Primary structure of the herpesvirus saimiri genome.

This report describes the complete nucleotide sequence of the genome of herpesvirus saimiri, the prototype of gammaherpesvirus subgroup 2 (rhadinoviruses). The unique low-G + C-content DNA region has 112,930 bp with an average base composition of 34.5% G + C and is flanked by about 35 noncoding high-G + C-content DNA repeats of 1,444 bp (70.8% G + C) in tandem orientation. We identified 76 major open reading frames and a set of seven U-RNA genes for a total of 83 potential genes. The genes are closely arranged, with only a few regions of sizable noncoding sequences. For 60 of the predicted proteins, homologous sequences are found in other herpesviruses. Genes conserved between herpesvirus saimiri and Epstein-Barr virus (gammaherpesvirus subgroup 1) show that their genomes are generally collinear, although conserved gene blocks are separated by unique genes that appear to determine the particular phenotype of these viruses. Several deduced protein sequences of herpesvirus saimiri without counterparts in most of the other sequenced herpesviruses exhibited significant homology with cellular proteins of known function. These include thymidylate synthase, dihydrofolate reductase, complement control proteins, the cell surface antigen CD59, cyclins, and G protein-coupled receptors. Searching for functional protein motifs revealed that the virus may encode a cytosine-specific methylase and a tyrosine-specific protein kinase. Several herpesvirus saimiri genes are potential candidates to cooperate with the gene for saimiri transformation-associated protein of subgroup A (STP-A) in T-lymphocyte growth stimulation.

The genome of human herpesvirus 6: maps of unit-length and concatemeric genomes for nine restriction endonucleases.

More than 50 fragments resulting from complete digestion of the DNA of human herpesvirus 6 (HHV-6, strain U1102) with BamHI, EcoRI, HindIII, KpnI, NruI, SalI or SmaI have been isolated as clones in M13, plasmid, cosmid and lambda vectors. Using these clones, maps have been constructed for the fragments produced by nine restriction enzymes from unit-length virus genomes and from their concatemeric precursors. The unit-length genome is a linear, double-stranded molecule of 161.5 kbp composed of a central segment of a largely unique sequence of 141 kbp (U) with a sequence of 10 kbp duplicated in the same orientation at both 'left' and 'right' genomic termini (i.e. 'left' and 'right' copies of the direct repeat; DRL and DRR). Adopting as standard an orientation in which the major capsid protein gene is 'left' of the gene for alkaline exonuclease, then the 'right' genome termini and DRL. U junctions occur close to or within repetitive (GGGTTA)n sequences. Repetitions of short sequence motifs are present in at least two other regions of the genome. One of these regions consists of a simple repeat (TC/G) of approximately 1.5 kbp in length and is unstable as clones in bacterial vectors. The second region is stably maintained in such vectors and consists of a tandem array of at least 25 copies of a 110 bp sequence containing a single KpnI site. Comparisons of fragments arising from unit-length DNA with those from virus DNA from the nuclei of infected cells have shown that the concatemeric junctions in intracellular DNA contain head-to-tail dimers of the terminal duplications (i.e. ...U1.DRR1.DRL2.U2...). The gross structure established here for the genome from the U1102 isolate of HHV-6 resembles closely that suggested by Pellett and his colleagues for the Z29 isolate and differs from that of the five previously characterized human herpesviruses. This structure of HHV-6 DNA bears a superficial resemblance to that proposed for DNA from channel catfish virus and equine cytomegalovirus.

Human herpesvirus 6 is closely related to human cytomegalovirus.

