TLR4 promotes fibrosis but attenuates tubular damage in progressive renal injury

Abstract

Toll-like receptors (TLRs) can orchestrate an inflammatory response upon activation by pathogen-associated motifs and release of endogenous stress ligands during tissue injury. The kidney constitutively expresses most TLRs, including TLR4. The function of TLR4 during the inflammation, tubular atrophy, and fibrosis that accompany progressive renal injury is unknown. Here, we subjected wild-type (WT) and TLR4-deficient mice to unilateral ureteral obstruction and observed elevated levels of TLR4 mRNA in the kidney after obstruction. One day after unilateral ureteral obstruction, TLR4-deficient mice had fewer proliferating tubular epithelial cells and more tubular damage than WT mice; however, TLR4-deficient mice developed considerably less renal fibrosis despite decreased matrix metalloproteinase activity and without significant differences in myofibroblast accumulation. In vitro, TLR4-deficient primary tubular epithelial cells and myofibroblasts produced significantly less type I collagen mRNA after TGF-beta stimulation than WT cells. The reduced fibrosis in TLR4-deficient mice associated with an upregulation of Bambi, a negative regulator of TGF-beta signaling. In conclusion, TLR4 attenuates tubular damage but promotes renal fibrosis by modulating the susceptibility of renal cells to TGF-beta. These data suggest that TLR4 signaling may be a therapeutic target for the prevention of renal fibrosis.

Figure 1.

Enhanced TLR4 mRNA expression after UUO and higher expression of danger ligands in TLR4−/− kidneys. (A through D) Expression levels of TLR4 mRNA (A) in WT obstructed kidneys and its endogenous danger ligands HMGB1 (B), GP96 (C), and biglycan (D) in WT (□) and TLR4−/− (■) mice after UUO. TLR4 is markedly upregulated upon UUO when compared with contralateral kidneys. HMGB1 and GP96 mRNAs are enhanced in TLR4−/− mice compared with WT mice, whereas biglycan levels are similar. Data are means ± SEM of six mice per group. *P < 0.05.

Figure 2.

TLR4 attenuates tubular injury after UUO. Semiquantitative score of tubular injury in WT (□) and TLR4−/− (■) mice after UUO. (Top) TLR4−/− mice develop more severe tubular injury when compared with WT mice 1, 3, and 7 days after UUO. Data are means ± SEM of seven to eight mice per group. Contral., contralateral kidneys of mice subjected to 14 days of UUO. *P < 0.05. (Bottom) Representative microphotographs of renal tissue of WT and TLR4−/− mice 1 day after UUO. Magnification, ×200.

Figure 3.

TLR4 deficiency affects early proliferation but not apoptosis of tubular epithelial cells upon UUO. (A and B) Proliferating (A) and apoptotic (B) cells in the obstructed and contralateral kidneys from WT (□) or TLR4−/− (■) mice after UUO. The number of proliferating (BrdU+) TECs is lower in the obstructed kidneys of TLR4−/− mice compared with WT mice 1 day after UUO, whereas there are no differences at later time points. The number of apoptotic (active caspase-3+) cells is similar in obstructed kidneys of WT and TLR4−/− mice at all time points. The number of positive stained TECs is counted in 10 nonoverlapping high-power fields. Because tubular atrophy is very severe 14 days after UUO in both groups, identification of epithelial cells is difficult; therefore, the total amount of positive cells is quantified at this time point. Representative pictures of WT and TLR4−/− mice are of 1 day after UUO. Data are means ± SEM of seven to eight mice per group. Contral., contralateral kidneys of mice subjected to 14 days of UUO. *P < 0.05. Magnification, ×400.

Figure 4.

TLR4 deficiency does not reduce macrophage accumulation after UUO. (Top) Quantitative analysis of interstitial macrophages in obstructed and contralateral kidneys of WT (□) and TLR4−/− (■) mice after UUO. In WT and TLR4−/− mice, an increase of interstitial macrophages is observed in the obstructed kidneys, peaking at 14 days after UUO. The amount of macrophages is higher in the TLR4−/− mice 3 days after UUO when compared with WT mice. (Bottom) Representative pictures of WT and TLR4−/− mice are of 14 days after UUO. Data are means ± SEM of seven to eight mice per group. Contral., contralateral kidneys of mice subjected to 14 days of UUO. *P < 0.05. Magnification, ×200.

Figure 5.

TLR4 deficiency reduces renal fibrosis but does not affect myofibroblast accumulation after UUO. (A and B) Quantitative analysis of total collagen deposition (A) and myofibroblast accumulation (B) in obstructed and contralateral kidneys of WT (□) and TLR4−/− (■) mice after UUO. (A) The amount of total collagen deposition progressively increases after UUO and is significantly lower in the TLR4−/− mice 7 and 14 days after UUO. (B) No difference is found in the accumulation of α-SMA–positive myofibroblasts between WT and TLR4−/− mice. The percentage of picrosirius red– and α-SMA–positive staining is quantified by digital analysis. Representative pictures of WT and TLR4−/− mice are of 14 days after UUO (A) or 7 days after UUO (B). Data are means ± SEM of seven to eight mice per group. Contral., contralateral kidneys of mice subjected to 14 days of UUO. *P < 0.05.

Figure 6.

TLR4 deficiency reduces MMP9 but not MMP2 activity after UUO. (A and B) Quantification of latent and active MMP2 (A) and MMP9 (B) in obstructed kidneys of WT (□) and TLR4−/− (■) mice after UUO. No differences are observed in the renal levels of latent or active MMP2 between WT and TLR4−/− mice. The levels of latent and active MMP9 are significantly lower in the kidneys of TLR4−/− mice 14 days after UUO. Data are means ± SEM of seven mice per group. *P < 0.05.

Figure 7.

TLR4 deficiency enhances renal Bambi mRNA expression after UUO. Bambi mRNA levels in obstructed kidneys of WT (□) and TLR4−/− (■) mice after UUO. Bambi expression is significantly enhanced in obstructed kidneys of TLR4−/− mice compared with WT mice 3 and 14 days after UUO. Data are means ± SEM of six to seven mice per group. Contral., contralateral kidneys of mice subjected to 14 days of UUO. *P < 0.05.

Figure 8.

Primary renal TECs and myofibroblasts produce type-I collagen in a TLR4-dependent manner after TGF-β stimulation. (A and B) Relative collagen type I mRNA expression by primary TECs (A) and primary myofibroblasts (B) from WT (□) and TLR4−/− (■) mice, when subjected to stimulation with 1 ng/ml TGF-β. Control cells remain for 72 hours in culture medium. TLR4−/− primary TECs and TLR4−/− primary myofibroblasts produce significantly less collagen type I mRNA compared with WT primary TECs and with WT primary myofibroblasts after TGF-β stimulation. *P < 0.05.