Lithium ameliorates tubule-interstitial injury through activation of the mTORC2/protein kinase B pathway

Abstract

Tubule-interstitial injury (TII) is a critical step in the progression of renal disease. It has been proposed that changes in proximal tubule (PT) albumin endocytosis plays an important role in the development of TII. Some reports have shown protective effects of lithium on kidney injury animal models that was correlated to proteinuria. We tested the hypothesis that lithium treatment ameliorates the development of TII due to changes in albumin endocytosis. Two experimental models were used: (1) TII induced by albumin overload in an animal model; (2) LLC-PK1 cells, a PT cell line. Lithium treatment ameliorates TII induced by albumin overload measured by (1) proteinuria; (2) collagen deposition; (3) area of tubule-interstitial space, and (4) macrophage infiltration. Lithium treatment increased mTORC2 activity leading to the phosphorylation of protein kinase B (PKB) at Ser473 and its activation. This mechanism enhanced albumin endocytosis in PT cells, which decreased the proteinuria observed in TII induced by albumin overload. This effect did not involve changes in the expression of megalin, a PT albumin receptor. In addition, activation of this pathway decreased apoptosis in LLC-PK1 cells, a PT cell line, induced by higher albumin concentration, similar to that found in pathophysiologic conditions. Our results indicate that the protective role of lithium treatment on TII induced by albumin overload involves an increase in PT albumin endocytosis due to activation of the mTORC2/PKB pathway. These results open new possibilities in understanding the effects of lithium on the progression of renal disease.

Fig 1. Lithium treatment reduces proteinuria induced by albumin overload due to the increase in albumin reabsorption.

Male BALB/C mice were separated into different experimental groups as described in the Materials and methods section. CONT, control group; BSA, group that received intraperitoneal injections 10 g/kg/day BSA for 7 days; BSA+LIT, group that simultaneously received BSA injection and 300 mg/kg/day lithium carbonate via gavage; LIT, lithium-treated group. A) Proteinuria (n = 7). B) Ratio of urinary protein and urinary creatinine (UPCr; n = 7). C) Urinary albumin content was assessed by immunoblotting. Urinary volume was adjusted for 10 μg of urinary creatinine. D) In vivo albumin reabsorption (n = 4). Mice received a single intravenous dose of 5 μg/g BSA-FITC used as a tracer. After 15 min, the animals were perfused with saline and the BSA-specific fluorescence intensity in the renal cortex was determined. The results are expressed as means ± SE. *P < 0.05 versus CONT group; #P < 0.05 versus BSA group.

Fig 2. Lithium treatment did not change the inhibition of megalin expression during albumin overload.

Mice were treated as described in Fig 1 (n = 4 per group). A) Representative megalin staining in the cortical area (bars represent 100 μm). B) Quantitative analyses. The results are expressed as means ± SE. *P < 0.05 versus CONT group.

Fig 3. Lithium blockade of the inhibitory effect of high albumin concentration on albumin endocytosis in PT cells measured by cell-associated fluorescence.

LLC-PK1 cells were grown in 24-well plates until 95% confluence was reached. Then, the cells were incubated with albumin and/or lithium under different experimental conditions. Albumin endocytosis measured by cell-associated fluorescence using albumin-FITC. A) Dose-response effect of lithium on albumin endocytosis (n = 9). The lithium concentrations used were 0.1, 1, 5,10, and 20 mM. B) Time-course effect of lithium on albumin endocytosis (n = 4). Times of incubation were 0, 2, 4, 6, and 12 h. C) Lithium treatment reversed the inhibitory effect of high albumin concentration on albumin endocytosis (n = 7). The cells were pre-incubated with 20 mg/mL albumin in the absence or presence of increasing lithium concentrations (5, 10, and 20 mM). The results are expressed as means ± SE. A) *P < 0.05 versus control (in the absence of lithium), #P < 0.05 versus 1 mM lithium; &P < 0.05 versus 10 mM lithium. B) *P < 0.05 versus control (in the absence of lithium). C) *P < 0.05 versus control (in the absence of albumin), #P < 0.05 versus albumin.

