|Home | About | Journals | Submit | Contact Us | Français|
Human immunodeficiency virus type 1 (HIV-1) subtype C is responsible for more than 56% of all infections in the HIV and AIDS pandemic. It is the predominant subtype in the rapidly expanding epidemic in southern Africa. To develop a relevant model that would facilitate studies of transmission, pathogenesis, and vaccine development for this subtype, we generated SHIVMJ4, a simian/human immunodeficiency virus (SHIV) chimera based on HIV-1 subtype C. SHIVMJ4 contains the majority of env, the entire second exon of tat, and a partial sequence of the second exon of rev, all derived from a CCR5-tropic, primary isolate envelope clone from southern Africa. SHIVMJ4 replicated efficiently in human, rhesus, and pig-tailed macaque peripheral blood mononuclear cells (PBMCs) in vitro but not in CEMx174 cells. To assess in vivo infectivity, SHIVMJ4 was intravenously inoculated into four rhesus macaques (Macaca mulatta). All four animals became infected as determined through virus isolation, PCR analysis, and viral loads of 107 to 108 copies of viral RNA per ml of plasma during the primary infection phase. We have established a CCR5-tropic SHIVMJ4/rhesus macaque model that may be useful in the studies of HIV-1 subtype C immunology and biology and may also facilitate the evaluation of vaccines to control the spread of HIV-1 subtype C in southern Africa and elsewhere.
Genetic variation among isolates is a potential obstacle to the development of an effective human immunodeficiency virus type 1 (HIV-1) vaccine. Multiple genotypes of HIV-1 exist within and between individuals, and these may display distinct properties of cytopathogenicity, host cell range, and replicative capacity (21, 45). HIV-1 strains have been classified phylogenetically into groups M (major), O (outlier), and N (non-M, non-O) (43). The groups have been proposed to represent three independent zoonotic transmissions from apes to humans (22). Viruses in the M group represent the majority of circulating strains in the global epidemic and are subdivided into genetic subtypes or clades designated A, A2, B, C, D, F1, F2, G, H, J, and K and several circulating recombinant forms (43). The global distribution of these subtypes is uneven, and amino acid differences between them can reach up to 24% in Env (23, 48; W. Janssens, A. Buve, and J. N. Nkengasong, Editorial, AIDS 11:705–712, 1997). This level of diversity raises the possibility of functional or biological differences between subtypes. A limited number of epidemiological studies have suggested that HIV-1 subtypes may differ in selected properties. In Senegal, West Africa, women infected with non-A subtypes had more rapid progression to disease than did those infected with subtype A (29). In a cross-sectional study in Nairobi, Kenya, women infected with HIV-1 subtype C had higher plasma RNA levels and significantly lower CD4 counts than did those infected with either subtype A or D (38). In Tanzania, HIV-1 genotypes were associated with differential probabilities of perinatal transmission; mothers infected with subtypes A, C, and intersubtype recombinants had higher odds of transmitting virus to their infants than did those infected with subtype D (42).
HIV-1 subtype C is estimated to comprise more than 50% of all infections in the pandemic (8, 18, 19, 50) and is the predominant subtype in southern Africa and parts of Asia, where the HIV and AIDS epidemic is growing fastest (26, 33, 40, 50; Janssens et al., Editorial). HIV-1 strains isolated from individuals soon after infection display the slow/low, non-syncytium-inducing phenotype and use CCR5 as the coreceptor for entry into cells (2, 12, 14, 17, 44, 52). Variants that utilize other chemokine receptors—typically CXCR4—and display the rapid/high phenotype develop later during HIV infection in many individuals infected with subtypes A, B, and D (13, 20, 49, 51). This switch in coreceptor use, which is governed by genetic changes in env, has been reported as being uncommon for subtype C viruses, even among symptomatic AIDS patients (6, 9, 40, 49). The failure of HIV-1 subtype C to undergo coreceptor utilization switch during the course of infection is consistent with the observation that HIV-1 subtype C V3 loop sequences show more genetic homogeneity than do those of other HIV-1 subtypes (15, 31).
