Identification and Characterization of Putative Nucleomodulins in Mycobacterium Tuberculosis

IDENTIFICATION AND CHARACTERIZATION OF PUTATIVE NUCLEOMODULINS IN MYCOBACTERIUM TUBERCULOSIS

Abstract The nuclear targeting of bacterial proteins that modify host cell gene expression, the so- called nucleomodulins, has emerged as a novel mechanism contributing to virulence of several intracellular pathogens. The goal of this study was to identify nucleomodulins produced by Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis (TB), and to investigate their role upon infection of the host. We first performed a screening of Mtb genome in search of genes encoding proteins with putative eukaryotic-like nuclear localization signals (NLS). We identified two genes of Mtb, Rv0229c and Rv3876, encoding proteins that are secreted in the medium by Mtb and are localized into the nucleus when expressed in epithelial cells or in human or murine macrophages. The NLSs of these two proteins were identified and found to be essential for their nuclear localization. The gene Rv0229c, a putative RNase, is present only in pathogen species of the Mtb complex and seems to have been recently acquired by horizontal gene transfer (HGT). Rv3876 appears more widely distributed in mycobacteria, and belongs to a chromosomal region encoding proteins of the type VII secretion system ESX1, essential for virulence. Ongoing studies are currently investigating the dynamics of these proteins upon infection of host cells, and their putative role in the modulation of host cell gene expression and Mtb virulence.

IDENTIFICATION ET CARACTERISATION DE NUCLEOMODULINES

PUTATIVES CHEZ LA BACTERIE MYCOBACTERIUM TUBERCULOSIS

Résumé Les nucléomodulines sont des protéines produites par des bactéries parasites intracellulaires et qui sont importées dans le noyau des cellules infectées pour y moduler l’expression génique et contribuer ainsi à la virulence de la bactéries. L’identification de nucléomodulines chez plusieurs espèces de bactéries pathogènes a fait émerger ce concept comme une stratégie supplémentaire utilisée par les parasites intracellulaires pour contourner les moyens de défense de l’hôte. Le but de cette thèse était d’identifier et d’analyser d’éventuelles nucléomodulines produites par l’agent étiologique de la tuberculose, la bactérie Mycobacterium tuberculosis (Mtb). Nous avons tout d’abord analysé le génome de Mtb, à la recherche de gènes dont les produits portent des séquences analogues aux séquences de localisation nucléaire des eucaryotes (NLS). Nous avons pu identifier deux gènes de Mtb, Rv0229c et Rv3876, qui codent des protéines sécrétées dans le milieu de culture et qui se localisent dans le noyau lorsqu’elles sont exprimées dans des cellules épithéliales ou dans des macrophages murins ou humains. Les NLS de ces deux protéines ont été identifiées et leur modification abolit la localisation nucléaire dans les cellules eucaryotes. Le gène Rv0229c est trouvé spécifiquement dans le génome des espèces pathogènes du complexe Mtb. Ce gène semble avoir été acquis récemment par l’ancêtre de Mtb, via un transfert génétique horizontal. Rv3876 est plus généralement distribué chez les mycobactéries, et appartient à une région génomique codant un système de sécrétion type VII, ESX1, essentiel pour la virulence de Mtb. Les travaux en cours visent à analyser la dynamique de ces deux protéines au cours de l’infection de macrophages ou de modèles animaux, et leur rôle en tant que modulateurs de l’expression génique des cellules infectées et dans la virulence de Mtb.

TABLE OF CONTENTS

ABSTRACT ...... i RÉSUMÉ ...... ii TABLE OF CONTENTS ...... iii LIST OF FIGURES ...... vii LIST OF TABLES ...... viii LIST OF ABBREVIATIONS ...... viiii ACKNOWLEDGEMENTS AND GRATITUDE ...... xiii ACKNOWLEDGEMENTS AND AUTHOR ATTRIBUTIONS ...... xv CHAPTER 1 INTRODUCTION ...... 1 1.1 Tuberculosis ...... 1 I. History of tuberculosis ...... 1 II. Epidemiology of tuberculosis ...... 1 III. Mycobacterium tuberculosis ...... 2 A. Taxonomy and characteristics ...... 2 B. Structure of the cell envelope of mycobacteria ...... 3 IV. TB treatment and drug resistance ...... 4 A. Drug resistance in TB today ...... 5 B. Current and future treatments of MDR- and XDR-TB ...... 6 1.2 Host responses to mycobacterium tuberculosis ...... 11 I. Immunity to Tuberculosis ...... 12 A. Innate immune system during Mtb infection ...... 12 B. Adaptive immune system during bacterial infection ...... 15 i. Cell-mediated immunity ...... 15 ii. Humoral immunity and B lymphocyte ...... 17 II. Evasion of the immune response ...... 18 A. Phagosome maturation ...... 18 B. Resistance mechanisms ...... 19 i. ROS and RNS ...... 19 ii. Heavy metals: Zinc (Zn)/copper (Cu) ...... 20

C. Nutritional mechanisms ...... 21 D. Cell death: necrosis vs apoptosis ...... 22 1.3 The concept of nucleomodulins ...... 24 I. Nuclear transport ...... 25 II. Nuclear Localization Signals (NLS) ...... 26 III. Nucleomodulins ...... 27 1.4 Statement of the problem ...... 31 1.5 Purpose ...... 32

CHAPTER 2 MATERIALS AND METHODS ...... 33 2.1 In-silico prediction of Mtb proteins targeted to the host cell nucleus ...... 33 2.2 Reagents and chemicals ...... 33 I. Commercial Reagents ...... 33 II. Enzymes and kits ...... 34 III. Oligonucleotides and Sequencing ...... 34 2.3 Cloning and expression ...... 34 I. Amplification of DNA fragments by PCR ...... 34 II. Plasmid constructions ...... 35 A. GFP-Fusion plasmid clones ...... 35 B. Mammalian expression vector constructs ...... 35 C. Generation of plasmid constructs for overexpression ...... 35 D. Plasmid constructs for overexpression under inducible promoter ...... 36 E. Site directed mutagenesis ...... 36 F. Construction of Knockout Mutant of Rv0229c and Rv3876 of Mtb ...... 37 G. Recombineering experiments in Mtb ...... 37 III. Ligation ...... 38 IV. Media and culture conditions for bacteria ...... 38 V. Preparation of chemically competent E. coli cells ...... 38 VI. Transformation of bacteria ...... 39 VII. Preparation of plasmids from small E. coli cultures ...... 39 VIII. Digestion of DNA with restriction enzymes ...... 39 VIIII. Agarose gel electrophoresis ...... 39

X. Purification of DNA fragment by gel extraction ...... 40 2.4 Cell culture and transfection ...... 40 I. Primary cells and cell lines ...... 40 A. Generation of human monocyte-derived macrophages ...... 40 B. Murine cell lines: Mouse monocyte-macrophage BALB/c cells ...... 41 C. Human embryonic kidney 293 (HEK 293) cells ...... 41 D. Cell passaging ...... 41 E. Assessing cell vitality ...... 41 F. Freezing and thawing cells ...... 42 II. Transfection of the constructed plasmids into host cells ...... 42 A. Transient transfection of HMDM ...... 42 B. Transient transfection of RAW 264.7 cells ...... 42 C. Transient transfection of HEK 293 cells ...... 43 III. Transfection efficiency analysis ...... 43 A. Fluorescence microscopy and imaging ...... 43 B. FACS analysis (Flow cytometry) ...... 43

CHAPTER 3 RESULTS ...... 44 3.1 Prediction of proteins with putative NLSs in Mtb ...... 44 3.2 Screening of nuclear targeting proteins of Mtb in eukaryotic cells ...... 46 3.3 Rv0229c encodes a protein of unknown function ...... 47 3.4 Rv3876 (EspI) encodes a putative ATPase, involved in the regulation of the ESX-1 secretion system ...... 48 3.5 Rv0229c was recently acquired by the ancestor of Mtb through horizontal genetic transfer as suggested by Comparative Genomics ...... 48 3.6 Identification of the functional NLS of Rv0229c and Rv3876...... 49 3.7 Rv0229c and Rv3876 are secreted proteins in Mtb ...... 51 3.8 Construction of genetic tools for functional characterization of Rv0229c and Rv3876 during infection by Mtb...... 52 A. Deletion of Rv0229c and Rv3876 ...... 52 B. Over-expression of Rv0229c and Rv3876 ...... 53 C. Expression of Flag tagged versions of Rv0229c and Rv3876 ...... 54

3.9 Transcriptome analysis ...... 54

CHAPTER 4 DISCUSSION ...... 56 REFERENCES ...... 62 APPENDICES ...... 79 APPENDIX A ...... 79 APPENDIX B ...... 80 1. Background ...... 80 1.1 Toxin-antitoxin (TA) systems in bacteria ...... 80 1.2 Gateway® cloning system ...... 81 2. Material and methods ...... 82 2.1 Generation of Rv1990c-1989c operon expression clones ...... 82 2.2 Cloning and expression ...... 83 2.3 Analysis of GFP expression ...... 83 I. Fluorescence intensity ...... 84 II. Fluorescence microscopy ...... 84 3. Results and perspectives ...... 84 3.1 Generation of expression clones by the Gateway®cloning system ...... 84 3.2. Testing the Rv1990c-1989c operon expression ...... 85 3.3. Dissection of the regulatory region of the Rv1990c promoter ...... 86 3.4 Expression of the Rv1990c-1989c operon in a variety of stress conditions .... 87 APPENDIX B: REFERENCES ...... 89

LIST OF FIGURES

Figure 1 Timeline showing historical highlights of scientific breakthroughs, and the ongoing public health issues, in the fight against TB ...... 1 Figure 2 Estimated TB incidence rates by country in 2013 ...... 2 Figure 3 The cell envelope of Mtb ...... 3 Figure 4 Current global pipeline of new tuberculosis drugs ...... 6 Figure 5 Mtb infection and granuloma formation ...... 12 Figure 6 Stages of phagosomal maturation ...... 18 Figure 7 Overview of nuclear protein import ...... 25 Figure 8 Bacterial strategies to control gene expression in the nucleus ...... 29 Figure 9 Mammalian expression vector ...... 34 Figure 10 E. coli and mycobacterium shuttle plasmid ...... 36 Figure 11 Genes associated with the Mtb ESAT-6 (esx) gene clusters ...... 45 Figure 12 Nuclear localization of mycobacterial nucleomodulin candidates fused with GFP in eukaryotic cells ...... 46 Figure 13 Nuclear localization of mycobacterial nucleomodulin candidates fused with GFP in RAW 264.7 macrophages ...... 47 Figure 14 Comparative Genomics of Rv0229c ...... 48 Figure 15 Rv0229c contains a functional nuclear localization signal ...... 49 Figure 16 Rv3876 contains a functional nuclear localization signal ...... 50 Figure 17 Rv0229c and Rv3876 are mycobacterial secretory proteins ...... 51 Figure 18 Over-expression of Rv0229cv or Rv3876 has no effect on multiplication of Mtb inside human macrophages ...... 53 Figure 19 Optimization of the GFP-reporter plasmid transfection ...... 54 Figure 20 Analysis of the RNAs co-immunoprecipitated with Rv0229c ...... 60

LIST OF TABLES

Table 1 Classification of drug used to treat TB ...... 5 Table 2 Different classes of NLS allow nuclear localization of bacterial nucleomodulins 27 Table 3 Strategies exploited by pathogens to modulate the host signaling pathways .... 29 Table 4 Strategies exploited by pathogens to modulate the host nuclear proteins ...... 29 Table 5 Strategies exploited by pathogens to modulate the host post-translation ...... 30 Table 6 Construction of the different plasmids used in this study ...... 35 Table 7 Construction of the different mutagenesis used in this study ...... 35 Table 8 Cloning primers used for plasmid constructions ...... 36 Table 9 Cloning primers used for mutagenesis ...... 37 Table 10 Primers used for Mtb mutant construction ...... 37 Table 11 Sequencing primers used in this study ...... 38 Table 12 Predicted nuclear targeting proteins with NLS sequences in Mtb H37Rv ...... 45

LIST OF ABBREVIATIONS

AG Arabinogalactan AIDS Acquired Immune Deficiency Syndrome AMs Alveolar macrophages APC Antigen Presenting Cells BAHD1 Bromo adjacent homology domain-containing protein 1 BCG Bacillus Calmette Guerin BSA Bovine Serum Albumin CIITA MHC class II transactivator CD4 Cluster of Differentiation 4 cDNA Complementary Deoxy Ribonucleic Acid CLRs C-type lectin receptor CMI Cell Mediated Immune Response DAMPs Damage-associated molecular patterns DC Dendritic Cells DC-SIGN Dendritic cell-specific intracellular-adhesion-molecule- grabbing non-integrin DMSO Dimethyl sulfoxide DOTS Directly Observed Treatment Short course EEA1 Early endosome antigen 1 eis Enhanced intracellular survival Emb Ethambutol ERK Extracellular signal-regulated kinase ESAT-6 Early Signs of Antigenic Target 6 kDa FACS Fluorescent Activated Cell Sorting FAS-I Fatty-Acid-Synthase systems I FAS-II Fatty-Acid-Synthase systems II GalN N‑acetylated galactosamine HAUSP Herpes virus-associated ubiquitin-specific protease HGT Horizontal gene transfer HIV Human Immunodeficiency Virus

ICL Isocitrate lyase IFN-β Interferon-beta IFN-γ Interferon-gamma Inh Isoniazid IL-2 Interleukin-2 IL-4 Interleukin-4 IL-5 Interleukin-5 IL-6 Interleukin-6 IL-10 Interleukin-10 IL-12 Interleukin 12 iNOS2 Inducible nitric oxide synthase IRAK IL-1 receptor-associated kinase IRF Interferon-regulatory factor JNK Jun N-terminal kinase kDa Kilo Dalton LM Lipomannan LntA Listeria nuclear targeted protein A LAM Lipoarabinomannan LpqH 19-kDa lipoprotein LTBI Latent TB infection MA Mycolic acids MDR-TB Multidrug-resistant TB ManLAM Mannosylated lipoarabinomannan MAP Mitogen-activated protein MAPKs Mitogen-activated protein kinases MAPc Mycolic acid-Arabinogalactan-Peptidoglycan complex mDAP meso-diaminopimelic acid MLS Malate synthase MOMP Mitochondrial outer membrane permeabilization MTBC Mtb complex MyD88 Myeloid differentiation primary response protein 88

NADPH Nicotinamide adenine dinucleotide phosphate NAGs N-acetylglucosamines NAM N-acetylmuramic acid NAG-NAM N‑acetyl glucosamine-N‑acetyl muramic acid NES Nuclear export signals NLRs NOD-like receptors NLS Nuclear localization signals NF Nuclear transcription factor NOD Nuclear oligomerization domain NOS2 Nitric oxide synthases 2 NPCs Nuclear pore complexes NTM Nontuberculous mycobacteria PAMPs Pathogen-Associated Molecular Patterns PG Peptidoglycan PGE2 Prostaglandin E2 PI Phosphatidyl-myo-inositol PI3P Phosphatidylinositol 3-phosphate PIMs phosphatidyl-myo-inositol mannosides PP2A Protein phosphatase 2A PRRs Pattern recognition receptors PTMs Post-translational modifications Pza Pyrazinamide RLU Relative fluorescence units RLRs RIG-I like receptors Rif Rifampicin RNI Reactive nitrogen intermediates ROI Reactive oxygen intermediates SRs Scavenger receptors SWI/SNF SWItch/Sucrose NonFermentable TAK1 TGFβ-activated protein kinase 1 TAL Transcription activator-like

xii

TB Tuberculosis TCA Tricarboxylic acid TCR T cell receptors TDR-TB Totally drug-resistant TB TDM Trehalose dimycolates TFs Transcription factors TLRs Toll-like receptors TMM Trehalose monomycolates TNF Tumor necrosis factor TRAF TNF receptor-associated factor VPS34 Human vacuolar protein-sorting 34 XDR-TB Extensively drug-resistant TB

“Uncertainty is the only certainty” -Buddha-

ACKNOWLEDGEMENTS AND GRATITUDE

I would like to express my sincerest gratitude to Dr. Olivier Neyrolles, CNRS research director, for giving me the opportunity to carry out my research project. It also gives me immense pleasure to thank my mentor and immediate supervisor, Prof Claude Gutierrez, for his support and continuous guidance.

Also, I want to express my thanks to Dr. Geanncarlo Lugo who has always played a key role in encouraging and coordinating this project and keeping me abreast of scientific developments and for his inexhaustible patience during the correction phase of this dissertation.

I would also like to thank Prof. Denis Hudrisier for helping me with important comments, meticulous suggestions and astute criticism during the practical phase. And thanks to all other lab colleagues: Assoc. Prof Yannick Poquet, Florence Levillain, Ingrid Mercier, Dr. Carine Duval, Dr. Alan Benard, Dr. Alexandre Gouzy, Anthony Troegeler, Alexia Dumas and Lucie Bernard for their scientific help during the research program.

Sincerest thanks are also to: My Dissertation Committee: Prof Matthieu Arlat and Assoc Prof Philippe Rousseau. My Thesis Defense Committee: Prof Olivier Dussurget, Prof Carlos Martin and Prof Matthieu Arlat.

I also appreciate my pervious lab of Dr. Mamadou Daffé, CNRS research director, and his team. Special thanks to Assoc. Prof Nathalie Eynard for her precious support and encouragement. I am also thankful to my former advisor Dr. Didier Zerbib who supported me for the first 3 years of my research program.

I would also like to thank The Royal Thai Government Scholarships, offered by the Ministry of Science and Technology, for its financial support.

xiv

I am also thankful for my friends here in Toulouse, especially to Clément for the amazing support and the wonderful friendship when I was alone, and to his family. GC (Grumpy cat), Denis (senseï) and his family, Alan (the boy), Yanos and his family, Ingrid (la plus belle dans l’IPBS) and her family, Alexia (Madame Ouin Ouin), Alex (the best of the best), Stevie (beau gosse), Anthony (try to make it) and Carine (KK). They are lab colleagues who became my dear friends. I am also thankful my new friends from California: Alexandra and Natsinet and special thanks to Nat (that’s interesting) for her extensive knowledge of the English language, proofreading of my manuscript and also for the great times we spent together, and all others who helped make my stay in France a very pleasant one.

Last, but not the least, I would also like to thank to my teacher Prof Somsak Pantuwatana for constant support, encouragement and for never retiring from my life. I am also very thankful to my friends in Thailand, Sompong, Supirada, Sawanya, my lovely student Kanookporn and her family for unwavering support and lasting friendship.

Finally, I would like to express a deep sense of gratitude to my family, especially to my parents, who have always stood by me in all situations and I owe my life to them for the constant love, encouragement, moral support and blessings.

ACKNOWLEDGEMENTS AND AUTHOR ATTRIBUTIONS

The project was conceived by Dr. Olivier Neyrolles. Experiments were designed by Prof Claude Gutierrez. Experiments were performed in biosafety level 3 facility by Prof Claude Gutierrez and Dr. Anna Grabowska. Protein secretion test in Mtb using western blot analysis by Dr. Daria Bottai, Pisa University, Italy. Co-immunoprecipitation test by Dr. Anthony Henras and Dr. Yves Roméo, LBME - UMR 5099 - CNRS/UPS.

Figure 1. Timeline showing historical highlights of scientific breakthroughs, and the ongoing public health issues, in the fight against TB (Kaufmann, 2005).

CHAPTER 1 INTRODUCTION

1.1 TUBERCULOSIS I. History of tuberculosis Tuberculosis (TB), historically known as “phthisis” or “consumption,” due to the disturbing weight loss experienced by patients suffering from this disease, is one of the oldest recorded human afflictions as evidenced, for example, by Egyptian mummies dating as far back to 1550–1080 BC. They were found to have spinal columns with signs of tubercular decay and mycobacterial DNA still detectable in the bones (Cambau and Drancourt, 2014; Daniel, 2006; Nerlich et al., 1997). It was not until 1834 that the term “tuberculosis” was coined for this disease based on the original description (in 1650) of the apparent lung nodules (characteristic of patients succumbing to TB) as “tubercles” (Cambau and Drancourt, 2014). Another important historical event happened when Jean Antoine Villemin demonstrated (in 1865) that TB was an infectious disease; he showed transmissibility of TB from humans to rabbits, from cattle to rabbits, and from rabbits to rabbits (Cambau and Drancourt, 2014). The history of TB was changed dramatically in 1882, when Robert Koch identified the bacillus as the etiological agent causing TB, for which he was awarded the Nobel Prize in Physiology or Medicine in 1905. Four years later, the bacillus was named Mycobacterium tuberculosis (Mtb) (Grange, 1982). This revolutionary finding gave way to later discoveries such as that of tuberculin, which is an extract of Mtb used in skin testing in animals and humans to identify TB infection (in 1890), the Bacillus-Calmette Guérin (BCG) vaccine (in 1921), and the first anti-TB drug (i.e. streptomycin, in 1943) (Kaufmann, 2005). Together, these breakthroughs offered hope for the eradication of the disease (Fig. 1).

II. Epidemiology of tuberculosis More than a century after the discovery of TB, despite advances in diagnostic, vaccination and treatment strategies, numerous challenges still remain. Today, TB ranks number eight on the list of the most common causes of death worldwide, and it is the second most common cause of death from an infectious disease after the human immunodeficiency virus (HIV /AIDS) (WHO, 2014). Mtb infection kills more people than any other bacterial pathogen. In 2013, there were more than 9 million TB cases accounting for over 1.5 million deaths. This

Figure 2. Estimated TB incidence rates by country in 2013 (WHO, 2014).

CHAPTER 1 INTRODUCTION 2 is especially evident in the case of co-infection with HIV, TB is a leading killer of HIV-positive people causing one fourth of all HIV-related deaths (WHO, 2014). The overall TB burden continues to rise as a result of the rapid growth of the world population, specifically in Africa and Asia. The six countries with the largest number of incident cases in 2013 were India, China, Nigeria, Pakistan, Indonesia and South Africa; India (24%) and China (11%) alone accounted for 35% of the world’s TB cases. Figure 2 illustrates the TB incidence rates by country in 2013 (WHO, 2014).

III. Mycobacterium tuberculosis A. Taxonomy and characteristics The bacillus causing TB in humans belongs to the genus Mycobacterium. Mycobacteria are classified as Gram-positive bacteria based on 16S ribosomal RNA sequence comparison, but are more closely related to Gram-negative bacteria when comparing the whole genome (Fu and Fu-Liu, 2002). Mtb is typically visualized by Ziehl-Neelsen (acidfast) staining and appears as a rod-shaped red bacillus. At 37°C and under optimal aerobic conditions and nutrients, Mtb has a generation time of 18–24 hours and forms a white to light yellow colony on agar within 3–4 weeks. Mtb does not form spores but has the capacity to become dormant, a non-replicating state characterized by low metabolic activity and phenotypic drug resistance (Gengenbacher and Kaufmann, 2012). The genome of the laboratory strain H37Rv of Mtb was the first completely sequenced in 1998 by Cole and coworker. The genome has a high guanine (G) and cytosine (C) content (65%), and comprises 4.4 mega base pairs (Mbp), encoding around 4,000 predicted proteins (Cole et al., 1998). Mycobacteria can be divided into three major groups for the purpose of diagnosis and treatment: i) the Mtb complex (MTBC) that cause TB in humans and animals, includes Mtb and the closely related species (i.e. M. bovis, M. bovis BCG, M. africanum, M. microti, and M. canetti (Richter et al., 2003), ii) M. leprae that causes leprosy, and iii) the nontuberculous mycobacteria (NTM) that groups all Mycobacterium species except the obligate pathogens MTBC and M. leprae, which are typically environmental organisms residing in soil and water and are in most cases nonpathogenic for humans or animals (van Ingen, 2013). Mycobacterium species that belong to NTM include human disease-associated bacteria: M. avium, M. kansasii, M. abscessus, M. chelonae, M. fortuitum, M. peregrinum and M. marinum. Other species are rarely the cause of disease, such as M. smegmatis and M. flavescens (Griffith et al., 2007). M. smegmatis and

Figure 3. The cell envelope of Mtb. The cell wall core is composed of three different covalently linked structures peptidoglycan (PG ‐ grey), arabinogalactan (AG ‐ blue) and mycolic acids (MAs ‐ green), forming a complex known as MAPc (Mycolic acid‐Arabinogalactan‐Peptidoglycan complex). The covalent linkage of MAs results in a hydrophobic layer of extremely low fluidity. This layer is also referred to as the mycomembrane. The outer part of the mycomembrane contains various free lipids, such as PIMs, LM and LAM, which are intercalated with the MAs. The outer layer is known as the capsule that mainly contains polysaccharides (Abdallah et al., 2007).

CHAPTER 1 INTRODUCTION 3

M. bovis BCG (an attenuated form of M. bovis) are often used as surrogate hosts to study Mtb gene expression (Gengenbacher and Kaufmann, 2012).

B. Structure of the cell envelope of mycobacteria The mycobacterial cell wall consists of inner and outer layers that surround the plasma membrane. The inner compartment is composed of three distinct macromolecules: peptidoglycan (PG), arabinogalactan (AG) and mycolic acids (MA). Interestingly, the PG is covalently attached to AG, which in turn is attached to MA; these three components form the cell wall core otherwise known as the Mycolic acid-Arabinogalactan-Peptidoglycan complex (MAPc). This complex remains as the insoluble residue when cell walls are disrupted (e.g. with various solvents), whereas other free lipids and proteins are solubilized. The insoluble core is essential for the viability of the cell and many of the drugs used to combat Mtb specifically target these complexes (Brennan, 2003; Hett and Rubin, 2008). The outer layer, called capsule, contains only minor amounts of lipids and consist of a loosely- bound structure of polysaccharides, proteins and noncovalently linked glycoconjugates, which is one of the unique features of the mycobacteria. Indeed, this last feature exhibits a high content of mannosylated molecules, including phosphatidyl-myo-inositol (PI) mannosides (PIMs), lipomannan (LM) and lipoarabinomannan (LAM) in particular (Kaur et al., 2009). Figure 3 illustrates the organization of the cell envelope of Mtb (Abdallah et al., 2007). The following is a brief description of some of the components of the inner and outer layers of the mycobacterial cell wall: • The PG (or murein), which is surrounding the plasma membrane, is made of peptides and glycan strands. The glycan strand consists of repeating N- acetylglucosamines (NAGs) linked to N-acetylmuramic acid (NAM) via peptide bridges to form N‑acetyl glucosamine-N‑acetyl muramic acid (NAG-NAM). The PG polymer can be also modified by glycolylation of NAM residues, or amidation of the d‑Glu and meso-diaminopimelic acid (mDAP) residues of the peptide side chain (Kieser and Rubin, 2014). • AG is the major polysaccharide of the mycobacterial cell wall and is important for cell wall integrity, which is composed of arabinan and galactan. AG is covalently attached to the PG layer by a phosphoryl-N-acetylglucosaminosyl-rhamnosyl linkage and can also be modified by the addition of succinyl or unusual

CHAPTER 1 INTRODUCTION 4

non‑N‑acetylated galactosamine (GalN) moieties. The GalN modifications are mostly present in pathogenic mycobacteria and are thought to promote infection efficiency (Hett and Rubin, 2008; Kieser and Rubin, 2014). • MAs are long-chain, α-acylated, β-hydroxylated fatty acids, which are formed from the condensation of a medium chain fatty acid (C20-C26) produced by Fatty- Acid-Synthase systems I (FAS-I) with a long meromycolic chain (C60-C90) produced by Fatty-Acid-Synthase systems II (FAS-II) (Cole et al., 1998). MAs are a major lipid component of the envelope and form the external mycomembrane (Hoffmann et al., 2008; Zuber et al., 2008). They are either inserted in the mycomembrane as trehalose dimycolates (TDM) and trehalose monomycolates (TMM) or linked covalently to the AG, itself bound to PG, to form the MAPc (Cole et al., 1998; Crick et al., 2001). The coordination of MAs and AG synthesis and insertion in the cell wall still unclear, recent report suggest that these macromolecules are produced and exported at the cell pole where new cell growth occurs (Carel et al., 2014; Meniche et al., 2014). • PIMs contain up to six mannose residues, which are unique glycolipids anchored through a PI moiety to the inner and outer membranes of the cell envelope of all Mycobacterium species (Boldrin et al., 2014). PIMs play important roles in the permeability barrier of the cell envelope, cell membrane integrity, the regulation of septation and cell division (Morita et al., 2005; Patterson et al., 2003). • LM and LAM are extensions of PIMs. LM is the mannan chain added to the PIM as a chain and a branch. Similarly, LAM is an additional arabinan group attached to the PIM (Chatterjee and Khoo, 1998). LAM from Mtb has short mannose containing oligosaccharide ‘‘caps’’ that are involved in mediating the binding of mycobacteria to, and subsequent entry into, macrophages (Brennan, 2003). Both LM and LAM show various immunomodulatory activities, and play critical roles in the integrity of cell wall and the pathogenesis of TB (Fukuda et al., 2013).

IV. TB treatment and drug resistance Streptomycin, first purified from Streptomyces griseus in 1944 by Schatz and Waksman and then introduced as TB chemotherapy in 1946 (Zumla et al., 2013), was the first antibiotic with proven activity against Mtb opening the doors as a viable approach for the generation

Table 1. Classification of drug used to treat TB. First‐line anti‐TB drugs Group 1. Oral: isoniazid (H/Inh), rifampicin/rifampin (R/Rif), pyrazinamide (Z/Pza), ethambutol (E/Emb), rifapentine (P/Rpt) or rifabutin (Rfb)

Second‐line anti‐TB Group 2. Injectable aminoglycosides: streptomycin (S/Stm), drugs kanamycin (Km), amikacin (Amk). Injectable polypeptides: capreomycin (Cm), viomycin (Vim). Group 3. Oral and injectable fluoroquinolones: ciprofloxacin (Cfx), levofloxacin (Lfx), moxifloxacin (Mfx), ofloxacin (Ofx), gatifloxacin (Gfx). Group 4. Oral: para‐aminosalicylic acid (Pas), cycloserine (Dcs), terizidone (Trd), ethionamide (Eto), prothionamide (Pto), thioacetazone (Thz), linezolid (Lzd).

Third‐line anti‐TB drugs Group 5. Clofazimine (Cfz), linezolid (Lzd), amoxicillin plus clavulanate (Amx/Clv), imipenem plus cilastatin (Ipm/Cln), clarithromycin (Clr)

CHAPTER 1 INTRODUCTION 5 of TB-specific therapy. The current chemotherapy of infectious TB is based on drugs classified into five groups based on evidence of efficacy, potency, drug class and experience of use (Table 1). The first-line anti-TB drugs (group 1) are currently recommended in a four drugs combination for the treatment of drug-susceptible TB; the second-line (groups 2-4) are reserved for drug-resistant TB; and the third-line (group 5) have unclear efficacy or undefined roles (Zumla et al., 2013). The current recommended treatment of drug-susceptible TB is a minimum of six months and achieves cure rates of >95% when administered under directly observed therapy (DOT). This treatment is done in two phases. The first comprises a two-month regimen of first-line four drugs comprising isoniazid (Inh), rifampicin (Rif), pyrazinamide (Pza) and ethambutol (Emb). The second is characterized by administration of Inh plus Rif for the following four months (ISTC, 2014). Despite the success of this treatment, the emergence of drug resistant Mtb strains and co-infection incidence with HIV/AIDS make TB a difficult challenge to manage, as described below.

A. Drug resistance in TB today The past 30 years have seen the worldwide appearance of multidrug-resistant TB (MDR- TB), followed by extensively drug-resistant TB (XDR-TB), and, most recently, of totally drug- resistant TB (TDR-TB). MDR-TB is defined by resistance to at least Inh and Rif, or more of the first-line anti-TB drugs (Inh, Rif, Pza, Emb and Rfb). In 2013, the WHO estimated that there were 480,000 incidences of drug-resistant TB, and 210,000 drug-resistant TB-related deaths worldwide (WHO, 2014). There are two forms of drug resistance in Mtb: genetic drug resistance and phenotypic drug resistance. Genetic drug resistance refers to genetic alterations, such as mutations in target genes. In many bacteria, drug resistance is the result of horizontal gene transfer (HGT) by plasmids or transposons. In the case of Mtb, all strains with acquired resistance that are currently known appeared through chromosomal mutations fixed by the selective pressure of antibiotics (Almeida Da Silva and Palomino, 2011). Phenotypic drug resistance or drug tolerance arises when genetically susceptible cells become resistant to antibiotic treatment without genetic alterations (Dartois, 2014). Treatment for MDR-TB is then prolonged, less effective, costly and therapeutically poorly tolerated.

Figure 4. Current global pipeline of new tuberculosis drugs (WHO, 2014).

CHAPTER 1 INTRODUCTION 6

XDR-TB is defined as resistant to at least Inh and Rif, plus any fluoroquinolone and at least one of the three injectable second-line drugs: Amk, Km and Cm. According to WHO in 2013, one hundred countries have reported XDR-TB, with an estimated 9% of patients with MDR- TB that have XDR-TB (WHO, 2014). The principles used for MDR-TB and XDR-TB treatment are the same. The XDR‑TB takes longer to treat than MDR‑TB and requires the use of third- line anti‑TB drugs, which are expensive and often have more side effects than first-line or second-line drugs. The XDR-TB is associated with a much higher mortality rate than MDR-TB due to the reduced number of effective treatment options (Velayati et al., 2009; Zumla et al., 2013). TDR-TB is referred to the TB caused by the Mtb strains resistant to all available first-line, as well as second-line TB drugs (Udwadia et al., 2012).

B. Current and future treatments of MDR- and XDR-TB A growing number of MDR-TB and XDR-TB strains, and the full emergence of TDR-TB strains, has had a tremendous negative impact on the success of treatment programs against TB. Together with the toxicity reported for the new generation of drugs, the current drug therapy and the BCG vaccine are largely ineffective at least in preventing adult pulmonary disease. There is no doubt that improvement is largely needed for anti-TB treatments. The last decade has seen the development of a promising TB drug pipeline (Fig. 4). Combining these new drugs with existing TB drugs offers hope for regimens that are better tolerated and are shorter in duration. A number of candidates in Phase II and Phase III clinical trials together with much activity in the TB therapeutic, such as Bedaquiline (Phase II) and delamanid (Phase III), represent promising treatment for the emerging MDR-TB (WHO, 2014). Nevertheless, new drug development and therapeutic combinations are still needed to eradicate various forms of TB and prevent the development of resistance. Several novel compounds are currently under investigation and new treatment regimens are in clinical trials including existing drugs that are redeveloped or repurposed for TB and new chemical compounds (Fig. 4).

CHAPTER 1 INTRODUCTION 7

• New anti-TB drug development i. Repurposed compounds FLUOROQUINOLONES Several members of this class of compounds have been used as second-line drugs for the treatment of MDR-TB. Fluoroquinolones are broad-spectrum antimicrobial drugs that trap gyrase and topoisomerase IV on DNA as ternary complexes, thereby blocking the movement of replication forks and transcription complexes (Drlica et al., 2009). For instance, gatifloxacin and moxifloxacin (Fig. 4) are currently in Phase III clinical trials to establish whether treatment of drug-susceptible TB can be shortened to 4 months by substitution of gatifloxacin for ethambutol, or moxifloxacin for ethambutol or isoniazid (Ma et al., 2010).

RIFAMYCINS Rifampicin has been the key component of the first-line treatment for TB chemotherapy for 40 years. This molecule prevents transcription by targeting the β subunit of RNA polymerase. Rifapentine (Fig. 4), another type of rifamycin, acts in the same way but has a much longer half-life than rifampicin with a better exposure and shorten treatment duration (Zumla et al., 2013).

OXAZOLIDINONES Oxazolidinones are a new class of drugs that inhibit protein synthesis by binding to the 23S rRNA in the 50S ribosomal subunit of bacteria. These compounds have a broad spectrum of activity against anaerobic and gram-positive aerobic bacteria, and mycobacteria. Linezolid, a first-generation oxazolidinone (Fig. 4), has been used in combination regimens for the treatment of MDR-TB. Sutezolid (PNU‑100480; Fig. 4) is an analogue of linezolid that has stronger bactericidal activity in the murine model than linezolid, and it is currently in Phase II clinical trials. AZD5847 (Fig. 4) is a next-generation oxazolidinone that acts like linezolid; the drug has bactericidal activity against Mtb inside macrophages, as well as in murine models of acute and chronic TB infection (Zumla et al., 2013).

CHAPTER 1 INTRODUCTION 8

ii. New chemical compounds DIARYLQUINOLINE The newly approved drug bedaquiline (TMC-207; Fig. 4) is an ATP synthase inhibitor that decreases the intracellular ATP levels. It is highly potent against drug-susceptible Mtb and MDR-TB (Zumla et al., 2013).

BENZOTHIAZINONES The 1,3-benzothiazin-4-ones (BTZs) (Fig. 4) are one of the most potent antimycobacterial agents that have activity against Mtb in vitro, ex vivo and murine TB models, by blocking enzymes important for the synthesis of the cell-wall arabinans (Makarov et al., 2009). Among the BTZ derived compounds is BTZ043 that is candidate for inclusion in combination therapies for both drug-sensitive and XDR-TB. Likewise, compared to BTZ043, the derivative PBTZ169 has improved potency, safety and efficacy in zebrafish and mouse models. When combined with other TB drugs, PBTZ169 showed additive activity against Mtb in vitro (Makarov et al., 2014).

NITROIMIDAZOLES Two newer nitroimidazoles, PA‑824 and Delamanid (OPC67683) (Fig. 4), are prodrugs in clinical development. PA‑824 is activated intracellularly by active nitrogen species, including nitric oxide (NO), which are the major effectors of its aerobic activity. The drug component has progressed to Phase II trials and is being assessed as a component of novel regimens. Delamanid is a nitro-dihydro-imidazooxazole whose mechanism of action is probably through inhibiting mycolic acid biosynthesis; it also kills intracellular TB by producing NO. The action of Delamanid seems more potent than that of PA‑824, and it is currently in Phase III for the treatment of MDR‑TB (Zumla et al., 2013).

SQ109 SQ109 (Fig. 4), a 1,2-ethylenediamine, is a derivative of ethambutol. Its mode of action is to target the mycobacterial membrane protein Large 3 (MmpL3) (Tahlan et al., 2012), an essential membrane transport protein belonging to the resistance, nodulation and division (RND) family. SQ109 inhibits mycolic acid biogenesis by blocking the transport of trehalose monomycolates (TMM) into the cell envelope (Zumla et al., 2013).

CHAPTER 1 INTRODUCTION 9

• TB vaccine development Achieving the goals to eliminate TB is requiring development of many aspects. Drug treatment is one strategy to control the disease; early and effective diagnosis is another critical aspect and development of an effective new vaccine is also of prime importance (Young et al., 2008). However, BCG vaccine is the only vaccine used in practice for the last 100 years, although it only protects younger children and exhibits poor efficacy in adults (WHO, 2014). The protection in children against pulmonary TB can reach up to 80%, whereas in adults only 50% are protected in the best scenario or no protection at all in the worst ones. In addition, BCG vaccination can lead to adverse effects (Principi and Esposito, 2015). Thus, the development of safe new vaccines is desired and considered as a priority to induce a long-lasting memory and promote better protection than the conventional BCG vaccine. There are three different approaches for new TB vaccines designers:

i. BCG replacements The main strategies used to develop new vaccines are based on replacing the BCG antigen of the BCG vaccine to confer longer protection with increased safety in immuno- compromised patients.

RECOMBINANT BCG (rBCG) STRAINS Methods to construct rBCG include over expression of promising Mtb immunodominant antigens expressed by BCG, such as VPM 1002, which is a recombinant BCG strain based on live Mtb vaccine. The recombinant BCG strain has a gene of Listeria monocytogenes coding for the protein listeriolysin (Hly), which is integrated into the BCG genome with inactivated urease C gene (BCG ΔureC::hly). Hly is responsible for pore formation in the macrophage phagosome and BCG antigens trapped inside the phagosome, consequently enters the MHC II pathway of antigen presentation for a better activation of the immune response (Grode et al., 2013; Principi and Esposito, 2015).

ATTENUATED MTB STRAINS The first live-attenuated Mtb-based vaccine is MTBVAC, which contains two independent deletions negatively affecting its virulence. The first one is in the gene encoding phoP, a key transcription factor for the regulation of Mtb virulence; this mutant strain showed a high

CHAPTER 1 INTRODUCTION 10 degree of safety, improved immunogenicity and protective efficacy compared to BCG in several animal models. The second mutation is in the gene encoding fadD26, one of the major mycobacterial virulence factors, which exhibits safety and biodistribution profiles similar to BCG, conferring superior protection in preclinical studies (Arbues et al., 2013).

MYCOBACTERIUM VACCAE M. vaccae (MV) is a saprophytic NTM that contains a number of antigenic epitopes common to Mtb, such as the heat shock proteins and LAM and some low molecular weight (<40 kDa) secreted antigens (Stanford et al., 1990). Heat-killed MV added to chemotherapy was helpful for the treatment of patients with never-treated TB that showed improved sputum conversion and X-ray appearances (Yang et al., 2011). In HIV-infected patients MV was well tolerated and no serious adverse events have been reported. In addition, MV was safe to prevent TB and immunogenic in HIV-infected patients (Montagnani et al., 2014).

ii. Booster vaccines This approach consists in preparing vaccines capable of boosting the immune response in subjects who have previously received the BCG vaccine.

VIRAL VECTORED VACCINES MVA85A is a live viral vector that cannot replicate in humans. It is a modified recombinant strain of Vaccinia virus Ankara expressing the Mtb antigen 85A (Ag85A), designed to enhance the immune response induced originally by BCG (McShane et al., 2005). A number of studies reported the viral vector is safe in healthy adults, Mtb and HIV infected patients, adolescents, children and infants (Montagnani et al., 2014).

PROTEIN-ADJUVANTED VACCINES Several protein-adjuvanted vaccines have been developed already. For instance, H1/IC31 is a recombinant subunit vaccine, composed by the hybrid protein of ESAT6 and Ag85B (Ag85B-ESAT6) that have been extensively evaluated in several animal models and combined with different adjuvants, such as cationic liposomal adjuvant formulation 01 (CAF01). The vaccine has been shown to be safe in healthy adults. No serious adverse events related to the vaccine have been reported (Montagnani et al., 2014).

CHAPTER 1 INTRODUCTION 11

iii. Therapeutic vaccines Many therapeutic TB vaccines including MV have been developed. However, the most widely studied therapeutic TB vaccine is the RUTI® vaccine. RUTI® is a non-live therapeutic vaccine constituted by detoxified, fragmented Mtb cells, delivered in liposomes developed by Archivel Pharma, Spain immunotherapeutic vaccine (Cardona, 2006). RUTI® is safe and enables the induction of a mixed Th1/Th2/Th3, polyantigenic response with no local or systemic toxicity (Cardona, 2006). The vaccine was designed to shorten the treatment of latent TB infection (LTBI) with isoniazid (inh) (Vilaplana et al., 2008). In the near future, the current ongoing clinical trials for several TB vaccine candidates will prove to be effective and become readily available.

