Human Papillomavirus Oncoprotein E7 Dysregulates Immune

HUMAN PAPILLOMAVIRUS ONCOPROTEIN E7 DYSREGULATES IMMUNE

RESPONSES THROUGH EPIGENETIC MANIPULATION

LOUIS J CICCHINI

B.S., University of Arizona, 2009

A thesis submitted to the

Faculty of the Graduate School of the

University of Colorado in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

Molecular Biology

2016

This thesis for the Doctor of Philosophy degree by

Louis J Cicchini

has been approved for the

Molecular Biology Program

Rytis Prekeris, Chair

Dohun Pyeon, Advisor

James Hagman

Thomas E. Morrison

Xiao-Jing Wang

Date: August 18, 2016

Cicchini, Louis J. (PhD, Molecular Biology)

Human Papillomavirus Oncoprotein E7 Dysregulates Immune Responses through

Epigenetic Manipulation

Thesis directed by Associate Professor Dohun Pyeon

ABSTRACT

High-risk human papillomaviruses (HPVs) are causally associated with multiple human cancers. Previous studies have shown that the HPV oncoprotein E7 induces immune suppression; however, the underlying mechanisms remain unknown. We report that, while expression of many proinflammatory chemokines increases throughout HPV-positive cancer progression, CXCL14 is dramatically downregulated by promoter hypermethylation in an E7- dependent manner. Our in vivo mouse models revealed that restoration of Cxcl14 expression in HPV-positive mouse oropharyngeal carcinoma cells clears tumors in immunocompetent syngeneic mice, but not in Rag1-deficient mice. Further, restoration of

Cxcl14 expression significantly increases natural killer (NK), CD4+ T, and CD8+ T cell infiltration into the tumor-draining lymph nodes in vivo. In vitro transwell migration assays show that restoration of Cxcl14 expression induces chemotaxis of NK, CD4+ T, and CD8+ T cells. These findings suggest that high-risk HPV E7 is likely to dysregulate host gene expression in order to persist by modulating host DNA methylation. To investigate the extent of gene expression dysregulated by HPV E7-induced DNA methylation, we performed parallel global gene expression and methylome analyses with HPV-positive and -negative normal immortalized keratinocyte lines, NIKS, NIKS-16, NIKS-18, and NIKS-16∆E7. We show that expression of the MHC class-I genes HLA-A, -B, -C, -E, and -G is downregulated in HPV-positive keratinocytes in an E7-dependent manner. Additional methylome analysis revealed hypermethylation at a distal CpG island (CGI) near the HLA-E gene in HPV16- positive keratinocytes (NIKS-16) cells compared with NIKS and NIKS-16∆E7 cells. The HLA-

E CGI functions as a promoter exhibiting transcription factor binding sites and DNase

iii hypersensitivity. Promoter activity of the HLA-E CGI is considerably decreased by DNA methylation. HLA-E protein expression also is downregulated by high-risk HPV16 and

HPV18 E7 expression, but not by low-risk HPV6 and HPV11 E7 expression. Conversely, demethylation in the HLA-E CGI by treatment with 5-aza-β’-deoxycytidine restores HLA-E protein expression in HPV-positive keratinocytes. Because HLA-E plays an important role in antiviral immunity, epigenetic downregulation of HLA-E by high-risk HPV E7 may contribute to HPV-induced immune suppression during HPV persistence. Our findings provide a new mechanistic understanding of virus-induced immune evasion that contributes to cancer progression.

The form and content of this abstract are approved. I recommend its publication.

Approved: Dohun Pyeon

TABLE OF CONTENTS

CHAPTER

I. INTRODUCTION……………………..……………..……………………………………….1

Human Papillomavirus: Disease and Global Implications ...... 1

The Molecular Biology of Papillomavirus ...... 3

Papillomavirus Oncogenes and Mechanisms of Transformation ...... 7

E5 and Dysregulation of Immune Signaling ...... 8

E6 Abrogation of p53 and Evasion of Apoptosis ...... 9

Transformation by E7 through Dysregulation of Host Gene Expression ...... 10

Papillomavirus-Induced Neoplastic Transformation ...... 15

Viral Persistence and Cancer Progression ...... 17

Evaluating General Risk Factors for HPV-Associated Cancer Progression ...... 17

HPV Persistence and Host Epigenetic Consequences ...... 18

Virus Evasion of Immune Detection ...... 21

Evasion of Resident Antigen Presenting Cells ...... 22

Dysregulation of Antigen Presentation ...... 25

Chemokines and Inflammation...... 27

Inflammation and HPV-Associated Cancer Progression ...... 27

CXCL14 is a Potential Tumor Suppressor ...... 28

Dissertation Goals and Objectives ...... 29

II. MATERIALS AND METHODS ...... 32

Cell and Tissue Culture ...... 32

Conventional PCR and Quantitative Reverse Transcription-PCR (RT-qPCR) ...... 33

Bisulfite Modification and Assessment of Methylated DNA ...... 34

Vectors and Plasmids ...... 35

Luciferase Reporter Vectors ...... 35

Lentiviral Vectors ...... 36

Enzyme-Linked Immunosorbent Assay (ELISA) ...... 36

Cell Migration Assays ...... 37

Scratch Assay ...... 37

Transwell Migration ...... 37

Mice and Treatment ...... 38

Flow Cytometry ...... 39

Antibodies ...... 39

Sample Preparation and Flow Analysis ...... 39

Array Preparation and Analysis ...... 40

Genome-wide Expression and DNA Methylation Arrays ...... 40

Data Processing and Statistical Analysis ...... 40

Bioinformatics ...... 41

Immunoblotting ...... 42

Statistical Analysis ...... 42

Generation of HPV16 Reporter Virions ...... 43

Reagent Acknowledgements ...... 43

III. SUPPRESSION OF ANTITUMOR IMMUNE RESPONSES BY HUMAN PAPILLOMAVIRUS THROUGH EPIGENETIC DOWNREGULATION OF CXCL14 ...... 44

Importance ...... 44

Introduction ...... 44

Results ...... 46

Proinflammatory Chemokines are Upregulated During CxCa Progression ...... 46

CXCL14 Expression is Downregulated in HPV-Associated Cancer Progression ...... 47

CXCL14 Downregulation in HPV-Positive Keratinocytes is Associated with Promoter Hypermethylation ...... 50

CXCL14 Expression Hinders Cell Migration in vitro ...... 53

Restoration of Cxcl14 Expression Clears HPV-Positive Tumors in Immunocompetent Mice, but not in Rag1-Deficient Mice ...... 55

Restored Expression of Cxcl14 Increases Natural Killer (NK), CD4+ T, and CD8+ T Cells in Tumor-Draining Lymph Nodes in vivo...... 57

Expression of Cxcl14 Induces Chemotaxis of NK, CD4+ T, and CD8+ T Cells in vitro ...... 59

Discussion ...... 62

IV. HIGH-RISK HUMAN PAPILLOMAVIRUS E7 ALTERS HOST DNA METHYLOME AND REPRESSES HLA-E EXPRESSION IN HUMAN KERATINOCYTES ...... 66

Importance ...... 66

Introduction ...... 66

Results ...... 68

The HPV oncoprotein E7 drives global gene expression changes in human keratinocytes ...... 68

The HPV oncoprotein E7 downregulates gene expression related to antigen presentation ...... 72

The HPV oncoprotein E7 dysregulates DNA methylation in human keratinocytes ...... 75

DNA methylation of the HLA-E CGI is significantly increased by the HPV oncoprotein E7 ...... 79

The promoter activity of the HLA-E CGI is regulated by DNA methylation ...... 82

HLA-E is downregulated by high-risk HPV E7, but not by low-risk HPV E7 ...... 87

Discussion ...... 88

V. DISCUSSION AND FUTURE DIRECTIONS ...... 93

Discussion ...... 93

CXCL14: The Epithelial Homeostatic Chemokine ...... 93

Immune Evasion and Viral Persistence ...... 100

vii

E7-Directed DNA Methylation as a Viral Strategy for Immune Evasion and Persistence ...... 106

Additional Future Directions ...... 113

Repression of CXCL14 may Induce Tolerance of HPV-Infection ...... 113

What is the nature of interactions between HPV E7 and DNMT1? ...... 115

Concluding Remarks ...... 117

REFERENCES ...... 120

APPENDIX ...... 157

A. Primers Used for PCR and RT-qPCR ...... 157

B. Primers Used for Methylation Specific PCR (MSP) ...... 158

C. Primers Used for Molecular Cloning ...... 159

D. Supplemental Figures for Chapter III ...... 160

E. Supplemental Figures for Chapter IV ...... 169

F. Detailed Protocol for Designing and Performing Methylation Specific PCR ...... 197

G. Detailed Protocol for Dual Luciferase Reporter Assay ...... 200

viii

LIST OF TABLES

TABLE

4-1. CCNA1 and TERT Exhibit Increased Methylation in HPV16-Positive Keratinocytes ...... 78

LIST OF FIGURES

FIGURE

1-1. HPV Infectious Lifecycle ...... 5

1-2. HPV16 Genome ...... 6

1-3. HPV E7 Activates Progression of the Cell Cycle Through Inhibition of Rb ...... 12

1-4. HPV-Associated Cervical Cancer Progression ...... 16

1-5. Host Immune Surveillance of the Skin ...... 23

3-1. CXCL14 Expression is Downregulated During HPV-Associated Cancer Progression ...... 48

3-2. CXCL14 Downregulation in HPV-Positive Epithelial Cells is Associated with CXCL14 Promoter Hypermethylation ...... 51

3-3. CXCL14 Expression Hinders Mobility of HPV-Positive Cancer Cells...... 54

3-4. Restoration of Cxcl14 Expression Clears HPV-Positive Tumors in Immunocompetent Mice, but not in Rag1-Deficient Mice ...... 56

3-5. Cxcl14 Expression Increases NK, CD4+ T and CD8+ T Cells in Tumor- Draining Lymph Nodes ...... 58

3-6. Cxcl14 Expression Restores Decreased Populations of NK, CD4+ T and CD8+ T Cells in TDLNs ...... 60

3-7. Cxcl14 Expression Induces Chemotaxis of NK, CD4+ T and CD8+ T Cells ...... 61

4-1. High-Risk HPV E7 Distinctly Alters Host Gene Expression in Normal Keratinocytes ...... 69

4-2. High-Risk HPV E7 Downregulates HLA-I Gene Expression in Normal Keratinocytes ...... 74

4-3. HPV16 E7 Alters Host Genome Methylation in Normal Keratinocytes...... 76

4-4. HPV16 E7 Is Necessary for Hypermethylation at a Distal HLA-E CpG Island .... 80

4-5. Promoter Activity of the HLA-E CGI is Repressed by DNA Methylation ...... 83

4-6. High-Risk HPV E7, but not Low-Risk HPV E7, is Sufficient for Downregulation of HLA-E Expression in Normal Keratinocytes ...... 85

5-1. CXCL14 Expression Reduces HPV Infectivity in Keratinocytes...... 98

5-2. CXCL14 acts to “Fine Tune” Signaling through the CXCR4/CXCL12 Signaling Axis ...... 101

5-3. MICA and MICB Expression in HPV-Positive Cancer and Keratinocytes ...... 105

5-4. Murine Cxcl14 Expression Decreases Myeloid-Derived Suppressor-Like Cells (MDSC-Like) in Rag1-/- Tumors and Spleens ...... 115

5-5. Model of HPV E7 Immune Suppression through Epigenetic Manipulation of the Host ...... 118

LIST OF ABBREVIATIONS

5-aza 5-aza-β’-deoxycytidine APC antigen presenting cell ATCC American type culture collection ATP adenosine triphosphate B bone marrow-derived (lymphocyte) B2M beta-2 microglobulin BCA bicinchoninic acid BRAK breast and kidney CD cluster of differentiation CDC Centers for Disease Control and Prevention CDK cyclin-dependent kinase CGI CpG island ChIP chromatin immunoprecipitation Chr chromosome CIN cervical intraepithelial neoplasia CM conditioned medium CMV cytomegalovirus CR conserved region CRE cyclic-AMP response element CxCa cervical cancer CXCL chemokine containing C-X-C motif, ligand CXCR chemokine containing C-X-C motif, receptor DAC decitabine DC dendritic cell DMEM Dulbecco's modified Eagle medium DMP differentially methylation position DMR differentially methylated region DNA deoxyribonucleic acid DNMT DNA methyltransferase dNTP deoxyribonucleic triphosphase E early EBV Epstein-Barr virus ECL enhanced chemiluminescence ECM extra cellular matrix EDTA ethylenediaminetetraacetic acid EF1 human elongation factor-1 EGF epidermal growth factor EGFR epidermal growth factor receptor ELISA enzyme-linked immunosorbent assay ER endoplasmic reticulum et al. ei alia (and others) FBS fetal bovine serum FDR false discovery rate FITC fluorescein isothiocyanate G1 gap 1 phase GCR glucocorticoid receptor gDNA genomic-DNA GE gene expression GEO Gene Expression Omnibus

xii

GPCR G-protein coupled receptors H3K27Ac histone 3 lysine 27 acetylation HAT histone acetyltransferase HDAC histone deacetylase HEK human embryonic kidney HFK human foreskin keratinocyte HG human genome HLA human leukocyte antigen HNC head and neck cancer HNSCC head and neck squamous cell carcinoma HPV human papillomavirus HSPG heparan sulfate proteoglycans i.e. id est (that is) IFN interferon IgG immunoglobulin G IL interleukin IRF interferon regulatory factor ISG interferon stimulatory gene KLK kallikreins L late LC Langerhans cells LN lymph nodes MAP mitogen-activated protein MAPK mitogen-activated protein kinase MCS multiple coding site MDSC myeloid-derived suppressor cells Me Methyl group MHC major histocompatibility complex MIC MHC class I polypeptide-related sequence miRNA microRNA MMP matrix metalloproteinases MOE murine oropharyngeal epithelial mRNA messenger RNA MSP methylation specific PCR ncRNA non-coding RNA NF-κB nuclear factor κB NIH National Institutes of Health NIKS normal immortalized keratinocytes NK natural killer NKT natural killer T OPSCC oropharyngeal squamous cell carcinoma ORF open reading frame PBS phosphate buffered saline PCA principal component analysis PCR polymerase chain reaction PE phycoerythrin PEI polyethylene amine pRb retinoblastoma protein qMSP quantitative methylation specific PCR Rag recombination-activating gene RBC red blood cell

xiii

RLU relative light units RNA ribonucleic acid RPMI Roswell Park Memorial Institute RT-PCR reverse transcription PCR RT-qPCR quantitative reverse transcription PCR S (DNA) synthesis SSC side scatter SCID severe combined immunodeficiency SWAN subset-quantile within array normalization T thymus TAP transporter associated with antigen processing TCGA The Cancer Genome Atlas TDLN tumor-draining lymph node TERT telomerase reverse transcriptase TF transcription factor TGF tumor growth factor TME tumor micro environment TNF tumor necrosis factor V-ATPase vacuolar-type H+ ATPase VLP virus-like particle VSVG vesicular stomatitis virus glycoprotein WHO World Health Organization

xiv

CHAPTER I

INTRODUCTION

Human Papillomavirus: Disease and Global Implications

Papillomaviruses represent a diverse group of small, non-enveloped, double- stranded DNA viruses belonging to the ancient family, Papillomaviridae. While papillomaviruses are known to infect a large variety of animals including birds, fish, mammals, and reptiles, each genotype is species-specific and highly tissue tropic (1, 2).

Over 180 human papillomavirus (HPV) genotypes have been sequenced and characterized

(3). This diverse group of viruses is further sub-divided based on tissue tropism (mucosal or cutaneous) and oncogenic risk (high- or low-risk) (1, 4, 5). To date, 25 HPV genotypes have been identified as causative agents in cancer (high-risk), while at least 11 genotypes have been classified as non-cancer causing (low-risk) (6, 7). Of the high-risk mucosotropic HPVs,

HPV16 and -18 are causal in a large majority (70-90%, depending upon anatomical location) of HPV-associated cancers (8–10). Of the cutaneotropic genotypes, HPV5 and -8 are examples of genotypes which infect skin and may cause epidermodysplasia verruciformis lesions (11, 12). While prophylactic vaccines are available for protection against several cancer-causing genotypes, rates of HPV-associated disease continues to grow word-wide, making HPV a continuing global health concern. As a significant cause of morbidity and mortality worldwide, high-risk HPVs are known to be causal in over 5% of all human cancers, including over 99% of cervical cancers, as well as about 25% of head and neck squamous cell carcinoma (HNSCC) (9, 13–15).

The World Health Organization (WHO) estimates that up to 80% of sexually active individuals will become infected with HPV within their lifetime, making HPV an incredibly successful human pathogen, and the most prevalent sexually transmitted disease worldwide

(16–18). While the majority of infections are cleared naturally by the host, 10-20% of infections persist; of these, a small fraction will progress to develop cancer (19). Worldwide,

1 the prevalence of high-risk HPV genotypes in women is estimated to range by region from

2% to 35%, with regions observing the highest prevalence in the developing world (18, 20,

21). Further, it is estimated that worldwide 610,000 new HPV-associated cancers will be diagnosed annually, and greater than 80% of these will occur in developing countries (18). It is estimated that 274,000 women die each year worldwide due to HPV-related cervical cancer, 88% of whom live in developing nations (22). Furthermore, the World Bank estimates that cervical cancer alone causes about 10% of all economic loss in developing and low-income countries (22). A recent report exploring the global burden of cervical cancer estimates that by the year 2030, cervical cancer incidents will increase by 2%, making HPV-associated disease a relevant and important global health concern (18).

HPV prevalence in healthy women in the United States is highest in women 20-30 years of age (prevalence over 40%) and an average rate of infection being 25% (23–26). Of these infections, HPV16 is the most prevalent and the rate of infection drops with age (27,

28). Cervical cancer rates in the United States have continued to decline primarily due to increased rates of screening and detection methods (29, 30). In 2012, the Centers for

Disease Control and Prevention (CDC) estimated that over 12,000 women were diagnosed with cervical cancer in the United States, and over 4,000 women died from the disease (31).

While rates of cervical cancer in the United States are dropping, rates of HPV-associated oropharyngeal squamous cell carcinoma (OPSCC) are increasing (32, 33). It is estimated that by the year 2020, HPV-positive OPSCC will comprise the majority of OPSCC in the

United States (32, 33). With the growing rate of HPV-positive OPSCCs in the United States, and the projected increase in cervical cancers worldwide, HPV remains a relevant global pathogen and important health risk.

On the forefront of prophylactic therapies designed to prevent cervical cancer are three relatively new HPV vaccines available through Cervarix, and Gardasil (27, 34). All available vaccines protect against HPV16 and 18, together causal in 70% of cervical

2 cancers. Protection against a handful of additional genotypes is also offered (27, 35). All

HPV vaccines currently available have proven to be safe and efficacious, protecting against

95% of HPV-associated anogenital diseases (36–38). Efficacy against HPV-associated

HNSCC, however, remains to be determined. Vaccination rate for girls of non-Hispanic white decent in the United States remains low at 30%, while some minorities have vaccination rates as little as 11% (39, 40). Relatively low vaccination rates in the United States are due, in part, to high cost, social stigma, and the propagation of misinformation (41, 42). A recent study from the Centers for Disease Control and Prevention aimed to assess incidents of cervical neoplasia and associated HPV genotypes since the introduction of the vaccine (43).

The study genotyped HPVs from over 13,000 cervical lesions and found that, while 49% of the women had received the HPV vaccine, the most prevalent HPV genotype was still

HPV16. Additionally, their study found seven additional oncogenic HPV genotypes associated with the cervical lesions (35, 39, 51, 56, 59, 66, 68) that are not protected against by any vaccine, prompting continued efforts to develop a pan-valent vaccine.

Additionally, the incidence of HPV-associated OPSCC in the United States is increasing at an epidemic rate, particularly in men, who are not encouraged to receive the HPV vaccine and thus offered no protection against HPV-associated disease (44, 45). Taken together, these observations suggest that, while vaccines protecting against HPV are available, HPV- associated disease will continue to be a global health burden for years to come.

The Molecular Biology of Papillomavirus

Papillomaviruses are a genetically diverse group of non-enveloped viruses containing a circular double-stranded DNA genome of about 8kb (3, 46). The general structure of the viral capsid is icosahedral, consisting of 72 capsomeres each composed of five L1 major capsid molecules (47). The viral genome and bound host-derived nucleosomes are held within the capsid by the minor capsid protein L2, a factor important for viral packaging as well as infection (48).

HPV is a highly tropic virus with each genotype preferentially infecting species- specific cutaneous or mucosal epithelial keratinocytes (49, 50). HPV infection occurs through direct contact with infected cutaneous (for example, hands and feet) or mucosal (for example, anogenital and oral mucosa) skin. HPV only infects the basal layer of undifferentiated keratinocytes and therefore must gain access to these cells through micro- tears in the skin (viral life cycle summary in Figure 1-1). The exact route of entry and the receptors involved are currently controversial; however, it is thought that integrins and heparan sulfate proteoglycans (HSPG) are important for entry (51, 52). Once the cell has been entered, the virus traffics to the nucleus via the endosome and Golgi apparatus where it begins genome maintenance and viral gene expression (53, 54).

Overall, HPV encodes eight to nine overlapping open reading frames (ORF, Figure

1-2). Transcription from the HPV genome is tightly regulated and organized into early (E) and late (L) gene expression (55). After trafficking of the viral genome into the nucleus, HPV begins expression of early genes E1 and E2, which are required for viral genome replication and tethering of the viral genome to host chromosomes (56, 57). More specifically, the viral

E1 gene encodes a helicase that is required for genome replication, while E2 encodes the major transcription factor that controls expression of viral genes. Importantly E2 inhibits expression of the viral oncogenes E6 and E7, which will be discussed in depth later (58, 59).

Once the genome is established within the host cell, the virus begins its replication program.

Ordinarily, basal epithelial keratinocytes divide vertically, pushing the daughter cells up to create a stratified protective barrier of cells. The daughter cells then arrest cell division and undergo programmed differentiation and keratinization, and in cutaneous skin these cells are eventually sloughed off naturally through a process known as cornification (Figure

1-1) (60). It should be noted that HPV requires a host cell to be actively cycling in order to gain access to the nucleus, thus only the basal layer of keratinocytes can be infected (61).

HPV therefore establishes infection in the basal keratinocytes, which then divide and move

Figure 1-1. HPV Infectious Lifecycle. The infectious lifecycle of HPV is shown, which occurs within the stratified epithelium (differentiation is estimated by the black triangle). HPV capsids (red icosahedral) gain access to the basal epithelium through micro-tears in the skin. The virus infects the most basal keratinocytes (red cells and nuclei) attached to the ECM (thatched pattern) and begins genome maintenance and early gene expression (quantified by blue triangle). As the cells divide and stratify upward, the virus begins to synthesize late structural proteins (quantified by green triangle). As cells are naturally slough off, new virions are released.

Figure 1-2. HPV16 Genome. Gene expression from HPV16 is complex and transcripts originate from all three reading frames (indicated by ORF staggering). Gene expression is divided into early (E) and late (L) genes. Early genes regulate viral gene expression (E2), cell cycle (E7), genome maintenance (E2), and replication (E1) among other functions. The late genes encode the major (L1) and minor (L2) capsid proteins. Three major cis regulatory elements are included within the genome: two poly-A signals, one for early (poly AE) and one for late (poly AL) genes, and an upstream regulatory region (URR) are shown. Red text denotes oncogenes (E5, E6, and E7) and blue text denotes capsid genes (L1, and L2).

6 vertically up through the stratified epithelium. These cells have a natural tendency to arrest cell cycle and differentiate; therefore, HPV must intervene to encourage continued DNA synthesis, avoid apoptosis and maximize virion production. To achieve this, HPV expresses three oncogenes: E5, E6, and E7. Briefly, E5 expression enhances epidermal growth factor

(EGF) and mitogen-activated protein (MAP) signaling, resulting in increased DNA synthesis and cell proliferation (Reviewed, (62)). E6 is a well characterized oncogene that targets the tumor suppressor p53 for degradation by the proteasome, thereby negating cell cycle arrest and apoptosis and allowing for continued viral propagation (Reviewed, (63)). Finally, the oncoprotein E7 binds and targets the tumor suppressor and cell cycle regulator pRb for degradation, thereby directing the cell into S phase, cell proliferation, and virus DNA synthesis (Reviewed, (64)). The functions of these oncogenes are important not only for virus propagation but for evasion of the host immune response and will be further discussed below.

As the infected cells move up through the stratified epithelium, they begin expression of late structural genes L1 and L2 (65). Additionally, amplification of the viral genome is dramatically increased, which will be bound by L2 proteins and self-assembled into mature

L1 virions (66, 67). In the upper keratinized layers of the epithelium, the viral E4 protein degrades the cytokeratin matrix, making the cells vulnerable to rupturing for virus release

(68, 69). As the host cell begins to die and mitochondrial function fades, the cellular environment becomes oxidizing, allowing for the L1 capsid proteins to form disulfide bonds, generating mature viral particles that can be released. Thus, the viral life cycle begins anew

(70).

Papillomavirus Oncogenes and Mechanisms of Transformation

A link between HPV and abnormal skin lesions was first explored in 1966 by the

Jablonska laboratory in Poland, who isolated HPV5 from epidermodysplasia verruciformis

(71). The association of HPV with genital warts and subsequently cervical cancer was

7 initially established by the zur Hausen laboratory in the early 1980s. Their original work identified a small viral DNA isolated from genital warts, recognizing the new virus as HPV6

(72, 73). Subsequent work from the zur Hausen laboratory performed on cervical cancer lesions identified another new papillomavirus, HPV16 (74). It has since been established that high-risk papillomavirus encodes three oncogenes (E5, E6, and E7), which perform unique functions that contribute to cellular transformation.

Like most cancers, HPV-associated cancer development requires decades to progress from HPV-infected cells to invasive disease. Recent cancer genomics studies of

HNSCCs have reported that HPV-positive HNSCCs have far fewer mutations (~5 per tumor) when compared to HPV-negative HNSCCs (>20 per tumor) (75, 76). These results indicate that viral factors replace multiple oncogenic processes usually achieved by multiple somatic mutations in HPV-negative cancer progression. For example, none of the HPV-positive

HNSCCs tested had a somatic mutation in TP53, but rather, p53 is efficiently degraded by the HPV oncoprotein E6. On the other hand, 78% of HPV-negative HNSCCs contain one or more mutations in the TP53 gene (75). Other studies showed that expression of the HPV oncoprotein E7 is required for continuous cancer growth and maintenance in vitro and in vivo (77–81), suggesting that the HPV oncoprotein E7 performs multiple functions in HPV- associated cancer progression. While the mechanisms by which HPV infection contributes to multiple steps of decades-long cancer progression is not fully understood, much work has been done to establish the role of HPV as a true driver of cell transformation and to uncover the oncogenic mechanisms employed by the each of the three encoded oncoproteins.

E5 and Dysregulation of Immune Signaling

HPV E5 is a Golgi and ER-resident small transmembrane homodimer (82). Of particular interest here are the immune evasion strategies employed by E5 in the context of an infected host cell. E5 dysregulates antigen presentation in infected cells using a multifaceted approach which underscores the importance of inhibiting antigen presentation

8 to HPV. Antigen presentation occurs in all nucleated cells, and acts as a mechanism for T lymphocytes to recognize and destroy cells expressing non-self proteins (for example, infected cells expressing viral proteins). E5 inhibits major histocompatibility complex class I

(MHC-I) maturation and transport to the cell surface by trapping MHC-I in the Golgi before they are shuttled to the cell surface (83, 84). This is accomplished through E5 inhibition of the proton pump V-ATPase, which alkalinizes the Golgi and inhibits vesicle budding (85).

Additionally, E5 inhibits activity of the chaperone protein calnexin, inhibiting maturation of

MHC-I complexes, again reducing surface expression (86). Finally, E5 is known to reduce

MHC surface expression comprised of classical human leukocyte antigen (HLA) subunits (A,

B, and C), which was shown to reduce recognition of infected cells by CD8+ T cells (87).

This evasion of immune recognition is an important viral strategy achieved by E5, underscored by its ability to negate T cell recognition of infected cells. Evasion of immune recognition allows the viral infection to persist, an important risk factor for HPV-associated cancer progression.

E6 Abrogation of p53 and Evasion of Apoptosis

The second of three oncoproteins encoded by papillomavirus is E6. E6 alone is sufficient to transform human mammary epithelial cells (88, 89) and co-expression with E7 in human keratinocytes, the natural host of HPV, is necessary and sufficient for transformation

(90, 91). Further evidence suggesting that E6 is a true driver of cancer comes from transgenic mouse studies performed by Arbeit et al. Here, transgenic mice expressing E6 and E7 driven by the tissue-specific keratin 14 promoter (expressing only in basal layers of stratified epithelium), develop tumors in multiple epithelial surfaces throughout the body

(92). Of note, experiments designed to determine the stage at which E6 contributes to cancer progression found that expression of E6 in transgenic mice results in relatively little increase in benign tumor abundance. However, when E6 is expressed in established benign lesions, there is a dramatic increase in malignant transformation. These findings suggest

9 that E6 promotes tumor progression to malignancy, rather than initiation of carcinogenesis

(93). Because E6 antagonized the p53 DNA damage response, E6 may act to promote tumor progression by allowing DNA damage and chromosomal rearrangements to accumulate, encouraging transition to a metastatic phenotype. Together, these observations solidify the role of HPV E6 as a potent oncogene.

The transformative properties of E6 are derived largely from its interaction with the host tumor suppressor p53, an important regulator of the cell cycle and apoptosis. E6 assembles complexes with the p53 tumor suppressor, leading to ubiquitination-dependent proteasomal degradation of p53 (94, 95). As expected, p53 degradation results in a cellular inability to arrest growth due to DNA damage as well as cellular evasion of apoptosis (96–

98). E6-mediated degradation of the tumor suppressor p53 therefore allows suprabasal infected epithelial cells to continue to divide by inactivating a process that ordinarily would activate p5γ and induce apoptosis. The abrogation of p5γ additionally will nullify the cell’s

DNA damage response, leading to an accumulation of somatic mutations that can contribute to malignant transformation.

In addition to its activation of p53 degradation, E6 carries out a myriad of functions that may lead to cellular transformation. First, E6 induces telomerase activity by activating transcription of the hTERT subunit, thereby extending the replicative potential of the cell

(99). Additionally, E6 inhibits NF-κB signaling, activates MAPK signaling, represses proapoptotic gene expression, and inhibits cell differentiation (100–103). The many functions performed by HPV E6 make clear its role as a potent oncoprotein.

Transformation by E7 through Dysregulation of Host Gene Expression

The transformative potential of the third HPV oncoprotein, E7, was discovered in the late 1980s. HPV16 E7 is sufficient to activate mouse NIH 3T3 fibroblast proliferation and malignant transformation (104). In vivo experiments revealed that, in mice, E7 expression alone is sufficient to form benign tumors (105). On the other hand, E6 promotes progression

10 of benign lesions to malignant tumors in the later stages of cancer development, but was not necessarily able to elicit benign tumor growth. However, E7 promotes initiation of carcinogenesis and synergizes with carcinogens to cause benign papillomas (93). These very important observations underscore the importance of E6 and E7, which complement each other as oncoproteins to bring potent neoplastic transformative potential to HPV.

The major oncogenic effects of E7 are due to its interactions with the Rb family of tumor suppressors. The pRb family of proteins are regulators of the E2F transcription factors which control the transition to G1/S phase during cell division (106). pRb is a cell cycle regulator that, when bound to E2F transcription factors, forms an pRb/E2F repressor complex (Figure 1-3). Once phosphorylated, pRb dissociates from E2F, trans-activating gene expression and forcing the cell to progress into S phase. E7 binds pRb proteins and targets them for proteasomal degradation, thereby releasing the cell from growth arrest, and allowing for viral DNA synthesis (107, 108). Additionally, HPV16 E7 binds directly to E2F1, activating transcription (109). The effects of E7-mediated modulation of cell cycle gene expression is exemplified in global gene expression studies comparing HPV-positive and negative cancer tissue, which shows a dramatic upregulation in cell cycle related genes in

HPV-positive cells (110, 111). The cell cycle activating function of E7 is necessary for successful viral genome replication. Suprabasal cells in the stratified epithelium ordinarily senesce and do not replicate DNA, which would not be conducive to productive HPV infection. Activation of cell division in these cells by E7 also contributes to cancer progression by sustaining their active replication, a hallmark of cancer.

The specific mechanisms differentiating the oncogenic potential of different E7 proteins remain to be fully elucidated. For instance, the oncogenic potential of low-risk HPV

E7 proteins has been shown to be weak, specifically in HPV6 and HPV11 (112–114). While it has been shown that high-risk HPV E7 proteins display a higher binding affinity for the pRb family of proteins compared to low-risk E7s, low-risk E7 proteins maintain the ability to

Figure 1-3. HPV E7 Activates Cell Cycle Progression Through Inhibition of Rb. The Rb family of tumor suppressors bind the E2F transcription factors, inhibiting cell cycle progression (left). The viral oncoprotein E7 binds pRb and targets it for degradation in proteosomes, thereby releasing E2F driven gene expression and leading to cell cycle progression (right).

12 target pRb for degradation and only partially explains differences in oncogenic potential

(115–117). Consistently, both high- and low- risk E7 proteins can push the cell into S phase for DNA synthesis, however low-risk E7 proteins appear to be insufficient to elicit sustained cell growth, inhibit senescence, or transform cells (118). Importantly, high-risk HPV16 and

18 E7 proteins repress transcription of classical antigen presentation genes HLA-A, -B and -

C while low-risk HPV6 and 11 could not (118, 119). While HPV6 and 18 E7 was sufficient to repress TAP-1 expression (involved in loading peptides onto antigen presentation complexes) HPV11 E7 was not sufficient to repress TAP-1 expression (118), suggesting that low-risk HPVs can partially disrupt antigen presentation through TAP-1 dysregulation.

Further delineating the distinct functions of high- and low-risk HPV E7 proteins will be important to more fully understand that mechanisms utilized by HPV to causes cancer.

The E7 oncoprotein may also contribute to neoplastic transformation through epigenetic remodeling of the host. Epigenetic markers can dramatically alter host gene expression leading to transformation and metastasis; therefore, changes in the epigenome can be classified as a hallmark of cancer (Reviewed, (120)). Similar to its association with

E2F1, low- and high-risk HPV E7 can directly bind E2F6, a repressive transcription factor which is part of the polycomb repressive complex, a well known chromatin remodeling complex (121–123). This interaction is linked to deregulation of the cell cycle through decreased abundance of polycomb repressive complexes overall, and likely resulting in a prolonged S phase, which is conducive to viral DNA replication. HPV E7 can also activate histone deacetylases (HDAC) as well as deactivate histone acetyltransferases (HAT) resulting in dysregulation of epigenetic histone markers and cancer cell proliferation (124–

126). Finally, high-risk E7 has been shown to directly bind and activate DNA methyltransferase I (DNMT1) in vitro, potentiating yet another mechanism for epigenetic remodeling of the host genome by HPV (127). HPV-associated cancers are known to carry distinct epigenetic markers, specifically in DNA methylation patterns, when compared to

HPV-negative cancers in the same anatomical location (128, 129). Our and other labs have shown that DNA methylation specific to HPV infection silences expression of individual genes (130, 131). Here, we showed that HPV E7 is both necessary and sufficient to direct hypermethylation of the CXCL14 chemokine promoter, silencing CXCL14 expression and thereby inhibiting immune responses to cancer progression (131). HPV evasion of immune detection in order to persist can therefore be partially attributed to E7-directed changes in host gene expression. Taken together, the observations describing the role of E7 in modifying host epigenetic control of gene expression identify E7 as an active, potent oncogene.

HPV E7 modulates cytokine signaling, contributing to tumor growth and evasion of the host immune response. For example, E7 expression alone is sufficient to abrogate the growth-suppressive effects of TGF-β in keratinocytes by re-activating c-myc expression and promoting cell proliferation (132). E7 expression also appears to desensitize keratinocytes to cell cycle inhibition and apoptosis by tumor necrosis factor (TNF) (133, 134). Finally, an immunoevasive and tumor promoting function of E7 can be seen by its ability to inhibit interferon signaling. E7 has been shown to hinder type-I interferon signal transduction in keratinocytes though inhibition of interferon regulatory factor-1 (IRF-1) by recruiting HDACs to IRF-1 regulatory elements, repressing interferon stimulated gene (ISG) expression (135,

136). Evasion of immune detection and evasion of anti-growth signals are two hallmarks of cancer that are directly due to functions of E7. Of particular interest is the repression of immune signaling by E7 through epigenetic rearrangements in the host genome. The multiple mechanisms utilized by E7 to manipulate host epigenetic control of gene expression present a relatively unexplored avenue that may explain the oncogenic potential of HPV.

The major oncogenic functions of E7 stem from its ability to modulate host gene expression through multiple mechanisms. Dysregulation of the cell cycle through releasing expression of E2F-driven genes and repressing CXCL14 expression through

14 hypermethylation of the CXCL14 promoter are two important examples. We hypothesize that E7 extensively modifies host gene expression through epigenetic modulation. A comprehensive analysis of host gene expression affected by E7 would be an important step to better understand HPV-associated disease progression.

Papillomavirus-Induced Neoplastic Transformation

As prevalent oncogenic pathogens, HPVs are causal in nearly 100% of cervical cancers, 25% of head and neck cancers including 70% of oropharyngeal cancers, and 5% of human cancers overall (9, 13–15). HPV-associated cancer development is possible due to the multiple functions of the viral oncoproteins, discussed previously. Viral persistence and evasion of immune detection are thought to be major risk factors for HPV-associated cancer development (4). Of the HPV-positive cancers, cervical cancer has been the most thoroughly studied; however, recent work elucidating the molecular patterns that distinguish

HPV-positive and -negative head and neck cancer is allowing for new insight into how HPV uniquely causes cancer. HPV-associated cancer progression of the cervix is a multi-step process: cervical intraepithelial neoplasia (CIN) 1-3, and squamous cell carcinoma/cancer

(Reviewed, (137)).

Disease stages of cervical cancer are segregated based on viral gene expression and disruption of tissue homeostasis and are summarized graphically in Figure 1-4.

Following infection, HPV can persist for decades, potentially leading to cancer. Importantly, because the virus can persist without immune clearance for decades, HPV must have evolved efficient immune evasive techniques to negate immune detection and clearance.

Approximately 70% of productive HPV infections persist and progress to CIN1 lesions, in which the epithelial architecture remains largely intact and exhibits only mild dysplasia, and viral gene expression remains regulated (138). While most CIN1 lesions regress and are cleared naturally (~90%), a small percentage will progress to CIN2/3, developing over the course of many years. CIN3 lesions exhibit severe dysplasia and viral oncogene

Figure 1-4. HPV-Associated Cervical Cancer Progression. HPV-associated cancer progression occurs over the course of years to decades. Once cells of the basal layer have been infected (red arrow, red nuclei), they begin the programmed viral lifecycle (See, Figure 1-1). Many infections will progress to CIN1 lesions wherein viral gene expression maintains regulation and only mild dysplasia is observed. The majority of CIN1 lesions will resolve and be cleared (green arrow) naturally. ~10% of CIN1 lesions will persist (maroon arrow) and progress to CIN2/3 lesions, in which viral gene expression becomes dysregulated: E6 and E7 are expressed to high levels and the stratified epithelium begins to lose its organized architecture. While many CIN2/3 lesions regress and are cleared naturally (green arrow), ~1% will continue to persist, disrupting tissue homeostasis until eventually the basement membrane is breached and the infected cells escape as metastatic cancer.

16 dysregulation: the viral episome is frequently integrated into the host genome, truncating the viral ORF encoding the E2 regulatory transcription factor and resulting in augmented E6 and

E7 oncogene expression (137, 138). Importantly, while CIN3 lesions are benign, they maintain a highly proliferative state and lose the ability to repair DNA sufficiently or undergo apoptosis due to extensive DNA damage, owing to the effects of E6 and E7 (139, 140).

CIN3 lesions may persist for >10 years, of which ~1% will further develop into cancer (138).

Once the basement membrane is breached by the infected cells, the lesion is considered squamous cell carcinoma, or invasive cancer. Because decades long persistence must occur before cancer is achieved, elucidating the mechanisms of viral evasion of host immune detection, or mechanisms of viral-induced immune tolerance remain attractive fields of study.

Viral Persistence and Cancer Progression

The effects of high-risk papillomavirus oncogenes on the host cells are profound.

Evasion of apoptosis, maintenance of constant growth signals, evasion of death signals, induction of telomerase activity, and preventing immune detection are hallmarks of cancer, all of which are achieved by the cooperative functions of three viral oncogenes, E5, E6, and

E7. The functions of these oncoproteins also allows for evasion of host immune detection and viral persistence potentiating development of cancer. While viral DNA integration into the host genome and relative viral load have been identified as potential risk factors in the development of HPV-associated cancer development, viral persistence and evasion of host immune defenses are arguably the most important risk factors.

Evaluating General Risk Factors for HPV-Associated Cancer Progression

Risk factors such as viral genotype (high- or low-risk), viral load, and viral DNA integration into the host genome are all important risk factors for HPV-associated cancer development. As discussed previously, viral genotype is the most important risk factor for cancer development, with HPV16 being the most prevalent genotype in HPV-associated

17 cancers (141). Studies assessing viral load in patients with different stage cervical lesions found that patients with CIN lesions are likely to carry a high viral load; however, viral load was found to be insufficient to act as a prognostic indicator of cancer risk (142, 143). These studies indicate that viral load is important for disease progression, but not a necessary factor. Work to determine the risk associated with viral genome integration has also been studied. Viral integration appears to be a random event with detrimental effects to the virus: the regulatory genes E1 and E2 are generally lost after integration due to truncation, disallowing viral replication and leaving expression of E6 and E7 unchecked. Continued overexpression of the E6 and E7 oncoproteins is a major contributing factor to cancer development due to their role in degrading the tumor suppressors p53 and pRb. While a strong correlation has been found between viral genome integration and cervical cancer development (144), additional research must be conducted to determine if lesions with integrated viral DNA are more likely to progress, and if other risk factors are involved such as the genomic location of integration, stability of viral mRNAs after integration, or the number of integration events. Finally, the role of tobacco and alcohol consumption in HPV- positive HNSCC cancer is debatable. While tobacco and alcohol consumption has been linked to an increase in the rate of HPV infection in the head and neck (145), alcohol and tobacco consumption are not necessarily risk factors for development of HPV-associated

HNSCC (146, 147). While the factors mentioned above may play roles in the development of HPV-associated cancer, arguably the most important risk factor is viral persistence.

HPV Persistence and Host Epigenetic Consequences

Studies have indicated that persistence of HPV infection and prolonged viral gene expression are major risk factors for HPV-associated cancer progression and development, but the precise mechanisms of this action are largely not understood (148–151). Cervical

HPV infections are typically cleared within the first two years of infection with lesions progressing and regressing until cleared, or cancer develops (Figure 1-4) (20, 152, 153).

With an average time to clearance of 9.4 months in a cohort study of University-age women,

HPV infections persist for relatively long periods of time (154). In a large scale meta-analysis of multiple studies linking HPV to cancer, it was concluded that persistence of HPV, especially high-risk genotypes 16 and 18, represents a significant risk factor for development of cervical neoplasia (155). Progression and regression of HPV-positive cervical lesions is very transient in nature (Figure 1-4) with the majority of CIN1, 2, and 3 lesions regressing and eventually clearing. As viral oncoproteins expressed from high-risk viruses are the true drivers of CIN lesion progression, the longer the virus persists and the longer viral oncoproteins are expressed, the more likely lesions are to progress to cancer.

Strikingly, HPV was found to persist in 20% of women in a cohort study assessing HPV prevalence in CIN lesions (156). This suggests that there are viral and/or host factors that determine whether viral infection will persist.

Recently, a link between viral persistence and accumulations of epigenetic lesions in the host genome has been established. DNA methylation is an important heritable epigenetic marker which is generally thought to be repressive (although there are exceptions), and it is intricately involved in gene regulation during development, homeostasis, and disease progression (Reviewed, (157)). Inhibition of gene expression by cytosine methylation (5-methylcytidine) can be achieved by blocking the binding of necessary transcription factors simply due to steric hindrance, or by recruitment of repressive 5-methylcytidine binding proteins, which can recruit HDAC repressive complexes

(158, 159). The transfer of the methyl group is achieved by the DNMT family of proteins, most of which are involved in development and meiosis. In tissue homeostasis, the DNMT1 methyltransferase is thought to be the primary methyltransferase for maintenance of 5- methylcytidines and can associate with HDAC complexes for further repressive epigenetic remodeling (160). As discussed previously, the viral oncoprotein E7 can directly bind the methyltransferase DNMT1, as well as HDAC and HAT complexes, modifying and potentially

19 directing their function. This leads to the hypothesis that HPV E7 can modulate the host epigenome in order to regulate host gene expression in turn establishing a more favorable host environment, leading to persistence. Accumulation of regional DNA methylation may take years to decades during cancer development; therefore, persistent viral E7 expression may be a key factor directing aberrant DNA methylation leading to cancer development.

The link between methylation and cancer has been known since the early 1980s, when it was discovered that DNA methylation was highly enriched in cancer cells compared to normal cells (161). More recently, methylation patterns in HPV-positive versus HPV- negative HNSCC patient samples has been shown to be unique suggesting that HPV- positive and -negative cancers exhibit different molecular markers (162). Using the identified unique set of differentially methylated sites, a recent study surveying changes in methylation patterns in patients with HNSCC and determined that the extent of methylation at certain loci can be used as a prognostic indicator for patient survival (163). While the gene regions associated with aberrant DNA methylation are not necessarily linked to disease progression, these studies indicate that HPV infection alone is sufficient to induce global epigenetic modifications that, as a whole, can influence the outcome of disease progression.

Multiple subsequent studies have further elucidated the large scale changes in host methylation distinct to HPV-positive cancers which alters host gene expression (129, 164,

165). While assessing methylation at different genomic loci over the course of HPV-induced immortalization of keratinocytes, a steady and gradual increase in methylation was observed at multiple host promoter elements (166). This supports the hypothesis that aberrant DNA methylation is accumulated over time during viral persistence and continued viral gene expression. Further supporting this hypothesis, we show here that accumulation of methylation at the promoter of the tumor suppressor chemokine CXCL14 occurs over the course of HPV-induced cellular transformation. This accumulation of promoter methylation is

E7-dependent, presumably through its interaction with DNMT1. CXCL14 promoter

20 methylation downregulates CXCL14 expression significantly. In HPV-positive tumors in mice, Cxcl14 downregulation decreases natural killer (NK) and CD8+ T cell surveillance near the tumor microenvironment, resulting in tumor growth. Finally, when Cxcl14 expression is restored in HPV-positive tumors in mice, NK and CD8+ T cells return to the tumor microenvironment and tumors are cleared. These observations not only elucidate a direct consequence of HPV-directed methylation of a host gene, CXCL14, they also underscore the importance of immune evasion techniques employed by the virus in order to persist to develop cancer.

Virus Evasion of Immune Detection

HPV evasion of immune detection allows for viral propagation and oncogene expression to go unnoticed by the host for long periods of time and is therefore closely linked to persistence and cancer development. While most HPV infections will be cleared within the first two years of infection, a small subset will persist for decades evading clearance by the host immune system (20, 152, 153). A primary HPV infection can persist for years and the HPV life cycle does not elicit an immune response through cytolysis, necrosis, or inflammation (Figure 1-1). Together, these observations suggest that an adaptive immune response to the infection is circumvented (167). Evidence for this evasion is further supported by the efficacy of HPV vaccines: 99% of women vaccinated against

HPV16 and -18 carry antibodies against the virus when tested four years post vaccination; additionally, the vaccine is estimated to be up to 100% efficacious against incident HPV16 and -18 anogenital infection (168, 169). Because the vaccine elicits a strong humoral immune response and protects against nearly 100% of HPV infections (of the genotypes vaccinated against) it stands to reason that an adaptive immune response is capable of eliminating the virus. A humoral response therefore must be evaded by HPV immediately upon infection to allow for viral replication and persistence. Additionally, when serum antibody levels against the HPV oncogenes E6 and E7 were evaluated in a cohort of over

600 women, it was found that 66% of women with invasive cervical cancer were seropositive, while only 10% of the control women tested seropositive (170). This indicates that, during a normal HPV infection cycle, the virus goes unnoticed to an adaptive immune response and it is not until HPV-positive cells disrupt the epithelial basement layer during cancer progression that the virus is detected (Figure 1-4) (170). Therefore, HPV must employ mechanisms of immune evasion, because recognition by the immune system is detrimental to productive infection and persistence.

Because establishment of infection requires actively dividing cells, papillomavirus will infect only the basal, undifferentiated epithelial cells of the skin or mucosal surfaces. These cells then undergo polar vertical division, pushing the infected daughter cells up through the stratified epithelium where they initiate a cascade of viral gene expression, culminating in infectious capsid production and release from cells that will be sloughed off naturally by the host. In this way, infection can be undetected due to two strategies employed by the virus.

First, the entire HPV lifecycle occurs in resident epidermal cells away from adaptive immune detectors in the dermis and peripheral blood, such as dendritic cells (DC), T cells, and natural killer (NK) cells (Figure 1-5). Second, as progeny virions are released without causing damage to the host, viral shedding does not alert the immune system of infection.

Infected primary keratinocytes in the basal layer of the epidermis are potentially susceptible to detection from immune cells of the dermis; however, viral gene expression only occurs in low amounts. In fact, investigators have hypothesized that the virus is latent in these cells, providing one mechanism of immune evasion (171).

Evasion of Resident Antigen Presenting Cells

While the HPV lifecycle is completed in the epidermis, the virus must contend with one major adversary at these sites: Langerhans cells (LC) Figure 1-5). LC are specialized immature dendritic cells that reside in mucosal and squamous epidermal tissue and are thought to exist as several subtypes that act as antigen-presenting cells (APCs) as well as

Figure 1-5. Host Immune Surveillance of the Skin. The HPV viral lifecycle is compartmentalized in the epidermis away from many adaptive and innate immune sensors. Infected cells (red cells, red nuclei) divide in the mid zone of the epidermis and viral particles are shed in the superficial zone. Resident Langerhans cells (LC) are specialized dendritic cells (DC) that reside primarily in the mid zone of the epidermis and therefore the most likely immune cells to discover infected cells. Fibroblasts generate the basement membrane and connective tissues of the dermis. T cells, natural killer (NK) and DCs patrol the dermis for pathogens. Capillaries (not shown) perforate the dermis, delivering T, NK, and DCs to sites of inflammation.

23 immunosuppressors to regulate T-cell activity in the skin (172–176). As the primary APC at the site of HPV infection, viral evasion of these cells is multifaceted, which begins upon first viral contact with LC. Recent studies have shown that, while HPV virus-like particles (VLPs) can be internalized by LC, they do not result in LC activation and are not presented to T cells (177, 178). These findings are contrary to observations that dendritic cells (DCs) are fully activated due to stimulus by HPV VLP, suggesting that HPV is specifically escaping detection by resident APCs (179, 180). Finally, a study comparing the activating potential of the major and minor HPV capsid proteins (L1 and L2, respectively) found that, while L1 stimulus alone activated LC, L2 stimulus did not, suggesting a specific role of the L2 protein in suppressing LC activation (181).

In addition to evasion of immediate detection by LC in the epidermis, HPV has obtained strategies to elude LC after infection has been established. For example, E6 expression alone in epithelial cells inhibits differentiation of contacting monocytes to LC in an in vitro co-culture model, suggesting another LC evasion tactic utilized by the virus (182).

E6 has also been shown to be sufficient to downregulate E-cadherin in keratinocytes and in

HPV-positive cervical lesions, reducing epithelial contact with LC, which then migrate away and are depleted from the HPV-positive lesion (183). Exclusion of immune cells from HPV- positive lesions by chemokines was also observed by Trimble et al. (184). While CCL19 and

CCL21 chemoattractants are upregulated in HPV-positive cervical lesions, the endothelial vasculature is not permissive to CD8+ T cell infiltration into the lesional tissue. This observation indicates that HPV-positive keratinocytes are able to signal to endothelial cells to downregulate the surface expression of vascular adhesion proteins in order to evade immune detection. Barring immune cells from investigating HPV-positive lesions may therefore be a highly effective immunoevasive mechanism, which leads to persistence and cancer progression.

Dysregulation of Antigen Presentation

Further evidence for immunosuppressive function of the HPV oncoproteins can be seen in cells expressing E7. For instance, LC are functionally deficient in migratory potential and antigen uptake in mice expressing E7 in epithelial tissues (185). Functions return in LC following isolation and stimulation in vitro, suggesting that E7 is modulating epithelial production of APC suppressive factors. This observation highlights another mechanism of viral suppression of LC (185). Even more striking, Tindle et al. reported that E7-expressing epithelium grafted onto mice of the same genetic background did not reject the transplant, while tissue-expressing antigenic human growth hormone elicited a strong CD8+ T cell response (186). This evidence supports the conclusion that E7 expression alone is sufficient for evasion of immune detection by HPV infected cells, which occurs, in part, through disruption of antigen presentation.

Antigen presentation in keratinocytes and indeed, all nucleated cells, involves the major histocompatibility complex (MHC) class-I proteins (Reviewed, (187)). Briefly, MHC-I complexes are comprised of a human leukocyte antigen (HLA) heavy chain subunit, which binds and presents the antigen, and beta-2 microglobulin (B2M) which stabilizes the complex. HLA proteins are divided into classical (A, B, and C) and non-classical (E, F, and

G) proteins. Classical HLA proteins are responsible for presenting peptide fragments from proteins generated by the cell to T lymphocytes for recognition of potentially foreign protein production. Non-classical HLA proteins are less well understood but appear to play important roles in antigen presentation for innate immunity as well as facilitate immune tolerance during pregnancy.

HPV E5 and E7 each dysregulate antigen presentation in infected cells, underscoring its importance to HPV. Specifically, E5 inhibits MHC class I maturation and transport to the cell surface by trapping it in the Golgi before it can be shuttled to the cell surface (83, 84). This is accomplished through E5 inhibition of the proton pump V-ATPase,

25 which alkalinizes the Golgi and inhibits vesicle budding (85). Additionally, E5 inhibits activity of the calnexin chaperone protein, inhibiting classical HLA maturation and surface expression (86). Finally, E5 reduces MHC surface expression of classical HLA subunits (A,

B, and C), which was shown to interfere with recognition of infected cells by CD8+ T cells

(87).

The effect of HPV E7 on MHC expression is also complex. E7 expression downregulates classical HLA (A, B, and C) surface expression, which incidentally increases cellular vulnerability to NK cell-mediated lysis. Presumably, this is due to decreased overall surface MHC expression (188). While E7 expression is sufficient to downregulate surface expression of MHC complexes, no changes in HLA transcription have previously been observed (189, 190). The importance of downregulation of MHC surface expression has been underscored in multiple studies. When presented on the cell surface, HPV-specific antigens can be identified by classical HLA-restricted CD8+ T cells, leading to lysis of the infected cell (191–194). It is therefore imperative for the virus to negate this response by modifying MHC surface expression for evasion of immune responses and elimination. Of note, while multiple mechanisms downregulate classical HLA surface expression in HPV- positive cells, modulation of non-classical HLA molecules (E, G, and F) has not been described. Of particular interest is HLA-E, which has been shown to regulate both NK and

CD8+ T cell activity.

HLA-E can deactivate or activate NK cells through interactions with NKG2A and

NKG2D receptors, respectively (195–198). Additional studies revealed that, not only can

HLA-E interact with NK cells, but HLA-E acts as a ligand for both CD8+ T and natural killer T

(NKT) cells, conferring both adaptive (i.e. T cells) and innate (i.e. NK cells) immune functions to HLA-E (199–201). HLA-E regulation of NK and T cells is complex and not completely understood. HLA-E binds specific viral and bacterial antigens for presentation to

NK and CD8+ T cells. This may either protect the cell from cytotoxicity, or activate a cytolytic

26 response (202–208). Therefore, expression of HLA-E poses a challenge to HPV with its potential to present HPV-specific or stress related antigens to NK or CD8+ T cells, leading to lysis of the infected cell. Here, we show that high-risk HPV E7-mediated methylation of a distal CpG island (CGI) 5’ of the HLA-E ORF can downregulate HLA-E expression in keratinocytes. This finding presents a potentially new mechanism of immune escape accomplished by HPV oncogene expression.

Chemokines and Inflammation

Inflammation and HPV-Associated Cancer Progression

Chronic inflammation and a risk for developing cancer are closely linked. In HPV- positive cervical cancer progression, increased inflammation was not observed in CIN1 or

CIN2 lesions, while CIN3 lesions positive for high-risk HPVs showed a modest increase in inflammation (209). While relatively mild, chronic inflammation in CIN lesions is hypothesized to be a risk factor for cancer development, as is common in other cancers

(210). In cancers of the oral cavity, alveolar bone loss (an indication of a strong immune response) was observed to be slightly increased in HPV-positive HNSCC compared to HPV- negative (211). Of note, the IL-6 and IL-8 inflammatory cytokines secretion was significantly higher in CIN lesions compared to normal cervical tissue; however, an increase in immune cell infiltrates into the lesion was not observed (212). IL-8 is a potent inducer of angiogenesis and therefore important for cancer development. IL-6 is secreted by T cells and macrophages to induce inflammation. The increase of IL-6 and -8 expression in CIN lesions suggests that while the immune system may have detected disease progression, it is unable to respond appropriately. IL-6 and IL-8 expression has also been shown to be induced in keratinocytes, driven by E6 and E7 expression (213) and IL-6 can in fact promote proliferation of HPV-positive cells through TGF-α. Indeed, HPV is known to dysregulate myriad additional immune signaling networks including downregulation of interferon, and upregulation of IL-10 and TGF-β, which together create a local immunosuppressive

27 environment (214). In contrast, HPV-positive cells in vitro-treated with nonsteroidal anti- inflammatory drugs arrested growth and underwent apoptosis (215). Taken together, these results indicate that HPV requires a certain low level of inflammatory signaling in order to persist but also inhibits immune cell infiltration into infected tissue, the mechanisms of which remain elusive. HPV has therefore evolved complex mechanisms to evade immune clearance.

To more thoroughly understand the deregulation of inflammatory signaling molecules in HPV-positive lesions, den Boon et al. assessed transcriptional changes in 128 cervical tissue samples from normal, low grade, high grade, and cervical cancer (111). Expression of estrogen receptor α (ERα) was found to incrementally decrease in epithelial cells throughout cancer progression; however, expression of ERα in the surrounding stromal tissue is considerably increased (111). ERα is closely linked to immune signaling, specifically through the NF-kB and AP-1 pathways, and dysregulation of which may present a strategy employed by HPV to avoid an immune response (216–218). Previous studies have shown that blocking ERα signaling clears HPV-associated cancerous lesions completely, concluding that ERα signaling is required for cervical cancer progression (219, 220). Further studies by den Boon et al. identified transcriptional changes in immunomodulating genes that correlated with epidermal ERα downregulation (using Spearman’s rank correlation), and found corresponding decreases in CXCR2, IL1R2, CXCL14, and CXCL12, and increases in

IL-24, ILF2/3, and IL-8 (111). Of particular interest is the CXCL14 chemokine, which has been shown to be dysregulated in various cancers.

CXCL14 is a Potential Tumor Suppressor

Chemokines are small, secreted signaling proteins that were first discovered as chemotactic molecules involved in recruiting nearby immune cells. While some proinflammatory chemokines are involved in immune cell infiltration and migration in response to various infections, other homeostatic chemokines play important roles for

28 normal tissue maintenance or development (221). The potential tumor suppressor chemokine CXCL14 is constitutively expressed in many epithelial organs in humans and mice, and in neural tissue during development (222–225). Expression of CXCL14 in embryogenesis and development is important for trophoblast and NK cell regulation, respectively (226, 227). In vitro analysis showed that CXCL14 is a modest chemoattractant for NK cells (228). CXCL14 is also thought to act as a chemoattractant for immature DCs; however, dispensing with CXCL14 expression in mice was shown to not affect DC activity

(229, 230). In tumors, decreased CXCL14 expression decreases DC infiltration into the tumors, suggesting a role for CXCL14 in tumor immunology (231). Due to the wide range of tissues in which CXCL14 is expressed, and its role in development and immune cell signaling, CXCL14 is an important homeostatic regulator and potential tumor suppressor; however, the precise mechanisms defining its role as a homeostatic factor remain to be elucidated.

Further investigation into the deregulation of chemokine signaling may provide insight into additional immune evasion strategies put forth by the virus. Here, we show that

CXCL14 is precipitously downregulated in HPV-positive cancers by promoter hypermethylation in an E7 dependent manner, which has profound effects on immune cell infiltration in the tumor environment.

Dissertation Goals and Objectives

HPVs have developed complex immune evasion strategies; however, the extent of these tactics remains to be explored. One strategy the virus could potentially use to avoid immune detection for decades during persistence and cancer development is to alter expression of cytokines in the infected epithelium. To better understand the roles of host immunity in HPV-associated cancer progression, we analyzed expression levels of all cytokines using our global gene expression datasets of CxCa progression (232) and HPV- positive and -negative HNSCCs (233). Notably expression of several chemokines was

29 progressively changed throughout CxCa progression, specifically in HPV-positive HNSCCs compared to HPV-negative HNSCCs. Chemokines are small chemotactic cytokines that cause the directed migration of immune cells (221). Deregulated chemokine networks in the tumor microenvironment (TME) alter immune cell infiltration, angiogenesis, tumor cell growth, survival, and migration, which lead to cancer progression (234). Recent laboratory studies and clinical trials have shown that restoring antitumor immune responses might be a promising therapeutic strategy to treat several cancers including HNSCCs (235–238). While initial studies have begun to explore a relationship between HPV infection and chemokine regulation, little is yet known about chemokine expression patterns altered by HPV during

HPV-associated cancer progression. Here we show that, while many proinflammatory chemokines are significantly increased, CXCL14 is significantly decreased in HPV-positive

HNSCC and CxCa patient tissues during cancer progression. Thus, the first major aim of this dissertation is to further explore the mechanism of CXCL14 downregulation and the consequences to tumor growth and immune cell surveillance.

