Identification and Characterization of Barrier Insulator Activity in the T-Cell Receptor Alpha Locus Control Region

Home , Insulator (genetics), Locus control region

City University of New York (CUNY) CUNY Academic Works

All Dissertations, Theses, and Capstone Projects Dissertations, Theses, and Capstone Projects

2-2019

Identification and Characterization of Barrier Insulator Activity in the T-Cell Receptor alpha Locus Control Region

Gayathri Devi Raghupathy The Graduate Center, City University of New York

How does access to this work benefit ou?y Let us know!

More information about this work at: https://academicworks.cuny.edu/gc_etds/3005 Discover additional works at: https://academicworks.cuny.edu

This work is made publicly available by the City University of New York (CUNY). Contact: [email protected]

IDENTIFICATION AND CHARACTERIZATION OF BARRIER INSULATOR ACTIVITY

IN THE T CELL RECEPTOR α LOCUS CONTROL REGION

GAYATHRI DEVI RAGHUPATHY

A dissertation submitted to the Graduate Faculty in Biology in partial fulfillment of the requirements for

the degree of Doctor of Philosophy, The City University of New York

2019

GAYATHRI DEVI RAGHUPATHY

IDENTIFICATION AND CHARACTERIZATION OF BARRIER INSULATOR ACTIVITY

IN THE T CELL RECEPTOR α LOCUS CONTROL REGION

Gayathri Devi Raghupathy

This manuscript has been read and accepted for the Graduate Faculty in Biology in satisfaction of the dissertation requirement for the degree of Doctor of Philosophy.

Date [Ben Ortiz]

Chair of Examining Committee

Date [Cathy Savage-Dunn]

Executive Officer

Supervisory Committee:

Laurel Eckhardt

Shubha Govind

Diana Bratu

Fei Li

THE CITY UNIVERSITY OF NEW YORK

iii

ABSTRACT

IDENTIFICATION AND CHARACTERIZATION OF BARRIER INSULATOR ACTIVITY

IN THE T CELL RECEPTOR α LOCUS CONTROL REGION

by Gayathri Devi Raghupathy

Advisor: Dr.Ben Ortiz

Genes of different spatiotemporal expression profiles are often juxtaposed in the genome. This organization raises risks of cross-regulatory influences from neighboring genes; for instance heterochromatin can spread over euchromatin or long-range acting enhancers can inappropriately activate genes. Gene regulatory elements such as Locus Control Regions (LCR) and Insulators prevent such cross-communications and allow for normal gene expression patterns. In transgenic systems, LCRs limit influences from surrounding chromatin by providing site-of-integration independent and specific spatiotemporal expression upon a linked transgene. The field’s understanding of the ability of an LCR to overcome chromatin influences and allow site-of- integration independent expression is minimal. Interestingly, this function of an LCR closely resembles that of barrier insulators. Barrier insulators prevent the spread of heterochromatin onto a euchromatin region and are characterized by their ability to suppress site-of-integration dependent chromatin influences upon a transgene. We hypothesize that the integration site- independence activity of LCRs is mediated by insulator-like DNA elements present within the

LCR. In support of this hypothesis, we identify a novel barrier insulator activity within the mouse T-cell receptor (TCR)-α LCR. A 4.0-kb compilation of TCRα LCR sub-elements insulates a linked transgene in barrier assay- a long-term culture of stably transfected T cell lines.

TCRα LCR-derived insulators enable maintenance of euchromatin and prevention of heterochromatin at a linked transgene. We find one element within the TCRα LCR that interacts iv with the USF1 transcription factor, which has been shown to have an important role in barrier insulation. In contrast to previously identified barrier insulators, the function of TCRα LCR- derived insulators does not require them to bi-laterally flank a gene. These data suggest that the

TCRα LCR-derived elements may support both known and novel mechanisms of barrier insulation.

v Acknowledgements

I would like to thank the CUNY Graduate Center for accepting me into the PhD program. My sincere thanks to Dr.Ben Ortiz for giving me an opportunity to work in his laboratory, where I evolved as a scientist. Thank you Ben for being available to answer my questions and bearing with me through my struggles with experiments. I would like to thank my committee members-

Dr.Eckhardt, Dr.Bratu, Dr.Li and Dr.Govind for guiding me through my research at various stages. My sincere thanks to Dr.Eckhardt, whose brilliant advice to focus on certain projects, helped me to not only save time and resources but to also learn to prioritize. A big thanks to Dr.

Li and his lab members, Dr.Qianhua Dong and Dr.Haijin He for supporting me throughout the collaboration. Dr.Li thank you for being very encouraging and appreciative of my efforts.

This project would not have existed if not for Armin Lahiji, who laid the groundwork for this research. My first day in the Ortiz lab, Armin was my teacher and since then he has been supportive and helpful through these years. I thank Martina Kucerova Levisohn for always being available (amidst her crazy schedule) to brain storm ideas and troubleshoot my experiments. I often refer to Martina’s lab notebooks for protocols, thanks to her for leaving behind meticulously written records. I also would like to thank my undergraduate students, Valia

Kostiyuk and Calvin Herman for being energetic and ready to help at all times in the middle of their classes and tests. They have helped me get through my dull days and I have learnt quite a lot from them in terms of aiming high, multi-tasking and handling pressure. I would like to thank

Zayd Daruwala, for creating a fun environment in the lab and for giving me cool cloning tips.

The best part of the last seven years in this program has been my best friend Jordana Lovett.

She has been my classmate, lab mate, friend, family, motivator and a ‘shrink’. We have had a shared experience throughout this program, since day 1 of our classes. Jordana has helped me bring out the hidden best side in me; she has unconsciously played a huge role in daring me to

vi make my dreams into reality. We both know that our journey together as friends and ‘dynamic duo’ does not end with this program; we are looking forward to doing bigger things together. I have so many more things to share and thank Jordana for; it might be way too long so I will stop here…

I would like to thank the Department of Biology at Hunter College and my friends, Livia,

Irena, Jyoti, Chong for being supportive and giving feedback on my work. Thanks to Inna and

Razib for making teaching convenient and less laborious. Thanks to Priscillia Lhmouad, Skok

Lab, NYULMC for being very helpful in optimizing in my experiments.

Thanks to my parents, who have been my emotional support system throughout this program.

While they did not get my experimental issues, their kind encouraging words, ‘there will be light at the end of the tunnel’ kept me going. My parents did not get to complete their education but they raised me to appreciate education and stay committed to it. My success of achieving this highest level of education truly belongs to them. I dedicate my PhD work to my parents,

Mr.Raghupathy and Mrs.Padma Raghupathy. Thanks to my brother, Hari for trying to understand what was wrong with my experiments and even attempting to fix it! Thanks to my in- law family for being very understanding and supportive during my research.

Last but not the least, I would like to thank my husband, Kishan without whose unconditional love and support, I would have been very stuck. Kishan, you have played a tremendous role in enabling me to get through every struggle that I had through this journey. You have supported me in small and big ways. Thanks for always believing in me and pushing me to stay positive and be my best. This would not have been possible without you! There is a very special person;

I would like to thank-my Anya baby. While I feared that my pregnancy would hinder me in wrapping-up, it indeed has powered me up with positivity and energy, Thank You Biology! .

vii Table of Contents

CHAPTER 1: INTRODUCTION 1

1.1 LOCUS CONTROL REGION 1

1.2 BARRIER INSULATORS 5

1.3 TCRα/DAD1 LOCUS 8

CHAPTER 2: MATERIALS AND METHODS 13

2.1 CHROMATIN IMMUNOPRECIPITATION 13

2.2 DNASE I SENSITIVITY-QUANTITATIVE PCR 15

2.3 SOUTHERN BLOTTING TO DETECT TANDEM INTEGRATION OF TRANSGENES 16

2.4 POLYMERASE CHAIN REACTION TO DETECT TANDEM INTEGRATION OF TRANSGENES 16

2.5 T CELL LINE MAINTENANCE AND TRANSFECTION 17

2.6 DATA MINING METHODS BASED ON CISTROME AND UCSC GENOME BROWSER 18

2.7 YEAST INSULATOR ASSAY 18

CHAPTER 3: IDENTIFICATION AND CHARACTERIZATION OF BARRIER INSULATOR ACTIVITY IN THE TCRα LCR 21

3.1 HALLMARK EPIGENETIC EVENTS ASSOCIATED WITH BARRIER INSULATORS 21

3.2 TCRα LCR ELEMENTS PRESERVE EUCHROMATIN STATUS AND PREVENT HETEROCHROMATIN ENCROACHMENT AT THE REPORTER GENE LOCUS 22

3.3 TCRα LCR ELEMENTS MAINTAIN REPORTER GENE LOCUS ACCESSIBILITY IN CHROMATIN 25

CHAPTER 4: NOVEL FEATURE OF INSULATION BY TCRα LCR ELEMENTS 27

4.1 CLASSICAL BARRIER INSULATORS INSULATE ONLY WHEN FLANKING A TRANSGENE 27

4.2 TCRα LCR ELEMENTS INSULATE THE TRANSGENE IN A UNILATERAL MANNER 27

viii CHAPTER 5: DETERMINING THE BINDING OF TCRα LCR TO THE INSULATOR FACTOR, USF1 30

5.1 ROLE OF USF1 IN BARRIER INSULATION 30

5.2 INSULATOR FACTOR USF1 BINDS TO THE HS1’ REGION, BUT NOT THE HS4 OR HS6 REGIONS OF THE TCRα LCR 31

5.3 TESTING SHORT HAIRPIN RNAs (SHRNA) TO KNOCKDOWN USF1 33

CHAPTER 6: IDENTIFICATION OF FACTORS THAT BIND TO THE TCRα LCR35

6.1 CANDIDATE FACTORS THAT BIND TO THE TCRα LCR 35

CHAPTER 7: CHROMATIN LANDSCAPE OF THE TCRα/DAD1 LOCUS 38

7.1 CHROMATIN MARKS ASSOCIATED WITH TCRα, LCR AND DAD1 IN THYMUS AND NON-THYMUS TISSUES 38

CHAPTER 8: ASSESSING THE INSULATOR ACTIVITY OF TCRα LCR IN A FISSION YEAST MODEL 42

8.1 FISSION YEAST: TESTING A NEW MODEL TO STUDY VERTEBRATE INSULATORS 42

8.2 HS6 ELEMENT OF THE TCRα LCR IS NOT SUFFICIENT TO INSULATE A TRANSGENE IN YEAST INSULATOR ASSAY 43

CHAPTER 9: CONCLUSION 45

9.1 MODEL OF INSULATION 46

9.2 FUTURE DIRECTIONS 48

9.3 SIGNIFICANCE 49

REFERENCES 51

ix LIST OF FIGURES

Figure 1. Position Effects Of Transgene Integration. 2 Figure 2. DNase I Hypersentive Sites. 2 Figure 3. Types Of Insulators. 4 Figure 4. The Chicken β-Globin Locus. 5 Figure 5. The TCRα LCR Locus. 9 Figure 6. Barrier Assay. 11 Figure 7. The TCRα LCR Insulates Reporter Gene Expression From Silencing In Barrier Assay. 12 Figure 8. TαLCR4.0 Is Sufficient To Maintain Euchromatin Marks At The Reporter Gene Used In The Barrier Assay. 23 Figure 9. Heterochromatin Marks Do Not Accumulate At The Lcr Sub-Element-Linked Reporter Gene Locus. 24 Figure 10. TαLCR4.0 Preserves Accessibility At The Reporter Gene Locus. 26 Figure 11. Status Of Transgene Integration In TαLCR4.0. 29 Figure 12. Role Of USF1 At The Chicken β-Globin Locus. 31 Figure 13. USF1 Is Recruited To The HS1’ Region Of The TαLCR. 32 Figure 14. Shrna Knockdown Of USF1 In VL3-3M2 Cell Line. 34 Figure 15. Factors Associated With The TαLCR4.0 HS Elements. 36 Figure 16. Chromatin (Heterochromatin Marks) Architecture Of The Mouse TCRα/DAD1 Locus. 39 Figure 17. Chromatin (Euchromatin Marks) Architecture Of The Mouse TCRα/DAD1 Locus. 40 Figure 18. Testing Fission Yeast Model To Study TCRα LCR. 44 Figure 19. Model Of Insulator Activity Mediated By TCRα LCR. 48

x CHAPTER 1: INTRODUCTION

Organization and compaction of the genome is a complex and incompletely understood process. The DNA helix is wrapped around core histones in several layers of hierarchical configuration to achieve a very tight organization1. The complexity of what results is a mosaic of heterochromatin and euchromatin, and juxtaposition of genes with completely different spatiotemporal expression profiles. Heterochromatin represents transcriptionally silenced regions of chromatin, marked by DNA that is less accessible to nucleases, whereas euchromatin is less condensed, accessible to nucleases and consists of transcriptionally active regions of chromatin.