1.2 HOST RESPONSES TO MYCOBACTERIUM TUBERCULOSIS Multi-cellular organisms (the host) are constantly exposed to a great variety of pathogens, such as bacteria, fungi, parasites and viruses. To prevent the entry of these microbes, the host counts on an elaborate first line of defense composed of physical and biochemical barriers. The physical barriers include epithelial surfaces such as skin and mucosal linings of the gastrointestinal, respiratory and urogenital tracts, among others, and the biochemical ones include the secretion of antimicrobial substances (e.g. lysozymes and defensins) found in tears, saliva and skin oils, for example. If this first line of defense is broken by a pathogen, the host’s immune system is alerted and deployed to control and destroy the intruder. The immune system is divided into two types of responses: innate and adaptive. The innate immunity is an ancient and universal system that detects “microbial non-self”, relying on the ability of the host to recognize conserved products derived from microbes that are not found in the host. These are known as “Pathogen-Associated Molecular Patterns” (PAMPs). Indeed, the innate immune recognition of PAMPs represents a major checkpoint in the decision to respond to a pathogen, and it is dependent on a limited number of highly conserved pattern recognition receptors (PRRs). At the moment these receptors are engaged, the innate immune system will attack and, for the most part, eliminate the pathogen by deploying cellular and molecular weaponry in a rapid fashion. Together, these factors will culminate in the inflammation of the affected area, leading to a successful clearance of the pathogen characterized by its destruction and proper collection of debris. Nevertheless, there are exceptional cases in which pathogens have evolved specialized

Figure 5. Mtb infection and granuloma formation. (A) TB infection starts with the inhalation of an aerosol droplet containing the bacilli, and followed by deposition of the bacteria in the lung alveolar space. (B) The alveoli consists of type I and type II epithelial cells. After initial bacterial multiplication in alveolar macrophages, the bacteria are thought to be taken up by DCs, which may carry Mtb to the draining thoracic lymph nodes. (C) In the draining lymph nodes, DCs carrying Mtb undergo apoptosis, and the released mycobacterial antigenic peptides are presented by the activated lymph node‐resident DCs to the specific naive T cells, including through cross‐presentation. On antigen recognition, activated T cells proliferate, become effector T cells, and leave the lymph node to reach the blood circulation through the efferent lymphatics and the thoracic duct (Griffiths et al., 2010). (D) Effector T cells originating in the draining lymph nodes home back from the blood through pulmonary capillaries to the site of inflammation under the influence of chemokines and other mediators. TB granulomas are typically well organized, with a central core of macrophages, which can also fuse into multinucleated giant cells or differentiate into foam cells, surrounded by other leukocytes such as neutrophils, dendritic cells, B and T cells, NK cells, fibroblasts and cells that secrete extracellular matrix components (Ramakrishnan, 2012).

CHAPTER 1 INTRODUCTION 12 mechanisms to evade and circumvent the innate immune system. For these types of cases, the host relies on the specific response orchestrated by the adaptive immune system. Adaptive immunity is activated when the information about the pathogen, derived from the site of infection, is delivered to secondary lymphoid organs, where naïve lymphocytes (B and T cells) are mainly located. Indeed, the adaptive immune response is distinguished by the clonal selection and differentiation of effector cells that are capable of specific recognition of a given pathogen, orchestrate the elimination of the pathogen and generate “memory” cells. Immunological memory not only represents the hallmark of the adaptive immunity, but also underlies the principle of vaccination that provides an enhanced protection to the host against a later challenge with the same infectious agent.

I. Immunity to Tuberculosis A. Innate immune system during Mtb infection The innate immune response plays an important role in the protection against TB as it provides the first “active” line of defense. In pulmonary TB, the bacillus is typically inhaled as small aerosol droplets into the body through the mouth or nose, entering subsequently the alveolar passages along the trachea, bronchus and bronchioles (Fig. 5a). The respiratory mucosa consists of (i) the epithelium, a layer of airway epithelial cells, which mediate the production of cytokines and effector molecules to prevent invasion; (ii) the lamina propria, a layer of connective tissue and immune cells, including lymphocytes and macrophages; and (iii) a coating of airway surface liquid, which consist of mucus, immunoglobulin A and other innate immune factors on the luminal surface (Lerner et al., 2015). Mtb that successfully passes through the upper airways will be delivered to the alveoli. The alveoli consists of type I and II epithelial cells as well as other immune cells such as alveolar macrophages (AMs), dendritic cells (DCs) and neutrophils (Fig. 5b), which are the main targets for mycobacteria. Type I epithelial cells form the walls of the alveolus, are involved with gas exchange, and sometime can be infected by Mtb; type II cells produce and secrete pulmonary surfactant, antimicrobial molecules, hydrolytic enzymes and hydrolases. There are molecular soluble factors such as the collectin family of surfactant proteins that enhance macrophages phagocytosis by binding Mtb and causing agglutination, or secreted hydrolases that affect interactions with macrophages and host immune responses by altering the cell wall of Mtb (Lerner et al., 2015).

CHAPTER 1 INTRODUCTION 13

In order for Mtb to be phagocytosed (engulfed) by AMs and DCs, the pathogen has to first be recognized by PRRs on the surface of the immune cells. Indeed, the recognition of PAMPs in Mtb, such as lipoarabinomannan (LAM), mannosylated lipoarabinomannan (ManLAM) and trehalose dimycolate (TDM), is an essential step for the proper binding by innate immune cells. Several families of PRRs exist that are each capable of recognizing a different type of PAMPs. These PRRs belong to several families: Toll-like receptors (TLRs), C-type lectin receptor (CLRs), scavenger receptors (SRs), nuclear oligomerization domain (NOD) or NOD- like receptors (NLRs), and RIG-I like receptors (RLRs) (Killick et al., 2013). The best-studied TLR family is composed of 10 members in humans (TLRs 1-10). TLRs 1, 2, 4, 5 and 6 are located on the cell surface, while TLRs 3, 7, 8 and 9 are typically located intracellularly, on endosomal membranes (Killick et al., 2013). The importance of TLRs in the immune response against Mtb is perhaps best illustrated by TLR2. This receptor is known to form heterodimers with either TLR1 or TLR6, which have been implicated in recognition of mycobacterial cell wall glycolipids like LAM, LM, 38-kDa, and 19-kD mycobacterial glycoprotein, and PIM, triacylated (TLR2/TLR1), or diacylated (TLR2/TLR6) lipoproteins (Kleinnijenhuis et al., 2011). Also, TLR2 is important to initiate the innate host defense through its stimulatory effects on tumor necrosis factor-! (TNF!), Interleukin-1-beta (IL-1β), and IL-12 production in macrophages. Moreover, mice that are deficient of TLR2 show defective granuloma formation and, when infected with high doses of Mtb, they show enhanced susceptibility to infection compared to the WT mice (Kleinnijenhuis et al., 2011). Among the different innate immune cell types, the macrophages and DCs are thought to be essential players in the control of immune response to Mtb, as they are both the major effector cells to combat Mtb and the primary replication sites for this pathogen. During activation through PAMPs, these cells initiate a signaling cascade via adapter proteins such as myeloid differentiation primary response protein 88 (MyD88), which results in the recruitment of interleukin-1 (IL-1) receptor-associated kinase (IRAK) 4, TNF receptor- associated factor (TRAF) 6, TGFβ-activated protein kinase 1 (TAK1) and mitogen-activated protein (MAP) kinase, leading to activation and nuclear translocation of transcription factors such as the nuclear transcription factor (NF)-κB, the main nuclear activator of proinflammatory cytokines (Kleinnijenhuis et al., 2011). Key pro-inflammatory cytokines include interferon-! (IFN!) and TNF!, which in combination with the generation of reactive nitrogen intermediates (RNI) and reactive oxygen intermediates (ROI) (Flynn and Chan,

CHAPTER 1 INTRODUCTION 14

2001), contribute to microbicidal activity that ensures intracellular Mtb killing in macrophages and DCs (Flynn and Chan, 2003). Apart from macrophages and DCs, neutrophils, natural killer (NK) cells and the complement system have been implicated in the innate immune response against Mtb. Neutrophils play an essential role in the innate immune response to Mtb infection, as they are the first cells to be recruited and participate in the protection by i) engulfment of pathogen, ii) secretion of reactive oxygen species (ROS) and iii) production of microbicidal granule molecules, such as the cathelicidin LL-37, a neutrophil peptide with antimycobacterial activity (Martineau et al., 2007; Nathan, 2006). In addition, infected neutrophils become apoptotic and are phagocytized by macrophages, thereby becoming a source of antigen-presentation material to create a link between innate and adaptive immunity (Persson et al., 2008). In the case of NK cells, which are large granular circulating lymphocytes, they are attracted to the sites of bacterial infection early on and play a role in amplifying the antimicrobial defense to TB. This is via recognition of infected macrophages through receptor molecules such as NKp44 and NKp46 (Vankayalapati and Barnes, 2009). NKs have cytotoxic functions exerted through perforin and granzyme (granulysin) or Fas/Fas ligand pathways to destroy infected host cells. Through the secretion of IFN-γ, a critical step for activation of macrophages, NKs induce the macrophage-derived production of cytokines such as IL-12, IL-15 and IL-18, which in turn activate CD8+ T-cells and their cytotoxic capacities. In this manner, NK cells also act as a link to the adaptive immune system and ensure a protective response against Mtb and other intracellular pathogens (Gupta et al., 2012; Vankayalapati and Barnes, 2009). Last, the complement system is the humoral arm of the body’s first line of defense against invasion by pathogens. It has been shown that M. bovis BCG, which has 99.9% gene sequence identity with Mtb, activates the three major complement pathways: classical, lectin and alternative. The classical pathway is activated when the complement protein C1q binds to the complement activator either directly or via bound antibodies; the lectin pathway is triggered by the binding of mannose-binding lectin (MBL) or ficolin to the bacterial surface; and the alternate pathway is turned on through the C3b deposition on the bacterial surface. Overall, there are three major immunological functions of complements: opsonization of pathogen, which facilitates their ingestion by macrophages; microbial cell lysis through the formation of the complement membrane

CHAPTER 1 INTRODUCTION 15 attack complex (MAC); and leukocyte recruitment by eliciting chemokine secretion (Carroll et al., 2009; Dunkelberger and Song, 2010; Gupta et al., 2012). However, although it is effective against most bacterial infections, innate immunity is not sufficient to control Mtb infection, and thus the adaptive immune system is needed to control and eliminate this pathogen from the host.

B. Adaptive immune system during bacterial infection The adaptive immune response consists of humoral immunity, which is mediated by B lymphocytes, and cell-mediated immunity, which is controlled by T cells. During Mtb infection, the adaptive immune response becomes detectable approximately two weeks after the challenge with this pathogen. Antigen presentation is a crucial process for the activation of the adaptive immunity against the pathogen. Pathogens are taken up by specialized cells known as antigen-presenting cells (APCs), which include macrophages, DCs and B cells. The APCs then process and present Mtb-antigens to several T lymphocytes including CD4+ T cells and CD8+ T cells, leading to the activation and proliferation of these cells. One of the end result is the keen production of cytokines and antibodies to combat the pathogen (Jasenosky et al., 2015).

i. Cell-mediated immunity CD4+ T cells CD4+ T cells are a subpopulation of T cells that express the CD4 receptor. They also express T cell receptors (TCR) that recognize Mtb-antigens presented by infected APCs on major histocompatibility complex (MHC) class II molecules. In general, CD4+ T cells are divided into two subsets depending on the cytokines produced: T-helper 1 (Th1) and Th2 cells. On the one hand, Th1 cells produce interferon (IFN)-γ, tumor necrosis factor (TNF)-α and interleukin (IL)-2, and contribute to macrophage activation and granuloma formation. IFN-γ has been reported to be essential for defense against Mtb infection (Cooper et al., 1993; Flynn et al., 1993; van de Vosse et al., 2004). The number of IFN-γ producing T cells has been considered to be a marker for induction of a protective immunity against Mtb (Walzl et al., 2011). On the other hand, Th2 produce IL-4, IL-5, and/or IL-13 and contribute to antibody-mediated immune responses or support humoral immunity, and down regulate Th1 responses (Russell, 2007). Therefore, Th2 are considered to have inhibitory effects for

CHAPTER 1 INTRODUCTION 16 protective immunity against Mtb. The T cell response to Mtb is activated in the local lung- draining lymph node (Fig. 5c), where circulating naïve T cells recognize antigens through different receptors (as described above) presented by APCs (Gupta et al., 2012). The infected macrophages and DCs secrete cytokines including IL-12, IL-23, IL-7, IL-15 and TNF-α, leading to the differentiation of Th1 cell and the attraction of leukocytes to the infection site (Dheda et al., 2010). Th1 cells response leads to the release of pro-inflammatory cytokines including IFN- γ, TNF-α and IL-2. The activated macrophages show increased microbicidal activity through the expression of the inducible nitric oxide synthase (iNOS), which is thought to enhance killing of intramacrophage bacilli through nitric oxide (NO) and reactive oxygen species (ROS) production (Fang, 2004; Flynn and Chan, 2003; MacMicking, 2009), and components of the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex (NOX2) (Forman and Torres, 2002; Russell, 2011).

Cytotoxic CD8+ T cells Cytotoxic CD8+ T cells are a subpopulation of T cells that express the CD8 receptor along with the TCR. Mtb are cross-presented to CD8+ T cells via MHC class I molecules. Specific activation of CD8+ cytotoxic T cells contributes to protective immunity against Mtb by secretion of cytokines such as IFN-γ and TNF-α. As discussed before, IFN-γ production is essential for CD8+ T-cell mediation of protective immunity distinguished by the killing of Mtb-infected cells through a cytolytic molecule, perforin, and the antimicrobial protein granulysin-mediated pathway resulting in necrotic cell death, or by induction of apoptosis via Fas ligand (Dheda et al., 2010; Weerdenburg et al., 2010). Alternatively, previous studies have reported a subset of CD8+T cells that coordinately expresses chemokine CCL5 (RANTES), perforin and granulysin, attracting Mtb-infected macrophages and leading to the killing of the intracellular pathogen (Stegelmann et al., 2005).

ii. Humoral immunity and B lymphocyte The role of humoral immunity in the protection against Mtb infection is thought to be less important than the role of T cells. Yet, B cells and antibody production have been shown to play a variety of potential protective roles at each stage of Mtb infection. Antibody defense mechanisms at the alveolar level during the initial step of infection include i) the increase serum permeability into the alveolar space, ii) opsonization, iii) complement activation, iv)

CHAPTER 1 INTRODUCTION 17 and FcR-mediated enhanced Mtb phagocytosis and killing of extracellular Mtb within the cavity, among other protective activities (Achkar et al., 2015). B cells, which are located at the germinal center of lymphoid organs, are thought to play a role in i) the defense against infection as APC, ii) activation of T cells, iii) production of cytokines that induce the development of the T-helper cell response, and iv) production of antibodies, which are involved in various aspects of the innate and adaptive immune response (Achkar et al., 2015). The past decade has evidenced that administration of monoclonal antibodies against several Mtb components such as arabinomannan, LAM, and 16 kDa α-crystallin, can protect mice against Mtb (Reljic et al., 2006; Teitelbaum et al., 1998; Williams et al., 2004).

Granulomas formation During Mtb infection, the granuloma provides a microenvironment able to inhibit the growth of Mtb. Although the granuloma is the site for mycobacterial killing, the pathogen has developed a variety of mechanisms to resist this macrophage-mediated killing. The granulomas are formed when innate and adaptive immune responses act together as a multi-cellular structure to encapsulate and control the bacteria. In most cases, granuloma formation results in a state of latency or dormancy of the bacillus, which remains at the necrotic center of the structure, exposed to a toxic and hypoxic extracellular environment (Ramakrishnan, 2012). The tuberculous granuloma is an organized aggregate of epithelioid cells with a central core of macrophages, which can also fuse into multinucleated giant cells or differentiate into foam cells that are characterized by lipid accumulation. In addition to the extracellular mycobacteria, the central necrotic areas also include dead and dying macrophages surrounded by neutrophils, dendritic cells, B and T cells, NK cells, fibroblasts and cells that secrete extracellular matrix components (Fig. 5d) (Griffiths et al., 2010; Ramakrishnan, 2012). Moving away from the necrotic center and the epitheloid-macrophage inner ring, the granuloma structure is then characterized by an outer ring containing a mixture of anti-inflammatory macrophages and T lymphocytes. Moreover, the outer ring can also be accompanied by aggregates containing lymphoid cells, in particular B cells. Altogether, although granuloma formation appears as a host strategy for entrapping the infection, Mtb has evolved adapt to this particular niche, and it is also beneficial for the pathogen in its initial spread after dormancy state.

Figure 6. Stages of phagosomal maturation. After pathogen uptake, different stages of maturation are recognized as (a) early, (b) intermediate, (c) late phagosomes, and (d) The formation of phagolysosomes (Flannagan et al., 2009).

CHAPTER 1 INTRODUCTION 18

II. Evasion of the immune response As part of the long and continuing evolution process, pathogens have evolved multiple strategies to evade and subvert the immune response in the host. In the case of Mtb, this intracellular pathogen has the capacity to modulate different microbicidal activities in effector cells, including the macrophage (Kaufmann, 2001; Russell, 2001). The following are some well documented ways by which Mtb evades and circumvents these effector cells:

A. Phagosome maturation As aforementioned, while macrophages are key effector cells in Mtb killing, they also provide a niche for Mtb multiplication. When Mtb is detected by macrophage surveillance at the site of the infection, the process of phagocytosis is initiated by recognition of the pathogen by cell-surface receptors that trigger its engulfment into plasma-membrane- derived intracellular vesicles called phagosomes. In order to kill the pathogen and then degrade, process, and present bacterial antigens on its surface, the macrophage exposes Mtb to a toxic intracellular environment inside the phagosomes, which undergo a maturation process starting with early and late endosomes, and the eventual fusion with lysosomes (called phagolysosome), as illustrated in Figure 6 (Flannagan et al., 2009). The Rab family GTPases mediate phagosome maturation process. For instance, Rab5 facilitates fusion of the nascent phagosome with early endosomes through the recruitment of early endosome antigen 1 (EEA1) and the type III PI3K human vacuolar protein-sorting 34 (VPS34), which generates phosphatidylinositol 3-phosphate (PI3P) on the cytosol-facing leaflet of the early endosomal membrane (Canton, 2014). PI3P interacts with Rab5 in retaining EEA1, and promotes the recruitment and subsequent activation of a number of proteins involved in phagosome maturation. Another important factor is Rab7 that mediates the fusion of the phagosome with late endosomes (Canton, 2014) and phagolysosome formation (Desjardins, 1995). This latter process involves the dramatic increase in phagosome lumen acidity from neutral (7.0) to an acidic (4.5) pH, mediated by proton (H+) pumps of the V-ATPase family and subsequent activation of a number of hydrolytic luminal enzymes (protease and lipase), which compromise bacterial membrane integrity (Flannagan et al., 2009). ROS and RNS, generated by the phagosomal enzymes NADPH phagocyte oxidase and the iNOS, are essential microbicidal factors to eliminate the captured microorganisms (Fang, 2004).

CHAPTER 1 INTRODUCTION 19

Altogether, these processes result in the clearance of the engulfed bacterium and antigen presentation to initiate an adaptive immune response. Mtb is able to bypass the macrophage killing machinery by mainly preventing phagosome acidification and blocking the fusion of the Mtb-containing phagosomes with lysosomes. This strategy allows Mtb to escape proteolytic degradation and antigen presentation. Indeed, several studies have demonstrated that Mtb blocks the phagosome maturation by the use of various macromolecules. For instance, ManLAM inhibits VPS34 and takeover of EEA1 (Fratti et al., 2001). LAM incorporated in the membrane rafts via its glycosylphosphatidylinositol (GPI) anchor results in reduced phagosomal maturation (Welin et al., 2008). The surface lipid trehalose dimycolate (TDM) sequesters rab7 and thereby prevents lysosomal fusion (Katti et al., 2008). The secreted tyrosine phosphatase (PtpA) interacts with host V-ATPase, which results in dephosphorylation of host vacuolar protein sorting 33B (VPS33B) and subsequent exclusion of V-ATPase from the phagosome during Mtb infection (Wong et al., 2011). The lipoamide dehydrogenase (LpdC) protein mediates the retention of coronin 1 on vacuoles leading to phagolysosome fusion arrest (Deghmane et al., 2007). These examples demonstre the capacity of Mtb to influence the phagosome maturation process to avoid its killing and degradation.

B. Resistance mechanisms i. Reactive Oxygen Species (ROS) and Nitrogen-Centered Free-Radical Species (RNS) Cytotoxic gases are one of the host antimicrobial arsenals employed by host cells. These gases are termed ROS and RNS (Nathan and Shiloh, 2000). Host cells derive their energy from the reduction of oxygen through metabolism of the mitochondrial electron-transport chain, generating downstream intermediates such as hydrogen peroxide (H2O2) and hydroxyl – radicals (•OH), and by-products such as superoxide ions (O2 ). These three species are referred to as ROS and they are produced within phagocytic cells via the NADPH oxidase 2 (NOX2), a major factor driving the respiratory burst. RNS refers to the nitrogenous products of nitric oxide synthases 2 (NOS2), whose isoform is often called iNOS and which generates + reactive oxygen intermediates (ROI) including nitric oxide (•NO) to nitrate (NO3 ) (MacMicking, 2012; Nathan, 2002). Interestingly, the targets of ROS and RNS include mycobacterial DNA, lipids and hemo-proteins (MacMicking et al., 2003; Nathan and Shiloh, 2000). To resist against intracellular intoxication by ROS and RNS, Mtb uses multiple

CHAPTER 1 INTRODUCTION 20 mechanisms. This includes the production of the antioxidant mycothiol, several oxygen and nitrogen detoxification enzymes, i.e. superoxide dismutases SodA and SodC, the catalase KatG, and the NADH-dependent peroxidase and peronitrate reductase, among others (Lugo- Villarino and Neyrolles, 2014). Other factors include the ‘‘enhanced intracellular survival’’ (eis) and NuoG that play critical roles in modulating ROS production in macrophages (Shin et al., 2010; Miller et al., 2010).

ii. Heavy metals: Zinc (Zn)/copper (Cu) Another recently discovered antimicrobial mechanism is the use of metal ions, such as zinc (Zn2+) and copper (Cu2+), to inhibit intracellular pathogens (Botella et al., 2012). The metal ions or trace elements are essential for immune system. Indeed, they are required for the differentiation, activation and performance of immune cells and deficiency in these elements leads to suppression in the activities of the immune cells, which can in turn lead to increased bacterial infections (Failla, 2003). Host cells utilize these metals to poison invading pathogens by actively increasing their concentrations within the phagosome (Botella et al., 2012). In response, Mtb produces the Zn and Cu efflux pumps, CtpC and CtpV, respectively, as a way to avoid poisoning by these metals (Botella et al., 2012). Consequently, mutant strains of Mtb genes encoding Zn and Cu efflux pumps, which are known to be expressed to high levels when the bacillus is exposed to these metals, display a heavy accumulation of Zn and Cu, and have impaired intracellular bacterial growth, underlying the importance of these genes for Mtb virulence (Botella et al., 2011; Wagner et al., 2005; Ward et al., 2010; White et al., 2009; Wolschendorf et al., 2011).

C. Nutritional mechanisms Nutrient acquisition is essential for growth and survival of all organisms. This is also true for Mtb upon entry into macrophages and other phagocytic cells, which constitute rich nutritional sources of carbon, nitrogen and trace elements or transition metals. To persist and survive within the host cells, Mtb exploits and manipulates nutritional sources in different ways. Perhaps the best known example is that of iron (Fe). Fe is an essential element both for pathogens and for their host, as it is often required as cofactor for enzymes in metabolic pathways (Schaible and Kaufmann, 2004) and it is well known that Mtb requires Fe for growth and survival in intracellular host cells (Gobin and Horwitz, 1996), and that

CHAPTER 1 INTRODUCTION 21 bacterial growth is exacerbated under Fe overload (Gangaidzo et al., 2001; Schaible et al., 2002). Fe is a first row transition metal and an abundant biomolecule, existing in the reduced ferrous (Fe2+) or oxidized ferric (Fe3+) oxidation states. However, Fe3+ is poorly soluble under aerobic, aqueous and neutral pH conditions. Furthermore, Fe in Fe2+ state can generate highly toxic free oxygen radicals, resulting in protein denaturation, damages on genomic DNA and lipid peroxidation. Therefore, in the host, iron is usually bound to specific proteins, such as transferrin and lactoferrin, or complexed to heme within hemeproteins, maintaining a low concentration of available iron (Schaible and Kaufmann, 2004). In addition, macrophages maintain an even lower concentration of iron within the endosome, via the divalent metal cation transporter natural resistance-associated macrophage protein 1 (Nramp1/SLC11A1), an integral membrane phosphoprotein of macrophage phagosomes that pumps iron out (Cellier et al., 2007; Li et al., 2011). To counteract this defense mechanism and acquire iron, mycobacteria synthesize and secrete two classes of iron-chelating molecules known as siderophores (iron carriers): the membrane-associated mycobactins and the secreted carboxymycobactins that are able to recruit Fe from host transferrin and lactoferrin (Banerjee et al., 2011; Boradia et al., 2014; De Voss et al., 1999; Ratledge and Dover, 2000; Rodriguez, 2006; Wagner et al., 2005). Another important nutrient is carbon and its metabolism is essential for Mtb to replicate and persist in the host. Fatty acids are the principal carbon source and main substrates for the tricarboxylic acid (TCA) cycle, which serves for energy production. Several enzymes are involved in the TCA cycle to prevent carbon loss during oxidation. Among these, two enzymes are critical to the cycle: (i) isocitrate lyase (ICL), which catalyses the cleavage of isocitrate into glyoxylate and succinate, and (ii) malate synthase (MLS), which mediates the condensation of glyoxylate and acetyl-CoA into malate. Bacteria use ICL as a metabolic enzyme to grow on even-chain fatty acids through an anaplerotic pathway called the glyoxylate shunt. The Mtb genome contains two genes (icl1, icl2) that express two isoforms of ICL: isocitrate lyase 1 (ICL1) and isocitrate lyase 2 (ICL2) (Gould et al., 2006; Munoz-Elias and McKinney, 2005). Both enzymes also function as a methylisocitrate lyase (MCL), and serve in a parallel pathway involved in the metabolism of propionyl CoA generated by β- oxidation of odd-chain fatty acids, called the methylcitrate cycle (Munoz-Elias and McKinney, 2006). An ICL-deficient Mtb strain was shown to be incapable of in vitro growth on fatty acids or of growth in macrophages, and it displayed an impairment of intracellular

CHAPTER 1 INTRODUCTION 22 replication and rapid elimination from the lungs of infected mice (Munoz-Elias and McKinney, 2005). The loss of bacterial survival of ICL-deficient Mtb strain growing in media containing even (acetate) and odd (propionate) chain fatty acids was attributed to the absence of the MCL activity (Eoh and Rhee, 2014). Nitrogen is yet another essential nutrient for bacteria, which in turn uses amino acids (such as aspartate and asparagine) as a nutrient source. Recent studies have reported that the transporter AnsP1 is essential for the uptake of aspartate by Mtb (Gouzy et al., 2013) and that a second transporter, AnsP2, together with the asparaginase AnsA, are crucial for intake and processing of asparagine (Gouzy et al., 2014). Indeed, Mtb mutant strains clearly demonstrate the requirement for these genes in mycobacterial virulence (Gouzy et al., 2014; Gouzy et al., 2013). Interestingly, whether the host cell possesses a mechanism based on aspartate and asparagine deprivation remains unknown. Recent evidence suggested, however, that tryptophan (Trp) deprivation (via targeted degradation) is an effective defense against Mtb. In response, Mtb can synthesize this amino acid under stress conditions to overcome starvation, highlighting the importance of amino acids as nutritional sources for the growth of Mtb within the host (Zhang et al., 2013).

D. Cell death: necrosis vs apoptosis Cell death is important for many processes in the body, including homeostasis, development and immune regulation. Three distinct cell death pathways are defined based on morphologic and biochemical features: (i) heterophagy (“eating of another”), commonly referred to as apoptosis; (ii) autophagy (“eating of itself”); and (iii) cell death that do not involve digestion, usually called necrosis (Vanden Berghe et al., 2014). The mode of host cell death after Mtb infection is crucial for the outcome of the disease. The most commonly forms of death observed whith Mtb infected macrophages are apoptosis and necrosis. Apoptosis was considered to be the standard cell death form during development, homeostasis, infection and pathogenesis. The morphology of apoptosis is characterized by cell shrinkage and chromatin condensation (pyknosis), nuclear fragmentation (karyorrhexis), and blebbing of the outer cell membrane that culminates in apoptotic body formation. Apoptosis in general does not induce an inflammatory response (Vanden Berghe et al., 2014). The members of the caspase family of cysteine proteases play important roles in signal transduction cascades in apoptosis (Creagh et al., 2003). Three distinct molecular

CHAPTER 1 INTRODUCTION 23 pathways can induce the apoptotic process; extrinsic, intrinsic and perforin/granzyme-based apoptosis. The “extrinsic” pathway is triggered by binding of the ligands like TNF-α, FasL and CD95L, among others, to their receptors TNFR, Fas/CD95 and DR3, respectively, on the cell surface. This activates caspases proteins that cleave various death substrates leading to apoptosis (Kalimuthu and Se-Kwon, 2013). The “intrinsic” pathway of apoptosis is trigged by intracellular death signals, functionally defined by mitochondrial outer membrane permeabilization (MOMP) (Behar et al., 2011). Mitochondrial enzyme cytochrome C plays a major role in this response. Cytochrome C binds to the cytosolic protein apoptotic protease activating Factor-1 (Apaf-1) and promotes apoptosis signaling (Kim et al., 2005). The Perforin/granzyme pathway utilizes the extrinsic mode of apoptosis and it is used by cytotoxic T lymphocytes as a mechanism to kill their target cells. Perforin creates a pore on the plasma membrane of target cells and allows granzyme B, which is pro-apoptotic enzyme to access the cytosol to subsequently promote apoptosis (Mattila et al., 2015). It has been proposed that the activation of the apoptotic program contributes to immune control of Mtb (Amaral et al., 2014; Behar et al., 2010; Danelishvili et al., 2011). Several studies have suggested that Mtb uses the inhibition of apoptosis as a virulence mechanism. A number of Mtb proteins, such as the serine/threonine kinase PknE inhibit apoptosis in human cells. Indeed, an Mtb mutant strain deficient in PknE displays an increase in apoptotic infected macrophages (Jayakumar et al., 2008). Another mycobacterial factor is the Rv3364c protein that is capable of inhibiting caspase-1 activation, and consequently the host cell apoptosis (Danelishvili et al., 2011). Moreover, the Mtb protein NuoG neutralizes NOX2-derived ROS subsequently resulting in the inhibition of extrinsic apoptosis in THP-1 macrophage (Miller et al., 2010). Similarly, the products of the Mtb Rv3654c and Rv3655c genes prevent extrinsic apoptosis (Danelishvili et al., 2010). Finally, Mtb-infected macrophages produce the eicosanoid lipid mediator prostaglandin E2 (PGE2), providing yet another mechanism to prevent apoptosis in macrophages (Behar et al., 2010). Necrosis was long considered to be an “accidental” cell death that occurred due to pathological or physiological conditions such as infections, toxins or injuries. Necrosis is characterized by cytoplasmic granulation, organelle and/or cellular swelling (oncosis), loss of outer cell membrane integrity, and swelling of organelles including mitochondria and cell nuclei with release of cytoplasmic and nuclear contents to the extracellular space. Necrosis is thought to be a primordial release of damage-associated molecular patterns (DAMPs),

CHAPTER 1 INTRODUCTION 24 causing inflammatory responses of the innate immune system that stimulate PRRs receptor on macrophages, DCs and NK cells to activate T cells and initiate immune responses (Taylor et al., 2008). As previously described, while the apoptotic program contributes to immune control of Mtb, Mtb-induced necrosis has been proposed to facilitate the spreading and persistence of the bacilli (Amaral et al., 2014; Behar et al., 2010; Divangahi et al., 2009; Mayer-Barber et al., 2014).

1.3 THE CONCEPT OF NUCLEOMODULINS The past decade has seen remarkable progress in the understanding of the interaction of microbial pathogens with their hosts (Finlay and Cossart, 1997; Stebbins and Galan, 2001). It has become clear that microbial pathogens have developed a variety of strategies to manipulate host cell functions for their own benefit. Pathogenic bacteria are dependent on virulence factors to colonize and replicate in their host. Virulence factors can be protein, lipid or carbohydrate in nature, and enable pathogenic organisms to evade host defenses, promote propagation and provide access to host tissues. There are two main categories of virulence factor: toxins and effectors (Cambronne and Roy, 2006). The primary difference between these two classes of virulence factors is their mechanism of secretion. Toxins are produced within the pathogen and are secreted to the extracellular milieu by type I, type II or type V secretion systems, and contact with host cells is not necessary for their secretion (Cambronne and Roy, 2006). Effector proteins are virulence factors that are produced in the bacteria and translocated directly into infected host cells, and this secretion may be mediated by type III, type IV or type VI secretion systems (Cambronne and Roy, 2006). Once inside the host cell, effector proteins function to subvert host cell functions, resulting in a multitude of virulence phenotypes. Bacterial toxins and effectors that modulate the eukaryotic cell cycle have been termed “cyclomodulins” (Oswald et al., 2005). Historically, the focus of the interface between the host and pathogens has been between the virulence factors and host cell proteins at the surface or cytoplasm, ultimately influencing the host response through signaling cascades inducing gene expression. However, recent studies have focused on the concept of hijacking host cell function by direct interaction of pathogen- derived proteins with host nuclear components. This new class of bacterial proteins that target the nucleus, known as the “nucleomodulins” (Bierne and Cossart, 2012), can promote survival of the pathogens by modulating the epigenome of host cells and thereby modify the

Figure 7. Overview of nuclear protein import. Targeting to the nucleus involves a NLS that is recognized by an importin, which shuttles the cargo through the NPC via its central channel. Binding of Ran‐GTP inside the nucleus causes the complex to disassemble. The karyopherin can be recycled to the cytoplasm while the cargo accumulates in the nucleus. During export, an exportin together with Ran‐GTP recognizes a NES. This ternary complex is transported through the NPC central channel to the cytoplasm, where nucleotide hydrolysis is stimulated, causing Ran‐GTP to be converted to Ran‐GDP, which then releases the karyopherin and cargo. (Modified from Katta et al., 2014; Stewart, 2007).

CHAPTER 1 INTRODUCTION 25 host transcriptome and proteome. Once nucleomodulins have reached the host cell’s nucleus, the remodeling of host gene expression occurs through a variety of mechanisms (Bierne and Cossart, 2012; Silmon de Monerri and Kim, 2014). However, in eukaryotes, the nuclear membrane provides a physical barrier to macromolecules translocating from and into cytoplasm. Thus, nuclear transport is an essential step for the regulation of gene expression and replication cycle by bacterial nucleomodulins.

I. Nuclear transport Most proteins that function in the nucleus are selectively imported from the cytosol through nuclear pore complexes (NPCs), large structures that form channels allowing proteins smaller than the diameter of the channel to passively diffuse into the nucleus. Globular proteins larger than ~40 kDa are unable to passively diffuse through the NPCs and therefore require active import, which is dependent on nuclear localization signals (NLS) or nuclear export signals (NESs) (Nigg, 1997). Nuclear import and export of macromolecules are mediated by receptors called karyopherins, also known as import-competent karyopherins (importin) and export-competent karyopherins (exportin). Nuclear import of proteins is generally initiated by the formation of a ternary complex with importin-α, importin-β and a cargo, where impotin-α serves as an adapter for importin-β and recognizes NLS within the cargos. The complex facilitates movement through the central NPC channel. Once the complex is in the nucleus, the small GTPase protein Ran (Ras-related nuclear protein) promotes nuclear import of the proteins bearing NLS by binding to Importin-β and displacing it from the complex. Ran is regulated by the GTPase activating protein (GAP) and the GDP/GTP exchange factor (GEF) to form Ran-GTP in the nucleus. The NLS-protein is thus free in the nucleoplasm. Importin-α is then recycled back to the cytoplasm through the nuclear exporter CAS, whereas Importin-β is separated and transported back to the cytoplasm together with Ran-GTP. Finally, cytoplasmic RanGAP (Ran GTPase activating protein) stimulates the Ran GTPase, generating Ran-GDP, which dissociates from the importins and thereby releases them for a new NLS-protein import round (Figure 7) (Katta et al., 2014; Stewart, 2007).

CHAPTER 1 INTRODUCTION 26

Over the last few years, a growing number of bacterial effectors have been identified and characterized to target the host nucleus and nuclear import, usually through the NPCs by an active process.

II. Nuclear Localization Signals (NLS) As previously mentioned, NLS is an amino acid sequence that marks a protein for cell nucleus import by active nuclear transport. In general, NLS consists of one or more short sequences of positively charged lysine or arginine residues exposed on the protein surface. Interestingly, different nuclear localized proteins may share the same NLS. There are two classes of NLS as described below.

A. Classical nuclear localization signal (cNLS) The classical NLS consists of a short sequence enriched in basic amino acids, which is recognized by importin-β through direct binding to an adaptor protein importin-α. The classical NLS is further classified into two major classes (Lange et al., 2007): i) Monopartite NLS. These sequences contain 3–5 basic amino acids. The simian virus 40 (SV40) T-antigen NLS (126PKKKRKV132) is the canonical monopartite cNLS (Kalderon et al., 1984). The monopartite is divided into two subclasses: • Class 1 contains NLS with at least 4 consecutive basic amino acids • Class 2 contains NLS with three basic amino acids, represented by K(K/R)X(K/R) as a putative consensus sequence where X indicates any amino acid. ii) Bipartite NLS. These sequences consist of two small clusters of basic amino acids separated by about a dozen amino acids. The best-studied bipartite cNLS is found in the Xenopus laevis protein, nucleoplasmin (150KRPAATKKAGQAKKK170) (Dingwall et al., 1988). A putative consensus sequence of the bipartite NLS has been defined as (K/R)(K/R)X10–

12(K/R)3/5, where X is any amino acid and (K/R)3/5 represents three lysine or arginine residues out of five consecutive amino acids (Dingwall and Laskey, 1991).

B. Nonclassical nuclear localization signal (ncNLS) There are many other types of NLS differing from the classical ones, and most of these sequences appear to be recognized by specific receptors of the importin family:

Table 2. Different classes of NLSs allow nuclear localization of bacterial nucleomodulins. Species Effector Subcellular Targeting References localization mechanism

Agrobacterium tumefaciens VirE2 Nucleus Bipartite NLS (Citovsky et al., 1992; Citovsky et al., 1994) VirD2 Nucleus Bipartite NLS (Howard et al., 1992; Rossi et al., 1993) Xanthomonas spp. AvrBs3 Nucleus Monopartite NLS (Van den Ackerveken et al., 1996; Yang and Gabriel, 1995; Szurek et al., 2002) AvrXa5 Nucleus Monopartite NLS (Yang and Gabriel, 1995; Zou et al., 2010) AvrXa7 Nucleus Monopartite NLS (Yang et al., 2000) AvrXa10 Nucleus Monopartite NLS (Yang and Gabriel, 1995) PthA Nucleus Monopartite NLS (Yang and Gabriel, 1995) Yersinia spp. YopM Nucleus Nonclassical NLS (Benabdillah et al., 2004) Ralstonia solanacearum PopP2 Nucleus Bipartite NLS (Deslandes et al., 2003) PopB Nucleus Bipartite NLS (Gueneron et al., 2000)

CHAPTER 1 INTRODUCTION 27

i) ncNLS recognized by importin-α These are defined as class 3 (KRX (W/F/Y) XXAF) and class 4 (P/R) XXKR (K/R) NLS that bind directly to the minor binding pocket of importin-α (Kosugi et al., 2009a). This is in contrast to the class 1 and class 2 cNLS that specifically bind to the major binding pocket of importin-α. Class 5 NLS has a LGKR (K/R)(W/F/Y) core sequence and is specific only to plants, in contrast with the other classes, all of which are functional in yeast, plants and mammals (Kosugi et al., 2009b). ii) ncNLS recognized by importin-β2 It has been reported that proline-tyrosine NLS (PY-NLS) binds directly to importin-β2 (Transportin-1). The PY-NLS contains loose sequence motifs (N-terminal hydrophobic or basic motifs and a C-terminal R/K/H-X2–5-P-Y motif) (Soniat and Chook, 2015).

III. Nucleomodulins An increasing number of studies indicate that various animal and plant pathogenic bacteria can deliver nucleomodulins to the nucleus of infected host cells and, interestingly, different NLSs have been identified as key elements for their entry into the host nucleus (see Table 2). Bacterial molecules nucleomodulins use diverse strategies to hijack nuclear processes, as described as follows. Therefore, the study of bacterial nucleomodulins can generate new insights into the long-term impact on the host by different human pathogens, and contribute to a better understanding of the regulation of the host nuclear dynamics.

A. Modulation of host signaling pathways In general, a signaling pathway is a biochemical cascade that comprises a series of chemical reactions. Acting on a receptor, a stimulus (first messenger) usually initiates a signaling pathway that is transduced to the cell interior through second messengers, which amplify the initial signal and ultimately to effector molecules, such as transcription factors. Thereby, the signaling pathway often results in a cell response to the initial stimulus. At each step of a signaling cascade, multiple controlling factors converge to regulate cellular actions, responding to different changes happening in their internal and external environments, including pathological situations such as a infection by a pathogen. For instance, one of the most well conserved signaling pathways in plants and animals is that controlled by the family

CHAPTER 1 INTRODUCTION 28 of TLR. Indeed, this family of receptors is known to control the innate immune response to pathogenic challenge. TLRs are a type of PRR and recognize molecules that are mainly shared by pathogens, yet distinguishable from host molecules, collectively known to as pathogen PAMP. These receptors are broadly expressed in sentinel cells, such as macrophages and dendritic cells, which recognize microbial PAMPs. Once a pathogen has gotten around a physical barriers (e.g. the skin, intestinal tract mucosa), it is recognized by TLRs and subsequently activate a rapid innate immune response (Akira and Takeda, 2004). Most of the TLR-induced responses go through the mitogen-activated protein kinases (MAPKs), which are a highly conserved family of serine/threonine protein kinases involved in a variety of fundamental cellular processes such as proliferation, differentiation, motility, stress response, apoptosis, and survival. Following pathogen infection or tissue damage, the stimulation of TLR and other PRRs on the cell surface and in the cytoplasm of innate immune cells activates members of each of the conventional MAPKs including the extracellular signal- regulated kinase (ERK), p38 and Jun N-terminal kinase (JNK) subfamilies. In conjunction with the activation of nuclear factor-κB (NFκB) and interferon-regulatory factor (IRF) transcription factors, MAPK activation induces the expression of multiple genes that together regulate the inflammatory response (Arthur and Ley, 2013). As a defense against a threatening immune response by the host, such as TLR-induced signaling pathways, some pathogens have evolved to modulate and control these signaling cascades in order to increase their fitness in the host. For instance, the virulence factor OspF of the human pathogen Shigella flexneri displays a phosphothreonine lyase activity, which removes phosphate groups from Erk2 and p38 from host MAPKs pathway and prevents MAPK-dependent H3S10 phosphorylation (Brennan and Barford, 2009; Li et al., 2007). Multiple effectors modulating host signaling pathways have been identified in several species of pathogens including bacteria, but also virus, yeasts and protozoan (Table 3).

B. Direct targeting of host nuclear proteins In eukaryotic cells, genomic DNA is folded with histone and non-histone proteins to form chromatin. Each chromatin unit, or nucleosome, contains 146 bp of DNA, which is wrapped around the 8 cores histone that includes two copies of H2A, H2B, H3, and H4. The primary functions of chromatin are not only to package DNA into a smaller volume to fit in the cell, but also to control gene expression and DNA replication. In mammals, chromatin is mainly

Table 3. Strategies exploited by pathogens to modulate the host signaling pathways.

Organism Effector protein Target molecule References

Shigella flexneri OspF MAPKs (Arbibe et al., 2007) Salmonella spp. SpvC MAPKs (Mazurkiewicz et al., 2008) Adenovirus 5 EB1‐55K DAXX (Schreiner et al., 2010) Chlamydia spp. Unknown ZNF23 (Soupene et al., 2012) Chlamydia trachomatis CT441 p65/ReI (Lad et al., 2007) Toxoplasma gondii Unknown STAT1 (Schneider et al., 2013)

Table 4. Strategies exploited by pathogens to modulate the host nuclear proteins.

