MODULATION OF PROTEIN FUNCTION THROUGH OPTOGENETIC REGULATION
OF SUBCELLULAR PROTEIN DISTRIBUTION
by
JESSICA IRENE SPILTOIR
B.S., University of Vermont, 2009
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Pharmacology Program
2017 This thesis for the Doctor of Philosophy degree by
Jessica Irene Spiltoir
has been approved for the
Pharmacology Program
by
J. David Port, Chair
Joshua C. Black
Mair E. A. Churchill
Patricia A. Ernst
Aaron M. Johnson
Chandra L. Tucker, Advisor
Date: May 19, 2017
ii Spiltoir, Jessica Irene (Ph.D., Pharmacology)
Modulation of Protein Function through Optogenetic Regulation of Subcellular Protein
Distribution
Thesis directed by Associate Professor Chandra L. Tucker
ABSTRACT
The field of optogenetics involves using light responsive proteins to regulate signaling events in order to probe cellular processes. Optogenetic tools are engineered based on naturally occurring photosensory proteins, and they have been shown to regulate biological processes such as ion channel activation, transcription, enzyme activity, signaling events, and many more. Scientists use these tools to harness the benefits of light, which offers superior spatial and temporal precision over traditional chemical means of control. In this thesis, I focus on using optogenetic tools to modulate protein function in subcellular compartments by regulating protein localization.
It has been shown many times that a protein’s functional efficiency is directly connected to its microenvironment. By controlling protein localization using optogenetic tools, it is hypothesized that a protein’s function can be altered using light illumination. Two optogenetic approaches were taken in order to modulate function: first, using light we sequestered proteins out of the cytosol in order to block their function, and second, we facilitated redistribution of the proteins in the nucleus to specific locations to stimulate or block function with light.
iii To sequester proteins from their activity in the cytosol, we synthesized a protein tag
with a engineered a short peroxisomal targeting sequence in the Jα-helix of the LOV domain, a blue light responsive photoreceptor. In the dark, the Jα-helix is docked to the core of the
LOV domain but when it is illuminated the Jα-helix unwinds and reveals the peroxisomal targeting sequence triggering peroxisomal trafficking of tagged proteins. This optogenetic tag was fused to cytosolic GFP in order to mediate trafficking of GFP into peroxisomes with light. This novel tag can be used to reduce protein activity in the cytosol and also to explore components of peroxisomal trafficking, which can lead to life-threatening peroxisomal biogenesis disorders when trafficking goes awry.
After exploring regulation of cytosolic proteins, we switched our focus to nuclear protein regulation. The compartmental switch led us to discover that fusion of photoreceptor cryptochrome 2 (CRY2) to a dimeric transcription factor induces clearing of the fusion protein in the nucleus upon illumination. This clearing phenotype is associated with a significant reduction of transcriptional activity, allowing us to develop an optogenetic tool to abrogate transcription using light. Reducing the multimeric state of CRY2-fused transcription factors retains the protein in the nucleus. We deleted the dimerization motif of a Gal4 DNA binding domain fused to CRY2 and co-expressed it with an activation domain fused to CIB1.
Illumination stimulates the interaction between CRY2 and CIB1 reconstituting a split Gal4 transcription factor at the GalUAS promoter to activate transcription. Thus, with the developed CRY2 optogenetic tools described, illumination can reduce or activate transcription.
The form and content of this abstract are approved. I recommend its publication.
Approved: Chandra L. Tucker
iv DEDICATION
To the main pillar in my life, my mom, Eileen O’Connor Spiltoir.
Without your persistent love and encouragement many of my life’s achievements would
never have blossomed. You are my angel on earth.
And to my father, Thomas Spiltoir, who inspires a sense of adventure and wonder in all
aspects of life. Let’s hit the trail.
v ACKNOWLEDGEMENTS
I would first like to thank my advisor, Chandra Tucker, for taking me in when I was adrift and having the patience to guide me through this portion of my scientific journey.
Second, I would like to thank Tim McKinsey for giving me a solid foundation on which to complete the rest of my thesis research. My range of knowledge and techniques would not be so broadly applicable without both experiences.
I would also like to recognize Jana Kraft, who inspired me from the beginning and supported me throughout. I would never have applied my life to research without your encouragement. And cheers to all the Lock Lab and AGTC co-workers for making research fun and exciting, which solidified the desire to pursue my degree.
There is just not enough space for my Pharmacology friends: Lyn Gatchalian, who has braved many an adventure with me through the past five years; Wallace Liu, who has always been a friendly face when I needed one; Rachel Fickes, whose kindhearted soul has kept me afloat; Kelsey Barcomb, for always being up for anything fun; Justine Masselli Tait, who has shown me how to ride the waves of life with humor; Ashley Bourke, whose passion is a light for all those around her; Sarah Haegar, who has helped me to the end with a listening ear and walk to the pub; and to all the other students who shared in happy times- thank you. Liz Bowen also deserves a gigantic thank you for everything she has done to support me through graduate school academically and personally.
A thank you goes out to everyone present in the Tucker and McKinsey Labs during my stay. Especially Gopal Pathak, who started the transcription project and taught me the finer points of brewing; Tiffany Cung, who made Thursdays my favorite day in lab; Qi Liu,
vi who is the best source for conversations on bear species and for Milano cookies when you need one; and Brad Ferguson, who taught me many techniques and so much more. Special thanks to Sarah Williams and Kim Demos-Davies, who not only shared scientific knowledge but also friendship and many cherished memories.
A huge shout out to: the Women in STEM officers especially Sara Flubacher and
Abigail Armstrong, who helped make Ashley and my vision a reality; the campus Courage
Classic team for giving me a reason to ride and keeping me in shape when graduate school gave me a reason to binge; and Ruff Rescue for giving me the best foster dogs so I wouldn’t feel the hole that Ollie left behind so fiercely.
There were many whose support was felt from far away and also through visits that lifted my spirits. Thanks to my friends from home that kept me grounded, motivated, and kept my supply of the best maple syrup afloat, Tiffany French and Julia Hobson, and to
Stacey Monahan, whose move to Denver could not have been at a better time. My love and thanks to those I am connected to for eternity, my sisters, Megan and Becca, and my dog,
Ollie.
Lastly, the Mahaffey family, who I could never thank enough for sharing their time me throughout my stay in Colorado. Jenny, you have been a wonderful coworker, the best neighbor, and you are simply an irreplaceable friend. You made it feel like I always had family across the street.
vii TABLE OF CONTENTS
CHAPTER Page
I. INTRODUCTION 1
Protein Function as it Relates to Localization 1
Optogenetics Overview 2
History of Soluble Optogenetic Systems 5
Considerations for Photoreceptors when Designing Optogenetic Tools 8
Soluble Optogenetic Photoreceptors 10
LOV Domains 10
AsLOV2 12
FKF1 13
Aureochrome 14
VVD 14
EL222 15
Cryptochromes 15
Phytochromes 17
Bacterial Phytochromes 18
UVR8 19
Dronpa 20
BLUF Domains 21
Applications of Soluble Optogenetic Systems 21
Recruitment to Specific Cellular Locations 28
Reconstituting Split Proteins 30 Reconstituting Split Transcription Factors 30
Reconstituting Split Enzymes 31
Inducing Protein Interactions 32
Caging/Sequestering and Releasing Proteins 33
Caging 33
Homodimerization Sequestering 34
Heterodimerization Sequestering 34
Thesis Statement 35
II. OPTICAL CONTROL OF PEROXISOMAL TRAFFICKING 37
Introduction 37
Peroxisomes and Peroxisomal Trafficking 37
Optogenetic Peroxisomal Trafficking Tool Design 40
LOV Domain Mutations That Alter Kinetics 41
Materials and Methods 43
Strains and Plasmid Construction 43
Indirect Immunofluorescence 45
Microscopy and Image Analysis 45
Cell Culture and Live Cell Imaging 46
Yeast Studies 47
Results 47
Optogenetic Regulation of Peroxisomal Trafficking 47
Tracking Peroxisomal Import Kinetics with Live-Cell Imaging 49
Peroxisomal Translocation is Due to Light-Dependent Binding to PEX5 51
ix Comparison of Peroxisomal Import with Mutants and Fluorescent 51
Reporters on LOV-PTS1
Cargo Protein Stays Relatively Stable at Peroxisome After 54
Trafficking
Peroxisomal Sequestering of a Cytosolic Reporter Protein 56
The LOV-PTS1 Tag is Unable to Traffic Rapamycin-Induced 56
Dimers into Peroxisomes
LOV-PTS1 with Actin Nucleating Factor, VCA, is Unable to Traffic 58
To Peroxisomes
Discussion 60
III. REGULATING TRANSCRIPTION WITH OPTOGENETICS 62
Introduction 62
History of Regulating Transcription 62
Regulating Transcription with CRY2 65
Materials and Methods 66
Constructs 66
Cell Culture and Light Treatment Conditions 69
Luciferase Assays 70
Quantitative RT-PCR 71
CRISPR/dCAS9-VP64 Studies 71
Immunoblotting 72
Immunofluorescence Staining 73
Imaging 73
x Image Analysis 74
Zebrafish Studies 75
Results 77
CRY2-Fused Transcription Factors are Redistributed Correlating 77
With Functional Loss
Dimerization Domain is Required for Nuclear Depletion and Functional 81
Loss
An Optimized Light-Activated System for Transcriptional Control 83
Light-Mediated Transcriptional Reduction 87
Utilizing the Light-Dependent Depletion System 93
Dual Optogenetic Control to Deplete Protein Robustly with Light 93
Discussion 94
IV. CRYPTOCHROME 2 IN THE MAMMALIAN NUCLEUS 97
Introduction 97
Materials and Methods 99
Constructs 99
Site Directed Mutagenesis 100
Cell Culture and Live Cell Imaging 100
Immunofluorescence Staining 100
Microscopy and Image Analysis 101
Fractionation 101
Immunoprecipitation 102
xi Immunoblotting 103
Mass Spectrometry 103
Results 105
Inconsistent PML Nuclear Body Colocalization 105
Tracking Light Stimulated Nuclear Clearing 106
Post-Translational Modification Effects on Nuclear Clearing 110
Pharmacological Inhibition Shows Translation is Important for Nuclear 115
Clearing
Discussion 118
V. CONCLUSIONS AND FUTURE DIRECTIONS 121
Perspective 121
Summary of Major Findings 121
Chapter II- Optical Control of Peroxisomal Trafficking 122
Chapter III-Regulating Transcription with Optogenetics 123
Chapter IV-Cryptochrome2 in the Mammalian Nucleus 126
Future Directions 127
REFERENCES 133
APPENDIX 149
A. BET Acetyl-lysine Binding Proteins Control Pathological Cardiac 149
Hypterophy
xii LIST OF TABLES
TABLE Page
I-1 Considerations of soluble photoreceptors used in optogenetic systems 11
I-2 Applications of soluble photoreceptor technologies 24
II-1 Primers for construct cloning 44
III-1 Construct characteristics 67
III-2 Toxicity testing in UAS-GFP embryos 92
IV-1 Percent of total fragments that had post-translational modifications 115
A-1 Primer sequences used for qPCR and chromatin immunoprecipitation. 153
LIST OF FIGURES
FIGURE Page
I-1 Optogenetic modes of using natural properties of photosensory proteins 4
I-2 Photodimerizer approaches to regulate protein function 22
II-1 Peroxisomal targeting with PTS1 or PTS2 39
II-2 Model of optogenetic peroxisomal trafficking tool 42
II-3 Optogenetic regulation of peroxisomal trafficking 48
II-4 Kinetics of photostimulated peroxisomal import in HeLa cells over four hours 50
II-5 Peroxisomal translocation is due to light-dependent binding to Pex5 52
II-6 Varying levels of peroxisomal localization of LOV-PTS1 tag mutants and 53
fluorescent reporters
II-7 Peroxisomal cargo protein is relatively stable over 24 hours 55
II-8 Light-triggered depletion of a cytosolic reporter protein 55
II-9 Rapamycin-induced dimerization blocks light-triggered peroxisomal import of 57
GFP-LOV-PTS1
II-10 Fusion of VCA to LOV-PTS1 tagged protein disrupts actin cytoskeleton fibers 59
and blocks peroxisomal import
III-1 Optogenetic split transcription tool is unable to induce activity with light 64
III-2 Nuclear loss is seen with GalBD-CRY2 but not with other CRY2 fusion proteins 76
III-3 CRY2-fused transcription factors are cleared from the nucleus with light 78
correlating with functional loss
xiv III-4 A CRY2-fused dimerization domain is required for light-dependent nuclear 80
depletion and functional loss of activity
III-5 Addition of a tetramerizing dsRed domain restores light dependent functional 82
loss and nuclear clearing to CRY-Gal∆DD-VP16
III-6 Optimization of split CRY2/CIB1 transcriptional system 84
III-7 Light-mediated reduction of transcription using optimized CRY-GalVP16l 88
III-8 Optimization of CRY-tTA light-mediated reduction system 91
III-9 Light regulation of endogenous IL1RN transcription using a CRISPR/ dCAS9- 92
based approach
III-10 In vivo testing of CRY-GalVP16l in GalUAS-GFP reporter zebrafish embryos 92
IV-1 Staining of PML with light induced clusters showed inconsistent co-localization 107
IV-2 No Light Dependent Protein Reduction Observed in the Nucleus 109
IV-3 Photoconvertible mEOS3.2 tracks nuclear pool during light stimulation 111
IV-4 Phosphorylation and ubiquitin marks detected with mass spectrometry 113
IV-5 Loss of individual phosphorylation or ubiquitination PTMs are not sufficient 115
to block nuclear clearing
IV-6 Cycloheximide blocks light triggered nuclear clearing 116
A-1 A small molecule BET protein inhibitor blocks cardiac hypertrophy 158
A-2 JQ1 blocks induction of fetal gene program in phenylephrine treated neonatal 160
rat ventricular myocytes
A-3 BET family member mRNA expression in purified cardiac myocytes and 161
fibroblasts
xv A-4 Chromatin immunoprecipitation of NRVM lysates show BRD4 and Phospho 163
-RNA Polymerase II enrichment at ANF transcription start site with PE
stimulation
A-5 JQ1 suppresses left ventricular hypertrophy in response to pressure overload 164
A-6 Model for the regulation of cardiac hypertrophy by BRD4 167
xvi LIST OF ABBREVIATIONS
Abbreviation Meaning
AKT Protein Kinase B
AsLOV2 Avena sativa Phototropin1 LOV2 domain
BLUF Blue-Light Using FAD
CIB1 Cryptochrome Interacting Basic-Helix-Loop-Helix 1
COP1 Constitutively Photomorphogenic 1
CRM1 Chromosome Region Maintenance 1
CRY Cryptochrome
CRY2 Cryptochrome 2
CTD C-terminal Tail Domain
FAD Flavin Adenine Dinucleotide
FKF1 Flavin-Binding, Kelch Repeat, F-Box 1
FMN Flavin Mononucleotide
FP Fluorescent Protein
GEF Guanine Nucleotide Exchange Factor
GFP Green Fluorescent Protein
GI Gigantea
iLID Improved Light-Inducing Dimer
IP Immunoprecipitate
LOV Domain Light, Oxygen, Voltage Domain
mCh mCherry Fluorescent Protein
xvii NBs Nuclear Bodies
NES Nuclear Export Signal
NLS Nuclear Localization Signal
nMag Negative Magnet
PAS Period-ARNT-Singleminded
PCB Phycocyanobilin
PEX Peroxisomal Biogenesis
Phy Phytochrome
PHR Photolyase Homology Region
PIF Phytochrome Interacting Factor
PΦB Phytochromobilin pMag Positive Magnet
PML Progressive Multifocal Leukoencephalopathy
PTS1 Peroxisomal targeting sequence 1
PTS2 Peroxisomal targeting sequence 2
TULIP Tunable Light-Controlled Interacting Protein
UVR8 UV-B Resistance 8
VMA Vacuolar Membrane ATPase
VVD Vivid
WASP Wiskott-Aldrich Syndrome Protein
Zdk Zdark
xviii Appendix Abbreviations
Abbreviation Meaning
α-MHC Alpha-myosin heavy chain
ANF Atrial natriuretic factor
ARVF Adult rat ventricular fibroblast
ARVM Adult rat ventricular myocyte
β-MHC Beta-myosin heavy chain
BET Bromodomain and extraterminal domain
BNP Brain natriuretic peptide
BRD Bromodomain
CDK9 Cyclin-dependent kinase 9
ChIP Chromatin immunoprecipitation
HAT Histone acetyltransferase
HDAC Histone deacetylase
LV Left ventricle
LVH Left ventricular hypertrophy
NRVF Neonatal rat ventricular fibroblast
NRVM Neonatal rat ventricular myocyte
P-TEFb Positive transcriptional elongation factor b
PE Phenylephrine
PKC Protein kinase C
PMA Phorbol-12-myristate-13-acetate
RV Right ventricle
xix SERCA2a Sarcoendoplasmic reticulum calcium ATPase 2a
TAC Transverse aortic constriction
TSS Transcription start site
xx CHAPTER I
INTRODUCTION1
Protein Function as it Relates to Localization
Eukaryotic cells are highly compartmentalized, with structural barriers that create functionally distinct regions of the cell. Each region has its own biochemical activities that require specific local environments for proteins to function optimally. Consequently, protein localization is critical for protein function, and when there is abnormal localization the protein may not function optimally.
Membrane bound organelles are one of the cellular barriers that create microenvironments within the cell. These endomembrane systems differ in their local environment in characteristics such as pH, substrate availability, protein concentration and diversity, and morphology. These factors lead to changes in protein function, for instance lysosomal acid hydrolases are inactive at neutral pH but are most effective at a pH of about 5, which is optimal due to the acidity of that specific organelle. Compartmentalization not only increases efficiency by providing the right conditions for proteins, but it also localizes proteins and substrates in the same microenvironment to enhance reactions. Organelle structures are not the only spatial regulation in the cell. Protein recruitment to specific subcellular locations can act similarly to membrane barriers, for example plasma membrane recruitment of AKT where it gets activated by membrane bound phosphoinositide-dependent kinases. These examples demonstrate how specificity and protein function are enhanced by spatial compartmentalization.
1 A portion of this introduction is under review for a book chapter in Optogenetics: Spiltoir and Tucker © Wiley-VCH
1 Endomembrane systems provide a protein trafficking challenge to ensure proper localization after translation. The signal for localization of a protein is generally encoded in the protein’s sequence, but proteins can also be post-translationally modified to alter their localization.
Peroxisomal proteins have specific sequences at their N- and C- termini to recruit import receptor proteins to shuttle them to the peroxisomal membrane for import where they stay.
Transcription factors have been well characterized for their change in translocation between the cytosol and nucleus. External stimuli signal cytosolic transcription factors to translocate to the nucleus through processes such as dephosphorylation or phosphorylation events that unmask a nuclear localization signal or hide a nuclear export signal respectively. There are also examples of localization facilitated through anchoring to/release of cytoplasmic proteins and modulation of the nuclear import machinery to control transcription factors. The cell has established specific signals that allow proteins to traffic to the proper location and bypass the barriers of the cell.
The experiments in this thesis are set up to explore the regulation of a protein’s localization and distribution in order to modulate its function. Using optogenetics we aim to control trafficking of proteins from the cytosol into specific organelles, in our case peroxisomes, to sequester protein activity from the cytosol. We also aim to regulate transcription by developing optogenetic tools to regulate transcription factor localization in the nucleus. As we develop these novel systems I have also sought to characterize and understand the mechanism behind these new tools.
Optogenetics Overview
Optogenetics is an innovative field of science that uses light to precisely regulate cellular processes. The field involves the use of naturally occurring photosensory proteins to
2 develop optogenetic tools allowing scientists to probe signaling pathways with
spatiotemporal precision unmatched by chemical means of regulation. Due to the spatial
precision in which light can be administered, subcellular regions can be independently
stimulated teasing apart local effects of activation. With the reversible nature of photosensory
proteins in conjunction with the ability to regulate light intensity and duration, the tools can
provide a dose dependent and reversible means of control. Many of the early years of
optogenetics focused on membrane bound ion-channel regulation. The first system used
optogenetics to modulate a cation channel, channelrhodopsin2 (ChR2), from green algae
Chlamydomonas reinhardtii in cell culture models (Nagel et al., 2003). The Tucker laboratory and many others have been developing new optogenetic tools in the recent decade using soluble protein domains that have opened possibilities for exploring global regions of the cell beyond membrane signaling.
Soluble optogenetic tools use photoreceptors in four broad but distinct modes: caging, association, dissociation, and oligomerization (Figure I-1). Caging has been used to mask peptides, enzyme, and other molecules within either a single domain (Figure I-1A) or between two photoreceptor domains (Figure I-1B). Dimerization is used to stimulate interactions (Figure I-1C) while dissociation is used to induce separation (Figure I-1D).
Oligomerization, while similar to dimerization, is a more dramatic clustering involving many molecules (Figure I-1E). These unique exchanges are the basis of soluble optogenetic tools that will be discussed throughout this thesis. After this introduction, the history of optogenetics, considerations to make while designing optogenetic tools, photoreceptors available, and more specific methods of utilization will be discussed.
3 A B hv hv dark dark
Single Domain Caging Dual Domain Caging CD
hv hv dark dark
Dimerization Dissociation
E hv
Oligomerization
Figure I-1: Optogenetic modes of using natural properties of photosensory proteins. Caging can be obtained with a (A) single component or (B) dual component photoreceptor approach. (C) Dimerization requires two proteins to interact, but those can be two photoreceptors or a photoreceptor and it interacting partner. (D) Dissociation is the loss of interaction between two proteins and that can occur between homo- or hetero-interacting proteins. (E) Oligomerization is a multivalent interaction between photoreceptors.
4 History of Soluble Optogenetic Systems
In 2002 the first demonstration of using soluble photosensitive proteins was published by Peter Quail and colleagues showing the ability to regulate transcription with light(Shimizu-Sato et al., 2002). A transcription factor was split between the DNA binding domain and the activation domain and fused to a red light responsive phytochrome B (phyB) and PIF3 respectively. Upon red light illumination phyB and PIF3 interact reconstituting the split transcription factor and stimulating transcription. This system was shown to be reversible, as the phyB/PIF3 interaction can be released with far red light. This was an extremely innovative approach that demonstrated a dynamic range of 50-fold transcriptional activation after 30 minutes in a reversible manner, but there was a pitfall. PhyB has a necessary cofactor of a bilin chromophore, phycocyanobilin (PCB), which is not naturally found in non-plant cells. When added to the media, the yeast absorbed the PCB, but access and delivery of PCB would be a deterrent to using this system.
A lull over the next six years was broken when two groups independently used the phyB/PIF3 light dependent interaction to develop novel light inducible systems. While chemical dimerizers had been used to reconstitute split enzymes, Tyszkiewicz and Muir developed the first optogenetically regulated split enzyme using the phyB/PIF3 interaction to reconstitute a split vacuolar ATPase (VMA) intein enzyme in yeast (Tyszkiewicz and Muir,
2008). Another group used the phyB/PIF3 interaction to explore actin assembly by recruiting a GDP-bound form of Rho family GTPase Cdc42 to Wiskott-Aldrich Syndrome Protein
(WASP) stimulating actin assembly with light (Leung et al., 2008). These systems demonstrated reconstitution of split proteins and inducible protein recruitment to specific cellular locations and were the foundation upon which many optogenetic systems were built.
5 In 2009 a novel pair of light inducible dimerizers, phyB and PIF6, were used to
bridge the gap into mammalian cells (Levskaya et al., 2009). This study was the first to
demonstrate morphologic changes of the cell membrane in response to light. Using the
phyB/PIF6 interaction, catalytic domains of Guanine Nucleotide Exchange Factors (GEFs)
were recruited to the plasma membrane stimulating lamellipodia and filopodia extensions.
Levskaya et al. were able to show with spatial precision that they could cause these membrane disruptions at specific locations on the membrane by precise application of light, and the reaction was reversible with far-red light. As with the other phytochrome systems the cofactor bilin needed to be added which was a disadvantage to this tool.
Within the next several years, blue light-regulated systems were developed for optogenetic approaches, which because their cofactor is ubiquitously found in eukaryotes, overcame phytochrome’s limitation of necessary addition of exogenous cofactors. In addition to novel dimerization tools, optogenetic caging tools were established with the LOV domain.
The caging property of the LOV domain uses structural changes to control access to small peptide sequences in light versus dark conditions. These novel blue light regulated tools included dimerization with FKF1/Gigantea (Yazawa et al., 2009), caging with AsLOV2 (Wu et al., 2009), and dimerization with cryptochrome 2(CRY2)/CIB1 (Kennedy et al., 2010).
Because the photosensory proteins in these systems do not require a cofactor addition, tools developed with these tools translate more easily to eukaryotic models.
The past five years have seen exponential growth of new optogenetic modules based on growing, diverse set of LOV domains. These include LOV-based: TULIPs which used caging and dimerization (Strickland et al., 2012); SsrA/SspB dimerizers (Lungu et al., 2012), which were later optimized to develop iLIDs and other variants that explore kinetics (Guntas
6 et al., 2015; Zimmerman et al., 2016); VVD/VVD dimerizers (Chen et al., 2013), which homodimerize upon illumination; pMag/nMag, based on engineered heterodimers of VVD
(Kawano et al., 2015); homodimerizing aureochrome-1 (Grusch et al., 2014); and EL222 homodimerizers (Motta-Mena et al., 2014). Due to the natural diversity of LOV domains
(Glantz et al., 2016) there are many options to explore when applying LOV photoreceptors to optogenetic tools.
Photoreceptors with unique activation wavelengths have also been added to the optogenetic toolkit. The UVR8/COP1 light dependent interaction uses UV light to stimulate association (Crefcoeur et al., 2013). A new phytochrome dimerizing tool was developed from bacterial BphP1/PpsR2 that associates with far red light and dissociates with red light
(Kaberniuk et al., 2016). In addition to the altered activation wavelengths differing from other phytochromes, this dimerization pair relies on biliverdin cofactor, which is more readily found in eukaryotic cells.
Photoreceptors’ light inducible dissociation properties have also proven useful for studying signaling events during the past five years. UVR8 dimerizes in the dark and monomerizes upon UV illumination, and this photoreceptor was the first to use dissociation as a tool to trap proteins in a dimerized form and release them as monomers (Chen et al.,
2013). Shortly following, phyB/PIF6’s far-red triggered dissociation was utilized to regulate signaling in yeast (Yang et al., 2013). Dronpa is a photoswitchable fluorescent protein that tetramerizes in dark and monomerizes with light, and this dissociation has been used to uncage molecules sequestered between two Dronpa modules (Zhou et al., 2012). A LOV- based dissociation tool, LOV-TRAP, was developed based on the strong dark interaction of
7 synthetic peptide, Zdark, with AsLOV2, and upon illumination this affinity is reduced
allowing the diffusion of the Zdark peptide (Wang et al., 2016).
CRY2 has low-level self-association in the dark, but upon illumination the protein
oligomerizes. This light dependent homo-interaction has been utilized to regulate cellular
events through clustering of fused proteins (Bugaj et al., 2013, 2015; Kim et al., 2014; Lee et
al., 2014; Ozkan-Dagliyan et al., 2013; Wend et al., 2014). Optimizing on the homo-
interaction, in a randomized mutagenesis screen the Tucker laboratory found a light
dependent enhanced oligomerization mutant, CRY2olig, that was able to disrupt cellular
signaling (Taslimi et al., 2014). These oligomerization tools developed have proven
efficatious for clustering proteins to stimulate and sequester their activity, demonstrating
their diverse function.
Considerations for Photoreceptors when Designing Optogenetic Tools
Each photosensory protein has unique characteristics that make it more or less ideal
for specific utilizations. When developing an optogenetic tool one must consider the overall
size, availability of potential cofactors, kinetic properties, and the wavelength of stimulation in order to create an optimal system. A substantial amount of engineering goes into each novel optogenetic tool, but making careful considerations when choosing a photoreceptor makes optimization more efficient.
Size can be a critical issue when considering engineering a soluble optogenetic tool because bulky tools can cause steric hindrance on small proteins and also the ability to dissociate throughout the cells can be obstructed by large protein tags. In addition, if a tool is going to be expressed in a model organism one way to deliver the tool is via viral packaging, which has a size limitation for synthesis of viral particles. This makes the small, yet versatile
8 LOV family of photoreceptors a popular choice for small protein tags, whereas the
phytochrome and cryptochrome modules add more bulk to a tool.
Availability of cofactors is another consideration. Some photoreceptors use
endogenous cofactors while others require exogenous cofactors to be added. The first
systems developed used phytochromes which require tetrapyrrole chromophores, which are
not found naturally in non-plant cells. New bacterial phytochromes that use biliverdin IXα have recently been developed as optogenetic photoreceptors. Biliverdin IXα is found in eukaryotic cells conferring the potential for red light sensitive photoreceptors to be used without exogenous cofactor addition (Kaberniuk et al., 2016). The blue light photoreceptors such as CRYs, BLUF domains, and the LOV domain family utilize cofactors endogenous to all organisms, positioning them to be used easily in any cell type. Other photoreceptors such as Dronpa and UVR8 are able to function without a cofactor making them ideal for universal use.
Kinetics are key considerations and cover rates of activation and reversibility. Each photoreceptor will eventually revert back to its dark state, but for some this is a fast process taking only seconds while others such as UVR8 can take over 8 hours. The phytochromes and Dronpa are distinct in their ability to be turned on and off with separate wavelengths giving them temporally the most precise control. Systems such as the LOV family have been well characterized to find mutations that can alter the on/off rates. This is one area in which the photosensory proteins can be mutated to change their properties to create an optimal tool, and the Tucker laboratory has been exploring mutations that alter the kinetics of CRY2
(Taslimi et al., 2016). The rates of kinetics of each photoreceptor will determine the temporal precision a system can achieve.
9 The photoreceptor’s wavelength of activation is especially important when frequent
illumination must be used and when determining other fluorescent tags that may need to be
detected. Certain wavelengths such as ultraviolet can have toxicity associated with exposure, so for longer illumination periods a longer wavelength is optimal. Another benefit of longer wavelengths, such as those that regulate red light photoreceptors, is the ability for the light to penetrate deeper into tissues of model organisms. Overall, the longer the wavelength the less harmless and the more penetrable, which is optimal.
Soluble Optogenetic Photoreceptors
There are a growing number of photoreceptors that have been used for optogenetics, and each has its own individual characteristics. This section details the photoreceptors that have been used to date and gives insights into the benefits and pitfalls of each. Table I-1 summarizes the characteristics of the photoreceptors used in optogenetic systems. The studies in this thesis use a combination of LOV and CRY2 based optogenetic tools, and thus they are described first.
LOV-Domains
Light-oxygen-voltage domains (LOV domains) are relatively small (~110 amino
acid) blue light responsive proteins. There are a multitude of LOV-domain containing
proteins found in the environment (Spencer T. Glantz et al., 2016), but only a few have been
characterized for optogenetic usage. LOV domains are flavin-binding structural motifs that
belong to the larger family of Period-ARNT-Singleminded (PAS) domain proteins. They
bind Flavin mononucleotide (FMN) or flavin adenine dinucleotide (FAD) as a chromophore
and show a peak absorbance at 450 nm (Zoltowski and Imaizumi, 2014). It acts to create
light dependence through formation of a covalent thioether bond between the LOV domain
10
11 and flavin (Moglich et al., 2010). The LOV domains that have been developed for
optogenetic use are engineered from naturally occurring proteins, but often synthetic
alterations have been made to change characteristics like kinetics and sensitivity. Below the
differences between LOV variants AsLOV2, FKF1, Aureochrome-1, VVD and EL222 are described.
AsLOV2
This photoreceptor is currently the most widely utilized of LOV based photosensory proteins, likely due to its small size and fast kinetics. This domain is the second LOV domain from the phototropin 1 protein of Avena sativa (AsLOV2). It contains a PAS core that utilizes
flavin mononucleotide (FMN) as a cofactor. Upon photoexcitation, a covalent bond is formed
between the FMN isoalloxazine ring and a conserved cysteine residue of LOV2 (Christie et
al., 2012; Salomon et al., 2000), this bond formation stimulates unwinding of a C-terminal
Jα-helix that is docked to the protein in the dark state (Halavaty and Moffat, 2007; Harper et
al., 2003). Structural changes within the Jα-helix have been utilized to cage peptide
sequences in a light dependent manner (Wu et al., 2009). The peroxisomal trafficking tool
detailed in Chapter II uses AsLOV2 properties to harness its caging abilities.
Multiple synthetic systems using AsLOV2 core as the photosensory protein have been
developed and further optimized. The tunable light-controlled interacting proteins (TULIPs)
system uses a combination of caging and dimerization through fusion of a PDZ-binding
peptide to the C-terminus of the Jα-helix, which cages the peptide in the dark but upon
illumination the Jα-helix unwinds allowing for an engineered PDZ domain to bind
(Strickland et al., 2012). This study also explored a series of mutations that alter kinetics of
reversion and the sensitivity of the tool. In a similar approach a bacterial SsrA peptide was
12 engineered into the Jα-helix to allow for light dependent interaction with its binding partner
SspB (Lungu et al., 2012). The SsrA/SspB tool was further optimized to increase dynamic range and decrease background interactions. These optimizations have led to systems including iLID nano, iLID micro, and iLID/SspB-milli, all with tunable kinetic properties
(Guntas et al., 2015; Zimmerman et al., 2016).
AsLOV2 was also used in the synthetic system, LOVTRAP, which allows for regulation of light-dissociation (Wang et al., 2016). LOVTRAP has two components: the
AsLOV2 domain and a mutated domain of the Z domain of protein A, Zdark (Zdk). Zdk was found in a randomized mutation screen selecting for peptides that have dark association with
AsLOV2. Wang et al. were able to demonstrate the ability to regulate signaling cascades by using the light initiated release tool (Wang et al., 2016). LOVTRAP, TULIPs, and the iLID variants are all examples of successful hybrids of naturally based AsLOV2 domains with engineered components.
FKF1
Flavin Binding, Kelch Repeat, F-Box 1 (FKF1) is an Arabidopsis thaliana protein that contains a LOV domain, and is known to interact with Gigantea (GI) (Imaizumi et al.,
2003; Sawa et al., 2007). Similar to AsLOV2, FKF1 uses FMN as the chromophore. While the FKF1/GI interaction was the first blue light optogenetic tool published (Yazawa et al.,
2009) the size of the components are extremely bulky (FKF1 is 619 amino acids and GI is
1173 amino acids). The minimal photoactivation domain is between amino acids 56-168 of
FKF1 suggesting that FKF1 can be reduced in size without altering photosensitivity; however, the interaction with GI requires a larger minimal domain. Other alterations to FKF1 include mutations to the endogenous nuclear localization signal (NLS) when utilizing the
13 protein in the cytosol. Compared with other LOV domain containing proteins, FKF1 has very slow reversion kinetics to dark state with a half time of over 100 hrs. This is a disadvantage for fine temporal or dose-dependent control, but a long-lasting dimerization can be advantageous if a more permanent interaction is desired.
Aureochrome
Aureochrome-1 is a LOV-domain containing protein found in Vaucheria frigida that differs from the previously described LOV-domain containing proteins because it homodimerizes upon illumination (Takahashi et al., 2007). Aureochrome-1 is composed of a
N-terminal basic region/leucine zipper (bZIP) DNA-binding domain and a single C-terminal
LOV domain and also binds a FMN cofactor (Takahashi et al., 2007). The bZIP dimerization has been utilized for optogenetic applications to bring proteins together (Grusch et al., 2014).
It is known that light stimulation induces conformational change leading to homodimerization, but the exact mechanism is currently not fully understood. Kinetic studies of AuLOV show reversion rates in the range of minutes, but like with other LOV domains, mutations have tuned the kinetics (Mitra et al., 2012).
VVD
Vivid (VVD) is a LOV-domain-containing protein from the fungus Neurospora crassa (Schwerdtfeger and Linden, 2003) that also homodimerizes when illuminated with blue light (Zoltowski and Crane, 2008; Zoltowski et al., 2007). It differs from other LOV domains utilized in optogenetics due to its use of flavin adenine dinucleotide (FAD) rather than FMN as a chromophore (Zoltowski et al., 2007). Light dependent dimerization of VVD was first used to regulate transcription (Wang et al., 2012). A mutation of I52C was able to increase VVD’s homodimerization property enhancing the gene induction (Nihongaki et al.,
14 2014). VVD’s kinetics are slow in comparison to other LOV domains with a dark reversion
half-life of 2.8 hr (Zoltowski et al., 2007).