The complex genome organization can cause potential inappropriate cross-regulatory interactions that can affect normal gene expression programs. For instance, heterochromatin could spread into a euchromatin region and lead to inappropriate silencing of active genes; or, long-range acting gene regulatory elements such as enhancers and silencers of one domain would be able to act on promoters located in another gene domain. In order to shield the genes from any such cross-interferences, gene regulatory elements such as Locus Control Regions (LCR) and insulators play an important role in partitioning gene expression into distinct domains2.

1.1 LOCUS CONTROL REGION

LCRs are cis-acting elements that confer upon a linked transgene the ability to overcome integration site-dependent position effects. Position effect is a phenomenon that occurs when regulatory influences from the surrounding chromatin environment subject a transgene introduced therein to variability in gene expression3 (Fig 1). The influence could be from enhancers, silencers, or heterochromatin of neighboring genes. LCRs modulate chromatin to overcome such cross-regulatory hindrances and permit accessibility of a linked gene to transcription factors in a finely regulated manner. 1

Figure 1. Position Effects of Transgene Integration. A transgene can randomly integrate into heterochromatin or euchromatin regions of the genome. Integration into heterochromatin yields little or no expression of the transgene; in contrast, a high level of expression is obtained in euchromatin regions. This phenomenon of gene expression influenced by neighboring chromatin is called position effect.

LCRs also support a copy number-dependent and predictable spatiotemporal pattern of linked gene expression. Structurally, LCRs tend to co-localize with a cluster of DNase I hypersensitive sites, which bind to specific factors and mediate different aspects of the LCR function either individually or synergistically (Fig 2).

Figure 2. DNase I Hypersentive Sites. Regions of chromatin (green circle) that are highly suspectible to degradation by DNase I enzyme are referred to as DNase I hypersensitive sites. LCRs encompass cluster of hypersensitive sites that bind to factors (colored circles under the arrow). Adaped from Wang et al.4.

2 The first LCR to be discovered was found in the β-globin locus. Two main findings led to the identification of an LCR: 1) β-globin genes along with its promoter and enhancer regions that were originally thought to comprise all necessary regulatory elements failed to show uniform or high levels of expression in transgenic mice5-7, and 2) β-thalassemia patients despite having intact globin genes did not express globin8,9. This led to the understanding that a missing regulatory region was causing such phenotypes. DNase I hypersensitivity assays led to the identification of a region upstream of β-globin genes, which was then identified as an essential cis-regulatory element called the locus control region10. The β-globin LCR regulates the developmental stage specific chromatin structure and transcription of the five-globin genes present in its locus11-13. The discovery of the β-globin LCR led to the identification of LCRs in other gene loci, some examples include, mouse T-cell receptor (TCR) α14, human CD215, human growth hormone16, human adenosine deaminase17. While LCRs have been studied in various gene loci, the mechanism of regulation exerted by an LCR on its associated gene(s) has not yet been elucidated. One avenue to achieve a greater understanding an LCR’s function lies in studying its ability to modulate chromatin and suppress position effects.

Position effect was first identified in Drosophila. Position effects might lead to various forms of disruptions of the associated gene including control by a different regulatory unit at the new location18, or silencing due to insertion into a heterochromatin domain. The ability of an LCR to suppress position effects and overcome heterochromatin structures has been studied in β-globin and hCD2 LCRs in that they protect transgenes integrated at or near the centromeric or pericentromeric regions from heterochromatin-induced position effects19,20. However, it is not yet clear how LCRs suppress position effects and modulate chromatin.

3 The ability of an LCR to overcome heterochromatin and suppress position effects resembles barrier insulators. Insulators are DNA-protein complexes that usually support one of two distinct functions, enhancer blocking or chromatin barrier (Fig 3).

Figure 3. Types of Insulators. A) Barrier insulators block the spread of heterochromatin and allow downstream gene to remain ‘on’ B) Enhancer-blocking insulators when present between an enhancer and a promoter, shields the promoter from activation. Adapted from Heger et al.21.

An enhancer-blocking insulator is a DNA element that, when placed between an enhancer and promoter, prevents the enhancer from acting on the gene transcribed from that promoter.

Chromatin barriers are characterized by their ability to suppress position effects, in that they prevent the spread of heterochromatin into a nearby euchromatin domain22. Other than their ability to suppress position effects, barrier insulators do not exhibit any other LCR-like functions such as support of copy-number dependent or specific spatiotemporal expression patterns.

Hence, it is believed that the barrier-type insulators are sub-elements of LCRs that mediate suppression of position effects. Links between barrier insulators and LCRs have not yet been established.

4 1.2 BARRIER INSULATORS

Barrier insulators have been identified in Drosophila, yeast (S. cerevisiae, S. pombe) and vertebrates. There are five well-characterized Drosophila insulators that have laid the groundwork for the study of insulators in other organisms. Below, I discuss in detail vertebrate insulators since this study focuses on barrier insulators within a mouse gene locus.

1.2.1 VERTEBRATE BARRIER INSULATORS Vertebrate barrier insulators have been identified in long-term cell culture assays that demonstrate their capacity to suppress position effects. In this assay, cells stably transfected with a reporter gene initially express but over a period of 60-90 days, they will be subject to silencing due to the encroachment of heterochromatin23. However, the reporter gene maintains its expression and remains in a euchromatin state when a barrier insulator flanks it. The first vertebrate barrier insulator called cHS4 (hypersensitive site) was identified in the chicken β- globin locus24. The β-globin genes lie in a complex neighborhood of two distinctly regulated loci. A 16kb heterochromatin domain lies 5’ to the cluster of four β-globin genes and the folate receptor gene lies upstream of the heterochromatin region25 (Fig 4).

Insulator

Figure 4. The Chicken β-Globin Locus. Vertical arrows indicate DNase I hypersensitive regions. FR refers to the folate receptor gene. Shaded box represents the cHS4 insulator. Adapted from Jin et al.26.

5 Heterochromatin and euchromatin structures have distinct repertoires of histone tail modifications that are established by histone modifying enzymes. Some examples of euchromatin marks include diacetylation of histone H3 lysine (K) 9, K14 and heterochromatin marks include methylation of H3K9, H3K27, H3K20 among others. In the chicken β-globin locus, the 16kb heterochromatin is enriched for dimethyl H3K9, whereas the active β-globin genes show elevated levels of acetylated H327. The euchromatin β-globin genes are protected from the potential spread of heterochromatin initiating from the upstream region by insulator element called cHS4.

The cHS4 insulator that is found between the transcriptionally accessible globin genes and the

16kb condensed chromatin possesses both barrier and enhancer-blocking activities24,2826,27,28. cHS4 is enriched for acetylated H3 and H4 histones27,2927,2927,29, and a host of other active chromatin modifications27,29. USF1 recruits histone-modifying enzymes to cHS4, and is essential for the barrier function30,31. Another essential factor for barrier activity called VEZF1 protects the promoter DNA-methylation mediated silencing32. Two HS- HSA and HSB present on the 5’ end of this locus also bind to USF1 and VEZF1 to protect the folate receptor gene from the downstream 16kb heterochromatin25,33. These findings have yielded a model that states that barrier elements bind to factors that recruit active histone-modifying enzymes in order to maintain a euchromatin structure and thereby prevent the spread of inactive chromatin within a locus. Research done on cHS4 led to identification of insulators in several other gene loci.

Gallagher et. al. found insulator activity in the human ankyrin-1 gene locus that encodes an erythroid membrane protein34. A 5’HS region within the promoter of this gene acts as a barrier insulator and bears the characteristics of cHS4 in that the insulator is hyperacetylated at H3 and

H4 and is bound by USF1. This barrier is hypothesized to function tissue specifically to prevent influences from nearby ankyrin promoters that drive expression in non-erythroid tissues and to protect the ankyrin locus from heterochromatin encroachment. The same group, found a third 6 example of a vertebrate insulator in the human spectrin gene, a main component of the erythrocyte plasma membrane35. A non-coding exon located within this gene acts as a barrier insulator in reporter assays, and is thought to regulate this locus in a tissue specific and developmental stage specific manner. The USF1 and USF2 enrichment seen in this exon region is thought to recruit histone-modifying enzymes to maintain euchromatin structure, similar to cHS4. The effect of loss of USF1 on insulation, or role of other factors at both of these insulators, has not been studied.

Recently, a barrier insulator was identified in the mouse TCRβ locus36. During T cell development, the variable (V), diversity (D) and joining (J) segments of the α and β gene loci undergo V(D)J recombination to generate diversity in antigen recognition. In the TCRβ locus, the Vβ segments are separated from the downstream Dβ, Jβ and Cβ segments by a set of trypsinogen genes. The neighboring trypsinogen genes are not expressed and exist in a heterochromatin state, while the V, D, J segments undergo active chromatin modulation, transcription and recombination during T cell development37. A barrier element that exists at the border of the last tyrpsinogen gene and the first Dβ segment prevents the 5’ spread of H3K9me2 heterochromatin mark from the trypsinogen genes to the 3’ Dβ segments. This barrier has the hallmark feature of elevated levels of euchromatin marks (H3ac, H3K4me2) further making a case for insulation mediated by the recruitment of active histone modifying enzymes. However not much is known about its binding partners and the mechanism of insulation.

The study of chicken β-globin insulator has provided some insights into the mechanism of insulation. While a few other vertebrate insulators have been studied, it is not yet clear if the mode of insulation, and factors involved in the chicken β-globin locus, represent a universal mechanism of action. While it is speculated that barrier insulators are sub-elements of LCRs, no study has addressed this hypothesis until now. Recently, our lab that studies the mouse TCRα

LCR used long-term cell culture based assays to demonstrate that the LCR exhibits a suppression 7 of position effects function similar to that of barrier insulators (Armin Lahiji, Doctoral thesis,

The Graduate Center, CUNY, 2013). We further explored this finding to identify characteristics features of barrier insulators.

1.3 TCRα/DAD1 LOCUS

Our lab studies a locus in the mouse chromosome 14 that harbors two differentially expressed genes: TCRα and Defender Against Death (DAD) 1. TCRs are cell surface proteins that mediate antigen recognition in T cells and play an important role in the cell-mediated response of the immune system. TCRs are important for proper development and differentiation of early T cells.

T cells express one of two types of TCRs, αβ or γδ with most circulating T cells expressing the

αβ TCR. The heterodimers αβ and γδ are encoded by TCR -α,-β,-γ and -δ genes respectively.

DAD1 is an anti-apoptotic gene that plays a role in N-linked glycosylation38,39. There exist major differences between the TCRα and DAD1 gene in terms of their developmental timing and tissue specificity of their expression patterns40. With respect to timing, TCRα is activated at a much later stage of embryogenesis than DAD140. TCRα is expressed only in a subset of T cells, whereas DAD1 is ubiquitously expressed40. Regulatory mechanisms that compartmentalize the transcriptional domains of these two genes seem critical to allow for their distinct patterns of expression and to prevent inappropriate cross-regulatory influences between them. Although the regulatory mechanisms involved are still not clear, the activity of the TCRα LCR that is located between TCRα and DAD1 is generally invoked to help explain the differential regulation of these genes14.

8 1.4 TCRα LCR

The TCRα LCR exerts full features of an LCR including site of integration independent, spatiotemporally specific and copy number dependent expression of a linked reporter gene. The

13-kb TCRα LCR contains nine DNase I hypersensitive sites14,40 (Fig 5).

Figure 5. The TCRα LCR Locus. Vertical arrows indicate DNase I hypersensitive regions of TCRα LCR. The open box depicts the classical enhancer (Eα, HS1) of TCRα. Solid boxes mark the exons of TCRα and DAD1 genes.

Of the nine HS, only HS1, HS1’, HS4 and HS6 have thus far been shown to contribute to different aspects of the LCR’s functions41-43. The deletion of two HS regions, HS7 and HS8 that lie upstream of HS1 has negligible effects on the TCRα-LCR activity41. The role of HS2, HS3 and HS5 in the LCR activity has not yet been identified.

HS1 DNA includes a T cell specific, classical enhancer (Eα) that is required for full TCRα expression44. The HS1’ region appears to be critical to the tissue specific expression of LCR- linked reporter gene and endogenous TCRα gene function41,45. The regions that are critical for position effect suppression and high-level transgene expression include both HS4 and HS6. The deletion of HS4 or HS6 from a TCRα LCR-driven transgene results in a reduction in transgene mRNA levels, along with other indicators of poor chromatin accessibility42,43. Of all the HS in

9 the TCRα LCR, it is highly possible that the HS4 and/or HS6 regions bear barrier insulator elements that enable it to suppress position effects.

The TCRα LCR that exhibits all features associated with LCR activity in transgenic mouse and in vitro stem cell differentiation models shows only partial LCR activity when introduced de novo into differentiated T cell lines (may be due to lack of development signals originating during differentiation)46. The partial LCR activity refers to the inability of the LCR to deliver copy number dependent expression of the linked transgene46. However, it does provide some degree of site of integration independence to gene expression. It was thought that the observed partial LCR function might represent barrier-like insulator activity. Hence, in order to formally test the insulation capacity of the TCRα LCR, a long-term, cell culture-based, barrier assay was setup.

