E s . . 4 k t. 1 >3... ..-,T‘L\.-L * v' -..azu.;.; géh 3M flay! ' .4 .-fi'.“’;‘ Kiir‘f‘, .399 ‘ ~11 Ms , . ' . ,_.' 1' fin "is“ I .‘ AA ‘- vqngr‘r; 1. WWW- ‘Jl up . a . 7 '.""-" J' “43:: 5‘4“ ~ . ‘.’» . ~ . {fiat-UV ,f: . . .v-4 .4 . . , ' "‘ >‘- ' I ‘ V 3 _ \ ’3‘ ‘ -‘. .ik . ,. r? . - um i‘ ‘ J'L . , pa “‘3 '9 ~ . \ A .~. ‘ . . u! E? “‘thh' : I, ~ 0 .ul ‘. . ‘ r \1 .. * .- _ M,» 2 J... ‘ .. . .I’ ' ‘>... 3:13? ‘.~“‘. C L‘".Z?o‘{w;zi::‘vr- ritizgém‘ " *3 .'iv“:‘,t’=‘-Tid§~zs 323': ‘ I'L‘t.’ ;_ . 2P 3: ~ > _ _ ‘ ,ii ' ‘,'.’fi',{i- "; ._‘ ~ =L’f;:?‘fi""~‘3~“* u t ._ “.5 ,.—.~m. ‘ .. ~ ~ 4 . ‘ fi ‘ t «WM .. .' 3:2 :2 ‘J V. n‘ qui " ¥~ 3 “‘01 ' v" {3.3.1 a)". “V f A ‘ '\ tfifiifi’fl‘ :4. ‘ ‘31:» T —u H 4 I. -A' u-u cm ..1 WH‘ ‘ 0" ' ' ‘l'.-'.’ .‘v’ 3% Az'tt‘fi‘ ) ;..'a . .1 O'- .r m - 2.‘ ... Mv on ,4: ”a A‘ ‘ nn' , . w. ‘ ,. £00?th . "I'u‘OCIO ‘ N“ ‘ . u .A.’ ‘. . _- ‘ J‘- . - r t'i‘hfl'pn . 'II-nO-U . V---" co. :‘125—2 * . 1'. '41:! w, flair!“ $311.» ' \‘v n [:2 . ‘1 F3"- . v {3" .'. . 3‘. » ~v— . «a X‘ ‘ 7‘ “,1; '1': ' ' 7 . a. L‘- p ‘. ~ D;I':«. ”on : ,5} ’: ",1 ‘7 .M’ga}? .‘fi-v. a '2: 1 ' .\ IL. ' J: 3?. "2r {.‘Qtt: ‘AI' ‘ ma: '0 O " 1‘. t ‘43? .I ,_ vw‘ff wt.’ J ‘ I’I’E 1 “ a.;:’ rm‘ ‘7. .v- n: f I: M _;..':*:‘- 3. TC- “31' ‘. ,-s.r..' - w.v .7 _..- , ';. §’ 'ul - ' s "‘1 D r d' f: 1 ,..- H... . , {£774 .4 .“ J . v wr . .. vag-T ,h- .._. 7 .ng- , - 3' ('_ ‘Piq r‘ ". . W .‘.‘_T , .': (31195539): _ .. ,. _' 3.319 ..~. ‘7 .. g 115-3 V r;>’:f.:sg'r1§" h : . '.. ‘ . r: ’i .' ' V ‘ _ 2 Ifi‘kfl gfigigiz'lrlfif, ' "1; 1 ' ;’~. ugfiplfiig'; ' 3?;51'31' ”Pa. ,. 5' : Iéié - ‘r " é "” ' 9W 3’: r" , ,i'éfI-g-Efiff’ . ~ . ' 4 1.7- ‘ . 3&3“: ; wk“; 5" v i 3.3%" . 8:3 ;3;W:¢u:§§?£{,rhg- ‘ R J" ' z _ . F . o't ‘ k . k; I ’1' if}?! V '1 ' . tan w 932’ = , 1 <.‘~—- ‘ V‘V...VVV. .. ..... w.-. nv w' 3 .‘.~._~. , ”—‘r—n. yuav-W. ~‘_ AN‘_‘ w...“ m5: Aft‘é ‘ t g. 3 1 F f! a." «1;. . . w ...~....... wag—fl m..-” '_... .1. I b , a V '- , .2.-.’LI§:.LZ.-Lnfa 25;? 2,5’23 ( r W LIBRARY , Michigan State IlllllllllllillHllHlHlllllllllllHlllllllllllllllllllllll L "“"m‘" , 841046 ' ~15» i‘ This is to certify that the dissertation entitled THE MAREK'S DISEASE HERPESVIRUS B ANTIGEN GLYPROTEIN COMPLEX: CHARACTERIZATION AND PROCESSING OF ITS PRECURSOR POLYPEPTIDE AND IDENTIFICATION AND CHARACTERIZATION OF THE GENE ENCODING IT presented by Idah Sithole has been accepted towards fulfillment of the requirements for Ph.D. Biochemistry degree in \Mv. Major fofessor my? 1/57! %M Q /l/£¢’M MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 —___‘_ __V___ PLACE N RETURN BOX to manual: chockoutfrom your record. TOAVOID FINESmunonorbda-oddodu. ll DATE DUE DATE DUE DATE DUE MSU In An Affirmdivo ActiorVEcpal Opportunity IMRIMOI‘I emails-o: THE MAREK'S DISEASE HERPESVIRUS B ANTIGEN GLYCOPROTEIN COMPLEX: CHARACTERIZATION AND PROCESSING OF ITS PRECURSOR POLYPEPTIDE AND IDENTIFICATION AND CHARACTERIZATION OF THE GENE ENCODING IT BY Idah Sithole A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1988 7 E 3v? r 5...; .J ABSTRACT BY Idah Sithole The Marek's disease herpesvirus (MDHV) B antigen (MDHV-B) glycoprotein complex was identified and molecularly characterized as a set of three glycoproteins, gp100, gp60, and gpu9, based on apparent molecular weight and immunoprecipitation with both polyclonal and monoclonal antibody. Tunicamycin inhibition of N-linked glycosylation revealed two putative precursor molecules of 88 kilodalton (kDa) (pr88) and UN kDa (prNU) molecular weight. Pulse-chase studies showed gplOO, a glycosylated intermediate, was processed to gp60 and ng9 and monensin inhibited this processing. Endo-BeN-acetylglucosaminidases F and H reduced gp100 to pr88 and reduced gpu9, and to a lesser extent gp60, to erU. The same monoclonal antibody recognized gp60, gpu9 and pruu. Cell-free translation of infected cell mRNA, and immunoprecipitation analysis using either polyclonal or monoclonal antibody, yielded a “A kDa percursor polypeptide. Collectively these data 1) showed that pruu was the initial gene product, 2) are consistent with the concept that pruu dimerizes to pr88, and 3) demonstrate that pr88 is a processing intermediate glycosylated to gplOO, which is then processed to gp60 and gpu9. The gene encoding prUN was identified by two approaches. A lgtll expression library was screened with the monoclonal antibody. In Southern blot analysis, DNA from positive recombinant phage hybridized to MDHV BamHI D and H fragments, located partly in the long inverted repeats, suggesting the gene is represented twice in the genome. Northern blot analysis revealed a 1.8 kb transcript, a size appropriate to encode a nu kDa polypeptide. Hybrid-selection of mRNA from these regions, and cell-free translation and immunoprecipitation analyses, revealed a nu kDa precursor polypeptide which co-migrated with pruu formed in vivo. When the MDHV-B gene was mapped within the BamHI H fragment by nuclease 81 protection, the 5' terminus was located ~550 bp from the unique SmaI site, the 3' terminus ~UHO bp from the unique Hind III site. This more precise localization will facilitate nucleotide sequence analysis of the gene's open reading frame and its 3' and 5' flanking regions. Copyright by IDAH SITHOLE 1988 To Mom and Dad ACKNOWLEDGEMENTS I would like to thank Dr. Leland F. Velicer for guidance, encouragement and financial support. I would like to thank Dr. J.B. Dodgson for his assistance, both academically and administratively. Special thanks to members of my guidance committee, Dr. P.J. Fraker, Dr. V.M. Maher and Dr. R.P. Hausinger fo their helpful suggestions. My special thanks to Dr. P.M. Coussens for the stimulating discussions, and intellectual input for the work in Chapters 3 and A, Dr. R.J. Isfort for initiating the MDHV-B antigen work in Chapter 2, Ruth A. Stringer for technical assistance, members of Dr. Velicer's lab for providing a pleasant atmosphere to work in and Moriko Ito, Chris Stewart and Paul Boyer for your friendship and help with Nuclease Sl mapping. I am grateful to USAID/ZIMMAN/IIE for the financial support, which made it possible for me to pursue this degree. Finally, my sincere gratitude to my family and Stella Konadu for everything. vi TABLE OF CONTENTS Introduction . . . . . . . . . . . . . . Chapter I: Literature Review . . . . An Overview on Herpesviruses . . . . Marek's Disease Herpesvirus System . References . . . . . . . . . . . . Chapter II: Molecular Characterization Herpesvirus B Antigen . . . . . . . . Abstract . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . Materials and Methods . . . . . . . . . Results . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . Chapter III: Synthesis and Processing of the Marek's Disease Herpesvirus 8 Antigen Glycoprotein Complex Abstract . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . Materials and Methods . . . . . . . . . Results . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . Chapter IV: Identification and Localization of Marek's Disease Herpesvirus B Antigen Abstract . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . Materials and Methods . . . . . . . . . Results . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . vii Page 13 25 3O 31 33 36 39 58 67 71 72 711 78 82 103 116 122 123 125 128 132 150 1‘55 LIST OF FIGURES Figure Page CHAPTER I 1 Organization of Herpesvirus Genomes . . . . . . . . . . . . . 5 2 Schematic representation of glyCOprotein processing In ”8‘1, 0 o o o o o o o o o o o o o o o o o o o o o o o o o o 8 3 PhYSICal map or HDHV DNA 0 o o o o o o o o o o o o o o o o o 16 CHAPTER II 1 SDS-PAGE identification and characterization of HVT-B and MDHV-B polypeptides immunoprecipitated with RoB . . . . . . . N1 2 Identification and characterization of the HVT-B and MDHV-B polypeptides immunoprecipitated with RoPM and ICS sera . . . UN 3 Blocking of RoB with MDHV-B purified by isoelectric focusing . . . . . . . . . . . . . . . . . . . . . . . . . . “7 “ [1uCngucosamine labeling of the three MDHV-B polypeptides . . . . . . . . . . . . . . . . . . . . . . . . 51 5 SDS-PAGE analysis of HVT-B and MDHV-B under reducing and nonreducing conditions . . . . . . . . . . . . . . . . . . . 55 CHAPTER III 1 Kinetics of MDHV-B processing as determined by pulse-chase analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 83 2 Effect of monensin pretreatment time on inhibition of MDHV-B processing . . . . . . . . . . . . . . . . . . . . . . . . . 86 3 The effect of monensin on MDHV-B processing as determined by pulse-chase analysis . . . . . . . . . . . . . . . . . . . . 88 u Deglycosylation of MDHV-B using endoglycosidase F . . . . . . 91 5 Deglycosylation of MDHV—B using endoglycosidase H (Endo-H) . . . . . . . . . . . . . . . . . . . . . . . . . . 93 6 Effect of protease inhibitors on appearance of putative MDHv-B precursors O O O O O O O O O O O O O O O O O O O O O O 96 7 Identification of the 353-methionine labeled MDHV-B precursor polypeptide pruu by cell-free translation and immunoprecipitation . . . . . . . . . . . . . . . . . . . . . 99 viii MDHV-B Processing Sequence Summary Physical Map of MDHV DNA Localization of the MDHV-B gene on the viral genome by Southern blot analysis Southern blot analysis to detect DNA subfragments carrying CHAPTER IV sequences encoding MDHV-B . Hybrid selection of MDHV-B-specific mRNA, cell-free translation translation and immunoprecipitation analysis Detection of MDHV-B mRNA by Northern blot analysis Localization of the MDHV-B mRNA termini by S1 nuclease protection ix 10“ 135 137 139 1U2 11111 1U7 INTRODUCTION Marek's disease (MD) of chickens is caused by the Marek's disease herpesvirus (MDHV), and is manifested by T-cell lymphomas and progressive demyelination of peripheral nerves. Symptoms of MD can be prevented by immunization with the non-pathogenic herpesvirus of turkeys (HVT). HVT represents the first successful vaccine against a naturally occurring tumor of any species. The basis of this immunity is not fully understood, although it's believed to be a two step mechanism involving both anti-viral and anti-tumor immunity. To more precisely understand the nature of this immunity, our laboratory and others have been studying antigens that exist in common between both viruses, especially the most prominent A and B antigens MDHV-A and MDHV-B. MDHV-B elicits partial protective immunity against MD. The long term objective is to further evaluate the role of MDHV-B in protective immunity against MD. The main objective of the research presented in this thesis was to molecularly characterize the MDHV-B polypeptides, identify and characterize the precursor polypeptide, elucidate the glycoprotein processing events, and to identify, characterize and localize the gene encoding the precursor polypeptide. MDHV-A, a secretory glycoprotein has been extensively studied in terms of the synthesis and processing, and its gene has been identified and sequenced. This antigen therefore served as a precedence for similar analyses with MDHV-B. The Literature Review (Chapter 1) describes certain aspects of herpesviruses such as classification and genome structure, and emphasizes the areas of glycoprotein processing and gene mapping which N are topics pertinent to the discussions in Chapters 3 and U. It also gives an overview of the Marek's disease herpesvirus system. The research data is presented in manuscript form in Chapters 2-1. Unfortunately some of the information intended for Materials and Methods, Introduction, Summary and Significance sections is therefore presented repeatedly throughout the thesis. Chapter 2 describes the identification and molecular characterization of the MDHV-B polypeptides. In Chapter 3, the primary gene product is identified and processing events are elucidated. Chapter A, goes into identification and characterization of the gene encoding the precursor polypeptide, which lays the ground work for nucleotide sequence determination. AN OVERVIEW ON HERPESVIRUSES: Classification of Herpesviruses. Herpesviruses have linear double stranded DNA that is enclosed in a capsid containing 162 capsomeres, and an envelope derived from the nuclear membrane. About 80 herpesviruses have been identified and all infect eucaryotic cells, ranging from fish, birds, amphibians, and rodents to primates, with humans having five different forms (70). There are three classes of the herpesviridae family: (i) the alphaherpesvirinae, (ii) betaherpesvirinae, and (iii) gammaherpesvirinae which have been divided according to their biological properties. The alphaherpesvirinae (herpes simplex virus 1 and 2 (RSV-1 and RSV-2), bovine mammalitis virus (BMV), etc.) are characterized by a short reproductive cycle, rapid spread in tissue culture, variable host range, excessive destruction of host cell, and ability to establish latency primarily in ganglia. The betaherpesvirinae subfamily is characterized by slow growth in tissue culture, restricted host range and the ability to enlarge cells (cytomegalia). The virus goes latent in secretory glands and the kidney. The prototype is cytomegalovirus (CMV). Gammaherpesvirinae consists of the lymphotropic viruses, e.g. Epstein Bar virus (EBV), Marek's disease virus, and Herpesvirus saimiri (HVS), which infect T or B lymphocytes (“8,53,70). Their host range is restricted to the same family or order to which the same natural host belongs, e.g. for the chicken, MDHV will infect turkey, duck and quail. The virus goes latent in lymphoid tissue. MDHV does not envelope nucleocapsids fully in tissue culture. All herpesviruses have a capsid 100 nm in diameter and is made up of 162 capsomeres. The mature virion ranges in diameter from 120 to 300 nm, the difference being due to the thickness of the tegument (the region between the capsid and the viral envelope) and pleomorphism of the envelope. This overview will describe some of the characteristics of herpesviruses and will give a more detailed coverage only on regulation of gene expression, glycoprotein processing and gene mapping. Genomes. Herpesviruses range in size from 120 kb for channel catfish virus (CCV) to 2H0 kb for CMV (70). The G + C content varies from 33% for CCV to 77% for monkey B-virus (70). All herpesviruses identified to date have structures which fall into one of the five illustrated in Fig. 1 (70). Interestingly MDHV and HVT have the type E genome common among alphaherpesviruses such as RSV-1 and RSV-2 and yet the latter belong to the gammaherpesvirus group usually typified by a type C genome based on their biological properties, i.e. lymphotropism. Gene Regulation. Herpesvirus genes are regulated in a cascade fashion, and their gene products have been classified into three general groups, (a, 8 and Y) based on the kinetics and their requirements for synthesis (70,80,86). a or immediate early genes are expressed early in the infectious cycle and do not require 33 2919 protein synthesis. In HSV these genes map at or near the long and short terminal repeat regions (70,86). These genes function to regulate gene expression at later times during infection. The B or early genes are expressed prior to DNA synthesis. These genes map throughout the genome (70,86). They are required for viral DNA synthesis and shut off a genes. Examples of 8 genes include the viral ribonucleotide reductase, the major DNA binding protein, viral DNA polymerase, and the viral thymidine kinase. Two classes of late genes have been detected. The BY class also known as the Fig. 1. Organization of Herpesvirus Genomes. The genome types A, B, C, D, and E are exemplified by the channel catfish herpesvirus (CCV), herpesvirus samiri (HVS), Epstein-Barr virus (EBV), pseudorabies virus (PRV) and herpes simplex virus (RSV), respectively. The horizontal lines represent unique sequences, and the vertical lines represent terminal sequences. Rectangles represent reiterated sequences larger than 1 kbp. The number represent molecular weight millions (MW x 106). B. Roizman and w. Batterson (70). . ______l.maww anoica— magma .. mum? o. 33 a an: a o a .. E _= ._ no 6.0 E ____ 9 ad __=___________________________ 2 2. L;_____________ __ F A w «o 39 Q— “— _ 3 age. "leaky late" genes are expressed prior to DNA synthesis at low levels and become abundantly expressed after DNA synthesis. The Y or late genes have an absolute requirement for DNA synthesis for their expression. This expression can be inhibited by DNA synthesis inhibitors such as phosphonoacetic acid which inhibits the viral DNA polymerase (U2). Y genes shut off 8 genes, and during the next round of infection induce a genes via components of the virion. RSV glycoprotein B (gB) is a typical example of a leaky late gene product whereas HSV gC is encoded by a late gene. As a result gene products are regulated at both the transcriptional and post transcriptional level. Viral mRNAs are transcribed in the nucleus using RNA polymerase II. The mRNAs are capped, methylated and polyadenylated, although nonpoly- adenylated RNAs of the same sequence have been observed. In general very little splicing has been observed in herpesvirus genes. The EBV gp350/220 (2) glycoproteins are encoded by the same gene which undergoes splicing to generate two mRNAs that share identical 5' and 3' termini. The functional significance of synthesizing both proteins is not known. Glycoprotein Synthesis and Processing. Viral proteins are synthesized on both membrane-bound and free ribosomes (80). It is estimated that herpesviruses encode anywhere up to 80 virus specific polypeptides (structural and non-structural). Glycoproteins can be found on the surfaces of infected cells and on the envelope of the mature virion. Because of their surface location, glycoproteins are presumed to be the primary target of host anti-viral, and anti-cellular immune responses following an infection. Glycoproteins have been extensively studied in terms of their synthesis and processing. Processing takes several forms Fig. 2. Schematic representation of glycoprotein processing in HSV. Spear (80). 33803 3252003023 “.9.st cocoa—3 02.023 «Boa 2.2 32358323 fiocoaza 0085.12 «02.2.95»..ng uoxETO c. 9.9983,. 2399:? oao_o>cw «Emmi . 6.00 mum 520:2 10 ranging from cleavage of a precursor polypeptide, oligomerization, glycosylation, sulfation, phosphorylation, and acylation to translocation of proteins across membranes (3,80). The extent to which a protein is processed is a reflection of the enzyme complement of the host cell. Membrane and secretory proteins are synthesized on membrane bound ribosomes (rough endoplasmic reticulum, RER). Both membrane and secretory proteins have a signal sequence on the amino terminal end of the polypeptide, which is used to anchor the nascent polypeptide on the RER. The signal sequence is cleaved co-translationally by a signal peptidase found in the lumen of the RER. Glycosylation is cotranslational and dimerization (oligomerization) may occur immediately after synthesis (Fig. 2). RSV-1 gB, C, D (80) and MDHV-A (29) have all been shown to possess a signal sequence. Co-translational glycosylation results in addition of N-linked oligosaccharides to the Asn-x-Ser/Thr consensus sequence, to yield high-mannose type glycoproteins while complex-type glycoproteins are generated by further trimming of the high mannose chain, and addition of other substituents such as fucose, sialic acid; palmitate or sulfate (36,80). Processing in the Golgi may result in O-linked glycosylation at serine and threonine residues (Fig. 2). Further processing, may involve cleavage of a precursor molecule to two smaller forms with formation of interchain disulfide bridges; as was reported for CMV (5), VZV (18), PRV (19) and BHV—i (82). Herpesviruses unlike other enveloped viruses; bud from the nuclear membrane (70,80). There are two modes of transport of glycoproteins to the plasma membrane: (i) fully enveloped virions with immature glycoprotein are glycosylated in situ and transported in cytoplasmic ll vesicles (30,31), (ii) and the remainder of the glycoproteins are transported in cellular organelles. Immature glycoproteins have been found associated with nuclear fractions. Infectious virions bearing immature glycoproteins have been detected in cells treated with: (i) monensin (31), (ii) ammonium ions (37), and cells deficient in a glycosyl transferase (76). Both these findings argue very strongly in favor of in situ glycosylation during transit. Gene Mapping. Various techniques have been used to determine the map locations of different genes in herpesvirus systems. Initially Ruyechan et al. (73) mapped HSV glycoprotein genes by using intertypic recombinants of RSV 1 and 2, which were generated using information obtained from restriction enzyme analyses. Preston and Cordingley (67) mapped the gene for alkaline exonuclease by microinjecting hybrid-selected RNA or cloned viral DNA into Xenopus oocytes and assaying for the exonuclease activity. Other genes mapped by hybrid-selection were the tk gene (68) and gpl of PRV (N9). Wigler et al. (87) on the other hand used cloned fragments in transfection assays and tested for the ability of tk‘ cells to restore enzyme activity. The gene for glycoprotein C (gC) of varicella zoster virus (VZV) was identified by cloning a plasmid library into the open reading frame (ORF) vector (12). gC specific monoclonal antibodies were used to screen colonies, and the resulting positive clones were used to hybrid-select RNA specific for gC. Cell-free translation of this RNA and immunoprecipitation revealed a polypeptide which co-migrated with the precursor polypeptide of gC. 12 Using the bacteriophage expression vector Agt11 (89), Mocarski gt gt. (50) have mapped the gene for ICP36 of CMV. They used a pool of monoclonal antibodies to screen a random lgt11 library, and confirmed the identity of the gene by hybrid-selection and cell-free translation. Petrovskis gt gt. (63) mapped the gene for gX by screening a Agt11-PRV library using antisera raised against denatured acetone-dried proteins. Fifty-eight transcripts were identified in VZV (58) by transcriptional mapping using Northern blot analysis. One approach towards the identification of herpesviral genes has been based on DNA sequence homology by computer search using the available sequence data. The genomes of EBV (1), VZV (10), have been sequenced in their entirety. Pellett gt gt. (62) detected DNA sequence homology between EBV gp110 and HSV gB-i. Hydropathy plots revealed conserved amino and carboxy terminal ends. Another approach is the immunological detection of cross-reactive antigens. The cross reactive anti-gB-1 antiserum immunoprecipitated gp110 from EBV infected cells, thus confirming the DNA sequence homology studies reported by Pellett E£.El° (62). In addition anti-gB-1 antiserum has been used to immunoprecipitate a homologous glycoprotein from equine herpesvirus type-1 (79) and BMV (79) infected cells. Using gB-1 monoclonal antibodies Edson gt gt. (11) detected gp63 from VZV infected cells. A slight variation to the immunological approach is the use of monoclonal antibody resistant (mar) mutants in order to map gp130 of BHV-1 (39). The mutants were restored to wild type virus by marker rescue using cloned DNA fragments. Once the mutation was localized, the DNA fragment 13 was hybridized to the RSV gB-1 probe. In this case the prediction was based on the similar biochemical properties of gp130 (82) and gBi (80). MAREK'S DISEASE HERPESVIRUS SYSTEM. Introduction. Marek's disease was first described by Marek in 1907 (U7). The disease is a herpesvirus-induced malignant lymphoma of the chicken. It is characterized by T-cell lymphomas and progressive nerve demyelination (59,60,61). MD is of great economic importance to the poultry industry, and is estimated to have cost $200 million per year (61) until the early 1970's when a live and effective vaccine was developed from the apathogenic herpesvirus of turkeys (56). MD represents the first neoplastic disease for which an effective vaccine has been developed, and also represents the first disease caused by a herpesvirus that responds to a vaccine. Marek's disease herpesvirus is important in tumor virology because it is one of the herpesvirus that has been shown to induce tumors in its natural host. MDHV and Epstein Barr virus belong to the gammaherpesvirus group and are both lymphotropic in nature (70). The similarities of these two systems make MD an attractive and important disease animal model system to study. 