mTOR Inhibition limits LPS induced acute kidney injury and ameliorates hallmarks of cellular senescence

Abstract

Sepsis-induced acute kidney injury (AKI) can lead to chronic renal dysfunction with accelerated renal aging. Activation of the mammalian target of rapamycin (mTOR) is implicated in the initiation and progression of renal injury. This study investigates the effectiveness of the mTOR inhibitor, rapamycin, in mitigating kidney injury and explores the underlying mechanisms. AKI was induced by intraperitoneal administration of a solution containing 10 mg/kg of lipopolysaccharide (LPS) in a mouse model. Two groups of endotoxemic mice received pre- and post- treatment with rapamycin. Whole-genome DNA methylation analysis was performed on renal proximal tubular epithelial cells (RPTEC). In the LPS-induced AKI mouse model, rapamycin treatment significantly reduced creatinine levels, preserved renal parenchyma, and counteracted the endothelial-to-mesenchymal transition (EndMT) by inhibiting the ERK pathway. Whole-genome DNA methylation analysis revealed that LPS induced aberrant methylation, particularly in genes associated with premature aging, including ectonucleoside triphosphate diphosphohydrolase 1 (ENTPD1/CD39) and wolframin ER transmembrane glycoprotein (WFS1). Accordingly, endotoxemic mice exhibited decreased CD39 expression and klotho down-regulation, both of which were reversed by rapamycin, suggesting an anti-aging effect in AKI. mTOR inhibition may represent a promising strategy to prevent accelerated renal aging in LPS-induced AKI and potentially slow the progression of chronic kidney disease.

Fig. 1

Recovery of renal damage in endotoxemic animals. (A) PAS staining showed significant proximal tubular epithelial cells damage (zoomed images), marked fibrin deposition, increased tubulointerstitial space, capillary rarefaction in numerous glomeruli, Bowman’s capsule expansion (zoomed images), and interstitial inflammatory infiltrates (zoomed images) at 24 h and 48 h after LPS infusion (LPS 24 h and LPS 48 h) compared to the control group (Sham group). LPS 48 h infusion induced more severe tubular and glomerular damage. Renal biopsies from endotoxemic animals receiving either rapamycin pre-treatment (RAPA/LPS) or post-treatment (LPS/RAPA) showed reduced inflammatory infiltrate and preservation of tubular and glomerular compartments (zoomed image). Rapamycin post-treatment also attenuated glomerular capillary loss, Bowman’s capsule expansion, and tubular injury, although the protective effect was less pronounced than that observed with pre-treatment (zoomed images). (B) Tubular and glomerular pathological scores were assessed as described in the Methods section and are expressed as mean ± SD from at least five mice for each group (Sham group n = 5; LPS 24 h group n = 8; LPS 48 h group n = 5; RAPA/LPS n = 8; LPS/RAPA n = 5). (C) Statistical significance was evaluated using one-way ANOVA, corrected for multiple comparisons of pairwise treatment group differences using Tukey’s method (****p < 0.0001 vs. Sham; ££££p < 0.0001 vs. LPS 24 h; ####p < 0.0001 vs. LPS 48 h).

Fig. 2

Anti-fibrotic role of rapamycin in LPS induced AKI. (A) Immunohistochemistry for collagen III showed increased collagen III deposition in endotoxemic mice (LPS 24 h and 48 h) at both interstitial and glomerular level (zoomed images). Notably, the duration of endotoxemia correlated with greater collagen III accumulation, indicative of interstitial fibrosis. Rapamycin pre-treatment (RAPA/LPS) significantly decreased collagen deposits (zoomed images). Similarly, rapamycin post-treatment (LPS/RAPA) also decreased collagen accumulation, although collagen III deposition remained detectable (zoomed images). (B) Quantitative analysis was performed as described in the Methods section. The mean intensity of strong positive (ISP) relative to the analyzed area was measured for collagen III immunohistochemistry. (Sham group n = 5; LPS 24 h group n = 8; LPS 48 h group n = 5; RAPA/LPS n = 8; LPS/RAPA n = 5). Error bars indicate standard deviation. ****p < 0.0001 vs. Sham; ££££p < 0.0001 vs. LPS 24 h; ####p < 0.0001 vs. LPS 48 h.