A synthetic version of VVD-VVD was engineered to confer heterodimerization to the
system, termed Magnets (pMag/nMag), with mutations of VVD at I52R/M55R (pMag) and
I52D/M55G (nMag) (Kawano et al., 2015). Heterodimerization occurs within seconds and
dissociation takes hours, while homodimerization is significantly reduced. Optimized
variants of Magnets have been designed including ‘High’ mutants (M135I and M165I) that
show enhanced dimerization but reduced dissociation kinetics, and ‘Fast’ mutants (I74V and
I85V) with a shortened photocycle but decreased dimerization. Combination of ‘High’ and
‘Fast’ mutations results in increased heterodimerization and faster system kinetics (Kawano
et al., 2015).
EL222
EL222 is a 208 amino acid domain designed from a LOV domain containing
transcription factor from bacteria, Erythrobacter litoralis (Motta-Mena et al., 2014; Nash et
al., 2011) The N-terminal LOV domain cages a helix-turn-helix (HTH) DNA binding domain that becomes exposed with light and permits dimerization. To use EL222 to modulate transcription, VP16 transcriptional activation domain was fused to the N-terminus of EL222
(LOV-HTH) which under illumination binds the HTH promoter (Motta-Mena et al., 2014).
With kinetics in the seconds and extremely low background, this LOV domain has potential for precise temporal tunability and propensity for high dynamic range.
Cryptochromes
Cryptochromes (CRYs) are found in bacteria and eukaryotes, but the optogenetic
tools use plant cryptochromes from Arabidiopsis thaliana. CRYs are blue light sensitive
15 photoreceptors that are known to regulate hypocotyl elongation, photoperiodic flowering
initiation, and other light dependent processes in plants. They are composed of a N-terminal
photosensory domain (photolyase homology region, PHR) with structural homology to DNA
photolyase, and a relatively unstructured C-terminal tail domain (CTD). CRYs use flavin
adenine dinucleotide (FAD) as their cofactor, similar to some LOV domains. FAD is thought
to undergo photoreduction upon light stimulation, leading to conformational changes within
the CRY2 protein but the exact mechanism has not been proven (Conrad et al., 2014).
The blue light dependent Arabidopsis CRY2 interaction with CIB1 (Liu et al., 2008) was first used as a light dependent dimerization tool to modulate protein localization, transcription, and reconstitute a split enzyme (Kennedy et al., 2010). The original CRY2 and
CIB1 proteins are relatively large, at 612 amino acids and 335 amino acids, respectively. A
498 residue tructation of CRY, CRY2PHR, expressed better than full length and in many cases reduces the background binding of CRY2 and CIB1 in the dark. A 170 residue truncation of CIB1, termed CIBN, retains interaction with CRY2 and can be substituted for
CIB1. Recent studies show a 535 amino acid truncation of CRY2 has improved functionality over full length and may be the most optimized version (Taslimi et al., 2016). In the same study a drastically shorted CIB1, residues 1-81, retained light dependent interaction with
CRY2 demonstrating a more minimal motif (Taslimi et al., 2016). Unless the constructs are to be used in the nucleus, the endogenous nuclear localization signals of CRY2 and CIB1 must be removed to sustain cytosolic localization (Kennedy et al., 2010). The CRY2-CIB1 interaction occurs within seconds and dissociation occurs within minutes, with a half-life of about five minutes. As with many photoreceptors, mutants modulating kinetics of CRY2 have been identified for use (Taslimi et al., 2016).
16 CRY2 is also known to have low levels of self-association in the dark and
oligomerizes upon light exposure. CRY2 oligomerization has been used successfully as an
optogenetic tool in multiple studies to activate or inactivate biological processes with light
(Bugaj et al., 2013, 2015; Chang et al., 2014; Kim et al., 2014; Lee et al., 2014; Ozkan-
Dagliyan et al., 2013; Taslimi et al., 2014; Wend et al., 2014). While this oligomerization has
proven useful, it can also disrupt CRY2-CIB1 dimerization tools, as discussed in Chapter III.
The Tucker laboratory explored further CRY2 oligomerization by engineering an E490G
mutation in CRY2PHR, termed “CRY2olig”, which has enhanced oligomerization with blue
light illumination and has been used to perturb cellular events (Taslimi et al., 2014).
Phytochromes
Phytochromes are red and far-red light responding plant photoreceptors. Two
optogenetic systems utilizing Arabidopsis thaliana phytochrome B (phyB) have been
developed: phyB/PIF3 and phyB/PIF6. The phyB/PIF3 dimerizers uses residues 1-621 of
phyB, a N-terminal photosensory domain which interacts with PIF3, a transcription factor
that specifically binds red light-stimulated phyB (Ni et al., 1999; Shimizu-Sato et al., 2002).
The phyB/PIF6 dimerizers use residues 1-908 of phyB which binds residues 1-100, the N-
terminal domain, of PIF6 (Levskaya et al., 2009).
PhyB has low affinity for PIF family proteins in the red light responsive (Pr) state, but
upon red light illumination it adopts a far-red-light responsive (Pfr) state with high affinity to
PIF proteins (Rockwell et al., 2006). An essential cofactor, bilin chromophore, phytochromobilin (PΦB), must be present to mediate light responsiveness. Phytochromobilin covalently links to phyB in dark and undergoes a cis-trans isomerization with photon absorption, resulting in conformational change of the protein to the Pfr state. In the Pfr state,
17 the molecule can absorb a second photon that reverts the protein back to the Pr state. The
kinetics for the photoswitching between these two states are within seconds.
There are multiple advantages to using phytochromes including the fast kinetics and
ability to control association and dissociation with different wavelengths of light. While red
light promotes the phyB-PIF interaction, far-red light is used for dissociation. Since longer
wavelengths are less toxic and penetrate tissues deeper, red and far red illumination for in
vivo applications is beneficial.
Disadvantages of the phyB/PIF systems include the requirement for the PΦB cofactor, which must be added exogenously to non-plant cells. Often a related cofactor phycocyanobilin (PCB) extracted from spirulina is substituted for the plant cofactor PΦB, (Li and Lagarias, 1994; Shimizu-Sato et al., 2002). While PCB is absorbed slowly by cells, the exogenous addition makes the systems difficult for use in tissues and model organisms, though studies are making the chromophore purification and delivery more efficient
(Buckley et al., 2016). Co-expressing biosynthetic enzymes that stimulate generation of the bilin cofactor is one way to circumvent this lack of cofactor, however, two different genes are required to synthesize PCB from heme (Gambetta and Lagarias, 2001; Levskaya et al.,
2005; Müller et al., 2013a). Another disadvantage is the phytochromes are also not tolerant of C-terminal fusions (Leung et al., 2008) and thus forces N-terminal fusions on the phyB for proteins of interest. These disadvantages have likely played a role in the limited widespread adoption of the phyB/PIF systems.
Bacterial Phytochrome
Recently a novel BphP1/RpPpsR2 interaction was developed for optogenetics using
a bacterial phytochrome, phBphP1, from Rhodopseudomonas palustris. This unique
18 phytochrome exists in a ground Pfr state but can be photoconverted to a Pr state with near- infrared illumination (~740-780 nm) where it can interact with transcriptional repressor
RpPpsR2 (Bellini and Papiz, 2012; Rottwinkel et al., 2010). These interacting proteins were used as the basis for a new red/far-red photodimerizer system (Kaberniuk et al., 2016).
There are innovative advantages to the BphP1-RpPpsR2 interaction such as use of a biliverdin chromophore that is endogenously found in mammalian cells. In addition, BphP1-
RpPpsR2 interaction is stimulated with 740 nm light and dissociation induced with 650 nm light. The near infrared (NIR) light at 740 nm penetrates even more deeply into tissues and organs than red light, which is a substantial advantage of this system. The kinetics are fast with dimerization occurring within seconds of light exposure. Light can also be used to induce dissociation; however, the only observed disadvantage is an overlap between Pr and
Pfr light sensitivities leading to only ~80% of the BphP1 protein dissociated upon illumination. This can be overcome by allowing full dark reversion of the Pr state to the ground Pfr state, but the kinetics are dependent on BphP1-RpPpsR2 ratios, occurring with a half-time of 15 min for an 8:1 molar ratio (Kaberniuk et al., 2016).
UVR8
Ultraviolet resistance locus 8 (UVR8) is an UV-B sensitive photoreceptor from
Arabidopsis thaliana that has been used in two optogenetic systems: UVR8-COP1 and
UVR8-UVR8. The photoreceptor utilizes internal tryptophan residues W233 and W285 to form a chromophore and therefore has no prosthetic cofactor. In the dark UVR8 is dimeric and UV-B illumination releases it to a monomeric state where it can interact with its binding
partner, COP1, to induce gene expression (Rizzini et al., 2011). UVR8 has been utilized in
19 optogenetic tools to stimulate dissociation of the UVR8-UVR8 dimer (Chen et al., 2013) and
to mediate association of UVR8-COP1 heterodimer (Crefcoeur et al., 2013).
Unique properties of UVR8 systems are the light stimulated dissociation and lack of
cofactor necessary to form a chromophore. Also, UV-B light as a stimulus enables the
cooperative imaging of most fluorescent proteins (FPs) due to the fact that FPs excitation spectrum do not generally overlap with the 280-315nm UV-B wavelength, which is an advantage for tracking cellular events. There are significant limitations to this system. The wavelength, while advantageous for imaging, can induce DNA damage and activate stress pathways. The kinetics of the system are fast, with dissociation to monomers occurring within seconds; however, reversibility requires a protein from plants (Heijde and Ulm, 2013) limiting the temporal tunability of the system.
Dronpa
Dronpa is unique because it is a photoswitchable protein engineered from a
tetrameric, 257 amino acid, Pectinidae coral protein (Ando et al., 2004). A mutant form,
Dronpa145N, undergoes changes in subunit association (tetramer to monomer) upon illumination with 488-500 nm light (Mizuno et al., 2010), and this property has been utilized for optogenetic applications (Zhou et al., 2012). In concert with changes in subunit association, Dronpa145N also shows changes in fluorescence upon illumination, making it a unique molecule to use in optogenetics. Dronpa is converted by strong excitation at 490 nm to a non-fluorescent dark state, which can be switched back by illumination at ~400 nm. In addition to the traceable nature of Dronpa, it also does not need an external cofactor. Like
GFP, Dronpa relies on the switch from cis- to trans-conformation of the chromophore formed by residues of the molecule, Cys-62–Tyr-63–Gly-64 (Andresen et al., 2007). The
20 switchable association/dissociation occurs within seconds allowing for fast kinetics of the
Dronpa system.
BLUF Domains
Sensors of Blue-Light Using FAD (BLUF) domains were first found in the N-
terminus of the AppA protein present in Rhodobacter sphaeroides (Gomelsky and Klug,
2002). There are many other BLUF containing proteins including PixD and YcgF, often
found in prokaryotes. They have binding properties in the dark and dissociate from their
complex upon illumination (Masuda and Bauer, 2002; Masuda et al., 2008; Tschowri et al.,
2009). As the name implies, these are blue light photosensory domains that use FAD as
cofactor. The kinetics of the BLUF domains varies from seconds to minutes for dark state
reversions (Kraft et al., 2003; Zirak et al., 2006). These domains have not yet been developed
for novel optogenetic tools but rather have been used to modulate their natural pathways;
however, they are thought to be potential photoreceptors for synthesizing new optogenetic
systems.
Applications of Soluble Optogenetic Systems
As introduced throughout this Chapter, protein function directly correlates with
protein localization. Optogenetics has utilized this principle to regulate protein function with
many different tools (Table I-2). Proteins of interest can be controlled by fusing them to photoreceptor proteins which are designed to control their localization. Optogenetic tools have utilized this engineering technique to modulate target proteins in distinct ways including: recruitment to specific subcellular locations, reconstitution of split protein, induction of protein interactions, and caging/sequestering and then releasing protein (Figure
21 Target Photosensitive Protein Associating Protein Anchoring Protein
A B
hv hv
N C
C D
hv hv
E F
hv hv
G H
hv hv
Figure I-2: Photodimerizer approaches to regulate protein function.
22 Figure I-2: Photodimerizer approaches to regulate protein function. Schematic showing approaches used to regulate protein activity using photodimerizers. The target protein regulated is colored in green, with photosensory proteins in blue, partner proteins in purple, and anchoring proteins in orange. A. Dimerization of two different proteins. Photodimerizers can be used to bring two different target proteins together with light. B. Reconstitution of a split protein. Activity is achieved by fusing the N-terminus and C-terminus of a split protein to the interacting photodimerizers, allowing functional reconstitution with light. C. Reconstitution of a split transcription factor. Variation of (B), where one of the photodimerizer partners is fused to a DNA binding domain, where it can bind to DNA at a promoter site. The partner photodimerizer is fused to a transcriptional activation domain. Light allows recruitment of the activation domain to this location, resulting in activation of transcription. D. Recruitment to anchored subcellular location. One of the photodimerizer components is fused to a protein or peptide allowing anchored subcellular localization (plasma membrane is shown as an example), while the other is fused to a target protein of interest. Light illumination allows recruitment of the target to the anchored location. E. Oligomerization. A target is fused to a photoreceptor (such as CRY2) that undergoes light dependent oligomerization, which can be used to induce or disrupt activity. F. Sequestration/release. Used with light-dissociated dimerizers, or light- associated/dissociated dimerizers such as phyB/PIF. A target protein is anchored (sequestered) at an inactive subcellular location, then released with light illumination to allow function. G. Dissociation of protein clusters. Used with light-dissociated dimerizers, protein assembles into clusters that impede trafficking or activity, which are dissociated/dissolved with light. H. Single chain caging. Interacting partners are presented on the same protein chain, along with a target protein or domain. Function of the target protein is sterically or allosterically impeded when photodimerizers are bound, but enabled when partners dissociate.
23 Table I-2: Applications of soluble photoreceptor technologies System Utilization Reference Phytochrome based systems Plant phytochrome B phyB/PIF3 (1-621/FL) Reconstitution of a split Gal4 transcription factor to regulate (Shimizu-Sato et gene expression in yeast al., 2002) phyB/PIF3 (1-621/FL, Reconstitution of split S. cerevisiae vacuolar ATPase intein (Tyszkiewicz and 1-621/1-100) in yeast Muir, 2008) phyB/PIF3 (1-651/1-100) Induced dimerization of GDP-bound form of Cdc42 with (Leung et al., WASP to stimulate actin assembly 2008) PhyB/PIF6 (1-908/1-100) Activation of Rho GTPase pathways by recruiting the (Levskaya et al., catalytic domain of Rho GEFs to the plasma membrane 2009) phyB/PIF3 (1-621/FL) Reconstitution of a split transcription factor to regulate gene (Hughes et al., expression in yeast 2012a) PhyB/PIF6 (1-908/1-100) Recruitment of catalytic domain of SOS to the plasma (Toettcher et al., membrane to activate Ras and the Ras/Raf/MEK/ERK 2013) pathway phyB/PIF6 (1-908/1-100) Recruitment of Gal80 to PM with 650 nm light, resulting in (Yang et al., alleviation of Gal80 transcriptional repression in the 2013) nucleus; Recruitment of Clb2-PIF6 to intracellular sites (nucleus, spindle pole body) to disrupt function. phyB/PIF6 (1-650/1-100) Reconstitution of a split transcription factor to regulate gene (Müller et al., expression in mammalian cells. 2013b) phyB/PIF6 (1-908, 1-650/1-100) Reconstitution of a split transcription factor to regulate gene (Müller et al., expression in mammalian cells. 2013c) phyB/PIF6 (1-908/1-100); Reconstitution of a split transcription factor to regulate gene (Pathak et al., phyB/PIF3 (1-621/1-100) expression in yeast 2014) phyB/PIF6 (1-908/1-100) Sequestration of Bem1 at inactive sites to prevent (Jost and Weiner, polarization and budding, probe consequences of release 2015) at different times phyB/PIF6 Targeted AAV viral particles to the nucleus for enhanced (Gomez et al., gene delivery by recruiting a PIF6-fused viral capsid protein 2015) to a phyB-NLS phyB/PIF3 (1-908/FL) Regulation of nuclear-cytoplasmic transport through (Beyer et al., phyB/PIF3 interaction in mammalian cells and zebrafish 2015) phyB/PIF6 (1-908/1-100) Recruitment of Gαq, Gαs to plasma membrane, inducing (Yu et al., 2016) activation of PLCβ phyB/PIF6 (1-908/1-100) Recruitment of apical polarity protein Pard3 to plasma (Buckley et al., membrane in zebrafish 2016) Bacteriophytochrome BphP1-PpsR2 Recruitment of catalytic domain of intersectin to the plasma (Kaberniuk et al., membrane; Reconstitution of a split transcription factor to 2016) induce gene expression in mammalian cells Cryptochrome based systems CRY2/CIB1 CRY2/CIB1, Recruitment of cargo to plasma membrane in mammalian (Kennedy et al., CRY2PHR/CIBN cells; Reconstitution of a split Cre DNA recombinase; 2010) Reconstitution of a split transcription factor to induce gene expression in yeast CRY2PHR/CIBN Recruitment of an inositol 5-phosphatase domain to the (Idevall-Hagren et plasma membrane to alter phosphoinositide metabolism; al., 2012) Recruitment of inter-SH2 domain of p85α, a regulatory subunit of PI3 kinase, to the plasma membrane CRY2PHR/CIBN Reconstitution of a split transcription factor to regulate gene (Hughes et al., expression in yeast 2012a) CRY2/CIB1 Reconstitution of a split transcription factor to modulate (Liu et al., 2012) gene expression in zebrafish
24 Table 1-2 continued
System Utilization Reference CRY2/CIB1 (LITEs: light-inducible Recruitment of transcriptional activator or epigenetic (Konermann et transcriptional effectors) modifiers to promoter sites using TALE DNA binding al., 2013) domains CRY2PHR/CIBN Recruitment of iSH2 domain of p85β, a regulatory subunit (Kakumoto and
of PI3kinase, to the plasma membrane to stimulate PIP3 Nakata, 2013) production CRY2/CIBN Reconstitution of a split Cre DNA recombinase in (Boulina et al., Drosophila 2013) CRY2PHR/CIBN Recruitment of an inositol 5-phosphatase domain to the (Giordano et al., plasma membrane to alter phosphoinositide metabolism 2013) CRY2PHR/CIBN Recruitment of eIF4E to PUF domain that binds mRNA to (Cao et al., 2014) regulate translation CRY2/CIBN Recruitment of a protease to its target cleavage sequence, (Wieland et al., resulting in release of a membrane-tethered transcription 2014) factor CRY2PHR/CIBN Recruitment of Raf1 to the plasma membrane to stimulate (Zhang et al., ERK signaling 2014) TULIPs Reconstitution of a split transcription factor to regulate gene (Pathak et al., expression in yeast 2014) CRY2PHR/CIBN Recruitment of cofilin to Lifeact to induce F-actin (Hughes and remodeling, filopodia, and directed cell motility Lawrence, 2014) CRY2PHR/CIBN Recruitment of catalytic domain of intersectin to plasma (Valon et al., membrane 2015) CRY2/CIBN Split dCas9/CRISPR system regulating endogenous genes (Polstein and in mammalian cells Gersbach, 2015) CRY2PHR/CIB1 Reconstitution of split dCas9/CRISPR system regulating (Nihongaki et al., endogenous genes in mammalian cells 2015a) CRY2/CIB1 Mitochondrial recruitment of Bax to stimulate Smac1 (Hughes et al., release, activating caspase cleavage and apoptosis 2015) CRY2/CIBN Mediation of actin polymerization by recruitment of Arpin (Maiuri et al., away from leading edge 2015) CRY2/CIBN Reconstitution of split transcription factor in Drosophila (Chan et al., 2015) CRY2/CIBN Reconsitution of a split Cre DNA recombinase in mouse (Schindler et al., introduced via AAV 2015) CRY2PHR/CIBN Recruitment of kinase domain of Akt to the plasma (Katsura et al., membrane to activate Akt signaling 2015) CRY2PHR/CIBN Recruitment of organelles (mitochondria, peroxisome, (Duan et al., lysosome) to cytoskeletal motors (kinesin, dynein) to 2015) regulate organelle positioning CRY2PHR/CIBN Recruitment of an inositol 5-phosphatase domain to the (Guglielmi et al., plasma membrane to alter phosphoinositide metabolism in 2015) Drosophila CRY2PHR/CIBN Reconstitution of split Gal4 transcription factor; recruitment (Hallett et al., of Tiam catalytic domain to plasma membrane 2016) CRY2PHR/CIBN Dimerization of Cav1.3S or Cav1.3L channels (Moreno et al., 2016) CRY2/CIB1 Recruitment of DNA methyltranferase to subtelomeric (Choudhury et al., regions to regulate methylation 2016) CRY2/CIB1; CRY2(535)/CIB1 Reconstitution of split Cre DNA recombinase using CRY2 (Taslimi et al., mutant L348F; Reconstitution of a split transcription factor 2016) to regulate gene expression in yeast CRY2/CIBN (EXPLORs) Recruitment of cargo protein into exosomes (Yim et al., 2016) CRY2PHR/CIBN Recruitment of Raf and Akt to the plasma membrane to (Ong et al., 2016) activate signaling CRY2/CIBN Recruitment of Raf1 to the plasma membrane to stimulate (Krishnamurthy et ERK signaling al., 2016)
25 Table 1-2 continued
System Utilization Reference CRY2PHR/CIBN Recruitment of PLEKHG3 to Lifeact, Recruitment of iSH2 (Nguyen et al., domain of p85β to the plasma membrane 2016b)
CRY2PHR/CIBN Recruitment of i-SH2 domain of p85α, a regulatory subunit (Xu et al., 2016) of PI3 kinase, to the plasma membrane; Recruitment of PH domain of Akt to PM CRY2 dimerization/ oligomerization CRY2/CRY2 Clustering of CRY2-LRP6 c-terminal domain to activate β- (Bugaj et al., catenin signaling; Clustering of CRY2-Rac1 or CRY2-RhoA 2013) to induce cytoskeletal changes CRY2/CRY2 Clustering CRY2-TopBP1 to activate ATR-mediated DNA (Ozkan-Dagliyan damage checkpoint signaling in the absence of DNA et al., 2013) damage CRY2/CRY2 and CRY2/CIBN Homodimerization of CRY2PHR-C-RAF or B-RAF; (Wend et al., heterodimerization of C-Raf and B-Raf through CRY2PHR- 2014) CIBN interaction CRY2PHR/CRY2PHR Dimerization of TrkB receptor to stimulate Trk signaling (Chang et al., 2014) CRY2PHR/CRY2PHR Dimerization of FGFR to stimulate MAPK, PI3K, and (Kim et al., 2014) phospholipase C signaling CRY2PHR/CRY2PHR/CIB1 (LARIAT: Use of a CaMKII-CIB1 along with target proteins fused to (Lee et al., 2014) light activated reversible inhibition by CRY2 to cluster and sequester activity (targeting: Vav2, assembled trap) Tiam1, Rac1, Cdc42, RhoG, PI3K) CRY2olig (CRY2PHR E490G) Clustering of CRY2olig-NCK (SH3 domains) and VCA (Taslimi et al., domain of N-WASP to induce changes in actin 2014) polymeration; Clustering of CRY2olig-clathrin light chain to disrupt endocytosis CRY2PHR/CRY2PHR (CLICR: Clustering of PLC-γ SH2 domain using CRY2PHR resulting (Bugaj et al., clustering indirectly using CRY2) in targeting of SH2 domain to endogenous binding partners 2015) and activation of receptor tyrosine kinases; Similar approach with talin F3 domain targeting integrins CRY2PHR/CRY2PHR/ Clustering of CIBN-Rab GTPases (Rab5, Rab11,Rab7, (Nguyen et al., CIBN Rab3a, Rab2a, Rab6a) with CRY2PHR 2016a) CRY2PHR/CRY2PHR; CRY2/CIB1 Homodimerization of CRY2PHR-BRAF and CRY2PHR- (Chatelle et al., CRAF; heterodimerization of B-RAF and C-RAF via 2016) CRY2PHR-CIBN interaction CRY2PHR/CRY2PHR Dimerization of FGFR to induce FGFR signaling pathways (Kim et al., 2016) CRY2/CRY2 Dimerization of DCC receptor to stimulate axon extension (Endo et al., through FAK and PLCϒ-1 signaling 2016) CRY2PHR/CRY2PHR/CIBN Utilized LARIAT (Lee et al. 2014) to cluster GFP binding (Nguyen et al., antibody VHH-CRY2 and GFP-PLEKHG3 2016b) LOV Based Systems FKF1/GI FKF1/GI Reconstitution of a split transcription factor to regulate gene (Yazawa et al., expression in mammalian cells; Plasma membrane 2009) recruitment of Rac1 FKF1/GI (LITEZ) Recruitment of a transcriptional activator to programmed (Polstein and zinc finger binding site Gersbach, 2012) TULIPs (AsLOV2pep-ePDZ) TULIPs Membrane recruitment of Ste5 to stimulate mating MAP (Strickland et al., kinase pathways; Membrane recruitment of Cdc24 to 2012) modulate cell polarity TULIPs Reconstitution of a split transcription factor to regulate gene (Pathak et al., expression in yeast 2014) TULIPs Reconstitution of a split transcription factor to regulate gene (Müller et al., expression in yeast 2014) TULIPs Reconstitution of split Gal4 transcription factor; Recruitment (Hallett et al., of Tiam DH/PH catalytic domain to PM 2016) TULIPs Recruitment of peroxisomes or recycling endosomes to (van Bergeijk et cytoskeletal motor proteins (kinesin, dynein or myosin) to al., 2015) regulate organelle positioning
26
Table 1-2 continued
System Utilization Reference
TULIPs Recruitment of RhoA GEF LARG to plasma membrane (Wagner and Glotzer, 2016) VVD dimerization VVD-VVD (LightON) Dimerization of a transcription factor to regulate gene (Wang et al., expression in mammalian cells 2012) VVD-VVD Dimerization of a transcription factor to regulate gene (Müller et al., expression in mammalian cells 2013b) VVD-VVD Reconstitution of T7 RNA Polymerase (Han et al., 2016) VVD-VVD Caspase-9 dimerization (Nihongaki et al., 2014) Magnets (nMag/pMag) Magnets (nMag/pMag) Recruitment of iSH2 domain of p85β, a regulatory subunit (Kawano et al., of PI3kinase, to the plasma membrane to stimulate PIP3 2015) production Magnets Split dCas9/CRISPR gene regulation system (Nihongaki et al., 2015b) Magnets Recruitment of Gαq to plasma membrane, inducing (Yu et al., 2016) activation of PLCβ Magnets Reconstitution of T7 RNA Polymerase (Han et al., 2016) Magnets Reconstitution of split Cre DNA recombinase (Kawano et al., 2016) AsLOV2-SsrA/SspB AsLOV2-ipaA/vinculinD1 Perturbation of actin co-sedimentation in vitro; (Lungu et al., AsLOV2-SsrA/SspB (oLID) Reconstitution of split Gal4 transcription factor in yeast 2012) iLIDs (improved AsLOV2-SsrA/SspB) Recruitment of Tiam and Intersectin catalytic domains to (Guntas et al., iLIDnano, iLIDmicro the plasma membrane 2015) iLID Recruitment of catalytic domain of Cdc42 GEF to plasma (O’Neill et al., membrane 2016) iLID Reconstitution of split Gal4 transcription factor; Recruitment (Hallett et al., of Tiam DH/PH catalytic domain to PM 2016) iLID/SspB_milli (A58V in SspB). sLID Recruitment of cargo to subcellular locations (Zimmerman et (slow-cycling N414L AsLOV2 mutant) al., 2016) EL222 dimerization EL222 Dimerization of EL222 to regulate gene expression in (Motta-Mena et mammalian cells al., 2014) Aureochrome dimerization Aureochrome 1 Dimerization of receptor tyrosine kinases (FGFR1, EGFR, (Grusch et al., and RET) 2014) LOVTRAP Zdk/AsLOV2 (LOVTRAP) Release of constitutively active Vav2, Rac1, or RhoA from (Wang et al., anchored sites through dissociation of Zdk/AsLOV2 2016) interaction. Dronpa based systems Dronpa Plasma membrane recruitment; caging Cdc42 GEF (Zhou et al., intersectin activity, caging HCV NS3-4A protease 2012) UVR8 based systems UVR8/COP1 Recruitment of cargo to chromatin; Reconstitution of a split (Crefcoeur et al., transcription factor to drive a reporter gene 2013) UVR8/COP1 Reconstitution of a split transcription factor to regulate gene (Müller et al., expression in mammalian cells. 2013b) UVR8/UVR8 Light mediated dissociation of UVR8 fused VSVG to (Chen et al., release from ER 2013) Combined systems CRY1/phyB Reconstitution of a split transcription factor to drive a (Hughes et al., reporter gene 2012b)
27 I-2). These modulations of proteins of interest are achieved through photoreceptor caging,
association, dissociation, and oligomerization.
Recruitment to Specific Cellular Locations
Regulating protein localization is a powerful way to regulate its activity and thus has
been the focus of many optogenetic tools. In order to regulate specific localization, a
dimerizing approach is used where one half of the dimer is fixed with a docking tag specific
to a subcellular location, and the protein of interest is fused to the other half of the
dimerization pair. Dimerization occurs upon illumination recruiting the target proteins to the
location of the anchored half of the interacting partners.
Recruitment of proteins to the plasma membrane is a broadly utilized approach demonstrating high efficiency at regulating a large variety of proteins. One of the first studies used the PhyB/PIF6 optogenetic system to recruit the DHPH catalytic domains of GDP-GTP exchange factors (GEFs) Tiam or intersectin to the plasma membrane, resulting in induction
of Rac1 or Cdc42 activity, respectively (Levskaya et al., 2009). This system was also able to
show spatial precision by localizing the DHPH domains to specific locations to regulate
membrane morphology. Due to the use of a phytochrome system the tool was also shown to
be reversible and could be activated/deactivation through multiple photocycles.
Many other photoreceptor systems demonstrated efficiency at regulating RhoGTPase
pathways through membrane recruitment, and others also showed robust modulation of
kinase signaling pathways. Using TULIPs and CRY2/CIB1 dimerization tools, MAP kinase
scaffold Ste5 was recruited to the plasma membrane in yeast resulting in activation of the
MAP kinase dependent mating pathway (Pathak et al., 2014; Strickland et al., 2012).
CRY2PHR/CIBN has been used to recruit multiple proteins to the plasma membrane
28 including: Raf1 which stimulated the ERK pathway and neurite outgrowth in PC12 cells
(Zhang et al., 2014) and Akt stimulating downstream signaling (Katsura et al., 2015).
Optogenetic tools have the advantage of temporal specificity for probing signaling pathways, and thus scientists began to use them to observe signaling dynamics in the cell. By recruiting the SOS catalytic domain to the plasma membrane using phyB/PIF6 dimerizers,
Ras-ERK signaling pathways were activated (Toettcher et al., 2013). This study demonstrated the ability to regulate signaling of a pathway by activating one singular node, in this case with SOS. By using this approach specific nodes of pathways can be probed, bypassing other signaling events that may complicate downstream effectors. In addition to the ability to stimulate a pathway at intermediate nodes, this study showed the ability to stimulate signaling nodes for precise, user-specified amounts of time, allowing for temporally specific activation/deactivation effects to be understood.
Another cellular pathway that has been regulated successfully with optogenetics is lipid metabolism signaling. Membrane recruitment with CRY2PHR/CIBN interaction was used for recruiting kinases or phosphatases to the plasma membrane. One application was to recruit an inositol 5-phosphatase domain or a PI3-kinase domain (inter-SH2 (iSH2) region of the p85α regulatory subunit of PI3-kinase) to the plasma membrane, which resulted in rapid, local control of phosphoinositide metabolism (Idevall-Hagren et al., 2012). In another experiment the recruitment of the inositol 5-phosphatase domain of OCRL resulted in
PI(4,5)P2 dephosphorylation, loss of membrane ruffling, and disappearance of clathrin coated endocytic pits, and opposite this the recruitment of PI3-kinase resulted in induction of
PI(3,4,5)P3 synthesis and membrane ruffling (Idevall-Hagren et al., 2012).
29 Other subcellular locations have also been targeted with optogenetic approaches.
Organelle cargo trafficking using cytoskeletal motor proteins was achieved with TULIP and
CRY2PHR/CIBN interactions. Both tools recruit cytoskeletal motor proteins to cargos,
including peroxisomes and mitochondria, resulting in light dependent organelle trafficking
(van Bergeijk et al., 2015; Duan et al., 2015). Multiple systems have also been developed to regulate shuttling between the nucleus and cytoplasm. One method fused phyB(1-908) and an NES to a target protein, resulting in cytosolic localization of the target in dark but induced nuclear targeting by stimulating interaction with nuclear PIF3 (Beyer et al., 2015). By fusing
CRY2PHR to cargo and anchoring CIBN to an exosome-associated tetraspanin protein, CD9,
CRY2/CIBN dimerization directed cargo into exosomes (Yim et al., 2016).
Reconstituting Split Proteins
This approach involves attaching split protein halves to a dimerizing pair in order to
confer light dependent reconstitution. It is important that the split protein halves do not have
sufficient affinity on their own. Splitting enzymes has been done with chemical dimerizers
previous to the advent of soluble optogenetic approaches. Split transcription factors are
commonly used in techniques probing protein-protein interactions, such as the yeast two-
hybrid assay (Fields and Song, 1989).
Reconstituting Split Transcription Factors
The first soluble optogenetic tool that demonstrated the ability to reconstitute a split
protein transcription factor used phyB and PIF3 to regulate a split Gal4 binding domain and
activation domain (Shimizu-Sato et al., 2002). Gal4 DNA binding domain was fused to phyB
and Gal4 activation domain was fused to PIF3. While the transcription factor is separated and
inactive in the dark, red-light-induced dimerization of phyB and PIF3 reconstituted the
30 protein and activated transcription. This was reversible with far-red light which dissociated
the fragments leading to inactivation. Light-dependent transcriptional regulation was shown
to be dose-dependent with relation to the intensity of light applied. Many others have
followed up on this technology with novel photoreceptors, but none have the optimal
characteristics for global use in many model systems.
Optogenetic photodimerizer systems have also been used to target endogenous gene
expression using TALE and CRISPR technologies. Zinc finger binding proteins were used to
target a transcriptional activator to specific DNA binding using a light dependentFKF1/GI
interaction (Polstein and Gersbach, 2012). A similar approach was taken with CRY2/CIB1
photoreceptors to recruit an activation domain to TALE DNA binding domains targeting an
endogenous promoter (Konermann et al., 2013). Multiple CRISPR approaches have used a
split dCas9/CRISPR to recruit a VP64 activation domain to an endogenous promoter site
(Nihongaki et al., 2015a, 2015b; Polstein and Gersbach, 2015).
Reconstituting Split Enzymes
Tools regulating genetic activation did not attempt to reconstitute split enzymes, but rather they split a multi-domain protein. The phyB/PIF3 system was again one of the first examples. The system uses the photodimerizers to reconstitute a split protein splicing enzyme, the yeast vacuolar ATPase intein (Tyszkiewicz and Muir, 2008). This system showed the ability to reconstitute split enzymes with light rather than the traditional chemical dimerizer methods.
Cre recombinase is a widely-used enzyme which allows recombination at loxP sites in the genome. The first publication of the CRY2/CIB interactions demonstrated the ability to reconstitute a split Cre recombinase (Kennedy et al., 2010). While this system had little to no
31 dark background, activation was not robust and took intense light exposures to achieve.
Recently, the Tucker laboratory improved upon this split Cre tool using site directed mutagenesis and truncations to create the PA-Cre2.0 system with higher dynamic range, which can be activated by a single pulse of light (Taslimi et al., 2016). In similar timing another photoactivateable Cre system using nMag and pMag photodimerizers was developed which showed robust light reponse and in vivo application (Kawano et al., 2016).
Inducing Protein Interactions
Inducing protein interactions is another method of regulating protein activity. In one of the first optogenetic examples of this technique the phyB and PIF3 interaction was used to modulate actin assembly by recruitment of a weak-affinity, GDP-bound form of Cdc42 to its effector, Wiskott-Aldrich Syndrome Protein (WASP) (Leung et al., 2008). This study showed that cellular events can be regulated through forcing interactions of proteins with naturally weak affinity for one another. It has also been demonstrated that using photodimerizers attached to membrane bound receptors confers light dependent activation through induced interaction. This approach was used by fusing CRY2PHR and CIBN to intracellular domains of tyrosine kinase receptors in order to induce synthetic interaction with: TrkB receptor (Chang et al., 2014) and FGFR receptor (Kim et al., 2014). The LOV domains of aureochrome-1 have also been used to dimerize receptor tyrosine kinases
(FGFR1, EGFR, and RET) (Grusch et al., 2014). CRY2PHR/CIBN interactions were used again to regulate downstream kinases, such as C-Raf and B-Raf through induced homodimerization and heterodimerization using approaches (Wend et al., 2014).
