Profound depletion of HIV-1 transcription in patients initiating antiretroviral therapy during acute infection

Abstract

Background: Although combination antiretroviral therapy (cART) initiated in the acute phase of HIV-1 infection may prevent expansion of the latent reservoir, its benefits remain controversial. In the current study, HIV-1 RNA transcription patterns in peripheral blood mononuclear cells (PBMC) were monitored during acute cART to assess the effect of early treatment on cellular viral reservoirs.

Methodology/principal findings: Acutely HIV-1 infected patients (n = 24) were treated within 3-15 weeks after infection. Patients elected to cease treatment after ≥1 year of therapy. HIV-1 DNA (vDNA), HIV-1 RNA species expressed both in latently and productively infected cells, unspliced (UsRNA), multiply spliced (MsRNA-tatrev; MsRNA-nef), and PBMC-associated extracellular virion RNA (vRex), expressed specifically by productively infected cells, were quantified in PBMC by patient matched real-time PCR prior, during and post cART. In a matched control-group of patients on successful cART started during chronic infection (n = 15), UsRNA in PBMC and vDNA were measured cross-sectionally. In contrast to previous reports, PBMC-associated HIV-1 RNAs declined to predominantly undetectable levels on cART. After cART cessation, UsRNA, vRex, and MsRNA-tatrev rebounded to levels not significantly different to those at baseline (p>0.1). In contrast, MsRNA-nef remained significantly lower as compared to pretreatment (p = 0.015). UsRNA expressed at the highest levels of all viral RNAs, was detectable on cART in 42% of patients with cART initiated during acute infection as opposed to 87% of patients on cART initiated during chronic infection (Fisher's exact test; p = 0.008). Accordingly, UsRNA levels were 105-fold lower in the acute as compared to the chronic group.

Conclusion: Early intervention resulted in profound depletion of PBMC expressing HIV-1 RNA. This is contrary to chronically infected patients who predominantly showed continuous UsRNA expression on cART. Thus, antiretroviral treatment initiated during the acute phase of infection prevented establishment or expansion of long-lived transcriptionally active viral cellular reservoirs in peripheral blood.

Conflict of interest statement

Competing Interests: HFG has served as a consultant and medical advisor for Abbott Laboratories, Bristol-Myers Squibb, Boehringer Ingelheim Pharmaceuticals, Gilead Sciences, GlaxoSmithKline, Pfizer, Tibotec Therapeutics, and ViiV Healthcare and has received unrestricted research and educational grants from Abbott Laboratories, Bristol-Myers Squibb, Gilead Sciences, Merck Sharp & Dohme, and Pfizer. No company was involved in the study design, in generation, analysis and interpretation of any data. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.

Figures

Figure 1. Map of HIV-1 amplica for qPCR assays.

Splice acceptors and donor sites are shown by red and blue vertical lines, respectively. Exons are depicted in grey bars. Blue dotted lines show documented or predicted splice events. Sense primers used in qPCR assays are depicted by red arrows antisense/cDNA primers by blue arrows. Fluorescent hydrolysis probes (FH-probes/TaqMan®-probes) are shown by green/pink bars. Assays for MsRNAs share a common antisense/cDNA primer and in some instances, when a probe for MsRNA-tatrev was not available, a common FH-probe 3′ of the splice-acceptor of the 2nd major intron. Note that this map does not show mRNAs encoding Env because these transcripts were not assessed in the current study.

Figure 2. Longitudinal course of HIV-1 virion production.

Levels of plasma viremia (A) and of PBMC-associated virions, vRex (B) of 24 acutely infected patients, were plotted against time after initiation of cART (0 weeks, left panels) or against time after cART cessation (defined as 0 weeks in the right panels). Note that some data points are depicted in the left panels are also shown in the right panels to facilitate visualization of the effects of cART cessation. The amount of vRex measurements is lower, because cell sampling was done less frequently than plasma HIV-RNA was measured. Data within the grey horizontal areas show PCR-negative samples below the detection limit of the assays applied. The black line in panel A depicts the clinically used threshold for plasma viremia of 50 RNA copies/ml.

Figure 3. Empirical distribution of times to virological end-points.

Time to event analysis showing the percentage of patients before reaching their first PCR-negative measurement after initiation of cART (A) or reaching measurable viral RNA levels above a defined threshold after cessation of cART (B). Thresholds are indicated in parentheses in panel B and were defined as approximately 3-fold elevation over mean on cART levels for cellular HIV-1 RNAs. Threshold of plasma viremia was set at 50 copies/ml. On average vRex (red lines), MsRNAs (magenta), UsRNA (blue) and plasma viremia (red broken lines) dropped to undetectable levels 4.5, 8.4, 13.5, and 24.6 weeks after initiation of cART, respectively (A) and rebounded within 17.9, 9.8, 8.0, and 4.2 weeks after treatment cessation (B), (dotted lines). Note that for analysis of plasma viremia only time-points were considered for which also UsRNA and MsRNA measurements had been available.

Figure 4. Virological parameters before, during, and after cART.

Viral nucleic acid levels at baseline (diamonds), and average values during the time windows defined as on cART (circles, vRex ≥2 weeks, plasma viremia, PBMC-associated UsRNA and MsRNAs ≥24 weeks) and post-cessation maximal values (triangles). Data-points within the grey horizontal areas show PCR-negative samples below the detection limit of the assays applied. P-values within panels show significance levels (Wilcoxon signed rank test) between baseline and post-cessation. Bars show medians and quartiles. The dotted line in panel A depicts the clinically used threshold for plasma viremia of 50 RNA copies/ml.

Figure 5. Viral nucleic acid expression in acute and chronic patients during cART.

Mean levels of vDNA (grey symbols) and UsRNA (blue symbols) (A) or viral transcription rates (B) of patients that had initiated cART during acute infection (acute, n = 24, circles) were assessed during the on-cART time window (≥24 weeks) and compared to single measurements in patients that had initiated cART during chronic infection (chronic, n = 15, triangles, median treatment time 50 weeks, range 36–183, see Table S1 for details). Data-points within the grey horizontal areas show PCR-negative samples below the detection limit of the assays applied. P-values within panels show significance levels (Mann-Whitney test) between the acute and the chronic group. Bars show medians and quartiles.