If 5“ L. 4:. ‘ 2. l a.» . ad. «1 HI: 4 i as .25." f». ‘ “2-; WJUDW t ‘. .. a . 3 . . Wmuwflyfiqumw :2 Raw nun... . .flmmfi. a. , .. :2“ ‘ 1a.? I .n. .5l- , J ‘ I Ib‘Afiu‘.‘.A-\’D)’ .e ‘ .xhflfafi.’ 1;)? R1 .tfklf 5! Gin! { .yn‘uom-‘jw- 55:3 llllljllllllllllllllllllllllllll 1293 01566 0750 :- LIBRARY Michigan State University This is to certify that the _ dissertation entitled \ "‘\€~'\\-.\%§Qxxxu§\ 0'5 M March's Q’sém UAW) QADWQJO‘é 0g 4—8»; QM)“ X'NWQNUK Us“); MW \ KW»\\ Ufivo it“( presented by ‘(U\ Q \4\4\ $30 k\ Ssugtwoa . has been accepted towards fulfillment of the requirements for (DMD degree in Aag'nhfil $Cx¢me @MUN Major professor Date 3006 2—“ 5 \qq 5 MS U i: an Affirmatiw Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES Mum on or baton data duo. DATE DUE DATE DUE DATE DUE MSU I. An Nflrmativo WM Opportunity Institution W ififif____._________._——.__.‘_._E.__f lDENTlF THE N HEF IDENTIFICATION AND MOLECULAR CHARACTERIZATION OF THE MAREK’S DISEASE VIRUS (MDV) HOMOLOG OF THE HERPES SIMPLEX VIRUS TYPE 1 (HSV-1) VP16 GENE By Mekki Boussaha A DISSERTATION submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Animal Science 1 996 VP16 pr VP16 o with a ABSTRACT by Mekki Boussaha A Marek's disease virus (MDV) gene encoding a homolog to the HSV—1 VP16 protein has been identified and sequenced (Yanagida et al., 1993). The MDV VP16 coding region is 1278 nucleotides long and capable of encoding a protein with a calculated molecular weight of 48,000 daltons. Predicted amino acid sequence of MDV VP16 shows considerable homology (greater than 30 % overall in all cases examined) with VP16 homologs of other alphaherpesvirus . However, MDV VP16 is 64 amino acids shorter than HSV-1 VP16, lacking a region corresponding to the HSV-1 VP16 carboxyl-terminal acidic activation domain. In this report, we show that the MDV VP16 molecule is able to transactivate both MDV and HSV-1 immediate early gene promoters. An examination of the deduced amino acid sequence revealed two potential transactivation domains within the amino-tenninal region of MDV VP16. These include a highly acidic domain (23 % acidic residues), defined by residues 1 to 60, and a proline-rich (23%) domain defined by amino acids 61 to 90. By comparing VP16 homologs using hydrophobic cluster analysis of several VP16 homologs, including MDV VP16, revealed conservation of bulky hydrophobic clusters critical for VP16 activity. Oi particular interest in these studies were bulky hydrophobic residues surrounding the MDV VP16 Phe43 residue. Substitutions of MDV VP16 Phe43 with aromatic residues mese anxne obser isde; genes seque is im; home been preserves MDV VP16 activity whereas substitutions of MDV VP16 Phe“3 with non- aromatic amino acids abolishes MDV VP16 transactivation function, similar to that observed with HSV-1 VP16. Furthermore, transactivation of MDV ICP4 promoters is dependent upon presence of an intact TAATGARAT element upstream of target genes. Deletion and mutational analysis within the MDV ICP4 gene promoter sequences revealed that the upstream ATGCAtATATTAT element at position 572 is important for MDV VP16 activation function. The ATGCAtATATTAT is highly homologous to the extended OCT-1 binding site ATGCAAATGARAT, which has been shown to be critical for activation functions of VP16 gene family members. To My Parents For Their Love and Support in wuiéeJLJiJuiuJAgzsicL-flgs IV lam very thank encouraQment of Robert 000k, Dr. tanks are also d Acknowledgments I am very thankful for the superb technical assistance, precious advice, and encouragment of Dr. Pau M. Coussens. I also thank Dr. Steven Triezenberg, Dr. Robert Cook, Dr. Bob silva, and Dr. Roger Maes for their technical advice. Special thanks are also due to my fellow students in Dr. Coussens lab. Chapte l) RNA I ll) - Ger ”ll - Ba Table of Contents Chapter 1: Transcriptional regulation: Lessons from the VP16 gene family I) RNA polymerase II (RNAPII) .................................................................................. 2 ll) - General transcription factors .............................................................................. 3 a - Transcription factor TFIlD ........................................................................ 3 b - Transcription factor TFIIA ....................................................................... 6 c - Transcription factor TFIIB ....................................................................... 7 d - Transcription factor TFIIF ....................................................................... 8 e - Transcription factor TFIIE ....................................................................... 9 f - Transcription factor TFIIH ........................................................................ 9 III) - Basal transcription by RNA polymerase II ..................................................... 10 a) Initiation complex formation ................................................................... 10 a-1) Template activation .................................................................. 10 3-2) Pre-initiation complex formation ............................................... 11 a-2-1) Stepwise model .......................................................... 11 a-2-2) Holoenzyme model ..................................................... 12 a-3) Open complex formation ........................................................... 14 b) Promoter escape by RNAPII .............................................................. 15 c) Elongation and termination ..................................................................... 16 IV) - Regulation of RNAPII transcription ................................................................. 18 VI a) Activators and their targets ...................................................................... 19 a-1) Activators and initiation complex formation ................................ 19 a-1-1) Activators and nucleosomal DNA ................................. 20 a-1-2) Activators and TBP ....................................................... 20 a-1-3)Activators and TAFs ...................................................... 21 a-1-4) Activators and TFIIB ................................................... 22 a-1-5) Activators and TFIIA ................................................... 22 a-2) Activators and promoter escape by RNAPII .............................. 24 a-2-1) Activators and TFIIH ................................................... 24 a-3) Activators and elongation and termination ................................. 24 b) VP16 gene family members ...................................................................... 25 Chapter 2: Marek's disease virus I) - Historical perspective ....................................................................................... 30 ll) - Biology of Marek's disease virus ..................................................................... 33 a) Wrus morphology and morphogenesis ..................................................... 33 b) Marek's disease virus serotypes ............................................................... 34 0) Sources of Marek's disease virus ............................................................ 35 Ill) - Pathology of Marek's disease virus ................................................................. 36 a) Productive infection .................................................................................. 36 b) Non-productive infection .......................................................................... 37 c) Pathogenesis ........................................................................................... 37 VII Vl-hh Chaph vhusl ABSTF INTRO MATEF RESUL D'SCU: FKMJRE Chapte d) Gross lesions ........................................................................................... 39 IV) - The molecular biology of MDV ........................................................................ 40 a) MDV genome structure ............................................................................. 40 b).Physical map of MDV .............................................................................. 42 V) - MDV gene expression ...................................................................................... 43 a) MDV immediate-early genes .................................................................... 44 b) MDV early genes ...................................................................................... 45 c) MDV late genes ........................................................................................ 46 Chapter 3: Identification and molecular characterization of Marek's disease virus homolog of the Herpes simplex virus VP16 gene. ABSTRACT ............................................................................................................ 49 INTRODUCTION .................................................................................................... 51 MATERIALS AND METHODS............ .................................................................... 55 RESULTS ............................................................................................................... 62 DISCUSSION ......................................................................................................... 73 FIGURES AND TABLES ....................................................................................... 77 Chapter 4: General conclusion and future research General conclusion ..................................................................................... 105 Future research ......................................................................................... 108 List of references ............................................................................................... 1 12 VIII Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figme FIQUre Figme FiQUre List of Figures Figure 1: Localization and cloning of the MDV VP16 gene ............................... 78 Figure 2: Confirmation of MDV VP16 gene location ........................................... 79 Figure 3: Nucleotide and amino acid sequneces of MDV VP16 ......................... 82 Figure 4: Comparison of amino acid sequences of VP16 homologs .................. 84 Figure 5: Detection of MDV VP16 transcripts ..................................................... 86 Figure 6: Detection of MDV VP16 gene product ................................................ 87 Figure 7: Map of pBK-CMVP16 .......................................................................... 88 Figure 8: Transactivation of MDV IE genes by MDV VP16 ................................ 89 Figure 9A: Schematic diagram of MDV ICP4 gene promoter ............................. 91 Figure SB: Effect of MDV ICP4 deletions on VP16 activity ................................ 93 Figure 10: Effect of MDV ICP4 promoter Oct1 mutants on VP16 activity ........... 95 Figure 11A: Hydrophobic cluster analysis .......................................................... 97 Figure 1 18: Linear alignment of several VP16 gene homologs ......................... 99 Figure 12A: Effect of Phe43 mutants on MDV VP16 activity ........................... 101 Figure 123: Western blot analysis of MDV VP16 mutants ............................... 102 IX Table 1' homolog Table 2 domain List of tables ' Table 1: Deduced amino acid comparisons between several VP16 homologs ........................................................................................................... 1 03 Table 2: Comparisons of putative MDV VP16 N-terminal activation domains with other known transactivation domains .......................................... 