xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJLAKNVJ9_O5YURM INGEST_TIME 2014-10-03T21:26:14Z PACKAGE UFE0046319_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
1 INVESTIGATION OF THE ROLE OF INSULATOR BINDING PROTEIN CTCF IN LYTIC HSV 1 REPLICATION By CAMERON LAINE LILLY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQU IREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 4
2 2014 Cameron Laine Lilly
3 This is dedicated to my friends and family without whom this would not have been possible Thank you for believing in me! I love you all!
4 ACKNOWLEDGMENTS I would like to start by thank ing my partner, Blair, for all of the support and love she has shown me. She has picked me up from the lab at 2 :00 a m brought me food and made me coffee when I was unable to get away from the bench and somehow managed to find the patience to not strangle me when I was frustrated and stressed out. Her support has be en absolutely indispensable. me nor when I fell asleep on her and may or may not have drooled Together we have built a home and a partnership, adopted a little striped dork with goofy ears and built a pack. I would also like to thank my family, as they too have shown unwavering support and love over my graduate ca reer My father has helped me achieve perspective and calmness when I was panicking and ultimately is responsible for unleashing me upon the world with an unquenchable curiosity and desire to know how things work. I would like to thank my mother for embo dying unconditional love and kindness to me throughout my life, even when I was trying her patience. My twin, my womb mate, my sister, the other Dr. Lilly, has always been positively perky and I thank h er for her joy and her support, and especially for th at face that she makes when she tries my beer. I also have to thank my third twin, partially due to the infrequency with which one gets to do so, but mostly because she is one of my dearest friends, and her of boundless enthusiasm and willingness to say, My family has taught me the value of a good meal shared with those you hold dear, especially when Leanne pterodactyls and we all laugh until we cry. Where would I be without Jenni? It was he r on her advice that I major ed in biochemistry during undergraduate, than I join ed my fraternity, and that I announce for
5 roller derby. Our friendship has withstood more than a decade, some of that time under intense strain and we have re forged our frien dship to be ever stronger I have to thank her for being my closest friend, and one who knows me better than almost anyone. I look forward to being friends with her for a long time yet. I, of course, have to thank my fellows of GHTS: Jeff, Jen, Nick, Terry, and Ian. Those friends are some of the greatest things I have found during my time in graduate school. I am truly grateful to have such excellent friends. They have supported me both inside and outside of the lab The members of my lab have also provided me with phenomenal support and wisdom over the course of my graduate career. Levi, Derek, Nat, Tang, Harald, Sanae and Dane have helped me shape countless hypotheses and ideas, eat several absurdly large birthday cakes, and shared many great tim es with me. Lastly, I would like to thank Dr. Dave Bloom. Dave believed in my potential during a tumultuous period early in my graduate career and brought me into his lab, and under his guidance I have learned an immeasurable amount. I would like to th ank Dave for the years of patient guidance, for never giving up on me, and for helping me learn how to think like a scientist. I would like to thank him for always considering my input, no matter my level of experience in the lab Most of all, I would li ke to thank Dave for the teaching he provided me.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 The Paradigm of Herpesviruses ................................ ................................ ............. 14 Herpes Simplex Virus Ty pe 1 ................................ ................................ ................. 15 Infection and Disease ................................ ................................ ....................... 15 Prevalence of HSV 1 ................................ ................................ ........................ 16 HSV 1 Lytic Infection ................................ ................................ ........................ 17 Viral Latency and Latent Transc ription ................................ ............................. 19 Regulation of Viral Gene Expression through Epigenetics ............................... 19 HSV 1 Latency and Chromatin ................................ ................................ ......... 21 Organization of Transcriptional Domains by the Insulator Protein CTCF ......... 22 CTCF and Gammaherpesviruses ................................ ................................ ..... 24 Lytic Replication and Host Chromatin Factors ................................ .................. 25 2 METHODS ................................ ................................ ................................ .............. 32 Cells and Viruses ................................ ................................ ................................ .... 32 In Vitro Infection with HSV 1 ................................ ................................ ................... 33 Chromatin Immunoprecipitation ................................ ................................ .............. 33 Real time PCR of ChIP ................................ ................................ ........................... 34 Calculation of Enrichment of CTCF ................................ ................................ ........ 34 SiRNA Knockdown of CTCF ................................ ................................ ................... 35 In Vitro Infection for siRNA E xperiments ................................ ................................ 36 Isolation of DNA from Infected Cells for R eal t ime PCR A nalysis ........................... 36 RNA Isolation for RT PCR and R eal time PCR analysis ................................ ......... 36 Reverse Transcription ................................ ................................ ............................. 37 Calculation of Q uantity of DNA and mRNA ................................ ............................. 37 3 RESULTS ................................ ................................ ................................ ............... 39 CTCF Binds to the HSV 1 Genome Early Dur ing Lytic Replication in Neuro2A Cells ................................ ................................ ................................ ..................... 39
7 CHIP of CTCF is Capable o f Detecting Established Differences of Enrichment in an Analogous System ................................ ................................ .......................... 43 Differences in CTCF Enrichment are Readily Detectable on BAC 16 DNA in 293 T Cells ................................ ................................ ................................ ........... 44 CTCF is Enriched on the HSV 1 Genome at Low Levels in 293T cells .................. 45 A Viral Mutant with an Abrogated Transcriptional Program Exhibits Altered CTCF Enrichment ................................ ................................ ................................ 47 Depletion of CTCF by siRNA Knockdown Alters HSV 1 Replication and Gene Expression ................................ ................................ ................................ ........... 53 Viral DNA accumulates to higher levels in cells depleted for CTCF ................. 54 Viral gene expression after depletion of CTCF by siRNA ................................ 57 4 DISCUSSION AND CONCLUSIONS ................................ ................................ ...... 70 LIST OF REFERENCES ................................ ................................ ............................... 83 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 91
8 LIST OF FIGURES Figure Page 1 1 Diagram of the HSV 1 Genome. ................................ ................................ ......... 28 1 2 Comparison of the transcriptional program of Lyt ic Replication versus Latency. ................................ ................................ ................................ .............. 28 1 3 Diagram of chromatin on th e HSV 1 genome during latency. ............................. 29 1 4 Insu lator function and CTCF binding. ................................ ................................ 30 1 5 Diagram of the CTCF binding motifs in the HSV 1 genome .............................. 31 3 1 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected at MOI 3. ....... 60 3 2 CTCF Chromatin Immunoprecipitation in Neur o2A cells infec ted at MOI 0.2. .... 61 3 3 CTCF Chromatin immunoprecipitation in 2932T cells harboring KSHV DNA maintained as BAC16. ................................ ................................ ........................ 62 3 4 CTCF Chromatin Im munoprecipitation in 293T cells ................................ .......... 63 3 5 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected with HSV 1 strain KOS. ................................ ................................ ................................ ......... 64 3 6 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected with HSV 1 ICP4 deletion virus KD6. ................................ ................................ .................... 65 3 7 Viral DNA in cells treated with siRNA again st CTCF and infected at MOI 3. ...... 66 3 8 Viral DNA in cells treated with siRNA against CTCF and infected at MOI 0.2. ... 67 3 9 Viral transcript levels in Neuro2A cells infected at MOI of 3 after treatment with either non targeting ( NT ) or anti CTCF siRNA. ................................ ........... 68 3 10 Viral transcript levels in Neuro2A cells infected at MOI of 0.2 after treatment with eith er non targeting ( NT ) or anti CTCF siRNA. ................................ ........... 69 4 1 Model of CTC F binding in Low and High Multiplicities. ................................ ....... 81 4 2 Potential to assemble CTCF onto incoming HSV 1 genetic material in order to establish chromatin and coordina te HSV 1 gene expression. ........................ 82
9 LIST OF TABLES Table page 2 1 Real time PCR Primer and Probe Sequences ................................ ................... 38
10 LIST OF ABBREVIATIONS BAC Bacterial Artificial Chromosome Bp Base Pairs CTCF CCCTC Binding Factor ChIP Chromatin Immunoprecipitation DNA De oxyribonucleic Acid E Early EBV Epstein Barr Virus HCF Host Cell Factor Hpi Hours Post Infection HSK Herpes Stromal Keratits HVEM Herpes Viral Entry Molecule IE Immediate Early KSHV Sarcoma associate Herpesvirus L Late LANA Latency Assoc iated Nuclear Antigen LAT Latency Associated Transcript MINE Myc Insulator Element MOI Multiplicity of Infection mRNA Messenger RNA NT Non targeting Orf Open Reading Frame PBS Phosphate Buffered Saline PCR Polymerase Chain Reaction PTMs Post translational Modifica tions
11 PFU Plaque Forming Units RNA Ribonucleic Acid rRNA Ribosomal RNA RT Reverse Transcription qPCR Real time PCR SUMO Small Ubiquitin like Modifier Tk Thymidine Kinase
12 Abstract of Dissertation Presented to the Graduate School of the Unive rsity of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INVESTIGATION OF THE ROLE OF INSULATOR BINDING PROTEIN CTCF IN LYTIC HSV 1 REPLICATION By Cameron Laine Lilly May 2014 Chair: David Bloom Major: Medical Sciences Lytic replication of Herpes Simplex Virus type 1 can proceed in many different tissues. Latency of this virus, however, can only occur in a very specialized subset of sensory neurons. During both lytic replication and late ncy, the HSV 1 genome subverts and appropriates cellular processes to associate with chromatin proteins such as histones. To keep differentially regulated regions of chromatin separate during latency, HSV 1 encodes a nu mber of DNA sequences which bind the cellular protein CTCF. One such sequence, known as B2, possesses the ability to perform enhancer blocking function. CTCF is known to coordinate gene expression during development and regulate genes in somatic cells and also known t o be involved in KSHV and EBV latency and in reactivation and this interaction has been shown to be remodeled during reactivation. During lytic replication of HSV 1 the binding of host chromatin proteins is less organized. Histone proteins are associated with the HSV 1 genome so it st ands to reason that other chromatin factors may assemble on the HSV 1 genome even if the genome is not destined for latency This, taken with what is known from the gammaherpesviruses studied, suggest CTCF may act in coordinating herpesviral gene
13 expression. In this study we strove to determine if CTCF is a factor in coordinating temporal gene expression during HSV 1 lytic infection This study showed that CT CF binds not only to CTCF binding sequences during lytic replication, but CTCF binding decreases dramatically as DNA re plication occurs Enrich ment of CTCF on the HSV 1 genome correlates with transcription from the genome as a mutant lacking a critical transcriptional a ctivator, KD6, has a static bind i ng pattern of CTCF. Finally, if CTCF is depleted from the cells prior to in fection, amounts of HSV 1 DNA increase and viral gene expression is altered as a result These studies suggest that CTCF association with the genome is mutual ly exclusive to lytic replication and give s CTCF a potential role for transcriptional control during lytic infection.
14 CHAPTER 1 INTRODUCTION The Paradigm of Herpesviruses Herpesvirus es (order Herpesviralis) are a family of viruses with large double stranded DNA genomes. This clade of viruses includes pathogens of a diverse breadth of species, ranging from invertebrates to reptiles, birds, primates, and humans (for review see (1) ) The Herpesviridae family contains the members of this family which infect mammals, and is further divided into three subfamilies delineated by the cell type required for the establishment of latent infection. Alphaherpesviruses establish latent infection s in neurons, betaherpesviruses in T cells, and gammaherpesvirus in B cells (2) The herpesvirus paradigm involves a biphasic life cycle. One phase of this life cycle is a lytic replication cycle. During lytic replication, viral genes are expressed in a temporal cascade: the Immediate Early (IE) genes expressed very quickly (catalyzed by proteins within the virus particle) and then catalyze the expression of the Early (E) class of genes which includes the viral DNA replication machinery, and lastly t he Late (L) gene class is expressed robustly after DNA replication and these genes encode the proteins needed to assemble infectious progeny virus. The second phase is the hallmark of the herpesviruses, which is known as latency. Transcriptionally distin ct from lytic replication, gene expression from the viral genome is tremendously limited and the genome is largely transcriptionally quiescent. Latency, as mentioned previously, is also only established in a specific type of host cell. During latency, th e genome of the latent virus exists as extrachromosomal episome s using similar mechanisms of transcriptional repression and activation as used on our cellular chromosomes. Latency is punctuated by the ability to reactivate: to respond to cellular stimuli and return to lytic replication in
15 order to periodically produce new virus, affording the virus an opportunity to spread to a new host. Herpes Simplex Virus Type 1 Infection and Disease Herpes simplex virus type 1 is a prototypical member the alphaherpes virus family and is a pathogen of humans. The infectious particle is made up of a number of structures: the nucleocapsid which contains the double stranded DNA genome, surrounded by an amorphous layer known as the tegument, rich in viral effector molecule s, and lastly the envelope which is derived from the host cell and contains viral proteins for cell adhesion and entry. HSV 1 packages a genome of approximately 152 kilobases which encodes for ~80 genes (3) The structure of the HSV 1 genome can be described as two regions, designated Long and Short each with respective unique segments flanked by inverted terminal repeats designated Repeat Long an d Repeat Short regions (Figure 1 1). These regions are joined by a small intervening sequence. Primary infection (lyti c infection) with HSV 1 in humans occurs typically at the orolabial mucosal epithelium or in the mucosal epithelium of the eye. In orolabial infection, a cold sore or fever blister can form as a consequence of infection but the majority of infecti ons are sub clinical (4) Virus produced in the periphery then infects nerve termini enervating the site of infection, where the viral nucleocapsid is transported down the axon to the cell body of the trigeminal ganglion via fast axonal retrograde transport. Upon reaching the cell body of the infected neuron, the genome is i n jec ted into the nucleus. Here, the virus is capable of taking two paths: lytic replication can ensue to produce additional infectious virus, or viral lytic gene transcription can be quelled and the viral genome can become latent. The factors involved in directing this
16 crossroads in viral biology are as of yet incompletely understood Upon stimulus, reactivation occurs and results in the production of infectious virus which is then trafficked anterograde down the axon. At the surface of the lips and mou th, the progeny virus is shed and can result in a cold sore or a fever blister though the large majority of rea ctivations are subclinical (5) In the eye, repeated reactivation events and the immune response thereto can result in scaring of the cornea, described as Herpes Stromal Keratitis (HSK). Th is can lead to blindness, necessitating corneal transplantation to recover sight; however, this does not address viral genomes harbored within the trigeminal ganglia, and does not preclude further reactivation events and recurrence of disease in the donor tissue. HSK and the difficulties treating it make HSV 1 the leading cause of infectious blindness in the United States (6) Indeed, treatment for HSV 1 disease centers on nucleoside analogues to interrupt lytic replication post reactivation, and do not effect clearance of latent HSV 1 genome within the infected individual. Prevalence of HSV 1 Within the populous of the United States, the overwhelming majority will likely be exposed and infected by HSV 1 during the course of their life. Studies comparing seropositivity and age have demonstrated that by the time the sample population reaches age 70 or greater, 90% of the individuals surveyed have evidence of infect ion with HSV 1. Age of incidence was determined to be fairly young in the interval studied, as almost 45% of the individuals between 12 19 years of age were seropositive, and total, in persons older than the age of 12, 67.6% were seropositive for HSV 1 (7) Despite the majority of populous being latently infected, only 1 6% of infected individuals present clinical symptoms (4) The relatively low prevalence of disease with