In conclusion of the first aim, we show that CXCL14 expression is dramatically repressed by E7-dependent DNA methylation. Thus, the second major aim of this dissertation is to elucidate the extent of DNA methylation-induced changes in host gene expression, as directed by HPV E7. Gene expression dysregulation is a well known strategy that viruses frequently employ to evade the host immune response (239–245). HPV-positive

HNC and CxCa progression has also shown distinct changes in host DNA methylation that alters host gene expression (129, 164, 165, 246). Here, we performed parallel global gene expression and methylome analyses to identify key host factors and pathways altered by

HPV-mediated DNA methylation in human keratinocytes. Additionally, we assessed genomic regions that are likely affected by HPV-directed changes in methylation. Here, we found that most class I major histocompatibility complex (MHC-I) molecules are downregulated in an

E7-dependent manner. Further, non-classical HLA-E, which regulates NK and CD8+ T cells,

30 is significantly downregulated by E7-mediated hypermethylation in a distal regulatory CpG island. These results suggest that HPV E7-mediated DNA methylation induces host immune evasion by downregulating HLA-E expression.

CHAPTER II

MATERIALS AND METHODS

Cell and Tissue Culture

Human keratinocytes NIKS (247), NIKS-16, NIKS-18 (248), and NIKS-16ΔE7 (249) cells were co-cultured with 3T3 feeder cells in E-complete medium as previously described

(250). Briefly, NIH 3T3 mouse fibroblasts were incubated in 4 μg/mL mitomycin C for 3-6 hours then seeded at approximately 1:4 by surface area onto dish containing NIKS cells.

NIKSE7(6), (11), (16), and (18) were generated through lentiviral transduction of the appropriate E7 expression cassette and stable cell lines were generated via puromycin selection. NIKS and NIKS derivatives were generated in 1999, and maintained under passage 50, as described (247, 248).

HeLa cells were obtained from the American Type Culture Collection (ATCC) and cultured in DMEM supplemented with 10% fetal bovine serum (FBS) according to the manufacturer’s recommendations. Demethylation was achieved through treatment with 5μM

5-aza-2deoxycytidine (DAC) for five days, refreshing medium daily.

W12E (derived from a low-grade precancerous cervical lesion with episomal HPV16)

(251), and W12G (low-grade with integrated HPV16) (252) cells were established in the

Paul Lambert laboratory and grown with 3T3 feeder mouse fibroblasts in E-Complete medium as described (233). Transformed W12GPXY cells derived from W12G cells were obtained from Dr. Sheila Graham (University of Glasgow) in 2011. W12GPXY cells were grown without 3T3 feeders in DMEM supplemented with 10% FBS.

CaSki cells were obtained from ATCC in 1987 (253) and validated by HPV early gene expression. Cells were cultures in DMEM supplemented with 10% FBS. Demethylation was achieved through treatment with 10μM DAC for six days, refreshing medium daily.

Mouse oropharyngeal epithelial (MOE) cell lines, MOE/shPTPN13 (transformed with

Ras and shRNA against Ptpn13) and MOE/E6E7 (transformed with Ras and HPV16 E6/E7)

32 were generated by Dr. John Lee in 2009 (254), and validated by assessing cytokeratin expression, the presence of the E6 and E7 expression vectors which confer resistance to puromycin, and activation of the MAPK pathway, a hallmark of E6 expression. All MOE variants were cultured in E-Complete medium (250) supplemented with 0.2 ng/mL tri-iodo- thyronine and 5 μg/mL transferrin (Fisher Scientific). Cxcl14 re-expressing MOE/E6E7 clones were established by limiting dilution. Briefly, murine Cxcl14 was cloned into the pCDH-CMV-MCS-EF1-Puro using primers located in Appendix C. MOE/E6E7 cells were transduced and selected with puromycin (40 μg / mL) for 4 days. Cells were trypsinized thoroughly and plated at a ratio of 0.3 cells per well of a 96-well plate. 2-3 weeks after plating, select cloneses and transfer to a 6-well plate, then transfer again to a T-25 flask when approximately 75% confluent. The confluent T-25 flask is considered passage 1.

Cxcl14 expression from each of the clones was determined by RT-qPCR using primers listed in Appendix A.

Mouse spleens were harvested from C57BL/6 mice (The Jackson Laboratory) and tissue was disrupted by passage through a 100-μm strainer (Becton Dickinson / Falcon).

Splenocytes were incubated in red blood cell lysis buffer (Sigma-Aldrich) according to the manufacturer’s instruction. Mononuclear cells were pelleted and cultured in RPMI 1640 medium containing 10% FBS and 10 ng/mL mouse recombinant IL-2 (mrIL-2, eBioscience) for 3 hours before plating for transwell analysis.

Conventional PCR and Quantitative Reverse Transcription-PCR (RT-qPCR)

Genomic DNA (gDNA) was isolated from cells using DNeasy Blood and Tissue Kit

(Qiagen). E7 variant expression was detected by PCR using GoTaq Green Master Mix

(Promega) followed by separation on agarose gels. Primers used for detection are listed in

Appendix A.

Total RNA for RT-qPCR was extracted using High Pure RNA Isolation Kit (Roche

Applied Science) and in column DNA digest was performed using the supplied RNAse

33 according to the manufacturer’s instructions. Complementary DNA (cDNA) was reverse transcribed using the Transcriptor First Strand cDNA Synthesis Kit (Roche Applied Science) and oligo-d(T)16 (supplied) from total RNA. RT-qPCR PCR was performed using the Bio-Rad

CFT Connect Real-time System and FastStart Universal SYBR Green Master (Rox) (Roche

Applied Science). Reaction mixtures consisted of 10 μL SYBR Green Master, 1.4 μL of each forward and reverse primers (100 μM), and diluted template cDNA brought to a final volume of 20 μL. Data were normalized by the level of β-actin (ACTB) for human targets, and Gapdh for mouse targets. Primers used for detection are listed in Appendix A.

Bisulfite Modification and Assessment of Methylated DNA

Genomic DNA (gDNA) was extracted from keratinocytes (Qiagen) and 500 ng or gDNA was bisulfite converted using EZ DNA Methylation Kit (Zymo Research) according to the manufacturer’s instruction. All MSP reactions were performed using Platinum Taq

(Invitrogen) and supplied reagents as follows: reaction mixtures consisted of 1 μL 10X

Buffer, 0.2 μL dNTPs, 0.3 μL MgCl2, 0.5 μL each forward and reverse primers (100 μM), 0.1

μL Platinum Taq Polymerase, diluted template bisulfite converted gDNA, brought to 10 μL with water.

MSP specificity and annealing temperatures were determined using methylated or unmethylated control gDNA. Briefly, control DNA was generated by incubating a mixture of gDNA from W12E, W12G, and W12GPXY cells with McrBC or MSssI methyltransferase followed by BstUI digestion (New England Biolabs). MSP was performed using an annealing temperature gradient for each primer set (specific for methylated or unmethylated DNA) using either McrBC or MSssI treated and bisulfite converted gDNA. The annealing temperature offering unique specificity to either methylated or unmethylated bisulfite converted gDNA was used for subsequent experiments. CXCL14 primers were specific only for methylated DNA and compared to non-specific control primer amplification. All primers for MSP are listed in Appendix B.

In vitro promoter methylation to assess the effect on transcription was performed using the MSssI methyltransferase (New England Biolabs) and methylation efficiency was validated by McrBC and BstUI digest (New England Biolabs) according to the manufacturer’s instruction. Constructs were transiently transfected into HEK 293FT cells using polyethyleneimine (PEI) using a 3:1 PEI to DNA ratio. Transcription was assessed by quantifying relative luciferase expression using the Dual-Luciferase Reporter Assay System

(Promega) according to the manufacturer’s instruction.

Bisulfite sequencing products were cloned into pGEM-T easy vector (Promega) according the manufacturer’s instruction. Twenty clones for each cell type, W12E, W12G, and W12GPXY were sequenced by Sanger sequencing. Sequencing was performed using the following forward and reverse primers (5’ to γ’): GGTTGGGAAGGTTTTTTTTT and

ACCCAACTCTACTCRACTTTCT.

Quantitative MSP (qMSP) was performed for CXCL14 methylation with bisulfite- converted gDNA using SYBR Green Master (Rox) (Roche). Relative DNA methylation was calculated using the 2-ΔΔCt equation using methylated amplification as the sample, control amplification as the control, and β-Actin amplification as the reference.

A detailed protocol for MSP with examples can be found in Appendix F.

Vectors and Plasmids

Luciferase Reporter Vectors

In vitro methylation analysis was achieved using the pCpGL and pCpGL-Basic luciferase reporter vectors (provided by Dr. Michael Rehli, University of Regensburg,

Germany) (255). Promoter elements were amplified from NIKS gDNA by PCR using GoTaq

Master Mix (Promega). See Appendix C for primer sequences. The CRE 4X repeat element was cloned into pCpGL as previously described using the following hybridized primers designed with BamHI and NcoI sticky ends (underlined): 5’-GATCCAGCCTGACGTCAGAG

AGCCTGACGTCAGAGAGCCTGACGTCAGAGAGCCTGACGTCAGAGAC-γ’, and 5’-

CATGGTCTCTGACGTCAGGCTCTCTGACGTCAGGCTCTCTGACGTCAGGCTCTCTGAC

GTCAGGCTG-γ’ (256). A new multiple cloning site was introduced to the pCpGL-Basic vector in place of the CMV enhancer by PCR-mediated mutagenesis using the following primers: 5’-GGGGGGATCCGTTCGAAGTCTAGAGGCTAGCCTGCAGGAATTCCTTTTAAT

CTGCTG-γ’ and 5’-GGGGGGATCCGCGATCGACTAGTGGAGAAGAGCATGCTTGAGG-γ’.

The HLA-E CGI element was then directionally cloned into this vector using the primers in

Appendix C. A detailed protocol for dual luciferase assays can be found in Appendix G.

Lentiviral Vectors

For generation of E7-expressing NIKS cells, E7 variants were PCR amplified from

HPV6, 11, 16, and 18 genomic DNA using the primers in Appendix C. The XbaI and BamHI restriction sites (underlined) were used to directionally clone the amplified sequence into the pCDH-CMV-MCS-EF1-Puro lentiviral expression vector (CD510B-1, System Biosciences).

Generation of lentivirus was achieved by transfecting 5.3 μg VSVG vector, 8 μg Δ8.2 vector, and 10 μg of the proviral plasmid (generated above) into 293 cells using a 3:1 ratio PEI to

DNA. Lentiviruses were harvested 48 hours post transfection.

For generation of CXCL14 and Cxcl14 expression vectors, ORFs were PCR amplified from cDNA generated from HFK and MOE cells, respectively, using primers found in Appendix C. The EcoRI and BamHI restriction sites (underlined) were used to directionally clone the amplified sequence into the pCDH-CMV-MCS-EF1-Puro lentiviral expression vector (CD510B-1, System Biosciences). Lentiviriuses were generated as above.

Enzyme-Linked Immunosorbent Assay (ELISA)

All ELISA reagents were purchased from R&D Systems. CXCL14 concentration was determined using DuoSet ELISA Development System for Human BRAK/CXCL14 according to the manufacturer’s specifications. High-binding 96 well ELISA plates were coated with anti-CXCL14 capture antibody and incubated overnight at room temperature. Wells were washed three times with PBS containing 0.1% Tween-20 and blocked with reagent diluents

36 containing bovine serum albumin. For sample collection, cell culture supernatant was collected after 48 hours incubation with NIKS, NIKS-16, W12E, W12G, and W12GPXY cells.

Culture supernatant was incubated in ELISA wells for one hour at room temperature. Wells were washed three times with PBS containing 0.1% Tween-20 and incubated with a biotinylated detection anti-CXCL14 antibody, followed by incubation with streptavidin-HRP, and HRP substrate reagent. Absorbance was determined using the Bio-Tek Synergy HT plate reader and analyzed using Bio-Tek KC4 software. CXCL14 concentration was determined from a standard curve of recombinant CXCL14 using the 4-parameter logistic regression model from ElisaAnalysis.com (LTG Ventures Pty Ltd). Statistical significance was assessed using the Student’s t-test.

Cell Migration Assays

Scratch Assay

The in vitro scratch assay mimics wound healing migratory behavior of epithelial cells across an artificially generated gap or “wound” (257). 6-well tissue culture plates were incubated with a 0.01% poly-L Lysine solution in PBS at room temperature for 4 hours. The plate was washed with water and allowed to dry. CaSki and MOE/E6E7 cells were plated at

1.5 X 105 cells / cm2 on the treated plates. After 12 hrs, confluent monolayers were

“scratched” using a p10 pipette tip. Cells were washed with PBS to remove debris and floating cells and supplied with fresh medium. Gap closure was monitored every 4 hours.

The width of the gap was measured using NIH Image J 1.48 (http://imagej.nih.gov/ij). At each time point, the same relative locations along the wound were monitored, guided by a grid on the plate. About γ0 different locations were measured per sample. Student’s t-test was used to determine significance.

Transwell Migration

The transwell migration assay was performed using 1 X 105 cells per well of an 8 μm

24-well transwell permeable support, and incubated overnight using FBS as a

37 chemoattractant. Percent migrated cells were calculated and Student’s t-test was used to determine statistical significance.

Spleens from C57BL/6 mice injected with MOE/E6E7 cells were harvested at 21 days post injection and mechanically disrupted through a 100-μm filter. Red blood cells

(RBC) were cleared by RBC lysis buffer (Sigma) and remaining splenocytes were rested at

37°C in RPMI 1640 medium containing 10% FBS and 10 ng/mL of mouse recombinant IL-2

(mrIL-2, eBioscience) for 3 hours. Conditioned media (CM) from the culture of MOE cells re- expressing Cxcl14 (clones 8 and 16) or containing vector were added into the bottom chamber of a transwell (γ μm pore size; Costar). Isolated splenocytes (β × 106 cells/mL) were resuspended in RPMI 1640 medium supplemented with mrIL-2 and added to the upper chamber of the transwell. Splenocytes in RPMI 1640 medium without mrIL-2 was used as a negative control. After 12 hour incubation at 37°C, splenocytes were harvested from the top and bottom chambers, stained with trypan blue, and counted using a hemocytometer. Cell populations were analyzed using flow cytometry as described below in Flow Cytometry.

Total cell populations were determined by applying the cell counts to the cell population percentages. Migration index for each cell type was calculated by: Percent cell migration = migrated cell number (bottom chamber) / total cell number (upper chamber + bottom chamber).

Mice and Treatment

Four to six week old, 20-25 g male C57BL/6J wild type or Rag1-/- mice (The Jackson

Laboratory) were maintained in accordance with the USDA guidelines. Tumors were initiated by injection of engineered MOE/E6E7 cells (1 x 105) subcutaneously into the rear right flank of mice (n = 10 per group). Tumor growth was measured weekly using previously established techniques (258). Tumor volume was calculated using the equation: volume =

(width)2 × depth. Animals were euthanized humanely when tumor size was greater than 1.5 cm in any dimension. Conversely, mice were considered tumor free when no measurable

38 tumor was detected for a period of 11 weeks. Survival graphs were calculated by standardizing for a tumor volume of 2,500 mm3.

Flow Cytometry

Antibodies

For mouse tissues, the following anti-mouse antibodies were purchased from eBioScience and used according to manufacture specifications: MHC-II (FITC conjugate, clone M5/114.15.2), CD11c (PE conjugate, clone N418), CD25 (PerCPCy5 conjugate, clone

PC61.5), CD4 (eF450 conjugate, clone RM4-5), F4/80 (APC conjugate, clone BM8), Gr1

(AF700 conjugate, RB6-8C5), CD16/CD32 (clone 93), and CD11b (ef450 conjugate, clone

M1/70). Anti-mouse CD45 (PerCP conjugate, clone 30-F11) and NKp46 (PECy7 conjugate, clone PC61.5) were purchased from Biolegend.

Anti-human antibodies were purchased from eBioscience for flow Cytometry on NIKS cells: HLA-E (PE conjugate, 3D12) and HLA-BC (APC conjugate, B1.23.2).

Sample Preparation and Flow Analysis

For each experimental group, spleen, TDLN, and tumor tissue were harvested from wild type and Rag1-/- mice at 21 and 23 days post injection, respectively. Tumor samples were incubated in collagenase (Sigma-Aldrich) for 30 min at 37°C to remove extracellular matrix and fibrous tissues. All tissue samples were passed through a 100 μm cell strainer

(Corning Life Sciences) and spleens were incubated in Red Blood Cell Lysing Buffer Hybri-

Max (Sigma-Aldrich) for 3 min at room temperature. The isolated cells were incubated with a panel of antibodies conjugated with unique fluorophores for 1 hr at room temperature and washed with PBS. Samples were passed through a γ5 μm cell strainer (Corning Life

Sciences) immediately before analysis on an LSRII flow cytometer (Becton Dickinson) using

FACSDiva collection software. All cells were assessed for viability by staining with

LIVE/DEAD® Fixable Aqua Dead Cell Stain (Life Technologies).

For analysis of HLA expression in NIKS cells, approximately 1 x 106 cells were trypsinized and pelleted. Cells were fixed in 4% paraformaldehyde, washed three times with

PBS (10 minutes each wash) followed by permeabilization in 0.1% saponin and three additional 10 minute washes in PBS. Cells were incubated in anti-human antibodies against

HLA-B/C or HLA-E overnight, followed by three 10 minutes washes in PBS. Flow cytometric analysis was performed on an LSRII flow cytometer (Becton Dickinson) and data were processed using FlowJo software.

Array Preparation and Analysis

Genome-wide Expression and DNA Methylation Arrays

For gene expression analysis, total RNA was extracted from NIKS, NIKS-16, NIKS-

18, and NIKS-16ΔE7 cells using RNeasy kit (Qiagen) and hybridized to Affymetrix Human

Genome U133 Plus 2.0 Array chip as previously described (110). For methylome analysis, genomic DNA (gDNA) was isolated from NIKS, NIKS-16, NIKS-18, and NIKS-16ΔE7 cells using the DNeasy kit (Qiagen). Bisulfite-converted gDNA was prepared using the EZ DNA

Methylation Kit (Zymo Research) and assessed using Illumina Infinium

HumanMethylation450 BeadChip Kits according to the manufacturer’s protocol.

Data Processing and Statistical Analysis

Data analysis was performed using R version 3.2.2 (259). Gene expression data quality was validated using QC plots observing RNA degradation and log density distribution for all probes for each sample. The data were filtered based on mas5 present/absent calls, where transcripts were discarded if they had at least one absent call. After filtering, 27,146 transcripts remained. Transcript expression was normalized using RMA (260). Two-sample

Student’s t-test was used to measure differential expression for each comparison where correction for multiple comparisons was performed using Benjamini-Hochberg False

Discovery Rate (FDR) within each comparison. Differentially expressed genes were identified based on absolute minimum change in expression of 30% and FDR < 0.05 (261).

Expression was visualized via heatmaps from normalized signal intensity data generated using Microsoft Excel (262).

Methylation detection p-values were calculated for each probe on the array where the methylated and unmethylated signals were compared to background signal using negative control probe intensities. For comparison with gene expression data, CpG probe annotation including nearby genes was determined using the minfi package. Non-significant detection p-values greater than 0.05 were filtered out. After filtering, 485,531 probes remained. Subset-quantile within array normalization (SWAN) provided by the minfi package was performed for each sample (263, 264). Differentially methylated positions (DMP) were determined by independent two-sample Student’s t-test on the M values for each comparison where β values were reported for significant CpG sites. To correct for multiple comparisons, p-values were adjusted using FDR within each comparison (261) and CpG probes with FDR < 0.05 were determined significant. Identification of differentially methylated regions (DMR) was performed using the bumphunter method where CpG probes within 500 bp were clustered (265). Significance was determined by permutation testing where 1,000 permutations were performed and a 99% quantile of the estimated area of the permuted clusters is used as a cutoff. Regions meeting the cutoff determined by the algorithm and those having a permutation p-value < 0.05 were considered significant.

The RNA-seq RSEM (RNA-seq by expectation maximization) counts of HLA-A, -B, -

C, -E, -F, and –G were obtained from the TCGA data through cBioPortal (cbioportal.org):

HPV-negative HNC, n = 243; HPV-positive HNC, n = 36 (129); CxCa, n = 309 (NCI, TCGA,

Provisional). P-values were determined by Mann-Whitney test.

Bioinformatics

100 bp of genomic sequence on both flanks of each DMR were fetched using the

Galaxy platform (usegalaxy.org) (266–268) and enriched TF binding motifs were identified using MEME Suite (269). Flanking regions were limited to 200 bp total due to MEME Suite

41 processing constraints. Relative TF binding site enrichment was performed using the

Analysis of Motif Enrichment (AME) tool with shuffled input sequences as control DNA. The

UCSC Genome Browser (hg19) was used to assess DNase I hypersensitivity (from

ENCODE V3 using 125 cell types), H3K27Ac (ENCODE using 7 cell types), and transcribed human RNAs (GenBank) near the HLA-E CGI. miRNA Target Prediction and Functional

Study Database (miRDB, mirdb.org) analysis mining was performed using default settings.

Pathway analysis was performed using Reactome (reactome.org). Submitted gene lists were restricted by FDR < 0.05 and magnitude 30% change in expression using NIKS expression as baseline.

Immunoblotting

Approximately 1 x 106 NIKS cells were trypsinized, pelleted and lysed in PBS supplemented with 1% SDS and cOmplete, EDTA-Free Protease Inhibitor Cocktail (Roche

Diagnostics). Lysates were passed through Qiashredder columns to shear DNA and reduce viscosity (Qiagen). Protein concentration was measured using Pierce BCA Protein Assay Kit

(Thermo-Fisher Scientific). 20 µg of whole cell lysate was separated by electrophoresis by

12% PAGE on Tris-glycine gels and transferred to 0.45 μm pore polyvinylidene difluoride

(PVDF) membranes (Millipore). Membranes were blocked overnight with 5% nonfat milk buffered in TBS-T, followed by incubation in the appropriate antibodies. Mouse anti-HPV-16

E7 (1:50, Santa Cruz Biotechnology Inc., clone ED17), mouse anti-HLA-E (1:500, AbD

Serotec, clone MEM-E/02), and mouse-anti β-actin (1:100,000, Cell Signaling Technology, clone 8H10D10) were used for protein detection. Donkey anti-mouse IgG conjugated to horseradish peroxidase (Jackson ImmunoResearch) was used for secondary detection with

ECL Western Blotting Substrate (Pierce).

Statistical Analysis

Student’s t test and one-way analysis of variance (ANOVA) were used to calculate significance for comparison of two matched groups and three or more unmatched groups,

42 respectively. The correlation coefficient (R2) was determined by linear regression using

Prism 6 (GraphPad). Results were considered statistically significant at a p-value of less than 0.05. Distributions of time to event outcomes (e.g. survival time) was summarized with

Kaplan-Meier curves, compared across groups using the log-rank test with α = 0.01. Tukey’s method for identifying outliers was used in box-and-whisker plots and outliers are identified.

Generation of HPV16 Reporter Virions

Production of HPV16 reporter virions (HPV16-LucF) was performed as described previously (270). Briefly, HEK 293 FT cells were co-transfected with p16SheLL expressing the HPV16 capsid proteins L1 and L2, and the pLucF reporter plasmid (271). 48 hours post- transfection, cells were lysed in PBS containing 0.25% Brij58 (Sigma-Aldrich), 9.5 mM

MgCl2, followed by 0.3% benzonase (Sigma-Aldrich) and 2 U/100μL Plasmid-Safe (epibio).

Virions were isolated by ultracentrifugation over an OptiPrep (Sigma-Aldrich) gradient.

Reagent Acknowledgements

I would like to thank Sheila Graham (University of Glasgow, United Kingdom) for kindly providing the W12GPXY cells. I also thank Michael Rehli (University of Regensburg,

Germany) for his generous gift of the CpG-free pCpGL and pCpGL-Basic plasmids. Thank you John Lee (Sanford Health, USA), for kindly gifting us the MOE cells and MOE variants.

Thank you also to Cody Warren (University of Colorado, USA) for generating the stable

NIKS cells expressing HPV E7 proteins. The p16SheLL and pLucF plasmids were kind gifts provided by John Schiller (National Cancer Institute). A special thanks to Mallory Meyers for her work generating the pCpGL variants for promoter methylation analysis.

CHAPTER III

SUPPRESSION OF ANTITUMOR IMMUNE RESPONSES BY HUMAN PAPILLOMAVIRUS

THROUGH EPIGENETIC DOWNREGULATION OF CXCL141

Importance

Human papillomaviruses (HPVs) are causally associated with over 5% of all cancers.

During decades of cancer progression, HPV persists, evading host surveillance. However, little is known about the immune evasion mechanisms driven by HPV. Here, we report that the chemokine CXCL14 is significantly downregulated in HPV-positive head/neck and cervical cancers. Using patient tissues and cultured keratinocytes, we found that CXCL14 downregulation is linked to CXCL14 promoter hypermethylation induced by the HPV oncoprotein E7. Restoration of CXCL14 expression in HPV-positive cancer cells clears tumors in immunocompetent syngeneic mice, but not in immunodeficient mice. Mice with

Cxcl14 expression show dramatically increased natural killer and T cells in the tumor draining lymph nodes. These results suggest that epigenetic downregulation of CXCL14 by

HPV plays an important role to suppress antitumor immune responses. Our findings may offer novel insights to develop preventive and therapeutic tools for restoring antitumor immune responses in HPV-infected individuals.

Introduction

Human papillomaviruses (HPVs) are causally associated with multiple human cancers, including cervical cancer (CxCa) and head and neck cancer (HNC), and result in about half a million deaths worldwide each year (272). Persistent infection of HPV is required for HPV-associated cancer development and therefore HPV must evade host immune surveillance (273). To evade host immune surveillance, HPV creates a local immune suppressive environment by inducing chemokine expression and diminishing the

1 This chapter was published with permission from Cicchini et al. 2016. Suppression of Antitumor Immune Responses by Human Papillomavirus through Epigenetic Downregulation of CXCL14. mBio, http://mbio.asm.org/content/7/3/e00270-16.

44 cytotoxic T cell response (273, 274). However, little is known about the mechanisms of disease progression driven by HPV-induced immune suppression.

To better understand the roles of host immunity in HPV-associated cancer progression, we analyzed expression levels of all known chemokines and chemokine receptors using our global gene expression datasets of CxCa progression (232) and HPV- positive and -negative HNCs (233). Deregulated chemokine networks in the tumor microenvironment (TME) alter immune cell infiltration, angiogenesis, tumor cell growth, survival, and migration, leading to cancer progression (234). Recent laboratory studies and clinical trials have shown that restoring antitumor immune responses may be a promising therapeutic strategy to treat several cancers including HNCs (235, 275, 276). While initial studies have begun to explore relations between HPV infection and chemokine regulation, little is yet known about chemokine expression patterns altered by HPV during cancer progression. Here we show that, while expression of many proinflammatory chemokines is increased, CXCL14 expression is significantly decreased in HPV-associated cancer progression.

CXCL14 is a chemokine distantly related to other CXC chemokines, showing 30% identity with CXCL2 and CXCL3 (277). CXCL14 functions as a potent angiogenesis inhibitor and a chemotactic factor for dendritic cells (DCs) and natural killer (NK) cells (278, 279).

While normal human epithelial cells constitutively express CXCL14, its expression is frequently reduced in cervical, prostate, and oral cancers (280–282). Restoration of Cxcl14 expression recruits DCs into tumors in vivo and in vitro (282, 283) and induces tumor necrosis (284). Importantly, Cxcl14 expression in HNC cells suppresses tumor growth from xenografts in athymic nude and SCID mice (285, 286). In addition, the rate of colorectal tumor formation and metastasis was significantly lower in Cxcl14 transgenic mice compared to isogenic wild type mice (287). Previous studies have shown that CXCL14 inhibits signaling of proinflammatory chemokines IL-8 (278) and CXCL12 (288), which are known to

45 promote cancer development and metastasis. Thus, CXCL14 has been suggested as a potential tumor suppressor having anti-inflammatory functions. CXCL14 expression is epigenetically regulated by promoter hypermethylation in colorectal cancer cells (283). In the current investigation we show that the CXCL14 promoter is highly methylated and its expression is downregulated in HPV-positive tissues and cells in an E7-dependent manner.

Importantly, restoration of murine Cxcl14 expression in HPV-positive mouse oropharyngeal epithelial (MOE) cells increases NK, CD4+ T, and CD8+ T cell infiltration into the tumor- draining lymph nodes (TDLN), and results in significant clearance of implanted HPV-positive

HNC cells in immunocompetent syngeneic mice.

Results

Proinflammatory Chemokines are Upregulated during CxCa Progression

To understand the mechanisms by which HPV deregulates host immune responses in the TME, we analyzed gene expression changes of known chemokines and their receptors in tissue epithelium during CxCa progression using our global gene expression data from human cervical tissue specimens of normal, cervical intraepithelial neoplasia

(CIN) 1/2, CIN3, or tumor tissues (GEO accession #GSE63514) (232). The results showed that fourteen chemokines and chemokine receptors increased at least 3-fold in during cancer progression (Appendix D-8). Expression of IL-8, CXCL9, CXCL11, CCL3, and

CCL19 mRNAs was progressively increased throughout disease progression (Appendix D-

1A). In contrast, expression of CXCL1, CXCL2, CXCL5, CXCL6, and CCL20 mRNA was significantly upregulated during the early transition from normal to CIN1/2 (Appendix D-1B), while CXCL13 and CCL8 mRNA expression significantly increased only in the later transition to invasive tumors (Appendix D-1C). Among chemokine receptors, CXCR2 mRNA expression was decreased by 12-fold and CXCR4 mRNA expression was upregulated nearly 7-fold throughout cancer progression (Appendix D-1D). To identify HPV-specific chemokine deregulation, we analyzed our previously published gene expression data of

HPV-positive and -negative HNCs (GEO accession # GSE6791) (233). This analysis revealed that expression of CXCL9, CXCL10, CXCL13, and CCL19 as well as CXCR4 mRNA was significantly upregulated in HPV-positive HNCs compared to HPV-negative

HNCs (Appendix D-2A-E), suggesting that HPV infection specifically modulates chemokine expression. Unlike increased expression during cervical cancer progression, the expression level of IL-8 mRNA was lower in HPV-positive HNCs by 2-fold compared to HPV-negative

HNCs (Appendix D-2F). Although HPV-positive cancers exhibit lower levels of IL-8 expression compared to HPV-negative cancers, our previous study showed that IL-8 expression was significantly increased in all HNCs compared to normal tissues (233). These results indicate that HPV-negative HNCs robustly upregulate IL-8 and CXCL1 expression more than HPV-positive HNCs by other mechanisms. To explore these changes of chemokine expression in vitro, we analyzed chemokine expression in cervical keratinocyte lines using reverse transcriptase quantitative PCR (RT-qPCR). We used W12E (derived from a low-grade precancerous cervical lesion with episomal HPV16), W12G (low-grade with integrated HPV16), and W12GPXY (transformed) cells, which sequentially mimic CxCa progression (289). As expected, all W12 cell lines express high levels of the HPV16 early gene transcript (Appendix D-2G). Expression levels of proinflammatory chemokines IL-8,

CXCL1, CXCL2, CXCL10, and CXCL11 were significantly increased in W12G and

W12GPXY cells compared to a normal immortalized keratinocyte line (NIKS) (Appendix D-

2H-L). These results from tissue specimens and cultured keratinocytes suggest that several proinflammatory chemokines, which are recognized as major players in cancer development, are upregulated during HPV-associated cancer progression.

CXCL14 Expression is Downregulated in HPV-Associated Cancer Progression

While over a dozen chemokines were highly upregulated, CXCL14 was the only chemokine decreased over 3-fold in CxCa progression (Figure 3-1A). CXCL14 mRNA

HNC A Cervical tissues B 1.2 Keratinocytes C 15 16 p < 0.005 1.0 p < 0.0001 -actin β 0.8 12 12 0.6

8 0.4 9 p < 0.0001 0.2 4 to mRNArelative 0.0 6 mRNA expression level (log2) level expression mRNA

mRNA expression level (log2) level expression mRNA E Y HPV- HPV+ 2 3 2 G al / er IKS 1 2 X m 1 IN c N P HNC HNC r N C n W W1 G o I a 2 N C C 1 W DECXCL14 2.5 1.4 CXCL14 * 1.2 2.0 -actin -actin β β 1.0 1.5 0.8 * * 1.0 0.6 * 0.4 0.5 * 0.2

mRNA relative to to mRNArelative * 0.0 to relative mRNA 0.0 NIKS NIKS-16 NIKS-18 NIKS-31 NIKS-16 NIKS NIKS- NIKS- ΔE7 16E7 18E7 F CXCL14 ELISA GH 600 0.5 Cxcl14 250 HPV16 E7

** 0.4 200 400 0.3 150 * 200 * 0.2 100

CXCL14 (pg/ml) CXCL14 * 0.1 50 * 0 NIKS NIKS-16 W12E W12G W12GPXY 0.0 0 HPV- HPV+ HPV- HPV+ mRNA relative to 1000mGAPDH to mRNArelative mRNA relative to 1000mGAPDH to mRNArelative MOE MOE MOE MOE

Figure 3-1. CXCL14 expression is downregulated during HPV-associated cancer progression. CXCL14 mRNA expression levels were analyzed from global gene expression data sets of (A) 128 cervical tissue samples in different disease stages (normal, n = 24; low- grade lesion, n = 36; high-grade lesion, n = 40; and cancer, n = 28) (111) and (C) 42 HNC (HPV-HNC, n = 26; HPV+HNC, n = 16) (233) tissue samples. Normalized fluorescence intensities (log2) of gene expression from each group are shown in box-and-whisker plots with Tukey's method for outliers (black triangles) noted as distinct data points. P-values were determined by one-way ANOVA analysis (A) or the Student’s t-test (C). Total RNA was extracted from (B) W12 cell lines and (D and E) NIKS keratinocyte lines. The expression levels of CXCL14 were measured by RT-qPCR. (F) Secreted CXCL14 was measured by ELISA using culture supernatant from NIKS, NIKS-16, W12E, W12G and W12GPXY cells. (G & H) Total RNA was extracted from NIKS, NIKS-HPV16E7, NIKS-HPV18E7 and mouse oropharyngeal epithelial (MOE) cell lines, MOE/shPTPN13 (HPV-negative) and MOE/E6E7 (HPV-positive). The expression levels of HPV16 E1^E4 mRNA transcript (G) and murine Cxcl14 mRNA (H) were measured by RT-qPCR. HPV16 E1^E4 and CXCL14 mRNA copy numbers were calculated using serially diluted standard plasmids and normalized by human ß-actin and murine Gapdh mRNA copy numbers. P-values were calculated by the Student’s t-test. *p < 0.0001, **p = 0.0002. Panels A and C were contributed by Dohun Pyeon, PhD, University of Colorado. Panels B and D were contributed by Tao Xu, PhD, St. Jude Hospital, TN.

expression was progressively decreased by about 21-fold from normal to cancer tissue. The downregulation of CXCL14 was consistently observed in the W12 cell culture model (Figure

3-1B). CXCL14 expression levels showed a significant inverse correlation with the expression levels of IL-8 and other proinflammatory chemokines in cervical tissue specimens and cultured keratinocytes (Appendix D-1A, Figure 3-1A, 3-1B, and Appendix

D-2H to D-2L). To determine whether CXCL14 downregulation is unique to HPV-positive cancers, we compared CXCL14 mRNA expression between HPV-positive and HPV- negative HNCs using the datasets from our previous global gene expression study (233).

The results showed that CXCL14 mRNA expression was significantly lower in HPV-positive

HNC compared to HPV-negative HNC (Figure 3-1C). We also confirmed downregulation of

CXCL14 mRNA expression in HPV-positive HNC and CxCa compared to HPV-negative

HNC using the TCGA RNA-seq data (290) (Appendix D-3A). A previous study reported that

CXCL14 expression was significantly decreased in HNCs compared to normal tissue (280).

Taken together, these results suggest that CXCL14 is further downregulated in HPV-positive

HNCs compared to HPV-negative HNCs and normal keratinocytes. To validate these observations using homogeneous keratinocyte culture models, we analyzed CXCL14 mRNA expression in NIKS cell lines with and without high-risk HPV genomes. We found that each high-risk HPV (HPV16, HPV18, or HPV31) was sufficient to inhibit CXCL14 expression

(Figure 3-1D). Of note, CXCL14 expression was not downregulated in NIKS-16∆E7 cells, which contain an E7-deficient HPV16 genome (291) (Figure 3-1D). Furthermore, CXCL14 mRNA expression was modestly but significantly downregulated in NIKS cells expressing only the E7 oncoprotein from HPV16 or HPV18 (Figure 3-1E). To detect secretion of the

CXCL14 protein in cell culture supernatant, we performed an ELISA using culture supernatant from NIKS and W12 cells. NIKS cells secreted a high level of CXCL14 protein, consistent with the previous study showing that normal keratinocytes constitutively express

CXCL14 (280). In contrast, CXCL14 levels secreted by NIKS-16 and W12 cells were significantly decreased, indicating that the CXCL14 mRNA levels in NIKS and W12 cells correlate with CXCL14 secretion in cell culture supernatant (Figure 3-1F). Taken together, these results suggest the HPV oncoprotein E7 is sufficient to suppress CXCL14 expression.

However, long term exposure is required for dramatic repression as seen in the W12GPXY cells and HPV-positive cancers.

Next, using HPV-positive and -negative MOE cells, we assessed the effect of the

HPV16 oncoproteins E6 and E7 on murine Cxcl14 expression. A protein sequence alignment demonstrated 98% homology between human CXCL14 and murine Cxcl14 within the C-X-C chemokine motif (data not shown). Two neutral amino acid substitutions are observed within the C-X-C motif: human CXCL14 I70 and V75, corresponding with murine

Cxcl14 V58 and M63, respectively. We determined expression levels of Cxcl14 mRNA in

MOE cell lines, MOE/shPTPN13 (Ras transformed, HPV-negative) and MOE/E6E7 (Ras transformed, expressing the HPV16 oncogenes E6 and E7) that form tumors in immunocompetent syngeneic C57BL/6 mice (292). Consistently, Cxcl14 expression was also significantly downregulated in MOE/E6E7 cells compared to MOE/shPTPN13 cells

(Figure 3-1G and 3-1H). Taken together, our results suggest that CXCL14 expression is specifically inhibited in HPV-positive cells, likely in an E7-dependent manner.

CXCL14 Downregulation in HPV-Positive Keratinocytes is Associated with Promoter

Hypermethylation

Previous studies have shown that CXCL14 expression is suppressed by DNA hypermethylation in the CXCL14 promoter region (284). To determine whether HPV induces

CXCL14 promoter hypermethylation, we analyzed the methylation status of the CXCL14 promoter in NIKS, NIKS-16, and W12 cell lines using methylation-specific PCR (MSP), as previously described (283). We found the CXCL14 promoter region was hypermethylated

Figure 3-2. CXCL14 downregulation in HPV-positive epithelial cells is associated with CXCL14 promoter hypermethylation. Genomic DNA was extracted from (A) NIKS, NIKS- 16, NIKS-16∆E7, W12E, W12G, and W12GPXY keratinocyte lines and (B) MOE/shPTPN13 (HPV-negative) and MOE/E6E7 (HPV-positive) cells. MSP was performed using specific primers and analyzed in 1.2% agarose gel as described in Experimental Procedures. MSP products of the control CXCL14 promoter and the hypermethylated CXCL14 promoter are indicated as “C” and “M”, respectively (B). (C) Bisulfite PCR products were cloned into the pGEM-T easy vector and sequenced. (D) CXCL14 expression was measured as described in Figure 3-1. (E & F) CaSki cells were treated with 10 μM decitabine for 6 days or a vehicle (H2O) control. RT-qPCR (E) and qMSP (F) were performed using total RNA and genomic DNA, respectively. CXCL14 mRNA copy numbers were normalized by ß-actin mRNA (E). -∆∆C Changes of CXCL14 promoter methylation were calculated using the 2 T method, and shown as a fold ratio of methylated signal over total signal (F). P-values were determined by Student’s t-test. Panels A-C were contributed by Tao Xu, PhD, St. Jude Hospital, TN.

51 in NIKS-16, W12E, W12G, and W12GPXY cells, but not in NIKS cells (Figure 3-2A).

Consistent with our results from cervical tissue specimens, the cervical keratinocyte lines

W12E, W12G, and W12GPXY showed gradually increasing levels of CXCL14 promoter hypermethylation during cancer progression (Figure 3-2A). To determine whether the HPV oncoprotein E7 affects CXCL14 promoter hypermethylation, we examined the methylation status of the CXCL14 promoter in NIKS-16∆E7 cells. Importantly, CXCL14 promoter hypermethylation was considerably less frequent in NIKS-16∆E7 cells (Figure 3-2A). These results indicate that the HPV16 E7 oncoprotein is necessary for HPV-induced CXCL14 promoter hypermethylation. Next, we analyzed the DNA methylation status of the CXCL14 promoter in HPV-positive vs. -negative MOE cell lines. Consistent with our results from the keratinocyte culture models, the CXCL14 promoter was hypermethylated in HPV-positive

MOE cells, but not in HPV-negative MOE cells (Figure 3-2B).

We determined the methylation status of the CpG island within the promoter region of CXCL14, using bisulfite sequencing on genomic DNA from NIKS, NIKS-16, W12E and

W12GPXY cells. Promoter amplicons were cloned from genomic DNA and 24 clones from each cell type were sequenced. Consistent with the MSP results above, there were no methylated cytidine residues detected in NIKS cells (Figure 3-2C). Conversely, DNA methylation in the CXCL14 promoter region appeared in NIKS-16 and W12E cells. A significantly higher frequency of CXCL14 promoter methylation was detected in the

W12GPXY cell line, showing that ~25% of the CXCL14 promoter clones contained multiple sites with DNA methylation (Figure 3- 2C). These results indicate that CXCL14 promoter hypermethylation is induced by high-risk HPVs and accumulated over the course of cancer progression. This implies that other unknown factors in addition to E7 may be necessary for accumulation of CXCL14 promoter hypermethylation in HPV-positive cells. To examine

CXCL14 promoter hypermethylation in HPV-positive cancer tissues, we analyzed CXCL14

DNA methylation data from 279 HNC and 309 CxCa tissue samples obtained from the

TCGA database (23). Consistent with our results from keratinocytes, CXCL14 DNA methylation is significantly increased in HPV-positive HNC and CxCa compared to HPV- negative HNC (Appendix D-3B). CXCL14 downregulation is highly correlated with CXCL14

DNA methylation in HPV-positive HNC and CxCa, but not in HPV-negative HNC (Appendix

D-3C to D-3E). These results indicate that CXCL14 mRNA expression is controlled by

CXCL14 promoter methylation in HPV-positive cancers. To verify CXCL14 downregulation by promoter hypermethylation, we determined whether the methylation inhibitor decitabine

(5-aza-β’-deoxycytidine) restores CXCL14 mRNA transcription (293). Unfortunately, decitabine was toxic to NIKS cells and ineffective in W12 cells at all concentrations tested.

Therefore, we examined consequences of DNA demethylation using an HPV16-positive

CxCa cell line (CaSki), which expresses low levels of CXCL14 transcripts (Figure 3-2D).

Decitabine treatment for 6 days significantly increased CXCL14 expression in CaSki cells, together with a 50% decrease in CXCL14 promoter methylation as determined by quantitative MSP (qMSP) (Figure 3-2E and 3-2F). These results suggest that reversing methylation at the CXCL14 promoter, even partially, drastically increases CXCL14 expression in HPV-positive cancer cells. Taken together, our results suggest that HPV downregulates CXCL14 expression in HPV-positive cells by facilitating promoter hypermethylation.

CXCL14 Expression Hinders Cell Migration in vitro

Previous studies have shown that CXCL14 interferes with IL-8 and CXCL12 signaling, which are important for tumor cell migration and invasion (278, 288). Consistently,

CXCL14 downregulation suppresses migration of colorectal and tongue cancer cell lines

(294, 295). To determine the effects of restoration of CXCL14 expression on HPV-positive cell migration, an in vitro scratch assay was performed. Unfortunately, NIKS and W12 cells differentiated and senesced when confluent, and therefore could not be used. Instead, we established CaSki and MOE/E6E7 cell lines re-expressing human CXCL14 or murine

Figure 3-3. CXCL14 expression hinders mobility of HPV-positive cancer cells. (A & B) CXCL14 re-expressing CaSki and MOE/E6E7 cell lines were established using lentiviral transduction of the human CXCL14 and murine Cxcl14 genes, respectively, and validated by RT-qPCR. CXCL14 and Cxcl14 mRNA copy numbers were normalized by human ß-actin or murine Gapdh mRNA, respectively. In vitro scratch assay was performed with the established CaSki (C & D) and MOE/E6E7 (E) cells. Images were captured at 0, 4, 8, and 12 hours post wounding, and the width of the wound gaps were measured using NIH Image J software. Representative data from three replicates of each group are shown. The initial wound gaps (white dashed bar) and representative gaps at indicated time points (solid white bar) are shown. The scale bars (black bar) indicate 500 μm. (F) Transwell migration assays were performed on CaSki cells re-expressing CXCL14 generated as in (A). The percentage of cells that migrated through the permeable supports is shown, using 0%, 2.5%, and 5% FBS as a generic chemoattractant. P-values were calculated using the Student’s t-test. *p < 0.0001, **p < 0.03.

Cxcl14, respectively, using lentiviral transduction (Figure 3-3A and 3-3B). The expression level of CXCL14 in CaSki cells was comparable to the level seen in NIKS (Figure 3-3A).

The results showed that restoration of CXCL14 expression in both CaSki and MOE/E6E7 cells significantly delayed wound closure (Figure 3-3C to 3-3E). While the gaps were filled within 8 hours with control CaSki and MOE/E6E7 cells, both CaSki and MOE/E6E7 cells re- expressing CXCL14 showed wide gaps of 50 to β00 μm at 1β hours post wounding. To further corroborate these results, we performed a transwell migration assay using CaSki cells re-expressing CXCL14 with FBS as a generic chemoattractant. The data revealed that

CXCL14 expression significantly reduced CaSki cell migration compared to the vector only control (Figure 3-3F). CXCL14 expression did not affect proliferation of CaSki and

MOE/E6E7 cells (data not shown). Taken together, these results suggest that CXCL14 downregulation in HPV-positive HNC and CxCa cells increases epithelial cell motility.

Restoration of Cxcl14 Expression Clears HPV-Positive Tumors in Immunocompetent

Mice, but not in Rag1-Deficient Mice

To determine whether CXCL14 suppresses HPV-positive tumor growth in vivo, we established ~20 clones of MOE/E6E7 cells expressing various levels of murine Cxcl14 using lentiviral transduction. Untransduced and vector-transduced MOE/E6E7 cells consistently showed an over 30-fold decrease of Cxcl14 mRNA expression compared to HPV-negative normal parental MOE cells (Figure 3- 4A). To define the in vivo effects of restored Cxcl14 expression, we tested two clones (8 and 16) of our Cxcl14 re-expressing MOE/E6E7 cells that had physiological levels of Cxcl14 mRNA expression comparable to parental MOE cells

(Figure 3-4A). Cxcl14 expression did not affect proliferation of MOE/E6E7 cells (data not shown). Wild type C57BL/6 mice were injected with 1 × 105 MOE/E6E7 cells, from our established clones, in the rear right flank. Tumor growth was monitored by measuring tumor volume for up to 11 weeks. Strikingly, restored Cxcl14 expression in MOE/E6E7 cells significantly suppressed tumor growth in wild type C57BL/6 mice, while vector-transduced

AB 104 6000 Vector ) 3 Cxcl14-clone 8

X) Cxcl14-clone 16 6 6 103 4000

102 p < 0.0001

GAPDH (10 GAPDH 2000 Expression relative to to relative Expression 1

10 (mm volume Tumor Parental MOE Vector Cxcl14 Cxcl14 MOE /E6E7 Clone 8 Clone 16 0 0 7 14 21 28 35 MOE/E6E7 clones Days CD 4000 Vector )

100 3 Cxcl14-clone 8 Cxcl14-clone 16 3000 Cxcl14-clone 16

2000 50 *p < 0.0001 Cxcl14-clone 8 1000 Percent survival Percent Vector p < 0.0001 (mm volume Tumor 0 0 0 20 40 60 80 0 7 14 21 28 35 Days Days

EF n.s. 3200 100 ) Cxcl14-clone 8 3 *p < 0.0001 2400

1600 50 Cxcl14-clone 16 **

800 Percent survival Percent

Vector p = 0.0015 (mm volume Tumor 0 0 - - - 0 20 40 60 80 e -/ e -/ e -/ p 1 p 1 p 1 ty g ty g ty g Days ld a ld a ld a i R i R i R W W W Vector Cxcl14 Cxcl14 clone 8 clone 16

Figure 3-4. Restoration of Cxcl14 expression clears HPV-positive tumor in immunocompetent mice, but not in Rag1-deficient mice. MOE/E6E7 cell clones containing the Cxcl14 gene or vector were established and Cxcl14 expression levels were determined by RT-qPCR (A). Two MOE/E6E7 cell clones re-expressing Cxcl14 (clones 8 and 16) and one vector containing MOE/E6E7 cell clone were injected into the rear right flank of C57BL/6 (B, D, & E) and Rag1-/- (C, D, & F) mice (n = 10, each group of wild type; n = 7, each group of Rag1-/-). Tumor growth was determined every week by the formula: volume = (width)2 × depth. P-value was determined by one-way ANOVA analysis (B & C) and the Student’s t-test (D). Survival rates of wild type and Rag1-/- mice were analyzed using a Kaplan-Meier estimator (E & F). Time-to-event was determined for each group (vector only, Cxcl14-clone 8, Cxcl14-clone 16) with the event defined as a tumor burden larger than 2,500 mm3. Deaths not associated with tumor (samples collected for flow cytometry) were censored. P-values were determined by the Log-rank test (E & F). Experiments for panels B- F were completed in collaboration with Dan Vermeer, Sanford Health.

MOE/E6E7 cells rapidly formed tumors (Figure 3- 4B). All ten mice transplanted with vector- transduced MOE/E6E7 cells succumbed to tumor burden within 5 weeks post injection

(Figure 3-4C). In contrast, five and seven out of ten mice transplanted with Cxcl14 re- expressing MOE/E6E7 clones 8 and 16, respectively, were tumor-free up to 11 weeks post injection (Figure 3- 4C and Appendix D-4A to D-4C). To determine whether adaptive immune responses are involved in Cxcl14-mediated tumor suppression, we examined the tumor growth from these clones in Rag1-deficient C57BL/6 mice (Rag1-/-). Tumor growth was moderately slowed by restored Cxcl14 expression in Rag1-/- mice up to 14 days post injection (Figure 3-4D). However, all fourteen Rag1-/- mice injected with clones 8 and 16 exhibited tumor growth and succumbed to tumor burden within 5 weeks post injection

(Figure 3-4D, 3-4E, and Appendix D-4D to D-4F). The results demonstrate no significant difference in tumor growth between wild type and Rag1-/- mice transplanted with vector control MOE/E6E7 cells at 21 days post injection (Figure 3-4F). These results indicate that

Cxcl14 expression is critical to trigger an adaptive immune response to clear implanted cancer cells in vivo.

Restored Expression of Cxcl14 Increases Natural Killer (NK), CD4+ T, and CD8+ T Cells in Tumor-Draining Lymph Nodes in vivo

To characterize immune cell infiltration regulated by Cxcl14 expression, we analyzed various immune cells in TDLNs and spleens harvested from the wild type C57BL/6 mice at

21 days post injection with vector or Cxcl14 re-expressing MOE/E6E7 cells. Using flow cytometry, we assessed populations of hematopoietic cells (CD45+) including NK cells

(NKp46+), CD4+ T cells (CD4+), CD8+ T cells (CD8+), antigen presenting cells (MHCII+), neutrophils (Gr1high), monocytes (Gr1mid), and macrophages (MHCII+, F4/80+). Our gating strategy for all interrogated cell types was based on cell populations detected in spleens and lymph nodes from C57BL/6 mice (Appendix D-5). Our data showed that percentages of NK,

CD4+ T, and CD8+ T cells were highly increased in TDLNs of the mice transplanted with

Figure 3-5. Cxcl14 expression increases NK, CD4+ T and CD8+ T cells in tumor draining lymph nodes. MOE/E6E7 cells with Cxcl14 (clones 8 and 16) or vector were injected into the rear right flank of C57BL/6 mice (n = 10, each group). Tumor-draining lymph nodes (TDLNs) were harvested from the mice at 21 days post injection. Percentage of immune cell populations defines the frequency of lymphocytes that were single cells and either NK (CD45+, NKp46+), CD4+ T (CD45+, CD4+), or CD8+ T (CD45+, CD8+) cells. Gating for flow cytometry was based on splenocyte populations and applied to TDLN samples as described in Appendix D-5. Representative flow cytometry diagrams are shown (A-C) and quantification of the indicated immune cells in each mouse tested appears (D-F). P-values were determined between vector and either clone 8 or clone 16 by the Student’s t-test. Data were collected in collaboration with Joseph Westrich, University of Colorado.

MOE/E6E7 cells re-expressing Cxcl14 (Figure 3-5). These results suggest that Cxcl14 increases infiltration of NK, CD4+ T, and CD8+ T cells into TDLNs, which may be critical for tumor clearance. This is consistent with our tumor growth results showing a moderate delay in tumor growth by restored Cxcl14 expression in Rag1-/- mice, in which NK cell infiltration is increased in the absence of T cells (data not shown). These results suggest that NK cells alone may not be sufficient to clear HPV-associated tumors (Figure 3-4D and 3-4E). In addition to increased NK, CD4+ T, and CD8+ T cell infiltration, monocytes were also modestly increased in TDLNs of the mice injected with MOE/E6E7 cells re-expressing

Cxcl14. Conversely, Cxcl14 expression did not change antigen presenting cells, neutrophils, and macrophages in TDLNs (Appendix D-6), and marginal or no changes of these immune cell populations were observed in spleens by Cxcl14 expression (Appendix D-7). These results indicate that Cxcl14 locally affects NK, CD4+ T, and CD8+ T cell infiltration near the

TME. To determine any difference in local and systemic immune responses altered by

Cxcl14, populations of NK, CD4+ T, and CD8+ T cells were compared between TDLNs and distal lymph nodes (LNs) in the same mice injected with MOE/E6E7 cells with or without restored Cxcl14 expression. We found that NK, CD4+ T, and CD8+ T cell populations were significantly decreased in TDLNs compared to distal LNs in mice injected with control

MOE/E6E7 cells (Figure 3-6). In contrast, mice injected with MOE/E6E7 cells re-expressing

Cxcl14 showed significantly restored NK, CD4+ T, and CD8+ T cell populations in TDLNs comparable to distal LNs (Figure 3-6). These results indicate that restoration of Cxcl14 expression reverses suppression of antitumor immune responses by locally recruiting NK,

CD4+ T, and CD8+ T cells.

Expression of Cxcl14 Induces Chemotaxis of NK, CD4+ T, and CD8+ T Cells in vitro

To determine whether expression of Cxcl14 in MOE/E6E7 cells induces chemotaxis of NK, CD4+ T, and CD8+ T cells, we performed an immune cell migration assay using the transwell system and splenocytes isolated from C57BL/6 mice. The results showed that

NK 60

p < 0.0001 40

20 Cell percentage Cell ns ns 0 Distal TDLN Distal TDLN Distal TDLN LN LN LN Vector Clone 8 Clone 16

CD4 50

Cell percentage Cell 10 p = 0.0003 p = 0.007 p = 0.005

0 Distal TDLN Distal TDLN Distal TDLN LN LN LN Vector Clone 8 Clone 16

CD8 40

10 Cell percentage Cell p < 0.0001 p = 0.01 p = 0.003

0 Distal TDLN Distal TDLN Distal TDLN LN LN LN Vector Clone 8 Clone 16

Figure 3-6. Cxcl14 expression restores decreased populations of NK, CD4+ T and CD8+ T cells in TDLNs. TDLNs (closed circle) and distal lymph nodes (distal LNs, open circle) were harvested from the mice injected with MOE/E6E7 cells re-expressing Cxcl14 (clones 8 and 16) or containing vector only. Percentage of NK, CD4+ T and CD8+ T cell populations were analyzed as described in Figure 3-5. P-values were determined between TDLN and distal LNs by the Student’s t-test. Data were collected in collaboration with Joseph Westrich and Jennifer Berger, University of Colorado.

NK p = 0.01 p = 0.02 25 25 CD4 p = 0.01 p = 0.01 20 20

15 15

10 10

5 5 Percent cell migration cell Percent Percent cell migration cell Percent 0 0 Negative Vector Clone 8 Clone 16 Negative Vector Clone 8 Clone 16 control CM CM CM control CM CM CM

p = 0.02 25 CD8 40 Neutrophil p = 0.01 20 30

15 20 10 10 5 Percent cell migration cell Percent Percent cell migration cell Percent 0 0 Negative Vector Clone 8 Clone 16 Negative Vector Clone 8 Clone 16 control CM CM CM control CM CM CM

Figure 3-7. Cxcl14 expression induces chemotaxis of NK, CD4+ T and CD8+ cells. Conditioned media (CM) from the culture of MOE cells with Cxcl14 (clones 8 and 16) or vector were added into the bottom chamber of a transwell and supplemented with IL-2. Splenocytes isolated from C57BL/6 mice were added to the upper chamber. After 12 hour incubation, migrated splenocytes to the bottom chamber were collected and analyzed by flow cytometry. Percentage of immune cell populations defines the frequency of immune cells that were single cells and either (A) NK (CD45+, NKp46+), (B) CD4+ T (CD45+, CD4+), (C) CD8+ T (CD45+, CD8+) cells, or (D) neutrophils (CD45+, Gr1high). P-values were determined between vector and CXCL14 expressing cells (clones 8 and 16) by the Student’s t-test. Experiments were contributed by Joseph Westrich, University of Colorado.

61 conditioned medium from cultured MOE/E6E7 cells re-expressing Cxcl14 (clones 8 and 16) significantly increased NK, CD4+ T, and CD8+ T cell chemotaxis, while conditioned medium from MOE/E6E7 cells containing vector only has little effect compared to the negative control (Figure 3-7A to 3-7C). Consistent with the in vivo immune cell infiltration (Fig. S6B), neutrophil migration was not affected by Cxcl14 expression (Figure 3-7D). These results suggest that Cxcl14 plays an important role in recruitment of NK, CD4+ T, and CD8+ T cells, which may enhance antitumor immune responses.

Discussion

Like most cancers, HPV-associated cancer development requires decades to progress from HPV-infected cells to invasive disease. Recent cancer genomics studies of

HNCs have reported that HPV-positive HNCs have far fewer oncogenic mutations (~5 per tumor) compared to HPV-negative HNCs (>20 per tumor) (296). These findings indicate that viral factors replace oncogenic processes usually triggered by multiple somatic mutations in

HPV-unrelated cancer progression. Other studies showed that continuous expression of the

HPV oncogene E7 is required for cancer growth and maintenance in vitro and in vivo (297,

298), suggesting that HPV E7 has multiple functions in HPV-associated cancer progression.

However, the mechanism by which HPV infection contributes to multiple steps of decades- long cancer progression is poorly understood.

Several proinflammatory chemokines such as IL-8, CXCL1, and CXCL12 drive cancer progression by facilitating tumor cell growth, survival, and migration, as well as by inducing angiogenesis (299). In our study, expression of proinflammatory chemokines IL-8,

CXCL1, CXCL2, and CCL3 was upregulated in the early stages of cancer progression

(Appendix D-1A and D-1B). These chemokines are also increased in HPV-negative HNCs but to higher levels than HPV-positive HNCs, suggesting that most HNCs might have increased levels of proinflammatory chemokine expression that is pivotal for tumor cell migration and angiogenesis (221).

We found that CXCL14 was significantly downregulated during CxCa progression and in HPV-positive HNCs compared to HPV-negative HNCs (Figure 3-1A and 3-1C).

Constitutively expressed CXCL14 is an important homeostatic chemokine in normal epithelial and neural tissue of mammals (223, 280, 300). Additionally, by directly binding to

IL-8, CXCL14 inhibits the ability of IL-8 to recruit endothelial cells and promote angiogenesis

(278), which is known to be essential for cancer progression. While specific receptors of

CXCL14 have not been identified, a recent study showed that CXCL14 binds to CXCR4 as a decoy ligand, inhibiting CXCL12 signal transduction through CXCR4 (288). This is an important signaling pathway for cell growth, angiogenesis, and metastasis in many cancers.

CXCL14 expression is frequently downregulated in cervical, prostate, colorectal, lung, and oral cancers (280–283, 286, 301, 302). Overexpression of CXCL14 has shown antitumor effects by suppressing tumor growth and cancer cell migration in breast, oral, lung, and liver cancers (284–286, 303, 304). Consistently, our results here show that restored CXCL14 expression in HPV-positive cells significantly suppresses tumor growth in vivo (Figure 3-4).

Additionally, the HPV oncoprotein E7 induces CXCL14 promoter hypermethylation and significantly downregulates CXCL14 expression (Figure 3-1 and 3-2). A previous study determined that HPV16 E7 activates the methyltransferase activity of DNMT1 (127). A preliminary experiment detected upregulation of DNMT1 expression in HPV-positive cancers and keratinocytes (data not shown). These observations suggest that CXCL14 promoter methylation may be mediated through interactions between E7 and DNMT1. CXCL14 expression in HPV-positive CaSki cells was significantly increased following treatment with decitabine, an FDA-approved DNMT inhibitor (305) (Figure 3-2E). Previous studies have shown that DNA hypermethylation is associated with suppression of various immune factors including downregulation of cancer testis antigen, and MHC class I and chemokine expression (306). Consistently, inhibition of DNA methylation by decitabine increases expression of cancer testis antigens and MHC molecules and enhances cytotoxic NK and T

63 cell antitumor activity (307–309). Decitabine treatment also activates expression of several different chemokines in a murine ovarian cancer model (310). Similarly, a recent study showed that decitabine treatment enhanced antitumor immune responses by increasing

CXCL9 and CXCL10 expression and effector T cell infiltration (311). Thus, reversing the promoter hypermethylation of CXCL14 could be a feasible approach for restoring antitumor immune responses to treat HPV-positive cancers.

