Insights to HIV-1 coreceptor usage by estimating HLA adaptation with Bayesian generalized linear mixed models

Area of Science:

• Virology
• Immunology
• Computational Biology

Background:

• Mechanisms of human immunodeficiency virus type I (HIV-1) coreceptor switching from CCR5 to CXCR4 remain unclear.
• High CD4+ T cell counts with CXCR4 usage present a puzzling scenario.
• Viral adaptation to the human leukocyte antigen (HLA) complex is hypothesized to drive coreceptor switch.

Purpose of the Study:

• To investigate the role of HLA adaptation in HIV-1 coreceptor switching.
• To develop and apply a novel computational method for estimating HLA adaptation.
• To assess the predictive value of HLA adaptation for coreceptor usage and switching.

Main Methods:

• Sequencing of viral gag and env genes and HLA class I and II alleles in 312 treatment-naive, subtype C HIV-1 patients.
• Development of Bayesian generalized linear mixed models (GLMMs) to estimate HLA adaptation, accounting for the full HLA repertoire and viral phylogenetic relatedness.
• Utilizing the geno2pheno[coreceptor] tool to assess the false positive rate (FPR) of CCR5-using variants.

Main Results:

• CXCR4-using HIV-1 variants demonstrated significantly higher HLA adaptation compared to CCR5-using variants (p=1.34e-2).
• Adapted CCR5-using variants exhibited a lower predicted false positive rate (FPR) by geno2pheno[coreceptor] than non-adapted variants (p=2.21e-2).
• Lower FPR is associated with CXCR4 usage, suggesting a link between adaptation and coreceptor switch.

Conclusions:

• HLA adaptation is a significant factor in HIV-1 coreceptor usage and switching.
• Estimating HLA adaptation can aid in predicting coreceptor usage and impending switches in CCR5-using HIV-1.
• Bayesian GLMMs offer a robust approach for modeling virus-host adaptation.