Frontiers in Bioinformatics (May 2024)

Decreased but persistent epigenetic age acceleration is associated with changes in T-cell subsets after initiation of highly active antiretroviral therapy in persons living with HIV

  • Mary E. Sehl,
  • Mary E. Sehl,
  • Elizabeth Crabb Breen,
  • Roger Shih,
  • Fengxue Li,
  • Joshua Zhang,
  • Peter Langfelder,
  • Peter Langfelder,
  • Steve Horvath,
  • Steve Horvath,
  • Steve Horvath,
  • Jay H. Bream,
  • Priya Duggal,
  • Jeremy Martinson,
  • Steven M. Wolinsky,
  • Otoniel Martinez-Maza,
  • Christina M. Ramirez,
  • Beth D. Jamieson

DOI
https://doi.org/10.3389/fbinf.2024.1356509
Journal volume & issue
Vol. 4

Abstract

Read online

IntroductionPersons living with HIV (PLWH) experience the early onset of age-related illnesses, even in the setting of successful human immunodeficiency virus (HIV) suppression with highly active antiretroviral therapy (HAART). HIV infection is associated with accelerated epigenetic aging as measured using DNA methylation (DNAm)-based estimates of biological age and of telomere length (TL).MethodsDNAm levels (Infinium MethylationEPIC BeadChip) from peripheral blood mononuclear cells from 200 PLWH and 199 HIV-seronegative (SN) participants matched on chronologic age, hepatitis C virus, and time intervals were used to calculate epigenetic age acceleration, expressed as age-adjusted acceleration residuals from 4 epigenetic clocks [Horvath’s pan-tissue age acceleration residual (AAR), extrinsic epigenetic age acceleration (EEAA), phenotypic epigenetic age acceleration (PEAA), and grim epigenetic age acceleration (GEAA)] plus age-adjusted DNAm-based TL (aaDNAmTL). Epigenetic age acceleration was compared for PLWH and SN participants at two visits: up to 1.5 years prior and 2–3 years after HAART (or equivalent visits). Flow cytometry was performed in PLWH and SN participants at both visits to evaluate T-cell subsets.ResultsEpigenetic age acceleration in PLWH decreased after the initiation of HAART but remained greater post-HAART than that in age-matched SN participants, with differences in medians of 6.6, 9.1, and 7.7 years for AAR, EEAA, and PEAA, respectively, and 0.39 units of aaDNAmTL shortening (all p < 0.001). Cumulative HIV viral load after HAART initiation was associated with some epigenetic acceleration (EEAA, PEAA, and aaDNAmTL), but even PLWH with undetectable HIV post-HAART showed persistent epigenetic age acceleration compared to SN participants (p < 0.001). AAR, EEAA, and aaDNAmTL showed significant associations with total, naïve, and senescent CD8 T-cell counts; the total CD4 T-cell counts were associated with AAR, EEAA, and PEAA (p = 0.04 to <0.001). In an epigenome-wide analysis using weighted gene co-methylation network analyses, 11 modules demonstrated significant DNAm differences pre- to post-HAART initiation. Of these, nine were previously identified as significantly different from pre- to post-HIV infection but in the opposite direction.DiscussionIn this large longitudinal study, we demonstrated that, although the magnitude of the difference decreases with HAART is associated with the cumulative viral load, PLWH are persistently epigenetically older than age-matched SN participants even after the successful initiation of HAART, and these changes are associated with changes in T-cell subsets.

Keywords