Extensive Homology Between the Herpes Simplex Virus Type 2 Glycoprotein F Gene and the Herpes Simplex Virus Type 1 Glycoprotein C Gene

Transcription

1 JOURNAL OF VIROLOGY, OCt. 1984, p X/84/ $02.00/0 Copyright C) 1984, American Society for Microbiology Vol. 52, No. 1 Extensive Homology Between the Herpes Simplex Virus Type 2 Glycoprotein F Gene and the Herpes Simplex Virus Type 1 Glycoprotein C Gene DONALD J. DOWBENKO AND LAURENCE A. LASKY* Department of Vaccine Development, Genentech, Inc., South San Francisco, California Received 12 March 1984/Accepted 29 June 1984 The region of the herpes simplex virus type 2 () genome which maps colinearly with the glycoprotein C (gc) gene has been cloned, and the DNA sequence of a 2.29-kilobase region has been determined. Contained within this sequence is a major open reading frame of 479 amino acids. The carboxyterminal three-fourths of the derived protein sequence showed a high degree of sequence homology to the gc amino acid sequence reported by Frink et al. (J. Virol. 45: , 1983). The amino-terminal region of the sequence, however, showed very little sequence homology to gc. In addition, the gc sequence contained 27 amino acids in the amino-terminal region which were missing from the protein. Computer-assisted analysis of the hydrophilic and hydrophobic properties of the derived sequence demonstrated that the protein contained structures characteristic of membrane-bound glycoproteins, including an amino-terminal signal sequence and carboxy-terminal hydrophobic transmembrane domain and charged cytoplasmic anchor. The protein sequence also contained seven putative N-linked glycosylation sites. These data, in conjunction with mapping studies of Para et al. (J. Virol. 45: , 1983) and Zezulak and Spear (J. Virol. 49: , 1984), suggest that the protein sequence derived from the genome corresponds to gf, the homolog of gc. The genomes of herpes simplex virus type 1 () and encode at least four different glycoproteins, ga/b, gc, gd, and ge, which are found on the surface of the infected cell and the virion (29). Three of the four glycoproteins, ga/b, gd, and ge, have been found to be structurally similar in the two virus types, based on immunological and biochemical criteria (9, 19, 20, 22, 23). For example, a recent analysis of the DNA sequences of the gd genes from and revealed that the gd proteins had an overall sequence homology of -85% (L. Lasky and D. Dowbenko, DNA, in press). Thus, it may be concluded that the primary sequences of these three glycoproteins have been relatively well conserved, since the two virus types diverged from each other. gc initially appeared to have no obvious homolog in. gc was thought to be type specific since antibodies against this glycoprotein were found to react almost exclusively with gc (24). In addition, no detectable immunological reactions could be demonstrated between gc and antisera made against (8). A protein having the same electrophoretic mobility as gc has been demonstrated in ; however, it did not map colinearly with gc (26). In contrast to, appears to encode yet another glycoprotein, termed gf (2, 3, 20, 32). Although gf had an electrophoretic mobility which was much faster than gc, initial mapping studies with recombinant viruses revealed that this protein was encoded by a region of the genome which was approximately colinear with the gene for gc (20, 32). Subsequent studies with finer structural mapping revealed a much closer colinearity between the gc and the gf coding regions (33). In addition, it has been recently demonstrated that a monoclonal antibody against gf cross-reacts * Corresponding author. weakly with gc (34) and that a polyclonal antiserum made against virion envelope proteins precipitates gf (32), suggesting a possible structural homology between the two glycoproteins. Thus, it appeared that a possible homolog to gc was the gf protein. The most conclusive proof of relatedness between two proteins is to demonstrate homology at the amino acid level. We have determined the DNA sequence of a 2.29-kilobase (kb) region of the genome which is colinear with the gc gene. Translation of a large open reading frame in this region demonstrates that a protein which has significant homology to gc is encoded in this region. We propose that this region encodes the gf gene and that the gf protein is the homolog of gc. MATERIALS AND METHODS Cells, virus, and DNA isolation. (strain G) was grown on HEp-2 cells after infecting the cell culture at an input multiplicity of 0.1 for 3 days at 33 C in Dulbecco modified Eagle medium containing 10% fetal bovine serum and antibiotics. DNA was isolated by proteinase K digestion, followed by CsCl ultracentrifugation as previously described (1). DNA manipulations. Restriction enzymes, DNA polymerase Klenow fragment, T4 DNA ligase, and T4 polynucleotide kinase were purchased from Bethesda Research Laboratories and were used according to the directions of the supplier. Molecular cloning of DNA restriction fragments. The EcoRI P fragment, which corresponds to map position of the genome, was isolated from EcoRIdigested DNA on 5% acrylamide gels. The isolated 154 fragment was cloned into EcoRI-digested puc9 (30). This plasmid was called puc-rip. The puc-rip subclone was then used to localize an Sacl fragment of the genome which contained the EcoRI P