17 the high prevalence of infection indicate s then that infection does not always correlate with disease, reflecting a mechanism for spread that does not rely on clinical disease; indeed the overwhelming majority (98%) of individuals studied shed viral DNA in tears and saliva in the absence of clinical disease (8) HSV 1 Lytic Infection Lytic replication of HSV 1 begins with a membrane binding event between the host cell and viral particle. Virally encoded glycoproteins embedded on the surface on the virion bind complementary receptors on the surface of the cell. T he players in this interaction, from the viral side, are glycoproteins gD, gB, gH, and gL, and they bind to cellular receptors HVE M (Herpes Viral Entry Molecule) and nectin 1, coordinated by gC binding to heparin sulfate (9, 10) Subsequent to binding to the host cell, the viral membrane and the cellular membrane fuse, spilling the viral tegument proteins and the nucleocapsid into the cytoplasm of the cell The tegument proteins serve to establish an intrace llular environment conducive to infection immediately upon entry without the need for transcription. The nucleocapsid is transported to the nucleus of the cell via microtubule mediated trafficking, where the genome of the virus is then inserted into the n ucleus. In the nucleus, the execution of the viral transcriptional program can ensue (11, 12) Carried into the cell with the tegument is a tegument protein known as VP16, which recrui ts two host cell proteins HCF (Host Cell Factor) and Oct 1, a transcription factor. This VP16 complex then transactivates transcription from the viral IE promoters, directed by Oct 1 which binds to a TAATGARAT motif in the promoter sequence VP16 which po ssesses a potent transactivation domain, and HCF which recruits transcriptionally activating chromatin modifiers (13) The immediate early genes ICP0,
18 ICP4, ICP22, ICP27, and ICP47 are subsequently expressed. The IE proteins function to control the host cell by inhibiting the intracellular factors requ ired for the immune response regulate RNA trafficking and splicing as well as to catalyze the further transcription of both IE and E classes of genes (14 21) Critical for the furthering of viral transcription is protein IC P4, as viruses mutated for ICP4 are incapable of replicating as a result in the defect in I E gene and subsequent classes of gene expression (22) ICP0, as well as contributing to viral transcription, is an E3 ubiquitin ligase known to disrupt subnuclear structu res known as ND10 bodies (23, 24) ND10 bodies have been shown to be involved in interferon inhibition of viral transcription (25) ND10 bodies are disrupted by ICP0 mediated ubiquitination of the ND10 protein PML, resulting in its degradation (23) ICP0 has also been shown to catalyze the ubiquitination of RNF8 and RNF168 to suppress the DNA damage response to the incoming viral DNA (26) Vir uses with mutations in ICP0 display decreased replication at a low multiplicity, however this can be overcome by performing infection at a high multiplicity (27 30) ICP4 as well as ICP0, 22, and 27 further the transcriptional program into the expression of E class genes. E genes are largely comprised of the machinery required to replicate the viral genome (e.g. thymidine kinase (tk), and the HSV 1 DNA polymerase [ pol ] ). Post DNA replication, the L class of genes are ex pressed robustly, and encode the proteins required to assemble new infectious particles which are assembled in the nucleus of the cell, then bud from the nu clear membrane through the Golgi and subsequently from the cell surface at membrane structures enriched for viral proteins (31)
19 Viral Latency and Latent Transcription Latency is a diametrically opposed state to that of lytic replication. Historically, the hallmark of latently infected neurons is the accumulation of a single transcript in great abundance: the Latency Associated Transcript (LAT ). The LAT is a ~8.5 kilobase non coding RNA transcript, which is then spliced into a 2.0 kb stable intron (32) This 2.0 kb intron accumulates in sensory neurons, and has been demonstrated to have a half life of up to 24 hours in cell culture (33) This 2.0 kb intron can be further processed into a 1.5kb intron, which is also detectable during latency. Despite intensive investigation, the function of the LAT 2.0 kb and 1.5 kb intron are unknown. Transcription from the LAT region of the genome has also been found to produce some other RNA products. Eight microRNAs have been found to be encoded within the LAT promoter and transcriptional region, 6 of which are co linear with the LAT primary region (34, 35) These microRNAs have been demonstrated in cell culture to regulate viral genes (36) During latency, lytic gene transcription can be detected with very sensitive RT PCR methods, indicating that perhaps abortive lytic transcription occurs without disrupting the latency program (37, 38) A comparison of the viral lytic gene expression and latent gene expression pa tterns are presented in Figure 1 2 Regulation of Viral Gene Expression through Epigenetics Gene expression is regulated by several mechanisms that change the expression of th e genetic information without changing the genetic sequence. One such method of controlling gene expression is to methylate the DNA itself at CpG bases within the DNA sequence. This covalent modification to the 5 position of the ring of cytosine physical ly blocks the DNA binding proteins such as transcriptional machinery
20 which regulate the target sequence, resulting in silencing of the methylated locus (39 43) The organization of the genetic information of eukaryotes is done in a manifold process involving both physical and functional organization. The firs t level of this is the histone protein complex, around which DNA is wound (44) This serves to compact that relatively large chromosomes into the small space of the nucleus of the cell. The histo ne not only serves as a spool around which the DNA can be physically compressed but as a template for regulation access to the encoded genetic information. Fully assembled canonical histones, octamers containing two copies each of histone proteins H2A, H2 B, H3, and H4 contain core domains which are buried within the core of the macromolecule (44) However, histone proteins themselves are subjected to specific post translational modifications (PTMs) on their C and N termini, which exist free from the DNA interaction domain of the histone macromolecule and exposed in space. These modifications correlate to the function of the given region of DNA associated with those histones; thus genomic regions which are associated with histones bearing the same post translational mark can be said to be functionally analogous, hence the so called (45) The histone code is generalized into two subclasses, euchromatin and heterochromatin. Euchromatic regions represent regions of chromatin that are transcriptionally accessible or capable, and are associated with histone PTMs that are mutually exclusive to those of heterochromatic regions. Heterochromatin is repressed o r transcriptionally inactive regions or silent regions. In order to change the functional status of a region of DNA, its histone code must also be changed to reflect the new fu nction. This can be done b y directly changing the modification of the histone
21 HSV 1 Latency and Chromatin HSV 1 is a champion of subverting, subduing, or appropriating cellular processes for infection. Regulation of HSV 1 DNA during latency is organized onto histone proteins and these histones bear PTMs which indicate they are regulated like host chromatin (46) During latency, the genes required for lytic infection are enriched in heterochromatic post translational modification s thought to repress lytic gene expression (47 49) The LAT promoter and enhancer region (Figure 1 3 ) however are the only regions in the genome enriche d for euchromatic marks, consistent with the known transcriptional activity the LAT locus (38, 50) These regions of differentially regulated HSV 1 chromatin are immediately proximal, as ICP0, a lytic gene, is immediately downstream and antisense to t In cellular systems, maintaining regions of differentially regulated chromatin is done by insulator elements, and so too are they in HSV 1: HSV 1 contains several sequences with in the viral genome that are capable of binding to CT CF (51) Within the viral genome, there have been seven CTCF binding sequences described an d are delineated B1 B7 (Figure 1 5 ). During latency, CTCF was detected by Chromatin Immunoprecipitation (ChIP) at B1 and B2, w hich flank the LAT promoter and enhancer region, B5 and B6, which flank lytic immediate early gene ICP4, and B7, present within the unique short region of the genome downstream of lytic gene ICP47. The B2 binding site was also vetted in classical lucifera se assays to possess enhancer blocking activity, making it a true insulator. The reigning hypothesis is that CTCF enrichment at the sequences is require d for the appropriate maintenance of latent viral chromatin and subsequently the control of lytic gene expression.
22 Organization of Transcriptional Domains by the Insulator P rotein CTCF Within the confines of the nucleus of the cell, making optimal usage of three dimensional space is paramount. One major consequence of a genome exhibiting complex regulation is the need to distinguish physically proximal regions of genetic information that may be functionally disparate. In order to maximize the efficiency of these processes, being able to organize and concentrate loci which will be regulated in parallel woul d be idea l To achieve this end, eukaryotes have a protein known as CTCF, which stands for CCCTC Binding factor, so named for the initial pentameric sequence it was discovered to bind (52) CTCF was originally identified as being crucial for the regulation of the proto oncogene c Myc, where it was initially purported to function as a transcriptional repressor by mutational analysis (52) CTCF, as a protein, contains eleven separate zinc finge r domains capable of binding to DNA. As a result, the binding sequences capable of being bound by CTCF are highly polymorphic allowing this singular pr otein to regulate a wide variety of loci (53) Sequences of DNA which bind CTCF have been found to be highly divergent (53) CTCF binding sites are classifie d as being insulators (54) Insulator elements exist to functionally delineate genes which are regulated differentially in a number of different capacities (Figure 1 4 ) One of the traditional canonical functions of an insulator is to block the long distance i nfluence of an enhancer sequence in a position dependent manner; an insulator will protect a promoter from an enhancer on the other side of it (Figure 1 4 A) (55) T his had classically been defined by utilizing plasmids encoding luciferase downstream of a promoter element and monitoring the luciferase expression of the same construct with and without an enhancer and an insulator element interjected between the promote r and enhancer (56) Should the introduction of the putative
23 insulator between the enhancer and promoter cause a decrease in luciferase expression, the sequence is said to perform a s an enhancer blocker. A related function of insulators is to provide a boundary element to regions of chromatin. This functions to keep heterochromatin from spreading into regions of active transcription and allow for regions that are euchromatic and he terochromatic to be juxtaposed on the same strand of DNA (Figure 1 4 B) Ultimately, all of these functions are carried out through the formation of chromatin loops: regions of chromatin which need to be separated are physically displaced from one another to form so called chromatin hubs (Figure 1 4 C) (57) These hubs allow for a complex regulation of chromatin regions on the same or even different chromosomes in the same physical space in the nucleus. R egulation of CTCF binding to the target sequence can be altered in a number of ways. One such way is, as previously mentioned, to methylate the target binding sequences, which tend to be G C rich, which prevents CTCF from binding. CTCF can itself be regu lated by post translational modifications. Phosphorylation of CTCF by protein kinase CK2 at serine residues in its c terminal can modulate how CTCF binding affects transcription of the regulated locus (58, 59) Modification of CTCF by poly(ADP ribosyl)ation is thought to assist in CTCF binding and segregation to imprinted loci, and has also been demonstrated to regulate CTCF mediated insulator functi on (60) CTCF is also subjected to post translation modification by the S m all U biquitin like Mo difier (SUMO) which seems to play a role in regulating the suppression of genes, and SUMOylation of CTCF at its N and C termini by polycomb protein Pc2 controls its role in suppression of the c Myc P2 promoter region (61)
24 CTCF and Gammaherpesviruses CTCF has repeatedly been demonstrated to be important for herpesvirus biology Sarcoma associated Sarcoma and Primary Effusion Lymphoma, contains three CTCF binding sites upstream of Orf73, which encodes for the KSHV latency protein LANA (Latency Associated Nuclear Antigen) (62) During KSHV Latency and episomal maintenance in quiesce nt cell culture models, CTCF binds to these binding sites to form a chromatin loop between the promoter of Orf 73 and Orf 50 which encodes RTA, which is the major lytic switch protein in KSHV (63) In this same study, it was also determined that disrupting the binding of CTCF to these motifs by mutation binding sites caused a defect in reactivation by chemical stimulus but not when reactivation induction was performed by transducing with RTA, ind icating that CTCF has a role in potentiating signaling based KSHV reactivation by coordinating RTA expression. CTCF has also been shown to be important for KSHV transcription, This was corroborated by single molecule analysis which suggests that in cells harboring the KSHV genome, nearly every genome is associated with CTCF at all three CTCF binding sites (64) Epstein relationship with CTCF. CTCF binds to a regions proximal to a pr omoter known as Cp, where it plays a role in regulating the polycistronic transcript encoding one of the EBV latency antigens, EBNA2 (65, 66) Subsequent genome wide studies of EBV latency revealed other loci of CTCF binding. This study focused on the Q promoter of EBV, and demonstra ted that mutational deletion of the CTCF binding site results in a loss of
25 transcription as a result of CpG methylation and the deposition of the heterochromatic mark H3K9 trimethylation (67) A rece nt report describes CTCF function in forming a chromatin loop between the viral origin of replication, OriP (near the Cp binding site), and a region overlapped by both the LMP1 and LMP2 transcripts, which encode membrane proteins detectible during EBV late ncy. Mutational analysis showed a loss of this chromatin loop, deregulation of LMP1 and 2 transcription, and again, H3K9 trimethylation deposition (68) These studies, combined with the studies in KSHV, describe CTCF as a factor critical for organizatio n of herpesviral genomes and coordination of transcription. Lytic Replication and Host Chromatin Factors Chromatin during HSV 1 productive infection in cultured cells has similarly been studied to investigate the potential for chromatin control of lytic g ene expression. Previous work has established that very early in lytic infection in vitro, histone proteins are associated with the HSV 1 genome and the DNA is organized into nucleosomes (69, 70) though the spacing of these nucleosomes is irregular. A nalysis of the post translational modifications of the histones associated with the HSV 1 genome during lytic replication indicates that the bulk of these are consistent with euchromatin, which stands to reason since the viral genome is tremendously transcriptionally active during lytic replication. However, there has been a consistent observation in these studies in that during vira l DNA replication, the enrichment of histones on the HSV 1 genome appears to be greatly reduced, and as infection proceeds into late time points where vial assembly is taking place, remains low (70) It remains unclear as to whether the euchromatic histone modifications that are associated with the lyt ic genes early in
26 infection are required for efficient IE gene expression, though studies interrupting chromatin modifiers that associate with HCF sugge st this is likely the case (71) ICP0 has been shown to interact with histone modifying proteins. During infection ICP0 causes the disruption of the suppressor REST/CoREST complex by displacing the histone modifying subunit, Histone Deacetylase 1. This causes a p artial change in both CoREST and HDAC1 localization from nuclear to cytoplasmic. This process appears to be dependent on phosphorylation of HDAC1 and CoREST by the viral phosphatase U S 3. Lysine Specific Demethylase 1, a histone demethylase enzyme, is als o a part of the REST/CoREST repressor complex, and infection with HSV 1 has been shown to disrupt LSD 1 association with the complex, as well as to localize with ICP8, the viral single stranded DNA binding protein (72 77) HCF 1, previously mentioned to be part of the VP16 transcriptional complex, is also a member of a host protein complex called MLL which is responsible for the deposition the euchromatic histone mark of trimethylation of lysine residue 4 on histone protein H3 (H3k4 3me). Though HCF 1 is not directly a chromatin modifying protein, these studies suggest that appropriation and disruption of chromatin binding and chromatin modifying proteins is critical for regulating the host nuclear response to the incoming viral DNA in order to carry out infection. Previous work investigating the occupancy and functionality of the CTCF binding motifs present in the HSV 1 genome have been carried out in systems outside of lytic replication. Amelio et al. (51) utilized ChIP and traditional transient insulator assays to confirm enrichment of CTCF in latency and the functionality of the insulator dubbed B2 respectively. Chen et al. (78) utilized a heterologous system in Drosophila melanogaster These tests thoroughly vetted B2 as an insulator, which
27 is of specific interest for latent biology due to its positioning in the intron o f the LAT, between the LAT enhancer and downstream lytic transcript ICP0. Both of these tests though remain outside of the requirements of replicaiton of the viral genome However, the reiterated elements in the genome which can potentially bind to CTCF remain present in during lytic replication, which is a nuclear phenomenon. As it has been established in previous studies that other chromatin proteins associate with the genome during lytic replication, it stands to reason CTCF may associate with the HSV 1 genome during lytic replication as well. The experiments presented in this dissertation were 1 in cell culture models.