CRY2 oligomerization properties were harnessed to activate biological processes normally triggered by protein oligomerization. TopBP1, a protein that activates ATR kinase,
32 was clustered to induce the DNA damage response in the absence of DNA damage through
CRY2 light induced oligomerization (Ozkan-Dagliyan et al., 2013). CRY2PHR clustering
was again used to stimulate β-catenin or Rho GTPase pathways by light-induced clustering
of CRY2PHR-LRP6c or CRY2PHR-Rac1/RhoA, respectively (Bugaj et al., 2013). In a
follow up study, binding domains fused to CRY2PHR showed increased binding avidity
upon light-induced clustering, providing a means to target and activate endogenous receptors
with light (Bugaj et al., 2015). CRY2olig fused SH3 domains of Nck were used to stimulate
actin polymerization with light (Taslimi et al., 2014).
Caging/Sequestering and Releasing Proteins
Caging
Caging is the idea of blocking an interaction in the dark by steric hindrance and then
releasing it. The most common way caging has been achieved is throught utilization of the
LOV domains. Small peptides are sequestered into the Ja-helix in the dark, and upon light illumination they are exposed and active. This was first used to control membrane perterbations and movement with light by caging Rho GTPase, Rac1, in the Ja-helix (Wu et al., 2009). Others have used the same strategy to cage degron sequences to produce light stimulated degradation tags (Bonger et al., 2014), nuclear localization sequences to stimulate translocation with light (Niopek et al., 2014; Yumerefendi et al., 2015). Another way caging was achieved was through the utilization of two Dronpa monomers which were appended to either side of an enzyme, to cage the catalytic site in dark. Light stimulation released the steric hindrance and exposed the catalytic site of the enzyme (Zhou et al., 2012). This approach was also used to stimulate Cdc42 with GEF proteins, as well as the hepatitis C virus NS3-4A protease (Zhou et al., 2012).
33 Homodimerization Sequesetering
Homodimerization sequestering involves interactions between the same
photoreceptors sequestering proteins from their functional location. UVR8 homointeractions
were the first to be used in order to sequester an endoplasmic reticulum(ER)-processed
protein, VSVG, from trafficking. Using a UVR8-VSVG fusion, in dark, UVR8-UVR8
interactions trapped the fused VSVG in the ER and upon illumination the UVR8 were
dissolved allowing release of VSVG into the ER for trafficking (Chen et al., 2013). This
work demonstrated the importance of protein release in probing cellular trafficking.
The LARIAT (light activated reversible inhibition by assembled trap) system takes advantage of CRY2’s propensity to cluster when associated with a multimeric protein and showed the ability to sequester activity of Vav2, Tiam1, Rac1, Cdc42, RhoG, and tubulin through clustering the proteins (Lee et al., 2014). CRY2olig’s clustering phenotype has also been used to cluster clathrin light chain resulting in light mediated disruption of endocytosis
(Taslimi et al., 2014).
Heterodimerization Sequestering
Heterodimerization sequestration is similar in theory to homodimerization, but it uses heterodimerizing photosensory proteins. Using the phyB/PIF6 interaction, a mitotic cyclin,
Clb2, was sequestered to specific locations within the cell. This involved fusion of Clb2 to
PIF6 and anchoring of phyB to subcellular locations such as the nucleus or spindle pole body. Red light resulted in sequestered activity and disruptions in Clb2-dependent pathways.
Far-red light released Clb2 from sequestration at specific times, for specific durations, testing whether recovery of Clb2 function at those times and durations was sufficient for rescue of
34 altered pathways (Yang et al., 2013). Bem1 was regulated using a similar approach to modulate activity of Cdc42 in yeast (Jost and Weiner, 2015).
LOVTRAP used Zdk to sequester fused proteins of interest to an anchored AsLOV2 location in the dark and then release them upon light illumination. This approach was used to regulate Rho GTPase signaling by sequestering constitutively active versions of Vav2, Rac1, or PI3K at the mitochondria in dark and releasing them with light (Wang et al., 2016). This system can be used with AsLOV2 mutants with half-lives from two seconds to over eight minutes, allowing tunability of sequestration/ release (Wang et al., 2016). These tools demonstrate the ability to regulate protein function through modulating the proteins’ location with light.
Thesis Statement
The studies in this thesis aim to further the field of optogenetics by creating novel tools to regulate protein function in the cytosol and nucleus. Chapter II details an approach to regulate cytosolic protein function with an AsLOV2 domain to create an inducible peroxiomal targeting tag that can be appended to proteins to sequester activity into peroxisomes. In Chapter III, nuclear tools were developed to regulate transcription through light initiated induction or sequestration of activity with CRY2 technologies. While it was clear how the cytosolic peroxisomal system functioned, Chapter IV provided further examination of the nuclear transcription tools to try to determine the mechanism behind the reaction to light. These systems combined aimed to develop novel approaches to control cellular processes that are universally useful to many fields of science such as transcription and protein activity, where other systems showed high background activity and low dynamic range of control over protein activities in dark versus light. Also, we aimed to uncover
35 cellular effects of using endogenous proteins used in optogenetic systems in mammalian cells.
36 CHAPTER II
OPTICAL CONTROL OF PEROXISOMAL TRAFFICKING2
Introduction
Peroxisomes and Peroxisomal Targeting
Peroxisomes are small organelles found in almost all eukaryotic cells, however their morphology differs depending upon the type of tissue and species. These organelles are about
0.1-1µm in diameter with a single lipid bilayer enclosing them from the rest of the cytosol
(van den Bosch et al., 1992). Peroxisomes are highly enzymatic, with over 50 enzymes known to exist within them. Major roles of these enzymes are detoxification of chemicals, such hydrogen peroxide, and beta-oxidation of fatty acids. Due to the role in detoxification of potentially hazardous chemicals, the membrane remains tightly closed off to the rest of the cytosol even during division through a process of highly regulated fission. Various species have other specializations for their peroxisomes including human, whose peroxisomes are also involved in the synthesis of cholesterol, bile acids, and ether lipids.
The number of peroxisomes per cell is highly variable depending upon not only the type of cell but also the level of metabolic requirements. Due to the role in detoxification and synthesis of bile acids in humans the liver’s hepatic cells are highly abundant in peroxisome, but other detoxifying cells such as those in the kidneys also are rich in peroxisomes. In comparison yeast cells may have one or two peroxisomes, but higher numbers (and size) can be induced in Saccharomyces cerevisiae by growing them with fatty acids as their sole
2 A portion of this research is published with permission from the American Chemical Society Synthetic Biology: Spiltoir et al. 5(7) (2016): 554-560, © ACS Publications
37 source of carbon (Kunau et al., 1995; Veenhuis et al., 1987). Due to the peroxisome’s ability
to adapt in number, morphology, and contents to environmental stimuli, the role of
peroxisomal biogenesis was studied in various organisms. This led to many years of heated
debate of how peroxisomes originated and replicated (Kunau, 2005). The debate ended when
conclusive evidence that de novo peroxisomes originate from the endoplasmic reticulum was published (Hoepfner et al., 2005) and replicated through a process of fission (Hoepfner et al.,
2001; South and Gould, 1999).
Since peroxisomes are enzymatic powerhouses of the cells they are essential for the health of the cells and organism. There are a group of 32 known genes called peroxins or
PEX genes that function in the maintenance and inheritance of peroxisomes in eukaryotes.
These genes are necessary for proper peroxisomal formation of the membrane, compartmentalization of proteins in the matrix, and proliferation of new peroxisomes.
Mutations in these genes are debilitating for peroxisomal biogenesis and can lead to a number of debilitating or fatal diseases in humans. Examples of peroxisomal biogenesis disorders are Zellweger spectrum syndrome, which is the most common and includes
Zellweger syndrome, infantile Refsum disease, and neonatal adrenoleukodystrophy. Other examples are rhizometlic chondrodysplasia punctate type 1 (RCDP-1) as well as multiple other less life threatening disorders such as hyperoxaluria type 1, pipecolic academia, acatalasia and a multitude more. Zellweger spectrum syndrome disorders are characterized by a multitude of symptoms involving a build up of large branch-chain fatty acids causing developmental brain disorder which lead to a multitude of other developmental malformations. Other disorders have a range of symptoms that can be relatively benign to life threatening, but most do not have a cure so treatments target symptom relief.
38
Figure II-1: Peroxisomal targeting with PTS1 or PTS2. Pex5 and Pex7 recognize PTS1 and PTS2 respectively. Pex7 complexes with Pex18 and Pex21 to form a receptor cargo complex. These import receptors shuttle the PTS containing cytosolic protein to the peroxisomal membrane for import into the peroxisomal lumen. Pex5 is known to be either recycled back out to the cytosol or polyubiquitinated for proteosomal degradation. Figure 1 was adapted from Paul and colleagues’ review (Paul et al., 2013).
39 While few organelles such as mitochondria have DNA within them, the peroxisome
relies on nuclear genes to synthesize all of its components and therefore has specific targeting
methods to transport peroxisomal proteins to the correct location. A short peptide sequence at
the N- or C- terminal tail of the targeted peroxisomal protein is able to signal peroxisomal
import receptors, shuttling chaperones, to bind the protein and traffic them to the
peroxisomal membrane where they are imported into the peroxisomal lumen. Membrane
bound peroxisomal proteins are recognized and imported by their own receptor and therefore
will not be discussed.
The peroxisomal targeting sequence 1 or 2 (PTS1 or PTS2) designates that the signal
is on the C- or N-terminal tail of the lumen targeted protein respectively. The PTS2 is
recognized by peroxisomal import receptor PEX7 which complexes with PEX18 and PEX21
to form the receptor cargo complex. The PTS1 containing proteins are recognized by
peroxisomal import receptor PEX5, which is currently not thought to complex with other
import proteins, but alone can shuttle targeted proteins to the peroxisome (Figure II-1). The
PTS2 is 26 amino acids in length, but the PTS1 has a minimal motif of three amino acids
serine-lysine-leucine (-SKL), which have been determined to be sufficient for peroxisomal
targeting for non-perixosomal proteins (Gould et al., 1987, 1989). This short PTS1 motif is
used for the majority of peroxisomal proteins and is ideal for the use in optogenetic systems
such as the one discussed in this chapter because it can easily be appended onto proteins and
masked.
Optogenetic Peroxisomal Trafficking Tool Design
The blue-light-responsive LOV2 domain of Avena sativa phototropin1 (AsLOV2) has been used to regulate activity and binding of diverse protein targets with light as discussed in
40 the introduction. In the dark, a C-terminal Jα-helix is bound tightly to the core of AsLOV2
(Harper et al., 2003), and blue light triggers covalent bond formation between a conserved
cysteine residue on the LOV domain and a flavin mononucleotide chromophore (Salomon et
al., 2000; Swartz et al., 2001). This results in unwinding of the Jα-helix and dissociation from
the LOV core, ultimately affecting phototropin kinase activity (Halavaty and Moffat, 2007;
Harper et al., 2003, 2004).
Many optogenetic tools have utilized these structural changes that occur between the
main LOV domain and the Jα-helix (Bonger et al., 2014; Guntas et al., 2015; Lee et al., 2008;
Lungu et al., 2012; Niopek et al., 2014; Pham et al., 2011; Rao et al., 2013; Renicke et al.,
2013; Schmidt et al., 2014; Strickland et al., 2010, 2012, 2008; Wu et al., 2009; Yumerefendi
et al., 2015). Our aim was to develop a light inducible peroxisomal targeting tool, and to do
this we cloned a PTS1 into the C-terminus of the Jα-helix of AsLOV2 (Figure II-1). A LOV-
PTS1 tag would be easy to fuse onto the C-terminus of proteins to regulate peroxisomal
targeting, and the small size of the tag would make it ideal for appending onto other proteins.
LOV Domain Mutations That Alter Kinetics
In this study, we use multiple variations of the AsLOV2 in order to alter the kinetics
of our tool to provide a longer half life and to stabilize dark state docking of the Jα-helix. A
V416I variant of AsLOV2 was shown to have a longer half-life of ~3 min (Strickland et al.,
2012; Zoltowski et al., 2009) and the core also contained mutations T406A and T407A which were shown to increase Jα-helix docking (Strickland et al., 2012) and G528A and N538E which have been shown to be helix-stabilizing mutations (Strickland et al., 2010). With those mutations in the main core specialized mutations were also used to alter the characteristics including a deletion, ΔK533, which renders the LOV domain constitutively active with
41 A Pex5
GFP GFP Peroxisome LOV Pex5 PTS1 Light LOV PTS1
Figure II-2: Model of Optogenetic Peroxisomal Trafficking Tool. (A) Schematic of caged LOV-PTS1 construct showing the C-terminal Jα-helix that dissociates from the LOV core upon illumination and uncages the PTS1 sequence. Pex5 binds PTS1 and shuttles the protein to the peroxisome.
α
42 the Jα-helix in an undocked state. A mutation to strengthen dark state docking of the Jα-helix is I532A which has been used in multiple studies to reduce background activity when using
AsLOV2 as the photoreceptor (Strickland et al., 2010, 2012).
Materials and Methods
Strains and Plasmid Construction
Relevant primer sequences used in cloning described (Table II-1). We used PCR to
amplify the ‘TULIP’ LOVpep construct pDS248 (Strickland et al., 2012), containing an erbin
binding peptide at the C-terminus of GFP-AsLOV2-Jα. This construct contains residues 404-
540 of AsLOV2 (numbering corresponds to the corresponding location in the wild-type
Arabidopsis phototropin 1 sequence), as well as additional mutations T406A, T407A, and
V416I in the AsLOV2 core. In addition, we added helix-stabilizing mutations G528A and
N538E. A canonical PTS1 peroxisomal targeting sequence, ‘LQSKL’, was added at the C-
terminus, replacing the erbin PDZ domain. To generate the constitutively active ΔK533
variant, we used pDS420 as template (Strickland et al., 2012). The GFP-LOV-PTS1 insert
was cloned into pcDNA3.1 at BamHI and EcoRI sites, using primers 1068f/1070r (LOV-
PTS1.1), 1068f/1072 (LOV-PTS1.2), 1068f/1071r (LOV-PTS1.3), or 1068f/1069r (LOV-
PTS1(∆K533)). To create a tetracycline regulated construct, a nuclear export signal was
cloned into pcDNA3.1-GFP-LOV-PTS1 at HindIII and BamHI to create pcDNA3.1-NES-
GFP-LOV-PTS1. NES-GFP-LOV-PTS1 was PCR-amplified and cloned into pTRE3G-luc
between SalI and EcoN1 using primers 1713f/ 1714r. Site directed mutagenesis was used to
change Ile532 to Ala using primers 1567f/ 1568r. The mCherry-AGTma construct was
generated by first inserting mCherry into pcDNA3.1 between BamHI and EcoRI using
43
44 primers 1347f/1373r, and then inserting the coding sequence for alanine: glyoxylate aminotransferase at EcoRV and XhoI using 1548f/1541r. For yeast two-hybrid studies, the
Gal4AD-Pex5 fusion protein was in pGADT7rec (Clontech). LOV-PTS1 was cloned by homologous recombination in pDBTrp (a version of pDBLeu (Invitrogen) with a TRP1 selection marker) (Tucker et al., 2009) using primers 1084f/1085r. The mCherry-FRB construct was generously provided by Dr. Matthew Kennedy (University of Colorado School of Medicine). 2xFKBP was amplified from 2xFKBP-GFP-homer1c (provided by Dr.
Kennedy) by PCR and cloned into the pcDNA3.1-GFP-LOV-PTS1.1 vector at KpnI and
BamHI sites.
Indirect Immunofluorescence
HEK293T and COS-7 cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100 in PBS with 5% normal goat serum, and incubated with an α-PMP70 monoclonal antibody (Sigma clone 70-18, SAB4200181, 1:100 dilution) and Cy3-conjugated goat anti-mouse secondary antibody (Jackson Immunoresearch 115-165-146, 1:500 dilution).
Microscopy and Image analysis
Imaging was performed using two systems:
1) An Olympus IX71 microscope equipped with a spinning disc scan head (Yokogawa
Corporation) with a 60x/NA 1.4 objective. Excitation illumination was delivered from an
AOTF controlled laser launch (Andor Technology) and images collected on a 1024 x 1024 pixel EM-CCD camera (iXon; Andor Technology). The emission bandpass filters were
525/30 (GFP), and 685/36 (mCherry). Metamorph software was used for collection of images on this system. For light-sensitive experiments, cells were protected from ambient light sources by focusing in the presence of filtered light (572/28 bandpass filter, Chroma).
45 2) A Zeiss AxioObserver Z1 Inverted Spinning Disc microscope with a 63x/NA 1.4
objective and HQ 480/40x and HQ565/30x (Chroma) bandpass filters. Excitation
illumination was delivered by a 3i Ablate!™ Model 3iL13 and image collected using a
Yokogawa CSU-X1CU camera. Slidebook 6 was used for collection of images on this
system.
ImageJ 1.45s and Fiji were used for image analysis. Percent of protein in puncta was
calculated by first determining the total fluorescence within the cell or cell region and
subtracting background. Then, the total fluorescence within peroxisomal puncta (background
subtracted as well was determined using ImageJ thresholding (Otsu’s method) to delineate
protein within puncta (also background subtracted). The % fluorescence within peroxisomes
was calculated by dividing the fluorescence within puncta by total fluorescence within each
cell or analyzed region.
Cell Culture and Live Cell Imaging
HeLa, HEK293T, and COS-7 cells were maintained in Dulbecco’s modified Eagle
medium (DMEM) supplemented with 10% FBS and 1% penicillin-streptomycin at 37o C
with 5% CO2. Cells were seeded onto a 35 mm glass bottom dish (live cell images) or
coverslips on a 12 well plate (fixed images) and transfected using Lipofectamine 2000
(Invitrogen) according to the manufacturer’s protocol. Dark samples were wrapped in
aluminum foil immediately after transfection, and all manipulations carried out using a red
safelight. Unless otherwise indicated, light-treated cells were illuminated using a custom
programmable blue LED light source (461 nm) with a 1 s pulse per 1 min interval, 5.8
2 mW/cm . Media in glass bottom dishes was changed to HBS with 1 mM CaCl2 directly before imaging. Live cell imaging was performed at 34o C.
46 Yeast Studies
Yeast two-hybrid studies were performed using strain AH109 (MATa, trp1-901, leu2-
3, 112, ura3-52, his3-200, gal4∆, gal80∆, LYS2::GAL1UAS-GAL1TATA-HIS3, GAL2UAS-
GAL2TATA-ADE2, URA3::MEL1UAS-MEL1TATA-lacZ, MEL1). Yeast were transformed with
indicated GalBD and GalAD fusion plasmids, then grown on SC -Trp/-Leu/-His/ +3mM 3-
AT for 3 days at 30˚C in light or dark. The AD-control used was a pGADT7rec-CIB1
construct that is not expected to bind Pex5.
Results
Optogenetic Regulation of Peroxisomal Trafficking
To test the model of controlling peroxisomal trafficking through a LOV-PTS1 tag we
developed a constitutively active form of the tag using AsLOV2(∆K533) fused to the C-
terminus of GFP. The ∆K533 mutant of AsLOV2 has a permanently unwound Jα-helix
(Strickland et al., 2012) and when an acyl-CoA oxidase PTS1 (-LQSKL) is fused to the end of the C-terminal Jα-helix this short peptide sequence would be continually available for
PEX5 interaction regardless of light illumination. Expression of the GFP-LOV2(∆K533)-
PTS1 showed a punctate pattern resembling peroxisomal localization (Figure II-3A). After realizing the feasibility of using the LOV-PTS1 tag we tested a series of Jα-helix truncations with AsLOV2 to screen for light dependent peroxisomal targeting (Figure II-3B). The
AsLOV2 domain used for initial studies also had T406A and T407A mutations that increase
Jα-helix docking and a V416I mutation
47 A B peroxisomal Amino Acid Sequence localization 527 530 535 540 545 Dark Light AsLOV2 EGVMLIKKTAENIDEA A KEL PDA *LOV-PTS1.1 EAVMLIKKTAEEIDEA L QSK L - + LOV-PTS1.2 EAVMLIKKTAEEIDEA SKL -- LOV-PTS1.3 EVMLIKKTAEILQSAE KL + +
CD GFP-LOV-PTS1 α-PMP70 Merged
40 Dark 30
20 in peroxisome
% Fluorescence 10 Light Dark Light (4 hr)
Figure II-3: Optogenetic Regulation of Peroxisomal Trafficking. (A) A GFP- LOV(ΔK533)-PTS1 fusion protein carrying a mutation that eliminates Jα-helix docking (ΔK533) shows peroxisomal localization even without blue-light illumination. Scale bar, 5µm. (B) Alignment of tested sequences with AsLOV2-Jα sequence. Numbering corresponds to amino acid residue in the plant AsLOV2 sequence. LOV-PTS1.1 (*) was used in further studies. Orange residues indicate regions where mutations were introduced. Yellow residues indicate the PTS1 signal. ‘+’ depicts < 5% of cells show peroxisomal localization; ‘-‘ depicts > 95% of cells show peroxisomal localization. (C) Localization of GFP-LOV-PTS1 in fixed COS-7 cells exposed to light or dark (representative cells shown). COS-7 cells were transfected with GFP-LOV-PTS1 and exposed to dark or blue light pulses for 24 hrs,18 hrs after transfection. Localization of the peroxisomal marker protein PMP70 is shown at left. Scale bar, 10µm (D) Quantification of peroxisomal fluorescence in HeLa cells. Cells were transfected with GFP-LOV-PTS1, incubated in dark for 18 hours, then imaged every 10 min for 3 hrs. During imaging, cells were exposed to additional LED light treatments. Dark treated quantification used the initial image capture at time 0, while light treated quantification used the last image in the series (at 3 hr). Data represents the average and standard deviation of one experiment, n = 8. This experiment was repeated several times with similar results.
48 that prolongs the illuminated state (Strickland et al., 2012; Zayner et al., 2012; Zoltowski et
al., 2009) in addition to mutations, G528A and N538E, that were shown to increase the
dynamic range of previously developed LOV-based tools (Strickland et al., 2010). Insertion
of the PTS1 in the Jα-helix truncations displayed varying levels of peroxisomal trafficking
abilities ranging from a inability to traffic GFP to the peroxisome (PTS1.2) to trafficking to
the peroxisome without light dependence (PTS1.3). The construct GFP-LOV-PTS1.1 with a
full acyl-CoA oxidase PTS1 starting at amino acid 543 displayed clear light dependent
trafficking to puncta, which colocalized precisely with a perixomal marker, PMP70 (Figure
II-3C). This PTS1.1 tag will herein after be referred to as LOV-PTS1 for simplicity.
Quantification of the peroxisomal import in HeLa cells showed low levels of background
(dark-state) localization in peroxisomes (1.9+/-1.9% total fluorescence in peroxisomes), but
within four hours of light stimulation the peroxisomal import was 31.7+/-6.5% of total
fluorescence in peroxisomes (Figure II-3D).
Tracking Peroxisomal Import Kinetics with Live-Cell Imaging
In order to visualize the kinetics of peroxisomal import with the LOV-PTS1 tag, a
live-cell imaging approach was taken. HeLa cells were transiently transfected with a NES-
GFP-LOV-PTS1 construct containing a nuclear export signal to deplete the nuclear pool of protein. After 18 hours the cells were exposed to blue light pulses and time-lapse microscopy was used to visualize import. Within 10-20 minutes of illumination peroxisomal puncta were detectable (Figure II-4A), similar to studies of peroxisomal trafficking of permeabilized or injected cells (Stephen, 1992; Wendland and Subramani, 1993). Quantification of the import over four hours shows the beginning of a plateau in import, which may represent a saturation of the import machinery available for trafficking. Experimentation with other light regimes
49 A Time (min) after 488nm light stimulation 0 30 60 90
NES-GFP- LOV-PTS1
LOV-PTS1
AGTma mCh- 120 150 180 AGTma
NES-GFP- LOV-PTS1
LOV-PTS1
AGTma
B 30
20
10 in peroxisome % Fluorescence 0 0 50 100 150 200 Time (minutes)
Figure II-4: Kinetics of photostimulated peroxisomal import in HeLa cells over four hours. (A) Time lapse images of representative HeLa cell expressing NES-GFP-LOV-PTS1 exposed to blue light pulses for 4 hrs,18 hrs after transfection. Peroxisomes are highlighted using a mCherry-AGTma reporter that localizes to the peroxisome. DNA was transfected at a ratio of 3:1 (NES-GFP-LOV-PTS1 to mCherry-AGTma). Scale bar, 10µm. (B) Quantification of image sequence in (A).
50 informed us that a singular pulse of blue light was unable to direct a detectable amount of
peroxisomal import but rather continuous pulsing was required. It is unclear how long the
PEX5 association with the PTS1 lasts after uncaging of the LOV-PTS1 occurs.
Peroxisomal Translocation is Due to Light-Dependent Binding to PEX5
Using a yeast two-hybrid approach we probed whether the light-dependent import of
GFP-LOV-PTS1 was indeed due to light-dependent binding to PEX5, the peroxisomal import receptor (Figure II-5). The yeast two-hybrid arrangement was LOV-PTS1 fused to a
Gal4 DNA binding domain which was tested for interaction with a PEX5-Gal4 activation
domain fusion protein (Figure II-5A). While LOV-PTS1 and PEX5 did not interact in the
dark, they interacted upon blue light illumination as detected by yeast growth from a HIS3
reporter gene (Figure II-5B).
Comparison of Peroxisomal Import with Mutants and Fluorescent Reporters on LOV-PTS1
In an effort to optimize the LOV-PTS1 tool we compared the peroxisomal import of
the original GFP-LOV-PTS1 with multiple AsLOV2 mutants and different fluorescent
reporters in HEK293T cells (Figure II-6A). After 24 hours of illumination the original GFP-
LOV-PTS1 showed 35.7 +/- 9% of total protein in the peroxisome with only 3.3 +/-3%
localized in the dark. The constitutively active GFP-LOV(∆K533)-PTS1 was able to achieve
53 +/-7% total protein in the peroxisome in the same timeframe, which was higher than a mCherry tagged version of a naturally peroxisomal protein, alanine: glyoxylate aminotransferase major allele (mCh-AGTma), at only 38 +/- 6% total protein in the peroxisome. We also tested light dependence of a mCherry fluorescent protein tagged with our LOV-PTS1 (mCh-LOV-PTS1), which underperformed the equivalent GFP version when illuminated with blue light but showed a much higher dark background level of import.
51 A AD Pex5 AD Pex5 Light LOV LOV PTS1 BD HIS3 BD HIS3 GalUAS GalUAS B Dark Light
LOV-PTS1 + AD-control
LOV-PTS1 Pex5 LOV-PTS1 BD-LOV-PTS1 + AD-vector + AD-PEX5
Figure II-5: Peroxisomal translocation is due to light-dependent binding to Pex5. (A) Schematic showing yeast two-hybrid assay testing interaction of LOV-PTS1 with PEX5. (B) BD-LOV-PTS1 does not interact with AD-PEX5 in the dark, but shows interaction under blue light illumination.
52 A 60
40
20
% Fluorescence in peroxisome Dark Light Dark Light
mCh-LOV- GFP-LOV- GFP-LOV mCh-AGTma PTS1 PTS1 (∆K533)-PTS1 BC Dark Blue Light 25
20
15 GFP-LOV-PTS1 10
5
0 Dark Light (4 hr) % Fluorescence in peroxisome I532A Mutant I532A I532A Mutant GFP-LOV-PTS1
D Time post initial light stimulation 0 min 30 min 60 min 90 min 120 min
Figure II-6: Varying Levels of Peroxisomal Localization of LOV-PTS1 Tag Mutants and Fluorescent Reporters. (A) Quantification of levels of peroxisomal protein with indicated PTS1-fused peroxisomal targets. HEK293T cells were tranfected and fluorescence quantified after 24 hrs. Light samples received 24hr blue light pulses. Data represents the average and standard deviation of one imaging experiment, n = 8. (b-d) Analysis of higher caging AsLOV2 mutant I532A in GFP-LOV-PTS1. (B) COS-7 cells were transfected with GFP-LOV-PTS1 or GFP-LOV-PTS1(I532A) and incubated in dark or 24 hrs light. Scale bar, 10μm (C) Quantification of peroxisomal localization in HeLa cells transfected with GFP- LOV-PTS1(I532A). Cells were incubated in dark 18 hours, then imaged for GFP fluorescence every 5 min for 3 hrs. The first image was used to quantify dark background, and the last image for light quantification. Cells were exposed to light pulses from an external LED during imaging. Data represents the average and standard deviation of one imaging experiment, n = 7. This experiment was repeated two times with similar results. (D) Image series of cells expressing GFP-LOV-PTS1(I532A) treated as in c.
53 In order to reduce the dark state background localization seen with our original
construct known mutations were explored that increase docking of the Jα-helix to the LOV domain in the dark state. Literature showed an I532A mutation on AsLOV2 was able to stabilize the dark state and reduce spontaneous unwinding of the Jα-helix (Strickland et al.,
2010). Using the I532A mutation on our LOV-PTS1 tag was able to eliminate detectable dark state localization but also decreased the light initiated import after four hours to about half, 16.2 +/-4% total fluorescence in the peroxisome, of the non-mutant version (Figure II-
6B and C). Using live-cell imaging the lack of visible puncta at the initiation of the experiment was seen and within 30 minutes of light exposure the peroxisomal puncta were detectable (Figure II-6D). While the overall light induced import is decreased with the I532A mutation, the spontaneous localization of the dark state was eliminated.
Cargo Protein Stays Relatively Stable at Peroxisomes After Trafficking
To test whether this tool could be used to sequester the protein more permanently in
the peroxisomes we used the GFP-LOV(I532A)-PTS1 and initiated import for 18 hours
followed by a dark chase for 24 hours with protein at steady state levels. The I532A mutation
was able to keep the background levels to 0.5% total protein in the peroxisomes after a total
of 48 hours post-transfection. The samples that were primed with peroxisomal import for 18
hours showed no significant difference between levels of peroxisomal protein between cells
quantified at the start of the dark chase or 24 hours after dark incubation (Figure II-7A).
These data suggest that over a time period of 24 hours levels of protein in the peroxisomes
are relatively stable.
54 A N.S. 60
40
20 % Fluorescence in peroxisome
0 light 24 light 24 dark
Figure II-7: Peroxisomal Cargo Protein is Relatively Stable Over 24 Hours. (A) Quantification of peroxisomal localization in HEK293T cells transfected with GFP-LOV- PTS1(I532A). Cells were either incubated in dark for the 48 hrs (‘Dark’), or in dark for 6 hrs, then in light for 18 hrs to induce peroxisomal import. Samples were quantified at this initial time (t=0), or after an additional 24hr dark incubation (t=24). Data represents the average and standard deviation, n = 8. N.S., p > .05.
A Dark Light (4 hr) Light (8 hr)
B 70 * 60 50 40 30 peroxisome 20 % Fluorescence in 10
Dark Light Light 4 hr 8 hr
Figure II-8: Light-triggered depletion of a cytosolic reporter protein. HEK293T cells expressing NES-GFP-LOV-PTS1 were incubated initially in the dark for 18 hrs. At 18 hrs, cycloheximide was added to block translation and light treatment was initiated. GFP localization was quantified 0, 4, and 8 hrs after initiation of light treatment. (A) Images of representative cells. Scale bar, 10µm. (B) Quantification of percent peroxisomal protein at 0, 4, or 8 hrs. Data represents average and standard deviation of one experiment; this experiment was repeated 2 times with similar results. Dark analysis was from 4 cells; 4 and 8 hr analysis was from 8-16 cells *, p-value < .005.
55 Peroxisomal Sequestering of a Cytosolic Reporter Protein
To more accurately depict how much protein is being sequestered into the
peroxisomes, new protein synthesis was blocked using cycloheximide. This
allowed for quantification of peroxisomal import of a singular bulk amount of protein. Using
cytosolic reporter NES-GFP-LOV-PTS1 there was only 3% peroxisomal localization in dark, but blue light illumination resulted in significant depletion of cytosolic protein into the peroxisome to 55 +/-6.8% of total cytosolic protein (Figure II-8A and B). This robust sequestering of the protein into peroxisomes was dose dependent with duration of light illumination. This suggested the LOV-PTS1 tag could be used as a light-regulated trafficking signal to sequester a protein of interest from the cytosol, a tactic that has been used with other optogenetic tools (Crefcoeur et al., 2013; Lee et al., 2014; Niopek et al., 2014; Yang et al.,
2013; Yumerefendi et al., 2015).
The LOV-PTS1 Tag is Unable to Traffic Rapamycin-Induced Dimers into Peroxisomes
With the notion of using the LOV-PTS1 tag to regulate endogenous proteins that
interact in the cell, we designed an experiment using proteins that dimerize upon the addition
of rapamycin. Multiple studies suggest that proteins can be imported into the peroxisome as a
‘piggybacking’ pair (Glover et al., 1994; Islinger et al., 2009; McNew and Goodman, 1994),
so our approach tests if our tag can deplete interacting proteins from the cytosol. We used
inducible dimerizers FRB and FKBP, which do not interact when coexpressed but dimerize
in the presence of rapamycin (Spencer et al., 1993). FKBP12 was fused to the N-terminus of
GFP-LOV-PTS1 and this was coexpressed with a mCherry-labeled FRB (mCh-FRB) (Figure
II-9A). When expressed alone or coexpressed with mCh-FRB in the absence of rapamycin, peroxisomal trafficking of FKBP-GFP-LOV-PTS1 could be triggered upon light
56 A
me? B
Figure II-9: Rapamycin-induced dimerization blocks light-triggered peroxisomal import of GFP-LOV-PTS1. (A) Schematic of constructs used in studies. (B) When coexpressed in COS-7 cells with mCherry-FRB, FKBP-tagged GFP-LOV-PTS1 is inducibly targeted to the peroxisome in the presence of a DMSO control, but targeting is blocked with addition of rapamycin, which induces FRB-FKPB dimerization. Constructs were transfected at a ratio of 2:1 (FKBP-GFP-LOV-PTS1 to mCherry-FRB). Translocation of FKBP-GFP- LOV-PTS1 expressed alone is not affected by DMSO or rapamycin. Scale bar, 10µm.
57 illumination; however, when the dimerizing constructs are co-expressed in the presence of
rapamycin, no peroxisomal import could be detected for either construct (Figure II-9B).
Testing the effect of rapamycin demonstrated that it has no effect on peroxisomal import.
These data suggest that the LOV-PTS1 is not sufficient for trafficking a multimeric
protein, in the form of dimerizing proteins, into the peroxisome. Recent studies have
suggested that peroxisomal import of oligomeric proteins can be inefficient (Freitas et al.,
2015), which proposes that oligomeric proteins or protein complexes may have other factors
that influence their import into peroxisomes that are currently not known. In addition, it
cannot be excluded that dimerization with FRB-mCh may sterically interfere with the
interaction of LOV-PTS1 and Pex5.
LOV-PTS1 with Actin Nucleating Factor, VCA, is Unable to Traffic to Peroxisomes
In a design to regulate actin cytoskeleton with the LOV-PTS1 tool, the Verprolin,
Central, Acidic (VCA), the C-terminal domain of the Wiskott-Aldrich Syndrome Protein,
was fused to the NES-GFP-LOV-PTS1. This domain in isolation is a constitutively active actin nucleation regulator, as the N-terminus acts to suppress its activity in the wildtype protein. We hypothesized VCA-NES-GFP-LOV-PTS1 would be available to stimulate actin polymerization in the dark state, but illumination would sequester the protein into peroxisomes and block VCA activity (Figure II-10A). The results revealed VCA-NES-GFP-
LOV-PTS1 was unable to be trafficked into peroxisomes upon light illumination (Figure II-
10B) and overexpression of the VCA containing construct disrupted actin cytoskeleton
(Figure II-10C and D). Peroxisomes are known to be associated with actin fibers, which are important for mobility of the organelle (Hoepfner et al., 2001; Schollenberger et al., 2010), it
58 A
VCA- VCA-
X"
B 488nm-Light-SRmulaRon- $AcRn- Control-- mCh NES$GFP$LOV$PTS1- $AcRn- mCh
VCA$NES$GFP$LOV$PTS1--
C NES-GFP-LOV-PTS1 + mCh-Actin
D VCA-NES-GFP-LOV-PTS1 + mCh-Actin
Figure II-10: Fusion of VCA to LOV-PTS1 tagged protein disrupts actin cytoskelton and blocks peroxisomal import. (A) Schematic of constructs used for experiments (B) When coexpressed in COS-7 cells with mCherry-Actin, NES-GFP-LOV-PTS1 is targeted to the peroxisome with light, but targeting is blocked by fusion of VCA. (C and D) Magnified images of the mCherry-Actin filaments with (C) NES-GFP-LOV-PTS1 alone and (D) VCA fused NES-GFP-LOV-PTS1 which shows disruption of the actin fibers.
59 cannot be discounted that the disruption of actin cytoskeleton may also interrupt proper trafficking to these organelles.