2 VOL. 52, 1984 gf SEQUENCE 155 fragment. Southern blot experiments (28) revealed that a 4.9- kb SacI fragment of contained the EcoRI P fragment. This fragment was isolated on 0.7% agarose gels and was cloned into a pbr322-derived plasmid which contained a unique SacI site (15). This plasmid was called pbrsaci-e. Further restriction enzyme analysis of pbrsaci-e demonstrated a 2.9-kb Sall fragment with sequences homologous to the EcoRI P fragment which was subcloned into Salldigested puc9 as described above. This plasmid was called pgc2sal2.9. DNA sequence analysis of cloned DNA. The majority of DNA sequences were determined by using the dideoxynucleotide chain termination technique. Various fragments were subcloned into the replicative form of the m13 phage vectors mp7, mp8, and mp9, and the DNA sequence was determined as described previously (18). In some cases, fragments were 32p labeled at their 5' ends with [y-32p]atp and T4 polynucleotide kinase, and the DNA sequence of the fragment was determined by using the chemical degradation method (16). Computer-assisted analysis of DNA and protein sequence data was performed by the HOM program (7). The hydropathy of the deduced amino acid sequences was analyzed, using a width of 12 amino acids and a jump of 1 (13). Southern blot analysis of DNA. Restriction endonuclease-digested DNA and plasmid DNA were fractionated on 1.5% agarose gels and blotted onto nitrocellulose Sal I Il TRL 0' Sm 0.4 UL Soc Bg 0.6., /AL,,P1 t. 1,, I I RI RI -Sol by standard procedures. The single-stranded ends of the SacII fragment, marked with a star in Fig. 1, were filled in with the Klenow fragment of DNA polymerase I, and the resultant blunt-ended fragment was ligated to SmaI-digested ml3mp8 replicative form (18) with T4 DNA ligase. The single-stranded DNA prepared from this ligation and transfection was used as a template for the synthesis of 32p_ labeled single-stranded probe DNA of high specific activity (109 cpml,lg) by using the Klenow fragment of DNA polymerase I. Hybridizations were performed by standard procedures (28). RESULTS Molecular cloning of the gf coding region of the genome. The strategy adopted for the isolation of the gf gene of was based on the assumption that this gene was colinear with the gc gene. This assumption was supported by the recent finding that a 75,000-dalton glycoprotein, gf, with antigenic relatedness to gc is found in and that the gene for this protein is approximately colinear with the gc gene (10, 32). In addition, the isolation of a monoclonal antibody which binds to both gc and gf further suggested that these two proteins may be homologous to each other (34). It was thus reasoned that DNA sequence analysis of the genomic region which is colinear with the gc gene would Pv maopunits IRL IRS Us TRs -c{ c pucri-p,,p HSV2 Genome Sac Bg RI RI SoIl Rs i SmRs Rs Sm Rs RsRsSm 11 I11 -I -I I I S * So sosa So so So sa So sosa So so sa So3 -.. RI Sa~s RI " Rsj I I pbrsac l-"e" Sol pgc2sal 2.9, o kbp... ATG.X TAA... Major Open Reading Frame FIG. 1. Cloning of pgc2sal2.9, the region which maps colinearly with gc. The region of the genome mapping from to 0.66 was cloned as an Sacl fragment (pbrsacl-e), using the 600-base-pair EcoRI P fragment as a probe. An Sall subclone of pbrsaci-e, pgc2sal2.9, was used for DNA sequence analysis. Arrows refer to the sequenced regions, and the location of a major 479-aminoacid open reading frame derived from the sequence is illustrated. Various restriction sites are illustrated, including the EcoRI sites which delineate the EcoRI P fragment and the BglIl site which is found at the right end of the BgllI N fragment (map position, ) (25). The SacII fragment marked with a star was used in Southern blotting experiments to investigate the deletion which appears in this region (see text). Other sites were used for DNA sequencing experiments. Abbreviations: Sm, SmaI; Sa, SacII; Rs, RsaI; Bg, BgII; Pv, PvuII; RI, EcoRI.

3 156 DOWBENKO AND LASKY J. VIROL. result in the derivation of protein sequence information the gc gene which maps between and of which would localize the gf gene. the genome (10). This fragment was isolated from an The 600-base-pair EcoRIP fragment of the genome EcoRI digest of DNA and cloned in the plasmid has been shown to map at position (25). This region puc9 (30), and its DNA sequence was determined (16, 18). is approximately colinear with the known coding region of Comparison of the resultant sequence with the gc * ** * ** ** * ** * ** ** * **** * G-GATGGGGCCCGGGTATAAATTCCGGAAGGGGACACGGGCTACCCTCACTACCGAGGGC 60 GTGCCGTGGA-CGGGTATAAAGGCCAGGGGGGCAGGCGGGC--CCATCACTGTT-AGGGT NTATA 1" -----> gc mrna 5' end * * * ** *** * * * * *** * *** ** * * * GCTTGGTCGGGAGGCCGCATCGAACGC-ACACCCCCATCCGGTGG---TC--CGTGTGGA 120 GTTAGGTTGGGAGGTGGCACAAAAAGCGACACACCCGTGTTGTAGTTGTCCGCGGGAGGC * ******* ** * ** ** *** * * ** GGTCGTTTTTCAGTGCCCGGTC-TCGCTTTGCCGGGAACGCTAGCCGATCCCTCGCAAGG 180 GGTGGTTTCCGGCAACCC--TCCTCGCTGCGCCGGGCGCGCCCACCGGTCCTTCGCGGGG *** ***** * * * * ** * GGGAGGCG---TCGGG-CATGGCCCCTGGGCGGGTGGGCCTTGCCGTGGTCCTGTGGGGC 240 GCCGGGGCTCTTCTGGTCATGGCCCTTGGACGGGTGGGCCTAACCGTGGGCCTGTGGGGC gt7hssv-2 gf initiation codons * * * * * ** **** * ********* * * * CTGTTGTGGCTCGGGGCGGGGGTGGCCGGGGGCTCGGAAACTGCCTCCACCGGGCCCACG 300 CTGCTGTGGGTGGGTGTGGTCGTGGTGCTGGCCAAT------GCCTCCCCCGGACGCACG * * * ** ** ** * * * * * *************** ATCACCGCGGGAGCGGTGACGAACGCGAGCGAGGCCCCCACATCGGGGTCCCCCGGGTCA 360 ATAACGGTGGGCCCGCGGGGGAACGCGAGCAATGCCGCCCCCTCG GCCGCCAGCCCGGAAGTCACCCCCACATCGACCCCAAACCCCAACAATGTCACACAAAAC ******** **** * *** ** * * * * * **** * * AAAACCACCCCCACCGAGCCGGCCAGCCCCCCAACAACCCCCAAGCCCACCTCCACGCCC GTCCCCCGGAACCGATCCGCCCCCCGAACCACACCCACGCCCCCCCAACCCCGC ***** **** ****** * ** * ********* ****** *** AAAAGCCCCCCCACGTCCACCCCCGACCCCAAACCCAAGAACAACACCACCCCCGCCAAG 540 AAGGCGACGAAAAGTAAGGCCTCCACCGCCAAACCGGCCCCGCCC------CCC---AAG * * **** * * ** ** ** *** * * * * ** TCGGGCCGCCCCACTAAACCCCCCGGG---CCCGTGTGGTGCGACCGCCGCGACCCATTG 600 ACCGGG---CCCCCGAAGACATCCTCGGAGCCCGTGCGATGCAACCGCCACGACCCGCTG GCCCGGTACGGCTCGCGGGTGCAGATCCGATGCCGGTTTCGGAATTCCACCCGCATGGAG 660 GCCCGGTACGGCTCGCGGGTGCAAATCCGATGCCGGTTTCCCAACTCCACCCGCACGGAG * * ** * *** ** * *** *** * * TTCCGCCTCCAGATATGGCGTTACTCCATGGGTCCGTCCCCCCCAATCGCTCCGGCTCCC 720 TCCCGCCTCCAGATCTGGCGTTATGCCACGGCGACGGACGCCGAGATCGGAACGGCGCCT ** * ** *** * ** * * * ** * GACCTAGAGGAGGTCCTGACGAACATCACCGCCCCACCCGGGGGACTCCTGGTGTACGAC 780 AGCTTAGAGGAGGTGATGGTAAACGTGTCGGCCCCGCCCGGGGGCCAACTGGTGTATGAC ** * * * AGCGCCCCCAACCTGACGGACCCCCACGTGCTCTGGGCGGAGGGGGCCGGCCCGGGCGCC 840 AGCGCCCCCAACCGAACGGACCCGCACGTGATCTGGGCGGAGGGCGCCGGCCCGGGCGCC ** * ** * * ** ****** * * ** GACCCTCCGTTGTATTCTGTCACCGGGCCGCTGCCGACCCAGCGGCTGATTATCGGCGAG 900 AGCCCGCGGCTGTACTCGGTCGTCGGGCCGCTGGGTCGGCAGCGGCTCATCATCGAAGAG * **** * * * * ** * * ** GTGACGCCCGCGACCCAGGGAATGTATTACTTGGCCTGGGGCCGGATGGACAGCCCGCAC 960 CTGACCTTGGAGACCCAGGGCATGTACTACTGGGTGTGGGGCCGGACGGACCGCCCGTCC * * * * * * * * GAGTACGGGACGTGGGTGCGCGTCCGCATGTTCCGCCCCCCGTCTCTGACCCTCCAGCCC 1020 H SV-2 GCGTACGGGACCTGGGTGCGCGTTCGCGTGTTCCGCCCTCCGTCGCTGACCATCCACCCC * * * CACGCGGTGATGGAGGGTCAGCCGTTCAAGGCGACGTGCACGGCCGCCGCCTACTACCCG 1080 CACGCGGTGCTGGAGGGCCAGCCGTTTAAGGCGACGTGCACGGCCGCCACCTACTACCCG * * * * * * ** * * ** * * CGTAACCCCGTGGAGTTTGACTGGTTCGAGGACGACCGCCAGGTGTTTAACCCGGGCCAG 1140 GGCAACCGCGCGGAGTTCGTCTGGTTCGAGGACGGTCGCCGGGTATTCGATCCGGCCCAG ** * * * ** * * ATCGACACGCAGACGCACGAGCACCCCGACGGGTTCACCACAGTCTCTACCGTGACCTCC 1200 ATACACACGCAGACGCAGGAGAACCCCGACGGCTTTTCCACCGTCTCCACCGTGACCTCC * * * * * GAGGCTGTCGGCGGCCAGGTCCCCCCGCGGACCTTCACCTGCCAGATGACGTGGCATCGC 1260 GCGGCCGTCGGCGGCCAGGGCCCCCCGCGCACCTTCACCTGCCAGCTGACGTGGCACCGC FIG. 2. Continued

4 * * * * *** *** GACTCCGTGACGTTCTCGCGACGCAATGCCACCGGGCTGGCCCTGGTGCTGCCGCGGCCA 13?0 GACTCCGTGTCGTTCTCTCGGCGCAACGCCAGCGGCACGGCATCGGTGCTGCCGCGGCCA * * ** * *** * * * ACCATCACCATGGAATTTGGGGTCCGGCATGTGGTCTGCACGGCCGGCTGCGTCCCCGAG 1380 HSV-? ACCATTACCATGGAGTTTACGGGCGACCATGCGGTCTGCACGGCCGGCTGTGTGCCCGAG * * * ** ** * GGCGTGACGTTTGCCTGGTTCCTGGGGGACGACCCCTCACCGGCGGCTAAGTCGGCCGTT 1440 GGGGTGACGTTTGCCTGGTTCCTGGGGGACGACTCCTCGCCGGCGGAGAAGGTGGCCGTC * * *** ** * **** * * * ** * ACGGCCCAGGAGTCGTGCGACCACCCCGGGCTGGCTACGGTCCGGTCCACCCTGCCCATT 1500 GCGTCCCAGACATCGTGCGGGCGCCCCGGCACCGCCACGATCCGCTCCACCCTGCCGGTC ** * * * * * * * * * * * TCGTACGACTACAGCGAGTACATCTGCTGGTTGACCGGATATCCGGCCGGGATTCCCGTT 1560 TCGTACGAGCAGACCGAGTACATCTGCCGGCTGGCGGGATACCCGGACGGAATTCCGGTC CTAGAGCACCACGGCAGTCACCAGCCCCCACCCAGGGA*CCCACCGAGCGGCAGGTGATC 1620 CTAGAGCACCACGGCAGCCACCAGCCCCCGCCGCGGGACCCCACCGAGCGGCAGGTGATC ** * * * * * * *** * * ** * * * ** GAGGCGATCGAGTGGGTGGGGATTGGAATCGGGGTTCTCGCGGCGGGGGTCCTGGTCGTA 1680 CGGGCGGTGGAGGGGGCGGGGATCGGAGTGGCTGTCCTTGTCGCGGTGGTTCTGGCCGGG * ** * * ** * ** *** ** ** * ** * * ACGGCAATCGTGTACGTCGTCCGCACATCACAGTCGCGGCAGCGTCATCGGCGGTAACGC 1740 ACCGCGGTAGTGTACCTCACCCACGCCTCCTCGGTGCGCTATCGTCGGCTGCGGTAACTC gc, HSV7 gf termination codons ** ** *** **** ******* * *** ** ***** ** * *** ** * * GAGACCCCCCCGTTACCTTTTTAATATCTATATAGTTTGGTCCCCCCTT---CTATCCCG 1800 CGGGGCCGGGCCCGGCCGCCGGT-TGTCTTCTTT-TCCACCCCTTCCGTCCCCCGTACCC * ************************** * * * ***** * * * CC CACCGCTGGGCGCTATAAAGCC-GCCACCCTCTC 1860 ACCACACCCCACCCCACCCCCCCGCCGTCCCCCGGGCGTTATAAGC--CGCCGCACTCGC "TATA 2" ** * ******* * * * * *** * * ** *** * TTCCCTCAGGTC---ATCCTTGGTC-GATCCCGAACGAGACACGGCGTGGAG---CAAAA 1920 TTTTCCCACCGGAAAATCCTCGGCCCGATCC-GAACGGCGCACGCCGCGTGGGCTCCAAA **** *** **** ** * ** ** ** ** ***** * ** ***** CGCCTCCCCCTGAGCC-GCTTTCCTACCAACACACCGGCATGCC----T-CT--G CGCCTCCGGAAGAGAGCGCCCCGCCCCGAT-ATTCAAGCCCGCGGTGGTGCTATGGCTTT second open readinj frame initiation codon -CGGGCATCGGAACAGCC-TACCGGCCCCTGGGCCCCGGGACACCCCCCATGCGGGCTCG 2040 CCGTGCTTCGGGACCCGCCTACCAGCCCCTCGCCCCGCGGCCTCCCCCGGCGCGGGCTCG 730 bp mrna initiation codon ** * * * * * ** * * *** * * * * GCTCCCCGCCGCGGCCTGGGTTGGCGTCGGGACCATCATCGGGGGAGTTGTGATCATTGC 2100 TGTTCCGGCCGTGGCCTGGATCGGCGTCGGAGCGATCGTCGGGGCCTTTGCGCTCGTCGC *C* * * * * * C* * * CGCGTTGGTCCTCGTGCCCTCGCGGGCCTCGTGGGCACTTTCCCCATGCGACAGCGGATG 2160 CGCGTTGGTTCTCGTACCCCCTCGGTCCTCGTGGGGACTCTGCCCGTGCGACAGCGGCTG * * *** * * ** * * * * * GCACGAGTTCAACCTCGGGTGCATATCCTGGGATCCGACCCCCATGGAGCACGAGCAGGC 2220 GCAGGAATTCAACGCGGGATGCGTCGCGTGGGACCCCACCCCCGTCGAGCACGAGCAGGC * * * * * * * * GGTCGGCGGCTGTAGCGCCCCGGCGACCCTGATCCCCCGCGCGGCTGCCAAACAGCTGGC 2280 GGTCGGCGGCTGCAGCGCGCCGGCCACCCTTATCCCCCGTGCGGCCGCCAAGCACCTGGC CGCCGTCGCACGCGTCCAGTCGGCAAGATCCTCGGGCTACTGGTGGGTGAGCGGAGACGG 2340 CGCTCTGACACGCGTCCAGGCGGAGAGATCGTCGGGTTACTGGTGGGTGAACGGAGACGG * * * * * * * ** * * * * CATTCGGGCCCGCCTGCGGCTCGTCGACGGCGTTGGCGGTATTGACCAGTTTTGCGAGGA 2400 CATCCGGACCTGTCTGAGACTCGTCGACAGCGTCAGTGGCATCGACGAGTTTTGCGAGGA GCCCGCCCTTCGCATATGCTACTATCCCCGCAGTCCCGGGGGCTTTGTTCAGTTTGTAAC 2460 GCTC TTCGACCCGCAACGCGCTGGGGCTGCCGTGA 2491 FIG. 2. DNA sequence obtained from pgc2sal2.9 compared with the DNA sequence of the HiSV-1 gc region (10). The gc region () and the sequence obtained from pgc2sal2.9 () were compared by the HOM program (7). Because various deletions were utilized to maximize sequence overlap, all positions, including spaces, have been numbered for clarity. Stars are placed over nonmatching nucleotides. The underlined A residue at position 43 of the sequence is the approximate transcriptional start site of the gc mrna (10). TATA 1 and TATA 2 are the probable transcriptional control regions for the gc mrna and the 730-base mrna, respectively (10, 17). The inserted T residue at position 1728 of the sequence was discovered by resequencing this region (M. Jackson, unpublished data) and was found to introduce an in-phase stop codon at positions 1735 to 1737 which was homologous to the stop codon for the position of a second initiation codon at positions 1975 to