Organism Effector Target molecule References protein

Toxoplasma gondii GRA16 HAUSP deubiquitinase and PP2A (Bougdour et al., 2013) phosphatase EBV EBNA3C Polycomb, mSin3A, NCoR, histone (Knight et al., 2003) deacetylases Shigella flexneri OspB, OspF Rb tumor suppressor proteins (Zurawski et al., 2009) Anaplasma AnkA SHP‐1 (Garcia‐Garcia et al., 2009) phagocytophilum Listeria monocytogenes LntA BAHD1 (Lebreton et al., 2011)

Figure 8. Bacterial strategies to control gene expression in the nucleus. To prevent chromatin remodeling and maintain a silenced state, Mtb LpqH lipoprotein binds to SWI/SWF remodeling complexes and blocks their function. L. monocytogenes regulates chromatin state via the effector protein LntA, which recruits heterochromatin regulator BAHD1 to recruit heterochromatin proteins and induce formation of heterochromatin. HIV, uses vpr protein to target p300/HAT complexes, causing them to dissociate from chromatin. Alternatively, some pathogens express proteins that directly bind DNA to induce transcription or prevent it. Hepatitis C virus expresses NS5A, which binds promoter regions of host genes. S. flexneri prevents transcription by sequestering host transcription factors, such as the Rb tumor‐suppressor proteins. Chromatin state is also regulated by histone post‐translational modifications, which can be modulated through manipulation of host enzymes or directly through secreted effector enzymes. For example, S. flexneri modulates the phosphorylation of histone H3S10 through the activity of OspF, a secreted phosphothreonine lyase. OspF removes phosphate groups from Erk2 and p38, two members of the MAPK pathway, which prevents MAPK‐dependent H3S10 phosphorylation. Gray line, DNA; red line, silenced promoter; red circles, histone PTMs. Me, cytosine methylation (Silmon de Monerri and Kim, 2014).

CHAPTER 1 INTRODUCTION 29 found as a condensed transcriptionally silent form called heterochromatin. In contrast, euchromatin is less condensed and contains the most actively transcribed genes. A number of proteins participate in shaping the chromatin structure including histones and other chromatin interacting proteins such as transcription factors (Grewal and Jia, 2007). Indeed, the regulation of gene transcription in eukaryotes is mainly controlled by enhancers and transcription factors (TFs), which are proteins that control which genes are turned on or off in the genome. These factors recognize specific DNA sequences and their binding to these sites activates transcription of their target genes, independent of their location or distance to the promoters of these genes. Upon binding, one TF can initiate a series of interactions between multiple proteins that may serve as activators or repressors of gene transcription (Levine, 2010). Several examples have been described of microbial effectors that modulate host gene transcription, through direct or indirect mechanisms (Bierne and Cossart, 2012) (Table 4 and Figure 8). Xanthomonas species can inject into cells of infected plants, through a type 3 secretion system, transcription activator-like (TAL) effectors. Once inside plant cells, TALs enter the nucleus, bind to TAL-specific DNA sequences and activate the expression of the target genes (Boch and Bonas, 2010; Bogdanove et al., 2010). For instance, the effector AvrBs3 from Xanthomonas campestris activates expression of the regulator Upa20, inducing cell hypertrophy and thereby facilitating bacterial colonization (Kay et al., 2007). In addition to such examples of a direct interaction of pathogen-derived proteins with DNA mimicking eukaryotic transcription factors, other effectors can indirectly influence gene transcriptional activity. For instance, Shigella flexneri secretes OspB and OspF effector proteins that interact with members of the human retinoblastoma protein Rb complexes (essential factors for normal cell growth) to prevent their binding to DNA (Zurawski et al., 2009). Another example is that of the parasite Toxoplasma gondii, which secretes the virulence factor GRA16 to control the activity of host nuclear proteins such as the herpes virus-associated ubiquitin-specific protease (HAUSP) and protein phosphatase 2A (PP2A). Upon binding, GRA16 induces the translocation of PP2A into nucleus and assembly with HAUSP into complex, resulting in important changes of host cell cycle functions and cell proliferation (Bougdour et al., 2013). Microbial effector proteins can also influence the histone modification by facilitating direct effects on chromatin structure. For example, the human pathogen Listeria monocytogenes is able to regulate immune response genes by

Table 5. Strategies exploited by pathogens to modulate the host post‐translationaly. Organism Effector protein Target molecule References

Chlamydia trachomatis NUE methyltransferase Host chromatin, (Pennini et al., 2010) histones Streptococcus pyogenes Ser/Thr phosphatase Host chromatin (Agarwal et al., 2012) SP‐STP Mtb 19 kDa lipoprotein SWI/SNF and (Pennini et al., 2007; LpqH C/EBPβ Pennini et al., 2006). Mtb Mycobacterial protein Histones (Sharma et al., 2015) Rv2966, 5‐ methylcytosine‐ specific DNA methyltransferase Legionella pneumophila RomA Histone H3 K4 (Rolando et al., 2013). methyltransferase Paramecium bursaria Chorella virus Histone H3K27 (Manzur et al., 2003) chorella virus methyltransferase

CHAPTER 1 INTRODUCTION 30 reversing the formation of heterochromatic regions (Lebreton et al., 2011). This is achieved by interfering with the function of bromo adjacent homology domain-containing protein 1 (BAHD1), a repressor protein that promotes the formation of heterochromatin by recruiting proteins involved in heterochromatin assembly (Bierne et al., 2009). L. Monocytogenes cells present in the cytoplasm of infected host epithelial cells secrete the effector protein called listeria nuclear targeted protein A (LntA). LntA is transported into the nucleus, binds BAHD1 and colocalizes with it at heterochromatic regions, ultimately resulting in impaired binding of BAHD1 to promoters and stimulation of type III interferon (IFN) response (Lebreton et al., 2011).

C. Post-translational modification of host nuclear proteins Post-translational modifications (PTMs) are known to be essential mechanisms used by eukaryotic cells to diversify their protein functions and dynamically coordinate their signaling networks. The discovery and investigation of PTMs such as methylation, acetylation, phosphorylation and SUMOylation, among many others, has established both nuclear and non-nuclear roles for PTMs. PTMs can directly impact cell function by modifying histones, enzymes (and their associated activity) and assembling protein complexes (Berger, 2007; Lee et al., 2010). Various bacterial effectors manipulate the host cell cycle by interference of PTMs (Table 5 and Fig. 8). Some act as “mimics” of host chromatin modifiers to directly induce histone modifications. These include, for example, the nuclear effector E (NUE) that is secreted by Chlamydia trachomatis, localizes to host cells nuclei, binds to chromatin and behaves as a histone methyltransferase (Pennini et al., 2010). Another secreted bacterial methyltransferase, Legionella pneumophila RomA, is a member of a group of genes encoding proteins with high similarity to eukaryotic proteins. Like NUE, RomA targets histones for methylation, inducing tri-methylation of histone H3K14 (Rolando et al., 2013).

D. Can Mtb modulate gene expression of infected host cells? In the case of Mtb, a protein able to modulate chromatin is involved during the classical inhibition of expression of some IFN-γ-responsive genes, such as the MHC class II transactivator (CIITA) (Kincaid and Ernst, 2003). Indeed, during Mtb infection, mycobacterial cell wall fractions containing 19-kDa lipoprotein (LpqH, Rv3763) prevent the binding of the

CHAPTER 1 INTRODUCTION 31

SWItch/Sucrose NonFermentable (SWI/SNF) chromatin-remodeling complex to chromatin, leading to the inactivation of CIITA expression (Pennini et al., 2007; Pennini et al., 2006). However, the mechanism of this effect is dependent on TLR2-induced MAPK signaling, and 19-kDA lipoprotein has not been shown to be a nucleomodulin. A recent study showed that the mycobacterial protein Rv2966c is a 5-methylcytosine- specific DNA methyltransferase that is secreted from the Mtb cells and localizes in part in the cytoplasm and in part in the nucleus of the host cells. Rv2966c methyltransferase binds to and methylates specific host DNA sequences, resulting in the repression of host gene expression. In addition, this activity seems to be controlled by the phosphorylation status of the protein Rv2966c, which is also shown to interact with host proteins, such as the mammalian DNA methyltransferase, DNMT3L (Sharma et al., 2015). Most importantly, this study demonstrated that proteins secreted by Mtb phagocyted into macrophages can be transported to the nucleus of the infected cells, strongly arguing in favor of the existence of nucleomodulins as a strategy for Mtb to subvert host response and increase its fitness during infection. This is confirmed by recent results, still ahead of print, that identified a histone acetyltransferase, produce by the Rv3423.1 gene of Mtb, that can reach the nucleus of infected cells and modify the chromatin (Jose et al., 2015).

1.4 STATEMENT OF THE PROBLEM Host organisms respond to infection by initiating inflammatory and immune responses in an attempt to clear out infectious microorganisms. Pathogens have adapted to alter host cell functions to their own advantage, to promote their survival, and in the case of intracellular pathogens, to generate a suitable environment for replication within the host cell. Pathogens use a variety of strategies to manipulate host cells to their benefit. In the past, most studies focused on the mechanisms used by bacteria to modify gene expression and cell proliferation via interactions with cell surface and/or cytoplasmic components of the infected cells. Recently, it has become increasingly evident that bacterial factors can also act directly within the nucleus, as evidenced by nucleomodulins. While there have been a few studies examining potential mycobacterial nucleomodulins (Jose et al., 2015), there is a great need to identify and characterize the role of this type of factors during TB pathogenesis.

CHAPTER 1 INTRODUCTION 32

1.5 PURPOSE The purpose of this study was to identify potential nucleomodulins in Mtb. To this end, we first performed a screening of Mtb genome, to identify genes encoding proteins with NLS that could be secreted out of the bacterial cells. Here, I report the identification of two novel mycobacterial proteins that are imported into the eukaryotic nucleus in an NLS-dependent manner, and I describe an approach to characterize their mode of action in the host cell.

CHAPTER 2 MATERIALS AND METHODS

2.1 IN-SILICO PREDICTION OF MTB PROTEINS TARGETED TO THE HOST CELL NUCLEUS In order to perform overall structure and functional predictions, the coding nucleotide and amino acid sequence of Mtb H37Rv strain were obtained from Tuberculist server (http://tuberculist.epfl.ch/). To look for the presence of NLS sequences, a search was performed in the server of the Comprehensive Microbial Resource (CMR) (http://cmr.jcvi. org/cgibin/CMR/shared/MakeFrontPages.cgi?page=cmr_search&search_type=motif&crumb s=genomes). Map NLS server (http://nls-mapper.iab.keio.ac.jp/cgibin/NLS_Mapper_ form.cgi) was used to calculate the “NLS score” of each candidate protein (Kosugi et al., 2009b). The secretion property of the candidate proteins was verified using review of the existing literature (Albrethsen et al., 2013; Gey van Pittius et al., 2006).

2.2 REAGENTS AND CHEMICALS I. Commercial Reagents RPMI 1640 medium, GlutaMAX™ Supplement, β-mercaptoethanol and sodium pyruvate were purchased from Gibco Laboratories, Life Technologies (Carlsbad, CA). Fetal bovine serum (FBS), cell dissociation solution, ampicillin sodium salt, streptomycin sulfate salt, Tween-80, Luria-Bertani (LB) Broth (Lennox), dimethylsulfoxide (DMSO), Fluoroshield™ with DAPI (4’-6-diamidino-2-phenylindole) and Middlebrook 7H10 agar were purchased from Sigma-Aldrich (St. Louis, MO). Kanamycin sulfate was purchased from Fluka chemic GmbHC (Buchs, Switzerland), hygromycin from Invitrogen (Carlsbad, CA) and albumin-dextrose- catalase supplement (ADC), oleic acid-albumin-dextrose-catalase supplement (OADC) and 7H9 culture media were from Becton, Dickinson and Company (Sparks, MD). Pierce™ 16% Formaldehyde (w/v), Methanol-free was purchased from Thermo Fisher Scientific (Rockford, IL).

Figure 9. Mammalian expression vector. ((A) pEGFP‐C1 is mammalian expression vector with CMV promoter, EGFP tag and kanamycin resistance (neomycin for mammalian selection). (B) pIRES‐hrGFP II vector contains a bicistronic expression cassette in which a protein of interest tagged with 3 copies of the FLAG®epitope can be expressed after cloning of its gene into the multiple cloning site (MCS). An internal ribosome entry site (IRES) followed by hrGFP II, encoding a humanized and optimized version of GFP, allows for the monitoring of the presence of the vector with green fluorescence. CHAPTER 2 MATERIALS AND METHODS 34

II. Enzymes and kits Restriction endonucleases were obtained from New England Biolabs (Frankfurt, Germany) and Fermentas (St. Leon-Rot, Germany). 2x Phusion Master Mix with GC Buffer A was purchased from Thermo Scientific (Foster City, CA). GeneJET Gel Extraction Kit and T4 DNA ligase were purchased from Fermentas (St. Leon-Rot, Germany). Mini prep kits for DNA isolation were obtained from Qiagen (Hilden, Germany). Gateway® BP Clonase® II Enzyme mix, Gateway® LR Clonase™ Enzyme Mixes and Lipofectamin® 2000 transfection reagent were purchased from Invtrogen, Life Technologies (Carlsbad, CA). ViaFect™ Transfection Reagent was purchased from Promega (Madison, WI).

III. Oligonucleotides and Sequencing All oligonucleotides were synthesized by Sigma-Aldrich and DNA sequenced by Eurofins MWG Operon (Ebersberg, Germany).

2.3 CLONING AND EXPRESSION I. Amplification of DNA fragments by PCR The PCR reagents from Themo scientific were used for amplification. A typical reaction was set up according to the conditions given below. The reaction mixture contained: 2x Phusion GC master mix 25 µl Primers Forward (10mM) 2.5 µl Primers Reverse (10 mM) 2.5 µl DMSO 5 µl Template DNA 0.10-10 ng MilliQ (MQ) water to 50 µl (final volume)

The amplification conditions is given below: Initial denaturation 95 °C for 5 min 1 cycle Denaturation 95 °C for 1 min Annealing 55 °C for 1 min 30 cycles Extension 72 °C for 2 min Final extension 72 °C for 5 min 1 cycle

Table 6. Cloning primers used for plasmid constructions.

Primers Sequence 5’‐3’ Site Rv0229c‐For GAATTGGAATTCAGATCTCGTGAGACAGCCGCGCCGGGCG BglII Rv0229c‐Rev GACAAGCTTGAGATCTCTAGTCGATCGTGCCCGCTG BglII Rv0286‐For GAATTGGAATTCAGATCTCATGGCCGCGCCCATCTGGATG BglII Rv0286‐Rev GACAAGCTTGAGATCTTTACTTGCTGTCGTGCGGTAAG BglII Rv1798‐For GAATTGGAATTCAGATCTCATGACACGCCCGCAGGCCGCCG BglII Rv1798‐Rev GACAAGCTTGAGATCTCTACTTCGTCTTCTTCTCCGCG BglII Rv3866‐For GAATTGGAATTCAGATCTCATGACGGGTCCGTCCGCTG BglII Rv3866‐Rev GACAAGCTTGAGATCTTCAGCCTCGGGAGGAGGCTTG BglII Rv3876‐For GAATTGGAATTCAGATCTCATGGCGGCCGACTACGACAAG BglII Rv3876‐Rev GACAAGCTTGAGATCTTCAACGACGTCCAGCCCTCTCG BglII Rv3882c‐For GAATTGGAATTCAGATCTCATGAGAAATCCTTTAGGGCTG BglII Rv3882c‐Rev GACAAGCTTGAGATCTCTACTTCGGCAGCGCCATCTG BglII Rv3888c‐For GAATTGGAATTCAGATCTCGTGACGAACCCGTGGAATG BglII Rv3888c‐Rev GACAAGCTTGAGATCTTCACTGCGCCGCCCGTTCGG BglII MMA2680‐For GAATTGGAATTCAGATCTCATGACCCGCGCACAATCAGCTG BglII MMA2680‐Rev GACAAGCTTGAGATCTCTAGTTCGTCTTCTTCTCTG BglII 0229‐IRES‐Fw AGGAATTCTGCAGATGCCGCCACCATGAGACAGCCGCGCCGGGCG EcoRV 0229‐IRES‐Rv GCCAGTGTGATGGATAAGCTTGTCGATCGTGCCCG EcoRV 3876‐IRES‐Fw AGGAATTCTGCAGATGCCGCCACCATGGCGGCCGACTACGACAAG EcoRV 3876‐IRES‐Rv GCCAGTGTGATGGATAAGCTTACGACGTCCAG EcoRV Nde‐His0229‐Fw GGGGCATATGCATCATCACCATCACCACGTGAGACAGCCGCGCCGGGCG NdeI Hind‐0229His‐Rv GGGGAAGCTTGTCGATCGTGCCCGCTGGCACG HindIII Nde‐His3876‐Fw GGGGCATATGCATCATCACCATCACCACGCGGCCGACTACGACAAG NdeI Hind‐3876His‐Rv GGGGAAGCTTACGACGTCCAGCCCTCTCG HindIII clo‐RBD‐rv0229cHis‐attB2 GGGGACAGCTTTCTTGTACAAAGTGGAGGAAGACAGGCTGCCCATGGTGAGACAGCCGCGCCGGGCG clo‐rv0229c6His‐attB3 GGGGACAACTTTGTATAATAAAGTTGTCAGTGGTGGTGGTGGTGGTGAAGCTTGTCG clo‐RBS‐rv3876His‐attB2 GGGGACAGCTTTCTTGTACAAAGTGGAGGAAGACAGGCTGCCCATGATGGCGGCCGACTACGACAAG clo‐rv3876His‐attB3 GGGGACAACTTTGTATAATAAAGTTGTCAGTGGTGGTGGTGGTGGTGAAGCTTACGACG

Table 7. Construction of the plasmids used in this study.

Final plasmid Template Primer Restriction Vector Encoded Promoter sits product pEGFP‐Rv0229c Mtb H37Rv Rv0229c‐For BglII pEGFP‐C1 EGFP‐Rv0229c CMV Rv0229c‐Rev pEGFP‐Rv0286 Mtb H37Rv Rv0286‐For BglII pEGFP‐C1 EGFP‐Rv0286 CMV Rv0286‐Rev pEGFP‐Rv1798 Mtb H37Rv Rv1798‐For BglII pEGFP‐C1 EGFP‐Rv1798 CMV Rv1798‐Rev pEGFP‐Rv3866 Mtb H37Rv Rv3866‐For BglII pEGFP‐C1 EGFP‐Rv3866 CMV Rv3866‐Rev pEGFP‐Rv3876 Mtb H37Rv Rv3876‐For BglII pEGFP‐C1 EGFP‐Rv3876 CMV Rv3876‐Rev pEGFP‐Rv3882c Mtb H37Rv Rv3882c‐For BglII pEGFP‐C1 EGFP‐Rv3882c CMV Rv3882c‐Rev pEGFP‐Rv3888c Mtb H37Rv Rv3888c‐For BglII pEGFP‐C1 EGFP‐Rv3888c CMV Rv3888c‐Rev pEGFP‐MMA2680 M. marinum MMA2680‐For BglII pEGFP‐C1 EGFP‐ CMV MMA2680‐Rev MMA2680 pVV16‐Rv0229c‐His pEGFP‐Rv0229c Nde‐His0229‐Fw NdeI, pVV16 Rv0229c‐His hsp60 Hind‐0229His‐Rv HindIII pVV16‐Rv3876‐His pEGFP‐Rv3876 Nde‐His3876‐Fw NdeI, pVV16 Rv3876‐His hsp60 Hind‐3876His‐Rv HindIII p3Flag‐Rv0229c‐IRES‐hrGFPII pEGFP‐Rv0229c 0229‐IRES‐Fw EcoRV pIRES‐GFPII Rv0229c CMV 0229‐IRES‐Rv p3Flag‐Rv3876‐IRES‐hrGFPII pEGFP‐Rv3876 3876‐IRES‐Fw EcoRV pIRES‐GFPII Rv3876 CMV 3876‐IRES‐Rv CHAPTER 2 MATERIALS AND METHODS 35

II. Plasmid constructions A. GFP-Fusion plasmid clones To construct the final expression vector of Mtb candidate genes tagged with GFP, the expression cassette was amplified using Mtb H37Rv genomic DNA as the template with the primer pair Forward and Reverse (Table 6). The primers also contained 15 bp identical to that adjacent to the insertion sites in the targeting vector, pEGFP-C1 (the intermediate vector, Fig. 9A). After sequence verification, the intermediate vector was linearized with BglII, and In-Fusion assembly reactions (using the Clontech In-Fusion PCR Cloning Kit, Mountain View, CA) were performed with this linearized intermediate vector and the PCR amplified expression cassette to generate the final expression vector (Table 7).

B. Mammalian expression vector constructs p3Flag-Rv0229c-IRES-hrGFP II and p3Flag-Rv3876-IRES-hrGFP II are variants of pIRES- hrGFP II (Agilent Technologies, Santa Clara, CA; Fig. 9B). They carry a bicistronic expression cassette allowing the simultaneous but separate expression of the FLAGTM-tagged protein of interest and of hrGFP (to visualize successfully transfected cells by green fluorescence). The open reading frames of Rv0229c and Rv3876 were amplified by PCR using primers containing a 15 bp overlap on both sides of the insertion site for In-Fusion recombineering. The forward primer was designed to add a unique EcoRV site, followed by a consensus Kozak sequence (GCCGCCACC) immediately 5’ to the ATG start codon Rv0229c or Rv3876. The reverse primer was designed to generate a unique EcoRV site 3’ to the Rv0229c and Rv3876 open reading frame (Table 6). The intermediate vector pIRES-hrGFP II was linearized with EcoRV, and In- Fusion assembly reactions were performed with this linearized intermediate vector and the PCR amplified expression cassette to generate the final expression vector. DNA sequencing was performed to verify the sequence of all the modules used during the construction (Sequencing primer see table 11).

C. Generation of plasmid constructs for overexpression Mtb Rv0229c and Rv3876 were amplified from the Mtb H37Rv genomic DNA using gene specific primers (Table 6). PCR products were digested with NdeI and HindIII (New England Biolabs, NEB) then cloned downstream to the heat shock protein 60 (hsp60) promoter of the

Figure 10. E. coli and mycobacterium shuttle plasmid. pVV16 contains two origins of replication, active either in E. coli or in GroEL2 gene. Cloning of an ORF without stop codon in 3’ allows for the construction of a variant protein carrying a 6his‐tag at its C‐terminal end.

Table 8. Cloning primers used for mutagenesis.

Primers Sequence 5’‐3’ Site

Rv0229c‐MluI‐Fw ACGTCTCAGCGAGTTCAACGCGTGCTGCAGCTTCTCGACACGCTGGCCGCCG MluI Rv0229c‐MluI‐Rv GCTTACAATTTACGCGTTAAGATACATTG MluI Rv0229c‐AAHAA‐Fw GATTCATGGCGAAACATCTGCTGCACATGCAGCTGCAGGCTTTAAACATGGCTCG BglII Rv0229c‐AAHAA‐Rv CGAGCCATGTTTAAAGCCTGCAGCTGCATGTGCAGCAGATGTTTCGCCATGAATC BglII Rv3876‐Fw‐Ala3 GCCGCAACCACTGGCGGTGCTGCAGCTGCGCGTGCAGCGCCGGATCTCGACG BglII Rv3876‐Rv‐Ala4 GCGTCGAGATCCGGCGCTGCACGCGCAGCTGCAGCACCGCCAGTGGTTGCGG BglII Rv3876‐Fw‐Ala5 GCGGCCAAGGGGCCGAAGGTGGCTGCAGTGGCGCCCCAGAAACCGAAGGCCACG BglII Rv3876‐Rv‐Ala6 GTGGCCTTCGGTTTCTGGGGCGCCACTGCAGCCACCTTCGGCCCCTTGGCCG BglII

CHAPTER 2 MATERIALS AND METHODS 36

Escherichia coli and mycobacterium shuttle plasmid, pVV16 (Stover et al., 1991) (Fig. 10). Plasmid pVV16-Rv0229c and pVV16-Rv3876 were verified by DNA sequencing.

D. Generation of plasmid constructs for overexpression under inducible promoter Expression of the genes Rv0229c and Rv3876 under the control of a Tet-inducible promoter was achieved after cloning in vectors of the Gateway system developed by the groups of S. Ehrt and D. Schnappinger, at Weill Cornell Medical College, New York (Ehrt et al., 2005). PCR were performed in a total volume of 50 µl as described above, pVV16-Rv0229c or pVV16-Rv3876 as the template with the primer pair Forward and Reverse (Table 6), which generated the full-length attB2 and attB3 sites flanking the ORFs. Gateway compatible amplified ORFs were recombined into the pDO23A vector (Ehrt et al., 2005) by “BP” recombination reaction (10 µl final volume): 20-50 fmoles PCR products, 2 µl BP clonase II Enzyme Mix (Invitrogen), 150 ng pDO23A vector plasmid and TE buffer (pH 8.0) were incubated at 25 °C for 1 h. Proteinase K 1 µl was added to the DNA samples and incubated at 37 °C for 10 min. BP reactions were used directly for bacterial transformation of E. coli MACH1 cells, and bacteria were plated on LB medium containing 100 μg/ml of ampicillin. Plasmid (“entry” clones) prepared from the transformants was tested by restriction enzymes, and positive clones were finally sequenced. These entry clones were set up in “LR” reactions for recombination of target genes into the destination vector. Entry clones were used for generation of pGMC-TetR-P1-Rv0229c or pGMC-TetR-P1-Rv3876 clones in LR recombination reaction (10 µl final volume): 2 μl of 5xLR clonase II Enzyme Mix (Invitrogen), 20-50 fmoles pEN23A vector, 20-50 fmoles pEN41A vector, 20-50 fmoles pEN12A vector, 60 ng pDE43- MCS vector (Ehrt et al., 2005), and TE buffer (pH 8.0), and incubation at 25 °C for 16 h. After addition of 1 µg proteinase K and incubation at 37 C for 10 min, the LR reactions were used for transformation of E. coli MACH1 cells and transformants were selected in plates containing 25 μg /ml streptomycin.

E. Site directed mutagenesis Site directed mutagenesis was performed using In-Fusion PCR Cloning Kit. The amino acids in the predicted NLS, functionally important positively charged amino acids were mutated to neutral alanine residues. For stepwise mutagenesis primer sets described in Table 8 were used. For Rv0229c double mutation (pEGFP-Rv0229cD) PCR products from

Table 9. Construction of putative NLS variants by directed mutagenesis. Final plasmid Template Primer Restriction Vector Encoded Promoter sits product pEGFP‐Rv0229cM1 pEGFP‐ Rv0229c‐MluI‐Fw MluI pEGFP‐C1 EGFP‐ CMV Rv0229c Rv0229c‐MluI‐Rv Rv0229cM1 pEGFP‐Rv0229cM2 pEGFP‐ Rv0229c‐For BglII pEGFP‐C1 EGFP‐ CMV Rv0229c Rv0229c‐AAHAA‐Rv Rv0229cM2 Rv0229c‐AAHAA‐Fw Rv0229c‐Rv pEGFP‐Rv0229cD pEGFP‐ Rv0229c‐For BglII pEGFP‐C1 EGFP‐ CMV Rv0229cM1 Rv0229c‐Rv Rv0229cD pEGFP‐ Rv0229cM2 pEGFP‐Rv3876M1 pEGFP‐ Rv3876‐For BglII pEGFP‐C1 EGFP‐ CMV Rv3876 Rv3876‐Rv‐Ala4 Rv3876M1 Rv3876‐Fw‐Ala3 Rv3876‐Rv pEGFP‐Rv3876M2 pEGFP‐ Rv3876‐For BglII pEGFP‐C1 EGFP‐ CMV Rv3876 Rv3876‐Rv‐Ala6 Rv3876M2 Rv3876‐Fw‐Ala5 Rv3876‐Rv

Table 10. Primers used for Mtb mutant construction. Primers Sequence 5’‐3’

0229cAm‐Fw GAATCCTCAAGTGCGCCACCG 0229c‐Av‐Rv GGTCGGGTTGCCGCATCGG 0229cZeo‐Av‐Fw CCACTGAGCGTCAGACCCACGTGCTCCCAGCGGGCACGATCGACTAG 0229cZeo‐Am‐Rv CAGTCGATCCACGTGGAGGCTGTCTCACTGATAGCCG 3876Am‐Fw CGGCGTCCAATACTCGAGG 3876‐Av‐Rv GCATCGCCATCCCGATGACGG 3876Zeo‐Av‐Fw CCACTGAGCGTCAGACCCACGTGCTCGGACGTCGTTGAGCGCACCTG 3876Zeo‐Am‐Rv CAGTCGATCCACGTGGAGGAAGAGCTTGTCGTAGTCGG Zeo‐Dir CTCCACGTGGATCGACTGCCAGGC Zeo‐Rev GAGCACGTGGGTCTGACGCTCAGTGG

CHAPTER 2 MATERIALS AND METHODS 37 pEGFP-Rv0229cM1 and pEGFP-Rv0229cM2 were amplified again with secondary adapter primers to generated variants of expression vector tagged with GFP (Table 9).

F. Construction of Knockout Mutant of Rv0229c and Rv3876 of Mtb The Mtb genes Rv0229c and Rv3876 were disrupted by using the “recombineering method” (van Kessel and Hatfull, 2007). The linear allelic exchange substrate (AES) was constructed using a three fragments fusion-PCR (Stover et al., 1991) with the indicated primers (Table 10). The AES was constituted, from 5’ to 3’, by the fusion of the 700 bps of DNA upstream from the initiator codon of Rv0229c/Rv3876, followed by the Zeocin (InvivoGen) resistance gene (Sh ble,) from the pZeo vector (Drocourt et al., 1990) and ended by the 700 bps of DNA downstream from the stop codon of Rv0229c and Rv3876. The substrate was subsequently introduced by electro-transformation into a recipient Mtb strain already containing a hygromycin-resistant derivative of the recombineering plasmid pJV53 encoding a phage recombinase (van Kessel and Hatfull, 2007). The construction and the selection of the mutant were essentially performed as previously described (van Kessel and Hatfull, 2007). The chromosomal structure of the mutant was verified by PCR with suitable primers amplifying the complete AES.

G. Recombineering experiments in Mtb Mtb H37Rv and H37Rv/pJV53H was grown in 7H9, OADC, Tween 80 and hygromycin (50 µg/ml) in agitated cultures at 37 °C. In mid-log phase of growth, acetamide was added (0,02 % final concentration) and electrocompetent cells, were prepared and electrotransformed with 100ng of AES, as described by (van Kessel and Hatfull, 2007). Electrotransformed cells were incubated for 72h in 7H9 ADC Tween without antibiotic and plated on selective plates: 7H11 OADC zeomycin (25 µg/ml). After 3 week at 37°C, zeomycin- resistant clones were picked and grown in 7H9 ADC tween. Their genomic DNA was extracted and tested for the allelic replacement with appropriate oligonucleotides. The plasmid pJV53H was then lost by serial rounds of culture without antibiotic and ZeoR HygroS clones were verified again by PCR and conserved.

Table 11. Sequencing primers used in this study.

Plasmid/PCR product Primers Sequence 5’‐3’

pEGFP‐Rv0229c CMVfor CGCAAATGGGCGGTAGGCGTG pEGFP‐Rv0286 pEGFPN1rev GTCCAGCTCGACCAGGATG pEGFP‐Rv1798 pEGFP‐Rv3866 pEGFP‐Rv3876 pEGFP‐Rv3882c pEGFP‐Rv3888c pEGFP‐MMA2680 pEGFP‐Rv0229cM1 pEGFP‐Rv0229cM2 pEGFP‐Rv0229cD pEGFP‐Rv3876 CMVfor CGCAAATGGGCGGTAGGCGTG pEGFP‐Rv3876M1 pEGFPN1rev GTCCAGCTCGACCAGGATG pEGFP‐Rv3876M2 Seq‐Rv3876‐1 GCTATCGCACAGACACCG Seq‐Rv3876‐2 CGCGTCAACGCATGCACC pVV16‐Rv0229c‐His pVV‐U ACAACTTGAGCCGTCCGTCG pVV‐D CAGTCGATCGTACGCTAGTTAAC pVV16‐Rv3876‐His pVV‐U ACAACTTGAGCCGTCCGTCG pVV‐D CAGTCGATCGTACGCTAGTTAAC Seq‐Rv3876‐1 GCTATCGCACAGACACCG Seq‐Rv3876‐2 CGCGTCAACGCATGCACC p3Flag‐Rv0229c‐IRES‐hrGFP II T3 AATTAACCCTCACTAAAGGG p3Flag‐Rv3876‐IRES‐hrGFP II T7 TAATACGACTCACTATAGGG

AES‐Rv0229c 0229cAm‐Fw GAATCCTCAAGTGCGCCACCG 0229c‐Av‐Rv GGTCGGGTTGCCGCATCGG

AES‐Rv3876 3876Am‐Fw CGGCGTCCAATACTCGAGG 3876‐Av‐Rv GCATCGCCATCCCGATGACGG

CHAPTER 2 MATERIALS AND METHODS 38

III. Ligation T4 DNA ligase joins blunt and cohesive ends by catalyzing the formation of phosphodiester bonds between 5’-phosphate and 3’-hydroxyl terminus in duplex DNA. It was used for cloning the DNA fragments into the plasmids. A molar ratio of fragments to vector (4:1) was used in the ligation reactions. T4 DNA ligation buffer and 5 units T4 DNA ligase were added to the mixture and incubated overnight at 16°C.

IV. Media and culture conditions for bacteria A. E. coli Bacteria (DH5α, stellar, MACH1 or DB3.1) were cultured using standard conditions at 37°C in LB medium. Liquid cultures were incubated on a shaker at 200 rpm. Selection for cells harbouring a resistance gene (ampicillin, kanamycin, hygromycin or streptomycin) on a transformed plasmid was performed with the antibiotic added to the medium at a concentration of 100 μg/ml ampicillin, 100 μg/ml kanamycin, 200 μg/ml hygromycin and 25 μg/ml streptomycin. Bacterial stocks were made by mixing 1.5 ml liquid culture with 300 μl 100% glycerol. The mixture was stock frozen at -80°C.

B. Mtb Mtb (H37Rv strain) was grown at 37 °C in Middlebrook 7H9 medium supplemented with 10% ADC, 0.5% glycerol and 0.05% Tween-80 or on Middlebrook 7H11 agar medium supplemented with 10% OADC. All cultures were performed in a BSL3 laboratory.

V. Preparation of chemically competent E. coli cells Bacteria were grown over night in 1 ml LB medium at 37°C on a shaker at 200 rpm. On the next day, they were transferred into 100 ml LB medium (Dilute the culture 1:100) and grown until the OD590 reached 0.3-0.5, then quickly chilled by immersion of the culture flask in ice and swirl. Cells were pelleted at 3000 rpm at 4°C for 5 minutes, resuspended in 10 ml of 0.1

M ice-cold CaCl2 and hold on ice for 30 min. After a centrifugation step (3000 rpm at 4°C for

5 minutes), the pellet was resuspended in 1 ml of 0.1 M ice-cold CaCl2 + 17% glycerol and aliquots were stored at -80°C.

CHAPTER 2 MATERIALS AND METHODS 39

VI. Transformation of bacteria 100 µl of competent bacteria (E. coli strain DH5α, stellar, MACH1 or DB3.1) was added to 25 µl ligation reaction (plasmid DNA after ligation in a reaction tube) or 2.5 µl of In-Fusion cloning reaction or 1 µl of Gateway® recombination cloning and incubated on ice for 20 to 30 minutes. A heat shock was carried out at 42°C for 45 seconds, followed by 3 minutes on ice. 1 ml of LB medium was added and the bacteria was revived at 37°C for 60 minutes while shaking. 100-250 µl of the bacterial suspension were spread on LB agar plates containing appropriate antibiotic and incubated at 37°C over-night.

VII. Preparation of plasmids from small E. coli cultures (Mini-Prep) Colonies were inoculated in 3 ml LB medium containing the required antibiotic and incubated at 37°C over-night. On the next day, 2 ml were transferred into a 2 ml tube. The cells were centrifuged at 13000 rpm for 3 minutes and the supernatant was removed. The plasmid isolation kits (Qiagen, Hilden, Germany) were used, according to the supplier specifications. 250 μl of P1 solution were added to the pellet and resuspended to homogeneity. Then, 250 μl of P2 solution were added and mixed gently by inverting the tube 4-6 times. 350 μl of P3 solution were added and mixed by inverting the tube 4-6 times. After centrifugation at 13000 rpm for 10 minutes, the supernatant (850 μl) was transferred to the QIAprep spin column. After centrifugation at 13000 rpm for 1 minute, it was washed twice by adding 500 μl BP buffer and 750 μl PE buffer. 50 μl water was added to elute DNA and centrifuged at 13000 rpm for 1 minute. The DNA concentration was determined with NanoDrop ND-1000 spectrophotometer and stored at -20 C.

VIII. Digestion of DNA with restriction enzymes Restriction enzymes and buffers from NEB or Fermentas were used as recommended. 1 to 5 μg of DNA was digested with 0.2-0.5 μl enzyme in appropriate buffer at 37°C for 1 hour.

VIIII. Agarose gel electrophoresis DNA was analyzed using horizontal submarine gel electrophoresis using 0.8-1% agarose gels in 1xTAE buffer. The agarose was dissolved completely by heating the solution in a microwave, cooled down to 60°C and casting it into a tray fitted with a comb. Solidified gel

CHAPTER 2 MATERIALS AND METHODS 40 was placed in an electrophoresis tank filled with 1xTAE buffer and DNA samples were mixed with 6x loading buffer (Fermentas) and loaded into the wells. The electrophoresis was carried out at a constant voltage of 80-100V for separation of DNA-fragments (30 to 45 minutes). DNA was stained with 25 μg/ml ethidium bromide (Sigma-Aldrich) and visualized under UV light. GeneRuler™ 1 kb and 100 bp Ladder (Fermentas) were used as a molecular size marker. The electrophoresed DNA was analysed on UVP Gel Documentation system and photographed.

X. Purification of DNA fragment by gel extraction DNA fragments were purified by running on an ethidium bromide-containing agarose gel, excising the band containing the fragment under UV illumination and subsequent gel extraction using GeneJET Gel Extraction Kit (Fermentas) following the manufacturer's instructions.

2.4 CELL CULTURE AND TRANSFECTION I. Primary cells and cell lines A. Primary cells: In vitro generation of human monocyte-derived macrophages (HMDM) Human peripheral blood (buffy coats) was obtained from healthy blood donors (Etablissement Français du Sang, EFS, Toulouse). Written informed consent to obtain the human blood was received from the donors, under EFS contract no. 21/PVNT/TOU/IPBS01/ 2009-0052. In accordance with articles L1243-4 and R1243-61 of the French Public Health Code, the contract was approved by the French Ministry of Science and Technology (agreement no. AC 2009-921). Peripheral blood mononuclear cells, from these buffy coats, were then immediately isolated by density gradient centrifugation using Ficoll-Paque™ Plus technique (Amersham Bioscience AB, Freiburg, Germany). Monocytes were then purified from freshly isolated PBMC by positive magnetic selection using anti-CD14 microbeads (Miltenyi Biotec). Monocytes were cultured at 4 × 106cells/ml per well in a 24-well plate (600 μl per well), or 1.6 millions of cells in a 6-well plate (1.2 ml per well) in RPMI-1640 GlutaMax™ supplement. After 2 hours at 37°C, attachment of at least 90% of the cells was verified by microscopy. 600 μl of RPMI-1640 GlutaMax™ supplement (or 1.2 ml in 6 wells plate) containing 2X FBS (Final concentration 10%; FBS was previously incubated at 56°C for

CHAPTER 2 MATERIALS AND METHODS 41

30 min in order to inactivate the complement), 2X Sodium pyruvate (Final concentration 1 mM) and 2X β-mercaptoethanol (Final concentration 50 μM) were then added in each well. The medium was then supplemented with 20 ng/ml of macrophage colony-stimulating factor (M-CSF) (PeproTech, Rocky Hill, NJ). Monocytes were incubated at 37°C in the presence of 5% CO2 and M-CSF for 7 days.

B. Murine cell lines: Mouse monocyte-macrophage BALB/c (RAW 264.7) cells RAW 264.7 cells were obtained from the American Type Tissue Culture Collection (ATCC) and maintained in RPMI-1640 GlutaMax™ supplemented with 10% FBS, 50 μM β- mercaptoethanol and 1 mM sodium pyruvate at 37°C in the presence of 5% CO2.

C. Human cell lines: Human embryonic kidney 293 (HEK 293) cells HEK 293 cells were obtained from the ATCC and maintained in RPMI-1640 GlutaMax™ supplemented with 10% FBS, 50 μM β-mercaptoethanol and 1 mM sodium pyruvate at 37°C in the presence of 5% CO2.

D. Cell passaging Cell cultures were split 1:10 in fresh medium every 2-4 days. Adherent cells were washed once with PBS and disaggregated by incubation with cell dissociation solution (1 ml per 25 cm2 culture vessel area) at 37°C for 10-15 min until cells detached. Cell dissociation solution was then inactivated by adding 9 ml medium with 10% FBS.

E. Assessing cell vitality The number of viable and dead cells was determined by trypan blue exclusion. The cell suspension was diluted with trypan blue solution (0.4%) and the cells counted in a Neubauer haemocytometer. The concentration of viable cells was then calculated using the equation: Number of viable cells/ml C= N x D x 104 N average of unstained cells per corner square (1mm containing 16 sub-squares) D dilution factor

CHAPTER 2 MATERIALS AND METHODS 42

F. Freezing and thawing cells Cells were harvested and suspended at 4-6x 106 cells/ml in cold freezing medium containing 10% FBS and 10% DMSO. Aliquots of 1 ml of these cell suspensions were placed on ice and frozen at -20°C for 2 h, then transferred to -80°C for 48 h. For long-term storage, samples were transferred to liquid nitrogen (-196°C). For thawing cells, the cryovial containing the frozen cells was removed from liquid nitrogen storage and immediately placed into a 37°C water bath. 9 ml of warm medium was added dropwise to the partially frozen cells, and the culture was transferred to flasks and incubated at 37°C in the presence of 5% CO2. Growth medium was changed as soon as cells were attached.

II. Transfection of the constructed plasmids into host cells A. Transient transfection of HMDM Neon® Transfection System (Invitrogen) was used for transient expression of various plasmids construct in HMDM. The experiments were performed in accordance with the manufacturer’s instructions, with the following parameters: 1000 V, 40 ms, 2 pulses, and 1 mg of DNA for 2x105 cells, transfected cells were plated on coverslips in 24-well plates and placed in incubator at 37°C in the presence of 5% CO2. After 12 h, culture medium was removed and replaced with fresh medium. Cells were used within 24 h of transfection.

B. Transient transfection of RAW 264.7 cells RAW 264.7 cells transfection with various plasmid constructs was carried out using Lipofectamin® 2000 transfection reagent according to Invitrogen’s instructions. In brief, for transfection in 24-well plates, for each well the following solutions were taken: • Solution A: 1 μg DNA was mixed with 50 μl RPMI-1640 GlutaMax™ supplement. • Solution B: 4 μl Lipofectamin was mixed with 50 μl RPMI-1640 GlutaMax™ supplement. Combined solutions A + B were incubated for 5 min at room temperature, 50 μl transfection mix was applied into 2x105 cells/well seeded on day prior to transfection. After 12 h, transfection mix was removed and replaced with normal growth medium. After 24 h, transient transfected cell were ready for further experiments.

CHAPTER 2 MATERIALS AND METHODS 43

C. Transient transfection of HEK 293 cells Transfection of various plasmids into HEK 293 cells was performed at 90 to 95% confluence on coverslips in 24-well plates using ViaFect™ transfection reagent at a reagent: DNA ratio of 3:1 according to the manufacturers' instructions. The transfected cells were incubated for 24 h in culture medium at 37°C in the presence of 5% CO2.