In our current study, we assessed the potential for CXCL14 to alter immune cell infiltration in TDLNs. We showed that restoration of Cxcl14 expression increases percentages of NK, CD4+ T, and CD8+ T cell populations in TDLN (Figure 3-5). Because tumor growth is only partially suppressed by Cxcl14 expression in Rag1-/- mice, our results indicate that both innate and adaptive immune responses play important roles in the antitumor functions of CXCL14. Consistently, a marked reduction in NK cell activity in uterine walls was observed in Cxcl14-/- mice, compared to Cxcl14+/- mice (302). In addition,

NK cell depletion increases the risk of colorectal cancer in Cxcl14 transgenic mice (287). On the other hand, the effects of CXCL14 on T cells are completely unknown. Both NK and

CD8+ T cells are well known as effector killer cells capable of eliminating virus-infected cells as well as cancer cells (312–315). NK cell activation induces CD8+ T cell responses through priming DCs, suggesting that NK cells may be the link between innate and adaptive immunity to induce antiviral and antitumor CD8+ T cell responses (315, 316). Thus, our findings suggest that CXCL14, secreted by epithelial cells, might be one of the key regulators for NK, CD4+ T and CD8+ T cells to drive tumor clearance during HPV-associated cancer progression.

In conclusion, our study suggests that CXCL14 plays an important role in antitumor immune responses to clear HPV-positive HNC. CXCL14 is a small, secreted protein that can be used as a therapeutic agent. Additionally, identification of the native CXCL14 receptor(s) would provide druggable targets to enhance CXCL14 functions. Thus, further

64 studies of the effects of CXCL14 on NK and T cells may provide a novel means of anticancer immunotherapy to treat HNCs.

CHAPTER IV

HIGH-RISK HUMAN PAPILLOMAVIRUS E7 ALTERS HOST DNA METHYLOME AND

REPRESSES HLA-E EXPRESSION IN HUMAN KERATINOCYTES2

Importance

Over 5% of all human cancers are caused by persistent human papillomavirus (HPV) infection which requires evasion of host immune responses; however, the mechanisms of immune evasion by HPV are not fully understood. Immune regulatory pathways are frequently altered by DNA methylation. Because our previous studies confirmed that chemokine-mediated antitumor immune response is suppressed by HPV E7-dependent promoter methylation, we analyzed global gene expression and DNA methylation altered by

HPV E7. Here, we show that high-risk HPV E7 downregulates MHC class I molecules and particularly, HLA-E downregulation is dependent on E7-directed methylation at a distal CpG- rich promoter element. HLA-E expression is restored by demethylating agent. HLA-E plays an important role in antiviral immunity by activating natural killer and CD8+ T cells.

Therefore, epigenetic repression of HLA-E expression by HPV E7 may be an important immune evasion mechanism employed by HPV for persistent infection.

Introduction

Human papillomaviruses (HPV) are small double-stranded DNA viruses with over

180 genotypes that infect mucosal and cutaneous basal epithelia (317). It has been estimated that up to 80% of sexually active individuals will become infected in their lifetime, making HPV the most common sexually transmitted pathogen (318). HPVs are classified as high- and low-risk genotypes based on their oncogenic potential (319). High-risk HPVs are causally associated with 5% of all human cancers including nearly all cervical cancer (CxCa)

2 Data, figures, and portions of text from this chapter are currently in submission to Scientific Reports, July 2016.

66 and about 25% of head and neck cancer (HNC), making HPV a significant cause of morbidity and mortality worldwide (15, 319).

HPVs exclusively infect basal keratinocytes of the epithelium. A recent study showed that primary HPV infections can persist for two years or longer, with an average time to clearance of 9.4 months (154). Similar studies have revealed that of the genotypes tested,

HPV16 is the most likely to persist (320). Given the propensity of HPV to persist without eliciting a strong immune response, it is very likely that the virus has evolved efficient immune evasion mechanisms.

Gene expression dysregulation is a well-known strategy that viruses frequently employ to evade the host immune response (321). Of note, Epstein-Barr virus (EBV) hijacks host cell epigenetic machinery to modulate host gene expression (322). These epigenetic manipulations are considered a hallmark of EBV-induced lymphomas, and persist even after infection is cleared (322, 323). HPV-positive HNC and CxCa progression exhibit distinct changes in host DNA methylation that alter host gene expression (129, 164). In a similar study, HPV-induced cell immortalization corresponded with hypermethylation at several host chromosomal loci including the telomerase subunit hTERT (166). Expression of hTERT is increased by promoter hypermethylation which correlates with HPV-associated transformation and cancer progression (324). Previously it has been described that E7 directly binds and activates DNA methyltransferase 1 (DNMT1), leading to a potential epigenetic mechanism of E7-mediated gene silencing (127, 130). Consistently, the HPV E7-

DNMT1 complex induces hypermethylation of the tumor suppressor cyclin A1 (CCNA1) promoter, an epigenetic marker strongly correlated with HPV-associated malignancy (130,

325). Our recent work revealed that the chemokine CXCL14 is significantly downregulated by E7-directed promoter hypermethylation (326). Restoration of CXCL14 expression in HPV- positive cancer cells prevents tumor formation in vivo and increases natural killer (NK) and

CD8+ T cells in the tumor-draining lymph nodes (326). Downregulation of CXCL14 is

67 therefore an important immune evasion mechanism employed by HPV, allowing for virus persistence. Thus, it is likely that HPV E7 dysregulates other host gene expression by modulating DNA methylation to establish persistent virus infection.

Here, we performed parallel global gene expression and methylome analyses to identify key host factors and pathways altered by HPV-mediated DNA methylation in human keratinocytes. Additionally, we assessed genomic regions that are likely affected by HPV- directed changes in DNA methylation. We demonstrated that most class I major histocompatibility complex (MHC-I) molecules are transcriptionally downregulated in an E7- dependent manner. Further, non-classical HLA-E, which regulates NK and CD8+ T cells, is significantly downregulated by E7-mediated hypermethylation in a distal regulatory CpG island (CGI). These results suggest that HPV E7-mediated DNA methylation modulates host immune responses by downregulating HLA-E expression.

Results

The HPV oncoprotein E7 drives global gene expression changes in human keratinocytes

To determine gene expression alterations in human keratinocytes by high-risk HPVs, we performed gene expression profiling in normal immortalized keratinocytes (NIKS) and their derivatives: NIKS-16 and NIKS-18 cells containing episomal HPV16 and HPV18 genomes, respectively. The NIKS-16ΔE7 cell line containing the HPV16 genome lacking E7 expression (249) was used to investigate the functions of the HPV oncoprotein E7. Total

RNA was extracted from NIKS, NIKS-16, NIKS-18, and NIKS-16E7 cells from three different passages each and global gene expression was analyzed with human genome

Affymetrix U133 Plus 2.0 microarrays (GEO accession #GSE83259). Principal component analysis (PCA) of complete mRNA expression profiles demonstrated that HPV-positive cells

(NIKS-16 and NIKS-18) clustered together distinctly from NIKS and NIKS-16ΔE7 cells (Fig.

4-1A).

Figure 4-1. High-Risk HPV E7 Distinctly Alters Host Gene Expression in Normal Keratinocytes.

Figure 4-1. High-Risk HPV E7 Distinctly Alters Host Gene Expression in Normal Keratinocytes. Gene expression profiles were assessed by microarray in triplicate for normal keratinocytes lines, NIKS, NIKS-16, NIKS-18, and NIKS-16ΔE7, in three different passages. (A) Principal component analysis data are shown for each replicate of NIKS (red circle), NIKS-16 (blue square), NIKS-18 (green triangle) and NIKS-16ΔE7 (black triangle) cells (B) Normalized expression of differentially expressed genes in both NIKS16 cells vs. NIKS cells and NIKS16 cells vs. NIKS16ΔE7 cells is shown by heat map (FDR adjusted p < 0.05 and a change > 30% magnitude in expression). Probe IDs are listed in Table S1. (C) Heat map of dysregulated cell cycle-related genes previously identified in HNC and CxCa patient tissue samples (110). (D) CDKN2A, MCM5, MCM7, and UHRF1 expression levels were determined by RT-qPCR and normalized to β-actin expression levels. Fold changes to NIKS cells are plotted. P-values were calculated by Student’s t test. *p < 0.0001, **p < 0.001, ***p < 0.01, ****p < 0.05. Data processing was performed in collaboration with Rachel Blumhagen, University of Colorado.

Our and other studies have shown that high-risk HPV infection significantly changes host gene expression patterns including dramatically increased DNA replication- and cell cycle-related gene expression (110, 111, 327). However, the extent of E7-specific gene expression changes has not been fully determined. To define E7-mediated gene expression changes in normal keratinocytes, we analyzed the global expression data and identified genes up- or downregulated in both NIKS-16 cells vs. NIKS cells and NIKS-16 cells vs.

NIKS-16ΔE7 cells. Genes exhibiting a false discovery rate (FDR)-adjusted p-value of less than 0.05 for each comparison, and genes exhibiting a change greater than 30% magnitude in expression are selected (Fig. 4-1B, Appendix E-5). To examine the physiological relevance of the HPV-specific gene expression changes in cultured keratinocytes, we analyzed the expression patterns of distinct cell cycle-specific genes which were previously identified using CxCa, HNC, and normal patient tissue samples (110). Consistent with the results from patient tissues, the majority of the cell cycle genes upregulated in HPV-positive cancers were markedly increased in NIKS-16 and NIKS-18 cells compared to NIKS cells, while none of the cell cycle genes upregulated in HPV-negative cancer were increased in

NIKS-16 and NIKS-18 cells (Fig. 4-1C). Of note, most of the upregulated genes in NIKS-16 and NIKS-18 cells were not changed or slightly downregulated in NIKS-16ΔE7 cells compared to NIKS cells. These results indicate that the distinct patterns of cell cycle dysregulation in HPV-positive cancers are largely caused by HPV oncoprotein E7 expression. Using RT-qPCR, we further validated expression changes of selected genes from Fig. 4-1C (CDKN2A and MCM7) and previously reported (UHRF1 and MCM5) (110,

111) (Fig. 4-1D). These results indicate a significant role of the HPV oncoprotein E7 in global gene expression changes during persistent HPV infection in human keratinocytes including cell cycle-related genes.

The HPV oncoprotein E7 downregulates gene expression related to antigen presentation

To understand the biological functions of the identified HPV16 E7-regulated genes

(Appendix E-5A and 5B), we performed pathway analysis using a peer reviewed web- based pathway database, Reactome (reactome.org). Consistent with our previous findings

(110, 111), the majority of the upregulated genes were involved in cell cycle progression (i.e.

MCM7 and CDK2), DNA replication (i.e. MCM4 and CCNA1), and DNA repair (i.e. BLM and

RAD1) (Appendix E-5A, Appendix E-1A). In contrast, the pathways of downregulated genes are diverse, suggesting that E7-mediated downregulation of gene expression is more complex than E7-mediated upregulation of gene expression, which might be largely affected by pRB degradation. Our analysis revealed that genes involved in antigen presentation, IL1 signaling and extracellular matrix degradation are significantly downregulated in NIKS-16 cells compared to NIKS and NIKS-16∆E7 cells (Appendix E-6B, Appendix E-1B). Various matrix metalloproteinases (MMPs) and kallikreins (KLKs) were significantly downregulated in

NIKS-16 cells compared to NIKS cells, but not in NIKS-16ΔE7 cells (Appendix E-6B,

Appendix E-2A). To further examine gene expression changes of MMPs by HPV in patient tissues, we analyzed The Cancer Genome Atlas (TCGA) data of HNC patients (129) obtained from cBioPortal (cbioportal.org). The results showed no significant differences in the expression levels of MMP1, 9, 10, and 28 between HPV-positive and HPV-negative

HNCs (data not shown). This suggests that MMP downregulation by E7 may be an early event in HPV-infected keratinocytes. Accordingly, MMP overexpression in HPV-positive cancers, which has previously been observed (328, 329), might be caused by other mechanisms such as -catenin and Ras activation during disease progression to invasive cancer (330, 331).

Importantly, immune response pathways were among the most significantly affected by HPV16 E7 expression (Appendix E-1B, Appendix E-6B). Several genes in antigen

72 presentation (e.g. HLA-B, HLA-E, SEC31A, ITGAV, CTSL2, and RNASEL) and IL1 signaling

(e.g. IL1B, IL1R1, IL1RN, and IL36G) were significantly changed in NIKS-16 and NIKS-18 cells, but not in NIKS-16ΔE7 cells, compared to NIKS cells (Appendix E-2B and E-2C).

Additional RT-qPCR validation showed E7-dependent dysregulation of genes involved in IL1 signaling (Appendix E-2D). Previous studies have shown that HPV16 E5 disrupts trafficking of MHC-I and –II complexes to the cell surface, and HPV16 E7 downregulates cell surface expression of MHC-I complexes (83, 84, 189). While multiple mechanisms of inhibiting HLA surface expression have been observed, HPV-mediated alterations in HLA gene expression is poorly understood. Thus, we further assessed expression of all HLA-I genes (HLA-A, -B, -

C, -E, -F, and –G) in the NIKS cell lines. The results showed that with the exception of HLA-

F, all HLA-I mRNA expression was significantly downregulated in NIKS-16 cells and HLA-A,

-C, and -E were also downregulated in NIKS-18 cells compared to NIKS cells (Fig. 4-2).

Notably, downregulation of the HLA-I genes was not observed in NIKS-16∆E7 cells (Fig. 4-

2). To determine any difference in HLA-I expression between HPV-positive and HPV- negative cancer tissues, we analyzed TCGA data of HNC and CxCa obtained from cBioPortal. Of note, the expression levels of HLA-C and HLA-E were significantly lower in

HPV-positive HNCs than HPV-negative HNCs (Fig. S3). Decreased HLA-E expression was also observed in CxCa compared to HPV-negative HNCs. Unfortunately due to the presence of numerous splice isoforms for HLA-I genes (332), we were unable to reliably detect amplicons from HLA-I mRNA by RT-qPCR. However, all array probe sets for HLA-A, -

B, -C, -E and -G detection consistently exhibit significant downregulation in HPV-positive keratinocytes in an E7-dependent fashion (Fig. 4-2). Overall downregulation of the HLA-I genes suggest that HPV16 E7 plays an important role in immune evasion of HPV-infected keratinocytes during early stages of persistent infection.

HLA-A HLA-B 12.5 11.5 p = 0.001 11.0 p = 0.0015 12.0 p = 0.01 10.5 11.5 10.0 11.0 p > 0.05 9.5 p < 0.01 mRNA expression (log2) mRNAexpression mRNA expression (log2) mRNAexpression 10.5 9.0 NIKS NIKS16 NIKS18 NIKS16 NIKS NIKS16 NIKS18 NIKS16 ΔE7 ΔE7 HLA-C HLA-E 12.0 9.5 p = 0.004 p = 0.0001 p = 0.001 11.5 9.0 p < 0.0001

11.0 8.5

10.5 8.0 p = 0.015 p = 0.0001 mRNA expression (log2) mRNAexpression 10.0 (log2) mRNAexpression 7.5 NIKS NIKS16 NIKS18 NIKS16 NIKS NIKS16 NIKS18 NIKS16 ΔE7 ΔE7 HLA-F HLA-G 6.0 10.0

9.5 p < 0.01 5.5

9.0

5.0 8.5 p < 0.01 mRNA expression (log2) mRNAexpression mRNA expression (log2) mRNAexpression 4.5 8.0 NIKS NIKS16 NIKS18 NIKS16 NIKS NIKS16 NIKS18 NIKS16 ΔE7 ΔE7

Figure 4-2. High-Risk HPV E7 downregulates HLA-I gene expression in normal keratinocytes. Normalized gene expression of HLA-A, -B, -C, -E, -F, and -D in NIKS (circle), NIKS-16 (square), NIKS-18 (triangle), and NIKS-16ΔE7 (inverse triangle).

Fluorescence intensity (log2) of each replicate is plotted and p values were calculated by Student’s t test.

The HPV oncoprotein E7 dysregulates DNA methylation in human keratinocytes

A previous study has shown that HPV infection distinctly alters the methylation patterns in HNC patients (333). Additionally, HPV16 E7 protein directly binds to DNMT1 and activates its enzymatic activity (127). We recently reported that HPV oncoprotein E7- dependent methylation of the CXCL14 promoter resulted in CXCL14 downregulation and inhibition of antitumor immune responses (326). These findings suggest that high-risk HPV

E7 is very likely to dysregulate host gene expression by modulating DNA methylation. To investigate the extent of gene expression dysregulated by HPV E7-induced DNA methylation, we analyzed the methylome of NIKS, NIKS-16, NIKS-18, and NIKS-16ΔE7 cell lines using Illumina Infinium HumanMethylation450 BeadChip arrays (GEO accession

#GSE83261). PCA of methylome profiles showed that each cell type clustered distinctly

(Fig. 4-3A). Given that the PCA from our gene expression analysis showed high similarity between NIKS-16 and NIKS-18 cells (Fig. 4-1A), the methylome data may discriminate the molecular patterns of different cell types more precisely than the gene expression data.

Nevertheless, the sample-by-sample variations in the triplicates of each cell line are much lower in the methylome data (Fig. 4-3A) than the sample-by-sample variations in gene expression data (Fig. 4-1A). This implies that DNA methylation could be a better biomarker than gene expression for early detection of high-risk HPV infection.

By assessing the relative methylation density at any given CpG site across the genome, we found that the NIKS cells tended to maintain β-values (the ratio of methylated probe intensity vs. the overall intensity) near 0 (0% methylation) or 1 (100% methylation) with uniform distribution between those two peaks on both flanks (Fig. 4-3B, black line). In contrast, both NIKS-16 and NIKS-18 cells exhibited an influx in hemi-methylation near β =

0.6 (Fig. 4-3B, red and orange lines). Of particular note, the methylation pattern of NIKS-

16ΔE7 cells was strikingly similar to the methylation pattern of NIKS cells but distinct from the methylation pattern of NIKS-16 and 18 cells (Fig. 4-3B, blue line). These results suggest

Figure 4-3. HPV16 E7 Alters Host Genome Methylation in Normal Keratinocytes. Global DNA methylation profiles in NIKS, NIKS-16, NIKS-18, and NIKS-16ΔE7 cells were analyzed using Illumina Infinium HumanMethylation450 BeadChip arrays. (A) Principal component analysis for each replicate of normalized data from NIKS (red circle), NIKS-16 (blue square), NIKS-18 (green triangle) and NIKS-16ΔE7 (black triangle) cells is shown. (B) Microarray data from NIKS (black), NIKS-16 (red), NIKS-18 (orange) and NIKS-16ΔE7 (blue) cells were normalized using SWAN and the relative methylation (β) density across the genome are plotted. β represents the ratio of methylated signal to total signal at a given CpG site. β near 0 or 1 indicates no methylation or complete methylation, respectively. Three pairwise comparisons are summarized by Venn diagrams showing the number of overlapping (C) differentially methylated positions (DMP, FDR adjusted p < 0.05) and (D) differentially methylated regions (DMR, FDR adjusted p < 0.05). DMPs are defined as a single differentially methylated CpG site between groups while DMRs consist of at least two DMP sites each within 500bp of another DMP. Data processing was performed in collaboration with Rachel Blumhagen, University of Colorado.

76 that the HPV oncoprotein E7 plays an important role in changing the global methylation patterns of host genome.

To validate our methylome array data, we assessed DNA methylation status at the

CCNA1 and TERT promoter regions that are known to be hypermethylated in HPV-positive cells. Consistent with previous findings, the CCNA1 and TERT promoter regions showed significantly increased methylation (24-β8% increase in β, p < 0.004) in NIKS-16 cells compared to NIKS cells (Table 4-1) (130, 166, 325, 334). However, NIKS-18 cells did not show consistent changes in CCNA1 and TERT promoter methylation. Given that the global methylome data show the distinct DNA methylation patterns between NIKS-16 and NIKS-18 cells, these results suggest that HPV16 and HPV18 may employ different mechanisms to modulate DNA methylation.

A genome-wide comparison of methylated CpG sites between NIKS and NIKS-16 cells revealed 5,190 differentially methylated positions (DMPs, defined as a single differentially methylated CpG site) and 1,307 differentially methylated regions (DMRs, defined as a cluster of two or more CpG sites permutation p-value < 0.05) (Fig. 4-3C and 4-

3D). To investigate DNA methylation specifically regulated by HPV16 E7, we identified 953

DMRs (red bold in Fig. 4-3D) in a comparison between NIKS-16ΔE7 cells to NIKS-16 cells and excluded the DMRs found in the NIKS-16ΔE7 to NIKS comparison from the 953 DMRs.

Using a more stringent DMR area p-value less than 0.01, we found 56 hypermethylated

DMRs (Appendix E-7A) and 47 hypomethylated DMRs (Appendix E-7B) that are dependent on HPV16 E7 expression. Additionally, regional methylation analysis near HLA-E identified two significantly hypermethylated DMRs across a total of 5 probed CpG sites (p ≤

0.004, Appendix E-7A), suggesting that HPV16 E7 may mediate DNA methylation of the

HLA-E gene.

To determine gene expression regulated by E7-mediated DNA methylation, we identified genes that show gene expression changes with an FDR adjusted p-value less

Table 4-1. CCNA1 and TERT Exhibit Increased Methylation in HPV16-Positive Keratinocytes.

NIKS to NIKS-16 NIKS to NIKS-18

Gene Illumina ID Δβ p Δβ p

CCNA1 cg07962128 28.0% 2.0E-05 -5.6% 9.5E-03

CCNA1 cg08296176 24.9% 1.1E-05 25.8% 5.6E-05

CCNA1 cg21587066 24.6% 3.7E-06 23.4% 1.0E-06

TERT cg21741223 28.3% 3.4E-03 -10.4% 5.0E-02

78 than 0.01 and associated DMPs with an FDR adjusted p-value less than or equal to 0.005 between NIKS and NIKS-16 cells (Appendix E-8). DMPs rather than DMRs were used in this analysis to reduce the possibility of type II errors: the locations of methylation array probes are predicted to be sentinel CpG sites and may not be clustered near additional probes, thus potential true positives may be eliminated from DMR classification. A total of

122 genes showed significant changes of both gene expression and DNA methylation comparing NIKS-16 cells to NIKS cells. This result is consistent with a previous global methylation study assessing epigenetic changes directed by EBV, showing that most DMPs are silent and relatively a small number of them contributed to changes in gene expression

(323). Our results suggest that E7-mediated DNA methylation regulates gene expression of a subset of host genes in human keratinocytes. Of note, HLA-E shows a significant decrease in gene expression and increase in DNA methylation (Figs. 4-2 and 4-4).

Consistent with the HLA-E gene expression results, the comparison between NIKS-16 and

NIKS-16ΔE7 cells showed a significant decrease in DNA methylation at HLA-E (24%) in

NIKS-16ΔE7 cells compared to NIKS-16 cells (Appendix E-9). These results suggest that downregulation of HLA-E gene expression shown in Fig. 4-2 is likely caused by HPV16 E7- induced HLA-E DNA methylation.

DNA methylation of the HLA-E CGI is significantly increased by the HPV oncoprotein

The larger aim of this study was to identify immune-related genes that are repressed transcriptionally by DNA methylation. Our transcriptome and methylome merged data sets

(FRD p < 0.01 and p < 0.005, respectively) revealed 122 genes exhibiting these criteria.

Here, we identified DNA methylation upstream of HLA-E in our methylome data which consistently showed hypermethylation in NIKS-16 cells compared to NIKS cells in an E7- dependent manner (Appendix E-8). HPV has previously been shown to downregulate HLA surface expression by inhibiting HLA trafficking and TAP-mediated antigen binding.

Figure 4-4. HPV16 E7 is Necessary for Hypermethylation at a Distal HLA-E CpG Island. (A) The difference in methylation (β) of all probed CpG dinucleotides in the HLA-E CpG island (CGI, chr6:30,434,030-30,434,730) between NIKS and NIKS-16 cells is shown. Positive and negative β indicates increased or decreased methylation in NIKS-16 cells compared to NIKS cells, respectively. (B) Methylation specific PCR (MSP) products were separated in 2% agarose gel to evaluate the methylation status of the HLA-E CGI using bisulfite-converted gDNA from NIKS, NIKS-16, and NIKS-16ΔE7 cells using primers listed in Appendix B. (C) Sequence logos of enriched transcription factor (TF) binding motifs. 100 bp regions flanking E7 sensitive DMRs (comparing NIKS to NIKS-16, p < 0.04, count 185) were assessed for enrichment of TF binding motifs using MEME Suite software and nucleotide frequencies in the submitted sequences as background (p < 0.05). (D) Schematic diagram of the potential mechanism of targeted E7-induced DNA methylation. E7 binds transcription factors (or complexes) through its CR1/2 domain and DNMT1 through its CR3 domain, leading to hypermethylation near specific transcription factor binding motifs.

Conversely, suppression of HLA transcription mediated by HPV, specifically E7, has not been described. Downregulation of antigen presentation molecules may act as an immune evasion strategy employed by HPV and DNA methylation may therefore be a key mechanism of HLA-E repression.

To further examine the DNA methylation in the HLA-E gene, we analyzed changes in

β-values between NIKS and NIKS-16 cells by scanning the CpG island (CGI) associated with the HLA-E gene. Unexpectedly, the HLA-E CGI containing the identified DMR is

~23,000 bases upstream of the HLA-E open reading frame (ORF), potentiating its functional role as an enhancer or distal promoter element. Our results from scanning the HLA-E CGI for DNA methylation showed a dramatic increase near the γ’ region of the CGI (Fig. 4-4A).

To validate the DNA methylation in the HLA-E CGI, we performed methylation-specific PCR

(MSP) using primer sets specific to the DMR at the γ’ region of the HLA-E CGI. Consistent with the methylome array data, the γ’ region of the HLA-E CGI was highly methylated in NIKS-16 cells compared to NIKS cells (Fig. 4-4B). Strikingly, DNA methylation of the HLA-E CGI was dramatically decreased in NIKS-16ΔE7 cells (Fig. 4-4B). This DNA methylation status of HLA-E is highly correlated with the decrease of HLA-E expression in

NIKS-16 cells, but not in NIKS-16ΔE7 cells (Fig. 4-2D). These results suggest that HLA-E gene expression is likely to be downregulated by HPV16 E7-mediated DNA methylation in

HPV-positive keratinocytes.

HPV16 E7 directly binds and activates DNMT1 through its CR3 zinc finger binding domain, providing a potential mechanism of E7-induced host DNA methylation (127).

However, it is unclear how E7-induced DNA methylation targets specific regions in the genome. We hypothesized that the regions near specific transcription factor (TF) binding sites are targeted by the E7-DNMT1 complex to induce DNA methylation. Supporting our hypothesis, a previous study revealed that HPV16 E7 recruits histone deacetylases

(HDACs) to IRF-1 regulatory promoter complexes thereby directing histone deacetylation to

81 silence IRF-1 responsive genes (136). To test our hypothesis, we compiled a list of hypermethylated DMRs, filtered as described above for E7-dependent hypermethylation (p <

0.04, count 185) and submitted DMRs with 100 bp of flanking sequence to MEME suite for analysis of enrichment of TF binding motifs. E7-dependent hypermethylated DMRs showed enrichment of EPAS1, FOXJ3, CDX2, IRF4, FOXF1, and glucocorticoid receptor (GCR) TF binding sites (Fig. 4-4C). Enrichment of IRF4, FOXF1 and GCR motifs imply that E7- mediated DNA methylation may be directed to TF binding motifs near immunoregulatory and developmental genes (335–338). Consistently, scanning the HLA-E CGI (containing the identified DMR) for TF binging sites identified an AGAACA motif, which is a consensus binding site for GCR (Fig. 4-4C). Previous studies have shown that GCR is involved in suppression of MHC-I (337) and MHC-II expression (339). Enrichment of methylation near specific TF binding sites implies that HPV E7 may direct DNA methylation by recruiting

DNMT1 methyltransferase near specific promoter elements through interactions with TFs

(Fig. 4-4D). Further analysis revealed that the HLA-E CGI contains sites for DNase I hypersensitivity and acetylated H3K27 histone markers, both indicative of active regulatory elements. Additionally, multiple small RNAs are transcribed from γ’ of the HLA-E CGI

(Appendix E-8). MicroRNA target prediction analysis of these small RNAs revealed 65 putative target cellular mRNAs (Appendix E-10), including histocompatibility 13 (HM13), which is involved in loading peptides onto HLA-E. HLA-E surface expression and stability require antigen binding, suggesting a potential mechanism of downregulating HLA-E surface expression (340). Taken together, the HLA-E CGI appears to be an active site for transcription of small RNAs and other regulatory elements which may have direct or indirect effects on HLA-E expression.

The promoter activity of the HLA-E CGI is regulated by DNA methylation

To assess the transcriptional regulation by DNA methylation at the HLA-E CGI, we employed a promoter reporter assay using a CpG-free firefly luciferase expression vector,

Figure 4-5. Promoter Activity of the HLA-E CGI is Repressed by DNA Methylation. (A) Schematic representation of pCpGL plasmid constructs. The HLA- E CGI (HLAE CGI) was directionally cloned into pCpGL constructs indicated by an arrow head at the γ’ end. (B) 293FT cells were transfected with indicated pCpGL constructs (panel A) along with a Renilla luciferase (RL) plasmid as a transfection control. Luciferase activity was measured 24 hours post transfection using the Dual Luciferase Reporter Assay (Promega). Representative data of three independent experiments is shown as a fold ratio (FL/RL) of relative light units (RLU) from quadruplicates. (C) The pCpGL plasmid containing the CXCL14 promoter element was incubated with M.SssI methyltransferase (MT) or buffer only (untreated). Samples were subsequently treated with buffer only (uncut), BstUI or McrBC methylation-sensitive endonucleases. Samples were separated in 0.7% agarose gel to verify in vitro methylation. (D) 293FT cells were transfected with M.SssI MT- treated (methylated) or buffer only control (unmethylated) pCpGL constructs (described in panel A) along with an RL plasmid as a transfection control. Luciferase activity was assessed as in panel B. Fold changes to the unmethylated reporter constructs are plotted. P values were calculated by Student’s t test. *p < 0.0005, **p < 0.005, ***p = 0.01.

83 pCpGL-Basic (255). We first determined the promoter and/or enhancer activity of the HLA-E

CGI by amplifying the HLA-E CGI (hg19, chr6:30,434,030-30,434,730) and cloning it into the pCpGL-Basic vector. Several different pCpGL-HLAE-CGI constructs were prepared to test the promoter activity of the HLA-E CGI in forward and reverse orientations (pCpGL-HLAE-

Fwd and pCpGL-HLAE-Rev). Additionally, we tested the enhancer activity of the HLA-E CGI using the EF1α promoter (pCpGL-HLAE-Fwd-EF1α and pCpGL-HLAE-Rev-EF1α) (Fig. 4-

5A). Each construct was transfected into 293FT cells along with a Renilla luciferase vector transfection control. Promoter activity was determined by relative luciferase activity. Our results revealed the strong promoter activity of the HLA-E CGI, showing near 200-fold and

130-fold increases in luciferase activity by insertion of the HLA-E CGI in forward or reverse orientations, respectively (Fig. 4-5B). The promoter activity of the HLA-E CGI was even 3- fold higher than the EF1α promoter.

We next tested if hypermethylation in the HLA-E CGI represses its promoter activity using in vitro DNA methylation. The pCpGL reporter constructs were methylated in vitro using the M.SssI CpG methyltransferase. To verify successful DNA methylation of the pCpGL reporter constructs, the methylated and unmethylated plasmids were digested with restriction enzymes BstUI and McrBC, which cut only unmethylated and methylated CpG motifs, respectively (Fig. 4-5C). Each methylated and unmethylated plasmid was transfected into 293FT cells and relative luciferase activity was measured as described above. The

CpG-free pCpGL-CMV-EF1α plasmid (unaffected by CpG methylation) was used as a negative control (255) and the pCpGL-CXCL14 promoter (repressed by CpG methylation) plasmid and pCpGL-CRE4X plasmid (activated by CpG methylation) were used as positive controls (256, 326). We found that in vitro DNA methylation dramatically decreased the luciferase activity of all HLA-E CGI containing reporter plasmids. As expected, while the luciferase activity of pCpGL-CMV-EF1α was unchanged, the luciferase activity of pCpGL-

CRE4X and pCpGL-CXCL14 were significantly increased and decreased by in vitro DNA

Figure 4-6. High-Risk HPV E7, but not Low-Risk HPV E7, is Sufficient for Downregulation of HLA-E Expression in Normal Keratinocytes.

Figure 4-6. High-Risk HPV E7, but not Low-Risk HPV E7, is Sufficient for Downregulation of HLA-E Expression in Normal Keratinocytes. (A, C-E, and H) NIKS and NIKS derivative cells were fixed, permeabilized, incubated with appropriate antibodies, and assessed by flow cytometry as described in Materials and Methods. (A) HLA-E and HLA-B/C protein expression in NIKS (grey) and NIKS-16 (red) cells. (B-D) NIKS cells stably expressing E7 from HPV6, 11, 16, and 18 (NIKS-6E7, NIKS-11E7, NIKS-16E7, and NIKS- 18E7, respectively) were generated by lentiviral transduction followed by selection in puromycin. (B) E7 gene expression was validated by RT-PCR using specific primers (Appendix A). HLA-E (C) and HLA-B/C (D) expression in NIKS cells (grey) and NIKS cells expressing high-risk (HPV16 and 18, red) or low-risk (HPV-6 and 11, blue) E7 was analyzed by flow cytometry. HLA-E expression in NIKS-16 cells mock treated or treated with 5 μM 5- aza for five days was assessed by flow cytometry (E) and immunoblot (F). (G) MSP was performed with gDNA from NIKS-16 cells mock treated (mock) and treated with 5-aza (5- aza) using specific primers (Appendix B). (H) NIKS-16 cells were treated with 50 U/mL interferon β (IFNβ, red) or interferon (IFN, blue) for 5 days and HLA-E expression was assessed by flow cytometry.

86 methylation, respectively (Fig 4-5D). These results suggest that HLA-E expression is regulated by DNA methylation in the HLA-E CGI.

HLA-E is downregulated by high-risk HPV E7, but not by low-risk HPV E7

To determine if HLA-E protein levels are decreased by gene expression downregulation in HPV-positive keratinocytes, HLA-E protein levels were analyzed in permeabilized NIKS and NIKS-16 cells by flow cytometry. As previous studies have shown that surface expression of HLA-I molecules is frequently downregulated in HPV-positive cancer cells and tissues (341), we also examined HLA-B/C expression in NIKS and NIKS-16 cells. Consistent with our mRNA expression data, protein expression of HLA-E as well as

HLA-B/C was dramatically decreased in NIKS-16 cells compared to NIKS cells (Fig. 4-6A).

As shown above, HPV16 E7 expression is necessary for downregulation of HLA-E gene expression (Fig. 4-2). To test if expression of high-risk E7 is sufficient for downregulation of

HLA-E expression in normal keratinocytes, we generated stable NIKS cell lines expressing

E7 oncoproteins from high-risk HPV genotypes (16 and 18) and low-risk HPV genotypes (6 and 11). E7 expression in each stable NIKS cell line was validated by RT-PCR (Fig. 4-6B) as antibodies detecting E7 from different genotypes are not available. Next, HLA-E protein expression was determined in each NIKS cell line expressing only the E7 gene from different genotypes. The results clearly showed that high-risk HPV16 E7 or HPV18 E7 expression was sufficient for downregulation of HLA-E protein expression in NIKS cells, while low-risk HPV6 E7 or HPV11 E7 expression rather increased HLA-E expression in

NIKS cells (Fig. 4-6C). In contrast, low-risk E7 expression moderately decreased HLA-B/C expression compared to substantial downregulation of HLA-B/C by HPV16 E7 expression

(Fig. 4-6D). These results highlight the distinct functions of high-risk and low-risk E7 oncoproteins in dysregulation of HLA-I expression in host cell.

Because we demonstrated that DNA demethylation at the HLA-E CGI was associated with HLA-E gene expression, we tested if the demethylating agent, 5-aza-β’-

87 deoxycytidine (5-aza) increases HLA-E protein expression in NIKS-16 cells. We found that

5-aza treatment dramatically restored HLA-E protein expression in NIKS-16 cells to a similar level observed in NIKS cells (Fig. 4-6E). Demethylation at the HLA-E CGI with 5-aza treatment was verified by MSP (Fig. 4-6F) and restored HLA-E protein expression is also confirmed by immunoblot (Fig. 4-6G). Additionally, we tested if interferon (IFN) treatment restores HLA-E expression in NIKS-16 cells, because IFNβ and IFN upregulate HLA-E expression (342, 343). Our results showed that HLA-E expression in NIKS-16 cells was highly increased by IFN, but not by IFNβ treatment (Fig. 4-6H). Taken together, our findings suggest that HLA-E expression is downregulated by the HPV oncoprotein E7- mediated DNA methylation and can be restored by treatment with a demethylating agent or

IFN.

Discussion

HPV oncogene E7 expression dramatically alters host gene expression through E2F transactivation by deactivating the pRB tumor suppressor (344). Our previous global gene expression studies using CxCa and HNC patient tissue samples have shown that expression of cell cycle-related genes is highly upregulated in HPV-positive cancers compared to normal tissue and HPV-negative HNC (110, 111). In contrast, estrogen receptor  and the chemokine CXCL14 are dramatically downregulated during HPV-driven cancer progression (111, 326). CXCL14 downregulation in HPV-positive cells is associated with the HPV oncoprotein E7-dependent promoter methylation. Previous studies confirmed that HPV alters global DNA methylation as a mechanism to silence host gene expression; however, these studies used genetically dissimilar HNC tissues and cell lines (333, 345).

Thus, the specific mechanisms of gene expression changes by HPV-directed DNA methylation remain to be uncovered. To address this, we performed parallel gene expression and methylome analyses using homogeneous normal keratinocytes, the native host of HPV.

We found that HPV oncoprotein E7 expression altered immune regulatory pathways involved in IL1 signaling and MHC-I expression. HPV is known to downregulate IL1B and

HPV E6 inhibits proteasome-dependent degradation of pro-IL-1β (346, 347). IL1 signaling is intricately linked to NF-κB and MAPK, which activates TNF family members and IFN expression to recruit immune cells to the site of infection (348). Accordingly, suppression of

IL1 signaling contributes to immune evasion of various viruses including vaccinia virus and

Kaposi’s sarcoma–associated herpesvirus (349, 350). Here, we show that IL1B, IL1R1,

IL1RN, and IL36G (IL1F9) expression is significantly altered in an E7-dependent manner, suggesting an E7 function in host immune evasion through suppression of IL1 signaling

(Appendix E-2B and E-2D).

In our methylome analysis, we found that the HLA-E CGI is hypermethylated in HPV- positive cells in an E7-dependent manner, correlating with downregulation of HLA-E expression. Previous studies have shown that DNA hypermethylation is an effective mechanism of repressing MHC-I and –II gene expression, which is reversed by methylation inhibitors (351, 352). The HPV oncoproteins E5 and E7 are known to interfere with trafficking of MHC proteins to the cell surface; however, any effect of HPV on HLA-E expression was unknown.

MHC expression is modulated by various cellular mechanisms. For example, the

HLA-G promoter contains a series of cis regulatory elements that govern tissue-specific expression (353). Distal promoter elements are scattered throughout MHC gene loci and activate transcription of non-coding RNAs (ncRNAs) that may regulate expression of the

MHC genes (354). One distal regulatory element has been characterized 25 kb upstream of the HLA-DRA promoter (355). Similarly, we here report that a distal CGI located 23 kb upstream of the HLA-E ORF exhibits strong promoter activity. We showed that the HLA-E

CGI remains hypomethylated in normal (NIKS) cells (β = 1β.4%), but hypermethylated in

HPV-positive (NIKS-16) cells (change in β >50%) and its methylation is linked to HLA-E

89 downregulation. Thus, it is possible that HPV E7-induced DNA methylation silences promoter activity of the distal HLA-E CGI, or silences expression of a regulatory ncRNA which modulates the HLA-E regulatory elements, as previously hypothesized (354).

Additionally, DNase I hypersensitivity (indicative of a relaxed chromatin structure), H3K27 acetylation (a histone marker synonymous with active transcription (356)), conserved TF binding sites, and small transcribed RNAs from this region (Fig. 4-4C, Appendix E-4) indicate that the HLA-E CGI is an active regulatory region for HLA-E expression.

It has been suggested that HPV E7 inhibition of STAT1 activation represses TAP1 transcription, which leads to decrease of surface expression of MHC-I molecules (357).

However, the TAP1 mRNA levels were not changed in either NIKS-16 or NIKS-18 cells compared to NIKS and NIKS16∆E7 cells (data not shown). This observation suggests that downregulation of HLA-B/C protein expression in NIKS-16 and NIKS-18 cells might not be mediated by E7-induced TAP1 downregulation. Here, our study showed that HLA-E expression is epigenetically regulated by DNA methylation and is restored by treatment of a demethylating agent. While methylation at the HLA-E CGI may not directly affect expression of HLA-E, our data suggest that HPV-directed DNA methylation can repress HLA-E expression. In contrast, there was no significant DNA methylation identified by our arrays in the HLA-B and -C genes by HPV16 or HPV18. Thus, HLA-B and -C downregulation is likely caused by HPV E7 is mediated by other unknown mechanisms, as both high- and low-risk

E7 suppresses HLA-B and -C expression while only high-risk E7 downregulates HLA-E expression. Additionally, the MHC-I transactivator NLRC5, which is necessary for MHC-I gene expression and repressed by DNA methylation in various cancers (358, 359), was downregulated in NIKS-16 and NIKS-18 cells (Appendix E-5B) when compared to NIKS and NIKS-16ΔE7 cells. This may in part explain E7-dependent downregulation of HLA-B/C expression, while further studies are essential to fully understand the mechanism.

HLA-E is constitutively expressed in various tissues (360). The NK cell inhibitory receptor NKG2A and the activating receptor NKG2C were initially found to bind HLA-E on the cell surface (196). HLA-E presentation of MHC leader sequence peptides to NK cells leads to NKG2A-mediated inhibition of NK effector function (196, 361). Additionally, HLA-E presentation of stress-inducible heat shock protein 60 peptides interferes with NKG2A recognition, leading to NK cell-mediated killing of stressed cells (362). In addition to regulating NK cells for innate immune responses, HLA-E also acts on CD8+ T and natural killer T (NKT) cells to modulate adaptive immune responses (199, 200). HLA-E activates or inhibits NK and CD8+ T cell functions by presenting a narrow range of viral and bacterial antigens (363). Of note, a viral-derived peptide presented by HLA-E on HIV-1-infected CD4+

T cells activates NK cells and induces cytolysis of virus-infected cells (364). In contrast, a viral peptide presented by HLA-E on the surface of hepatitis C virus (HCV)-infected cells inactivates NK cells but elicits HLA-E-restricted CD8+ T cell responses (365). These findings imply that HLA-E may also present HPV-derived peptides to NK or CD8+ T cells, leading to lysis of the HPV-infected cells. We have previously found that high-risk HPV E7 significantly reduces NK and CD8+ T cell infiltration into the HPV-positive tumor microenvironment through CXCL14 downregulation by its promoter hypermethylation (326). Together, these results suggest that HPV may employ multiple mechanisms to evade antiviral NK and CD8+

T cell activity by dysregulating DNA methylation.

In contrast, HLA-E is upregulated in several cancers, including HNC, CxCa, breast, rectal, colon, and ovarian cancers (129). Accordingly, it has been proposed that the high levels of HLA-E may inhibit NK cell activation caused by downregulation of classical HLA-I expression on cancer cells. The better survival rate of ovarian cancer patients with infiltrating

CD8+ T cells disappears when NKG2A signaling is activated by high HLA-E expression

(366). HLA-E expression is also linked to poor clinical outcome and low overall survival in breast and colon cancer patients (367, 368). Further, knockdown of HLA-E expression

91 enables NKG2D-mediated lysis of glioma cells by NK cells (369). Our results showing that high-risk HPV E7 downregulates HLA-E expression in normal keratinocytes imply dual roles of HLA-E that induces antiviral immunity in normal cells but suppresses antitumor immunity in cancer cells.

Indeed, previous studies have shown that HLA-E plays an important role in antiviral immune responses. HLA-E presentation of viral peptides elicits cytotoxic responses of NK and CD8+ T cells that kill virus-infected cells (208, 364, 365). We report here that high-risk

HPV E7 significantly downregulates HLA-E expression in normal keratinocytes, while low- risk HPV E7 increases HLA-E expression. The epigenetic repression of HLA-E expression by E7 suggests a previously undescribed immune evasion mechanism employed by high- risk E7, but not by low-risk E7. We also show that HLA-E expression can be restored through treatment with the demethylating agent, 5-aza, as well as IFN. This may provide a new therapeutic approach to treat HPV-positive lesions by activating the HLA-E mediated antitumor immune responses of NK and CD8+ T cells.

CHAPTER V

DISCUSSION AND FUTURE DIRECTIONS

Discussion

HPV-associated cancers are known to carry significantly fewer oncogenic mutations than HPV-negative tumors in the same anatomical region ( ~5 vs. >20) (296). This is an important observation which suggests that HPV expresses viral factors that replace cellular oncogenic events. Studies have revealed that E7 may act as such a factor: continuous E7 expression is required for cancer maintenance (297, 298), suggesting that E7 manipulates cellular processes leading to and maintaining oncogenic transformation. Importantly, E7 is known to dysregulate host gene expression through abrogation of the cell cycle regulator pRB, as well as manipulating multiple epigenetic regulatory factors including HDACs, HATs, and DNMT1 (124–127). Regulation of host gene expression through modulation of the epigenome is an efficient mechanism to promote a favorable environment for viral replication and evasion of immune surveillance. Here, we show that, while many other chemokines are upregulated over the course of cervical cancer progression, CXCL14 is precipitously downregulated epigenetically in an E7-dependent manner (Figure 3-1A and 3-

1C). Additionally we show that HPV E7 expression induces large-scale changes to the host methylome and transcriptome. One additional target suppressed by DNA methylation (as described here) is the MHC-I subunit, HLA-E (Figure 4-6). Together, these observations reveal important mechanisms by which HPV E7 evades immune recognition and disrupts tissue homeostasis for viral persistence.

CXCL14: The Epithelial Homeostatic Chemokine

Chemokines and Homeostasis

Homeostasis is a characteristic of structured cellular architecture, which involves maintenance of healthy tissue as well as repairing damage. Chemokines and their receptors are known to both regulate tissue development and maintain homeostasis. For instance,

CCR7 and CXCR5 regulate trafficking of T and B cells to LN and spleen organs, and CCR7 and CXCR5 are necessary during development of lymphoid organs (370–372). CCL25 is another example, which is constitutively expressed by gut epithelium and recruits the α4β7- positive effector T cell subset, via CCR9, into the mucosa of the small intestine for surveillance (373, 374). While homeostatic chemokines are required for tissue development and immune surveillance, dysregulation of chemokines and their receptors can lead to cancer progression and metastasis. The well-studied CXCR4/CXCL12 signaling axis is an example, which is frequently dysregulated in breast cancer, colon cancer, lung cancer and others (375–377). Inhibition of CXCR4 signaling in cancer cells results in reduced metastasis and cancer cell growth, highlighting a homeostatic role for CXCR4/CXCL12 signaling. Additionally, CXCR4 and CCR7 overexpressing breast cancer cells tend to metastasize to CXCL12 and CCL21 (ligands for CXCR4 and CCR7, respectively) - expressing tissues due to increased invasive potential (378). This again highlights how dysregulated homeostatic chemokine signaling plays a role in cancer development and metastasis.

It is interesting to consider the opposing nature of CXCL14 downregulation in HPV- positive cancer development compared to dramatic upregulation of chemokines and chemokine receptors observed in other cancers. For instance, chemokine receptors CXCR3

(379), CXCR4 (375), CCR7 (380), CCR9 (381), and CCR10 (382) are all overexpressed in various cancers (i.e. breast, melanoma, prostate, leukemia). Overexpression of chemokine receptors (involved in cell motility) has led some to hypothesize that cancer cells may use these receptors as a mechanism to “’hijack the chemokine receptor-mediated cell migration highway” in order to achieve metastasis (300). Others have shown that chemokines CXCL1

(GRO-1) and CXCL8 (IL-8) are overexpressed in cancers, leading to increased proliferation

(383, 384). Conversely, we show that CXCL14 expression is nearly completely abrogated by

HPV E7, which leads to tumor growth and immune evasion. We therefore describe

94 repression of chemokine expression as a mechanism of immune escape (exclusion of NK and T cells from lesional tissue) rather than a homing mechanism for metastasis or a mechanism of unlimited growth potential.

Similar to our results, the homeostatic keratinocyte chemokine CCL27 is significantly downregulated in actinic keratosis skin tumors (385). Reduction of CCL27 expression was linked to reduced T cell infiltration into the TME and increased tumor growth in mice (385).

CXCL14 expression in epithelial tissues may act in a similar manner to maintain immune surveillance, suppression of which allows for sustained tumor growth. Inhibiting epithelial cell migration (Figure 3-3) and directing immune surveillance (Figure 3-5) represent dual roles for tissue homeostasis achieved by CXCL14, in turn compounding the tumorigenic and metastatic effects of HPV E7-mediated CXCL14 repression.

Fine-Tuning CXCL14 Expression for Homeostasis

CXCL14 is constitutively expressed in epithelial organs and is thought to be an important homeostatic chemokine (223, 280, 300), due to its roles in development (302), immune cell recruitment (282), and regulating angiogenesis (278). Further, CXCL14 is known to directly bind and inhibit IL-8, blocking angiogenesis (386). Control of angiogenesis is key factor in tissue regulation and maintenance and therefore solidifies the role of

CXCL14 as a homeostatic chemokine. A native receptor for CXCL14 has not been discovered; however, some reports show that CXCL14 can act as a decoy ligand for the

CXCR4/CXCL12 signaling axis (288) which is important in cell growth and metastasis in cancer. These claims are however disputed, which will be discussed later. Thus, the exact mechanisms behind the function(s) of CXCL14 remain to be elucidated.

In cancer, CXCL14 expression is frequently dysregulated. CXCL14 expression is upregulated in prostate and pancreatic cancers (387, 388) while others have shown dramatic downregulation in prostate, lung, oral, and breast cancers (389–391). Further, in our syngeneic mouse model we observed that, while restoration of Cxcl14 to near

95 physiological expression levels resulted in tumor clearance in over 50% of mice (Figure 3-

4), preliminary data revealed that dramatic overexpression of Cxcl14 did not impart any ability for tumor clearance (data not shown). Indeed, when mice were injected with

MOE/E6E7 cells overexpressing Cxcl14 by almost 1,000-fold, tumors grow with near identical kinetics to the MOE/E6E7 cells, which have repressed Cxcl14 expression. This observation may be used as a basis for further studies elucidating the effect of differing concentrations of CXCL14 on tissue homeostasis and tumor growth.

The optimal concentration of proteins or signaling molecules can be termed a

“Goldilocks” concentration. Abundances of these signaling proteins and molecules outside the Goldilocks concentration can have adverse effects are observed, occurs during development and homeostasis. For instance, a Goldilocks concentration of retinoic acid is required for inner ear development (392), and optimal expression of the E3 ubiquitin-protein ligase BRE1 is necessary for epigenetic control of transcription (393). Our observations suggest an optimal amount of Cxcl14 expression is required in order to maintain homeostasis, disruption of which contributes to cancer development. Dysregulation of

CXCL14 expression (either up- or downregulated) in cancer is therefore a candidate to explore to understand the mechanisms of cancer development and progression.

These observations may be important if development of CXCL14 as a therapeutic agent is perused. For instance, in our mouse syngeneic tumor model, HPV-positive tumors may be injected with variable levels of CXCL14 to assess tumor clearance. Because

CXCL14 appears to work at an optimal Goldilocks concentration, a controlled release gel or polymer may be necessary when administering CXCL14 to tumors for concentration control.

Controlled release of chemotherapeutics is a current strategy employed by oncologists and is therefore an attractive candidate for future clinical applications for CXCL14 (394).

CXCL14, Wound Repair, and Infection

We have shown that CXCL14 regulates immune surveillance as well as epithelial cell migration, highlighting its role as a homeostatic chemokine. Exciting additional data show that CXCL14 overexpression in keratinocytes make cells remarkably refractory to HPV infection (Figure 5-1). HPV infection is dependent upon viral capsid interactions with the basement membrane at the basolateral side of epithelial cells (395). As CXCL14 inhibits epithelial cell migration and HPV infection, it is tempting to speculate that, as a homeostatic chemokine, CXCL14 encourages the formation of tight and other cellular junctions (such as gap junctions and desmosomes), disallowing HPV to traverse to the basolateral side of the cell for successful infection. This is further supported by the observation that infectivity is drastically reduced in mostly confluent monolayers of cells (even cells that are still dividing in a monolayer, such as HEK 293 cells, lab observations not shown), suggesting that the peripheral cells are those that are most susceptible to infection. This potential function for

CXCL14 could easily be assessed using immunofluorescence in 3D raft cultures with

CXCL14 supplemented medium, or using exogenously overexpressed or knocked-out

CXCL14. During wound repair, epithelial cells are known to alter their gene expression profiles (including cytokine expression) and cells nearest the injury migrate into the wound and proliferate in order to return the tissue to its homeostatic state (396, 397). CXCL14 expression in wound healing would therefore be exciting to observe: is CXCL14 downregulated near the wound periphery, allowing for both cell migration and HPV infection? Our results from this study indicate a potential function for CXCL14 in epithelial tissue homeostasis as well as HPV infection.

Figure 5-1. CXCL14 Expression Reduces HPV Infectivity in Keratinocytes. CaSki cells were stably transduced with lentiviral vectors expressing human CXCL14 (CaSki CXCL14) or an empty vector control (CaSki) and selected for one week using Blasticidin. Cells were seeded onto 96 well plates at a density of 7 x 103 cells per well and infected with HPV16- LucF reporter virus at a 1:200 ratio of virus to media. 48 hours post infection luciferase activity was assessed and plotted as relative light units (RLU) per second. Student’s T test was used to determine p = 0.0109. Data are averaged from three combined biological replicates of 4 technical replicates each.

CXCL14 and Metastasis

Previous studies have shown that CXCL14 is downregulated in a variety of carcinomas, including breast, prostate, cervical, and oral cancers (280–283, 286, 301, 302).

Because CXCL14 is often dysregulated in carcinomas, it is likely that CXCL14 plays an important role specifically in epithelial tissue homeostasis and disruption of which may contribute to cancer progression. Consistently, we show that downregulation of CXCL14 significantly increases epithelial cell motility in scratch and transwell cell culture models

(Figure 3-3). Downregulation of CXCL14 in carcinomas may therefore contribute to cell motility, invasion, and metastasis.

CXCL14 appears to work oppositely regarding the motility of epithelial cells with respect to the chemokines mentioned. For instance, in breast cancer, when CXCR4 is dramatically upregulated in conjunction with CXCL12 epigenetic downregulation, cancer cells are more likely to metastasize to other organs expressing CXCL12 (389, 398, 399). We have observed no evidence that CXCL14 can act as a chemoattractant to epithelial cells by transwell (Figure 3-3) and have shown that CXCL14 can only act as a chemoattractant to certain immune cells (Figure 3-7). While the receptor for CXCL14 is unknown, other chemokine receptors are G-protein coupled receptors (GPCR), which are known to stimulate cell motility during development as well as cancer (400). Because CXCL14 can act as a chemoattractant to immune cells but is inhibitory to epithelial movement, it is tempting to speculate that CXCL14 may act through cell type-specific receptors.

Determining the effect of CXCL14 repression on metastasis would be a feasible future direction, which may have further clinical applications. In epithelial cells, for instance, downregulation of CXCL14 leads to increased epithelial cell motility. It would therefore be interesting to determine if CXCL14 expression affects invadopodia formation or dissociation of tight junctions, as is observed in cells highly expressing CXCR4. Further experiments

Immune Evasion and Viral Persistence

Persistence through Deregulation of CXCL14 and CXCL12 Signaling

There is no known native receptor for CXCL14, although some have speculated that it can act as a “fine tuning” mechanism for the CXCR4/CXCL1β signaling axis, a pathway important in development and immune signaling (Figure 5-2) (401, 402). This function of

CXCL14 is attractive as CXCL12 is involved in tumor cell migration and metastasis (288).

CXCL12 is overexpressed in cancer associated fibroblasts and is known to recruit endothelial cells for angiogenesis, and enhanced CXCL12 signaling has been shown to promote tumor growth and metastasis (403–405). If CXCL14 represses CXCL12 signaling, it would therefore be advantageous for HPV to repress CXCL14 activity in order to promote cell cycle progression for viral DNA synthesis, and angiogenesis for nutrients in order to persist. Unfortunately the interaction between CXCL14 and CXCR4 is contested and remains to be resolved (406). Nonetheless, the potential remains that CXCL14 acts to regulate CXCR4/CXCL12 signaling and should be further explored for a better understanding of how CXCL14 repression may function to aid in viral persistence and immune evasion. This may ultimately lead to identification of therapeutic targets for the myriad cancers that downregulate CXCL14.

CXCL14 Repression Dysregulates Innate and Adaptive Immune Responses to Cancer

The anti-tumor effect of CXCL14 has been shown by our and other groups to be multifaceted. Restoration of CXCL14 expression in lung cancer mouse xenografts induces tumor necrosis (389), and restored expression of CXCL14 in a nude mouse xenograft model showed slowed human HNSCC tumor growth (390). Additionally, expression of CXCL14 in human OSCC cells xenografted into severe combined immunodeficiency (SCID) mice

(completely lacking functional T and B cells) suppressed tumor growth, suggesting a pivotal role for innate immune clearance of the xenograft (407). CXCL14 was also shown to inhibit

100

Figure 5-2. CXCL14 acts to “Fine Tune” Signaling through the CXCR4/CXCL12 Signaling Axis. Canonical (“normal”) CXCR4/CXCL1β signaling (center) regulates stem and immune cell movement and has been implicated in tumor growth and metastasis. As a tuning mechanism, CXCL14 (left) is proposed to bind CXCR4, which is internalized and made unavailable for signaling. Additionally, CXCR7 has been shown to bind and sequester CXCL12 as an additional regulatory layer.

101 proliferation and metastasis of breast cancer cells, suggesting yet another tumorsuppressive role for CXCL14 (391). Dysregulated expression of CXCL14 has been shown to be a prognostic indicator for overall patient survival in colorectal cancer, suggesting that dysregulation of CXCL14 expression is indeed important for cancer development (408).

Here, we show that restored expression of Cxcl14 in a syngeneic mouse tumor model restores NK, CD4+ T, and CD8+ T cell populations in the TDLN to near-physiological levels

(Figure 3-4 and Figure 3-6). These data are paired with the observation that tumors were cleared in about 50% of the mice injected with clone 8 and clone 16 MOE cells re- expressing Cxcl14 to near-physiological levels (Figure 3-4). It is important to observe consistent tumor growth from multiple clones re-expressing Cxcl14 in order to rule out a clonal effect on tumor growth, or differential presentation of a tumor specific antigen, which would result in differences in tumor growth or clearance. When Cxcl14 re-expressing cells were injected into Rag1-/- mice, which do not develop functional T or B cells due to a deficiency in the Rag1 recombinase, tumor growth is stymied only by a week, at which point tumors grow with similar kinetics to MOE/E6E7 cells (Figure 3-4). Delayed tumor growth in the absence of an adaptive immune response has not been shown before, and strongly suggests that innate recognition of the tumor is a pivotal first step in order to fully eliminate the tumors however it is not sufficient to clear tumors. While CXCL14 is involved in NK cell migration, the effect of CXCL14 on CD4+ and CD8+ T cells has been previously undescribed. Our results indicate that CXCL14 expression is intimately involved in both innate and adaptive immune cell regulation.

As we have observed decreased NK cells near tumors with repressed Cxcl14 expression (131), and Cxcl14 is known to regulate NK cell movement during development

(302), it is tempting to speculate that NK cells are the major factor here linking an innate and adaptive immune response against HPV-positive tumors. NK and T cells are both known to regulate each other’s functions through direct (i.e. receptor interaction) or indirect (i.e.

102 chemokines or cytokines) signaling (409, 410). NK cells function to identify tumor cells primarily through interactions (or lack of interactions) with their NKG2D receptor (411).

Briefly, tumor specific antigens are recognized by NKG2D, resulting in co-stimulation of NK and CD8+ T cells, as well as increased NK cell adhesion to the cancer cell through the

CD226 adhesion molecule followed by cytolysis of the cancer cell (411, 412). Of note, some cancers secrete soluble MHC class I polypeptide-related sequence (MIC) A and B protein

(cleaved from the surface by MMP9), a ligand for NKG2D, resulting in downregulation of

NKG2D and CD226. MICA/B expression can therefore be used as a prognostic biomarker for some cancers (413, 414). Downregulation of NK cell receptors CD226 and NKG2D has been shown to reduce NK cell-mediated cytolysis of cancer cells (412, 415). The CD226 and NKG2D proteins are therefore vital to identify and kill cancer cells. Once activated, NK

+ cells secrete IFN which skews maturation of CD4 T cells to Th1 cells, and induces maturation of DCs which in turn activate CD8+ T cells for tumor clearance (416–419). Both

NK and CD8+ T cells are well known as effector killer cells capable of eliminating virus- infected cells as well as cancer cells (296–299). NK cell activation induces CD8+ T cell responses through priming DCs, further suggesting that NK cells may be the link between innate and adaptive immunity to induce antiviral and antitumor CD8+ T cell responses (221,

299).

Here, we show that tumor growth from Cxcl14 re-expressing cells is repressed for a short time in T cell-deficient Rag1-/- mice but tumor growth later returns to kinetics seen in

Cxcl14-repressed (MOE/E6E7) cells (Figure 3-4). Taken together, we hypothesize that

Cxcl14 re-expressing cells recruit NK cells to the TME to kill tumor cells resulting in inhibited tumor growth. In the context of a fully functional immune system (for instance our syngeneic

C57BL/6 mouse model), activated NK cells then secrete IFN which matures DCs, skews

+ + maturation of CD4 T cells to Th1 cells, and activates CD8 T cells, resulting in an influx of these immune cells and successful tumor clearance. In Rag1-/- mice, however, an adaptive

103 immune response is not elicited, the tumor cells proliferate faster than the NK cells can eliminate them, and tumor growth is only shortly delayed.

