Distinct intestinal microbial signatures linked to accelerated systemic and intestinal biological aging

Abstract

Background: People living with HIV (PLWH), even when viral replication is controlled through antiretroviral therapy (ART), experience persistent inflammation. This inflammation is partly attributed to intestinal microbial dysbiosis and translocation, which may lead to non-AIDS-related aging-associated comorbidities. The extent to which living with HIV - influenced by the infection itself, ART usage, sexual orientation, or other associated factors - affects the biological age of the intestines is unclear. Furthermore, the role of microbial dysbiosis and translocation in the biological aging of PLWH remains to be elucidated. To investigate these uncertainties, we used a systems biology approach, analyzing colon and ileal biopsies, blood samples, and stool specimens from PLWH on ART and people living without HIV (PLWoH) as controls.

Results: PLWH exhibit accelerated biological aging in the colon, ileum, and blood, as measured by various epigenetic aging clocks, compared to PLWoH. Investigating the relationship between microbial translocation and biological aging, PLWH had decreased levels of tight junction proteins in the intestines, along with increased microbial translocation. This intestinal permeability correlated with faster biological aging and increased inflammation. When investigating the relationship between microbial dysbiosis and biological aging, the intestines of PLWH had higher abundance of specific pro-inflammatory bacteria, such as Catenibacterium and Prevotella. These bacteria correlated with accelerated biological aging. Conversely, the intestines of PLWH had lower abundance of bacteria known for producing the anti-inflammatory short-chain fatty acids, such as Subdoligranulum and Erysipelotrichaceae, and these bacteria were associated with slower biological aging. Correlation networks revealed significant links between specific microbial genera in the colon and ileum (but not in feces), increased aging, a rise in pro-inflammatory microbe-related metabolites (e.g., those in the tryptophan metabolism pathway), and a decrease in anti-inflammatory metabolites like hippuric acid.

Conclusions: We identified specific microbial compositions and microbiota-related metabolic pathways that are intertwined with intestinal and systemic biological aging. This microbial signature of biological aging is likely reflecting various factors including the HIV infection itself, ART usage, sexual orientation, and other aspects associated with living with HIV. A deeper understanding of the mechanisms underlying these connections could offer potential strategies to mitigate accelerated aging and its associated health complications. Video Abstract.

Keywords: Aging clocks; Biological aging; Gut; HIV; Intestines; Metabolome; Microbiome.

Fig. 1

Accelerated intestinal and blood biological aging in PLWH on ART. A Study design schematic. B Chronological age comparison between PLWoH and PLWH on ART, displaying median and interquartile range (IQR). Statistical analysis was performed using the Mann–Whitney U test. C-E Dot-plots illustrating the significant acceleration in biological age for blood (C), ileum (D), and colon (E) based on multiple epigenetic clocks: Horvath1, Horvath2, Hannum, PhenoAge, DNAmTL, GrimAge, and DunedinPACE for blood; Horvath1, Horvath2, Hannum, and DNAmTL for ileum; Horvath1 and DunedinPACE for colon. Positive graph values indicate accelerated biological aging differences per clock, while negative values for DNAmTL signify telomere reduction differences. Both median and IQR are depicted, with statistical analysis via Mann–Whitney U tests. F Dot-plots contrasting tissue-specific accelerated biological age differences in blood, ileum, and colon utilizing Horvath1. Statistical assessments were conducted using the non-parametric Friedman test, corrected using the Two-stage step-up method of Benjamini, Krieger, and Yekutieli

Fig. 2

The rate of acceleration in biological age, as assessed by epigenetic clocks, corresponds to other established and emerging methods for measuring biological aging. A Spearman's rank correlation heatmap displaying the correlations between telomere lengths in blood (rows) and DNA methylation-based biological aging in blood, ileum, and colon (columns), as gauged by various epigenetic clocks including Horvath1, Horvath2, Hannum, PhenoAge, DNAmTL, and GrimAge. Positive and negative correlations are illustrated in red and blue, respectively. B Dot plots depict elevated inflammatory markers in plasma for PLWH on ART compared to PLWoH. Data is represented by medians and interquartile ranges (IQR), with each dot signifying an individual. Comparisons were drawn using the Mann–Whitney U test. C Heatmaps of Spearman’s rank correlations between plasma-based inflammatory markers of aging (rows) and accelerated epigenetic age in blood, ileum, and colon (columns), as estimated by the indicated epigenetic clocks. Positive and negative correlations are illustrated in red and blue, respectively. * P < 0.05, ** P < 0.01, *** P < 0.001