5 158 DOWBENKO AND LASKY sequence (10) revealed a remarkable degree of sequence homology between the EcoRI P fragment and the 3' end of the gc coding region. Thus, the EcoRI P fragment was subsequently used as a probe to isolate an Sacl restriction endonuclease fragment from genomic DNA that overlapped the EcoRI P fragment sufficiently to include the remainder of the gene which was homologous to the gc gene. Figure 1 illustrates the steps taken to isolate a 2.9-kb Sall fragment from the genome which contained the EcoRI P fragment and which was used for subsequent DNA sequence analysis. DNA sequence analysis of the EcoRI-P region of the genome. The 4.3-kb SacI E fragment, which was isolated from the genome based upon its sequence homology to the EcoRI P fragment, was further digested to give a 2.9- kb Sall fragment which was termed pgc2sal2.9. Figure 1 illustrates the fragments from pgc2sal2.9 which were subjected to DNA sequence analysis, using either the dideoxynucleotide sequencing procedure (18) or the chemical degradation procedure (16). In addition, Fig. 1 shows the position of the EcoRI P fragment within pgc2sal2.9 as well as the position of a BglII site which corresponds to the right-hand end of the BglII N fragment at position of the genome (25). Figure 2 illustrates the derived DNA sequence of this region of the genome as well as a sequence comparison with the DNA sequence of the gc gene region of (10). The overall sequence homology between these two fragments was -68%. However, certain regions of the sequence showed either a much higher or lower degree of sequence homology than others. For example, the sequences between positions 0 and 570 of the and HSV- 2 sequences showed only 51% homology, whereas the region FIG. 3. Southern blot analysis of genomic DNA and pgc2sal2.9 DNA. The region spanning the 81-base-pair region missing in the sequence (Fig. 2) ( positions 346 to 426) was analyzed with the SacII fragment marked with a star in Fig. 1 which overlaps the deleted region. Lanes 1 to 3 are restriction digests of genomic DNA, and lanes 4 to 6 are restriction enzyme digests of pgc2sal2.9. The digested DNAs were electrophoresed on 1.5% agarose gels, denatured, blotted onto nitrocellulose, and probed with the 32P-labeled SacII fragment. (The arrow shows the position of the 564-base-pair HindlIl fragment ofa DNA.) Lanes 1 and 6, SmaI plus BglII; lanes 2 and 5, SmaI; lanes 3 and 4, SacII. J. VIROL. between positions 570 and 1740 showed a much higher degree of sequence homology (80%). An additional highly homologous region (70%) was also found at the end of the two sequences from position 1975 to position In addition to the nucleotide sequence changes, the two genomes showed various deletions or insertions when compared with each other. The most notable was an 81-base-pair region found at positions 346 to 426 of the gc sequence which is missing from the genome. From this overall sequence comparison, it appeared that there was a high degree of sequence homology between the gc region and the region sequenced here. Frink et al. (10) have found that the 5' end of the 2,520- base mrna encoding gc maps to the underlined A residue at position 43 (Fig. 2). In addition, they pointed out an AT-rich TATA box (17) sequence -22 base pairs 5' to this residue. Comparison of the two sequences shows that the and sequences both contained the identical sequence, CGGGTATAAA, in this region (Fig. 2). This sequence is identical to that reported previously by Whitton et al. (31), which is found to occur at the TATA box regions in many of the and sequences determined thus far. This conserved sequence is also followed by a G- rich region in both virus genomes. In addition to this putative transcriptional-control region, a second TATA box was found in both sequences at positions 1845 to 1849 (Fig. 2). This second TATA box has been hypothesized to control the transcription of a 730-base mrna in the genome (10). Both and contain this sequence surrounded by GC-rich flanking sequences, including a CGGGCG sequence which is similar to the CGGG sequence preceding the first TATA box. In addition, both genomes encode open reading frames 3' to these second TATA boxes, which will be discussed below. To determine whether the 81-base-pair deletion described above was actually found in the genome or whether it was an artifact of cloning or sequencing, Southern blot analysis of the genomic DNA and the cloned DNA was performed. A 32P-labeled probe was prepared from an SacII fragment (see fragment in Fig. 1) which spans the region missing the 81 nucleotides. If the genomic DNA is missing the 81-base-pair region, then an SmaI-BgIII fragment spanning this region will be 576 base pairs, an SmaI fragment will be 662 base pairs, and an SacII fragment will be 195 base pairs. The predicted restriction sites surrounded the region missing the 81 base pairs in both the genomic DNA and the cloned DNA (Fig. 3). In addition, the genomic fragments and the cloned fragments comigrated exactly, demonstrating that the deletion is not an artifact of cloning or sequencing. Analysis of the major open reading frame within the 2.9-kb SalI fragment. Analysis of the potential coding sequences within the 2.9-kb Sall DNA fragment of revealed an open reading frame of 479 amino acids which began with the methionine encoded at positions 199 to 201 of the sequence and ended at the TAA termination codon at positions 1735 to 1737 of the sequence (Fig. 2). Both the gc protein and the open reading frame initiate at approximately the same position in the two sequences, relative to the TATA box homologies (Fig. 2). In addition, although it initially appeared that the open reading frame found in this region terminated 12 codons before the gc gene, resequencing of the carboxyterminal region of the gc gene sequence (M. Jackson, unpublished data) of strain F revealed that the sequence reported by Frink et al. (10) was missing a thymi-