The simian/human immunodeficiency virus (SHIV) infection of rhesus macaques has been an important model to study the role of various HIV-1 genes in transmission and pathogenesis and to test vaccine immunogens. The majority of SHIV clones have been constructed by replacing the envelope gene region of the simian immunodeficiency virus (SIV) pathogenic clone SIVmac239 with the counterpart from HIV-1 subtype B (28, 32, 35, 41, 46). The available SHIV chimeras fail to reflect the genetic diversity of the HIV-1 epidemic, which is dominated by non-B HIV-1 subtypes. Only one HIV-1 subtype C envelope-based SHIV chimera, designated SHIVCHN19, has been reported (10). SHIVCHN19 replicated in pig-tailed macaque (Macaca nemestrina) peripheral blood mononuclear cells (PBMCs) but failed to replicate in rhesus macaque PBMCs in vitro, even on CD8+ cell depletion. Furthermore, SHIVCHN19 requires the full tat and rev genes to be from the parental subtype B clone for replication.
In this paper we report the construction of SHIVMJ4, a chimera that contains a primary isolate HIV-1 subtype C envelope from Botswana, southern Africa. The majority of env, the entire second exon of tat, and part of the second exon of rev of SHIVMJ4 are all derived from HIV-1 subtype C. SHIVMJ4 replicates efficiently in human, pig-tailed macaque, and rhesus macaque PBMCs. Intravenous inoculation of SHIVMJ4 stock into four rhesus macaques resulted in productive infection as determined by virus isolation, positive proviral DNA PCR, and plasma viral RNA load quantification. The SHIVMJ4-macaque model may facilitate studies of HIV and AIDS transmission and pathogenesis for HIV-1 subtype C as well as evaluations of vaccine preparations against this predominant subtype.
A subtype B env-based SHIV molecular clone designated SHIV-89.6PD-MC1 was used as the molecular backbone for the insertion of the subtype C envelope. SHIV-89.6PD is a plasma-derived pathogenic SHIV stock that was generated by the in vivo passage of SHIV-89.6 (34). To identify genetic changes that may be responsible for the increased virulence of SHIV-89.6PD over SHIV-89.6, genomic DNA was isolated from SHIV-89.6PD-infected CEMx174 cells using the TurboGen genomic DNA purification kit (Invitrogen, Carlsbad, California). PCR amplification was performed on the genomic DNA to amplify a 500-bp fragment in the env/nef region of SHIV-89.6PD using the primers SHIV8440 (5′-GAGAGAGACAGAGACAGATCC-3′) and SHIV9565 (5′-TCTCTCTTCAGCTGGGTTTCT-3′). We focused on this region because studies have suggested that genetic changes in this region may be responsible for increased pathogenicity of SHIV and SIV (30, 47). The PCR products were then cloned into pCRII (Invitrogen) and sequenced manually. Sequence analysis of five clones of the env/nef amplicons revealed a 144-bp deletion in the HIV-1 env gene located immediately upstream of the SIV nef gene (Fig. (Fig.1).1). Such a deletion not only preserves the initiation codon of SIV nef but also creates an in-frame fusion between the preceding HIV-1 env gene and the sequences of the SIV env gene that overlap with the SIV nef gene.
To create the SHIV-89.6PD-MC1 molecular clone, the sequenced env/nef fragment described above was exchanged with the counterpart region of SHIV-89.6 by using BamHI and NcoI, which are located immediately downstream and upstream of the 8440 and 9565 primers, respectively. SHIV-89.6PD-MC1 therefore has a hybrid envelope glycoprotein in which 48 amino acids of the C terminus of HIV-1 gp41 are replaced by 53 amino acid residues from the C terminus of SIV gp41. When transfected into CEMx174 cells, SHIV-89.6PD-MC1 replicates to higher titers than SHIV-89.6 does (data not shown).
