle weakness. Supported by that all three interfaces are required for biochemical and cell-based functional ex- the assembly of MxA at membrane surfaces periments, we proposed a molecular model and for its antiviral activity. Based on these for helical dynamin oligomers (Fig. 1b). results, we suggested a model how MxA as- Furthermore, we showed how the domain

84 MDC Research Report 2016 CANCER RESEARCH

Oligomerization of dynamin and septin-like GTPases. a) Models of ring-like oligomers of (a) the MxA GTPase or (b) the dynamin GTPase. c) Oligomeric ring model of the EHD2 ATPase. Oligomerization is solely mediated via the GTPase domains, but not the helical domains. d) Linear GTP-dependent oligomer of GIMAP2 which assembles via two interfaces in the GTPase domain into a linear scaffold.

MDC Research Report 2016 85 CANCER RESEARCH

interplay contributes to the assembly of an amino-terminal GTPase domain, followed auto-inhibited dynamin tetramer which is by a helical domain and a carboxy-terminal regulatory Eps15 homology domain. EHDs curvature. Finally, we introduced a dynamic are found at the plasma membrane and at modelrecruited how to GTPmembrane binding surfaces and hydrolysis of defined in- vesicular and tubular endocytic membrane duces scission of the vesicle neck. structures. They are implicated in several - Dynamin-related GTPases involved tably during the recycling of receptor pro- in mitochondrial remodelling teinsmembrane to the plasma trafficking membrane. pathways, most no A new focus of our group has been the structural and functional characterization We previously showed that EHD2 oligomer- of mitochondrial membrane remodelling izes in ring-like structures around tubulat- events. Mitochondria are dynamic membra- ed liposomes. By solving the crystal struc- nous structures which continuously under- ture of an EHD2 dimer, we found that stable go scission and fusion. These processes are dimerization of EHD2 is mediated via the crucial for the maintenance of respiratory and metabolic activity and for the preserva- binding site at the tip of the helical domains tion of mitochondrial DNA. Malfunctioning andGTPase showed domains. that Wealso identified the N-terminus the lipid- of of this process has been linked to numerous EHD2 can insert into the membrane, there- pathologies, such as neurodegenerative dis- by regulating membrane recruitment and/ eases, cardiomyopathy and cancer. Three or oligomerization of EHD2. Based on func- dynamin-like proteins, dynamin-like-pro- tional experiments, we suggested a model tein 1 (DNM1L), OPA1 (optical atrophy 1) for the EHD2 oligomeric rings (Fig. 1c). and mitofusin are implicated in mitochon- These rings have a remarkably different ar- drial dynamics. Unlike in endocytosis, mi- chitecture from those formed by MxA and tchondrial membrane fusion and division involves the double-membrane system of as a novel integral component of caveolae mitochondria. Consequently, mitochondrial openingdynamin. up Eventually, a wide range we of identified cell-based EHD2 ap- dynamin-related GTPase act on distinct proaches to study the physiological func- mitochondrial membranes, e.g. mitofusin tion of this protein. mediates fusion of the outer mitochondrial membrane whereas OPA1 catalyzes fusion GIMAPs – cellular scaffolds of the inner mitochondrial membrane. controlling lymphocyte survival - We determined the crystal structure of the mitochondrial membrane scission protein GTPase family of the in Immunity vertebrates. Associated The seven Pro hu- DNM1L. Based on a range of biochemical teins (GIMAPs) comprise a septin-related and cell-based experiments, we deduced expressed in cells of the immune system. an alternative assembly mode of DNM1L at man GIMAP members are predominantly the outer mitochondrial membrane for the apoptosis by controlling the activity of Bcl2 constriction of mitochondria. These results proteins.Some GIMAPs were proposed to regulate may have implications for mitochondria- based diseases in which DNM1L function is We determined four crystal structures of compromised. - - Oligomerization and cellular func- binationa representative with biochemical member, GIMAP2, experiments, in dif tion of Eps15 homology domain- thisferent work nucleotide-loading elucidated the molecular states. In basis com containing ATPases (EHDs) of GTP-dependent oligomerization via the EHDs are ubiquitously expressed dyna- GTPase domains (Fig. 1d). We also showed min-related ATPases which are built of an

that GIMAP2 in Jurkat T cells localizes to 86 MDC Research Report 2016 CANCER RESEARCH

the surface of lipid droplets which are cellular storage and signalling compartments.

Interestingly, GIMAP2 bound GTP with high Usingaffinity co-localization but did not hydrolyze experiments, it. we re-

- cently identified GIMAP7 at the surface of lipid droplets. In contrast to GIMAP2, GI MAP7 efficiently hydrolyzed GTP. It also- merstimulated elucidated the theGTPase mechanism activity ofof dimeriza GIMAP2.- tion-inducedThe structure GTPase of a GTP-bound activation. GIMAP7 These data di -

suggested that GIMAP7 acts as GTPase ac- ulatedtivating in protein a variety for GIMAP2.of different Furthermore, anaplastic largewe found cell lymphoma GIMAPs differentially lines compared down-reg to their T-cell progenitors, implicating a function of Selected Publications

certain GIMAPs as tumor suppressors. Olal, D.* and Daumke, O.*, 2016. Structure of the Hantavirus Nucleoprotein Provides

Reubold,T.F., Faelber,K.*, Plattner,N., Posor,Y., Ketel,K., Curth,U., Schlegel,J., Anand,R., Manstein,D.J.,Insights into the Noe,F., Mechanism Haucke,V., of RNADaumke,O.*, Encapsidation. and Eschenburg,S.* Cell Rep. 14, (2015).2092-2099. Crystal struc-

Daumke,O.*, Roux,A.*, and Haucke,V.* (2014). BAR domain scaffolds in dynamin-medi- ture of the dynamin tetramer. Nature 525, 404-408. Frohlich,C., Grabiger,S., Schwefel,D., Faelber,K., Rosenbaum,E., Mears,J., Rocks,O., andated Daumke,O.* membrane fission.(2013). Cell Structural 156, 882-892. insights into oligomerization and mitochondrial

Faelber,K.*, Posor,Y., Gao,S., Held,M., Roske,Y., Schulze,D., Haucke,V., Noe,F., and Daumke,O.*remodelling (2011).of dynamin Crystal 1-like structure protein. of EMBO nucleotide-free J. 32, 1280-1292. dynamin. Nature 477, 556- 560. * corresponding author/s

Group Leader Alexej Dick Prof. Dr. Oliver Daumke Marie Fricke Stephan Grunwald Scientists Manuel Hessenberger Dr. Tobias Bock Dr. Katja Fälber Dr. Oleg Ganichkin Dagnija Tupiņa Dr. Claudia Matthäus TechnicalAudrey Xavier Assistants Dr. Jeffrey Noel Franziska Bartelt Dr. Daniel Olal Carola Bernert Jeanette Schlegel PhD students Sabine Werner Arthur Alves de Melo Arasu Balasubramaniyam Secretariat Sivanandam Birgit Cloos

MDC Research Report 2016 87 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Michela Di Virgilio

DNA Repair and Maintenance of Genome Stability

DNA double-strand breaks (DSBs) occur as a by-product of DNA replication and in pathway shortly after their formation as a consequenceDSBs become ofcommitted DNA end toprocessing. a specific Arepair ma- response to ionizing radiation but are also jor regulatory point of DSB repair pathway generated as obligate intermediates of class switch recombination (CSR) in mature DNA ends, which interferes with direct liga- B lymphocytes. CSR is a deletional recom- tionchoice by NHEJ, is the and initiation predisposes of 5’-3’ cells resection to homol of- ogy-dependent repair (Fig. 1). The key regu- bination reaction that changes the constant lator of DNA end processing is the DNA repair region of the antibody molecule, altering its factor 53BP1, whose association with DSBs in effector function. This process has the func- G1 promotes NHEJ by preventing DNA end re- tion of generating highly efficient antibodies section (Fig. 1). B lymphocytes lacking a functional 53BP1 protein are unable to protect against pathogens, and it occurs through DSBs from processing, and as a consequence, the regulated formation and repair of DSBs. they cannot engage into NHEJ-mediated re- Mammalian cells employ two major options pair of programmed CSR breaks. This results in impaired class switching and immunode- to repair DSBs. In one case, DNA ends are protected from extensive processing and - directly re-ligated. This pathway is known as cytes.ficiency. On Therefore,the contrary, DNA 53BP1 end ability protection to pro by- non-homologous end-joining (NHEJ), and it tect53BP1 DNA is beneficialends against in switchingresection Brepresents lympho a barrier to homologous repair during DNA is the pathway of choice in G1 phase of the replication. Under physiological conditions, cell cycle. Alternatively, the cell can profit this barrier is removed by the HR protein from the availability of a homologous DNA BRCA1, which displaces 53BP1 and allows molecule generated by DNA replication, and for resection and appropriate repair of DSBs by HR (Fig. 1). BRCA1 gene is frequently mu- repair the broken DNA by copying the infor- tated in hereditary breast and ovarian can- mation in the duplicate DNA template. This cers, and the inability to inactivate DNA end process is called homologous recombina- protection by 53BP1 in the absence of a functional BRCA1 prevents DNA end processing tion (HR), and requires extensive resection and physiological repair of DNA replication- of DNA ends to occur. NHEJ and HR cannot associated breaks. As a consequence, DSBs always be used interchangeably. Repair by are channeled into alternative NHEJ reactions end-joining is essential for CSR in mature B lymphocytes, whereas homologous recombination is crucial to faithfully repair DSBs generated during DNA replication, and therefore it ensures preservation of genome integrity in this context. Figure 1. 53BP1 role in the regulation of DNA end processing.

88 MDC Research Report 2016 CANCER RESEARCH

Figure 2. Identification of phospho- dependent 53BP1 interactors: H/(H+L) ratio distribution of proteins identified by wild-type (heavy, H) versus phosphomutant (Light, L) 53BP1 SILAC pull-down. Error bars represent the standard deviation of the H/(H+L) mean value for all the peptides identified for each individual protein. Adapted from Di Virgilio et al., Science 2013.

leading to aberrant chromosomal rearrange- Our lab employs a wide variety of techniques ments. Accumulation of unrepaired breaks from classic biochemistry, molecular and cel- and chromosomal aberrations is known as lular biology approaches to state-of-the-art genomic instability, which is a predisposing proteomics and genomics methodologies to factor to carcinogenesis. Under these con- dissect the pathways and mechanisms that ditions, DNA end protection by 53BP1 has determine the balance between DNA end pathological consequences. resection and protection, and that ensure appropriate DSB repair outcomes. The molecular mechanisms underlying 53BP1 function in the regulation of DNA end processing remain elusive. We have previously determined that in addition to its recruitment to the chromatin surrounding the break sites, DNA damage-induced phosphorylation Selected Publications of 53BP1 is required for its ability to both protect DNA ends during CSR in B cells, and Di Virgilio M to promote genomic instability in BRCA1- Sanderactivity S,differentially Chu VT, Yasuda control T, Franklin B cells inA, theGraf germinal R, Calado center DP, Li light S, Imami and dark K, Selbach zones. M, Im - munity , Bullinger L, Rajewsky K. PI3 Kinase and FOXO1 transcription factor Callen E., Di Virgilio M., Kruhlak M.J., Nieto-Soler M., Wong N., Chen H.T., Faryabi R.B., Polato 2015 F., DecSantos 15;43(6):1075-86. M., Starnes L.M., Epub Wesemann 2015 Nov D.R., 24. Lee J.E., Tubbs A., Sleckman phosphorylationdeficient cells (Bothmer of 53BP1 et wasal., Mol crucial Cell for2011). the B.P., Daniel J.A., Ge K., Alt F.W., Fernandez-Capetillo O., Nussenzweig M.C., and Nus- This finding suggested that DSB-induced senzweig A. 53BP1 mediates productive and mutagenic DNA repair through distinct phospho-protein interactions. Cell Di Virgilio M., Callen E., Yamane A., Zhang W., Jankovic M., Gitlin A.D., Feldhahn N., proteomicrecruitment approach of yet-to-be-identified for phospho-dependent DNA end Resch W., Oliveira T.Y., Chait B.T., Nussenzweig2013 Jun 6;153(6):1266-80. A., Casellas R., Robbiani Epub 2013 D.F., andMay Nus 30.- 53BP1protection interactors, factors. Bywe using have a subsequently SILAC-based - lin Class Switching. Science senzweig M.C. Rif1 Prevents Resection of DNA Breaks and Promotes Immunoglobu 53BP1 effectors in DNA end protection, in- Bothmer A., Robbiani D.F., Di Virgilio M. J., Chen H.T., Bosque D., Callen2013 E., Nussenzweig Feb 8;339(6120):711-5. A., and Nussenzweig Epub 2013 M.C. Jan Regulation 10. cludingidentified the several DNA damage potential response candidates proteins as of DNA end joining, resection, and immunoglobulin, Bunting S.F., class Klein switch I.A., Feldhahnrecombination N., Barlow by 53BP1. Mol Cell

top candidates (Fig. 2 and Di Virgilio et al., Sci- 2011 May 6;42(3):319-29. Rif1, PTIP, and its stable partner Pa1 as the indeed recruited to DSBs via interaction with Start of the Group: September 2014 phosphorylatedence 2013). We showed 53BP1, that and Rif1 mediate and PTIP 53BP1- are dependent end protection during physiologi- Group Leader Technical Assistant cal (CSR) and pathological (genomic instabil- Dr. Michela Di Virgilio Lisa Keller ity) DNA repair, respectively (Di Virgilio et al., Scientist Secretariat Dr. Ali Rahjouei Jasmin Bonkowski thatScience are 2013;required Callen for 53BP1et al., Cell to promote2013). These liga- Dr. Devakumar Sundaravinayagam tionfindings of DNA indicate ends during that the CSR molecular and the aberrant events chromosomal rearrangements in BRCA1- PhD students mutant cells are, to some extent, distinct, and Matteo Andreani that multiple 53BP1 effectors are required to Sandhya Balasubramanian mediate DNA end protection in these differ- Veronica Delgado Benito ent repair contexts.

MDC Research Report 2016 89 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Gaetano Gargiulo

Molecular Oncology

Our research focuses on understanding the Group summary molecular mechanisms regulating tumor ho- Cancer arises in normal cells by means of genetic and epigenetic alterations. We want meostasis and response to anti-cancer therapy. to understand how tumor tissues balance We do so by modeling human cancers in labo- self-growth and interactions with the host. ratory animals using genetic alterations de- Both processes can ensure cancer survival scribed in patients. We wish to exploit animal and govern critical cell decisions such as to whether self-renew, differentiate or die (i.e. models as surrogate or “targeted” patients to homeostasis). Ultimately, tumor homeosta- ultimately identify novel anti-cancer approaches sis can be viewed as the dark side of the and biomarkers for response. normal tissue self-regulation. We also want to know how do cancer cells Our group has been very recently established deal with the therapy-induced stress. here at MDC. Supported by a Helmholtz Young Learning about these mechanisms will Investigator Grant, we started recruiting scien- enable us to identify cancer-specific vul- tists and students between June and December nerabilities, which in turn may pave the way to identifying more effective treat- 2016. Before my appointment at MDC, I was ments. trained at the European Institute of Oncology in Milan (EIO, Italy) and at the Netherlands Cancer Approach Institute in Amsterdam (NKI, The Netherlands). One of the approaches we take is to model human cancers in laboratory animals using In my most recent work, I contributed to de- genetic alterations found in patients (see velop an in vivo screening procedure to functionally characterize cancer genes during are used to study: •example genotype-to-molecular in Figure 1). In turn, phenotype these models con- gliomagenesis. Alongside, I was also involved nections (Fig. 1b) in understanding the role of tumor cell identity • molecular mechanisms of tumor growth in lung cancer progression. Currently, we are and response to therapy (Fig. 1c) translating the concept that changing the tumor • target discovery and validation (Fig. 1d) cell identity in vivo using epigenetic drugs can introduce specific vulnerabilities into cancer - cells that can be therapeutically exploited. In tientsIn the longto ultimately run, we aimidentify to exploit novel animal anti- parallel, we have started a long-term endeavor cancermodels treatments as “surrogate” and biomarkers or “targeted” for pare- sponse. in which we are using a novel entry point into Mechanistically, we focus on genetic and combating Glioblastoma, the most aggressive epigenetic control of gene expression form of brain tumors. In the fall, the group mis- in cancer cells. We make use of a combination of experimental and computational sion was reinforced by an European Research approaches among which: adult stem cells Council Starting Grant, which will further boost genetic engineering, in vivo tumor model- our activities in that direction. ing, in vivo genetic screens and genome-

90 MDC Research Report 2016 CANCER RESEARCH

notably at the single cell and single molecule level. These technologies include – but Aswide we thrive binding, to achieve occupancy a conceivably & expression rapid are not limited to – RNA-seq, ATAC-seq as translationprofiling by of high our experimental throughput sequencing.efforts into well as Perturb-seq like approaches. We are clinical oncology, we leverage our results in a good position to engage in this endeav- against large publicly available repositories or as the MDC is well equipped with all of the required state-of-art technologies. information. containing patients’ molecular and clinical Focus on solid tumors Currently, we will focus our work on solid

particular, the lab established a long-term Recent Publications researchtumors such program as brain dealing and lung with cancers. the Glio In- Ferone, G., Song, J.-Y., Sutherland, K. D., Bhaskaran, R., Monkhorst, K., Lambooij, blastoma Multiforme (GBM). The GBM is the most common primary brain tumor, Squamous Cell Carcinoma from Different Cells of Origin. Cancer Cell, 30(4), 519–532. J.-P.,http://doi.org/10.1016/j.ccell.2016.09.001 et al. (2016). SOX2 Is the Determining Oncogenic Switch in Promoting Lung Koppens, M. A. J., Bounova, G., Gargiulo, G., Tanger, E., Janssen, H., Cornelissen- - Steijger, P., et al. (2016). Deletion of Polycomb Repressive Complex 2 From Mouse tientsand is (precision currently medicine) incurable. and It is be urgent able to Gastroenterology, 151 devise treatments best fitting individual pa doi.org/10.1053/j.gastro.2016.06.020 therapy. Both tumor heterogeneity and re- Intestine Causes Loss of Stem Cells. (4), 684–697.e12. http:// dosage makes the (kind of) poison. Molecular & Cellular Oncology, 00–00. http://doi. predict the patients’ response to the chosen- Gargiulo, G., Citterio, E., & Serresi, M. (2016). Polycomb and lung cancer: When the ly affect GBM clinical management. As men- Serresi, M., Gargiulo, G., Proost, N., Siteur, B., Cesaroni, M., Koppens, M., et al. (2016). tionedsistance above, to available we approach treatments these significant problems org/10.1080/23723556.2016.1152345 Epithelial-Mesenchymal Transition in Non-Small-Cell Lung Cancer. Cancer Cell, Polycomb29(1), 17–31. Repressive http://doi.org/10.1016/j.ccell.2015.12.006 Complex 2 Is a Barrier to KRAS-Driven Inflammation and

patientsby creating at andmolecular characterizing level and “humanized” exploiting vivo shRNA screens in solid tumors. Nature Protocols theseanimal modelsmodels ofin GBM state-of-the-art accurately reflecting genetic Gargiulo, G., Serresi, M., Cesaroni, M., Hulsman, D., & Van Lohuizen, M. (2014). In , 9(12), 2880–2902. http://doi. - org/10.1038/nprot.2014.185 tify molecular biomarkers for response to standard-of-carescreens in vivo. In for this patients case, we with aim thisto iden dis- Start of the Group: June 2016 ease as well as to uncover mechanisms of intrinsic and acquired resistance. Group Leader For lung cancer, we are interested in those Dr. Gaetano Gargiulo tumor subtypes driven by the Kras onco- Associate Staff Scientist gene, for which effective treatments are Lab manager Dr. Michela Serresi currently lacking. Melanie Liekweg Office manager Technologies PhD Students Daniela Keyner To characterize our animal models and Juan Carlos Company Nevado their response to treatment as well as to Quang Vinh Phan Scientist perform functional genetics, we aim at ex- Matthias Jürgen Schmitt Dr. Celine Sin

ploiting the so-called ‘omics’ technologies, MDC Research Report 2016 91 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Udo Heinemann

Macromolecular Structure and Interaction

The Macromolecular Structure and Interaction heterotetrameric p97:ASPL complexes. The Laboratory combines macromolecular crys- p97-ASPL interaction observed in crystals tallography with biochemical and biophysical and validated by biochemical and cellular approaches to elucidate the molecular basis assays serves as a paradigm for a new type of biological processes at atomic detail. One of AAA+ ATPase regulation by UBX-domain major area of research is concerned with the proteins. Finally, the Helmholtz Protein regulation of gene expression at the tran- Sample Production Facility produces pro- scriptional as well as translational level. Post- teins, protein crystals and crystal structures transcriptionally, mRNA homeostasis is en- for collaborating laboratories. Our crystal- sured by RNA-binding proteins that regulate lographic research is greatly facilitated by mRNA maturation, transport, translation and privileged access to the synchrotron stor- degradation. Among these, the human pro- age ring BESSY in Berlin. tein ROQUIN1/RC3H1 regulates the expression of genes in the immune system by bind- Molecular basis of ing to conserved sequence/structure motifs gene-expression control in the 3’ untranslated region of their mRNAs. - The crystal structure of the ROQUIN1/RC3H1 lated at the levels of transcription initiation ROQ domain reveals a winged helix-turn- andIn eukaryotic mRNA processing, cells, gene expressiontransport, translais regu- tion and degradation. We are studying the helix fold common in DNA-binding proteins molecular basis of gene-expression control and a dimeric ROQ arrangement presenting by crystallographic and biochemical analy- a basic dimer surface. Mutational analysis ses of classical transcription factors as well supports the hypothesis that this ROQUIN1/ as proteins and microRNAs involved in maintaining mRNA homeostasis. RC3H1 surface is involved in RNA binding. In a second major area of research, we address problems of intracellular protein transport and 1)ROQUIN1, was reported also knownto control as the RC3H1 lifespan (RING of mRNAsfinger and by CCCH-typebinding to zinc constitutive finger domains decay sorting. In all eukaryotes, the highly abun- - dant AAA+ ATPase p97/CDC48 is involved - in multiple cellular processes including the tebrateelements mRNAs (CDEs) found within to contain their 3’ CDEs untrans are transfer of polyubiquitinated protein chains lated region (3’UTR). Among the >50 ver from the endoplasmic reticulum to cytosolic those encoding tumor necrosis factor-α proteasomes for their degradation. We have (TNF-α) and ICOS, the inducible T-cell co- determined crystal structures of human p97 stimulator. ROQUIN1/RC3H1 therefore has important functions in inflammation, bound to the extended UBX domain of the toimmunity extended, and natively auto-immunity. unfolded Inpeptide addition re- adapter protein ASPL and reveal that ASPL to the RING and zinc-finger domains and remodels functional p97 hexamers to form ROQ domain known to mediate CDE bind- gions, ROQUIN1/RC3H1 contains a novel

92 MDC Research Report 2016 CANCER RESEARCH

A dimeric ROQ domain of ROQUIN1/RC3H1 is involved in target RNA binding. (a) Dimeric arrangement of the ROQ domain as observed in the crystal structure. (b) The ROQ dimer interface is highly positively charged (blue surface) and is predicted to be involved in RNA binding. (c, d) An engineered, disulfide-stabilized ROQ domain dimer shows enhanced ICOS mRNA binding under oxidizing conditions (-DTT), but not under reducing conditions (+DTT). From Schuetz et al. (2014)

ing. Crystal structure analysis reveals a of mRNA targets, and a novel mode of binding winged helix-turn-helix (wHTH) fold of the ROQ domain, which is decorated by four ad- - RC3H1in the 3’UTR can modulate of the A20 the mRNA activity was of revealed. the NF- monly found in proteins binding double- This work demonstrates that ROQUIN1/ ditional α-helices. wHTH domains are com RC3H1 ROQ domains are present in a di- κBMolecular pathway through basis itsof regulatorintracellular A20. mericstranded arrangement, DNA. In crystals, which is the stabilized ROQUIN1/ by transport polar interactions and creates a highly basic - surface suited for RNA binding. Mutational close functionally distinct compartments analysis demonstrated that basic amino- suchIn eukaryotic as the nucleus, cells, membrane the endoplasmic bilayers enre- acid residues present in this surface region ticulum (ER), the Golgi apparatus, and lysosomes. Many proteins and other molecules binding. Binding to the stem-loop structure travel to their destination within cellular ofare, the indeed, CDE could required be enhanced for ICOS by mRNA an engi CDE- compartments or outside the cell by crossing membrane bilayers. For proteins, these transport steps are associated to matura- toneered a model disulfide where bond, proteins that associate stabilizes in thedi- tion and quality control processes, which mericROQUIN1/RC3H1 fashion on CDEs dimer, of target lending mRNAs. support ensure that defective macromolecules are recognized and degraded along their path. proteins, we collaborate closely with the We study structural aspects of two differ- On ROQUIN1/RC3H1 and other RNA-binding ent transport processes in yeast and human cells, the tethering of lipid vesicles at their thelaboratory MDC groups of Markus of Stefan Landthaler Kempa, of Jana BIMSB/ Wolf target membranes and the retro-transloca- MDC. In a wider collaboration, also involving tion of folding incompetent protein chains binding sites were mapped in a large number from the ER lumen into the cytoplasm. and Claus Scheidereit, ROQUIN1/RC3H1

MDC Research Report 2016 93 CANCER RESEARCH

The conserved N-terminal domain of Vps18 (Vps18NTD) folds into a seven-bladed β-propeller. (A) Schematic of

Vps18NTD showing the seven propeller blades in rainbow colors. (B) In the crystal lattice, two Vps18NTD propel- lers face each other in an arrangement stabilized by polar interactions. (C, D) Close-ups of the hydrogen-

bonded contacts within the Vps18NTD interface. From Behrmann et al. (2014)

HOPS, the homotypic fusion and vacuole protein sorting complex of yeast, is a ca- shown to be critical for HOPS stability and nonical tethering complex. HOPS shares function.as a seven-bladed The structure β-propeller, and interactions which was of four subunits with CORVET (class C core vacuole/endosome tethering), the other provide a framework for the architecture hexameric tethering complex of the endoly- andthe Vps18assembly β-propeller of the HOPS in the and crystal CORVET lattice ves- sosomal pathway. Both HOPS and CORVET icle tethering complexes. AAA+ ATPases are abundant proteins in eukaryotic cells, which are recruited to their contain five subunits that are structurally cellular site of activity by adaptor proteins similar to subunits of the vesicle coat COP II and convert the energy from ATP hydrolysis low-resolutionwith predicted amino-terminalcryo-electron β-propellersmicroscopy into mechanical work. A proteomics study in structureand carboxy-terminal of HOPS has α-helicalbeen published, domains. but A the laboratory of Erich Wanker (MDC) iden- high-resolution structural data for any part of HOPS or CORVET were not available until most tightly binding adaptor of the human recently. The crystal structure of the ami- tified the UBX-domain protein ASPL as the no-terminal domain of the HOPS subunit structure analysis revealed an unexpected remodelingAAA+ ATPase of p97 functional (CDC48 p97 in yeast). hexamers Crystal by the laboratory of Christian Ungermann (Os- ASPL and the formation of p97:ASPL hetero- nabrück)Vps18, determined has now revealed in collaboration its structure with tetramers with reduced ATPase activity and

94 MDC Research Report 2016 CANCER RESEARCH

Crystal structure of a p97:ASPL heterotetramer. (a) The C-terminal domain of ASPL (ASPL-C) bound in 2:2 stoichiometry to the N-D1 fragment of human p97. (b) ASPL-C consists of an extended UBX domain (eUBX) containing strand β0 and several α-helices in addition to the canonical UBX domain. (c) ASPL-C wraps around the p97 N-domain in a geometry that is incompatible with the p97 hexamer structure. (d) An extended α-helical lariat is a key structural element for p97 remodeling by ASPL. Hotspots of the ASPL- p97 interaction include residues D351 (e), P437 and P438 (f) of ASPL-C. Modified after Arumu- ghan et al., 2016.

impaired function in endoplasmic reticu- Selected Publications lum-associated protein degradation (ERAD). We have therefore discovered a novel inacti- Behrmann, H., Lürick, A., Kuhlee, A., Kleine Balderhaar, H., Bröcker, C., Kümmel, D., vation mechanism for p97, which may serve as starting point for the development of new Engelbrecht-Vandré,HOPS complex stability S., Gohlke,and function. U., Raunser, J. Biol. S.,Chem. Heinemann, 289, 33503-33512. U.# & Ungermann, C.# anti-cancer compounds. (2014) Structural identification of the Vps18 β-propeller reveals a critical role in the Roquin binding to target mRNAs involves a winged helix-turn-helix motif. Nat. Com- mun.Schuetz, 5:5701. A., Murakawa, Y., Rosenbaum, E., Landthaler, M. & Heinemann, U. (2014) Protein Sample Production Murakawa, Y., Hinz, M., Mothes, J., Schuetz, A., Uhl, M., Wyler, E., Yasuda, T., Mastro- Facility buoni, G., Friedel, C.C., Dölken, L., Kempa, S., Schmidt-Supprian, M., Blüthgen, N., The Helmholtz Protein Sample Production Facility (Helmholtz PSPF) is a joint activ- Backofen, R., Heinemann,Nat. Commun. U., Wolf, 6 J., Scheidereit, C. & Landthaler, M. (2015) RC3H1 ity between the MDC and the Helmholtz- Arumughan,post-transcriptionally A.*, Roske, regulates Y.*, Barth, A20 C., mRNALleras Forero,and modulates L., Bravo-Rodriguez, the activity of K., the Redel, IKK/ A., Kostova,NF-κB pathway. S., McShane, E., Opitz, R.,:8367. Faelber, K., Rau, K., Mielke, T., Daumke, O., Selbach, Braunschweig. The PSPF offers expertise in proteinZentrum production für Infektionsforschung and biophysical (HZI) charac in- M.,that Sanchez-Garcia, disrupts functional E., Rocks, p97 hexamers. O., Panáková, Nat. D., Commun. Heinemann, 7:13047. U.# & Wanker, E.E.# terization for structural biology to external Kang,(2016) Y.*, Quantitative Gohlke, U.*, interaction Engström, mappingO.*, Hamark, reveals C., Scheidt, an extended T., Kunstmann, UBX domain S., Heinein ASPL- and internal partners, using various host - mann,ride interactions. U., Widmalm, J. Am. G., Santer,Chem. Soc. M.# 138 & Barbirz, S.# (2016) Bacteriophage tailspikes requirements. *Equaland bacterial contribution, O-antigens #corresponding as model authors system to study weak-affinity protein-polysaccha Thesystems Helmholtz tailored PSPF to specificoffers a experimentalplatform for , 9109-9118. initiating collaborative projects on topics that are initially outside the area of exper- Group Leader Graduate and Prof. Dr. Udo Heinemann Undergraduate Students past, these co-operations have frequently Zeynep Kurt yieldedtise of thecrystal Heinemann structures laboratory. and molecular In the insight on proteins linked to human disease Scientists Pavan Kumar Reddy Vennapusa or target proteins for pharmacological in- Dr. David Carter Mehmet Hanifi Kurt tervention with small molecules. Recently, Dr. Ulrich Gohlke Technical Assistants a series of structure analyses has helped to Dr. Piotr Henryk Malecki Tracy Alice Dornblut identify inhibitors of the protein tyrosine Dr. Anja Schütz Andreas Knespel phosphatase SHP2 in collaboration with the Silke Kurths laboratories of Jörg Rademann and Gerhard PhD students Dr. Yvette Roske Wolber (both FU Berlin), Jens von Kries Janett Tischer (MDC/FMP Screening Facility) and Walter Ankur Garg Birchmeier (MDC). These inhibitors block QianqianSofia Banchenko Ming Secretariat cell motility and growth of cancer cells in Sasa Petrovic Birgit Cloos vitro and in vivo.

MDC Research Report 2016 95 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Uta E. Höpken

Microenvironmental Regulation in Autoimmunity and Cancer

Our group dissects cellular requirements and gression of B cell lymphomas was delayed. molecular pathways which contribute to the Lymphoma-exposed DCs upregulated immunomodulatory cytokines, growth factors transformation of secondary lymphoid organs and the CCAAT/enhancer binding protein (SLOs) toward lymphoma/leukemia-permissive - niches in preclinical mouse models. We aim to responsive to lymphoma imposed cytokine identify stromal elements in SLOs that collabo- changesβ (C/EBPβ). and C/EBPβ-deletedunable to promote DCs lymphoma were un cell survival. We conclude that DCs have rate with lymphoma/leukemia B cells and study lymphoma-promoting functions and act lymphoma-induced cellular and molecular re- directly by the provision of growth factors, modeling. Secondly, we study chronic autoim- provided that they express the transcription mune or pathogen-induced immune reactions that lead to lymphoid neogenesis and develop- Cellularfactor C/EBPβ. stromal elements in B cell fol- ment of associated lymphomas. Based on the licles that collaborate with leukemia B identification of pivotal stromal cell types and cells in an indolent B-CLL mouse model V. Stache, F. Brand, T. Börding (collaboration signaling pathways, we will probe pharmaco- with A. Rehm, MDC) logical interventions targeted at the crosstalk We characterized non-hematopoetic stro- between lymphoma cells and the microenvi- mal cells in the B cell zone of SLOs that col- ronment. laborate with leukemia B cells in an indolent Eµ-Tcl1 leukemia mouse model. This relationship involved chemokine-mediated Immunosurveillance and inter guidance of leukemic cells into promoting actions between tumor cells and niches, followed by a nurse-like function of its microenvironment the stromal follicular dendritic cell (FDC)- Cellular interactions and molecular com- network (Figure 1). Crosstalk inhibition of munication pathways in the lymphoma - survival niche F. Scholz, A. Graband (collaboration with A. lymphotoxin(LT)/LTRβ signaling or inhibi- Rehm, A. Leutz, MDC) tion of the CXCL13/CXCR5 signaling axis- Survival and progression of lymphoma cells turesretarded indicative lymphoma of a progression.skewed polarization In addi is dependent on nodal access and micro- oftion, monocytes we identified and geneneutrophils. expression Depletion signa anatomical localization We showed that the of macrophages or neutrophils resulted in tumor retardation. Leukemia cells se- Myc B cell lymphoma nodal dissemination. chemokine receptor CCR7 is crucial for Eμ-- neutrophils and macrophages into a tumor tic reticular cells and dendritic cells (DCs) supportivecreted IL-10 phenotype. and by that Also, rendered neutrophils splenic re- withinMalignant the cellsT cell lodged zone. According adjacent to to fibroblas their dif- vealed increased expression of B-cell acti- ferentiation status, DCs trigger immunity or vating factors which supports leukemia cell tolerance in T lymphocytes. DC-mediated B survival. Hence, immune cells can also be cell lymphomagenesis remained enigmatic. primed towards a lymphoma-supporting - milieu.

In DC-depleted mice we observed that pro 96 MDC Research Report 2016 CANCER RESEARCH

Figure 1. Leukemia cell homing and cellular interactions in the lymphoma niche. Red–labeled Eμ-Tcl1 leukemia cells intermingle with the FDC-network. Spleen sections were stained for CD21+CD35+ FDCs (green) and B220+ B cells (blue), scale bar, 100 μm. The interference with CXCL13/ CXCR5-governed leukemia cell homing and lymphotoxin-mediated tumor-stroma crosstalk provides new targets in lymphoma therapy.

CAR-T cell immunotherapy for their development. Th cell transcription multiple myeloma and B cell - lymphoma (collaboration with A. Rehm, J. Bluhm, O. factors other than RORγt may induce differ Daumke, MDC, and W. Uckert, Humboldt- entiation of Il-17-producing Th cells under University) tissue-specific conditions. Chimeric Antigen Receptors (CARs) are sin- Selected Publications: gle-chain antibodies, anchored on the T cell surface and containing intracellular signal Lenz, P, Lenz G, Höpken UE*, Rehm, A* (2016). Splenic marginal zone granulocytes domains. The whole construct confers T cells Gätjenacquire M, an Brand, accentuated F, Grau, neutrophil M, Gerlach, B K,cell-helper Kettritz, R,phenotype Westermann in chronic J, Anagnostopoulos lymphocytic I, leukemia. Cancer Research Wichner K, Stauss D, Kampfrath B, Krüger K, Müller G, Rehm A, Lipp M, Höpken UE generated CAR-T cells directed against the B 76: 5253-65; *equal contribution cellwith maturation antigen-specific antigen cytolytic (BCMA) capacity.for the treat We- FASEBJ ment of B cell disorders, in particular mul- (2016).30:761-774 Dysregulated development of IL-17- and IL-21-expressing follicular T helper tiple myeloma (MM), mature Non-Hodgkin cells and increased germinal center formation in the absence of RORγt. Cseresnyes Z, Hauser A, Lenz P, Hehlgans T, Brink R, Westermann J, Dörken B, Lipp Lymphoma (B-NHL), and autoimmune dis- HeinigM, Lenz K, G, Gätjen Rehm M,A, Grau,Höpken M, UEStache, (2014). V, Anagnostopoulos Access to follicular I, Gerlach, dendritic K, cells Niesner is a pivotalR, eases. Our anti-BCMA CAR-T cell constructs step in murine chronic lymphocytic leukemia B cell activation and proliferation. Cancer Discovery 4: 1449-1465 cytolytic capacity and their ef- Rehm A, Gätjen M, Gerlach K, Scholz F, Mensen A, Gloger M, Heinig K, Lamprecht in vitro in vivo B, Mathas S, Begay V, Leutz A, Lipp M, Dörken B, Höpken UE (2014). Dendritic cell- confers human T cells with antigen-specific- transplantation models of MM and B-NHLs. Nature Communication 5: 5057-5070 ficacy has been validated in pre-clinical xeno mediated survival signals in Eμ-Myc B cell lymphoma depend on the transcription factor C/EBPβ. The tissue-specific micromilieu Höpken UE (2014). Lymphotoxin-α keeps the gate open for T cell infiltration in cHL. regulates plasticity of T helper cell Patents:Blood 124: 2897-2898 subpopulations K. Wichner against CD269 (BCMA). Autoimmunity is associated with a strong PCT-Anmeldung beim Europäischen Patentamt: PCT/EP2013/072857 “Antibody genetic component, but onset and persis- tence of autoimmune diseases requires an additional environmental trigger. The bal- Group Leader Tessa Lara Skroblyn ance between immunity and tolerance is Priv. Doz. Dr. Uta E. Höpken Vanessa Stache regulated by molecular factors including the (Permanent Fellow) Undergraduate Students involved in functional lymphoid organogen- Scientists Teresa Börding* nuclear hormone receptor RORγt which is Dr. Kristina Heinig* Annika Graband* Lutz Menzel Edward Michaelis* germinalesis. RORγt-/- center mice formation, exhibited and accumulationincreased au- Dr. Katharina Wichner* Lydia Verlaat* toantibodyof follicular production. T helper (Tfh) Deregulated cells, spontaneous B cell responses were linked to enhanced production PhD Students Technical Assistants Franziska Brand Branka Kampfrath* * part of the as a key transcription factor for the genera- Florian Scholz Kerstin Krüger period of IL-17. Because RORγt has been described Heike Schwede reported

tion of inflammatory Th17 cells, we revealed an unknown transcriptional flexibility in MDC Research Report 2016 97 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Ralf Kühn

IPS cell based disease modelling

Understanding how genetic variation and muta- are bound by the generic nuclease Cas9 and tions lead to the onset of human diseases is an guide the complex to the complementary DNA sequence, followed by the induction of important question in biomedical research and the basis for developing therapeutic interven- in mammalian cells mainly by the NHEJ re- tions. We are using the CRISPR/Cas nuclease paira double-strand pathway which break reconnects (DSB). DSBs open are fixedDNA technology in human induced pluripotent stem ends imprecisely and frequently leads to the loss of multiple base pairs at the target cells (hiPSCs) as central technology to inves- site. Such small, randomly sized deletions tigate human genetic disease mechanisms. are often used for the creation of frame hiPSCs are cell lines can be indefinitely propa- shift mutations within coding regions leading to gene inactivation (knockout). The gated but also differentiated in vitro into many cell types of the human body. The research nucleotide replacements by gene targeting group on hiPSC -based disease modeling is requiresgeneration the of alternative precise modifications pathway of suchhomol as- setting up a technology platform enabling the ogy directed repair (HDR), able to read new sequences from externally delivered, DNA engineering, differentiation and phenotyping of hiPSCs. The lab is presently focusing on the op- timisation of CRISPR/Cas induced mutagenesis in hiPSCs and their directed differentiation into dopaminergic neurons. The first topic requires the efficient delivery and action of Cas9 nuclease, gene specific sgRNAs and customized DNA repair templates molecules in hiPSCs. In the second working area we are defining the optimal combination of external and internal cues specifying the differentiation of hiPSCs into dopaminergic neurons to study genetic lesions leading to the loss of these cells in Parkinson`s disease.

Gene editing using the CRISPR/ Cas nuclease system Programmable nucleases are used for edit- Figure 1: Gene editing in mammalians cells using the ing the sequence of target genes by breakup CRISPR/Cas nuclease system. of the DNA at preselected positions. The lat- Cas9 nuclease together with sgRNA create a double-strand est generation of nucleases is provided by break (DSB) in the target gene. DSBs are either fixed by non-homologous end joining (NHEJ) repair which recon- that uses short, single guide (sg) RNAs for nects DNA ends imprecisely and deletes multiple base pairs DNAthe CRISPR/Cas9 sequence recognition bacterial anddefense can besystem pro- (knockout) or by homology directed repair (HR), reading grammed towards new targets by adaption new sequences from a targeting vector into the DSB site (knockin).

of the sgRNA first 20 nucleotides. sgRNAs 98 MDC Research Report 2016 CANCER RESEARCH

to introduce Parkinson`s disease risk alleles into `healthy´ genetic backgrounds or the correction of such alleles in patient-derived hiPS cells. For phenotypic analysis mutant and control cells are converted using lineage- - ated cells, such as dopaminergic neurons. The differentiatedspecific transcription cultures factorscan be intostudied differenti by live cell imaging for disease associated phenotypes, ultimately building a relational map of genotypes, envirotypes and phenotypes (Fig. 2). Since fall 2014 we focus on the set up of this hiPS cell technology platform for the si-

(multiplex engineering) by optimizing the deliverymultaneous of Cas9, modification sgRNAs andof up HDR to three templates genes into hiPSCs and for the differentiation of hiP- SCs into dopaminergic neurons. Presently we

the mutagenesis of two genes and can dif- Figure 2: Technology platform for gene editing and ferentiateachieve in 1/3 up toof the 30% cells of transfectedinto dopaminergic hiPSCs differentiation of human IPS cells. neurons. To support the analysis of cultures Using CRISPR/Cas gene editing, disease related mutations - can be introduced into the genome of human induced pluripotent stem (iPS) cells. Upon the directed differentiation by live cell imaging we use fluorescent report of iPS cells into e.g. neurons, cultures can be studied by live ers by knockin into marker genes. In addition, cell imaging for disease associated phenotypes, ultimately Caswe aimbased for conditional,extending the i.e. utility inducible of the and hIPSC cell building a relational map of genotypes, envirotypes and technology platform by establishing CRISPR/ phenotypes. cells and the three dimensional differentia- tiontype-specific of hiPSCs gene into knockoutorganoids in which differentiated recapitu- templates into the DSB site (knockin) (Fig. late aspects of fetal brain development. 1). Only a small fraction of DSBs in mammalian cells is repaired by HDR, making pre-

constructs indicating the ratio of NHEJ or HDRcise gene for the editing repair inefficient. of DSBs Usingwe found reporter that Selected Publications the suppression of the NHEJ enzyme DNA N., and Kühn, R. (2014). Simple Derivation of Transgene-Free iPS Cells by a Dual - Pertek,Recombinase A., Meier, Approach. F., Irmler, Mol. M., Biotechnol. Beckers, J., 56,Skylaki, 697–713. S., Endele, M., Wurst, W., Prakash, latingligase DSBIV by repair a small by molecule HDR. This inhibitor intervention or by Kühn, R., and Chu, V.T. (2015). Pop in, pop out: a novel gene-targeting strategy for facilitatesadenoviral the proteins introduction is sufficient of knockin for stimu se- Chu, V.T., Weber, T., Wefers, B., Wurst, W., Sander, S., Rajewsky, K., and Kühn, R. use with CRISPR-Cas9. Genome Biol. 16, 244. as required for the precise modelling of pa- tientquence derived modifications mutations. in cell Furthermore, lines and mice, this Truong,(2015). IncreasingD.-J.J., Kühner, the K.,efficiency Kühn, R., of Werfel,homology-directed S., Engelhardt, repair S., Wurst, for CRISPR-Cas9- W., and Ortiz, and ongoing work on the active promotion O.induced (2015). precise Development gene editing of an in intein-mediated mammalian cells. split-Cas9 Nat. Biotechnol. system for 33, gene 543–548. therapy. of HDR provides a basis for the future appli- Buchholz, F., Friedrich, B., Gräb-Schmidt, E., Kühn, R., Müller, A., Müller-Röber, B.,Nucleic Propping, Acids P., Res. Pühler, 43, 6450–6458. A. ,Schlegelberger,B., Taupitz, J., Vogel, J., Winnacker, E.-L. gene therapy. (2015). The opportunities and limits of genome editing. Nationale Akademie der cation of the CRISPR/Cas system in somatic Studying disease pathways by Wissenschaften Leopoldina. ISBN 978-3-8047-3493-7 gene editing in human IPS cells Human induced pluripotent stem cells (hiP- Start of the Group: May 2014 Graduate and Undergraduate SCs) are early embryonic cell lines which can student be established by the redifferentation of skin Group Leader Eric Danner - Dr. Ralf Kühn (Guest, BIH research group) Technical Assistants propagatedfibroblasts throughbut also the differentiated expression of in repro vitro Johanna Kuchta intogramming major factors.cell types hiPSCs such canas neurons be indefinitely or car- Scientists Jana Roßius diomyocytes. By combining the hiPS cell and Dr. Sanum Bashir Anja Zimmer - Dr. Anne-Kathrin Dreyer cal research can investigate human genotype/ Dr. Saniye Yumlu Secretariat phenotypeCRISPR/Cas relations. nuclease The technologies nuclease-assisted biomedi Feraye Kocaoglu assembly of genotypes enables for example

MDC Research Report 2016 99 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Achim Leutz

Cell Differentiation and Tumorigenesis

The inherent and extensive self-renewal Synopsis: Structure & Function property of somatic stem cells vanishes of C/EBP The C/EBP family of transcription factors upon asymmetric division into differentiating harbor cell lineage instructive and trans- progeny, but may recur in tumor stem cells differentiation potential. They regulate and cause cancer. Central issues in tumor stem cell self renewal, lineage commit- stem cell biology and in the application of highly related family members. Targeted somatic stem cell based therapy are the mousement and genetics cell maturation. revealed that C/EBPα,β,δ,ε, the combined are disclosure of molecular mechanisms in- - volved in the generation of somatic stem bryonic lethality, whereas other compound C/EBPdeletion deletions of C/EBPα displayed and β causes more subtleearly em ef- cells, processes involved in self-renewal and cues to differentiation into functional the most essential factors of the C/EBP fam- cells. All these processes are regulated by ily.fects. The These structure data suggestof C/EBPs that is C/EBPα,βhighly modu are- transcription factors that integrate signals lar and adapts to signal dependent regulatory multitasking. The C-terminal basic into epigenetic programs. CCAAT Enhancer DNA binding leucine zipper domains (bZip) Binding Proteins (C/EBP) represent an in- can dimerize and bind to cis-regulatory teresting family of pioneering transcriptional palindromic ATTGC•GCAAT DNA sequence key regulators of cell lineage commitment, motifs. The C/EBP N-termini harbor strong trans-activation domains (TAD), whereas self-renewal, and cell differentiation in many the center sequences represent regulatory organs and cell types. We study how signal domains (RD). Both, TAD and RD consists dependent alterations of the C/EBP struc- of several conserved short peptide regions (CR) that are intervened by less conserved ture determine C/EBP functions in proliferation, differentiation and tumorigenesis. overall C/EBP structure suggests structural intrinsically disordered regions (IDR). The with co-factors and gene regulatory protein flexibility and adaptability to interactions

arecomplexes. single exon Interestingly, genes that theare pioneeringlocated on differenttranscription chromosomes. factors C/EBPα Nevertheless, and C/EBPβ vari-

regulated alternative translation initiation. ous C/EBPα,β isoforms can be generated by- pression interferes with cell differentiation andDysregulation may cause tumorigenic of C/EBPα,β cell isoforms transforma ex-

oftion. post-translational In addition to alternative amino acid translation, side chain C/EBPα,β can be decorated by a multitude

modifications (PTM). Signaling events that 100 MDC Research Report 2016 CANCER RESEARCH

sense growth factor stimulation, nutrition, or stress regulate both isoform transla-

- tivetion translation and post-translational and amino acid modifications side chain of C/EBPα,β. The combination of alterna functions by controlling the C/EBP interactome.modifications Hence, orchestrate how signal diversedependent C/EBP C/

areEBP essentialmodifications for the are understandingestablished and of how C/ EBPmodifications biology. Our determine research C/EBP focuses functions on the dynamic regulation of C/EBPs, their ability to balance cell proliferation and cell differentiation and the function of C/EBPs in disease.

Defective C/EBPβ isoform switching causes many diseases - ternative translation initiation at consecu- Isoforms of C/EBPα,β are generated by al- stream open reading frames (uORF) in the tive “in frame” AUG-start codons. Small up mediate appropriate start codon selection ofC/EBPβ individual mRNA isoforms. translation We leader have usedsequence tar- Translation of the single exon C/EBPβ transcript is regulat- geted mouse genetics to explore the imed by an upstream open reading frame (uORF). The uORF portance and vital function of individual senses the activity of the translational machinery and directs translation initiation to alternative start sites. The - resulting LAP*, LAP, and LIP protein isoforms have distinct isoforms of C/EBPβ and the consequences gene regulatory and epigenetic potential. Mouse strains, of defective C/EBPβ isoform switching. Mu- as shown below, that cannot switch between isoform have formsrine gene express replacement either only “knockin” the long mutants (LAP*, been constructed and develop many diseases. that can not switch between C/EBPβ iso

strainsLAP) or develop the truncated many diseases C/EBPβ and isoformdisplay defects(LIP), as in shown bone inhomeostasis, Figure 1. These liver mouseregen- oncoproteins. We explored the importance eration, mammary gland development, inof uORF mediated protein expression con- nate immune response, and metabolism. trol in the onco-developmentally important group of tyrosine kinases (TK) that from a single allele causes cancer in many are important regulators of cell fate and Expression of the C/EBPβ LIP isoform, even- proliferation control and are frequently

tissues and suggests LIP as a dominant on- TK transcripts harbor uORFs. Experimental generation,co-protein. Incell summary, differentiation, switching metabolic between explorationoverexpressed of the in tumors. impact ofAlmost TK-uORFs 64% ofthen all adaptation,C/EBPβ isoforms and prevention is important of cancer. in tissue re showed that removal of their uORFs indeed enhanced TK protein expression and may Importance of protein expression relate to increased cancer susceptibility. control by upstream open reading These results suggest driver functions for frames (uORF) uORF loss-of-function mutations in proto- Bioinformatic analysis of transcript leader oncogenes and disclose a novel mechanism sequence regions of the genome revealed - that evolutionary conserved uORFs occur ence with translational initiation and uORF regulationfor TK overexpression may thus represent in cancer. a potential Interfer target for treatment. Mechanistically,in many transcripts. uORFs Ribosomeconsume functional profiling pre-initiationconfirmed that complexes uORF are indeedand thus translated. curb the Post-translational C/EBP production of critical proteins encoded modifications provide an index downstream on the same mRNA. Accord- of epigenetic functions ingly, loss-of-function uORF mutations Seeking for an explanation for the multi- in proto-oncogenes may cause enhanced tasking C/EBP functionality, it came to our expression of potentially harmful proto- attention that many amino acid residues in

MDC Research Report 2016 101 CANCER RESEARCH

Schematic representation of lympho-myeloid trans-differentiation assay. Various myeloid cell types can be distinguished by flow cytometry (FACScan), reflecting the epigenetic potential of the C/EBPβ structure. Underneath: Plots show that lineage assignment during trans-differentiation depends on modifications of C/ EBPβ. LAP* and LAP as in Figure 1, KK>A indicates a C/EBPβ point mutant with previously identified lysine (K) PTM targets mutated into alanine (A).

- C/EBPβ structure and PTM in trans-differentiation analysisC/EBPα,β by represent mass spectrometry targets of (in post-trans collabo- C/EBPs are downstream of signaling cas- rationlational with modifications Gunnar Dittmar, (PTM). MDC) Systematic identi- cades and play important roles in hematopoietic lineage commitment and in estab- - ylation,fied dozens S-, T-, of and novel Y-phosphorylation. PTM on C/EBPα Bio andwe found that ectopic expression of C/EBPs chemicalβ, including and K-acetylation molecular genetic and K-,analysis R-meth re- inlishing many the cell innate types immune may activate system. myeloidInitially genes. Accordingly, C/EBPs entail the epigenetic competence to orchestrate gene regulatory,vealed that chromatinalteration ofmodifying individual and C/EBPβ chro- expression and instruction of the myeloid matinPTM sitesremodeling modifies complexes. interaction For with example, gene cell fate in a signal dependent fashion. The methylation of distinct K- and R-residues in trans-differentiation capacity of C/EBP to - convert B lymphoid progenitors into myeloid cells was employed to examine how complexes.the N-terminal We C/EBPβalso observed TAD changes signal depen the interaction with Mediator, SWI/SNF, and G9a cell fates. The results of structural-func- PTM that regulate global gene regulatory the structure of C/EBPβ directs alternative- functions,dent cross-talk such as between an activator various or repressor. C/EBPβ We think that C/EBP may act as a meeting trans-differentiationtional analysis showed into thatvarious C/EBPβ cell types, iso platform or assembly spot to coordinate in- suchforms as and monocytes/macrophages, PTMs pre-define the outcome dendrit of- teractions and diverse functions of protein ic cells, and granulocytes. Some of these complexes involved in shaping chromatin results are shown in Figure 2. Moreover, and regulating genes. point mutations of distinct PTM-sites may

102 MDC Research Report 2016 CANCER RESEARCH

- actions with other transcription factors and tinct chromatin remodeling complexes and co-factors in order to direct gene regulatory toalter induce the different ability of cell C/EBPβ types (Stoilova to recruit et dis al., and epigenetic machineries in a context spe- - suggest that the PTM pattern and the struc- 2013, PLoS ONE, 8:e65169). These findings numbercific fashion. of combinatorial The multitude interactions of C/EBP modi that modular properties to integrate and rewire fications comprises an extraordinarily high- epigenetictural flexibility functions of C/EBPβ during differentiation are adaptable dent functional and evolutionary plasticity. into diverse innate immune cells (Schön- Weimplicate anticipate an amplification that unraveling of C/EBP the PTM-de depen- pendent C/EBP interaction network will help - to elucidate basic and complex C/EBP func- ings?heit et Observations al. 2015, J Molof trans-differentiation Biol. 427:670-87). tions such as in hematopoietic cell lineage wereWhat initially are the regarded implications merely of theseas the find re- choice, cell differentiation, regeneration, innate immune regulation, and tumorigen- physiological correspondence. However, esis. incoherentsult of artificial cell lineages manipulation, and differentiation with little states were reported to emerge in pre-neoplastic tissues. Such alterations may occur Selected Publications leukemia/lymphoma known as ‘lineage inin epithelia, known as ‘metaplasia’ or in- mental cell reprogramming. Metaplasia and Cirovic B, Schonheit J, Kowenz-Leutz E, Ivanovska J, Klement C, Pronina N, Begay V, fidelity’ and may echo results from experi- Leutzstemcr.2016.12.015 A (2017) C/EBP-Induced Transdifferentiation Reveals Granulocyte-Macrophage WethmarPrecursor-like K, Schulz Plasticity J, Muro of EM,B Cells. Talyan Stem S, CellAndrade-Navarro Reports 8: 346-359. MA, Leutz DOI A10.1016/j. (2016) - Comprehensive translational control of tyrosine kinase expression by upstream open torylineage of trans-differentiationinfidelity are examples and of epigenetic dysregu reprogramming.lated cell differentiation We hypothesize that reflect that a histhe Regalo G, Forster S, Resende C, Bauer B, Fleige B, Kemmner W, Schlag PM, Meyer TF, remarkable trans-differentiation potential Machadoreading frames. JC, Leutz Oncogene A (2016) 35: C/EBPbeta 1736-1742. regulates DOI 10.1038/onc.2015.233 homeostatic and oncogenic gastric of C/EBPs plays a role in leukemic lineage Begay V, Smink JJ, Loddenkemper C, Zimmermann K, Rudolph C, Scheller M, - Steinemanncell proliferation. D, Leser J Mol U, MedSchlegelberger (Berl) 94: 1385-1395. B, Stein H, et DOI al. (2015)10.1007/s00109-016-1447-7 Deregulation of the mine the C/EBP interactome in conjunc- tioninfidelity. with its Studies trans-differentiation are underway tocapacity. deter Schellerendogenous M, Schonheit C/EBPbeta J, Zimmermann LIP isoform predisposes K, Leser U, Rosenbauerto tumorigenesis. F, Leutz J Mol A (2013) Med (Berl) Cross These studies will address the question 93: 39-49. DOI 10.1007/s00109-014-1215-5 whether reprogramming is involved in tu- talk between Wnt/beta-catenin and Irf8 in leukemia progression and drug resistance. 2013, EMBO Mol Med, 5:1154-64). The Journal of experimental medicine 210: 2239-2256. DOI 10.1084/jem.20130706 morigenic transformation (Regalo & Leutz, A C/EBPβ ‘Indexing Code’ in Group Leader partner choice Prof. Dr. Achim Leutz

Scientists Graduate and Undergraduate OurC/EBPs results may suggest regulate that and an direct “Indexing protein-pro Code”- Dr. Elisabeth Kowenz-Leutz students oftein interactions. post-translational The functional modifications consequenc on- Dr. Jörg Schönheit* Michael Gerloff* es of individual C/EBP-PTM sites can be ex- Christine Klement* amined in trans-differentiation tests (Figure Dr. Susann Förster* Nina Pronina* 2). Discerning the underlying mechanisms Dr. AlexandraJelena Ivanovska* Patmanidi* Anne-Margarethe Schmidt* Dr. Alexander Mildner* of differentially interacting partner proteins. Dr. Klaus Wethmar* Technical Assistants Weremains therefore difficult employ and requiresvarious methods the knowledge of sys- Maria Hofstätter tematic differential proteomics to discover PhD students Maria Knoblich PTM-dependent interactors of C/EBP (to- Undine Hill Victoria Malchin gether with Gunnar Dittmar, Mass Spectrom- Radoslaw Wesolowski Ruth Zarmstorff etry Core Unit). To date, our screens have Branko Cirovic Sarah Jaksch* revealed many transcription factors and epi- Julia Schulz Romy Leu* genetically important proteins that differen- Milda Masionyte* Secretariat * part of the tially interact with distinct C/EBP isoforms or Dorothea Dörr* period C/EBP-PTMs. Now, we explore how the Thuy Linh Nguyen* Sylvia Sibilak reported -

modifications on C/EBPs orchestrate inter MDC Research Report 2016 103 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Antonio Pezzutto

Molecular Immunotherapy

Goal of our work is to translate results of basic in CML qPCR for bcr/abl is a well standard- immunological research into preclinical mod- ized method to quantify MRD, even at very low levels. Therefore, minimal residual CML els and clinical trials. For this purpose, we is an ideal clinical setting for a) vaccination have a close cooperation with the MDC re- - search groups working on tumor immunology, ciated antigens in order to eradicate residual against leukemia-specific and leukemia-asso T.Blankenstein and W. Uckert: several projects leukemia cells and b) to monitor vaccine ef- were developed in joint efforts up to clinical on a successfully concluded clinical vaccina- trials. Our strategies cover passive and active tionficacy trial even using at a in subclinical vitro-generated, level. Following bcr-abl immunotherapy. We have performed several positive dendritic cells (DC) in patients with chronic myeloid leukemia (CML) we have clinical studies with dendritic cells in chronic started a clinical DC-vaccination trial in CML myeloid leukemia and renal cancer, gene- patients with minimal residual cytogenetic modified tumor cell vaccines for treatment of or molecular disease upon treatment with advanced metastatic renal cancer, lymphoma tyrosine-kinase inhibitors. The aim of the study is to improve cytogenetic/molecular radioimmunotherapy. In the field of DNA-vaccination we are optimizing immunogenicity of T cell responses against bcr/abl, proteinase-3 a vaccine directed against the HER-2 protein remission by the induction of CD4+ and CD8+ overexpressed in breast cancer. We will soon molecular remissions have become an attrac- and WT-1. In CML, long lasting „complete“ start a clinical trial with genetic engineered patients may remain bcr/abl-negative even T cells expressing a MAGE-A1 specific T cell tive goal of therapy since up to 40% of these receptor (developed by the MDC Blankenstein afterDNA cessation Vaccination: of TKI treatment.preclinical models group), and are further developing adoptive T Jörg Westermann, Tam Nguyen-Hoai cell therapy with TCR-engineered to recognize B-cell specific antigens and mutated antigens DNA vaccination offers several advantages for leukemia and lymphoma therapy. over the use of peptides as vaccines (DNA directly targets the endogenous presentation Dendritic cell-based vaccination in pathwaycovers several and contains MHC-I and immunostimulatory MHC-II epitopes, minimal residual CML under thera- CpG sequences). Furthermore, DNA vaccines py with tyrosine kinase inhibitors can be easily produced on large-scale for the J.Westermann and J.Kopp, in cooperation with Alexander Scheffold (DRFZ). cooperation with the group of M. Lipp (Mo- lecularuse in clinicalTumor trialsGenetics) across we HLA have barriers. explored In Minimal residual disease (MRD) is the only the possibility of recruiting immune cells by clinical setting in which active immuno- using plasmid DNA coding for the chemo- therapy using vaccines may hold promise for kines CCL19, CCL21 and CCL4 as possible ad- improving (molecular) remission and/or preventing relapse. Chronic Myeloid Leukemia and CCL21 with tumor antigens results in enhancementjuvants. In particular, of a Th1-polarized coexpression immune of CCL19 re- which standard treatment with tyrosine ki- sponse with substantial improvement of the nase(CML) inhibitors is a paradigm is able ofto achieve„targeted deep therapy“ molecu in- protective effect of the vaccine. Furthermore, lar remissions in most patients. Furthermore, in cooperation with S. Mathas we are studying

104 MDC Research Report 2016 CANCER RESEARCH

cell carcinoma of cervix and head and neck preclinical model. Further improvement of and in ovarian cancer, with a particularly fo- the vaccineadjuvant potency effect of has STATs been and achieved IRFs in using our cus on patients responding to treatment with a gene-gun for intradermal vaccine application (cooperation with O.Hohn/S. Norley, the are directly isolated from the fresh tumor samples/Ascites/pleuralcheck-point inhibitors. Tumor effusions specific for single-T cells results have been extended to a preclinical cell TCR sequencing and cloning. Moreover, modelRobert-Koch using Institute).the human These breast-cancer encouraging as- isolated T cells are tested for recognition of sociated antigen Her-2/neu as a target antigen, with the aim of developing a vaccine for Her2/neu+ tumors (particularly breast can- mutated tumor specific epitopes identified in cer) in the clinical situation of minimal resid- cellthe contextcarcinomas, of the isolatedBIH Cooperative T cells Researchare also ual disease after successful standard therapy. testedGrant. for In caserecognition of HPV of associated HPV associated squamous antigens. Finally, a clinical study using TCR-gene Identification of tumor-reactive T-cells for adoptive T-cell transfer tumor-associated antigen MAGE-A1 (gener- in solid tumors, lymphoma and atedmodified in T.Blankenstein T cells specifically group) recognizing is in advanced the leukemia preparation. GMP-expansion of T cells for Simone Rhein, Antonia Busse, Sara Boiani, treatment of MAGE-A1 positive multiple my- Neşe Çakmak, Corinna Grunert and Özcan eloma is done in the GMP laboratory of the Cinar in cooperation with T.Blankenstein, ERC by J.Kopp in cooperation with E.Kieback H.Schreiber, G.Willimsky and W.Uckert (MDC, and M.Obenaus (MDC Blankenstein group). Charité)

We focus on the treatment of different ma- Patents / Patent applications lignancies by establishing adoptive therapy and A. Pezzutto) PCT-Anmeldung PCT/EP2013/069569 “CD22-BINDING PEPTIDES” (O. Schmetzer antigenswith T cell such receptor as CD22 gene (S.Rhein), transfer. InCD79a different and Selected Publications CD79bprojects, (S.Boiani), we focus kappa on B-cell and lineage-specificlambda immu- Flörcken A, Grau M, Wolf A, Weilemann A, Kopp J, Dörken B, Blankenstein T, Pezzutto noglobulin light chains (A.Busse, Ö.Cinar), and also explore possible targets for myeloid leukemias such as FLT3 (N.Cakmack). So far, A,cell Lenz vaccine P, Lenz in advanced G, Westermann renal cell J. (2015). cancer: Gene tumor-induced expression immunosuppressionprofiling of peripheral and a blood mononuclear cells during treatment with a gene-modified allogeneic tumor these antigens (CD22 and CD79a/b) have Flörcken A, Kopp J, van Lessen A, Movassaghi K, Takvorian A, Jöhrens K, Möbs M, been targeted mainly by antibodies but not by Schönemannpossible role forC, Sawitzki NF-κB. Int. B, Egerer J. Cancer K, Dörken136: 1814-1826. B, Pezzutto A, Westermann J. (2013). Allogeneic partially HLA-matched dendritic cells pulsed with autologous tumor cell natural T cell receptors against intracellular antigensT cell therapy: might we be believesuperior that as the compared efficacy toof Nguyen-Hoailysate as a vaccine T, Kobelt in metastatic D, Hohn O, renal Vu MD, cell Schlag cancer: PM, a clinical Dörken phase B, Norley I/II study. S, Lipp Hum M, the CAR-recognized surface antigens which WaltherVaccin Immunother. W, Pezzutto 9:1217-27.A, Westermann J. (2012). HER2/neu DNA vaccination by intra- might be modulated in vivo giving origin to dermal gene delivery in a mouse tumor model: Gene gun is superior to jet injector in inducing CTL responses and protective immunity. Oncoimmunology. 1:1537-1545. Scholz C, Pinto A, Linkesch W, Linden O, Viardot A, Keller U, Hess G, Lastoria S, Lerch cooperation with T.Blankenstein (MDC), we K, Frigeri F, Arcamone M, Stroux A, Frericks B, Pott C and Pezzutto A. 90Yttrium- usesurface-antigen-negative a humanized mouse escapeexpressing variants. the hu In- man TCR gene repertoire. This mouse can Ibritumomab-Tiuxetan as First Line Treatemnt for Follicular Lymphoma: 30 months Nguyen-Hoai,of follow-up Data T., Baldenhofer, from an International G., Sayed Ahmed, Multicenter M.S., PhasePham-Duc, II Clinical M., Vu, Trial. M.D., (2013) Lipp, J antigens (which have not undergone thymic M.,Clin Doerken, Oncol 31: B., 308-313 Pezzutto, A., and Westermann, J. (2012). CCL21 (SLC) improves selection)generate high-affinity upon immunization T cells against with peptides, human tumor protection by a DNA vaccine in a Her2/neu mouse tumor model. Cancer Gene Therapy 19: 69-76. protein, DNA or RNA. The mouse is transgenic for the human HLA-A2 antigen. Clon- - Group Leader tion that can be transduced in patients T cells Prof. Dr. Antonio Pezzutto foring generationof reactive Tof cells large allows amounts TCR ofidentifica lineage- (Guest, Charité) - Özcan Cinar Scientists Corinna Grunert specific T cells with high therapeutic poten Dr. Antonia Busse withtial. In T.Blankenstein, the context of MPa cooperative Klötzel, M.Hummel, project of Dr. Tam Hoai Nguyen Graduate and Undergraduate H.Schreiber,the Berlin Institute W.Uckert) of Healthwe also (cooperation evaluate Dr. Simone Rhein student Prof. Dr. Joerg Westermann Neeta Donnimath

CARD11,mutation-specific EZH2, A20, epitopes as possible of key regulatorstarget for PhD students Other of lymphocyte pathways such as MYD88,- Sara Boiani Margarete Gries ect, A.Busse and C.Grunert are targeting tu- Nese Cakmak Nicole Hellmig TCR recognition (Ö.Cinar). In a different proj

mor specific antigens expressed in squamous MDC Research Report 2016 105 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Klaus Rajewsky

Immune Regulation and Cancer

Human B cell malignancies represent a major medical problem because of their abundance transgene, they fail to initiate a proliferative responseBCR-deficient to Toll-like B cells receptorare kept alive(TLR) by engage a Bcl2- and limitations of available therapies. We study ment (Otipoby et al. 2015). As in the case of B cell lymphomagenesis in comparison to nor- BCR-dependent B cell survival, the prolifera- mal B cell physiology, through genetic mouse models generated by conditional gene target- modeltive response emerges of thein BCR-deficientwhich B cells cells integrate could ing. Recent achievements include the identifi- be rescued by PI3K signaling. Altogether a cation of antagonizing PI3 Kinase and FOXO1 antigenic determinants) and co-receptors activities as key elements of germinal center B (respondingsignals from to the T cell BCR help (recognizing or pathogen-asso specific- ciated molecular patterns) in the control of cell differentiation, and their cooperativity in B - cell lymphomagenesis. Additional new insights formed mouse B cells the BCR controls the relate to the mechanism of immune surveil- their proliferative response. In Myc-trans- lance of Epstein-Barr-Virus (EBV) infected and al.competitive Nature, in fitness press 2017).of the tumor cells (collab transformed B cells, as well as various aspects oration with S. Casola, IFOM, Milan; Varano et of normal B cell development including its con- The germinal center reaction and trol by miRNAs. A major new direction of our B cell lymphomagenesis work comes from the availability of CRISPR/ S. Sander, V.T. Chu, T. Yasuda, A. Franklin, R. Graf, D. Calado*, K. Imami+, M. Selbach+, M. Cas-mediated targeted mutagenesis. We have Di Virgilio+, L. Bullinger+, E. Kabrani *Former improved this tool to introduce tailored muta- group members +Collaborators tions and insertions, and used it to generate genetically modified mice by zygote mutagen- - esis. A newly generated mouse strain allowing siveIn Trounds cell-dependent of proliferation immune and responsesselection conditional expression of the Cas9 endonucle- inantigen-specific histological structures B cells undergocalled germinal succes ase is being used for genetic screens in pri- centers (GCs). Proliferation is accompanied by somatic hypermutation (SHM) and class- mary cells, and we have added gene editing in switching (CSR) of the antibodies expressed human cells to our agenda. - ity antibodies are selected into the pool of The interplay between B cell anti- memoryby the cells, B cells and and cells long-lived expressing plasma high-affin cells. gen receptor and Toll-like receptors Because of their extensive proliferation ac- in the B cell proliferative response companied by DNA breaks introduced by K. Otipoby*, A. Waisman*, E. Derudder, L. SHM and CSR, GC B cells are at the origin of Srinivasan*, A. Franklin *Former group most B cell lymphomas. We had earlier mod- members eled a typical GC-derived malignancy, Burkitt lymphoma, in mice, by targeted induction of We had earlier shown that the B cell antigen receptor (BCR) is a key survival determi- Cancer Cell 2012). The resulting lymphomas nant for mature B cells, depending on sig- closelyc-MYC andresembled PI3K activity human in Burkitt,GCs (Sander including et al. tertiary mutations selected during tumor pathway. We now demonstrated that when progression. Among the latter we discov- nals of the phosphoinositide 3-kinase (PI3K)

106 MDC Research Report 2016 CANCER RESEARCH

Figure 1. (A) Non-synonymous FOXO1 mutations in murine and human GC B cell derived lymphomas (FHD, forkhead DNA- binding domain; NLS, nuclear localization signal; NES, nuclear export sequence; TAD, trans- activation domain). (B) Spleen sections showing GC B cells (red) surrounded by follicular B cells (green). In FOXO1 deficient GCs the network of follicular dendritic cells (blue) which demarcates the light zone in the wild type extends throughout the GC.

Figure 2. (A) After primary EBV infection in humans, some latently infected B cells persist for life. In immunocompromised patients EBV+ B cells expand and lymphomas develop. (B) Mouse model for the conditional activation of the oncogenic EBV proteins LMP1 and LMP2A. Mouse B cells which have turned on LMPs form characteristic colonies similar to colonies of human EBV infected cells (C) and develop monoclonal lymphomas in immunocompromised mice (D).

MDC Research Report 2016 107 CANCER RESEARCH

Figure 3. CRISPR/Cas9-mediated mutagenesis. (A) A transgenic „traffic light“ reporter (TLR) system to monitor DNA double strand break repair through NHEJ or HDR. NHEJ leads to RFP reporter expression, HDR in the presence of a Venus template to that of the Venus reporter. Suppression of NHEJ results in upregulation of HDR (red). (B) Scheme of C57BL/6 zygote mutagenesis. HDR observed in ~20% of mutagenic events. For details see text and Chu et al 2015; 2016.

ered recurrent activating mutations of the a recent study we showed that a fraction of human DLBCLs exhibit activation of the alternative NF-kB pathway, and that activation FOXO1 transcription factor (TF; Fig. 1A). As of this pathway plays indeed an oncogenic this resulted in co-expression of FOXO1 and role in a mouse model of DLBCL (Zhang et al. PI3K signaling, mutually exclusive in normal 2015). This complements previous evidence signalingphysiology in the where GC reaction. PI3K inactivates We found FOXO1, a clear that activation of the canonical NF-kB path- we investigated the role of FOXO1 and PI3K way, together with other oncogenic events, the so-called GC Light Zone (LZ), where mu- drives DLBCL of the ABC subtype. Ongoing division of labor, with PI3K activity limited to work explores the role of other recurrent expression largely restricted to the Dark Zone mutations in DLBCL in tumor pathogenesis, (DZ),tant B where cells are GC selected, B cells proliferateand FOXO1 andnuclear un- such as activating mutations in signaling pathways emanating from the BCR and TLRs. regulator of the GC DZ, controlling more than We are also developing mouse models of MM dergo SHM. FOXO1 turned out to be a master and other tumor entities through sequential DZs were not formed (Fig. 1B), and, although activation of oncogenic events recurrently B60% cell ofproliferation DZ-specific and genes; SHM in were its absence, unaffected, GC cellular selection and CSR in GCs were profoundly disturbed. We concluded that proper identifiedEpstein-Barr-Virus in the human. driven patholo- sequential proliferation/mutation and se- gies and immune surveillance; Hodgkin lymphoma (collaboration with the Janz/Mathas/Dörken lection of GC cells is controlled by the PI3K- group) FOXO1 antagonism and essential for efficient T. Yasuda, T. Sommermann, T. Weber, T. Wirtz, antibody affinity maturation in GCs (Sander S. Li, M. Janz+ mutationset al. 2015). are Interestingly, dependent on the concomitant Burkitt-like lymphoma cells carrying activating FOXO1- Since our discovery many years ago that the PR/Cas9 mediated targeted mutagenesis tumor cells in Hodgkin Lymphoma (HL), the (unpublishedPI3K and FOXO1 data). activity, Present as shownstudies by explore CRIS the mechanism of this unusual cooperation cells, originate from pro-apoptotic GC B cells and its role in lymphoma pathogenesis. which,so-called in HodgkinEBV+ cases, & Reed-Sternberghave been rescued (HRS) by expression of the EBV proteins LMP1 and Mouse models of Diffuse Large LMP2A, we develop mouse models of EBV B Cell Lymphoma and Multiple infection and the immune surveillance of Myeloma EBV-infected B cells, as well as EBV patholo- B. Zhang*, D. Calado*, Z. Wang+, S. Fröhler+, K. Köchert*, Y. Qian+, S. Koralov*, M. Schmidt- - Supprian*, Y. Sasaki*, C. Unitt+, S. Rodig+, W. proliferativegies such as Disorder X-linked (PTLD), Lymphoproliferative and EBV+ B Chen+, R. Dalla-Favera+, F.W. Alt+, L. Pasqualuc- cellSyndrome lymphomas (XLP), including Post-Transplant HL. Human Immuno B cells ci+, K. Schmidt, U. Sack*, W. Winkler, M. Janz+ infected and transformed by EBV rapidly

Various attempts have been and are being and NK cells, leaving behind a small pool of latentlyexpand, infected but are efficientlycells. When eliminated immune bysur T-

made to model DLBCL and MM in mice. In 108 MDC Research Report 2016 CANCER RESEARCH

veillance is compromised, the EBV infected ects, which are close to completion. These cells cannot be controlled and lethal patholo- - in B cell homeostasis and B1/B2 subset described that both EBV immune surveillance termination,relate to NF-kB the signaling role and and functional BCR specificity impact andgies pathologiesensue (Fig. 2A).can beIn earliermodeled work in mice we de by - methylation (with Y. Bergmann and H. Ceder, - Hebrewof Tet enzymes University), in lineage-specific and an unexpected, DNA devi- daB cell-specific et al. Cold expressionSpring Harb of Symp the EBV Quant proteins Biol tal role of miRNA seed matches in the ubiq- 2013).LMP1 and Fig. -2A2B-D (Zhang shows et our al. Cellconditional, 2012; Yasu Cre uitously expressed pro-apoptotic Bim gene recombinase dependent LMP alleles and the (with C. Birchmeier, M. Landthaler and N. expansion of LMP+ B cells and tumor forma- - tion in the absence of immune surveillance. ated by A. Franklin with O. Daumke and M. Di Ongoing work explores the immunogenicity Virgilio.Rajewsky). An AID-related project was initi and T cell recognition of LMP+ B cells and tumors, as well as the molecular mechanisms by which LMP+ B cells develop into monoclonal lymphomas in vivo. We have also devel- Selected Publications oped mouse models of acute EBV infection and a monogenic inherited fatal EBV-driven - lymphoproliferative human disease. With re- Chu, V.T., Weber, T., Wefers, B., Wurst, W., Sander, S., Rajewsky, K.*, Kühn, R.* Increas- spect to HL we refer the reader to the report ingresponding the efficiency authors of homology-directed repair for CRISPR-Cas9-induced precise of M. Janz and colleagues. Otipobygene editing K.L., inWaisman mammalian A., Derudder cells. Nature E., Srinivasan Biotechnology L., Franklin 2015. A.,33(5):543-8. Rajewsky K.*Cor The B-cell antigen receptor integrates adaptive and innate immune signals. Proceedings of the National Academy of Sciences of the United States of America 2015. 112 (39): Gene editing in somatic cells and 12145-12150. mouse zygotes V.T. Chu, T. Weber, R. Graf, S. Sander, - + Sandertor activity S., Chu differentially V.T., Yasuda control T., Franklin B cells A., in Graf the germinalR., Calado center D.P., Li light S., Imami and dark K., Selbach zones. T. Sommermann, R. Kühn M., Di Virgilio M., Bullinger L., Rajewsky K. PI3 kinase and FOXO1 transcription fac

- Immunity 2015. 43 (6): 1075-1086. PR/Cas9 revolution on our work, we have, in Chuimmune V.T., Grafcells R.,using Wirtz CrispRGold T., Weber and T., Favret a C57BL/6 J., Li X., Cas9 Petsch transgenic K., Tran mouse N.T., Sieweke line. Proceed M.H., - Berekings of C., the Kühn National R., Rajewsky Academy K. of Efficient Sciences CRISPR-mediated of the United States mutagenesis of America in 2016. primary collaborationRealizing the with profound R. Kühn, impact attempted of the to CRIS en- 113(44):12514-12519. hance homology-directed DNA repair (HDR) Wirtz T., Weber T., Kracker S., Sommermann T., Rajewsky K., Yasuda T. Mouse model at the expense of non-homologous end join- for acute Epstein-Barr virus infection. Proceedings of the National Academy of Sci- - genesis, in order to optimize this approach ences of the United States of America 2016. 29;113(48):13821-13826. ing (NHEJ) in CRISPR/Cas9-mediated muta- cations into genomic sequences. Chemical Group Leader PhD students ortowards genetic the inhibition introduction of the of preciseNHEJ pathway modifi Prof. Dr. Klaus Rajewsky Robin Graf indeed led to a strong increase of HDR and Eleni Kabrani almost complete suppression of NHEJ (Fig. Scientists Kristin Schmidt Dr. Jana Busch (guest from the Timm Weber Sieweke group) Wiebke Winkler (jointly with the We3A; Chualso etsucceeded al. 2015, publishedin the rapid together generation with Dr. Van Trung Chu Janz/Mathas/Dörken group) ofa similar mice carrying paper from knock-out the Ploegh or conditional lab at MIT). al- Dr. Emmanuel Derudder (until Nov Tristan Wirtz 2015) C57BL/6 zygotes (Fig. 3B). Particularly useful Dr. Andrew Franklin (Marie Curie Lab Manager forleles our through future work CRISPR/Cas9 is a strain mutagenesis carrying a con of- Fellow) Claudia Große ditional Cas9 allele. We have shown that this Dr. Verena Labi (until Dec 2014) Dr. Anja Köhler (until Aug 2014) Dr. Shuang Li (until Dec 2014, jointly and genetic screens in primary cells through with the Janz/Mathas/Dörken group) Technical Staff transfectionstrain allows of efficient guide-RNAs targeted (Chu, mutagenesis Graf et al. Dr. Kaaweh Molawi (guest from the Janine Cernoch (until April 2016) 2016). We are in the process of extending our Sieweke group, until March 2016) Jeremy Favret (guest from the work to gene editing in human cells. Dr. Sandrine Sander (until Nov 2015) Sieweke group) Dr. Thomas Sommermann Jenniffer Pempe Other work Dr. Ngoc-Tung Tran (Alexander von Kerstin Petsch MDC lab members involved: E. Derudder, Humboldt Fellow) A. Franklin, R. Graf, V. Labi, T. Yasuda Dr. Tomoharu Yasuda Secretariat Hannah Dunham (until Dec 2014) Due to space limitations, we refer the reader Riikka Lauhkonen-Seitz to the literature and only mention a few proj-

MDC Research Report 2016 109 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Armin Rehm

Translational Tumorimmunology

Studying cell biological mechanisms that define previous results on the immune-tempering the capacity of T cells to mediate protection effects of EBAG9, we are analyzing whether its targeted deletion stimulates the cytolytic from infections or tumors, we have identified activity of CTLs against leukemias. We pos- gene functions that amplify the secretion of ef- + T cells fector molecules. Because a high local effector - function of adoptively transferred T cells (ATT) tulateing us thatto combine EBAG9-loss EBAG9-targeted confers CD8 deletion with an enhanced cytolytic efficiency, allow- is key for tumor eradication, this approach can ciency provides T cells with improved sen- complement current efforts to select for tumor- sitivitywith TCR-engineering. toward weak tumor Thus, antigens. EBAG9-defi The antigen specific T cells. Our strategies to boost cytolytic capacity and its predisposition to T cell avidity and thus, improve ATT converge at enterrelationship the memory between T cell a T pool cell’s is heighthenedunder inves- the final steps of effector functions, as mediat- tigation. ed by the secretion of cytotoxic molecules. This principle can be applied to either T cell receptor Chimeric antigen receptor (CAR) T cells directed against target (TCR) or chimeric antigen receptor (CAR) -en- antigens expressed in hematologic gineered T cells. Moreover, our work is devoted malignancies to the elucidation of tumor-stroma interactions in hematological diseases. We combine animal and Wolfgang Uckert (Humboldt-University) JuliaIn collaboration Bluhm with Uta E. Höpken (MDC) models, cellular and molecular biology techniques as well as gene expression profiling to express CARs has proven tremendous suc- gain insight into the complex network of tumor cessThe adventin B cell of leukemia genetically treatment, modified despite T cells that the fact that patients were multiresistant against cells and a benign infrastructure within second- several lines of chemo- and antibody thera- ary lymphoid organs. These approaches are instrumental in gaining a profound understand- was administered successfully for salvage ing of mechanisms that cause lymphoma to be pies. In these cases, ATT using CAR-T cells cell neoplasia which are currently incurable, addicted to the local microenvironment. andtherapy. thus We are have suitable identified targets several for thetypes devel of B- opment of CAR-dependent immunotherapy. Adoptive Immunotherapy: exploit- Based on antibody sequences, we designed ing cell biological enhancers of - the secretory pathway tivity against Multiple Myeloma and other B- CARs with an efficient in vitro and in vivo ac Gerald Willimsky and Dana Hoser (Charité), aimed at a clinical translation phase. WolfgangIn collaboration Uckert with and UtaMario E. Höpken Bunse (MDC),(Hum- cell non-Hodgkin’s lymphoma. This project is boldt-University, MDC), Thomas Willnow Tumor and microenvironment: (MDC) Cellular and molecular prerequisites for onset and progression of Enhancing the release of gran- leukemia and lymphoma zymes to target leukemias Anthea Wirges, Julia Bluhm Leutz (all MDC), Georg Lenz (University Hos- - pitalIn collaboration Münster), Holger with Uta Gerhardt E. Höpken, and AchimAnne- Clemence Vion (MDC) Central to the efficiency of ATT is the ampli fication of functional avidity. Related to our 110 MDC Research Report 2016 CANCER RESEARCH

Figure 1. The transcription factor C/EBPβ controls the capacity of DCs to maintain Eμ-Myc B cell lymphoma survival. Tumor- derived factors induce a phenotyping skewing of DCs, characterized by C/EBPβ up-regulation and followed by cytokine and growth factor secretion. In addition, lymphoma-induced immunosuppression depends on the expression of C/EBPβ in context of DCs (for details, see Rehm et al. (2014), Nature Communications 5, 5057-5070)

Involvement of DCs and their tran- modal chemo-/immunotherapies with VEGF scriptional regulation in shaping a - lymphoma survival niche withinhibitors specialized showed endothelial no benefit cell in subsets non-Hodg that model tumor cells lodged within the T cell linekin’s the lymphoma. blood and Lymph lymphatic nodes vasculature. are equipped To In an aggressive Myc-driven B cell lymphoma- dissect lymphoma-induced alterations, we ular cells (FRCs), but also with DCs. Function- - ally,zone deletion and intermingled of DCs delayed with fibroblastic progression retic of thelial cells and FRCs. Selected gene functions these lymphomas. Lymphoma-exposed DCs haveapplied been gene validated expression in mouse profiling models of and endo on upregulated immunomodulatory cytokines, human lymphoma specimen. To elucidate growth factors and the transcription factor C/ - ploring whether vascular growth pattern in the preferential translation of the LAP/LAP* lymphomalymph node-specific are different peculiarities, from solid we tumors. are ex EBPβ. Moreover, Eμ-Myc lymphomas induced - were unresponsive to lymphoma-associated aptations of normally quiescent endothelial isoforms of C/EBPβ. C/EBPβ-deficient DCs cellsIn addition, during lymphomagenesis.we aim to identify metabolic ad Myc lymphoma cell survival. T cell prolifera- tioncytokine in lymphoma-bearing changes and unable mice to was improve impaired. Eμ- Selected Publications This immune suppression was reverted by Gätjen, M., Brand, F., Grau, M., Gerlach, K., Kettritz, R., Westermann, J., Anagnosto-

granulocytes acquire an accentuated neutrophil B cell helper phenotype in chronic functionsthe DC-restricted as pivotal deletion steps for of the C/EBPβ. creation Thus, of poulos,lymphocytic I., Lenz, leukemia. P., Lenz, Cancer G., Höpken, Res. 76, U.E.*, 5235-5265 Rehm, A.* (2016). Splenic*equal marginal contribution zone we have identified the C/EBPβ-controlled DC Herda, S., Raczkowski, F., Mittrücker, H.-W., Willimsky, G., Gerlach, K., Kühl, A.A., a lymphoma growth promoting and immu- Breiderhoff, T. Willnow, T.E., Dörken, B., Höpken, U.E., Rehm, A. (2012). The sorting nosuppressive niche. More recently, we analyzed the chronic lymphocytic leukemia (CLL) B cell-imposed stromal alterations along their receptor Sortilin exhibits a dual function in exocytic trafficking of interferon-γ and R.,granzyme Cseresnyes, A in TZ., cells. Hauser, Immunity A., Lenz, 37, P., 854-866 Hehlgans, T., Brink, R., Westermann, J., Dörken, - Heinig,B., Lipp, K., M., Gätjen, Lenz, G.,M., Rehm, Grau, M.,A.*, Stache, Höpken, V., U.E. Anagnostopoulos, *(2014). Access I., to Gerlach, follicular K., dendriticNiesner, ginal zone, B cell follicle and FDC network. We cells is a pivotal step in murine chronic lymphocytic leukemia B cell activation and trafficking route including the splenic mar- proliferation. Cancer Discovery 4, 1449-1465 *equal contribution ser capture microdissected spleens obtained Rehm, A. **, Gätjen, M., Gerlach, K., Scholz, F., Mensen, A., Gloger, M., Heinig, K., Lam- precht, B., Mathas, S., Begay, V., Leutz, A., Lipp, M., Dörken, B., Höpken, U.E.** (2014). fromperformed leukemic gene mice. expression A splenic profiling tissue remod on la- eling predominated by the expansion and Dendritic cell-mediated survival signals in Eμ-Myc B **cell shared lymphoma corresponding depend on authors the polarized differentiation of monocytes and the transcription factor C/EBPβ. Nature Communications 5, 5057-5070 - Wichner, K., Stauss, D., Kampfrath, B., Krüger, K., Müller, G., Rehm, A., Lipp, M., Höp-

neutrophils was identified. This innate im- ken,FASEB U.E. J. 30,(2015). 761-774 Dysregulated development of IL-17 and IL-21 expressing follicular helper T cells and increased germinal center formation in the absence of RORγt. lishedmune cellCLL infiltration disease macrophages was not associated exhibited with an M2a systemic polarization, inflammation. whereas Instead,neutrophils in estabwere skewed toward a neutrophil B cell-helper-like Group Leader Marcel Gätjen* phenotype. Myeloid cells provided additional Dr. Armin Rehm growth factors, but were also involved in the (Permanent Fellow) Undergraduates structural integrity of the spleen necessary Christof Bauer* for leukemia B cell homing. Scientist Maria Zschummel Dr. Lutz Menzel Maddi Arsuaga Azpiroz* Remodeling of the lymph node vasculature during lymphomagenesis PhD Students Technical Assistants Marleen Gloger, Lutz Menzel Marleen Gloger Kerstin Gerlach Anti-angiogenic strategies have become an Anthea Wirges Heike Schwede important therapeutic modality for solid tu- Julia Bluhm * part of the period reported mors. However, attempts to combine multi-

MDC Research Report 2016 111 CANCER RESEARCH Photo: Feriet Tunç Feriet Photo:

Oliver Rocks

Spatio-temporal control of Rho GTPase signaling

Cells frequently change their shape and their This cycle is controlled by three classes of attachment to the environment in order to regulatory proteins: the large families of activating RhoGEFs (Rho guanine nucleo- move or to move or to adopt to specialized tide exchange factors) and inactivating tissue structures. These events are driven RhoGAPs (Rho GTPase activating proteins), by coordinated changes of the cytoskeleton and are fundamental during embryonic nucleotide dissociation inhibitors), which sequesterand by RhoGDI the membrane proteins anchored (Rho guanine Rho development, immune surveillance or wound GTPases into the cytosol (Fig. 1). The Rho repair. Our lab is interested in the family of family comprises about 20 genes, with the Rho GTPase proteins which have emerged most prominent members RhoA, Rac1 and as master regulators of the cytoskeleton. Cdc42. Numerous Rho-regulated signaling pathways have been outlined over the past We combine advanced microscopy with cell years, many of which drive central aspects biological and genetic approaches to study of cellular morphodynamics. The different the precision control that ensures proper members of the Rho family thereby have distinct functions and can crosstalk with Rho signaling. Specificity of Rho GTPase each other, resulting in a complex interplay function is achieved by the huge family of of pathways. A precise spatio-temporal con- RhoGEF and RhoGAP regulatory proteins. trol of their activities is thus essential. The We aim to study in detail how these proteins couple Rho activity to specific environmental major challenge in the field is to understand cues and functional contexts in a cell. Our Thesehow this mechanisms signaling specificity are of great is achievedclinical rel by- lab also investigates the dynamics of local evancethe RhoGEF/GAP/GDI since deregulated regulatory Rho signaling proteins. is Rho signaling complex formation at the a key contributor to cancer metastasis and many other diseases. membrane at the single molecule level. Insights into this spatio-temporal regulation Schematic representation of the molecular switch func- of Rho signaling will provide a deeper tion of Rho GTPases and the function of their regulatory understanding of morphodynamic processes proteins. in cells and organisms, both in normal and disease settings.

