Atypical Skeletal Muscle Profiles in Human Immunodeficiency Virus-Infected Asymptomatic Middle-Aged Adults

Abstract

Background: Human immunodeficiency virus (HIV)-infected individuals are at increased risk of age-associated functional impairment, even with effective antiretroviral therapy (ART). A concurrent characterization of skeletal muscle, physical function, and immune phenotype in aviremic middle-aged HIV-infected adults represents a knowledge gap in prognostic biomarker discovery.

Methods: We undertook a prospective observational study of 170 middle-aged, HIV-infected ambulatory men and women with CD4+ T-cell counts of at least 350/µL and undetectable plasma viremia while on effective ART, and uninfected control participants. We measured biomarkers for inflammation and immune activation, fatigue, the Veterans Aging Cohort Study mortality index, and physical function. A subset also received a skeletal muscle biopsy and computed tomography scan.

Results: Compared to the uninfected participants, HIV-infected participants displayed increased immune activation (P < .001), inflammation (P = .001), and fatigue (P = .010), and in a regression model adjusting for age and sex displayed deficits in stair-climb power (P < .001), gait speed (P = .036), and predicted metabolic equivalents (P = .019). Skeletal muscle displayed reduced nuclear peroxisome proliferator-activated receptor-γ coactivator 1α-positive myonuclei (P = .006), and increased internalized myonuclei (P < .001) that correlated with immune activation (P = .003) and leukocyte infiltration (P < .001). Internalized myonuclei improved a model for HIV discrimination, increasing the C-statistic from 0.84 to 0.90.

Conclusions: Asymptomatic HIV-infected middle-aged adults display atypical skeletal muscle profiles, subclinical deficits in physical function, and persistent inflammation and immune activation. Identifying biomarker profiles for muscle dysregulation and risk for future functional decline in the HIV-infected population will be key to developing and monitoring preventive interventions.

Clinical trials registration: NCT03011957.

Figure 1.

Consolidated Standards of Reporting Trials (CONSORT) flow diagram of study enrollment and characterization in the Muscle and Aging in Treated Chronic human immunodeficiency virus (HIV) (MATCH) cohort. The baseline characterization period was between April 2015 and August 2016. A total of 215 adult men and women from the Boston metropolitan area, between the ages of 50 and 65 years, consented. A total of 170 participants (71 HIV+-infected and 99 uninfected) were found eligible and enrolled. Participants were excluded from the analyses if they were outside of the target age range, had detectable viremia (>200 copies/mL), had a CD4 count <350 cells/µL, were nonambulatory, were taking anabolic agents, or had a recent infection. Participants were excluded from the biopsy and computed tomographic (CT) scan procedures based on lack of consent, medical ineligibility, and closure of enrollment.

Figure 2.

Immune activation and circulating inflammatory biomarkers. Data were plotted as natural log (Ln) vs human immunodeficiency virus (HIV) status. The Mann-Whitney U test was used to compare the 2 groups. Bars represent medians. P < .05 was considered statistically significant. Immune activation was based on flow cytometric expression of human leukocyte antigen – antigen D related (HLA-DR) and CD38 (A), CD38 (B), and HLA-DR (C) on CD8 T cells. Circulating inflammatory biomarkers were measured in serum including C-reactive protein (CRP; D), soluble CD14 (sCD14; E), and soluble CD163 (sCD163; F).

Figure 3.

Cross-sectional area (CSA) of mid-thigh muscle. Data were plotted as natural log (Ln) vs human immunodeficiency virus (HIV) status. Bars represent medians. The Mann-Whitney U test was used for between-group comparisons. Computed tomographic images of mid-thigh muscle were analyzed using ImageJ software. A, Anterior CSA measurements (Hounsfield units [HU] from –190 to 100) based on sex (B) and HIV status of women (C) and men (D) are shown.

Figure 4.

Muscle fiber type and size distribution based on sex and human immunodeficiency virus (HIV) status. The unpaired t test was used to compare the 2 groups within each bin. *P < .05; **P < .01. Muscle sections were stained with slow-twitch type I myosin heavy chain (MYH) and laminin. Images were taken with the ×20 objective lens in a grid of 15 × 10. Myofibers were analyzed with the image analysis software Cell Profiler. Laminin signals were used to outline myofibers. Fibers with a MYH median intensity <0.3 and those ≥0.3 were assigned as type II fast-twitch (fMYH) and type I slow-twitch (sMYH) fibers, respectively. The distribution of myofiber size (minimum Feret diameter [MFD], defined as the minimum distance of parallel tangents at opposing borders of the muscle fiber [60]) is shown based on sex (A and D; 36 male and 18 female) and HIV status of female (B and E; 7 HIV⁺ and 11 HIV^–) and male (C and F; 22 HIV⁺ and 14 HIV^–).

Figure 5.

Skeletal muscle profiles based on peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α) localization, internalized nuclei, and collagen. The Mann-Whitney U test was used to compare the 2 groups. P < .05 was considered statistically significant. A–C, All muscle fibers were fluorescently stained (A) with type I collagen (red), Hoechst (blue), and PGC-1α (reddish-brown) stained with peroxidase (B). C, Images were merged showing collagen (magenta), Hoechst (yellow), and PGC-1α (reddish-brown) with evident of internalized myonuclei in skeletal muscle of human immunodeficiency virus–infected (HIV⁺) and –uninfected (HIV^–) Muscle and Aging in Treated Chronic HIV (MATCH) participants (orange circles). Measurements of nuclear PGC-1α (D), internalized myonuclei (E), and collagen (F) are shown. Bars represent medians; 29 HIV⁺ and 25 HIV^–. Representative muscle fibers from HIV⁺ participants are shown.

Figure 6.

Receiver operating characteristic (ROC) curves associated with human immunodeficiency virus (HIV) infection. Curves are based on multivariate logistic regression models of the prediction of risk with the use of composite scores for inflammation and immune activation with or without internalized myonuclei. The area under the ROC curve (AUC) was used to assess internalized myonuclei as a discriminating biomarker for HIV status. As a primary analysis, we considered models with composite immune activation score and inflammation score as biomarkers with and without internalized myonuclei.