Abstract
Early HIV infection involves the transmission of a small number of virions and the subsequent expansion and spread of the virus. The early phases of infection represent an important opportunity for viral control, although relatively little is understood about the determinants of the rate of spread or the "bottlenecks" the virus must negotiate during this time. Here, we use the barcoded simian immunodeficiency virus (SIV) infection model to analyze the kinetics of early SIV infection in vitro and in vivo. We analyze the trajectories of multiple individual barcode lineages across in vitro infection of cell lines and primary cells and after mucosal and intravenous infection of macaques. We observe a large distribution in lineage sizes when an infection is founded by multiple different viral lineages, which we presume results from bottleneck-induced variability in early growth between lineages. Surprisingly, we find that lineage heterogeneity is similar after mucosal and intravenous infection, indicating that dissemination bottlenecks present in the genital tract are not grossly different from those encountered during seeding of systemic tissues. Comparing in vitro and in vivo diversity, we estimate that inherent heterogeneity in viral production by the first infected cells may contribute to 23%-44% of the dissemination bottleneck. However, the majority of the heterogeneity in viral spread in vivo (56%-77%) likely arises from heterogeneity in spread in local microenvironments.IMPORTANCEA brief window for viral clearance exists immediately following HIV transmission, where low levels of virus and infected cells are present. Understanding the bottlenecks encountered by the virus during this window is essential when designing therapies for this early period. We used a barcoded simian immunodeficiency virus infection model to track the early dissemination of multiple viral lineages in rhesus macaques. We observed up to a 10(5)-fold difference in the viral load of transmitted lineages within a single animal 2 weeks after infection, suggesting that early bottlenecks induced very different trajectories of initial growth. By comparing lineage size diversity under multiple experimental conditions, we conclude that around a quarter of the diversity in lineage size results from early cellular infection events, with the remainder likely attributable to differences between initial anatomical sites of infection in vivo.