A sequence of 21,858 base pairs from the genome of human herpesvirus 6 (HHV-6) strain U1102 is presented. The sequence has a mean composition of 41% G + C, and the observed frequency of CpG dinucleotides is close to that predicted from this mononucleotide composition. The sequence contains 17 complete open reading frames (ORFs) and part of another at the 5' end of the sequence. The predicted protein products of two of these ORFs have no recognizable homologs in the genomes of other sequenced human herpesviruses (i.e., Epstein-Barr virus [EBV], human cytomegalovirus [HCMV], herpes simplex virus [HSV], and varicella-zoster virus [VZV]). However, the products of nine other ORFs are clearly homologous to a set of genes that is conserved in all other sequenced herpesviruses, including homologs of the alkaline exonuclease, the phosphotransferase, the spliced ORF, and the major capsid protein genes. Measurements of similarity between these homologous sequences showed that HHV-6 is clearly most closely related to HCMV. The degree of relatedness between HHV-6 and HCMV was commensurate with that observed in comparisons between HSV and VZV or EBV and herpesvirus saimiri and significantly greater than its relatedness to EBV, HSV, or VZV. In addition, the gene for the major capsid protein and its 5' neighbor are reoriented with respect to the spliced ORFs in the genomes of both HHV-6 and HCMV relative to the organization observed in EBV, HSV, and VZV. Three ORFs in HHV-6 have recognizable homologs only in the genome of HCMV. Despite differences in gross composition and size, we conclude that the genomes of HHV-6 and HCMV are closely related.

A comparative analysis of the sequence of the thymidine kinase gene of a gammaherpesvirus, herpesvirus saimiri.

We present the nucleotide sequence of a region from the genome of the A + T-rich gammaherpesvirus, herpesvirus saimiri (HVS), which includes the coding sequences for the viral thymidine kinase (TK) gene. The organization of genes in this region resembles the homologous region of the Epstein-Barr virus (EBV) genome and is very compact, using overlapping coding sequences and with nucleotides shared by initiation and termination codons of neighbouring reading frames. The HVS TK is predicted to contain a 527 residue polypeptide with the first part of the presumptive nucleotide-binding site [(L, I, V)(F, Y)(I, L)(D, E)(G)(X)(X)(G)(L, I, V, M)(G)(K)(T, S)(T, S)] located at residues 212 to 224. This motif is close to the amino terminus of the TK polypeptides of alphaherpesviruses and the polypeptides of the cellular and poxvirus-encoded enzymes. The corresponding reading frame of the human gammaherpesvirus (EBV) also has a long amino-terminal extension but significant amino acid sequence similarities between the HVS and EBV sequences are not observed until the region of the nucleotide-binding site. Comparisons of these homologous carboxy-terminal sequences of the HVS- and EBV-encoded proteins with those from six alphaherpes viruses and proteins encoded by Marek's disease virus (MDV) and the herpesvirus of turkeys (HVT) confirm that the HVS and EBV sequences are products of a distinct lineage. The sequences of alphaherpesvirus enzymes than to those of HVS and EBV. Comparison of these 10 highly divergent TK sequences extends and refines the identification of essential features of this family of herpesvirus enzymes and defines 19 positions at which all sequences have identical residues.

Conservation of glycoprotein H (gH) in herpesviruses: nucleotide sequence of the gH gene from herpesvirus saimiri.