2E5“ Research on MD was greatly hampered by difficulties in obtaining pure viral DNA samples. When the virus was first identified it was found to be highly cell-associated in tissue culture (5U,61). Most of the particles formed were naked nucleocapsids, and very little cell-free virus could be found in culture supernatants. The feather follicle epithelium is the only site where fully enveloped infectious cell-free virus is produced (5U,61). Attempts to isolate viral DNA free of cellular 14 DNA by CsCl equilibrium density gradient centrifugation failed because of the close similarities in the buoyant densities (1.705 and 1.700 g/cm3 respectively). The viral DNA used in the cloning experiments and in the determination of physical properties (described below) was isolated from naked nucleocapsids found in culture supernatants (16,34). The virus particles were purified by centrifugation in sucrose gradients, lysed with detergents, treated with pronase and further purified by centrifugation to equilibrium in CsCl. MDHV DNA is a linear double stranded molecule of 1.2 x 108 daltons molecular weight (40) which is slightly larger than that of the prototype herpes simplex virus DNA. It has a sedimentation coefficient of 568 (A0) when compared to T“ bacteriophage and RSV DNA. In addition to the 565, there is another species of 708 sedimentation coefficient corresponding to 6 x 107 daltons, which suggests the presence of gaps or nicks in the DNA (U0). RVT DNA also has a sedimentation coefficient of 56S. The density of MDHV and RVT DNA in 0301 is 1.705 and 1.707 g/cm3 respectively (u0). MDHV DNA has 3 guanine plus cytosine content of “6%. Electron microscopic studies (7) revealed the presence of inverted repeats which lead to the conclusion that the structure of both MDRV and RVT is similar to that of alphaherpesviruses. The genome consists of two covalently linked unique components, L and S, and both are bordered by inverted repeats (7). While the RSV DNA can exist in four isomeric forms both MDRV and RVT DNA's have no isomers (7,1U). Both MDRV and RVT genomes have been cloned into cosmid and plasmid vectors (1N,16,21,72). First, EcoRi fragments of the MDHV-GA strain were cloned into pBR328 (16). Twenty five clones were identified as containing viral DNA inserts, however these clones accounted for only N5% of the n: 15 whole genome. Soon after, 27 of the 29 BamRI-digested MDHV-GA restriction fragments were cloned into the cosmid pRC79, and the plasmids pACYC18A and pBR322 (1N). Detailed restriction enzyme maps for BamR1, SmaI and BglI (Fig. 3) have been published (1“). Nineteen RVT BamR1-digested fragments have been cloned into both cosmid and plasmid vectors (21). EcoR1 fragments of the RPRS-1 strain have been cloned into pBR322 (72). Twelve positive clones were identified and shown to cover 50% of the genome. The related herpesviruses found in the MD system have been classified into three serotypes, type 1 virulent MDRV; type 2, naturally occuring avirulent; and type 3 viruses such as RVT (6). All three serotypes are extensively related immunologically, sharing a large number of polypeptides (25,77,83,8u) that are apparently common antigens including the most prominent common antigens A and 8. Despite this strong antigenic relationship the most reliable of early nucleic acid studies showed 1-uz homology (N3). Using less stringent hybridization conditions and methods that greatly improved the DNA reassociation kinetics Gibbs §£.El° (16) showed MDHV and RVT to be closely related, with 70-80% homology at the nucleotide level, and this extended to 90-95% of the respective genomes. This finding could readily eXplain the similar protein profiles and close antigenic relationship observed between MDHV and RVT. There is no data available for serotype 2, using these new hybridization conditions. Vaccinal Immunity. Since the early 1970's MD has been successfully controlled by immunization with live virus. Three types of vaccines are available: (1) the attenuated MDHV, (ii) the naturally apathogenic MDHV, and (iii) naturally apathogenic RVT. The pathogenic strain RPRS-16, was attenuated by serial passage in tissue culture, and used as the first Fig. 3. Physical map of MDHV DNA, Bam R1, Bgl I and Sma I restriction endonuclease maps of MDHV DNA. DR, Direct repeat. The location on the map is expressed in (MW x 106) from the left end of the genome. Fukuchi gt gt. (1“). l6 ..- 31131117 . 0 n n 0 a 11.253 :3... .c 4 ... .u. .....fiz a. .12. .3. u .33... o o... a u.. .. flown! to. cm. m: cm. 0.» c... 2. co 3 9. on ow 0.. .im 00.x 3.3 . l7 18 commercial vaccine (61). The CV1 988 is a naturally apathogenic virus which is widely used in the Netherlands (61). RVT is apathogenic in both chickens and turkeys and is now the most extensively used vaccine worldwide (56). The RVT vaccine is not 100% efficient and occasional failures have been reported (61). Failures result mainly from (i) administration of incorrect dose, (ii) challenge by pathogenic virus before normal vaccinal immunity has developed, (iii) delayed onset of vaccinal immunity due to the presence of maternal antibody, and (iv) overattenuation of MDHV and RVT. Three targets have been proposed to be involved in vaccinal immunity; (1) the MDRV virion, (ii) the infected cell, and (iii) the tumor cell. Studies to determine the mechanism of protective immunity involved immunization with subcellular or soluble fractions of cells productively infected with MDHV and RVT, including highly purified cell plasma membrane (32,33), detergent soluble and insoluble antigen (UN), and partially purified glycoproteins (88). All of these preparations, elicited protective immunity at a significant level leading to the conclusion that anti-viral immunity plays an important role in the prevention of MD. However these early studies did not include molecular characterization of the antigens involved. Immunization with lymphoid cells containing the Marek's disease associated tumor surface antigen (MATSA) (6A,65,71) but not the other antigens, also elicited protective immunity. The hypothesis is that protection is achieved by two forms of immunity, one against virus- specific antigens seen in productively infected cells and one against tumor-specific MATSA, or an unknown antigen on the surface of tumor cells. In 19 Subsequent work focused on characterizing the two mechanisms of protective immunity (51,66), studying the effect of immunization with inactivated virus and MATSA on early pathogenic events. Both immunogens achieved significant protection against subsequent tumor development although viral antigen appeared superior to tumor antigen. Recently Ono gt gt. (57) reported that antigen gB, when purified by affinity chromatography from RVT infected cells, achieves partial protection against MD. Proteins Antigens. Viral-specific antigens have been found in the feather follicle epithelium, in cells of the spleen, bursa and thymus and in permissive cell cultures infected with the virus and occasionally in lymphoblastoid cell-lines (5“). Different techniques have been used in the study of these antigens including immunodiffusion (8), and immunofluorescence (IF) (23), immunoprecipitation and SDS-PAGE (17,83,84) and ELISA (enzyme-linked immunosorbent assay) (“1). Three MDV-associated antigens A, B and C have been detected by immunodiffusion of sonicated cell extracts and culture fluid (8,N5,85). Using immunofluorescence, virus-specific antigens were detected on the surface of infected cells in the nucleus and cytoplasm (23), and feather follicle epithelium (61). Immunoprecipitation and SDS-PAGE analyses in one and two-dimensional electrophoresis led to the detection of 35 virus-specific polypeptides, 11 of which were common between MDHV and RVT (83,8U). More recently Ikuta gt gt. have reported more than no proteins specific to and cross reactive between MDHV and HVT ranging in molecular weight from 19,000 (19K) to 350,000 (350K) daltons (2U). (I) E] ("I ‘l /'\ 3 20 Monoclonal antibodies with MDHV and RVT have been isolated and used for the identification of type common and type specific epitopes of MDRV and RVT specific proteins (26,77). MDHV-A. MDHV-A was detected by immunodiffusion analysis of cell extracts and culture fluids (A6). 8 and C antigens were present only in the cell extracts (8,85). Both A and B were shown to be common between MDHV and RVT (56), and were identified as glycoproteins (26,28,u5,77). MDHV-A could be purified free of B antigen (MDHV-B) by isoelectric focusing since their isoelectric points differ by two pR units making it possible to prepare rabbit anti-MDRV-A (RoA) possesing no reactivity with MDHV-B (U5,N6,85). MDHV-A is a secretory glycoprotein of apparent molecular weight 57-65,000 daltons (29). The primary gene product is an unglycosylated precursor polypeptide of “H.000 daltons (pruu), which is seen after tunicamycin treatment of infected cells (29) and cell-free translation (29) (labelled with 35S-methionine, and immunoprecipitated with RaA). RVT-A is similar in molecular weight to MDHV-A (29). The staphylococcus V8 protease and tryptic peptide data (Naidu; Stringer and Velicer, unpublished data) showed identical profiles. The gene encoding MDRV-A has been identified (27), and sequenced (Coussens and Velicer, manuscript submitted), and similar work is almost complete for RVT-A (Coussens and Velicer, manuscript in preparation). Comparison of the MDHV-A and RVT-A sequence data reveals third base pair changes in the genes, which can readily explain why the restriction enzyme patterns are different, whereas the antigenic similarities are retained (unpublished data, Coussens and Velicer). Great importance was given to A antigen because it had been shown earlier that attenuation of 21 the virus by serial passaging in culture resulted in loss of this antigen (9). Later it was shown that naturally apathogenic strains, such as RVT, produced A antigen and some pathogenic strains lacked A antigen without attenuation (69). Also, viruses failing to produce A antigen failed to produce membrane antigen (MA), suggesting that the two antigens were related or identical (”2). It seems however that attenuation of virus in culture and loss of A antigen were totally unrelated and occured purely by coincidence. More recently Velicer gt gt. (unpublished data) have shown for the attenuated JM strain, that the A antigen gene is present but not expressed, strongly suggesting a mutation in the promoter region, or other regulatory gene. MDHV-B. MDHV-B could be purified free of MDHV-A by isoelectric focusing since their isoelectric points differ by two pH units making it possible to prepare rabbit anti MDHV-B (RaB) possesing no reactivity with MDHV-A (85). While its physical properties were compared to MDHV-A, no information on molecular characterization was obtained (85). Silva and Lee (77) identified glycoproteins of 100,000, 60,000 and ”9,000 daltons in common between MDHV and RVT, using monoclonal antibodies shown to have virus neutralizing activity. Ikuta gt gt. (26) also identified what they call gB antigens, as a similar size set of three polypeptides (MDHV 115/110,000, 63,000, 50,000 and RVT 110,000, 62,000 and 52,000 daltons), with a monoclonal antibody also having virus neutralizing activity. However these three glycoproteins were assumed to be the MDHV-B glycoprotein complex. MDHV-B was then shown to consist of three glycoprotein, designated gp100, gp60, and ng9 (28), based on immunoprecipitation with three antisera previously shown to have MDHV-B Ti 01‘ It 22 activity in immunodiffusion. Tunicamycin inhibition of glycosylation resulted in identification of two putative MDHV-B precursor polypeptides pr88 and pruu (28). Silva and Lee (77) detected only pruu using a 6 hr labelling period; whereas Ikuta gt gt. (25) detected a pr83 and pr88 for MDHV and a pr90 for RVT using a 10 minute pulse. It is quite possible that the pr83 they detect is analogous to the virus-specific p79 that we detect in our cell lysates as a result of non-specific trapping. Immunization of chickens with gB partially purified by affinity chromatography, resulted in production of neutralizing antibodies (57) and partial protection against MD (57). gB from serotypes 1, 2, and 3 was shown to be present on the surface of infected cells, using their monoclonal antibodies (23). This strongly suggests that B antigen is a component of the virion envelope, because viral neutralization is usually associated with envelope glycoproteins. Enzymes. Two enzyme activities have been reported in MDHV-infected cell cultures (A). First, the MDHV-specific DNA polymerase has been detected in the nucleus 2“ hours post infection, and in the cytoplasm 72 hrs post infection (A). This enzyme can be distinguished from the cellular DNA polymerase on the basis of sedimentation coefficient, catalytic and inhibitory properties, and chromatographic profile on phosphocellulose columns. The RVT DNA polymerase has been isolated and has similar properties to the MDHV-specific enzyme. Secondly, an MDRV (35) and RVT (35) specific thymidine kinase (TK) was also reported. These two enzyme differ from the RSV TK and the cellular TX on the basis of electrophoretic mobility, isoelectric point and substrate specificity. It has been reported however, that the RVT TK has similar properties to the host mitochondrial TK except for TK sensitivity to inhibition by dCTP and sedimentation coefficient (35). Other Proteins. Early and late membrane antigen (MA) have been identified on the basis of IF (AZ). Four tumor-associated surface antigens have been detected: Marek's disease tumor-associated antigen (MATSA) (6A), the Ia-like antigen (75), chicken fetal antigen (52), and Forssman antigen (22). MATSA is heterogeneous amongst the lympho- blastoid cell lines. This is a glycoprotein molecule which is thought to be a cellular component induced during transformation, as it is not found in productively infected cells. Ikuta gt gt. (2“) have identified at least 10 MDHV-specific phospho- rylated polypeptides from lysates of infected cells using polyclonal antisera specific for serotype 1. Using monoclonal antibodies, they immunoprecipitated four phosphorylated polypeptides of M3K; 39K and 2UK from MDHV-BC-i strain; and “UK from its non-oncogenic counterpart. In an IF test these monoclonal antibodies reacted with a cytoplasmic antigen (2“). Two virus-specific nuclear antigens (NA) were detected by an IF test using monoclonal antibodies. One NA is related to the 135K DNA binding protein, whereas, the other NA is related to the 1N5K MDHV, and 155K RVT specific proteins. Both NAs are common amongst the three serotypes. State of Viral Genome in Transformed Cells and Tumors. MD tumors in chickens are devoid of cells producing virus-specific antigens or virus particles. However DNA cRNA hybridization experiments (42) showed that these tumors contained viral genomes. When a number of tumors from difl numi cont tun: 24 different chickens were examined they were found to contain a variable number of genomes. Interestingly, different tumors from the same bird contained the same number of viral genomes per cell, indicating that the tumors were monoclonal in origin. It appears that the viral genome in MD tumors exist in two states, either as an episome or integrated in the host chromosome (20,3N,81). Tanaka gt gt. (81) found a covalently closed circular genome in the non-producer cell line MKT-1. Kashka-Dierich gt gt. (3“) looked at both the producer cell-line MSB-i and the non-producer cell-line (MDCC-HP1) and found both episomal DNA and DNA tightly associated with cellular DNA and could not be dissociated by mechanical shearing. Rziha and Bauer (7“) locked at two producers and one non-producer cell line and demonstrated the presence of free and integrated viral DNA. This result is similar to that observed with EBV in Raji cells (55). Hughes gt gt. (20) using Southern blot analysis of DNA from different chromosomal preparations, showed that the viral DNA was associated with at least two minichromosomes. There are no RVT transformed cell-lines available, although a lymphoblastoid cell-line co-transformed with RVT and MDHV has been established (Hirai, unpublished data). Phosphonoacetic acid, a DNA synthesis inhibitor which blocks the herpesvirus specific DNA polymerase had no effect on the replication of resident viral DNA genomes in transformed cells (“2). Using a double thymidine block to synchronize cells, Lau and Nonoyama (38) showed that the viral DNA in transformed cells replicates only once during the cell cycle. The results of the above mentioned experiments indicate that the viral DNA in transformed cells is regulated by cellular functions. These findings argue in favor of viral DNA integrating into host chromosomes. 10. 11. 12. 13. 1A. 15. 16. 17. LIST OF REFERENCES Baer, R., A.T. Bankier, M.D. Biggin, P.L. Deininger, P.J. Farrell, T.G. Gibson, G. Ratfull, G.S. Hudson, S.C. Satchwell, C. Sequin, P.S. Tuffnell, and 8.0. Barrell. (198“) Nature (London) 310:207-211 Beisel, C., Tanner, J., Matsuo, T., Thorley-Lawson, D., Kezdy, F., and Kieff, E. (1985) J. Virol. 5M:665-67U. Blobel, G., Walter, P., Chang, C.N., Goldman, B.M., Erikson, A.R., and Lingappa, V.R. (1979) Symp. Soc. Exp. Biol. 33:9-36. Boezi, J.A., Lee, L.F., Blakesley, R.W., Koenig, M., and Towle, R.C. (197A) J. Virol. A 1029-1219. Britt, W.J. and Auger, D. (1986) J. Virol. 58:185-191. Bulow, V. von, Biggs, P.M., and Frazier, J.A. (1975) In: Oncogenesis and Rerpesviruses. (G. de-The, M.A. Epstein, and H. zur Rausen, eds.), International Agency for Research in Cancer, Lyon. No. 11, pp. 329-336. Cebrian, J., Kaschka-Dierich, C., Bertholet, N., and Sheldrick, P. (1982) Proc. Natl. Acad. Sci. USA 79:555-558. Chubb, R.C. and Churchill, V.F. (1968) Vet. Rev. 83:1-7. Davison, A.J. and Scott, J.E. (1986) J. Virol. 67:1759-1816. Churchill, A E., Chubb, R.C., and Baxendale, w. (1969) J. Gen. Virol. u:557-56u. Edson, C.M., Hosler, B.A., Respess, R.A., Water, D.J., and Thorley-Lawson, D.A. (1985) J. Virol. 56:333-336. Ellis, R.W., Keller, P.M., Lowe, R.S., and Zivin, R.A. (1985) J. Virol. 53:81-88. Emini, B.A., Luka, J., Armstrong, M.E., Keller, P.M., Ellis, R.W., and Pearson, G.R. (1987) Virol. 157:552-555. Fukuchi, K., Sudo, M., Lee, Y.S., Tanaka, A., and Nonoyama, M. (198“) J. Virol. 51:102-109. Fukuchi, K., Tanaka, A., Schierman, L.W., Witter, R.L., and Nonyama, M. (1985b) Proc. Natl. Acad. Sci. USA 82:751-754. Gibbs, C., Nazerian, K., Velicer, L.F., and Kung, R.J. (198A) Proc. Natl. Acad. Sci. USA 81:3365-3369. Glaubiger, C., Nazerian, K., and Velicer, L.F. (1983) J. Virol. “5:1228-1239. 25 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 3A. 35. 36. 26 Grose, C., Edwards, D.P., Weigle, K.A., Friedrichs, W.E., and McGuire, W.L. (198A) Virology 132:138-1U6. Hampl, R., Ben-Porat, T., Ehrlicher, L., Habermehl, K.-O., and Kaplan, A.S. (198M) J. Virol. 52:583-590. Hughes, S.H., Stubblefield, E., Nazerian, K., and Varmus, H.E. (1980) Virology 105 23u-2u0. Igarashi, T., Takahashi, M., Donovan, J., Jessip, J., Smith, M., Hirai, K., Tanaka, A., and Nonoyama, M. (1987) Virol. 157:351-358. Ikuta, K., Kitamoto, N., Shoji, R., Kato, S., and Naiki, M. (1981) J. Gen. Virol. 52:1“5-151. Ikuta, K., Nakajima, K., Naito, M., Ann, S.H., Ueda, S., Kato, S., and Hirai, K. (1985b) J. Cancer 35:257-26“. Ikuta, K., Nishi, Y., Kato, S., and Hirai, K. (1981) Virology 11A:277-281. Ikuta, K., Ueda, 8., Kato, S., and Hirai, K. (1984) Microbiol. Immunol. 28:923-933. Ikuta, K., Ueda, S., Kato, S., and Hirai, K. (1984) J. Virol. “9:1019-1017. Isfort, R.J., Kung, H.-J., and Velicer, L.F. (1987) J. Virol. Isfort, R.J., I. Sithole, H.J. Kung, and L.F. Velicer (1986) J. Virol. 59:911-H19. Isfort, R.J., R.A. Stringer, R.-J. Kung, and L.F. Velicer (1986) J. Virol. 57 uou-u7u. Johnson, D.C. and Spear, P.G. (1983) Cell 32:987-997. Johnson, D.C. and Spear, P.G. (1982) J. Virol. “3:1102-1112. Kaaden, O.R., Dietzschold, B., and Ueberschar, S. (197“) Med. Microbiol. Immunol. 159:261-269. Kaaden, G.R. and Dietzschold, B. (197A) J. Gen. Virol. 25:1-10. Kaschka-Dierich, C., Nazerian, K., and Thomssen, R. (1979) J. Gen. Virol. NH 271-280. Kit, 8., Rorgensen, C.N., Leung, W.C., Trkula, D., and Dubbs, D.R. (1973) Intervirology 2:299-311. Kornfeld, R. and Kornfeld, S. (1985) Ann. Rev. Biochem. 59:631-66“. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 1"8. 49. 50. 51. 52. 53. 54. 55. 27 Kousoulas, K.G., Bzik, D.J., DeLuca, N., and Person, S. (1983) Virology 125:468-474. Lau, R.Y., and Nonoyama, M. (1980) J. Virol. 33:912-914. Lawrence, W.C., D'urso, R.C., Kundel, C.A., Whitbeck, J.C., and Bello, L.J. (1986) J. Virol. 60:405-414. Lee, L.F., Kieff, E.D., Bachenhemier, S.L., Roizman, B., Spear, P.G., Burmester, B.R., and Nazerian, K. (1971) J. Virol. 7:289-294. Lee, L.F., Liu, X., and Witter, R.L. (1983) J. Immunol. 130:1003-1006. Lee, L.F., Nazerian, K., Leinbach, S.S., Reno, J.M., and Boezi, J.A. (1976) J. Natl. Cancer Inst. 56:823-827. Lee, Y., Tanaka, A., Silver, 8., Smith, M., and Nonoyama, M. (1979) Virology 93:277-280. Lesnik, F. and Ross, L.J.N. (1975) Int. J. Cancer 16:153-163. Long, P.A., Clark, J.L., and Velicer, L.F. (1975) J. Virol. 15:1192-1201. Long, P.A., Kaveh-Yamini, P., and Velicer, L.F. (1975) J. Virol. 15:1182-1191. Marek, J. (1907) Dtsch. Tieraerztl. Wochenschr. 15:1192-1201. Melendez, L.V., Daniel, M.D., Hunt, R.D., and Garcia, F.G. (1968) Lab. Anim. Care 18:374-381. Mettenleiter, T., Luckacs, N., and Rziha, R.J. (1985) J. Virol. 56:307-311. Mocarski, E., Pereira, L., and Michael, N. (1985) Proc. Nat. Acad. Sci. USA. 82:1266-1270. Murthy, K.K., Dietert, R.R., and Calnek, B.W. (1979) Int. J. Cancer 24:349-354. Murthy, K.K. and Calnek, B.W. (1978) J. Natl. Cancer Inst. 61:849-854. Nazerian, K. (1970) J. Natl. Cancer Inst. 44:1257-1267. Nazerian, K. and Lee, L.F. (1974) J. Gen. Virol. 25:317-321. Nonoyama, M., and Pagano, J.S. (1972) Nature (London) New Biol. 238:169-171. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 28 Okazaki, W., Purchase, R.C., and Burmester, B.R. (1970) Avian Dis. 14:413-429. Ono, K., Takashima, M., Ishikawa, T., Hayashi, M., Yoshida, I., Konobe, T., Ikuta, K., Nakajima, K., Ueda, S., Kato, S., and Hirai, K. (1985) Avian Dis. 29:533-539. Ostrove, J.M., Reinhold, W., Fan, C.-M., Zorn, 8., Ray, J., and Straus, S.E. (1985) J. Virol. 56:600-606. Pappenheimer, A.M., Dunn, L.C., and Cane, V. (1929) J. Exp. Med. 49:63-86. Pappenheimer, A.M., Dunn, L.C., and Seidlin, S.M. (1929) J. Exp. Med. 49:87-102. Payne, L.M. (1982) In: The Rerpesviruses. (B. Roizman, ed.), Vol. 1, pp. 347-431 Plenum, New York. Pellett, P.E., Biggin, M.D., Barrell, B., and Roizman, B. (1985) J. Virol. 56:807-813. Petrovskis, B.A., J.G. Timmins, and Post, L.E. (1986) J. Virol. 60:185-193. Powell, P.C. (1985) p. 238-261. tg B.W. Calnek and J.L. Spencer (ed.), Proceedings of the International Symposium on Marek's Disease. American Association of Avian Pathologists, Inc., Kennett Square, PA. Powell, P.C. (1975) Nature 257:684-685. Powell, P.C. and Rowell, J.G. (1977) J. Natl. Cancer Inst. 59:919-924. Preston, C.M. and Cordingley, M.G. (1982) J. Virol. 43:386-394. Preston, C.M. and McGeoch, D.J. (1981) J. Virol. 38:593-605. Purchase, R.C., Burmester, B.R., and Cunningham, G.R. (1971) Infect. Immun. 3:295-301. Roizman, B. and Batterson, W. (1985) In: Rerpesviruses and Their Replication. (B.N. Fields et al., ed.), Raven Press, New York, 25:497-526. Ross, L.J.N. (1977) Nature 268:644-646. Ross, L.J.N., Milne, B., and Biggs, P.M. (1983) J. Gen. Virol. 64:2785-2790. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 830 84. 85. 86. 87. 88. 89. 90. 29 Ruyechan, W.T., Morse, L., Knipe, D., and Roizman, B. (1979) J. Virol. 29:677-683. Rziha, R.-J., and Bauer, B. (1982) Arch. Virol. 72:211-216. Schat, K.A., Chen, G.R., Shek, W.R., and Calnek, B.W. (1982) J. Natl. Cancer Inst. 69:715-720. Serafini-Cessi, F., Dall'Olio, F., Scannavini, M., and Campadelli-Fiume, G. (1983b) Virology 131:59-70. Silva, R.F. and Lee, L.F. (1985) In: Proceedings of the International Symposium of Marek's Disease. (B.W. Calnek and J.L. Spencer, eds). Pp. 101-110. American Association of Avian Pathologists, Inc., Kennett Square, Pennsylvania. Silva, R.F. and Lee, L.F. (1984) Virology 136:307-320. Snowden, B.W., Kinchington, P.R., Powell, K.L., and Ralliburton, I.W. (1985) J. Gen. Virol. 66:231-247. Spear, P.G. (1985) In: The Herpes Viruses, Vol. 3, (B. Roizman, ed.), pp. 315-356. Plenum Press, New York. Tanaka, A., Silver, 8., and Nonoyama, M. (1978) Jpn. J. Vet. Res. 26:57-67. Van Drunen Littel-Van Den Hurk, S., Van Den Hurk, J.V., Gilchrist, J.E., Misra, V., and Babiuk, L.A. (1984) Virology 135:466-479. Van Zaane, D., Brinkhof, J.M.A., and Gielkens, A.L.J. (1982b) Virology 121:133-146. Van Zaane, D., Brinkhof, J.M.A., Westenbrink, F., and Gielkens, A.L.J. (1982a) Virology 121:116-132. Velicer, L.F., Yager, D.R., and Clark, J.L. (1978) J. Virol. 