Decreased Expression of Connexin 43 Blunts the Progression of Experimental GN

Abstract

GN refers to a variety of renal pathologies that often progress to ESRD, but the molecular mechanisms underlying this progression remain incompletely characterized. Here, we determined whether dysregulated expression of the gap junction protein connexin 43, which has been observed in the progression of renal disease, contributes to GN progression. Immunostaining revealed de novo expression of connexin 43 in damaged glomeruli in patients with glomerular diseases as well as in mice after induction of experimental GN. Notably, 2 weeks after the induction of GN with nephrotoxic serum, mice with a heterozygous deletion of the connexin 43 gene (connexin 43+/-) had proteinuria, BUN, and serum creatinine levels significantly lower than those of wild-type animals. Additionally, the connexin 43+/- mice showed less crescent formation, tubular dilation, monocyte infiltration, and interstitial renal fibrosis. Treatment of cultured podocytes with connexin 43-specific blocking peptides attenuated TGF-β-induced cytoskeletal and morphologic changes and apoptosis as did treatment with the purinergic blocker suramin. Finally, therapeutic treatment of GN mice with connexin 43-specific antisense oligodeoxynucleotide improved functional and structural renal parameters. These findings suggest that crosstalk between connexin 43 and purinergic signaling contributes to podocyte damage in GN. Given that this protein is highly induced in individuals with glomerular diseases, connexin 43 may be a novel target for therapeutic treatment of GN.

Keywords: connexins; glomerular disease; podocyte; renal fibrosis.

Graphical abstract

Figure 1.

Cx43 is induced in human glomerulopathies and experimental GN. In control human biopsies, faint staining for Cx43 can be observed in tubules, whereas basal expression also occurs in the glomerular endothelium of control biopsies. (A) In C3 and IgA glomerulopathies, Cx43 is highly induced within the injured glomerulus. (B) Cx43 is highly upregulated 8 and 15 days post-NTS administration in mice and mainly localized within the injured glomerulus. (C) This upregulation was confirmed by Western blot. (D) ChIP assays showed that Cx43 upregulation is triggered by p-STAT1 and p-cJun binding on Cx43 promoter. p-STAT1 was enriched in three positions on the Cx43 promoter, whereas p-cJUN was enriched in two. Different positions are indicated by P and sequential numbers. Positions and sequences are given in Table 2. *P<0.05; **P<0.01.

Figure 2.

Cx43+/− mice are protected against NTS-induced GN. (A) Survival curve showed reduced mortality rates in Cx43+/− mice. Renal function was evaluated by proteinuria expressed as (B) grams of protein per 1 mmol creatinine, (C) BUN, and (D) plasma creatinine measurements. All three parameters reveal a slower progress of the disease in Cx43+/− animals. (E) Mason Trichrome staining shows that renal structure was preserved in Cx43+/− versus Cx43+/+, because they developed (F) less tubular dilation and (G) fewer glomerular crescents. Crescents are expressed as the percentage of glomeruli presenting cellular crescents. *P<0.05; **P<0.01; ***P<0.001.

Figure 3.

Cx43+/− mice develop less renal inflammation and fibrosis. (A and B) Immunohistochemistry for F4/80 and (C) quantitative PCR for VCAM-1 at day 15 indicate that Cx43+/− mice develop less inflammation compared with wild-type mice. At the same time point, (D) quantitative PCR for collagen I and (E) quantification of Sirius red staining show restriction of the fibrotic response in Cx43+/−mice. Quantitative PCR data are expressed in graphs as arbitrary units (a.u.) that represent the ratio of the target gene to internal control gene (HPRT). *P<0.05; **P<0.01; ***P<0.001.

Figure 4.