Signaling Specificity of Rho GTPases Rho GTPases are molecular switches that

in the activated state they bind effector pro- teinscycle betweenand thereby an ‘on’ relay and anincoming ‘off’ state. signals Only further downstream of a signaling pathway.

112 MDC Research Report 2016 CANCER RESEARCH

microscales and employ model organisms - cal processes in vivo. We aim for a broader understandingthat closely reflect of the the function underlying of the biologi con-

scaffolds, ‘how they dynamically assemble, howtext-specific they relay signaling signals to scaffolds, the cytoskeleton, signaling and how they communicate with other cel-

lularTemporal signaling framework pathways’. of Rho GTPase signaling Very little is known also about the temporal control of Rho signaling. Most Rho fam-

be membrane associated to properly exert theirily GTPases signaling are lipidfunction. modified Exactly and how need on to membranes the formation of Rho signaling complexes is initiated, maintained and ter- minated and whether a precise temporal control is important for proper signaling is - ied in cells. We employ state-of-the-art imag- ingunclear techniques and technically at the single difficult molecule to be level stud to Modular domain architecture of the human analyze the dynamics of membrane interac- RhoGEF and RhoGAP proteins. tion of Rho GTPases and its modulation by regulatory and effector proteins. We thereby aim to elucidate the temporal framework in RhoGEFs and RhoGAPs which Rho signaling complexes operate and The genome encodes over 145 different RhoGEFs and GAPs which encode a diverse spectrum of subcellular targeting to find novel modes of signal regulation. and protein interaction domains, besides Selected Publications their catalytic domains (Fig. 2). This mul- Eccles, RL., Czajkowski, MT., Barth, C., Müller, PM., McShane, E., Grunwald, S., Beau- tidomain architecture enables them to as- dette, P., Mecklenburg, N., Volkmer, R., Zühlke, K., Dittmar, G., Selbach, M., Hammes, A., Daumke, O., Klussmann, E., Urbé, S., Rocks, O. (2016) Bimodal Antagonism of PKA signaling by ARHGAP36. Nat Commun. - Rho activity to critical sites of cytoskeletal ms12963. rearrangement,semble specific proteinsuch as complexes focal adhesions, to target Rocks, O., Gerauer, M., Vartak, N., Koch, S., Huang, Oct 7;7:12963. ZP., Pechlivanis, doi: 10.1038/ncom M., Kuhlmann, cell junctions, endocytic vesicles or mi- totic spindles. Likewise, the regulators can also connect Rho GTPases to other signal J.,Proteins. Brunsveld, Cell L., Chandra, A., Ellinger, B., Waldmann, H., Bastiaens, PI. (2010) The Dekker,Palmitoylation FJ.*, Rocks, Machinery O.*, Vartak*, Is a Spatially N., Menninger*, Organizing S., System Hedberg, for C., Peripheral Balamurugan, Membrane R., transduction pathways. However, the ma- Wetzel, S., Renner, 141(3):458-471 S., Gerauer, M., Schölermann, B., Rusch, M., Kramer, JW., Rauh, jority of RhoGEFs and GAPs is only poorly characterized and the cellular repertoire inhibition of APT1 affects Ras localization and signaling. Nat Chem Biol 6(6):449-56 D.,*equal Coates, contribution GW., Brunsveld, L., Bastiaens, PI., Waldmann, H. (2010) Small-molecule of spatio-temporal control of Rho GTPases Rocks, O. not fully explored. Therefore, we have re- signals: one ship, three anchors, many harbors. Curr Op Cell Biol - Rocks, O., Peyker, A.,A. &Kahms, Bastiaens, M., Verveer, PI. (2006) PJ., Spatio-temporalKoerner, C., Lumbierres, segregation M., Kuhlmann, of Ras terization of all mammalian RhoGEFs and 18(4):351-7 - cently carried out a first systematic charac Science 307:1746- GAPs, yielding a complete cDNA library of J.,1752 Waldmann, H., Wittinghofer, A. & Bastiaens, PI. (2005) An Acylation Cycle Regu all regulators and datasets of their inter- lates Localization and Activity of Palmitoylated Ras Isoforms. actome, subcellular localization, overexpression and knockdown phenotypes and Group Leader Rebecca Eccles Dr. Oliver Rocks Paul Markus Müller unique resource ‘one the one hand to study thesubstrate control specificities. of Rho signaling We now at theexploit systems this Scientist Graduate and Undergraduate level and on the other hand to characterize Dr. Maciej Czajkowski students individual RhoGEFs and GAPs implicated Kristin Frensemeier in different signaling contexts. These are PhD students Philipp Trnka currently cell adhesion, guided cell migra- Carolin Barth tion and G protein coupled receptor/cAMP Matti Baumann Technical Assistants - Juliane Brümmer Marlies Grieben croscopy and biosensors to obtain informa- Lisa Spatt tionsignaling’. on appropriate To this end, spatial we use and live temporal cell mi

MDC Research Report 2016 113 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Claus Scheidereit

Signal Transduction in Tumor Cells and Development

Our laboratory focuses on gene expression - and signal transduction processes, which ing cascades are activated by various cell- surfacespecific familyand intracellular members. Canonical receptors signal with are involved in tumorigenesis and develop- rapid kinetics. Signal transmission involves ment. The nuclear factor-κB (NF-κB) tran- the formation of K63-linked and linear scription factors are critical regulators of - inflammation, immunity, stress response, teins, which mediate recruitment of regula- polyubiquitin chains at TRAF or RIP pro cell proliferation, differentiation and survival. complex. The latter is activated via phos- NF-κB activity is controlled by inhibitory IκB phorylationtory protein ofcomponents, activation loop including serines. the Cata IKK- proteins and the IκB kinase (IKK) complex. The NF-κB/IκB/IKK signaling system is of Negativelytically active feedback IKK regulationtriggers degradation of canonical of relevance for the pathogenesis of a wide IκBs and liberation of prototypic p50-p65. spectrum of human diseases. - uitinatingsignaling pathwaysenzymes, such is provided as A20 or by CYLD. NF-κB dependent expression of IκBα and deubiq The molecular constituents of IKK and NF-κB signaling cascades triggeringNon-canonical C-terminal NF-κB signalingprocessing is inducedof NF- by a specific set of receptors, classically p105), NFKB2 (p52/p100), RelA (p65), c- pathway proceeds with a slow kinetics and RelThe andNF-κB RelB, family which comprises form homo- NFKB1 and (p50/ het- κB2/p100 (see below). The non-canonical erodimers and regulate numerous physiological processes by controlling inducible levelsrequires are NF-κB-inducing kept low by a kinasedestruction (NIK) com and- IKKα. In unstimulated cells NIK expression - gene expression programs. NF-κB activity is- doubleplex containing strand breaks TRAF andand cIAP depends proteins. on nu- regulated by IκB proteins, e.g. the cytoplas A third IKK pathway is activated by DNA p50mic inhibitorand p52, areIκBα generated or the nuclear by proteasomal co-activa processingtors Bcl-3 andof IκB-ζ.their Twoprecursor NF-κB subunits,proteins, clear shuttling and SUMO-modification of p105 and p100, respectively. IKKγ,Processing as well as of on NF- theκ kinaseB precursors ATM. p105 and p100 (Patrick Beaudette, Inbal Ipenberg, Buket Yilmaz) NF-κB activation requires the IκB kinase (IKK) complex, composed of two serine- threonine kinases (IKKα and IKKβ) and of their processing products p52 and p50, widethe regulatory variety of extra- subunit and IKKγ intracellular (also known sig- NF-κB p100 and p105 are the precursors- nalsas NEMO). to catalyze The phosphorylation IKK complex integrates of various a ing nuclear translocation and DNA-bind- but can also act as cytoplasmic IκBs, block The mechanisms that control processing intoIκB anddistinct NF-κB sub-pathways, proteins as well which as ofactivate other oring complete of other destruction Rel/NF-κB familyof the members.precursor substrates. NF-κB signaling can be divided

114 MDC Research Report 2016 CANCER RESEARCH

Figure 1: The LTβR pathway triggers interdependent proteolysis of the NF-κB precursor proteins p100 and p105. NIK and IKKα regulate p105 proteolysis through C-terminal degron serines in p100. The co-generation of p50 and p52 depends on VCP/p97 ATPase activity. Quantitative mathematical models predict direct signal responsiveness of p100-p105 complexes.

proteins by the ubiquitin-proteasome sys- ous protein degradation and processing tem have been considered as independent processes. We could show that VCP/p97 - ATPase activity is required to promote derstood. Using mass-spectrometry-based quantitativeevents; however, analyses they remain(cooperation poorly with un - Gunnar Dittmar) we could demonstrate that taneousLTβR-triggered processing processing of p100 and of p105 p100 might and havep105 important(Yilmaz et implications al. 2014; Figure for the 1). regulaSimul- - tion of target genes and for the physiologi- ingactivation of both of p100 non-canonical and p105. NF-κBThe precursors signaling cal and pathological functions of precursor- formby LTβR hetero-complexes. stimulates simultaneous Stimulus-depen process- dent proteolysis strictly depends on the signal response domain (SRD) of p100, sug- Canonicalderived NF-κB and subunits non-canonical (see below). NF- gesting that the SRD-targeting proteolytic κB/IKK gene networks in lympho- machinery acts in cis and in trans. Separa- magenesis tion of cellular pools by isotope labeling re- (Linda von Hoff, Eva Kärgel, Kívia A. Pontes vealed synchronous dynamics of p105 and de Oliveira) p100 proteolysis – predominant degradation of preexisting precursors and process- - ing of newly synthesized molecules (Yilmaz bers contributes to various lymphoid ma- lignanciesDeregulation via of promoting, NF-κB and amongstIκB family others, mem and colleagues, quantitative mathematical proliferation and survival. However, it re- modelset al. 2014). were In developed cooperation that with describe Jana Wolf the mained unknown whether and how differ- dynamics of the system and predict that p100-p105 complexes are signal responsive distinct sets of target genes and control se- lectiveent NF-κB oncogenic dimers functions.and nuclear IκBs regulate - (Yilmaz et al. 2014; Figure 1). phoma (HL) cell lines as well as primary beIt is achieved. still puzzling We howcould partial demonstrate proteasomal that HodgkinWe could anddemonstrate Reed-Sternberg that Hodgkin’s cells display lym bothproteolysis precursors of the interact NF-κB with precursors the ATPase can VCP/p97, a well-known modulator of vari- which triggers cell survival and growth by high constitutive activity of IKK/NF-κB,

MDC Research Report 2016 115 CANCER RESEARCH

Figure 2: DNA damage activates NF-κB through ATM and PARP-1 signaling. PARP-1 is activated and sequesters PIASy, ATM and IKKγ, which shuttles between nucleus and cytoplasm, resulting in IKKγ SUMOylation. After cytoplasmic export, ATM binds to TRAF6 to induce K63-linked poly-Ub formation. Subsequent steps are TAK1 activation, IKKγ mono-ubiquitination and IKK activation. DNA damage induced NF-κB regulates cellular apoptosis and senescence and may cause che- motherapy resistance.

inducing an anti-apoptotic and pro-prolif- which are believed to mediate largely operative gene program. HL cells display con- posing cellular reactions. While p53 in- stitutive canonical as well as non-canonical promotes cell survival. To analyze the func- - tionalduces cell-cycleimpact of arrestboth transcription or cell death, factorsNF-κB cursorsIKK/NF-κB p100 signaling and p105 and occurs overexpress in a simulta Bcl-- in more detail, we performed global gene 3. We found that processing of NF-κB pre expression studies and observed an unex- - blastsneous, (seeNIK-dependent above). To determinemanner, as cistromes observed and p53 mediated gene regulation. The andfor LTβR target signaling gene signatures in mouse of embryo canonical fibro as mechanisticpected interdependence details of this between cross-talk NF-κB are currently under investigation.

andwell microarray as non-canonical analyses. NF-κB We speciesfound that and have led to an improved understanding of p50-RelAof Bcl-3, and we p52-RelB performed activate ChIP-sequencing or repress Recent findings from our group and others transcription of distinct genes and have pathway (Hinz et al. 2010). We showed that selective as well as overlapping oncogenic the DNA damage induced NF-κB signaling functions. The type of gene regulation is de- DNA double strand breaks requires a bi- termined by the distance and subunit occu- furcatedactivation pathway of IKK and triggered NF-κB inby response poly(ADP- to ribose)-polymerase-1 (PARP-1) and ataxia regulation patterns distinguish Hodgkin telangiectasia mutated (ATM) as DNA pancies of NF-κB binding regions. The gene strand break reading heads (Figure 2). The malignancies and may contribute to the understandinglymphoma from of otherlymphomagenesis NF-κB-associated (de Oliveira et al. 2016). proteingenotoxic complex IKK pathway formations depends that areon uniquenot in- modifications, transport processes and- Mechanism of IKK and NF-κB ing NEMO-SUMOylation, nuclear export of activation by genotoxic stress ATM,volved and in “classical”PAR-dependent IKK pathways, protein interacinclud- (Cristina Brischetto, Daniel Heinze, Michael Hinz, Marina Kolesnichenko, Nadine Mikuda, - Michael Willenbrock) tions. These specific requirements of the pathway-selectiveDNA damage induced inhibitors, NF-κB signalingwhich might cas DNA lesions evoke multiple cellular re- havecade offerthe potential the possibility to be used to find to overcomegenotoxic sponses including DNA repair, cell cycle checkpoint control as well as activation of cooperation with the screening unit of the therapy resistance in cancer treatments. In-

the transcription factors NF-κB and p53, Leibniz-Institute for Molecular Pharmacol 116 MDC Research Report 2016 CANCER RESEARCH

ogy (FMP) we could identify small molecu- bud formation and patterning, as well as lar weight compounds, which interfere se- hair bud down-growth.

activation. Further preclinical studies such ectodermal organogenesis, we have inves- aslectively target with validation DNA damage and testing induced candidate NF-κB To unravel further functions of NF-κB in- inhibitors are ongoing (collaboration with natures during hair follicle development, Jens von Kries and Marc Nazaré). tigated NF-κB-dependent target gene sig- Yet, several steps of the genotoxic signaling tional role in hair follicle induction, includ- ingwhich ECM revealed remodeling, that NF-κB orchestration has a multifunc of WNT, understood and unknown essential com- ponentscascades and that activating activate IKKas well are as yet negative poorly feedback regulators are likely involved. SHH and TGFβ2 signaling and regulation of- - trolsthe stem processes cell marker essential LHX2. for We placode showed down- that etry strategies to identify novel pathway growth,the NF-κB-LHX2-TGFβ2 such as E-cadherin signaling down-regula axis con- We utilized SILAC-based mass spectrom tion (Toman et al. 2016). Furthermore, we cooperation with the groups of Udo Heine- mann,regulators. Markus In aLandthaler parallel study, and Jana based Wolf, on it a in hair follicle regeneration (hair cycle) and could be demonstrated that the RNA decay- inwere regeneration able to define of the a key intestinal function epithelium. for NF-κB promoting protein RC3H1 (also known as Roquin) binds preferentially to short-lived and DNA damage-induced mRNAs, indicating a role of this RNA-binding protein in the Selected Publications post-transcriptional regulation of the DNA de Oliveira, K.A., Kaergel, E., Heinig, M., Fontaine, J.F., Patone, G., Muro, E.M., Mathas, damage response. Notably, RC3H1 affects S., Hummel, M., Andrade-Navarro, M.A., Hübner, N., and Scheidereit, C. (2016) A

- p50 and p52 revealed by genome-wide analyses. Genome Med Tomann,roadmap P.,of Paus,constitutive R., Millar, NF-κB S.E., activityScheidereit, in Hodgkin C., and lymphoma:Schmidt-Ullrich, Dominant R. (2016) roles Lhx2 of sultedexpression in increased of the NF-κB A20 pathwayprotein expression, regulators 8(1):28 IκBα and A20. Knockdown of RC3H1 re growth. Development Murakawa,is a direct NF-κB Y., Hinz, target M., Mothes,gene that J., promotesSchuetz, A., primary Mastrobuoni, hair follicle G., Friedel, placode C.C., down- Dölken, thereby interfering with IκB kinase and NF- L., Kempa, S., Schmidt-Supprian, 2016 1;143(9):1512-22 M., Bluethgen, N., Heinemann, U., Wolf, J., Scheide- reit, C. and Landthaler, M. (2015) RC3H1 posttranscriptionally regulates A20 mRNA pathwayκB activities, (Murakawa demonstrating et al. 2015). that RC3H1 Nature Commun 6:7367. can modulate the activity of the IKK/NF-κB Role of Eda-A1/Edar/NF-κB sig- and modulates the activity of the IKK/NF-κB pathway. naling in development and organ Yilmaz, Z.B., Kofahl, B., Beaudette, P., Baum, K., Ipenberg, I., Weih,Cell F., Rep Wolf, 9, 1756–1769. J., Dittmar, Hinz,G., Scheidereit, M., Stilmann, C. (2014) M., Çöl Quantitative Arslan, S., Khanna, Dissection K.K., and Dittmar, Modeling G., and of theScheidereit, NF-κB p100- C. regeneration p105 Module Reveals Interdependent Precursor Proteolysis. - (Karsten Krieger, Ruth Schmidt-Ullrich*, Mol Cell 40, 63-74. Philip Tomann) *Principal Scientist (2010) A cytoplasmic ATM-TRAF6-cIAP1 module links nuclear DNA damage signal ing to ubiquitin-mediated NF-κB activation. Group Leader embryonic development and organ regen- Prof. Dr. Claus Scheidereit eration,To elucidate various the in loss-of-function, vivo function of NF-κBgain-ofin Karsten Krieger* function and reporter mouse models were Scientists NadineInbal Ipenberg Mikuda generated in our laboratory. We previously Ahmet Bugra Tufan* - Dr. Michael Hinz Michael Willenbrock duction of hair follicle and secondary lymph Dr. MarinaIrene Coin* Kolesnichenko nodedemonstrated development a key androle infor the NF-κB morphogen in the in- Dr. Kivia A Pontes de Oliveira Technical Assistants esis of ectodermal organs such as teeth or Dr. Philipp Tomann* Lydia Blankenstein* mammary glands. Further studies mainly Dr. Ruth Schmidt-Ullrich used the hair follicle as a model organ. Es- Dr. Zekiye Buket Yilmaz* Dr. Eva Kärgel sential signals that regulate early hair fol- DorisInge Krahn Lange* PhD students as well as TNF family member ectodyspla- Patrick Beaudette Secretariat sinlicle A1development (EDA-A1) andinclude its WNT/β-catenin,receptor EDAR. Cristina Brischetto* Daniela Keyner We demonstrated an interdependence of Daniel Heinze* Linda von Hoff * part of the time reported signaling which is required for focal hair WNT/β-catenin and EDA-A1/EDAR/NF-κB

MDC Research Report 2016 117 CANCER RESEARCH Photo: private Photo:

Clemens A. Schmitt

Cancer Genetics and Cellular Stress Responses in pathogenesis and treatment of lymphatic malignancies

The increasingly faster and detailed mo- found in human diffuse large B-cell lym- lecular dissection of cancer on one side phoma (DLBCL), and particularly enriched for in those primarily or secondarily local- and the rapidly growing list of lesion-based therapeutic agents on the other side are - about to make the vision of “personalized tion,izing butto the also CNS. tumor-suppressive NF-κB may be oncogenicregarding cancer precision medicine” a clinical reality, itswith role respect in mediating to cell survival cellular and senescence. inflamma although many obstacles need to be over- mutations in vivo, where they accelerated come before. In particular, tumor hetero- Eµ-We mycmodeled transgenic distinct DLBCL-derivedlymphoma develop NF-κB- geneity, bioinformatics processing of mul- ment. Moreover, we found lymphomas with tiple omics data layers and their functional associated CNS manifestation to overex- interpretation remain major challenges. We vice versa - not only employ genetically tractable mouse motepress lymphomaan NF-κB target CNS tropism. gene signature, Next, we and,will models of cancer and apply integrative , distinct NF-κB mutants to pro- tor mechanisms, and seek to identify novel “cross-species” bioinformatics, but focus functionally dissect NF-κB-governed effec on conceptually novel – synthetic lethal and “restore & target“ – therapies whose target vulnerabilities (as therapeutic “Achilles’ principles are cell states with their associ- omics-basedHeels”) in reverse drug (in)sensitivitygenetics models, model as well and currentlyas in our expanded “PanOmics”-dissected repository of forwardDLBCL ated vulnerabilities (such as cellular senes- material-based patient-derived xenograft cence) rather than single molecular lesions. “co-clinical trial-like“ functional and multi- omics-based(PDX) mouse analyses models (Fig.in mouse A). Importantly, models ac- Mechanistic dissection, functional company a new investigator-initiated trial modeling, and clinical exploitation headed by this group leader, in which previ- of DLBCL-derived pathogenetic ously untreated DLBCL patients receive the components (e.g. the BCR/NF-κB signaling network) in lymphoma ®) and the proteasome blocker development, CNS tropism, and Bruton’sBortezomib tyrosine (Velcade kinase®) as inhibitor a proximal/dis Ibrutinib- therapy (Imbruvic Soyoung Lee, Maurice Reimann, Aitomi Bittner, Animesh Battacharya and collabora- anti-CD20tal BCR/NF-κB Rituximab double-targeting plus CHOP standard “novel/ tion partners (Claus Scheidereit, Jana Wolf, immunochemotherapynovel” small compound backbone expansion (termed of the Nikolaus Rajewsky and others) an intense multi-omics-based molecular Hyperactivating mutations in the B-cell re- explorationthe “ImbruVeRCHOP” prior to and study), during and exposure undergo to the trial medication (Fig. B).

ceptor (BCR)/NF-κB cascade are frequently 118 MDC Research Report 2016 CANCER RESEARCH

Data-dense “PanOmics” investigation of drug (in)sensitivity in a transgenic mouse lymphoma model followed by cross-species exploitation in human data sets and humanized PDX mouse models (A). Data-dense multiple-omics-based analyses of DLBCL patients in the ImbruVeRCHOP IIT, flanked by mouse model (transgenic and PDX)-based “co-clinical trials” that clinically and molecularly dissect components of the poly-agent regimen in order to identify determinants of sensitivity and resistance (B).

The repressive H3K9me3 mark as not only correct a central malignant prop- essential senescence relay and erty, but also render cHL susceptible to therapeutic target in lymphoma clinically established therapies tailored and melanoma Yong Yu, Bin Yue, Maja Milanovic Jan R. Dörr, NHL), thereby broadening (chemo-free) Soyoung Lee, and collaboration partners for B-cell Non-Hodgkin’s lymphoma (B-- tion with the FMP, we successfully con- ductedtherapeutic a high-throughput options for cHL. pharmacological In collabora terminal cell-cycle arrest that precludes screening to identify drugs that promote pre-malignantOncogene-induced cells from senescence further (OIS)expansion is a re-expression of the B-cell phenotype in via transcriptional silencing of prolifera- cHL cells and sensitized them to clinically tion-related genes in senescence-associated approved antibodies (such as anti-CD20 heterochromatin foci (SAHF), which are Rituximab) and small compound inhibitors characterized by the repressive trimethylated histone H3 lysine 9 (H3K9me3) modi- - - (such as Ibrutinib and Idelalisib) – thereby methylases – namely the Jumonji domain providing proof-of-efficacy for this concep containingfication. We 2C now(JMJD2C) found and H3K9-active the lysine-spe de- tually novel “Restore & Target” strategy. - Selected Publications nescence and permit direct transformation Milanovic, M., J.H.M. Däbritz, Z. Zhao, J.R. Dörr, Y. Yu, L. Dimitrova, D. Lenze, M. Metzner, undercific demethylase-1 oncogenic Ras (LSD1) or Braf. – to Accordingly, disable se K. Pardon, B. Dörken, M. Hummel, A. Trumpp, and C.A. Schmitt. 2016. Senescence- melanoma cell lines and primary melanoma associated reprogramming promotes cancer stemness. In revision. samples present with high-level H3K9 de- Yu, Y., B. Yue, S. Ji, P. Lohneis, L. Li, D. Dhawan, N. Qutob, M. Werner-Klein, M. Reimann, A. Elkahloun, S. Treitschke, B. Dörken, F. Mallette, Y. Samuelis, C.A. Schmitt, and S. Lee. methylase activity, and respond to targeted H3K9 demethylases disable oncogene-induced senescence and serve as therapeutic inhibitor therapy in vitro and in vivo with se- targets for senescence restoration in melanoma. In revision. nescence restoration. Similarly, H3K9-active Du, J., M. Neuenschwander, Y. Yu, J.H.M. Däbritz, N.-R. Neuendorff, K. Schleich, A. demethylase expression countered therapy- Bittner, M. Milanovic, G. Beuster, S. Radetzki, E. Specker, M. Reimann, F. Rosenbauer, S. Mathas, P. Lohneis, M. Hummel, B. Dörken, J.P. von Kries, S. Lee, and C.A. Schmitt. 2016. Pharmacological restoration and therapeutic targeting of the B-cell phenotype in clas- However, detrimental features of senescent Blood. Epub ahead of print. cells,induced such senescence as associated (TIS) reprogramming in lymphoma. Däbritz, J.H.M., Y. Yu, M. Milanovic, M. Schönlein, M.T. Rosenfeldt, J.R. Dörr, A.M. sical Hodgkin’s lymphoma. into cancer stemness, ask for their second- Kaufmann, B. Dörken, and C.A. Schmitt. 2016. CD20-targeting immunotherapy promotes cellular senescence in B-cell lymphoma. Mol Cancer Ther ary co-therapeutic elimination, a conceptu- Dörr, J.R., Y. Yu, M. Milanovic, G. Beuste, C. Zasada, J.H.M. Däbritz, J. Lisec, D. Lenze, A. ally novel strategy we are currently investi- Gerhardt, K. Schleicher, S. Kratzat, B. Purfürst, S. Walenta, W. Mueller-Klieser, 15: 1074-1081. M. Gräler, M. Hummel, U. Keller, A.K. Buck, B. Dörken, L. Willmitzer, M. Reimann, S. Kempa, S. Lee, and C.A. Schmitt. 2013. Synthetic lethal metabolic targeting of cellular senescence in cancer therapy. Nature, 501:421-425. gatingPharmacological (“Enforce & Eliminate”). restoration and therapeutic targeting of the B-cell phenotype in classical Hodgkin’s Group Leader Clinician Scientists lymphoma Prof. Dr. Clemens Schmitt Dr. Aitomi Bittner, M.D. Yong Yu, Kolja Schleich, Soyoung Lee, and M.D (Guest, Charité) Dr. Sophy Denker, M.D. collaboration partners (Jens Peter von Kries, Stephan Mathas) Vice Group Leader Graduate students Dr. Soyoung Lee, Ph.D. Dimitri Belenki - Damaris Anell Rendon though originating from B-cells, is charac- Lab Scientists Bin Yue terizedClassical by Hodgkin’s the virtual lymphomalack of gene (cHL), products al Dr. Dorothy N. Fan, Ph.D. whose expression constitutes the B-cell Dr. Maja Milanovic, Ph.D. Technicians phenotype. Epigenetic repression of B-cell- Dr. Maurice Reimann, Ph.D. Nadine Burbach Dr. Kolja Schleich, Ph.D. Sven Maßwig contribute to the lost B-cell phenotype in Dr. Yong Yu, Ph.D. Dana Tschierske cHL.specific Restoring genes wasthe previouslyB-cell phenotype reported may to

MDC Research Report 2016 119 CANCER RESEARCH Photo: J.-M. Huron Photo:

Michael Sieweke

Stem cell and macrophage biology

Our research is located at the interface of tory and metabolic diseases. Macrophages immunology and stem cell research, with are present in essentially every tissue of the body, where they can play trophic roles in the overarching question of how cell iden- tissue homeostasis, metabolism and repair. tity and cell numbers are controlled. In many Macrophages cannot only originate from he- organs of the body cells are frequently lost matopoietic stem cells via blood monocyte and need to be replaced as part of normal intermediates but can also be derived from embryonic progenitors and be maintained tissue maintenance or in response to injury. long term in tissues by local proliferation (re- In most cases new functional cells are gener- viewed in Sieweke and Allen, Science 2013). ated from tissue-specific stem cells that are Macrophages could thus be considered sys- themselves un-specified but can self-renew. temic tissue guardians and changes in their self-renewal capacity could have major im- As stem cells differentiate into functionally pact on health and disease of parenchymal mature cells, this self-renewal capacity is typically lost. Macrophages, a specific cell type of development and maintainance of tissue macrophagescells. In this frameworkby differentiation we investigate from hema the- the immune system, represent a rare excep- topoietic stem cells or local self-renewal and tion to this pathway, since they can be main- the role of macrophage activation states in in- tained independently of blood stem cells by local proliferation of mature cells. We investi- flammationStimulation and of tissue macrophage regeneration. differ- gate the mechanisms that enable such self- entiation from hematopoietic stem renewal without loss of differentiated function. cells The manipulation of macrophage self-renewal We could show that M-CSF, a myeloid cyto- might have therapeutic benefit in cancer, in- kine important for macrophage proliferation and polarization can also act directly flammatory disease and tissue regeneration. on hematopoietic stem cells to stimulate Insight into the underlying mechanisms could the myeloid differentiation pathway that provide methods for the generation of large gives rise to macrophages. Previously it had been widely assumed that cell fate choice in numbers of macrophages, as source of de- stem cells is largely determined by stochas- fined cellular therapies for these applications. tic mechanisms. Using multiple methods of Finally, the understanding of the difference of single cell analysis we could demonstrate normal self-renewal to tumorigenic transfor- that M-CSF can directly instruct hematopoietic stem cells (HSC) to adopt a myeloid fate mation is also of critical importance for cancer (Mossadegh,Sarrazin et al., Nature 2013) in research. an integrated circuit with the transcription factor MafB (Sarrazin et al., Cell 2009). These mechanisms appear to be active in emergen- Macrophages as tissue guardians cy hematopoiesis when physiological levels Macrophages have recently entered the lime- of M-CSF are high, such as during infections. light of immunology and beyond (Geissman et We are now developing projects to investi- al., Science 2010, Sieweke and Allen, Science 2013), because they are involved in many populations and could show that the mecha- physiological and pathological processes of nismgate how can bethis used influences to treat mature neutropenic macrophage condi- major interest beyond their immune functions after hematopoietic stem cell transplan- - tation (Kandalla, JEM 2016).

tions, such as regeneration, cancer, inflamma 120 MDC Research Report 2016 CANCER RESEARCH

Macrophage self-renewal activates gene network shared with stem cells

the macrophage transcription factors MafB The self-renewal potential of both embryonic stem cells and dif- andWe previouslycMaf enables discovered self-renewal that deficiencyof differenti for- ferentiated macrophages is dependent on a shared network of ated functional macrophages without malig- self-renewal genes (left) that are controlled by distinct, lineage- nant transformation or stem cell intermedi- specific enhancers (right). In quiescent macrophages, the tran- ates (Aziz et al, Science 2009). We have now scription factor MafB binds and represses these enhancers. Loss investigated the gene regulatory mechanisms of MafB expression results in enhancer activation and enables that enable macrophage self-renewal capac- macrophage self-renewal. (Figure: Serena Billie) ity to be integrated into the overall program of epigenetic macrophage identity by ChiPseq Patents / Patent applications analysis of gene regulator enhancer elements. Our results demonstrate that self-renewal Method for generating, maintaining and expanding monocytes, and/or macrophages and/or dendritic cells in long term culture, delivered 2014 in macrophages involves MafB/cMaf down- Method for expanding monocytes, delivered 2015 regulation and concomitant activation of a A method for inducing extended self-renewal of functionally differentiated somatic self-renewal gene network that is centered on cells, delivered 2015 Myc and Klf2. We observed that the network Methods and compositions for use in preventing or treating myeloid cytopenia and is shared with embryonic stem cells but con- Methods for expanding a population of Alveolar Macrophages in a long term culture, trolled from a completely distinct set of cell related complications, filed 2013

- Selectedfiled 2015 Publications atedtype withspecific self-renewal enhancers genes (Soucie are etalready al. Science pres- ent2016, in quiescent fig.1). Macrophage cells but can enhancers become activated associ O. Matcovitch-Natan, D. R. Winter, A. Giladi, S Vargas Aguilar, A. Spinrad, S. Sarrazin, H. Ben-Yehuda, E. David, F. Zelada Gonzalez, P. Perrin, H. Keren-Shaul, M. Gury, D. when direct repression by Maf transcription Lara-Astaiso, C. A. Thaiss, M. Cohen, K. Bahar Halpern, K. Baruch, A. Deczkowska, factors is relieved either by genetic inactivation or in resident macrophage populations Microglia development follows a stepwise program to regulate brain homeostasis. ScienceE. Lorenzo-Vivas, S. Itzkovitz, E. Elinav, M. Sieweke*, M. Schwartz*, Ido Amit* that express constitutively or transiently low Prashanth K. Kandalla, Sandrine Sarrazin, Kaaweh Molawi, Carole Berruyer, David Redelberger,,353(6301):aad8670 Anne Favel, Chrstophe (2016) Bordi, Sophie de Bentzmann, Sieweke* joint lastMH. author M- a general molecular rationale for the compat- CSF improves protection againsgt bacterial and fungal infestions after hematopoietic ibilitylevels ofof MafBself-renewal and cMaf. and Our differentiated findings provide cell stem/progenitor cell transplantation, J Exp Med, 213 (11) 2269-2279 (2016) functions. We also observed that these pro- Gimenez G, VanHille L, Beniazza M, Favret J, Berruyer C, Perrin P, Hacohen N, Andrau JC, cesses are relevant for maturation of resident SoucieFerrier EL,P, Dubreuil Weng Z, P,Geirsdóttir Sidow A, Sieweke L, Molawi MH. K, MaurizioActivation J, ofFenouil Self-Renewal R, Mossadegh-Keller Gene Network N, macrophages, such as microglia in vivo (Mat- Science, 351 (6274):aad5510 (2016) kovitch et al. Science 2016). Molawi K, Wolf Y, Kandalla PK, Favret J, Hagemeyer, Frenzel K, Pinto AR, Klapproth K,on Henri a Macrophage-Specific S, Malissen B, Rodewald Enhancer H-R, Platform. Rosenthal NA, Bajenoff M, Prinz M, Jung S, Sieweke MH. Progressive replacement of embryo-derived cardiac macrophages with Macrophages in tissue regeneration age. J Exp Med Mossadegh-Keller N, Sarrazin S, Kandalla PK, Espinosa L, Stanley RE, Nutt SL, Moore , 211(11) :2151-8 (2014) model organisms and organs that macro- cells. Nature phagesIt has been are demonstratedrequired for tissue recently regeneration in several J, Sieweke MH; M-CSF instructs myeloid lineage fate in single haematopoietic stem but the relative contribution of self-renewing , 9 ;497(7448) :239-43 (2013) tissue macrophages and macrophages derived Group Leader PhD students Dr. Michael Sieweke Clara Busch (BIH-Einstein Fellow, Helm- Noushine Mossadegh-Keller asfrom a highly infiltrating relevant inflammatory in vivo system monocytes for under has- holtz-INSERM Joint Research Stephanie Vargas Aguilar standingremained the unknown.Investigating regulation of tissue regeneration, the heart Group) we have shown that embryo-derived cardiac Technical Assistants macrophages show declining self-renewal Senior scientists Wilfried Cathou with age and are progressively substituted Dr. Carole Berruyer Jeremy Favret by monocyte-derived macrophages. Replace- Dr. Sandrine Sarrazin Noushine Mossadegh-Keller ment of embryo-derived cells by less proliferative monocyte-derived macrophages coin- Post-docs Bioinformaticians cides with the loss of functional regeneration Dr. Magdalena Canali Gregory Giminez capacity of the heart and might therefore be Dr. Prashanth Kumar Kandalla Julien Maurizio an important factor in this process (Molawi Dr. Bérengère de Laval et al., JEM 2014). We are currently develop- Administrative Assistant ing projects investigating this question and Karin Geiser extending it to other organs such as the lung.

MDC Research Report 2016 121 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Thomas Sommer

Intracellular Proteolysis

Mechanisms of Protein Homeostasis misfolded proteins in a process termed ER associatedcific components protein of thedegradation UPS eliminate (ERAD). such pre-existing proteins are constantly endan- The ERAD pathway cooperates with the geredIn all by cells, misfolding both newly and aggregation. synthesized The and Unfolded Protein Response (UPR) in main- accumulation of such damaged proteins taining homeostasis in the ER. The UPR perturbs cellular homeostasis and pro- measures the degree of misfolding in the vokes aging, pathological states, and even ER and regulates folding and proteolytic ca- cell death. To avert these dangers, cells have pacities according to the load with aberrant developed protein quality control (PQC) proteins (Fig. 1). ERAD appears to be con- strategies that counteract protein damage served from yeast to mammals. Thus, many basic principles and components have been systems channel misfolded proteins either discovered in the model organism Saccha- intoin a re-folding compartment-specific pathways or manner.initiate their PQC romyces cerevisiae. Key components of the proteolysis by the Ubiquitin Proteasome - System (UPS). Key components of the UPS nations or Ubiquitin Ligase complexes such involved in cellular homeostasis are spe- asERAD the HRD pathway Ligase, are the specific Doa10 E2/E3 Ligase combi or the Asi-Complex. These membrane-bound E3 (E2) enzymes and Ubiquitin Ligases (E3 complexes recognize signals contained in enzymes).cific combinations The work of Ubiquitinof this group Conjugating focuses misfolded proteins of the ER-lumen and the ER-membrane. Proteins committed for selectively dispose aberrant proteins with- degradation are exported from the ER in outon how affecting specific correctly complexes folded of polypeptides.E2’s and E3’s a process termed protein dislocation and ubiquitylated (Fig. 2). However, substrates One of the best investigated PQC path- of these ligases are not limited to the ER, ways exist in the endoplasmic reticulum but also include proteins of the cytosol and (ER). The ER is a cellular organelle through the nucleoplasma. which a large number of proteins travel - membranes, exocytic and endocytic com- tin Ligases exist in other compartments or partments,on their way or tothe their cell exterior. final destination The ER is ina cellIn addition types, which to that, are further involved specific in maintain Ubiqui- major folding compartment of eukaryotic ing protein homeostasis as well. An example - for that are skeletal muscle cells. Skeletal erones, which assist in protein folding and muscle atrophy, a combination of reduced maturation.cells. It hosts Still, an arrayprotein of biogenesis molecular chapis an muscle mass and strength, accompanies error-prone process. A considerable frac- many diseases and negatively affects the tion of all newly synthesized polypeptides course of these illnesses. Muscle atrophy is fail to attain their native conformation due caused by a disturbed protein homeostasis to mutations, transcriptional and transla- with an increase in protein degradation, tional errors, folding defects, or imbalanced which is predominantly mediated by the subunit synthesis. Mature proteins can be - damaged by environmental stress condi- uitin ligase family, which consists of three tions, such as high-energy radiation, chemi- highlyUPS. The homologues muscle RING-finger proteins (MuRF1, (MuRF) 2ubiq and cal insults, or metabolic by-products. Spe- 3) are important factors to maintain mus-

122 MDC Research Report 2016 CANCER RESEARCH

Figure 1: Protein homeostasis in the ER. Proteins transported into the ER associate with molecular chaperones, only correctly folded proteins can leave the ER for their final destinations. Misfolded proteins are retained in the ER and are either subjected to re-folding or dislocated from the ER for destruction by the UPS (ERAD pathway). The UPR measures the content of misfolding in the ER, signals it to the nucleus, and up- regulates chaperones and UPS components of the ERAD branch.

cular protein homeostasis. MuRF1, is highly progressive assembly of ubiquitin chains up-regulated during atrophy, ubiquitylates is managed by the responsible Ubiquitin muscular proteins such as myosin heavy Ligases. Only recently ubiquitin binding chain (MyHC) and actin and directs them to activity has emerged as an important fac- proteasomal degradation. tor in Ub-chain formation. The Ubc7 activator Cue1 carries a ubiquitin binding CUE domain, which substantially stimulates and characterized components of these UbiquitinIn the last Ligasedecade, complexes the group hasusing identified genet- by the E2 enzyme Ubc7. Our results from - NMR-basedK48-linked analysis polyubiquitination in combination mediated with cation strategies. However, this laboratory interaction studies and in vitro ubiquitina- hasics, molecularnow established biology, powerful and protein in vitro purifi ap- tion reactions point out that two parameters proaches and quantitative methods us- accelerate ubiquitin chain assembly: the increasing number of CUE binding sites and the biochemical principles of the action of the position of CUE binding within a grow- Ubiquitining purified Ligases. components to understand ubiquitin moiety adjacent to the acceptor ubiquitining chain. facilitate In particular chain interactions elongation. withThese a Projects in the laboratory: data indicate a novel mechanism for ubiquitin binding in which Cue1 positions Ubc7 and The CUE domain of Cue1 aligns the distal acceptor ubiquitin for rapid polyu- growing ubiquitin chains with biquitination. Disrupting this mechanism re- Ubc7 for rapid elongation sults in dysfunction of the ERAD pathway by Maximilian von Delbrück and Lukas Pluska a delayed turnover of some protein quality in cooperation with the laboratory of Volker control substrates (von Delbrück et al., 2016, Dötsch (Goethe University Frankfurt) MOL CELL).

Ubiquitination mostly produces polymeric signals. Amazingly, little is known how the

MDC Research Report 2016 123 CANCER RESEARCH

Figure 2: Hrd ubiquitin ligase as an example of a quality control element. ER-luminal modul of the ligase (Hrd3, Yos9) associates with the chaperone Kar2. Der1, Usa1, and Hrd1 collectively build the membrane module of the ligase, which most likely forms a dislocation channel. The cytosolic module consisting of Hrd1 and Cue1/Ubc7 decorates dislocated proteins with poly-ubiquitin, which leads to recruitment Ubx2 and Cdc48/Ufd1/Npl4. This AAA-ATPase most likely provides energy for dislocation and transfers substrates to the 26S proteasome.

Sequential poly-ubiquitylation by to conjugate Ub not only to lysine but also different E2 enzymes expands the to other residues increases the target range range of a quality control E3 ligase. of the Doa10 quality control Ub ligase. This Annika Weber and Ernst Jarosch in coopera- work provides new insights on how the tion with the laboratories of Tommer Ravid combined activity of specialised Ub con- (Hebrew University) and Gunnar Dittmar jugating enzymes contributes to both the (MDC) - uitylation (Weber et al., 2016, MOL CELL). Quality control ubiquitin ligases label mal- versatility and specificity of protein ubiq folded polypeptides with ubiquitin (Ub) The WD repeat-containing protein chains for proteasomal degradation. These WDR42A is a new co-factor of the enzymes must on the one hand facilitate MuRF1 ubiquitin ligase in skeletal the attachment of Ub to a large set of highly muscle atrophy. diverse clients and on the other hand me- Marcel Nowak in cooperation with the labo- diate accurate formation of uniform lysine ratories of Jens Fielitz (ECRC), Erich Wanker (MDC) and Gunnar Dittmar (MDC) for the recognition by the proteolytic machinery.48-linked Studying poly- Ub, the which yeast is aDoa10 prerequisite Ub li- The MuRF proteins are suggested to func- gase pathway in vitro and in vivo we show tion as indivudual ubiquitin ligases im- that different ubiquitin conjugating en- plicated in degradation of Myosin heavy zymes take over distinct functions during chain (MyHC) or actin by the proteasome. substrate processing: Ubc6 mounts Ub on MuRF1 is highly up-regulated under skel- Doa10 client proteins, which primes them etal muscle atrophy and mice that lack MuRF1 are resistant to atrophy. The molecular function of the MuRF proteins and for lysine 48-linked poly-ubiquitylation by Ubc7. Importantly, the propensity of Ubc6 124 MDC Research Report 2016 CANCER RESEARCH

the regulation of their activity is still only a complex encompassing the yeast lectin- poorly understood. Using high throughput like protein Htm1 and the oxidoreductase

Pdi1 converts Man GlcNAc2 on glycopro- we found several known and also new teins into the Man GlcNAc 7 8 2 MuRFYeast-2-Hybrid associated and proteins. SILAC-AP-MS Among those screens we the Htm1/Pdi1 complex processes both - unfolded and native proteins signal. albeit In with vitro, a

(WD)identified repeat-containing DDB1- and CUL4-associated protein that spefac- expression of HTM1 causes glycan trim- tor 8 (DCAF8), a tryptophan-aspartic acid mingpreference on misfolded for the former.and folded In vivo, proteins, elevated but and MuRF3 in cells. This association made only degradation of the non-native species itcifically highly bindslikely thatand co-localizesMuRF1 may withnot function MuRF1

as an individual E3 enzyme but function as Man7GlcNAc2 structure does not inevitably integral parts of larger cullin-type ubiqui- commitis accelerated. a protein Thus, for ERAD. modification The function with of a tin ligase complex. Cullin-type E3 enzymes Htm1 in ERAD relies on its association with play a role in cell cycle control and other Pdi1, which appears to regulate the access essential cellular processes and are char- to substrates. Our data support a model in - which the balanced activities of Pdi1 and ing co-immunoprecipiation we found that - acterized by a central ‘cullin’ subunit. Us cient removal of misfolded secretory glyco- proteinsHtm1 are (Köhler crucial etdeterminants al., 2016, JBC). for the effi OurMuRF1 results is complexed demonstrates with that DCAF8, MuRF1 DDB1, is Cul4A, and the RING finger protein Rbx1.- gase complex in differentiated C2C12 skel- etalpart ofmuscle a specific myotubes. cullin-type Furthermore ubiquitin we li Selected Publications highly up-regulated during denervation-in- Mehnert, M., Sommer, T., and Jarosch, E. (2014) Der1 promotes movement of mal- ducedcould showmuscle that atrophy. MuRF1 Finally, and we DCAF8 showed are folded proteins through the endoplasmic reticulum membrane. Nature Cell Biol., 16, that siRNA-mediated down-regulation of Denzel, M.S., Storm, N.J., Gutschmidt, A., Baddi, R., Hinze, Y., Jarosch, E., Sommer, T., 77-86 prevented dexamethasone-induced atro- Hoppe, T., and Antebi, A. (2014) Hexosamine pathway metabolites enhance protein DCAF8 in C2C12 skeletal muscle myotubes phy and MyHC degradation. These results Pfeiffer, A., Stephanowitz, H., Krause, E., Volkwein, C., Hirsch, C., Jarosch, E., and Som- establish a cullin-type ubiquitin ligase commer,quality T. (2016) control A and complex prolong of life.Htm1 Cell, and 156, the 1167-1178oxidoreductase Pdi1 accelerates degrada- - tion of misfolded glycoproteins J. Biol. Chem. 29, 12195-12207. pant in skeletal muscle atrophy. This specif- von Delbrück, M., Kniss, A., Rogov, V.V., Pluska, L., Bagola, K., Löhr, F., Güntert, P., Sommer, T. and Dötsch, V. (2016) The CUE domain of Cue1 aligns growing ubiquitin icprising ligase DCAF8 complex and could MuRF1 represent as a vital a partici target for the development of new anti-atrophic drugs. These results imply that WDR42A E.,chains (2016) with Sequential Ubc7 for rapidpoly-ubiquitylation elongation. Mol by Cell specialized 62, 918-928 conjugating enzymes ex- functions as an important co-factor of the Weber, A., Cohen, I., Popp, O., Dittmar, G., Reiss, Y., Sommer, T., Ravid, T., and Jarosch, MuRF1 ubiquitin ligase in skeletal muscle pands the versatility of a quality control ubiquitin ligase. Mol. Cell 63, 827-839. atrophy, which makes it a highly interesting target for the development of new anti- Group Leader atrophic drugs. Prof. Dr. Thomas Sommer

A complex of Htm1 and the oxido- Scientists Susanne Kreßner reductase Pdi1 accelerates degra- Dr. Maren Berger Christian Lips dation of misfolded glycoproteins Dr. Ernst Jarosch Lukas Pluska Anett Köhler, Christian Hirsch, and Ernst Dr. Anett Köhler Jarosch in cooperation with the laboratory Dr. Marcel Nowak Technical Assistants of Eberhard Krause (FMP) Dr. Benjamin Sünkel Mandy Mustroph Dr. Annika Weber Corinna Volkwein Angelika Wittstruck - PhD students sylIn the and ER, two an enzymaticN-acetylglucosamine cascade generates residues a Alba Ferri Blázquez Secretariat specific N-glycan structure of seven manno (Man7GlcNAc2) on misfolded glycoproteins Sara Knape Sylvia Klahn to facilitate their disposal. We show that

MDC Research Report 2016 125 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Wolfgang Uckert

Molecular Cell Biology and Gene Therapy

Adoptive T cell therapy (ATT) of cancer requires quent generation of TCR-engineered T cells the preparation of antigen-specific T cells that enabled us to characterize TCR properties. recognize and eradicate tumor cells. While the use of tumor-infiltrating lymphocytes showed withOur approach various diseases led to the and identification cancer. of novel some efficacy of ATT for some tumor entities, the TCRs specific for viral antigens associated use of T cells engineered with T cell receptors Non-Viral Sleeping Beauty trans poson gene transfer for ATT (TCR) or chimeric antigen receptors (CAR) paves J. Clauß, M. Bunse in collaboration with the way for a broader clinical application. Our Z. Iszvak (MDC) and Z. Ivics (PEI) group focuses on several aspects of TCR/CAR Transposon-based vectors are an alternative gene therapy (Fig. 1) including the identification of to viral vectors for fast and convenient genet- suitable target antigens, the generation and isola- ic engineering of T cells. The use of transpo- tion of antigen-specific TCRs, the improvement of sons requires transfection of DNA into T cells vector systems for the generation of gene-engi- as they lack the ability of viral vectors to pass genetic material through the cell membrane. neered T cells, the development of safety mea- We found that DNA transfection by electro- sures to limit side effects of TCR gene therapy, poration affected the cell viability in a dose- and the evaluation of ATT in pre-clinical models. dependent manner and the responsiveness of

molecular mechanisms behind these obser- Identification and isolation of vationsT cells towardswere investigated CD3/CD28 and stimulation. an improved The antigen-specific TCRs for ATT protocol was developed requiring only a F. Lorenz, E. Hilgenberg, K. Dudaniec in collabo- limited amount of transfected DNA. Stable ration with D. J. Schendel (Munich) expression of TCRs and CARs in human T

A bottleneck of TCR gene therapy is the avail- size-reduced transposon vector in combina- ability of potent TCRs, which target therapeu- cells (up to 50%) was achieved by applying a tically relevant cancer antigens. We developed an experimental approach that enables the stimulation,tion with the the hyperactive engineered SB100XT cells mediated Sleeping rapid detection of immunogenic T cell epitopes cytokineBeauty transposase.secretion and Upontumor antigen-specificcell killing. Cur- of target antigens and the isolation of antigen- rently, we are working on the development of an improved transposon-based expression stimulated with antigen-expressing dendritic vector encoding genetically optimized TCR or cells,specific which TCRs. process To do and so, present donor Tonly cells crucial were CAR genes in combination with miRNAs tar- peptides on MHC complexes at the cell surface. geting the endogenous TCR chains. - plex receive proliferation signals and expand. RNAi-mediated TCR knockdown ToDonor identify T cells these specific T cells for without a peptide:MHC knowing com the prevents autoimmunity in TCR exact epitope and MHC restriction, we screened gene therapy M. Bunse in collaboration with T. Schumacher (Amsterdam) the T cells for specificity by using a MHC library pre-dominant(artificial antigen TCR presentingchains of T cells cells expressing respond- the donor’s MHC class I alleles). To identify TCR heterodimer can form potentially self- performed next generation sequencing. Subse- reactiveIn TCR-engineered mixed TCR dimers, T cells composed the transduced of ening to a specific peptide:MHC combination, we

126 MDC Research Report 2016 CANCER RESEARCH

Schematic representation of TCR (A) and CAR gene therapy (B). T cells are isolated from a cancer patient and modified ex vivo using transfer vectors encoding antigen-specific TCR or CAR genes. Engineered T cells are then expanded and reinfused into the patient. CAR: chimeric antigen receptor; LTR: long terminal repeat; TCR: T cell receptor.

dogenous and transferred TCR chains. To pre- genic mouse model of established tumors to vent the formation of mixed TCR dimers, we evaluate two CDK4 mutations (R24C, R24L) developed a retrovirus vector that employs as targets for TCR-gene therapy. T cells, which RNAi to silence the endogenous TCR and to facilitate expression of an RNAi-resistant, showed similar reactivity to R24C and R24L in therapeutic TCR. By using mouse and human vitro,were but modified expanded with and a patient-derivedrejected tumors TCR,only TCR-engineered T cells we showed that TCR in response to R24L in vivo. Such differences expression without RNAi results in heterodi- in neoantigen quality might explain why can- merization of transduced and endogenous cer immunotherapy induces tumor regres- TCR chains while the amount of mixed TCR sion in some individuals, while others do not respond, despite similar mutational load.

modeldimers of is TCR significantly gene transfer-induced reduced when graft- using Patents / Patent applications the RNAi-TCR replacement vector. In a mouse received T cells transduced with a standard versus-host-disease (TI-GVHD) all mice that Europäischereceptors by Patentanmeldungmeans of an MHC cellAnmeldenummer: library EP 15159212.8, Titel: Method of detecting new immunogenic T cell epitopes and isolating new antigen-specific T cell retrovirus vector developed lethal TI-GVHD Selected Publications preventeddue to formation when expression of mixed TCRof the dimers. endoge In- Bunse M, Bendle GM, Linnemann C, Bies L, Schulz S, Schumacher TN, Uckert W. nouscontrast, TCR TI-GVHDwas abrogated was by almost RNAi. completelyTherefore, (2014). RNAi-mediated TCR knockdown prevents autoimmunity in mice caused by the use of the RNAi-TCR replacement vector instead of standard transfer vectors liberates mixed TCR dimers following TCR gene transfer. Molec Ther 22: 1983-1991. TCR gene therapy from potential autoim- Obenausthe cancer-testis M, Leitão antigen C, Leisegang MAGE-A1 M, Chen from X,antigen-negative Gavvovidis I, van hosts. der BruggenNature Biotechnol P, Uckert W, mune side effects. Schendel33: 402-407. DJ, Blankenstein T. (2015). Optimal-affinity human T cell receptors against Blankenstein T, Leisegang M, Uckert W, Schreiber H. (2015). Targeting cancer-specif- Targeting tumor neoantigens by Leisegang M, Kammertoens T, Uckert W*, Blankenstein T*. (2016). Targeting human TCR gene therapy ic mutations by T cell receptor gene therapy. Curr Opin Immunol 33: 112-119. M. Leisegang in collaboration with H. *equal contribution Schreiber (Chicago) and T. Blankenstein (MDC) Kiebackmelanoma E, Hilgenbergneoantigens E, by Stervbo T cell receptorU, Lampropoulou gene therapy. V, Shen J Clin P, Bunse Invest M, 126: Jaimes 854-858. Y, Boudinot P, Radbruch A, Klemm U, Kühl AA, Liblau R, Hoevelmeyer N, Anderton T cell responses in successful cancer immu- SM, Uckert W, Fillatreau S. (2016). Thymus-derived regulatory T cells are positively selected on natural self-antigen through TCR interactions of high functional avidity. notherapy seem directed towards neoantigens created by somatic mutations. Targeting neoantigens by TCR gene therapy requires a Immunity 44:114-1126. complex and personalized approach that we Group Leader Prof. Dr. Wolfgang Uckert Christoph Kemna* (Guest, Humboldt Universität) Maria Lietz* explore in pre-clinical models. In a melanoma Peter Meyerhuber* formouse a somatic model wemutation analyzed natively the efficacy expressed of ATT in Scientist anusing established TCR-engineered tumor. TAlthough cells with the specificity targeted Dr. Mario Bunse Technical Assistants antigen and the therapeutic TCR provided ex- Dr. Ellen Hilgenberg* Carolin Genehr cellent conditions for ATT, therapy was foiled Dr. Elisa Kieback* Janina Hauchwitz by escape of antigen-negative tumor variants. Dr. Matthias Leisegang* Christin Hesse Our analysis showed that successful muta- Dr. Felix Lorenz Kordelia Hummel Michaela Naschke levels of the targeted neoantigen or ATT has Graduate and Undergraduate Christina Schulz* totion-specific be combined TCR with gene radiotherapy therapy requires to increase high Students antigen levels. Julian Clauß Secretary - Krystyna Dudaniec* Erika von Bühler (SFB TR36) nogenic mutations in cyclin-dependent kinase 4In (CDK4) another that model naturally we analyzed occur indifferent human immu mela- Julian Gabrysch* * part of the time reported noma. We used a syngeneic HLA-A2-trans- Inan Edes

MDC Research Report 2016 127 CANCER RESEARCH Photo: David Ausserhofer/MDC Photo:

Jana Wolf

Mathematical Modelling of Cellular Processes

Complex diseases are often characterized by non-canonical branch. Both signaling branch- an accumulation of multiple perturbations, such es can be activated by a range of stimuli and lead to the activation of different members of as mutations or over-expression of proteins, - in rather large and complex cellular networks. estingly, both branches act on different time The consequences of these perturbations can scalesthe NF-κB but are transcription linked via shared factor pathway family. Inter com- hardly be analyzed by pure reasoning. Here, ponents and target genes. We are interested in the regulation of the overall system and the mathematical modeling and computational cross-talk between the two branches. analysis contribute to a deeper understanding Homeostasis and activation of the overall NF- of the regulatory systems and provide a better basis for the interpretation of high-throughput on the processing of the precursor proteins p100κB/IKK and signaling p105. We pathway used a mass-spectrome critically depend- data and identification of effective drug targets. try based quantitative analysis and modeling approach to dissect the processes of precur- Our group develops and analyses mathemati- sor processing (cooperation with the groups cal models of mammalian signaling pathways of Claus Scheidereit and Gunnar Dittmar, now and gene-regulatory networks in normal and - disease states. For our investigations we use gations showed the dependence of canonical tools such as simulations, bifurcation analyses precursorLuxembourg p105 Institute levels onof Health).non-canonical Our investi stim- and sensitivity analyses. These give insights ulation, establishing a new link between the into the dynamical properties of systems and two signaling branches. help to identify most sensitive processes and We developed a core model of the canonical critical regulations. Another important as- core model reproduces the dynamical and differences in signaling and gene-regulatory NF-κB/IKK signaling branch (Fig. 1). This networkspect is the since investigation these are critically of cell-type involved specific in the reduced model complexity a bifurcation analysisstationary could behavior be performed, of NF-κB very demonstrating well. Due to side-effects of drugs. Mainthe prediction projects inof thethe lastefficiency years haveand possiblefocused on signal transduction, in particular on the which intracellular parameters can influence ofthe dynamics the NF-κB in response dynamics. to Wethe showedsame stimu that- lus.NF-κB Of isparticular capable interestof exhibiting are oscillatory distinct types and projectsNF-κB/IKK we pathwayclosely collaborate and the Wnt/β-catenin with experi- non-oscillatory modes of behavior (Fig. 2). mentalpathway, partners and gene in- expression. and outside In of many the MDC. of our

Modeling the NF-κB/IKK pathway We identified the total NF-κB concentration dynamicsand the IκBα (Fig. transcription 2). rate as two critical Claus Scheidereit at the MDC we aim for a parameters that modulate the type of NF-κB In a collaborative effort with the group of Investigation of oscillating systems - - plicatedsystems levelin several understanding human diseases, of the IKK/NF-such as nalling, occur in a wide variety of cellular κB signaling pathway. This pathway is im processes,Oscillatory e.g.dynamics, in calcium as found and p53in NF-κB signaling sig responses, in metabolic pathways or within inflammatory and autoimmune diseases as well as cancer. It consists of a canonical and a 128 MDC Research Report 2016 CANCER RESEARCH

Core model of the signal transduction in the canonical NF B pathway.

κ gene-regulatory networks, such as the circa- Analysis of the possible modes of dynamical behaviour of dian system. Since it is of central importance the NF- B core model shown in Fig. 1. (A) The bifurcation analysis shows the ranges of the I Bα transcription rate for on the behaviour of these systems a number steady statesκ and oscillations. Examples of the dynamics ofto studiesunderstand have the been influence undertaken of perturbations to examine are shown in (B) for specific values:κ i) resulting in strongly their robustness. The period of circadian oscil- damped behaviour, ii) showing sustained oscillations and lations seems to be very robust to provide reli- iii) weakly damped oscillations. (C) Dynamics for the range of I Bα transcription rate values. oscillations is on the other hand very sensitive toable allow timing; for thefrequency-encoded period of intracellular transduction calcium teinκ RC3H1 on the system. Model simulations

(EMBL) and Ralf Steuer (Humboldt University system is affected by post-transcriptional reg- Berlin)of signals. we In examined cooperation the withrobustness Antonio of Politi the ulation,and experiments opening new show possibilities that the NF-κB/IKKof manipu- oscillatory period towards perturbations of - kinetic parameters in diverse oscillatory systems and how design principles, such as type ubiquitinlating NF-κB/IKK ligase F-box-protein, system. An important which is regu dif- of feedback and formalism of reaction kinet- ferentiallylator of the expressed Wnt/β-catenin in wild-type pathway and is cancerthe E3 cells. We investigated the consequences of the

forics, theimpact characteristics these findings. of oscillatory Our work systems.highlights the importance of specific design principles casecomplex of oncogenic regulation pathway by β-TrCP mutations. and discussed Analysis of target gene expression the suitability of β-TrCP as drug targets in the - lates many physiological processes by control- Selected Publications lingThe the Wnt/β-catenin differential expression signaling of pathway target genes. regu F. Lorenzin, U. Benary, A. Baluapuri, S. Walz, L. A. Jung, B. von Eyß, C. Kisker, J. Wolf, M. The deregulation of the pathway is associated with various types of cancer. We developed dependent gene regulation, eLife. 5, e15161. Eilers & E. Wolf (2016), Different promoter affinities account for specificity in MYC- and reaction kinetics impact the robustness of cellular oscillations. Plos Comp Biol. pathway that include regulatory processes K. Baum, A.Z. Politi, B. Kofahl, R. Steuer & J. Wolf (2016), Feedback, mass conservation ofmathematical target gene modelsexpression of theunder Wnt/β-catenin normal and oncogenic conditions. The target gene MYC is 12(12), e1005298. a well-known proto-oncogenic transcription Murakawa Y.; Hinz M.; Mothes J.; Schuetz A.; Uhl M.; Wyler E.; Yasuda T.; Mastrobuoni - G.; Friedel C.C.; Doelken L.; Kempa S.; Schmidt-Supprian M.; Bluethgen N.; Backofen factor that regulates the expression of hun- R.;Nature Heinemann Communications U.; Wolf J.; 6: Scheidereit 7367. C.; Landthaler M. (2015), RC3H1 post-transcrip tionally regulates A20 mRNA and modulates the activity of the IKK/NF-{κ}B pathway. signal transduction: a mechanistic model. BioEssays 37 (4): 452-462. Eilersdreds ofat genesBiocenter influencing University cell of fate Würzburg and driving we Mothes J.; Busse D.; Kofahl B.; Wolf J. (2015), Sources of dynamic variability in NF-{κ}B cell proliferation. In collaboration with Martin Yilmazp100-p105 Z.B.; moduleKofahl B.; reveals Beaudette interdependent P.; Baum K.; precursor Ipenberg proteolysis,I.; Weih F.; Wolf* Cell Reports J.; Dittmar* 9 (5): expression patterns can be explained by dis- G.;1756-1769. Scheidereit* (* co-senior C. (2014), author) Quantitative dissection and modeling of the NF-{κ}B found evidence that cancer type specific gene to the promoters of target genes. Sometinct MYC target levels genes and of different signaling affinities pathways of MYC are Group Leader PhD student themselves regulators of the pathway estab- Dr. Jana Wolf Fabian Konrath lishing interlocked negative and positive feed- - Scientists Master student Dr. Katharina Baum Mareike Simon andback A20 loops. that Important both negatively regulators regulate of the the ca Dr. Uwe Benary pathway.nonical NF-κB/IKK Together signalingwith the branchgroups areof ClausIκBα Dr. Dorothea Busse Secretariat Dr. Bente Kofahl Sabine Froese MDC) we studied the effect of post-transcrip- Dr. Janina Mothes Petra Haink tionalScheidereit regulation and Markus by the LandthalermRNA binding (BIMSB/ pro-

MDC Research Report 2016 129 130 MDC Research Report 2016 Diseases of the Nervous System

Coordinator: Carmen Birchmeier-Kohler

MDC Research Report 2016 131 Diseases of the Nervous System

Coordinator: Carmen Birchmeier-Kohler

he nervous system is fundamental to developed new protocols to produce spinal - T sense the environment, to move and gione joined the program as Delbrück Fel- react tothe stimuli, body’s and ability to generate to maintain and itself, control to low.neuron Alessandro types. In is addition,a specialist Alessandro on embryonic Pri behavior. Hence, disorders of the nervous stem cells and induced pluripotent stem system that manifest themselves as neuro- cells, which he uses to model mitochondrial logical and psychiatric disease constitute DNA disorders. The success of the Neuro- a major challenge for affected individuals, their families, and for society. Determining that young researchers received attractive the molecular basis of normal nervous sys- offersscience for Program positions is also from reflected outside in and the factleft tem function, and discovering the changes the MDC. Jochen Meier became professor responsible for inherited or acquired de- - schweig where he moved in 2015. Kathrine against nervous system disease. Pooleat the Instituthas an offerfür Zoologie to return of theto herTU Braunnative fects are important priorities in the fight Australia as a Senior Lecturer at the Univer- Research in the Neuroscience Department sity of New South Wales in Sidney, and will of the MDC focuses on molecular and cel- move in March 2016. Björn Schroeder will lular analysis of the central and peripheral take a position in industry. nervous system. Themes and expertise covered by the department are broad and interdisciplinary, and comprise genetics, by 2013 and aims to support research in neurophysiology, proteomics and system systemsThe Berlin medicine Institute at of the Health Charité was and founded MDC. biology, biochemistry, cell biology and stem cell research. This provides many opportu- the use of high throughput genomic tech- nities for interactions and collaborations, nologiesIt provides for infrastructure translational andresearch. support Cur for- and ensures an optimal environment for rently, seven group leaders in the Neuro- young researchers and group leaders. We science Program participate in cooperative are delighted that one of them, James Pou- let, has now received his second ERC grant - in 2015, a consolidator grant in addition to networksers: E. Wanker funded andby the F. BIH.Heppner/Charité) For instance, a the starting grant in 2010. He was awarded consortiumuses system on biology Alzheimer’s approaches disease to (Speak model tenure in 2016 and will continue his work at the MDC. new strategies for dementia treatment. A networkproteostasis on congenital networks disease and aims (Speakers: to find We were able to recruit young scientists C. Rosenmund/Charité and C. Birchmei- to the department during the report period. Niccolo Zampieri joined the Neurosci- cause congenital disorders by analyzing ence Department in 2014. He trained as a er) aims to define novel mechanisms that postdoc in the laboratory of Tom Jessell at addition, smaller research networks were Columbia University in New York, where regulatory i.e. non-coding sequences. In- he explored the functions of cadherins in tration in glioblastoma (H. Kettenmann in the nervous systemused and used rabies collaborationestablished, one with to C. analyze Harms/Charité) immune infiland virus to map sensory and motor circuits. a second that investigates core mechanisms Mina Gouti joined the MDC in the beginning in the pathophysiology of psychotic disor- of 2016. She trained in the laboratory of ders (J. Poulet and S. Jacob/Charité). These - - don and uses mouse embryonic stem cells tional research and collaborations between toJames generate Briscoe neurons at the Crickand muscle Institute cells. in Lon She BIH activities have strengthened transla

Charité and MDC researchers. In addition, 132 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

an initiative to facilitate translation of aca- ment. The group of Helmut Kettenmann demic discoveries into therapy (SPARK - rophages of the brain, in glioma progres- defined the function of microglia, the mac- forBERLIN) translational was recently projects. established Two projects at theby late microglia activity (reviewed in Nat. BIH that provides mentoring and funding Neurosciencesion and identified 2015). molecules The group that of Carmen modu Birchmeier systematically analyzed the Gary Lewin were selected in the first SPARK network that controls the gene expression ScientistsBERLIN call. of the Neuroscience Program par- program in mature pancreatic beta cells ticipate in research and graduate schools, and showed that combinatorial binding of - - rod1 and Foxa2) maintains the mature gene ternationalamong them Medical the Helmholtz Neuroscience International program, expressionthree transcription program (in factors collaboration (Insm1, Neuwith theResearch MyoGrad School school MolNeuro, and the the newly Charité estab In- the groups of N. Rajewsky and W. Chen, Jia lished Einstein-Center for Neuroscience. et al., Embo J 2015). Huntingtin protein The curriculum of these graduate schools with expanded polyglutamine tracts causes encompasses lectures, journal clubs and courses. Furthermore, the schools organize a pathological hallmark of the disorder. The retreats and symposia for graduate stu- Huntington’s disease, and its aggregates are- tering approach to identify novel hunting- for instance courses on communication and tingroup interaction of Erich partners.Wanker used One aof network those pro fil- presentationdents, and they skills. provide ‘soft skills’ training, teins, CRMP1, was found to be a suppressor of huntingtin misfolding and neurotoxicity (Stroedicke et al., Genome Research 2015). Scientific highlights of the program

During the report period, several important

Neuroscience Program. Regulation of cell volumefindings is were critical published for many by memebrscellular and of theor- ganismal functions but the molecular identity of the volume-regulated anion channel VRAC had remained unknown. Thomas

as indispensable component of VRAC (Voss etJentsch al., Science and his 2014), group and identified their subsequent LRRC8A work characterized the subunit composition of this channel (Planells-Cases et al., Embo J 2015). The groups of Gary Lewin and James Poulet were able to identify a somatosensory circuit in the cortex, which is required for cooling perception in mice (Milenkovic et al., Nat. Neuroscience 2014). Gliomas are complex tumors composed of neoplastic and non-neoplastic cells, which each individually contribute to cancer formation, progression and response to treat-

MDC Research Report 2016 133 DISEASES OF THE NERVOUS SYSTEM Photo: Thomas Müller Photo:

Carmen Birchmeier-Kohler

Developmental Biology / Signal Transduction

Our lab focuses on functional analysis of Craniofacial and trunk skeletal muscles are genes important in development and stem evolutionarily distinct and derive from cranial and somitic mesoderm, respectively. Dif- cells, concentrating on muscle, the nervous ferent regulatory hierarchies act upstream of system and endocrine organs. We observe myogenic regulatory factors in cranial and that many genes control both, developmen- somitic mesoderm, but the same core regu- tal and regenerative processes in the adult. latory network - MyoD, Myf5 and Mrf4 - ex- ecutes the differentiation program. We com- For our work, we use mouse genetics and pared the role of Notch signals in head and stem cells culture, and define molecular trunk myogenesis, and observed that the ma- mechanisms by genomic and proteomic jor Notch function, the suppression of MyoD and premature myogenic differentiation, is technologies. conserved during muscle development in the trunk and head.

Muscle stem cells and their niche D. Bröhl, I. Lahmann, M. T. Czajkowski (former member of the laboratory), J. Griger, T. Zyrianova, in collaboration with H.P. Rahn (Preparative Flow Cytometry Facility, MDC) and B. Purfürst (EM Facility, MDC).

Skeletal muscle growth and regeneration rely on myogenic progenitor cells. Satellite cells are progenitor/stem cells of the postnatal muscle and reside between basal lam-

This anatomical position is called the stem cellina niche. and plasma membrane of myofibers. We have previously shown that generation and maintenance of satellite cells depends on Notch signals. For instance, elimination of Rbpj, the transcriptional mediator of Notch signals, results in premature differ- Figure 1: Notch signals suppress prema- entiation of myogenic progenitors and the ture differentiation of muscle progenitor formation of tiny muscle groups that lack cells in cranial muscle satellite cells (Vasyutina et al., PNAS 2007). Section of a head of a mouse which shows skel- This drastic effect of a loss of Notch signals etal muscle (myosin, red), myogenic progenitor is rescued by the additional mutation of the cells (Pax7, green) and nuclei (DAPI, blue). (A) muscle differentiation factor MyoD. Thus, a Dll1 mutant mice show small muscle groups major role of the Notch pathway is to sup- that no longer contain progenitor cells. (B) In press MyoD in myogenic progenitors (Bröhl the Dll1/MyoD double mutant, muscle groups are larger and full of progenitor cells.

et al., 2012; Czajkowski et al., 2014). 134 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 2: MAPK activity suffices to replace Shp2 signaling during peripheral myelination (A) Analysis of nerves from control and mutant mice by EM. When Shp2 is mutated (Dhhcre;Shp2flox/flox) in Schwann cells, Schwann cells are present and wrap the nerves but do not form myelin. Myelination occurs when MAPK is activated in mutant Schwann cells (Dhhcre;Shp2flox/flox; Mek1DD). (B) Quantification of myelin thickness (G-ratio).

Notch signaling has a second role and is need- nals provided by Nrg1 (an EGF-like growth ed for colonization of the satellite cell niche in factor). This signal is mediated in Schwann cells by ErbB tyrosine kinase receptors and of double mutants (mutant for Notch signal- the tyrosine phosphatase Shp2. ingthe trunk.components In particular, like Rbpj myogenic or Dll1 progenitorsand MyoD) Developmental myelination depends on do not take up a satellite cell position under the lamina, and locate instead in the inter- sensory and motoneurons and presented in stitial space where they poorly contribute type III Nrg1 that is produced by peripheral that arise during development due to a lack - orthe dampenedaxonal membrane. Nrg1/ErbB2 Myelination signals deficitscannot lularto muscle matrix growth. around The satellite colonization cells. deficitSatellite is be compensated at later time points. Nrg1 cellscaused do by thus insufficient contribute generation to the formation of extracel of is also required for effective nerve repair in adulthood and acts during an early phase cranial muscle progenitors do not require of remyelination. Remyelination and nerve Notchtheir own signals niche to in colonize trunk muscle. their niche. In contrast, Thus, repair are severely delayed in the absence the major Notch function - suppression of of Nrg1 but ultimately occur even when premature myogenesis - is conserved in head the signal is lacking. Thus, myelination and and trunk muscle progenitor cells, but other remyelination are controlled by similar sig- functions like satellite cell homing diverge in nals, but differences exist (cf. Birchmeier cranial- and somite-derived muscle (Bröhl et and Bennett, 2016). Erk/MAPK transmit signals provided by tyrosine kinase receptors, but other path- Neuregulin/ErbBal., 2012; Czajkowski et signaling al., 2014). in de- - velopmental myelin formation and gamma also act downstream. All of these nerve repair pathwaysways like are PI3kinase/Akt, known to be Cdc42 activated or PLC by M. Sheean (former member of the laborato- Nrg1/ErbB and were suggested to control ry), C. Cheret, T. Müller, in collaboration with - E. McShane and M. Selbach (Cell Signaling/ dent myelination program like myelin gene Mass Spectrometry, MDC), J. Walcher and expressionspecific aspects or morphogenic of the Nrg1/ErbB-depen movements G. R. Lewin (Molecular Physiology, MDC), S. during myelination. Our previous work us- Hoelper and M. Krüger (MPI Bad Nauheim), ing mouse genetics demonstrated that loss A. N. Garratt (Charité, Berlin), D. Meijer of ErbB3/Shp2 severely impairs Schwann (Erasmus University, Rotterdam), K. Rajew- cell development and disrupts the myelina- sky and W. Birchmeier (Cancer Program, tion program. We recently found that acti- MDC), D. Bennett and F. Fricker (University vated MAPK compensates for the absence of Oxford, UK). of ErbB3 or Shp2 during Schwann cell development and myelination (Sheean et al., Myelin is essential for rapid and accurate - conduction of electrical impulses by axons place Nrg1/ErbB singaling during myelina- in the central and peripheral nervous sys- tion.2014). Thus, MAPK activity suffices to re tem. Myelin is formed in the early postnatal period and myelin formation in the peripheral nervous system depends on axonal sig-

MDC Research Report 2016 135 DISEASES OF THE NERVOUS SYSTEM

Figure 3: Insm1 maintains mature pancreatic beta-cell function (A) Western blot analysis after introduction of a Insm1 mutation in mature beta-cells; low Insm1 protein in mutant (RIP-creERT2;Insm1flox/flox) compared to control cells (Insm1flox/flox or RIP- creERT2;Insm1flox/+). (B) Conditional mutant mice showed a severe deficit in glucose-stimulated insulin secretion. Animals received a glucose injection that results in insulin release in control but not in mutant mice.