The SHIV-89.6PD-MC1 molecular clone was used as the backbone for insertion of the subtype C envelope. SHIV-89.6PD-MC1 exists in two different plasmids (Fig. (Fig.2).2). The 5′ proviral clone consists entirely of sequences derived from the SIVmac239 clone. No genetic alteration was carried out on this plasmid. To replace the subtype B env sequence in SHIV-89.6PD-MC1 with a subtype C envelope, we digested the 3′ proviral clone with KpnI (nucleotide 5925 by HXBc2 numbering) and BamHI (nucleotide 8053). The counterpart envelope fragment of HIV-1 subtype C, MOLE1, previously shown to be functional and to mediate cell entry by the coreceptor CCR5 (36, 37) was used to replace the 89.6PD KpnI-BamHI fragment. The inserted subtype C envelope was then sequenced and found to be identical to the parent clone.
293T cells were propagated in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum (FCS). 293T cells were cultured in six-well flat-bottom plates in 2,000 μl of culture medium. CEMx174 cells were grown in RPMI 1640 supplemented with 10% FCS. Human PBMCs were obtained from HIV-1-negative donors and separated by density gradient centrifugation on lymphocyte separation medium (LSM) (ICN Biomedical, Corp., Aurora, Ohio). The PBMCs were cultured for 72 h prior to infection in RPMI 1640 supplemented with 10% FCS, 5 μg of phytohemagglutinin (PHA; Sigma, St Louis, Mo.) per ml, and 20 U of interleukin-2 (Becton Dickinson Labware, Bedford, Mass.) per ml. Just before infection, the PBMCs were washed and then maintained throughout the infection period in RPMI 1640 supplemented with 10% FCS and 20 U of interleukin-2 per ml.
Rhesus and pig-tailed monkey blood was purchased from Tulane Regional Primate Research Center (New Orleans, La.). PBMCs were separated by density gradient centrifugation on LSM. Monkey PBMCs were propagated overnight prior to infection in RPMI 1640 supplemented with 10% FCS, 5 μg of concanavalin A (ConA; Sigma) per ml, and 20 U of interleukin-2 per ml. Just before infection, the cells were washed and resuspended in RPMI 1640 with interleukin-2 but without ConA. Typically, sampling was done and half the medium was replaced from infected cells every 4 days.
To determine whether the SHIVMJ4 construct yields infectious progeny virions, we linearized the 5′ proviral construct with SphI and ClaI. The SphI site is unique in the SIVmac239 genome, while ClaI is located in the cellular flanking sequences. The 3′ clone was digested with SphI and ApaI. DNA was transfected after overnight ligation of the linearized 5′ and 3′ SHIV plasmids. CEMx174 cells were transfected with 5 μg of total DNA using the Fugene 6 transfection reagent (Boehringer Mannheim, Indianapolis, Ind.). CEMx174 cells were split 1:3 on days 7 and 14. SHIV-89.6PD-MC1, SHIV-HXBc2 (32), and SHIV-89.6 (41) were used as the positive controls.
The alternative approach was to transfect the ligated SHIV DNA into subconfluent 293T cells. At 72 h posttransfection, 2 × 106 PHA-stimulated human PBMCs were added to the 293T cells. 293T and human PBMCs were cocultured for an additional 72 h. Nonadherent cells (mostly PBMCs) were then aspirated and cultured separately for an additional 21 days, while the adherent 293T cells were discarded. Infection of the target cells was monitored for up to 14 days by quantification of the viral core protein in tissue culture supernatant using the SIV p27 antigen capture enzyme-linked immunosorbent assay (ELISA) kit (Coulter Beckman, Miami, Fla.).
Culture supernatants were filtered through 0.45-μm-pore-size filter units (Nalgene, Rochester, N.Y.), and SHIV virions were quantified by p27 antigen ELISA. Equivalent amounts of virus (as determined by the amount of p27 antigen in culture supernatant) were used to infect 2 × 107 PHA- or ConA-stimulated human or monkey PBMCs. At 24 h post infection, the cells were washed three times with phosphate-buffered saline and fresh medium was then added. Infection of the target cells was monitored for up to 14 days by p27 ELISA.