27:205-217. Wagner, E.K. (1985) In: The Rerpesviruses, Vol. 3. (Roizman, 8., Ed.) Plenum Press, New York, pp. 45-103. Wigler, M., Silverstein, S., Lee, L., Pellicer, A., and Cheng, Y. (1977) Cell 11 233-232. Wyn-Jones, A.P. and Kaaden, O.R. (1979) Infect. Immun. 25:54-59. Young, R. and Davis, R. (1983) Proc. Natl. Acad. Sci. USA 80:1194-1198. Zezulak, K.M. and Spear P.G. (1984a) J. Virol. 50:258-262. CHAPTER II Molecular Characterization of Marek's Disease Herpesvirus B Antigen Isfort, R.J., I. Sithole, H.-J. Kung, and L.F. Velicer. J. Virol. 59:411-419. 30 ABSTRACT The Marek's disease herpesvirus (MDHV) B antigen (MDHV-B) was identified and molecularly characterized as a set of three glycoproteins of 100,000, 60,000, and 49,000 apparent molecular weight (gp100, gp60, and gp49, respectively) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) after immunoprecipitation from [3SSJmethionine- labeled infected cells by specific rabbit antiserum directed against MDHV-B (RaB), as previously determined by immunodiffusion. Further identification was accomplished by blocking this immunoprecipitation with highly purified MDHV-B. The same set of three polypeptides was also immunoprecipitated from [3SSJmethionine- and 1nC-labeled infected cells with two other sera shown to have anti-B activity, i.e., rabbit anti-MDHV- infected-cell plasma membrane (RaPM) and immune chicken serum from birds naturally infected with MDHV. The three herpesvirus of turkeys (RVT) B-antigen (RVT-B) glycoproteins immunoprecipitated with all three sera containing anti-B activity were also shown to be identical in size to those of MDHV-B by immunoprecipitation and SDS-PAGE. These data serve to clarify the molecular identification of the polypeptides found in common between MDHV and HVT by linking them to MDHV-B, previously identified only by immunodiffusion, and to a similarly sized set of immunologically related common glycoproteins called gp100, gp60, and gp49, detected with monoclonal antibody by other workers. Tunicamycin inhibition of N-linked glycosylation resulted in either nonglycosylated or O-linked glycosylated putative precursors of MDHV-B and RVT-B with apparent molecular weights of 88,000, called pr88, and 44,000, tentatively called pr44, both immuno- precipitable with all three sera. However, the relationships of these two 31 32 polypeptides to each other and to the overall precursor-processing relationship of the MDHV-B complex remains to be elucidated. The three fully glycosylated B-antigen polypeptides were not connected by disulfide linkage. Collectively, the data presented here and by others support the conclusion that all three glycoproteins now identified as gp100, gp60, and gp49 have MDHV-B determinants. Finally, detection of the same three poly- peptides with well-absorbed RaPM, which was directed against purified infected-cell plasma membranes, suggests that at least one component of the B-antigen complex has a plasma membrane location in the infected cell. These preliminary data point to the future membrane biochemistry and membrane immunology experiments needed to understand the MDHV system, and they may explain the high level of immunogenicity of MDHV-B in the infected chicken, as shown by its immunoprecipitation with immune chicken serum. INTRODUCTION Marek's disease (MD) is a lymphoproliferative disease of chickens, caused by MD herpesvirus (MDHV), which results in T-cell lymphomas and peripheral nerve demyelination (14,18,19). This disease was a major cause of economic loss to the poultry industry (21) until the early 1970s, when a live vaccine was developed from the antigenically related yet apathogenic herpesvirus of turkeys (RVT) (16). Because the protection caused by RVT appears to have an immunologic basis (15,20,23,24), there has been considerable interest in detecting antigens held in common between MDHV and RVT. In the MDHV system, six antigenic proteins are detected by immunodiffusion and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) analysis with convalescent-chicken serum (23). These six proteins have been found to have an immunologic relationship to six equally or similarly sized proteins in cells infected with RVT (23). These antigenically related proteins are referred to collectively as the MDHV-RVT common antigens. It is believed that one or more of these common antigenic proteins are responsible for the vaccinal immunity RVT confers on chickens in protecting them from MD (23). Of the six MDHV-RVT common antigenic proteins, only one, the common A antigen, as it was originally named based on immunodiffusion analysis (2,3), has been extensively characterized. It is a 44,000-dalton (Da) glycoprotein at the physical and chemical levels (11,12), and a 57,000- to 65,000-Da glycoprotein (gp57-65) at the molecular level as determined by immunoprecipitation and SDS-PAGE analysis (4,7). Recently, the gene encoding this antigen was identified (R.J. Isfort, R.J. Kung, and L.F. 33 34 Velicer, manuscript in preparation). 0f the other common antigenic proteins, the common B antigen, also originally named based on immuno- diffusion analysis (2,3), has been characterized only physically and chemically (25). Specifically, the MDHV B antigen (MDHV-B) has been identified as a glycoprotein that is found primarily in the cell lysate; resistant to a pH of 2.0, eluted from concanavalin A by a-methylmannoside, resistant to 1 M urea and 0.05% Brij; and has a pI of 4.54, a sedimentation coefficient of 4.48, and an apparent molecular size of 58,000 or 69,000 Da, as determined either by gel filtration on Sephadex G-200 or from the sedimentation coefficient, respectively (25). In the characterization process, MDHV-B has been shown to be separate and distinct from MDHV A antigen (MDHV-A), and the two glycoproteins can be separated from each other because of a large difference in their iso- electric points (25). As a result of its ability to extensively purify the antigen, a highly specific antiserum has been prepared against MDHV-B by the injection of rabbits with isoelectrically focused B antigen (25). In fact, because of the sensitivity of the immune response, preparation of R08, which is monospecific upon immunodiffusion analysis, is used as a rigorous criterion for purification (25). However, a distinct and useful molecular identification of the B antigen with this antiserum against MDHV-B has not yet been performed. This paper reports the molecular identification and characterization of the B antigen by SDS-PAGE as a set of three glycosylated polypeptides that apparently have the same immunological activity, provides evidence for their immunoprecipitable putative precursor polypeptides lacking N-linked glycosylation, and gives preliminary evidence for a possible cell plasma membrane location of this MDHV-RVT common antigenic protein. As 35 this study was in progress, Ikuta et al. (5) and Silva and Lee (23,24) reported finding a very similarly sized family of immunologically related polypeptides with their monoclonal antibodies. In fact, Silva and Lee (23,24) include what appear to be the same three polypeptides in their group of five common proteins found in cell lysates. Since the set of three polypeptides reported here appears to be the same as that of Silva and Lee, the size estimates already published by them (23) were adapted for consistency of nomenclature. This study, then, serves to establish the identity of all these polypeptides as an MDHV-B complex, providing identification in terms of antisera previously prepared with purified MDHV-B as determined by immunodiffusion analysis (25). The work summarized above also helps to identify as MDHV-B the proteins previously detected by monoclonal antibody (5,23,24). MATERIALS AND METHODS Cells and viruses. Duck embryo fibroblast (DEF) cells were seeded in 100-mm-diameter plastic dishes (107 cells per dish) with 10 ml of standard growth medium as described previously (4,7). MDHV-infected DEF cells (GA strain, passage 26) frozen in liquid nitrogen or RVT-infected DEF cells (F0126 strain, passage 17) were used to infect 75 to 85% confluent DEF monolayers at a dilution of 1:8 so that maximum cytopathic effects were noticeable at 48 to 72 h. The medium for infected-cell monolayers was changed to 10 ml of growth medium without calf serum at 24 h after infection. Radioactive labeling by standard methods. Infected cells and uninfected control cells were initially labeled at 48 h after infection by washing the monolayer with Hanks buffered saline solution (GIBCO Labora- tories) three times and then incubating the cells for 24 h in 5 ml of Dulbecco minimal essential medium (GIBCO) containing 1/20 of the normal concentration of unlabeled methionine and 50 uCi of [3SSJmethionine per ml (1,000 Ci/mmol; Amersham Corp.). More recently a variation--4 h of labeling in the presence of labeled methionine only--was found to be superior for labeling the polypeptides emphasized in this study. After being labeled, the cells were subjected to three washes with sterile phosphate-buffered saline, followed by the addition of 5 ml of ice-cold detergent buffer (27), incubation for 5 min at 4°C, and removal by scraping with a rubber policeman. The lysates were then clarified by centrifugation at 3,000 rpm (International Centrifuge model RP6) for 10 min and recentrifugation at 30,000 rpm for 1 h in a type 30 or SW50.1 rotor (Beckman Instruments, Inc.). The clarified lysates were stored at 36 37 -20°C. Labeling in the presence of tunicamycin (TM) was essentially the same as described above, except that TM (Sigma Chemical Co.) was added to the labeling medium at a concentration of 2 ug/ml. Labeling with [1uCngucosamine was performed as previously reported (4,7). Immunoprecipitation with RaB and RoPM. Rabbit antiserum against MDHV-B, i.e., rabbit anti-B (RoB), was prepared as described previously (25). Rabbit antiserum against infected-cell plasma membranes, i.e., rabbit anti-plasma membrane (RaPM), was prepared with plasma membranes purified from MDHV-infected cells as described by Kaaden and Deitzschold (8). Immunodiffusion analysis of this serum at the time it was prepared revealed that the major immunologic reactivity was against MDHV-B (Velicer, unpublished data). Immunoprecipitations were performed by the methods of Witte and Wirth (27), as previously adapted to this system (4,7). Briefly, labeled cell lysate was first precleared with a volume of normal rabbit serum (NRS) equal to the volume of immune serum to be used (see below). After incubation overnight on ice, a 10% Staphylococcus aureus suspension (9) was added at 10 ul/ul of serum, and this mixture was incubated on ice for 1 h. After incubation, the material was spun at top speed in an Eppendorf centrifuge (model 5414) for 3 min. The supernatant was then removed, and to it was added the selected volume of either RaB or RoPM. In these studies, usually 5 ul of RaB or RaPM (and NRS in the preclearance step described above) was reacted with 200 or 400 pl of cell lysate radiolabeled for 24 or 4 h, respectively. After incubation on ice overnight, the 10% §. aureus suspension was added as described above, and the samples were incubated for 1 h on ice. As described above, the antigen-antibody-t. aureus complexes were harvested by centrifugation, washed twice in detergent buffer, and collected by centrifugation at top 38 speed in an Eppendorf centrifuge for 3 min. The precipitate was then suspended in sample buffer for electrophoresis. Blocking experiments. To block immunoprecipitation of MDHV-B by the RoB serum, the method previously employed for MDHV-A was used (4). Breifly, various amounts of unlabeled, isoelectrophoretically purified B antigen (25) were incubated with the R08 serum overnight on ice prior to the addition of the R08 sera to precleared cell lysates. SDS-PAGE. Discontinuous SDS-PAGE was performed by the method of Laemmli (10) as modified by the Roeffer Corp. Briefly, electrophoresis was performed as previously reported from this laboratory by Glaubiger et al. (4), except that a constant current of 30 mA and a sample buffer containing twice the concentration of SDS and B-mercaptoethanol were used (7). Electrophoresis under nonreducing conditions was the same as described above, except that the 2-mercaptoethanol was deleted from the sample buffer (13). Standard molecular size markers included [1“Cmeosin (200 kilodaltons [kDa]) phosphorylase B (92.5 kDa), bovine serum albumin (68 kDa), ovalbumin (43 kDa), and chymotrypsinogen A (25.7 kDa), all from Bethesda Research Laboratories, Inc. Molecular sizes were calculated by interpolation between standard proteins by the method of Weber and Osborn (26). Fluorography was performed by the method of Bonner and Laskey (1). Autoradiography of the dried gels was carried out at -70°C for the times indicated in each figure legend. RESULTS The assay used for molecular characterization of MDHV-B was immuno- precipitation with the highly specific RaB serum (25) followed by SDS-PAGE, performed essentially by the procedure used for the molecular characterization of MDHV-A and its precursor with RoA (4,7). The immuno- precipitation step was critical because (i) the use of highly specific RaB serum was the only way to distinguish a quantitatively minor protein from the background of more abundant cellular and viral proteins, and (ii) RaB serum was prepared with highly purified MDHV-B and was known to be specific for the B antigen by its ability to specifically immuno- precipitate the B antigen, as identified by immunodiffusion analysis (25). RVT-infected cells were also used in comparison for identification of their B antigens, since immunodiffusion studies revealed a common RVT-MDHV B antigen (12,25). Therefore, immunoprecipitation analysis of both types of infected-cell lysates should yield a very similarly or identically sized protein that can be identified as the B antigen, as was done successfully with A antigen (4). In preliminary studies, MDHV- and HVT-infected-cell lysates labeled for 24 h in the absence of TM were analyzed with R08, and three virus- specific proteins were detected that migrated with apparent molecular sizes of 100,000, 60,000, and 49,000 Da (data not shown). Appropriate controls with uninfected-cell lysate and NRS were negative for these three polypeptides; equivalent controls are presented below. In the preliminary studies, the three polypeptides of interest were not labeled well in relation to cell proteins, and some nonspecific trapping of the latter existed with the X-ray exposure times needed to detect the three 39 XI 40 polypeptides. This trapping was especially a problem in visualizing the 49-kDa polypeptide. This polypeptide was not well resolved in these early linear gels (data not shown), possibly because of distortions in that region of the gels due to a slight overload of immunoglobulin heavy chains. However, whenever the 60- and 100-kDa polypeptides were seen, there was also a large thickening or darkening or both in the area of the non- specifically trapped ubiquitous band of that approximate size, also seen in control cells with R08 or in control or infected cells with NRS. These problems were partly resolved by adopting a 4-h labeling period, which gave a much better ratio of the three virus-specific labeled polypeptides to nonspecifically trapped cell protein (Fig. 1). In addition, a gradient gel (Fig. 1) was tried to resolve the remainder of the difficulty. In this gel, the 49-kDa MDHV—B and RVT B-antigen (RVT-B) polypeptides were clearly separated from the ubiquitous bands just below them, and the 60- and 100-kDa polypeptides were still well resolved (Fig. 1, lanes 3 and 5). Although size estimates can vary slightly between laboratories, the 49-, 60-, and 100-kDa sizes determined in this study are very similar to the previous estimates by Silva et al. (23,24), whose numbers were adopted to maintain consistency in reporting and communicating. The polypeptide labeled p79 bears no relationship to B antigen; its identity will be further addressed below. In conclusion, the MDHV-RVT B antigen appears to consist of a family of three polypeptides of 100,000-, 60,000-, and 49,000- kDa molecular sizes. Since the B antigen was previously identified as a glycoprotein by its abilities to be labeled with [14Cngucosamine (25), to bind to concanavalin A, and to elute with a-methyl-D-mannoside (25), TM was used to inhibit N-linked glycosylation and to study the precursor protein(s) Fig. 1. SDS-PAGE identification and characterization of RVT-B and MDHV-B polypeptides immunoprecipitated with R08. Lysates were prepared from cells labeled with [358]methionine for 4 h with (+) or without (-) TM (2 ug/ml) as described in Materials and Methods. Immunoprecipitation with RaB and SDS-PAGE analyses were done as described in Materials and Methods. Analysis was on a gradient gel of 7.5 to 20% acrylamide. Fluorographic exposure was for 24 h. M, Molecular size markers; CON, control; INF, infected; k, kilodaltons; *, tentative nomenclature. 41 MDHV HVT M CON INF INF TM—-)--l-—+—+ Zing/ml 200k—— ‘2'“.‘5; ‘ gplOO 92 5k —- ~ g: pr88 68k - - p79 "‘ "' -' 9060 — - - — gp49 25.711 —- M 123456 42 43 that might result. When MDHV- and HVT-infected cells were treated with TM at concentrations adequate to inhibit glycosylation of MDHV-A (7), none of the three MDHV-B or RVT-B polypeptides described above were seen; instead, proteins of 88,000 and 44,000 Da immunoprecipitable with the R08 serum were observed (Fig. 1, lanes 4 and 6). Immunoprecipitation analysis of the same HVT- and MDHV-infected-cell lysates with NRS resulted in no nonspecifically trapped 88,000-Da molecule (data not shown; Fig. 2A). With the improved labeling procedure and improved separation on a gradient gel (Fig. 1), the possible second MDHV-B precursor polypeptide was more clearly seen at 44,000 Da. In this case, the 44-kDa polypeptide, tentatively named pr44 (see Discussion), was seen in two lanes designated infected plus TM (Fig. 1, lanes 4 and 6), whereas it is absent from the lane designated control plus TM (Fig. 1, lane 2). There is also a faint 44-kDa band in the control-minus-TM lane (Fig. 1, lane 1). The conclusions that this polypeptide is different from, and not specifically immunoprecipitated like, pr44 are supported by the facts that it is (1) either not always present (Fig. 2, NRS and control lanes) or present in variable amounts, usually less than (data not shown; Fig. 1, lanes 1, 3, and 5) pr44 itself (Fig. 1, lanes 4 and 6), and (ii) present only in the absence and not the presence of TM (Fig. 1, lanes 1 and 2, respectively; other data not shown), whereas pr44 is seen only with plus-TM samples. The existence of pr44 is also clearly shown in Fig. 2A. The pattern of three MDHV-B polypeptides and two MDHV-B polypeptides in the absence and presence of TM, respectively, as seen upon gradient gel analysis (Fig. 1), is now a common feature when the 4-h labeling period is used and good resolution is achieved, even in linear gels (Fig. 2A). Fig. 2. Identification and characterization of the HVT-B and MDHV-B polypeptides immunoprecipitated with RaPM and 108 sera. [3SSJmethionine labeling was for 4 h as described in the legend to Fig. 1. Lysate formation and immunoprecipitation analysis was as for Fig. 1, 2, and 3, except for the variations described below. (A) Immunopre- cipitation with NRS and RaPM. RaPM was first absorbed with normal-cell sonic extract (three cycles) by sonicating a cell pellet in 500 pl of serum, incubating it overnight on ice, and centrifuging it to remove cell debris. Fluorographic exposure was for 24 h. (B) Immunopre- cipitation with immune chicken serum (ICS) plus rabbit anti-chicken IgG (RaC). RAC was added at a ratio of 0.5:1 (i.e., 0.5 ul of RAG/1 ul of 108) to allow the use of §. aureus for the collection of immunopre- cipitated protein, as is described for Fig. 1, 2, and 3. Fluorographic exposure times are as indicated. CON, Control; *, tentative nomen- clature; k, kilodaltons. 44 Esau +I+I+1+I +|+ITE$ +1 + I+I+|+ n+1 H x? I i - 1.. n/xwmflm M I’D. lane 8 - 1! I I 003 l - - x I lllll K at , .Ixmm $.81 Ix 85 I- U . H>I 2.52 20 #2.. >102 200 ._.>I >752 200 .5: >19. :00 E¢N1o5mo xo >21x¢ :3 Zavom mmz .m .4. 45 46 To further identify the 100,000-, 60,000-, and 49,000-0a proteins as MDHV-B, their abilities to be specifically immunoprecipitated with the RaB serum were blocked in a competition experiment. Both 25 and 5 ul of an unlabeled isoelectrophoretically purified MDHV-B preparation (25) (MDHV-B titer, 1 U/ml; exact protein concentration undetermined due to very limited supply and ampholite interference with protein analysis) were adequate to completely block immunoprecipitation of the labeled 60,000-0a molecule, the most prominent component of the three glycoproteins in the B-antigen complex (Fig. 3). However, a volume of only 1 ul was unable to block immunoprecipitation completely, and partial immunoprecipitation of gp60 occurred, as indicated by its appearance at its gel position, illustrating the sensitivity and specificity of the procedure. Virtually the same results were seen for gp49, the only exception being some radioactivity at 49,000 Da when 25 and 5 ul were used (Fig. 3). Actually, since this gel system is an older one (preliminary data discussed above), not the newer gradient gel (Fig. 1), this radioactivity is most likely the background radioactivity often seen in this region with linear gels and seen here due to intentional overexposure (Fig. 3) and not to gp49, which appears to be blocked as extensively as gp60. Although there is a considerable blocking effect for gp100, it does not appear as extensive as that for gp60 and gp49. That result could be explained two ways. First, it is possible that the same degree of extensive blocking occurred, but upon intentional overexposure, a nonspecifically trapped protein appeared at about 100 kDa in the lanes labeled 25 and 5 ul, as apparently occurred at 49 kDa (see above). On the other hand, while the form or forms of MDHV-B present in the purified B-antigen preparation used for blocking are not known at this time, it is possible that only the smaller forms were Fig. 3. Blocking of RaB with MDHV-B purified by isoelectric focusing. Lysate preparation and analysis were as described in the legend to Fig. 1, except that the earlier 24-h labeling period as used (see Results) and RaB was first incubated with various volumes of MDHV-B obtained from isoelectric focusing as described in Materials and Methods. Analysis was on a linear gel of 7.5% acrylamide. Fluorographic exposure was for 4 days. 47 ul purified MDHV-B O 25 O 5 O I ..l g' in ~ "1 Lu, eacl was Stud data Mfiu~ “-A‘J 49 solubilized by sonic extraction (25), and that effect may result in less complete blocking of gp100. Nevertheless, 25 and 5 pl of purified MDHV-B blocks gp100 distinctly more than 1 pl (seen more clearly on lighter exposure), a result not seen with other proteins that show some apparent blocking. The word apparent is used because p79 is a known sticky protein (see below), and 8 antigen is also known to be sticky (25) and may aggregate with other proteins easily. Despite these problems inherent in work with MDHV-B and not faced with MDHV-A (4), it appears that there is specificity of blocking, even of gp100, and the blocking studies serve as significant support for identification of the three polypeptides as MDHV-B. When this experiment was repeated to block HVT-B, the same results were observed (data not shown). Therefore, the immunoprecipitation of both MDHV-B and HVT-B by RaB serum is specific, and the two common antigens each exist as identically sized sets of three proteins at 100,000, 60,000, and 49,000 Da. As another means of molecular characterization, immunoprecipitation was performed with a special serum previously shown by immunodiffusion studies to have primarily anti-B-antigen activity (Velicer, unpublished data). This serum, RaPM, was made with plasma membranes isolated from MDHV-infected cells (8). In their development of the membrane purifi- cation procedure, the authors used enzymatic analysis to confirm that the membrane fraction isolated was plasma membrane (8). Since the rabbits could react with normal-cell as well as virus-encoded proteins, there was considerable background radioactivity (data not shown) until adsorption was performed (three cycles) with normal DEF cells sonicated in the presence of RaPM. After adsorption, this serum is still positive for MDHV-B in immunodiffusion (I. Sithole, R.A. Stringer, and L.F. Velicer, 50 unpublished data). As can be seen, background immunoprecipitation of control cell materials is virtually nonexistent (Fig. 2A). When lysates of MDHV- and HVT-infected cells labeled in the absence or presence of TM were analyzed by immunoprecipitation with adsorbed RaPM serum, the same set of three B-antigen glycoproteins (gp100, gp60, and gp49) or two poly- peptides (pr88 and pr44), respectively, was precipitated (Fig. 2A). These proteins were identical when compared directly to the B antigen pre- cipitated by RaB (Fig. 1 and Fig. 4, [35$]methionine; Sithole, data not shown). This immunoprecipitation was specific, since these proteins were not observed by immunoprecipitation analysis of control cell lysates with RaPM and were not immunoprecipitated from infected-cell lysates by NRS (Fig. 2A). Since the same polypeptides were immunoprecipitated by the two rabbit sera with known anti-B activity (Fig. 1, 2A, and 3) and the three poly- peptides described by Silva and Lee with monoclonal antibody were also immunoprecipitated with convalescent chicken plasma (23,24), this study was extended to include the same type of reagent, here called immune chicken serum (ICS), which also has anti-B activity (25), to see if the correlation would continue. Although chicken serum does not bind well to S. aureus protein A, a small amount of rabbit anti-chicken serum (RAC) (at a ratio of 0.5 volume RAC to 1 volume chicken serum, 1/12 the 6:1 ratio optimal for double antibody precipitation; J. Seibel, R.A. Stringer, and L.F. Velicer, unpublished data) allows doing S. aureus protein A-based immunoprecipitation with ICS with minimal nonspecific trapping. Since ICS is from naturally infected birds (i.e., horizontal transmission of virus), the only foreign protein the birds could specifically react to would be that made due to expression of the viral genome during infection. As Fig. 4. [1”C1glucosamine labeling of the three MDHV-B polypeptides. [1uCngucosamine labeling in the presence (+) and absence (-) of TM 2 pg/ml was as described in Materials and Methods. [3581methionine labeling was for 4 h as for Fig. 1. Subsequent lysate preparation, immunoprecipitation with rabbit serum RaB, and SDS-PAGE analysis was as for Fig. 1, 2, and 3. Fluorographic exposures were for 64 days. Photography was varied slightly to emphasize the [358]methionine-labeled gp100, gp60, and gp49 on the left half of the photograph and to emphasize the same molecules labeled with [1uCngucosamine on the right half. CON, control; k, kilodaltons. 