28 Figure 1 1 Diagram of the HSV 1 Genome. R L : Re peat Long, UL: Unique Long, R S Repeat Short, U S : Unique Short. Figure 1 2. Comparison of the transcriptional program of Lytic Replication versus Latency. During Lytic replication the genome is linear and ~80 genes are expressed in a temporal ca scade, whereas during latency, the genome is circularized and only one major gene product can be detected, the LAT. Lytic Replication
29 Figure 1 3 Diagram of chromatin on the HSV 1 genome during latency. Heterochromatin (red) is present across the e ntire genome, whereas euchromatin (green) can only be found on the LAT promoter (L. Pro) and the LAT enhancer (L. Enh). L. Enh L. Pro
30 A B C Figure 1 4 I n sul ator function and CTCF binding. A) Enhancer blocking: CTCF prevents enhancer funct ion (green arrows) across insulator elements (blue box es ) with CTCF binding sequences ( red boxes) B) Barrier Elements: heterochromatin spread (red circles) into euchromatin (green circles) is blocked by insulator elements. C) Chromatin hub formation: CT CF organizes chromatin into loops in order to coordinate gene expression in proximal regions which are differentially expressed. Enh : Enhancer. CBS: CTCF binding sequences Insulator
31 Figure 1 5 Diagram of the CTCF binding motifs in the HS V 1 genome The top section shows position of the CTCF binding sites in the genome, and bottom f ocus es on the internal repeat structure of the genome showing placement of CTCF binding sites and select transcripts in the region All but one such motif are in the repeat regions of the genome and thus exist in two co pies each per unit length genome. LAT ICP4 ICP0 B1 B2 B3 B4 B5 B6 B7 ICP47 B4 B2 B3 B1 B1 B2 B3 B4 B5 B6 B7 B6 B5
32 CHAPTER 2 METHODS Cells and Viruses HSV 1 strain s 17syn+ and KOS were obtained from J. Stevens. ICP4 knockout virus KD6 was obtained from L. Feldman. Virus stocks of 17 syn+ and KOS were grown and titrated on rabbit ski (Life Technologies) supplemented with 5% calf serum and 100 units/ml streptomycin, 100 g/ml penicillin and 292 ng/ml L glutamine. Virus stocks of KD6 were grown on Vero cells which express ICP4 (Life Technologies) supplemented with 10% fetal bovine serum and 100 units/ml streptomycin, 100 g/ml penicillin and 292 ng/ml L glutamine. Neuro2A cells were obtained from ATCC. RSCs were obtained from J. Stevens. E5 cells were obtained from L. Feldman. 293T BAC16 cells and 293T cells were a generous gift from R. Renne. Cells were counted prior to seeding using trypan blue exclusion and a hemocytometer. Neuro2A cells (N2A) were cultured in minimal essential m edia supplemented with 10% Fetal Bovine serum and 100 units/ml streptomycin, 100 g/ml penicillin and 292 ng/ml L glutamine. 293T cells were cultured in 293T cells were grown in Dulbecc clone) supplemented with 10% fetal bovin e serum, 100 units/ml streptomycin, 100 g/ml penicillin, 292 ng/ml L glutamine. For experiments in 293T BAC16 cells 293T cells stably tra nsfected with BAC16 (293T BAC16 which cont ains the KSHV genome) were grown in Dulbecc c lone) supplemented with 10% fetal bovine serum, 100 units/ml streptomycin, 100 g/ml penicillin, 292 ng/ml L glutamine, and 200m/l Hygromycin B (Cellgro) to maintain selection of the BAC.
33 In Vitro Infection with HSV 1 2x10 6 cells were used for ChIP experiments, whereas between 1.48x10 5 and 1.58 x 10 5 cells were used for siRNA experiments Growth media was decanted and replaced with the appropriate dilution of virus in order to yield either a multiplicity of infection of 3 pfu/cell or 0.2 pfu/cell, i n an inoculum volume of 500 l. After a one hour incubation at 37C, the infection inoculum was removed and the cells were either harvested as the 0 hour time point, or complete growth media was replaced and cells were returned to the incubator until time of harvest for ChIP processing. Chromatin Immunoprecipitation ChIP was performed as previously reported with the following alterations (50) Media was decanted from the cells at the appropriate time and the monolayers were washed with phosphate buffered saline (PBS) supplemented with HALT protease inhibitor cocktail (Thermo). 500l of PBS was then added to the dish and the cells were harvested and transferred to a 1.5ml mic roc entrifuge tube. S amples were then formaldehyde cross linked as previously reported. Prior to chromatin shearing by sonication, samples were lysed in ChIP Lysis buffer for 45 60 minutes at 4C. Shearing was performed via sonication either by probe sonica tion as described previously or using a Bioruptor Twin (Diagenode) instrument. DNA fragment size was analyzed by agarose gel electrophoresis and was determined to be between 500 bp and 1,000 bp. For experiments where 293T BAC16 chromatin was added to HSV 1 infected 293T cell chromatin, sheared chromatin from 293T BAC16 cells was added to sheared chromatin from HSV 1 infected 293T cell at a concentration of 1% (vol/vol). Chromatin was then diluted in ChIP Dilution Buffer and pre cleared using salmon sperm DNA Protein A agarose beads (Millipore) at 4C. Rabbit anti CTCF (Millipore) antibody was added to
34 the sheared, pre cleared chromatin at a dilution of 1.85l of antibody per ml of pre cleared chromatin and incubated overnight with shaking at 4C. Immunoc hromatin complexes were then captured by adding fresh salmon sperm DNA protein A agarose beads and incubating with shaking for 2h at 4C. The immunochromatin protein A agarose b ) were pelleted by centrifugation and the super Bead bound Immunochromatin complexes were washed and eluted from the beads, and chromatin from both bound and unbound fractions was treated with RNase A and Proteinase K. DNA was then isolated using a Qiaquick nucleotide removal kit. Real time PCR of ChIP All real time PCR analysis described herein was performed using Taq Man Universal Fast Mix, No Amp erase UNG (Applied Biosystems) on a Step One Plus real time PCR instrument (Applied Biosystems). All primer probe sets were either obtained ready made from or designed and synthesized by Applied Biosystems and were used according to manufacturer specifications. Primer and probe s equences can be found in Table 1 PCR cycle conditions are as follows: cycle 1: 95 C 20s, cycles 2 40: 95C 1s, 60C 20s. Fluorescence thresholds were set in linear range of amplification for standard curve dilutions of cellular or viral DNA. For relative quantification, PCR of samples was performed in triplicate and compared to a 10 fold dilution series of either cellular or viral DNA. Relative quantities for bound and unbound fractions for each sample were then calculated from the standard curve generated from these dilutions of DNA. Calculation of Enrichment of CTCF ChIP validatio n was performed by analyzing the calculated relative quantity of two cellular DNA loci; comparing the MINE region upstream of c Myc (79) which is
35 known to bind CTCF, to a negative control region ( upstream Hox gene A5 in Neuro2A cells, or t he 18s rRNA gene in 293T and 293T BAC16 cells ) Samples with positive over negative fold enrichment greate r than or equal to 2 were said to have valid precipitation and were further analyzed for viral targets. For HSV and BAC16 targets, enrichment is calculated as the ratio of the bound fraction divided by bound plus unbound or B/B+U, divided by the B/B+U fo r negative cellular control s iRNA Knockdown of CTCF Cells were transfected with ON TARGET plus SMARTpool anti CTCF siRNAs (Thermo Scientific) according to manufacturer protocols. Briefly, either equimolar amounts of each of the 4 siRNAs comprising the S MARTpool or ON TARGET plus Non Targeting Control pool siRNAs (Thermo Scientific) were added to serum free, antibiotic free MEM supplemented with non essential amino acid and mixed gently. Dharmafect 1 transfection reagent (Thermo Scientific) was added to serum free, antibiotic free MEM supplemented with non essential amino acids and mixed gently. After a 5 minute incubation at room temperature, media containing the siRNA mixture was added to the media containing Dharmafect 1, mixed gently, and incubated a t room temperature for 20 minutes. Antibiotic free complete growth media was then added to this mixture yielding a final concentration of 25nM total siRNA (6.25nM of each respective siRNA construct). Growth media was decanted from Neuro2A cells cultured in 24 well dishes and the transfection media was added to the cells 6 hours post transfection, the transfection media was removed and complete growth medium supplemented with antibiotics was added to the cells. Knockdown was confirmed by western blot. Infections were performed 48 hours post transfection.
36 In Vitro Infection for siRNA E xperiments 48 hours post transfection with siRNAs, growth media was decanted and replaced with the appropriate dilution of virus in order to yield either a multiplicity o f infection of 3 pfu/cell or 0.2 pfu/cell, in an inoculum volume of 500 l. After a one hour incubation at 37C, the infection inoculum was removed and the cells were either harvested as the 0 hour time point, or complete growth media was replaced and cel ls were returned to the incubator until time of harvest. Isolation of DNA from Infected Cells for R eal time PCR A nalysis DNA was isolated from HSV 1 infected cells as described previously (80) Briefly, at the appropriate time post infection, growth media was removed and cells were harvested. Cells we pelleted by centrifugation at 16,000 x g at 4C for 40 min. Cells were resuspended in 200l of PCR Lysis buffer containing 10mM Tris HCl, 1mM EDTA, 0.001% Triton x 100 (vol /vol), 0.001% SDS (weight/vol ) and 50 mg/ml proteinase K (Fisher), then incubated at 50C overnight. Proteinase K was inac tivated by incubation at 94C for 20 minutes. Samples were then PCR ready. RNA Isolatio n for RT PCR and R eal time PCR A nalysis RNA from infected Neuro2A cells was isolated Trizol LS reagent followed by DirectZol RNA Miniprep kit (Zymo Research) according to manufacturer instructions Briefly, growth media was removed from infected cells and 200 l of Trizol LS was added directly to each well of the culture dish. Cells were then scraped off the dish and lysate was mixed by pipetting before being transfer red into a 1.5ml microcentrifuge tube. Samples were then processed with the DirectZol RNA Miniprep kit according to Trizol LS and mixed by vortexing. Samples were then ad ded to manufacturer supplied
37 microcolumns and centrifuged at 16,000xg for 1 minute, and the flow through discarded. Columns were then washed twice with manufacturer supplied RNA PreWash buffer, and once with RNA Wash Buffer by adding buffer to the column and centrifuging as before. Elution of RNA from miniprep column was performed using 25 l nuclease free water supplemented with SUPERase IN (Ambion) at a concentration of 1 U/l to prevent degradation of RNA. Isolated RNA was DNase treated utilizing TURB O DNA free (Ambion). Measurement of RNA concentration was performed on NanoDrop 2000 (Thermo) instrument. Reverse Transcription Reverse transcription was performed using random decamer priming (Ambion) and utilizing Omniscript reverse transcriptase (Qiag ins truction. Briefly, 100ng of DNa se treated RNA was added to reverse transcription reactions containing 10M random decamers, dNTP mixture and buffer supplied by the manufacturer, 1U/l SUPERase IN and Omniscript reverse t ranscriptase. Reactions were incubated at 37C for one hour. Tandem reactions were performed lacking Omniscript enzyme as no RT controls. Synthesized cDNA was analyzed directly by real time PCR. Calculation of Q uantity of DNA and mRNA For assessing the increase in DNA as infection progressed, relative quantities generated from real time PCR for HSV 1 pol (UL30) at each time point were recorded. Mean relative quantity for each time point was calculated, then mean relative quantity at each time point was normalize the relative quantity of pol detectible at 0 hours post infection. Thus, the data are presented as fold over input at the given time points. Each time point for each condition was repeated in replicates of 4 to 6.
38 Calculation of mRNA expressi on was performed as follows. Relative quantity reported from real time PCR analysis for cellular 18s rRNA and viral gene targets were collected for both reverse transcription reactions containing Omniscript and those without. Relative quantity of no RT c ontrol reactions was subtracted from relative quantity of RT reactions to obtain adjusted quantity, and viral targets are presented as adjusted quantity of viral gene targets divided by adjusted quantity of the cellular control. Table 2 1 Real time PCR Prim er and Probe Sequences Primer Forward Primer Probe Reverse Primer B1 CCCCGGCGGATTTTGTTG CATGCGTCGCCCAACC GCATGTGATCGTTGGGAATGAC B2 TGTGGTGCCCGTGTCTTTC ACTTTTCCCCTCCCCGACACG GCCCACTACACCAGCCAAT B5 TTTATTGCGTCTTCGGGTCTCA AAGCGCCCCGCCCC GCGGCGCGTTCGA B6 GCGGGAGTCGCAGAGG CCGACGCCGTCCGCT CGATGCGATCCCGATCCC B7 TGGCTGCTCCGCTAAAAGAC ACACGCGCGTCCTT AAGGCTGGGTGCAAATTGC UL20 CCATCGTCGGCTACTACGTTAC CCCGCACCGCCCAC CGATCCCTCTTGATGTTAACGTACA ICP0 CCGTGTGCACGGATGAGAT CTGCGCTGCGACACC CATGCACGGGATGCAGAAG ICP4 CACGGG CCGCTTCAC CCGACGCGACCTCC GCGATAGCGCGCGTAGA Tk CACGCTACTGCGGGTTTATATAGAC CACCACGCAACTGC CGTGGGACACCTTCAGCTT Pol AGAGGGACATCCAGGACTTTGT ACCGCCGAACTGAGCA CAGGCGCTTGTTGGTGTAC BSNEG CCTGAACAGCTTCCTCTTGGTTT CAACGTGATGCCCGAGGTC CAGGCCACTGCAGATTGG BS1 TCGGGAAA TCTGGTCTGACAAC ACTGCCACCGCCTCC GTCACTACGGGTATTGCATAATGTG Mouse MINE GACCTCCGCCCTCGTT AACGCTGTGGTCTCTG TGAAAGTAAAGTAAGTGTGCCCTCTAC Human MINE CAAAATCCAGCATAGCGATTGGTT CTCCCCGCGTTTGC TGCCTCCAGGCCTTTGC UPHOXA5 AGCAGCAGGGCCAATTCT CCCGCGATGCACCC GCTGCCCAAGCC AGCTT
39 C HAPTER 3 RESULTS CTCF Binds t o t he HSV 1 Genome Early During Lytic Replication i n Neuro2A Cells CTCF has been previously demonstrated to bind to the HSV 1 genome during latency (51) As interplay with CTCF has been demonstrated for other herpesviruses (reviewed in Chapter 1), the following experiments were designed to determine if CTCF binds to the HSV 1 genome during lytic replication using ChIP. To address the biological significance of any observed bindi ng to viral gene regulation, two approaches were applied. The first, a genetic approach, examined CTCF binding in a virus mutated in ICP4, which has an abrogated transcriptional program that does not progress past IE expression (81) Secondly, CTCF was depleted from the host cell by siRNA, and viral DNA and mRNA were analyzed using real t ime PCR. In the first set of experiments, a mouse neuroblastoma cell line, Neuro2A, was infected at an MOI of 3 with HSV strain 17syn+ and the infected cells harvested at 0, 3, 6, and 9 hours post infection and chromatin extracted. The isolated chromatin was analyzed by ChIP for CTCF enrichment at five CTCF binding sites on the HSV 1 genome dubbed B1, B2, B5, B6, and B7 ( Figure 1 4 ) (51) and one viral negative control, UL20 (Figure 3 1). At 0 hpi, all of the CTCF bi nding sites assayed showed modest levels of enrichment of CTCF, ranging from ~2.84 fold to ~5.07 fold over UpHoxA5. Of the five CTCF binding sites, B1 appeared to have slightly higher enrichment. also enriched at roughly equivalent levels to the CTCF binding sites, despite being thousands of base pairs away from the nearest characterized CTCF binding site By 3 hpi, after immediate early gene expression is well underway and when early gene expres sion is beginning,
40 enrichment at B1 decreases to ~61.5% (~1.62 fold decrease) of the levels at 0 hpi and enrichment at B2 also decreased by a smaller margin to ~87 6% (~1.14 fold decrease). The other HSV 1 loci assay, however, increased slightly in enric hment; the increases ranged from ~7.4% increase at B7 to ~31.4% increase at B6. At 6 hpi, enrichment at the loci assayed has decreased to between ~ 29 6% at B1 (~3.78 fold decrease) and ~57.1% at B7 (~1.75 fold decrease) of the enrichment seen at 0h A similar degree of enrichment is observed by 9 hpi, with the ratios being essentially identical to those observed at 6 hpi. By 9 hpi, enrichment at all sites assayed is further decreased, with observed decreases in enrichment compared to 0 hpi ranging fr om 4.75 fold at B1 to as much as 9.22 fold at B6, with enrichment of all but the B1 site being either at or below that of the cellular negative control. T hese data indicate that there is an overall decrease in enrichment of CTCF at the previously characte rized CTCF insulator sites on the HSV 1 genome over the course of the lytic infection. When infection was carried out in Neuro2A cells at a lower multiplicity, an MOI of 0.2, an interesting result came forth (Figure 3 2). At 0 hpi, the mean enrichment, compared to the higher multiplicity experiment (Figure 3 1), was greatly increased, with the mean enrichment ranging from 12.0 to as high as 30.3 fold compared to UpHoxA5. As in the case of the high multiplicity experiment though, B1 remained the most enr iched viral target examined, and enrichment at the UL20 site was observed at levels comparable to most of the viral CTCF binding sites analyzed. As infection progressed to 3 hpi, the overall enrichment of CTCF on the HSV 1 genome decreased, much like in t he high multiplicity study. The enrichment at the B1 site remains higher than other targets analyzed. The enrichment at 3 hpi, compared to 0 hpi ranges from ~54% (~1.85
41 fold decrease) at B5 to ~75% at UL20 (~1.33 fold decrease). At 6 hpi, the overwhelm ing majority of enrichment is no longer observed. Unlike the high multiplicity experiment, enrichment above the level of cellular control can be measured here. Compared to 0 hpi, enrichment at 6 hpi ranges from ~10.3% (~9.70 fold) at B5 to ~19.5% (~5.13 fold) at UL20 By 9 hpi, as was seen in the MOI 3 samples, the majority of regions assayed for enrichment showed less enrichment than the cellular negative UpHoxa5 (B2, B5, B6, B7). The B1 and UL20 regions remained slightly more enriched than the cellula r negative control at 2.6 and 1.4 fold over UpHoxa5, respectively. In the low multiplicity infections, after 9 hpi, enrichment reached between 4.4% (~22.6 fold decrease) and 8.6% ( ~ 11.6 fold decrease) at B6 and B1 respectively. As was observed in the MOI 3 experiment, enrichment at UL20 remained comparable to that at the CTCF binding sites investigated. In comparison low multiplicity infection yielded an enrichment of CTCF on the genome greater than high multiplicity infection which ranged from ~7.6 fo ld greater at B2 up to ~17.1 fold greater at B6. Low multiplicity infection also differed from high in which regions underwent the greatest reduction in CTCF enrichment over time, as B2 and B6 both decreased to just ~4.9% of their initial enrichment, wher eas the greatest change in a high multiplicity infection occurred at B1, which dropped to ~27.6% of the enrichment observed at 0 hpi. Two conclusions can be wrought from the data presented in these experiments: first, CTCF associates with the genome very q uickly after infection in an apparent nonspecific fashion, and second, this enrichment decreases markedly over time. Potentially, the cell is organizing the incoming genetic material. The apparent non
42 specificity of this association, as demonstrated by t he precipitation of UL20 from these samples, could be explained in a number of ways. The UL20 signal could represent background for the assay; that it is truly nonspecific and the ChIP is very low efficiency This is argued against by the cellular contr ols, and that UL20 is detected as being enriched compared to the cellular negative region UPHOXA5 and by the ChIP validation as well. It is also possible that CTCF, which has a high polymorphic binding sequence, is binding to an unidentified DNA element n ear to the UL20 primer probe binding site, indicating a specific precipitation at the UL20 locus. Thirdly, CTCF could be binding the genome and orienting it into a complicated three dimensional structure, which brings the UL20 region in close proximity to CTCF enriched areas din three dimensional space. This could mean during crosslinking, this region may be linked to areas of CTCF enrichment, and thus precipitated during ChIP. This possibility will be further elaborated on in the discussion chapter In summary, CTCF is enriched on the HSV 1 genome very early in the infection process, but to a much greater degree in a low multiplicity infection as compared to a high multiplicity infection. In general, CTCF enrichment on the HSV 1 genome decreases over t ime, but not equally: some of the binding sites in each multiplicity infection showed a greater decrease than others. This suggest s that CTCF mediated regulation of some of these loci is specific in a multiplicity dependent fashion. However, the ChIP exp eriments lack the ability to discern biological function, and assessing the specificity of the precipitation would lend validity to this hypothesis.
