Host genetic and viral determinants of HIV-1 RNA set point among HIV-1 seroconverters from sub-saharan Africa

Abstract

We quantified the collective impact of source partner HIV-1 RNA levels, human leukocyte antigen (HLA) alleles, and innate responses through Toll-like receptor (TLR) alleles on the HIV-1 set point. Data came from HIV-1 seroconverters in African HIV-1 serodiscordant couple cohorts. Linear regression was used to determine associations with set point and R(2) to estimate variation explained by covariates. The strongest predictors of set point were HLA alleles (B*53:01, B*14:01, and B*27:03) and plasma HIV-1 levels of the transmitting partner, which explained 13% and 10% of variation in set point, respectively. HLA-A concordance between partners and TLR polymorphisms (TLR2 rs3804100 and TLR7 rs179012) also were associated with set point, explaining 6% and 5% of the variation, respectively. Overall, these factors and genital factors of the transmitter (i.e., male circumcision, bacterial vaginosis, and use of acyclovir) explained 46% of variation in set point. We found that both innate and adaptive immune responses, together with plasma HIV-1 levels of the transmitting partner, explain almost half of the variation in viral load set point.

Importance: After HIV-1 infection, uncontrolled virus replication leads to a rapid increase in HIV-1 concentrations. Once host immune responses develop, however, HIV-1 levels reach a peak and subsequently decline until they reach a stable level that may persist for years. This stable HIV-1 set point represents an equilibrium between the virus and host responses and is predictive of later disease progression and transmission potential. Understanding how host and virus factors interact to determine HIV-1 set point may elucidate novel mechanisms or biological pathways for treating HIV-1 infection. We identified host and virus factors that predict HIV-1 set point in people who recently acquired HIV-1, finding that both innate and adaptive immune responses, along with factors that likely influence HIV-1 virulence and inoculum, explain ∼46% of the variation in HIV-1 set point.

FIG 1

Longitudinal HIV-1 RNA set point measurements by HLA concordance, HLA alleles, source partner HIV-1 RNA, and TLR polymorphisms. Points represent individual HIV-1 set point measurements, and lines were created by fitting seroconverter set point measurements for the indicated group against time. Infection for participants were subgrouped according to HLA concordance between source partner and seroconverter (A), the presence or absence of the specific HLA allele (B*1401, B*2703, or B*5301; the dashed line indicates the presence of at least one allele) (B), the quartile of the source partner's plasma HIV-1 RNA level (in log₁₀ copies/ml) (C), and the presence or absence of a specific TLR variant (TLR2 rs3804100 or TLR7 rs179012, with red indicating the absence and blue indicating the presence of at least 1 allele) (D).

FIG 2

Contributions of virus and host/source partner characteristics to explained variation in HIV-1 seroconverter set point levels. Variation explained by each factor was determined as the decrease in the coefficient of determination (R²) when the variable or group of variables was removed from a multiple-variable ordinary least-squares regression model.