The basis for the switch from CCR5 to CXCR4 coreceptor usage seen in ϳ50% of human immunodeficiency virus type 1 (HIV-1) subtype B-infected individuals as disease advances is not well understood. Among the reasons proposed are target cell limitation and better immune recognition of the CXCR4 (X4)-tropic compared to the CCR5 (R5)-tropic virus. We document here X4 virus emergence in a rhesus macaque (RM) infected with R5-tropic simian/human immunodeficiency virus, demonstrating that coreceptor switch can happen in a nonhuman primate model of HIV/AIDS. The switch to CXCR4 usage in RM requires envelope sequence changes in the V3 loop that are similar to those found in humans, suggesting that the R5-to-X4 evolution pathways in the two hosts overlap. Interestingly, compared to the inoculating R5 virus, the emerging CXCR4-using virus is highly neutralization sensitive. This finding, coupled with the observation of X4 evolution and appearance in an animal with undetectable circulating virus-specific antibody and low cellular immune responses, lends further support to an inhibitory role of antiviral immunity in HIV-1 coreceptor switch.The human immunodeficiency virus (HIV) enters target cells via interaction of the viral glycoprotein with the cellular receptor CD4 and two principal coreceptors, CCR5 (R5 viruses) and CXCR4 (X4 viruses) (2). Most HIV type 1 (HIV-1) transmission results in a predominantly R5 virus infection. With time, X4 variants arise and coexist with R5 virus variants in ϳ50% of subtype B-infected individuals, and this event is associated with rapid CD4 ϩ T-cell loss and disease progression (22, 37). The determinant(s) of phenotypic change from R5 to X4 maps largely to the V3 loop of the envelope gp120 (6, 18, 39) and can be inferred by analysis of the amino acid sequence of this region (11). Although the underlying basis for virus coreceptor switch late in infection remains ill defined, several hypotheses that include changes in target cell populations during the course of infection and/or differential immune recognition of X4 and R5 viruses have been proposed (31,34). Furthermore, it is unclear whether X4 viruses evolve during the course of infection or are transmitted but selected against early in infection.We have used infection of rhesus macaques (RM) with simian/human immunodeficiency viruses (SHIV) expressing the envelopes of X4 and R5 HIV-1 isolates to study the impact of coreceptor usage in virus transmission and pathogenesis. We previously reported that both X4 and R5 SHIVs can be transmitted intravenously or intravaginally but showed that the basis for the immunodeficiencies caused by these viruses is different. Whereas primary infection with X4 SHIV caused severe and sustained peripheral blood and secondary lymphoid tissue CD4ϩ T-cell loss, infection with R5 SHIV resulted in transient loss of CD4 ϩ T cells at these sites (15, 17). Thus, infection with SHIVs of different coreceptor usage recapitulates the different stages of HIV infection in humans: R5 SHIV provides a model of early infection ...

To examine the pathway of the coreceptor switching of CCR5-using (R5) virus to CXCR4-using (X4) virus in simian-human immunodeficiency virus SHIV SF162P3N -infected rhesus macaque BR24, analysis was performed on variants present at 20 weeks postinfection, the time when the signature gp120 V3 loop sequence of the X4 switch variant was first detected by PCR. Unexpectedly, circulating and tissue variants with His/Ile instead of the signature X4 V3 His/Arg insertions predominated at this time point. Phylogenetic analysis of the sequences of the C2 conserved region to the V5 variable loop of the envelope (Env) protein showed that viruses bearing HI insertions represented evolutionary intermediates between the parental SHIV SF162P3N and the final X4 HR switch variant. Functional analyses demonstrated that the HI variants were phenotypic intermediates as well, capable of using both CCR5 and CXCR4 for entry. However, the R5X4 intermediate virus entered CCR5-expressing target cells less efficiently than the parental R5 strain and was more sensitive to both CCR5 and CXCR4 inhibitors than either the parental R5 or the final X4 virus. It was also more sensitive than the parental R5 virus to antibody neutralization, especially to agents directed against the CD4 binding site, but not as sensitive as the late X4 virus. Significantly, the V3 loop sequence that determined CXCR4 use also conferred soluble CD4 neutralization sensitivity. Collectively, the data illustrate that, similar to human immunodeficiency virus type 1 (HIV-1) infection in individuals, the evolution from CCR5 to CXCR4 usage in BR24 transitions through an intermediate phase with reduced virus entry and coreceptor usage efficiencies. The data further support a model linking an open envelope gp120 conformation, better CD4 binding, and expansion to CXCR4 usage.Entry of human immunodeficiency virus type 1 (HIV-1) into target cells requires the CD4 receptor and one of two coreceptors, CCR5 or CXCR4 (2). CCR5-using (R5) virus predominates early in infection, but in about 50% of subtype B-infected individuals, CXCR4-tropic (X4) virus appears and coexists with R5 viruses, and this is associated with more rapid decline of CD4 ϩ T cells and poorer prognosis (3,5,11,12,58,66). The basis for X4 emergence late in infection remains ill defined, but among the hypotheses proposed are mutation by chance, CCR5 bearing target cell limitation, and differential immune recognition of X4 and R5 viruses (43, 53). Furthermore, it is unclear whether X4 viruses evolve during the course of infection or were present at time of transmission but preferentially suppressed early in infection.In HIV-1-infected individuals and in tissue culture systems, the pathway to coreceptor switching transitions through intermediates with the ability to use CXCR4 in addition to CCR5 (12,50,57,60,61). Compared to the early or inoculating R5 viruses, these R5X4 dual-tropic viruses often display a loss in replicative fitness as well as less efficient use of the CCR5 coreceptor in vitro (30,50). It has been sug...

NK cell-mediated immune surveillance against infected DC may be effective only during early HIV infection and may not be restored by HAART.