Adaptive mutations in the V3 loop of gp120 enhance fusogenicity of human immunodeficiency virus type 1 and enable use of a CCR5 coreceptor that lacks the amino-terminal sulfated region.

To identify sites in gp120 that interact with the CCR5 coreceptor and to analyze the mechanisms of infection, we selected variants of the CCR5-dependent JRCSF molecular clone of human immunodeficiency virus type 1 (HIV-1) that adapted to replicate in HeLa-CD4 cells that express the mutant coreceptor CCR5(Y14N) or CCR5(G163R), which were previously shown to bind purified gp120-CD4 complexes only weakly. Correspondingly, these mutant CCR5s mediate infections of wild-type virus only at relatively high cell surface concentrations, demonstrating a concentration-dependent assembly requirement for infection. The plots of viral infectivity versus concentration of coreceptors had sigmoidal shapes, implying involvement of multiple coreceptors, with an estimated stoichiometry of four to six CCR5s in the active complexes. All of the adapted viruses had mutations in the V3 loops of their gp120s. The titers of recombinant HIV-1 virions with these V3 mutations were determined in previously described panels of HeLa-CD4 cell clones that express discrete amounts of CCR5(Y14N) or CCR5(G163R). The V3 loop mutations did not alter viral utilization of wild-type CCR5, but they specifically enhanced utilization of the mutant CCR5s by two distinct mechanisms. Several mutant envelope glycoproteins were highly fusogenic in syncytium assays, and these all increased the efficiency of infection of the CCR5(Y14N) or CCR5(G163R) clonal panels without enhancing virus adsorption onto the cells or viral affinity for the coreceptor. In contrast, V3 loop mutation N300Y was selected during virus replication in cells that contained only a trace of CCR5(Y14N) and this mutation increased the apparent affinity of the virus for this coreceptor, as indicated by a shift in the sigmoid-shaped infectivity curve toward lower concentrations. Surprisingly, N300Y increased viral affinity for the second extracellular loop of CCR5(Y14N) rather than for the mutated amino terminus. Indeed, the resulting virus was able to use a mutant CCR5 that lacks 16 amino acids at its amino terminus, a region previously considered essential for CCR5 coreceptor function. Our results demonstrate that the role of CCR5 in infection involves at least two steps that can be strongly and differentially altered by mutations in either CCR5 or the V3 loop of gp120: a concentration-dependent binding step that assembles a critical multivalent virus-coreceptor complex and a postassembly step that likely involves a structural rearrangement of the complex. The postassembly step can severely limit HIV-1 infections and is not an automatic consequence of virus-coreceptor binding, as was previously assumed. These results have important implications for our understanding of the mechanism of HIV-1 infection and the factors that may select for fusogenic gp120 variants during AIDS progression.