43 CHIP of CTCF is Capable o f Det ecting Established Differences o f Enrichment i n a n Analogous System It becam e clear that the HSV 1 viral internal negative control region was either precipitating with CTCF or represented a very high background for the assay. In order to assess the validity of the immunoprecipitation, an experiment was devised in a heterologous b ut related system. associated Herpesvirus (KSHV) can be maintained as a Bacterial Artificial Chromosome or BAC in 293T cells (82) KSHV, like HSV 1, has CTCF binding sites within its genome; indeed this appears to be a hallmark of herpesviruses (62, 83 85) This system is highly advantageous as the cells are very simple to culture to l arge volume, all cells contain comparable numbers of the KSHV genomes, and there is a large and thorough literature base characterizing the binding of CTCF to these KSHV BACs. Upstream of the gene Orf 73, which encodes the Latency Associate Nuclear Antige n (LANA), there are three CTCF binding sites known to be occupied during both KSHV latency and quiescence in BACs. Additionally, in the experiments characterizing these binding sites, a region of very low CTCF enrichment was discovered (62, 63) BAC16 is a KSHV BAC constructed in response to the discovery that the widely used BAC36 contained a large gene duplication in the terminal repeat region (86) The design of this experiment was as follows: ChIP was performed for CTCF in 293T cells stably transfected with BAC16, and assess the ability to recapitulate the known findings. ChIP and real time PCR analysis of these sites were performed to determine the specificity and resolution of the CTCF precipitation.
44 Differences in CTCF Enrichment are Readily Detectable on B AC 16 DNA in 293T Cells Previous experiments demonstrated that enrichment of CTCF on the HSV 1 genome during lytic infection in cell culture, even though several of the HSV CTCF sites were more abundantly enric hed in CTCF than others, the viral negative control UL20 was enriched despite being thousands of bases away from a CTCF binding motif. In this set of experiments, we sought to confirm the specificity of the CTCF precipitation utilizing the heterologous BA C16 system described above. Thus, determining regions where purported high and low enrichment of CTCF was relatively straightforward. The 293T BAC16 ChIPs were validated by comparing the MINE upstream of c Myc to that of the 18s rRNA gene, an average fold enrichment of 3.2 of MINE/c Myc enrichment; these ChIPs validated (Figure 3 3 A). The validated ChIPs from 6 independent experiments were then analyzed using two primer probe sets designed against BAC16. The first primer set BS1, lies upstream of ORF73 a nd is known to be enriched for CTCF in this system. The second primer set, called BSNEG, amplifies a region within KSHV gene ORF34, and has been established to be barren of CTCF (Figure 3 3 B). Real time PCR analysis of precipitated chromatin from 293T BAC16 cells shows a marked enrichment as measured with the BS1 primers, 17.55 fold over 18s. Enrichment as measured by BSNEG, however, shows enrichment barely over that of the cellular negative control, just 1.42 fold over 18s. Comparison between BS1 and BSNEG shows that a difference of more than 10 fold enrichment can be readily detected using this CTCF ChIP This ChIP is in accordance with the previously established high levels of enrichment at the CTCF sites in the Orf 73 promoter region on BACs maint ained in cell culture compared to the negative control
45 region though these reports have shown a much larger difference, nearly ~100 fold (63) Enrichment in this prior work was calculated as fold over IgG isotype control so the differences between these experiments and previously published d ata may be explained by differences in the methodology. CTCF is Enriched on the HSV 1 Genome at Low Levels in 293T C ells Chromatin from 293T BAC16 cells could then be used as a powerful tool while investigating CTCF enrichment on the HSV 1 genome. By infecting normal 293T cells in a manner parallel to previous experiments and then supplementing in a small percentage of chromatin from 239T BAC16 cells prior to immunoprecipitation, the BAC16 targets could be analyzed in tandem with the HSV 1 targets. By performing this 1 data yielded and provide us an o ther system in which we could study CTCF binding events to the HSV 1 genome during lytic infection. In order to confirm the ability to specifically precipitate HSV 1 DNA from infected cells in culture, 293T cells were infected with HSV 1 at a multiplicity of 0.2. This multiplicity was chosen as in previous experiments, enrichment of CTCF on HSV 1 was highest at lower multipli city. Samples were harvested at 0, 3, 6, and 9 hpi, as before, and analyzed by real time PCR (Figure 3 4). Strikingly, when compared with enrichment of CTCF in Neuro2A cells (Figure 3 1), enrichment on the HSV 1 genome is quite low, even at time 0. Inde ed even at 0 hpi, when enrichment was highest in Neuro2A cells, the majority of CTCF binding sites in the HSV 1 genome are enriched at levels comparable to both the cellular negative control and the KSHV negative control region. The B1 and B5 regions both appear to be somewhat enriched for CTCF at 0 hpi, whereas enrichment at B6 is actually below that of the cellular negative control. The B6 and B7 regions also show low levels of
46 enrichment, both roughly two fold over cellular negative control. Consistent with previous results, UL20 is comparably enriched to the CTCF binding sites in the HSV 1 genome. An additional observation can be made that the KSHV negative site, BSNEG, is also similarly enriched to the majority of HSV 1 genome targets, while BS1 show s marked enrichment. By 3 hours post infection in 293T cells, a slight increase in enrichment can be observed at the B1 and B5 regions, despite the remainder of the HSV 1 targets changing very little. The B5 remains more enriched than other HSV 1 targets analyzed. At 6 hpi, we observed a familiar phenomenon: enrichment of CTCF on the HSV 1 genome plummeted. In stark contrast to work in Neuro2A cells, however, HSV 1 site B5 remained enriched at this time point, at 2.59 fold enriched compared to 18s. The B7 region, as well, showed some modest enrichment at this time point. By 9 hpi, enrichment at all HSV 1 targets except B5 is at or below the level of the cellular negative control. The external controls in this experiment, the BAC16 targets BS1 and BSNE G, demonstrated that the precipitations were effective in demonstrating known enrichment patterns both from the previous experiment and published data This experiment successfully validated the ChIP assay in determining CTCF enrichment both on the KSHV B AC and on the HSV 1 genome. It also demonstrated that there may be a difference in the amount of CTCF that can be assembled during a lytic infection as compared to a quiescent or latent infection. From this it could be extrapolated that th e levels of CTC F enrichment on the HSV 1 genome during lytic infection would be lowe r than that of latent infection, and indeed recent reports suggest that enrichment during latency in sensory neurons can approach ~100 fold greater than cellular negative controls (87) Thus, CTCF is enriched on the HSV 1 genome during
47 lytic replication at a lower lev el than that observed during latency, but the experiments thus far have yet to establish regulatory significance for the levels of enrichment discovered at the CTCF binding sites on the HSV 1 genome. A Viral Mutant w ith a n Abrogated Transcriptional Program Exhibits Altered CTCF E nrichment The experiments so far have demonstrated that CTCF is indeed enriched on the HSV 1 genome, and the enrichment wanes to barely discernable levels as infection progresses. This decreasing enrichment begins early in infect ion, by 3 hpi, when IE class genes are robustly expressed at E class of genes can be detected. It was unclear at this point what the mechanism for evacuation of CTCF from the HSV 1 genome during infection may be. At cellular loci, control of CTCF occupan cy of binding elements is executed in many different ways. One such way is through transcription, in some instances, can be intimately involved in the process of flux in CTCF occupancy, both as a consequence and as a driving force behind eviction. For exa mple, the lysozyme locus in the chicken contains a CTCF binding element that is nearly constitutively bound, unless the cell is stimulated with lipopolysaccharide (88) Stimulation induces a massive chromatin remodeling event in which nucleosomes are repositioned into the CTCF bindi ng site, and CTCF is evicted. Moreover, it was shown that this eviction process required to stimulate gene expression from this locus is dependent directly on transcriptional elongation. The HSV 1 transcriptional program is well characterized. Catalysis o f viral transcription is mediated by a protein carried into the infected cell in the tegument of the virion known as VP16, which, upon entering the cell, translocates to the nucleus, binds
48 to cell transcription factors Oct 1 and HCF 1, and binds to the pro moter elements of HSV 1 immediate early genes (see for more details see Chapter 1) (89) These immediate early genes include ICP4, which is a critical transactivator for downstream gene expressi on of viral genes in the early class which contain the machinery necessary for DNA replication. ICP4 deletions are lethal and viruses containing inactivating mutations of ICP4 must be propagated on complementing cell lines (22) The following experiments were performed in order to establish a role for viral transcription in the perceived eviction of CTCF on the HSV 1 genome. A genetic approach was utilized by analyzing an ICP4 null virus known as KD6 in tandem with its parent wild type strain KOS (81) KD6 contains a large deletion in ICP4, rendering it functionless: the viral transcriptional program is greatly abrogated and the virus is unable to replicate. With these tools we investigated events seen early in infection, between 0 and 3 hours post infection of the cells, in order to focus on the potential role for viral transcription mediate d by ICP4 expression rather than the effect of DNA replication on the enrichment of CTCF on the viral DNA. In this experiment, early infection was the focus since in previous work, CTCF enrichment in Neuro2A cells had been highest at 0 and 3 hpi. Additio nally, the transcriptional defect in KD6 resulting from the ICP4 deletion occurs in this window as well. A low multiplicity was used in this experiment since in previous work in Neuro2A cells, a low multiplicity gave the greatest possibility for resolutio n between positive and negative enrichment. The experiment was performed as follows. Neuro2A cells were infected either with the ICP4 deletion virus KD6 or with the parental strain KOS at an MOI of 0.2. ChIP of CTCF was performed at 0 and 3 hours post in fection and enrichment on the HSV 1
49 genome was performed as before. It should be noted that since the ICP4 mutant used 1 strain KOS, which varies significantly from HSV strain 17+ used in the previous CTCF ChIP experiments, an analysis of CTCF binding was also performed using this wild type HSV 1 strain. ChIP from both KOS and KD6 infected Neuro2A cells validated readily (Figure 3 5 A and 3 6 A). KOS cells at 0 hpi displayed an enrichment pattern for CTCF that was distinct f rom that seen in previous experiments using HSV 1 strain 17+ (Figure 3 6 B). Most prominent is the robust enrichment of CTCF at B5, downstream of ICP4, especially when taken into account that the next most enriched target is less than half as enriched (B5 versus B1, 24.4 fold versus 10.35 fold, respectively). The next most enriched viral target is UL20, at 8.5 fold more enriched that UpHoxA5, and B2 and B6 have similar mean enrichments at 4.78 and 4.5 fold, respectively. At 3 hpi, enrichment on the KOS genome changed drastically compared to 0 hpi. Shown in Figure 3 5A, one of the most striking observations is that from sample to sample, the variability increases dramatically. Additionally, enrichment at all loci except for B1 increased as infection pro gressed. The pattern of mean enrichment has also changed: B7 now displays the highest level of enrichment at 19.16 fold over negative control which represents a 4.26 fold increase in enrichment. Enrichment of CTCF has dropped at B5 to enrichment of 17.77 fold above the cellular negative control. Enrichment increased at B2 and B6, enrichment at B6 more than doubled. B1, formerly second most enriched at 0 hpi, has comparatively little enrichment for CTCF, with a mean of just 3.38 fold over the cellular neg ative control. This represents a 3.1 fold decrease in the enrichment at 3 hpi compared to 0 hpi at B1. Lastly, enrichment at UL20
50 increased as well from 0 to 3 hpi by roughly half again as much This demonstrates a markedly different interaction with CT CF for strain KOS than for strain 17+ Most enriched at 0 hpi for 17 was region B1, which was the least enriched at the same time point in KOS. B5, which exhibited the greatest enrichment at 0 hpi in strain KOS displayed only modest enrichment in infecti on in strain 17+ when compared with other loci assayed. This is highly suggestive that perhaps different strains of HSV 1 have different relationships with CTCF during the lytic infection. Enrichment of CTCF as measured by ChIP on the KD6 genome at 0 hpi displays a similar pattern to that of its parent strain (Figure 3 6B). B5, which is proximal to but outside of the lesion in ICP4, remains the most enriched viral target assayed at 0 hpi, at 27.58 fold over UpHoxA5. B6 is the least enriched again at 8.2 7 fold over UpHoxA5, nearly identical to the enrichment at UL20 in this assay. B7 and B1 exhibit very similar enrichment as well, at 10.93 and 10.65 fold over UpHoxa5, respectively. B2, interestingly, in the absence of ICP4, appears to accumulate CTCF a t a levels higher than that of B1, in stark contrast to what was observed for the parent strain (Figure 3 5). KD6, at 3 hpi, bears much more resemblance to its respective 0 hpi enrichment than does strain KOS. Enrichment at the B1, B6, and B7 sites chan ged very little, decreasing slightly at B1 and B7, and increasing slightly at B6. Enrichment at B5 is also slightly decreased despite remaining the most enriched target of analysis on the viral genome, at 20.97 fold over upHoxA5. The most striking increa se, however, occurs at B2, which nearly doubled in enrichment from 13.52 fold over UphoxA5 to 21.71 fold. UL20 shows the least enrichment for CTCF at 3 hpi, dropping to just 5.45 fold over UphoxA5 This experiment highlights that there is an interplay b etween CTCF
51 enrichment on the HSV 1 genome and viral transcription as a transcriptional defect resulted in a markedly diffe rent enrichment pattern of CTCF, and that transcription may play a role in the remodeling of CTCF binding. Strain KOS exhibits a dis tinct enrichment pattern of CTCF at 0 hpi compared to strain 17+ In this experiment, the CTCF binding element immediately downstream of ICP4 displays robust enrichment compared to any other target measured, ranging from 2.35 fold greater enrichment than the next highest B1 to 14.8 fold greater enrichment than B6 which is barely more enriched for CTCF than the cellular negative control. It is an attractive hypothesis that perhaps this strain of HSV 1 has a specific recruitment pattern of CTCF during lyti c replication, but this would be argued by the enrichment pattern at 3 hours post infection, where enrichment is wildly variable and the majority of the CTCF binding sites assayed have comparable enrichment to that of UL20 In strain KOS at 3 hours pos t infection, the least enriched viral targets are those which reside within the repeat long region of the genome, and encompass the LAT locus. In the absence of the major viral immediate early transactivator protein ICP4, however, CTCF enrichment on the viral genome appears more static over time. The 0 hpi enrichment pattern that in KD6 shows that B5, immediately downstream of ICP4 is greatly enriched for CTCF, a levels that are comparable to those seen in for both wild type strains KOS and 17+. This s hows that even in the presence of a large deletion in the proximal ICP4 region, CTCF enrichment is unperturbed. Enrichment at B6, however, which is on the opposite flanking side of ICP4 demonstrates enrichment comparable to that of UL20 but is greater tha n that observed at B6 in strain KOS However, with the terminal abrogation of the viral transcriptional program, enrichment of CTCF changes
52 very little by 3 hpi. Interestingly, B2, downstream of the LAT locus, is increased in enrichment at 3 hpi, and B5 decreases to where the two are enriched at nearly the same level. What is markedly absent in KD6 is the drastic changes in enrichment seen in the parent strain KOS. B1, and B7 are nearly unchanged, decreasing in mean enrichment, where B6 increases to th e point where the three regions are enriched almost equivalently. This suggests that in the absence of ICP4, the changes in CTCF enrichment seen in wild type KOS which takes place by 3 hours post infection cannot or does not occur. There are several pote ntial explanations for the differences observed in the flux of CTCF enrichment on KOS versus that of KD6. First, there may be an HSV 1 gene product downstream of ICP4 transactivation which is responsible for orchestrating how CTCF interacts with the HSV 1 genome. CTCF is known to be post translationally modified (59, 60) and perhaps part of the DNA metabolism machinery which remains unexpressed in KD6 infection has a role in modifying and removing CTCF from the genome. Another possible explanation is that the restru cturing of CTCF from the HSV 1 genome in this case is more intrinsically tied to the transcriptional machinery itself, much like in the case of the chicken lysozyme locus (88) However, this argument is complicated by the observation that in the wild type virus KOS CTCF enrichment a t the majority of loci assayed increased, as opposed to the observed decrease in previous experiments. Surveying strain KOS for CTCF enrichment at later time points, and correlating these changes with transcription in each strain, would be crucial for und erstand how this particular strain of virus copes with CTCF binding to the genome during replication.