In our Rag1-/- study, we observed that an innate immune response is unlikely sufficient to clear Cxcl14 re-expressing tumors; however, we did observe an increase in NK cells in the tumors of these mice. A potential mechanism for the inability of NK cells to clear tumors can be hypothesized when assessing MICA/B expression in HPV-positive cancer progression. Expression of MICA/B (ligands for NKG2D which downregulate NKG2D and

CD226 expression) in cervical cancer increase significantly over the course of cancer progression, but an increase in expression is not observed in an early infection model comparing NIKS-16 (HPV16-positive keratinocytes) cells to NIKS (normal uninfected keratinocytes) (Figure 5-3). Additionally, we observed an increase in MMP9 expression in

HPV-positive cancers and HPV-positive NIKS-16 cells. MMP9 solubilizes MICA/B by cleaving it from the cell surface. Delayed tumor growth in our Rag1-/- model may therefore be due to buildup of soluble MICA/B, desensitizing NK cells and leading to tumor development. This suggests an effective immune escape mechanism (in addition to downregulation of Cxcl14) utilized by HPV-positive cancer to avoid detection and clearance by NK cells, and this effect is increased over the course of persistence and cancer progression. The abundance of NKG2D receptors on NK cells in Rag1-/- mice, and the abundance of MICA/B in early and late stage Rag1-/- tumors could therefore be the basis for follow-up studies. This hypothesis is further supported by data from Toro-Arreola et al. who showed that other cervical cancer cell lines express high surface levels of MICA/B (420) and

Jimenez-Perez et al. found that NK cells downregulate NKG2D surface expression after contact with HPV-positive cells with upregulated MICA/B expression (421). Additionally, others have observed an increase in soluble MICA expression in CIN1 lesions compared to normal cervical tissue (422). Downregulation of NK cell NKG2D receptor through

104

Figure 5-3. MICA and MICB Expression in HPV-Positive Cancer and Keratinocytes. Relative expression levels of MICA and MICB are shown. (A) Gene expression data were analyzed from a global gene expression study of 128 cervical tissue samples in different disease stages: normal (n = 24); low-grade lesion (n = 36); high-grade lesion (n = 40); and cancer (n = 28) (111). (B) Gene expression data from triplicates of global gene expression arrays from NIKS and NIKS-16 cells are shown. Normalized fluorescence intensities (log2) of gene expression from each group are graphed. Data for panel A were collected by Dohun Pyeon, PhD, University of Colorado.

105 overexpression of soluble MICA/B in HPV-positive cancer cells may therefore act as an important immune evasion mechanism. This hypothesis begs the question: would restored

Cxcl14 expression in late-stage (equivalent to CIN3 or a high-grade HPV-positive lesion)

HPV-positive lesions (which are expressing high levels of MICA/B and therefore may have less-functional cytotoxic NKG2D-expressing NK cells) be sufficient to resolve tumors?

Perhaps Cxcl14 expression was sufficient to clear HPV-positive tumors on our syngeneic mouse model due to the fact that Cxcl14 was present immediately upon injection and repression of NK cell function had not yet developed. Follow-up studies to further delve into the mechanisms of HPV-induced repression of innate and adaptive immune responses should therefore be conducted.

Potential for Therapeutic Targets

Our study suggests that CXCL14 plays an important role in antitumor immune responses to clear HPV-positive tumors. CXCL14 is a small, secreted protein that carries the potential to be used as a therapeutic agent. Elucidating the mechanisms utilized by

CXCL14 to recruit NK and T cells (i.e. receptors, intermediates, co-signaling molecules etc) would be an exciting frontier for further studies. Additionally, identification of the native

CXCL14 receptor(s) would provide for druggable targets to potentially enhance or fine-tune

CXCL14 function. Thus, further studies of the effects of CXCL14 on NK and T cells may provide a novel means of anti-cancer immunotherapy to treat HPV-positive cancers.

E7-Directed DNA Methylation as a Viral Strategy for Immune Evasion and Persistence

Expression of CXCL14 is known to be negatively regulated by promoter hypermethylation, and we and others have shown that demethylation at the promoter activates CXCL14 expression (131, 423). Additionally, we show here that E7-dependent methylation at the HLA-E CGI represses expression of HLA-E (Figure 4-6). As discussed previously, DNA methylation is an important marker for cancer development and provides for a mechanism by which HPV may modulate host gene expression. Here we show that the

106

CXCL14 promoter is hypermethylated in an E7-dependent manner. Noting that CXCL14 is considerably downregulated in cervical cancer, we hypothesized that E7, through an interaction with DNMT1, hypermethylates the CXCL14 promoter in order to suppress gene expression. CXCL14 has been shown to be an important homeostatic chemokine, but also a potent recruiting factor for NK and T cells. Additionally, HLA-E is known to regulate NK and

CD8+ T cell cytolytic activity against virus-infected cells and tumors cells. Thus, repression of

HLA-E expression may act to aid in viral persistence leading to cancer development. The action of directing methylation at specific host genes may therefore be an important mechanism employed by HPV to evade immune detection and persist.

Viral Persistence Leads to CXCL14 Methylation and Repression.

Previous work has demonstrated that persistence of HPV infection and prolonged viral gene expression are major risk factors for HPV-associated cancer progression and development, but the precise mechanisms of this action remain poorly understood (148–

151). Prolonged viral oncoprotein expression is one potential mechanism of such action.

The HPV oncogene E7 expression dramatically alters host gene expression through E2F transactivation by deactivating the pRb tumor suppressor (Figure 1-3) (344). Our previous global gene expression studies using CxCa and HNC patient tissue samples have shown that expression of cell cycle-related genes is highly upregulated in HPV-positive cancers compared to normal tissue and HPV-negative HNC (Figure 4-1) (110, 111). In contrast, estrogen receptor alpha and the chemokine CXCL14 are dramatically downregulated during

HPV-driven cancer progression (111, 326). With many of the dysregulated genes discussed, changes in expression occur gradually over time suggesting that long-term expression of viral genes during persistence is key (Figure 3-1, Appendix D).

Here we show that CXCL14 downregulation in HPV-positive cells is associated with

HPV E7-induced promoter methylation (Figure 3-1). Matching our in vivo tissue data showing a gradual decline in CXCL14 transcription over the course of cervical cancer

107 progression, we show in our in vitro W12 cell culture model for cervical cancer progression that methylation at the CXCL14 promoter also increases gradually over the course of disease progression (Figure 3-2). We therefore hypothesize that E7-directed methylation at the CXCL14 promoter is a mechanism of repression and that methylation slowly accumulates over time as HPV persists. Further supporting this hypothesis, we show that short-term stable expression of high-risk E7 is sufficient to slightly downregulate CXCL14 expression but only at ~25% (Figure 3-1E). This is in comparison to long term E7 exposure, as observed in the W12 cells (Figure 3-1B), which exhibit almost complete repression of

CXCL14 in W12GPXY cells compared to W12E cells. Finally, we show that in HNSCC patient sample data taken from TCGA, CXCL14 mRNA expression is inversely correlated to methylation at the CXCL14 promoter, and that this hypermethylation is dependent on HPV- associated cancers (Appendix D-3). This shows that gradual increases in methylation at the

CXCL14 promoter proportionally reduce CXCL14 expression, consistently suggesting that long term exposure to E7 is required for dramatic repression of CXCL14. Taken together, this suggests that persistent expression of HPV E7 is a critical factor in repression of

CXCL14 expression which is an efficient mechanism of immune evasion.

Epigenetic modifications directed by HPV E7 for virus persistence are reminiscent of similar strategies employed by other viruses in order to evade immune detection. For instance, EBV induces methylation of its own genome during latency by recruiting methyltransferases to promoter elements in order to silence gene expression (424, 425).

EBV-directed hypermethylation of host and viral DNA results in extensive epigenetic modifications which are required for viral latency and subsequent reactivation. Additionally,

EBV encodes two proteins, LMP1 and LMP2A which upregulate host DNA methyltransferases which are thought to hypermethylate tumor suppressor promoters (426).

Similarly, HPV E7 upregulates DNMT1 expression and likely directs methylation to host promoters in order to evade immune detection. Modified host epigenomes are not unique to

108

EBV and HPV: in fact, this is a common strategy used by other tumor-associated viruses, such as hepatitis C virus, hepatitis B virus, Merkel cell polyomavirus (426).

It is interesting to consider why E7-directed DNA methylation at the CXCL14 promoter is inefficient and takes decades to accumulate: if repression of CXCL14 is advantageous to virus immune evasion, why is promoter methylation slow to accumulate?

Perhaps we are seeing an early gain-of-function trait obtained by E7 and throughout evolution CXCL14 promoter methylation may become more efficient and more quickly established in early HPV infection. Determining the exact mechanism of E7-directed promoter hypermethylation (E7 domains involved, necessary cellular factors, and site specific methylation) would provide clues to understand the origin of this trait over the course of evolution. It would also be appealing to assess whether low-risk HPV are capable of interacting with DNMT1, and direct DNA methylation.

HPV Persistence through Tolerance

HPV is capable of imparting immunoevasive properties to infected host cells. We have shown that expression of many chemokines are increased throughout HPV-associated cancer progression (Appendix D-1). Additionally, CIN3 lesions exhibit mild inflammation

(209) however immune cells are not capable of eliciting a response against the infected tissue. LC and DCs are excluded from HPV-positive tissue, limiting host immune recognition

(427). NK and CD8+ T cells are capable of directly eliminating HPV-positive cells; however, detection can be evaded by HPV (428, 429) through mechanisms only partially understood.

Others have shown that E7 expression alone is sufficient to confer immune tolerance to tissue in mouse xenograft models (430); however, the mechanisms of this action are not well understood. Additionally, immunizing mice against E7 elicits a poor CD8+ T cell response in E7 transgenic mice, but elicits a strong CD8+ T cell response when immunization is followed by tumor challenge (431). These data suggest that E7 expression in stratified epithelial tissue is sufficient to evade surveillance of cytotoxic T cells, but E7-

109 expressing cells in circulation (i.e. tumor cell suspension) can be recognized and destroyed.

One mechanism of this action is thought to occur through mast cell-mediated tolerance. E7- expressing cells upregulate CCL2 and CCL5 chemokine expression, recruiting mast cells and inducing tolerance (432). Mast cells work with regulatory T lymphocytes (Treg) to generate an immunosuppressive environment, and in mast cell-deficient mice, E7- expressing cells are recognized and killed by CD8+ T cells (433). Additionally, HPV is dysregulates the IL-10 and TGF-β signaling networks which together create a local immunosuppressive environment for virus persistence (214). Here we show an additional mechanism of immune exclusion from HPV-infected tissue, through E7-directed repression of CXCL14 expression. We show that CXCL14 can directly recruit NK, CD4+ T, and CD8+ T cells and that repression of CXCL14 may potentially induce immune tolerance simply by excluding immune detector cells from exploring the HPV-positive TME. With HPV infections and cancer progression lasting years to decades, HPV has developed complex and potentially redundant processes for immune evasion and immune tolerance, some of which remain to be fully explored.

Viral Persistence through Downregulation of HLA Expression Early in Infection

Previous studies have shown that HPV alters global DNA methylation, which may be used as a potential mechanism to alter host gene expression (333, 345). This mechanism is therefore unlikely to be isolated to CXCL14 repression and may be used in a large scale as an immune evasive strategy. Here, we performed parallel global gene expression and methylome analyses to identify key host factors and pathways altered by HPV-mediated

DNA methylation in human keratinocytes using our NIKS culture model for early HPV infection. We found that most major histocompatibility complex class I (MHC-I) molecules are downregulated in an E7-dependent manner (Figure 4-2). Further, non-classical HLA-E, which regulates NK and CD8+ T cells, is significantly downregulated by E7-mediated hypermethylation in a distal regulatory CpG island (CGI). These results suggest that HPV

110

E7-mediated DNA methylation induces host immune evasion by downregulating HLA-E expression.

Viral peptides loaded onto HLA-E and presented to NK and CD8+ T cells have been shown to elicit a cytotoxic response (208, 364, 365). HLA-E, therefore, represents an important host defense mechanism for both an innate and adaptive immune response.

While downregulation of MHC subunits may lead to evasion of NK cell (HLA-E specifically, depending on antigens presented) or CD8+ T cell detection (HLA-A through -E), it may expose the cell to NK cell-mediated “missing-self” cytolysis. This risk is potentially avoided by HPV through downregulation of CXCL14. We have shown that CXCL14 can act as a direct chemoattractant to NK cells, and they are excluded from the HPV-positive TME.

Additionally, preliminary data show that MICA/B are upregulated along with MMP9, which cleaves MICA/B from the cell surface to generate a soluble ligand which downregulates

NKG2D expression on NK cell surfaces. Therefore, the combination of CXCL14 downregulation and MICA/B upregulation may represent a combined strategy for evasion

NK cell mediated lysis due to downregulation of MHC surface expression or “missing self”.

Here, we show that high-risk HPV E7 can successfully downregulate both mRNA and protein expression of HLA-E. Low-risk HPV6 and 11 E7 proteins appear to upregulate

HLA-E expression, suggesting differences in immune evasive capacities between high-risk and low-risk HPV genotypes. HPV6 and 11 are low-risk mucosotropic genotypes while 16 and 18 are high-risk; therefore, the discrepancies observed in E7 function may partially explain differences in the ability of high- and low-risk viruses to persist and develop cancer.

The dramatic reduction of HLA-E expression by high-risk E7 makes it tempting to speculate that HLA-E is capable of eliciting a potent immune response against HPV-infected cells and therefore must be subverted in order for viral persistence. As HPV downregulates expression of MHC-I α-subunits both transcriptionally and post-transcriptionally, it is then possible that class-I HLA leader peptide sequences are not available to load onto HLA-E to

111 deactivate NK cells, which would result in HLA-E degradation and NK cell-mediated lysis. It is possible that HPV encodes a decoy HLA leader peptide which is processed and presented by HLA-E, however the dramatic E7-mediated downregulation of HLA-E suggests that HPV negates HLA-E antigen presentation to NK and CD8+ T cells by another mechanism (downregulation). We could then speculate that HPV must downregulate HLA-E expression due to HPV’s inability to present decoy peptides, and HLA-E’s potential to present HPV-specific viral peptides to NK and CD8+ T cells. It would be attractive to assess antigens presented by HLA-E HPV-infected cells through an MHC tetramer assay. If HLA-E is capable of presenting a viral peptide to NK and CD8+ T cells which can elicit killing of virus-infected tumor cells, this could be used as a therapeutic tool.

Follow-up experiments may examine the effect of restoration of HLA-E expression in an early infection mouse model using the murine HLA-E ortholog, H2-Qa1, which has also been shown to act as a ligand for mouse NKG2A receptors on NK cells (434). We have previously shown that an influx of NK cells coincides with the clearance of HPV-positive tumors. We therefore suspect that NK cell recognition of HPV-positive cells through HLA-E may lead to immune clearance. It is tempting to speculate that restoration of CXCL14 and

HLA-E expression in lesions would recruit and aid in NK cell-mediated lysis of the tumor cells, synergistically adding to cytolytic activity against early HPV-positive lesions. It would be appealing to investigate a correlation between repression of HLA-E in early HPV-infected tissue and increased risk of developing cervical cancer (or HNSCC). We would suspect that lesions with lower HLA-E expression are more likely to progress to higher-grade lesions. If true, assessment of HLA-E expression or CGI methylation status could be used as an attractive biomarker for prognosis.

Finally, while CXCL14 promoter methylation was shown to accumulate over large expanses of time, methylation at the HLA-E CGI required relatively short-term expression of

HPV E7. This suggests that perhaps repression of HLA-E expression (or other genes

112 identified in the arrays) is paramount in order for HPV to evade early immune detection, and that repression of CXCL14 through promoter methylation is a trait gained later in evolutionary time. Disparities in methylation efficiency at different promoters may be due to differences in binding affinity between E7 and transcription factors or complexes necessary for methylation at specific genomic loci (hypothesized in Figure 4-4). Follow-up experiments may then aim to define the exact mechanisms of E7-directed promoter methylation, which may shed light on temporal discrepancies in the accumulation of methylation at different genomic elements.

Additional Future Directions

Repression of CXCL14 may Induce Tolerance of HPV-Infection

CXCL14 downregulation by HPV may induce tolerance of host immune responses caused by chronic inflammation. As discussed, chronic inflammation is a known risk factor for cancer development. CIN3 HPV-positive cervical lesions have been shown to exhibit moderate but sustained inflammation (209), suggesting a potential mechanism of cancer progression and immune tolerance (210). Our and other studies have shown that chemokine expression is increased in HPV-positive lesions; however an immune response is not launched to clear the infection.

An additional potential mechanism of immune evasion through tolerance can be mediated through myeloid-derived suppressor cells (MDSC), which secrete immune suppressive signaling molecules such as NOS, TGF-β, and IL-10 (435). In addition to differential abundances of NK and CD8+ T cells in the TDLN near tumors with or without

Cxcl14 expression, we observed differences in abundances of myeloid-derived suppressor- like cells (Gr1high, CD11b+) in Rag1-/- mice (Figure 5-4). While the overall abundance of macrophages in the spleens of mice re-expressing Cxcl14 was increased, MDSC-like cells in the spleens and tumors of mice re-expressing Cxcl14 was significantly reduced. MDSC represent a combination of myeloid cells that develop from exposure to chemokines in the

113

Figure 5-4. Murine Cxcl14 Expression Decreases Myeloid-Derived Suppressor-Like Cells (MDSC-Like) in Rag1-/- Tumors and Spleens. MOE/E6E7 cell clones expressing Cxcl14 (8 and 16) and one vector containing MOE/E6E7 cell clone were injected into the rear right flank of Rag1-/- mice (n = 4, each group). Regional tumor (A) and spleen (B & C) were harvested at 23 days post injection, homogenized, and analyzed by flow cytometry. Relative abundance of MDSC-like cells (A & B) was determined using anti-MHCII, anti-Gr1, and anti-CD11b antibodies, while macrophage abundance (C) was determined using anti- MCHII and anti-F4/80 antibodies. Data collected in collaboration with Joe Westrich, University of Colorado.

114

TME (436, 437). Signaling from MDSC can reduce L-selectin expression on T cells, disallowing infiltration into the TME, and recruit Treg cells for additional immune suppression

(435). These results imply the possibility that Cxcl14 expression in HPV-positive MOE cells reduces MDSC, thereby allowing for NK and CD8+ T cell antitumor activity.

One major caveat to these preliminary results is that they were performed only in

Rag1-/- deficient mice—follow-up studies would therefore need to confirm the same observation in wild type mice. Future studies would aim to isolate the described MDSC-like cells by FACS sorting, and assess their ability to repress NK and T cell activity in vitro.

Additional in vivo studies may use anti-Gr1 antibodies to deplete MDSC in mice with repressed Cxcl14 expression. MDSC are known to repress T cell proliferation and an influx of MDSC is frequently observed in tumors and TDLN (438). Given the abundance of MDSC- like cells seen here as a function of Cxcl14 expression, it is possible that HPV regulates yet another facet of the host immune response through chemokine dysregulation and should be further explored.

What is the nature of interactions between HPV E7 and DNMT1?

A previous study has shown that recombinant E7 can bind and activate purified

DNMT1 in vitro (127). We therefore hypothesize that DNMT1 transactivation by HPV E7 and/or a direct interaction between DNMT1 and E7 triggers CXCL14 promoter hypermethylation, leading to CXCL14 downregulation (among many other genes, as observed in Chapter IV). This interaction would therefore act as an intermediary between

HPV E7 and viral persistence, immune tolerance, and cancer progression. We can further utilize our methylation reporter system (Figure 4-5) to assess the ability of HPV E7 to promote hypermethylation at the CXCL14 promoter by transfection into NIKS cell lines +/-

HPV genome or E7. The sufficiency of E7 to methylate the CXCL14 promoter can be determined by MSP. The necessity of DNMT1 for E7-mediated methylation of the CXCL14

115 promoter can also be determined using shRNA or CRISPR to knock-down or knock-out

DNMT1 expression.

We can investigate the mechanism of E7-induced CXCL14 promoter methylation by utilizing a series of E7 functional mutants (obtained from Karl Munger, Tufts University) and the promoter methylation reporter system. As the DNMT1 promoter contains an E2F- responsive element (439) and DNMT1 expression is upregulated by HPV E7, DNMT1 activation may occur via E2F-mediated transactivation by E7. To test this hypothesis, we will use HPV16 E7 mutants that cannot bind to and inactivate pRb (440, 441). If activation of

DNMT1 is simply due to overexpression, none of the E7 mutants will increase DNMT1 expression or CXCL14 promoter methylation, while wild type HPV16 E7 does.

DNMT1 enzymatic activity is more likely, however, to be regulated posttranslationally. A previous study has shown that HPV16 E7 binds to DNMT1 protein and activates methyltransferase activity mediated by the CR3 zinc-finger domain of E7 (127).

Thus, HPV16 E7 would induce CXCL14 promoter hypermethylation by enhancing DNMT1 methyltransferase activity via a direct interaction. We hypothesized that the E7-DNMT1 complex can localize to certain TF or complexes in order to direct methylation (Figure 4-4).

To test this hypothesis we can determine the methylation status of the CXCL14 or HLA-E promoter region in NIKS cells using E7 functional mutants which are defective in the CR3 protein binding domain (442–445). ChIP experiments can be utilized to pull-down E7-

DNMT1 complexes and determine if enrichment indeed occurs near certain transcription factor binding sites, as hypothesized.

Elucidating the mechanism of E7-dependent CXCL14 promoter hypermethylation may lead to identifying therapeutic targets for HPV-associated cancers and would therefore be a candidate for future studies. Identifying, for instance, an E7 domain that is necessary to bind a TF complex for directed methylation could provide a potential druggable target. On

116 the other hand, if DNMT1 activation is due to an E7-induced overabundance, DNMT inhibitors, or novel E7 inhibitors may be explored.

Concluding Remarks

HPV is a prevalent global pathogen that remains a significant cause of morbidity and mortality, especially in the developing world. The HPV oncoproteins perform myriad functions to aid the virus in its lifecycle, some of which may impart devastating consequences to the host. The viral oncoprotein E7 is a key player which can repress a host immune response to viral infection, leading to persistence and disease progression (Figure

5-5). This study aimed to further elucidate the mechanism by which E7 evades a host immune response, evasion of which may lead to cancer progression. In Chapter III, we discuss the effect of viral infection on chemokines expression. E7 expression is both necessary and sufficient to direct hypermethylation at the CXCL14 promoter, leading to repression of CXCL14 expression. Repression of CXCL14 expression in HPV-positive tumors dysregulates CD4+ T, CD8+ T, NK and MDSC-like cell localization which has devastating results for the host in our in vivo mouse model (Figure 5-5). CXCL14 is an important homeostatic tumor suppressing chemokine and repression of which facilitates cell migration and immune cell regulation, leading to viral persistence, immune tolerance, and cancer progression. This study highlights the importance of CXCL14 regulation in HPV- positive cancer development, and provides for new avenues of therapeutic discovery.

Chapter IV discusses the extent of E7-dysregulation of the host methylome, and its effect on gene expression. Here, we show that E7 expression directs specific changes to host gene expression through multiple mechanisms targeting specific pathways, including immune cell regulation, cell cycle regulation, and DNA damage responses. We show that methylation at a distal HLA-E CGI represses HLA-E expression, which is known to regulate

NK and CD8+ T cell activity, downregulation of which may therefore play an important role in viral persistence and disease progression (Figure 5-5). Repression of HLA-E through

117

Figure 5-5 Model of HPV E7 Immune Suppression through Epigenetic Manipulation of the Host. Beginning top right, HPV expresses E7 upon viral infection. E7 interacts with DNMT1, directing specific changes to the host methylome, leading to repression of CXCL14 and HLA-E. CXCL14 repression (potentially in wound healing) increases cell motility, potentially downregulating cell-cell junctions and increasing HPV infectivity. CXCL14 expression inhibits MDSC accumulation, and recruits NK and CD8+ T surveillance for homeostatic tissue maintenance. HPV-repression of CXCL14 results in tumor growth through dysregulation of immune surveillance. HPV-repression of HLA-E may have significance consequences by dysregulating NK and CD8+ T cell activity near infected keratinocytes, leading to maintenance of productive infection. Solid black lines indicate known direct interactions of consequences. Black dashed lines indicate known direct or indirect consequences. Dashed red lines indicate potential interactions.

118

promoter hypermethylation is therefore likely an additional mechanism employed by HPV to evade immune detection in order to persist.

Elucidating the mechanisms of viral evasion of the host immune response through

CXCL14 and HLA-E downregulation is an important step to fully understand how HPV persistence leads to cancer progression. Revealing how E7 specifically directs hypermethylation in the host genome should be further explored as a potential therapeutic target. Additionally, detection of HPV-directed hypermethylation in the host genome through

HNSCC patient saliva samples is an appealing and inventive non-invasive approach to therapeutics and should be perused. Overall, the viral oncoprotein E7 is a master manipulator of the host immune response and further studies into the mechanisms of immune dysregulation by E7 will be key in order to combat HPV-associated disease.

119

REFERENCES

1. Bernard H-U, Burk RD, Chen Z, van Doorslaer K, zur Hausen H, de Villiers E-M. 2010. Classification of papillomaviruses (PVs) based on 189 PV types and proposal of taxonomic amendments. Virology 401:70–9.

2. Bravo IG, Félez-Sánchez M. 2015. Papillomaviruses: Viral evolution, cancer and evolutionary medicine. Evol Med public Heal 2015:32–51.

3. de Villiers E-M, Fauquet C, Broker TR, Bernard H-U, zur Hausen H. 2004. Classification of papillomaviruses. Virology 324:17–27.

4. zur Hausen H. 2002. Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2:342–50.

5. Mistry N, Wibom C, Evander M. 2008. Cutaneous and mucosal human papillomaviruses differ in net surface charge, potential impact on tropism. Virol J 5:118.

6. Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, Benbrahim- Tallaa L, Guha N, Freeman C, Galichet L, Cogliano V, WHO International Agency for Research on Cancer Monograph Working Group. 2009. A review of human carcinogens--Part B: biological agents. Lancet Oncol 10:321–2.

7. Muñoz N, Bosch FX, de Sanjosé S, Herrero R, Castellsagué X, Shah K V, Snijders PJF, Meijer CJLM, International Agency for Research on Cancer Multicenter Cervical Cancer Study Group. 2003. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 348:518– 27.

8. Ault KA. 2006. Epidemiology and Natural History of Human Papillomavirus Infections in the Female Genital Tract. Infect Dis Obstet Gynecol 2006:1–5.

9. Tommasino M. 2014. The human papillomavirus family and its role in carcinogenesis. Semin Cancer Biol 26:13–21.

10. de Sanjose S, Quint WG, Alemany L, Geraets DT, Klaustermeier JE, Lloveras B, Tous S, Felix A, Bravo LE, Shin H-R, Vallejos CS, de Ruiz PA, Lima MA, Guimera N, Clavero O, Alejo M, Llombart-Bosch A, Cheng-Yang C, Tatti SA, Kasamatsu E, Iljazovic E, Odida M, Prado R, Seoud M, Grce M, Usubutun A, Jain A, Suarez GAH, Lombardi LE, Banjo A, Menéndez C, Domingo EJ, Velasco J, Nessa A, Chichareon SCB, Qiao YL, Lerma E, Garland SM, Sasagawa T, Ferrera A, Hammouda D, Mariani L, Pelayo A, Steiner I, Oliva E, Meijer CJ, Al- Jassar WF, Cruz E, Wright TC, Puras A, Llave CL, Tzardi M, Agorastos T, Garcia-Barriola V, Clavel C, Ordi J, Andújar M, Castellsagué X, Sánchez GI, Nowakowski AM, Bornstein J, Muñoz N, Bosch FX, Retrospective International Survey and HPV Time Trends Study Group. 2010. Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study. Lancet Oncol 11:1048–56.

120

11. Jablonska S, Biczysko W, Jakubowicz K, Dabrowski J. 1968. On the viral etiology of epidermodysplasia verruciformis Lewandowsky-Lutz. Electron microscope studies. Dermatologica 137:113–25.

12. Pass F, Reissig M, Shah K V, Eisinger M, Orth G. 1977. Identification of an immunologically distinct papillomavirus from lesions of epidermodysplasia verruciformis. J Natl Cancer Inst 59:1107–12.

13. Younai FS. 2016. Current Trends in the Incidence and Presentation of Oropharyngeal Cancer. J Calif Dent Assoc 44:93–100.

14. Marur S, Forastiere AA. 2016. Head and Neck Squamous Cell Carcinoma: Update on Epidemiology, Diagnosis, and Treatment. Mayo Clin Proc 91:386–96.

15. de Martel C, Ferlay J, Franceschi S, Vignat J, Bray F, Forman D, Plummer M. 2012. Global burden of cancers attributable to infections in 2008: a review and synthetic analysis. Lancet Oncol 13:607–15.

16. International Agency for Research on Cancer. 2007. The World Health Organization IARC Monographs on the Evaluation of Carcinogenic Risks to Humans Volume 90.

17. Hariri S, Unger ER, Sternberg M, Dunne EF, Swan D, Patel S, Markowitz LE. 2011. Prevalence of genital human papillomavirus among females in the United States, the National Health And Nutrition Examination Survey, 2003-2006. J Infect Dis 204:566–73.

18. Forman D, de Martel C, Lacey CJ, Soerjomataram I, Lortet-Tieulent J, Bruni L, Vignat J, Ferlay J, Bray F, Plummer M, Franceschi S. 2012. Global burden of human papillomavirus and related diseases. Vaccine 30 Suppl 5:F12–23.

19. Schiffman M, Castle PE. 2005. The promise of global cervical-cancer prevention. N Engl J Med 353:2101–4.

20. Crow JM. 2012. HPV: The global burden. Nature 488:S2–3.

21. Bruni L, Diaz M, Castellsagué X, Ferrer E, Bosch FX, de Sanjosé S. 2010. Cervical human papillomavirus prevalence in 5 continents: meta-analysis of 1 million women with normal cytological findings. J Infect Dis 202:1789–99.

22. LIVESTRONG, American Cancer Society. 2010. The Global Economic Cost of Cancer. Am Cancer Soc Tech Rep No. 00544.

23. Kulasingam SL, Hughes JP, Kiviat NB, Mao C, Weiss NS, Kuypers JM, Koutsky LA. 2002. Evaluation of human papillomavirus testing in primary screening for cervical abnormalities: comparison of sensitivity, specificity, and frequency of referral. JAMA 288:1749–57.

24. Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. 1998. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 338:423–8.

121

25. Richardson H, Kelsall G, Tellier P, Voyer H, Abrahamowicz M, Ferenczy A, Coutlée F, Franco EL. 2003. The natural history of type-specific human papillomavirus infections in female university students. Cancer Epidemiol Biomarkers Prev 12:485–90.

26. Dunne EF, Unger ER, Sternberg M, McQuillan G, Swan DC, Patel SS, Markowitz LE. 2007. Prevalence of HPV infection among females in the United States. JAMA 297:813–9.

27. Harper DM, Vierthaler SL, Santee JA. 2010. Review of Gardasil. J Vaccines Vaccin 01.

28. Baseman JG, Koutsky LA. 2005. The epidemiology of human papillomavirus infections. J Clin Virol 32 Suppl 1:S16–24.

29. National Institutes of Health. 1996. Cervica Cancer. NIH Consens Statement 14:1– 38.

30. Lees BF, Erickson BK, Huh WK. 2016. Cervical cancer screening: evidence behind the guidelines. Am J Obstet Gynecol 214:438–43.

31. US Cancer Statistics Working Group. 2015. United States Cancer Statistics: 1999– 2012 Incidence and Mortality Web-based Report. Dep Heal Hum Serv Centers Dis Control Prev Natl Cancer Inst.

32. Chaturvedi AK, Engels EA, Pfeiffer RM, Hernandez BY, Xiao W, Kim E, Jiang B, Goodman MT, Sibug-Saber M, Cozen W, Liu L, Lynch CF, Wentzensen N, Jordan RC, Altekruse S, Anderson WF, Rosenberg PS, Gillison ML. 2011. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol 29:4294–301.

33. Auluck A, Hislop G, Bajdik C, Poh C, Zhang L, Rosin M. 2010. Trends in oropharyngeal and oral cavity cancer incidence of human papillomavirus (HPV)- related and HPV-unrelated sites in a multicultural population: the British Columbia experience. Cancer 116:2635–44.

34. Monie A, Hung C-F, Roden R, Wu T-C. 2008. Cervarix: a vaccine for the prevention of HPV 16, 18-associated cervical cancer. Biologics 2:97–105.

35. Gee J, Weinbaum C, Sukumaran L, Markowitz LE. 2016. Quadrivalent HPV vaccine safety review and US safety monitoring plans for nine-valent HPV vaccine. Hum Vaccin Immunother 1–12.

36. Macartney KK, Chiu C, Georgousakis M, Brotherton JML. 2013. Safety of human papillomavirus vaccines: a review. Drug Saf 36:393–412.

37. De Vincenzo R, Conte C, Ricci C, Scambia G, Capelli G. 2014. Long-term efficacy and safety of human papillomavirus vaccination. Int J Womens Health 6:999–1010.

122

38. Muñoz N, Kjaer SK, Sigurdsson K, Iversen O-E, Hernandez-Avila M, Wheeler CM, Perez G, Brown DR, Koutsky LA, Tay EH, Garcia PJ, Ault KA, Garland SM, Leodolter S, Olsson S-E, Tang GWK, Ferris DG, Paavonen J, Steben M, Bosch FX, Dillner J, Huh WK, Joura EA, Kurman RJ, Majewski S, Myers ER, Villa LL, Taddeo FJ, Roberts C, Tadesse A, Bryan JT, Lupinacci LC, Giacoletti KED, Sings HL, James MK, Hesley TM, Barr E, Haupt RM. 2010. Impact of human papillomavirus (HPV)-6/11/16/18 vaccine on all HPV-associated genital diseases in young women. J Natl Cancer Inst 102:325–39.

39. Beltran R, Simms T, Lee HY, Kwon M. 2016. HPV Literacy and Associated Factors Among Hmong American Immigrants: Implications for Reducing Cervical Cancer Disparity. J Community Health 41:603–11.

40. Yi JK, Anderson KO, Le Y-C, Escobar-Chaves SL, Reyes-Gibby CC. 2013. English proficiency, knowledge, and receipt of HPV vaccine in Vietnamese-American women. J Community Health 38:805–11.

41. Zimet GD, Rosberger Z, Fisher WA, Perez S, Stupiansky NW. 2013. Beliefs, behaviors and HPV vaccine: correcting the myths and the misinformation. Prev Med (Baltim) 57:414–8.

42. Agosti JM, Goldie SJ. 2007. Introducing HPV vaccine in developing countries--key challenges and issues. N Engl J Med 356:1908–10.

43. Tomljenovic L, Spinosa JP, Shaw CA. 2013. Human papillomavirus (HPV) vaccines as an option for preventing cervical malignancies: (how) effective and safe? Curr Pharm Des 19:1466–87.

44. Pytynia KB, Dahlstrom KR, Sturgis EM. 2014. Epidemiology of HPV-associated oropharyngeal cancer. Oral Oncol 50:380–6.

45. Ward G, Mehta V, Moore M. 2015. Morbidity, mortality and cost from HPV-related oropharyngeal cancer: Impact of 2-, 4- and 9-valent vaccines. Hum Vaccin Immunother 1–5.

46. Van Doorslaer K. 2013. Evolution of the papillomaviridae. Virology 445:11–20.

47. Chen XS, Garcea RL, Goldberg I, Casini G, Harrison SC. 2000. Structure of small virus-like particles assembled from the L1 protein of human papillomavirus 16. Mol Cell 5:557–67.

48. Wang JW, Roden RBS. 2013. L2, the minor capsid protein of papillomavirus. Virology 445:175–86.

49. Egawa K. 2003. Do human papillomaviruses target epidermal stem cells? Dermatology 207:251–4.

50. Doorbar J. 2006. Molecular biology of human papillomavirus infection and cervical cancer. Clin Sci (Lond) 110:525–41.

123

51. Letian T, Tianyu Z. 2010. Cellular receptor binding and entry of human papillomavirus. Virol J 7:2.

52. Joyce JG, Tung JS, Przysiecki CT, Cook JC, Lehman ED, Sands JA, Jansen KU, Keller PM. 1999. The L1 major capsid protein of human papillomavirus type 11 recombinant virus-like particles interacts with heparin and cell-surface glycosaminoglycans on human keratinocytes. J Biol Chem 274:5810–22.

53. Day PM, Thompson CD, Schowalter RM, Lowy DR, Schiller JT. 2013. Identification of a role for the trans-Golgi network in human papillomavirus 16 pseudovirus infection. J Virol 87:3862–70.

54. Zhang W, Kazakov T, Popa A, DiMaio D. 2014. Vesicular trafficking of incoming human papillomavirus 16 to the Golgi apparatus and endoplasmic reticulum requires -secretase activity. MBio 5:e01777–14.

55. Chen J, Xue Y, Poidinger M, Lim T, Chew SH, Pang CL, Abastado J-P, Thierry F. 2014. Mapping of HPV transcripts in four human cervical lesions using RNAseq suggests quantitative rearrangements during carcinogenic progression. Virology 462- 463:14–24.

56. Kim K, Lambert PF. 2002. E1 protein of bovine papillomavirus 1 is not required for the maintenance of viral plasmid DNA replication. Virology 293:10–4.

57. Van Tine BA, Dao LD, Wu S-Y, Sonbuchner TM, Lin BY, Zou N, Chiang C-M, Broker TR, Chow LT. 2004. Human papillomavirus (HPV) origin-binding protein associates with mitotic spindles to enable viral DNA partitioning. Proc Natl Acad Sci U S A 101:4030–5.

58. Hughes FJ, Romanos MA. 1993. E1 protein of human papillomavirus is a DNA helicase/ATPase. Nucleic Acids Res 21:5817–23.

59. McBride AA. 2013. The papillomavirus E2 proteins. Virology 445:57–79.

60. Eckhart L, Lippens S, Tschachler E, Declercq W. 2013. Cell death by cornification. Biochim Biophys Acta 1833:3471–80.

61. Pyeon D, Pearce SM, Lank SM, Ahlquist P, Lambert PF. 2009. Establishment of human papillomavirus infection requires cell cycle progression. PLoS Pathog 5:e1000318.

62. DiMaio D, Mattoon D. 2001. Mechanisms of cell transformation by papillomavirus E5 proteins. Oncogene 20:7866–73.

63. Mantovani F, Banks L. 2001. The human papillomavirus E6 protein and its contribution to malignant progression. Oncogene 20:7874–87.

64. Yim E-K, Park J-S. 2005. The role of HPV E6 and E7 oncoproteins in HPV- associated cervical carcinogenesis. Cancer Res Treat 37:319–24.

124

65. Doorbar J, Egawa N, Griffin H, Kranjec C, Murakami I. 2015. Human papillomavirus molecular biology and disease association. Rev Med Virol 25 Suppl 1:2–23.

66. Maglennon GA, McIntosh P, Doorbar J. 2011. Persistence of viral DNA in the epithelial basal layer suggests a model for papillomavirus latency following immune regression. Virology 414:153–63.

67. Holmgren SC, Patterson NA, Ozbun MA, Lambert PF. 2005. The minor capsid protein L2 contributes to two steps in the human papillomavirus type 31 life cycle. J Virol 79:3938–48.

68. Doorbar J, Ely S, Sterling J, McLean C, Crawford L. 1991. Specific interaction between HPV-16 E1-E4 and cytokeratins results in collapse of the epithelial cell intermediate filament network. Nature 352:824–7.

69. Wang Q, Griffin H, Southern S, Jackson D, Martin A, McIntosh P, Davy C, Masterson PJ, Walker PA, Laskey P, Omary MB, Doorbar J. 2004. Functional analysis of the human papillomavirus type 16 E1=E4 protein provides a mechanism for in vivo and in vitro keratin filament reorganization. J Virol 78:821–33.

70. Buck CB, Thompson CD, Pang Y-YS, Lowy DR, Schiller JT. 2005. Maturation of papillomavirus capsids. J Virol 79:2839–46.

71. Jablonska S, Fabjanska L, Formas I. 1966. On the viral etiology of epidermodysplasia verruciformis. Dermatologica 132:369–85.

72. Gissmann L, zur Hausen H. 1980. Partial characterization of viral DNA from human genital warts (Condylomata acuminata). Int J cancer 25:605–9.

73. de Villiers EM, Gissmann L, zur Hausen H. 1981. Molecular cloning of viral DNA from human genital warts. J Virol 40:932–5.

74. Dürst M, Gissmann L, Ikenberg H, zur Hausen H. 1983. A papillomavirus DNA from a cervical carcinoma and its prevalence in cancer biopsy samples from different geographic regions. Proc Natl Acad Sci U S A 80:3812–5.

75. Agrawal N, Frederick MJ, Pickering CR, Bettegowda C, Chang K, Li RJ, Fakhry C, Xie T-X, Zhang J, Wang J, Zhang N, El-Naggar AK, Jasser SA, Weinstein JN, Treviño L, Drummond JA, Muzny DM, Wu Y, Wood LD, Hruban RH, Westra WH, Koch WM, Califano JA, Gibbs RA, Sidransky D, Vogelstein B, Velculescu VE, Papadopoulos N, Wheeler DA, Kinzler KW, Myers JN. 2011. Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1. Science 333:1154–7.

125

76. Stransky N, Egloff AM, Tward AD, Kostic AD, Cibulskis K, Sivachenko A, Kryukov G V, Lawrence MS, Sougnez C, McKenna A, Shefler E, Ramos AH, Stojanov P, Carter SL, Voet D, Cortés ML, Auclair D, Berger MF, Saksena G, Guiducci C, Onofrio RC, Parkin M, Romkes M, Weissfeld JL, Seethala RR, Wang L, Rangel-Escareño C, Fernandez-Lopez JC, Hidalgo-Miranda A, Melendez- Zajgla J, Winckler W, Ardlie K, Gabriel SB, Meyerson M, Lander ES, Getz G, Golub TR, Garraway LA, Grandis JR. 2011. The mutational landscape of head and neck squamous cell carcinoma. Science 333:1157–60.

77. Hall AHS, Alexander KA. 2003. RNA interference of human papillomavirus type 18 E6 and E7 induces senescence in HeLa cells. J Virol 77:6066–9.

78. Gu W, Putral L, Hengst K, Minto K, Saunders NA, Leggatt G, McMillan NAJ. 2006. Inhibition of cervical cancer cell growth in vitro and in vivo with lentiviral-vector delivered short hairpin RNA targeting human papillomavirus E6 and E7 oncogenes. Cancer Gene Ther 13:1023–32.

79. Magaldi TG, Almstead LL, Bellone S, Prevatt EG, Santin AD, DiMaio D. 2012. Primary human cervical carcinoma cells require human papillomavirus E6 and E7 expression for ongoing proliferation. Virology 422:114–24.

80. Jabbar SF, Abrams L, Glick A, Lambert PF. 2009. Persistence of high-grade cervical dysplasia and cervical cancer requires the continuous expression of the human papillomavirus type 16 E7 oncogene. Cancer Res 69:4407–14.

81. Jabbar SF, Park S, Schweizer J, Berard-Bergery M, Pitot HC, Lee D, Lambert PF. 2012. Cervical cancers require the continuous expression of the human papillomavirus type 16 E7 oncoprotein even in the presence of the viral E6 oncoprotein. Cancer Res 72:4008–16.

82. Oetke C, Auvinen E, Pawlita M, Alonso A. 2000. Human papillomavirus type 16 E5 protein localizes to the Golgi apparatus but does not grossly affect cellular glycosylation. Arch Virol 145:2183–91.

83. Ashrafi GH, Haghshenas MR, Marchetti B, O’Brien PM, Campo MS. 2005. E5 protein of human papillomavirus type 16 selectively downregulates surface HLA class I. Int J cancer 113:276–83.

84. Zhang B, Li P, Wang E, Brahmi Z, Dunn KW, Blum JS, Roman A. 2003. The E5 protein of human papillomavirus type 16 perturbs MHC class II antigen maturation in human foreskin keratinocytes treated with interferon-gamma. Virology 310:100–8.

85. Schapiro F, Sparkowski J, Adduci A, Suprynowicz F, Schlegel R, Grinstein S. 2000. Golgi alkalinization by the papillomavirus E5 oncoprotein. J Cell Biol 148:305– 15.

86. Gruener M, Bravo IG, Momburg F, Alonso A, Tomakidi P. 2007. The E5 protein of the human papillomavirus type 16 down-regulates HLA-I surface expression in calnexin-expressing but not in calnexin-deficient cells. Virol J 4:116.

126

87. Campo MS, Graham S V, Cortese MS, Ashrafi GH, Araibi EH, Dornan ES, Miners K, Nunes C, Man S. 2010. HPV-16 E5 down-regulates expression of surface HLA class I and reduces recognition by CD8 T cells. Virology 407:137–42.

88. Wazer DE, Liu XL, Chu Q, Gao Q, Band V. 1995. Immortalization of distinct human mammary epithelial cell types by human papilloma virus 16 E6 or E7. Proc Natl Acad Sci U S A 92:3687–91.

89. Band V, De Caprio JA, Delmolino L, Kulesa V, Sager R. 1991. Loss of p53 protein in human papillomavirus type 16 E6-immortalized human mammary epithelial cells. J Virol 65:6671–6.

90. Hawley-Nelson P, Vousden KH, Hubbert NL, Lowy DR, Schiller JT. 1989. HPV16 E6 and E7 proteins cooperate to immortalize human foreskin keratinocytes. EMBO J 8:3905–10.

91. Münger K, Phelps WC, Bubb V, Howley PM, Schlegel R. 1989. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. J Virol 63:4417–21.

92. Arbeit JM, Münger K, Howley PM, Hanahan D. 1994. Progressive squamous epithelial neoplasia in K14-human papillomavirus type 16 transgenic mice. J Virol 68:4358–68.

93. Song S, Liem A, Miller JA, Lambert PF. 2000. Human papillomavirus types 16 E6 and E7 contribute differently to carcinogenesis. Virology 267:141–50.

94. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. 1990. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell 63:1129–36.

95. Huibregtse JM, Scheffner M, Howley PM. 1993. Localization of the E6-AP regions that direct human papillomavirus E6 binding, association with p53, and ubiquitination of associated proteins. Mol Cell Biol 13:4918–27.

96. Kessis TD, Slebos RJ, Nelson WG, Kastan MB, Plunkett BS, Han SM, Lorincz AT, Hedrick L, Cho KR. 1993. Human papillomavirus 16 E6 expression disrupts the p53-mediated cellular response to DNA damage. Proc Natl Acad Sci U S A 90:3988– 92.

97. Foster SA, Demers GW, Etscheid BG, Galloway DA. 1994. The ability of human papillomavirus E6 proteins to target p53 for degradation in vivo correlates with their ability to abrogate actinomycin D-induced growth arrest. J Virol 68:5698–705.

98. Liu HC, Chen GG, Vlantis AC, Tse GM, Chan ATC, van Hasselt CA. 2008. Inhibition of apoptosis in human laryngeal cancer cells by E6 and E7 oncoproteins of human papillomavirus 16. J Cell Biochem 103:1125–43.

99. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA, Klingelhutz AJ. 1998. Both Rb/p16INK4a inactivation and telomerase activity are required to immortalize human epithelial cells. Nature 396:84–8.

127

100. Patel D, Huang SM, Baglia LA, McCance DJ. 1999. The E6 protein of human papillomavirus type 16 binds to and inhibits co-activation by CBP and p300. EMBO J 18:5061–72.

101. Gao Q, Singh L, Kumar A, Srinivasan S, Wazer DE, Band V. 2001. Human papillomavirus type 16 E6-induced degradation of E6TP1 correlates with its ability to immortalize human mammary epithelial cells. J Virol 75:4459–66.

102. Giampieri S, García-Escudero R, Green J, Storey A. 2004. Human papillomavirus type 77 E6 protein selectively inhibits p53-dependent transcription of proapoptotic genes following UV-B irradiation. Oncogene 23:5864–70.

103. Tan MJA, White EA, Sowa ME, Harper JW, Aster JC, Howley PM. 2012. Cutaneous β-human papillomavirus E6 proteins bind Mastermind-like coactivators and repress Notch signaling. Proc Natl Acad Sci U S A 109:E1473–80.

104. Tanaka A, Noda T, Yajima H, Hatanaka M, Ito Y. 1989. Identification of a transforming gene of human papillomavirus type 16. J Virol 63:1465–9.

105. Herber R, Liem A, Pitot H, Lambert PF. 1996. Squamous epithelial hyperplasia and carcinoma in mice transgenic for the human papillomavirus type 16 E7 oncogene. J Virol 70:1873–81.

106. Frolov M V, Dyson NJ. 2004. Molecular mechanisms of E2F-dependent activation and pRB-mediated repression. J Cell Sci 117:2173–81.

107. Huh K, Zhou X, Hayakawa H, Cho J-Y, Libermann TA, Jin J, Harper JW, Munger K. 2007. Human papillomavirus type 16 E7 oncoprotein associates with the cullin 2 ubiquitin ligase complex, which contributes to degradation of the retinoblastoma tumor suppressor. J Virol 81:9737–47.

108. Boyer SN, Wazer DE, Band V. 1996. E7 protein of human papilloma virus-16 induces degradation of retinoblastoma protein through the ubiquitin-proteasome pathway. Cancer Res 56:4620–4.

109. Hwang SG, Lee D, Kim J, Seo T, Choe J. 2002. Human papillomavirus type 16 E7 binds to E2F1 and activates E2F1-driven transcription in a retinoblastoma protein- independent manner. J Biol Chem 277:2923–30.

110. Pyeon D, Newton MA, Lambert PF, den Boon JA, Sengupta S, Marsit CJ, Woodworth CD, Connor JP, Haugen TH, Smith EM, Kelsey KT, Turek LP, Ahlquist P. 2007. Fundamental differences in cell cycle deregulation in human papillomavirus-positive and human papillomavirus-negative head/neck and cervical cancers. Cancer Res 67:4605–19.

111. den Boon JA, Pyeon D, Wang SS, Horswill M, Schiffman M, Sherman M, Zuna RE, Wang Z, Hewitt SM, Pearson R, Schott M, Chung L, He Q, Lambert P, Walker J, Newton MA, Wentzensen N, Ahlquist P. 2015. Molecular transitions from papillomavirus infection to cervical precancer and cancer: Role of stromal estrogen receptor signaling. Proc Natl Acad Sci U S A 112:E3255–64.

128

112. Barbosa MS, Vass WC, Lowy DR, Schiller JT. 1991. In vitro biological activities of the E6 and E7 genes vary among human papillomaviruses of different oncogenic potential. J Virol 65:292–8.

113. Halbert CL, Demers GW, Galloway DA. 1992. The E6 and E7 genes of human papillomavirus type 6 have weak immortalizing activity in human epithelial cells. J Virol 66:2125–34.

114. Münger K, Werness BA, Dyson N, Phelps WC, Harlow E, Howley PM. 1989. Complex formation of human papillomavirus E7 proteins with the retinoblastoma tumor suppressor gene product. EMBO J 8:4099–105.

115. Ciccolini F, Di Pasquale G, Carlotti F, Crawford L, Tommasino M. 1994. Functional studies of E7 proteins from different HPV types. Oncogene 9:2633–8.

116. Gage JR, Meyers C, Wettstein FO. 1990. The E7 proteins of the nononcogenic human papillomavirus type 6b (HPV-6b) and of the oncogenic HPV-16 differ in retinoblastoma protein binding and other properties. J Virol 64:723–30.

117. Zhang B, Chen W, Roman A. 2006. The E7 proteins of low- and high-risk human papillomaviruses share the ability to target the pRB family member p130 for degradation. Proc Natl Acad Sci U S A 103:437–42.

118. Roman A, Munger K. 2013. The papillomavirus E7 proteins. Virology 445:138–168.

119. Georgopoulos NT, Proffitt JL, Blair GE. 2000. Transcriptional regulation of the major histocompatibility complex (MHC) class I heavy chain, TAP1 and LMP2 genes by the human papillomavirus (HPV) type 6b, 16 and 18 E7 oncoproteins. Oncogene 19:4930–5.

120. Sharma S, Kelly TK, Jones PA. 2010. Epigenetics in cancer. Carcinogenesis 31:27– 36.

121. McLaughlin-Drubin ME, Huh K-W, Munger K. 2008. Human Papillomavirus Type 16 E7 Oncoprotein Associates with E2F6. J Virol 82:8695–8705.

122. Trimarchi JM, Fairchild B, Wen J, Lees JA. 2001. The E2F6 transcription factor is a component of the mammalian Bmi1-containing polycomb complex. Proc Natl Acad Sci U S A 98:1519–24.

123. Attwooll C, Oddi S, Cartwright P, Prosperini E, Agger K, Steensgaard P, Wagener C, Sardet C, Moroni MC, Helin K. 2005. A novel repressive E2F6 complex containing the polycomb group protein, EPC1, that interacts with EZH2 in a proliferation-specific manner. J Biol Chem 280:1199–208.

124. Baldwin A, Huh K-W, Münger K. 2006. Human papillomavirus E7 oncoprotein dysregulates steroid receptor coactivator 1 localization and function. J Virol 80:6669– 77.

129

125. Brehm A, Nielsen SJ, Miska EA, McCance DJ, Reid JL, Bannister AJ, Kouzarides T. 1999. The E7 oncoprotein associates with Mi2 and histone deacetylase activity to promote cell growth. EMBO J 18:2449–58.

126. Finzer P, Ventz R, Kuntzen C, Seibert N, Soto U, Rösl F. 2002. Growth arrest of HPV-positive cells after histone deacetylase inhibition is independent of E6/E7 oncogene expression. Virology 304:265–73.

127. Burgers WA, Blanchon L, Pradhan S, de Launoit Y, Kouzarides T, Fuks F. 2007. Viral oncoproteins target the DNA methyltransferases. Oncogene 26:1650–5.

128. Anayannis NVJ, Schlecht NF, Belbin TJ. 2015. Epigenetic Mechanisms of Human Papillomavirus-Associated Head and Neck Cancer. Arch Pathol Lab Med 139:1373– 8.

129. Cancer Genome Atlas Network. 2015. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature 517:576–82.

130. Chalertpet K, Pakdeechaidan W, Patel V, Mutirangura A, Yanatatsaneejit P. 2015. Human papillomavirus type 16 E7 oncoprotein mediates CCNA1 promoter methylation. Cancer Sci 106:1333–40.

131. Cicchini L, Westrich JA, Xu T, Vermeer DW, Berger JN, Clambey ET, Lee D, Song JI, Lambert PF, Greer RO, Lee JH, Pyeon D. 2016. Suppression of Antitumor Immune Responses by Human Papillomavirus through Epigenetic Downregulation of CXCL14. MBio 7:e00270–16.

132. Pietenpol JA, Stein RW, Moran E, Yaciuk P, Schlegel R, Lyons RM, Pittelkow MR, Münger K, Howley PM, Moses HL. 1990. TGF-beta 1 inhibition of c-myc transcription and growth in keratinocytes is abrogated by viral transforming proteins with pRB binding domains. Cell 61:777–85.

133. Villa LL, Vieira KB, Pei XF, Schlegel R. 1992. Differential effect of tumor necrosis factor on proliferation of primary human keratinocytes and cell lines containing human papillomavirus types 16 and 18. Mol Carcinog 6:5–9.

134. Vieira KB, Goldstein DJ, Villa LL. 1996. Tumor necrosis factor alpha interferes with the cell cycle of normal and papillomavirus-immortalized human keratinocytes. Cancer Res 56:2452–7.

135. Perea SE, Massimi P, Banks L. 2000. Human papillomavirus type 16 E7 impairs the activation of the interferon regulatory factor-1. Int J Mol Med 5:661–6.

136. Park JS, Kim EJ, Kwon HJ, Hwang ES, Namkoong SE, Um SJ. 2000. Inactivation of interferon regulatory factor-1 tumor suppressor protein by HPV E7 oncoprotein. Implication for the E7-mediated immune evasion mechanism in cervical carcinogenesis. J Biol Chem 275:6764–9.

137. Doorbar J. 2005. The papillomavirus life cycle. J Clin Virol 32 Suppl 1:S7–15.

130

138. Steenbergen RDM, de Wilde J, Wilting SM, Brink AATP, Snijders PJF, Meijer CJLM. 2005. HPV-mediated transformation of the anogenital tract. J Clin Virol 32 Suppl 1:S25–33.

139. Middleton K, Peh W, Southern S, Griffin H, Sotlar K, Nakahara T, El-Sherif A, Morris L, Seth R, Hibma M, Jenkins D, Lambert P, Coleman N, Doorbar J. 2003. Organization of human papillomavirus productive cycle during neoplastic progression provides a basis for selection of diagnostic markers. J Virol 77:10186–201.

140. von Knebel Doeberitz M. 2002. New markers for cervical dysplasia to visualise the genomic chaos created by aberrant oncogenic papillomavirus infections. Eur J Cancer 38:2229–42.

141. Centers for Disease Control and Prevention (CDC). 2012. Human papillomavirus- associated cancers - United States, 2004-2008. MMWR Morb Mortal Wkly Rep 61:258–61.

142. Ylitalo N, Sørensen P, Josefsson AM, Magnusson PK, Andersen PK, Pontén J, Adami HO, Gyllensten UB, Melbye M. 2000. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet (London, England) 355:2194–8.

143. Lorincz AT, Castle PE, Sherman ME, Scott DR, Glass AG, Wacholder S, Rush BB, Gravitt PE, Schussler JE, Schiffman M. 2002. Viral load of human papillomavirus and risk of CIN3 or cervical cancer. Lancet (London, England) 360:228–9.

144. Klaes R, Woerner SM, Ridder R, Wentzensen N, Duerst M, Schneider A, Lotz B, Melsheimer P, von Knebel Doeberitz M. 1999. Detection of high-risk cervical intraepithelial neoplasia and cervical cancer by amplification of transcripts derived from integrated papillomavirus oncogenes. Cancer Res 59:6132–6.

145. Kumar R, Rai AK, Das D, Das R, Kumar RS, Sarma A, Sharma S, Kataki AC, Ramteke A. 2015. Alcohol and Tobacco Increases Risk of High Risk HPV Infection in Head and Neck Cancer Patients: Study from North-East Region of India. PLoS One 10:e0140700.

146. Applebaum KM, Furniss CS, Zeka A, Posner MR, Smith JF, Bryan J, Eisen EA, Peters ES, McClean MD, Kelsey KT. 2007. Lack of Association of Alcohol and Tobacco with HPV16-Associated Head and Neck Cancer. JNCI J Natl Cancer Inst 99:1801–1810.

147. Smith EM, Hoffman HT, Summersgill KS, Kirchner HL, Turek LP, Haugen TH. 1998. Human papillomavirus and risk of oral cancer. Laryngoscope 108:1098–1103.

148. Muñoz N, Bosch FX, de Sanjosé S, Tafur L, Izarzugaza I, Gili M, Viladiu P, Navarro C, Martos C, Ascunce N. 1992. The causal link between human papillomavirus and invasive cervical cancer: a population-based case-control study in Colombia and Spain. Int J cancer 52:743–9.

131

149. Bosch FX, Manos MM, Muñoz N, Sherman M, Jansen AM, Peto J, Schiffman MH, Moreno V, Kurman R, Shah K V. 1995. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. International biological study on cervical cancer (IBSCC) Study Group. J Natl Cancer Inst 87:796–802.

150. Schiffman MH. 1995. New epidemiology of human papillomavirus infection and cervical neoplasia. J Natl Cancer Inst 87:1345–7.

151. Schlecht NF, Kulaga S, Robitaille J, Ferreira S, Santos M, Miyamura RA, Duarte- Franco E, Rohan TE, Ferenczy A, Villa LL, Franco EL. 2001. Persistent human papillomavirus infection as a predictor of cervical intraepithelial neoplasia. JAMA 286:3106–14.

152. Peto J, Gilham C, Deacon J, Taylor C, Evans C, Binns W, Haywood M, Elanko N, Coleman D, Yule R, Desai M. 2004. Cervical HPV infection and neoplasia in a large population-based prospective study: the Manchester cohort. Br J Cancer 91:942–53.

153. 2006 Chapter 5 Updating the natural history of HPV and anogenital cancer.

154. Winer RL, Hughes JP, Feng Q, Xi LF, Cherne S, O’Reilly S, Kiviat NB, Koutsky LA. 2011. Early natural history of incident, type-specific human papillomavirus infections in newly sexually active young women. Cancer Epidemiol Biomarkers Prev 20:699–707.

155. Koshiol J, Lindsay L, Pimenta JM, Poole C, Jenkins D, Smith JS. 2008. Persistent Human Papillomavirus Infection and Cervical Neoplasia: A Systematic Review and Meta-Analysis. Am J Epidemiol 168:123–137.

156. Kreimer AR, Katki HA, Schiffman M, Wheeler CM, Castle PE. 2007. Viral Determinants of Human Papillomavirus Persistence following Loop Electrical Excision Procedure Treatment for Cervical Intraepithelial Neoplasia Grade 2 or 3. Cancer Epidemiol Biomarkers Prev 16:11–16.

157. Robertson KD. 2005. DNA methylation and human disease. Nat Rev Genet 6:597– 610.

158. Bird A, Nan X, Ng H-H, Johnson CA, Laherty CD, Turner BM, Eisenman RN. 1998. Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex. Nature 393:386–389.

159. Jones PL, Veenstra GJ, Wade PA, Vermaak D, Kass SU, Landsberger N, Strouboulis J, Wolffe AP. 1998. Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet 19:187–91.

160. Fuks F, Burgers WA, Brehm A, Hughes-Davies L, Kouzarides T. 2000. DNA methyltransferase Dnmt1 associates with histone deacetylase activity. Nat Genet 24:88–91.

161. Feinberg AP, Tycko B. 2004. The history of cancer epigenetics. Nat Rev Cancer 4:143–53.

132

162. Kostareli E, Holzinger D, Bogatyrova O, Hielscher T, Wichmann G, Keck M, Lahrmann B, Grabe N, Flechtenmacher C, Schmidt CR, Seiwert T, Dyckhoff G, Dietz A, Höfler D, Pawlita M, Benner A, Bosch FX, Plinkert P, Plass C, Weichenhan D, Hess J. 2013. HPV-related methylation signature predicts survival in oropharyngeal squamous cell carcinomas. J Clin Invest 123:2488–501.