Previously mentioned insulator elements including chicken β-globin, human ankyrin and spectrin genes34,35 were studied by utilizing long-term cell culture based barrier assay. Barrier assays are designed to investigate the ability of an element to overcome position effects and successfully insulate a randomly integrated reporter gene from heterochromatin encroachment.

In this assay, a cell line of choice is stably transfected with a construct bearing an insulator element along with a reporter gene and the expression of the reporter gene expression is then monitored over 90 days. This assay is conducted for long periods of time since the establishment of a stable inheritable state of un-insulated reporter gene silencing is dependent on its site of integration. In the absence of an insulator, the reporter gene expression is variable and extinguishes in less than 90 days due to heterochromatin encroachment, whereas in the presence of an insulator the expression of the reporter is maintained.

A similar assay was used to test the ability of TCRα LCR to act as a barrier insulator (Armin

Lahiji, Doctoral thesis, The Graduate Center, CUNY, 2013) (Fig 6). In this assay, a mouse T cell line (VL3-3M2) was stably transfected with constructs bearing either the full HS1-HS6 regions 10 (TαLCR7.4), or HS1, HS1’, HS4 and HS6 regions (TαLCR4.0) or HS1 only (Eα) (Fig 7a). Each clone was then maintained in duplicate cultures, one with and one without drug (selection marker-neomycin) selection for 90 days and reporter gene expression (YFP) was monitored by flow cytometry. Removal of drug selection allows for the detection of reporter gene silencing and maintaining another set in drug serves as a ‘day 0’ or non-silencing control

Figure 6. Barrier Assay. VL3 cells were stably transfected with constructs bearing a drug selection marker (NEOR), reporter gene (YFP) and different HS of the LCR. Clones were maintained in the presence and absence of neomycin, and YFP expression was monitored by flow cytometry for 90 days. If YFP is not insulated (left), there might exist two types of cell populations- one that is silenced due to heterochromatin (coil) and hence not resistant (✓) to neomycin, another that is not subject to silencing and resistant to neomycin (✗) Silencing of YFP can be detected by flow cytometry only in the presence of neomycin. If insulated from heterochromatin encroachment, YFP expression remains the same in the presence or absence of neomycin.

After 90 days, the assay showed that the TαLCR7.0 and TαLCR4.0 clones successfully maintained the expression of the reporter gene and prevented it from silencing. A construct bearing only the HS1 enhancer region (Eα-only) that served as a negative control showed 11 significant loss of reporter gene expression. Moreover, a 4kb compilation of the LCR consisting of HS1, HS1’, HS4 and HS6 was sufficient for this function (Fig 7b).

A )

B )

Figure 7. The TCRα LCR Insulates Reporter Gene Expression from Silencing in Barrier Assay. A) Constructs tested in the barrier assays. The sub-elements of the LCR included in each of the constructs (Eα-only, TαLCR7.4 and TαLCR4.0 constructs) are indicated. An SV40 promoter-driven neomycin resistance gene upstream of the Vα17 promoter-driven YFP reporter gene enabled selection for stable transfectants B) The TCRα LCR insulates reporter gene expression from silencing in barrier assay. VL3-3M2 cells were stably transfected with TαLCR7.4, TαLCR4.0 or Eα-only and clones were generated. Graph of flow cytometry analyses of average %YFP level maintenance (over time) in multiple VL3-3M2 T cell clones bearing the indicated reporter construct. “Day” is the number of days that cells are cultured in the absence of drug selection. YFP maintenance is the ratio of mean fluorescence intensity (MFI) in ‘no drug’ to that of ‘with drug’ selection. MFI ratios for a specific ‘day’ in clones within each construct were averaged. The number of clones (n) assessed for each construct is indicated. p-values are calculated using a 2-tailed student test.

12 The results of this assay led to the central hypothesis of this thesis: the TCRα LCR acts as a barrier insulator element to maintain the reporter gene in euchromatin state and prevent it from silencing. We propose to identify the chromatin state of the reporter gene linked to LCR elements, since the hallmark feature of a barrier insulator lies in its ability to modulate and establish euchromatin domain. Overall, identifying barrier insulator elements within an LCR will enable understanding of mechanism of action of LCRs. Investigation of insulator activity exerted by TCRα LCR will also enable understanding regulation and chromatin organization at other multi gene loci. Moreover, a novel insulator in the mouse TCRα locus can be eventually tested for translational applications in the field of gene therapy to overcome integration site-based silencing of therapeutic genes.

CHAPTER 2: MATERIALS AND METHODS

2.1 CHROMATIN IMMUNOPRECIPITATION

Chromatin immunoprecipitation (ChIP) protocol was done as previously described in47. 5x107

VL3-3M2 cells were treated with formaldehyde at a final concentration of 1% for 10 min at 37oC in order to crosslink protein-DNA complexes. This was followed by quenching the fixation with

125 mM glycine for 5 min at room temperature. Cells were harvested and washed three times with 1X phosphate buffered saline (PBS). Chromatin was extracted by re-suspending cell pellets in 2 ml of 1X RIPA buffer containing 5 µl of protease inhibitor cocktail (Sigma) and 1 µl of

100mM PMSF (Sigma). Extracted chromatin was subject to sonication for 25 cycles of 1 min on/1 min off at 4oC to obtain fragments ranging from 100-600 bp. The sonicated chromatin sample was centrifuged at 13.5x10000g for 20 min at 4oC, to pellet cell debris.

13 400 µg of chromatin was immunoprecipitated with antibody overnight at 4oC. To the incubated chromatin-antibody complex, 50 µl of 25% slurry of protein A/G plus agarose beads

(Santa Cruz) (beads were pre-cleared with 0.3 mg/ml of salmon sperm DNA for 30 min at 4oC) was added and rotated for 2h at 4oC. Following that, beads were washed 1X in three different conditions using wash buffer 1 (0.1% SDS, 1% Triton X, 20 mM Tris pH 8.1, 150 mM NaCl), wash buffer 2 (0.1% SDS, 1% Triton X, 20 mM Tris pH 8.1, 500 mM NaCl), wash buffer 3

(0.25M LiCl, 1% IGEPAL, 1% Deoxycholate, 1 mM EDTA, 10 mM Tris pH 8.1) and two final washes with Tris-EDTA pH 8 buffer. Elution was done in 0.1M NaHCO3 and 1% SDS supplemented with 1mg/ml of proteinase K (Fisher scientific) overnight at 65oC. DNA extraction was done using a PCR purification kit (Qiagen) according to manufacturer’s protocol. 10% of the DNA isolated from the sonicated chromatin lysates was used as DNA input controls in subsequent quantitative PCR (qPCR) analyses.

To prepare chromatin for ChIP from freshly isolated adult mouse thymocytes

(macrodissection), 10x106 cells were cross-linked with formaldehyde and neutralized with glycine as described earlier. Chromatin was extracted by re-suspending cell pellets in 0.35 ml of lysis buffer (10 mM Tris- HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 0.1% Na-

Deoxycholate, 0.5% N-lauroylsarcosine) containing 5 µl of protease inhibitor cocktail (Sigma).

Extracted chromatin was subjected to sonication (Bioruptor, Diagenode) for 10 cycles of 30 sec on/30 sec off at 4oC to obtain fragments ranging from 100-600 bp. The sonicated chromatin sample was supplemented with 1% Triton X-100 and 0.1% SDS followed by centrifugation at

13,500g for 5 min at 4oC, to pellet cell debris. Protein A beads (Dynabeads, Invitrogen) were incubated with rabbit polyclonal anti-USF (5 µg, ab180717) or anti-IgG (ab37415) antibodies for one hour at room temperature. Sonicated chromatin was then incubated with the antibody-bound beads for 4 hours at 4oC followed by two washes with buffer 1 (20 mM Tris-HCl/pH 7.4,

150 mM NaCl, 0.1% SDS, 1% Triton X-100, 2 mM EDTA) and buffer 2 (250 mM LiCl, 14 1% Triton X-100, 0.7% Deoxycholate, and 10 mM Tris -HCl, 1 mM EDTA). Elution and reverse crosslinking was performed in 0.5% SDS, 300 mM NaCl, 5 mM EDTA, 10 mM Tris HCl pH 8.0 and treated with proteinase K at 55oC for 1 hour followed by overnight incubation at 65oC. DNA extraction was carried out as described earlier, and analysis by qPCR was performed as described below. Primers are listed in table 1.

qPCR was performed using the Dynamo SYBR Green qPCR kit (Thermo Fisher F410L) in an

Applied Biosystems Viia7 Real-Time PCR system. ChIP DNA was used as template in qPCR reactions and the thermocycling parameters were set following the manufacturer's protocol

(annealing temperature: 60oC). Samples were analyzed in triplicates and quantification was conducted using the comparative Ct (threshold cycle) method. p-values were calculated by performing two tailed student t-test.

2.2 DNASE I SENSITIVITY-QUANTITATIVE PCR

Nuclei were prepared as previously described 42. Briefly, 1.5 x 107 cells were harvested, washed and re-suspended in hypotonic buffer A (10 mM HEPES pH 7.6, 15 mM KCl, 2 mM MgCl2 and

0.1 mM EDTA pH 8.0). Cells were then lysed in 0.5% IGEPAL added to buffer A. Nuclei were collected and re-suspended in digestion buffer (60 mM KCl, 15 mM NaCl, 15 mM Tris pH 7.5,

5 mM MgCl2, 300 mM sucrose and 5% glycerol). Buffer A and digestion buffer were supplemented with 5 µl of protease inhibitor cocktail (Sigma) and 1 µl of 100mM PMSF

(Sigma). 2.5 x 106 cells were used for each titration point. Titration was performed with DNase I

(Worthington), with the concentration ranging from 0.5-2.0 units. Reactions were carried out on ice for 10 min and stopped by adding stop buffer (5% SDS, 100mM EDTA). Proteinase K was added to the sample and incubated overnight at 55oC. This was followed by phenol-chloroform extraction and ethanol precipitation of the DNA. This was followed by qPCR assay and analysis

15 as previously described in48. Briefly, DNA concentration was measured (Nanodrop) and 50 ng was used as DNA template in qPCR experiments (as described above in the qPCR section).

Primers are listed in table 1. The DNase I insensitive gene, neurofilament (Nf)-M was used as an internal control. Standard curves were generated for each sample using 50 ng, 5 ng and 0.5 ng of

DNA template. The fraction of intact copies of the target DNA region remaining was calculated using the standard curve and graphed.

2.3 SOUTHERN BLOTTING TO DETECT TANDEM INTEGRATION OF TRANSGENES

DNA from TαLCR4.0 clones was isolated by phenol-choloroform extraction method. 15µg of isolated DNA was digested with BglII enzyme overnight followed by ethanol precipitation.

Digested DNA was then separated on an agarose gel, and transferred to nitrocellulose membrane overnight. An 827bp BglII fragment liberated from pLCRc14 was labeled with [α-32P] dCTP

(RadPrime DNA Labeling System, Life Technologies). This labeled DNA fragment served as the hybridizing probe for the southern blot.

2.4 POLYMERASE CHAIN REACTION TO DETECT TANDEM INTEGRATION OF TRANSGENES

To detect the head to tail integrants in TαLCR4.0 and TαLCR7.4, a forward primer at the 3’ end of the LCR and a reverse primer in the 5’ end of neomycin gene were designed. For the tail-tail integrants, forward primer in the 3’ end of neomycin and reverse primer in the 5’ end of SV40 promoter was designed. As a control to indicate that the clones are integrated with the plasmid, primers to the YFP region were also included. Primers are listed in table 1. PCR reaction was conducted with high fidelity DNA polymerase from NEB.

16 2.5 T CELL LINE MAINTENANCE AND TRANSFECTION

T cell cultures were maintained as previously described 46. Briefly, mouse VL3-3M2 T cell line49 was cultured in RPMI 1640 (Gibco) supplemented with 5% FBS (Hyclone), 1% Penicillin-

Streptomycin (Cellgro), 1% Glutagro (Cellgro) and 54µM β-mercaptoethanol (Sigma).

Transfection was accomplished via electroporation (Biorad) according to manufacturer’s protocol at 950µF and 0.3kV. 20µg of lentiviral shRNA vectors for USF1 along with a non- targeting control shRNA vector were linearized and transfected (TRCN0000071893,

TRCN0000071894, TRCN0000071896, TRCN0000071897 and control from GE Dharmacon).

24h post-transfection, stable transfectants were selected by using puromycin at a concentration of

1.5µg/ml for 10days.

Following the generation of stable transfectants, RT-qPCR and Western blots were conducted.

RNA was harvested from cells according to manufacturer’s protocol (Qiagen, RNA kit). 0.5µg of RNA was subject to reverse transcription reaction according to manufacturer’s protocol

(Qiagen-RT kit). RT-qPCR was done according to the protocol mentioned in the earlier sections.

Primers are listed in table 1. mRNA levels calculated by delta-delta Ct method were normalized to actin levels and expressed as fold change compared to control vector.