Fig. 3

Intestinal permeability and microbial translocation are associated with increased rates of accelerated biological aging. A-B Ileum (A) and colon (B) samples from PLWoH and PLWH on ART depict ZO-1 or occludin expression (green). Nuclei were stained with DAPI (blue). Images, captured at 40 × magnification on a Zeiss Axio Observer 7 microscope, have a scale bar of 20 μm. Tight junction scores are presented as medians with IQR. Statistical significance was determined using Mann–Whitney U tests. C Elevated microbial translocation markers in PLWH on ART plasma compared to PLWoH, represented as median with IQR. Statistical significance determined using Mann–Whitney U tests. D Heatmaps of Spearman's rank correlations between tight junction integrity (top rows) and microbial translocation (bottom rows) with accelerated biological aging (left columns) and inflammatory markers (right columns). Positive and negative correlations are colored in red and blue, respectively. * P < 0.05, ** P < 0.01, *** P < 0.001. E–H Scatter plots display Spearman's rank correlations between HIV DNA/RNA levels in various tissues and tight junction scores: (E) HIV DNA in colon vs occludin score in colon, (F) HIV DNA in PBMC vs occludin score in colon, (G) HIV RNA in PBMC vs occludin score in colon, and (H) HIV RNA in PBMC vs ZO-1 score in ileum

Fig. 4

Living with HIV is associated with distinct intestinal and fecal microbial dysbiosis, characterized by a reduction in butyrate-producing bacteria. A Alpha diversity indices (Richness, Shannon, Faith’s phylogenetic diversity) reveal reduced colon microbiome diversity in PLWH on ART vs. controls. B Butyrate-producing microbiota's relative abundance in feces, ileum, and colon for both study groups are illustrated. Medians and IQR are depicted, with significance derived from Mann–Whitney U tests. C Differential bacterial abundance between tissue and fecal samples on a logarithmic scale. Comparisons include colon vs. feces (red), ileum vs. feces (blue), and ileum vs. colon (green). Significance markers: FDR < 0.05 (closed circles) and FDR > 0.05 (open circles). Adjustments made for multiple tests using the Benjamini–Hochberg method. D Log-scale differences in bacterial abundance across colon (triangle), ileum (square), and feces (circle) between PLWoH and PLWH on ART. Analysis incorporated bacterial taxa with > 1% mean relative abundance. Linear models estimated abundance changes, and adjustments for multiple tests used the Benjamini–Hochberg method. Significance markers: FDR < 0.05 (purple), FDR < 0.1 (cyan), and P < 0.05 (green)

Fig. 5

Microbial signature in mucosa linked to accelerated biological aging. A Spearman’s rank correlation analysis depicts associations between colonic microbiomes enriched (top rows) or depleted (bottom rows) in PLWH on ART and accelerated biological aging in the colon and blood (columns). Red and blue signify correlations with P < 0.05. White indicates P > 0.05. *FDR < 10%. B Plots represent correlations between specific bacterial taxa in the colon and accelerated epigenetic aging: Catenibacterium vs. DunedinPACE (colon) [left]; Prevotella 9 vs. DunedinPACE (colon) [middle]; and Erysipelotrichaceae UCG-003 vs. GrimAge (blood) [right]. C-D Spearman’s rank correlation heatmaps display association of ileal microbiota (C) and fecal microbiota (D) with accelerated aging in different regions. Red indicates positive and blue indicates negative correlations with P < 0.05; white spaces show P > 0.05. E Spearman’s rank correlation heatmap illustrates correlations between microbiomes in the colon (top), ileum (middle), and feces (bottom) with tight junction integrity, microbial translocation, and inflammatory markers. F Spearman’s rank correlation heatmap depicts associations between pro-inflammatory, butyrate, and SCFA-producing microbiota in the ileum or colon with HIV DNA/RNA levels. Red signifies positive and blue indicates negative correlations. * P < 0.05, ** P < 0.01, *** P < 0.001. G Spearman’s rank correlations are demonstrated between the relative abundance of pro-inflammatory microbiome (ileum) vs. HIV DNA/RNA in ileum and PBMCs (top). The bottom panel shows correlations of SCFA-producing microbiome (ileum) vs. HIV DNA/RNA in ileum and PBMCs

Fig. 6

Correlation network reveals specific links between the mucosal microbiome, microbiota-associated metabolites, and accelerated biological aging. A A heatmap displays the relative abundance of metabolites associated with gut health in both stool and plasma samples from PLWH on ART and PLWoH. The color gradient, from blue to red, signifies the normalized metabolite values, with red representing higher abundance and blue indicating lower abundance. Differences between groups were assessed using the Mann–Whitney U test. B Spearman’s rank correlation heatmap demonstrates relationships between gut-specific metabolites (rows) and indicators of accelerated biological aging, tight junction integrity, microbial translocation, and inflammatory markers (columns). The top panel highlights metabolites upregulated in PLWH on ART, while the bottom section presents metabolites that are downregulated in this group. Positive and negative correlations are represented by red and blue, respectively. Key: * P < 0.05, ** P < 0.01, *** P < 0.001. C Circos plots visualize Spearman’s rank correlations among tissue-specific microbiome (yellow), metabolites derived from plasma and stool samples (green), and indicators of accelerated biological aging specific to both tissue and blood (blue). Red lines denote significant positive correlations, while blue lines signify negative correlations. Only correlations manifesting a 3-way association are included. Emphasized lines hint at their potential functional relevance. 5-HIAA stands for 5-Hydroxyindole-3-acetic acid, K/T ratio stands for Kynurenine/Tryptophan ratio, and Q/T ratio stands for Quinolinic acid/Tryptophan ratio