51 35s METHIONINE MDHV-infected 52 '40 GLUCOSAMINE MDHV-infected CON R R RNKR O‘RPOL SMB +++ R NozR RPoz SMB 1:112 lcn was and g pepti many In 0; inzun HVT- (23‘). 53 would be expected, this serum is much more complex and reacts with a number of proteins from the infected cell, but with none from uninfected cells (Fig. 28). By looking at both the 4- and 24-h exposures, the characteristic sets of three HVT-B and MDHV-B glycoproteins (gp100, gp60, and gp49) present when TM was not used and two HVT-B and MDHV-B poly- peptides (pr88 and pr44) present when TM was used can be seen despite the many other immunoprecipitated proteins. Thus, the correlation holds true. In our hands, the third serum with anti-B activity, as determined by immunodiffusion, immunoprecipitated the same set of polypeptides from both HVT- and MDHV-infected cells, thereby confirming the data of Silva and Lee (23). The ICS serum also immunoprecipitated a 79,000-Da polypeptide from MDHV- and HVT-infected cell lysates but not from control cell lysates (Fig. 2B). This virus-specific polypeptide is singled out for attention since it is one of the most extensively labeled proteins in infected cells, as reported by Silva and Lee (23). This extensive labeling and an apparent ability to be nonspecifically trapped or adsorbed quite readily (23) seemed to contribute to its frequent but more variable appearance (sometimes not seen at all and always much less than when immuno- precipitated by ICS) even with NRS (data not shown), but only from infected cells (Fig. 1, 2A, and 5). TM has no effect on its apparent molecular size (Fig. 28), indicating an absence of N-linked glycosylation and contributing to its identification as the nonspecifically trapped protein not only without, but also with, TM (Fig. 1, 2A, and 5). The facts that it is specific for infected cells and is immunoprecipitated by ICS from naturally infected birds strongly suggest that the 79,000-Da molecule is virus encoded. H4. ‘IV 54 As might be expected from the results of glucosamine labeling of MDHV-B seen by immunodiffusion (25) and of gp100, gp60, and gp49 detected with monoclonal antibodies (23), the three polypeptides immunoprecipitated with RaB, RaPM (Fig. 4), and ICS (data not shown) were labeled with [1“C3glucosamine. The [3SSJmethionine lanes on the left side of Fig. 4 are included to establish the position of the 100-, 60-, and 49-kDa poly- peptides in the same linear gel. All control lanes on the right half of the fluorogram were negative for the [1uCngucosamine. Thus, the only positive labeling seen is in the positions of the three polypeptides of interest in the R08 and RaPM lanes. The same results were obtained with ICS (data not shown). Since RaB is a less potent serum and immuno- precipitated [3SSJmethionine-labeled molecules much more poorly than did RaPM (Fig. 4, left half), it was not surprising that the faint [1uCngucosamine-labeled molecules were not immunoprecipitated with RaB as well as with the latter serum. As was also expected, no glucosamine labeling of the 49-, 60-, and 100-kDa polypeptides occurred in the presence of TM. In agreement with the results of Silva and Lee (23), the p79 immunoprecipitated with ICS was not labeled with [1"Cngucosamine (data not shown). Since gp100, gp60, and gp49 were specifically immunoprecipitated with the RaB and RaPM sera known to have anti-B activity, it was of interest to us to understand more about the relationship of these three polypeptides to each other. An experiment was performed to test the hypothesis that the three peptides, or two of the three, are linked together by disulfide bridge formation. This test was done by excluding the B-mercaptoethanol from the sample buffer (13), since it is conceivable that the polypeptides are attached by disulfide bridges which are released by the reducing Fig. 5. SDS-PAGE analysis of HVT-B and MDHV-B under reducing and nonreducing conditions. [358]methionine labeling was for 4 h as for Fig. 1 and 4. All other methods were as for Fig. 2, 3, and 5, except that nonreducing conditions were established by the elimination of B- mercaptoethanol from some samples as indicated in the figure. -, Absence of TM; +, presence of TM; M, 1"'C-labeled molecular size markers; CON, control; k, kilodaltons. 55 —CON— —MDHV— —HVT— TMZuq/ml--++--++-—++ ZME .+-+—+—.+—+_+_ -—~ 200k— _ — gplOO 92.5k _ pr 88 68 k— _ 1 - . — gp60 57 conditions created by the B-mercaptoethanol in the sample buffer. The deletion of B-mercaptoethanol from the sample buffer did not result in aggregation of the three (or two of the three) polypeptides into a larger- molecular-weight form (Fig. 5). Instead, gp100, gp60, and gp49 are all visible at exactly the same gel positions as their B-mercaptoethanol- treated counterparts. Therefore, it appears that these three glyco- proteins are not connected to each other by disulfide bridges, but are three separate forms of MDHV-B. An interesting result of deleting the B-mercaptoethanol from the sample buffer was that the lane width decreased in comparison to that of the B-mercaptoethanol-containing samples (Fig. 5). However, this decrease appeared to have no effect on the resolution and migration of the three MDHV-B glycoproteins. DISCUSSION The work reported here was an extension of the initial purification and physicochemical characterization of MDHV-B by Velicer et al. (25) and used the highly specific polyvalent RaB antiserum prepared against this highly purified antigen as part of the earlier study. Thus, this study gives a molecular identity to MDHV-B previously identified only in terms of immunodiffusion analysis. Interestingly, the antigen, as identified by immunoprecipitation and SOS-PAGE analysis (Fig. 1), appears to consist of three glycoproteins (data discussed below) of 100, 60 and 49 kDa (gp100, gp60, and gp49, respectively) sharing one or more common antigenic determinant(s). That this set of three polypeptides represents the MDHV-B previously seen only upon immunodiffusion (25) is further supported by their consistent immunoprecipitation with RaPM (Fig. 2A) and ICS (Fig. 28), two sera with anti-B activity as determined by immunodiffusion (25; Velicer, unpublished data). In all studies, HVT-infected cells were analyzed in parallel with MDHV-infected cells, and in every instance, the molecules with B-antigen immunological reactivity (gp100, gp60, gp49, pr88, and pr44; see below) were virtually identical in size on SDS-PAGE. The finding that poly- peptides encoded by both viruses were immunoprecipitated by sera with anti-B activity was not unexpected; it has long been known that B antigen is a common antigen (2,3,25). The evidence that MDHV-B and HVT-B poly- peptides are virtually identical in size is important to our overall understanding of the extent of homology between MDHV and HVT. If differences in B antigen do exist, they most likely are small, and their 58 ‘1‘) W1 (f If 59 demonstration will require more refined analysis, i.e., possibly tryptic peptide analysis or, more likely, sequence analysis. A glucosamine labeling experiment was included in this study for completeness, since (i) the B antigen originally detected by immuno- diffusion was shown to contain glucosamine (25), (ii) MDHV-B was shown to consist of three molecules, and (iii) the three same-sized molecules immunoprecipitated by monoclonal antibody (discussed below) are glyco- proteins (23). As expected, all three molecules immunoprecipitated by the three sera with anti-B activity were glycosylated (RaB and RaPM in Fig. 2; ICS data not shown). Clearly, the terminology gp100, gp60, and gp49 originally suggested by Silva and Lee (23,24) can be extended to the three molecules seen in this study as well. The importance of this work with polyclonal sera is further enhanced when it is realized that, in two laboratories, the use of several independently prepared monoclonal antibodies identified what appears to be the same set of three glycoproteins (5,23,24). Especially significant is the fact that a single monoclonal antibody can immunoprecipitate all three glycoproteins, results consistent with, but not proving, the fact that all three share the same antigenic determinant(s). The sizes reported by Ikuta et al. (5) are slightly larger than those reported by Silva and Lee (23) and vary to a slight extent between MDHV- and HVT-infected cells. However, since minor size variations are common, the three glycoproteins of Ikuta et al. (5) are most likely the same as those reported here based on identification with polyclonal antibody known to have MDHV-B activity and those detected by Silva and Lee (23) with another monoclonal antibody. Ikuta et al (5) tentatively identify their set of three molecules as gB, since these molecules are detected with monoclonal antibody different from 60 that detecting gA, their tentative name for what they also call A antigen, based on its release into the culture medium. However, they present no experimental evidence or other basis for relating their polypeptides to B antigen. Silva and Lee (23,24) include their set of three glycoproteins (gp100, gp60, and gp49) immunoprecipitable with monoclonal antibody as part of the five common viral proteins immunoprecipitated from MDHV- infected cells (the sixth antigen is MDHV-A in the culture medium of these cells) with chicken serum against MDHV and HVT, but no attempt is made to speculate on the molecular nature of MDHV-B. In their initial discussion of common antigens that might be candidates for the study of immune protection by HVT, they refer to gp100, gp60, and gp49 in addition to A and B antigens (23). In recent discussions regarding the relationship of the data from this study to their own data (Silva and Lee, personal communication), it was clear that they now view gp100, gp60, and gp49 as various forms of B antigen. In view of the above facts, the particular value of the work reported here is in the experimental evidence clearly identifying gp100, gp60, and gp49 (23,24) as MDHV-B and HVT-B. This result then clarifies a major aspect of the MDHV-HVT work in progress in several laboratories, namely, the identification, quantification, and characterization of common antigens that might be involved in protective immunity (23,24). In fact, Ono et al. (17) have recently reported that the glycoproteins purified by affinity chromatography with monoclonal antibody to gB (5), as discussed above, elicited partial protection against MD. Also, demonstration that three of the five intracellular common polypeptides identified by Silva and Lee (23) appear to be forms of MDHV-B reduces by two the number of common antigens, and possibly genes, that have to be dealt with. Thus, 61 the clarification provided herein with polyvalent antisera makes even more valuable and meaningful the more precise work that can be done with monoclonal antibodies (5,6,17,23,24), since glycoproteins such as gp100, gp60, and gp49 now have an identity in terms of earlier work in this system (25). Important questions perceived in the course of this study concern the possibilities that (i) not all three polypeptides are MDHV-B, but exist in a complex so that one or two non-B glycoproteins immunoprecipitate because they are attached to at least one MDHV-B glycoprotein, and that (ii) a disulfide-linked complex exists even if all three glycoproteins are forms of MDHV-B. The idea of a disulfide-linked complex is based in part on the finding of such results with two pseudorabies virus glycoproteins (13). Based on the same approach, the results presented here (Fig. 5) indicate that gp100, gp60, and gp49 are not linked by disulfide bridge formation. The results summarized above do not rule out the possibility of the three proteins being held in some sort of complex or aggregate other than one held together by disulfide bridges. In fact, there is now evidence in support of this concept (Silva, personal communication). Infected-cell lysate was layered on a bovine serum albumin gradient which was centri- fuged and fractionated. When the fractions were subjected to immuno- precipitation with monoclonal antibody known to precipitate gp100, gp60, and gp49, all three glycoproteins were immunoprecipitated from the same fractions and were lower in the gradient than would be predicted for a single soluble protein. These data suggest that the three glycoproteins exist as some sort of complex in the lysis buffer, which would explain why all three glycoproteins could be immunoprecipitated with monoclonal antibodies (5,6,23,24). 62 Although more rigorous confirmation will come from future studies in this system, the existing data strongly suggest that all three glyco- proteins (gp100, pg60, and gp49) are forms of B antigen. Two laboratories (5,6,24) report pulse-chase studies that indicate that gp100 is processed to generate gp60 and gp49 (discussed further below), and in one case (Silva, personal communication), the results of limited Staphylococcus V8 protease digestions are consistent with (but do not prove) that conclusion. Furthermore, we now have very preliminary data (Sithole and Velicer, unpublished data) indicating that monensin inhibition of glycoprotein processing reduces the appearance of gp60 and gp49, whereas the appearance of gp100 still occurred. Although it is recognized that the nature of the complex described above remains to be determined and that the apparent processing of gp100 to yield gp60 and gp49, as discussed above, should be shown more rigorously, such studies are beyond the scope of this paper. With five papers already published on the three glycoproteins without any immuno- logical correlation with well-established immunology of the system (5,6, 17,23,24), including several papers suggesting that these glycoproteins elicit virus-neutralizing antibody and one paper (17) reporting that they elicit protective immunity, the work reported here starting with RaB is particularly timely and valuable. First, a clear link is now established with the B antigen of past studies (25) so that it is clear which antigen is being used in virus neutralization and protection studies. Second, the physical and chemical methods and data worked out for B antigen (25) are now available to other workers in this field as a result of the clarifying data incorporated into this report. Thus, as efforts are made to provide more rigorous and in-depth analyses of gp100, gp60, and gp49 at all 63 levels, our full range of knowledge of B antigen can be applied to this future work. Since the MDHV-B originally defined by immunodiffusion and the three proteins now identified as MDHV-B were found to be glycoproteins, TM was used to inhibit N-linked glycosylation of the three polypeptides. Two proteins (pr88 and pr44) immunoprecipitable with sera having anti-B activity were found in the presence of TM (Fig. 1 and 2). Two immuno- precipitable forms could exist, despite only one trUe precursor, if the following occurred. It is possible that pr88 is the nonglycosylated or O-linked glycosylated precursor of gp100, which could be cleaved to form gp60 and gp49, as appears to be the case (see above), in a manner analogous to the precursor processing of pseudorabies virus glycoproteins (13). The smaller molecules with MDHV-B activity, appearing both in vivo after TM inhibition (24; Fig. 1 and 2) and in vitro after cell-free translation (Sithole, Isfort, and Velicer, unpublished data), may exist if the larger precursors are unstable without sugar residues and therefore are rapidly degraded. A situation similar to this has been reported for the influenza virus glycoprotein hemagglutinin (22). Thus the nomen- clature for pr44 is tentative (Fig. 1,and 2A), since many questions exist concerning the true nature of the glycoprotein, the clarification of which requires experiments beyond the scope of this preliminary characterization study. In contrast, the identity of the protein designated pr88 as a precursor polypeptide seems more certain, since molecules of that size were found after TM treatment by this study, by Silva and Lee (personal communication), and by Ikuta et al. (6). As indicated above, data exist indicating that the largest molecule of the set of three is processed to generate the two smaller molecules (6,24). A logical prediction would be 64 that pr88 is glycosylated to yield gp100. However, in these very preliminary pulse-chase studies, the failure to detect pr88 during a 10-min pulse can be explained by rapid glycosylation and processing, as occurred with MDHV-A (7), steps that would be missed with a pulse of that length. These preliminary pulse-chase reports, the presence of pr88 after TM inhibition, and the discovery of an approximately same-sized molecule after cell-free translation (Sithole, Isfort, and Velicer, unpublished data) allow construction of precursor-processing and glycosylation models that can be tested experimentally by appropriate pulse-chase studies (7) and supplemented by tryptic peptide analysis. Such studies are planned for the future, but are clearly out of the scope of this paper, which started as a simple attempt at molecular characterization of MDHV—B. Finally, an experiment was performed to take advantage of an antiserum (RaPM) available in this laboratory which was made against purified infected-cell plasma membranes and was reactive with MDHV-B. This antiserum was also able to immunoprecipitate the three B-antigen glycoproteins, as well as the two polypeptides made in the presence of TM (Fig. 2A). In fact, it is a more potent serum than RaB (Fig. 4) and, when absorbed, shows even less background (Fig. 2A) than RaB (Fig. 1). Thus, as a result of these studies, RaPM has become our preferred serum for other use. Also, these data may mean that at least one form of B antigen is an integral part of the plasma membrane of infected cells in such a manner that a key antigenic determinant is accessible to the host immune system. This finding could explain why ICS has a high titer for B antigen. These results are potentially significant in that they are consistent with the reports of Ikuta et al. (5,6) that monoclonal antibody against what they 65 term HVT-gB and MDHV-g8 and of Silva and Lee (24) that monoclonal antibody capable of immunoprecipitating gp100, gp60, gp49 can neutralize HVT or MDHV. If the set of three polypeptides called gB (5,6) is the same as gp100, gp60, and gp49 identified in this study and by Silva and Lee (24), the virus neutralization results (5,6,24) suggest that some form of MDHV-B is the primary antigen responsible for generating virus-neutralizing antibody. The relationship between this role of MDHV-B in virus neutra- lization and its recently reported (17) role of protection remains to be elucidated. Determining the relationship of gp100, gp60, and gp49 to the plasma membrane of infected cells should facilitate learning more about the possible role of at least one form of MDHV-B in virus neutralization and immune protection (8,17,28). Determining which form is part of the membrane is also important, because this form may contain antigenic determinants essential for virus neutralization and protection that are different from the one(s) in common between gp100, gp60, and gp49. In summary, the data reported here are significant for several reasons. (i) The data contribute to the molecular identification and characterization of MDHV-B as planned, but in doing so, the facts serve as a valuable link identifying the antigen as originally studied by immuno- diffusion with polypeptides immunoprecipitated by monoclonal antibody. (ii) The molecular identification work done thus far is already the basis of gene identification studies now in progress. (iii) The discoveries of MDHV-B as a complex of three glycoproteins and of what could be either one or two true unglycosylated precursors indicate that the antigen is considerably more complex, than originally expected. As a result, this study serves to identify unanswered questions regarding this complex, formulate a precursor-to-product processing model, and allow the design of 66 ways to test the model. (iv) The demonstration that at least one form of MDHV-B is apparently part of the infected-cell plasma membrane leads to important questions, and the experiments to answer them, regarding the role of MDHV-B in the immune response and possibly in protective immunity in the MD system. With one report (17) already indicating that gB, which was earlier defined as all three glycoproteins (5,6), elicits partial protection against MD, such questions become of paramount importance. Therefore, the work reported here, although originally intended as the first attempt at molecular identification and characterization of MDHV-B, is already useful as the foundation for studies in progress or planned that will reveal much about the molecular biology and immunobiology of the complex but interesting MDHV system. LITERATURE CITED Bonner, W.M., and R.A. Laskey. 1974. A film detection method for triton labeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 46:83-88. Chubb, R.C., and A.E. Churchill. 1968. Precipitating antibodies associated with Marek's disease. Vet. Res. 83:4-7. Churchill, A.E., R.C. Chubb, and W. Baxendale. 1969. The attenuation with loss of antigenicity, of the herpes-type virus of Marek's disease (strain HPRS-16) on passage in cell culture. J. Gen. Viol. 5:557'563. Glaubiger, C., K. Nazerian, and L.F. Velicer. 1983. Marek's disease herpesviruses. IV. Molecular characterization of Marek's disease herpesvirus A antigen. J. Virol. 45:1228-1234. Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 1984. Identification with monoclonal antibodies of glycoproteins of Marek's disease virus and herpesvirus of turkeys related to virus neutralization. J. Virol. 33:1014-1017. Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 1984. Processing of glycoprotein gB related to neutralization of Marek's disease virus and herpesvirus of turkeys. Microbiol. Immunol. 28:923-933. Isfort, R.J., R.A. Stringer, H.-J. Kung, and L.F. Velicer. 1986. Synthesis, processing, and secretion of the Marek's disease herpesvirus A antigen glycoprotein. J. Virol. 51:464-474. Kaaden, O.-R., and B. Dietzschold. 1974. Alterations of the immunological specificity of plasma membrane from cells infected with Marek's disease and turkey herpesvirus. J. Gen. Virol. 25:1-10. 67 10. 11. 12. 13. 14. 15. 16. 68 Kessler, S.W. 1975. Rapid isolation of antigens from cells with a staphylococcal protein A-antibody absorbent: parameters of the interaction of antibody—antigen complexes with protein A.J. Immunol. 11::1617-1624. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 221:680-685. Long, P.A., J.L. Clark, and L.F. Velicer. 1975. Marek's disease herpesviruses. II. Purification and further characterization of Marek's disease herpesvirus A antigen. J. Virol. 15:1192-1201. Long, P.A., P. Kaveh-Yamini, and L.F. Velicer. 1975. Marek's disease herpesviruses. I. Production and preliminary characterization of Marek's disease herpesvirus A antigen. J. Virol. 12:1182-1191. Lukacs, N., H.-J. Thiel, T.C. Mettenleiter, and H.-J. Rziha. 1985. Demonstration of three major species of pseudorabies virus glycoproteins and identification of a disulfide-linked glycoprotein complex. J. Virol. 53:166—173. Marek, J. 1907. Multiple Nervenentzuendung (Polyneuritis) bei Huchnern. Dtsch. Tieraerztl. Wochenschr. 15:417-421. Nazerian, K. 1980. Marek's disease: a herpesvirus-induced malignant lymphoma of the chicken, p. 665-682. In G. Kein (ed.), Viral oncology. Raven Press, Publishers, New York. Okazaki, W., R.C. Purchase, and B.R. Burmester. 1970. Protection against Marek's disease by vaccination with a herpes-virus of turkeys (HVT). Avian Dis. 14:413—429. 17. 18. 19. 20. 21. 22. 23. 69 Ono, K., M. Takashima, T. Ishikawa, M. Hayashi, I. Yoshida, T. Konobe, K. Ikuta, K. Nakajima, S. Ueda, S. Kato, and K. Hirai. 1984. Partial protection against Mareks' disease in chickens immunized with glycoprotein gB purified from turkey-herpesvirus-infected cells by affinity chromatography coupled with monoclonal antibodies. Avian. Dis. 22:533'539. Pappenheimer, A.M., L.C. Dunn, and V. Cane. 1929. Studies on fowl paralysis (neurolymphomatosis gallinarum). I. Clinical features and pathology. J. Exp. Med. 49:63-86. Pappenheimer, A.M., L.C. Dunn, and S.M. Seidlin. 1929. Studies on fowl paralysis (neurolymphomatosis gallinarum). II. Transmission experiments. J. Exp. Med. 49:87-102. Payne, L.N. 1982. Biology of Marek's disease virus and the herpes- virus of turkeys, p. 347-431. In B. Roizman (ed.), The herpesvirus, Plenum Publishing Corp, New York. Purchase, H.G. 1977. The etiology and control of Marek's disease of chickens and the economic impact of a successful research program, p. 63-81. In J.A. Romberger, J.D. Anderson, and R.L. Powell (ed.), Beltsville Symposium on Agricultural Research, vol. 1. Virology in agriculture. Alanheld, Osmun & Co. Publishers, Inc. Montclair, N.J. Schwarz, R.T., J.M. Rohrschneider, and M.F.G. Schmidt. 1976. Suppression of glycoprotein formation of Semliki Forest, influenza, and avian sarcoma virus by tunicamycin. J. Virol. 19:782-791. Silva, R.F., and L.F. Lee. 1984. Monoclonal antibody-mediated immunoprecipitation of protein from cells infected with Marek’s disease virus or turkey herpesvirus. Virology 136:307-320. 24. 25. 26. 27. 28. 70 Silva, R.F., and L.F. Lee. 1985. Isolation and partial characterization of three glycoproteins common to Marek's disease virus and turkey herpesvirus-infected cells, p. 101-110. In B.W. Calnek and J.L. Spencer (ed.), Proceedings of the International Symposium on Marek's Disease. American Association of Avian Pathologists, Inc., Kennett Square, Pa. Velicer, L.F., D.R. Yager, and J.L. Clark. 1978. Marek's disease herpesviruses. III. Purification and characterzation of Marek's disease herpesvirus B antigen. J. Virol. 21:205-217. Weber, K., and M. Osborn. 1969. The reliability of molecular weight determination by dodecyl-sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244:4406-4412. Witte, 0.N., and D.F. Wirth. 1979. Structure of the murine leukemia virus envelope glycoproteins precursor. J. Virol. 29:735-743. Wyn-Jones, A.P., and O.-R. Kaaden. 1979. Induction of virus- neutralizing antibody by glycoproteins isolated from chicken cells infected with a herpesvirus of turkeys. Infect. Immun. 25:54-59. CHAPTER III Synthesis and Processing of the Marek's Disease Herpesvirus B Antigen Glycoprotein Complex Sithole, I., L.F. Lee, and L.F. Velicer. To be submitted to J. Virol. 71 ABSTRACT The Marek's disease herpesvirus B antigen complex was previously immunologically identified and molecularly characterized as a set of three glycoproteins designated goiOO, gp60, and gp49 based on apparent molecular weight and immunoprecipitation with both polyclonal and monoclonal antibody. Immunoprecipitation analysis, previously with polyclonal and more recently with monoclonal antibody, of infected cell lysates labeled with 358 methionine in the presence of tunicamycin, an inhibitor of N-linked glycosylation, revealed two putative precursor molecules of 88 kilo daltons (kDa) (pr88) and 44 kDa (pr44). High resolution pulse-chase studies revealed that gp100 was a glycosylated intermediate, which was processed to yield gp60 and gp49. This cleavage was inhibited by monensin, an inhibitor of glycoprotein processing. Endo-B-N-acetylglucosaminidases F and H (Endo-F, Endo-H) reduced gp100 to pr88, indicating that the latter is an intermediate in the biosynthetic pathway. These same enzymes reduced gp49, and to a lesser extent gp60, to pr44, suggesting that pr44 is their polypeptide backbone. Significant support for this concept is the fact that the same monoclonal antibody recognized all three molecules, gp60, gp49 and pr44. In the presence of monensin terminal addition of complex sugars was also prevented, since gp60 was replaced by a slightly faster migrating component which was insensitive to both Endo-F and Endo-H. Cell-free translation of infected cell mRNA, followed by immunoprecipitation analysis using either polyclonal or monoclonal antibody, resulted in detection of a putative unglycosylated precursor polypeptide of 44 kDa. Since pr88 was not the initial precursor polypeptide of the MDHV-B 72 73 complex its existence may have resulted from dimerization of pr44. Again detection of both pr88 and pr44 with the same monoclonal antibody is consistent with this interpretation. These collective data obtained from the cell-free and In XLXE studies with polyclonal and monoclonal antibody reactive with MDHV-B are consistent with the concept that pr44, the initial gene product, dimerizes to form pr88; and demonstrate that pr88 is actually a processing intermediate glycosylated to gp100, another processing intermediate, which is then processed to gp60 and 8949. INTRODUCTION Marek's disease (MD) is a lymphoproliferative disease of chickens that was first described by Marek in 1907 (28,37). The disease is caused by the Marek's disease herpesvirus (MDHV), and represents the first neoplastic disease for which an effective vaccine has been developed (32,34,37). The mechanism of this immunity is not fully understood, however it appears to be a two step process involving both antiviral and antitumor-cell immunity (31,32). Experimental evidence suggests antiviral immunity is superior to antitumor cell immunity, and that if the former occurs the latter is not necessary (31). The vaccine virus, herpesvirus of turkeys (HVT) (34), and MDHV are extensively related immunologically (34,44,45), sharing a large number of polypeptides that are apparently common antigens (14,40,44,45), including the two most prominent common antigens, A and B (32,33,37). In an effort to determine the mechanism of this protective immunity at a molecular level, these two common antigens have been the focus of recent immunologic and molecular biologic studies in this laboratory (10,15,16,17). The MDHV A antigen (MDHV-A), a secretory glycoprotein of 57-65 (gp57-65) (kDa), has been extensively studied with regard to its synthesis, processing and secretion (15). The gene encoding MDHV gp57-65 (MDHV-A) was identified (16) and sequenced (P.M. Coussens and L.F. Velicer, Manuscript submitted), and similar work is almost complete for HVT gp57-65 (HVT-A) (P.M. Coussens and L.F. Velicer, Manuscript in preparation). The processing studies with MDHV-A (15) serve as a useful background for designing similar studies with other antigens in this system. 74 75 The B antigens of MDHV, (MDHV-B) and HVT (HVT-B) were molecularly characterized as a complex of three immunologically related glyco- proteins, gp100, gp60 and gp49, one of which may be a membrane component, based on studies with polyclonal sera containing anti-MDHV—B activity (17). Tunicamyacin inhibition of N-linked glycosylation resulted in either nonglycosylated, or O-linked glycosylated, putative precursor polypeptides of MDHV-B and HVT-B, called pr88 and pr44, being immunoprecipitated with the same polyclonal sera (17). In that same study the identity, as MDHV-B, of similar sized polypeptides, previously detected by others using their (13,40) monoclonal antibodies, was also accomplished (17); thus making it possible to relate their virus neutralization, immunoprotection and pulse-chase studies to MDHV-B. Silva and Lee (41) have carried out pulse-chase studies with 10 min pulses and even longer chase intervals, using HVT-infected cells and their monoclonal antibody reactive with the three MDHV-B glycoproteins. They reported that the disappearance of gp100 coincided with the appearance of gp60 and gp49, suggesting the former molecule may be cleaved to generate the latter two molecules. However they state that confirmation of the relationship of gp100 to gp60 and gp49 must await further experiments. Ikuta £2 EI. (13), using HVT-infected cells, have also performed pulse-chase studies with similar time intervals, using both a monoclonal antibody and rabbit anti-MDV-gB antisera. They found essentially the same sequence of events as Silva and Lee (see above), and identified the larger glycoproteins as the precursor form of B antigen which is processed to the two smaller glycoproteins (13). However in both pulse-chase studies no smaller unglycosylated precursor forms were detected, supporting the suggestion that cells may need to be 76 pulsed for even shorter periods of time to visualize rapidly occurring early events. This may be especially true in view of the very rapid processing of MDHV-A where most of the glycosylation occurs within 1 min (15). Since their pulse-chase experiments did not detect an unglycosylated precursor polypeptide, Silva and Lee (41) used lysates from tunicamycin-treated infected cells to resolve this point. Their monoclonal antibody detected a 44 kDa molecule they called pr44, based on their interpretation that it was a precursor polypeptide. In their study a pr88 molecule was not reported, although reexamination of their data (41, Fig. 4) suggests that there may have been some protein of pr88 size present, but the amount was clearly less than pr44 (R.F. Silva, personal communication). On the other hand Ikuta SE EI. (13) detected what appears to be their equivalent of pr88, but no pr44, in tunicamycin-treated infected cells, using their monoclonal antibody. While our laboratory has more recently found both putative precursor polypeptides after tunicamycin treatment (summarized above), full analysis of their possible role in the MDHV-B precursor-processing relationships was beyond the scope of the previous study (17). The purpose of this study was to examine the entire MDHV-B precursor-processing question in more depth using approaches not previously tried. Thus, the analyses of MDHV-B pr44 synthesis and processing reported here were performed using a combination of higher resolution pulse-chase studies, two glycosylation inhibitors (tunicamycin and monensin), two endoglycosidases (F and H), protease inhibitors, and cell-free translation; and serve to provide a more complete analysis of processing events than previously existed. Of 77 particular significance was the cell-free translation analysis followed by immunoprecipitation, which revealed that pr44 is the initial unglycosylated precursor polypeptide MDHV-B. Of equal significance was the use of monoclonal antibody, which established that pr44 was a component of all larger molecules with MDHV-B reactivity. The sequence of events this combination of approaches and reagents elucidated involves the apparent dimerization of pr44 to pr88, followed by glycosylation of this intermediate to gp100 and subsequent processing of gp100 to yield gp60 and gp49. MATERIALS AND METHODS Cells and viruses. The preparation, propagation, and infection of small scale duck embryo fibroblast (DEF) cell cultures with MDHV was generally performed as first reported by Glaubiger §£,§l (10), and more recently by Isfort BE EI (17). One exception was that the MDHV strain GA used was at passage level 6, following isolation of cell-free virus from feathers tips obtained from infected birds (38). Radiolabeling of proteins. Infected and non-infected cells were labeled with 3SS-methionine at 48 hrs post-infection for 4 hrs by standard methods as previously described (17). Pulse-chase labeling was by methods already established for this system (15), with the following minor modification. Preincubation without methionine was for 40 min (15), pulse labeling was by incubation with 250 pCi/ml 35S-methionine (specific activity 1000 pCi/mmol; Amersham Corp.) for 1, 4 or 15 min, and chases were for the various times indicated in the figure legends. For labeling in the presence of tunicamycin (Sigma Chemical Co.) or monensin (Calbiochem), cells were preincubated with the drugs for 1 hr as reported (15), or at various times as indicated in the figure legends. Both drugs were present throughout the remaining course of each experiment. Antisera. Preparation and characterization of the polyclonal antisera RaA, RaB, RaPM and ICS has been described previously (15,17). The monoclonal antibody IAN86.17, an IgG1 isotype, is a cloned derivative from hybridoma 1AN86, and it immunoprecipitated gp100, gp60 78 79 and gp49 (data not shown), as previously reported for 1AN86 (40,41). However, while this cloned monoclonal antibody reacted poorly in immunofluorescence with MDHV-infected monolayer cell cultures, it exhibited a strong reaction with the surface antigen of infected cells in suspension. Also it neutralized cell-free HVT, but not MDHV (data not shown). This monoclonal antibody was used because of its immunopre- cipitation properties, and especially because of its ability to recognize an epitope present in pr44 (41). Immunoprecipitation and SDS-PACE. Immunoprecipitation was carried out as previously described (17), except that lysate preparation throughout was with detergent buffer containing 300 pg/ml of phenyl- methylsulfonylfluoride (PMSF), and a second antibody was added as needed with mouse monoclonal antibody (as described below). The immunopre- cipitates were washed, and resuspended in sample buffer, and analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PACE) (23). Acrylamide concentrations were 7.5% for all gels presented. Molecular sizes were calculated by interpolation between protein standards by the method of Weber and Osborn (46). Fluorography was performed by the method of Bonner and Laskey (4). Autoradiography of the dried gels was carried out at -70°C for the times indicated in the figure legends. Enzyme digestions. Immunoprecipitated proteins bound to staphylococcus protein A were eluted by incubating the pellets for 20 min at 37°C in 50 pl sodium citrate buffer (50 mM, pH 5.5) containing 0.2% SDS, then boiling for 2 min, and centrifuging for 3 min. The 3!”. w“ L (I) p ) 8O supernatant was collected, 10 units endo-B-N-acetylglucosaminidase H, also called endoglycosidase H, (Endo-H) (Miles Laboratories) was added, and the mixture was incubated at 37°C for 16-24 hr(1). For digestion with endo-B-N-acetylglucosaminidase-F, also called endoglycosidase F, (Endo-F) (New England Nuclear) the immunoprecipitated proteins were eluted in 100 pl sodium phosphate (100 mM pH 6.1) containing 0.1% SDS, 0.5% NP-40, and 50 mM EDTA. One quarter unit Endo-F was added to the sample and incubated as above. Control samples without Endo-H or Endo-F were treated the same way, and after incubation all samples were precipitated with ice-cold acetone for 1 hr at 4°C and centrifuged in an Eppendorf microfuge (model 5414) at top speed for three min. The precipitates were washed with 100% ethanol, vacuum dried briefly and resuspended in sample buffer for electrOphoresis. Isolation of total cellular RNA. Total cellular RNA was isolated from MDHV-infected DEF cells at 48 hr postinfection using the guanidinium isothiocyanate procedure described by Chirgwin SE El (6) and modified by Maniatis (26). Cells were harvested, pelleted, and resuspended in 4 ml guanidinium isothiocyanate buffer (4 M guanidinium isothiocyanate, 1% sodium sarkosyl and 0.1 M 2-mercaptoethanol) per 108 cells. Eight ml of the resulting lysate was layered on 3 ml of 5.7 M cesium chloride in a Tris-EDTA solution (TE) consisting of 10 mM Tris hydrochloride (26), 1 mM EDTA (pH 7.5). The RNA in the lysate was pelleted at 25,000 rpm for 20 hr at 22°C in a Beckman SW41 rotor. The pelleted RNA was solubilized in TE with 0.1% SDS, extracted with an equal volume of phenol-chloroform, ethanol precipitated, and solubilized in H20 (26). 81 Cell-free translation. Total cellular RNA from MDHV-infected cells was translated by use of a rabbit reticulocyte lysate system from Bethesda Research Laboratories. The RNA (10 pg) was translated in a 30 ml reaction volume as recommended by the supplier. An equal volume of detergent buffer was added to stop the reaction. Ten pl of this diluted translation mixture was analyzed directly by SDS-PAGE, the rest was immunoprecipitated, as described above, with undiluted antisera or an ascites fluid stock solution (previously diluted 1:50) at 1/20 volume, (i.e., for 50 pl of translation mixture, 2.5 pl of antisera or ascites fluid stock solution was used). To ensure successful immunoprecipi- tation (see above) rabbit anti-mouse IgG (Sigma) was added along with mouse ascites fluid stock solution at a ratio of 1:25, and RAC was added along with ICS as previously described (17). it nu RESULTS Kinetics of MDHV-B processing. Initially pulse-chase analyses were performed with one minute pulses, as previously done for MDHV-A processing studies (15). The intent was to identify and analyze early rapid processing events, should they exist for MDHV-B, as in the case for MDHV-A. However, during preliminary experiments it became apparent that the appearance of molecules of interest was sufficiently slow (unpublished data) that a somewhat longer pulse interval could be used to facilitate isotope conservation and still critically analyze early MDHV-B processing events. Thus for all subsequent pulse-chase analyses infected cells were pulsed 4 min. and chased at various times (Fig. 1). The first molecule with MDHV-B immunological reactivity that could be detected, gp100, was easily seen at the 12 min chase, although as early as zero min chase trace amounts appeared as a heterogeneous slightly smaller putative glycosylation intermediate which acquired gp100 properties as glycosylation proceeded. The intensity of the gp100 band increased to a maximum at 45 min of chase and declined with increasing chase periods. In contrast gp60 and gp49 were first seen after a 30 min chase and subsequently increased in amount, with the majority of the processing completed by 120 min. It is important to note that the above events coincide in a manner consistent with a precursor-product relationship. The results of these pulse-chase studies with short pulses indicate that gp100 is processed relatively slowly to yield gp60 and gp49, confirming the earlier observation by Silva and Lee (41) and Ikuta gt EI. (13). 82 Fig. 1. Kinetics of MDHV-B processing as determined by pulse-chase analysis . A 4 min pulse with 35S-methionine was followed by chases of various times as shown above each lane. Immunoprecipitation analysis of 1 ml cell lysate was done with RuPM. Fluorographic exposure was for 8 days. Molecular sizes are indicated in kilodaltons. 83 g minutes hours ' chase after 4 m'lnuto ”18?: r 03 612304512 3 4 Label gpIOO 92.51: 681: r gp60 9949 43k 84 85 A careful examination of the data in Fig. 1 reveals two lines of evidence that could be interpreted to mean that some gp49 forms directly by glycosylation of a smaller precursor polypeptide, such as the pr44 described below. One observation is the appearance of an immunoprecipi- table polypeptide slightly smaller than gp49, but increasing in size, at the 45 min and 1 hour pulse intervals. The second observation is the accumulation of more radioactivity in the gp49 band than in the gp60 t>aand at the 2, 3 and 4 hr chase intervals. Densitometic analysis (data not shown) confirms this visual observation. If gp100 is being processed to yield gp60 and gp49, the latter two should be present in equimolar amounts. The greater accumulation of gp49 could be explained if some was formed by a more direct route. While it is tempting to speculate about an alternate processing route, another explanation could be differential release of gp60 and gp49 from membranes by the detergent buffer used in this laboratory (17), and/or varying degrees of Sue ceptability to protease degradation. In any event, if an alternate route does exist it would have to be a minor pathway, and would not change the obvious existance of gp100 as a processing intermediate. _P_r_‘ocessing in the presence of monensin. Monensin, a monovalent lonoPhore, inhibits transport of glycoproteins specified by vesicular stomatitis virus, Sindbis virus and herpes simplex virus (8,18,19,20). In addition this ionophore blocks both the processing of N-linked 011808 accharides and the addition of O-linked oligosaccharides to herpes simplex virus type 1 glycoproteins (19,20) and gp118 and gp98-62 of VZV (29’ 30) . To further examine the processing of MDHV-B, lysates from in feeted cells labeled in the presence and absence of monensin were Fig. 2. Effect of monensin pretreatment time on inhibition of MDHV-B processing. Immunoprecipitation analysis was done with RaPM and infected (INF) or uninfected control (CON) cell lysates labeled with 358-methionine for 4 hr with (+) or without (-) monensin (0.5 pM). Fluorographic exposure was for 1 day. Molecular sizes are indicated in kilodaltons. 86 9 2.54 ‘— 68k ‘— 43k -1 25.71:— + + Monensin 0.5 1.1M 16 24 hrs pretreatment ~07:- o-Qrfiomorflr . WM— quOO - ... ... - .. —°p60 up... .. ~— —op49 87 Fig. 3. The effect of monensin on MDHV-B processing as determined by pulse-chase analysis. A 15 min pulse with 35S-methionine was followed by chases of various times as shown above each lane. Immuno- precipitation analysis of the cell lysate (1 ml) with (+) and without (-) monensin (0.5 pH) was done with RaPM. Fluorographic exposure was for 8 days. Molecular sizes are indicated in kilodaltons. 88 minutes chase after 15 minute pulse 4 hr 6304560153045 Label .. — — —+++ + +- Monensin —gp100 .11] , , fit . 3... _ - ._ . ..l-N: <. \ fl? - 68k — . ' , . ' ' . .. . I, ; .-. *5? 31 ' —gp49 43k— 1.“ - 89 90 immunoprecipitated with RaPM and analyzed by SDS-PACE. At a previously determined optimal monensin concentration of 0.5 pM (data not shown), and at various times of pretreatment, the amount of gp60 and gp49 were reduced relative to the amount of gp100, and gp60 was replaced by a slightly faster migrating component (Fig. 2); suggesting inhibition of either O-linked glycosylation or addition of terminal sialic acid This preliminary finding suggested that monensin inhibited residues. processing of gp100 to gp60 and gp49. This was further confirmed by a [modified pulse-chase experiment performed in the presence of monensin. .Since only later processing events needed to be analyzed over a critical 15eriod of time the pulse interval was increased from 4 min (Fig. 1) to 15 min to facilitate radioisotope incorporation, and only four chase intervals were employed (Fig. 3) based on information gained in Fig. 1. 