III. TRANSFECTION EFFICIENCY ANALYSIS A. Fluorescence microscopy and imaging After transfection, the coverslips containing transected cells were washed with 1x PBS, fixed with 4 % formaldehyde in PBS for 20 min at room temperature and washed again with 1x PBS. The coverslips were mounted with Fluoroshield™ mounting medium containing DAPI and observed with a large field Leica DMIRB fluorescence microscope at 63X magnification (Leica HCX Plan Apox63/1.40-Oil). GFP (488 nm/509 nm) was revealed with a GFP filter cube (excitation: BP 470/40; dichroic mirror, 495; emission: LP 525/50) whereas DAPI (358 nm/463 nm) was visualized with a DAPI filter cube (excitation: BP 350–50; dichroic mirror, 400; emission LP 460/50). Images were captured with a CoolSnap HQ2 CCD camera (1 pixel = 6.45 µm), acquired with MetaMorph 7.1.7 software and processed with ImageJ software (National Institutes of Health, Bethesda, MD). All different images were acquired with the same exposure time of 200 ms. Image processing consisted in equivalent adjustments of brightness and contrast on complete images.

B. FACS analysis (Flow cytometry) Cells were harvested and washed two times in 1x PBS. For each FACS reaction 0.5-1x 106 cells were resuspended in 500 μl 1xPBS. Sample acquisition was performed using a LSR-II cytometer (BD Bioscience) and analyzed in FlowJo 8.8.6 software (Ashland, OR).

CHAPTER 3 RESULTS

3.1 Prediction of proteins with putative NLSs in Mtb Over the past decade, a new field of science called computational genomic and bioinformatics has emerged. Its focus is to develop computational tools and databases for organizing and extracting biological meaning from the comparison of large collections of genomes that could be of valuable information to users. Indeed, there are numerous resources for studies of prokaryotic genomes available to the public (Field et al., 2005). Based on some of these tools, I developed a screening for genes of Mtb encoding proteins harboring a putative eukaryotic-like NLS and predicted to be nucleomodulins. In order to look for the presence of NLS sequences, a search was performed in the server of the Comprehensive Microbial Resource (CMR) (http://cmr.jcvi.org/cgibin/CMR/shared/ MakeFrontPages.cgi?page=cmr_search&search_type=motif&crumbs=genomes), scanning the Mtb genome with the consensus sequences for different NLS types (Peterson et al., 2001). Over 300 genes of Mtb H37Rv were predicted to encode proteins harboring a putative NLS in their amino acid sequence. The protein sequences were then analyzed using the Map NLS server (http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi) to calculate the “NLS score” of each candidate protein. I selected medium to high NLS scores (ranging from 1-10). These candidates were ranked on four different levels of NLS scoring: 8- 10, for proteins predicted to be exclusively localized to the nucleus; 7-8, for partial localization to the nucleus; 3-5, for localization to both the nucleus and the cytoplasm; and 1-2, for localization predicted to be cytoplasmic (Kosugi et al., 2009b). The next step of the screening, based on the assumption that effector proteins produced in the bacteria must be translocated into infected host cells, was to focus only on those proteins either demonstrated (Albrethsen et al., 2013) or supposed to be secreted. For this, the secretion property of the candidate proteins was verified using the Tuberculist server (http://tuberculist.epfl.ch/) along with a careful review of the existing literature (Albrethsen et al., 2013; Gey van Pittius et al., 2006). In mycobacteria, several protein secretion systems have been described (Bitter et al., 2009; Champion and Cox, 2007). In particular, the type VII secretion pathway (T7SS), also known as the ESX pathway, is able to secrete a series of specific proteins across the complex

Table 12. Predicted nuclear targeting proteins with NLS sequences in Mtb H37Rv.

No. Accession Predicted NLS Targeting mechanism NLS score ESAT‐6 (esx) gene Identified Number sequences cluster regions in

1 Rv0229c RRRR Monopartite class 1 4‐6 ‐ CF* 2 Rv0284 KKMR Monopartite class 2 3‐3.5 3 MF, WCL 3 Rv0286 RRRR Monopartite class 1 5 3 ‐ 4 Rv1798 KKTK Monopartite class 2 3‐5 5 WCL 5 Rv3866 RRDIGGVMVR Bipartite 3 1 WCL FVVCRRGDR 6 Rv3876 RRRK Monopartite class 1 4‐10 1 MF 7 Rv3881c KRFR Monopartite class 2 3‐3.5 1 CF 8 Rv3882c RRRR Monopartite class 1 3 1 WCL 9 Rv3885c RRRK Monopartite class 1 3‐3.7 2 WCL 10 Rv3888c KKSR Monopartite class 2 3‐4 2 MF, WCL 11 Rv3894c KKKK Monopartite class 1 5 2 WCL

* Starvation condition NLS score: 3‐5 localized to the nucleus and the cytoplasm, 6‐7 partially localized to the nucleus, 8‐10 exclusively localized to the nucleus Abbreviations: CF (Culture filtrates), WCL (Whole cell lysates), MF (Cell membrane fraction)

Figure 11. Genes associated with the Mtb ESAT‐6 (esx) gene clusters. Ten gene candidates were members of the genetic loci encoding the ESX‐1 to ESX‐5 type VII secretion systems (Modified from Gey van Pittius et al., 2006).

CHAPTER 3 RESULTS 45 envelope of mycobacteria (see (Houben et al., 2014), for a review). Mtb carries up to five of these secretion systems, namely ESX-1 to ESX-5. These secretion machineries are encoded by genes or operons clustered in five different regions of the genome, each encoding one of the five ESX systems (Fig. 11). At least three of these systems are involved in Mtb virulence and/or viability in the infected hosts. The first ESX system to be discovered, ESX-1, is known to secrete the early secreted antigenic target protein 6 (ESAT-6; also named EsxA) along with the 10 kDa culture filtrate protein (CFP-10, also named EsxB) and a few other proteins, so-called Esp. Notably, part of the genomic region encoding ESX-1 is deleted in the attenuated strain of the Mycobacterium bovis bacille Calmette-Guérin (BCG) vaccine (Brodin et al., 2006), creating what is named the region of difference 1 (RD1) between Mtb and BCG. The other ESX systems are thought to secrete specific sets of proteins. For instance, each system seems to contain a pair of genes encoding two small proteins analogous to EsxAB. In addition, it has been shown that the ESX-5 system is able to export proteins of the so-called PE and PPE families (Daleke et al., 2011). Among the ∼300 candidates conserved after the first part of our screen, we noticed that ten genes were members of the genetic loci encoding the ESX-1 to ESX-5 type VII secretion systems (i.e. Rv0284, Rv0286, Rv1798, Rv3866, Rv3876, Rv3881c, Rv3882c, Rv3885c, Rv3888c and Rv3894c; see Figure 11). We thus kept these candidates for further analysis. In addition, we used data of published proteomic analysis that investigated the comparative exported proteome during exponential growth or starvation conditions (Albrethsen et al., 2013; de Souza et al., 2011) to complete our screen. The products of nine of the ten conserved candidate genes belonging to ESX systems were identified in this study. Only one was detected in the culture filtrates (CF), whereas eight others were seen in whole cell lysates (WCL) or in cell membrane fraction (MF), and another one (Rv0286) not detected at all. Nevertheless, we decided to conserve all of them for the rest of the screening. This decision was based on the fact that well characterized secreted proteins, such as EsxB or EsxA, can be detected not only in culture filtrates, but also in the membrane fraction. In addition, it is also possible to miss some truly secreted proteins if they are produced and/or exported in low amounts. Thus, this type of analysis cannot be totally exhaustive. Finally, one additional candidate, Rv0229c, was also kept because it was detected in the culture filtrate, during the starvation conditions (Albrethsen et al., 2013; de Souza et al.,

Figure 12. Nuclear localization of mycobacterial nucleomodulin candidates fused with GFP in eukaryotic cells. (A) Fluorescence images of HEK 293 cells transiently expressing EGFP‐tagged Mtb protein candidates, (B) Bar plots show quantifications of the percentages of cells exhibiting predominantly nuclear or cytoplasmic protein localization from transient transfection of HEK 293 cells. All experiments were repeated independently twice, with triplicate for each experiment. At least 200 total cells were counted for each protein candidate. (C) Human monocyte‐derived macrophages were transiently transfected with protein candidates fused with GFP and GFP‐alone vector (pEGFP‐C1) as a control. The cells were fixed in 4% paraformaldehyde 24 h post transfection and mounted in Fluoroshield mounting medium containing DAPI, nuclei were stained with DAPI showing blue fluorescent stains. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. CHAPTER 3 RESULTS 46

2011), bringing the total number of candidates to eleven. Table 12 summarizes the properties of the eleven candidates conserved after our screen.

3.2 Screening of nuclear targeting proteins of Mtb in eukaryotic cells To demonstrate that the NLS-containing protein candidates actually translocate into the host cell nucleus, I first tried to clone the eleven Mtb gene candidates into a eukaryotic expression vector, in fusion with EGFP, a variant of the Green Fluorescent Protein reporter. This was done in order to visualize easily the sub-cellular location of the hybrid proteins. The first step for this cloning was the amplification by PCR from the Mtb H37Rv genome, using oligonucleotides containing appropriate recombining sequences on both sides of the amplified ORF. While seven gene candidates (Rv0229c, Rv0286, Rv1798, RV3866, Rv3876, Rv3882c and Rv3888c) were successfully amplified by PCR, the remaining four gene candidates could not be obtained in spite of repeated attempts with various conditions and thermostable polymerases. I then cloned the remaining seven wild-type gene candidates into pEGFP-C1, a vector expressing EGFP under the control of a CMV strong and constitutive promoter. The cloning was performed using the In-Fusion recombination system and generated a hybrid protein with the EGFP at the N-terminal end, and the product of the cloned ORF at the C-terminus. The seven recombinant plasmids were constructed and verified by sequencing. The hybrid proteins could then be over-expressed after transient transfection into HEK293 cells. After 24 hours of transfection, the GFP-tagged protein candidates were examined under a large field Leica DMIRB fluorescence microscope at 63X magnification. Interestingly, two of the GFP-tagged protein candidates showed a robust nuclear localization: pEGFP-Rv0229c and pEGFP-Rv3876. They showed a nuclear localization in 100 % of the transfected cells (GFP positive). A third candidate, pEGFP-Rv1798, also exhibited an apparent nuclear localization, but only in a fraction (85%) of the transfected cells (Fig. 12A-B). By contrast, the remaining four GFP-tagged proteins were localized only in the cytoplasm, indicating that the putative NLS identified in silico were not functional in HEK cells (Fig. 12A-B). Of note, the green fluorescence was observed for each of these four GFP- tagged candidates, thus demonstrating that the hybrid proteins were expressed. The three GFP-tagged proteins that yielded nuclear fluorescence in HEK cells were then expressed in leukocytes, such as human monocyte-derived macrophages (HMDM) and the RAW 264.7 mouse macrophage cell line. Again, pEGFP-Rv0229c and pEGFP-Rv3876 localized

Figure 13. Nuclear localization of mycobacterial nucleomodulin candidates fused with GFP in RAW 264.7 macrophages. (A) Fluorescence microscopy was used to analyze RAW 264.7 macrophages transiently transfected with the indicated GFP‐ tagged proteins, Rv0229c, Rv3876 and Rv1798. Cells were transfected with GFP‐alone (vector pEGFP‐C1) as a control. GFP‐ tagged protein MMA‐2680 was used as a negative control. Approximately 24 h after transfection, cells were fixed with 4% paraformaldehyde for 20 min and mounted in Fluoroshield mounting medium containing DAPI (blue), which is a counter‐ stain for DNA. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. (B) Bar plots of the percentages of cells exhibiting predominantly nuclear or predominantly cytoplasmic fluorescence after transient transfection of RAW 264.7 macrophages. All experiments were repeated twice independently, with triplicate for each experiment. At least 100 total cells were counted for each protein candidate.

CHAPTER 3 RESULTS 47 into the nucleus of all the transfected cells (Fig. 12C and Fig. 13). By contrast, pEGFP-Rv1798 failed to show a nuclear localization and the fluorescence was found distributed predominantly in the cytoplasm in leukocytes (Fig. 12C and Fig. 13). As controls, we used the cloning vector pEGFP-C1 alone, expressing intact GFP or the recombinant plasmid pEGFP- MMA-2680, carrying MMA-2680, the homologous of Rv1798 present in the genome of Mycobacterium marinum, fused to the GFP gene. Neither intact GFP nor the product of MMA-2680 carry a predicted NLS sequence and, as expected, the fluorescence produced from the two plasmids resides entirely into the cytoplasm (Fig. 12A-B and Fig 13). In conclusion, we could identify two genes of Mtb, Rv0229c and Rv3876, which encode proteins with functional NLS that can drive their transport into the nucleus of eukaryotic cells. We note that both EGFP-Rv0229c and EGFP-Rv3876 exhibit a sub-nuclear localization in HEK cells and in HMDM, showing green spots inside the nucleus. However, we do not know the nature of the sub-domains or the functional significance of this localization and further analysis will have to address this point.

3.3 Rv0229c encodes a protein of unknown function with a putative RNase signature The gene Rv0229c, extending from positions 274306 to 274986 on the chromosome of Mtb H37Rv, encodes a hypothetical protein of 226 amino acids with a calculated molecular mass of 25183.8 Da, a very basic isoelectric point (estimated at 11.7853), and of unknown (so far) function (Tuberculist database). Analysis of this protein in the Conserved Domain Database (CDD), at NCBI (http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml), reveals a putative domain of the PIN family encompassing the second half of the protein. The PIN domain belongs to a large superfamily of nucleases, found in eubacteria, archaea and eukaryotes. Two main divisions are described in this superfamily. The first division contains 5’ nucleases involved in DNA replication, repair and recombination. The second division, the VapC-Smg6 family, contains eukaryotic proteins involved in mRNA degradation or pre-rRNA processing, and also a large number of bacterial or archaeal ribonuclease toxins, members of toxin-antitoxin systems and generally involved in regulation of translation (see for instance (Sharp et al., 2012)). Therefore, although it remains to be experimentally demonstrated, the product of Rv0229c may be a ribonuclease.

Figure 14. Comparative Genomics of Rv0229c. (A) and (B) Comparison of the homologous genomic region encompassing Rv0229c between closely related Mycobacterium species. (C) Evolutionary scenario explaining the organization of the genomic regions (adapted from Veyrier et al., 2011).

CHAPTER 3 RESULTS 48

Another interesting feature of the Rv0229c protein is the presence of a hydrophobic segment at its N-terminal end, which could be a transmembrane segment (TMHMM prediction). This hydrophobic segment is not predicted to be a cleavable signal sequence by the SignalP program (http://www.cbs.dtu.dk/services/SignalP/).

3.4 Rv3876 (EspI) encodes a putative ATPase, involved in the regulation of the ESX-1 secretion system The gene Rv3876, extending from positions 4353010 to 4355010 on the chromosome of Mtb H37Rv, encodes a protein of 666 amino acids, with calculated molecular mass of 70645.1 Da and isoelectric point of 10.2754. This gene, also named espI, belongs to a putative operon composed of Rv3876 (espI) and Rv3877 (eccD1), located nearby the esxBA operon within the esx-1 locus (Fig. 14). Conserved in species of the Mtb complex (MTBC) and in related pathogenic species such as M. leprae and M. marinum, the ESX-1 secretion system allows the secretion of a number of so-called Esp proteins and is necessary for virulence of Mtb (Bitter et al., 2009; Brodin et al., 2006; Simeone et al., 2009). The product of Rv3876 is a protein very rich in proline and alanine, and harbors a putative ATPase domain of the FlhG family (Das et al., 2011), leading to the hypothesis that it could exert a regulatory role on the ESX-1 secretion system. This was confirmed recently by Zhang et al. (2014) who reported that Rv3876/EspI is not required for ESX-1-mediated secretion and virulence upon macrophage infection, but that it is able to negatively regulate the ESX-1 system in response to low cellular ATP.

3.5 Rv0229c was recently acquired by the ancestor of Mtb through horizontal genetic transfer as suggested by Comparative Genomics Comparative genomics of the region harboring Rv0229c between closely related Mycobacterium species demonstrate a very good synteny in this region. However, homologues of Rv0229c are present only in species of the MTBC, such as Mtb or M. bovis BCG, but they are absent from the genome of M. kansassii, M. avium or M. marinum (Fig. 14). In addition, whereas no gene is present between the homologues of Rv0228 and Rv0230 in the genome of M. kansassii or M. avium (Fig. 14A), a genomic island carrying several unrelated genes is present in place of a homologue of Rv0229c in the genome of M.

Figure 15. Rv0229c contains a functional nuclear localization signal. (A) The amino acid sequence of Rv0229c and predicted NLS sites (in red) are indicated. (B) The table shows the constructed plasmid sequences of the wild type and mutant derivatives of Rv0229c. (C) RAW 264.7 macrophages were transiently transfected with GFP‐tagged wild‐type Rv0229c (pEGFP‐Rv0229cWT), GFP‐tagged NLS mutants Rv0229c (pEGFP‐Rv0229cM1, pEGFP‐Rv0229cM2 and pEGFP‐ Rv0229cD) constructs, or vector alone (pEGFP‐C1) as a control. (D) HEK 293 cells were transiently transfected with GFP‐ tagged wild‐type Rv0229c and mutant derivatives. The cells were fixed in 4% paraformaldehyde 24 h post transfection and mounted in Fluoroshield mounting medium containing DAPI (blue) stained nuclei. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X

CHAPTER 3 RESULTS 49 marinum (Fig. 14B). Two mutually exclusive evolutionary pathways can be proposed to explain the present state of these genomic regions. In the first scenario, the gene Rv0229c was present in the ancestor of Mtb and has been lost during the course of evolution by the branches that yielded M. avium and M. kansassii. However, in light of a probable evolutionary pathway of the ancestors of Mtb (Veyrier et al., 2011), this would imply that the deletion of the homologue of Rv0229c occurred several times and independently in branches that led to different species. The simplest scenario that could explain the genomic organizations observed in the present species would be if the gene Rv0229c was absent from the common ancestor of Mtb and other related species such as M. avium and M. kansassii, and had been acquired by horizontal gene transfer in the branch that yielded Mtb and M. bovis BCG. In the meantime, a different genetic island had been acquired at the same locus in the branch that yielded M. marinum (Fig. 14C). In Mtb, the two ORFs of Rv0230c and Rv0229c overlap: the UGA stop codon of Rv0330c contains part of the GUG start of the Rv0229c translation start. Therefore, it is very likely that the insertion of the horizontally acquired Rv0229c gene created an operon in which Rv0229c is co-transcribed with Rv0230c, the common promoter being upstream from the Rv0230c gene. This promoter remains to be identified and a recent global RNAseq analysis did not detect a transcription start upstream from Rv0229c or Rv0230c (Cortes et al., 2013), suggesting that these genes might be transcribed at low level in standard growth medium.

3.6 Identification of the functional nuclear localization signals of Rv0229c and Rv3876 To confirm that the predicted NLS sequences contained in Rv0229c and Rv3876 are in fact responsible for their nuclear localization in eukaryotic cells, I mutated these sequences, by directed mutagenesis of the pEGFP-Rv0229c and pEGFP-Rv3876 plasmids, replacing by alanines the charged amino acid residues present in the predicted NLS (Fig. 15 and 16). In the case of Rv0229c, fusion of the complete gene to the 3’ end of the gene encoding EGFP generated a GFP-tagged protein with a predicted size of ∼33 kDa. This is smaller than most predicted actively transported proteins (∼40 kDa), suggesting the possibility that this protein may passively diffuse into the nucleus of eukaryotic cells. However, the sequence of Rv0229c product contains at least two motifs composed of basic amino acids that exhibit

. Figure 16. Rv3876 contains a functional nuclear localization signal. (A) The amino acid sequence of Rv3876 and NLS sites are indicated. (B) The table shows the constructed plasmid sequences of the wild type and mutant derivatives of Rv3876. (C) RAW 264.7 macrophages were transiently transfected with GFP‐tagged wild‐type Rv3876 (pEGFP‐Rv3876WT), GFP‐ tagged NLS mutant Rv3876 (pEGFP‐Rv3876M1 and pEGFP‐Rv3876M2) constructs, or vector alone (pEGFP‐C1) as a control. (D) HEK 293 cells were transiently transfected with GFP‐tagged wild‐type Rv3876 and mutant derivatives. The cells were fixed in 4% paraformaldehyde 24 h post transfection and mounted in Fluoroshield mounting medium containing DAPI. Nuclei (blue) were counterstained by DAPI. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X.

CHAPTER 3 RESULTS 50 significant score when analyzed with various NLS prediction programs, and may therefore be involved in active transport into the nucleus (Fig. 15A). To address this question, I constructed three variants of this protein candidate (Fig. 15B). In the first mutant, three positively charged amino acids (71RRRR74) were changed into neutral alanine residues (71AAAR74), generating the plasmid called thereafter pEGFP-Rv0229cM1. In the second mutant, four positively charged amino acids (71RRHRR74) were changed into alanine (71AAHAA74), generating pEGFP-Rv0229cM2. Finally, a third mutant was constructed in which both motifs were mutated, yielding plasmid pEGFP-Rv0229cD. These mutant constructs were then transfected into both HEK 293 cells and RAW 264.7 macrophages. The fluorescent microscopy images clearly showed that modifying the two basic motifs in pEGFP-Rv0229cD prevented nuclear transport. Indeed, with the EGFP-Rv0229cD hybrid protein the fluorescence was found to be present predominantly in the cytoplasmic region in both cell types, ruling out the possibility of a passive diffusion of a small enough hybrid protein into the nucleus. By contrast, the hybrid proteins expressed from pEGFP-Rv0229cM1 and pEGFP- Rv0229cM2 displayed almost exclusive nucleus localization in RAW 264.7 macrophages and in HEK cells (Fig. 15C-D). Therefore, it appears that both motifs can be used as NLS, and each of them is sufficient to allow the transport of the hybrid proteins to the nucleus. We note that in macrophages, M1 modification perturbed the sub-nuclear localization of the fluorescence, suggesting that in addition to being involved in nuclear import, the 71RRRR74 motif could be involved in this sub-nuclear localization. However, a careful and quantitative analysis of this phenomenon still needs to be performed to confirm this observation. In the case of Rv3876, fusion of the complete Rv3876 gene to the 3’ end of the gene encoding EGFP, yielding plasmid pEGFP-Rv3876, generated a hybrid protein with a predicted size of ∼78 kDa, and hence too big to passively diffuse into the mammalian cell nucleus. Similar to Rv0229c, two motifs containing basic amino acids were predicted as putative NLS (Fig. 16A), and both were modified independently by directed mutagenesis of plasmid pEGFP-Rv3876. In the first one, called thereafter pEGFP-Rv3876M1, the four positively charged amino acids 334RRRK337 were changed into 334AAAA337. In the second one, pEGFP- Rv3876M2, the three positively charged amino acids 360KKVK363 were changed into 360AAVA363 (Fig. 16B). Interestingly, when I analyzed the subcellular localization after transient transfection of HEK 293 cells and RAW 264.7 macrophages, pEGFP-Rv3876M1 was excluded from the nucleus, whereas pEGFP-Rv3876M2 was found present predominantly in the nucleus,

Figure 17. Rv0229c and Rv3876 are mycobacterial secretory proteins. Rv0229c (A) and Rv3876 (B‐C) were fused with 6‐His‐ tag and over‐expressed in wild type H37Rv Mtb (A‐B) or MtbΔRD1 (C) after cloning in the pVV16 vector. Different clones (A‐ D) were tested. Culture filtrates were obtained as supernatants of bacterial cultures during early‐log phase and mid‐log phase, recovered by centrifugation, and concentrated using filters with a 3‐kD cutoff. Total lysates were obtained by shaking bacterial pellets with 106‐µm acid‐washed glass beads and centrifugation at 2300 x g for 30 min. Cell wall, cytosolic membrane fractions were prepared from total lysates by 45‐min centrifugation at 17,500 x g, followed by ultracentrifugation at 350,000 x g for 90 min. Proteins were quantified, and analyzed by Western blotting using monoclonal anti‐ His‐tagged antibodies. Mouse anti‐GroEL2 antibodies were used for lysis controls.

CHAPTER 3 RESULTS 51 exhibiting a sub-nuclear localization identical to that of the EGFP-Rv3876 wild-type protein (Fig. 16C and D). Therefore, the motif 334AAAA337 is essential for the nuclear transport of Rv3876, and is a major determinant of its NLS.

3.7 Rv0229c and Rv3876 are secreted proteins in Mtb Delivery of bacterial effector protein to the cytosolic compartments of host cells is a prerequisite for nuclear targeting of these proteins, thus it was necessary to verify that nucleomodulin candidates are indeed secreted. The protein product of Rv0229c has been found in culture filtrate during starvation conditions (Albrethsen et al., 2013). In an independent proteomic study, Rv0229c has been identified both in membrane fraction and in culture filtrate, with its first 30 amino-acids forming a putative (but not demonstrated) cleavable signal sequence (de Souza et al., 2011). Therefore, it is very likely that this protein can be secreted by Mtb. By contrast, although Rv3876 protein is encoded by a gene of the ESX-1 region (Bottai et al., 2011; Brodin et al., 2006), Rv3876 protein has been found in membrane fraction and whole cell lysate, but not in the culture filtrate (De Souza et al., 2011), and not identified in the culture filtrate (Albrethsen et al., 2013). However, the protein profile of the Mtb culture filtrate is highly dependent on the cultivation time (Andersen et al., 1991), and we next wanted to perform experimental analysis to determine whether Mtb secretes Rv0229c and Rv3876. To this end, I fused both proteins with 6-His-tag and over-expressed them in Mtb using the pVV16 expression vector. pVV16-Rv0229c and pVV16-Rv3876 were constructed and verified by DNA sequencing. Our collaborator Dr. Daria Bottai, at Pisa University (Italy), introduced both plasmids into wild-type H37Rv and analyzed the culture filtrate and whole cell extract of the resulting Mtb strains grown to early-log or mid-log phase in Sauton medium by Western blot analysis using antibodies against the 6-Histidine tag. In agreement with previously published data (de Souza et al., 2011), the results confirmed the presence of Rv0229c in the culture filtrate during mid-log phase of bacterial growth, whereas the protein was not detected in the growth medium during early-log phase (Fig. 17A). Most interestingly, these experiments demonstrated that the Rv3876 protein is also present in the medium, both during early and mid-log phase (Fig. 17B). The plasmid pVV16-Rv3876 was also introduced into a strain of Mtb carrying a deletion of the Region of Difference 1 (MtbΔRD1), and thus deleted of the ESX-1 secretion system (Bottai et al., 2011). In the Mtb RD1

CHAPTER 3 RESULTS 52 knockout strain (MtbΔRD1) the secretion of His6-Rv3876 was significantly reduced (Fig. 17C), although not completely abolished. Lysis controls using antibodies against GroEL2 indicated that the samples were not contaminated with cytosolic or membrane-associated proteins (Fig. 17). Altogether, this data indicates that Mtb has the capacity to secrete both Rv0229c and Rv3876, further arguing for their role as potential nucleomodulins.

3.8 Construction of genetic tools for functional characterization of Rv0229c and Rv3876 during infection by Mtb A. Deletion of Rv0229c and Rv3876 Neither Rv0229c nor Rv3876 are essential in Mtb, since both genes they can be disrupted by transposon or antibiotic resistance cassette insertions (Griffin et al., 2011; Zhang et al., 2014). Therefore, we decided to construct mutants of these genes in H37Rv, by recombineering of a zeocin resistance cassette replacing most of the ORF of each gene (Jamet et al., 2015). I constructed allelic exchange segments (AES) by amplification of ∼500 bp upstream and downstream DNA fragments surrounding a ∼700 bp Zeocin resistance cassette. The two AES were used to transform an H37Rv strain derivative carrying pJV53H. pJV53H is derived from pJV53, expressing the recombinase of bacteriophage Che9c and that conferring, resistance to kanamycin (van Kessel and Hatfull, 2007), by exchange of the antibiotic gene, leading to, a plasmid conferring resistance to hygromycin (a gift from W. Malaga). The transformation mixtures were plated on 7H9/ADC/tween medium supplemented with Zeocin (25 µg/ml) and incubated 3 weeks at 37°C. ZeocinR clones were obtained, verified by PCR amplification to carry Rv0229c::ZeoR or Rv3876::ZeoR insertion replacing the wild-type sequences, and cured from pJV53H by successive rounds of growth without selection. ZeoR HygS clones were finally obtained and verified again for the presence of the gene disruption by PCR. In the case of Rv0229c, as this gene stands as the second and last one of a putative operon with Rv0230c (see 3.7), the strain carrying the Rv0229c::ZeoR replacement is very likely to be a single mutant. The construction of a complemented strain is still in progress. By contrast, the strain carrying an Rv3876::ZeoR insertion is deficient not only for Rv3876, but also for the gene Rv3877; these genes are known to be co-transcribed (Zhang et al., 2014). Therefore, we decided to construct and introduce in the deletion strain plasmids expressing

Figure 18. Over‐expression of Rv0229cv (A) or Rv3876 (B) has no effect on multiplication of Mtb inside human macrophages. Human macrophages were infected with MOI of 0.1, and incubated 7 days. Samples were harvested at the indicated time points, lyzed with triton X100 and the colony forming units (CFU) were counted after plating serial dilutions on 7H11 OADC tween agar. For Rv3876 (B), anhydro‐tetracycline (200 µg/ml) was added to the culture of bacterial cells one day before infection, and was maintained in the cell culture medium during the infection.

CHAPTER 3 RESULTS 53 either Rv3876, Rv3877 or the operon Rv3876-Rv3877, under the control of the promoter present upstream from Rv3876 (Zhang et al., 2014). These strains are still under construction. When obtained, these deleted and complemented strains will be used to infect both macrophages in culture and mice, in order to test the role of Rv0229c or Rv3876 during different stages of infection.

B. Over-expression of Rv0229c and Rv3876 Plasmids allowing over-expression of Rv0229c and Rv3876 have first been constructed by cloning into the shuttle expression vector pVV16 (BEI catalog # NR-13402). In this plasmid, the cloned genes are expressed under the control of the strong and constitutive promoter of the Mtb GroEL2 gene. We noticed that Rv0229c cloned in this vector did not affect the growth of wild-type H37Rv. Then, H37Rv/pVV16-Rv0229c was used to infect human macrophages (experiment performed in the group by A. Grabowska). The results of this experiment showed no significant difference between the control strain H37Rv/pVV16 as compared to the over-expressing strain H37Rv/pVV16-Rv0229c (Fig. 18). Similar experiments could not be performed with Rv3876 because H37Rv/pVV16-Rv3876 transformants yielded very small colonies on 7H9 OADC tween + kanamycin plates, demonstrating a partial toxicity that might hamper the multiplication of the strain into infected macrophages. We then decided to construct inducible expression vectors, using integrative plasmids of the Gateway family, expressing the gene of interest under the control of a Tet-inducible promoter developed by the groups of S. Ehrt and D. Schnappinger, at Weill Cornell Medical College, New York (Ehrt et al., 2005). These plasmids are integrative vectors that are present in single copy into the chromosome of Mtb, at the bacteriophage L5 integration site (glyV). I constructed such plasmids (named pGMC-TetR-P1-Rv0229c and pGMC-TetR-P1-Rv3876) that express Rv0229c or Rv3876 in a Tet-inducible manner, and these plasmids were introduced by transformation into wild-type H37Rv. The resulting strains were used to infect human macrophages, and the CFU were followed in the absence or presence of Anhydro- tertacycline (200 µg/ml), to induce expression of the protein of interest. Again, no significant difference was seen when comparing the induced or uninduced conditions, indicating that either none of the tested protein affects multiplication of Mtb into macrophages, or that the basal level of expression of both protein is sufficient to fully exert its function. The inducible plasmids and the control empty vector will be introduced into the deleted strains (see 3.8-A),

Figure 19. Optimization of the GFP‐reporter plasmid transfection. (A) Flow cytometric assessment of the GFP detection after transfection with pIRES‐hrGFP II, p3Flag‐Rv0229c‐IRES‐hrGFP II, or p3Flag‐Rv3876‐IRES‐hrGFP II in HEK cells. (B) Comparison of transfection efficiency of pIRES‐hrGFP II by varying the concentration of Lipofectamine® 2000 reagent in RAW 264.7 macrophages. (C) HEK 293 cells were transiently transfected with plasmid alone pIRES‐hrGFP II and its derivatives. (D) RAW 264.7 macrophages were transiently transfected with pIRES‐hrGFP II and its derivatives. After transfection, the cells were fixed in 4% paraformaldehyde 24 h post transfection and mounted in Fluoroshield mounting medium containing DAPI (blue) stained nuclei. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X.

CHAPTER 3 RESULTS 54 and the infection experiments will be done again, allowing us to compare over-expression of Rv0229c or Rv3876 to their absence.

C. Expression of Flag tagged versions of Rv0229c and Rv3876 In order to investigate the targets of Rv0229c or Rv3876 in the nucleus of eukaryotic cells, we decided to express FLAGTM-tagged versions of Rv0229c and Rv3876 that could be amenable to experiments of co-immunoprecipitation. Indeed, the FLAGTM tag (DYKDDDDK) is specifically recognized by antibodies and well-established protocols are available to help identifying the partners of proteins of interest (Einhauer and Jungbauer, 2001). I cloned both Rv0229c and Rv3876 into the expression vector pIRES-hrGFP II (Agilent Technologies), which can express under the control of a strong CMV promoter the proteins of interest carrying 3 tandem copies of the FLAGTM tag at the N-terminal end. The presence of an internal ribosome entry sequence (IRES) and a humanized copy of the GFP gene also allow the production of GFP from the same transcript. The two plasmids p3Flag-Rv0229c-IRES-hrGFP II and p3Flag-Rv3876-IRES-hrGFP II were constructed and verified by DNA sequencing.

3.9 Transcriptome analysis In order to understand whether Rv0229c and Rv3876 are able to modulate the host cell gene expression, we attempted to perform a cDNA microarray technology analysis. For this end, I tried to optimize the transfection of RAW 264.7 cells (as previously described) with empty plasmid (pIRES-hrGFP II), p3Flag-Rv0229c-IRES-hrGFP II, or p3Flag-Rv3876-IRES-hrGFP II (see section 3.8-C). First, we tested the expression vectors in 293 HEK cells using the ViaFect™ transfection reagent, and notice that the transfection was quite optimal as detected by a flow cytometry approach (Fig. 19 A). However, when I tested these conditions on RAW 264.7 cells, the vectors were not transfected at all (data not shown). For this reason, I used a different transfection reagent called Lipofectamine® 2000 reagent. At best, this transfection reagent yielded poor transfection efficiency (about 3.5% of positive cells) of the vectors in RAW 264.7 cells (Fig. 19 B). As a control, we used a fluorescent microscopy-based approach to confirm that the expression vectors did work in transfected HEK 293 and RAW 264.7 cells (Fig. 19 C-D). Altogether, in order to perform the transcriptome analysis, we first need to optimize the transient transfection conditions in RAW 264.7 cells to a level that yields at

CHAPTER 3 RESULTS 55 least 10%, and then isolate and purify the GFP-positive cells by a FACS-sorting approach. To accomplish this, either we can test additional transfection reagents, or change the macrophage cell line for another type of immune cell-line.

CHAPTER 4 DISCUSSION

Mtb, the etiological agent of TB in humans, is a facultative intracellular pathogen that can parasitize host cells. A better understanding of Mtb physiology and virulence and of host defense mechanisms against Mtb may help develop novel intervention strategies, including new drugs and host-targeted therapies. Our laboratory has a keen interest in the characterization of Mtb virulence factors required for intracellular parasitism, and of host cell pathways involved in the control of mycobacterial infection. More specifically, we became interested in investigating whether the bacillus might target the nucleus, a cell compartment long considered to be mostly “safe” from direct bacterial attacks. Indeed, nucleomodulins, proteins secreted by different bacterial pathogens that are imported into the nucleus of infected cells to subvert their gene expression, have been characterized in a growing number of bacterial pathogens, such as for example, Listeria monocytogenes, Clostridium perfringens, Streptococcus pneumoniae, Helicobacter pylori and Anaplasma phagocytophilum, among others (Bierne and Cossart, 2012; Bierne et al., 2012). To date, a quick search of the literature using Pubmed with the keywords “Tuberculosis AND nucleomodulin” yields one published article dating October 2015. In this recent study, the authors compare the chromatin extracted from macrophages infected, or not, with Mtb and report the isolation and identification of a putative nucleomodulin that i) is secreted by Mtb, ii) translocates into the host cell nucleus, iii) binds chromatin, and iv) causes hyperacetylation of histone H3 to upregulate gene expression (Jose et al., 2015). To our knowledge, this is the very first study describing a bonafide mycobacterial nucleomodulin, and no literature was available on this subject when my thesis was started. For these reasons, my thesis work initiated a new project in our laboratory aiming at the identification and characterization of mycobacterial nucleomodulins that might enter the nucleus of the macrophage and modulate its response to favor pathogen colonization and/or persistence. The results obtained during this thesis identified two nuclear localization signals (NLS)- containing Mtb proteins, Rv0229c and Rv3876, that are secreted by Mtb and imported into the nucleus in an NLS-dependent manner when expressed in macrophages. In addition, this study set up the basis for the exploration of the biological function of these proteins, and their role in mycobacterial virulence by constructing the tools and improving the

CHAPTER 4 DISCUSSION 57 experimental protocols necessary to address these questions by a combination of genetics, transcriptomics/epigenomics or cell biology approaches. I performed an in silico analysis to screen for Mtb effector proteins candidates with the potential to be nucleomodulins, because we reasoned that such proteins might carry recognizable NLS-like sequences. As a first step, we note, on the one hand, that our approach could miss some true nucleomodulins. For instance, the recently identified histone acetyltransferase encoded by Rv3423.1 (Jose et al., 2015) was not incorporated in our genomic survey since this gene, which encodes a very small protein (76 aa) and partly overlaps the gene Rv3423 (in anti-sense), was not annotated in Tuberculist. In addition, some proteins may carry atypical NLS-like sequences and would not be predicted. On the other hand, due to the degenerated nature of eukaryotic NLS, this first step yielded many candidates (∼300) and most of them are probably false positive. Then, focusing on proteins with high-enough NLS scores AND putatively secreted, another expected feature of nucleomodulins, drastically reduced the number of candidates to only 11. Here again, we note that available results in the literature and predictions for the secretory status of a given protein must be taken cautiously. For instance, although the product of Rv3876 was identified only in the membrane fraction and whole cell lysate and not in the culture filtrate by proteomic analysis (de Souza et al., 2011), western blot experiments unambiguously demonstrated that this protein can be secreted by Mtb in the growth medium. In contrast, our data were in perfect agreement with those of the literature, demonstrating the secretion of the Rv0229c product, occurring in late exponential phase (Albrethsen et al., 2013; de Souza et al., 2011) (Fig. 17). The presence of a recognizable putative NLS does not necessary mean that this sequence is functional. Indeed, this motif must be exposed on the surface of a protein to be accessible and used for translocation of the protein to the nucleus (see above, 1.3-I). Therefore, all proteins with predicted NLS needed to be experimentally verified. We then cloned the candidate genes into the mammalian/prokaryotic shuttle-expression vector pEGFP-C1 to generate translational fusions with the GFP reporter, and expressed the hybrid proteins in eukaryotic cells by transient transfection of HEK 293 epithelial cell lines or macrophages, such as HMDM and RAW 264.7 cells. The subcellular localization of GFP fusion recombinant proteins was determined by fixing the cells and observation under a fluorescence microscope (Fig. 12 and Fig. 13). This approach has been used previously, for instance to

CHAPTER 4 DISCUSSION 58

demonstrate that the YopMC-ter NLS of Yersinia species is crucial for the nuclear targeting of an EGFP-YopM fusion protein (Benabdillah et al., 2004). Quite expectedly, the majority of the seven candidates conserved and tested in our screening were not able to drive the GFP protein to the nucleus. Only two strong candidates, Rv0229c and Rv3876, were finally shown to reach the nucleus and were further analyzed. The molecular mass of Rv3876 protein fused with GFP (∼78 kDa) was greater than 40 kDa, hence, the recombinant protein can only enter the nucleus by interacting with the eukaryotic transport machinery to be actively transported through the NPC, driven by the RanGDP/GTP gradient. Consistently, mutations modifying the positively charged amino acids in the predicted NLS abolished nuclear localization of the hybrid GFP-Rv3876 protein, confirming the necessity of active transport and importance of the NLS motif (Fig. 16C and D). As for Rv0229c, the molecular mass of GFP-Rv0229c hybrid protein (∼33 kDa), was within the exclusion limits of the NPC, suggesting the possibility of passive diffusion of this protein into nuclei of host cells. However, mutations changing simultaneously the positively charged amino acids of two different motifs within the Rv0229c protein abolish the nuclear localization (Fig. 15C and D). First, this demonstrates that an active transport into the nucleus is also necessary for the GFP-Rv0229c hybrid protein. Second, it shows that i) two possible NLS exist in the Rv0229c sequence and ii) each of these motifs is sufficient for the nuclear import. Delivery of bacterial proteins to the cytoplasm of host cells is a prerequisite for nuclear targeting. For these reasons, we first tested whether Rv0229c and Rv3876 were actually secreted by Mtb. Interestingly, analyses of the culture filtrate proteins of Mtb by western blotting confirmed the presence of both Rv0229c and Rv3876 proteins, indicating that these candidates are probably secreted proteins. All the data gained so far on the nuclear import of Rv0229c and Rv3876 have been obtained in artificial systems, after expression from a plasmid vector transfected in eukaryotic cells. Future studies will have to determine whether these proteins actually translocate into the cell nucleus during infection of macrophages by Mtb. Genetic tools are being constructed that will allow to address this point. For instance, a variant of H37Rv expressing constitutively a DsRed fluorescent protein will be transformed with a pVV16- Rv0229c-GFP plasmid expressing a hybrid protein Rv0229c-GFP. This strain will be used to

CHAPTER 4 DISCUSSION 59 infect HMDM and fluorescent microscopy analysis will analyze whether green fluorescence is visible in the nucleus of infected macrophages (carrying red Mtb). Of course, as Mtb is present inside phagosomes, the proteins secreted out of the bacterial cells must cross the phagosome membrane to be able to reach the macrophage nucleus. This must be possible because the localization in the nucleus of infected macrophage of a protein produced by infecting mycobacteria has been demonstrated in the case of the histone acetyltransferase Rv3423.1 (Jose et al., 2015) or the 5-methylcytosine-specific DNA methyltransferase Rv2966c (Sharma et al., 2015). Although the underlying mechanism is not firmly established, virulent Mtb are known to secrete small highly immunogenic proteins such as ESAT6, a substrate of the type VII secretion system ESX-1, that induce rupture of the phagosome membrane, allowing the presence of mycobacterial proteins in the cytosol of infected macrophages (Brodin et al., 2006; Simeone et al., 2012; Simeone et al., 2009; Simeone et al., 2015). Mtb can also directly deliver specific virulence factors to host cells through the ESX-5 secretion system (Abdallah et al., 2011). Future work will also have to investigate the kinetics of nuclear import of GFP-Rv0229c and GFP-Rv3876, and also their precise localization in the nucleus. Indeed, we noted that both exhibited a sub-nuclear localization in HEK 293 cells and in HMDM, showing a “spotty” appearance inside the nucleus (Fig. 12A and C). Such behavior has already been observed with nucleomodulins. For instance, SET-domain nucleomodulins (from Legionella and Burkholderia species) translocate into the nucleus showing a “spotty” appearance within a subnuclear structure, the nucleolus, which is the site of both ribosomal RNA (rRNA) synthesis and the assembly of ribosomal subunit, and target and modify rDNA chromatin. This ultimately activates rDNA transcription and promotes bacterial intracellular survival (Li et al., 2013). Therefore, it would be highly interesting to characterize the nature of the sub- domains or the functional signification of this nuclear localization for both EGFP-Rv0229c and EGFP-Rv3876. For instance, FISH experiments using specific markers of nucleolus or of speckle in parallel with the localization of EGFP-Rv0229c or EGFP-Rv3876 will be able to address this point. A crucial point in this project will be to identify the functional consequences of the expression of Rv0229c or Rv3876 in infected hosts. We have started the analysis of the transcriptomic effects of the expression of the two proteins in transfected macrophages. Although we did not have the time to complete these experiments, we believe that some

Figure 20. Analysis of the RNAs co‐immunoprecipitated with Rv0229c. HEK 293 cells were transfected with the empty vector pIRES‐hrGFP II or with p3Flag‐Rv0229c‐IRES‐hrGFP II, expressing the 3‐FLG‐Rv0229c protein. Total cell lysates were obtained and immunoprecipitated with antibodies against FLAG tag. Left: proteins of the total lysate, or the immunoprecipitated fraction, were analyzed by western blotting using antibodies against FLAG tag. Right: RNAs present in the total lysates, or the immunoprecipitated fraction, were radiolabeled with terminal transferase and γ‐33P ATP, and analyzed on polyacrylamide denaturing gels. Autoradiography of the gel is shown. (Yves Roméo, LBME, Toulouse).

