High glucose induces CCL20 in proximal tubular cells via activation of the KCa3.1 channel

Abstract

Background: Inflammation plays a key role in the development and progression of diabetic nephropathy (DN). KCa3.1, a calcium activated potassium channel protein, is associated with vascular inflammation, atherogenesis, and proliferation of endothelial cells, macrophages, and fibroblasts. We have previously demonstrated that the KCa3.1 channel is activated by TGF-β1 and blockade of KCa3.1 ameliorates renal fibrotic responses in DN through inhibition of the TGF-β1 pathway. The present study aimed to identify the role of KCa3.1 in the inflammatory responses inherent in DN.

Methods: Human proximal tubular cells (HK2 cells) were exposed to high glucose (HG) in the presence or absence of the KCa3.1 inhibitor TRAM34 for 6 days. The proinflammatory cytokine chemokine (C-C motif) ligand 20 (CCL20) expression was examined by real-time PCR and enzyme-linked immunosorbent assay (ELISA). The activity of nuclear factor-κB (NF-κB) was measured by nuclear extraction and electrophoretic mobility shift assay (EMSA). In vivo, the expression of CCL20, the activity of NF-κB and macrophage infiltration (CD68 positive cells) were examined by real-time PCR and/or immunohistochemistry staining in kidneys from diabetic or KCa3.1-/- mice, and in eNOS-/- diabetic mice treated with the KCa3.1 channel inhibitor TRAM34.

Results: In vitro data showed that TRAM34 inhibited CCL20 expression and NF-κB activation induced by HG in HK2 cells. Both mRNA and protein levels of CCL20 significantly decreased in kidneys of diabetic KCa3.1-/- mice compared to diabetic wild type mice. Similarly, TRAM34 reduced CCL20 expression and NF-κB activation in diabetic eNOS-/- mice compared to diabetic controls. Blocking the KCa3.1 channel in both animal models led to a reduction in phosphorylated NF-κB.

Conclusions: Overexpression of CCL20 in human proximal tubular cells is inhibited by blockade of KCa3.1 under diabetic conditions through inhibition of the NF-κB pathway.

Figure 1. KCa3.1 blocker TRAM34 inhibited CCL20 synthesis and NF-κB activation induced by HG in human renal proximal tubular cells.

HK2 cells were treated with either control, HG (25 mM) or high glucose (25 mM) combined with DMSO (vehicle control) or TRAM34 (4 µM) for 6 days. (A) RT-PCR results showed that TRAM34 inhibited HG-induced CCL20 mRNA expression. (B) ELISA result showed that TRAM34 inhibited the induction of CCL20 in HG treated HK2 cells. Nuclear extracts of those cells were used to test NF-κB binding activity by EMSA. (C) EMSA results showed that TRAM34 inhibited HG-induced NF-κB activation in HK2 cells. (D) The quantitive result for NF-κB binding acivity. Results are presented as means ± SEM. *P<0.05 and **P<0.01, n = 3.

Figure 2. Upregulation of CCL20 expression in HG-stimulated HK2 cells was inhibited by NF-κB inhibition.

HK2 cells were treated with either control, HG (25 mM) or HG (25 mM) combined with NF-κB inhibitor PDTC (25 µM) for 6 days. RT-PCR results showed that PDTC inhibited HG-induced CCL20 mRNA expression in HK2 cells. Results are presented as means ± SEM. ** P<0.01, n = 3.

Figure 3. Blockade of KCa3.1 suppressed CCL20 expression in diabetic mice.

Quantitative RT-PCR showed increased mRNA levels of CCL20 (A), in the kidneys of diabetic KCa3.1+/+ mice but reversed in diabetic KCa3.1-/- mice (n = 8). Quantitative RT-PCR showed increased mRNA expression of CCL20 (B) in the kidneys of diabetic eNOS-/- mice compared to control mice but reduced with TRAM34 treatment (DM+TRAM34) (n = 6). Representative images (C) show immunohistochemical staining of CCL20 in the renal cortex from control KCa3.1+/+ mice, diabetic KCa3.1+/+ mice and diabetic Kca3.1-/- mice (n = 8). Representative images (E) show immunohistochemical staining of CCL20 in the renal cortex from control mice, diabetic eNOS-/- mice and diabetic eNOS-/- mice treated with TRAM34. (D, F) The quantitation of CCL20 expression in mice kidneys. Results are presented as mean ± SEM. *P<0.05 and **P<0.01. Original magnification: ×200.

Figure 4. Blockade of KCa3.1 prevented diabetes-induced macrophage infiltration into Kidney.

(A) Quantitative RT-PCR showed increased mRNA levels of CD68 in the kidneys of diabetic KCa3.1+/+ mice but reversed in diabetic KCa3.1-/- mice (n = 8). (B) Quantitative RT-PCR showed increased mRNA levels of CD68 in the kidneys of diabetic eNOS-/- mice but reversed in diabetic eNOS-/- mice treated with TRAM34 (n = 6). (C) Immunohistochemical analysis showed increased CD68 in diabetic KCa3.1+/+ kidneys compared to control mice and reversed expression of CD68 in diabetic KCa3.1-/- kidneys (n = 8). (E) Immunohistochemical analysis showed increased CD68 in diabetic eNOS-/- kidneys compared to control mice and reversed expression of CD68 in diabetic eNOS-/- kidneys treated with TRAM34 (n = 6). (D, F) The quantitation of CD68 expression in mice kidneys. Results are presented as mean ± SEM. *P<0.05 and **P<0.01. Original magnification: ×400.

Figure 5. Blockade of KCa3.1 reversed diabetes-induced NF-κB activation in diabetic kidneys.

(A) Immunohistochemical analysis showed increased phosphorylated NF-κB-P65 in diabetic KCa3.1+/+ kidneys compared to control mice and reversed activation of NF-κB in diabetic KCa3.1-/- kidneys (n = 8). (C) Immunohistochemical analysis showed increased NF-κB activation in diabetic eNOS-/- kidneys compared to control mice and reversed activation of NF-κB in diabetic eNOS-/- kidneys treated with TRAM34 (n = 6). (B, D) The quantitation of phosphorylated NF-κB expression in mice kidneys. Results are presented as mean ± SEM. *P<0.05 and **P<0.01. Original magnification: ×200.