6 A. C C C CC CCCN C N C CNC C CC * * * ** ****** * * * *** * ** gc 1 MAPGRVGLAVVLWGLLWLGAGVAGGSETASTGPTITAGAVI EAPTS gf MALGRVGLAVGLWGLLWVGVVVVLL: GRTITVGPRqgWANAAPS NCNCC N N CNCN CCNC **************************** ***** * * **** **** gc 50 GSPGSAASPEVTPTSTPNPNT:.::PTEPASPPTTPKPTSTPKSPPT gf VPRPffPRTTPTPPQPRKATKS NCCNC NNN C C CCC N N C N ***** ***** * * * * **** * * * * gc 101 STPDPKPKI'-N AKSGRPTKPPG-PVi4DRRDPLARYGSRVQI P F gf 71 KASTAKPAPW--P-KTG-PPKTSSEPV qnrhdplarygsrvqiiqgfp; C C CCCCNCC CC N CC CC N N * * ******* ** * ** ** * * gc 150 g. tmefrlqiwrysmgpsppiapapdleevl APPGGLLVYDSAPg gf 117 ffrtesrlqiwryatatdaeigtapsleevm ffappggqlvydsa E C C N C CN C C CC NC * * * * ** * *** ** gc 200. PHVLWAEGAGPGADPPLYSVTGPLPTQRLIIGEVTPATQGMYYLAWGR gf 167,.fJPHVIWAEGAGPGASPRLYSVVGPLGRQRLIIEELTLETQGMYYWVWGR C N NC C CC C C N NC * ** * ** * * * ** gc 250 MDSPHEYGTWVRVRMFRPPSLTLQPHAVMEGQPFKA raaayyprnpve gf 217 TDRPSAYGTWVRVRVFRPPSLTIHPHAVLEGQPFKA CAATYYPGNRAE N N C N C N C C C C C * * * * * * * * * * * gc gf FDWFEDDRQVFNPGQIDTQTHEHPDGFTTVSTVTSEAVGGQVPPRTF FVWFEDGRRVFDPAQIHTQTQENPDGFSTVSTVTSAAVGGQGPPRTFTEC C C C C C CCN C * * ****** gc 350 MTWHRDSVTFSRR. LALVLPRPTITMEFGVRHV C A C PEGVTFA gf 317 LTWHRDSVSFSRR : TASVLPRPTITMEFTGDHA A C PEGVTFA C C C CC C NC C C C CCC N C * * * ** * ** * * * *** * * gc 400 WFLGDDPSPAAKSAVTAQE C HPGLATVRSTLPISYDYSEYIFWLTGYP gf 367 WFLGDDSSPAEKVAVASQT C RPGTATIRSTLPVSYEQTEYIj LAGYP N N C CC CC C C CC C * * * ** ** * * ** * gc 450 AGIPVLEHHGSHQPPPRDPTERQVI EAIEWVGIGI'GVLAAGVLVVTAI VY gf 417 DGIPVLEHHGSHQPPPRDPTERQVIRAVEGAGIGVAVLVAVVLAGTAVVY CCCC NN N NN **** ** * ** gc gf VVRTSQSRQRHRR LTHASSVRYRRLR B. C C C C C CCCCC ******************** * * * * * ***** 1 1 MAFRASGPAYQPLAPRPPPARARVPAVAWIGVGAIVGAFALVAALVLVP MRARLPAAAWVGVGTIIGGVVIIAALVLVP C C C C C C CC C * * * * * * ** * PRSSWGLCPCDSGWQEFNAGCVAWDPTPVEHEQAVGGCSAPATLIPRAAA SRASWALSPCDSGWHEFNLGCISWDPTPMEHEQAVGGCSAPATLIPRAAA C CC CC N CN C C N C * ** ** * ** * * * * KHLAALTRVQAERSSGYWWVNGDGIRTCLRLVDSVSGIDGFCEEL KQLAAVARVQSARSSGYWWVSGDGIRARLRLVDGVGGIDQFCEEPALRIC 131 YYPRSPGGFVQFVTSTRNALGLP C. N C CCCNN CN NN C NNNCNN NNNN N C CC ****** *** ***** ** * ** * ****** **** * * ** frame 3 1 MSHKTKPPPPSRPAPQQPPSPP-PRPKAPPRPPPTPNPRTTPPPPSRAAP frame 3 1 MPP--PRSPGTDPPPEPHPRPPNPARRRKVRPPPPNRPRPPRPGP NCN N N C C C CC N C N C NN C *** ** * * * * ** * * * * ** * frame 3 frame LNPPG-PCGATAATHWPGTARGCRSDAGFGIPPAWSSASRYGVTPWVRPP RRHPRSPCDATATTRWPGTARGCKSDAGFPTPPARSPASRSGVMPRRRTP C C NCC * * *** frane 3 frame QSLRLPT RSERRLA FIG. 4. (A) Translation of the large open reading frame and comparison with the gc amino acid sequence. The single-letter amino acid symbols were used. gc refers to the gc sequence, and gf refers to the open reading frame sequence. The proteins were compared by the HOM program, which maximized homologies by inserting gaps where necessary (7). Stars are placed over 159

7 160 DOWBENKO AND LASKY J. VIROL. 3.0 gc-l o I gc-2(gf) 2.0r lo Amino Acid Position FIG. 5. Hydropathy analysis of the gc protein and the major open reading frame protein. The hydropathy of each protein was determined by the program of Hopp and Woods (13). Hydrophobic regions are above the midline, and hydrophilic regions are below the midline. Stretches of 12 amino acids were analyzed, and the average hydropathy was calculated. Putative asparagine-linked glycosylation sites (14) are indicated by closed circles. gc-1, gc protein hydropathy; gc-2 (gf), major open reading frame protein hydropathy. dine nucleotide after position 1727 and that insertion of this residue resulted in a translated gc protein terminating at the same place as the open reading frame (1735 to 1737; Fig. 2). Thus, when taking the various deletions and insertions into account, the gc gene and the open reading frame show a very high degree of overlap (Fig. 2). Figure 4 illustrates the high degree of sequence homology between the gc gene and the 479-amino-acid open reading frame. The first 19 amino acids contain -80% sequence homology, with the changes in the first 25 amino acids being all conservative with respect to charge. From residue 124 of gc (residue 90 of the sequence) to the end of both proteins there is -74% sequence homology, with 75% of the amino acid changes being conservative with respect to charge. Five putative N-linked glycosylation sites (NXS or NXT [14]) are conserved between the two proteins, and all seven cysteine residues are located in homologous positions relative to the C terminus. In addition to the overall conservation of sequences in the carboxynonhomologous amino acids. Putative N-linked glycosylation sites (NXS or NXT) (14) are shaded, and cysteine residues are boxed. Only amino acids, and not spaces, are numbered. (B) Translation of the second open reading frame () and comparison with the 730-base mrna protein. 730 is the incomplete amino acid sequence of the second open reading frame from positions 1975 to 2406 of the sequence shown in Fig is the amino acid sequence derived for the protein encoded by the 730-base mrna of (10). (C) Translation of the open reading frame homologous to the frame 3 polypeptide of (10). frame 3 refers to the open reading frame found by Frink et al. (10) which overlaps the gc sequence and which corresponds to positions 407 to 727 of the gc sequence shown in Fig. 2. frame 3 refers to a homologous open reading frame found in the sequence at positions 333 to 727 in Fig. 2. Conserved amino acid changes, with respect to charge, are marked C, and nonconserved changes, with respect to charge, are marked N.

8 VOL. 52, 1984 terminal three-fourths of the proteins, there are also large regions of contiguous amino acid sequence homology up to 20 residues in length (i.e., positions 385 to 405 of the sequence and 352 to 372 of the sequence). It may be concluded from this sequence comparison that the open reading frame in this region of the genome encodes a protein which is homologous to gc. Although the protein encoded in this region shows a remarkable degree of sequence homology to the gc sequence, there are several notable differences between the two sequences. The most striking difference is a deletion of 27 amino acids in the sequence which is found in the gc sequence from residues 50 to 76 (Fig. 4A) and which corresponds to the 81-base-pair deletion described above. In addition to this large deletion, both sequences show minor deletions of one or two amino acids. All of these deletions are found in the amino-terminal regions of the proteins. In addition to these deletions, there are a large number of amino