104 AGP ALV BHV csr CHX CKC cw cm en: DHFR on DR E genes EBV EHV QBlC/DIEIIIK AGP ALV BHV CEF CHX CKC CMV CTD CTF DHFR DPI DR E genes EBV EBNA-2 EHV gBICIDIEIIIK List of Abbreviations Agar-gel precipitin Avian Ieukosis virus Bovine herepsvirus Chicken embryo fibroblast Cyclohexamide Chicken kidney cells Cytomegalovirus CarboxyI-terminal domain Cellular transcription factor Dihydrofolate reductase Days post-infection Direct repeat Early genes Epstein-Barr virus EBV nuclear antigen 2 Equine herpesvirus Glycoprotein BlC/D/ElllK XI GTF HCA HCF HSV-1 I2 HVT IE IF IRL IRS L genes MD MDV NER NaOH ORF PAA PCR PEG pp1 4138 RNAPII General transcription factor Hydrophobic cluster analysis Host cell factor Herpes simplex virus type 1 and 2 Herpesvirus of turkey Immediate early lmmunofluorescence Inverted repeat long Inverted repeat short Kilobasepairs Late genes Marek's disease Marek's disease virus Nucleotide excision repair Sodium hydroxide Open reading frame Phosphonoacetic acid Polymerase chain reaction polyethelene glycol Phosphoprotein 14/38 RNA polymerase II XII SDS SRB TAF TBP ' IFHAENIVET TRS UL US VP16 vaDV SDS SRB TAF TBP TFIIAIBIDIEIFIHIK TK TRL TRS UL US VP16 vaDV VZV Sodium dodecyl-sulfate Suppressor regulatory protein TBP-associated factor TATA-binding protein Transcription factors ll Thymodine Kinase Terminal repeat long Terminal repeat short Unique long Unique short Virion protein 16 very virulent MDV Varicella-zoster virus XIII Chapter 1 Transcription regulation: Lessons from the VP16 gene family Differentiation and development in eukaryotes is largely regulated at the level of transcription. The transcriptional control regions of eukaryotic genes often consist of networks of DNA-binding sites for factors that regulate the rate of transcription initiation at promoters. The promoter of a typical eukaryotic protein- encoding gene contains a TATA box, which is located 25 to 30 base pairs upstream of the transcription start site and an initiation sequence surrounding the start site. These elements comprise the core promoter region which is recognized by the general transcription machinery. In addition, promoter-specific factors bind to specific arrays of DNA recognition sites unique to each gene, both in the promoter proximal region and distal to the transcriptional start site. Signalling between transcriptional regulators and basal transcription factors requires co-activatcrs that are not required for basal transcription but, rather, assist or mediate activation and perhaps repression of transcription (Pugh and Tjian, 1990). Initiation of messenger RNA transcription by the enzyme RNA polymerase II (RNAPII) is a complex process that requires assembly of a multi-component pre- initiation complex near the transcription start site (Buratowski et al., 1989; Zawel and Reinberg, 1993). This complex is responsible for promoter recognition and response to regulatory signals (Sawadago and Sentenac, 1990; Roeder, 1991). At least six transcription factors are required, in addition to RNAPII, for accurate and eficbnt {l-E,4 n-IRNAl E thattnu pdwnen banscnb snufllsu enlymes andregL ”flhflahc banscnp Rembeq RNAPH always c 140kDa found a: pobpept An canNDXyL eukaDYoh 2 efficient initiation of transcription. These transcription factors consist of TFIIA, -B, -D, -E, -F, and -H (Matsui et al., 1980; Samuels et al., 1982; Davison et al., 1983). I) - RNA polymerase II (RNAPII) Eukaryotic cells contain three nuclear DNA-dependent RNA polymerases that transcribe different classes of genes (Roeder and Rutter,1969). RNA polymerase I synthesizes ribosomal RNA precursors. RNA polymerase II (RNAPII) transcribes protein-coding genes, while RNA polymerase III transcribes genes for small stable RNAs such as SS ribosomal RNA and transfer RNA. Each of these enzymes requires a collection of additional factors for selective promoter recognition and regulated transcription initiation. The molecular basis of transcription and its regulation are best understood for genes transcribed by RNAPII with general transcription factors (GTFs) (Roeder, 1991; Hori and Cari, 1994; Maldonado and Reinberg, 1995). Eukaryotic RNAPII enzymes share three major features. First, RNAPII is generally composed of 10 :l: 2 subunits. Second, the RNAPII enzyme always contains two large subunits with molecular sizes of approximately 220 and 140 kDa. Finally, RNAPII contains three subunits of 14 to 28 kDa that are also found associated with RNA polymerases I and III in all eukaryotes. These polypeptides are called "common" or "shared" subunits. An interesting feature of RNAPII is that its largest subunit has a unique carboxyI-terrninal domain (CTD) that is absent in prokaryotic RNA polymerases, and eukaryotic RNA polymerases l and III (Corden et al., 1990). An array of tandem heptapel single-let largest SL The CTD and is pr from the accomplis (RNAPHA showed it transcriptn CDKsubu "l - Generl a-1 t0the TAT; Dyke 9t al., cornplex rec Binding of ' “(inscription luncfiwa' ml Rembe’Q, 1 c 3 heptapeptide repeats of the sequence Tyr-Ser-Pro-Thr—Ser—Pro-Ser WSPTSPS; single-letter amino acid code), located at the carboxyl-terminal domain (CTD) of the largest subunit of RNAPII constitute a highly conserved region essential for viability. The CTD of RNAPII is phosphorylated at different stages of the transcription cycle, and is probably involved in transcriptional regulation. For instance, transcription from the murine dihydrofolate reductase (DHFR) promoter can only be accomplished by the form of RNAPII that contains the hypophosphorylated CTD (RNAPIIA), but not by the form that lacks it (RNAPIIB). Akoulitchev et al. (1995) showed that the CTD, but not its phosphorylation, is required for initiation of transcription, and that transcription requires CTD kinase activity provided by the CDK subunit of TFIIH. ll) - General transcription factors a - Transcription factor TFIID The first step in assembly of initiation complexes is binding of TFIID to the TATA element (Sawadago and Roeder, 1985; Nakajima et al., 1988; Van Dyke et al., 1988; Buratowski et al., 1989). TFIID is a site-specific, DNA-binding complex required for selective initiation of transcription by eukaryotic RNAPII. Binding of TFIID at the promoter region launches sequential recruitment of transcription factors TFIIA, TFIIB, TFIIF/RNAPII, TFIIE, and TFIIH to establish a functional multi-protein complex capable of transcriptional initiation (Zawel and Reinberg, 1993). Trans TBP (Pugh referred to . Tjian, 1991: can replacx Horikoshi e terminal do of structure etal., 1990 67 amino a a basic regr species pre direct lepe, 19930. TI elelllertts, ace‘ESsible The SequenCe‘ eMential rc the N‘lermr 4 Transcription factor TFIID contains the TATA-binding protein, referred to as TBP (Pugh and Tjian, 1992), tightly associated with several other polypeptides referred to as TBP-associated factors or TAFs (Dynlacht et al., 1991; Pugh and Tjian, 1991; Tanese et al., 1991). The C-tenninal 180 amino acid domain of TBP can replace a TFIID fraction for basal transcription in Mitre (Hoey et al., 1990; Horikoshi et al., 1990; Petersen et al., 1990). Amino acid comparisons of the C- terrninal domain of TBP from numerous species exhibit considerable conservation of structure (Hoffmann et al., 1990a, b; Peterson et al., 1990; Kao et al., 1990; Gash et al., 1990). The C-tenninal domain of TBP contains two DNA-binding repeats (66- 67 amino acids each). These two direct repeats are nearly 40% identical and flank a basic region. The three—dimensional and crystal structures of TBP from numerous species predicted that TBP resembled a molecular saddle in which the C-tenninal direct repeats straddle the DNA (Nikolov et al., 1992; Kim et al., 1993a; Kim et al., 1993b). The inner surface of the saddle interacts with the minor groove of TATA elements, causing prominent distortion of DNA. The outer protein surface is accessible for interactions with other transcription factors. The N-terminal regions of TBP from different species vary greatly in size, sequence, and amino acid composition. The N-terminal region of TBP has no essential role in transcription. Indeed, some strains of S. cerevesiae can grow when the N-terminal region of TBP is either completely missing (Cormack et al., 1991; Gill and Tjian, 1991; Poon et al., 1991; Reddy and Hahn, 1991; Zhou et al., 1991), or replaced with a very different N-terminal region of human TBP (Cormack et al., 1991; Gill ar TBP (iAFs) (Dyi studies has with TBP ( 1988a, b; interactior al., 1994), (Chiang a dTAFH40, mmmdbr by which still Uncle interactjo (Hisatak. (Weinzle interactn dTAFHSC El al., 1 dTAFHBC el al., 1. Tilan, 1 1991; Gill and Tjian, 1991). TBP is tightly associated with polypeptides called'TBP-associated factors (T AFs) (Dynlacht et al., 1991; Pugh and Tjian, 1991; Tanese et al., 1991). Several studies have demonstrated direct interactions of transcriptional activators not only with TBP (Sawadago and Roeder, 1985a; Abmayr et al., 1988; Horikoshi et al., 1988a, b; Stringer et al., 1990), but also with specific TAFs. These include interactions of Sp1 with Drosophila TAF.,110 (dTAF..10) (Hoey et al., 1993; Gill et al., 1994), interactions of USF, Tat, CTF, and E1a with human TAF..55 (hTAF.,55) (Chiang and Roeder, 1995), interactions of VP16, p53, NF-KB p65 with dTAF..60, dTAF..40, hTAF..80 and hTAF..31, (Goodrich et al., 1993; Thut et al., 1995), and interactions of a steroid receptor with hTAF..30 (Chen et al., 1994). The mechanism by which TAFs transduce or relay signals from activators to the basal machinery is still unclear. TAFs can also interact with other TAFs and/or with GTFs. TAF-TAF interactions include interactions of hTAF..80 with hTAF..250, hTAF.,31, and hTAF..20 (Hisatake et al., 1995), and interactions of dTAF..60 with dTAF..250 and dTAF..40 (Weinzierl et al., 1993; Kokubo et al., 1994; Thut et al., 1995). TAF-GTF interactions include dTAF..40 that interacts with TFIIB (Goodrich et al., 1993), dTAF..60 that interacts with TBP (Weinzierl et al., 1993; Kokubo et al., 1994; Thut et al., 1995), dTAF..110 that interacts with TFIIA (Yokomori et al., 1993a), and dTAF..80 that interacts with TBP, TFIIE and the RAP74 subunit of TFIIF (Hisatake et al., 1995), and hTAF250 interacts with RAP74 subunit of TFIIF (Ruppert and Tjian, 1995). Although the mechanism by which these interactions enhance tanscnphc TFllD bind and (2) rec (Honkoshi in yeast ( Complex DeJong mechanig entirely u lime folio and high transcnp TECOQmZ 1989; A: TFIIA m factors ( ”Home l"Ostroz. aiSQ afie 6 transcription is unclear, they could be involved in (1) recruitment and stability of TFIID binding on the promoter (Abmayr et al., 1988; Lieberman and Berk, 1994), and (2) recruitment and functional modulation of other general transcription factors (Horikoshi et al., 1988a; Choy and Green, 1993). b - Transcription factor TFIIA TFIIA has been identified as a two-subunit (43 and 12 kDa) complex in yeast (Ranish and Hahn, 1991), and as a three-subunit (37, 19, and 13 kDa) complex in human and Drosophila (Cortes et al., 1992; Yokomori et al., 1993b; DeJong and Roeder, 1993; Coulombe et al., 1992; A50 et al., 1994). The mechanism by which TFIIA influences transcription initiation by RNAPII is not entirely understood. TFIIA can enter the transcription pre-initiation complex at any time following TBP binding. TFIIA is not required for basal transcription using TBP and highly purified factors (Ma et al., 1993). TFIIA is capable of stimulating basal transcription by altering the conformation of TBP, and thus enhancing its ability to recognize and stably associate with various TATA elements (Buratowski et al., 1989; Aso et al., 1994; Lee et