Figure 4: Insm1 cooperates with Neurod1 and Foxa2 to control gene expression Insm1, Neurod1 and Foxa2 binding sites in Insm1 cooperates with Neurod1 chromatin from a beta-cell line were defined and Foxa2 to maintain mature by ChIP-seq and bioinformatics. (A) Binding pancreatic β-cell function sites on the glucose transporter Glut2 (Slc2a2) S. Jia, T. Müller, D. Blasevic (former member gene that is needed for glucose-stimulated of the laboratory), in collaboration with A. insulin secretion. (B) Analysis of all Insm1 Ivanov and N. Rajewsky (Systems Biology, binding sites in the genome demonstrated MDC), W. Sun and W. Chen (Scientific Ge- large degrees of overlap with Neurod1 and nomics Platform, MDC), and M. Poy (Molecu- Foxa2 binding sites. lar Mechanisms of Metabolic Disease, MDC).

Key transcription factors control the gene expression program in mature pancreatic beta-cells, but their integration into regula-

tory networks is little understood. Insm1 encodes a zinc finger factor and we found 136 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

cells and other endocrine cell types in the Selected Publications pancreas,that Insm1 intestine, controls pituitary,differentiation lung andof beta- ad- renal medulla. We recently showed that Insm1 cooperates with Neurod1 and Foxa2 to maintain mature pancreatic β-cell and together bind regulatory sequences in Divergentfunction. Jia and S, Ivanovconserved A, Blasevic roles of D,Dll1 Müller signaling T, Purfürst in development B, Sun W, Chenof craniofacial W, Poy MN, and Rajewsky N, Birchmeier C. EMBO J. 2015 May 12;34(10):1417-33. theInsm1, genome Neurod1 of mature and Foxa2 pancreatic directly beta-cells. interact trunk muscle. Czajkowski MT, Rassek C, Lenhard DC, Bröhl D, Birchmeier C. Dev Biol. Colonization of the satellite cell niche by skeletal muscle progenitor cells depends on2014 Notch Nov signals. 15;395(2):307-16. Bröhl D, Vasyutina E, Czajkowski MT, Griger J, Rassek C, Rahn Insm1 ablation in mature beta-cells in mice dependentresulted in genes pronounced in developing deficits and in insulin adult 10.1016/j.devcel.2012.07.014. ActivationHP, Purfürst of B, MAPK Wende overrides H, Birchmeier the termination C. Dev Cell. of 2012myelin Sep growth 11;23(3):469-81. and replaces doi: Nrg1/ secretion and gene expression. Insm1-- ErbB3 signals during Schwann cell development and myelination. Sheean ME, lar, when the mutation was introduced in McShane E, Cheret C, Walcher J, Müller T, Wulf-Goldenberg A, Hoelper S, Garratt AN, adult,beta-cells mature markedly beta-cells differed. assumed In an particu imma- Krüger M, Rajewsky K, Meijer D, Birchmeier W, Lewin GR, Selbach M, Birchmeier C. ture beta-cell character in respect to gene Neuregulin/ErbB signaling in developmental myelin formation and nerve repair. expression and function. BirchmeierGenes Dev. 2014C and Feb Bennett. 1;28(3):290-303. DLH Current Topics in Developmental Biology (CTDB). 2016, Birchmeier, C. and D. L. Bennett. “Neuregulin/ErbB Signaling in Developmen- and discovered that many binding sites are We defined Insm1 binding sites in beta-cells, tal Myelin Formation and Nerve Repair.” Curr Top Dev Biol 116: 45-64 (2016). and Foxa2. Binding sites co-occupied by Group Leader cooperatively bound by Insm1, Neurod1- Prof. Dr. Carmen Simon Gonzalez-Hirschfeld cated in intergenic and intronic sequences Birchmeier-Kohler Madeleine Larrosa Insm1/Neurod1/Foxa2 were mainly lo data indicated that many correspond to Senior Scientist Graduates and Undergraduates enhancerand comparison sequences. with Furthermore, histone modification sites co- Dr. Thomas Müller Rozina Aktar occupied by all three factors represent high Lydia Sippach - Scientists Dr. Dominique Bröhl Technical Assistants affinity binding sites. Remarkably, combina Dr. Cyril Cheret Bettina Brandt correlatedtorial binding and sites explained of Insm1, gene Neurod1 expression and Dr. Joscha Griger Sven Buchert Foxa2 but not binding sites of Insm1 alone Dr. Luis Rodrigo Hernandez Carola Griffel cells. Moreover, human genomic sequences Miranda Vivian Schulz correspondingchanges observed to murine in Insm1 sites mutant co-occupied beta- Dr. Shiqi Jia Mandy Terne

in single nucleotide polymorphisms associ- Dr. Pierre-Louis Ruffault Secretariat atedby Insm1/Neurod1/Foxa2 with glycolytic traits. Thus, were combinato enriched- Dr. TatianaInes Lahmann Zyrianova Dr. Timkehet Teffera - PhD and MD students Animal Attendants maturerial Insm1/Neurod1/Foxa2 gene expression program binding in identi beta- Charlotte Chaimowicz Claudia Päseler cells,fies regulatory and disruption sequences of this that network maintain results the Jie-Shin Chen Petra Stallerow in functional failure.

MDC Research Report 2016 137 DISEASES OF THE NERVOUS SYSTEM Photo: privat Photo:

Mina Gouti

Research Focus

We use human and mouse pluripotent stem in the mouse embryo suggested that a cells (PSCs) to model embryo development common bipotent progenitor, that can give descendants to both the spinal cord and in vitro and understand the mechanisms paraxial mesoderm, exists in vivo. These that regulate cell fate decisions during neu- cells, called neuromesodermal progenitors romuscular system development. (NMPs), reside in the caudal lateral epiblast region of the embryo during embryonic development. NMP cells are important for During embryonic development spinal cord axis elongation and correct tissue growth motor neurons are generated with high but have been largely ignored in the stem precision along the anterior – posterior (AP) axis and establish connections with embryonic development we have recently skeletal muscles to control movement. Pre- succeededcell field. Followingin generating the NMP cues fromcells in mouse vitro vious studies have shown that development from mouse and human pluripotent stem and survival of motor neurons and muscles cells (Figure 1). The in vitro generation of depends on each other but until recently these cells opens up new opportunities for their generation were considered as inde- the study and treatment of neuromuscular pendent events. Striking evidence coming diseases as it gives unprecedented access to from clonal lineage analysis experiments the simultaneous development of both neu-

Our focus is to understand how these two tissuesral and aremesodermal generated cell and types interact in the in “dish”.space and time during human development. This will allow us to unravel the mechanisms of human embryo development and evaluate how defects in early development of these tissues may predispose to disease in adult life. To address these questions we are using gain and loss of gene function ap-

sequencing technologies (single cell RNA- seq)proaches as well (CRISPR-Cas9), as live cell imaging next techniques. generation

Figure 1: Progressive differentiation of NMPs during develop- Neuromesodermal Progenitor ment of the Central Nervous System. (A) Schematic of a wild type Cells (NMPs) embryo, looking down onto the surface of the posterior end of the em- During normal embryonic development bryo. The posterior gradients of Wnt and Fgf signals (red), emanating part of the posterior spinal cord, muscles from the caudal lateral epiblast (CLE), oppose the activity of retinoic and bones are generated from a neurome- acid (RA, green), which is produced by the developing somites. Neu- sodermal progenitor population (NMP). romesodermal progenitor (NMP) cells (orange dots) located in the CLE NMPs are a transient population, arising transit either through a preneural tube (PNT) stage (blue dots) to spi- early in gastrulation, which can be identi- nal cord progenitors (green dots), or through a presomitic mesoderm in vivo by the co-expression of the tran- stage (PSM) to form the somites (red cells). Green and red arrows show scription factors Sox2 and Brachyury (T/ the distinct developmental paths of NMP cells. (B) Timely activation of Bra)fied . During development, FGF and Wnt Wnt/Fgf signaling in mouse and human differentiating ES cells results signalling emanating from the posterior in the generation of NMP cells expressing Brachyury (T/Bra) and Sox2 part of the embryo are instrumental for the protein, equivalent to those found in the CLE region of the embryo. induction and maintenance of NMP cells

138 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 2: Single cell analysis reveals the differentiation route of NMPs towards the neural and mesodermal lineage. Single-cell RNA- sequencing analysis of NMP cells isolated from mouse embryos (A) and in vitro NMP cells (B) defines their molecular signature and resemblance. (C) Hierarchical clustering of single NMP cells reveals four distinct clusters. One associated with NMP identity (cerulean), a transitioning NMP cluster (light green), a neural progenitor (green) and a mesodermal cluster (red). (D) Pseudoteporal ordering of single cells depicts the two different developmental trajectories towards the neural and mesodermal lineages. (E) Analysis of single cell RNA-sequencing data predicts a gene regulatory network that balances the generation of neural and mesodermal tissue to facilitate orderly axis extension during embryonic development.

(Figure1). Using the same signals, we have recently succeeded in generating NMP cells in vitro from mouse and human embryonic Figure 3: NMPs as a tool for disease modeling and drug screen- et al., 2014). Single cell RNA-sequencing ing. Patient derived somatic cells can be reprogrammed to generate analysisstem cells revealed (ESCs) (Goutithat in et vitroal., 2015; generated Gouti induced pluripotent stem cells (iPS cells) carrying the disease related NMPs closely resemble the in vivo counter- phenotype. Directed differentiation of iPS cells through an NMP inter- parts (Gouti, 2017). Furthermore, in vitro mediate stage results in the generation of neurons (Tuj1+) and muscle NMP cells can then be coaxed into either cells (Desmin+). Neuromuscular cultures generated from patient spinal cord neurons and / or paraxial me- specific iPS cells can be used for disease modeling and high-throughput drug screening. at the correct time windows. We are using thissoderm system cells to by gain exposure a better to specificunderstanding signals of how NMP cells decide between the spinal cord or paraxial mesodermal lineage (Fig- ure 2). Overall, our results illustrate the importance of mimicking normal embryonic

development for the efficient generation of specificModeling cell typesHuman from Neuromuscular PSCs. Selected Publications Diseases Using Patient iPS Cells. Gouti, M., Delile, J., Stamataki, D., Wymeersch, F. J., Huang, Y., Kleinjung, J., Wilson, V., The ability to culture induced pluripotent - stem (iPS) cells and edit their genome of- cation during vertebrate trunk development. Developmental Cell In Press. fers advantages for regenerative medicine Gouti,Briscoe, M., J. Metzis,(2017). V., A andgene Briscoe, regulatory J. (2015). network The balances route to neural spinal andcord mesoderm cell types: specifia tale of in vi- signals and switches. Trends Genet 31 Gouti, M., Tsakiridis, A., Wymeersch, F.J., Huang, Y., Kleinjung, J., Wilson, V., and Briscoe, tro system, it provides a unique opportunity , 282-289. toapplications. understand In the conjunction pathology with of neuromus our - cular diseases. The last years, remarkable J.PLoS (2014). Biol In12 vitro, e1001937. generation of neuromesodermal progenitors reveals distinct roles progress has been made in the generation for wnt signalling in the specification of spinal cord and paraxial mesoderm identity. of spinal cord motor neurons (MNs) from human PSCs. However, the ability of these Start of the Group: April 2016 Technical Assistants MNs to connect and control human skel- Loreen-Claudine Kabuss etal muscle remains uncertain. We use the Head of the Group Johanna Schmid NMP differentiation system to generate in Dr. Mina Gouti parallel nerve and muscle cells in the dish Graduate and Undergraduate and understand the origins of disease and Postgraduates Students the contribution of either lineage (neural – Marie-Sophie Bittner Filip Angelov muscle) to its pathology (Figure 3). Apart Jorge Miguel Faustino Martins from implications in disease modeling and drug screening, our novel culture system SecretariatIris Caroline Hube enables new promises for regenerative Beate Valeske medicine.

MDC Research Report 2016 139 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

Annette Hammes

Molecular Pathways in Cortical Development

We study the regulation of morphogen activ- progenitor cells from mitogenic stimuli. Loss ity during neural development and in the adult of LRP2 in mice leads to ectopic proliferation in the CMZ progenitor niche (Christ et al., cortical stem cell niche. Deficits in morphogen signaling lead to severe malformations of LRP2 to act in a context dependent manner the central nervous system (CNS), including as2015). activator Our findingsor inhibitor document of the SHH the abilitypathway. of holoprosencephaly and neural tube closure Overall the data substantiate the emerging concept that auxiliary receptors are critical defects. Another current project in our lab is modulators of morphogen delivery and signal addressing the question whether migratory reception in target tissues. neuronal precursors need mechanical cues for proper differentiation and pathfinding. New insights into receptor mediated uptake of folic acid during We hypothesize that besides the well estab- neurulation lished chemical signaling, migrating neurons Dr. Nora Mecklenburg, Dr. Esther Kur need mechanosensitive protein complexes to transduce motion and interaction with their closure of the cranial neural tube. This defect environment into electrical signals. isLRP2 unrelated deficient to embryos the etiology suffer of from holoprosen impaired- cephaly and hinted at an additional role for LRP2 in the dorsal domain of the developing Sophisticated regulation of mor- brain. Neural tube closure defects (NTDs) phogen activity in the embryonic have been attributed to impaired folic acid

and adult brain (vitamin B9) metabolism in mice and humans. in collaboration with AG Prof. Thomas We therefore asked whether LRP2 might be Willnow required for delivery of folic acid to neuroepithelial cells during neurulation. Our results LRP2 (low density lipoprotein receptor re- showed that LRP2 mediates endocytic uptake lated protein 2) is a multifunctional endocytic of folic acid and its binding protein FOLR1 (fo- - late receptor 1), a crucial step for proper neu- lar mechanism underlying the holoprosen- cephalyreceptor. (HPE) We previously phenotype clarified of mouse the embryos molecu thatral tube LRP2 closure teams (Figure up with 1; Kurother et al.,receptors 2014). morphogen signaling as a SHH co-receptor formingThese findings co-receptor substantiate complexes our to hypothesisensure not todeficient patched for 1 in this the receptor. ventral forebrain LRP2 promotes neuro- epithelium (Christ et al., 2012). The recep- manner but also to ensure precisely timed tor is also important for neuronal progenitor andonly titratedligand specificity uptake of in signaling a tissue dependentmolecules. proliferation and neurogenic output in the Moreover, to better understand the molecu- adult cortical stem cell niche by maintaining lar mechanisms of LRP2 dependent NTDs, we balanced morphogen activity (Gajera et al., will further investigate whether other factors - such as altered morphogen signaling or dis- ary marginal zone (CMZ), LRP2 antagonizes turbed cilia function in addition to impaired SHH2010). activity In the byretinal mediating stem cell endocytic niche, the clear cili- folic acid uptake modulate the NTD pheno- ance of the morphogen to protect quiescent

type in LRP2 deficient mouse embryos. 140 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 1. Immunohistological detection of sFOLR1-A488 and folic acid-Cy3 on coronal head sections of E8.5 whole-embryo cultures. Uptake of these added ligands was only seen in wild-type but not in Lrp2-/- rostral neural folds. The model depicts LRP2 dependent endocytic uptake of folic acid into neuroepithelial cells. Soluble FOLR1 and GPI-anchored FOLR1 bind folic acid. Internalization of these complexes relies on the interaction with the co-receptor LRP2.

Understanding the genetic causes The role of mechanotransduction of neural tube defects (NTDs) in the motility and migration of Fabian Paul, Deborah Kohler, Dr. Nora neuronal precursor cells Mecklenburg Carina Fürst in collaboration with AG Prof Gary Lewin Identifying novel pathways contributing to NTDs Cell migration is a fundamental process in - tifactorial etiology of human CNS anomalies by chemical signals. However there are still sinceLRP2 deficientthe severity mice of are the ideal Lrp2 to-/- study NTD the pheno mul- manythe developing open questions brain and concerning strongly the influenced mecha- type varies strongly depending on the genetic nisms underlying neuronal migration. We background. This suggests a strong impact of hypothesize that neuronal precursors use mechanotransduction to probe their physi- We generated RNAseq data from Lrp2 mutant cal environment during migration. First re- embryosmouse strain on C57BL/6N associated background genetic modifiers. suffering cordings of mechanically gated currents in from severe CNS anomalies and from mu- neuroepithelial cells provide evidence that tants on FVB/N background with a mild phe- neuronal precursors are indeed mechanosen- notype. The aim is to identify differentially sitive. We will further characterize mechanical signaling in neuronal development using modulating the severity of NTDs. The genetic mouse genetics, electrophysiology, and high- regulated pathways and new modifier genes resolution imaging. used to evaluate analogous gene-gene inter- actionsmodifiers in identifiedNTD patient in our data studies in collaboration will be also with Dr. Angela Kaindl and Dr. Gregory Wul- czyn (Charité, Berlin). The goal is to improve genetic risk assessment and to gain deeper insights in the pathogenesis of human NTDs. Selected Publications

Eccles RL, Czajkowski MT, Barth C, Müller PM, McShane E, Grunwald S, Beaudette P, The role of LRPs in WNT-signaling Mecklenburg N, Volkmer R, Zühlke K, Dittmar G, Selbach M, Hammes A, Daumke O, dependent development of the CNS Klussmann E, Urbé S, Rocks O. Bimodal antagonism of PKA signaling by ARHGAP36. Another LRP family member, the WNT recep- Nature Commun Nasrallah R, Fast EM, Solaimani P, Knezevic K, Eliades A, Patel R, Thambyrajah R, tor LRP6, has been implicated in neural tube Unnikrishnan A,. Thoms2016 Oct J, Beck 7;7:12963. D, Vink CS, Smith A, Wong J, Shepherd M, Kent D, Roy- defects in mice and NTD patients. To test for gene-gene interactions among Lrp gene fam- - choudhurital hematopoiesis R, Paul usingF, Klippert endoglin J, Hammes regulatory A, Willnow elements T, Göttgensas molecular B, Dzierzak probes. E,Blood Zon. LI, ily members, linked to the WNT pathway, we Lacaud2016 Aug G, 23.Kouskoff V, Pimanda JE. Identification of novel regulators of developmen -/- -/- are currently analyzing Lrp4 Lrp5 and Christ A, Christa A, Eule JC, Bachmann S, Wallace V, Hammes A, Willnow TE. LRP2 Lrp4-/- Lrp6-/- compound mutant embryos. acts as SHH clearance receptor to protect the retinal margin from mitogenic stimuli. ; - Dev Cell etrant ;CNS anomalies in these compound mu- Kur E, Mecklenburg N, Cabrera RM, Willnow TE, Hammes A. LRP2 mediates folate uptake in. 2015 the developing Oct 12;35(1):36-48.2015 neural tube. J Cell Sci. tantInitial mouse data show embryos more compared severe and to highly single penmu- tants, suggesting important new functions for 2014 May 15;127(Pt 10):2261-8. LRP4 and LRP5 in the developing brain that Group Leader may be synergistic with LRP6 activity. We will Dr. Annette Hammes-Lewin test the hypothesis that the phenotypes seen (Permanent Fellow) in Lrp4 Lrp6 and Lrp4 Lrp5 compound mu- Graduate Students tants are linked to disturbed canonical WNT/ Scientists Fabian Alexander Paul ß-catenin; dependent and/or; non-canonical Dr. Nora Mecklenburg WNT planar cell polarity pathways. Carina Iris Fürst

MDC Research Report 2016 141 DISEASES OF THE NERVOUS SYSTEM Photo: Stefan Jentsch Stefan Photo:

Thomas Jentsch

Physiology and Pathology of Ion Transport

Ion transport across cellular membranes is crucial for cellular homeostasis and has integrative functions such as transepithelial transport or signal transduction. We study ion transport at various levels: from biophysical and structure-function analysis of transport proteins to their role in the cell and the organism. We have discovered several human ‘channelopathies’ and have generated and analyzed many mouse models. Until recently, we have focused on CLC Cl- channels and transporters, Anoctamin Ca2+-activated Cl- channels, KCNQ K+ channels, and KCC K+-Cl- cotransporters. Their disruption led to pathologies ranging from epilepsy, deafness, and neurodegeneration to osteopetrosis and kidney stones. We are particularly interested in the control of neuronal excitability, sensory physiology and ion homeostasis in endosomes and lysosomes. After >3 years of intense effort, we achieved in 2014 another ambitious goal and identified LRRC8 heteromers as constituting the Fig. 1: The Journal of Physiology com- volume-regulated anion channel VRAC. memorated the discovery of CLC channels This ubiquitously expressed channel with and transporters by Thomas Jentsch 25 years ago with a special issue. The cover im- many proposed functions has been known age shows the CLC chloride channel of E. coli biophysically for >30 years, but the underly- (PDB ID: 1KPK) embedded in a lipid bilayer ing proteins have remained unknown. This membrane (J Physiol, 593, 2015). breakthrough opens up an entire new field. We are now focusing much of our effort on VRAC. We will continue to investigate CLCs, but will stop working on Anoctamins, KCNQs and KCCs after successfully concluding ongoing projects.

142 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Fig. 2: Schematic diagram of VRAC’s function in regulatory volume decrease. VRAC is closed under resting conditions (top left) and opens after osmotic swelling. The resulting efflux of chloride and organic osmolytes decreases cellular osmolarity, resulting in water efflux and regulatory volume decrease.

CLC Cl- channels and transporters unc mice to investigate whether the endo- Anja Blessing, Tony Daubitz, Corinna somal ClC-3 needs, like ClC-5 and ClC-7, Göppner, Sabrina Jabs, Maja Hoegg-Beiler, exchange activity for its biological role. Carmen Ludwig, Ian Orozco, Sonali Saha, To- We showed that ClC-2 disruption leads to bias Stauber, Till Stuhlmann, Florian Ullrich, leukodystrophy in mice and others identi- Janis Vogt, Stefanie Weinert CLCN2 mutations in a form of human leukodystrophy. Human MLC-type leuko- The CLC gene family, discovered by us 25 encephalopathyfied can be caused by muta- years ago (Fig. 1), encodes plasma mem- tions in the membrane protein MLC1 and brane Cl- channels and Cl-/H+ exchangers in GlialCAM, a cell adhesion molecule that of intracellular membranes. We generated physically interacts with MLC1. GlialCAM - also binds to ClC-2 and changes its current - properties. GlialCAM directs both MLC1 larmouse CLCs models are Cl for-/H most+-exchangers, CLCs, identified performed an and ClC-2 to cell-cell contacts of transfect- cillarystructure-function β-subunits, analysis, discovered and that uncovered vesicu ed cells. We have generated Glialcam‑/‑ and several pathologies resulting from their Glialcam dn /dn mice, which carry a mutation dysfunction. Vesicular CLCs were believed found in human leukodystrophy. Compari- to be Cl- channels that facilitate vesicu- son with Mlc1‑/‑ and Clcn2‑/‑ mice showed that GlialCAM is important for the localiza- currents, but surprisingly they are Cl-/H+- tion of ClC-2 and MLC1 also in vivo. ClC-2 larexchangers. acidification Mice by in shunting which we proton converted pump currents were reduced in oligodendrocytes ClC-5 and ClC-7 to uncoupled Cl- channels of all three mouse models, suggesting that unexpectedly displayed roughly the same the lack of glial Cl- channels contributes to phenotype as the respective KOs, suggest- the pathology of all three forms of leuko- ing a role of these CLCs in vesicular Cl- ac- dystrophy. However, the loss of GlialCAM cumulation. However, some phenotypes of has additional pathogenic effects. We are Clcn7unc/unc mice were less severe than in Clcn7‑/‑ mice, indicating ei- pinpoint the cell type in which loss of ClC-2 therthe ‘uncoupled’a rescue of Cl-/H+-exchange by the isnow pathogenic. studying cell-type specific ClC-2 KOs to ClC-7unc Cl- conductance, or an ion transport-independent function of the ClC-7 pro- K+-Cl- cotransporters tein (that is present in Clcn7unc/unc, but not in Kathrin Gödde Clcn7‑/‑ mice). To distinguish between these possibilities, we created mice with a point Previously, we constitutively disrupted all mutation that totally abolishes ClC-7 ion KCl-cotransporter isoforms (KCC1-4) in transport. These mice showed that some mice. We now focused on KCC2 which low- phenotypes of Clcn7-/- mice depend on the ers the cytoplasmic Cl- concentration in absence of the ClC-7 protein rather than neurons, a prerequisite for the inhibitory the loss of ion transport. We characterized action of GABA- and glycine-receptor Cl- - channels. After having published our work tients and osteopetrotic cattle and titrated thefunctionally amount ClC-7of osteoclast-expressed mutants identified inClC-7 pa cerebellar granule and Purkinje cells, we with respect to osteopetrosis in transgenic on mice with specific KCC2 elimination in mice. We are currently analyzing Clcn3unc/ now disrupted Kcc2 specifically in mitral cells of the olfactory bulb, the first relay MDC Research Report 2016 143 DISEASES OF THE NERVOUS SYSTEM

the vomeronasal organ (VNO) that is e.g. responsible for the detection of pheromones. The signal transduction in the VNO is different from the MOE and it expresses both Ano1 and Ano2. We generated mice lacking both Ano1 and -2 in OSNs and cross them to mice lacking the upstream Trpc2 cation channel to genetically and functionally dissect the poorly understood signal transduction cascade in the VNO.

(4) KCNQ K+ channels Pawel Fidzinski, Matthias Heidenreich, Ian Orozco, Sebastian Schütze

K+ Ofthat the regulate five different neuronal isoforms excitability. of KCNQ We (Kv7) previously channels, showed KCNQ2-5 that KCNQ2 mediate and ‘M-currents’ -3 underlie a form of human epilepsy, and that KCNQ4 mutations can cause human deafness which we then analysed with mouse models. We also demonstrated a role of KCNQ4 and -5 in the vestibular organ. We now focused on the very sensitive D-hair skin

adapting mechanosensors, where we had demonstratedmechanoreceptors. a role Inof contrastKCNQ4 in to mice rapidly and Fig. 3: VRAC: dual role in drug resistance. VRAC regu- men, we found KCNQ3 in the very sensitive lates cell volume by releasing Cl- and organic osmolytes skin D-hair nerve endings where it damp- like taurine, and also mediates uptake of cancer drugs like ens excitation. Our recent work on the CNS cisplatin. Cisplatin enters cells by passive diffusion and function of KCNQ5, which combines immu- through VRACs containing subunit LRRC8D (red). It kills nohistochemistry, slice electrophysiology, cells by modifying nuclear DNA, and opens VRACs possibly in vivo registrations and behavioural stud- by generating reactive oxygen species in mitochondria, ies, indicates a role of KCNQ5 in dampening thus further stimulating cisplatin uptake and apoptotic synaptic inhibition, whereas KCNQ2 and -3 volume decrease (EMBO J, 34: 24, 2015). dampen excitation.

VRAC volume-regulated anion channels Andreia Cruz e Silva, Tony Daubitz, Deborah station in olfactory information process- Elger, Maja Hoegg-Beiler, Jennifer Lück, ing. Synaptic inhibition of mitral cells was Darius Lutter, Jonas Münch, Rosa Planells- markedly reduced and behavioural experi- Cases, Mayya Polovitskaya, Momsen Reincke, ments showed that these mice could not Tobias Stauber, Till Stuhlmann, Florian distinguish closely similar odours. These Ullrich, Felizia Voss, Pingzheng Zhou, Joanna studies are complemented with in vivo Ziomkowska electrophysiology. Swelling-activated Cl- currents mediated Anoctamin (TMEM16) Ca2+-acti by a postulated channel named VRAC (vol- vated Cl- channels ume-regulated anion channel) (a.k.a. VSOR Jonas Münch, Sebastian Albrecht

After having shown that Ano2 is the long- itsor VSOAC)properties have and first suggested been described physiological in the sought Ca2+-activated Cl- channel of olfac- roles.1980s This and hundredschannel was of papers(controversially) described tory sensory neurons (OSN) in the main believed to also conduct organic osmolytes olfactory epithelium (MOE), but that it is like taurine and was postulated to not only dispensable for olfaction, we now focus on play a role in cell volume regulation (Fig.

144 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

2), cell division and migration, but also in apoptosis, cancer drug resistance, glio- Pending Patents transmitter release, and stroke. Progress in

this area, however, had stalled because de- VRAC modulators spite many attempts the molecular identity LRRC8-Comprising protein complexes and methods for identification of of VRAC remained unknown until our iden- ChlorideInventors: transporter Jentsch, Voss, ClC-7 Stauber; and cell-based Applicants: screening MDC/FVB method EP2919009 (A1), filed on 2014-06-17; US2015253303 (A1) filed on 2015-03-10 subunittification byof LRRC8Patapoutian heteromers and incolleagues. 2014 and Inventors: Jentsch, Stauber; Applicants: MDC/FVB EP2420830 (A1), filed on 2011-08-16 Wethe used parallel a genome-wide identification siRNA of the screen LRRC8A in Selected Publications a functional assay that relied on swelling- Gödde K., Gschwend O., Puchkov D., Pfeffer C.K., Carleton A., Jentsch T.J. (2016). Disruption of Kcc2-dependent inhibition of olfactory bulb output neurons suggests its importance in odour discrimination. Nature Comm. 7, 12043 induced iodide influx and subsequent YFP Planells-Cases R., Lutter D., Guyader C., Gerhards N.M., Ullrich F., Elger D.A., Kucu- 4fluorescence transmembrane quenching. domains Weand identified17 cyto- L.F., Brummelkamp T.R., Borst P., Rottenberg S., Jentsch T.J. (2015). VRAC channel plasmicLRRC8A asLeucine-Rich an essential Repeats subunit. (henceIt displays its kosmanoglu A, Xu G., Voss F.K., Reincke S.M., Stauber T., Blomen V.A., Vis D.J., Wessels - Fidzinskicomposition P., Korotkova determines T., its Heidenreich substrate specificity M., Maier N.,and Schuetze cellular S.,resistance Kobler O., to Zuschrat Pt-based- teranti-cancer W., Schmitz drugs. D., PonomarenkoEMBO J. 34, 2993-3008. A., Jentsch T.J. (2015). KCNQ5 K+ channels control name LRR Containing 8A). Expression stud hippocampal synaptic inhibition and fast network oscillations. Nature Comm. 6, VRACies and activity KO cells but generated needs at byleast CRISPR-Cas9 one other 6254. showed that LRRC8A is not sufficient for Weinert S., Jabs S., Hohensee S., Chan W.L., Kornak U., Jentsch T.J. (2014). Transport activity and presence of ClC-7/Ostm1 complex account for different cellular func- - torymember volume (LRRC8B-E). decrease and We that showed they also that Voss F.K., Ullrich F., Münch J., Lazarow K., Lutter D., Mah N., Andrade-Navarro M.A., LRRC8 heteromers are crucial for regula vontions. Kries EMBO J.P., Rep. Stauber 15, 784-791. T., Jentsch Our work (Science 2014) opened an entire an essential component of the volume-regulated anion channel VRAC. Science 244, T.J. (2014). Identification of LRRC8 heteromers as mediate swelling-activated taurine efflux.- 634-638. offield. VRAC, Our which next papermediates (EMBO apoptotic J. 2015) volume con Group Leader Karina Oberheide decrease,firmed the impairs hypothesis the progressionthat downregulation of drug- Prof. Dr. Dr. Thomas J. Jentsch Mayya Polovitskaya (*) induced apoptosis. With cisplatin, a widely (Joint with FMP / Cross Function) Sebastian Schütze used anticancer drug, there was an unfore- Till Stuhlmann seen twist: We discovered that it enters Scientists Florian Ullrich cells through VRAC (Fig. 3). Uptake of cis- Dr. Kathrin Gödde Dr. Maja Hoegg-Beiler MD Student Dr. Sabrina Jabs (*) Momsen Reincke (*) ofplatin, databases as well revealed as taurine a clinically efflux, was important greatly Dr. Anna Oliveras Martínez (*) correlationstimulated by of thereduced LRRC8D tumor subunit. expression Analysis of Technical Assistants - Dr. Rosa Planells-Cases Carolin Backhaus platin treated patients with ovarian can- Dr. SonaliIan Orozco Saha (*) Karolin Fuchs (*) cer.LRRC8D The observed with poor dependence survival in of cis/carbo substrate Dr. Tobias Stauber (*) Anika Günther (*) Dr. Janis Vogt Johanna Jedamzick shows that these proteins form the pore Dr. Felizia Voss Janet Liebold ofselectivity VRAC. Our on lab LRRC8 is now subunit tackling composition many ex- Dr. Stefanie Weinert(*) Nicole Krönke (*) citing aspects of VRAC structure/function, Dr. Pingzheng Zhou (*) Ruth Pareja Alcaraz physiology, and role in disease. Dr. Joanna Ziomkowska (*) Katrin Räbel Patrick Seidler PhD students Anyess von Bock Sebastian Albrecht Andrea Weidlich (*) Anja Blessing Silke Zillmann (*) Andreia Cruz e Silva Tony Daubitz (*) Animal Care Attendant Deborah Elger (*) Petra Göritz Corinna Göppner Carmen Ludwig (*) Research Coordinator Darius Lutter Dr. Norma Nitschke Jennifer Lück (*) Jonas Münch (*) = part of the period

MDC Research Report 2016 145 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

Helmut Kettenmann

Cellular Neurosciences

Our goal is to understand the role of glial space mediating signals between the brain cells in physiology and pathology. We ana- endothelium and neurons. They form intimate contact with synapses and neuronal lyze how glial cells communicate among activity results in astrocyte responses. Mi- each other and with neurons. We determine croglial cells are immune-competent cells how astrocytes and oligodendrocytes are in the brain and their functional role is best connected via gap junctions and how the during pathologic events. The present re- panglial network influences neuronal func- searchdefined program as the firstis focused responsive on four elements topics: tion. Another focus of our research is on (1) the extent of gap junction coupling microglial cells, the immune mediators and within the glial network and its impact on pathologic sensors of the central nervous brain function (2) the functional expression of transmitter receptors in microglia (3) system. We study the expression of trans- the response of microglial cells to brain in- mitter receptors in microglial cells and how jury and (4) the interaction of gliomas with activation of these receptors influences microglia and astrocytes. microglial functions. Within the context of Structural and functional analyses pathology we are currently studying the of panglial coupling networks and importance of microglial cells in stroke, its impact on brain signaling Alzheimer’s disease and schizophrenia. An- Niklas Meyer, Christiane Nolte (in collaboration with Christian Steinhäuser, University other line of research addresses the ques- of Bonn) tion how microglia interact with glioma cells. We aim to understand this interaction on a Gap junction coupling among glial cells molecular level for the purpose of identifying plays an important role for brain homeostasis, brain communication and brain energy therapeutic targets. supply. Most previous studies have focused on the network formed by astrocytes. We now characterize communication between Introduction all types of glial cells, in particular astro- The central nervous system contains two cytes and oligodendrocytes. We found that major cell populations, neurons and glial microglial cells are not part of the panglial network. We now address the question are essential for the proper functioning of whether these panglial networks are het- thecells. brain. It has The become different evident types that of glial cells erogeneous with respect to brain region, comparing different white and gray matter the myelin-forming cells of the central regions. We are characterizing junctional nervousfulfil distinct system tasks. and Oligodendrocytesensure rapid signal are properties and their modulation by neu- conduction in the white matter. The role of rotransmitters. Moreover, we now address the question whether the coupled network guiding structures during development and has functional implications for energy sup- representastrocytes importantis less well elementsdefined; they for controlprovide- ply to neurons. We aim to present a more ling the composition of the extracellular comprehensive view of the panglial net-

146 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

adult microglia also respond to the mus- carinicA subpopulation acetylcholine (11%) receptor of freshly agonist isolated carbachol with a Ca2+ increase, indicating the expression of functional receptors. The carbachol-sensitive population increased in microglia/brain macrophages isolated from the tissue of mouse models for stroke

and multiple sclerosis. As an example, mi- crogliaand Alzheimer’s cultured disease,from adult but notand for neonatal glioma brain contained a small carbachol-sensitive subpopulation which was increased in microglia isolated from stroke tissue. This increase correlated with an upregulation of the muscarinic receptor subtype 3 (M3). Carbachol is a chemoattractant for microglia and down-regulates their phagocytic ac-

Microglialtivity. (Funded cells by DFG show and anBIH). intrinsic spontaneous electrical activity un- correlated to neuronal responses Satellite microglial cells are closely associated Laura Maria Korvers, Marina Matyash, with neurons. On the left neurons are shown in Marcus Semtner, Stefan Wendt red, microglia in green. The right upper image shows such a satellite microglia/neuron pair. A subpopulation of microglia, called satel- Below the contact zone between microglia and - neuron is illustrated. phological soma-to-soma association with alite neuron, microglia indicative is defined of a direct by a closefunctional mor - sis revealed closely attached plasma mem- work and its heterogeneity and function in branesinteraction. of satellite Indeed, microglia ultrastructural and neuron. analy the brain. (Funded by DFG). potentials and postsynaptic potentials of Microglial express functional theEvoked associated local fieldneuron potentials did not lead or to action any receptors for neurotransmitter/ correlated membrane potential change in neurohormones the satellite microglia as revealed by dou- Amanda Costa, Meron Maricos, Martin ble patch-clamp recordings in cortex and Mersch, Susanne A. Wolf hippocampus. Microglia, however, show spontaneous transient membrane depo- We have addressed the question whether larizations which were not correlated with microglia express neurotransmitter re- neuronal activity. These events could be di- ceptors to sense neuronal activity. We vided into fast and slow rising depolarisa- tions which exhibited different characteris- the neurotransmitters/neurohormones tics in satellite and non-satellite microglia. GABA,have previouslyadrenaline, identified dopamine, receptors bradykinin, for A second form of spontaneous activity are serotonin, endothelin-1, substance P and transient increases in intracellular Ca2+ histamine. We are characterizing these re- which we study in cultured microglial cells. sponses not only in microglia in culture, These Ca2+ increases are due to release but also in situ and in freshly isolated cells. from internal stores.

MDC Research Report 2016 147 DISEASES OF THE NERVOUS SYSTEM

The response of microglial cells to esis and behavior. Moreover genome wide brain injury analysis of microglia by RNAseq isolated Masataka Ifuku, Philipp Jordan, Susanne A. Wolf, Stefan Wendt change in the transciptome that was nor- malizedfrom PolyI:C by minocycline animals revealed treatment. a significant These During any kind of disease or any pathological event such as after trauma, stroke - or in multiple sclerosis, the resting microg- liafindings dysregulation strongly as support an adjunctive the idea therapy to use lial cell transforms into an activated form inanti-inflammatory schizophrenic patients. drugs to target microg characterized by an ameboid morphology. Activated microglia can proliferate, migrate What are the properties of glioma to the site of injury, phagocytose, and re- associated microglia/brain macro- lease a variety of factors like cytokines, phages chemokines, nitric oxide (NO) and growth Alice Buonfiglioli, Jan-Frederik Fischer, factors. We are studying microglial proper- Felipe de Almeida Sassi, Susanne A. Wolf ties in different mouse models of disease (In collaboration with Michael Synowitz, University of Kiel, Dolores Hambardzumyan, (in collaboration with Matthias Endres and Emory University, Atlanta, Eric Holland, including Alzheimer’s disease and stroke Fred Hutch Cancer Center, Seattle and David Gutmann, Washington University, St. Louis) Christoph Harms, Neurology, Charité). In microglialcollaboration function with such the FMP as NO we production. identified Gliomas comprise the majority of cerebral Wedrug will like test compounds these novel that compounds change specific in the tumors and patients have a poor progno- above mentioned mouse models for their sis since there is essentially no concept therapeutic potential.(Funded by DFG) for successful treatment. A large fraction of the cells within a glioma are brain mac- Microglia as therapeutic target in rophages/microglial cells (GAMs) which psychiatric disorders Dilansu Gueneykaya, Daniele Mattei, mix of intrinsic microglia and monocytes Susanne A. Wolf invadedcan amount from up the to periphery. 30%. These We cells have are es a- tablished a procedure to isolate, purify and Many studies discuss an immunological culture human and mouse GAMs. Corre- component in the onset and progression of sponding control (non-glioma–associated psychiatric disorders. While the peripheral microglial cells) human cells are obtained immune system has been intensively inves- from patients with epileptic seizures that tigated, little is known about the role of mi- underwent temporal lobe surgery or from croglia. We thus aim to study the microglia post-mortem isolated microglia. We have phenotype derived from rodent models of determined the genome wide expression schizophrenia including epigenetic factors. To model a schizophrenia-like endopheno- and found that these cells acquire a distinct type, we used maternal immune stimula- profile of the GAMs from mouse and human- tion by injecting polyinosinic-polycytidilic sical scheme of the M1 or M2 phenotype, a phenotype which does not fit into the clas to mimic an anti-viral immune response. - acid (Poly I:C) into pregnant rats or mice croglialdefinition populations originating will from indicate the macrophage new path- increase constituting the factors correlat- waysfield. ofGenetic glioma profiling interactions of the with different GAMs. miWe ingWe best identified with decreases microglia in IL-1β net-neurogene and TNF-α- also determine how bulk glioma cells and sis and impairment in pre-pulse inhibition glioma stem cells differentially affect the GAM phenotype and found that GAMs pro- Treatment with the antibiotic minocycline normalizedof a startle responsemicroglial in cytokine the Poly production I:C model. in the hippocampus and rescued neurogen- mote glioma stem cells via IL-6 signaling (Funded by DFG, BIH and NIH).

148 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Scheme of interaction between glioma and microglia/brain macrophages. Glioma release the inactive pro-MMP2 which is cleaved into active MMP2 by microglial MT1-MMP. Its expressionn is stimulated by versicans released from glioma cells. Versican activates TLR2 signaling in microglial cells which leads to the up regulation of MT1-MMP. TLR2 signaling in microglia also triggers MMP9 release. This cascade promotes glioma growth (modified after Hambardzumyan et al., 2016).

Selected Publications

Wolf SA, Boddeke HW, Kettenmann H. (2017) Microglia in Physiology and Disease. Microglia/brain macrophages Annu Rev Physiol. 79: 619–643. promote glioma growth Hambardzumyan D, Gutmann D H and Kettenmann H (2015) The role of microglia/ Alice Buonfiglioli, Jan-Frederik Fischer, macrophages in glioma maintenance and progression. Nat Neurosci. 19:20-27. Verena Haage, Felipe de Almeida Sassi, Su- Kettenmann H, Kirchhoff F and Verkhratsky A (2013) Microglia: New roles for the sanne A. Wolf (In collaboration with Michael Synowitz, University of Kiel, and Dolores synaptic stripper. Neuron 77: 10 – 18. Hambardzumyan, Emory University) Szulzewsky F, Pelz A, Feng X, Synowitz M, Markovic D, Langmann T, Holtman IR, Wang X,M1 Eggen and M2 BJ, polarizationBoddeke HW, and Hambardzumyan highly express GpnmbD, Wolf SA,and Kettenmann Spp1. PLoS One. H (2015) 2015 Glioma-Feb We have found that microglial cells strong- associated microglia/macrophages display an expression profile different from Wendt S, Wogram E, Korvers L, Kettenmann H. (2016) Experimental Cortical Spread- ly promote glioma growth and invasion. 6;10(2):e0116644. doi: 10.1371/journal.pone.0116644. eCollection 2015. There is an interesting interplay between Neurosci. 36:6165-6174. microglial and glioma cells. Glioma cells re- ing Depression Induces NMDA Receptor Dependent Potassium Currents in Microglia. J lease the metalloprotease MMP2 which is important for degradation of extracellular Group Leader MD matrix and promotes invasion. This metal- Prof. Dr. Helmut Kettenmann Le Dong loprotease is, however, released in an inac- Jan-Frederik Fischer tive, larger form and it needs to be cleaved Scientists Paul Triller to acquire its activity. This cleavage is ac- Dr. Christiane Nolte complished by the ectoenzyme MT1-MMP. Technical Assistants An endogenous ligand for Toll-like recep- Dr. Marina Matyash Regina Piske tors (TLR), versican, is released from glio- Dr. MarcusMasataka Semtner Ifuku Nadine Scharek ma and activates TLR2 toll-like receptors in Dr. Susanne A. Wolf Michaela Seeger-Zografakis the surrounding GAMs. TLR2 signaling trig- Maren Wendt gers the expression of MT1-MMP in GAMs. PhD students The activation of the TLR2 pathway also Felipe de Almeida Sassi Secretariat induces the release of MMP-9 from GAMs. Amanda Costa Birgit Jarchow Thus, glioma cells exploit GAMs to promote Dilansu Gueneykaya their invasion. Thus interfering with TLR Verena Haage Other receptors or their intracellular pathways Philipp Jordan Jieqi Chen reduced the rapid expansion of glioma cells Laura Maria Korvers Meino Alexandra Gibson as we could show with TLR2-interfering Meron Maricos Vera Heinemann antibodies and microglia cells have become Daniele Mattei Stefanie Korthals a new target for glioma research and for po- Niklas Meyer Victoria Rothe Stefan Wendt

tential therapy. (Funded by DFG and NIH).

MDC Research Report 2016 149 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

Gary Lewin

Molecular Physiology of Somatic Sensation

Somatic sensation includes all those sensa- Molecular Basis of Mechanotrans- tions that we consciously feel after stimula- duction tion of the body, e.g. touch, warmth, cool- Caglar Gök, Carina Fürst, Raluca Fleischer, Regina Herget, Mirko Moroni, Kate Poole, ing, or even limb movement. We experience Christiane Wetzel, Martha Rocio Servin- these sensations as a direct result of the Vences activation of sensory neurons that are located in the dorsal root ganglia (DRG). In Mechanotransduction is the process whereby receptor proteins present in the end- our group we are interested in the molecu- ings of sensory neurons are able to detect lar mechanisms that allow these neurons mechanical stimulation of the tissue they to transduce these varied stimuli. Sensory innervate. We have used information from neurons can, for example, detect changes genetic experiments with the nematode worm C. elegans to identify possible verte- in temperature of the skin in the non-nox- brate candidate proteins that might detect ious (not painful) as well as the noxious mechanical stimuli. Genetic screens for range (painful heat, or cold). They can also touch insensitive worms have turned up around 15 genes whose function is neces- detect gentle movement of the skin as well sary to confer touch sensitivity. These genes as intense mechanical stimulation of the were named mec for mechanically insensi- skin that is normally harmful. The nature tive and we have focused on identifying the of the transduction molecules involved in role mammalian orthologs of these genes play in touch sensation. The mec genes in transforming mechanical and thermal stimuli C. elegans have been proposed to work to- into electrical signals that lead to percep- gether in a mechanotransduction complex. tion is a major focus in the lab. Mecha- An essential component of this complex is nosensitive channels required for touch the membrane protein MEC-2 that forms a hairpin in the membrane and might regu- sensation have recently been identified in late the activity of the mechanotransduction sensory neurons in work from our group. channels. We have cloned and characterized It is increasingly clear that mechanosensi- vertebrate homologues of mec genes and have created mouse mutant alleles to char- tive channels serve as important initiators of acterize the in vivo function of these genes. critical signalling events in cell types outside MEC-2 is a member of a large family of proof sensory systems, a new area being ex- teins that contain a stomatin-like domain. plored in the lab. A member of this family called STOML3 (stomatin like protein-3) was cloned by our group, and we generated a mouse model

STOML3 mutant mice many mechanorecep- torswith (ora null touch mutation receptors) of the in STOML3 the skin locus. do not In work in the absence of the STOML3 protein.

150 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 1 Dramatic loss of mechanosensitivity in slowly adapting mechanoreceptors (SAM are type I Merkel cell receptors) after conditional deletion of the Piezo2 gene in sensory neurons (A). A cardinal feature of all SAMs is high sensitivity to moving stimuli. (B) is a quantitatve measure of the receptors velocity sensitvity which is virtually absent after Piezo2 gene deletion. (B) data reproduced from Ranade et al 2014 Nature.

The central core of any mechanotransducer et al 2014). The use of this newly developed is an ion channel or channels(s), unfortu- pili technique allows us to measure with ex- nately the molecular nature of this channel tremely high precision fast mechanotransduction in almost any cell type. The lab is sensory neurons. There is some evidence now investigating the physiological role of has not yet been definitively identified in which belong to the Deg/ENaC family of ion in a variety of non-sensory cell types. that acid sensing ion channels (ASICs)- “new and old” mechanically gated channels duction of mechanical stimuli by sensory Stomatin domain proteins, mo- channels have an influence on the trans lecular membrane scaffolds with are genetic interactions between stoma- multiple roles in physiology and tin-domainneurons. Indeed proteins we haveand shownacid sensing that there ion pathophysiology channels. Recently, two new mechanosen- Caglar Gök, James Hall, Kate Poole, Jane Reznick mammals called Piezo1 and Piezo2. Piezo 2sitive is aion huge channels trimeric have membrane been identified protein in Stomatin-domain proteins can regulate that is expressed in touch receptors of the both ion channels and transporter proteins. The structural and molecular basis of such - regulation remained a mystery until re- tutedorsal in rootCalifornia ganglia. we Inrecently collaboration showed withthat cently. Together with the Daumke group we Piezo2researchers is absolutely at the Scripps required Research for touch Insti re- have used structural biology techniques and structure function studies to examine the the loss of function in mice with conditional molecular mechanisms of stomatin-domain deletionceptor function of the Piezo2 (Ranade gene et al in 2014). touch Indeed recep- - tors has remarkable similarities to the sen- raphy techniques the structure of the mam- sory phenotype of STOML3 mutant mice malianprotein function.stomatin-protein By using wasX-ray solved: crystallog We generated in our lab. showed that the basic building block of this Using a newly developed assay based on protein is a banana-shaped dimer that can elastomeric pili we could also show that the form the basis of higher order oligomeric sensitivity of Piezo1 and Piezo2 ion chan- structures in the membrane. The oligomer- nels to mechanical displacement is exqui- ization of stomatin-domains appears to be sitely tuned by the STOML3 protein (Poole necessary for their ability to modulate both

MDC Research Report 2016 151 DISEASES OF THE NERVOUS SYSTEM

ion channels like Piezo1 and Piezo2 (Poole these proteins may function to regulate etASIC al 2014).proteins Ongoing as well work as mechanosensitive in the lab seeks theing cellsadaptation of the innerof the ear. mechanotransduc In both cell types- to examine the precise in vivo function of tion channels. We are currently working on other poorly understood stomatin-domain further hearing genes that may also affect containing proteins such as STOML1. The cutaneous mechanosensation. The same use of mouse models and advanced cellular genes as we study in the mouse are also imaging techniques has revealed unexpect- mutated in humans and it is possible that ed roles for stomatin-like proteins in meta- the perception of cutaneous touch stimuli bolic diseases. is altered in such patients. We have measured psychometric functions in healthy The activity of mechanosensitive ion chan- and hearing impaired people in order to nels expressed by sensory neurons can be describe quantitatively differences in the measured using high-resolution electro- perception of touch. We carried out a large physiology techniques. We have recently twin study which showed that touch and shown that such ion channels in the mem- also thermosensation are both heritable in branes of cultured DRG neurons can be ac- humans. The results of these studies have tivated by stimuli in the nanometer range also revealed that there appears to be a (Poole et al 2014). We have also gathered strong link between the sense of touch and considerable evidence that the mechano- the sense of hearing. We could demonstrate sensitive channels are actually opened via a that one gene, mutations in which cause protein tether that attaches to laminin-con- severe hearing loss and blindness is also taining extracellular matrices. We have re- associated with poor touch performance in cently also shown that STOML3 is required humans. This gene USH2A encodes a large to maintain the high sensitivity of endog- extracellular protein that is thought to form enous Piezo2 currents in primary sensory a structural link within the sterocillia of in- neurons. We also developed small mol- ner ear hair cells. The role of this protein in ecules that inhibit the oligomerization of touch receptors remains to be investigated. STOML3 and these drugs work in vivo and Our ability to sense temperature changes in vitro to decrease the sensitivity of Piezo in the skin is actually quite remarkable. Humans can easily detect skin cooling of can reversibly silence touch receptors in just 1o theion channels.mouse and Indeed can be these used small to treat molecules painful Poulet we have now shown that mice also mechanical hyper sensitivity in mice that have suchC or less.an exquisite In collaboration temperature with Jamessense follows nerve lesions (Wetzel et al 2016). and have started to identify the sensory afferents that relay this information to the Touch, hearing, and thermal cortex (Milenkovic et al 2014). We are now sensation using optogenetic and optochemical tools Regina Herget, Johannes Kühnemund, Rabih to determine which sensory afferents con- Moshourab and Jan Walcher tribute to the perception of warming and cooling (Frank et al., Nat. Comm.6, 2015, Hereditary deafness is a relatively common phenomenon and a large number of genes humans.doi: 10.1038/hcomms8118). We have established We a constantlybattery of lead to deafness in mouse and man. We are psychometrictry to apply our testing basic findingsprocedures in mousethat en to- workinghave been with identified several deaf that mutant when mutatedmice to able us to measure very precisely touch per- examine whether genes required for nor- formance in healthy volunteers as well as mal mechanotransduction in the inner ear may also be required for normal cutaneous c-kit receptor as an essential regulator of sensation. Our data indicate that members thedefined heat patienttransduction cohorts. mechanism We identified in mouse the of the unconventional myosin protein fam- nociceptive sensory neurons. As a follow ily have a common function in sensory neu- up to these studies we could recently show rons and in hair cells, mechanotransduc- that humans being treated with drugs that

152 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

interfere with c-kit signaling actually modi- evolved a remarkably unique carbohydrate fy noxious heat perception (Ceko et al 2014 metabolism that may help them survive severe hypoxic stress. The metabolic basis of mole-rat resistance to hypoxic stress is a ma- ThePain DOI:Naked 10.1016/j.pain.2014.03.010). Mole-Rat an jor new research direction in the lab. extremeophile mammal Damir Omerbasic, Tania Kosten, Jane Selected Publications Reznick, Valerie Begay-Müller Wetzel C, Pifferi S, Picci C, Gök C, Hoffmann D, Bali KK, Lampe A, Lapatsina L, Fleischer R, St. John Smith E, Bégay V, Moroni M, Estebanez L, Kühnemund J, Walcher The naked mole rat is an unusual subterra- J, Specker E, Neuenschwander M, von Kries JP, Haucke V, Kuner R, Poulet JFA, Schmo- ranzer J, Poole K, Lewin GR. Small molecule inhibition of mechanotransduction modulates touch perception and can control neuropathic pain. Nat. Neurosci. 2017 known poikilothermic mammal (ie. cold blooded),nean rodent it lives in many in colonies respects. with It is an the insect- only like social structure, and it is also the lon- FaulkesFeb 20(2): CG, 209-218. Selbach M,doi: Lewin 10.1038/nn.4454. GR. Hypofunctional (epub TrkAahead Accounts of print) for the Absence of gest-lived rodent species known (lifetimes Omerbašić D, Smith ES, Moroni M, Homfeld J, Eigenbrod O, Bennett NC, Reznick J, Milenkovic,Pain Sensitization N., Zhao, in theW.-J., African Walcher, Naked J., Albert, Mole-Rat. T., Siemens, Cell Rep. J., 2016 Lewin, Oct G.R., 11;17(3):748- Poulet, J.F.A., this animal has normal acute pain respons- 2014.758. doi: A somatosensory 10.1016/j.celrep.2016.09.035. circuit for cooling perception in mice. Nat. Neurosci. 17, esin excessit displays of 32 no yrs). hypersensitivity Interestingly, (so although called Poole, K., Herget, R., Lapatsina, L., Ngo, H.-D.D., Lewin, G.R., Herget, R., Lapatsina, L.,1560–1566. Ngo, H.-D.D., doi:10.1038/nn.3828 Lewin, G.R., Herget, R., Lapatsina, L., Ngo, H.-D.D., Lewin, G.R., 2014. and chemical stimuli. What is particularly strikinghyperalgesia) in the to naked a variety mole-rat of inflammatory is that the animals completely lack a neuronal or be- Ranade,Tuning Piezo S.S., Woo, ion channels S.-H., Dubin, to detect A.E., Moshourab,molecular-scale R.A., movements Wetzel, C., Petrus, relevant M., for Mathur, fine J., Bégay,touch. Nat.V., Coste, Commun. B., Mainquist, 5, 3520. doi:10.1038/ncomms4520J., Wilson, A.J., Francisco, A.G., Reddy, K., Qiu, Z., Wood, havioral response to acid. We suspect that J.N., Lewin, G.R., Patapoutian, A., 2014. Piezo2 is the major transducer of mechanical at the heart of this specialized adaptation lies in distinct gene variants encoding ion forces for touch sensation in mice. Nature 516, 121–125. doi:10.1038/nature13980 channels and associated channels that are required for the transduction of painful Group Leader Graduate and Undergraduate stimuli. We are at present cloning and char- Prof. Dr. Gary Lewin students acterizing genes coding ion channels from Jedrzej Blaszkiewicz the naked mole rat to address this issue. Scientists Johannes Kühnemund We have cloned and started to character- Dr. Valerie Begay-Müller Ella Middelhoff ize the naked mole rat capsaicin receptor, Dr. Mirko Moroni an ion channel called TRPV1, as well as the Dr. Rabih Moshourab Technical Assistants tyrosine kinase receptor trkA the activation Dr. Kathryn Poole (Cécile-Vogt Kathleen Barda of which by NGF can potently potentiate Fellow) Maria Braunschweig TRPV1 (Omerbasic et al 2016). We were Dr. Jane Reznick Liana Kosizki very interested in understanding how na- Dr. Frederick Schwaller Anke Scheer ked mole-rats are completely insensitive to Dr. Jeffrey Stear Heike Thränhardt acid. We could show that this insensitivity Dr. Christiane Wetzel can be explained by a unique variant of the Secretariat nociceptor-expressed voltage gated sodium PhD students Sophia Kanter

channel NaV1.7 which is potently blocked Ole Eigenbrod Cornelia Stärkel by protons in this species leading to behav- Raluca Fleischer ioral insensitivity to acid (Smith et al 2011 Permanent Fellow Hammes-Lewin) Dr. Annette Hammes-Lewin CaglarCarina GökFürst Iris (with Annette see also Chapter Molecular Path- However,Science DOI: our 0.1126/science.1213760). interest in naked mole-rat James Hall ways in Cortical Development physiology goes beyond the study of sensory Julia Haseleu systems. We are studying molecular chang- Tetiana Kosten Animal Attendant es in naked mole rats that may account for Jan Walcher Franziska Kressin their complete lack of thermogenesis, and Martha Rocio Servin-Vences uniquely low metabolic rates. We have re- (with Kate Poole) cently discovered that naked-mole-rats have

MDC Research Report 2016 153 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

James Poulet

Neural Circuits and Behaviour

Our goal is to understand neural mecha- The mouse forepaw system nisms of sensory processing and percep- The mouse forepaw has a similar structure and function to the human hand (Figure tion. We examine brain regions associated 1). Mice use their forepaws not only for with sensing and movement of the mouse locomotion but also for manipulating ob- forepaw, with a special focus on neocorti- jects and sensing their immediate tactile cal circuits. To investigate the link between environment. The mouse forepaw has 5 neural activity and sensory perception we a thumb. The glabrous skin is innervated record and manipulate genetically identified withdigits, a albeitfull complement the first more of primary of a stump sensory than neurons in mice performing trained behav- afferent neurons processing the multiple modalities of somatosensation including ioural tasks with their forepaw. light touch, proprioception, deep touch, temperature and pain.

Peripheral and central circuits of thermal perception The temperature of an object provides important information for animals performing tactile tasks. Temperature and touch input are thought to closely interact in the brain but the primary sensory afferents in the skin and central circuits they engage to enable the perception of surface temperature

et al., 2014), we delivered high-resolution thermalwere unclear. and tactile In this stimuli project to the (Milenkovic forepaw digits (Figure 2). We showed that mice are able to detect small amplitude (~ 2ºC) cooling stimuli, making their thermosensory performance equivalent to that of humans. We went on to show that primary somatosensory cortical neurons respond both to tactile and thermal stimulation of the forepaw and are required for non-noxious Figure 1. The mouse forepaw system. A. Image of the glabrous skin of cooling perception. Using mutant mice, we the right forepaw, mice have 5 digits (D1-D5) and 5 pads (P1-3, Thenar showed that the transient receptor poten- pad, TH, and the HypoThenar pad, HT), scale bar 1 mm. B. Cartoon of the - mouse forepaw system and the sensory pathway in blue from forepaw to sary for mild cold temperature perception. somatosensory cortex (S1) via spinal cord, brainstem and thalamus. Finally,tial melastatin in collaboration 8 (TRPM8) with protein the lab is of neces Gary Lewin at the MDC, we performed primary

- sensory nerve recordings. It was thought that sensory afferent C fibers were only in 154 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 2. Mild cold thermal perception in mice is mediated by primary somatosensory cortex. A.Setup. B. Mice lick in response to forepaw cooling C. Pharmacological silencing of S1 abolishes cooling perception. D. TRPM8 mutant mice cannot perceive cooling. E. Intrinsic optical imaging of S1 shows overlapping response to cooling and touch. F. Sensory afferent responses to cooling. G. Averaged membrane potential responses to cooling and touch in 4 example cortical neurons.

volved in pain processing, but we show that and low spontaneous false alarms rates, giv- they are the likely drivers for non-noxious ing us a novel, highly sensitive psychophysi- cooling perception and cortical thermal re- cal testing platform. We currently examine sponses in mice. the activity of subsets of cortical neurons during this behaviour and their impact on In vivo monosynaptic transmission perception and forepaw movement. between sensory cortex neurons Monosynaptic connections between cortical neurons underpin cortical sensory Selected Publications processing but little is known about mono- Zhao W.J., Kremkow J. and Poulet J.F.A. (2016) Translaminar cortical membrane synaptic transmission in vivo. Recently we potential synchrony in behaving mice. Cell Reports developed a method to make some of the Jouhanneau J-S. Kremkow, J. Dorrn, A. and Poulet J.F.A. excitatory transmission between layer 2 pyramidal neurons., 15: 2387-2399. Cell Reports, 13: 1-9. neurons in vivo (Jouhanneau et al., 2015). Jouhanneau J.S., Ferrarese L., Estebanez L., Audette N., Brecht (2015) M., In Barth vivo monosynaptic A.L. and Wefirst show recordings that layer of synaptically 2 cortical connectedpyramidal Poulet J.F.A. excitatory neurons targeted by POm. Neuron neurons form a sparsely connected network Milenkovic N., (2014) Zhao W.-J., Cortical Walcher fosGFP J., expressionAlbert T., Siemens reveals J-E, broad Lewin receptive G. Poulet field with an overrepresentation of bidirectional J.F.A. (2014) A somatosensory circuit for cooling, 84: perception 1-14. in mice. Nature connections. The majority of unitary excit- Neuroscience, 17: 1560-1566. atory postsynaptic potentials were small in Barth A., and Poulet J.F.A. (2012) Experimental evidence for sparse cortical coding. Trends Neurosci., 35: 345-55.

amplitude (< 0.5 mV), with a small minority less> 1 mV. reliable In comparison and showed to prior little in short-termvitro work, Group Leader PhD students synaptic connectionsdepression. inOur “Downstates” approach werewill Dr. James Poulet Leiron Ferrarese help bridge the gap between monosynaptic (Joint with NeuroCure) Diana Hoffmann connectivity and function and allow investi- Tobias Leva gations into the impact of network activity Scientists Ricardo Paricio Montesinos on synaptic transmission. Dr. Mario Carta Dr. Jean-Sebastien Jouhanneau Technical Assistant A sensory-triggered goal-directed Dr. Jens Kremkow Janett König reaching task for mice Dr. Nevena Milenkovic-Zujko Charlene Memler To monitor sensory perception we have Dr. Florian Rau developed a new behavioural platform. Dr. Philipp Schnepel Secretary Multiple mice are trained simultaneously Dr. Annapoorani Udhayachandran Janin Bergmann to report a tactile stimulus delivered to the Dr. Mikkel Vestergaard forepaw by reaching and pressing a target Dr. Birgit Voigt Lab Manager sensor using the same paw. Mice perform Dr. Phillip Wisinski-Bokiniec Dr. Evgeny Bobrov the task with short latency, high accuracy

MDC Research Report 2016 155 DISEASES OF THE NERVOUS SYSTEM Photo: private Photo:

Alessandro Prigione

Mitochondria and cell fate reprogramming

Induced pluripotent stem cells (iPSCs), gen- Mitochondrial reprogramming in erated through reprogramming of somatic cell fate conversion and neural specification cells back into an embryonic-like pluripotent Recent advances in stem cell technologies state, hold great promises for biomedical reveal the plasticity of the mammalian epig- research. enome. We have previously demonstrated that We are interested in dissecting the contribut- upon reprogramming of somatic cells to in- ing role of mitochondria and energy metabo- ducedmitochondria pluripotent are also stem drastically cells (iPSCs), reconfigured a process lism in enabling cellular conversion. By build- that is paralleled by a metabolic switch towards ing a mitochondria-centered model of cell glycolysis. This glycolytic shift represents an early reprogramming event, preceding the ex- fate reprogramming and differentiation with pression of genes controlling pluripotency and a targeted neuronal focus, we wish to unveil self-renewal. Thus, metabolic transformation the relevance of mitochondrial modulation might control cell fate identity. Modulation of for cell identity, neuronal commitment, repro- bioenergetics is particular important for the development of the nervous systems, which gramming, and rejuvenation. consumes almost half of the whole-organism Additionally, we seek to apply the iPSC tech- basal metabolic rate. However, the regulatory nology for modeling neurological diseases af- role of mitochondria and energy metabolism fecting the mitochondria either directly, such in dictating neural commitment and orchestrating its epigenetic rewiring remains to be as mitochondrial DNA (mtDNA) disorders, or indirectly, like Huntington’s disease (HD). the potentially instructive role of bioenerget- The development of alternative modeling icinvestigated.In restructuring this for project, cell fate we conversion seek to dissect and approaches is highly needed for conditions affecting the nervous system, whose under- iPSCs,neural neural specification progenitors, in particular. neurons, We and initially as- standing has been hindered by the inability trocytes)focus on definedto map cellulartheir mitochondrial/meta states (fibroblasts,- to sample live neuronal cells. This is particu- bolic signature and the downstream effects on the epigenetic landscape. We then analyze larly important for mtDNA disorders, which the mechanistic function of mitochondria and lack viable modeling tools due to the hurdles energy metabolism during reprogramming to associated with engineering mtDNA. Repro- iPSCs and upon glial and neuronal differen- gramming-derived neurons can be eventu- tiation, by altering nutrient availability and by chemical and genetic manipulation of critical ally employed as disease-relevant cell type- mitochondrial and metabolic regulators. Fi- specific cellular systems for the discovery of nally, we wish to integrate the obtained data to novel disease-modifying therapeutic strate- computationally build a metabolic map of cell identities. This will be coupled to mitochondri- gies for these untreatable brain disorders. al-related high throughput assays to possibly identify enablers of metabolic-driven cellular conversions. Overall, these studies may lead

156 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

(A) Schematic cartoon of our disease modeling approach.(B) Following integration-free iPSC generation (upper image), patient and control cells are coaxed into astrocytes (middle image) and neurons (lower image). (C) Assessment of mitochondria-related functions include bioenergetic profiling and calcium imaging.

to a better understanding of the importance disease states and to in silico predict the tar- of mitochondria and metabolism for brain gets of potential interventional strategies. The development and for the control of cell fate computational predictions will be validated in follow-up cellular experiments. Finally, based

Anplasticity iPSC-systems and specification. biology approach to establish scalable assays amenable to high- for modeling and discovery of throughputon the identified and focused disease pathways,compound wescreen seek- therapeutic strategies of inherited basal ganglia disorders biology approach may represent an innova- The project aims to generate patient-derived tiveings. platform If successful, for drug this discovery iPSC-driven of complex systems neuronal cell models of two genetic disorders genetic brain diseases. affecting the basal ganglia and lacking effective treatments. Maternally inherited Leigh syn- Selected Publications due to point mutations within the MT-ATP6 genedrome of (MILS) the mitochondrial is an infantile encephalopathyDNA (mtDNA). - - Lorenz C.; Lesimple P.; Bukowiecki R.; Zink A.; Inak G.; Mlody B.; Singh M.; Semtner erative disorder caused by the CAG triplet re- M.; Mah N.; Aure K.; Leong M.; Zabiegalov O.; Lyras E.M.; Pfiffer V.; Fauler B.; Eich horstiPSC-derived J.; Wiesner neural B.; Huebnerprogenitors N.; Prillerare an effectiveJ.; Mielke drug T.; Meierhofer discovery D.;model Izsvak for Z.; neurologi Meier - peatHuntington’s expansion disease within (HD) the gene is a encoding neurodegen the J.C.;cal mtDNA Bouillaud disorders. F.; Adjaye Cell J.; Stem Schuelke Cell. M.;doi:10.1016/j.stem.2016.12.013 Wanker E.E.; Lombes A.; Prigione A. Human protein Huntingtin. Basal ganglia are highly dependent on mitochondrial-based energy 19 (4): 419-420. (2016-10-06) production. Hence, the two diseases repre- Mlody B.; Prigione A. A glycolytic solution for pluripotent stem cells. Cell Stem Cell- sent in a way two paradigmatic examples, one tion and in iPSC models of brain disorders. Semin Cell Dev Biol. (2016-02-11) Mlody B, Lorenz C, Inak G, Prigione A. Energy metabolism in neuronal/glial induc - exhibiting a direct mitochondrial implication and one where mitochondrial dysfunctions Wang J.; Xie G.; Singh M.; Ghanbarian A.T.; Rasko T.; Szvetnik A.; Cai H.; Besser D.; Pri are considered a secondary insult. Human gione A.; Fuchs N.V.; Schumann G.G.; Chen W.; Lorincz M.C.; Ivics Z.; Hurst L.D.; Izsvak Z. Primate-specific endogenous retrovirus-driven transcription defines naive-like basal ganglia neurons (GABAergic and dopa- stem cells. Nature 516 (7531): 405-409 (2014-12-18) minergic) and glia cells (astrocytes) will be Prigionethrough earlyA.; Rohwer glycolytic N.; Hoffmanshift and S.;up-regulation Mlody B.; Drews of PDK1-3 K.; Bukowiecki and PKM2. R.; Stem Bluemlein Cells 32 K.; obtained from induced pluripotent stem cells Wanker(2): 364-376 E.E.; Ralser(2014-02) M.; Cramer T.; Adjaye J. HIF1{alpha} modulates reprogramming - broblasts using integration-free Sendai-based reprogramming.(iPSCs) generated The from chance patient to andidentify control a dis fi- Group Leader Master Students ease phenotype within the reprogramming- Dr. Alessandro Prigione Lamia Bouguerne derived basal ganglia cells and to develop (Delbrück Fellow) Franziska Dinter potential therapeutic strategies might be di- Preethi Kannan minished when using only a conventional re- Scientists Megan Leong Dr. Barbara Mlody Oleksandr Zabiegalov study, we therefore broaden the levels of in- Dr. Pawel Lisowski vestigationsductionist molecular by combining biology standard approach. function In this- TA PhD students Franziska Wiedemann driven analyses (systematic transcriptomics, Raul Bukowiecki proteomics,al and biochemical and metabolomics) assays with global to unravelOMICS- TA Trainee the disease mechanisms without an a priori Carmen Lorenz Denise Heidler knowledge. This would allow the generation AnnikaGizem Inak Zink Celina Kassner of computational models for the two neuronal

MDC Research Report 2016 157 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

Fritz G. Rathjen

Developmental Neurobiology

The correct wiring of the brain is of critical to the complexity of neuronal circuits and importance for its function. It is orches- is essential for an integrated processing of trated by multiple molecular factors and information. Consequently, impairment of cellular mechanisms during developmental periods. The research of Fritz G. Rathjen’s interestaxon branching to unravel causes molecular severe mechanisms deficits in group focuses on several topics of the “wir- implicatedbrain function. in the It regulation is therefore of axon of essential branching” problem. For example, we investigate ing. the role of cGMP signaling on axon branch- Our previous investigations showed that ing, of signaling systems implicated in the cGMP signaling is essential for axon branch- regulation of interneuronal synapses and ing of dorsal root ganglia (DRG) and cranial the function of Ig cell adhesion proteins on sensory ganglia (CSG) in vivo. The cGMP signaling cascade is composed of the ligand calcium waves in the fine tuning of circuits. CNP, the receptor guanylyl cyclase Npr2 Moreover, we study neuronal mechanisms implicated in ultrasonic vocalizations of the absence of any of these components, ax- onsand of the sensory serine/threonine neurons no longer kinase bifurcate cGKI. In mice. and instead individual axons only turn in a descending and ascending direction of the spinal cord or hindbrain (Figure 1).