acid changes in the amino-terminal region of the proteins which are clustered between residues 29 and 123 of the gc sequence (residues 31 to 90 of the HSV- 2 sequence). Only 30% of the amino acids in this region are homologous, with much of this homology due to conserved proline residues. Altogether, 43% of the amino acid substitutions found in this region are nonconservative with respect to charge. The only other regions which showed such a large number of changes are a carboxy-terminal hydrophobic domain (residues 476 to 496 of the sequence and 443 to 463 of the sequence) in which the proteins are 55% homologous, but in which all the changes are conserved, uncharged, hydrophobic amino acids, and the carboxy termini of the proteins in which the sequences are only 25% homologous, but in which the overall amino acid composition is similar (residues 500 to 512 of the sequence and 467 to 479 of the sequence). Although five of the putative N-linked glycosylation sites are conserved between the two proteins, the gc sequence contains two more sites than the sequence (nine versus seven total). The gc sequence contains two N-linked glycosylation sites in the 27 amino acids deleted from the sequence, and an overlapping pair of sites between residues 109 and 112 (Fig. 4). The sequence contains two N-linked glycosylation sites not found in the sequence, one of which is proximal to the amino terminus. To examine more fully the possible structural homologies between the and sequences, hydropathy analysis was performed (13). Both proteins exhibited an extraordinary degree of structural homology, based on the hydrophilic and hydrophobic properties of the amino acid sequences (Fig. 5). Each shows an N-terminal hydrophobic domain followed by a stretch of hydrophilic amino acids which contain either six of nine total () or three of seven total () putative N-linked glycosylation sites. The peaks and valleys which follow this hydrophilic region are very similar in both proteins, including the hydrophilic domain containing the final N-linked glycosylation site. The carboxy termini of both proteins show a very hydrophobic 20-residue region followed by a hydrophilic carboxy terminus. The 27 contiguous amino acids found exclusively in the gc protein appear to encode a relatively hydrophilic region between residues 50 and 76 (Fig. 5). In conclusion, this analysis reveals that the hydropathic features of both the gc and the protein are very similar and that the least conserved amino-terminal regions of the proteins are found in hydrophilic regions which have the potential to be highly glycosylated. gf SEQUENCE 161 Analysis of the second open reading frame. Translation of the final 431 base pairs of the sequence (Fig. 2) (residues 1975 to 2406) revealed a second open reading frame of 105 amino acids. Although the sequence information reported here is insufficient to contain the entire second open reading frame, comparison of this sequence with the open reading frame encoded by the 730-base mrna of reported by Frink et al. (10) also revealed a high degree of sequence homology. The two sequences showed 75% sequence homology in the overlapping regions, with -90% of the amino acid changes being conservative with respect to charge (Fig. 4B). The major difference between the two sequences was a 19-amino-acid N-terminal region which was found in the but not the sequence. Thus, although the function of the protein encoded in this region is unknown, the proteins from and show a considerable degree of sequence homology. Analysis of other open reading frames. The work of Frink et al. (10) demonstrated the existence of two other potential reading frames which overlapped the gc gene region. Analysis of the sequence determined here reveals that this sequence encodes a potential polypeptide which is homologous to the frame 3 polypeptide of the sequence. These open reading frames map approximately colinearly in the two genomes, with the open reading frame between positions 407 and 727 and the open reading frame between positions 333 and 727 (Fig. 2). Although the amino-terminal regions of the two open reading frames bear little obvious sequence homology, the carboxy halves of the proteins are -65% homologous, with the majority of the changes between the two proteins being conservative with respect to charge. In addition, analysis of the amino-terminal half of the sequences reveals that the majority of the amino acids encoded by this region are either charged residues or prolines in both proteins. Thus, although no functional or physical evidence exists for these proteins, the conservation of length, amino-terminal amino acid composition, and carboxy-terminal sequence suggest that this region of both the and genomes may encode a protein which is expressed during infection. A similar search for a homolog of the frame 1 open reading frame of the genome (10) in the sequence again revealed a homologous region. However, the protein was found to terminate after only 49 amino acids of the potential 151-amino-acid coding region. Thus, although it is possible that this shortened protein may still be functional during infection, it is more likely that either the protein is specific for or that this potential reading frame is not expressed in either virus. DISCUSSION The results reported here demonstrate that the genome encodes a colinearly mapping homolog of gc. The colinearity of the sequences found here is strengthened by the finding of a sequence 3' of the major open reading frame which apparently encodes a homolog of the 730-base-pair mrna (10). Previous mapping of the gf gene with recombinant viruses in conjunction with specific restriction enzyme fragments (32, 33) and the properties described here for the major open reading frame in this region of the genome (including several potential N-linked glycosylation sites and an apparent amino-terminal signal sequence [5] as well as a putative carboxy-terminal transmembrane domain [28]) allow the conclusion that the protein described here is the glycoprotein gf. In addition, the size of the translated HSV-