To generate viral stock for animal inoculation, 2 × 107 rhesus macaque PBMCs were infected with virus equivalent to 5 ng of total p27 antigen. The cells were washed the following day, and fresh medium was added. Tissue culture supernatants were pooled from days 8 and 10 postinfection and frozen at −80°C as viral stock. The 50% tissue culture infective dose of this viral stock in human PBMCs was determined by end-point dilution with microtiter plates.
Four colony-bred young-adult male rhesus macaques free of simian type D retroviruses, SIV, HIV-2, and simian T-lymphotropic virus were housed at the California Regional Primate Research Center, Davis, Calif. in accordance with American Association for Accreditation of Laboratory Animal Care standards. Each animal was inoculated intravenously with a cell-free stock of SHIVMJ4 that was equivalent to 200 50% tissue culture infective doses. The animals were physically examined and monitored regularly for opportunistic infections and other clinical signs of disease by the California Regional Primate Research Center veterinary staff.
Blood samples were drawn from infected animals at 1, 2, 4, 6, 8, 12, and 16 weeks postinoculation. PBMCs were separated from plasma by centrifugation on LSM. The plasma viral load assay was measured by the branched-DNA assay (Bayer Diagnostics, Emeryville, Calif.). The amount of p27 antigen in plasma was determined by ELISA. To isolate virus from PBMCs of infected animals, separated monkey PBMCs were cocultured with PHA-stimulated human PBMCs. Samples were taken and half of the culture medium was replaced every 3 to 4 days. Virus isolation from the infected monkeys was assessed by detection of p27 antigen in human-monkey PBMC culture supernatants using the p27 antigen ELISA kit.
Nested PCR was carried out on genomic DNA from PBMCs by using SIV gag-specific primer pairs as previously described (27). The genomic DNA, isolated using the QIAmp DNA isolation kit (Qiagen, Chatsworth, Calif.), was quantitated by spectrophotometry, and 0.6 μg of DNA was used for PCR.
Culture supernatants from 293T cells transfected with a full-length HIV-1 clone expressing subtype C envelope (C.96BW15) were centrifuged at 800 × g for 15 min. Supernatants were then filtered through 0.45-μm-pore-size filter units, overlaid on a 4-ml 20% sucrose cushion, and centrifuged at 20,000 rpm (Beckman SW28 rotor) for 2 h. The supernatant was then discarded, and any remaining fluid was removed using cotton swabs. The viral pellet was resuspended in 200 μl of lysis buffer (0.15 M NaCl, 0.05 M Tris HCl, [pH 7.2], 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS]). The recovered protein was then mixed with reducing buffer (0.08 M Tris-HCl [pH 6.8], 0.1 M dithiothreitol, 2% SDS, 10% glycerol, 0.2% bromophenol blue), boiled for 3 min, and resolved on an SDS–12% polyacrylamide gel.
Resolved proteins were transferred onto nitrocellulose membranes (Schleicher & Schuell, Keene, N.H.) by electroblotting. Viral proteins were visualized by immunoblotting with a 1:100 dilution of plasma obtained from the SHIV-infected animals at various time points. A positive control was run with sera from an HIV-1 subtype C-infected person.
To construct SHIVMJ4, the 3′ plasmid of SHIV-89.6PD-MC1 was genetically altered (Fig. (Fig.2).2). The resultant 3′ proviral clone (p3′SHIV/MOLE1) contains vpu, exon 1 of tat, exon 1 of rev, and the 5′ signal peptide region of env (39 amino-terminal residues) that are derived from HIV-1 subtype B HXBc2 clone. The majority of the env sequence, including the transmembrane region of gp41, the entire exon 2 of tat, and part of exon 2 of rev, is derived from pSVIII/MOLE1 plasmid that encodes a functional subtype C envelope clone (36, 37). Similar to the parental clone SHIV-89.6PD-MC1, the 53 amino acid residues of the gp41 carboxyl terminus of SHIVMJ4 are derived from the SIVmac239.