Fig. 3

Rapamycin prevents EC dysfunction. (A) When activated by LPS, renal CD31⁺ (red) EC acquired myofibroblast marker α-SMA (green) within renal vessels, glomeruli and interstitium (LPS 24 h and LPS 48 h). Rapamycin pre-treatment (RAPA LPS) and post-treatment (LPS RAPA) restored the EC phenotype in all renal compartments. (B) Quantitative analysis was obtained as described in the Methods section. Magnification 630x. To-Pro 3 was used to counterstain nuclei (blue). (Sham group n = 5; LPS 24 h group n = 8; LPS 48 h group n = 5; RAPA/LPS n = 8; LPS/RAPA n = 5). ****p < 0.0001 vs. Sham; ££££p < 0.0001 vs. LPS 24 h; ####p < 0.0001 vs. LPS 48 h. (C,D) Primary endothelial cells (HUVEC) were exposed to LPS (4 µg/ml) for 48 h. For rapamycin pre-treatment, rapamycin (5nM) was added in medium 1 h before the LPS exposure and then maintained for 48 h. Post-Treatment was performed by incubating rapamycin 6 h after LPS stimulation. Upon LPS stimulation, EC showed a significant reduction in specific EC markers and an increased expression of dysfunctional-fibroblast markers, as determined by flow cytometry analysis. In the presence of rapamycin, EC preserved their phenotype. (C) Results are representative of three independent experiments. (****p ≤ 0.0001 vs. Basal; ££££p ≤ 0.0001 vs. LPS); (D) Representative plot.

Fig. 4

TLR4-mTORC1 axis induces EndMT through ERK signaling. (A,B) Cultured EC were stimulated with LPS (4µg/ml) for 30’. For rapamycin pre-treatment, rapamycin (5nM) was added in medium 30’ before the LPS exposure. Post-Treatment was performed incubating rapamycin after 30’ of LPS stimulation. EC were also pretreated with AKT-inh or ERK-inh for 30’ followed by a short stimulation with LPS (30’). LPS increased the expression of phosphorylated ERK, AKT, mTORC-1 and p70S6K. Rapamycin (RAPA) pre and post-stimulation significantly decreased p-mTORC1 and p-p70S6K but did not abrogate phosphorylation of AKT and ERK. Pre-treatment of EC with ERK-inh reduced LPS-induced phosphorylation of mTORC1 and p70S6K. (A) Representative plot of flow cytometry analysis; (B) Results are representative of three independent experiments. (C,D) EC were cultured with LPS or ERK-inh for 48 h; EC were pre-treated with ERK-inh for 1 h followed by LPS for 47 h. FACS analysis showed that EC did not modify their phenotype also upon LPS stimulation. (C) Representative plot of flow cytometry analysis; (D) Results are representative of three independent experiments; (****p ≤ 0.0001 vs Basal; ££££p ≤ 0.0001 vs LPS).

Fig. 5

LPS-associated changes in DNA methylation as indicated by whole-genome bisulfite assay in RPTEC. (A) Chart showing the number and the frequency of methylated regions (tiling regions) identified in RPTEC at basal level. The left vertical axis represents the number of methylated regions per each chromosome. The right vertical axis indicates the cumulative percentage of the total number of occurrences. The red concave curve is the cumulative function indicating that the 50% of the total methylated regions in RPTEC are covered by chromosomes 1, 8, 6, 2, 7 and 5. (B) DNA methylation levels at the single CpG sites for RPTEC at basal level (blue line) and for RPTEC stimulated by LPS (orange line). LPS increased the DNA methylation in the overall genome. (C) Graph showing the mean DNA methylation levels of tiling regions, shared for chromosomes, in RPTEC (blue line) and in LPS-stimulated RPTEC (red line). The central axis shows the mean β methylation value. (D) Scatterplot of the CpG site methylation comparison, colored according to the combined ranks of a given site. Values are represented as mean differences (mean.diff) between stimulated and unstimulated RPTEC for each CpG site. (Combined rank: difference in mean methylation levels of stimulated and non-stimulated RPTEC, the quotient in mean methylation and the t test are ranked for all regions. This value aggregates them using the maximum, i.e., worst rank of a site among the three measures.)

Fig. 6

CD39 and WFS1 gene expression is regulated by the DNA methylation. (A) Methylation levels of CD39 and WFS1 in LPS-stimulated RPTEC or cells treated with 1 µM 5-aza-2’-deoxycytidine compared to basal condition. The percentage of PMR was calculated as described in the Materials and methods. (B) Gene expression levels of CD39 and WFS1 evaluated by qRT-PCR in the LPS-stimulated RPTEC or treated with 1 µM 5-aza, compared to untreated RPTEC. Expression levels were significantly different in LPS-stimulated RPTEC compared to untreated cells. Gene expression levels were normalized to GAPDH as a housekeeping gene. Results are presented as mean ± standard deviation (SD), n = 3. **p < 0.001, *p < 0.01.

