Deep sequencing of the HIV latent reservoir to discern mechanisms of clonal proliferation and HIV persistence

PROJECT SUMMARY/ABSTRACT The HIV latent reservoir (LR) is the major barrier to cure of HIV. The persistence of the LR in blood and tissues in people living with HIV (PLWH) during ART is due to cellular proliferation of infected cells. Three mechanisms are postulated to drive infected cell proliferation: integration of the provirus in or near genes associated with

cancer or cell proliferation; engagement of the infected CD4+ T cell receptor with its cognate antigen, leading to proliferative bursts; and cytokine-mediated homeostatic proliferation. Recent evidence suggests that oncogene- driven proliferation does not contribute significantly to infected cell proliferation. Meanwhile, antigen-driven

proliferation of infected cells appears to account for many or most of the most-highly expanded, or dominant, infected cell clones. However, the degree to which homeostatic proliferation versus antigen-driven proliferation maintains the overall reservoir is difficult to discern. Because infected cells are so clonal, shallow sampling

depths still reveal dominant clones (likely maintained by antigenic stimulation) while underestimating the extent

of the non-dominant clones (the ‘tail’ of the clonality distribution, likely maintained by homeostatic proliferation). We will fill this knowledge gap by generating a near-full length HIV provirus sequencing dataset deep enough to accurately assess clonality richness and by developing models to more precisely quantify the fractional

contribution of antigen-driven versus homeostatic proliferation to reservoir maintenance. In Aim 1, we will fully characterize the intact and defective provirus clonality distributions in PLWH on ART by generating near-full length HIV provirus sequencing datasets deep enough to approximate clonality richness saturation. We will test

the hypothesis that provirus clonality distributions mirror those of uninfected memory CD4+ T cells, with a small number of dominant clones and a large number of non-dominant clones, and that they can be modeled with similar power-law exponents. In Aim 2, we will first generate longitudinal characterizations of in vivo antigen-

driven and homeostatic CD4+ T cell proliferation of uninfected CD4s in PLWH on ART using T cell receptor repertoires. Then we will use this information to construct mathematical models to estimate the fractional contributions of antigen-driven and homeostatic proliferation to the persistence of the reservoir in PLWH on ART.

We will test the hypothesis that both antigen-driven and homeostatic proliferation contributes significantly to the maintenance of the HIV LR. This work and the resulting models will help the HIV cure field design future therapeutic strategies and help the field predict more precisely outcomes when HIV cure strategies such as anti-

proliferative therapies or personalized therapeutic vaccines are trialed.