CHAPTER 4 DISCUSSION 60 improvement of the transfection efficiency coupled to sorting of the transfected cells will allow their finalization. Infection of mice with deleted and complemented strains of Mtb will allow to determine the importance of these genes not only for the initial stage of multiplication within macrophages, but also for the spreading in an infected host, for long term survival within this host, and also for a possible effect on the immune response and physiopathological output. At this stage, one can only speculate on the molecular function of Rv3876 and Rv0229c. In the case of Rv0229c, the strong PIN domain signature harbored by this protein suggests that it could be an RNase. We initiated a collaboration with a team specialized in the metabolism of rRNAs in eukaryotic cells (Anthony Henras and Yves Roméo, Laboratory of Eukaryotic Molecular Biology, CNRS/University of Toulouse), to try and identify the substrates of Rv0229c by biochemical approaches. HEK 293 cells were transfected with p3Flag-Rv0229c- IRES-hrGFP II and crude extracts of the transfected cells were immunoprecipitated with antibodies against FLAGTM tag. Then, radioactive labeling of the RNAs present in the precipitated pellet and gel electrophoresis analysis allow visualization of RNAs co- precipitated with the protein of interest. As shown in Figure 20, the presence of Rv0229c resulted in an increased amount of several species of RNAs in the pellet of the immunoprecipitation. Although no band seems absolutely specific of the presence of Rv0229c, these very preliminary results (the experiment has only been performed once) suggest that Rv0229c is indeed an RNA-binding protein. Refinement of this experiment, including separation of nuclei from the rest of the cell lysate and a “soft” disruption of the nuclei to avoid destruction of macromolecular complexes, followed by a global sequencing of the precipitated RNAs will have to be performed. For its part, Rv3876 (EspI), is known to harbor an ATPase domain, and seems to be involved inside the bacterial cells in the regulation of secretion through ESX-1 in response to the level of intracellular ATP (Zhang et al., 2014). This could be achieved by protein-protein interactions depending on the presence of ATP, and suggests that experiments such as co- immunoprecipitation followed by proteomic analysis could help in identifying the partners of Rv3876 within the nucleus of infected cells.

CHAPTER 4 DISCUSSION 61

Finally, although much work remains to be done, we believe that our work opened a new and attractive research area, identifying novel mechanisms contributing to Mtb virulence, through manipulation of the genetic expression of the infected host cells.

REFERENCES

Abdallah, A.M., Bestebroer, J., Savage, N.D., de Punder, K., van Zon, M., Wilson, L., Korbee, C.J., van der Sar, A.M., Ottenhoff, T.H., van der Wel, N.N., et al. (2011). Mycobacterial secretion systems ESX-1 and ESX-5 play distinct roles in host cell death and inflammasome activation. J Immunol 187, 4744-4753.

Abdallah, A.M., Gey van Pittius, N.C., Champion, P.A., Cox, J., Luirink, J., Vandenbroucke- Grauls, C.M., Appelmelk, B.J., and Bitter, W. (2007). Type VII secretion--mycobacteria show the way. Nat Rev Microbiol 5, 883-891.

Achkar, J.M., Chan, J., and Casadevall, A. (2015). B cells and antibodies in the defense against Mycobacterium tuberculosis infection. Immunol Rev 264, 167-181.

Agarwal, S., Jin, H., Pancholi, P., and Pancholi, V. (2012). Serine/threonine phosphatase (SP- STP), secreted from Streptococcus pyogenes, is a pro-apoptotic protein. J Biol Chem 287, 9147-9167.

Akira, S., and Takeda, K. (2004). Toll-like receptor signalling. Nat Rev Immunol 4, 499-511.

Albrethsen, J., Agner, J., Piersma, S.R., Hojrup, P., Pham, T.V., Weldingh, K., Jimenez, C.R., Andersen, P., and Rosenkrands, I. (2013). Proteomic profiling of Mycobacterium tuberculosis identifies nutrient-starvation-responsive toxin-antitoxin systems. Mol Cell Proteomics 12, 1180-1191.

Almeida Da Silva, P.E., and Palomino, J.C. (2011). Molecular basis and mechanisms of drug resistance in Mycobacterium tuberculosis: classical and new drugs. J Antimicrob Chemother 66, 1417-1430.

Amaral, E.P., Ribeiro, S.C., Lanes, V.R., Almeida, F.M., de Andrade, M.R., Bomfim, C.C., Salles, E.M., Bortoluci, K.R., Coutinho-Silva, R., Hirata, M.H., et al. (2014). Pulmonary infection with hypervirulent Mycobacteria reveals a crucial role for the P2X7 receptor in aggressive forms of tuberculosis. PLoS Pathog 10, e1004188.

Andersen, P., Askgaard, D., Ljungqvist, L., Bennedsen, J., and Heron, I. (1991). Proteins released from Mycobacterium tuberculosis during growth. Infect Immun 59, 1905-1910.

Arbibe, L., Kim, D.W., Batsche, E., Pedron, T., Mateescu, B., Muchardt, C., Parsot, C., and Sansonetti, P.J. (2007). An injected bacterial effector targets chromatin access for transcription factor NF-kappaB to alter transcription of host genes involved in immune responses. Nat Immunol 8, 47-56.

Arbues, A., Aguilo, J.I., Gonzalo-Asensio, J., Marinova, D., Uranga, S., Puentes, E., Fernandez, C., Parra, A., Cardona, P.J., Vilaplana, C., et al. (2013). Construction, characterization and preclinical evaluation of MTBVAC, the first live-attenuated M. tuberculosis-based vaccine to enter clinical trials. Vaccine 31, 4867-4873.

REFERENCES 63

Arthur, J.S., and Ley, S.C. (2013). Mitogen-activated protein kinases in innate immunity. Nat Rev Immunol 13, 679-692.

Banerjee, S., Farhana, A., Ehtesham, N.Z., and Hasnain, S.E. (2011). Iron acquisition, assimilation and regulation in mycobacteria. Infect Genet Evol 11, 825-838.

Behar, S.M., Divangahi, M., and Remold, H.G. (2010). Evasion of innate immunity by Mycobacterium tuberculosis: is death an exit strategy? Nat Rev Microbiol 8, 668-674.

Behar, S.M., Martin, C.J., Booty, M.G., Nishimura, T., Zhao, X., Gan, H.X., Divangahi, M., and Remold, H.G. (2011). Apoptosis is an innate defense function of macrophages against Mycobacterium tuberculosis. Mucosal Immunol 4, 279-287.

Benabdillah, R., Mota, L.J., Lutzelschwab, S., Demoinet, E., and Cornelis, G.R. (2004). Identification of a nuclear targeting signal in YopM from Yersinia spp. Microb Pathog 36, 247-261.

Berger, S.L. (2007). The complex language of chromatin regulation during transcription. Nature 447, 407-412.

Bierne, H., and Cossart, P. (2012). When bacteria target the nucleus: the emerging family of nucleomodulins. Cell Microbiol 14, 622-633.

Bierne, H., Hamon, M., and Cossart, P. (2012). Epigenetics and bacterial infections. Cold Spring Harb Perspect Med 2, a010272.

Bierne, H., Tham, T.N., Batsche, E., Dumay, A., Leguillou, M., Kerneis-Golsteyn, S., Regnault, B., Seeler, J.S., Muchardt, C., Feunteun, J., et al. (2009). Human BAHD1 promotes heterochromatic gene silencing. Proc Natl Acad Sci U S A 106, 13826-13831.

Bitter, W., Houben, E.N., Bottai, D., Brodin, P., Brown, E.J., Cox, J.S., Derbyshire, K., Fortune, S.M., Gao, L.Y., Liu, J., et al. (2009). Systematic genetic nomenclature for type VII secretion systems. PLoS Pathog 5, e1000507.

Boch, J., and Bonas, U. (2010). Xanthomonas AvrBs3 family-type III effectors: discovery and function. Annu Rev Phytopathol 48, 419-436.

Bogdanove, A.J., Schornack, S., and Lahaye, T. (2010). TAL effectors: finding plant genes for disease and defense. Curr Opin Plant Biol 13, 394-401.

Boldrin, F., Ventura, M., Degiacomi, G., Ravishankar, S., Sala, C., Svetlikova, Z., Ambady, A., Dhar, N., Kordulakova, J., Zhang, M., et al. (2014). The phosphatidyl-myo-inositol mannosyltransferase PimA is essential for Mycobacterium tuberculosis growth in vitro and in vivo. J Bacteriol 196, 3441-3451.

Boradia, V.M., Malhotra, H., Thakkar, J.S., Tillu, V.A., Vuppala, B., Patil, P., Sheokand, N., Sharma, P., Chauhan, A.S., Raje, M., et al. (2014). Mycobacterium tuberculosis acquires iron by cell-surface sequestration and internalization of human holo-transferrin. Nat Commun 5, 4730.

REFERENCES 64

Botella, H., Peyron, P., Levillain, F., Poincloux, R., Poquet, Y., Brandli, I., Wang, C., Tailleux, L., Tilleul, S., Charriere, G.M., et al. (2011). Mycobacterial p(1)-type ATPases mediate resistance to zinc poisoning in human macrophages. Cell Host Microbe 10, 248-259.

Botella, H., Stadthagen, G., Lugo-Villarino, G., de Chastellier, C., and Neyrolles, O. (2012). Metallobiology of host-pathogen interactions: an intoxicating new insight. Trends Microbiol 20, 106-112.

Bottai, D., Majlessi, L., Simeone, R., Frigui, W., Laurent, C., Lenormand, P., Chen, J., Rosenkrands, I., Huerre, M., Leclerc, C., et al. (2011). ESAT-6 secretion-independent impact of ESX-1 genes espF and espG1 on virulence of Mycobacterium tuberculosis. J Infect Dis 203, 1155-1164.

Bougdour, A., Durandau, E., Brenier-Pinchart, M.P., Ortet, P., Barakat, M., Kieffer, S., Curt- Varesano, A., Curt-Bertini, R.L., Bastien, O., Coute, Y., et al. (2013). Host cell subversion by Toxoplasma GRA16, an exported dense granule protein that targets the host cell nucleus and alters gene expression. Cell Host Microbe 13, 489-500.

Brennan, D.F., and Barford, D. (2009). Eliminylation: a post-translational modification catalyzed by phosphothreonine lyases. Trends Biochem Sci 34, 108-114.

Brennan, P.J. (2003). Structure, function, and biogenesis of the cell wall of Mycobacterium tuberculosis. Tuberculosis (Edinb) 83, 91-97.

Brodin, P., Majlessi, L., Marsollier, L., de Jonge, M.I., Bottai, D., Demangel, C., Hinds, J., Neyrolles, O., Butcher, P.D., Leclerc, C., et al. (2006). Dissection of ESAT-6 system 1 of Mycobacterium tuberculosis and impact on immunogenicity and virulence. Infect Immun 74, 88-98.

Cambau, E., and Drancourt, M. (2014). Steps towards the discovery of Mycobacterium tuberculosis by Robert Koch, 1882. Clin Microbiol Infect 20, 196-201.

Cambronne, E.D., and Roy, C.R. (2006). Recognition and delivery of effector proteins into eukaryotic cells by bacterial secretion systems. Traffic 7, 929-939.

Canton, J. (2014). Phagosome maturation in polarized macrophages. J Leukoc Biol 96, 729- 738.

Cardona, P.J. (2006). RUTI: a new chance to shorten the treatment of latent tuberculosis infection. Tuberculosis (Edinb) 86, 273-289.

Carel, C., Nukdee, K., Cantaloube, S., Bonne, M., Diagne, C.T., Laval, F., Daffe, M., and Zerbib, D. (2014). Mycobacterium tuberculosis proteins involved in mycolic acid synthesis and transport localize dynamically to the old growing pole and septum. PLoS One 9, e97148.

Carroll, M.V., Lack, N., Sim, E., Krarup, A., and Sim, R.B. (2009). Multiple routes of complement activation by Mycobacterium bovis BCG. Mol Immunol 46, 3367-3378.

Cellier, M.F., Courville, P., and Campion, C. (2007). Nramp1 phagocyte intracellular metal withdrawal defense. Microbes Infect 9, 1662-1670.

REFERENCES 65

Champion, P.A., and Cox, J.S. (2007). Protein secretion systems in Mycobacteria. Cell Microbiol 9, 1376-1384.

Chatterjee, D., and Khoo, K.H. (1998). Mycobacterial lipoarabinomannan: an extraordinary lipoheteroglycan with profound physiological effects. Glycobiology 8, 113-120.

Citovsky, V., Warnick, D., and Zambryski, P. (1994). Nuclear import of Agrobacterium VirD2 and VirE2 proteins in maize and tobacco. Proc Natl Acad Sci U S A 91, 3210-3214.

Citovsky, V., Zupan, J., Warnick, D., and Zambryski, P. (1992). Nuclear localization of Agrobacterium VirE2 protein in plant cells. Science 256, 1802-1805.

Cole, S.T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S.V., Eiglmeier, K., Gas, S., Barry, C.E., 3rd, et al. (1998). Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537-544.

Cooper, A.M., Dalton, D.K., Stewart, T.A., Griffin, J.P., Russell, D.G., and Orme, I.M. (1993). Disseminated tuberculosis in interferon gamma gene-disrupted mice. J Exp Med 178, 2243-2247.

Cortes, T., Schubert, O.T., Rose, G., Arnvig, K.B., Comas, I., Aebersold, R., and Young, D.B. (2013). Genome-wide mapping of transcriptional start sites defines an extensive leaderless transcriptome in Mycobacterium tuberculosis. Cell Rep 5, 1121-1131.

Creagh, E.M., Conroy, H., and Martin, S.J. (2003). Caspase-activation pathways in apoptosis and immunity. Immunol Rev 193, 10-21.

Crick, D.C., Mahapatra, S., and Brennan, P.J. (2001). Biosynthesis of the arabinogalactan- peptidoglycan complex of Mycobacterium tuberculosis. Glycobiology 11, 107R-118R.

Daleke, M.H., Cascioferro, A., de Punder, K., Ummels, R., Abdallah, A.M., van der Wel, N., Peters, P.J., Luirink, J., Manganelli, R., and Bitter, W. (2011). Conserved Pro-Glu (PE) and Pro-Pro-Glu (PPE) protein domains target LipY lipases of pathogenic mycobacteria to the cell surface via the ESX-5 pathway. J Biol Chem 286, 19024-19034.

Danelishvili, L., Everman, J.L., McNamara, M.J., and Bermudez, L.E. (2011). Inhibition of the Plasma-Membrane-Associated Serine Protease Cathepsin G by Mycobacterium tuberculosis Rv3364c Suppresses Caspase-1 and Pyroptosis in Macrophages. Front Microbiol 2, 281.

Danelishvili, L., Yamazaki, Y., Selker, J., and Bermudez, L.E. (2010). Secreted Mycobacterium tuberculosis Rv3654c and Rv3655c proteins participate in the suppression of macrophage apoptosis. PLoS One 5, e10474.

Daniel, T.M. (2006). The history of tuberculosis. Respir Med 100, 1862-1870.

Dartois, V. (2014). The path of anti-tuberculosis drugs: from blood to lesions to mycobacterial cells. Nat Rev Microbiol 12, 159-167.

REFERENCES 66

Das, C., Ghosh, T.S., and Mande, S.S. (2011). Computational analysis of the ESX-1 region of Mycobacterium tuberculosis: insights into the mechanism of type VII secretion system. PLoS One 6, e27980. de Souza, G.A., Leversen, N.A., Malen, H., and Wiker, H.G. (2011). Bacterial proteins with cleaved or uncleaved signal peptides of the general secretory pathway. J Proteomics 75, 502-510.

De Voss, J.J., Rutter, K., Schroeder, B.G., and Barry, C.E., 3rd (1999). Iron acquisition and metabolism by mycobacteria. J Bacteriol 181, 4443-4451.

Deghmane, A.E., Soualhine, H., Bach, H., Sendide, K., Itoh, S., Tam, A., Noubir, S., Talal, A., Lo, R., Toyoshima, S., et al. (2007). Lipoamide dehydrogenase mediates retention of coronin- 1 on BCG vacuoles, leading to arrest in phagosome maturation. J Cell Sci 120, 2796-2806.

Desjardins, M. (1995). Biogenesis of phagolysosomes: the 'kiss and run' hypothesis. Trends Cell Biol 5, 183-186.

Deslandes, L., Olivier, J., Peeters, N., Feng, D.X., Khounlotham, M., Boucher, C., Somssich, I., Genin, S., and Marco, Y. (2003). Physical interaction between RRS1-R, a protein conferring resistance to bacterial wilt, and PopP2, a type III effector targeted to the plant nucleus. Proc Natl Acad Sci U S A 100, 8024-8029.

Dheda, K., Schwander, S.K., Zhu, B., van Zyl-Smit, R.N., and Zhang, Y. (2010). The immunology of tuberculosis: from bench to bedside. Respirology 15, 433-450.

Dingwall, C., and Laskey, R.A. (1991). Nuclear targeting sequences--a consensus? Trends Biochem Sci 16, 478-481.

Dingwall, C., Robbins, J., Dilworth, S.M., Roberts, B., and Richardson, W.D. (1988). The nucleoplasmin nuclear location sequence is larger and more complex than that of SV-40 large T antigen. J Cell Biol 107, 841-849.

Divangahi, M., Chen, M., Gan, H., Desjardins, D., Hickman, T.T., Lee, D.M., Fortune, S., Behar, S.M., and Remold, H.G. (2009). Mycobacterium tuberculosis evades macrophage defenses by inhibiting plasma membrane repair. Nat Immunol 10, 899-906.

Drlica, K., Hiasa, H., Kerns, R., Malik, M., Mustaev, A., and Zhao, X. (2009). Quinolones: action and resistance updated. Curr Top Med Chem 9, 981-998.

Drocourt, D., Calmels, T., Reynes, J.P., Baron, M., and Tiraby, G. (1990). Cassettes of the Streptoalloteichus hindustanus ble gene for transformation of lower and higher eukaryotes to phleomycin resistance. Nucleic Acids Res 18, 4009.

Dunkelberger, J.R., and Song, W.C. (2010). Complement and its role in innate and adaptive immune responses. Cell Res 20, 34-50.

Ehrt, S., Guo, X.V., Hickey, C.M., Ryou, M., Monteleone, M., Riley, L.W., and Schnappinger, D. (2005). Controlling gene expression in mycobacteria with anhydrotetracycline and Tet repressor. Nucleic Acids Res 33, e21.

REFERENCES 67

Einhauer, A., and Jungbauer, A. (2001). The FLAG peptide, a versatile fusion tag for the purification of recombinant proteins. J Biochem Biophys Methods 49, 455-465.

Eoh, H., and Rhee, K.Y. (2014). Methylcitrate cycle defines the bactericidal essentiality of isocitrate lyase for survival of Mycobacterium tuberculosis on fatty acids. Proc Natl Acad Sci U S A 111, 4976-4981.

Failla, M.L. (2003). Trace elements and host defense: recent advances and continuing challenges. J Nutr 133, 1443S-1447S.

Fang, F.C. (2004). Antimicrobial reactive oxygen and nitrogen species: concepts and controversies. Nat Rev Microbiol 2, 820-832.

Field, D., Feil, E.J., and Wilson, G.A. (2005). Databases and software for the comparison of prokaryotic genomes. Microbiology 151, 2125-2132.

Finlay, B.B., and Cossart, P. (1997). Exploitation of mammalian host cell functions by bacterial pathogens. Science 276, 718-725.

Flannagan, R.S., Cosio, G., and Grinstein, S. (2009). Antimicrobial mechanisms of phagocytes and bacterial evasion strategies. Nat Rev Microbiol 7, 355-366.

Flynn, J.L., and Chan, J. (2003). Immune evasion by Mycobacterium tuberculosis: living with the enemy. Curr Opin Immunol 15, 450-455.

Flynn, J.L., Chan, J., Triebold, K.J., Dalton, D.K., Stewart, T.A., and Bloom, B.R. (1993). An essential role for interferon gamma in resistance to Mycobacterium tuberculosis infection. J Exp Med 178, 2249-2254.

Forman, H.J., and Torres, M. (2002). Reactive oxygen species and cell signaling: respiratory burst in macrophage signaling. Am J Respir Crit Care Med 166, S4-8.

Fratti, R.A., Backer, J.M., Gruenberg, J., Corvera, S., and Deretic, V. (2001). Role of phosphatidylinositol 3-kinase and Rab5 effectors in phagosomal biogenesis and mycobacterial phagosome maturation arrest. J Cell Biol 154, 631-644.

Fu, L.M., and Fu-Liu, C.S. (2002). Is Mycobacterium tuberculosis a closer relative to Gram- positive or Gram-negative bacterial pathogens? Tuberculosis (Edinb) 82, 85-90.

Fukuda, T., Matsumura, T., Ato, M., Hamasaki, M., Nishiuchi, Y., Murakami, Y., Maeda, Y., Yoshimori, T., Matsumoto, S., Kobayashi, K., et al. (2013). Critical roles for lipomannan and lipoarabinomannan in cell wall integrity of mycobacteria and pathogenesis of tuberculosis. MBio 4, e00472-00412.

Gangaidzo, I.T., Moyo, V.M., Mvundura, E., Aggrey, G., Murphree, N.L., Khumalo, H., Saungweme, T., Kasvosve, I., Gomo, Z.A., Rouault, T., et al. (2001). Association of pulmonary tuberculosis with increased dietary iron. J Infect Dis 184, 936-939.

Gengenbacher, M., and Kaufmann, S.H. (2012). Mycobacterium tuberculosis: success through dormancy. FEMS Microbiol Rev 36, 514-532.

REFERENCES 68

Garcia-Garcia, J.C., Rennoll-Bankert, K.E., Pelly, S., Milstone, A.M., and Dumler, J.S. (2009). Silencing of host cell CYBB gene expression by the nuclear effector AnkA of the intracellular pathogen Anaplasma phagocytophilum. Infect Immun 77, 2385-2391.

Gey van Pittius, N.C., Sampson, S.L., Lee, H., Kim, Y., van Helden, P.D., and Warren, R.M. (2006). Evolution and expansion of the Mycobacterium tuberculosis PE and PPE multigene families and their association with the duplication of the ESAT-6 (esx) gene cluster regions. BMC Evol Biol 6, 95.

Gobin, J., and Horwitz, M.A. (1996). Exochelins of Mycobacterium tuberculosis remove iron from human iron-binding proteins and donate iron to mycobactins in the M. tuberculosis cell wall. J Exp Med 183, 1527-1532.

Gould, T.A., van de Langemheen, H., Munoz-Elias, E.J., McKinney, J.D., and Sacchettini, J.C. (2006). Dual role of isocitrate lyase 1 in the glyoxylate and methylcitrate cycles in Mycobacterium tuberculosis. Mol Microbiol 61, 940-947.

Gouzy, A., Larrouy-Maumus, G., Bottai, D., Levillain, F., Dumas, A., Wallach, J.B., Caire- Brandli, I., de Chastellier, C., Wu, T.D., Poincloux, R., et al. (2014). Mycobacterium tuberculosis exploits asparagine to assimilate nitrogen and resist acid stress during infection. PLoS Pathog 10, e1003928.

Gouzy, A., Larrouy-Maumus, G., Wu, T.D., Peixoto, A., Levillain, F., Lugo-Villarino, G., Guerquin-Kern, J.L., de Carvalho, L.P., Poquet, Y., and Neyrolles, O. (2013). Mycobacterium tuberculosis nitrogen assimilation and host colonization require aspartate. Nat Chem Biol 9, 674-676.

Grange, J.M. (1982). Koch's Tubercle Bacillus - a centenary reappraisal. Zentralblatt fur Bakteriologie, Mikrobiologie und Hygiene 1 Abt Originale A, Medizinische Mikrobiologie, Infektionskrankheiten und Parasitologie = International journal of microbiology and hygiene A, Medical microbiology, infectiousdiseases, parasitology 251, 297-307.

Grewal, S.I., and Jia, S. (2007). Heterochromatin revisited. Nat Rev Genet 8, 35-46.

Griffin, J.E., Gawronski, J.D., Dejesus, M.A., Ioerger, T.R., Akerley, B.J., and Sassetti, C.M. (2011). High-resolution phenotypic profiling defines genes essential for mycobacterial growth and cholesterol catabolism. PLoS Pathog 7, e1002251.

Griffith, D.E., Aksamit, T., Brown-Elliott, B.A., Catanzaro, A., Daley, C., Gordin, F., Holland, S.M., Horsburgh, R., Huitt, G., Iademarco, M.F., et al. (2007). An official ATS/IDSA statement: diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir Crit Care Med 175, 367-416.

Griffiths, G., Nystrom, B., Sable, S.B., and Khuller, G.K. (2010). Nanobead-based interventions for the treatment and prevention of tuberculosis. Nat Rev Microbiol 8, 827-834.

Grode, L., Ganoza, C.A., Brohm, C., Weiner, J., 3rd, Eisele, B., and Kaufmann, S.H. (2013). Safety and immunogenicity of the recombinant BCG vaccine VPM1002 in a phase 1 open- label randomized clinical trial. Vaccine 31, 1340-1348.

REFERENCES 69

Gueneron, M., Timmers, A.C., Boucher, C., and Arlat, M. (2000). Two novel proteins, PopB, which has functional nuclear localization signals, and PopC, which has a large leucine-rich repeat domain, are secreted through the hrp-secretion apparatus of Ralstonia solanacearum. Mol Microbiol 36, 261-277.

Gupta, A., Kaul, A., Tsolaki, A.G., Kishore, U., and Bhakta, S. (2012). Mycobacterium tuberculosis: immune evasion, latency and reactivation. Immunobiology 217, 363-374.

Hett, E.C., and Rubin, E.J. (2008). Bacterial growth and cell division: a mycobacterial perspective. Microbiol Mol Biol Rev 72, 126-156, table of contents.

Hoffmann, C., Leis, A., Niederweis, M., Plitzko, J.M., and Engelhardt, H. (2008). Disclosure of the mycobacterial outer membrane: cryo-electron tomography and vitreous sections reveal the lipid bilayer structure. Proc Natl Acad Sci U S A 105, 3963-3967.

Houben, E.N., Korotkov, K.V., and Bitter, W. (2014). Take five - Type VII secretion systems of Mycobacteria. Biochim Biophys Acta 1843, 1707-1716.

Howard, E.A., Zupan, J.R., Citovsky, V., and Zambryski, P.C. (1992). The VirD2 protein of A. tumefaciens contains a C-terminal bipartite nuclear localization signal: implications for nuclear uptake of DNA in plant cells. Cell 68, 109-118.

ISTC (2014). TB CARE I, The Hague, 3 edn (International Standards for Tuberculosis Care ).

Jamet, S., Quentin, Y., Coudray, C., Texier, P., Laval, F., Daffe, M., Fichant, G., and Cam, K. (2015). Evolution of mycolic acid biosynthesis genes and their regulation during starvation in Mycobacterium tuberculosis. J Bacteriol.

Jasenosky, L.D., Scriba, T.J., Hanekom, W.A., and Goldfeld, A.E. (2015). T cells and adaptive immunity to Mycobacterium tuberculosis in humans. Immunol Rev 264, 74-87.

Jayakumar, D., Jacobs, W.R., Jr., and Narayanan, S. (2008). Protein kinase E of Mycobacterium tuberculosis has a role in the nitric oxide stress response and apoptosis in a human macrophage model of infection. Cell Microbiol 10, 365-374.

Jose, L., Ramachandran, R., Bhagavat, R., Gomez, R.L., Chandran, A., Raghunandanan, S., Omkumar, R.V., Chandra, N., Mundayoor, S., and Kumar, R.A. (2015). Hypothetical protein Rv3423.1 of Mycobacterium tuberculosis is a histone acetyltransferase. Febs J.

Kalimuthu, S., and Se-Kwon, K. (2013). Cell survival and apoptosis signaling as therapeutic target for cancer: marine bioactive compounds. Int J Mol Sci 14, 2334-2354.

Katta, S.S., Smoyer, C.J., and Jaspersen, S.L. (2014). Destination: inner nuclear membrane. Trends Cell Biol 24, 221-229.

Katti, M.K., Dai, G., Armitige, L.Y., Rivera Marrero, C., Daniel, S., Singh, C.R., Lindsey, D.R., Dhandayuthapani, S., Hunter, R.L., and Jagannath, C. (2008). The Delta fbpA mutant derived from Mycobacterium tuberculosis H37Rv has an enhanced susceptibility to intracellular antimicrobial oxidative mechanisms, undergoes limited phagosome maturation and activates macrophages and dendritic cells. Cell Microbiol 10, 1286-1303.

REFERENCES 70

Kaufmann, S.H. (2001). How can immunology contribute to the control of tuberculosis? Nat Rev Immunol 1, 20-30.

Kaufmann, S.H. (2005). Robert Koch, the Nobel Prize, and the ongoing threat of tuberculosis. N Engl J Med 353, 2423-2426.

Kaur, D., Guerin, M.E., Skovierova, H., Brennan, P.J., and Jackson, M. (2009). Chapter 2: Biogenesis of the cell wall and other glycoconjugates of Mycobacterium tuberculosis. Adv Appl Microbiol 69, 23-78.

Kay, S., Hahn, S., Marois, E., Hause, G., and Bonas, U. (2007). A bacterial effector acts as a plant transcription factor and induces a cell size regulator. Science 318, 648-651.

Kieser, K.J., and Rubin, E.J. (2014). How sisters grow apart: mycobacterial growth and division. Nat Rev Microbiol 12, 550-562.

Killick, K.E., Ni Cheallaigh, C., O'Farrelly, C., Hokamp, K., MacHugh, D.E., and Harris, J. (2013). Receptor-mediated recognition of mycobacterial pathogens. Cell Microbiol 15, 1484-1495.

Kim, H.E., Du, F., Fang, M., and Wang, X. (2005). Formation of apoptosome is initiated by cytochrome c-induced dATP hydrolysis and subsequent nucleotide exchange on Apaf-1. Proc Natl Acad Sci U S A 102, 17545-17550.

Kleinnijenhuis, J., Oosting, M., Joosten, L.A., Netea, M.G., and Van Crevel, R. (2011). Innate immune recognition of Mycobacterium tuberculosis. Clin Dev Immunol 2011, 405310.

Knight, J.S., Lan, K., Subramanian, C., and Robertson, E.S. (2003). Epstein-Barr virus nuclear antigen 3C recruits histone deacetylase activity and associates with the corepressors mSin3A and NCoR in human B-cell lines. J Virol 77, 4261-4272.

Kosugi, S., Hasebe, M., Matsumura, N., Takashima, H., Miyamoto-Sato, E., Tomita, M., and Yanagawa, H. (2009a). Six classes of nuclear localization signals specific to different binding grooves of importin alpha. J Biol Chem 284, 478-485.

Kosugi, S., Hasebe, M., Tomita, M., and Yanagawa, H. (2009b). Systematic identification of cell cycle-dependent yeast nucleocytoplasmic shuttling proteins by prediction of composite motifs. Proc Natl Acad Sci U S A 106, 10171-10176.

Lad, S.P., Li, J., da Silva Correia, J., Pan, Q., Gadwal, S., Ulevitch, R.J., and Li, E. (2007). Cleavage of p65/RelA of the NF-kappaB pathway by Chlamydia. Proc Natl Acad Sci U S A 104, 2933-2938.

Lange, A., Mills, R.E., Lange, C.J., Stewart, M., Devine, S.E., and Corbett, A.H. (2007). Classical nuclear localization signals: definition, function, and interaction with importin alpha. J Biol Chem 282, 5101-5105.

Lebreton, A., Lakisic, G., Job, V., Fritsch, L., Tham, T.N., Camejo, A., Mattei, P.J., Regnault, B., Nahori, M.A., Cabanes, D., et al. (2011). A bacterial protein targets the BAHD1 chromatin complex to stimulate type III interferon response. Science 331, 1319-1321.

REFERENCES 71

Lee, J.S., Smith, E., and Shilatifard, A. (2010). The language of histone crosstalk. Cell 142, 682-685.

Lerner, T.R., Borel, S., and Gutierrez, M.G. (2015). The innate immune response in human tuberculosis. Cell Microbiol 17, 1277-1285.

Levine, M. (2010). Transcriptional enhancers in animal development and evolution. Curr Biol 20, R754-763.

Li, B., Carey, M., and Workman, J.L. (2007). The role of chromatin during transcription. Cell 128, 707-719.

Li, T., Lu, Q., Wang, G., Xu, H., Huang, H., Cai, T., Kan, B., Ge, J., and Shao, F. (2013). SET- domain bacterial effectors target heterochromatin protein 1 to activate host rDNA transcription. EMBO Rep 14, 733-740.

Li, X., Yang, Y., Zhou, F., Zhang, Y., Lu, H., Jin, Q., and Gao, L. (2011). SLC11A1 (NRAMP1) polymorphisms and tuberculosis susceptibility: updated systematic review and meta- analysis. PLoS One 6, e15831.

Lugo-Villarino, G., and Neyrolles, O. (2014). Manipulation of the mononuclear phagocyte system by Mycobacterium tuberculosis. Cold Spring Harb Perspect Med 4, a018549.

Ma, Z., Lienhardt, C., McIlleron, H., Nunn, A.J., and Wang, X. (2010). Global tuberculosis drug development pipeline: the need and the reality. Lancet 375, 2100-2109.

MacMicking, J.D. (2009). Recognizing macrophage activation and host defense. Cell Host Microbe 5, 405-407.

MacMicking, J.D. (2012). Interferon-inducible effector mechanisms in cell-autonomous immunity. Nat Rev Immunol 12, 367-382.

MacMicking, J.D., Taylor, G.A., and McKinney, J.D. (2003). Immune control of tuberculosis by IFN-gamma-inducible LRG-47. Science 302, 654-659.

Makarov, V., Lechartier, B., Zhang, M., Neres, J., van der Sar, A.M., Raadsen, S.A., Hartkoorn, R.C., Ryabova, O.B., Vocat, A., Decosterd, L.A., et al. (2014). Towards a new combination therapy for tuberculosis with next generation benzothiazinones. EMBO Mol Med 6, 372- 383.

Makarov, V., Manina, G., Mikusova, K., Mollmann, U., Ryabova, O., Saint-Joanis, B., Dhar, N., Pasca, M.R., Buroni, S., Lucarelli, A.P., et al. (2009). Benzothiazinones kill Mycobacterium tuberculosis by blocking arabinan synthesis. Science 324, 801-804.

Manzur, K.L., Farooq, A., Zeng, L., Plotnikova, O., Koch, A.W., Sachchidanand, and Zhou, M.M. (2003). A dimeric viral SET domain methyltransferase specific to Lys27 of histone H3. Nat Struct Biol 10, 187-196.

REFERENCES 72

Martineau, A.R., Newton, S.M., Wilkinson, K.A., Kampmann, B., Hall, B.M., Nawroly, N., Packe, G.E., Davidson, R.N., Griffiths, C.J., and Wilkinson, R.J. (2007). Neutrophil-mediated innate immune resistance to mycobacteria. J Clin Invest 117, 1988-1994.

Mattila, J.T., Maiello, P., Sun, T., Via, L.E., and Flynn, J.L. (2015). Granzyme B-expressing neutrophils correlate with bacterial load in granulomas from Mycobacterium tuberculosis-infected cynomolgus macaques. Cell Microbiol 17, 1085-1097.

Mayer-Barber, K.D., Andrade, B.B., Oland, S.D., Amaral, E.P., Barber, D.L., Gonzales, J., Derrick, S.C., Shi, R., Kumar, N.P., Wei, W., et al. (2014). Host-directed therapy of tuberculosis based on interleukin-1 and type I interferon crosstalk. Nature 511, 99-103.

Mazurkiewicz, P., Thomas, J., Thompson, J.A., Liu, M., Arbibe, L., Sansonetti, P., and Holden, D.W. (2008). SpvC is a Salmonella effector with phosphothreonine lyase activity on host mitogen-activated protein kinases. Mol Microbiol 67, 1371-1383.

McShane, H., Pathan, A.A., Sander, C.R., Goonetilleke, N.P., Fletcher, H.A., and Hill, A.V. (2005). Boosting BCG with MVA85A: the first candidate subunit vaccine for tuberculosis in clinical trials. Tuberculosis (Edinb) 85, 47-52.

Meniche, X., Otten, R., Siegrist, M.S., Baer, C.E., Murphy, K.C., Bertozzi, C.R., and Sassetti, C.M. (2014). Subpolar addition of new cell wall is directed by DivIVA in mycobacteria. Proc Natl Acad Sci U S A 111, E3243-3251.

Miller, J.L., Velmurugan, K., Cowan, M.J., and Briken, V. (2010). The type I NADH dehydrogenase of Mycobacterium tuberculosis counters phagosomal NOX2 activity to inhibit TNF-alpha-mediated host cell apoptosis. PLoS Pathog 6, e1000864.

Montagnani, C., Chiappini, E., Galli, L., and de Martino, M. (2014). Vaccine against tuberculosis: what's new? BMC Infect Dis 14 Suppl 1, S2.

Morita, Y.S., Velasquez, R., Taig, E., Waller, R.F., Patterson, J.H., Tull, D., Williams, S.J., Billman-Jacobe, H., and McConville, M.J. (2005). Compartmentalization of lipid biosynthesis in mycobacteria. J Biol Chem 280, 21645-21652.

Munoz-Elias, E.J., and McKinney, J.D. (2005). Mycobacterium tuberculosis isocitrate lyases 1 and 2 are jointly required for in vivo growth and virulence. Nat Med 11, 638-644.

Munoz-Elias, E.J., and McKinney, J.D. (2006). Carbon metabolism of intracellular bacteria. Cell Microbiol 8, 10-22.

Nathan, C. (2002). Inducible nitric oxide synthase in the tuberculous human lung. Am J Respir Crit Care Med 166, 130-131.

Nathan, C. (2006). Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol 6, 173-182.

Nathan, C., and Shiloh, M.U. (2000). Reactive oxygen and nitrogen intermediates in the relationship between mammalian hosts and microbial pathogens. Proc Natl Acad Sci U S A 97, 8841-8848.

REFERENCES 73

Nerlich, A.G., Haas, C.J., Zink, A., Szeimies, U., and Hagedorn, H.G. (1997). Molecular evidence for tuberculosis in an ancient Egyptian mummy. Lancet 350, 1404.

Nigg, E.A. (1997). Nucleocytoplasmic transport: signals, mechanisms and regulation. Nature 386, 779-787.

Oswald, E., Nougayrede, J.P., Taieb, F., and Sugai, M. (2005). Bacterial toxins that modulate host cell-cycle progression. Curr Opin Microbiol 8, 83-91.

Patterson, J.H., Waller, R.F., Jeevarajah, D., Billman-Jacobe, H., and McConville, M.J. (2003). Mannose metabolism is required for mycobacterial growth. Biochem J 372, 77-86.

Pennini, M.E., Liu, Y., Yang, J., Croniger, C.M., Boom, W.H., and Harding, C.V. (2007). CCAAT/enhancer-binding protein beta and delta binding to CIITA promoters is associated with the inhibition of CIITA expression in response to Mycobacterium tuberculosis 19-kDa lipoprotein. J Immunol 179, 6910-6918.

Pennini, M.E., Pai, R.K., Schultz, D.C., Boom, W.H., and Harding, C.V. (2006). Mycobacterium tuberculosis 19-kDa lipoprotein inhibits IFN-gamma-induced chromatin remodeling of MHC2TA by TLR2 and MAPK signaling. J Immunol 176, 4323-4330.

Pennini, M.E., Perrinet, S., Dautry-Varsat, A., and Subtil, A. (2010). Histone methylation by NUE, a novel nuclear effector of the intracellular pathogen Chlamydia trachomatis. PLoS Pathog 6, e1000995.

Persson, Y.A., Blomgran-Julinder, R., Rahman, S., Zheng, L., and Stendahl, O. (2008). Mycobacterium tuberculosis-induced apoptotic neutrophils trigger a pro-inflammatory response in macrophages through release of heat shock protein 72, acting in synergy with the bacteria. Microbes Infect 10, 233-240.

Peterson, J.D., Umayam, L.A., Dickinson, T., Hickey, E.K., and White, O. (2001). The Comprehensive Microbial Resource. Nucleic Acids Res 29, 123-125.

Principi, N., and Esposito, S. (2015). The present and future of tuberculosis vaccinations. Tuberculosis (Edinb) 95, 6-13.

Ramakrishnan, L. (2012). Revisiting the role of the granuloma in tuberculosis. Nat Rev Immunol 12, 352-366.

Ratledge, C., and Dover, L.G. (2000). Iron metabolism in pathogenic bacteria. Annu Rev Microbiol 54, 881-941.

Reljic, R., Clark, S.O., Williams, A., Falero-Diaz, G., Singh, M., Challacombe, S., Marsh, P.D., and Ivanyi, J. (2006). Intranasal IFNgamma extends passive IgA antibody protection of mice against Mycobacterium tuberculosis lung infection. Clin Exp Immunol 143, 467-473.

Richter, E., Weizenegger, M., Rusch-Gerdes, S., and Niemann, S. (2003). Evaluation of genotype MTBC assay for differentiation of clinical Mycobacterium tuberculosis complex isolates. J Clin Microbiol 41, 2672-2675.

REFERENCES 74

Rodriguez, G.M. (2006). Control of iron metabolism in Mycobacterium tuberculosis. Trends Microbiol 14, 320-327.

Rolando, M., Sanulli, S., Rusniok, C., Gomez-Valero, L., Bertholet, C., Sahr, T., Margueron, R., and Buchrieser, C. (2013). Legionella pneumophila effector RomA uniquely modifies host chromatin to repress gene expression and promote intracellular bacterial replication. Cell Host Microbe 13, 395-405.

Rossi, L., Hohn, B., and Tinland, B. (1993). The VirD2 protein of Agrobacterium tumefaciens carries nuclear localization signals important for transfer of T-DNA to plant. Mol Gen Genet 239, 345-353.

Russell, D.G. (2001). Mycobacterium tuberculosis: here today, and here tomorrow. Nat Rev Mol Cell Biol 2, 569-577.

Russell, D.G. (2007). Who puts the tubercle in tuberculosis? Nat Rev Microbiol 5, 39-47.

Russell, D.G. (2011). Mycobacterium tuberculosis and the intimate discourse of a chronic infection. Immunol Rev 240, 252-268.

Schaible, U.E., Collins, H.L., Priem, F., and Kaufmann, S.H. (2002). Correction of the iron overload defect in beta-2-microglobulin knockout mice by lactoferrin abolishes their increased susceptibility to tuberculosis. J Exp Med 196, 1507-1513.

Schaible, U.E., and Kaufmann, S.H. (2004). Iron and microbial infection. Nat Rev Microbiol 2, 946-953.

Schneider, A.G., Abi Abdallah, D.S., Butcher, B.A., and Denkers, E.Y. (2013). Toxoplasma gondii triggers phosphorylation and nuclear translocation of dendritic cell STAT1 while simultaneously blocking IFNgamma-induced STAT1 transcriptional activity. PLoS One 8, e60215.

