Acquisition of CD4-dependence by CD4-independent SIV passaged in human peripheral blood mononuclear cells

Chemokine receptors (CKRs), the primordial receptors for primate lentiviruses, are sufficient to mediate virus-cell fusion. Several different fusogenic CKRs and related receptors provide a broad potential host cell range, presumably advantageous for viral spread within a given infected individual, and across species. By contrast, the additional constraint of obligatory CD4 binding, just prior to CKR engagement, radically restricts potential host cells within an individual (or lymph node microenvironment), and might also limit xenotransmission, as CD4 sequences vary among primates. In spite of these potential drawbacks, CD4 dependent entry for SIV and HIV is the rule rather than the exception, and is generally thought to have evolved by selection for 1) stabilization of virus-cell surface interactions, and 2) conformational shielding of readily neutralized CKR binding epitopes. CD4 binding residues of SIV and HIV envelope are recessed, (relatively hidden from immune detection) and may exhibit a strong degree of automimicry, thus benefitting from self tolerance.Documented evolution, within individual macaques, of neutralization-resistant CD4-dependent SIV, derived from CD4-independent inocula, supports these ideas, but does not explain CD4's exclusive role as the penultimate receptor-even more striking, given the wide diversity of CKRs and other surface molecules that can serve as actual fusion receptors for SIV. We, therefore, explored the additional, non-exclusive, hypothesis that surface CD4 on leukocytes is a marker of a more favorable host cell environment, as compared to CD8, NK, or B cell surface markers.

We demonstrate progressive in vitro evolution of two SIV strains to CD4-dependence (and CXCR4 tropism) in normal human PBMCs (hPBMCs). The two CD4-independent strains of SIV tested developed nearly complete CD4 dependence over several months of serial passage in hPBMCs, correlating with a limited number of non-synonymous env region mutations, some previously reported to be determinants of CD4-dependency. The initial ability of SIV stocks to grow to significant (albeit, relatively low) levels in CD4(-), CD14(-) cells was also lost with long term passage. Rapid emergence and subsequent prominence of G → A and A → G mutations within env regions associated with CD4 dependence was seen.

Progressive acquisition of strict CD4 tropism, independent of immunoselection, supports the idea that surface CD4 identifies optimal host cells having intracellular environments most favorable to viral replication. The prominence of mutations involving G to A, or A to G, suggests that APOBEC 3 mediated infidelity may facilitate rapid switching of cell surface receptor usage within SIV swarms encountering fluctuating availability of optimal CD4+CKR+ targets. These observations of non-immune selection are compatible with, and may accelerate, simultaneous selection for previously described CD4-dependent neutralization resistance in vivo.