al., 1992; lmbalzano et al., 1994). In this regard, TFIIA may stimulate transcription by competing for TBP binding with negative factors (NC1, Dr1lNC2, HMG-1, and ADI), which prevent formation of functional pre-initiation complexes (Meisterernst et al., 1991; Meisterernst and Roeder, 1991; Inostroza et al., 1992; Ge and Roeder, 1994; Auble and Hahn, 1993). TFIIA may also affect transcription activation by sequence-specific activators (Meistrernst et al., 199‘ directly (Lieberrr 1994). the RNAF roughly 00 with a put. TFIIF (Ha through inti Carboxyl~tei Contains twi by a Short, , pr 0lTlOter Co in solution (+ DNA and Rn 1993; Ha 6t transcription s several transc Robens 91 al., 7 al., 1991; Meisterernst and Roeder, 1991; Ma et al., 1993). Indeed, TFIIA interacts directly with GAL4-AH (Wang et al., 1992), the Epstein-Barr virus activator Zta (Lieberman and Berk, 1991), and HSV—1 VP16 (Yokomori et al., 1994; Ozer et al., 1994). c - Transcription factor TFIIB TFIIB binds to the TBP-DNA complex, where it recruits and positions the RNAPII/TFIIF complex (Buratowski et al., 1989). TFIIB has two domains that roughly correlate with its two functions. TFllB consists of an amino-terminal region, with a putative metal-binding site that is essential for recruitment of RNAPII and TFIIF (Ha et al., 1993; Yamashita et al., 1993; Hisatake et al., 1993), probably through interactions with RAP30 of TFIIF (Ha et al., 1993), and a protease-resistant carboxyl-terminal core (T Flch) composed of two long amino-acid repeats. TFIIBc contains two similar domains (each encoded by one of the two repeats) connected by a short, random-coil peptide. TFIIBc is sufficient for interactions with the TBP- promoter complex (Ha et al., 1993; Yamashita et al., 1993), and can bind RNAPII in solution (Ha et al., 1993). TFIIB functions as the "bridging" factor between TBP- DNA and RNAPIIfTFIIF complexes (Buratowski and Zhou, 1993; Barberis et al., 1993; Ha et al., 1993). In combination with RNAPII, TFIIB determines the transcription start site (Pinto et al., 1992; Li et al., 1994). TFIIB is also the target of several transcriptional activators. In this regard, acidic— (Lin and Green, 1991; Roberts et al., 1993), and proline-rich (Kim and Roeder, 1994) activation domains can inte stages 0 consists These ar associatir specificall similarities MCCraken Sequence region 4 at A5 Predicte DNA (T an e RAP74 Cont by a hi9th 8 can interact directly with and recruit TFIIB into the pre-initiation complex. d - Transcription factor TFIIF TFIIF controls activity of RNAPII at both the initiation and elongation stages of transcription (Tan et al., 1994). Mammalian TFIIF is a heterodimer that consists of two subunits of 30 and 74 kDa each (Conaway and Conaway, 1993). These are designated RAP30 and RAP74 (RAP stands for RNA polymerase II- associating protein). RAP30 binds to RNAPII and prevents it from binding non- specifically to DNA (Killeen and Greenblatt, 1992). RAP30 shares sequence similarities to the Escherichia coli 07° protein, and B. subtilis o“3 (Sopta et al., 1989; McCraken and Greenblatt, 1991). The RAP30 carboxyI-terminal region shares sequence similarities with the cryptic DNA binding domain present in conserved region 4 at the carboxyl-terminal domain of a bacterial 0 factor (Garrett et al., 1992). As predicted by sequence alignment, this region was found to be capable of binding DNA (Tan et al., 1994). The TFIIF RAP74 has been less extensively characterized. RAP74 contains globular amino-terminal and carboxyl-terminal domains separated by a highly charged central region. RAP74 is required for efficient transcription initiation by RNAPII (Kephart et al., 1994; Yonaha et al., 1993). However, only the RAP74 amino-terminal domain is required for transcriptional elongation (Kephart et al., 1994). Some reports suggested that RAP74 may be a helicase, although no ATP-dependent helicase activity or ATPase has been detected (Flores et al., 1990). or) andt 1991). ' (lnostroz transcript form of R etal., 19$ etal., 199 thus medl recruitmen TFIIH-3350 ERGO-3 (n 1993i The Clealance a f'Tr.‘ comma, r aCiiVItieS (Ge: e - Transcription Factor TFIIE TFIIE is a heterotetramer composed of two 56 kDa (Jr-subunits (T FllE- cr) and two 34 kDa B—subunits (TFIIE-B) (Ohkuma et al., 1991; Inostroza et al., 1991). TFIIE enters the pre-initiation complex after recruitment of TFIIF/RNAPII (Inostroza et al., 1991). TFIIE interacts with several components of the RNAPII transcription complex. Indeed, TFIIE can bind selectively to the non-phosphorylated form of RNAPII (RNAPlla) and to both subunits of TFIIF as well as TFIID (Maxon et al., 1994). Interactions of TFIIE with RNAPII are mediated by TFIIE-or (Maxon et al., 1994). TFIIE can also interact directly with TFIIH (Maxon et al., 1994), and thus mediate TFIIH-recruitment to the RNAPII transcription complex. TFIIH- recruitment mediates subsequent phosphorylation of the CTD of RNAPII by the TFIIH-associated kinase activity (Maxon et al., 1994). TFIIE can also interact with ERGO—3 (Maxon et al., 1994), a DNA repair helicase of TFIIH (Schaeffer et al., 1993). These multiple interactions suggest that TFIIE is critical for both promoter clearance and transcription-coupled DNA repair (Drapkin and Reinberg, 1994). f - Transcription factor TFIIH TFIIH is one of the last factors to be recruited into the initiation complex. TFIIH is a multisubunit complex with several associated enzymatic activities (Gerard et al., 1991; Gileadi et al., 1992; Serizawa et al., 1992; Feaver et al., 1993). TFIIH possesses a kinase activity that is capable of phosphorylating the CTD of RNAPII, DNA-dependent ATPase and DNA helicase activities (Serizawa et al., 199 (Schael HoIoTFi athrees form is a complex 14), and RNAPII ( The CTD 1994). C Complex ( reSlulatory a)| 10 al., 1993). TFIIH is also involved in nucleotide excision repair (NER) of DNA (Schaeffer et al., 1993). S. cerevisiae TFIIH exists in two forms. One form, HoloTFIIH, is composed of a six subunit core/SSLZ complex tightly associated with a three subunit CTD-kinase complex designated TFIIK (Svejstrup et al., 1995). This form is active in transcription. The other form, repairsome, is a multi-component complex of core/SSL2 and all other S. cerevisiae NER genes (RAD1, 2, 4, 10, and 14), and is active in NER (Svejstrup et al., 1995). TFIIH plays an important role in RNAPII CTD-phosphorylation, open-complex formation and promoter clearance. The CTD kinase resides with MO15lCdk7, a cyclin-dependent kinase (Feaver et al., 1994). Cyclin H, the regulatory partner of MO15/Cdk7 is also found in the TFIIH complex (Serizawa et al., 1995). TFIIH can be targeted by several trancriptional regulatory proteins, such as HSV-1 VP16 and cellular protein p53 (Xiao et al., 1994) III) - Basal transcription by RNAPII a) Initiation complex formation a-1) Iemplate activation DNA in eukaryotic cells is packaged into chromatin. The primary subunit of chromatin is the nucleosome core particle which consists of 146 bp of DNA wrapped around core histone octamer (Kornberg, 1974). The core histone octamer contains two (H2A-H2B) heterodimers and one (H3-H4)2 tetramer (Eickbush and Moudrianakis, 1978). Chromatin structure can affect the process of RNAPII transcription by (1) hindering access of the general transcription machinery to promote transcriptio 1992). Th ”transcripti is unclear. SWllSNFr factors acc 1992; Pete al.,1994). on an ATF SWl/SNF (SNF2)‘ 3' can assist 1994) an HerkOWitz comDOSeC ZaWe! am 11 to promoter sequences, and (2) blocking the binding of sequence-specific transcription activator proteins to their promoter-binding sites (Felsenfeld et al., 1992). The mechanism by which RNA polymerases can unwind DNA to form a "transcription bubble" and elongate past a nucleosome without major dislocations is unclear. Recent biochemical and genetic studies imply that cellular factors - the SWIISNF complex - are required to alter chromatin structure to allow transcription factors access to their binding sites (Kruger and Herskowitz, 1991; Hirschhom et al., 1992; Peterson and Herskowitz, 1992; Winston and Carlson, 1992; lmbalzano et al., 1994). Interactions of SWIISNF complex with nucleosomal DNA is dependent on an ATPase activity which is intrinsic to the complex (COte et al., 1994). In yeast, SWIISNF complex is composed of five subunits encoded by SWI1 (ADR6), SWI2 (SNF2), SWI3, SNF5, and SNF6 (Winston and Carlson, 1992). SWIISNF complex can assist binding of basal transcription factors (Kwon et al., 1994; lmbalzano et al., 1994) and transcription activator proteins (COté et al., 1994; Peterson and Herkowitz, 1992; Yoshinaga et al., 1992; Laurent and Carlson, 1992) to nucleosomal DNA, in vitro. aid) Stepwise model Pre-initiation complexes are elaborate structures composed of general transcription factors and RNAPII (Buratowski et al., 1989; Zawel and Reinberg, 1993). In the stepwise model, pre-initiation complexes assemble in (Buratowski binding and factor stably et al., 198i templates, (Buratowsi whether or TFIIB acts TFIIF/RNA resulting ; RNAPII),1 and TFIIF identified titat gens RNA DOIy The feiitu 12 assemble in a cascade fashion starting with binding of TFIID to the promoter (Buratowski et al., 1989). TBP and TAF subunits of TFIID participate in DNA binding and promoter recognition (Verrijzer et al., 1995). No other transcription factor stably interacts with the promoter in the absence of TFIID binding (Van Dyke et al., 1988, 1989; Buratowski et al., 1989). After binding of TFIID to DNA templates, the basal factors TFIIA and TFIIB are recruited to the promoter (Buratowski et al., 1989). TFIIB can associate with the TFIID-promoter complex whether or not TFIIA is present (Buratowski et al., 1989; Peterson et al., 1990). TFIIB acts as a molecular bridge, allowing subsequent recruitment of the TFIIF/RNAPII complex to TFIID-DNA complexes (Buratowski et al., 1989). The resulting assembled complex (composed of TFIID, TFIIA, TFIIB, TFIIF, and RNAPII), is sufficient to initiate transcription. Recruitment of additional factors, TFIIE and TFIIH, appears to direct later steps such as promoter clearance, elongation, and transcription-coupled DNA repair (Goodrich and Tjian, 1994). a-2-2) Holoenzyme model Genetic and biochemical studies in yeast have recently identified a large pre-assembled complex, termed RNA polymerase II holoenzyme, that consists of RNAPII, a subset of GTFs, nine suppressor regulatory proteins of RNA polymerase B (SRB 2-11), GAL11, SUG1, and additional unidentified proteins. The feature that distinguishes the RNA polymerase II holoenzyme model from the stepwise model is stable interactions of RNAPII with the SRB proteins and GTFs independ 1994). lr initiation polymera polymera: polymera: regulatory polymeras forms of F smaller Rh form and Purification 0ftranscrip transcriptio feaiUres' nc 1992). SligE direct iniEra Called the n holoenzyme “Untied med suggeSied F 13 independently of DNA template, in Mitre (Kim et al., 1994; Koleske and Young, 1994). In the holoenzyme model, the two major regulatory steps in transcription initiation are formation of a TFIID-bound promoter and association of the RNA polymerase II holoenzyme with TFIID-DNA complex. Several forms of RNA polymerase II holoenzyme have been described. The larger form of RNA polymerase II holoenzyme contains RNAPII, TFIIB, TFIIF, TFIIH, and SRB regulatory proteins (Koleske and Young, 1994). The second form of RNA polymerase II holoenzyme contains RNAPII, TFIIF, and SRB proteins. These two forms of RNA polymerase II holoenzyme may exist simultaneously in vim, or the smaller RNA polymerase II holoenzyme may, in fact, be a subcomplex of the larger form and