For the Western blot analysis, cells were harvested and protein samples were prepared in

‘magic SDS buffer’ (Skok lab, NYU). Samples were boiled for 5min at 95C and separated on a

SDS-PAGE followed by transfer onto a nitrocellulose membrane. Membrane was blocked in 5% milk solution for 2hours at RT, followed by incubation with primary antibody (USF1, Abcam ab180717) for 2hours at RT. Membrane was washed in TBS with 0.1% tween followed by incubation with anti-rabbit IgG- HRP (Fisher 18-8816-33) for 2hours at RT. Signal was detected using an enhanced chemiluminescent kit (Fisher PI34080).

17 2.6 DATA MINING METHODS BASED ON CISTROME AND UCSC GENOME BROWSER

Published ChIP-seq data sets that were submitted on UCSC genome browser

(https://genome.ucsc.edu/) or Cistrome (http://www.cistrome.org/) database were accessed.

Datasets were viewed on mm8 or mm10 UCSC genome browser or downloaded from cistrome website and viewed either on UCSC browser or IGB viewer. Accession numbers of p300,

STAT3, FoxO1 and IRF4 are GSM994520, GSM580756, GSM1141666 and GSM1309511 respectively. Accession numbers for G1E and megakaryocytes datasets are GSM946531 and

GSM946523, respectively. Accession numbers for H3K4me3, H3K9ac, H3K27ac and H3ac datasets are GSM580758, GSM773490, GSM1309521 and GSM580757, respectively.

2.7 YEAST INSULATOR ASSAY

To generate pL5-URA and pL5-HS6-URA, the following steps were done. L5 was obtained from S. Pombe genomic DNA using primers that specifically amplified a ClaI-SpeI fragment in the dgIII repeat region in centromere 350. The primers also incorporated sites for restriction enzymes SacII and EagI (forward: 5’-aatgatccgcggtactcccaactgc-3’ and reverse:

5’-gtaaatcggcggaccagagttgcc-3’). The PCR product of L5 was ligated into pBSK cut with SacII and EagI. Following that, URA extracted by digesting pREP2-URA plasmid (Li Lab, NYU) with

NotI-Xba was ligated into pBSK-L5 plasmid. Originally ade6 was obtained by PCR amplification from yeast gDNA and inserted into BamhI-SalI site of pBSK-L5-URA plasmid, however after identifying that the ade6 had a single point mutation from that yeast strain, a replacement strategy was designed to insert wild type form of ade6. The mutant version of ade6 in pBSK-L5-Ura plasmid was liberated by digestion with SalI followed by blunting of the plasmid and then digesting with NotI. A BamhI fragment of wild type ade6 obtained from pBlue- ade6-cnp1 plasmid (Li Lab, NYU) was blunted and digested with NotI. This fragment obtained

18 from pBlue-ade6-cnp1 was then ligated into the blunt-NotI fragment of pBSK-L5-URA generating pBSK-L5-ura-ade6. HS6 of the TCRα LCR was obtained by amplifying pLCRc with primers that incorporated EagI and NotI. Primers are listed in table 1. pBSK-L5-HS6-Ura was obtained by inserting the amplified HS6 into the EagI site of the pBSK-L5-Ura plasmid.

Plasmids were linearized with BaeI before transformation.

S.Pombe was grown overnight at 30°C in PMG media with 225 mg/l supplements (ade, leu, his, lys, ura). Cells grown to a density of 10x106/ml was harvested by spinning at 3000rpm for

5min followed by a wash in ice-cold water. A second wash was performed in ice-cold 1M sorbitol. In order to increase electrocompetency, the washed yeast cells were incubated in DTT buffer (0.6M sorbitol, 20mM HEPES ph7.5 and 25mM DTT) for 15min. Following incubation, cells were washed three times in 1M sorbitol and finally resuspended in ice-cold 1M sorbitol at a density of 1x109cells/ml. 100ng of linearized plasmids were added to 40µl of cell suspension and incubated on ice for 5min. 0.9ml of 1M sorbitol was added to the incubated cells that was then transferred into pre-chilled cuvettes. Electroporation was done in a Biorad apparatus (1.5kV,

200 ohms and 25µF). Cells were then plated onto ade6- selective media plates and incubated at

32°C for 4-6days. Transformants were then selected, serially diluted and plated on the following plates: PMG, PMG-ura, FOA, and PMG-ade, and incubated for 4-6days at 30°C. Yeast strain that was mutant for ade6 and ura (ade6-ura-) and another ade6+ ura+ strain were used as controls to ensure growth on respective minimal media plates.

19 Table 1: Primer List

Forward Primer Reverse Primer USF1 ChIP HS1’ 5’-catcctctggaaagaggagtta-3’ 5’-tcttttctgcacctgtggtt-3’ HS1’ non-canonical 5’-gtctctcagggtcctaggaagt-3’ 5’-gagaaagccttttggtggagta-3’ HS4 5’-aaggggtttttactctctgagc-3’ 5’-gcgtcactaggtgccttgtat-3’ HS6 5’-gagactgatttgatccagtgtg-3’ 5’-gaagctgattaccaaatgaccc RORγT 5’-ccggttgtaccacactggtt-3’ 5’-gctcgaaatcccctctcctg-3’ β-globin 5’- ggacaggtcttcagcctcttga-3’ 5’-cagatgcttgtgatagctgcct-3’ Heterochromatin and euchromatin ChIP VαYFP-1 5’-aaatcctgtcacttcagctagc-3’ 5’-gaccgggagatgtattcaggaa-3’ VαYFP-2 5’-gccttgtccacagggagatt-3’ 5’-cgggtcttgtagttgccgt-3’ GAPDH 5’-aggtgaaaatcgcggagtg-3’ 5’-agcatccctagacccgtaca-3’ β-globin 5’-ggacaggtcttcagcctcttga-3’ 5’-cagatgcttgtgatagctgcct-3’ DNaseI sensitivity assay YFP 5’-atggtgagc aagggcgaggag-3’ 5’-catgccgagagtgatcccggc-3’ Neurofilament 5’-gctgggtgatgcttacgacc-3’ 5’-gcggcatttgaaccactctt-3’ Tandem integration of transgenes Head to tail-1 5’-gtaatcagcttcaggggaacac-3’ 5’-caagctcttcagcaatatcacg-3’ Head to tail-2 5’-caacagtggccagtaccactaa-3’ 5’-ctccttccgtgtttcagttagc-3’ Tail to tail-1 5’-gaggctaactgaaacacggaag-3’ 5’-gaaacgatcctcatcctgtctc-3’ Tail to tail-2 5’-catacctcgctctgctaatcct-3’ 5’-cagtcatagccgaatagcctct-3’ USF1 shRNA transfection Actin 5’-gagcacagcttctttgcagct-3’ 5’-agcctggatggctacgtacat-3’ USF1 5’-gttgttaccacccagggctc-3’ 5’-cccacccttattccccgaag-3’ Yeast plasmid cloning pL5 with SacII and EagI 5’-aatgatccgcggtactcccaactgc-3’ 5’-gtaaatcggcggaccagagttgcc-3’ pBSK-L5-HS6-Ura with 5’-gaaaacggccgatgtcaattgaaaaagt-3’ 5’-aattgcggccgcagctctgtgtctctca-3’ EagI and NoI

CHAPTER 3: IDENTIFICATION AND CHARACTERIZATION OF BARRIER INSULATOR ACTIVITY IN THE TCRα LCR

3.1 HALLMARK EPIGENETIC EVENTS ASSOCIATED WITH BARRIER INSULATORS

In order to understand how the TCRα LCR maintains reporter gene expression in aforementioned barrier assay, it is essential to examine the hallmark molecular events associated with the function of barrier elements. The proposed model of barrier insulator function states that insulators recruit histone-modifying enzymes to maintain the active chromatin state and overcome heterochromatin mediated silencing of an associated gene. Mutskov et al, studied the temporal order of events associated with the silencing of a reporter gene stably integrated into a cell line over a period of 100 days51. They show that recruitment of histone deacteylases

(HDAC) to deactetylate histone H3 and H4 is the first step in inactivating the transgene. This is then followed by recruitment of histone methylatransferases (HMT) and DNA methyltransferases (DNMT) to methylate their substrates, H3K9 and CpG respectively. Finally, a locked state of transgene inactivation is brought by the recruitment of heterochromatin protein

(HP1) and methyl CpG binding protein. Therefore, since silencing of a reporter gene is associated with loss of euchromatin and gain of heterochromatin marks, histone modifications of the reporter gene in previously mentioned Eα only (uninsulated) and TαLCR4.0 (insulated) constructs were examined.

21 3.2 TCRα LCR ELEMENTS PRESERVE EUCHROMATIN STATUS AND PREVENT HETEROCHROMATIN ENCROACHMENT AT THE REPORTER GENE LOCUS

Chromatin immunoprecipitation (ChIP) assay was optimized to examine euchromatin and heterochromatin associated histone marks on the YFP reporter gene. ChIP performed with acetyl

H3 antibody (euchromatin) showed that the YFP reporter gene in Eα-only clones had a significant loss of acetyl H3 modification after drug withdrawal compared to the clones in drug selection (Fig 8a). In contrast, TαLCR4.0 clones showed no loss of acetyl H3 on removal of drug selection (Fig 8b). Thus, the ChIP results correlate with previously obtained flow cytometry data for both Eα-only and TαLCR4.0 (Armin Lahiji, Doctoral thesis, The Graduate Center, CUNY,

2013).

With respect to heterochromatin modifications, ChIP was done with antibodies to di- methylated H3K9. The YFP reporter gene in TαLCR4.0 clones showed no enrichment of heterochromatin mark both with and without drug conditions (Fig 9a). Out of three Eα-only clones that were studied, only one Eα-only clone (H5C3, Fig 9a) showed significant enrichment of di-methylated HK9. Overall, these ChIP results demonstrate that the YFP reporter gene is maintained in euchromatin status in the TαLCR4.0 clones. Thus, the TCRα LCR exhibits the characteristic feature of a barrier insulator in preventing silencing of a reporter gene at the epigenetic level.

Figure 8. TαLCR4.0 is Sufficient to Maintain Euchromatin Marks at the Reporter Gene Used in the Barrier assay. ChIP assays of acetylated histone H3 at the VαYFP reporter gene in stably-transfected VL3-3M2 T cell clones in the presence (+) and absence (-) of drug selection. Withdrawal of drug enabled detection of acetylated H3 euchromatin loss in the Eα-only (A). In contrast, this mark was maintained in the presence of TαLCR4.0 (B). ChIP signals were 23 quantified by qPCR. VαYFP ChIP signals were normalized to GAPDH (internal control) and expressed as a fraction of the signal seen in the presence of drug selection.

Figure 9. Heterochromatin Marks do not Accumulate at the LCR Sub-element-linked Reporter Gene Locus. ChIP assays of di-methylated histone H3K9 at the YFP (VαYFP) reporter gene in stably-transfected VL3-3M2 T cell clones in the presence (+) and absence (-) of drug selection. The heterochromatin mark was not gained in the presence of the TαLCR4.0. In contrast, withdrawal of drug, enabled detection of heterochromatin mark di-methylated H3K9 accumulation in one out of three Eα-only clones [H5C3 (A)]. ChIP signals were quantified by qPCR. VαYFP ChIP signals were normalized to β-globin (internal control)52 and expressed as a fraction of the signal seen in the presence of drug selection.

24 3.3 TCRα LCR ELEMENTS MAINTAIN REPORTER GENE LOCUS ACCESSIBILITY IN

CHROMATIN

To corroborate ChIP results, a method employing a DNase I sensitivity assay followed by quantitative PCR (qPCR) 48 was carried out to study the chromatin accessibility at the YFP reporter locus in transfected cell clones. Figure 10 shows the fraction of copies of intact DNA remaining (as determined by qPCR) at each DNase I titration point in both TαLCR4.0 and Eα- only clones. In the TαLCR4.0 clones, the YFP locus is readily digested by increasing amount of

DNase I both in the presence and absence of drug selection, indicating its accessibility under both conditions. In contrast, the chromatin at the YFP locus in Eα-only clones is accessible in the presence of drug selection, but relatively resistant to DNase I in the absence of drug selection

(Fig 10).

The latter result is indicative of an inaccessible chromatin status of the reporter gene in the absence the TCRα LCR elements. Thus, TαLCR4.0 is able to maintain an accessible chromatin conformation at the reporter gene, even in the absence of drug selection pressure to maintain an open locus configuration.

Figure 10. TαLCR 4.0 Preserves Accessibility at the Reporter Gene Locus. DNase I sensitivity qPCR assay of Eα-only (A,B) and TαLCR 4.0 (C,D) clones post-barrier assay (dotted line represents ‘no drug’ and solid line represents ‘with drug’ conditions). qPCR was done to determine the fraction of copies remaining after DNase I titration. Y-axis shows signal from qPCR for reporter gene (YFP) normalized to a DNase I resistant heterochromatin control, neurofilament (Nf-M).

