Central memory CD4+ T cells are preferential targets of double infection by HIV-1

BACKGROUND: Template switching between two distinct HIV-1 RNA genomes during reverse transcription gives rise to recombinant viruses that greatly expand the genetic diversity of HIV-1 and have adverse implications for drug resistance, immune escape, and vaccine design. Virions with two distinct genomes are produced exclusively from cells infected with two or more viruses, or ‘doubly infected’ cells. Previous studies have revealed higher than expected frequencies of doubly infected cells compared to frequencies based on chance alone, suggesting non-random enhancement of double infection. METHODS: We investigated double infection of unstimulated primary CD4+ T cells using reporter viruses carrying genes for different fluorescent proteins, EGFP and mCherry, combined with sophisticated modeling techniques based on Poisson distribution. Additionally, through the use of multiparameter flow cytometry we examined the susceptibility of naïve and memory subsets of CD4+ T cells to double infection by HIV. RESULTS: Using our double infection system, we confirm non-random enhancement of multiple infection events. Double infection of CD4+ T cells was not found to be a consequence of suboptimal provirus expression rescued by Tat in trans—as has been reported in cell lines—but rather due to a heterogeneous cell population in which only a fraction of primary peripheral blood CD4+ T cells are susceptible to HIV infection regardless of viral titer. Intriguingly, double infection of CD4+ T cells occurred preferentially in memory CD4+ T cells—particularly the central memory (T(CM)) subset—but was not a consequence of SAMHD1-mediated restriction of HIV infection in naïve cells. CONCLUSIONS: These findings reveal that double infection in primary CD4+ T cells is primarily a consequences of cellular heterogeneity and not rescue of suboptimal provirus expression by Tat in trans. Additionally, we report a previously unappreciated phenomenon of enhanced double infection within primary T(CM) cells and suggest that these long-lived cells may serve as an archive that drive ongoing viral recombination events in vivo.