To determine if SHIVMJ4 generates infectious viral progeny, we initially transfected CEMx174 cells. These cells were first tested because they traditionally yield high titers of SHIV stock in vitro. SHIVMJ4 failed to replicate in CEMx174 cells, while the CXCR4-utilizing SHIV-HXBc2 and the dualtropic SHIV-89.6 and 89.6-PD-MC1 all replicated efficiently in these cells (data not shown). We hypothesized that because the SHIVMJ4 envelope had previously been shown to be functional (36, 37), the inability of SHIVMJ4 to replicate in CEMx174 cells might be cell type specific and not necessarily due to an intrinsic noninfectious phenotype of this SHIV clone.
To test SHIVMJ4 replication in primary cells, we transfected DNA into 293T cells followed by coculture, 72 h later, with previously PHA-stimulated human PBMCs. This protocol resulted in efficient replication of all the tested viral clones (data not shown). It is possible that 293T cells were not completely eliminated from the PBMC culture and therefore continued to shed virus into the culture supernatant. To rule out this possibility, we filtered supernatants from the initial PBMC coculture through 0.45-μm-pore-size filter units, quantified the virus by p27 antigen ELISA, and used the supernatant to infect fresh human PBMCs. Cell-free supernatants for all the SHIV clones tested were able to productively infect freshly stimulated PBMCs (Fig. (Fig.3).3).
Two species of macaque monkeys, rhesus and pig-tailed, are commonly used as nonhuman primate models of HIV and AIDS. Some SHIV chimeras have been shown to replicate to persistently high titers in these animals and cause an AIDS-like syndrome. We investigated whether SHIVMJ4 would replicate in PBMCs from rhesus or pig-tailed macaques. SHIVMJ4 was able to replicate efficiently in both rhesus and pig-tailed macaque PBMCs (Fig. (Fig.4).4). Supernatants from days 8 and 10 of the rhesus macaque PBMC culture were pooled to make up the stock for animal inoculation.
Rhesus macaques are the most commonly used nonhuman primates for HIV studies. We therefore attempted infection of rhesus macaques with SHIVMJ4 since we had established that their PBMCs could be productively infected in vitro. Four juvenile male rhesus macaques, designated 29634, 29744, 30236, and 30301, were inoculated intravenously with SHIVMJ4 stock and monitored for infection over a 16-week period. Blood samples were taken from the inoculated animals at 1, 2, 4, 6, 8, 12, and 16 weeks. Cocultivation of PHA-stimulated human PBMCs with monkey PBMCs obtained at 2, 4, 6, 8, 12, and 16 weeks postinoculation resulted in productive viral replication (Table (Table1).1). Replication at 2 weeks postinoculation was more rapid, with p27 in culture supernatant reaching a peak concentration faster (day 10) for all four macaques (data not shown). The peak p27 concentration for the 1-, 4-, 6-, 8-, 12-, and 16-week-postinoculation samples was on day 14 for all four animals (data not shown). Plasma p27 levels were detectable for macaques 29634 and 29744 at 2 weeks postinfection, and for macaques 30236 and 30301 at 1 and 2 weeks postinfection but were below the detection limit in all subsequent postinoculation bleeds (Table (Table1).1). PCR of genomic DNA using gag primers was positive at all time points tested postinoculation (Table (Table1).1).
The plasma viral load was determined using the branched-DNA assay. All four macaques showed robust viral replication during the first 2 weeks postinoculation, and viral loads peaked at 22 × 106 to 62 × 106 viral RNA copies/ml (Table (Table2).2). Samples taken at subsequent weeks showed a fall in the viral burden, with the load falling by at least 3 log units from the peak in all animals by week 16 postinfection.
In all four animals, CD4 cell counts remained fairly stable, with no dramatic drop observed over the 16-week infection period (Table (Table3).3). There were also no notable changes observed in CD4/CD8 ratios over this period. So far, the animals have maintained their weight and appetite and have generally been scored as healthy by veterinary staff with no signs of clinical AIDS (data not shown).