'Ihe appropriateness of the conditions is shown in Fig. 3, lanes 1-4, \uhere normal processing (Fig. 1) can be seen in the absence of monensin, ‘41th gp100 first appearing and becoming processed to gp60 and gp49. In t;he presence of monensin (Fig. 3, lanes 5-8), however, gp100 accumulates Eind no gp60 or gp49 can be found. Digestion with Endoglycosidase F and H. Endoglycosidase H (Endo-H) .F>r‘edominantly cleaves between the two proximal N-acetylglucosamine r'esidues of high mannose oligosaccharides (1,22). Complex oligo- E3éa«ccharides are resistant to Endo-H, whereas intermediate oligo- E3€3430harides exhibit intermediate sensitivity. Endoglycosidase F ( E3I'ldo-F) cleaves both high mannose and complex oligasaccharides but GOes not affect O-linked sugars (1,22). To further evaluate Egilmb'czoprotein processing, Endo-F and Endo-H were used to deglycosylate Fig. 4. Deglycosylation of MDHV-B using endoglycosidase F. SDS-PAGE analysis of MDHV-B polypeptides labeled with 3SS-methionine in the presence (+) and absence (-) of TM (2 pg/ml) immunoprecipitated with RaPM, eluted and incubated in the presence (+) and absence (-) of Endo- glycosidase F. Fluorographic exposure was for 1 day. Molecular sizes are indicated in kilodaltons. 91 —9960 -w _, 49 43k - - "1" p?44 Fig. H. Deglycosylation of MDHV—B using endoglycosidase F. SDS-PAGE analysis of MDHV-B polypeptides labeled with 35S-methionine in the presence (+) and absence (-) of TM (2 ug/ml) immunoprecipitated with RaPM, eluted and incubated in the presence (+) and absence (‘) of Endo- glycosidase F. Fluorographic exposure was for 1 day. Molecular sizes are indicated in kilodaltons. 91 NDO’F Fig. 5. Deglycosylation of MDHV-B using endoglycosidase H (Endo-H) SDS-PAGE analysis of MDHV-B polypeptides labeled with 3SS-methionine in the presence (+) and absence (-) of TM (2 ug/ml) or monensin (0.5 uM), immunoprecipitated with RaPM, eluted and incubated in the presence (+) and absence (-) of Endo—H. Fluorographic exposure was for 1 day. Molecular sizes are indicated in kilodaltons. 93 Endo- H Monensin TM "..--4%? 43k - ‘V' pr 257(k- 94 9S immunoprecipitated MDHV-B glycoproteins (Fig. u and 5, respectively). Both Endo-F and H, cleaved gp100 to pr88. Also ng9, and to a lesser extent gp60, were cleaved to pruu by the same enzymes. Effect of Protease Inhibitors. With two putative precursor polypeptides detected in 1119, and only one seen in vitro (preliminary cell-free translation data obtained with polyclonal sera, not shown). there was concern that pruu might be an artifact of pr88 degradation by cellular proteases in the former situation. We examined the effect of four classes of protease inhibitors (2,39) on the appearance of pruu in 1119, synthesized in the presence of tunicamycin. As can be seen in Fig. 6, in all cases the ratio of pruu to pr88 remained very similar, as would be expected if pruu is a precursor polypeptide and not a degradation product of pr88. Cell-Free Translation. Interpretation of data obtained from cell-free translation, and immunoprecipitation analysis of the product is aided by careful choice of antisera and ideally some prior information concerning the putative precursor polypeptides. While the RaB and RcPM sera generated in this laboratory have been useful in characterization of the MDHV-B polypeptides, and they detect the putative precursor polypeptide pruu made in tunicamycin treated cells, they are of low titer (17). Thus even the higher titered of the two sera, RaPM, must be used in quantities such that the amount of IgG heavy chain present causes a distortion of protein migration in that region of the gel, which in turn interferes with normal migration of MDHV-B prUU. Fortunately there have been a panel of monoclonal antibodies prepared Fig. 6. Effect of protease inhibitors on appearance of putative MDHV-B precursors. SDS-PAGE analysis of cell lysates labeled with 3SS-methionine in the presence of TM (2 ug/ml) and harvested in the presence (+) and absence (-) of various protease inhibitors of the metalloprotease (EDTA), serine protease (PMSF, TPCK, TLCK, aprotinin, leupeptin), thiol protease (aprotinin) and acidic protease (pepstatin) classes. Fluorographic exposure was for 1 day. Molecular sizes are indicated in kilodaltons. 96 25.7k —' "i protease _ _ - Inhibitor + + + -+ + + + |—_§§L|fl§—.' metalloprotease |thiol ' acidic 97 98 that neutralize MDHV and immunoprecipitate the MDHV-B polypeptides gp100, gp60 and ng9 (MO). The monoclonal antibody IAN86.17 was of particular interest because the antibody from the original hybridoma, IAN86, was one of the more extensively used and better characterized members of the panel (“0), and because it could immunoprecipitate pruu from talnicamycin-treated cells (“1), implying it should be possible to use its derivative (IAN86.17) to do the same from cell-free translation products. This antibody was first compared directly with RcPM (analysis in the same gel) in respect to their ability to immunoprecipitate the knOWTl MDHV-B polypeptides (Fig. 7A). As would be predicted from the indepnsndent analyses with these two antibodies (17,“0), both immunopre- cipiteated gp100, gp60, and ng9 from MDHV-infected cells labeled in the absewuce of tunicamycin (Fig. 7A, 3rd and Nth lanes). Both RaPM and IAN86.‘V7 also immunoprecipitated pr88 and pruu formed in the presence of tunicmunycin (Fig. 7A, 5th and 6th lanes). As predicted, there is no distorfl:ion of protein migration with IAN86.17. This feature, and its abilit)! to immunoprecipitate both pr88 and pruu, made it the reagent of choice .for identification of the MDHV-B precursor polypeptide made by Gen‘f'r‘ee translation. Chloe work shifted to analysis of the products resulting from cell'fT‘ee translation of MDHV-infected cell mRNA another point required clamfication. The MDHV-A precursor polypeptide is very nearly the size (15.16) of MDHV-B's prAU. In fact, its previous size estimate as pr”? Was t>ased on the size of the ovalbumin marker previously being estimated as ”6 l O 5...?“— *0 tém 25 3:200:08 5.3 Eta mu>Ioz *0 cozooEEmE .U ¢¢ha ID x”? , Illeea miss? 3.3 m->za2ll . 5 [3% N lehmm Lam Eal‘ Ixndm + I < 8 I mo. 5:3 31a 9232 2.5.8 .m ' [.003 al..-- 8. .... looaa ‘. 1'24. o.+ 2| JLLLI 2| LL. Z O U ....z. 323323 92.5.2 5.3 3:02: 33.25 6:23:06 28 63.033 3 3386004 100 101 that confusion could exist if both were present in the same gel due to non-specific trapping. Based on our previous cell-free translation studies (15,16) the MDHV-A precursor will definitely be present in any translation product immunoprecipitated with ICS, which has both MDHV-A and MDHV-B reactivity. Therefore intentional co-immunoprecipitation of the two precursor polypeptides was done with ICS, with and without prior immunoprecipitation by RaA, to differentiate the MDHV-B pruu from the MDHV-A precursor. As can be seen in Fig. 78 (left lane), when ICS was used alone there was a doublet of radioactive protein in the vicinity of the “3 kDa ovalbumin marker. In the case where immunoprecipitation was performed first with RaA (Fig. 78, right lane). the lower band was eliminated, leaving only the upper band to be immunoprecipitated with ICS. Based on RuA's known reactivity with MDHV-A's precursor poly- peptide (15,16) the lower band that was eliminated was that molecule, and its apparent molecular weight position in SDS-PAGE gels is more appropriately presented as “3 kDa now that the ovalbumin marker is reported as M3 kDa. Since RoA has no reactivity with MDHV-B poly- peptides (15,16) the remaining molecule was MDHV-B pruu, and its precise location in relation to the “3 kDa marker was established. Now that the actual differences in the apparent molecular weight of these two percursor polypeptides has been shown in the same gel lane, it is clear that MDHV-B pruu can be identified without confusion. Immunoprecipitation analysis of the product of cell-free translation of MDHV-infected cell mRNA, with the monoclonal antibody IAN86.17, revealed a precursor polypeptide of NH kDa (Fig. 7C, right lane) ivhich co-migrated with pruu seen after immunoprecipitation of lysates; labeled with 35S-methionine in the presence of tunicamycin (Fig. 102 7C, center lane). In contrast, the larger putative precursor molecule, pr88, also detected by immunoprecipitation analysis of these lysates from tunicamycin treated cells (Fig. 7C, center lane), was not seen after cell-free translation and immunoprecipitation analysis (Fig. 7C, right lane). This observation establishes pruu as the primary product of the MDHV-B gene. DISCUSSION Processing of the MDHV-B glycoproteins. The already published detailed study of MDHV-A antigen synthesis and processing (15), which was done with the polyclonal RoA, serves as a precedent for similar analyses involving other MDHV-encoded antigens. In the current study MDHV-B synthesis and processing events were elucidated by using similar high resolution pulse-chase studies, a variety of other methods new to the system and a combination of the polyclonal antibody, RaPM, and the monoclonal antibody, IAN86.17. These two sera were useful because they both were known to immunoprecipitate the MDHV-B glycoproteins gp100, gp60, and ng9, and the MDHV-B putative precursor polypeptides pr88 and prUU. The results obtained with this combination of approaches and reagents is summarized in Fig. 8. While the pulse chase studies were set up in a manner very similar to those used for MDHV-A (15), some results obtained during MDHV-B processing analysis were very different. First, no immunoprecipitable MDHV-B putative percursor polypeptides (either pruu or pr88) are seen (Fig. 1), apparently because glycosylation is cotranslational. In contrast, some MDHV-A precursor polypeptide was detected after a 1 min pulse, despite rapid glycosylation kinetics consistent with co-translational glycosylation (15). The first MDHV-B molecule seen was gp100, and even then it was present in an immunoprecipitable form in only trace amounts in the first three minutes of chase after a A minute Pulse. Labeled precursor molecules must have been present from the start; since MDHV-B intermediates and products appear later, but they apparenutly are not in an immunoprecipitable form at these early times. 103 Fig. 8. MDHV-B processing sequence summary. pruu is the primary gene product. pr88 and gp100 are processing intermediates and gp60 and gpu9 are the final mature forms of MDHV-B. Heavy arrows denote biological pathway. 104 106 Presumably during co-translational glycosylation, the precursor molecules were membrane-bound and were not solubilized efficiently by our detergent buffer; hence the failure to detect the signal, as was found to be the case with HSV-i precursor to gB (pr) (NZ) and with the latent membrane protein, lmp63, of Epstein Barr Virus (EBV) (27). The presumption that MDHV-B precursor polypeptides were undetectable at these early times, was supported by pulse-chase studies performed in the presence of tunicamycin (data not shown), where the putative precursor molecule pr88 is detected at the same time (3 min chase after a N min pulse) that gpIOO appears in the absence of tunicaymcin. The comparison of early processing events of MDHV-A and -B that involved primarily glycosylation warrant further discussion. The heterogeneous MDHV-B molecules slightly smaller than gp100 that were seen first (Fig. 1) were most likely glycosylation intermediates. While some details differ, such diffuse glycosylation intermediates were also seen with MDHV-A (15). MDHV-B is glycosylated to its full-sized gp100 form after a 3 min chase following a N min pulse. Similarly MDHV-A reaches its full sized form after a 6 min chase following a 1 min pulse. Possibly when similarities occur, such as glycosylation kinetics, their existence reflects the involvement of the same cellular (DEF) glycosylation mechanism. On the other hand, when differences occur, such as failure to detect primary precursor polypeptides (see above), fundamental polypeptide differences may be involved, since MDHV-A is predominantly secreted (15), whereas MDHV-B appears to be membrane associated (17). The pulse-chase studies reported here suggest gp1OO is processed to form gp60 and ng9. The gp1OO molecule increases in amount to a maximum 107 at MS min of chase and then decreases. This decrease coincides very closely with the appearance of gp60 and gpu9, in a manner consistent with a precursor-product relationship, thus confirming the earlier observations of Silva and Lee (N1) and Ikuta gt El (13). This interpretation is further substantiated in this study by the observation that monensin inhibited processing of gp100 to gp60 and gpu9 (Fig. 3). Use of this inhibitor leads to a very important conclusion that gp100 is actually an intermediate in the overall biosynthesis of MDHV-A. Since processing is slow, and gp100 is an intermediate, some of the molecule is always present in the cell. This conclusion raises an important new question. Up until now we (17) and others (1M,UO) have referred to the MDHV-B as a complex of the three glycoproteins, designated gp100, gp60, and ng9 (17,“0). This new data suggests that the mature forms of MDHV-B are only gp60 and gpu9 (summarized in Fig. 8), since gp100 is apparently an intermediate. The previous interpretation that it is part of the final complex can be understood when one considers the slow processing kinetics of gp100 and the long (u hour) labeling periods used previously (17). Possibly the profile of three polypeptides seen previously simply reflects the presence of a significant amount of radiolabeled intermediate gp100 under the conditions previously used. On the other hand when chase intervals as long as 20 hrs were used (data not shown) gp100 did not reduce to zero, although less was detected than was found after the u hr chase presented in Fig. 1. Final resolution of this question may require determination of the arrangement of MDHV-B polypeptides in the infected cell nuclear and plasma membranes, a project clearly beyond the scope of this current study. 108 Endoglycosidase digestion of the MDHV-B glycoproteins. The use of endoglycosidase digestion alone and in combination with monensin, was also a new approach to study MDHV-B processing, and also generated additional information that contributes to our overall understanding of this system. The fact that gn100, and not pr88, was sensitive to Endo-H and Endo-F, and that the former yields the latter upon digestion, suggests that pr88 is the immediate precursor of gp100, and thus is an intermediate in the biosynthetic pathway. The fact that gp60 exhibited partial sensitivity to both Endo-F and especially Endo-H indicated that it was undergoing some modifications such as the addition of complex sugars or O-linked glycosylation (8,19,20). At lower concentrations of monensin ($0.5 uM) (data not shown) and at increased times of its pretreatment (Fig. 2), gp60 was replaced by a faster migrating molecule. This molecule (Fig. 2) does not comigrate with any of the Endo-H or Endo-F resistant gp60 species (Fig. A and 5). However gpu9 was completely sensitive to both Endo-F and Endo-H, indicating that only N-linked sugars were added to the gpu9 portion of the molecule before cleavage. Based on the above work, it appears that pr88 undergoes cotranslational addition of N-linked high mannose sugars to yield gp100, which in turn is processed to gp60 and gpu9. Apparently gp60 and gpN9 are differentially glycosylated, with ng9 retaining N-linked high mannose sugars (Bndo-H and Endo-F sensitive) and with gp60 presumably undergoing trimming and possibly addition of O-linked and complex sugars as suggested by the effect of monensin and partial sensitivity to Endo-H and Endo-F. 109 Identification of pruu as the ppimary MDHV-B precursor polypeptide. The role of pruu as the primary precursor polypeptide becomes clear when the full range of data now available is considered together. First, the cell-free translation data (Fig. 7C) clearly indicates that only pruu, and not pr88, is made 1p yippp. This is not due to premature termina- tion of translation because other molecules larger than pr88 are made in our cell-free translation system. Also, the pruu comigrates with the pruu found in cells after tunicamycin treatment (Fig. 7C). Second, studies on the effect of protease inhibitors on pruu and pr88 (Fig. 6) indicate that both molecules are made ip 1119, since the former is not a degradation product of the latter. Third, preliminary partial proteolytic cleavage analysis of prUN and pr88 (data not shown) suggest that the two molecules are related. Fourth, the results obtained by deglycosylation of gp60 and ng9 by Endo-F and H (Fig. u and 5) and the immunoprecipitation of all three molecules with the same monoclonal antibody (U0,UI, Fig. 7A), indicates that pruu is the polypeptide backbone of each glycoprotein. Fifth, the gene encoding MDHV-B has been identified, and analysis by hybrid-selection, cell-free translation and immunoprecipitation (I. Sithole, P.M. Coussens, and L.F. Velicer, Manuscript in preparation) revealed a nu kDa polypeptide that comigrates with pruu seen after tunicamycin treatment. Sixth, the fact that pruu from both the 1p vippp and pp 1119 sources are immunoprecipitated by the same monoclonal antibody (Fig. 7C) indicates that they are identical. Finally Northern blot analysis with a MDHV-B gene probe (I. Sithole, P.M. Coussens, and L.F. Velicer, Manuscript in preparation), detected a transcript of 1.8kb, a size appropriate to encode a polypeptide of an kDa, as is the case for the MDHV-A precursor polypeptide (16). The 110 conclusion based on these data is that pruu is the MDHV-B primary gene product. Thus it is tempting to speculate (Fig. 8) that the initial processing step involves formation of pr88 as a dimer of two pruu molecules. The finding that only pruu, and not pr88, forms in a cell-free translation system, suggests that this hypothesized dimerization may occur on the rough endoplasmic reticulum. The concept of glycoprotein oligomerization is not without precedence in biology. Malek-Hedayat pp 21. (25) have characterized an endogenous lectin from cultured soybean cells which can be found as both a 30 kDa monomer and a 60 kDa dimer. Surprisingly the dimer was stable under conditions such as heating in buffers containing SDS, B-mercapto- ethanol, and urea. The two forms are interconvertible, with the extent of interconversion dependent on the conditions of prolonged incubation. The interconversion establishes that the dimer forms by noncovalent association (25). The fact that the dimer can be found in root extracts is consistent with the idea that it may be more than an artifact that occurs during storage. Glycoprotein B (gB) encoded by herpes simplex virus can be found in both virions or infected cells in the form of detergent stable oligomers (7) that have certain similarity to MDHV-B in respect to the processing data reported here. Specifically, for both HSV gB (7) and MDHV pr88 dimerization occurred within minutes of poly- peptide synthesis, and did not depend on N-linked glycosylation, since they both formed in the presence of tunicamycin. However HSV's gB differs in one important way, it can be dissassociated by heat (7). whereas MDHV-B's pr88 and gp100 can withstand boiling in SDS sample buffer. 111 Despite this difference one is tempted to speculate that dimerization is a step that is needed prior to glycosylation, since MDHV-B's pr88 forms in the presence of tunicamycin. Experiments to resolve questions concerning the detailed nature of the dimerization process, such as the nature of the bonds, the part of the peptide involved, and in the case of MDHV-B, the mechanism of separation of the dimer (i.e. the processing of gp100 to yield gp60 and ng9) are beyond the scope of the current work on MDHV-B. Possibly dimerization of pruu to pr88 and subsequent processing of gp100 occur as the molecules move through different environments within the cell and the MDHV-B dimers resistant to disassociation by heat represents some difference in the properties of the polypeptides of MDHV-B and HSV gB unrelated to some sort of common processing mechanism. Answering such questions will have to be reserved for future studies. In closing this comparative analysis, it must be emphasized that the names of these two herpesvirus polypeptides include the letter designation B for entirely unrelated reasons, not because they are homologues. The comparison of these two herpesvirus polypeptides is therefore in relation to a possible common glycoprotein processing mechanism, not because they are genes conserved between the two herpesviruses. In fact we recently learned (T. O'tal, P.M. Coussens and L.F. Velicer, unpublished data) that the MDHV homologue of gB is located in a different region of the MDHV genome than the gene encoding MDHV-B (I. Sithole, P.M. Coussens, and L.F. Velicer, manuscript in preparation). 112 The possible role of MDHV-B in MDHV membrane biology and immunobiology. Our pulse-chase data (Fig. 1) indicate mature MDHV-B may not be the three polypeptides as we have recently reported (17). The results (summarized in Fig. 8) indicate ng9 and gp60 are the mature fully processed final form of MDHV-B, whereas gp100 is an intermediate that is processed slowly and therefore small amounts are present at all times. Now that we know that gp1OO is really an intermediate, and that a slow processing sequence is involved, we can more accurately plan future experiments to follow the processing of MDHV-B polypeptides into the nuclear and plasma membranes. We predict that gp60 and gpu9 will both be in the membrane as mature forms of MDHV-B and they will appear slowly according to the kinetics seen for their processing. This leads to an interesting question, does gp100 move to the membrane where it is then processed, or does processing occur first, and then do ng9 and gp60 move to the membrane? Based on the monensin data, processing appears to occur in the Golgi apparatus, which would suggest that gp100 is processed to gp60 and gpu9, which then move to the membrane. Alternatively, disruption of the Golgi (19,20) by monensin could prevent transport of gp10O to a site where processing can occur. We do not know if processing is brought about by a cellular or virus-specific protease, or even if proteolysis is involved. Processing of a larger molecule to two smaller molecules has been observed for Rous sarcoma virus (21) and for at least four members of the herpesvirus family, pseudorabies virus (12,2u), bovine herpesvirus (“3), varicella zoster virus (11) and cytomegalovirus (3,5,9). In some cases the processing involves polypeptide cleavage and formation of disulfide linkages (3,5,9,11,12,U3). However gp60 and gpu9 are not linked by disulfide 113 bridges (17). pp ppppp artifactual cleavage has been documented to be cell-line specific for herpes simplex virus glycoproteins (36,“7). Although artifactual cleavage has not been demonstrated in human fibroblasts, gp16O of HCMV is cleaved to gp116 and gp55 (5). In this MDHV study all steps involving cell lysis and immunoprecipitation were carried out at uoc in buffer containing the protease inhibitor PMSF to prevent nonspecific proteolytic cleavage. The evidence that virus neutralization (VN) could be achieved with monoclonal antibody that also immunoprecipitated the three glycoproteins identified as the MDHV-B complex (17), as originally presented by Silva and Lee (“0) and Ikuta pp pl (1“), suggests that some form of MDHV—B is the primary antigen responsible for generating virus neutralizing antibody pp vivo. Ono pp pl (35) reported that glycoproteins purified by affinity chromatography with monoclonal antibody to gB (their terminology for the B antigen complex) elicited partial protection against MD. While the relationship between viral neutralization and immunoprotection remains to be elucidated, it is assumed to be highly significant (35). On the other hand cell-mediated immunity (CMI) may be a factor as well. In either case the probable membrane location of these glyCOproteins becomes important. Determining the relationship of the MDHV-B polypeptides to the cell membranes is part of the planned future experiments on the kinetics of their insertion into the plasma and nuclear membranes. Regardless of the relative importance of VN and CMI, one must ask if all three glycoproteins are involved in protective immunity or if just a single one is involved? Since they all seem to contain the same pruu polypeptide backbone, either individually (gp60 and ng9) or in a dimer form (gp100) they should all have the same 114 epitopes that are determined by the polypeptide backbone. Possibly their role will be determined more by their location in respect to the surface of relevant membranes. The significance of the ideas presented above become more apparent when one takes into consideration the observation (L.F. Lee, data not shown) that the monoclonal antibody used in this study, IAN86.17, reacts well with the surface of MDHV-infected cells in a membrane immunofluoresence assay. Clearly the epitope recognized by this antibody is exposed on the cell surface. However, presenting a more complete answer to these questions concerning possible membrane locations and possible epitopes involved in protective immunity are beyond the scope of this paper. Summary and significance. In summary (Fig. 8) the data reported in this study; (1) further confirms the existence of, and more clearly defines the nature of, the MDHV-B glycoprotein complex, (2) determines the nature of the MDHV-B precursor polypeptide, and (3) demonstrates the MDHV-B processing sequence more rigorously than has been previously reported. Key observations here are that gp100 is really a processing intermediate not a precursor, and its processing yields gp60 and gpu9. The final process, membrane insertion, has not been elucidated, and will be the basis of future membrane biology studies. However, the primary sequence of events has been determined, which will facilitate planning future studies. The data presented is also significant in that it contributes to our understanding of the molecular mechanism of synthesis and processing of the MDHV-B precursor polypeptide. This work not only establishes pruu as the primary gene product, it provides strong evidence that a dimerization step may occur before the glycosylation and 115 other processing events occur. The nature and function of this dimerization process will also be the basis of future studies. Knowing the nature of the primary gene product will facilitate other studies, such as gene identification by hybrid selection and cell-free translation analysis, and epitope mapping, which are in progress or planned, respectively. These new studies, will contribute greatly to our understanding of the role of MDHV-B in the immunobiology of the MDHV system, just as predicted with MDHV-A (15). LITERATURE CITED Balachandran, N., and L.M. Hutt-Fletcher. 