163. Kostareli E, Hielscher T, Zucknick M, Baboci L, Wichmann G, Holzinger D, Mücke O, Pawlita M, Del Mistro A, Boscolo-Rizzo P, Da Mosto MC, Tirelli G, Plinkert P, Dietz A, Plass C, Weichenhan D, Hess J. 2016. Gene promoter methylation signature predicts survival of head and neck squamous cell carcinoma patients. Epigenetics 11:61–73.

164. Sakane J, Taniyama K, Miyamoto K, Saito A, Kuraoka K, Nishimura T, Sentani K, Oue N, Yasui W. 2015. Aberrant DNA methylation of DLX4 and SIM1 is a predictive marker for disease progression of uterine cervical low-grade squamous intraepithelial lesion. Diagn Cytopathol 43:462–70.

165. Teschendorff AE, Liu X, Caren H, Pollard SM, Beck S, Widschwendter M, Chen L. 2014. The dynamics of DNA methylation covariation patterns in carcinogenesis. PLoS Comput Biol 10:e1003709.

166. Schütze DM, Kooter JM, Wilting SM, Meijer CJLM, Quint W, Snijders PJF, Steenbergen RDM. 2015. Longitudinal assessment of DNA methylation changes during HPVE6E7-induced immortalization of primary keratinocytes. Epigenetics 10:73–81.

167. Stanley M. 2006. Immune responses to human papillomavirus. Vaccine 24:S16–S22.

168. De Vincenzo R, Conte C, Ricci C, Scambia G, Capelli G. 2014. Long-term efficacy and safety of human papillomavirus vaccination. Int J Womens Health 999.

169. Lehtinen M, Paavonen J, Wheeler CM, Jaisamrarn U, Garland SM, Castellsagué X, Skinner SR, Apter D, Naud P, Salmerón J, Chow S-N, Kitchener H, Teixeira JC, Hedrick J, Limson G, Szarewski A, Romanowski B, Aoki FY, Schwarz TF, Poppe WAJ, De Carvalho NS, Germar MJ V, Peters K, Mindel A, De Sutter P, Bosch FX, David M-P, Descamps D, Struyf F, Dubin G. 2012. Overall efficacy of HPV-16/18 AS04-adjuvanted vaccine against grade 3 or greater cervical intraepithelial neoplasia: 4-year end-of-study analysis of the randomised, double-blind PATRICIA trial. Lancet Oncol 13:89–99.

170. Combes J-D, Pawlita M, Waterboer T, Hammouda D, Rajkumar T, Vanhems P, Snijders P, Herrero R, Franceschi S, Clifford G. 2014. Antibodies against high-risk human papillomavirus proteins as markers for invasive cervical cancer. Int J Cancer 135:2453–61.

171. Maglennon GA, McIntosh PB, Doorbar J. 2014. Immunosuppression facilitates the reactivation of latent papillomavirus infections. J Virol 88:710–6.

172. Henri S, Guilliams M, Poulin LF, Tamoutounour S, Ardouin L, Dalod M, Malissen B. Disentangling the complexity of the skin dendritic cell network. Immunol Cell Biol 88:366–75.

133

173. Heath WR, Carbone FR. 2009. Dendritic cell subsets in primary and secondary T cell responses at body surfaces. Nat Immunol 10:1237–44.

174. Igyarto BZ, Jenison MC, Dudda JC, Roers A, Müller W, Koni PA, Campbell DJ, Shlomchik MJ, Kaplan DH. 2009. Langerhans cells suppress contact hypersensitivity responses via cognate CD4 interaction and langerhans cell-derived IL-10. J Immunol 183:5085–93.

175. Romani N, Clausen BE, Stoitzner P. 2010. Langerhans cells and more: langerin- expressing dendritic cell subsets in the skin. Immunol Rev 234:120–41.

176. Iwasaki A. 2007. Mucosal dendritic cells. Annu Rev Immunol 25:381–418.

177. Da Silva DM, Woodham AW, Skeate JG, Rijkee LK, Taylor JR, Brand HE, Muderspach LI, Roman LD, Yessaian AA, Pham HQ, Matsuo K, Lin YG, McKee GM, Salazar AM, Kast WM. 2015. Langerhans cells from women with cervical precancerous lesions become functionally responsive against human papillomavirus after activation with stabilized Poly-I:C. Clin Immunol 161:197–208.

178. Da Silva DM, Movius CA, Raff AB, Brand HE, Skeate JG, Wong MK, Kast WM. 2014. Suppression of Langerhans cell activation is conserved amongst human papillomavirus α and β genotypes, but not a µ genotype. Virology 452-453:279–86.

179. Fausch SC, Da Silva DM, Rudolf MP, Kast WM. 2002. Human papillomavirus virus- like particles do not activate Langerhans cells: a possible immune escape mechanism used by human papillomaviruses. J Immunol 169:3242–9.

180. Herman L, Hubert P, Herfs M, Kustermans G, Henrotin Y, Bousarghin L, Boniver J, Delvenne P. 2010. The L1 major capsid protein of HPV16 differentially modulates APC trafficking according to the vaccination or natural infection context. Eur J Immunol 40:3075–84.

181. Guha M, Mackman N. 2002. The phosphatidylinositol 3-kinase-Akt pathway limits lipopolysaccharide activation of signaling pathways and expression of inflammatory mediators in human monocytic cells. J Biol Chem 277:32124–32.

182. Iijima N, Goodwin EC, Dimaio D, Iwasaki A. 2013. High-risk human papillomavirus E6 inhibits monocyte differentiation to Langerhans cells. Virology 444:257–62.

183. Matthews K, Leong CM, Baxter L, Inglis E, Yun K, Bäckström BT, Doorbar J, Hibma M. 2003. Depletion of Langerhans cells in human papillomavirus type 16- infected skin is associated with E6-mediated down regulation of E-cadherin. J Virol 77:8378–85.

184. Trimble CL, Clark RA, Thoburn C, Hanson NC, Tassello J, Frosina D, Kos F, Teague J, Jiang Y, Barat NC, Jungbluth AA. 2010. Human papillomavirus 16- associated cervical intraepithelial neoplasia in humans excludes CD8 T cells from dysplastic epithelium. J Immunol 185:7107–14.

134

185. Abd Warif NM, Stoitzner P, Leggatt GR, Mattarollo SR, Frazer IH, Hibma MH. 2015. Langerhans cell homeostasis and activation is altered in hyperplastic human papillomavirus type 16 E7 expressing epidermis. PLoS One 10:e0127155.

186. Tindle RW, Herd K, Doan T, Bryson G, Leggatt GR, Lambert P, Frazer IH, Street M. 2001. Nonspecific down-regulation of CD8+ T-cell responses in mice expressing human papillomavirus type 16 E7 oncoprotein from the keratin-14 promoter. J Virol 75:5985–97.

187. Mackay IR, Rosen FS, Klein J, Sato A. 2000. The HLA System. N Engl J Med 343:702–709.

188. Bottley G, Watherston OG, Hiew Y-L, Norrild B, Cook GP, Blair GE. 2008. High- risk human papillomavirus E7 expression reduces cell-surface MHC class I molecules and increases susceptibility to natural killer cells. Oncogene 27:1794–9.

189. Hoos A, D’Incan C, Gissmann L, Altmann A, Momburg F, Nindl I, Osen W, Schönning BH, Jochmus I. 1996. Human papillomavirus type 16 (HPV 16) E7 and major histocompatibility complex (MHC) class I and II expression in human keratinocytes in culture. Arch Virol 141:449–58.

190. Jochmus I, Dürst M, Reid R, Altmann A, Bijward KE, Gissmann L, Jenson AB. 1993. Major histocompatibility complex and human papillomavirus type 16 E7 expression in high-grade vulvar lesions. Hum Pathol 24:519–24.

191. Bauer M, Wagner H, Lipford GB. 2000. HPV type 16 protein E7 HLA-A2 binding peptides are immunogenic but not processed and presented. Immunol Lett 71:55–9.

192. Peng S, Trimble C, He L, Tsai Y-C, Lin C-T, Boyd DAK, Pardoll D, Hung C-F, Wu T-C. 2006. Characterization of HLA-A2-restricted HPV-16 E7-specific CD8(+) T-cell immune responses induced by DNA vaccines in HLA-A2 transgenic mice. Gene Ther 13:67–77.

193. Riemer AB, Keskin DB, Zhang G, Handley M, Anderson KS, Brusic V, Reinhold B, Reinherz EL. 2010. A conserved E7-derived cytotoxic T lymphocyte epitope expressed on human papillomavirus 16-transformed HLA-A2+ epithelial cancers. J Biol Chem 285:29608–22.

194. Peng S, Frazer IH, Fernando GJ, Zhou J. 1998. Papillomavirus virus-like particles can deliver defined CTL epitopes to the MHC class I pathway. Virology 240:147–57.

195. Navarro F, Llano M, Bellón T, Colonna M, Geraghty DE, López-Botet M. 1999. The ILT2(LIR1) and CD94/NKG2A NK cell receptors respectively recognize HLA-G1 and HLA-E molecules co-expressed on target cells. Eur J Immunol 29:277–83.

196. Braud VM, Allan DS, O’Callaghan CA, Söderström K, D’Andrea A, Ogg GS, Lazetic S, Young NT, Bell JI, Phillips JH, Lanier LL, McMichael AJ. 1998. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 391:795–9.

135

197. Lee N, Llano M, Carretero M, Ishitani A, Navarro F, López-Botet M, Geraghty DE. 1998. HLA-E is a major ligand for the natural killer inhibitory receptor CD94/NKG2A. Proc Natl Acad Sci U S A 95:5199–204.

198. Lanier LL. 2008. Up on the tightrope: natural killer cell activation and inhibition. Nat Immunol 9:495–502.

199. García P, Llano M, de Heredia AB, Willberg CB, Caparrós E, Aparicio P, Braud VM, López-Botet M. 2002. Human T cell receptor-mediated recognition of HLA-E. Eur J Immunol 32:936–44.

200. Moretta L, Romagnani C, Pietra G, Moretta A, Mingari MC. 2003. NK-CTLs, a novel HLA-E-restricted T-cell subset. Trends Immunol 24:136–43.

201. Pietra G, Romagnani C, Falco M, Vitale M, Castriconi R, Pende D, Millo E, Anfossi S, Biassoni R, Moretta L, Mingari MC. 2001. The analysis of the natural killer-like activity of human cytolytic T lymphocytes revealed HLA-E as a novel target for TCR alpha/beta-mediated recognition. Eur J Immunol 31:3687–93.

202. Salerno-Gonçalves R, Fernandez-Viña M, Lewinsohn DM, Sztein MB. 2004. Identification of a human HLA-E-restricted CD8+ T cell subset in volunteers immunized with Salmonella enterica serovar Typhi strain Ty21a typhoid vaccine. J Immunol 173:5852–62.

203. Harriff MJ, Cansler ME, Toren KG, Canfield ET, Kwak S, Gold MC, Lewinsohn DM. 2014. Human lung epithelial cells contain Mycobacterium tuberculosis in a late endosomal vacuole and are efficiently recognized by CD8+ T cells. PLoS One 9:e97515.

204. Romagnani C, Pietra G, Falco M, Mazzarino P, Moretta L, Mingari MC. 2004. HLA-E-restricted recognition of human cytomegalovirus by a subset of cytolytic T lymphocytes. Hum Immunol 65:437–45.

205. Romagnani C, Pietra G, Falco M, Millo E, Mazzarino P, Biassoni R, Moretta A, Moretta L, Mingari MC. 2002. Identification of HLA-E-specific alloreactive T lymphocytes: a cell subset that undergoes preferential expansion in mixed lymphocyte culture and displays a broad cytolytic activity against allogeneic cells. Proc Natl Acad Sci U S A 99:11328–33.

206. Ulbrecht M, Modrow S, Srivastava R, Peterson PA, Weiss EH. 1998. Interaction of HLA-E with peptides and the peptide transporter in vitro: implications for its function in antigen presentation. J Immunol 160:4375–85.

207. Nattermann J, Nischalke HD, Hofmeister V, Ahlenstiel G, Zimmermann H, Leifeld L, Weiss EH, Sauerbruch T, Spengler U. 2005. The HLA-A2 restricted T cell epitope HCV core 35-44 stabilizes HLA-E expression and inhibits cytolysis mediated by natural killer cells. Am J Pathol 166:443–53.

208. Mazzarino P, Pietra G, Vacca P, Falco M, Colau D, Coulie P, Moretta L, Mingari MC. 2005. Identification of effector-memory CMV-specific T lymphocytes that kill CMV-infected target cells in an HLA-E-restricted fashion. Eur J Immunol 35:3240–7.

136

209. Castle PE, Hillier SL, Rabe LK, Hildesheim A, Herrero R, Bratti MC, Sherman ME, Burk RD, Rodriguez AC, Alfaro M, Hutchinson ML, Morales J, Schiffman M. 2001. An association of cervical inflammation with high-grade cervical neoplasia in women infected with oncogenic human papillomavirus (HPV). Cancer Epidemiol Biomarkers Prev 10:1021–7.

210. Koutsky LA, Holmes KK, Critchlow CW, Stevens CE, Paavonen J, Beckmann AM, DeRouen TA, Galloway DA, Vernon D, Kiviat NB. 1992. A cohort study of the risk of cervical intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus infection. N Engl J Med 327:1272–8.

211. Tezal M, Scannapieco FA, Wactawski-Wende J, Hyland A, Marshall JR, Rigual NR, Stoler DL. 2012. Local Inflammation and Human Papillomavirus Status of Head and Neck Cancers. Arch Otolaryngol Neck Surg 138:669.

212. Tjiong MY, van der Vange N, ten Kate FJ, Tjong-A-Hung SP, ter Schegget J, Burger MP, Out TA. 1999. Increased IL-6 and IL-8 levels in cervicovaginal secretions of patients with cervical cancer. Gynecol Oncol 73:285–91.

213. Hsiao Y-P, Yang J-H, Wu W-J, Lin M-H, Sheu G-T. 2013. E6 and E7 of human papillomavirus type 18 and UVB irradiation corporately regulate interleukin-6 and interleukin-8 expressions in basal cell carcinoma. Exp Dermatol 22:672–674.

214. Torres-Poveda K. 2014. Role of IL-10 and TGF-β1 in local immunosuppression in HPV-associated cervical neoplasia. World J Clin Oncol 5:753.

215. Karl T, Seibert N, Stöhr M, Osswald H, Rösl F, Finzer P. 2007. Sulindac induces specific degradation of the HPV oncoprotein E7 and causes growth arrest and apoptosis in cervical carcinoma cells. Cancer Lett 245:103–11.

216. Jakacka M, Ito M, Martinson F, Ishikawa T, Lee EJ, Jameson JL. 2002. An estrogen receptor (ER)alpha deoxyribonucleic acid-binding domain knock-in mutation provides evidence for nonclassical ER pathway signaling in vivo. Mol Endocrinol 16:2188–201.

217. Jakacka M, Ito M, Weiss J, Chien PY, Gehm BD, Jameson JL. 2001. Estrogen receptor binding to DNA is not required for its activity through the nonclassical AP1 pathway. J Biol Chem 276:13615–21.

218. Cunningham M, Gilkeson G. 2011. Estrogen receptors in immunity and autoimmunity. Clin Rev Allergy Immunol 40:66–73.

219. Chung S-H, Lambert PF. 2009. Prevention and treatment of cervical cancer in mice using estrogen receptor antagonists. Proc Natl Acad Sci U S A 106:19467–72.

220. Nair HB, Luthra R, Kirma N, Liu Y-G, Flowers L, Evans D, Tekmal RR. 2005. Induction of aromatase expression in cervical carcinomas: effects of endogenous estrogen on cervical cancer cell proliferation. Cancer Res 65:11164–73.

221. Allen SJ, Crown SE, Handel TM. 2007. Chemokine: receptor structure, interactions, and antagonism. Annu Rev Immunol 25:787–820.

137

222. Meuter S, Moser B. 2008. Constitutive expression of CXCL14 in healthy human and murine epithelial tissues. Cytokine 44:248–55.

223. Banisadr G, Bhattacharyya BJ, Belmadani A, Izen SC, Ren D, Tran PB, Miller RJ. 2011. The chemokine BRAK/CXCL14 regulates synaptic transmission in the adult mouse dentate gyrus stem cell niche. J Neurochem 119:1173–82.

224. García-Andrés C, Torres M. 2010. Comparative expression pattern analysis of the highly conserved chemokines SDF1 and CXCL14 during amniote embryonic development. Dev Dyn 239:2769–77.

225. Hromas R, Broxmeyer HE, Kim C, Nakshatri H, Christopherson K, Azam M, Hou YH. 1999. Cloning of BRAK, a novel divergent CXC chemokine preferentially expressed in normal versus malignant cells. Biochem Biophys Res Commun 255:703–6.

226. Kuang H, Chen Q, Zhang Y, Zhang L, Peng H, Ning L, Cao Y, Duan E. 2009. The cytokine gene CXCL14 restricts human trophoblast cell invasion by suppressing gelatinase activity. Endocrinology 150:5596–605.

227. Cao Q, Chen H, Deng Z, Yue J, Chen Q, Cao Y, Ning L, Lei X, Duan E. 2013. Genetic deletion of Cxcl14 in mice alters uterine NK cells. Biochem Biophys Res Commun 435:664–70.

228. Starnes T, Rasila KK, Robertson MJ, Brahmi Z, Dahl R, Christopherson K, Hromas R. 2006. The chemokine CXCL14 (BRAK) stimulates activated NK cell migration: implications for the downregulation of CXCL14 in malignancy. Exp Hematol 34:1101–5.

229. Schaerli P, Willimann K, Ebert LM, Walz A, Moser B. 2005. Cutaneous CXCL14 targets blood precursors to epidermal niches for Langerhans cell differentiation. Immunity 23:331–42.

230. Meuter S, Schaerli P, Roos RS, Brandau O, Bösl MR, von Andrian UH, Moser B. 2007. Murine CXCL14 is dispensable for dendritic cell function and localization within peripheral tissues. Mol Cell Biol 27:983–92.

231. Shurin G V, Ferris RL, Ferris R, Tourkova IL, Perez L, Lokshin A, Balkir L, Collins B, Chatta GS, Shurin MR. 2005. Loss of new chemokine CXCL14 in tumor tissue is associated with low infiltration by dendritic cells (DC), while restoration of human CXCL14 expression in tumor cells causes attraction of DC both in vitro and in vivo. J Immunol 174:5490–8.

232. Den Boon J, Pyeon D, Wang S, Horswill M, Schiffman M, Sherman M, Zuna R, Wang Z, Hewitt S, Pearson R, Schott M, Chung L, He Q, Lambert P, Walker J, Newton M, Wentzensen N, Ahlquist P. 2015. Molecular transitions from papillomavirus infection to cervical precancer and cancer: role of stromal estrogen receptor signaling. Proc Natl Acad Sci U S A In Press.

138

233. Pyeon D, Newton MA, Lambert PF, den Boon J a, Sengupta S, Marsit CJ, Woodworth CD, Connor JP, Haugen TH, Smith EM, Kelsey KT, Turek LP, Ahlquist P. 2007. Fundamental differences in cell cycle deregulation in human papillomavirus-positive and human papillomavirus-negative head/neck and cervical cancers. Cancer Res 67:4605–4619.

234. Grivennikov SI, Greten FR, Karin M. 2010. Immunity, inflammation, and cancer. Cell 140:883–899.

235. Gildener-Leapman N, Ferris RL, Bauman JE. 2013. Promising systemic immunotherapies in head and neck squamous cell carcinoma. Oral Oncol 49:1089– 1096.

236. Powles T, Eder JP, Fine GD, Braiteh FS, Loriot Y, Cruz C, Bellmunt J, Burris H a, Petrylak DP, Teng SL, Shen X, Boyd Z, Hegde PS, Chen DS, Vogelzang NJ. 2014. MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature 515:558–562.

237. Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJ, Robert L, Chmielowski B, Spasic M, Henry G, Ciobanu V, West a N, Carmona M, Kivork C, Seja E, Cherry G, Gutierrez a J, Grogan TR, Mateus C, Tomasic G, Glaspy J a, Emerson RO, Robins H, Pierce RH, Elashoff D a, Robert C, Ribas a. 2014. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature.

238. Herbst RS, Soria JC, Kowanetz M, Fine GD, Hamid O, Gordon MS, Sosman J a, McDermott DF, Powderly JD, Gettinger SN, Kohrt HE, Horn L, Lawrence DP, Rost S, Leabman M, Xiao Y, Mokatrin a, Koeppen H, Hegde PS, Mellman I, Chen DS, Hodi FS. 2014. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature 515:563–567.

239. Lyles DS. 2000. Cytopathogenesis and inhibition of host gene expression by RNA viruses. Microbiol Mol Biol Rev 64:709–24.

240. Wagner RR, Huang AS. 1966. Inhibition of RNA and interferon synthesis in Krebs-2 cells infected with vesicular stomatitis virus. Virology 28:1–10.

241. Krug RM. 1993. The regulation of export of mRNA from nucleus to cytoplasm. Curr Opin Cell Biol 5:944–9.

242. Plotch SJ, Bouloy M, Ulmanen I, Krug RM. 1981. A unique cap(m7GpppXm)- dependent influenza virion endonuclease cleaves capped RNAs to generate the primers that initiate viral RNA transcription. Cell 23:847–58.

243. Zhu H, Cong JP, Mamtora G, Gingeras T, Shenk T. 1998. Cellular gene expression altered by human cytomegalovirus: global monitoring with oligonucleotide arrays. Proc Natl Acad Sci U S A 95:14470–5.

244. Fernández M, Quiroga JA, Carreño V. 2003. Hepatitis B virus downregulates the human interferon-inducible MxA promoter through direct interaction of precore/core proteins. J Gen Virol 84:2073–82.

139

245. Gyöngyösi E, Szalmás A, Ferenczi A, Póliska S, Kónya J, Veress G. 2015. Transcriptional regulation of genes involved in keratinocyte differentiation by human papillomavirus 16 oncoproteins. Arch Virol 160:389–98.

246. Jiménez-Wences H, Peralta-Zaragoza O, Fernández-Tilapa G. 2014. Human papilloma virus, DNA methylation and microRNA expression in cervical cancer (Review). Oncol Rep 31:2467–76.

247. Allen-Hoffmann BL, Schlosser SJ, Ivarie CA, Sattler CA, Meisner LF, O’Connor SL. 2000. Normal growth and differentiation in a spontaneously immortalized near- diploid human keratinocyte cell line, NIKS. J Invest Dermatol 114:444–55.

248. Flores ER, Allen-Hoffmann BL, Lee D, Sattler CA, Lambert PF. 1999. Establishment of the human papillomavirus type 16 (HPV-16) life cycle in an immortalized human foreskin keratinocyte cell line. Virology 262:344–54.

249. Flores ER, Allen-Hoffmann BL, Lee D, Lambert PF. 2000. The human papillomavirus type 16 E7 oncogene is required for the productive stage of the viral life cycle. J Virol 74:6622–31.

250. Griffin LM, Cicchini L, Xu T, Pyeon D. 2014. Human keratinocyte cultures in the investigation of early steps of human papillomavirus infection. Methods Mol Biol 1195:219–38.

251. Stanley MA, Browne HM, Appleby M, Minson AC. 1989. Properties of a non- tumorigenic human cervical keratinocyte cell line. Int J cancer 43:672–6.

252. Jeon S, Allen-Hoffmann BL, Lambert PF. 1995. Integration of human papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J Virol 69:2989–97.

253. Baker CC, Phelps WC, Lindgren V, Braun MJ, Gonda MA, Howley PM. 1987. Structural and transcriptional analysis of human papillomavirus type 16 sequences in cervical carcinoma cell lines. J Virol 61:962–71.

254. Spanos WC, Nowicki P, Lee DW, Hoover A, Hostager B, Gupta A, Anderson ME, Lee JH. 2009. Immune response during therapy with cisplatin or radiation for human papillomavirus-related head and neck cancer. Arch Otolaryngol Head Neck Surg 135:1137–46.

255. Klug M, Rehli M. Functional analysis of promoter CpG methylation using a CpG-free luciferase reporter vector. Epigenetics 1:127–30.

256. Rishi V, Bhattacharya P, Chatterjee R, Rozenberg J, Zhao J, Glass K, Fitzgerald P, Vinson C. 2010. CpG methylation of half-CRE sequences creates C/EBP binding sites that activate some tissue-specific genes. Proc Natl Acad Sci 107:20311–20316.

257. Liang C-C, Park AY, Guan J-L. 2007. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protoc 2:329–333.

140

258. Hoover AC, Spanos WC, Harris GF, Anderson ME, Klingelhutz AJ, Lee JH. 2007. The role of human papillomavirus 16 E6 in anchorage-independent and invasive growth of mouse tonsil epithelium. Arch Otolaryngol Head Neck Surg 133:495–502.

259. R Development Core Team. 2011. R: a language and environment for statistical computing. R Found Stat Comput ISBN: 3-90.

260. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. 2003. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4:249–64.

261. Benjamini Y, Hochberg Y. 1995. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc 57:289–300.

262. Metsalu T, Vilo J. 2015. ClustVis: a web tool for visualizing clustering of multivariate data using Principal Component Analysis and heatmap. Nucleic Acids Res 43:W566– 70.

263. Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, Irizarry RA. 2014. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 30:1363–9.

264. Maksimovic J, Gordon L, Oshlack A. 2012. SWAN: Subset-quantile within array normalization for illumina infinium HumanMethylation450 BeadChips. Genome Biol 13:R44.

265. Jaffe AE, Murakami P, Lee H, Leek JT, Fallin MD, Feinberg AP, Irizarry RA. 2012. Bump hunting to identify differentially methylated regions in epigenetic epidemiology studies. Int J Epidemiol 41:200–9.

266. Goecks J, Nekrutenko A, Taylor J, Galaxy Team. 2010. Galaxy: a comprehensive approach for supporting accessible, reproducible, and transparent computational research in the life sciences. Genome Biol 11:R86.

267. Blankenberg D, Von Kuster G, Coraor N, Ananda G, Lazarus R, Mangan M, Nekrutenko A, Taylor J. 2010. Galaxy: a web-based genome analysis tool for experimentalists. Curr Protoc Mol Biol Chapter 19:Unit 19.10.1–21.

268. Giardine B, Riemer C, Hardison RC, Burhans R, Elnitski L, Shah P, Zhang Y, Blankenberg D, Albert I, Taylor J, Miller W, Kent WJ, Nekrutenko A. 2005. Galaxy: a platform for interactive large-scale genome analysis. Genome Res 15:1451–5.

269. Bailey TL, Elkan C. 1994. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2:28–36.

270. Pyeon D, Lambert PF, Ahlquist P. 2005. Production of infectious human papillomavirus independently of viral replication and epithelial cell differentiation. Proc Natl Acad Sci U S A 102:9311–6.

141

271. Johnson KM, Kines RC, Roberts JN, Lowy DR, Schiller JT, Day PM. 2009. Role of heparan sulfate in attachment to and infection of the murine female genital tract by human papillomavirus. J Virol 83:2067–74.

272. de Martel C, Ferlay J, Franceschi S, Vignat J, Bray F, Forman D, Plummer M. 2012. Global burden of cancers attributable to infections in 2008: a review and synthetic analysis. Lancet Oncol.

273. Tindle RW. 2002. Immune evasion in human papillomavirus-associated cervical cancer. Nat Rev Cancer 2:59–65.

274. Bergot AS, Ford N, Leggatt GR, Wells JW, Frazer IH, Grimbaldeston MA. 2014. HPV16-E7 expression in squamous epithelium creates a local immune suppressive environment via CCL2- and CCL5- mediated recruitment of mast cells. PLoS Pathog 10:e1004466.

275. Pardoll DM. 2012. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12:252–264.

276. Hamid O, Robert C, Daud A, Hodi FS, Hwu W-J, Kefford R, Wolchok JD, Hersey P, Joseph RW, Weber JS, Dronca R, Gangadhar TC, Patnaik A, Zarour H, Joshua AM, Gergich K, Elassaiss-Schaap J, Algazi A, Mateus C, Boasberg P, Tumeh PC, Chmielowski B, Ebbinghaus SW, Li XN, Kang SP, Ribas A. 2013. Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med 369:134–144.

277. Hromas R, Broxmeyer HE, Kim C, Nakshatri H, Christopherson K, Azam M, Hou YH. 1999. Cloning of BRAK, a novel divergent CXC chemokine preferentially expressed in normal versus malignant cells. Biochem Biophys Res Commun 255:703–706.

278. Shellenberger TD, Wang M, Gujrati M, Jayakumar A, Strieter RM, Burdick MD, Ioannides CG, Efferson CL, El-Naggar AK, Roberts D, Clayman GL, Frederick MJ. 2004. BRAK/CXCL14 is a potent inhibitor of angiogenesis and a chemotactic factor for immature dendritic cells. Cancer Res 64:8262–8270.

279. Starnes T, Rasila KK, Robertson MJ, Brahmi Z, Dahl R, Christopherson K, Hromas R. 2006. The chemokine CXCL14 (BRAK) stimulates activated NK cell migration: implications for the downregulation of CXCL14 in malignancy. Exp Hematol 34:1101–1105.

280. Frederick MJ, Henderson Y, Xu X, Deavers MT, Sahin AA, Wu H, Lewis DE, El- Naggar AK, Clayman GL. 2000. In vivo expression of the novel CXC chemokine BRAK in normal and cancerous human tissue. Am J Pathol 156:1937–1950.

281. Ito S, Ozawa S, Ikoma T, Yajima N, Kiyono T, Hata RI. 2010. Expression of a chemokine BRAK/CXCL14 in oral floor carcinoma cells reduces the settlement rate of the cells and suppresses their proliferation in vivo. Biomed Res 31:199–206.

142

282. Shurin G V, Ferris RL, Ferris R, Tourkova IL, Perez L, Lokshin A, Balkir L, Collins B, Chatta GS, Shurin MR. 2005. Loss of new chemokine CXCL14 in tumor tissue is associated with low infiltration by dendritic cells (DC), while restoration of human CXCL14 expression in tumor cells causes attraction of DC both in vitro and in vivo. J Immunol 174:5490–5498.

283. Song EY, Shurin MR, Tourkova IL, Gutkin DW, Shurin G V. 2010. Epigenetic Mechanisms of Promigratory Chemokine CXCL14 Regulation in Human Prostate Cancer Cells. Cancer Res 70:4394–4401.

284. Tessema M, Klinge DM, Yingling CM, Do K, Van Neste L, Belinsky SA. 2010. Re- expression of CXCL14, a common target for epigenetic silencing in lung cancer, induces tumor necrosis. Oncogene 29:5159–5170.

285. Ozawa S, Kato Y, Kubota E, Hata R-I. 2009. BRAK/CXCL14 expression in oral carcinoma cells completely suppresses tumor cell xenografts in SCID mouse. Biomed Res 30:315–318.

286. Ozawa S, Kato Y, Komori R, Maehata Y, Kubota E, Hata R-I. 2006. BRAK/CXCL14 expression suppresses tumor growth in vivo in human oral carcinoma cells. Biochem Biophys Res Commun 348:406–412.

287. Hata R-I, Izukuri K, Kato Y, Sasaki S, Mukaida N, Maehata Y, Miyamoto C, Akasaka T, Yang X, Nagashima Y, Takeda K, Kiyono T, Taniguchi M. 2015. Suppressed rate of carcinogenesis and decreases in tumour volume and lung metastasis in CXCL14/BRAK transgenic mice. Sci Rep 5:9083.

288. Tanegashima K, Suzuki K, Nakayama Y, Tsuji K, Shigenaga A, Otaka A, Hara T. 2013. CXCL14 is a natural inhibitor of the CXCL12-CXCR4 signaling axis. FEBS Lett.

289. Aasen T, Hodgins MB, Edward M, Graham S V. 2003. The relationship between connexins, gap junctions, tissue architecture and tumour invasion, as studied in a novel in vitro model of HPV-16-associated cervical cancer progression. Oncogene 22:7969–7980.

290. Cancer Genome Atlas N. 2015. Comprehensive genomic characterization of head and neck squamous cell carcinomas. Nature 517:576–582.

291. Flores ER, Allen-Hoffmann BL, Lee D, Lambert PF. 2000. The human papillomavirus type 16 E7 oncogene is required for the productive stage of the viral life cycle. J Virol 74:6622–6631.

292. Spanos WC, Nowicki P, Lee DW, Hoover A, Hostager B, Gupta A, Anderson ME, Lee JH. 2009. Immune response during therapy with cisplatin or radiation for human papillomavirus-related head and neck cancer. Arch Otolaryngol Head Neck Surg 135:1137–1146.

293. Jones PA, Taylor SM. 1980. Cellular differentiation, cytidine analogs and DNA methylation. Cell 20:85–93.

143

294. Cao B, Yang Y, Pan Y, Jia Y, Brock M V, Herman JG, Guo M. 2013. Epigenetic silencing of CXCL14 induced colorectal cancer migration and invasion. Discov Med 16:137–147.

295. Sato K, Ozawa S, Izukuri K, Kato Y, Hata R. 2010. Expression of tumour- suppressing chemokine BRAK/CXCL14 reduces cell migration rate of HSC-3 tongue carcinoma cells and stimulates attachment to collagen and formation of elongated focal adhesions in vitro. Cell Biol Int 34:513–522.

296. Stransky N, Egloff AM, Tward AD, Kostic AD, Cibulskis K, Sivachenko A, Kryukov G V, Lawrence MS, Sougnez C, McKenna A, Shefler E, Ramos AH, Stojanov P, Carter SL, Voet D, Cortés ML, Auclair D, Berger MF, Saksena G, Guiducci C, Onofrio RC, Parkin M, Romkes M, Weissfeld JL, Seethala RR, Wang L, Rangel-Escareño C, Fernandez-Lopez JC, Hidalgo-Miranda A, Melendez- Zajgla J, Winckler W, Ardlie K, Gabriel SB, Meyerson M, Lander ES, Getz G, Golub TR, Garraway LA, Grandis JR. 2011. The mutational landscape of head and neck squamous cell carcinoma. Science (80- ) 333:1157–1160.

297. Jabbar SF, Abrams L, Glick A, Lambert PF. 2009. Persistence of high-grade cervical dysplasia and cervical cancer requires the continuous expression of the human papillomavirus type 16 E7 oncogene. Cancer Res 69:4407–4414.

298. Jabbar SF, Park S, Schweizer J, Berard-Bergery M, Pitot HC, Lee D, Lambert PF. 2012. Cervical Cancers Require the Continuous Expression of the Human Papillomavirus Type 16 E7 Oncoprotein Even in the Presence of the Viral E6 Oncoprotein. Cancer Res.

299. Raman D, Baugher PJ, Thu YM, Richmond A. 2007. Role of chemokines in tumor growth. Cancer Lett 256:137–165.

300. Zlotnik A, Burkhardt AM, Homey B. 2011. Homeostatic chemokine receptors and organ-specific metastasis. Nat Rev Immunol 11:597–606.

301. Maehata Y, Ozawa S, Kobayashi K, Kato Y, Yoshino F, Miyamoto C, Izukuri K, Kubota E, Hata R-I, Lee MC-I. 2010. Reactive oxygen species (ROS) reduce the expression of BRAK/CXCL14 in human head and neck squamous cell carcinoma cells. Free Radic Res 44:913–924.

302. Cao Q, Chen H, Deng Z, Yue J, Chen Q, Cao Y, Ning L, Lei X, Duan E. 2013. Genetic deletion of Cxcl14 in mice alters uterine NK cells. Biochem Biophys Res Commun 435:664–670.

303. Wang W, Huang P, Zhang L, Wei J, Xie Q, Sun Q, Zhou X, Xie H, Zhou L, Zheng S. 2013. Antitumor efficacy of CXCL14 in hepatocellular carcinoma in vitro and in vivo. Cancer Sci.

304. Gu XL, Ou ZL, Lin FJ, Yang XL, Luo JM, Shen ZZ, Shao ZM. 2012. Expression of CXCL14 and its anticancer role in breast cancer. Breast Cancer Res Treat 135:725– 735.

144

305. Saunthararajah Y, Sekeres M, Advani A, Mahfouz R, Durkin L, Radivoyevitch T, Englehaupt R, Juersivich J, Cooper K, Husseinzadeh H, Przychodzen B, Rump M, Hobson S, Earl M, Sobecks R, Dean R, Reu F, Tiu R, Hamilton B, Copelan E, Lichtin A, Hsi E, Kalaycio M, Maciejewski J. 2015. Evaluation of noncytotoxic DNMT1-depleting therapy in patients with myelodysplastic syndromes. J Clin Invest 125:1043–1055.

306. Saleh MH, Wang L, Goldberg MS. 2015. Improving cancer immunotherapy with DNA methyltransferase inhibitors. Cancer Immunol Immunother.

307. Chan HW, Kurago ZB, Stewart CA, Wilson MJ, Martin MP, Mace BE, Carrington M, Trowsdale J, Lutz CT. 2003. DNA methylation maintains allele-specific KIR gene expression in human natural killer cells. J Exp Med 197:245–255.

308. Bruniquel D, Schwartz RH. 2003. Selective, stable demethylation of the interleukin-2 gene enhances transcription by an active process. Nat Immunol 4:235–240.

309. Northrop JK, Thomas RM, Wells AD, Shen H. 2006. Epigenetic remodeling of the IL-2 and IFN-gamma loci in memory CD8 T cells is influenced by CD4 T cells. J Immunol 177:1062–1069.

310. Wang L, Amoozgar Z, Huang J, Saleh MH, Xing D, Orsulic S, Goldberg MS. 2015. Decitabine Enhances Lymphocyte Migration and Function and Synergizes with CTLA-4 Blockade in a Murine Ovarian Cancer Model. Cancer Immunol Res 3:1030– 1041.

311. Peng D, Kryczek I, Nagarsheth N, Zhao L, Wei S, Wang W, Sun Y, Zhao E, Vatan L, Szeliga W, Kotarski J, Tarkowski R, Dou Y, Cho K, Hensley-Alford S, Munkarah A, Liu R, Zou W. 2015. Epigenetic silencing of TH1-type chemokines shapes tumour immunity and immunotherapy. Nature 527:249–253.

312. Vivier E, Tomasello E, Baratin M, Walzer T, Ugolini S. 2008. Functions of natural killer cells. Nat Immunol 9:503–510.

313. Demers KR, Reuter MA, Betts MR. 2013. CD8(+) T-cell effector function and transcriptional regulation during HIV pathogenesis. Immunol Rev 254:190–206.

314. Caserta S, Borger JG, Zamoyska R. 2012. Central and Effector Memory CD4 and CD8 T-Cell Responses to Tumor-Associated Antigens. Crit Rev Immunol 32:97–126.

315. Mocikat R, Braumüller H, Gumy A, Egeter O, Ziegler H, Reusch U, Bubeck A, Louis J, Mailhammer R, Riethmüller G, Koszinowski U, Röcken M. 2003. Natural killer cells activated by MHC class ILow targets prime dendritic cells to induce protective CD8 T cell responses. Immunity 19:561–569.

316. Kalinski P, Nakamura Y, Watchmaker P, Giermasz A, Muthuswamy R, Mailliard RB. 2006. Helper roles of NK and CD8+ T cells in the induction of tumor immunity. Polarized dendritic cells as cancer vaccines. Immunol Res 36:137–146.

317. Egawa N, Egawa K, Griffin H, Doorbar J. 2015. Human Papillomaviruses; Epithelial Tropisms, and the Development of Neoplasia. Viruses 7:3863–3890.

145

318. Centers for Disease Control and Prevention. 2016. Genital HPV Infection-CDC Fact Sheet.

319. Tommasino M. 2014. The human papillomavirus family and its role in carcinogenesis. Semin Cancer Biol 26:13–21.

320. Louvanto K, Syrjänen KJ, Rintala MAM, Grénman SE, Syrjänen SM. 2010. Genotype-specific clearance of genital human papillomavirus (HPV) infections among mothers in the Finnish family HPV study. J Clin Microbiol 48:2665–71.

321. Adhya D, Basu A. 2010. Epigenetic modulation of host: new insights into immune evasion by viruses. J Biosci 35:647–63.

322. Niller HH, Szenthe K, Minarovits J. 2014. Epstein-Barr virus-host cell interactions: an epigenetic dialog? Front Genet 5.

323. Birdwell CE, Queen KJ, Kilgore PCSR, Rollyson P, Trutschl M, Cvek U, Scott RS. 2014. Genome-wide DNA methylation as an epigenetic consequence of Epstein- Barr virus infection of immortalized keratinocytes. J Virol 88:11442–58.

324. de Wilde J, Kooter JM, Overmeer RM, Claassen-Kramer D, Meijer CJLM, Snijders PJF, Steenbergen RDM. 2010. hTERT promoter activity and CpG methylation in HPV-induced carcinogenesis. BMC Cancer 10:271.

325. Chujan S, Kitkumthorn N, Siriangkul S, Mutirangura A. 2014. CCNA1 promoter methylation: a potential marker for grading Papanicolaou smear cervical squamous intraepithelial lesions. Asian Pac J Cancer Prev 15:7971–5.

326. Cicchini L, Westrich JA, Xu T, Vermeer DW, Berger JN, Clambey ET, Lee D, Song JI, Lambert PF, Greer RO, Lee JH, Pyeon D. 2016. Suppression of Antitumor Immune Responses by Human Papillomavirus through Epigenetic Downregulation of CXCL14. MBio 7:e00270–16.

327. Rusan M, Li YY, Hammerman PS. 2015. Genomic landscape of human papillomavirus-associated cancers. Clin Cancer Res 21:2009–19.

328. Zhu D, Ye M, Zhang W. 2015. E6/E7 oncoproteins of high risk HPV-16 upregulate MT1-MMP, MMP-2 and MMP-9 and promote the migration of cervical cancer cells. Int J Clin Exp Pathol 8:4981–9.

329. Kaewprag J, Umnajvijit W, Ngamkham J, Ponglikitmongkol M. 2013. HPV16 oncoproteins promote cervical cancer invasiveness by upregulating specific matrix metalloproteinases. PLoS One 8:e71611.

330. Brabletz T, Jung A, Dag S, Hlubek F, Kirchner T. 1999. beta-catenin regulates the expression of the matrix metalloproteinase-7 in human colorectal cancer. Am J Pathol 155:1033–8.

146

331. Yang JQ, Zhao W, Duan H, Robbins ME, Buettner GR, Oberley LW, Domann FE. 2001. v-Ha-RaS oncogene upregulates the 92-kDa type IV collagenase (MMP-9) gene by increasing cellular superoxide production and activating NF-kappaB. Free Radic Biol Med 31:520–9.

332. Thierry-Mieg D, Thierry-Mieg J. 2006. AceView: a comprehensive cDNA-supported gene and transcripts annotation. Genome Biol 7 Suppl 1:S12.1–14.

333. Sartor MA, Dolinoy DC, Jones TR, Colacino JA, Prince MEP, Carey TE, Rozek LS. 2011. Genome-wide methylation and expression differences in HPV(+) and HPV(-) squamous cell carcinoma cell lines are consistent with divergent mechanisms of carcinogenesis. Epigenetics 6:777–787.

334. Longo ALB, Rettori MM, de Carvalho AC, Kowalski LP, Carvalho AL, Vettore AL. 2014. Evaluation of the methylation profile of exfoliated cell samples from patients with head and neck squamous cell carcinoma. Head Neck 36:631–7.

335. Tsuboi K, Iida S, Inagaki H, Kato M, Hayami Y, Hanamura I, Miura K, Harada S, Kikuchi M, Komatsu H, Banno S, Wakita A, Nakamura S, Eimoto T, Ueda R. 2000. MUM1/IRF4 expression as a frequent event in mature lymphoid malignancies. Leukemia 14:449–56.

336. Nilsson G, Kannius-Janson M. 2016. Forkhead Box F1 promotes breast cancer cell migration by upregulating lysyl oxidase and suppressing Smad2/3 signaling. BMC Cancer 16:142.

337. Dennis GJ, Mond JJ. 1986. Corticosteroid-induced suppression of murine B cell immune response antigens. J Immunol 136:1600–4.

338. Snyder DS, Unanue ER. 1982. Corticosteroids inhibit murine macrophage Ia expression and interleukin 1 production. J Immunol 129:1803–5.

339. Celada A, McKercher S, Maki RA. 1993. Repression of major histocompatibility complex IA expression by glucocorticoids: the glucocorticoid receptor inhibits the DNA binding of the X box DNA binding protein. J Exp Med 177:691–8.

340. Lee N, Goodlett DR, Ishitani A, Marquardt H, Geraghty DE. 1998. HLA-E surface expression depends on binding of TAP-dependent peptides derived from certain HLA class I signal sequences. J Immunol 160:4951–60.

341. Matsumoto K, Leggatt GR, Zhong J, Liu X, de Kluyver RL, Peters T, Fernando GJP, Liem A, Lambert PF, Frazer IH. 2004. Impaired antigen presentation and effectiveness of combined active/passive immunotherapy for epithelial tumors. J Natl Cancer Inst 96:1611–9.

342. Derré L, Corvaisier M, Charreau B, Moreau A, Godefroy E, Moreau-Aubry A, Jotereau F, Gervois N. 2006. Expression and release of HLA-E by melanoma cells and melanocytes: potential impact on the response of cytotoxic effector cells. J Immunol 177:3100–7.

147

343. Wolpert F, Happold C, Reifenberger G, Florea A-M, Deenen R, Roth P, Neidert MC, Lamszus K, Westphal M, Weller M, Eisele G. 2015. Interferon-β Modulates the Innate Immune Response against Glioblastoma Initiating Cells. PLoS One 10:e0139603.

344. Buitrago-Perez A, Garaulet G, Vazquez-Carballo A, Paramio J, Garcia-Escudero R. 2009. Molecular Signature of HPV-Induced Carcinogenesis: pRb, p53 and Gene Expression Profiling. Curr Genomics 10:26–34.

345. Marsit CJ, Christensen BC, Houseman EA, Karagas MR, Wrensch MR, Yeh R-F, Nelson HH, Wiemels JL, Zheng S, Posner MR, McClean MD, Wiencke JK, Kelsey KT. 2009. Epigenetic profiling reveals etiologically distinct patterns of DNA methylation in head and neck squamous cell carcinoma. Carcinogenesis 30:416–422.

346. Karim R, Meyers C, Backendorf C, Ludigs K, Offringa R, van Ommen G-JB, Melief CJM, van der Burg SH, Boer JM. 2011. Human papillomavirus deregulates the response of a cellular network comprising of chemotactic and proinflammatory genes. PLoS One 6:e17848.

347. Niebler M, Qian X, Höfler D, Kogosov V, Kaewprag J, Kaufmann AM, Ly R, Böhmer G, Zawatzky R, Rösl F, Rincon-Orozco B. 2013. Post-translational control of IL-1β via the human papillomavirus type 16 E6 oncoprotein: a novel mechanism of innate immune escape mediated by the E3-ubiquitin ligase E6-AP and p53. PLoS Pathog 9:e1003536.

348. Netea MG, Simon A, van de Veerdonk F, Kullberg B-J, Van der Meer JWM, Joosten LAB. 2010. IL-1beta processing in host defense: beyond the inflammasomes. PLoS Pathog 6:e1000661.

349. Alcamí A, Smith GL. 1992. A soluble receptor for interleukin-1 beta encoded by vaccinia virus: a novel mechanism of virus modulation of the host response to infection. Cell 71:153–67.

350. Gregory SM, Davis BK, West JA, Taxman DJ, Matsuzawa S, Reed JC, Ting JPY, Damania B. 2011. Discovery of a viral NLR homolog that inhibits the inflammasome. Science 331:330–4.

351. Moreau P, Mouillot G, Rousseau P, Marcou C, Dausset J, Carosella ED. 2003. HLA-G gene repression is reversed by demethylation. Proc Natl Acad Sci U S A 100:1191–6.

352. Serrano A, Tanzarella S, Lionello I, Mendez R, Traversari C, Ruiz-Cabello F, Garrido F. 2001. Rexpression of HLA class I antigens and restoration of antigen- specific CTL response in melanoma cells following 5-aza-β’-deoxycytidine treatment. Int J cancer 94:243–51.

353. da Silva JS, Slowik R, Bicalho M da G. 2013. Considerations on regulatory sequences of the distal promoter region of the HLA-G gene. Hum Immunol 74:473–7.

354. Stulberg MJ, Wright PW, Dang H, Hanson RJ, Miller JS, Anderson SK. 2007. Identification of distal KIR promoters and transcripts. Genes Immun 8:124–30.

148

355. Krawczyk M, Peyraud N, Rybtsova N, Masternak K, Bucher P, Barras E, Reith W. 2004. Long distance control of MHC class II expression by multiple distal enhancers regulated by regulatory factor X complex and CIITA. J Immunol 173:6200–10.

356. Tie F, Banerjee R, Stratton CA, Prasad-Sinha J, Stepanik V, Zlobin A, Diaz MO, Scacheri PC, Harte PJ. 2009. CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila Polycomb silencing. Development 136:3131–41.

357. Zhou F, Chen J, Zhao K-N. 2013. Human papillomavirus 16-encoded E7 protein inhibits IFN--mediated MHC class I antigen presentation and CTL-induced lysis by blocking IRF-1 expression in mouse keratinocytes. J Gen Virol 94:2504–14.

358. Yoshihama S, Roszik J, Downs I, Meissner TB, Vijayan S, Chapuy B, Sidiq T, Shipp MA, Lizee GA, Kobayashi KS. 2016. NLRC5/MHC class I transactivator is a target for immune evasion in cancer. Proc Natl Acad Sci U S A 113:5999–6004.

359. Yao Y, Wang Y, Chen F, Huang Y, Zhu S, Leng Q, Wang H, Shi Y, Qian Y. 2012. NLRC5 regulates MHC class I antigen presentation in host defense against intracellular pathogens. Cell Res 22:836–47.

360. Wei XH, Orr HT. 1990. Differential expression of HLA-E, HLA-F, and HLA-G transcripts in human tissue. Hum Immunol 29:131–42.

361. Kaiser BK, Barahmand-Pour F, Paulsene W, Medley S, Geraghty DE, Strong RK. 2005. Interactions between NKG2x immunoreceptors and HLA-E ligands display overlapping affinities and thermodynamics. J Immunol 174:2878–84.

362. Michaëlsson J, Teixeira de Matos C, Achour A, Lanier LL, Kärre K, Söderström K. 2002. A signal peptide derived from hsp60 binds HLA-E and interferes with CD94/NKG2A recognition. J Exp Med 196:1403–14.

363. Foroni I, Rita A, Filipe B, Santos M, Lima M, Bruges-Armas J. 2014. HLA-E, HLA- F and HLA-G — The Non-Classical Side of the MHC Cluster.HLA and Associated Important Diseases. InTech.

364. Davis ZB, Cogswell A, Scott H, Mertsching A, Boucau J, Wambua D, Le Gall S, Planelles V, Campbell KS, Barker E. 2016. A Conserved HIV-1-Derived Peptide Presented by HLA-E Renders Infected T-cells Highly Susceptible to Attack by NKG2A/CD94-Bearing Natural Killer Cells. PLoS Pathog 12:e1005421.

365. Schulte D, Vogel M, Langhans B, Krämer B, Körner C, Nischalke HD, Steinberg V, Michalk M, Berg T, Rockstroh JK, Sauerbruch T, Spengler U, Nattermann J. 2009. The HLA-E(R)/HLA-E(R) genotype affects the natural course of hepatitis C virus (HCV) infection and is associated with HLA-E-restricted recognition of an HCV- derived peptide by interferon-gamma-secreting human CD8(+) T cells. J Infect Dis 200:1397–401.

366. Gooden M, Lampen M, Jordanova ES, Leffers N, Trimbos JB, van der Burg SH, Nijman H, van Hall T. 2011. HLA-E expression by gynecological cancers restrains tumor-infiltrating CD8+ T lymphocytes. Proc Natl Acad Sci U S A 108:10656–61.

149

367. de Kruijf EM, Sajet A, van Nes JGH, Natanov R, Putter H, Smit VTHBM, Liefers GJ, van den Elsen PJ, van de Velde CJH, Kuppen PJK. 2010. HLA-E and HLA-G expression in classical HLA class I-negative tumors is of prognostic value for clinical outcome of early breast cancer patients. J Immunol 185:7452–9.

368. Zeestraten ECM, Reimers MS, Saadatmand S, Dekker J-WT, Liefers GJ, van den Elsen PJ, van de Velde CJH, Kuppen PJK. 2014. Combined analysis of HLA class I, HLA-E and HLA-G predicts prognosis in colon cancer patients. Br J Cancer 110:459–468.

369. Wischhusen J, Friese MA, Mittelbronn M, Meyermann R, Weller M. 2005. HLA-E protects glioma cells from NKG2D-mediated immune responses in vitro: implications for immune escape in vivo. J Neuropathol Exp Neurol 64:523–8.

370. Hardtke S, Ohl L, Förster R. 2005. Balanced expression of CXCR5 and CCR7 on follicular T helper cells determines their transient positioning to lymph node follicles and is essential for efficient B-cell help. Blood 106:1924–31.

371. López-Giral S, Quintana NE, Cabrerizo M, Alfonso-Pérez M, Sala-Valdés M, De Soria VGG, Fernández-Rañada JM, Fernández-Ruiz E, Muñoz C. 2004. Chemokine receptors that mediate B cell homing to secondary lymphoid tissues are highly expressed in B cell chronic lymphocytic leukemia and non-Hodgkin lymphomas with widespread nodular dissemination. J Leukoc Biol 76:462–71.

372. Wengner AM, Höpken UE, Petrow PK, Hartmann S, Schurigt U, Bräuer R, Lipp M. 2007. CXCR5- and CCR7-dependent lymphoid neogenesis in a murine model of chronic antigen-induced arthritis. Arthritis Rheum 56:3271–83.

373. Wurbel MA, Philippe JM, Nguyen C, Victorero G, Freeman T, Wooding P, Miazek A, Mattei MG, Malissen M, Jordan BR, Malissen B, Carrier A, Naquet P. 2000. The chemokine TECK is expressed by thymic and intestinal epithelial cells and attracts double- and single-positive thymocytes expressing the TECK receptor CCR9. Eur J Immunol 30:262–71.

374. Johansson-Lindbom B, Agace WW. 2007. Generation of gut-homing T cells and their localization to the small intestinal mucosa. Immunol Rev 215:226–42.

375. Williams SA, Harata-Lee Y, Comerford I, Anderson RL, Smyth MJ, McColl SR. 2010. Multiple functions of CXCL12 in a syngeneic model of breast cancer. Mol Cancer 9:250.

376. Su L, Zhang J, Xu H, Wang Y, Chu Y, Liu R, Xiong S. 2005. Differential expression of CXCR4 is associated with the metastatic potential of human non-small cell lung cancer cells. Clin Cancer Res 11:8273–80.

377. Speetjens FM, Liefers GJ, Korbee CJ, Mesker WE, van de Velde CJH, van Vlierberghe RL, Morreau H, Tollenaar RA, Kuppen PJK. 2009. Nuclear localization of CXCR4 determines prognosis for colorectal cancer patients. Cancer Microenviron 2:1–7.

150

378. Müller A, Homey B, Soto H, Ge N, Catron D, Buchanan ME, McClanahan T, Murphy E, Yuan W, Wagner SN, Barrera JL, Mohar A, Verástegui E, Zlotnik A. 2001. Involvement of chemokine receptors in breast cancer metastasis. Nature 410:50–6.

379. Wu S, Gessner R, Taube T, Korte A, von Stackelberg A, Kirchner R, Henze G, Seeger K. 2006. Chemokine IL-8 and chemokine receptor CXCR3 and CXCR4 gene expression in childhood acute lymphoblastic leukemia at first relapse. J Pediatr Hematol Oncol 28:216–20.

380. Li YM, Pan Y, Wei Y, Cheng X, Zhou BP, Tan M, Zhou X, Xia W, Hortobagyi GN, Yu D, Hung M-C. 2004. Upregulation of CXCR4 is essential for HER2-mediated tumor metastasis. Cancer Cell 6:459–69.

381. Amersi FF, Terando AM, Goto Y, Scolyer RA, Thompson JF, Tran AN, Faries MB, Morton DL, Hoon DSB. 2008. Activation of CCR9/CCL25 in cutaneous melanoma mediates preferential metastasis to the small intestine. Clin Cancer Res 14:638–45.

382. Scala S, Giuliano P, Ascierto PA, Ieranò C, Franco R, Napolitano M, Ottaiano A, Lombardi ML, Luongo M, Simeone E, Castiglia D, Mauro F, De Michele I, Calemma R, Botti G, Caracò C, Nicoletti G, Satriano RA, Castello G. 2006. Human melanoma metastases express functional CXCR4. Clin Cancer Res 12:2427– 33.

383. Yang G, Rosen DG, Zhang Z, Bast RC, Mills GB, Colacino JA, Mercado-Uribe I, Liu J. 2006. The chemokine growth-regulated oncogene 1 (Gro-1) links RAS signaling to the senescence of stromal fibroblasts and ovarian tumorigenesis. Proc Natl Acad Sci U S A 103:16472–7.

384. Sparmann A, Bar-Sagi D. 2004. Ras-induced interleukin-8 expression plays a critical role in tumor growth and angiogenesis. Cancer Cell 6:447–58.

385. Pivarcsi A, Müller A, Hippe A, Rieker J, van Lierop A, Steinhoff M, Seeliger S, Kubitza R, Pippirs U, Meller S, Gerber PA, Liersch R, Buenemann E, Sonkoly E, Wiesner U, Hoffmann TK, Schneider L, Piekorz R, Enderlein E, Reifenberger J, Rohr U-P, Haas R, Boukamp P, Haase I, Nürnberg B, Ruzicka T, Zlotnik A, Homey B. 2007. Tumor immune escape by the loss of homeostatic chemokine expression. Proc Natl Acad Sci U S A 104:19055–60.

386. Shellenberger TD, Wang M, Gujrati M, Jayakumar A, Strieter RM, Burdick MD, Ioannides CG, Efferson CL, El-Naggar AK, Roberts D, Clayman GL, Frederick MJ. 2004. BRAK/CXCL14 is a potent inhibitor of angiogenesis and a chemotactic factor for immature dendritic cells. Cancer Res 64:8262–70.

387. Schwarze SR, Luo J, Isaacs WB, Jarrard DF. 2005. Modulation of CXCL14 (BRAK) expression in prostate cancer. Prostate 64:67–74.

388. Wente MN, Mayer C, Gaida MM, Michalski CW, Giese T, Bergmann F, Giese NA, Büchler MW, Friess H. 2008. CXCL14 expression and potential function in pancreatic cancer. Cancer Lett 259:209–17.

151

389. Tessema M, Klinge DM, Yingling CM, Do K, Van Neste L, Belinsky SA. 2010. Re- expression of CXCL14, a common target for epigenetic silencing in lung cancer, induces tumor necrosis. Oncogene 29:5159–70.

390. Ozawa S, Kato Y, Komori R, Maehata Y, Kubota E, Hata R-I. 2006. BRAK/CXCL14 expression suppresses tumor growth in vivo in human oral carcinoma cells. Biochem Biophys Res Commun 348:406–12.

391. Gu X-L, Ou Z-L, Lin F-J, Yang X-L, Luo J-M, Shen Z-Z, Shao Z-M. 2012. Expression of CXCL14 and its anticancer role in breast cancer. Breast Cancer Res Treat 135:725–35.

392. Frenz DA, Liu W, Cvekl A, Xie Q, Wassef L, Quadro L, Niederreither K, Maconochie M, Shanske A. 2010. Retinoid signaling in inner ear development: A “Goldilocks” phenomenon. Am J Med Genet Part A 152A:2947–2961.

393. Wozniak GG, Strahl BD. 2014. Catalysis-dependent stabilization of Bre1 fine-tunes histone H2B ubiquitylation to regulate gene transcription. Genes Dev 28:1647–52.

394. Wolinsky JB, Colson YL, Grinstaff MW. 2012. Local drug delivery strategies for cancer treatment: Gels, nanoparticles, polymeric films, rods, and wafers. J Control Release 159:14–26.

395. Richards KF, Mukherjee S, Bienkowska-Haba M, Pang J, Sapp M. 2014. Human papillomavirus species-specific interaction with the basement membrane-resident non-heparan sulfate receptor. Viruses 6:4856–79.

396. Zahm JM, Kaplan H, Hérard AL, Doriot F, Pierrot D, Somelette P, Puchelle E. 1997. Cell migration and proliferation during the in vitro wound repair of the respiratory epithelium. Cell Motil Cytoskeleton 37:33–43.

397. Iizuka M. 2011. Wound healing of intestinal epithelial cells. World J Gastroenterol 17:2161.

398. Wendt MK, Johanesen PA, Kang-Decker N, Binion DG, Shah V, Dwinell MB. 2006. Silencing of epithelial CXCL12 expression by DNA hypermethylation promotes colonic carcinoma metastasis. Oncogene 25:4986–97.

399. Wendt MK, Cooper AN, Dwinell MB. 2008. Epigenetic silencing of CXCL12 increases the metastatic potential of mammary carcinoma cells. Oncogene 27:1461– 71.

400. Cotton M, Claing A. 2009. G protein-coupled receptors stimulation and the control of cell migration. Cell Signal 21:1045–53.

401. Tanegashima K, Suzuki K, Nakayama Y, Tsuji K, Shigenaga A, Otaka A, Hara T. 2013. CXCL14 is a natural inhibitor of the CXCL12-CXCR4 signaling axis. FEBS Lett 587:1731–5.

402. Hara T, Tanegashima K. 2014. CXCL14 antagonizes the CXCL12-CXCR4 signaling axis. Biomol Concepts 5:167–73.

152

403. Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T, Naeem R, Carey VJ, Richardson AL, Weinberg RA. 2005. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121:335–48.

404. Zheng H, Fu G, Dai T, Huang H. 2007. Migration of endothelial progenitor cells mediated by stromal cell-derived factor-1alpha/CXCR4 via PI3K/Akt/eNOS signal transduction pathway. J Cardiovasc Pharmacol 50:274–80.

405. Kryczek I, Wei S, Keller E, Liu R, Zou W. 2007. Stroma-derived factor (SDF- 1/CXCL12) and human tumor pathogenesis. Am J Physiol Cell Physiol 292:C987–95.