A major open question is to what extent the mechanism underlying axon bifurcation of DRG or CSG neurons can be generalized. Our current investigation concentrate on the branching of axons from mesencephalic The impact of cGMP signaling on trigeminal neurons (MTN) which also ex- axon branching, sensory percep- - tion and vocalization cialized group of cells that simultaneously innervatepress Npr2 motor and cGKI. neurons MTNs of are the a hindbrain very spe Npr2-mediated cGMP signaling is implicated and jaw closing muscles and therefore coin axonal bifurcation of sensory axons ordinate the movement of maxilla and man- dible (Figure 1). During extension of axons to their target - Npr2-mediated signaling affects palmi- es most, if not all, axons generate different toylation of specific proteins in growth cones typesfields atof embryonic branches. orThese early postnatalarbors enable stag an individual neuron to innervate several To get mechanistic insights into the pro- distinct neurons which might be located cess of axonal bifurcation we have begun to in different parts of the brain. Therefore, study downstream events of Npr2-mediat- axon branching enormously contributes ed signaling. Biochemical as well as immu-

158 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure1: Axon bifurcation of three types of sensory neurons is controlled by the CNP-Npr2- cGKI signaling cascade: DRG, CSG and MTN neurons. A) Wildtype, B) Mutant. In the absence of CNP, Npr2 or cGKI sensory axons are unable to bifurcate when entering the spinal cord or hindbrain. CG – cranial ganglia; DRG – dorsal root ganglia; MTN – mesencephalic trigeminal neurons; FB – forebrain; MB – mid- brain; HB – hindbrain; SC – spinal cord.

nocytological studies indicated that the ki- Axon bifurcation is essential for a correct representation of the periphery within the membranes in the vesicular compartment CNS nase cGKI is associated with palmitoyl-rich- - We have also begun to examine the impact mitoylationsuggesting that - a reversible cGKI might post-translational regulate intra of disturbed axon bifurcation on sensory cellular protein trafficking by protein S-pal perception. Painful stimuli are detected by proteins from intracellular membranes to sensory neurons which reveal the above de- modification involved in the shuttling of- scribed disturbed axon bifurcation. Howev- onstrated by a biochemical approach that er, Npr2 has also a critical function for lon- cGMPthe plasma signaling membrane. regulated Indeed, S-palmitoylation we dem gitudinal bone growth in that its absence of a narrowed number of proteins. Candi- causes a disproportionate growth accom- dates are currently tested by genetic methods whether they control axonal bifurcation. ofpanied Npr2 bywhich a dwarfism. was then To selectively overcome inacti these- vatedobstacles in sensory we have neurons generated by athe floxed Wnt1-Cre allele Degradation of cGMP by PDE2A (phosphodi- driver. This approach resulted in animals esterase 2A) with normal-sized bones but with impaired

To elucidate the signaling components es- mice revealed a decrease in acute nocicep- sential for the regulation of cGMP levels tiveaxon pain bifurcation. sensation Initial suggesting studies that on the these skin we investigated the localization of PDE in is incompletely represented in the spinal embryonic DRGs. PDE2A was found in a cord in the absence of Npr2. pattern overlapping with that of Npr2- and Npr2 mediated cGMP signaling is implicated in ultrasonic vocalization aftercGKI. CNPMeasurements stimulation of resultedcGMP levels in a in strong DRGs increasefrom wild-type of cGMP and in PDE2A-deficientthe PDE2A knockout mice Since Npr2-mediated cGMP signaling is im- that was not compensated by other PDEs. plicated in a broad variety of physiological However, this increase in cGMP levels did functions and since the guanylyl cyclase not result in altered branching pattern of Npr2 in addition to its localization in senso- DRG axons indicating that axon bifurcation ry neurons is expressed by groups of motor is stable to a rise in cGMP levels. neurons in the hindbrain at early embry-

MDC Research Report 2016 159 DISEASES OF THE NERVOUS SYSTEM

Figure 2: Analysis of calcium waves in the developing retina of the mouse. A) A flat-mounted mouse retina (P2) is loaded with the calcium-sensitive dye Fura-2 and fluorescent intensities were analyzed in regions encircled as indicated (ROI – regions of interest) in B). C) Example fluorescence traces from representative retinal ganglion cells are shown. D) Fluorescence intensities were converted to heat maps to indicate the number of waves over time.

onic and postnatal stages we explored its the fundamental role of synaptogenesis in role in ultrasonic vocalization of mice. Mice neuronal circuit formation, the molecular are able to generate an array of ultrasonic components orchestrating synapse forma- vocalizations in a variety of behavioral con- tion in the brain are not fully understood texts, for example infants produce so-called in contrast to the neuromuscular junction. isolation-induced calls when separated Here Lrp4 acts as agrin receptor resulting from their mothers. These isolation calls in the activation of the protein tyrosine kinase receptor MuSK. We now demonstrate Coordinating movements of the larynx, the that Lrp4 is also expressed in the cortex vocalcan be tract elicited and very the reliablyrespiratory and systemquantified. are with high levels at developing periods. required for vocalization. Analysis of synapse formation in microis- land cultures by patch clamp recordings Our studies reveal that in Npr2- or CNP- indicated that Lrp4 also mediates synapse formation on cortical neurons upon application of soluble agrin suggesting that Lrp4 reduceddeficient suggesting mice the number that cGMP and frequencysignaling might have a function on cortical neurons eitherof isolation interferes ultrasonic with calls the are differentiation significantly similar to that one at the neuromuscular junction. function of neurons. Our current investiga- of specific neurons or modulates an acute CALEB is a transmembrane protein with an the neuronal populations that are essen- EGF-like domain which is selectively local- tialtions to concentrateproduce ultrasounds. on the identification These genetic of ized in the postsynaptic compartment of studies might contribute to characterize central neurons. Previous investigations neuronal circuits underlying call usage and demonstrated an impaired presynaptic structure. - Synapse differentiation by the tionfunction process in CALEB-deficient of supernumerary mice byclimbing patch Lrp4-agrin complex and by CALEB clamp recordings. In addition, the elimina to be distorted in the absence of CALEB. Development and plastic changes of synaps- fiber synapses on Purkinje cells was found- es are of critical importance for the connec- natal stages suggesting a transient role for tivity and functionality of the brain. Despite CALEBThese deficits in synapse are maturation.restricted to early post

160 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Role of Ig cell adhesion proteins of the CAR subfamily

The group of Dr Rathjen has also a long standing interest in the functional characterization of cell adhesion proteins of Selected Publications - rently, we are concentrating on the CAR Hannes Schmidt, Stefanie Peters, Katharina Frank, Lai Wen, Robert Feil, and Fritz G. Rathjen (2016) Dorsal root ganglion axon bifurcation tolerates increased cGMP lev- (coxsackievirusthe immunoglobulin and (Ig)adenovirus superfamily. receptor) Cur els: the role of phosphodiesterase 2A and scavenger receptor Npr3, Eur. J. Neurosci. 44: 2991-3000. - Gohar Ter-Avetisyan, Fritz G. Rathjen, and Hannes Schmidt (2014) Bifurcation of ax- subgroup of IgCAMs which is composed of ons from cranial sensory neurons is disabled in the absence of Npr2-induced cGMP perfamily) and CLMP (CAR-like membrane signaling. J. Neurosci. 34: 737-747. CAR itself, BT-IgSF (brain and testes Ig su protein). Our genetic studies reveal that Aline Winkelmann, Nicola Maggio, Joanna Eller, Gürsel Caliskan, Marcus Semtner, these components control a variety of cel- Ute Häussler, René Jüttner, Tamar Dugladze, Birthe Smolinsky, Sarah Kowalczyk, Ewa lular and organelle functions ranging from Chronowska, Günter Schwarz, Fritz G. Rathjen, Gideon Rechavi, Carola Haas, Akos Kulik, Tengis Gloveli, Uwe Heinemann, Jochen Meier (2014) Changes in neural net- the peristaltic of the intestine and of the ureter to the embryonic development of the René Jüttner, Dirk Montag, Rogerio B. Craveiro, Aleksei Babich, Petra Vetter, and Fritz heart and male fertility. work homeostasis trigger neuropsychiatric symptoms, J. Clin. Invest. 124 : 696-711. premature stages during postnatal development of the cerebellum in the absence of G. Rathjen (2013) Impaired presynaptic function and elimination of synapses at Regulation of calcium waves in Christopher Patzke, Klaas Max, Joachim Behlke, Jadwiga Schreiber, Hannes Schmidt, developing retinae in the absence ArminCALEB Dorner,(CSPG5/neuroglycan Stephan Kröger, C), Mechthild Eur. J.Neurosci. Henning, 38: 3270-3280.Albrecht Otto, Udo Heinemann of IgCAMs and Fritz G. Rathjen (2010) The Coxsackie-Adenovirus receptor (CAR) reveals complex homophilic and heterophilic binding activities on neural cells, J. Neurosci. During early phases of development a coarse pattern of connectivity is estab- 30, 2897-2910. lished in the brain and with the help of neuronal activity the coarse connections Group Leader Prof. Dr. Fritz Rathjen developing visual system correlated bursts ofare action then fine-tuned.potentials detected For example, as calcium in the Scientists Laura Wolf* waves propagate among neighboring reti- Dr. Rene Jüttner Azadeh Mohtashamdola- nal cells. These calcium waves are believed Dr. Philip Tröster* tshahi* to provide cues within their activity im- Dr. Gohar Ter-Avetisyan Stefan Marre*

segregation. Similar spatio-temporal pat- PhD students Technical Assistants ternsportant of forcalcium the refinement waves have of been eye-specific reported Alexandre Dumoulin Anne Banerjee* in other regions of the nervous system as Florian Hetsch* Madlen Driesner well suggesting that this type of neuronal Hanna Langhorst* Mechthild Henning Claudia Matthäus* neuronal circuits in target areas. We hy- Johannes Oppermann* Secretariat activity is important for the refinement of- Laura Schütz Birgit Cloos (part time) ate intimate cell-cell contacts modulate the Philip Tröster* frequencypothesize thatand somepattern IgCAMs of calcium which waves. medi Animal Attendant We have therefore recorded spontaneous Graduate and Undergraduate Karola Bach (part time) calcium waves in the developing retina in students the absence of different cell adhesion mol- Shoaib Afzal* *Part of the time reported TAG1, DSCAML1 and others (Figure 2). ecules of the IgSF such as NCAM, L1, CAR, Ina Sohm*

MDC Research Report 2016 161 DISEASES OF THE NERVOUS SYSTEM Photo: NeuroCure Photo:

Dietmar Schmitz

Role of GPRIN1 on kainate-receptors function

Our mental capacities, the possibility of These neurotransmitters have a large spec- making coherent decisions, solve complex trum of cellular actions, which are achieved - problems, and display elaborated behav- ingly, one neurotransmitter can activate iors, rely on the specialized way in which diverseby activation sets of of receptors specific receptors. that may Interestcontrib- the brain treats information. Brain cells use ute differently to neuronal communication. a wide range of neurotransmitters to com- Glutamate, for example, activates ionotropic receptors (leading fast excitation) municate. as well as metabotropic receptors (act as modulators). There are three main types of ionotropic glutamate receptor, NMDA, AMPA and kainate receptors (KARs). While AMPA and NMDARs mediate most excitatory postsynaptic currents, KAR contribution to excitability is localized to discrete synapses, and they serve other functions. These range from neuronal development to excitability and modulation of neurotransmitter release. They are found prepostand perisynaptically. Many of KARs functions are mediated by activation of heterotrimeric G proteins. The complexity of functions and polarized localization of KARs are dominated by intricate protein associations. We want to identify KAR-interacting proteins that are of key importance to understand their physiology in the healthy and diseased brain, and to develop targeted pharmacology. One salient characteristic of KARs is their slow kinetics. These receptors share structural conformation and sensitivity to many drugs with AMPARs, but their currents differ in fundamental ways. KARs yields excitatory postsynaptic currents (EPSCs) with smaller amplitude, and slower rise and Figure1- Schematic structure of Gprin1 protein, showing decay time than AMPARs. To understand membrane targeting region, and main phosphorylation re- difference in kinetics of these receptors gions. Variants that interact with different proteins Gao/i studies of recombinant DNA expressed in or GluK2 are shown (A). Currents elicited by application heterologous systems have been of great fast of glutamate (10mM for 200 ms) to excised outside-out help. However, currents from recombinant patches of Hek298 cells, transfected with GluK2 (blue) or KAR DNA do not recapitulate endogenous GluK2 and Gprin1 (red) (B) (t-test- ** p< 0.01).

currents. Indeed, in heterologous systems, 162 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 2- Photography showing that Gprin1 is enriched in the cerebral cortex, the hippocampus, the cerebellum and the olfactory bulb in mice (A). Drawing showing that obliteration of Gprin1 leads to brain volume reduction of about 30% compared to wildtype (B). The hippocampal formation is rounder in Gprin1 mutants and the volume is reduced (C). This can be reflected in a thinner CA1 layer, with CA1 cells smaller and unhealthy (D). (t-test- **p<0.01 or *p<0.05)

AMPARs and KARs currents are comparable. shown by Sholl analysis(Fig.3D). These re- This suggests that KARs synaptic proteins sults are important since they validate the confer them particular properties. Exten- Gprin1 mutants as a model to study KARs sive elucidation of these proteins is needed. physiology in brain slices and in vivo. We found a KAR-interacting protein, the G protein-regulated inducer of neurite out- KARs suggests that they associated with a growth 1 (Gprin1) that is associated with setIn summary,of synaptic the proteins diversity that of confer functions them of neurite development and interacts with GTP bounded Gao/i, the inhibitory G alpha subunit. We used immunoprecipitation isspecific essential properties to generate and modesadvances of actions.in this and found the region that binds to GluK2 Extensive identification of these proteins (Fig.1A). We use a system to determine the a partner of GluK2. Association with Gprin1 kinetics of GluK2 in response to glutamate. changesfield. Our the data kinetics support of a therole receptors for Gprin1 and as We use a fast perfusion tool, installed on an Gprin1 obliteration affects postnatal brain electric piezo, allowing fast application of development, possibly by orchestrating the glutamate to patches of cells that express function of neurotransmitter receptors, in- GluK2 recombinant DNA. We found that cluding KARs. Next, we will elucidate the Gprin1 slows down the rise time of GluK2 role of Gprin1 and endogenous KARs func- currents (Fig.1B). This effect may account tion and its potential role in neuronal de- in part for the slow kinetics observed in velopment. hippocampal GluK2. - ation we generated a transgenic mouse, whichTo extend lacks these Gprin1 findings (Gp1-/-). to the We native found situthat Selected Publications Gprin1 is expressed mainly in the hippo- Stempel AV, Stumpf A, Zhang HY, Ozdogan T, Pannasch U, Theis AK, Otte DM, Wojtalla campus, cortex, cerebellum and olfactory

A, Racz I, Ponomarenko A, Xi ZX Zimmer A, Schmitz D. Cannabinoid Type 2 Receptors Gprin1 leads to a decrease in brain volume, Mediate a Cell Type-Specific Plasticity in the Hippocampus. Neuron 2016; 90:795- withbulb (Fig.2A).all regions Importantly, measured (hippocampus, obliteration of 809 Johenning FW, Theis AK, Pannasch U, Ruckl M, Rudiger S, Schmitz D. Ryanodine cortex and cerebellum) reduced in the same Receptor Activation Induces Long-Term Plasticity of Spine Calcium Dynamics. PLoS proportion (Fig. 2B). The changes for the Biol 2015; 13:e1002181. Beed P, Gundlfinger A, Schneiderbauer S, Song J, Bohm C, Burgalossi A, Brecht M, Vida hippocampus are shown in detail (Fig.2C). I, Schmitz D. Inhibitory gradient along the dorsoventral axis in the medial entorhinal Albeit the size reduction, all layers of the cortex. Neuron 2013; 79:1197-1207. hippocampus are present. Finally, there are MaierDugladze N, Tejero-Cantero T, Schmitz D, Whittington A, Dorrn AL, MA, Winterer Vida I, GloveliJ, Beed PS,T. Segregation Morris G, Kempter of axonal R, and Poulet less cells in the CA1 and their cell body is JF,somatic Leibold activity C, Schmitz during D. fast Coherent network phasic oscillations.Science2012;336:1458-1461 excitation during hippocampal ripples. smaller (Fig.2D). Gprin1 accumulates in the membrane of cultured hippocampal Liu KS, Siebert M, Mertel S, Knoche E, Wegener S, Wichmann C, Matkovic T, MuhammadNeuron 2011; K, 72:137-152 Depner H, Mettke C, Buckers J, Hell SW, Muller M, Davis GW, Schmitz neurons(Fig.3A). We analyzed the morphology of neurons of the hippocampus in CA1 D*,Beed Sigrist P, Bendels SJ*. RIM-binding MH, Wiegand protein, HF, Leibold a central C, Johenning part of the FW, active Schmitz zone, D. is Analysisessential of for area of Gp1-/- mice(Fig.3B). There is no dif- neurotransmitter release. Science 2011; 334:1565-1569 | *corresponding authors ference in neurite length but an increased angle of bifurcation between sister branch- excitatory microcircuitry in the medial entorhinal cortex reveals cell-type-specific differences. Neuron 2010; 68:1059-1066 es in the apical dendrites(Fig.3C). This is interesting because we know that there are Group Leader Scientist less neurons in CA1 and the brain is smaller. Prof. Dr. Dietmar Schmitz Dr. Nutabi Camargo But it may indicate that the neurons that (Guest Charité) manage to survive grow normal neurites, as

MDC Research Report 2016 163 DISEASES OF THE NERVOUS SYSTEM Photo: Sigrid Schnögl/MDC Sigrid Photo:

Erich Wanker

Proteomics and Molecular Mechanisms of Neurodegenerative Diseases

My group’s research focusses on ‘Neu- roproteomics’, the protein-based investi- relevant interactions between the triple A ATPaseal., Cell, VCP/p972005). Recently, and an adaptorwe identified protein highly that gation of neurodegenerative diseases. In effects a fundamental structural change in detailed, hypothesis-driven studies we ad- VCP from a homohexamer to a heterodimer dress mechanisms of protein misfolding and with far-reaching functional implications aggregation, with the aim of understanding (Arumugham et al. 2016). We are constantly - the molecular mechanisms of neurodegen- tion and validation of protein-protein inter- eration in Huntington’s, Alzheimer’s and developing new methods for the identifica- further diseases causally related to the mis- cence-based co-immunoprecipitation assay actions, most recently DULIP, a dual lumines folding of proteins. In particular, we aim to for interactome mapping in mammalian cells, (Trepte et al., J Mol Biol, 2015.) elucidate the molecular principles by which abnormally folded proteins, their complexes As interactomics core facility of the Max and aggregates cause cellular toxicity and Delbrück Center, we offer screenings on our platform to interested colleagues. neuronal dysfunction. More recently, we have started new lines of - translational research establishing methods to detect disease-relevant misfolded protein identifyIn our efforts and tocharacterize promote translation modulators of baof species in patient biosamples. These inves- proteinsic research misfolding into benefits cascades for in patients,disease (Eh we- tigations are directed at the development of predictive disease markers which are a Bieschke et al., Nat Chem Biol, 2011). We prerequisite for the clinical investigation of havernhoefer previously et al., Nat demonstrated Struct Mol Biol, that 2008; ex- new disease-modifying therapies targeting panded polyglutamine (polyQ) sequences neurodegeneration before symptoms of ir- trigger misfolding and aggregation of N- reversible neuronal damage arise. terminal huntingtin fragments in vitro and in vivo IntegraMent – Interactome net- et al., Cell, 1997). works and perturbed cellular func- (Scherzinger et al., Cell, 1997; Davis tions in schizophrenia - An “integrated understanding of causes ogy, in particular using interactomics approaches,Our second which field areof activity also mainly is systems applied biol to - neurodegenerative disease processes. Pre- Ment,and mechanisms funded by ofthe mentalFederal disorders” Ministry of is viously, we developed an automated yeast Educationsought in the and collaborative Research under project its systemsIntegra two-hybrid (Y2H) system, which we used medicine initiative, e:med. Our role in this to generate a focused protein-protein international consortium is to map interac- action network for the huntingtin protein tions of proteins involved in this family of diseases. Such efforts are crucial to the al., Mol. Cell, 2004), as well as a large inter- understanding of these underresearched actionrelevant map to Huntington’sof the human diseaseproteome (Goehler (Stelzl et diseases, since the cellular functions of

164 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 1: An interaction network of proteins involved in schizophrenia. The network is based on a list of 286 protein coding genes implicated in the pathogenesis of the disease and contains 4,302 proteins forming nearly 8,000 connections. The network is significantly more connected than randomly generated networks. Red: Protein coding genes primarily associated with schizophrenia. Blue: First degree interacting partners of proteins encoded by schizophrenia-associated genes.

the involved proteins, their relationships, tional methods to identify disease-relevant and their roles in disease mechanisms are clusters in the network as the basis for in- depth validation studies that are to follow. been pursued: First, the computational pre- dictionlargely unknown.of interaction Three networks specific for goals proteins have ERA-NET Neuron and Berlin In- with known relevance to disease. Second, stitute of Health: Collaborative the systematic validation of the involved translational research to fight protein-protein interactions. Third, the in- Alzheimer’s disease vestigation of selected interactions in dis- - ease models and patient biosamples. This nated since, two translational consortia: The systems-biology driven approach will allow ERA-NETIn 2013-14 Neuron we launched, programme and fundshave coordiefforts to establish new methods for the discovery – functional protein interaction networks of novel drug candidates or diagnostic ap- andthe consortiummay also identify to define novel – for biomarkers the first time for the molecular diagnosis of mental diseases - that currently can only be diagnosed via as- aly,proaches we study for Alzheimer’saggregated amyloid-beta,disease. Together one ofwith the researchers polypeptides in France,suspected Belgium of triggering and It

whichsessment are of implicated psychiatric in symptoms. the pathogenesis We first one hand, we are interested to target it ef- ofcompiled schizophrenia, a list of 286from protein-coding which we generated genes fectivelyneurodegeneration with chemical in Alzheimer’s.substances to On create the - work connecting 4,302 proteins via 7,991 approaches to detect the molecular changes interactionsa schizophrenia-specific (Figure 1). We interaction integrated gene net ittherapeutic causes before effects; patients on the show other, symptoms we develop of expression datasets and applied computa- dementia. This is one of the most important

MDC Research Report 2016 165 DISEASES OF THE NERVOUS SYSTEM

Figure 2: Electron microscopy image of amyloid-beta aggregates extracted from postmortem brain. Magnetic beads in combination with special antibodies are used to extract aggregates of the amyloid-beta polypeptide from biological samples. The large sphere is one of the beads in the reaction, floating above it, as it were, we see thin fibrils of amyloid-beta.

goals pursued by Alzheimer researchers in academia and industry at the moment: Find- ing prognostic disease markers is hoped to diseasein Alzheimer’s. markers Systematicor chemical approaches modulators are of facilitate pre-symptomatic treatment of the flanked by in-depth studies of potential new disease. We have laid groundwork with re- disturbedInvestigating molecular protein profiles aggregation in the disease. a way to extract aggregated peptide from and neurodegeneration related to postmortemgard to the amyloid-beta brain samples peptide in order by finding to de- Huntington’s disease in fruit flies termine those of its properties that are relevant to pathogenesis (Figure 2). disease. This means neurological distur- More systematic, large scale experimen- bancesHuntington’s adversely disease affect is movement a neuromuscular abilities tation has been started in the framework of patients while also causing progressive psychiatric and cognitive symptoms. Under- Health: Together with six other basic and lying these symptoms is neurodegeneration, clinicalof the newlyresearch founded groups Berlin at the Institute Max Del of- which in turn is caused by a gene mutation brück Center and Charité University Medi- in the huntingtin protein. Mutated hunting- cine, we have started systems biology ef- tin contains an elongated stretch of over 40 repeats of the amino acid glutamine, whereas

forts to define changes in protein networks

Figure 3: Immunofluorescence images of Drosophila melanogaster brains. Brains of 9-days-old flies without (left) and with induction of mutated huntingtin expression for 6 days (right). Mutated huntingtin aggregates visualised as green dots.

166 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

the unmutated gene has under 40. This extended polyglutamine stretch causes the Patents / Patent applications protein to misfold and aggregate. A useful system to study the disease mechanism of huntingtin aggregation and neurodegenera- Wanker, E.; Wiglenda, T.; Babila, J.; Böddrich, A.; Schmidt, M.; Schiele, F. (2011) AzoEnhancers compounds of protein reduce aggregation, formation WO2011020883and toxicity of amyloid beta aggregation Expression of mutant huntingtin with 96 Wanker,intermediates, E.; Wiglenda, WO2011117305 T.; Babila, J.; Boeddrich, A.; Schmidt, M.; Schiele, F. (2011) polyglutaminetion is the fruitfly repeats Drosophila using the melanogaster. conventional UAS-GAL4 expression system results in an Selected Publications early expression in the Drosophila lifecycle. A. Arumughan, Y. Roske, C. Barth, L. L. Forero, K. Bravo-Rodriguez, A. Redel, S. This early pan-neuronal expression leads Kostova, E. McShane, R. Opitz, K. Faelber, K. Rau, T. Mielke, O. Daumke, M. Selbach, E. to formation of aggregates in the larvae and Sanchez-Garcia, O. Rocks, D. Panáková, U. Heinemann, and E. E. Wanker, “Quantitative pupae stage which results in neurodegenera- - A.interaction Buntru, P. mapping Trepte, K. reveals Klockmeier, an extended S. Schnoegl, UBX domain and E. E.in Wanker, ASPL that “Current disrupts Approaches functional gate the molecular sequence of events that p97 hexamers.,” Nat. Commun., vol. 7, p. 13047, Oct. 2016. leadstion even to the in onsetfreshly of hatched neurodegeneration, flies. To investi we P. Trepte, A. Buntru, K. Klockmeier, L. Willmore, A. Arumughan, C. Secker, M. Zenkner, employed an inducible expression system Toward Quantitative Mapping of the Interactome.,” Front. Genet., vol. 7, p. 74, 2016. L.Biol., Brusendorf, 2015. K. Rau, A. Redel, and E. E. Wanker, “DULIP : A Dual Luminescence-Based M.Co-Immunoprecipitation Stroedicke, Y. Bounab, N.Assay Strempel, for Interactome K. Klockmeier, Mapping S. Yigit, in Mammalian R. P. Friedrich, Cells,” G. J. Mol. called GeneSwitch. Induced expression of Chaurasia, S. Li, F. Hesse, S.-P. Riechers, J. Russ, C. Nicoletti, A. Boeddrich, T. Wiglenda, brain.mutated To huntingtin visualise huntingtin for six days aggregates in adult flies in C. Haenig, S. Schnoegl, D. Fournier, R. K. Graham, M. R. Hayden, S. Sigrist, G. P. Bates, J. Priller, M. A. Andrade-Navarro, M. E. Futschik, and E. E. Wanker, “Systematic led to the formation of aggregates in the fly was used, showing them as green spots (Fig- fly heads, immunofluorescence microscopy C.interaction Kayatekin, network K. E. S. Matlack,filtering identifiesW. R. Hesse, CRMP1 Y. Guan, as aS. novel Chakrabortee, suppressor J. Russ, of huntingtin E. E. ures 3). We also investigated the relation be- misfolding and neurotoxicity.,” Genome Res., vol. 25, no. 5, pp. 701–13, 2015. tween the quantity of aggregates and their Wanker, J. V Shah, and S. Lindquist, “Prion-like proteins sequester and suppress

the toxicity of huntingtin exon 1.,” Proc. Natl. Acad. Sci. U. S. A., vol. 111, no. 33, pp. 12085–90, Aug. 2014. sixtoxicity days inof amutant life span Huntingtin assay. Healthy expression fruit flies die live to about 100 days, while flies undergoing Group Leader Daniela Kleckers* Prof. Dr. Erich Wanker Susanne Köppen* atA anplatform age of about for 38large-scale days. investi- Susanne Kostka* gation of protein-protein interac- Scientists Sabrina Kruse* tions facilitates regional and inter/ Dr. Annett Böddrich Nancy Neuendorf national collaboration Dr. Alexander Buntru Kirstin Rau* - Dr. Annika Deckert* Alexandra Redel ful to enable new discoveries of molecular Christian Hänig Franziska Wiedemann, trainee* connectionsInteractomics in approaches pathways, arecellular highly compart power- Dr. Franziska Schindler Martina Zenkner ments or disease mechanisms. New cutting- Christopher Secker (guest – Charité) edge assays are constantly being estab- Dr. Martin Strödicke* Students* lished in our group and offered as service or Dr. Philipp Trepte Simon Berberich in the framework of collaborative projects Dr. Thomas Wiglenda Svenja Bolz to researchers in and outside the Max Del- Nea Dierolf PhD students Christin Donner - Mohammed Ahmed (DAAD Fellow) Nicole Groenke brück Center. Recently, we reported DULIP:- Kathrin Andrich Regine Hasenkopf pingA Dual in Mammalian Luminescence-based Cells (Trepte Co-Immuno et al., J Mol Anup Arumughan Melanie Humpenöder precipitation Assay for Interactome Map Lisa Maria Diez protein-protein interaction screening meth- Sha Jin Loni Klaus odBiol, of 2015).high utility, DULIP e.g., is of a detecting second-generation the effects Konrad Klockmeier HannahIsabelle JansenNiederlechner of point mutations on interaction strength. Nadja Schräpel Simona Kostova Anne Tempelmeier FRET-based assays for researchers inter- NadineYoung-In Strempel Ko Lindsay Willmore estedWe can to runuse yeastnetwork two-hybrid, biology approaches. DULIP and Anne Steinhof We put great effort in validation and bench- Secretariat marking of our systems to provide high-con- Technical Assistants Erika Pisch Lydia Brusendorf Please approach us if you are interested in Sabrina Golusik* (guest – Charité) Coordinator screeningfidence, highly your reliableprotein interactionof interest fordatasets. inter- Gerlinde Grelle* Sigrid Schnögl action partners. So far we have performed Anne Hahmann, trainee* screenings for over 20 collaborating labs in Christin Hesse, trainee* * part of the period reported Berlin, Germany, Europe and the USA.

MDC Research Report 2016 167 DISEASES OF THE NERVOUS SYSTEM Photo: David Ausserhofer/MDC Photo:

Niccolò Zampieri

Development and Function of Neural Circuits

A fundamental question in neuroscience regards the mechanisms used during development to wire the nervous system. Just by peeking at the anatomy of the nervous system it is evident that neuronal subtypes are assigned highly stereotyped positions, where neurons grouped in the same structure not only share positional coordinates but typically receive connections from similar inputs and send projections to common targets. Precise spatial organization appears to represent a major strategy to simplify the problem of connectivity. In addition, specific molecular cues, or labels, have a prominent role in wiring neural circuits and indeed several families of molecules involved in axonal guidance and synapse Figure 1: Digital reconstruction of motor neuron formation have been identified. However, organization. Upper panels, immunohistochemical the contributions and interplay of these two analysis a single section of a mouse lumbar spinal cord strategies - neuronal location and label - showing motor neuron subtypes identified by expres- are still not well defined. Understanding the sion of the transcription factors Isl1/2 (LMCm) and Hb9 (LMCl). Lower panels, 3D reconstructions of the mechanisms that control neuronal position- organization of LMCm and LMCl neurons of control and ing is therefore an important step in defining mutant mice with defect in motor neuron organization. the logic that drives neural circuit assembly.

model system the mouse spinal cord where motorIn order neurons to tackle controlling this question the we activity use as of a

structures, termed pools, which occupy ste- reotypedlimb muscles and invariant are clustered positions into across defined in-

movements, muscle groups with different biomechanicaldividuals. In order functions to generate have to coordinated be activated in precise patterns. Thus, the selectivity with which sensory neurons and interneurons form connections with different motor pools is critical for the timely activation of muscle contraction. The mechanisms that

168 MDC Research Report 2016 DISEASES OF THE NERVOUS SYSTEM

Figure 2: Rabies mapping of motor circuits organization. Strategy, representative image and mapping of pre-motor interneurons in the spinal cord labelled via rabies virus transsynaptic tracing.

dictate these patterns of connectivity re- - main largely obscure. Recent studies have tively study motor neuron organization, we areadhesion developing molecules. novel Inmethods order toto quantitagenerate to motor neuron connectivity is achieved, three-dimensional maps of motor neuron inshown part, that through the specificity the ability of sensory of sensory neuron ax- positions in the whole spinal cord (Figure ons to project to precise areas within the spinal cord. These results suggest that the assembly of motor circuits by taking advan- spatial organization of motor neurons sub- tage1). In of a parallelviral and approach, anatomical we aretracing analyzing tech- types might be indeed a critical constraint niques to highlight connectivity patterns between motor neurons and their synaptic of motor neuron settling position in deter- partners in control and mutants mice to de- for sensory input specificity. The influence an indication that the elaborate positional the assembly of neural circuits (Figure 2). organizationmining sensory of motor input specificityneuron cell provides bodies fine the role of precise cell body position for has critical roles for the formation of selective connections in motor circuits. Selected Publications

Zampieri N., Jessell TM., Murray A. (2014). “Mapping sensory circuits by anterograde - herin/catenin adhesive signaling as a key Demireva EY, Shapiro L, Jessell TM, Zampieri N. (2011). “Motor neuron position and playerIn previous in the workcontrol we of identifiedmotor neuron the cadpo- transsynaptic transfer of recombinant rabies virus.” Neuron, Feb 19; 81(4):766-778. Ciatto C, Bahna F, Zampieri N, VanSteenhouse HC, Ranscht B, Katsamba P, Ahlsen G, Posy sitioning. Classical cadherin expression topographic order imposed by β- and γ-catenin activities.” Cell, Oct 28; 147(3):641-651. distinguishes motor neurons at a molecu- S, Vendome J, Jessell TM, Honig B, Shapiro L. (2010). “T-cadherin structures reveal a novel lar level and perturbation of cadherin adhesive function prevents segregation and cadherin adhesive binding mechanism.” Nat. Struct. Mol. Biol. Mar;17(3):339-47.

document a key role for cadherin/catenin Start of the Group: September 2014 inclustering orchestrating of motor motor pools. pools These organization. findings However, it is still not clear how cadherin Group Leader signaling controls motor neuron position- Dr. Niccolò Zampieri Postdoc ing and whether other signaling systems (Joint with Neuro-Cure) Felipe Sassi are involved. By combining molecular and genetic approaches, ongoing experiments PhD students Technical Assistant Carola Dewitz contributions of different adhesion systems Stephan Dietrich focusingare defining on theclassical molecular cadherins mechanisms and more and AdministrativeIsabelle Werner Assistant recently extending the analysis on members Sophie Skarlatou Jasmin Bonkowski of immunoglobulin superfamily of cell-cell Sofia Pimpinella

MDC Research Report 2016 169 170 MDC Research Report 2016 Berlin Institute for Medical Systems Biology

Nikolaus Rajewsky

MDC Research Report 2016 171 Berlin Institute for Medical Systems Biology

Nikolaus Rajewsky

revalent and devastating diseases tumors and stem cells. The Bioinformatics such as cancer, cardiovascular dis- Platform is headed by Altuna Akalin who P ease, metabolic diseases and neurodegenerative disorders are usually multi- Light Mircoscopy Platform was established factorial diseases involving many genes. To bystarted Andrew in spring Woehler 2014. in 2016. In addition, a new extend our understanding of the interplay of functions involving different genes in Scientific group leaders at BIMSB health and disease, an important goal of systems biology is to analyze gene regula- - tory networks. To this end we collect ge- standing new Junior and Senior group lead- nome-wide quantitative data to integrate ersBIMSB since was 2009. fortunate First to to recruit join was several Markus out them into predictive models. This challenge Landthaler ‚RNA Biology and post-tran- is at the heart of Medical Systems Biology scriptional regulation‘. Junior groups inves- and is by nature highly interdisciplinary, tigate aspects of gene regulatory mecha- combining molecular and cellular biology, nisms and use model systems which were biochemistry, mathematics, physics and en- not established on campus before: Marina gineering. The past ten years have particu- - larly highlighted that cells employ multiple sophisticated levels of gene regulation, sun,Chekulaeva ‘Gene regulation ‘Non coding and RNA’s cell fate and decision mecha which include organization of chromo- nisms of cytoplasmic regulation’, Baris Tur-- somes in the nucleus, epigenetic gene regu- ogy of neural tissue differentiation in Dro- lation, transcription, post-transcriptional in C. elegans’, Robert Zinzen ‘Systems biol gene regulation and post-translational - sophila’, Jan Philipp Junker ‘Quantitative- els interact and regulate each other and are Developmental Biology’, Stephan Preibisch modifications.involved in disease Importantly, development. all of these At levthe Schwarz‘Microscopy, ‘Evolutionary Image Analysis and Cancer and Genom Model- ing of Developing Organisms’ and Roland

MDC, the grand challenge and Scientific ics’. In 2017, two new Junior groups will- theseMission levels of the of gene“Berlin regulatory Institute controlfor Medical into join BIMSB: Dario Lupianez ‘Epigenetics comprehensiveSystems Biology” and (BIMSB) predictive is to models integrate of Seniorand Sex Group Development’ leaders Uwe and Ohler Christoph ‘Computa Lip- how gene regulatory changes are linked to pert ‘Statistical Genomics’. In addition to- phenotypes. bo ‘Epigenetic regulation and chromatin ar- tional regulatory genomics’, and Ana Pom Scientific Technology Platforms She is a computational biologist studying RNAchitecture’, structure. Irmtraud Benedikt Meyer Obermayer joined in ‘Com2015.- High throughput technologies are key for - putation and Theory’ and Scott Lacadie many research aspects at BIMSB. We have- groups.‘Functions of enhancer during zebrafish de therefore established essential Scientific velopment’ are Delbrück Fellows in BIMSB andTechnology integrated Platforms these technologies for ‘omics’-technolo in scien- Publications gies. The newly recruited PIs have installed - directstific projects, the Genomics often inplatform collaboration which offers with ing and highly-cited papers. 256 publica- diverseother labs applications, at the BIMSB/MDC. including Sascha genome Sauer and tions/citedBIMSB scientists more havethan published13500 times. outstand Often, transcriptome sequencing and assembly, - - ting edge methods to quantify the metabo- orativeBIMSB spirit. publications For example, feature the two Landthaler or more lomeand ChIP-Seq. as well as Stefan the proteome, Kempa develops with a specut- andBIMSB Rajewsky authors, labs demonstrating published a study the collabin Cell cial focus on aspects of the metabolism of (Rybak-Wolf et al. 2014) where they discov-

172 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY (copyright MDC) (copyright

Additional Activities

- gram to train the next generation of Sys- temBIMSB Biologists. has established The program an Exchange allows proPhD students to work on collaborative research

at the Center for Genomics and Systems Bi- ered a new class of binding sites for a key project be tween BIMSB labs and partners enzyme in small RNA biology (Dicer). Other - recent highlights: The Ohler lab discovered esology the at annual New York Berlin University. Summer In Meeting addition and to that human promoters are intrinsically bi- Summerseminar andSchools, lecture as series,well as BIMSB joint organizretreats directional (Duttke et al. Mol Cell 2015). with the MRC-Clinical Sciences Centre (UK). The Pombo lab found that methylation of - marks early transcription in mammalian treatsIn spring to discuss and fall science, 2015, BIMSB technologies, PIs and col all- RNA POLII non-consensus lysine residues laborationsBIMSB teams, and respectively, strategies. held 2 day re leading journals were published by the Pombocells (Dias lab (Nat et al Rev eLIFE Mol Cell 2015). Biol Reviews 2015) and in Grants, Awards & Peer reviews the Rajewsky lab (Nat Rev Genetics 2015). Other key publications originate from the Landthaler lab (Murakawa et al Nature - Comm. 2015), Rajewsky lab (Rybak-Wolf et edIn for addition extramural to the funding. start-up Since funding its start, of al. Mol. Cell 2015), Delbrück Fellow Ober- BIMSB, the labs have successfully compet € additional funds from the DFG, BMBF, EC, et al. Genome Biol 2015) and the Ohler lab BIMSB labs have acquired a total of 32 Mio mayer with several BIMSB labs (Mackowiak- are: ERC starting grants (Tursun, Junker), theNIH, Gottfried GIF, and others.Wilhelm Some Leibniz of the Prize highlights (Ra- already(Calviello 6 patentet al. Nat applications Methods with2015). focus In ad on jewsky), Human Frontiers Program grant newdition technologies to publications, and diagnostic BIMSB groups principles filed (Rajewsky), the German Epigenome Con- of molecular components of gene regulatory networks. bioinformatics infrastructure in Germany). Model of the new sortium ‘DEEP’, and de.NBI (BMBF funded BIMSB building in the The new BIMSB building center of Berlin

the Campus of the HU Life Sciences in the centerThe new of Berlin, building will for accommodate BIMSB, located up to 25 on research groups, approximately half experimental and half computational. The building is currently in construction and will

support high-tech labs, flexible work spaces and communication areas. BIMSB groups- edwill by be competitively moving in 2018. selected The art architectural works for theand insidescientific and concept outside will of the be building.complement The overall theme for the art projects is to ‚break walls‘ between disciplines and people.

MDC Research Report 2016 173 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Altuna Akalin

BIMSB Bioinformatics / Mathematical Modelling

Recently, high-throughput methods enable regions with abnormality and classify samples searchers to do a much more thorough charac- according to those abnormalities. We are using statistical and machine-learning methods terization of genomes and epigenomes. Complex to that end, and have developed and have been diseases like cancer can only be understood by computational analysis of such high-throughput - maintaining the popular software, “methylKit”. data. The amount of data generated calls for - cerIn addition, samples we(SNPs, are alsoCNVs focusing and gene other expression) genom application of statistics and machine-learning in toics developprofiling computational data frequently methods generated to forclassify Can a broad scale. In the light of this, our research interests are in direction of understanding the throughput genomics data sets and develop methodstumor subtypes. to investigate In summary, epigenomic we analyze and high- ge- genetic and epigenetic control mechanisms of nomic abnormalities and differences in vari- cellular differentiation and complex diseases ous normal and cancer cell types. through computational analysis and integration of high-throughput datasets, such as microar- Computational analysis of long- range gene regulation rays and next generation sequencing. Ultimately, Inga Patarcic we want to discover genomic and epigenomic Gene regulation is a complex event where anomalies and interaction between them that multiple partners interact with each other to lead to disease. In addition, we will strive to keep achieve gene regulation. The level of expression of a gene is initially controlled by tran- most up-to-date expertise in computational biol- scriptional regulatory mechanisms, such as ogy and data analysis, and develop tools that enhancers and insulators which can act upon provide analysis techniques that are relevant, their target genes from large distances. Muta- tions on enhancers or structural variations reproducible and well-documented. that disrupt enhancer target pairing are asso-

(Epi-)genomics and disease full picture on long-range gene regulation, one Katarzyna Wreczycka, Jonathan Ronen, mustciated take with into various account diseases. all the Indifferent order towhole- have Alexander Gosdschan genome data sets that provide information on Discovering which abnormalities lead to com- the location and effect of the long-range regu- plex diseases such as cancer remains as a chal- lators. We are working on to develop methods lenge. To complicate the matters, most cases that take into account multi-level gene regu- are unique to their selves and do not allow for lation information (CAGE, RNA-seq, Histone simple all-embracing modeling of the disease. - However, our previous work has shown that certain types of cancers share mutations on long-rangemodifications, regulatory DNA methylation) elements. to more pre cisely define the location and activity of the new class of cancers are distinguished by cer- Omics data integration and pathways relating to DNA modifications. These machine learning approaches which paves way for different approaches in Vedran Franke, Katarzyna Wreczycka, tain abnormalities in their epigenomic profiles, Brendan Osberg developing computational methods to analyze Most successful approaches in modeling com- patient stratification and drug design. We are- plex phenomena, such as complex diseases or

epigenomic profiles of samples to identify re 174 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

gene regulation, make use of different types of data. We believe that only through integration of multi-layered data, complex biological processes can be understood in a system- a) Examples of DNA methylation abnormalities in two different - subtypes of acute myeloid leukemia. b) Examples of data integra- ing, integration, analysis and visualization of tion and visualization techniques using in-house software. diversewide level. data This sets. requires We have efficient built a comprehenpreprocess- - sualizing genomics. With the help of this tool sive tool, “genomation”, for integrating and vi RNA-seq, CAGE, etc.) can be integrated and visualizeddiverse genomics over the data regions sets (ChIP-seq, of interest BS-seq, in the genome. The tool can achieve data set spe-

based on genomics data to uncover genome- cific normalization and clustering of regions added more statistics functionality, parallel- izedwide data patterns. processing In the and last optimized couple of visualizayears, we- tion functions for faster graphics generation.

of big datasets can be solved by translating computerIn principal, science integration and statisticsand analysis research problem to biology, and developing those methods fur-

other platforms and groups at MDC, we are aimingther where to use necessary. statistics and In collaboration machine-learning with techniques to uncover relationships in high- dimensional biological data from diverse sources.

Infrastructure for RNA Bioinformatics Bora Uyar, Dilmurat Yusuf, Dan Muntenau, Ricardo Wurmus (in Collaboration with Ohler & Rajewsky Labs) We are part of the “RNA Bioinformatics Cen- Selected Publications

Uyar B, Yusuf D, Wurmus R, Rajewsky N, Ohler U and Akalin A (2017). “RCAS: an - ter” (RBC), which is an initiative supported by Acids Research. doi: 10.1093/nar/gkx120. The“German center Network deals with for developing Bioinformatics methods Infra for BlinRNA K,centric Dieterich annotation C, Wurmus system R, Nikolaus for transcriptome-wide Rajewsky, Markus regions Landthaler, of interest.” Akalin Nucleic A. structure – de.NBI”. DoRiNA 2.0—upgrading the doRiNA database of RNA interactions in post-transcrip- all RNA-related data sets not limited to tran- tional regulation. Nucleic Acid Research , 2015 scriptome analysis but also RNA structure Rampal R *, Akalin A *, Madzo J *, Vasanthakumar A *, Pronier E, Patel J, Li Y, Ahn, et - TET2 Function in Acute Myeloid Leukemia. Cell Reports, 2014 ( * Equal contribu- andanalysis, the analysis prediction of protein-RNA of ncRNA targets,interactions. defi al.tion) DNA Hydroxymethylation Profiling Reveals that WT1 Mutations Result in Loss of Akalin A - Withinnition andthis classificationframework, we of RNAare developing transcripts kit to summarize, annotate and visualize genomic intervals. Bioinformatics , 2014 bioinformatics tools and maintaining local (# Co-corresponding * # , Franke author, V *, Vlahoviček * Equal contribution) K, Mason CE, AkalinSchübeler A*, D.Kormaksson genomation: M, a Li tool S, infrastructure (such as online bioinformatics Garrett-Bakelman FE, Figueroa ME, Melnick A, Mason CE. methylKit: a comprehen- workbench, Galaxy) for RNA biology. We have also developed and currently maintaining the *sive (Co-corresponding R package for the author) analysis of genome-wide DNA methylation profiles. Genome version 2 of DoRiNA database. This database Biology, 2012 Oct 3;13(10):R87 includes RNA binding protein interactions - Group Leader NA predictions from multiple experimental Dr. Altuna Akalin andfrom computational CLIP-seq experiments resources. as We wells also as recent miR- ly developed a toolkit, RCAS, for annotation of Scientists regions of interest on the transcripome. RCAS Dr. Vedran Franke Katarzyna Wreczycka is an automated system that provides dynam- Dr. Brendan Osberg Inga Patarcic Dr. Bora Uyar Technical Assistants that contain transcriptomic regions. On top Dr. Dilimulati Yusufujiangaili Dan Munteanu ofic genomethese, we annotations are integrating for custom tools developedinput files Ricardo Wurmus at Ohler and Rajewsky labs to galaxy work- PhD students bench. This will enable wet-lab researchers Jonathan Ronen Staff to analyze their own data without the need Alexander Gosdschan Sabrina Deter for programming knowledge.

MDC Research Report 2016 175 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: privat Photo:

Marina Chekulaeva

Non-coding RNAs and mechanisms of cytoplasmic gene regulation

Our broad interest is in understanding the Background and overall aims molecular mechanisms that regulate intracellular RNA localization, translation, and Identification of the neurite-localized proteome, transcriptome and translatome stability and roles of RNA-binding proteins - and non-coding RNAs in these processes. partments is a fundamental biological pro- We are particularly interested in the subcel- Targeting proteins to specific cellular com lular aspect of these processes. The subcel- cell growth and migration, synaptic plasticity,cess. and It a underlies vast range formation of other biological of body axis,pro- lular localization and translation of mRNAs cesses. Proper subcellular localization are fundamental to the establishment of the of proteins can be achieved (a) by trans- body axis, cell migration, and synaptic plastic- porting proteins with molecular motors as parts of ribonucleoprotein com- ity, which serves as a foundation of learning plexes (RNPs) or vesicular organelles, and memory. In mRNAs, they are mediated (b) through mRNA localization and local by specific cis-regulatory elements, so called translation or (c) via local translation of zip-codes. These elements are bound by equally distributed mRNAs. - amples for each mechanism have been de- trans-acting factors (RBPs, miRNAs), which scribed in the literature, but it is Specific unclear ex to regulate transport, stability or translation. So what extent each contributes to the overall far, our knowledge is restricted to only a few protein distribution asymmetry. One reason examples of zip-codes and regulatory fac- for this is that most genome-wide studies focused on a particular level of gene expres- tors. We aim to identify these elements on sion (transcriptome, proteome, or trans- a genome-wide scale and dissect molecu- lated transcriptome) or a single cellular lar mechanisms underlying their function. To compartment (e.g. axon without comparison to the cell bodies). Therefore, we aim identify proteins and RNAs that are differen- to determine the extent by which each of tially localized and translated between neu- these separate localization mechanisms rites and soma of neuronal cells, we devel- contributes to the overall asymmetry of oped a neurite/soma fractionation scheme in neuronal protein distribution. For that, we developed an approach allowing us to combination with mass spectrometry, RNAs- separate neurons on subcellular compart- eq, Riboseq and bioinformatic analyses. Using ments – cell bodies (soma) and neurites this approach, we quantified 7323 proteins (axons and dendrites) – and analyzed them and the entire transcriptome and identified by RNAseq, Riboseq and mass spectrometry (MS, Fig. 1A). hundreds of neurite-localized proteins and locally translated mRNAs. Our results demon- Identification of neurite-localized RBPs strate that mRNA localization is the primary and dissecting their roles in local RNA metabolism mechanism for protein localization in neurites RNA localization and local translation are and may account for more than a half of the mediated by RNA-binding proteins (RBPs), neurite-localized proteome. cis-regulatory ele-

interacting with specific 176 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Fig. 1. An outline of the workflow, including accomplished and future objectives: Identification of local proteome, transcriptome, translatome and deciphering roles of RBPs and miRNAs. (A) Scheme illustrating neurite/soma separation using porous membrane and fluorescent images of the induced neurons taken above and below the membrane. Separation is followed by omics analyses to identify local proteome, transcriptome, translatome, epitranscriptome and RBPs. Roles of potential regulators of local RNA metabolism (local RBPs, AGO, miRNAs) are analyzed via CRISPR/Cas- mediated KOs followed by spatial omics analyses to define the regulated step of metabolism and (B) PAR-CLIP and RIP to identify the targets of investigated RBPs.

ments in RNAs. RBPs have the potential to bind to their targets via nearly-perfect bind and regulate hundreds to thousands of matches and AGO cleaves the target, simi- targets. So far, our knowledge is restricted larly to plants. At the same time, it pos- to only a few examples of such RBPs, so - our next aim is to identify neurite-tar- tion about the primary mechanisms of the geted RBPs with potential to regulate miRNAsesses GW182pathway protein,open. Using leaving a combination the ques local RNA metabolism and dissect their of pulldowns, MS, mutational analysis and mechanisms of function. We used a neu- reporter assays in cultured human cells, rite/soma fractionation scheme combined we have shown that Nematostella recruits the CCR4-NOT deadenylation complexes via its W-motifs, thereby inhibiting GW182 with MS for identification of RBPs enriched translation and promoting mRNA decay. knockoutsin neurites (KO)and a for combination dissection of of PAR-CLIP, selected RBPsRNA-IP functions and CRISPR/CAS-mediated (Fig. 1B). RBP of miRNA silencing via the CCR4-NOT com- plexOur findingis conserved suggests for thatat least the 600 mechanism million Deciphering the roles of miRNAs in years and was already present in ancient mRNA localization and local translation animals (Mauri et al., 2016). The miRNA pathway provides an alternative way to recruit a complex of proteins Selected Publications targeting half of all eukaryotic genes and Pamudurti NR*, Bartok O*, Jens M*, Ashwal-Fluss R*, Stottmeister C, Ruhe L, functionto a specific as guides, group byof pairingmRNAs. with miRNAs partially are Hanan M, Wyler M, Perez-Hernandez D, Ramberger E, Shenzis S, Samson M, complementary sites in their target mRNAs, Dittmar G, Landthaler M, Chekulaeva M, Rajewsky N and Kadener S (2017). thereby recruiting a complex of proteins Translation of circRNAs. Molecular Cell, 66(1):9-21.e7 which represses mRNA function. Thus, our Mauri M, Kirchner M, Aharoni R, Ciolli Mattioli C, van den Bruck D, Gutko- vitch N, Modepalli V, Selbach M, Moran Y, and Chekulaeva M# (2016). Con- next aim is to dissect the roles of miRNAs servation of miRNA-mediated silencing mechanisms across 600 million years in subcellular localization and transla- of animal evolution. Nucleic Acids Research nar/gkw792. tion of mRNAs in neurons on a genome- Chekulaeva M# and Landthaler M# (2016)., 45 Eyes (2), on 938-950, translation. doi:10.1093/ Molecular wide level. For that, we knocked down the Cell miRNA pathway in neurons and are testing Pamudurti NR*, Bartok O*, Jens M*, Ashwal-Fluss R*, Stottmeister C, Ruhe L, how RNA localization and local translation Hanan, 63(6):918-25. M, Wyler M, doi:Perez-Hernandez 10.1016/j.molcel.2016.08.031 D, Ramberger E, Shenzis (review). S, Samson M, are affected, using a neurite/soma fraction- Dittmar G, Landthaler M, Chekulaeva M, Rajewsky N and Kadener S (2017). Translation of circRNAs. Molecular Cell, 66(1):9-21.e7 ation scheme. Chekulaeva M*, Mathys H*, Zipprich J, Attig J, Colic M, Parker R, and Filipo- Mechanisms and conservation of miR- CCR4–NOT through conserved W-containing motifs. Nature Structural and . NA-mediated silencing wicz W (2011). miRNA repression involvesRecommended GW182-mediated by Faculty1000 recruitment of Our current knowledge about the mecha- Molecular Biology 8(11): 1218-26. nisms of miRNA silencing is restricted to (*equal contribution; # co-correspondent) higher animals (vertebrates, arthropods, Group Leader PhD students miRNAs mediate translational repression Dr. Marina Chekulaeva Camilla Ciolli Mattioli andnematodes) decay of and their plants. target In mRNAs. higher animals,We and David van den Bruck others have shown that this mechanism re- Postdocs Larissa Ruhe - Esteban Peguero-Sanchez Alessandra Zappulo ily that function downstream of Argonaute Marta Mauri (AGO)lies on tothe recruit proteins the of CCR4-NOT GW182/TNRC6 deadenyl fam- ation complex via conserved tryptophane- Inga Loedige Secretariat

in cnidarian species Nematostella, miRNAs containing motifs (W-motifs). Interestingly, Ines Krock

MDC Research Report 2016 177 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Jan Philipp Junker

Quantitative developmental biology

Biology is a study in contrasts between tages compared to traditional techniques variation and stability. On the one hand, such as in situ generation of complex spatial patterns wide, it can be readily applied to mutants or other rare samples, hybridization: and it isIt suitableis genome- for during development proceeds with striking comparing individual embryos. Tomo-seq precision: the different organs are generally is highly complementary to microscopy- formed at the correct position, at the right time, and with a defined size. On the other al.,based PLoS techniques Genet, 2015) like single-molecule that offer absolute FISH hand, regulation should also not be too (Junker et al., Dev Cell, 2014; Soh, Junker et rigid, since embryos need to flexibly adjust to environmental perturbations and cor- workquantification (Karoline andHoller), sub-single-cell we have success spatial- fullyresolution expanded for selectedtomo-seq genes. to sub-single-cell In ongoing rect errors caused by noisy gene expres- resolution by identifying maternally depos- sion. Focusing on the zebrafish as a model ited transcripts that are localized within system, we study the interplay between variation and stability in embryonic and adult genome-wide sub-single-cell spatial data- the zebrafish oocyte. Our aim is to use this- tissues. Using single-cell transcriptomics, nisms that are responsible for intracellular we aim to understand how cell type diver- set to identify the 3’UTR-mediated mecha sity in vertebrate animals is generated and we aim to combine single-cell RNA-seq with mRNA localization. In other ongoing work regulated, and which compensatory mech- mechanisms that retain information about the immediate spatial proximity of cells in anisms are triggered upon perturbation. To a tissue, enabling us to systematically iden- this end, we develop novel experimental tify the effect that the local niche has on cell and computational approaches for spatially- fate decisions. resolved transcriptomics and massively par- Massively parallel lineage tracing allel lineage analysis on the single cell level. Besides spatial position, lineage history We use these approaches in combination constitutes another main determinant of with light microscopy and tools from zebraf- cell identity. Single-cell RNA-seq can systematically identify all different cell types ish developmental biology. in a tissue or organ. However, it does not yield direct information about lineage relationships and differentiation trajectories. Spatially-resolved transcriptomics - We recently developed a strategy for spatially-resolved transcriptomics in 3D (Junk- andWe have lineage developed tracing the in thousandsfirst method of that single al er et al., Cell, 2014). This method, called cellslows simultaneousfrom the same transcriptome animal (Bo Hu,profiling Nina - Mitic, Bastiaan Spanjaard). Our approach, cal techniques, novel approaches for low- input‘tomo-seq’, RNA-seq, combines and mathematical traditional histologi image - reconstruction to generate transcriptome- atedcalled at ‘scartrace’, the target usessite ofthe Cas9 short as insertions heritable wide 3D maps of gene expression (Figure cellularor deletions barcodes (“genetic for lineagescars”) tracingthat are (Fig cre- 1). Tomo-seq has three important advan-

ure 2). In successful pilot experiments we 178 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 1. The tomo-seq method for spatially resolved transcriptomics. A. The sample of interest (here: zebrafish embryo at somitogenesis stages) is cryosectioned, and RNA from individual sections is extracted, reverse transcribed, and linearly amplified. B. The approach can be extended to 3D by generating datasets for embryos sectioned along 3 different axes followed by computational image reconstruction.

marked individual cells with genetic scars in a GFP transgene at early developmental stages by injecting Cas9 and sgRNA into

that the resulting scars have very high complexity,1-cell stage suggesting zebrafish embryos.that we can We use found the

of individual cells. (Junker et al., BioRxiv, 2016).CRISPR/Cas9 We are system currently for expanding efficient barcoding this system to achieve greater control of scarring dynamics, to add mechanistic information Figure 2. Massively parallel single-cell lineage tracing. about cumulative signaling intensity on top A. Cells are marked with indels (“genetic scars”). Each cell will of lineage information, and to place lineage carry a different scar as nonhomologous end joining of Cas9- trees in 3D space. induced double-strand breaks is a highly variable process. B. We inject Cas9 and sgRNA into 1-cell stage zebrafish embryos, which Variability and robustness leads to cell labeling at early developmental stages. The scars are We have recently received an ERC starting transmitted to daughter cells and serve as lineage readouts. grant (start date: January 2017) to use the methods described above for studying variability and robustness during vertebrate Selected Publications

Junker, J.P., Spanjaard, B., Peterson-Maduro, J., Alemany, A., Hu, B., Florescu, M., and project (Roberto Moreno), we study cause van Oudenaarden, A. (2016). Massively parallel whole-organism lineage tracing using anddevelopment consequence and inof adultstochastic life. In heart an ongoing inver- sion along the left/right axis, a phenom- Soh, Y.Q.S., Junker, J.P., Gill, M.E., Mueller, J.L., van Oudenaarden, A., and Page, D.C. (2015). ACRISPR/Cas9 Gene Regulatory induced Program genetic for scars. Meiotic Biorxiv, Prophase doi: 10.1101/056499 in the Fetal Ovary. PLoS Genet 11, enon that happens with a surprisingly high e1005531. - Junker, J.P., Peterson, K.A., Nishi, Y., Mao, J., McMahon, A.P., and van Oudenaarden, A. ish. Our preliminary data links this phe- (2014). A Predictive Model of Bifunctional Transcription Factor Signaling during Embry- nomenonfrequency toof variablearound 5% assignment in wildtype of cellszebraf to the three germ layers during gastrulation. Junker, J.P., Noël, E.S., Guryev, V., Peterson, K.A., Shah, G., Huisken, J., McMahon, A.P., Ber- ezikov,onic Tissue E., Bakkers, Patterning. J., and Developmental van Oudenaarden, Cell 31, A. (2014).448–460. Genome-wide RNA Tomography As a next step, we will use the scartrace method to reveal how lineage trees adapt Junker, J.P., and van Oudenaarden, A. (2014). Every cell is special: genome-wide studies to developmental perturbations such as re- in the Zebrafish Embryo. Cell 159, 662–675.

add a new dimension to single-cell biology. Cell 157, 8–11. the developing heart. Furthermore, we will determineduction of specifichow adult progenitor organs regenerate pool sizes af in- Start of the Group: November 2015 PhD students Bo Hu system for this line of research, as many or- Group Leader Karoline Holler ganster injury. recover The fully zebrafish even after is anmassive ideal injury.model Dr. Jan Philipp Junker Nina Mitic We will focus on the heart and brain, two - Postdocs Technician Roberto Moreno-Ayala Jana Richter (theorgans brain that has regenerate adult stem efficiently cells, whilein the theze Bastiaan Spanjaard heartbrafish, has but not). through different mechanisms Administrative assistant Regina Philipp

MDC Research Report 2016 179 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Scott Lacadie

Dynamics and functions of the enhancer landscape during zebrafish development and cardiogenesis

The temporal and spatial regulation of Transcriptional regulation during gene expression are key determinants of zebrafish development Regulation of gene expression at the tran- cellular identity within multicellular eukary- scriptional level centers around regions of otes. Such fine spatiotemporal control open chromatin where transcription fac- occurs via the binding of transcription factors bind to DNA and post-translational tors within accessible, non-protein-coding regions of chromatinized DNA. Studies suggesting that the majority of human disease-associated genetic variants fall in such regions of the genome highlight the importance of gene regulation for proper tissue formation and function. Pressing challenges are to understand how gene regulatory regions function and how they are impared in disease. These challenges must be met in the context of dynamic, multicellular systems that reflect the true complexity of biology. The zebrafish as a vertebrate genetic model provides an ideal system for in vivo testing of transcriptional regulatory regions wherein many tissues are highly parallel to their human counterparts and can be easily manipulated with medium to high throughput. Our goal is to identify, classify, and functionally characterize the temporal and spatial dynamics of transcriptional regulatory regions during Figure 1: Transgenic zebrafish are being used zebrafish embryonic development and to to define cell-specific open chromatin regions and the post-translational modification state use the zebrafish to mechanistically dis- of their associated nucleosomes. Plotted are sect human regulatory regions harbouring ATAC-seq data we generated on cells from the disease variants. embryo body and whole embryo ChIP-seq data for H3K4me1, H3K4me3, and H3K27ac taken from Bogdanovic et al., 2012.

180 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 2

- Directionality of human promoters - One frequent and intriguing characteristic latorymodifications element. of Since histones much within of our nucleounder- of human promoters is divergent transcrip- standingsomes reflect about the transcriptional activity state of regulation the regu tion, in which unstable, reverse-oriented comes from homogenous cancer cell cul- transcripts occur upstream of gene tran- tures, it is imperative that future studies are - carried out in models where cellular make- - up and signaling activity accuratley capture erscription regions start and sites. determined We used that DNaseI a sizeable hyper the intricate variation of biological systems. fractionsensitivity of tohuman define promoters boundaries indeed of promot lack Genome-wide assessment of regulatory ar- divergent transcription. From our studies it is clear that promoters showing divergent isolated from animals has recently become trancription contain two core promoters, morechitecture feasible within with specificthe advent cell of populations techniques one at each edge of the open chromatin re- such as ATAC-seq for the measurement of gion arranged in opposite direction, while DNA accessibility (Buenrostro et al., 2013) unidirectional promoters contain only one core promoter at the forward edge. Re- verse core promoters drive the production techniquesand BiTS-ChIP to vertebrate for genome-wide biology occupancywill great- of these unstable transcripts, but are not lyprofiling deepen (Bonn our understanding et al., 2012). Applying of tissue these spe- necessary for positioning of the upstream nucleosome, where transcriptional activ- We perform genome-wide assays for chro- ity is associated with a unique combina- cific gene regulation. - the activity of regulatory regions within cations. We envision reverse directed core matin architecture to define and classify tion of histone post-translational modifi development (Figure 1). We use the DNA se- transcription through the competition with quencesspecific cellfrom populations these regions during to predict zebrafish tran- forwardpromoters core as promotersspecific regulators for basal of RNA forward poly- scription factor activity and to shed light on merase machinery. the combinatorial code of transcriptional - nology is used to assess the in vivo func- Selected Publications tionregulation. of entire Cutting regulatory edge CRISPR/Cas9 regions, predicted tech Duttke, S.H.C., Lacadie, S.A.*** binding sites, and candidate trans factors. C., Heinz, S., Kadonaga, J.T., Ohler, U. (2015). Perspectives on Unidirectional versus , Ibrahim, M.M., Glass, C.K., Corcoran, D.L., Benner, Duttke, S.H.C., Lacadie, S.A.*** regions orthologous to human promoters Divergent Transcription. Mol. Cell 60, 348–349. The identification of zebrafish regulatory Heinz, S., Kadonaga, J.T., Ohler, U. (2015). Human promoters are intrinsically direc- and enhancers will reveal fundamental , Ibrahim, M.M., Glass, C.K., Corcoran, D.L., Benner, C., aspects of how such regions must be or- Lacadie, S.A., and Ohler, U. (2015). JAMM: A Peak Finder for Joint ganized to maintain function throughout tional. Molecular Cell, 57, 674–684. evolution. Using these orthology relation- LacadieIbrahim, S.A.M.M., Analysis of NGS Replicates. Bioinformatics (Oxford, England) 31(1), 48-55. , Ibrahim M.M., Gokhale S.A., Ohler U.(2016). Divergent transcription methods, we test human disease-associated and epigenetic directionality of human promoters. FEBS J, 283(23):4214-4222. regulatoryships together genetic with variants zebrafish to transgenesisdissect their *** = Co-first author mechanism in disease formation. Our sys- Group Leader tems biology approach combines the lat- Dr. Scott Lacadie est sequencing-based technologies with (BIH Delbrück Fellow) elegant computation and in vivo vertebrate developmental genetics to produce mecha- Postdoc Technician nistic insights into both basic biology and Dubravka Vucicevic Antje Hirsekorn human disease (Figure 2).

MDC Research Report 2016 181 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Markus Landthaler

RNA Biology and Posttranscriptional Regulation

Our main interest is the understanding of Post-transcriptional regulation by posttranscriptional regulatory networks RNA-binding proteins Roberto Arsie, Karol Rogowski, Ulrike Zinnall controlling gene expression in humans. Na- scent mRNA stably associates with a large Mammalian genomes encode several hun- number of mRNA-binding proteins to form dred RNA-binding proteins, each containing ribonucleoprotein (RNP) complexes. These one or multiple domains able to recognize target mRNA in a sequence and/or struc- complexes play a key role in the regulation ture dependent manner. The association of of spatial and temporal changes in protein these proteins with transcripts regulates the synthesis by controlling transport, storage, biogenesis and translation of mRNA. For the majority of mRNA-binding proteins the tar- decay and translation of mRNAs. Deregula- get transcripts and their function in RNA me- tion and failed coordination of these mecha- tabolism are unknown, limiting our under- nisms contribute to pathophysiologicial standing of post-transcriptional regulatory development and conditions. A prerequisite in the characterization of known and novel for a systems level understanding of post- mRNA-bindingprocesses. In particular,proteins. By we combining are interested maps transcriptional regulation is a transcriptome- of functional RNA-protein interactions with wide high-resolution map of mRNA-protein cell-based and biochemical assays, we deter- contacts that allow us to study how these mine the dynamic assembly of RNA-binding proteins on their target mRNAs as well as the interactions control the fate of mRNA tran- elements and mechanisms guiding mRNA scripts. To achieve this goal we use photo- maturation, localization, turnover and pro- crosslinking approaches (PAR-CLIP [Hafner tein synthesis (Figure 1).

& Landthaler et al. Cell 2010]) and protein occupancy profiling on mRNA [Schueler et al. Genome Biol. 2014] in combination with next-generation sequencing to identify functional RNA-protein interactions at single nucleotide resolution. In addition we employ mRNA interactome capture in combination with mass spectrometry [Baltz et al. Mol Cell 2012] to characterize changes in the composition of mRNP complexes in response to intra- and extracellular signals.

Figure 1: Posttranscriptional regulation of A20/TNFAIP3 mRNA by the RNA-binding protein RC3H1

182 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

proteome using quantitative proteomics (mRNA interactome capture) and to display the protein occupancy on mRNA transcripts by next-generation sequencing. Application to a human embryonic kidney cell line identi-

novel mRNA interactors. Ourfied closeobservations to 800 proteins indicate with the more presence than of240 a large number of mRNA binders with diverse molecular functions participating in combinatorial posttranscriptional gene-expression networks. More recently we described a global comparison of protein occupancy pro-

cell types, and provided evidence for altered mRNAfiles on metabolism mRNA transcriptomes as a result fromof differential different protein-mRNA contacts. Our work demon- - ing on protein coding transcripts for assess- ingstrated the thecis-regulatory value of protein mRNA occupancy sequence profilspace and its dynamics in growth, development and disease. We use the described methods to monitor Figure 2: Identification of the mRNA-bound the dynamic changes of the mRNA-protein proteome and protein occupancy profiling on interactome to capture differentially binding mRNA. proteins and occupied cis-regulatory mRNA regions as a consequence of intra- and extracellular signals. These data provide insights Specificity and function of RNA into the mechanisms that lead to changes in helicases mRNA turnover and protein synthesis in re- Nicolai Kastelic sponse to genotoxic stress, viral infection and during embryonic stem cell differentiation RNA helicases are a family of highly con- (Figure 2). served proteins that utilize NTP hydrolysis to unwind RNA structures and/or remodeling of ribonucleoprotein complexes. RNA helicases Selected Publications participate in all biological processes that in- Landthaler M#. (2016) volve RNA metabolism, including transcrip- Genome Research 26, 1000-1009. tion, splicing and translation and have been MurakawaWessels HH, Y, ImamiHinz M, K, Mothes Baltz A, J, Small Schuetz S, Selbach A, Yasuda M, T,Ohler Mastrobuoni U#, G, Friedel CC, implicated in disease states such as tumori- DölkenThe early L, Rajewskyfly embryo K, mRNA-bound Kempa S, Schmidt-Supprian proteome. M, Heinemann U, Wolf J, Scheide- genesis and the replication of viruses. We reit C, Landthaler M Communications 6, 7367. (2015) RC3H1 represses the IKK/NF-κB negative feedback functional interactions of helicases, their he- Rybak-Wolfregulator A20 A, byJens binding M, Murakawa to a 3’UTR Y, Herzog composite M, Landthaler structure-sequence M#, Rajewsky element. N#. Nature(2014) licase-inactiveare using our variants PAR-CLIP and approach RNA that to are define typi- A variety of Dicer substrates in human and C.elegans. Cell 159:1153-1167. Gregersen LH, Schueler M, Munschauer M, Mastrobuoni G, Chen W, Kempa S, Dieterich C, RNA target sites provides the foundation for Landthaler M biochemicalcally transient and in nature.reverse Thegenetic identification approaches of . (2014) MOV10 is a 5’ to 3’ RNA helicase contributing to to investigate the remodeling mechanism of SchuelerUPF1 mRNA M, Munschauer target degradation M, Gregersen by translocation LH, Finzel alongA, Loewer 3’UTRs. A, Chen Molecular W, Landthaler Cell 54: 573-585. ribonucleoprotein complexes by helicases. M#, - scriptome. Genome Biology 15:R15 These studies give insights into the deter- Dieterich C# (2014) Differential protein occupancy profiling of the mRNA tran minants of target RNA selection, functional interactions with other RNA-interacting pro- Group Leader teins and the physiological role of RNA heli- Dr. Markus Landthaler Marcin Kolinski cases. Ruben Lopes Pereira Abreu Scientists Karol Rogowski Dynamics of the mRNA-bound Dr. Miha Milek Ulrike Zinnall proteome in response to intra- and extracellular signals Dr. Emanuel Wyler Technical Assistants Nicolai Kastelic , Marcin Kolinski, Miha Milek, Dr. Igor Minia Ouidad Benlasfer Emanuel Wyler PhD students Roberto Arsie Secretariat We developed a photoreactive nucleotide-en- Orsalia Hazapis Sabrina Deter - Nicolai Kastelic cation approach to identify the mRNA-bound hanced UV crosslinking and oligo(dT) purifi

MDC Research Report 2016 183 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: private Photo:

Irmtraud M. Meyer

RNA Structure and Transcriptome Regulation

My group develops dedicated computa- From all we know, complex organisms like tional methods for investigating how RNA processing their primary transcripts into structural features and RNA-RNA interac- different,ours have functionaldeveloped productsmore refined according ways for to tions regulate the expression of protein- the status of the corresponding cell. coding and non-coding genes. Our primary focus are the transcriptomes of eukary- Another surprise in the wake of the human genome sequencing project was the otic model organisms such as the fruit fly (Drosophila melanogaster), the mouse encodes protein information. Moreover, (Mus musculus) and the human (Homo manyrealisation genes that do onlynot encode a small a fraction protein (<2%) prod- sapiens). For this, we aim to combine p5)) and even the primary transcripts of computational analyses developed by us protein-codinguct (25180 so-called genes RNAcontain genes a seemingly(GRCh38. with large scale data sets generated using disproportionate fraction of non-coding state-of-the-art high-throughput sequenc- nucleotides (introns, untranslated regions). This is surprising, given that introns are ex- ing and investigation protocols. cised before the remaining, shortened RNA sequence is translated into the corresponding protein. When the human genome sequence was released more than a decade ago, it came My group develops dedicated computa- as a surprise to many that the number of tional methods and algorithms to under- protein-coding genes was not radically dif- stand how the transcriptome is regulated. ferent from the corresponding gene count Our main goal is to understand the role that for the more humble organism Caenorhab- RNA structure and interactions between ditis elegans (C. elegans). The current gene different transcripts play in regulating tran- - scripts in the living cell. For this, we study sus 20447 for C. elegans (WBcel235)) are transcriptome data from model organisms stunninglycounts (20313 similar. for human The gene (GRCh38.p5) count itself ver is - thus only a poor measure for the complex- man. ity of the corresponding organism. such as the fruit fly, the mouse and the hu Beyond the one-dimensional view The genome in any cell of a living organism of transcripts: is not a static entity, but generates diverse functional products as function of time. illustrate the Central Dogma of Biology by respond to internal and external changes. depictingMore often transcripts than not, figuresas linear in molecules textbooks TheThese primary define theproducts status of thatthe cellgenome and alsoare inside a eukaryotic cell, with transcrip- transcripts (RNA sequences). These are tion and splicing seemingly happening consecutively. What we know from many but need to be processed before they yield dedicated experiments, however, is that typically not the final functional products, processes such as splicing, RNA editing and

the final products (proteins and RNAs). 184 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 1: Arc-plot for the HDV ribozyme made using R-Chie (www.e-rna.org). Each arc represents one pair of base- paired alignment columns. Arcs and the alignment at the top show the alternative structure and the active structure; those at the bottom the inhibitory alternative structure. The left legend specifies the percentage of canonical base-pairs for each arc. The right legend colour- codes the nucleotides and specifies the evolutionary evidence supporting each arc.

RNA structure formation can happen co- that the sequences of structured RNAs not transcriptionally, i.e. while the transcript emerges from the genome. Similarly to structure, but also on transient structural protein information, information on RNA featuresonly encode of their information co-transcriptional on their final folding RNA structure can be directly encoded in the transcript itself. We thus expect that RNA - structural features and RNA-RNA interac- scriptspathway from (Meyer related and Miklós,organisms BMC also Mol have Biol tions are widely used for regulating gene similar2004). Itco-transcriptional turns out that orthologous folding pathways tran expression on transcript level. and that distinct transient RNA structure features can be as conserved and function- Modelling RNA structures in vivo: - - ally relevant as those of the final RNA struc for detecting the RNA structural features ture (see Figure 1; Steif and Meyer, RNA Bi thatIn order are functionallyto devise computational relevant in vivo, methods it is Biologyology 2012; 2015). Zhu, This Steif, has Proctor obvious and implica Meyer,- worth acknowledging the complexity of tionsNucl. Acidsfor many Res. 2013;state-of-the-art Zhu and Meyer, methods RNA the cellular environment and the impact in RNA secondary structure prediction as this may have on the structure forma- these typically assume that any given tran- tion process (Lai, Proctor and Meyer, RNA script folds into exactly one (but not more) 2013). By devising the new RNA second- functional RNA structure. A probabilistic ary structure prediction program CoFold method developed earlier by us (Transat, (Proctor and Meyer, Nucl. Acids Res. 2013), see Wiebe and Meyer, PLoS Compbio. 2010) we showed that it is possible to capture the aims to address this problem and has al- overall effects of the speed and direction- lowed us to detect individual, conserved ality of transcription in vivo. Our method RNA secondary structure features of pseudo-knotted structures, ribo-switches and especially for long transcripts (> 200 nt) transient structures which are otherwise suchyields as significantly ribosomal RNAs. improved We know predictions, already

notoriously difficult to predict. MDC Research Report 2016 185 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 2: (A) Genomic context of identified editing sites. (B) Distribution of conversion types for four tissue types. (C) Percentage of common editing sites between pairs of tissues. (Bottom) Gene CG5850 is differentially expressed between head (blue) and digestive system (red) and editing and splicing may affect each other. X-axis: exons of the gene, y-axis: number of reads normalized by library size. Arrows show editing sites. The purple box is predicted to be alternatively expressed.

RNA structure features involved in splicing regulation: order to explore the link between RNA struc- tureSch√∂ning and alternative et al., Nucl. splicing Acids onRes. a transcrip2008). In- tome-wide scale, we recently analysed tis- are known to encode functional RNA struc- tureViral ingenomes protein-coding such as regionsHepatis-C as oneand majorHIV-1 constraint for their genomes it to remain pipelinesue-specific that high-throughput explicitly captures transcriptome the require- short. We contributed early on to these mentdata from for double-stranded the fruit fly. Using regions, a new we analysis iden- studies by showing that these RNA structures can be reliably predicted provided well as numerous (244) regions where RNA the know protein context is explicitly taken editingtified around and alternative 2000 novel splicing editing are likely sites asto into account (Pedersen, Meyer, Forsberg, - tected RNA editing sites are approximately Pedersen, Forsberg, Meyer and Hein, Mol. threeinfluence times each more other likely (see toFigure occur 2). in The exons de Simmonds and Hein, Nucl. Acids Res. 2004; with multiple splice sites than unique ones Functional RNA structures overlapping and conserved RNA structure features over- protein-codingBiol. Evol. 2004; regions, Watts et however, al., Nature are 2009). not the preserve of viral genomes, but can also Figure 3). We therefore conclude that local regulate the alternative splicing and transla- RNAlap 39%structure (96/244) features of have these the regions potential (see to tion of eukaryotic protein-coding genes e.g. regulate alternative splicing via structural in Arabidopsis thaliana, mouse and human changes induced by RNA editing (Mazloom- ian and Meyer, RNA Biology 2015).

(Meyer and Mikl√≥s, Nucl. Acids Res. 2005; 186 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

0102030405060708090 100 Aligned Species (top to bottom)

Conservation D. grimshawi Covariation D. virilis One−sided D. mojavensis Invalid D. willistoni Unpaired D. ananassae Gap D. pseudoobscura D. persimilis D. sechellia D. simulans D. melanogaster (OregonR) D. melanogaster (Reference) D. yakuba D. erecta

Figure 3: (Top) Arc-plot for the highlighted region of the Cip4 gene containing a predicted, conserved RNA secondary structure overlapping RNA editing sites (red arrows) that could influence alternative splicing via structural changes. The left legend colour-codes the nucleotides according to the evidence supporting each arc, see also Figure 1. Figure made using R-Chie (www.e-rna.org). (Bottom) Gene structure of the Cip4 gene with grey box highlighting the structure- containing part at the top.

Trans RNA-RNA interactions cated computational analyses and methods regulating the transcriptome: developed by us with transcriptome-wide data sets generated via state-of-the-art se- RNAs not only have the potential to form quencing and experimental protocols to RNA structure, but can also interact with gain a global and detailed understanding of other RNAs in trans. This involves the same how RNA structure and RNA-RNA interac- simple structural building blocks, namely tions regulate eukaryotic gene expression hydrogen bonds between complementary on transcript level.

thus much more straightforward to evolve ornucleotides design a desired ({G,C}, {A,U}RNA structure and {G, U}).or trans It is the University of British Columbia in Van- RNA interaction using an RNA than a pro- couver,I will be Canada, joining and the lookMDC forward early 2016 to setting from tein. We thus expect many trans RNA-RNA up collaborations with new colleagues at interactions beyond those already known (miRNA-mRNA, snoRNA-rRNA, etc) to be community in Berlin, Germany and Europe discovered. We have shown how the pow- (Contact:the BIMSB [email protected]). and MDC and the wider scientific er of the comparative approach can be harnessed in the context of coding gene prediction, RNA gene prediction and RNA Selected Publications: structure prediction and the prediction of - trans RNA-RNA interactions. Based on our tion prediction methods. Nucleic Acids Research 44(7):e61. recent survey of methods for predicting de Lai D, Meyer IM (2016) A comprehensive comparison of general RNA-RNA interac novo trans RNA-RNA interactions (Lai and Mazloomianregulating alternative A, Meyer splicing.IM (2015). RNA Genome-wide Biology 12(12):1391-1401 identification and characterization Meyer, 2016), we intend to focus on com- of tissue-specific RNA editing events in D. melanogaster and their potential role in putational methods that can better handle RNA Biology 12(1):5-20 full-length transcripts and that can differ- Zhu JY, Meyer IM (2015). Four RNA families with functional transient structures. entiate between different kinds of evolu- structure formation. RNA 19(11):1461-1473 tionary constraints. Lai D, Proctor JR, Meyer IM (2013). On the importance of cotranscriptional RNA method that takes co-transcriptional folding into account. Nucleic Acids Research Proctor41(9):e102 JR, Meyer IM (2013). CoFold: an RNA secondary structure prediction made dramatic advances through the in- ventionIn the last of new few years,transcriptome-wide transcriptomics inves has- tigation protocols (e.g. CLASH, SeqZip) as Start of the Group: January 2016 Group members: well as a new generation of sequencing Stefan Stefanov techniques (e.g. nanopore and SMRT se- Group Leader Carlo M. Massimo - Peter Menzel ger read-lengths. We want to combine dedi- quencing) that allow for significantly lon Prof. Dr. Irmtraud Meyer

MDC Research Report 2016 187 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Benedikt Obermayer

Evolutionary Modeling and Inference in Computational Systems Biology

As more animal genomes get sequenced, tween a miRNA and its target site, but rather computational biology gains new power to to be conveyed by the collective regulation by multiple miRNAs on entire target sets. understand biological function from an evolu- With the function of the network conserved tionary perspective. Developing methods for throughout evolution, this implies that in- network inference from heterogeneous and dividual target sites do not evolve indepen- noisy data, we use evolutionary correlations - ary correlations between vertebrate miRNA to understand collective functions of post- targetdently sites,of each we other. corroborate Indeed, the using conservation evolution transcriptional gene regulation and simplify of combinatorial regulation and identify dif- protein folding. Further, we use evolutionary ferent regulatory strategies for the targeting of different gene sets. We also developed and modeling and inference to study small open analyzed methods to distinguish direct from reading frames (ORFs). So-called micropep- indirect correlations in amino acid substitu- tides encoded by these ORFs have recently tion patterns given alignments from phyloge- emerged as important regulators of devel- netically closely related species, which can be used to simplify and aid computational pro- opment and physiology, but computational tein folding. identification is challenging due to their small size. We predict hundreds of novel con- Small open reading frames served small ORFs from their characteristic that a large fraction of the genome is tran- evolutionary signatures. In several ongoing scribed,Genome-wide partly into expression canonical profiling coding mRNAs, shows collaborations, we now investigate coding or and partly into RNAs without long ORFs. The regulatory functions of small ORFs in mam- functions of most of these long non-coding RNAs (lincRNAs) are poorly understood. malian cell lines and in the model organisms However, recent observations suggest that D. rerio, D. melanogaster, and C. elegans. small ORFs on lincRNAs are often translated and produce peptides detectable via mass Inference of regulatory networks Gene regulatory networks are densely inter- micropeptides have essential biological roles connected systems of trans-acting regulators inspectrometry. physiology Inor fact,development. some of theYet, encodedtransla- and cis-acting sequence elements. This inter- tion of small ORFs could also exert regulatory dependency creates informative correlations functions, for instance via nonsense-medi- in gene expression levels or conservation ated decay. Despite numerous publications patterns. We develop methods for network on this subject, there is still little consensus inference using observed correlations, with a about the abundance of functional sORFs in particular focus on techniques to reduce bias animals, although complete gene catalogs are from confounding factors such as stochastic essential to explain the molecular basis of or- noise, indirect correlations due to covariates, ganismal phenotypes. or heterogeneity in the underlying samples. We applied these methods to understand features of post-transcriptional regulation In a collaboration between the Rajewsky andby microRNAs (miRNAs). This regulation ap- cRNAsGiraldez using labs, complementarywe identified hundreds methods: of smallribo- pears to depend less on individual links be- ORFs encoded by human and zebrafish lin

some profiling was used to detect translated 188 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Prediction of conserved small ORFs in animals. A: searching the transcriptome for small ORFs, we evaluate conservation signatures from multiple species alignments and use machine learning to predict conserved small ORFs. B: hundreds of small ORFs are predicted in different transcriptome regions of 5 animal species. C: nucleotide-level conservation is different for translated ORFs in 5’UTRs or lincRNAs than for annotated small protein-coding genes.

small ORFs, and evolutionary conservation was used to predict putatively functional cod- conserved across vertebrates. With in-situ hybridizationbrafish knock-out data model and morpholino for a 25aa peptideknock- we performed a much more comprehensive downs indicating functions in the brain, we ing small ORFs. Building on these findings, also use mammalian cell lines to study mo- in the transcriptomes of human, mouse, ze- lecular functions in neuronal development study and identified conserved smallC. elegans ORFs. and differentiation by means of RNA-seq and About 2,000 putatively functional animal smallbrafish, ORFs fruit flyare and predicted the nematode by this rigorous the Selbach lab, we use computational protein computational screen. Systematic homol- CRISPR/Cas genome editing. Together with

conserved candidates of high interest for docking and affinity purification followed by functionalogy analysis follow-up identified studies. a number Further, of we widely lev- collaborations,mass spectrometry we investigate to find protein micropeptides interaction eraged a large amount of published experi- withpartners possible of selected roles micropeptides. in heart disease In further using - animal models and clinical samples, we study pression of conserved small ORFs on multiple functions of small ORFs in immunity, and we mental data to confirm and characterize ex screen small ORF knockout phenotypes in D. and mass spectrometry. Togetherlevels, such with as the RNA-seq, Ohler and ribosome Landthaler profiling labs, we investigated translated small ORFs in melanogaster using CRISPR/Cas. noncoding transcriptome regions. While we Selected Publications found highly conserved and translated mi- L. Calviello, N. Muhkerjee, E. Wyler, H. Zauber, A. Hirsekorn, M. Selbach, M. Landthal- er, B. Obermayer, and U. Ohler, “Detecting actively translated open reading frames in there was generally only limited overlap be- S.D. Mackowiak, H. Zauber, C. Bielow, D. Thiel, K. Kutz, L. Calviello, G. Mastrobuoni, tweencropeptides conserved encoded and intranslated 5’UTRs or small lincRNAs, ORFs. ribosome profiling data”, Nat. Meth., 13:165 (2016)

N. Rajewsky, S. Kempa, M. Selbach, and B. Obermayer*, “Extensive identification and analysis of conserved small ORFs in animals”, Genome Biol., 16:799 (2015) areIn contrastactually tooften known only small conserved protein-coding in their J. Phys., 16:123017 (2014) B. Obermayer and E. Levine*, “Inverse Ising inference with correlated samples”, New positionsgenes, translated but not ORFs their in encoded 5’UTRs orpeptide lincRNAs se- B. Obermayer and E. Levine*, “Exploring the miRNA regulatory network using evolu- quences, suggesting that only some have cod- A.A. Bazzini*, T.G. Johnstone, R. Christiano, S.D. Mackowiak, B. Obermayer, E.S. Flem- ing, but many others have regulatory func- tionary correlations”, PLoS Comput. Biol., 10:e1003860 (2014) tions. We will combine these computational of small ORFs in vertebrates using ribosome footprinting and evolutionary conser- ing, C.E. Vejnar, M.T. Lee, N. Rajewsky*, T.C. Walther, and A.J. Giraldez*, “Identification * (co)-corresponding author vation”, EMBO J. 33:981 (2014) linefindings model with system high-throughput for neuronal data differentia from RNA- tion.expression Developing and translation computational profiling models in a cellfor translational regulation by small ORFs and Group Leader other cis-acting sequence elements, we aim Dr. Benedikt Obermayer to obtain a quantitative understanding of (Delbrück Fellow) Undergraduate student post-transcriptional gene regulation in this Denise Thiel* system. Graduate student Marco Uhrig* *part of the period reported - In collaboration with Ferdinand Le Noble at KIT, we use CRISPR/Cas to create a ze MDC Research Report 2016 189 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Uwe Ohler

Computational Regulatory Genomics

According to current understanding, com- eration and analysis with computational plexity in higher organisms is not achieved modeling. We frequently frame questions by a more complex repertoire of parts, i. e. learning approaches to make testable pre- genes, but instead by the more complex dictions.as classification These predictionsproblems and are use often machine vali- regulation of the parts. The expression of dated experimentally, either by us or our genes is tightly controlled on several levels friends and collaborators. At the same time, inspection of model parameters may allow — a large number of protein and RNA fac- us to draw conclusions about the impor- tors and DNA and RNA sequence elements tance of particular biological features. enable the precise regulation of interacting gene products. It is a key challenge to Quantifying gene expression at different layers of gene regulation decipher these complex networks of play- The rise of deep sequencing technologies ers and interactions, and to move biology has changed the world of genomics research from case studies to an integrated, global in ways we are only beginning to appreci- approach. addition to determining the DNA sequence Our lab develops and applies genomics ofate. a genome,One striking many aspect biological is its processesversatility; can in and computational approaches to understand mechanisms of gene regulation in - cationbe quantified of mRNA by means steady-state of sequencing. levels via eukaryotic organisms. Computational biol- (paired-end)One popular RNA application sequencing is the (RNA-seq). quantifi ogy has become indispensable to analyze A single RNA-seq data set contains mil- and ultimately make sense of large-scale lions of small (~100 nucleotide) reads, and data sets that look at the phenomenon of it is a well-known challenge to assemble these into the complete set of all alterna- gene regulation from different angles. Our tive splice isoforms of expressed genes. To long term goal is to investigate how regu- guide and improve this process, comple- latory networks enable the correct devel- mentary information can be used. We developed an isoform annotation algorithm opment of complex organisms, with their that is informed by genomic features (in- multitude of cell types that carry out differ- cluding start/stop codon, splice sites, cod- ent functions despite the same genome. ing potential) as well transcript features from RNA-seq (Majoros et al, Bioinformat-

Computational biology and gene annotation of splice isoforms to those that regulation containics 2014). a consistent In this way, open we reading restricted frame the Understanding gene regulation emerged as (ORF). While we cannot annotate non-cod- a crucial challenge after the initial sequencing transcripts in this manner, the annota- ing of the human genome, as misregulation tions of protein-coding genes improved sig- of gene expression is at the heart of many genes where RNA-seq data alone does not context, the particular strength of a multi- providenificantly, enough particularly coverage. for lowly expressed disciplinarydevelopmental group defects like oursand diseases. lies in the In com this- bination of genomics-oriented data gen- RNA-sequencing protocols go beyond the Using clever modifications and extensions,

190 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

steady-state transcriptome and quantify gene expression regulation from nascent transcription to protein synthesis. For instance, to determine which, and how strongly, ORFs are translated, ribosome - ing ribosomes. As the ribosomes move one codonprofiling at a maps time, the resulting positions reads of translat exhibit characteristic, three-nucleotide periodic footprints along the transcriptome. We have developed RiboTaper, a rigorous statistical

on the basis of this periodicity (Calviello approach that identifies translated regions We used RiboTaper to derive an extensive et al, Nature Methods, in press; Figure 1). covered ORFs for >11,000 protein-coding genesmap of from cell-type a single specific experiment. translation Mass spec that-

achieved excellent coverage of the cellular proteome.trometry dataAlthough confirmed they frequently that RiboTaper contain short ORFs, our analysis also suggested that only relatively few of the currently annotated long non-coding RNAs appear to encode stable polypeptides.