53 Taken in concert, these data suggest that CTCF binding early in infection seems to be temporally linked to transcription and that by altering the transcri ptional program of HSV 1, the pattern of CTCF binding to the genome can be subsequently altered during lytic infection in culture. Depletion o f CTCF b y siRNA Knockdown Alters HSV 1 Replication a nd Gene Expression The experiments presented thus far in th is document have demonstrated that in neuronal like cells, CTCF is enriched on the HSV 1 genome early in infection, this enrichment wanes as infection progresses, and by altering the transcriptional program of HSV 1 utilizing a genetic approach, the decrea se in enrichment can be arrested. This suggests that viral transcription and CTCF binding to the HSV 1 genome may be interconnected. The progressive decrease of CTCF enrichment from the genome through HSV DNA replication would also suggest that perhaps C TCF occupancy on the HSV 1 genome is mutually exclusive to DNA replication. CTCF binding to herpesviral genomes has also been extensively studied in associated herpesvirus (KSHV), CTCF is known to bind to the KSHV genome at three binding sites directly upstream of Orf73 which encodes that Latency Associated Nuclear Antigen. This binding site was shown to form a chromatin loop which isolated the region of the KSHV genome whi ch is transcriptionally active during latency (63) Moreover, the LANA promoter is also brought into space proximal the Orf 50, which encodes the major switch protein RTA, hypothesized to poise th e RTA promoter or expression (subsequently driving reactivation). Indeed, when reactivation is induced either by ectopic expression of RTA or by stimulating with sodium butyrate, CTCF binding to
54 these sites in the latency region nearly completely disrupte d, and consequently the chromatin loops formed are also lost as lytic gene expression progresses. Though the process of reactivation from latency is not perfectly identical to a de novo infection, it appears that for herpesviral replication and lytic gene expression, CTCF must be removed from the genome. In order to determine whether the relatively low levels of CTCF binding to the HSV 1 genome during the early phases of the lytic infection plays a biological role, an siRNA approach was used to deplete CTC F from host cells prior to infection with HSV 1 strain 17syn+, at both a low and high multiplicities of infection. By measuring the amount of viral DNA by real time PCR analysis and the expression of viral genes at the mRNA level, we proposed that the bio logical contribution of CTCF to HSV 1 transcription and replication could be discerned. Viral DNA A ccumulates to H igher l evels in C ells D epleted for CTCF To determine the biological consequence of CTCF binding to the HSV 1 genome during lytic infection, Ne uro2A cells were treated with a pool of siRNAs directed against CTCF for 48 hours prior to infection. Knockdown was confirmed with western blot analysis of cellular lysates, and confirmed nearly total ablation of CTCF in the cell by western blot ( data not shown ). Cells were then infected with HSV 1 strain 17syn+ at a multiplicity of either 3 or 0.2. For cells infected at MOI of 3, lysates were collected at 0, 3, 6 and 9 hpi, in order to encompass a single round of replication. For the low multiplicity experiment, lysates were collected at 1, 6, 12 and 24 hpi, encompassing 2 3 rounds of replication. To assay increases in viral DNA, lysates were analyzed by real time PCR to detect the viral pol gene, UL30.
55 In Neuro2A cells treated with non targeting siR NAs and infected at a high multiplicity (MOI = 3), viral increases in DNA are detected at 6 hours post infection (Figure 3 7). There appears to be no increase by 3 hours post infection, as would be expected since early gene expression would have only just begun by this time. By 6 hpi, the amount of viral DNA pol signal is almost 3 times that of the input displayed at 0 hpi. By the latest time point measured in this assay, the quantity of HSV 1 DNA as measure by PCR is nearly 7.97 fold greater than that o f the input levels as determined by 0 hpi Cells infected at a multiplicity of 3 treated with siRNAs against CTCF, however, display a different magnitude to the increasing viral DNA over time (Figure 3 8). Again at 3 hpi, little increase in viral DNA i s detected. However, by 6 hpi, a dramatic increase in viral DNA can be observed, reaching levels 5.2 fold over input. By 9 hours, the difference between cells treated with non targeting siRNA and those treated with siRNAs targeting CTCF becomes the most pointed. By 9 hpi, viral DNA reached levels 14.1 fold greater than that of input DNA, indicating that over a 9 hour replicative cycle at this multiplicity, knock down of CTCF yields a nearly two fold increase in the amount of viral DNA produced. The low MOI study provided similar results. Cells transfected with non targeting siRNAs showed an increase at of viral DNA to about 2 fold compared to input. By 12 hours post infection, samples displayed 8.5 times as much viral DNA. At 24 hours post infection, after seve ral rounds of viral replication, non targeting siRNA treated cells exhibited a 39.4 fold increase in viral DNA over 0 hpi. In cells with depleted CTCF, at 6 hpi we see an increase of 2.54 fold over input. Viral DNA in these samples continues to increase by 12 hpi, by which point levels have reached 14.1 fold over input levels at 0
56 hpi. By 24 hours post infection, however, the amount of DNA pol signal has reached a level 76.7 time that of 0 hpi, indicating a massive increase in the amount of viral DNA in the sample. At a high multiplicity, depletion of cells of CTCF appears to increase replication in a 9 hour infection period by nearly 2 fold compared to cells treated with non targeting siRNAs. The anti CTCF siRNA treatment seemed to begin to accelerate the accumulation of HSV 1 DNA by 6 hours post infection, reaching 1.7 fold more DNA in CTCF depleted cells versus non targeting treated cells. By the end of a single round of replication, this difference remains nearly consistent, which suggests that perh aps overcoming the early binding of CTCF moves the time table for accumulation of viral DNA forward. Low multiplicity infection coupled with depletion of CTCF by siRNA corroborates this finding. In the lower multiplicity infection, the difference in non t argeting siRNA treated cells and cells depleted for CTCF is less pronounced at 6 hpi, treated cells exhibiting just 1.36 fold more DNA than non targeting treated cells. By 12 hpi, the difference is much more pronounced. Cells treated with siRNA displayed a 14.1 fold increase in DNA compared to 0 hpi, whereas non targeting siRNA treated cells showed just an 8.5 fold increase in DNA. The difference of siCTCF treated cells compared to non targeting at this point is again about 1.7 fold. By 24 hpi, the diffe rence in viral DNA present between treatments increases further still. Compared to non targeting siRNA treated cells, cells treated with CTCF siRNA showed nearly 2 fold more HSV 1 viral DNA at the latest time point In summary, depletion of CTCF by siRN A treatment yielded an increase in viral DNA indicating perhaps that CTCF is inhibitory to infection.
57 As demonstrated in previous experiments, altering the transcriptional program of HSV 1 altered the CTCF enrichment on the genome, then depletion of CTCF may consequently effect viral transcription. Viral G ene E xpression after D epletion of CTCF by siRNA Analysis of HSV 1 mRNA levels was performed using RT qPCR for genes from each temporal class of expression: ICP0 and ICP4 from the IE class, thymidine kinas e from the E class, and UL20 from the L class of genes. Viral gene targets were normalized to the cellular 18s rRNA. (Figures 3 8 and 3 9) At MOI of 3 as show in in Figure 3 8, ICP0 expression can be detected at very low levels at 0 hpi in Neuro2A cells treated with both non targeting and anti CTCF siRNAs, indicating that transcription of ICP0 is initiated quickly after the HSV 1 genome reaches the nucleus of the infected cell. Even at this early time point, there is a difference between the amounts of ICP0 between treatments: cells depleted for CTCF have accumulated roughly 60% as much ICP0 mRNA as cells treated with non targeting siRNA. This deficit of ICP0 mRNA remains consistent throughout the course of the infection, CTCF depleted cells displaying between 50 65% the amount of ICP0 as non targeting treated cells. ICP4 transcription also shows a differential pattern when CTCF is knocked down prior to infection. ICP4 is undetectable at 0 hpi, but by 3 hpi ICP4 is expressed in both treatments, at whic h point ICP4 mRNA has accumulated to only 78% of the level of non targeting siRNA treated cells. By 6 hpi, the ICP4 level in siRNA treated cells is just 68% of that than in cells transfected with non targeting siRNAs. By 9 hpi, ICP4 transcript detectible in siRNA treated cells has dropped to just 50% of that of non targeting siRNA treated cells. Expression of thymidine kinase can be detected in Neuro2A cells as early as 3 hpi. Knockdown of CTCF in infected cells prior to infection
58 had little effect on th e thymidine kinase mRNA levels at 0, 3, and 6 hpi. However, by 9 hpi, post replication, the levels of tk mRNA detectible in cells treated with anti CTCF siRNA are again decreased compared to non targeting siRNA treated cells. UL20 expression in siRNA tre ated cells shows a different trend than any of the previous transcripts analyzed. Detectible by 3 hpi, cells depleted for CTCF have roughly 25% more UL20 mRNA than non targeting siRNA treated cells, and this relative increase persists over the course of t he infection and reaches a difference of more than 30% by 9 hpi. In cells depleted for CTCF infected at a low multiplicity we see a slightly different pattern off change in mRNA detectible post infection (Figure 3 9). In this experiment, the effect on I CP0 transcription in siRNA treated cells was opposite of that in high multiplicity infection. At 0 hpi, ICP0 transcript was undetectable in either treatment, however by 6 hpi a difference could be discerned, as siCTCF treated cells showed 4.7 fold more IC P0 transcript. At 12 and 24 hpi, ICP0 levels in siRNA treated cells remained higher than those in non targeting treated cells, at 2.1 and 2.0 fold higher, respectively. ICP4 transcript levels, however, displayed a much different trend, as cells treated with siRNA against CTCF generally accumulated less ICP4. In this experiment, ICP4 transcript levels were beyond the limit of detection for RT qPCR until 12 hpi, at which point non targeting siRNA treated cells exhibited ~3 fold more ICP4 mRNA than cells t reated with siRNAs against CTCF. This 3 fold increase was carried through to the 24 hpi time point, despite both treatment groups displaying an increase in ICP4 transcription. Transcript levels for tk between treatment groups display a difference at 6 hp i, where anti CTCF siRNA treated cells show a 2.4 fold increase in tk transcript. The
59 difference between the two treatment groups, though, decreases as time post infection progresses, dropping to just 1.15 increase compared to non targeting treated cells at 12 hpi, and by 24hpi, amounts of tk transcripts detectible between treatment groups is equal. UL20 transcript in either treatment group at this MOI was not detectible until 12hpi, at which point non targeting siRNA treated cells show robust expression, were as siCTCF treated cells did not. However, by 24 hours, cells treated with siRNA against CTCF have recovered UL20 transcript levels comparable to those in non targeting siRNA treated cells Taken in concert, these data suggest that CTCF does indeed play a role in regulating HSV 1 transcription, and this role may be different depending on multiplicity of infection.