406. Otte M, Kliewer A, Schütz D, Reimann C, Schulz S, Stumm R. 2014. CXCL14 is no direct modulator of CXCR4. FEBS Lett 588:4769–75.

407. Ozawa S, Kato Y, Kubota E, Hata R-I. 2009. BRAK/CXCL14 expression in oral carcinoma cells completely suppresses tumor cell xenografts in SCID mouse. Biomed Res 30:315–8.

408. Zeng J, Yang X, Cheng L, Liu R, Lei Y, Dong D, Li F, Lau QC, Deng L, Nice EC, Xie K, Huang C. 2013. Chemokine CXCL14 is associated with prognosis in patients with colorectal carcinoma after curative resection. J Transl Med 11:6.

409. Kerdiles Y, Ugolini S, Vivier E. 2013. T cell regulation of natural killer cells. J Exp Med 210:1065–1068.

410. Crouse J, Xu HC, Lang PA, Oxenius A. 2015. NK cells regulating T cell responses: mechanisms and outcome. Trends Immunol 36:49–58.

411. Nausch N, Cerwenka A. 2008. NKG2D ligands in tumor immunity. Oncogene 27:5944–5958.

412. Zhang Z, Su T, He L, Wang H, Ji G, Liu X, Zhang Y, Dong G. 2012. Identification and functional analysis of ligands for natural killer cell activating receptors in colon carcinoma. Tohoku J Exp Med 226:59–68.

413. Holdenrieder S, Stieber P, Peterfi A, Nagel D, Steinle A, Salih HR. 2006. Soluble MICA in malignant diseases. Int J cancer 118:684–7.

414. Kumar V, Yi Lo PH, Sawai H, Kato N, Takahashi A, Deng Z, Urabe Y, Mbarek H, Tokunaga K, Tanaka Y, Sugiyama M, Mizokami M, Muroyama R, Tateishi R, Omata M, Koike K, Tanikawa C, Kamatani N, Kubo M, Nakamura Y, Matsuda K. 2012. Soluble MICA and a MICA variation as possible prognostic biomarkers for HBV-induced hepatocellular carcinoma. PLoS One 7:e44743.

415. Tahara-Hanaoka S, Shibuya K, Onoda Y, Zhang H, Yamazaki S, Miyamoto A, Honda S-I, Lanier LL, Shibuya A. 2004. Functional characterization of DNAM-1 (CD226) interaction with its ligands PVR (CD155) and nectin-2 (PRR-2/CD112). Int Immunol 16:533–8.

153

416. Gerosa F, Baldani-Guerra B, Nisii C, Marchesini V, Carra G, Trinchieri G. 2002. Reciprocal activating interaction between natural killer cells and dendritic cells. J Exp Med 195:327–33.

417. Morandi B, Bougras G, Muller WA, Ferlazzo G, Münz C. 2006. NK cells of human secondary lymphoid tissues enhance T cell polarization via IFN-gamma secretion. Eur J Immunol 36:2394–400.

418. Laouar Y, Sutterwala FS, Gorelik L, Flavell RA. 2005. Transforming growth factor- beta controls T helper type 1 cell development through regulation of natural killer cell interferon-gamma. Nat Immunol 6:600–7.

419. Mocikat R, Braumüller H, Gumy A, Egeter O, Ziegler H, Reusch U, Bubeck A, Louis J, Mailhammer R, Riethmüller G, Koszinowski U, Röcken M. 2003. Natural killer cells activated by MHC class I(low) targets prime dendritic cells to induce protective CD8 T cell responses. Immunity 19:561–9.

420. del Toro-Arreola S, Arreygue-Garcia N, Aguilar-Lemarroy A, Cid-Arregui A, Jimenez-Perez M, Haramati J, Barros-Nuñez P, Gonzalez-Ramella O, del Toro- Arreola A, Ortiz-Lazareno P, Hernandez-Flores G, Bravo-Cuellar A, Daneri- Navarro A, Jave-Suarez LF. 2011. MHC class I-related chain A and B ligands are differentially expressed in human cervical cancer cell lines. Cancer Cell Int 11:15.

421. Jimenez-Perez MI, Jave-Suarez LF, Ortiz-Lazareno PC, Bravo-Cuellar A, Gonzalez-Ramella O, Aguilar-Lemarroy A, Hernandez-Flores G, Pereira-Suarez AL, Daneri-Navarro A, del Toro-Arreola S. 2012. Cervical cancer cell lines expressing NKG2D-ligands are able to down-modulate the NKG2D receptor on NKL cells with functional implications. BMC Immunol 13:7.

422. Garcia-Chagollan M, Jave-Suarez LF, Haramati J, Sanchez-Hernandez PE, Aguilar-Lemarroy A, Bueno-Topete MR, Pereira-Suarez AL, Fafutis-Morris M, Cid-Arregui A, del Toro-Arreola S. 2013. Substantial increase in the frequency of circulating CD4+NKG2D+ T cells in patients with cervical intraepithelial neoplasia grade 1. J Biomed Sci 20:60.

423. Song EY, Shurin MR, Tourkova IL, Gutkin DW, Shurin G V. 2010. Epigenetic mechanisms of promigratory chemokine CXCL14 regulation in human prostate cancer cells. Cancer Res 70:4394–401.

424. Niller HH, Szenthe K, Minarovits J. 2014. Epsteine Barr virus host cell interactions: an epigenetic dialog? Front Genet 5.

425. Woellmer A, Arteaga-Salas JM, Hammerschmidt W. 2012. BZLF1 governs CpG- methylated chromatin of Epstein-Barr Virus reversing epigenetic repression. PLoS Pathog 8:e1002902.

426. Minarovits J, Demcsák A, Banati F, Niller HH. 2016. Epigenetic Dysregulation in Virus-Associated Neoplasms. Adv Exp Med Biol 879:71–90.

154

427. Amador-Molina A, Hernández-Valencia J, Lamoyi E, Contreras-Paredes A, Lizano M. 2013. Role of Innate Immunity against Human Papillomavirus (HPV) Infections and Effect of Adjuvants in Promoting Specific Immune Response. Viruses 5:2624–2642.

428. Renoux VM, Bisig B, Langers I, Dortu E, Clémenceau B, Thiry M, Deroanne C, Colige A, Boniver J, Delvenne P, Jacobs N. 2011. Human papillomavirus entry into NK cells requires CD16 expression and triggers cytotoxic activity and cytokine secretion. Eur J Immunol 41:3240–52.

429. Nakagawa M, Kim KH, Moscicki A-B. 2005. Patterns of CD8 T-Cell Epitopes within the Human Papillomavirus Type 16 (HPV 16) E6 Protein among Young Women Whose HPV 16 Infection Has Become Undetectable. Clin Vaccine Immunol 12:1003– 1005.

430. Frazer IH, De Kluyver R, Leggatt GR, Guo HY, Dunn L, White O, Harris C, Liem A, Lambert P. 2001. Tolerance or immunity to a tumor antigen expressed in somatic cells can be determined by systemic proinflammatory signals at the time of first antigen exposure. J Immunol 167:6180–7.

431. Tindle RW. 2002. Immune evasion in human papillomavirus-associated cervical cancer. Nat Rev Cancer 2:59–65.

432. Bergot A-S, Ford N, Leggatt GR, Wells JW, Frazer IH, Grimbaldeston MA. 2014. HPV16-E7 Expression in Squamous Epithelium Creates a Local Immune Suppressive Environment via CCL2- and CCL5- Mediated Recruitment of Mast Cells. PLoS Pathog 10:e1004466.

433. de Vries VC, Pino-Lagos K, Elgueta R, Noelle RJ. 2009. The enigmatic role of mast cells in dominant tolerance. Curr Opin Organ Transplant 14:332–337.

434. Vance RE, Kraft JR, Altman JD, Jensen PE, Raulet DH. 1998. Mouse CD94/NKG2A is a natural killer cell receptor for the nonclassical major histocompatibility complex (MHC) class I molecule Qa-1(b). J Exp Med 188:1841–8.

435. Marvel D, Gabrilovich DI. 2015. Myeloid-derived suppressor cells in the tumor microenvironment: expect the unexpected. J Clin Invest 125:3356–64.

436. Youn J-I, Nagaraj S, Collazo M, Gabrilovich DI. 2008. Subsets of myeloid-derived suppressor cells in tumor-bearing mice. J Immunol 181:5791–5802.

437. Talmadge JE, Gabrilovich DI. 2013. History of myeloid-derived suppressor cells. Nat Rev Cancer 13:739–752.

438. Gabrilovich DI, Nagaraj S. 2009. Myeloid-derived suppressor cells as regulators of the immune system. Nat Rev Immunol 9:162–174.

439. Kimura H, Nakamura T, Ogawa T, Tanaka S, Shiota K. 2003. Transcription of mouse DNA methyltransferase 1 (Dnmt1) is regulated by both E2F-Rb-HDAC- dependent and -independent pathways. Nucleic Acids Res 31:3101–3113.

155

440. Münger K, Basile JR, Duensing S, Eichten A, Gonzalez SL, Grace M, Zacny VL. 2001. Biological activities and molecular targets of the human papillomavirus E7 oncoprotein. Oncogene 20:7888–98.

441. Lee JO, Russo AA, Pavletich NP. 1998. Structure of the retinoblastoma tumour- suppressor pocket domain bound to a peptide from HPV E7. Nature 391:859–65.

442. Todorovic B, Hung K, Massimi P, Avvakumov N, Dick FA, Shaw GS, Banks L, Mymryk JS. 2012. Conserved region 3 of human papillomavirus 16 E7 contributes to deregulation of the retinoblastoma tumor suppressor. J Virol 86:13313–23.

443. Massimi P, Pim D, Banks L. 1997. Human papillomavirus type 16 E7 binds to the conserved carboxy-terminal region of the TATA box binding protein and this contributes to E7 transforming activity. J Gen Virol 78 ( Pt 10:2607–13.

444. Mannhardt B, Weinzimer SA, Wagner M, Fiedler M, Cohen P, Jansen-Dürr P, Zwerschke W. 2000. Human papillomavirus type 16 E7 oncoprotein binds and inactivates growth-inhibitory insulin-like growth factor binding protein 3. Mol Cell Biol 20:6483–95.

445. Shin M-K, Balsitis S, Brake T, Lambert PF. 2009. Human papillomavirus E7 oncoprotein overrides the tumor suppressor activity of p21Cip1 in cervical carcinogenesis. Cancer Res 69:5656–63.

446. Pyeon D, Pearce SM, Lank SM, Ahlquist P, Lambert PF. 2009. Establishment of human papillomavirus infection requires cell cycle progression. PLoS Pathog 5:e1000318.

156

APPENDIX A

Primers Used for PCR and RT-qPCR

Gene Orientation Primer Sequence 5' → 3' CDKN2A Forward GTGGACCTGGCTGAGGAG CDKN2A Reverse TCTTTCAATCGGGGATGTCT CXCL1 Forward CTTCCTCCTCCCTTCTGGTC CXCL1 Reverse GAAAGCTTGCCTCAATCCTG CXCL10 Forward AGGAACCTCCAGTCTCAGCA CXCL10 Reverse CAAAATTGGCTTGCAGGAAT CXCL11 Forward AGTGTGAAGGGCATGGCTAT CXCL11 Reverse GCCTTGCTTGCTTCGATTTG CXCL14 Forward GGAAATGAAGCCAAAGTACCC CXCL14 Reverse AGGCGTTGTACCACTTGATGA Cxcl14 Forward ATTAGAATTCATGAGGCTCCTGGCGGCCGC Cxcl14 Reverse ATCGGGATCCCTATTCTTCGTAGACCCTGC CXCL2 Forward GCTTCCTCCTTCCTTCTGGT CXCL2 Reverse GGGCAGAAAGCTTGTCTCAA DNMT1 Forward TTCTGTTAAGCTGTCTCTTTCCA DNMT1 Reverse TGCTGAAGCCTCCGAGAT E7 (HPV 11) Forward TAAGCATCTAGAATGCATGGAAGACTTGTTACC E7 (HPV 11) Reverse TGCTTAGGATCCTTATGGTTTTGGTGCGCA E7 (HPV 16) Forward TAAGCATCTAGAATGCATGGAGATACACCTACA E7 (HPV 16) Reverse TGCTTAGGATCCTTATGGTTTCTGAGAACAGATG E7 (HPV 18) Forward TAAGCATCTAGAATGCATGGACCTAAGGCA E7 (HPV 18) Reverse TGCTTAGGATCCTTACTGCTGGGATGCACA E7 (HPV 6) Forward TAAGCATCTAGAATGCATGGAAGACATGTTACC E7 (HPV 6) Reverse TGCTTAGGATCCTTAGGTCTTCGGTGCGC Gapdh Forward AATGACCCCTTCATTGACCTC Gapdh Reverse ATGGGATTTCCATTGATGACA HLA-E Forward TAGCTCCCTCCTTTTCCACC HLA-E Reverse GGATCTGTGGTCTCTGGAGC HPV16 E1^E4 Forward AAATGACAGCTCAGAGGAGGAG HPV16 E1^E4 Reverse GAGTCACACTTGCAACAAAAGG IL-8 Forward AAGAAACCACCGGAAGGAAC IL-8 Reverse AGCACTCCTTGGCAAAACTG MCM5 Forward GACATCCAGGTCATGCTCAA MCM5 Reverse AGGGATCTTCACCAGGTGTG MCM7 Forward GGTCAGTTCTCCACTCACGG MCM7 Reverse CATACATTGATCGACTGGCG UHRF1 Forward GCCTGCAGAGGCTGTTCTAC UHRF1 Reverse AGGAGCTGGATGGTGTCATT β-actin Forward TCACCCACACTGTGCCCATCTA β-actin Reverse TGAGGTAGTCAGTCAGGTCCCG

157

APPENDIX B

Primers Used for Methylation Specific PCR (MSP)

Annealing Target Orientation Specificity Temperature Sequence (5' → 3') CXCL14 Forward Methylated TTAATGAGTTCGTTCGTTGCGAG 61.9°C CXCL14 Reverse Methylated ACCAAAAACCTCATACTAACC CXCL14 Forward Control GAGTTTGTTTGTTGTGAGGGTA 61.9°C CXCL14 Reverse Control ACCAAAAACCTCATACTAACC Cxcl14 Forward Methylated GAGATCGTAGTATTTAGCGTTAAGC 62°C Cxcl14 Reverse Methylated AAAAAACAAAAAAACAAAAAAAACG Cxcl14 Forward Unmethylated GGAGATTGTAGTATTTAGTGTTAAGTGT 62°C Cxcl14 Reverse Unmethylated AAAACAAAAAAACAAAAAAAACACC HLA-E Forward Methylated CGTTTTTTCGGGGCGGGAGTCGCG 65°C HLA-E Reverse Methylated CTAAAAAAAACGAAACCCTCGCTATCGACG HLA-E Forward Unmethylated TGTGGTGTTTTTTTGGGGTGGGAGTTGTG 65°C HLA-E Reverse Unmethylated CCAACTAAAAAAAACAAAACCCTCACTATCAACA HOXD10 Forward Methylated TCGCGGAATTCGATTTATTTTTTCGT 62.9°C HOXD10 Reverse Methylated GCTTCGACCCCTAAACCCAACCGAC HOXD10 Forward Unmethylated GGGTTTGTGGAATTTGATTTATTTTTTTGT 62.9°C HOXD10 Reverse Unmethylated CCACTTCAACCCCTAAACCCAACCA

158

APPENDIX C

Primers Used for Molecular Cloning

Target Gene Orientation Primer Sequence (5' → 3') Restriction Site(s) CXCL14 Forward ATTTGAATTCATGTCCCTGCTCCCACGCCG EcoRI CXCL14 Reverse CGATGGATCCCTATTCTTCGTAGACCCTGC BamHI Cxcl14 Forward ATTAGAATTCATGAGGCTCCTGGCGGCCGC EcoRI Cxcl14 Reverse ATCGGGATCCCTATTCTTCGTAGACCCTGC BamHI E7 (HPV6) Forward TAAGCATCTAGAATGCATGGAAGACATGTTACC XbaI E7 (HPV6) Reverse TGCTTAGGATCCTTAGGTCTTCGGTGCGC BamHI E7 (HPV8) Forward TAAGCATCTAGAATGCATGGAAGACTTGTTACC XbaI E7 (HPV8) Reverse TGCTTAGGATCCTTATGGTTTTGGTGCGCA BamHI E7 (HPV16) Forward TAAGCATCTAGAATGCATGGAGATACACCTACA XbaI E7 (HPV16) Reverse TGCTTAGGATCCTTATGGTTTCTGAGAACAGATG BamHI E7 (HPV18) Forward TAAGCATCTAGAATGCATGGACCTAAGGCA XbaI E7 (HPV18) Reverse TGCTTAGGATCCTTACTGCTGGGATGCACA BamHI Target Promoter Orientation Primer Sequence (5' → 3') Restriction Sites CXCL14 Forward GGGGGGATCCCAGCCTCTCCACCGTCTGG BamHI CXCL14 Reverse GGGGCCATGGACCGGAGGGGCGCGGCG NcoI HLA-E CGI Forward GGGGGGATCCGGGCTTCCGGCTTGGCCGCG BamHI GGGGTTCGAAGCGATCGGATGCTTCGAGAATCTCTTGA HLA-E CGI Reverse BstBI; PvuI ACCCGGGAGG HOXD10 Forward GGGGAGATCTCTCCCAGAGGTCCGGCCCCG BglII HOXD10 Reverse GGGGAAGCTTTGTAATTGTGGATGCCTCTGCCGACCAC HindIII

159

APPENDIX D

Supplemental Figures for Chapter III3

A IL-8 CXCL9 CXCL11 CCL3 CCL19 15 15 10 16 * * 15 * ** * * * 8 * * 12 * 10 10 10 6 8 4

5 5 4 5 2

0 0 mRNA expression level (log2) level mRNAexpression l l l r l a /2 3 2 3 a /2 3 a /2 3 1 N er al / 1 N 1/2 IN3 ce 1 N m I c 1 IN m I cer rma IN C n I cer r IN C rm IN C IN C o C a C o C an o ancer or C an N C CIN an N C N C C N C Norm C B CXCL1 CXCL2 CXCL5 CXCL6 CCL20 16 15 16 15 15 * * * * * 12 12 10 10 10 8 8 5 4 5 4 5 0 0 mRNA expression level (log2) level mRNAexpression l 2 2 l 2 3 l 2 l 3 a / er al / / / /2 N m 1 IN3 c m 1 IN3 cer ma 1 IN cer ma IN3 cer I r IN C n r N C r N C r C N1 C o a I o o IN1 orma CI ancer N C C No C Can N CI Can N C Can N C

CDCXCL13 CCL8 CXCR2 CXCR4 15 12 * * 15 * 8 * * * * 10

10 * 6 8 10 6 4 5 4 2 5 mRNA expression level (log2) level mRNAexpression l (log2) level mRNAexpression l l al /2 3 r a /2 3 r a /2 3 r a /2 3 r m 1 IN ce m 1 IN ce m 1 IN ce m 1 IN ce r C n r N C n r N C n r N C n o a o I a o I a o I a N CIN C N C C N C C N C C

Appendix D-1. Chemokine Expression is Deregulated in HPV-Associated Cancer Progression. Chemokines and chemokine receptors with significant changes of expression in CxCa progression are shown: (A) IL-8, CXCL9, CXCL11, CCL3, CCL19; (B) CXCL1, CXCL2, CXCL5, CXCL6, CCL20; (C) CXCL13, CCL8; and (D) CXCR2, CXCR4. The gene expression data were analyzed from a global gene expression study of 128 cervical tissue samples in different disease stages: normal (n = 24); low-grade lesion (n = 36); high-grade lesion (n = 40); and cancer (n = 28) (111). Normalized fluorescence intensities (log2) of gene expression from each group are shown in box-and-whisker plots with Tukey's method for outliers (black triangles) noted as distinct data points. P-values were calculated between each transition (normal to CIN1/β, CIN1/β to CINγ, and CINγ to cancer) by the Student’s t- test. *p<0.05 (A-D). These data were generated by Dohun Pyeon PhD, and Tao Xu PhD, University of Colorado.

3 Figures within Appendix D are published with permission from Cicchini et al. 2016. Suppression of Antitumor Immune Responses by Human Papillomavirus through Epigenetic Downregulation of CXCL14. mBio, http://mbio.asm.org/content/7/3/e00270-16.

160

ABCCXCL9 CXCL10 CXCL13 14 16 15 p=0.02 p=0.036 14 p=0.0003 12 12 10 10 10

8 8 6 5 mRNA expression level (log2) level mRNAexpression HPV- HNC HPV+ HNC (log2) level mRNAexpression HPV- HNC HPV+ HNC (log2) level mRNAexpression HPV- HNC HPV+ HNC

DEFCCL19 CXCR4 IL-8 12 14 p=0.01 14 p=0.03 p<0.008

10 12 12

10 10 8

8 8 6 6 6 mRNA expression level (log2) level expression mRNA HPV- HNC HPV+ HNC (log2) level expression mRNA HPV- HNC HPV+ HNC (log2) level expression mRNA HPV- HNC HPV+ HNC

GHIHPV16 early gene IL-8 CXCL1 2.0 20 25 ***

-actin 1.5 *** 20 β 15 -actin (fold) -actin -actin (fold) -actin β β *** 15 1.0 10 10 ** 0.5 5 5 * mRNA relative to to relative mRNA 0.0 0 0 mRNA relative to to relative mRNA mRNA relative to to relative mRNA S E G Y K 2 2 X S E G Y S E G Y I 1 1 K 2 2 X K 2 X N W I 1 12 1 P W N NI W W G 12GP W1 W 2 W 1 W12GP W JKLCXCL2 CXCL10 CXCL11 20 *** 8 2.5 *** *** 2.0 15 6 -actin (fold) -actin -actin (fold) -actin -actin (fold) -actin β β β 1.5 * 10 *** 4 * ** 1.0 5 2 ** 0.5

0 mRNA relative to to relative mRNA 0.0 0 to relative mRNA mRNA relative to to relative mRNA S E G Y S E Y E G Y 2 2 X G 2 X IK 1 1 K 2 2 X 1 P N P 1 1 P NIKS W12 W G W W G NI W W G 2 2 2 1 1 1 W W W Appendix D-2. Proinflammatory Chemokines are Upregulated in HPV-Positive HNCs and Keratinocytes. (A-F) Gene expression levels of chemokines and chemokine receptors in head and neck were analyzed with HPV-positive (n = 16) and HPV-negative (n = 26) HNCs from our previous global gene expression study (233), as described in Fig. 1 and Supplementary Figure S1. Normalized fluorescence intensities (log2) of gene expression from each group are shown in box-and-whisker plots with Tukey's method for outliers (black circle) noted as distinct data points. P-values shown on each panel were calculated between HPV-negative and HPV-positive HNCs by the Student’s t-test. (G-L) Total RNA was extracted from NIKS, W12E, W12G, and W12GPXY keratinocyte lines. (G) HPV16 early gene transcript E1^E4 was measured by RT-qPCR, as previously described (446) . (H-L) mRNA expression of IL-8, CXCL1, CXCL2, CXCL10, and CXCL11 were measured by RT- qPCR using specific primers (Appendix A), and normalized by β-actin mRNA. Data are shown as fold changes (± SD) to the mRNA level in NIKS cells. P-values were determined by the Student’s t-test. *p<0.05, **p<0.001, ***p<0.0001. These data were generated by Dohun Pyeon, PhD, University of Colorado and Tao Xu, PhD, St Jude Hospital, TN.

161

mRNA expression DNA methylation p < 0.0001 p < 0.0001 6 p < 0.0001 n.s. 1.0×10 1.0 p = 0.0001 p = 0.03 100000 0.8 10000 1000 0.6

100 0.4

10 values Beta 0.2 1

RSEM normalized counts normalized RSEM 0.1 0.0 HPV- HPV+ CxCa HPV- HPV+ CxCa HNC HNC HNC HNC

HPV+ HNC HPV- HNC 100000 1000000 R2 = 0.3853 R2 = 0.0672 100000 10000 10000 1000 1000 100 100 mRNA expression (counts) mRNAexpression mRNA expression (counts) mRNAexpression 10 10 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8 DNA methylation (b-values) DNA methylation (b-values)

CxCa 100000 R2 = 0.1389 10000

1000

100

mRNA expression (counts) mRNAexpression 1 0.0 0.2 0.4 0.6 0.8 1.0 DNA methylation (b-values)

Appendix D-3. CXCL14 Downregulation Correlates with Increased CXCL14 Promoter Methylation in HPV-Positive HNC and CxCa. The TCGA data sets of CXCL14 RNA-seq RSEM counts (mRNA expression) and beta-values (DNA methylation) were retrieved from cBioPortal (cbioportal.org): HPV- HNC, n = 243; HPV+ HNC, n = 36 (129); CxCa, n = 309 (NCI, TCGA, Provisional). (A & B) Shown are box-and-whisker plots with Tukey's method for outliers (black triangles) noted as distinct data points. P-values were determined by the Student’s t-test. n.s., not significant. Correlations between CXCL14 mRNA expression and DNA methylation were analyzed within HPV+ HNC (C), HPV- HNC (D), and CxCa (E). The correlation coefficient (R2) was determined by linear regression using Prism software. These data were generated by Dohun Pyeon, PhD, University of Colorado.

162 aVector bCxcl14-clone 8 c Cxcl14-clone16 3000 3000 3000 ) ) 3 ) 3 3

2000 2000 2000

1000 1000 1000 Tumor volume (mm volume Tumor Tumor volume (mm volume Tumor Tumor volume (mm volume Tumor 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Days Days Days dRag-vector eRag-clone 8 f Rag-clone 16 3000 3000 3000 ) ) ) 3 3 3

2000 2000 2000

1000 1000 1000 Tumor volume (mm volume Tumor (mm volume Tumor (mm volume Tumor 0 0 0 0 20 40 60 80 0 20 40 60 80 0 20 40 60 80 Days Days Days Appendix D-4. Restoration of Cxcl14 Expression Suppresses Tumor Growth in vivo. MOE/E6E7 cell clones re-expressing Cxcl14 (clones 8 and 16) and a vector containing MOE/E6E7 cell clone were injected into the rear right flank of wild type C57BL/6 (A-C) and Rag1-/- (D-F) mice (n = 10 for wild type, n = 7 for Rag1-/- each group). Tumor growth was determined every week by the formula: volume = (width)2 × depth. Tumor growth curves of each mouse are shown. These data were collected in collaboration with Dan Vermeer, Sanford Health.

163

Appendix D-5. Gating Strategy for Flow Cytometry. Whole spleen from a C57BL/6 mouse was homogenized and stained with a panel of antibodies conjugated to unique fluorophores. Single stain and no stain controls were used for fluorescence compensation. A generous large cell gate (forward scatter vs. side scatter area), single cell gate (side scatter area vs. side scatter width), and CD45+ gate (side scatter area vs. CD45) were applied as parental gates before determining antigen presenting cell (side scatter area vs. MHCII), neutrophil (side scatter high, Gr1high), monocyte (side scatter low, Gr1mid), and macrophage (MHCII+, F4/80+) populations. A small cell lymphocyte gate (side scatter area vs. forward scatter), and single cell (side scatter area vs. side scatter width) were applied as parental gates to determine NK cell (CD45+, NKp46+), CD4+ T cell (CD45+, CD4+), and CD8+ T cell (CD45+, CD8+) populations. A representative example of the overall gating strategy is shown and was applied to TDLN and spleen harvested from C57BL/6 mice injected with MOE/E6E7 cells with Cxcl14 or vector.

164

A MHC II B Neutrophil 80 0.3

60 0.2

40 0.1

20 0.0 Cell percentage Cell Cell percentage Cell

0 -0.1 Vector Clone 8 Clone 16 Vector Clone 8 Clone 16

Cxcl14 Cxcl14 C D Monocyte Macrophage 0.5 2.5 p = 0.001 p = 0.004 0.4 2.0

0.3 1.5

0.2 1.0 Cell percentage Cell 0.1 percentage Cell 0.5

0.0 0.0 Vector Clone 8 Clone 16 Vector Clone 8 Clone 16

Cxcl14 Cxcl14

Appendix D-6. Immune Cell Populations in TDLN Altered by Cxcl14 Expression. MOE/E6E7 cells with Cxcl14 (clones 8 and 16) or the vector were injected into the right flank of C57BL/6 mice (n = 10, each group). TDLNs were harvested at 21 days post injection. Percentages of antigen presenting cell (A), neutrophil (B), monocyte (C), and macrophage (D) populations were determined by flow cytometry using specific antibodies as described in Materials and Methods. P-values were determined between vector alone and clone 8 or clone 16 by the Student’s t-test. These data were collected in collaboration with Joe Westrich, University of Colorado.

165

NK CD4 25 30

20 20 15 p = 0.005 10 p = 0.08 10

Cell percentage Cell 5 percentage Cell

0 0 Vector Clone 8 Clone 16 Vector Clone 8 Clone 16

Cxcl14 Cxcl14

CD8 MHC II 30 80

60 20

10 p = 0.004 20 Cell percentage Cell percentage Cell

0 0 Vector Clone 8 Clone 16 Vector Clone 8 Clone 16

Cxcl14 Cxcl14

Neutrophil Monocyte 8 2.0

6 1.5 4 1.0 2 0.5 Cell percentage Cell Cell percentage Cell 0

-2 0.0 Vector Clone 8 Clone 16 Vector Clone 8 Clone 16

Cxcl14 Cxcl14

Macrophage 2.5

2.0

1.5

1.0

Cell percentage Cell 0.5

0.0 Vector Clone 8 Clone 16

Cxcl14

Appendix D-7. Immune Cell Populations in Spleen Altered by Cxcl14 Expression.

166

Appendix D-7. Immune Cell Populations in Spleen Altered by Cxcl14 Expression. MOE/E6E7 cells with Cxcl14 (clones 8 and 16) or the vector were injected into the right flank of C57BL/6 mice (n = 10, each group). Spleens were harvested at 21 days post injection. Percentages of NK cell (A), CD4+ T cell (B), CD8+ T cell (C), antigen presenting cell (D), neutrophil (E), monocyte (F), and macrophage (G) populations were determined by flow cytometry using specific antibodies as described in Materials and Methods. P-values were determined between vector alone and clone 8 or clone 16 by the Student’s t-test. These data were collected in collaboration with Joe Westrich, University of Colorado.

167

Cervix Head and neck Gene Symbol Expression level (Log2) Fold Expression level (Log2) Fold Normal Early Late Cancer Cancer/normal HPV+HNC HPV-HNC HPV+/- IL8 5.277 7.316 9.255 10.482 36.886 9.521 10.543 -2.031 CXCL9 2.971 4.327 5.705 6.653 12.835 10.602 9.605 1.995 CXCL1 6.316 8.324 10.212 9.568 9.527 9.079 9.771 -1.616 CXCR4 7.941 8.732 9.082 10.729 6.907 10.839 9.422 2.671 CXCL10 6.407 7.482 8.19 9.154 6.713 11.427 9.699 3.315 CXCL11 3.365 4.144 4.214 5.975 6.105 7.326 6.426 1.867 CXCL2 3.216 4.169 5.215 5.673 5.491 7.375 7.157 1.163 CCL19 3.62 3.962 4.728 5.99 5.169 8.313 6.941 2.587 CCL3 3.175 3.502 4.234 5.425 4.757 8.386 8.190 1.146 CXCL13 2.31 2.411 3.353 4.361 4.144 9.586 8.328 2.392 CCL20 3.532 4.707 5.864 5.544 4.033 10.414 9.317 2.140 CCL8 3.011 3.498 3.5 4.978 3.910 7.194 7.245 -1.036 CXCL6 5.867 7.261 8.418 7.671 3.492 6.554 6.717 -1.119 CCR1 3.592 4.234 4.18 5.197 3.042 8.124 7.836 1.221 CCR1 4.588 5.004 5.525 5.986 2.635 8.046 7.734 1.241 CCL11 2.983 3.049 3.246 3.865 1.843 6.963 7.055 -1.066 CXCL16 7.82 8.214 8.803 8.667 1.799 9.653 9.710 -1.040 CXCL3 3.956 4.267 4.934 4.8 1.795 6.776 6.388 1.308 CCL18 3.413 3.391 3.6 4.202 1.728 9.000 8.333 1.587 CCR7 4.276 4.264 4.541 4.885 1.525 8.207 7.579 1.545 CCL4 6.184 6.469 6.48 6.772 1.503 9.167 8.856 1.240 CCR2 2.057 2.26 2.431 2.484 1.344 7.292 6.876 1.334 CCL22 2.504 2.62 3.206 2.724 1.165 8.778 8.698 1.057 CCR10 2.964 2.941 2.919 3.181 1.162 6.059 6.118 -1.042 CCL13 2.82 2.82 2.628 3.013 1.143 6.994 7.143 -1.109 CCR6 3.252 3.384 3.245 3.388 1.099 5.384 5.101 1.217 CCL21 2.828 2.515 2.761 2.961 1.097 6.853 6.867 -1.010 CCL28 6.634 6.705 6.943 6.741 1.077 7.873 7.876 -1.002 CCR5 5.722 5.95 5.932 5.798 1.054 10.640 10.195 1.361 CCR9 1.423 1.422 1.416 1.411 -1.008 6.599 6.600 -1.001 CCR4 2.2 2.214 2.197 2.187 -1.009 7.466 7.499 -1.023 CCR2 1.773 1.773 1.765 1.756 -1.012 6.284 6.451 -1.123 CCL23 3.68 3.678 3.59 3.661 -1.013 4.146 4.185 -1.028 CCL17 3.169 3.201 3.281 3.145 -1.017 6.019 6.024 -1.004 CCR3 1.786 1.79 1.77 1.76 -1.018 6.534 6.672 -1.100 CCR8 2.332 2.326 2.311 2.303 -1.020 7.761 7.846 -1.060 CCL25 2.414 2.415 2.397 2.384 -1.021 7.088 7.329 -1.182 CCL24 2.367 2.365 2.349 2.333 -1.024 6.073 6.145 -1.051 CXCR5 2.171 2.167 2.151 2.135 -1.025 6.562 6.565 -1.002 CCL7 3.562 3.603 3.551 3.524 -1.027 5.387 5.523 -1.099 CCL27 3.6 3.609 3.582 3.557 -1.030 6.446 6.396 1.035 CXCR1 (IL8RA) 2.285 2.268 2.261 2.241 -1.031 8.648 8.745 -1.069 CCL1 2.607 2.608 2.588 2.558 -1.035 6.016 6.195 -1.133 CCL16 2.179 2.173 2.093 2.091 -1.063 7.074 7.237 -1.120 CXCR6 4.513 4.674 4.737 4.4 -1.081 7.704 6.984 1.647 CCL14/CCL15 3.9 3.359 3.251 3.784 -1.084 7.643 7.769 -1.091 CCL2 8.624 9.347 8.719 8.458 -1.122 8.311 7.745 1.480 CXCR3 4.883 4.927 4.892 4.631 -1.191 8.115 7.946 1.125 CX3CL1 6.316 6.198 7.436 5.911 -1.324 8.693 8.304 1.309 CCL5 7.534 8.052 8.067 7.125 -1.328 9.037 8.180 1.810 CXCL5 4.254 5.357 5.732 3.658 -1.512 5.474 6.021 -1.462 CCL26 4.713 4.142 4.165 4.018 -1.619 7.047 7.344 -1.229 CXCL12 7.871 6.837 6.372 7.091 -1.717 7.854 7.259 1.511 CXCR7 10.698 11.061 11.132 9.902 -1.736 10.199 10.441 -1.183 CXCL17 11.107 11.597 11.911 9.98 -2.184 9.011 9.549 -1.452 CX3CR1 6.686 6.405 5.776 4.903 -3.441 7.243 6.901 1.267 CXCR2 (IL8RB) 9.508 9.192 7.523 5.86 -12.536 6.435 6.417 1.012 CXCL14 12.001 11.062 9.811 7.594 -21.215 9.271 10.949 -3.200 Appendix D-8. Chemokine Expression Profiles of Tissue Specimens from Cervix and Head/Neck.

168

APPENDIX E

Supplemental Figures for Chapter IV4

Appendix E-1. Summary of biological pathways related to HPV16 E7-dependent gene expression changes.

4 Figure and tables in this appendix are currently in submission to Scientific Reports, July 2016.

169

Appendix E-1. Summary of biological pathways related to HPV16 E7-dependent gene expression changes. Pathways analysis was performed using Reactome (version 56, reactome.org) with genes specifically upregulated (A, Appendix E-5A) or downregulated (B, Appendix E-5B) by HPV16 E7 described in Fig 1B. The global pathway diagrams shown are automatically generated by the Reactome database. Yellow nodes and lines indicate pathways and their connections significantly altered by HPV16 E7 expression. Pathway clusters showing significant E7-dependent changes are indicated with red text and specific biological pathways significantly changed (FDR-adjusted p < 0.01) are listed in Appendix E- 6.

170

Appendix E-2. Gene expression involved in matrix metalloproteinases, IL1 signaling, and antigen presentation is dysregulated by high-risk E7. Normalized gene expression levels in the pathways of (A) matrix metalloproteinases (MMP1, MMP9, MMP10, MMP28); (B) IL1 signaling (IL1B, IL1RN, IL1R1, IL36G) and (C) antigen presentation (Sec31A, ITGAV, CTSL2, RANSEL) are shown. (D) Expression of IL1B, IL1R1, IL1RN and IL36G was validated by RT-qPCR with total RNA extracted from NIKS, NIKS-16, NIKS-18, and NIKS- 16ΔE7 cells and normalized by β-actin expression. P values were calculated by Student’s t test and indicated on the graphs (A-C) or by asterisks as follows (D): *p < 0.0001, **p < 0.0005, ***p < 0.005, ****p < 0.05.

171

HLA-A HLA-B HLA-C 250000 ns 500000 250000 ns ns 200000 400000 200000 ns ns 150000 300000 150000 p < 0.05

100000 200000 100000

50000 100000 50000 RSEM normalized counts normalized RSEM RSEM normalized counts normalized RSEM 0 0 counts normalized RSEM 0 HPV- HPV+ CxCa HPV- HPV+ CxCa HPV- HPV+ CxCa HNC HNC HNC HNC HNC HNC

HLA-E HLA-F HLA-G 40000 5000 70000 p < 0.0001 p < 0.0001 p = 0.002 60000 4000 ns p = 0.02 30000 50000 3000 40000 20000 ns 30000 2000 20000 10000 1000 10000 RSEM normalized counts normalized RSEM counts normalized RSEM RSEM normalized counts normalized RSEM 0 0 0 HPV- HPV+ CxCa HPV- HPV+ CxCa HPV- HPV+ CxCa HNC HNC HNC HNC HNC HNC

Appendix E-3. HLA-E expression is significantly downregulated in HPV-positive cancers compared to HPV-negative cancers. The RNA-seq RSEM (RNA-seq by expectation maximization) counts of HLA-A, -B, -C, -E, -F, and –G were obtained from the TCGA data through cBioPortal (cbioportal.org): HPV-negative HNC, n = 243; HPV-positive HNC, n = 36 (129); CxCa, n = 309 (NCI, TCGA, Provisional). Normalized RSEM counts are shown in scatter plots. P values were determined by Student’s t test. Data were collected by Dohun Pyeon, PhD, University of Colorado.

172

Appendix E-4. Regulatory elements and non-coding RNAs in the HLA-E CGI. Predicted regulatory elements were identified near the HLA-E CGI in chromosome 6 (Chr:6) using the UCSC Genome Browser (version hg19, genome.ucsc.edu). The HLA-E CGI is shown (green bar). The locations of ncRNAs identified by RNA sequencing, DNase I hypersensitivity, and transcription factor binding sites identified by ChIP-seq (EZH2, ELF1, UBTF, CCNT2, GABPA, E2F6, MAX, HMGN3 and ELF1) are labeled and indicated as black bars. The locations of histone H3 lysine 27 acetylation (H3K27Ac) are shown with different colors indicating different cell types (GM12878, H1-hESC, HSMM, HUVEC, K562, NHEK, NHLF) in which acetylation was observed.

173

p<0.05 1.3-fold p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity (log2) (log2) Probe ID Gene ID Probe ID Gene ID NIKS1 NIKS1 NIKS NIKS16 NIKS16Δ NIKS NIKS16 NIKS16Δ 8 E7 8 E7 231372_at LOC153328 5.41 12.07 5.06 5.23 219383_at PRR5L 4.82 6.07 5.84 4.85 214414_x_at HBA1 /// HBA2 4.92 8.38 5.34 5.33 204018_x_at HBA1 /// HBA2 6.27 7.51 6.44 6.59 202237_at NNMT 5.85 9.05 10.21 6.85 225223_at SMAD5 8.06 9.29 8.19 7.79 205668_at LY75 4.88 7.65 5.95 5.01 236918_s_at LRRC34 4.49 5.70 3.72 3.27 207039_at CDKN2A 7.24 9.85 9.95 7.35 200884_at CKB 8.36 9.56 9.62 9.06 227276_at PLXDC2 5.73 8.20 8.49 5.60 208850_s_at THY1 5.98 7.18 6.43 6.13 236297_at --- 3.08 5.52 5.98 3.03 214022_s_at IFITM1 7.66 8.85 9.06 6.07 206204_at GRB14 4.25 6.64 6.53 4.17 206631_at PTGER2 5.48 6.66 6.18 5.71 202238_s_at NNMT 5.77 8.14 9.16 6.60 202295_s_at CTSH 8.36 9.54 7.81 6.76 226132_s_at MANEAL 5.31 7.67 5.86 6.10 219298_at ECHDC3 6.22 7.39 5.74 5.75 209644_x_at CDKN2A 8.97 11.31 11.28 9.08 202149_at NEDD9 6.07 7.23 8.07 6.10 238455_at --- 4.08 6.41 6.58 4.04 220807_at HBQ1 6.19 7.34 6.41 6.93 218966_at MYO5C 5.46 7.74 8.40 6.45 241925_x_at --- 3.34 4.48 4.56 3.04 211745_x_at HBA1 /// HBA2 6.08 8.33 6.44 6.55 204783_at MLF1 6.53 7.66 6.77 6.32 205885_s_at ITGA4 3.97 6.15 6.05 2.86 229332_at HPDL 5.89 7.00 6.25 6.05 205884_at ITGA4 4.08 6.23 6.22 3.12 242283_at DNAH14 5.30 6.41 5.77 5.06 1554640_at PALM2 3.95 6.04 5.68 4.01 219315_s_at TMEM204 6.76 7.87 7.42 6.70 213416_at ITGA4 5.99 8.07 8.17 4.62 206354_at SLCO1B3 4.92 6.02 4.79 4.25 204114_at NID2 6.63 8.68 6.59 6.60 218711_s_at SDPR 7.41 8.50 8.70 7.79 217414_x_at HBA1 /// HBA2 6.20 8.23 6.32 6.52 221942_s_at GUCY1A3 4.91 6.01 6.90 5.50 1569191_at ZNF826 3.11 5.11 2.57 2.45 221648_s_at --- 6.97 8.05 7.53 6.51 210448_s_at P2RX5 6.06 8.04 6.81 6.22 226925_at ACPL2 6.32 7.40 8.31 4.92 243366_s_at --- 5.24 7.21 7.34 3.85 223824_at RNLS 4.69 5.75 6.10 4.58 1554542_at LOC153328 5.68 7.56 5.71 5.71 202948_at IL1R1 5.16 6.22 5.68 5.54 220051_at PRSS21 6.71 8.52 8.31 6.28 235428_at --- 4.55 5.60 5.45 4.88 206698_at XK 4.92 6.72 6.18 5.82 229139_at JPH1 8.13 9.18 8.67 7.71 219884_at LHX6 6.87 8.65 7.97 7.46 221916_at NEFL 6.77 7.82 6.53 7.05 237450_at LOC389332 5.90 7.64 5.75 5.88 204510_at CDC7 7.23 8.27 8.21 7.43 211596_s_at LRIG1 5.97 7.69 7.75 6.17 209167_at GPM6B 5.24 6.28 6.36 4.02 213107_at TNIK 4.93 6.62 6.80 5.67 241503_at FAM81A 4.84 5.88 5.38 5.06 1552289_a_at CILP2 5.56 7.23 5.78 6.53 208767_s_at LAPTM4B 11.05 12.09 11.65 11.69 205187_at SMAD5 5.94 7.58 6.01 5.67 242100_at CHSY3 3.47 4.50 3.38 3.53 235451_at SMAD5 8.58 10.14 8.63 8.31 219745_at TMEM180 6.95 7.99 7.41 7.28 243110_x_at NPW 4.73 6.30 4.62 5.11 239761_at GCNT1 6.50 7.53 7.25 6.10 209733_at MID2 5.01 6.55 6.08 5.14 205505_at GCNT1 6.35 7.38 7.29 6.15 209008_x_at KRT8 9.33 10.84 11.12 10.21 202035_s_at SFRP1 6.09 7.10 6.75 6.11 209458_x_at HBA1 /// HBA2 6.04 7.53 6.20 6.37 1554679_a_at LAPTM4B 10.29 11.29 10.87 10.79 1552288_at CILP2 4.85 6.33 4.92 5.69 202995_s_at FBLN1 6.23 7.23 6.73 6.40 230003_at --- 4.05 5.52 5.62 3.71 203799_at CD302 3.25 4.25 3.31 3.05 208510_s_at PPARG 4.84 6.31 6.92 4.92 1552946_at ZNF114 5.82 6.82 5.92 5.54 213912_at TBC1D30 3.77 5.23 4.27 4.66 206103_at RAC3 7.98 8.95 8.82 8.03 232331_at --- 5.33 6.78 7.17 5.58 226748_at LYSMD2 8.16 9.12 8.84 8.26 211699_x_at HBA1 /// HBA2 6.10 7.54 6.27 6.31 204784_s_at MLF1 8.27 9.24 8.60 8.45 202037_s_at SFRP1 9.44 10.87 10.74 9.86 201387_s_at UCHL1 7.15 8.11 7.24 7.09 225516_at SLC7A2 6.68 8.11 7.61 6.99 206299_at FAM155B 7.23 8.20 7.68 7.50 1570153_at C13orf38 5.11 6.54 5.83 4.60 218755_at KIF20A 9.03 9.98 9.94 8.96 227641_at FBXL16 7.09 8.51 7.99 7.61 220060_s_at C12orf48 5.86 6.81 6.59 5.70 1564706_s_at GLS2 4.61 6.03 4.79 4.52 205743_at STAC 7.04 7.99 6.91 6.81 207057_at SLC16A7 4.36 5.78 5.75 3.86 228977_at LOC729680 7.54 8.49 7.88 7.58 209170_s_at GPM6B 4.46 5.87 5.78 3.34 218500_at C8orf55 7.33 8.28 7.79 7.20 211828_s_at TNIK 4.21 5.62 5.72 4.39 215091_s_at GTF3A 10.93 11.88 11.51 11.06 225681_at CTHRC1 6.15 7.54 7.79 6.36 222904_s_at TMC5 4.93 5.86 5.66 4.66 202036_s_at SFRP1 7.23 8.61 8.38 7.49 1565951_s_at CHML 4.73 5.65 5.58 4.64 228885_at MAMDC2 4.37 5.74 4.30 3.59 201338_x_at GTF3A 11.06 11.98 11.64 11.25 225219_at SMAD5 9.05 10.41 9.18 8.88 219615_s_at KCNK5 5.73 6.64 6.33 5.74 222717_at SDPR 6.60 7.96 8.23 7.21 226889_at WDR35 7.25 8.15 7.78 7.28 204042_at WASF3 6.42 7.77 7.01 6.78 221586_s_at E2F5 7.59 8.47 8.00 7.74 226865_at --- 5.36 6.70 7.71 5.32 227165_at SKA3 7.71 8.60 8.34 7.52 205188_s_at SMAD5 7.33 8.66 7.32 7.23 242519_at CCDC152 4.14 5.02 5.14 4.31 205110_s_at FGF13 4.15 5.48 3.96 4.00 221911_at ETV1 3.36 4.25 4.39 3.52 219569_s_at TMEM22 7.19 8.52 6.19 7.50 204717_s_at SLC29A2 5.58 6.45 6.51 5.45 229797_at MCOLN3 3.50 4.79 6.42 3.44 205531_s_at GLS2 5.66 6.54 5.64 5.67 1555486_a_at PRR5L 6.22 7.49 7.61 6.88 203543_s_at KLF9 5.62 6.49 6.71 5.75 Appendix E-5A. E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

174

p<0.05 1.3-fold p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity Probe ID Gene ID NIKS1 Probe ID Gene ID NIKS1 NIKS NIKS16 NIKS16Δ NIKS NIKS16 NIKS16Δ 8 E7 8 E7 223843_at SCARA3 4.81 5.68 4.94 4.42 210215_at TFR2 5.60 6.32 5.58 5.74 203453_at SCNN1A 10.08 10.95 11.38 10.53 204023_at RFC4 9.19 9.91 9.94 8.59 201601_x_at IFITM1 7.89 8.75 9.06 7.02 206752_s_at DFFB 6.40 7.12 6.94 6.36 204127_at RFC3 9.43 10.30 10.06 9.16 202637_s_at ICAM1 7.33 8.05 8.41 7.34 201860_s_at PLAT 7.95 8.81 9.02 7.45 219733_s_at SLC27A5 7.23 7.95 7.81 7.32 212510_at GPD1L 8.43 9.29 9.03 8.49 221436_s_at CDCA3 8.34 9.06 8.86 8.18 202638_s_at ICAM1 6.05 6.91 7.30 5.94 232242_at --- 3.25 3.97 3.79 3.49 229641_at CCBE1 7.20 8.05 8.36 7.25 223605_at SLC25A18 4.53 5.24 5.56 4.63 204128_s_at RFC3 8.20 9.05 8.84 7.86 223381_at NUF2 8.47 9.19 9.10 8.03 219785_s_at FBXO31 /// LOC100 6.18 7.04 6.57 6.04 226482_s_at TSTD1 8.61 9.32 9.13 8.92 202994_s_at FBLN1 6.83 7.68 7.32 6.95 203386_at TBC1D4 7.90 8.61 8.42 7.80 240304_s_at TMC5 4.78 5.63 5.29 5.01 1553810_a_at KIAA1524 5.24 5.94 5.78 4.86 210692_s_at SLC43A3 9.15 9.98 10.29 9.17 225291_at PNPT1 9.46 10.16 9.85 9.46 204521_at C12orf24 8.53 9.36 9.18 8.71 218911_at YEATS4 7.09 7.79 7.59 7.03 242888_at --- 7.20 8.03 7.68 6.79 212771_at FAM171A1 7.54 8.24 8.05 6.31 219494_at RAD54B 7.88 8.70 8.37 7.96 235749_at UGGT2 4.94 5.64 5.18 4.33 205229_s_at COCH 3.49 4.31 3.74 3.66 223307_at CDCA3 8.98 9.68 9.47 8.66 227444_at ARMCX4 2.63 3.44 2.61 2.69 1552619_a_at ANLN 9.38 10.08 9.92 8.79 219152_at PODXL2 6.28 7.10 6.55 6.60 202147_s_at IFRD1 7.73 8.43 8.49 7.69 217979_at TSPAN13 9.15 9.97 8.93 9.05 1556429_a_at WDR67 5.73 6.43 6.01 5.70 204146_at RAD51AP1 8.33 9.15 8.98 8.14 1569342_at GLI3 4.82 5.51 4.98 4.98 234023_s_at CENPJ 4.34 5.16 4.82 3.82 231855_at KIAA1524 6.64 7.33 7.01 5.99 207828_s_at CENPF 8.96 9.77 9.88 8.81 223062_s_at PSAT1 10.01 10.70 10.45 9.80 211519_s_at KIF2C 7.94 8.75 8.59 8.07 234307_s_at KIF26A 5.24 5.92 5.49 5.53 228876_at BAIAP2L2 6.83 7.64 7.22 7.09 225786_at NCRNA00201 5.41 6.09 6.02 5.60 222714_s_at LACTB2 8.77 9.57 8.90 8.88 204782_at --- 5.19 5.88 5.42 5.17 239503_at --- 4.91 5.72 5.14 4.61 213109_at TNIK 3.81 4.49 4.66 4.03 229404_at TWIST2 5.48 6.28 5.28 5.56 217640_x_at SKA1 6.39 7.07 6.96 5.84 229538_s_at IQGAP3 7.36 8.15 8.19 7.16 223545_at FANCD2 7.60 8.28 8.15 7.34 209891_at SPC25 7.09 7.88 7.67 6.65 207891_s_at HAUS7 /// TREX2 6.29 6.97 6.66 6.10 204588_s_at SLC7A7 6.00 6.79 5.92 5.78 226287_at CCDC34 8.23 8.90 8.70 8.02 205485_at RYR1 5.63 6.41 5.61 5.59 212599_at AUTS2 8.11 8.78 8.46 7.93 242069_at CBX5 6.04 6.83 6.54 6.09 202983_at HLTF 7.96 8.63 8.48 7.57 202855_s_at SLC16A3 8.44 9.22 9.29 8.04 227196_at RHPN2 7.22 7.89 7.63 6.99 215765_at LRRC41 4.74 5.52 5.17 4.81 220892_s_at PSAT1 8.12 8.79 8.48 7.94 234944_s_at FAM54A 6.99 7.77 7.45 7.00 218701_at LACTB2 7.93 8.61 7.95 8.15 218691_s_at PDLIM4 6.69 7.47 6.86 6.71 202384_s_at TCOF1 8.31 8.98 8.92 8.24 202856_s_at SLC16A3 9.22 10.00 10.06 8.90 244427_at KIF23 4.01 4.68 4.26 3.74 242324_x_at CCBE1 6.42 7.20 7.26 6.58 219735_s_at TFCP2L1 7.92 8.59 8.29 8.09 202146_at IFRD1 7.26 8.03 8.11 7.01 226661_at CDCA2 8.62 9.28 9.23 8.29 222037_at MCM4 9.02 9.79 9.67 8.99 201843_s_at EFEMP1 8.69 9.35 9.17 8.12 204794_at DUSP2 7.09 7.86 7.46 6.90 222857_s_at KCNMB4 6.38 7.04 6.82 6.45 230033_at C19orf51 7.59 8.36 7.48 7.85 235684_s_at SESN3 4.93 5.59 4.86 5.04 203387_s_at TBC1D4 8.09 8.85 8.72 7.97 208159_x_at DDX11 8.09 8.75 8.61 8.08 219922_s_at LTBP3 7.01 7.77 7.39 7.33 224932_at CHCHD10 9.04 9.70 9.61 8.95 216331_at ITGA7 6.18 6.94 6.49 6.08 211089_s_at NEK3 5.69 6.36 6.24 5.74 204215_at C7orf23 8.17 8.93 8.69 7.88 227212_s_at PHF19 8.61 9.27 9.39 8.19 211300_s_at TP53 7.62 8.38 8.29 7.92 222685_at HAUS6 7.77 8.44 8.34 6.94 226912_at ZDHHC23 5.89 6.64 6.27 6.09 1564467_at FAM161A 5.21 5.87 5.34 5.14 219312_s_at ZBTB10 3.03 3.79 3.22 2.87 230876_at ZNF883 4.77 5.42 3.93 4.53 223275_at PRMT6 7.70 8.46 8.16 7.46 244324_at C18orf54 6.97 7.62 7.51 6.76 230250_at PTPRB 3.69 4.44 3.77 3.50 224959_at SLC26A2 9.68 10.33 10.23 9.65 225081_s_at CDCA7L 8.29 9.03 8.96 7.92 207199_at TERT 6.15 6.80 6.08 6.32 216870_x_at DLEU2 6.46 7.20 6.65 6.17 235178_x_at ESCO2 3.48 4.13 3.67 3.47 224548_at HES7 6.62 7.35 7.12 6.72 243805_at CCBE1 5.09 5.73 5.89 5.01 215425_at BTG3 4.49 5.22 5.00 4.57 218168_s_at CABC1 7.91 8.55 8.50 7.83 205733_at BLM 6.90 7.63 7.46 6.90 1556190_s_at PRNP 3.97 4.62 4.78 4.09 1553654_at SYT14 3.25 3.99 4.00 3.57 222608_s_at ANLN 10.19 10.83 10.67 9.54 1553120_at CLSPN 4.99 5.72 5.41 4.68 220840_s_at C1orf112 7.09 7.73 7.64 6.94 205395_s_at MRE11A 8.03 8.77 8.77 8.16 219544_at C13orf34 8.30 8.94 8.92 8.03 219288_at C3orf14 9.32 10.05 9.82 9.00 238133_at --- 4.29 4.93 5.01 4.43 202107_s_at MCM2 9.49 10.21 10.29 9.11 219537_x_at DLL3 6.51 7.15 6.58 6.71 Appendix E-5A. (continued) E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

175

p<0.05 1.3-fold p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity Probe ID Gene ID NIKS1 Probe ID Gene ID NIKS1 NIKS NIKS16 NIKS16Δ NIKS NIKS16 NIKS16Δ 8 E7 8 E7 229442_at C18orf54 7.58 8.21 8.05 7.02 204240_s_at SMC2 8.88 9.45 9.32 8.50 212619_at TMEM194A 5.45 6.08 5.64 5.03 222632_s_at LZTFL1 6.86 7.43 7.17 6.59 218005_at ZNF22 8.41 9.04 8.84 7.74 244786_at SNHG10 4.72 5.29 5.23 4.72 219985_at HS3ST3A1 7.38 8.01 6.22 7.01 203210_s_at RFC5 7.14 7.71 7.75 7.05 223296_at SLC25A33 7.90 8.53 8.26 7.90 223417_at RAD18 6.74 7.31 7.15 6.53 227607_at STAMBPL1 7.56 8.19 7.94 7.34 205899_at CCNA1 8.72 9.28 7.90 6.55 221958_s_at WLS 8.45 9.08 9.19 8.42 211814_s_at CCNE2 7.28 7.85 7.92 6.91 218252_at CKAP2 9.35 9.98 9.84 9.12 220120_s_at EPB41L4A 3.80 4.36 4.23 3.64 1553299_at DUSP5P 4.51 5.13 4.50 4.40 220993_s_at GPR63 4.76 5.33 5.04 4.61 219433_at BCOR 5.54 6.16 5.75 5.62 205961_s_at PSIP1 7.39 7.95 8.07 6.63 232899_at FAM41C /// RPL23A 6.93 7.55 7.30 7.07 216568_x_at --- 5.72 6.28 6.52 5.89 228752_at EFCAB4B 4.85 5.47 4.81 4.90 224715_at WDR34 8.68 9.24 9.33 8.18 241733_at C18orf54 4.97 5.59 5.31 4.43 223305_at TMEM216 7.41 7.98 7.89 7.38 227211_at PHF19 7.84 8.46 8.72 7.56 201614_s_at RUVBL1 9.41 9.97 9.82 9.20 239069_s_at --- 7.58 8.19 7.88 7.69 205883_at ZBTB16 4.38 4.95 5.78 4.55 233110_s_at BCL2L12 8.06 8.67 8.48 7.64 235588_at ESCO2 6.45 7.01 6.74 6.15 223513_at CENPJ 7.33 7.95 7.84 6.93 243840_at LOC100289612 6.83 7.39 7.33 6.64 213427_at RPP40 9.12 9.73 9.19 9.22 205429_s_at MPP6 6.77 7.33 7.09 6.25 210567_s_at SKP2 7.18 7.80 8.05 6.77 226614_s_at FAM167A 7.43 7.99 7.84 7.38 227818_at CCDC21 7.72 8.33 7.85 7.83 205677_s_at DLEU1 8.59 9.15 8.94 8.70 209324_s_at RGS16 6.05 6.66 6.59 6.23 233823_at FAM184B 4.75 5.31 4.89 4.87 221652_s_at C12orf11 10.17 10.78 10.53 10.37 223759_s_at GSG2 6.52 7.08 7.06 6.49 223245_at STRBP 7.28 7.89 7.75 6.92 203588_s_at TFDP2 7.44 8.00 8.22 7.31 208691_at TFRC 11.37 11.98 11.81 11.30 210486_at ANKMY1 7.09 7.65 7.25 7.11 219217_at NARS2 8.73 9.34 8.92 8.68 223570_at MCM10 7.36 7.91 7.75 7.38 218602_s_at HAUS6 5.83 6.44 6.15 5.14 242310_at --- 3.95 4.51 4.19 4.04 204492_at ARHGAP11A 5.57 6.17 5.92 5.04 1053_at RFC2 8.13 8.68 8.59 7.78 226299_at PKN3 6.33 6.93 6.67 6.13 241866_at SLC16A7 3.12 3.68 3.91 3.10 203818_s_at SF3A3 8.71 9.32 9.25 8.69 1555313_a_at MCF2 3.90 4.45 3.65 3.01 205603_s_at DIAPH2 6.36 6.96 6.56 6.48 210517_s_at AKAP12 5.96 6.51 7.26 4.09 239253_at --- 4.54 5.14 5.21 4.48 219622_at RAB20 7.67 8.22 8.17 7.43 235253_at RAD1 7.00 7.60 7.66 6.90 230973_at SH2D5 7.93 8.49 7.81 8.03 235014_at LOC147727 6.42 7.02 6.50 6.47 204012_s_at LCMT2 7.85 8.41 7.99 7.80 204093_at CCNH 9.28 9.88 9.78 9.31 224963_at SLC26A2 7.75 8.30 8.12 7.61 229269_x_at SSBP4 7.96 8.56 8.42 8.13 223211_at HACL1 8.15 8.69 8.51 7.90 205412_at ACAT1 10.36 10.95 10.42 10.50 202633_at TOPBP1 8.27 8.82 8.81 7.68 235020_at TAF4B 8.04 8.64 8.45 7.71 1554878_a_at ABCD3 8.48 9.03 8.67 8.14 219118_at FKBP11 9.60 10.20 10.11 9.12 219262_at SUV39H2 4.14 4.68 4.46 4.21 235205_at LOC346887 7.05 7.65 7.77 7.13 204767_s_at FEN1 9.72 10.26 10.32 9.50 222930_s_at AGMAT 6.00 6.60 6.31 5.71 219510_at POLQ 7.52 8.06 7.83 6.95 214240_at GAL 10.17 10.76 10.41 10.17 224200_s_at RAD18 6.38 6.93 6.70 5.70 218404_at SNX10 7.13 7.72 7.75 7.11 227680_at ZNF326 9.02 9.56 9.33 8.93 213761_at MDM1 5.53 6.12 5.70 5.33 214727_at BRCA2 6.46 7.00 7.04 5.83 219588_s_at NCAPG2 8.43 9.02 9.08 7.74 1553715_s_at FAM195A 8.52 9.07 8.76 8.63 210869_s_at MCAM 8.01 8.59 8.90 6.71 238615_at ERLIN2 4.47 5.01 4.71 4.35 221520_s_at CDCA8 8.79 9.37 9.27 8.75 224734_at HMGB1 8.43 8.97 8.64 7.77 240365_at LOC647946 4.89 5.48 5.66 4.88 201842_s_at EFEMP1 10.83 11.38 11.23 10.31 209754_s_at TMPO 6.32 6.90 6.74 5.30 227370_at FAM171B 3.65 4.20 5.49 3.53 215739_s_at TUBGCP3 5.96 6.54 6.63 5.78 219650_at ERCC6L 7.77 8.31 8.16 7.59 226895_at NFIC 9.38 9.96 9.98 9.11 204768_s_at FEN1 8.92 9.47 9.49 8.63 235273_at DYX1C1 5.22 5.80 5.27 4.82 227376_at GLI3 7.89 8.43 8.36 7.77 223130_s_at MYLIP 7.10 7.68 7.31 7.15 208149_x_at DDX11 8.08 8.62 8.48 7.94 224523_s_at C3orf26 10.02 10.60 10.43 9.71 202666_s_at ACTL6A 9.19 9.73 9.48 8.61 208795_s_at MCM7 9.34 9.91 9.83 8.77 221551_x_at ST6GALNAC4 7.32 7.86 7.23 6.98 201664_at SMC4 10.47 11.04 11.04 9.93 231270_at CA13 8.10 8.64 8.51 8.19 203696_s_at RFC2 8.26 8.84 8.75 7.89 213253_at SMC2 6.91 7.45 7.26 6.36 212022_s_at MKI67 8.76 9.33 9.44 8.45 203925_at GCLM 9.85 10.39 10.04 9.31 225861_at FAM195A 8.43 9.01 8.73 8.53 209087_x_at MCAM 7.63 8.16 8.66 6.81 210983_s_at MCM7 9.89 10.47 10.40 9.25 209285_s_at C3orf63 7.71 8.25 8.19 7.67 219257_s_at SPHK1 8.40 8.97 8.57 8.33 204244_s_at DBF4 8.94 9.47 9.46 8.13 225943_at NLN 7.96 8.53 8.26 7.90 223219_s_at CNOT10 7.61 8.15 8.07 7.35 222697_s_at ABHD10 8.33 8.90 8.63 8.10 1552711_a_at CYB5D1 8.33 8.87 8.66 8.38 Appendix E-5A. (continued) E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