26 CHAPTER 4: NOVEL FEATURE OF INSULATION BY TCRα LCR ELEMENTS

4.1 CLASSICAL BARRIER INSULATORS INSULATE ONLY WHEN FLANKING A TRANSGENE

Reporter genes subject to long-term cell culture barrier assay are at the risk of silencing by heterochromatin encroachment from the 3’ or 5’ side. Hence, generally constructs are designed such that two or three copies of the candidate insulator element bi-laterally flank the reporter gene53. Moreover, when Pikaart et al. conducted a barrier assay with chicken cHS4 insulator linked only to the 5’ end of the reporter gene, they observed no protection from silencing (as opposed to bi-lateral flanking of the reporter gene)23. It was concluded that perhaps cHS4 insulator does not target the reporter gene to an active chromatin, but protects it from heterochromatin encroachment that can occur on either side. Following this study, flanking the reporter gene with insulators became the standard procedure for barrier assays.

4.2 TCRα LCR ELEMENTS INSULATE THE TRANSGENE IN A UNILATERAL MANNER

In our study, the constructs subject to barrier assay are designed such that the candidate insulator elements of the TCRα LCR are linked only to 3’ of the reporter gene. This design mimics the endogenous TCRα and LCR arrangement. However, plasmids tend to integrate in a tandem array fashion and that might result in the LCR flanking the reporter gene. We performed carefully designed southern blotting and PCR experiments to identify the integration status in

TαLCR7.4 and TαLCR4.0 clones (Fig 11). A Southern blot probe detecting the HS6 region at the 3’end of the LCR was designed such that it would differentiate endogenous HS6 and single copy from ‘head to tail’ tandem integrants. Primers for the PCR experiments were designed such that a forward primer was at the 3’end of the LCR and a reverse primer at the 5’ end of the neomycin gene. PCR amplification will not be obtained for single copy integrants as these 27 primers are in the opposite directions and are ~7kb apart. If a head-tail integration existed, then the primers will yield a product, since the LCR region of an integrated construct would lie 5’ to the neomycin gene of the second copy. Following these techniques, 11 out of 12 of TαLCR7.4 and TαLCR4.0 were identified as single copy integrants (TαLCR4.0 were analyzed by both

Southern blot and PCR shown in Fig 11).

28 A)

Figure 11. Status of Transgene Integration in TαLCR4.0. A) Three panels of PCR gels containing VL3 (non-transfected) and TαLCR4.0 clones amplified with two sets of head to tail primers and YFP primers. Cartoon at the bottom of the gel shows a head to tail integration event and approximate position of forward and reverse primers. Black boxes highlight the head to tail band generated with L1C1 clones. Sizes of the PCR products are indicated. B) Southern blot consisting of TαLCR4.0 clones with and without drug along with VL3 (non-transfected) control. The presence of endogenous band (8.3kb) is shown in all the samples. L1C1 (+ and -) show a head to tail band (5.4kb). Bands present in L2C3 and L6C1 other than the endogenous bands indicate multiple single copy integration events. Transgene bands are not seen in L3C1 and L5C1, perhaps because of an integration with a band size above ~12kb that may be difficult to be separated/visualized. However, these clones have been confirmed to contain the transgene by PCR and flow cytometry. Construct below indicates restriction sites for BglII (B) enzyme used in digesting the genomic DNA and location of binding of probe (see method section 2.3). N indicates Not I enzyme used in linearizing the plasmid at the time of transfection.

Moreover, the Southern blot (Fig 11b) shows that the banding pattern on clones with and without drug is the same. This eliminates any concerns regarding sub-clonal drift or any alterations in the integration status of transgene over a period of time. This unilateral mode of

29 insulation is not to be confused with enhanced expression, since other than HS1, none of the other elements-HS1’, HS4 or HS6 support enhancer activity44.

CHAPTER 5: DETERMINING THE BINDING OF TCRα LCR TO THE INSULATOR FACTOR, USF1

5.1 ROLE OF USF1 IN BARRIER INSULATION

USF1 is a ubiquitously expressed transcription factor that belongs to the family of Basic-

Helix-Loop-Helix-Leucine Zipper proteins (bHLH-LZ). USF1 binds to canonical site CACGTG as well as the non-canonical CACNTG. USF1 has been shown to regulate gene transcription and bind to several other transcription factors54. USF1 is also involved in chromatin modulating activities by recruiting histone-modifying complexes31. Moreover, USF1 has been shown to interact particularly with heavily acetylated regions of the chromatin55. The role of USF1 in indirectly modulating DNA topology was demonstrated by its interaction with the topoisomerase

III gene 56. Thus, USF1 seems to play an important role in regulating chromatin accessibility and transcription activation.

As mentioned earlier, USF1 is essential for mediating insulation in the chicken β-globin locus

(Fig 12). Knockdown of USF1 or its binding sites disrupts the recruitment of active histone modifications and thereby prevents the barrier function of chicken β-globin insulator. USF1 associates with histone modifying complexes, SET7/9 (histone methyl transferase), PCAF

(histone acetyl transferase) and PRMT1 (H4R3 specific histone methyl transferase)30,31.

Moreover, USF1 is also thought to mediate the recruitment of histone acetyl transferases p300 and CBP28. USF1 has been shown to bind to insulators found in the human ankyrin and spectrin genes34,35. However, the outcome of loss of the USF1 factor or its binding sites in these insulators is yet to be determined. Therefore, with the available information thus far, it is unclear 30 if USF1 is a ‘universal’ barrier factor. In principle, any factor that can recruit histone-modifying complexes to maintain a euchromatin structure can potentially function as a barrier factor.

Figure 12. Role of USF1 at the Chicken β-Globin Locus. The chicken β-globin cHS4 insulator is bound by USF1 that recruits active chromatin modifying complexes including SET, p300/CBP and PCAF. The DNA-protein complex assembled at the cHS4 insulator allows for the maintenance of euchromatin state of β-globin genes and prevention of spread of heterochromatin from the adjacent domain (HE: histone-modifying enzymes, SF: silencing factors, Ac: acetylated, Me: methylated). Adapted from Wei et al.57.

5.2 INSULATOR FACTOR USF1 BINDS TO THE HS1’ REGION, BUT NOT THE HS4 OR HS6 REGIONS OF THE TCRα LCR

To determine if USF1 plays a role in the insulation exhibited by the TCRα LCR, we scanned for USF1 consensus site within the TCRα LCR region. The search revealed that HS1’ contains both canonical (CACGTG) and non-canonical binding sites (CACGGG), and HS4 and HS6 regions each contain a non-canonical binding site. In order to determine if the USF1 factor binds to these HS in the TCRα LCR region, ChIP experiment with α-USF1 antibody was conducted in mouse thymocytes. Primers to a region in the retinoic acid-related orphan receptor γT (RORγT) promoter, known to bind to USF1 in T cells 58, were used as a positive control. Primers to a region of the inactive β-globin gene served as the negative control for these ChIP experiments52.

ChIP data showed that only the HS1’ with the canonical binding site was associated with the

USF1 factor. The non-canonical regions in HS1’, HS4 and HS6 do not bind to the USF1 (Fig

13).

Figure 13. USF1 is Recruited to the HS1’ Region of the TCRα LCR. ChIP assays using USF1 antibody were conducted on chromatin from ex vivo mouse thymocytes from three separate individual wild type mice. HS1’-1 and HS1’-2 indicate results from primers to detect USF1 binding at canonical and non-canonical USF1 sites, respectively. HS4 and HS6 indicate primers to detect USF1 binding at a non-canonical site within each region. IgG, globin and RORγT were included as background, negative and positive controls, respectively. The qPCR reactions were conducted in triplicate for each experiment. USF1 ChIP signals were quantified and expressed in terms of fold enrichment over IgG background signal.

Transgenes linked to HS1 and HS1’ alone exhibit very poor expression levels in transgenic mice indicative of the lack of ability of these HS to suppress position effects 14,59. Thus these sites are poor candidates for barrier insulator elements. As such, we predict that it is highly unlikely that HS1’ bound USF1 alone is sufficient for transgene insulation. However, the finding that the barrier factor, USF1 binds to HS1’ necessitates the investigation of the role of HS1’ in insulation by either knocking down USF1 or removing the HS1’ in insulator assays. Both USF1 dependent and independent mechanisms of insulation might be important at this locus.

Exploration of the role of USF1 and identification of factors that bind to HS4 and HS6 will enable better understanding of potential USF1 dependent and independent regulation mediated at this locus.

32 5.3 TESTING SHORT HAIRPIN RNAs (SHRNA) TO KNOCKDOWN USF1

In order to identify the role of USF1 in insulation, we aim to study the effect of knockdown

USF1 protein in the (insulated) TαLCR4.0 clones. Short hairpin RNA (shRNA) mediated knockdown of USF1 was optimized in an initial experiment using a mouse T cell line (VL3).

Knockdown of gene expression by RNA interference is a widely used biological tool. Expression of shRNA induces degradation of target mRNA. Four sets of shRNA vectors (U1, U2, U3 and

U4) targeting USF1 were tested for their ability to knockdown USF1 in VL3 cells. A control vector to a non-targeting region was included. shRNA vectors were stably transfected into the cells and knockdown of USF1 was analyzed by RT-qPCR (reverse transcriptase qPCR) and

Western blot assays (Fig 14a,b). U1 shRNA vectors yielded the highest reduction in the levels of

USF1 mRNA (87.8% reduction in mRNA levels compared to control vector) and protein.

33 A)

Figure 14. shRNA Knockdown of USF1 in VL3-3M2 Cell Line. VL3-3M2 cells were transfected with non-targeting control shRNA vector or four shRNA vectors targeting different sites of USF1 mRNA (U1, U2, U3 and U4). After selection in puromycin, RNA and cell lysate was harvested. A) cDNA obtained from RNA was subject to qPCR using primer sets specific to USF1 and actin. Relative mRNA expression level was obtained by delta-delta Ct method, normalized to actin and control vector. B) Western blot was performed using antibodies against USF1 and GAPDH (loading control).

34 CHAPTER 6: IDENTIFICATION OF FACTORS THAT BIND TO THE TCRα LCR

6.1 CANDIDATE FACTORS THAT BIND TO THE TCRα LCR

In order to identify what other factors might play a role in the insulator activity other than the

HS1’ bound USF1, we directed our focus to factors that bind to HS4 and HS6 regions. Findings from previous studies strongly suggest that HS4 and HS6 that drive position effect suppression are the ideal candidates for insulator elements. It has been shown that the absence of HS4 disrupts TCRα LCR’s ability to suppress position effects on a linked transgene in vivo. HS6 has been studied in more detail. The functional regions of HS6 that is responsible for overcoming position effects have been mapped to three thymic footprints (TF123) and a 316bp region60. In the absence of rest of the LCR (in constructs), HS6 allows for suppression of position effects in a tissue-unrestricted manner60. Specifically, the HS6-316bp region, but not the TF123 region, suppresses position effects in non-lymphoid cells (fibroblast)42. TF2 and TF3 bind lymphoid specific factors, AML-1/RUNX and Elf-1 respectively. It is highly relevant to note both of these proteins interact with chromatin remodeling complexes61,62. Collectively, our prior and present findings support the hypothesis that the uncommonly strong insulation capacity of the TCRα

LCR is a product of novel synergy between multiple distinct barrier insulator elements residing within the HS4 and HS6. By utilizing ChIP-seq databases, we identified the three additional factors that bind to the HS6 region: Foxo1, STAT3 and IRF4 (Fig 15).

Figure 15. Factors Associated with the TCRα LCR HS Elements. Using publicly available datasets, factors binding to the TCRα LCR regions were identified. Arrows indicate the regions within each HS bound by specific factors.

The five HS6 DNA-binding factors mentioned here could play a role in TCRα LCR-derived insulation by different mechanisms. AML1/Runx1 and STAT3 are known to interact with the p300 histone acetyltransferase complex61,63. Therefore, by analogy to the role of USF at vertebrate insulators30, one or both of these factors may function by recruiting this complex to the gene locus. This notion is supported by available ChIP seq data indicating that p300 is indeed enriched at HS4 and HS6 DNA in T cells. The IRF4 factor has been shown to promote histone acetylation at target gene loci64 and has also been found to work cooperatively with

STAT3 in regulating gene expression in T cells65. Finally, both Elf-1 and FoxO1 are members of the winged helix class of transcription factors. Several winged helix factors have been found to act as “pioneer factors” capable of engaging silent chromatin to make it available to activating transcription factors66,67. The most prominent member of this class of pioneer factors is the

FoxA1/HNF3a factor, which has the ability to regulate nucleosome positioning to facilitate enhancer function68,69. In a similar fashion, Elf-1 has been shown to be associated with a specifically positioned nucleosome at an enhancer of the IL2Rα gene, and collaborates with 36 HMG-I(Y) to regulate nucleosome architecture and enhanceosome assembly 70. FoxO1 has the capacity to disrupt histone-DNA contacts and thus de-condense compacted chromatin arrays71.

In summary, this information suggests that multiple molecular mechanisms of action may be playing a role in the function of the TCRα LCR-derived insulator elements to establish or maintain euchromatin. Future experiments would be designed to study the effect of knockdown of these factors on barrier insulation.