Transcriptional control: from proximal initiation to distal regulation

elements are located in different regions relativeIn eukaryotes, to the gene transcriptional they control. regulatory The core promoter, a ~100 nucleotide long region centered on the transcription start site (TSS), is contained within the proximal pro-

has been traditionally assumed to perform themoter function region todirectly recruit upstream the basal of transcrip a gene. It- tion complex to the correct genomic location, and thus to contain similar sequence

non-coding regulatory regions such as en- Figure 1. RiboTaper, a computational method hancerspatterns containacross all different genes. In combinations contrast, distal of to detect actively translated open reading frames (ORFs) from RiboSeq data. (A) The transcription factors (TFs) and thus convey workflow. Top: Using known ORFs, the optimal sequence patterns recognized by specific- read size and offset from the read start can be sarman, Development 2010). determined, which indicates the precise P-site Thecondition-specific application of activity deep (Ohlersequencing and Was has of the translated codon. Middle: Example P- shown transcriptional activity at many site profiles for three exons. Bottom: Only the places in the genome, going substantially coding exon exhibits a significant 3-nucleotide beyond annotated protein-coding genes. periodicity. (B) ORF annotations include both coding and upstream ORFs. initiation for different types of pervasive In ongoing work, we examine transcription

MDC Research Report 2016 191 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 2. Many human promoters are divergently transcribed. (A) By mapping transcription start sites (TSS) of nascent RNA transcripts (5’GRO; blue) to regions of open chromatin (DNase I; red), TSS locations could be precisely mapped to the edges of proximal promoters. In this way, promoters showing transcription in forward-only or both directions were identified. (B) Divergent and unidirectional promoters exhibit significant differences in chromatin and sequence composition.

transcription. For instance, many mammalian promoters show divergent transcrip- seq provides coarse-resolution maps of tion in both directions, i.e. in the forward gions andsites in DNA. Specifically, DNase- direction towards the annotated gene and which are universal hallmarks of all types in the reverse direction away from the gene. ofnucleosome-free transcriptional sites regulatory or “open elements chromatin”, in- Notably, this phenomenon is not conserved cluding proximal promoters. We are now across all eukaryotes (cf. Core et al, Cell Rep investigating DNase-seq data at nucleotide 2012). We could show that human core pro- resolution, to investigate whether it is pos- moters are inherently unidirectional and sible to pinpointing individual TF binding that reverse-oriented transcripts originate from distinct reverse-directed core promot- etevents al, Nucleic (protected Acids “footprints” Res 2014). in To the this DNase end, Figure 2). A new algorithm enabled us to weprofiles) built withinsimple openbaseline chromatin models (Yardimci that de- ers (Duttke, Lacadie et al, Mol Cell 2015; convolved DNase experimental bias from genuine binding signal at putative foot- Bioinformaticsuse DNase I accessibility 2015), which to precisely revealed define that - forproximal expressed promoter genes borders across (Ibrahim different et cell al, able data is frequently not deep enough to - reliablyprints. In identify this manner, genuine we binding, found that and avail that directional. These unidirectional promot- some TF binding events are inherently hard erstypes, are about depleted 25-45% at theirof promoters upstream are edges uni to identify, for instance when the TF recog- of the distinct reverse core promoters. Di- nition sequence overlaps with the experi- vergent transcription is thus not an inher- mental bias. ent property of the transcription process but rather the consequence of the presence RNA regulation: identifying target of two core promoters. We are currently sites of regulatory proteins examining whether reverse directed tran- The last decade has seen an outstanding increase in appreciation of the role of post- of the forward counterpart. transcriptional gene regulation. RNA-bind- Forscripts a long have time, an influence the large on non-coding the expression ge- ing proteins (RBPs) and small, regulatory nomic sequence space of higher eukaryotes RNAs such as microRNAs (miRNAs) recog- - tory regions that are targeted by TFs. Sev- eralmade recent it difficult genomic to pinpoint approaches distal provide regula tonize map specific the sitesprecise in targettarget RNA sites transcripts of post- transcriptionaland influence their regulators fate. It via is cross-linking now feasible - “global regulatory profiles”, which reflect where regulatory factors target specific re followed by deep sequencing (CLIP-seq). In 192 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

previous work, we have implemented spe- bers of proteins, as well as accounting for - regulatory effects at different layers of gene expression. ascific for approaches follow-up prediction for reliable, of regulatory high-resolu se- The described studies frequently resulted quencetion identification motifs (Ascano of binding et al, Nature sites, as2012). well from joint work and had the support from While we are developing the next genera- a number of excellent colleagues, including tion of algorithms to deal with a wider ar- - ray of experimental methods and improved protocols, we have continued to apply exist- James Kadonaga (UC San Diego); Tom Tus ing methods in several collaborative stud- Landthaler,chl (Rockefeller Benedikt University); Obermayer Philip and Benfey, Mat- ies, for instance to determine the target thiasGreg SelbachCrawford, (MDC). (Duke University); Markus sites of Regnase-1/Roquin (Mino et al, Cell 2015) and RBPMS (Farazi et al, RNA 2014).

Putting the pieces together: Selected Publications networks involving different Wessels HH regulatory factors or mechanisms Ohler U* Our work on DNA and RNA-based regula- Genome Res ,26:1000-9, Imami K, Baltz 2016. AG, Kolinski M, Beldovskaya A, Selbach M, Small S, tion converges in projects where we aim Calviello, L,Landthaler Mukherjee M*. N The, Wyler mRNA-bound E, Zauber H, proteome Hirsekorn of the A, Selbach early fly M, embryo. Landthaler to dissect quantitative contributions of M, Obermayer B, Ohler U. Detecting actively translated open reading frames in ribo- multiple regulatory factors or mechanisms, Duttke SH*, Lacadie SA*, Ibrahim MM, Glass CK, Corcoran DL, Benner C, Heinz S, - Kadonagasome profiling JT*, Ohlerdata. Nat U*. MethodsHuman promoters 13:165-70, are 2016. intrinsically directional. Mol Cell rect expression. For instance, several RBPs targetand how so-called they work AU-rich together sequence to define elements cor Mukherjee N*, Jacobs NC, Hafner M, Kennington EA, Nusbaum JD, Tuschl T, Blackshear57:674-84, 2015.PJ, Ohler U* RNA-binding protein ZFP36. Genome Biol 15:R12, 2014. mRNAs. ZFP36, also known as tristetra- Yardımcı GG, Frank CL, .GlobalCrawford target GE*, mRNA Ohler specification U*. Explicit DNase and regulation sequence by bias the prolin(AREs) or in TTP, the 3’and untranslated ELAVL1, also region known of modeling enables high-resolution transcription factor footprint detection. Nucleic as HuR, both interact with AREs, but they have antagonistic roles. We determined Acids Res 42:11865-78, 2014. - tegrated it with existing data for ELAVL1, Group Leader Henriette Miko andZFP36 examined binding the preferences combinatorial using regulation CLIP, in Prof. Dr. Uwe Ohler Alina-Cristina Munteanu of AU-rich elements by ZFP36 and ELAVL1 Hans-Hermann Wessels (Mukhkerjee et al, Genome Biol 2014). Scientists Using partial correlation analysis, we were Jose Maria Muino Acuna Graduate and Undergraduate able to quantify the association between Philipp Boss students ZFP36 binding and differential target RNA Mahsa Ghanbari Ekatarina Eroshok abundance upon ZFP36 overexpression, Sucheta Arun Gokhale Lea Heinze independent of effects from confounding Scott Lacadie Wanja Kassuhn - Neelanjan Mukherjee Clemens Messerschmidt Michal Szczesniak Jonas Peters carefulfeatures. investigation We identified showed thousands that of ZFP36 over Dubravka Vucicevic Julia Philipp lapping ZFP36 and ELAVL1 binding sites; Christine Rebecca Worsley Hunt Maja Schuster sequence variants, representing a subset Camille Schwartz ofdegrades the sequences transcripts ELAVL1 through interacts specific withARE PhD students to stabilize transcripts. Thus, ZFP36 and Sophia Bauch Technical Assistants ELAVL1 bind an overlapping spectrum of Christian Martin Burkert Antje Hirsekorn RNA sequences, yet with differential rela- Lorenzo Calviello tive preferences that dictate combinatorial Dina Hafez Secretariat regulatory potential. Motivated by these Michaela Liemen results, we are currently expanding our Aslihan Karabacak Regina Keiper integrative models to include larger num- Mahmoud Ibrahim

MDC Research Report 2016 193 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Ana Pombo

Epigenetic regulation and chromatin architecture

Our laboratory aims to understand mechanisms of RNAPII regulation in pluripotent gene expression which act at multiple levels, from and differentiated cells the local action of transcription factors, to long- Robert A. Beagrie, Giulia Caglio, João Dias, Ana Miguel Fernandes, Carmelo Ferrai, range chromatin looping events that connect regu- Izabela Harabula, Alexander Kukalev, Kelly latory DNA sequences with the genes they regu- J. Morris, Marta Oliveira, Tiago Rito, Marisa late, to how whole chromosomes are positioned Saponaro, Markus Schueler, Konstantina within cell nuclei. Binding of transcription factors at Skourti-Stathaki, Elena Torlai Triglia gene promoters helps recruit chromatin remodel- Appropriate recruitment of the RNA pro- ers and the transcription machinery to transcription cessing machinery and chromatin remod- units. Upon recruitment, RNA polymerases either elers to chromatin is mediated by post- remain primed for subsequent activation (e.g. after environmental stress, during differentiation or - disease), or directly transcribe the DNA template translational modification of the C-terminal domain (CTD) of the largest RNAPII sub into RNA molecules, which code for proteins or for Rep. 10, 1213). The CTD is an unusual dis- orderedunit, RPB1 domain (Brookes composed & Pombo of an 2009, heptapep EMBO- structural RNAs that constitute many enzymatic tide consensus sequence, Tyr1-Ser2-Pro3- complexes. Thr4-Ser5-Pro6-Ser7, repeated 52 times An increasingly important aspect of gene regula- in mammals. All aminoacids are targets for tion are the mechanisms by which distant regulatory DNA sequences physically interact with gene we know most about Serine phosphorylation.post-translational modifications, of which promoters to recruit or activate RNA polymerase After recruitment of unphosphorylated II (RNAPII), the enzyme responsible for transcribing protein-coding genes, and many structural and phosphorylated on Ser5 residues (S5p), regulatory RNAs. Much is known about the tran- followedRNAPII toby promoters,sequential phosphorylation the CTD becomes on scription cycle of protein-coding genes by RNAPII, Ser7 (S7p) and Ser2 (S2p) in the transition which entails a fascinating and highly coordinated to elongation. S5p recruits the capping enzyme and H3K4 histone methyltransferases sequence of processes that promote nascent (HMTs), which mark active promoters. S7p transcript maturation concomitant with transcrip- is thought to mark the transition to elongation. As the nascent RNA emerges from the poly- tion, whereas S2p recruits H3K36 HMT, to merase, it is capped on its 5’ end. The polymerase mark coding regions, and the splicing and progresses through the coding region and introns polyadenylation machineries. The regula- are spliced out from the nascent transcript. Upon termination, the addition of a protective polyA tail tion of RNAPII at the transition between- to the transcript 3’ end also occurs on chromatin. noacids,initiation most and elongationoften found is atalso position influenced 7 of These co-transcriptional events are mediated by theby CTD most modifications C-terminal heptapeptideat non-canonical repeats ami exquisite post-translational modifications of the RNAPII enzyme itself, which coordinate the timely canonical repeats of the CTD with Ser7- to-Lys7(Figure 1).substitutions In 2015, we (i.e. showed Tyr1-Ser2-Pro3- that non- recruitment of appropriate machineries at each Thr4-Ser5-Pro6-Lys7) are both mono- and stage of the transcription cycle.

di-methylated, and that this modification 194 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 1: Two novel post-translational modifications of RNA polymerase II (RNAPII) mark the earliest stages of the transcription cycle. Mouse RPB1 C-terminal domain is composed of 52 heptapeptide repeats; non-consensus amino acids are enriched for at the distal region. Chromatin occupancy of RNAPII-K7me1, K7me2, K7ac, 8WG16, S5p, S7p and S2p, and mRNA-seq profiles are shown for active gene Rpl13. A balance between K7me and K7ac fine-tunes gene expression. Adapted from Dias et al. (2015).

occurs at the earliest stages of the transcription cycle at active genes in mamma- foundmonoubiquitinylate co-associated on H2AK119. chromatin In at devel 2012,- opmentalwe showed regulator that Polycomb genes and(Brookes RNAPII et areal. lian cells (Dias et al. 2015 eLIFE). 2012). We used chromatin immunoprecipi- associatedRegulation withof RNAPII active by transcription post-translational cycles. modifications extends beyond the changes sequencing, to map the phosphorylation at inducible genes, before activation, such tation (ChIP) coupled with next generation Phosphorylated RNAPII has been reported across the genome. Surprisingly, not only - state of RNAPII and Polycomb occupancy- velopmentalas heat-shock. regulator Our group genes has in identified mouse ES a lator genes under Polycomb repression, but cells,novel which poised are form repressed of RNAPII by at Polycomb silent de itsdid presence we find RNAPII was positively at developmental correlated regu with complexes, but become activated through the amount and occupancy of Polycomb. To early differentiation and cell commit- characterize the extent of transcriptional ment (Stock et al. 2007, Nature Cell Biol. activity at poised genes, which are targets of Polycomb repression, we are analyzing Complexes, PRC1 and PRC2, are important chromatin-bound RNAs using genome- for9, 1428). maintaining Two major the poised Polycomb state Repressorof primed wide approaches, and CAGE-seq data. developmental regulator genes. Polycomb Our unbiased genome-wide analyses of - tains Ezh2, which methylates H3K27, and PRC1complexes contains modify Ring1A histone and tails; Ring1B, PRC2 which con regulationRNAPII and in mouse Polycomb ES cells, occupancy which were also identified novel gene targets of Polycomb

MDC Research Report 2016 195 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 2: (A) Scheme of mouse embryonic stem cells (ESC), neuronal progenitor cells (NPC) and differentiated neurons. (B) A hierarchical topology of meta-domains (metaTADs) are found by clustering individual topological domains (TADs). (C) Contact map of a sub-chromosomal regions (Hi-C data). (D) Meta-domain trees for chr6 in ESC and NPC. Two examples of highly similar (green) and dissimilar regions (red) are indicated (Adapted from Fraser et al., 2015.).

Genome architecture: mecha- cohort of genes which are active in ES cells nisms of 3D chromatin folding butmore whose unexpected. expression In particular, is dampened we byfound Poly a- Mariano Barbieri, Robert A. Beagrie, Giulia comb repression (Brookes et al. 2012 Cell Caglio, Carmelo Ferrai, Rieke Kempfer, Doro- Stem Cell). This group of genes includes thee Kraemer, Alexander Kukalev, Gesa Loof, many genes important in cell cycle, cancer, Tiago Rito, Julieta Ramirez Cuellar, Markus signaling and energy metabolism. Our cur- Schueler, Leo Serebreni, Marta Slimak- rent studies aim to better understand their Mastrobuoni, Christoph Thieme, Elena Torlai regulation in the transition from pluripo- Triglia tency to differentiated cellular states. To probe for the dynamic changes in poly- Long-range gene regulation works through merase and chromatin conformation that chromatin looping mechanisms that bring accompany re-programming of gene ex- together the promoters of genes with dis- pression during cell commitment, we have extended our analyses to neural percursor - cells and differentiated neurons and to car- says).tant regulatoryThis contact, regions mediated (e.g. by enhancers; transcrip- diac lineages. Our current bionformatics tionreviewed factors in and Beagrie other & chromatin Pombo 2016 associated Bioes analyses aim to understand how Polycomb components, promotes the recruitment of repression acts to establish poised states of the transcriptional machinery to the gene promoter, acting as the starting point for and differentiation. the synthesis of transcripts. Genes that RNAPII occupancy at genes in pluripotency

196 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

regulate early development are often ex- by simulating chromatin occupancy maps pressed at different stages of differentia- of epigenetic features as interaction sites. tion and in cell types, and are accordingly - ing to study the cell-to-cell variability and - physicalWe also apply distances single that cell fluorescencecharacterize imagchro- tionregulated to contacts through with specific their respective interactions en- matin contacts. We currently develop novel hancerswith different (which regulatory have been regions. described In up addi to approaches to map chromatin contacts at 1 Mbp from gene promoters), genes also co- the single cell level, which can be applied associate with other genes in 3D space. Our in tissues (such as biopsies) to help inves- group has shown that Active and Polycomb- tigate the close relationship between chro- repressed genes associate, respectively, matin folding and genome function. -

with ‘active transcription factories’ (con Selected Publications taining RNAPII-S5p, S2p) and with ‘Poised Arner E, Daub CO, Vitting-Seerup K, Andersson R, …,Ferrai C, ..., Morris KJ, ..., transcription factories’ (containing RNAPII- Pombo A, …, Hayashizaki Y (2015) Transcribed enhancers lead waves of coordi- andS5p suggestedand Polycomb; that such Ferrai associations et al. 2010 estab PLoS- nated transcription in transitioning mammalian cells. Science 347, 1010-4. Biol.; Brookes et al. 2012 Cell Stem Cell), Pombo A, Dillon N (2015) Three-dimensional genome architecture: players and important for silencing in the primed state, mechanisms. Nature Rev Mol Cell Biol, 16, 245-257. orlish for specific robust chromatinactivation upontopologies induction. that are Fraser J*, Ferrai C*, Chiariello AM*, Schueler M*, Rito T*, Laudanno G*, Barbieri M, Moore BL, Kraemer DCA, Aitken S, Xie SQ, Morris KJ To explore the changes in chromatin con- Hayashizaki Y, Carninci P, Forrest ARR, the FANTOM Consortium, Semple CA1, Dostie tacts that accompany the transcriptional J1, Pombo A1, Nicodemi M1 (2015) Hierarchical folding and, Itoh reorganization M, Kawaji J, Jaegerof chro I,- mosomes are linked to transcriptional changes in cellular differentiation. Molecular reprograming of mouse ES cells during Systems Biology 1joint corresponding authors) neuronal differentiation, we mapped chro- Dias JD, Rito T*, Torlai Triglia E*, Kukalev A, Ferrai C, Chotalia M, Brookes E, matin contacts genome wide by Hi-C, and Kimura H1, Pombo 11, A1 852. (*joint first and - matched CAGE data (Fraser et al. 2015 sine residues marks early transcription in mammalian cells. eLife 4, e11215. (*joint second and 1joint corresponding (2015) Methylation authors) of RNA polymerase II non-consensus ly Molecular Systems Biology). Our analy- Beagrie RA, Pombo A (2016) Gene activation by metazoan enhancers: diverse ses have shown that chromatin contacts mechanisms stimulate different steps of the transcription cycle. Bioessays extend across the length of each chromo- 38, 881- some (across tens of megabases), and that 893. the hierarchical folding and reorganization of chromosomes are linked to transcrip- Group Leader tional changes during cellular differentia- Prof. Dr. Ana Pombo tion (Figure 2). Polymer physics modeling Rieke Kempfer demonstrated that hierarchical folding of Scientists Dorothee Kraemer Dr. Mariano Barbieri Gesa Loof Dr. Robert A. Beagrie Elena Torlai Triglia Wechromatin apply different promotes approaches efficient to packagingstudy the Dr. Tiago M.C. Rito Marta Oliveira (co-supervised by Dr. mechanismswithout loss ofunderlying contact specificity. complex chroma- Dr. Carmelo Ferrai Alexandra Moreira, Univ Porto, Portugal) tin folding patterns associated with long- Dr. Alexander Kukalev range regulation, and how they might be Dr. Kelly J. Morris Graduate and Undergraduate established through different chromatin Dr. Markus Schueler students epigenetic states. With Prof. Mario Nicode- Dr. Konstantina Skourti-Stathaki Julieta Ramirez Cuellar Dr. Marta Slimak-Mastrobuoni available data on chromatin contacts, both Dr. Christoph Thieme Marisa Saponaro mi (Univ. Naples, Italy), we have shown that- LeonidIzabela SerebreniHarabula PhD students originated by 3C approaches and fluores Giulia Caglio Secretariat cence in situ hybridization (FISH), can be João Rodrigo Dias Regina Keiper Curr.explained Op. Cell with Biol.). the Strings Currently, & Binders we are Switch using Ana Miguel Fernandes (co-super- the(SBS) SBS model model (Nicodemi to explore &the Pombo 3D topology 2014 vised by Dr. Esteban Mazzoni, Michaela Liemen - NY Univ, NY, USA) SusanInes Krock Lübcke tal disease in mouse ES cells and neurons of specific genomic loci affected in congeni

MDC Research Report 2016 197 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Stephan Preibisch

Microscopy, Image Analysis & Modeling of Developing Organisms

Our lab aims at understanding more about Lightsheet microscopy data how gene regulatory networks define The relatively small size of C. elegans and the fact that cells are densely packed in nervous the development of organisms. We ap- system requires high-resolution light mi- proach the understanding of these complex croscopy acquisitions in order to unambigu- systems by developing computational and ously detect all nuclei for mapping. We em- experimental solutions for in-toto visualiza- ploy lightsheet microscopy (Huisken et. al.) to maximize optical resolution while mini- tion, statistical analysis, and perturbation of developing specimen. We combine light & the full potential of lightsheet microscopy electron microscopy, transcription imaging, wemizing develop photo multiview damage. image In order reconstruction, to exploit genetics and deep sequencing with com- image analysis and data management software (Fig. 2 and publications). All resulting puter vision and software development. image data can be integrated into the ‘vir- While our lab is focused on C. elegans as a model organism we additionally collaborate descriptive in-toto representations. tual nematodes’ thereby creating powerful, with different labs on various model or- Single-molecule RNA imaging ganisms providing our computational and We study gene regulation in C. elegans com- imaging expertise. with single-molecule RNA imaging (Fig. bining deep sequencing and ChIP-seq data Ercan (NYU) we currently focus on uncov- High-resolution atlases of ering3). In collaborationmolecular mechanisms with the lab underlying of Sevinc C. elegans We combine different types of microscopic acquisitions into one common computational representation of C. elegans exploiting its high degree of stereotypicity. The

acquisitions of C. elegans are high-resolu- tionbasis spatial for the atlases integration acquired of live using and Serial fixed Section Transmission Electron Microscopy (Fig. 1). Following the reconstruction of the datasets by aligning thousands of two- dimensional image tiles in three-dimen- Figure 1: Acquisition of an entire C. elegans larvae in sional space (Cardona et al.), we segment dauer stage using transmission electron microscopy and annotate the complete volume to cre- (acquired at HHMI Janelia). The imaged specimen has ate computational representations of entire a length of around 500um and a diameter of 25um. The dataset consists of 25.000 individual images covering 561 physical sections totaling a size of approximately allowsC. elegans direct at specific mapping developmental of lower resolution stages. 1 terabyte. The atlas of cellular location and function lightThe fixed microscopy lineage ofdata C. elegans onto thesesubsequently ‘virtual resulting from the ongoing reconstruction will serve as basis for the mapping of light microscopy acquisitions.

nematodes’ using the nuclear locations. 198 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 2: Multiview (MV) deconvolution of Figure 3: Single molecule FISH experiments in C. elegans. lightsheet microscopy data. (a) C. elegans em- (a,b) co-staining for two different isoforms of a gene bryo in 4 cell stage, (b) Drosophila embryo in highlighting the spatial expression pattern in different blastoderm stage. The upper row shows normal developmental stages. (c) computational representation of fusion, the lower row the result of MV deconvo- individual mRNA isoforms and nuclear locations from (b). lution. (c,d) MV Deconvolution of a C. elegans in (d) straightened C. elegans larva used for computational L1 stage, arrows mark the cross-sections shown comparison and statistical pattern analysis. as insets. The upper two rows shows orthogo- nally acquired raw data, the lower row the result of MV deconvolution capable of identifying all 558 nuclei unambiguously.

transcriptional repression of the C. elegans - Kaynig V., Longair M., Preibisch S., Rueden cess compensating the presence of a second C.,Schindelin Saalfeld J., S.,Arganda-Carreras Schmid B., Tinevez I., Frise J.Y., E., X-chromosome, which is an important pro- ing candidate selection based on existing ge- Cardona A. Nat. Meth. 9 X-chromosomenome-wide datasets, in hermaphrodites. we use single-molecule Follow Hartenstein V., Eliceiri K., Tomancak P. & , 676-82 (2012). to visualize individual mRNAs to measure fluorescent in-situ hybridization (smFISH)- Selected Publications sion between wildtype and mutant animals fine spatio-temporal differences in expres Carlas Smith*, Stephan Preibisch*, Aviva Joseph, Bernd Rieger, Sjoerd Stallinga, Eugene

aat four-dimensionalsingle-cell level. The model smFISH of C. experiments elegans de- Myers,Cell Biology Robert 209(4), H. Singer 609-619. and David Grunwald (2015), “Nuclear accessibility of β-actin mRNA measured by 3D single-molecule real time (3D-SMRT) microscopy”, Journal of velopmentperformed onacquired fixed specimen by lightsheet are mapped microsco to- Stephan Preibisch, Fernando Amat, Evangelia Stamataki, Mihail Sarov, Robert H. py using the location of nuclei as reference. TobiasSinger, Pietzsch*,Eugene Myers, Stephan Pavel Preibisch*, Tomancak Pavel (2014), Tomancak “Efficient and Bayesian-based Stephan Saalfeld* Multi-View (2012), at dissecting the mechanism of transcrip- Deconvolution”, Nature Methods 11(6), 645-648. Stephan Preibisch*, Stephan Saalfeld*, Johannes Schindelin, Pavel Tomancak (2010), tionalBy studying repression specific at the mutants molecular we are level. aiming “ImgLib2 – Generic image processing in Java”, Bioinformatics 28(22), 3009-3011.

Software development Stephan“Software Preibisch, for bead-based Stephan registration Saalfeld, Pavel of selective Tomancak plane (2009), illumination “Globally microscopy Optimal Stitch data”,- We develop software solutions for image Nature Methods 7(6), 418-419. - ing of Tiled 3D Microscopic Image Acquisitions”, Bioinformatics 25(11), 1463-1465. taViewer (publications) that we provide asprocessing open-source based to on the ImgLib2 community. and BigDa Most Start of the Group: September 2015 software is made available as user-friendly plugins in Fiji (Schindelin et al.), capable Group Leader of utilizing cluster infrastructure and GPU Dr. Stephan Preibisch hardware. We develop versatile solutions Klim Kolyvanov applicable to many different types of image Postdoc Dhana Friedrich (visting, Prof. Dr. data and biological questions. Dr. Nikita Vladimirov Alexander Löwer, TU Darmstadt) Dr. Varun Kapoor (shared with David Hörl (visiting, Prof. Dr. Heinrich References Leonardt, LMU Munich) Cardona A., Saalfeld S., Schindelin J., Argan- Dr. Yann Le Poul (shared with Dr. Simone Reber, IRI Berlin) - Prof. Dr. Nicolas Gompel, LMU Munich) Technician Andrea Grybowski Oneda-Carreras 7 I., Preibisch S., Longair M., To PhD students Huiskenmancak P.,J., Hartenstein Swoger J., V.Del & Douglas Bene F., R. WittPLoS- Ella Bahry Secretariat , e38011 (2012). 305, 1007–9 Laura Breimann Regina Philipp (2004). brodt J. & Stelzer E.H. Science

MDC Research Report 2016 199 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Nikolaus Rajewsky

Systems Biology of Gene Regulatory Elements

My lab uses experimental (molecular biol- Introduction ogy, biochemistry) and computational meth- A major lesson from recent genomics is ods (bioinformatics, computational biology) that metazoans share to a large degree the to dissect, systems-wide, function of gene same repertoire of protein-encoding genes. regulation in metazoans. One major focus is to understand more about post tran- within a species, between species, or be- tweenIt is thought healthy that and differences diseased between animals cells are scriptional gene regulation. We investigate in many cases due to differences in when, general mechanisms and principles of gene where and how genes are turned on or off. regulation by microRNAs and RNA binding Gene regulatory information is to a large degree hardwired into the non-coding proteins in cell lines and in vivo. For example, parts of the genome. Our lab focuses on we are studying function and mechanisms decoding transcriptional regulation (iden- of post-transcriptional gene regulation during early development in C. elegans. We use transcription factors in non-coding DNA) andtification post-transcriptional and characterization control of targetsmediated of CrispX-Cas to mutate individual regulatory by RNA binding proteins and small, non- sites in vivo. Furthermore, we have estab- coding RNAs, in particular microRNAs. lished planaria as a model system in our lab. microRNAs are a recently discovered large Planaria are famous for their almost unlimited class of regulatory genes, present in virtually all metazoans. They have been shown ability to regenerate any tissue via pluripotent adult stem cells. We investigate molecular mechanisms of pluripotency and the encodingto bind to mRNAs specific and, cis-regulatory by unknown sites mecha in 3’- nisms,untranslated to repress regions protein (3’ UTRs) production of protein- of role of post-transcriptional gene regulation in their target mRNAs. Our understanding planarian stem cell biology and regeneration. of the biological function of animal mi- More recently, we have started to quantify croRNAs is just beginning to emerge, but RNAs in single cells and implemented state- it is clear that microRNAs are regulating or involved in a large variety of biological of-the-art technology in our lab (microfluid- processes and human diseases, such as ics). We routinely capture RNA of thousands developmental timing, differentiation, sig- of individual cells in 15 minutes. We develop nalling, homeostasis of key metabolic gene computational methods to analyze these products such as cholesterol, cardiovascular diseases and cancer. Overall, however, data and start to apply the entire approach it is clear that miRNAs are only a small to understand more about gene expression part of the entire post transcriptional gene in neuronal systems and tumors, which are regulation apparatus used by cells, and we are beginning to systematically explore the both well known to harbor important hetero- largely unknown function of RNA binding geneity at the single cell level. proteins.

200 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Figure 1: Possible functions of circRNAs: (a) binding miRNAs or RNA binding proteins (RBPs) to change their local free concentration (b) transport of miRNAs/RBPs inside the cell (c) acting in trans (d) involved in assembly of other molecules (e) involved in regulating splicing (f) no function.

helped to elucidate the function of microR- gene regulation and in particular of the NAs in pancreatic beta cells (insulin secre- justIt is emerging clear that universe a better of understandingnon-coding RNAs of tion), in liver (cholesterol level), and other can only come by integrating various data systems. We have shown that microRNAs sources (comparative sequence analysis, mRNA expression data, protein-protein in- signatures on hundreds of genes (Sood et teractions, mutant phenotypes from RNAi al.,leave PNAS cell 2006), type specificand that mRNAhuman expressiongenotyped screens, polymorphism data, experimen- SNP data can be used to explicitly demonstrate and quantify the contribution of mi-

alonetally definedis only a gene partial regulatory description networks, of how Rajewsky, Nature Genetics 2007). We have cellsChIP-chip function. data, For etc) example, since each to dataunderstand source furthercroRNA developedtargets to human computational fitness (Chen methods and microRNA function, we not only need to (miRDeep) to predict miRNAs from high identify their targets but also to decode throughput sequencing data (Friedlaender how microRNAs are transcriptionally regulated. A major focus of the lab is therefore also pioneered approaches that allowed to in developing methods that integrate differ- experimentallyet al., Nature Biotechnology assay, genome-wide, 2008). We havethe ent data sources and methods to produce impact of miRNAs on protein synthesis - how, when, and where genes are regulated. going effort is currently to use and develop global and yet specific predictions about- several(Selbach keyet al., high- Nature throughput 2008). A majortechnolo on- tion and functional description of gene reg- gies for in vivo studies in C. elegans and ulatoryThis will networks. ultimately We lead will tocontinue the identifica to test, - LAC), RNA sequencing, and new methods planaria: high-throughput proteomics (SI systems,develop and such “translate” regeneration these in methodsplanaria and of binding sites of RNA binding proteins. earlytheir embryogenesis predictions using in C. specific elegans. biological Verythat allowrecently the published genome-wide new identification biochemical and computational methods to obtain context-dependent, in-vivo and precise maps of genome-wide miRNA:target interactions andSpecifically, could later we haveon show developed that microRNAs one of the (Grosswendt et al. Molecular Cell 2014). veryfirst microRNAlikely regulate target (in findinga functionally algorithms important way) thousands of genes within Circular RNAs

- We have very recently shown that single- vertebrates, flies, and nematodes (Krek et- stranded circular RNAs (circRNAs) are a putationalal., Nature Biology Genetics 2006). 2005; We Lall have et al.,further Cur widespread class of animal RNAs with tis- rent Biology 2006; Gruen et al., PloS Com

MDC Research Report 2016 201 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

Development in C. elegans

Although C. elegans is one of the most famous model systems for developmental biology, it has been impossible to use most high-throughput technologies to study differential gene expression and networks during very early embryogenesis (for example the oocyte to one-cell embryo transition upon fertilization). To overcome this problem, we have developed a novel meth-

Figure 2: Single Cell Sequencing via “Drop-Seq”. We have at precise stages during embryogenesis via implemented a recent advance in single-cell sequenc- FACSod (“eFACS”) sorting that (Stoeckius, allows us Maaskola to sort embryos et al., ing technology (adapted from Macosko et al. Cell 2015). Nature Methods 2009). For example, we Each bead carries oligos which can bind the polyA tail of can now routinely obtain ~60,000 one-cell mRNAs/lncRNAs. Single cells enter in suspension, beads in lysis buffer. Each oligo contains a bead-specific barcode. FACS run, enough to apply virtually any Therefore, the contents of all droplets can be amplified, se- high-throughputstage embryos (at method a purity of >98%)interest. in oneWe quenced as one library, and de-convolved computationally have used eFACS to assay the dynamics of later on. The approach is the fastest and most inexpensive small RNA expression during embryogene- method to date. We capture 4,000 “transcriptomes’ in 15 sis. We discovered a wealth of orchestrated, minutes runtime (Alles et al. unpublished). - - ingsspecific open changes the door between for many and computational within virtu - andally functional all classes follow of small up RNAs.studies. These For exam find- pression (Memczak et al. Nature 2013). ple, we have shown that sperm transmits circRNAssue and developmentalare usually expressed stage specific from ex- RNA into the zygote, and that hundreds of ons, are found in the cytoplasm, and are paternal mRNAs/small RNAs are present unusually stable. For a particular circRNA in the one-cell embryo (Stoeckius et al., (CDR1as) we could show that this circRNA submitted). Very recently we have estab- can function as a sponge to sequester miR- lished CrisR-Cas in C. elegans (Froehlich et - al. unpublished) and are currently probing tigating possible functions of circRNAs and the function and regulatory mechanisms of mechanisms7, a neuronal ofspecific circRNA miRNA. biosynthesis We are inves(Ash-

et al. Cell Reports 2015, Rybak-Wolf et al. Stembinding cell sites biology in 3’ untranslated regions. Molecularwal-Fluss etCell al. 2015).Molecular We Cellcollaborate 2014, Ivanov with the Kadener lab, Hebrew University, Jerusa- We used massive next generation sequenc- lem). However, it is also possible that many ing to identify miRNAs and piRNAs in S. circRNAs are inert by-products of splicing (Figure 1). Even in this case, because of their unusual stability and differential ex- biology.mediterranea. A number We also of these identified miRNAs miRNAs are pression, it will be extremely interesting to that seem specifically linked to stem cell- study circRNAs as possible biomarkers for amidi et al., PNAS 2009). We have further processes (Memczak et al PloS One 2015). developedconserved inexperimental humans (Friedlaender and computa & Ad- Generally we have found that hundreds of tional methods that allowed us to assemble circRNAs are highly abundant and often the planarian transcriptome via next generation sequencing of mRNAs (Adamidi et that we have examined (Rybak-Wolf et al. al., Genome Research 2011). We have used Molecularspecifically Cell expressed 2015. Wein each have neural also develtissue- FACS and subsequent RNA-seq and shotgun oped assays to pull down circRNAs after proteomics to obtain planarian stem cells RNA:protein crosslinking with subsequent mass-spec to identify proteins which are expressed in them. Comparison to mamma- lianand embryonic to define which stem cells genes revealed are specifically that mo- non-crosslinked pull down as a control, lecular mechanisms for pluripotency are Stottmeisterspecifically bound et al. unpublished). to circRNAs (usingThese theas- deeply conserved throughout life and that says allow the mechanistic understanding planarian stem cells are indeed informative of knock-down phenotypes that we ob- for mammalian stem cell biology (Oenal et serves for circRNAs. al. EMBO J 2012). Part of this work involved

202 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

collaborations with the Sanchez lab (Utah), Chen, Dieterich, and Kempa labs (MDC). Currently we are studying the role of al- freevivo concentration.bound passively We by have DICER- during expression the past ternative splicing in pluripotency. We have yearsof this developed transcripts simple therefore quantitative inhibits DICER mod- discovered that two alternative splice regu- els which can be used to study systemati- lators bind proximal to the same set of ex- cally and quantitatively such competition ons to antagonistically regulate alternative effects- expression of a RNA with binding splicing. One of these factors is expressed sites can lead, under certain stoichiomet- in stem cells, the other one in differenti- ric conditions, to reduced availability and therefore binding of the regulator to other control functionally important and cell type binding sites in the transcriptome (Jens and ated cells. In this way both factors together Rajewsky Nature Reviews Genetics 2015). (Solana et al submitted). specific inclusion or exclusion of exons Towards systematic decoding of the “post-transcriptional regulatory code”

We have started to use biochemical methods Selected Publications to investigate mechanisms and function of Memczak S, Jens M, Elefsinioti A, Torti F, Krueger J, Rybak A, Maier L, Mackowiak SD, post-transcriptionalsuch as PAR-CLIP (Hafner gene etregulation. al., Cell 2010) For Gregersen LH, Munschauer M, Loewer A, Ziebold U, Landthaler M, Kocks C, le Noble example, we have studied the human RNA F, Rajewsky N (2013). Circular RNAs are a large class of animal RNAs with regula-

Rybak-Wolf A, Jens M, Murakawa Y, Herzog M, Landthaler M (co-corresponding), functionally relevant targets of HuR and Rajewskytory potency. N* (2014).Nature 495,A variety 333-338. of Dicer substrates in Human and C. elegans. Cell couldbinding show protein that HuR.HuR regulatesWe identified mRNA >4000 pro- 1153-67. cessing and alternative splicing (Lebedeva et al., Molecular Cell 2011), previously un- EY, Landthaler M, Dieterich C, Rajewsky N (2015). Analysis of intron sequences known functions of HuR. Very interestingly, Ivanovreveals A,hallmarks Memczak of S, circRNAs Wyler E, biogenesisTorti F, Porath in animals. HT, Orejuela Cell Reports MR, Piechotta 170-7. M, Levanon we also found that HuR directly and strongly controls the expression of one conserved Rybak-WolfM, Refojo D, A,Kadener Stottmeister S, Rajewsky C, Glažar N (2015). P, Jens M,Circular Pino N, RNAs Giusti in S,the Hanan mammalian M, Behm brain M, human microRNA (miR-7). Ultimately we Bartok O, Ashwal-Fluss R, Herzog M, Schreyer L, Papavasileiou P, Ivanov A, Öhman need to study regulatory relationships not Jens M and Rajewsky N (2015). Competition between target sites or regulators shapesare highly post-transcriptional abundant, conserved, gene and regulation. dynamically Nature expressed. Reviews Molecular Genetics 113-26 Cell 870-85. - (Analysis Paper). ciple experiment, we have shown that it is possiblein cell lines to biochemically but in vivo. In identify a proof targets of prin of RBPs in C. elegans and at nucleotide resolu- Group Leader tion (Jungkamp et al., Molecular Cell 2011). Prof. Dr. Nikolaus Rajewsky Panagiotis Papavasileiou Marta Rodriguez-Orejuela to systematically identify the targets and to Scientists Christin Stottmeister studyThis new the function method (“iPAR-CLIP”)of any RBP in C. allows elegans us Dr Filippos Klironomos Marcel Schilling One immediate goal was to identify the tar- Dr Christine Kocks (Associated Kathrin Theil Jonathan Fröhlich of small RNA pathways in vivo (Rybak-Wolf Dr Markus Meyer * Petar Glazar gets of DICER and other central components DrPI) Agnieszka Rybak-Wolf Marco Uhrig (supervised by B. Dr Benedikt Obermayer (Del- Obermayer) et al Cell 2014). Together with a DICER PAR- brueck Fellow) CLIP in human cell lines performed by the Dr Jordi Solana-Garcia Diploma Students Landthaler lab, we discovered that DICER Dr Ulrike Ziebold Jonathan Alles* harborhas not smallonly “active” hairpin targetstructures sites which are Dr Evegenia Afanasieva* cleaved. In fact, many mRNAs and lncRNAs Dr Nikolaos Karaiskos* Technical Assistants - Salah Ayoub sivespecifically sites are bound biologically by DICER active but not and cleaved that Graduate Students Margareta Herzog they(“passive” are enriched sites). We in mRNAscould show which that encode pas Andrei Filipchyk Luisa Maier Stephanie Grosswendt example proteins found in P-bodies (or Secretariat granules).proteins which These interactmolecular with networks DICER, are for Marvin Jens Alex Tschernycheff conserved from C. elegans to man. We also SebastianAndranik Ivanov*Mackowiak* found that a lncRNA is highly covered with Sebastian Memczak (* part of the time reported)

passive DICER binding sites and indeed in MDC Research Report 2016 203 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: Keih Happell Keih Photo:

Roland Schwarz

Evolutionary and Cancer Genomics

Our group was founded in October 2016 and investigates the aetiology and func- construct all possible paths of converting define edit operations on sequences and to- tional implications of intra-tumour heterogeneity using computational approaches, one such sequence into another. In MED- with a focus on chromosomal rearrange- traryICC, FSTslength define on the biologicallygenome, which meaningful are then ments, somatic copy-number alterations usedevents to of compute amplification the minimum and deletion number of arbi of such events to transform one genome into and changes in chromatin structure. To this another as an estimate of evolutionary di- end, we develop and apply machine learn- vergence. ing and statistical methods to reconstruct We applied this method to a new cohort of patients under treatment for high-grade the evolutionary history of cancer in the serous ovarian cancer (HGSOC). We dem- patient and associate germline and somatic onstrated ongoing somatic copy-number variation with changes in gene regulation. Our group currently consists of two PhD mutator phenotypes and demonstrated associationsevolution in between HGSOC, different identified degrees potential of students and one visiting PhD student from London. We collaborate with a wide range (Schwarz et al. 2015). of experimental labs within and outside the ITH and patient progression and survival MDC to tackle the most pressing questions Establishing between-genome phasing for copy-number variation in cancer genomics. The accuracy of tree inference from genomic rearrangements depends on the quality of the phasing of copy-numbers: the assign- Previous work – reconstructing ment of major and minor copy-numbers the evolutionary history of cancer to the two physical parental alleles. So far Cancer is characterised by ubiquitous so- phasing has been done using evolutionary matic alterations, such as single nucleotide criteria alone, a heuristic and computation- variants (SNVs), somatic copy-number al- ally expensive procedure which impedes terations (SCNAs) and genomic rearrange- probe-level resolution tree reconstruction ments, that shape the genetic landscape (Schwarz et al. 2014). of cancer and contribute to intra-tumour We are currently developing and applying - a novel phasing algorithm for SCNAs from - cancersheterogeneity with substantial (ITH) and SCNAs, genome we plastic devel- ing the shared genetic background of mul- ity. To understand the aetiology of ITH in tiplemulti-region samples tumour from the profiles same (Figurepatient we1). Usas- sign copy-numbers to physical alleles based anoped algorithm MEDICC for(Minimum the reconstruction Event Distance of thefor on the bi-allelic frequency (BAF) distribu- evolutionaryIntra-tumour history Copy-number of cancer Comparisons), in the pa- tion of heterozygous single-nucleotide poly- - previously established evolutionary phas- tient from allele-specific copy-number pro ingmorphisms algorithm (SNPs). this provides In combination a new, accuratewith our anfiles algorithmic (Schwarz etframework al. 2014). heavily The algorithm used in and fast phasing method which leverages naturalis based language on finite-state processing, transducers that allows (FSTs), to the available SNP data effectively. This is a

204 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

A:Figure Schematic 1 – Reference of the reference phasing forphasing: allele-specific Using a copy-number profiles the two parental alleles can be distinguished commonbased on geneticthe BAF background, patterns. Only amplifications heterozygous on probes (white) are informative. B: Example of an inverted BAF pattern in HGSOC: Within a seg- mental duplication of CN 3 (red square), sample 2 shows an inverted BAF pattern compared to - pened on the opposite allele. Breakpoints are theindicated reference, as green indicating vertical the lines, amplification average total hap DNA content (logR) is shown in yellow.

crucial step towards probe-level resolution on the same NGS read. Our preliminary re- tree inference on genomic rearrangement sults demonstrate a strong effect of SCNAs

genetic heterogeneity for routine applica- ASE variability, with germline and somatic tionsevents in in translational cancer and exactcancer quantification research. of pointson ASE, mutations accounting both for contributing about 80% roughlyof total equally to the remaining variability. Assessing the effect of somatic variation on the cancer regulatory non-coding SNVs contribute to cancer gene landscape using deep learning regulationIt is so far throughlargely uncleardisruption to whichof transcrip extent- With the tools in hand to accurately reconstruct tumour evolution, the question chromatin states. To estimate such func- remains which somatic events are func- tionaltion factor effects binding of individual sites or modificationSNVs de-novo, of we train a convolutional neural network, progression, and which do not. As part of - tional, i.e. contribute to cancer fitness and genetic features, including transcription Whole Genomes (PCAWG) working group, to predict regulatory effects on 918 epi wethe are TCGA/ICGC currently investigating Pan-Cancer the Analysis effect of different types of somatic variation on the tofactor predict binding the sites, regulatory DNase Ieffect sensitivity of ~10M and regulatory landscape across different can- phasedhistone modifications.SNVs from PCAWG, We use using this model ran- cer types. Using 1200 whole-genome se- domised null variants to identify SNVs that quenced cases with tumour and matched normal samples and associated RNAseq were significant positively and negatively associated with specific regulatory marks. readoutsprofiles, and we mapquantify somatic the effects and germlineof differ- Selected Publications entvariants mutation to allele-specific types. expression (ASE) Spatial and temporal heterogeneity in high-grade serous ovarian cancer: a phyloge- To be able to assess the directionality of netic analysis. Schwarz RF, Ng CK, Cooke SL, Newman S, Temple J, Piskorz AM, Gale D, Sayal K, phasing of both germline as well as somatic Murtaza M, Baldwin PJ, Rosenfeld N, Earl HM, Sala E, Jimenez-Linan M, Parkinson the change in allele-specific expression, - mosomal loss or gain, the phase of germ- SchwarzCA, Markowetz RF, Trinh F, Brenton A, Sipos JD. B, Brenton– PLoS Med. JD, Goldman 2015 Feb N, 24;12(2) Markowetz F. – PLoS Comput linevariants variants is essential. is directly In segmentsevident from of chro the Phylogenetic quantification of intra-tumour heterogeneity. Cancer evolution: mathematical models and computational inference. BAF, where within a SCNA segment SNPs on Biol. 2014 Apr 17;10(4) the same haplotype co-cluster in the corresponding BAF bands. We combine this Beerenwinkel N, Schwarz RF, Gerstung M, Markowetz F. – Syst Biol. 2015 Jan;64(1) phasing with statistical phasing on regions Start of the Group: October 2016 Marina Petkovic lacking SCNAs to achieve near-to-complete Tom Watkins long-range phasing for germline and SC- Group Leader NAs. This allows for the accurate aggrega- Dr. Roland Schwarz London, UK) (visitor from The Francis Crick Institute, tion of allelic expression to quantify gene- level ASE. We then phase individual somatic PhD Students Secretary variants to the nearest germline heterozy- Maria Stella de Biase gous site, in instances were both co-occur Ines Krock

MDC Research Report 2016 205 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Baris Tursun

Gene regulation and cell fate decision in C. elegans

Understanding mechanisms which safe- the ectopic induction of neuronal fates in guard cellular identities is key to improv- non-neuronal tissues such as germ, epider- mal, and intestinal cells. ing the reprogramming of differentiated cells. We use Caenorhabditis elegans as The FACT complex safeguards cell a genetic model organism to systemati- fates and counteracts reprogram- cally study the molecular basis of cell fate ming - maintenance and protection. C. elegans erone complex FACT and the associated provides genetic tractability and visualiza- histoneWe identified chaperone subunits SPT6 of theto histoneact as barrichap- tion of cell conversions in vivo. We utilized ers for induced cellular reprogramming. Further characterization of these factors large-scale genetic screens to identify fac- revealed that reprogramming upon loss of tors safeguarding cell fates and acting as FACT or SPT6 depends on the activity of a barrier for direct conversion of cells to a different identity (Transdifferentiation / Di- for the phosphorylation of Histone H3, therebythe Aurora-Kinase triggering AIR-2. activation AIR-2 ofis requiredsilenced rect Reprogramming). Additionally, we are studying the interconnection of Aging and that the FACT complex protects repressed Cellular Reprogramming (Figure A). Our chromatin domains.to prevent Our activation findings of suggest gene findings could help overcome the limita- type of gene regulation links FACT and the tions of Direct Reprogramming in order to Aurora-Kinaseexpression, triggered B to safeguarding by AIR-2. This of newcel- generate the cellular material for prospec- lular identities and, importantly, may be tive tissue replacement therapies. conserved because these factors have more - man counterparts. Moreover, in collabora- tionthan with 90% S. amino Diecke acid (Stem homology Cell Facility) with huwe Genetic screens to identify barri- showed that FACT acts as a barrier for iPSC ers for direct reprogramming A straightforward strategy to identify factors counteracting direct reprogramming reprogramming also in human fibroblasts (barrier factors) is to perform genetic germ(Fig. B).”cells Notably, against wereprogramming previously identified (Tursun screens, which can easily be applied to that the histone chaperone LIN-53 protects C. elegans. Using mutagenesis and RNAi study, the CAF-1 complex, which consists of screens we systematically interrogate the et al., 2011 Science). In an analogous mouse entire genome of animals for ‘barrier fac- the LIN-53 homolog CAF-1p48/RBBP4/7, allow mis-expression of cell fate-inducing etwas al., recently Nature identified2015). The by marked another conserva group as- mastertors’ in transcriptionvivo. We use transgenicfactors combined worms withthat tiona barrier of function for iPSC for reprogramming the same factor, (Cheloufi from

conversions in vivo. Our mutagenesis and - whole-genomefluorescent fate RNAi reporters, screens to (knock-down monitor cell lularworm reprogramming, to mouse, gives can us be confidence translated that to highermore C. organisms elegans findings, and, importantly, with regard human to cel 160 novel factors, whose depletion allows tissues. of 20K individual genes) identified around

206 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

with Lamin mutations (Figure D). Many laminopathy patients suffer from myopathy and some show symptoms of premature aging that parallel the phenotypes of worms

- alitylacking causes LIN-53. the symptoms Hence, it of is laminopathy conceivable patients.that affecting Currently, LIN-53 we levelsare collecting and function more

homeostasis and aging regulation in both wormsevidence and that humans.Our LIN-53 plays aim a role is into musclebetter understand the mechanisms which safeguard cell fates in differentiated cells and The histone chaperone LIN-53 is a in the context of Aging. Our investigations barrier for cellular reprogramming and regulates Aging and muscle mechanisms due to the fact that the C. el- maintenance may lead to the identification of conserved As mentioned above, the histone chap- human genome. egans genome is ~60% homologous to the as a barrier for reprogramming in worms anderone mammals. LIN-53 (CAF-1p48/RBBP4/7)Furthermore, our ongoing acts Selected Publications of cellular reprogramming revealed that Notchcharacterization signaling promotes of LIN-53 reprogramming in the context by counteracting the chromatin-regulating Seelkreprogramming S*, Kalchhauser of germ I*, cellsHargitai in C. B, elegans. Hajduskova elife. M,2016 Gutnik 5. pii: S, e15477Ciosk R#, Tursun B#. Polycomb Repressive Complex 2 (PRC2), CochellaNotch signaling L*, Tursun antagonizes B*, Hsieh PRC2-mediated YW, Johnston RJ, silencing Chunag onCF, specific Hobert genesO. Two to distinct promote types of neuronal asymmetries are controlled by the Caenorhabditis elegans tran-

Tursun B. Cellular reprogramming processes in Drosophila and C. elegans. Curr Opin Seelkwith whichet al., in LIN-53 revision). associates Yet in different in order tis to- scription factor die-1. Genes Dev. 2014; 28: 34-48 safeguard germ cell identity (Figure C; Patel T*, Tursun B*, Rahe, DP, Hobert O. Removal of Polycomb Repressive Complex from counteracting reprogramming. Fur- Genet Dev. 2012; 22: 475-484. sues,ther LIN-53characterization appears toof play lin-53 distinct mutants roles Tursun2 makes B#, C. elegans Patel T, germKratsios cells P, susceptibleHobert O#. Directto direct conversion conversion of intoC. elegans specific germ somatic cells cell types. Cell Rep. 2012; 2: 1178-1186 maintenance and normal aging. Mutants showrevealed strong that motility LIN-53 isdefects crucial in for addition muscle into specific neuron types. Science 2011; 331: 304-308 to a shortened lifespan. Genetic analysis of *co-first, #corresponding the muscle and aging phenotypes revealed Group Leader Andreas Ofenbauer Dr. Baris Tursun Stefanie Seelk premature aging by affecting the chromatin remodelingthat loss of LIN-53complex connects NuRD and myopathy the histone and Scientist Undergraduate Dr. Anna Reid Anne Sommermeier the muscle and aging phenotypes can be deacetylase complex SIN3. Intriguingly, PhD students Technical Assistants - Selman Bulut Alina El-Khalili ininguncoupled human by samples specifically of laminopathy depleting NuRD pa- Nida ul Fatima Sergej Herzog or SIN3, respectively. Moreover, by exam Mei He Marlon Kazmierczak Secretariat proteinstients (collaboration are diminished with or Simonemis-localized Spuler’s to Ena Kolundzic thegroup), cytoplasm we discovered in primary that human human myoblasts LIN-53 Ines Krock

MDC Research Report 2016 207 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY Photo: David Ausserhofer/MDC Photo:

Robert P. Zinzen

Systems biology of neural tissue differentiation

How do cells acquire and maintain their identity? This question remains a central is- sequencing (SEQ) to identify the genome- tation (ChIP) followed by next generation sue in developmental and stem cell biology. TFs. We aim to identify target genes and the Our lab aims to understand the molecular CRMswide bindingthrough which profiles TFs of act. key This neurogenic is crucial mechanisms undifferentiated embryonic for understanding the regulatory network cells employ to organize into distinct cell that shape the NS. A more encompassing view of genome types and tissues. With the advent of mod- regulation can be gained by investigating ern ‘-omics’ approaches, it is becoming possible to investigate embryonic development - hancers,chromatin promoters, state dynamics. etc., but Specific also ofhistone their on genome wide scales. No longer limited to regulatorymodifications state. are notHowever, only predictive since chroma of en- asking how individual genes are regulated, tin features are employed differentially we can now ask how the genome is regulated to coordinate development. Further- - tionthroughout of the cell the populations embryo, identificationof interest. BiTS of more, new methodologies allow increased the tissue specific signatures requires isola- resolution of these approaches – from tissue - specificity to even individual cells. ulations(Batch isolation by FACS, of which Tissue are Specific then usedchroma for tin) allows the isolation of defined cell pop Because of its genetic tractability and ex- signatures in the NS and its components perimental accessibility, we approach this withChIP-SEQ. exquisite We are resolution uncovering in space the chromatin and time. question in embryonic development of the The functional integration of TF binding and Drosophila melanogaster. Using ner- chromatin state data remains an important vous system (NS) development as a model, goal to assess how TFs shape their chroma- wefruit focus fly on how cell populations subdivide tin environment and how the chromatin environment regulates the interaction of TFs and NS components. Our aim is to under- with the genome. standand give the rise molecular to specific mechanisms neuroblasts underly (NBs)- - Transcriptomes resolved in space nation of NBs, and their differentiation. & time ing NS specification, the regulated delami P. Wahle, A. McCorkindale Transcription factor & chromatin landscapes A. Guimaraes, A. Glahs orA caveat predict of transcriptional ChIP-SEQ studies responses. is that binding To as- - sessdata thecannot effect definitively individual identify TFs have target on globalgenes tiation in development is driven largely via transcription in vivo, we have established spatiotemporalCoordinated specification regulation of and gene differen expres- genotypes to isolate homozygous mutant sion. Whether a cell expresses a particular embryos and comprehensively assess the effect of key neurogenic TFs on developmen- regulatory elements (CRMs) interpret the tal transcriptomes and lineage decisions in cellulargene is environment, directed by howe.g. which that gene’stranscrip cis-- - tion factors (TFs) are recruited to CRMs. scriptomic and TF binding, we aim to identi- We are utilizing chromatin immunoprecipi- fythe the NS. regulatory By combining effects cell of typeneurogenic specific TFs. tran A

208 MDC Research Report 2016 BERLIN INSTITUTE FOR MEDICAL SYSTEMS BIOLOGY

regulatory level that remains understudied are noncoding RNAs (ncRNAs). We are systematically assessing both the spatial and the temporal expression dynamics of various types of ncRNAs (i.e. micro-, circular-, and long ncRNAs) in order to investigate their functional roles in neurogenesis.

Toward single cell resolution Philipp Wahle, with the Rajewsky lab, esp. Nikos Karaiskos New methodologies make it possible to extract transcriptome information from

earlysingle 6000-cellcells. In a embryo,pilot project sequenced with Nikolaus its in- dividualRajewsky’s cells, lab, and we havemapped deconstructed them back the to their original embryonic positions. What emerges is a virtual embryo – a single-cell transcriptome-resolved blueprint of development: individual cells and positions can be queried for their global gene expression programs (www.dvex.org). This yields unprecedented insight into the combinatorial gene expression programs that drive development, cellular heterogeneity and even reveals evolutionary gene expression divergence. Applying this proof-of-principle to nervous system development promises new insights into the regulation of neurogenic fate decisions. Dissection of the Drosophila NS. A) Early signs of NS subdivision in Network extension the cellularizing embryo. B) Schematic representation. C) BiTS work- S. Krüger, A. Guimaraes flow. D) Embryos stained for tissues and cell types to be sorted as indi- A central aim of the lab is to gain a mechanis- cated at (L->R) st.9, st.11, and st.13. E) Sample tracks of developmental tic understanding of how individual CRMs transcriptomic data from neuronal (red) and non-neuronal cells (blue) translate their cellular environment into for (L->R) a ubiquitous, neuronal, non-neuronal, and a gene showing tissue-specific temporal regulation. that our provisional networks of NS devel- opmentregulatory are (in)activity; still very incomplete. however, Toit is extend clear Selected Publications - - tionthe NS of networks,individual we CRMs have together created flieswith thatthe lates chromatin state to drive cell fate decisions. EMBO Journal 35 (1): 1-3 (2016-01- proteinsallow for thattissue-specific bind to them biochemical in vivo. Usingisola Glahs04) A.; Zinzen R.P. Putting chromatin in its place: the pioneer factor NeuroD1 modu mass spectrometry to identify the bound -

proteins, we are particularly interested in Bonn S.; Zinzen R.P.; Perez-Gonzalez A.; Riddell A.; Gavin A.C.; Furlong E.E. Cell type how CRM activity is affected by both, the specific chromatin immunoprecipitation from multicellular complex samples using temporal dynamics of CRM-protein inter- BiTS-ChIP. Nature Protocols 7 (5): 978-994 (2012-05) actions across development, as well as in Zinzen R.P.; Bonn S.; Girardot C.; Gustafson E.H.; Perez-Gonzalez A.; Delhomme N.; - Ghavi-Helm Y.; Wilczynski B.; Riddell A.; Furlong E.E. Tissue-specific analysis of chromatin state identifies temporal signatures of enhancer activity during embryonic- plexes. With this method, we are not only dictsdevelopment. spatio-temporal Nature Geneticscis-regulatory 44 (2): activity. 148-156 Nature (2012-02) 462 (7269): 65-70 (2009- 11-05) approachingthe tissue specificity a mechanistic of CRM-protein understanding com Zinzen R.P.; Girardot C.; Gagneur J.; Braun M.; Furlong E.E. Combinatorial binding pre

are able to probe novel interactors for their Zinzen(2006-07-11) R.P.; Senger K.; Levine M.; Papatsenko D. Computational models for neurogenic gene expression in the Drosophila embryo. Current Biology 16 (13): 1358- 1365 globalof individual roles. CRMs, but once identified, we Group Leader Sabrina Krüger biology approach to uncover the interdepen- Dr. Robert P. Zinzen Philipp Wahle denciesOverall, theamong lab’s connected aim is to regulatoryutilize a systems levels Ana Luisa Ferreira Guimaraes Silva (e.g. TF binding dynamics, chromatin land- Scientist scape, transcriptional output, and the regu- Dr. Alexandra McCorkindale Technical Assistants latory networks involved) to establish a pre- Sarah Ugowski dictive model of NS development: we want PhD students Claudia Kipar to understand the regulatory interactions Ruben Lopes Pereira Abreu driving tissue development in the developing (joint with M. Landthaler) Secretariat embryo to a degree that allows modeling the Alexander Glahs global effects of systematic perturbations. Ines Krock

MDC Research Report 2016 209 210 MDC Research Report 2016 The Experimental and Clinical Research Center (ECRC)

Friedrich C. Luft

MDC Research Report 2016 211 The Experimental and Clinical Research Center (ECRC)

Friedrich C. Luft

he ECRC is a joint project of the MDC and the Charité Medical Fac- Klinikum-Berlin, a large private hospital T ulty. The mission of the ECRC is to adjacentgroup leaders to the are campus. clinicians at the HELIOS- pursue basic science directly relevant to Recruitment, mentoring, and interactions clinical medicine. The mission statement are our main tools at the ECRC. To entice could be: “We bring together the courage - and imagination of basic researchers and ated a research fellowship for clinicians at clinician/scientists”. The idea of bringing theclinicians MDC years into aago. scientific This fellowship career, we for initi two novel ideas to patient care more quickly years with extension enables clinicians to

concept is not new. For instance, insulin additional degree. The program has been wasis termed discovered “translational” in 1921 and research.given to a dia The- veryaccrue popular scientific and expertise was thereafter and possibly adapted an betic patient in 1922. The Nobel Prize was institution-wide by the Charité. Subsequent awarded for this work in 1923. The ECRC is there to keep that goal alive, but we are two possibilities are attachment to an exist- ingcareer ECRC development group or establishing is a problem; a new however, inde- panel founding the ECRC recognized that pendent group. Another challenge is bring- thenot naïve.ECRC needed The 2005 dedicated international well-trained review ing MDC scientists and clinician/scientists clinicians heading research groups. These together. We have developed a partnership clinicians deserved a state-of-the-art infra- granting system to that end to foster such structure. Equally important, the clinicians collaborations. required world-class basic scientists, who The ECRC has a facility that exists nowhere are committed to working on clinical prob- else in Berlin, although that state-of-af- lems with clinicians that are generally less fairs may change in the future. We have a knowledgeable in the science. These mar- designated clinical research unit (CRU) to riages are seldom made in heaven. Who study humans, outside of the health-care remembers Bertram Collip, the biochemist system and solely for the purposes of science. This unit is operative since 2000 and could actually be given to a human? With- has been extremely productive in fostering outwho him, purified no treatment, insulin so nothat prize, the compound and more investigator-initiated mechanistic research. importantly no cure. Much very valuable clinical research can be The ECRC was designed as an interface be- performed in a patient-care setting. Most tween MDC scientists and Charité clinician/ scientists. We are located in a previous hos- this venue. For instance, comparing tablet pital building named after Robert Rössle, a clinical research viewed by the public fits Charité pathologist in the Virchow line of compared to old cancer treatment Y, or coat- succession. ECRC groups acquire laborato- edA with stent tablet Q versus B, or bare-metal new cancer stent treatment R needs X ry space and infrastructure through compe- no special unit because extra resources to test these questions are minimal. More and the contract can be renewed, again on tition. The space is provided for five years Just suppose we develop a new mesenchy- is measured in publications and grant sup- difficult are studies involving “new ideas”. port.a competitive Not all, but basis. most Scientific ECRC groups productivity are led syndrome. The commercial application is by clinicians who also have patient-care re- notmal obvious,stem-cell-derived the condition therapy is not for that fragile-X wide- sponsibility at one of the Charité campuses spread, and the health-care system is not and/or in our outpatient facility. Two of our enamored with the idea of supporting this

212 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ) Photo: David Ausserhofer/MDC Photo:

Fig. 1. Philipp Du Bois and Jens Fielitz explain their project to Angela Merkel, while Friedrich Luft looks on.

endeavor. That is where the CRU steps in. process is today conducted under the The uniqueness of the ECRC is the mission. We are at the source of medically related This idea was not lost on the reviewers of basic science. The MDC is the best broad- trendy catchword “translational” research. based institute that Germany has to offer. - the Berlin-Institute of Health reviewers of via that source. Our medical school has the 2012. These reviewers specifically recom entireIdeas outsidecity of theBerlin MDC at couldits disposal. be channeled Thus, ourmended design more was financial the prototype. outlays forOur patient- CRU is patient recruitment (from 4 million inhab- modeledoriented after research the US in thegeneral form clinical of CRUs; re- itants) is possible as is nation-wide recruit- search centers and it has functioned well since 2000. We have succeeded in making our job. We admit that we cannot guarantee the CRU independent of any patient-care beatingment. We the view diabetes “translational team of Banting, research” Best, as endeavors that concern third-party payers McLeod, and Collip. But this goal is a worth- - while aim. cians, nurses, scientists, technical experts, and(cross students. financing). We have recruited physi Research on human subjects in Our CRU is located in an earlier hospital the Clinical Research Unit (CRU) building and relies on excellent facilities Patient/proband-oriented research is the included the former new bone-marrow transplant unit. We are well equipped for to humans or in revealing physiological cardiovascular and metabolic studies with orfinal pathological step in bringing mechanisms medical in innovations man. This resources not available elsewhere, in- bench-to-bedside and bedside-to-bench cluding a normobaric hypoxia chamber, a

MDC Research Report 2016 213 ECRC (JOINTLY WITH CHARITÉ)

total-body metabolic chamber, and a large supporting our outpatient clinics covering gamut of techniques including microdialy- - sis, spiro-ergometry, muscle-nerve record- ics, we have excellent ultrasound capabili- ings, lower-body negative pressure, insulin tiesvarious including disciplines. state-of-the-art In the outpatient echocardiog clin- and glucose-clamp studies, and indirect raphy and a laboratory aimed at biobanking calorimetry. We also conduct detailed body samples appropriately. We have added a composition measurements. We adapt our genetic screening and mutation-detection capability that meets German patient-care clinical investigator at the Charité is eligible standards and have a physician with the tocapabilities use our facility to the investigator’sand its infrastructure needs. Any after applying to and receiving approval by such investigations. required additional board certification for can now add biobanking and data-manage- Human Data Management mentour Institutional strategies that Advisory have been Committee. added to the We For data management of all human data, program. we use REDCap, a clinical data manage- We have not focused on inpatient studies ment system (also called electronic case but could do so if required. As a result, for report form) developed at Vanderbilt Uni- pharmacological investigations we cur- versity in 2004. The REDCap Consortium is rently are not doing phase 1 studies but - conduct phase 2 and phase 3 clinical trials, ners in 71 countries and is currently used although our focus has been primarily on composed of 873 active institutional part physician-investigator initiated mechanis- 116,000 users. Thus, REDCap is one of the tic research. We have cooperated actively mostin more frequently than 89,000 used clinical projects data with manage over- with the MDC magnetic resonance imaging ment systems worldwide. Our group has center, particularly with Jeanette Schulz- joined the REDCap consortium. The entire Menger. Our mission is also directed at program can be inspected at: www.project-

Fig. 2. Angela Merkel PhD, a physicist, tackles the atomic force microscope at the ECRC. Photo: David Ausserhofer/MDC Photo:

214 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

Fig. 3. Our CRU organizational concept is adapted from the general clinical research center idea formulated by NIH. We supply infrastructure and various services for any investigator approved by the Institutional Advisory Committee.

redcap.org. We now manage all our outpa- Recent CRU Publications tient contacts with an electronic medical records system that we installed in 2014. Heusser K, Tank J, Brinkmann J, Schroeder C, May M, Großhennig A, Wenzel D, Died- rich A, Sweep FC, Mehling H, Luft FC, Jordan J. Preserved autonomic cardiovascular - ally respected informatics and healthcare researchThe Regenstrief organization, Institute recognized is an internation for its regulation with cardiac pacemaker inhibition: a crossover trial using high-fidelity Lindschaucardiovascular C, Mai phenotyping. K, Boschmann J Am M, Heart Rahn Assoc.G, Movsesian 2016;5(1). MA, Müller T, Doescher A, Gnoth role in improving quality of care, increasing TokaS, Mühl O, TankA, Toka J, Schächterle HR, Wefeld-Neuenfeld C, Aydin A, MaassY, Utz W, PG, Töpper Elitok A,S, Bartels-KleinJordan J, Schulz-Menger E, Hollfinger J, I, - Klussmann E, Bähring S, Luft FC. Clinical effects of phosphodiesterase 3A mutations in ing medical errors and enhancing patient efficiency of healthcare delivery, prevent Haufe S, Kaminski J, Utz W, Haas V, Mähler A, Daniels MA, Birkenfeld AL, Lichtinghagen R,inherited Luft FC, hypertension Schulz-Menger with J, Engeli brachydactyly. S, Jordan Hypertension.J. Differential response 2015;66(4):800-8. of the natriuretic renowned informatics and healthcare re- peptide system to weight loss and exercise in overweight or obese patients. J Hyper- searchsafety. organization, The Regenstrief is dedicated Institute, to the Inc., im a- provement of health through research that Schroeder C, Heusser K, Zoerner AA, Großhennig A, Wenzel D, May M, Sweep FC, tens. 2015;33(7):1458-64. enhances the quality and cost-effectiveness Mehling H, Luft FC, Tank J, Jordan J. Pacemaker current inhibition in experimental human cardiac sympathetic activation: a double-blind, randomized, crossover study. Clin of health care, by conducting health-services research (HSR). HSR works to improve Fischer R, Konkel A, Mehling H, Blossey K, Gapelyuk A, Wessel N, von Schacky C, healthcare delivery and health outcomes DechendPharmacol R, Ther.Muller 2014;95(6):601-7. DN, Rothe M, Luft FC, Weylandt K, Schunck WH. Dietary omega-3 of patients through research on patient safety, quality of care, patient-provider fatty acids modulate the eicosanoid profile in man primarily via the CYP-epoxygenase communication strategies, applied health pathway. J Lipid Res. 2014;55(6):1150-1164. information technologies, system rede- Clinical Research Unit staff sign, and medical education research. This Friedrich C. Luft, MD healthcare information technology can be Michael Boschmann, MD inspected at: www.regenstrief.org. Heidrun Mehling, MD Natalia Rakova, MD, PhD Education Jochen Steiniger, PhD ECRC members have not only access to all Anja Mähler, PhD MDC facilities but also are welcome at all Nadine Krüger - Gabriele Rahn tion to internal laboratory meetings, there Doris Merkert isMDC a monthly seminars ECRC that happenseminar daily. to introduce In addi Anke Strauss members to what other researchers in the Alexander Krannich, PhD (Biostatistical consultant) building are doing. Finally, we have a week- Robert Preissner, MD (Bioinformatics consultant) ly Clinical Journal Club. At this function, Friedrich C. Luft presents every paper pub- Outpatient clinics lished in N Engl J Med and Lancet from the Skeletal muscle disturbances (congenital and acquired) following week. The material is announced Psychiatry (Memory) in a brief summary ahead of time over email Psychiatry (Cognitive assessments) portals of the MDC and ECRC. The presen- Medical Genetics (Multidisciplinary) tation can be accessed as well. The written Neurology (Skeletal muscle disorders) material is in English, while the presentation (medical student and house-staff level) Cardiology (Cardiomyopathy) is held in German. Outside visitors are wel- NephrologyNeurology (Inflammatory (Diabetes registry) CNS disorders) come and the material is available weekly Metabolism (Diabetes and lipids) Pediatric allergy (Atopic dermatitis) weekly at http://www.charite-buch.de/). via Webex or YouTube (load the file down

MDC Research Report 2016 215 ECRC (JOINTLY WITH CHARITÉ) Photo: David Ausserhofer / MDC David Ausserhofer Photo: Photo: David Ausserhofer / MDC David Ausserhofer Photo:

Dominik N. Müller Ralf Dechend

Hypertension-induced Target-Organ Damage

The group’s major research interests are the and M2 macrophage action.1,5 The outcome of renin-angiotensin system2-4 the immune sys- this increased TH17 activation resulted in an 1-5 accelerated and more severe autoimmune dis- tem and how both systems cause hyperten- ease (multiple sclerosis) in mice on a high-salt sion-induced target-organ damage. Recent diet.5 work has extended the concept analyzing how Macrophages are an important innate im- environmental factors such as a high salt affect homeostatic functions beyond host defense. the microbiome, immune cells, target organs, mune cell type, which fulfills a plethora of- and autoimmunity. In a translational approach, ous spectrum of different activation states the Dechend/Müller laboratory focuses primar- Functional diversity is reflected by a continu ily on the placenta, heart, and kidneys. The and M(LPS) also referred to as “M1” and- group also cooperates closely with MDC sci- matoryM(IL-4+IL-13) poles alsoof macrophage referred to asactivation, “M2” may rebe- entists and Charité clinician scientists and has spectively.viewed as Similar the extreme to T cells, pro- physiologically and anti-inflam in- also been a resource for young clinicians and creased environmental sodium concentrations doctoral students beginning their careers in ex- - perimental and clinical cardiovascular research. ryaffected activation pro- ofand macrophages anti-inflammatory with LPS activation was sig- of macrophages differentially. Pro-inflammato The immune system, salt, and hypertension-induced target organ mediatornificantly andboosted effector in the molecule presence expression. of high NaCl damage concentrations and enhanced inflammatory Hypertension induces target-organ dam- activationIn contrast, was blunted for murine in vitro M2 macrophagesand in vivo.1 general, the immune system is traditionally activated with IL-4+IL-13 recently found that consideredage; however, to thebe a mechanisms complex biological are unclear. system In - cantlyFor instance, reduced M(IL-4+IL-13) ability to suppress cells activatedCD4+ T cell in invasion of foreign microorganisms, such as theproliferation presence inof vitroadditional and aNaCl high-salt had adiet signifi de- bacteriathat wards and off viruses. disease, We typically and others by fighting found that the layed cutaneous wound healing. Salt effect on the immune system plays a pivotal role in the tonicity.1 target-organ damage. affectedM(IL-4+IL-13) M2 macrophages activation was via independent an Akt-mTOR of pathogenesisTogether with of Angiotensin Jens Titze (Vanderbilt (Ang) II-induce Uni- metabolic Instead, signaling we pathway found evidence and changes that salt in versity), we address the regulatory interaction - between immune cells, lymph vessels, and in- knowledged that adaptions of cellular meta- terstitial matrix components for maintenance boliccellular programs metabolism are crucial (Figure for 1). proper It is immune now ac of internal environmental composition, blood- pressure regulation, cardiovascular target- simply adopted to meet the energetic needs of organ damage, and immunity. We provided cell activation. Metabolic “rewiring” is not evidence that sodium, a known risk factor for effector functions and fate decisions of adap- cardiovascular disease, interacts with the im- tiveblasting immune lymphocytes; cells. One instead, future research it also controls goal of mune system.1,5 our laboratory is to elucidate the role of salt on induced an increase in the cytokine-driven immuneometabolism and immune cell func- Increased NaCl concentrations tion.

TH17 activation in naïve T cells and alter M1 216 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

Putative mechanisms how sodium affects IL-4 and IL-13-induced M2 macrophage activation.

