The existence of the latent HIV reservoir is a major barrier to HIV cure [1]. Understanding what maintains the latent reservoir and how latently infected cells are eliminated will help advance HIV cure efforts. For the majority of individuals living with HIV, levels of plasma genomic HIV RNA and cellular HIV DNA and mRNA rise rapidly and peak within the first month following infection, then plateau [2]. After antiretroviral therapy (ART) is initiated, HIV RNA levels decline rapidly (by 10,000-fold) and can become undetectable by clinical assays. HIV DNA levels also decline, but less impressively (10to 20-fold) [2-4]. Modelling demonstrates that HIV DNA decay kinetics fit a 3-slope curve over three time intervals: zero to seven months, eight to thirty-two months and >32 months, with slopes of decay of 0.131, 0.016 and 0.0021 log10 copies/10 peripheral blood mononuclear cells (PBMC)/month over the respective time periods [5]. HIV controllers are a small group (approximately 1%) of individuals who are able to control viral replication without ART to very low or undetectable levels of HIV-1 RNA by clinical assays for long periods of time [6]. The mechanisms involved in their viral control likely include a combination of both viral and host factors [7]. The longitudinal kinetics of infected cell decay (i.e. of HIV DNA) in HIV controllers are less well described. In their recent article in the Journal of the International AIDS Society, Avettand-Fenoel et al. described HIV DNA kinetics in 202 HIV controllers (defined as having HIV RNA <400 copies/mL without ART) from the ANRS-CODEX cohort [8]. The median HIV DNA was 1.5 log10copies/10 6 PBMC, which is much lower than the 3.3 log10 copies/10 6 PBMC in ART-na€ıve individuals during primary HIV infection in the ANRS-PRIMO cohort [9], but similar to the predicted 1.6 log10 copies/10 6 PBMCs after five years of uninterrupted ART in individuals started on ART within 15 days after HIV infection in the same cohort [5]. Mathematical modelling of HIV DNA dynamics in the ANRS-CODEX HIV controller cohort revealed a significant decline in 46% of participants. HLA-B*27/B*57 alleles and lower levels of plasma HIV RNA and HIV DNA at the entry visit into the cohort were independently associated with HIV DNA decline. The authors postulated that intrinsic resistance of host cells to HIV infection, lower levels of immune activation leading to fewer potential target cells, lower residual HIV replication and a shorter half-life of infected cells may all have contributed to the observed decline in HIV DNA levels. By contrast, nonHLA-B*27/B*57 alleles or persistent HIV RNA ≥1 log10 copies/mL during follow-up were associated with increases in HIV DNA over time in the same controller cohort. Characterization of the HIV proviruses by DNA sequencing and analyses of clonal expansion of infected cells in the controllers could provide additional insight into the mechanisms behind the divergent HIV DNA dynamics that were observed [10]. Such analyses would differentiate clonal expansion of cells with identical proviruses from ongoing cycles of infection as the cause of rising HIV DNA levels. Although a decline in HIV DNA levels in HIV controllers is intriguing, it does not necessarily imply a reduction in the latent but replication-competent (intact) proviral reservoir that can produce infectious virus and lead to viral rebound. This is because unintegrated linear and episomal DNAs, in addition to integrated proviruses, are also detected by total HIV DNA quantification, neither of which can produce infectious virus. HIV controllers have also been found to have a higher proportion of unintegrated DNA to total HIV DNA when compared to both treated and untreated non-controllers [11]. In addition, the vast majority of integrated proviral DNA is defective and cannot lead to the production of infectious virions [12]. Though assays designed to quantify replication-competent provirus may be able to provide a more accurate estimate of the latent reservoir, these assays are labour intensive and may lack the dynamic range to detect small fluctuations [13]. Newer PCR-based assays of intact provirus should provide further insight into whether the changes observed in HIV DNA levels in HIV controllers in the ANRS-CODEX cohort parallel those of the intact reservoir. Hsu DC and Mellors JW Journal of the International AIDS Society 2019, 22:e25254 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25254/full | https://doi.org/10.1002/jia2.25254

