Impact of clinical reverse transcriptase sequences on the replication capacity of HIV-1 drug-resistant mutants.

Abstract

We have shown that the HIV-1 laboratory strain NL4-3 that contains P236L [a reverse transcriptase mutation conferring resistance to the nonnucleoside reverse transcriptase inhibitor (NRTI) delavirdine] replicates more slowly than wild-type NL4-3. Other NNRTI-resistance mutations, such as K103N and Y181C, do not reduce the replication capacity of NL4-3 as much as P236L and develop more frequently in HIV-1 isolates from patients failing delavirdine. However, a minority of patients on delavirdine therapy still have isolates with P236L. We postulated that reverse transcriptase (RT) sequences from these patient isolates contain other mutations that compensate for the adverse effect of P236L. To test this hypothesis, we created 15 chimeric NL4-3 isolates that contained delavirdine-resistant RT sequences derived from eight patient isolates and characterized their replication kinetics. Nine of 10 patient-derived clones containing P236L replicated as slowly as NL4-3 with P236L. In contrast, three of five clones that did not have P236L (but had either K103N or Y181C) replicated significantly better than NL4-3 with P236L. Thus, the majority of patients who acquire P236L during delavirdine therapy do not have RT mutations that compensate for the replication defect conferred by P236L. We hypothesize that HIV-1 isolates with P236L may have a compensatory mutation outside RT. Alternatively, variants of HIV-1 with reduced replication fitness may be selected during antiretroviral therapy, suggesting that stochastic events rather than viral replication fitness may determine which drug-resistant mutants emerge early during antiretroviral failure. In some isolates, it appears that the background RT sequence can contribute significantly to the replication fitness of drug-resistant HIV-1 variants.