60 A B Figure 3 1 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected at MOI 3. A) Relative quantities of immunoprecipit ated DNA as measured by real rime PCR for cellular positive (MINE) and negative (UPHOXA5) controls. Means are presented as black bars B) Enrichm ent of CTCF as determined by ChI P at targets on the HSV 1 genome at 0, 3, 6 and 9 hours post infection Means are presented as height of bars and error bars represent standard deviation 0.004 0.001 0.000 0.002 0.004 0.006 0.008 MINE UPHOXA5 Relative Quantity N2A MOI 3 VALIDATIONS -1.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 0h 3h 6h 9h Relative Quantity B/B+U Normalized To UPHOXA5 B/B+U CTCF ChIP in Neuro2A cells, MOI 3 B1 B2 B5 B6 B7 UL20
61 A B Figure 3 2 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected at MOI 0.2. A) Relative quantities of immunoprecipitated DNA as measured by real rime PCR f or cellular positive (MINE) and negative (UPHOXA5) controls. Means are presented as black bars. B) Enrichment of CTCF as determined by C hI P at targets on the HSV 1 genome at 0, 3, 6 and 9 hours post infection. Means are presented as height of bars and err or bars show standard deviation 0.0032 0.0008 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 MINE UPHOX Relative Quantity N2A MOI 0.2 VALIDATIONS 0 5 10 15 20 25 30 35 40 45 0h 3h 6h 9h Relative Quantity B/B+U Normalized To UPHOXA5 B/B+U CTCF ChIP in Neuro2A cells, MOI 0.2 B1 B2 B5 B6 B7 UL20
62 A B Figure 3 3 CTCF Chromatin immunoprecipitation in 2932T cells harboring KSHV DNA maintained as BAC16. A) Relative quantities of immunoprecipitated DNA as measured by real rime PCR for cellular positive (MINE) and negative (18s) controls. Means are presented as black bars. B) Enrichm ent of CTCF as determined by ChI P at targets on BAC16 DNA Means are presented as height of bars and error bars show standard deviation. 0.0016 0.0005 0 0.0005 0.001 0.0015 0.002 0.0025 Mine 18s Relative Quantity 293T BAC16 VALIDATIONS 0 5 10 15 20 25 Viral Targets B/B+U Normalized to 18s B/B+U Enrichment of CTCF on BAC16 BS1 BSNEG
63 A B Figure 3 4 CTCF Chromatin Imm unoprecipitation in 293T cells. A) Relative quantities of immunoprecipitated DNA as measured by real rime PCR for cellular positive (MINE) and negative (18S) controls. Means are presented as black bars. B) Enrich ment of CTCF as determined by ChI P at targe ts on the HSV 1 genome at 0, 3, 6 and 9 hours post infection. Sheared chromatin from 293T BAC16 cells was adde d prior to immunoprecipitation and analyzed at BS1 and BSNEG as an external control. M eans are presented as height of bars and error bars represe nt standard deviation 0.0022 0.0007 0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 MINE 18s Relative Quantity 0 5 10 15 20 25 30 0h 3h 6h 9h Relative Quantity B/B+U Normalized to 18s B/B+U Enrichment of CTCF on the HSV 1 in 293T cells B1 B2 B5 B6 B7 UL20 BS1 BSNEG
64 A B Figure 3 5 CTCF Chromatin Immunoprecipitation in Neuro2A cells infected with HSV 1 strain KOS. A) Relative quantities of immunoprecipitated DNA as measured by real rime PCR for cellular positive (MINE) and negative ( UPH OXA5 ) controls. Means are presented as black bars. B) Enrich ment of CTCF as determined by ChI P at targets on the HSV 1 genome at 0 and 3 post infection. Means are presented as height of bars, and error bars show standard deviation. 0.0018 0.0004 0 0.001 0.002 0.003 0.004 Mine UPHOX Relative Quantity N2A KOS Validations 0 5 10 15 20 25 30 35 40 0h 3h Relative Quantity B/B+U Normalized to UPHOXA5 B/B+U Enrichment of CTCF on HSV 1 Strain KOS B1 B2 B5 B6 B7 UL20
65 A B Figure 3 6. CTCF Chromatin Immunoprecipitation in Neuro2A cells infected with HSV 1 ICP4 deletion virus KD6. A) Relative quantities of immunoprecipitated DNA as measured by real rime PCR for cellular positive (MINE) and negative (UPHOXA5) controls. Means are presente d as black bars. B) Enrichment of CTCF as determined by ChIP at targets on the HSV 1 genome at 0 and 3 post infection. Means are presented as height of bars, and error bars show standard deviation. 0.0015 0.0006 0 0.0005 0.001 0.0015 0.002 0.0025 Mine UPHOX Relative Quantity 0 5 10 15 20 25 30 35 0h 3h Relative Quantity B/B+U Normalized to UPHOXA5 B/B+U B1 B2 B5 B6 B7 UL20
66 Figure 3 7 Viral DNA in cells treated with siRNA against CTCF and infected at MOI 3. Neuro2A cells were either transfected with a non targeting control pool of siRNA (NT) or ON TARGET SMARTpool siRNA against CTCF (siCTCF) and then infected with HSV 1 at a MOI of 3. At 0, 3, 6 and 9 hours post transfect ion, DNA was harvested and v iral DNA was quantified by real time PCR for the viral pol gene UL30. Quantity of viral DNA detected was normalized to 0 hpi which is considered representative of input. Error bars show standard deviation. 0 2 4 6 8 10 12 14 16 18 0 3 6 9 12 Viral DNA Normalized to 0h Hours Post Infection Viral DNA in siRNA Treated Cells, MOI 3 NT siCTCF
67 Figure 3 8 Vira l DNA in cells treated with siRNA against CTCF and infected at MOI 0.2. Neuro2A cells were either transfected with a non targeting control pool of siRNA (NT) or ON TARGET SMARTpool siRNA against CTCF (siCTCF) and then infected with HSV 1 at a MOI of 3. A t 0, 6 12 and 24 hours post transfection, DNA was harvested and viral DNA was quantified by real time PCR for the viral pol gene UL30. Quantity of viral DNA detected was normalized to 0 hpi which is considered representative of input. Error bars show s tandard deviation. 0 20 40 60 80 100 120 140 0 6 12 18 24 30 Viral DNA Normalized to 0h Hours Post Infection Viral DNA in siRNA Treated Cells, MOI 0.2 NT siCTCF
68 Figure 3 9 Viral transcript levels in Neuro2A cells infected at MOI of 3 after treatment with either non targeting ( NT ) or anti CTCF siRNA. Viral mRNA amoun ts were normalized to 18s rRNA. Mean is shown as height of bars, and error bars show standard deviation. 0 500 1000 1500 2000 2500 3000 0h 3h 6h 9h Viral mRNA Normalized to 18s ICP0 NT siCTCF 0 100 200 300 400 500 600 700 800 0h 3h 6h 9h Viral mRNA Normalized to 18s ICP4 NT siCTCF -200 0 200 400 600 800 1000 1200 1400 1600 1800 0h 3h 6h 9h Viral mRNA Normalized to 18s tk NT siCTCF 0 50 100 150 200 250 300 350 400 0h 3h 6h 9h Viral mRNA Normalized to 18s UL20 NT siCTCF
69 Figure 3 10 Viral transcript levels in Neuro2A cells infected at MOI of 0.2 after treatment with either non targeting ( NT ) or anti CTCF siRNA. Viral mRNA amounts were normalized to 18s rRNA. Mean is shown as height of bars, and error bars show standard deviation 0 100 200 300 400 500 600 700 800 900 0h 6h 12h 24 Viral mRNA Normalized to 18s ICP0 NT siCTCF -20 0 20 40 60 80 100 120 140 160 0h 6h 12h 24 Viral mRNA Normalized to 18s ICP4 NT siCTCF -50 0 50 100 150 200 250 300 350 400 0h 6h 12h 24 Viral mRNA Normalized to 18s tk NT siCTCF -20 0 20 40 60 80 100 120 140 160 180 0h 6h 12h 24 Viral mRNA Normalized to 18s UL20 NT siCTCF
70 CHAPTER 4 DISCUSSION AND CONCLUSIONS The experiments described in this dissertation set out to determine if the insulator binding protein CTCF, which binds to the HSV 1 genome during latency as disc ussed in Chapter 1, plays a role in the lytic replication of the virus. This was assessed first by determining if CTCF was enriched on the HSV 1 strain 17syn+ genome during lytic infection utilizing ChIP. CTCF was found to bind to the CTCF binding eleme nts as well as to distal regions of the genome very early in the infection process, and this enrichment was greater in a lower multiplicity infection than in a higher multiplicity. The validity of the precipitation was confirmed with the cellular controls as well as the utilization of an external control utilizing KSHV BAC16 chromatin and infection in 293T cells, which also suggested that 293T cells were unable to assemble CTCF onto the genome as efficiently as Neuro2A cells. Subsequently, in order to ass ess the biological relevance of this binding, a viral mutant defective for transcription was discovered to have altered CTCF binding compared to its widl type parent strain KOS This experiment also showed that strain KOS and strain 17syn+ show differenti al patter n s of CTCF enrichment. Lastly, utilizing siRNA to deplete CTCF from the cell prior to infection yielded greater amounts of viral DNA during replication and altered the transcription in ways that were dependent on MOI. The Effect of Multiplicity o f Infection on CTCF A ssociation with the HSV 1 Genome Chromatin immunoprecipitation of CTCF on the HSV 1 genome shows that CTCF binds to the HSV genomes early in infection. The kinetics of this enrichment appear to be independent of the amount of incoming HSV 1 genetic material, as in both
71 high and low multiplicity infections, CTCF enrichment on the HSV 1 genome is highest at the 0 hour time point. The method of defining 0 hpi in this case is after a 1 hour adsorption incubation with the infection inoculu m, is enough time for the initial portion of infection to take place: most of the input genome s are already in the nucleus by the time the 0 h time point is collected. In these experiments, at the 0 hour time, CTCF enrichment was highest, and then enrichme nt waned as time passed. By 6 hours, when DNA replication would be progressing, enrichment of CTCF plummeted in both low and high multiplicity infections (Figure 4 1). By the time progeny virus would be produced, enrichment was at very low levels. The lo w enrichment on the HSV 1 genome during a high MOI infection as compared to a low MOI infection could potentially be explained by the cell maintaining a limited pool of CTCF available to associate with the HSV 1 genome, resulting in a titration free CTCF a nd lower enrichment with the increase in incoming viral genomes. Th ese data agree with data measuring histone association with the HSV 1 genome during lytic infection, where early association followed by waning enrichment was also observed (69, 70) There are two possible explanations for the transient association of CTCF with the HSV 1 genomes early in infection and its los s after replication, which are not mutually exclusive. In the first scenario, the initial enrichment of CTCF on the HSV 1 genome titrates all of the available or unbound CTCF inside the infected nucleus. This would explain the dramatic drop in CTCF enrichment a t 6 hours, when the nucleus is flooded with newly synthesized HSV 1 genomes, and the cell is unable to synthesize CTCF protein due to viral shut down of the host transcription and translation machinery. This explanation could not, however, explain the red uction of CTCF enrichment by 3 hours post infection, as DNA replication
72 is not yet underway A second possibility is that CTCF binding to the HSV 1 genome interferes with efficient replication of viral genomes and must then be removed prior to replication This, evacuation of CTCF occurs on the HSV 1 genome in a mechanism either coupled with the transcriptional program of the virus or that there is a gene product responsible for eviction of CTCF. opinion, work best in concert. At the initiation of infection, there is a limited available pool of free CTCF in the nucleus which can bind to the HSV 1 genome, and as viral transcription progresses more and more is evicted from the pre replicative DNA. Then as DNA replica tion begins, the remainder of CTCF is evicted from the template genomes, and the amount of DNA unenriched for CTCF is greatly increased as replication progresses. Due to highly polymorphic nature of the binding sequence for CTCF, this single protein is uni quely capable of regulating vastly disparate regions of chromatin in seemingly mutually exclusive ways via the formation of loop domains. Flux in CTCF binding, for example can correlate with a change in physical conformation of the chromatin that is intri nsically tied to the change in transcriptional status; restructuring of chromatin loops mediated by CTCF may permit or prevent transcription at the target locus. Thus, in the venue of HSV 1 infection, it stands to reason that CTCF function would be simila rly multifaceted. During latency in wild type HSV 1 (strain 17syn+), when transcription is almost exclusively limited to the repeat long encoding the LAT, CTCF is shown to be enriched on the genome in a specific fashion, its binding correlating with known boundaries between regions of active transcription and the proximal lytic genes which are transcriptionally inactive. CTCF occupancy on the HSV 1 genome during
73 latency also appears to be specific: in all cases studied, enrichment at CTCF binding motifs w as greater than that at lytic loci (51) During reactivation of HSV 1 in the mouse model, CTCF enrichment drops dramatically over the first hours post reactivation, considerably prior to when reactivation stimulated DN A replication could take place (87) With this extensive investigation into the events that can occur after HSV 1 DNA has established latency, the questions of the involvement of CTCF in the initial infection of the host cell had remained unanswered. The evidence in the studies presented here provides the first investigation of the initial events in HSV 1 infection that CTCF binds to HSV 1 DNA as it enters the nucleus prior to the activation of the viral transcriptional program (Figure 3 1 and 3 2). The data presented in Chapter 3 also correlates well with the observed eviction of C TCF following reactivation stimulus: CTCF occupancy of the HSV 1 genome may in fact not be compatible with efficient lytic gene expression. In fact, in the reactivation study, at 3 hours post stimulus, there was a depreciation of CTCF enrichment on the 1 7syn+ genome that parallels our observations during lytic replication in Neuro2A cells (87) CTCF is known to play roles in both potentiating and 1 biology is consistent with the cellular biology (52, 79, 90) As CTCF can be post translationally modified by cellular Pc2, which is an E3 family ligase of the ubiquitin related polypeptide group SUMO, which corr esponds to transcriptional repression (61) ICP0 has been shown to mediate proteosomal degradation of SUMOylated proteins in HSV 1 infection by recognizing SUMO via its C terminal domain and subsequent ubiquitination of the SUMO group (91) This suggests that if SUMO ylated CTCF is bound to the HSV 1 genome, ICP0 may be involved in
74 recognizing it and orchestrating its removal from the genome. It is no t know n however, if CTCF is degraded during HSV 1 infection. Were CTCF to be subjected to widespread degradation medi ated by ICP0, a change in the ability to reliably validate the ChIP via cellular controls would be expected, and this was not observed. This could potentially be explained by recruitment of largely SUMOylated CTCF to the HSV 1 genome as compared to the ce llular positive control, and thus the posited ICP0 mediated degradation of CTCF would effect CTCF bound to the HSV 1 genome preferentially This warrants the investigation of CTCF in viruses mutated in ICP0 which can be overcome viral transcriptional d efect as a result of mutation by infection at a high MOI (30) Determining the post translational modification status of the CTCF interacting with the HSV 1 genome and attempti ng to identify players in the removal of HSV 1 from the genome represent a next step in the process of understanding this interaction. The Relationship between HSV 1 Transcription and CTCF Occupancy The interplay between viral transcription and enrichmen t of CTCF was underscored by the studies of the transcriptionally defunct mutant KD6. When compared to its parent strain KOS during the course of the study presented within this document, the CTCF binding pattern in Neuro2A cells was remarkably static in the absence of E class gene expression. This suggests that either active transcription from the viral genome or perhaps a gene product downstream of ICP4 is responsible for orchestrating the global deregulation of CTCF that appears to occur prior to and d uring replication. These hypotheses are not mutually exclusive: potentially transcription of the immediate early genes is responsible for eviction of CTCF from the genome prior to the robust expression of ICP4, which is in the process of catalyzing E gene expression by the 3 hpi time point.