Pathogenesis of preeclampsia Preeclampsia is a leading cause of maternal death, and perinatal morbidity and mortality resulting from premature delivery and intrauterine growth restriction. New onset of hypertension and proteinuria arises secondary

and immunological dysfunction. We induced Tregto maternal in our established endothelial transgenic and inflammatory rat model for preeclampsia by application of super- - Selected Publications

Binger KJ, Gebhardt M, Heinig M, Rintisch C, Schroeder A, Neuhofer W, Hilgers hypertensionagonistic monoclonal and proteinuria, antibody but for it CD28. substan In- K, Manzel A, Schwartz C, Kleinewietfeld M, Voelkl J, Schatz V, Linker RA, Lang F, tiallyduction improved of Treg fetalhad nooutcome influence by ameliorating on maternal Voehringer D, Wright MD, Hubner N, Dechend R, Jantsch J, Titze J, Muller DN. High intra uterine growth restriction. J Clin Invest. By whole genome expression analysis of Geisbergersalt reduces S, the Maschke activation U, Gebhardt of IL-4 and M, IL-13Kleinewietfeld stimulated M, macrophages. Manzel A, Linker RA, placental tissue using microarrays, we found Chidgey2015;125:4223-38. A, Dechend R, Nguyen G, Daumke O, Muller DN*, Wright MD*, Binger KJ*. CD74 (Cluster of differentiation 74) in pla- New role for the (pro)renin receptor in T-cell development. Blood (*shared authorship) cental macrophages to be one of the most 2015;126:504-7. down-regulated genes in placentas from G, Staff AC, Müller DN, Hünig T, Felderhoff-Müser U, Herse F, LaMarca B, Dechend RPrzybyl. Regulatory L, Ibrahim T cells T, ameliorateHaase N, Golic intrauterine M, Rugor growthJ, Luft FC, retardation Bendix I, Serdarin a transgenic M, Wallukat rat macrophages lacking CD74 showed a shift to- model for preeclampsia. Hypertension preeclamptic women. Naïve and activated- Choi M, Przybyl L, Binger KJ, Haase N, Wilck 2015;65:1298-306. N, Heuser A, Fokuhl V, Ruland J, Lucas PC, duced macrophage adhesion to trophoblasts. MarkóMcAllister-Lucas L, Henke N, LM, Park Luft JK, FC, Spallek Dechend B, Qadri R, Müller F, Balogh DN A, Apel IJ, Oravecz-Wilson KI, CD74-knockoutwards a pro-inflammatory mice showed signature disturbed and pla re- Hypertension Kleinewietfeld M, Manzel A, Titze J, Kvakan H, Linker. Bcl10RA, Müller Mediates DN Angiotensin II- cental morphology, reduced junctional zone Induced Cardiac Damage and Electrical Remodeling. 2014;64:1032-9. and smaller placentas with fetal growth re- Pathogenic Th17 Cells. Nature *, Hafler DA*. striction. These data show that altered mac- Sodium Chloride Drives Experimental Autoimmune Disease by the Induction of rophage-trophoblast interaction is involved Patents 2013;496:518-22 (*shared authorship) in preeclampsia. We found altered genomic imprinting in G, Schmidt C, Fischer R, Muller DN (all MDC), Puli N and Falck JR (UT Southwestern, gene expression in preeclampsia. Statistical 1. Novel Eicosanoid Derivatives. EP-1874, 2009. Inventors: Schunck WH, Wallukat

analyses of dependence between imprint- Dallas); WO 2010/081683 A1 ing to probe distance showed that the dif- 2.Marko Patent L, Bartholomaeusapplication, 2015. H, BaloghDE 102015115192.8 A. ‘Short-chain fatty acids for the ference was driven by altered expression of treatment of cardiovascular disease’ Inventors: Muller DN, Wilck N, Dechend R, the imprinted transcription factor Distal-less Group Leaders PhD/MD Students imprinting as one underlying mechanism for Priv. Doz. Dr. Ralf Dechend Sabrina Geisberger homeobox 5 (DLX5). We identified loss of Prof. Dr. Dominik N. Müller Kristin Kräker into trophoblasts reduced proliferation and Hendrik Bartolomaeus increasedits dysregulation. metabolic Overexpression activity. Transcriptome of DLX-5 Scientists - Priv. Doz. Dr. Florian Herse Technical Assistants pression showed several dysregulated path- (lab manager) Juliane Anders analysis of trophoblasts with DLX5 overex Dr. Nadine Haase Ute Gerhard Julia Rugor stronglyways, which clusters suggests with thatthat DLX5of preeclamptic is involved May-Britt Köhler placentas,in stress and responses. are separated DLX5 transcriptomefrom the tro- Clinician Scientists Ilona Kamer Dr. Michaela Golic Jutta Meisel can model the differential gene expression Dr. Nicola Wilck Gabriele N’diaye seenphoblast in preeclampsia wild-type cells and suggesting might be that a potent DLX5 Dr. Lajos Marko Technical Assistants Trainees transcription factor involved in pathogenesis Dr. Andras Balogh Yves Marcel of the disease. We perform these experiments Dr. Natalia Rakova in close collaboration with Prof. Zsuzsanna Study Nurse Heike Schenck from Oslo. Izsvak (MDC) and Prof. Ann-Cathrine Staff

MDC Research Report 2016 217 ECRC (JOINTLY WITH CHARITÉ) Photo: David Ausserhofer/MDC Photo:

Maik Gollasch

Blood Vessel Function and Target-Organ Damage / Registry Diabetic Nephropathy

Our group focuses on ion channels, primarily in TMEM16a and TRP channels vascular smooth muscle cells (VSMC), to clarify TMEM16a, TRPC6, and TRPV1 channels are expressed in the vasculature. We used mechanisms contributing to hypertension and cardiovascular disease. Potassium channels, chloride mice and found that TMEM16a downregu- channels and transient receptor potential (TRP) latessmooth agonist-induced muscle specific vasoconstrictions TMEM16a deficient and channels have received special attention. In col- thereby contributes to blood pressure regulation. Our current research in this area is di- laboration with Björn Schroeder, we investigated rected towards identifying role of TMEM16a the role of smooth muscle TMEM16a chloride and ryanodine receptor isoforms in collat- channels. We showed that VSMC TMEM16 channels function as regulator of agonist-dependent Thomas Willnow, we have examined the role oferal the arterial choroid networks. plexus in In the collaboration pathogenesis with of arterial constriction and systemic blood pressure. multiple sclerosis, and found that claudin 3 In collaboration with Thomas Jentsch and Mi- (CLDN3) may be regarded as a crucial and chael Bader, we provided definitive evidence that novel determinant of blood-cerebrospinal myogenic responses of mouse mesenteric and that TRPV1 and TRPV4 channels play a role renal arteries rely on ligand-independent, mecha- fluid barrier integrity. We have also found noactivation of AT1R subtype a. We are currently TRPV1 channels can contribute to ischemia/ studying biased agonism and ryanodine receptor in regulating renal blood flow. We found that

isoforms (Figure 1) in VSMC potassium channels reperfusion (I/R) -induced kidney injury. Eicosanoids, H2S and inflammation gating and myogenic tone. A particular attention is EETs serve as endothelial-derived hyperpo- paid to various KCNQ potassium channel isoforms larizing factors (EDHF), but may also affect in periadventitial vasoregulation. Thomas Jentsch mechanisms. Our current research in this is helping us here. We also have a project focus- areacardiovascular is directed function towards by identifying anti-inflammatory the role ing on the perivascular adipose tissue (PVAT) as a of H2S producing enzymes as regulator of va- source for relaxing factors. Here, we collaborate - laboration with Michael Bader, we found that with Wolf-Hagen Schunck, who has peaked our angiotensin-convertingsodilatory EETs and nitric enzyme oxide (NO). 2 (ACE2) In col interests in eicosanoids. We also collaborate with regulates vascular function by modulating Huang Yu, Hong Kong, China. Together, we developed the novel concept that perivascular adipose diabetes mellitus, insulin-induced relaxation ofnitric arteries oxide is releaseimpaired and and oxidative the level stress.of ortho- In tissue (PVAT) function requires considering het- tyrosine (o-Tyr), an oxidized amino acid is erogeneous PVAT as a specialized organ that can increased. We found that elevated levels of differentially regulate vascular function depending o-Tyr contribute to vasomotor dysfunction in on its anatomical location. Finally, in collabora- diabetes mellitus. By an improved tag-switch tion with the German Institute of Human Nutrition (DifE), we are studying human diabetic nephropa- rolemethod, of Anti-AT1 we identified receptor thioredoxin and –ETA toreceptor act as thy with a focus on genetics. antibodiesdepersulfidase. in pulmonary We also identifiedarterial hyperten a novel- sion associated with systemic sclerosis. Both

218 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

Intracellular calcium release is visualized by calcium sensitive dyes and Nipkow spinning disc microscopy. Three freshly-isolated cells are shown in this figure: The upper two cells are isolated from wild-type (WT) mice; the lower cell is isolated from a ryanodine receptor type 2 deficient (KO) mouse. The stimulator of intracellular ryanodine receptors caffeine (10 mM) induced a strong calcium release signal in WT, but not in KO cells.

antibodies may contribute to pulmonary hy- www.charite-buch.de/rdn/). We are continu- pertension via increased vascular endothelial ously recruiting patients for our Registry and reactivity and induction of pulmonary vascu- - ies. Through these studies, we hope to iden- tifyrenal/vascular novel mechanisms disease-specific leading to family increased stud hypertensionlopathy. In collaboration and target-organ with Ralfdamage. Dechend, Our cardiovascular risk and target-organ damage datawe also suggest identified that evena novel short-term role of vitamin severe D vi in- and novel treatment targets. - pertension and impacts on renin-angiotensin Primary funding systemtamin D components deficiency may that directly could contribute promote hy to The Deutsche Forschungsgemeinschaft (GO 766/13-1, GO766/15-1; target-organ damage. GO766/17-1; GO766/18-1, FOR1368), DAAD, Alexander von Hum- boldt Stiftung, ERA-EDTA and cooperative grants between the MDC Vasodilator signals from and ECRC have funded this work. perivascular adipose tissue Selected Publications produced in the perivascular adipose tissue Tsvetkov D, Tano JY, Kassmann M, Wang N, Schubert R, Gollasch M. The role of (ADRF).We have Our identified recent work a vasorelaxing showed that KCNQ factor channels could represent the subtype of K v DPO-1 and XE991-sensitive potassium channels in perivascular adipose tissue- mediatedPMC4973012. regulation of vascular tone. Front Physiol. 2016 Aug 4;7:335. doi: H S, could represent ADRF. However, other 2 Riehle10.3389/fphys.2016.00335. M, Büscher AK, Gohlke PubMed BO, Kaßmann PMID: 27540364;M, Kolatsi-Joannou PubMed M, Central Bräsen PMCID: JH, Nagel channels involved. The “third gas”, namely M, Becker JU, Winyard P, Hoyer PF, Preissner R, Krautwurst D, Gollasch M, Weber S, alterations in the paracrine control of arterial Harteneck C. TRPC6 G757D Loss-of-function mutation associates with FSGS. J Am toneadipokines by periadventitial may also play adipose a role. Wetissue identified in animal models of hypertension and metabolic ChenSoc Nephrol. L, Kaßmann 2016 M, Sep;27(9):2771-83. Sendeski M, Tsvetkov doi: D,10.1681/ASN.2015030318. Marko L, Michalick L, Riehle PubMed M, disease. KCNQ and cystathionine gamma-ly- LiedtkePMID: 26892346; WB, Kuebler PubMed WM, Harteneck Central PMCID: C, Tepel PMC5004639. M, Patzak A, Gollasch M. Functional transient receptor potential vanilloid 1 and transient receptor potential vanilloid 4 channels along different segments of the renal vasculature. Acta Physiol (Oxf). 2015 the role of Kv channels and H2S. ADRF and its aseputative deficient targets mice (KCNQ are available channels) to us might to clarify rep- - resent exciting new targets for the develop- reichFeb;213(2):481-91. M, Pathan AR, Anistandoi: 10.1111/apha.12355. YM, Alenina N, Rusch PubMed NJ, Bader PMID: M, 25069877.Jentsch TJ, Gollasch ment of drugs for treatment of cardiovascu- Schleifenbaum J, Kassmann M, Szijártó IA, Hercule HC, Tano JY, Weinert S, Heiden lar and metabolic disorders. Overall, our and M. Stretch-activation of angiotensin II type 1a receptors contributes to the myogenic other data indicate that dysfunctional peri- response of mouse mesenteric and renal arteries. Circ Res. 2014 Jul 7;115(2):263-- vascular adipose tissue (PVAT) contributes to enbaum72. doi: 10.1161/CIRCRESAHA.115.302882. J, Vitzthum H, Gollasch M, Ehmke H, PubMedSchroeder PMID: BC, Hübner 24838176. CA. Disruption of Heinze C, Seniuk A, Sokolov MV, Huebner AK, Klementowicz AE, Szijártó IA, Schleif cardiovascular risk. vascular Ca2+-activated chloride currents lowers blood pressure. J Clin Invest. 2014 Genetic renal diseases Feb;124(2):675-86. doi: 10.1172/JCI70025. PubMed PMID: 24401273; PubMed Central PMCID: PMC3904609. - Group Leader Dmitry Tsvetkov (ERA-EDTA and sponsibilities has been a focus on clinical Univ.-Prof. Dr. med. Dr. rer. nat. DAAD fellow) geneticsAn outgrowth related of to Maik renal Gollasch’s diseases. clinical We have re Maik Gollasch Ning Wang performed functional analyses of mutations Graduate students and Associated Scientists emphasis on TRPC6 channels in focal and clinical fellows Dr. Mario Kaßmann segmentalcausing familiar glomerulosclerosis kidney diseases with(FSGS). specific We Gang Fan Dr. Johanna Schleifenbaum - Chen Lan Dr. Yulia Komleva (DAAD fellow) man family, supporting the overall concept Dr. Marwan Mannaa Dr. Jean-Yves Tano (Alexander von thatidentified CD2AP-associated a unique CD2AP nephropathy mutation inis ana Ger au- Sophie Nickel Humboldt fellow) tosomal dominant form of FSGS in man. We Weiying Kong are also studying human diabetic nephropa- Artur Kuczmanski Technicians thy with a focus on genetics. For these pur- Yoland-Marie Anistan poses, we recently established an Outpatient Kornelia Buttke Kidney Clinic at Charité Campus Buch and the Xiaoming Lian Nadine Wittstruck Registry of Diabetic Nephropathy (http:// Dr. Istvan A. Szijarto

MDC Research Report 2016 219 ECRC (JOINTLY WITH CHARITÉ) Photo: David Ausserhofer/MDC Photo:

Ralph Kettritz

Nephrology and Inflammatory Vascular Diseases

Our group is interested in vascular biology tential for collateral tissue damage. We stud- with special focus on autoimmune vasculitis. relevance to ANCA vasculitis: 1) NSPs as tis- Patients develop autoantibodies against neu- sue-damagingied NSPs in two enzymes inflammatory that are settings transferred with trophil and monocyte constituents that reside from the neutrophil to endothelial cells (EC), in the cells’ cytoplasm and are called anti- neutrophil cytoplasmic autoantibodies (ANCA). - cytes.and 2) NSPs as regulator of proinflammatory The major ANCA target antigens are myeloper- IL-1βWe generation observed that in neutrophilsall three NSPs and are monotrans- oxidase (MPO) and proteinase 3 (PR3). What ferred to EC where they accumulate and re- breaks tolerance is not clear, but once ANCA are generated they activate neutrophils and usedtain their terminal-amine enzymatic activity.isotopic In labeling collaboration of sub- monocytes. These activated cells adhere to with Gunnar Dittmar’s group at the MDC, we- the endothelium and cause severe necrotizing thelial NSP substrates. We characterized the vasculitis. This highly inflammatory process can extendedstrates (TAILS) cleavage to identify pattern several and generated novel endo cor- affect small vessels of every organ; however, on the gene ontology analysis we showed that the kidneys and lungs are the most life-threat- NSPsresponding released IceLogos from ANCA-activatedfor each protease. neutro Based- ening manifestations. Patients develop rapidly phils entered the EC, disturbed the cytoskel- progressing necrotizing crescentic glomeru- - thelial permeability increase with increased lonephritis (NCGN) and often require dialysis. etal architecture (figure), and caused endo- Timely immunosuppressive treatment improves gated these effects. renal function. However, treatment, including albumin flux. Serine protease inhibitors abro steroids, cytotoxic drugs and more recently ANCA-stimulatedOur second projectmonocytes identified and neutrophils NSPs as CD20-targeted B-cell biological, is not disease inimportant vitro and regulators in an ANCA of mouse IL-1β generationmodel. Active in specific. Translational research is needed to - develop more specific and less toxic treatment ternatives to the classical caspase-1 depen- NSPs, and particularly PR3, were efficient al protocols. Oxidative and non-oxidative effector mechanisms participate in ANCA vasculitis and modeldent inflammasome we showed thatand promotedinactive or pro-IL1b lacking provide potential novel treatment targets. NSPsprocessing protected into IL-1β.against Using NCGN an and ANCA this mouse effect

Neutrophil serine proteases (NSPs) was accompanied by a significantly reduced accelerate vascular inflammation protectedIL-1β generation. from the Treatment disease ofsuggesting mice with that the Neutrophil granules contain large amounts of specific IL-1 receptor antagonist, anakinra, NSPs that include human neutrophil elastase, a treatment target in this model. PR3, and cathepsin G. Once the neutrophil is IL-1β is a critical inflammatory mediator and activated (e.g. by ANCA), these NSPs are re- Phagocyte NADPH oxidase (Phox) leased into the environment where they act restrains the inflammasome as powerful proteolytic enzymes. NSPs shape Several clinical antioxidant trials in a variety - of diseases have failed indicating that reac-

the inflammatory response and have high po 220 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

IceLogo for proteinase 3 (PR3) depicting amino acids that were most likely (above the x-axis) or least likely (below the x-axis) present in the substrate at the indicated P or P’ site. Active neutrophil serine proteases, including PR3 are released from ANCA-stimulated neutrophils, accumulate in EC, and cause actin cytoskeleton disruption and permeability increase. Serine protease inhibition by PMSF abrogates this effect. Phalloidin (green) and DAPI (blue) staining.

tive oxygen species (ROS) have more complex CD177/mPR3 phenotype. We have strong functions. We tested the hypothesis that ROS evidence that DNA methylation plays an important role. down-regulating caspase-1, thereby reducing produced by Phox restrain inflammation by group to elucidate the crystal structure of CD177We alsoand teamed the CD177:PR3 up with Oliver complex. Daumke’s We IL-1β generation and limiting ANCA-induced found previously that, on CD177pos/mPR3high NCGN. In an MPO-ANCA disease model, mice neutrophils, PR3 and CD177 form of a larger transplanted with Phox-deficient BM, showed signaling complex within lipid rafts. Recruit- increased renal IL-1β levels and accelerated- ment of the transmembrane b2-integrin disease. Therapeutic IL-1β receptor blockade- - cientabrogated monocytes aggravated and neutrophils NCGN. IL-1β challenged genera cellular signaling pathways and neutrophil tion was strongly accelerated in Phox-defi activationCD11b/CD18 by thenPR3-ANCA. allows initiationWe believe of intra that generation was abrogated when caspase-1 detailed structural information will help to waswith blocked. MPO-ANCA The andcausative this acceleratedrole of caspase-1 IL-1β therapeutically interfere with PR3 binding - to CD177, or, alternatively, with anti-PR3 91phox antibody-binding to CD177-presented PR3. rescued.was firmly Our established study indicated by showing that Phox-gen that gp- Either approach would have the potential to erated/caspase-1 ROS down-regulate double-deficient caspase-1, mice keepwere- prevent neutrophil activation by PR3-ANCA.

Selected Publications receptoring the inflammasome blockade provided in check also inand this thereby study Jerke U, Hernandez DP, Beaudette P, Korkmaz B, Dittmar G, Kettritz R: Neutrophil protectionlimiting ANCA-induced from the disease inflammation. and is a promis IL-1-

serine proteases exert proteolytic activity on endothelial cells. Kidney Int 2015, - 88:764-75. ing treatment strategy. Importantly, our data Schreiber A, Luft FC, Kettritz R: Phagocyte NADPH Oxidase Restrains the Inflamma questionGenetic the regulation benefit of antioxidants. of the PR3- some in ANCA-Induced GN. J Am Soc Nephrol 2015;26:411-24. Schreiber A, Eulenberg-Gustavus C, Bergmann A, Jerke U, Kettritz R. Lessons from a presenting neutrophil surface double-transgenicKettritz R. Neutral serineneutrophil protease approach of neutrophils. to induce antiproteinaseImmunol Rev. 2016, 3 antibody-mediated 273(1):232-48. receptor CD177 and disruption of vasculitis in mice. J Leukoc Biol. 2016 100(6):1443-1452. the CD177/PR3 complex on neu- Schreiber A, Pham CT, Hu Y, Schneider W, Luft FC, Kettritz R: Neutrophil serine prote- trophils as a treatment strategy - PR3 is the major autoantigen in European ases promote IL-1beta generation and injury in necrotizing crescentic glomerulone - phritis. J Am Soc Nephrhol 2012;23:470-82. Group Leader membranepatients with receptor. vasculitis. CD177 We is identified restricted neu to Prof. Dr. Ralph Kettritz atrophil-specific variable, but CD177in a given as aindividual PR3-presenting stable neutrophil subset. Only this CD177-express- Graduate students and clinical Associated scientists ing neutrophil subset presents high PR3 fellows Prof. Friedrich C. Luft amounts on the neutrophil surface. The per- PD Dr. Adrian Schreiber centage of CD177pos/mPR3high neutrophils has PD Dr. Mira Choi Technical Assistants clinical implications for ANCA vasculitis risk Dr. Uwe Jerke Susanne Rolle and clinical outcomes. We are working on Dr. Claudia Eulenberg-Gustavus Sylvia Krüger the genetic and epigenetic regulation of this

MDC Research Report 2016 221 ECRC (JOINTLY WITH CHARITÉ) Photo: David Ausserhofer / MDC David Ausserhofer Photo: Photo: David Ausserhofer / MDC David Ausserhofer Photo:

Stephan Mathas Martin Janz

Biology of Malignant Lymphomas

A detailed molecular understanding of the as well as growth and survival pathways. oncogenic process is fundamental for the - stand the hierarchical order and coopera- development of new therapeutic strategies tiveIn this activity part ofof our such work, TF weactivities aim to inunder HRS for human lymphomas. We aim to gain insight cells using a combinatorial approach that into the network of oncogenic defects in B integrates the analysis of accessible chro- and T lymphoid cells by using a wide range of matin studies with global gene expression and epigenetic data. This search revealed cell and molecular biology techniques com- enrichment of DNA binding motifs for the bined with high-throughput and screening ap- inducible transcription factors AP-1, NF-kB, proaches as well as in vivo mouse models. Our - special interest is focused on the identification scape of cHL. Moreover, these motifs were STAT and IRFs in the active chromatin land- and characterization of deregulated signaling ing a chronic activity of the corresponding events and aberrant transcription factor activi- specifically occupied in HRS cells, indicat ties in lymphoid malignancies. Using classical infactors. cooperation Apart fromwith the confirming group of theC. Scheide role of- Hodgkin lymphoma (cHL) and anaplastic large reit,AP-1, MDC) NF-kB and (both STAT defined factors in for HL cHL by ourpatho lab- cell lymphoma (ALCL) as model systems, we - analyze the role of lineage infidelity, cellular reprogramming and activation of lineage-inap- genesis at a genome-wide level, a key find ing of this analysis was the identification of propriate survival signals in lymphomagenesis. defectoverexpressed of HRS cells. IRF5, Our a central work provides regulator evi of- In multiple myeloma we study the impact of inflammatory gene expression, as a specific tumor – microenvironment interactions as well - as of intrinsic genetic and functional defects on siondence program, that IRF5 which in combination offers an explanationwith NF-kB orchestrates the HL-specific gene expres the malignant transformation of terminally dif- cHL. We are currently transferring the ap- ferentiated B cells by employing in vitro and in proachfor the describedunique inflammatory herein to other phenotype lymphoid of vivo strategies. Collectively, our work does not malignancies to identify key transcriptional regulators of the respective lymphoma en- only address fundamental aspects of lympho- tities. ma pathogenesis, but also aims at providing Apart from the direct modulation of cellu- therapeutic solutions. lar processes like growth or apoptosis, even single TFs can enforce malignant transfor- Deregulated transcription factor mation by disrupting the physiologic differ- activities as a driving force of ma- entiation program. A number of lymphoid lignant transformation and lineage tumors display a phenotype that is in ac- infidelity in human lymphomas cordance with such a reprogramming pro- Aberrant transcription factor (TF) activities cess, including cHL, PEL and ALCL. Among are key regulators of malignant transforma- these entities, cHL constitutes the most

cHL provides a unique example among lym- phoidtion of malignancies hematopoietic given cells. the In thisexceptional regard, haveprominent shown examplethat the activity for lineage of the infidelity TF E2A diversity and strength of deregulated TFs isand disrupted reprogramming. in the malignant In previous cells work, of cHL we in the malignant Hodgkin-/Reed-Sternberg and ALCL. This defect is directly linked to a (HRS) cells that target cellular differentiation reprogramming process that results in the

222 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

up-regulation of genes usually suppressed with PDLIM2 promoter methylation, al- in lymphoid cells, including the myeloid tered splicing and genomic alterations, and CSF1 receptor (CSF1R) gene. Remarkably, contributes to the activation of NF-kB and CSF1R expression in HL cells is not medi- AP-1 in these lymphoma cells. Thus, inac- ated by the canonical myeloid promoter, but is activated by an upstream long terminal repeat (LTR) element, providing the contributingtivation of PDLIM2 to lymphoma promotes pathogenesis. activation of inflammatory signaling pathways, thereby

first example of such an alternative proto- ofoncogene new treatment activation. strategies. In addition, For ourexample, work inhibitionhas significant of CSF1Rimpact activityon the developmentresulted in reduced proliferation or cell death of lymphoma cells. Ongoing projects include work on the molecular mechanisms of the reprogramming processes in lymphoid cells, their therapeutic implications and, in cooperation with the group of K. Rajewsky (MDC), the generation of transgenic mouse models to study the impact of lymphoma- associated factors on cell growth, apoptosis and differentiation in vivo.

Identification and characterization of molecular defects in B and T cell-derived malignancies Figure 1. Key signaling pathways in cHL. Major signaling pathways Apart from our work on deregulated TF in HRS cells are canonical and non-canonical NF-kB (defined in cHL by - groups Dörken and Scheidereit), and JAK/STAT signaling. Further key cation of additional genetic and functional players are IRF5, which together with NF-kB coordinates HL-specific networks, we are interested in the identifi- gene expression (defined in cHL by group Dörken), and high-level ac- ticular, we focus on the pathogenesis of B tivation of AP-1/CREB members (JUN, JUNB, ATF3) (defined in cHL by cell-deriveddefects of B malignanciesand T cell lymphomas. such as cHL In parand groups Dörken and Scheidereit). Green arrows indicate interactions. multiple myeloma as well as T cell-derived malignancies (ALCL and cutaneous lym- Selected Publications investigate the mechanisms leading to the phoproliferative disorders). It is our aim to- Janz, M, Rodig, S, Rajewsky, K. (2012). The cell cycle regulator c-Myc is essential for the ate the contribution of breakpoint-proxi- Calado, DP, Sasaki, Y, Godinho, SA, Pellerin, A, Köchert, K, Sleckman, BP, de Alborán, IM, Kreher*, S, Bouhlel*, MA, Cauchy, P, Lamprecht, B, Li, S, Grau, M, Hummel, F, Köchert, malt(2;5)-translocation gene deregulation in ALCLto ALCL and pathogen to evalu- formation and maintenance of germinal centers. Nat. Immunol. 13, 1092-1100. esis and for prognosis and development of Dörken, B, Bonifer, C, Mathas, S. (2014). Mapping of transcription factor motifs in new treatment strategies for ALCL. Scheidereit, C, Giefing, M, Siebert, R, Rajewsky, K, Lenz, G, Cockerill, PN, Janz M, Furthermore, to identify factors that are Natl. Acad. Sci. USA 111, E4513-4522. *) equal contribution. Ullrich,active chromatin K, Blumenthal-Barby, identies IRF5 F, asLamprecht, key regulator B, Köchert, in classical K, Lenze, Hodgkin D, Hummel, lymphoma. M, Proc. tumor microenvironment, we screened involved in the formation of the HL-specific Mathas, S, Dörken, B, Janz, M. (2015). The IL-15 cytokine system provides growth and Hodgkin cell lines with a focus on cytokines Behringer,survival signals K, Goergen, in Hodgkin H, Hitz, lymphoma F, Zijlstra, and JM, enhances Greil, R, Markova,the inflammatory J, Sasse, S,phenotype Fuchs, M, of Topp,HRS cells. MS, Soekler,Leukemia M, 29, Mathas, 1213-1218. S, Meissner, J, ... Engert, A. (2015). Omission of dacarba- andgene cytokine expression receptors. profiles With of HRS this and approach, B non- zine or bleomycin, or both, from the ABVD regimen in treatment of early-stage favour-

the corresponding receptor components in able Hodgkin’s lymphoma (GHSG HD13): an open-label, randomised, non-inferiority Hodgkinwe identified cell anlines. up-regulation We demonstrated of IL-15 thatand Köchert,trial. Lancet K, Krappmann, 385, 1418-1427. D, Klapper, W, Hummel, M, Wenzel, SS, Lenz, G, Janz, M, Dörken, - Wurster, KD, Hummel, F, Richter, J, Giefing, M, Hartmann, S, Hansmann, ML, Kreher, S, sistance, stimulates JAK/STAT signaling in classical Hodgkin and anaplastic large cell lymphoma. Leukemia 31, 602-613. andIL-15 is involved promotes in growth the regulation and apoptosis of HRS cell- re B, Siebert, R, Mathas, S. (2017). Inactivation of the putative ubiquitin-E3 ligase PDLIM2

Group Leaders Undergraduate Students ofspecific the complex gene expression. interactions In between summary, tumor this Prof. Dr. Stephan Mathas Nikolai Schleußner cellsstudy and identified the microenvironment IL-15 as an important in HL. part PD Dr. Martin Janz Henrike Sczakiel Finally, we postulated that a defect of the cellular degradation machinery might be Technical Assistants involved in the unusually strong activa- Graduate students Franziska Hummel tion of transcription factors, including NF- Mariantonia Costanza Simone Lusatis Shuang Li* Ute Nitschke Katrin Ullrich* Brigitte Wollert-Wulf recurrentkB, AP-1, inactivation STATs and IRFs,of the in putative cHL. By ubiq the- Wiebke Winkler analysis of microarray data, we identified a Kathrin Wurster* * part of the period reported

uitin E3-ligase PDLIM2 in cHL and ALCL. Loss of PDLIM2 expression was associated MDC Research Report 2016 223 ECRC (JOINTLY WITH CHARITÉ) Photo: private Photo:

Friedemann Paul

Outpatient Clinic for Neuroimmunology

Friedemann Paul is head of the Out-patient We could show that despite differences in Clinic of Neuroimmunology at the ECRC caring disease course and clinical characteristics primary progressive and relapsing remit- for more than 1500 patients each year suffering ting MS exhibit identical lesion morphology from neuroimmunological disease like Multiple Sclerosis (MS), Neuromyelitis optica (NMO) Susac syndrome or Myasthenia gravis. In coin ultra-high field (7 Tesla) MRI. operation with the NeuroCure Clinical Research Center his research is dedicated to translational medicine running numerous trials to develop new diagnostic and therapeutic approaches in neuroimmunological disorders. The main focus is on developing so-called neuroprotective (nerve cell protecting) treatment and establishing modern examination procedures such as MRI (magnetic resonance imaging) and OCT (optical coherence tomography). He works closely with basic scientists, clinical neurolo- gists and neuro-imaging researchers on local, national and international level.

Multiple Sclerosis Multiple sclerosis (MS) is the most common - tral nervous system that primarily affects chronic inflammatory disease of the cen

systemyoung adults.falsely Itrecognizes is caused tissue by autoimmune of its own processes, meaning the body’s immune

tonervous nerve cellssystem and as axons “foreign” with andtheir causes envelop an- inginflammatory protective reaction,layer. A variety resulting of neurologiin damage- cal symptoms may occur: paralysis, visual disturbances, abnormal sensations or balance problems. 7 Tesla MRI images of brain structures. A 7T T1w MPRAGE The neuroimmunology team focuses on shows high-resolution anatomical images. B 7T SWI developing and establishing new disease depicts tiny brain veins. C,D 7T T2*w FLASH (resolution course and prognostic markers for MS. 0.2mm x 0.2mm x 2mm) depict highly myelinated struc- This includes, for example, improving pos- tures like the optic radiation (white arrows) or the stripe of Gennari (black arrows). Tiny veins emerge in the periven- tricular white matter (black arrowheads). Image quality of establishingsibilities in the new field imaging of magnetic techniques resonance such 7T gradient echo images may be reduced due to inhomoge- asimaging optical (MRI) coherence and ultra-high tomography field MRI,(OCT). or neities or artifacts (asterisks).

224 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

at present widely used immunosuppres- sant therapy upcoming therapeutic strate- - cal interventions in the immune system. giesThe focus neuroimmunology on highly specific grouppharmacologi is considered one of the leading centres in NMO research both nationally and internationally. New diagnostic approaches are devel-

therapeutic studies conducted. We were ableoped, to e.g., introduce with OCT a paradigm or MRI to imaging differenti and- ate between MS and NMO cerebral lesions Neuromyelitis optica (NMO) versus Multiple Sclerosis (MS) lesion morphology. 7T T2*w that NMO lesions compared to MS lesions FLASH images from exemplary NMO (A) and lackin ultra-high a central vein field and MRI rarely based show on thea T2*w- fact MS (B) patients. A small central vein shows hypointense rim. Our growing cohort study within the inner third of most MS lesions of NMO patients offers a wide opportunity (white arrows). An acute MS lesion is charac- to investigate clinical as well as pathopyh- terised by a hypointense rim (white arrow- siolocial aspects of the disease. heads) and a surrounding edema (asterisks). A central vessel is not visible in NMO lesions Primary funding sources (black arrowheads) despite using high spatial German Research Foundation (DFG Exc 257 and other grants) and resolution of 0.2mm x 0.2mm x 2mm. the German Ministry of Education and Research (Competence Net- work Multiple Sclerosis KKNMS and other grants), research grant from the Guthy Jackson Charitable Foundation/National Multiple Sclerosis Society of the USA, European Union Seventh Framework Programme (FP7/2007-2013), EU funding program ERACOSYS- MED, Walter und Ilse Rose Stiftung, Leifheit-Stiftung, Sanitätsrat Dr. Arthur Arnstein Stiftung, Federal Ministry for Economic Affairs and Technology (ZIM funding line)

Because there is as yet no cure for MS, treatment strategies centre primarily on Selected Publications improving symptoms and modifying the Weygandt M, Meyer-Arndt L, Behrens JR, Wakonig K, Bellmann-Strobl J, Ritter K, - Scheel M, Brandt AU, Labadie C, Hetzer S, Gold SM, Paul F, Haynes JD. Proc Natl velops and conducts clinical trials with the aimdisease of improving course. In treatmentthis context options the team in deMS process] Acad Sci U S A. 2016 Nov 22;113(47):13444-13449. PMID: 27821732 [PubMed – in and to learn more about the mechanism of A, Cross AH, Morgan TE, Wei M, Paul F, Bock M, Longo VD. Cell Rep. 2016 Jun action of these therapies next to new patho- Choi IY, Piccio L, Childress P, Bollman B, Ghosh A, Brandhorst S, Suarez J, Michalsen physiological aspects of the disease itself. in process] Free PMC Article Finke7;15(10):2136-46. C, Prüss H, Heine doi: 10.1016/j.celrep.2016.05.009.J, Reuter S, Kopp UA, Wegner F, PMID:Then Bergh 27239035 F, Koch [PubMed S, Jansen – Neuromyelitis optica O, Münte T, Deuschl G, Ruprecht K, Stöcker W, Wandinger KP, Paul F, Bartsch T. JAMA Neuromyelitis optica (NMO) has long been Pache F, Zimmermann H, Mikolajczak J, Schumacher S, Lacheta A, Oertel FC, Bell- regarded as a variant of MS. Today we know mann-StroblNeurol. 2017 J, Jan Jarius 1;74(1):50-59. S, Wildemann doi: B, 10.1001/jamaneurol.2016.4226. Reindl M, Waldman A, Soelberg K, Asgari N, that NMO is an independent disease, which - is based on different disease mechanisms. Ringelstein M, Aktas O, Gross N, Buttmann M, Ach T, Ruprecht K, Paul F, Brandt AU; Typical manifestations of NMO are, on Finkein cooperation C, Kopp UA, with Pajkert the Neuromyelitis A, Behrens JR, Optica Leypoldt Study F, WuerfelGroup (NEMOS). JT, Ploner J CJ,Neuroinflam Prüss one the hand, unilateral or bilateral op- mation. 2016 Nov 1;13(1):282. - tic neuritis, often leading to severe vision H, Paul F. Biol Psychiatry. 2016 May 1;79(9):727-34. doi: 10.1016/j.bio problems and, on the other hand, long- psych.2015.02.024. PMID:25866294. Biol Psychiatry. 2016 May 1;79(9):727-34. doi: 10.1016/j.biopsych.2015.02.024. PMID:25866294 which can lead to severe paralysis, abnor- Group Leader Dr. med. Ludwig Rasche malsegmented sensations, inflammation gait disorders, in the and spinal bladder cord, Prof. Dr. med. Friedemann Paul PD Dr. med. Michael Scheel dysfunction. Dr. med. Felix Schmidt The disease is caused by an autoim- Scientists Sophie Schumacher (M.D. student) mune response to a cellular water channel Dr. med. Susanna Asseyer known as aquaporin-4. Antibodies against Dr. med. Judith Bellmann-Strobl Study assistants this water channel can be detected in most Dr. med. Alexander Brandt Gritt Stoffels patients with NMO. As with MS, there is cur- Dr. med. Joseph Kuchling Bettina Zwingenberger rently no cure for NMO. As successor of the

MDC Research Report 2016 225 ECRC (JOINTLY WITH CHARITÉ) Photo: Bierderlack/Charité Photo:

Jeanette Schulz-Menger

Cardiac MRI

The CMR group at the Experimental Clinical (Univ. of Virginia, USA), Tobias Schaeffter Research Center (ECRC) and Clinics for Car- (PTB, Berlin, Germany)

diology and Nephrology of the HELIOS Clinics We have installed a post-processing lab (CMR Berlin Buch has focused research on in vivo reading and development) giving us the op- assessment of functional and structural myo- portunity to speed up processes. During the cardial abnormalities related to non-ischemic last year we developed standardized procedures (SOP) for quantitative analysis to im- diseases and ischemic heart disease. Our prove and introduced a systematic quality main hypothesis is, that the in-vivo differentia- check of all data. That goes along with the tion of myocardial structural injury is possible in preserved ejection fraction and will help to 2011 we founded an academic outpatient departmentintroduction of of cardiology staff scientist focused structure. on myo In- explain remodeling processes. Furthermore, cardial disease at the ECRC and established we understand cardiovascular disease as a research database dedicated on patient systemic disorders and aim to understand the oriented research. During the last year a fruitful cooperation between the Muscular mechanism and the interaction of vascular group (headed by Prof Spuler, ECRC) was es- and cardiac disorders. In many projects, we tablished. That drives the understanding of are treading the path from the phantom/model including development of techniques into The group has several outreach activities. Jea- “neuro-cardiology” in muscular dystrophy.- clinical research. Developing and establishing innovative imaging biomarker will allow for HCMRnette Schulz-Menger trial (risk assessment is e.g. thefor patients Principal with In quantification of structural integrity and in-vivo HCM)vestigators and is of co-heading Germany of the the Focus NIH sponsored Group for evaluation of prognostic relevant processes. - tional Cohort. Furthermore, she was elected asCardiovascular the president Imagingof Society of for the Cardiovascular German Na As Cardiovascular Magnetic Resonance Magnetic Resonance. Florian von Knobels- (CMR) has the unique possibility to differ- dorff and she were members of the writing entiate myocardial injury, we are focusing group in societal position statements.

scanner with the advantage to reach out to Myocardial tissue injury on clinical research using a 1.5 clinical MRI- and integrity was able to expand the research activities by The assessment of myocardial tissue changes developinghigher field translationalstrengths at the research MDC. The tools group ap- during follow-up is an ongoing challenging plying advanced imaging modalities in close task in cardiovascular research and clinical collaboration with the Berlin Ultra-High-Field cardiology and has been our main focus since Facility (B.U.F.F.). This could be established years. The in vivo differentiation of myocardi- fast, because the head of the Experimental MR-Group (Prof Niendorf) has a strong inter- (patho-)physiological mechanism is the driv- est and experience in CMR-research. A strong ingal structure, force of our cell research. integrity CMR and identificationhas the capabil of- collaboration with scientists from all over the ity to differentiate between the various forms world allows us to proof and use dedicated of myocardial injuries (e.g. edema, hyperemia techniques for myocardial differentiation. Examples for cooperation are Peter Kell- fraction. During the last years we have pre- paredand fibrosis) the ground already to the in preservednext step towards ejection

man (NIH, Bethesda, USA), Frederick Epstein 226 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

Quantitative assessment of focal fibrosis in hypertrophic cardiomyopathy.

- markers for a more precise and personalized ing assessment of extracellular volume. The therapy guiding in this patient cohort underlyingquantification approach of structural is parametric changes mapping. includ Cardiac MR at Ultra Highfield The close collaboration with B.U.F.F. at MDC areAs a currently first step working we published on the recently assessment (2014) of allowed to publish several joint paper in- non-ischemicnormal values heartat different diseases. field Kind strengths and dos and- cardiac function of the right ventricle in vol- factor for quantitative approaches. We have cluding the worldwide first paper assessing age of contrast-media is another influencing 2015 in a prospective randomized trial. The patientsunteers atat 7.07.0Tesla Tesla was in 2013.accepted In Decemberfor publi- understandingfaced its potential of cardiac influence function as published is usually in cation.2015 the We worldwidewere able to first identify paper subtle in cardiac myo- focused on the left ventricle. We could show, cardial changes . that in case of thoracic deformation the right ventricle is predominantly altered and may Selected Publications

Schmacht L, Traber J, Grieben U, Utz W, Dieringer MA, Kellman P, Blaszczyk E, von growing interest in the understanding dia- stolicprofit dysfunction from surgical as a correction. prognostic There factor. isWe a Myotonic Dystrophy Type 2 Patients With Preserved Ejection Fraction: Detection by could show that the analysis of myocardial Knobelsdorff-Brenkenhoff F, Spuler S, Schulz-Menger J. Cardiac Involvement in von Knobelsdorff-Brenkenhoff F, Karunaharamoorthy A, Trauzeddel RF, Barker AJ, deformation reveals differences between dif- BlaszczykCardiovascular E, Markl Magnetic M, Schulz-Menger Resonance. CircJ. Evaluation Cardiovasc of AorticImaging. Blood 2016 Flow Jul;9(7). and Wall ferent types of left ventricular hypertrophy and volunteers. These preliminary results at rest and stress are another connection to our ProthmannShear Stress M, in von Aortic Knobelsdorff-Brenkenhoff Stenosis and Its Association F, Töpper With A,Left Dieringer Ventricular MA, Remodeling. Shahid E, GraesslCirc Cardiovasc A, Rieger Imaging. J, Lysiak 2016 D, Thalhammer Mar;9(3):e004038. C, Huelnhagen T, Kellman P, Niendorf T, interest in the assessment of the aorta. Schulz-Menger J.High Spatial Resolution Cardiovascular Magnetic Resonance at 7.0 Tesla in Patients with Hypertrophic Cardiomyopathy – First Experiences: Lesson Vascular MR As mentioned, we are understanding cardio- Traber J, Wurche L, Dieringer MA, Utz W, von Knobelsdorff-Brenkenhoff F, Greiser A,Learned Jin N, Schulz-Menger from 7.0 Tesla. J.PLoS Real-time One. 2016 phase Feb contrast 10;11(2) magnetic resonance imaging for vascular diseases as systemic disorders and assessment of haemodynamics: from phantom to patients. Eur Radiol. 2015 Jul 19. so far we have already published assessment [Epub ahead of print] of vascular changes of the aorta including von Knobelsdorff-Brenkenhoff F, Hennig P, Menza M, Dieringer MA, Foell D, Jung B, Schulz-Menger J. Myocardial dysfunction in patients with aortic stenosis and

wall itself is known to be affected by differ- [Epub ahead of print] entquantification hemodynamic of wall factors alteration. promoting The aortic hypertensive heart disease assessed by MR tissue phase mapping. JMRI. 2015 Dec

the impact of interventional replacement of Group Leader Marcel Prothmann remodeling. In 2015 we were able to publish- Prof. Dr. Jeanette Schulz- Dr. Agnieszka Toepper - Menger dicatingthe aortic a valverelation (TAVI) between applying left 4D-flow ventricular im Technical Assistants and remodelingaging. (see figand 2) aortic First furtherwall changes results in are aortic in Study Nurses (all part time) stenosis. Another aspect is the accurate quan- Scientist Denise Kleindienst Edyta Blaszyk Kerstin Kretschel arrhythmias. We have performed systemic Dr. Simone Fritschi Evelyn Polzin analysistification fromof flow phantom velocity alsoexperiments in patients to with pa- Stephanie Funk Annette Koehler-Rohde tient assessment for evaluation of real-time Dr. Fabian Muehlberg Elke Nickel

flow. That research is intended to identify MDC Research Report 2016 227 ECRC (JOINTLY WITH CHARITÉ) Photo: private Photo:

Simone Spuler

Muscle Research Unit and Out-patient Clinic for Muscle Diseases

We are interested in muscle sciences. Muscle stem cells – Satellite cells In our outpatient clinic at the ECRC we care Skeletal muscle harbors its own stem cells called satellite cells located in the stem cell for >1500 patients with genetic and ac- niche between sarcolemma and basal lam- quired skeletal muscle diseases. Therapies ina. Satellite cells are essential for muscle for muscular dystrophies are not available. - Our research is motivated and directed ellite cells are removed from their micro- environmentregeneration. Ifthey muscle become is injured activated or if andsat towards changing this by investigating the differentiate into myoblasts and myotubes. potential of muscle stem cells in regenera- Earlier attempts to use the therapeutic po- tion and restoring muscle function and by tential of satellite cells failed because differentiation interfered with their ability to possible pharmacologic intervention based regenerate muscle. We are trying to pre- on detailed understanding of the molecular serve the regenerative capacity of satellite pathophysiology of muscle wasting. Our cells yet expanding them. We also aim to ge- main disease models are muscular dystro- netically repair mutations in genes causing muscular dystrophy. phies due to mutations in the dysferlin and

the lamin A/C genes. Equally important are Dysferlinopathies the recruitment of excellent young scientists Dysferlin, a large protein located at the sar- to the field as well as the translation of our colemma, is involved in membrane repair after microinjury, a physiological conse- research into the public. MyoGrad is a grad- quence of normal muscle activity. Mutations uate program dedicated solely to muscle in dysferlin lead to muscular dystrophy. We sciences (DFG GK1631, UFA). found that dysferlin is essential for normal glucose metabolism in muscle (Schöwel et al. in revision).

Acquired myopathies: Critical illness myopathy and statin- induced myopathy

coenzym-A-reductase (statins) are pre- scribedInhibitors to ofmillions 3-hydroxy-3-methylglutaryl- of patients of whom

We are exploring genetic and structural aspects10% develop of this musclecommon related myopathy. side Critical effects. illness myopathy occurs as a severe side

large translational and interdisciplinary ap- proacheffect of that intensive involved care several unit treatment. ECRC groups In a we have shown that the failure to activate AMPK and to translocate GLUT4 to the sar- - cular atrophy. colemma contributes significantly to mus

228 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

Funding sources Deutsche Forschungsgemeinschaft (MyoGrad IGK1631; SP 1152/11-1), German French University UFA; Jain Foundation; Foundation Gisela Krebs

www.muskelforschung.de

Patents

PCT-Anmeldung (WO 2016/030371). Cultivating human muscle satellite cells.

Selected Publications

Moshourab R, Palada V, Grunwald S, Grieben U, Lewin GR, Spuler S. A Molecular

Signature of Myalgia in Myotonic Dystrophy 2. EBioMedicine. 2016 May;7:205-11. doi: 10.1016/j.ebiom.2016.03.017. Epub 2016 Mar 14. PubMed PMID: 27322473; -PubMed Central PMCID: PMC4909324. HarrisDíaz-Manera E, Bladen J, Eagle CL, Mayhew M, Grieben A, James U, Harms M, Bettinson M, Jones KJ,K, MooreLochmüller U, Smith H, Mendell FE, Rufi JR, bachMori-Yoshimura L, Cnaan A, M,Bharucha-Goebel Paradas C, Pegoraro DX, Blamire E, Pestronk AM, BravverA, Salort-Campana E, Carlier PG, E, Schreiber-Day JW, Katz O, Semplicini C, Spuler S, Stojkovic T, Straub V, Takeda S, Rocha CT, Walter MC,

Bushby K; Jain COS Consortium.. The Clinical Outcome Study for dysferlinopathy: An international multicenter study. Neurol Genet. 2016 Aug 4;2(4):e89. doi: 10.1212/ NXG.0000000000000089. eCollection 2016 Aug. PubMed PMID: 27602406; PubMed S.Central Human PMCID: satellite PMC4994875. cells have regenerative capacity and are genetically manipulable. J Pax7 positive human muscle precursor cells (satellite cells) Marg A, Escobar H, Gloy S, Kufeld M, Zacher J, Spuler A, Birchmeier C, Izsvák Z, Spuler after 5 weeks in culture Philippi S, Lorain S, Beley C, Peccate C, Précigout G, Spuler S*, Garcia L*. Dysferlin res- cueClin byInvest. spliceosome-mediated 2014; 124: 4257-4265. pre-mRNA trans-splicing targeting introns harbouring

weakly defined 3’ splice sites. Hum. Mol. Genet. 2015; 2:4049-60. Escobar H, Schoewel V, Spuler S, Marg A*, Izsvák Z*. Full-length dysferlin transfer by *equalthe hyperactive contribution Sleeping Beauty transposase restores dysferlin-deficient muscle. Mol Therap Nucl Acid 2016; 5:e277

Group Leader Doctoral Students Prof. Dr. med. Simone Spuler Leonie Heidt (M.D. student) Henning Langer (Ph.D. student,

Consultant Out-patient Clinic Jakub Malcher (Ph.D. student) together with Stefan Kempa, BIMSB) Dr. med. Ulrike Grieben Eric Metzler (Ph.D. student) Tobias Opialla (Ph.D. student, Administrative Manager Susanne Wissler Vinko Palada (Ph.D. student) together with Stefan Kempa, BIMSB)

Scientists Technicians Dr. rer. nat. Helena Escobar Stefanie Belz Dr. rer. nat. Stefanie Grunwald Kornelia Gräning Dr. rer. nat. Andreas Marg Stephanie Meyer-Liesener Dr. med. Verena Schöwel Adrienne Rothe

MDC Research Report 2016 229 ECRC (JOINTLY WITH CHARITÉ) Photo: private Photo:

Ulrike Stein

Translational Oncology of Solid Tumors

Metastatic dissemination of primary tumors (MS). We found kinases essential for MACC1 is directly linked to patient survival and criti- Y-phosphorylation, and entire signalosomes of receptor tyrosine kinases binding pY- cally limits successful therapy in many tumor MACC1.We used a MACC1 promoter reporter entities. Our translational concepts aim at the for high throughput screenings (HTS) using identification of key molecules in tumor pro- the FMP-ChemBioNet (J. von Kries, FMP) gression and metastasis for improved prog- and DKFZ/EMBL libraries (N. Gunkel, DKFZ, J. Lewis, EMBL). The S100A4 promoter re- nosis and therapy of solid cancers. We discovered MACC1, S100A4 and transcriptional targets or protein binding partners thereof porter was used for HTS using the NCI/ andNIH-Lopac S100A4 library transcriptional (R.H. Shoemaker, inhibitors. NCI/NIH as new diagnostic, prognostic and predictive Bethesda MD). We identified the first MACC1 key players for tumor progression and me- Preclinical developments for tastasis. Biomarker development was done in understanding and intervening in established and patient-derived 3D cultures, tumor progression and metastasis We established in vivo intervention strate- cell line-derived and patient-derived xeno- gies targeting MACC1 and S100A4 for me- grafts (PDX) and newly generated genetically tastasis inhibition. shRNA acting on MACC1 engineered mouse models. We exploited this or S100A4 or on transcriptional or post- knowledge for improved prognosis and treat- translational targets thereof decreased metastasis. The small molecule transcriptional ment response prediction using tissue and blood of cancer patients. Novel therapeutic HTS, restricted MACC1- or S100A4-induced approaches are currently tested in clinical metastasisinhibitors of in MACC1 mice. Thisor S100A4, repositioning identified of alby- ready FDA-approved drugs for the new in- trials to treat patients with metastatic dis- dication of metastasis restriction paves the ease using small molecule inhibitors acting on way for clinical trials. these biomarkers. Based on our MS data, we showed MACC1- induced metastasis restriction in mice when Cancer cell signaling of MACC1 deleting MACC1 pY sites or inhibiting MACC1- and S100A4 for pathways-based phosphorylating kinases using drugs tested interventions MACC1 was discovered in our group acting from tumors and metastases of CRC patients as decisive driver for tumor progression and (B.in clinicalRau, Charité, trials. J. WeHoffmann, generated epo), PDX and models evalu- metastasis for a broad variety of solid can- ated biomarkers and pathways for progres- cers. By transcriptomics, we found MACC1 sion, metastasis, and response prediction totarget genes, e.g. SPON2, which induce wards conventional and targeted therapies. metastasis and are prognostic for colorectal cancer (CRC) patients. We explored the mouse models vil-MACC1, vil-MACC1/ApcMin miRNA landscape for MACC1- and S100A4- We(B. Jerchow, generated MDC). the firstvil-MACC1/Apc transgenic Min MACC1 mice - showed increased number and size of tu- genetic regulations, interplays for metas- mors, possessed an accelerated adenoma- tasisinduced prognostication metastasis, identified and patient miRNAs survival. epi carcinoma-sequence, an invasive phenotype linked to reduced survival, and increased Dittmar, MDC) and phospho-interactome (B. Wnt and pluripotency signaling with up- Küster,We identified TU Munich) the MACC1 by mass interactome spectrometry (G. regulation of Nanog and Oct4. These pluri-

230 MDC Research Report 2016 ECRC (JOINTLY WITH CHARITÉ)

MACC1 induces cell motility, proliferation, and scattering in cell culture and metastasis in mouse models. It acts as transcription factor regulating e.g. Met expression and also interacts with proteins via e.g. its SH3 domain or proline-rich motif. By microarrays we identified the MACC1 transcriptome. By HTS using the MACC1 promoter we identified first transcriptional MACC1 inhibitors. By MS we identified the MACC1 interactome and phospho-interactome for novel interventions.

Patents / Patent applications

for prognosing and treating colorectal cancer. US Provisional US62/422,573 potency markers are regulated by MACC1 Stein(15.11.2016) U, Herrmann P, Ilm K, Shanmugam K, Rohr U. Compositions and methods and strongly correlate with MACC1 levels in Stein U, Schwabe H, Walther W, Schlag PM. MACC1-polypeptide and use thereof for the diagnostic and therapy of tumors. CA 2,526,240 (16.9.2014), EP1646649B1

MACC1-induced tumor progression in CRC B2 (14.12.2010) acts,CRC patients.at least in We part, provide via the first newly evidence discovered that Stein(27.2.2013), U, Juneja JP2006-519895 M, Walther W. Schlag(13.5.2011), PM, von AU2004259281 Kries JP, Kobelt (12.5.2011), D. Pharmacologic US7,851,168 inter- MACC1/Nanog/Oct4 axis. vention of tumor growth and metastasis via the MACC1 pathway by using rottlerin.

Stein U, Lederer A, Schlag PM, Herrmann P, Seehofer D. MACC1 as a prognostic Clinical translation for improved EP14179874.4 (5.8.2014) prognosis, prediction and restric- Stein U, Walther W, Sack U, Scudiero D, Schlag PM, Shoemaker RH. Niclosamide tion of tumor progression and forbiomarker treatment in hepatobiliaryof cancer metastasis. tumors. US14/112,521 US14/291,831 (12.11.2013), (30.5.2014) EP2012/057049 metastasis We demonstrated MACC1 as metastasis in- Selected(18.4.2012), PublicationsEP11162875.6 (18.4.2011) ducer, prognostic biomarker for CRC and as a target for metastasis inhibition. MACC1 Lemos C, Hardt MS, Juneja M, Voss C, Forster S, Jerchow B, Haider W, Blaker H, Stein U. MACC1 induces tumor progression in transgenic mice and colorectal cancer outperforms KRAS, BRAF, and MMR muta- patients via increased pluripotency markers Nanog and Oct4. Clin Cancer Res. 2016, tion, and improves prognosis combined

with KRAS13 mutation. We showed highest 22:2812-24. prognostic value of MACC1 at the CRC tumor Schmidgene of MACC1,F, Wang drivesQ, Huska colorectal MR, Andrade-Navarro cancer metastasis MA, in Lemm mice M,and Fichtner is prognostic I, Dahlmann for invasion front (V. Kölzer, Uni Bern). We estab- M,colorectal Kobelt D,cancer Walther patient W, Smith survival. J, Schlag Oncogene. PM, Stein 2016, U. 35:5942-52.SPON2, a newly identified target lished a novel, non-invasive liquid biopsy test Lederer A, Herrmann P, Seehofer D, Dietel M, Pratschke J, Schlag P, Stein U. Metasta- demonstrating the prognostic value of circu- sis-associated in colon cancer 1 is an independent prognostic biomarker for survival lating MACC1 transcripts in patient blood in Dahlmann M, Okhrimenko A, Marcinkowski P, Osterland M, Herrmann P, Smith J, Heizmannin Klatskin CW, tumor Schlag patients. PM, Stein Hepatology. U. RAGE 2015, mediates 62:841-50. S100A4-induced cell motility via regulated genes, demonstrated their metas- MAPK/ERK and hypoxia signaling and is a prognostic biomarker for human colorec- tasis-inducingprospective studies. potential We in identified mice and MACC1- their tal cancer metastasis. Oncotarget. 2014, 5:3220-33. prognostic value for CRC patient survival. We Steinmetastasis. U, Walther Nature W, Med.Arlt F, 2009, Schwabe 15:59-67. H, Smith J, Fichtner I, Birchmeier W, Schlag PM. and driver for progression and metastasis in MACC1, a newly identified key regulator of HGF/Met signaling, predicts colon cancer tumorsconfirmed and MACC1 blood from as prognostic hepatobiliary, biomarker gastric, pancreatic, breast cancer and glioblastoma Group Leader patients (A. Lederer, S. Burock, C. Denkert, Prof. Dr. Ulrike Stein Maria Haustein* K. Detjen, Charité, C. Hagemann, Uni Würz- burg), and as predictive marker for patient Senior Scientists PD Dr. Wolfgang Kemmner ManishaKatharina Juneja* Ilm Prof. Dr. Wolfgang Walther JessicaFrancesca Pahle* Imbastari* Ventanastratification Tucson for AZ).adjuvant We showed treatments the imporin CRC- Harikrishnan Radhakrishnan tancestage IIof (HoffmannS100A4 for La CRC Roche in the Pleasanton context of CA, its Scientists receptor RAGE, which itself and combined Dr. Mathias Dahlmann Fabian Zincke* with S100A4 serves as prognostic biomark- Dr. Clara Faria Alvares de Lemos* Inna Zaimenko* er for metastasis and survival. Therapeutic Dr. Dennis Kobelt Technical Assistants translation is done in a prospective clinical Dr. Giridhar Mudduluru* Jutta Aumann* - Dr. Christoph Treese* Pia Herrmann Janice Smith phase II trial treating CRC patients for metas PhD/MD students Cynthia Voss Keilholz,tasis inhibition Charité,). with Therapeutic a newly identified, outcome FDA- is Müge Erdem* monitoredapproved small by circulating molecule S100A4(NCT02519582; transcripts U. Markus Hardt* *part of time reported in patient blood.

MDC Research Report 2016 231 232 MDC Research Report 2016 Scientific Infrastructures & Technology Platforms

Jutta Steinkötter

MDC Research Report 2016 233 Scientific Infrastructures & Technology Platforms

Jutta Steinkötter

echnology Platforms at the MDC sup- Strategic developments of the past years port established and front-end ap- initiated the establishment and combina- T plication of technologies, provide in- tions of expertise, resources and services strumentation and methodologies and are for new Technology Platforms. engaged in collaborative research projects and technology developments. Driven by BioinformaticsBIMSB established and PlatformsLight Microscopy. for Genomics, These the portfolio of our Technology Platforms Integrative Proteomics and Metabolomics, developsthe demands hand of inthe hand MDC’s with scientific progress topics, and Platforms, where know-how and expertise innovation in research and with new de- platform concepts were blueprints for BIH velopments on the technology market. at the MDC, such as Genomics, Metabolomics - andcontributed Proteomics to the set-up of BIH Platforms forms is to provide state-of-the-art services researchers and projects. Additional technol- The concept of the MDC’s technology plat ogies and infrastructures serving the were needs established of the BIH in applications and co-development of meth- Pluripotent as well as specific tailor-made technology Stem Cell Platform and the Biobank are host- technological innovation. This approach is edcollaboration with one branch with theat the BIH: MDC the and a second aodologies powerful at technology the forefront support of scientific for research and branch at the Charité. And last but not least, - another collaboratively and jointly operated tions and patentable innovation. research infrastructure is the Chemical Biol- groups and results in high-profile publica ogy of the FMP, supported by the MDC.

Shown are human induced pluripotent stem cells derived cardiomyocytes. The structure of the sarcomeres were stained using antibodies against cardiac markers alpha actinin (red) and cardiac troponin (green). The nucleus was stained with DAPI (blue). Graphic: Sebastian Diecke, MDC Diecke, Sebastian Graphic:

234 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Executive Management Jutta Steinkötter

Scientific Board Holger Gerhardt, Markus Landthaler, Matthias Selbach Graphic: Anje Sporbert, MDC Anje Graphic:

3D reconstruction of the DNA (blue) replication machinery in the cell nucleus visualized by Central Technology Platforms and proteins (green) involved in DNA synthesis and Core Facilities incorporated nucleotides (red). Advanced Light Microscopy Anje Sporbert

their portfolio of technologies and instru- Biobank (MDC and BIH) Tobias Pischon mentsMost of and the thereforeMDC’s Platforms increase are the expanding potential for collaborations and technology innova- Bioinformatics tion. The Genomics Platform for example Altuna Akalin provides not only standard service for Ge- Chemical Biology (MDC and FMP) - lumina and PacBio Systems but is also de- Electron Microscopy velopingnome and applications Transcriptome for single Sequencing cell analysis, on Il BettinaJens Peter Purfürst von Kries & Marc Nazaré - idics systems. This major trend shapes re- Genomics (MDC and BIH) searchbased onin celltranslational sorting and and various basic areasmicroflu and Sascha Sauer many approaches are directed towards Magnetic Resonance Thoralf Niendorf decisions and mechanisms of disease. Ex- Metabolomics (BIH) pandingcell type capacities identification, for animal tracking, phenotyping cell fate Jennifer Kirwan were planned and constructed in the past Light Microscopy years and in 2017, the upcoming reporting Andrew Woehler period, the in vivo Pathophysiology will offer more space for sophisticated imaging, Pathophysiology physiology, behavior and molecular studies Arnd Heuser of mice. More structural work will be sup- Pluripotent Stem Cells (MDC and BIH) ported through the EM Platform, which re- Sebastian Diecke cently installed a new CryoEM. Preparative Flowcytometry As shown in the platform reports, they sub- Hans-Peter Rahn stantially contribute to innovation and publication output of the institute, and based Proteomics (MDC and BIH) Gunnar Dittmar on recruitments, investments and strategic expansions of the past years, they can offer Proteomics / Metabolomics state-of-the-art instrumentation, knowl- Stefan Kempa edge and expertise for research projects at Transgenics the MDC and in collaboration with local, na- Ralf Kühn tional and international research partners.

MDC Research Report 2016 235 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Anje Sporbert

Advanced Light Microscopy

The Advanced Light Microscopy (ALM) scope booking, news, reporting of incidents technology platform provides scientists with and mailing lists. The software can readily be adapted to suit the needs of other core access to state-of-the-art microscopes and facilities. imaging techniques and supports research collaborative imaging projects we can projects with customized scientific, method- - ological and technical expertise. ALM cur- odologicalIn support covering steps includ- ingoffer project very customized planning, thescientific optimization and meth of rently supports about 200 researchers from sample preparation and image acquisition, more than 45 groups, mainly from the MDC - and BIMSB, but also from ECRC, Charite and tablishment of new imaging methods. image analysis workflows and even the es FMP. Selected projects & ALM currently hosts 14 high-end, cutting Method development edge imaging systems that cover confo- Label-free imaging – polarised second cal laser scanning, multi-photon, wide- harmonic generation (SHG), A. Margine- - anu: 2-photon microscopy can be used not

3D-stimulatedfield fluorescence, emission total depletion internal (STED) reflec organs at higher depths, but also to gener- tion (TIRF), light-sheet, spinning-disk and ateonly other to excite effects fluorescence such as SHG. in SHG tissues imaging and has been applied to the study of changes Thismicroscopy, portfolio fluorescence of advanced lifetimetechniques imaging per- in the extracellular matrix. Because SHG (FLIM) as well as laser microdissection. yields information on the orientation of specimens as well as live imaging of cells, - tissuemits the preparations, imaging of aorganoids, wide range small of fixed organisms and animals. intravitalcollagen fibers,2-photon adding microscope polarisation has imagbeen Advanced image analysis and processing offers even more insights. The ALM’s- ing tools are available on dedicated image ments of cartilage samples obtained from analysis workstations for 3D reconstruc- micemodified (Fig.1A, to permit collaboration polarised AG SHG Poole). measure - A. ing, tracking, stitching, and deconvolution. Margineanu/M. Richter: Fluorescence mi- Regulartions, colocalization, seminars quantification,covering theoretical count croscopyFluorescence measures lifetime not imaging only the (FLIM), intensity principles and main applications of light/ and cellular distribution of dyes, but also

month. Each ALM user receives in-depth, the lifetime. This method is sensitive to the individualizedfluorescence microscopy practical aretraining given in each the other properties of fluorophores such as safe handling of microscopes to ensure the to detect intracellular protein-protein in- correct acquisition of data. Regular per- teractions.fluorophore The environment energy transfer and can from be aused do- formance tests of our instrumentation are nor to an acceptor (labeling the proteins another measure we take to ensure and -

acquired with ALM setups. providesof interest) a moreoccurs robust only whenapproach the two then fluo ra- Themaintain ALM developedthe high quality core facilityof scientific manage data- tiometricrophores imaging are at distances since only <10 the nm.reduction FLIM ment software to offer easy access to us- is also more quantitative, providing donor- support and troubleshooting. The software acceptorof the donor distances lifetime and must interacting be detected. popula It- providesers and toolsonline for user efficient registration, communication, micro-

tion fractions. FLIM-FRET has been applied 236 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

at ALM to identify substrates that could interact with the titin kinase catalytic domain (Fig.1B, collaboration AG Gotthardt) and to identify interacting compartments of the peptide loading complex (collaboration AG Klötzel).

molecules using dyes or genetically-encod- FLIM can also detect biologically active- cence lifetimes upon binding the molecule ofed interest. biosensors One which biosensor change applied their at fluores ALM is Peredox, which detects changes in the cytoplasmic redox ratio of NAD(P)H/NAD+.

HTogether and FAD with gives the an indication biosensor, of FLIM the energy of the metabolismendogenous of cellular cells. fluorophores NAD(P) Cleared specimens for confocal and lightsheet microscopy, M. Richter/A. Sporbert:

- Fig1. A) Polarised SHG from cartilage in parallel (red) and parentImaging by entire removing organs scattering or specimens substances greatly perpendicular (green) detection. Colour-coded images show the suchbenefits as lipidsfrom rendering and matching tissues the more refractive trans calculated polarisation images (sample M. Servin-Vinces) B) index of the sample to the optical setup. Nu- Protein-protein interactions in situ using FLIM imaging. merous clearing techniques have been pub- Reduced GFP lifetime (red) indicates specific interaction (sample lished in the last years. We have optimized F. Rudolph) C) High resolution 3D reconstruction of 480 the high-resolution imaging of cleared rat stitched images (sample F. Heuberger) D) Laser-assisted hearts, mouse embryos and spinal cords for membrane injury of cardiomyocyte (sample L. vanVuuren). further analysis by 3D reconstructions (collaborations with AG Müller, W. Birchmeier (Fig.1C), Spagnoli). Laser-assisted membrane injuries, A. Sporbert: High-power lasers can be used to Selected Publications cut and catapult cells and tissues (through laser-microdissection microscopy) and also Sporbert A. and fast diffusion with a constant ratio throughout S-phase in replicating mamma- to create targeted cellular injuries. Uses Zessin P.J.M.; ; Heilemann M. PCNA appears in two populations of slow include a range of biophysical and mecha- Margineanu A., Chan J.J., Kelly D.R., Warren S.C., Flatters D., Kumar S., Katan M., nistic studies targeting cells or their com- Dunsbylian cells, C.W., Scientific French Reports P.M.W. (2016)6: 18779 Screening (2016-01-13) for protein-protein interactions using ponents. An assay for reversible membrane 6 damage was implemented in skeletal mus- WarrenFörster resonanceS.C., Margineanu energy transferA, Katan andM., Dunsbyfluorescence C., French lifetime P.M. imaging (2015) Homo-FRETmicroscopy, cle cells and cardiomyocytes to evaluate the Scientific Reports, : 28168 repair capacity of cell membranes (collabo- Based Biosensors and Their Application to Multiplexed Imaging of Signalling Events ration AG Morano, Fig.1D). in LiveCseresnyes Cells, Int. J. Z. Mol. Sci., 16(7):14695-14716 In-vivo/intravital 2-photon microscopy Heinig K.; Gaetjen M.; Grau M.; Stache V.; Anagnostopoulos I.; Gerlach K.; Niesner - (A. Margineanu) for small animals and in- R.A.;dritic cells is a pivotal; Hauser step in A.E.; murine Lenz chronic P.; Hehlgans lymphocytic T.; Brink leukemia R.; Westermann B cell activation J.; tact tissue preparations is invaluable in Doerken B.; Lipp M.; Lenz G.; Rehm A.; Hoepken U.E. (2014) Access to follicular den Sporbert A.; Cseresnyes Z. the validation of results that have been ob- and proliferation , Cancer Discovery 4: 1448-1465 tained by standard techniques in physiolog- resolution imaging of cellular; Heidbreder structures onM.; standard Domaing confocal P.; Hauser microscopes S.; Kaltschmidt by three- B.; ical environments. Examples of projects we Kaltschmidt C.; Heilemann M.; Widera D. (2013) Simple method for sub-diffraction have supported are the multi-color imaging photon absorption of quantum dots PLoS ONE 8 (5) of mouse brain (AG Priller), mouse brain stem (AG Siffrin) and spleen (AG Höpken). Group Leader Quantum dot triexciton imaging (K. Dr. Anje Sporbert Grohmann): Higher order excitated states can be generated in some types of commer- Scientist Technical Assistants cially available Quantum Dots (QD) by con- Dr. Anca Margineanu Matthias Richter ventional lasers. Detection of these tri-exci- Dr. Vitali Matiash (DFG grant, tons combined with deconvolution results part-time, 01/2014-04/2015) Secretariat in a up to 1.7 fold higher spatial resolution Konstantin Grohmann (since Manuela Brandenburg of QD labeled subcellular structures using 11/2015) Cornelia Stärkel standard confocal microscopes.

MDC Research Report 2016 237 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Tobias Pischon

Biobank

The biobank offers scientists a platform for have the capacity to store 2.5 million 250 state of the art storage and access of liq- tubes with 700 µL, 1000 µL and 2 mL. The uid biosamples at ultra-low temperatures. systemµL tubes; guarantees however, thatthe system there is can no interrupalso hold- Samples are stored at temperatures below tion of the cold chain during the process of -160°C in the vapor phase of liquid nitrogen transferring the samples into the biobank. in a fully automated system. This allows high The tubes with biological samples are preferably transported to the biobank on racks quality and reliability for storage as well as in boxes on dry ice. The boxes are opened in swift identification, fast access and efficient a dedicated room at a temperature of -20°C retrieval of samples for analysis. The biobank and the racks are immediately put from dry also provides advice and consultation for - mated robot system transfers the racks to projects with biospecimens regarding de- theice intoliquid a nitrogen-80°C buffer tanks system. at a temperature A fully auto of sign, planning and implementation of studies -20°C, and the tracks are put into the tanks (e.g. SOPs). As part of the Berlin Institute of for storage at temperatures below -160°C. The system also allows to store and retrieve Health (BIH) the biobank also stores sam- single tubes or sets or collection of tubes. ples of BIH research projects, cohort stud- For these instances, a fully automated tube ies and trials. In addition, the biobank stores picker is integrated into the system that biological samples from 30.000 participants The biobank is equipped with a laboratory of the Metropolitan area Berlin-Brandenburg works at a temperature of -80°C. of the German National Cohort. that is able to provide full documentation aboutinformation the samples management stored within system the (LIMS) sys- Focus tem, including but not limited to character- The biobank is equipped to support basical- istics of the samples, temperature, storage ly any type of research study that requires condition, as well as about research proj- storage of liquid samples at ultra-low tem- ects and analysis results.

suitable for projects that require both, the Contribution to major research abilityperatures that (below collected -160°C). biosamples It is particularly may be initiatives stored over longer time periods but also The MDC stores biological samples from that they may quickly be made available for 30.000 participants of the Metropolitan analysis if necessary. For example, this in- area Berlin-Brandenburg of the German Na- cludes cohort studies, clinical trials, but also tional Cohort (GNC). The GNC includes par- experimental studies (e.g. stem cells). ticipants from the general population aged 20-69 years who are followed up for the Equipment and infrastructure The biobank primarily consists of a fully-au- recruitment of participants into the study, tomated liquid nitrogen storage system. The includingincidence aof detailed major chronicphenotyping diseases. and collecInitial- samples are stored in the vapor phase of liq- tion of biological samples, has been started uid nitrogen at temperatures below -160°C. To avoid repeated freeze-thaw cycles sam- addition, there will be a re-assessment of all ples are preferably stored in tubes with in 2014 and will be completed by 2019. In small volumes. The system currently holds Every 2-3 years questionnaires on lifestyle 2 large dimension liquid nitrogen tanks that changesstudy participants and new betweenmajor diseases 2018 and will 2023. be

238 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photos: Horst Krüger/MDC Photos:

collected. Mortality follow-up and systematic linkage with disease registries will be Core Facility will store a wide range of dif- performed. The GNC is thus optimally deformation system (LIMS). The BIH Biobank signed to study the association of baseline research project under safe and quality-con- biomarkers as well as changes in biomarker trolledferent typesconditions of specimens and will make collected them in avail BIH- concentrations over time, with risk of inci- dent diseases. will also provide advice and consultation The MDC also stores biological samples able to BIH researchers. The BIH Biobank- from 1200 participants of the BioCog proj- ing design, planning and implementation of ect (Biomarker Development for Postopera- studiesfor BIH projects(e.g. SOPs), with data biospecimens protection regardissues, and ethical issues. A number of services are which is supported within FP7 by the Eu- or will be offered, such as tissue sectioning ropeantive Cognitive Union. ImpairmentThe aim of BioCog in the is Elderly), to dis- and histology as well as immunohistochem- cover, identify and develop biomarkers for istry, DNA or RNA extraction, etc. the prediction of postoperative cognitive

occurs frequently after surgery among per- sonsimpairment within (POCI)the general in elderly population, people. POCIand includes postoperative delirium as well as - Cog, biological samples are being collected beforepostoperative surgery cognitive as well asdysfunction. on different In timeBio points after surgery, thus allowing detailed investigations for prediction of adverse outcomes. Head of the Group Prof. Dr. Tobias Pischon As part of the Berlin Institute of Health Assistants (BIH) the biobank also stores samples of Scientist Dr. Claus Derz biobankBIH research at two projects, locations cohort (Charité studies CVK and Dr. Jürgen Janke Marit Kanter MDC)trials. Thewith BIH a synchronizedBiobank Core laboratoryFacility is onein-

MDC Research Report 2016 239 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Altuna Akalin

Bioinformatics Platform, BIMSB

Bioinformatics is the application of compu- contains annotation and genome-wide da- tational methods to the management and tasets for human and most popular model - analysis of biological data. By using bioin- vide space and instructions for users to link formatics, high-dimensional data can be theirorganisms. data to In the the local last year,UCSC we browser started for to easypro translated to biological knowledge. With the visualization. advent of high-throughput technologies in We also provide a shiny server which contains statistical web applications and interac- biology, computational analysis needs of the tive documents (http://shiny.mdc-berlin.de). labs are on the rise. This creates a challenge The server lets users manipulate the data in in method development, application and also - sumptions, and customize their analysis for a demand in training new individuals. real-time, they can sort, filter, or change as BIMSB Bioinformatics platform creates and On top of these, in 2016, we started to provide maintains bioinformatics tools and data- thetheir Galaxy specific workbench needs. for bioinformatics. bases, and studies biology using those Galaxy is a framework for doing bioinformatics without the knowledge of programming. in-house or publicly available computational The users simply link or upload their data to techniques. In addition, the platform pro- their galaxy session and select the tool and vides collaboration opportunities, helpdesk, the parameters they want to run using a web and training for MDC scientists. The platform browser based interface. Our Galaxy server is available at http://galaxy.mdc-berlin.net. also provides maintanence for hardware Until now, we will held two tutorials on how that hosts BIMSB web apps, group compute to use the galaxy framework and these will servers and virtual machines. The platform continue in the future.

aims to help MDC scientists address their Scientific software repositories for biological questions using and developing reproducible research computational methods. Since December 2014 the Bioinformatics Platform is using GNU Guix to satisfy the Databases and web apps - - vironments of our HPC users. GNU Guix is tains and develops databases and web apps anrequirements advanced functional for scientific package software manager, en forThe bioinformatics. BIMSB Bioinformatics With the platform explosion main of offering support for per-user software pro- genomics data, visualizing genomics annotation in harmony with high-throughput data- versions and package variants, transaction- sets became more and more important. We alfiles, upgrades simultaneous and roll-backs, installation and a of controlled different are maintaining external and internal data- build environment to maximize reproduc- bases and web applications that contain ana- ibility. lytical tools for bioinformatics. Over the past years, we have added hundreds We are maintaining the DoRiNA database (http://dorina.mdc-berlin.de) for RNA bind- - andof package machine-learning definitions tofor the tools quickly and librariesgrowing sociated with our local copy of UCSC genome used in the field of bioinformatics, statistics, browsering protein interactions;(http://genome.mdc-berlin.de this database is as). packages available are listed at http://guix. Our local copy of the UCSC genome browser mdc-berlin.decollection of scientific software in Guix. The

240 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Selected Publications

Uyar B, Yusuf D, Wurmus R, Rajewsky N, Ohler U and Akalin A (2017). “RCAS: an Guix allows us to offer tailor-made bioinformatics software environments for individual Acids Research. doi: 10.1093/nar/gkx120. users, projects, and whole groups. These BlinRNA K,centric Dieterich annotation C, Wurmus system R, Nikolaus for transcriptome-wide Rajewsky, Markus regions Landthaler, of interest.” Akalin Nucleic A. DoRiNA 2.0—upgrading the doRiNA database of RNA interactions in post-transcrip- environments are portable across different tional regulation. Nucleic Acid Research , 2015 machines, easily reproducible, and safely iso- Akalin A*, Kormaksson M, Li S, Garrett-Bakelman FE, Figueroa ME, Melnick A, Ma- lated from changes, providing a reliable basis son CE. methylKit: a comprehensive R package for the analysis of genome-wide DNA for reproducible research. * (Co-corresponding author)

Akalinmethylation A, Garrett-BakelmanFE, profiles. Genome Biology, KormakssonM, 2012 Oct BusuttilJ, 3;13(10):R87 et al. Base-pair resolution Consultation and training DNA methylation sequencing reveals profoundly divergent epigenetic landscapes in The platform also offers knowledge-base, help-desk and training for MDC scientists. PlessyC*,Pascarella G*, Bertin N*, Akalin A*, CarrieriC, VassalliA, LazarevicD, et al. Promoteracute myeloid architecture leukemia.. of PLoSmouse Genetics olfactory 2012;8(6):e1002781. receptor genes. Genome Research. 2011 We provide a regular help-desk session, Dec 22. * (Equal Contribution) users are welcome to ask any bioinformat- icscalled question “Bioinformatics and would walk-in not need clinics”, a prior where appointment. We can direct users to right tools Group Leader PhD Students and methods or different groups who have Dr. Altuna Akalin Jonathan Ronen

could be used to initiate collaborations. Scientists Katarzyna Wreczycka Thethe rightplatform expertise. offers regular In addition, training this opportu avenue- Dr. Vedran Franke Inga Patarcic nities, such as courses on R programing, Unix Dr. Brendan Osberg Administrative assistant and SGE, and version controls system git. Dr. Bora Uyar Sabrina Deter Since 2015, we organize international hands- Dr. Dilimulati Yusufujiangaili on courses on computational genomics. Most Complete list of team members can be of the applicants for these courses are from IT staff found in chapter Berlin Institute for Dan Muntenau Medical Systems Biology, Group “Bioin- total, we have provided 11 courses/tutorials Ricardo Wurmus formatics, Mathematical Modelling” other institutes abroad and in Germany. In

to 188 participants. MDC Research Report 2016 241 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: Silke Oßwald/FMP Silke Photo: Oßwald/FMP Silke Photo:

Jens Peter von Kries Marc Nazaré

Chemical Biology Platform Screening Unit – Medicinal Chemistry – Drug Design

The Chemical Biology Platform provides The genome wide RNA-interference col- interdisciplinary scientific support in Screen- lections (siRNA) target mouse (four siRNA molecules per gene, pooled) or human ge- ing, Medicinal Chemistry and Computational nomes (three separate siRNA molecules). Chemistry and is jointly operated by the MDC From 40.000 (mouse) to 140.000 (human) and the Leibniz Institut für Molekulare Phar- transfections are required for a genome- makologie, which is located on our joint Cam- wide screen, if not limited to the so called pus in Berlin-Buch. Jens Peter von Kries, Marc Nazaré and Ronald Kühne provide expertise “druggableHighlight genome”.Projects in the identification and chemical optimization Chronic Pain and Mechano-Receptors of drugs and the generation of chemical tools Mechanoreceptors in the skin allow the perception of the slightest brush by transfor proof-of-concept studies or of labeled forming movement into electrical signals biosensor molecules. A broad range of state controlled by the stomatin-like protein-3 of the art technologies for High-Throughput- Screening in small molecule libraries actually inhibitors(STOML3). ofIn STOML3collaboration oligomerization with the group re- containing about 66.000 drug-like compounds ducingof Gary theLewin sensitivity we identified in sensory small-molecule neurons or in RNAi-libraries targeting human or mouse and silencing mechanoreceptors in vivo genomes are available for screening ap- which may allow to treat tactile-driven pain following neuropathy. proaches. The platform enables to identify and optimize bioactive small molecules as tools to Identification of the gene for the Volume- unravel molecular mechanisms in health and Regulated Anion Channel VRAC disease and contributes to identify and vali- The group of Thomas Jentsch and the date novel therapeutic targets. gene encoding for a channel protein that regulatesScreening cell Unit volume. identified For athis long-sought purpose, Screening Compound Collections (Small Molecule Drugs & RNAi) by ~140.000 transfections with interfering The screening compound collections are RNAmRNA using of 21.687 a cellular human reporter genes was system targeted for continuously curated and extended and provide currently more than 66.000 compounds. About 30.000 compounds have protein.iodide influx The Volume-Regulated and intra-cellular Anion quenching Chan- been selected for screening according to nelof anwas iodide-sensitive stimulated by a yellowchange fluorescentof medium their structural similarity with drugs from containing different salt conditions. The in- - hibition of quenching provided a read out tures of approved drugs. Additional 3.300 for high-speed kinetic imaging and the iden- compoundsthe World Drug are Indexidentical containing with approvedsubstruc drugs or experimental clinical candidates components of VRAC. suitable for a potential drug repurposing. tification of LRRC8 heteromers as essential 9.000 compounds are donations from a net- Drugs for functional dissection of differ- work of chemists and 20.000 compounds ent paths for NFKB activation were derived from a natural product collec- The laboratory of Claus Scheidereit has a mation from Analyticon Discovery. jor interest in understanding the molecular

242 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

mechanisms in regulation of NFKB signaling in development and disease like in cancer

Screening with automated microscopes and or inflammation. Applying High-Content--

the fluorescent detection of NFKB translo inducedcation to NFKB the nucleus,activation we and identified therefore drugs pro- videwhich unique specifically tools to interfere functionally only dissect with DNA the Marc Nazaré molecular mechanisms behind. The drugs -

roundsidentified of weresynthesis, further testing validated and by analysis chemi cal synthesis and derivatization. Iterative - tionshowed of the a target well-defined proteins. structure-activity- relationship which will assist the identifica Development of specific inhibitors of the - tyrosine phosphatase Shp-2 tive TPH inhibitors that are able to modu- - lateidentified physiological and further serotonin developed levels. highly The ac cance in signal transduction and disease pathology,In stark contrast phosphatases to their validatedare notoriously signifi our inhibitors allowed us to elucidate the bindingX-ray co-crystal mode and structures to reveal obtainedthe structural with The protein tyrosine phosphatase Shp-2 playsdifficult a critical to inhibit role using in growth small molecules.factor-me- protein-ligand interaction of these inhibi- diated processes, primarily by promoting tors.determinants Several inhibitors for the remarkablyare currently efficient under- the activation of the RAS/ERK signaling pathway. Aberrant gain-of-function mutations are associated with several meta- going in vivo efficacy studies in mice. Selected Publications Birchmeier, a re-scaffolding approach that Du, J., Neuenschwander, M., Yu, Y., Däbritz, J.H.M., Neuendorff, N.-R., Schleich, K., involvesstatic cancers. replacing In collaboration the former withframework Walter Bittner, A., Milanovic, M., Beuster, G., Radetzki, S., et al. 2017. Pharmacological of a tyrosine phosphatase Shp2 inhibitor restoration and therapeutic targeting of the B-cell phenotype in classical Hodgkin

the discovery of novel structural classes lymphoma. Blood 129, 71–81. and(W. Birchmeiereliminated several et al., PNAS,chemical 2008), liabilities, led to Kozian, D. H.; von Haeften, E.; Joho, S.; Czechtizky, W.; Anumala, U. R.; Roux, P.; Dudda, i.e. unfavorable structural features. These Wetzel,A.; Evers, C., A.; Pifferi, Nazare, S., Picci,M.; Modulation C., Gök, C., ofHoffmann, Hexadecyl-LPA-Mediated D., Bali, K.K., Lampe, Activation A., Lapatsina, of Mast L., novel compounds are not only active in a Fleischer,Cells and MicrogliaR., Smith, byE.S.J., a Chemical et al. 2016. Probe Small-molecule for LPA5. ChemBioChem inhibition of 2016,STOML3 17, oligomer861-865.- sub-micromolar range in the Shp2-enzyme ization reverses pathological mechanical hypersensitivity. Nat. Neurosci. Voss, F.K.,Ullrich, F., Münch, J., Lazarow, K., Lutter, D., Mah, N., Andrade-Navarro, MA., assay, but are also effective in the low micromolar range on hepatocyte growth fac- an Essential Component of the Volume-Regulated Anion Channel VRAC. Science 344: tor (HGF)-stimulated canine MDCK-C cells, von Kries, J.P., Stauber, T., and Jentsch, TJ. 2014 Identification of LRRC8 Heteromers as as well as human pancreatic tumor cells for 634-8. epithelial-mesenchymal transition (EMT), a Hu, H.-Y.; Lim, N.-H.; Juretschke, H.-P.; Ding-Pfennigdorff, D.; Florian, P.; Kohlmann, - hallmark of cancer cell dissemination. M.;trophic Kandira, lesions. A.; PeterChem.Sci. von Kries,2015, J.;6, Saas,6256-6261. J.; Rudolphi, K. A.; Wendt, K. U.; Nagase, H.; Plettenburg, O.; Nazare, M.; Schultz, C.; In vivo visualization of osteoarthritic hyper Tryptophan hydroxylase (TPH) inhibitors The neurotransmitter serotonin [5-hy- Group Leader Screening Unit Group Leader Medicinal Chemistry droxytryptamine (5-HT)] is causally in- Dr. Jens Peter von Kries Dr. Marc Nazaré volved in multiple aspects of mood control in the central nervous system, such Scientists Scientists as regulating sleep, anxiety, drug abuse, Dr. Katina Lazarow** Dr. Upendra Anumala* - Dr. Martin Neuenschwander rotonin regulates vascular tone, gut motil- Dr. Silke Radetzki*** Dr. Hassen Bel Abed** ity,and primary food intake. hemostasis, In peripheral and cell-mediated tissues, se M. Sc. Marc Wippich Dr. Isabel Fernández Bachiller* immune responses, and is associated with Dr. André Horatscheck* diseases like irritable bowel syndrome and Technical Assistants Dr. PeterMaría Lindemann**Pascual López-Alberca*** carcinoid syndrome. The biosynthesis of Carola Seyffarth Dr. Vera Martos serotonin is a highly-regulated two-step Sabrina Kleissle** Dr. Edgar Specker process, starting with the essential amino Romy Leu** acid L-tryptophan (Trp), while tryptophan Andreas Oder*** Technical Assistants hydroxylase (TPH) is the initial and rate- Keven Mallow** limiting enzyme in the biosynthesis of se- * part of the reporting period Sandra Miksche** Team funded by FMP, Dr. Lioudmila Perepelittchenko Michael Bader and Udo Heinemann (both Jessica Przygodda* MDC),rotonin. and In collaborationJens von Kries with (FMP), the groups we have of third party or **MDC ***BIH.

