In order to better understand how SIV is transmitted Takehisa et al. (2009) undertook several experiments to determine the phylogenetic relationship between SIVgor and SIVcpz. These experiments depended primarily on sequence homology comparisons, a commonly-used and well-accepted approach for determining phylogenetic relationships.
The specific aims are as follows:
Whole genome sequence homology comparisons will be performed between different strains of SIVgor, SIVcpz, and HIV-1, to establish the relative similarity and thus reveal phylogenetic relationships.
SIVgor sequence will be examined for evidence of recombination. Should a recombination signature be found, it can be used to help search for the most recent common ancestor.
There are multiple strains of SIVcpz infecting chimpanzees from central and eastern Africa, and previous research has shown these strains are geographically-specific. SIVgor sequence comparisons with SIVcpz strains isolated from wild chimpanzee populations will be used to determine whether the SIVgor strains arose through local inter-species transmissions or were derived from a common ancestor.
4. A full-length SIVgor consensus sequence will be assembled and used to produce an infection-competent SiVgor virus. This laboratory virus will then be characterized in terms of its requirements for infecting CD4 T cells.
Materials and Methods
SIVgor RNA was derived from fecal samples collected from wild western gorillas, and then subjected to reverse transcriptase and PCR amplification. The amplified products were sequenced and aligned to generate a full-length consensus sequences representing the four different strains. There was no mention of whether multiple independent reverse transcriptase and PCR amplifications were used to control for mistakes in these laboratory procedures. PCR mistakes are common, depending on which thermostable polymerase is used.
A full-length copy of the SIVgor genome was assembled using multiple overlapping PCR fragments, from multiple PCR amplifications. The sequence of the full length clone was validated using previous SIV and strain-specific primers. This is a standard approach used by laboratories around the world. Direct cloning of isolated RNA molecules could be attempted, but this approach is much more labor intensive and likely to fail. Given the urgency surrounding HIV research efforts, the use of PCR is reasonable.
To produce infectious viruses, the full-length clone was amplified in a bacterial host and then transfected into 293T cells. Infectivity of SIVgor and SIV strains from chimpanzees were assessed using a HeLa cell derivative genetically-modified to express CD4, CCR5, and CXCR4 and containing the luciferase and ?-galactosidase reporter genes under the control of an HIV-1 LTR element. Assays depended on lysing the cells and measuring reporter gene activity. Infectivity of primary chimpanzee and human CD4 T cells isolated from peripheral blood mononuclear cells was also determined. Viral activity was measured by assaying viral reverse transcriptase activity in the cell culture supernatants every three days. Blocking assays involved incubating viral particles with various inhibitors, soluble CD4, and blocking antibodies in the presence of the reporter cell line. This approach seems straight forward and the inclusion of primary CD4 T cells helps to validate the results of in vitro assays relying on genetically-modified cell lines.
Consensus and strain-specific primers were able to amplify three strains derived from the fecal matter recovered from the same region in western Africa. A fourth strain (BQ664), which was derived from gorilla fecal matter located 400 km away, did not amplify as easily. It's possible that the sequence of the fourth strain was too degraded for RT/PCR to be effective, or the PCR primer annealing site had diverged. Aligning the sequences revealed significant differences, including altered amino acid coding sequence, but none of the substitutions introduced a stop or frameshift codon. The authors could have pursued different primer approaches for both the reverse transcriptase reaction and the PCR, such as using random hexamers for primers. The sequence so obtained could be aligned to produce a consensus sequence.
The three SIVgor clones from which full-length sequence was derived varied in length by as much as 109 nt. All three contained nine full-length open reading frames and the predicted conserved cis-regulatory elements. Deduced amino acid sequences for the nine proteins revealed significant divergence, but no obvious SIVgor-specific signatures. Programs designed to predict protein function found strong similarities between the SIVgor-predicted functional sites and those found for SIVcpz and HIV-1 proteins, suggesting function has been conserved despite significant sequence divergence. The homology between SIVgor, and SIVcpz or HIV-1, varied depending on which sequence was used in the comparison.
