Background:  Anti-HIV agents targeting CCR5 have shown significant potency in cell culture, primate models, and phase II clinical trials. HIV resistance to CCR5 antagonists in cell culture has been observed in the absence of coreceptor switch but is unclear whether inhibition of HIV-1 replication with a CCR5 antagonist will lead to an increased rate of emergence of CXCR4 variants depending on the cell culture conditions or the intrinsic capacity of a particular strain to switch coreceptor use after selective CCR5 drug pressure.

Methods:  R5 HIV-1 strains were evaluated for their susceptibility to CXCR4, CCR5, and reverse transcriptase inhibitors (AMD3100, Tak-779, and AZT, respectively) and cultured in lymphoid cells expressing high levels of CXCR4 and low levels of CCR5 in the absence or presence of these drugs. Cell cultures were passaged and prolonged for up to 200 days and syncytium formation was recorded weekly. Viral co-receptor use was evaluated in MT-2 cells and U87-CD4⁺ cells expressing the appropriate co-receptor at different times post infection.

Results:  Virus replication was reminiscent of slow replicating, non-syncytium-inducing phenotype but could be blocked by the reverse transcriptase inhibitor AZT or the CCR5 antagonist Tak-779. In the absence of drug pressure, 3 of 6 strains used were able to switch from the R5 to the X4 phenotype and showed increased replication rate. Co-receptor switch could be delayed by AZT or Tak-779. However, under similar drug pressure, outgrowth of virus could be detected faster in presence of Tak-779, but not in the presence of AZT and was concomitant to its ability to use CXCR4. Conversely, treatment with AMD3100 prevented the emergence of CXCR4-using virus. In peripheral blood mononuclear cells (CCR5⁺, CXCR4⁺), passage of HIV-1 for as long as 200 days in the presence of Tak-779 did not lead to co-receptor switch.

Conclusions:  Experimental conditions strongly determine the outcome of CCR5 drug pressure on the emergence of CXCR4-using virus. A lower replication rate (ie, AZT or Tak-779 vs no drug treatment) will slow co-receptor switch. However, under similar replication conditions (ie, AZT vs Tak-779) CCR5 drug pressure will induce a faster emergence of CXCR4-using HIV. A cell culture model of the evolution of HIV-1 co-receptor use may be relevant to assess the propensity of clinical isolates to develop drug resistance through a change in virus co-receptor.