75 By utilizing siRNA to deplete cells of CTCF, this interaction between CTCF and viral transcription can be assessed from the other side of the equation. In both low and high multiplicity, transcript levels detectible be tween NT and CTCF targeted siRNA can be discerned. The striking difference, though, is how depletion of CTCF effected given transcripts between the differing multiplicities of infection. In the high multiplicity infection, both ICP0 and ICP4 genes stud ied showed a lower transcript abundance as a result of knockdown. Low multiplicity infection shows a different story: ICP0 accumulates to higher levels in siRNA treated cells compared to NT whereas ICP4 is lower in treated cells compared to NT This would suggest that CTCF plays a role to control the levels of IE transcript in different directions depending on amount of incoming genetic material, and its association with the genome helps to fine tune the transcription of immediate early genes. Early gene transcription here was largely unaffected by CTCF knockdown, as tk expression only differed between treatments late in a high multiplicity infection UL20 expression in cells depleted for CTCF at high multiplicity is consistent with L gene expressi on being coordinated with DNA replication, as it shows somewhat increased expression in knockdown conditions which also resulted in higher DNA replication. However during a low multiplicity infection, in which knockdown of CTCF also yielded more DNA than in NT treated cells, UL20 transcript accumulated to equivalent levels only at 24 hpi. It is a notable caveat of this set of experiments that the data are presented normalized to the cellular transcript 18s, rather than compared or normalize to the amoun t of HSV 1 DNA in the infection, which makes comparison between the multiplicities in terms of absolute quantities difficult. If gene expression from the HSV 1 genome is correlated with CTCF binding, this would suggest
76 that perhaps at differing MOI, the i nteraction differs: perhaps CTCF forms different inter or intra genomic loops in a low multiplicity infection, resulting in differing gene expression very early during the infection process Since enrichment is generally much higher in low multiplicity infection compared to high, perhaps the accumulation of CTCF at these loops changes the dynamics of these loop domains, or indeed it may change their shape entirely Investigating chromatin loops formed by the HSV 1 genome during low and high multiplici ty infection using 3C analysis could be used to investigate this hypothesis. The final experimental set presented forth in this series supports the hypothesis that deposition of CTCF on the genome is a process that must be overcome by the virus in order to replicate. By depleting the nucleus of CTCF, HSV 1 is able to achieve a toehold in the nuclear architecture slightly ahead of schedule. This allows viral DNA replication to take place to a higher level and apparently begin earlier, which stands to reason as a mitigating factor no long has to be removed. Strain Differences in CTCF Enrichment When comparing the two different laboratory passaged HSV 1 wild type strains used in these experiments, strain KOS appears to bind CTCF with a different pattern than does strain 17syn+ at very early times post infection, establishing CTCF directly downstream of ICP4 at higher levels than other loci assayed. What is most curious when comparing these two strains of HSV 1 is that KOS, unlike 17syn+, displays an overall i ncrease in the enrichment of CTCF from 0 hpi to 3 hpi. This suggests that despite these two strains both being classified as wild type virus, they exhibit vastly different replication mechanics. Indeed, these two strains have long noted differences in th e animal models. Strain 17syn+ is highly virulent and transcribes the LAT more
77 abundantly KOS, however, is attenuated: orders of magnitude more virus can be utilized in mouse infections without causing mortality and KOS reactivates inefficiently from lat ency in animal models (92) This certainly suggests that there is basic virology yet to be established in truly differentiating these two strains. KOS has yet to be investigated in latent bind ing of CTCF as well, and a comprehensive investigation of the way in which these two strains of HSV 1 interplay with CTCF is warranted. It would be beneficial to again compare strain 17syn+ and KOS in their ability to remove CTCF and examine regions of gen ome differences in an attempt to identify potential viral factors which may take an active role in the removal of CTCF, or important cis elements that may be involved which may differ between the two strains. Cell Type Differences in Enrichment of CTCF on HSV 1 Employing the BAC16 system to assist in the investigation of CTCF enrichment on the HSV 1 genome elucidated the potential that perhaps different cell types interact differently with the HSV 1 genome. Latency of HSV 1 is only possible in a highly res tricted cell type: sensory neurons. When infecting 293T cells with HSV 1, the enrichment was quite low compared to that of Neuro2A cells. While neither 293T cells nor Neuro2 A cells are truly normal cell s it is conspicuous that the more neuronal like cel l displayed higher enrichment of CTCF on the HSV 1 genome. It is also known that neuronal cells replicate HSV less efficiently and produce less virus than other cells, suggesting perhaps a slightly more repressed phenotype of HSV 1 infection (93) Thus, cells can be envisioned to exist on a spectrum displaying the capacity to repress the HSV 1 genome. Far at one end would be the ultimate site of HSV 1 repression of lytic genes, the sensory neuron, in which lytic gene expression is tightly controlled during latency. On the other end of the spectrum would be cells which support high levels of
78 HSV 1 replication, such as epithelial cells which are less equipped to ste m the tide of HSV 1 replication. It is important to note that even on the extremes of this seemingly diametrical opposition that it is not an all or nothing situation: there are latently infected neurons which express lytic genes albeit at low levels, and it is possible that even in tissue culture some cells harbor a less successful replicative cycle that those around them. Thus, even similar cells can be viewed as occupying a small range within the hypothetical gradient. The experiments presented in th is dissert ation indicate that th at Neuro2A cells and 293T cells occupy different areas of this spectrum, with 293Ts being closer to the permissive side and Neuro2A being closer to the repressed side (Figure 4 2). Previous studies have been limited to the u sage of animal models to study the interplay of CTCF and HSV 1. Sensory ganglia are a milieu of a variety of cell types. In studying latency or reactivation, one important caveat of the system is that the vast majority of the cells harvested in either a trigeminal ganglia or dorsal root ganglia are not latently infected with HSV 1 (94) This presents difficulty in attempting to study the role of differential cell types in the biology of HSV 1. However, when comparing the levels of enrichment observed in latently infected mouse sensory g anglia to that observed in replication in cell culture, despite the presence of a large number of cells not harboring latent HSV 1 DNA, enrichment appears to much greater compared to lytic replication. It is worth noting, however, that it is difficult to draw direct conclusions between these sets of experiments due to the utilization of different validation primer probe sets between investigators. In Chapter 3, the utilization of chromatin from the heterologous 293T BAC16 system as an external control allo wed for the verification that
79 this lower level of enrichment in lytic culture is a real result, and not artifact of the assay. Utilizing cell culture models to investigate enrichment of CTCF in an epithelial cell provides an opportunity to investigate tis sues otherwise not possible, such as that from humans. Despite the utility of animal models, the mouse model in particular is limited in its reproduction of disease in the native host. For example, as the mouse undergoes very little detectable spontaneou s reactivation comparatively. Investigation of CTCF enrichment in a differential tissue from neuronal cells demonstrated that perhaps part of the tissue specificity of HSV 1 latency may be the levels of initial deposition of CTCF. Model for the Function of Early Lytic CTCF Enrichment The experiments presented here suggest that CTCF plays a small determining role in the kinetics of HSV 1 replication in neuronal cells. The nucleus of the infected cell detects the incoming viral nucleic acid and attempts to assemble it into chromatin in order to gain control of expression of the genes encoded therein. In a lytic infection, the cells starts this arms race off at a disadvantage as tegument protein vhs has already begun to destroy host mRNA and VP16 has begun t o recruit host factors to catalyze transcription from the viral genome. This is a race the cell will ultimately lose. This suggests that cells that are more permissive to harboring/enforcing latent HSV 1 DNA (epithelial cells as opposed to neurons) may be capable of assembling chromatin factors onto the genome more effectively than others. How it is that the intracellular tug of war between the virus and cell plays out so differently in sensory neurons to result in latency? It would be worthwhile to corr oborate the findings of this document by comparing CTCF enrichment during lytic replication in a system where latency can be established, such as comparing the footpad epithelial cells to the DRG neurons in the mouse footpad model of latency. The mouse fo otpad model could further be pursued by
80 utilizing mice which contain a conditional knockout of CTCF, inducible by Cre mediated recombination. By co infecting with HSV 1 and AAV containing a Cre expressing transgene, the effect of depletion of CTCF from t he neuron on establishment of latency can be discerned. AAV could similarly be employed to deliver siRNA to the ganglia in wild type mice. Other models to study lytic replication of HSV 1 in neurons such as the trigeminal ganglia neuron culture system ma y also be useful in investigating this phenomenon further (95) In summary, the experiments described in th is dissertation provide evidence that CTCF robustly bind s to the HSV 1 genome early during infection of neuronal cells. This suggest that CTCF association with the genome may be a component of the cellular decision between lytic replication versus latency, and provide basis for continued investigation into the role of CTCF in other viral processes. Elucidation of the attempts by the cell to enforce chromatin on the HSV 1 genome can help widen the understanding the complicated process of latency establishme nt, and perhaps a return to latency after reactivation has occurred affording a greater understanding of HSV 1 biology.
81 Figure 4 1 Model of CTCF binding in Low and High Multiplicities. In Low multiplicity infection, the majori ty of HSV 1 genomes in nuclei are bound by CTCF, whereas in high multiplicity, the amount of free CTCF is titrated by the incoming HSV 1 genomes. As DNA replication progresses, the nucleus is flooded with viral DNA which is unbound by CTCF Low Multiplicity High Multiplicity DNA Replication
82 Figure 4 2 Potential to assemble CTCF onto incoming HSV 1 genetic material in order to establish chromatin and coordinate HSV 1 gene expression. Different cell types are differentially capable of depositing CTCF on the HSV 1 genome in an attempt to organize and potentially silence the HSV 1 genome. Sensory neurons are singularly capable of forcing HSV 1 into latency. More neuronal like cells such as Neuro2A have some potential, but are unable to stem lytic replication. 293T cells in contrast are le ss able to assemble CTCF on the genome and therefore have limited potential to silence HSV 1. Lytic (Permissive) Sensory Neurons Neuro2A Cells Threshold 293T cells
83 LIST OF REFERENCES 1. Davison AJ 2010. Herpesvirus systematics. Vet. Microbiol. 143 :52 69. 2. D avison AJ 2007. Overview of Classification I n : Arvin A, Campadelli Fiume G, Mocarski E, et al., editors Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge University Press, Cambridge. Chapter 1. Available from http://www.ncbi.nlm.nih.gov/books/NBK47406/ 3. Kinchington PR, Leger AJS, Guedo n J MG, Hendricks RL 20 12. Herpes simplex virus and varicella zoster virus, the house guests who never leave. Herpesv iridae 3 :5. 4. Liesegang TJ 2001. Herpes simplex virus epidemiology and ocular importance. Cornea 20 :1 13. 5. Feldman LT, Ellison AR, Voytek CC, Yang L, Krause P, Margolis TP 2002. Spontaneous molecular reactivation of herpes simplex virus type 1 laten cy in mice. Proc. Natl. Acad. Sci. 99 :978 983. 6. Jones C 2003. Herpes Simplex Virus Type 1 and Bovine Herpesvirus 1 Latency. Clin. Microbiol. Rev. 16 :79 95. 7. Xu F, Schillinger JA, Sternberg MR, Johnson RE, Lee FK, Nahmias AJ, Markowitz LE 2002. Sero prevalence and Coinfection with Herpes Simplex Virus Type 1 and Type 2 in the United States, 1988 1994. J. Infect. Dis. 185 :1019 1024. 8. Kaufman HE, Azcuy AM, Varnell ED, Sloop GD, Thompson HW, Hill JM 2005. HSV 1 DNA in Tears and Saliva of Normal Adult s. Invest. Ophthalmol. Vis. Sci. 46 :241 247. 9. Spear PG, Eisenberg RJ, Cohen GH 2000. Three Classes of Cell Surface Receptors for Alphaherpesvirus Entry. Virology 275 :1 8. 10. Spear PG, Longnecker R 2003. Herpesvirus Entry: an Update. J. Virol. 77 :101 79 10185. 11. 1993. The virion transactivator of herpes simplex virus. Semin. Virol. 4 :145 155. 12. Sodeik B, Ebersold MW, Helenius A 1997. Microtubule mediated transport of incoming herpes simplex virus 1 capsids to the nucleus. J. Cell Biol. 136 :1007 1021. 13. Boehmer PE, Lehman IR 1997. Herpes Simplex Virus Dna Replication. Annu. Rev. Biochem. 66 :347 384.
84 14. DeLuca NA, Schaffer PA 1985. Activation of immediate early, early, and late promoters by temperature sensitive and wild type forms of herpes simplex virus type 1 protein ICP4. Mol. Cell. Biol. 5 :1997 2008. 15. Fraser KA, Rice SA 2007. Herpes simplex virus immediate early protein ICP22 triggers loss of serine 2 phosphorylated RNA polymerase II. J. Virol. 81 :5091 5101. 16. Poon APW, Roizman B 2005. Herpes simplex virus 1 ICP22 regulates the accumulation of a shorter mRNA and of a truncated US3 protein kinase that exhibits altered functions. J. Virol. 79 :8470 8479. 17. Smith IL, Hardwicke MA, Sandri Goldin RM 1992. Evidence that th e herpes simplex virus immediate early protein ICP27 acts post transcriptionally during infection to regulate gene expression. Virology 186 :74 86. 18. Hardy WR, Sandri Goldin RM 1994. Herpes simplex virus inhibits host cell splicing, and regulatory prote in ICP27 is required for this effect. J. Virol. 68 :7790 7799. 19. Frh K, Ahn K, Djaballah H, Semp P, van Endert PM, Tamp R, Peterson PA, Yang Y 1995. A viral inhibitor of peptide transporters for antigen presentation. Nature 375 :415 418. 20. Neumann L, Kraas W, Uebel S, Jung G, Tamp R 1997. The active domain of the herpes simplex virus protein ICP47: a potent inhibitor of the transporter associated with antigen processing. J. Mol. Biol. 272 :484 492. 21. Easterfield AJ, Austen BM, Westwood OM 2001. Inhibition of antigen transport by expression of infected cell peptide 47 (ICP47) prevents cell surface expression of HLA in choriocarcinoma cell lines. J. Reprod. Immunol. 50 :19 40. 22. DeLuca NA, McCarthy AM, Schaffer PA 1985. Isolation and characteri zation of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate early regulatory protein ICP4. J. Virol. 56 :558 570. 23. Everett RD 2000. ICP0, a regulator of herpes simplex virus during lytic and latent infection. BioEssays 22 :7 61 770. 24. Boutell C, Sadis S, Everett RD 2002. Herpes Simplex Virus Type 1 Immediate Early Protein ICP0 and Its Isolated RING Finger Domain Act as Ubiquitin E3 Ligases In Vitro. J. Virol. 76 :841 850. 25. 2000. Interferon Coordinately Inhibits the Disruption of PML Positive ND10 and Immediate Early Gene Expression by Herpes Simplex Virus. J. Interferon Cytokine Res. 20 :805 815.
85 26. Lilley CE, Chaurushiya MS, Boutell C, Landry S, Suh J, Panier S, Everett RD Stewart GS, Durocher D, Weitzman MD 2010. A viral E3 ligase targets RNF8 and RNF168 to control histone ubiquitination and DNA damage responses. EMBO J. 29 :943 955. 27. Everett RD, Boutell C, Orr A 2004. Phenotype of a Herpes Simplex Virus Type 1 Mutan t That Fails To Express Immediate Early Regulatory Protein ICP0. J. Virol. 78 :1763 1774. 28. Stow ND, Stow EC 1986. Isolation and Characterization of a Herpes Simplex Virus Type 1 Mutant Containing a Deletion within the Gene Encoding the Immediate Early Polypeptide Vmw110. J. Gen. Virol. 67 :2571 2585. 29. Cai W, Schaffer PA 1992. Herpes simplex virus type 1 ICP0 regulates expression of immediate early, early, and late genes in productively infected cells. J. Virol. 66 :2904 2915. 30. Chen J, Silverstein S 1992. Herpes simplex viruses with mutations in the gene encoding ICP0 are defective in gene expression. J. Virol. 66 :2916 2927. 31. Mingo RM, Han J, Newcomb WW, Brown JC 2012. Replication of Herpes Simplex Virus: Egress of Progeny Virus at Specialize d Cell Membrane Sites. J. Virol. 86 :7084 7097. 32. Farrell MJ, Dobson AT, Feldman LT 1991. Herpes simplex virus latency associated transcript is a stable intron. Proc. Natl. Acad. Sci. 88 :790 794. 33. Thomas DL, Lock M, Zabolotny JM, Mohan BR, Fraser NW 2002. The 2 Kilobase Intron of the Herpes Simplex Virus Type 1 Latency Associated Transcript Has a Half Life of Approximately 24 Hours in SY5Y and COS 1 Cells. J. Virol. 76 :532 540. 34. Cui C, Griffiths A, Li G, Silva LM, Kramer MF, Gaasterland T, Wang X J, Coen DM 2006. Prediction and Identification of Herpes Simplex Virus 1 Encoded MicroRNAs. J. Virol. 80 :5499 5508. 35. Umbach JL, Nagel MA, Cohrs RJ, Gilden DH, Cullen BR 2009. Analysis of Human Alphaherpesvirus MicroRNA Expression in Latently Infect ed Human Trigeminal Ganglia. J. Virol. 83 :10677 10683. 36. Umbach JL, Kramer MF, Jurak I, Karnowski HW, Coen DM, Cullen BR 2008. MicroRNAs expressed by herpes simplex virus 1 during latent infection regulate viral mRNAs. Nature 454 :780 783. 37. Chen SH, Kramer MF, Schaffer PA, Coen DM 1997. A viral function represses accumulation of transcripts from productive cycle genes in mouse ganglia latently infected with herpes simplex virus. J. Virol. 71 :5878 5884.
