Relationship between HLA class I-driven evolution in Gag, Pol and Nef and clinical markers of HIV disease: A multi-center collaborative study

Background: The relationship between immune escape and HIV disease is incompletely understood. Here we examine the relationship between CTL selection pressure and plasma viral load (pVL) by defining HLA class I-associated polymorphisms in Gag, Pol and Nef in a large cohort of untreated HIV-1 subtype B-infected persons. Methods: HLA-associated polymorphisms were identified in >1200 chronically-infected, treatment-naïve individuals from Canada (HOMER cohort), the USA (ACTG5142/5128 protocols), and Western Australia (WAHC), using phylogenetically-corrected analysis methods incorporating a multivariate adjustment for HLA linkage disequilibrium and a q-value correction for multiple tests. In individuals for whom clinical data were available (N~550), correlations between pVL and HLA-associated polymorphisms were assessed using Spearman’s rank test. Results: >1000 unique HLA-associated polymorphisms were identified, but the number of codons involved varied by protein: 22%, 16% and 48% of Gag, Pol and Nef codons harbored an HLA-associated polymorphism. Roughly 60% of polymorphisms mapped in or near known or predicted CTL epitopes; a further 20% likely represent compensatory mutations. A modest inverse correlation was observed between the total number of HLA-associated sites in Gag (but not Pol/Nef) and pVL (R=-0.13;p=0.002), suggesting that Gag CTL targeting contributes more substantially to viremia control than targeting other proteins. A positive correlation was observed between the proportion of escaped Gag (but not Pol/Nef) sites and pVL (R=0.11;p=0.01), suggesting that the consequences of escape in Gag may be greater than other proteins. Conclusion: Results support strong CTL-mediated selection on HIV, the effects of which vary by protein. The list of >1000 HLA-associated polymorphisms represents the most comprehensive to date for HIV-subtype-B. The associations between pVL and CTL selection pressure/escape in Gag suggest that in vivo CTL targeting of Gag may be import