Schreiner, S., Wimmer, P., Sirma, H., Everett, R.D., Blanchette, P., Groitl, P., and Dobner, T. (2010). Proteasome-dependent degradation of Daxx by the viral E1B-55K protein in human adenovirus-infected cells. J Virol 84, 7029-7038.

Sharma, G., Upadhyay, S., Srilalitha, M., Nandicoori, V.K., and Khosla, S. (2015). The interaction of mycobacterial protein Rv2966c with host chromatin is mediated through non-CpG methylation and histone H3/H4 binding. Nucleic Acids Res 43, 3922-3937.

Sharp, J.D., Cruz, J.W., Raman, S., Inouye, M., Husson, R.N., and Woychik, N.A. (2012). Growth and translation inhibition through sequence-specific RNA binding by Mycobacterium tuberculosis VapC toxin. J Biol Chem 287, 12835-12847.

Shin, D.M., Jeon, B.Y., Lee, H.M., Jin, H.S., Yuk, J.M., Song, C.H., Lee, S.H., Lee, Z.W., Cho, S.N., Kim, J.M., et al. (2010). Mycobacterium tuberculosis eis regulates autophagy, inflammation, and cell death through redox-dependent signaling. PLoS Pathog 6, e1001230.

REFERENCES 75

Silmon de Monerri, N.C., and Kim, K. (2014). Pathogens hijack the epigenome: a new twist on host-pathogen interactions. Am J Pathol 184, 897-911.

Simeone, R., Bobard, A., Lippmann, J., Bitter, W., Majlessi, L., Brosch, R., and Enninga, J. (2012). Phagosomal rupture by Mycobacterium tuberculosis results in toxicity and host cell death. PLoS Pathog 8, e1002507.

Simeone, R., Bottai, D., and Brosch, R. (2009). ESX/type VII secretion systems and their role in host-pathogen interaction. Curr Opin Microbiol 12, 4-10.

Simeone, R., Sayes, F., Song, O., Groschel, M.I., Brodin, P., Brosch, R., and Majlessi, L. (2015). Cytosolic access of Mycobacterium tuberculosis: critical impact of phagosomal acidification control and demonstration of occurrence in vivo. PLoS Pathog 11, e1004650.

Soniat, M., and Chook, Y.M. (2015). Nuclear localization signals for four distinct karyopherin- beta nuclear import systems. Biochem J 468, 353-362.

Soupene, E., Rothschild, J., Kuypers, F.A., and Dean, D. (2012). Eukaryotic protein recruitment into the Chlamydia inclusion: implications for survival and growth. PLoS One 7, e36843.

Stanford, J.L., Rook, G.A., Bahr, G.M., Dowlati, Y., Ganapati, R., Ghazi Saidi, K., Lucas, S., Ramu, G., Torres, P., Minh Ly, H., et al. (1990). Mycobacterium vaccae in immunoprophylaxis and immunotherapy of leprosy and tuberculosis. Vaccine 8, 525-530.

Stebbins, C.E., and Galan, J.E. (2001). Structural mimicry in bacterial virulence. Nature 412, 701-705.

Stegelmann, F., Bastian, M., Swoboda, K., Bhat, R., Kiessler, V., Krensky, A.M., Roellinghoff, M., Modlin, R.L., and Stenger, S. (2005). Coordinate expression of CC chemokine ligand 5, granulysin, and perforin in CD8+ T cells provides a host defense mechanism against Mycobacterium tuberculosis. J Immunol 175, 7474-7483.

Stewart, M. (2007). Molecular mechanism of the nuclear protein import cycle. Nat Rev Mol Cell Biol 8, 195-208.

Stover, C.K., de la Cruz, V.F., Fuerst, T.R., Burlein, J.E., Benson, L.A., Bennett, L.T., Bansal, G.P., Young, J.F., Lee, M.H., Hatfull, G.F., et al. (1991). New use of BCG for recombinant vaccines. Nature 351, 456-460.

Szurek, B., Rossier, O., Hause, G., and Bonas, U. (2002). Type III-dependent translocation of the Xanthomonas AvrBs3 protein into the plant cell. Mol Microbiol 46, 13-23.

Tahlan, K., Wilson, R., Kastrinsky, D.B., Arora, K., Nair, V., Fischer, E., Barnes, S.W., Walker, J.R., Alland, D., Barry, C.E., 3rd, et al. (2012). SQ109 targets MmpL3, a membrane transporter of trehalose monomycolate involved in mycolic acid donation to the cell wall core of Mycobacterium tuberculosis. Antimicrob Agents Chemother 56, 1797-1809.

Taylor, R.C., Cullen, S.P., and Martin, S.J. (2008). Apoptosis: controlled demolition at the cellular level. Nat Rev Mol Cell Biol 9, 231-241.

REFERENCES 76

Teitelbaum, R., Glatman-Freedman, A., Chen, B., Robbins, J.B., Unanue, E., Casadevall, A., and Bloom, B.R. (1998). A mAb recognizing a surface antigen of Mycobacterium tuberculosis enhances host survival. Proc Natl Acad Sci U S A 95, 15688-15693.

Udwadia, Z.F., Amale, R.A., Ajbani, K.K., and Rodrigues, C. (2012). Totally drug-resistant tuberculosis in India. Clin Infect Dis 54, 579-581. van de Vosse, E., Hoeve, M.A., and Ottenhoff, T.H. (2004). Human genetics of intracellular infectious diseases: molecular and cellular immunity against mycobacteria and salmonellae. Lancet Infect Dis 4, 739-749. van Ingen, J. (2013). Diagnosis of nontuberculous mycobacterial infections. Semin Respir Crit Care Med 34, 103-109. van Kessel, J.C., and Hatfull, G.F. (2007). Recombineering in Mycobacterium tuberculosis. Nat Methods 4, 147-152.

Van den Ackerveken, G., Marois, E., and Bonas, U. (1996). Recognition of the bacterial avirulence protein AvrBs3 occurs inside the host plant cell. Cell 87, 1307-1316.

Vanden Berghe, T., Linkermann, A., Jouan-Lanhouet, S., Walczak, H., and Vandenabeele, P. (2014). Regulated necrosis: the expanding network of non-apoptotic cell death pathways. Nat Rev Mol Cell Biol 15, 135-147.

Vankayalapati, R., and Barnes, P.F. (2009). Innate and adaptive immune responses to human Mycobacterium tuberculosis infection. Tuberculosis (Edinb) 89 Suppl 1, S77-80.

Velayati, A.A., Masjedi, M.R., Farnia, P., Tabarsi, P., Ghanavi, J., Ziazarifi, A.H., and Hoffner, S.E. (2009). Emergence of new forms of totally drug-resistant tuberculosis bacilli: super extensively drug-resistant tuberculosis or totally drug-resistant strains in iran. Chest 136, 420-425.

Veyrier, F.J., Dufort, A., and Behr, M.A. (2011). The rise and fall of the Mycobacterium tuberculosis genome. Trends Microbiol 19, 156-161.

Vilaplana, C., Ruiz-Manzano, J., Gil, O., Cuchillo, F., Montane, E., Singh, M., Spallek, R., Ausina, V., and Cardona, P.J. (2008). The tuberculin skin test increases the responses measured by T cell interferon-gamma release assays. Scand J Immunol 67, 610-617.

Wagner, D., Maser, J., Lai, B., Cai, Z., Barry, C.E., 3rd, Honer Zu Bentrup, K., Russell, D.G., and Bermudez, L.E. (2005). Elemental analysis of Mycobacterium avium-, Mycobacterium tuberculosis-, and Mycobacterium smegmatis-containing phagosomes indicates pathogen-induced microenvironments within the host cell's endosomal system. J Immunol 174, 1491-1500.

Walzl, G., Ronacher, K., Hanekom, W., Scriba, T.J., and Zumla, A. (2011). Immunological biomarkers of tuberculosis. Nat Rev Immunol 11, 343-354.

REFERENCES 77

Ward, S.K., Abomoelak, B., Hoye, E.A., Steinberg, H., and Talaat, A.M. (2010). CtpV: a putative copper exporter required for full virulence of Mycobacterium tuberculosis. Mol Microbiol 77, 1096-1110.

Weerdenburg, E.M., Peters, P.J., and van der Wel, N.N. (2010). How do mycobacteria activate CD8+ T cells? Trends Microbiol 18, 1-10.

Welin, A., Winberg, M.E., Abdalla, H., Sarndahl, E., Rasmusson, B., Stendahl, O., and Lerm, M. (2008). Incorporation of Mycobacterium tuberculosis lipoarabinomannan into macrophage membrane rafts is a prerequisite for the phagosomal maturation block. Infect Immun 76, 2882-2887.

White, C., Lee, J., Kambe, T., Fritsche, K., and Petris, M.J. (2009). A role for the ATP7A copper-transporting ATPase in macrophage bactericidal activity. J Biol Chem 284, 33949- 33956.

WHO (2014). Global Tuberculosis Report (World Health Organization).

Williams, A., Reljic, R., Naylor, I., Clark, S.O., Falero-Diaz, G., Singh, M., Challacombe, S., Marsh, P.D., and Ivanyi, J. (2004). Passive protection with immunoglobulin A antibodies against tuberculous early infection of the lungs. Immunology 111, 328-333.

Wolschendorf, F., Ackart, D., Shrestha, T.B., Hascall-Dove, L., Nolan, S., Lamichhane, G., Wang, Y., Bossmann, S.H., Basaraba, R.J., and Niederweis, M. (2011). Copper resistance is essential for virulence of Mycobacterium tuberculosis. Proc Natl Acad Sci U S A 108, 1621- 1626.

Wong, D., Bach, H., Sun, J., Hmama, Z., and Av-Gay, Y. (2011). Mycobacterium tuberculosis protein tyrosine phosphatase (PtpA) excludes host vacuolar-H+-ATPase to inhibit phagosome acidification. Proc Natl Acad Sci U S A 108, 19371-19376.

Yang, X.Y., Chen, Q.F., Li, Y.P., and Wu, S.M. (2011). Mycobacterium vaccae as adjuvant therapy to anti-tuberculosis chemotherapy in never-treated tuberculosis patients: a meta-analysis. PLoS One 6, e23826.

Yang, Y., and Gabriel, D.W. (1995). Xanthomonas avirulence/pathogenicity gene family encodes functional plant nuclear targeting signals. Mol Plant Microbe Interact 8, 627-631.

Yang, B., Zhu, W., Johnson, L.B., and White, F.F. (2000). The virulence factor AvrXa7 of Xanthomonas oryzae pv. oryzae is a type III secretion pathway-dependent nuclear- localized double-stranded DNA-binding protein. Proc Natl Acad Sci U S A 97, 9807-9812.

Young, D.B., Perkins, M.D., Duncan, K., and Barry, C.E., 3rd (2008). Confronting the scientific obstacles to global control of tuberculosis. J Clin Invest 118, 1255-1265.

Zhang, M., Chen, J.M., Sala, C., Rybniker, J., Dhar, N., and Cole, S.T. (2014). EspI regulates the ESX-1 secretion system in response to ATP levels in Mycobacterium tuberculosis. Mol Microbiol 93, 1057-1065.

REFERENCES 78

Zhang, Y.J., Reddy, M.C., Ioerger, T.R., Rothchild, A.C., Dartois, V., Schuster, B.M., Trauner, A., Wallis, D., Galaviz, S., Huttenhower, C., et al. (2013). Tryptophan biosynthesis protects mycobacteria from CD4 T-cell-mediated killing. Cell 155, 1296-1308.

Zou, H., Zhao, W., Zhang, X., Han, Y., Zou, L., and Chen, G. (2010). Identification of an avirulence gene, avrxa5, from the rice pathogen Xanthomonas oryzae pv. oryzae. Sci China Life Sci 53, 1440-1449.

Zuber, B., Chami, M., Houssin, C., Dubochet, J., Griffiths, G., and Daffe, M. (2008). Direct visualization of the outer membrane of mycobacteria and corynebacteria in their native state. J Bacteriol 190, 5672-5680.

Zumla, A., Nahid, P., and Cole, S.T. (2013). Advances in the development of new tuberculosis drugs and treatment regimens. Nat Rev Drug Discov 12, 388-404.

Zurawski, D.V., Mumy, K.L., Faherty, C.S., McCormick, B.A., and Maurelli, A.T. (2009). Shigella flexneri type III secretion system effectors OspB and OspF target the nucleus to downregulate the host inflammatory response via interactions with retinoblastoma protein. Mol Microbiol 71, 350-368.

APPENDIX A

PUBLISHED WORK: Mycobacterium tuberculosis Proteins Involved in Mycolic Acid Synthesis and Transport Localize Dynamically to the Old Growing Pole and Septum This article was published after two years of thesis work (2011-13) under supervision of Dr. Didier ZERBIB within the laboratory of Dr. Mamadou DAFFE’s group at IPBS-CNRS. We investigated the localization of Mtb enzymes belonging to the FAS-II complexes that are involved in mycolic acids (MA) biosynthesis. Using a combination of molecular genetic, fluorescence- and video-microscopy approaches, we found that the Mtb transmembrane transport protein 3 (MmpL3), which is the MA transporter and phospholipid-interacting protein Wag31, is essential for the septum formation in M. smegmatis. In order to compare the dynamics between the FAS-II proteins, MmpL3 and Wag31, we constructed bacteria expressing these genes fused to fluorescent probes: i) MmpL3 and the FAS-II proteins (i.e. MabA, InhA, KasA and KasB) coupled to the green fluorescent protein (GFP) at their C- terminus under the control of the acetamidase promoter (PamiE) of the inducible vector pLAM 12, and ii) the Wag31 fused to mCherry at its N-terminus under the control of the constitutive promoter pHSP60 of the integrative vector pMV361. Based on this approach, our study clearly demonstrated that FAS-II proteins, MmpL3 and Wag31, are co-localized at cell poles, suggesting that these macromolecules are produced and conveniently exported to the same cellular region to contribute MA biosynthesis.

Mycobacterium tuberculosis Proteins Involved in Mycolic Acid Synthesis and Transport Localize Dynamically to the Old Growing Pole and Septum

Cle´ment Carel1,2., Kanjana Nukdee1,2., Sylvain Cantaloube1,2,Me´lanie Bonne1,2, Cheikh T. Diagne1,2, Franc¸oise Laval1,2, Mamadou Daffe´ 1,2, Didier Zerbib1,2* 1 Centre National de la Recherche Scientifique, Institut de Pharmacologie et de Biologie Structurale, Toulouse, France, 2 Universite´ de Toulouse, Universite´ Paul Sabatier, Toulouse, France

Abstract Understanding the mechanism that controls space-time coordination of elongation and division of Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis (TB), is critical for fighting the tubercle bacillus. Most of the numerous enzymes involved in the synthesis of Mycolic acid - Arabinogalactan-Peptidoglycan complex (MAPc) in the cell wall are essential in vivo. Using a dynamic approach, we localized Mtb enzymes belonging to the fatty acid synthase-II (FAS-II) complexes and involved in mycolic acid (MA) biosynthesis in a mycobacterial model of Mtb: M. smegmatis. Results also showed that the MA transporter MmpL3 was present in the mycobacterial envelope and was specifically and dynamically accumulated at the poles and septa during bacterial growth. This localization was due to its C-terminal domain. Moreover, the FAS-II enzymes were co-localized at the poles and septum with Wag31, the protein responsible for the polar localization of mycobacterial peptidoglycan biosynthesis. The dynamic localization of FAS-II and of the MA transporter with Wag31, at the old-growing poles and at the septum suggests that the main components of the mycomembrane may potentially be synthesized at these precise foci. This finding highlights a major difference between mycobacteria and other rod-shaped bacteria studied to date. Based on the already known polar activities of envelope biosynthesis in mycobacteria, we propose the existence of complex polar machinery devoted to the biogenesis of the entire envelope. As a result, the mycobacterial pole would represent the Achilles’ heel of the bacillus at all its growing stages.

Citation: Carel C, Nukdee K, Cantaloube S, Bonne M, Diagne CT, et al. (2014) Mycobacterium tuberculosis Proteins Involved in Mycolic Acid Synthesis and Transport Localize Dynamically to the Old Growing Pole and Septum. PLoS ONE 9(5): e97148. doi:10.1371/journal.pone.0097148 Editor: Laurent Kremer, Universite´ de Montpellier 2, France Received December 16, 2013; Accepted April 15, 2014; Published May 9, 2014 Copyright: ß 2014 Carel et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by the Centre National de la Recherche Scientifique (CNRS). CC was supported by the French ministry of research; KN was supported by a fellowship from the Thailand government. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] . These authors contributed equally to this work.

Introduction In Mtb, two Fatty-Acid-Synthase systems (FAS), FAS-I [6] and FAS-II [7], are devoted to the synthesis of common fatty acids and The mycobacterial cell envelope is of an exceptional complex- of specific, long-chain, a-acylated, b-hydroxylated fatty acids, ity. One of the major challenges for Mycobacterium tuberculosis (Mtb) namely mycolic acids (MA) [8,9]. Mycolic acids constitute the is probably achieving the synthesis of this thick multilayer barrier major lipid component of the envelope and form the external [1,2]. The Mtb envelope is partly responsible for its innate mycomembrane [10,11]. They are either inserted in the resistance to antibiotics and plays a major role in both its virulence mycomembrane as trehalose dimycolates and monomycolates and persistence. Tuberculosis (TB) reappeared in the 1990s and (TDM and TMM) or linked covalently to the underlying remains one of the main causes of lethality worldwide. The arabinogalactan (AG), itself bound to peptidoglycan (PG) to form principal challenge is to fight the increasing number of multi-drug the MAPc [12,13]. Mycolic acids stem from the condensation of a resistant (MDR) Mtb clinical isolates. The alternate periods of medium chain-length fatty acid (C24-C26) produced by FAS-I with growth and dormancy of Mtb during infection [3] render the a long meromycolic chain (up to C60) produced by FAS-II (Figure discovery of new effective antibiotics difficult. Mtb division, as well S1) [14,15]. FAS-II is comprised of specialized and interconnected as the regulation of its cell cycle and the maintenance of cell wall protein complexes [16,17,18] (Figure S2). The MA biosynthetic homeostasis, are nonetheless weak points of the bacillus. Under- pathway is assumed to be cytoplasmic, a notion recently reinforced standing the mechanisms that guarantee spatial and temporal by the discovery of a MA transporter in mycobacteria: the coordination of Mtb envelope biogenesis during the cell cycle is a potential trans-membrane protein MmpL3 [19,20,21]. We crucial step toward deciphering the role of this major player of Mtb hypothesized that at least part of the FAS-II multi-protein pathogenesis [4,5]. complexes may exist as stable entities and could be involved in the biogenesis of MAPc during bacterial growth.

PLOS ONE | www.plosone.org 1 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

The cell wall elongation process of actinobacteria [4] is different Results from that of other rod-shaped models such as Escherichia coli or Bacillus subtilis [22]. In the latters, nascent PG is made all along the FAS-II complex components are located at one bacterial side of the bacterium during elongation and at the septum during pole the division process [23] whereas in the former, the lateral PG Mtb genes encoding the FAS-II reductases MabA and InhA, derives from polar biosynthesis during elongation [24]. Among and the condensing enzymes KasA and KasB, were merged to the mycobacteria, the polar localization of PG synthesis is mainly due C-terminal end of gfp and expressed under the control of the PamiE to the phospholipid-interacting protein Wag31, the ortholog of the inducible promoter in the non-pathogenic bacterium model M. B. subtilis DivIVA protein [25,26,27]. Wag31, by targeting smegmatis mc2155 (Msm). In a first series of experiments, large field negatively-curved membranes [28], is responsible for the polar fluorescence microscopy images, hereafter called static images, localization and the septal re-localization of several PG biosyn- were acquired six hours after induction. GFP alone emitted a thesis proteins [29,30]. Wag31 is mainly located at the ‘‘old pole’’ bright and diffuse fluorescence throughout the cytoplasm (Figure 1), which is the active locus among mycobacteria and other (Figure 2A). Conversely, the reductase fusions yielded intense foci actinobacteria where new molecular material of the lateral cell wall is located mainly at one bacterial pole (Figure 2A). In the strains added [31,32]. The ‘‘septal pole’’ (Figure 1), which represents the expressing either GFP-KasB or GFP-KasA, polar foci were also future pole and the developing septum, does not participate in detectable although together with a diffuse fluorescence in the lateral PG biosynthesis [33]. cytoplasm (Figure 2A). To quantify foci formation, a polar index We demonstrate herein that reductases from the FAS-II core as (PI) was calculated and defined as the ratio of the number of well as the MA transporter MmpL3 are located at the active bacteria containing at least one polar focus reported to the total mycobacterial old poles and at the position of the septum prior to number of fluorescent bacteria (N). Over one thousand bacteria division. The condensing enzymes form polar foci but are also able expressing each fusion were analyzed on several replicates. Two to diffuse in the cytoplasm. FAS-II localization is correlated with groups were found (Figure 2B). The first group, e.g. MabA the dynamics of cell division, with the localization of Wag31 and (89.47%61.20, N = 1100) and InhA (63.86%61.56, N = 2000), with the dynamics of the MA transporter MmpL3. These results had a high PI while the second group, e.g. KasA (13.34%61.40, describe the first observation of the localization of sophisticated N = 900) and KasB (21.02%62.40, N = 1700), had a lower PI. machinery for fatty acid biosynthesis. In addition, they reveal the Due to the inaccuracy of quantifying and comparing static images existence of a possible link between two central and essential from different snapshots, the total amount of fluorescence in live processes: bacterial division and whole envelope biogenesis. bacterial culture was assessed by fluorometry as a function of Mycolic acids may be synthesized at the active poles and septa induction time. This enabled to verify whether PI was correlated in order to be transported, probably by MmpL3, to the external or not with fusion expression levels. Eight hours after induction (Figure 2C), the strain containing GFP alone (PI = 0%) emitted the mycomembrane. maximum fluorescence (280 RFU). MabA (PI = 89%) and KasB (PI = 21%) yielded intermediate fluorescence levels (45 RFU and 35 RFU, respectively) whereas KasA (PI = 13%) and InhA (PI = 63%) yielded very low levels (below 10 RFU). The absence of correlation between fluorescence levels and PI values (as compared between Figure 2B and Figure 2C) was further confirmed by Western blotting experiments performed on the same cultures (Figure 2D). Western blotting results were perfectly correlated with fluorometry results (see comparison between Figure 2D and Figure 2C) and thus were not correlated with PI values, indicating that foci formation was not due to protein overexpression but rather due to preferential localization of FAS-II at the pole.

Dynamics of polar localization of FAS-II proteins In order to analyze foci formation and localization and avoid potential problems due to protein accumulation, a dynamic analysis was implemented. The goal was to observe and follow the very early stages of foci formation during bacterial division, starting just after the beginning of the induction of GFP-fusion expression with a low amount of inducer (0.02% acetamide). Time-lapse microscopy experiments were designed using a recent and efficient agar pad microscopy technique [34]. In the GFP- producing strain (Movie S1), the signal was bright and diffuse in which no foci were observed. In the GFP-MabA producing strain (Movie S2), at the beginning of induction, the first foci (Figure 3, Figure 1. Schematic representation of mycobacterial polar foci a, 3 h; and b, 5 h), were observed at the septal poles of the growth and establishment of bacterial poles. Two division cycles mother bacterium. These foci appeared at mid-cell at the position (D1 and D2) are presented. The active biosynthesis of lateral of the next division septum. These types of foci were often peptidoglycan at the poles and biosynthesis of septal peptidoglycan apparently split in two after the end of the division process at the future pole (septal poles) are symbolized with curved arrows. The 1-2 1-2 old pole (in red) and the new pole (in blue); are both represented and (Figure 3, foci a , 5 h; b , 9 h). Thereafter, they became refer to the first division event (D1). established at the new poles of the first pair of daughter bacteria doi:10.1371/journal.pone.0097148.g001 (Figure 3, foci a, b, c, 9 h to 13 h). At the end of induction the

PLOS ONE | www.plosone.org 2 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

Figure 2. Polar localization of FAS-II proteins. (A) Enlarged images of wide-field microscopy experiments on GFP-fusion expressing bacteria. Bright field images (BF; leftmost column) together with GFP fluorescence images (GFP, middle column) of Msm expressing GFP fusions with MabA, InhA, KasA or KasB, or GFP alone (Ø) were acquired after 6 hours of induction. The merged images (right columns) allowed visualizing the polar localization of the fusions. Total optical magnification: 630 X. Scale bar: 5 mm. (B) Scatter dot plot representation of polar indexes of Msm strains expressing each GFP fusion after 6 hours of induction. Data were collected from wide-field microscopy images and subsequently expressed and evaluated as described in Materials and Methods. Statistical t-tests were performed to assess the differences between the polar indices of GFP-MabA (black circles) and GFP-InhA (black triangles) and between GFP-KasA (black squares) and GFP-KasB (open diamonds). Each data point corresponds to an individual experiment. Each series of data points for a given fusion represents 500 to1500 bacteria analyzed. The p values of indicated unpaired t- tests are symbolized by asterisks (***, p,0.0001), (*, p = 0.0264). (C) Total fluorescence of Msm cultures expressing GFP fusions. Experimental values are represented as means 6 SEM. Fluorescence intensities, expressed in RFU (as defined in the text and in the Materials and Methods) of liquid cultures of Msm expressing either GFP alone (Ø), GFP-MabA (black circles), GFP-KasB (open diamonds), GFP-KasA (black squares) or GFP-InhA (black triangles) were plotted against the length of induction in hours (h). The curve fit was obtained by using a second order polynomial nonlinear regression (quadratic) calculated with GraphPad Prism 5.0 software. (D) Western Blot analysis of GFP-fusions. Total protein content of Msm culture samples expressing either GFP alone or the GFP-fusions with the indicated proteins was analyzed by Western blotting using anti-GFP antibodies. The blots were performed 2 hours (left lanes), 4 hours (middle lanes) or 6 hours (right lanes) after acetamide induction of the same cultures depicted in panel C. The molecular weights of the relevant marker bands are indicated in kDa. doi:10.1371/journal.pone.0097148.g002 images resembled the static images from the first experiments pole structure in order to ensure elongation of the lateral (Figure 2A) where this protein was present at only one pole, mycomembrane. leading to conclude that these foci correspond to dynamic protein complexes. The same results were obtained with the InhA fusions FAS-II proteins are localized with the old pole marker (Movie S3), displaying foci which were first visible at the septal protein Wag31 poles and thereafter establishing at one preferential pole. Because Since FAS-II localization appeared to follow the reported of the lower level of expression of the KasA fusion (Movie S4), it localization of the polar protein Wag31 [29,35], a recognized ‘‘old was more difficult to follow foci dynamics at the beginning of pole marker’’, Wag31-FAS-II colocalization was therefore studied. induction although in the micro-colonies, GFP-KasA foci were For purposes of clarity, the word colocalization is used in this study also first visible at the septal poles and then appeared very rapidly to describe the localization of two proteins at the same focus but at the poles. Finally, the KasB fusion (Movie S5) was diffuse in the without implying the existence of any molecular interaction. A cytoplasm at the early stage of induction and the foci were only strain was constructed bearing a single but ectopic chromosomal visible at one pole at the end of the experiment. There was no copy of a mCherry-Wag31 fusion under the control of the typical envelope labeling by FAS-II fusions observed at any time, constitutive promoter pHSP60 of the integrative vector pMV361. indicating that the FAS-II protein tested are probably cytoplasmic. This type of construct, using the pHSP60 promoter, has been We thus hypothesized that these proteins ultimately participate in successfully used previously to study Wag31 localization [36]. When this strain was observed in fluorescence microscopy (Figure

PLOS ONE | www.plosone.org 3 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

bacteria expressing each fusion (N) stemming from several individual slides (n.6). The polar indices of MabA (93.9360.59 N = 800), KasA (23.8764.56 N = 500) and KasB (13.4961.43 N = 500) were not statistically different from the previously observed indices (see comparison between Figure 2B and 4B) whereas the PI of GFP-InhA (35.9962.14 N = 700) was slightly reduced. While there was no dramatic effect of the presence of the mCherry-Wag31 fusion on the percentage of localization of the FAS-II proteins, there was the presence of colocalization of both Wag31 and FAS-II fusions foci. To quantify this phenomenon on a large number of bacteria a colocalization index was defined as being the percentage of bacteria exhibiting Wag31-GFP colocalization relative to the number of GFP-mCherry-positive bacteria. Colocalization indices (Figure 4C) were very high for all FAS-II proteins: MabA (83.31%63.27, N = 800), InhA (93.93%62.60, N = 700), KasA (89.71%63.37, N = 500) and KasB (89.52%63.60, N = 500). In order to evaluate the distribution of foci within individual bacteria, a more in-depth analysis was performed on a subset of bacteria by scanning more than 100 organisms of each type. On each scan (Figure 4D), the maximum fluorescence was plotted against the normalized length of the bacteria. A dot represented on a graph can refer to a focus (at the pole) or simply to the maximum of diffuse fluorescence along the cell. Localization of the reductases and colocalization with Wag31 at the old poles were clearly confirmed (Figure 4D). Furthermore, the maximum florescence emitted by KasA and KasB, which displayed the lowest polar indices, were clearly preferentially observed in proximity of the old poles along with Wag31, thus suggesting an actual preference for this localization (Figure 4D). The distribution of the polar foci clouds represented only about 20% of total cell size, i.e. less than 1 mm, probably due to light diffusion. This distribution was similar for FAS-II proteins and for Wag31, already known to be localized close to the polar membranes. In contrast, the maximum accumulation of GFP alone was never found at the old pole suggesting a polar exclusion of GFP alone. These analyses clearly demonstrate that the FAS-II proteins were preferentially located at the old pole with Wag31. Since Wag31 is acknowledged to be an old pole marker, we can conclude that the polar FAS-II proteins were located at the active Figure 3. Dynamics of GFP-MabA localization. Images were growing ‘‘old pole’’ of the bacteria. extracted from the movie presented as movie S2. The time interval after the starting point of the movie is indicated in hour (h). Merge bright FAS-II localization dynamics follow Wag31 dynamics with field and GFP channel images (BF-GFP, left column) and GFP channel images (GFP, right column) were extracted in order to follow the no molecular interaction dynamics of representative GFP-MabA foci. Two bacteria and their In order to compare FAS-II and Wag31 dynamics, time-lapse daughter cells were schematized (far right). Representative foci (in experiments were performed on bacteria expressing both types of green) are labeled (a to f) and followed, when possible, in the daughter fusions. The expression of mCherry-Wag31 being constitutive, cells. Superscript numbering represents potential foci splitting. White arrows indicate the localization of the main foci on GFP images. polar Wag31 foci were visible in the mother cells at the beginning Magnification and scale are identical to those indicated in Figure 2A. of the experiments. As observed on static images (Figure 4A), GFP doi:10.1371/journal.pone.0097148.g003 alone was diffuse in the cytoplasm (Movie S6). In this movie, the presence of Wag31 at the old pole and its re-localization to the S3), the mCherry fusion was more visible at one pole as also septum were clearly visible (Movie S6). In the GFP-FAS-II commonly observed by others [36,37]. This pole has been shown producing strains (Movies S7-S10), bacterial division was observed to represent the old pole (Figure 1) [29]. The fusions were together with Wag31 and FAS-II dynamics. The dynamics of GFP introduced into the mCherry-Wag31 producing strain after which foci were analyzed in detail for the GFP-MabA fusion (Movie S7 the fluorescence of both GFP and mCherry was analyzed as and Figure 5). Prior to the detection of GFP fluorescence, performed above on static images. As observed in the absence of mCherry-Wag31 mainly observed at the old poles then, as the mCherry-Wag31 (Figure 2A), GFP remained diffuse whereas elongation progressed (1 h, bacteria 1 and 2), Wag31 was relocated to the septa as previously observed [29,30]. The GFP GFP-MabA and GFP-InhA yielded intense polar foci, mainly at fusions were first detected at these new poles and slightly at the old one pole (Figure 4A). GFP-KasA and GFP-KasB still appeared poles (2 h, bacteria 11 and 12). Progressively thereafter, establish- diffuse with a lower extend of polar spotting (Figure 4A). The ment was observed at the old poles (see after 5 h, bacteria 11.1). At overall localization pattern of the FAS-II fusions was comparable the end of the experiments, GFP fluorescence was primarily to that of the previous fusions (Figure 2A). The statistical relevance observed at the old poles together with the brighter Wag31 of the localization (Figure 4B) was analyzed as above on individual fluorescence. Bacterium no. 2 presented here (Figure 5) displayed

PLOS ONE | www.plosone.org 4 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

Figure 4. Polar colocalization of FAS-II proteins with Wag31. (A) Enlarged images of wide-field microscopy experiments on GFP-fusion and

PLOS ONE | www.plosone.org 5 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

mCherry-Wag31 fusion-expressing bacteria. GFP fluorescence images (GFP; leftmost column) together with mCherry fluorescence images (Che, middle column) of Msm::mCherry-wag31 expressing GFP fusions with MabA, InhA, KasA, KasB, or GFP alone (Ø) were acquired after 6 hours of induction. The merged images (right columns) allowed visualizing the polar colocalization of the fusions. Magnification and scale bars are identical to those indicated in Figureoˆ 2A. (B) Bar graph representation of GFP polar indices of Msm::mCherry-wag31 strains expressing each GFP fusion after 6 hours of induction. Experimental values are represented as mean 6 SD. Data were processes as described in Figureoˆ 2B. Statistical t-tests were performed to determine the differences between the polar indices of GFP-InhA (dark grey bar) and GFP-KasA (light grey bar) and of GFP-KasA and GFP-KasB (white bar). The p values of indicated unpaired t-tests are symbolized by asterisks (*, p = 0.0271 for InhA-KasA), (*, p = 0.0186 for KasA-KasB). (C) Bar graph representation of GFP-mCherry colocalization indices of Msm::mCherry-wag31 strains expressing each GFP fusion after 6 hours of induction. Experimental values are represented as means 6 SD. Data were processed as in panel B; no significant differences were found between colocalization indices. (D) Analysis of maximal mCherry-Wag31 (Che-Wag31, left column) and GFP-FAS-II fusion (GFP fusion, right column) fluorescence position within individual Msm::mCherry-wag31 bacteria expressing each GFP fusion after 6 hours of induction. Each type of bacterium was scanned on both channels. Each dot represents the position of maximum fluorescence of the scans expressed in % of the highest values. Each dot was plotted against its position in the bacterium with the length of bacteria reported to 1. The names of the FAS-II proteins fused with GFP are indicated. Ø refers to the Msm::mCherry-wag31 strain containing GFP alone. doi:10.1371/journal.pone.0097148.g004

exactly the same dynamics with a delayed appearance at the beginning of the induction. Because of the lower expression of the InhA fusion, the dynamics of this fusion was more difficult to follow although the majority of GFP foci were established with the brighter Wag31 foci (Movie S8). Finally, the KasA fusion (Movie S9) and the KasB fusion (Movie S10) were mainly diffuse in the cytoplasm as already observed on static images and foci were observed only at the end of the experiments. When GFP fusion expression was induced, new proteins entered the system via the pole in construction (septal pole) and then established progressively to the old pole with Wag31. To ascertain whether dynamic FAS-II-Wag31 colocalization was due, or not, to molecular interactions, in vitro Co-immunoprecipitation (Co-IP) experiments were performed with Wag31 against FAS-II proteins. In vitro 35S-labeled HA-tagged FAS-II proteins (h-proteins) bound to magnetic beads coated with anti- HA antibodies were used to trap labeled c-Myc-tagged Wag31 protein (c-Wag31). No Co-IP bands were observed, except with Wag31 itself (Figure S4A). When the Co-IP experiments were performed in the reverse direction (Figure S4B), with h-Wag31 against c-FAS-II proteins, only non-specific binding was observed with FAS-II proteins (Figure S4B, IPØ). This non-specific binding was not observed when known negative controls were used (Figure S4C) including human c-Lamin (c-Lam) or mycobacterial Nat protein (Arylamine N acetyltransferase, [38]). This indicates that FAS-II proteins localize to the old pole and follow Wag31 dynamics without any direct molecular interaction with Wag31.

In vivo activity of GFP-fusions FAS-II proteins are organized in specialized complexes (Figure S2) interconnected with one another [16,17,18]. We showed above that representative members of theses complexes were located at the old poles during elongation and at the septa prior to division. It was thus tempting to postulate that the biosynthesis of mycolic acids may occur at these precise foci. To date, there is no known biochemical clue for locating MA biosynthesis in vivo. Thus, to address this question, we chose to analyze GFP-fusion activity in vivo. The kasB gene is the only FAS-II gene that is not essential in both M. marinum [39] and Mtb [40]. In the absence of KasB, which is part of the E2-FAS-II complex [17] and is responsible for the Figure 5. Dynamic of GFP-MabA/Wag31 colocalization. Images final step of meromycolate elongation, the MA are shortened by were extracted from the movie presented as movie S7. The time interval two to four carbons [40]. For the present purposes, we constructed after the starting point of the movie is indicated in hours (h). Images a kasB-KO mutant of Msm (Figure S5) that was viable, indicating were extracted from three channels (bright field, BF, left column; GFP, that kasB is also ‘‘non-essential’’ in Msm. In this mutant, the alpha- middle column; mCherry, Che, right column) in order to follow the and epoxy-MA were shortened by four carbons (Figure 6A, 6B) dynamics of representative GFP-MabA and mCherry-Wag31 foci. Two bacteria, numbered 1 and 2, and their daughter cells, identified by whereas the alpha’-MA, which appeared as KasB independent, superscript numbering are schematized on the right. Magnification: 630 were not affected (Figure S6). The mutant phenotype was reverted X. Scale bar: 5 mm. by complementation with the wild type kasB gene and, more doi:10.1371/journal.pone.0097148.g005 importantly, by the GFP-KasB fusion (Figure 6A). The localization

PLOS ONE | www.plosone.org 6 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

Figure 6. Analysis of GFP-KasB and GFP-InhA in vivo activities. (A) MALDI-TOF mass spectra of a-mycolic acid methyl esters extracted from wild type (Msm mc2155, WT), mutant (Msm mc2155 DkasB, DB) and complemented (Msm mc2155 DkasB/kasB, DB/B; Msm mc2155 DkasB/gfp-kasB, DB/ gB) strains. The accurate masses (m/z), together with the length of the odd carbon-numbered mycolates are indicated. (B) TLC analysis of mycolic acid methyl esters extracted from the strains analyzed and described in Figureoˆ 6A. The migration position of the a-mycolates (alpha), the a’-mycolates (alpha’) and of the epoxy-mycolates (epoxy) are indicated. (C) Growth analysis of colonies of Msm mc2155 derivatives spotted onto agar plates containing increasing amounts of isoniazid (INH). Bacterial colonies were photographed either under white light (first three rows) or under blue light illumination allowing to reveal GFP fluorescence (last row). Bacteria containing the vector alone (Ø) or expressing the Mtb inhA gene, or the GFP fusion with InhA are represented. (D) Two colonies of the thermosensitive strain Msm inhAts containing either the pLAM12 vector (L12), the pLAM12::GFP-InhA plasmid (L12 GFP-InhA) or the pLAM12::InhA plasmid (L12 InhA) were grown at 30uC (left) or 42uC (right) on agar plates containing acetamide (0.2%). doi:10.1371/journal.pone.0097148.g006 pattern of GFP-KasB in the complemented mutant was identical in our system, to test the GFP-fusion complementation of to that observed with the wild type (data not shown). The GFP- interrupted genes. MabA and kasA genes exist within operons KasB fusion was active in vivo. whereas inhA lies at the end of an operon. Nevertheless, we had Given that the other FAS-II genes (mabA, inhA, kasA) are some inkling with regard to InhA. InhA is the primary INH essential [41,42] and part of operons, it was not easily achievable, (isoniazid, also known as isonicotinylhydrazine) target and its

PLOS ONE | www.plosone.org 7 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes overproduction induces an increase in the level of INH resistance by trapping the INH-NAD adducts in its catalytic pocket, thus reducing INH concentration within the bacteria [43]. We first verified that the overproduction of the wild type inhA gene in our strain indeed increased INH resistance (18.75 fold; 75 mg/ml, Figure 6C) whereas the production of GFP alone did not affect this level (4 mg/ml). A clear increase in resistance threshold, (12.5 fold; 50 mg/ml), was also observed when GFP-InhA was produced. The GFP-InhA fusion was likely able to bind the INH-NAD adducts in vivo thereby inducing an increase in cell resistance to INH. Finally, to demonstrate that the GFP-InhA fusion was active, the complementation of an inhA thermosensitive mutant Msm strain (mc22359; inhAts; [44]) was tested (Figure 6D). At 42uC, in the presence of acetamide, the inhAts strain containing the void vector pLAM12 was unable to grow. In contrast, when either the wild type InhA protein or, more interestingly, the GFP-InhA fusion was expressed, the complemented strains were thermo-resistant indicating that the fusion was also able to complement the InhA deficiency. These results demonstrate that the GFP-InhA as well as the GFP-KasB fusion may be active in vivo while forming foci.

