Detection of Simian Immunodeficiency Virus in Diverse Species and of Human Immunodeficiency Virus Type 2 by Using Consensus Primers within the pol Region

ABSTRACT Human immunodeficiency virus type 2 (HIV-2) is the result of cross-species transmission of simian immunodeficiency virus (SIV) from sooty mangabey monkeys to humans. Primer pairs (intHIV-2/SIV) based on a region of integrase that has considerable homology across HIV-2 and SIV lineages were designed to develop a broadly cross-reactive molecular assay to detect lentivirus infection in primates. The intHIV-2/SIV primers detect HIV-2 and simian viruses SIVcpz, SIVsmm, SIVsyk, SIVagm, and SIVmnd. The primers are also capable of amplifying some HIV-1 strains. Additionally, sequences from the integrase amplicons were of sufficient genetic diversity to permit not only phylogenetic clustering of all simian viruses to their respective lineages but also HIV type and group classification. Thus, the primers described here provide a method to detect primate lentiviruses from diverse species of nonhuman primates, as well as from persons infected with HIV-1 and HIV-2.

Zoonotic transmission is an important factor in the emergence of retroviruses and other infectious agents in humans (2,14,29,32,33). At least 20 different nonhuman primate species in Africa have been shown to harbor simian immunodeficiency viruses (SIVs) (14). Thus, African primates represent an extremely large reservoir of lentiviruses that can potentially infect other species including humans (14,29). Indeed, the extensive phylogenetic relatedness among many strains of human immunodeficiency virus type 1 (HIV-1) and HIV-2 and primate lentiviruses has elucidated the simian origin of AIDS (2,7,9,12,14,29). These studies have further established that HIV-1 infection, the main cause of the worldwide AIDS pandemic (16), was the result of cross-species transmission of SIV from chimpanzees (Pan troglodytes troglodytes) to humans (12,15,(31)(32)(33). Likewise, the HIV-2 epidemic seems to have emerged via cross-species transmission of SIV from sooty mangabey monkeys (Cercocebus sp.) (4,8,14).
Current evidence indicates that the SIV counterparts of HIV-1 and HIV-2 were introduced into the human population multiple times (at least seven transmission events have been suggested) (14,29). Yet, the HIV-1 group M viruses appear to have arisen from just one such zoonotic transmission (12). Thus, while cross-species transmission of primate retroviruses to humans occurs relatively frequently, the subsequent spread of the retroviruses in the human population is rare (14,29,32,33). Nevertheless, this scenario represents a unique opportunity to study both the emergence of new human retroviruses and the genetic diversities of these human and simian viruses. The primate lentiviruses for which full-length genomic sequences are available fall into five major equidistant phylogenetic lineages: (i) SIVcpz from chimpanzees, together with HIV-1; (ii) SIVsmm from sooty mangabey monkeys together with HIV-2; (iii) SIVagm from four species of African green monkeys; (iv) SIVsyk from Sykes monkeys; and (v) SIVmnd and SIVlhoest from mandrills and l'Hoest monkeys, respectively (3,5,7,10,(12)(13)(14)(15)31). Recently, an additional SIV (SIVcol) from Guereza colobus monkeys (Colobus guereza), representing a sixth lineage of primate lentivirus, has been identified (8). Therefore, studies are needed to determine whether transmission of simian lentiviruses other than SIVcpz and SIVsmm to humans is occurring, particularly in regions where SIV infection in nonhuman primates is highly prevalent.
These studies require the development of molecular detection assays that can detect a wide range of lentivirus infections in both nonhuman primates and humans. We have recently described serologic and molecular diagnostic assays that permit detection of highly divergent HIV-1 strains and their simian counterpart, SIVcpz (21,25,35,36). Here we report a sensitive and broadly reactive PCR-based molecular screening tool for detecting HIV-2 and SIVs from diverse species of primates. More importantly, sequences generated from these PCR products can be reliably used for phylogenetic classification within the HIV-2 and SIV lineages, thus permitting identification of the zoonotic source of infection.
PCR. DNA preparation was done by proteinase K digestion in a Tris-Triton buffer, and RNA was extracted by using the QIAamp viral RNA kit according to the manufacturer's protocol (Qiagen Inc., Valencia, Calif.). Protocols for RNA extraction and conditions for RT-PCR and PCR are described elsewhere (35,36), except that an annealing temperature of 50°C was used for both the primary and secondary intHIV-2/SIV primer sets. All reactions were carried out with appropriate negative controls to detect possible contamination.
Sequence and phylogenetic analysis. Selected nested-PCR products were purified with the QIAquick PCR purification kit (Qiagen) and sequenced with Big-dye terminators (Perkin-Elmer, Foster City, Calif.) on an automated 377 DNA sequencer (Applied Biosystems, Foster City, Calif.). Sequences were aligned with CLUSTAL W (version 1.74) after editing, and phylogenetic trees were constructed by the neighbor-joining method using the PHYLIP, version 3.5c, package.
Nucleotide sequence accession numbers. The GenBank accession numbers for the sequences obtained in this study are AF395546 to AF395571.