appears to be a consequence of instability of the latter one during purification. RNA polymerase II holoenzymes are capable of site-specific initiation of transcription when supplemented with the missing GTFs. and are responsive to transcriptional activators (Kim et al., 1994; Koleske and Young, 1994). These features, not observed with purified RNAPII and GTFs alone (Flanagan et al., 1991; 1992), suggest a model in which transcriptional activators may function through direct interactions with the RNA polymerase II holoenzyme. Indeed, a subcomplex, called the mediator of activation, can be dissociated from the RNA polymerase II holoenzyme by using monoclonal anti-CTD antibodies (Kim et al., 1994). The purified mediator contains SRB proteins, SUG1, GAL11, and TFIIF. Genetic studies suggested previously that GAL11, SUG1, and SRB proteins are involved in transcriptional regulation (Fassler and Winston, 1989; Himmelfarb et al., 1990; Nishizawe and bioct between important initiation. tightly ass phosphoryl is not yet c interactions Reinberg, 1 °°mplex as: becofiles un to as the 0p, TFIIB, TFIIF, t0 the tra ”8 Cr 0Den of melte bond hydrolys ATP may be u DhOSphOMate l4 Nishizawa et al., 1990; Vallier and Carlson, 1991; Yu and Fassler, 1993). Genetic and biochemical interactions have described physical and functional interactions between the RNAPII CTD and SRB regulatory proteins, and thus may have important implications for the function(s) of the RNAPII CTD in transcription initiation. SRB10 and SRB11 encode kinase and cyclin-like polypeptides that are tightly associated with the holoenzyme and appear to have roles in CTD phosphorylation (Liao et al., 1995). The role of CTD phosphorylation in transcription is not yet clear (Li and Kornberg, 1994; Serizawa et al., 1993), but it may disrupt interactions between RNAPII and GTFs to stimulate promoter clearance (Zawel and Reinberg, 1992; Dahmus and Dynan, 1992; O'Brien et al., 1994). as?) Open £0111an tarmatlhn After RNAPII and the GTFs assemble on promoter DNA (closed complex assembly), a short stretch of DNA around the transcription start site becomes unwound and serves for abortive transcription. This process is referred to as the open complex formation. Minimal initiation complexes containing TBP, TFIIB, TFIIF, and RNAPII are efficient of cycling short abortive transcripts specific to the transcription start site, indicating that the DNA around the start site is in the open or melted configuration. Transition to a fully open complex requires ATP B—v bond hydrolysis, TFIIE and TFIIH (Layboum and Dahmus, 1990; Wang et al., 1992). ATP may be used by DNA helicases to melt the start site and/or by CTD kinases to phosphorylate the CTD of RNAPII. TFIIH has both ATP-dependent helicase and CTD-kina be targett reported t Jiang et at b) I escapes th Minimal init. in cycling st of TFIIE an. comWent it eiongation c Tun,1994) (GOOGrich an the melted , transcript forr followmg p O ,y’ the initiation , PhOSphOrylan.C CTD hyperphc 15 CTD-kinase activities (Drapkin and Reinberg, 1994). Open complex formation may be targeted by transcriptional activators. Indeed, GAL4—VP16 derivatives were reported to facilitate the process of open complex formation (\Nang et al., 1992; Jiang et al., 1994). b) Emmeter escape by RNAEII Promoter clearance by RNAPII is the event during which RNAPII escapes the promoter to elongate primary transcripts (Goodrich and Tjian, 1994). Minimal initiation complexes containing TBP, TFIIB, TFIIF, and RNAPII are efficient in cycling short abortive transcripts specific to the transcription start site. Addition of TFIIE and TFIIH results in formation of active transcription complexes that are competent to advance through the promoter clearance stage and proceed into an elongation complex, upon addition of nucleoside triphosphates (Goodrich and Tjian, 1994). Production of extended transcripts requires hydrolyzable ATP (Goodrich and Tjian, 1994). TFIIH-associated helicase and ATP hydrolysis stretch the melted region of DNA, probably in the direction of transcription, prior to transcript formation. The RNAPII system requires a pair of GTFs (T FllE/TFIIH) following polymerase entry because this pair somehow modifies polymerase and/or the initiation complex. Such modifications could involve the CTD of RNAPII. Phosphorylation of RNAPII CTD generates at least two RNAPII isoforms, one with CTD hyperphophorylated (RNAPIIO), the other with CTD non-phosphorylated (RNAPIIA) (Dahmus, 1981; Cadena and Dahmus, 1987; Baskaran et al., 1993). TFIIH conta (Drapkin an: and TFIIE transcription 1992; O‘Brli (Cadena ar supercoiling and ATP l clearance ( associated Open comp l6 TFIIH contains a kinase activity capable of phosphorylating the CTD of RNAPII (Drapkin and Reinberg, 1994). RNAPIIA interacts with TBP (Usheva et al., 1992) and TFIIE (Maxon et al., 1994), and participates in complex assembly and transcription of sequences proximal to the promoter (Lu et al., 1991; Chestnut et al., 1992; O'Brien et al., 1994). RNAPIIO, however, catalyzes RNA chain elongation (Cadena and Dahmus, 1987; O'Brien et al., 1994; Weeks et al., 1993). Negative supercoiling can, however, functionally replace the requirements for TFIIE, TFIIH, and ATP hydrolysis, indicating that negative supercoiling expedites promoter clearance (Goodrich and Tjian, 1994). These findings suggested a role of TFIIH- associated kinase activity and supercoiling in promoter escape by RNAPII, but not open complex formation as previously suggested (Schaeffer et al., 1993). o) Elongation and termination The mechanism of transcriptional elongation of eukaryotic messenger RNA synthesis is not completely understood. Recently, however, several developments have offered new insights into possible transcriptional elongation mechanisms. Biochemical studies have defined several factors involved in the elongation process by RNAPII. These elongation factors include TFIIS (Sll), TFIIF, and the elongin/SIII complex. Mechanistic studies indicate that S" is capable of increasing the overall rate of duplex DNA transcription by RNAPII (Chen et al., 1992). SII does not increase the catalytic rate of nucleotide addition to a growing transcript. Rather, Sll appears to facilitate passage of RNAPII through a variety of transcriptio sltes).whic throughter RNAPII cc 1991; Kan endonucle polymerasi 1993). Cle Mn”), and interaction: TFll RNAPII (B GTFS beca MEChanistii bYRNAPll SlabiIiZes I (Kitajlma e, transcnptior arrest at the elongin/s“, In ad( Complex iS a l7 transcriptional blocks (nucleoprotein complexes, intrinsic arrest sites, and pausing sites), which can lead RNAPII to pause, terminate, or resume transcription and pass through terrninators (Rudd et al., 1994). Upon encountering a block to transcription, RNAPII concludes elongation, but can be reactivated by Sll (Kerpolla and Kane, 1991; Kane, 1994; Reines, 1994). Sll-dependent read-through is proceeded by endonucleolytic cleavage and re-extension of nascent transcripts held in the polymerase active site (Reines, 1992; Izban and Luse, 1992; Wang and Hawley, 1993). Cleavage of nascent transcripts by Sll requires: (i) divalent cations (Mgz*or Mn”), and is inhibited by low concentrations of or-amanitin; and (ii) physical interactions between RNAPII and the cleaved RNA. TFIIF and elongin/Slll increase the overall rate of RNA chain elongation by RNAPII (Bengal et al., 1991; Garrett et al., 1994). TFIIF is unique among other GTFs because of its ability to support both transcription initiation and elongation . Mechanistic studies indicated that TFIIF increases the rate of RNA chain elongation by RNAPII (Bengal et al., 1991; Garrett et al., 1994). The RAP74 subunit of TFIIF stabilizes binding of TFIIF to RNAPII and increases TFIIF elongation activity (Kitajima et al., 1994). TFIIF is not capable of releasing RNAPII from arrest at transcription blocks. Rather, TFIIF decreases the likelihood that RNAPII will suffer arrest at these sites (Cu and Reines, 1995). The mechanism by which TFIIF and elonginlSlll increase the elongation rate is not completely understood. In addition to transcription blocks that can halt elongation, RNAPII elongation complex is also the target of transcription-coupled nucleotide excision repair (NER) (Mellon et largest sul (Schaeffer TFB1/p62 mammalia Drapkin e' VI) - Regi Trz deveIOprr Combinati Promoter and Tjian factors it trarlSCript poii’mera SurroUndi achieVe a or more 0' regulatory UnClear, 18 (Mellon et al., 1987). TFIIH is closely involved in NER (Schaeffer et al, 1993). The largest subunit of mammalian TFIIH is the product of the NER gene XPB/ERCC3 (Schaeffer et al., 1993). The products of the known NER genes RAD3/XPB, TFB1/p62, and SSL1/p44 are also among the subunits of S. cerevisiae and mammalian TFIIH (Feaveret al, 1993; Schaefferet al., 1994; Humbert et al., 1994; Drapkin et al., 1995). VI) - Regulation of RNAPII transcription Transcriptional activation of eukaryotic protein-coding genes during development or in response to extracellular signals can be accounted by the combinatorial effect of upstream activator proteins targeted to enhancer and promoter regions by sequence-specific DNA-binding (Maniatis et al.,1987; Mitchell and Tjian, 1989). Key players in this process are sequence-specific transcription factors that select genes to be activated and assist assembly of a stable transcription pre-initiation complex at the start site of mRNA synthesis by RNA polymerase II (Ptashne and Gann,1990). GTFs and RNAPII bind in the surroundings of initiation start sites to support a basal level of transcription. To achieve an induced level of transcription, factors of the regulatory family bind to one or more of a variety of DNA sequence elements located farther away. How these regulatory factors operate to boost the frequency of initiation by RNAPII remains unclear. transcriptio major step elongation all of these II. The me over RNA F tfanscriptiOr regulatory I Complex fon ‘dismptionc f0"nation - n iniiiation c0” RNAP“ CTD tranSCTiDIiOn f 19 a) Actuators and their targets How transcriptional activators act to increase the frequency of transcription initiation is unclear. Transcription by RNAPII can be divided into three major steps: (1) Initiation complex formation, (2) promoter clearance, and (3) elongation and termination. Transcriptional regulatory factors may target some or all of these discrete steps to increase the rate of transcription by RNA polymerase II. The mechanisms through which transcriptional activators exert their influence over RNA polymerase II machinery is the focus of the following sections. a-1) Actuators and initiation complex formation As a first step to comprehend the mechanism of activation of transcription initiation complex assembly, it is important to identify interactions of regulatory proteins with individual transcription factors. Transcripton initiation complex formation can be subdivided into three major steps: (1) template activation - disruption of nucleosomal DNA by the SWIISNF complex; (2) pre-initiation complex formation - recruitment of TFIID, TFIIA, TFIIB, and TFIIF/RNAPII to form the closed initiation complex; and (3) open complex formation - ATP B—v bond hydrolysis and RNAPII CTD phosphorylation may be required for this transitition. Transcriptional activators were reported to interact directly with several of the aforementioned transcription factors which are involved in the three phases of transcription initiation complex assembly. These interactions are the focus of the following sections. interferes transcriptic 1986; Wort transcriptio nucleosom (Workman SWIISNF c enabling Ira aftd Herkow et al., 1992) (KWOH et al. 20 a-1-1) Actuators and nucleosomal DNA The assembly of naked template DNA into nucleosomes interferes with the ability of DNA to be transcribed, by blocking access of transcription factors to the DNA template (T suda et al., 1986; Knezetic and Luse, 1986; Workman and Roeder, 1987; Paranjape et al., 1994). Association of several transcriptional activators with nucleosomal DNA can induce a change in nucleosomal structures which facilitate recruitment of GTFs to the template DNA (Workman et al., 1988; Workman et al., 1990; Workman et al., 1991). The SWIISNF complex mediates an ATP-dependent disruption of nucleosomal DNA, enabling transcriptional activator proteins, such as GAL4, Drosophila flz (Peterson and Herkowitz, 1992), mammalian glucocorticoid and estrogen receptor (Yoshinaga et al., 1992), LexA-GAL4 (Laurent and Carlson, 1992), GAL4-VP16 and GAL4-AH (Kwon et al., 1994), to bind to DNA within a nucleosome core. a-1-2) Activators and IBE TBP was the first general transcription factor shown to interact with an activation domain (Stringer et al., 1990; lngles et al., 1991 ). Several other activation domains have, subsequently, been reported to bind TBP, in vitro. These include (1) viral activators: E1A, Zta, Tat, Tax1, and IE2; (2) cellular proteins: E2F-1, p53, PU.1, Sp1, Oct-1, Oct-2, c-Rel, and c-Fos; and (3) yeast protein GAL4 (Lee et al., 1991; Lieberman and Berk, 1991; Kashanchi et al., 1994; Caron et al., 1993; Hagemeier et al., 1992; Emili and lngles, 1995; Truant et al., 1993; Hagemeien Melcher and TBP may t promoter. 