Localization of the Mycolic acid export machinery Recently, the potential membrane protein MmpL3 has been identified in Mtb as an MA transporter [19,45]. It is involved in the transport of TMM to the external mycomembrane. Because of the localization of major components of FAS-II observed herein, we hypothesized that MA may be synthesized at these loci to be subsequently transported at the actively growing envelope regions. GFP was therefore merged to the MmpL3 C-terminal given that this domain was predicted to be intra-cytoplasmic (Toppred2, [46], http://bioweb.pasteur.fr/seqanal/interfaces/toppred.html). MmpL3, a RND family protein [47], was predicted to be organized in twelve transmembrane helices. The localization of this fusion was analyzed as above. Upon performing time-lapse experiments with MmpL3-GFP alone (Movie S11), or in the presence of mCherry-Wag31 (Figure 7, extracted from Movie S12), a very clear envelope staining was observed together with preferential labeling of one pole and a very bright and flashing Figure 7. Dynamic localization of MmpL3. Images were extracted labeling of the septum, likely due to the presence of the double from the movie presented as movie S12. The time interval after the membrane at this position. The first conclusion was that MmpL3 starting point of the movie is indicated in hours (h). Images were might indeed be in the membrane or at least, in the envelope. Its extracted from three channels (bright field, BF, left column; GFP, middle column; GFP and mCherry, GFP-Che, right column). The dynamics of the C-terminal domain, fused with the GFP, could be intra- localization of the GFP fusions was followed in Msm::mCherry-wag31 cytoplasmic because GFP is not fluorescent when it is located in bacteria expressing the MmpL3-GFP fusions. The presence of MmpL3 in another compartment [48]. As a localization control, a GFP fusion septa is indicated by white arrows in the right column. The ‘‘croissant’’- of the first ten transmembrane helices of MmpL3 (MmpL3-N10) shaped accumulation of MmpL3 at the poles is indicated by white was constructed and its localization analyzed in vivo as described arrows in the middle column. The position of mCherry-Wag31 at the above. Upon performing time-lapse experiments with MmpL3- poles and septa is clearly visible. Magnification: 630 X. Scale bar: 5 mm). doi:10.1371/journal.pone.0097148.g007 N10-GFP in the presence of mCherry-Wag31 (Figure 8, extracted from Movie S13), the preferential MmpL3 polar localization was not observed. MmpL3-N10, which remained in the membrane, or membrane curvature. This polar localization of the MA at least in the envelope, was no more accumulated at the pole. The transporter appeared to be due to a MmpL3 domain located polar accumulation (P) of MmpL3 was quantified as described in between the end of helix 10 and the C-terminal end of the protein. Materials and Methods and compared to that of MmpL3-N10 (Figure 8B). MmpL3-N10 localization was not polar (P = 1.16 6 Discussion 0.15 N = 113) whereas MmpL3 was enriched by more than 5 fold The main objective of the present study was to determine at the pole (P = 5.3560.38 N = 192). whether the biogenesis of mycolic acids is coordinated or not with The aspect of the images of the polar localization of MmpL3 bacterial envelope elongation. Through the use of a mycobacterial differed significantly from that observed with the FAS-II model of Mtb, M. smegmatis, we were able to localize MA components. The former was ‘‘crescent-shaped’’, suggesting a biosynthesis FAS-II proteins in vivo and compare their localization membrane localization of MmpL3. MmpL3 was located in the dynamics to that of the polar protein Wag31. Wag31 is a main envelope, probably in the cytoplasmic membrane, and was component of mycobacterial elongation and division [29,37] and preferentially located at the growing bacterial tips containing the has been found to be associated with the curved membranes of old larger amounts of Wag31. In addition, MmpL3 was visible very poles as well as of the division septum [49]. Herein; proteins of the early in the septa, likely before the appearance of Wag31 and

PLOS ONE | www.plosone.org 8 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

microscopy images (as in Figureoˆ 7 and Figureoˆ 8A) by scanning individual bacteria at the level of the membrane, the cytoplasm, and the poles as described in the Materials and Methods. The y-axis represents the polar accumulation (P) of either the complete (Wt) or the N-terminal portion (N10) of the MmpL3 transporter. A statistical t-test was performed to determine the difference between the polar accumulation of MmpL3-GFP (black bar, P = 5.35360.3810 N = 192) and the absence of polar accumulation of MmpL3-N10-GFP (white bar, P = 1.16060.1572 N = 113), with N indicating the number of bacteria analyzed. The unpaired t-test p value is symbolized by asterisks (***, p, 0.0001). doi:10.1371/journal.pone.0097148.g008 FAS-II complex core [16,17,18] were observed to form foci associated with both the old poles and septal poles. While the FAS- II condensing enzymes were mainly diffuse, they were also able to form foci at the same locations. In dynamic studies, newly- synthesized ectopic FAS-II proteins were first located at the septal region suggesting that they may participate very early in pole construction. Thereafter, they were established rapidly at the old poles together with polar Wag31 foci [29]. Such polar localization of FAS-II proteins as well as their dynamic localization to the old poles along with Wag31 suggest the existence of a spatiotemporal coordination between MA biosynthesis and Wag31-controlled polar PG synthesis. A key question was to determine whether foci, which were observed herein in bacteria expressing ectopic GFP fusions, correspond to actual complexes or to protein aggregates or inclusion bodies (IB). In rod-shaped bacterium E.coli, which does not display the same polar cell-wall elongation process as mycobacteria, the poles are not the active sites for PG synthesis [22]. In E.coli, the poles and the regions between the chromosomes are the areas where protein aggregates accumulate [50]. Further- more, these foci do not separate after division and are preferentially accumulated in the mother cell during cellular aging [51]. In contrast, in mycobacteria, poles are very active loci where lateral PG synthesis [22] and most likely free MA synthesis [52], occur during bacterial elongation. In addition, poles are also involved in protein export [53], DNA transfer [32], chromosomal partitioning and DNA replication [54,55] as well as phosphorylation by Ser/Thr protein kinases [56]. Mycobacterial poles are thus certainly not suitable for detrimental protein accumulation. A second argument against polar IB formation is the fact that asymmetric FAS-II polar localization was observed in the present model. Actinobacteria exhibit apical (polar) growth [24] and the distribution of polar proteins involved in PG biosynthesis is asymmetric with a preference for old poles [29]. Even though old pole-containing mycobacteria do not grow faster [57], polar growth itself appears to be asymmetric with the old pole being more active than the new pole [31]. Indeed, we observed herein a localization preference of FAS-II proteins to the old pole containing Wag31 whereas GFP alone seemed excluded from these poles. The foci clouds of FAS-II and Wag31 were located at the same position in the bacteria, at the very end of the organism. Figure 8. Dynamic localization of MmpL3-N10. (A) Images were As commonly observed in E.coli, inclusion bodies are otherwise extracted from the movie presented as movie S13. The time interval after the starting point of the movie is indicated in hours (h). Images more likely to be found further inside the bacterial tips. Finally, were extracted from three channels (bright field, BF, left column; GFP three additional observations in the present study advocate against and mCherry, GFP-Che, middle column; GFP, right column). The IB formation. First, there was no quantitative correlation between dynamics of the localization of the GFP fusions was followed in foci formation and GFP-fusion expression levels. Secondly, two of Msm::mCherry-wag31 bacteria expressing the MmpL3-N10-GFP fusions. the four fusion proteins tested (KasB and InhA) were active in vivo The presence of MmpL3-N10 in septa is indicated by white arrows in the right column. The position of mCherry-Wag31 at the poles and while forming foci. Thirdly, during the division process, we septa is clearly visible. Magnification: 630 X. Scale bar: 5 mm. (B) Bar observed foci splitting which can be compared to the active graph representation of the localization of MmpL3 and MmpL3-N10 in polarisome splitting in Streptomyces [58]. These finding strongly Msm strains expressing each GFP fusion after 6 hours of induction. Data suggest that the FAS-II foci revealed the presence of FAS-II are represented as mean 6 SEM. Data were collected from wide-field complexes mainly at the old poles and septal poles.

PLOS ONE | www.plosone.org 9 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

Mycolic acids are believed to be synthesized in the cytoplasm In conclusion, the polar localization of external membrane lipid and transported to the periplasm as TMM by the trans-membrane biosynthesis enzymes observed in the present study constitutes a RND-family transporter MmpL3 [19]. As expected, in view of its novel observation in bacteria and a fortiori in mycobacteria. secondary structure and function, MmpL3 was observed at the Moreover, the observation of the dynamic localization of the MA level of the mycobacterial envelope. It is likely inserted in the transporter MmpL3 strongly reinforces the possibility of a plasma membrane similarly to that proposed in the original positioning of MA biosynthesis in these regions. In keeping with MmpL3 study [19]. Moreover, RND family proteins such as gram the existing literature, the present data further suggest the negative efflux pumps for example are inserted in the inner existence of the polar localization of complete MAPc synthesis membrane and driven by the proton motive force [59] solely and its coordination with bacterial cell elongation. Peptidoglycan present at this level of the envelope. In addition, the MmpL3 C- is already known to be synthesized at these positions [70] as is terminal GFP fusion used herein was fluorescent, indicating that likely arabinan [52]. Moreover, mycolic acids are linked to the C-terminus domain of MmpL3 is probably located in the arabinogalactan at these positions [68,69]. Present findings also cytoplasm. GFP functions as an efficient reporter for analysis of show that FAS-II proteins, MmpL3 and Wag31 are present in the protein topology since it is fluorescent when located in the same region, at the same time in the bacteria; furthermore, it has cytoplasm and non-fluorescent when transported outside the recently been demonstrated that INH can affect the coupling of plasma membrane [48]. MmpL3-GFP was clearly more abundant division and elongation [57]. Altogether, all the above data further in the membrane of Wag31-containing old poles while displaying a place the pieces of the puzzle together and support the proposed crescent-shaped fluorescence. This typical shape is similar to that model of localized envelope biogenesis machinery coordinated observed for DivIVA, an ortholog of Wag31 in B.subtilis [28]. with the process of cell elongation and division. This coordinated Moreover, this shape was very different from the shape of the process may ultimately represent the Achilles’ heel of the GFP-FAS-II foci, suggesting that FAS-II was not inserted in the pathogenic bacterium Mycobacterium tuberculosis. membrane as in the case for MmpL3 or occupying the inside of the membrane as in the case of Wag31. Finally, the most Materials and Methods impressive images were obtained when the bacteria formed a Strains and culture conditions septum: MmpL3 yielded a very intense labeling at the position of the septum probably due to its insertion into the double membrane Plasmid constructions were performed in the Escherichia coli K12 of the septum. This MmpL3 localization pattern strongly derivative Top10-F’ (Invitrogen) using classical cloning procedures according to enzyme and product manufacturers’ recommenda- resembled that reported for membrane Ser/Thr protein kinase tions (NEBiolabs; Fermentas; Promega,). Culture media (Luria- PknB [56]. Indeed, Wag31 has been shown to be regulated by Bertani (LB) broth, or LB-Agar) were supplemented when needed PknA and PknB [29]. PknA and B have been involved in the with either kanamycin (50 mg/mL), hygromycin-B (200 mg/mL), inhibition of many MA biosynthesis proteins such as KasA [60], or chloramphenicol (12.5 mg/mL). Mycobacterium smegmatis mc2155 mtFabH [61], MabA [62], InhA [63,64] and the methyltransferase (ATCC 700084) was cultured in Middlebrook based media PcaA [65]. This suggests the existence of a common regulatory (Difco). Liquid cultures were grown in Middlebrook 7H9 (0.05% network between FAS-II, Wag31 and the cell wall elongation Tween 80, 10% ADC (BBL)) whereas bacteria were plated on process. From the present data and in view of the literature, we Middlebrook 7H10 agar (10% OADC (BBL)). Msm inhAts (mc2 can propose that mycolic acids are probably primarily exported at 2 2359; [44]) is a derivative of Msm mc 155 bearing a punctual the poles and septa in proximity to Wag31 and FAS-II foci and at mutation in the inhA gene rendering both the protein and the the locus of PG biosynthesis. The disappearance of the polar strain thermosensitive. When needed, media were supplemented accumulation of MmpL3 upon deletion of its C-terminal end with either kanamycin (50 mg/mL), hygromycin-B (150 mg/mL) suggests that this region is probably involved in its localization. It or zeocin (10 mg/ml). also suggests that MmpL3 may be localized through a diffusion- capture mechanism using its C-terminal end as signal [66]. Plasmid constructions A puzzling question was to ascertain whether the polar The four Mtb FAS-II genes (inhA, rv1484; mabA, rv1483; kasA, localizations of FAS-II and MmpL3 reflect MA biosynthesis and rv2245; kasB, rv2246) were isolated from the previously described export localization. Some reports can indeed be interpreted in this pGAD-T7::fas-II and pGBK-T7::fas-II derivatives [17] as either manner when compared with our results. Mtb polar lysis has been Nde1-BamH1 (kasA and mabA), Nde1-EcoR1 (inhA), or Nde1-Msc1 observed in very early EM images of isoniazid-treated wild-type fragments (kasB). They were cloned under the control of the bacilli [67]. Isoniazid targets InhA and if we assume that synthesis acetamidase promoter (P ) at the corresponding sites of the and export of mycolic acids are primarily polar, InhA inhibition amiE inducible vector pLAM12 [71] to produce the pLAM12::fas-II will leave ‘‘holes’’ in the polar mycomembrane and ultimatly derivatives (Table S1). For the construction of the fluorescence induce specific unipolar lysis as previously reported [67]. More control vectors, the egfp and the mCherry genes were amplified recently, polar lysis was also observed when arabinan synthesis was respectively from pEGFPN1 (Clontech) and from pmCherry-N1 inhibited with benzothiazinone [31,52]. Finally, a link with MA (Clontech) using specific pairs of primers creating a Nde1 site at the synthesis was also attributed following MmpL3 localization. ATG position of each gene and a second site (egfp, HindIII; mcherry, Because MmpL3 is accumulated at the pole, MA export (and BamH1) at the end of each gene (Table S2). These genes were then probably MA synthesis) can be polar. By using unnatural trehalose cloned at the same sites in pLAM12 to yield pLAM12::gfp and analogs, two separate reports have shown that Ag85 activity, the pLAM12::mCherry (Table S1). For construction of N-terminal GFP mycoloyl-transferase that links mycolic acids to arabinogalactan, is fusions with the FAS-II proteins, the egfp gene was amplified from concentrated in the envelope and at the poles of mycobacteria the pEGFPN1 vector using a second set of primers creating in- with a pattern strongly resembling that of MmpL3 herein [68,69]. phase Nde1 sites both at the ATG and STOP positions (Table S2). The linking of mycolic acids to the cell wall may occur mainly at The egfp gene was inserted at the Nde1 sites of each of the four these positions. As suggested by the present results, the C-terminal FAS-II genes in the pLAM12 derivatives. For construction of the domain of MmpL3 may be involved in this localization. Wag31 fusion, the wag31 gene (rv2145c) was amplified by PCR

PLOS ONE | www.plosone.org 10 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes from Mtb H37Rv genomic DNA using specific primers (Table S2) Construction of the Mycobacterium smegmatis kasB creating a Nde1 site at the ATG position and cloned as an Nde1- knockout strain BamH1 fragment under the control of the pHSP60 promoter of The Msm kasB gene (MSMEG_4328) was disrupted by using the the replicative vector pMV261-Nde [17], a derivative of pMV261 ‘‘recombineering method’’ developed by van Kessel and Hatfull [72]. Using the same strategy as above, the mCherry gene was [71]. The linear allelic exchange substrate (AES; Figure S5) was cloned, as an intermediate, at the Nde1 site of pMV261::wag31 to constructed using a three fragments fusion-PCR [73] with the yield a N-terminal fusion with Wag31. The fusion was further indicated primers (Table S2). The AES was constituted, from 59 to transported at an equivalent position of the integrative vector 39, by the fusion of the 800 bps of DNA upstream from the pMV361 [72]. The same wag31 Nde1-BamH1 DNA fragment was initiator codon of kasB, followed by the Zeocin (InvivoGen) also cloned at the equivalent positions of the Matchmaker III yeast resistance gene (Sh ble,) from the pZeo vector [74] and ended by two-hybrid cloning vectors pGAD-T7 and pGBK-T7 (Clontech) the 800 bps of DNA downstream from the stop codon of kasB. The to produce pGAD-T7::wag31 and pGBK-T7::wag31. In the same substrate was subsequently introduced by electro-transformation manner, the Mtb nat gene (rv3566) was cloned in the Matchmaker into a recipient Msm strain already containing a hygromycin- III vectors as Nde1-BamH1 fragments. Finally, the mmpL3 gene resistant derivative of the recombineering plasmid pJV53 encoding (rv0206c) and its N-terminal end containing only the first ten trans- for phage recombinases [71]. The construction and the selection of membrane helices (Mp3-N10) were cloned at the Nde1 site of the mutant were essentially performed as previously described pLAM12::gfp to construct the C-terminal MmpL3-GFP and [71]. The chromosomal structure of the mutant (Figure S5) was MmpL3-N10-GFP fusion plasmids pLAM12::mmpL3::gfp and verified by PCR with suitable primers (Table S2 and Figure. S5). pLAM12::mp3-N10::gfp. All constructs were verified by restriction After curing of the recombineering plasmid, the mutant strain was analysis and sequencing (Millegen). transformed by appropriate plasmid to produce the control strain (Msm-DkasB/pLAM12) and the complemented strains (Msm- Analysis of GFP-fusion expression in vivo DkasB/pLAM12::kasB and Msm DkasB/pLAM12::gfp-kasB)). Msm liquid cultures were induced with 0.2% acetamide in early log phase (OD590nm = 0.5) and samples were collected every two Mycolic acid analysis hours. Fluorescence at 525 nm (excitation at 485 nm) and MA was performed essentially as described [75]. Bacterial OD590nm were measured in micro-titration plates (Optiplate-96 well) with a FLx800 fluorescence reader (Biotek Instruments). residues obtained after lipid extraction with organic solvents were Acquisition of fluorescence in the sample (F(s)) was performed in saponified at 110uC for 3 h in a mixture of KOH (40%) and triplicate on serial dilutions of each culture. The data, expressed in methoxyethanol (C3H8O2) (1:7, v/v). After acidification, fatty relative fluorescence units (RLU), were corrected for the variation acids were extracted with diethyl ether and methylated with an in fluorescence of the medium (F(m)) and the background ethereal solution of diazomethane. The mycolate patterns of the fluorescence of the Msm/pLAM12 control strain (F(c)). Fluores- strains were determined by analytical high performance thin-layer cence (F) was also reported to the OD of the sample (OD(s)) and of chromatography (HPTLC) on Silica gel 60 (Merck) using the control strain (OD(c)) using the following formula: F = [(F(s) – dichloromethane 100% as eluent and revealed with rhodamine F(m))/OD(s))]/[(F(c) – F(m))/(OD(c)]. The means 6 SD (standard B (Sigma). The different classes of mycolates were separated by deviation) of the data were plotted against the time of induction preparative TLC (Macherey-Nagel Silica Gel 60), using the same using Graphpad Prism 5 software (GraphPad Software Inc.). The eluent. The lengths of the various types of purified mycolates were same culture samples were analyzed by Western blotting during determined by MALDI-TOF mass spectrometry as previously the course of induction. Aliquots of known and equivalent OD described. MALDI mass spectra were acquired in reflectron mode were lysed, fractionated on 10% SDS-PAGE and appropriate using an Applied Biosystems 4700 Analyzer equipped with a amounts of total protein were blotted using mouse monoclonal Nd:YAG laser (355 nm wavelength, ,500-ps pulse and 200Hz anti-GFP (Roche Applied science ref: 11814460001) as primary repetition rate). The shots (2500 total) were accumulated in antibodies. The blots were revealed by bioluminescence (ECL positive ion mode and MS data were acquired using the detection Kit, Millipore) and visualized by fluoroimaging. instrument default calibration. Mycolate samples were dissolved in chloroform at a concentration of 1 mM, and were directly In vitro Co-immunoprecipitation (Co-IP) spotted onto the target plate as 0.5 ml droplets, followed by the In vitro transcription/translation of the genes of interest with the addition of 0.5 ml matrix solution. Samples were allowed to TnT Quick Coupled Transcription/Translation System (Pro- crystallize at room temperature. The matrix used was 2,5- mega) was performed as described previously [18] using super- dihydroxybenzoic acid (10 mg ml21) in CHCl3/CH3OH (1 : 1). coiled DNA from pGAD-T7 or pGBK-T7 derivatives expressing either the FAS-II genes [16,17] or the wag31 and nat genes as Microscopy and video-microscopy substrates (Table S1). This system allowed generating L-[35S]- Msm liquid cultures expressing either the GFP protein fusions methionine-labeled proteins tagged with either the HA (hemag- alone, or the GFP fusions together with the mCherry-Wag31 glutinin) epitope (pGAD-T7 derivatives) or the c-Myc epitope fusion were induced in log phase (OD590nm 0.5) with 0.2% (pGBK-T7). The products were named h-proteins (HA-tagged) or acetamide during 6 hours. Bacteria were pelleted, concentrated c-proteins (c-Myc-tagged). For Co-IP experiments, the h-proteins (20X) in PBS, and spread (10 mL) on poly-L-lysine coated glass and the c-proteins were mixed together with goat anti-mouse IgG slides (PolyScience). Slides were mounted with Dako mounting magnetic beads (Dynabeads M-450) coated with monoclonal anti- medium and observed with a large field Leica DMIRB fluores- HA antibodies (Sigma). Proteins ratios were adjusted to 1:1 as cence microscope at 63X magnification (Leica HCX Plan Apo described by assessing their specific activities (phosphorimaging) x63/1.40-Oil). GFP (488 nm/509 nm) was revealed with a GFP- and correcting for their differences in the number of methionine ET filter cube (excitation: BP 470/40; dichroic mirror, 495; residues. The co-IP reactions were performed as previously emission: LP 525/50) whereas mCherry fluorescence (587 nm/ described [18] and analyses performed by separation on SDS- 610 nm) was visualized with a N2.1 Leica filter cube (excitation: PAGE (4-20%) and phosphorimaging. BP 515–560; dichroic mirror, 580; emission LP 590). Images were

PLOS ONE | www.plosone.org 11 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes captured with a CoolSnap HQ2 CCD camera (1 pixel = 6.45 mM), channels were extracted from the time-lapse image stacks and acquired with MetaMorph 7.1.7 software and processed with combined as described in the supplemental movie legends, after ImageJ software (National Institutes of Health, Bethesda, MD). All which they were converted to AVI (Audio Video Interleave) different images were acquired with the same exposure time format at 0.5 to 1.5 frames per second using JPEG compression. (200 ms). Image processing consisted in equivalent adjustments of Minimal image processing was performed as described above for brightness and contrast on complete images. Gamma and LUT static images. (Look Up Table) values were not modified and were left as linear on each channel. For statistical analysis and quantification of the Supporting Information polar indices, multiple slides (n) of independent experiments were used. An average of 1500 bacteria (N; see text) were counted from Figure S1 The mycolic acid biosynthesis pathway. The a minimum of three different slides and a maximum of 16 slides substrates and products of the Fatty Acid Synthase of type I (FAS- (3,n,16). Student t-tests were performed on raw data using the I) and II (FAS-II) are indicated together with the biochemical GraphPad Prism 5.0 software. Polar indices were expressed as reactions (black arrows) necessary to achieve the biosynthesis of percentages of polar foci in the fluorescent population and mature mycolic acid (see [8] and [9] for review). The enzymes presented as mean with standard deviation (SD). For precise responsible for each reaction are identified in colored circles. The analysis of colocalization, each channel (GFP and mCherry) on cofactors necessary for reduction reactions to occur (NADPH, microscopy images from individual bacteria producing both types NADH) were omitted for reasons of clarity. The proximal and of fusion (greater than 100) were scanned using the ROI manager distal positions of the meromycolic chain modifications by MA- of ImageJ software (NIH-USA). Thereafter, the GFP and the Mtfs are indicated as P and D respectively. Intermediate reactions mCherry fluorescence profiles of each bacterium were obtained not detailed (acyl-chains activation) are symbolized by interrupted with the Plot Profile plugin. The maximum fluorescence levels of arrows. The synthesis of the meromycolic chain is initiated by the each scan of a given strain were normalized to the highest condensation by the mtFabH protein of the acyl-CoA products of fluorescence value in this strain and plotted against their position FAS-I with malonyl-ACP by MtFabD to give a keto-acyl-ACP. along the long axis of the bacterium. The total length of each After reduction by the keto-acyl-ACP reductase MabA, followed bacterium was reported to 1. by dehydration by the hydroxyl-acyl-dehydratases HadAB or In order to quantify the polar localization of MmpL3, images HadBC and reduction by the enoyl-ACP reductase InhA, the acyl- from wide field fluorescence microscopy of Msm expressing either chain enters into a new cycle of elongation through condensation MmpL3-GFP or MmpL3-N10-GFP were analyzed with ImageJ by the keto synthase KasA or KasB with a new malonyl-ACP unit. software (NIH-USA) using the Plot Profile function. On each After its synthesis, the meromycolic chain is adenylated and ligated individual bacterium, three plots along 0.7 mm width lines were by FadD32 onto Pks13, which is the terminal condensing enzyme obtained: the first inside the cytoplasm (Background Fluorescence, that links the meromycolic chain to a carboxylated alpha chain FBG), the second along the long axis of the bacterium and passing produced by FAS-I. The remaining keto function of the generated through one pole (Polar Fluorescence, FPO), and the third along mycolic motif is then reduced by CmrA to yield the mycolic acid. the short axis of the bacterium and crossing the membranes (TIF) (Membrane fluorescence, F ). On each bacterium, the maxi- MB Figure S2 The Mycolic Acid Biosynthesis Interactome. mum fluorescence (FMax) of each plot was collected. The analysis was performed on more than 100 individual bacteria. The polar The Mycolic Acid Biosynthesis Interactome (M.A.B.I.) is com- localization fold ratio was calculated with the formula, [|F Max- posed of three types of complexes: (i) the ‘initiation FAS-II’ (I-FAS- PO II) contains mtFabH interacting with a core (MabA, InhA and FBGMax|]/[|FMBMax-FBGMax|], yielding a value of 1 when there was no polar localization and a value superior to 1 when the mtFabD) and represents the link between FAS-I and FAS-II; (ii) fluorescent protein was accumulated at the membrane pole. two ‘elongation FAS-II’ (E-FAS-II) complexes comprise the core Polarization was expressed as mean 6 SEM (Standard Error to interacting preferentially with either KasA and HadAB (E1-FAS- the Mean), with N represent here the number of bacteria II) or KasB and HadBC (E2-FAS-II); these two complexes are analyzed. thought to be capable of elongating acyl-AcpM to produce full- Time-lapse video microscopy was performed using an agar pad length meromycoloyl-AcpM, (iii) the ‘termination FAS-II’ (T-FAS- method essentially as previously described [34]. The bacteria were II) involves Pks13 interacting with KasB and is thought to grown to mid-log phase, and aliquots (400 mL) were applied onto condense the a-branch with the meromycolic branch. Our the coverslip of 35 mm glass bottom dishes (MatTek). After working hypothesis is that the formation of each complex type 5 minutes, the liquid was removed by aspiration and the bacteria may be successive and that the acyl-ACP substrates may be were covered by 7H9 medium (10% ADC) at 37uC containing channeled from one type of complex to another as a function of its 0.6% Noble agar (Sigma), along with the appropriate antibiotics elongation status. The FAS-II initiation complex (I-FAS-II), the and the inducer (acetamide, 0.02%). After 1 hour solidification, elongation complexes 1 and 2 (E1-FAS-II and E2-FAS-II) and the the plates were pre-incubated for three hours at 37uC. The termination complex (T-FAS-II) are represented in grey. The fluorescence was detectable in control GFP-producing bacteria. condensing enzymes KasA, KasB and MtFabH (FabH) are in This time was chosen as the starting point (t = 0) for all time lapse orange. The interactions between the HadAB (AB) and HadBC experiments. Thereafter, the plates were observed as above using (BC) dehydrase heterodimers (in green) with the MA-Mtf (in violet) the Leica DMIRB florescence microscope equipped with a 37uC are symbolized by curved arrows. The core (in yellow) symbolizes humid chamber. GFP and mCherry fluorescence was visualized the reductases MabA and InhA together with the malonylCoA with a GFP/mCherry-ET filter cube (Chroma Technology; ACP transacylase MtFabD. The terminal condensing enzyme excitation, no selection; dichroic filter, SP490-BP570/40; emission Pks13 is depicted in blue. The direction of trafficking of the BP522/44-BP633/54). Individual frames (100 ms exposure) were elongating substrate is symbolized by arrows emanating from acyl- acquired at regular time intervals (from 10 min to 2 hours) with CoA (from FAS-I) toward the mature mycolic acid. Interrupted MetaMorph 7.1.7 software during 14 to 20 hours. The represen- arrows symbolize omitted steps of substrate modifications. tative movies were generated using ImageJ software. The required (TIF)

PLOS ONE | www.plosone.org 12 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

Figure S3 Localization of mCherry-Wag31 in Table S1 Description of plasmids. Msm::mcherry-wag31. Enlarged images of wide-field micros- (DOCX) copy experiments on mCherry-Wag31 fusion-expressing bacteria Table S2 Oligonucleotide sequences of PCR primers. as presented in Figure 2 and Figure 4. Brightfield images (BF; (DOCX) leftmost column) together with mCherry fluorescence images (Che, middle column) of Msm mc2 155::pMV361::mCherry (Ø) and Movie S1 Localization of GFP. Growth of Msm derivatives 2 Msm mc 155::pMV361::mCherry-wag31 (Wag31) were acquired expressing GFP alone under the control of the pamiE promoter was 6 hours after dilution of the main culture and in the presence of followed by fluorescence time lapse video-microscopy. The movie anhydro-tetracyclin (50 ng/mL) for the latter. The merged images (left panel) obtained by merging the bright field channel (in red) (right columns) allowed visualizing the localization of Che-Wag31 with the GFP channel (in green) is presented with the movie (right mainly at one pole of the bacteria (the old pole). Total optical panel) displaying only the GFP channel. AVI file (0.8 frames per magnification: 630 X. Scale bar: 5 mm. second) was obtained as described in Materials and Methods at 0.8 (TIF) frames per second with one frame per hour. Time intervals are indicated in hours (h). Scale bar: 5 mm. Figure S4 Analysis of FAS-II/Wag31 interactions by co- (AVI) immunoprecipitation. L-[35S]-methionine-labeled h-proteins (HA tagged; h-protein) and c-proteins (c-Myc tagged; c-protein) Movie S2 Localization of GFP-MabA. Growth of Msm from in vitro transcription translation reactions as well as the Co-IP derivatives expressing the GFP-MabA fusion under the control of reaction products were fractionated by SDS-PAGE (4–20%) the pamiE promoter was followed by fluorescence time lapse video- followed by phosphorimaging analysis. The gels containing the microscopy. Movie treatment is identical to that of Movie S1 h-Wag31 protein alone or the c-Wag31 alone are represented in (AVI) the leftmost lanes. The names of the c-proteins or h-proteins used Movie S3 Localization of GFP-InhA. Growth of Msm in the Co-IP experiments are indicated above each panel. These derivatives expressing the GFP-InhA fusion under the control of proteins were run alone (-), or after Co-IP (IP). (A) c-Wag31 was the pamiE promoter was followed by fluorescence time lapse video- tested against h-FAS-II. (B) h-Wag31 was tested against c-FAS-II microscopy. Movie treatment is identical to that of Movie S1 proteins; IPØ refers to a Co-IP performed between a given c- (AVI) protein and a void extract (obtained with the pGAD-T7 empty Movie S4 Msm vector). (C) h-Wag31 was tested against human laminC (c-Lam) Localization of GFP-KasA. Growth of derivatives expressing the GFP-KasA fusion under the control of with Mtb Arylamine N-acetyltransferase (c-Nat) used as negative the p promoter was followed by fluorescence time lapse video- control. The black arrowheads represent a Co-IP band while the amiE microscopy. Movie treatment is identical to that of Movie S1 open arrowheads indicate the position of a missing Co-IP band except that the AVI file was obtained at 1.5 frames per second corresponding to a negative interaction. When the sizes of both with one frame obtained every 15 min. proteins were too close to each other, non-labeled h-proteins were (AVI) used for Co-IP experiments as indicated with an asterisk. (TIF) Movie S5 Localization of GFP-KasB. Growth of Msm derivatives expressing GFP-KasB alone under the control of the Figure S5 Construction and characterization of p promoter was followed by fluorescence time lapse video- M.smegmatis DkasB. (A) The wild type region (WT) of the amiE microscopy. Movie treatment is identical to that of Movie S1 Msm chromosome comprising kasA (open arrow), kasB (grey arrow) (AVI) and accD6 (open arrow) is depicted as a solid line together with the positions for hybridization of PCR primers (Wt-Fwd and Wt-Rev). Movie S6 Localization of mCherry-Wag31 and GFP. The AES product synthesized by fusion-PCR is also presented Growth of Msm::pMV361 mCherry-Wag31 derivatives expressing with the primers used for its synthesis. The Zeocin resistance gene GFP alone under the control of the pamiE promoter was followed (Sh ble; ble) is represented as a black arrow. The structure of the by fluorescence time lapse video-microscopy. The movie (left chromosomal region after the allelic exchange is illustrated panel) obtained by merging together the bright field channel (in together with the positions of PCR primers used for its blue), the GFP channel (in green) and the mCherry channel (in characterization. The lengths of each PCR product are indicated red) is presented with the movie displaying only the merged GFP in base pairs (bps) and represented by dashed lines. (B) Ethydium and the mCherry channels (right panel). The AVI file was bromide staining of an agarose gel of PCR products of wild-type obtained as described in Materials and Methods at 0.5 frames per (Wt) and DkasB (D) Msm genomic DNA. The primer pairs used for second. With one frame obtained every hour. The time intervals amplification are indicated above the gel. The migration positions are indicated in hours (h). Scale bar: 5 mm. of control linear DNA are indicated along with their respective (AVI) sizes (in kbp). Movie S7 Localization of mCherry-Wag31 and GFP- (TIF) MabA. Growth of Msm::pMV361 mCherry-Wag31 derivatives Figure S6 MALDI-TOF analysis of a’- and epoxy- expressing the GFP-MabA fusion under the control of the pamiE mycolic acid methyl esters. Mass spectra of a’-mycolic acid promoter was followed by fluorescence time lapse video-micros- (left column) and epoxy-mycolic acid methyl esters (right column) copy. Movie treatment is identical to that of Movie S6 except that extracted from wild type (Msm mc2155, WT), mutant (Msm the AVI file was obtained at 1.5 frames per second with one frame mc2155 DkasB, DB) and complemented (Msm mc2155 DkasB/kasB, obtained every 10 min. DB/B; Msm mc2155 DkasB/gfp-kasB, DB/gB) strains. The accurate (AVI) masses (m/z), together with the length of the even carbon- Movie S8 Localization of mCherry-Wag31 and GFP- numbered epoxy-mycolates and the odd carbon-numbered a’- InhA. Growth of Msm::pMV361 mCherry-Wag31 derivatives mycolates are indicated. expressing the GFP-InhA fusion under the control of the pamiE (TIF) promoter was followed by fluorescence time lapse video-micros-

PLOS ONE | www.plosone.org 13 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

copy. Movie treatment is identical to that of Movie S6 except that pamiE promoter was followed by fluorescence time lapse video- the AVI file was obtained at 1.5 frames per second with one frame microscopy. Movie treatment is identical to that of Movie S6. obtained every 15 min. (AVI) (AVI) Movie S13 Localization of mCherry-Wag31 and Movie S9 Localization of mCherry-Wag31 and GFP- MmpL3-N10-GFP. Growth of Msm::pMV361 mCherry-Wag31 KasA. Growth of Msm::pMV361 mCherry-Wag31 derivatives derivatives expressing the GFP fusion of the membrane domain of expressing the GFP-KasA fusion under the control of the pamiE MmpL3 (Mp3-N10-GFP) under the control of the pamiE promoter promoter was followed by fluorescence time lapse video-micros- was followed by fluorescence time lapse video-microscopy. Movie copy. Movie treatment is identical to that of Movie S6 except that treatment is identical to that of Movie S6. the AVI file was obtained at 1.5 frames per second with one frame (AVI) obtained every 15 min. (AVI) Acknowledgments Movie S10 Localization of mCherry-Wag31 and GFP- We thank John McKinney and Neeraj Dhar (EPFL; Lausanne, KasB. Growth of Msm::pMV361 mCherry-Wag31 derivatives Switzerland), Jean-Yves Bouet and Jeroˆme Rech (LMGM-CNRS; expressing the GFP-KasB fusion under the control of the pamiE Toulouse, France), and Serges Maze`res (TRI platform; IPBS-CNRS- promoter was followed by fluorescence time lapse video-micros- UPS; Toulouse, France) for very initial help on time lapse microscopy copy. Movie treatment is identical to that of Movie S6. experiments setup. We are particularly grateful to Lucie Spina for technical (AVI) assistance on HPTLC mycolic acid analysis. We thank JC van Kessel and G Hatfull for providing the pLAM12 plasmid from the recombineering Movie S11 Localization of MmpL3-GFP. Growth of Msm system [71] and W. Jacobs for providing the Msm inhAts mutant strain [44]. derivative expressing the MmpL3-GFP fusion under the control of Thank to Mr. Pierre Pothier for the editing of the manuscript. the pamiE promoter was followed by fluorescence time lapse video- microscopy. Movie treatment is identical to that of Movie S1 Author Contributions (AVI) Conceived and designed the experiments: DZ CC. Performed the Movie S12 Localization of mCherry-Wag31 and experiments: CC KN SC MB CD FL DZ. Analyzed the data: DZ CC MmpL3-GFP. Growth of Msm::pMV361 mCherry-Wag31 deriva- MD. Wrote the paper: DZ. tives expressing the MmpL3-GFP fusion under the control of the

References 1. Daffe M, Draper P (1998) The envelope layers of mycobacteria with reference to tuberculosis are essential for mycobacterial viability. Mol Microbiol 54: 1161– their pathogenicity. Adv Microb Physiol 39: 131–203. 1172. 2. Minnikin DE, Kremer L, Dover LG, Besra GS (2002) The methyl-branched 18. Cantaloube S, Veyron-Churlet R, Haddache N, Daffe M, Zerbib D (2011) The fortifications of Mycobacterium tuberculosis. Chem Biol 9: 545–553. Mycobacterium Tuberculosis FAS-II Dehydratases and Methyltransferases 3. Ernst JD (2012) The immunological life cycle of tuberculosis. Nat Rev Immunol Define the Specificity of the Mycolic Acid Elongation Complexes. PLoS One 12: 581–591. 6: e29564. 4. Hett EC, Rubin EJ (2008) Bacterial growth and cell division: a mycobacterial 19. Grzegorzewicz AE, Pham H, Gundi VA, Scherman MS, North EJ, et al. (2012) perspective. Microbiol Mol Biol Rev 72: 126–156. Inhibition of mycolic acid transport across the Mycobacterium tuberculosis 5. Letek M, Fiuza M, Villadangos A, Mateos L, Gil J (2012) Cytoskeletal proteins plasma membrane. Nat Chem Biol 8: 334–341. of actinobacteria. International journal of cell biology 2012: 905832. 20. Zumla A, Hafner R, Lienhardt C, Hoelscher M, Nunn A (2012) Advancing the 6. Boehringer D, Ban N, Leibundgut M (2013) 7.5 A˚ Cryo-EM Structure of the development of tuberculosis therapy. Nat Rev Drug Discov 11: 171–172. Mycobacterial Fatty Acid Synthase. Journal of molecular biology 825: 841–849. 21. Varela C, Rittmann D, Singh A, Krumbach K, Bhatt K, et al. (2012) MmpL 7. Gago G, Diacovich L, Arabolaza A, Tsai SC, Gramajo H (2011) Fatty acid genes are associated with mycolic acid metabolism in mycobacteria and biosynthesis in actinomycetes. FEMS Microbiol Rev 35: 475–497. corynebacteria. Chemistry & biology 19: 498–506. 8. Takayama K, Wang C, Besra GS (2005) Pathway to synthesis and processing of 22. Daniel RA, Errington J (2003) Control of cell morphogenesis in bacteria: two mycolic acids in Mycobacterium tuberculosis. Clin Microbiol Rev 18: 81–101. distinct ways to make a rod-shaped cell. Cell 113: 767–776. 9. Marrakchi H, Lane´elle M-A, Daffe´ M (2014) Mycolic acids: structures, 23. den Blaauwen T, de Pedro M, Nguyen-Diste`che M, Ayala J (2008) biosynthesis, and beyond. Chemistry & biology 21: 67–85. Morphogenesis of rod-shaped sacculi. FEMS microbiology reviews 32: 321–344. 10. Zuber B, Chami M, Houssin C, Dubochet J, Griffiths G, et al. (2008) Direct 24. Letek M, Fiuza M, Ordo´n˜ez E, Villadangos A, Ramos A, et al. (2008) Cell visualization of the outer membrane of mycobacteria and corynebacteria in their growth and cell division in the rod-shaped actinomycete Corynebacterium native state. J Bacteriol 190: 5672–5680. glutamicum. Antonie Van Leeuwenhoek 94: 99–109. 11. Hoffmann C, Leis A, Niederweis M, Plitzko JM, Engelhardt H (2008) Disclosure 25. Edwards D, Thomaides H, Errington J (2000) Promiscuous targeting of Bacillus of the mycobacterial outer membrane: cryo-electron tomography and vitreous subtilis cell division protein DivIVA to division sites in Escherichia coli and sections reveal the lipid bilayer structure. Proc Natl Acad Sci U S A 105: 3963– fission yeast. The EMBO journal 19: 2719–2727. 3967. 26. Cha JH, Stewart GC (1997) The divIVA minicell locus of Bacillus subtilis. 12. Crick DC, Mahapatra S, Brennan PJ (2001) Biosynthesis of the arabinogalactan- J Bacteriol 179: 1671–1683. peptidoglycan complex of Mycobacterium tuberculosis. Glycobiology 11: 107R– 27. Fla¨rdh K (2003) Essential role of DivIVA in polar growth and morphogenesis in 118R. Streptomyces coelicolor A3(2). Molecular microbiology 49: 1523–1536. 13. Cole S, Brosch R, Parkhill J, Garnier T, Churcher C, et al. (1998) Deciphering 28. Lenarcic R, Halbedel S, Visser L, Shaw M, Wu L, et al. (2009) Localisation of the biology of Mycobacterium tuberculosis from the complete genome sequence. DivIVA by targeting to negatively curved membranes. The EMBO journal 28: Nature 393: 537–544. 2272–2282. 14. Kremer L, Baulard A, Besra GS (2000) Genetics of Mycolic Acid Biosynthesis. 29. Kang CM, Nyayapathy S, Lee JY, Suh JW, Husson RN (2008) Wag31, a In: Hatfull G, Jacobs WR, editors. Molecular Genetics of Mycobacteria. homologue of the cell division protein DivIVA, regulates growth, morphology Washington, D.C.: ASM Press.pp. 173–190. and polar cell wall synthesis in mycobacteria. Microbiology 154: 725–735. 15. Asselineau C, Asselineau J, Laneelle G, Laneelle MA (2002) The biosynthesis of 30. Jani C, Eoh H, Lee JJ, Hamasha K, Sahana MB, et al. (2010) Regulation of mycolic acids by Mycobacteria: current and alternative hypotheses. Prog Lipid polar peptidoglycan biosynthesis by Wag31 phosphorylation in mycobacteria. Res 41: 501–523. BMC Microbiol 10: 327. 16. Veyron-Churlet R, Bigot S, Guerrini O, Verdoux S, Malaga W, et al. (2005) 31. Aldridge BB, Fernandez-Suarez M, Heller D, Ambravaneswaran V, Irimia D, et The biosynthesis of mycolic acids in Mycobacterium tuberculosis relies on al. (2012) Asymmetry and aging of mycobacterial cells lead to variable growth multiple specialized elongation complexes interconnected by specific protein- and antibiotic susceptibility. Science 335: 100–104. protein interactions. J Mol Biol 353: 847–858. 32. Wirth SE, Krywy JA, Aldridge BB, Fortune SM, Fernandez-Suarez M, et al. 17. Veyron-Churlet R, Guerrini O, Mourey L, Daffe M, Zerbib D (2004) Protein- (2012) Polar assembly and scaffolding proteins of the virulence-associated ESX-1 protein interactions within the Fatty Acid Synthase-II system of Mycobacterium secretory apparatus in mycobacteria. Mol Microbiol 83: 654–664.