Fig. 7

SA-β Gal staining in RPTEC and validation of target genes. (A) SA-β-Gal activity in early passage RPTEC exposed to LPS (4 µg/ml) for 48 h. For the pre-treatment group, rapamycin (5 nM) was added to the medium 1 h before LPS exposure and maintained for 48 h. Post-treatment was performed by incubating rapamycin 6 h after LPS stimulation. The inhibition of senescence was more pronounced in the pre-treatment group compared to post-treatment. SA-β-gal + cells were observed following LPS exposure with senescent RTEC appearing enlarged and morphologically distinct from the normal cells at the same passage, showing the formation of larger and polynucleated cells. Untreated cells are referred to as basal, whike cells exposed to H₂O₂ were used as a positive control for senescence. Representative images were acquired using phase contrast microscopy. Magnification 40X; scale bar: 400 μm (B) Quantification of SA-β-Gal + cells cultures. The ratio of cells positive for SA-β-gal activity was calculated by examining five non-overlapping fields per condition (6-well plate). Results are presented as mean ± SD from three independent experiments (**p < 0.001). (C–F) Real-time PCR analysis of four genes (KLOTHO, p21, CD39 and WFS1) differentially expressed in RPTECs under basal condition, after 48 h of stimulation, pre- and post-treatment with rapamycin, as well as rapamycin alone. Data are expressed as mean ± standard deviation (SD), n = 3. (KLOTHO *p < 0.05; p21, CD39 and WFS1 **p < 0.001and *p < 0.01).

Fig. 8

Rapamycin treatment restores CD39 expression in LPS-induced AKI. (A) Immunohistochemistry for CD39 revealed a significant decrease of tubular CD39 expression in LPS mice (LPS 24 h), (zoomed image). At 48h, LPS exposure resulted in an even more pronounced reduction of tubular (zoomed image) and glomerular CD39 expression (zoomed image) compared to the 24h endotoxemic group. Rapamycin treatment restored CD39 expression, with pre-treatment (RAPA/LPS) being more effective than post-treatment (LPS/RAPA) in rescuing tubular CD39 levels (zoomed image). (B) Quantitative analysis was obtained as described in the Methods section. The mean intensity of strong positive (ISP) related to the area analyzed was measured for CD39 immunohistochemistry. Error bars indicate standard deviation. ****p < 0.0001 vs. Sham; ££££p < 0.0001 vs. LPS 24 h. (Sham group n = 5; LPS 24 h group n = 8; LPS 48 h group n = 5; RAPA/LPS n = 8; LPS/RAPA n = 5) (C,D) WB analysis revealed a significant decrease of CD39 expression after 24 h of LPS infusion, compared to sham level. A 48 h LPS infusion showed a marked reduction of CD39 protein. Rapamycin pre-treatment protected the kidney in course of endotoxemic AKI, restoring CD39 expression. However, post-treatment did not significantly reactivate CD39 expression. β-actin protein expression was used for normalization. (C) This gel is representative of three animals for group with similar results. (D) Data are expressed as mean ± standard deviation (SD). (Sham group n = 5; LPS 24 h group n = 8; LPS 48 h group n = 5; RAPA/LPS n = 8; LPS/RAPA n = 5). Statistically significant differences were assessed one-way ANOVA, corrected for multiple comparison of pairwise treatment group differences using Tukey’s method (****p < 0.0001 vs. Sham; ££p < 0.01 vs. LPS 24 h; ###p < 0.001 vs. LPS 48 h).

Fig. 9

Characterization of tubular dysfunction and rapamycin effects. (A) Immunohistochemistry for klotho revealed a significant decrease of tubular klotho expression in LPS mice and its restoration by rapamycin treatment. After 48 h from LPS infusion, tubular klotho expression was dramatically decreased. Rapamycin pre-treatment (RAPA/LPS) more than post- (LPS/RAPA), restored tubular klotho expression. (B) The quantitative analysis was obtained as described in the Methods section. Mean number of strong positive (NSP) related to the area analyzed was measured for klotho immunohistochemistry. Error bars indicate standard deviation. (Sham group n = 5 LPS 24 h group n = 8 LPS 48 h group n = 5 RAPA/LPS n = 8 LPS/RAPA n = 5). ****p < 0.0001 vs. Sham; ££££p < 0.0001 vs. LPS 24 h; ####p < 0.0001 vs. LPS 48 h. (C,D) WB analysis revealed a significant decrease of klotho expression after 24 h of LPS infusion, compared to sham level. After 48 h from LPS infusion, there was a marked reduction of klotho expression. The rapamycin treatment, particularly in pre-treated group, preserved renal parenchyma, restoring klotho expression. β-actin protein expression was used for normalization. (C) This gel is representative of three animals for group with similar results. (D) Data are expressed as mean ± standard deviation (SD) and statistical differences were assessed one-way ANOVA, corrected for multiple comparison of pairwise treatment group differences using Tukey’s method. (Sham group n = 5 LPS 24 h group n = 8 LPS 48 h group n = 5 RAPA/LPS n = 8 LPS/RAPA n = 5) ****p < 0.0001 vs. Sham; ££p < 0.0001 vs. LPS 24 h; ###p < 0.0001 vs. LPS 48 h.