CD32-expressing CD4 T cells are phenotypically diverse and can contain proviral HIV DNA

Abstract

Efforts to both characterize and eradicate the HIV reservoir have been limited by the rarity of latently infected cells and the absence of a specific denoting biomarker. CD32a (Fc?RIIa) has been proposed to be a marker for an enriched CD4 T cell HIV reservoir, but this finding remains controversial. Here, we explore the expression of CD32 on CD3+CD4+ cells in participants from two primary HIV infection studies and identify at least three distinct phenotypes (CD32low, CD32+CD14+, and CD32high). Of note, CD4 negative enrichment kits remove the majority of CD4+CD32+ T cells, potentially skewing subsequent analyses if used. CD32high CD4 T cells had higher levels of HLA-DR and HIV co-receptor expression than other subsets, compatible with their being more susceptible to infection. Surprisingly, they also expressed high levels of CD20, TCRa?, IgD, and IgM (but not IgG), markers for both T cells and naïve B cells. Compared with other populations, CD32low cells had a more differentiated memory phenotype and high levels of immune checkpoint receptors, programmed death receptor-1 (PD-1), Tim-3, and TIGIT. Within all three CD3+CD4+CD32+ phenotypes, cells could be identified in infected participants, which contained HIV DNA. CD32 expression on CD4 T cells did not correlate with HIV DNA or cell-associated HIV RNA (both surrogate measures of overall reservoir size) or predict time to rebound viremia following treatment interruption, suggesting that it is not a dominant biomarker for HIV persistence. Our data suggest that while CD32+ T cells can be infected with HIV, CD32 is not a specific marker of the reservoir although it might identify a population of HIV enriched cells in certain situations. © 2018 Martin, Pace, Thornhill, Phetsouphanh, Meyerowitz, Gossez, Brown, Olejniczak, Lwanga, Ramjee, Kaleebu, Porter, Willberg, Klenerman, Nwokolo, Fox, Fidler and Frater.