Cx43 downregulation preserves glomerular structure. Electron microscopy images from glomeruli of Cx43+/+ and Cx43+/− NTS-treated mice at (A) low and (B) high magnification. PBS-treated mice show intact glomerular basement membrane and podocytes (P) with preserved foot processes (white arrows). (B, center panel) Cx43+/+ NTS mice showed thickening of glomerular basement membrane (white arrows) and disorganization of fenestrated endothelium (black arrow). (B, right panel) In Cx43+/− NTS mice, fenestrated endothelium (black arrow) and foot processes (white arrow) were substantially preserved. Scale bars, 5 μm in A; 2 μm in B.

Figure 5.

Cx43 is expressed by injured podocytes. Double immunofluorescence for Cx43 (green) and the podocyte markers nephrin and nestin (red) in (A) control and (B) damaged podocytes indicate localization of Cx43 in damaged podocytes after 15 days of NTS administration. (C) At the same time point, Cx43+/− mice show preserved expression of nephrin. (D) Immunofluorescence for WT-1 indicates increased number of podocytes per glomerulus in Cx43+/− versus Cx43+/+ mice. Results are expressed as number positive for WT-1 staining nuclei per glomerulus. *P<0.05; **P<0.01; ***P<0.001.

Figure 6.

Cx43 blockade preserves podocyte phenotype in vitro. (A and B) TGF-β (5 ng/ml) increases Cx43 expression in the E.11 cell line at 24- and 48-hour time points. Quantitative PCR for (C) fibrotic, (D) podocyte, and (E and F) migratory markers after treatment with TGF-β in the presence or absence of GAP26. GAP26 alone had no effect on the mRNA levels of the studied genes. Western blot for the phosphorylated forms of kinases involved in (G and H) migration and (G and I) cytoskeletal rearrangement. GAP26 alone had no effect on the phosphorylation of the kinases studied. Quantitative PCR data are expressed in graphs as arbitrary units (a.u.) that represent the ratio of the target gene to internal control gene (HPRT). *P<0.05; **P<0.01; ***P<0.001; ^#P=0.05.

Figure 7.

Cx43 and purinergic signaling blockade protects podocytes in vitro. (A) TUNEL staining indicates that (B) TGF-β–induced apoptosis is ameliorated by GAP26. (C) Measurements of ATPi levels further confirm the antiapoptotic effect of Cx43 blockade. Results are expressed in relative luminosity units. (D) TGF-β induces activation of P2Y2 receptor mRNA. Blockade of purinergic signaling with suramin (E) limits upregulation of collagen I, integrin-β3, and synaptopodin mRNAs and (F) decreases phosphorylation of pak1/2. Quantitative PCR data are expressed in graphs as arbitrary units (a.u.) that represents the ratio of the target gene to internal control gene (HPRT). *P<0.05; **P<0.01; ***P<0.001.

Figure 8.

AS ODN against Cx43 preserve renal function after NTS-GN. (A and B) Administration of AS ODN decreases Cx43 expression. The NTS-induced increase of (C) body weight, (D) proteinuria, (E) creatininemia, and (F) BUN is attenuated in wild-type mice treated with Cx43 AS ODN. Body weight increase is presented as weight (grams) gained since the beginning of the protocol. *P<0.05; **P<0.01; ***P<0.001.

Figure 9.

AS ODN against Cx43 preserve renal structure after NTS-GN. (A) Masson Trichrome reveals (B) increased glomerular crescents and fibrin deposits that are significant reduced in AS-treated mice. (C) Quantitative PCR for nephrin indicates preserved glomerular structure after AS treatment. Renal fibrosis, assessed with quantitative PCR for (D) collagen I and (E) Sirius red staining, is diminished after treatment. Quantitative PCR data are expressed in graphs as arbitrary units (a.u.) that represents the ratio of the target gene to internal control gene (HPRT). *P<0.05; **P<0.01; ***P<0.001; ^#P=0.05.

Figure 10.

Proposed model of the Cx43 deleterious effects on podocytes during the progression of GN. Our study shows that de novo expression of Cx43 in podocytes exacerbates ATP release, which in turn, mediates via P2R signaling cell damage. Podocyte damage can be illustrated by either foot process effacement or cell death, thus contributing to the progression of glomerular disease.