Smad3 Mediates Renal Fibrosis via GPX4-Dependent Ferroptosis

Abstract

TGF-β/Smad3 signaling is a key pathway leading to the cell death and renal fibrosis. Here we report a new mechanism through which Smad3 mediates renal fibrosis by downregulating the glutathione peroxidase 4 (GPX4), a central inhibitor for ferroptosis. In patients with chronic kidney disease (CKD) and a mouse model of unilateral ureteral obstruction (UUO), progressive renal fibrosis was associated with the overactive Smad3 signaling and the development of ferroptosis identified by decreased GPX4 while increasing two ferroptosis biomarkers including the Transferrin receptor 1 (TFR1) and 4-Hydroxynonenal (4-HNE). Mechanistically, we uncovered that Smad3 could bind directly to GPX4 to repress its transcription while increasing TFR1 and 4-HNE expression, which was abolished when this binding site was mutated. This novel finding was functionally confirmed in the UUO mice and mouse embryonic fibroblasts (MEFs) in which deletion of Smad3 protected against UUO and transforming growth factor-β1 (TGF-β1)-induced loss of GPX4, upregulation of TFR1 and 4-HNE, and progressive renal fibrosis in vivo and in vitro. Importantly, we also found that GPX4 was a downstream target gene of Smad3 and functioned to protect against Smad3-mediated renal fibrosis as silencing GPX4 restored UUO-induced severe renal fibrosis in Smad3 KO mice and in TGF-β1-stimulated Smad3 KO MEFs and SIS3-treated HK-2 cells. Thus, GPX4 is protective in renal fibrosis. Smad3 mediates renal fibrosis via a mechanism associated with GPX4-dependent ferroptosis. The protective effect of GPX4 on Smad3-mediated renal pathologies suggests that targeting the Smad3/GPX4 axis may be a novel therapy for CKD.

Keywords: Ferroptosis; GPX4; Renal fibrosis; Smad3; TGF-β1..

Figure 1

Smad3 signaling is highly activated in CKD patients with progressive renal fibrosis and is associated with decreased GPX4 while increasing ferroptosis. (A) Representative photomicrographs of kidney sections (periodic acid-Schiff, Masson's trichrome, and periodic acid-silver methenamine staining) from patients with MCD, IgAN, and DKD. (B) Representative photomicrographs of immunohistochemistry staining of renal α-SMA, Col-1, and FN. (C) Representative photomicrographs of immunohistochemistry staining of renal p-Smad3, GPX4, 4-HNE and TFR1. (D) Representative transmission electron microscope (EM) photomicrographs of pathological changes in the mitochondria. (E) Correlation of GPX4 with p-Smad3 and the fibrosis marker Col-1 in the kidney of patients with CKD. (F) Quantitation of renal GPX4, p-Smad3, 4-HNE, TFR1, α-SMA, and Col-1 protein expression. Each dot presents one patient, and data were represented as mean ± SD. ***P < 0.001. Scale bar, 100 μm.

Figure 2

Smad3 is highly activated in the mouse UUO kidney with progressive renal fibrosis, which is associated with decreased renal GPX4 while increasing ferroptosis. (A) Representative photomicrographs of kidney sections (periodic acid-Schiff, Masson's trichrome, and periodic acid-silver methenamine staining) from Sham and UUO mice. (B) Representative photomicrographs of immunohistochemistry staining of renal α-SMA, Col-1, and FN. (C) Representative photomicrographs of immunohistochemistry staining of renal p-Smad3, GPX4, 4-HNE, TFR1, and representative transmission electron microscope photomicrographs. (D) Quantitation of renal p-Smad3, GPX4, 4-HNE, TFR1, α-SMA, Col-1, and FN protein expression. (E) Western blotting for p-Smad3, Smad3, GPX4, 4-HNE, TFR1, α-SMA, Col-1, and FN. (F) Quantitation of renal p-Smad3, GPX4, α-SMA, Col-1, and FN protein expression. (G) Real-time PCR for renal GPX4, α-SMA, Col-1, and FN mRNA expression. Data are presented as mean ± SD for groups of 6 mice. *P < 0.05, **P < 0.01, ***P < 0.001; Scale bar, 50 μm.

