Evolution of a TRIM5-CypA splice isoform in old world monkeys

Abstract

The TRIM family proteins share a conserved arrangement of three adjacent domains, an N-terminal RING domain, followed by one or two B-boxes and a coiled-coil, which constitutes the tripartite-motif for which the family is named. However, the C-termini of TRIM proteins vary, and include at least nine evolutionarily distinct, unrelated protein domains. Antiviral restriction factor TRIM5alpha has a C-terminal B30.2/SPRY domain, which is the major determinant of viral target specificity. Here, we describe the evolution of a cyclophilin-A encoding exon downstream of the TRIM5 locus of Asian macaques. Alternative splicing gives rise to chimeric transcripts encoding the TRIM motif fused to a C-terminal CypA domain (TRIM5-CypA). We detected TRIM5-CypA chimeric transcripts in primary lymphocytes from two macaque species. These were derived in part from a CypA pseudogene in the TRIM5 locus, which is distinct from the previously described CypA insertion in TRIM5 of owl monkeys. The CypA insertion is linked to a mutation in the 3' splice site upstream of exon 7, which may prevent or reduce expression of the alpha-isoform. All pig-tailed macaques (M. nemestrina) screened were homozygous for the CypA insertion. In contrast, the CypA-containing allele was present in 17% (17/101) of rhesus macaques (M. mulatta). The block to HIV-1 infection in lymphocytes from animals bearing the TRIM5-CypA allele was weaker than that in cells from wild type animals. HIV-1 infectivity remained significantly lower than SIV infectivity, but was not rescued by treatment with cyclosporine A. Thus, unlike owl monkey TRIMCyp, expression of the macaque TRIM5-CypA isoform does not result in increased restriction of HIV-1. Despite its distinct evolutionary origin, Macaca TRIM5-CypA has a similar domain arrangement and shares approximately 80% amino-acid identity with the TRIMCyp protein of owl monkeys. The independent appearance of TRIM5-CypA chimeras in two primate lineages constitutes a remarkable example of convergent evolution. Based on the presence of the CypA insertion in separate macaque lineages, and its absence from sooty mangabeys, we estimate that the Macaca TRIM5-CypA variant appeared 5-10 million years ago in a common ancestor of the Asian macaques. Whether the formation of novel genes through alternative splicing has played a wider role in the evolution of the TRIM family remains to be investigated.

Figure 1. The macaque TRIM5 locus.

A. Schematic depiction of the primate TRIM5 locus including the seven coding exons (grey shaded regions) and introns, and the nucleotide sequence in the region of the 3′ss G/T SNP at the terminus of intron 6. Sequencing analysis confirmed that an NsiI restriction site was linked to the G/T change, and PCR amplification followed by NsiI digestion was used as an allelic discrimination assay to survey multiple individuals from two species of macaque. B. PCR/NsiI allelic discrimination in rhesus macaques (M. mulatta). C. Pedigree depicting genotype of rhesus macaque 173-02 (homozygous T/T) along with its dam (220-97; heterozygous G/T) and sire (76-99; heterozygous G/T). D. PCR+NsiI screening of sixteen Pig-tailed macaques (M. nemestrina). E. A second PCR screen of genomic DNA samples for the presence of a CypA insertion downstream of TRIM5. A gel revealing the presence of the insertion in 173-02 (T/T), 220-97(G/T), and 76-99 (G/T) (lanes 2–4) is shown. The insertion was not found in two wild type (G/G) individuals (lanes 5 and 6), but was present in two pig-tailed macaques (both T/T) (lanes 7 and 8).

Figure 2. Single-cycle infection assays.

Cells were infected at a low M.O.I., to reduce possible effects of saturation. Virus stocks were first titered by serial dilution and infection of CRFK cells (Figure S1), and equivalent infectious units of HIV-1 and SIV were used for parallel infections of macaque PBMC. All experiments were performed in triplicate. A. Infection of activated PBMC from 23 rhesus macaques, including 19 G/G, 1 T/T and 3 G/T individuals, with VSV-pseudotyped HIV-1. B. Same as in A, but using VSV-pseudotyped SIVmac239. C. Single cycle HIV and SIV infection of BLCL derived from a rhesus macaque homozygous for the TRIM5-CypA allele, in the absence and presence of cyclosporine A. D. Single cycle infectivity on immortalized BLCL lines from seven pig-tailed macaques. The BLCL line from each individual animal was tested in triplicate with each of the two viruses. Bars indicate mean infectivity +/−SEM for all seven cell lines.

Figure 3. A TRIM5-CypA transcript of macaques is generated by alternative splicing.

A. Messenger RNA sequence spanning the junction between the TRIM5 exon-6 and CypA sequences. B. Predicted amino-acid sequence of rhesus TRIM5-CypA aligned with pig-tailed macaque TRIM5-CypA and owl monkey TRIMCyp. Residues that differ from rhesus TRIM5-CypA are highlighted in blue. The region encoded by exon-7 is boxed in red; as the result of alternative splicing, this sequence is present in owl monkey TRIMCyp, but is missing from old world monkey TRIM5-CypA. C. Cartoon depicting predicted protein sequences of the old world monkey TRIM5-CypA protein (top), the owl monkey TRIM-Cyp protein (middle) and wild type primate TRIM5α (bottom).

Figure 4. 5′ junction of the inserted CypA pseudogene.

The first 56 nucleotides of the insertion are underlined. The first nucleotide of the inserted sequence is indicated with an arrow, and occurs just after nucleotide position 671,500 of rhesus macaque chromosome 14 (accession # NW_001100384, based on M. mulatta reference assembly Mmu1 01212), in or near the 3′UTR of TRIM5. The 3′ss AG dinucleotide and the first methionine codon in the CypA pseudogene are in boldface. Splicing from the end of TRIM5 exon-6 occurs 35 bases upstream of the AUG, but maintains a continuous open reading frame.