1985. Synthesis and processing of glycoprotein gG of Herpes simplex virus type 2. J. Virol. 5N:825-832. Bartles, J.E., L.T. Braiterman, and A.L. Hubbard. 1985. Biochemical characterization of domain-specific glycoproteins of the rat hepatocyte plasma membrane. 260:12792-12807. Benko, B.M., and W. Gibson. 1986. Primate cytomegalovirus glycoproteins: Lectin-binding properties and sensitivities to glycosidases. J. Virol. 59:703-713. Bonner, W.M., and R.A. Laskey. 197“. A film detection method for triton labeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. “6:83-88. Britt, W.J., and D. Auger. 1986. Synthesis and processing of the envelope gp55-116 complex of human cytomegalovirus. J. Virol. 58:185-191. Chirgwin, J.M., A.G. Przybyla, R.J. MacDonald, and W.J. Rutter. 1979. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:529A-5299. Claesson-Welsh, L., and P.G. Spear. 1986. Oligomerization of Herpes simplex virus glycoprotein B. J. Virol. 60:803-806. Dall'olio, F., N. Malagolini, V. Speziali, G. Campadelli-Fiume, and F. Serafina-Cessi. 1985. Sialylated oligosaccharides O-glycosidically linked to glycoprotein C from Herpes simplex virus type 1. J. Virol. 56:127-13U. 10. 11. 12. 13. 1“. 15. 16. 117 Farrar, G.H., and P.J. Greenaway. 1986. Characterization of glycoprotein complexes present in human cytomegalovirus envelopes. J. Gen. Virol. 67:1U69-1U73. Glaubiger, C., K. Nazerian, and L.F. Velicer. 1983. Marek's disease herpesviruses. IV. Molecular characterization of Marek's disease herpes virus A antigen. J. Virol. “5:1228-123u. Grose, C., D.P. Edwards, K.A. Weigle, W.E. Friedrichs, and W.L. McGuire. 198M. Varicella-zoster virus-specific gpluo: A highly immunogenic and disulfide-linked structural glycoprotein. Virology 132:138-1u6. Hampl, P., T. Ben-Porat, L. Ehrlicher, K.-O. Habermehl, and A.S. Kaplan. 198“. Characterization of the envelope proteins of pseudorabies virus. J. Virol. 52:583-590. Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 198A. Processing of glycoprotein gB related to neutralization of Marek's disease virus and herpesvirus of turkeys. Microbiol. Immunol. 28(8):923-933. Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 198M. Identification with monoclonal antibodies of glycoproteins of Marek's disease virus and herpesvirus of turkeys related to virus neutralization. J. Virol. U9z101u-1017. Isfort, R.J., R.A. Stringer, H.-J. Kung, and L.F. Velicer. 1986. Synthesis, processing, and secretion of the Marek's disease herpesvirus A antigen glycoprotein. J. Virol. 57:“6u-M7H. Isfort, R.J., H.-J. Kung, and L.F. Velicer. 1987. Identification of the gene encoding Marek's disease herpesvirus A antigen. J. Virol. 61:261U-2620. 17. 18. 19. 20. 21. 22. 23. 2“. 118 Isfort, R.J., I. Sithole, H.-J. Kung, and L.F. Velicer. 1986. Molecular characterization of Marek's disease herpesvirus B antigen. J. Virol. 59:“11-U19. Johnson, D.C., and M.J. Schlesinger. 1980. Vesicular stomatitis virus and Sindbis virus glycoprotein transport to the cell surface is inhibited by ionophores. Virology 103:u07-N2N. Johnson, D.C., and P.G. Spear. 1983. O-linked oligosaccharides are acquired by Herpes simplex virus glycoproteins in the Golgi apparatus. Cell 32:987-997. Johnson, D.C., and P.G. Spear. 1982. Monensin inhibits the processing of Herpes simplex virus glycoproteins, their transport to the cell surface, and the egress of virions from infected cells. J. Virol. “3:1102-1112. Klemenz, R., and H. Diggelmann. 1979. Extracellular cleavage of the glyCOprotein precursor of Rous sarcoma virus. J. Virol. 29:285-292. Kornfeld, R., and S. Kornfeld. 1985. Assembly of asparagine-linked oligosaccharides. Ann. Rev. Biochem. 5U:631-66U. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage Tu. Nature (London) 227:680-685. Lukacs, N., H.-J. Thiel, T.C. Mettenleiter, and H.-J. Rziha. 1985. Demonstration of three major species of pseudorabies virus glycoproteins and identification of a disulfide-linked glycoprotein complex. J. Virol. 53:166-173. 25. 26. 27. 28. 29. 30. 31. 32. 33. 119 Malek-Hedayat, S.S.A. Meiners, T.N., Metcalf, III, M. Schindler, J.L. Wang, and S.-C. Ho. 1987. Endogenous lectin from cultured soybean cells. J. Biol. Chem. 262:7825-7830. Maniatis, T., E.F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Mann, K.P., and D. Thorley-Lawson. 1987. Posttranslational processing of the Epstein-Barr virus-encoded p63/LMP protein. J. Virol. 61:2100-2108. Marek, J. 1907. Multiple Nerventzuendung (Polyneuritis) bei Huchnern. Dtsch. Tieraerztl. Wochenschr. 15:”17-"21. Montalvo, E.A., and C. Grose. 1986. Neutralization epitope of varicella zoster virus on native viral glycoproteins gp118 (VZV glycoprotein gp118). Virology 1u9:23o-2u1. Montalvo, E.A., R.T. Parmley, and C. Grose. 1985. Structural analysis of the varicella-zoster virus gp98-gp62 complex: Posttranslational addition of N-linked and O-linked oligosaccharide moieties. J. Virol. 53:761-770. Murthy, K.K., and B.W. Calnek. 1979. Pathogenesis of Marek‘s disease: Effect of immunization with inactivated viral and tumor-associated antigens. Infect. and Immun. 26:5“7-553. Nazerian, K. 1980. Marek's disease: a herpesvirus-induced malignant lymphoma of the chicken, p.665-682. In G. Klein (ed.), Viral oncology. Raven Press, New York. Nonoyama, M. 1982. The molecular biology of Marek's disease herpesvirus, 333-3U6. In B. Roizman (ed.), The herpesviruses, vol. 1, Publishing Corp., New York. 3U. 35. 36. 37. 38. 39. ”0. H1, 120 Okazaki, W., H.G. Purchase, and B.R. Burmester. 1970. Protection against Marek's disease by vaccination with a herpesvirus of turkeys (HVT). Avian Dis. iu:ui3-u29. Ono, K., M. Takashima, T. Ishikawa, M. Hayashi, I. Yoshida, T. Konobe, K. Ikuta, K. Nakajima, S. Ueda, S. Kato, and K. Hirai. 198A. Partial protection against Marek's disease in chickens immunized with glycoproteins gB purified from turkey-herpesvirus- infected cells by affinity chromatography coupled with monoclonal antibodies. Avian. Dis. 29:533-539. Pereira, L., D. Dondero, and B. Roizman. 1982. Herpes simplex virus glycoprotein gA/B: Evidence that the infected vero cell products comap and arise by proteolysis. J. Virol. uu:88-97. Payne, L.N. 1982. Biology of Marek's disease virus and the herpesvirus of turkeys, p.3u7-u31. In B. Roizman (ed.), The herpesviruses, chap. 8, vol. 1. Plenum Publishing Corp., New York. Sharma, J.M., and H.A. Stone. 1972. Genetic resistance to Marek's disease. Delineation of the response of genetically resistant chickens to Marek's disease virus infection. Avian Dis. 16:89U-906. Shin, J., and T.H. Ji. 1985. Composition of cross-linked 125l- follitropin-receptor complexes. J. Biol. Chem. 260:12822-12827. Silva, R.F., and L.F. Lee. 1984. Monoclonal antibody-mediated immunoprecipitation of proteins from cells infected with Marek's disease virus or turkey herpesvirus. Virology 136:307-320. Silva, R.F., and L.F. Lee. 1985. Isolation and partial characterization of three glycoproteins common to Marek's disease virus and turkey herpesvirus-infected cells, p.101-110. In B.W. 112 112 1111. 4'5. I (7‘ 0 U7. U2. U2. uu. U5. U6. “7. 121 Calnek and J.L. Spencer (ed.), Proceedings of the International Symposium on Marek's Disease. American Association of Avian Pathologists, Inc., Kennett Square, PA. Spear, P.G. 1976. Membrane proteins specified by Herpes simplex viruses. I. Identification of four glycoprotein precursors and their products in type 1-infected cells. J. Virol. 17:991-1008. Van Drunen Littel-Van Den Hurk, S., J.V. Van Den Hurk, J.E. Gilchrist, V. Misra, and L.A. Babiuk. 198U. Interactions of monoclonal antibodies and bovine herpesvirus type 1 (BVH-1) glycoproteins: Characterization of their biochemical and immunological properties. Virology 135:U66-U79. Van Zaane, D., J.M.A. Brinkhof, F. Westenbrink, and A.L.J. Gielkens. 1982. Molecular-biological characterization of Marek's disease virus. I. Identification of Virus-specific Polypeptides in Infected Cells. Virology 121:116-132. Van Zaane, D., J.M.A. Brinkhof, and A.L.J. Gielkens. 1982. Molecular-biological characterization of Marek's disease virus. II. Differentiation of Various MDV and HVT Strains. Virology. 121:133-1U6. Weber, K., and M. Osborn. 1969. The reliability of molecular weight determination by dodecyl-sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 2UU:UUOS—UU12. Zezulak, K.M., and P.G. Spear. 198U. Limited proteolysis of herpes simplex virus glycoproteins that occurs during their extraction from vero cells. J. Virol. 50:258-262. CHAPTER IV Identification and Localization of the Gene Encoding Marek's Disease Herpesvirus B Antigen Sithole, I., P.M. Coussens, and L.F. Velicer. Manuscript in preparation for submission to J. Virol. 122 Marek couple monocl ABSTRACT The gene which encodes for the precursor polypeptide prUU of the Marek's disease herpesvirus (MDHV) B antigen (MDHV-B) glycoprotein complex was physically mapped to the viral genome. The MDHV-B specific monoclonal antibody 1AN86.17, which had previously been shown to specifically imrunoprecipitate the MDHV-B prUU, was used to screen a total Agt11 MDHV genomic expression library. The positive Agtll clones were used to probe Southern blots of the BamHI plasmid clones of the viral genome. The gene was localized to both the 12.0 and 5.5 kbp BamHI D and H fragments, respectively, apparently in the portion of each that map in the unique long inverted repeats. Hybrid selection with these two DNA fragments, and cell-free translation of the selected RNA revealed a precursor polypeptide of UU kilodaltons (kDa) immunoprecipi- table with antisera reactive with MDHV-B, confirming this localization. Subsequent analysis was performed with the MDHV-B gene on the H fragment only. Northern blot analysis revealed a 1.8 kb transcript, a size consistent with encoding a UU kDa polypeptide. Transcription of the mRNA proceeds in the leftward direction on the H fragment. The 5' end of the gene maps approximately 550 bp from the unique SmaI site located 5' to the gene. The 3' end maps approximately UUO bp from the unique HindIII site located downstream of the termination codon. This more precise localization has made it possible to proceed with nucleotide sequence analysis. This is the second report of an MDHV gene being mapped to the MDHV viral genome. This opens the way for the use of recombinant DNA technology to study the nature of a gene encoding a 123 viru immui 124 virus-specific membrane glycoprotein already shown to be involved in the immunoprevention in this oncogenic herpesvirus system. POlYPe (MDHV- common this lg whose s charact identif submitt (HVT~A; Th: lilentlf: based or tation 1.: (D lcr al \0 (F glycopPOf or H‘liny INTRODUCTION Marek's disease (MD), an oncogenic disease of the chicken, is caused by the Marek's disease herpesvirus (MDHV), which results in T-cell lymphomas and peripheral nerve demyelination (12,17,18), MD can be prevented by immunization with the apathogenic herpesvirus of turkeys HVT (15). MDHV and HVT are immunologically related, sharing a large number of polypeptides (6,22,25,26) including the most prominent A antigens (MDHV-A and HVT-A) and B antigens (MDHV-B and HVT-B) (1U,19). These common antigens have been the focus of considerable molecular studies in this laboratory in recent years. MDHV-A is a secretory glycoprotein whose synthesis, processing and secretion (10) has been extensively characterized. The gene encoding MDHV gp57-65 (MDHV-A) has been identified (8) and sequenced (P.M. Coussens and L.F. Velicer, Manuscript submitted), and a similar analysis is being completed for HVT gp57-65 (HVT-A) (P.M. Coussens and L.F. Velicer, Manuscript in preparation). The immunologically related MDHV-B and HVT-B antigens were identified as a set of three glycoproteins designated gp100, gp60, ng9 based on an apparent molecular weight on SDS-PAGE and immunoprecipi- tation with polyclonal sera containing anti-MDHV-B activity (9). Ikuta pp pp. (6) and Silva and Lee (23) have reported finding a similar set of glycoproteins with their monoclonal antibodies. Tunicamycin inhibition of N-linked glycosylation resulted in the appearance of two putative precursor polypeptides pr88 and prUU immunoprecipitable with the same polyclonal sera (9). All indications point to these three glycoproteins defined as MDHV-B making up the antigen gB that was reported to elicit carti that tides glyco (Sith transl PolycL in pre: 00-mjg} infect, tunica; initial dirtPM: p"Rees: pPOCeSE appear a DPOQE °f MDHi defiied disQOVE 126 partial protection against MD (16). This discovery makes it imperative that MDHV-B be studied indepth in terms of the unique set of polypep- tides that exists and the gene encoding the primary gene product. Our most recent pulse-chase studies revealed that gp100 is a glycosylated intermediate which is processed to yield gp60 and ng9 (Sithole, Lee and Velicer, manuscript in preparation), thus confirming similar findings by Silva and Lee (23) and Ikuta pp pl. (7). Treatment of gp100 with endoglycosidases reduces it to pr88 indicating that the latter is also an intermediate in the biosynthetic pathway. With the same enzymes ng9, and to a lesser extent gp60, are reduced to prUU, suggesting the latter prUU is their polypeptide backbone. Cell-free translation of infected cell RNA followed by immunoprecipitation with polyclonal and monoclonal antibody (Sithole, Lee and Velicer manuscript in preparation) revealed a precursor polypeptide of UU kDa which co-migrates with prUU seen after immunoprecipitation of lysates of infected cells labeled with 3SS-methionine in the presence of tunicamycin. Collectively, these data; (1) showed that prUU is the initial gene product, (2) are consistent with the concept that prUU dimerizes to form pr88; and (3) then demonstrate that pr88 is a processing intermediate which is glycosylated to gp100, another processing intermediate which is then processed to gp60 and ng9, which appear to be final mature forms of MDHV-B. The discovery that gp100 is a processing intermediate, and that gp60 and ng9 are the mature forms of MDHV-B, suggest that the MDHV-B complex might be more appropriately defined as a two, rather than a three, glycoprotein complex. The discovery that prUU is MDHV-B's precursor polypeptide is important in terms of plans to identify the gene encoding MDHV-B. apprc local 127 The study reported here describes the construction and immunoscreening of a lgt11 genomic expression library, and its use, along with hybrid selection and cell-free translation, to identify the gene encoding MDHV-B prUU. In addition we report the results of an appropriate combination of molecular biology approaches to further localize and characterize the gene. MATERIALS AND METHODS Cells and Viruses. The preparation, propagation, and infection of small scale duck embryo fibroblast (DEF) cell cultures with MDHV was generally performed as first reported by Glaubiger pp pl (5), and more recently by Isfort pp pp (9). One exception was that the MHDV strain GA used was at passage level 6, following isolation of cell-free virus from feathers tips obtained from infected birds (21). Restriction enzymes, DNA prpparation, gel electrophoresis and Southern blotting. All restriction and DNA modifying enzymes were obtained from either Bethesda Research Laboratories (BRL, Gaithersburg, MD) or Boehringer Mannheim Biochemicals (BMB, Indianapolis, IN), and used according to the manufacturer's instructions. Preparation of genomic, plasmid and phage DNA was by standard procedures (11). The MDHV BamHI library (U) was a gift from M. Nonoyama, Showa University Research Institute. Two ug of plasmid DNA from each of the 27 BamHI clones were digested with BamHI to free the inserts, and the digests were analyzed on 0.6% agarose gel. The gel was stained with 0.5 ug/ml ethidium bromide solution for U5 mins, rinsed in water and photographed under ultraviolet light. DNA in the gel was denatured, neutralized, and then transferred to nitrocellulose by the method of Southern (2U). The filters were probed with (a-32P)dCTP (Amersham Corp. >UOO Ci/mmol) nick-translated probes from the positive lgt11 clones (see below). 128 129 Construction and screening of the Agt11 genomic expression library. The library was constructed by a modification of the procedure described by Young and Davis (28,29). Total genomic DNA from MDHV-infected DEF cells was partially digested with HaeIII to yield fragments in the range O-7.2kb, the maximum size that the vector can accept. The inserts were ligated to phosphorylated EcoRI linkers using TU DNA ligase, and digested with an excess of EcoRI. The mixture was passed over a Sephadex G-50 column, then run in an 0.8% agarose gel to purify the inserts free of excess linkers. The inserts were cloned into the unique EcoRI site of the dephosphorylated Agt11 arms, and packaged pp plppp into phage heads (Promega Biotech). The resulting phage were used to infect Y109O r'm+ cells. A library was obtained from 1 pg of DNA which contained 1 x 106 independent recombinants. Induction was with 10 mM isopropyl-thio-B-D-galactopyranoside (IPTG) (Sigma Chemicals) in conjunction with a nitrocellulose overlay of the plaques as specified by the manufacturer. The filters were screened with monoclonal antibody 1AN86.17, known to be reactive with MDHV-B polypeptides (Sithole, Lee and Velicer, Manuscript in preparation). Isolation of total cellular RNA. Total cellular RNA was extracted from MDHV-GA infected DEF cells at U8 hrs postinfection using the guanidinium isothiocyanate procedure described by Chirgwin pp pp. (2), and modified by Maniatis pp pl. (11), and already in use in this laboratory (Sithole, Lee and Velicer, Manuscript in preparation). Northern blotting. Twenty pg of control and infected cell RNA were run in 1.2% formaldehyde denaturing agarose gels as described by Maniatis (f Im the Ba the go hibrid hibric Select 130 pp pl. (11). Electrophoresis was performed at 60 volts for 7 hrs. The gel was stained with ethidium bromide (2 ug/ml) for 30 mins, destained in running buffer (11) and photographed under ultraviolet light. RNA was transferred to nitrocellulose filters by the method of Southern (2U). The filter was baked at 80°C in vacuo for 2.5 hrs. Hybridizations to detect viral RNA with DNA probes 32P-dCTP labeled by nick translation (20), were performed at U2°C in 50% formamide, 10% dextran sulfate, 5X Denhardt's solution and 100 ug/ml salmon sperm DNA. Following hybridization filters were washed in 2 x SSC, 0.1% SDS at room temperature for 20 mins, then 1 x SSC, 0.1% SDS at 65°C four times, 15 mins each. Molecular weight markers were an RNA ladder obtained from Bethesda Research Laboratories. Hybrid selection and cell-free translation. Ten ug of DNA from each of the BamHI D and H fragments, the EcoRI-c subfragment of the latter, and the positive AgtiiHRU clone were bound to nitrocellulose filter and hybridized to 100 pg MDHV-infected DEF cell RNA in a 100 pl reaction and hybridized at 50°C as described by Maniatis pp pl. (11). The hybrid selected RNA was eluted and translated in vitro in a 30 ul reaction volume, using 50 uCi 3SS-methionine (Amersham Corp., 1000 Ci/mmol) and a rabbit reticulocyte lysate system used according to manufacturer's specifications (BRL). All reactions were performed in the presence of 1U/ul RNasin to inhibit RNase activity (Promega Biotech). The reaction was stopped using an equal volume (30 ul) of detergent buffer (9). Ten ul was analyzed directly by SDS-PAGE (see below), and the rest was immunoprecipitated with a mouse ascites fluid stock solution (previously diluted 1:50) at 1/20 volume (i.e. 50 ul of translation mixture with 131 2.5 ul of the stock solution of the monoclonal antibody 1AN86.17). With this mouse ascites fluid, one ul of a second antibody, rabbit anti-mouse IgG (Sigma), was used along with the mouse ascites fluid stock solution at a ratio of 1:25 to ensure successful immunoprecipitation (see below). Immunoprecipitation and SDS-PAGE analysis. The general immunoprecipi- tation and SDS-PAGE methods previously described (9) were used for analysis of lysates of tunicamycin-treated cells with the monoclonal antibody IAN86.17 (Sithole, Lee and Velicer, Manuscript in preparation) and cell-free translation products prepared as described above. S1 Nuclease mapping. To map the 5' terminus 100 ug of total cellular RNA was hybridized to an EcoRI-SmaI fragment (Fig. 5, probe III) labeled at the dephosphorylated 5' end using 32P-Y-ATP (ICN >7,000 Ci/mmole) and polynucleotide kinase. as described by Maniatis, pp pp. (11). To map the 3' terminus, 100 ug of total cellular RNA was hybridized to the EcoRI-HindIII fragment (Fig. 5, probe I) as described by Maniatis pp pl. (11). Hybridizations were carried out at 50°C for 3 hr, and digestions with nuclease Si (1 U/ul) were at 37°C for 15 min (11). Protected DNA fragments were analyzed on 5% denaturing polyacrylamide gels (11). pBR322 digested with Hian, and end-labeled with 32P-a-dATP (>uoo Ci/mmol Amersham Corp.) using Klenow fragment, was used as size markers (11). ’- FE RESULTS Preliminary identification and localization of the MDHV-B gene py_hybrid selection and cell-free translation. Initial attempts at MDHV-B gene identification were with our polyclonal RaPM antisera, which had previously been shown to immunoprecipitate the two putative precursor polypeptides, prUU and pr88, formed in the presence of tunicamycin, and our established hybrid selection and cell-free translation procedures. These methods, along with polyclonal RaA, allowed successful identifica- tion of the gene encoding MDHV-A (gp57-65) in this system (8). The rationale was to proceed with what had worked well previously. Preliminary data (not shown) obtained with the above method using the BamHI MDHV library, resulted in two clone pools containing DNA capable of selecting mRNA that could be translated to yield a precursor polypeptide of approximately UU kDa immunoprecipitable with RaPM. Pool rearrangement narrowed this to the BamHI D and H fragments. The drawback with interpreting the results obtained here was the severe distortion of bands in the AU kDa region as a result of protein overload by IgG heavy chain which migrates in this region of the gel. While we were convinced that we had detected prUU as the MDHV-B precursor, and thus had localized the MDHV-B gene to the BamHI D and H fragments, the distortion problem interfered with migration of the polypeptide at exactly the expected size position on the gel (Sithole, Lee, and Velicer, Manuscript in preparation). Therefore, for a variety of reasons, we sought to obtain supporting data by alternate methods, and to localize the gene even further. 