Fig 4. Lithium blockade of the inhibitory effect of high albumin concentration on albumin endocytosis in PT cells measured by confocal fluorescence microscopy.

The cells were pre-incubated with 20 mg/mL albumin and/or 20 mM lithium. A) Representative image. Green (BSA-FITC) indicates endocytic albumin, and blue (DAPI) indicates the cell nucleus. Scale bar represents 20 μm. DIC, differential interference contrast. B) Quantitative analysis (n = 4). The results are expressed as means ± SE. The control was taken as 100%. *P < 0.05 versus control (in the absence of lithium), #P < 0.05 versus albumin.

Fig 5. Lithium treatment increases albumin endocytosis due to activation of the mTORC2/PKB/GSK3β pathway.

LLC-PK1 cells were grown in 24-well plates until 95% confluence was reached. Then, the cells were incubated with different compounds as indicated. A) The stimulatory effect of lithium on albumin endocytosis depends on the PI3K/PKB pathway (n = 6). The cells were pre-incubated with 10−7 M wortmannin (Wort, PI3K inhibitor) or 10−6 M MK-2206 (PKB inhibitor) for 30 min before the addition of 20 mM lithium. B) Dose-response of lithium on PKB (Ser473) and mTORC2 (Ser2481) phosphorylation (n = 3). C) Lithium treatment modulates the inhibitory effect of higher albumin concentration on PKB (Ser473), GSK3β (Ser9) or mTORC2 (Ser2481) phosphorylation in LLC-PK1 cells (n = 4). D) Lithium treatment modulates the inhibitory effect of higher albumin concentration on PKB (Ser473), GSK3β (Ser9) or mTORC2 (Ser2481) phosphorylation in the TII animal model (n = 6). Representative immunoblotting images are shown. Densitometric quantification was obtained as a ratio of phosphorylated and total protein. Relative expression is represented as a percentage of control. The results are expressed as means ± SE. A) *P < 0.05 versus control (in the absence of lithium), #P < 0.05 versus 20 mM lithium. B) *P < 0.05 versus control (in the absence of lithium). C,D) *P < 0.05 versus control (in the absence of albumin and lithium) or CONT group, #P < 0.05 versus albumin or the BSA group.

Fig 6. Lithium treatment protects PT cells against apoptosis induced by high albumin concentrations.

LLC-PK1 cells were grown in 24-well plates until reach 95% confluence was reached. Then, the cells were incubated with 20 mg/mL albumin and/or 20 mM lithium. A) Representative dot plot of the experiments (n = 8). B) Quantitative analysis showing the frequency of annexin-V-positive cells. *P < 0.05 versus control (in the absence of albumin), #P < 0.05 versus 20 mM albumin.

Fig 7. Lithium treatment ameliorates tubule-interstitial injury induced by albumin overload.

Mice were treated as described in Fig 1. A) Representative images of Bowman´s capsule space. B) Quantitative analysis related to A. C) Representative images of tubule-interstitial space. D) Quantitative analysis related to C (n = 5). E) Representative Picrosirius staining (red color) in the cortical area. F) Quantitative analysis related to C (n = 8). Scale bars represent 100 μm. The results are expressed as means ± SE. *P < 0.05 versus the CONT group, #P < 0.05 versus the BSA group.

Fig 8. Lithium treatment induces change in M1/M2 phenotype macrophages.

Mice were treated as described in Fig 1. A) Representative areas of macrophage infiltration by F4/80 staining in the cortical area. Arrows denote positive brown staining. Scale bars represent 100 μm. B) Quantitative analysis related to E (n = 4). C) iNOS expression (n = 8). D) Arginase-1 expression (n = 8). Both iNOS and arginase-1 bands were detected using specific antibodies. The OD related to iNOS or arginase-1 bands was normalized to the OD related to the β-actin band. The results are expressed as means ± SE. *P < 0.05 versus the CONT group.