2 [huNCZ,F initiation co Of a numbe interactions al.,1994), (Chiang ar dTAFH40' i intErection mechanisr et at. 19. modulatiOr Gl'eenI 19 21 Hagemeier et al., 1993; Emili et al., 1994; Zwilling et al., 1994; Metz et al., 1994; Melcher and Johnston, 1995). Interactions between transcriptional activators and TBP may be involved in (1) recruitment and stability binding of TBP on the promoter, and/or (2) conquer the inhibitory effects of certain proteins (NC1, Dr1lNC2, HMG1, and ADI) that prevent TBP from initiating assembly of the pre- initiation complex (Kraus et al., 1994). a-1-3) Activators and IAE,,s Several studies have demonstrated direct interactions of a number of transcriptional activators with TAFll subunits of TFIID. These include interactions of Sp1 with Drosophila TAF..110 (dTAF..110) (Hoey et al., 1993; Gill et al., 1994), interactions of USF, Tat, CTF, and E1a with human TAF..55 (hTAF..55) (Chiang and Roeder, 1995), interactions of VP16, p53, NF-KB p65 with dTAF..60, dTAF..40, hTAF..80 and hTAF..31, (Goodrich et al., 1993; Thut et al., 1995), and interactions of a steroid receptor with hTAF..30 (Chen et al., 1994). Although the mechanism by which these interactions enhance transcription is unclear, they could be involved in (1) recruitment and stability of TFIID binding on the promoter (Abmayr et al., 1988; Lieberman and Berk, 1994), and (2) recruitment and functional modulation of other general transcription factors (Horikoshi et al., 1988; Choy and Green, 1993). a rate limiti 1991). Aci Roeder, 19 increase tr First, the a creating a VP16 to Tl that Opens binding of release d moleculec mmmdmr fOHOWing rounds of 22 a-1-4) Actuators and 15118 TFIIB-association with TFIID-promoter DNA complex is a rate limiting step in transcription pre-initiation complex formation (Lin and Green, 1991). Acidic- (Choy and Green, 1994; Kim et al., 1994) and proline-rich (Kim and Roeder, 1994) activation domains can interact directly with TFIIB. How activators increase transcription rates by interacting with TFIIB may be a bipartate process. First, the amino- and carboxyl-tenninal domains of TFIIB interact with one another, creating a closed structure which prevents TFIIF/RNAPII recruitment. Binding of VP16 to TFIIB appears to induce a conformational change in the structure of TFIIB that opens up the molecule and exposes surfaces within TFIIB which are critical for binding of TFIIF/RNAPII complexes (Roberts and Green, 1994). Second, TFIIB release during transcriptional elongation interferes with recruitment of another molecule of RNAPII for a subsequent round of transcription initiation (Reines, 1991). Interactions of activators with TFIIB may increase the rate of TFIIB-reassociation, following promoter clearance, facilitating RNAPII-recruitment and subsequent rounds of transcription initiation. a-1-5) Actuators and IEIIA Several lines of evidence indicated that TFI IA stimulates activator-mediated transcription (Ma et al., 1993; Ozer et al., 1994; Yokomori et al., 1994). Transcriptional activators that interact with TFIIA include VP16, Sp1, NTF-1, and Zta. The role of TFIIA in activated transcription may be two fold. (1) TFIIA may act as an such Zta. 1993). ( mmmdm activator-' Liebermar TFIIA-pror increased result of E include int and the ac protein lots the TBP 3L are require Colitacts rr. interactiOnS the “time n ‘ ConSlSiEnt V 23 act as an anti-repressor of TAF subunits of TFIID and in conjunction with activators, such Zta, overcome a slow step in pre-initiation complex formation (Chi and Carey, 1993). (2) TFIIA may increase stability of the TFIID-TFIIB-promoter complex. Interactions of TFIIA with transcriptional activators may facilitate formation of activator-TFIlD-TFIIA-promoter complexes (Wang et al., 1992; Chi and Carey, 1993; Lieberman and Berk, 1994). Transcriptional activators stabilize the activator-TFIID- TFIIA-promoter complex relative to the TFIID-TFllA-promoter complex. The increased stability of the activator-TFIID-TFIIA-promoter complex is likely to be the result of DNA-protein and protein-protein interactions. DNA-protein interactions include interactions between the activator DNA-binding domain and its binding sites and the activation domain-induced downstream TFIID-DNA interactions. Protein- protein interactions include interactions between the activator activation domain and the TBP subunit of TFIID (Lieberman and Berk, 1991). Because TAFs and TFIIA are required for formation of stable activator-TFIlD-TFIIA complexes, additional contacts may occur between the activator activation domain, TFIIA, and TAFs. Interactions of TFIIA with TAF110 of Dmsophila TFIID (Yokomori et al., 1993a), and the human co-activator PC4 (Ge and Roeder, 1994, Kretzschmar et al., 1994) are consistent with a specialized role of TFIIA in activation. a-2)Acti1atorsandpromoterescapebyRNAEll a-2-1)Acti1atorsandIEllH TFIIH is the only GTF known to possess enzymatic activities repair (Dr: with trans: VP16 and et al., 195 transcriptii transcriptic rate of trar Several ac elongation F et al., 1994) the ehicienc transcription These factoI “Woman With TFIIH. bi VPj I 24 activities involved in transcription initiation, transcriptional elongation, and DNA repair (Drapkin and Reinberg, 1994). Several reports indicate that TFIIH interacts with transcription activator proteins. These include interactions of TFIIH with HSV-1 VP16 and cellular protein p53 (Xiao et al., 1994), and with the EBV EBNA2 (Tong et al., 1995). These interactions may influence the role(s) TFIIH plays during transcription, although no stimulation of TFIIH-associated enzymatic activities by transcriptional regulatory proteins has yet been reported. a-3)Acti1ators and elongation and termination Activators can stimulate transcription by increasing not only the rate of transcription initiation but also the efficiency of transcriptional elongation. Several activators have been implicated in regulation of the transcriptional elongation phase. These include GAL4-VP16, GAL4-AH, and GAL4-E1 a (Yankulov et al., 1994). The mechanism by which transcriptional regulatory proteins increase the efficiency of elongation is unclear. Activators can, however, stimulate RNAPII transcriptional elongation by sequestering factors that are required during this step. These factors include Sll, TFIIF, and SIII/elongin. Transcriptional activators can also stimulate transcription-coupled nucleotide excision repair (NER) by interacting with TFIIH. b) Mmefamilymembers Mechanisms of eukaryotic gene regulation often involve associations betwee enhanc and pro insight i simplex VP16 (al comprise VP16 is assembly Virion env. ii Specific; 25 between protein factors and cis-regulatory elements within gene promoters and enhancers. Often, viruses exploit cellular mechanisms of gene regulation to control and promote expression of their own genes. In this regard, viruses offer a unique insight into eukaryotic transcriptional control mechanisms. In the case of herpes simplex virus type 1 (HSV-1), immediate-early (IE) genes are transactivated by VP16 (also called Vmw65, ICP25, or-TIF) (Campbell et al., 1984). HSV-1 VP16 is comprised of 490 amino acids and encodes a gene product of 54 kDa. HSV-1 VP16 is synthesized during the late phase of gene expresion. During virion assembly, HSV-1 VP16 is incorporated into the tegument between the capsid and virion envelope. HSV-1 VP16 is subsequently released during infection, whereupon it specifically induces transcription of viral IE genes (Post et al., 1981, Campbell et al., 1984). Specific induction of IE genes by VP16 requires at least one cis-acting DNA sequence motif, TAATGARAT (R = purine) (McKnight et al., 1987; O'Hare and Goding, 1988). HSV-1 VP16, which has no substantial affinity for double-stranded DNA (Marsden et al., 1987), functions by forming a multi-component complex on TAATGARAT sites in IE gene promoters (T riezenberg et al., 1988; Werstuck and Capone, 1989), together with the cellular POU domain protein Oct-1 and at least one other cellular factor called CFF, VCAF, or HCF (Katan et al., 1990; Xiao and Capone, 1990; Kristie et al., 1989). Mutational analysis of HSV-1 VP16 has identified a highly acidic domain (defined by 80 amino acids at the carboxyl terminus) as a potent transcriptional activating region (T riezenberg et al., 1988; Greaves and O'Hare, 1989). Activation domains have been typically classified into acidic (HSV Jun), and 9' based upor Recent stuc artificial dor more sensit acids than t aromatic re acidic-rich p (Blair et al., al., 1994) a Pattern of b, were the m mUtationai s phenl’ialanir transcription Pile“? With l whereas Sui: HSV‘i VP15 gel/Ere These i’TClud, 26 acidic (HSV-1 VP16, yeast GAL4 and GCN4), glutamine-rich (Sp1, Oct-1, Oct-2 and Jun), and proline-rich (CTF/NF 1) (Mitchell and Tjian, 1989;). However, classification based upon the most conspicuous attributes of these domains may be deceptive. Recent studies on VP16 (Regier et al., 1993; Cress and Triezenberg, 1991) and the artificial domain AH (Ruden, 1994) uncovered that most activation domains are more sensitive to alterations in specific patterns of hydrophobic and aromatic amino acids than to mutations of acidic amino acids. The importance of hydrophobic and aromatic residues for activation is supported by studies of the mammalian cell acidic-rich protein p53 (Lin et al., 1994), the foamyvirus acidic-rich activator Bel-1 (Blair et al., 1994), as well as the prototypical glutamine-rich activators Sp1 (Gill et al., 1994) and Oct-1 (Tanaka and Herr, 1994). Thus, it appears that a specific pattern of bulky hydrophobic and aromatic amino acids may be more critical than were the more conspicuous and abundant acidic amino acids. In this regard, mutational studies of the HSV-1 VP16 acidic activation domain uncovered that a phenylalanine residue at position 442 (Phe‘m) is critical for HSV-1 VP16 transcriptional activity (Cress and Triezenberg, 1991). Substitutions of HSV-1 Phe442 with hydrophobic and aromatic amino acids restored HSV-1 VP16 activity, whereas substitutions of HSV-1 Phe442 with non-aromatic amino acids abrogated HSV-1 VP16 activity (Cress and Triezenberg, 1991). Several VP16 homologs from numerous herpesviruses have been identified. These include VP16 from several strains of HSV-1 (Pellet et al., 1985; Dalrymple et al., 1985), HSV-2 (Cress and Triezenberg, 1990), Varicalla-zoster virus (VZV) 27 (Davison and Scott, 1986; McKee et al., 1990), equid herpesvirus type 1 (EHV-1) (Purewal et al., 1992), equid herpes virus type 4 (EHV-4) (Purewal et al., 1994), BHV-1 (Carpenter and Misra, 1992), and MDV VP16 (Yanagida et al., 1993). VZV open reading frame 10 (ORF10) is the most thoroughly studied among these VP16 homologs. VZV encodes a protein of 410 amino acids that is homologous to HSV- 1 VP16 (Davison and Scott, 1986). While VZV ORF10 and HSV-1 VP16 show considerable amino acid homology within their amino-tenninal regions, VZV ORF10 is 80 amino acids shorter than VP16, lacking sequences similar to that of the HSV-1 VP16 acidic activation domain (McKee et al., 1990). A recent study (Moriuchi et al., 1993) shows that the VZV ORF10 protein acts as a transactivator for both VZV and HSV-1 IE promoters in transient-expression assays. VZV ORF10 can also substitute for the transactivation function of HSV-1 VP16. Cell lines expressing VZV ORF 10 are able to complement an HSV-1 VP16 mutant which lacks the transactivation function of VP16 (Moriuchi et al., 1993). Interestingly, the transcriptional activation domain of VZV ORF10 maps to the amino—terminal region of the protein (Moriuchi et al., 1995). Hydrophobic cluster analysis (HCA) and sequence alignment of HSV-1 VP16 and VZV ORF10, using the HSV—1 VP16 Phe442 as a guide, revealed conservation of a pattern of bulky hydrophobic residues. Of particular interest were bulky hydrophobic clusters surrounding the VZV ORF10 Phe28 residue (Moriuchi et al., 1995). Similar to Phe442 of HSV-1 VP16, substitutions of VZV ORF 10 Phe28 with hydrophobic and aromatic amino acids preserved VZV ORF10 activity whereas substitutions of VZV ORF10 Phe28 with non-aromatic amino 28 acids abolished VZV ORF10 transactivation function, similar to that observed with HSV-1 VP16 (Moriuchi et al., 1995). Thus it appears that a common transactivation mechanism may exist between VZV and HSV-1 VP16 gene family members, despite significant differences in protein sequences and spatial arrangements of activator domains. BHV-1 VP16 is also capable of transactivating IE gene promoters (Misra et al., 1994). Unlike the activation domain of the HSV-1 VP16, the carboxyl-terminus of BHV-1 VP16 is largely hydrophobic in character (Misra et al., 1994). When fused to the DNA-binding domain of GAL4, the BHV-1 VP16 transactivation domain displays a poor stimulation of promoters containing GAL4 binding sites (Misra et al., 1994). Deletions of the entire C-tenninus of BHV-1 VP16 does not entirely destroy its transactivation functions (Misra et al., 1994), indicating that other sequences contribute to gene activation. EHV-1 gene12, another homolog of the HSV-1 VP16, is 479 amino acids in length. Like VZV ORF10, EHV-1 gene12 lacks sequences similar to those of the HSV-1 VP16 acidic transactivation domain (T elford et al., 1992). EHV-1 gene12 is, however, capable of transactivating both homologous and heterologous (HSV-1) IE gene promoters. Mutational analysis of the EHV-1 gene12 have identified the C- terminal seven amino acids, that are similar to the extreme C-terrninal region of HSV-1 VP16, as a potent transactivation subdomain. MDV VP16 is another homolog of the HSV-1 VP16 gene (Yanagida et al., 1993; Boussaha and Coussens, unpublished data). Identification and functional 29 characterization of MDV VP16 are the focus of this thesis project. Chapter 2 Marek's disease virus 1) - Historical perspective Marek's disease (MD) is a naturally occurring lymphoproliferative and peripheral nerve demyelinating disease of chickens (Marek, 1907). Lesions in the nervous system, described as a “polyneuritis”, were first reported in 1907 by Joseph Marek (Marek, 1907). Lymphomas involving the gonads, muscle, skin, and various visceral organs are attributes of MD (Payne, 1982). During the early 1960's, it was discovered that the etiologic agent of MD is a highly cell-associated avian herpesvirus, Marek’s disease virus (MDV) (Churchill and Biggs, 1967; Nazerian and Burrnester, 1968). MDV is transmitted horizontally in dust and dander, and has high mortality which led to condemnation of infected birds, until development of successful live-virus vaccines in the early 1970's (Churchill et al., 1969; Okazaki et al., 1970). Prior to widespread vaccination, MD was a disease of devastating economic proportions in the poultry industry. MDV remains as an important pathogen in intensive poultry operations, particularly with the emergence of very virulent strains of MDV that are resistant to current vaccines (Writer, 1985). For several decades, there was a great deal of confusion regarding the nature of MD. Based on similarities between many of the resulting gross lesions, MD lymphoma (caused by a herpesvirus) and lymphoid Ieukosis (caused by a retrovirus), were once classified under the same heading, "Avian Leukosis 30 Complex 1961; C lymphon neoplasr primarily Avian Le 1967; So as to whr neOplast intermixe exIlerime generally some Sig the natur SiSillificar transmis, lympilom etiologic; (MDV) (c 1958); a. attenUateC turkeys (o 31 Complex" (Anonymous, 1951; Cottral, 1952; Chubb and Gordon, 1957; Biggs, 1961; Campbell, 1954, 1961). It was not until the late 1960's that visceral lymphomas were shown to include two pathologically discrete lymphoid neoplasmas. MD lymphoma is now known to be caused by a herpesvirus and is primarily restricted to T-cells, while lymphoid Ieukosis is caused by a retrovirus, Avian Leukosis virus (ALV), and is largely restricted to B-cells (Churchill and Biggs, 1967; Solomon et al, 1968; Nazerian and Burrnester, 1968). Confusion also existed as to whether MD lesions were inflammatory or neoplastic. Both inflammatory and neoplastic lesions occur in affected nerves of infected birds and are sometimes intermixed. Finally, there was confusion over transmissibility of MD. Transmission experiments using tumor or blood cells from affected birds gave variable and generally controversial results (Olson, 1940). In spite of all of these difficulties, some significant research, in the 1960's, contributed to a greater understanding of the nature of MD, and led to virtual control of the disease by vaccination. Three significant events were accomplished during this period: (1) The experimental transmission of MD by inoculation of susceptible young chicks with blood or lymphoma cells from infected birds (Biggs and Payne, 1963); (2) Recognition of the etiologic agent of MD as a highly cell-associated herpesvirus, Marek’s disease virus (MDV) (Churchill and Biggs, 1967; Solomon et al., 1968; Nazerian and Burrnester, 1968); and (3) production of live virus vaccines containing either tissue-culture- attenuated MDV (Churchill et al., 1969) or the antigenically related herpesvirus of turkeys (Okazaki et al., 1970). 32 In addition to its enormous economic relevance to the commercial poultry industry, MD has served as an exquisite model for herpesvirus oncology. MDV is one of numerous herpesviruses that are involved in the induction of tumors in their natural hosts. These herpesviruses include several human herpesviruses (Epstein- Barr virus, human cytomegalovirus, and human herpesvirus type 6), as well as representatives from a long list of animal herpesviruses (Zur Hausen, 1980; Nahmias and Norild, 1980; Naegle and Granoff, 1980, Falk, 1980, Hinze and Chipman, 1971). Unfortunately, many of these oncogenic herpesviruses cannot be studied experimentally in their natural host system. This is especially true for human herpesviruses where there is no appropriate experimental model. MD, however, has captured attention as a model for herpesvirus oncology for several major reasons. Well-characterized genetic strains of chickens have been produced that range from extremely susceptible to remarkably resistant to the disease. Specific pathogen-free experimental animals are readily available. Third, a large spectrum of virus strains ranging from very virulent to non-oncogenic and non- pathogenic have been isolated. Finally, MD is a naturally occurring disease that can be reproduced experimentally using natural methods of exposure in the natural host. Various virological and immunological aspects of MDV infection can serve as a model of herpesvirus infections in human and in other animals. The development of herpes virus of turkey (HVT), or attenuated MDV as effective vaccines against the disease, provides the first case of a means to control a naturally occurring malignant lymphomatous disease by vaccination in any species (Churchill et al., 1969; 33 Okazaki et al., 1970). 2) - Biology of Marek's disease virus A) - Virus morphology and morphogenesis MDV is classified as a Gal/id herpesvirus 1 (Matthews, 1979). The MDV genome is a linear double-stranded DNA molecule with a molecular weight of 1.2 x 10" (Lee et al., 1971), which is slightly larger than that of the prototype herpes simplex virus (HSV). In sucrose gradients, the sedimentation coefficient of MDV DNA is 568 when compared to that of T4 phage DNA and HSV DNA (Lee et al., 1971). However, in alkaline gradients, the sedimentation coefficient of MDV DNA was determined to be 708 with a molecular weight of 6 x 107. The differences in S values implied the presence of gaps and/or nicks in the MDV DNA molecule (Lee et al., 1971). The buoyant density of MDV DNA in CsCl, is 1.705 g/cma, corresponding to an overall guanine + cytosine (G+C) content of approximately 46% (Lee et al., 1971 ), similar to the G+C content of chicken genomes. Similarity of G+C contents prevents effective separation of viral DNA from cellular DNA by buoyant density centrifugation. The MDV capsid consists of 162 capsomeres arranged in icosahedral symmetry (Nazerian and Burmester, 1968). MDV nucleocapsids consist of a core of intact viral DNA, approximately 85-100 nm in diameter, and a protein shell, the capsid, assembled in the nucli of infected cells. Few or no enveloped virions are produced in most cell types, but rather, naked intranuclear particles are present in 34 infected cells (Nazerian and Burmester, 1968; Hamdy et al., 1974). The feather- follicle epithelium is unique in that it is the only tissue in which MDV infection is fully productive (Calnek et al., 1970; Witter et al., 1972). Enveloped virions are produced in cells undergoing keratinization, and these virions are spread when feathers are molted or when dead cells are lost in the form of dander. These cell- free particles contain infectious virions which contaminate the environment and therefore are important epizootiologically. B) - Marek’s disease virus serotypes MDV has been classified into three serotypes based on agar-gel precipitin (AGP) tests and indirect immunofluorescence (IF) antibody assays (Bulow and Biggs, 1975; Schat and Calnek, 1978). MDV serotype 1 (MDV1), includes all oncogenic viruses and their attenuated derivatives. These include very virulent (wMDV), virulent, and attenuated strains of MDV. wMDV strains, such as RB1 B and Md/5, are responsible for many outbreaks of MDV in vaccinated chickens. Only vaccinated genetically resistant lines of chickens are resistive to wMDV infection. MDV strains GA and JM are classified as virulent strains of MDV and can result in a high incidence of MD in genetically susceptible, unvaccinated chickens. Continued in intro passage of wMDV and virulent MDV isolates attenuates their oncogenic potential, modifies their growth in yitro and their antigenic features. Following attenuation, these viruses are immunogenic, and appear to be protective when used as vaccines against oncogenic MDV strains (Churchill et al., 1969). 