Phylogenetic Relatedness of Circulating HIV-1C Variants in Mochudi, Botswana

Background
Determining patterns of HIV transmission is increasingly important for the most efficient use of modern prevention interventions. HIV phylogeny can provide a better understanding of the mechanisms underlying HIV transmission networks in communities.
Methods
To reconstruct the structure and dynamics of a local HIV/AIDS epidemic, the phylogenetic relatedness of HIV-1 subtype C env sequences obtained from 785 HIV-infected community residents in the northeastern sector of Mochudi, Botswana, during 2010–2013 was estimated. The genotyping coverage was estimated at 44%. Clusters were defined based on relatedness of HIV-1C env sequences and bootstrap support of splits.
Results
The overall proportion of clustered HIV-1C env sequences was 19.1% (95% CI 17.5% to 20.8%). The proportion of clustered sequences from Mochudi was significantly higher than the proportion of non-Mochudi sequences that clustered, 27.0% vs. 14.7% (p = 5.8E-12; Fisher exact test). The majority of clustered Mochudi sequences (90.1%; 95% CI 85.1% to 93.6%) were found in the Mochudi-unique clusters. None of the sequences from Mochudi clustered with any of the 1,244 non-Botswana HIV-1C sequences. At least 83 distinct HIV-1C variants, or chains of HIV transmission, in Mochudi were enumerated, and their sequence signatures were reconstructed. Seven of 20 genotyped seroconverters were found in 7 distinct clusters.
Conclusions
The study provides essential characteristics of the HIV transmission network in a community in Botswana, suggests the importance of high sampling coverage, and highlights the need for broad HIV genotyping to determine the spread of community-unique and community-mixed viral variants circulating in local epidemics. The proposed methodology of cluster analysis enumerates circulating HIV variants and can work well for surveillance of HIV transmission networks. HIV genotyping at the community level can help to optimize and balance HIV prevention strategies in trials and combined intervention packages.

背景
明确HIV传播模式对于最优化利用现代预防干预手段愈发重要。HIV系统发育分析（HIV phylogeny）有助于深入理解社区内HIV传播网络的潜在作用机制。
方法
为重构当地HIV/AIDS流行的结构与动态特征，本研究对2010-2013年间于博茨瓦纳莫丘迪东北部地区785名HIV感染社区居民所获取的HIV-1 C亚型env基因序列（env sequences）的系统发育相关性进行了评估。基因分型覆盖率经估算为44%。聚类的划分依据为HIV-1C env序列的相关性以及分支的自展支持率（bootstrap support）。
结果
聚类后的HIV-1C env序列总体占比为19.1%（95%置信区间（confidence interval, CI）：17.5%~20.8%）。莫丘迪本地序列的聚类占比（27.0%）显著高于非莫丘迪序列的聚类占比（14.7%）（p=5.8×10^-12；费希尔精确检验（Fisher exact test））。大部分莫丘迪本地聚类序列（90.1%；95%CI：85.1%~93.6%）仅存在于莫丘迪特有聚类中。未发现任何莫丘迪序列与1244条非博茨瓦纳HIV-1C序列发生聚类。本研究共枚举了莫丘迪地区至少83种不同的HIV-1C变异株或HIV传播链，并重构了其序列特征。20名经基因分型的血清转换者（seroconverters）中有7名分别属于7个不同的聚类。
结论
本研究阐明了博茨瓦纳某社区HIV传播网络的核心特征，证实了高采样覆盖率的重要性，并强调需通过大范围HIV基因分型来明确当地流行中社区特有及社区混合病毒变异株的传播情况。本研究提出的聚类分析方法可有效枚举循环传播的HIV变异株，适用于HIV传播网络的监测工作。社区层面的HIV基因分型有助于优化并平衡临床试验与联合干预方案中的HIV预防策略。