Discussion
These data together show the ability of a LOV-PTS1 tag to control peroxisomal localization of proteins of interest with light. The highest dynamic range was achieved with an acyl-CoA oxidase PTS1 (-LQSKL) fused within the Jα-helix of AsLOV2 starting at amino acid A543. This placement mimic sites used in the optogenetic TULIP dimerizers (Strickland et al., 2012) and is close to where others have caged short peptide sequences such as the degron sequence (-RRRG) (Bonger et al., 2014) and a ssrA peptide for dimerization (Lungu et al., 2012). Multiple groups have also successfully caged variations of nuclear localization signals at Ile539, Asp540, or to cage a bipartite NLS, Lys544 (Niopek et al., 2014;
Yumerefendi et al., 2015), which have allowed them to control cellular trafficking between compartments in the cell similar to the way our tool was intended. Our results and others show the comprehensive way in which the AsLOV2 caging mechanism can be utilized by fusing short peptides into the Jα-helix within residues 539-544. Each tool has slightly different kinetics and dynamic range, which suggests even with the same photoreceptor each tool needs to be experimentally tested for novel uses. Also, as we saw with our truncations
(Figure II-3B), placement of the peptide can produce drastically different results, so cloning optimization is necessary.
The LOV-PTS1 tag demonstrated some inefficiency throughout these studies. The tool was not able to completely sequester the cytosolic reporter protein within eight hours, despite new protein synthesis being turned off. Potential saturation of the import machinery may be at play. Due to the fact that we are highly overexpressing these proteins, there is the
60 possibility that expressing proteins at lower levels equivalent to endogenous genes we may see better results. Our attempts to regulate interacting proteins and cytoskeletal regulators were unsuccessful, which suggests that the PTS1 is not universally sufficient for import into the peroxisome. There may be specific requirements to importing multimeric proteins into
peroxisomes.
With optimization, we think our tool could have use in studying various disease states
and remaining biological questions related to peroxisomal trafficking. With the LOV-PTS1
tag, users can inducibly target proteins to the peroxisome at specific times, or within defined
locations, and to study kinetics of peroxisomal import and to track peroxisomal cargo. The
various peroxisomal biogenesis diseases such as Zellweger Syndrome Spectrum, where
mutations cause aberrant peroxisomal function, can be explored through trafficking specific
mutations on PEX proteins to peroxisomes during biogenesis to tease apart their roles.
Mislocalization of peroxisomal proteins, as with metabolic disease primary hyperoxaluria
(Danpure et al., 1989), can be explored. Since the peroxisomal import machinery can reach
saturation (Wendland and Subramani, 1993), we posit that the LOV-PTS1 module may also
be utilized in competition experiments, broadly blocking peroxisomal protein import at
specific times and locations.
We also envision this tool being used broadly to deplete protein from the cytosol,
however this process is slow and does not completely remove the protein from the cytosol
within eight hours, but, these timescales, are comparable to other approaches to deplete
protein activity with light through inducible degradation (Bonger et al., 2014; Renicke et al.,
2013). With the ubiquitous nature of peroxisomes and the conserved mechanism of import
we believe this tool can be optimized to many systems and novel uses.
61 CHAPTER III
REGULATING TRANSCRIPTION WITH OPTOGENETICS3
Introduction
History of Regulating Transcription
As the first step to gene expression, transcription is a key regulatory factor in the cell
and thus the means to precisely control it is a powerful tool. The first inducible system was a
crude method of control using a heat shock promoter as the regulatory agent upstream of a
gene. Activation was achieved by heat shocking the model organism, usually a transgenic
animal (Blochlinger et al., 1991; González-Reyes and Morata, 1990; Ish-Horowicz and
Pinchin, 1987; Ish-Horowicz et al., 1989; Schneuwly et al., 1987; Steingrimsson et al., 1991;
Struhl, 1985). A more spatially specific, but less tunable method of gene regulation was to
use a well characterized, specific promoter to regulate the gene, such as a tissue specific
promoter which limited transcription to a subset of cells (Parkhurst and Ish-Horowicz, 1991;
Parkhurst et al., 1990; Zuker et al., 1988). These methods often involve transgenic models
and lack the fine-tuning that present transcriptional regulation systems have achieved.
Bidirectional control over transcription was achieved when a chemically regulated
system was developed using the control elements of a tetracycline-resistance operon with a
tetracycline transactivator protein to control a downstream reporter gene with the presence or
absence of tetracycline or doxycycline (Gossen and Bujard, 1992; Gossen et al., 1995). The
Tet-On/Tet-Off regulatory systems demonstrated dose-dependent chemical control over
3 A portion of this research is under review by Nucleic Acids Research: Pathak and Spiltoir et al. © Oxford University Press
62 transcription, which revolutionized the ability to regulate genes in vivo. Similar regulatory systems such as the GalUAS regulatory system using Gal4 transcription factor were developed showing similar tunability (Brand and Perrimon, 1993). While the dose dependent nature of these chemical systems was a substantial improvement over previous systems, they lacked spatial resolution and fast kinetics. In addition, chemicals can be difficult to completely remove from cells and the most aggressive methods of chemical removal involving replating of the cells. In a study it was found that doxycycline, the Tet-system regulator, binds nonspecifically to extracellular matrix and cell components leading to a slow release of the chemical throughout washes (Rennel and Gerwins, 2002).
Light regulation provided an alternative method to potentially overcome the limitations of chemical means of regulation. One of the first soluble optogenetic systems focused on using light to regulate transcription in yeast by reconstituting a split Gal4 transcription factor using the phytochrome B and PIF3 light dependent interaction (Shimizu-
Sato et al., 2002). This study paved the way for optimizing usage of light to regulate transcription with a variety of optogenetic photoreceptors (Crefcoeur et al., 2013;
Konermann et al., 2013; Liu et al., 2012; Motta-Mena et al., 2014; Müller et al., 2013b;
Polstein and Gersbach, 2012; Wang et al., 2012; Yazawa et al., 2009). While these systems have shown great promise in regulating transcription, the level of background and dynamic range leave room for optimization in order to create a universal system that works in all cell types and model system.
63 A VP16 CIB1 CRY2 GalBD reporter
B 12 Blue light Dark Reporter 8 Only 4
Luciferase activity
(Fold Over Reporter) 0
Figure III-1: Optogenetic split transcription tool is unable to induce activity with light. (A) Model of the split transcription system. (B) Luciferase activity of mammalian split transcriptional system using GalBD-CRY2 and VP16AD-CIB1. HEK293T cells were CRY2(∆NLS)-mCh-tTA transfected with the AD and BD constructs and a GalUAS-luciferase reporter, then incubatedCRY2(∆NLS)- in dark or exposed to light pulses for 18 hrs before assaying for luciferase activity. Fold increase in luciferase activity is shown compared to reporter only controls. Data represents average and error (s.d.) for 3 independentα experiments. Nuclear:
α
α Nuclear:cytosolic ratio
α Nuclear: CRY2(∆NLS)
64 Regulating Transcription with CRY2
The Tucker laboratory has worked extensively to develop novel uses for the
Arabidopsis thaliana cryptochrome 2 (CRY2) photoreceptor including utilizing its light dependent interaction with CIB1 to reconstitute a split transcription factor in yeast, similar to the methods above (Hughes et al., 2012a; Kennedy et al., 2010; Pathak et al., 2014). With low background and a high dynamic range these systems showed great promise, but while transitioning the CRY2-CIB1 split transcription factor tool into mammalian cells our group found the ability to regulate transcription was severely diminished (Figure III-1); this phenomenon was seen in other studies as well (Chan et al., 2015; Konermann et al., 2013;
Liu et al., 2012). Mammalian cell results demonstrated high background activity (Chan et al.,
2015; Konermann et al., 2013).. The DNA binding domain fused to CIB1, which is a transcriptional activator (Liu et al., 2008, 2012), allows for spontaneous activation of the reporter gene. Our system was designed with the Gal4 DNA binding domain fused to CRY2 and the activation domain bound to CIB1 (Figure III-1A), which reduced background but abolished the light dependent activation.
The aim of this study was to determine why the tool that functioned so robustly in yeast was unable to stimulate transcription in mammalian cells. While exploring the possible mechanisms, we discovered that the fusion of CRY2 to multimeric transcription factor DNA binding domains stimulates a nuclear clearing phenotype. This nuclear clearing is associated with the loss of transcriptional function, which led to the optimization of a novel optogenetic tool to suppress transcription with light by fusing an intact multimeric transcription factor to
CRY2. Using the information about the clearing of protein in the nucleus, a second generation of the split transcription factor tool was developed that removed the dimerization
65 domain from the DNA binding domain fused to CRY2. This improved system demonstrated
induction of transcription with light with low background and high dynamic range. The
systems developed in this aim not only improved upon current optogenetic regulation of
transcription, but also the nuclear clearing information could lead to optimized higher
dynamic range of previously developed optogenetic tools in the nucleus such as the light
dependent CRISPR-dCas9 systems (Nihongaki et al., 2015a, 2015b; Polstein and Gersbach,
2015).
Materials and Methods
Constructs
Summarized construct features are provided for constructs described below (Table
III-1). GalBD-CRY2 (B689) and VP16-CIB1 (B692) were PCR amplified from yeast two-
hybrid plasmids (Hughes et al., 2012a; Kennedy et al., 2010) and cloned into pCMV-Gal4
(Addgene 24345) after removing Gal4 by Kpn I/Not I digestion (for GalBD-CRY2) or Kpn
I/Age I digestion (for VP16-CIB1). CRY-GalVP16 (B695) was generated by digesting
CRY2-CreN (Kennedy et al., 2010), containing CRY2(+NLS) downstream of a mCherry-
IRES sequence, with Not I and Xma I to remove CreN, and inserting a 573bp fragment of
Gal4BD-VP16AD (also digested at Not I/Xma I sites) containing residues 413 to 454 of
VP16. To generate CRY(∆NLS)-mCh-tTA (B714), tTA2 (from pBT224_(pCA-tTA2),
Addgene 36430) was amplified by PCR and inserted into CRY(∆NLS)-mCh(Kennedy et al.,
2010) at Bsr GI/Not I sites. The galUAS-luciferase reporter was pGL2-GAL4UAS-Luc
(Addgene 33020); the tetO-luciferase reporter was pTRE3G-luciferase (Clontech).
CRY-GalΔDD-VP16 (B1012) is equivalent to CRY-GalVP16 but missing residues
66-95 of the Gal4 binding domain (comprising the majority of the dimerization domain). To
66 67 generate this, a fragment containing Gal4BD residues 1-65 and an engineered ‘SR’ linker containing an Xba I site, and a second fragment containing Gal4BD residues 96-147 through the end of VP16 were amplified, then joined by overlap extension PCR. The GalBD(Δ66-
95)-VP16 PCR product was ligated into B695 (CRY-GalVP16) at Not I and Xma I sites.
GalVP16 (B1018) and Gal∆DD-VP16 (B1019) were generated from B695 and B1012, respectively, by digesting with Not I and Sal I to remove the CRY2 fragment, followed by blunt end fill-in with T4 DNA Polymerase and ligation. To generate CRY-dsRed-GalΔDD-
VP16 (B1017), flanking Not I sites were added to dsRed by PCR, and this fragment was ligated between CRY2 and Gal(ΔDD) at the Not I site of GalBD(Δ66-95)-VP16.
CRY-Gal(1-65) (B1013) contains a fusion of CRY2 with Gal4(1-65), and was cloned by digesting CRY-Gal(ΔDD)-VP16 with Xba I and Sma I to remove Gal4(96-147) and
VP16, blunt end fill-in with T4 DNA Polymerase, followed by ligation. CIB1-VP16-CIB1
(B1015) was cloned by amplifying CIB1 using PCR to add flanking Kpn I sites, then ligating this fragment into B692 (VP16-CIB1) at a Kpn I site N-terminal to VP16. CIB1-VP16
(B1014) was generated by amplifying CIB1-VP16 from CIB1-VP16-CIB1 using PCR, then ligating into pcDNA3.1+ at Eco RV and Xba I sites. CIB1-VP64 (B1016) was cloned by removing CreC in the plasmid CIB1-CreC(N1)(Taslimi et al., 2016) (Addgene 75367) at Bsp
EI and Bsr GI sites and replacing with a SV40 NLS-VP64 fragment amplified from pcDNA- dCas9-VP64 (Addgene 47107).
To generate CRY-GalVP16l (B694), first mCherry was removed in the construct
CRY(∆NLS)-mCh(Kennedy et al., 2010) (cut with Xma I/Not I) and replaced with the Gal4
DNA binding domain (residues 1-147). Next, a fragment containing residues 1-147 of Gal4
DNA binding domain, a glycine-serine (G-S) linker, then the VP16 activation domain
68 (residues 413-490) was inserted at Bsr GI-Not I sites to generate CRY-GalVP16l. To
generate CRY-tTA (B702), CRY2(∆NLS)-mCh (Kennedy et al., 2010), containing CRY2 with a mutated NLS sequence in the pmCherryN1 vector (Clontech), was digested at Xma I and Not I sites to remove mCherry, and replaced with tTA from pBT224_(pCA-tTA2)
(Addgene 36430) digested with Xma I and Not I. CRY2olig(FL)-tTA (B703) was generated
by inserting full length CRY2olig (E490G CRY2∆NLS) between the Xho I and Xma I sites of CRY-tTA. CRY2-dCas9-VP64 was cloned in the backbone pcDNA3.1 using Gibson
Assembly. The dCas9 fragment was amplified from pcDNA3.1-dCas9-VP64 (Addgene
47107) using the primers 5’-GAAACGCAAA GTTGGGCGCG CCCGCGGAAT
GGATAAGAAATACTC-3’ and 5’- CGTCCAGCGC GTCGGCGCGC CCAACTTTGC
GTTT-3’and cloned into pcDNA3.1-CRY2FL-VP64 (Addgene 60554) at the Pac I site.
pGL2-GalUAS-GFP-HA (B1007) contains GFP (GFPuv variant) followed by the
nuclear export signal (NES) of Id1 (amplified from GFP-Id1NES by PCR) inserted at Hind
III/Eco RI sites into pGL2-GalUAS-Luc (Addgene 33020). A HA epitope was ligated in frame after GFP at Avr II/Cla I sites to yield pGL2-GalUAS-GFP-HA. pGL2-GalUAS-GFP-
LOVdeg (B710) is in the same backbone, and contains GFP-NES followed by a LOV-degron sequence (PCR amplified from pBMN HAYFP-LOV24, Addgene 49570) inserted at the Bsr
GI site at the C-terminus of GFP. The CMV-GFP-LOVdeg construct (B711) was generated
by inserting the GFP-LOVdeg fragment from pGL2-GalUAS-LOVdeg into the pcDNA3.1
backbone after the CMV promoter (Hind III site).
Cell Culture and Light Treatment Conditions
HEK293T cells were maintained in Dulbecco’s modified Eagle medium (DMEM)
o supplemented with 10% FBS at 37 C with 5% CO2. HEK293T cells were transfected using
69 calcium phosphate or Lipofectamine 2000 or 3000 (Invitrogen), according to the
manufacturer’s protocol. Cells were transfected with 1 µg DNA for each construct (for
experiments in 12-well plates) or scaled up accordingly for larger plates. DNA was
transfected at a ratio of 1:1 (2 constructs, 2 µg total DNA per well), 1:1:1 (3 constructs, 3 µg
total per well), or 1:1:1:1 (4 constructs, 4 µg total DNA). Nonspecific ‘stuffer’ DNA (for
example, pmCherry-N1 empty vector) was included in transfections as needed to ensure all
transfection wells contained equal amounts of DNA. Dark samples were wrapped in
aluminum foil immediately after transfection. Light-treated cells were illuminated using a
custom programmable blue LED light source (461 nm, 7.4-18 mW/cm2). Pulse length and
interval used was 2s pulse every 3 min for all experiments (1.1% duty cycle), unless
differently indicated. Light dosage experiments shown in Figure III-6F used a custom built
LED light source designed for 24-well plates and low illumination levels (Gerhardt et al.,
2016) and used constant blue light rather than pulses.
Luciferase Assays
Measurement of luciferase activity (Fluc) was carried out using the Luciferase Assay
System (Promega) according the manufacturer’s protocol. A Modulus microplate reader
(Turner Biosystems) was used to quantify luminescence. Fluc activity was normalized using
a cotransfected RSV-lacZ control in which ß-galactosidase activity was determined using an
ONPG assay (Clontech Laboratories, protocol #PT3024-1) using o-Nitrophenyl-β-D-
Galactopyranoside (Thermo Scientific) as a substrate. Fluc/lac activity is represented as fold
increase over a control containing only the luciferase reporter.
70 Quantitative RT-PCR
Total RNA was prepared using TRI reagent (Molecular Research Center) following
the manufacturer’s protocol. RNA was verified by calculation of OD280/260 ratio using an
Eppendorf spectrophotometer. To remove genomic DNA, total RNA was treated with RQ1
RNAse free DNAse (Promega). High Capacity cDNA Archive Kit (Applied Biosystems) was
used to synthesize cDNA from DNAse- treated total RNA (2 µg of total RNA) using random
hexamers. Quantification of firefly luciferase mRNA and mCherry mRNA (internal control)
was performed on Applied Biosystems 7300 real-time PCR system using Power SYBR
Green Master Mix (Applied Biosystems). Samples were run in triplicate, and the average
cycle threshold (CT) was calculated. The average luciferase CT value for each sample was
normalized to the corresponding average mCherry CT value to obtain a ΔCT value. The fold change in luciferase mRNA expression relative to reporter only samples was calculated using the comparative CT (ΔΔCT) values. Primer sequences used for firefly luciferase were 5’-
GCTATTCTGA TTACACCCGA GG-3’ and 5’-TCCTCTGACA CATAATTCGC C-3’ and
for mCherry were 5’-CACTACGACG CTGAGGTCAA-3’ and 5’-GTAGTCCTCG
TTGTGGGAGG-3’.
CRISPR/dCAS9-VP64 Studies
HEK293T cells were transfected using Lipofectamine 3000 (Invitrogen) with 255 ng
gRNA expression plasmid (63.75 ng of plasmid DNA encoding each of four gRNAs
targeting IL1RN or the off-target control HBG1) and 245 ng of pcDNA3.1-CRY-dCas9-
VP64. Primer sequences and standard curves were previously described (Perez-Pinera et al.,
2013). IL1RN gRNAs were CR1 (TGTACTCTCTGAGGTGCTC, at -29), CR2
(ACGCAGATAAGAACCAGTT, at -180), CR3 (CATCAAGTCAGCCATCAGC, at -113),
71 CR4 (GAGTCACCCTCCTGGAAAC). HBG1 control gRNAs were CR1 (
GCTAGGGATGAAGAATAAA, -26), CR2 (TTGACCAATAGCCTTGACA, -101),
(TGCAAATATCTGTCTGAAA, -163), (AAATTAGCAGTATCCTCTT, -209). As a
negative control cells were transfected with 500 ng of an empty pCMV plasmid. Cells were
either illuminated with blue light using a custom-built 6x4 LED array (1 s pulses every 15 s,
450 nm, 16 mW/cm2) starting four hours after transfection or incubated in the dark for the entire experiment. Three days after transfection total mRNA was purified using Qiagen
RNeasy spin prep columns. cDNA was synthesized using the SuperScript® VILO cDNA
Synthesis Kit (Life Technologies). Relative levels of cDNA were detected using Quanta
PerfeCta® SYBR® Green FastMix® (Quanta Biosciences) and CFX96 Real-Time PCR
Detection System (Bio-Rad). Raw data was normalized to GAPDH levels and cells transfected with an empty plasmid control using the ΔΔCT method.
Immunoblotting
For total cell preps, cells were washed in 1x PBS, collected, and lysed in 2x Laemmli
sample buffer by boiling. Proteins were separated by electrophoresis on an SDS-PAGE gel
and transferred to nitrocellulose membranes, followed by probing with primary antibodies:
GFP (G1544, Sigma-Aldrich, dilution 1:3750), beta-actin (sc-47778, Santa Cruz
Biotechnology, dilution 1:1000), mCherry (PA5-34974, Thermo Scientific, dilution 1:1000),
α-tubulin (926-42213, LI-COR, dilution 1:1000), or Gal4DBD (sc-577, Santa Cruz
Biotechnology, dilution 1:1000). IRDye® secondary antibodies (dilution 1:15,000) and an
Odyssey® FC Imager (LI-COR) were used to detect and visualize the target protein in the
membrane.
72 Immunofluorescence Staining
Cells grown on coverslips were washed with PBS then fixed with 4%
paraformaldehyde in PBS for 15 min. Coverslips were washed with PBS and
permeabilized/blocked in 5% Normal Goat Serum (Jackson ImmunoResearch Laboratories)
and 0.1% Triton® X-100 (Amresco) in PBS for one hour. Cells were incubated at room
temperature for two hours with anti-Gal4BD primary antibody (Santa Cruz Biotechnology,
sc-577) in permeablization buffer. Cells were washed in PBS and incubated with
AlexaFluor® 488 Goat anti-rabbit IgG (Invitrogen) for one hour. Cells were washed in PBS
and mounted on slides with FluoromountG® (SouthernBiotech) .
Imaging
HEK293T cells were cultured in DMEM with 10% FBS, then seeded onto 35-mm
glass bottom dishes or on coverslips in 12-well plates prior to transfection. Calcium
phosphate or Lipofectamine 2000 (Life Technologies) was used for transfection, according to
the manufacturer’s protocol. For imaging of fixed cells, cells were washed in PBS, fixed in
4% paraformaldehyde, and mounted on glass slides using Fluoromount-G (Southern
Biotech). For live cell imaging, cells were grown in glass bottom dishes, then media was
exchanged for HBS with 1 mM CaCl2 and samples were imaged at 33.5°C. Samples were protected from ambient light sources using a red safelight and by focusing in the presence of filtered light (572/28 bandpass filter, Chroma). Cell imaging was performed on either: 1) An
Olympus IX71 microscope equipped with a spinning disc scan head (Yokogawa Corporation of America) with a 60x/NA 1.4 objective. Excitation illumination was delivered from an
AOTF controlled laser launch (Andor Technology) and images collected on a 1024 x 1024 pixel EM-CCD camera (iXon; Andor Technology) using Metamorph (Molecular Devices)
73 software. 2) A Ziess AxioObserver Z1 Inverted Spinning Disc Confocal Microscope with a
63x/NA 1.4 objective. Excitation illumination was delivered by a 3i Ablate!™ Model 3iL13 and image collected using a Yokogawa CSU-X1CU camera. Slidebook 6 was used for all collection of images on this system.
Spatial control studies were performed with HEK293T cells on 10cm plates and were imaged with an UltraThin LED Illuminator (GelCompany, TBL-01) using an iImage GelDoc imaging stage with PureShot imaging software on an iPhone 6S platform.
Image Analysis
After initial image collection, ImageJ 1.45s was used for all image analysis. For nuclear:cytoplasmic quantification, confocal image Z-stacks were first compiled into a maximal projection. Total fluorescence in either the nucleus or cytosol was calculated by manually selecting either the nucleus or cytosolic region of interest and quantifying average fluorescence and area within these compartment. A separate region of the image was used to calculate background, which was subtracted from the average fluorescence. Average fluorescence was multiplied by area and calculated for nuclear vs cytosolic regions, then these were divided to obtain a ratio of nuclear: cytosolic protein.
For zebrafish image quantification, we calculated the total fluorescence for each embryo (Average Fluorescence x Area), after background subtraction. All embryos
(including control uninjected embryos) showed significant autofluorescence, as indicated on graph. To control for variation in size between embryos in different groups, we normalized to a single, representative size, equal to the average. Nine embryo images were analyzed for light/dark samples, and seven for controls. Statistical significance for experiments comparing two populations was determined using a two-tailed unpaired Student's t-test.
74 Zebrafish Studies
A UAS-GFP transgenic zebrafish line (Tg(5xUAS:EGFP)nkuasgfp1a) was
used(Asakawa et al., 2008). This line was generated by the Kawakami laboratory (National
Institute of Genetics, Japan) and was a kind gift from Maximiliano Suster (University of
Bergen). The permits for the experiments were obtained from the Office of the Regional
Government of Southern Finland in agreement with the ethical guidelines of the European
convention.
Embryos were staged in hours post-fertilization (hpf). Developing embryos were
injected at the one cell stage then incubated for 3 hr at 28°C in room light to recover from
injections. After 3 hr (at 3 hpf) embryos were either placed under blue light (10s pulse every
3min, 5mW/cm2) or put in the dark. The light source was a custom LED array (470 nm).
Light treatment was maintained for the duration of the experiment. At indicated times, embryos were fixed in 4% paraformaldehyde in PBS, then washed with PBS and dechorionated. Embryos were subsequently incubated with 50% glycerol for 1h and transferred to 80% glycerol in PBS. Embryos were kept in 80% glycerol in PBS at 4°C, then placed on glass slides for imaging. Imaging was carried out using a Leica TCS SP2 AOBS
Confocal microscope (Leica Microsystems, Mannheim, Germany). Stacks of images were taken at 1.1µm intervals (z-axis) using an Argon laser and compiled to maximum intensity projection images using Leica Confocal Software. For toxicity testing (Supplemental Table
1), embryos were injected with 20pg CRY-GalVP16, treated with the same light/dark conditions as above, and assayed for viability at 27 hpf. Data is expressed as number of live embryos/number of embryos injected. Each group contained n = 20-43 embryos and each experiment was repeated three times.
75 ABGalBD-CRY2 CRY2-mCherry 100 80 Dark Dark 60 40 20 (18 hr) (18 hr) Percent Nuclear (%)
Blue Light 0 Blue Light Dark Light
Figure III-2: Nuclear loss is seen with GalBD-CRY2 but not with other CRY2 fusion proteins. (A) Representative HEK293T cells expressing GalBD-CRY2 kept in dark or exposed to blue light pulses for 18 hrs, immunostained using an anti-Gal4BD antibody. Scale bar, 10 µm (B) HEK293T Cells expressing CRY2-mCherry were kept in dark or exposed to 18 hours blue light pulses Scale bar, 10 µm. Cells show a slight reduction of percent nuclear protein after blue light, as quantified in graph at right. Data represents average and error (s.d., n=15). Blue light pulse regime for both was 2s flash every 3 min, 461 nm.
α
α
α
α
CRY2(∆NLS)- CRY2(∆NLS)-mCh-tTA Nuclear: Nuclear:cytosolic ratio 76
CRY2(∆NLS) Nuclear: Results
Transcriptional activation has worked robustly in yeast models with various
transcription factors (Hughes et al., 2012a; Kennedy et al., 2010; Pathak et al., 2014), but we
were unable to repeat this in mammalian cells using Gal4 DNA binding domain-CRY2
(GalBD-CRY2) and VP16 activation domain-CIB1 (VP16-CIB1) fusion proteins (Figure III-
1). To determine the cause of inefficiency in the mammalian system we tracked localization
of the GalBD-CRY2 fusion protein. In the dark the protein showed robust nuclear
localization, but after illumination with 18 hours of light the nucleus appeared to have
cleared (Figure III-2A). The phenomenon is not universal to CRY2 fusion, as CRY2 fused to
mCherry stay predominantly nuclear after eighteen hours of illumination (Figure III-2B).
CRY2-fused Transcription Factors are Redistributed Correlating with Functional Loss
In order to more directly track the correlation between localization and function of the
CRY2-transcription factor fusion proteins, an intact transcription factor was fused to CRY2.
Synthesized constructs included a CRY2 fused Gal4BD-VP16AD synthetic transcription
factor (Sadowski et al., 1988) (CRY-GalVP16) and a ‘tTA’ (Tet-OFF) transcriptional
activator (Gossen and Bujard, 1992) (CRY∆NLS-mCh-tTA). While both CRY2-intact transcription factor constructs showed robust activity in the dark, blue light exposure was able to significantly decrease the luciferase levels to 3.6% of dark level for CRY-GalVP16
(Figure III-3A) and 6.8% of dark levels for CRY∆NLS-mCh-tTA (Figure III-3B). Steady- state protein levels were not diminished with light for both constructs suggesting that protein loss was not the mechanism by which functional loss occurred (Figure III-3C). This led us to track the redistribution that occurs with the total protein after light exposure and both CRY-
77 Percent Nuclear (%)
A BC CRY2 Gal-VP16 CRY2 mCh tTA Dark Light kDa 150 100 α-Gal4 28-fold reduction 75 100 15-fold reduction 100 80 80 α-tub 50 60 60 40 40
(normalized) Dark Light kDa
20 (normalized) 20
Luciferase activity 150 0 Luciferase activity 0 α-mCh DarkBlue Reporter DarkBlue Reporter 100 75 Light Light α-tub 50
D E 2 CRY-GalVP16 CRY2(∆NLS)- E CRY2(∆NLS)-mCh-tTA CRY-GalVP16 mCh-tTA
5 Dark Blue light 1 4 Nuclear: Dark
3 cytosolic ratio 0 2 0 1 23 4 F Time in light post dark 1 F exposure (hrs) 0 1.2 Nuclear:cytosolic ratio 0.8
(18 hr) CRY- 0.4 Blue Light GalVP16 Nuclear: CRY2(∆NLS) 0 mCh-tTA cytosolic ratio 0 2 4 6 8 Time in dark post light stim (hrs)
Figure III-3: CRY2-fused transcription factors are cleared from the nucleus with light correlating with functional loss. (A-B) Luciferase activity of HEK293 cells expressing CRY-GalVP16 and a GalUAS-luciferase reporter (A) or CRY2∆NLS-mCh-tTA and a 7xtetO-luciferase reporter (B) incubated 18 hrs in dark or with blue light pulses. Data represents average and error (s.d.) for 3 independent experiments. (C) Immunoblot of HEK293T cells expressing CRY-GalVP16 (top) or CRY2∆NLS-mCh-tTA (bottom) and exposed to dark or light pulses for 18 hr. Samples were also blotted with α-tubulin as a loading control. (D) Representative immunostaining (CRY-GalVP16, Gal4BD antibody) or fluorescence (CRY2(∆NLS)-mCh-tTA) images showing localization of CRY2-fused proteins exposed to dark or light for 18 hrs. The ratio of nuclear:cytosolic protein from multiple cells is quantified in graph at right. Data represents average and error (s.d., n=10). Scale bar, 10 µm. (E) Kinetics of nuclear depletion. Cells expressing CRY2 fusion constructs were incubated in dark for 16 hrs, then exposed to blue light pulses for indicated times before fixation. The ratio of nuclear:cytosolic protein from fixed cells was then quantified. Data represents average and error (s.d., n=4). (F) Reversibility of nuclear depletion. HEK293T cells expressing CRY2(∆NLS)-mCh-tTA were treated with light for 18 hrs, then incubated in dark for indicated times before fixing and quantifying nuclear:cytoplasmic ratio as in (E). Data represents average and error (s.d., n=8). Panels A and B are results from Dr. Gopal Pathak.
78 GalVP16 and CRY∆NLS-mCh-tTA were redistributed in the nucleus, exhibiting a nuclear
cleared phenotype similar to GalBD-CRY2 (Figure III-3D).
To probe the kinetics of the nuclear clearing phenotype we ran a timecourse tracking
the fluorescent reporter-tagged CRY∆NLS-mCh-tTA protein or by immunostaining CRY-
GalVP16 protein using an anti-Gal4BD antibody. With both constructs, nuclear clearing occurred over several hours (Figure III-3E), which gives insight into the mechanism of clearing. One consideration is that the protein is being diluted out through cell division; however, HEK293T cells double at a rate of approximately 20 hours, which does not fit the results of the timecourse. Triggered export of nuclear export signal (NES) containing proteins via the CRM1 pathways has been recorded on a much faster timescale in the realm of minutes (Niopek et al., 2016) suggesting it is not a rapid export pathway mechanism. The slower clearing over four hours suggests mechanisms such as degradation or protein sequestration.
This was followed up with explorations of the reversion kinetics to observe the nuclear protein rate of recovery. Nuclear clearing was induced with illumination after which cells were switched to dark and fluorescence was tracked using the CRY∆NLS-mCh-tTA construct. The recovery was heterogeneous and on a much slower timescale than the clearing with a portion of the cells fully recovered by eight hours (Figure III-3F). Cells left overnight showed a less heterogeneous fully recovered phenotype (data not shown). This slow recovery may be from protein diffusing back into the nucleus from the cytosolic pool or new protein synthesis.
79 A C Time post initial blue light (min) 0 15 30 45 60 75 90 1000
!! 100 !!
10 Luciferase activity (Fold OVer Reporter) (Fold OVer B 1 Gal4BD 147 CRY-Gal-VP16 CRY2DNA Bind DD VP16 1 50 94
CRY-Gal -VP16 CRY2 ∆DD VP16 ∆DD GalVP16 1 65 96 147 Reporter Only CRY-GalVP16Gal∆DD-VP16 D E CRY-Gal∆DD-VP16 CRY-Gal CRY-Gal 6000 Blue Light Dark 30 ** -VP16 ∆DD-VP16 5000 20 4000
10 Dark 3000 Ratio D:L activity Ratio D:L
2000 0 -
∆DD Luciferase Activity Luciferase 1000 Gal (Fold Over Reporter) Gal-VP16 VP16
0 Blue Light CRY2-Gal CRY-Gal Reporter VP16 ∆DD-VP16 Only
Figure III-4: A CRY2-fused dimerization domain is required for light-dependent nuclear depletion and functional loss of activity. (A) Live cell imaging showing formation of CRY2(∆NLS)-mCh-tTA clusters in the nucleus upon initial light exposure, which coalesce into larger puncta over the 90 min timecourse. Scale bar, 10 µm. (B) Schematic showing constructs used in (D) and (E). CRY-Gal∆DD-VP16 contains Gal4BD residues 1-147 but is missing residues 66-95, which are important for dimerization.∆DD (C) Functional analysis of dimerization∆DD domain deleted Gal4BD-VP16AD fusion constructs. Indicated constructs were transfected along with a GalUAS-luciferase reporter into HEK293T cells, then incubated in dark for∆DD 24 hrs before quantification of luciferase activity. While a construct (Gal∆DD) containing Gal4 residues 1-147 but with a deletion in the Gal4 DNA binding domain dimerization domain (residues 66-95) was not functional, activity could be restored with fusion to CRY2, which shows some degree of self-association in dark. Activity is expressed as fold over reporter only control. Data represents the average and error (s.d., n=3) for a single experiment, and experiments were∆DD ∆DDrepeated twice with similar results. (D) Luciferase activity of cells expressing indicated constructs exposed to 18 hours dark or blue light pulses. Data represents average and error (s.e.m.) from 4 independent experiments. Inset shows the ratio of activity in dark to activity in light. **, p-value <.05. (E) Localization of protein fusions in HEK293TGal∆DD-VP16 cells, Gal∆DD-VP16assayed by immunohistochemistryGal∆DD-VP16Gal∆DD-VP16 using an anti-Gal4BD antibody. Cells were treated as in (D).
80 Dimerization Domain is Required for Nuclear Depletion and Functional Loss
It has been shown that multivalency of tagged proteins and high concentrations of
CRY2-tagged proteins can lead to clustering (Bugaj et al., 2013; Lee et al., 2014; Mas et al.,
2000; Ozkan-Dagliyan et al., 2013; Taslimi et al., 2014; Yu et al., 2009). We used live cell
imaging to track CRY∆NLS-mCh-tTA in response to light, which showed an immediate clustering phenotype in the nucleus (Figure III-4A). The clusters coalesce and appear to be dynamic over 1.5 hrs of imaging. Both of the transcription factors fused to CRY2 are known to dimerize, and Gal4 has a well-characterized dimerization domain between residues 50-94
(Carey et al., 1989). To determine if this dimerization motif could play a role in the clustering and clearing of proteins in the nucleus we synthesized a construct with a deletion of amino acids 66-95 (Gal∆DD) in the DNA binding domain of GalVP16 (Figure III-4B).
Gal∆DD-VP16 has only a minimal dimerization motif (Marmorstein et al., 1992) and cannot stimulate transcription on its own (Figure III-4C), demonstrating the dimerization domain’s importance in DNA binding. CRY2 fusion to Gal∆DD-VP16 is able to restore a larger
portion of lost activity (Figure III-4C), likely due to the slight self-association of CRY2 that
has been seen in dark (Taslimi et al., 2016) mimicking the dimerization domain.