We present the nucleotide sequence of the glycoprotein H (gH) gene of herpesvirus saimiri (HVS), a representative of the T lymphotropic herpesviruses of New World monkeys, and compare the predicted amino acid sequence with sequences of homologous proteins from four human herpesviruses. The HVS gH gene is located within a block of genes encoding products conserved in all herpesvirus subgroups as represented by the human herpesviruses herpes simplex virus, varicella-zoster virus, cytomegalovirus and Epstein-Barr virus. In agreement with the biological grouping of HVS as a lymphotropic gammaherpesvirus, its gH amino acid sequence shows greatest similarity to that of the B lymphotropic Epstein-Barr virus, although the nucleotide sequences of their respective gH genes show little similarity given different G + C compositions of 31% and 54%. The similarity observed between the gH amino acid sequences of the two representative gammaherpesviruses is greater than that between the two human alphaherpesviruses varicella-zoster virus and herpes simplex virus. The members of the gH family range in size from 706 to 743 amino acid residues for the beta- and gammaherpesviruses, to 838 to 841 for the alphaherpesviruses, giving non-glycosylated precursors with Mr values of 78,322 to 93,651. The difference in size is due to heterogeneity in the poorly conserved N-terminal regions of the larger alphaherpesviruses compared to the smaller beta- and gammaherpesvirus molecules. Greatest similarity is observed in the C-terminal halves of the proteins including residues surrounding four conserved cysteine residues, a conserved N-linked glycosylation site (within the sequence NGTV) 13 to 18 residues proximal to the membrane-spanning sequences, and a short cytoplasmic domain of seven or eight residues for the beta- and gammaherpesviruses' and 14 or 15 residues for the alphaherpesviruses' gH. Thus, the representatives of all subgroups of herpesviruses, including those with a non-human host, encode gH homologues. Together with the observation that gH of these viruses are major targets for virus neutralization by antibody, this suggests that this glycoprotein family is essential among all herpesviruses and represents a major component involved in herpesvirus infectivity.

Conservation of sequence and function between the product of the 52-kilodalton immediate-early gene of herpesvirus saimiri and the BMLF1-encoded transcriptional effector (EB2) of Epstein-Barr virus.

We present a sequence of 2,220 nucleotides from a region of the genome of herpesvirus saimiri (HVS) which includes the coding and putative regulatory sequences for the 52-kilodalton (kDa) immediate-early (IE) phosphoprotein of the virus. The amino acid sequence predicted for this protein shows it to be homologous to the EB2 transcriptional effector encoded by the BMLF1 open reading frame of Epstein-Barr virus (EBV), the IE 68-kDa protein of varicella-zoster virus, and the IE 63-kDa (alpha 27) protein of herpes simplex virus (HSV). By measuring the function of the HVS 52-kDa-protein gene in transient expression assays, we also showed that it can substitute with comparable efficiency for the EB2 product of EBV in the EB1-dependent activation of the EBV DR promoter. The alpha 27 gene of HSV was an inefficient trans-activator in similar assays. We conclude that the IE 52-kDa protein of HVS is structurally and functionally more similar to the homologous protein of the human lymphotropic virus, EBV, than to the corresponding proteins from the neurotropic viruses, varicella-zoster virus and HSV.

Conservation of gene organization in the lymphotropic herpesviruses herpesvirus Saimiri and Epstein-Barr virus.

By analyses of short DNA sequences, we have deduced the overall arrangement of genes in the (A + T)-rich coding sequences of herpesvirus saimiri (HVS) relative to the arrangements of homologous genes in the (G + C)-rich coding sequences of the Epstein-Barr virus (EBV) genome and the (A + T)-rich sequences of the varicella-zoster virus (VZV) genome. Fragments of HVS DNA from 13 separate sites within the 111 kilobase pairs of the light DNA coding sequences of the genome were subcloned into M13 vectors, and sequences of up to 350 bases were determined from each of these sites. Amino acid sequences predicted for fragments of open reading frames defined by these sequences were compared with a library of the protein sequences of major open reading frames predicted from the complete DNA sequences of VZV and EBV. Of the 13 short amino acid sequences obtained from HVS, only 3 were recognizably homologous to proteins encoded by VZV, but all 13 HVS sequences were unambiguously homologous to gene products encoded by EBV. The HVS reading frames identified by this method included homologs of the major capsid polypeptides, glycoprotein H, the major nonstructural DNA-binding protein, thymidine kinase, and the homolog of the regulatory gene product of the BMLF1 reading frame of EBV. Locally as well as globally, the order and relative orientation of these genes resembled that of their homologs on the EBV genome. Despite the major differences in their nucleotide compositions and in the nature and arrangements of reiterated DNA sequences, the genomes of the lymphotropic herpesviruses HVS and EBV encode closely related proteins, and they share a common organization of these coding sequences which differs from that of the neurotropic herpesviruses, VZV and herpes simplex virus.