37 CHAPTER 7: CHROMATIN LANDSCAPE OF THE TCRα/DAD1 LOCUS

7.1 CHROMATIN MARKS ASSOCIATED WITH TCRα, LCR AND DAD1 IN THYMUS AND NON-THYMUS TISSUES

In the endogenous context, barrier insulators have been shown to prevent the spread of heterochromatin from one gene locus to another. For example, the chicken cHS4 insulator separates the euchromatin β-globin gene from the nearby 16kb condensed chromatin. In the case of TCRα/DAD1 locus, we hypothesized that the TCRα LCR would act as a barrier to separate the ubiquitously expressed DAD1 from TCRα in tissue types where it is not expressed and might assume a heterochromatin state. In order to explore this hypothesis, publicly available ChIP- sequencing (ChIP-seq) datasets were analyzed to study the chromatin landscape of the

TCRα/DAD1 locus.

In support of our hypothesis, we found datasets that showed that TCRα is associated with heterochromatin in two non-thymus cell types: erythroid and megakaryocytes. TCRα is enriched with H3K27me3 heterochromatin mark in both these cell types whereas as expected, DAD1 is devoid of such marks (Fig 16).

Figure 16. Chromatin (heterochromatin marks) Architecture of the Mouse TCRα/DAD1 Locus. Publicly available ChIP-sequencing datasets were viewed on the mm8 and mm10 UCSC genome browser and images of tracks obtained are shown (https://genome.ucsc.edu/). The positions of the TCRα and DAD1 exons are indicated. H3K27me3 ChIP-sequencing experiments conducted in erythroid progenitor cell line (G1E) and megakaryocytes are depicted as peak traces (black).

These instances suggest that TCRα LCR might function as typical barriers protecting the euchromatin of DAD1 from potential encroachment by the heterochromatin enriched in TCRα locus. However, TCRα does not exist in a heterochromatin state in other cell types (an example or two?) that were examined, suggesting that such a potential role of the LCR would not be universally required to protect DAD1 in all tissues.

Examination of the euchromatin status of the TCRα and DAD1 genes revealed as expected that TCRα is enriched for euchromatin marks in T cells, and DAD1 is euchromatin in both T and non-T cells. Interestingly, we noticed ‘qualitative’ differences in the accumulation of these marks across this locus. Analysis of euchromatin marks including H3K4me3, H3K27ac, H3ac and

H3K9ac showed that these marks were strictly restricted only to the promoter region of DAD1 gene, however enrichment for these marks were spread throughout the body of TCRα and LCR

39 regions (Fig 17). A distinct ‘boundary’ in the accumulation of euchromatin marks in this locus suggests a lack of spread of active chromatin marks from the TCRα/LCR region into the body of the DAD1 gene.

Figure 17. Chromatin (euchromatin marks) Architecture of the Mouse TCRα/DAD1 Locus. Publicly available ChIP-sequencing datasets were viewed on the mm8 and mm10 UCSC genome browser and images of tracks obtained are shown (https://genome.ucsc.edu/). The position of the TCRα and DAD1 exons are indicated. Dotted box through the tracks highlights the genomic location of the TCRα LCR (HS1 through HS6). ChIP-sequencing experiments conducted for H3K4me3, H3K9ac, H3K27ac and H3ac (diacetylated [K9/K14] histone H3) in CD4+ T cells were obtained via Cistrome database (http://www.cistrome.org/) and viewed on UCSC browser as bigwig peaks. 40

We hypothesize that there may be a need to prevent the spread of active chromatin marks from the TCRα locus into the DAD1 gene. This might be of importance especially during V-D-J recombination; a process that involves highly regulated chromatin accessibility to the RAG recombinase in that the PHD domain of RAG2 specifically recognizes H3K4me3 modification72.

The concept of spread of active chromatin is not as commonly explored or understood as that of the spread of heterochromatin, however few examples of such occurrences do exist. Specifically, a barrier insulator prevents the spread of active chromatin in the TCRβ locus; deletion of the barrier results in spread of euchromatin and disruption of V-D-J recombination process73. In conclusion, it would be interesting to further explore the non-typical role of TCRα LCR insulator elements in preventing euchromatin spread from TCRα to DAD1 in T cells.

41 CHAPTER 8: ASSESSING THE INSULATOR ACTIVITY OF TCRα LCR IN A FISSION YEAST MODEL

8.1 FISSION YEAST: TESTING A NEW MODEL TO STUDY VERTEBRATE INSULATORS

Schizosaccharomyces pombe is a unicellular eukaryote that is a widely chosen to study chromatin organization. S.pombe can be easily manipulated and homologous recombination occurs efficiently. Moreover the chromatin organization and epigenetic mechanisms, specifically heterochromatin of S.pombe is similar to that of constitutive heterochromatin in humans74. The

14.1Mbp genome of S.pombe is contained within three chromosomes. The centromeres of

S.pombe can be closely compared to vertebrates in that the kinetochore is surrounded by pericentromeric heterochromatin. The three centromeres of fission yeast have similar structural units consisting of a central domain, flanked by inner repeats and outer repeats. The central core domain contains euchromatin regions marked by methylated H3K4 interposed with Cnp1

(centromeric protein A (CENPA) homolog)75,76. The outer repeats contain heterochromatin regions with methylated H3K9 bound to swi6 (HP1 homolog)77. tRNA barrier elements that exist within the inner repeats separate the euchromatin and heterochromatin regions78,79. L5 element, a

DNA region isolated from the outer centromeric repeats region can ectopically induce pericentromeric heterochromatin on an associated reporter gene50. However, when a tRNA barrier is inserted between L5 and a reporter gene, heterochromatin formation is blocked and the reporter is maintained in active chromatin status. Recently, a human tDNA insulator was also shown to block the L5 from inducing heterochromatin on a reporter gene80.

In our study we utilized S.pombe model to determine if the TCRα LCR insulator can protect the reporter gene from silencing induced by L5. Considering the resemblance in heterochromatin assembly between vertebrates and S.pombe, we rationalized that the TCRα LCR can be tested 42 for its insulation capacity against the L5 element. However, a main drawback was that none of the aforementioned TCRα LCR candidate factors have known homologs in S.Pombe.

Nevertheless, given the convenience and speed of the working with S.pombe compared to the 90 day VL3 barrier assay, we designed this novel experiment.

8.2 HS6 ELEMENT OF THE TCRα LCR IS NOT SUFFICIENT TO INSULATE A TRANSGENE IN YEAST INSULATOR ASSAY

As previously emphasized, HS6 is a strong candidate insulator element given its ability to suppress position effects and bind to chromatin modulating factors. As an initial pilot experiment, we tested the ability of HS6 in insulating a reporter gene (uracil, ura4+) from L5 mediated silencing. S.pombe uracil auxotrophs can grow only when plated on uracil media or transformed with a construct bearing ura4. In the S.pombe barrier assay, when the ura4+ gene is linked to the L5 element, it results in silencing of ura4+ gene. To visualize this phenomenon, the transformed cells are grown on plates containing or lacking uracil (+/-URA), or containing FOA

(+FOA-5-Fluoroorotic acid). 5-FOA is a compound that gets converted to 5-fluorouracil by an enzyme coded by ura4+. The converted form of 5-FOA is toxic to cells, therefore when ura4 is not silenced, cells do not grow on FOA media. Thus, when ura4+ is silenced by L5, growth occurs only in +FOA and +URA plates and no growth in –URA plates. In contrast, when an insulator protects ura4 from L5-silencing, growth occurs in –URA plates but not +FOA plates.

A construct bearing HS6 placed between L5 and ura4+ was designed. As controls, constructs with L5 linked to ura4+ or a construct with ura4+ alone were used. In these constructs, ade6 located 3’ to the ura4+ reporter gene was used as transformation marker. Each of these constructs was transformed into ade6- S.pombe strains, followed by selection of transformed colonies (on ade6- plates) that were then cultured on –URA, +URA and +FOA plates. In the L5-HS6- ura4+ plate, lack of growth in +URA and presence of colonies in +FOA plates was observed indicating

43 that HS6 was unable to protect ura4 from silencing. The results from constructs bearing HS6 linked to L5 or the L5 element alone, showed similar patterns of growth. The control, ura4+ alone construct was able to grow on -URA plates and lacked growth in +FOA plates, confirming that the assay was functional (Fig 18).

Figure 18. Testing Fission Yeast Model to Study TCRα LCR. Growth assay was conducted on serially diluted cells (highest to lowest concentration from left to right lane) from the following strains- ura4+, ura4-, L5+ ura4+ and L5+HS6 ura4+ in media lacking URA (-URA) and in media containing FOA (+FOA).

These are some reasons that might explain the failure of HS6 in insulating ura4+ from silencing. i) The candidate factors that bind to HS6 do not have homologs in S.Pombe.

Transforming the silenced clones with a rescue plasmid bearing cDNA of each of the candidate factors could potentially save the reporter gene from silencing. ii) There is a possibility of synergistic interaction of HS6 with HS4 and/or HS1’. A construct bearing all three HS could be tested in this assay, and it would also be important to test a stuffer fragment of the same size to resolve any issues related to the possible inability of the L5 element to induced heterochromatin beyond a certain distance. Furthermore, a potential scenario wherein the three HS together do not overcome silencing might indicate the need for vertebrate factors beyond the identified candidate factors.

44 CHAPTER 9: CONCLUSION

LCRs are multifaceted cis-elements that provide spatiotemporal and copy number dependent expression upon a linked reporter gene. The key aspect of an LCR’s function lies in its ability to suppress position effects that allows it to establish or maintain a linked gene in a euchromatin state. Barrier insulators share this characteristics feature with LCRs. LCRs have been often hypothesized to contain barrier insulator elements that mediate suppression of position effects.

The TCRα LCR contains HS4 and HS6 elements that exhibit hallmark features of a barrier insulator, however a direct link between the LCR and insulator has yet not been established42,43,60,81. Here, we show that the TCRα LCR might contain HS with barrier-like functions that suppress position effects in a long-term cell culture assay. The insulated reporter gene exhibits active chromatin state as demonstrated both by ChIP and DNase sensitivity assays.

We also show that a minimal LCR construct, TαLCR4.0 containing HS1, HS1’, HS4 and HS6 was sufficient to protect the reporter gene from position effects. Since HS1-HS1’ do not possess suppression of position effects function, we strongly hypothesize that HS4 and/or HS6 are the key candidate barrier insulator elements. Furthermore, the LCR insulates transgene in a novel, unilateral fashion that is distinct from what is known thus far in the field. To understand the role of insulation in the endogenous locus, the chromatin landscape of the TCRα/DAD1 was explored. Preliminary study showing distinct separation of euchromatin in TCRα and DAD1 genes, led to the development of a hypothesis in which the TCRα LCR might prevent in-cis spreading of active chromatin from TCRα to DAD1 in T cells. We believe that this novel aspect of insulation might play a role in epigenetically controlled processes such as VDJ recombination.

45 9.1 MODEL OF INSULATION We propose several models that take into consideration novel and known mechanisms of barrier insulation (Fig 19). a) As mentioned in earlier sections, barrier insulators function by recruiting histone-modifying complexes to establish and maintain the linked transgene in open chromatin conformation. HS1’ site of the LCR is bound by USF1, a transcription factor that has been proven essential for recruiting histone-modifying complexes to the cHS4 insulator. Since HS1’ does not exhibit suppression of position effects, it is unlikely that HS1’-bound USF1 alone mediates insulation at this locus, we identified additional factors that bind to HS4 or HS6 by utilizing publically available ChIP-seq databases. The database search yielded three factors that bind to HS6; these include FoxO1, STAT3 and IRF4. Similar to USF1, these factors have an ability to recruit histone modifying complexes or act as pioneer factors in establishing euchromatin. Moreover, previous research in our lab has shown that AML1/RUNX1 and Elf-1 both of which are known to interact with chromatin modifying complexes bind to HS6. These findings strengthen the proposed theory that the LCR recruits active histone-modifiers to maintain the transgene in a euchromatin status. b) The protection of reporter gene from position effects in a unilateral manner suggests that the

TCRα LCR insulator might bear unknown mechanisms of insulation. We hypothesize that the

LCR might exert unilateral mode of insulation either by looping or tethering the reporter gene to sub-nuclear compartments.

In the looping model, TCRα LCR might associate with certain factors to form loops with a downstream target region, thereby protecting the reporter gene within the looped region from heterochromatin effects and recruit active histone-modifying complexes to maintain the transgene in open chromatin. In fact, LCRs have been proposed to function by folding into a loop allowing for synergistic interactions between HS and factors82. For example, chromatin loops

46 form between the β-globin promoter and its LCR to drive developmental stage specific expression of β-globin genes83. CTCF84, a highly conserved eukaryotic zinc finger transcription factor is widely implicated in DNA loop formation. The HS1’ region of the TCRα LCR binds to

CTCF but the removal of CTCF has no effect on LCR activity43. Hence, future experiments could be designed to identify other factors that might facilitate looping of the LCR.