We assessed macaques 29634 and 29744 for measurement of antibody responses to the HIV-1 subtype C envelope glycoprotein by Western blot analysis using a viral lysate from an HIV-1 molecular clone. Macaque 29744 appeared to show weak anti-envelope antibody reactivity 2 weeks postinoculation (Fig. (Fig.5).5). Both animals were definitively positive for anti-gp41, gp120, and gp160 antibodies by 6 weeks postinoculation.
We have constructed SHIVMJ4, a chimera that bears the majority of env, the entire exon 2 of tat, and a portion of exon 2 of rev from HIV-1 subtype C. SHIVMJ4 is the first SHIV construct to include sequences other than env from HIV-1 subtype C. Tat and Rev are early regulatory proteins of HIV-1 that are frequently targeted by helper and cytotoxic T-lymphocyte (CTL) responses in HIV-1-infected individuals (1, 7). SHIVMJ4 may therefore be a useful model to study anti-Tat and -Rev immune responses in HIV-1 subtype C infections and to evaluate the effectiveness of vaccine preparations that include Tat and Rev.
SHIVMJ4 replicates efficiently in human, rhesus, and pig-tailed macaque PBMCs but not in CEMx174 cells. This is the first report of an HIV-1 subtype C-based SHIV molecular clone that can replicate in vitro in rhesus macaque PBMCs. SHIVMJ4 may therefore allow in vitro studies of HIV-1 subtype C using rhesus PBMCs. We did not investigate the biological reasons for the failure of SHIVMJ4 to replicate after transfection into CEMx174 cells. This issue may require further studies because this cell line is commonly used to propagate SHIV clones and isolates in tissue culture.
Intravenous inoculation of SHIVMJ4 stock into rhesus macaques resulted in productive infection. In all four rhesus macaques, SHIVMJ4 isolates obtained 2 weeks postinoculation replicated faster to peak titers than did isolates from subsequent weeks. This faster replication at 2 weeks postinfection may reflect the possibility that a higher percentage of separated monkey PBMCs introduced into culture harbor infectious provirus, consistent with the viral load peak at 2 weeks postinoculation. Residual monkey immune responses at subsequent weeks postinoculation, when the animals may have started developing antiviral immune responses, might also play a role. Differences in viral isolate replication at various times postinoculation may also reflect phenotypic differences in PBMCs due to their being obtained from different donors.
It is important to note that during the 16-week observation period, none of the four animals experienced a dramatic decline in the CD4 lymphocyte count. This result is not totally unexpected because the SHIVs that have been associated with a rapid depletion in CD4 cell count and with disease either are dualtropic or use CXCR4 as the coreceptor for cell entry (28, 30). In contrast, previous studies showed that the envelope in SHIVMJ4 used only CCR5 and not CCR1, CCR2(b), CCR3, or CXCR4 as a coreceptor for cell entry (36, 37). It will be interesting to see whether the infected animals develop CD4 depletion and clinical AIDS-like disease on long-term follow-up. As expected, there was a primary surge of viral replication, which reached a peak at 2 weeks postinoculation as determined by the plasma HIV-1 RNA load. The load subsided over the next several weeks. Antibody responses to HIV-1 envelope could be clearly detected by 6 weeks post inoculation in the 2 animals that were tested. Antibodies to the p27 Gag protein were detectable by 4 weeks postinoculation (data not shown).