MDC Research Report 2016 243 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Bettina Purfürst

Electron Microscopy

In the period reported, we realized more than 25 EM-projects from nearly all research areas of the institute and the related organizations. There is a constant high demand for phenotyping on the ultrastructural level, mainly with the mouse system and cell cultures, but also with other specimen like the naked mole rat, Drosophila and yeast. Interestingly, we had a range of projects to the ultrastructure of mitochondria over the last years, and I will highlight some of these issues and results here:

Mitochondria are rich of proteins and appear as darkly stained, compact organells - following classical plastic embedding and ize,In cellalthough cultures, single the cells inner should mitochondrial be better contrast (A, heart tissue). Following even structure is much more difficult to visual short periods (30 min !) of stress like oxy- theaccessible cells may for often the fixanssuffer from than suboptimal tissue. Cell conditionscultures are concerning always artificial the medium, systems oxygen and evengen shortage, burst, as tissue shown damage in B for or heart insufficient tissue supply, cell density, passage number and afterfixation an conditionsexperimentally they tendinduced to swellcardiac or infarction (collaboration with M. Gotthar- the “good morphology in the light micro- dt). This has to be considered carefully in others. In many cases, we could not confirm phenotyping (projects on heart and skel- tendscope” to ondie, the as ultrastructuralshown for the mitochondrial level. Cells in Morano, T. Röpke, A. Heuser, B. Schröder, E. structureculture often in MCF-7 grow cells suboptimal in E and andF (collabo finally- Klussmannetal muscle andwith J. theFielitz). groups of G. Lewin, I. ration with W. Chen, other projects with O. Rocks, F. Wanker and O. Daumke). The inner membrane structure of mito- - Recently, we have started again to analyze cal EM-preparations, not always preserved ultrathin cryosections from yeast, follow- well.chondria Cryosectioning is, even in according strongly fixedto Tokuyasu classi ing the protocols developed by Frank Vogel yield here much better results (C-H). With years ago. And we could immediately show this method, the membranes appear white, clear differences in the inner membrane as known from negative contrast, and show architecture: in some mutants, the contacts - between the outer and inner membranes sue the arrangement of the inner mitochon- were lost (G and H, collaboration with O. drialmuch membranes more details. can In normal,vary considerably, wild type tis as Daumke). shown for heart (C) and brain cortex (D), - ects with G. Lewin and E. Klussmann). and can be quantified in most cases (proj

244 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

EM-images from different projects showing mitochondria A,B: mouse heart, plastic embedding, mitochondria (m) are compact, darkly stained organells (A), which tend to swell under stress (B, induced cardiac infarction) C-H: cryosections, mouse heart (C) and brain (D), MCF-7 cell cultures, growing optimal (E) or too dense (F), yeast cells depicting the transitions between the outer and inner membranes in the wild type (G, arrowheads) which lack in mutants (H) Bar = 200 nm

Selected Publications

RFJ Marsman, CR Bezzina, F Freiberg, AO Verkerk, ME Adriaens, S Podliesna, C Chen,

Bishopric, EM Lodder, JMT de Bakker, R Fischer, R Coronel, AAM Wilde, M Gotthardt, B Purfürst, B Spallek, TT Koopmann, I Baczko, CG dos Remedios, AL George, N H - duction and arrhythmia vulnerability in the setting of myocardial ischemia. JACC 63: CA549-559 Remme. (2014) Coxsackie and adenovirus receptor (CAR) is a modifier of cardiac con

Cell differentiation mediated by co-culture of human umbilical cord blood stem cells M Stecklum, A Wulf-Goldenberg, B Purfürst, A Siegert, M Keil, K Eckert, I Fichtner. K Walentin, C Hinze1, M Werth, N Haase, S Varma, R Morell, A Aue, E Pötschke, D Warburton,with murine A hepatic Qiu, J Barasch, cells. In BVitro Purfürst, Cell Dev C Dieterich, Biol Anim.: E Popova,51(2):183-91 M Bader, (2015) R Dechend, AC Staff, ZY Yurtdas, E Kilic, KM Schmidt-Ott. A Grhl2-dependent gene network controls trophoblast branching morphogenesis. Development 142:1125-1136 (2015)

Reimann, B Purfürst, MA Dieringer, H Waiczies, M Dathe, A Pohlmann, T Niendorf. SAnchoring Waiczies, dipalmitoylS Lepore, K phosphoethanolamineSydow, S Drechsler, M-C to Ku, nanoparticles C Martin, D boostsLorenz, cellular I Schütz, up HM-

- take and Fluorine-19 magnetic resonance signal. Sci Reports 5:8427 (2015) Jß-cell Shiqi, function. A Ivanov, EMBO D Blasevics, J 34(10):1417-33 B Purfürst, (2015)W Sun, W Chen, M Poy, N Rajewsky, C Birch meier. Insm1 cooperates with Neurod1 and Foxa2 to maintain mature pancreatic

Group Leader Margit Vogel* (part-time) Dr. Bettina Purfürst Christina Schiel*

Scientists Secretariat Dr. Severine Kunz* Manuela Brandenburg* Cornelia Stärkel* Technical Assistants Marianne Vannauer* (part-time) * part of the period reported

MDC Research Report 2016 245 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: privat Photo:

Sascha Sauer

Genomics, BIMSB

Novel technologies such as next genera- NGS instruments and applications tion sequencing have revolutionized genome equipped with short-read next generation research. The Scientific Genomics Platform The Scientific Genomics Platform is currently- at the MDC focuses on development and ing two HiSeq 4000, one HiSeq 2500, three implementation of new methods and their sequencing systems from Illumina includ application in pilot studies to enable the Moreover, for long-read sequencing SMRT NextSeq 500 (Illumina), and one MiniSeq. provision of cutting-edge methodologies for a broad range of life science researchers. instruments from Pacific Biosciences such as- The platform consists of established units for an RSII and a very recently purchased Sequel system are available. So far, the RSII has for ex diverse genomics applications – particularly whole genome sequencing of organisms and forample novel been approaches used for specificfor lineage projects tracing such using as using highly parallel sequencing and associ- long-read sequencing of large PCR products. ated bioinformatic data analysis. In general, the platform has established and continuously Whole-genome sequencing provides support to cater a large spectrum preparation kit features the rapid creation of of state-of-the-art sequencing technologies gDNAThe Illumina libraries Nextera from very whole-genome low input sample library and protocols, which are relevant for systems concentration (> 50ng genomic DNA). During wide approaches to the genome, the epig- library creation, a process termed ‘tagmen- enome and the transcriptome. This includes DNA, making fragments amenable to direct methods ranging from whole genome (re) tation’ simultaneously fragments and tags sequencing, targeted genome sequencing, PCR amplification. A sequencing output of various epigenomic profiling methods to the to cover the whole genome with >20x, can be achieved~80Gb per by genome generating sample, ~200 which millions is required paired- many variants of transcriptome sequencing. end 2x100/150nt reads. A particular future focus of the group lays on single cell genomics approaches. The es- Exome sequencing The Agilent SureSelect exome enrichment tablished technology platforms are open to system is routinely used to prepare exome internal and external scientists and were very sequencing libraries. With a sequencing out- recently extended in capacity and biomedical applications through tight scientific coopera- fall within the targeted regions+/- 100bp and put of ~10Gb, an average of ~80% of all reads tion with the Berlin Institute of Health. The 20x. A sequencing output of ~10Gb per exome platform especially supported and collabo- sample~95% ofcan all be targeted achieved bases on a thoroughly are covered opti >=- rated with numerous research groups at the MDC and the Charité, as well as with other the SureSelect V5 enrichment system, gen- eratingmized workflow ~200 million using singlethe HiSeq end systems 100/150nt and external institutions such as the DKFZ and reads for each probe. Furthermore, newer the IZW. Further information can be found at: methods such as as Nextera Rapid Capture http://genomics.mdc-berlin.de Exome are being used.

246 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

RNA-seq Single cell applications A large number of varying protocols are be- The group continuously implements new ing applied to analyze RNA at large-scale. methods. Since autumn 2016 in particular RNA samples are processed in content-de- various complementary single cell genomics approaches were set up including: or paired-end libraries are created using – Fluidigm C1 fined batches, and standardized single-end- – FACS-seq (based on BD Aria III and Melody) lunina Truseq Chemistry, and never proto- colsestablished provided protocols by alternative (e.g. the vendors). optimized De Il- pending on the project, various sequencing – Dolomite’s Drop-seq system lengths and depths are being applied using – 1CellBio’s inDrop system The– 10X focus Genomics’ was so Chromium far on single system cell tran- devices for total or for RNA sequencing, respectively.the most efficientFuture development short-read will sequencing be ded- aims to extend the portfolio to DNA analy- icated to decrease the amount of required sesscriptome on the levelanalyses. of single In the cells, future as wellthe group as on RNA material, increasing throughput by ap- simultaneous analysis of (epi)genomic and propriate robotics and reducing costs. transcriptomics features.

Small RNA-seq Platform, as the predecessor of the group selection is performed for RNAs between leaderIn addition position, to providing Prof. Wei Scientific Chen, Dr. Genomics Sauer is Starting from usually 1 μg total RNA, a size- pursuing independent, third-party-funded cluding miRNAs. Subsequent library prep- research his associated Laboratory of Func- aration18 and 30steps nt tocomprise: enrich for adapter small RNAs, ligation in tional Genomics, Nutrigenomics and Sys- tems Biology. library is obtained by a further size selec- tionand PCRstep. amplification. Usually 1x50nt The final single-end small RNA is performed with a minimum of few million reads per sample, or more, depending on the research question.

ChIP-seq Selected Publications

the following library preparation steps are performed:Starting from end-repair, about 10 a-tailing, ng of ChIP-DNA,adapter li- Simandi Z, Horvath A, Wright LC, Cuaranta-Monroy I, De Luca I, Karolyi K, Sauer S, - WallmeierDeleuze JF, J,Gudas Shiratori LJ, Cowley H, Dougherty SM, Nagy GW, L. Edelbusch OCT4 Acts C, as Hjeij an Integrator R, Loges NT, of Pluripotency Menchen T, Olbrichand Signal-Induced H, Pennekamp Differentiation. P, Raidt J, Werner Mol Cell. C, Minegishi 2016 Aug K, 18;63(4):647-61. Shinohara K, Asai Y, Takaoka ment. After quality control of the sequenc- K, Lee C, Griese M, Memari Y, Durbin R, Kolb-Kokocinski A, Sauer S, Wallingford JB, inggation, library, purification up samples and a arefinal labeled PCR enrich with unique barcodes and pooled for sequenc- Docking Machinery and Primary Ciliary Dyskinesia with Left-Right Body Asymmetry ing. Usually 1x50nt single-end sequencing Hamada H, Omran H. TTC25 Deficiency Results in Defects of the Outer Dynein Arm with a minimum output of 40 million reads Randomization. Am J Hum Genet. 2016 Aug 4;99(2):460-9 for each sample is performed. Future devel- YouNeural X, Vlatkovic circular RNAs I, Babic are A, derived Will T, Epsteinfrom synaptic I, Tushev genes G, Akbalik and regulated G, Wang by M, development Glock C, opment will be dedicated to decrease the Quedenau C, Wang X, Hou J, Liu H, Sun W, Sambandan S, Chen T, Schuman EM, Chen W.

and plasticity. Nat Neurosci. 2015 Apr;18(4):603-10. - throughput by appropriate robotics and re- Wang J, Xie G, Singh M, Ghanbarian AT, Raskó T, Szvetnik A, Cai H, Besser D, Prigione A, ducingamount costs. of ChIP-DNA material, increasing Fuchs NV, Schumann GG, Chen W, Lorincz MC, Ivics Z, Hurst LD, Izsvák Z. Primate-spe cific endogenous retrovirus-driven transcription defines naive-like stem cells. Nature. 2014 Dec 18;516(7531):405-9. DNA methylation analysis TattikotaO‘Carroll D,SG, Eliasson Rathjen L, T, ZavolanMcAnulty M, SJ, Gloyn Wessels AL, Ferrer HH, Akerman J, Shalom-Feuerstein I, van de Bunt R, M, Aberdam Methylome–wide analysis was mainly per- Hausser J, Esguerra JL, Musahl A, Pandey AK, You X, Chen W, Herrera PL, Johnson PR, formed using Reduced Representation Bi- D, Poy MN. Argonaute2 mediates compensatory expansion of the pancreatic β cell. - Cell Metab. 2014 Jan 7;19(1):122-34. od was selected because DNA methylation patternsulfite Sequencing can be analyzed (RRBS). on The the RRBS nucleotide meth Head of the Group Engineer Dr. Sascha Sauer Claudia Quedenau - Wei Chen until July 2016 tions.level in Moreover CpG island this and method CpGI is shores feasible with with a Sascha Sauer since August 2016 Technical Assistants acoverage very low of amount around of 10 DNA % of for all sequencing CpG posi Madlen Sohn a minimum of 50 million reads for each Scientist samplewith of 100 nt lenghts. Recent de- Cornelius Fischer (part time) Part-time Secretary velopments were dedicated to apply Whole Maria Metsger(part time) Sabrina Deter

Genome Bisulfite Sequencing (WGBS). MDC Research Report 2016 247 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Thoralf Niendorf

Magnetic Resonance (B.U.F.F.)

B.U.F.F. builds on collaborations with our partners from the Charité, the Leibniz Institute for Molecu- lar Pharmacology (FMP), the German Metrol- ogy Institute (PTB), Siemens Healthcare, Bruker Biospin and other (inter)national institutions. To learn more about the mission of B.U.F.F. please visit https://www.mdc-berlin.de/BUFF.

Equipment and research infrastructure B.U.F.F. provides state-of-the-art infrastructure for interdisciplinary imaging projects and is equipped with small animal (9.4 T) and - ners (Figure 1) Figure 2: Overview of the volunteer/patient preparation whole-bodylaboratory and human supports (7.0 T small & 3.0 scaleT) MR animal scan areas (top left), the MR hardware development laborato- housing. To foster. B.U.F.F. translational runs an research S1 certified and ries (top right) and the clinical imaging setup (bottom clinical studies B.U.F.F. is equipped with re- left) available at B.U.F.F. bottom right: Members of B.U.F.F. ception areas and changing rooms to comfort congratulating the 1000th volunteer of the population im- and guide volunteers and patients (Figure aging study of the national cohort (NAKO) for supporting 2, left). All technical prerequisites needed epidemiological and medical research. for clinical studies are setup including emergency equipment and extra patient moni toring units. Members of B.U.F.F. have access Research focus to infrastructure for hardware and MR pulse Research at B.U.F.F. concentrates on the devel- sequence development. (Figure 2, right). opment of enabling MR technology through

Figure 1: Picture photographs illustrating the of the UHF-MR (1-4). The focus is on new ways of whole-body human (7.0 T & 3.0 T) and small animal (9.4 T) mappingThoralf Niendorf’sanatomy, morphology, group on experimental microstruc- MR scanner environments at B.U.F.F. ture, function, physiology and metabolism. These efforts are designed to characterize (patho)physiological and biophysical processes to promote transfer from basic research to (pre)clinical studies and vice versa. To highlight the anatomical detail and spatial Figure 3 sur- veys examples of in vivo mice phenotyping andresolution from in capabilities vivo human of MRIimaging. The hard work of B.U.F.F. members resulted in 65 peer- reviewed publications during 2014-16, 70+ abstract contributions to the 2014/15/16

head of B.U.F.F. (Thoralf Niendorf) recently receivedISMRM meetingsan ERC Advanced plus one Grant US patent. by the TheEu- ropean Research Council.

248 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Figure 3: Illustration of the anatomical detail provided by non-invasice MRI.Top: In vivo MRI of the mice heart (resolution 50 µm), of the mice brain (resolution 30 µm), of the vessel tree of the mice brain (resolution 30 µm) and of the mice kidney (resolution 30 µm).Bottom: In vivo MRI of the human heart (resolution 1.0 mm), of the human brain (resolution 0.5 mm), of the vessel tree of the human brain (0.5 mm) and of the human liver (1.0 mm) obtained at 7.0 T.

Contributions to major research Public engagement initiatives initiatives B.U.F.F. actively contributes to public engagement Members of B.U.F.F. form a key/integral part with open tours to the ‘Lange Nacht der Wissen- of major research initiatives and support the schaften’ (Long Night of Science). B.U.F.F. is also engaged into disseminating its research to the public by supporting the Life Science Learning Lab Helmholtz Alliance for ImagingHemodynamics and Curing Environmentalof Acute Kidney DiseasesInjury, the (ICEMED), imaging program the DFG fruitful partnership with the Robert-Havemann researchof the German group Center FOR 1368 for Cardiovascularon Re- Gymnasium,(German: “Gläsernes Berlin which Labor”). we B.U.F.F. support runs through a very search (DZHK), the individualized Medicine school presentations, on-site visits and lab-tours. (iMED) initiative of the Helmholtz Associa- For further details about B.U.F.F.s public engage- tion, the BMBF project on targeted radiofre- ment please visit www.mdc-berlin.de/BUFF. quency heating (THERAHEAT), the BMBF funded validation project on the diagnosis of Basic research and clinical science renal diseases (renalMROXY), the German Ul- opportunities trahigh Field Imaging EU The B.U.F.F. team is always looking for collabora- COST Action PARENCHIMA on renal imaging, tors committed to the value and power of non- the DFG funded project (GUFI) on Fluorine network, MR the tech - invasive imaging. We are ready to enter into fair nologies to study cellular therapies, the Marie collaborative enterprises aiming at moving the Sklodowska Curie project SODIUMMRI-4-EU gradu- attracting extramural funding with the goal to ate school on quantitative imaging fosterneedle multi-disciplinary by first class imaging and translational sciences and re by- and(EU projectthe European 752489), Ultrahigh-Field the DFG funded Imaging search at the MDC. Further details please visit Network in Neurodegenerative Diseases(BIOQIC) (EU- www.mdc-berlin.de/BUFF or contact the head of B.U.F.F. (Thoralf Niendorf, thoralf.niendorf@mdc- berlin.de, +49 30 9406 4505). FIND). B.U.F.F. is part of the imaging facilities Germanof the Berlin National Institute ofCohort Health (BIH).at B.U.F.F. Selected Publications B.U.F.F. is home of the Berlin site of the imag- Niendorf T, Paul K, Oezerdem C, Graessl A, Klix S, Huelnhagen T, Hezel F, Rieger J, ing program (5) of the German National Co- Waiczies H, Frahm J, Nagel A, Oberacker E, Winter L. W(h)ither Human Cardiac and hort (NAKO) which accommodates 30.000 Body Magnetic Resonance at Ultrahigh Fields? Technical Advances, Practical Consid- volunteers to be included in a comprehensive MR examination. 6000 of these volunteers are Wintererations, L, Applications, Oberacker E, andOzerdem Clinical C, JiOpportunities Y, von Knobelsdorff-Brenkenhoff NMR Biomed 2015, DOI:F, Weidemann 10.1002/ examined at B.U.F.F. This team recently cel- nbm.3268 [epub ahead of print]. ebrated the examination of the 3000th volun- G, Ittermann B, Seifert F, Niendorf T. On the RF heating of coronary stents at 7.0 teer . Niendorf T, Pohlmann A, Reimann HM, Waiczies H, Peper E, Huelnhagen T, Seeliger (Figure 3) E,Tesla Schreiber MRI. Magn A, Kettritz Reson R,Med Strobel 2015;74(4):999-1010. K, Ku MC, Waiczies S. Advancing Cardiovascular, - Teaching and education initiatives Members of B.U.F.F. are chairing and orga- NiendorfNeurovascular, T, Pohlmann and Renal A, Arakelyan Magnetic K,Resonance Flemming Imaging B, Cantow in Small K, Hentschel Rodents J, Using Grosenick Cryo D,genic Ladwig Radiofrequency M, Reimann CoilH, Klix Technology. S, Waiczies Front S, Seeliger Pharmacol E. How 2015;6:255. bold is blood oxygen- nizing the Annual Symposium of Ultrahigh ation level-dependent (BOLD) magnetic resonance imaging of the kidney? Opportu- Field Magnetic Resonance: Clinical Needs, Research Promises and Technical Solutions Bamberg F, Kauczor HU, Weckbach S, Schlett CL, Forsting M, Ladd SC, Greiser KH, Webernities, challenges MA, Schulz-Menger and future J, directions.Niendorf T, Acta Pischon Physiol T, Caspers (Oxf) 2015;213(1):19-38. S, Amunts K, Berger K, which attracts about 250 MR technology Bulow R, Hosten N, Hegenscheid K, Kroncke T, Linseisen J, Gunther M, Hirsch JG, leaders and distinguished clinical experts, all Kohn A, Hendel T, Wichmann HE, Schmidt B, Jockel KH, Hoffmann W, Kaaks R, Reiser bridging disciplinary boundaries. On behalf of Design, and Technical Background. Radiology 2015:142242. the European Society for Magnetic Resonance MF, Volzke H. Whole-Body MR Imaging in the German National Cohort: Rationale, in Medicine and Biology members of B.U.F.F. chaired and organized a one week workshop Group Leader Secretary RF coil design: Build your own which is fully Prof. Dr. Thoralf Niendorf Rosita Knispel (over)booked every year.

MDC Research Report 2016 249 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: Christian Kruppa/BIH Christian Photo:

Jennifer Kirwan

Metabolomics, BIH Platform at MDC

The Berlin Institute for Health Metabolomics • Thermo Quantiva: a triple quadrupole mass Facility undertakes metabolomics research spectrometer allows the measurement of with clear translational medicine potential. speeds make it particularly useful for tar- A dedicated and highly skilled team, we col- getedsmall moleculesanalysis and and it lipids.can be Its combined fast scanning with laborate with researchers and clinicians at liquid chromatography for prior separation the Charité and MDC to advance medical of complex biological matrices. • Thermo Q-Exactive (based on Orbitrap knowledge, especially the interplay between technology): a high resolution, highly heart and brain health. mass accurate Fourier transformation mass spectrometer that is used for lipi- domics and metabolomics analyses We are equipped with state-of-the-art, high • Agilent 1290 two-dimensional liquid resolution, highly mass accurate mass spec- chromatography system: for detection, trometry instrumentation enabling both - targeted and non targeted metabolomics ecules with similar chemical properties: to be undertaken. An automated sample especiallyidentification useful and for lipids separation of mol preparation robot and a separate derivati- • sation robot enables robust and reproduc- Spectrometer: an ion mobility technology ible epidemiological scale analyses to be thatOwlstone enables Field separation Asymmetric of Ioncompounds Mobility undertaken. based on cross sectional area to enable improved analysis by mass spectrometry Technology and applications of mixed isomeric species. Metabolomics aims for a global analysis of • To support epidemiological size metabo- all metabolites within a cell or biological lomics studies we have developed a ro- system (for example, blood, urine, cells, botic system to facilitate the preparation saliva etc) and measures the perturbations of the required standard mixtures for that occur in disease states or other chang- metabolomics analyses and to perform es in conditions e.g. nutrition, exercise, the extraction and sample preparation different genotypes etc. A variety of mass procedures in an automated way. This ro- spectrometers and related instrumentation is used to enable to enable a wide range of lab automation, as it includes the han- molecules to be measured, often at widely dlingbotic systemof solid is substances a significant and step liquids towards in variable concentrations and with very different physicochemical properties. a single system specifically designed for Facilities include: Collaborationsour workflows. • Leco GCxGC-Electron impact-Time of We are currently collaborating with several Flight mass spectrometer: a gas chroma- different research groups, both local and tography based system that allows the international. Examples of projects we are measurement of small molecules and is currently involved in include: the metabo- especially useful for polar molecules. A lomics changes seen in Alzheimer models previously validated method enabling a as the disease develops, the metabolomics quantitative, targeted analysis of up to of depressive illness and the lipid changes associated with heart disease. We concen- involved in central carbon metabolism is trate on projects that have relevance to im- possiblenearly one on hundredthis system. defined metabolites

proving people’s health or quality of life. 250 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Figure 1: The Berlin Institute for Health Metabolomics Facility is composed of a range of state-of-the-art technolo- Recognising the importance of the “value gies which are used in combination to globally map the metabolome (the range of metabolites in a biological system) work closely with other -omic platforms to and assess how the metabolome is altered by disease states. improveadded” approachthe biological of systems information biology, we gain we from any individual study. This is particularly notable with our close collaboration with the Bioinformatics Core Facility where strategic cooperation to embed staff members in each group will enable the transfer of knowledge between both facilities.

External to the MDC, we support the development of precision medicine to improve medical treatment and prognosis. The Head of the Facility, Dr Jennifer Kirwan is a member of the Working Party in Precision Medicine and Pharmacogenomics for the international Metabolomics Society.

Method development Metabolomics is still a relatively young science and methods are constantly being Selected Publications developed to enlarge the range and extent The Head of the Group took over only two months before the end of this reporting of the analyses we can perform. Over the period, but we anticipate new publications will follow very soon! next few years, we are working closely with other groups to develop and validate new - Start of the Group: October 2016 lems. The majority of these methods will havemethods applications to answer to specificother projects research thus prob en- Group Leader abling a rapid expansion of the abilities of Dr. Jennifer Kirwan the Metabolomics Facility.

MDC Research Report 2016 251 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Andrew Woehler

Light Microscopy Platform, BIMSB

Light microscopy has long been an essential technology, the platform will work towards technology for the progress of life sciences microscopy-based solutions for systems biologists that include automated high- research. Advances in this technology have throughput microscopy, as well as devel- kept pace with the demands of modern inves- opment of computational methods for the tigation. We now have the ability to monitor analysis and management of large biologi- the structure and function of living systems cal imaging data sets. at scales that vary by orders of magnitude. Research Focus Modern light microscopy allows us to monitor the development of whole organisms, the In addition to providing support for MDC structure and function of single cells, as well continuing to development microscopy- basedand BIMSB approaches research to projects,study the our regulatory group is as the localization and dynamics of single mechanisms synaptic plasticity. The abil- molecules. ity of the neuronal synapse to weaken or strengthen over time is key to information processing and storage in the CNS. Changes was recently established to support re- searchThe BIMSB projects Light for MDC Microscopy scientists Platform at the that occur on the millisecond to second time scale,in the known efficiency as short of synaptic term plasticity transmission (STP), The new facility will host several high- are regulated by a number of processes at new BIMSB research center in Berlin Mitte. the axon terminals (or presynaptic boutons). live-cell, light sheet, and super resolution Here, upon arrival of an action potential, microscopes.end instruments We includingwill also provide confocal, expert TIRF, synaptic vesicles (SVs) fuse with the plasma support for project planning, sample prep- membrane releasing neurotransmitters aration, data analysis and visualization as into the synaptic cleft where they bind to post synaptic receptors to propagate signal. Vesicular membrane and protein contents additionwell as trainingto hosting for core advance light fluorescencemicroscopy methods such as FRET, FLIM, FRAP, FCS. In are then endocytosed, recycled, and refilled

Figure 1. Electrochemical gradient measured at the single vesicle level. A) A schematic representation of labeled SVs imaged using a custom-built wide-field UV- flash uncaging / TIRF microscope. B) SV phenotyping is performed by on-stage immuno-labeling against VGLUT1 or VGAT. C) Averaged spH response to NEP-ATP uncaging is shown with a representative image of spH-SVs. D) Averaged VF.2.1.Cl response and a representative image of VF2.1.Cl-labeled SVs are presented.

252 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Figure 2. Imaging synaptic release at hippocampal boutons. A) Primary neurons expressing a glutamate sensor were imaged, during which 15 APs were elicited at 10 Hz. Pseudo-color map represents synaptic release. B) Traces from repeated stimuli sampled from a single synapse are plotted in gray, the mean in black, and the deconvoled amplitudes below. C) PPR is mapped back to the respective synapse. D) Analysis of spatial correlation reveals a PPR length constant of 4.1 µm.

with neurotransmitter before they are able postsynaptic response, careful investiga- to again contribute to synaptic transmis- tion of STP requires precise control of single sion. We are currently investigating several presynaptic boutons. This has limited such investigations to large synapses such as the

assteps well in as thisthe priming cycle including before fusion. acidification boutons. Optical approaches utilizing cal- and refilling of SVs with neurotransmitters ciumcalyx andof held pH andsensitive hippocampal reporters mossy have beenfiber Regulation of the SV proton developed to help address these limitations electrochemical gradient - temporal kinetics to monitor STP. We have successfullyhowever do developednot provide an sufficient approach SNR using nor a ofIn neurotransmitterorder to maintain moleculesrobust synaptic within func sec- genetically encoded glutamate sensor to di- ondstion, SVsafter must endocytosis. be refilled This with process thousands is fu- rectly monitor the release of glutamate into eled by a proton electrochemical gradient the synaptic cleft (Fig 2). This approach al- generated across the SV membrane by the - of synapses simultaneously to characterize mitter molecules that carry different net propertieslows us to monitorsuch as pairedstimulus pulse trains facilitation at 100’s vacuolar ATPase. It is unclear how trans (Taschenberger et al. 2016). We are con- a common mechanism while maintaining tinuing the development of this and other chargecharges neutrality can be efficiently and osmotic sequestered balance. We by microscopy-based approached to offer developed an approach to study the gen- greater insight into the molecular machin- eration and maintenance of proton electro- ery and signaling processes at play during chemical gradient in single SVs. By imaging STP at the many synapses inaccessible by electrophysiology-based techniques. - tionpurified we are SVs able labeled to reliably with either characterize luminal changespHluorin in or pH VF2.1.Cl or membrane under TIRF potential, illumina re- Selected Publications - Farsi Z, Woehler A - Synapse Using FRET-Based Biosensors. Methods Mol Biol. cagespectively NPE(caged)-ATP (Fig 1). By couplingand measure in epiflores the fast Taschenberger H, Woehler. Imaging A Activity-Dependent, Neher E. Superpriming Signaling of synaptic Dynamics vesicles at the as Neuronal a com- cent UV excitation we are able to flash un- mon basis for intersynapse variability and modulation of synaptic2017;1538:261-275. strength. Proc Natl tential generation in these diffraction lim- Acad Sci U S A kinetics of acidification and membrane po Farsi Z, Preobraschenski J, van den Bogaart G, Riedel D, Jahn R, Woehler A. Single- ited organelles. Furthermore we perform vesicle imaging. 2016 reveals Aug different 2;113(31):E4548-57. transport mechanisms between glutamatergic and on stage immuno-labeling to characterize GABAergic vesicles. Science the molecular composition of the SVs mea- Woehler A, Lin KH, Neher E. Calcium-buffering effects of gluconate and nucleo- sured (i.e. Glutamaterigics vs GABAergic) to . 2016 Feb 26;351(6276):981-4. J Physiol. 2014 Nov study how the presence and abundance of Huatides, Y, as Woehler determined A, Kahms by a novel M, Haucke fluorimetric V, Neher titration E, Klingauf method. J. Blocking endocytosis different SV constituents affect the proton enhances15;592(22):4863-75. short-term synaptic depression under conditions of normal availability of electrochemical gradient that drives neu- vesicles. Neuron

. 2013 Oct 16;80(2):343-9.

Opticalrotransmitter readout refilling for (Farsi synaptic et al. 2016). Start of the Group: May 2016 Scientists plasticity Dr. Zohreh Farsi Glutamatergic synapses show large varia- Group Leader Marie Walde tions in strength and short-term plasticity Dr. Andrew Woehler (STP). The current state of the art approach Technical Assistant to studying STP relies on patch clamp-based Agnieszka Klementowicz

electrophysiology. In addition to recording MDC Research Report 2016 253 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: Peter Himsel/MDC Peter Photo:

Arnd Heuser

Pathophysiology

The Pathophysiology Platform provides Understanding the complex physiological MDC researchers with a central resource interactions between the cardiovascular, respiratory and central nervous systems is for the study of animal disease models. We critical for the development of many human use validated approaches and state-of-the- pharmaceuticals and therapies. We focus on art instrumentation that allow for sensitive techniques which minimize animal use and screening of phenotypic variations. Despite enable researchers to obtain comparable levels of information from fewer animals or numerous advances in research techniques to obtain more information from the same that avoid the use of animals, such as in number of animals, thereby reducing future vitro methods and advanced computer use of animals. As such, we use a variety of non-invasive in vivo test tools that allow modelling, there is still the need for a whole monitoring of physiological and bioelec- system approach to preclinical research trical variables (e.g. blood pressure, heart and for data from conscious and uncon- rate, ECG and respiration) in conscious scious animals. animals. Non-invasive, whole body imaging of small animals using techniques such as High-Frequency Ultrasound and Time- Domain-Nuclear-Magnetic-Resonance imaging is helping to reduce the number of animals used in basic research and drug development. We continuously try to improve

minimize actual or potential pain, suffering, distressscientific or procedures lasting harm and husbandryand/or improve which animal welfare in situations where the use of animals is unavoidable.

High-Frequency Ultrasound Ultrasound is a powerful imaging modality that enables non-invasive, real-time visualization of organs and tissues. We perform sonographic studies of small animals using

transducers produce frequencies as high as a High-Frequency Ultrasound system. Its- tion of structures as small as 30µm with a maximum70MHz, which temporal has permitted resolution the ofidentifica 740fps. This makes the platform ideal for pre-clin- Veterinary Blood Tests ical research because it permits a study of The Pathophysiology Platform can assist research investi- many organs, tissues, and the behavior of gators who would benefit from the availability of services biological systems that need to be studied that replicates for animals the laboratory tests that are at this tiny scale. The non-invasive nature of routine for humans. Examples include blood chemistry, ultrasound means that it can be used with urinalysis, and hematology. no harm to animal or human subjects. We

254 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

The Methodological Spectrum of the Pathophysiology Platform. High Frequency Ultrasound Imaging:(A) Parasternal Long Axis View of a Murine Heart,(B) Micro- Ultrasound Imaging of a Murine Em- bryo at E12.5,(C) Contrast Imaging of a Melanoma Tumor,(D) 3D-Color Doppler of a Murine Testicle.

have established a wide variety of proto- times of spin rearrangements are recorded cols to analyze the phenotypes of mice and rats: For cardiovascular studies, ultrasound relaxation patterns can be correlated to the imaging can be used to detect the presence densitywhen the of magneticbody compounds, field is removed. i.e. lean tissue,These of localized or generalized hypertrophy or thinning of the myocardium of the left ven- their proportional composition. The NMR- tricle (LV) and the presence of regional or whole-bodyfat and free fluids,composition giving concreteanalyzer valuesallows of a global wall motion abnormalities associated with systolic dysfunction. The analysis whole body components of small animals. - non-invasive and reliable quantification of - atedof transmitral with LV diastolic flow profiles dysfunction. allows the Strain de Selected Publications analysistection of provides abnormal a highlyfilling patternssensitive associspeck- Embryonic cardiomyocytes can orchestrate various cell protective mecha- le-tracking based echocardiographic imag- nisms to survive mitochondrial stress. Magarin M, Pohl T, Lill A, Schulz H, ing technique that offers quantitation of the Blaschke F, Heuser A, Thierfelder L, Donath S, Drenckhahn JD. J Mol Cell Cardiol. velocity of the walls, displacement, strain, 2016 Aug;97:1-14 strain rate, and time to peak analysis. Be- CD74-Downregulation of Placental Macrophage-Trophoblastic Interactions in Preeclampsia. Przybyl L, Haase N, Golic M, Rugor J, Solano ME, Arck PC, Gauster cause of its non-invasive nature ultrasound M, Huppertz B, Emontzpohl C, Stoppe C, Bernhagen J, Leng L, Bucala R, Schulz H, can be used in longitudinal studies for tu- Heuser A, Weedon-Fekjær MS, Johnsen GM, Peetz D, Luft FC, Staff AC, Müller DN, mor diagnostics and therapy monitoring. Dechend R, Herse F. Circ Res. 2016 Jun 24;119(1):55-68. Muscle RING-finger 2 and 3 maintain striated-muscle structure and function. J 3D-Ultrasound imaging allows the reli- Cachexia Sarcopenia Muscle. Lodka D, Pahuja A, Geers-Knörr C, Scheibe RJ, Nowak able delineation of tumor boundaries and - ser A, Kraft T, Bassel-Duby R, Olson EN, Dittmar G, Sommer T, Fielitz J. J Cachexia SarcopeniaM, Hamati J, KöhnckeMuscle. 2016 C, Purfürst May;7(2) B, Kanashova T, Schmidt S, Glass DJ, Morano I, Heu addition, ultrasound imaging is ideal for Impaired myocardial development resulting in neonatal cardiac hypoplasia al- studyingthe assessment pregnancy of tumor and heterogeneity.development Inin ters postnatal growth and stress response in the heart. Drenckhahn JD, Strasen mice and rats. J, Heinecke K, Langner P, Yin KV, Skole F, Hennig M, Spallek B, Fischer R, Blaschke F, Heuser A, Cox TC, Black MJ, Thierfelder L. Cardiovasc Res. 2015 Apr Bcl10 mediates angiotensin II-induced cardiac damage and electrical remodel- Body Composition Analysis ing. Time-Domain-Nuclear-Magnetic-Reso- nance imaging (TD-NMR) is a powerful Lucas Markó PC, McAllister-Lucas L, Henke N, Park LM, JK, LuftSpallek FC, B,Dechend Qadri F, R, Balogh Müller A, DN. Apel Hypertension. IJ, Oravecz-Wilson 2014 Nov;64(5):1032-9.KI, Choi M, Przybyl L, Binger KJ, Haase N, Wilck N, Heuser A, Fokuhl V, Ruland J, solids and liquids in objects to be studied. Withouttool for theanesthesia non-invasive or other quantification preparations, of Group Leader animals such as mice and rats can be tested Dr. Arnd Heuser - ids and lean tissue within a few minutes. Technical Assistants Secretariat Afterfor body a stimulated composition adjustment of fat, free of body nuclear flu Stefanie Schelenz Cornelia Stärkel particle spins within a high energetic mag- Martin Taube

netic field, specific patterns of relaxation MDC Research Report 2016 255 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Sebastian Diecke

Pluripotent Stem Cells

This figure summerizes the services offered by the stem cell core facility.

256 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

The Pluripotent Stem Cell Core Facility at the Max Delbrück Center is a joint project of mediated gene targeting. cell lines using CRISPR/CAS9- or TALEN- in Berlin Buch. The facility was established With the German Stem Cell Network (GSCN) inthe September Berlin Institute 2014 ofand Health is located and the on MDC the and others, we have initiated a network of stem cell core facilities for iPS derivation 2014 the laboratories, including S1 and S2 across Germany. Here the goal is to estab- cellfirst culturefloor of labs, the Walterwere set Friedrich up and Haus.special In- lish common standards through the sharing ized equipment was purchased to provide of cell lines, tools and protocols which will

edge stem cell technologies. These include across different experimental frameworks. aMDC/BIH comprehensive scientists range with of the methods latest, cutting- for cell Welead are to higherconvinced efficiency that this and will comparability fundamen- cultures and molecular biology that are - required to derive, maintain, characterize nity across Germany. and differentiate human induced pluripo- tally benefit the stem cell research commu tent stem cells (hiPSC). For example, as a component to ensure quality control for Shown are human induced newly derived iPSCs, an iScan System from pluripotent stem cell derived - cardiomyocytes. The structure lel analysis of chromosome stability and a of the sarcomeres were visual- Illumina was purchased to allow the paral- ized by immunofluorescence acterization, a QuantStudio 6 real-time PCR staining using antibodies confirmation of their identity. For cell char- against cardiac markers cence microscopes are available. Further- alpha actinin (red) and more,system, to a carrydigital out PCR genetic machine engineering and fluores on cardiac troponin (green). the cells, transfection devices (Neon and The nucleus was stained Amaxa, Nucleofection) were purchased. with DAPI (blue). The cell culture labs of the core facility are equipped with class 2 safety cabinets and a special picking hood. This hood is essential for many of the steps involved in manual cell manipulations, such as selecting clones after reprogramming, the genetic manipulation of hiPSCs, or the removal of cells that have spontaneously differentiated.

Selected Publications the facility provides hands-on training in Diecke S basicIn addition stem cell to setting culture up techniques the infrastructure, for scien- M, Shrestha R, Wu H, Ma N, Shao NY, Seeger T, Woo NA, Wilson KD, Matsa E, Porteus tists who are planning to work with hPSCs. KarakikesMH, Sebastiano I, Termglinchan V, Wu JC. A V,Comprehensive Cepeda DA, Lee TALEN-Based J, , Hendel Knockout A, Itzhaki Library I, forAmeen

- - sisting of theoretical and practical sections Generating Human Induced Pluripotent Stem Cell-Based Models for Cardiovascular wereIn 2014-2016 offered and five successfully stem cell conducted. courses con DieckeDiseases. S, CircLu J, Res. Lee 2017J, Termglinchan Feb 28. pii: V, CIRCRESAHA.116.309948. Kooreman NG, Burridge PW, doi: Ebert 10.1161/CIR AD, Churko JM,CRESAHA.116.309948. Sharma A, Kay MA, Wu PMID:28246128 JC. Novel codon-optimized mini-intronic plasmid for The services of the core facility are summa- - Dieckeefficient, S ,inexpensive, Jung SM, Lee and J, Ju xeno-free JH. Recent induction technological of pluripotency. updates and Sci clinical Rep. 2015 applica Jan- tion of primary cells and the reprogram- 28;5:8081. doi: 10.1038/srep08081. ming,rized indifferentiation figure one and and include genome the editingextrac Burridgetions of induced PW, Matsa pluripotent E, Shukla stem P, Lin cells. ZC, ChurkoKorean JM,J Intern Ebert Med. AD, 2014Lan F, Sep;29(5):547- Diecke S, Huber of human pluripotent cells. Additionally we 57. doi: 10.3904/kjim.2014.29.5.547. Epub 2014 Aug 28. help and guide scientists who perform their own experiments under our supervision. B, Mordwinkin NM, Plews JR, Abilez OJ, Cui B, Gold JD, Wu JC. Chemically defined Redmergeneration T* of, Diecke human S* cardiomyocytes., Grigoryan T, Quiroga-Negreira Nat Methods. 2014 A, Birchmeier Aug;11(8):855-60. W, Besser doi: D. 10.1038/nmeth.2999. Epub 2014 Jun 15. Recently, the core facilities established a and can replace OCT4 during somatic cell reprogramming. EMBO Rep. 2011 Jul feeder-free cultivation system of hiPSC us- (*Shared co-first authors) E-cadherin is crucial for embryonic stem cell pluripotency - 1;12(7):720-6. doi: 10.1038/embor.2011.88. tions including homemade, chemically de- ing a range of chemically defined condi Start of the Group: September 2014 Scientists neuronal, endothelial and cardiac differen- Narasimha Swamy Telugu tiationfined media. protocols Moreover, in close we cooperation have established with Group Leader several research groups on the campus. The Dr. Sebastian Diecke Technical Assistants core facility also successfully performed (Joint with BIH) initial genome engineering experiments Norman Krüger - SandraAnna Iwanska Wegener eral genes and the generation of reporter including specific knock-out studies of sev

MDC Research Report 2016 257 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Hans-Peter Rahn

Preparative Flowcytometry

Fluorescence-activated cell sorting (FACS) is a specialized type of flow cytometry. It primary murine hematopoietic cells as well richment of specific subsets of tumor and was designed in 1971 by Len Herzenberg of various cell types in order to analyse the to provide a fast method for physical sepa- biologicalas of genetically functions modified of candidate subpopulations genes in ration of heterogeneous mixtures of cells particular with relevance to lymphoma- into distinct subsets based on their light genesis. A recent example of such a work which required cell sorting experiments of scattering and fluorescent characteristics. various cell types is the characterization of The FACS Core Facility at the MDC mainly provides assistance in operating such cell with NF-kB as key regulators in classical Hodgkintranscription lymphoma factors (Kreher, IRF5 in Bouhlel combination et al. sorters. These sorters can isolate up to 2014). 20,000 cells per second for large-scale cell isolation and allow single cells to be pre- Similarly, the group of Klaus Rajewsky gen- cisely spotted into microtiter plates or onto erated various mouse models of lymphomas and fatal lymphoproliferative disor- slides for analytical purposes. In addition, the staff of the FACS Core Facility assists in the underlying mechanisms of pathogen- the design and setup of more specialized esis.ders Theirand uses work flow relies cytometry on genetically to elucidate modi- or sophisticated experiments that use ad- - ditional state of the art analytical flow cy- tion,fied mouse they enhanced strains, for genome instance editing conditional using tometers. The facility operates one 3-colour loss- or gain-of-function mutants. In addi laser line analyzer (BD LSR), one 5-colour Cas9-mediatedthe CRISPR/Cas9 mutations technology or in homologous mammalian laser line analyzer (BD Fortessa), and four recombinationcells to increases events the efficiency (Chu, Weber of CRISPR/ et al. high-capacity digital cell sorters (BD Aria). 2015).

Stromal cells are pivotal for the onset and survival of lymphoma B cells, but which Sorted cells are routinely prepared for gene stromal cell subpopulations in the lymph expression analysis by microarrays, RNA nodes contribute to the formation of a lym- sequencing, quantitative real-time PCR, phoma growth-promoting niche had re- DNA sequencing, live-cell imaging or adop- mained elusive. The groups of Armin Rehm tive transfer experiments in animals. Nu- and Uta E. Höpken have applied a combination of cell enrichment using magnetic bead and the FMP have used the facility over the and FACS to isolate very rare stromal cell lastmerous few groupsyears for from such the purposes. MDC, BIMSB, ECRC populations of mesenchymal, myeloid and endothelial origin and analyzed their gene The greatest number of requests for all pre- expression on a genome wide scale. This parative cell sorting at the MDC arise from was a prerequisite for the analysis of spe- primary human and murine cells, which - are characterized by their immunopheno- typic parameters. For example, the group andcific other tumor-induced lymphoma-accessory gene expression cell types pro of Bernd Dörken routinely uses FACS en- (Rehm,files in stroma-located Gatjen et al. 2014). monocytic, dendritic

258 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

FACS cannot only be used for the analysis of the haematopoietic lineage, but also gained in importance for other cells populations. Reliable cell surface biomarkers are becoming available for such cell populations, and they also can be labeled by transgenic technologies. For example, the group of Walter Birchmeier uses FACS to isolate and analyze stem cells from normal tissues and from tumors. They isolated stem cells in the skin or the mammary gland, as well as rare cell populations present during heart development, or cancer stem cells that represent rare cell types present in tumors (Holland, Gyorffy et al. 2013).

Stem cells called Satellite cells also exist in the adult muscle. Satellite cells are needed for muscle growth and for the regeneration of injured muscle tissue. Although satellite cells account for only a small fraction of the muscle tissue, they can be isolated and enriched by FACS. Thus, changes in gene expression in these stem cells can be assessed in mouse mutants, and the isolated cells can be cultured for a more detailed characterization. The group of Carmen Birchmeier sorts satellite cells to investigate how signaling pathways control their development and function (Brohl, Vasyutina et al. 2012). Left: Inside view of a FACSAria (flow cell with sort block Another use of FACS technology is the gen- and deflection plates), Right: Schematic view of the separation principle pluripotent stem (iPS) cells. For example, theeration Pluripotent and genetic Stem modification Cell Core Facility of induced generates human pluripotent stem cells that Selected Publications - Brohl, D., E. Vasyutina, M. T. Czajkowski, J. Griger, C. Rassek, H. P. Rahn, B. Purfurst, H. ing a disease-causing mutation, and these Wende and C. Birchmeier (2012). “Colonization of the satellite cell niche by skeletal cellsare modifiedare enriched by introducing using FACS. or They correct are 23 Chu, V. T., T. Weber, B. Wefers, W. Wurst, S. Sander, K. Rajewsky and R. Kuhn (2015). then used by the groups of Norbert Hüb- muscle progenitor cells depends on Notch signals.” Dev Cell (3): 469-481. ner, Thomas Willnow or Erich Wanker to 33 - Holland,“Increasing J. D., the B. efficiencyGyorffy, R. of Vogel, homology-directed K. Eckert, G. Valenti, repair L. for Fang, CRISPR-Cas9-induced P. Lohneis, S. Elez- derlie diseases. Moreover, these cells can kurtaj,precise U. gene Ziebold editing and in W. mammalian Birchmeier cells.” (2013). Nat “Combined Biotechnol Wnt/beta-catenin, (5): 543-548. Met, bedefine tagged the withmolecular various mechanisms reporters encoding that un 5(5): 1214-1227. Kreher,and CXCL12/CXCR4 S., M. A. Bouhlel, signals P. Cauchy, characterize B. Lamprecht, basal breast S. Li, cancer M. Grau, and F. Hummel,predict disease K. Ko- that are formed when these cell types dif- outcome.” Cell Rep ferentiate.fluorescent Theproteins groups to enrich of Simone for cell Spuler types chert,Cockerill, I. Anagnostopoulos, M. Janz, B. Dorken, K. C.Johrens, Bonifer M. and Hummel, S. Mathas J. Hiscott, (2014). S. “Mapping S. Wenzel, of P. transcrip Lenz, M. - Schneider, R. Kuppers, C. Scheidereit, M. Giefing, R. Siebert, K. Rajewsky, G. Lenz, P. N. study the mechanisms underlying cell fate 111(42): E4513-4522. decisions.and Zsuzsanna Izsvak use this approach to Rehm,tion factor A., M. motifs Gatjen, in K.active Gerlach, chromatin F. Scholz, identifies A. Mensen, IRF5 M. as Gloger,key regulator K. Heinig, in classical B. Lam- precht,Hodgkin S. lymphoma.” Mathas, V. Begay, Proc Natl A. Leutz, Acad M. Sci Lipp, U S A B. Dorken and U. E. Hopken (2014). “Dendritic cell-mediated survival signals in Emu-Myc B-cell lymphoma depend on The group of Nikolaus Rajewsky devised 5: 5057. a method to collect large amounts of precisely staged C. elegans embryos by FACS the transcription factor C/EBPbeta.” Nat Commun which are then used for high-throughput Group Leader genomic and biochemical analyses. This Dr. Hans-Peter Rahn new method already contributed towards a more complete understanding of gene Technical Assistants Secretariat regulatory networks during early C. elegans Kirstin Rautenberg Cornelia Stärkel development.

MDC Research Report 2016 259 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Currently N.N.

Proteomics

Cells communicate with their environment concentration of these players, avoiding the and react to different stimuli and changes in sequencing of unrelated, not regulated proteins in the cell. A method that allows the the environment. These reactions can either selection of a limited number of proteins is be based on transcriptional activity or chemi- selected reaction monitoring (SRM). This cal modifications of existing proteins, so technique has the advantage of a high sensi- called post-translational modifications. In or- tivity combined with short run times on the liquid chromatography systems. Therefor der to gain insight into the regulation of these opening the possibility of measuring large protein-interaction networks it is necessary to quantities of different samples in a short identify proteins as well as their modifications time period and quantifying all components of the cascade. and to measure the protein expression levels Several collaborations of the core facility within a cell. This requires the identification within the MDC are now based on selected of several hundred or thousand proteins and reaction monitoring experiments. The core their quantification in a complex mixture. Ad- facility used a tripple quadrupol mass- spectrometer for this type of analysis (Q- vances in modern mass-spectrometry based TRAP 5500). One example of this technique proteomics made it possible to match all these criteria. The technique quickly rose to - involves the research on the inflammatory be the default method for large-scale protein proteinsNF-κB pathway. p100 and For p105 this projectand their a quantifi proteo- identification in life sciences. lyticallycation method shorted for versions the two p52 NF-κB and precursor p50 was The core facility mass-spectrometry offers developed. By using specialized SRM tech- a wide range of proteomic methods for the niques it was possible to quantify these molecules in cell extracts. This quantitative data identification and quantification of proteins was used to generate mathematical models, and peptides. Besides the identification of single proteins in solutions and SDS gel slices, the mass-spectrometry core facility uses a which allowed the identification of a new signalingNon-targeted complex proteomic in the NF-κB system. number of proteomic techniques in different approaches collaborations with MDC groups at the MDC. The increased sensitivity of mass-spectrom- In general the techniques can be divided into eters called for different ways to analyzed two groups: targeted and discovery or non- measurements are done quantitatively. For targeted proteomics. thisthe recordedseveral methods data. In for the the core introduction facility all of isotopic labels have been established. Targeted proteomics Among them (is) metabolic labeling. Meta- Monitoring the regulation of proteins un- bolic labeling of cells offers great advantages der different conditions can provide deep insight into regulatory networks, which of interest are therefore cultured in media, underly different signal transduction cas- whichfor the containquantification isotopic of labeled proteins. (non The radio cells- cades. For many cascades the major players active) amino acids. The additional mass - of the amino acids can later be detected by corporating this knowledge into a targeted mass- spectrometry. Comparison of two dif- strategyof these pathwaysallows to are monitor already changes identified. in theIn ferent samples is possible by analyzing the

260 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Model of the processing of the two NF-kB molecules p100 and p015 to the p52 and p50, respectively. Activation by lymphotoxin stimulation leads to the induced processing of the the two precursor-proteins. Using targeted mass- spectrometry the absolute amounts of p100, p105, p52 and p50 at different time points after stimulation were determined and used in the mathematical models.

sample on a high accuracy orbitrap mass spectrometer (Q-Exactive). The results of Selected Publications these measurements are relative ratios for each protein detected in the two different ubiquitin landscape. Proteomics, n/a–n/a. http://doi.org/10.1002/pmic.201500290 samples. This techniques is currently used Beaudette, P., Popp, O., & Dittmar, G. (2015). Proteomic techniques to probe the (2015). Neutrophil serine proteases exert proteolytic activity on endothelial cells. immuno-precipitations and large scale pro- Jerke, U., Hernandez, D. P., Beaudette, P., Korkmaz, B., Dittmar, G., & Kettritz, R. by the core facility for the quantification of Kanashova, T., Popp, O., Orasche, J., Karg, E., Harndorf, H., Stengel, B., et al. (2015). DifferentialKidney International. proteomic http://doi.org/10.1038/ki.2015.159 analysis of mouse macrophages exposed to adsorbate-loaded For samples, which cannot be cultured in heavy fuel oil derived combustion particles using an automated sample-preparation mediatein identifications. containing heavy amino acids, another technique is available. This technique Tuorto,workflow. F., Herbst,Analytical F., Alerasool,and Bioanalytical N., Bender, Chemistry. S., Popp, http://doi.org/10.1007/s00216- O., Federico, G., et al. (2015). relies on a highly reproducible liquid chro- The015-8595-4 tRNA methyltransferase Dnmt2 is required for accurate polypeptide synthe- matography, since two different chromatographic runs have to match. sis during haematopoiesis. The EMBO Journal, 34(18), 2350–2362. http://doi. org/10.15252/embj.201591382 Identification of post-translational Yılmaz,interdependent Z. B., Kofahl, precursor B., Beaudette, proteolysis. P., Baum, Cell Reports, K., Ipenberg, 9(5), I.,1756–1769. Weih, F., et http://doi.al. (2014). modifications Quantitativeorg/10.1016/j.celrep.2014.11.014 dissection and modeling of the NF-κB p100-p105 module reveals The regulation of protein activity or interactions with other proteins can be modulated - Group Leader Evelyn Ramberger no-acids in the peptides sequence. For the Dr. Gunnar Dittmar Sylvia Zerath Gurevich understandingby the modification of the of molecular side groups mechanisms, of the ami until September 2016 which regulate these proteins, it is necessary Graduate and Undergraduate to gain insight into the different post-trans- Scientist students Dr. Daniel Perez Hernandez Mercè Conill-Cortés - Dr. Oliver Popp Ortal Hayat trometrylational modificationsbased proteomic of analysis. proteins. The These core Rahel Lewin facilitymodifications actively can pursues be detected the development by mass-spec of PhD students Miru Hans Werner - Gülkiz Baytek Sebastian Günther Patrick Beaudette new methods for the identification of modi Günther Kahlert Secretariat thefication core sites facility by ubiquitin-likenow provides proteins. specialized For Tamara Kanashova Manuela Brandenburg methodsthe identification for improved of phosphorylationsensitivity. sites

MDC Research Report 2016 261 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Stefan Kempa

Proteomics / Metabolomics, BIMSB

The ‘integrative proteomics and metabo- spectrometer. Therefore, we have installed lomics’ research and technology platform three different mass spectrometers coupled to several chromatographic systems to is an integral part of the Berlin Institute for achieve a high coverage of the biochemical Medical Systems Biology (BIMSB). We are network of cells and organs (Fig. 1). an Interdisciplinary team consisting of biologists, engineers, chemists/pharmacologists • GCxGC-EI-ToF, GC-Q Exactive, gas chromatography based systems that allows and bio-informaticians. the measurement of small molecules (intermediates of the metabolism) Our core interest is to decode the regulation • QQQ-LC-MS, a triple quadrupole mass of metabolism at the molecular level. The spectrometer allows the measurement chemical space of metabolites spans a wide of small molecules, lipids and peptides range of physicochemical properties. The (proteomics) size of the molecule, its polarity and com- • Q-Exactive, a fourier transformation mass spectrometer that is used for pro- method and its detectability by the mass teomics and metabolomics analyses plexity finally determines the separation

Figure 1. Overview about the mass spectrometers installed at BIMSB and related methods. During the last years we have installed a GC-ToF-MS, a triple-quadrupole-MS and a LTQ-Orbitrap-Velos. The scheme shows the mass spectrometers, established methods and related metabolic pathways that can be analyzed.

262 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Figure 2. Dynamic analysis of glucose and glutamine metabolism in HCC tissues of ASVB mice. Stable isotope labeled glucose (13C glucose) and glutamine (13C glutamine) were applied i.p. and tumor and liver tissue were taken after 10 min. 13C carbon incorporation into central metabolites was analyzed using GC-MS based metabolomics (in vivo pSIRM).

Using the established mass spectrometry based platforms we are able to quantitatively measure metabolites (small molecules) lipids and proteins from variety

of sample species. In addition to our own scientific projects, we collaborate with our colleagues at the BIMSB and MDC and apply proteomics and metabolomics techniques;-

• StableMetabolic isotope profiling resolved (targeted, metabolomics non-target (in celled, lipid cultures profiling) and in vivo)

and targeted proteomics) Selected Publications • Proteome analyses (SILAC, in vivo SILAC Dejure FR#, Royla N#, Herold S, Kalb J, Walz S, Ade CP, Mastrobuoni G, Vanselow J, Metabolic Regulation

continuously adjusted to the physiological Schlosser(*,#equal) A, Wolf E, Kempa S* and Eilers M* 2017 The 3’-UTR of MYC couples RNA vonpolymerase Eyss B, Jaenicke II function LA, to Kortlever ribonucleotide RM, Royla levels, N, Wieseaccepted KE, in Letschert EMBO Journal S, McDuffus 2017 LA, Central metabolism is highly flexible and - and its regulation appears at the (post-) chondrial Dynamics Limits YAP/TAZ Function in Mammary EpithelialCells and Breast transcriptional,program of the cell,(post-) organ translational and organism; and Sauer M, Rosenwald A, Evan GI, Kempa S, Eilers M. A MYC-Driven Change in Mito Bielow C, Mastrobuoni G, Kempa S. Proteomics Quality Control: Quality Control Soft- allosteric level. Central metabolism is a Cancer. Cancer Cell. 2015 Dec 14;28(6):743-57. metabolic super pathway including e.g. Pietzke M, Zasada C, Mudrich S, Kempa S. Decoding the dynamics of cellular metabo- glycolysis, pentose phosphate pathway, tri lismware and for MaxQuantthe action ofResults. 3-bromopyruvate J Proteome Res. and 20152-deoxyglucose Dec 28. using pulsed stable carboxylic acid cycle, amino acid metabolism, glycogen and nucleotide metabolism. Liu L, Ulbrich J, Müller J, Wüstefeld T, Aeberhard L, Kress TR, Muthalagu N, Rycak L,isotope-resolved Rudalska R, Moll metabolomics. R, Kempa S*, CancerZender Metab.L, Eilers 2014 M, Murphy Jun 30;2:9. DJ. Deregulated MYC Our goal is to characterize the metabolic expression induces dependence upon AMPK-related kinase 5. Nature. 2012 Mar reprogramming events during cellular differentiation and transformation at the mo- 28;483(7391):608-12.Patents: (*corresponding for metabolomics) lecular level. We aim to understand (i) the metabolic differences of stem cells and can- method for analysing metabolic products in biological samples, 2015, US Patent cer cells. (ii) How metabolism can serve as Kempa,20,150,330,969 Stefan; Pietzke, Matthias; Zasada, Christin;Short term isotope pulse labeling a target for cancer treatment. Currently, we focus our technology development and research on decoding the crosstalk Group Leader Sabrina Geissenberger, AG between metabolism and post-transcrip- Dr. Stefan Kempa Müller, ECRC tional gene regulation. RNA or DNA binding Ana Rita Teixeira da Costa, AG proteins as well as RNA and DNA itself can Scientists be target of a number of modifying enzymes. Dr. Christin Zasada Dr. Raphaela Fritsche Meyer,Technical MPI Assistant for Infection Biology substantially to the post-transcriptional reg- Dr. Guido Mastrobuoni Jenny Grobe These metabolic modifications contribute- PhD students Staff atulatory (post-) code. transcriptional In addition, and the (post-) metabolic transla sta- Birte Arlt Sabrina Deter tus of the cell can influence gene expression between metabolism and gene regulation we Martin Forbes Alumni (2015-2017) aretional developing levels. In ordernew metabolomics to analyze the crosstalkand pro- OlgaJulia VvedenskayaLöffler Dr. Henning Kuich teomics approaches. Using comprehensive Nadine Royla Dr. Chris Bielow - Dr. Matthias Pietzke - Joint PhD students Amit Kumar sponsivenucleotide and profiling protective and metabolic RNA-protein feedback inter Henning Langer, AG Spuler, ECRC Dr. Tobias Opialla action studies, we identified a nucleotide re

mediated by the 3’UTR of MYC. MDC Research Report 2016 263 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES Photo: David Ausserhofer/MDC Photo:

Ralf Kühn

Transgenic Core Facility

Engineering of the mouse genome to create targeted mutants is a key technology for biomedical research. The Transgenic Core Facility (TCF) provides a central service for all kinds of reproductive technologies, supporting MDC research groups and their collaborators in their work with genetically modified mouse lines. One focus of the TCF is the rederivation of mouse lines by embryo transfer to exclude pathogens and and the cryopreservation of lines using embryo or sperm freezing. The second focus of the TCF is the generation of new mouse lines by microinjection of genetically modified embryonic stem cells into blastocysts or of DNA & RNA molecules into zygotes which induce genetic modifications in one-cell embryos.

Rederivation and cryopreservation of mouse lines The Transgenic Core Facility (TCF) offers in- house service for all kinds of reproductive technologies to support MDC research groups and their collaborators which are working

the planning phase, we support users with thegenetically design modifiedand layout mouse of their lines. projects. Starting One at focus of our work is the rederivation by embryo transfer for the import of mouse lines Figure 1 Pronuclear microinjection. from external facilities and for the transfer of A zygote fixed to the holding capillary (a) is penetrated - with the injection capillary (filled with DNA/RNA solution gen-free facility, without the burden caused and moved further to place the capillary`s tip in middle of byimmune opportunistic deficient strainsinfections. into theFurthermore, new patho a pronucleus (b). When the injection pressure is applied a we cryo-preserve the germ cells of precious swelling of the injected pronucleus is observed visible (c), mouse lines and bring frozen lines back to indicating successful microinjection. life. Cryopreservation is an important tool to prevent gene drifting, the loss of colonies in ing number of mouse models worldwide, we case of infection and to save cage space and see an increase in the number of organisa- resources occupied by lines which are not tions who offer to send frozen material in- actively used for research. Since most groups stead of live mice. These lines are revitalised choose to conserve mouse sperm this is the at the MDC follwing a variety of protocols de- main form of germ plasm we freeze down pending on the method of cryo-preservation. and store in liquid nitrogen. Due to the grow-

There are individual projects that do not fit 264 MDC Research Report 2016 CENTRAL TECHNOLOGY PLATFORMS AND CORE FACILITIES

Figure 2 Gene editing in mouse zygotes using CRISPR/Cas. Zygotes are microinjected with Cas9 mRNA, sgRNA and a gene targeting vector. Upon the production of Cas9 protein the nuclease binds together with sgRNA to the selected target site and creates a double-strand break. DSBs can be repaired by homology-directed repair (HDR) or non-homologous end-joining (NHEJ), leading to a knockin or knockout allele. Founder mutants (F0) are further mated to establish a mutant line.

into any of the above that can be supported by the TCF and we will help whenever it comes to the production, isolation, manipulation, culture or retransfer of pre-implantation embryos. Moreover, we can give advice on cloning and targeting strategies, ES cell culture and derivation and coat color genetics.

Generation of genetically modified mouse lines A second focus of the TCF is the generation of

is traditionally accomplished by gene tar- getingnew genetically in embryonic modified stem mouse(ES) cells lines. and This the generation of germline chimaeras obtained through the microinjection of recombinant - gineered nucleases, introduced into zygotes byES pronuclearcells into blastocysts. microinjection In recent (Fig. 1),years able en to create targeted double-strand breaks (DSB) in the genome of the one-cell embryo, have emerged as powerful tool for the single step production of targeted mutants, independent of ES cells. Proof of principle was provided Patents / Patent applications of which have been displaced by the more with zinc-finger nucleases and TALENs, both Record 2014/15/16 21 new mouse lines produced by injection of recombinant ES cell lines system. This system is composed of the ge- versatile and efficient CRISPR/Cas9 nuclease target sites by short sgRNAs which include 20 21735 new mouse mouse lines lines preserved produced by spermby pronuclear freezing microinjection (incl. 18 projects using CRISPR/Cas) nucleotidesneric Cas9 nuclease complementary that is guided to the totarget specific se- 172 mouse lines transferred by hygienic rederivation or revitalised by in vitro quence. Gene editing is achieved by endoge- fertilisation nous DSB repair pathways, either imprecisely by non-homologous end joining (NHEJ) caus- Selected Publications ing small deletions, or by homology-directed Brandl, C., Ortiz, O., Röttig, B., Wefers, B., Wurst, W., Kühn, R. (2015). Creation of repair (HDR) using repair template vectors for the precise insertion of new sequences mouse embryos. FEBS Open Bio 5, 26–35. Chu,targeted V.T., genomicWeber, T., deletions Wefers, B., using Wurst, TALEN W., Sander,or CRISPR/Cas S., Rajewsky, nuclease K., and pairs Kühn, in one-cell R. - (knockin) (Fig. 2). In mouse zygotes, CRISPR/ ing small deletions and knockout mutations Chu,(2015). V.T, Increasing Weber, T., Graf,the efficiency R., Sommermann, of homology-directed T., Petsch, K., repairSack, U., for Volchkov, CRISPR-Cas9- P., byCas9 the has NHEJ been repair efficiently pathway, used reaching for generat fre- induced precise gene editing in mammalian cells. Nat. Biotechnol. 33, 543–548.

Rajewsky, K., Kühn, R. (2015) Efficient generation of Rosa26 knock-in mice using Cas9/sgRNA microinjections, even in inbred CRISPR/Cas9 in C57BL/6 zygotes. HYPERLINK “http://www.ncbi.nlm.nih.gov/ quencies of up to 50% in pups derived from pubmed/26772810”BMC Biotechnol. 2016 Jan 16;16(1):4. - HDR events with co-injected targeting vec- Chu,mune V.T., cells Graf, using R., CrispRGoldWirtz, T., Weber, and a T., C57BL/6 Favret, J., Cas9 Li, X., transgenic Petsch, K., mouse Tran, line.N.T., Proc.Sieweke, Natl. torsbackgrounds occur less such frequently as C57BL/6. in zygotes In contrast, at rates M.H.,Acad. Berek,Sci. U.S.A. C., et 113, al. (2016).12514–12519. Efficient CRISPR-mediated mutagenesis in primary im

and template plasmids for the preparation of Group Leader of 5–10%. The TCF offers advice, protocols- Dr. Ralf Kühn jection into zygotes and embryo transfer into Sabine Manz fosterCRISPR/Cas females, reagents, and provides performs help the with microin geno- Technical Assistants David Schreiber Rainer Kabisch addition, we aim for developing and testing Katja Becker Administrative Assistance newtyping protocols of the resultingand reagents founder which mutants. increase In Andrea Leschke Christine Krause the recovery of knockin mutants.

MDC Research Report 2016 265 Photo: Harry Schnitger /MDC Annex

MDC Research Report 2016 267 ANNEX

Appointments and Recruitments

Research Groups Matthias SELBACH received W3 Professorship at Charité 2014 Matthias Selbach studied biology at the University of Münster and earned his PhD in the lab of Thomas F. Holger GERHARDT began work at MDC, Charite, - DZHK and BIH in Berlin - Several research institutions in Berlin have jointly suc- thiasMeyer Mann at the in MaxOdense, Planck Denmark, Institute and for followed Infection him Biol to ceeded in attracting the angiogenesis specialist Holger ogy, Berlin. In 2004, he proceeded to the lab of Mat Gerhardt to the capital city. Since September he is re- a postdoctoral fellow in a renowned proteomics lab, quantitativethe Max Planck mass Institute spectrometry of Biochemistry became ain cornerstone Munich. As - of his work. Since 2007, he headed the independent mentalsearch groupCardiovascular leader at ResearchMDC and atat theBerlin Charité Institute Berlin. of Furthermore,Health (BIH) and he isconcurrently integrated intoW3 Professorthe German of ExperiCenter at the MDC. From 2010, he was also an associate pro- for Cardiovascular Research (DZHK). After studying junior group “Cell signalling and mass spectrometry” biology and receiving his PhD in 2000, Gerhardt worked appointment to a full professorship for Proteomics as a postdoctoral fellow in der lab of Christer Betsholtz atfessor the Charité at the Charité.in cooperation In 2014, with Selbach the MDC. accepted His team the in Sweden until 2004. His most recent position was at combines expertise in proteomics and cell biology to study biological systems at the global scale. Recent re-

the London Research Institute, UK, and concurrently at Leuven,the Vesalius Belgium. Research Gerhardt Center investigates of the Flanders the formation Institute search highlights are the first systematic analysis of theof blood Biotechnology vessels in (VIB)organism at the and Catholic disease Universitydevelopment of cadeimpact in ofmammalian microRNAs cells. on protein synthesis and the first and is seeking ways to stop pathological vessel growth. global quantification of the entire gene expression cas Hans Schildbach endowed professorship for Kai Young-Ae LEE appointed to W3 Professorship for SCHMIDT-OTT Molecular Pediatric Allergology at Charité The nephrologist and clinical researcher Kai Schmidt- With Young-Ae Lee, the MDC succeeded to recruit an Ott accepted the appointment to a W2 professorship for excellent clinician scientist to its faculty. The joint ap- Urogenital Cancer Research at Charité. The professor- pointment to the full professorship for Molecular Pe- ship is jointly funded by the Foundation for Urological diatric Allergology by the MDC and the Charité is sup- Research and the MDC. Schmidt-Ott joined the MDC in 2007 and has worked as a Junior-Professor for Molecu- and Networking Fund of the Helmholtz President. lar Medicine at Charité since 2009, having carried out ported through additional funding from the Impulse- postdoctoral work in the laboratory of Jonathan Bar- lecular geneticist. Her research is concerned with the asch at Columbia University, New York, USA. By means Lee is a board-certified pediatrician and human mo of this joint appointment, Schmidt-Ott can continue genetic etiology of allergic disease in childhood and his career pathway with focus on the development of geneticswith disease and risk genomics prediction. of chronic Lee received inflammation, her medical the renal and uronary tract and intensify the cooperation degree from Free University Berlin after several years of MDC and Charité towards the development of new of clinical and research training in molecular genetics approaches in prevention, diagnostics and therapy of and genetic epidemiology at distinguished academic acute renal injury. institutions, including Harvard University and the Uni- versity of California, San Francisco. She completed her postgraduate clinical training in Pediatrics and Aller-

appointed Assistant Professor in 2002, and promoted gology at Charité Children’s Hospital, Berlin. She was

to Associate Professor in 2008.

268 MDC Research Report 2016 ANNEX

Altuna AKALIN joined the MDC/BIMSB the Transgenic Core Facility at MDC and established

using induced pluripotent stem cell lines and designer wasIn March recruited 2014, as Altuna head Akalinof the frombioinformatics Friedrich Miescherplatform nucleasein parallel technologies his own BIH-funded to study diseaseresearch mechanisms group that inis Institute for Biomedical Research in Basel, Switzerland,- human cells by reverse genetics.

PhDand independentstudies to the research University group of Bergen,leader at Norway, BIMSB. and Af Niccolo ZAMPIERI recruited from Columbia Univer- spentter studies some intime Istanbul, for a collaboration Turkey, Akalin project moved on for high- his sity throughput sequencing data analysis as a visiting stu- Niccolo Zampieri was jointly recruited by MDC and Charité/NeuroCure from Columbia University, New From 2010, Akalin stayed at Weill Cornell Medical York, USA. He was offered a NeuroCure group leader dent researcher at RIKEN Yokohama Institute, Japan. position with labs at the MDC starting in September 2014. Zampieri studied Biology at the University of Mi- aCollege, postdoctoral New York, fellow first in asthe a labpostdoctoral of Dirk Schuebeler and then asat lano where he also received his PhD degree. He com- a senior research associate. In 2013, he continued as on the computational analysis and integration of high- University, USA, and moved to Columbia University for throughputFriedrich Miescher datasets, Institute. such asAkalin’s microarrays research and focusses next pleted his PhD thesis in Moses Chao’s lab at New York generation sequencing, to understand the genetic and understanding the mechanisms that control neuronal epigenetic control mechanisms of cancer. his postdoctoral work. Zampieri’s research is aiming at-

Michela Di VIRGILIO started Helmholtz Young In- thatpositioning drives neural and synaptic circuit assembly. specificity during develop vestigators Group ment of the nervous system and at defining the logic Since September 2014, Michela DiVirgilio is heading 2015 an independent research group at MDC. Di Virgilio is a specialist in DNA repair and maintenance of genome Wei CHEN received full professorship for Function- stability and her group is co-funded from the Helm- al Genomics and Systems Biology at Charité holtz program. Having studied Biology at the Univer- After studies in Biochemistry and Medical Genetics at sity of Milano, she completed her PhD thesis under the supervision of Jean Gautier at Columbia Univer- scientist Wei Chen completed his PhD study at Max sity, New York, USA. She then joined the lab of Michel Xiamen and Sichuan University in China, the Chinese Nussenzweig at the Rockefeller University, New York, he focused on the analysis of structural variations in USA, for her postdoctoral training. thePlanck human Institute genome for and Molecular their roles Genetics, in causing Berlin human where congenital disorders. During this period, Chen be- Ralf KÜHN recruited to MDC and Berlin Institute of came interested in the development of novel genomics Health (BIH) Ralf Kühn was recruited to the MDC from Helmholtz- established a research group in the same institute that Zentrum München where he had worked as a staff technology. Right after finishing his PhD thesis, Chen-

Developmental Genetics with focus on genome engi- became one of the pioneers in the field of next genera neeringscientist and and diseaseresearch modeling group leader in mice, in the embryonic Institute and for thetion scratch sequencing one of (NGS). the leading In 2009, NGS he facilities moved in his Europe. lab to induced pluripotent stem cells. Kühn completed his Chenthe newly accepted established an appointment BIMSB at MDC to a and full set professor up from- PhD thesis in the lab of Klaus Rajewsky at the University ship for Functional Genomics and Systems Biology at - Charité starting in August 2015. Wei Chen accepted an netics in Cologne, he held a position as Head of Mouse offer for a full professorship at Southern University of Geneticsof Cologne. and After Principle a postdoc Scientist time at at Artemis the Institute Pharmaceu for Ge- Science and Technology of China, Shenzhen, and left ticals GmbH, Cologne. From May 2014, Kühn took over

the MDC/BIMSB in June, 2016.

MDC Research Report 2016 269 ANNEX

Irmtraud MEYER appointed to full professorship at Free University Berlin Humanto the Howard Frontier Hughes Science Medical Program Institute Fellow Janelia under Farm su- University of British Columbia, Vancouver, Canada, re- pervisionResearch Campusof Robert in Ashburn,H. Singer, USA. Albert In 2012,Einstein he becameCollege ceivedThe computational the call for the scientist W3 professorship Irmtraud “Bioinforma Meyer from of Medicine, New York, and Eugen Myers, Max Planck

the Free University Berlin and a research group leader September 2015, Stefan Preibisch was recruited to the positiontics of RNA at StructureMDC in 2015. and Transcriptome The joint recruitment Regulation” is coat Institute for Molecular Cell Biology and Genetics. In digital representation of the nematode Caenorhabditis elegansMDC/BIMSB. in the His dauer team diapause is aiming thatat creating will combine a complete mi- onfunded transcriptome from the Impulse regulation and byNetworking RNA folds, Fund RNA-RNA of the croscopy data of all scales as well as high‐resolution interactionsHelmholtz President. and RNA-protein The focus complex of Meyer’s formation. research After is imaging of transcription. studies in physics and mathematics, she did her PhD under supervision of Richard Durbin at the University 2016 of Cambridge, UK. She continued her academic career in the UK as a postdoctoral researcher at the University Michael GOTTHARDT received W3 professorship at the Charité Cambridge. Meyer started working at MDC in January - 2016.of Oxford and at the European Bioinformatics Institute, chael Gotthardt accepted an appointment to the full professorshipIn August 2016, for the Experimental MDC research und group Translational leader Mi Recruitment of Jan Philipp JUNKER for a junior Cardiology at the Charité Medical Faculty Berlin. Mi- group leader position at BIMSB chael Gotthardt studied medicine at the University of Heidelberg and completed his thesis at the Max Planck Netherlands, began as a new junior group leader at - Jan Philipp Junker from Hubrecht Institute, Utrecht, cal internships in Berlin and Boston, he worked as a quantitative developmental biology at the interface postdoctoralInstitute for Molecularfellow at the Genetics University in Berlin. of Texas After in Dallas. clini betweenMDC’s BIMSB experimental in November and theoretical 2015. His teamscience. works His reon- Michael Gotthardt has been leading a reserach team at search goal is to develop a comprehensive understand- the MDC since 2002. Until today, he is also holding a Re- ing of the spatiotemporal interplay between transcrip- search Assistant Professorship at the Washington State University in Pullman and a Research Associate Profes- expression for establishing and maintaining cell fates sorship at the University of Tucson in Arizona. Michael andtion complexfactor binding, body plans. chromatin Junker modifications, studied Physics and at gene the Gotthardt is continuing his research group at the MDC Technical University Munich where he also received his in order to increase the understanding of the cause and PhD degree. Then he worked as a postdoctoral fellow in progression of inherited heart and muscle diseases and the team of Alexander van Oudenaarden at Massachu- aid in developing novel therapeutic strategies.

Gaetano GARGIULO started Helmholtz Young Inves- setts Institute of Technology, USA. Since 2009, he was tigators Group at MDC RecruitmentSenior Postdoc ofat Hubrecht Stephan Institute. PREIBISCH for a junior Gaetano Gargiulo was recruited in September 2016 group leader position at BIMSB Stephan Preibisch studied computer science at the establish with co-funding of the Helmholtz Association Technical University Dresden. Then he applied for the from the Netherlands Cancer Institute in Amsterdam to - MDC. Together with his team, Gargiulo works on the lar Cell Biology and Genetics in Dresden and joined the moleculara junior research characterization group on “Molecular of the Gliobastoma, Oncology” whichat the interdisciplinaryPhD program of theteam Max of PlanckPavel Tomancak. Institute for He Molecu contin- is a brain tumor with a very bad clinical prognosis. ued there for a short postdoc term before he moved Gaetano Gargiulo is generating so-called Glioblastoma

270 MDC Research Report 2016 ANNEX

avatars as disease models for an improved treatment. Technology Platforms Gargiulo did his graduate studies at the University of - 2014 lar and Cell Biology from the University of California Milano in Italy and received his PhD degree in Molecu Sebastian DIECKE - gotat Berkeley. additional He training spent hisin genomics first postdoc and epigenomics.term in Eileen ford School of Medicine had beenin California Interim Directorbefore he of was the Furlong’s group at the EMBL in Heidelberg, where he Stem Cell Core at the Cardiovascular Institute of Stan Mina GOUTI recruited from Francis Crick Institute subsidiary of the BIH Core Facility Stem Cells which is Since April 2016, Mina Gouti is heading an indepen- locatedrecruited on to Campus the BIH/MDC. Buch. In 2014, he established the dent junior group at the MDC, which is also part of the 2016 her team are using human and mouse pluripotent stem cellsHelmholtz to model Program and understand “Personalized normal Medicine”. and pathologi Gouti and- Since mid-2016, Jennifer KIRWAN is heading the BIH cal development of the neuromuscular system. After Metabolomics Core Facility, which is also located studies in Molecular Biology at the University of Hert- on Campus Buch. Before she moved to Berlin, she had gathered experience in managing the Metabolomics Fa- - cility at the University of Birmingham in UK for over ence,fordshire, Technology Mina Gouti and finished Medicine. her Formaster her thesisPhD thesis,in the Minagroup Gouti of Amanda moved Fisher to the atBiomedical the Imperial Research College Founda of Sci- tion of the Academy of Athens. After two postdoctoral Saschafive years. SAUER took the leadership of Genomics Tech- - nology Platforms of BIH and BIMSB in August 2016. tigator Scientist in the lab of James Briscoe at the Fran- Dr. Sauer, who is also principal investigator in the “Eu- terms in Athens and London, she became Senior Inves - Rolandcis Crick InstituteSCHWARZ in London.starts new BIMSB Junior Re- ularropean Genetics Sequencing in Berlin and where Genotyping he held an Infrastructure”, independent search Group groupwas recruited leader position. from the Max Planck Institute for Molec Roland Schwarz studied Computer Science at the Uni- versity of Applied Sciences in Würzburg and became Andrew WOEHLER is biomedical engineer by train- graduate student in the bioinformatics team of Tobias ing. He was leading a junior group within the Cluster of Excellence and DFG Research Center for Nanoscale postdoctoral fellow at the Cancer Research UK Cam- Microscopy and Molecular Physiology of the Brain at Müller at the University’s Biocenter. He worked as a subsequently with a Marie Curie Fellowship at the Eu- Göttingen when he was offered the leadership of the bridge Institute in the group of Florian Markowetz and BIMSBthe Max Light Planck Microscopy Institute for Technology Biophysical Platform Chemistry start in- group of Nick Goldman. Schwarz accepted the offered ing from May 2016. ropean Bioinformatics Institute in Hinxton, UK, in the his group “Evolutionary and Cancer Genomics“ in Oc- toberposition 2016. as juniorHe is interested group leader in how at BIMSBsomatic and alterations started - ness and progression in the clinic and to what extent somaticand genomic variation rearrangements interacts with contribute the germline to tumor genetic fit background to shape the regulatory landscape of cancer.