When functionally compared with the robust dark activity of CRY-GalVP16, which is
diminished with light, CRY-Gal∆DD-VP16 showed decreased dark activity with no
reduction of activity upon light stimulation but rather a small increase (Figure III-4D). We
hypothesize that the mild increase may be due to clustering of CRY-Gal∆DD-VP16 at the
promoter region. This opposing functional result suggests the dimerization domain plays an
important role in the loss of activity seen with the intact transcription factor. Correlating with
81 ∆DD ∆DD
Gal∆DD-VP16
CRY-Gal∆DD-VP16
∆DD
∆DD
∆DD
A CRY-dsRed- CRY2 dsRed GalBD ∆DD VP16 Gal -VP16 ∆DD 1 65 BC D CRY-dsRed- Blue Light Dark Dark 800 Gal∆DD-VP16 Blue Light 1200 10 ***
600
Dark 800 5 400 Ratio D:L activity Ratio D:L 400 0 - -
∆DD ∆DD 200 Luciferase Activity Luciferase Luciferase Activity Luciferase
Gal Gal (Fold Over Reporter) (Fold Over Reporter) VP16 Blue Light 0 dsRed-VP16
CRY- CRY-dsRed Reporter only CRY-dsRedCRY-mCh Gal∆DD-VP16 Reporter only Gal∆DD-VP16 Gal∆DD-VP16Gal∆DD-VP16
Figure III-5: Addition of a tetramerizing dsRed domain restores light dependent functional loss and nuclear clearing to CRY-Gal∆DD-VP16. (A) Schematic of construct used in (B) and (C). (B) Representative HEK293T cells expressing CRY-dsRed-Gal∆DD- VP16, exposed to 18 hr dark or blue light pulses, and assayed for localization by immunohistochemistry using an anti-Gal4BD antibody. (C) Luciferase activity of cells expressing CRY-Gal∆DD-VP16 alone or with an added back multivalent domain (CRY- dsRed-Gal∆DD-VP16) assayed after being exposed to 18 hours dark or blue light pulses. Data represents average and error (s.e.m.) for 3 independent experiments. Inset shows the ratio of activity in dark to activity in light. ***, p-value <.005. (D) Comparison of effect of adding back a tetramerizing dsRed domain versus a monomeric mCherry domain to CRY- Gal∆DD-VP16. Luciferase assay was carried out as in (C). Data represents average and error (s.d., n=3) from one experiment, and experiments were repeated two times with similar results.
82 the opposing functional results, CRY-Gal∆DD-VP16 shows no light dependent clearing in nuclear localization with light treatment (Figure III-4E).
Further examination was done to show that the dimerization motif was the necessary factor for the functional loss and phenotypic changes in the nucleus. An add-back experiment was performed by inserting a tetrameric dsRED domain into CRY-Gal∆DD-VP16 (Figure
III-5A) and testing for functional and phenotypic changes. The add-back of a multimeric domain, CRY2-dsRED-Gal∆DD-VP16, was able to restore nuclear clearing (Figure III-5B) and functional loss (Figure III-5C) upon light illumination. Due to the changes in size of the protein with the deletion of the dimerization motif we tested an add back of a monomeric mCherry domain, which is 29 kDa in comparison to the 28kDa dsRED domain, to prove that size was not the causative factor. CRY2-mCh-Gal∆DD-VP16 displayed no light-dependent loss of activity, similar to CRY2-Gal∆DD-VP16 (Figure III-5D). Together these data show the importance of the dimerization motif in the light dependent nuclear clearing and functional loss.
An Optimized Light-Activated System For Transcriptional Control
With the newly realized importance of the dimerization domain in nuclear clearing, in conjunction with the clearing of GalBD-CRY2 (Figure III-2A), we redesigned the split transcription system (Figure III-1). The Gal4 DNA binding domain (residues 1-147) was replaced with a construct with a deletion of the critical portion of the dimerization motif,
Gal4∆DD (Gal4 residues 1-65) (Figure III-6A). This optimized construct was retained in the nucleus after light illumination (Figure III-6A) as shown by immunostaining with anti-Gal4 antibody. When coexpressed with VP16-CIB1, CRY-Gal(1-65) showed light dependent activation of transcription with low dark activity (Figure III-6B). In an effort to increase the
83 A B C
CRY-Gal(1-65) Blue Light 300 101.8x 160 Blue Light Dark CRY2 Gal4 Dark 1 65 120 55.4x 200 27.6x Dark Blue Light 80 15.1x 100 40 Luciferase Activity Luciferase Luciferase Activity Luciferase (Fold Over Reporter) (Fold Over Reporter)
D GalBD-CRY2 VP16-CIB1 CIB1-VP16CIB1-VP64 Reporter Only CRY2-Gal(1-65) Reporter Only 12 hr 24 hr + VP16-CIB1+ VP16-CIB1 kDa CIB1-VP16-CIB1 α-tub 50 E F Patterned light GFP reporter Illumination intensity (mW/cm2) 37 α- kDa 0 0.17 0.42 0.72 GFP 25 -GFP 25 α 50 α−Tubulin α Dark Light Dark Light G 100
Light 75 Dark
50 Light Light 25 Dark Luciferase activity Dark
(normalized, % of max) Dark only 0 0 6 12 Hours after initial timepoint
Figure III-6: Optimization of split CRY2/CIB1 transcriptional system.
84 Figure III-6: Optimization of split CRY2/CIB1 transcriptional system. (A) Schematic and immunostaining of new CRY2-BD construct (CRY-Gal(1-65)) assayed in split transcriptional system. HEK293T cells expressing CRY2-Gal(1-65) were kept in dark or exposed to blue light pulses for 18 hrs, then assayed for immunohistochemistry using an anti- Gal4BD antibody. (B) Luciferase activity of HEK293T cells expressing indicated constructs and a GalUAS-luciferase reporter and exposed to dark or blue light pulses as in (A) for 18 hours. Data represents average and error (s.d., n=3). Experiments were repeated two times with similar results. (C) Comparison of activation domain fusions. HEK293T cells were transiently transfected with CRY2-Gal(1-65) and indicated AD-fusion constructs, along with a GalUAS-luciferase reporter and tested for activity as in (B). Data represents average and error (s.d., n=3). Experiments were repeated three times with similar results. (D) Immunoblot of HEK293T cells expressing a GalUAS-GFP-HA reporter with CRY2-Gal(1-65) and CIB1- VP64 incubated in dark or blue light for 18 hours. (E) Spatial regulation. HEK293T cells expressing CRY-Gal(1-65), VP64-CIB1, and a GalUAS-GFP reporter were exposed to 18 h patterned blue light, followed by imaging of GFP reporter expression. Scale bar, 1 cm. (F) Dose-dependent regulation. HEK293T cells expressing constructs as in (E) were illuminated with varying light intensities for 18 hrs, followed by immunoblotting using an anti-GFP antibody or anti-tubulin control. (G) Temporal regulation of split transcriptional system. HEK293T cells were transfected with CRY-Gal(1-65), VP64-CIB1, a GalUAS- luciferase reporter, and a RSV-lacZ control. Cells were incubated for 6 hours in dark after transfection, followed by 12 hours in light to induce reporter expression. Samples were analyzed for luciferase and beta-galactosidase activity at this timepoint (t = 0 hr), maintained in light or dark for 6 h (t=6h timepoint), or maintained for 12 additional hours in either of 4 conditions: 6h light/6h light, 6h light/6h dark, 6h dark/6h light, or 6h dark/6h dark. Light samples were exposed to pulsed blue light (2s every 3 min) during treatment. The ‘dark only’ control sample indiated in red was maintained in dark after transfection and not exposed to light for the duration of the experiment. Data represents results from two independent timecourse experiments. For each experiment, luciferase levels measured from each timepoint were normalized to the beta-galactosidase levels at that timepoint, and then expressed as a percent of the 12 h (6h light/ 6h light) timpoint. Data represents average and error (s.e.m.) (0h timepoint., n=8; 6h timepoint., n=12; 12h timepoint., n=6).
85 dynamic range the activation domain construct was also redesigned. The original VP16-CIB1
construct showed 15.1 fold activation over dark levels, a CIB1-VP16-CIB1 construct
containing an additional copy of CIB1 at the N-terminus, showed only a modest 2-fold
increase in dynamic range (27.6 fold activation over dark levels). This double tag of CIB1
had been shown to be beneficial in the recruitment of a CRY-activation domain construct in
previous literature (Polstein and Gersbach, 2015). A N-terminal CIB1 fusion (CIB1-VP16)
achieved 55.5 fold over dark levels but also had higher dark background. Replacing VP16
with a stronger activation domain, VP64 (CIB1-VP64), provided the highest level of
activation, 101.8 fold activation over dark levels, with only a small increase in dark
background over VP16-CIB1 (Figure III-6C). In another method of tracking activation of
reporter genes a GalUAS-GFP reporter was expressed with CRY-Gal(1-65) and CIB1-VP64
resulting in no detectable background in dark and light duration dose-dependent induction of
reporter expression (Figure III-6D).
In order to define the spatiotemporal limitations as well as tunability of the optimized
split transcription tool we carried out additional experiments probing spatial and temporal
specificity along with varied intensities of light. Using a patterned mask we blocked areas of
illumination, allowing light to penetrate in a polka-dot pattern on a monolayer of HEK293T
cells on a 10 cm plate expressing CRY-Gal(1-65) and CIB1-VP64 with a GalUAS-GFP reporter resulting in patterned GFP expression where cells had been illuminated (Figure III-
6E). Tunability of expression was probed using the same constructs expressed in cells exposed to increasing light intensities before they were harvested and immunoblotted, which displayed increasing amounts of reporter expression (Figure III-6F). Temporal regulation was tested in conjunction with reversibility by putting HEK293T cells through a series of
86 light/dark incubations and harvesting cells to track activity. Cells expressed CRY-Gal(1-65) and CIB1-VP64 with a GalUAS-luciferase reporter in light for 12 hours to initially induce reporter activity, then they were exposed to four different treatment regimes with two periods within each treatment: dark/dark, dark/light, light/dark, light/light. Each treatment lasted 12 hours total and timepoints were harvested at 0 hours, 6 hours (post-period 1) and 12 hours
(post-period 2) (Figure III-6G). Cells incubated in light the total 12 hours showed the highest levels of luciferase activity. In comparison cells incubated in 6 hours light then 6 hours dark were less transcriptionally active, and cells with a 6 hours dark then 6 hours light-regime showed even less activation. The least transcriptionally active were cells incubated in dark for 12 hours. These temporal activation patterns suggest that there is a small but detectable lag time to turning off the system and turning on the system; however, it also demonstrates the ability to tune activation of the system with varying durations of light. With the activation kinetics and tunability of this system, it is well suited to regulating genes of interest that are altered over a period of hours as the timescale of our system suggests.
Light-Mediation Transcriptional Reduction
It was recognized that the functional loss seen with CRY2 fused to intact transcription factors could be harnessed as an optogenetic tool to inducibly decrease transcription with light. Optimization of constructs included placing the CRY-GalVP16 system on a stronger promoter and appending a longer version of VP16 (VP16l). The dark activity of CRY-
GalVP16l was about 2-fold better than intact Gal4 at activating a GalUAS-luciferase reporter and after 24 hours of illumination there was a 37-fold reduction in luciferase activity to
2.7+/-0.6% of dark levels (Figure III-7A). This correlated with a 10-fold reduction in
GalUAS-luciferase mRNA with light from dark levels (Figure III-7B). The light dependent
87 A B 10-fold 37-fold reduction 1000 reduction 100 GalBD- 750 CRY2 VP16ADl 75 500 50 Luciferase
250 (% of Dark) GalUAS 25
Luciferase activity
Luciferase mRNA
(fold increase over control) Light: - + - + - + Light Reporter Gal4 CRY- - + only GalVP16l CRY- GalVP16l
CDExperimental Strategy time 0 10 16 hours post harvest transfection (in blue light) 7.5
5 (normalized)
2.5 Dark Blue Dark CHX Luciferase reporter activity
light 0
(additional 8 hr incubation) Dark
Blue light Dark CHX
Initial (time 0) E F 100 GFP reporter Dark GalUAS 75 Light Patterned light GFP reporter Dark 50 Dark Light 25 Light Luciferase activity
(normalized, % of max) 0 0 6 12 Hours after initial timepoint G H Light Treatment Protocol (24h) 24h Blue light 24h 100
8h 75 6h 4h 50 2h Dark Time in
kDa Normalized dark (hrs): 0 2 4 6 8 24 band intensity 25 37 α-GFP 25 0 2 4 6 8 24 α-Tub α 50 Time in dark (hrs)
Figure III-7: Light-mediated reduction of transcription using optimized CRY- GalVP16l.
88 Figure III-7: Light-mediated reduction of transcription using optimized CRY- GalVP16l. (A) Schematic and luciferase activity of optimized CRY-GalVP16l system. HEK293T cells were transiently transfected with a GalUAS-luciferase reporter and control DNA (reporter only), CRY-GalVP16l, or intact Gal4, grown for 24 hours in dark or pulsed blue light, then assayed for luciferase activity. Fold increase in luciferase activity is shown compared to reporter controls. Data represents average and error (s.d., n=3). (B) Quantitative RT-PCR showing luciferase mRNA levels in cells transfected as in (A) and harvested after 22 hours. Shown is the average and range from two independent experiments (experiments were repeated two additional times with similar results). (C) Experimental strategy for (D) (D) Comparison of light inhibition of transcription with use of cycloheximide to halt protein synthesis. HEK293T cells were transiently transfected with a GalUAS-luciferase reporter and CRY-GalVP16 and incubated in dark for 16 hours, then exposed to 100 µg/ml cycloheximide, blue light, or dark for 8 additional hours. Initial (time 0) sample was harvested 16 hours after transfection. Graph represents average and error (s.d.) for three separate experiments. Samples were normalized relative to the activity at initial time 0 harvest. (E) Spatial control. HEK293T cells expressing a GalUAS-EGFP reporter and CRY- GalVP16l show patterned reduction of GFP expression in light. Cells exposed to 18 h patterned blue light, followed by imaging of GFP reporter expression. Scale bar, 1 cm. (F) Temporal regulation of CRY-GalVP16l. HEK293T cells were transfected with CRY- GalVP16l, a GalUAS-luciferase reporter, and a RSV-lacZ control. Cells were illuminated for 18 hours in dark to induce reporter expression. Samples were analyzed for luciferase and beta-galactosidase activity at this timepoint (t = 0 hr), maintained in light or dark for 6h (t=6h timepoint), or maintainedfor 12 additional hours in either of 4 conditions: 6h light/6h light, 6h light/6h dark, 6h dark/6h light, or 6h dark/6h dark. Light samples were exposed to pulsed blue light (2s pulse every 3 min) during treatment. Data represents results from two independent timecourse experiments. For each experiment, luciferase levels measured from each timepoint were normalized to the beta-galactasidase levels at that timepoint, then expressed as a percent of the max activity (12 h (6h dark/ 6h dark) timpoint. Data represents average and error (s.d.). (G) HEK293T cells were transiently transfected with CRY- GalVP16l and GalUAS-GFP, grown under blue light, and transferred to dark for indicated times before harvesting all samples at 24 hours and quantifying GFP reporter levels by immunoblotting using anti-GFP and anti-tubulin a antibodies. (H) Graph below shows quantification of immunoblot data from two independent experiments (average and error, s.d.). Panels A, B, C, D, G, and H are results from Dr. Gopal Pathak.
89 decrease in expression was similar to using cycloheximide to block translation suggesting an efficient block with light (Figure III-7C). Spatial control with the optimized CRY-GalVP16l construct was tested by delivery of patterned light across a 10 cm plate, which displayed robust expression of GalUAS-GFP reporter in dark and undetectable levels where illuminated
(Figure III-7D). Temporal control was tested with two approaches. First, a similar time series as used with the optimized split system was set up with cells expressing the optimized CRY-
GalVP16l with the GalUAS-luciferase reporter (Figure III-7E), and second, cells expressing
CRY-GalVP16l and GalUAS-GFP reporter were exposed to varying durations of dark to tune activation (Figure III-7F). Both experiments resulted in dose dependent activation of transcription with varying levels of light exposure. These data suggest the optimized CRY2-
GalVP16l system is both spatially and temporally tunable for activation.
Optimization of the tetracycline transcactivator (tTA) transcriptional regulation system involved removal of any extraneous domains such as the mCherry fluorescent tag.
Cells expressing the optimized CRY-tTA with a tetO-luciferase reporter (Figure III-8A) showed robust transcription in the dark, and blue light was able to reduce luciferase levels
44-fold to 2.3% of dark levels (Figure III-8B). Light dependent block of transcription was compared to that of using the chemical doxycycline to control the system, and while light was unable to reach the same level of depletion (4-fold over background with CRY-tTA + light, compared with 2-fold over background with tTA + doxycycline) there was still a sizable reduction. This loss of luciferase activity correlated with an 8-fold reduction in luciferase mRNA levels with light (Figure III-8C). It was hypothesized that an enhanced clustering phenotype may increase the efficiency of the depletion system due to the correlation of protein aggregation and loss of function. To test this, CRY2(FL)olig, a mutated
90 AB CRY2-tTA 300
44-fold 200 reduction CRY2 TetR-VP16 (tTA) 100
Luciferase activity Luciferase
7xtetO (fold increase over control) 0 Light: - + --- + Dox: -----+ Reporter tTA CRY- only tTA C 8-fold D 120 reduction CRY2-tTA CRY2oligonly -mChtTA-tTA C 8-fold D 120 reduction 25.5x 100 100 75 80
(% of Dark) 50 Luciferase activity 60
Luciferase mRNA 40
(normalized)
(% of Dark) 25 Luciferase activity 20
Luciferase mRNA
0 Light: - + Dark Blue Reporter CRY-tTA Light Only
Figure III-8: Optimization of CRY-tTA light-mediated reduction system. (A) Schematic of optimized CRY-tTA system. (B) Luciferase activity of HEK293T cells transiently transfected with a 7xtetO-luciferase reporter and either CRY-tTA, tTA, or control DNA (reporter only). Luciferase activity was measured after 24 hr incubation in dark or blue light pulses. Doxycycline samples included 0.5 µM doxycycline. Data represents average and error (s.d.) from three independent experiments. (C) Quantitative RT-PCR showing the levels of luciferase mRNA in HEK293T cells transfected as in (B) and harvested after 22 hr. Data represents average and standard deviation from 3 independent assays. (D) CRY2olig-tTA shows reduced luciferase reporter activity upon blue light illumination. CRY2olig-tTA was transfected along with a tetO-luciferase reporter into HEK293T cells, then incubated in dark for 18 hrs in dark or blue light before quantification of luciferase activity. Activity is expressed as percent of dark levels. Data represents the average and error (s.d.) for three independent experiments. Panels A, B, and C are results from Dr. Gopal Pathak.
91 A 100
75
50
(normalized) 25
IL1RN mRNA IL1RN mRNA
Light: - + - + gRNA: HBG IL1RN + CRY-dCAS9-VP64
Figure III-9: Light regulation of endogenous IL1RN transcription using a CRISPR/ dCAS9-based approach. HEK293T cells were transfected with CRY-dCas9-VP64 and IL1RN or HBG1-targeted gRNAs and exposed to light (1s pulses every 15 s) or dark for 3 days. Data represents average and range for two independent experiments (3 replicates each). Results from the Gersbach Laboratory.
AB Light Dark Control *** 2000 CRY2 GalBD- n.s. VP16ADl 12 hpf 1500
EGFP 1000 5xGal UAS 27 500
per zebrafish embryo hpf Relative fluorescence 0
Dark Light Control Figure III-10: In vivo testing of CRY-GalVP16l in GalUAS-GFP reporter zebrafish embryos. (A) Embryos were injected with 20 pg CRY-GalVP16l DNA at the one-cell stage and then incubated (after a 3 hr recovery period) in the dark or light (10s pulse every 3 min, 470 nm, 5 mW/cm2). GFP images of representative embryos were acquired at 12 hpf and 27 hpf . Controls show images of non-injected embryos at 12 and 27 hpf. (B) Quantification of zerafish embryo fluorescence. Embryos treated as described in (A) were imaged then quantified for total fluorescence (n=7-9 embryos each condition). Control embryos showed significant autofluorescence as indicated. ***, p-value < .005. n.s., not significant. Results from the Rossi Laboratory.
Table III-2. Toxicity testing in UAS-GFP embryos UAS- Injection Injection Injection GFP 1 2 3 Total % 24h dark 19/22 27/31 31/40 77/93 83 24h light 19/20 30/32 40/43 89/95 94 non-inj 19/20 30/31 38/40 87/91 96
92 version of CRY2 (E490G) that shows enhanced clustering, was utilized (Taslimi et al.,
2014). When substituted on CRY2-tTA, CRY2(FL)olig-tTA showed similar functional loss, to 3.9% of dark levels, which is not significantly different when compared with CRY2-tTA
(Figure III-8D).
Utilizing the Light-Dependent Depletion Systems
Relevant uses of optogenetic tools are key to their success, so we tested the ability for light to negatively regulate an endogenous gene promoter by fusing CRY2 to the
VP16 transcriptional activator. In order to regulate an endogenous genomic loci we used dCas9 (a D10A, H840A catalytically inactive Cas9) and CRISPR guide RNA technologies
(Cheng et al., 2013; Gilbert et al., 2013; Maeder et al., 2013; Perez-Pinera et al., 2013).
Fusion of CRY2 to dCas9 and activation domain, VP64, generated CRY2-dCas9-VP64 which was coexpressed with a gRNA targeting human IL1RN promoter in HEK293T cells.
Theoretically, this construct would activate the IL1RN gene in the dark and upon light stimulation the construct is hypothesized to cluster resulting in associated functional loss.
Similar to our Gal and tTA regulated systems the dark levels of IL1RN mRNA levels were robust, but with illumination the mRNA levels were reduced by approximately 4-fold (Figure
III-9). This suggests that endogenous genes may be targeted using this approach, and also this suggests possible improvements to current optogenetic CRISPR regulated systems that may have reduced efficiency due to similar nuclear clearing phenomena that we see with our intact Gal and tTA system.
Functional testing was also done in a vertebrate model of zebrafish, Danio rerio, to probe the dynamic range of light induced reduction of transcription in another experimental model. Embryos were microinjected at the one-cell stage with the CRY-GalVP16l plasmid
93 (20 pg DNA/embryo) then exposed to light or placed in the dark. The GFP expression in
zebrafish embryos incubated in the light was significantly reduced in comparison with embryos that were incubated in the dark, and this difference was visible and quantifiable at both 12 hours and 27 hour post-fertilization (Figure III-10). Due to the uneven division of injected DNA, a mosaic pattern appears throughout the embryo of GFP expression. There was minimal toxicity at the injected dose of 20 pg DNA (Table III-2). It has been observed that the full length VP16 activation domain, as with that in CRY-GalVP16l, may have mild toxicity (Ogura et al., 2009). The light induced reduction of GFP expression in zebrafish and mRNA levels with the CRY2-dCas9-VP64 CRISPR study show the potential for a wide range of applications of this CRY2 transcription technology.
Discussion
In this aim we wanted to understand why a split transcription tool that worked robustly in yeast was unable to translate to mammalian cells, and we determined that CRY2 fusion to dimeric transcription factors led to light-induced nuclear clearing of the fusion protein rendering it unable to perform its function. Further characterization of the dimerization domain of the Gal4 transcription factor showed that removal of the dimerization motif retains the protein in the nucleus, but an add back of another multimeric domain can rescue the nuclear clearing. This dependence of nuclear clearing on the multimeric state of the CRY2 fused proteins led to the development of two novel approaches to regulating transcription including activation, with a split transcription factor, and reduction, with an intact transcription factor.
Reconstitution of the split Gal4-VP16 with the CRY2-CIB1 interaction conferred light dependent transcriptional regulation, but only after the DNA binding domain
94 dimerization motif had been truncated, rendering it non-functional. Deletion of the dimerization domain eliminated the light dependent nuclear clearing of GalBD-CRY2. Due to the proper compartmentalization of the protein, light stimulated transcription was robust and was optimized with coexpression of a CIB1-VP64 activation domain. We demonstrated spatiotemporal control of the system in addition to tunability with duration of light as well as intensity of light, and thus with the low background of the system we believe this tool will be universally beneficial in many areas of genetic research.
Using the nuclear loss of the intact transcription factor, a second approach to regulate transcription with light was developed with a single-component transcription factor fused
CRY2 to confer light dependence. This light dependent reduction of transcription not only works with multiple transcription factors fused to CRY2, including Gal4-VP16 and the tet- transactivator, but it also showed promise in regulating endogenous gene loci by employing the CRISPR/dCas9 technology. In addition, the system shows clear light dependent reduction of reporter genes in a vertebrate model of zebrafish. This universal pattern suggests this tool can be used with a variety of multivalent nuclear proteins. The single component system has the advantage of being small, which as discussed in the introduction, means better efficiency in viral packing and more ease of use in vivo. While many groups have attempted systems to activate transcription with light, this single component system is a novel approach to reducing transcription with light.
The implications of the nuclear clearing phenotypes may extend to other systems, as suggested by our results with CRY-dCas9-VP64, as multiple groups have used CRY2 fusion proteins in the nucleus (Chan et al., 2015; Konermann et al., 2013; Nihongaki et al., 2015a;
Polstein and Gersbach, 2015; Taslimi et al., 2016). Due to the fact that fusion of the DNA
95 binding domain to CIB1, a known transcriptional activator, causes high background, multiple groups have sought to fuse CRY2 to the DNA binding domain and seen low dynamic range of their tool (Chan et al., 2015; Konermann et al., 2013; Liu et al., 2012). This may be explained by our findings and could be due to nuclear clearing of their CRY2 fused protein.
The knowledge of the effects of multimeric nuclear proteins fused to CRY2 is an example of how understanding how the tool works helps to optimize them and further improve other tools that may have similar mechanisms.
It remained unclear whether the clearing phenotype indicates a loss of protein in the nucleus or a simply a redistribution of protein to other nuclear compartments. The live cell imaging clearly shows clusters initiate in the nucleus and fixed cell imaging shows nuclear clearing over several hours; however steady state protein levels are unchanged which suggests an overall redistribution of protein within the cell. The following chapter describes studies investigating the mechanism by which the protein is being cleared, to develop a better understanding of where the protein redistributes to and what signaling events may play a role.
96 CHAPTER IV
CRYPTOCHROME 2 IN THE MAMMALIAN NUCLEUS
Introduction
In Chapter III a single component optogenetic transcription tool was described that demonstrated a light stimulated nuclear clearing phenotype that was associated with a reduction of transcription. It was also shown that this phenomenon was dependent upon a dimerization motif in the transcription factor, but the mechanism of clearing was never determined. It is important to understand how an optogenetic tool works in order to understand implications of the tool in the system it is being used. Thus, the aim of this portion of the thesis was to uncover possible mechanisms by which the clearing in the nucleus occurs.
It was hypothesized that two potential scenarios may be occurring, either individually or in conjunction with one another: the first, the nuclear protein was either being shuttled out of the nucleus or being degraded in the nucleus after clustering, and the second, the light illuminated CRY-transcription factor fusion proteins were clustered in the cytosol and beyond the size limit to traffic into the nucleus acting to enhance the nuclear cleared phenotype. Multiple approaches were taken to uncover possible mechanisms of action including probing for post-translational modifications (PTMs), assaying cellular pathways that were speculated to be involved, and tracking the protein via nuclear/cytosolic fractionation and live cell imaging with a photoconvertible protein.
We started to examine possible mechanisms by trying to characterize where in the nucleus the clustered protein localized. Using a straightforward microscopy approach, we
97 looked at potential co-localization with nuclear bodies (NBs). These small compartments in
the nucleus, known as nuclear bodies, are known for separating nuclear processes.
Promyelocytic Leukemia (PML) nuclear bodies are similar in size (0.3-1 micron) and
number (10-30) to the clustered CRY2-mCh-tTA. They are known to have a diverse array of
proteins that can be found within them at any certain time. PML is one of the few proteins
that can always be detected in PML nuclear bodies. PML nuclear bodies have been
implicated in many cellular disorders and have been thought to sequester, modify or degrade
protein partners (Lallemand-Breitenbach and de Thé, 2010). They are also thought to
regulate genome stability, DNA repair, control transcription and play a role in viral defense.
Other nuclear bodies that have similar characteristic to our clusters are paraspeckles, which are thought to function in mRNA regulation and RNA editing, and nuclear stress bodies, which function in regulation of transcription and splicing under cell stress (Dundr and
Misteli, 2010). There are many other nuclear bodies that the clusters could co-localize with, but the morphological differences in size and number suggest otherwise. PML nuclear body co-localization could suggest a potential mechanism for degradation of the transiently expressed protein in the nucleus.
CRY2 is known to be post-translationally modified in its native plant, Arabidopsis thaliana, with phosphorylation sites that determine regulation and function in the plant
(Shalitin et al., 2002) and thus it could be postulated that PTMs may play a role in degradation. In a study looking at the sensitivity of plant purified CRY2 to blue light it was
discovered that three serines (588, 599, and 605) in the CRY C-terminal Extension (CCE)
region were phosphorylated upon illumination leading to functional changes (Wang et al.,
2015). While plant biology may be relevant to mammalian cells, that would require
98 homologous kinases and pathways for regulation. No studies to date have addressed
regulation of CRY2 in mammalian cells through post-translational modifications.
There are a series of nuclear transport receptors and signals that play a role in protein
translocation in and out of the nucleus. One of the commonly referred to proteins is the chromosome region maintenance 1 (CRM1), also known as exportin 1, which recognizes a leucine rich Nuclear Export Signal (NES) and shuttles the protein containing this sequence out of the nucleus (Nguyen et al., 2012). A known inhibitor of this pathway, leptomycin B, blocks this transport pathway. CRM1 has many exportin family members that are responsible for trafficking cargo out of the nucleus through nuclear pore complexes, but most exportins have specific cargo that they carry (Kohler and Hurt, 2007). It has also been shown that some macromolecules, such as ribonucleoproteins, are excreted from the nucleus through budding of the nuclear envelope (Speese and Ashley et al., 2012). These types of protein translocation out of the nucleus are possible targets of the nuclear clearing phenotype.
Materials and Methods
Constructs
Experiments examining translocation from the nucleus used the CRY2(dNLS)-mCh-
tTA construct used in Chapter III due to its distinctive phenotype. For live cell imaging with
photoswitchable proteins, mCherry was removed from this construct and replaced with an
mEOS3.2 photoswitchable domain. The fractionation experiments were carried out with
CRY2-GalVP16 (described in Chapter III) due to its distinct switch to a nuclear cleared
phenotype and GalDBD epitope for probing by Western Blot. A NLS-containing
CRY2(+NLS)-mCh-tTA was engineered with an HA tag to make CRY2(+NLS)-HA-mCh- tTA, which was used for immunoprecipitation followed by mass spectrometry.
99 Site-Directed Mutagenesis
Samples were amplified using Phusion enzyme with primers containing the desired
mutant sequence according to the manufacturer’s protocol. PNK enzyme was added to
phosphorylate the ends, and a ligation was done with T4 Ligase according to the
manufacturer’s protocol.
Cell Culture and Live Cell Imaging
HeLa, HEK293T, and COS-7 cells were maintained in Dulbecco’s modified Eagle
medium (DMEM) supplemented with 10% FBS and 1% penicillin-streptomycin at 37o C
with 5% CO2. Cells were seeded onto a 35mm glass bottom dish (live cell images) or
coverslips on a 12-well plate (fixed images) and transfected using Lipofectamine 2000
(Invitrogen) according to the manufacturer’s protocol. Dark samples were wrapped in
aluminum foil immediately after transfection, and all manipulations carried out using a red
safelight. Unless otherwise indicated, light-treated cells were illuminated using a custom
programmable blue LED light source (461 nm) with a 1 s pulse per 1 min interval, 5.8
2 mW/cm . Media in glass bottom dishes was changed to HBS with 1 mM CaCl2 directly before imaging. Live cell imaging was performed at 34o C.
Immunofluorescence Staining
Cells grown on coverslips were washed with PBS then fixed with 4%
paraformaldehyde in PBS for 15 min. Coverslips were washed with PBS and
permeabilized/blocked in 5% Normal Goat Serum (Jackson ImmunoResearch Laboratories)
and 0.1% Triton® X-100 (Amresco) in PBS for one hour. Cells were incubated at room
temperature for two hours with anti-PML primary antibody (Santa Cruz Biotechnology, H-
238) in permeabilization buffer. Cells were washed in PBS and incubated with AlexaFluor®
100 488 Goat anti-rabbit IgG (Invitrogen) for one hour. Cells were washed in PBS and mounted on slides with FluoromountG® (SouthernBiotech) .
Microscopy and Image analysis
Imaging was performed using two systems: 1) An Olympus IX71 microscope equipped with a spinning disc scan head (Yokogawa Corporation) with a 60x/NA 1.4 objective. Excitation illumination was delivered from an AOTF controlled laser launch
(Andor Technology) and images collected on a 1024 x 1024 pixel EM-CCD camera (iXon;
Andor Technology). The emission bandpass filters were 525/30 (GFP), and 685/36
(mCherry). Metamorph software was used for collection of images on this system. For light- sensitive experiments, cells were protected from ambient light sources by focusing in the presence of filtered light (572/28 bandpass filter, Chroma). 2) A Zeiss AxioObserver Z1
Inverted Spinning Disc microscope with a 63x/NA 1.4 objective and HQ 480/40x and
HQ565/30x (Chroma) bandpass filters. Excitation illumination was delivered by a 3i
Ablate!™ Model 3iL13 and image collected using a Yokogawa CSU-X1CU camera.
Slidebook 6 was used for collection of images on this system. ImageJ 1.45s and Fiji were used for image analysis.
Fractionation
After 18 hours of blue light or dark treatment cells were washed with cold PBS and then harvested into 1mL of cold PBS and placed in a 1.5mL microcentrifuge tube. An aliquot was removed to be used as the “Total” sample. The remainder was spun down at 10,000 rpm for 10 s to pellet the cells. The supernatant was removed and the cell pellet was resuspended in 9x volume of the cell pellet with homogenization medium (0.25M Sucrose, 5mM MgCl2,
10mM Tris-HCl pH7.4). This cell suspension was homogenized using a Potter-Elvehjem
101 homogenizer on ice with 6 strokes and then passed through cheesecloth into a new 1.5mL
microcentrifuge tube. The tube was then centrifuged at 600g for 10 min. at 4°C. The top
supernatant was collected as the “Cytosolic Fraction”. After the remaining supernatant was
removed the pellet was resuspended in half the volume of the homogenization buffer from
the preceeding step and centrifuged again at 600g for 10 min. at 4°C, after which the
supernatant was discarded. The pellet was resuspended in 9 volumes of hypertonic sucrose
buffer (2.2M Sucrose, 1mM MgCl2, 10mM Tris-HCl pH7.4) and centrifuged at 80,000xg for
80 min. at 4°C. The sucrose was removed and the remaining pellet considered the “Nuclear
Fraction”. The fractions were mixed with an equal volume of 2x Laemmli Buffer and boiled
for 10 minutes.
Immunoprecipitation
After 3 hour light/dark treatment of a monolayer of HEK293Ts the cells were washed
twice with PBS and then membranes were lysed in RIPA Buffer (25mM Tris•HCl, pH 7.6,
150mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS))
with inhibitors: 1x Halt Protease, 1mM PMSF in EtOH, 0.2mM Sodium Orthovanadate,
10mM Sodium Fluoride, and 5mM beta-glycerophosphate. Lysed cells were collected in a pre-chilled 1.5mL microcentrifuge tube and incubated on ice for 10 min. DNA was sheered with 5-7 passes through a 25-gauge needle. Samples were clarified with a 10 min full speed
spin at 4°C. Supernatant was tranferred to a new pre-chilled 1.5mL microcentrifuge tube.
100-500μg whole cell extract was combined with 1-5 µg antibody in final volume of
500µL. This mixture was incubated at 4°C on a rotator for 1-4 hours. 24µl cold, cleaned
Protein G Magnetic Beads were added to Antibody/Extract Mixture and incubated for 1hr on
rotator at 4°C. Beads were pelleted using a magnetic rack, supernatant was removed and
102 discarded, and 500µl Wash Buffer (10mM Tris pH 7.4, 1mM EDTA, 1mM EGTA; pH 8.0,
150mM NaCl, 1% Triton X-100) was added. This wash step was repeated 4-6 times. All
Wash Buffer was removed and ~20µl of 2x Reducing Loading Buffer (2.5mL 1M Tris pH
6.8, 5mL of 100% Glycerol, 2.5mL beta-mercaptoethanol, 1.2g SDS, 500µL H2O, bromophenol blue to final conc. of 0.02% (w/v)) was added. Sample was vortexed and kept on ice to bring to Mass Spectrometry Lab or frozen and used for Western Blot.
Immunoblotting
For total cell preps, cells were washed in 1x PBS, collected, and lysed in 2x Laemmli
sample buffer with boiling. Proteins were separated by electrophoresis on an SDS-PAGE gel
and transferred to nitrocellulose membranes, followed by probing with primary antibodies:
Histone H3 (H0164, Sigma-Aldrich, dilution 1:5000), α-tubulin (926-42213, LI-COR, dilution 1:1000), or Gal4DBD (sc-577, Santa Cruz Biotechnology, dilution 1:1000).
