Search

Chimpanzees and gorillas are the only nonhuman primates known to harbor viruses closely related to HIV-1. Phylogenetic analyses showed that gorillas acquired the simian immunodeficiency virus SIVgor from chimpanzees, and viruses from the SIVcpz/SIVgor lineage have been transmitted to humans on at least four occasions, leading to HIV-1 groups M, N, O, and P. To determine the geographic distribution, prevalence, and species association of SIVgor, we conducted a comprehensive molecular epidemiological survey of wild gorillas in Central Africa. Gorilla fecal samples were collected in the range of western lowland gorillas (n = 2,367) and eastern Grauer gorillas (n = 183) and tested for SIVgor antibodies and nucleic acids. SIVgor antibody-positive samples were identified at 2 sites in Cameroon, with no evidence of infection at 19 other sites, including 3 in the range of the Eastern gorillas. In Cameroon, based on DNA and microsatellite analyses of a subset of samples, we estimated the prevalence of SIVgor to be 1.6% (range, 0% to 4.6%), which is significantly lower than the prevalence of SIVcpzPtt in chimpanzees (5.9%; range, 0% to 32%). All newly identified SIVgor strains formed a monophyletic lineage within the SIVcpz radiation, closely related to HIV-1 groups O and P, and clustered according to their field site of origin. At one site, there was evidence for intergroup transmission and a high intragroup prevalence. These isolated hot spots of SIVgor-infected gorilla communities could serve as a source for human infection. The overall low prevalence and sporadic distribution of SIVgor could suggest a decline of SIVgor in wild populations, but it cannot be excluded that SIVgor is still more prevalent in other parts of the geographical range of gorillas.

Western lowland gorillas (Gorilla gorilla gorilla) are infected with a simian immunodeficiency virus (SIVgor) that is closely related to chimpanzee and human immunodeficiency viruses (SIVcpz and HIV-1, respectively) in west central Africa. Although existing data suggest a chimpanzee origin for SIVgor, a paucity of available sequences has precluded definitive conclusions. Here, we report the molecular characterization of one partial (BQ664) and three full-length (CP684, CP2135, and CP2139) SIVgor genomes amplified from fecal RNAs of wild-living gorillas at two field sites in Cameroon. Phylogenetic analyses showed that all SIVgor strains clustered together, forming a monophyletic lineage throughout their genomes. Interestingly, the closest relatives of SIVgor were not SIVcpzPtt strains from west central African chimpanzees (Pan troglodytes troglodytes) but human viruses belonging to HIV-1 group O. In trees derived from most genomic regions, SIVgor and HIV-1 group O formed a sister clade to the SIVcpzPtt lineage. However, in a tree derived from 5' pol sequences ( approximately 900 bp), SIVgor and HIV-1 group O fell within the SIVcpzPtt radiation. The latter was due to two SIVcpzPtt strains that contained mosaic pol sequences, pointing to the existence of a divergent SIVcpzPtt lineage that gave rise to SIVgor and HIV-1 group O. Gorillas appear to have acquired this lineage at least 100 to 200 years ago. To examine the biological properties of SIVgor, we synthesized a full-length provirus from fecal consensus sequences. Transfection of the resulting clone (CP2139.287) into 293T cells yielded infectious virus that replicated efficiently in both human and chimpanzee CD4(+) T cells and used CCR5 as the coreceptor for viral entry. Together, these results provide strong evidence that P. t. troglodytes apes were the source of SIVgor. These same apes may also have spawned the group O epidemic; however, the possibility that gorillas served as an intermediary host cannot be excluded.

Studies of simian immunodeficiency viruses (SIVs) in their endangered primate hosts are of obvious medical and public health importance, but technically challenging. Although SIV-specific antibodies and nucleic acids have been detected in primate fecal samples, recovery of replication-competent virus from such samples has not been achieved. Here, we report the construction of infectious molecular clones of SIVcpz from fecal viral consensus sequences. Subgenomic fragments comprising a complete provirus were amplified from fecal RNA of three wild-living chimpanzees and sequenced directly. One set of amplicons was concatenated using overlap extension PCR. The resulting clone (TAN1.24) contained intact genes and regulatory regions but was replication defective. It also differed from the fecal consensus sequence by 76 nucleotides. Stepwise elimination of all missense mutations generated several constructs with restored replication potential. The clone that yielded the most infectious virus (TAN1.910) was identical to the consensus sequence in both protein and long terminal repeat sequences. Two additional SIVcpz clones were constructed by direct synthesis of fecal consensus sequences. One of these (TAN3.1) yielded fully infectious virus, while the second one (TAN2.69) required modification at one ambiguous site in the viral pol gene for biological activity. All three reconstructed proviruses produced infectious virions that replicated in human and chimpanzee CD4(+) T cells, were CCR5 tropic, and resembled primary human immunodeficiency virus type 1 isolates in their neutralization phenotype. These results provide the first direct evidence that naturally occurring SIVcpz strains already have many of the biological properties required for persistent infection of humans, including CD4 and CCR5 dependence and neutralization resistance. Moreover, they outline a new strategy for obtaining medically important "SIV isolates" that have thus far eluded investigation. Such isolates are needed to identify viral determinants that contribute to cross-species transmission and host adaptation.

Here we describe, for the first time, recombinants between two highly divergent major groups of human immunodeficiency virus type 1 (HIV-1), M and O, within a Cameroonian woman infected with three different HIV-1 strains, a group O virus, a subtype D virus, and a recently reported IBNG (A/G)-like recombinant virus. Using nested extra-long PCR amplification, we sequenced from the pol region to the env region including accessory genes of the viral genome obtained from the patient's uncultured peripheral blood mononuclear cells and examined the phylogenetic position of each gene. Compared with sequential blood samples obtained in 1995 and 1996, there were multiple segmental exchanges between three HIV-1 strains (O, D, and IBNG) and all the recombinants appeared to be derived from a common M/O ancestor. Importantly, recombination between groups M and O occurred, even though the homology between these two groups is 69, 76, 68, and 55% in the gag, pol, vif-vpr, and env regions, respectively. Recombination between strains with such distant lineages may contribute substantially to generating new HIV-1 variants.