176

p<0.05 1.3-fold p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity Probe ID Gene ID NIKS1 Probe ID Gene ID NIKS1 NIKS NIKS16 NIKS16Δ NIKS NIKS16 NIKS16Δ 8 E7 8 E7 228566_at RPRD1A 7.10 7.63 7.42 6.72 238590_x_at TMEM107 8.91 9.40 9.08 8.90 209849_s_at RAD51C 8.35 8.88 8.91 8.23 226496_at ZCCHC7 7.69 8.18 8.04 7.34 233252_s_at STRBP 7.43 7.96 7.94 7.18 218984_at PUS7 10.02 10.50 10.16 9.87 224495_at TMEM107 5.01 5.54 4.99 4.73 227017_at ERICH1 8.30 8.78 8.54 7.93 209507_at RPA3 9.54 10.07 10.26 9.16 215136_s_at EXOSC8 9.05 9.53 9.60 8.81 1555910_at PTCD2 8.29 8.82 8.85 8.14 220937_s_at ST6GALNAC4 7.87 8.35 7.69 7.62 210573_s_at POLR3C 7.75 8.28 8.07 7.74 243010_at MSI2 6.35 6.83 6.83 6.39 218772_x_at TMEM38B 7.71 8.24 8.01 7.29 212061_at SR140 8.10 8.59 8.75 7.51 211376_s_at NSMCE4A 9.21 9.74 9.79 9.29 212719_at PHLPP1 7.22 7.70 7.66 7.04 238055_at ATP8B1 4.79 5.31 4.89 4.65 1552787_at HELB 4.02 4.50 4.14 4.03 213150_at HOXA10 6.77 7.29 7.13 6.48 1569086_at EML2 5.85 6.33 5.97 5.77 204170_s_at CKS2 11.65 12.18 12.15 11.19 222701_s_at CHCHD7 10.05 10.53 10.25 10.09 232816_s_at DDX11 6.25 6.77 6.56 6.14 228671_at TMEM201 6.81 7.29 6.97 6.79 59705_at SCLY 6.85 7.37 7.17 6.79 1553533_at JPH1 5.44 5.92 5.68 5.29 215220_s_at TPR 4.49 5.01 5.01 4.40 220083_x_at UCHL5 7.54 8.02 8.01 7.63 208886_at H1F0 8.60 9.13 9.12 8.66 243529_at MARS2 7.38 7.86 7.50 7.35 225869_s_at UNC93B1 7.42 7.94 7.65 7.46 218997_at POLR1E 8.84 9.31 9.11 8.53 217010_s_at CDC25C 5.38 5.90 5.98 5.24 211707_s_at IQCB1 7.29 7.77 7.49 6.79 209337_at PSIP1 7.86 8.38 8.54 6.83 202502_at ACADM 9.17 9.64 9.41 9.24 225096_at C17orf79 9.62 10.15 9.76 9.55 204252_at CDK2 8.78 9.25 9.44 8.41 220027_s_at RASIP1 8.27 8.79 8.94 8.22 202531_at IRF1 7.93 8.40 8.63 7.84 222781_s_at C9orf40 7.97 8.49 8.23 7.63 213097_s_at DNAJC2 9.75 10.23 10.17 9.45 222735_at TMEM38B 7.04 7.56 7.26 6.73 206600_s_at SLC16A5 7.05 7.52 7.30 6.98 213116_at NEK3 7.05 7.56 7.47 7.05 200951_s_at CCND2 8.88 9.35 9.51 8.86 1554452_a_at C7orf68 8.35 8.87 8.50 8.37 213186_at DZIP3 6.00 6.47 6.52 5.72 212203_x_at IFITM3 11.20 11.72 12.02 10.78 212021_s_at MKI67 8.75 9.22 9.29 8.51 223165_s_at IP6K2 7.94 8.45 8.29 7.88 1552947_x_at ZNF114 5.44 5.91 5.63 5.32 218006_s_at ZNF22 7.08 7.59 7.29 6.07 203196_at ABCC4 7.48 7.94 7.72 7.45 213338_at TMEM158 7.28 7.79 7.22 7.35 202780_at OXCT1 8.49 8.96 9.14 8.45 203711_s_at HIBCH 7.52 8.04 7.79 7.62 231319_x_at KIF9 8.45 8.91 8.79 8.30 1557014_a_at C9orf122 5.71 6.22 5.68 5.15 203836_s_at MAP3K5 7.02 7.48 7.51 6.91 218437_s_at LZTFL1 6.73 7.25 6.95 6.46 220238_s_at KLHL7 6.63 7.09 6.73 6.01 1556344_at LOC150051 5.58 6.09 5.43 5.71 202963_at RFX5 8.62 9.09 9.54 8.61 227793_at MIRLET7D 6.85 7.36 7.22 6.55 223758_s_at GTF2H2 7.74 8.21 8.03 7.49 204281_at TEAD4 8.44 8.95 8.89 8.45 208896_at DDX18 9.49 9.95 9.70 9.38 232596_at DIAPH3 7.07 7.58 7.52 6.65 213869_x_at THY1 6.19 6.65 6.29 6.16 219117_s_at FKBP11 10.04 10.55 10.48 9.56 235039_x_at LIN9 4.63 5.09 5.27 4.60 204825_at MELK 10.35 10.86 10.81 9.88 204416_x_at APOC1 6.55 7.01 6.53 6.22 228401_at ATAD2 8.10 8.60 8.58 7.55 212694_s_at PCCB 9.15 9.61 9.26 8.97 206316_s_at KNTC1 6.95 7.46 7.36 6.93 203976_s_at CHAF1A 7.78 8.24 8.29 7.72 206348_s_at PDK3 5.83 6.33 5.29 5.89 205930_at GTF2E1 8.80 9.25 9.05 8.43 205761_s_at DUS4L 7.72 8.22 8.11 7.41 228707_at CLDN23 4.57 5.03 5.44 4.51 222267_at TMEM209 5.78 6.28 6.35 5.63 218479_s_at XPO4 6.78 7.24 7.02 6.72 203422_at POLD1 7.76 8.27 8.27 7.67 237435_at --- 6.09 6.55 6.30 5.20 221591_s_at FAM64A 7.71 8.21 8.08 7.44 206653_at POLR3G 8.39 8.85 8.62 8.29 208897_s_at DDX18 9.42 9.92 9.63 9.45 209165_at AATF 9.64 10.10 9.87 9.70 213346_at C13orf27 8.80 9.30 9.23 8.53 212603_at MRPS31 8.74 9.20 9.00 8.69 220885_s_at CENPJ 6.31 6.81 6.64 6.07 202833_s_at SERPINA1 3.97 4.42 4.46 3.91 215416_s_at STOML2 9.74 10.24 10.06 9.45 226586_at ANKS6 7.82 8.27 7.99 7.70 202145_at LY6E 9.40 9.89 9.70 9.16 219317_at POLI 6.66 7.11 7.12 6.54 219860_at LY6G5C 5.30 5.80 5.19 5.14 209845_at MKRN1 7.45 7.90 7.86 7.08 225793_at LIX1L 8.27 8.77 8.53 8.14 238700_at --- 6.29 6.74 6.40 5.79 203520_s_at ZNF318 7.36 7.86 7.63 7.31 218593_at RBM28 9.09 9.54 9.29 8.62 218594_at HEATR1 9.62 10.12 9.90 9.73 201663_s_at SMC4 9.48 9.93 9.82 8.76 225777_at C9orf140 8.30 8.80 8.72 8.14 234749_s_at POC1A 6.66 7.11 6.82 6.60 218756_s_at DHRS11 7.14 7.64 7.24 7.25 202240_at PLK1 8.09 8.54 8.55 8.04 223298_s_at NT5C3 9.71 10.21 10.02 9.44 226070_at C9orf142 7.40 7.85 7.67 7.41 1554086_at TUBGCP3 5.43 5.92 5.41 5.30 204466_s_at SNCA 8.53 8.98 9.27 7.75 220841_s_at AHI1 5.86 6.35 6.26 5.94 44040_at FBXO41 7.06 7.51 7.04 6.94 209677_at PRKCI 6.66 7.15 6.98 6.11 218631_at AVPI1 9.21 9.65 9.49 9.19 224731_at HMGB1 11.27 11.76 11.54 10.79 241713_s_at DYX1C1 5.96 6.40 5.91 5.62 208368_s_at BRCA2 4.76 5.25 5.13 4.64 218633_x_at ABHD10 8.13 8.58 8.37 7.89 Appendix E-5A. (continued) E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

177

p<0.05 1.3-fold p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity Probe ID Gene ID NIKS1 Probe ID Gene ID NIKS1 NIKS NIKS16 NIKS16Δ NIKS NIKS16 NIKS16Δ 8 E7 8 E7 205565_s_at FXN 7.71 8.16 7.88 7.16 220742_s_at NGLY1 8.43 8.85 8.44 8.28 1553652_a_at C18orf54 4.52 4.97 5.08 4.04 240436_at LOC650794 5.60 6.01 5.36 5.63 202779_s_at UBE2S 11.00 11.44 11.40 10.50 235168_at PIGM 6.16 6.57 6.20 6.00 202957_at HCLS1 4.18 4.62 4.72 4.14 214507_s_at EXOSC2 9.17 9.58 9.49 8.84 242157_at CHD9 6.06 6.50 6.30 6.12 228526_at LOC100292443 5.60 6.01 5.83 5.33 242685_at GTPBP8 4.34 4.79 4.59 4.38 213053_at HAUS5 6.99 7.40 7.37 6.85 212247_at NUP205 9.28 9.73 9.65 8.91 229674_at SERTAD4 6.65 7.06 7.42 6.60 227718_at PURB 7.26 7.70 7.68 6.99 230521_at C9orf100 5.84 6.26 6.35 5.69 201289_at CYR61 10.52 10.96 11.26 10.33 218980_at FHOD3 7.66 8.07 8.34 7.10 1554239_s_at ZADH2 7.77 8.21 7.95 7.42 229099_at C11orf83 8.33 8.74 8.96 8.23 204317_at GTSE1 6.35 6.79 6.82 6.31 213325_at PVRL3 5.74 6.15 6.49 3.41 218529_at CD320 8.20 8.64 8.40 8.19 200895_s_at FKBP4 10.58 10.98 10.70 10.58 204267_x_at PKMYT1 9.02 9.46 9.21 8.92 209161_at PRPF4 9.61 10.02 9.93 9.27 209153_s_at TCF3 9.29 9.73 9.62 9.07 231431_s_at --- 7.83 8.23 7.97 7.78 211165_x_at EPHB2 6.19 6.63 6.73 5.80 233999_s_at TTC26 6.53 6.94 6.71 5.96 213133_s_at GCSH /// LOC10032 10.28 10.72 10.33 10.27 226284_at ZBTB2 8.25 8.66 8.61 7.92 227642_at TFCP2L1 9.36 9.80 9.74 9.30 1567080_s_at CLN6 8.85 9.26 8.95 8.62 213149_at DLAT 8.62 9.06 8.80 8.59 200679_x_at HMGB1 11.09 11.50 11.39 10.70 213181_s_at MOCS1 6.11 6.54 6.83 6.13 218160_at NDUFA8 10.11 10.52 10.50 9.99 225105_at C12orf75 10.15 10.59 9.96 10.19 1569030_s_at NUB1 8.67 9.07 9.00 8.29 202446_s_at PLSCR1 9.54 9.98 9.97 9.10 235079_at --- 4.58 4.99 4.82 4.17 218244_at NOL8 9.28 9.71 9.76 8.98 201675_at AKAP1 9.72 10.12 9.95 9.67 206205_at MPHOSPH9 6.44 6.87 6.70 6.24 235509_at C8orf38 9.90 10.30 10.06 9.89 220214_at ZNF215 6.98 7.41 7.22 6.84 211623_s_at FBL 11.11 11.51 11.52 10.51 225567_at --- 7.74 8.18 7.88 7.39 231259_s_at --- 7.92 8.32 8.25 7.62 214431_at GMPS 9.86 10.30 10.34 9.71 226781_at C7orf55 7.50 7.90 7.70 7.49 226499_at NRARP 8.14 8.57 8.30 7.63 209162_s_at PRPF4 9.10 9.50 9.38 8.75 233827_s_at SUPT16H 8.24 8.68 8.51 7.81 225802_at TOP1MT 8.74 9.14 8.73 8.56 225344_at NCOA7 9.97 10.40 10.25 9.47 224467_s_at PDCD2L 8.37 8.77 8.68 8.14 229490_s_at --- 6.25 6.69 6.66 5.94 217993_s_at MAT2B 10.57 10.97 11.01 10.58 205995_x_at IQCB1 7.50 7.94 7.76 7.44 210764_s_at CYR61 9.98 10.37 10.80 9.95 205190_at PLS1 8.39 8.83 8.93 7.82 1554480_a_at ARMC10 8.22 8.62 8.32 7.66 225348_at SFRS13A 7.51 7.94 7.64 7.29 209645_s_at ALDH1B1 7.70 8.10 7.89 7.57 212654_at TPM2 6.41 6.85 6.89 6.21 234311_s_at GTPBP10 7.74 8.14 8.07 7.55 219133_at OXSM 8.57 9.00 8.82 8.52 223273_at C14orf142 7.99 8.38 8.28 7.92 202251_at PRPF3 8.06 8.49 8.46 8.05 236265_at SP4 7.07 7.46 7.67 6.91 228217_s_at PSMG4 9.50 9.93 9.71 9.18 227261_at KLF12 6.35 6.74 7.07 5.10 218981_at ACN9 7.83 8.26 8.06 7.54 205875_s_at TREX1 7.52 7.91 7.94 7.48 217426_at --- 5.71 6.14 5.75 5.53 218653_at SLC25A15 8.34 8.73 8.53 8.27 202330_s_at UNG 9.43 9.86 9.79 9.32 1555618_s_at SAE1 9.58 9.97 9.86 9.27 218273_s_at PDP1 7.56 7.99 7.80 6.95 226590_at ZNF618 6.63 7.02 6.69 5.91 208107_s_at LOC81691 6.00 6.43 6.10 5.84 1558152_at LOC100131262 9.29 9.68 9.59 9.00 209680_s_at KIFC1 8.30 8.72 8.64 8.19 201074_at SMARCC1 10.21 10.60 10.42 9.94 1555004_a_at RBL1 5.30 5.73 5.80 5.14 203832_at SNRPF 10.87 11.26 11.24 10.75 223490_s_at EXOSC3 8.57 9.00 8.80 8.08 213558_at PCLO 4.43 4.82 4.72 4.18 224879_at C9orf123 9.83 10.26 9.92 9.59 209646_x_at ALDH1B1 7.31 7.70 7.48 7.08 1557331_at POLR1B 6.71 7.13 6.94 6.55 1559946_s_at RUVBL2 11.05 11.44 11.40 10.75 200844_s_at PRDX6 11.87 12.29 12.27 11.87 218408_at TIMM10 9.43 9.82 9.94 9.33 209811_at CASP2 6.83 7.25 7.18 6.52 1552256_a_at SCARB1 9.85 10.24 9.97 9.83 225202_at RHOBTB3 6.11 6.53 7.84 4.65 1563315_s_at ERICH1 7.33 7.72 7.55 7.12 242671_at --- 5.69 6.11 5.94 5.62 225371_at GLE1 8.37 8.76 8.52 8.00 201919_at SLC25A36 9.27 9.69 9.49 9.02 225237_s_at MSI2 6.30 6.69 6.51 6.13 204805_s_at H1FX 8.33 8.75 8.73 7.87 205379_at CBR3 7.08 7.47 7.52 7.01 201051_at ANP32A 10.25 10.67 10.57 10.09 225887_at C13orf23 8.28 8.66 8.61 8.05 203150_at RABEPK 9.31 9.73 9.38 9.12 222683_at RNF20 8.99 9.37 9.13 8.58 227349_at HELLS 7.20 7.62 7.48 6.81 218866_s_at POLR3K 9.27 9.66 9.41 9.17 224415_s_at HINT2 8.83 9.25 9.27 8.81 209482_at POP7 9.41 9.80 9.72 9.15 1555225_at C1orf43 6.91 7.32 7.45 6.45 225153_at GFM1 10.18 10.57 10.28 9.95 225578_at C13orf37 10.01 10.42 10.35 9.96 203782_s_at POLRMT 8.41 8.79 8.63 8.40 218481_at EXOSC5 8.21 8.62 8.40 8.15 221770_at RPE 8.20 8.59 8.36 8.11 201315_x_at IFITM2 10.52 10.93 11.43 10.07 203564_at FANCG 7.71 8.10 8.09 7.30 201075_s_at SMARCC1 9.29 9.71 9.50 8.95 211673_s_at MOCS1 6.63 7.01 7.20 6.63 Appendix E-5A. (continued) E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

178

p<0.05 1.3-fold Normalized Fluorescence Intensity Probe ID Gene ID NIKS1 NIKS NIKS16 NIKS16Δ 8 E7 243166_at SLC30A5 2.75 3.13 2.83 2.70 205598_at TRAIP 6.21 6.59 6.31 6.06 235117_at CHAC2 8.80 9.18 9.10 8.68 Appendix E-5A. (continued) E7 Dependent Genes Significantly Upregulated (p<0.05, >1.3 fold)

179

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 219795_at SLC6A14 8.72 4.19 5.19 8.42 204750_s_at DSC2 7.55 5.96 5.82 7.35 203757_s_at CEACAM6 9.52 5.84 6.22 10.07 206385_s_at ANK3 6.63 5.06 6.44 8.04 223120_at FUCA2 7.84 4.20 9.21 8.46 205783_at KLK13 10.07 8.50 7.85 9.51 211657_at CEACAM6 9.57 6.22 6.79 9.87 225540_at MAP2 7.05 5.49 6.15 7.07 1553973_a_at SPINK6 10.52 7.18 6.70 11.30 204952_at LYPD3 9.24 7.69 7.62 9.41 205990_s_at WNT5A 7.76 4.45 3.93 6.45 205900_at KRT1 8.09 6.55 6.25 10.75 230835_at KRTDAP 12.09 8.94 8.86 13.34 220723_s_at CWH43 7.90 6.36 5.64 8.41 41469_at PI3 12.82 9.81 10.49 13.15 206953_s_at LPHN2 5.07 3.54 5.56 6.24 203691_at PI3 13.24 10.33 11.02 13.60 220664_at SPRR2C 7.16 5.64 5.82 7.25 232082_x_at SPRR3 11.66 8.75 9.33 12.44 205067_at IL1B 11.64 10.13 11.10 11.23 206642_at DSG1 6.64 3.74 3.71 8.35 1568868_at CYP27C1 6.63 5.12 4.79 5.78 209720_s_at SERPINB3 8.86 5.97 7.24 9.18 226939_at CPEB2 8.20 6.71 7.16 8.77 205513_at TCN1 7.78 4.93 5.06 6.63 231733_at CARD18 6.82 5.33 5.10 6.79 220230_s_at CYB5R2 9.04 6.20 6.64 8.78 220317_at LRAT 6.16 4.68 4.33 5.60 209800_at KRT16 11.82 9.00 8.28 12.85 204751_x_at DSC2 9.76 8.28 8.12 10.02 208539_x_at SPRR2B 12.03 9.23 9.41 12.52 212327_at LIMCH1 6.69 5.21 8.78 6.67 209719_x_at SERPINB3 11.02 8.29 9.42 11.35 238733_at --- 5.86 4.38 5.44 6.03 213796_at SPRR1A 12.47 9.75 9.40 13.66 209459_s_at ABAT 6.54 5.07 4.23 6.87 203021_at SLPI 11.01 8.30 9.60 12.02 206421_s_at SERPINB7 11.02 9.55 9.14 11.23 205863_at S100A12 8.53 5.99 6.35 9.75 218175_at CCDC92 7.13 5.67 5.89 6.59 220322_at IL1F9 9.43 6.92 7.16 9.77 222484_s_at CXCL14 9.26 7.82 7.10 9.69 214599_at IVL 10.69 8.19 8.50 11.44 202342_s_at TRIM2 7.35 5.92 6.29 7.94 1556793_a_at FAM83C 7.64 5.15 7.90 7.94 222838_at SLAMF7 6.48 5.04 5.11 6.77 204351_at S100P 10.27 7.77 10.41 11.51 231867_at ODZ2 10.27 8.84 8.86 9.67 214549_x_at SPRR1A 12.89 10.40 10.08 13.72 202708_s_at HIST2H2BE 6.67 5.24 5.24 6.98 219995_s_at ZNF750 7.08 4.62 5.70 6.60 204779_s_at HOXB7 5.54 4.11 4.26 5.50 235272_at SBSN 9.44 7.14 7.51 11.05 1555673_at LOC730755 6.94 5.51 6.29 8.20 214146_s_at PPBP 6.41 4.12 3.78 7.89 1564307_a_at A2ML1 10.02 8.60 8.01 10.64 231033_at --- 8.24 5.98 6.05 7.96 39402_at IL1B 11.39 9.97 10.91 10.97 231148_at IGFL2 8.62 6.36 6.34 9.42 239587_at --- 4.78 3.37 4.75 4.20 228726_at SERPINB1 9.07 6.84 6.78 8.40 1554921_a_at SCEL 7.35 5.93 6.31 7.79 213572_s_at SERPINB1 10.31 8.11 8.11 9.54 1556499_s_at COL1A1 6.31 4.91 5.33 6.98 213680_at KRT6B 13.88 11.69 11.14 14.03 218002_s_at CXCL14 9.16 7.77 7.06 9.64 205185_at SPINK5 9.56 7.44 7.58 10.58 201341_at ENC1 10.12 8.74 8.86 9.81 214974_x_at CXCL5 6.18 4.06 6.21 5.60 206193_s_at CDSN 6.79 5.41 5.14 9.96 220026_at CLCA4 6.50 4.39 4.89 6.08 227735_s_at C10orf99 6.99 5.61 5.48 8.35 212268_at SERPINB1 10.60 8.50 8.57 9.97 230205_at ZNF561 7.29 5.93 7.36 6.98 212531_at LCN2 9.62 7.53 8.85 9.79 218280_x_at HIST2H2AA3 /// HI 9.27 7.91 8.02 9.45 211906_s_at SERPINB4 6.98 4.91 5.89 6.13 202341_s_at TRIM2 6.46 5.11 5.51 7.13 239430_at IGFL1 9.98 7.96 8.08 10.89 203571_s_at C10orf116 10.77 9.43 9.42 11.72 222242_s_at KLK5 11.13 9.12 9.58 10.92 225646_at CTSC 11.23 9.88 10.14 10.74 227443_at C9orf150 8.02 6.02 6.07 7.33 237690_at GPR115 6.37 5.03 5.61 6.07 210397_at DEFB1 8.15 6.16 6.79 7.96 203608_at ALDH5A1 5.25 3.91 5.21 5.60 205470_s_at KLK11 9.82 7.83 7.54 9.98 215465_at ABCA12 8.49 7.16 7.18 8.77 205064_at SPRR1B 13.67 11.69 11.36 14.04 202488_s_at FXYD3 9.99 8.66 8.80 9.84 210413_x_at SERPINB3 /// SERPINB 9.50 7.52 8.14 9.52 1553031_at GPR115 7.71 6.39 6.96 7.38 209723_at SERPINB9 5.69 3.74 6.41 5.75 227449_at EPHA4 5.23 3.92 3.61 5.71 213240_s_at KRT4 9.14 7.23 6.96 10.93 212657_s_at IL1RN 9.70 8.39 8.33 9.96 1553505_at A2ML1 7.73 5.84 5.90 8.52 201117_s_at CPE 5.40 4.11 4.96 5.05 232056_at SCEL 7.43 5.54 6.06 8.04 227717_at ARHGEF37 8.72 7.43 7.64 8.78 202831_at GPX2 9.49 7.62 7.03 9.07 238067_at TBC1D8B 5.02 3.74 4.07 4.20 1555854_at AKR1C1 /// AKR1C2 6.65 4.82 4.30 5.99 213455_at FAM114A1 8.36 7.08 7.23 8.39 214455_at HIST1H2BC 6.19 4.35 4.59 6.53 241418_at LOC344887 7.40 6.12 5.59 7.56 206884_s_at SCEL 9.61 7.78 8.25 10.44 209118_s_at TUBA1A 8.97 7.69 7.64 8.81 215808_at KLK10 8.37 6.54 6.73 8.30 223720_at SPINK7 6.10 4.83 4.95 8.68 212013_at PXDN 6.92 5.13 6.47 7.82 225987_at STEAP4 7.00 5.73 7.33 8.42 206584_at LY96 6.70 4.92 6.35 6.36 202917_s_at S100A8 14.08 12.82 13.22 14.28 229933_at C1orf74 9.01 7.27 7.81 8.54 235427_at --- 5.52 4.27 4.73 5.40 205916_at S100A7 11.05 9.34 10.09 12.05 226817_at DSC2 10.97 9.72 9.71 11.36 226188_at HSPC159 9.38 7.66 8.39 10.06 218990_s_at SPRR3 7.10 5.86 6.61 10.03 231234_at CTSC 8.31 6.61 7.05 7.84 225502_at DOCK8 4.78 3.54 3.11 5.03 205778_at KLK7 9.36 7.67 7.69 10.19 210609_s_at TP53I3 10.58 9.35 9.41 10.92 Appendix E-5B. E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

180

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 212328_at LIMCH1 5.70 4.01 7.81 5.72 219267_at GLTP 10.41 9.18 9.24 10.82 236313_at CDKN2B 8.67 6.98 6.86 8.19 225391_at LOC93622 8.10 6.87 8.47 8.40 235200_at ZNF561 7.75 6.06 7.56 7.52 236534_at BNIPL 5.34 4.11 4.16 5.69 220724_at CWH43 7.27 5.59 4.90 7.81 206166_s_at CLCA2 10.03 8.81 9.07 10.37 214290_s_at HIST2H2AA3 /// HIST2 10.38 8.71 8.83 10.43 226177_at GLTP 11.92 10.71 10.70 12.36 209160_at AKR1C3 10.39 8.73 8.63 9.80 229400_at HOXD10 7.62 6.41 4.43 7.47 211597_s_at HOPX 6.97 5.32 5.92 9.97 202284_s_at CDKN1A 10.37 9.17 9.43 9.97 239381_at KLK7 9.64 8.01 8.06 10.57 206271_at TLR3 5.14 3.93 5.31 4.92 212012_at PXDN 7.92 6.33 7.38 8.83 222892_s_at TMEM40 8.91 7.70 8.22 9.14 203535_at S100A9 13.02 11.42 12.22 13.37 206164_at CLCA2 10.34 9.15 9.44 10.69 1554195_a_at C5orf46 7.36 6.17 6.37 8.04 209792_s_at KLK10 13.39 12.46 12.38 13.36 210619_s_at HYAL1 7.45 6.26 6.68 7.70 222408_s_at YPEL5 9.76 8.82 9.11 9.89 206165_s_at CLCA2 11.45 10.27 10.54 11.73 204159_at CDKN2C 5.18 4.25 4.71 4.81 220944_at PGLYRP4 8.08 6.91 6.92 8.76 228708_at RAB27B 9.10 8.17 8.37 8.67 243871_at --- 8.00 6.82 6.77 8.73 216973_s_at HOXB7 5.22 4.29 4.63 5.35 212977_at CXCR7 7.37 6.20 6.39 7.12 220414_at CALML5 6.49 5.56 5.95 8.41 217528_at CLCA2 10.35 9.19 9.47 10.71 204241_at ACOX3 7.29 6.36 6.65 7.23 215945_s_at TRIM2 6.77 5.61 6.02 7.11 202796_at SYNPO 7.55 6.62 7.21 7.62 212992_at AHNAK2 9.69 8.54 8.72 10.21 209949_at NCF2 7.23 6.30 5.94 8.31 222859_s_at DAPP1 9.25 8.11 8.66 9.10 226697_at FAM114A1 7.51 6.59 6.81 7.16 219655_at C7orf10 8.80 7.66 8.00 8.60 212441_at KIAA0232 8.25 7.33 7.63 8.22 236193_at HIST1H2BC 6.65 5.51 5.36 7.17 209751_s_at TRAPPC2 /// TRAP 7.96 7.04 7.85 7.47 1558846_at PNLIPRP3 9.68 8.55 8.17 9.89 210592_s_at SAT1 12.53 11.62 12.21 12.19 203921_at CHST2 7.59 6.46 7.31 9.03 1557389_at --- 8.12 7.20 7.76 7.93 235075_at DSG3 10.69 9.56 9.59 10.29 220431_at TMPRSS11E 6.56 5.64 5.82 7.03 236950_s_at LOC157381 5.09 3.97 4.36 5.72 205759_s_at SULT2B1 7.52 6.60 6.94 8.38 239135_at CPPED1 4.58 3.46 3.63 5.26 236009_at PERP 8.50 7.59 7.52 8.88 201295_s_at WSB1 6.77 5.66 5.94 6.67 224414_s_at CARD6 8.32 7.41 8.11 8.19 219909_at MMP28 6.89 5.78 5.55 6.31 211317_s_at CFLAR 7.54 6.63 6.92 7.38 233565_s_at SDCBP2 8.39 7.27 8.11 8.57 205466_s_at HS3ST1 7.60 6.69 7.14 7.39 238847_at --- 5.96 4.85 4.45 5.65 216243_s_at IL1RN 8.37 7.47 7.28 8.65 1553602_at MUCL1 5.92 4.81 4.93 7.24 213172_at TTC9 3.62 2.71 2.69 3.64 205625_s_at CALB1 5.88 4.78 4.10 6.56 214475_x_at CAPN3 5.41 4.50 4.53 5.54 238430_x_at SLFN5 8.66 7.57 7.81 8.22 222223_s_at IL1F5 6.50 5.59 5.67 7.18 213988_s_at SAT1 11.09 10.00 10.68 10.61 202263_at CYB5R1 11.62 10.71 10.81 11.80 238542_at ULBP2 9.42 8.34 8.50 9.46 217783_s_at YPEL5 10.13 9.23 9.55 10.43 201425_at ALDH2 9.18 8.11 8.85 9.47 204464_s_at EDNRA 4.27 3.37 3.16 4.23 214370_at S100A8 9.43 8.36 8.73 9.47 201249_at SLC2A1 7.56 6.66 6.40 7.71 201963_at ACSL1 7.82 6.75 7.62 8.01 236119_s_at SPRR2G 7.66 6.76 6.78 8.95 209569_x_at D4S234E /// FOXP1 9.20 8.13 8.96 9.12 219681_s_at RAB11FIP1 7.13 6.24 6.23 6.72 203139_at DAPK1 8.07 7.01 7.31 8.42 223878_at INPP4B 4.27 3.37 3.89 4.39 206191_at ENTPD3 9.80 8.74 8.56 9.26 227531_at --- 9.46 8.56 8.93 9.20 229125_at KANK4 6.69 5.63 4.95 7.21 227001_at NIPAL2 8.83 7.94 7.99 8.80 204475_at MMP1 9.78 8.72 6.80 10.37 226725_at --- 8.80 7.91 8.25 8.53 234331_s_at FAM84A 7.93 6.87 7.09 8.18 235198_at OSTM1 7.82 6.94 7.55 7.53 229614_at ZNF320 5.25 4.20 4.31 4.72 226853_at BMP2K 7.17 6.29 6.74 6.80 207373_at HOXD10 5.78 4.73 4.02 5.62 222858_s_at DAPP1 8.24 7.36 7.73 8.06 217497_at TYMP 7.36 6.31 6.54 7.54 1564630_at EDN1 7.25 6.38 6.90 6.98 200696_s_at GSN 10.42 9.38 9.28 10.45 219410_at TMEM45A 8.85 7.98 8.31 9.48 204141_at TUBB2A 11.46 10.41 10.09 11.15 219503_s_at TMEM40 7.43 6.56 6.89 7.77 213836_s_at WIPI1 6.95 5.91 6.46 7.62 222686_s_at CPPED1 6.59 5.73 5.80 6.88 205595_at DSG3 11.07 10.02 10.06 10.96 203788_s_at SEMA3C 7.61 6.76 6.79 7.22 209460_at ABAT 7.14 6.11 5.52 7.53 230278_at --- 4.54 3.69 3.73 4.09 222162_s_at ADAMTS1 7.77 6.74 7.01 7.39 202842_s_at DNAJB9 8.62 7.76 7.75 8.48 209373_at MALL 10.00 8.97 8.87 10.30 224604_at C4orf3 10.10 9.25 9.23 10.14 211361_s_at SERPINB13 8.49 7.47 6.82 8.31 207275_s_at ACSL1 7.48 6.63 7.25 7.74 222016_s_at ZNF323 5.99 4.96 5.19 6.11 203303_at DYNLT3 10.04 9.19 9.46 10.15 236449_at CSTB 7.73 6.71 6.68 7.83 210239_at IRX5 7.04 6.20 5.99 7.02 213780_at TCHH 5.07 4.06 4.59 6.25 202411_at IFI27 6.40 5.55 5.55 6.48 223544_at TMEM79 8.47 7.45 7.69 8.60 204137_at GPR137B 7.48 6.64 6.63 7.17 1558378_a_at AHNAK2 7.24 6.22 6.31 7.41 209369_at ANXA3 10.96 10.12 10.46 10.72 219476_at C1orf116 9.05 8.04 8.26 8.94 225611_at MAST4 9.02 8.18 8.50 8.66 230333_at --- 8.53 7.52 7.88 8.26 209173_at AGR2 6.43 5.59 6.28 7.77 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

181

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 217315_s_at KLK13 8.28 7.27 7.00 7.74 1554253_a_at LASS3 8.30 7.47 7.40 8.12 211316_x_at CFLAR 7.63 6.63 6.98 7.62 210564_x_at CFLAR 7.62 6.79 7.07 7.65 210074_at CTSL2 10.70 9.70 9.56 10.86 266_s_at CD24 11.00 10.17 9.99 10.91 202310_s_at COL1A1 6.78 5.77 6.27 7.31 209570_s_at D4S234E /// FOXP 8.04 7.21 7.80 7.76 231439_at FAM84A 5.56 4.57 4.92 5.67 225915_at CAB39L 7.07 6.24 6.43 6.83 211862_x_at CFLAR 7.68 6.68 7.04 7.55 206448_at ZNF365 6.68 5.85 5.40 6.77 225647_s_at CTSC 12.08 11.09 11.37 11.73 203789_s_at SEMA3C 10.99 10.16 10.24 10.76 204858_s_at TYMP 9.17 8.17 8.32 9.07 202967_at GSTA4 7.59 6.77 7.17 8.09 1552502_s_at RHBDL2 6.23 5.24 5.22 6.29 224567_x_at MALAT1 11.27 10.45 10.49 11.16 238017_at SDR16C5 7.98 7.00 8.23 8.87 220782_x_at KLK12 7.08 6.26 6.34 8.64 213400_s_at TBL1X 6.49 5.51 6.71 6.90 226636_at PLD1 6.56 5.74 5.93 6.26 224480_s_at AGPAT9 6.22 5.25 5.56 6.06 206969_at KRT34 5.30 4.49 4.63 6.08 208485_x_at CFLAR 7.72 6.75 7.09 7.56 201286_at SDC1 11.95 11.13 11.10 11.90 212325_at LIMCH1 6.03 5.06 7.94 5.95 211668_s_at PLAU 10.55 9.74 9.63 10.36 242705_x_at LOC100289219 7.37 6.41 7.03 7.40 218610_s_at CPPED1 4.17 3.36 3.55 4.37 209365_s_at ECM1 8.41 7.45 7.14 9.68 219290_x_at DAPP1 8.44 7.63 7.98 8.28 203827_at WIPI1 7.07 6.11 6.74 7.85 209637_s_at RGS12 7.03 6.23 6.54 6.97 201579_at FAT1 10.93 9.99 10.40 10.89 215785_s_at CYFIP2 5.85 5.05 5.37 5.71 1555786_s_at C14orf34 7.79 6.85 6.76 8.79 218638_s_at SPON2 7.17 6.37 6.23 7.78 244050_at PTPLAD2 7.27 6.33 6.96 7.40 231726_at PCDHB14 3.31 2.51 2.40 3.49 201116_s_at CPE 7.78 6.84 7.55 7.67 1555416_a_at ALOX15B 9.46 8.66 8.42 9.91 209508_x_at CFLAR 7.59 6.65 6.98 7.44 210944_s_at CAPN3 7.13 6.34 6.43 7.15 1554250_s_at TRIM73 6.07 5.27 5.43 5.81 219508_at GCNT3 7.87 7.13 7.36 9.18 202242_at TSPAN7 5.87 5.08 5.12 6.20 213273_at ODZ4 6.82 6.09 7.19 6.83 203037_s_at MTSS1 11.41 10.61 10.88 11.44 209668_x_at CES2 9.68 8.95 8.74 10.33 226605_at DGKQ 7.88 7.09 7.30 7.89 221291_at ULBP2 8.04 7.30 7.35 8.09 227568_at HECTD2 5.70 4.91 5.45 5.61 212753_at PCGF3 7.94 7.21 7.72 7.84 206400_at LGALS7 /// LGALS7B 11.82 11.03 11.21 12.20 244353_s_at SLC2A12 6.26 5.53 5.94 6.42 225613_at MAST4 9.07 8.28 8.57 8.81 228964_at PRDM1 6.78 6.05 6.09 6.74 227702_at CYP4X1 4.80 4.01 5.51 4.95 222878_s_at OTUB2 5.46 4.73 4.73 5.66 206714_at ALOX15B 10.07 9.28 9.06 10.56 203381_s_at APOE 8.31 7.58 8.21 8.67 219532_at ELOVL4 7.13 6.35 6.71 8.01 1554252_a_at LASS3 7.19 6.46 6.37 7.11 219225_at PGBD5 6.61 5.82 6.44 6.63 211824_x_at NLRP1 6.68 5.95 6.22 6.59 209283_at CRYAB 8.44 7.66 7.73 9.45 214703_s_at MAN2B2 7.81 7.08 7.16 7.78 205479_s_at PLAU 10.65 9.87 9.77 10.56 202575_at CRABP2 9.29 8.56 9.38 9.88 214823_at ZNF204P 5.18 4.40 4.62 5.39 203192_at ABCB6 6.24 5.52 6.51 6.64 225912_at TP53INP1 8.25 7.47 7.94 8.32 231817_at USP53 7.31 6.58 6.98 7.09 203585_at ZNF185 8.75 7.98 7.94 9.35 225949_at NRBP2 9.01 8.29 7.97 8.91 218834_s_at TMEM132A 9.25 8.48 8.74 9.05 219369_s_at OTUB2 7.19 6.47 6.51 7.35 220620_at CRCT1 5.91 5.14 5.32 8.47 226125_at LOC100288152 7.50 6.77 6.88 7.56 203335_at PHYH 7.97 7.21 7.36 8.30 223696_at ARSD 6.95 6.22 6.54 7.21 241359_at TLCD2 6.79 6.02 6.15 6.75 201734_at CLCN3 9.57 8.85 9.23 9.60 210020_x_at CALML3 7.64 6.87 6.75 8.97 227342_s_at MYEOV 8.26 7.54 7.92 8.33 219998_at HSPC159 6.97 6.20 6.62 7.78 201201_at CSTB 12.98 12.26 12.26 13.38 212062_at ATP9A 7.26 6.50 7.15 7.66 229189_s_at --- 6.10 5.38 5.87 6.66 201294_s_at WSB1 7.73 6.98 7.24 7.68 229285_at RNASEL 6.25 5.53 5.50 6.24 224901_at SCD5 8.40 7.64 8.54 8.91 225955_at METRNL 6.46 5.75 5.98 6.72 202982_s_at ACOT1 /// ACOT2 4.38 3.62 3.51 4.55 209667_at CES2 9.73 9.02 8.78 10.36 214123_s_at C4orf10 6.50 5.75 6.11 6.47 238439_at ANKRD22 8.31 7.61 8.42 8.81 205823_at RGS12 8.04 7.29 7.57 7.91 218802_at CCDC109B 7.98 7.28 8.00 7.89 240024_at SEC14L2 7.73 6.98 7.42 7.91 220573_at KLK14 7.16 6.46 6.47 7.83 220159_at ABCA11P 6.16 5.40 5.91 5.91 211548_s_at HPGD 4.79 4.09 4.18 5.82 238654_at VSIG10L 6.41 5.66 5.58 7.78 224828_at CPEB4 7.04 6.34 6.18 7.00 204980_at CLOCK 7.90 7.15 7.36 7.70 205016_at TGFA 10.05 9.35 9.57 9.82 241994_at XDH 10.42 9.67 10.00 10.51 228316_at C2orf63 6.15 5.45 5.97 6.64 223577_x_at MALAT1 7.54 6.80 6.82 7.33 1569868_s_at EME2 7.74 7.05 7.06 7.80 202351_at ITGAV 11.02 10.27 10.49 10.93 214109_at LRBA 8.14 7.44 8.06 8.15 237737_at LOC100289026 4.75 4.01 4.76 4.70 215071_s_at HIST1H2AC 7.89 7.19 7.47 8.35 220945_x_at MANSC1 8.25 7.50 8.31 7.97 210301_at XDH 8.51 7.82 8.21 8.49 216470_x_at PRSS1 /// PRSS2 /// P 7.42 6.68 7.08 7.34 230083_at USP53 7.80 7.11 7.45 7.67 212741_at MAOA 10.05 9.31 9.49 9.93 214791_at SP140L 7.21 6.51 6.90 6.92 208651_x_at CD24 10.30 9.55 9.44 10.30 212573_at ENDOD1 9.18 8.49 8.14 9.24 222486_s_at ADAMTS1 6.48 5.74 5.65 6.20 214848_at --- 9.12 8.43 8.28 9.45 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

182

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 1553534_at NLRP10 4.84 4.15 4.08 4.83 209107_x_at NCOA1 6.70 6.08 6.47 7.09 212841_s_at PPFIBP2 5.63 4.94 4.64 6.33 232366_at KIAA0232 5.06 4.43 4.53 5.08 213174_at TTC9 6.98 6.29 6.26 7.10 205932_s_at MSX1 7.02 6.39 5.89 7.15 225334_at C10orf32 9.04 8.35 8.44 9.17 218652_s_at PIGG 9.60 8.97 9.56 9.61 220468_at ARL14 5.64 4.95 4.86 5.50 230435_at LOC375190 5.77 5.15 5.19 6.27 235763_at SLC44A5 6.73 6.04 6.14 6.60 219799_s_at DHRS9 7.45 6.83 6.96 7.72 238710_at TMEM86A 5.90 5.22 5.41 7.29 235072_s_at --- 9.92 9.30 9.31 9.93 221645_s_at ZNF83 5.90 5.21 5.47 5.76 212807_s_at SORT1 5.42 4.80 5.70 5.96 226425_at CLIP4 8.99 8.31 8.55 9.20 205717_x_at PCDHGA1 /// PCDH 8.36 7.74 7.94 8.47 202489_s_at FXYD3 12.11 11.43 11.46 12.12 218651_s_at LARP6 6.70 6.08 5.87 6.78 217730_at TMBIM1 11.11 10.43 10.35 11.23 242317_at HIGD1A 7.89 7.27 7.01 7.70 227450_at ERP27 5.60 4.92 5.38 6.15 207950_s_at ANK3 5.49 4.87 5.36 5.90 224602_at C4orf3 11.24 10.56 10.70 11.34 211026_s_at MGLL 7.99 7.37 7.89 8.06 201250_s_at SLC2A1 10.46 9.78 9.53 10.67 213288_at MBOAT2 8.34 7.73 7.65 8.66 223341_s_at SCOC 9.03 8.35 8.67 8.77 204777_s_at MAL 5.78 5.17 4.95 7.77 202974_at MPP1 6.60 5.92 6.67 6.89 225102_at MGLL 8.06 7.44 8.03 8.16 222853_at FLRT3 9.95 9.27 9.14 9.96 235988_at GPR110 5.40 4.78 5.19 5.54 228084_at PLA2G12A 7.35 6.67 7.27 7.24 1568932_at --- 6.64 6.02 6.01 7.18 211066_x_at PCDHGA1 /// PCDHGA 8.80 8.13 8.33 8.80 236798_at --- 6.59 5.97 5.86 6.75 209772_s_at CD24 10.26 9.59 9.52 10.30 209771_x_at CD24 13.08 12.46 12.31 13.12 209106_at NCOA1 5.03 4.37 4.56 5.61 227020_at YPEL2 5.84 5.22 5.23 6.60 206746_at BFSP1 6.10 5.43 5.35 6.55 226657_at C17orf103 6.00 5.39 5.31 6.19 1552757_s_at C9orf66 3.61 2.94 3.21 3.58 218552_at ECHDC2 7.23 6.62 6.80 7.26 235678_at GM2A 8.85 8.18 8.71 9.01 236248_x_at TADA2B 6.58 5.97 6.39 6.50 213509_x_at CES2 9.68 9.01 8.75 10.34 213421_x_at PRSS3 9.11 8.50 8.85 9.24 40016_g_at MAST4 9.10 8.43 8.63 8.86 205376_at INPP4B 9.61 9.00 9.38 9.61 212463_at CD59 9.69 9.03 9.57 9.77 217744_s_at PERP 12.53 11.92 11.80 12.74 209005_at FBXL5 7.44 6.78 7.34 7.27 219597_s_at DUOX1 8.25 7.64 7.44 8.51 203382_s_at APOE 7.58 6.92 7.47 7.82 209270_at LAMB3 11.77 11.16 10.67 12.02 203961_at NEBL 8.05 7.39 7.18 8.48 226568_at FAM102B 5.13 4.52 4.11 5.58 201287_s_at SDC1 11.70 11.04 10.98 11.64 213353_at ABCA5 6.53 5.92 6.12 6.54 212659_s_at IL1RN 8.08 7.42 7.35 8.41 240284_x_at EMG1 7.53 6.93 7.08 7.88 210935_s_at WDR1 9.44 8.78 9.21 9.32 201130_s_at CDH1 9.62 9.01 8.99 9.70 214536_at SLURP1 6.46 5.80 5.62 7.98 201481_s_at PYGB 11.05 10.44 10.27 11.07 214736_s_at ADD1 8.97 8.31 8.68 8.89 223278_at GJB2 12.34 11.74 11.69 12.42 208030_s_at ADD1 8.82 8.17 8.54 8.69 204971_at CSTA 13.28 12.67 12.65 13.41 204341_at TRIM16 10.65 10.00 9.97 10.72 233687_s_at KLK8 /// KLK9 7.72 7.11 7.22 7.93 236101_at --- 7.21 6.56 7.11 7.11 238028_at C6orf132 7.60 7.00 6.94 7.60 209079_x_at PCDHGA1 /// PCDHGA 8.59 7.94 8.25 8.74 205157_s_at KRT17 13.39 12.79 12.73 13.65 235197_s_at OSTM1 6.80 6.15 6.56 6.61 236489_at GPR110 4.95 4.35 4.69 5.11 209318_x_at PLAGL1 5.82 5.17 7.21 6.86 241769_at --- 5.38 4.78 4.90 5.20 229860_x_at C4orf48 8.66 8.00 8.49 8.58 212503_s_at DIP2C 5.55 4.95 5.38 6.06 216379_x_at CD24 13.02 12.37 12.23 13.10 234989_at --- 9.48 8.88 9.99 9.53 201884_at CEACAM5 5.12 4.47 4.45 5.33 202074_s_at OPTN 9.93 9.33 9.36 9.88 1558687_a_at --- 6.96 6.31 6.04 8.12 237732_at PRR9 5.82 5.22 5.24 8.53 203108_at GPRC5A 10.43 9.78 9.76 10.47 217971_at MAPKSP1 9.81 9.22 9.71 9.82 219561_at COPZ2 7.71 7.06 8.07 7.90 225283_at ARRDC4 9.11 8.51 8.61 9.94 207002_s_at PLAGL1 4.96 4.31 6.14 5.68 220753_s_at CRYL1 6.53 5.94 5.98 6.74 202073_at OPTN 8.62 7.97 8.04 8.57 210563_x_at CFLAR 8.44 7.85 8.03 8.52 209425_at AMACR /// C1QTNF3 5.48 4.83 4.99 5.83 238584_at IQCA1 3.38 2.79 3.05 3.52 229146_at C7orf31 5.67 5.03 5.03 5.41 206833_s_at ACYP2 7.43 6.84 7.46 7.90 224829_at CPEB4 6.26 5.62 5.64 6.18 1558281_a_at TMEM184A 6.91 6.32 6.26 6.94 212338_at MYO1D 8.25 7.61 7.65 8.18 203600_s_at FAM193A 7.89 7.30 7.87 8.09 207463_x_at PRSS3 8.68 8.04 8.33 8.74 226671_at LAMP2 8.41 7.82 8.06 8.34 200611_s_at WDR1 11.16 10.52 10.99 11.16 203407_at PPL 9.52 8.93 9.02 10.20 203821_at HBEGF 9.20 8.56 9.00 8.98 224983_at SCARB2 9.83 9.24 9.73 9.72 232995_at --- 6.73 6.09 6.48 7.02 225246_at STIM2 7.25 6.66 7.23 7.17 220191_at GKN1 5.46 4.82 4.87 5.38 212692_s_at LRBA 9.62 9.04 9.43 9.70 238567_at SGPP2 6.98 6.35 8.15 8.03 226909_at ZNF518B 8.27 7.68 8.05 8.19 208650_s_at CD24 11.29 10.66 10.53 11.24 204518_s_at PPIC 7.87 7.28 7.77 8.06 226185_at CDS1 9.63 9.00 9.34 9.59 205807_s_at TUFT1 8.61 8.03 8.50 8.77 213279_at DHRS1 10.00 9.37 9.37 10.40 210113_s_at NLRP1 7.25 6.67 6.96 7.06 220066_at NOD2 6.62 5.99 6.20 7.05 201990_s_at CREBL2 6.64 6.06 6.29 6.67 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

183

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 203962_s_at NEBL 8.10 7.52 7.24 8.30 204797_s_at EML1 4.84 4.30 4.23 6.16 225667_s_at FAM84A 9.37 8.79 8.98 9.64 91826_at EPS8L1 7.60 7.06 7.13 7.72 218380_at LOC728392 /// NLRP1 6.63 6.05 6.17 6.78 242103_at TMEM86A 6.25 5.71 5.77 7.27 204682_at LTBP2 8.13 7.55 7.88 8.17 217767_at C3 10.60 10.07 11.15 11.08 201735_s_at CLCN3 8.92 8.34 8.77 8.89 242079_at RGS12 6.97 6.43 6.68 6.82 201412_at LRP10 10.67 10.09 10.27 10.48 225001_at RAB3D 9.87 9.34 9.42 9.96 209691_s_at DOK4 7.52 6.94 6.69 7.56 213982_s_at RABGAP1L 7.07 6.54 6.73 7.08 223194_s_at SLC22A23 6.89 6.32 6.50 7.66 1554897_s_at RHBDL2 6.85 6.31 6.19 6.94 213272_s_at TMEM159 8.15 7.58 7.45 8.07 227326_at MXRA7 7.27 6.74 7.18 7.15 226198_at TOM1L2 8.40 7.82 7.55 8.60 212570_at ENDOD1 7.63 7.10 6.80 7.48 200609_s_at WDR1 11.36 10.79 11.35 11.43 203036_s_at MTSS1 7.76 7.23 7.35 7.83 230523_at QSOX1 5.66 5.08 4.97 6.11 226726_at MBOAT2 9.84 9.31 9.14 9.85 209260_at SFN 11.96 11.38 11.36 12.06 210118_s_at IL1A 11.24 10.71 11.15 11.45 212779_at KIAA1109 7.60 7.03 7.40 7.47 211822_s_at NLRP1 6.51 5.98 6.22 6.56 35820_at GM2A 10.56 9.99 10.27 10.77 221269_s_at SH3BGRL3 12.04 11.51 11.44 12.15 224261_at --- 5.64 5.07 5.21 5.57 202630_at APPBP2 7.78 7.25 7.66 7.95 239061_at TPRXL 6.96 6.39 6.39 6.86 222006_at LETM1 8.81 8.28 8.59 8.75 202912_at ADM 10.48 9.91 9.80 10.58 213895_at EMP1 7.12 6.59 6.63 7.25 238689_at GPR110 6.44 5.88 6.43 6.62 1553695_a_at NLRX1 7.62 7.09 7.20 7.89 209732_at CLEC2B 5.52 4.95 5.20 6.28 208812_x_at HLA-C 11.26 10.74 10.99 11.15 221541_at CRISPLD2 5.42 4.86 4.76 5.43 234316_x_at KLK12 7.13 6.60 6.66 8.39 219648_at MREG 9.05 8.48 8.71 8.95 1564010_at CAST 5.27 4.75 4.74 5.25 231941_s_at MUC20 3.91 3.34 3.62 4.10 202281_at GAK 8.07 7.55 7.94 8.11 227747_at MPZL3 8.02 7.46 7.51 8.35 209498_at CEACAM1 5.22 4.70 5.55 5.68 242323_at PLA2G12A 8.50 7.94 8.39 8.52 205402_x_at PRSS2 8.10 7.58 7.75 8.12 224928_at SETD7 10.38 9.82 10.27 10.48 222758_s_at TMEM132A 7.33 6.81 6.91 7.21 204636_at COL17A1 10.22 9.67 9.58 10.45 219432_at EVC 6.94 6.42 6.56 6.93 215800_at DUOX1 7.21 6.66 6.67 7.24 230778_at --- 7.89 7.38 7.55 7.93 236224_at RIT1 6.38 5.82 6.08 6.33 215606_s_at ERC1 7.94 7.43 7.65 8.10 227961_at CTSB 10.38 9.82 10.12 10.67 227052_at --- 7.82 7.31 7.42 8.16 220260_at TBC1D19 6.60 6.05 6.50 6.64 220145_at MAP9 7.47 6.96 7.17 7.49 212737_at GM2A 11.97 11.41 11.67 12.16 204733_at KLK6 6.43 5.92 6.38 8.02 209420_s_at SMPD1 6.59 6.03 6.19 6.52 229018_at C12orf26 7.70 7.19 7.20 7.59 1552370_at C4orf33 6.03 5.48 5.94 5.86 1569157_s_at ZNF846 3.51 3.00 3.26 3.62 227272_at C15orf52 9.03 8.47 8.60 9.06 207317_s_at CASQ2 5.81 5.30 5.16 5.98 202421_at IGSF3 9.80 9.25 9.29 9.78 208191_x_at PSG4 5.58 5.07 5.12 6.02 227752_at SPTLC3 6.35 5.80 5.85 6.81 213627_at MAGED2 8.54 8.03 8.19 8.49 209140_x_at HLA-B 10.55 10.00 10.43 10.61 221036_s_at APH1B 7.09 6.58 6.69 7.32 209596_at MXRA5 4.70 4.15 4.20 5.42 1553077_at SDR9C7 4.65 4.14 3.90 6.26 209267_s_at SLC39A8 8.74 8.19 8.92 8.98 225784_s_at ZC4H2 6.55 6.04 6.80 6.72 226474_at NLRC5 5.90 5.35 5.55 5.84 219423_x_at TNFRSF25 8.07 7.56 7.56 8.30 230316_at SEC14L2 7.57 7.02 7.23 7.78 243582_at SH3RF2 7.07 6.57 7.08 7.07 201733_at CLCN3 5.82 5.27 5.53 5.65 229711_s_at MDM2 9.03 8.53 8.96 9.04 224558_s_at MALAT1 9.25 8.70 8.83 9.64 226440_at DUSP22 7.26 6.75 6.73 7.19 225579_at PQLC3 8.18 7.63 6.80 8.37 202996_at POLD4 8.23 7.72 7.75 8.24 209123_at QDPR 10.31 9.76 10.16 10.25 225301_s_at MYO5B 8.90 8.39 8.57 9.15 228029_at ZNF721 6.36 5.81 6.44 6.21 204242_s_at ACOX3 7.25 6.75 6.83 7.31 235651_at TTC22 7.40 6.86 7.14 8.14 240119_at TEPP 5.18 4.68 4.94 5.16 227142_at PLEKHG5 7.63 7.09 7.10 7.68 226126_at TBCK 8.05 7.55 7.87 8.38 202908_at WFS1 7.54 7.00 7.11 7.60 206192_at CDSN 6.68 6.18 5.89 9.04 1554283_at CCRN4L 6.39 5.85 6.04 6.38 226152_at TTC7B 7.55 7.04 7.31 7.47 207324_s_at DSC1 3.31 2.76 2.93 3.78 226502_at ELMOD2 8.33 7.83 8.18 8.31 224818_at SORT1 7.46 6.92 7.71 8.17 204517_at PPIC 9.28 8.78 9.28 9.41 210249_s_at NCOA1 6.55 6.01 6.32 7.10 50400_at PAOX 5.79 5.29 4.91 5.83 205330_at MN1 7.14 6.59 6.51 7.32 235352_at --- 6.32 5.82 5.80 6.48 225177_at RAB11FIP1 7.75 7.21 7.24 7.69 212221_x_at IDS 9.03 8.53 8.55 9.28 210398_x_at FUT6 5.83 5.29 5.42 6.00 223431_at CNO 8.15 7.65 8.02 8.12 209939_x_at CFLAR 8.72 8.19 8.39 8.92 202439_s_at IDS 6.52 6.03 5.80 6.71 235534_at --- 7.80 7.26 7.35 7.81 1552797_s_at PROM2 10.51 10.01 10.07 10.75 225019_at CAMK2D 8.26 7.72 7.58 8.57 214726_x_at ADD1 8.22 7.72 7.95 8.16 219991_at SLC2A9 9.09 8.55 8.93 9.02 209166_s_at MAN2B1 8.60 8.11 8.12 8.88 210359_at MTSS1 5.32 4.78 4.98 5.31 230093_at RSPH1 5.17 4.68 5.00 5.38 234335_s_at FAM84A 3.40 2.86 2.98 3.45 210473_s_at GPR125 9.26 8.77 9.13 9.39 40225_at GAK 9.68 9.15 9.53 9.70 204447_at ProSAPiP1 6.40 5.91 6.04 6.55 235746_s_at --- 4.60 4.06 4.10 4.46 205966_at TAF13 5.85 5.36 5.32 5.75 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