RESULTS AND DISCUSSION
Analysis of a cloned fragment from HIV-2 (GB122) revealed that the intHIV-2/SIV primers reliably amplified as few as 25 copies of HIV-2 per PCR (Fig. 1). Moreover, these primers also detected SIVcpz with the same sensitivity (Fig. 1). We next examined the cross-reactivity of these primers with HIV-1. Amplifications were performed with HIV-1 group M molecular clones with inputs of 100, 50, and 25 copies per PCR. All of the HIV-1 group M subtypes tested with the exception of subtypes C and D were amplified at 25 copies per reaction (Fig. 1); subtypes C and D could only be amplified at 100 copies per reaction. Thus, HIV-2, SIVcpz, and all HIV-1 subtype molecular clones were consistently amplified at 100 copies per reaction. In addition, RNA from the culture supernatant of an HIV-1 group N virus was amplified (Table 1). We next tested the amplification efficiency of the primers for 12 HIV-2 primary isolates. The intHIV-2/SIV primers amplified all samples, which included nine HIV-2 subtype A specimens, two subtype B specimens, and one subtype A/B recombinant specimen ( Table 1). As cultured isolates from most of the specimens were not available, we performed PCR amplification on DNA derived from uncultured peripheral blood mononuclear cells (PBMC). Analysis of primary PBMC from 10 HIV-2-infected individuals revealed that DNA of those from 8 was amplified by using these primers (Table 2), giving an amplification efficiency of 80%. These results are comparable to those for the primers identified in the protease region, which are highly specific for HIV-2 detection only (26).
The ( Table 2). Amplification with RNA from the same samples with protease primers, previously shown to amplify as few as 10 cDNA copies from PBMC (24,26), resulted in only 43% amplification efficiency (data not shown). Further, amplification of RNA from 34 HIV-1 plasma samples by using intHIV-2/SIV primers resulted in amplification of RNA from 19 (56%) samples. In comparison intM-Z primers, previously shown to amplify HIV-1 and SIVcpz efficiently (36), amplified RNA from all specimens (100%) ( Table 2). Previous studies have developed HIV-2-specific primers that allow detection of HIV-2 from DNA in persons dually infected with HIV-1 and HIV-2 (18,24,26). More recently, qualitative and quantitative PCR-based assays for specific detection of HIV-2 have been developed (1,6,30). The intHIV-2/SIV primers described here were designed to be as cross-reactive as possible, so that any new variants of HIV-2 or HIV-1 or crossspecies transmission of SIVs from nonhuman primates to humans could be detected.
The intHIV-2/SIV primers detected 80% (10 of 12) of the PBMC from HIV-2-infected persons but detected viremia in only 36% (5 of 14) of plasma samples. While the sensitivity of amplification from primary DNA was good, one of the limitations of the intHIV-2/SIV primers is that they are not sensitive for detection of HIV-2 plasma viremia. However, using previously characterized HIV-2 protease primers resulted in comparable efficiency of amplification of HIV-2 from viral RNA (24,26). The lower sensitivity of HIV-2 detection from plasma may be due to low copy numbers of HIV-2, and indeed several studies have established that HIV-2-infected persons have significantly lower viremia than those infected with HIV-1 (9,27,28).
Since the primers had broad specificity for detection of HIV-2, HIV-1, and SIVs, we next examined if phylogenetic analysis of amplified sequences would permit distinction of HIV-2, HIV-1, and SIV lineages. The products amplified by using intHIV-2/SIV from cultured isolates of HIV-2 and SIVs, plasma of HIV-2-infected persons, and selected HIV-1s were sequenced, and a 420-bp segment was used for phylogenetic analysis. Figure 2 depicts the relationship of HIV-2 and SIV sequences derived from the amplified integrase products. The overall topology of the tree remains the same as those previously demonstrated for full-length integrase sequences. HIV-2 and several SIVs amplified and sequenced from this study (Fig.  2) revealed distinct cluster patterns similar to those of the previously characterized reference clusters. As expected, HIV-2 and SIV from sooty mangabeys formed a monophyletic cluster. Sequences from SIVrcm, SIVmnd, SIVagm, and SIVcpz were highly divergent and represented distinct lineages (2,14). Similar phylogenetic analysis of sequences amplified from persons infected with HIV-1 group M or group O revealed clustering with their respective group M or group O sequences (Fig. 3) (20), although the subtype designation within the group M sequences was not reliable (data not shown). Thus, despite the small fragment size, the phylogenetic analysis of this region provides an adequate clustering pattern to identify the correct lineages of SIVs, as well as the HIV type and group.
The intHIV-2/SIV primers were able to amplify specimens from five major lineages of primate lentiviruses. Additionally, HIV-1 and HIV-2 sequences were correctly identified. How-      3. Phylogenetic position of HIV-1 sequences from this study (boldface and underlined) and previously reported HIV-1 and SIVcpz strains in integrase regions. The topology shows an overall branching order consistent with previously reported phylogenies for full-length sequences. Trees were derived from nucleotide sequence alignments (consensus lengths of 412 bp) using the neighbor-joining method. Horizontal branch lengths are drawn to scale, with the bar indicating 0.10 nucleotide substitution per site. Numbers at the nodes indicate the percentages of bootstrap values (out of 500) in which the cluster to the right is supported (only values 70% or above are shown).