9 162 DOWBENKO AND LASKY 2 protein (-52,000 daltons) is similar to that reported for the endoglycosidase H-treated native size for gf (54,000 daltons) (32). Finally, the large extent of amino acid sequence homology as well as the conservation of several potential N-linked glycosylation sites and of all seven cysteine residues indicates structural homology between gc and gf. These results, then, confirm previous mapping studies (32, 33) and strongly suggest that the gc protein and the gf protein are homologous to each other. Our results help explain previous results which demonstrated that the gf and gc proteins were mainly type specific but did have type-common determinants (22, 24, 32, 34). Since several previous studies (8, 22, 24) demonstrated that these proteins induced predominantly type-specific antibodies, it is reasonable that the most antigenic regions of the proteins are found within the more divergent N-terminal sequences which follow the putative hydrophobic signal sequences. The hydrophilic nature of the divergent regions, along with their high content of potential N-linked glycosylation sites (14), suggests that these regions would be located on the surface of the protein. Exposure of these divergent sequences to the outside of the proteins may be responsible for the generation of type-specific antibodies directed against these divergent epitopes. However, typecommon antibodies could likely also be generated by the more highly conserved carboxy-terminal three-fourths of the proteins, since hydrophilic regions conserved between gc and gf could be exposed to the outside of the proteins and may be, in one case, glycosylated (residues 363 to 366 of gc and 330 to 332 of gf). Thus, gc and gf share both type-specific and type-common determinants, but it appears that the type-specific determinants are more antigenic. Although an explanation of the type-specific and typecommon determinants of gc and gf must at this point be speculative, it is possible that the proteins have at least two functions, one of which is important for the viability of both viruses, the type-common domain, and one of which is specific for each virus type, the type-specific domain. Although the function(s) of gc and gf is at present unknown and viable gc minus mutants of have been isolated in vitro (6), it is not clear that either gc or gf is indispensable to the viruses during in vivo infection of the human host and the establishment of latency. It is possible that at least some of the biological differences between and, including predilection for site of infection and virulence, may be due to the marked structural differences between the amino-terminal regions of gc and gf. It may be concluded, even in the absence of any functional knowledge of these proteins, that different selective pressures must be operating on the divergent and conserved domains of gc and gf. Previous sequence comparison of the gd genes of and (Lasky and Dowbenko, in press) demonstrated that the amino-terminal signal sequence (5) and the carboxyterminal transmembrane domain (27) are able to tolerate a large number of mutations as long as the substituted amino acids are hydrophobic. The gc and gf sequence comparison demonstrates a similar finding in the carboxy-terminal putative transmembrane domain (27) from residues 476 to 496 of gc and 443 to 463 of gf. The large number of heterologous hydrophobic substitutions in this region suggests that, as in gd, any amino acid which is lipid soluble can be tolerated in this region. In contrast to gd, however, the amino-terminal signal sequence of gc and gf are highly homologous in the first 19 residues. Thus, either this region has an important conserved function other than direction of the glycoproteins into the rough endoplasmic reticulum (5), or there may be an overlapping gene or other functional sequence in this region of the genome which must be conserved (12). Although insufficient sequence is presented here for a complete comparison, the region 5' to the start of HSV- 1 gc mrna transcription shows an identical CGGGTATAA sequence in both the and genomes. In addition, both sequences are followed by a G-rich region immediately preceding the start of transcription. Thus, as was previously found for the gd genes of and, upstream sequence homologies exist between the two virus types, which suggests the possibility that these regions are involved in transcriptional regulation of these genes. Interestingly, the second TATA box homology found in both virus genomes, which probably controls transcription of the 730-base mrna (10, 17), also shows a relatively high degree of sequence homology in and. These TATA boxes are preceded by CG-rich sequences which are similar, but not identical, to those preceding the first TATA regions shown in Fig. 2, and they are both followed by a 14-base-pair region showing -80% sequence homology. The entire region of homology surrounding this region is only 33 base pairs, with an overall sequence homology of -75%. If this region is involved in transcriptional regulation of the 730-base mrna, then it appears that a relatively short sequence may be sufficient for recognition by transcriptional regulatory elements. In conclusion, the results reported in this paper demonstrate that the gc and gf glycoproteins are highly homologous and that they encode type-common and type-specific domains. Since the two proteins do show significant sequence homology and since they apparently map colinearly, we favor the proposal of Zezulak and Spear (32, 33) to rename gf as gc or gc-2. In addition, the sequencing data reported here open the way for a functional analysis of the gc-1 and gc-2 proteins by the interchange of various type-specific regions between the two proteins in vitro and expression of the chimeric sequences in mammalian cells (4) or by reincorporation of these regions back into the virus (11). ACKNOWLEDGMENTS We thank D. Kleid, E. Patzer, and C. Simonsen for careful reading of the manuscript, and Jeanne Arch for excellent manuscript preparation. LITERATURE CITED J. VIROL. 