184

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 203390_s_at KIF3C 7.91 7.42 7.58 7.81 1569112_at SLC44A5 6.56 6.11 6.14 6.66 223497_at FAM135A 8.56 8.07 8.37 8.83 201276_at RAB5B 8.75 8.29 8.43 8.82 228948_at EPHA4 4.37 3.88 3.62 4.79 225299_at MYO5B 5.72 5.27 5.37 5.72 209882_at RIT1 8.24 7.75 7.94 8.31 1552620_at SPRR4 7.65 7.19 7.08 9.29 236201_at --- 5.74 5.25 5.04 5.79 241931_at XG 5.61 5.15 5.23 5.63 219680_at NLRX1 8.76 8.27 8.25 9.08 209126_x_at KRT6B 13.73 13.28 13.09 13.80 225286_at ARSD 7.36 6.87 7.07 7.71 244663_at --- 3.52 3.07 3.29 3.82 234725_s_at SEMA4B 9.68 9.20 9.28 9.65 207335_x_at ATP5I 10.70 10.25 10.71 10.67 213656_s_at KLC1 9.12 8.63 8.75 9.01 208025_s_at HMGA2 10.68 10.23 10.77 10.96 227193_at --- 8.13 7.64 7.64 8.42 203904_x_at CD82 10.02 9.57 9.69 10.12 203870_at USP46 7.21 6.73 6.98 7.11 226692_at SERF2 7.46 7.01 7.15 7.59 212223_at IDS 8.52 8.04 7.95 8.59 37152_at PPARD 9.11 8.66 8.72 9.06 222469_s_at TOLLIP 7.62 7.13 7.16 7.63 202447_at DECR1 10.01 9.56 9.48 9.99 202636_at RNF103 8.41 7.92 7.94 8.69 213980_s_at CTBP1 9.44 8.99 9.50 9.42 220187_at STEAP4 4.57 4.09 4.67 5.16 225220_at SNHG8 9.59 9.15 9.67 9.81 204058_at ME1 9.75 9.27 8.88 9.75 218432_at FBXO3 8.27 7.82 7.96 8.44 215821_x_at PSG3 6.56 6.08 5.84 6.64 219492_at CHIC2 9.38 8.93 9.44 9.45 219696_at DENND1B 7.19 6.71 6.81 7.43 210065_s_at UPK1B 10.78 10.33 9.90 11.30 1555270_a_at WFS1 6.71 6.23 6.36 6.90 209442_x_at ANK3 6.44 5.99 6.25 7.44 227410_at FAM43A 5.49 5.01 5.24 5.46 201131_s_at CDH1 11.90 11.46 11.47 12.03 241455_at C6orf132 7.31 6.83 6.74 7.60 203718_at PNPLA6 8.27 7.83 8.04 8.27 201924_at AFF1 8.66 8.19 8.71 8.57 223423_at GPR160 5.32 4.88 5.16 6.30 202180_s_at MVP 9.64 9.17 9.12 9.91 203460_s_at PSEN1 9.27 8.82 8.83 9.29 212357_at FAM168A 7.24 6.77 6.92 7.20 222067_x_at HIST1H2BD 7.81 7.37 7.63 8.12 228846_at MXD1 7.58 7.11 7.11 7.94 229190_at --- 5.10 4.65 5.00 5.57 210561_s_at WSB1 10.62 10.15 10.20 10.70 229297_at --- 6.89 6.44 6.35 6.99 1553081_at WFDC12 7.08 6.61 6.60 7.84 227345_at TNFRSF10D 8.34 7.89 8.13 8.44 228865_at C1orf116 9.45 8.97 8.95 9.47 225314_at OCIAD2 11.64 11.20 11.51 11.72 232837_at KIF13A 6.90 6.42 6.46 7.11 220149_at C2orf54 5.72 5.28 5.35 5.70 227098_at DUSP18 6.99 6.52 6.36 7.09 1557114_a_at LOC284385 5.82 5.38 5.94 6.09 226485_at VSIG10 7.56 7.08 6.89 7.92 221843_s_at KIAA1609 9.62 9.18 9.29 9.81 218017_s_at HGSNAT 8.83 8.36 8.35 8.86 217995_at SQRDL 11.73 11.29 11.74 11.96 224622_at TBC1D14 9.09 8.62 9.07 9.15 209727_at GM2A 8.16 7.72 7.91 8.20 220999_s_at CYFIP2 4.11 3.64 3.68 4.08 212135_s_at ATP2B4 8.50 8.06 8.05 8.83 209215_at MFSD10 8.90 8.43 8.77 8.83 635_s_at PPP2R5B 7.19 6.75 6.72 7.29 218231_at NAGK 8.82 8.35 8.39 9.13 235417_at SPOCD1 6.55 6.11 6.18 6.52 229566_at LOC645638 7.02 6.55 6.62 7.39 225224_at C20orf112 6.98 6.54 7.52 7.31 204778_x_at HOXB7 6.71 6.24 6.23 6.85 214604_at HOXD11 7.23 6.80 5.60 7.22 226145_s_at FRAS1 3.96 3.49 4.38 4.33 200945_s_at SEC31A 10.81 10.38 10.74 10.84 33494_at ETFDH 6.79 6.32 6.51 6.82 228910_at --- 9.15 8.72 8.82 9.47 218773_s_at MSRB2 9.39 8.92 8.88 9.33 214686_at ZNF266 8.20 7.76 8.14 8.37 210847_x_at TNFRSF25 7.45 6.99 6.93 7.69 218845_at DUSP22 8.84 8.40 8.46 8.81 214279_s_at NDRG2 5.37 4.91 4.94 5.81 226622_at MUC20 5.67 5.24 5.35 5.91 203912_s_at DNASE1L1 9.48 9.02 8.91 9.68 223594_at TMEM117 7.73 7.29 7.25 7.80 213110_s_at COL4A5 9.40 8.94 9.29 9.70 214772_at C11orf41 4.56 4.13 4.48 4.55 204332_s_at AGA 7.86 7.39 7.55 7.97 208682_s_at MAGED2 8.40 7.97 8.18 8.38 221655_x_at EPS8L1 7.47 7.01 7.02 7.40 218779_x_at EPS8L1 7.19 6.75 6.69 7.24 235882_at VPS53 6.39 5.93 5.67 6.50 214088_s_at FUT3 7.96 7.52 7.82 8.41 209492_x_at ATP5I 10.94 10.48 11.00 10.97 212236_x_at KRT17 13.91 13.48 13.38 14.12 211911_x_at HLA-B 10.02 9.56 9.84 10.11 214203_s_at PRODH 7.40 6.97 6.69 7.95 218727_at SLC38A7 7.30 6.84 6.77 7.41 232381_s_at DNAH5 6.80 6.36 6.43 6.75 242998_at RDH12 5.41 4.95 4.92 5.63 226791_at KIFC2 8.23 7.80 7.53 8.62 242951_at --- 5.68 5.22 5.18 7.70 227134_at SYTL1 8.50 8.07 7.93 8.60 222750_s_at SRD5A3 8.62 8.16 8.52 8.81 1557212_at FLJ33065 6.94 6.50 6.74 7.02 205499_at SRPX2 8.15 7.69 7.84 8.25 213764_s_at MFAP5 9.85 9.42 9.15 10.37 201989_s_at CREBL2 8.60 8.14 8.20 8.56 200904_at HLA-E 8.57 8.14 8.11 8.56 201186_at LRPAP1 8.93 8.47 8.75 8.93 212359_s_at KIAA0913 8.14 7.71 7.79 8.25 218434_s_at AACS 9.47 9.01 8.98 9.46 212863_x_at CTBP1 10.98 10.55 11.02 11.01 213401_s_at --- 4.82 4.36 4.65 4.81 203528_at SEMA4D 5.18 4.75 5.37 5.38 213462_at NPAS2 6.53 6.08 6.73 6.57 205833_s_at PART1 6.12 5.69 5.57 6.27 200920_s_at BTG1 9.82 9.36 9.43 10.01 213765_at MFAP5 9.84 9.41 9.16 10.45 210608_s_at FUT2 6.52 6.06 6.24 6.72 213275_x_at CTSB 9.86 9.43 9.72 10.12 220274_at IQCA1 6.20 5.74 5.84 6.32 234305_s_at GSDMC 8.46 8.03 8.27 8.76 232968_at FANK1 4.82 4.36 4.51 4.76 1554895_a_at RHBDL2 6.39 5.97 5.95 6.44 207922_s_at MAEA 9.80 9.35 9.79 9.83 200759_x_at NFE2L1 10.19 9.77 9.79 10.18 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

185

p<0.05 1.3-fold Normalized Fluorescence Intensity Normalized Fluorescence Intensity NIKS1 Probe ID Gene ID NIKS NIKS16 8 NIKS16ΔE7 Probe ID Gene ID NIKS NIKS16 NIKS18 NIKS16ΔE7 224836_at TP53INP2 7.29 6.86 6.95 7.77 227252_at LRP10 7.72 7.34 7.23 7.72 200606_at DSP 13.20 12.78 12.78 13.27 214246_x_at MINK1 10.34 9.96 10.19 10.58 210616_s_at SEC31A 10.30 9.88 10.18 10.43 238513_at PRRG4 10.66 10.27 10.46 10.80 201828_x_at FAM127A 9.88 9.45 9.25 10.01 218634_at PHLDA3 8.92 8.53 8.61 8.95 212112_s_at STX12 9.20 8.77 8.72 9.21 239448_at --- 4.16 3.78 3.88 4.21 209424_s_at AMACR /// C1QTNF3 6.98 6.56 6.60 7.09 242722_at LMO7 6.34 5.96 5.84 6.53 209310_s_at CASP4 8.26 7.83 8.11 8.39 228951_at SLC38A7 7.46 7.08 6.93 7.49 205627_at CDA 9.83 9.41 9.23 10.07 230188_at NIPAL4 8.88 8.49 8.87 8.88 204062_s_at ULK2 6.83 6.41 6.11 6.96 218319_at PELI1 7.86 7.48 7.82 8.01 212717_at PLEKHM1 8.73 8.31 8.15 8.89 218939_at LETM1 7.78 7.39 7.66 7.78 205159_at CSF2RB 4.03 3.61 3.74 4.79 222392_x_at PERP 13.59 13.21 13.11 13.73 227203_at FBXL17 7.33 6.91 7.06 7.35 218373_at AKTIP 8.56 8.17 8.29 8.73 1569144_a_at C9orf169 8.20 7.78 8.36 8.33 1553212_at KRT78 5.57 5.18 5.13 6.05 213405_at RAB22A 8.77 8.35 8.45 8.76 227274_at SYNJ2BP 8.54 8.16 8.16 8.66 223540_at PVRL4 7.18 6.76 6.91 7.71 226049_at ERC1 8.43 8.04 8.22 8.55 207291_at PRRG4 8.76 8.34 8.46 8.95 207394_at ZNF137 3.98 3.60 3.95 4.38 223010_s_at OCIAD1 10.84 10.42 10.67 10.92 240861_at --- 5.74 5.35 5.41 5.93 222670_s_at MAFB 8.15 7.73 7.47 9.18 225752_at NIPA1 6.68 6.30 6.61 6.98 227962_at ACOX1 8.39 7.98 7.92 8.55 225160_x_at MDM2 8.49 8.10 8.52 8.50 233586_s_at KLK12 6.98 6.56 6.44 8.50 230504_at CEACAM19 7.17 6.79 6.84 7.27 209216_at WDR45 8.86 8.44 8.55 8.97 223894_s_at AKTIP 7.86 7.48 7.73 8.07 228423_at MAP9 7.96 7.55 7.74 8.04 211034_s_at C12orf51 7.17 6.79 6.90 7.46 223072_s_at INO80B /// WBP1 7.99 7.57 7.67 8.05 227185_at LOC643988 7.18 6.77 7.34 7.79 230454_at ICA1L 5.26 4.84 5.03 5.24 226297_at --- 10.90 10.48 10.65 11.07 203392_s_at CTBP1 10.64 10.23 10.74 10.70 201296_s_at WSB1 10.23 9.82 9.91 10.34 1563805_a_at FAM83C 6.69 6.28 6.91 6.73 221808_at RAB9A 9.73 9.32 9.38 9.93 218186_at RAB25 10.00 9.59 9.78 10.30 1553770_a_at SLAMF9 6.90 6.49 6.62 7.07 1556123_a_at --- 6.01 5.59 5.73 6.08 201015_s_at JUP 11.78 11.37 11.20 11.99 205574_x_at BMP1 7.90 7.49 7.35 7.94 210716_s_at CLIP1 7.49 7.09 6.81 7.55 1557241_a_at --- 5.62 5.22 5.12 5.83 227594_at ZMYM6 7.00 6.59 6.59 7.01 205158_at RNASE4 4.80 4.39 4.56 4.80 235095_at CCDC64B 7.42 7.02 7.09 7.73 209018_s_at PINK1 8.70 8.30 8.07 9.09 225258_at FBLIM1 10.22 9.82 10.16 10.21 208774_at CSNK1D 8.68 8.28 8.47 8.84 220291_at GDPD2 8.30 7.90 7.76 8.58 224329_s_at CNFN 6.68 6.28 6.40 8.26 224990_at C4orf34 6.65 6.26 6.40 7.04 227461_at STON2 9.08 8.68 8.84 9.17 219104_at RNF141 8.46 8.07 8.10 8.82 212546_s_at FRYL 8.21 7.81 7.94 8.25 211423_s_at SC5DL 9.58 9.19 9.24 9.66 1554608_at TGOLN2 4.60 4.20 4.31 4.64 232322_x_at STARD10 8.64 8.24 8.30 9.00 202607_at NDST1 8.99 8.59 8.63 9.10 33646_g_at GM2A 8.26 7.86 8.07 8.32 1559827_at LOC401074 7.53 7.14 6.46 7.63 215395_x_at PRSS1 /// TRY6 7.84 7.44 7.63 8.01 227241_at MUC15 3.51 3.11 3.15 3.87 202389_s_at HTT 8.00 7.61 7.86 8.12 230131_x_at ARSD 6.53 6.13 6.19 6.69 1557342_a_at LOC400931 6.35 5.95 6.16 6.44 226024_at COMMD1 8.65 8.26 8.22 8.87 218531_at TMEM134 7.84 7.45 7.33 7.92 228275_at --- 6.85 6.46 6.22 6.89 241782_at NEBL 3.82 3.43 3.37 4.24 212070_at GPR56 10.04 9.65 9.67 10.17 Appendix E-5B. (continued) E7 Dependent Genes Significantly Downregulated (p<0.05, >1.3 fold)

186

p<0.01

# Entities # Reactions Pathway identifier Pathway name Entities FDR Submitted entities found found found

RAD1;MCM10;MCM4;RAD1;RPA3;MCM7;RFC5;CDC25C;RFC2;CDC25C;CDK2;CDC25C;RFC4;MCM2; Activation of ATR in response to R-HSA-176187 16 1.29E-07 7 RFC3;DBF4;CDC7;RFC2;CLSPN;RFC5;RFC4;RFC3;CLSPN;DBF4;MCM2;MCM10;MCM4;CDC7;MCM7;R replication stress PA3;CDK2

KIF23;KIF2C;TPR;ESCO2;CCND2;SKP2;TP53;BRCA2;KIF2C;RAD51C;CDC25C;CENPF;RAD51C;CENPJ;D BF4;SKA1;CCND2;PLK1;MRE11A;GTSE1;TPR;HAUS7;HAUS5;RUVBL1;HAUS6;RUVBL2;E2F5;CENPF;M CM10;NCAPG2;SPC25;CCNH;BLM;CCNE2;BRCA2;SMC2;LIN9;SMC4;TMPO;HAUS6;RBL1;HAUS5;CDC 25C;HAUS7;PKMYT1;KNTC1;CDC7;POLD1;SPC25;PKMYT1;SKP2;TUBGCP3;MRE11A;MCM2;CCNE2;T ERT;SKA1;MCM4;KIF23;MCM7;SKA1;E2F5;C13orf34;CCNH;PLK1;NUP205;CDKN2A;SMC2;SMC4;NUP R-HSA-1640170 Cell Cycle 65 3.29E-06 233 205;RPA3;MRE11A;CDKN2A;KNTC1;RUVBL2;RUVBL1;TP53;KNTC1;RFC2;CLSPN;RFC5;RFC4;RFC3;C DCA8;CLSPN;TERT;ESCO2;TFDP2;ERCC6L;CDC7;CCNA1;KIF20A;PLK1;NCAPG2;TMPO;TUBGCP3;RAD 1;FEN1;KIF20A;GTSE1;TFDP2;CENPF;NUF2;RBL1;TOPBP1;CENPJ;MCM10;MCM4;POLD1;TOPBP1;RA D1;MCM7;BLM;RFC5;NUF2;RFC2;CDK2;CDKN2A;RFC4;CDC25C;RFC3;MCM2;ERCC6L;LIN9;FEN1;CD CA8;DBF4;CCNA1;RPA3;CDK2

KIF23;TPR;KIF2C;ESCO2;CCND2;TP53;SKP2;KIF2C;CDC25C;CENPF;CENPJ;DBF4;SKA1;CCND2;PLK1;G TSE1;HAUS7;TPR;HAUS5;HAUS6;E2F5;CENPF;MCM10;NCAPG2;SPC25;CCNH;CCNE2;SMC2;LIN9;SM C4;TMPO;HAUS6;RBL1;HAUS5;CDC25C;HAUS7;PKMYT1;KNTC1;POLD1;SPC25;CDC7;PKMYT1;TUBG CP3;SKP2;CCNE2;MCM2;SKA1;MCM4;KIF23;MCM7;SKA1;E2F5;C13orf34;CCNH;PLK1;NUP205;CDKN R-HSA-69278 Cell Cycle, Mitotic 55 1.50E-05 183 2A;SMC2;SMC4;NUP205;RPA3;CDKN2A;KNTC1;TP53;RFC2;KNTC1;RFC5;RFC4;RFC3;CDCA8;ESCO2; TFDP2;ERCC6L;CDC7;CCNA1;KIF20A;PLK1;NCAPG2;TMPO;TUBGCP3;KIF20A;FEN1;GTSE1;RBL1;NUF 2;CENPF;TFDP2;CENPJ;MCM10;MCM4;POLD1;MCM7;NUF2;RFC5;RFC2;CDK2;CDKN2A;CDC25C;RFC 4;RFC3;MCM2;ERCC6L;LIN9;FEN1;CDCA8;DBF4;CCNA1;CDK2;RPA3

Homologous DNA Pairing and Strand RAD1;BLM;TOPBP1;RAD51AP1;BRCA2;RPA3;TOPBP1;RAD1;BRCA2;MRE11A;RAD51C;BLM;RFC5;RF R-HSA-5693579 13 7.19E-05 7 Exchange C2;RFC4;RFC3;RAD51C;MRE11A;RAD51AP1;RFC2;RFC5;RFC4;RFC3;MRE11A;RPA3

POLD1;FEN1;RFC2;FEN1;POLD1;RFC5;MCM4;RFC4;RPA3;RFC3;MCM7;MCM2;RFC5;MCM4;RFC2;M R-HSA-69190 DNA strand elongation 11 4.65E-04 14 CM7;RFC4;MCM2;RFC3;RPA3

CCNA1;RAD1;FEN1;BLM;TOPBP1;RAD51AP1;POLD1;BRCA2;TOPBP1;RAD1;RPA3;BRCA2;MRE11A;R R-HSA-5693538 Homology Directed Repair 20 6.96E-04 44 AD51C;BLM;RFC5;CDK2;RFC2;RFC4;POLQ;RFC3;RAD51C;POLD1;MRE11A;RFC2;RAD51AP1;FEN1;CL SPN;RFC5;RFC4;RFC3;POLQ;CLSPN;MRE11A;CCNA1;RPA3;CDK2

Presynaptic phase of homologous DNA RAD51C;RAD1;BLM;TOPBP1;MRE11A;RFC2;RFC5;BRCA2;RFC4;RAD1;RPA3;TOPBP1;RFC3;MRE11A; R-HSA-5693616 11 6.96E-04 4 pairing and strand exchange BRCA2;BLM;RAD51C;MRE11A;RFC5;RFC2;RFC4;RFC3;RPA3

RAD1;GTSE1;BLM;TOPBP1;MCM10;MCM4;TP53;RAD1;TOPBP1;RPA3;MCM7;MRE11A;BLM;RFC5;C DC25C;CDK2;CDC25C;RFC2;PKMYT1;CDC25C;RFC4;RFC3;MCM2;DBF4;CDC7;MRE11A;TP53;PKMYT R-HSA-69481 G2/M Checkpoints 22 6.96E-04 18 1;RFC2;CLSPN;RFC5;RFC4;RFC3;GTSE1;CLSPN;DBF4;MRE11A;MCM2;MCM10;MCM4;CDC7;MCM7; RPA3;CDK2

HDR through Homologous RAD1;BLM;TOPBP1;RAD51AP1;POLD1;BRCA2;RPA3;TOPBP1;RAD1;BRCA2;MRE11A;BLM;RAD51C; R-HSA-5685942 14 9.08E-04 16 Recombination (HRR) RFC5;RFC2;RFC4;RFC3;RAD51C;POLD1;MRE11A;RAD51AP1;RFC2;RFC5;RFC4;RFC3;MRE11A;RPA3

CCNA1;RAD1;GTSE1;BLM;TOPBP1;CCNE2;MCM10;MCM4;TP53;RAD1;TOPBP1;RPA3;MCM7;MRE11 A;BLM;RFC5;CDC25C;CDK2;CDC25C;RFC2;PKMYT1;CDC25C;RFC4;MCM2;RFC3;DBF4;CDC7;MRE11 R-HSA-69620 Cell Cycle Checkpoints 24 1.39E-03 29 A;TP53;PKMYT1;RFC2;CLSPN;RFC5;RFC4;RFC3;GTSE1;CLSPN;DBF4;MRE11A;MCM2;CCNE2;MCM10; MCM4;CDC7;CCNA1;MCM7;RPA3;CDK2

Appendix E-6A. E7 Dependent Pathways Significantly Upregulated (p<0.01)

187

p<0.01

# Entities # Reactions Pathway identifier Pathway name Entities FDR Submitted entities found found found

CCNA1;E2F5;CCNH;TFDP2;CCNH;RBL1;CCNE2;CDKN2A;MCM10;CCND2;MCM4;SKP2;RPA3;MCM7;L R-HSA-453279 Mitotic G1-G1/S phases 20 1.47E-03 47 IN9;RBL1;CDKN2A;CDK2;PKMYT1;CDKN2A;MCM2;DBF4;CDC7;CCND2;LIN9;PKMYT1;SKP2;DBF4;E2F 5;CCNE2;MCM2;MCM10;TFDP2;MCM4;CDC7;CCNA1;MCM7;RPA3;CDK2

HDR through Single Strand Annealing RAD1;BLM;TOPBP1;MRE11A;RFC2;RFC5;RFC4;RPA3;TOPBP1;RFC3;RAD1;MRE11A;BLM;MRE11A;R R-HSA-5685938 10 1.66E-03 4 (SSA) FC5;RFC2;RFC4;RFC3;RPA3

HDR through Homologous CCNA1;RAD1;BLM;TOPBP1;RAD51AP1;POLD1;BRCA2;TOPBP1;RAD1;RPA3;BRCA2;MRE11A;RAD51 R-HSA-5693567 Recombination (HR) or Single Strand 18 1.92E-03 36 C;BLM;RFC5;CDK2;RFC2;RFC4;RFC3;RAD51C;POLD1;MRE11A;RAD51AP1;RFC2;CLSPN;RFC5;RFC4;R Annealing (SSA) FC3;CLSPN;MRE11A;CCNA1;RPA3;CDK2

POLD1;FEN1;RUVBL2;RUVBL1;RFC2;FEN1;POLD1;RFC5;RFC4;RPA3;RFC3;RUVBL1;TERT;RUVBL2;RF R-HSA-180786 Extension of Telomeres 10 1.92E-03 19 C5;TERT;RFC2;RFC4;RFC3;RPA3

Activation of the pre-replicative DBF4;CDC7;MCM10;MCM4;RPA3;MCM7;DBF4;MCM2;MCM10;MCM4;CDK2;CDC7;MCM7;MCM2;R R-HSA-68962 9 1.92E-03 8 complex PA3;CDK2

CCNA1;RAD1;FEN1;BLM;TOPBP1;RAD51AP1;TP53;POLD1;BRCA2;RAD1;TOPBP1;RPA3;BRCA2;MRE R-HSA-5693532 DNA Double-Strand Break Repair 21 2.76E-03 84 11A;RAD51C;BLM;RFC5;CDK2;RFC2;RFC4;POLQ;RFC3;RAD51C;POLD1;MRE11A;TP53;RFC2;RAD51 AP1;FEN1;CLSPN;RFC5;RFC4;RFC3;POLQ;CLSPN;MRE11A;CCNA1;RPA3;CDK2

PCNA-Dependent Long Patch Base R-HSA-5651801 7 3.71E-03 6 POLD1;FEN1;RFC2;FEN1;POLD1;RFC5;RFC4;RFC3;RPA3;RFC5;RFC2;RFC4;RPA3;RFC3 Excision Repair

R-HSA-156711 Polo-like kinase mediated events 7 6.04E-03 13 PLK1;PLK1;CENPF;LIN9;CENPF;CDC25C;LIN9;PLK1;CENPF;CDC25C;PKMYT1;PKMYT1;CDC25C

Regulation of TP53 Activity through CCNA1;TAF4B;RAD1;BLM;TOPBP1;TAF4B;TP53;RPA3;TOPBP1;RAD1;MRE11A;BLM;RFC5;CDK2;RFC R-HSA-6804756 14 7.25E-03 24 Phosphorylation 2;RFC4;RFC3;SUPT16H;SUPT16H;MRE11A;TP53;RFC2;RFC5;RFC4;RFC3;MRE11A;CCNA1;RPA3;CDK2

CCNA1;FEN1;MCM10;MCM4;POLD1;RPA3;MCM7;RFC5;RFC2;CDK2;RFC4;MCM2;RFC3;DBF4;POLD R-HSA-69306 DNA Replication 16 8.33E-03 29 1;CDC7;RFC2;FEN1;RFC5;RFC4;RFC3;DBF4;MCM2;MCM10;MCM4;CDC7;CCNA1;MCM7;RPA3;CDK2

Appendix E-6A. (continued) E7 Dependent Pathways Significantly Upregulated (p<0.01)

188 p<0.01 # Entities # Reactions Pathway identifier Pathway name Entities FDR Submitted entities found found found Endosomal/Vacuolar R-HSA-1236977 51 1.94E-14 4 HLA-C;HLA-B;CTSL2;HLA-C;HLA-E;HLA-B;HLA-E pathway Antigen Presentation: Folding, assembly and SEC31A;HLA-C;HLA-B;HLA-C;HLA-E;HLA- R-HSA-983170 51 1.94E-14 12 peptide loading of class I B;SEC31A;SEC31A;HLA-E MHC Antigen processing-Cross HLA-C;ITGAV;HLA-B;CTSL2;HLA-C;HLA-E;HLA-B;HLA- R-HSA-1236975 53 1.94E-14 9 presentation E;NCF2;ITGAV;NCF2 R-HSA-1236974 ER-Phagosome pathway 50 1.94E-14 2 HLA-C;HLA-B;HLA-C;HLA-E;HLA-B;HLA-E HLA-C;HLA-B;CAMK2D;HLA-C;HLA-E;TRIM2;HLA- R-HSA-877300 Interferon gamma signaling 53 1.94E-14 2 B;CAMK2D;TRIM2;HLA-E;SP140L Interferon alpha/beta HLA-C;HLA-B;HLA-C;HLA-E;IFI27;HLA-B;HLA- R-HSA-909733 52 1.94E-14 1 signaling E;IFI27;RNASEL;RNASEL HLA-C;HLA-B;CAMK2D;HLA- R-HSA-913531 Interferon Signaling 55 4.19E-10 3 E;TRIM2;RNASEL;RNASEL;SP140L;HLA- C;CAMK2D;TRIM2;IFI27;HLA-B;HLA-E;IFI27 Immunoregulatory CLEC2B;CDH1;HLA-C;CDH1;HLA-B;COL1A1;CDH1;HLA- interactions between a R-HSA-198933 57 5.54E-07 16 E;SLAMF7;SLAMF7;C3;MAL;COL17A1;C3;HLA-C;HLA-B;HLA- Lymphoid and a non- E;COL17A1;COL1A1;CLEC2B Lymphoid cell Class I MHC mediated SEC31A;HLA-C;CDH1;HLA-B;HLA- R-HSA-983169 antigen processing & 58 3.02E-06 25 E;SEC31A;WSB1;NCF2;WSB1;NCF2;ITGAV;CTSL2;HLA-C;HLA- presentation B;FBXO3;SEC31A;HLA-E;HECTD2;FBXO3;ITGAV;HECTD2 CDH1;CDH1;COL1A1;CAPN3;PSEN1;COL4A5;COL4A5;LAMB3; Degradation of the ADAMTS1;COL17A1;KLK7;BMP1;CTSB;CAST;CDH1;CTSB;BMP1 R-HSA-1474228 26 3.45E-04 80 extracellular matrix ;PRSS3;PRSS2;COL17A1;CAPN3;MMP1;PRSS1;PSEN1;KLK7;CTS L2;ADAMTS1;COL1A1;PRSS1;CAST;LAMB3;MMP1;PRSS2 Activation of Matrix R-HSA-1592389 11 1.63E-03 24 CTSL2;KLK7;PRSS3;PRSS2;PRSS1;MMP1;PRSS1;MMP1;PRSS2 Metalloproteinases COL1A1;CTSB;PRSS3;PRSS2;COL17A1;COL4A5;COL4A5;MMP1 R-HSA-1442490 Collagen degradation 15 4.15E-03 24 ;PRSS1;CTSL2;COL17A1;KLK7;CTSB;COL1A1;MMP1;PRSS2 Appendix E-6B. E7 Dependent Pathways Significantly Downregulated (p<0.01)

189

Gene Area p-value Chr Start End Probes MIR615 8.50E-05 12 54427293 54427884 6 LTB4R2 2.60E-04 14 24780682 24780926 6 QPCT 4.10E-04 2 37571687 37571732 4 TNF 4.30E-04 6 31543638 31543686 4 HLA-E 4.50E-04 6 30434406 30434421 3 SALL1 5.10E-04 16 51183533 51184152 3 RASGRP2 5.30E-04 11 64511786 64512032 3 ZIC4 6.00E-04 3 147125758 147125926 5 MEI1 7.90E-04 22 42095441 42095536 3 PCDH10 8.10E-04 4 134069670 134070235 3 ZSCAN5A 8.30E-04 19 56879554 56879571 3 ARHGAP6 1.00E-03 X 11157158 11157323 2 MGMT 1.10E-03 10 131214543 131214585 3 TRPC5 1.20E-03 X 111325092 111325176 3 SLC15A1 1.30E-03 13 99405102 99405124 3 PROM1 1.40E-03 4 16085401 16085606 3 MACROD1 1.40E-03 11 63828713 63828762 2 GM140 1.50E-03 1 181287301 181287636 3 NR2E1 2.10E-03 6 108495678 108495985 3 MSRB2 2.20E-03 10 23385780 23385979 2 AKAP8L 2.60E-03 19 15530737 15530870 2 LOC401242 2.70E-03 6 28832567 28832571 2 ARX 2.70E-03 X 25029360 25029629 2 PTPRO 3.20E-03 12 15475231 15475236 2 FAM110A 3.40E-03 20 825408 825489 3 DDX25 3.50E-03 11 125774311 125774447 3 RUNX3 3.70E-03 1 25257587 25257624 3 DDX25 4.10E-03 11 125774082 125774090 2 LINC01164 4.20E-03 10 133507998 133508039 3 HLA-E 4.50E-03 6 30434523 30434525 2 STK32C 4.60E-03 10 134043618 134043635 2 L3MBTL4 4.70E-03 18 6414974 6414978 3 RAB9B 4.80E-03 X 103087380 103087808 2 LINC00862 4.90E-03 1 200271342 200271455 2 PFDN6 5.20E-03 6 33258129 33258248 3 BCOR 5.80E-03 X 39866890 39867034 2 ARL14EP 6.10E-03 11 30344373 30344399 2 GRAMD1B 6.20E-03 11 123495493 123495671 2 GCA 6.20E-03 2 163200302 163200405 2 MTRNR2L1 6.40E-03 17 22020759 22020806 2 APCS 6.60E-03 1 159556997 159557257 2 ADRA2A 7.30E-03 10 112838983 112839302 2 ACADL 7.40E-03 2 211090323 211090349 2 LIN28A 7.60E-03 1 26737318 26737470 2 TMEM192 7.80E-03 4 165980112 165980278 2 KIRREL2 7.90E-03 19 36347441 36347443 2 OCA2 7.90E-03 15 28341714 28341783 2 ZNF702P 7.90E-03 19 53496695 53496738 2 BDKRB2 8.00E-03 14 96670628 96670634 2 ZNF131 8.00E-03 5 43191975 43191993 2 KIF26A 8.10E-03 14 104690151 104690201 2 MIR3679 8.20E-03 2 134785497 134785688 2 ATP6AP1 8.40E-03 X 153664903 153664905 2 SNORD114-31 9.10E-03 14 101459254 101459547 2 SMOC2 9.20E-03 6 169273007 169273049 2 OR1A1 9.50E-03 17 3117953 3117981 2 Appendix E-7A. Significantly Hypermethylated DMRs (NIKS-16 to NIKS, area p < 0.01)

190

Gene Area p-value Chr Start End Probes HOXD10 4.30E-05 2 176980837 176981469 7 GJD4 1.50E-04 10 35894222 35894430 5 GATA1 6.80E-04 X 48645024 48645037 2 RP11-40F8.2 9.40E-04 X 23017656 23017887 2 ZNF157 1.50E-03 X 47229908 47230278 3 SYTL3 1.50E-03 6 159070959 159071008 3 RPS6KA2 1.60E-03 6 167284353 167284983 4 GSTCD 1.60E-03 4 106631906 106631934 3 ERICH1-AS1 1.90E-03 8 1108682 1108732 2 LIN7A 2.40E-03 12 81332540 81332744 2 C17orf99 2.50E-03 17 76141038 76141249 3 TRH 2.60E-03 3 129692887 129692899 2 LHFPL1 2.90E-03 X 111924209 111924370 2 SERPING1 3.00E-03 11 57365701 57365725 3 FABP3 3.50E-03 1 31845904 31845959 4 C6orf118 4.00E-03 6 165715586 165715592 2 ACP5 4.10E-03 19 11689726 11689855 3 DOK6 4.20E-03 18 67070170 67070324 2 TKTL1 4.30E-03 X 153560525 153560828 2 MAGEB4 4.40E-03 X 30259099 30259354 2 SLITRK5 4.40E-03 13 88326244 88326752 3 MGMT 4.70E-03 10 131568811 131569018 2 RAB9B 5.10E-03 X 103088089 103088500 2 P2RY14 5.10E-03 3 150996078 150996184 2 PAX9 5.50E-03 14 37123778 37124018 2 VN1R4 5.60E-03 19 53771286 53771332 2 NIP7 5.80E-03 16 69374664 69374895 2 TRIM40 6.10E-03 6 30113786 30113877 3 CDK5R2 6.80E-03 2 219825837 219826145 2 NSUN7 6.90E-03 4 40751717 40751843 3 CD1D 7.10E-03 1 158151058 158151363 3 C6orf123 7.10E-03 6 168147329 168147420 3 RP11-80F22.15 7.20E-03 16 34808939 34809256 2 GLUD2 7.30E-03 X 120182714 120183084 2 LOC729041 7.60E-03 1 46913532 46914023 3 MGLL 7.70E-03 3 127542386 127542388 2 OR4D1 7.80E-03 17 56232302 56232357 2 GLYATL1 8.10E-03 11 58710158 58710266 2 ATP4B 8.40E-03 13 114305448 114305746 2 CHKB-CPT1B 8.60E-03 22 51017162 51017432 3 ADRA2C 8.60E-03 4 3775787 3775898 2 C17orf58 8.90E-03 17 65990429 65990431 2 CCND2 8.90E-03 12 4381788 4381800 3 ABCB7 9.30E-03 X 74376811 74376828 2 DOC2B 9.40E-03 17 5999 6018 2 LOC286083 9.60E-03 8 1309967 1310046 2 DYDC2 9.60E-03 10 82116203 82116208 2 Appendix E-7B. Significantly Hypomethylated DMRs (NIKS-16 to NIKS, area p < 0.01)

191

Gene expression Methylation Gene ID Probe ID Fold Change (log2) FDR-adjusted p Probe ID Chage i β NIKS ea β FDR-adjusted p ABAT 209459_s_at 1.47 6.9E-03 cg07439546 29.90% 61.50% 1.9E-03 ABCC3 239217_x_at 0.43 6.9E-03 cg26623266 -27.50% 80.60% 4.3E-03 ACP6 218795_at -0.81 5.9E-03 cg27244432 -11.40% 18.20% 2.7E-03 ACSL1 201963_at 1.07 5.2E-03 cg27571769 -40.80% 87.50% 4.5E-03 ACSL1 207275_s_at 0.85 9.8E-03 cg27571769 -40.80% 87.50% 4.5E-03 ADAR 201786_s_at -0.14 8.0E-03 cg11650704 25.10% 63.10% 2.2E-03 ANAPC5 211036_x_at -0.15 7.2E-03 cg15417488 26.50% 66.20% 4.5E-03 APLP2 208703_s_at 0.37 4.4E-03 cg24860589 -16.50% 92.60% 3.6E-03 APLP2 211404_s_at 0.30 4.6E-03 cg24860589 -16.50% 92.60% 3.6E-03 APOE 203381_s_at 0.73 4.4E-03 cg16471933 27.20% 59.90% 3.3E-03 ASPHD1 1553997_a_at -0.65 8.3E-03 cg02488299 34.80% 20.60% 4.3E-03 B4GALT7 222191_s_at 0.19 8.5E-03 cg03493123 56.50% 30.60% 2.0E-03 BAK1 203728_at 0.28 8.3E-03 cg10531711 21.30% 67.90% 4.5E-03 BTBD11 228570_at 0.65 6.7E-03 cg01797169 35.80% 15.50% 2.3E-03 BTBD11 238692_at 0.71 9.1E-03 cg01797169 35.80% 15.50% 2.3E-03 C10orf57 218174_s_at 0.42 8.7E-03 cg03909471 -14.30% 33.80% 2.5E-03 C15orf41 232506_s_at -0.19 8.2E-03 cg26942580 -37.30% 59.10% 3.4E-03 C16orf80 217957_at -0.27 5.6E-03 cg09408116 -56.90% 75.90% 4.7E-03 C3orf14 219288_at -0.73 5.4E-03 cg01181009 41.70% 18.20% 4.6E-03 CCND2 200953_s_at -0.28 8.0E-03 cg15249639 -44.70% 53.20% 1.3E-03 CD3EAP 205264_at -0.34 7.2E-03 cg23347323 18.70% 59.60% 4.4E-03 CENPV 226611_s_at -1.30 5.7E-03 cg18185665 36.00% 21.70% 3.9E-03 CENPV 226610_at -1.06 8.3E-03 cg18185665 36.00% 21.70% 3.9E-03 CKAP4 200998_s_at 0.35 8.3E-03 cg26477549 -41.80% 61.00% 3.2E-03 CLN5 204085_s_at 0.52 3.2E-03 cg18817318 -40.00% 49.80% 2.5E-03 CPEB4 224831_at 0.74 7.2E-03 cg25122699 -36.60% 47.30% 4.0E-03 CRNKL1 219913_s_at -0.43 6.1E-03 cg06039489 61.10% 24.60% 3.5E-03 CTBP1 203392_s_at 0.41 3.1E-03 cg02384251 26.20% 66.30% 3.3E-03 CYP27C1 1568868_at 1.50 8.1E-03 cg04946274 -14.50% 76.80% 4.7E-03 DAZAP1 229813_x_at -0.24 6.1E-03 cg00716812 31.50% 58.10% 1.9E-03 DDAH1 209094_at -2.53 4.4E-03 cg14985076 -29.30% 58.10% 3.4E-03 DENND3 212974_at -0.38 8.7E-03 cg16494567 -17.00% 88.50% 3.3E-03 DGCR2 214198_s_at 0.33 4.4E-03 cg16621987 -34.80% 46.40% 1.6E-03 DLL3 219537_x_at -0.63 9.1E-03 cg01519877 29.10% 20.20% 2.4E-03 DRG1 202810_at -0.38 8.4E-03 cg01809797 48.70% 11.10% 4.4E-03 DYNLT3 203303_at 0.85 4.1E-03 cg20119635 78.00% 8.90% 4.8E-03 EIF3K 221494_x_at -0.26 6.6E-03 cg04292549 24.90% 28.20% 2.6E-03 EIF3K 210501_x_at -0.27 7.1E-03 cg04292549 24.90% 28.20% 2.6E-03 ELL2 214446_at 0.27 9.8E-03 cg19493250 35.60% 60.30% 2.5E-03 EPS8L1 91826_at 0.54 6.0E-03 cg00462362 32.70% 59.60% 2.6E-03 EPSTI1 227609_at -1.36 8.2E-03 cg16327891 19.60% 70.90% 1.9E-03 ESPL1 38158_at -0.58 6.7E-03 cg25696949 -29.90% 94.00% 4.7E-03 FAM155B 206299_at -0.96 6.8E-03 cg22650104 58.20% 20.50% 1.8E-03 FAM171A1 212771_at -0.70 4.6E-03 cg13905757 -28.70% 57.90% 3.2E-03 FRMD4A 225163_at 0.26 8.9E-03 cg07629776 -51.50% 59.40% 1.5E-03 GNB5 207124_s_at -0.34 7.4E-03 cg22134372 -21.00% 94.20% 3.1E-03 GPR115 1553031_at 1.31 8.0E-03 cg10844275 -45.70% 82.00% 2.0E-03 GPR115 237690_at 1.34 8.5E-03 cg10844275 -45.70% 82.00% 2.0E-03 GPR137B 204137_at 0.85 9.4E-03 cg07396581 -29.30% 94.60% 1.9E-03 HLA-E 200904_at 0.43 4.4E-03 cg15816267 11.40% 12.40% 3.0E-03 HLA-E 217456_x_at 0.40 7.4E-03 cg15816267 11.40% 12.40% 3.0E-03 HOXD10 229400_at 1.21 5.4E-03 cg03918304 -22.50% 33.20% 5.0E-03 HS3ST1 205466_s_at 0.91 9.9E-03 cg09997532 -36.90% 94.70% 2.1E-03 INPP4B 235046_at 0.70 9.2E-03 cg00183107 20.00% 8.90% 2.5E-03 ITPK1 210197_at -0.53 6.8E-03 cg04983685 32.10% 10.70% 2.0E-03 JARID2 203298_s_at 0.38 6.5E-03 cg14682005 35.30% 31.10% 3.2E-03 JTB 210434_x_at -0.29 8.7E-03 cg22332037 15.70% 77.90% 1.9E-03 JTB 200048_s_at -0.32 8.9E-03 cg22332037 15.70% 77.90% 1.9E-03 KATNB1 203162_s_at -0.32 5.5E-03 cg02534657 17.40% 56.70% 3.4E-03 KLK11 205470_s_at 1.99 3.8E-03 cg09923953 7.90% 84.00% 4.2E-03 KLRG2 244264_at -1.48 6.9E-03 cg03537942 -42.80% 48.00% 4.9E-03 KRT1 205900_at 1.54 8.3E-03 cg03348792 32.90% 59.30% 1.9E-03 Appendix E-8. Combined Transcriptome (p < 0.01) and Methylome DMPs (p < 0.005) for the NIKS-16 to NIKS Comparison

192

Gene expression Methylation Gene ID Probe ID Fold Change (log2) FDR-adjusted p Probe ID Chage i β NIKS ea β FDR-adjusted p LAGE3 219061_s_at -0.75 6.7E-03 cg26581933 -73.50% 77.30% 4.5E-03 LTBP3 219922_s_at -0.76 5.6E-03 cg03101763 -17.00% 46.10% 3.7E-03 MAD2L2 223234_at -0.29 6.2E-03 cg16695380 -40.80% 45.10% 1.7E-03 MAMDC2 228885_at -1.38 8.3E-03 cg14403200 -12.60% 64.50% 2.0E-03 MAP2 225540_at 1.56 3.8E-03 cg17167021 -25.80% 94.90% 4.7E-03 MAP7D1 217943_s_at 0.22 6.7E-03 cg19002149 33.40% 57.00% 2.1E-03 MARK3 202568_s_at 0.29 6.8E-03 cg16295401 20.60% 5.70% 3.0E-03 MBOAT2 213288_at 0.62 6.5E-03 cg25371267 -25.20% 90.40% 4.6E-03 MCM10 223570_at -0.56 9.2E-03 cg17109175 -31.90% 45.00% 4.6E-03 MEST 202016_at -2.74 5.4E-03 cg03588221 -41.80% 57.80% 3.3E-03 MFSD1 218109_s_at 0.59 7.9E-03 cg11632438 -8.00% 16.40% 4.8E-03 MRPL34 221692_s_at -0.41 7.2E-03 cg08702612 -15.20% 33.90% 3.9E-03 MSRB2 218773_s_at 0.47 4.0E-03 cg07834396 78.00% 13.50% 2.4E-03 MYD88 209124_at 0.38 8.5E-03 cg24368167 -3.30% 10.20% 1.6E-03 MYO1B 212365_at 0.15 6.5E-03 cg15096140 -50.80% 65.20% 1.3E-03 NADSYN1 232946_s_at 0.24 9.6E-03 cg25623977 39.70% 10.70% 2.0E-03 NARS 200027_at -0.19 6.4E-03 cg20711496 55.20% 28.60% 2.0E-03 NAV1 224772_at 0.64 3.1E-03 cg13071729 41.00% 3.40% 1.4E-03 NDEL1 208093_s_at 0.20 9.4E-03 cg12173651 -54.90% 62.00% 2.6E-03 NDST1 202607_at 0.40 3.5E-03 cg19836174 -36.50% 46.00% 4.9E-03 NDUFA9 208969_at -0.33 9.7E-03 cg13719771 26.30% 66.10% 4.2E-03 NDUFS2 201966_at -0.13 5.8E-03 cg24049880 28.40% 63.90% 2.8E-03 NHP2 209104_s_at -0.26 7.9E-03 cg05654329 -59.90% 92.20% 2.9E-03 PARP9 223220_s_at 0.33 9.2E-03 cg00959259 26.70% 67.60% 4.2E-03 PDLIM5 212412_at 0.63 9.7E-03 cg18009021 15.00% 62.10% 1.8E-03 PLEKHA7 228450_at -0.62 5.4E-03 cg15084286 -10.10% 18.60% 3.2E-03 PLEKHG5 227142_at 0.55 9.9E-03 cg17329304 27.90% 34.50% 1.9E-03 PLXDC2 227276_at -2.47 2.7E-03 cg24022375 -10.00% 14.70% 1.7E-03 PSMG4 228217_s_at -0.43 8.7E-03 cg04767753 41.20% 21.40% 2.8E-03 PTP4A1 200730_s_at -0.16 4.1E-03 cg14717759 18.80% 57.90% 3.7E-03 RAB27B 228708_at 0.93 3.8E-03 cg16546376 31.20% 59.40% 2.0E-03 RBM22 218134_s_at -0.25 6.6E-03 cg01394116 -56.00% 63.90% 1.9E-03 RBMS3 238447_at 1.36 5.6E-03 cg26082870 53.80% 37.80% 4.7E-03 SART1 200051_at -0.46 5.7E-03 cg01415344 15.90% 64.50% 4.9E-03 SCCPDH 201825_s_at -0.55 9.2E-03 cg26309691 -25.00% 37.40% 3.7E-03 SCLY 59705_at -0.53 9.5E-03 cg04003327 -28.30% 90.80% 3.0E-03 SH3BGRL 201312_s_at 2.97 3.1E-03 cg00715011 71.70% 5.20% 2.2E-03 SLC16A3 202855_s_at -0.78 6.7E-03 cg20531020 -36.80% 73.20% 3.0E-03 SLC16A3 202856_s_at -0.78 7.8E-03 cg20531020 -36.80% 73.20% 3.0E-03 SLC16A3 217691_x_at -0.36 8.3E-03 cg20531020 -36.80% 73.20% 3.0E-03 SLC29A1 201801_s_at -0.63 6.6E-03 cg22330763 9.90% 54.10% 3.0E-03 SLC43A3 210692_s_at -0.83 9.0E-03 cg09197783 25.10% 68.20% 2.6E-03 SLC43A3 213113_s_at -0.82 9.8E-03 cg09197783 25.10% 68.20% 2.6E-03 SMAD5 235451_at -1.57 5.4E-03 cg07162551 -18.20% 66.60% 3.2E-03 SMAD5 205188_s_at -1.33 6.8E-03 cg07162551 -18.20% 66.60% 3.2E-03 SNX10 218404_at -0.59 3.9E-03 cg12965553 51.20% 39.60% 2.4E-03 SUPT16H 217815_at -0.35 6.7E-03 cg11120865 -27.80% 88.60% 2.0E-03 TAF7 201023_at 0.68 5.8E-03 cg15028904 33.40% 4.60% 1.7E-03 TMCC1 213352_at 0.71 9.4E-03 cg06509598 29.20% 61.90% 3.4E-03 TMEM30B 213285_at -0.77 8.3E-03 cg19705215 19.30% 41.30% 4.7E-03 TNFRSF8 206729_at -1.66 6.2E-03 cg00569500 33.60% 54.80% 4.5E-03 TNIK 213107_at -1.69 9.4E-03 cg12354192 13.40% 74.50% 3.3E-03 TRAF3IP3 231932_at 1.81 5.2E-03 cg01997629 28.40% 62.00% 2.5E-03 TRIM44 217759_at -0.37 2.7E-03 cg02760962 40.50% 21.90% 4.0E-03 TRPC1 205802_at -0.66 8.3E-03 cg10503359 -18.90% 79.30% 4.6E-03 TSEN54 225879_at -0.36 3.5E-03 cg06665453 -36.40% 52.90% 2.5E-03 TSPAN13 217979_at -0.82 5.4E-03 cg11384744 -42.80% 68.00% 3.2E-03 VARS2 226200_at -0.62 6.2E-03 cg27057509 -42.80% 66.20% 3.7E-03 ZAK 225665_at 0.36 9.9E-03 cg15136424 -48.70% 54.20% 2.6E-03 ZNF750 219995_s_at 2.46 9.3E-03 cg27354296 47.00% 12.70% 1.7E-03 Appendix E-8. (continued) Combined Transcriptome (p < 0.01) and Methylome DMPs (p < 0.005) for the NIKS-16 to NIKS Comparison

193

Gene expression Methylation Gene ID Probe ID Fold Change (log2) FDR-adjusted p Probe ID Chage i β NIKS ea β FDR-adjusted p U133 Plus 2.0 Infinium 450 Gene Log2 Fold Change p (FDR) Chage i β NIKS ea β p (FDR) Probe ID Probe ID ANK3 206385_s_at -2.98 9.17E-03 cg24760597 36.87% 57.47% 1.22E-03 APOE 203381_s_at -1.09 9.18E-03 cg01032398 -22.96% 62.14% 3.68E-03 ATP5D 213041_s_at 0.15 4.84E-03 cg07294734 -12.17% 10.71% 1.41E-03 BRE 205550_s_at -0.43 9.41E-03 cg12886188 -30.74% 62.37% 9.74E-04 C10orf76 218891_at -0.32 9.41E-03 cg03049303 -8.20% 5.74% 8.73E-04 C3 217767_at -1.01 9.41E-03 cg14279361 -24.78% 54.31% 7.82E-04 CCNA1 205899_at 2.74 9.41E-03 cg05089090 -70.59% 8.62% 5.02E-04 CDKN2A 209644_x_at 2.23 5.62E-04 cg03079681 -17.84% 7.56% 4.41E-03 CDKN2A 207039_at 2.50 6.24E-03 cg03079681 -17.84% 7.56% 4.41E-03 CHST2 203921_at -2.57 9.41E-03 cg17024542 16.16% 72.88% 2.92E-03 COL16A1 204345_at -0.47 9.69E-03 cg09251959 -23.00% 14.00% 2.41E-03 CSNK1D 208774_at -0.56 9.41E-03 cg26103512 -42.59% 6.76% 5.74E-04 CTBP1 203392_s_at -0.48 9.41E-03 cg20743744 32.59% 87.03% 6.56E-04 CTDSPL 201906_s_at 0.73 4.61E-03 cg13333722 -35.67% 59.58% 2.68E-03 CTDSPL 201904_s_at 0.76 9.41E-03 cg13333722 -35.67% 59.58% 2.68E-03 CTSH 202295_s_at 2.78 2.51E-03 cg20059407 11.74% 96.10% 5.73E-04 CYBA 203028_s_at 1.61 9.91E-03 cg19790294 -58.94% 20.97% 7.71E-04 DAB2IP 225020_at 0.45 9.21E-03 cg14582957 -30.41% 62.40% 1.45E-03 DBNL 222429_at 0.43 9.41E-03 cg22785672 37.95% 54.69% 4.44E-04 DDR1 208779_x_at -0.22 9.41E-03 cg00466425 15.75% 36.80% 4.35E-03 DGKQ 226605_at -0.79 9.20E-03 cg04771133 12.78% 91.51% 9.30E-04 DISP2 229579_s_at -0.97 9.41E-03 cg17018786 21.27% 83.51% 1.92E-03 DUOX1 215800_at -0.58 9.41E-03 cg12434875 -29.45% 64.66% 5.41E-04 ELOVL4 219532_at -1.66 9.41E-03 cg05492839 31.60% 33.70% 8.32E-04 ESD 209009_at 0.22 9.41E-03 cg03031868 -13.33% 6.24% 7.49E-04 EVPL 204503_at -0.73 6.24E-03 cg02678768 -47.18% 11.69% 4.07E-03 EXD2 218363_at -0.19 9.07E-03 cg00925766 30.11% 88.79% 2.30E-03 FAM101B 226905_at 0.63 9.41E-03 cg13951796 -28.66% 63.11% 1.41E-03 FAM135A 223497_at -0.76 9.41E-03 cg23194609 -54.11% 6.93% 1.34E-03 GFM1 225158_at 0.90 7.73E-03 cg17481912 -41.43% 13.12% 1.36E-03 GGCT 215380_s_at 0.36 9.21E-03 cg00670698 16.30% 53.76% 8.23E-04 GLTP 226177_at -1.64 9.41E-03 cg20045970 -57.79% 14.32% 2.20E-03 GOLIM4 204324_s_at 0.75 9.41E-03 cg27357134 10.69% 92.84% 2.13E-03 HLA-E 200904_at -0.42 9.41E-03 cg09326440 -24.08% 6.52% 4.83E-04 HPGD 211548_s_at -1.73 9.21E-03 cg04555941 -51.85% 9.17% 4.56E-04

Appendix E-9. Combined Transcriptome (p < 0.01) and Methylome DMPs (p < 0.005) for the NIKS-16ΔE7 to NIKS-16 Comparison

194

Gene expression Methylation Gene ID Probe ID Fold Change (log2) FDR-adjusted p Probe ID Chage i β NIKS ea β FDR-adjusted p U133 Plus 2.0 Infinium 450 Gene Log2 Fold Change p (FDR) Chage i β NIKS ea β p (FDR) Probe ID Probe ID HSPB8 221667_s_at -2.21 6.24E-03 cg16493126 -16.18% 73.89% 2.81E-03 IGSF3 202421_at -0.53 9.41E-03 cg12473570 -7.65% 47.26% 1.90E-03 ITGB5 201125_s_at 0.54 9.41E-03 cg18770216 60.07% 86.07% 5.96E-04 KDM4C 209984_at 0.20 9.53E-03 cg13935079 -43.68% 25.73% 3.46E-03 KIAA0494 201778_s_at -0.24 8.88E-03 cg05638830 20.69% 93.08% 3.11E-03 KLK7 239381_at -2.57 8.88E-03 cg26172504 27.41% 39.03% 3.72E-04 KLK7 205778_at -2.52 8.88E-03 cg26172504 27.41% 39.03% 3.72E-04 KRT1 205900_at -4.20 6.24E-03 cg02425372 -26.09% 54.60% 6.87E-04 KRT10 210633_x_at -1.14 7.37E-03 cg13464004 29.89% 85.94% 4.08E-03 KRT10 207023_x_at -1.43 9.41E-03 cg13464004 29.89% 85.94% 4.08E-03 KRT10 213287_s_at -1.30 9.41E-03 cg13464004 29.89% 85.94% 4.08E-03 KRT6B 213680_at -2.34 9.41E-03 cg02068361 5.22% 15.50% 4.99E-03 KRTDAP 230835_at -4.40 9.84E-03 cg22498784 45.59% 56.72% 1.06E-03 LCN2 212531_at -2.27 7.58E-03 cg22438810 -9.14% 5.82% 7.32E-04 LOC401397 224723_x_at 0.52 7.84E-03 cg02425644 33.31% 62.76% 1.66E-03 MAFK 226206_at 0.59 9.55E-03 cg12359001 -42.59% 22.81% 1.55E-03 MAMDC2 228885_at 2.16 2.90E-03 cg02515233 -24.61% 62.97% 1.64E-03 MAN2B1 209166_s_at -0.77 9.69E-03 cg14132016 -23.80% 11.09% 5.60E-04 MTSS1 203037_s_at -0.83 9.69E-03 cg19523819 24.01% 77.01% 1.28E-03 MYO5C 218966_at 1.29 6.24E-03 cg01140395 -44.48% 33.05% 1.10E-03 NAMPT 217739_s_at 0.45 7.86E-03 cg05004518 -39.47% 34.06% 2.67E-03 NDRG3 224368_s_at -0.41 9.41E-03 cg11861654 -40.35% 14.50% 2.06E-03 NIPA1 225752_at -0.68 8.88E-03 cg00207833 -33.77% 19.11% 2.47E-03 NPEPL1 218822_s_at -0.54 9.41E-03 cg20022869 -27.50% 55.25% 1.05E-03 OGFOD2 44617_at -0.20 9.41E-03 cg21745287 38.78% 50.00% 2.29E-03 OTUD3 213216_at -0.40 6.40E-03 cg18856004 21.74% 81.66% 6.73E-04 P4HB 1564494_s_at -0.22 8.76E-03 cg17462329 -20.29% 71.17% 1.75E-03 P4HB 200654_at -0.26 9.41E-03 cg17462329 -20.29% 71.17% 1.75E-03 PAFAH1B3 203228_at 0.40 9.44E-03 cg08400147 -27.63% 6.64% 7.46E-04 PGLS 218387_s_at -0.37 9.44E-03 cg11912570 -16.43% 79.05% 1.50E-03 PLXDC2 227276_at 2.60 9.41E-03 cg19431623 35.35% 89.60% 4.62E-04 PNMA1 218224_at -0.41 6.24E-03 cg09238801 25.73% 76.19% 1.84E-03 POF1B 1555383_a_at -1.15 9.41E-03 cg08238865 -64.98% 6.74% 7.55E-04 PON2 210830_s_at 0.47 9.41E-03 cg18672670 -11.24% 4.03% 3.03E-03 PON2 201876_at 0.47 9.55E-03 cg18672670 -11.24% 4.03% 3.03E-03 PSTPIP2 219938_s_at 0.49 9.44E-03 cg00532474 -29.07% 60.25% 6.51E-04 RAB8B 222846_at -0.48 9.41E-03 cg17556578 -11.78% 5.63% 3.41E-03

Appendix E-9. (continued) Combined Transcriptome (p < 0.01) and Methylome DMPs (p < 0.005) for the NIKS-16ΔE7 to NIKS-16 Comparison

195

Target Score Gene Symbol Gene Description 87 DLG3 discs, large homolog 3 (Drosophila) 80 PDGFRA platelet-derived growth factor receptor, alpha polypeptide 78 COX19 cytochrome c oxidase assembly homolog 19 (S. cerevisiae) 78 ZNF14 zinc finger protein 14 77 FNBP4 formin binding protein 4 74 PHAX phosphorylated adaptor for RNA export 73 SUGP2 SURP and G patch domain containing 2 72 MAX MYC associated factor X 72 PIGX phosphatidylinositol glycan anchor biosynthesis, class X 68 RCC2 regulator of chromosome condensation 2 68 LEPREL4 leprecan-like 4 67 LILRB1 leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 1 67 FFAR2 free fatty acid receptor 2 65 HM13 histocompatibility (minor) 13 65 TNFSF10 tumor necrosis factor (ligand) superfamily, member 10 64 TONSL tonsoku-like, DNA repair protein 63 PPM1B protein phosphatase, Mg2+/Mn2+ dependent, 1B 63 WDR92 WD repeat domain 92 62 ELMOD1 ELMO/CED-12 domain containing 1 62 PPA2 pyrophosphatase (inorganic) 2 61 NMNAT1 nicotinamide nucleotide adenylyltransferase 1 60 C9orf66 chromosome 9 open reading frame 66 59 RPS6KA3 ribosomal protein S6 kinase, 90kDa, polypeptide 3 58 TPRG1L tumor protein p63 regulated 1-like 58 C6orf132 chromosome 6 open reading frame 132 58 KIAA1191 KIAA1191 57 FUT6 fucosyltransferase 6 (alpha (1,3) fucosyltransferase) 57 STMN3 stathmin-like 3 57 DHODH dihydroorotate dehydrogenase (quinone) 57 ABHD15 abhydrolase domain containing 15 57 ISY1 ISY1 splicing factor homolog (S. cerevisiae) 57 NAIF1 nuclear apoptosis inducing factor 1 57 RAP1A RAP1A, member of RAS oncogene family 56 CD96 CD96 molecule 56 CEP104 centrosomal protein 104kDa 56 B3GNT6 UDP-GlcNAc:betaGal beta-1,3-N-acetylglucosaminyltransferase 6 (core 3 synthase) 56 GNB4 guanine nucleotide binding protein (G protein), beta polypeptide 4 55 PPIL6 peptidylprolyl isomerase (cyclophilin)-like 6 55 LOC102725035 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102724997 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102725034 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102725031 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102725029 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102725027 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LOC102725015 leukocyte immunoglobulin-like receptor subfamily B member 3-like 55 LILRB3 leukocyte immunoglobulin-like receptor, subfamily B (with TM and ITIM domains), member 3 55 NLRP12 NLR family, pyrin domain containing 12 55 PPP2R3A protein phosphatase 2, regulatory subunit B'', alpha 55 ARIH2OS ariadne homolog 2 opposite strand 55 CERS5 ceramide synthase 5 55 CPN2 carboxypeptidase N, polypeptide 2 55 ALPP alkaline phosphatase, placental 55 MRPS10 mitochondrial ribosomal protein S10 54 TMIGD2 transmembrane and immunoglobulin domain containing 2 53 NFATC2IP nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 2 interacting protein 53 MINOS1 mitochondrial inner membrane organizing system 1 52 SGSM1 small G protein signaling modulator 1 52 NEK8 NIMA-related kinase 8 52 MGAT2 mannosyl (alpha-1,6-)-glycoprotein beta-1,2-N-acetylglucosaminyltransferase 52 ZNF549 zinc finger protein 549 51 ANKRD62 ankyrin repeat domain 62 51 MANEAL mannosidase, endo-alpha-like 51 PTK6 protein tyrosine kinase 6 50 GNPNAT1 glucosamine-phosphate N-acetyltransferase 1 50 HAUS3 HAUS augmin-like complex, subunit 3 Appendix E-10. Predicted Target mRNAs for HLA-E CGI Small RNA

196

APPENDIX F

Detailed Protocol for Designing and Performing Methylation Specific PCR

Flow Chart

PCR Protocol for Platinum Taq

197

Protocol for Optimization

198

199

APPENDIX G

Detailed Protocol for Dual Luciferase Reporter Assay

200