Abstract
We use mathematical models to determine possible mechanisms contributing to the evolution and rise of virulent CXCR4-tropic HIV in vivo. The models predict that the ability of the virus to specialize on a given target cell type depends on the exact fitness landscape of the viral mutants. Because this fitness landscape varies between people, this may explain why the evolution of fully CXCR4-tropic strains only occurs in about 50% of infected patients. Assuming that CXCR4-tropic HIV may evolve, we investigate the effect of different immune responses on the rise of such virulent strains. If we assume that CXCR4-tropic HIV is more cytopathic than CCR5-tropic virus, virulent CXCR4-tropic mutants remain suppressed at low levels both in the absence of an immune response, and in the presence of responses that act on the virus before integration into the host genome. On the other hand, this difference in cytopathogenicity is reduced by the presence of immune responses acting on infected cells, allowing CXCR4-tropic HIV to coexist with the CCR5-tropic virus. These results may help to interpret experimental data and are discussed with reference to the literature.