中文翻译：

---

HIV 控制者中 HIV DNA 的纵向变化：它们意味着什么？

HIV 潜伏宿主的存在是 HIV 治愈的主要障碍[1]。了解什么维持着潜在的储存库以及如何消除潜在感染的细胞将有助于推进 HIV 治疗工作。对于大多数 HIV 感染者，血浆基因组 HIV RNA 和细胞 HIV DNA 和 mRNA 的水平迅速上升，并在感染后的第一个月内达到峰值，然后趋于平稳 [2]。抗逆转录病毒治疗 (ART) 开始后，HIV RNA 水平迅速下降（下降 10,000 倍），临床检测无法检测到。HIV DNA 水平也会下降，但下降幅度较小（10 至 20 倍）[2-4]。建模表明，HIV DNA 衰减动力学在三个时间间隔内符合 3 条斜率曲线：0 到 7 个月、8 到 32 个月和 > 32 个月，衰减斜率分别为 0.131、0.016 和 0。0021 log10 拷贝/10 个外周血单核细胞 (PBMC)/月在相应的时间段内 [5]。HIV 控制者是一小群（大约 1%）的个体，他们能够在没有 ART 的情况下将病毒复制控制到非常低或无法通过临床检测长时间检测到的 HIV-1 RNA 水平 [6]。其病毒控制所涉及的机制可能包括病毒和宿主因素的组合 [7]。HIV 控制者中受感染细胞衰变（即 HIV DNA）的纵向动力学描述较少。在他们最近发表在国际艾滋病协会杂志上的文章中，Avettand-Fenoel 等人。描述了来自 ANRS-CODEX 队列的 202 个 HIV 控制者（定义为 HIV RNA <400 拷贝/mL，没有 ART）中的 HIV DNA 动力学[8]。HIV DNA 中位数为 1.5 log10copies/10 6 PBMC，这远低于 ANRS-PRIMO 队列中初次 HIV 感染期间未接受 ART 个体的 3.3 log10 拷贝/10 6 PBMC [9]，但与五年后预测的 1.6 log10 拷贝/10 6 PBMC 相似在同一队列中，个体在 HIV 感染后 15 天内开始接受 ART 治疗 [5]。ANRS-CODEX HIV 控制者队列中 HIV DNA 动力学的数学模型显示，46% 的参与者显着下降。在进入队列时，HLA-B*27/B*57 等位基因和较低水平的血浆 HIV RNA 和 HIV DNA 与 HIV DNA 下降独立相关。作者假设宿主细胞对 HIV 感染的内在抵抗力，较低的免疫激活水平导致潜在的靶细胞减少，较低的残留 HIV 复制和较短的受感染细胞半衰期可能都是观察到的 HIV DNA 水平下降的原因。相比之下，在随访期间，非 HLA-B*27/B*57 等位基因或持续性 HIV RNA ≥1 log10 拷贝/mL 与同一对照队列中 HIV DNA 随时间增加有关。通过 DNA 测序对 HIV 原病毒进行表征，并对控制器中受感染细胞的克隆扩增进行分析，可以进一步了解所观察到的不同 HIV DNA 动力学背后的机制 [10]。此类分析将区分具有相同原病毒的细胞的克隆扩增与持续感染周期，这是导致 HIV DNA 水平升高的原因。尽管 HIV 控制者的 HIV DNA 水平下降很有趣，它并不一定意味着可以产生传染性病毒并导致病毒反弹的潜在但具有复制能力（完整）的原病毒库减少。这是因为除了整合的原病毒外，未整合的线性和游离 DNA 也可以通过总 HIV DNA 定量检测到，这两种 DNA 都不能产生感染性病毒。与接受治疗和未接受治疗的非控制者相比，还发现 HIV 控制者的未整合 DNA 占总 HIV DNA 的比例更高 [11]。此外，绝大多数整合的前病毒 DNA 是有缺陷的，不能导致产生感染性病毒粒子 [12]。虽然旨在量化具有复制能力的原病毒的分析可能能够提供对潜在储库的更准确估计，这些检测是劳动密集型的，可能缺乏检测小波动的动态范围 [13]。新的基于 PCR 的完整原病毒检测应能进一步了解 ANRS-CODEX 队列中 HIV 控制者的 HIV DNA 水平变化是否与完整病毒库的变化平行。Hsu DC 和 Mellors JW Journal of the International AIDS Society 2019, 22:e25254 http://onlinelibrary.wiley.com/doi/10.1002/jia2.25254/full | https://doi.org/10.1002/jia2.25254 1002/jia2.25254/全| https://doi.org/10.1002/jia2.25254 1002/jia2.25254/全| https://doi.org/10.1002/jia2.25254