Prolonged Antibiotic Use in a Preclinical Model of Gulf War Chronic Multisymptom-Illness Causes Renal Fibrosis-like Pathology via Increased micro-RNA 21-Induced PTEN Inhibition That Is Correlated with Low Host Lachnospiraceae Abundance

Abstract

Gulf War (GW) veterans show gastrointestinal disturbances and gut dysbiosis. Prolonged antibiotic treatments commonly employed in veterans, especially the use of fluoroquinolones and aminoglycosides, have also been associated with dysbiosis. This study investigates the effect of prolonged antibiotic exposure on risks of adverse renal pathology and its association with gut bacterial species abundance in underlying GWI and aims to uncover the molecular mechanisms leading to possible renal dysfunction with aging. Using a GWI mouse model, administration of a prolonged antibiotic regimen involving neomycin and enrofloxacin treatment for 5 months showed an exacerbated renal inflammation with increased NF-κB activation and pro-inflammatory cytokines levels. Involvement of the high mobility group 1 (HMGB1)-mediated receptor for advanced glycation end products (RAGE) activation triggered an inflammatory phenotype and increased transforming growth factor-β (TGF-β) production. Mechanistically, TGF-β- induced microRNA-21 upregulation in the renal tissue leads to decreased phosphatase and tensin homolog (PTEN) expression. The above event led to the activation of protein kinase-B (AKT) signaling, resulting in increased fibronectin production and fibrosis-like pathology. Importantly, the increased miR-21 was associated with low levels of Lachnospiraceae in the host gut which is also a key to heightened HMGB1-mediated inflammation. Overall, though correlative, the study highlights the complex interplay between GWI, host gut dysbiosis, prolonged antibiotics usage, and renal pathology via miR-21/PTEN/AKT signaling.

Keywords: Gulf War Illness; Lachnospiraceae spp.; PTEN; Prolong antibiotics; TGF-β; miR-21; renal fibrosis.

Figure 1

Analysis of Western blot data revealed that prolonged antibiotic administration (neomycin and enrofloxacin) in GWI mice resulted in increased kidney inflammation and elevated levels of pro-inflammatory cytokines, namely, IL-1β and IL-17A. Representative (A) depicts protein expression of phosphorylated-p65, total-p65, pro-IL-1β, cleaved IL-1 β, IL-17A, and β-actin in kidney tissue of CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). (B–D) represents densitometry analysis of phosphorylated-p65 normalized with total-p65 (B), cleaved IL-1β normalized with pro-IL-1β (C), and IL-17A normalized with β-actin (D) in CONTROL, GWI, GWI+AB, and AB mice plotted as bar graph. The data are presented as the mean ±  SD (SD: standard deviation) and statistical significance was measured using one-way ANOVA between all the groups, with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 2

Immunohistochemical analysis and mRNA expression levels demonstrated that prolonged antibiotic administration (neomycin and enrofloxacin) in GWI mice led to an increase in the production of pro-inflammatory cytokines. Representative immunohistochemistry images (A) depict immunoreactivity of TNF-α in kidney sections from CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). Images were captured at a 40× magnification (scale bar = 50 µm). Immunoreactivity was denoted by the presence of red arrows and morphometric analysis was calculated as %ROI. (B) The bar graph represents the morphometric analysis of TNF-α in CONTROL, GWI, GWI+AB, and AB mice. The data are presented as the mean ±  SD (SD: standard deviation) of %ROI (mean value calculated from five different fields in each sample). Bar graphs (C–F) illustrate the normalized mRNA expression levels of different pro-inflammatory cytokines, namely IL-1 β (C), IL-17A (D), and TNF-α (E), and chemokines, specifically MCP-1 (F) against GAPDH. These expression levels are presented as fold changes relative to the CONTROL derived from kidney tissue samples of CONTROL, GWI, GWI+AB, and AB mice. The experiment was performed in triplicates and the data are represented as mean ± SEM (SEM: standard error mean). Statistical significance was determined using one-way ANOVA with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 3

Immunofluorescence analysis demonstrated that prolonged antibiotic administration in GWI mice led to an augmentation in RAGE activation mediated by HMGB1. Representative (A) immunofluorescence images depict co-localization events (yellow) of HMGB1 (in red) and RAGE (in green) within kidney sections from CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). The kidney sections were additionally counterstained with DAPI (in blue), and all images were acquired at a 40× magnification (scale bar = 50 µm). Immunoreactivity was denoted by the presence of white arrows and morphometric analysis was calculated as %ROI. (B) The bar graph represents HMGB1-RAGE co-localization events in CONTROL, GWI, GWI+AB, and AB mice. The data are presented as the mean ±  SD (SD: standard deviation) of %ROI (mean value calculated from five different fields in each sample) and statistical significance was measured using one-way ANOVA between all the groups, with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 4

Immunohistochemical analysis demonstrated that prolonged antibiotic administration in GWI mice led to an increase in renal TGF-β production. Representative immunohistochemistry images (A) depict immunoreactivity of TGF-β in kidney sections from CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). Images were captured at a 40× magnification (scale bar = 50 µm). Immunoreactivity was denoted by the presence of red arrows and morphometric analysis was calculated as %ROI. (B) The bar graph represents a morphometric analysis of TGF-β in CONTROL, GWI, GWI+AB, and AB mice. The data are presented as the mean ± SD (SD: standard deviation) of %ROI (mean value calculated from five different fields in each sample). Statistical significance was determined using one-way ANOVA with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 5

Prolonged antibiotic administration in GWI mice mediated the upregulation of micro-RNA 21 (miR-21), a signature miRNA for renal fibrosis, and correlated with Lachnospiraceae abundance. (A) represents miR-21 expression analyzed by qRT-PCR in the kidney tissue of CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). The miR-21 expression was normalized using miR-103-3p as endogenous control. The bar graph is represented as fold change against CONTROL. (B) represents a bar graph showing the percent relative abundance of Lachnospiraceae spp. in CONTROL, GWI, GWI+AB, and -AB groups following whole genome sequencing of mouse fecal pellets. (C) depicts the correlation analysis of Lachnospiraceae spp. relative abundance and miR-21 levels in renal tissue of CONTROL, GWI, GWI+AB, and AB groups. Pearson’s linear regression is shown in red with 95% confidence bands. The experiment was performed in triplicates and the data are represented as mean ± SEM (SEM: standard error mean). Statistical significance was determined using one-way ANOVA with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 6

The protein expression of PTEN, a target of miR-21, is perturbed due to prolonged antibiotic treatment in GWI mice, thereby modulating AKT signaling, as PTEN functions as a negative regulator of AKT signaling. (A) represents Western blot analysis of the relative protein expression of PTEN, phosphorylated-AKT, total-AKT, and β-actin in kidney tissue of CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). (B,C) represents a bar graph illustrating the densitometry analysis of PTEN normalized with β-actin (B) and phosphorylated-AKT normalized with total-AKT (C) in CONTROL, GWI, GWI+AB, and AB mice plotted as a bar graph. The data are presented as the mean ± SD (SD: standard deviation) and statistical significance was measured using one-way ANOVA between all the groups, with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.

Figure 7

Prolonged antibiotic treatment in GWI mice triggers the activation of AKT signaling, resulting in increased extracellular matrix deposition (ECM) and the induction of epithelial–mesenchymal transition (EMT)-like phenotype. Representative immunohistochemistry images (A) depict immunoreactivity of Fibronectin and α-SMA in kidney sections from CONTROL, GWI (persistent Gulf War Illness), GWI+AB (persistent Gulf War Illness with prolonged antibiotics exposure), and AB mice (prolonged antibiotics exposure only) (n = 6/group). Images were captured at a 40× magnification (scale bar = 50 µm). Immunoreactivity was denoted by the presence of red arrows and morphometric analysis was calculated as %ROI. (B,C) represents a bar graph illustrating the morphometric analysis of Fibronectin (Figure 6B) and α-SMA (Figure. 6C) in CONTROL, GWI, GWI+AB, and AB mice. The data are presented as the mean ±  SD (SD: standard deviation) of %ROI (mean value calculated from five different fields in each sample). Statistical significance was determined using one-way ANOVA with the Bonferroni–Dunn post hoc test. p < 0.05 was considered statistically significant.