IRDye® secondary antibodies (dilution 1:15,000) and an Odyssey® FC Imager (LI-COR) were used to detect and visualize the target protein in the membrane.
Mass Spectrometry
The UC Denver Mass Spectrometry and Proteomic Core prepped and ran the
immunoprecipitated samples with the following protocol. Bead-eluted CRY2-HA-tTA
samples were loaded onto a 1.5-mm-thick NuPAGE Bis-Tris 4−12% gradient gel
(Invitrogen). The BenchMark Protein Ladder (Invitrogen) was used as a protein molecular
mass marker. The electrophoretic run was performed by using Mes SDS running buffer, in an
X-Cell II mini gel system (Invitrogen) at 200 V, 120 mA, and 25 W per gel for 30 min. The
gel was stained by using Simply Blue Safe Stain (Invitrogen) stain and destained with water
according to the manufacturer’s protocol. After excision, gel pieces were destained in 200µL
103 of 25 mM ammonium bicarbonate in 50% vol/vol acetonitrile (ACN) for 15 min and washed
twice with 200µL of 50% (vol/vol) ACN. Disulfide bonds in proteins were reduced by
incubation in 10 mM DTT at 60 °C for 30 min, and cysteine residues were alkylated with 20
mM iodoacetamide in the dark at room temperature for 45 min. Gel pieces were subsequently
washed with 100µL of distilled water followed by the addition of 100µL of ACN and dried
by SpeedVac (SavantThermoFisher). 100 ng of trypsin was then added to each sample and
allowed to rehydrate the gel plugs at 4 °C for 45 min and incubated at 37 °C overnight. The
tryptic mixtures were acidified with formic acid (FA) to a final concentration of 1%. Peptides
were extracted two times from the gel plugs by using 1% FA in 50% ACN. The collected
extractions were pooled with the initial digestion supernatant, and the volume was reduced
by using SpeedVac. Samples were measured on an LTQ Orbitrap XL mass spectrometer
(Thermo Fisher Scientific) coupled to an Eksigent nanoLC-2D system through a nano-
electrospray LC−MS interface. Eight microliters of sample was injected into a 10-µL loop by using the autosampler. To desalt the sample, material was flushed out of the loop and loaded onto a trapping column (ZORBAX 300SB-C18, dimensions 5×0.3mm×5µm) and washed with 5% (vol/vol) ACN and 0.1% FA ata flow rate of 10µL/min for 5 min. At this time, the trapping column was put online with the nano-pump at a flow rate of 350nL/min. The mobile phase included water with 0.1% FA (solvent A) and 99.9% ACN with 0.1% FA (solvent B).
A 90-min gradientfrom 6% ACN to 40% (vol/vol) ACN was used to separate the peptides.
Peptides were separated on a house-made 100µm inner diameter ×150 mm fused silica capillary packed with JupiterC18resin (Phenomex). Data acquisition was performed by usingthe instrument supplied Xcalibur (version 2.0.6) software. The mass spectrometer was operated in the positive ion mode; the peptide ion masses were measured in the Orbitrap
104 mass analyzer, whereas the peptide fragmentation was performed by using either higher
energy collisional dissociation (HCD) or electron transfer dissociation (ETD) in the linear ion
trap analyzer by using default settings. Ten most intense ions were selected for fragmentation
in each scan cycle; fragmented masses were excluded from further sequencing for 90 s.
MS/MS spectra were extracted from raw data files and converted into Mascot generic format
(MGF) files by using a PAVA script (University of California, San Francisco). These mgf
files were then independently searched against the Arabidopsis thaliana subset of the
SwissProt database by using an in-house Mascot server (Version 2.2.06, Matrix Science).
Mass tolerances were ± 15 ppm for MS peaks, and ± 0.6 Da for MS/MS fragment ions. For
all HCD spectra, fragment ion tolerances were set to 0.05 Da. Trypsin specificity was used,
allowing for one missed cleavage. Met oxidation, protein N-terminal acetylation, and peptide
N-terminal pyroglutamic acid formation were allowed for variable modifications while
carbamidomethyl of Cys was set as a fixed modification.
Results
Inconsistent PML Nuclear Body Colocalization
Due to the enhanced clustering of CRY2-mCh-tTA in the nucleus with light, we
examined whether these puncta could be localizing to nuclear bodies. The characteristic size and number of puncta formed when clusters are stimulated suggested PML nuclear bodies
(NBs), which have similar morphology. Cells that showed robust clustering of CRY2-mCh- tTA were stained with anti-PML antibody and visualized. Results showed that colocalization of protein aggregates with PML was inconsistent, with puncta that displayed colocalization and others that were localized to distinctly separate regions (Figure IV-1A and B). Since
PML NBs are known to associate with the nuclear matrix, it is possible that colocalization
105 observed is due to similar localization on the matrix proteins. Because the cells needed to be
fixed and stained in order to detect PML NBs, the cells were treated with light for 30 minutes
to stimulate clustering prior to staining. In previous experiments exploring clustering, we
have had difficulty visualizing the clusters with a fixed-cell imaging approach, and again
after fixing the cells there were few cells maintained on the coverslip with robust clusters to observe potential colocalization. A live-cell approach with a fluorescently tagged nuclear body protein may result in more conclusive results. This follow up experiment would allow for visualization of enhanced nuclear clustering and the potential to see if the colocalized puncta are stable or dynamic in their observed proximity.
Tracking Light Stimulated Nuclear Clearing
Being able to determine the relative quantities of protein in the cytosol versus nucleus
would allow us to determine if CRY2-mCh-tTA or CRY2-GalVP16 was being degraded in
the nucleus upon light stimulation or if was being exported into the cytosol. We attempted to
quantify this phenotypic switch with a nuclear/cytosolic fractionation using multiple
established protocols. Typical protocols use a series of buffers, spins, and occasional
homogenization steps to lyse the outer membrane of the cell, collect the nuclei, then lyse the
nuclei resulting in lysates from the cytosol and nuclear fractions. While multiple protocols
using this approach successfully fractionated our control proteins (Histone H3, nuclear; a-
Tubulin, cytosol), highly variable and unexpected results were seen with our protein of
interest, likely due an insolubility of the protein aggregates seen upon illumination. This led
us to track our protein throughout the fractionation steps using the mCh tag in CRY2-mCh-
tTA, which gave us a potential explanation for our unforeseen results. When exposed to blue
106
A CRY2-mCh-tTA Anti-PML Composite
B CRY2-mCh-tTA Anti-PML Composite
Figure IV-1: Staining of PML with light induced clusters showed inconsistent co- localization. HEK293T cells expressing CRY2-mCh-tTA were exposed to blue light pulses for 30 minutes before they were fixed and stained with PML. (A) A representative cell shows inconclusive results for colocalization. A blow up of the composite image shows the nucleus. The blue arrow indicated a point where colocalization may by occurring and the white arrows show distinctly different localization for PML and CRY2-mCh-tTA. Scale Bar, 10 µM. (B) Magnification of two representative nuclei shows further lack of colocalization between PML and CRY2-mCh-tTA aggregates. Scale Bar, 10 µM.
107 light the CRY2-transcription factor proteins cluster, and while this is enhanced in the nucleus it also occurs in the cytosol. While tracking the CRY2-mCh-tTA during a fractionation, we observed that large protein clusters that formed after illumination were spun down with our intact nuclei during our nuclear isolation step. These large fluorescent clusters were not part of the nuclei and did not appear to be adhered to them. Multiple washes were unable to remove these aggregates from our intact nuclear pellet. We used a new approach combining a traditional hypotonic salt solution fractionation protocol with a sucrose gradient in an attempt to remove these aggregates. While fewer aggregates were found in the nuclear pellet, there was no light-dependent loss of CRY2-GalVP16 in the nucleus (Figure IV-2). The sucrose gradient method was able to improve upon the methods previously tested, but there were still aggregates in the nuclear pellet. Although we believe the aggregates may be complicating our results, it cannot be discounted that the protein may not actually leave the nucleus. Rather it may get redistributed to the periphery where it is difficult to detect with microscopy.
In another approach to track nuclear clearing, a photoconvertible fluorescent protein domain was inserted into CRY2-mCh-tTA. mEOS3.2 is a green fluorescent protein that can be photoconverted to a red fluorescent protein using a 415nm wavelength pulse (Zhang et al.,
2012). We replaced mCherry with mEOS3.2 (CRY2-mEOS-tTA) to utilize the photoconverting properties to tease apart if protein is shuttled out of the nucleus, being degraded, or getting sequestered at the nuclear periphery. The un-photoconverted green fluorescent population shows similar distribution as CRY2-mCh-tTA but is unchanged throughout the entire one hour experiment (Figure IV-3A). Live-cell imaging was performed
108 Total Cyto Nuclear
Dk BL Dk BL Dk BL kDa -100 Anti-Gal4 -75
Anti-!Tub -50
Anti-HistoneH3 -15
30000.030
25000.025
20000.0 20
15000.0 15
(x10^3) Intensity 10 10000.0
Normalized Band 5000.0 5
0.0 0 Dk BL Dk BL Cyto Nuclear
Figure IV-2: No Light Dependent Protein Reduction Observed in the Nucleus. Nuclear/cytosolic fractionation was done on HEK293T cells expressing CRY2-GalVP16 exposed to 18 hours blue light pulses or kept in the dark. An immunoblot was performed with the lysates with anti-Gal4 to probe CRY2-GalVP16 and controls: anti-histone H3 and anti-aTubulin. The fractionation controls, anti-histone H3 and anti-aTubulin, cleanly separated. Quantification of the individual immunoblot is depicted on the graph below.
109 with blue light pulses to stimulate clearing, and while a nuclear pool was photoconverted, the fluorescence decreases over the hour-long experiment (Figure IV-3B). Imaging in the green channel suggests this loss of fluorescence is not due to the cell shifting out of focus. More frequent imaging hastens the loss of fluorescence suggesting a photobleaching effect during imaging rather than degradation.
The temporal control of photoconversion is extremely precise, but the spatial control needs optimization for subcellular photoconversion. Photoconversion of the nucleus targets a small amount of cytosolic protein above and below the targeted area of the nucleus, but cells with high concentrations of expressing protein have a global photoconversion. The complication remains: a large number of molecules must be photoconverted to have suitable resolution, but cells primed for efficient photoconversion are those with the highest concentrations, which globally photoconvert upon stimulation. Throughout troubleshooting experiments the best spatial control was achieved through a low intensity, high duration laser pulse of 415nm wavelength. While more optimization may lead to better results, the loss of fluorescence in the nucleus could mean it is being degraded but it could also suggest photobleaching of the red channel.
Post-Translational Modification Effects on Nuclear Clearing
CRY2 is endogenously found in the nucleus where it is known to undergo phosphorylation, ubiquitination, photobody formation, and degradation in plant cells (Zuo et al., 2012). While it is not yet understood how all of those interplay, it is thought that phosphorylation regulates periodic protein levels in plants. Phosphorylation of mouse CRY2 at S557 has been shown to play a role in mouse circadian rhythm (Hirano et al., 2014) suggesting that there are mammalian kinases that act on CRY2. However, plant and
110 A 0min 10min 50min
B 0min 10min 20min
30min 40min 50min
Figure IV-3: Photoconvertible mEOS3.2 tracks nuclear pool during light stimulation. (A) Un-converted protein is unchanged throughout the 50 minute experiment. The star depicts where nucleus was photoconverted with a 415nm laser pulse. Panel 1 is t=0, panel 2 is t=10min (time of photoconversion), and panel 3 is t=60min. (B) Panels are t= 0, 10, 20, 30 , 40, and 50 min. time lapse images. Photoconversion occurs at 10 min. Photoconverted pool decreases in fluorescence over 50 minutes post-photoconversion. It is unclear whether this is due to protein degradation or fluorescent signal loss. Scale Bar, 10 µM.
111 mammalian CRYs have little homology, and thus implications of overexpressing plant CRY2
in mammalian cells remains unknown.
In order to investigate modifications that may play a role in nuclear clearing, we
synthesized a nuclear specific construct containing an HA-tag for immunoprecipitation (IP).
We used a full length CRY2 with an intact NLS for our photoreceptor (CRY2+NLS-HA- mCh-tTA). With lysates from HEK293T cells expressing our nuclear protein exposed to 18- hours light stimulus or kept in dark, we performed an IP-immunoblot and probed for ubiquitination. Our input showed ubiquitination but we were unable to detect ubiquitin modifications in our IP lanes (data not shown). Since ubiquitin is an unreliable epitope to detect on Western Blots, this result was not assumed to be a definitive negative.
In a more sensitive approach, we used mass spectrometry to detect post-translational modifications. HEK293T cells expressing CRY2+NLS-HA-mCh-tTA were exposed to blue light or kept in dark, then an IP was performed followed by analysis via mass spectrometry through the UC Denver Mass Spectrometry/Proteomics Core. Analysis of the data was done to detect differentially regulated modifications with light. PTMs detected in the screen include phosphorylation marks at S245 (2% of total fragments detected had modification),
Y618 (15% of total fragments), T1037 (8% of total fragments) and T1039 (4% of total fragments), and ubiquitination marks at K219 (7% of total fragments), K375 (10% of total fragments), and K483 (100% of total fragments, only one fragment detected) (Figure IV-4A;
Table IV-1). The Y618 mark falls on the HA tag and therefore was discarded as a potential residue of interest. Acetylation was also screened but no marks were detected. This assay was only done once with one sample per treatment. These PTM screen hits are potentially regulated marks and need further validation to prove their presence and significance.
112 A Phorphorylation Marks 245 618 1037 1039 S Y T T N C Ubiquitination Marks 219 375 483 K K K N C
B Serine Glutamine
Tyrosine Phenylalanine
Threonine Asparagine
Lysine Arginine
Figure IV-4: Phosphorylation and ubiquitin marks detected with mass spectrometry. (A) Mass spectrometry resulted in hits for phosphorylation (peach) at S245, Y618, T1037 and T1039, and ubiquitination (green) at K219, K375, and K483. (B) Mutational strategy for blocking these residues from getting modified.
113 Following up on this work, mutations were made to the residues to test the role of
detected PTMs in nuclear clearing. We designed serine to glutamine, tyrosine to
phenylalanine, and threonine to asparagine mutations for phospho-mutants and lysine to
arginine for ubiquitin-mutants (Figure IV-4B). These mutations were built on the DNLS
(CRY2-mCh-tTA) construct so that the effect on light stimulated clearing could be visualized. We expressed these mutants in HEK293T cells and exposed them to light or dark conditions for 18 hours and then fixed the cells. Imaging was performed and each mutant displayed nuclear clearing phenotypes when exposed to light (Figure IV-5). This does not rule out the possibility that multiple residues may need to be mutated in tandem to block the clearing phenotype, as they may function together.
Pharmcological Inhibition Shows Translation is Important for Nuclear Clearing
There are many mechanisms involved in the regulation of protein levels in the cell
including rate of translation, proteosomal degradation, and autophagy. To determine which
mechanisms may play a role in the regulation of protein in the nucleus we applied an array
of chemical inhibitors of pathways including cycloheximide to block translation, MG132 to
block proteosomal degradation, and chloroquine to block autophagy. In addition we used
doxycycline to block the transactivator from binding the promoter to determine if the binding
of the protein to DNA played a role. DMSO was used to dissolve the inhibitors and therefore
was included as a control. Chemical inhibitors were applied to HEK293T cells expressing
CRY2-mCh-tTA for 30 minutes before exposure to 8 hours of light stimulation or dark
incubation. The cells were then fixed and imaged. Cycloheximide was the only inhibitor to
change the phenotype of the nuclear clearing, resulting in the protein to be retained in the
nucleus (Figure IV-6A). While cycloheximide is commonly used in research as a protein
114 Table IV-1: Percent of Fragments Detected that Showed Post-Translational Modifications
Phophophorylation PTMs Ubiquitination PTMs
Percent of Total Percent of Total Amino Acid Fragments Amino Acid Fragments Detected Detected
S245 2% K219 7%
Y618 15% K375 10%
T1037 8% K483 100%
T1039 4%
A Phospho-mutants S245Q T1037N T1039N
B Ubiquitin-mutants K219R K375R K483R
Figure IV-5: Loss of individual phosphorylation or ubiquitination PTMs are not sufficient to block nuclear clearing. Cells expressing CRY2-mCh-tTA with described mutation show nuclear clearing after 18 hours (A) Respresentative HEK293 cells for (A) phospho-mutants and (B) ubiquitin-mutants. Scale Bar, 10 µM.
115 A Dark 8 Hrs Light CHX
B 8 Hrs Light DOX Chloroquine DMSO MG132
Figure IV-6: Cycloheximide blocks light triggered nuclear clearing. (A) HEK293T cells treated with protein synthesis blocker, cycloheximide (CHX), and exposed to light or dark for 8 hours shows a block of nuclear clearing. (B) This block was not seen with other inhibitors such as doxycycline (DOX), chloroquine, MG132 or DMSO control after an identical 8-hour light exposure. Scale Bar, 10 µM.
116 synthesis inhibitor, there are off-target effects including DNA damage and activation of cell
death and stress pathways that may also be playing a role in blocking the clearing phenotype. Therefore, other methods of protein inhibition, such as high doses of antibiotics such as rifampicin, could be used to confirm its necessity. Cells treated with doxycycline,
MG132, chloroquine, and DMSO showed relatively uniform levels of nuclear clearing
(Figure IV-6B) suggesting a lack of a role for DNA binding, proteosomal degradation, and autophagy in the mechanism of nuclear clearing. As with cycloheximide discussed above, each pharmacological inhibitor used has considerations that must be taken into account when using them. MG132 blocks degradation of ubiquitinated proteins that are tagged for proteosomal degradation, but it can also stimulate cell death pathways and has off target inhibition of calpains and cathepsins. Chloroquine blocks autophagy as it binds the lysosome which opens the possibility that nuclear export can still occur. An early stage autophagy inhibitor such as bafilomycin A would be more appropriate as it blocks the autophagosome formation, but blocking autophagy can lead to alternate cell death pathway activation when the cells are stressed. Due to off target effects, these inhibitors must be precisely controlled for in order to extract the relevance of each mechanism they control.
Others inhibitors were tested to probe the roles of CRM1 export pathways and transcription in the clearing phenotype. A singular test of leptomycin B (CRM1 inhibitor) and actinomycin D (transcription inhibitor) showed no difference in nuclear clearing as a control after 8 hours of light exposure. In future experiments, tracking a protein known to export via the CRM1 exportin pathway while testing the leptomycin B is an important control to ensure the inhibitor is functioning as expected. Since many NLS containing proteins are exported via this pathway, a result seen with leptomycin B would not rule out
117 that one of these proteins may shuttle the protein out of the nucleus or play a role in the transport as well. Actinomycin D is a globally acting inhibitor of transcription, therefore, a result with this inhibitor would suggest that a transcriptional product is involved but does not rule out products with a short RNA and protein half-life that may play a role. In future experiments running a gel and measuring mRNA output in the cells would demonstrate the efficiency of the inhibitor. While repetition of this experiment would be necessary to conclude their lack of roles, these data suggest that these pathways are not essential for the nuclear clearing phenotype.
Discussion
While we were unable to determine a specific mechanism for the clearing of nuclear
CRY2-fusion proteins, we discovered that there may be differentially regulated PTMs that could be explored for changes in nuclear clearing and also that active translation is necessary for the clearing to occur. Understanding the mechanism is necessary for explaining how the single component transcription system functions, and that will give insights into how this tool can be used in the future. Because as described in the previous chapter CRY2/CIB1 split yeast transcription systems work very efficiently, this phenomenon may be specific to mammalian cells. Understanding more about the biology behind the mechanism could propose ways to make it a more universal system.
While the results of this aim raised more questions than they answered, there were some very intriguing outcomes of the experiments. Most notably, there may be a translational product that affects clearing, but active transcription is not necessary. This suggests there is already mRNA ready to be translated upon stimulation that is involved in the mechanism of clearing. Following this study with a proteomics approach to determine what proteins are
118 activately translated after blue light illumination may help identify which proteins are necessary for the clearing that are upregulated. This conclusion hinges on the cycloheximide result, and it is known that cycloheximide can have off-target effect such as DNA damage and upregulation of cellular stress pathways. One of the cycloheximide side effects may be playing a role in the block of nuclear clearing, and so another inhibitor of protein synthesis should be tested to confirm new protein translation is required for CRY2-nuclear protein clearing.
While we were unable to track the protein with the photoconvertible approach at this time using a mEOS3.2 domain, but there are other photoconvertible protein domains that may have better expression and spatial resolution. Dendra2 also switches from green to red and has been codon optimized for expression in mammalian cells. The spatial precision of photoconversion may be enhanced with another photoconvertible protein as well. This would give us an accurate way to track translocation across the nuclear membrane.
As discussed, the proteomic analysis of post-translational modifications gave us starting points for further functional analysis, however the reliability of a singular run with one sample per treatment group is not conclusive and more replicates should be done to enhance the statistical probablility that these modifications are authentic. There are ways to enrich for phorphorylation sites prior to running the mass spectrometry that gives a better probability of getting positive identifications.
There were many questions raised by these studies and the mechanism of nuclear clearing is still unknown. However, we feel it is important to not only understand how optogenetic systems are functioning in the cells but also to appreciate the other effects of overexpressing a plant protein in a mammalian cell. It is possible that pathways involving
119 cellular stress are activated when aggregates of the proteins form in the nucleus, and this could have negative effects if these tools are being used to determine cellular mechanisms that these off target pathways may effect.
120 CHAPTER V
PERSPECIVE, CONCLUSIONS, AND FUTURE DIRECTIONS
Perspective
The last ten years exhibited exponential growth in the field of optogenetics, and there continue to be improvements made through novel applications of known photoreceptors, the characterization of new photoreceptors, and the enhancement of hardware used to regulate these systems. Optogenetics earned accolades as a discovery tool across many diverse fields such as the Method of the Year in 2010 (Nat. Methods Editorial, 2011) and it was highlighted as a breakthrough of the decade in the same year (Science Staff, 2010). Developing new soluble optogenetic tools and exploring their mechanisms is important to further the field, but also investigating the effects of expressing these engineered plant-based proteins in cell culture and in vivo models is important so scientists can better predict how an optogenetic tool may behave in their preferred system. The studies described in this Thesis contribute novel optogenetic approaches to regulate perosixomal trafficking and control transcriptional reduction and activation. Further they begin to probe the effects of illuminating cryptochrome 2 (CRY2) fusion proteins in the nucleus, as a start to understanding the mechanism of nuclear CRY2 tools.
Summary of Major Findings
The tools developed in this thesis demonstrate novel approaches to regulate protein function through modulation of protein localization. The first aim focused on inducibly sequestering proteins into peroxisomes in order to isolate them from the cytosol. The second aim concentrated on another subcellular compartment, the nucleus, in order to regulate
121 transcription by controlling transcription factor localization. Lastly, the biology behind how
the plant based photoreceptors respond to illumination in mammalian cells was explored.
Chapter II- Optical Control of Peroxisomal Trafficking
The goal of the first aim was to develop a light dependent peroxisomal trafficking
system to regulate cytosolic/peroxisomal protein activity and localization. By caging a
peroxisomal targeting sequence (PTS1) into the Jα-helix of the AsLOV2 domain we were able to confer light control to peroxisomal trafficking. Using a yeast two-hybrid assay we demonstrated that the light initiated peroxisomal import was due to a light dependent interaction of the PTS1 with PEX5 peroxisomal import receptor. As a demonstration GFP was tagged with the LOV-PTS1 tag to show the efficiency of peroxisomal trafficking.
Optimizations were made to the tag by mutating the AsLOV2 to significantly reduce dark activity with a previously characterized I532A mutation; however, this also decreased trafficking efficiency by about 50%. Adding a nuclear export signal eliminated the nuclear pool, which localized the tagged protein to the cytosol where it could interact with PEX5 and be shuttled to the peroxisome. The last optimization made to the system gave more power over the amount of protein expressed by putting the GFP-LOV-PTS1 under control of a tetracycline-regulated inducible promoter. Being able to turn off transcription allowed for more precise quantification of how much protein is trafficked into peroxisomes, which was determined to be about 77% of cytosolic protein over 24 hours. This was an improvement over the original light inducible construct, which traffics 30-35% of GFP into the peroxisome over the same timescale; however new protein synthesis in the cytosol obstructs those numbers. Using a constitutively active LOV domain showed only 50% protein trafficked to
122 the peroxisome suggesting that new protein synthesis, as well as possible saturation of the
trafficking machinery, may be at play.
Fusing the tag to other proteins showed that more optimization would be necessary to
adapt the tag to specific systems. Surprisingly, fusing LOV-PTS1 onto mCherry was
unsuccessful in initiating light induced peroxisomal trafficking. Using an inducible
dimerization pair, we hypothesized that we could use our trafficking tag on one half of the
dimer and a mCherry fluorescent tag on the other half to traffic the pair into the peroxisome.
While the tag induced peroxisomal trafficking of the fused half, it was unable to traffic the
dimerized pair into the peroxisome. Literature suggests this was not due to the size of the
complex when dimerized, as the dimerization tags are relatively small, but it cannot be
discounted. A last effort was made to fuse a modulator of actin cytoskeleton, the VCA
domain of N-WASP, to the GFP-LOV-PTS1, but again trafficking was not observed when
illuminated. While other factors such as VCA binding actin and blocking trafficking may be
at play, it was also noted that overexpression of VCA-GFP-LOV-PTS1 disrupted the actin cytoskeleton and may alter the cellular cytoskeleton in a way that blocks the trafficking machinery from properly operating.
Chapter III- Regulating Transcription with Light
The second aim probes the mechanism behind a light induced transcriptional
regulation system. The first optogenetic system studied in this aim involves a light mediated
reduction in transcription using a single component cryptochrome 2 (CRY2)-transcription
factor fusion. Using knowledge gained by investigating this CRY2-transcription factor fusion
tool, we were able to optimize a second optogenetic system involving a split transcription
123 factor system that is reconstituted with the stimulation of blue light giving precise control over turning transcription on.
Fusion of CRY2 to a transcription factor initiates clearing of the protein in the nucleus upon illumination, and this phenotype is associated with a significant reduction in transcription. This clearing occurs upon illumination when CRY2 is fused to tTA and Gal4 transcription factors with nuclei 60% cleared over four hours. Time lapse imaging shows that the clearing starts with a distinct clustering phenotype when illuminated, and these clusters coalesce over time before disappearing. This phenomenon is similar to oligomerization seen with other CRY2-multimeric protein fusions (Lee et al., 2014). Reversion of the clearing in the nucleus is slow and heterogeneous at eight hours, but full reversion is seen at 18 hours.
Western blot shows that total CRY2 transcription factor fusion protein levels are unchanged between dark and light stimulated cells, suggesting that the fusion proteins are not being depleted but rather redistributed.
Due to the similarities of clustering phenotypes with other CRY2-multimeric protein fusions, the dimeric properties of the transcription factors were explored. Gal4 has a well- characterized dimerization domain that when deleted from the CRY2-Gal4 protein ablates the nuclear clearing phenotype after light stimulation and abrogates the functional reduction in transcription. The dimerization deletion (AA 66-95) shows the critical function of the multimeric properties of the transcription factor fused to CRY2, but to further test this we added back a tetrameric dsRED domain and monomeric mCherry domain. The dsRED domain was able to restore light dependent functional loss as well as the nuclear clearing phenotype, whereas the monomeric mCherry domain was unable to restore the original phenotype. The dimerization of the transcription factor is playing a critical role in the
124 clearing process and functional reduction of transcription when fused to CRY2, but what occurs to the protein in order to lead to the nuclear loss in phenotype and function is still unknown.
Originally Gal4-CRY2 and VP16-CIB1 were used to generate a split transcription system, which has a robust light dependent induction of transcription in yeast (Kennedy et al., 2010; Pathak et al., 2014) but has no significant light dependent effect on transcription in mammalian cells. Utilizing the knowledge gained from the previous studies we were able to revive the idea of a mammalian split transcription system with the hypothesis that the dimerization domain of Gal4 led to the depletion of the CRY-DBD protein from the nucleus.
Deletion of the Gal4 dimerization domain in a split transcription factor system robustly increased the amount of transcription stimulated with blue light and reduced the background
“dark” transcription occurring in HEK293T cells
While the original CRY-Gal(1-147) cleared from the nucleus in 18 hours of light illumination, the dimerization deletion CRY-Gal(1-65) was retained in the nucleus over the same time span. Both Gal4-CRY2 and CRY2-Gal4 showed no light dependent induction of transcription when expressed with VP16-CIB, but illumination of the co-expressed VP16-
CIB1 and CRY-Gal(1-65) resulted in robust transcription. The split transcription system was further optimized by altering the activation domain construct demonstrating that orientation matters and VP64 was superior to VP16 with a 101x fold activation level over dark when expressed with CRY-Gal(1-65) (VP16-CIB < CIB-VP16-CIB < CIB-VP16 < CIB-VP64).
This optimized split transcription system was expressed in a GalUAS-GFP transgenic zebrafish vertebrate model, which showed light dependent GFP expression with no visible dark background.
125 Chapter IV- Cryptochrome 2 in the Mammalian Nucleus
The mechanism behind the peroxisomal trafficking LOV-PTS1 tag and split
transcription factor tools were well understood, but how the nuclear clearing and associated
functional loss in the single component transcription factor tool occurred was unknown.
Trying to tease apart the mechanism and learn how cells were responding to the illumination
of overexpressed clustered proteins in the nucleus gave us many insights into what does not
occur, and in the end, never solidified exactly what mechanism was at play.
Various methods of tracking the protein as it cleared from the nucleus were all
inconclusive. Nuclear/cytosolic fractionation was unable to cleanly separate the separate
compartments due to insoluble aggregates complicating the method. An effort to track the
clearing using a photoconvertible green-to-red protein tag also gave inconclusive results due to the photo-instability of the protein along with technical issues with the microscope over the extended live cell imaging capture. This has not resolved the difference between protein being redistributed within the nucleus versus leaving the nucleus.
As rapid degradation in the nucleus seemed most likely and it is known that CRY2 is regulated by phosphorylation in Arabidopsis, post-translational marks were assessed through western blot and mass spectrometry on an immunoprecipitated predominantly nuclear CRY2- tTA (CRY2+NLS-HA-mCh-tTA). A pan-ubiquitin antibody was unable to detect ubiquitination on the immunoprecipitated protein; however, mass spectrometry results of the immunoprecipitated protein detected sites of ubiquitination and phosphorylation. Site- directed mutagenesis of the modified amino acids on CRY2 was not sufficient to block the clearing phenotype. While none of these mutations individually are sufficient to block the
126 phenotype, this does not discount the possibility that these modified residues in tandem or in
series may play a role and should be further addressed.
A series of inhibitors were used to probe specific pathways that were potential
mechanisms of the light initiated nuclear clearing phenotype. Cycloheximide, an inhibitor of
translation, blocked the nuclear clearing phenotype in the presence of light. This experiment
was followed up with an inhibitor of transcription, actinomycin D, which did not block
clearing. Results of these inhibitor experiments suggest something mechanistically necessary is translated upon illumination or protein clustering but the message for translation may already be present in the cells. Multiple other inhibitors were tested that had no effect on nuclear clearing including doxycycline, MG132, chloroquine, nocodazole and leptomycin B, which suggested that DNA binding, proteosomal degradation, autophagy, cell replication, and the CRM1 nuclear export pathway were not individually necessary for the clearing phenotype.
Future Directions
Optogenetics is a field that is constantly in flux with new tools being introduced at a steady rate. To keep tools relevant, they need to be optimized and have broad utility. The
AsLOV2 domain is a well characterized optogenetic photosensory protein and has many
LOV family members that are continually being improved upon. The Tucker laboratory is
working to advance CRY2 tools in multiple ways: testing truncations of CRY2 and its
interacting partner CIB1, finding mutations that alter the kinetics of CRY2’s lit state and
reduce dark state interactions, and finding novel utilizations. These innovations to CRY2
would be able to enhance not only future tools but also previously developed tools. With the
127 development of our new optogenetic systems we demonstrated novel optogenetic approaches but also raised new questions.
While the mechanism of the inducible peroxisomal trafficking tool seems relatively simple, the failed attempts to traffic proteins, other than GFP, into the peroxisome offer complications to be explored and optimized. Experimentation on tool size limitations should be tested by using a series of small tags such as GFP in a 1x, 2x, 3x, etc. addition until a maximum size capability is reached. A tetrameric protein, catalase, which is 240kDa, is known to traffic in a folded state to the peroxisome for import (Léon et al., 2006), so while it is unlikely that size is the limiting factor, synthetic systems can differ from naturally occurring cellular processes.
Utilization of this peroxisomal trafficking tool will be forefront in the future of this novel system. The permanence of the sequestration into peroxisomes is a caveat to using this tool, but this permanent removal may have a use for those not looking for a reversible system. This particular tool may prove most useful for laboratories studying peroxisomal biogenesis disorders and peroxisomal trafficking. Being able to traffic peroxisomal proteins in a local and user defined manner could help uncover the role of various PEX genes in the biogenesis of new peroxisomes.
The optimized split transcription tool shows improvement over known systems with low background, high dynamic range, tunability and spatial precision. The future of this tool lies in utilization to regulate genes of interest with spatial and temporal precision without the side effects of chemically regulated systems. Since there are many options for genetic regulation with optogenetic systems, a full comparison to other systems is still needed.
128 Perhaps the most important knowledge to come out of the transcription factor tools was the role of the dimerization motif in clearing nuclear CRY2 fusion proteins. This discovery allowed us to optimize our split transcription system and develop the single component system. We think this information will have broader implications on CRY2 optogenetic tools used in the nucleus. With CRISPR technology revolutionizing research, it could be advantageous to revisit optimization of the previously developed CRISPR/dCAS9 optogenetic systems which showed low levels of activation, which could be due to the clearing phenomenon. To support this, we saw that we were able to negatively control a
CRISPR tool targeting gene loci in cell culture suggesting that these CRISPR constructs may be cleared from the nucleus (Figure III-9).
The single component system represents a novel approach to negatively regulating transcriptional activity with light, but it is unable to fully block transcription. Additional optimization may improve this system to allow enhanced light/dark differences and reduced background. The technology behind the single component system is truly the future of this tool, as the clearing of the protein from the nucleus in order to block nuclear activity may be applied to other nuclear effectors, for instance proteins that regulate chromatin structure. We believe this technology may be a general-use tool for many nuclear proteins.
The nuclear clearing phenomenon raised many questions that are still unanswered, and our attempts to probe the mechanism deeper led us to additional questions. With the potential implications for this nuclear clearing to be a globally used optogenetic technology, it is important that the mechanism be well understood. We have yet to answer how this clearing is facilitated and what potential signaling events play a role. It is still unclear where the nuclear protein redistributes upon illumination.
129 Staining experiments to determine if the clusters formed in the nucleus co-localized with any known nuclear bodies showed that they were not conclusively PML bodies, but there are multiple other nuclear bodies that could be examined. In addition, PML could be knocked-down to inhibit PML nuclear body formation and clusters could be induced to confirm if PML nuclear bodies are necessary for aggregates. To look for colocalization of other nuclear bodies we could stain for coilin to test if the clusters colocalize with cajal bodies, PSP2 to test for paraspeckles localizatin, or HSF1 to determine if they colocalize with nuclear stress granules. Knowing where these clusters localize could give insights into the mechanism of control.
To follow up on previously explored experiments, it would be important to test the importance of the post-translationally modified amino acids identified in the mass spectrometry screen in tandem and in series with one another. This could be done through the same mutagenesis approach, but inserting multiple mutations onto each construct to test different combinations. This would further answer our question addressing if the identified
PTMs may play a role in the clearing.
To follow up on the cycloheximide result, which showed that inhibition of protein synthesis may block nuclear clearing, it would be important to confirm with other protein synthesis inhibitors that this is a necessary mechanism and not a side-effect of cycloheximide. This experiment has not ruled out the possibility that a protein involved in clearing may have a very short half-life since we are imaging at 8 hours post illumination.
Images at shorter time periods may show partial clearing if a protein with a short half-life is involved in the nuclear clearing mechanism.
130 A proteomics approach should be taken to follow up on the result where cycloheximide was able to block the nuclear clearing but actinomycin D was not. This suggests that protein synthesis is necessary for the mechanism of clearing. Through a mass spectrometry screen of samples exposed to light or dark we can identify proteins that are upregulated by light exposure which may give us information on pathways that are activated.
It is hypothesized that the aggregates formed may induce cell stress pathways, and this experiment could help answer that question.
A more targeted mass spectrometry approach would be to immunoprecipitate samples exposed to light versus dark and follow this with mass spectrometry to identify potential binding partners that may affect light/dark responses. These targets would need to be verified using immunoprecipitation followed by immunoblotting to rule out false positives.