A comparison between SHIVMJ4 and SHIVCHN19, a previously described HIV-1 subtype C-based SHIV chimera (10), is shown in Fig. Fig.6.6. SHIVCHN19 required that both exons of tat and rev be derived from HIV-1 subtype B for it to be infectious. It was hypothesized that there may be a functional incompatibility of tat or rev of subtype C strains on SIV-responsive elements or a less efficient interaction between subtype C gp41 and SIV matrix protein during viral packaging. It was suggested that this effect might be subtype C specific because other subtype C-based SHIV constructs generated using Indian isolates showed a similar phenotype (10). Our study suggests that at least parts of tat and rev can be derived from HIV-1 subtype C. A significant difference between SHIVMJ4 and SHIVCHN19 is that in SHIVMJ4, the carboxyl terminus of the envelope is derived from SIV. Whether or not this genetic change is responsible for the enhanced replication of SHIVMJ4 in rhesus macaque cells will require further investigation, but it has been shown that amino acid changes in the carboxyl terminus of SHIV-89.6 may be responsible for its increased pathogenicity after in vivo passage (30).
We have previously shown by the use of full-length clones that HIV-1 subtype C from Botswana displayed greater average variability across the entire genome than did HIV-1 subtype B viruses (39). This is in contrast to subtype C sequences from Asia, which show tight phylogenetic clustering (33). The unprecedented diversity of southern African isolates raises the possibility that vaccinating against these viruses will be a greater challenge if we assume that sequence relatedness between an immunogen and the circulating viruses improves the efficacy of the immunogen. SHIVMJ4 may serve as a prototype challenge virus in vaccine studies that explore how intersubtype and intrasubtype C diversity might affect the effectiveness of potential envelope-based vaccines.
It has been reported that although cross-clade CTL recognition is widespread, recognition can also be highly clade specific (16). It has also been demonstrated that the CTL epitope recognition pattern appears to differ between Africans infected with HIV-1 subtype C and Caucasians infected with subtype B (24). It has been demonstrated that the magnitude and quality of the immune response to HIV-1 can be improved by combining recombinant envelope glycoproteins from different genetic subtypes (5). In another study, vaccination of macaques with a polyvalent subtype B-based envelope glycoprotein vaccine resulted in broader neutralizing-antibody responses and a lower viremia than did vaccination of animals with a monovalent vaccine (11). These studies point to the need to include as many representatives of circulating viruses as possible to formulate a product that may be highly efficacious against HIV-1 viruses that are rapidly spreading. Vaccination and challenge experiments with non-B-subtype-based reagents will further inform the process of a rational approach for a global HIV vaccine.
SHIVMJ4 infection of rhesus macaques may become an important animal model to study the pathogenic sequale of infection with viruses bearing subtype C envelope. The reported preferential utilization of the chemokine receptor CCR5 may, for example, imply a higher transmission potential. Studies using animal models have pointed to the possibility that infection with X4 versus R5 variants of HIV-1 may result in distinct pathogenic outcomes (3, 4, 25). SHIVMJ4 infection of macaques may be an attractive model to study the implications of these findings for transmission and clinical outcome with subtype C, which does not appear to frequently develop X4 variants.
In conclusion, we have generated SHIVMJ4, a subtype C-based SHIV chimera that utilizes the CCR5 chemokine receptor as the coreceptor for cell entry. SHIVMJ4 chimeric viruses efficiently replicate in human, pig-tailed macaque, and rhesus macaque PBMCs. We established a rhesus macaque infection model using SHIVMJ4. For the first time, an HIV-1 subtype C-based SHIV molecular clone that replicates in vitro and in vivo in rhesus macaque cells has been developed. The HIV-1 subtype C sequences in this chimera are derived from a primary isolate from Africa, the region worst affected by the HIV and AIDS epidemic. The SHIVMJ4-macaque model described here may be important and useful in studies of transmission and pathogenesis of HIV-1 subtype C and may facilitate the development, characterization, and evaluation of vaccines against this predominant HIV-1 subtype.
Thanks to C. J. Miller and Ding Lu for useful discussions and technical assistance. The monkey experiments were performed at the California Regional Primate Center (CRPRC) with the assistance of the Immunology Core Laboratory and support from CRPRC. We thank Christopher Mullins for assistance with immunoblotting experiments and Chanc E VanWinkle for editorial assistance.
This study was supported in part by NIH grants AI47067 and AI43255 and by grant TW00004 from the Fogarty International Center.