1. Anderson, K. P., J. R. Stringer, L. E. Holland, and E. K. Wagner Isolation and localization of herpes simplex virus type 1 mrna. J. Virol. 30: Balachandran, N., D. Harnish, R. A. Killington, S. Bacchetti, and W. E. Rawls Monoclonal antibodies to two glycoproteins of herpes simplex virus type 2. J. Virol. 39: Balachandran, N., D. Harnish, W. E. Rawls, and S. Bacchetti Glycoproteins of herpes simplex virus type 2 as defined by monoclonal antibodies. J. Virol. 44: Berman, P., D. Dowbenko, L. Lasky, and C. Simonsen Detection of antibodies to herpes simplex virus with a continuous cell line expressing cloned glycoprotein D. Science 222: Blobel, G Intracellular protein topogenesis. Proc. Natl. Acad. Sci. U.S.A. 77: Cassai, E., R. Manservigi, A. Corallini, and M. Terni Plaque dissociation of herpes simplex viruses: biochemical and

10 VOL. 52, 1984 biological characters of viral variants. Intervirology 6: Dayhoff, M. (ed.) Atlas of protein sequence and structure, vol. 5, suppl. 2, p National Biochemical Research Foundation, Silver Spring, Md. 8. Eberle, R., and R. J. Courtney Assay for type-specific and type-common antibodies to herpes simplex virus types 1 and 2 in human sera. Infect. Immun. 31: Eisenberg, R. J., M. Ponce de Leon, L. Pereira, D. Long, and G. H. Cohen Purification of glycoprotein gd of herpes simplex virus types 1 and 2 by use of monoclonal antibody. J. Virol. 41: Frink, R. J., R. Eisenberg, G. Cohen, and E. K. Wagner Detailed analysis of the portion of the herpes simplex virus type 1 genome encoding glycoprotein C. J. Virol. 45: Gibson, M. G., and P. G. Spear Insertion mutants of herpes simplex virus have a duplication of the glycoprotein D gene and express two different forms of glycoprotein D. J. Virol. 48: Hall, L. M., K. G. Draper, R. J. Frink, R. H. Costa, and E. K. Wagner Herpes simplex virus mrna species mapping in EcoRI fragment I. J. Virol. 43: Hopp, T., and K. Woods Prediction of protein antigenic determinants from amino acid sequences. Proc. Natl. Acad. Sci. U.S.A. 78: Hubbard, S., and R. Ivatt Synthesis and processing of asparagine-linked oligosaccharides. Annu. Rev. Biochem. 50: Kleid, D., D. Yansura, B. Small, D. Dowbenko, D. Moore, M. Grubman, P. McKercher, D. Morgan, B. Robertson, and H. Bachrach Cloned viral protein vaccine for foot and mouth disease: responses in cattle and swine. Science 214: Maxam, A., and W. Gilbert Sequencing end-labeled DNA with base-specific chemical cleavages. Methods Enzymol. 65: McKnight, S., and R. Kingsbury Transcriptional control signals of a eukaryotic protein coding gene. Science 217: Messing, R., R. Crea, and P. Seeburg A system for shotgun DNA sequencing. Nucleic Acids Res. 9: Norrild, B., H. Ludwig, and R. Rott Identification of a common antigen of herpes simplex virus, bovine herpes mammillitis virus, and B virus. J. Virol. 26: Para, M. F., L. Goldstein, and P. G. Spear Similarities and differences in the Fc-binding glycoprotein (ge) of herpes simplex virus types 1 and 2 and tentative mapping of the viral gene for this glycoprotein. J. Virol. 41: Para, M. F., K. M. Zezulak, A. J. Conley, M. Weinberger, K. Snitzer, and P. G. Spear Use of monoclonal antibodies against two 75,000-molecular-weight glycoproteins specified by gf SEQUENCE 163 herpes simplex virus type 2 in glycoprotein identification and gene mapping. J. Virol. 45: Pereira, L., D. Dondero, B. Norrild, and B. Roizman Differential immunologic reactivity and processing of glycoproteins ga and gb of herpes simplex virus types 1 and 2 made in Vero and HEP-2 cels. Proc. Natl. Acad. Sci. U.S.A. 78: Pereira, L., T. Klassen, and J. R. Baringer Type-common and type-specific monoclonal antibody to herpes simplex virus type 1. Infect. Immun. 29: Powell, K., A. Buchan, C. Sim, and D. Watson Typespecific protein in herpes simplex virus envelope reacts with neutralizing antibody. Nature (London) 249: Roizman, B The structure and isomerization of herpes simplex virus genomes. Cell 16: Ruyechan, W. T., L. S. Morse, D. M. Knipe, and B. Roizman Molecular genetics of herpes simplex virus. II. Mapping of the major viral glycoproteins and of the genetic loci specifying the social behavior of infected cells. J. Virol. 29: Sabatini, D., G. Kreibach, T. Morimoto, and M. Adesnik Mechanism for the incorporation of proteins in membranes and organelles. J. Cell Biol. 92: Southern, E. M Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98: Spear, P Herpes viruses, p In H. A. Blough and J. M. Tiffaney (ed.), Cell membranes and viral envelopes, vol. 2. Academic Press, Inc., London. 30. Vieira, J., and J. Messing The puc plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19: Whitton, J., F. Rixon, A. Easton, and J. Clements Immediate-early mrna-2 of herpes simplex virus types 1 and 2 is unspliced: conserved sequences around the 5' and 3' termini correspond to transcription regulatory signals. Nucleic Acids Res. 18: Zezulak, K. M., and P. G. Spear Characterization of a herpes simplex virus type 2 75,000-molecular-weight glycoprotein antigenically related to herpes simplex virus type 1 glycoprotein C. J. Virol. 47: Zezulak, K. M., and P. G. Spear Mapping of the structural gene for the herpes simplex virus type 2 counterpart of herpes simplex virus type 1 glycoprotein C and identification of a type 2 mutant which does not express this glycoprotein. J. Virol. 49: Zweig, M., S. D. Showalter, S. V. Bladen, C. J. Heilman, Jr., and B. Hampar Herpes simplex virus type 2 glycoprotein gf and type 1 glycoprotein gc have related antigenic determinants. J. Virol. 47: