Histone Isoforms: from Epigenetic Regulation to Cancer Screening

Histone Isoforms: From Epigenetic Regulation To Cancer Screening

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Rajbir Singh

Ohio State Biochemistry Graduate Program

The Ohio State University

2013

Dissertation Committee:

Dr. Mark R Parthun, Advisor

Dr. Michael A Freitas

Dr. Matthew Ringel

Dr. Robert Snapka

Rajbir Singh

2013

Abstract

Epigenetic regulation comprises the changes in gene expression that are not mediated by the DNA sequence. Molecular mechanisms that mediate epigenetic regulation include

DNA methylation and chromatin modifications. Several lines of evidence indicate that both mechanisms act in concert to provide stable and heritable gene expression patterns in higher eukaryotic genomes. Since epigenetic alterations are recognized to occur in various developmental disorders and cancer, understanding this intrinsic complexity of chromatin structure is critical in understanding and treating these diseases. In our study, we have uncovered a new level of complexity in human chromatin due to the presence of functionally distinct isoforms of the replication-dependent core histones that can influence cell proliferation and tumorigenicity. In our effort to understand the epigenetic changes associated with the leukemic state, the histone pool from B-cells of Chronic

Lymphocytic Leukemia (CLL) patients was isolated and the mass spectrometry analysis was carried out to map the entire spectrum of changes involved in the metastatic state.

The most significant change associated with CLL was the relative decrease in abundance of specific histone isoforms. These variants differ by 1-2 residues from their canonical counterparts. Following this we analyzed the blood samples from 129 CLL patients using real time PCR to correlate the respective downregulation of protein levels with mRNA expression and determine its possible role as a marker of cancer progression. Statistical

ii analysis confirmed that downregulation of some of these isoforms corresponds to an increased time to treatment of CLL from diagnosis. We extended our study to bladder and breast cancers and identified various histone isoforms that are changed in the malignant state. Knockdown studies revealed the distinct role of some of these isoforms in cell proliferation and tumorigenesis as evident by various growth kinetic and proliferation assays. Additionally, our study uncovers various mechanisms by which the histone cluster is regulated at the post-transcriptional level. Therefore, given the complex nature of chromatin and integration of several pathways within, the study reveals another level of complexity that could highlight some unknown aspects of epigenetic regulation.

The outcomes of this study will not only provide the clinicians with novel insights into the cancer biology, but will also bring to light the mechanisms for maintenance and transmission of epigenetic integrity and the disorders during which such integrity is compromised.

iii

Dedicated to my family and friends

Acknowledgments

With great pleasure I would like to thank many people who have helped me throughout my study, and research work at the Ohio State University. First of all, I express my deep sense of gratitude and sincere thanks to my advisor, Dr. Mark Parthun for his guidance, valuable suggestions, intellectual support and encouragement. I was fortunate to have him as a mentor under whose erudite guidance this investigation was carried out. He nurtured my personality to adopt a socio-scientific temperament, which enabled the metamorphosis of my raw skills into scientific expertise. It was his able supervision, advice, and guidance from the very early stage of this research as well as giving me extraordinary experiences throughout the work, which has resulted in fruitful outcome. Above all, his truly scientist intuition has made him as a constant oasis of ideas and passions in science, which inspired and enriched my growth as a student and a researcher. I would specially like to thank Dr Parthun for the enormous effort he has put in to help me improve my writing skills. I feel bereft of words to acknowledge his contribution to shape my academic perceptivity. It would be unfair if I forget the inspiration, help and encouragement from my committee members Drs. Michael Freitas, Matthew Ringel and Robert Snapka during the course of this study. I would like to thank them for all their input and support during my research. Not only their valuable input helped me in my research work, but also in correcting both my stylistic and scientific errors. All of them were easy to approach for their intellectual support. I’d like to thank Dr Snapka for his excellent guidance not only with respect to my dissertation work, but with my future career as well. I’m thankful to Dr Ringel for al his mentoring, encouragement and scientific input. I would like to thank Dr. Freitas for teaching me the mass spectrometry basics and MassLynx software that helped me with the data analysis. I thank my fellow lab mates Prabakaran, Pei and Paula for all their support, help and for making my time in lab enjoyable. I’d specially like to thank Prabakaran for all his help in lab related and non-lab related issues during my time in the lab. He has been a huge support throughout and helped me to get a better understanding of all techniques and the process of science in general. I would also like to thank all past members Devi, Raghuvir, Erica, Huanyu, Zhongqi, Neha and Amy for their support and valuable comments and suggestions during all my presentations. I would like to acknowledge the help from Sean, Bei and Kelly with the mass spectrometry. I’d appreciate their patience and promptness for running the samples as well as data analysis. Additionally I’d also like to thank the support from our departmental faculty members Drs Rafael-Fortney, Schoenberg and Kolb for valuable guidance and allowing me to use their lab resources.

Thanks are also due to all my close friends and colleagues for their untiring help, cooperation and understanding during the progress of this thesis. I would like to thank my friends Cordelia, Dwivedis’, Gargs’, Jhandis’, Meenu, Nan, Naphade, Patels’, Pig, Plappallys’, Rajarapus’, Srivastavas’, Veer and Vinayak for all the partying, loitering and other time-waste activities leading to an unforgettable stay here in columbus. An additional thanks to Meenu for designing an algorithm to reduce time in mass spectrometry data analysis. Not to forget all my friends back home Babloo, Kabir, Rupinder, Sumeet and Vikas for all their support and encouragement throughout. Last but certainly not the least, I am grateful to my parents, my wife Neha and baby Rayansh for their love and constant support for all of my endeavors. Their help and cooperation for all and everything is whole-heartedly acknowledged.

Vita

August 4, 1980 ...... Born-Amritsar, Punjab (India)

1998- 2001…………...... Bachelors in Industrial Microbiology, Guru

Nanak Dev University, Amritsar, Punjab (India)

2001-2003 ...... Masters in Microbiology, Guru Nanak Dev

University, Amritsar, Punjab (India)

2003-2006 ...... Research Associate, Institute of Microbial

Technology, Chandigarh, India

2007- present ...... Graduate Research Associate, Department of

Molecular and Cellular Biochemistry, The Ohio

State University

Publications

1. Rajbir Singh, Amir Mortazavi, Kelly H. Telu, Prabakaran Nagarajan, David M. Lucas, Jennifer M. Thomas-Ahner, Steven K. Clinton, John C. Byrd, Michael A Freitas and Mark R Parthun. Increasing the complexity of chromatin: functionally distinct roles for replication-dependent histone H2A isoforms in cell proliferation and carcinogenesis (Nucleic Acids Research, Accepted, In press)

2. Rajbir Singh, Sean W Harshman, Amy Stark, Amir Mortazavi, David Lucas, Jennifer M. Thomas-Ahner, Steven K Clinton, John C Byrd, Michael A Freitas and Mark

vii

R Parthun. Proteomic profiling identifies specific histone species associated with tumorigeneis (In preparation)

3. Debarati Paul, Rajbir Singh and Rakesh K. Jain. Chemotaxis of Ralstonia sp. SJ98 towards PNP in Soil. (Environmental Microbiology, 2006, 8 (10): 1797-1804).

4. Rajbir Singh, Debarati Paul and Rakesh K. Jain. Biofilms: Implications in Bioremediation. (Trends in Microbiology, 2006, 14 (9): 389-397).

Fields of Study

Major Field: Ohio State Biochemistry Program

viii

Table of Contents

Page

Abstract ...... ii

Dedication ...... iv

Acknowledgments ...... v

Vita ...... vii

List of Tables ...... xi

List of Figures ...... xiii

Chapters:

1. Introduction ...... 1

1.1. Histones and Nucleosome ...... 1 1.2. Histone variants ...... 3 1.3. An introduction to histone isoforms ...... 4 1.4. Histones and cancer ...... 5 1.5. Histone isoforms and cancer ...... 6 1.6. Analytical approaches for histone expression ...... 7 1.7. Histone H2A isoforms and Chronic Lymphocytic Leukemia (CLL) ...... 9 1.8. Trends and Prospects ...... 10 1.9. Research Questions/Hypotheses ...... 11

2. Proteomic Profiling Identifies Specific Histone Species Associated With Tumorigenesis ...... 16

2.1. Abstract ...... 16 2.2. Introduction ...... 17 ix

2.3. Materials and Methods ...... 20 2.4. Results……………… ...... 22 2.5. Discussion ...... 31

3. Increasing The Complexity Of Chromatin: Functionally Distinct Roles For Replication-Dependent Histone H2A Isoforms In Cell Proliferation And Carcinogenesis ...... 68

3.1. Abstract ...... 68 3.2. Introduction ...... 69 3.3. Materials and Methods ...... 73 3.4. Results………………...... 78 3.5 Discussion ...... 85

4. Conclusion and future directions ...... 104

4.1. Conclusion ...... 104 4.2. Future directions ...... 108

Bibliography ...... 112

List of Tables

Table Page

1.1- Histone H2A isoforms with the corresponding protein name and molecular weight ....13

1.2- Histone H2B isoforms with the corresponding protein name and molecular weight .....14

1.3- Histone H3 isoforms with the corresponding protein name and molecular weight ...... 15

1.4- Histone H4 isoforms with the corresponding protein name and molecular weight ...... 15

2.1- Lesser significantly changed histone H2A isoforms with their expression trend in

CLL ...... 43

2.2- Less significantly changed histone H2B isoforms with their expression trend in CLL .44

2.3- Less significantly changed histone H3-1 isoforms with their expression trend in CLL 45

2.4- Less significantly changed histone H3-2 isoforms with their expression trend in CLL 46

2.5- Lesser significantly changed histone H4 isoforms with their expression trend in CLL .47

2.6- Histone H2A isoforms with their expression trend in bladder cancer ...... 48

2.7- Histone H2B isoforms with their expression trend in bladder cancer ...... 49

2.8- Histone H3-1 isoforms with their expression trend in bladder cancer ...... 50

2.9- Histone H3-2 isoforms with their expression trend in bladder cancer ...... 51

2.10- Histone H4 isoforms with their expression trend in bladder cancer ...... 52

2.11- Histone H2A isoforms with their expression trend in breast cancer ...... 53

2.12- Histone H2B isoforms with their expression trend in breast cancer ...... 54

2.13- Histone H3-1 isoforms with their expression trend in breast cancer ...... 55 xi

2.14- Histone H3-2 isoforms with their expression trend in breast cancer ...... 56

2.15- Histone H4 isoforms with their expression trend in breast cancer ...... 57

2.16- Correlation of H2A isoforms with Zap-70 ...... 58

2.17- Correlation of H2B isoforms with Zap-70 ...... 59

2.18- Correlation of H3-1 isoforms with Zap-70 ...... 60

2.19- Correlation of H3-2 isoforms with Zap-70 ...... 61

2.20- Correlation of H4 isoforms with Zap-70 ...... 62

2.21- Histone H2A-Summary results for treatment ...... 63

2.22- Histone H2B-Summary results for treatment ...... 64

2.23- Histone H3-1-Summary results for treatment ...... 65

2.24- Histone H3-2-Summary results for treatment ...... 66

2.25- Histone H4-Summary results for treatment ...... 67

xii

List of Figures

Figure Page

2.1 The complement of histone H2A is altered in CLL patient samples ...... 36

2.2 Histone H2A complement is altered in bladder and breast cancers ...... 37

2.3 The complement of Histone H2B in different cancers ...... 38

2.4 The complement of histone H3 isoforms derived from peak 1 ...... 39

2.5 The complement of histone H3 isoforms derived from peak 2 ...... 40

2.6 The complement of histone H4 isoforms is changed in different cancers ...... 41

2.7 Expression levels of histone H2A isoforms after Ras-mediated transformation of

HFF cells ...... 42

3.1 Replication-dependent histone H2A isoforms are altered in bladder cancer cells ....90

3.2 Immunofluorescent staining using antibodies targeted against canonical H4 and

H2A 1C ...... 91

3.3 Immunofluorescent staining for metaphase chromosomes using antibodies

targeted against canonical H4 and H2A 1C ...... 92

3.4 Distinct roles of replication-dependent histone H2A isoforms in cell proliferation

and carcinogenesis ...... 93

3.5 Histone H2A 1C influences cell proliferation and carcinogenesis in U2OS cells .....95

xiii

3.6 Replication-dependent histone H2A gene expression in healthy individuals and

CLL patients ...... 97

3.7 The 5’ UTR of HIST1H2AC is a regulatory element ...... 98

3.8 Identification of a repeated element in the H2A 1C 5’ UTR necessary for

repressive activity ...... 99

3.9 Stained agarose gel showing a change in restriction digestion pattern after

bisulfite treatment ...... 101

3.10 Amplicon mean values for DNA methylation for normal B-cells, PBLs and CLL

patient samples ...... 102

3.11 H2A 1C and H2A 1B/E share a common amino acid change ...... 103

xiv

CHAPTER 1

Introduction

The history of chromatin dates back to 1884 when Albert Kossel discovered some proteins that were associated with the nucleic acids (1). Originally they were thought to be an initial complication in the isolation of nucleic acids and were ruled out for having any biological significance. It was only in the last few decades that their real significance came to light. Histone machinery plays an important role in all DNA-dependent cellular processes including DNA replication, transcription, repair and recombination. Not surprisingly, many diseases including cancer or autoimmune disorders may result when histones are introduced into the chromatin machinery, either at the wrong time or at the wrong place (2).

1.1 Histones and Nucleosome

Nucleosomes consist of about 146 bp of DNA wrapped 1.65 turns around an octameric histone core, comprised of 2 copies each of histones H2A, H2B, H3 and H4 that are arranged in a central (H3-H4)2 tetramer flanked on either side by two H2A-H2B dimers

(3-5). Nucleosomes are further stabilized by linker histone H1 and assembled into higher order tertiary structure known as chromatin (4,6-8). Overall, there are 14 contact points

1 between histones and DNA, which enable a very stable interaction under physiological conditions (4,9,10). The bulk of histones become highly expressed just prior to S-phase and are then repressed at the completion of DNA replication. They are the only known protein coding mRNAs produced in mammalian cells that lack a poly(A) tail. The core histones have a globular domain and protruding N-terminal tails, both of which are subjected to several post-translational modifications (4,5,7).

There are three known mechanisms that modulate nucleosome structure and chromatin dependent cellular processes. First, a wide range of core histone post-translational modifications, including acetylation, ubiquitination, phosphorylation, methylation, sumoylation and ribosylation, affect histone interactions with non-histone proteins as well as histone- histone and histone-DNA interactions (2,5,11). These modifications alter the structure and dynamics of the nucleosome and to affect chromosome function. They may either alter the electrostatic charge of the histone resulting in a structural change in histones or their binding to DNA, or additionally, may provide binding sites for several protein recognition modules. So far, there are over 60 residues on histones that are known to be modified and eight distinct types of modifications (6,12). Given that there could be multiple modifications at the same site, and different types and sites of modification, a given nucleosome at times could exist in thousands of distinct variations, each with a defined functional significance.

Second, various chromatin remolding complexes can disrupt and remodel the nucleosome structure which increases DNA accessibility (13). Based on their modes of action, two groups of chromatin-modifying complexes have been identified: (i) ATP-dependent 2 complexes, which use the energy of ATP hydrolysis to either move, eject or restructure nucleosomes, and (ii) covalent histone modifications by specific enzymes, i.e., histone acetyltransferases (HATs), deacetylases, methyltransferases, and kinases (14,15). In addition to actively controlling gene expression, chromatin remodeling plays a role in several key biological processes like apoptosis, chromosome segregation as well as DNA replication and repair. Abnormalities in chromatin remodeling proteins are found to be associated with several human diseases and therefore chromatin modifier pathways have been the subjects for their therapeutic potential (16-18). Lastly, the composition of nucleosomes can be altered by substituting the major histones with specialized variants.

1.2 Histone variants

The histone complement as such is not homogeneous but is rather composed of multiple sequence variants. Histone variants can be distinguished on several grounds.

Conventionally, the variants have been defined as the nonallelic fraction of histones that are replication independent and constitutively expressed throughout the cell cycle. They may differ from canonical histones by a handful of amino acid changes to the incorporation of large non-histone domains. Additionally, the replication-independent genes also differ from their canonical counterparts in that they are found as single genes dispersed throughout the genome and they generate transcripts with normal poly(A) tails

(19-21). As expected from their medium to large sequence divergence from core histones, the variants play distinct roles, and are not limited to packaging the DNA. Histone variants are an important contributor to the structural and functional heterogeneity that is 3 essential for the proper regulation of chromatin. Evidence for this in higher organisms comes from the presence of replacement variants that, unlike canonical histones, are expressed throughout the cell cycle and serve as a source of chromatin components needed during repair or recombination of DNA or to replace histones lost through turnover in quiescent cells (22). Their array of functions includes, but is not limited to silencing, antisilencing, DNA repair, tumor suppression and nucleosome stability. Some well-characterized examples of replication-independent histone variants are histones

H3.3, H2A.X, CenH3, H2A.Z and macroH2A (20,23-26). Till date, our definition of variants has been restricted to histones that meet the above-mentioned sets of conditions.

Here, we introduce another category of variants that lies within the histone gene clusters, follow the same set of regulations as canonical histones, and therefore, were considered to be functionally equivalent.

1.3 An introduction to histone isoforms

The histone genes are encoded by several large clusters located on chromosomes 1 and 6, encompassing a total of 72 histone genes. About 80 percent of these genes are encoded by the large cluster on chromosome 6, also known as histone cluster 1 (55 total histone genes). The other clusters of histone genes (histone clusters 2, 3 and 4) encode 12, 4 and

1 histone genes, respectively (27). These 72 genes are known to encode 5 distinct histone subtypes (H2A - 16 genes, H2B - 22 genes, H3 - 14 genes, H4 - 14 genes and H1 - 6 genes) (Tables 1.1-1.4). However, this redundancy is not absolute; many of these genes encode products, which differ in their primary sequence by a few residues. Thus, there is 4 heterogeneity within the histone gene cluster itself. For instance, 16 of the replication- dependent histone H2A genes code for 11 different protein species (Table 1.1). For example Hist1H2AC differs by T16S and K99R substitutions from the canonical H2A.

However, this difference is not insignificant, as even small amino acids have the tendency to disrupt the nucleosomal organization and not surprisingly, many of them have been shown to alter structure and/or functional capabilities of histones (23,28).

Therefore, the presence of distinct replication-dependent histone variants has the potential to significantly expand the complexity of mammalian chromatin structure and needs to be examined. From this point onwards, we will refer this particular subset of variants as ‘isoforms’ to distinguish them from the replication-independent histone variants. Given the fact that chromatin structure and dynamics are altered in many disease stages, it may not be unreasonable to infer that these particular isoforms could have a distinct role in human pathogenesis.

1.4 Histones and cancer

Recent advances have highlighted the mechanisms for how the translation of epigenetic information and the misregulation of these events can lead to cancer development and progression (11,29-31). Epigenetics and epigenetically regulated genes have emerged as potential targets in cancers. According to Daphne (32), a new approach that targets histone modification, termed `histonomics´, could be the next wave in the treatment of cancer. Cancer is a disease of gene expression, which may be modulated by the presence and/or modification of histones and their variants. Indeed, aberrant histone modifications 5 are already known hallmarks of many different types of cancers (33-37). For example, missense mutations of p300 histone acetyltransferases and loss of heterozygosity at the p300 locus are associated with colorectal and breast cancers. In addition, the hypermethylation of CpG islands of tumor-suppressor genes leads to their transcriptional inactivation and the loss of their normal cellular functions, thus contributing to a tumor phenotype. Hypermethylation may also lead to histone deacetylation and chromatin condensation, thereby instigating a transcriptionally silenced state. Intriguingly, tissue- specific and imprinted genes can show loss of DNA methylation, which may contribute to cancer cell phenotypes (11,30,38). Thus, the complexity of epigenetic regulation, may contribute to tumorigenesis in a context-dependent manner. However, it is difficult to functionally address the specific roles of epigenetic modifications in tumorigenesis as primary tissue samples are not genetically tractable systems (11,30). Thus, the dissection of mechanistic details of histone modifications in oncogenesis remains challenging and there is a need for more rigorous methods that are capable of profiling histones present in clinically relevant materials.

1.5 Histone isoforms and cancer

Currently, there is no report highlighting the significance of replication-dependent histone isoforms, although the levels of some of these have been known to be altered in the malignant state (39-41). The levels of Hist1H2AC specifically have been reported to change in many diseases including human papillomaviruses hyperplasias, AIDS, multiple sclerosis and cancer (42-44). In one study, treatment of HCT116 cells with Nutlin-3a, the 6 levels of the tumor-suppressor p53 increases, with a concomitant decrease in Hist1H2AC levels (45,46). Other researchers have found a decrease in the Hist1H2AC levels in MCF-

7 cells treated with estradiol (a potent mitogen) and quercetin (47,48). Additionally, the levels of Hist1H2AC have also been shown to decline following a cell cycle arrest (49).

Taken together, these studies indicate that the levels of Hist1H2AC in the cell may contribute to the control of proliferation rate and/or tumor suppression.

1.6 Analytical approaches for histone expression

Traditionally, histone analysis has relied upon the use of specific antibodies targeted against particular histones, histone variants and their PTMs. However, there are several disadvantages associated with antibody-based methods. In addition to the concern of antibodies cross-reacting with similar modifications on the same histone protein or on a different histone protein, a major problem is the recognition of unmodified histones and/or non-histone proteins with similar modifications (50,51). In the case of highly modified histones, a problem may arise because of epitope occlusion. The latter may be defined as the ability of certain nonintended PTMs to block the intended recognition of a

PTM by a site-specific antibody. The financial constraints involved in manufacturing and validating the antibodies also cannot be overlooked. Additionally, the antibodies have a limitation to differentiate the proteins with a highly similar primary sequence. This was a particular concern for us given that histone isoforms differ by a few residues only as compared to the canonical histones. To avoid these complications, we relied upon the use

7 of mass spectrometry as a tool for analyzing histones in both cell lines and patient samples.

The introduction of mass spectrometry based techniques is one of the most promising developments in the proteomic analysis of histones and their modifications. In addition to compensating for the disadvantages posed by traditional methods, they have added speed, accuracy and high resolving power to biomolecule analysis, thus facilitating a multi- dimensional viewpoint of histone analysis (52-54). Mass spectrometry provides a detailed, more unbiased method for discovering, screening and quantifying histone and non-histone proteins. Coupled with standard biological techniques, MS offers the unique capacity to compare protein and modification expression in the context of 'normal' cells versus the diseased state (55). Nowadays, MS instrumentation and techniques are becoming more and more accessible even to the non-MS-trained scientists. MS also offers the ability to examine the combinatorial nature of PTMs in a high-throughput fashion. Given all the advantages and the reliability, MS-based methods are arguably stronger approaches than immunoassays, and serve as complementary experiments to genomic studies. Matrix Assisted Laser Desorption Ionization (MALDI) and Electrospray

Ionization (ESI) have a wide spectrum of applications, not only because of their ability to provide direct amino acid sequence information and detection of biomolecules with high molecular weights, but also due to their ability to differentiate multiple modifications on a peptide with a same primary sequence (56-59). Therefore, Mass spectrometry based histone analysis has lead to identification of several PTMs not only in the histone tails, but also in the globular core domain of histones (38,57).

1.7 Histone H2A isoforms and Chronic Lymphocytic leukemia (CLL)

CLL is the second most common type of adult leukemia. Patients with CLL suffer from a humoral immune deficiency that becomes progressively worse as the disease advances with complications including autoimmune disorders, secondary cancers, and Richter’s transformation (41,60). Over the past decade, significant effort has been put forth to identify possible biomarkers that could predict disease progression and a possible response to chemoimmunotherapy. Cytogenetic abnormalities and the positive expression of ZAP-70 (zeta associated protein) are most commonly used (60-62), however, each of these markers has its own limitations especially as a single parameter. With the help of our collaborators, we characterized an LC-MS based protein profiling strategy in order to assess histone isoform distributions in primary tumor cells (41). Additionally, our research group carried out extensive method validations to rule out any experimental variances that would impact measurement of histone relative abundance, including sample preparation, processing time and biological variability within individuals. Using

B-cells isolated from 40 CLL patients and 4 healthy volunteers, our group observed a significant decrease in the relative abundance of histone H2A isoforms with molecular weights 14,018 and 14,046 respectively, in primary CLL cells as compared to normal B cells. Further, in order to assess the potential clinical implications of these results and the ability of relative H2A abundance to serve as either a diagnostic or prognostic biomarker for CLL, sensitivity and specificity values for the LC-MS based protein profiling strategy were calculated and compared to analogous values for other reported CLL biomarkers such as ZAP-70 expression, IgVH and p53 mutational status, and genomic profiling. The

9 data suppported a correlation between global chromatin modifications and the CLL phenotype. Bivariate statistical analysis showed a clear demarcation between primary

CLL cells and normal B cells. The downregulation of histone H2A isoforms thus displayed a good potential to be served as prognostic tool. Therefore, we repeated our study with a larger cohort of well-annotated patient samples to corroborate these changes.

1.8 Trends and prospects

The present study aims to develop and validate a quantitative prognostic tool for the conclusive detection and monitoring of cancer development and progression. In a broad manner, the study also determines how chromatin structure affects nuclear processes, and how chromatin structure is regulated to accommodate these processes. The results obtained will set stage for a greater understanding of chromatin assembly and dynamics during replication and transcription. Beyond their interest as histone isoforms, further studies into their functional role may also integrate into a pathway leading to tumorigenesis considering their downregulation in the cancerous state. Additionally, our efforts will help to unveil the position of these epigenetic events within the tumorigenesis cascade in order to gain more insight into their contribution to tumor phenotype.

1.9 Research questions/Hypotheses

Histones are one of the most evolutionarily conserved proteins known and despite the redundancy in histone gene clusters, it is speculated that the histone complement is functionally equivalent and is regulated as a group. However, minor differences do exist at protein levels and it is reasonable to hypothesize that some of these genes may encode for products that are functionally distinct to their canonical counterparts. These differences may also dictate different structural and/or functional properties to the nucleosome. In context of our previous report (41), we can conclude that histone isoform levels may change specifically in cancer. However, the key question of whether the change in abundance is a consequence of the malignant state or if the changes in histone levels mechanistically integrate to carcinogenesis or tumor progression remains elusive.

We hypothesized that specific changes in abundance of histone isoforms could have a defined role in tumorigenesis. Additionally, the depletion of a particular histone isoform could display a phenotype that may be a characteristic of the malignant state.

There are many reports that show a significant correlation between global histone modifications status and clinical outcome for different cancers (36,63), but the histone isoform abundance has never been examined. As we and our collaorators have shown, downregulation of histone H2A isoforms displayed potential to serve as prognostic tool

(41). Therefore, we repeated our study with a larger cohort of well-annotated patient samples to corroborate these changes. Additionally, we extended the scope of our study to test our second hypothesis for if there is some pattern or correlation between histone isoform abundance and cancer progression. To analyze the histone abundance in solid

11 tumor types, we used bladder and breast cancer cell lines. To eliminate the results due to heterogeneity within different cell lines, we used cell lines derived from the same progenitors. For breast cancer, we used M1-M4 cell lines that display linearity in terms of aggressiveness. Similarly for bladder cancer, we used normal bladder epithelium cells and cell lines with an increased aggressiveness and metastatic potential (chapter 1, materials). These two cancers are among the leading causes of female and male deaths worldwide, respectively. Extensive statistical evaluations were carried out to corroborate the results. Our findings are significant and can be applied to identify changes in chromatin that are a function of disease or treatment.

Gene$Name(s)$$ Protein$Name$ Molecular$Weight$ HIST1H2AH$ H2A$1H$ 13,817$ HIST1H2AJ$ H2A$$1J$ 13,847$ HIST2H2AC$ H2A$$2C$ 13,899$ HIST2H2AB$ H2A$$2B$ 13,906$ HIST1H2AG/I/K/L/M$ H2A$ 14,002$ HIST2H2AA$ H2A$2A$ 14,006! HIST1H2AC$ H2A$$1C$ 14,016$ HIST1H2AD$ H2A$$1D$ 14,018$ HIST3H2A$ H2A/3$ 14,032$ HIST1H2AB/E$ H2A$$1B/1E$ 14,046$ HIST1H2AA$ H2A$$1A$ 14,102$

Table 1.1- Histone H2A isoforms with the corresponding protein name and molecular weight

Gene$Name(s)$$ Protein$Name$ Molecular$Weight$ HIST1H2BC$ H2B$$1C$ 13,745$ HIST1H2BK$ H2B$$1K$ 13,801$ HIST1H2BH$ H2B$$1H$ 13,803$ HIST1H2BJ$ H2B$$1J$ 13,815$ HIST1H2BE/F/G/I/O$ H2B$ 13,817$ HIST3H2BB$ H2B$$3B$ 13,819$ HIST2H2BE$ H2B$$2E$ 13,831$ HIST1H2BN$ H2B$$1N$ 13,833$ HIST1H2BD$ H2B$$1D$ 13,847$ HIST1H2BB$ H2B$$1B$ 13,861$ HIST1H2BL$ H2B$$1L$ 13,863$ HIST1H2BM$ H2B$$1M$ 13,900$ HIST1H2BA$ H2B$$1A$ 14,079$ HIST2H2BA$ H2B$$2A$ N/A$ HIST2H2BB$ H2B$$2B$ N/A$ HIST2H2BC$ H2B$$2C$ N/A$ HIST2H2BD$ H2B$$2D$ N/A$

HIST3H2BA$ H2B$$3A$ N/A$

Table 1.2- Histone H2B isoforms with the corresponding protein name and molecular weight

Gene$Name(s)$$ Protein$Name(s)$Molecular$Weight$

HIST2H3A/B/C$ H3.2$ 15,298$

HIST1H3A/B/C/D/E/F/G/H/I/J$ H3.1$ 15,315$

HIST3H3$ H3.1T$ 15,419$

Table 1.3- Histone H3 isoforms with the corresponding protein name and molecular weight

Molecular$ Gene$Name(s)$$ Protein$Name(s)$ Weight$ HIST1H4G$ H4$1G$ 10,920$

HIST1H4I$$ H4$1I$$ 11,250$ $ HIST1H4A/B/C/D/E/F/H/J/K/L$ HIST2H4$ H4$$$ 11,278$ HIST4H4$ $

Table 1.4- Histone H4 isoforms with the corresponding protein name and molecular weight

CHAPTER 2

Proteomic Profiling Identifies Specific Histone Species Associated with

Tumorigenesis

2.1 Abstract

Alterations in the patterns of histone post-translational are a hallmark of many types of cancer. We have used a liquid chromatography/mass spectrometry-based approach to comprehensively characterize the histone proteome in primary tumor samples from chronic lymphocytic leukemia (CLL) patients and bladder and breast cancer tissue culture models. In general, the CLL patient samples had histone profiles that were distinct from those of the bladder and breast cancer cells. When compared to CD19+ B cells from healthy donors, the CLL histone proteome showed the highest degree of variability in histones H2A and H2B. This variability was in both post-translationally modified species and in unmodified, non-allelic replication-dependent histone isoforms.

The presence of a modified species of H2A with a mass of 14,063 Da was associated with a significantly shorter time to treatment in CLL patients. Contrary to global histone analyses reported from other tumor types, CLL patients showed no significant changes in the methylation or acetylation of histones H3 and H4 suggesting that alterations in these modifications are not a common feature of all tumor types.

2.2 Introduction

The packaging of eukaryotic genomes with histones to form chromatin is essential for the necessary condensation and protection of DNA. Chromatin structure also plays a critical role in the regulation of cell proliferation through the modulation of gene expression and epigenetic inheritance. The regulation of chromatin structure occurs through three basic mechanisms; ATP-dependent chromatin remodeling, histone post-translational modifications and histone variant incorporation. Not surprisingly, mutations that influence each of these mechanisms have been found to be associated with cancer.

Given the central role of histones in chromatin, there have been a number of studies that have examined histone proteins to identify alterations, particularly post-translational modifications, which correlate with cancer. These studies have identified several post- translational modifications whose abundance are altered in cancerous cells or, importantly, has prognostic value. For example, decreased levels of histone H4 lysine 16 acetylation (H4 K16Ac) and H4 lysine 20 trimethylation (H4 K20me3) are found in a number of human cancers, while low global levels of histone H3 lysine 9 dimethylation

(H3 K9me2) and H3 lysine 18 acetylation (H3 K18Ac) predict poorer outcome in prostate, lung, kidney and pancreatic cancer (36-38,63-68).

Studies of global levels of histone species associated with cancer utilize two basic approaches, each with its own advantages. The most common method of analysis involves the use of modification-specific antibodies. The use of antibodies has the advantage of being able to roughly quantitate changes in site-specific modifications.

Importantly, using modification-specific antibodies for immunohistochemical staining of 17 tumor tissues allows for single cell resolution of histone modification changes. The second method of analysis is liquid chromatography coupled with mass spectrometry

(LC/MS). LC/MS benefits from being an unbiased approach that is not dependent on the existence of antibodies whose specificities are not completely characterized.

In addition, LC/MS is capable of shedding light on a layer of chromatin complexity that is not visible with modification-specific antibodies. Variations in the histone proteins are typically considered to be the result of the incorporation of replication-independent histone variants, such as histone H2AZ and histone H3.3. However, the replication- dependent histones also consist of multiple distinct polypeptides. Each of the replication- dependent histones is encoded by multiple genes that are found in several large clusters in the human genome (27,69). Interestingly, the multiple replication-dependent histone genes do not all encode identical proteins. For example, there are 16 genes that code for replication-dependent histone H2A and they produce 11 distinct polypeptides. These histone species will be referred to as histone isoforms to distinguish them from the more familiar histone variants.

Given the complicated nomenclature that has arisen for the replication-dependent histones, we will use a systematic nomenclature based on the gene name for each isoform. The replication-dependent histone genes are named based on their identity and their location in the genome. The first part of the name refers to the histone cluster (e.g.

HIST1, HIST2 or HIST3), the second part indicates the type of histone (e.g. H2A, H2B,

H3, H4 or H1), and finally, the multiple copies of each histone type are designated alphabetically based on their order within each cluster (centromere distal to proximal). 18

Hence, HIST1H2AC refers to the third histone H2A gene in histone cluster 1. We will refer to the protein isoforms based on this genomic nomenclature. For example, the product of the HIST1H2AC gene will be referred to as H2A 1C (for HIST1H2AC).

Likewise, the 14,046 Da form of H2A is encoded by two genes, HIST1H2AB and

HIST1H2AE and will be called H2A 1B/E. For each of the replication-dependent histones, there is one protein form that is encoded by at least 5 genes. As these isoforms are typically the most abundant and thought of as the canonical histones, they will simply will be referred to as H2A, H2B and H4 (see tables 1.1-1.4 for complete list of genes/protein names). The exception to this nomenclature will be histone H3, where the existing nomenclature (H3.1, H3.2, H3.3 and H3.1t) is well-established (70).

The replication-dependent histone isoforms typically vary from each other by a small number of amino acids. This high degree of identity makes them very challenging to study using typical techniques that would require the generation of isoform-specific antibodies. However, they can readily resolved by LC/MS due to the change in mass

(Tables 1.1-1.4). In fact, a recent report identified alterations in specific histone H2A isoforms in B cells isolated from patients with chronic lymphocytic leukemia (CLL) (41).

We have used LC/MS to quantify the relative abundance of every detectable histone specifies in a large set of primary tumor samples. By comparing the histone profiles between normal B-cells and B-cells isolated form CLL patients, we have identified a number of specific histone species whose levels are significantly altered in the CLL cells.

We have also comprehensively analyzed histone profiles from cell culture models of bladder and breast cancer to determine whether histone alterations are specific to CLL or 19 whether they are a more general feature of tumorigenic cells. In addition, correlation of the histone profiles with detailed clinical data has allowed us to identify a specific histone

H2A species (14,062 MW) whose presence is indicative of a poor prognosis in CLL.

2.3 Materials and methods

Cell lines and culture conditions- M1-M4 cells were a gift from Dr. Tsonwin Hai. The

M1 cells are immortalized but non-transformed human breast epithelial cells; M2 cells are Ras-transformed MCF10A cells selected by neomycin, M3 cells are Ras-transformed

MCF10A cells, malignant but not metastatic breast epithelial cells and M4 cells are Ras- transformed MCF10A cells which are malignant as well as metastatic. The cells were grown in DMEM/F-12 with 5% horse serum (Invitrogen), 0.029 M sodium bicarbonate

(Sigma), 10 mM HEPES (Sigma), 10 µg/ml insulin (Sigma), 10ng/ml EGF (Millipore),

0.5 µg/ml hydrocortisone (Sigma), 100 ng/ml cholera toxin (Calbiochem) and 1% penicillin/streptomycin (Sigma). Bladder cancer cell lines were a gift from Amir

Mortazawi’s lab. hTERT cells are immortalized breast epithelial cells, RT4 cells are transformed but non-malignant, T24 cells are malignant but non-metastatic and UM-UC-

3 cells are malignant and highly metastatic. The cells were grown in DMEM, supplemented with 10% FBS and 1% penicillin/streptomycin (Sigma). All the cell types as described above were incubated at 37 °C in a humidified atmosphere with 5% CO2 and

95% air.

Mass Spectrometry- Histones extracted from primary CLL cells and cell lines were subjected to LC-MS analysis. Characterization was performed by HPLC separation

(Dionex, Waltham, MA) in line with a MicroMass Q-TOF (MicroMass, Milford, MA) mass spectrometer. Approximately 20 μg of histones were separated on a 1.0 x 150mm

C18 column (Discovery Bio wide pore C18 column, 5 μm, 300 Å, Supelco, USA) with a modified gradient described by Wang et al (71). Mobile phase A contained 0.05 % TFA

(Pierce, Rockford, IL) in HPLC water (J.T. Baker, Center Valley, PA). Mobile phase B contained 0.05 % TFA in acetonitrile (EMD Millipore, Billerica, MA). Starting at 20 %

B, the gradient increased linearly to 30 % at 2 minutes, 35 % at 10 minutes, 50 % B at 30 minutes, 60 % at 35 minutes and 95 % at 36 minutes. The column was held at 95 % B for

4 minutes. Column equilibration was conducted for 15 minutes at 20 % B. Peaks corresponding to the histone H1’s (H1.5 RT: 24.5-25.2, H1.2-H1.4 RT: 26.9-27.6) were deconvoluted by the MaxEnt algorithm and analyzed using the MassLynx software 4.0

(Waters Corp., Milford, MA). Statistical analysis was performed using Prism GraphPad software (GraphPad Software Inc., La Jolla, CA).

Statistical Analysis- For each histone, the expression for a particular fragment was measured relative to the total expression across all fragments. For each fragment within a histone, the nonparametric wilcoxon rank sum test was used to compare 1) the proportion expressed between normal donor and CLL patient samples and 2) the proportion expressed between patients samples with CLL and with positive Zap70 protein expression (>20%) versus patient samples with CLL and negative Zap70 protein expression (<20%). Time to treatment was measured from the date of diagnosis until the date of first treatment, censoring patients who had not yet started treatment at the date of last follow-up. Associations between expression of a fragment (using median expression

21 to group patients as high and low expressers) and time to treatment were screened using univariable proportional hazards models; hazard ratios with confidence intervals were estimated from the model. Differences in time to treatment between high and low expressers were shown graphically using Kaplan-Meier plots. All tests were 2-sided and statistical significance was declared for any fragment with p<0.05. To strongly control type I error, p-values from tests within a histone were adjusted using Holm’s method. For comparison of multiple columns in bladder and breast cancer samples, one-way ANOVA was used.

Ras transformation- The plasmids for activating the Ras pathway and negative control were a gift from Dr Tsonwin Hai’s lab. For immortalization of the cells, SV40 T-antigen immortalization plasmid was used. Each of the plasmids was transfected into cell lines and Human Foreskin Fibroblast (HFF) cells using Lipofectamine (Invitrogen). Histone extraction and mass spectrometry was carried out as described previously (72).

2.4 Results

Changes in global levels of histone H2A species in tumor tissues and models

Chronic lymphocytic leukemia. Histones were isolated from CLL patient samples

(n=96) or CD19+ cells from healthy volunteers (n=5) by acid extraction. The extracted histones were analyzed on an ESI LC-TOF MS coupled with a Waters HPLC. For each core histone peak eluting from the LC, we selected a specific mass range that encompassed all of its isoforms in an unmodified state as well as the possible spectrum of modifications that could occur. We then correlated the relative abundance of each of

22 each molecular species in that mass range with clinical parameters in CLL patient samples and disease aggression in cell lines.

The unmodified replication-dependent histone H2A isoforms have masses that range from 13,817 to 14,102 (Table 1.1). A spectrum of H2A from a representative normal

CD19+ B-cell sample is shown in Figure 2.1-A. Each of the peaks in the spectra represents a histone isoform or a post-translationally modified histone isoform with a defined molecular weight. In our previous study, we focused on the three most abundant forms of H2A, which have molecular masses of 14,002 Da, 14,018 Da and 14,046 Da

(41). Using our nomenclature, we refer to these forms as H2A, H2A 1C and H2A 1B/E, respectively. In this previous collection of patient samples, we observed that H2A 1C and H2A 1B/E were found at decreased levels in a high percentage of the patient samples. Using a significantly larger pool of patient samples, we have found that a decrease in the relative abundance of H2A 1C and H2A 1B/E is not consistently observed

(Figure 2.1-B). The levels of H2A 1C and H2A 1B/E were highly variable and we observed both increased and decreased levels of these H2A isoforms in the CLL patient samples relative to the healthy controls.

Examination of all detectable H2A isoforms identified several that showed statistically significant changes in abundance in the CLL patient samples (Figure 2.1-C and Table

2.1). There was a greater than 2-fold decrease in several H2A isoforms. Based on the molecular masses of these peaks, it is not possible to definitively determine the identity these species but they are highly modified, containing methylation and acetylation, as

23 well as the possibility of phosphorylation. In addition, there were modest increases in the unmodified forms of canonical H2A and H2A 1H.

The most significantly altered isoform of H2A had a molecular mass of 14,063 Da

(Figure 2.1-C and 2.1-D). This mass is compatible with post-translationally modified versions of several H2A isoforms. For example, this mass is consistent with the H2A 1H

Me3Ac3Phos1, H2A 2C Me3Ac1Phos1 or H2A 2A Me1Ac1. The behavior of this species was intriguing. This species was easily detectable in all of the healthy controls samples (Figure 2.1-D). For the CLL patient samples, there appeared to be two distinct populations. The majority of CLL patients had completely undetectable levels of this species while a small population had normal levels. Intriguingly, when these two populations were compared using the clinical data available for these samples, a clear difference was observed where the patients that had undetectable levels of this H2A species had a statistically significant increase in the time to treatment. No other H2A isoform showed a statistically significant correlation with any other clinical parameter, including Zap70 status (Tables 2.16 and 2.21).

Bladder and breast cancer models. To determine which changes in histone profiles observed in CLL patients are disease-specific and which might be more generally indicative of a cancerous state, we characterized the complement of histone proteins in two solid tumor model systems. The first is a set of cell lines that represent a range of bladder cancers. The cells that we have examined are normal bladder epithelium, normal bladder epithelial cells that have been immortalized with hTERT, a cell line derived from a non-invasive bladder cancer (RT4) and a cell line derived from an invasive bladder 24 cancer (T24). Several H2A species displayed statistically significant changes across these cell lines (Figure 2.2-A). The most apparent change was in the levels of H2A 1C, which dropped dramatically when the normal bladder epithelial cells became immortalized. Decreased levels of this isoform were previously observed in a set of CLL patient samples suggesting that a subgroup of multiple tumors may involve a downregulation of this H2A isoform (41).

Two other, less abundant, H2A species also decreased in abundance as the bladder cells became more tumorigenic, including unmodified H2A 2C and an acetylated/methylated form of H2A 1J. In addition, one species, unmodified H2A 1J, displayed increased levels as the cells became more tumorigenic (Figure 2.2-B). The H2A species with a molecular mass of 14,063 Da whose presence was indicative of a poor prognosis in CLL, did not show significant changes in the bladder cancer cells suggesting that changes in this species may be disease-specific (Table 2.6).

We also the analyzed the histone complement in a set of breast cancer cell lines, designated M1, M2, M3 and M4, that represent a spectrum of subtypes found with this disease, with M1 representing immortalized normal breast epithelium and M3 and M4 representing highly aggressive metastatic disease (73). There are a number of H2A species that vary among these cell lines (Figure 2.2-C and Table 2.11). Interestingly, the levels of H2A 1C and H2A 1B displayed a statistically significant increase as the cells become more tumorigenic. Changes in less abundant H2A species were also observed, such as the mono-methylated form of H2A 1J which decreased in abundance as the breast cancer cells became more aggressive (Figure 2.2-D). 25

Changes in global levels of histone H2B species in tumor tissues and models

Chronic lymphocytic leukemia. Replication-dependent histone H2B genes also encode for a large family of distinct polypeptides, with 22 genes yielding a total of 18 proteins

(Table 1.2). A typical spectrum of histone H2B from healthy CD19+ B cells is shown in

Figure 2.3-A. While a number H2B species displayed altered abundance in the CLL patient samples, these species were all very low abundance species that are also highly modified (Figure 2.3-B and Table 2.7). In addition, the changes observed in these minor peaks did not correlate with any of the clinical parameters or outcomes in the CLL patient population (Tables 2.17 and 2.22).

Bladder and breast cancer models. In contrast to the CLL patient samples, there were significant changes in the relative abundance of several of the most abundant replication- dependent H2B in both the bladder and breast cancer cell culture models (Figure 2.3-C and 2.3-D). In the bladder cancer cell lines, there was a progressive increase in the H2B species with a M.W. of 13,775 Da, which represents the dimethylated form of H2B 1C, as the cells became more aggressive. The histone H2B profile in the breast cancer cells also showed changes in the abundance of methylated forms of H2B 1B with the mono- methylated form increasing when the cells became tumorigenic and the di-mthylated form decreasing in the highly aggressive M4 cells (Figure 2.3-D and 2.3-E). Also of note is that the profile of the replication-dependent H2B in the normal bladder and breast epithelium differs dramatically from the profile of the normal B cells. This suggests that there may be important tissue-specific changes in the complement of this histone.

Changes in global levels of histone H3 species in tumor tissues and models 26

The global analysis of histone H3 is complicated by a number of factors. First, under the

LC conditions used, H3 elutes in two distinct peaks (data not shown). Hence, spectra from these peaks have been analyzed separately. These populations of histone H3 will be referred to as peak 1 (retention time ~36 minutes) and peak 2 (retention time ~42 minutes) based on the order in which they elute from a C18 column. Second, histone H3 is the most highly modified of the core histones. In particular, the high degree of methylation on histone H3 leads to the appearance of multiple peaks that are separated by

14 Da (Figure 2.4-A and 2.5-A).

Histone H3 Peak 1:

Chronic lymphocytic leukemia. A representative spectra of peak 1 histone H3 is shown in Figure 2.4-A. We observed down-regulation of several peak 1 histone H3 species in the CLL tissue samples (Figure 2.4-B and Table 2.3). Intriguingly, while changes in H3 acetylation and methylation have been linked to a number of cancers, the most significantly down-regulated species in peak 1 were also phosphorylated (Figure 2.4-B and 2.4-C). However, changes in the abundance of these isoforms did not correlate with any clinical parameters (Tables 2.18 and 2.23).

Bladder and breast cancer models. Comparison of the peak 1 histone H3 profiles in the series of bladder cancer cell lines indicated that these histones were largely unchanged. Only a single peak 1 H3 species, unmodified H3.2, had a significant change in abundance with its abundance increasing in the most aggressive bladder cancer cell lines (Figure 2.4-D, 2.4-F and Table 2.8). The majority of the peak 1 histone H3 species were also unchanged in the breast cancer cell lines (Figure 2.4-E, 2.4-G and Table 2.13). 27

The most significantly altered species was a decreased level of the mono-methylated, mono-phosphorylated form of the lone histone H3 isoform found in histone cluster 3

(H3.1t, Figure 2.4-D). Intriguingly, the abundance of this species, which was originally thought to be testis-specific, was also significantly down-regulated in the CLL patient samples (74,75).

Histone H3 Peak 2:

Chronic lymphocytic leukemia. The peak 2 population of histone H3 contained an entirely different complement of species and, based solely on the masses observed, there is no obvious explanation of why this pool of molecules has a distinct LC elution profile

(Figure 2.5-A). When comparing peak 2 H3 from healthy and CLL B cells, there was only a single species that showed a statistically significant change in abundance (Figure

2.5-B and Table 2.4). Again, this was a phosphorylated form of H3.1t but, in this case, it was also di-methylated. As was the case for the peak H3 species, this modification did not correlate with any clinical parameters (Tables 2.19 and 2.24).

Bladder and breast cancer models. There was also only a single peak 2 histone H3 species whose abundance was significantly altered in the bladder cancer cell lines relative to the normal bladder epithelium (Figure 2.5-C and Table 2.9). The unmodified form of histone H3.2 was present at relatively low levels in the normal bladder epithelium and displayed an increase in abundance in the bladder cancer cell lines. In the breast cancer cell lines, there was a statistically significant increase in the levels of the most abundant peak 2 H3 species. These were the methylated and methylated/acetylated forms of histone H3.1 (Figure 2.5-D and Table 2.14). 28

Changes in global levels of histone H4 species in tumor tissues and models

Chronic lymphocytic leukemia. Among the core histones, histone H4 contains the least variation. There are no replication-independent forms of histone H4 and of the 14 replication-dependent H4 genes, 12 encode identical protein products (Table 1.4). The other two genes encode minor, but detectable, forms of H4 (Table 1.4 and Figure 2.6-A).

Similar to the situation with H3, comparison of healthy and CLL B cells showed that H4 levels are remarkably unchanged, with only two very low abundance H4 species showing a statistically significant change (Figure 2.6-B and Table 2.5). Therefore, unlike several solid tumors, changes in the post-translational modification stats of histones H3 and H4 are unlikely to play a significant role in CLL (Tables 2.20 and 2.25).

Bladder and breast cancer models. Relative to the normal bladder epithelial cells, several relatively low abundance H4 species were altered in the bladder cancer cells

(Figure 2.6-C and 2.6-E). In contrast, there were dramatic changes in high abundance peaks as the breast cancer cell lines became more aggressive (Figure 2.6-D and 2.6-F).

Most apparent was the increased abundance of the di-methylated and acetylated form of

H4. There were also increased amounts of forms of H4 that contained higher levels of acetylation (Figure 2.6-D and 2.6-F). While these acetylated H4 species clearly correlate with disease progression, it will be important to determine whether they play a causative role.

We have performed a comprehensive proteomic analysis of the core histones from

CD19+ B cells from healthy individuals and CLL patient samples as well as from bladder and breast cancer cell line models. Using LC/MS, we have quantitated the relative 29 abundance of every detectable histone peak to identify histone species whose abundance correlates with the presence or severity of disease. Our results identify a specific histone

H2A species whose absence correlates with an increased time to treatment in CLL patients. In addition, we demonstrated that the abundance of the vast majority of histone species do not significantly change in primary tumor samples and cell model systems suggesting that at least some malignancies are not accompanied large-scale alterations in the global histone proteome.

Histone H2A 1C levels change after Ras-mediated transformation

As the downregulation of histone H2A isoforms was evident in bladder cancer especially after immortalization of the cells (Figure 2.2 and 3.1-B), we were interested in determining if the immortalization or transformation of normal cells would lead to the same fate. To address this question, we used SV40 T-antigen plasmid to immortalize the primary HFF cells and additionally, we carried out Ras-mediated transformation of the

HFF cells. We hypothesized that if in bladder cancer samples, the expression of H2A isoforms goes down in a linear manner with the cancer aggression, the immortalized and transformed HFF cells should also show a decreased H2A isoform levels. Additionally, we also transfected a set of one immortalized (293T) and one cancer cell line (HeLa) to check if Ras-transformation would lead to an increased aggression in these cell lines with a concomitant change in the levels of histone H2A isoforms.

Surprisingly, we observed an increase in the levels of H2A 1C as the HFF cells become aggressive. The levels of other histone isoforms either do not change significantly or

30 change randomly (Figure 2.7). For the 293T and HeLa cell lines, the H2A isoform levels remain unchanged (data not shown)

2.5 Discussion

A number of studies have identified global changes in histone post-translational modifications that correlate with the presence of specific tumors or have prognostic value in patients with specific cancers (63-66,76-80). These studies have led to the idea that global changes in histone post-translational modifications are a common feature of most human malignancies (68). Our comprehensive analysis of histone proteomics indicates that CLL is distinct from other tumor types studied in a number of respects.

Previous global analyses of histones in cancer have focused on the post-translational modification patterns of histones H3 and H4. These studies have identified several specific modifications that strongly correlate with tumorigenesis. In particular, decreases in the levels of H4 lysine 16 acetylation and lysine 20 methylation and H3 lysine 18 acetylation, lysine 4 methylation and lysine 9 methylation have been seen in a large number of tumor types and tumor cell models (63-66,76-80). However, we have found that the abundance of the post-translationally modified forms of H3 and H4 highly stable across a large number of CLL patient samples with most of the changes that we seen limited to relatively low abundance species. This suggests that alterations of H3 and H4 are not universally conserved hallmarks of cancer.

The lack of change in the patterns of H3 and H4 modification inn CLL patient samples may have a number of explanations. From a biological perspective, CLL may be distinct from the other tumor types that have been explored, such as prostate and lung cancer, in that the CLL cells are not highly proliferative. This would suggest that decreases in H3 and H4 acetylation and methylation could be more related to cell proliferation. There may also be technical reasons that could prevent our analyses from detecting changes in

H3 and H4 modification. Our study has relied on mass spectrometry rather than immunological methods. This allows us to measure the relative abundance of a specifically modified histone species but does not allow s to localize the modifications.

Hence, if there were a decrease in the level of H4 lysine 16 acetylation in the CLL cells but a concomitant increase in the acetylation of another lysine residue on H4, there would be no net change in acetylation detected. In addition, heterogeneity in the levels of modification in a population of cells can be detected by immunocytochemistry but may not be as detectable when the histones are collectively examined by mass spectrometry.

We also observed relatively minor changes in the pattern of histone H3 in the bladder cancer and breast cancer tumor models. Interestingly, an increase in the unmodified form of histone H3.2 was observed in both chromatographic peaks of H3 from the more highly tumorigenic bladder cancer cells. H3.2 is nearly identical to H3.1 and, as yet, there has been no unique function ascribed to this histone variant.

There was a more dramatic change in the profile of histone H4 in the breast cancer cell lines. There was a much higher level of the 11,349 Da species in the M4 cells relative to the M1 cells. Based on this molecular weight, this species is likely to be the di- 32 methylated/mono-acetylated form of H4. Based on previous analyses, the majority of this species typically contains H4 lysine 20 dimethylation and lysine 16 acetylation (38).

Progressive increases in this methylated/acetylated species of H4 as the breast cancer cells become more aggressive are contrary to immunohistochemical analysis of breast tumor samples. This difference may be due to the inability of in vitro models to accurately recapitulate in vivo tumors or may be due to differences in the types of changes that can be detected by immunohistochemistry relative to mass spectrometry described above.

One of the most important reasons for the use of mass spectrometry to characterize the histone proteome is that it is an unbiased approach that does not rely on the existence of modification-specific antibodies. As there are relatively few immunological reagents specific for the characterization of post-translational on histones H2A and H2B, our analysis is the most comprehensive study of these histones in cancer, to date. Our results indicate that the patterns of histones H2A and H2B found in both primary tumor tissue samples and cancer cell lines models is highly dynamic. These changes were observed both between healthy and tumor cells and between different types of tumors. For example, the pattern of H2B observed in B cells was very different from that observed in the breast cancer cell lines. This suggests the possibility that there may be tissue specific variations in the patterns of histone H2A and H2B species.

The dynamic nature of histone H2A is not limited to its pattern of post-translational modifications. We also detect a wide range of variation in the abundance of specific replication-dependent isoforms of histone H2A. While the functional analysis of these 33 distinct replication-dependent H2A isoforms is only beginning, the observation that their abundances can be altered in primary tumor tissue raises the intriguing possibility that they may encode functionally distinct molecules (72).

The most clinically relevant species in the histone complement of CLL cells may be that with a molecular mass of 14,065 Da. The CLL patient samples could be divided into two groups based on the presence of absence of this species. Those patients who contained a detectable level of this species had a significantly shorter time to treatment that this species correlates with more aggressive disease. While this species may serve as a prognostic biomarker for CLL, identification of the molecular make-up of this species will be needed to determine whether this form of histone H2A plays a functional role in disease progression.

In addition, we did cell transformation assays to find out if the fluctuating levels of these isoforms in cancer can be reproduced artificially in the lab. Based on our bladder cancer data, our initial hypothesis was that if the levels of Hist1H2AC are changing in a linear manner to cancer aggression, it could have been playing a role in inhibiting cell proliferation. However, for HFF cells, we observed an increase in H2A 1C levels with cancer aggression. A possible explanation is that the H2A 1C levels are induced in response to cancer to inhibit cell proliferation but some intracellular pathway nullifies their activity or their effect is by counterbalanced by some proliferation-associated protein in the cell. This may or may not be the case, given that there are multiple levels of control involved and complex integration of cancer pathways. Another possibility is that

H2A 1C may be playing a different role in different cancers. Also, we observed that 34

Hist1H2AC levels don’t change after the Ras-transformation of various cancer cell lines.

This indicates that H2A isoforms may have a threshold with cancer aggression, beyond which their expression level does not change. Nevertheless, the study is still in infancy and needs supporting data from CHIP-seq and microarray studies to pinpoint the exact functional role of these isoforms.

A" 14003 14018

14046

13847 14033 14065 13817 13899 13930 13834 13996 14132 14197

13800 13840 13880 13920 13960 14000 14040 14080 14120 14160 14200 B" C" H2A"1C" H2A"1B/1E"

D" E"

1.00& & Raw$P$=$0.0004$ & Adjusted$P$=$0.015$ && 0.75& & & & 0.50& H2A&14063&=&0&(n&=&55)& & & && 0.25& & & SURVIVAL&DISTRIBUTION&FUNCTION& & H2A&14063&>&0&(n&=&9)& 0.00& 0.0 &&&2.5 &&&&&&&5.0 &&&&&&&&&&&7.5 &&&&&&&&&&&&&10.0 &&&&&&&&&&&&&&&&&12.5&&&&&&&&&&&&&&&&&&&15.0 &17.5& YEARS&FROM&DIAGNOSIS&

Figure 2.1- The complement of histone H2A is altered in CLL patient samples. A) A representative spectra of histone H2A from CD19+ B cells isolated from a healthy volunteer. The peaks that are significantly changed in CLL, bladder and breast cancers are highlighted with the appropriately colored arrows. B) Dot plots showing the relative abundance of H2A 1C and H2A 1B/1E in normal vs CLL patient samples. C) A list of most significantly changed H2A species in CLL. A double arrow represents a species that was upregulated or downregulated by at least 2-fold. A single arrow represents a species changed by <50%. Based on the molecular weight of each peak, the potential identities are listed. The adjusted P-values were obtained after Bonferroni corrections. D) Representative spectras showing the presence or absence of the peak with molecular weight 14063 Da in CLL patient samples. E) KM curve showing the comparison of time to treatment in patients with and without the 14063 Da peak.

A) C) EXPRESSION) MOLECULAR) EXPRESSION) MOLECULAR) EXPRESSION) GENE)IDENTITY) P8VALUE) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) WEIGHT) TREND) NORMAL)) 13817& H2A&1H& 0.004& 13847& H2A&1J& ! 0.001& 13832& H2A&1H/&Me& 0.003& 13899& H2A&2C& " 0.001& 13847& H2A&1J& 0.006& 13932& H2A&1J/&Me3Ac/Ac2& ABSENT&IN&NORMAL& 0.015& 13932& H2A&1J/&Me3Ac/Ac2& 0.001& 14018& H2A&1C& " 0.037& 14018& H2A&1C& 0.001& H2A&1D/&Me&&H2A&2A/&Me2&&&H2A& B) 14032& 2B/&Me3Ac2/Ac3&&Hist3H2A& <0.001& 14045& H2A&1B& <0.001& D) )))))hTERT) ) ) ) M1) ) ) )))))))))RT4) ) ) ) ) ) ) M2) ) ) )))))))))T24) ) ) ) ) ) ) ) )) M3) )UM)UC3) ) ) ) ) M4)

Figure 2.2- Histone H2A complement is altered in bladder and breast cancers. A) List of most significantly changed species in bladder cancer. The expression level shown is relative to the normal bladder epithelium cells. Expression trend displays the pattern of expression among different cell lines starting from immortalized to most metastatic. Based on the molecular weight of each peak, the potential identities are listed. B) Representative spectra of one of the H2A isoforms that increases in abundance as the cancer aggressiveness increases. C) List of most significantly changed H2A species in breast cancer. D) The same isoform that increases in bladder cancer shows a decreased abundance in breast cancer cell lines. One way ANOVA was applied to test significance. Alterations in histone species are summarized by the following key. / - The isoform increases or decreases in a linear manner to disease aggression. / - the expression level of isoform is changed in first or second half of cell lines. / - the expression level of the isoform is changed in the least or most aggressive cell line only and remains constant for the rest of the group.

A) 13775

13759

14048 14056 13902 13961 13987 1402114032

13700 13750 13800 13850 13900 13950 14000 14050 14100 14150 14200 B) E) Molecular) Expression) Raw)P) Adjusted)P) weight) Histone)idenHty) Trend)(relaHve) to)normal)) 14032& H2B&1L+&Me3Ac3& "" <.0001& 0.0026& H2B&1B+&Me3Ac2/Ac3&& 13987& H2B&1D+&MeAc3&&H2B&1N+&Me2Ac3&& " 0.0002& 0.0086& M1) H2B&1L+&Me3Ac3&&&&& ) H2B+&Me3Ac/Ac2&&&H2B&1B+&Me3/Ac&&& ) H2B&1D+&MeAc&& ) 13902& 0.0003& 0.0121& &H2B&1H+&MeAc2&&&H2B&1N+&Me2Ac&&& "" ) H2B&2E+&Me2Ac&& ) &H2B&3B+&Me3Ac/Ac2& H2B&1J+&P&Me2Ac3&&H2B&1K+&P" ) 14048& " 0.0009& 0.0328& Me3Ac3&&& M2) H2B&1L+"P"Me2Ac2&&&H2B&1M+& ) 14055& !! 0.0009& 0.0317& Me2Ac3& ) H2B&1J+&P&Me3Ac2/P&Ac3&&&H2B&1M+"P" ) 14021& !! 0.0011& 0.0388& Me3/P&Ac&&&H2B&1K+"P"MeAc3& ) ) C) MOLECULAR) EXPRESSION) GENE)IDENTITY) P5VALUE) ) WEIGHT) TREND) M3) 13759& H2B&1C+&Me& <0.001& ) 13775& H2B&1C+&Me2& <0.001& ) 13961& H2B&1L+&MeAc2& 0.038& ) D) ) EXPRESSION) MOLECULAR) EXPRESSION) ) GENE)IDENTITY) (RELATIVE)TO) P5VALUE) WEIGHT) TREND) M4) NORMAL)) 13775& H2B&1C+&Me2& ! 0.031&

Figure 2.3 - The complement of Histone H2B in different cancers. A) A representative spectra of histone H2B from CD19+ B cells isolated from a healthy volunteer. The peaks that are significantly changed in CLL, bladder and breast cancers are highlighted with appropriately colored arrows. B) List of most significantly changed isoforms in CLL. The adjusted P-values were obtained after Bonferroni corrections. C) List of most significantly changed isoforms in bladder cancer. The expression level shown is relative to the normal bladder epithelium cells. Expression trend displays the pattern of expression among different cell lines starting from immortalized to most metastatic one. D) Isoforms that are most significantly changed in breast cancer. For bladder and breast cancers, one-way ANOVA was applied. E) A spectra showing a change in H2B isoforms in breast cancer. Table notations are as described in the legend to Figure 2.2

A) 15295

15536 15438 15494 15452 15509 15457 15474 15522

B)15250 15300 15350 15400 15450 15500 15550 Molecular) Expression) F) G) Raw)P) Adjusted)P) weight) Histone)idenHty) Trend)(relaHve) to)normal)) 15494& H3-&P&MeAc2& !! 0.0002& 0.0089& 15510& H3/3-&P"Me& !! 0.0003& 0.0165& M1) 15522& H3-&P&Me3Ac2/P"Ac3& !! 0.0004& 0.0169& ) H3-&P&MeAc&&& )))))hTERT) ) 15452& ! 0.0006& 0.0271& H3&2A-&Me2Ac3& ) ) 15438& H3&2A-&MeAc3& !! 0.0007& 0.0322& ) ) M2) ) C) ) ) ) ))))))) ) Normal) ) ) ) M3) )))))))))RT4) ) ) ) ) ) &&&&&&&&CLL) ) ) ) ) M4) )))))) ) D) EXPRESSION) MOLECULAR) EXPRESSION) ) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) NORMAL)) )))))))))T24) 15297& H3&2A& 0.031& ) ) E) MOLECULAR) EXPRESSION) ) GENE)IDENTITY) P8VALUE) WEIGHT) TREND) ) 15457& H3-&MeAc3& 0.039& ) 15476& H3&2A-&P&MeAc2&&Hist3H3-&MeAc& 0.034& ) 15510& Hist3H3-&P&Me& 0.008& ) ) )UM)UC3)

Figure 2.4- The complement of histone H3 isoforms derived from peak 1. A) A representative spectra of histone H3 from CD19+ B cells isolated from a healthy volunteer. The peaks that are significantly changes in CLL, bladder and breast cancers are highlighted with colored arrows. B) List of most significantly changed isoforms in CLL. The adjusted P-values were obtained after Bonferroni corrections. C) A spectra showing a change in H3 isoform abundance in a CLL patient and healthy volunteer. D) List of most significantly changed isoforms in bladder cancer. The expression level shown is relative to the normal bladder epithelium cells. Expression trend displays the pattern of expression among different cell lines starting from immortalized to most metastatic one. E) Examples of isoforms that are most significantly changed in breast cancer. Based on the molecular weight of each peak, the potential identities are listed. For bladder and breast cancers, one-way ANOVA was applied. F) and G) Representative spectra showing a change in H3-peak 1 isoforms in bladder and breast cancers, respectively. Table notations are as described in the legend to Figure 2.2

15344 A) 15358 15371

15300 15526

B) 15250 15280 15300 15320 15340 15360 15380 15400 15420 15440 15460 15480 15500 15520 15550 Molecular) Expression)Trend) Histone)idenHty) Raw)P) Adjusted)P) weight) (relaHve)to)normal)) 15528& H3/3-!P!Me2& "" 0.0003& 0.0168& C) EXPRESSION) MOLECULAR) EXPRESSION) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) NORMAL)) 15298& H3&2A& ! 0.040& D) MOLECULAR) EXPRESSION) GENE)IDENTITY) P8VALUE) WEIGHT) TREND) 15344& H3-Me2& 0.029& 15358& H3-&Me3/Ac& 0.038& 15371& H3-&MeAc& 0.014&

Figure 2.5 - The complement of histone H3 isoforms derived from peak 2. A) A representative spectra of histone H3 from CD19+ B cells isolated from a healthy volunteer. The peaks that are significantly changes in CLL, bladder and breast cancers are highlighted with colored arrows. B) List of most significantly changed isoforms in CLL. The adjusted P-values were obtained after Bonferroni corrections. C) List of most significantly changed isoforms in Bladder cancer. The expression level shown is relative to the normal bladder epithelium cells. Expression trend displays the pattern of expression among different cell lines starting from immortalized to most metastatic one. D) Examples of isoforms that are most significantly changed in breast cancer. Based on the molecular weight of each peak, the potential identities are listed. For bladder and breast cancers, one way ANOVA was applied. Table notations are as described in the legend to Figure 2.2

A) 11307 11348

11334 11322 11380 11391 11420 11446 11463 11056 11068 11126 11280 11408

10900 10950 11000 11050 11100 11150 11200 11250 11300 11350 11400 11450 11500 B) Molecular) Expression)Trend) Histone)idenRty) Raw)P) Adjusted)P) weight) (relaRve)to)normal)) E) F) 11056# H4#1G&#P"MeAc# "" 0.0007# 0.0350# H4&#MeAc3### 11419# "" 0.0008# 0.0380# H4#1I&#Me3Ac3# ))))) )))))hTERT) C) EXPRESSION) MOLECULAR) EXPRESSION) P8 GENE)IDENTITY) (RELATIVE)TO) ) M1) WEIGHT) TREND) VALUE) NORMAL)) ) ) 11068) H4#1G&#P#Me2Ac# 0.022# ) ) H4#1G&#P#Me3Ac2/P# ABSENT#IN# ) 11124) 0.034# ) Ac3# NORMAL# )))))) ) ! 11280) H4##H4#1I&#Me2# 0.040# ) ) H4&#MeAc2##H4#1I&# 11378) " 0.020# )))))))))RT4) ) Me3Ac2/Ac3# ) H4&#Me2Ac2##H4#1I&# M2) 11389) " 0.032# ) MeAc3## ) H4&#Me3Ac/#Ac3##H4# ) ) 11405) ! 0.041# 1I&#Me2Ac3# ) ) )))) ) D) MOLECULAR) EXPRESSION) ) GENE)IDENTITY) P8VALUE) ) WEIGHT) TREND) )))))))))T24) ) 11056) H4#1G&#P#MeAc# 0.027# ) ) 11306) H4&#Me2##H4#1I&#MeAc# <0.001# ) M3) 11322) H4&#Me3/Ac##H4#1I&#Me2Ac# 0.051# ) ) 11333) H4&#MeAc##H4#1I&#Me3Ac/Ac2# <0.001# ) ) 11348) H4&#Me2Ac##Hist1H4&#MeAc2# 0.004# ) ) 11378) H4&#MeAc2##H4#1I&#Me3Ac2/Ac3# <0.001# ) ) )UM)UC3) 11389) H4&#Me2Ac2##H4#1I&#MeAc3## 0.014# ) 11405) H4&#Me3Ac/#Ac3##H4#1I&#Me2Ac3# <0.001# ) 11446) H4&#Me3Ac3# 0.006# ) H4&#P#Me3Ac2/P#Ac3##H4#1I&#P# M4) 11465) 0.029# Me2Ac3#

Figure 2.6- The complement of histone H4 isoforms is changed in different cancers. A) A representative spectra of histone H3 from CD19+ B cells isolated from a healthy volunteer. The peaks that are significantly changes in CLL, bladder and breast cancers are highlighted with colored arrows. B) List of most significantly changed isoforms in CLL. The adjusted P-values were obtained after Bonferroni corrections. C) List of most significantly changed isoforms in Bladder cancer. The expression level shown is relative to the normal bladder epithelium cells. Expression trend displays the pattern of expression among different cell lines starting from immortalized to most metastatic one. D) Examples of isoforms that are most significantly changed in breast cancer. Based on the molecular weight of each peak, the potential identities are listed. For bladder and breast cancers, one way ANOVA was applied. E) and F) Spectra showing a change in H2B isoforms in bladder and breast cancer respectively. Table notations are as described in the legend to Figure 2.2

A B

Figure 2.7- Expression levels of Histone H2A isoforms after Ras-mediated transformation of HFF cells. A) H2A 1B; B) H2A 1C and C) H2A. The value on y-axis is relative expression level after normalization to total histone H2A pool.

Expression)trend) Molecular) Histone)iden3ty) (rela3ve)to) Raw)P) Adjusted)P) weight) normal)) H2A%1A%%H2A%1B*%MeAc%%H2A%1D*%Me3Ac/Ac2%%H2A% 14103% ! 0.0020% 0.0629% 2A*%MeAc2%%H2A/3*%Me2Ac% 14147% H2A%2A*%MeAc3%%H2A%2C*%P%Me3Ac3% ! 0.0045% 0.1400% 13916% H2A%1J*%Me2Ac%%H2A%1H*%MeAc2% "" 0.0054% 0.1607% 13986% H2A%1J*%MeAc3% "" 0.0092% 0.2668% 14070% H2A%2B*%P%Me3Ac/P%Ac2% " 0.0097% 0.2709% 14018% H2A%1C% " 0.0097% 0.2709% H2A%1A*%Me%%H2A%1B*%Me2Ac%%H2A%1D*%MeAc2%%H2A/ 14116% !! 0.0138% 0.3590% 3*%Me3Ac/Ac2% H2A*%Me2Ac3%%H2A%1A*%MeAc%%H2A%1B*%Me2Ac2%%H2A% 14157% 1C*%MeAc3%%H2A%1D*%MeAc3%%H2A%2A*%P%Me2Ac%%H2A/ !! 0.0161% 0.4014% 3*%Me3Ac/Ac2% 13932% H2A%1J*%Me3Ac/Ac2% " 0.0187% 0.4479% 13943% H2A%1H*%Me3Ac2/Ac3% ! 0.0215% 0.4941% 14112% H2A%1D*%P%Me%%H2A%2B*%P%Me3Ac2/P%Ac3% " 0.0311% 0.6846% 13906% H2A%2B%%% " 0.0326% 0.6846% H2A%1D*%Me%%H2A%2A*%Me2%%H2A%2B*%Me3Ac2/Ac3%% 14033% ! 0.0464% 0.9280% H2A/3% 14142% H2A*%MeAc3%%H2A%1C*%Me3Ac2/Ac3%%H2A%2A*%P%MeAc% " 0.1027% 1.0000% 14025% H2A%1J*%P%MeAc2%%H2A%2C*%Me3Ac2/Ac3% ! 0.1051% 1.0000% 14184% H2A%1C*%Me3Ac3%%H2A%2A*%P%MeAc2% ! 0.1467% 1.0000% 13899% H2A%2C% #$ 0.1468% 1.0000% 13875% H2A%1J*%Me2% #$ 0.1737% 1.0000% 14046% H2A%1B/1E% #$ 0.1975% 1.0000% 13940% H2A%1H*%P%Me3/P%Ac%%H2A%1J*%P%Me%%H2A%2C*%Me3/Ac% #$ 0.2293% 1.0000% 14083% H2A%2B*%P%MeAc2% #$ 0.2835% 1.0000% H2A%1A*%Me2Ac%%H2A%1B*%Me3Ac2/Ac3%%H2A%1D*% 14173% #$ 0.3030% 1.0000% Me2Ac3%%H2A%2A*%Me3Ac3%%H2A/3*%MeAc3% 13847% H2A%1J% #$ 0.3085% 1.0000% 13858% H2A%1H*%Me3/Ac% #$ 0.4101% 1.0000% H2A*%Me3Ac3%%H2A%1B*%P%Me3/P%Ac%%H2A%1C*%Me2Ac3%% 14168% H2A%1D*%P%Me2Ac%%H2A%2A*%P%Me3Ac/P%Ac2%%H2A/3*%P% #$ 0.4535% 1.0000% MeAc% 13958% H2A%1H*%MeAc3%%H2A%1J*%Me2Ac2% #$ 0.4991% 1.0000% 14057% H2A*%MeAc%%H2A%1C*%Me3/Ac%%H2A%2B*%P%Me2Ac% #$ 0.5460% 1.0000% 13890% H2A%1J*%Me3/Ac% #$ 0.6328% 1.0000% 13911% H2A%1H*%P%Me% #$ 0.6469% 1.0000% 13831% H2A%1H*%Me% #$ 0.8565% 1.0000% 14096% 0.9784% 1.0000% H2A*%P%Me%%H2A%1J*%P%Me3Ac3% #$

Table 2.1- Lesser significantly changed histone H2A isoforms with their expression trend in CLL

Molecular) Expression) Raw)P) weight) Histone)iden3ty) Trend)(rela3ve) Adjusted)P) to)normal)) H2B)%Me2Ac3%%H2B%1H)%Me2Ac3%%H2B%1L)%P%Me2%%H2B%1M)%Me2Ac%%H2B% 13971% ! 0.0025% 0.0821% 2E)%MeAc3% 13873% H2B)%MeAc%%H2B%1H)%Me2Ac%%H2B%2E)%Me3/Ac% ! 0.005% 0.1616% 14002% H2B%1B)%MeAc3%%H2B%1D)%Me2Ac3%%H2B%1L)%MeAc3%%H2B%1N)%Me3Ac3% ! 0.0068% 0.2109% H2B)%MeAc3%%H2B%1H)%Me2Ac3%%H2B%1L)%P%Me%%H2B%1M)%MeAc%%H2B%2E)% 13957% !! 0.0076% 0.2268% Me3Ac/Ac2% H2B)%Me2Ac%%H2B%1B)%Me2%%H2B%1D)%Me3/Ac%%H2B%1H)%Me3Ac/Ac2%%H2B% 13888% ! 0.0107% 0.3116% 1N)%MeAc%%H2B%2E)%MeAc%%H2B%3B)%Me2Ac% H2B)%Me3Ac2/Ac3%%H2B%1B)%Me3Ac/Ac2%%H2B%1D)%MeAc2%%H2B%1H)% 13944% ! 0.0118% 0.3296% MeAc3%%H2B%2E)%Me2Ac2%%H2B%3B)%Me3Ac2/Ac3% 14079% H2B%1A%%H2B%1M)%P%MeAc2%%H2B%2E)%P%Me3Ac3% "" 0.0188% 0.5074% H2B)%Me2%%H2B%1C)%MeAc2%%H2B%1H)%Me3/Ac%%H2B%1J)%Me2%%H2B%1K)% 13844% "" 0.0216% 0.5604% Me3/Ac%%H2B%2E)%Me% H2B%1B)%P%Me3Ac2/P%Ac3%%H2A%1D)%P%MeAc3%%H2B%1M)%Me3Ac3%%H2B%1N)% 14067% "" 0.022% 0.5604% P%Me2Ac3%%H2B%3B)%P%Me3Ac3% 13759% H2B%1C)%Me% ! 0.0237% 0.5685% H2B%1C)%P%Me2Ac3%%H2B%1J)%P%Me3Ac/P%Ac2%%H2B%1K)%P%MeAc2%%H2B%1L)% 13977% " 0.0259% 0.5964% Me2Ac2% H2B%1C)%P%MeAc3%%H2B%1J)%P%Me2Ac%%H2B%1K)%P%Me3Ac/P%Ac2%%H2B%1L)% 13963% "" 0.0272% 0.5993% MeAc2% 13819% H2B%%H2B%3B% " 0.0357% 0.7505% 13898% H2B%1C)%Me2Ac3%%H2B%1H)%P%Me%%H2B%1J)%Me3Ac/Ac2%%H2B%1K)%MeAc2% "" 0.0448% 0.8957% 13934% H2B%1L)%Me2Ac% ! 0.048% 0.9124% 13839% H2B%1C)%P%Me% " 0.0686% 1.0000% 13881% H2B%1C)%P%MeAc% " 0.0744% 1.0000% 13855% H2B%1C)%P%Me2%%H2B%1J)%Me3/Ac%%H2B%1K)%MeAc% " 0.0758% 1.0000% H2B)%Me2Ac2%%H2B%1H)%Me3Ac2/Ac3%%H2B%1J)%Me2Ac2%%H2B%1K)%Me3Ac2/ 13928% " 0.0991% 1.0000% Ac3%%H2B%1M)%Me2%%H2B%2E)%MeAc2%%H2B%3B)%P%Me2% 13831% H2B)%Me%%H2B%1H)%Me2%%H2B%2E% " 0.1067% 1.0000% 13911% H2B)%P%Me%%H2B%1H)%P%Me2% " 0.1674% 1.0000% H2B)%P%MeAc%%H2B%1C)%P%Me3Ac2/P%Ac3%%H2B%1H)%P%Me2Ac%%H2B%1J)%P% 13952% " 0.2034% 1.0000% MeAc%%H2B%1K)%P%Me2Ac%%H2B%2E)%P%Me3/P%Ac% 13790% H2B%1C)%Me3/Ac% #$ 0.2599% 1.0000% 13805% H2B%1H% #$ 0.3085% 1.0000% 13747% H2B%1C% #$ 0.3942% 1.0000% H2B)%P%Me3/P%Ac%%H2B%1D)%P%Me%%H2B%1H)%P%MeAc%%H2B%1J)%Me3Ac2/Ac%% 13940% #$ 0.4271% 1.0000% H2B%1K)%MeAc3%%H2B%1N)%P%Me2%%H2B%2E)%P%Me2%%H2B%3B)%P%Me3/P%Ac% 13921% H2B%1C)%P%MeAc2%%H2B%1J)%P%Me2%%H2B%1K)%P%Me3/P%Ac%%H2B%1L)%MeAc% #$ 0.4542% 1.0000% 14095% H2B%1B)%P%Me2Ac3%%H2B%1D)%P%Me3Ac3% #$ 0.6328% 1.0000% H2B)%P%Me2Ac2%%H2B%1H)%P%Me3Ac2/P%Ac3%%H2B%1J)%P%Me2Ac2%%H2B%1K)%P% 14008% #$ 0.684% 1.0000% Me3Ac2/P%Ac3%%H2B%1M)%P%Me2%%H2B%2E)%P%MeAc2% 14083% H2B%1L)%P%MeAc3% #$ 0.7489% 1.0000% 14092% H2B%1A)%Me%%H2B%1M)%P%Me2Ac2% #$ 0.9196% 1.0000% 13775% H2B%1C)%Me2% #$ 0.9784% 1.0000% 13877% 0.9926% 1.0000% H2B%1L)%Me% #$

Table 2.2- Less significantly changed histone H2B isoforms with their expression trend in CLL 44

Molecular) Expression) Raw)P) weight) Histone)iden3ty) Trend)(rela3ve) Adjusted)P) to)normal)) 15424% H3.1)%P%Me2%%H3.2)%Me3Ac2/Ac3% !! 0.0206% 0.8855% 15410% H3.1)%P%Me%%H3.2)%Me2Ac2% !! 0.0218% 0.9145% 15489% H3/3)%Me2Ac% " 0.0224% 0.9168% 15537% H3.1)%P%MeAc3% !! 0.0227% 0.9168% 15396% H3.2)%MeAc2% ! 0.0230% 0.9168% 15367% H3.2)%Me2Ac% ! 0.0301% 1.0000% 15466% H3.1)%P%Me2Ac%%H3.2)%Me3Ac3% ! 0.0316% 1.0000% 15448% H3.2)%P%Me2Ac%%H3/3)%Me2% "" 0.0459% 1.0000% 15527% H3/3)%P%Me2% "" 0.0591% 1.0000% 15461% H3.2)%P%Me3Ac/P%Ac2%%H3/3)%Me3/Ac% " 0.0753% 1.0000% 15353% H3.2)%MeAc% ! 0.0958% 1.0000% 15382% H3.2)%Me3Ac/Ac2% #$ 0.1396% 1.0000% 15390% H3.2)%P%Me% #$ 0.1628% 1.0000% 15533% H3.1)%P%MeAc3%%H3.2)%P%Me2Ac3%%H3/3)%Me2Ac2% #$ 0.1913% 1.0000% 15543% H3/3)%P%Me3/P%Ac%%H3/3)%Me3Ac2/Ac3% #$ 0.2232% 1.0000% 15296% H3.2% #$ 0.2365% 1.0000% 15431% H3/3)%Me% #$ 0.2416% 1.0000% 15404% H3.2)%P%Me2% #$ 0.2449% 1.0000% 15434% H3.2)%P%MeAc%%H3/3)%Me% #$ 0.2582% 1.0000% 15325% H3.2)%Me2% #$ 0.3299% 1.0000% 15547% H3.1)%P%Me2Ac3%%H3.2)%P%Me3Ac3%%H3/3)%Me3Ac2/ 0.3853% 1.0000% Ac3% #$ 15515% H3/3)%P%Me%%H3/3)%MeAc2% #$ 0.5108% 1.0000% 15485% H3.1)%Me3Ac3% #$ 0.5296% 1.0000% 15317% H3.1% #$ 0.5328% 1.0000% 15505% H3.1)%P%Me2Ac2%%H3.2)%P%Me3Ac2/P%Ac3%%H3/3)% 0.6012% 1.0000% Me3Ac/Ac2% #$ 15501% H3/3)%Me3Ac/Ac2% #$ 0.7419% 1.0000% 15310% H3.2)%Me% #$ 0.7454% 1.0000% 15457% H3.1)%MeAc3% #$ 0.8348% 1.0000% 15476% H3.2)%P%MeAc2%%H3/3)%MeAc% #$ 0.8354% 1.0000% 15339% 0.8422% 1.0000% H3.2)%Me3/Ac% #$

Table 2.3- Less significantly changed histone H3-1 isoforms with their expression trend in CLL

Expression) Molecular) Histone)iden3ty) Trend)(rela3ve) Raw)P) Adjusted)P) weight) to)normal)) 15329& H3.1)Me& !! 0.0010& 0.0542& 15344& H3.1)Me2& !! 0.0011& 0.0572& 15514& H3/3)&P&Me&& "" 0.0014& 0.0684& 15543& H3/3)&P&Me3/P&Ac&&H3/3)&Me3Ac2/Ac3& " 0.0014& 0.0684& 15457& H3.1)&MeAc3& " 0.0017& 0.0811& 15472& H3.1)&Me2Ac3& " 0.0038& 0.1807& 15358& H3.1)&Me3/Ac& ! 0.0048& 0.2222& 15468& H3.1)&Me2Ac3&&H3.2)&Me3Ac3& !! 0.0049& 0.2222& 15314& H3.1&&H3.2)&Me& !! 0.0063& 0.2751& 15442& H3.1)&Me2Ac3/Ac3& " 0.0063& 0.2751& 15386& H3.1)&Me2Ac& ! 0.0071& 0.2909& 15538& H3.1)&P&MeAc3& !! 0.0079& 0.3148& 15298& H3.2& !! 0.0138& 0.5390& 15400& H3.1)&Me3Ac/Ac2& ! 0.0155& 0.5898& 15485& H3.1)&Me3Ac3& "" 0.0243& 0.8796& 15371& H3.1)&MeAc& ! 0.0526& 1.0000& 15392& H3.2)&P&Me& ! 0.0681& 1.0000& 15534& H3.1)&P&MeAc3&&H3.2)&P&Me2Ac3&&& ! 0.1019& 1.0000& 15453& H3.1)&MeAc3&&H3.1)&P&MeAc&&H3.2)&Me2Ac3&& ! 0.1072& 1.0000& 15503& Hist2H3)&Me3Ac/Ac2&&Hist2H3)&P&Me3Ac2/P&Ac3& ! 0.1073& 1.0000& 15493& H3.1)&P&MeAc2& ! 0.1206& 1.0000& 15548& H3.1)&P&MeAc3&&H3.2)&P&Me3Ac3& ! 0.1455& 1.0000& 15489& H3.2)&P&MeAc2&&HIst3H3)&Me2Ac& ! 0.1502& 1.0000& 15519& H3.1)&P&Me3Ac2/P&Ac3&&H3.2)&P&MeAc3&&H3/3)&MeAc2& ! 0.1842& 1.0000& 15421& H3.1)&P&Me2&&H3.2)&P&Me3/P&Ac&&H3/3& ! 0.2059& 1.0000& 15414& H3.1)&MeAc2& ! 0.2125& 1.0000& 15461& H3.2)&P&Me3Ac/P&Ac2&&H3/3)&Me3/Ac& ! 0.2582& 1.0000& 15436& H3.2)&MeAc3&&H3.1)&P&Me3/P&Ac& #$ 0.3320& 1.0000& 15407& H3.1)&P&Me&&H3.2)&P&Me2& #$ 0.3350& 1.0000& 15428& H3.1)&Me2Ac2& #$ 0.4079& 1.0000& 15478& H3.1)&P&Me3Ac/P&Ac2& #$ 0.6698& 1.0000& 15449& H3.2)&P&Me2Ac&&H3/3)&Me2& #$ 0.7669& 1.0000& 15506& H3.1)&P&Me2Ac2&&H3.2)&P&Me3Ac2/P&Ac3& #$ 0.8519& 1.0000& 15464& 0.9461& 1.0000& H3.1)&P&Me2Ac&&H3.2)&P&Me3Ac/P&Ac2&& #$

Table 2.4- Less significantly changed histone H3-2 isoforms with their expression trend in CLL

Expression) Molecular) Histone)iden3ty) Trend)(rela3ve) Raw)P) Adjusted)P) weight) to)normal)) 11005$ H4$1G($Me3Ac/Ac2$ ! 0.0015$ 0.0721$ 11124$ H4$1G($P$Me3Ac2/P$Ac3$ ! 0.0022$ 0.1015$ 11405$ H4($Me3Ac/$Ac3$$H4$1I($Me2Ac3$ ! 0.0074$ 0.3256$ 11497$ H4($P$MeAc3$$H4$1I($P$Me3Ac3$ " 0.0081$ 0.3487$ 10946$ H4$1G($Me2$ ! 0.0082$ 0.3487$ 11250$ H4$1I$ ! 0.0088$ 0.3594$ 10990$ H4$1G($Me2Ac$ ! 0.0137$ 0.5362$ 11002$ H4$1G($Me3Ac/Ac2$ " 0.0149$ 0.5668$ 11068$ H4$1G($P$Me2Ac$ ! 0.0170$ 0.6274$ 11019$ H4$1G($MeAc2$ ! 0.0175$ 0.6301$ 11034$ H4$1G($Me2Ac2$ ! 0.0222$ 0.7775$ 10974$ H4$1G(MeAc$ ! 0.0294$ 0.9701$ 11378$ H4($MeAc2$$H4$1I($Me3Ac2/Ac3$ " 0.0295$ 0.9701$ 11526$ H4($P$Me3Ac3$ " 0.0358$ 1.0000$ 11433$ H4($Me2Ac3$ ! 0.0471$ 1.0000$ 11306$ H4($Me2$$H4$1I($MeAc$ " 0.0491$ 1.0000$ 11389$ H4($Me2Ac2$$H4$1I($MeAc3$$ " 0.0757$ 1.0000$ 11511$ H4($P$Me2Ac3$ " 0.0919$ 1.0000$ 11348$ H4($Me2Ac$$H4$1I($MeAc2$ " 0.1016$ 1.0000$ 11048$ H4$1G($Me3Ac2/Ac3$ ! 0.1388$ 1.0000$ 11446$ H4($Me3Ac3$ ! 0.1483$ 1.0000$ 11484$ H4($P$Me3Ac2/P$Ac3$$H4$1I($P$Me2Ac3$ ! 0.1507$ 1.0000$ 11322$ H4($Me3/Ac$$H4$1I($Me2Ac$ " 0.1635$ 1.0000$ 11333$ H4($MeAc$$H4$1I($Me3Ac/Ac2$ #$ 0.1915$ 1.0000$ 11043$ H4$1G($P$Me3/P$Ac$ #$ 0.2674$ 1.0000$ 11028$ H4$1G($P$Me2$ #$ 0.2947$ 1.0000$ 11076$ H4$1G($Me2Ac3$ #$ 0.3041$ 1.0000$ 11265$ H4$1I($Me$ #$ 0.3381$ 1.0000$ 11292$ H4($Me$$H4$1I($Me3/Ac$ #$ 0.3381$ 1.0000$ 10921$ H4$1G$ #$ 0.3463$ 1.0000$ 11363$ H4($Me3Ac/Ac2$$H4$1I($Me2Ac2$ #$ 0.4032$ 1.0000$ 11014$ H4$1G($P$Me$ #$ 0.5174$ 1.0000$ 11373$ H4($P$Me$$H4$1I($P$Me3/P$Ac$ #$ 0.6505$ 1.0000$ 11280$ H4$$H4$1I($Me2$ #$ 0.6759$ 1.0000$ 11039$ H4$1G($P$Me3/P$Ac$ #$ 0.7055$ 1.0000$ 10965$ H4$1G($Me3/Ac$ #$ 0.7801$ 1.0000$ 11384$ 0.8352$ 1.0000$ H4($P$Me2$$H4$1I($P$MeAc$ #$

Table 2.5- Lesser significantly changed histone H4 isoforms with their expression trend in CLL

EXPRESSION) MOLECULAR) EXPRESSION) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) NORMAL)) 13817% H2A%1H% ! "# 0.962% 13832% H2A%1H-%Me% ! 0.477% 13847% H2A%1J% $ 0.001% 13858% H2A%1H-%Me3/Ac% ! 0.554% 13876% H2A%1J-%Me2% ! /\/\% 0.872% ABSENT IN /\/\% 13891% H2A%1J-%Me3/Ac% NORMAL 0.438% 13899% H2A%2C% ! 0.001% 13906% H2A%2B%%% ! /\/\% 0.784% 13915% H2A%1J-%Me2Ac%%H2A%1H-%MeAc2% ! /\/\% 0.523% ABSENT IN 13932% H2A%1J-%Me3Ac/Ac2% NORMAL 0.015% ABSENT IN 13936% H2A%2B-%Me2% NORMAL 0.549% 13941% H2A%1H-%P%Me3/P%Ac%%%%%H2A%1J-%P%Me%%%%%H2A%2C-%Me3/Ac% ! 0.310% 13958% H2A%1H-%MeAc3%%H2A%1J-%Me2Ac2% ! 0.303% 13987% H2A%1J-%MeAc3% ! 0.730% ABSENT IN /\/\% 13996% H2A%1H-%P%MeAc2%%%%%H2A%1J-%P%Me2Ac%%%%%H2A%2C-%MeAc2% NORMAL 0.805% 14004% H2A% $ "#% 0.302% 14011% H2A%1J-%P%Me3Ac/P%Ac2%%H2A%2C-%Me2Ac2% ! /\/\% 0.829% 14018% H2A%1C% ! 0.037% 14025% H2A%1J-%P%MeAc2%%H2A%2C-%Me3Ac2/Ac3% ! 0.605% 14032% H2A%1D-%Me%%H2A%2A-%Me2%%H2A%2B-%Me3Ac2/Ac3%%%H2A/3% $ "# 0.977% 14045% H2A%1B% ! 0.267% 14063% H2A%2C-%P%Me3Ac/P%Ac2%%H2A%2A-%MeAc%% /\/\% /\/\% 0.589% 14070% H2A%2B-%P%Me3Ac/P%Ac2% ! /\/\% 0.890% 14083% H2A%2B-%P%MeAc2% ! /\/\% 0.240% 14097% H2A-%P%Me%%H2A%1J-%P%Me3Ac3% /\/\% 0.439% H2A%1A%%H2A%1B-%MeAc%%H2A%1D-%Me3Ac/Ac2%%H2A%2A-% 14103% /\/\% 0.893% MeAc2%%H2A-%Me2Ac% 14112% H2A%1D-%P%Me%%H2A%2B-%P%Me3Ac2/P%Ac3% /\/\% /\/\% 0.420% H2A%1A-%Me%%%%H2A%1B-%Me2Ac%%%%H2A%1D-%MeAc2%%%%%H2A/3-% 14117% $ "# 0.998% Me3Ac/Ac2% 14131% H2A%2A-%Me3Ac2/Ac3%%H2A%2C-%P%Me2Ac3% /\/\% /\/\% 0.70% H2A-%Me2Ac3%%H2A%1A-%MeAc%%H2A%1B-%Me2Ac2%%H2A%1C-% 14157% MeAc3%%H2A%1D-%MeAc3%%H2A%2A-%P%Me2Ac%%H2A/3-% $ 0.874% Me3Ac/Ac2% H2A%1A-%Me2Ac%%H2A%1B-%Me3Ac2/Ac3%%H2A%1D-%Me2Ac3%% 14171% /\/\% 0.868% H2A%2A-%Me3Ac3%%H2A/3-%MeAc3% ABSENT IN /\/\% 14184% H2A%1C-%Me3Ac3%%H2A%2A-%P%MeAc2% NORMAL 0.942% H2A%1A-%P%Me%%H2A%1B-%P%Me2Ac%%H2A%1D-%P%MeAc2%%H2A% 14198% /\/\% /\/\% 0.564% 2A-%P%Me2Ac2%%H2A/3-%P%Me3Ac/P%Ac2%

Table 2.6- Histone H2A isoforms with their expression trend in bladder cancer

EXPRESSION)LEVEL) MOLECULAR) EXPRESSION) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) NORMAL)) 13759& H2B&1C+&Me& 0.149& 13775& H2B&1C+&Me2& ! 0.031& 13790& H2B&1C+&Me3/Ac& " 0.440& 13819& H2B&&H2B&3B& ! /\/\& 0.016& 13832& H2B+&Me&&H2B&1H+&Me2&&H2B&2E& ! 0.050& H2B+&Me2&&&&&&&H2B&1C+&MeAc2&&&&&&&H2B&1H+&Me3/Ac&&H2B&1J+& 13845& ! #$ 0.975& Me2&&&&&&&H2B&1K+&Me3/Ac&&&&&&&H2B&2E+&Me& & H2B+&Me3/Ac&&&&&H2B&1C+&Me2Ac2&&&&&H2B&1K+&MeAc&&&&H2B&1J+& 13858& ! /\/\& 0.224& Me3/Ac&&&&&H2B&1H+&MeAc&&&&&H2B&2E+&Me2&& 13873& H2B+&MeAc&&&&&&H2B&1H+&Me2Ac&&&&&&&H2B&2E+&Me3/Ac& " /\/\& 0.468& H2B+&Me2Ac&&H2B&1B+&Me2&&H2B&1D+&Me3/Ac&&H2B&1H+&Me3Ac/ 13889& /\/\& /\/\& 0.162& Ac2&&H2B&1N+&MeAc&&H2B&2E+&MeAc&&H2B&3B+&Me2Ac& H2B+&Me3Ac/Ac2&&&&H2B&1B+&Me3/Ac&&&&H2B&1D+&MeAc&&&&H2B& 13903& 1H+&MeAc2&&&&H2B&1N+&Me2Ac&&&&H2B&2E+&Me2Ac&&H2B&3B+& /\/\& /\/\& 0.872& Me3Ac/Ac2& ABSENT IN 13933& H2B&1L+&Me2Ac& NORMAL 0.410& H2B+&P&Me3/P&Ac&&&H2B&1D+&P&Me&&&H2B&1H+&P&MeAc&&&H2B&1J+& 13942& Me3Ac2/Ac&&&H2B&1K+&MeAc3&&H2B&1N+&P&Me2&&&H2B&2E+&P&Me2&&& " 0.523& H2B&3B+&P&Me3/P&Ac& H2B+&MeAc3&&H2B&1H+&Me2Ac3&&H2B&1L+&P&Me&&H2B&1M+&MeAc&& 13957& " /\/\& 0.492& H2B&2E+&Me3Ac/Ac2& H2B+&Me2Ac3&&H2B&1H+&Me2Ac3&&H2B&1L+&P&Me2&&H2B&1M+& 13973& " /\/\& 0.665& Me2Ac&&H2B&2E+&MeAc3& H2B&1B+&Me3Ac2/Ac3&&H2B&1D+&MeAc3&&H2B&1N+&Me2Ac3&&H2B& 13988& #$ #$ 0.994& 1L+&Me3Ac3&&&&& H2B&1B+&MeAc3&&H2B&1D+&Me2Ac3&&H2B&1L+&MeAc3&&H2B&1N+& 14002& ! 0.875& Me3Ac3& 14033& H2B&1L+&Me3Ac3& ! /\/\& 0.402& ABSENT IN /\/\& 14047& H2B&1J+&P&Me2Ac3&&H2B&1K+&P&Me3Ac3&&& NORMAL 0.735& 14054& H2B&1L+&P&Me2Ac2&&H2B&1M+&Me2Ac3& " 0.953& H2B&1B+&P&Me3Ac2/P&Ac3&&H2B&1D+&P&MeAc3&&H2B&1M+&Me3Ac3&& 14069& " /\/\& 0.909& H2B&1N+&P&Me2Ac3&&H2B&3B+&P&Me3Ac3& 14083& H2B&1L+&P&MeAc3& " 0.678& 14092& H2B&1A+&Me&&H2B&1M+&P&Me2Ac2& /\/\& /\/\& 0.804& 14096& H2B&1B+&P&Me2Ac3&&H2B&1D+&P&Me3Ac3& /\/\& /\/\& 0.626&

Table 2.7- Histone H2B isoforms with their expression trend in bladder cancer

EXPRESSION)LEVEL) MOLECULAR) GENE)IDENTITY) (RELATIVE)TO) TREND) P8VALUE) WEIGHT) NORMAL)) 15297) H3.2) 0.031) 15310) H3.28Me) /\/\) 0.040) 15317) H3.1) /\/\) /\/\) 0.901) 15325) H3.28)Me2) ! /\/\) 0.534) 15339) H3.28)Me3/Ac) ! "# 0.939) 15356) H3.18)Me3/Ac) ! /\/\) 0.873) 15367) H3.28)Me2Ac) ! 0.902) 15382) H3.28)Me3Ac/Ac2) ! /\/\) 0.486) 15396) H3.28)MeAc2) ! /\/\) 0.339) 15404) H3.28)P)Me2) $) /\/\) 0.850) 15410) H3.18)P)Me))H3.28)Me2Ac2) ! /\/\) 0.326) 15424) H3.18)P)Me2))H3.28)Me3Ac2/Ac3) ! /\/\) 0.600) 15431) H3/38)Me) /\/\) /\/\) 0.905) 15438) H3.28)MeAc3) $) /\/\) 0.957) 15448) H3.28)P)Me2Ac))H3/38)Me2) $) /\/\) 0.769) 15452) H3.18)P)MeAc))H3.28)Me2Ac3) /\/\) /\/\) 0.335) 15457) H3.18)MeAc3) $ /\/\) 0.922) 15461) H3.28)P)Me3Ac/P)Ac2))H3/38)Me3/Ac) $ /\/\) 0.911) 15466) H3.18)P)Me2Ac))H3.28)Me3Ac3) $ /\/\) 0.636) 15480) H3.18)P)Me3Ac/P)Ac2) $ /\/\) 0.775) 15485) H3.18)Me3Ac3) Absent)in)normal) /\/\) 0.539) 15501) H3/38)Me3Ac/Ac2) $ /\/\) 0.913) H3.18)P)Me2Ac2))H3.28)P)Me3Ac2/P)Ac3)) 15505) $ /\/\) 0.942) H3/38)Me3Ac/Ac2) 15510) H3/38)P)Me) $ /\/\) 0.997) 15515) H3/38)P)Me))H3/38)MeAc2) $ /\/\) 0.950) 15522) H3.18)P)Me3Ac2/P)Ac3) $ /\/\) 0.892) 15527) H3/38)P)Me2) $ /\/\) 0.266) H3.18)P)MeAc3))H3.28)P)Me2Ac3))H3/38) 15533) $ /\/\) 0.312) Me2Ac2) 15537) H3.18)P)MeAc3) $ /\/\) 0.888) 15543) H3/38)P)Me3/P)Ac))H3/38)Me3Ac2/Ac3) ! 0.974) H3.18)P)Me2Ac3))H3.28)P)Me3Ac3))H3/38) 15547) $ /\/\) 0.931) Me3Ac2/Ac3)

Table 2.8- Histone H3-1 isoforms with their expression trend in bladder cancer

EXPRESSION)LEVEL) MOLECULAR) EXPRESSION) GENE)IDENTITY) (RELATIVE)TO) P8VALUE) WEIGHT) TREND) NORMAL)) 15298) H3.2% ! 0.040% 15314) H3.1%%H3.2)%Me% ! /\/\) 0.522% 15329) H3.1)%Me% ! "# 0.591% 15344) H3.1)Me2% ! /\/\) 0.506% 15358) H3.1)%Me3/Ac% ! 0.419% 15371) H3.1)%MeAc% ! 0.563% 15386) H3.1)%Me2Ac% ! /\/\% 0.508% ABSENT%IN%CANCER% 15392) H3.2)%P%Me% N.A.% N.A.% CELL%LINES% 15400) H3.1)%Me3Ac/Ac2% ! /\/\ 0.239% ABSENT%IN%CANCER% 15407) H3.1)%P%Me%%H3.2)%P%Me2% N.A.% N.A.% CELL%LINES% 15414) H3.1)%MeAc2% ! /\/\) 0.568% 15421) H3.1)%P%Me2%%H3.2)%P%Me3/P%Ac%%H3/3% /\/\) /\/\) 0.965% 15428) H3.1)%Me2Ac2% ! /\/\) 0.263% 15436) H3.2)%MeAc3%%H3.1)%P%Me3/P%Ac% /\/\) /\/\) 0.983% 15442) H3.1)%Me2Ac3/Ac3% ! /\/\) 0.751% ABSENT%IN%CANCER% 15449) H3.2)%P%Me2Ac%%H3/3)%Me2% N.A.% N.A.% CELL%LINES% ABSENT%IN%CANCER% 15453) H3.1)%MeAc3%%H3)%P%MeAc%%H3.2)%Me2Ac3%% N.A.% N.A.% CELL%LINES% 15457) H3.1)%MeAc3% ! /\/\) 0.971% ABSENT%IN%CANCER% 15464) H3.1)%P%Me2Ac%%H3.2)%P%Me3Ac/P%Ac2%% N.A.% N.A.% CELL%LINES% 15468) H3.1)%Me2Ac3%%H3.2)%Me3Ac3% $ /\/\) 0.977% 15472) H3.1)%Me2Ac3% ABSENT%IN%NORMAL% /\/\) 0.986% 15478) H3.1)%P%Me3Ac/P%Ac2% /\/\) /\/\) 0.948% 15485) H3.1)%Me3Ac3% !) /\/\) 0.762% 15493) H3.1)%P%MeAc2% /\/\) /\/\) 0.955% 15503) Hist2H3)%Me3Ac/Ac2%%Hist2H3)%P%Me3Ac2/P%Ac3% $) /\/\) 0.895% ABSENT%IN%CANCER% 15510) H3/3)%P%Me% N.A.% N.A.% CELL%LINES% 15514) H3/3)%P%Me%% /\/\) /\/\) 0.877% 15524) H3.1)%P%Me3Ac2/P%Ac3% $ 0.951% 15528) H3/3)%P%Me2% ! /\/\) 0.784% 15534) H3.1)%P%MeAc3%%H3.2)%P%Me2Ac3%%% /\/\) /\/\) 0.986% 15538) H3.1)%P%MeAc3% $ /\/\) 0.928% /\/\) 15543) H3/3)%P%Me3/P%Ac%%H3/3)%Me3Ac2/Ac3% ABSENT%IN%NORMAL% 0.852%

Table 2.9- Histone H3-2 isoforms with their expression trend in bladder cancer

EXPRESSION) MOLECULAR) EXPRESSION) GENE)IDENTITY) LEVEL)(RELATIVE) P8VALUE) WEIGHT) TREND) TO)NORMAL)) 10921) H4#1G# ! /\/\) 0.203# 10946) H4#1G*#Me2# /\/\) /\/\) 0.121# 10965) H4#1G*#Me3/Ac# "# "# 0.964# 10974) H4#1G*MeAc# $ 0.539# 10990) H4#1G*#Me2Ac# $ /\/\ 0.167# ABSENT#IN# 11002) H4#1G*#Me3Ac/Ac2# 0.582# NORMAL# 11019) H4#1G*#MeAc2# $ /\/\) 0.043# 11028) H4#1G*#P#Me2# $ 0.757# 11034) H4#1G*#Me2Ac2# $ 0.183# ABSENT#IN# 11039) H4#1G*#P#Me3/P#Ac# /\/\) 0.435# NORMAL# 11043) H4#1G*#P#Me3/P#Ac# !) /\/\) 0.878# 11048) H4#1G*#Me3Ac2/Ac3# ! /\/\) 0.135# 11056) H4#1G*#P#MeAc# /\/\) /\/\) 0.484# 11068) H4#1G*#P#Me2Ac# 0.022# 11076) H4#1G*#Me2Ac3# /\/\) /\/\) 0.389# ABSENT#IN# 11124) H4#1G*#P#Me3Ac2/P#Ac3# 0.034# NORMAL# 11250) H4#1I# /\/\) /\/\) 0.678# 11265) H4#1I*#Me# $ /\/\) 0.847# 11280) H4##H4#1I*#Me2# $ 0.040# 11292) H4*#Me##H4#1I*#Me3/Ac# $ /\/\) 0.123# 11306) H4*#Me2##H4#1I*#MeAc# ! 0.857# 11322) H4*#Me3/Ac##H4#1I*#Me2Ac# /\/\) /\/\) 0.541# 11333) H4*#MeAc##H4#1I*#Me3Ac/Ac2# "# "# 0.980# 11348) H4*#Me2Ac##H4#1I*#MeAc2# ! 0.602# 11363) H4*#Me3Ac/Ac2##H4#1I*#Me2Ac2# "# "# 0.840# 11373) H4*#P#Me##H4#1I*#P#Me3/P#Ac# $ 0.122# 11378) H4*#MeAc2##H4#1I*#Me3Ac2/Ac3# ! 0.020# 11389) H4*#Me2Ac2##H4#1I*#MeAc3## ! 0.032# 11405) H4*#Me3Ac/#Ac3##H4#1I*#Me2Ac3# $ 0.041# 11419) H4*#MeAc3##H4#1I*#Me3Ac3# $ 0.924# ABSENT#IN# 11433) H4*#Me2Ac3# 0.637# NORMAL# 11446) H4*#Me3Ac3# "# "# 0.989# 11484) H4*#P#Me3Ac2/P#Ac3##H4#1I*#P#Me2Ac3# /\/\) 0.913# 11497) H4*#P#MeAc3##H4#1I*#P#Me3Ac3# /\/\) /\/\) 0.801#

Table 2.10- Histone H4 isoforms with their expression trend in bladder cancer

MOLECULAR) EXPRESSION) GENE)IDENTITY) P5VALUE) WEIGHT) TREND) 13817% H2A%1H% 0.004% 13832% H2A%1H,%Me% 0.003% 13847% H2A%1J% 0.006% 13885% H2A%1H,%Me2Ac% /\/\% 0.133% 13899% H2A%2C% 0.276% 13915% H2A%1J,%Me2Ac%%H2A%1H,%MeAc2% /\/\% 0.005% 13932% H2A%1J,%Me3Ac/Ac2% 0.001% 13948% H2A%2B,%Me3/Ac% %/\/\ 0.076% 13969% H2A%1J,%P%Me3/P%Ac%%H2A%2C,%Me2Ac% /\/\% 0.795% 13982% H2A%1H,%P%Me3Ac/P%Ac2%%%H2A%1J,%P%MeAc%%%H2A%2C,%Me3Ac/Ac3% /\/\% 0.004% 14004% H2A% !"% 0.129% 14018% H2A%1C% 0.001% 14032% H2A%1D,%Me%%%%%%H2A%2A,%Me2%%%%%H2A%2B,%Me3Ac2/Ac3%%%%%H2A/3% <0.001% 14045% H2A%1B% <0.001% 14083% H2A%2B,%P%MeAc2% /\/\% 0.381% H2A%1A%%H2A%1B/1E,%MeAc%%H2A%1D,%Me3Ac/Ac2%%H2A%2A,%MeAc2%% % 14103% 0.123% H2A/3,%Me2Ac% 14121% H2A%2B,%P%MeAc3% !"% 0.203% 14151% H2A,%P%Me2Ac%%Hist1H2AC,%P%MeAc% 0.466% H2A%1A,%Me2Ac%%H2A%1B/1E,%Me3Ac2/Ac3%%H2A%1D,%Me2Ac3%%%H2A%2A,% /\/\% 14171% 0.610% Me3Ac3%%H2A/3,%MeAc3% /\/\% 14184% H2A%1C,%Me3Ac3%%H2A%2A,%P%MeAc2% 0.157%

Table 2.11- Histone H2A isoforms with their expression trend in breast cancer

MOLECULAR) EXPRESSION) GENE)IDENTITY) P5VALUE) WEIGHT) TREND) 13759& H2B&1C+&Me& <0.001& 13775& H2B&1C+&Me2& <0.001& 13790& H2B&1C+&Me3/Ac& /\/\& 0.002& 13819& H2B&&H2B&3B& /\/\& 0.371& H2B+&Me2&&&&&&&H2B&1C+&MeAc2&&&&&&&&H2B&1H+&Me3/Ac&& 13845& 0.271& H2B&1J+&Me2&&&&&H2B&1K+&Me3/Ac&&&&H2B&2E+&Me& H2B+&Me3/Ac&&&&H2B&1C+&Me2Ac2&&&&H2B&1K+&MeAc&&&&&&& 13858& 0.264& H2B&1J+&Me3/Ac&&&&&H2B&1H+&MeAc&&&&&&&H2B&2E+&Me2&& 13873& H2B+&MeAc&&H2B&1H+&Me2Ac&&H2B&2E+&Me3/Ac& !"& 0.663& H2B+&Me2Ac&&H2B&1B+&Me2&&H2B&1D+&Me3/Ac&&H2B&1H+& 13889& Me3Ac/Ac2&&H2B&1N+&MeAc&&H2B&2B/2E+&MeAc&&H2B& 0.123& 3B+&Me2Ac& 13908& H2B&1C+&P&Me3Ac/P&Ac2&&H2B&1K+&P&Me2&&H2B&1J+&P&Me& /\/\& 0.321& 13927& H2B&1L+&Me2Ac& /\/\& 0.006& H2B+&P&Me3/P&Ac&&&&H2B&1D+&P&Me&&&&H2B&1H+&P&MeAc&&&&&&& 13942& H2B&1J+&Me3Ac2/Ac&&&&&H2B&1K+&MeAc3&&&&H2B&1N+&P& !"& 0.469& Me2&&&&&H2B&2E+&P&Me2&&&&&&H2B&3B+&P&Me3/P&Ac& 13961& H2B&1L+&MeAc2& 0.038& H2B+&Me2Ac3&&H2B&1H+&Me2Ac3&&H2B&1L+&P&Me2&&H2B& 13973& 0.280& 1M+&Me2Ac&&H2B&2E+&MeAc3& H2B&1C+&P&Me3Ac3&&H2B&1K+&P&Me2Ac2&&H2B&1J+&P& 13992& 0.353& MeAc2& H2B&1H+&P&Me3Ac2/P&Ac3&&H2B+&P&Me2Ac2&&H2B&3B+&P& 14010& Me2Ac2&&H2B&1N+&P&MeAc2&&H2B&1D+&P&Me3Ac/P&Ac2&& 0.357& H2B&1B+&P&Me2Ac&& 14028& H2B&1B+&Me3Ac3&&H2B&1L+&P&Me3Ac/P&Ac2& 0.304& 14047& H2B&1J+&P&Me2Ac3&&H2B&1K+&P&Me3Ac3&&& /\/\& 0.392& H2B&1J+&P&Me3Ac3&&H2B+&P&Me3Ac3&&H2B&1M+&P& 14063& 0.568& Me3Ac/P&Ac2& /\/\& 14077& H2B&2E+&P&Me3Ac3&&&&&&&&H2B&1M+&P+MeAc2&&&&&&&&&H2B&1A& /\/\& 0.777& 14096& H2B&1B+&P&Me2Ac3&&H2B&1D+&P&Me3Ac3& 0.423& 14133& H2B&1M+&P&Me2Ac3&&H2B&1A+&MeAc& /\/\& 0.300& 14148& H2B&1M+&P&Me3Ac3&&H2B&1A+&Me2Ac& !"& 0.911&

Table 2.12- Histone H2B isoforms with their expression trend in breast cancer

MOLECULAR) GENE)IDENTITY) TREND) P3VALUE) WEIGHT) 15316% H3.1% 0.117% 15331% H3.1*Me% 0.179% 15347% H3.2*%Me3/Ac% 0.080% 15356% H3.1*%Me3/Ac% 0.103% 15387% H3.1*%Me2Ac% /\/\% 0.524% 15400% H3.1*%Me2Ac/Ac2% /\/\% 0.602% 15415% H3.1*%MeAc2% 0.851% 15457% H3.1*%MeAc3% 0.039% 15476% H3.2*%P%MeAc2%%H3/3*%MeAc% 0.034% 15494% H3.1*%P%MeAc2% /\/\% 0.001% 15510% H3/3*%P%Me% 0.008%

Table 2.13- Histone H3-1 isoforms with their expression trend in breast cancer

MOLECULAR) EXPRESSION) GENE)IDENTITY) P5VALUE) WEIGHT) TREND) 15298) H3.2% 0.131% 15314) H3.1%%H3.2(%Me% 0.213% 15329) H3.1(%Me% 0.174% 15344) H3.1(Me2% 0.029% 15358) H3.1(%Me3/Ac% 0.038% 15371) H3.1(%MeAc% 0.014% 15386) H3.1(%Me2Ac% /\/\% 0.031% 15400) H3.1(%Me3Ac/Ac2% /\/\ 0.086% 15414) H3.1(%MeAc2% /\/\) 0.138% 15428) H3.1(%Me2Ac2% 0.332% 15442) H3.1(%Me2Ac3/Ac3% /\/\) 0.149% 15457) H3.1(%MeAc3% 0.261% 15464) H3.1(%P%Me2Ac%%H3.2(%P%Me3Ac/P%Ac2%% !"% 0.962% 15472) H3.1(%Me2Ac3% /\/\) 0.096% 15485) H3.1(%Me3Ac3% 0.512% 15503) Hist2H3(%Me3Ac/Ac2%%Hist2H3(%P%Me3Ac2/P%Ac3% /\/\) 0.072% /\/\) 15534) H3.1(%P%MeAc3%%H3.2(%P%Me2Ac3%%% 0.297%

Table 2.14- Histone H3-2 isoforms with their expression trend in breast cancer

MOLECULAR) EXPRESSION) GENE)IDENTITY) P5VALUE) WEIGHT) TREND) 11005) H4#1G&#Me3Ac/Ac2# /\/\) 0.451# 11019) H4#1G&#MeAc2# /\/\) 0.690# 11034) H4#1G&#Me2Ac2# 0.291# 11043) H4#1G&#P#Me3/P#Ac# /\/\) 0.242# 11056) H4#1G&#P#MeAc# 0.027# 11068) H4#1G&#P#Me2Ac# 0.511# 11100) H4#1G&#P#MeAc2# 0.645# 11114) H4#1G&#P#Me2Ac2## 0.552# 11124) H4#1G&#P#Me3Ac2/P#Ac3# 0.538# 11280) H4##H4#1I&#Me2# 0.141# 11292) H4&#Me##H4#1I&#Me3/Ac# 0.113# 11306) H4&#Me2##H4#1I&#MeAc# <0.001# 11322) H4&#Me3/Ac##H4#1I&#Me2Ac# 0.051# 11333) H4&#MeAc##H4#1I&#Me3Ac/Ac2# <0.001# 11348) H4&#Me2Ac##H4#1I&#MeAc2# 0.004# 11363) H4&#Me3Ac/Ac2##H4#1I&#Me2Ac2# !" 0.985# 11378) H4&#MeAc2##H4#1I&#Me3Ac2/Ac3# <0.001# 11389) H4&#Me2Ac2##H4#1I&#MeAc3## 0.014# 11405) H4&#Me3Ac/#Ac3##H4#1I&#Me2Ac3# <0.001# 11446) H4&#Me3Ac3# 0.006# 11465) H4&#P#Me3Ac2/P#Ac3##H4#1I&#P#Me2Ac3# 0.029# 11479) H4&#P#MeAc3##H4#1I&#P#Me3Ac3# /\/\) 0.447#

Table 2.15- Histone H4 isoforms with their expression trend in breast cancer

Patient$Samples$Prior$to$ All$Patient$Samples$(n$=$83) Treament$and$with$Clinical$ Outcome$Data$(n$=$64)

H2A$ H2A$ Raw$P1 Adjusted$P FDR Raw$P Adjusted$P FDR Fragment Fragment _4103 14103 0.0105 0.40 0.34 0.0394 1.00 0.30 _3847 13847 0.0296 1.00 0.34 0.0360 1.00 0.30 _3932 13932 0.0296 1.00 0.34 0.0337 1.00 0.30 _4033 14033 0.0355 1.00 0.34 0.0149 0.57 0.30 _4096 14096 0.0881 1.00 0.63 0.1904 1.00 0.90 _4004 14004 0.1057 1.00 0.63 0.0200 0.74 0.30 _4083 14083 0.1202 1.00 0.63 0.3698 1.00 0.93 _3899 13899 0.1339 1.00 0.63 0.2897 1.00 0.93 _4025 14025 0.1498 1.00 0.63 0.7032 1.00 0.93 _4184 14184 0.1703 1.00 0.65 0.0830 1.00 0.53 _3911 13911 0.2412 1.00 0.77 0.2204 1.00 0.93 _3916 13916 0.2571 1.00 0.77 0.4381 1.00 0.93 _3865 13865 0.2811 1.00 0.77 0.8090 1.00 0.93 _4063 14063 0.3025 1.00 0.77 0.3903 1.00 0.93 _4147 14147 0.3343 1.00 0.77 0.3353 1.00 0.93 _4018 14018 0.3372 1.00 0.77 0.7615 1.00 0.93 _4057 14057 0.3465 1.00 0.77 0.6612 1.00 0.93 _3817 13817 0.4160 1.00 0.87 0.1587 1.00 0.86 _3987 13987 0.4591 1.00 0.87 0.6612 1.00 0.93 _3890 13890 0.4814 1.00 0.87 0.5392 1.00 0.93 _3940 13940 0.4973 1.00 0.87 0.7197 1.00 0.93 _4168 14168 0.5124 1.00 0.87 0.3013 1.00 0.93 _4197 14197 0.5815 1.00 0.87 0.7704 1.00 0.93 _4116 14116 0.5989 1.00 0.87 0.8132 1.00 0.93 _4173 14173 0.6074 1.00 0.87 0.7411 1.00 0.93 _3875 13875 0.6141 1.00 0.87 0.8587 1.00 0.93 _3906 13906 0.6673 1.00 0.87 0.6320 1.00 0.93 _4157 14157 0.6728 1.00 0.87 0.4527 1.00 0.93 _4132 14132 0.6742 1.00 0.87 0.5044 1.00 0.93 _3996 13996 0.7027 1.00 0.87 0.9396 1.00 0.95 _4112 14112 0.7077 1.00 0.87 0.6034 1.00 0.93 _3943 13943 0.7513 1.00 0.87 0.6613 1.00 0.93 _4070 14070 0.7629 1.00 0.87 0.9516 1.00 0.95 _4046 14046 0.7839 1.00 0.87 0.6130 1.00 0.93 _4142 14142 0.8121 1.00 0.87 0.8874 1.00 0.94 _3831 13831 0.8259 1.00 0.87 0.8445 1.00 0.93 _3858 13858 0.9013 1.00 0.93 0.6400 1.00 0.93 _3958 13958 0.9417 1.00 0.94 0.7718 1.00 0.93

Table 2.16- Correlation of H2A isoforms with Zap-70 58

Patient$Samples$Prior$to$ All$Patient$Samples$(n$=$83) Treament$and$with$Clinical$ Outcome$Data$(n$=$64)

H2B$Fragment H2B$Fragment Raw$P1 Adjusted$P FDR Raw$P Adjusted$P FDR

_3944 13944 0.0264 1.00 0.41 0.0990 1.00 0.64 _3775 13775 0.0317 1.00 0.41 0.4453 1.00 0.85 _3747 13747 0.0476 1.00 0.41 0.0758 1.00 0.64 _3987 13987 0.0478 1.00 0.41 0.0624 1.00 0.64 _3928 13928 0.0635 1.00 0.41 0.0654 1.00 0.64 _3957 13957 0.0669 1.00 0.41 0.0903 1.00 0.64 _3839 13839 0.0823 1.00 0.41 0.4739 1.00 0.85 _4095 14095 0.1021 1.00 0.41 0.2418 1.00 0.79 _3963 13963 0.1057 1.00 0.41 0.5276 1.00 0.86 _3971 13971 0.1064 1.00 0.41 0.3894 1.00 0.85 _3831 13831 0.1224 1.00 0.43 0.3709 1.00 0.85 _3819 13819 0.1675 1.00 0.48 0.1882 1.00 0.76 _3877 13877 0.1682 1.00 0.48 0.1391 1.00 0.76 _3790 13790 0.1732 1.00 0.48 0.0947 1.00 0.64 _3759 13759 0.1943 1.00 0.51 0.2161 1.00 0.77 _4092 14092 0.2643 1.00 0.63 0.6904 1.00 0.90 _3898 13898 0.2725 1.00 0.63 0.3086 1.00 0.80 _4002 14002 0.3211 1.00 0.70 0.1950 1.00 0.76 _3805 13805 0.3465 1.00 0.70 0.2806 1.00 0.80 _3977 13977 0.3604 1.00 0.70 0.8125 1.00 0.96 _4032 14032 0.3880 1.00 0.72 0.1746 1.00 0.76 _4021 14021 0.5105 1.00 0.87 0.8953 1.00 0.96 _3940 13940 0.5338 1.00 0.87 0.7668 1.00 0.93 _3921 13921 0.5341 1.00 0.87 0.2862 1.00 0.80 _4048 14048 0.5733 1.00 0.89 0.4627 1.00 0.85 _3952 13952 0.6148 1.00 0.92 0.4806 1.00 0.85 _4079 14079 0.6504 1.00 0.92 0.8932 1.00 0.96 _4083 14083 0.6687 1.00 0.92 0.5205 1.00 0.86 _3934 13934 0.6872 1.00 0.92 0.6954 1.00 0.90 _3911 13911 0.7314 1.00 0.94 0.9579 1.00 0.96 _4067 14067 0.7699 1.00 0.94 0.9580 1.00 0.96 _4008 14008 0.7901 1.00 0.94 0.9578 1.00 0.96 _3881 13881 0.7952 1.00 0.94 0.4757 1.00 0.85 _3844 13844 0.8550 1.00 0.97 0.6833 1.00 0.90 _3873 13873 0.9126 1.00 0.97 0.7420 1.00 0.93 _3888 13888 0.9453 1.00 0.97 0.5804 1.00 0.87 _3855 13855 0.9490 1.00 0.97 0.5805 1.00 0.87 _4055 14055 0.9563 1.00 0.97 0.9423 1.00 0.96 _3902 13902 0.9700 1.00 0.97 0.6365 1.00 0.90

Table 2.17- Correlation of H2B isoforms with Zap-70

Patient%Samples%Prior%to% All%Patient%Samples%(n%=%71) Treatment%and%with%Clinical% Outcome%Data%(n%=%54) H3.1%Fragment H3.1%Fragment Raw%P1 Adjusted%P Raw%P Adjusted%P _5267 15267 0.0032 0.16 0.0013 0.06 _5281 15281 0.0070 0.34 0.0023 0.11 _5404 15404 0.0362 1.00 0.0186 0.86 _5515 15515 0.0518 1.00 0.2994 1.00 _5498 15498 0.0833 1.00 0.1053 1.00 _5390 15390 0.0890 1.00 0.1743 1.00 _5332 15332 0.1084 1.00 0.0263 1.00 _5310 15310 0.1116 1.00 0.0451 1.00 _5527 15527 0.1129 1.00 0.1577 1.00 _5296 15296 0.1252 1.00 0.0366 1.00 _5537 15537 0.1301 1.00 0.3385 1.00 _5457 15457 0.1401 1.00 0.3853 1.00 _5317 15317 0.1508 1.00 0.1834 1.00 _5533 15533 0.1781 1.00 0.0605 1.00 _5547 15547 0.1801 1.00 0.0178 0.84 _5274 15274 0.1846 1.00 0.0861 1.00 _5489 15489 0.2017 1.00 0.3185 1.00 _5510 15510 0.2698 1.00 0.2327 1.00 _5485 15485 0.2742 1.00 0.5223 1.00 _5431 15431 0.2769 1.00 0.5835 1.00 _5472 15472 0.2823 1.00 0.6901 1.00 _5396 15396 0.3103 1.00 0.9654 1.00 _5434 15434 0.3245 1.00 0.4177 1.00 _5424 15424 0.3654 1.00 0.3227 1.00 _5522 15522 0.4071 1.00 0.5518 1.00 _5347 15347 0.4979 1.00 0.6086 1.00 _5461 15461 0.5156 1.00 0.6522 1.00 _5444 15444 0.5257 1.00 0.503 1.00 _5501 15501 0.5274 1.00 0.8786 1.00 _5325 15325 0.5643 1.00 0.2632 1.00 _5339 15339 0.5721 1.00 0.1475 1.00 _5376 15376 0.5889 1.00 0.3675 1.00 _5448 15448 0.6079 1.00 0.2536 1.00 _5410 15410 0.6447 1.00 0.8283 1.00 _5303 15303 0.6582 1.00 0.7031 1.00 _5260 15260 0.7270 1.00 0.6788 1.00 _5438 15438 0.7294 1.00 0.6457 1.00 _5476 15476 0.7416 1.00 0.8702 1.00 _5505 15505 0.8205 1.00 0.9066 1.00 _5480 15480 0.8530 1.00 0.9792 1.00 _5452 15452 0.8896 1.00 0.708 1.00 _5466 15466 0.8945 1.00 0.8417 1.00 _5416 15416 0.9021 1.00 0.8097 1.00 _5494 15494 0.9259 1.00 0.3572 1.00 _5353 15353 0.9265 1.00 0.4823 1.00 _5367 15367 0.9540 1.00 0.1962 1.00 _5543 15543 0.9906 1.00 0.6195 1.00 _5382 15382 0.9908 1.00 0.3624 1.00 _5289 15289 1.0000 1.00 0.5971 1.00

Table 2.18- Correlation of H3-1 isoforms with Zap-70 60

Patient%Samples%Prior%to% All%Patient%Samples%(n%=%71) Treatment%and%with%Clinical% Outcome%Data%(n%=%54)

H3.2%Fragment H3.2%Fragment Raw%P1 Adjusted%P Raw%P Adjusted%P _5371 15371 0.0042 0.23 0.0014 0.08 _5414 15414 0.0199 1.00 0.0225 1.00 _5442 15442 0.0320 1.00 0.0398 1.00 _5358 15358 0.0369 1.00 0.0336 1.00 _5329 15329 0.0380 1.00 0.0622 1.00 _5485 15485 0.0424 1.00 0.0295 1.00 _5294 15294 0.0515 1.00 0.0836 1.00 _5400 15400 0.0587 1.00 0.0122 0.66 _5344 15344 0.0618 1.00 0.0163 0.86 _5298 15298 0.0651 1.00 0.0726 1.00 _5493 15493 0.0678 1.00 0.1879 1.00 _5386 15386 0.1091 1.00 0.0398 1.00 _5261 15261 0.1177 1.00 0.1181 1.00 _5314 15314 0.1281 1.00 0.1845 1.00 _5499 15499 0.1552 1.00 0.0808 1.00 _5428 15428 0.2305 1.00 0.0282 1.00 _5457 15457 0.2366 1.00 0.0913 1.00 _5449 15449 0.2384 1.00 0.2335 1.00 _5506 15506 0.2456 1.00 0.1199 1.00 _5267 15267 0.2563 1.00 0.8006 1.00 _5534 15534 0.2869 1.00 0.2328 1.00 _5519 15519 0.2967 1.00 0.5221 1.00 _5283 15283 0.2968 1.00 0.3185 1.00 _5472 15472 0.3151 1.00 0.3101 1.00 _5538 15538 0.3471 1.00 0.3060 1.00 _5335 15335 0.3475 1.00 0.7825 1.00 _5468 15468 0.3494 1.00 0.1398 1.00 _5548 15548 0.4354 1.00 0.5012 1.00 _5461 15461 0.4526 1.00 0.2822 1.00 _5503 15503 0.4959 1.00 0.8179 1.00 _5351 15351 0.5124 1.00 0.7625 1.00 _5407 15407 0.5137 1.00 0.5717 1.00 _5464 15464 0.5392 1.00 0.9171 1.00 _5478 15478 0.5633 1.00 0.3356 1.00 _5543 15543 0.5870 1.00 0.6450 1.00 _5272 15272 0.6490 1.00 0.7408 1.00 _5321 15321 0.6582 1.00 0.5255 1.00 _5489 15489 0.6832 1.00 0.7170 1.00 _5279 15279 0.6900 1.00 0.4237 1.00 _5453 15453 0.6942 1.00 0.9033 1.00 _5303 15303 0.7165 1.00 0.7601 1.00 _5524 15524 0.7966 1.00 0.7754 1.00 _5436 15436 0.8034 1.00 0.7587 1.00 _5510 15510 0.8069 1.00 0.8335 1.00 _5514 15514 0.8084 1.00 0.6322 1.00 _5289 15289 0.8766 1.00 0.9501 1.00 _5264 15264 0.8832 1.00 0.6813 1.00 _5365 15365 0.8937 1.00 0.5830 1.00 _5528 15528 0.9346 1.00 0.6172 1.00 _5421 15421 0.9399 1.00 0.8441 1.00 _5306 15306 0.9562 1.00 0.7232 1.00 _5256 15256 0.9812 1.00 0.6768 1.00 _5392 15392 0.9899 1.00 0.5480 1.00 _5276 15276 1.0000 1.00 0.9404 1.00 _5378 15378 1.0000 1.00 0.9505 1.00

Table 2.19- Correlation of H3-2 isoforms with Zap-70 61

Patient#Samples#Prior#to# All#Patient#Samples#(n#=#84) Treatment#and#with#Clinical# Outcome#Data#(n#=#65) H4#Fragment H4#Fragment Raw#P1 Adjusted#P Raw#P Adjusted#P _1014 11014 0.0020 0.10 0.0082 0.41 _1002 11002 0.0062 0.30 0.0193 0.95 _1039 11039 0.0300 1.00 0.0756 1.00 _1542 11542 0.0862 1.00 0.0703 1.00 _1005 11005 0.1125 1.00 0.0462 1.00 _0965 10965 0.1465 1.00 0.0611 1.00 _1250 11250 0.1560 1.00 0.2054 1.00 _1019 11019 0.1620 1.00 0.1566 1.00 _1322 11322 0.1889 1.00 0.2018 1.00 _1484 11484 0.2228 1.00 0.2919 1.00 _1433 11433 0.2347 1.00 0.1370 1.00 _1043 11043 0.2478 1.00 0.3482 1.00 _1438 11438 0.2584 1.00 0.0503 1.00 _1461 11461 0.3023 1.00 0.5850 1.00 _1373 11373 0.3258 1.00 0.1880 1.00 _1511 11511 0.3810 1.00 0.2381 1.00 _1497 11497 0.3868 1.00 0.4146 1.00 _1333 11333 0.3917 1.00 0.8642 1.00 _1048 11048 0.4022 1.00 0.2202 1.00 _1384 11384 0.4065 1.00 0.5235 1.00 _1348 11348 0.4432 1.00 0.4298 1.00 _0974 10974 0.4482 1.00 0.8229 1.00 _1056 11056 0.4629 1.00 0.7161 1.00 _1519 11519 0.4812 1.00 0.2743 1.00 _1503 11503 0.4926 1.00 0.6451 1.00 _1292 11292 0.5098 1.00 0.7622 1.00 _1405 11405 0.5425 1.00 0.8985 1.00 _1535 11535 0.5751 1.00 0.8332 1.00 _1280 11280 0.6061 1.00 0.8849 1.00 _1034 11034 0.6565 1.00 0.9370 1.00 _1516 11516 0.6792 1.00 0.3698 1.00 _1488 11488 0.6924 1.00 0.3041 1.00 _0921 10921 0.7093 1.00 0.5847 1.00 _1076 11076 0.7164 1.00 0.8538 1.00 _1389 11389 0.7231 1.00 0.7322 1.00 _1526 11526 0.7400 1.00 0.6978 1.00 _1124 11124 0.7981 1.00 0.8848 1.00 _1265 11265 0.8121 1.00 0.8642 1.00 _0990 10990 0.8254 1.00 0.8843 1.00 _1419 11419 0.8401 1.00 0.9895 1.00 _0946 10946 0.8533 1.00 0.7507 1.00 _1363 11363 0.8541 1.00 0.8230 1.00 _1531 11531 0.8680 1.00 0.8344 1.00 _1028 11028 0.8713 1.00 0.7119 1.00 _1446 11446 0.9250 1.00 0.4532 1.00 _1378 11378 0.9321 1.00 0.6737 1.00 _1068 11068 0.9356 1.00 0.8952 1.00 _1473 11473 0.9428 1.00 0.7571 1.00 _1546 11546 0.9749 1.00 0.5982 1.00 _1306 11306 0.9893 1.00 0.8849 1.00

Table 2.20- Correlation of H4 isoforms with Zap-70

H2A Hazard(Ratio:( Lower(95%(CL Upper(95%(CL P Adjusted(P High(vs.(Low 14063 5.03 1.87 13.52 0.0004 0.0152 14197 0.45 0.21 0.95 0.0316 1.0000 13916 0.46 0.22 0.96 0.0342 1.0000 14132 0.47 0.23 0.97 0.0362 1.0000 13911 0.49 0.24 1.02 0.0501 1.0000 14046 1.93 0.92 4.04 0.0779 1.0000 14096 0.56 0.27 1.15 0.1073 1.0000 13958 1.78 0.86 3.70 0.1176 1.0000 14025 1.70 0.83 3.50 0.1447 1.0000 14018 1.71 0.81 3.62 0.1537 1.0000 13865 1.65 0.78 3.45 0.1835 1.0000 13943 0.54 0.21 1.43 0.2087 1.0000 14184 1.56 0.77 3.17 0.2186 1.0000 13890 0.65 0.32 1.33 0.2372 1.0000 13899 1.42 0.70 2.89 0.3239 1.0000 13932 1.42 0.68 2.96 0.3493 1.0000 14157 0.72 0.36 1.45 0.3546 1.0000 13831 0.72 0.36 1.46 0.3608 1.0000 13858 0.75 0.37 1.52 0.4278 1.0000 13817 0.75 0.36 1.55 0.4299 1.0000 14083 0.75 0.36 1.59 0.4544 1.0000 14057 0.78 0.39 1.56 0.4743 1.0000 14142 0.77 0.37 1.58 0.4749 1.0000 14112 0.77 0.38 1.57 0.4750 1.0000 14147 0.80 0.39 1.61 0.5256 1.0000 14116 1.23 0.61 2.47 0.5614 1.0000 13906 1.19 0.58 2.42 0.6400 1.0000 14168 1.15 0.57 2.33 0.6945 1.0000 14103 0.89 0.43 1.82 0.7401 1.0000 13875 1.11 0.55 2.23 0.7726 1.0000 14173 0.91 0.45 1.83 0.7841 1.0000 13996 1.09 0.53 2.24 0.8109 1.0000 14070 1.09 0.54 2.21 0.8161 1.0000 13940 1.08 0.53 2.17 0.8361 1.0000 14004 0.94 0.47 1.89 0.8570 1.0000 13987 0.96 0.47 1.94 0.9047 1.0000 13847 1.03 0.51 2.06 0.9420 1.0000 14033 1.02 0.50 2.08 0.9580 1.0000

Table 2.21- Histone H2A-Summary results for treatment

H2B Hazard(Ratio:( Lower(95%(CL Upper(95%(CL P Adjusted(P High(vs.(Low

13881 1.71 0.84 3.48 0.13 1.00 13971 1.69 0.84 3.41 0.14 1.00 14055 1.69 0.83 3.42 0.14 1.00 14032 0.55 0.22 1.34 0.18 1.00 14002 1.61 0.78 3.33 0.19 1.00 14008 0.64 0.31 1.31 0.22 1.00 14092 0.64 0.32 1.31 0.22 1.00 13747 1.53 0.75 3.10 0.24 1.00 13805 1.51 0.73 3.13 0.27 1.00 14021 0.68 0.33 1.41 0.30 1.00 13759 0.71 0.35 1.42 0.33 1.00 13934 1.41 0.70 2.84 0.33 1.00 13987 1.41 0.69 2.87 0.34 1.00 14083 0.71 0.34 1.45 0.34 1.00 14067 1.40 0.69 2.85 0.34 1.00 14095 0.72 0.35 1.46 0.36 1.00 13831 0.74 0.36 1.50 0.40 1.00 14048 1.32 0.65 2.69 0.44 1.00 13963 0.77 0.38 1.54 0.46 1.00 13921 1.28 0.64 2.58 0.49 1.00 13957 0.78 0.39 1.58 0.49 1.00 13977 0.78 0.38 1.59 0.50 1.00 13819 1.23 0.60 2.50 0.57 1.00 13928 1.22 0.61 2.46 0.57 1.00 13790 0.82 0.41 1.65 0.58 1.00 13940 0.83 0.41 1.68 0.61 1.00 13775 0.85 0.42 1.73 0.65 1.00 13902 1.15 0.56 2.36 0.71 1.00 13873 1.14 0.57 2.27 0.72 1.00 13877 0.89 0.43 1.84 0.74 1.00 13844 0.91 0.45 1.85 0.79 1.00 13898 0.91 0.45 1.86 0.80 1.00 13944 1.09 0.54 2.20 0.80 1.00 13839 0.92 0.46 1.85 0.82 1.00 13855 0.95 0.47 1.90 0.87 1.00 14079 0.98 0.49 1.96 0.95 1.00 13911 1.02 0.50 2.07 0.96 1.00 13888 1.01 0.50 2.02 0.98 1.00 13952 1.00 0.50 2.01 1.00 1.00

Table 2.22- Histone H2B-Summary results for treatment

H3_1 Hazard)Ratio:) Lower)95%)CL Upper)95%)CL P Adjusted)P High)vs.)Low 15289 0.37 0.16 0.87 0.019 0.911 15325 0.40 0.18 0.89 0.020 0.965 15332 2.50 1.11 5.61 0.022 1.000 15339 0.48 0.22 1.06 0.063 1.000 15416 0.47 0.20 1.10 0.073 1.000 15303 0.52 0.22 1.24 0.134 1.000 15367 0.59 0.27 1.31 0.189 1.000 15267 0.60 0.27 1.31 0.194 1.000 15424 0.62 0.28 1.34 0.216 1.000 15527 1.68 0.73 3.87 0.217 1.000 15404 0.63 0.29 1.37 0.237 1.000 15515 1.57 0.71 3.50 0.262 1.000 15457 1.58 0.69 3.59 0.272 1.000 15476 0.65 0.29 1.43 0.276 1.000 15510 0.66 0.30 1.45 0.292 1.000 15489 0.66 0.30 1.45 0.300 1.000 15485 1.51 0.69 3.28 0.301 1.000 15522 0.68 0.31 1.51 0.340 1.000 15296 0.69 0.31 1.53 0.362 1.000 15347 1.50 0.60 3.76 0.388 1.000 15543 0.73 0.33 1.62 0.437 1.000 15533 0.73 0.33 1.62 0.440 1.000 15505 0.74 0.34 1.63 0.455 1.000 15260 1.34 0.61 2.92 0.467 1.000 15547 0.75 0.34 1.68 0.483 1.000 15431 1.41 0.52 3.82 0.494 1.000 15444 1.30 0.59 2.87 0.523 1.000 15390 0.78 0.34 1.80 0.555 1.000 15353 0.79 0.36 1.75 0.567 1.000 15537 1.25 0.57 2.73 0.574 1.000 15501 0.80 0.37 1.76 0.584 1.000 15498 0.80 0.35 1.82 0.594 1.000 15281 0.83 0.39 1.80 0.640 1.000 15434 0.81 0.32 2.04 0.653 1.000 15410 0.84 0.38 1.86 0.670 1.000 15310 0.84 0.38 1.86 0.672 1.000 15472 0.85 0.39 1.86 0.688 1.000 15480 0.86 0.40 1.85 0.692 1.000 15438 1.16 0.54 2.52 0.702 1.000 15466 1.14 0.51 2.52 0.752 1.000 15494 1.12 0.51 2.45 0.783 1.000 15448 0.90 0.41 2.00 0.804 1.000 15382 0.91 0.41 2.00 0.808 1.000 15461 1.09 0.51 2.37 0.819 1.000 15317 0.92 0.42 2.04 0.843 1.000 15396 1.08 0.49 2.38 0.852 1.000 15376 1.08 0.46 2.50 0.865 1.000 15274 1.04 0.43 2.47 0.937 1.000 15452 1.00 0.46 2.18 0.994 1.000

Table 2.23- Histone H3-1-Summary results for treatment

H3_2 Hazard)Ratio:) Lower)95%)CL Upper)95%)CL P Adjusted)P High)vs.)Low 15371 2.70 1.20 6.09 0.012 0.682 15499 2.42 1.08 5.44 0.028 1.000 15468 2.41 1.05 5.55 0.033 1.000 15449 2.36 1.02 5.43 0.040 1.000 15489 2.30 1.02 5.20 0.040 1.000 15442 2.11 0.96 4.68 0.060 1.000 15400 2.04 0.92 4.53 0.072 1.000 15428 1.97 0.89 4.37 0.088 1.000 15386 1.95 0.88 4.30 0.093 1.000 15365 0.52 0.22 1.19 0.113 1.000 15524 0.59 0.27 1.30 0.187 1.000 15344 1.67 0.75 3.70 0.201 1.000 15506 1.65 0.74 3.70 0.220 1.000 15414 1.61 0.73 3.55 0.238 1.000 15485 1.58 0.73 3.44 0.242 1.000 15503 0.60 0.25 1.44 0.250 1.000 15276 1.59 0.71 3.54 0.255 1.000 15289 1.56 0.72 3.41 0.258 1.000 15279 1.84 0.62 5.50 0.266 1.000 15298 1.52 0.70 3.31 0.285 1.000 15314 1.53 0.69 3.40 0.290 1.000 15407 1.48 0.67 3.26 0.332 1.000 15392 1.52 0.62 3.73 0.354 1.000 15306 1.45 0.65 3.20 0.360 1.000 15519 1.44 0.66 3.16 0.361 1.000 15358 1.43 0.65 3.17 0.369 1.000 15528 1.43 0.65 3.17 0.372 1.000 15534 1.37 0.62 3.03 0.430 1.000 15267 0.74 0.33 1.64 0.458 1.000 15378 1.37 0.60 3.12 0.459 1.000 15461 1.50 0.50 4.51 0.471 1.000 15335 1.33 0.61 2.88 0.473 1.000 15514 1.31 0.60 2.84 0.494 1.000 15294 0.74 0.30 1.79 0.498 1.000 15548 1.31 0.60 2.86 0.503 1.000 15493 1.27 0.58 2.77 0.548 1.000 15472 0.81 0.36 1.82 0.608 1.000 15303 0.81 0.32 2.04 0.654 1.000 15478 1.15 0.53 2.50 0.722 1.000 15321 1.14 0.52 2.51 0.742 1.000 15421 1.14 0.47 2.80 0.771 1.000 15283 1.12 0.52 2.42 0.779 1.000 15272 1.12 0.51 2.46 0.780 1.000 15538 0.90 0.41 1.96 0.784 1.000 15543 1.11 0.51 2.42 0.787 1.000 15436 0.88 0.35 2.25 0.795 1.000 15510 1.09 0.51 2.37 0.820 1.000 15261 1.10 0.50 2.42 0.824 1.000 15329 1.07 0.48 2.37 0.874 1.000 15457 1.06 0.49 2.30 0.889 1.000 15256 1.05 0.48 2.28 0.909 1.000 15453 0.96 0.44 2.07 0.916 1.000 15264 1.02 0.45 2.29 0.968 1.000 15351 0.99 0.46 2.15 0.979 1.000 15464 1.01 0.38 2.71 0.984 1.000

Table 2.24- Histone H3-2-Summary results for treatment

H4 Hazard'Ratio:' Lower'95%'CL Upper'95%'CL P Adjusted'P High'vs.'Low 11014 2.66 1.26 5.62 0.008 0.390 11005 2.37 1.16 4.86 0.015 0.735 11076 2.54 1.17 5.50 0.015 0.735 11002 0.44 0.21 0.90 0.020 0.954 11433 0.46 0.22 0.93 0.027 1.000 11484 0.55 0.27 1.14 0.101 1.000 11497 0.55 0.26 1.15 0.107 1.000 11384 0.57 0.28 1.15 0.111 1.000 11535 1.72 0.83 3.55 0.139 1.000 11461 1.69 0.83 3.46 0.144 1.000 11043 1.64 0.80 3.39 0.173 1.000 11048 0.62 0.30 1.25 0.176 1.000 11034 1.63 0.80 3.34 0.178 1.000 11292 0.65 0.32 1.29 0.214 1.000 11265 1.54 0.76 3.12 0.227 1.000 10965 0.65 0.32 1.34 0.240 1.000 11405 0.67 0.33 1.36 0.267 1.000 11363 1.43 0.71 2.89 0.317 1.000 11438 0.71 0.35 1.41 0.324 1.000 11250 1.37 0.69 2.73 0.368 1.000 11068 1.37 0.68 2.77 0.384 1.000 11039 0.74 0.37 1.47 0.391 1.000 11306 0.75 0.37 1.51 0.419 1.000 11028 1.31 0.65 2.65 0.449 1.000 11419 1.29 0.64 2.61 0.479 1.000 11373 1.27 0.63 2.55 0.499 1.000 11546 1.28 0.63 2.60 0.503 1.000 11516 0.79 0.39 1.58 0.505 1.000 11056 1.26 0.63 2.49 0.515 1.000 11531 0.82 0.41 1.62 0.566 1.000 11322 0.82 0.41 1.63 0.568 1.000 11488 1.22 0.61 2.45 0.580 1.000 10990 0.83 0.42 1.66 0.602 1.000 11333 0.84 0.42 1.68 0.614 1.000 10974 0.84 0.42 1.68 0.623 1.000 11542 0.84 0.42 1.71 0.637 1.000 11473 0.87 0.43 1.74 0.690 1.000 10921 0.88 0.44 1.79 0.730 1.000 11124 1.12 0.57 2.23 0.742 1.000 11280 1.12 0.56 2.26 0.744 1.000 11378 0.89 0.45 1.77 0.746 1.000 11389 0.91 0.46 1.81 0.796 1.000 10946 0.92 0.46 1.82 0.801 1.000 11503 1.07 0.53 2.16 0.856 1.000 11348 0.94 0.47 1.88 0.866 1.000 11019 1.04 0.52 2.07 0.919 1.000 11511 1.04 0.51 2.12 0.921 1.000 11526 0.98 0.50 1.95 0.961 1.000 11519 0.99 0.48 2.01 0.967 1.000 11446 1.00 0.50 2.00 0.996 1.000

Table 2.25- Histone H4-Summary results for treatment

CHAPTER 3

Increasing The Complexity Of Chromatin: Functionally Distinct Roles For

Replication-Dependent Histone H2A Isoforms In Cell Proliferation And

Carcinogenesis

3.1 Abstract

Replication-dependent histones are encoded by multigene families found in several large clusters in the human genome and are thought to be functionally redundant. However, the abundance of specific replication-dependent isoforms of histone H2A are altered in patients with chronic lymphocytic leukemia (CLL). Similar changes in the abundance of

H2A isoforms are also associated with the proliferation and tumorigenicity of bladder cancer cells. To determine whether these H2A isoforms can perform distinct functions, expression of several H2A isoforms was reduced by siRNA knockdown. Reduced expression of the HIST1H2AC locus leads to increased rates of cell proliferation and tumorigenicity. We also observe that regulation of replication-dependent histone H2A expression can occur on a gene-specific level. Specific replication-dependent histone

H2A genes are either up- or down-regulated in CLL tumor tissue samples. In addition, discreet elements are identified in the 5’ UTR of the HIST1H2AC locus that that confer translational repression. Taken together, these results indicate that replication-dependent

68 histone isoforms can possess distinct cellular functions and that regulation of these isoforms may play a role in carcinogenesis.

3.2 Introduction

At the most fundamental level, chromatin is composed of a repeated structure known as the nucleosome. Each nucleosome consists of ~147 base pairs of DNA wrapped around a protein complex called the histone octamer that contains two molecules of each of the four core histones (H2A, H2B, H3 and H4). The importance of chromatin structure for the packaging and regulation of eukaryotic genomes is evidenced by the extraordinary conservation of this structure throughout eukaryotic evolution. In fact, the core histones are among the most highly conserved eukaryotic proteins with many residues being completely invariant (81). However, despite this seeming uniformity, one of the most important characteristics of chromatin structure is complexity, which is necessary for encoding all of the regulatory information necessary for the proper execution of nuclear processes and for epigenetic inheritance.

The complexity of chromatin is derived from two main sources: the post-translational modification of histones and the presence of histone variants. Histones are subject to multiple forms of post-translational modification (6). Further complexity is derived from the fact that the cellular complement of most histones is not homogeneous but, rather, is composed of multiple primary sequence variants (19,20,82).

Histone variants can be distinguished on a number of levels. The first is the distinction between replication-dependent and replication-independent histones. Replication- 69 dependent histones become highly expressed just prior to S-phase and are then repressed at the completion of DNA replication (83). Interestingly, the DNA replication-dependent histone genes are found in several large clusters that contain dozens of histone genes and they are the only protein coding mRNAs produced in mammalian cells that lack a poly(A) tail. Instead of a poly(A) tail, these messages contain a short, highly conserved stem loop structure in their 3’ UTR and their processing and stability are regulated by the stem loop binding protein (SLBP) which specifically interacts with this structure (69,84).

The DNA replication-dependent histones are used for the assembly of chromatin structure during DNA replication. Hence, the packaging of genomic DNA with the DNA replication-dependent histones is the “ground state” at which chromatin structure begins.

There are also a large number of core histone genes that are constitutively expressed throughout the cell cycle and, hence, are known as replication-independent histone variants. The replication-independent histones differ in primary sequence from the replication-dependent histones with these variations ranging from only a handful of amino acid changes to the incorporation of large non-histone domains. Well- characterized examples of replication-independent histone variants include histones H3.3,

H2AX, H2AZ and macroH2A (82). In addition to changes in protein sequence, the replication-independent histone genes also differ from their replication-dependent counterparts in that they are found as single genes dispersed throughout the genome and they generate transcripts with normal poly(A) tails.

While the replication-dependent core histones are considered to be the “canonical” histones, there is actually a wide range of primary sequence variations within this group 70

(69). To distinguish these histone variants from the replication-independent histone variants, they will be referred to here as histone isoforms. Each of the replication- dependent histones is encoded by multiple genes (H2A - 16 genes, H2B - 22 genes, H3 -

14 genes, H4 - 14 genes and H1 - 6 genes). Hence, the presence of distinct replication- dependent histone isoforms has the potential to significantly expand the complexity of mammalian chromatin structure. However, these core histone isoforms have not been studied in detail as they have been presumed to encode functionally equivalent molecules.

Mass spectrometry based analysis of the histone H2A complement in HeLa cells indicated that the most abundant replication-dependent isoforms were the products of several specific genes (52). The most abundant form is encoded by five distinct genes

(HIST1H2AG, HIST1H2AI, HIST1H2AK, HIST1H2AL and HIST1H2A, see Figure 3.1-

A). The second most abundant species is encoded by a single gene, HIST1H2AC, and the third most abundant species is encoded by two genes, HIST1H2AB and HIST1H2AE.

The nomenclature that has developed to describe these histone variants has not systematically addressed the naming of these replication-dependent histone isoforms

(69,70). Therefore, they will be referred to using a nomenclature that is based on the more systematic naming of the genes that encode these proteins. The names of the replication-dependent histone genes provide important information about the location of the gene. The first part of the gene name refers to the histone gene cluster in which it resides (i.e. HIST1, HIST2 or HIST3). The next part of the name refers to the specific type of histone encoded by the gene (i.e. H1, H2A, H2B, H3 or H4). Finally, the multiple

71 copies of each histone type are designated alphabetically based on their order within each cluster (centromere distal to proximal). Therefore, HIST1H2AC refers to the third histone

H2A gene present in histone cluster 1. In referring to the H2A protein isoforms, the most abundant form of H2A (encoded by 5 genes) will be considered to be the “canonical” histone H2A and will simply be referred to as H2A. Based on the genetic nomenclature, the second most abundant isoform will be referred to as H2A 1C (for HIST1H2AC).

Likewise, the third most abundant isoform will be referred to as H2A 1B/E (Figure 3.1-

A).

The replication-dependent histone isoforms are generally thought to be functionally interchangeable. However, in a recent study, the LC/MS profile of core histones isolated from B cells from patients with chronic lymphocytic leukemia (CLL) was compared with the profile of histones isolated from CD19+ B cells from healthy individuals.

Surprisingly, a decrease in the abundance of H2A 1C and H2A 1B/E was observed in a high percentage of the CLL samples (41). We now report that similar changes in H2A

1C and H2A 1B/E were found to in bladder cancer cells. This led us to hypothesize that replication-dependent isoforms can be functionally distinct. We present two lines of evidence to support this hypothesis. First, siRNA knockdown of H2A 1C and H2A 1B/E, but not canonical H2A, led to increased cell proliferation and tumorigenicity. Second, the expression of replication-dependent H2A isoforms can be subject to individualized levels of regulation. The 5’ UTR (untranslated region) of the mRNA encoding H2A 1C uniquely imparts translational repression. We also demonstrate that this repression maps to a specific duplicated sequence element found in this 5’ UTR. These observations

72 suggest that replication-dependent histone isoforms can be functionally distinct and play specific roles in the regulation of cell growth.

3.3 Materials and methods

Cell growth- Human non-invasive (RT4), and invasive (T24) bladder cancer cell lines, and 293TN and U2OS cell lines were purchased from the American Type Culture

Collection (ATCC; Manassas, VA). Immortalized (hTERT) and normal human bladder epithelial cells were obtained from Dr. Margareta Knowles (St James’s University

Hospital, Leeds, UK) and Lonza Walkersville, Inc (Walkersville, MD), respectively.

293TN and U2OS cells were cultured in DMEM (Sigma, St. Louis, MO), and RT4 and

T24 cells were cultured in RPMI 1640 medium (GIBCO, Grand Island, NY); all supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT), 2 mM L- glutamine, and penicillin (10 U/ml) and streptomycin (10 mg/ml) (GIBCO, Grand Island,

NY). hTERT and normal human bladder epithelial cells were grown in Keratinocyte

Serum Free Medium (KSFM; GIBCO, Grand Island, NY) with supplements of bovine pituitary extract and epidermal growth factor plus 30 ng/ml Cholera toxin (Sigma

Chemical Company, St. Louis, MO). All the other cell lines and primary cells used in our study were also purchased from ATCC. The cells were grown in DMEM, supplemented with 10% FBS and 1% penicillin/streptomycin (Sigma). For the Human Foreskin

Fibroblast (HFF) cells, the growth medium was supplemented with 12.5% FBS. All the cell types as described above were incubated at 37 °C in a humidified atmosphere with

5% CO2 and 95% air.

Histone Extraction- Cells were seeded, grown for 24-48 hr (based on our previous experiences and the growth kinetics of each cell line), harvested by scraping and then snap frozen. Histones were extracted as described previously (71). Briefly, cell pellets were resuspended in 1 mL of NP-40 lysis buffer [10 mM Tris-HCl (pH 7.4), 10 mM

NaCl, 3 mM MgCl2, 0.5% NP-40, 0.15 mM spermine, 0.5 mM spermidine, 1 mM PMSF, protease inhibitor cocktail (1:1000)] and incubated on ice for 5 min. The nuclei were pelleted at 483 x g for 15 min at 4 °C and the pellet washed with 1 mL TBS [10 mM Tris-

HCl (pH 7.4), 150 mM NaCl]. Sulfuric acid (H2SO4) (0.2 M) was added to the washed pellet to extract the histones and it was vortexed and incubated on ice for 30 min. The solution was centrifuged at 12,045 x g for 15 min at 4 °C to remove the cellular debris.

Eighty percent acetone was added to the supernatant and precipitated at -20 °C overnight.

The precipitated histones were centrifuged at 12,045 x g for 15 min at 4 °C, allowed to air-dry for 10 min and resuspended in HPLC water.

LC-MS- Protein concentration was determined by Bradford analysis. Thirty micrograms of purified histones were characterized by LC-MS analysis. LC-MS analysis was performed with a Dionex U3000 HPLC (Dionex; Sunnyvale, CA) coupled to a

MicroMass Q-Tof (MicroMass, Whythenshawe, UK). Reversed-phase separation was carried out on a Discovery Bio Wide Pore C18 column (1.0 mm x 150 mm, 5 µm, 300 Å;

Supelco, USA). Mobile phases A and B consisted of water and acetonitrile with 0.05% trifluoroacetic acid, respectively. The flow rate was 25 µL/min and the gradient started at

20% B, increased linearly to 30% B in 2 min, to 35% B in 8 min, 50% B in 20 min, 60%

B in 5 min and 95% B in 1 min. After washing at 95% B for 4 min, the column was

74 equilibrated at 20% B for 30 min and a blank was run between each sample injection.

The cone voltage on the Q-Tof was 25 V. LC-MS data was deconvoluted using

MassLynx 4.1

Real time PCR- RNA was isolated from the cell lines using the Trizol (Invitrogen). The

RNA was reverse transcribed using High Capacity cDNA Reverse Transcription Kit

(Applied Biosystems) and 50 ng of cDNA was used as template for qPCR reaction. The primers for all the histone isoforms and Taqman mastermix were purchased commercially (Applied Biosystems). The reaction was carried out on an Applied

Biosystems 7300 Real Time PCR systems instrument and relative quantitation carried out using 2-ΔΔCt method (85). siRNA transfection- The siRNAs for each histone isoform were from Applied

Biosystems. Transfection was carried out using siPORT NeoFX transfection reagent

(invitrogen) as per manufacturer’s instructions.

Cell proliferation assay- For each cell type the initial count was carried out using a hemocytometer and the cells were plated in 6-well plates in triplicate for every 12 hour reading. At each interval, the cells were trypsinised and counted in a Z1 coulter particle counter (Beckman Coulter).

Soft agar assay- For soft agar assay, 1% base agar was prepared and allowed to cool to

40˚C in a water bath and mixed with DMEM, supplemented with 15% FBS to make a final agar concentration of 0.5%. 5 ml of this molten mixture was added in a 100 mm plate and allowed to solidify for 10 min. For preparation of top agarose layer, 1.5% agarose was prepared and mixed 1:1 with DMEM. Each of the siRNA transfected cells 75 and control cells were trypsinized and added to this agar-DMEM mixture to a final agarose concentration of 0.7% and 400,000 cells per plate. The plates were incubated at

37˚C and images were taken after 4-7 days using Zeiss Axioskop Widefield Light

Microscope. For quantitative soft agar assay, we used CytoSelect™ 96-Well Cell

Transformation Assays kit (Cell Biolabs) as per manufacturer’s instructions.

Luciferase Assay- Oligos corresponding to the 5' UTR of each of the histone H2A isoforms were designed and cloned upstream of the promoter of the pcDNA-Luc vector.

The cloned vectors were then transfected into different cell lines using lipofectamine

(invitrogen). Each of the isoforms was also co-transfected with a Renilla luciferase vector as an internal standard and the transfection was carried out in triplicates in 6-well plates. After 36 hrs, the cells were harvested, lysed and luciferase levels estimated using

Dual Glo luciferase assay kit (Promega) as per manufacturers protocol. For HuR knockdown, cell lines were treated with siHuR (Santa Cruz Biotechnology) for 6-8 hrs before transfection with the luciferase vector. Luciferase levels were estimated as described above and one set of cell lines was also used for western analysis using HuR antibody (Santa Cruz Biotechnology).

Generation of stable cell lines- Lentiviral constructs encoding each of the histone isoforms with an N-terminal FLAG and His tags were purchased commercially

(Invivogen). Transfection and subsequent transduction was carried out using

Lipofectamine as per manufacturer’s protocol.

Localization studies- Stable cell lines containing each of the H2A isoforms were grown overnight in a chamber slide. The slides were then washed with PBS, and 76 immunostaining was carried out using anti-FLAG antibody (Sigma) and Cy-3 and FITC conjugated secondary antibodies (Jackson Labs). The cells were then counterstained with

DAPI and visualized under fluorescent microscope (Olympus). To precisely pinpoint their location on the chromosomes (heterochromatic or euchromatic region), metaphase- staining procedure was carried out. The cells were synchronized by serum starvation and colcemid (Invitrogen) was added to arrest the cells in metaphase. The cells were then lysed on glass slides, air-dried and scanned under a microscope. The regions with appropriate chromosome spreads were encircled with a hydrophobic marker, followed by the immunostaining and visualization of the cells as described above.

Common Bisulfite Restriction Analysis (COBRA)- The cell lines for COBRA analysis were a gift from Dr Jacob lab. Four distinct cell lines were used- Mec 1 wild type, Mec 1 with DNA methyltansferase 1 (Dnmt1) knocked out, Mec 1 with DNA methyltansferase

3b (Dnmt 3b) knocked out and double knockout (dKO). Bisulfite treatment of genomic

DNA from all the Mec1 cell types was carried out using the standard protocol (sadri).

MThe treated samples were subjected to PCR. The reaction conditions included 1 cycle at

94OC for 4 min after which Taq polymerase was added, followed by 45 cycles of 94OC for

30 sec, 55OC for 30 sec and 72OC for 30 sec and final extension for 10 min at 72OC. The amplified product was digested using suitable restriction enzymes and the digestion pattern was examined for each histone H2A isoform.

MassARRAy methylation analysis- The genomic DNA from CD19+ B cells from healthy donors, PB lymphocytes from healthy donors and unselected CLL samples was subjected to bisulfite treatment followed by PCR with T7-promotortagged reverse

77 primers. After SAP treatment, in vitro transcription was carried out and the generated transcript was subjected to an enzymatic base specific cleavage. The fragment mass was then determined by MALDITOF MS and the EpiTYPER software was used to generate an information report. Amplicon mean values were plotted using the Prism software.

3.4 Results

Characterization of the histone H2A complement in bladder cancer cells. To determine whether alterations in replication-dependent histone H2A isoforms is a phenomenon unique to CLL or whether it might be a more common aspect of tumorigenic cells, we analyzed the complement of H2A proteins in a completely unrelated tissue and tumor type. Procedures that had previously been validated were used where histones purified by acid extraction were resolved and analyzed by LC/MS (41).

The relative abundance of each histone H2A isoform was designated as the area under the curve representing each isoform divided by the area representing the total of all of the

H2A isoforms. Figure 3.1-B shows the LC/MS profile of histone H2A from normal human bladder epithelial cells, immortalized human bladder epithelial cells (hTERT) and non-invasive (RT4) and invasive (T24) human bladder cancer cell lines. It is clear that there was a significant decrease in the relative abundance of H2A 1C and H2A 1B/E as the bladder cells acquired the ability to proliferate more vigorously and that there was a further decrease as the cells become tumorigenic. These results suggested that regulation of the abundance of specific replication-dependent H2A isoforms was not limited to CLL and, hence, may be a more general aspect of carcinogenesis.

Localization pattern of histone isoforms. Since histone isoforms show a differential phenotype, we hypothesized them to have a distinct localization pattern. From the whole cell immunofluoresence study, we can clearly observe that each of the histone H2A isoform is located in the nucleus (Figure 3.3). Additionally, our initial analysis with the metaphase staining revealed a differential localization pattern on the chromosomes

(Figure 3.4). However, the staining procedure did not work out well even after our repeated attempts and with the use of multiple primary and secondary antibodies.

Functional analysis of replication-dependent histone H2A isoforms. To determine whether replication-dependent histone H2A isoforms are functionally distinct, we used an siRNA strategy to down-regulate the H2A isoforms and determine whether a decrease in specific isoforms leads to distinct effects on cell growth and proliferation. 293TN cells were transfected with either control siRNAs, a mixture of siRNAs that target all five genes that encode canonical H2A, a pair of siRNAs that target the H2A 1B/E genes or an siRNA targeting the H2A 1C gene. The specificity of the siRNA knockdown was confirmed using real time PCR assays. As seen in figure 3.5-A, the abundance of only the targeted mRNAs was specifically decreased.

The effect of the siRNA knockdowns on the rate of cell proliferation was tested by measuring the growth rate of cells treated with each of the siRNAs. As seen in Figure

3.5-B, siRNA knockdown of the canonical H2A genes had no effect on cell proliferation, knock down of H2A 1B/E caused a modest increase in cell proliferation and siRNA knockdown of H2A 1C resulted in a marked increase in the rate of cell proliferation.

To determine whether the siRNA knockdown of the H2A isoforms had an impact on cell tumorigenicity, soft agar assays were performed following the siRNA treatment of cells.

Strikingly, the knockdown of H2A 1C induced the formation of large colonies (Figure

3.5-C). Knockdown of H2A 1B/E resulted in a small increase in colony formation while depletion of canonical H2A had a slightly negative impact on colony formation. The effect of H2A isoform knockdown on cell proliferation in soft agar was also quantified using a fluorescent dye-based assay with similar results observed (Figure 3.5-D).

To determine whether the effect of H2A isoform modulation on cell proliferation and tumorigenicity was cell type specific, we repeated the analyses using a different cell line.

As seen in Figure 3.6-A, specific siRNA-based depletion of H2A isoforms was also achieved in U2OS cells. Strikingly, the effect of this H2A isoform on cell proliferation and growth in soft agar was very similar to that seen with 293TN cells. Namely, siRNA knockdown of H2A 1C caused a significant increase in cell growth (Figure 3.6-B). and a dramatic increase in the colony size observed during growth on soft agar, while siRNA knockdown of H2A 1B/E has a smaller effect on both phenotypes (Figure 3.6-B-D).

If replication-dependent histone H2A isoforms are functionally equivalent, depletion of any of the isoforms would be predicted to have a similar effect on cell proliferation. Our results clearly show that this is not the case and that knockdown of specific isoforms can have different effects on cellular phenotypes.

Regulation of replication-dependent histone H2A isoform gene expression. The genes encoding the replication-dependent histone isoforms are grouped together in several large clusters. These histone clusters localize to Cajal Bodies in the nucleus, 80 which are thought to be the sites at which the histone genes are transcribed. Based on the clustering and co-localization of the histone genes, it is presumed that they are transcriptionally regulated as a unit through common mechanisms. However, if replication-dependent isoforms, such as H2A 1C and H2A 1B/E, have distinct cellular functions, it would be predicted that there might be isoform specific gene regulatory mechanisms. Therefore, we have begun to analyze the expression of these genes at the mRNA level.

Total RNA was isolated from CD19+ B cells selected from healthy volunteers. In addition, RNA was isolated from the B cells of >100 CLL patients. Real time PCR assays were used to measure mRNA levels for the 8 replication-dependent H2A genes that encode the canonical H2A, H2A 1C and H2A 1B/E (Figure 3.7). In addition to an internal control (PGK1), the PCR reactions were also normalized to a control reaction that contained equal quantities of each H2A gene so that the absolute abundance of the mRNAs could be compared. In cells from the healthy individuals, most of the H2A isoform mRNAs are found at similar levels. The striking exception is HIST1H2AE, which is present at levels that are approximately 20-fold higher than the other isoforms.

Also, while most of the replication-dependent H2A isoforms did not significantly change in the CLL patient cells, three of the isoforms showed significant changes in abundance in these tumor cells. The HIST1H2AB and HIST1H2AM mRNAs were down-regulated in CLL and the HIST1H2AC mRNA was upregulated. These results indicate that individual replication-dependent histone H2A isoform genes can be subject to individual regulation at the mRNA level.

Replication-dependent H2A isoforms may also be subject to post-transcriptional regulation. To explore this idea, we have begun to dissect the H2A transcripts to identify functional regulatory elements. As would be expected from the high level of protein sequence conservation, the coding sequences of the eight histone H2A genes that encode the majority of H2A isoforms are nearly identical. Outside of the coding sequences, there are two blocks of sequence variation that could potentially provide platforms for regulation. The first is the 5’ untranslated region (UTR) and the second is a short stretch of the 3’ UTR between the stop codon and the invariant stem-loop.

To determine whether the 5’UTRs of the replication-dependent histone H2A genes influence gene expression, each 5’UTR was cloned directly upstream of the coding sequence of a luciferase reporter construct. Each vector was transfected into cells (along with a renilla reporter plasmid for normalization) and the luciferase levels were determined by fluorescence intensity. As seen in Figure 3.8-A, the 5’ UTR of the

HIST1H2AC locus was unique in its ability to alter luciferase expression (note that the mRNA for two of the H2A isoforms have no 5’ UTR). In multiple cell lines, the

HIST1H2AC 5’ UTR caused a significant down-regulation of luciferase (typically 3- to

5-fold) suggesting that this sequence could mediate translational repression. Importantly, a scrambled version of the HIST1H2AC 5’ UTR, which contains the same base composition but with a randomized sequence, does not alter expression when fused to luciferase (Figure 3.8-B). Taken together, these results demonstrate that replication- dependent histone H2A genes are subject to individualized regulation at the transcriptional and post-transcriptional levels.

Analysis of the HIST1H2AC 5’ UTR. To determine whether the HIST1H2AC 5’ UTR contains discrete sequences that mediate its regulatory activity; we performed a deletion analysis of this element. 10 base deletions were made across the length of the 5’ UTR in the context of the luciferase fusion (Figure 3.9-A). As seen in figure 3.9-B, two sets of contiguous deletions resulted in a loss of translational repression. This suggested that two regions of the 5’ UTR were important for its regulatory activity. These regions span bases 21-50 and 71-88. Conversely, deletion of bases 1-20 and 51-70 had little impact on the repression mediated by the HIST1H2AC 5’ UTR. Importantly, identical results were observed in multiple cell lines (Figure 3.9-B).

Comparison of the sequences between 21-50 and 71-88 of the HIST1H2AC 5’ UTR indicated that they each region contained a highly similar 15/16 base sequence (Figure

3.9-C). These sequences span bases 28-42 and 71-86 of the HIST1H2AC 5’ UTR and are identical at 13 positions and the position of these sequences is entirely consistent with the results of the deletion analysis. To more precisely probe the functional significance of this sequence, mutations were constructed that altered the sequence of either the first element (MUT1, Figure 3.9-D), the second element (MUT2, Figure 3.9-D) or both elements (MUT1 + MUT2). As shown in Figure 3.9-D, neither the MUT1 nor MUT2 construct showed any decrease in repression. However, mutating both sites caused a complete loss of translational repression. These results indicate that this conserved sequence plays a critical role in the post-transcriptional regulation of the HIST1H2AC gene.

The most striking aspect of this conserved sequence element is that it contains poly U stretches. In many contexts, this type of poly U stretch has been found to be a binding site for the HuR protein (86,87). While HuR binding sites are predominantly found in the

3’ UTR of HuR-regulated genes, there are also a significant number of mRNAs that contain HuR-binding sites in their 5’ UTR (88,89). In fact, genomic analyses of HuR target mRNAs detected binding of HuR to the 5’ UTR of the HIST1H2AC mRNA and, notably, HIST1H2AC was the only replication-dependent isoform of H2A associated with HuR (89). Intriguingly, 5’ UTR HuR binding has been shown to mediate translational repression (90-92). Therefore, we sought to determine whether HuR played a role in the regulation mediated by the HIST1H2AC 5’ UTR. However, knockdown of

HuR protein levels by siRNA treatment had no effect on the translational-repression mediated by the HIST1H2AC 5’ UTR indicating that the conserved sequence element identified in this regulatory region does not function solely through HuR (Figure 3.9-D).

Histone isoforms are differentially regulated. DNA methylation and histone modifications are the major epigenetic mechanisms that can affect gene expression in mammals (93,94). Also, in cancer cells, hypermethylation of DNA in the promoter CpG islands of tumor suppressor genes is thought to play a crucial role in carcinogenesis (95).

Based on our initial analysis that H2A 1C levels are governed by its 5’-UTR, we tried to probe if the isoforms also differ in terms of their post-translational modifications. To analyze the methylation levels for histone isoforms, we used COBRA. This technique is based on the fact that bisulfite treatment of genomic DNA converts unmethylated cytosine residues to uracil, whereas methylated cytosine residues remain unchanged.

Then, the sequence of interest is amplified by PCR, followed by a restriction digest with a restriction enzyme that contains a CpG dinucleotide in its restriction site. Because of bisulfite treatment-induced changes of the DNA sequence, the enzyme cleaves the PCR fragment only if it was originally methylated. Subsequent electrophoretic separation of

DNA fragments on acrylamide gels gives semiquantitative information on DNA methylation. Using this analysis, we were able to point out that histone isoforms do have different methylation status. Out of the five genes that encode the canonical H2A, only

Hist1H2AI is methylated and similarly, out of two genes encoding H2A 1B, only

Hist1H2AE is methylated (Figure 3.10). The sequence of other histone isoforms did not change after bisulfite treatment and were therefore left undigested (data not shown). This indicates that each of the histone isoforms is governed by its own set of regulations.

Furthermore, using MassARRAy methylation analysis, we were able to identify the set of isoforms that show an altered expression level in CLL (Figure 3.11). Notably, methylation levels for Hist1H2AI and Hist1H2AE are changed in both cell lines as well as the CLL patient samples, whereas other histone isoforms are not changed significantly.

3.5 Discussion

While the important role of histone variants in the regulation of chromatin structure has been firmly established, a functional role for the primary sequence diversity represented by the replication-dependent families of histone isoforms has not been addressed. Taken together, our results provide several lines of evidence that support the hypothesis that replication-dependent histone H2A isoforms are functionally distinct proteins. First, the 85 relative abundance of specific replication-dependent isoforms varied between normal tissue samples and tumor tissue samples at both the protein and mRNA levels. Second, artificial manipulation of replication-dependent H2A isoform levels by siRNA knockdown resulted in changes in the rates of cell proliferation and tumorigenicity in vitro. Third, replication-dependent H2A isoforms can be subject to individualized mechanisms of regulation. This is exemplified by the presence of specific elements in the 5’ UTR of HIST1H2AC mRNA that mediate translational repression.

The idea that the replication-dependent histone isoforms are functionally interchangeable stems largely from the high degree of sequence identity found between the isoforms.

H2A 1C and H2A 1B/E each differ from canonical H2A at two positions (Figure 3.12-A).

The threonine at position 16 is changed to serine in H2A 1C and the alanine at residue 40 is changed to serine in H2A 1B/E. In addition, the second change is shared by both H2A

1C and H2A 1B/E. This change is at position 99 where canonical H2A contains a lysine residue and H2A 1C and H2A 1B/E each contain an arginine residue. Interestingly, the presence of a lysine at residue 99 in canonical H2A is seen in many primate species, while in most mammals, including mice, the canonical H2A contains an arginine at this position (27). As depicted in figure 3.12-B, this residue is located in the center of the nucleosome face. While changing lysine to arginine is considered to be a conservative substitution in most proteins, the identities of these residues in the core histones are highly conserved throughout eukaryotic evolution (81). This may be due to the important role that these residues play in the regulation of chromatin structure as sites of post- translational modification. In fact, lysine 99 has recently been identified as a site of

86 methylation on histone H2A (96). In addition, the other changes in H2A 1C and H2A

1B/E either eliminate or introduce a potential site of phosphorylation. Hence, while the number of amino acid changes in H2A 1C and H2A 1B/E are small, we hypothesize that these differences have the potential to confer distinct functions to these isoforms.

The presence of multiple replication-dependent histone isoform genes in large clusters is likely to serve a number of purposes. Given the large amount of histone proteins need to assemble newly replicated DNA into chromatin, it is likely that the output of multiple histone genes is required to produce sufficient quantities of histone protein. In addition, the proximity of the replication-dependent histone genes facilitates their clustering near

Cajal bodies allowing for the coordinate regulation of this important gene family (84).

Given the high degree of sequence identity between the replication-dependent histone genes, techniques such as Northern blotting would only be able to look at the bulk levels of histone mRNAs and would not be able to discriminate between the expression of distinct genes. Our results suggest that expression of the different replication-dependent histone genes with the same cluster is not uniform. For example, transcript from the

HISTH2AE gene is far more abundant than the other H2A genes. In addition, the levels of some, but not all, replication-dependent H2A isoform genes can be significantly up- or down-regulated in tumor tissue samples. Whether these effects are the result of transcriptional or post-transcription of histone gene expression will be important to determine.

Histone mRNA 3’ end formation has clearly been shown to be the critical factor in the regulation of histone gene expression (84). However, our results suggest that the 5’ UTR 87 of replication-dependent histone genes can also play a regulatory role. While the 3’ stem-loop structure is ubiquitously present in replication-dependent histone genes, the regulatory elements identified in the 5’ UTR of the HIST1H2AC mRNA appears to be gene-specific. Like many RNA regulatory elements, the 5’ UTR of HIST1H2AC contains a high concentration of AU- and GU-rich sequences. While, the well- characterized ARE binding protein HuR does not appear to be involved in the

HISTH2AC 5’ UTR, there are many proteins that have been shown to interact with similar sequences that are now candidate regulators of this element. The identification of this regulatory element may help to identify novel regulators of histone gene expression.

Furthermore, using cell lines expressing tagged version of the histone isoforms, we tried to probe if the isoforms show a distinct localization pattern. The rationale is that the localization of the histone variants links with their biological function and therefore the activity of the locus may be used as a marker for the existence of variants and vice- versa. Intriguingly, many studies targeted at functional characterization of the variants have exploited this specific correlation. For example MacroH2A1 regulates transcription when located in the transcribed regions even though it is a marker for repressed autosomal chromatin (97). Our initial effort to target these isoforms, identified their nuclear localization, but our attempts to go for a metaphase staining proved futile as multiple combinations of antibodies as well as incubation conditions failed to highlight the specific location of these isoforms.

Last, but certainly not the least, our study consolidated the idea that histone cluster is not a single unit as it was thought to be. As described earlier, based on the clustering and co-

88 localization of the histone genes, the histone gene cluster was thought to be operated and regulated as single entity. Our study indicates that this is not exactly the case and there is lot of heterogeneity involved. Based on our COBRA analysis, we can infer, at least qualitatively, that H2A isoforms differ in their methylation levels and combined with the

5’-UTR study of Hist1H2AC, it can be concluded that the histone isoforms are differentially regulated. This is further consolidated by our MassARRAy methylation analysis. The latter also adds the possibility to explore the use of histone isoform methylation levels as a marker for the malignant state. Further efforts are needed to add a direction and broaden our outlook towards these isoforms, the unearthing of whom may also reveal some clues to the unsolved ambiguities of epigenetic regulation and human disease.

Figure 3.1- Replication-dependent histone H2A isoforms are altered in bladder cancer cells. A) The table lists the genes encoding replication-dependent histone H2A isoforms along with nomenclature used to name their respective proteins and the molecular weight of each isoform. The Hugo Gene Naming Consortium (HGNC) identification number is given in parentheses. B) LC/MS analysis of histone H2A isolated from the indicated cells. Normal bladder is normal human bladder epithelial cells; hTERT is immortalized human bladder epithelial cells; RT4 is non-invasive bladder cancer cells; and T24 is invasive bladder cancer cells.

A B

D C

Figure 3.2- Immunofluorescent staining using antibodies targeted against canonical H4 and H2A 1C. A) DAPI stained nuclei; B) Cy-3 stained H2A 1C nuclei; C) FITC targeted against canonical H4 and D) Merged image

Figure 3.3- Immunofluorescent staining for metaphase chromosomes using antibodies targeted against canonical H4 and H2A 1C. A) Cy-3 stained canonical H2A 1C chromosomes; B) Merged image showing foci of H2A 1C in red against a DAPI background and C) FITC conjugated antibody targeted against canonical H4

Figure 3.4- Distinct roles of replication-dependent histone H2A isoforms in cell proliferation and carcinogenesis. A) 293TN cells were transfected with siRNAs targeting the indicated histone H2A loci. qRT-PCR assays were used to measure each H2A mRNA before (blue bars) and after (red bars) transfection. B) 293TN cells were transfected with the indicated siRNAs (or mock transfected). Cell proliferation was measured by counting cell numbers at the indicated times. C) 293TN cells were transfected with the indicated siRNAs and plated in soft agar media. Plates were photographed following 5 days of growth. D) 293TN cells treated with the indicated siRNAs were plated in soft agar in microtiter wells. Following 6 days, cell growth was quantitated by solubilizing the cells in the presence of Cytoselect (Cytoselect 96 well cell transformation assay kit, Cell Biolabs, Inc.) and measuring relative fluorescence intensity.

Continued…..

Figure 3.4- Continued...

Figure 3.5- Histone H2A 1C influences cell proliferation and carcinogenesis in U2OS cells. A) U2OS cells were transfected with siRNAs targeting the indicated histone H2A loci. qRT pcr assays were used to measure each H2A mRNA before (blue bars) and after (red bars) transfection. B) Cells were transfected with the indicated siRNAs (or mock transfected). U2OS cell proliferation was measured by counting cell numbers at the indicated times. C) U2OS cells were transfected with the indicated siRNAs and plated in soft agar media. Plates were photographed following 5 days of growth. D) U2OS cells treated with the indicated siRNAs were plated in soft agar in mitrotiter wells. Following 6 days, cell growth was quantitated by solubilizing the cells in the presence of Cytoselect (Cytoselect 96 well cell transformation assay kit, Cell Biolabs, Inc.) and measuring relative fluorescence intensity.

Continued…..

Figure 3.5- Continued…

Figure 3.6- Replication-dependent histone H2A gene expression in healthy individuals and CLL patients. Real time PCR assays were used to measure the abundance of the indicated histone H2A genes in either healthy individuals or CLL patients (N=129). The relative expression of each isoform was determined by normalizing the PCR reactions to a control reaction that contained an equal amount of each H2A target cloned into a vector (50 fg each). H2A isoforms showing a significant change in abundance in CLL patients (p<0.01) are indicated by an asterisk.

Figure 3.7- The 5’ UTR of HIST1H2AC is a regulatory element. A) The 5’UTR of each of the indicated H2A isoforms was cloned upstream of a luciferase reporter construct. Each construct was transfected into the indicated cell lines (with a renilla reporter for normalization) and luciferase levels determined by fluorescence. B) Luciferase levels were determined as in A) using constructs that contained either the wild type HIST1H2AC 5’ UTR (1C) or a scrambled version that contains the same base composition but in a random order (scramble).

Figure 3.8- Identification of a repeated element in the H2A 1C 5’ UTR necessary for repressive activity. A) Sequence of the H2A 1C 5’ UTR separated into 10 base segments. B) Luciferase constructs containing the indicated H2A 1C 5’ UTR sequences were transfected into U2OS cells and assayed for fluorescence intensity. Control contains no insert while the H2A 1C contains the entire H2A 1C 5’ UTR. The scrambled construct contains a randomized version of the H2A 1C sequence. The remaining constructs contain a deletion of the indicated bases from the H2A 1C 5’ UTR. The fluorescence signal of the control construct was set to 1 for each cell line. C) Highly similar sequence elements identified in the indicated regions of the H2A 1C 5’ UTR. D) Sequences of the H2A 1C 5’ UTR containing either the MUT1 or MUT2 mutation (top, altered sequences shown in outlined text). Lucifierase constructs containing the indicated H2A 1C 5’ UTR sequences were transfected into U2OS cells and assayed for fluorescence intensity (bottom left). Aliquots of cells were also treated with an siRNA directed against HuR 48 prior to luciferase assay. Efficiency of HuR knockdown was assayed by Western blot analysis of whole cell extracts using the indicated antibodies (bottom right).

Continued…..

Figure 3.8- Continued…

100

WT Un Dn-/- Un 3b-/-Un dKOUn WT Un Dn-/- Un 3b-/-Un dKOUn

Hist1H2AE Hist1H2AI

Figure 3.9- Stained agarose gel showing a change in restriction digestion pattern after bisulfite treatment. Left panel shows Hist1H2AE and right panel shows Hist2H2AI after digestion with AciI. Abbreviations used include- WT: Wild type, Dn-/-: Dnmt1 knockout, 3b-/-:Dnmt 3b knockout, dKO: double knockout and Un: Undigested control

101

HIST1H2AC HIST1H2AE-1 0.10 0.6

0.08 0.4 A 0.06 B

0.04 0.2 methylation DNA 0.02 methylation DNA 0.00 0.0

PBL CLL PBL CLL CD19+ B CD19+ B

HIST1H2AG-1 HIST1H2AI-1 HIST1H2AK-3 0.25 0.8 0.20 0.20 0.6 D 0.15 C 0.15 E 0.4 0.10 0.10 0.2 0.05 methylation DNA 0.05 methylation DNA methylation DNA 0.00 0.0 0.00

PBL CLL PBL CLL PBL CLL CD19+ B CD19+ B CD19+ B

HIST1H2AL-1 HIST1H2AM-1 0.4 0.3 0.3 F 0.2 G 0.2

0.1 0.1 DNA methylation DNA DNA methylation DNA

0.0 0.0

PBL CLL PBL CLL CD19+ B CD19+ B

Figure 3.10- Amplicon mean values for DNA methylation for normal B-cells, PBLs and CLL patient samples. A) Hist1H2AC, B) Hist1H2AE, C) Hist1H2AG, D)- Hist1H2AI, E) Hist1H2AK, F) Hist1H2AL and G) Hist1H2AM

102

Figure 3.11- H2A 1C and H2A 1B/E share a common amino acid change. A) Protein sequence alignment of H2A, H2A 1C and H2A 1B/E. Positions of identity are marked with an *. Positions of divergence are highlighted in red. B) The crystal structure of the nucleosome with residue 99 of histone H2A shown in yellow as a space-filling model.

103

CHAPTER 4

Conclusions & Future Directions

4.1 Conclusions

This dissertation had two major foci: To explore if any of the histone isoforms carry any clinical significance and/or correlation with disease parameters and secondly, to probe if the replication-dependent histone H2A isoforms are functionally distinct.

In chapter 2, we followed a comprehensive proteomics approach to analyze the histone levels in cancer cell lines and CLL patient samples. The protocol involved isolation of histones from various primary cells and cell lines and using mass spectrometry to probe the specific protein levels. Rationale of using cell lines comes from the fact that cell lines don't contain the non-cancerous cells found in primary tumors, thus making the cultivated lines ideal for finding specific changes in the cancer genome or proteome. Not surprisingly, many researchers have found changes in cell lines as reproducible in primary cells, especially for short-term goals (98). Moreover, there is a limit to primary cells and patient samples that you can get. Therefore, many recent researches have employed and validated the use of cell lines to map the spectrum of epigenetic changes in cancer and other diseases (36,98). Though the importance of specific histone isoforms hasn’t been recognized so far, but various global ‘omics’ studies and microarrays have identified many isoforms as being upregulated or downregulated in the malignant state.

104

Based on these studies and using our own model systems, we can infer that histone isoforms levels are changed in response to cancerous state in both primary tissue as well as cell lines. In addition, many of the isoforms also show a linear correlation to cancer stage. We can conclude here that histone isoforms do have the potential to serve as prognostic candidates and could be explored further. However, our findings support a need to examine the defined functional role that these isoforms play in cancer and additionally, their expression levels should be corroborated in multiple cancer types and models.

In chapter 3, we have examined the functional role of specific H2A isoforms in tumorigenesis. We started by analyzing the mRNA expression of histone H2A isoforms in CLL patients using real time PCR and correlating it with the corresponding protein levels. Followed by this, we extended our analysis to the post-transcriptional levels to probe how the histone isoform levels are regulated in the cell, relative to their canonical counterparts. Additionally, we did knockdown studies to reveal their functional role in the cell and cell transformation studies to probe if the change in abundance can be induced synthetically or is it a natural response to the malignant state. At the mRNA level, we noticed some correlation of the protein level with mRNA of some of the isoforms in CLL patient samples, but overall it didn’t carry much clinical significance.

Intriguingly, mRNA from one of the isoforms showed an inverse correlation with the protein levels, suggesting a post-transcriptional regulation. After further analysis, we found that this particular isoform was regulated by its 5’-untranslated region, which is unique to the replication dependent histones. With the knockdown experiments, we

105 observed that some of the isoforms show a differential phenotype, particularly H2A 1C, whose knockdown caused an increased proliferation in the cells. Despite the fact that it differs just by two amino acids from the canonical H2A, it displayed a role that is different from packaging the DNA, an observation that is again distinctive to the replication-dependent histones.

Additionally, we did not observe any significant cell cycle change after knockdown of the isoforms (data not shown), but we were able to identify that cell proliferation has a direct link with the H2A 1C levels. The precise nature of this association, however, needs to be investigated, as the levels of this isoform have been known to decrease as well as increase in different cancers. Nevertheless, the isoform do carries the potential to be used as a biomarker and could be explored further. Lastly, our study identified a difference in methylation levels for some histone isoforms, even when they are coding for the same protein. This uncovers another level of complexity in epigenetic regulation, which may address some unanswered links between epigenetics and disease.

The inference from our study is that there is variability in the level of histone isoforms in the cell and they are regulated differentially to the core histones. In a broader sense, the histone gene cluster does not operate as a single unit as it was previously known, which may open up an otherwise overlooked aspect of the epigenetic regulation. The heterogeneity thus uncovered may also highlight some missing components in the structure and operation of epigenetic machinery. Overall, our project plays a role in enhancing the understanding of factors affecting chromatin structure and regulation and gives a better understanding of epigenetic regulation in biological systems. We used a

106 combination of molecular biology and proteomics to interrogate the characteristics of histone isoforms associated with cancer. More advances in our systems and future work involving characterization of the precise biological role of histone isoforms will enable a greater understanding of the mechanism of epigenetic regulation. This may also enhance our ability to develop novel interventions, which will help to design novel efficacies against the cancer and other diseases.

107

4.2 Future Directions

One of the questions that remain unanswered from our study is that whether cancer is causing the histone isoforms to change in abundance or if change in abundance of histone isoforms is dictating cancer progression. An apparent way to address this issue is to specifically knockdown each of the isoforms and study the phenotypic effects. However, most of our work involving the characterization of functional role of H2A isoforms was carried out using an siRNA approach, which carries its own set of disadvantages. In addition to being expensive, one of the major hurdles of using synthetic siRNA is the limited duration of post-transfection effects, which restricts the number of assays that can be carried out. Another drawback with siRNA mediated approach is that many cell types, including primary cells are difficult to transfect. This may specifically create problems when the targeted protein is thought to play a role in cell proliferation and/or tumor- suppression.

Given the functional role that these isoforms could play in cancer, an appropriate approach will be to validate the shRNAs for the stable knockdown of the specific histone isoform. Once the constructs are made, they can be reproduced inexpensively and introduced into a wide variety of cells. This will enable a multitude of assays to be carried out including effects on nucleosome stability, cell cycle and translational output.

This will also facilitate to carry out microarray studies to check the global effect of depletion of the histone isoforms. The profile thus generated will help to pinpoint the site and route by which these isoforms integrate into the cancer pathway. However, to rule

108 out any discrepancy due to passage, handling or other biological variability, the results should be confirmed in multiple cell types and cell lines.

Another subject of investigation that could enhance our understanding of these isoforms is to determine the precise location and how they are deposited in the nucleosome. Due to a high similarity in their primary structure with the canonical histones, it is difficult to generate a primary antibody that is specific to these isoforms. An initial step will be to generate an an epitope-tagged version of each of these isoforms and to generate stable cell lines expressing this construct. Our initial approaches to determine the location of these isoforms using an epitope-tagged approach were unsuccessful, partly due to the difficulty in maintenance of this construct in long-term. Additionally, the accumulation of excess histone proteins in cells leads to genomic instability caused by inappropriate electrostatic interactions between the histones and diverse negatively charged molecules in the cell (99). A better approach will involve the use of conditional gene expression systems like tet-inducible cell lines so that these potential risks can be minimized.

After the appropriate cell lines are generated, they can be used with a combination of various immunofluorescence-based techniques to pinpoint the location. These tagged cell lines may also be used for immunopreciptation of specific histone isoforms to identify the proteins/chaperones that are specifically interacting with the isoforms relative to the canonical histones. This information will also allow us to determine if these isoforms are deposited into the nucleosome via a replication dependent or independent pathway.

However, the epitope tags display a fluctuation in their response towards various fluorescent antibodies and immunoprecipitation protocols. Therefore, different tags can

109 be tested or alternatively, constructs can be designed with multiple tags at N- or C- terminal sites including tags to enable affinity purification. This will facilitate the immunofluorescence as well as immunoprecipitation studies to be carried out using the same construct, thus adding precision and specificity to the study. Our initial analysis during this approach was unsuccessful, but with more technological advancements, including the use of confocal microscope may be useful.

Given our preliminary study has identified a distinct role of these isoforms, it may not be surprising if each of these isoforms has a defined pathway or chaperone assigned specifically for its deposition into the chromatin. The key issue is to understand where, when and by what context are these H2A isoforms deposited into the nucleus, what other proteins are involved in the assembly process and what implications does this cascade of events have for structural integrity and chromatin dynamics. The results obtained hereby will uncover many aspects of the deposition of histone isoforms in the cell, for which no comprehensive information is available till date. Moreover, it may lead to discovery of novel chaperones/assembly factors that may be dedicated specifically in the assembly of these histone isoforms. In addition, there is no report regarding the assembly process, therefore any data generated will be novel in itself and may provide a foundation for the otherwise unclear world of chromatin assembly and dynamics.

Our study is novel and warrants further investigation into the mechanistic details. This will be accomplished by implementing a combinatorial approach which includes bringing together expertise from various clinical, chemical and biological domains and technical advances including microarrays as well as computational approaches and databases to

110 integrate the data generated during the course of the study. The outcomes will not only provide the clinicians with novel insights into the cancer biology, but will also bring to light the mechanisms for maintenance and transmission of epigenetic integrity and the disorders during which such integrity is compromised.

111

Bibliography

1. Olby, R. (1978) The origins of modern biochemistry. Nature, 274, 310-312. 2. Portela, A. and Esteller, M. (2010) Epigenetic modifications and human disease. Nature biotechnology, 28, 1057-1068. 3. Groth, A., Corpet, A., Cook, A.J., Roche, D., Bartek, J., Lukas, J. and Almouzni, G. (2007) Regulation of replication fork progression through histone supply and demand. Science, 318, 1928-1931. 4. Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F. and Richmond, T.J. (1997) Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature, 389, 251-260. 5. Luger, K. and Richmond, T.J. (1998) The histone tails of the nucleosome. Current opinion in genetics & development, 8, 140-146. 6. Kouzarides, T. (2007) Chromatin modifications and their function. Cell, 128, 693- 705. 7. Tremethick, D. (2006) Chromatin: the dynamic link between structure and function. Chromosome research : an international journal on the molecular, supramolecular and evolutionary aspects of chromosome biology, 14, 1-4. 8. Tremethick, D.J. (2007) Higher-order structures of chromatin: the elusive 30 nm fiber. Cell, 128, 651-654. 9. Li, B., Carey, M. and Workman, J.L. (2007) The role of chromatin during transcription. Cell, 128, 707-719. 10. Li, G. and Reinberg, D. (2011) Chromatin higher-order structures and gene regulation. Current opinion in genetics & development, 21, 175-186. 11. Wang, G.G., Allis, C.D. and Chi, P. (2007) Chromatin remodeling and cancer, Part I: Covalent histone modifications. Trends in molecular medicine, 13, 363- 372. 12. Almouzni, G., Khochbin, S., Dimitrov, S. and Wolffe, A.P. (1994) Histone acetylation influences both gene expression and development of Xenopus laevis. Developmental biology, 165, 654-669. 13. Martens, J.A. and Winston, F. (2003) Recent advances in understanding chromatin remodeling by Swi/Snf complexes. Current opinion in genetics & development, 13, 136-142. 14. Hassan, A.H., Neely, K.E. and Workman, J.L. (2001) Histone acetyltransferase complexes stabilize swi/snf binding to promoter nucleosomes. Cell, 104, 817-827. 15. Vignali, M., Hassan, A.H., Neely, K.E. and Workman, J.L. (2000) ATP- dependent chromatin-remodeling complexes. Molecular and cellular biology, 20, 1899-1910.

112

16. Greer, E.L. and Shi, Y. (2012) Histone methylation: a dynamic mark in health, disease and inheritance. Nature reviews. Genetics, 13, 343-357. 17. Skene, P.J. and Henikoff, S. (2013) Histone variants in pluripotency and disease. Development, 140, 2513-2524. 18. Van Beekum, O. and Kalkhoven, E. (2007) Aberrant forms of histone acetyltransferases in human disease. Sub-cellular biochemistry, 41, 233-262. 19. Henikoff, S., Furuyama, T. and Ahmad, K. (2004) Histone variants, nucleosome assembly and epigenetic inheritance. Trends Genet, 20, 320-326. 20. Kamakaka, R.T. and Biggins, S. (2005) Histone variants: deviants? Genes & development, 19, 295-310. 21. Suto, R.K., Clarkson, M.J., Tremethick, D.J. and Luger, K. (2000) Crystal structure of a nucleosome core particle containing the variant histone H2A.Z. Nature structural biology, 7, 1121-1124. 22. Brown, D.T. (2001) Histone variants: are they functionally heterogeneous? Genome biology, 2, REVIEWS0006. 23. Ahmad, K. and Henikoff, S. (2002) Histone H3 variants specify modes of chromatin assembly. Proceedings of the National Academy of Sciences of the United States of America, 99 Suppl 4, 16477-16484. 24. Sarma, K. and Reinberg, D. (2005) Histone variants meet their match. Nature reviews. Molecular cell biology, 6, 139-149. 25. Sporn, J.C., Kustatscher, G., Hothorn, T., Collado, M., Serrano, M., Muley, T., Schnabel, P. and Ladurner, A.G. (2009) Histone macroH2A isoforms predict the risk of lung cancer recurrence. Oncogene, 28, 3423-3428. 26. Talbert, P.B. and Henikoff, S. (2010) Histone variants--ancient wrap artists of the epigenome. Nature reviews. Molecular cell biology, 11, 264-275. 27. Marzluff, W.F., Gongidi, P., Woods, K.R., Jin, J. and Maltais, L.J. (2002) The human and mouse replication-dependent histone genes. Genomics, 80, 487-498. 28. Hake, S.B. and Allis, C.D. (2006) Histone H3 variants and their potential role in indexing mammalian genomes: the "H3 barcode hypothesis". Proceedings of the National Academy of Sciences of the United States of America, 103, 6428-6435. 29. Marks, P., Rifkind, R.A., Richon, V.M., Breslow, R., Miller, T. and Kelly, W.K. (2001) Histone deacetylases and cancer: causes and therapies. Nature reviews. Cancer, 1, 194-202. 30. Wang, G.G., Allis, C.D. and Chi, P. (2007) Chromatin remodeling and cancer, Part II: ATP-dependent chromatin remodeling. Trends in molecular medicine, 13, 373-380. 31. Yang, X., Ferguson, A.T., Nass, S.J., Phillips, D.L., Butash, K.A., Wang, S.M., Herman, J.G. and Davidson, N.E. (2000) Transcriptional activation of estrogen receptor alpha in human breast cancer cells by histone deacetylase inhibition. Cancer research, 60, 6890-6894. 32. Chung, D. (2002) Histone modification: the 'next wave' in cancer therapeutics. Trends in molecular medicine, 8, S10-11. 33. Archer, S.Y. and Hodin, R.A. (1999) Histone acetylation and cancer. Current opinion in genetics & development, 9, 171-174. 113

34. Gonzalo, V., Castellvi-Bel, S., Balaguer, F., Pellise, M., Ocana, T. and Castells, A. (2008) [Epigenetics of cancer]. Gastroenterologia y hepatologia, 31, 37-45. 35. Kristeleit, R., Stimson, L., Workman, P. and Aherne, W. (2004) Histone modification enzymes: novel targets for cancer drugs. Expert opinion on emerging drugs, 9, 135-154. 36. Seligson, D.B., Horvath, S., McBrian, M.A., Mah, V., Yu, H., Tze, S., Wang, Q., Chia, D., Goodglick, L. and Kurdistani, S.K. (2009) Global levels of histone modifications predict prognosis in different cancers. The American journal of pathology, 174, 1619-1628. 37. Seligson, D.B., Horvath, S., Shi, T., Yu, H., Tze, S., Grunstein, M. and Kurdistani, S.K. (2005) Global histone modification patterns predict risk of prostate cancer recurrence. Nature, 435, 1262-1266. 38. Fraga, M.F., Ballestar, E., Villar-Garea, A., Boix-Chornet, M., Espada, J., Schotta, G., Bonaldi, T., Haydon, C., Ropero, S., Petrie, K. et al. (2005) Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer. Nature genetics, 37, 391-400. 39. Kavak, E., Unlu, M., Nister, M. and Koman, A. (2010) Meta-analysis of cancer gene expression signatures reveals new cancer genes, SAGE tags and tumor associated regions of co-regulation. Nucleic acids research, 38, 7008-7021. 40. Parssinen, J., Alarmo, E.L., Khan, S., Karhu, R., Vihinen, M. and Kallioniemi, A. (2008) Identification of differentially expressed genes after PPM1D silencing in breast cancer. Cancer letters, 259, 61-70. 41. Su, X., Lucas, D.M., Zhang, L., Xu, H., Zabrouskov, V., Davis, M.E., Knapp, A.R., Young, D.C., Payne, P.R., Parthun, M.R. et al. (2009) Validation of an LC- MS based approach for profiling histones in chronic lymphocytic leukemia. Proteomics, 9, 1197-1206. 42. Kupferman, M.E., Patel, V., Sriuranpong, V., Amornphimoltham, P., Jasser, S.A., Mandal, M., Zhou, G., Wang, J., Coombes, K., Multani, A. et al. (2007) Molecular analysis of anoikis resistance in oral cavity squamous cell carcinoma. Oral oncology, 43, 440-454. 43. Olivares, I., Ballester, A., Lombardia, L., Dominguez, O. and Lopez-Galindez, C. (2009) Human immunodeficiency virus type 1 chronic infection is associated with different gene expression in MT-4, H9 and U937 cell lines. Virus research, 139, 22-31. 44. Singh, M.K., Scott, T.F., LaFramboise, W.A., Hu, F.Z., Post, J.C. and Ehrlich, G.D. (2007) Gene expression changes in peripheral blood mononuclear cells from multiple sclerosis patients undergoing beta-interferon therapy. Journal of the neurological sciences, 258, 52-59. 45. Pirngruber, J. and Johnsen, S.A. (2010) Induced G1 cell-cycle arrest controls replication-dependent histone mRNA 3' end processing through p21, NPAT and CDK9. Oncogene, 29, 2853-2863. 46. Pirngruber, J., Shchebet, A., Schreiber, L., Shema, E., Minsky, N., Chapman, R.D., Eick, D., Aylon, Y., Oren, M. and Johnsen, S.A. (2009) CDK9 directs H2B

114

monoubiquitination and controls replication-dependent histone mRNA 3'-end processing. EMBO reports, 10, 894-900. 47. Buterin, T., Koch, C. and Naegeli, H. (2006) Convergent transcriptional profiles induced by endogenous estrogen and distinct xenoestrogens in breast cancer cells. Carcinogenesis, 27, 1567-1578. 48. van Erk, M.J., Roepman, P., van der Lende, T.R., Stierum, R.H., Aarts, J.M., van Bladeren, P.J. and van Ommen, B. (2005) Integrated assessment by multiple gene expression analysis of quercetin bioactivity on anticancer-related mechanisms in colon cancer cells in vitro. European journal of nutrition, 44, 143-156. 49. Ellinger-Ziegelbauer, H., Fostel, J.M., Aruga, C., Bauer, D., Boitier, E., Deng, S., Dickinson, D., Le Fevre, A.C., Fornace, A.J., Jr., Grenet, O. et al. (2009) Characterization and interlaboratory comparison of a gene expression signature for differentiating genotoxic mechanisms. Toxicological sciences : an official journal of the Society of Toxicology, 110, 341-352. 50. Britton, L.M., Gonzales-Cope, M., Zee, B.M. and Garcia, B.A. (2011) Breaking the histone code with quantitative mass spectrometry. Expert review of proteomics, 8, 631-643. 51. Britton, L.M., Newhart, A., Bhanu, N.V., Sridharan, R., Gonzales-Cope, M., Plath, K., Janicki, S.M. and Garcia, B.A. (2013) Initial characterization of histone H3 serine 10 O-acetylation. Epigenetics : official journal of the DNA Methylation Society, 8. 52. Boyne, M.T., 2nd, Pesavento, J.J., Mizzen, C.A. and Kelleher, N.L. (2006) Precise characterization of human histones in the H2A gene family by top down mass spectrometry. J Proteome Res, 5, 248-253. 53. Siuti, N. and Kelleher, N.L. (2007) Decoding protein modifications using top- down mass spectrometry. Nature methods, 4, 817-821. 54. Thomas, S.N., Wan, Y., Liao, Z., Hanson, P.I. and Yang, A.J. (2011) Stable isotope labeling with amino acids in cell culture based mass spectrometry approach to detect transient protein interactions using substrate trapping. Analytical chemistry, 83, 5511-5518. 55. Cosgrove, M.S. (2007) Histone proteomics and the epigenetic regulation of nucleosome mobility. Expert review of proteomics, 4, 465-478. 56. Davie, J.K., Edmondson, D.G., Coco, C.B. and Dent, S.Y. (2003) Tup1-Ssn6 interacts with multiple class I histone deacetylases in vivo. The Journal of biological chemistry, 278, 50158-50162. 57. Galasinski, S.C., Louie, D.F., Gloor, K.K., Resing, K.A. and Ahn, N.G. (2002) Global regulation of post-translational modifications on core histones. The Journal of biological chemistry, 277, 2579-2588. 58. Horn, D.M., Zubarev, R.A. and McLafferty, F.W. (2000) Automated reduction and interpretation of high resolution electrospray mass spectra of large molecules. Journal of the American Society for Mass Spectrometry, 11, 320-332. 59. Lo, W.S., Trievel, R.C., Rojas, J.R., Duggan, L., Hsu, J.Y., Allis, C.D., Marmorstein, R. and Berger, S.L. (2000) Phosphorylation of serine 10 in histone

115

H3 is functionally linked in vitro and in vivo to Gcn5-mediated acetylation at lysine 14. Molecular cell, 5, 917-926. 60. Damle, R.N., Wasil, T., Fais, F., Ghiotto, F., Valetto, A., Allen, S.L., Buchbinder, A., Budman, D., Dittmar, K., Kolitz, J. et al. (1999) Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood, 94, 1840-1847. 61. Oscier, D.G., Gardiner, A.C., Mould, S.J., Glide, S., Davis, Z.A., Ibbotson, R.E., Corcoran, M.M., Chapman, R.M., Thomas, P.W., Copplestone, J.A. et al. (2002) Multivariate analysis of prognostic factors in CLL: clinical stage, IGVH gene mutational status, and loss or mutation of the p53 gene are independent prognostic factors. Blood, 100, 1177-1184. 62. Wiestner, A., Rosenwald, A., Barry, T.S., Wright, G., Davis, R.E., Henrickson, S.E., Zhao, H., Ibbotson, R.E., Orchard, J.A., Davis, Z. et al. (2003) ZAP-70 expression identifies a chronic lymphocytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome, and distinct gene expression profile. Blood, 101, 4944-4951. 63. Schneider, A.C., Heukamp, L.C., Rogenhofer, S., Fechner, G., Bastian, P.J., von Ruecker, A., Muller, S.C. and Ellinger, J. (2011) Global histone H4K20 trimethylation predicts cancer-specific survival in patients with muscle-invasive bladder cancer. BJU international, 108, E290-296. 64. Mosashvilli, D., Kahl, P., Mertens, C., Holzapfel, S., Rogenhofer, S., Hauser, S., Buttner, R., Von Ruecker, A., Muller, S.C. and Ellinger, J. (2010) Global histone acetylation levels: prognostic relevance in patients with renal cell carcinoma. Cancer science, 101, 2664-2669. 65. Ellinger, J., Kahl, P., von der Gathen, J., Rogenhofer, S., Heukamp, L.C., Gutgemann, I., Walter, B., Hofstadter, F., Buttner, R., Muller, S.C. et al. (2010) Global levels of histone modifications predict prostate cancer recurrence. The Prostate, 70, 61-69. 66. Elsheikh, S.E., Green, A.R., Rakha, E.A., Powe, D.G., Ahmed, R.A., Collins, H.M., Soria, D., Garibaldi, J.M., Paish, C.E., Ammar, A.A. et al. (2009) Global histone modifications in breast cancer correlate with tumor phenotypes, prognostic factors, and patient outcome. Cancer research, 69, 3802-3809. 67. Barlesi, F., Giaccone, G., Gallegos-Ruiz, M.I., Loundou, A., Span, S.W., Lefesvre, P., Kruyt, F.A. and Rodriguez, J.A. (2007) Global histone modifications predict prognosis of resected non small-cell lung cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology, 25, 4358-4364. 68. Waldmann, T. and Schneider, R. (2013) Targeting histone modifications-- epigenetics in cancer. Current opinion in cell biology, 25, 184-189. 69. Albig, W., Trappe, R., Kardalinou, E., Eick, S. and Doenecke, D. (1999) The human H2A and H2B histone gene complement. Biological chemistry, 380, 7-18. 70. Talbert, P.B., Ahmad, K., Almouzni, G., Ausio, J., Berger, F., Bhalla, P.L., Bonner, W.M., Cande, W.Z., Chadwick, B.P., Chan, S.W. et al. (2012) A unified

116

phylogeny-based nomenclature for histone variants. Epigenetics & chromatin, 5, 7. 71. Wang, L., Harshman, S.W., Liu, S., Ren, C., Xu, H., Sallans, L., Grever, M., Byrd, J.C., Marcucci, G. and Freitas, M.A. (2010) Assaying pharmacodynamic endpoints with targeted therapy: flavopiridol and 17AAG induced dephosphorylation of histone H1.5 in acute myeloid leukemia. Proteomics, 10, 4281-4292. 72. Singh, R., Mortazavi, A., Telu, K.H., Nagarajan, P., Lucas, D.M., Thomas-Ahner, J.M., Clinton, S.K., Byrd, J.C., Freitas, M.A. and Parthun, M.R. (2013) Increasing the complexity of chromatin: functionally distinct roles for replication-dependent histone H2A isoforms in cell proliferation and carcinogenesis. Nucleic acids research. 73. Tang, B., Vu, M., Booker, T., Santner, S.J., Miller, F.R., Anver, M.R. and Wakefield, L.M. (2003) TGF-beta switches from tumor suppressor to prometastatic factor in a model of breast cancer progression. The Journal of clinical investigation, 112, 1116-1124. 74. Witt, O., Albig, W. and Doenecke, D. (1996) Testis-specific expression of a novel human H3 histone gene. Experimental cell research, 229, 301-306. 75. Andersen, J.S., Lam, Y.W., Leung, A.K., Ong, S.E., Lyon, C.E., Lamond, A.I. and Mann, M. (2005) Nucleolar proteome dynamics. Nature, 433, 77-83. 76. Ellinger, J., Kahl, P., Mertens, C., Rogenhofer, S., Hauser, S., Hartmann, W., Bastian, P.J., Buttner, R., Muller, S.C. and von Ruecker, A. (2010) Prognostic relevance of global histone H3 lysine 4 (H3K4) methylation in renal cell carcinoma. International journal of cancer. Journal international du cancer, 127, 2360-2366. 77. Rogenhofer, S., Kahl, P., Mertens, C., Hauser, S., Hartmann, W., Buttner, R., Muller, S.C., von Ruecker, A. and Ellinger, J. Global histone H3 lysine 27 (H3K27) methylation levels and their prognostic relevance in renal cell carcinoma. BJU Int, 109, 459-465. 78. Behbahani, T.E., Kahl, P., von der Gathen, J., Heukamp, L.C., Baumann, C., Gutgemann, I., Walter, B., Hofstadter, F., Bastian, P.J., von Ruecker, A. et al. Alterations of global histone H4K20 methylation during prostate carcinogenesis. BMC Urol, 12, 5. 79. Rogenhofer, S., Kahl, P., Holzapfel, S., A, V.O.N.R., Mueller, S.C. and Ellinger, J. Decreased levels of histone H3K9me1 indicate poor prognosis in patients with renal cell carcinoma. Anticancer Res, 32, 879-886. 80. Ellinger, J., Kahl, P., von der Gathen, J., Heukamp, L.C., Gutgemann, I., Walter, B., Hofstadter, F., Bastian, P.J., von Ruecker, A., Muller, S.C. et al. (2012) Global histone H3K27 methylation levels are different in localized and metastatic prostate cancer. Cancer investigation, 30, 92-97. 81. Malik, H.S. and Henikoff, S. (2003) Phylogenomics of the nucleosome. Nat Struct Biol, 10, 882-891. 82. Talbert, P.B. and Henikoff, S. Histone variants--ancient wrap artists of the epigenome. Nat Rev Mol Cell Biol, 11, 264-275. 117

83. Wu, R.S., and Bonner, W.M. (1981) Separation of basal histone synthesis from S- phase hstone synthesis in dividing cells. Cell, 27, 321-330. 84. Marzluff, W.F., Wagner, E.J. and Duronio, R.J. (2008) Metabolism and regulation of canonical histone mRNAs: life without a poly(A) tail. Nature reviews. Genetics, 9, 843-854. 85. Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 25, 402-408. 86. Ma, W.J., Cheng, S., Campbell, C., Wright, A. and Furneaux, H. (1996) Cloning and characterization of HuR, a ubiquitously expressed Elav-like protein. The Journal of biological chemistry, 271, 8144-8151. 87. Ma, W.J., Chung, S. and Furneaux, H. (1997) The Elav-like proteins bind to AU- rich elements and to the poly(A) tail of mRNA. Nucleic acids research, 25, 3564- 3569. 88. Lebedeva, S., Jens, M., Theil, K., Schwanhausser, B., Selbach, M., Landthaler, M. and Rajewsky, N. (2011) Transcriptome-wide analysis of regulatory interactions of the RNA-binding protein HuR. Molecular cell, 43, 340-352. 89. Uren, P.J., Burns, S.C., Ruan, J., Singh, K.K., Smith, A.D. and Penalva, L.O. (2011) Genomic analyses of the RNA-binding protein Hu antigen R (HuR) identify a complex network of target genes and novel characteristics of its binding sites. The Journal of biological chemistry, 286, 37063-37066. 90. Kullmann, M., Gopfert, U., Siewe, B. and Hengst, L. (2002) ELAV/Hu proteins inhibit p27 translation via an IRES element in the p27 5'UTR. Genes & development, 16, 3087-3099. 91. Meng, Z., King, P.H., Nabors, L.B., Jackson, N.L., Chen, C.Y., Emanuel, P.D. and Blume, S.W. (2005) The ELAV RNA-stability factor HuR binds the 5'- untranslated region of the human IGF-IR transcript and differentially represses cap-dependent and IRES-mediated translation. Nucleic acids research, 33, 2962- 2979. 92. Yeh, C.H., Hung, L.Y., Hsu, C., Le, S.Y., Lee, P.T., Liao, W.L., Lin, Y.T., Chang, W.C. and Tseng, J.T. (2008) RNA-binding protein HuR interacts with thrombomodulin 5'untranslated region and represses internal ribosome entry site- mediated translation under IL-1 beta treatment. Molecular biology of the cell, 19, 3812-3822. 93. Kondo, Y., Shen, L., Cheng, A.S., Ahmed, S., Boumber, Y., Charo, C., Yamochi, T., Urano, T., Furukawa, K., Kwabi-Addo, B. et al. (2008) Gene silencing in cancer by histone H3 lysine 27 trimethylation independent of promoter DNA methylation. Nature genetics, 40, 741-750. 94. Kondo, Y., Shen, L. and Issa, J.P. (2003) Critical role of histone methylation in tumor suppressor gene silencing in colorectal cancer. Molecular and cellular biology, 23, 206-215. 95. Bayliss, J., Hilger, A., Vishnu, P., Diehl, K. and El-Ashry, D. (2007) Reversal of the estrogen receptor negative phenotype in breast cancer and restoration of

118

antiestrogen response. Clinical cancer research : an official journal of the American Association for Cancer Research, 13, 7029-7036. 96. Zhang, K.a.T., H. (2003) Analysis of core histones by liquid chromatography- mass spectrometry and peptide mapping. J. Chromatogr. B, 783, 173-179. 97. Gamble, M.J., Frizzell, K.M., Yang, C., Krishnakumar, R. and Kraus, W.L. (2010) The histone variant macroH2A1 marks repressed autosomal chromatin, but protects a subset of its target genes from silencing. Genes & development, 24, 21- 32. 98. Borrell, B. (2010) How accurate are cancer cell lines? Nature, 463, 858. 99. Singh, R.K., Liang, D., Gajjalaiahvari, U.R., Kabbaj, M.H., Paik, J. and Gunjan, A. (2010) Excess histone levels mediate cytotoxicity via multiple mechanisms. Cell cycle, 9, 4236-4244.

119