PLOS ONE | www.plosone.org 14 May 2014 | Volume 9 | Issue 5 | e97148

In Vivo Localization of Mycobacterial FAS-II Complexes

33. Flardh K (2010) Cell polarity and the control of apical growth in Streptomyces. 54. Maloney E, Madiraju M, Rajagopalan M (2009) Overproduction and Curr Opin Microbiol 13: 758–765. localization of Mycobacterium tuberculosis ParA and ParB proteins. Tubercu- 34. Joyce G, Robertson BD, Williams KJ (2011) A modified agar pad method for losis (Edinb) 89 Suppl 1: S65–69. mycobacterial live-cell imaging. BMC Res Notes 4: 73. 55. Maloney E, Madiraju SC, Rajagopalan M, Madiraju M (2011) Localization of 35. Kang CM, Abbott DW, Park ST, Dascher CC, Cantley LC, et al. (2005) The acidic phospholipid cardiolipin and DnaA in mycobacteria. Tuberculosis (Edinb) Mycobacterium tuberculosis serine/threonine kinases PknA and PknB: substrate 91 Suppl 1: S150–155. identification and regulation of cell shape. Genes Dev 19: 1692–1704. 56. Mir M, Asong J, Li X, Cardot J, Boons GJ, et al. (2011) The extracytoplasmic 36. Santi I, Dhar N, Bousbaine D, Wakamoto Y, McKinney JD (2013) Single-cell domain of the Mycobacterium tuberculosis Ser/Thr kinase PknB binds specific dynamics of the chromosome replication and cell division cycles in mycobac- muropeptides and is required for PknB localization. PLoS Pathog 7: e1002182. teria. Nat Commun 4: 2470. 57. Wakamoto Y, Dhar N, Chait R, Schneider K, Signorino-Gelo F, et al. (2013) 37. Nguyen L, Scherr N, Gatfield J, Walburger A, Pieters J, et al. (2007) Antigen 84, Dynamic persistence of antibiotic-stressed mycobacteria. Science 339: 91–95. an effector of pleiomorphism in Mycobacterium smegmatis. J Bacteriol 189: 58. Richards DM, Hempel AM, Flardh K, Buttner MJ, Howard M (2012) 7896–7910. Mechanistic basis of branch-site selection in filamentous bacteria. PLoS Comput 38. Munshi T, Gupta A, Evangelopoulos D, Guzman JD, Gibbons S, et al. (2013) Biol 8: e1002423. 59. Blair JM, Piddock LJ (2009) Structure, function and inhibition of RND efflux Characterisation of ATP-dependent Mur ligases involved in the biogenesis of cell pumps in Gram-negative bacteria: an update. Current opinion in microbiology wall peptidoglycan in Mycobacterium tuberculosis. PLoS One 8: e60143. 12: 512–519. 39. Gao LY, Laval F, Lawson EH, Groger RK, Woodruff A, et al. (2003) 60. Molle V, Brown AK, Besra GS, Cozzone AJ, Kremer L (2006) The Condensing Requirement for kasB in Mycobacterium mycolic acid biosynthesis, cell wall Activities of the Mycobacterium tuberculosis Type II Fatty Acid Synthase Are impermeability and intracellular survival: implications for therapy. Mol Differentially Regulated by Phosphorylation. J Biol Chem 281: 30094–30103. Microbiol 49: 1547–1563. 61. Veyron-Churlet R, Molle V, Taylor RC, Brown AK, Besra GS, et al. (2009) The 40. Bhatt A, Fujiwara N, Bhatt K, Gurcha SS, Kremer L, et al. (2007) Deletion of Mycobacterium tuberculosis {beta}-Ketoacyl-Acyl Carrier Protein Synthase III kasB in Mycobacterium tuberculosis causes loss of acid-fastness and subclinical Activity Is Inhibited by Phosphorylation on a Single Threonine Residue. J Biol latent tuberculosis in immunocompetent mice. Proc Natl Acad Sci U S A 104: Chem 284: 6414–6424. 5157–5162. 62. Veyron-Churlet R, Zanella-Cleon I, Cohen-Gonsaud M, Molle V, Kremer L 41. Bhatt A, Kremer L, Dai AZ, Sacchettini JC, Jacobs WR, Jr. (2005) Conditional (2010) Phosphorylation of the Mycobacterium tuberculosis beta-ketoacyl-acyl depletion of KasA, a key enzyme of mycolic acid biosynthesis, leads to carrier protein reductase MabA regulates mycolic acid biosynthesis. J Biol Chem mycobacterial cell lysis. J Bacteriol 187: 7596–7606. 285: 12714–12725. 42. Parish T, Roberts G, Laval F, Schaeffer M, Daffe M, et al. (2007) Functional 63. Molle V, Gulten G, Vilcheze C, Veyron-Churlet R, Zanella-Cleon I, et al. complementation of the essential gene fabG1 of Mycobacterium tuberculosis by (2010) Phosphorylation of InhA inhibits mycolic acid biosynthesis and growth of Mycobacterium smegmatis fabG, but not Escherichia coli fabG. J Bacteriol 189: Mycobacterium tuberculosis. Mol Microbiol 78: 1591–1605. 3721–3728. 64. Khan S, Nagarajan SN, Parikh A, Samantaray S, Singh A, et al. (2010) 43. Banerjee A, Dubnau E, Quemard A, Balasubramanian V, Um KS, et al. (1994) Phosphorylation of enoyl-acyl carrier protein reductase InhA impacts mycobac- inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium terial growth and survival. J Biol Chem 285: 37860–37871. tuberculosis. Science 263: 227–230. 65. Corrales R, Molle V, Leiba J, Mourey L, de Chastellier C, et al. (2012) 44. Vilcheze C, Morbidoni HR, Weisbrod TR, Iwamoto H, Kuo M, et al. (2000) Phosphorylation of Mycobacterial PcaA Inhibits Mycolic Acid Cyclopropana- Inactivation of the inhA-encoded fatty acid synthase II (FASII) enoyl-acyl carrier tion: CONSEQUENCES FOR INTRACELLULAR SURVIVAL AND FOR protein reductase induces accumulation of the FASI end products and cell lysis PHAGOSOME MATURATION BLOCK. The Journal of biological chemistry of Mycobacterium smegmatis. J Bacteriol 182: 4059–4067. 287: 26187–26199. 45. Tahlan K, Wilson R, Kastrinsky DB, Arora K, Nair V, et al. (2012) SQ109 66. Rudner D, Pan Q, Losick R (2002) Evidence that subcellular localization of a Targets MmpL3, a Membrane Transporter of Trehalose Monomycolate bacterial membrane protein is achieved by diffusion and capture. Proceedings of Involved in Mycolic Acid Donation to the Cell Wall Core of Mycobacterium the National Academy of Sciences of the United States of America 99: 8701– tuberculosis. Antimicrob Agents Chemother 56: 1797–1809. 8706. 46. von Heijne G (1992) Membrane protein structure prediction. Hydrophobicity 67. Bardou F, Que´mard A, Dupont MA, Horn C, Marchal G, et al. (1996) Effects of analysis and the positive-inside rule. Journal of molecular biology 225: 487–494. isoniazid on ultrastructure of Mycobacterium aurum and Mycobacterium 47. Domenech P, Reed MB, Barry CE, 3rd (2005) Contribution of the tuberculosis and on production of secreted proteins. Antimicrobial Agents And Mycobacterium tuberculosis MmpL Protein Family to Virulence and Drug Chemotherapy 40: 2459–2467. Resistance. Infect Immun 73: 3492–3501. 68. Backus KM, Boshoff HI, Barry CS, Boutureira O, Patel MK, et al. (2011) Uptake of unnatural trehalose analogs as a reporter for Mycobacterium 48. Feilmeier BJ, Iseminger G, Schroeder D, Webber H, Phillips GJ (2000) Green tuberculosis. Nat Chem Biol 7: 228–235. fluorescent protein functions as a reporter for protein localization in Escherichia 69. Swarts BM, Holsclaw CM, Jewett JC, Alber M, Fox DM, et al. (2012) Probing coli. Journal of bacteriology 182: 4068–4076. the mycobacterial trehalome with bioorthogonal chemistry. J Am Chem Soc 49. He Z, De Buck J (2010) Localization of proteins in the cell wall of 134: 16123–16126. Mycobacterium avium subsp. paratuberculosis K10 by proteomic analysis. 70. Thanky NR, Young DB, Robertson BD (2007) Unusual features of the cell cycle Proteome Sci 8: 21. in mycobacteria: Polar-restricted growth and the snapping-model of cell division. 50. Winkler J, Seybert A, Ko¨nig L, Pruggnaller S, Haselmann U, et al. (2010) Tuberculosis (Edinb) 87: 231–236. Quantitative and spatio-temporal features of protein aggregation in Escherichia 71. van Kessel JC, Hatfull GF (2007) Recombineering in Mycobacterium coli and consequences on protein quality control and cellular ageing. The tuberculosis. Nat Methods 4: 147–152. EMBO journal 29: 910–923. 72. Stover CK, de la Cruz VF, Fuerst TR, Burlein JE, Benson LA, et al. (1991) New 51. Lindner A, Madden R, Demarez A, Stewart E, Taddei F (2008) Asymmetric use of BCG for recombinant vaccines. Nature 351: 456–460. segregation of protein aggregates is associated with cellular aging and 73. Heckman K, Pease L (2007) Gene splicing and mutagenesis by PCR-driven rejuvenation. Proceedings of the National Academy of Sciences of the United overlap extension. Nature protocols 2: 924–932. States of America 105: 3076–3081. 74. Drocourt D, Calmels T, Reynes J, Baron M, Tiraby G (1990) Cassettes of the 52. Makarov V, Manina G, Mikusova K, Mollmann U, Ryabova O, et al. (2009) Streptoalloteichus hindustanus ble gene for transformation of lower and higher Benzothiazinones Kill Mycobacterium tuberculosis by Blocking Arabinan eukaryotes to phleomycin resistance. Nucleic acids research 18: 4009. Synthesis. Science 324: 801–804. 75. Laval F, Laneelle MA, Deon C, Monsarrat B, Daffe M (2001) Accurate 53. Carlsson F, Joshi SA, Rangell L, Brown EJ (2009) Polar localization of virulence- molecular mass determination of mycolic acids by MALDI-TOF mass related Esx-1 secretion in mycobacteria. PLoS Pathog 5: e1000285. spectrometry. Anal Chem 73: 4537–4544.

PLOS ONE | www.plosone.org 15 May 2014 | Volume 9 | Issue 5 | e97148

Figure 1. Schematic representations of the TA types. (A) Type I: the toxin is mRNA and antitoxin is a sRNA, anti‐sense of the toxin mRNA. (B) Type II: toxin and antitoxin are proteins. (C) Type III: the RNA antitoxin inhibits toxicity by directly binding the toxic protein. (D) Type IV: the toxin protein inhibits the target molecules (orange), whereas the antitoxin molecules counteract these effects. (E) Type V: the antitoxin cleaves the mRNA of the small toxin‐encoding ORF. Toxin genes and proteins are depicted in red color, antitoxins (AT) in blue, DNA as sinus curves (Schuster and Bertram, 2013).

APPENDIX B

Construction of tools to study the expression of Rv1990c-1989c operon This study was part of the project “Role of Horizontally acquired toxin-antitoxin system Rv1990c-1989c in Mtb”. The major goal of this study was to develop and adapt the Gateway® system as a tool to investigate the expression of the Rv1990c-1989c operon. The Gateway® system approach was designed with the unique purpose of performing a high- throughput genetics screen. Advantages in the use of this approach include the easy and fast cloning of different ORFs into destination vectors without use of restriction and ligation enzymes. The entry clones permit easy shuttling of bacterial ORFs into fluorescent protein expression vectors for determination of subcellular localization of bacterial proteins in bacterial cells. This study was successfully carried out as detailed below.

1. BACKGROUND 1.1 Toxin-antitoxin (TA) systems in bacteria Horizontal gene transfer (HGT) is a major mechanism contributing to microbial evolution (Ochman et al., 2000). Bacteria are always facing environmental stresses, as demonstrated by the huge variety of their ecological niches. Through adaptive evolution they have developed multiple systems to ensure survival, including the toxin–antitoxin (TA) systems (Hayes, 2003; Gerdes et al., 2005; Hayes and Van Melderen, 2011). Prokaryotic TA systems are frequently found on plasmids or chromosomes both in bacteria and archaea. TA system usually comprise bicistronic operons encoding two small genes; one for a toxic component and a second for a cognate antitoxin (Hayes, 2003; Gerdes et al., 2005; Leplae et al., 2011). So far, five types of Toxin-Antitoxin loci have been identified (Fig. 1). • Type I, the antitoxin is a small RNA (sRNA) which can hybridize with the toxin mRNA to prevent protein synthesis (Fozo et al., 2010). • Type II, the antitoxin is a protein that binds to and inhibits the toxin protein (Leplae et al., 2011). • Type III, the antitoxin is an RNA, but rather than being an anti- sense RNA, it interacts directly with the toxin protein, preventing its activity (Blower et al., 2012).

Table 1. Cloning primers used for plasmid constructions. Construction Templates Primers Sequence 5’‐3’ pEN41A‐5'RFP RFP ORF pLAM12‐RFP Fw‐B4‐GRFP GGGGACAACTTTGTATAGAAAAGTTGTTACTTGTACAGCTCGTCCATGCCG Rv‐SD‐GRFP AGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG P1 prom pEN12A‐P1 Fw‐GFP‐SD‐P1 GCTCACCATGGTGAATCCTCCTCTCATTTCGGGCGGCGAATCTC Rev‐B1‐P1 GGGGACTGCTTTTTTGTACAAACTTGGGCGGCACGATCCGCGACGTG pEN12A‐Prom‐ GFP Frg P1990c Cosmid I35 Fw‐B1‐P1990c GGGGACAAGTTTGTACAAAAAAGCAGGCTGCTGCTGGCTCCCAGTAGAC Rv‐GFP‐P1990c GCTCACCATGGTGAATCCTCCTCCGGGCAGCCTGTCTTCCTTTTG Frg GFP pMV361‐GFP Fw‐P1990‐GFP CAAAAGGAAGACAGGCTGCCCGGAGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG Rv‐B2‐GRFP GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATG pEN41A‐ 5'mTurquoise Turquoise ORF pJYB 240‐mTurq Fw‐B4‐mTurq GGGGACAACTTTGTATAGAAAAGTTGTTACTTGTAGAGTTCGTCCATGCCG Rv‐SD‐mTurq AGGAGGATTCACCATGGTCAGCAAGGGTGAGGAACTC P1 prom pEN12A‐P1 Fw‐mTurq‐SD‐P1 GCTGACCATGGTGAATCCTCCTCTCATTTCGGGCGGCGAATCTC Rev‐B1‐P1 GGGGACTGCTTTTTTGTACAAACTTGGGCGGCACGATCCGCGACGTG pEN12A‐Prom‐ mVenus Frg P1990c Cosmid I35 Fw‐B1‐P1990c GGGGACAAGTTTGTACAAAAAAGCAGGCTGCTGCTGGCTCCCAGTAGAC Rv‐GFP‐P1990c GCTCACCATGGTGAATCCTCCTCCGGGCAGCCTGTCTTCCTTTTG Frg mVenus pJYB 234‐mVen Fw‐P1990‐GFP CAAAAGGAAGACAGGCTGCCCGGAGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG Rv‐B2‐GRFP GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATG Promoter region Δup pEN12A‐P1990c‐GFP Fw2‐B1‐P1990c GGGGACAAGTTTGTACAAAAAAGCAGGCTCCGAAACTGAATGCCCATCG Rv‐GFP‐P1990c GCTCACCATGGTGAATCCTCCTCCGGGCAGCCTGTCTTCCTTTTG Fw‐P1990‐GFP CAAAAGGAAGACAGGCTGCCCGGAGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG Rv‐B2‐GRFP GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATG Δdw pEN12A‐P1990c‐GFP Fw‐B1‐P1990c GGGGACAAGTTTGTACAAAAAAGCAGGCTGCTGCTGGCTCCCAGTAGAC Rv2‐GFP‐P1990c GCTCACCATGGTGAATCCTCCTCCCTATTGATTTGTCATGTATTATGACACGAACCGAGG Fw2‐P1990‐GFP ACATGACAAATCAATAGGGAGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG Rv‐B2‐GRFP GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATG ΔdwG pEN12A‐P1990c‐GFP Fw‐B1‐P1990c GGGGACAAGTTTGTACAAAAAAGCAGGCTGCTGCTGGCTCCCAGTAGAC Rv3‐GFP‐P1990c GCTCACCATGGTGAATCCTCCTCCCTATTGATTTGTCATGTGTTATGACACGAACCGAGG Fw2‐P1990‐GFP ACATGACAAATCAATAGGGAGGAGGATTCACCATGGTGAGCAAGGGCGAGGAG Rv‐B2‐GRFP GGGGACCACTTTGTACAAGAAAGCTGGGTTTACTTGTACAGCTCGTCCATG

APPENDIX B 81

• Type IV, both the toxin and and antitoxin are proteins, but they do not dirtectly interact with each other. The type IV antitoxin interferes with the binding of the toxin to its target, thereby preventing its activity (Masuda et al., 2012). • Type V, both the toxin and antitoxin are proteins. The antitoxin cleaves the toxin mRNA, preventing its translation (Wang et al., 2012).

Most bacterial species harbor several TA systems in their genome. Surprisingly, Mtb genome is particularly rich in such systems. Indeed, although only four TA systems have been identified in M. smegmatis, and none is predicted in the genome of M. marinum, at least 77 TA systems are predicted in the genome of Mtb H37Rv, of which33 have already been experimentally demonstrated, according to TADB (http://bioinfo-mml.sjtu.edu.cn/ TADB/), which is a web-based resource for Type II TA loci in bacteria and archaea (Shao et al., 2011). The operon Rv1990c-1989c of Mtb H37Rv encodes one of the predicted and so far not analyzed experimentally TA systems. Furthermore, neither the putative toxin Rv1989c nor the putative antitoxin Rv1990c belong to a family of proteins already characterized. These reasons prompted our team to study this system, and preliminary results available in the group when I started my PhD work had demonstrated that this operon encodes a bona fide TA system. My part of the work on this system was to construct and utilize genetic tools designed to follow the expression of the operon Rv1990c-1989c. I have adapted plasmids of the Gateway® system to express fluorescent protein reporters and used the resulting constructions to monitor the expression of the Rv1990c-1989c operon both in vitro and after infection of macrophages.

1.2 Gateway® cloning system The Gateway® technology is a universal cloning method based on the bacteriophage λ site-specific recombination system (Landy, 1989; Ptashne, 1992). Integration of the genome of λ into the chromosome of E. coli occurs through recombination between specific sites (att sites): a site on the phage genome, attP, and a site on the bacterial genome, attB. Recombination between attB and attP (thereafter named the BP reaction) is catalyzed by λ integrase (Int), in the presence of E. coli Integration Host Factor (IHF) protein. It generates hybrid sites, so-called attR and attL, for the sites present on the right and left sides of the

Table 2. Plasmids

E.coli Plasmid Resistance Source 1 DB3.1 ‐ Invitrogen 2 DB3.1 pDO221A Cm Amp/Carb Ehrt et al., 2005 3 DB3.1 pDO23A Cm Amp/Carb Ehrt et al., 2005 4 DB3.1 pDO41A Cm Amp/Carb Ehrt et al., 2005 5 MACH1 pEN12A‐P1 Amp/Carb Ehrt et al., 2005 6 MACH1 pEN41A‐T02 Amp/Carb Ehrt et al., 2005 7 DB3.1 pDE43‐MCS Cm Strep Ehrt et al., 2005 8 MACH1 pEN23A‐MluI Amp/Carb Ehrt et al., 2005 9 MACH1 pEN23A‐1990c Amp/Carb This study 10 MACH1 pEN23A‐1990c‐1989c Amp/Carb This study 11 MACH1 pEN41A‐RFP Amp/Carb This study 12 MACH1 pEN12A‐P1990c‐GFP Amp/Carb This study 13 MACH1 pDE43‐RFP‐GFP‐φ Cm Strep This study 14 MACH1 pDE43‐RFP‐GFP‐1990c Cm Strep This study 15 MACH1 pDE43‐RFP‐GFP‐1990c‐1989c Cm Strep This study 16 MACH1 pEN41A‐mTurquois Amp/Carb This study 17 MACH1 pEN12A‐P1990c‐mVenus Amp/Carb This study 18 MACH1 pDE43‐mTurq‐mVen‐φ Cm Strep This study 19 MACH1 pDE43‐mTurq‐mVen‐1990c Cm Strep This study 20 MACH1 pDE43‐mTurq‐mVen‐1990c‐1989c Cm Strep This study 21 MACH1 pDE43‐RFP‐Δup‐GFP‐φ Cm Strep This study 22 MACH1 pDE43‐RFP‐Δdw‐GFP‐φ Cm Strep This study 23 MACH1 pDE43‐RFP‐ΔdwG‐GFP‐φ Cm Strep This study 24 MACH1 pDE43‐RFP‐Δup‐GFP‐1990c Cm Strep This study 25 MACH1 pDE43‐RFP‐Δdw‐GFP‐1990c Cm Strep This study

APPENDIX B 82 integrated λ genome, respectively. Excision of the λ genome from the chromosome of E. coli occurs through recombination between attR and attL and necessitates not only λ Integrase and IHF, but also a second protein from λ, the excisionase (Xis). This reaction is thereafter named the LR reaction. All att sites contain a central 7 bp overlap region, which defines the integrase cleavage site (Hartley et al., 2000). Several variants of the att sites (named attB1, B2, B3 and B4 and attP1, P2, P3 and P4), generated by point mutations, are still active and exhibit a very high specificity (i.e. B1 recombines only with P1, and L1 only with R1) allowing the utilization of several of these sites in LR reactions. This is the basis of the MultiSite Gateway® Three-Fragment kit, using the four types of att sites to allow simultaneous recombination between three DNA fragments in a single LR reaction (see Figure 2B in results section). The Gateway cloning technology has been optimized with the use of destination vector backbones carrying a gene from the F plasmid, cdd, encoding a lethal toxin in E. coli, which ensures a very high efficiency of cloning. The enzymatic mixes necessary for the BP or LR reactions are commercially available (Invitrogen life Technologies) and the Gateway cloning system has been widely used to clone many types of DNA, such as PCR fragments, cDNA, or genomic DNA and is compatible with a large collection of vectors for other studies (Dupuy et al., 2004).

2. MATERIAL AND METHODS 2.1 Generation of Rv1990c-1989c operon expression clones using Gateway® recombination cloning technology Individual PCR reactions were performed in a 50 μl reaction volume using DNA template as shown in table 1. The PCR products were cloned into the Entry vectors pDO41A, pDO221A or pDO23A (Table 2) through BP recombination reactions, generating a series of entry clones, named pEN41A, pEN12A and pEN23A respectively (Table 2). When used in LR recombination reactions, pEN41A plasmids provide the 5’ elements of the final expression clone, whereas the pEN12A and pEN23A plasmids provide the central and 3’ element, respectively (Fig. 2B).

Figure 2. Gene organization and construction of Gateway® plasmids to investigate gene expression. (A) Genetic constructions to analyze expression of the Rv1990c‐1989c operon. (B) Schematic illustration of the Multisite Gateway® System: construction by modular cloning. Top row: BP recombination reactions allow construction of entry clones carrying the 5’, middle and 3’ modules, respectively. Middle row: a single reaction of recombination (LR cloning) allows assembly of the different modules on a destination vector. During the LR cloning step, the toxic ccd gene present on the destination vector is replaced by the multisite construction. Bottom row: final modular expression clone.

APPENDIX B 83

BP reaction was set up according to the manufacturer’s instructions: The reaction mixture contained: attB PCR product (20-50 fmoles) 1-7 µl MultiSite Entry vector (150 ng) 1 µl BP Clonase™ II enzyme mix 2 µl TE buffer to 10 µl (final volume)

The reaction mix were incubated at 25 °C for 1 h, then proteinase K 1 µl was added to the DNA samples and incubated at 37 °C for 10 min. BP reactions were used directly for bacterial transformation of E. coli MACH1 cells, and bacteria were plated on LB medium containing 100 μg/ml of ampicillin. To generate the final expression vector (Table 2), LR reaction mixture contained: Entry clones (supercoiled, 20-25 fmoles) pEN41A (carrying gene of interest) 1 µl pEN12A (carrying gene of interest) 1 µl pEN23A (carrying gene of interest) 1 µl pDE43-MCS (supercoiled, 60 ng/μl) 1 µl LR Clonase™ Plus enzyme mix 2 µl TE buffer to 10 µl (final volume) The reaction mix were incubated at 25 °C for 16 h (or overnight), then proteinase K 1 µl was added to the DNA samples and incubated at 37 °C for 10 min. LR reactions were used directly for bacterial transformation of E. coli MACH1 cells, and bacteria were plated on LB medium containing 25 μg/ml of streptomycin.

2.2 Cloning and expression See Materials and Methods in thesis.

2.3 Analysis of GFP expression I. Fluorescence intensity

M. smegmatis liquid cultures in early log phase (OD590 ∼ 0.5) were analyzed in a total volume of 200 μl in a 96-well black microtitre plates (Greiner Bio-One GmbH, Frickenhausen, Germany) with a microplate reader CLARIOstar® (BMG Labtech, Ortenberg, Germany). The

Figure 3. Analysis of Rv1990c‐1989c operon regulation. (A) Fluorescence microscopy was used to visualize M. smegmatis harboring the three plasmids, pDE43‐RFP‐GFP‐1990c‐1989c, pDE43‐RFP‐GFP‐1990c and pDE43‐RFP‐GFP‐φ, which produce RFP constructively (1st panel) and GFP (2nd panel) expressed under the control of the promoter of the operon. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. (B) Bar plots of total fluorescence of M. smegmatis expressing GFP, the experiments were repeated three times independently, with triplicates for each experiment. The fluorescence intensity was normalized to the cell density and expressed in RFUs. Data are averages and error bars represent standard deviations. The p values of indicated unpaired t‐tests are symbolized by asterisks (***, p <0.0001), (ns, non‐significant). (C) and (D): The same three plasmids were introduced in wild type Mtb (D) or in a Mtb strain deleted of the Rv1990c‐1989c operon (C). Bacterial cells were fixed with 4%PFA, and observed under large field fluorescence microscopy.

APPENDIX B 84 fluorescence spectra were analyzed with excitation and emission at 483-14 nm and 530-30 nm respectively. Acquisition of fluorescence (F) and OD590 of the sample was performed in triplicate of each culture. The data expressed in relative fluorescence units (RLU = F/OD590). The data were plotted using Graphpad Prism 5 software (GraphPad Software Inc.).

II. Fluorescence microscopy M. smegmatis harboring the expression clones were pelleted, concentrated (20X) in PBS, and spread (10 µl) on poly-L-lysine coated glass slides (PolyScience, Warrington, PA). For Mtb, the bacteria were fixed with 4 % formaldehyde in PBS for 2 h at room temperature and washed twice with 1xPBS. Slides were mounted with Dako mounting medium (Agilent Technologies, Santa Clara, CA) and observed with a large field Leica DMIRB fluorescence microscope at 63X magnification (Leica HCX Plan Apo x63/1.40-Oil). GFP (488 nm/509 nm) was revealed with a GFPET filter cube (excitation: BP 470/40; dichroic mirror, 495; emission: LP 525/50), Red Fluorescent Protein (RFP) (587 nm/610 nm) was visualized with a N2.1 Leica filter cube (excitation: BP 515-560; dichroic mirror, 580; emission LP 590), mTurquoise (434 nm/474 nm) was visualized with a CFP Leica filter cube (excitation: BP 436/20; dichroic mirror, 455; emission LP 480/40) and mVenus (515 nm/528 nm) was visualized with a YFP Leica filter cube (excitation: BP 500/20; dichroic mirror, 515; emission LP 535/30). Images were captured with a CoolSnap HQ2 CCD camera (1 pixel = 6.45 µm), acquired with MetaMorph 7.1.7 software and processed with ImageJ software (National Institutes of Health, Bethesda, MD). All different images were acquired with the same exposure time (200 ms). Image processing consisted in equivalent adjustments of brightness and contrast on complete images.

3. RESULTS AND PERSPECTIVES 3.1 Generation of expression clones by the Gateway®cloning system The Gateway® cloning system takes advantage of the site-specific recombination properties of bacteriophage λ (Hartley et al., 2000). Bacteriophage λ integrase protein recognizes compatible recombination sites (att sites), present in the Gateway® cloning vectors and, introduced on both sides of a DNA sequence of interest. In vitro recombination, using purified λ integrase, then allows for the cloning of this fragment of interest into a so-

Figure 4. Regulation of the Rv1990c‐1989c TA system. Co‐transcription of the antitoxin Rv1990c and toxin Rv1989c from the same promoter is autoregulated by antitoxin Rv1990c, either alone or in complex with the toxin Rv1989c.

APPENDIX B 85 called destination vector of the Gateway® cloning system. The group of S. Ehrt and D. Schnappinger, at Weill-Cornell College of Medicine, NY, USA, has adapted a series of plasmids based on the Gateway® system for cloning into mycobacteria. In order to measure the expression of operons of interest in M. smegmatis and Mtb, I constructed new plasmids for the Gateway® site specific recombinational cloning system based on two fluorescent reporter proteins: GFP and RFP. The Gateway® system allows construction of composite plasmids with 3 different modules. In the first module, I introduced the ORF encoding RFP, under the control of a constitutive promoter, P1. Then, RFP can be used as a standard. In the second module, DNA fragments carrying a promoter to be monitored upstream from the ORF encoding GFP can be inserted. Finally, in the third module, one can insert any gene whose product is supposed to modulate the expression of the tested promoter (Fig. 2A). We wanted to investigate the expression and mechanism of regulation of the Rv1990c- 1989c operon, encoding an as yet uncharacterized toxin/antitoxin system of Mtb. Therefore,

I constructed plasmids in which the PRv1990c promoter transcribes the GFP gene (Fig 2A).

Three different constructs were made, with the PRv1990c promoter transcribing either only GFP (GFP-φ), an operon composed of GFP followed by the antitoxin Rv1990c (GFP-1990c), or an operon with GFP, the antitoxin Rv1990c and the toxin Rv1989c (GFP-1990c-1989c) (Fig. 2A). The resulting plasmids, named pDE43-RFP-GFP-φ, pDE43-RFP-GFP-1990c and pDE43- RFP-GFP-1990c-1989c, respectively, were then introduced into M. smegmatis and Mtb by transformation. They stably integrate in single copy into the chromosome of the recipient strains, through recombination at the Att site of the L5 mycobacteriophage (glyV) catalized by the phage integrase, encoded by the plasmid. The resulting strains allowed for the monitoring of the expression of RFP (as a standard) and GFP to ultimately study the transcription initiating at the PRv1990c promoter.

3.2. Testing the Rv1990c-1989c operon expression I first performed Fluorescence microscopy of M. smegmatis harboring the expression vectors pDE43-RFP-GFP-1990c-1989c, pDE43-RFP-GFP-1990c or pDE43-RFP-GFP-φ. In these bacteria RFP is produced constructively, providing a marker for transformed bacterial cells, and GFP reflects transcription from the PRv1990c promoter. The results of these experiments

(Fig2) showed that the PRv1990c promoter was expressed at a sufficient level to yield green fluorescence in M. smegmatis (Fig. 3A-B) in the absence of Rv1990c and Rv1989c, whereas

Figure 5. Genetic dissection of the promoter region of Mtb Rv1990c‐1989c. (A) Primers designed for PCR amplification of different DNA fragments encompassing the promoter. (B) Expression of RFP reporter (1st panel) and GFP (2nd panel) expressed under the control of the various promoter regions upon transformation of the indicated derivatives of pDE43‐ RFP‐GFP‐φ in M. smegmatis. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. APPENDIX B 86 the presence of Rv1990c resulted in a much reduced expression, even more reduced in the presence of both Rv1990c and Rv1989c. The three reporter plasmids were also introduced in Mtb wild type strain (Fig. 3D) and in a Mtb strain deleted of the chromosomal copy of the Rv1990c-Rv1989c operon (Fig. 3C). Again, our data demonstrated that the expression of the promoter was strongly inhibited in the presence of Rv1990c and even more when both Rv1990c and Rv1989c were present. Altogether, these data demonstrated that the antitoxin Rv1990c represses its own promoter. This negative autoregulation appears even more efficient when both the toxin and the antitoxin are produced (Fig. 4). Similar autoregulation has already been observed in several type II Toxin/antitoxin systems.

3.3. Dissection of the regulatory region of the Rv1990c promoter

In the genetic constructions used in section 3.2, the PRv1990c promoter was carried on a 261-bp DNA fragment extending from oligonucleotide Fw1 to Rv1 (Fig. 5A). In order to localize more precisely the promoter and the operator sequence(s) necessary for the expression and regulation of the Rv1990c-Rv1989c operon, I generated variants of the promoter region, that were inserted in front of the GFP reporter, in plasmids similar to those described in section 3.1. First, I constructed shorter versions of the promoter fragment, deleting either the upstream (Δup) of downstream regions (Δdw) (Fig. 5A) and these shorter promoter regions were introduced in front of the GFP gene in pDE43 plasmids either alone or with Rv1990c. In addition, we identified a putative −10 element upstream of the Rv1990c ORF that was suspected to be the actual -10 element of the PRv1990c promoter (Fig. 5A). To confirm this prediction, I introduced a point mutation (A to G) modifying this -10 element. The final constructions, pDE43-RFP-Δup-GFP-φ, pDE43-RFP-Δdw-GFP-φ, pDE43-RFP-ΔdwG-GFP-φ, pDE43-RFP-Δup-GFP-1990c and pDE43-RFP-Δdw-GFP-1990c, were transformed in M. smegmatis and promoter activities were measured using the fluorescent signal of RFP and GFP by fluorescent microscopy. As shown in Fig. 5B and 6A, and summarized in Fig. 6B, the results of these experiments confirmed that the suspected -10 element is necessary for the

PRv1990c promoter activity. Indeed, the point mutation changing a consensus position into a non-consensus one abolished the activity of the promoter. In addition, we demonstrated that deleting the sequences upstream from this promoter affected neither the activity of the promoter nor its repression by Rv1990c. In contrast, although the deletion of the sequences

Figure 6. Autoregulation of the variants of the promoter region of Mtb Rv1990c‐1989c in M. smegmatis. (A) Expression of RFP reporter (1st panel) and GFP (2nd panel) expressed under the control of the promoter of the operon. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. (B) Summary of the results (see section 3.3).

APPENDIX B 87 downstream from the promoter did not affect its expression in the absence of Rv1990c (Fig. 5B), it abolished the repression by Rv1990c (Fig. 5A). Therefore, at least part of the operator (Rv1990c binding site) necessary for the autorepression lies between the limits of Rv1 and Rv2. We note that a direct repeat of the sequence GACAA is present in this region and is in part deleted with the Rv2 construction. This motif is likely to constitute the binding site of Rv1990c.

3.4 Expression of the Rv1990c-1989c operon is induced in a variety of stress conditions. Several studies have demonstrated that transcription of the TA operon is activated by various stressful conditions including nutrient starvation (Betts et al., 2002), hypoxia (Ramage et al., 2009; Voskuil et al., 2003) and in host phagocytes (Tailleux et al., 2008). To study the effect of stress condition on expression of Rv1990c-Rv1989c, a Mtb strain harboring pDE43-RFP-GFP-1990c-1989c has been used to evaluate the expression of the operon. Our results showed that the operon is upregulated during nutrient starvation

(suspension in PBS buffer) and oxidative stress (H2O2 and NO), as illustrated in Figure 7. Additional data demonstrated that neither acidic nor basic pH conditions resulted in induction of the operon (data not shown). Thus, under normal growth conditions, the antitoxin, either alone or bound to its cognate toxin, is repressing its own transcription and, in response to specific stress conditions, this repression is relieved, allowing induction of the operon expression. In addition, we constructed a Mtb strain derivative carrying a variant of pDE43-RFP-GFP-1990c-1989c, but with the blue fluorescent protein mTurquoise replacing RFP and the yellow fluorescent protein mVenus replacing GFP. When this strain, Mtb/pDE43- mTurq-mVenus-1990c-1989c, was used to infect mouse and human derived macrophages, we observed that the expression of mVenus was very low in the bacterial cells at the time of infection but was induced upon 1 or two days after infection (Fig. 8), demonstrating that expression of the Rv1990c-1989c operon is induced by stressful conditions encountered in the host. This suggests that the Rv1990c-1989c operon of Mtb may govern cell division decisions during infection and could potentially participate in the establishment of the physiological state of persistent bacterial cells.

Figure 7. Effect of stress conditions on expression of the Rv1990c‐1989c operon in Mtb. Fluorescence microscopy was used to visualize Mtb harboring pDE43‐mTurq‐mVenus‐1990c‐1989c. mTurquoise is produced constructively (1st panel) and mVenus (2nd panel) is expressed under the control of the promoter of the operon. (A) Nutrient starvation (Day 0, 1, 2 and 5 in PBS buffer). (B) Oxidative stress with NO (0.5 mM, day 1) or H2O2 (5 mM, day 3). Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X.

Figure 8. Expression of the Rv1990c‐1989c operon is induced upon infection of macrophages. Fluorescent microscopy image showing the induction of Rv1990c‐1989c operon after 24 h in both mouse and human macrophages. mTurquoise is produced constructively (1st panel) and mVenus (2nd panel) expressed under the control of the promoter of the operon. Fluorescent microscopy was performed using large field Leica DMIRB, magnification 630X. APPENDIX B 88

In summary, I created a reporter system for detection of expression of any gene of interest in mycobacteria by the Gateway® cloning system. I validated this reporter system by investigating the regulation of the Rv1990c-1989c operon of Mtb. I could demonstrate the autoregulation of the operon that belongs to the type II family of Toxin/antitoxin systems, and performed a genetic dissection of its promoter and operator sequences. Finally, I demonstrated that this operon is upregulated upon exposure to stressful conditions, such as nutrient starvation, oxidative stress and during macrophage infection. One can envision that in vivo tests in animal models will be a strong evidence to support this study.

APPENDIX B: REFERENCES

Betts, J.C., Lukey, P.T., Robb, L.C., McAdam, R.A., and Duncan, K. (2002). Evaluation of a nutrient starvation model of Mycobacterium tuberculosis persistence by gene and protein expression profiling. Mol Microbiol 43, 717-731.

Blower, T.R., Short, F.L., Rao, F., Mizuguchi, K., Pei, X.Y., Fineran, P.C., Luisi, B.F., and Salmond, G.P. (2012). Identification and classification of bacterial Type III toxin-antitoxin systems encoded in chromosomal and plasmid genomes. Nucleic Acids Res 40, 6158-6173.

Dupuy, D., Li, Q.R., Deplancke, B., Boxem, M., Hao, T., Lamesch, P., Sequerra, R., Bosak, S., Doucette-Stamm, L., Hope, I.A., et al. (2004). A first version of the Caenorhabditis elegans Promoterome. Genome Res 14, 2169-2175.

Fozo, E.M., Makarova, K.S., Shabalina, S.A., Yutin, N., Koonin, E.V., and Storz, G. (2010). Abundance of type I toxin-antitoxin systems in bacteria: searches for new candidates and discovery of novel families. Nucleic Acids Res 38, 3743-3759.

Gerdes, K., Christensen, S.K., and Lobner-Olesen, A. (2005). Prokaryotic toxin-antitoxin stress response loci. Nat Rev Microbiol 3, 371-382.

Hartley, J.L., Temple, G.F., and Brasch, M.A. (2000). DNA cloning using in vitro site-specific recombination. Genome Res 10, 1788-1795.

Hayes, F. (2003). Toxins-antitoxins: plasmid maintenance, programmed cell death, and cell cycle arrest. Science 301, 1496-1499.

Hayes, F., and Van Melderen, L. (2011). Toxins-antitoxins: diversity, evolution and function. Crit Rev Biochem Mol Biol 46, 386-408.

Landy, A. (1989). Dynamic, structural, and regulatory aspects of lambda site-specific recombination. Annu Rev Biochem 58, 913-949.

Leplae, R., Geeraerts, D., Hallez, R., Guglielmini, J., Dreze, P., and Van Melderen, L. (2011). Diversity of bacterial type II toxin-antitoxin systems: a comprehensive search and functional analysis of novel families. Nucleic Acids Res 39, 5513-5525.

Masuda, H., Tan, Q., Awano, N., Wu, K.P., and Inouye, M. (2012). YeeU enhances the bundling of cytoskeletal polymers of MreB and FtsZ, antagonizing the CbtA (YeeV) toxicity in Escherichia coli. Mol Microbiol 84, 979-989.

APPENDIX B: REFERENCES 90

Ochman, H., Lawrence, J.G., and Groisman, E.A. (2000). Lateral gene transfer and the nature of bacterial innovation. Nature 405, 299-304.

Ptashne, M. (1992). A Genetic Switch: Phage Lambda and Higher Organisms (Cambridge, Mass, Cell Press and Blackwell Scientific Publications).

Ramage, H.R., Connolly, L.E., and Cox, J.S. (2009). Comprehensive functional analysis of Mycobacterium tuberculosis toxin-antitoxin systems: implications for pathogenesis, stress responses, and evolution. PLoS Genet 5, e1000767.

Schuster, C.F., and Bertram, R. (2013). Toxin-antitoxin systems are ubiquitous and versatile modulators of prokaryotic cell fate. FEMS Microbiol Lett 340, 73-85.

Shao, Y., Harrison, E.M., Bi, D., Tai, C., He, X., Ou, H.Y., Rajakumar, K., and Deng, Z. (2011). TADB: a web-based resource for Type 2 toxin-antitoxin loci in bacteria and archaea. Nucleic Acids Res 39, D606-611.

Tailleux, L., Waddell, S.J., Pelizzola, M., Mortellaro, A., Withers, M., Tanne, A., Castagnoli, P.R., Gicquel, B., Stoker, N.G., Butcher, P.D., et al. (2008). Probing host pathogen crosstalk by transcriptional profiling of both Mycobacterium tuberculosis and infected human dendritic cells and macrophages. PLoS One 3, e1403.

Voskuil, M.I., Schnappinger, D., Visconti, K.C., Harrell, M.I., Dolganov, G.M., Sherman, D.R., and Schoolnik, G.K. (2003). Inhibition of respiration by nitric oxide induces a Mycobacterium tuberculosis dormancy program. J Exp Med 198, 705-713.

Wang, X., Lord, D.M., Cheng, H.Y., Osbourne, D.O., Hong, S.H., Sanchez-Torres, V., Quiroga, C., Zheng, K., Herrmann, T., Peti, W., et al. (2012). A new type V toxin-antitoxin system where mRNA for toxin GhoT is cleaved by antitoxin GhoS. Nat Chem Biol 8, 855-861.

Auteur : Kanjana NUKDEE

Titre : Identification et caracterisation de nucleomodulines putatives chez la bactérie Mycobacterium tuberculosis

Directeur de thèse : Pr Claude GUTIERREZ

Lieu et date de soutenance : Salle Fernand Gallais (LCC, 205 route de Narbonne, 31077 Toulouse, FRANCE), le 10 décembre 2015

Résumé : Les nucléomodulines sont des protéines produites par des bactéries parasites intracellulaires et qui sont importées dans le noyau des cellules infectées pour y moduler l’expression génique et contribuer ainsi à la virulence de la bactéries. L’identification de nucléomodulines chez plusieurs espèces de bactéries pathogènes a fait émerger ce concept comme une stratégie supplémentaire utilisée par les parasites intracellulaires pour contourner les moyens de défense de l’hôte. Le but de cette thèse était d’identifier et d’analyser d’éventuelles nucléomodulines produites par l’agent étiologique de la tuberculose, la bactérie Mycobacterium tuberculosis (Mtb). Nous avons tout d’abord analysé le génome de Mtb, à la recherche de gènes dont les produits portent des séquences analogues aux séquences de localisation nucléaire des eucaryotes (NLS). Nous avons pu identifier deux gènes de Mtb, Rv0229c et Rv3876, qui codent des protéines sécrétées dans le milieu de culture et qui se localisent dans le noyau lorsqu’elles sont exprimées dans des cellules épithéliales ou dans des macrophages murins ou humains. Les NLS de ces deux protéines ont été identifiées et leur modification abolit la localisation nucléaire dans les cellules eucaryotes. Le gène Rv0229c est trouvé spécifiquement dans le génome des espèces pathogènes du complexe Mtb. Ce gène semble avoir été acquis récemment par l’ancêtre de Mtb, via un transfert génétique horizontal. Rv3876 est plus généralement distribué chez les mycobactéries, et appartient à une région génomique codant un système de sécrétion type VII, ESX1, essentiel pour la virulence de Mtb. Les travaux en cours visent à analyser la dynamique de ces deux protéines au cours de l’infection de macrophages ou de modèles animaux, et leur rôle en tant que modulateurs de l’expression génique des cellules infectées et dans la virulence de Mtb.

Abstract : The nuclear targeting of bacterial proteins that modify host cell gene expression, the so-called nucleomodulins, has emerged as a novel mechanism contributing to virulence of several intracellular pathogens. The goal of this study was to identify nucleomodulins produced by Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis (TB), and to investigate their role upon infection of the host. We first performed a screening of Mtb genome in search of genes encoding proteins with putative eukaryotic-like nuclear localization signals (NLS). We identified two genes of Mtb, Rv0229c and Rv3876, encoding proteins that are secreted in the medium by Mtb and are localized into the nucleus when expressed in epithelial cells or in human or murine macrophages. The NLSs of these two proteins were identified and found to be essential for their nuclear localization. The gene Rv0229c, a putative RNase, is present only in pathogen species of the Mtb complex and seems to have been recently acquired by horizontal gene transfer (HGT). Rv3876 appears more widely distributed in mycobacteria, and belongs to a chromosomal region encoding proteins of the type VII secretion system ESX1, essential for virulence. Ongoing studies are currently investigating the dynamics of these proteins upon infection of host cells, and their putative role in the modulation of host cell gene expression and Mtb virulence.

Mots-clés : Tuberculose, Mycobacterium tuberculosis, nucleomodulines, NLS, transfert génétique horizontal

Discipline : Microbiologie

Intitulé du laboratoire : Institut de Pharmacologie et de Biologie Structurale, CNRS-UMR 5089, 205 route de Narbonne, 31077 Toulouse, FRANCE