MDC Research Report 2016 271 ANNEX

European Research Council (ERC) Grants

ince 2007 the European Research Council (ERC) has been funding scientists in basic research S with one of the most important and prestigious MDC have been awarded 21 of the highly sought-after funding programs in Europe. In total, scientists of the 16 grant winners are currently working at the MDC. ERC grants, 18 of them were raised directly at the MDC. ERC Advanced Grants ERC Consolidator Grants

Funding period: 5 years Funding period: 5 years

Advanced Grants support established researchers who Consolidator Grants are intended for independent are seen as independent research leaders in their own scientists at a stage in their career at which they are right. To qualify for an application, researchers need an consolidating their research teams and programs. The grant supports researchers with up to € 2.75 million. ten years. An Advanced Grant supports researchers for 5excellent years with scientific € 2.5 million track record (up to over € 3.5 a periodmillion of in at excep least- James Poulet tional cases). Neural mechanisms of multisensory perceptual bind- Thomas Willnow ing2016 | ACoolTouch

Metabolic basis of neurodegenerative disease Oliver Daumke 2014 | BeyOND Zsuzsanna Izsvák Structural basis of mitochondrial inner membrane shape2014 | and MitoShape dynamics - man2011 disease | TRANSPOSOstress Impact of stress-induced transposon activities on hu Thomas J. Jentsch ERC Starting Grants

- Funding period: 5 years ganelles2011 | CYTOVOLION Ion homeostasis and volume regulation of cells and or Starting Grants support young and aspiring researcher Gary Lewin - dence. The grant supports scientists with up to € 2 mil- Molecular exploitation of an extremophile mammal lion.leaders and promote their early scientific indepen 2011 | EXTREMOPHILE MAMMAL Klaus Rajewsky Gaetano Gargiulio

Modeling lymphoma pathogenesis in mice - from basic Glioblastoma Subtype Avatar Models for Target Disco mechanisms2010 | LYMPHOMA to pre-clinical models very2016 and | iGBMavatars Biology

Martin Lohse* Jan Philipp Junker

Towards the Quantal Nature of Receptor/cAMP Signals Quantitative analysis of variability and robustness in 2008 | TOPAS spatial2016 | SPACEVARpattern formation

272 MDC Research Report 2016 ANNEX

Michela Di Virgilio ERC Proof of Concept Grant

2014 | DNAendProtection Funding period: 18 months BarisDNA end Tursun protection in Immunity and Cancer Gary Lewin - ing2014 Cell | EPROWORM Fate Reprogramming Safeguarding Cell Identities: Mechanisms Counteract Francesca2016 | Spagnoli Holger Gerhardt* Exploring the Potential for Therapeutic Lineage Repro- Reverse engineering of vascular patterning through gramming2014 | TheLiRep of Diabetes mosaic2012 | REshape in vivo analysis of endothelial cell shape regulation Martin Lohse*

Michael Gotthardt FRET-based receptor screening assays 2013 | FRESCA A systems and targeted approach to alternative splic- ing2011 in | theCardioSplice developing and diseased heart: Translating * ERC Grant recipients have not been at the MDC at the time they basic cell biology to improved cardiac function received the grant. ** ERC Grant recipients received the funding at their time at the MDC, Jan-Erik Siemens** but left the research center in the meantime. (since February 2013 Medical Faculty of Heidelberg)

Peripheral and Central Mechanisms of Temperature Detection2011 | ThermoReg and Core Body Thermoregulation

James Poulet

Brain states, synapses and behaviour 2010 | BrainStates Matthew Poy

- cularization2010 | IsletVasc Molecular Mechanisms Regulating Pancreatic Islet Vas Francesca Spagnoli

Mechanisms underlying cell fate decision between pan- creas2009 |and HEPATOPANCREATIC liver

MDC Research Report 2016 273 ANNEX

Awards

2014 2015

Hua YU (Beckman Research Institute of City of Yehudit BERGMAN (HUJI Jerusalem) Hope) Humboldt Research Award (for a collaboration with Klaus Rajewsky and within the Alexander von Humboldt Foundation SignGeneHelmholtz Research International School) Fellow Award (for a collaboration with Thomas Blankenstein) Robson Augusto SOUZA DOS SANTOS (Federal Michael SIEWEKE (CIML; CNRS) University of Minas Gerais Belo Horizonte) Georg Forster Research Award (for a collaboration with Klaus Rajewsky) Alexander von Humboldt Foundation Einstein BIH Visiting Fellow (for a collaboration with Michael Bader) Jane HOLLAND Curt Meyer Memorial Price Amanda FISHER (ICL London) Berlin Cancer Association (for a collaboration with Nikolaus Rajewsky) Nikolaus RAJEWSKY Helmholtz International Fellow Award Honorary PhD Ulrich MÜLLER (Scripps Research Institute)

(for a collaboration with Gary Lewin) MichaelRome’s Sapienza SIEWEKE University Einstein BIH Visiting Fellow Elected EMBO member Jichang WANG and Zsuzsanna IZSVAK GSCN Publication of the Year Clemens SCHMITT and Soyoung LEE German Stem Cell Network (GSCN) Johann Georg Zimmermann Research Award MHH-plus-Stiftung (Medizinische Hochschule Michael BADER Hannover) Gold Medal E.K. Frey – E. Werle Foundation

Florian HERSE Rösslin-Preis Deutsche Gesellschaft für Gynäkologie und Geburtshilfe (DGGG)

Michela DI VIRGILIO

Female Independency Award UlrikeBerlin School STEIN, of Peter Integrative SCHLAG, Oncology Ulrich (BSIO)ROHR (Hoffmann-La Roche) Felix Burda Award for Colon Cancer Research

Marvin JENS Springer Award for Outstanding Theses in Systems Biology

274 MDC Research Report 2016 ANNEX

2016

Nir FRIEDMAN (HUJI Jerusalem) Humboldt Research Award Alexander von Humboldt Foundation (for a collaboration with Nikolaus Rajewsky and within the SignGene Research School)

Jan Philipp JUNKER ARCHES Prize Minerva Foundation (MPG/BMBF)

Markus LANDTHALER Science Award GlaxoSmithKline Foundation

Roland SCHWARZ Preis der Berlin-Brandenburgischen Akademie der Wissenschaften, gestiftet von der Monika Kutzner Stiftung zur Förderung der Krebsforschung

Matthias LEISEGANG Curt Meyer Memorial Price Berlin Cancer Association

Véronique GEBALA Helmholtz Doctoral Prize Helmholtz Association

Felix LORENZ, Elisa KIEBACK, Julian CLAUß and Inan EDES Grand Prize One Start Life Sciences Startup Award

MDC Research Report 2016 275 ANNEX

Selected Highlight Publications

2014

Argonaute2 Mediates Compensatory Expansion Activation of MAPK overrides the termination of of the Pancreatic β Cell. Tattikota, S. G., Rathjen, T., myelin growth and replaces Nrg1/ErbB3 signals during Schwann cell development and myelina- M., et al. (2014). Cell Metabolism, 19 (1), 122–134. tion. Sheean, M. E., McShane, E., Cheret, C., Wal- https://doi.org/10.1016/j.cmet.2013.11.015McAnulty, S. J., Wessels, H.-H., Akerman, I., van de Bunt, cher, J., Müller, T., Wulf-Goldenberg, A., et al. (2014). MOV10 Is a 5′ to 3′ RNA Helicase Contributing to Genes & Development, 28 (3), 290–303. https://doi. UPF1 mRNA Target Degradation by Translocation org/10.1101/gad.230045.113 along 3′ UTRs. Gregersen, L. H., Schueler, M., Mun- Tuning Piezo ion channels to detect molecular- schauer, M., Mastrobuoni, G., Chen, W., Kempa, S., et al. scale movements relevant for fine touch. Poole, (2014). Molecular Cell, 54 K., Herget, R., Lapatsina, L., Ngo, H.-D., Lewin, G. R. org/10.1016/j.molcel.2014.03.017 (2014). Nature Communications, 5, 3520. https://doi. Blood Flow and Bmp Signaling (4), 573–585.Control Endocardial https://doi. Chamber Morphogenesis. Dietrich, A.-C., Lombardo, Identification of LRRC8 Heteromers as an Essential V. A., Veerkamp, J., Priller, F., Abdelilah-Seyfried, S. Componentorg/10.1038/ncomms4520 of the Volume-Regulated Anion Chan- (2014). Developmental Cell, 30 (4), 367–377. https:// nel VRAC. Voss, F. K., Ullrich, F., Münch, J., Lazarow, K., doi.org/10.1016/j.devcel.2014.06.020 Lutter, D., Mah, N., et al. (2014). Science, 344 LRP2 Acts as SHH Clearance Receptor to Protect the Retinal Margin from Mitogenic Stimuli. Christ, A., RNA-binding protein RBM20 represses splicing (6184), to Christa, A., Klippert, J., Eule, J. C., Bachmann, S., Wal- orchestrate634–638. https://doi.org/10.1126/science.1252826 cardiac pre-mRNA processing. Maatz, lace, V. A., et al. (2015). Developmental Cell, 35 (1), H., Jens, M., Liss, M., Schafer, S., Heinig, M., Kirchner, M., et al. (2014). The Journal of Clinical Investigation, 124 PI3 Kinase and FOXO1 Transcription Factor Activ- ity36–48. Differentially https://doi.org/10.1016/j.devcel.2015.09.001 Control B Cells in the Germinal A somatosensory circuit for cooling perception in Center Light and Dark Zones. Sander, S., Chu, V. T., mice.(8), 3419–3430. Milenkovic, https://doi.org/10.1172/JCI74523 N., Zhao, W.-J., Walcher, J., Albert, Yasuda, T., Franklin, A., Graf, R., Calado, D. P., et al. T., Siemens, J., Lewin, G. R., Poulet, J. F. A. (2014). Na- (2015). Immunity, 43 ture Neuroscience, 17 (11), 1560–1566. https://doi. org/10.1016/j.immuni.2015.10.021 Sequential Poly-ubiquitylation (6), 1075–1086. by Specialized https://doi. Conju- Human satellite cells have regenerative capacity gating Enzymes Expands the Versatility of a Quality andorg/10.1038/nn.3828 are genetically manipulable. Marg, A., Esco- Control Ubiquitin Ligase. bar, H., Gloy, S., Kufeld, M., Zacher, J., Spuler, A., et al. O., Dittmar, G., Reiss, Y., Sommer, T., et al. (2016). Molec- (2014). The Journal of Clinical Investigation, 124 (10), ular Cell, 63 Weber, A., Cohen, I., Popp, molcel.2016.07.020 A Variety of Dicer Substrates in Human and C. ele- Kinetic Analysis (5), 827–839.of Protein https://doi.org/10.1016/j. Stability Reveals Age- gans.4257–4265. Rybak-Wolf, https://doi.org/10.1172/JCI63992 A., Jens, M., Murakawa, Y., Herzog, M., Dependent Degradation. McShane, E., Sin, C., Zau- Landthaler, M., Rajewsky, N. (2014). Cell, 159 (5), 1153– 1167. https://doi.org/10.1016/j.cell.2014.10.040 Cell, 167 Primate-specific endogenous retrovirus-driven cell.2016.09.015ber, H., Wells, J. N., Donnelly, N., Wang, X., et al. (2016). transcription defines naive-like stem cells. Wang, Der1 promotes (3), 803–815.e21. movement https://doi.org/10.1016/j. of misfolded proteins - through the endoplasmic reticulum membrane. nik, A., et al. (2014). Nature, 516 (7531), 405–409. Mehnert, M., Sommer, T., Jarosch, E. (2013). Nature Cell Bi- J., Xie, G., Singh, M., Ghanbarian, A. T., Raskó, T., Szvet ology, 16 Changes in neural network homeostasis trigger neu- https://doi.org/10.1038/nature13804 ropsychiatric (1), 77–86. symptoms. https://doi.org/10.1038/ncb2882 Winkelmann, A., Maggio, N., Eller, J., Caliskan, G., Semtner, M., Häussler, U., et al. (2014). The Journal of Clinical Investigation, 124 (2),

696–711. https://doi.org/10.1172/JCI71472 276 MDC Research Report 2016 ANNEX

2015 2016

Identification of human T-cell receptors with opti- Lineage-specific enhancers activate self-renew- mal affinity to cancer antigens using antigen-neg- al genes in macrophages and embryonic stem ative humanized mice. Obenaus, M., Leitão, C., Leise- cells. K., Maurizio, J., Fenouil, R., et al. (2016). Science, 351 et al. (2015). Nature Biotechnology, 33 (4), 402–407. (6274), Soucie, aad5510–aad5510. E. L., Weng, Z.,https://doi.org/10.1126/ Geirsdóttir, L., Molawi, gang, M., Chen, X., Gavvovidis, I., van der Bruggen, P., science.aad5510 Increasing the efficiency of homology-directed re- Blood flow drives lumen formation by inverse mem- pairhttps://doi.org/10.1038/nbt.3147 for CRISPR-Cas9-induced precise gene edit- brane blebbing during angiogenesis in vivo. Geba- ing in mammalian cells. Van Trung Chu, Weber, T., Wefers, B., Wurst, W., Sander, S., Rajewsky, K., Kühn, (2016). Nature Cell Biology, 18 (4), 443–450. https:// R. (2015). Nature Biotechnology, 33 la, V., Collins, R., Geudens, I., Phng, L.-K., Gerhardt, H. Targeting human melanoma neoantigens by T cell Translational regulation shapes the (5), molecular 543–548. receptordoi.org/10.1038/ncb3320 gene therapy. Leisegang, M., Kammertoens, landscapehttps://doi.org/10.1038/nbt.3198 of complex disease phenotypes. Schafer, T., Uckert, W., Blankenstein, T. (2016). The Journal of S., Adami, E., Heinig, M., Rodrigues, K. E. C., Kreuchwig, Clinical Investigation, 126 F., Silhavy, J., et al. (2015). Nature Communications, 6, SORLA facilitates insulin receptor (3), 854–858. signaling https://doi. in adi- Alk1 and Alk5 inhibition by Nrp1 controls vascular org/10.1172/JCI83465pocytes and exacerbates obesity. Schmidt, V., Schulz, sprouting7200. https://doi.org/10.1038/ncomms8200 downstream of Notch. - don, E., Dubrac, A., Ragab, A., Narloch, J., Vizán, P., et al. (2016). The Journal of Clinical Investigation, 126 (7), (2015). Nature Communications, 6, Aspalter,7264. https://doi. I. M., Gor N., Yan, X., Schürmann, A., Kempa, S., Kern, M., et al. Bimodal antagonism of PKA signalling by ARH- Angiotensin II Induces Skeletal Muscle Atrophy by GAP36.2706–2720. Eccles, https://doi.org/10.1172/JCI84708 R. L., Czajkowski, M. T., Barth, C., Müller, Activatingorg/10.1038/ncomms8264 TFEB-Mediated MuRF1 Expression- P. M., McShane, E., Grunwald, S., et al. (2016). Nature Novelty and Significance. Bois, Du, P., Tortola, C. P., Communications, 7 Lodka, D., Kny, M., Schmidt, F., Song, K., et al. (2015). ncomms12963 Circulation Research, 117 (5), 424–436. https://doi. Preventing tumor escape, 12963. by https://doi.org/10.1038/ targeting a post-proteasomal trimming independent epitope. Textor, A., Crystal structure of the dynamin tetramer. Reu- Schmidt, K., Kloetzel, P.-M., Weißbrich, B., Perez, C., Charo, bold,org/10.1161/CIRCRESAHA.114.305393 T. F., Faelber, K., Plattner, N., Posor, Y., Ketel, K., J., et al. (2016). Journal of Experimental Medicine, 213 (11), Curth, U., et al. (2015). Nature, 525 Quantitative interaction mapping reveals an extend- High salt reduces the activation of IL-4–(7569), and 404–408. IL-13– ed2333–2348. UBX domain https://doi.org/10.1084/jem.20160636 in ASPL that disrupts functional stimulatedhttps://doi.org/10.1038/nature14880 macrophages. Binger, K. J., Gebhardt, M., p97 hexamers. Arumughan, A., Roske, Y., Barth, Heinig, M., Rintisch, C., Schroeder, A., Neuhofer, W., et C., Forero, L. L., Bravo-Rodriguez, K., Redel, A., et al. al. (2015). The Journal of Clinical Investigation, 125 (2016). Nature Communications, 7, 13047. https://doi.

Meta-analysis identifies seven susceptibility loci in- Titin-truncating variants affect heart function in dis- volved(11), 4223–4238. in the atopic https://doi.org/10.1172/JCI80919 march. easeorg/10.1038/ncomms13047 cohorts and the general population. Schafer, S., Gordillo, J., Rüschendorf, F., Bauerfeind, A., Strachan, de Marvao, A., Adami, E., Fiedler, L. R., Ng, B., Khin, E., et D. P., Ben D Spycher, et al. (2015). Marenholz, Nature Communica I., Esparza-- al. (2016). Nature Genetics, 49 (1), 46–53. https://doi. tions, 6 Detecting actively translated open reading frames Small-molecule inhibition of STOML3 oligomeriza- in ribosome, 8804. profilinghttps://doi.org/10.1038/ncomms9804 data. Calviello, L., Mukherjee, tionorg/10.1038/ng.3719 reverses pathological mechanical hypersensi- N., Wyler, E., Zauber, H., Hirsekorn, A., Selbach, M., et tivity. Wetzel, C., Pifferi, S., Picci, C., Gök, C., Hoffmann, al. (2015). Nature Methods, 13 (2), 165–170. https:// D., Bali, K. K., et al. (2016). Nature Neuroscience, 20 (2),

doi.org/10.1038/nmeth.3688 209–218. https://doi.org/10.1038/nn.4454 MDC Research Report 2016 277 ANNEX

Scientific Meetings and Conferences

2014

Date Event Host / Organizer

15.-17.05.2014 European Worm Meeting Dr. Baris Tursun

12.-14.06.2014 7th Berlin Summer Meeting Prof. Nikolaus Rajewsky „From Systems Biology to Systems Medicine“

20.06.2014 5th Annual Scientific Symposium ,,Ultrahigh Field Prof. Thoralf Niendorf Magnetic Resonance Clinical Need, Research Promises, Technical Solutions”

29.09.-03.10.2014 36th FGMR Discussion Meeting, GDCh Prof. Helmut Oschkinat (FMP) Prof. Thoralf Niendorf

20.-22. 11.2014 International DZHK-Symposium “Receptors, G proteins DZHK, MDC and integration of Ca2+ signaling in the cardiovascular system

2015

Date Event Host / Organizer

28.-29.05.2015 Brain Tumor Meeting Dr. Baris Tursun

04.-06.06.2015 8th Berlin Summer Meeting Prof. Nikolaus Rajewsky Localization of cellular Processes

26.06.2015 6th Annual Scientific Symposium ,,Ultrahigh Field Prof. Thoralf Niendorf Magnetic Resonance Clinical Need, Research Promises, Technical Solutions”

10.07.2015 Frontiers in Developmental Neuroscience Prof. Carmen Birchmeier Symposium in Honour of Carmen Birchmeier

07.-09.10.2015 ECBS & ICBS joint meeting 2015 “Bringing Chemistry to Dr. Ronald Frank (FMP) Life” Prof. Michael Bader, Prof. Gary Lewin

278 MDC Research Report 2016 ANNEX

2016

Date Event Host / Organizer

01.-03.06.2016 European Worm Meeting Dr. Baris Tursun

24.06.2016 7th Annual Scientific Symposium ,,Ultrahigh Field Prof. Thoralf Niendorf Magnetic Resonance Clinical Need, Research Promises, Technical Solutions”

12.09.2016 Frontiers in DNA Repair Dr. Michela di Virgilio

13.-14.10.2016 9th Berlin Summer Meeting Prof. Nikolaus Rajewsky Circular RNAs and RNA Modifications

4.11.2016 Mechanisms of Molecular and Cellular Immunity 1964- Prof. Klaus Rajewsky 2016 Symposium hosted by Klaus Rajewsky

28.11.2016 Dechema Colloqium BBB with MDC Games of tag – Letting protein modifications out of the lab

MDC Research Report 2016 279 ANNEX

Selected Seminars 2014 – 2016

MDC LECTURE ELENA TIMOFÉEFF-

02.11.15 Frederik W. ALT (Boston Children’s RESSOVSKY LECTURE Hospital, Harvard Medical School, Howard Notable Women in Science and Medicine Hughes Medical Institute) Chromosomal Loop Domains, Genomic 29.04.14 Nine KNOERS (University Medical Center Rearrangements, and Cancer Utrecht) The potential of next generation Tom RAPOPORT (Harvard Medical School, Howard Hughes Medical Institute) impact and clinical utility 18.12.15 How an organelle gets into shape sequencing for renal disorders: scientific 10.06.14 Hua Eleanore YU (Cancer 03.11.16 Michel C. NUSSENZWEIG (Howard Immunotherapeutics Program, Hughes Medical Institute, The Rockefeller Comprehensive Cancer Center) University) STAT3 pathway: from fundamental discoveries to clinical translation

25.11.16 BotondThe HIV ROSKAProblem (Friedrich Miescher 26.03.15 Anne-Lise BOERRESEN-DALE (Oslo Institute for Biomedical Research (FMI) University Hospital)

and repair personalized treatment of breast cancer The first steps of vision: cell types, circuits Towards integrated “omics” for 03.12.15 Hannelore EHRENREICH (Max Planck Institute of Experimental Medicine) Neuropsychiatric diseases: Early steps

04.03.16 Annetowards CORCORAN biological (Babrahamdisease definition Institute) Local and global epigenetic regulation of immunoglobulin V(D)J recombination

06.07.16 Cornelia WEYAND (Stanford University) The Warburg Effect – from cancer to autoimmune disease

07.09.16 Emmanuelle CHARPENTIER (Max Planck Institute for Infection Biology)

Genome Engineering with CRISPR-Cas9 and -Cpfl: lessons learned from bacteria

280 MDC Research Report 2016 ANNEX

LUNCHTIME SEMINAR BIMSB DISTINGUISHED SERIES SPEAKERS DISTINGUISHED SPEAKERS 21.04.15 Matthias HENTZE (EMBL) Amir ORIAN (Technion-Israel Institute of RNA-binding proteins, metabolism and a Technology) new function of the genome 28.08.14 maintains the differentiated identity and 05.05.15 Eric WESTHOF (IBMC/CNRS) tissueIdentification homeostasis of a genetic network that How far can we predict RNA architectures and RNA binding of proteins Sebastian KADENER (Hebrew University) 26.05.15 Anders KROGH (University of Copenhagen) 18.09.14 factors determine the production rate of Bayesian transcriptome assembly improves circRNAsLinear splicing efficiency and trans-acting transcripts in RNAseq Yehudit BERGMAN (Hebrew University detection and quantification of alternative Medical School) 01.09.16 Oliver HOBERT (Howard Hughes Medical 18.06.15 Epigenetic programming of allelic choice Institute, Columbia University) A regulatory map of the C. elegans nervous 19.11.15 Angela de PACE (Systems Biology system Department at Harvard Medical School) Precision and plasticity in animal 06.12.16 Wolf REIK (Babraham Institute) transcription Epigenetic reprogramming in mammalian development 29.09.16 Martyn GOULDING (Salk Institute) Genetic approaches for dissecting the circuitry for touch and movement

24.11.16 Stirling CHURCHMAN (Harvard) Gene expression regulation at high resolution

MDC Research Report 2016 281 ANNEX

CANCER LECTURE NEUROSCIENCE SEMINAR

01.04.14 Angelika EGGERT (Charité – SERIES Universitätsmedizin Berlin) DISTINGUISHED SPEAKERS On the way to Systems Medicine Approaches in Neuroblastoma 16.04.14 Nils BROSE (Max Planck Institute of Experimental Medicine) 13.06.14 Michael KARIN (UCSD School of Medicine) synaptogenesis to autism spectrum disorders Neuroligins at inhibitory synapses – from 16.06.14 Hans Inflammation, SCHREIBER Immunity (University and Cancer of Chicago) Directing T cells to eradicate cancer with 11.06.14 Martyn GOULDING (Molecular Neurobiology Laboratory, The Salk Institute) 13.09.14 YinonTCRs, CARs BEN-NERIAH or fusion (Theproteins Hebrew University) in the spinal cord Making sense of the sensorimotor circuits 14.08.15 Robert G. ROEDER (The Rockefeller 01.08.14 Hannah MONYER (Universitätsklinikum University) Heidelberg, DKFZ) Local and long-range GABAergic inhibition diverse transcriptional co-activators in animal Mechanistic cells studies of the function of and its function for spatial coding in the 13.10.15 Jerry L. WORKMAN (Stowers Institute for 29.09.15 Alisonhippocampal-entorhimal LLOYD (MRC LMCB, formation UCL, London Medical Research) and visiting NeuroCure Fellow) Serine and SAM responsive complex nerve regeneration crosstalk by sensing cellular metabolism Multicellular complexity of peripheral SESAME regulates histone modification 11.12.15 Ryoichiro KAGEYAMA (Kyoto University) 01.12.15 Eduardo MORENO (University of Bern) neurogenesis Oscillatory control of somitogenesis and Regulating the cellular composition of our 22.01.16 Gerd KEMPERMANN (Center for 09.02.16 Andreasbodies using STRASSER fitness fingerprints (The Walter and Eliza Regenerative Therapies (CRTD) / DFG Hall Institute of Medical Research) research center at TU Dresden) New neurons and hippocampal develop novel cancer therapies individuality Learning the mechanisms of cell death to 05.09.16 Michael KARIN (USCD School of Medicine) 08.07.16 Elena CATTANEO (Laboratory of Stem Cell Biology and Pharmacology of chronic liver disease and cancer Neurodegenerative Diseases, University of Metabolic stress and inflammation in Milan) 09.11.16 Michael RAPE (UC Berkeley) Disease research: combining genetics with Neuronscell biology from and stem evolution cells for Huntington’s Ubiquitin-dependent control of stem cell fate and function

282 MDC Research Report 2016 ANNEX

WOLLENBERGER SPECIAL Cardiovascular and Metabolic Disease Seminar Series

20.10.14 Luc MAROTEAUX (INSERM U839, UPMC, Institut du Fer à Moulin) Serotonin from development to pathophysiology: The 5-HT2B receptor contribution

11.03.15 Alvin H. SCHMAIER (Case Western Reserve University) Mas agonism in thrombosis

23.03.15 Antonio BALDINI (University Federico II; Institute of Genetics and Biophysics)

targeting chromatin: the case of congenital heartBalancing disease gene haploinsufficiency by

10.06.16 Matthias RIEF (Technical University Munich) Mechanics of single protein molecules

MDC Research Report 2016 283 ANNEX

Facts and Figures

Personnel (as of December 31, 2016)

Staff and Guests 1660

thereof in research 1291 thereof PhD students 363

MDC Budget 2016 (as of December 31, 2016) 3 % Technology Transfer Core Funding 87,5 million Euro 17,1 million Euro Extramural Funding 30,7 million Euro Investments Total 135,3 million Euro

23 % Extramural Funding 65 % Core Funding

12 % Investments

284 MDC Research Report 2016 Other Revenues

Technology Transfer

DZHK Project Funding

BIH Project Funding

Third-Party Project Funding ANNEX

Other Revenues 27 % Other Revenues Technology Transfer

DZHK Project Funding 53 % Third-Party BIH Project Funding 3 % Project Funding Technology Transfer Third-Party Project Funding 5 % DZHK Project Funding

12 % BIH Project Funding

Source (no. of projects) Others (42) BMBF (36) Foundations (46) DFG (74)

HGF (25) Industry (13)

EU (31) EU (31)

Industry (13) HGF (25) Fundations (46)

DFG (74) Others (42)

Budget in million BMBFEuro (36)

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 4,0

MDC Research Report 2016 285 ANNEX

Organizational Structure

Supervisory Board

The Supervisory Board is the supervisory commit- Members tee of the MDC and is responsible for monitoring the Prof. Dr. Peter-André Alt Freie Universität Berlin determineslegality, effectiveness, which decisions and economic of the Board efficiency of Directors of the requiremanagement the prior of all approval the activities of the of Supervisorythe corporation. Board It Prof. Dr. Andreas Busch and may issue instructions about important matters Bayer Pharmaceuticals

affairs of the corporation. Prof. Dr. Patrick Cramer involving political aspects of the research and financial The Supervisory Board consists of no more than twelve Prof.Max Planck Dr. Horst Institute Domdey for Biophysical Chemistry representatives from the federal government, a re BioM GmbH presentativemembers. Its of membersthe State of include, Berlin, amongthe chairman others: of twothe - MinR Dr. Jan Grapentin boldt Universität zu Berlin and the Freie Universität Federal Ministry of Education and Research Berlin,Scientific and Advisory two MDC Board, representatives. the Presidents of the Hum Prof. Dr. Annette Grüters-Kieslich Charité-Universitätsmedizin Berlin Chairperson Prof. Dr. Sabine Kunst MinDir’in Bärbel Brumme-Bothe Humboldt-Universität zu Berlin Federal Ministry of Education and Research Prof. Dr. Hannah Monyer Deutsches Krebsforschungszentum Vice-Chairperson Prof. Dr. Claus Scheidereit Dr. Hans Reckers Max-Delbrück-Center for Molecular Medicine Senat Chancellery, Berlin Dr. Jana Wolf Max-Delbrück-Center for Molecular Medicine

286 MDC Research Report 2016 ANNEX

Scientific Advisory Board

Chairperson

Prof. Dr. Patrick Cramer

Max Planck Institute for Biophysical Chemistry Members

Prof. Dr. Federico Caligaris-Cappio

Prof.University Dr. Stefanie Scientific Dimmeler Institute San Raffaele, Milano Goethe-University Frankfurt

Prof. Dr. Eckart D. Gundelfinger

Prof.Leibniz Dr. Institute Nancy Hynes for Neurobiology

Prof.Friedrich Dr. Elisa Miescher Izaurralde Institute for Biomedical Research

Prof.Max Planck Dr. Kerstin Institute Krieglstein for Developmental Biology University of Freiburg

Prof. Dr. Nine Knoers University Medical Centre Utrecht

Prof. Dr. Anna Tramontano † Sapienza University of Rome

* As of 31.12.2016

MDC Research Report 2016 287 ANNEX

Organization Chart of the MAX DELBRÜCK CENTER FOR MOLECULAR MEDICINE in the Helmholtz Association Corporation under public law

Scientific Advisory Board Supervisory Board Representatives & Mediators Chair Chair Animal Welfare Officers: Nadja Daberkow-Nitsche, Claudia Gösele, Prof. Dr. Patrick Cramer Bärbel Brumme-Bothe (Fed. Min. Education and Research) Karin Jacobi, Boris Jerchow, Anne Zintzsch (MPI for Biophysical Chemistry) Anti-Corruption Officer: Ina Herrmann Data Protection Officer: Ulrike Ohnesorge Disabilities Officer: Marina Winterfeld Staff Council Hazardous Goods Officer: Yvette Wefeld-Neuenfeld Scientific Council Board of Directors Dagmar Gerhard Risk Management Officer: Katrin Rosswog Chair IT Security Officer: Dr. Torsten Johr Prof. Dr. Gary Lewin Chair, Scientific Director Administrative Director Women’s Representative Research Ombudsman: Prof. Dr. Jens Reich Prof. Dr. Martin Lohse Dr. Heike Wolke Dr. Christiane Nolte Ph.D. Ombudspersons: Dr. Daniela Panáková, Prof. Dr. Matthias Selbach

Science Administration Administration and Infrastructure

Cardiovascular and Cancer Disorders of the Berlin Institute for Experimental and Clinical Scientific Coordination Administrative Coordination Metabolic Disease Research Nervous System Medical Systems Biology Research Center (with Charité) Dr. Gwendolyn Billig Dana Lafuente Katrin Rosswog Coordinator: Coordinator: Coordinator: Coordinator: Director: Dr. Nuria Cerdá-Esteban Dr. Gesa Schäfer Stefanie Rach Prof. Dr. Norbert Hübner Prof. Dr. Claus Scheidereit Prof. Dr. Carmen Birchmeier Prof. Dr. Nikolaus Rajewsky Prof. Dr. Friedrich C. Luft

Management: Management: Management: Management: Head of administration: Personnel Dr. Kathrin Saar Dr. Michael Hinz Dr. Michael Strehle Dr. Jutta Steinkötter Dr. Regina Jünger (Charité) Communications Vera Glaßer (interim) Dr. Barbara Wiskow (interim) Program-Management: Prof. Dr. Michael Bader Prof. Dr. Walter Birchmeier Prof. Dr. Carmen Birchmeier Dr. Marina Chekulaeva Dr. Cornelia Maurer Dr. Martin Falcke Prof. Dr. Thomas Blankenstein Dr. Mina Gouti Dr. Jan Philipp Junker Finances Prof. Dr. Holger Gerhardt Prof. Dr. Peter Daniel Dr. Annette Hammes-Lewin (PF) Dr. Stefan Kempa Animal Facilities PD Dr. Jens Fielitz Andreas Feldo Prof. Dr. Michael Gotthardt Prof. Dr. Oliver Daumke Prof. Dr. Thomas Jentsch (cr) Dr. Markus Landthaler Prof. Dr. Maik Gollasch Dr. Karin Jacobi Prof. Dr. Norbert Hübner Dr. Michela Di Virgilio Prof. Dr. Helmut Kettenmann Prof. Dr. Irmtraud Meyer Prof. Dr. Ralph Kettritz Dr. Zsuzsanna Izsvák Dr. Gaetano Gargiulo Prof. Dr. Gary Lewin Prof. Dr. Uwe Ohler PD Dr. Sabine Klaassen Prof. Dr. Thomas Jentsch (cr) Prof. Dr. Udo Heinemann Dr. James Poulet Prof. Dr. Ana Pombo Internal Audit Prof. Dr. Friedrich C. Luft (cr) Uta Brinker Dr. Enno Klußmann (PF) Dr. Uta Höpken (PF) Prof. Dr. Fritz G. Rathjen Dr. Stephan Preibisch Research Funding Prof. Dr. Dominik N. Müller (cr) / Kirstin Müller Prof. Dr. Young-Ae Lee Dr. Ralf Kühn Dr. Dietmar Schmitz Prof. Dr. Nikolaus Rajewsky Dr. Ralf Dechend Dr. Ioannis Legouras Prof. Dr. Friedrich C. Luft (cr) Prof. Dr. Achim Leutz Dr. Björn Schröder Dr. Roland Schwarz PD Dr. Il-Kang Na Prof. Dr. Ingo L. Morano Dr. Stephan Mathas / Prof. Dr. Matthias Selbach (cr) Prof. Dr. Matthias Selbach (cr) Prof. Friedemann Paul Funding Resources Management DZHK Prof. Dr. Dominik N. Müller (cr) Dr. Martin Janz Prof. Dr. Erich Wanker Dr. Baris Tursun PD Dr. Adrian Schreiber Prof. Dr. Thoralf Niendorf Prof. Dr. Antonio Pezzutto Dr. Niccolò Zampieri Dr. Jana Wolf (cr) Magdalena Milsch-Hetzig Prof. Dr. Jeanette Schulz-Menger International Office Dr. Daniela Panáková Prof. Dr. Klaus Rajewsky Dr. Robert Zinzen PD Dr. Volker Siffrin Prof. Dr. Tobias Pischon Dr. Armin Rehm (PF) Dr. Oksana Seumenicht Prof. Dr. Simone Spuler IT Dr. Matthew Poy Dr. Oliver Rocks Prof. Dr. Ulrike Stein (cr) Prof. Dr. Kai Schmidt-Ott Prof. Dr. Claus Scheidereit Dr. Alf Wachsmann Prof. Dr. Matthias Selbach (cr) Prof. Dr. Clemens Schmitt Graduate Office Dr. Francesca Spagnoli Prof. Dr. Matthias Selbach (cr) Library Prof. Dr. Matthias Treier Prof. Dr. Michael Sieweke Dr. Michaela Herzig Prof. Dr. Thomas Willnow Prof. Dr. Thomas Sommer Dr. Dorothea Busjahn Prof. Dr. Ulrike Stein (cr) Prof. Dr. Wolfgang Uckert Technical Facility Management Dr. Jana Wolf Biological and Laboratory Safety Construction: Ralf Streckwall Dr. Christian Klein Operation: Michael Arnold

Purchasing Dr. Carsten Oehme Core Facilities Central Services Advanced Light Microscopy Light Microscopy Electron Microscopy iPS Magnetic Resonance Genomics Rüdiger Mehlhase Dr. Anje Sporbert Dr. Andrew Woehler Dr. Bettina Purfürst Dr. Sebastian Diecke Prof. Thoralf Niendorf Dr. Sascha Sauer

Mass Spectrometry Pathophysiology Preparative Flowcytometry Transgenics Bioinformatics Proteomics / Metabolomics Legal Affairs and Technology Transfer N.N. Dr. Arnd Heuser Dr. Hans-Peter Rahn Dr. Ralf Kühn Dr. Altuna Akalin Dr. Stefan Kempa Dr. Claudia Thurow

Legend cr = cross-reference PF = permanent fellow Issue: December, 2016 Mailing address: Robert-Rössle-Strasse 10, D-13125, Berlin www.mdc-berlin.de/organigram-E

288 MDC Research Report 2016 ANNEX

Organization Chart of the MAX DELBRÜCK CENTER FOR MOLECULAR MEDICINE in the Helmholtz Association Corporation under public law

Science Administration Administration and Infrastructure

Legend cr = cross-reference PF = permanent fellow Issue: December, 2016 Mailing address: Robert-Rössle-Strasse 10, D-13125, Berlin www.mdc-berlin.de/organigram-E

MDC Research Report 2016 289 ANNEX

Index

A Bähring, Sylvia �� 39 A Pontes de Oliveira, Kivia ��117 Bahry, Ella ��199 Abdelwahab, Ahmed �� 43 Bakina, Olya �� 17 Adami, Eleonora �� 21 Balasubramanian, Sandhya �� Adler, Carolin �� 53 Balasubramaniyam -Sivanandam, Arasu �� Afanasieva, Evegenia ��203 Balogh, Andras ��21789 Afzal, Shoaib ��161 �� 9587 Ahmed, Mohammed ��167 Banerjee, Anne ��161 Akalin, Altuna �� 175, 241 Bangalore,Banchenko, Megha Sofia ��7 Akcay, Duygu �� 67 Barbieri, Mariano ��197 Akilli-Öztürk, Özlem �� 79 Barda, Kathleen ��153 Akram Kamal Ghobrial, Monica �� 43 Bartels-Klein, Eireen �� 13 Aktar, Rozina ��137 Bartelt, Claudia �� 53 Albrecht, Sebastian ��145 Bartelt, Franziska �� Alenina, Natalia ��7 Barth, Carolin ��113 Alles, Jonathan ��203 Bartolini, Laura �� 4987 Alvares de Lemos, Clara Faria ��231 Bartolomaeus, Hendrik ��217 Alves de Melo, Arthur �� Bashammakh, Saleh ��7 Alves, Daniele ��7 Bashir, Sanum �� 99 Amiri, Behnam ��879 Bastos de Oliveira, Marta �� 13 Anders, Juliane ��217 Bauch, Sophia ��193 Anders, Kathleen �� Bauer, Christof ��111 Andrade Cabrera, Santiago Patricio �� 13 Bauer, Daniel �� 79 Andreani, Matteo �� 83 Baum, Katharina ��129 Andrée, Christel �� 39 Baumann, Matti ��113 Andrich, Kathrin ��16789 Baytek, Gülkiz ��261 Anell Rendon, Damaris ��119 Beagrie, Robert A. ��197 Angelov, Filip ��139 Beaudette, Patrick �� 117, 261 Anistan, Yoland-Marie ��219 �� Annibale, Paolo �� 35 Becker, Katja ��265 Anton, Selma �� 35 Becker, MikeIsabell ��837 Anumala, Upendra ��243 Begay-Müller, Valerie ��153 Anwar, Rabia �� 25 Beis, Daniel ��7 ��7, 27, 63 Bel Abed, Hassen ��243 Araragi, Naozumi ��7 Belenki, Dimitri ��119 Arlt,Apostel-Krause, Birte �� Iris 263 Bellmann-Strobl, Judith ��225 Arsie, Roberto �� Belz, Stefanie ��229 Arsuaga Azpiroz, Maddi ��111 Benary, Uwe ��129 Arumughan, Anup ��167183 Benlasfer, Ouidad �� Asaro, Antonino �� 71 Berberich, Simon ��167 Aschmann, Mehrshad �� 67 Berger, Maren ��125183 Asseyer, Susanna ��225 Bergmann, Janin ��155 Aumann, Jutta ��231 Bernert, Carola �� Aydin, Atakan �� 39 Berruyer, Carole ��121 Ayoub, Salah ��203 Bielow, Chris ��26387 Bikou, Maria �� 21 Bilbao, Carlos �� 49 B �� 25 Bach, Karola ��161 Birchmeier-Kohler, Carmen ��137 Backhaus, Carolin ��145 Birchmeier,Bilic, Ilija Walter �� 79 Bader, Michael ��7 Bishop, Christopher �� 55 Bahlke, Yvonne �� 47 Bittner, Aitomi ��119

290 MDC Research Report 2016 ANNEX

Bittner, Marie-Sophie ��139 Chama, Laura ��7 Blachut, Susanne �� 21 Chamakuri, Nagalah ��9 Blankenstein, Lydia ��117 Chekulaeva, Marina ��177 Blankenstein, Thomas �� Chen, Jie-Shin ��137 Blaszkiewicz, Jedrzej ��153 Chen, Jieqi ��149 Blaszyk, Edyta ��22783 Chen, Wei ��247 Blessing, Anja ��145 �� Block, Franziska �� 67 Cheret, Cyril ��137 Bluhm, Julia ��111 Chilkunda,Chen, Xiaojing Shreevathsa Srinivasan �� 2583 Blum, Maximilian �� 39 Choi, Mira ��221 Bobrov, Evgeny ��155 Christ, Annabel �� 71 Bock, Andreas �� 35 Chu, Van Trung ��109 Bock, Tobias �� Cinar, Özcan ��105 Böddrich, Annett ��167 Ciolli Mattioli, Camilla ��177 Boettger, Adelheid ��877 Cirovic, Branko ��103 Bogdanow, Boris �� 61 Clauß, Julian ��127 Boiani, Sara ��105 Cloos, Birgit �� Bolz, Svenja ��167 ��117 Bonkowski, Jasmin �� Collins, Russell Thomas ��87, 95, 16113 Börding, Teresa �� 97 Coin,Company Irene Nevado, Juan Carlos �� 91 Boschmann, Michael ��89, 215169 Conill-Cortés, Mercè ��261 Boss, Philipp ��193 Coralluzzo, Violeta �� 49 Botoni, Fernando ��7 Costa, Aline ��7 Bouguerne, Lamia ��157 Costa, Amanda ��149 Brand, Franziska �� 97 Costanza, Mariantonia ��223 Brandenburg, Manuela �� 237, 245, 261 Coxam, Baptiste Claude �� 13 Brandt, Alexander ��225 Cozzitorto, Corinna �� 63 Brandt, Bettina ��137 Cruz e Silva, Andreia ��145 Braunschweig, Maria ��153 �� 49 Breimann, Laura ��199 Czajkowski, Maciej ��113 Brischetto, Cristina ��117 Csályi, Kitti Dóra Bröhl, Dominique ��137 Brouwer-Lehmitz, Antje Susan �� 67 D Brümmer, Juliane ��113 da Costa Goncalves, Susanne ��7 Brusendorf, Lydia ��167 Daher, Hannes ��7 Buchert, Sven ��137 Dahlmann, Mathias ��231 Bukowiecki, Raul ��157 Dalda, Malgorzata Anna �� 25 Bulut, Selman ��207 Danner, Eric �� 99 Bunatyan, Lena �� 71 Daubitz, Tony ��145 Bunse, Mario ��127 Dauksaite, Vita �� 17 Buntru, Alexander ��167 Daumke, Oliver �� Burbach, Nadine ��119 de Almeida Sassi, Felipe ��149 Burkert, Christian Martin ��193 de Biase, Maria Stella ��20587 Busch, Clara ��121 de Laval, Bérengère ��121 Busch, Jana ��109 de Souza, Valeria ��7 Busse, Antonia ��105 Deak, Veronika �� 27 Busse, Dorothea ��129 Dechend, Ralf ��217 Buttke, Kornelia ��219 Deckert, Annika ��167 Delgado Benito, Veronica �� Dema, Alessandro �� 27 C Denker, Sophy ��11989 Caglio, Giulia ��197 Derudder, Emmanuel ��109 Cai, Huiqiang �� 25 Derz, Claus ��53, 239 Cakmak, Nese ��105 Deter, Sabrina �� Calviello, Lorenzo ��193 Devaraj, Anantharam �� 25 Camargo, Nutabi ��163 Dewitz, Carola ��175, 183, 241, 247, 169263 Canali, Magdalena ��121 Dhamodaran, Arunraj �� Cano Rincon, Elena �� 13 Dhandapani, Priyavathi ��7 Carlo-Spiewok, Anne-Sophie �� 71 Di Virgilio, Michela �� 83 Carta, Mario ��155 Dias, João Rodrigo ��197 Carter, David �� 95 Dick, Alexej �� 89 Cathou, Wilfried ��121 Diecke, Sebastian ��257 Cavalcante, Paula ��7 Dierolf, Nea ��16787 Cernoch, Janine ��109 Dietrich, Stephan ��169 Chaimowicz, Charlotte ��137 Diez, Lisa Maria ��167

MDC Research Report 2016 291 ANNEX

Dinter, Franziska ��157 Franke, Vedran �� 175, 241 Dittmar, Gunnar ��261 Franklin, Andrew ��109 Dolati, Setareh ��9 Frensemeier, Kristin ��113 Domaing, Petra �� 43 Fricke, Marie �� Dong, Le ��149 Friedhoff, Nicolai ��9 Donner, Christin ��167 Friedrich, Dhana ��19987 Donnimath, Neeta ��105 Friese, Christian �� Dornblut, Tracy Alice �� 95 Fritsche, Raphaela ��263 Dörr, Dorothea ��103 Fritschi, Simone ��22783 dos Santos Periquito, Joao �� 47 Froese, Sabine ��61, 129 Dreyer, Anne-Kathrin �� 99 Fröhlich, Janine �� 17 Driesner, Madlen ��161 Fröhlich, Jonathan ��203 Duchene, Johan ��7 Fuchs, Karolin ��145 Dudaniec, Krystyna ��127 Funk, Stephanie ��227 Duft, Karolin �� 43 �� 141, 153 Dumoulin, Alexandre ��161 Dunham, Hannah ��109 Fürst, Carina Iris Dusek, Petr �� 47 G Gabrysch, Julian ��127 Gamal Mohamed El Said, Mariam �� 43 E Ganichkin, Oleg �� Ebert, Madlen �� 67 Garcia-Perez, Angelica �� 25 Eccles, Rebecca ��113 Garg, Ankur �� 9587 ��127 Gargiulo, Gaetano �� 91 Eichhorn, Barbara �� 53 Gärtner, Angelika �� Edes,Eigenbrod, Inan Ole ��153 Gärtner, Carolin �� 43 El-Khalili, Alina ��207 Gätjen, Marcel ��11183 Elger, Deborah ��145 Gauert, Anton �� 55 Els, Antje �� 47 �� Erdem, Müge ��231 Gebala, Véronique �� 13 Erdmann, Beate �� 47 Geelhaar,Gavvovidis, Andrea Ioannis �� 2783 Eroshok, Ekatarina ��193 Geisberger, Sabrina ��217 Escobar, Helena ��229 Geiser, Karin ��121 Escot, Anne Sophie �� 63 Geissenberger, Sabrina ��263 Esparza Gordillo, Jorge �� 31 Genehr, Carolin ��127 Eulenberg-Gustavus, Claudia ��221 Gerhard, Cathrin ��7 Gerhard, Mathias �� 21 Gerhard, Ute ��217 F Gerhardt, Holger �� 13 Faber, Allison �� 21 Gerlach, Kerstin ��111 Fälber, Katja �� Gerloff, Michael ��103 Falcke, Martin ��9 Ghanbari, Mahsa ��193 Fan, Dorothy N. ��11987 Gibson, Meino Alexandra ��149 Fan, Gang ��219 Gilbert, Stephen ��9 Fang, Liang �� 79 Giminez, Gregory ��121 Farsi, Zohreh ��253 Glahs, Alexander ��209 Fatima, Nida ul ��207 Glazar, Petar ��203 Faustino Martins, Jorge Miguel ��139 Glöde, Julia �� 53 Favret, Jeremy �� 109, 121 Gloger, Marleen ��111 �� 49 Gödde, Kathrin ��145 �� 53 Gohlke, Ulrich �� 95 Fechner,Fendler, AnnikaInes �� 79 Gök, Caglar ��153 Feinkohl, Insa ��243 Gokhale, Sucheta Arun ��193 Ferrai, Carmelo ��197 Goldbrich-Hannig, Beate �� 17 Ferrarese,Fernández LeironBachiller, �� Isabel 155 Golic, Michaela ��217 Ferri Blázquez, Alba ��125 Gollasch, Maik ��219 Filipchyk, Andrei ��203 Golusik, Sabrina ��167 Fink, Claudia �� 17 Gomez, Lorena ��7 Fischer, Cornelius ��247 Gonzales Lopez, Adrian �� 47 Fischer, Jan-Frederik ��149 Gonzalez-Hirschfeld, Simon ��137 Fishman, Veniamin ��7 Göppner, Corinna ��145 Flachmeier, Christina �� 31 Göritz, Petra ��145 Fleischer, Raluca ��153 Gosdschan, Alexander ��175 Forbes, Martin ��263 Gotthardt, Michael �� 17 Förster, Susann ��103 Gouti, Mina ��139

292 MDC Research Report 2016 ANNEX

Govindarajan, Thirupugal �� 17 Heinze, Daniel ��117 Graband, Annika �� 97 Heinze, Lea ��193 Graf, Robin ��109 Hellmig, Nicole ��105 Graf, Yasmine ��7 Henning, Lisa Kimberly �� Gräning, Kornelia ��229 Henning, Mechthild ��161 Grelle, Gerlinde ��167 Hergeselle, Martha �� 6183 Grieben, Marlies ��113 Hernandez Miranda, Luis Rodrigo ��137 Grieben, Ulrike ��229 Herrmann, Pia ��231 Gries, Margarete ��105 Herse, Florian ��217 Griffel, Carola ��137 Herzog, Margareta ��203 Griger, Joscha ��137 Herzog, Sergej ��207 Grigoryan, Tamara �� 79 Hesse, Christin �� 127, 167 Grimmer, Benjamin ��7 Hesse, Evelyn �� Grinat, Johanna �� 79 Hessenberger, Manuel �� Grobe, Jenny ��263 Hetsch, Florian ��16183 Groenke, Nicole ��167 Heuberger, Julian �� 7987 Grohmann, Konstantin ��237 Heuser, Arnd ��255 Grosche, Sarah �� 31 Heynen, Gustaaf �� 79 Große, Claudia ��109 Hilgenberg, Ellen ��127 Grosswendt, Stephanie ��203 Hill, Undine ��103 Grueger, Sabine ��7 Hinz, Michael ��117 Grunert, Corinna ��105 Hinz, Tanja �� 53 Grunwald, Stefanie ��229 Hinze, Christian �� 57 Grunwald, Stephan �� Hinze, Florian �� 17 Grunz, Katharina �� 21 Hirsekorn, Antje �� Grushevskyi, Yevgenii �� 3587 Hödke, Daniela �� 53 Grybowski, Andrea ��199 Hoegg-Beiler, Maja ��181, 145193 Guedes Simões, Mariana �� 49 Hoffmann, Diana ��155 Gueneykaya, Dilansu ��149 Hofstätter, Maria ��103 Guimaraes Silva, Ana Luisa Ferreira ��209 Hohnstädter, Franzy �� 53 Günther, Anika ��145 Holland, Jane �� 79 Günther, Sebastian ��261 Holler, Karoline ��179 �� 13 Hönig, Katrin �� H Hollfinger,Höpken, Uta Irene E. �� 97 Haage, Verena ��149 Horatscheck, André ��24383 Haase, Hannelore �� 43 Hörl, David ��199 �� 53 Horn, Sabrina �� Haase, Nadine ��217 Hoser, Dana �� Hafez,Haase, DinaInes ��193 Hsu, Meng-Tung �� 83 Hahmann, Anne ��7, 167 Hu, Bo ��17983 Hainer, Cornelia ��7 ��13983 Haink, Petra �� 49, 57, 61, 129 Hübner, Norbert �� 21 Halbeisen, Paula �� 57 Hube,Hügel, IrisStefanie Caroline ��7 Hall, James ��153 Hülnhagen, Till �� 47 Hallscheidt, Eileen Maren �� 27 Hummel, Franziska ��223 Hamann, Wiebke �� 17 Hummel, Kordelia ��127 Hammes-Lewin, Annette �� 141, 153 Hummel, Oliver �� 21 Hänig, Christian ��167 Humpenöder, Melanie ��167 ��197 Hüttemeister, Judith �� 17 Hardt, Markus ��231 Haseleu,Harabula, Julia Izabela ��153 Hasenkopf, Regine ��167 I Hasselbach, Andrea �� 47 ��193 Hauchwitz, Janina ��127 ��149 Haustein, Maria ��231 Ibrahim, Mahmoud ��231 Hayat, Ortal ��261 Ifuku, Masataka �� 61 Hazapis, Orsalia �� Ilm, Katharina ��231 He, Mei ��207 Imami, Koshi ��157 Heidler, Denise ��157183 Imbastari, Francesca ��117 Heidt, Leonie ��229 Inak, Gizem �� 35 Heinemann, Udo �� 95 Ipenberg, Inbal ��203 Heinemann, Vera ��149 Isbilir, Ali ��103 Heinig, Kristina �� 97 Ivanov, Andranik ��257 Heinig, Matthias �� 21 Ivanovska, Jelena �� 25 Iwanska, Anna Izsvák, Zsuzsanna MDC Research Report 2016 293 ANNEX

Keller, Lisa �� J Kemmner, Wolfgang ��231 Jabs, Sabrina ��145 Kemna, Christoph ��12789 Jacoby, Ben �� 53 Kempa, Stefan ��263 Jaeschke, Lina �� 53 Kempfer, Rieke ��197 Jaksch, Sarah ��103 Kettenmann, Helmut ��149 Janke, Jürgen ��53, 239 Kettritz, Ralph ��221 ��167 Keyner, Daniela ��91, 117 Janz, Martin ��223 Kieback, Elisa �� Jarchow,Jansen, Isabelle Birgit ��149 Kienitz, Tina ��7 Jarosch, Ernst ��125 Kipar, Claudia ��83, 127209 Järve, Anne ��7 Kirchgraber, Gabriel Simon �� 57 Jedamzick, Johanna ��145 Kirwan, Jennifer ��251 Jens, Marvin ��203 Klahn, Sylvia �� Jentsch, Thomas J. ��145 Klasen, Christian �� 67 Jeran, Stephanie �� 53 Klaus-Bergmann, Alexandra ��83,13, 125 79 Jerke, Uwe ��221 Klaus, Loni ��167 Ji, Yiyi �� 47 Kleckers, Daniela ��167 Jia, Shiqi ��137 Kleindienst, Denise ��227 Jin, Sha ��167 Kleissle, Sabrina ��243 Johann, Lisa ��7 Klement, Christine ��103 Jordan, Philipp ��149 Klempin, Friederike ��7 Jouhanneau, Jean-Sebastien ��155 �� 25 Ju, Hyoungseok �� 71 Klippert, Julia �� 71 Judis, Carmen-Anisia �� 17 Klinge,Klironomos, Isabel Filippos ��203 Juneja, Manisha ��231 Klix, Sabrina �� 47 Jung, Simone �� 79 Klockmeier, Konrad ��167 Junker, Jan Philipp ��179 Klußmann, Enno �� 27 Jüttner, Rene ��161 Knape, Sara ��125 Knespel, Andreas �� 95 Knispel, Rosita ��47, 249 K Knoblich, Maria ��103 Kaada, Manuela �� 43 Knurr, Genevieve �� Kabisch, Rainer ��265 ��167 Kabrani, Eleni ��109 Kobelt, Dennis ��23183 Kabuss, Loreen-Claudine ��139 Kocaoglu,Ko, Young-In Feraye �� 17, 55, 99 Kahlert, Günther ��261 Koch, Katharina Sarah �� 43 Kalb, Birgit �� 31 Kocks, Christine ��203 Kaldjob, Louise ��7 Koehler-Rohde, Annette ��227 Kaldrack, Joanna �� 17 Kofahl, Bente ��129 Kallfaß, Carsten �� 67 Kofent, Julia �� 63 ��217 Köhler, Anett ��125 Kammerer, Sarah �� 79 Köhler, Anja ��109 Kammertöns,Kamer, Ilona Thomas �� Köhler, May-Britt ��217 Kampf, Kristin �� 71 Kolesnichenko, Marina ��117 Kampfrath, Branka �� 9783 Kolinski, Marcin �� Kanashova, Tamara ��261 Kolundzic, Ena ��207 Kannan, Preethi ��157 Kolyvanov, Klim ��199183 Kanter, Marit ��53, 239 Komleva, Yulia ��219 Kanter, Sophia ��153 Kondrashkina, Aleksandra �� 25 Kapoor, Varun ��199 Kong, Weiying ��219 Karabacak, Aslihan ��193 König, Janett ��155 Karaiskos, Nikolaos ��203 Konigorski, Stefan �� 53 Karamatskos, Karin �� Konrath, Fabian ��129 Karber, Mirjam �� 39 Köppen, Susanne ��167 Karczewski, Karin �� 4383 Korthals, Stefanie ��149 Kärgel, Eva ��117 Korvers, Laura Maria ��149 Kaßmann, Mario ��219 Kosel, Frauke �� 79 Kassner, Celina ��157 Kosizki, Liana ��153 Kassuhn, Wanja ��193 Kosten, Tetiana ��153 Kastelic, Nicolai �� Kostka, Susanne ��167 Kathrin, Saar, ��3 Kostova, Simona ��167 Kazmierczak, Marlon ��207183 Kowenz-Leutz, Elisabeth ��103 Keil, Albrecht ��7 Kox, Stefanie �� 47 Keiper, Regina �� 193, 197 Kraemer, Dorothee ��197

294 MDC Research Report 2016 ANNEX

��117 Leschke, Andrea ��265 Kräker, Kristin ��217 Leu, Romy �� 103, 243 Krahn,Krannich, Inge Alexander ��215 Leutz, Achim ��103 Krause, Christine ��265 Leva, Tobias ��155 Kremkow, Jens ��155 Lewin, Gary ��153 Krenz, Lisa �� 47 Lewin, Rahel ��261 Krenzin, Stephanie �� 53 Li, Shuang �� 109, 223 Kressin, Franziska ��153 ��219 Kreßner, Susanne ��125 Liebold, Janet ��145 Kretschel, Kerstin ��227 Liekweg,Lian, Xiaoming Melanie ��71, 91 Kreuchwig, Franziska �� 21 Liemen, Michaela �� 193, 197 Krieger, Karsten ��117 Lietz, Maria ��127 Krivokharchenko, Alexander ��7 Lindemann, Peter ��243 �� 177, 197, 205, 207, 209 Lips, Christian ��125 Krönke, Nicole ��145 Liskovykh, Mikhail ��7 Krüger,Krock, Ines Katrin �� 53 Lisowski, Pawel ��157 Krüger, Kerstin �� 97 Liss, Martin �� 17 Krüger, Nadine ��215 Liu, Na ��7 Krüger, Norman ��257 ��177 Krüger, Sabrina ��209 ��263 Krüger, Sylvia ��221 Lohse,Loedige, Martin Inga �� 35 Kruse, Christine �� 71 Löffler,Loof, Gesa Julia ��197 Kruse, Sabrina ��167 Lopez, Natalia �� 25 Ku, Min-Chi �� 47 Lorenz, Carmen ��157 Kuchling, Joseph ��225 Lorenz, Felix ��127 Kuchta, Johanna �� 99 Lorenzen, Theodor �� 49 Kuczmanski, Artur ��219 Lübcke, Susan ��197 Kuhle, Verona �� 71 Lück, Jennifer ��145 Kühn, Ralf ��99, 265 Ludwig, Carmen ��145 Kühnemund, Johannes ��153 Luft, Friedrich C. ��39, 215, 221 Kuich, Henning ��263 Luganskaja, Tatjana �� 57 Kukalev, Alexander ��197 Lundberg, Felix �� 25 Kumar Kandalla, Prashanth ��121 Lusatis, Simone ��223 Kumar, Amit ��263 Lutter, Darius ��145 Kunz, Severine ��245 Lutter, Steffen ��43, 53 �� 95 Kurt, Zeynep �� 95 Kurt,Kurths, Mehmet Silke �� Hanifi 95 M Kusnierz, Marie-Kristin �� 53 Maaß, Philipp �� 39 Kutz, Kamila �� 61 Maatz, Henrike �� 21 Mackowiak, Sebastian ��203 Magg, Andreas ��7 L Maghsodi, Sara ��7 Labi, Verena ��109 Mähler, Anja ��215 Lacadie, Scott �� Mahmoodzadeh, Shokoufeh �� 43 ��137 �� 35 Lan, Chen ��181, 219193 Maier, Luisa ��203 Lan,Lahmann, Linxiang Ines �� 79 Malcher,Maiellaro, Jakub Isabella ��229 Landthaler, Markus �� Malchin, Victoria ��103 Langanki, Reika ��7 Malecki, Piotr Henryk �� 95 Lange, Doris ��117183 Malik, Anna �� 71 Langer, Henning �� 229, 263 Malinsky, Fernanda ��7 Langhorst, Hanna ��161 Mall, Sabine �� 53 ��7 Mallis, Lisa ��7 Larcheveque, Oriane �� 43 Mallow, Keven ��7, 243 Larrosa,Lapidous, Madeleine Irina ��137 Mannaa, Marwan ��219 Latz, Meike ��7 Manske, Melanie �� Lauhkonen-Seitz, Riikka ��109 Mansour, Fatma A. A. �� 57 Lauschke, Christina �� 53 Manz, Sabine ��26583 Lazarow, Katina ��243 Marcel, Yves ��217 Le Poul, Yann ��199 �� 31 Lee, Soyoung ��119 Marg, Andreas ��229 Lee, Young-Ae �� 31 Margineanu,Marenholz, Ingo Anca ��237 Leisegang, Matthias ��127 Mari, Tommaso �� 61 Leong, Megan ��157 Maricos, Meron ��149

MDC Research Report 2016 295 ANNEX

Marko, Lajos ��217 Moshourab, Rabih ��153 Markus, Carolin Elfriede �� 57 Mosienko, Valentina ��7 Marre, Stefan ��161 Mossadegh-Keller, Noushine ��121 Martos, Vera ��243 Mostafa, Mohamed Hassan �� 27 Masionyte, Milda ��103 Mothes, Janina ��129 Massimo, Carlo M. �� Motta, Daisy ��7 Maßwig, Sven ��119 Mücke, Michael Benedikt �� 21 Mastrobuoni, Guido ��263187 Mudduluru, Giridhar ��231 Matanovic, Anja �� 31 Muehlberg, Fabian ��227 Mathas, Stephan ��223 Mueller, Andrea ��7 Matiash, Vitali ��237 Mueller, Laura �� 63 Mattei, Daniele ��149 Mühl, Astrid �� 39 Matthäus, Claudia �� Muino Acuna, Jose Maria ��193 Matthes, Susann ��7 Mukherjee, Neelanjan ��193 Matyash, Marina ��87, 161149 Müller, Anita �� 21 Mauri, Marta ��177 Müller, Dominik N. ��217 Maurizio, Julien ��121 Müller, Marion �� 79 McCorkindale, Alexandra ��209 Müller, Nicola Victoria �� 49 Mecklenburg, Nora ��141 Müller, Paul Markus ��113 Mehling, Heidrun ��215 Müller, Thomas ��137 Meier, Katja �� 13 Münch, Jonas ��145 Meisel, Jutta ��217 Munteanu, Alina-Cristina ��193 Memczak, Sebastian ��203 Munteanu, Dan �� 175, 241 Memler, Charlene ��7, 155 Mustroph, Mandy ��125 Menzel, Lutz ��97, 111 Myszczyszyn, Adam �� 79 Menzel, Peter �� Merkert, Doris ��215 Merks, Anne Margarete ��17,187 49 N Messerschmidt, Clemens ��193 ��217 Metsger, Maria ��247 Nadler-Holly, Michal �� 61 Metzler, Eric ��229 Nagy,N’diaye, Enikö Gabriele Eva �� 25 Meußer, Birgit �� 39 Narayanavari, Suneel �� 25 Meyer-Liesener, Stephanie ��229 Naschke, Michaela ��127 Meyer, Alexander �� 49 Naumann, Heike �� 63 �� Nazaré, Marc ��243 Meyer, Katrina �� 61 Neto, Filipa �� 13 Meyer, MarkusIrmtraud ��203187 Neubert, Wilhelm ��9 Meyer, Niklas ��149 Neuendorf, Nancy ��167 Meyerhuber, Peter ��127 Neuendorf, Sandra �� 25 Michaelis, Edward �� 97 Neuenschwander, Martin ��243 Middelhoff, Ella ��153 Nguyen, Tam Hoai ��105 Mieth, Christin �� 21 Nguyen, Thuy Linh ��103 Miguel, Ana ��197 Nickel, Elke ��227 Miko, Henriette ��193 Nickel, Sophie ��219 Miksche, Sandra ��243 Nickl, Bernadette ��7 Mikuda, Nadine ��117 Niederlechner, Hannah ��167 Milanovic, Maja ��119 Niendorf, Thoralf ��47, 249 Mildner, Alexander ��103 Nimptsch, Katharina �� 53 Milek, Miha �� Nitschke, Norma ��145 Milenkovic-Zujko, Nevena ��155 Nitschke, Ute ��223 Milojkovic, Ana ��183 Nitschmann, Kathleen �� 53 Ming, Qianqian �� 95 Nitz, Katrin ��7 �� 83 Noel, Jeffrey �� Mitic, Nina ��179 Nolte, Christiane ��149 Mlody,Minia, IgorBarbara ��157183 Nowak, Marcel ��12587 Mohr, Till-Yong �� 31 Mohtashamdolatshahi, Azadeh ��161 Molawi, Kaaweh ��109 O �� 39 Obenaus, Matthias �� Möller, Jan �� 35 Oberacker, Eva �� 47 Molé Illas, Rosana �� 43 Oberheide, Karina ��14583 Moreno Garcia, Sarah �� 53 Obermayer, Benedikt �� Moreno-Ayala,Morano, Ingo Roberto ��179 Oder, Andreas ��243 Moroni, Mirko ��153 Ofenbauer, Andreas ��189, 207203 Morris, Kelly J. ��197 Ohler, Uwe ��193

296 MDC Research Report 2016 ANNEX

Olal, Daniel �� Polovitskaya, Mayya ��145 Oliveira Batista, Rogerio ��7 Polzin, Evelyn ��227 Oliveira, Aline ��877 Pombo, Ana ��197 Oliveira, Marta ��197 Poncette, Lucia �� Oliveras Martínez, Anna ��145 Poole, Kathryn ��153 Oloyo, Ahmed Kolade �� 43 Popova, Elena ��837 Omerbasic, Damir �� 61 Popp, Oliver ��261 Opialla, Tobias �� 229, 263 Poulet, James ��155 Oppermann, Johannes ��161 Poy, Matthew �� 55 ��145 Preibisch, Stephan ��199 Ortegon, Clara �� 43 Preissner, Robert ��215 Orozco,Osberg, IanBrendan �� 175, 241 Prigione, Alessandro ��157 Özerdem, Celal �� 47 Prochnov, Victoria �� 35 Pronina, Nina ��103 Prothmann, Marcel ��227 P Przygodda, Jessica ��243 Pahle, Jessica ��231 Purfürst, Bettina ��245 Pakula, Hubert �� 79 Palada, Vinko ��229 Panakova, Daniela ��17, 49 Q Pantzlaff, Tatjana �� 71 Qadri, Fatimunnisa ��7 Papafotiou, Georgios �� Qi, Jingjing �� 79 Papavasileiou, Panagiotis ��203 Quedenau, Claudia ��247 Pareja Alcaraz, Ruth ��35, 14583 Quinn, Hazel �� 79 Parente Fernandes, Joao Miguel �� 61 Paricio Montesinos, Ricardo ��155 ��243 R Päseler, Claudia ��137 Räbel, Katrin ��145 Pascual López-Alberca, �� María 175, 241 Rabelo, Luiza ��7 Patmanidi, Alexandra ��103 Radetzki, Silke ��243 Patone,Patarcic, Giannino Inga �� 21 Radhakrishnan, Harikrishnan ��231 Paul, Fabian Alexander ��141 Radke, Michael �� 17 Paul, Friedemann ��225 Radscheit, Kathrin �� 25 Paul, Katharina �� 47 Raghunathan, Vaishnavi �� 25 Paysen, Hendrik �� 47 Rahjouei, Ali �� Peeva, Polina ��7 Rahn, Gabriele ��215 Peguero-Sanchez, Esteban ��177 Rahn, Hans-Peter ��25989 Pempe, Jenniffer ��109 Rajewsky, Klaus ��109 Pereira Abreu, Ruben Lopes �� Rajewsky, Nikolaus ��203 Perepelittchenko, Lioudmila ��243 Rakova, Natalia �� 215, 217 Perets, Ekaterina ��183, 20927 Ramberger, Evelyn ��261 Perez Hernandez, Daniel ��261 Ramezani, Feryal �� 47 Petermann, Markus ��7 Ramirez Cuellar, Julieta ��197 Peters, Jonas ��193 Ranjan, Ashish ��7 Petkovic, Marina ��205 Raphael, Moritz �� 47 Petrovic, Sasa �� 95 Rasche, Ludwig ��225 Petsch, Kerstin ��109 Rasko, Tamas �� 25 Petzold, Stefanie �� Rathjen, Fritz ��161 Pezzutto, Antonio ��105 Rätzer, Jenny �� 67 Pham, Michael �� 4783 Rau, Florian ��155 Phan, Quang Vinh �� 91 Rau, Kirstin ��167 Philipp, Julia ��193 Rautenberg, Kirstin ��259 Philipp, Regina ��13, 179, 199 Reddy Vennapusa, Pavan Kumar �� 95 Pietzke, Matthias ��263 Redel, Alexandra ��167 ��169 Rehm, Armin ��111 Pinart Gilberga, Mariona �� 53 Reid, Anna ��207 Pimpinella,Pircher, Christian Sofia �� Reimann, Henning �� 47 Pisch, Erika ��167 Reimann, Maurice ��119 Pischon, Tobias ��53, 23983 Reincke, Momsen ��145 Piske, Regina ��149 Retzlaff, Kristin �� Pivovarov, Misha �� 47 Reznick, Jane ��153 Planells-Cases, Rosa ��145 Rharass, Tareck �� 4983 Plehm, Ralph ��7 Rhein, Simone ��105 Pluska, Lukas ��125 Riazy, Leili �� 47 Pohlmann, Andreas �� 47 Richter, Annelie �� 53

MDC Research Report 2016 297 ANNEX

Richter, Jana ��49, 179 Schmidt, Anne-Margarethe ��103 Richter, Matthias ��237 Schmidt, Felix ��225 Riepenhausen, Thorsten ��7 Schmidt, Karin �� Rintisch, Carola �� 21 Schmidt, Kristin ��109 Rito, Tiago M.C. ��197 Schmidt, Sabine �� 2183 Rocks, Oliver ��113 Schmidt, Vanessa �� 71 Rodriguez-Orejuela, Marta ��203 Schmitt, Clemens ��119 Rogowski, Karol �� Schmitt, Matthias Jürgen �� 91 Rohloff, Michael �� 47 Schmitz, Dietmar ��163 Rojas Rusak, Fabio Enrique ��18367 Schneider, Valentin �� 21 Rolle, Susanne ��221 Schnepel, Philipp ��155 Römer, Christine �� 25 Schnögl, Sigrid ��167 Ronen, Jonathan �� 175, 241 Scholz, Florian �� 97 Rösch, Marion �� Schön, Christian �� Roske, Yvette �� 95 Schönheit, Jörg ��103 Roßius, Jana �� 9983 Schöwel, Verena ��22983 Rothe, Adrienne ��229 Schrade, Katharina �� 27 Rothe, Michael �� Schräpel, Nadja ��167 Rothe, Victoria ��149 Schreiber, Adrian ��221 Rother, Franziska ��837 Schreiber, David ��265 Rousselle, Anthony ��7 Schrezenmeier, Eva Vanessa �� 57 Royla, Nadine ��263 Schriever, Cindy �� 43 Rudolph, Franziska �� 17 Schueler, Markus ��197 �� 71 Schulz-Menger, Jeanette ��227 Ruffault, Pierre-Louis ��137 Schulz, Christina ��49, 127 Ruffert,Rudolph, Janett Ina-Maria �� 57 Schulz, Julia ��103 ��7 Schulz, Maike �� 27 Rugor, Julia ��217 Schulz, Vivian ��137 RufinoRuhe, Larissa Estrela, �� Gabriel 177 Schumacher, Sophie ��225 Rüschendorf, Franz �� 21 Schunck, Wolf-Hagen �� 39 Ruzittu, Silvia �� 63 Schuster, Maja ��193 Rybak-Wolf, Agnieszka ��203 Schütz, Anja �� 95 Schütz, Laura ��161 Schütze, Sebastian ��145 S Schwabe, Daniel ��9 Saha, Sonali ��145 Schwaller, Frederick ��153 Sander, Sandrine ��109 Schwartz, Camille ��193 Santos, Robson ��7 Schwarz, Roland ��205 Saponaro, Marisa ��197 Schwede, Heike ��97, 111 Sarrazin, Sandrine ��121 Schwedhelm, Carolina �� 53 Sassi, Felipe ��169 Sczakiel, Henrike ��223 Sauer, Sascha ��247 Secker, Christopher ��167 Sauer, Sascha ��247 Seeger-Zografakis, Michaela ��149 Schächterle, Carolin �� 27 Seelk, Stefanie ��207 Schäfer, Sebastian �� 21 Seidler, Patrick ��145 Scharek, Nadine ��149 Seifert, Annika �� 79 Scheel, Michael ��225 Selbach, Matthias �� 61 Scheer, Anke ��153 Semtner, Marcus ��149 Scheidereit, Claus ��117 Serebreni, Leonid ��197 Schelenz, Stefanie ��255 Serradas Duarte, Teresa Silva �� 47 Schenck, Heike ��217 Serresi, Michela �� 91 Scheuplein, Vivian �� Servin-Vences, Martha Rocio ��153 Schiel, Christina ��245 Seyffarth, Carola ��243 Schiele, Phillip ��837 Sibilak, Sylvia ��103 Schihada, Hannes �� 35 Sieweke, Michael ��121 Schilling, Marcel ��203 Sihn, Gabin ��7 Schindler, Franziska ��167 Simon, Mareike ��129 Schlegel, Jeanette �� Sin, Celine �� 91 Schleich, Kolja ��119 Singh, Manvendra �� 25 Schleifenbaum, Johanna ��21987 Sippach, Lydia ��137 Schleußner, Nikolai ��223 Skarlatou, Sophie ��169 Schmeißer, Maria �� 71 Skorna-Nussbeck, Madeleine ��7 Schmid, Johanna ��139 Skourti-Stathaki, Konstantina ��197 Schmidt-Ott, Kai �� 57 Skroblyn, Tessa Lara �� 97 Schmidt-Ullrich, Ruth ��117 Slimak-Mastrobuoni, Marta ��197

298 MDC Research Report 2016 ANNEX

Smith, Janice ��231 Thiel, Denise �� 161 Thieme, Christoph ��197 Sohn, Madlen ��247 Thomas, Jane Joy ��18979 Solana-Garcia,Sohm, Ina Jordi ��203 Thränhardt, Heike ��153 Sommer, Christian �� 61 Thuß, Theresa �� 31 Sommer, Thomas ��125 Tischer, Janett �� 95 Sommermann, Thomas ��109 Todiras, Mihail ��7 Sommermeier, Anne ��207 Toepper, Agnieszka ��227 Sousa Rosa, Andre �� 13 Tomann, Philipp ��117 Spagnoli, Francesca �� 63 Torlai Triglia, Elena ��197 Spanjaard, Bastiaan ��179 Tran, Ngoc-Tung ��109 Spatt, Lisa ��113 Treese, Christoph ��231 Specker, Edgar ��243 Treier, Anna-Corina �� 67 Sporbert, Anje ��237 Treier, Mathias �� 67 Spuler, Simone ��229 Trepte, Philipp ��167 Stache, Vanessa �� 97 Triller, Paul ��149 Stallerow, Petra ��137 Trnka, Philipp ��113 Stangl, Oliver �� 47 Tröster, Philip ��161 Stärkel, Cornelia �� 153, 237, 245, 255, 259 Tschernycheff, Alex ��203 Stauber, Tobias ��145 Tschierske, Dana ��119 Stauch, Maren �� 21 Tseng, Yu-Ting �� 67 Stear, Jeffrey ��153 Tsvetkov, Dmitry ��219 Stefanov, Stefan �� Tufan, Ahmet Bugra ��117 Stein, Ulrike ��231 �� Steinbrecher, Astrid ��18753 Tursun, Baris ��207 Steinhof, Anne ��167 Tupiņa, Dagnija 87 Steiniger, Jochen ��215 Stendal, Manuela �� 53 U Stevanovic, Katarina �� 25 Uckert, Wolfgang ��127 Stoffels, Gritt ��225 Udhayachandran, Annapoorani ��155 Stottmeister, Christin ��203 Ugowski, Sarah ��209 Strauss, Anke ��215 �� 13 Strempel, Nadine ��167 Uhrig, Marco �� Strödicke, Martin ��167 Ullrich,Uhlenhut, Florian Ines ��145 Ströhl-Sattler, Anna-Maria �� 43 Ullrich, Katrin ��189, 223203 Stuhlmann, Till ��145 �� 13 Subkhangulova, Aygul �� 71 Uyar, Bora �� 175, 241 Suhr, Frank �� 43 Unterweger, Iris Sun, Chuanbo �� 25 Sundaravinayagam, Devakumar �� V Sünkel, Benjamin ��125 Valenti, Giovanni �� 79 Swinarski, Marie �� 4989 Valeske, Beate ��25, 139 Swolinsky, Jutta �� 57 van den Bruck, David ��177 Szczesniak, Michal ��193 van der Laan, Eveline �� 79 ��219 van Heesch, Sebastiaan �� 21 Szvetnik, Attila �� 25 van Vuuren, Laura �� 43 Szijarto,Szymborska-Mell, Istvan A. Anna Bozena �� 13 Vannauer, Marianne ��245 Vargas Aguilar, Stephanie ��121 Vasiljevic, Djordje �� 61 T Veauthier, Anette �� 53 Tamyra-Lesniak, Olga ��7 Verlaat, Lydia �� 97 Tano, Jean-Yves ��35, 219 Vestergaard, Mikkel ��155 Taube, Martin ��255 Viera, Carlos �� 61 Teffera, Timkehet ��137 Vierheller, Janine ��9 Teichmann, Karin �� 53 Vigolo, Emilia �� 57 Teixeira da Costa, Ana Rita ��263 Viol, Ricarda Merle �� 57 Teller, Janine �� 67 Vion, Anne-Clemence �� 13 Telugu, Narasimha Swamy ��257 Vladimirov, Nikita ��199 Tempelmeier, Anne ��167 Vogel, Margit ��245 Ter-Avetisyan, Gohar ��161 Vogel, Regina �� 79 Terne, Mandy ��137 Vogt, Janis ��145 Textor, Ana �� Voigt, Birgit ��155 Thakkar, Meghna Hiren �� 17 Volkwein, Corinna ��125 Theil, Kathrin ��20383 von Bock, Anyess ��145 Thiede, Katerina �� von Bühler, Erika ��127

83 MDC Research Report 2016 299 ANNEX

von Hoff, Linda ��117 Woischwill, Christiane �� 43 von Kries, Jens Peter ��243 Wolf, Jana ��129 Voss, Cynthia ��231 Wolf, Laura ��161 Voss, Felizia ��145 Wolf, Susanne A. ��149 Vucicevic, Dubravka �� Wolff, Janine ��7 �� 27 Wollert-Wulf, Brigitte ��223 Vvedenskaya, Olga ��181, 263193 Worsley Hunt, Christine Rebecca ��193 Vukićević, Tanja Wreczycka, Katarzyna �� 175, 241 Wszolek-Meißner, Karolina ��7 W Wührmann, Lara Lee �� Wahle, Philipp ��209 Würfel, Jens �� 47 Waiczies, Helmar �� 47 Wurmus, Ricardo �� 175, 24183 Waiczies, Sonia �� 47 Wurster, Kathrin ��223 Walcher, Jan ��153 Wyler, Emanuel �� Walde, Marie ��253 Walentin, Katharina �� 57 183 Walther, Wolfgang ��231 X Wang, Jichang �� 25 �� Wang, Ning ��219 Wang, Zhen �� 55 Xavier, Audrey 87 Wanker, Erich ��167 Y Wasinski, Frederick ��7 �� 55 Wasser, Martina �� 25 Yasuda, Tomoharu ��109 Watkins, Tom ��205 Yilmaz,Yan, Xin Zekiye Buket ��117 Weber, Annika ��125 �� 67 Weber, Timm ��109 Yong, Guo �� 25 Wefeld-Neuenfeld, Yvette �� 39 Yin,Yu, Yong Xiushan ��119 Wegener, Sandra ��257 Yue, Bin ��119 Weidlich, Andrea ��145 Yumlu, Saniye �� 99 Weinberger, Oliver �� 47 Yurtdas, Zeliha Yesim �� 57 Weinert, Stefanie ��145 Yusufujiangaili, Dilimulati �� 175, 241 Wendt, Maren ��149 Wendt, Stefan ��149 Wenz, Daniel �� 47 Z ��169 Zabiegalov, Oleksandr ��157 Werner, Miru Hans ��261 Zadora, Julia �� 25 Werner, SabineIsabelle �� 231 Wesolowski, Radoslaw ��103 Zampieri, Niccolò ��169 Wesselmann, Maike �� 5387 Zaniewska,Zaimenko, Inna Magdalena ��7 Wessels, Hans-Hermann ��193 Zappulo, Alessandra ��177 Westen, Christel �� Zarmstorff, Ruth ��103 Westermann, Joerg ��105 Zasada, Christin ��263 Westphal, Christina �� 3983 Zauber, Henrik �� 61 Wethmar, Klaus ��103 Zenkner, Martina ��167 Wetzel, Christiane ��153 Zerath Gurevich, Sylvia ��261 Wichner, Katharina �� 97 Zhang, Juan �� 63 Wiedemann, Franziska �� 157, 167 Zhou, Luyu ��7 Wiglenda, Thomas ��167 Zhou, Pingzheng ��145 Wilck, Nicola ��217 Zhou, Zhimin �� 25 Willenbrock, Michael ��117 Zhu, Qionghua �� 79 Willimsky, Gerald �� Ziebold, Ulrike ��203 Willmore, Lindsay ��167 Ziehm, Matthias �� 61 Willnow, David �� 6383 Zillmann, Silke ��145 Willnow, Thomas �� 71 Zimmer, Anja �� 99 Winkler, Wiebke �� 109, 223 Zincke, Fabian ��231 Winter, Lukas �� 47 Zink, Annika ��157 Wippich, Marc ��243 Zinnall, Ulrike �� Wirges, Anthea ��111 Zinzen, Robert P. ��209 Wirtz, Tristan ��109 Ziomkowska, Joanna ��145183 Wisinski-Bokiniec, Phillip ��155 Zschummel, Maria ��111 Wissler, Susanne ��229 Zühlke, Kerstin �� 27 Wittstruck, Angelika ��125 Zummach, Ramona �� 39 Wittstruck, Nadine ��219 Zwingenberger, Bettina ��225 �� 79 Zyrianova, Tatiana ��137 Woehler, Andrew ��253 Wiznerowicz, Irmgard

300 MDC Research Report 2016 Common Facilities

A 8 Gate House with Café Max and apartments A 9 Reception gate A 13 Gläsernes Labor / Life Science Learning Lab; Campus Info Center A 14 Cafeteria

MDC Guesthouses B 54 Hans Gummel House B 61 Salvadore Luria House with kindergarden Research

Max Delbrück Center for Molecular Medicine (MDC) C 27 Walter Friedrich House C 31.1 -2 Max Delbrück House C 31.3 -5 Energy and Infrastructure Center South (EZS) C 83 MDC.C / Max Delbrück Communications Center C 84 Hermann von Helmholtz House B 63, 64 Research Services B 88 Ultrahigh Field Facility B 89 Max Rubner House A10 Library

Leibniz-Institut für Molekulare Pharmakologie (FMP) C 81 FMP

Facilities shared by MDC and FMP C87 Timoféeff-Ressovsky-Haus Clinical Research

B 42-53 Experimental and Clinical Research Center (ECRC) Companies

A 7 Eckert & Ziegler AG, Eckert Wagniskapital und Frühphasenfinanzierung (EWK) A 15 Car mechanics, EZAG, Gardener B 55 OCVH - Oskar-und-Cécile-Vogt-Haus BBB-post office, 8sens.biognostic, FILT, MRI, ConGen, E.R.D.E., ART-CHEM, LIPIDOMIX, Roboklon, Myelo, Fresenius, Patent lawyers Dr. Baumbach, L.O.T., WISE, Octreopharm, Institut E & G, NIKON, Hermes-Analytik, Steinbeis-Transfer-Institut, I.M.S.M., AJ Innuscreen B 64 epo D 16/23 Eckert & Ziegler AG D 79 Erwin Negelein House Akademie der Gesundheit, BioTeZ, imaGenes, AviTop, GlycoUniverse D 80 Otto Warburg House ALRISE, Silence Therapeutics, Celares, OMEICOS Therapeutics D82 Karl Lohmann House EPO GmbH D85 Arnold Graffi House BBB, stratec, aokin, Biosyntan, emp, ascenion, MerLion, KFFB, Prof. Wanker, Nutrineu, AMAL, pi-PharmaImage, GLYCOTOPE, S. Langmacker accountant www.mdc-berlin.de