132 133 In an attempt to localize the gene even further, both the D and H fragments were partially digested with the restriction enzyme Sau3A and cloned into the unique BamHI site of the expression vector pUCl9 (27). The colonies were immunoscreened with RaPM and probed with 125I-protein A. This technique was cumbersome and we had very little success with it. We then constructed a Agtll library (25,26, described below) which facilitates rapid and efficient screening of a large number of recombinant clones. Since we again had very little success screening with RaPM, and since ICS was a very complex antiserum, we opted to screen the Agtil library with RaB. This antiserum also failed to detect any recombinant clones expressing MDHV-B precursor polypeptides, thus confirming our earlier results, that none of our polyclonal antisera detected MDHV-B on Western blots (data not shown). During this time, as part of the study to determine the MDHV-B precursor processing relationships, the true precursor of MDHV-B was found to be prUU using IAN86.17, the newly cloned derivative (Sithole, Lee, and Velicer, Manuscript in preparation) of the monoclonal antibody IAN86 (22). Thus, all the gene identification work that follows was done with unequivocal knowledge that prUU was the MDHV-B primary gene product, and with the prospects of trying yet another antisera with anti-B activity and the ability to immunoprecipitate prUU (Sithole, Lee, and Velicer, Manuscript in preparation). Construction and screening of the Agt11 genomic expression library. With the information and new reagent available, as described above, the Agt11 method was repeated because of its potential for rapidity and efficiency. MDHV genomic DNA was partially digested with the 134 restriction enzyme HaeIII. The fragments were ligated to EcoRI linkers, and cloned into the unique EcoRI site of lgt11, and expressed as B- galactosidase fusion protein (28,29). A library was obtained from 1 pg of DNA, which contained 1 x 106 independent recombinants. A total of U0,000 plaques were screened with the monoclonal antibody 1AN86.17 and three positive clones were identified (data not shown) and designated Agt11HRU, lgtiiHRS, lgt11HR6. Virus in each clone was plaque purified and propagated to produce DNA for probe preparation. Mapping the location of the gene by Southern blot analysis. DNA from the three progeny viruses was nick-translated and used to probe Southern blots of the 27 BamHI plasmid clones (Fig. 1, top). In each case the DNA hybridized to the BamHI D and H fragments, only the data obtained with lgt11HRU is shown (Fig. 2). The additional hybridization of probe, as seen in seven other lanes between J and T, is the cosmid vector pHC79. Note that this occurs only in lanes where the cosmid is present; and in every case it is at a size position (of the cosmid) larger than the MDHV DNA clone present (J to T). BamHI D and H are located in the unique long terminal repeats, suggesting that the gene is present in two copies. The BamHI D and H fragments were next digested with EcoRI and SmaI (Fig. 1, bottom) based on the detailed restriction enzyme mapping of these two fragments by Fukuchi pp pp. (U), and probed with the same Agt11HRU clone by Southern blot analysis. Subfragments EcoRI a, and SmaI a from D and EcoRI c and SmaI a from H were positive (Fig. 1 and 3, bottom). Gene localization by hybrid selection and cell-free translation. To more rigorously ascertain the identity of the gene, BamHI fragments D, Fig. 1 Physical Map of MDHV DNA. The top part is a BamHI restriction enzyme map of MDHV DNA. The bottom part is the detailed restriction enzyme maps of BamHI D and H. Fukuchi pp pl. (U). 135 I .233 m a a... w w m a 22%... w m m 8.3.. 01. m. m. w Tu m. 1 , 6.2330 7 o _~ _ . _ . w. T _ . 1 v o o a o L z. 1__ . H «E S a... ...: ‘1 J {1111 ‘ 3.19 1‘ U 1‘ . ow. co. 3131948,... 3: on. 3. on 3 i av i ow , lo .3. 08.3 .t < .....3. x. z a a a u new}... 53W.- ..6» .3. u ...: u u .6 o «:4 .3. 136 Fig. 2 Localization of the MDHV-B gene on the viral genome by Southern blot analysis. BamHI-digested MDHV plasmid clones were analyzed on a 0.6% agarose gel, and probed with DNA from a positive AgtiiHRU clone. DNA isolation restriction enzyme digestion, agarose gel electrophoresis, Southern blotting, nick translation and hybridization were performed as described in the text. Exposure was for 16 hr. 137 3353ngan 2.2.3 Amocozzmaéiov mmmmma Armamoomfi $553 2: 3:20 2583 .1 58 >10? ad I Qm my. 3“ I .1 rec ¢.mH a 10d. ‘0 - .0 04mm anus! lllllllllllllllllllllllll as: .m “imam onmwmfozssnx «xi arm... xoumooofiw 138 Fig. 3 Southern blot analysis to detect DNA subfragments carrying sequences encoding MDHV-B. MDHV BamHI D and H fragments were digested with EcoRI (E) and SmaI (S), or undigested (-) and analyzed on a 0.8% agarose gel, Southern blotted and probed with DNA from a lgtllHRU clone (Fig. 2), as described in Fig. 2 and the text. Exposure was for 18 hr for the left lane and U5 min. for the right five lanes. 139 Barn HI Fragments L D i H I l n |-|E'S|_|EIS| Size ‘ (kb) 23.5 1 ...... E 6.6 "" - 4.4 ___ 2.3 2.0 “—0.5 140 141 and H, EcoRI c subfragment of H, and DNA from the immunoreactive Agt11HRU clone were used to perform hybrid selection and the selected RNA was translated in a rabbit reticulocyte lysate system. The translation mixture was analyzed by immunoprecipitation with the monoclonal antibody 1AN86.17 (Fig. U), which has already been shown to immunoprecipitate the MDHV-B prUU made by cell-free translation (Sithole, Lee and Velicer, Manuscript in preparation). In each of the four lanes identified by the DNA used for hybrid selection, a single polypeptide of UU kDa was detected that was not present in the minus DNA lane (Fig. U). This polypeptide co-migrates with the UU kDa polypeptide seen after immunoprecipitation analysis of the product formed when unselected infected cell RNA is translated pp ppppp (data not shown). In both cases (Fig. U, and Sithole, Lee and Velicer, Manuscript in preparation), these two polypeptides synthesized by cell-free translation co-migrate with prUU seen after immunoprecipitation of lysates labeled with 35S-methionine in the presence of tunicamycin, using the monoclonal antibody 1AN86.17. The conclusions are that the UU kDa polypeptide is MDHV-B's prUU, and that at least part of its gene is in each form of DNA used (Fig. U). Identification of the transcript of the MDHV gene py_Northern blot analysis. Northern blot analysis of infected and uninfected cell RNA was performed by using the purified 950 bp EcoRI c subfragment of BamHI H as a nick-translated probe. A single abundant 1.8kb transcript was detected in the infected cell RNA lane (Fig. 5). Upon prolonged exposure (20-fold longer). this probe recognized a large transcript of 9.5kb (data not shown). The significance of this larger sized but Fig. U Hybrid selection of MDHV-B-specific mRNA, cell-free translation and immunoprecipitation analysis. Ten pg of DNA from BamHI D and H, the EcoRI c subfragment of BamHI H, and Agt11HRU was bound to separate nitro- cellulose filters and hybridized to 100 pg total cellular RNA in a 100 pl reaction volume, at 50°C for 3 hr. The eluted RNA was cell-free translated and immunoprecipitated with the monoclonal antibody IAN86.17. Lysates (LYS*) prepared from cells labeled with 35S-methionine in the presence of tunicamycin (TM) 2 pg/ml, were also subjected to immunopre- cipitation with IAN86.17 as described in the text. SDS-PAGE analysis was as described in the text, with 10% acrylamide. Fluorographic exposure was for 8 days for the left lane and 6U days for right five lanes. 142 <1‘ I C) (E :T:' E 3 < E E = E mien-3’1 . l —200k —92.5k -68k 2» ——43k —25.7k '— 18.4k — l4.3k 143 Fig. 5 Detection of MDHV-B mRNA by Northern blot analysis. 20 pg of control (CON) and infected (INF) total cellular RNA were loaded per lane. The blot was probed with the EcoRI c subfragment of the BamHI H fragment as described in the text. Exposure was for 1 hr. 144 Inf Con R Size (kb) — 9.5 - 7.5 - 4.4 - 2.2 “L4 143'. 146 low-abundance transcript is not known at this moment. However Deatly pp pl. (3) have reported finding a 9.5kb transcript in pseudorabies virus (PRV) infected cells. This PRV RNA has been shown not to encode any protein. These transcripts were observed as being retained in the cell nucleus, and function neither as messenger RNA or messenger RNA DPGCUPSOPS . S1 nuclease mapping of the 5' and 3' termini of the MDHV-B gene. Preliminary nuclease S1 mapping studies with probe II (Fig. 6, map) resulted in the entire fragment being protected (data not shown). This result also revealed the direction of transcription to be leftward (data not shown). Having determined that all of probe II is protected from nuclease Si digestion, the entire EcoRI c subfragment was used as probe. Again the entire subfragment was protected, from nuclease S1 digestion, suggesting the gene includes the entire EcoRI c subfragment (data not shown). These results also suggested that the gene spans the two EcoRI sites (Fig. 6, map) and, if no splicing occurred, that the 5' and 3' termini should map in the EcoRI-Smal and EcoRI-HindIII fragments, respectively, (Fig. 6, map). Following denaturation, hybridization to RNA, and S1 nuclease digestion, a protected fragment of 3' labeled probe I (~U60bp in size) was detected (Fig. 68, right lane), determining the position of the arrow above probe I (Fig. 6, map). In the right lane the upper band is reannealed probe, appearing to be slightly out of position due to a distortion in the gel. When probe III was 5' labeled and hybridized to RNA and digested as described above, two strongly hybridizing fragments (~700bp and 3Upr) and one minor species (~UOObp) were detected (Fig. 6A, left lane), determining the positions of the three arrows above Fig. 6 Localization of the MDHV-B mRNA termini by S1 nuclease protection. The orientation and limits of the gene were determined by using probes I, II and III. Probes I and II were 3' labeled with (a-32P)dATP using the Klenow fragment of p. all DNA polymerase I. Probe III was dephosphorylated with calf intestinal alkaline phosphatase and 5' labeled with (7-32P)ATP and polynucleotide kinase. In both panels A and B, DNA was hybridized at 50°C for 3 hr to control (CON) or infected (INF) total cellular RNA, digested (+) or not digested (-) with S1 nuclease (1U/pl). Panel A involves analysis of the 5' terminus through use of 5' labeled probe III. Panel B involves analysis of the 3' teminus through use of 3' labeled probe I. pBR322 digested with Hian and end labeled with (a-32P)dATP using the Klenow fragment of E. pplp DNA polymerase I, were used as size standards. Electrophoresis was on denaturing gel of 5% acrylamide. Exposure for panel A was U days and for panel B was 2 days. 147 A. 5' terminus B. 3' terminus mRNA mRNA 1 C C l N O O N F N Sl Nuclease N F Size + + - (I U er ul) '- t 4' Size (bp) [J ,I I I I} p {I l I I I I (bp) 1,632 - probel :. \— A 1 probe Ill J - II 517— 506” 396— " — 5 I 7 344" “506 —396 “344 B H E T E S 8 Hr; 1 7 i 1 1 ti U l IIIL : 3' labelled % ' * probes [ I II II: ”E III , 5 labelled fl probe 148 149 probe III (Fig. 6, map). Based on preliminary nucleotide sequence data (not shown), and size of the mRNA transcript detected (Fig. U), the arrow determined by the 3Upr fragment (left one of three above probe III, Fig. 6, map) would give rise to the predicted mRNA size. DISCUSSION This study was carried out in order to determine the location of the gene encoding MDHV-B within the MDHV genome. This was accomplished by two totally different approaches; (1) hybrid selection and cell-free translation followed by immunoprecipitation analysis, and (2) immunoscreening of a Agtll expression library followed by Southern blot analysis. The former approach was actually carried out in two stages, first in a more preliminary fashion with one of our polyclonal antisera reactive with MDHV-B glycoproteins and precursor polypeptide, and subsequently for purposes of confirmation, using a more recently available monoclonal antibody with the same reactivity. In all cases, the gene encoding MDHV-B was found in two locations, the BamHI D and H fragments. Clearly the evidence accumulated indicates that the gene identified in this study is that encoding MDHV-B. By combining immunoscreening of the Agt11 library, Southern blotting, hybrid- selection and cell-free translation followed by immunoprecipitation, Northern blotting and 81 nuclease mapping we have confirmed the identity of, and mapped the location of, the gene encoding the MDHV-B precursor polypeptide. The Agt11 library was constructed from genomic DNA rather than cDNA, so that all possible open reading frames would be expressed. Also, each DNA insert is equally represented, as opposed to cDNA, which is synthesized pp ppppp from mRNA, whose abundant transcripts would strongly influence the make-up of the cDNA library. Construction of a genomic DNA library is a reasonable approach since splicing of mRNA 150 151 transcripts is relatively uncommon among glycoprotein genes in herpes- virus systems. To date only the EBV ED350/220 mRNA is spliced (1). To screen the lgtll expression library (13), we used a mouse monoclonal antibody which was immunoreactive with MDHV-B glycoproteins and its precursor polypeptide, prUU (Sithole, Lee and Velicer, Manuscript in preparation). Three immunoreactive Agtll MDHV recombinant plaques were detected. All three recombinant Agtll clones contained a 500bp insert, and each were used individually, to probe Southern blots of the BamHI plasmid clones. In all three cases (only one is shown, Fig. 2) the signal hybridized to both fragments D and H on the MDHV genome. Since parts of these two fragments map in the unique long inverted repeats, the MDHV-B gene apparently is present as two copies per genome in these two repeat regions. More detailed Southern blot analysis of fragments BamHI D and H digested with the restriction enzymes SmaI and EcoRI using the same lgtll probe (Fig. 3), localized the gene to the SmaI a, and EcoRI a subfragments of the D fragment, and SmaI a and EcoRI c subfragments of the H fragment (Fig. 1, bottom) (U). All subsequent work involves the gene found in the BamHI H fragment, only. There are obviously many interesting questions regarding a second copy of the same gene, but its detailed analysis is beyond the scope of the current study. While construction and screening of Agtll libraries affords an efficient and rapid method for identifying genes, it is possible to misidentify a gene using this procedure. Thus, hybrid-selection, cell-free translation and immunoprecipitation analyses, with the monoclonal antibodies, were used to confirm the identity of the gene. In all cases a precursor polypeptide of UU kDa was detected. This unglycosylated precursor polypeptide comigrated with the unglycosylated 152 prUU seen after immunoprecipitation of lysates labeled in the presence of tunicamycin (Fig. U). Use of this approach, where the MDHV-B precursor polypeptide prUU is actually visualized, strongly supports our conclusion, resulting from immunoscreening of the lgt11 library and Southern blotting, that the gene identified encodes the precursor polypeptide of MDHV-B. Nuclease S1 mapping of the 5' terminus (Fig. 6A) resulted in two strongly hybridizing protected fragments and one minor protected species of sizes ~700, ~3UU and ~UOObp, respectively, whereas mapping of the 3' terminus (Fig. 68) revealed a single protected fragment ~U60bp in size. In addition the entire EcoRI c subfragment (U, Fig. 1, bottom and the region between the two EcoRI sites on the map of Fig. 6) was protected (data not shown) indicating no obvious splicing occurred in this region of the gene. Northern blot analysis (Fig. 5) using the EcoRI c subfragment as probe revealed a 1.8kb transcript. This is the size of mRNA appropriate to encode a precursor polypeptide of UU kDa, if we assume an average molecular weight of 120 per amino acid residue and a poly A tract of ~200 bases. A 1.8kb transcript would arise if transcription initiated at or near a site where the 3Upr fragment maps. Based on the 81 nuclease analysis, and predictions based on transcript size discussed above, the gene is sufficiently localized to permit determination of its entire nucleotide sequence (work currently in progress). In terms of the MDHV-B gene's unique dual location, there is no glycoprotein gene in other herpesvirus genomes that maps in the unique long inverted repeat regions. Preliminary evidence suggests that the gene encoding the immunologically related HVT-B does not map in the inverted repeat regions (data not shown). 153 The fact that three different laboratories (6,23, Sithole, Lee and Velicer, Manuscript in preparation) have immunoprecipitated gp100, gp60 and ng9 with monoclonal antibodies known to have virus neutralizing activity, suggests that some form, (if not all forms) of MDHV-B, is(are) responsible for generating neutralizing antibody pp Kipp, since they all have prUU as the polypeptide backbone and have at least one common epitope. Also gB, one laboratory's nomenclature for the same three MDHV glycoproteins (16), partially purified by affinity chromatography elicited partial protection against MD in chickens. While the relationship between virus neutralization and protective immunity has not been elucidated, it is assumed to be highly significant (16). On the other hand cell-mediated immunity could be involved. Clearly, learning more about the role of MDHV-B in these immune phenomena is of high priority in the MD system. Identification and characterization of the MDHV-B gene will be extremely valuable to the MDHV system in several different ways. (1) We have generated a lgtll expression library that could be used to identify other genes for which monoclonal antibodies are available. (2) The entire gene, including the 5' and 3' flanking regions, has been subcloned into the vector pCla12N (data not shown). This will be used for production of antigen in expression vectors, to evaluate further its role in protective immunity against MD, and in the development of new generation vaccines for use in the poultry industry. (3) Finally the complete nucleotide sequence is being determined and will prove useful in designing site-directed mutagenesis experiments for epitope mapping and other functional studies. This is only the second gene identified and cloned in the MDHV system, the other being MDHV-A (8). Since there 154 is evidence that MDHV-B plays at least a partial role in the immunoprevention of MD (16), its continued study by methods of recombinant DNA technology should yield much useful information regarding the nature and function of this oncogenic herpesvirus encoded membrane antigen. LITERATURE CITED Beisel, C., J. Tanner, T. Matsuo, D. Thorley-Lawson, F. Kezdy, and E. Kieff. 1985. Two major outer envelope glycoproteins of Epstein-Barr Virus are encoded by the same gene. J. Virol. 5Uz665-67U. Chirgwin, J.M., A.G. Przybyla, R.J. MacDonald, and W.J. Rutter. 1979. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 18:529U. Deatly, A.M., L. Feldman, and T. Ben-Porat. 198U. The large late virus transcripts synthesized in herpesvirus suis (pseudorabies) virus-infected cells are not precursors of mRNA. Virology 135:U52-U65. Fukuchi, K., A. Tanaka, and M. Nonoyama. 1985. The Structure of Marek's Disease Virus DNA, p. 68-83. In B. W. Calnek and J. L. Spencer (ed.), Proceedings of the International Symposium on Marek's Disease. American Association of Avian Pathologists, Inc., Kennett Square, Pa. Glaubiger, C., K. Nazerian, and L.F. Velicer. 1983. Marek's disease Herpesviruses. IV. Molecular Characterization of Maerk's Disease Herpesvirus A Antigen. J. Virol. U5:1228-123U. Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 198U. Identification with monoclonal antibodies of glycoproteins of Marek's disease virus and herpesvirus of turkeys related to virus neutralization. J. Virol. U9:101U-1017. 155 I” 12. 13. 1U. 10. 11. 12. 13. 1U. 15. 156 Ikuta, K., S. Ueda, S. Kato, and K. Hirai. 198U. Processing of glycoprotein gB related to neutralization of Marek's disease virus and herpes virus of turkeys. Microbiol. Immunol. 28(8):923-933. Isfort, R.J., H.-J. Kung, and L.F. Velicer. 1987. Identification of the gene encoding Marek's disease herpesvirus A antigen. J. Virol. 61:261U-2620. Isfort, R.J., I. Sithole, H.-J. Kung, and L.F. Velicer. 1986. Molecular characterization of Marek's disease herpesvirus B antigen. J. Virol. 59:U11-U19. Isfort, R.J., R.A. Stringer, H.-J. Kung, and L.F. Velicer. 1986. Synthesis, processing, and secretion of the Marek's disease herpesvirus A antigen glycoprotein. J. Virol. 57:U6U—U7U. Maniatis, T., E.F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Marek, J. 1907. Multiple Nerventzuendung (Polyneuritis) bei Huchnern. Dtsch. Tieraerztl. Wochenschr. 15:U17-U21. Mocarski, E.S., L. Pereira, and N. Michael. 1985. Precise localization of genes on large animal virus genomes: Use of lgt11 and monoclonal antibodies to map the gene for a cytomegalovirus protein family. Proc. Natl. Acad. Sci. USA 82:1266-1270. Nazerian, K. 1980. Marek's disease: a herpesvirus-induced malignant lymphoma of the chicken, p.665-682. In G. Klein (ed.), Viral oncology. Raven Press, New York. Okazaki, W., H.G. Purchase, and B.R. Burmester. 1970. Protection against Marek's disease by vaccination with a herpesvirus of turkeys (HVT). Avian Dis. 1U:U13-U29. 16. 17. 18. 19. 20. 21. 22. 23. 157 Ono, K., M. Takashima, T. Ishikawa, M. Hayashi, I. Yoshida, T. Konobe, K. Ikuta, K. Nakajima, S. Ueda, S. Kato, and K. Hirai. 198U. Partial protection against Marek's disease in chickens immunized with glycoproteins gB purified from turkey-herpesvirus- infected cells by affinity chromatography coupled with monoclonal antibodies. Avian. Dis. 29:533-539. Pappenheimer, A.M., L.C. Dunn, and V. Cane. 1929. Studies on fowl paralysis (neurolymphomatosis gallinarum). I. Clinical features and pathology. J. Exp. Med. U9:63-86. Pappenheimer, A.M., L.C. Dunn, and S.M. Seidlin. 1929. Studies on fowl paralysis (neurolymphomatosis gallinarum). II. Transmission experiments. J. Exp. Med. U9:87-102. Payne, L.N. 1982. Biology of Marek's disease virus and the herpesvirus of turkeys, p.3U7-U31. In B. Roizman (ed.), The herpesviruses, vol. 1. Plenum Publishing Corp., New York. Rigby, P.W.J., M. Dieckmann, C. Rhodes, and P. Berg. 1977. Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I. J. Mol. Biol. 113:237-251. Sharma, J.M., and H.A. Stone. 1972. Genetic resistance to Marek's disease. Delineation of the response of genetically resistant chickens to Marek's disease virus infection. Avian Dis. 16:89U-906. Silva, R.F., and L.F. Lee. 198U. Monoclonal antibody-mediated immunoprecipitation of proteins from cells infected with Marek's disease virus or turkey herpesvirus. Virology 136:307-320. Silva, R.F., and L.F. Lee. 1985. Isolation and partial characterization of three glycoproteins common to Marek's disease 2U. 25. 26. 27. 28. 29. 158 virus and turkey herpesvirus-infected cells, p.101-110. In B.W. Calnek and J.L. Spencer (ed.), Proceedings of the International Symposium on Marek's Disease. American Association of Avian Pathologists, Inc., Kennett Square, PA. Southern, E.M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. Van Zaane, D., J.M.A. Brinkhof, F. Westenbrink, and A.L.J. Gielkens. 1982. Molecular-biological characterization of Marek's disease virus. I. Identification of Virus-specific Polypeptides in Infected Cells. Virology 121:116-132. Van Zaane, D., J.M.A. Brinkhof, and A.L.J. Gielkens. 1982. Molecular-biological characterization of Marek's disease virus. II. Differentiation of Various MDV and HVT Strains. Virology. 121:133-1U6. Vieira, J. and Messing, J. 1982. The pUC plasmids, and M13mp7- derived system for insertion, mutagenesis and sequencing with synthetic universal primers. Gene 19:259-268. Young, R.A., and R.W. Davis. 1983. Yeast RNA polymerase II genes: Isolation with antibody probes. Science 222:778-782. Young, R.A., and R.W. Davis. 1983. Efficient isolation of genes by using antibody probes. Proc. Natl. Acad. Sci. USA 80:119U-1198. "‘lIIIIIIIIIIIIIIIES