Candidate host epigenetic marks predictive for HIV reservoir size, responsiveness to latency reversal, and viral rebound

Abstract

Objective: This study aimed to identify candidate host epigenetic biomarkers predicting latency reversal agents (LRA) efficacy and HIV-1 rebound kinetics during analytical treatment interruption (ATI).

Design: Retrospective longitudinal epigenetic profiling study from 13 people with HIV (PWH) on virologically suppressive antiretroviral therapy (ART) that participated in a LRA (HDAC inhibitor) clinical trial (NCT01680094) and a subsequent optional ATI to monitor for viral recrudescence after ART cessation.

Methods: Genome-wide DNA methylation (DNAm) in purified CD4+ T cells was measured at single-nucleotide resolution using the Infinium MethylationEPIC array. HIV-1 DNA and RNA measures were previously assessed by PCR-based methods and the association of DNAm levels at regulatory sites of the human genome were examined with reservoir size, responsiveness to LRA, and time to viral rebound following ATI.

Results: A distinct set of 15 candidate DNAm sites in purified CD4+ T cells at baseline pre-LRA and pre-ATI significantly correlated with time to viral rebound. Eight of these DNAm sites occurred in genes linked to HIV-1 replication dynamics including (SEPSECS, cg19113954), (MALT1, cg15968021), (CPT1C, cg14318858), (CRTAM, cg10977115), (B4GALNT4, cg04663285), (IL10, cg16284789), (TFPI2, cg19645693), and (LIFR, cg26437306); with the remaining sites at intergenic regions containing regulatory elements. Moreover, baseline DNAm states related to total HIV-1 DNA levels and the fold change in unspliced cell-associated HIV RNA following LRA treatment.

Conclusion: Preexisting host epigenetic states may determine HIV-1 rebound kinetics and reservoir maintenance. These findings suggest integrating a suite of DNA methylation markers to improve optimal participant selection and drug regimen in future HIV cure clinical trials.

Conflict of interest statement

CONFLICT OF INTEREST

LCN has served as a scientific advisor for Abbvie, ViiV and Cytodyn for work unrelated to this project. OSS has served as a scientific advisor for Abbvie, Gilead, Mologen AG, and Immunocore for work unrelated to this project. All other authors declare no competing interests.

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.

Figures

Figure 1:

Time points of DNA methylation analysis of CD4+ T lymphocytes from study participants in a latency-reversal clinical trial (NCT01680094). Study design of 20 mg panobinostat dosing three times per week every other week for 8 week duration. Baseline, early (8 h postdosing), late (following three dosing regimens), and at follow-up (post 4 weeks after last dosing). Red star indicates timepoint of DNA methylation utilized to relate to HIV-1 reservoir size, responsiveness to LRA (Panobinostat), and days to viral rebound following ATI.

Figure 2:

Correlogram of DNA methylation markers of host CD4+ T lymphocytes pre-ATI related with time to viral rebound during ATI (>1000 copies per mL). Solid color square indicates a Spearman correlation with P

Figure 3:

DNA methylation sites associated with cell associated unspliced HIV-1 increase following LRA and HIV-1 total DNA. A. Pre-ATI DNA methylation levels at annotated enhancer region of HDAC4 gene and B. open chromatin 5’UTR region of HDAC9 correlated with magnitude of HIV-1 RNA following LRA treatment. C. Pre-ATI DNA methylation levels at ATP13A3, D. BTN3A2, E. MGRN1, and F. CD86 genes relate to log10 total HIV-1 DNA levels assayed pre-ATI. Methylation β values represent a quantification at each CpG site of β:=M/(M+U+a), where M>0 and U>0 denote the methylated and unmethylated signal intensities.

Figure 4:

Venn diagram of overlap between differentially methylated loci comparing baseline, early Panobinostat treatment, late Panobinostat treatment, and post Panobinostat time points.

Figure 5:

Differentially methylated loci impacted by early and late Panobinostat treatment in CD4+ T cells. The grey-scaled box represents panobinostat treatment period (on-panobinostat during the first (early) and third (late) treatment cycle). Post timepoint is four weeks post panobinostat. Data are shown as median and error bars as interquartile range. *P