Background:  We studied the hypervariable domains in longitudinal samples from 14 men enrolled in the San Francisco Men’s Health Study with the goal of examining more closely the timing of HIV-1 co-receptor usage changes, and the rates of env variant population turnover during chronic infection. Subjects with rapid and slow rates of CD4⁺ T-cell loss were compared.

Methods:  Over 2.5 to 8 years, 6 subjects experienced rapid rates of CD4⁺ T cell loss, and 8 subjects experienced slow rates of CD4⁺ T cell loss. HIV-1 RNA was isolated from plasma or serum collected semiannually, and the V3, V1-V2, and V4-V5 regions were analyzed using specific heteroduplex tracking assays (HTA). V3 regions were cloned and sequenced.

Results:  Shifts in V3 HTA patterns and sequence substitutions indicative of X4 variant emergence were seen for 5 of the 6 subjects with rapid loss of CD4 cells and 4 of the 8 subjects with slow loss of CD4 cells. Surprisingly, 3 of the subjects with slow CD4 loss and X4 shifts were ccr5∆32 heterozygotes, while all slow-loss subjects without X4 shifts were wild type for CCR5. In the rapid-loss subjects with X4 shifts, the V3-HTA pattern shifts occurred after the rapid loss of CD4 cells began. HTA analysis of the V1-V2 and V4-V5 env populations in these same subjects showed dramatic changes in the variant populations over time. The subjects with rapid CD4 loss had significantly less change in their env populations than the subjects with slow CD4 loss (p <0.01). Of the rapid-loss subjects whose CD4 count was <150/µL for at least 2 years, 2 showed stabilization of their V1-V2 and V4-V5 populations as reflected by low levels of total change in HTA pattern and low HTA indices (a novel measure of new band emergence and band distribution); this stabilization was not observed in the other subjects. For both subjects, the stabilization occurred after the emergence of X4 variants.

Conclusions:  Our data indicate that X4 emergence is associated with rapid loss of CD4⁺ T cells, but occurs after the rapid loss begins, which suggests that X4 variants arise as a result of, rather than a cause of, the rapid loss of CD4 cells. X4 emergence showed a trend toward association with ccr5∆32 heterozygosity in this small study. The stabilization of env variant populations at low CD4 counts following periods of rapid viral evolution suggests that selective pressure on env, likely from new immune responses, is minimal when CD4 counts drop dramatically and remain low for extended periods of time.