The phylogeny of the SIVgor strains revealed a close, monophyletic relationship within this cluster. The BQ664 strain, which was found at a location distant from the other three strains, proved to be not as closely related to the other three, but still more closely related to other SIVgor strains than HIV-1 or SIVcpz. Comparison of sequences coding for Gag, Pol1, Pol2, and Env proteins, revealed that all four strains were found to be more closely related to HIV-1 (O group) than SIVcpz. A comparison of the SiVgor/HIV-1 to the two geographically-distinct SIVcpz strains, SIVcpzPts and SIVcpzPtt, suggests the SIVgor strains were derived from chimpanzees within the same geographic location as the western gorillas (SIVcpzPtt).
A scan for signatures of recombination found the N-terminal region of Pol had diverged significantly from SIVcpz. This result can be explained in two ways, either the SIVgor common ancestor experienced a recombination event or an as yet unidentified SIVcpz strain went through a recombination event. Through a process of comparing the evolutionary distance using the sequence for Gag and Pol, between different SIVcpz strains and SIVgor, it was found that the SIVcpzPtt strains MT145 and GAB2 were probably the source of the recombination event. This result suggests that the inter-species transmission of SIVcpz that resulted in the emergence of SIVgor occurred through these strains.
The monophyletic grouping of the SIVgor strains based on sequence comparisons suggests intra-species transmission occurred from a single event. The possibility that the SIVgor strains arose from multiple local transmission events, while unlikely, could still explain the above findings. To investigate this possibility further samples of chimpanzee fecal matter were collected from the site where the three closely SIVgor strains were found. Of the 77 chimpanzee samples, 9 contain SIVcpz. Sequence comparisons revealed the two chimpanzee sequences that were eventually amplified and sequenced were from two distinct SIVcpz strains distantly related to SIVgor. These results suggest that SIVgor was not derived from these strains and therefore argues against local transmission and for a single inter-species transmission event.
The most recent common ancestor for HIV-1 group O. has been dated to 1920, with a confidence interval between 1896 and 1942. The rate of substitution was calculated to be 1.05 x 10-3 per year. Applying these findings to SIVgor divergence predicts divergence from the most recent common ancestor occurred around 1864, approximately 60 years before HIV-1.
The infectious potential of the laboratory-synthesized SIVgor virus was comparable to SIVcpz. When incubated in the presence of inhibitors, entry into the cell was found to occur via the CCR5 receptor and required CD4 expression on the surface of the cell.
The authors concluded that the recombination signature they identified in the Pol coding sequence allowed them to determine that SIVcpzPtt was the most likely ancestral lineage for both SIVgor and HIV-1 group O, which suggests inter-species transmission that gave rise to the current strains of SIVgor occurred between chimpanzees and gorillas sharing the same geographic location approximately 100-200 years ago. Additional support for this conclusion came from finding the same recombinant pol sequence present in two SIVcpz strains.
What remains unclear from these results is whether HIV-1 was derived from gorillas or chimpanzees, since a closely-related lineage has yet to be found in the extant gorilla and chimpanzee populations. The isolation and sequencing of distantly related SIVcpz strains within the same geographic location where the three full-length SIVgor strains were found, suggests the HIV-1 primate reservoir lies elsewhere.
Characterization of the intact SIVgor virus infectivity revealed traits similar to primary HIV-1 and SIVcpz, and its reliance on CD4 and CCR5 for cell entry supports this conclusion.
While all of the above conclusions may eventually turn out to be true, there is considerable weakness underlying the findings due to the low number of strains analyzed and the limited sampling of geographic locations. I'm sure the authors are aware of the need for further research and that their conclusions may turn out to be wrong, but in the discussion the authors seem to make predictions unwarranted by the strength of their data.
The primary limitation of the findings is the low number of SIVgor strains successfully amplified, which precludes having a high degree of confidence in the intra-strain and inter-lineage phylogenetic analysis conducted here. The proper controls for reverse transcriptase and PCR amplifications were not mentioned in the Methods section, and therefore consensus sequence information for the different strains could contain mistakes.
By the end of the paper it seemed clear that SIVgor was more closely related to HIV-1 than SIVcpz, but this finding seemed to be…