Knockdown experiments of binding partners could tease apart which partners play an active role in the nuclear clearing process.
In conclusion, optogenetics is a dynamic field where tools are constantly being developed, and as optogenetic systems become more widespread throughout molecular and cellular biology it will be important to understand the mechanisms of each tool being used.
Furthermore, since optogenetics uses engineered proteins that are often not endogenous to the cell, it will be important to probe what potential consequences overexpression of these tools have in cell culture and other animal models.
131 REFERENCES
Ando, R., Mizuno, H., and Miyawaki, A. (2004). Regulated fast nucleocytoplasmic shuttling observed by reversible protein highlighting. Science (80-. ). 306, 1370–1373.
Andresen, M., Stiel, A.C., Trowitzsch, S., Weber, G., Eggeling, C., Wahl, M.C., Hell, S.W., and Jakobs, S. (2007). Structural basis for reversible photoswitching in Dronpa. Proc. Natl. Acad. Sci. U. S. A. 104, 13005–13009.
Asakawa, K., Suster, M.L., Mizusawa, K., Nagayoshi, S., Kotani, T., Urasaki, A., Kishimoto, Y., Hibi, M., and Kawakami, K. (2008). Genetic dissection of neural circuits by Tol2 transposon-mediated Gal4 gene and enhancer trapping in zebrafish. Proc. Natl. Acad. Sci. U. S. A. 105, 1255–1260.
Bellini, D., and Papiz, M.Z. (2012). Structure of a bacteriophytochrome and light-stimulated protomer swapping with a gene repressor. Structure 20, 1436–1446. van Bergeijk, P., Adrian, M., Hoogenraad, C.C., and Kapitein, L.C. (2015). Optogenetic control of organelle transport and positioning. Nature 518, 111–114.
Beyer, H.M., Juillot, S., Herbst, K., Samodelov, S.L., Müller, K., Schamel, W.W., Römer, W., Schäfer, E., Nagy, F., Strähle, U., et al. (2015). Red Light-Regulated Reversible Nuclear Localization of Proteins in Mammalian Cells and Zebrafish. ACS Synth. Biol. 4, 951–958.
Blochlinger, K., Jan, L.Y., and Jan, Y.N. (1991). Transformation of sensory organ identity by ectopic expression of Cut in Drosophila. Genes Dev. 5, 1124–1135.
Bonger, K.M., Rakhit, R., Payumo, A.Y., Chen, J.K., and Wandless, T.J. (2014). General method for regulating protein stability with light. ACS Chem. Biol. 9, 111–115.
van den Bosch, H., Schutgens, R.B., Wanders, R.J., and Tager, J.M. (1992). Biochemistry of peroxisomes. Annu Rev Biochem 61, 157–197.
Boulina, M., Samarajeewa, H., Baker, J.D., Kim, M.D., and Chiba, A. (2013). Live imaging of multicolor-labeled cells in Drosophila. Development 140, 1605–1613.
Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415.
Buckley, C.E., Moore, R.E., Reade, A., Goldberg, A.R., Weiner, O.D., and Clarke, J.D.W. (2016). Reversible Optogenetic Control of Subcellular Protein Localization in a Live Vertebrate Embryo. Dev. Cell 36, 117–126.
Bugaj, L.J., Choksi, A.T., Mesuda, C.K., Kane, R.S., and Schaffer, D. V (2013). Optogenetic protein clustering and signaling activation in mammalian cells. Nat. Methods 10, 249–252.
132 Bugaj, L.J., Spelke, D.P., Mesuda, C.K., Varedi, M., Kane, R.S., and Schaffer, D. V (2015). Regulation of endogenous transmembrane receptors through optogenetic Cry2 clustering. Nat. Commun. 6, 6898.
Bush, E.W., and McKinsey, T.A. (2010). Protein acetylation in the cardiorenal axis: The promise of histone deacetylase inhibitors. Circ. Res. 106, 272–284.
Cao, J., Arha, M., Sudrik, C., Schaffer, D. V., and Kane, R.S. (2014). Bidirectional regulation of mRNA translation in mammalian cells by using PUF domains. Angew. Chemie - Int. Ed. 53, 4900–4904.
Carey, M., Kakidani, H., Leatherwood, J., Mostashari, F., and Ptashne, M. (1989). An amino- terminal fragment of GAL4 binds DNA as a dimer. J. Mol. Biol. 209, 423–432.
Chan, Y.-B., Alekseyenko, O. V, and Kravitz, E.A. (2015). Optogenetic Control of Gene Expression in Drosophila. PLoS One 10, e0138181.
Chang, K.-Y., Woo, D., Jung, H., Lee, S., Kim, S., Won, J., Kyung, T., Park, H., Kim, N., Yang, H.W., et al. (2014). Light-inducible receptor tyrosine kinases that regulate neurotrophin signalling. Nat. Commun. 5, 4057.
Chatelle, C. V, Hövermann, D., Müller, A., Wagner, H.J., Weber, W., and Radziwill, G. (2016). Optogenetically controlled RAF to characterize BRAF and CRAF protein kinase inhibitors. Sci. Rep. 6, 23713.
Chen, D., Gibson, E.S., and Kennedy, M.J. (2013). A light-triggered protein secretion system. J. Cell Biol. 201, 631–640.
Cheng, A.W., Wang, H., Yang, H., Shi, L., Katz, Y., Theunissen, T.W., Rangarajan, S., Shivalila, C.S., Dadon, D.B., and Jaenisch, R. (2013). Multiplexed activation of endogenous genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res. 23, 1163– 1171.
Choudhury, S.R., Cui, Y., Narayanan, A., Gilley, D.P., Huda, N., Lo, C.-L., Zhou, F.C., Yernool, D., and Irudayaraj, J. (2016). Optogenetic regulation of site-specific subtelomeric DNAmethylation. Oncotarget 7.
Christie, J.M., Gawthorne, J., Young, G., Fraser, N.J., and Roe, A.J. (2012). LOV to BLUF: Flavoprotein contributions to the optogenetic toolkit. Mol. Plant 5, 533–544.
Conrad, K.S., Manahan, C.C., and Crane, B.R. (2014). Photochemistry of flavoprotein light sensors. Nat. Chem. Biol. 10, 801–809.
Crefcoeur, R.P., Yin, R., Ulm, R., and Halazonetis, T.D. (2013). Ultraviolet-B-mediated induction of protein-protein interactions in mammalian cells. Nat. Commun. 4, 1779.
133 Danpure, C.J., Cooper, P.J., Wise, P.J., and Jennings, P.R. (1989). An enzyme trafficking defect in two patients with primary hyperoxaluria type 1: peroxisomal alanine/glyoxylate aminotransferase rerouted to mitochondria. J Cell Biol 108, 1345–1352.
Devereux, R.B., Wachtell, K., Gerdts, E., Boman, K., Nieminen, M.S., Papademetriou, V., Rokkedal, J., Harris, K., Aurup, P., and Dahlöf, B. (2004). Prognostic significance of left ventricular mass change during treatment of hypertension. JAMA 292, 2350–2356.
Duan, L., Che, D., Zhang, K., Ong, Q., Guo, S., and Cui, B. (2015). Optogenetic Control of Molecular Motors and Organelle Distributions in Cells. Chem. Biol. 22, 671–682.
Dundr, M., and Misteli, T. (2010). Biogenesis of nuclear bodies. Cold Spring Harb. Perspect. Biol. 2.
Editorial (2011). Method of the Year 2010 optogenetics. Nat. Methods 8, 1.
Endo, M., Hattori, M., Toriyabe, H., Ohno, H., Kamiguchi, H., Iino, Y., and Ozawa, T. (2016). Optogenetic activation of axon guidance receptors controls direction of neurite outgrowth. Sci. Rep. 6, 23976.
Espinoza-Derout, J., Wagner, M., Shahmiri, K., Mascareno, E., Chaqour, B., and Siddiqui, M.A.Q. (2007). Pivotal role of cardiac lineage protein-1 (CLP-1) in transcriptional elongation factor P-TEFb complex formation in cardiac hypertrophy. Cardiovasc. Res. 75, 129–138.
Espinoza-Derout, J., Wagner, M., Salciccioli, L., Lazar, J.M., Bhaduri, S., Mascareno, E., Chaqour, B., and Siddiqui, M.A.Q. (2009). Positive transcription elongation factor b activity in compensatory myocardial hypertrophy is regulated by cardiac lineage protein-1. Circ. Res. 104, 1347–1354.
Fields, S., and Song, O. (1989). A novel genetic system to detect protein-protein interactions. Nature 340, 245–246.
Filippakopoulos, P., Qi, J., Picaud, S., Shen, Y., Smith, W.B., Fedorov, O., Morse, E.M., Keates, T., Hickman, T.T., Felletar, I., et al. (2010). Selective inhibition of BET bromodomains. Nature 468, 1067–1073.
Freitas, M.O., Francisco, T., Rodrigues, T.A., Lismont, C., Domingues, P., Pinto, M.P., Grou, C.P., Fransen, M., and Azevedo, J.E. (2015). The peroxisomal protein import machinery displays a preference for monomeric substrates. Open Biol. 5, 140236.
Gambetta, G.A., and Lagarias, J.C. (2001). Genetic engineering of phytochrome biosynthesis in bacteria. Proc. Natl. Acad. Sci. U. S. A. 98, 10566–10571.
Gardin JM, and Lauer MS (2004). Left ventricular hypertrophy: The next treatable, silent killer? JAMA 292, 2396–2398.
134 Gerhardt, K.P., Olson, E.J., Castillo-Hair, S.M., Hartsough, L.A., Landry, B.P., Ekness, F., Yokoo, R., Gomez, E.J., Ramakrishnan, P., Suh, J., et al. (2016). An open-hardware platform for optogenetics and photobiology. Sci. Rep. 6, 35363.
Gilbert, L.A., Larson, M.H., Morsut, L., Liu, Z., Brar, G.A., Torres, S.E., Stern-Ginossar, N., Brandman, O., Whitehead, E.H., Doudna, J.A., et al. (2013). CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442–451.
Giordano, F., Saheki, Y., Idevall-Hagren, O., Colombo, S.F., Pirruccello, M., Milosevic, I., Gracheva, E.O., Bagriantsev, S.N., Borgese, N., and De Camilli, P. (2013). XPI(4,5)P2- Dependent and Ca2+-Regulated ER-PM interactions mediated by the extended synaptotagmins. Cell 153, 1494–1509.
Glover, J.R., Andrews, D.W., and Rachubinski, R.A. (1994). Saccharomyces cerevisiae peroxisomal thiolase is imported as a dimer. Proc. Natl. Acad. Sci. U. S. A. 91, 10541– 10545.
Gomelsky, M., and Klug, G. (2002). BLUF: A novel FAD-binding domain involved in sensory transduction in microorganisms. Trends Biochem. Sci. 27, 497–500.
Gomez, E.J., Gerhardt, K., Judd, J., Tabor, J.J., and Suh, J. (2015). Light-Activated Nuclear Translocation of Adeno-Associated Virus Nanoparticles Using Phytochrome B for Enhanced, Tunable, and Spatially Programmable Gene Delivery. ACS Nano acsnano.5b05558.
González-Reyes, A., and Morata, G. (1990). The developmental effect of overexpressing a Ubx product in Drosophila embryos is dependent on its interactions with other homeotic products. Cell 61, 515–522.
Gossen, M., and Bujard, H. (1992). Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc. Natl. Acad. Sci. U. S. A. 89, 5547–5551.
Gossen, M., Freundlieb, S., Bender, G., Muller, G., Hillen, W., and Bujard, H. (1995). Transcriptional activation by tetracyclines in mammalian cells. Science (80-. ). 268, 1766– 1769.
Gould, S.G., Keller, G.A., and Subramani, S. (1987). Identification of a peroxisomal targeting signal at the carboxy terminus of firefly luciferase. J. Cell Biol. 105, 2923–2931.
Gould, S.J., Keller, G.A., Hosken, N., Wilkinson, J., and Subramani, S. (1989). A conserved tripeptide sorts proteins to peroxisomes. J. Cell Biol. 108, 1657–1664.
Grusch, M., Schelch, K., Riedler, R., Reichhart, E., Differ, C., Berger, W., Inglés-Prieto, Á., and Janovjak, H. (2014). Spatio-temporally precise activation of engineered receptor tyrosine kinases by light. EMBO J. 33, 1713–1726.
Guglielmi, G., Barry, J.D., Huber, W., and De Renzis, S. (2015). An Optogenetic Method to Modulate Cell Contractility during Tissue Morphogenesis. Dev. Cell 35, 646–660.
135 Guntas, G., Hallett, R. a., Zimmerman, S.P., Williams, T., Yumerefendi, H., Bear, J.E., and Kuhlman, B. (2015). Engineering an improved light-induced dimer (iLID) for controlling the localization and activity of signaling proteins. Proc. Natl. Acad. Sci. 112, 112–117.
Halavaty, A.S., and Moffat, K. (2007). N- and C-terminal flanking regions modulate light- induced signal transduction in the LOV2 domain of the blue light sensor phototropin 1 from Avena sativa. Biochemistry 46, 14001–14009.
Hallett, R.A., Zimmerman, S.P., Yumerefendi, H., Bear, J.E., and Kuhlman, B. (2016). Correlating in Vitro and in Vivo Activities of Light-Inducible Dimers: A Cellular Optogenetics Guide. ACS Synth. Biol. 5, 53–64.
Han, T., Chen, Q., and Liu, H. (2016). Engineered photoactivatable genetic switches based on the bacterium phage T7 RNA polymerase. ACS Synth. Biol. acssynbio.6b00248.
Harper, S.M., Neil, L.C., and Gardner, K.H. (2003). Structural basis of a phototropin light switch. Science 301, 1541–1544.
Harper, S.M., Christie, J.M., and Gardner, K.H. (2004). Disruption of the LOV-Jalpha helix interaction activates phototropin kinase activity. Biochemistry 43, 16184–16192.
Heijde, M., and Ulm, R. (2013). Reversion of the Arabidopsis UV-B photoreceptor UVR8 to the homodimeric ground state. Proc. Natl. Acad. Sci. U. S. A. 110, 1113–1118.
Hill, J.A., and Olson, E.N. (2008). Cardiac plasticity. N. Engl. J. Med. 358, 1370–1380.
Hirano, A., Kurabayashi, N., Nakagawa, T., Shioi, G., Todo, T., Hirota, T., and Fukada, Y. (2014). In vivo role of phosphorylation of cryptochrome 2 in the mouse circadian clock. Mol. Cell. Biol. 34, 4464–4473.
Hoepfner, D., Van Den Berg, M., Philippsen, P., Tabak, H.F., and Hettema, E.H. (2001). A role for Vps1p, actin, and the Myo2p motor in peroxisome abundance and inheritance in Saccharomyces cerevisiae. J. Cell Biol. 155, 979–990.
Hoepfner, D., Schildknegt, D., Braakman, I., Philippsen, P., and Tabak, H.F. (2005). Contribution of the endoplasmic reticulum to peroxisome formation. Cell 122, 85–95.
Hughes, R.M., and Lawrence, D.S. (2014). Optogenetic engineering: Light-directed cell motility. Angew. Chemie - Int. Ed. 53, 10904–10907.
Hughes, R.M., Bolger, S., Tapadia, H., and Tucker, C.L. (2012a). Light-mediated control of DNA transcription in yeast. Methods 58, 385–391.
Hughes, R.M., Vrana, J.D., Song, J., and Tucker, C.L. (2012b). Light-dependent, dark- promoted interaction between arabidopsis cryptochrome 1 and phytochrome b proteins. J. Biol. Chem. 287, 22165–22172.
136 Hughes, R.M., Freeman, D.J., Lamb, K.N., Pollet, R.M., Smith, W.J., and Lawrence, D.S. (2015). Optogenetic Apoptosis: Light-Triggered Cell Death. Angew. Chemie - Int. Ed. 54, 12064–12068.
Idevall-Hagren, O., Dickson, E.J., Hille, B., Toomre, D.K., and De Camilli, P. (2012). PNAS Plus: Optogenetic control of phosphoinositide metabolism. Proc. Natl. Acad. Sci. 109, E2316–E2323.
Imaizumi, T., Tran, H.G., Swartz, T.E., Briggs, W.R., and Kay, S.A. (2003). FKF1 is essential for photoperiodic-specific light signalling in Arabidopsis. Nature 426, 302–306.
Ish-Horowicz, D., and Pinchin, S.M. (1987). Pattern abnormalities induced by ectopic expression of the Drosophila gene hairy are associated with repression of ftz transcription. Cell 51, 405–415.
Ish-Horowicz, D., Pinchin, S.M., Ingham, P.W., and Gyurkovics, H.G. (1989). Autocatalytic ftz activation and metameric instability induced by ectopic ftz expression. Cell 57, 223–232.
Islinger, M., Li, K.W., Seitz, J., Völkl, A., and Lüers, G.H. (2009). Hitchhiking of Cu/Zn superoxide dismutase to peroxisomes--evidence for a natural piggyback import mechanism in mammals. Traffic 10, 1711–1721.
Jost, A.P.T., and Weiner, O.D. (2015). Probing Yeast Polarity with Acute, Reversible, Optogenetic Inhibition of Protein Function. ACS Synth. Biol. 4, 1077–1085.
Kaberniuk, A.A., Shemetov, A.A., and Verkhusha, V. V (2016). A bacterial phytochrome- based optogenetic system controllable with near-infrared light. Nat. Methods 13, 1–15.
Kakumoto, T., and Nakata, T. (2013). Optogenetic Control of PIP3: PIP3 Is Sufficient to Induce the Actin-Based Active Part of Growth Cones and Is Regulated via Endocytosis. PLoS One 8, 1–17.
Katsura, Y., Kubota, H., Kunida, K., Kanno, A., Kuroda, S., and Ozawa, T. (2015). An optogenetic system for interrogating the temporal dynamics of Akt. Sci. Rep. 5, 14589.
Kawano, F., Suzuki, H., Furuya, A., and Sato, M. (2015). Engineered pairs of distinct photoswitches for optogenetic control of cellular proteins. Nat. Commun. 6, 6256.
Kawano, F., Okazaki, R., Yazawa, M., and Sato, M. (2016). A photoactivatable Cre–loxP recombination system for optogenetic genome engineering. Nat. Chem. Biol. 1–8.
Kennedy, M.J., Hughes, R.M., Peteya, L. a, Schwartz, J.W., Ehlers, M.D., and Tucker, C.L. (2010). Rapid blue-light-mediated induction of protein interactions in living cells. Nat. Methods 7, 973–975.
Kim, J.M., Lee, M., Kim, N., and Heo, W. Do (2016). Optogenetic toolkit reveals the role of Ca2+ sparklets in coordinated cell migration. Proc. Natl. Acad. Sci. U. S. A. 113, 5952– 5957.
137 Kim, N., Kim, J.M., Lee, M., Kim, C.Y., Chang, K.Y., and Heo, W. Do (2014). Spatiotemporal control of fibroblast growth factor receptor signals by blue light. Chem. Biol. 21, 903–912.
Kohler, A. and Hurt, E. (2007). Exporting RNA from the nucleus to the cytoplasm. Nature Reviews Mol. Cell Biol. 8, 761-773.
Konermann, S., Brigham, M.D., Trevino, A.E., Hsu, P.D., Heidenreich, M., Cong, L., Platt, R.J., Scott, D. a, Church, G.M., and Zhang, F. (2013). Optical control of mammalian endogenous transcription and epigenetic states. Nature 500, 472–476.
Kraft, B.J., Masuda, S., Kikuchi, J., Dragnea, V., Tollin, G., Zaleski, J.M., and Bauer, C.E. (2003). Spectroscopic and mutational analysis of the blue-light photoreceptor AppA: A novel photocycle involving flavin stacking with an aromatic amino acid. Biochemistry 42, 6726– 6734.
Krishnamurthy, V. V., Khamo, J.S., Mei, W., Turgeon, A.J., Ashraf, H.M., Mondal, P., Patel, D.B., Risner, N., Cho, E.E., Yang, J., et al. (2016). Reversible optogenetic control of kinase activity during differentiation and embryonic development. Development 4085–4094.
Kunau, W.H. (2005). Peroxisome biogenesis: End of the debate. Curr. Biol. 15.
Kunau, W.H., Dommes, V., and Schulz, H. (1995). Beta-Oxidation of fatty acids in mitochondria, peroxisomes, and bacteria: A century of continued progress. Prog. Lipid Res. 34, 267–342.
Lallemand-Breitenbach, V., and de Thé, H. (2010). PML nuclear bodies. Cold Spring Harb. Perspect. Biol. 2.
Lee, J., Natarajan, M., Nashine, V.C., Socolich, M., Vo, T., Russ, W.P., Benkovic, S.J., and Ranganathan, R. (2008). Surface sites for engineering allosteric control in proteins. Science (80-. ). 322, 438–442.
Lee, S., Park, H., Kyung, T., Kim, N.Y., Kim, S., Kim, J., and Heo, W. Do (2014). Reversible protein inactivation by optogenetic trapping in cells SUPPLEMENTARY. Nat. Methods 11, 633–636.
Léon, S., Goodman, J.M., and Subramani, S. (2006). Uniqueness of the mechanism of protein import into the peroxisome matrix: Transport of folded, co-factor-bound and oligomeric proteins by shuttling receptors. Biochim. Biophys. Acta - Mol. Cell Res. 1763, 1552–1564.
Leung, D.W., Otomo, C., Chory, J., and Rosen, M.K. (2008). Genetically encoded photoswitching of actin assembly through the Cdc42-WASP-Arp2/3 complex pathway. Proc. Natl. Acad. Sci. U. S. A. 105, 12797–12802.
138 Levskaya, A., Chevalier, A.A., Tabor, J.J., Simpson, Z.B., Levery, L.A., Levy, M., Davidson, E.A., Scouras, A., Ellington, A.D., Marcotte, E.M., et al. (2005). Engineering Escherichia coli to see light. Nature 438, 441–442.
Levskaya, A., Weiner, O.D., Lim, W.A., and Voigt, C.A. (2009). Spatiotemporal control of cell signalling using a light-switchable protein interaction. Nature 461, 997–1001.
Li, L., and Lagarias, J.C. (1994). Phytochrome assembly in living cells of the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U. S. A. 91, 12535–12539.
Liu, H., Yu, X., Li, K., Klejnot, J., Yang, H., Lisiero, D., and Lin, C. (2008). Photoexcited CRY2 interacts with CIB1 to regulate transcription and floral initiation in Arabidopsis. Science (80-. ). 322, 1535–1539.
Liu, H., Gomez, G., Lin, S., Lin, S., and Lin, C. (2012). Optogenetic Control of Transcription in Zebrafish. PLoS One 7, 1–5.
Lloyd-Jones, D., Adams, R., Carnethon, M., De Simone, G., Ferguson, T.B., Flegal, K., Ford, E., Furie, K., Go, A., Greenlund, K., et al. (2009). Heart disease and stroke statistics-- 2009 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 119.
Long, C.S., Henrich, C.J., and Simpson, P.C. (1991). A growth factor for cardiac myocytes is produced by cardiac nonmyocytes. Cell Regul. 2, 1081–1095.
Lungu, O.I., Hallett, R. a, Choi, E.J., Aiken, M.J., Hahn, K.M., and Kuhlman, B. (2012). Designing photoswitchable peptides using the AsLOV2 domain. Chem. Biol. 19, 507–517.
Maeder, M.L., Linder, S.J., Cascio, V.M., Fu, Y., Ho, Q.H., and Joung, J.K. (2013). CRISPR RNA-guided activation of endogenous human genes. Nat. Methods 10, 977–979.
Maiuri, P., Rupprecht, J.F., Wieser, S., Ruprecht, V., Bénichou, O., Carpi, N., Coppey, M., De Beco, S., Gov, N., Heisenberg, C.P., et al. (2015). Actin flows mediate a universal coupling between cell speed and cell persistence. Cell 161, 374–386.
Marmorstein, R., Carey, M., Ptashne, M., and Harrison, S.C. (1992). DNA recognition by GAL4: structure of a protein-DNA complex. Nature 356, 408–414.
Mas, P., Devlin, P.F., Panda, S., and Kay, S.A. (2000). Functional interaction of phytochrome B and cryptochrome 2. Nature 408, 207–211.
Masuda, S., and Bauer, C.E. (2002). AppA is a blue light photoreceptor that antirepresses photosynthesis gene expression in Rhodobacter sphaeroides. Cell 110, 613–623.
Masuda, S., Hasegawa, K., Ohta, H., and Ono, T.A. (2008). Crucial role in light signal transduction for the conserved Met93 of the BLUF protein PixD/Slr1694. Plant Cell Physiol. 49, 1600–1606.
139 McNew, J.A., and Goodman, J.M. (1994). An oligomeric protein is imported into peroxisomes in vivo. J. Cell Biol. 127, 1245–1257.
Mitra, D., Yang, X., and Moffat, K. (2012). Crystal structures of Aureochrome1 LOV suggest new design strategies for optogenetics. Structure 20, 698–706.
Mizuno, H., Dedecker, P., Ando, R., Fukano, T., Hofkens, J., and Miyawaki, A. (2010). Higher resolution in localization microscopy by slower switching of a photochromic protein. Photochem. Photobiol. Sci. 9, 239.
Moglich, A., Yang, X., Ayers, R.A., and Moffat, K. (2010). Structure and function of plant photoreceptors. Annu Rev Plant Biol 61, 21–47.
Moon, K.J., Mochizuki, K., Zhou, M., Jeong, H.S., Brady, J.N., and Ozato, K. (2005). The bromodomain protein Brd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II-dependent transcription. Mol. Cell 19, 523–534.
Moreno, C.M., Dixon, R.E., Tajada, S., Yuan, C., Opitz-Araya, X., Binder, M.D., and Santana, L.F. (2016). Ca2+ entry into neurons is facilitated by cooperative gating of clustered Cav1.3 channels. Elife 5, 1–26.
Motta-Mena, L.B., Reade, A., Mallory, M.J., Glantz, S., Weiner, O.D., Lynch, K.W., and Gardner, K.H. (2014). An optogenetic gene expression system with rapid activation and deactivation kinetics. Nat. Chem. Biol. 10, 196–202.
Müller, K., Engesser, R., Timmer, J., Nagy, F., Zurbriggen, M.D., and Weber, W. (2013a). Synthesis of phycocyanobilin in mammalian cells. Chem. Commun. 49, 8970–8972.
Müller, K., Engesser, R., Schulz, S., Steinberg, T., Tomakidi, P., Weber, C.C., Ulm, R., Timmer, J., Zurbriggen, M.D., and Weber, W. (2013b). Multi-chromatic control of mammalian gene expression and signaling. Nucleic Acids Res. 41.
Müller, K., Engesser, R., Metzger, S., Schulz, S., Kämpf, M.M., Busacker, M., Steinberg, T., Tomakidi, P., Ehrbar, M., Nagy, F., et al. (2013c). A red/far-red light-responsive bi-stable toggle switch to control gene expression in mammalian cells. Nucleic Acids Res. 41.
Müller, K., Engesser, R., Timmer, J., Zurbriggen, M.D., and Weber, W. (2014). Orthogonal optogenetic triple-gene control in mammalian cells. ACS Synth. Biol. 3, 796–801.
Nagel, G., Szellas, T., Huhn, W., Kateriya, S., Adeishvili, N., Berthold, P., Ollig, D., Hegemann, P., and Bamberg, E. (2003). Channelrhodopsin-2, a directly light-gated cation- selective membrane channel. Proc. Natl. Acad. Sci. U. S. A. 100, 13940–13945.
Nash, A.I., McNulty, R., Shillito, M.E., Swartz, T.E., Bogomolni, R.A., Luecke, H., and Gardner, K.H. (2011). Structural basis of photosensitivity in a bacterial light-oxygen- voltage/helix-turn-helix (LOV-HTH) DNA-binding protein. Proc. Natl. Acad. Sci. 108, 9449–9454.
140 Nguyen, K.T., Halloway, M.P., Altura, R.A. (2012). The CRM1 nuclear export protein in normal develeopment and disease. Int J Biochem Mol Biol. 2012; 3(2): 137-151
Nguyen, M.K., Kim, C.Y., Kim, J.M., Park, B.O., Lee, S., Park, H., and Heo, W. Do (2016a). Optogenetic oligomerization of Rab GTPases regulates intracellular membrane trafficking. Nat. Chem. Biol. 12, 1–8.
Nguyen, T.T.T., Park, W.S., Park, B.O., Kim, C.Y., Oh, Y., Kim, J.M., Choi, H., Kyung, T., Kim, C.-H., Lee, G., et al. (2016b). PLEKHG3 enhances polarized cell migration by activating actin filaments at the cell front. Proc. Natl. Acad. Sci. 113, 201604720.
Ni, M., Tepperman, J.M., and Quail, P.H. (1999). Binding of phytochrome B to its nuclear signalling partner PIF3 is reversibly induced by light. Nature 400, 781–784.
Nihongaki, Y., Suzuki, H., Kawano, F., and Sato, M. (2014). Genetically engineered photoinducible homodimerization system with improved dimer-forming efficiency. ACS Chem. Biol. 9, 617–621.
Nihongaki, Y., Yamamoto, S., Kawano, F., Suzuki, H., and Sato, M. (2015a). CRISPR-Cas9- based photoactivatable transcription system. Chem. Biol. 22, 169–174.
Nihongaki, Y., Kawano, F., Nakajima, T., and Sato, M. (2015b). Photoactivatable CRISPR- Cas9 for optogenetic genome editing. Nat Biotechnol 33, 755–760.
Niopek, D., Benzinger, D., Roensch, J., Draebing, T., Wehler, P., Eils, R., and Di Ventura, B. (2014). Engineering light-inducible nuclear localization signals for precise spatiotemporal control of protein dynamics in living cells. Nat Commun 5, 4404.
Niopek, D., Wehler, P., Roensch, J., Eils, R., and Di Ventura, B. (2016). Optogenetic control of nuclear protein export. Nat Commun 7, 10624.
O’Neill, P.R., Kalyanaraman, V., and Gautam, N. (2016). Subcellular optogenetic activation of Cdc42 controls local and distal signaling to drive immune cell migration. Mol. Biol. Cell 27, 1442–1450.
Ogura, E., Okuda, Y., Kondoh, H., and Kamachi, Y. (2009). Adaptation of GAL4 activators for GAL4 enhancer trapping in zebrafish. Dev. Dyn. 238, 641–655.
Ong, Q., Guo, S., Duan, L., Zhang, K., Collier, E.A., and Cui, B. (2016). The Timing of Raf/ERK and AKT Activation in Protecting PC12 Cells against Oxidative Stress. PLoS One 11, e0153487.
Ozkan-Dagliyan, I., Chiou, Y.Y., Ye, R., Hassan, B.H., Ozturk, N., and Sancar, A. (2013). Formation of arabidopsis cryptochrome 2 photobodies in mammalian nuclei: Application as an optogenetic DNA damage checkpoint switch. J. Biol. Chem. 288, 23244–23251.
141 Palmer, J.N., Hartogensis, W.E., Patten, M., Fortuin, F.D., and Long, C.S. (1995). Interleukin-1beta induces cardiac myocyte growth but inhibits cardiac fibroblast proliferation in culture. J. Clin. Invest. 95, 2555–2564.
Parkhurst, S.M., and Ish-Horowicz, D. (1991). Mis-regulating segmentation gene expression in Drosophila. Development 111, 1121–1135.
Parkhurst, S.M., Bopp, D., and Ish-Horowicz, D. (1990). X:A ratio, the primary sex- determining signal in Drosophila, is transduced by helix-loop-helix proteins. Cell 63, 1179– 1191.
Pathak, G.P., Strickland, D., Vrana, J.D., and Tucker, C.L. (2014). Benchmarking of optical dimerizer systems. ACS Synth. Biol. 3, 832–838.
Paul, P., Simm, S., Blaumeiser, A., Scharf, K.-D., Fragkostefanakis, S., Mirus, O., and Schleiff, E. (2013). The protein translocation systems in plants - composition and variability on the example of Solanum lycopersicum. BMC Genomics 14, 189.
Perez-Pinera, P., Kocak, D.D., Vockley, C.M., Adler, A.F., Kabadi, A.M., Polstein, L.R., Thakore, P.I., Glass, K.A., Ousterout, D.G., Leong, K.W., et al. (2013). RNA-guided gene activation by CRISPR-Cas9-based transcription factors. Nat. Methods 10, 973–976.
Pham, E., Mills, E., and Truong, K. (2011). A synthetic photoactivated protein to generate local or global Ca(2+) signals. Chem. Biol. 18, 880–890.
Polstein, L.R., and Gersbach, C.A. (2012). Light-inducible spatiotemporal control of gene activation by customizable zinc finger transcription factors. J Am Chem Soc 134, 16480– 16483.
Polstein, L.R., and Gersbach, C.A. (2015). A light-inducible CRISPR-Cas9 system for control of endogenous gene activation. Nat. Chem. Biol. 11, 198–200.
Prinjha, R.K., Witherington, J., and Lee, K. (2012). Place your BETs: The therapeutic potential of bromodomains. Trends Pharmacol. Sci. 33, 146–153.
Rao, M.V., Chu, P.-H., Hahn, K.M., and Zaidel-Bar, R. (2013). An optogenetic tool for the activation of endogenous diaphanous-related formins induces thickening of stress fibers without an increase in contractility. Cytoskeleton (Hoboken). 70, 394–407.
Renicke, C., Schuster, D., Usherenko, S., Essen, L.-O., and Taxis, C. (2013). A LOV2 domain-based optogenetic tool to control protein degradation and cellular function. Chem. Biol. 20, 619–626.
Rennel, E., and Gerwins, P. (2002). How to make tetracycline-regulated transgene expression go on and off. Anal. Biochem. 309, 79–84.
142 Rizzini, L., Favory, J.-J., Cloix, C., Faggionato, D., O’Hara, A., Kaiserli, E., Baumeister, R., Schäfer, E., Nagy, F., Jenkins, G.I., et al. (2011). Perception of UV-B by the Arabidopsis UVR8 protein. Science (80-. ). 332, 103–106.
Rockman, H.A., Ross, R.S., Harris, A.N., Knowlton, K.U., Steinhelper, M.E., Field, L.J., Ross, J., and Chien, K.R. (1991). Segregation of atrial-specific and inducible expression of an atrial natriuretic factor transgene in an in vivo murine model of cardiac hypertrophy. Proc. Natl. Acad. Sci. U. S. A. 88, 8277–8281.
Rockwell, N.C., Su, Y.S., and Lagarias, J.C. (2006). Phytochrome structure and signaling mechanisms. Annu Rev Plant Biol 57, 837–858.
Roger, V.L., Go, A.S., Lloyd-Jones, D.M., and Benjamin, E.J. (2011). Heart Disease and Stroke Statistics--2012 Update: A Report From the American Heart Association.
Rottwinkel, G., Oberpichler, I., and Lamparter, T. (2010). Bathy phytochromes in rhizobial soil bacteria. J. Bacteriol. 192, 5124–5133.
Sadowski, I., Ma, J., Triezenberg, S., and Ptashne, M. (1988). GAL4-VP16 is an unusually potent transcriptional activator. Nature 335, 563–564.
Salomon, M., Christie, J.M., Knieb, E., Lempert, U., and Briggs, W.R. (2000). Photochemical and mutational analysis of the FMN-binding domains of the plant blue light receptor, phototropin. Biochemistry 39, 9401–9410.
Sano, M., and Schneider, M.D. (2004). Cyclin-dependent kinase-9: An RNAPII kinase at the nexus of cardiac growth and death cascades. Circ. Res. 95, 867–876.
Sano, M., Abdellatif, M., Oh, H., Xie, M., Bagella, L., Giordano, A., Michael, L.H., DeMayo, F.J., and Schneider, M.D. (2002). Activation and function of cyclin T-Cdk9 (positive transcription elongation factor-b) in cardiac muscle-cell hypertrophy. Nat. Med. 8, 1310–1317.
Satoh, N., Suter, T.M., Liao, R., and Colucci, W.S. (2000). Chronic alpha-adrenergic receptor stimulation modulates the contractile phenotype of cardiac myocytes in vitro. Circulation 102, 2249–2254.
Sawa, M., Nusinow, D.A., Kay, S.A., and Imaizumi, T. (2007). FKF1 and GIGANTEA complex formation is required for day-length measurement in Arabidopsis. Science (80-. ). 318, 261–265.
Schindler, S.E., McCall, J.G., Yan, P., Hyrc, K.L., Li, M., Tucker, C.L., Lee, J.-M., Bruchas, M.R., and Diamond, M.I. (2015). Photo-activatable Cre recombinase regulates gene expression in vivo. Sci. Rep. 5, 13627.
Schmidt, D., Tillberg, P.W., Chen, F., and Boyden, E.S. (2014). A fully genetically encoded protein architecture for optical control of peptide ligand concentration. Nat. Commun. 5, 3019.