86 38. Giordani NV, Neumann DM, Kwiatkowski DL, B hattacharjee PS, McAnany PK, Hill JM, Bloom DC 2008. During Herpes Simplex Virus Type 1 Infection of Rabbits, the Ability To Express the Latency Associated Transcript Increases Latent Phase Transcription of Lytic Genes. J. Virol. 82 :6056 6060. 39. Razin A, Riggs AD 1980. DNA methylation and gene function. Science 210 :604 610. 40. Zhang XY, Ehrlich KC, Wang RY, Ehrlich M 1986. Effect of site specific DNA methylation and mutagenesis on recognition by methylated DNA binding protein from human placenta. Nu cleic Acids Res. 14 :8387 8397. 41. Prendergast GC, Ziff EB 1991. Methylation sensitive sequence specific DNA binding by the c Myc basic region. Science 251 :186 189. 42. Nan X, Campoy FJ, Bird A 1997. MeCP2 is a transcriptional repressor with abundant binding sites in genomic chromatin. Cell 88 :471 481. 43. 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 :187 191. 44. Luger K, Mder AW, Richmond RK, Sargent DF, Richmond TJ 1997. Crystal 389 :251 260. 45. Jenuwein T, Allis CD 2001. Translating the Histone Code. Science 293 :1074 1080. 46. Deshmane SL, Fraser NW 1989. During latency, herpes simplex virus type 1 DNA is associated with nucleosomes in a chromatin structure. J. Virol. 63 :943 947. 47. Kwiatkowski DL, Thompson HW, Bloom DC 2009. The Polycomb Group Protein Bmi1 Binds to the Herpes Simplex Virus 1 Latent Genome and Maintains Repressive Histone Marks during Latency. J. Virol. 83 :8173 8181. 48. Cliffe AR, Garber DA, Knipe DM 2009. Transcription of the Herpes Simplex Virus Latency Associated Transcript Promotes the Formation of Facultative Heterochromatin on Lytic Promoters. J. Virol. 83 :8182 819 0. 49. Cliffe AR, Coen DM, Knipe DM 2013. Kinetics of facultative heterochromatin and polycomb group protein association with the herpes simplex viral genome during establishment of latent infection. mBio 4 50. Kubat NJ, Amelio AL, Giordani NV, Bloom D C 2004. The Herpes Simplex Virus Type 1 Latency Associated Transcript (LAT) Enhancer/rcr Is
87 Hyperacetylated during Latency Independently of LAT Transcription. J. Virol. 78 :12508 12518. 51. Amelio AL, McAnany PK, Bloom DC 2006. A Chromatin Insulator Like Element in the Herpes Simplex Virus Type 1 Latency Associated Transcript Region Binds CCCTC Binding Factor and Displays Enhancer Blocking and Silencing Activities. J. Virol. 80 :2358 2368. 52. Lobanenkov VV, Nicolas RH, Adler VV, Paterson H, Klenova EM, P olotskaja AV, Goodwin GH 1990. A novel sequence specific DNA binding protein which interacts with three regularly spaced direct repeats of the CCCTC flanking sequence of the chicken c myc gene. Oncogene 5 :1743 1753. 53. Kim TH, Abdullaev ZK, Smith AD, Ching KA, Loukinov DI, Green RD, Zhang MQ, Lobanenkov VV, Ren B 2007. Analysis of the Vertebrate Insulator Protein CTCF Binding Sites in the Human Genome. Cell 128 :1231 1245. 54. Gaszner M, Felsenfeld G 2006. Insulators: exploiting transcr iptional and epigenetic mechanisms. Nat. Rev. Genet. 7 :703 713. 55. Bell AC, West AG, Felsenfeld G 1999. The Protein CTCF Is Required for the Enhancer Blocking Activity of Vertebrate Insulators. Cell 98 :387 396. 56. Recillas Targa F, Bell AC, Felsenfeld G 1999. Positional enhancer blocking globin insulator in transiently transfected cells. Proc. Natl. Acad. Sci. 96 :14354 14359. 57. Phillips JE, Corces VG 2009. CTCF: Master Weaver of the Genome. Cell 137 :1194 1211. 58. Klenov a EM, Chernukhin IV, El Kady A, Lee RE, Pugacheva EM, Loukinov DI, Goodwin GH, Delgado D, Filippova GN, Len J, Morse HC, Neiman PE, Lobanenkov VV 2001. Functional Phosphorylation Sites in the C Terminal Region of the Multivalent Multifunctional Transcrip tional Factor CTCF. Mol. Cell. Biol. 21 :2221 2234. 59. El Kady A, Klenova E 2005. Regulation of the transcription factor, CTCF, by phosphorylation with protein kinase CK2. FEBS Lett. 579 :1424 1434. 60. Yu W, Ginjala V, Pant V, Chernukhin I, Whitehead J, Docquier F, Farrar D, Tavoosidana G, Mukhopadhyay R, Kanduri C, Oshimura M, Feinberg AP, Lobanenkov V, Klenova E, Ohlsson R 2004. Poly(ADP ribosyl)ation regulates CTCF dependent chromatin insulation. Nat. Genet. 36 :1105 1110. 61. MacPherson MJ, Beatty L G, Zhou W, Du M, Sadowski PD 2009. The CTCF Insulator Protein Is Posttranslationally Modified by SUMO. Mol. Cell. Biol. 29 :714 725.
88 62. Stedman W, Kang H, Lin S, Kissil JL, Bartolomei MS, Lieberman PM 2008. Cohesins localize with CTCF at the KSHV latenc y control region and at cellular c myc and H19/Igf2 insulators. EMBO J. 27 :654 666. 63. Kang H, Wiedmer A, Yuan Y, Robertson E, Lieberman PM 2011. Coordination of KSHV Latent and Lytic Gene Control by CTCF Cohesin Mediated Chromosome Conformation. PLoS P athog 7 :e1002140. 64. Darst RP, Haecker I, Pardo CE, Renne R, Kladde MP 2013. Epigenetic associated herpesvirus. Nucleic Acids Res. 41 :2993 3009. 65. Chau CM, Zhang X Y, McMahon SB, Lieberman PM 2006. Regulation of Epstein Barr Virus Latency Type by the Chromatin Boundary Factor CTCF. J. Virol. 80 :5723 5732. 66. Day L, Chau CM, Nebozhyn M, Rennekamp AJ, Showe M, Lieberman PM 2007. Chromatin Profiling of Epstein Barr Virus Latency Control Region. J. Virol. 81 :6389 6401. 67 Tempera I, Wiedmer A, Dheekollu J, Lieberman PM 2010. CTCF Prevents the Epigenetic Drift of EBV Latency Promoter Qp. PLoS Pathog 6 :e1001048. 68. Chen H S, Martin KA, Lu F, Lupey LN, Mueller JM, Lieberman PM, Tempera I 2013. Epigenetic Deregulation of the LMP1/LMP2 Locus of Epstein Barr Virus by Mutation of a Single CTCF Cohesin Binding Site. J. Virol. JVI.02209 13. 69. Kent JR, Zeng P Y, Atanasiu D, Gardner J, Fraser NW, Berger SL 2004. During Lytic Infection Herpes Simplex Virus Type 1 Is Associate d with Histones Bearing Modifications That Correlate with Active Transcription. J. Virol. 78 :10178 10186. 70. Oh J, Fraser NW 2008. Temporal Association of the Herpes Simplex Virus Genome with Histone Proteins during a Lytic Infection. J. Virol. 82 :3530 3537. 71. Oh J, Ruskoski N, Fraser NW 2012. Chromatin Assembly on Herpes Simplex Virus 1 DNA Early during a Lytic Infection Is Asf1a Dependent. J. Virol. 86 :12313 12321. 72. Poon APW, Liang Y, Roizman B 2003. Herpes Simplex Virus 1 Gene Expression Is A ccelerated by Inhibitors of Histone Deacetylases in Rabbit Skin Cells Infected with a Mutant Carrying a cDNA Copy of the Infected Cell Protein No. 0. J. Virol. 77 :12671 12678. 73. Hagglund R, Roizman B 2004. Role of ICP0 in the Strategy of Conquest of th e Host Cell by Herpes Simplex Virus 1. J. Virol. 78 :2169 2178.
89 74. Gu H, Liang Y, Mandel G, Roizman B 2005. Components of the REST/CoREST/histone deacetylase repressor complex are disrupted, modified, and translocated in HSV 1 infected cells. Proc. Natl. Acad. Sci. U. S. A. 102 :7571 7576. 75. Poon APW, Gu H, Roizman B 2006. ICP0 and the US3 protein kinase of herpes simplex virus 1 independently block histone deacetylation to enable gene expression. Proc. Natl. Acad. Sci. U. S. A. 103 :9993 9998. 76. Gu H, Roizman B 2007. Herpes simplex virus infected cell protein 0 blocks the silencing of viral DNA by dissociating histone deacetylases from the CoREST REST complex. Proc. Natl. Acad. Sci. U. S. A. 104 :17134 17139. 77. Gu H, Roizman B 2009. Engagement of the Lysine Specific Demethylase/HDAC1/CoREST/REST Complex by Herpes Simplex Virus 1. J. Virol. 83 :4376 4385. 78. Chen Q, Lin L, Smith S, Huang J, Berger SL, Zhou J 2007. CTCF Dependent Chromatin Boundary Element between the Latency Associated Transcript and ICP0 Promoters in the Herpes Simplex Virus Type 1 Genome. J. Virol. 81 :5192 5201. 79. Gombert WM, Krumm A 2009. Targeted Deletion of Multiple CTCF Binding Elements in the Human C MYC Gene Reveals a Requirement for CTCF in C MYC Expression. PLoS ONE 4 :e6109. 80. Flores O, Nakayama S, Whisnant AW, Javanbakht H, Cullen BR, Bloom DC 2013. Mutational Inactivation of Herpes Simplex Virus 1 MicroRNAs Identifies Viral mRNA Targets and Reveals Phenotypic Effects in Culture. J. Virol. 87 :6589 6603. 81. Dobs on AT, Margolis TP, Sedarati F, Stevens JG, Feldman LT 1990. A latent, nonpathogenic HSV 1 galactosidase in mouse neurons. Neuron 5 :353 360. 82. Warden C, Tang Q, Zhu H 2010. Herpesvirus BACs: Past, Present, and Future. BioMed Res. Int. 2011 83. Bloom DC, Giordani NV, Kwiatkowski DL 2010. Epigenetic regulation of latent HSV 1 gene expression. Biochim. Biophys. Acta BBA Gene Regul. Mech. 1799 :246 256. 84. Takacs M, Banati F, Koroknai A, Segesdi J, Salamon D, Wolf H, Niller HH, Minarovits J 2010. Epigenetic regulation of latent Epstein Barr virus promoters. Biochim. Biophys. Acta 1799 :228 235. 85. Tempera I, Lieberman PM 2010. Chromatin Organization of Gammaherpesvirus Latent Genomes. Biochim. Biophys. Acta 1799 :23 6 245.
90 86. Brulois KF, Chang H, Lee AS Y, Ensser A, Wong L Y, Toth Z, Lee SH, Lee H R, Myoung J, Ganem D, Oh T K, Kim JF, Gao S J, Jung JU 2012. associated herpesvirus Bacterial artificial chromosom e clone. J. Virol. JVI.01019 12. 87. Ertel MK, Cammarata AL, Hron RJ, Neumann DM 2012. CTCF Occupation of the Herpes Simplex Virus 1 Genome Is Disrupted at Early Times Postreactivation in a Transcription Dependent Manner. J. Virol. 86 :12741 12759. 88. L efevre P, Witham J, Lacroix CE, Cockerill PN, Bonifer C 2008. The LPS Induced Transcriptional Upregulation of the Chicken Lysozyme Locus Involves CTCF Eviction and Noncoding RNA Transcription. Mol. Cell 32 :129 139. 89. Simmen KA, Newell A, Robinson M, Mi lls JS, Canning G, Handa R, Parkes K, Borkakoti N, Jupp R 1997. Protein interactions in the herpes simplex virus type 1 VP16 induced complex: VP16 peptide inhibition and mutational analysis of host cell factor requirements. J. Virol. 71 :3886 3894. 90. Vo strov AA, Quitschke WW 1997. The Zinc Finger Protein CTCF Binds to the Protein Precursor Promoter EVIDENCE FOR A ROLE IN TRANSCRIPTIONAL ACTIVATION. J. Biol. Chem. 272 :33353 33359. 91. Boutell C, Cuchet Loureno D, Vanni E, O rr A, Glass M, McFarlane S, Everett RD 2011. A Viral Ubiquitin Ligase Has Substrate Preferential SUMO Targeted Ubiquitin Ligase Activity that Counteracts Intrinsic Antiviral Defence. PLoS Pathog 7 :e1002245. 92. Thompson RL, Cook ML, Devi Rao GB, Wagner E K, Stevens JG 1986. Functional and molecular analyses of the avirulent wild type herpes simplex virus type 1 strain KOS. J. Virol. 58 :203 211. 93. Bloom DC, Stevens JG 1994. Neuron specific restriction of a herpes simplex virus recombinant maps to the U L5 gene. J. Virol. 68 :3761 3772. 94. Mehta A, Maggioncalda J, Bagasra O, Thikkavarapu S, Saikumari P, Valyi Nagy T, Fraser NW, Block TM 1995. In situ DNA PCR and RNA hybridization detection of herpes simplex virus sequences in trigeminal gangliaof latent ly infected mice. Virology 206 :633 640. 95. Bertke AS, Swanson SM, Chen J, Imai Y, Kinchington PR, Margolis TP 2011. A5 positive primary sensory neurons are nonpermissive for productive infection with herpes simplex virus 1 in vitro. J. Virol. 85 :6669 66 77.
91 BIOGRAPHICAL SKETCH Cameron L. Lilly was born in Colorado Springs, CO where grew up alongside his twin sister. He attended Fountain Valley School, where his interest in the sciences was piqued. His next foray in to academics was at Knox College in Galesburg, IL where he majored in Biochemistry and minored in Japanese. in research began under the tutelage of Dr. Rob Ewy, studying gene expression in plants. Cameron spent a summer research internship in the SuP URBS program at Michigan State University in the Insect Microbiology Lab headed Dr. Ed Walker, where he performed analysis of microbial biofilm populations in mosquito rearing habitats. After completing his undergraduate degree, Cameron then pursued his doctorate at the University of Florida in the lab of Dr. David Bloom studying lytic replication of HSV 1 and chromatin proteins. He then moved from his graduate lab to a position with Dr. Grant McFadden, also at UF, to study the oncolytic potential of My xoma virus