35 MDV serotype 2 (MDV2), consists of non-oncogenic, naturally occurring chicken herpesviruses. A cell-associated lymphotrcpic herpesvirus of turkey (HVT), comprises a third serologic type (MDV3) related to, but different from, both MDV serotypes 1 and 2. HVT is apathogenic for both chickens and turkeys and is used as an effective vaccine against MD in chickens. MDV2, MDV3, and attenuated MDV1 have been successfully used individually and in combination to produce monovalent, bivalent, or trivalent vaccines against MDV-induced tumors (Churchill et al., 1969; Okazaki et al., 1970; Witter, 1985). C) - Source of Marek’s disease virus MDV is present in a cell-associated form in many tissues of MDV- infected chickens (Philips and Biggs, 1972). Non-enveloped MDV can be isolated from viable whole cells, buffy coat cells, kidney cells, or lymphoma cells (Witter et al., 1969). The feather-follicle epithelium is the only tissue from which cell-free MDV can be produced. Cell-associated and cell-free MDV can be propagated in primary fibroblast cells obtained from various avian embryos. Cell-associated and cell-free MDV produce cytopathic plaques characteristic of herpesviruses within a few days when inoculated onto tissue-culture monolayers of chick kidney cells (CKC) (Churchill, 1968), duck embryo fibroblasts (Solomon et al., 1968), and chick embryo fibroblasts (CEF) (Nazerian, 1970). Plaques consist of foci of refractile rounded or fusiforrn cells (Churchill, 1968). Initial absorption of both cell-associated and cell- free MDV to cultured cells is rapid. Approximately, 50% of virus is absorbed within 36 30 minutes at 37°C (Churchill and Biggs, 1967; Sharma et al., 1969). 3) - Pathology of Marek’s disease virus The understanding of MDV pathology is aided by knowledge of the types of virus-cell interactions that occur. These interactions consist of (1) productive infections, in which virions are completely or partially formed, resulting in cell death; and (2) non-productive infections, in which there is either none or very limited expression of viral genomes, resulting in viability of infected cells and, in some instances, neoplastic transformation. A) - Productive infection Productive infections are characterized by synthesis of viral DNA, viral proteins, and virions. Productive infections can be subdivided into fully-productive and semi-productive infections. F ully-productive infections by MDV only occur in the feather-follicle epithelium, from which cell-free infectious virions can be produced (Calnek et al., 1970a; Nazerian and Witter, 1970). Production of cell-free virions from feather-follicle epithelium is associated with envelopment of virions in cytoplasmic inclusion bodies. Semi-productive MDV infections (restrictive or abortive infection) are seen in essentially all other tissues, including spleen, bursa, thymus and peripheral blood lymphocytes (Calnek et al., 1970; Schat et al., 1978). Infected tissues contain cells that express viral antigens, naked virions in the nucleus, and limited enveloped virions, mainly associated with nuclear membranes. 37 Virus is not released in an infectious form, rather, it spreads from cell to cell. Infection of in intro-cultured cells by MDV is also of the semi-productive type and is closely cell-associated. B) - Non-productive infection Non-productive infections are usually characterized as latent infections and occur primarily in transformed T-cells. Latent infections are first detected at the end of the early cytolytic infection cycle. Latent infections are usually restricted to T-lymphocytes and are characterized by the presence of intact virus genomes, with little or no active viral gene expression. In MDV and HVT, apparently latent infections of splenic and peripheral blood cells persist, probably, for the lifetime of the bird. Virus particles are not observed in these cells in Miro. Yet virus can be rescued in Mitre (Adlinger and Calnek, 1973; Payne and Rennie, 1973; Calnek et al., 1970), and by inoculation of susceptible birds. The ultimate response in serotype 1 MDV-induced MD is transformation. Transformation by MDV appears to be restricted to T-lymphocytes (Calnek et al., 1970; Nazerian, 1973). C) - Pathogenesis The pattern of events which occurs sequentially in antibody-free, genetically susceptible chickens which ultimately die from tumors after infection with an oncogenic strain of MDV can be generally divided into four stages: (1) early cytolytic infection; (2) latent infection; (3) permanent immunosuppression and late .38 cytolytic infection ; and (4) transformation. MDV infection usually occurs via the respiratory tract. As early as 1 to 2 days post-infection (DPI), virus can be detected in spleen, thymus and bursa of Fabricius. By 3 days post-infection, a productive-restrictive infection is detected. Few or no enveloped virions are produced. Rather, naked intranuclear particles are present in infected cells. The necrotizing effects of early infection lead to acute inflammatory changes, termed “acute reticulitis” (Payne et al., 1976). Histologic features in thymus and bursa include cytolysis of lymphocytes, loss of cortical cells (thymus) or follicular structure (bursa), infiltration by macrophages and granulocytes, and reticular cell hyperplasia. The result is atrophy of bursa and thymus, and consequently immunosuppression. Early cytolytic infection reaches a peak by 4 to 5 DPI and declines by 6 to 7 DPI (Shek et al., 1983). B-cells are found to be a primary early target for cytolytic MDV infection. Five to seven days post-infection, there is a switch in the type of infection seen in lymphocyte populations. During the second week of infection when virus activity subsides in lymphoid organs, a variety of other tissues, notably those of epithelial origin, become infected. Focal necrosis and intranuclear inclusion bodies may be seen in many organs, including the kidney, pancreas, adrenal gland, and proventriculus (Calnek and Hitchmer, 1969). These focal sites typically exhibit productive-restrictive infections. At the same time that these areas are developing focal infections, there is a reappearance of infection in the central lymphoid organs. Lymphocytes may become cytolytically infected at these sites. At about 2 to 3 weeks post-infection, a permanent immunosuppression involving bot' immunosupr cytolytic infe lymphocytes muscle, and al., 1976; C (1) Type I lymphoblas Schwann c scattered ir cell prolifer iris (leadin tumors) (F deveIOpm. Rennie, 1! D) are claSS ehiargem‘ are USUall a gray 0r 39 involving both humoral and cell-mediated immune responses develops. Permanent immunosuppression could result from loss of responsive lymphocytes due to cytolytic infections. The ultimate response in MD is neoplastic transformation of lymphocytes. Gross lymphomas involving almost any of the visceral organs, skin, muscle, and nerves develop in most cases after 3 weeks post-infection (Payne et al., 1976; Calnek, 1980). There are two types of lesions (Payne and Biggs, 1967): (1) Type A (neoplastic) lesions consist of both infiltrative and proliferative lymphoblastic cells, reticulum cells, small and medium cells, and sometimes Schwann cells; and (2) type B (inflammatory) lesions which are characterized by scattered infiltrative lymphocytes, plasma cells, and macrophages, and Schwann cell proliferation. Other sites of infiltrative and/or proliferative lesions in MD are the iris (leading to blindness), muscle and skin (resulting in grayish-white lymphoid tumors) (Payne et al., 1976; Calnek and Vtfitter, 1978). Blood changes include development of lymphomatosis due to increased numbers of T cells (Payne and Rennie, 1967; Evans and Patterson, 1971). D) - Gross lesions Clinical signs of MD usually appear at 3 to 4 weeks post-infection, and are classified as either classical or acute. Classical MD is characterized by enlargement of the celiac, autonomic, and other peripheral nerves. Affected nerves are usually 2 to 3 times their normal thickness, lose their cross-striations, and show a gray or yellow discoloration (Goodchild, 1969; Sugiyama et al., 1973). Nerve enlargement acute form of including the muscle are these organ in bursal a Rennie, 1E atrophy of infection ( 4i - The aDproxi (Lee e: WUSQY lympl has gen. 10. 40 enlargement is caused by lymphomatous and/or inflammatory infiltrations. The acute form of MD is usually characterized by lymphoma formation. Various organs including the ovary, testis, liver, spleen, kidney, heart, lung, skin, and skeletal muscle are common targets for acute infection. Tumors may become visible in these organs by 3 to 4 weeks post-infection. By 5 weeks post-infection, reduction in bursal and thymus weight may occur, due to lymphoid atrOphy (Payne and Rennie, 1973). Very virulent strains of MDV (wMDV), cause even more severe atrophy of the bursa and thymus, and may result in cell death by 8 to 10 days post- infection (Witter et al., 1980). 4) - The molecular biology of MDV A) - MDV genome structure The MDV genomic DNA is a linear double-stranded DNA molecule of approximately 165 to 180 kilo-basepairs (kbp) in length and contains nicks and gaps (Lee et al., 1971; Hirai et al., 1979; Cebrian et al., 1982; Fukuchi et al., 1984; Wilson and Coussens, 1991). Originally, MDV was classified as a gamma-herpesvirus based on its lymphotrcpic nature, similar to Epstein-Barr virus (EBV). Recently, however, MDV has been reclassified as an alpha-herpesvirus based on genome structure and gene collinearity with other alpha-herpesviruses, including herpes simplex virus type 1 (HSV-1) and varicella-zoster virus (VZV) (Buckmaster et al., 1988; Roizman 1992; Brunovskis and Velicer, 1992; Velicer and Brunovskis, 1992). Similar to HSV-1 and 41 VZV, the genome structure of MDV belongs to the Herpesviridae group E genome family (Cebrian et al., 1982). MDV genomes consist of covalently linked unique long (UL) and unique short (Us) regions, each flanked by internal inverted repeats (IRL and IRS) and terminal repeats (T RL and TRS) (Cebrian et al., 1982; Fukuchi et al., 1984). The MDV DNA molecule also contains several direct repeats (DR1-5 sequences). These DR sequences are located within the internal or terminal repeat regions (Hirai, 1988). DR1 is a tandem direct repeat of a 132 bp repeat unit located within the TRL and IRL of BamHI D and H, respectively (Maotani et al., 1986). Copy number of the 132 bp repeat unit within TRL and IRL regions of oncogenic MDV strains is one to three each, while that of attenuated derivatives is 3 to 100. For example wMDV strain Md5 has only two 132 bp repeat units. MDV1 DNA in different lymphoblastoid cell lines contain either two or three repeat units. These findings suggested that two copies of the 132 bp repeats may be necessary or sufficient for induction and maintenance of oncogenic transformation by MDV1. DR2 is located within the BamHI F fragment of the UL region and consists of a direct repeat of approximately 1.4 Kbp (Fukuchi et al., 1984). DR2 does not appear to differ in size and number of repeats between oncogenic and attenuated forms of MDV1 strain DNAs. DR3 is located within IRS, and TRs adjacent to the U-S junction and terminal repeats. DR3 consists of a direct 178 bp repeat. The 178 bp sequence of DR3 does not have any homology to the 132 bp sequence of DR1 and is amplified as much as 50 fold during viral replication of both oncogenic and non- oncogenic MDV1 strain DNAs. DR4 is located within IRS and TRS adjacent to the 42 