143 Schneuwly, S., Klemenz, R., and Gehring, W.J. (1987). Redesigning the body plan of Drosophila by ectopic expression of the homoeotic gene Antennapedia. Nature 325, 816– 818.
Schollenberger, L., Gronemeyer, T., Huber, C.M., Lay, D., Wiese, S., Meyer, H.E., Warscheid, B., Saffrich, R., Peränen, J., Gorgas, K., et al. (2010). RhoA Regulates Peroxisome Association to Microtubules and the Actin Cytoskeleton. PLoS One 5.
Schwerdtfeger, C., and Linden, H. (2003). VIVID is a flavoprotein and serves as a fungal blue light photoreceptor for photoadaptation. EMBO J. 22, 4846–4855.
Shalitin, D., Yang, H., Mockler, T.C., Maymon, M., Guo, H., Whitelam, G.C., and Lin, C. (2002). Regulation of Arabidopsis cryptochrome 2 by blue-light- dependent phosphorylation. Nature 417, 763–767.
Shimizu-Sato, S., Huq, E., Tepperman, J.M., and Quail, P.H. (2002). A light-switchable gene promoter system. Nat. Biotechnol. 20, 1041–1044.
South, S.T., and Gould, S.J. (1999). Peroxisome synthesis in the absence of preexisting peroxisomes. J. Cell Biol. 144, 255–266.
Spencer, D.M., Wandless, T.J., Schreiber, S.L., and Crabtree, G.R. (1993). Controlling Signal Transduction with Synthetic Ligands. Science (80-. ). 262, 1019–1024.
Spencer T. Glantz, Eric J. Carpenter, Michael Melkonian, Kevin H. Gardner, Edward S. Boyden, Gane Ka-Shu Wong, Brian Y. Chow, Glantz, S.T., Carpenter, E.J., Melkonian, M., et al. (2016). Functional and topological diversity of LOV domain photoreceptors. Proc. Natl. Acad. Sci. 113, 201509428.
Speese, S.D., Ashley, J., Jokhi, V., Nunnari, J., Barraia, R., Li, Y. Ataman, B., Koon, A., Chang, Y., Li, Q., Moore, M.J., Budnik, V. (2012). Nuclear envelope budding enables large ribonucleoprotein particle export during synaptic Wnt signaling. Cell 149(4), 832-846.
Staff, T. new (2010). Insights of the decade. Stepping away from the trees for a look at the forest. Introduction. Science 330, 1612–1613.
Steingrimsson, E., Pignoni, F., Liaw, G.J., and Lengyel, J.A. (1991). Dual role of the Drosophila pattern gene tailless in embryonic termini. Science (80). 254, 418–421.
Stephen, J. (1992). Transport of microinjected proteins into peroxisomes of mammalian cells : inability of Zellweger cell lines to import proteins with the SKL tripeptide peroxisomal targeting signal . Transport of Microinjected Proteins into Peroxisomes of Mammalian Cells :
Strickland, D., Moffat, K., and Sosnick, T.R. (2008). Light-activated DNA binding in a designed allosteric protein. Pnas 105, 10709–10714.
Strickland, D., Yao, X., Gawlak, G., Rosen, M.K., Gardner, K.H., and Sosnick, T.R. (2010). Rationally improving LOV domain-based photoswitches. Nat Methods 7, 623–626.
144 Strickland, D., Lin, Y., Wagner, E., Hope, C.M., Zayner, J., Antoniou, C., Sosnick, T.R., Weiss, E.L., and Glotzer, M. (2012). TULIPs: tunable, light-controlled interacting protein tags for cell biology. Nat. Methods 9, 379–384.
Struhl, G. (1985). Near-reciprocal phenotypes caused by inactivation or indescrimnate expression of the Drosophila segmentation gene ftz. Nature 318, 677–680.
Swartz, T.E., Corchnoy, S.B., Christie, J.M., Lewis, J.W., Szundi, I., Briggs, W.R., and Bogomolni, R.A. (2001). The photocycle of a flavin-binding domain of the blue light photoreceptor phototropin. J. Biol. Chem. 276, 36493–36500.
Takahashi, F., Yamagata, D., Ishikawa, M., Fukamatsu, Y., Ogura, Y., Kasahara, M., Kiyosue, T., Kikuyama, M., Wada, M., and Kataoka, H. (2007). AUREOCHROME, a photoreceptor required for photomorphogenesis in stramenopiles. Proc. Natl. Acad. Sci. U. S. A. 104, 19625–19630.
Taslimi, A., Vrana, J.D., Chen, D., Borinskaya, S., Mayer, B.J., Kennedy, M.J., and Tucker, C.L. (2014). An optimized optogenetic clustering tool for probing protein interaction and function. Nat. Commun. 5, 4925.
Taslimi, A., Zoltowski, B., Miranda, J.G., Pathak, G.P., Hughes, R.M., and Tucker, C.L. (2016). Optimized second-generation CRY2–CIB dimerizers and photoactivatable Cre recombinase. Nat. Chem. Biol. 12, 1–8.
Toettcher, J.E., Weiner, O.D., and Lim, W.A. (2013). Using optogenetics to interrogate the dynamic control of signal transmission by the Ras/Erk module. Cell 155, 1422–1434.
Tschowri, N., Busse, S., and Hengge, R. (2009). The BLUF-EAL protein YcgF acts as a direct anti-repressor in a blue-light response of Escherichia coli. Genes Dev. 23, 522–534.
Tucker, C.L., Peteya, L.A., Pittman, A.M., and Zhong, J. (2009). A genetic test for yeast two- hybrid bait competency using RanBPM. Genetics 182, 1377–1379.
Tyszkiewicz, A.B., and Muir, T.W. (2008). Activation of protein splicing with light in yeast. Nat. Methods 5, 303–305.
Valon, L., Etoc, F., Remorino, A., Di Pietro, F., Morin, X., Dahan, M., and Coppey, M. (2015). Predictive Spatiotemporal Manipulation of Signaling Perturbations Using Optogenetics. Biophys. J. 109, 1785–1797.
Veenhuis, M., Mateblowski, M., Kunau, W.H., and Harder, W. (1987). Proliferation of microbodies in Saccharomyces cerevisiae. Yeast 3, 77–84.
Wagner, E., and Glotzer, M. (2016). Local RhoA activation induces cytokinetic furrows independent of spindle position and cell cycle stage. J. Cell Biol. 213, 641–649.
145 Wang, H., Vilela, M., Winkler, A., Tarnawski, M., Schlichting, I., Yumerefendi, H., Kuhlman, B., Liu, R., Danuser, G., and Hahn, K.M. (2016). LOVTRAP: an optogenetic system for photoinduced protein dissociation. Nat. Methods 13, 755–758.
Wang, Q., Barshop, W.D., Bian, M., Vashisht, A.A., He, R., Yu, X., Liu, B., Nguyen, P., Liu, X., Zhao, X., et al. (2015). The blue light-dependent phosphorylation of the CCE domain determines the photosensitivity of Arabidopsis CRY2. Mol. Plant 8, 631–643.
Wang, X., Chen, X., and Yang, Y. (2012). Spatiotemporal control of gene expression by a light-switchable transgene system. Nat Methods 9, 266–269.
Wend, S., Wagner, H.J., Muller, K., Zurbriggen, M.D., Weber, W., and Radziwill, G. (2014). Optogenetic control of protein kinase activity in mammalian cells. ACS Synth. Biol. 3, 280– 285.
Wendland, M., and Subramani, S. (1993). Cytosol-dependent peroxisomal protein import in a permeabilized cell system. J. Cell Biol. 120, 675–685.
Wieland, M., Müller, M., Kyburz, A., Heissig, P., Wekenmann, S., Stolz, F., Ausländer, S., and Fussenegger, M. (2014). Engineered UV-A light-responsive gene expression system for measuring sun cream efficacy in mammalian cell culture. J. Biotechnol. 189, 150–153.
Wu, Y.I., Frey, D., Lungu, O.I., Jaehrig, A., Schlichting, I., Kuhlman, B., and Hahn, K.M. (2009). A genetically encoded photoactivatable Rac controls the motility of living cells. Nature 461, 104–108.
Xu, Y., Nan, D., Fan, J., Bogan, J.S., and Toomre, D. (2016). Optogenetic activation reveals distinct roles of PI(3,4,5)P3 and Akt in adipocyte insulin action. J. Cell Sci. 104–125.
Yang, X., Jost, A.P.-T., Weiner, O.D., and Tang, C. (2013). A light-inducible organelle- targeting system for dynamically activating and inactivating signaling in budding yeast. Mol. Biol. Cell 24, 2419–2430.
Yazawa, M., Sadaghiani, A.M., Hsueh, B., and Dolmetsch, R.E. (2009). Induction of protein- protein interactions in live cells using light. Nat. Biotechnol. 27, 941–945.
Yim, N., Ryu, S.-W., Choi, K., Lee, K.R., Lee, S., Choi, H., Kim, J., Shaker, M.R., Sun, W., Park, J.-H., et al. (2016). Exosome engineering for efficient intracellular delivery of soluble proteins using optically reversible protein–protein interaction module. Nat. Commun. 7, 12277.
Yu, G., Onodera, H., Aono, Y., Kawano, F., Ueda, Y., Furuya, A., Suzuki, H., Sato, M., Wettschureck, N., Offermanns, S., et al. (2016). Optical manipulation of the alpha subunits of heterotrimeric G proteins using photoswitchable dimerization systems. Sci. Rep. 6, 35777.
146 Yu, X., Sayegh, R., Maymon, M., Warpeha, K., Klejnot, J., Yang, H., Huang, J., Lee, J., Kaufman, L., and Lin, C. (2009). Formation of nuclear bodies of Arabidopsis CRY2 in response to blue light is associated with its blue light-dependent degradation. Plant Cell 21, 118–130.
Yumerefendi, H., Dickinson, D.J., Wang, H., Zimmerman, S.P., Bear, J.E., Goldstein, B., Hahn, K., and Kuhlman, B. (2015). Control of protein activity and cell fate specification via light-mediated nuclear translocation. PLoS One 10, 1–19.
Zayner, J.P., Antoniou, C., and Sosnick, T.R. (2012). The amino-terminal helix modulates light-activated conformational changes in AsLOV2. J. Mol. Biol. 419, 61–74.
Zhang, K., Duan, L., Ong, Q., Lin, Z., Varman, P.M., Sung, K., and Cui, B. (2014). Light- mediated kinetic control reveals the temporal effect of the Raf/MEK/ERK pathway in PC12 cell neurite outgrowth. PLoS One 9.
Zhang, M., Chang, H., Zhang, Y., Yu, J., Wu, L., Ji, W., Chen, J., Liu, B., Lu, J., Liu, Y., et al. (2012). Rational design of true monomeric and bright photoactivatable fluorescent proteins. Nat. Methods 9, 727–729.
Zhou, X.X., Chung, H.K., Lam, A.J., and Lin, M.Z. (2012). Optical Control of Protein Activity by Fluorescent Protein Domains. 338, 619–621.
Zimmerman, S.P., Hallett, R., Bourke, A.M., Bear, J.E., Kennedy, M.J., and Kuhlman, B. (2016). Tuning the Binding Affinities and Reversion Kinetics of a Light Inducible Dimer Allows Control of Transmembrane Protein Localization. Biochemistry acs.biochem.6b00529.
Zirak, P., Penzkofer, A., Schiereis, T., Hegemann, P., Jung, A., and Schlichting, I. (2006). Photodynamics of the small BLUF protein BlrB from Rhodobacter sphaeroides. J. Photochem. Photobiol. B Biol. 83, 180–194.
Zoltowski, B.D., and Crane, B.R. (2008). Light activation of the LOV protein vivid generates a rapidly exchanging dimer. Biochemistry 47, 7012–7019.
Zoltowski, B.D., and Imaizumi, T. (2014). Structure and Function of the ZTL/FKF1/LKP2 Group Proteins in Arabidopsis. In The Enzymes, pp. 213–239.
Zoltowski, B.D., Schwerdtfeger, C., Widom, J., Loros, J.J., Bilwes, A.M., Dunlap, J.C., and Crane, B.R. (2007). Conformational switching in the fungal light sensor Vivid. Science (80-. ). 316, 1054–1057.
Zoltowski, B.D., Vaccaro, B., and Crane, B.R. (2009). Mechanism-based tuning of a LOV domain photoreceptor. Nat. Chem. Biol. 5, 827–834.
Zuker, C.S., Mismer, D., Hardy, R., and Rubin, G.M. (1988). Ectopic expression of a minor Drosophila opsin in the major photoreceptor cell class: Distinguishing the role of primary receptor and cellular context. Cell 53, 475–482.
147 Zuo, Z.C., Meng, Y.Y., Yu, X.H., Zhang, Z.L., Feng, D.S., Sun, S.F., Liu, B., and Lin, C.T. (2012). A study of the blue-light-dependent phosphorylation, degradation, and photobody formation of Arabidopsis CRY2. Mol. Plant 5, 726–733.
148 APPENDIX A
BET ACETYL-LYSINE BINDING PROTEINS CONTROL PATHOLOGICAL
CARDIAC HYPERTROPHY3
Overview
Cardiac hypertrophy is an independent predictor of adverse outcomes in patients with
heart failure, and thus represents an attractive target for novel therapeutic intervention. JQ1,
a small molecule inhibitor of bromodomain and extraterminal (BET) acetyl-lysine reader
proteins, was identified in a high throughput screen designed to discover novel small
molecule regulators of cardiomyocyte hypertrophy. JQ1 dose-dependently blocked agonist-
dependent hypertrophy of cultured neonatal rat ventricular myocytes (NRVMs) and reversed
the prototypical gene program associated with pathological cardiac hypertrophy. JQ1 also
blocked left ventricular hypertrophy (LVH) and improved cardiac function in adult mice
subjected to transverse aortic constriction (TAC). The BET family consists of BRD2, BRD3,
BRD4 and BRDT. BRD4 protein expression was increased during cardiac hypertrophy, and
hypertrophic stimuli promoted recruitment of BRD4 to the transcriptional start site (TSS) of
the gene encoding atrial natriuretic factor (ANF). Binding of BRD4 to the ANF TSS was
associated with increased phosphorylation of local RNA polymerase II. These findings
define a novel function for BET proteins as signal-responsive regulators of cardiac
hypertrophy, and suggest that small molecule inhibitors of these epigenetic reader proteins
have potential as therapeutics for heart failure.
3 A portion of this research is published with permission from the Journal of Molecular and Cellular Cardiology: Spiltoir et al. Oct(63) (2013): 175-179, © Elsevier
149 Introduction
Approximately 6 million Americans suffer from heart failure, placing an economic burden on the United States that is projected to rise to nearly $100 billion annually by 2030
(Roger et al., 2011). The 5-year mortality rate following first admission for heart failure is
42.3%, further highlighting an urgent need for novel therapeutic approached (Lloyd-Jones et al., 2009). A hallmark of heart failure is cardiomyocyte hypertrophy. Cardiac hypertrophy has historically been viewed as a compensatory mechanism that normalizes wall stress and
enhances cardiac performance. However, long-term suppression of left ventricular
hypertrophy (LVH) is associated with reduced morbidity and mortality in patients with
cardiovascular disease (Devereux et al., 2004; Gardin JM and Lauer MS, 2004), and thus
cardiac hypertrophy is now recognized as an attractive target for novel therapeutic
intervention (Hill and Olson, 2008).
Acetylation of lysine residues in core histones within chromatin provides an
important post-translational mechanism for regulating gene expression, and enzymes that
govern lysine acetylation have emerged as critical regulators of cardiac hypertrophy. Genetic
and pharmacological manipulation of histone deacetylases (HDACs) and histone
acetyltransferases (HATs) in rodent models of heart failure has revealed several key roles for
these enzymes as positive and negative regulators of pathological cardiac growth (Bush and
McKinsey, 2010). However, it is not known whether acetyl-lysine ‘reader’ proteins, which
recruit multi-protein complexes to chromatin via bromodomains, function in the control of
cardiac hypertrophy. Here, we demonstrate that selective small molecule inhibition of
bromodomain and extraterminal (BET) family proteins blocks agonist-dependent
hypertrophy of cultured cardiac myocytes, and is efficacious in a mouse model of pressure
150 overload-induced LVH. These findings establish a novel role for acetyl-lysine binding
proteins in the control of pathological cardiac remodeling, and suggest potential for BET
protein inhibitors for the treatment of heart failure.
Materials and Methods
Experimental animals
Animal experiments were done in accordance with the Institutional Animal Care and Use
Committee at the University of Colorado Denver. Ten week-old male C57BL6 mice (Jackson
Laboratories) were used for transverse aortic constriction (TAC). TAC and sham surgeries
were performed as previously described, using a 27 gauge needle to guide suture
constriction(Rockman et al., 1991). JQ1 was delivered via intraperitoneal injection at a
concentration of 50 mg/kg in a 1:4 DMSO: 10% (2-Hydroxypropyl)-β-cyclodextrin (Sigma
Aldrich; 389145) vehicle starting one day post TAC surgery. Treatment groups were weight- matched prior to start of study.
Hemodynamic analysis
Echocardiographic analyses were performed the day the animals were euthanized using a
Vevo770 System equipped with a 30 mHz frequency mechanical transducer (VisualSonics).
Hearts were imaged in the two-dimensional parasternal short axis. M-mode images were
recorded to measure LV wall dimensions and internal diameter at the level of the papillary
muscles. For analyses, animals were anesthetized using 2% isoflurane and their body
temperature was maintained at 37°C. For data from all in vivo studies, GraphPad Prism
software was used to generate graphs and analyze data. One-way ANOVA with Newman-
Keuls post-hoc test (p<0.05) was used to determine statistical differences between groups.
151 Tissue procurement
After hemodynamic recordings, mice were sacrificed by exsanguination and hearts were excised and placed in ice-cold saline. RV was dissected from LV by cutting along the septum and the outer wall of the LV. A 50 mg biopsy from the LV was flash-frozen in liquid nitrogen for subsequent biochemical and gene expression analyses. Tibia was cleaned of cartilage and length was measured.
Cell culture
Neonatal rat ventricular myocytes (NRVMs) and neonatal rat ventricular fibroblasts
(NRVFs) were isolated from the hearts of 1-3 day-old Sprague Dawley rats (Charles River), as previously described (Long et al., 1991; Palmer et al., 1995). Cell counting and viability was assayed using a Vi-Cell Cell Viability Analyzer (Beckman Coulter). Cells were incubated overnight on 10-cm plates or 96-well clear-bottom plates (Greiner) coated with
0.2% gelatin (Sigma; G9391) in DMEM with 10% calf serum, 2mM L-glutamine, and penicillin-streptomycin. The following morning, cells were washed with serum-free medium and maintained in DMEM supplemented with L-glutamine, penicillin-streptomycin and
Neutridoma-SP (0.1%; Roche Applied Science), which contains albumin, insulin, transferrin, and other defined organic and inorganic compounds. Adult rat ventricular myocytes
(ARVMs) and adult rat ventricular fibroblasts (ARVFs) were obtained from female SD rats, as described previously(Satoh et al., 2000). ARVMs were plated at a density of 100 to 150 cells/mm2 on laminin-coated 10-cm plates and maintained in serum-free DMEM supplemented with albumin (2 mg/ml), 2,3-butanedione monoxime (1 mg/ml), L-carnitine (2 mM), creatine (5 mM), penicillin-streptomycin (100 µg/ml), triiodothyronine (1 pM), and
152 Table A-1: Primer sequences used for qPCR and chromatin immunoprecipitation.
!
Species Target Forward Primer Reverse Primer Rat β2M 5’- 5’-TCTCGGTCCCAGGTGAC TCACACCCACCGAGACCGAT GGTTT-3’ GT-3’ BNP 5’-GGTGCTGCCCCAGATGATT- 5’-CTGGAGACTGGCTAGGA 3’ CTTC-3’ α-Sk 5’- 5’-GACATGACGTTGTTGGC Actin CCACCTACAACAGCATCATGA GTACA-3’ AGT-3’ SERCA 5’-GGCCAGATCGCGCTACA-3’ 5’- GGGCCAATTAGAGAGCA GGTTT-3’ α-MHC 5’-CCTGTCCAGCAGAAAGAG- 5’- 3’ CAGGCAAAGTCAAGCAT TCATATTTATTGTG-3’ β-MHC 5’-CGCTCAGTCATGGCGGAT- 5’-GCCCCAAATGCAGCCA 3’ T-3’ ANF 5’- 5’- TSS GAGACAGTGACGGACAAAG CTTGGTGATGGAGAAGGA G-3’ GC-3’ Mouse 18S 5’- 5’- GCCGCTAGAGGTGAAATTCTT CTTTCGCTCTGGTCCGTC A-3’ TT-3’ α-Sk 5’- 5’- Actin GAGACCACCTACAACAGCAT GACATGACGTTGTTGGCG C TACA-3’ AT-3’ ANF 5’- 5’- GCCGGTAGAAGATGAGGTCA GCTTCCTCAGTCTGCTCA TG-3’ CTCA-3’
153 taurine (5 mM). In all studies, cells were treated in maintenance media for 48 hrs in the
absence or presence of agonists.
Gene expression analysis
Total RNA was isolated from cells and tissue using TRIzol (Life Technologies; 15596) and
tissues were homogenized using a Bullet Blender (Next Advance). For analysis of BET
family member mRNA expression in primary neonatal and adult cardiac myocytes and
fibroblasts, RNA was prepared from freshly isolated cells prior to culture. A Verso™ cDNA
Synthesis Kit (Thermo Scientific) was used to convert 500ng of RNA to cDNA. Quantitative
PCR was performed using a DyNAmo Flash SYBR Green qPCR Kit (ThermoScientific; F-
415) on a StepOne qPCR Instrument (Applied Biosystems). Sequences of primers used for
qPCR are shown in Table A1. Relative expression levels were determined with 2-ΔΔCT method after primers were validated through a standard curve of serial dilutions from pooled cDNA. One-way ANOVA with Tukey’s post-hoc test (P<0.05) was used to determine
statistical differences between groups.
Protein analysis
Protein lysates were prepared in PBS (pH7.4) containing 0.5% Triton X-100, 300mM
NaCl and Halt™ Protease Phosphatase Inhibitor cocktail (ThermoScientific; 1861280).
Tissues were homogenized using a Bullet Blender (Next Advance) and cells were sonicated
prior to clarification by centrifugation. Protein concentrations were determined using a BCA
Protein Assay Kit (ThermoScientific). Proteins were resolved by SDS-PAGE, transferred to
nitrocellulose membranes and probed with primary antibodies specific for calnexin (Santa
Cruz Biotechnology; sc-11397), BRD2 (Cell Signaling Technology; #5848), BRD3 (Santa
Cruz Biotechnology; sc-81202) or BRD4 (Bethyl Laboratories; A301-985A). Proteins were
154 detected using SuperSignal West Pico Chemiluminescent Substrate (ThermoScientific;
34080) and a FluorChem HD2 Imager (Alpha Innotech).
Histological analysis
Myocyte cross sectional area was quantified using latitudinal mid-sections of the LV treated with neuraminidase type V (Sigma) and stained with fluorescein labeled peanut agglutinin
(10mg/ml; Vector Laboratories). Images were captured with an AxioVert 200 inverted microscope using an AxioCam MRc digital camera, and analyzed with AxioVision Release
4.8.1 imaging software (Zeiss, Germany). Approximately 100-125 myocytes were analyzed and averaged for each animal. Analysis focused on the epicardium and endocardium, where the best cross sections of myocytes were present. Data was analyzed using GraphPad Prism software, with a one-way ANOVA with Newman-Keuls post-hoc test (P<0.05) to determine statistical differences between groups.
For fibrosis analysis, sectioned tissue was rehydrated and collagen was stained using picrosirus red dye (Chromaview, Richard-Allen Scientific). All histological analyses were carried out in a blinded manner using an Axiovert 200 inverted microscope with a digital camera equipped with AxioVision imaging software (Zeiss, Germany). Quantification of picrosirius red staining was completed by determining the average stained pixels2 per total pixels2 in images of the LV.
Indirect immunofluorescence
NRVMs were grown on gelatin-coated glass coverslips and fixed with 3.7% paraformaldehyde in PBS. Cells were permeabilized with NP-40 (0.1%), blocked with 1 %
BSA Fraction V (Fisher) and stained with antibodies against α-actinin (Sigma) and ANF
(Phoenix Pharmaceuticals). Coverslips were incubated with fluorescein-conjugated anti-
155 mouse and Cy3-conjugated anti-rabbit antibodies (Jackson ImmunoResearch Laboratory) and
mounted on slides using SlowFade® Gold antifade reagent containing DAPI (Invitrogen).
Images were captured using a Nikon Eclipse Ti inverted microscope with an Andor
Technology digital camera and NIS Element software.
Operetta high content imaging
Fixed and stained cells on 96-well, clear-bottomed plates (Greiner) were imaged on
an automated fluorescence microscopy system (Operetta, Perkin Elmer) using a 20X
objective. Thirty fields were imaged in each well of the 96-well plates. Three channels were
collected for each field, corresponding to fluorescein (a-actinin), Cy3 (ANF), and DAPI
(nuclei). Images were analyzed using a custom algorithm in the Harmony high-content analysis software package (Perkin Elmer). Briefly, objects were initially defined using the nuclear channel, then cytoplasm was segmented using the fluorescein channel. Mean fluorescein intensity was calculated for each cell. Cells were selected using threshold values for mean fluorescein fluorescence in order to filter out residual fibroblasts or outlying bright objects. Perinuclear masks were defined for each valid cell, and total Cy3 fluorescence was calculated for each mask. Finally, cell area was calculated for each valid cell based on the fluorescein channel.
Chromatin immunoprecipitation
NRVMs were processed using the Magnify ChIP kit (Invitrogen). After 16 hours of
treatment with PE, cells were trypsinized, washed with PBS and crosslinked with 1%
formaldehyde in PBS for 10 minutes. After crosslinking, cells were washed with PBS and
lysed at a concentration of 1x106 cells per 50µl of buffer. Chromatin was sheared using a
Bioruptor UCD 200 (high setting, 16 cycles, 30 seconds on and 30 seconds off). Sheared and
156 cleared chromatin (4x105 cell equivalents) was used for each immunoprecipitation.
Antibodies recognizing BRD4 (Bethyl, A301-985A) or phospho-Ser-2 of RNA polymerase II
(Covance, H5) were used for immunoprecipitation. Normal IgG (Invitrogen) was used as a negative control. The presence of ANF TSS sequence in purified and eluted DNA was quantified using DyNAmo Flash SYBR Green qPCR Kit (ThermoScientific) on a StepOne qPCR Instrument (Applied Biosystems). Primer sequences are shown in Table A1. Results were normalized to sheared and purified input DNA.
Results
To facilitate discovery of novel therapeutics for heart failure, we developed a high throughput assay of cardiomyocyte hypertrophy. The assay is based on the use of primary neonatal rat ventricular myocytes (NRVMs) and a high content imaging platform that enables simultaneous quantification of cell size and expression of atrial natriuretic factor (ANF) protein as a biomarker of pathological cardiac hypertrophy (Figure A-1A). Screening of small molecule libraries yielded several ‘hit’ compounds, which will be described in detail elsewhere. JQ1, a compound that blocks binding of bromodomain and extraterminal (BET) proteins to acetyl-lysine (Filippakopoulos et al., 2010), was identified as a potent inhibitor of
NRVM hypertrophy. JQ1 dose-dependently suppressed cardiac hypertrophy mediated by phenylephrine (PE), which stimulates the α1-adrenergic receptor, and phorbol-12-myristate-
13-acetate (PMA), which stimulates hypertrophy intracellularly by activating protein kinase
C (PKC) (Figure A-1B and C). JQ1 was well tolerated by NRVMs; representative images of cells exposed to PE and PMA in the absence or presence of JQ1 are shown (Figure A-1D).
Pathological cardiac hypertrophy is associated with a prototypical gene program that includes induction of ANF and brain natriuretic peptide mRNA expression, downregulation
157
Figure A-1: A small molecule BET protein inhibitor blocks cardiac hypertrophy. (A) A high throughput assay of cardiomyocyte hypertrophy based on culture of primary neonatal rat ventricular myocytes (NRVMs) on 96-well plates and high content imaging to quantify cell size and expression of atrial natriuretic factor (ANF) protein as a biomarker of pathological cardiac hypertrophy. (B, C) A hit compound, JQ1, dose-dependently blocked hypertrophy in response to treatment with phenylephrine (PE; 10 µM) or phorbol myristate acetate (PMA; 50 nM). (D) Images of ANF and a-actinin staining in NRVMs treated with agonists in the absence or presence of JQ1 (1 µM). Scale bar = 10 µm.
158 of sarcoendoplasmic reticulum calcium ATPase (SERCA2a), and β- and α-myosin heavy chain isoform switching. JQ1 completely blocked induction of this gene program in PE- treated NRVMs (Figure A-2A-E).
There are 47 bromodomain-containing proteins, and JQ1 has been shown to selectively inhibit binding of BET family bromodomains to acetyl-lysine (Filippakopoulos et al., 2010). Quantitative PCR and immunoblotting was performed to determine whether BET family members (BRD2, BRD3, BRD4 and BRDT) are expressed in the heart. Messenger
RNA transcripts for BRD2, BRD3 and BRD4 were detected in isolated neonatal and adult rat cardiac myocytes and fibroblasts; BRDT mRNA was downregulated in adult cells (Figure A-
3A). BRD2, BRD3 and BRD4 protein was present in homogenates from cultured NRVMs, and BRD4 protein expression was upregulated following PE treatment (Figure A-3B).
Given the increased expression of BRD4 in response to a hypertrophic signal, this
BET family member was chosen for further investigation. BRD4 has previously been shown to recruit the positive transcriptional elongation factor b (P-TEFb) complex to transcriptional start sites (TSSs) (Moon et al., 2005). P-TEFb is a complex containing cyclin-dependent kinase 9 (CDK9) and cyclin T, which is able to phosphorylate the carboxy-terminal domain
(CTD) of RNA polymerase II (RNA pol II) and thereby stimulate transcription elongation.
P-TEFb and its negative regulators, 7SK RNA and Hexim/CLP-1, have all been shown to regulate cardiac hypertrophy (Espinoza-Derout et al., 2007, 2009; Sano and Schneider, 2004;
Sano et al., 2002). Using the ANF promoter as a prototype, we tested the hypothesis that
BRD4 is recruited to target genes and regulates RNA pol II phosphorylation in response to hypertrophic stimuli. Cultured NRVMs were stimulated with PE for 16 hours and harvested for chromatin immunoprecipitation (ChIP) analysis using antibodies directed against BRD4
159
A B C
D E
Figure A-2: JQ1 blocks induction of fetal gene program in phenylephrine treated neonatal rat ventricular myocytes. NRVMs were stimulated for 48 hrs with PE in the absence or presence of JQ1 and quantitative PCR was performed to assess expression of (A) ANF, (B) brain natriuretic peptide (BNP), (C) sarcoendoplasmic reticulum Ca2+ ATPase 2a (SERCA2a), (D) β-myosin heavy chain (β-MyHC) and (E) α-myosin heavy chain (α-MyHC). N = 3 plates of cells per condition; *P<0.05 vs. unstimulated controls.
160
A
B
Figure A-3: BET family member mRNA expression in purified cardiac myocytes and fibroblasts. (A) Quantitative PCR was performed to assess expression of BET family members in purified neonatal rat ventricular myocytes (NRVMs), neonatal rat ventricular fibroblasts (NRVFs), adult rat ventricular myocytes (ARVMs) and adult rat ventricular fibroblasts (ARVFs). BRDT transcripts were not detected (ND) in adult myocytes or fibroblasts. (B) Immunoblot analysis of BET proteins in NRVMs treated with PE for 48 hrs. Calnexin is a loading control.
161 or phosphorylated RNA pol II (Figure A-4A). PE treatment led to increased abundance of
BRD4 at the ANF TSS (Figure A-4B), and this correlated with enhanced phosphorylation of
the CTD of RNA pol II (Figure A-4C). Together, these data support the notion that BET
proteins integrate upstream signals with the transcriptional program for cardiac growth.
To address the role of BET proteins in cardiac remodeling in vivo, mice were
subjected to transverse aortic constriction (TAC) and administered JQ1 or vehicle control
every other day for four weeks. Consistent with results obtained with cultured cardiac
myocytes (Figure A-3B), pressure overload stimulated BRD4 protein expression in the LV
(Figure A-5A). TAC promoted LV systolic dysfunction, which was rescued by JQ1, as
determined by echocardiographic analysis (Figure A-5B and C). Echocardiography also
revealed that JQ1 normalized LV and septal wall thickness in mice subjected to TAC,
suggesting that the compound reduced cardiac hypertrophy (Figure A-5D and E). Consistent
with this, JQ1 normalized LV weight-to-tibia length ratio, myocyte cross-sectional area and expression of ANF in the LV (Figure A-5F – I). JQ1 also blunted pathological LV interstitial fibrosis in animals subjected to TAC (Figure A-5J-K).
Discussion
The findings of this study define a novel function for BET acetyl-lysine binding proteins in the control of cardiac hypertrophy. We propose a model in which BET proteins serve as nodal regulators of the transcriptional program for cardiac hypertrophy. According to this model, pro-hypertrophic stimuli stimulate gene transcription by promoting recruitment of BET proteins (e.g., BRD4) to regulatory regions of downstream target genes. Since BRD4 associates with the P-TEFb complex, recruitment of BRD4 to these regulatory elements
162
A B C
Figure A-4: Chromatin immunoprecipitation of NRVM lysates show BRD4 and Phospho-RNA Polymerase II enrichment at ANF transcription start site with PE stimulation. (A) Chromatin immunoprecipitation (ChIP) strategy. (B, C) BRD4 and phospho-RNA Pol II association with the ANF transcription start site. IP with normal IgG was used as a negative control. N = 3 plates of cells per condition (4 pooled plates per N); *P<0.05 vs. unstimulated controls.
163 J
KJ
Figure A-5: JQ1 suppresses left ventricular hypertrophy in response to pressure overload.
164 Figure A-5: JQ1 suppresses left ventricular hypertrophy in response to pressure overload. (A) Immunoblot analysis of BET proteins in LVs of mice subjected to TAC for four weeks using a 25 gauge or 27 gauge needle to establish suture diameter. Calnexin is a loading control. Mice were subjected to TAC or sham surgery and were dosed IP with JQ1 (50 mg/kg) every other day for four weeks. (B – E) M-mode echocardiographic images were used to quantify ejection fraction, LV posterior wall thickness, and interventricular septum thickness. (F) LV weight-to-tibia length ratios were determined at necropsy. (G) LV sections were stained with fluorescein-conjugated peanut agglutinin to assess myocyte cross- sectional area. Scale bar = 10µm. (H) Quantification of myocyte cross-sectional area (µm2); >100 myocytes were quantified per the indicated number of LVs. (I) Quantitative PCR analysis of ANF mRNA expression. N = 3 plates of cells per condition; *P<0.05 vs. sham or unstimulated controls. (J-K) JQ1 attenuates LV fibrosis. (J) LV sections were stained with picrosirius red dye to reveal deposition of interstitial collagen (red). Images were captured at 40X magnification; scale bar = 10 µm (K) Collagen fraction was quantified. Four weeks of TAC was insufficient to increase collagen fraction in a statistically significant manner.
165 results in CDK9-mediated phosphorylation of RNA pol II and enhanced transcriptional elongation (Figure A-6).
Of the four BET proteins, only BRD2, BRD3 and BRD4 are expressed in adult cardiac myocytes. Increased expression of BRD4 (Figures A-3B and 5A) and recruitment of
BRD4 to the ANF promoter (Figure A-4B) in response to a hypertrophic signal suggests a prominent role for this BET family member in the control of cardiac remodeling. However, the lack of induction of BRD2 and BRD3 expression during cardiac hypertrophy does not exclude the possibility that these proteins also control heart muscle growth. For example,
BRD2/3 activity, interacting partners or subcellular localization could be affected by hypertrophic stimuli.
Regulation of ANF expression by BRD4 is an example of how BET proteins can affect gene expression during cardiac hypertrophy. However, it should be noted that ANF itself does not promote hypertrophy. We propose that BET proteins are required for induction of critical pro-hypertrophic genes. BET protein-focused ChIP-sequencing should lead to the discovery of these target genes and will provide further mechanistic insight into the involvement of BET acetyl-lysine binding proteins in the regulation of cardiac growth.
BET inhibitors are in clinical development for the treatment of cancer and atherosclerosis and have shown promise in pre-clinical models of inflammation and HIV latency (Prinjha et al., 2012). Our data suggest novel potential for small molecule BET inhibitors for the treatment of pathological cardiac remodeling and heart failure.
166
Figure A-6: Model for the regulation of cardiac hypertrophy.
167