Another model by which the TCRα LCR can exert its unilateral mode of insulation is by targeting the reporter gene to regions of sub-nuclear compartments. There exist regions (nuclear pore, nuclear lamina) within the nucleus that act as sites of active chromatin hubs with high levels of transcription activity. These active chromatin environments are not conducive to the formation or presence of heterochromatin regions. In yeast silent type mating locus

Homothallic Mating Left (HML), the boundary element physically tethers the HML locus to the nuclear pore complex to block spreading of heterochromatin 85. This mechanism is also seen in

Drosophila insulators. Gypsy insulator, which is well studied for its enhancer-blocking insulator activity, also has the ability to suppress heterochromatin silencing due to position effects86. It is speculated that the insulator might suppress position effects by the same mechanism as that of its enhancer-blocking activity. Factors binding to Gypsy drive the formation of insulator bodies that attach to nuclear lamina87,88. Chromatin fibers associated with these complexes are clustered into loops segregating the enhancer-promoter interactions. Perhaps, the Gypsy barrier insulator might function in a similar fashion in that sequestering of chromatin fibers to nuclear lamina region would prevent influence from heterochromatin regions.

The HS1’ region of the LCR has been shown to be associated with nuclear matrix; a region between HS3 and HS4 sites also shows moderate association89. The relevance of this finding is not yet known and it would be interesting to explore if this finding has any significance to insulation mediated by TCRαLCR. While these theories suggest that HS1’ might have a strong role to play in insulation, it is highly unlikely that HS1’ alone is sufficient for insulation, since 47 this element does not suppress position effect, a hallmark feature of barrier insulators. It is possible that HS1’ collaborates with HS4/HS6 in exerting insulator activity.

Figure 19. Model of Insulator Activity Mediated by TCRαLCR. The LCR might form a loop with the transgene promoter or maintain the euchromatin state by recruiting active histone- modifying complexes (Ac). Alternatively, the LCR might bind to factors that tether it to nuclear matrix region in order to insulate the transgene from heterochromatin (coils). CTCF protein that binds to HS1’ might play a role in the formation of the loop.

9.2 FUTURE DIRECTIONS There are many unanswered questions and theories to be explored in order to obtain a better understanding of mechanism of insulation and regulation at the TCRα/DAD1 locus. The finding that the TCRαLCR4.0 is sufficient for insulation suggests that either one or all of the HS (HS1’,

HS4 HS6) are important. Barrier assays can be designed to test the individual and synergistic functions of the HS. 90-day barrier assay is a lengthy experiment prone to potential flow cytometry errors/variations. Experimental models that test the insulator capacity of HS in a quick and reliable way need to be designed. With this goal, fission yeast model was explored in this study. The initial attempt done in this study showed that the HS6 site of the LCR was insufficient in insulating the heterochromatin-inducing L5 element. However, these experiments can be continued in two different ways: a) testing the role of HS1’, HS4 and HS6 together (in place of

48 HS6-alone) in the fission-yeast model (control constructs consisting of spacer elements may have to be included to eliminate any issues related to distance from L5 to the LCR), and b) rescuing the originally tested HS6-L5-URA yeast clones with cDNA of candidate insulator factors

(AML-1/RUNX, FOXO1, STAT3, IRF4).

Having identified that USF1, an important barrier factor binds to HS1’, the next logical step lies is in determining the effect of loss of this factor in insulation mediated by TCRα LCR.

Preliminary experiments have identified an shRNA (targeted at USF-1 mRNA) that effectively knocks down USF1 in VL3 cell line. As next steps, the USF1-shRNA construct can be stably transfecting into TαLCR4.0 clones and loss or maintenance of YFP expression levels would inform about the effect of USF1 on barrier activity.

The ability of the LCR to form loops can be explored by carefully designing transgene-LCR constructs for testing in 3C experiments. Overall, identification of the HS and factors involved in insulation can help streamline experiments to narrow on the mechanism of insulation.

9.3 SIGNIFICANCE The role of TCRα LCR in the regulation of differentially expressed TCRα and DAD1 genes is not yet clear. Identifying and understanding the insulation properties of the LCR, aims at explaining how the TCRα and DAD1 genes are maintained as distinct gene expression domains without any promiscuous communication. High-level genome organization consists of complex arrangement of genes with different spatiotemporal expressions, insulators are said to define transcriptional activity into separate domains. Studying the action of insulators within a single gene loci such as the TCRα/DAD1 locus, can be extrapolated to understand insulation at the 3D genome level. Moreover, there are not many well-studied vertebrate insulators other than the chicken β-globin insulator; identification of TCRα LCR insulator serves a great advantage in understanding the mechanism of insulation. 49 Study of the TCRα LCR is not only beneficial in terms of basic science but also in translational science. In gene therapy, therapeutic genes often delivered in a lentiviral or retroviral vectors are subject to silencing based on the nature of chromatin environment at the site of integration90. Insulators can be utilized to protect the therapeutic gene from such position effects. Chicken β-globin insulator was shown to improve the probability of expression of retroviruses subject to random integration events in a murine erythroleukemia cell line90. While met with some level of success, the study showed that complete protection of retroviruses from silencing was not observed especially in certain regions that might be centromeric heterochromatin. Hence, identification of additional insulators such as that of TCRα LCR can be of great benefit if proven successful in insulating a therapeutic gene. Overall, this study has laid the groundwork to expand knowledge about mechanism of LCR’s action and gene regulation at other complex multi gene loci.

REFERENCES

1 Kornberg, R. D. Structure of chromatin. Annual review of biochemistry 46, 931-954, doi:10.1146/annurev.bi.46.070177.004435 (1977). 2 Barkess, G. & West, A. G. Chromatin insulator elements: establishing barriers to set heterochromatin boundaries. Epigenomics 4, 67-80, doi:10.2217/epi.11.112 (2012). 3 Kellum, R. & Schedl, P. A position-effect assay for boundaries of higher order chromosomal domains. Cell 64, 941-950 (1991). 4 Wang, Y. M. et al. Correlation between DNase I hypersensitive site distribution and gene expression in HeLa S3 cells. PLoS One 7, e42414, doi:10.1371/journal.pone.0042414 (2012). 5 Magram, J., Chada, K. & Costantini, F. Developmental regulation of a cloned adult beta- globin gene in transgenic mice. Nature 315, 338-340 (1985). 6 Townes, T. M., Lingrel, J. B., Chen, H. Y., Brinster, R. L. & Palmiter, R. D. Erythroid- specific expression of human beta-globin genes in transgenic mice. The EMBO journal 4, 1715-1723 (1985). 7 Kollias, G., Wrighton, N., Hurst, J. & Grosveld, F. Regulated expression of human A gamma-, beta-, and hybrid gamma beta-globin genes in transgenic mice: manipulation of the developmental expression patterns. Cell 46, 89-94 (1986). 8 Kioussis, D., Vanin, E., deLange, T., Flavell, R. A. & Grosveld, F. G. Beta-globin gene inactivation by DNA translocation in gamma beta-thalassaemia. Nature 306, 662-666 (1983). 9 Driscoll, M. C., Dobkin, C. S. & Alter, B. P. Gamma delta beta-thalassemia due to a de novo mutation deleting the 5' beta-globin gene activation-region hypersensitive sites. Proc Natl Acad Sci U S A 86, 7470-7474 (1989). 10 Grosveld, F., van Assendelft, G. B., Greaves, D. R. & Kollias, G. Position-independent, high-level expression of the human beta-globin gene in transgenic mice. Cell 51, 975-985 (1987). 11 Groudine, M., Kohwi-Shigematsu, T., Gelinas, R., Stamatoyannopoulos, G. & Papayannopoulou, T. Human fetal to adult hemoglobin switching: changes in chromatin structure of the beta-globin gene locus. Proceedings of the National Academy of Sciences of the United States of America 80, 7551-7555 (1983). 12 Forrester, W. C., Thompson, C., Elder, J. T. & Groudine, M. A developmentally stable chromatin structure in the human beta-globin gene cluster. Proceedings of the National Academy of Sciences of the United States of America 83, 1359-1363 (1986). 13 Forrester, W. C., Takegawa, S., Papayannopoulou, T., Stamatoyannopoulos, G. & Groudine, M. Evidence for a locus activation region: the formation of developmentally stable hypersensitive sites in globin-expressing hybrids. Nucleic acids research 15, 10159-10177 (1987). 14 Diaz, P., Cado, D. & Winoto, A. A locus control region in the T cell receptor alpha/delta locus. Immunity 1, 207-217 (1994). 15 Greaves, D. R., Wilson, F. D., Lang, G. & Kioussis, D. Human CD2 3'-flanking sequences confer high-level, T cell-specific, position-independent gene expression in transgenic mice. Cell 56, 979-986 (1989).

51 16 Jones, B. K., Monks, B. R., Liebhaber, S. A. & Cooke, N. E. The human growth hormone gene is regulated by a multicomponent locus control region. Molecular and cellular biology 15, 7010-7021 (1995). 17 Aronow, B. J. et al. Functional analysis of the human adenosine deaminase gene thymic regulatory region and its ability to generate position-independent transgene expression. Molecular and cellular biology 12, 4170-4185 (1992). 18 Eszterhas, S. K., Bouhassira, E. E., Martin, D. I. & Fiering, S. Transcriptional interference by independently regulated genes occurs in any relative arrangement of the genes and is influenced by chromosomal integration position. Molecular and cellular biology 22, 469-479 (2002). 19 Milot, E. et al. Heterochromatin effects on the frequency and duration of LCR-mediated gene transcription. Cell 87, 105-114 (1996). 20 Festenstein, R. et al. Locus control region function and heterochromatin-induced position effect variegation. Science (New York, N.Y.) 271, 1123-1125 (1996). 21 Heger, P. & Wiehe, T. New tools in the box: an evolutionary synopsis of chromatin insulators. Trends Genet 30, 161-171, doi:10.1016/j.tig.2014.03.004 (2014). 22 Sun, F. L. & Elgin, S. C. Putting boundaries on silence. Cell 99, 459-462 (1999). 23 Pikaart, M. J., Recillas-Targa, F. & Felsenfeld, G. Loss of transcriptional activity of a transgene is accompanied by DNA methylation and histone deacetylation and is prevented by insulators. Genes & development 12, 2852-2862 (1998). 24 Chung, J. H., Whiteley, M. & Felsenfeld, G. A 5' element of the chicken beta-globin domain serves as an insulator in human erythroid cells and protects against position effect in Drosophila. Cell 74, 505-514 (1993). 25 Prioleau, M. N., Nony, P., Simpson, M. & Felsenfeld, G. An insulator element and condensed chromatin region separate the chicken beta-globin locus from an independently regulated erythroid-specific folate receptor gene. The EMBO journal 18, 4035-4048, doi:10.1093/emboj/18.14.4035 (1999). 26 Jin, C. & Felsenfeld, G. Distribution of histone H3.3 in hematopoietic cell lineages. Proc Natl Acad Sci U S A 103, 574-579, doi:10.1073/pnas.0509974103 (2006). 27 Litt, M. D., Simpson, M., Gaszner, M., Allis, C. D. & Felsenfeld, G. Correlation between histone lysine methylation and developmental changes at the chicken beta-globin locus. Science (New York, N.Y.) 293, 2453-2455, doi:10.1126/science.1064413 (2001). 28 Chung, J. H., Bell, A. C. & Felsenfeld, G. Characterization of the chicken beta-globin insulator. Proc Natl Acad Sci U S A 94, 575-580 (1997). 29 Litt, M. D., Simpson, M., Recillas-Targa, F., Prioleau, M. N. & Felsenfeld, G. Transitions in histone acetylation reveal boundaries of three separately regulated neighboring loci. The EMBO journal 20, 2224-2235, doi:10.1093/emboj/20.9.2224 (2001). 30 Huang, S., Li, X., Yusufzai, T. M., Qiu, Y. & Felsenfeld, G. USF1 recruits histone modification complexes and is critical for maintenance of a chromatin barrier. Molecular and cellular biology 27, 7991-8002, doi:10.1128/MCB.01326-07 (2007). 31 West, A. G., Huang, S., Gaszner, M., Litt, M. D. & Felsenfeld, G. Recruitment of histone modifications by USF proteins at a vertebrate barrier element. Mol Cell 16, 453-463, doi:10.1016/j.molcel.2004.10.005 (2004). 32 Dickson, J. et al. VEZF1 elements mediate protection from DNA methylation. PLoS Genet 6, e1000804, doi:10.1371/journal.pgen.1000804 (2010). 33 Ma, M. K., Heath, C., Hair, A. & West, A. G. Histone crosstalk directed by H2B ubiquitination is required for chromatin boundary integrity. PLoS Genet 7, e1002175, doi:10.1371/journal.pgen.1002175 (2011).

52 34 Gallagher, P. G. et al. Mutation of a barrier insulator in the human ankyrin-1 gene is associated with hereditary spherocytosis. J Clin Invest 120, 4453-4465, doi:10.1172/JCI42240 (2010). 35 Gallagher, P. G. et al. An insulator with barrier-element activity promotes alpha-spectrin gene expression in erythroid cells. Blood 113, 1547-1554, doi:10.1182/blood-2008-06- 164954 (2009). 36 Carabana, J., Watanabe, A., Hao, B. & Krangel, M. S. A barrier-type insulator forms a boundary between active and inactive chromatin at the murine TCRbeta locus. Journal of immunology (Baltimore, Md. : 1950) 186, 3556-3562, doi:10.4049/jimmunol.1003164 (2011). 37 Tripathi, R., Jackson, A. & Krangel, M. S. A change in the structure of Vbeta chromatin associated with TCR beta allelic exclusion. Journal of immunology (Baltimore, Md. : 1950) 168, 2316-2324 (2002). 38 Nakashima, T. et al. Molecular cloning of a human cDNA encoding a novel protein, DAD1, whose defect causes apoptotic cell death in hamster BHK21 cells. Molecular and cellular biology 13, 6367-6374 (1993). 39 Makishima, T. et al. The highly conserved DAD1 protein involved in apoptosis is required for N-linked glycosylation. Genes to cells : devoted to molecular & cellular mechanisms 2, 129-141 (1997). 40 Hong, N. A. et al. A targeted mutation at the T-cell receptor alpha/delta locus impairs T- cell development and reveals the presence of the nearby antiapoptosis gene Dad1. Molecular and cellular biology 17, 2151-2157 (1997). 41 Ortiz, B. D., Cado, D. & Winoto, A. A new element within the T-cell receptor alpha locus required for tissue-specific locus control region activity. Molecular and cellular biology 19, 1901-1909 (1999). 42 Harrow, F., Amuta, J. U., Hutchinson, S. R., Akwaa, F. & Ortiz, B. D. Factors binding a non-classical Cis-element prevent heterochromatin effects on locus control region activity. The Journal of biological chemistry 279, 17842-17849, doi:10.1074/jbc.M401258200 (2004). 43 Gomos-Klein, J., Harrow, F., Alarcón, J. & Ortiz, B. D. CTCF-Independent, but Not CTCF-Dependent, Elements Significantly Contribute to TCR-α Locus Control Region Activity. The Journal of Immunology 179, 1088-1095, doi:10.4049/jimmunol.179.2.1088 (2007). 44 Winoto, A. & Baltimore, D. Alpha beta lineage-specific expression of the alpha T cell receptor gene by nearby silencers. Cell 59, 649-655 (1989). 45 Bassing, C. H. et al. T cell receptor (TCR) alpha/delta locus enhancer identity and position are critical for the assembly of TCR delta and alpha variable region genes. Proc Natl Acad Sci U S A 100, 2598-2603, doi:10.1073/pnas.0437943100 (2003). 46 Lahiji, A. et al. Complete TCR-alpha gene locus control region activity in T cells derived in vitro from embryonic stem cells. Journal of immunology (Baltimore, Md. : 1950) 191, 472-479, doi:10.4049/jimmunol.1300521 (2013). 47 Rosso, M., Polotskaia, A. & Bargonetti, J. Homozygous mdm2 SNP309 cancer cells with compromised transcriptional elongation at p53 target genes are sensitive to induction of p53-independent cell death. Oncotarget 6, 34573-34591, doi:10.18632/oncotarget.5312 (2015). 48 McArthur, M., Gerum, S. & Stamatoyannopoulos, G. Quantification of DNaseI- sensitivity by real-time PCR: quantitative analysis of DNaseI-hypersensitivity of the mouse beta-globin LCR. J Mol Biol 313, 27-34, doi:10.1006/jmbi.2001.4969 (2001).

53 49 Groves, T. et al. In vitro maturation of clonal CD4+CD8+ cell lines in response to TCR engagement. Journal of immunology (Baltimore, Md. : 1950) 154, 5011-5022 (1995). 50 Partridge, J. F., Scott, K. S., Bannister, A. J., Kouzarides, T. & Allshire, R. C. cis-acting DNA from fission yeast centromeres mediates histone H3 methylation and recruitment of silencing factors and cohesin to an ectopic site. Curr Biol 12, 1652-1660 (2002). 51 Mutskov, V. & Felsenfeld, G. Silencing of transgene transcription precedes methylation of promoter DNA and histone H3 lysine 9. The EMBO journal 23, 138-149, doi:10.1038/sj.emboj.7600013 (2004). 52 Kingsley, P. D. et al. "Maturational" globin switching in primary primitive erythroid cells. Blood 107, 1665-1672, doi:10.1182/blood-2005-08-3097 (2006). 53 Recillas-Targa, F. et al. Position-effect protection and enhancer blocking by the chicken beta-globin insulator are separable activities. Proc Natl Acad Sci U S A 99, 6883-6888, doi:10.1073/pnas.102179399 (2002). 54 Corre, S. & Galibert, M. D. Upstream stimulating factors: highly versatile stress- responsive transcription factors. Pigment Cell Res 18, 337-348, doi:10.1111/j.1600- 0749.2005.00262.x (2005). 55 Vettese-Dadey, M. et al. Acetylation of histone H4 plays a primary role in enhancing transcription factor binding to nucleosomal DNA in vitro. The EMBO journal 15, 2508- 2518 (1996). 56 Han, S. Y., Kim, J. C., Suh, J. M. & Chung, I. K. Cell type-dependent regulation of human DNA topoisomerase III alpha gene expression by upstream stimulatory factor 2. FEBS Lett 505, 57-62 (2001). 57 Wei, G. H., Liu, D. P. & Liang, C. C. Chromatin domain boundaries: insulators and beyond. Cell Res 15, 292-300, doi:10.1038/sj.cr.7290298 (2005). 58 Ratajewski, M., Walczak-Drzewiecka, A., Salkowska, A. & Dastych, J. Upstream stimulating factors regulate the expression of RORgammaT in human lymphocytes. Journal of immunology (Baltimore, Md. : 1950) 189, 3034-3042, doi:10.4049/jimmunol.1200519 (2012). 59 Kouskoff, V., Signorelli, K., Benoist, C. & Mathis, D. Cassette vectors directing expression of T cell receptor genes in transgenic mice. Journal of immunological methods 180, 273-280 (1995). 60 Ortiz, B. D., Harrow, F., Cado, D., Santoso, B. & Winoto, A. Function and factor interactions of a locus control region element in the mouse T cell receptor-alpha/Dad1 gene locus. Journal of immunology (Baltimore, Md. : 1950) 167, 3836-3845 (2001). 61 Westendorf, J. J. & Hiebert, S. W. Mammalian runt-domain proteins and their roles in hematopoiesis, osteogenesis, and leukemia. J Cell Biochem Suppl 32-33, 51-58 (1999). 62 Ito, Y. Molecular basis of tissue-specific gene expression mediated by the runt domain transcription factor PEBP2/CBF. Genes to cells : devoted to molecular & cellular mechanisms 4, 685-696 (1999). 63 Hedrich, C. M. et al. Stat3 promotes IL-10 expression in lupus T cells through trans- activation and chromatin remodeling. Proceedings of the National Academy of Sciences 111, 13457-13462, doi:10.1073/pnas.1408023111 (2014). 64 Lazorchak, A. S., Schlissel, M. S. & Zhuang, Y. E2A and IRF-4/Pip promote chromatin modification and transcription of the immunoglobulin kappa locus in pre-B cells. Molecular and cellular biology 26, 810-821, doi:10.1128/MCB.26.3.810-821.2006 (2006). 65 Kwon, H. et al. Analysis of interleukin-21-induced Prdm1 gene regulation reveals functional cooperation of STAT3 and IRF4 transcription factors. Immunity 31, 941-952, doi:10.1016/j.immuni.2009.10.008 (2009). 54 66 Zaret, K. S. & Mango, S. E. Pioneer transcription factors, chromatin dynamics, and cell fate control. Curr Opin Genet Dev 37, 76-81, doi:10.1016/j.gde.2015.12.003 (2016). 67 Lalmansingh, A. S., Karmakar, S., Jin, Y. & Nagaich, A. K. Multiple modes of chromatin remodeling by Forkhead box proteins. Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms 1819, 707-715, doi:http://dx.doi.org/10.1016/j.bbagrm.2012.02.018 (2012). 68 Cirillo, L. A. et al. Opening of compacted chromatin by early developmental transcription factors HNF3 (FoxA) and GATA-4. Mol Cell 9, 279-289 (2002). 69 Shim, E. Y., Woodcock, C. & Zaret, K. S. Nucleosome positioning by the winged helix transcription factor HNF3. Genes & development 12, 5-10 (1998). 70 Reeves, R., Leonard, W. J. & Nissen, M. S. Binding of HMG-I(Y) imparts architectural specificity to a positioned nucleosome on the promoter of the human interleukin-2 receptor alpha gene. Molecular and cellular biology 20, 4666-4679 (2000). 71 Hatta, M. & Cirillo, L. A. Chromatin opening and stable perturbation of core histone:DNA contacts by FoxO1. The Journal of biological chemistry 282, 35583-35593, doi:10.1074/jbc.M704735200 (2007). 72 Krangel, M. S. Beyond Hypothesis: Direct Evidence That V(D)J Recombination Is Regulated by the Accessibility of Chromatin Substrates. Journal of immunology (Baltimore, Md. : 1950) 195, 5103-5105, doi:10.4049/jimmunol.1502150 (2015). 73 Majumder, K. et al. Lineage-specific compaction of Tcrb requires a chromatin barrier to protect the function of a long-range tethering element. J Exp Med 212, 107-120, doi:10.1084/jem.20141479 (2015). 74 Trojer, P. & Reinberg, D. Facultative heterochromatin: is there a distinctive molecular signature? Mol Cell 28, 1-13, doi:10.1016/j.molcel.2007.09.011 (2007). 75 Cam, H. P. et al. Comprehensive analysis of heterochromatin- and RNAi-mediated epigenetic control of the fission yeast genome. Nat Genet 37, 809-819, doi:10.1038/ng1602 (2005). 76 Takahashi, K., Chen, E. S. & Yanagida, M. Requirement of Mis6 centromere connector for localizing a CENP-A-like protein in fission yeast. Science (New York, N.Y.) 288, 2215-2219 (2000). 77 Nakayama, J., Rice, J. C., Strahl, B. D., Allis, C. D. & Grewal, S. I. Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science (New York, N.Y.) 292, 110-113, doi:10.1126/science.1060118 (2001). 78 Scott, K. C., Merrett, S. L. & Willard, H. F. A heterochromatin barrier partitions the fission yeast centromere into discrete chromatin domains. Curr Biol 16, 119-129, doi:10.1016/j.cub.2005.11.065 (2006). 79 Scott, K. C., White, C. V. & Willard, H. F. An RNA polymerase III-dependent heterochromatin barrier at fission yeast centromere 1. PLoS One 2, e1099, doi:10.1371/journal.pone.0001099 (2007). 80 Raab, J. R. et al. Human tRNA genes function as chromatin insulators. The EMBO journal 31, 330-350, doi:10.1038/emboj.2011.406 (2012). 81 Gaszner, M. & Felsenfeld, G. Insulators: exploiting transcriptional and epigenetic mechanisms. Nature reviews. Genetics 7, 703-713, doi:10.1038/nrg1925 (2006). 82 Li, Q., Peterson, K. R., Fang, X. & Stamatoyannopoulos, G. Locus control regions. Blood 100, 3077-3086, doi:10.1182/blood-2002-04-1104 (2002). 83 Carter, D., Chakalova, L., Osborne, C. S., Dai, Y. F. & Fraser, P. Long-range chromatin regulatory interactions in vivo. Nat Genet 32, 623-626, doi:10.1038/ng1051 (2002). 84 Bell, A. C., West, A. G. & Felsenfeld, G. The protein CTCF is required for the enhancer blocking activity of vertebrate insulators. Cell 98, 387-396 (1999). 55 85 Ishii, K., Arib, G., Lin, C., Van Houwe, G. & Laemmli, U. K. Chromatin boundaries in budding yeast: the nuclear pore connection. Cell 109, 551-562 (2002). 86 Roseman, R. R., Pirrotta, V. & Geyer, P. K. The su(Hw) protein insulates expression of the Drosophila melanogaster white gene from chromosomal position-effects. The EMBO journal 12, 435-442 (1993). 87 Gerasimova, T. I., Byrd, K. & Corces, V. G. A chromatin insulator determines the nuclear localization of DNA. Mol Cell 6, 1025-1035 (2000). 88 Gerasimova, T. I. & Corces, V. G. Polycomb and trithorax group proteins mediate the function of a chromatin insulator. Cell 92, 511-521 (1998). 89 Magdinier, F., Yusufzai, T. M. & Felsenfeld, G. Both CTCF-dependent and -independent insulators are found between the mouse T cell receptor alpha and Dad1 genes. The Journal of biological chemistry 279, 25381-25389, doi:10.1074/jbc.M403121200 (2004). 90 Rivella, S., Callegari, J. A., May, C., Tan, C. W. & Sadelain, M. The cHS4 insulator increases the probability of retroviral expression at random chromosomal integration sites. J Virol 74, 4679-4687 (2000).