US and consists of a direct 200 bp repeat (Hirai et al., 1984). DR3 and DR4 may not be associated with a loss of oncogenicity. DR5 is a putative terminal direct repeat at each end of the MDV DNA molecule. DR5 may contain signals for cleavage of replicative-form genomes to yield virion DNA. B) - Physical map of Marek's disease virus Physical maps of the three MDV serotypes have been reported (Fukuchi et al., 1984; Ono et al., 1992). Restriction endonuclease patterns are very different between the three serotypes, despite their antigenic similarities (Hirai et al., 1979; Ross et al., 1983). Initially MDV serotype 1 and HVT viral DNAs were shown to share less than 5% homology by DNA-DNA reassociation kenitics and Southern blot hybridization (Hirai et al., 1979). Gibbs et al. (1984), however, estimated that the homology between DNAs of MDV serotype 1 and HVT under less-stringent conditions ranged from 70 to 80% at the nucleotide level. Restriction enzyme profiles of the MDV serotype 2 DNA were also shown to differ greatly from those of the MDV serotype 1 and HVT DNAs (Hirai et al., 1984). Gene identification and mapping indicated that genes encoded in the unique long and unique short regions of the MDV genome are collinear with those of HSV-1 and VZV (Buckmaster et al., 1988; Brunovskis and Velicer, 1992). To date, two MDV glycoproteins, A and B antigen, have been characterized as HSV-1 9C and 9B homologs, respectively (Coussens and Velicer, 1988; Isfort et al., 1987; Ross et al., 1989; Chen and Velicer, 1992). Thirty-five MDV genes were also identified 43 by comparison to VariceIIa-zoster virus (VZV) (Buckmaster et al., 1988). Recently, homologs of HSV ICP4, ICP27, VP16 and 9K have been identified and sequenced from serotype 1 MDV (Anderson et al., 1992; Yanagida et al., 1993; Ren et al., 1994). Genes located in the repeat regions are highly unique and specific to MDV. In this regard, a 38 kDa phosphoprotein (pp38) and a fosflun oncogene homolog (meq), both expressed in MDV transformed cell lines, were identified within the BamHI H and BamHI l2 fragments of the IRL region of MDV genomes (Chen et al., 1992; Cui et al., 1991). A 14 kDa MDV protein encoded by a cDNA spanning BamHl H and I2 regions has also been identified (Hong and Coussens, 1994). The potential relationship between unique genes in MDV and MDV tumorigenicity has attracted much interest in further exploring MDV repeat regions. 5) - MDV gene expression As with other herpesviruses, MDV gene expression is regulated in a cascade fashion (Maray et al., 1988). Three major kenetic classes of MDV genes are expressed: (1) immediate-early (IE or or), (2) early (E or [3), and (3) late genes (L or v). A) - MDV Immediate-early genes Immediate-early genes are expressed immediately after infection. IE gene expression does not require de norm viral protein synthesis and hence IE mRNAs are synthesized in the presence of metabolic inhibitors such as 44 cyclohexamide (CHX). IE gene products are required for subsequent activation of early and late virus gene expression, and feedback regulation of their own expression. Two MDV IE genes homologous to HSV-1 ICP4 and ICP27 were identified within the MDV genome (Anderson et al., 1992; Ren et al., 1994). The MDV ICP4 genes are located within the IRs and TRS regions of the MDV genome. The MDV ICP4 is 4,245 nucleotides in size (Anderson et al., 1992). The predicted structure of the MDV ICP4 gene product is similar to its counterparts in HSV-1 and VZV. In fact, the MDV ICP4 gene product contains five regions in which regions 2 and 4 are the most conserved, and an amino-terminal serine-rich domain (Anderson et al., 1992). The serine-rich domain of MDV ICP4 is unique in that it is flanked on both sides by proline- and basic-rich stretches. Whereas the serine-rich region of the HSV-1 and VZV ICP4 products are preceded by proline- and basic-rich amino acid residues, but followed by acidic rich regions. A number of potential regulatory sites were identified within or adjacent to the MDV ICP4 sequence. These include an ICP4 binding site, Oct-1 site, and TAATn3A sequence similar to the HSV-1 VP16/Oct1 recognizing motif, TAATGARAT (Anderson et al., 1992). Although the MDV ICP4 gene product is capable of homologous promoter activation, the precise function of MDV ICP4 in MDV infected cells is still unclear. The MDV homolog of the HSV-1 ICP27 gene is located within the EcoRI B fragment of the MDV genome (Ren et al., 1994). The MDV ICP27 gene is 1,419 nucleotides in size and could potentially produce a product with a molecular weight 45 of 54.5 kDa (Ren et al., 1994). The carboxyl-terminal region of MDV ICP27, which likely contains its functional domain, is highly conserved when compared to other herpesviruses, including HSV-1 and VZV. The carboxyl—terminal domain of MDV ICP27 contains a zinc-finger motif highly similar to that of the HSV-1 ICP27. The Zing-finger motif of the HSV-1 ICP27 gene product is involved in DNA, RNA, and protein-protein interactions (Smith et al., 1991). Intensive research is now in progress to investigate the functional properties of MDV ICP27. A MDV unique IE gene - pp14 - has been recently identified within the BamHI l2 fragment of the MDV genome (Hong et al., 1994). The pp14 gene encodes a product with a molecular weight of 14 kDa and is expressed in Iytically-infected cells with oncogenic strains (GA and Md11), or their attenuated derivatives, as well as in Iatently MDV-infected and transformed MSB—1 cell lines. B) - MDV Early genes Early genes are the next class of genes expressed in lytic MDV infection and their synthesis requires the activity of at least one IE protein. Early genes encode proteins required for nucleotide precursor metabolism, and viral DNA replication. Expression of E genes is enhanced in the presence of drugs that inhibit viral DNA synthesis, such as phosphonoacetic acid (PAA). Several MDV early genes homolgous to those of HSV-1 have been identified. These include thymidine kinase (T K), DNA polymerase, DNA-binding protein, etc. (Buckmaster et al., 1988). A MDV unique gene, phosphoprotein 38 (pp38), has also been identified within the BamHl H frag has been ide Nakajima el expressed ir cells (lkutae and is relati (Nakajima . reported tl POiypeptidi 9938 (Cui I c). protein syi re I 4:444 4.9-0 4444444 4.44.44.44.40 Po 4.4 4. .4 44> 4 4.. .4. :00 Q 4.44.04.44.— 44.444444 M I I I B I HMO .A I I Iwm I HM I I I IMm I NM 0 .1 A 0 no... l--- .....44 o3 t--- ---------- a: t--- .....44440 a: 9444 440.44.44.44“. one i... ---------- 44» aoo44z4444 o4>4444444 4444444444 44» ..... -i--- ---------- ---------- 4.44 94----4444 444.4o4o444 040.444.4444 ..... -x.----$....$-:::-: 444 4444044444 4444o4 44c 4444444444 --- ..... ----- ---------- 4404944404 44> ---------- ---------- 449444444o 4 4 4 4. 4 4 4 4 > 4 > 4M4 4 4 4 4403??» 4 WIN..- - - N4 4 4M. - - - - l 4 4. .4 4 o as? 4 4 4 4 adheumu 4 4.4 4.....- - 44....4 .44,” 4.4.4 44aa4>o>4.M.4 44>4n,,..wmmwm>4 “4,4.M---mm444w 44444449404444 4444......444044 444._.444W.,.44fl . ..----....-44. 44994444404 «Hafiz—44444 .- 4 ‘I _.4.4M4 43o-44.wi44m443 mmM444amoo4 4444....o444 44M w..mz4>>44wawmm ..a.$4>44---- 44.43.444.44”... 44M4 4M004444flmhm ,4 $4444.44... 1.1.4.}.44444444. -444 .M9444-4.w..44.w .4 4W4>444444 M...4.,4M----44»4M 44m 44444a>4>...4M 144M444: - - -4o M4...4m4444.4.4.4.? A “molt 4.44 3044444404 ”.4... 44am“ 44$4444A4>9 “M44..4.,Q.Wa>4.4..mm4 44.4...>4oou.4..4..w 4.4...4M moMHammaMnom 4444.....44444? 4444404444.... >444... 4444440444.: .44....44...M4.4..444Mo 444444444444... 0 4 4M 4. >. .44 4.4 4W4 n o 44 4 4.4 4...? 4 .4 4.4% 4 4 41> .444 u 4 4.4.4 onn oHn non ann run me> >II OHHIO >N> NHZIO >II 9HH> >MI wdmb bat Udm> >II OHIIO >N> NHBIO >II wHH> >w= QHH> >9! uHm> >II OHIIO >N> NHZIO >II odmb >mx UHH> >9! wHA> >II OHIIO >N> NHBIG >II oHA> >mI oHA> >9! odm> >In OHIIO >N> NHZNO >II th> >II 0HA> bnx oHA> >II OHIIO >N> NHIIO >II me> >MI DHH> >98 85 Figure 5: Detection of the MDV VP16 transcript. Poly(A)*mRNA was isolated, transferred to supported nylon membranes, and hybridized as detailed in Materials 8 and Methods. (A) Map of the three MDV DNA probes used for Northern blot analysis. (B) RNA was hybridized with each of the DNA probes. Lane 1: uninfected CEF; Lane 2: CEF infected with MDV strain GAp9; Lane 3: CEF infected with MDV strain Md11p16. The primary (7.8, 4.6, and 2.5 Kb) MDV VP16 transcripts are indicated by an arrow to the right. Prediction of the map of the transcripts, according to our results and results published previously (Yanagida et al,1993) 86 MDV VP1 6 P1. _ P2 P3_ UL49 VP1 6 UL47 UL46 Ep—r—q—E ‘— II- ————'- 7. 2. 4. QUIO M '- 87 Figure 6: Detection of MDV VP16 gene product by Western blot analysis. Proteins were extracted from uninfected CEF and CEF infected with MDV GAp9 strain. 10 ug of proteins were subjected to Western blot analysis using a poly-clonal antibody raised against the HSDV-1 VP16 gene product. 88 Flgure 7: Map of pBKCMVP16. The 1.615 SpeI-Sall fragment, which contains the entire MDV VP16 coding region was cloned into pBK-CMV, as described in Materials and Methods. 89 3000 %2000- .% a 7’ 8 y; z / / 01000- é? 0 Z é émmfia :gegezesege 33%9%§%&% g m+ + + + + :2 5 5 a “a :2 .- E o. 9. FigureB: Transactivation of MDV and HSV-1 IE gene promoters by MDV VP16. CEF cells were cotransfected with 3 ug of plasmid DNA containing either the MDV ICP4 (from low and high passage), pp38, and 93 gene promoters or the HSV-1 ICP4 promoter linked to the CAT reporter gene, and 5 ug of pBKCMVP16. Controls were transfected with an amount of pBK-CMV DNA equivalent to that of pBKCMVP16. 90 Figure 9A: Schematic diagram indicating relative positions of potential transcription factor binding sites in the MDV ICP4 promoter sequences (upper figure). Deletion versions of the ICP4 are below the wild type MDV ICP4 gene promoter. 91 .01; IIOEOHMI 3552.3 in... . 43.5.5. 23. .9: 29.4.03. 7 . is... 4.352.. 3X70)... _ IE I 43.53. >5. .9: 8.6.94 3:33 92 Figure 98: Effect of deletions within the MDV ICP4 gene promoter sequences on basal and MDV VP16 activated transcription. Lane 1: mock, lane 2: pCAT, lane 3: pCAT+VP16. Lane 4 &5: MDV lCP4 in the absence and presence of VP16, respectively. Lanes 6 & 7: pSH3CAT in the absence and presence of VP16, respectively. Lanes 8 & 9: pAH3CAT in the absence and presence of VP16. Lanes 10 & 11: PSXbCAT in the absence and presence of VP16. 93 6000 5000 — — - 4000 3000 2000 «:58 .240 24> + 4404444 4.40864 24> + 440244 4.48144 24> + 4.404144 44854 24> + 44o. «40. 24> + 4404 4404 4.00! 94 Figure 10: Effect of point mutations within the MDV ICP4 promoter Oct-1 site on MDV VP16 activated transcription. Lane 1: mock, lane 2:pCAT, lane 3: pCAT+VP16. MDV ICP4, Oct1A-, abd Oct1B- expressions were assessed in the absence (lanes 4, 6, and 8), or in the presence on MDV VP16 (lanes 5, 7, and 9). 95 24> + 4.4 :02 24> + 3:59 @ .< 2.00 e40. 24>+>404 _ _ _ _ w m m m m 0 5 4 3 2 1 4:500 2&0 96 Figure 11A: Hydrophobic Cluster analysis of several HSV-1 VP16 gene homologs. HCA profiles show the characteristic horse shoe-shaped cluster and the centrally located Phe residues. 97 441 451 440 450 | 25 BHVUM8 VP16-2 VPl6-l EHV-l GENIZ MDVUIAS VZVGEN 10 98 Figure 118: Linear alignment of several HSV-1 VP16 homologs using the six hydrophobic residues surrounding HSV-1 VP16 Phe‘” residue as a guide. Shading indicates conserved residues. Boldface letters indicate conserved bulky hydrophobic residues. Dark shading denotes conserved acidic residues. Open ovals represent residues conserved among homologs but not present in HSV-1 VP16. 99 + 4.8.... W w/M 24> + <94 W%////////////% 2.3.... %///////////////% 24>+44o_ . A «42 // 3.0.964 + ._.