Clonal dynamics of HIV reservoirs through the lens of CD4+ T cells

The HIV reservoir consists of latently infected cells that persist despite suppressive antiretroviral therapy (ART) and is the barrier to cure. Despite the rapid mutation of HIV during untreated infection, HIV reservoirs have highly clonal sequence distributions that resemble those of CD4+ T cell receptor (TCR) sequence repertoires. We hypothesize the same mechanisms driving T cell dynamics sustain HIV during ART. Here, we develop power-law models of CD4+ TCR rank abundance distributions and contrast these with published models of HIV reservoir clonality. We find TCR distributions can be split into more and less clonal subgroups that are hypothetically related to antigenically responding vs homeostatically proliferating clonotypes. We show that existing experiments would be likely to sample HIV reservoir cells from the more clonal subgroup, potentially biasing the field towards thinking that HIV reservoirs are sustained by antigenic proliferation. We then calculate sample sizes that would be required to overcome this bias in future experiments. Finally, we demonstrate a stochastic model that explains HIV reservoir decay based on T cell expansion and contraction clonal dynamics. This model suggests that reservoir cells must be slightly more likely to die than uninfected CD4+ T cells. Our modeling helps characterize the dominant mechanisms sustaining the HIV reservoir, informing rational targets for future interventions.