MCP-1 directly induces renal tubulointerstitial fibrosis independently of monocytes/macrophages infiltration

Abstract

Background: Previous studies have demonstrated the importance of monocyte chemoattractant protein-1 (MCP-1) and its receptor, C-C chemokine receptor 2 (CCR2), in the pathogenesis of various inflammatory and fibrotic diseases via the recruitment and activation of monocytes/macrophages. Recently, however, accumulating in vitro evidence has indicated that MCP-1 per se may act directly on renal cells via CCR2. Therefore, the results of a number of former studies showing the impacts of MCP-1/CCR2 blockade on renal injury may be partly attributed to a direct inhibitory effect of MCP-1 on renal cells, but it has never been clarified in vivo to date. Since all previous studies to explore the effect of MCP-1 on diverse renal diseases were performed using MCP-1-neutralizing antibody, CCR2 antagonist, or MCP-1 or CCR2 knockout technique, which accompanied a significant reduction in monocytes/macrophages infiltration within the kidney, it was hard to define whether the beneficial influence of MCP-1/CCR2 inhibition on renal injury was attributed to the direct inhibitory effect of MCP-1 on renal cells or was just a consequence of a significant decrease in inflammatory cells infiltration.Purpose: This study was undertaken to investigate the direct impact of the MCP-1/CCR2 on renal fibrosis in vivo by using monocytes/macrophages-depleted mice. In addition, the direct effect of MCP-1 on extracellular matrix (ECM) synthesis along with the role of CCR2 was also examined in cultured tubular epithelial cells (NRK-52E).Methods: In vitro, NRK-52E cells were incubated in DMEM media containing 5.6 mM glucose (Control, Con) or recombinant MCP-1 (10 ng/ml) with or without RS102895 (2 and 10 M), a specific chemical inhibitor of CCR2, and CCR2 siRNA (final concentrations: 10, 25, and 50 nM). After 72 hours, cells were harvested. In vivo, Thirty-six male C57BL/6J mice, weighting 25-30 g, were divided into six groups: Group 1, control mice treated with intravenous PBS only (N = 6) (Control, Con); Group 2, mice injected with liposome vehicle (LV) and lenti-empty virus intravenously (N = 6); Group 3, mice with liposome-clodronate (LC) and lenti-empty virus (N = 6); Group 4, mice with LV and lenti-MCP-1 virus (N = 6); Group 5, mice with LC and lenti-MCP-1 virus (N = 6); and Group 6, mice with LC, lenti-MCP-1 virus, and RS102895 (N = 6). LV and LC in a volume of 200 µl PBS were injected intravenously every 5 days for 4 weeks, and lenti-empty and lenti-MCP-1 virus were injected intravenously at a dose of 1.5 x 107 transfection units every 5 days, three times with LC or LV. RS102895 was delivered at a dose of 3 mg/kg/day for 4 weeks via subcutaneously-implanted osmotic mini-pumps. Either diphtheria toxin (DT) diluted in PBS (10 ng/g of body weight) (N = 24) or PBS (N = 6) was given intraperitoneally to Cd11b-diphtheria toxin receptor (DTR) mice on the day before lenti-MCP-1 virus injection and every 3 days. Lenti-MCP-1 was also injected at 3 days after the first administration. Six from DTR mice injected with lenti-MCP-1 virus and DT were treated with RS102895 (3 mg/kg/day) via osmotic mini-pumps. DTR mice were sacrificed at 10 days after the first lenti-MCP-1 virus injection. The protein expression of fibronectin, type I collagen and CCR2 in cultured NRK-52E cells, and the whole kidney were evaluated by Western blot, and the mRNA expression of fibronectin and type I collagen was assessed by real-time PCR. MCP-1 concentrations in serum and the whole kidney were determined by ELISA. Peripheral blood cell counts were conducted and immunohistochemistry (IHC) for fibronectin, type I collagen and F4/80 and Masson’s trichrome staining were examined.Results: Compared to Con cells, the protein expression of fibronectin and type I collagen were significantly increased in NRK-52E cells exposed to MCP-1, and these increases were significantly abrogated by RS102895 or CCR2 siRNA (P < 0.05). Two days after administration of LC and DT, there was a significant reduction in peripheral blood monocyte counts in C57BL/6J mice and DTR mice, respectively, compared to Con mice (P < 0.001). While administration of lenti-MCP-1 virus alone induced a significant increase in the number of infiltrated monocytes/macrophages in the kidney (P < 0.001), a concomitant treatment with LC or DT significantly attenuated the increase in monocytes/macrophages infiltration in the kidney (P < 0.001) compared to Con mice. The significant increases in fibronectin and type I collagen protein expression (P < 0.01), assessed by Western blot, in LC- or DT-treated mice infected with lenti-MCP-1 virus were significantly ameliorated by CCR2 inhibition using osmotic mini-pumps containing RS102895 (P < 0.05). The significant increases in IHC staining scores for fibronectin and type I collagen within the tubulointerstitium in monocytes/macrophages-depleted mice infected with lenti-MCP-1 virus (P < 0.001) were significantly attenuated by RS102895 treatment (P < 0.001).Conclusions: These findings suggest that the MCP-1/CCR2 system is directly involved in MCP-1-induced renal fibrosis and blockade of the MCP-1/CCR2 system can be a promising approach to treat various kidney diseases such as diabetic nephropathy, of which MCP-1-induced renal fibrosis is involved in the pathogenesis.