Investigation of nucleic acid sequences associated with antiretroviral resistance in whole blood, leukocyte and plasma samples from HIV-1 positive patients
Project no. 28519 · Status: Kapanmış (Yayın bekleniyor) · 2023
Summary
<p class="ql-align-justify"><strong style="color: black;">ABSTRACT</strong></p><p class="ql-align-justify"><strong style="color: black;">Objective:</strong><span style="color: black;"> This study aimed to compare antiretroviral resistance-associated nucleic acid sequences in whole-blood, leukocyte, and plasma samples from HIV-1-positive patients. It also sought to characterize the distribution of resistance-associated variants across biological compartments, identify low-frequency mutations, and delineate intrapatient viral population heterogeneity. Beyond the conventional consensus-based approach, the study further investigated viral quasispecies detected by deep sequencing and evaluated their potential association with drug resistance.</span></p><p class="ql-align-justify"><strong style="color: black;">Materials and Methods:</strong><span style="color: black;"> Sixteen HIV-1-positive patients were included, and plasma, whole-blood, and leukocyte samples were obtained simultaneously from a single blood draw. Viral RNA was extracted from plasma, whereas total nucleic acid was isolated from whole blood and the leukocyte fraction. The HIV-1 </span><em style="color: black;">pol</em><span style="color: black;"> region was amplified by one-step RT-PCR followed by nested PCR, and the resulting amplicons were sequenced on the Illumina MiniSeq platform using 2 × 150 bp paired-end reads. Consensus-level drug resistance mutations were interpreted using SierraLocal/Stanford HIVdb, whereas low-frequency variants were analyzed with LoFreq using a minimum coverage threshold of 100× and a variant frequency range of 2%–20%. APOBEC-associated G-A hypermutation was assessed using Hypermut 3.0. Haplotype reconstruction was performed with CliqueSNV, and compartmentalization was evaluated using TN93-based F</span><sub style="color: black;">ST</sub><span style="color: black;">, Slatkin-Maddison, and Snn tests. Phylogenetic analysis was performed using the maximum-likelihood method.</span></p><p class="ql-align-justify"><strong style="color: black;">Results:</strong><span style="color: black;"> Of the 45 samples, the P07 leukocyte sample and the P12 and P14 plasma samples were excluded because of amplification failure, whereas the P15 plasma sample was excluded due to a low mapping rate (0.39%) and insufficient target-region coverage (23.18%). The mean breadth of coverage across the target region was 99.5%, the mean sequencing depth was 6,441×, and 97.2% of the target region was covered at a depth of at least 100×. Although only a limited number of drug resistance mutations were detected at the consensus level, low-frequency variant analysis identified at least one such variant in 10 of 16 patients (62.5%). Intercompartmental comparisons suggested the presence of archived variants that were undetectable in plasma but detectable in cellular DNA. In the discordance analysis, 22 of 25 identified mutations (88%) were discordant, with the largest proportion detected in the leukocyte compartment. APOBEC hypermutation screening yielded only two positive results among 93 tests, both in the P06_WB sample. Haplotype reconstruction identified a total of 515 haplotypes across 44 samples; the mean number of haplotypes per sample was 11.7, the mean Shannon diversity index was 1.59, and the mean dominant haplotype frequency was 0.47. Compartmentalization was prominent at the haplotype level: F</span><sub style="color: black;">ST</sub><span style="color: black;"> analysis was significant in 73.7% of compartment pairs, the Slatkin-Maddison test in 87.5% of patients, and the Snn test in 93.8%.</span></p><p class="ql-align-justify"><strong style="color: black;">Conclusion:</strong><span style="color: black;"> Plasma RNA consensus genotyping alone may miss a substantial proportion of low-frequency and compartment-specific resistance-associated variants in HIV-1. Although proviral DNA-based analyses provide complementary information, they should be interpreted cautiously because of APOBEC-associated artifacts and preanalytical variability. Detection of reservoir-associated low-frequency variants may require lowering variant-frequency thresholds within an analytically reliable range of 2%–20%. The findings should be interpreted in conjunction with haplotype analysis, compartmentalization testing, and treatment history. Prospective studies in larger cohorts are warranted to clarify the clinical significance of these variants.</span></p><p class="ql-align-justify"><strong style="color: black;">Keywords:</strong><span style="color: black;"> HIV drug resistance, next-generation sequencing, low frequency variants</span></p><p class="ql-align-justify"></p><p></p>