Natural polymorphisms in HIV-1 CRF01_AE strain and profile of acquired drug resistance mutations in a long-term combination treatment cohort in northeastern China

Abstract

Abstract Background The impacts of genetic polymorphisms on drug resistance mutations (DRMs) among various HIV-1 subtypes have long been debated. In this study, we aimed to analyze the natural polymorphisms and acquired DRM profile in HIV-1 CRF01_AE-infected patients in a large first-line antiretroviral therapy (ART) cohort in northeastern China. Methods The natural polymorphisms of CRF01_AE were analyzed in 2034 patients from a long-term ART cohort in northeastern China. The polymorphisms in 105 treatment failure (TF) patients were compared with those in 1148 treatment success (TS) patients. The acquired DRM profile of 42 patients who experienced TF with tenofovir/lamivudine/efavirenz (TDF/3TC/EFV) treatment was analyzed by comparing the mutations at TF time point to those at baseline. The Stanford HIVdb algorithm was used to interpret the DRMs. Binomial distribution, McNemar test, Wilcoxon test and CorMut package were used to analyze the mutation rates and co-variation. Deep sequencing was used to analyze the evolutionary dynamics of co-variation. Results Before ART, there were significantly more natural polymorphisms of 31 sites on reverse transcriptase (RT) in CRF01_AE than subtype B HIV-1 (|Z value| ≥ 3), including five known drug resistance-associated sites (238, 118, 179, 103, and 40). However, only the polymorphism at site 75 was associated with TF (|Z value| ≥ 3). The mutation rate at 14 sites increased significantly at TF time point compared to baseline, with the most common DRMs comprising G190S/C, K65R, K101E/N/Q, M184 V/I, and V179D/I/A/T/E, ranging from 66.7 to 45.2%. Moreover, two unknown mutations (V75 L and L228R) increased by 19.0 and 11.9% respectively, and they were under positive selection (Ka/Ks > 1, log odds ratio [LOD] > 2) and were associated with several other DRMs (cKa/Ks > 1, LOD > 2). Deep sequencing of longitudinal plasma samples showed that L228R occurred simultaneously or followed the appearance of Y181C. Conclusion The high levels of natural polymorphisms in CRF01_AE had little impact on treatment outcomes. The findings regarding potential new CRF01_AE-specific minor DRMs indicate the need for more studies on the drug resistance phenotype of CRF01_AE.