Sequence and genomic analysis of a Rhesus macaque rhadinovirus with similarity to Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8

Abstract

We have sequenced the long unique region (LUR) and characterized the terminal repeats of the genome of a rhesus rhadinovirus (RRV), strain 17577. The LUR as sequenced is 131,364 bp in length, with a G+C content of 52.2% and a CpG ratio of 1.11. The genome codes for 79 open reading frames (ORFs), with 67 of these ORFs similar to genes found in both Kaposi's sarcoma-associated herpesvirus (KSHV) (formal name, human herpesvirus 8) and herpesvirus saimiri. Eight of the 12 unique genes show similarity to genes found in KSHV, including genes for viral interleukin-6, viral macrophage inflammatory protein, and a family of viral interferon regulatory factors (vIRFs). Genomic organization is essentially colinear with KSHV, the primary differences being the number of cytokine and IRF genes and the location of the gene for dihydrofolate reductase. Highly repetitive sequences are located in positions corresponding to repetitive sequences found in KSHV. Phylogenetic analysis of several ORFs supports the similarity between RRV and KSHV. Overall, the sequence, structural, and phylogenetic data combine to provide strong evidence that RRV 17577 is the rhesus macaque homolog of KSHV.

FIG. 1

Map of the genome of RRV 17577. The ORFs of RRV 17577 are represented by arrows indicating the direction of transcription. Divergent loci are indicated by black bars above the scale; their assignment is based on the work of Nicholas et al. (44). ORF numbers are presented above or below the appropriate ORF. ORFs resident in the divergent loci are named, as are the ORFs used for bootstrap analysis. Terminal repeat sequences are represented by the dark hatched boxes at either end of the LUR. Textured boxes within the LUR represent repeat sequences, which have been named for the appropriate divergent loci. BamHI, EcoRI, and HindIII deduced restriction maps are presented. Digest products are numbered based on the sizing presented in Table 1. Fragments marked with asterisks contain terminal repeat sequences; the sizes listed for these fragments represent the portion of the fragment in the LUR. The hash marks on the map above the repeat unit rDL-E indicate that there is some ambiguity about the length of the repeats. ORFs found in DL-D2 and DL-F are not shown on the map, pending verification of transcription from these regions.

FIG. 2

Restriction digests of RRV 17577 genomic DNA. Genomic DNA (15 μg) was precipitated and then resuspended and distributed for three restriction digests. DNA was digested overnight with 10 units of either BamHI, EcoRI, or HindIII at 37°C overnight. The digests were supplemented with 10 units of the appropriate enzyme and incubated for a further 4 h before the digest products were separated on a 0.7% agarose gel. Numbering of fragments is based on the data in Table 1. The numbers 5 + 14, 11 + 12, and 1 + 15 indicate restriction fragments that are attached to the terminal repeats. The band in the BamHI digest at 11 kb is a residual partial digestion product. The faint bands seen at 1 kb in the BamHI and EcoRI lanes are artifacts of reproduction; only the band in the HindIII lane is visible on the original photograph. Bands smaller than 1 kb were too faint to see. The absence of band 12 in the BamHI digest lane is discussed in the text. The standards used are 1-kb standards (Life Technologies); the largest band is 12 kb, and the smallest band visible in the figure is 1 kb.

FIG. 3

Comparison of gB from RRV 17577 and RRV H26-95. The deduced sequences for gB from the two strains of RRV were aligned by ClustalW. No variations occurred between the two except in the regions between residues 274 and 540. These variations in gB account for all of the variations at the protein level between the two strains.

FIG. 4

ClustalW alignment of UDG. The deduced sequences of UDG from seven viruses (Table 3) were aligned with ClustalW, version 1.4, as implemented by MacVector, version 6.01. Blosum 30 was used as the scoring matrix for the pairwise alignment and for the multiple sequence alignment. The gap introduction penalty was 10, and the gap extension penalty was 0.1. Identities, shown in white on black, are defined as columns containing the same residue in all rows.

FIG. 5

Phylogenetic trees for gammaherpesvirus proteins. Six proteins were selected for phylogenetic analysis with the Phylip programs. Proteins were aligned by ClustalW. Columns containing any gaps were removed. Unaligned N- and C-terminal residues were removed. The adjusted alignments were then used to generate 100 subsets with the program Seqboot. Trees were generated from each of these subsets with the program Protpars. A final consensus tree was generated with the program Consense. The numbers represent the number of Protpars trees, out of 100, which contained the same sequences to the right of the branch point as are found in the consensus tree. Proteins used, residues remaining after gaps were removed, and number of identities in the alignment are as follows: ssDBP, 1,072 residues, 117 identities; gB, 788 residues, 123 identities; Pol, 972 residues, 253 identities; MCP, 1,318 residues, 223 identities; Hel, 759 residues, 166 identities; UDG, 240 residues, 69 identities.

FIG. 6

Phylogenetic tree of the gammaherpesviruses. A concatenated file was made by merging all of the sequences used to generate the individual trees in Fig. 5. The concatenated file was then processed in the same fashion as described in the legend to Fig. 5. The outgroup CMV branch has been removed from the figure.

FIG. 7

Comparison of RRV vIRF R10 to two human IRFs. RRV vIRF R10 was aligned to human ICSBP (59) and ISGF3γ (57). Identities between R10 and either of the human sequences are highlighted in white on black. The conserved tryptophans found in the DNA binding domain of mammalian IRFs are marked by asterisks.

FIG. 8

Alignment of RRV vIRFs and KSHV K9. All eight RRV vIRFs and KSHV K9 were aligned by ClustalW. Identities and similarities are highlighted in white on black. Similarities are ranked as having the same residue in greater than half of the rows in a column. Tryptophans found in R8 and R12 that correspond to conserved tryptophans in mammalian IRFs are marked with asterisks.