Figure 3

Bioinformatics analysis reveals that decreased renal GPX4 expression and the development of ferroptosis are associated with TGF-β/Smad3 signaling in the fibrotic kidney of CKD patients and UUO mice. (A) Three different renal fibrosis datasets (GSE66494, GSE118339, and GSE152250) were obtained from the NCBI for KEGG and GO analyses. C57BL/6 mice were used to establish a renal fibrosis model using UUO surgery. The kidneys were used for RNA sequencing (RNA-seq) 7 days later. (B) - (D) Over-representation analysis of GO Biological Processes and KEGG on significantly (adjusted P-value < 0.05) differentially expressed genes with upregulated (log2-fold change > 0.5) and downregulated (log2-fold change < -0.5) expression in kidney tissues from patients with CKD (B) or from UUO mice (C), and TGF-β1-treated tubular epithelial cells (TEC) (D). (E) Volcano plot depicting differentially expressed genes (adjusted P-value < 0.05 and 1 log2-fold change cutoff) in kidney tissues between Sham and UUO mice. (F) Heatmap and hierarchical clustering of selected differentially expressed genes involved in ferroptosis and TGF-β/Smad3 signaling. Each experiment was performed in duplicate, and kidney samples were used from two Sham or UUO mice.

Figure 4

GPX4 is a Smad3 target gene for fibrosis and is negatively regulated by TGF-β1 via a Smad3-dependent manner in HK-2 cells and MEFs. (A) A Smad3 binding site was predicted on the promoter region of the GPX4 genomic sequence by the JASPAR database. (B) Dual-luciferase reporter assay shows the transcriptional regulation of Smad3 on GPX4 expression in HEK293T cells. Note that mutation of the Smad3 binding site protects against the transcriptional repression of GPX4. (C) The ChIP assay showed the physical binding of Smad3 on the GPX4 promoter genomic sequence, which is enriched after TGF-β1 treatment in HK-2 cells. (D, E) Western blot and quantitative analysis show that treatment with TGF-β1 (5 ng/mL) induces activation of Smad3 (p-Smad3) and upregulation of Col-1 and FN, which is associated with inhibition of GPX4 expression in MEFs. (F, G) Western blot analysis detects that that deletion of Smad3 enhances GPX4 expression but inhibits TGF-β1-induced fibrosis including Col-1, and FN in Smads3 KO MEFs. Data are presented as mean ± SD for at least 3 independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5

Mice null for Smad3 are protected from UUO-induced loss of GPX4, thereby inhibiting ferroptosis and progressive renal fibrosis in a mouse model of UUO. (A) Representative photomicrographs of kidney sections from Smad3 WT and Smad3 KO UUO mice on 7 days after surgery. Sections were stained with PAS and antibodies against GPX4, 4-HNE, TFR1, α-SMA, Col-1, and FN by immunohistochemistry. (B) Quantitation of renal GPX4, 4-HNE, TFR1, α-SMA, Col-1, and FN protein expression. (C) Western blotting for expression of GPX4, α-SMA, Col-1, and FN in the UUO kidney. (D) Quantitation of renal GPX4, α-SMA, Col-1, and FN protein expression. (E) Real-time PCR for renal GPX4, α-SMA, Col-1, and FN mRNA expression. Data are presented as mean ± SD for groups of 6 mice. *P < 0.05, **P < 0.01, ***P < 0.001. Scale bar, 50 μm.

Figure 6

Kidney-specifically silencing GPX4 restores the severity of ferroptosis and renal fibrosis in the UUO kidney of Smad3 KO mice. (A, B) Representative photomicrographs and quantitative analysis for histology (PAS) and expression of GPX4, 4-HNE, TFR1 and fibrotic markers including α-SMA, Col-1, FN from Smad3 WT/KO mice at 7 days after UUO. (C, D) Western blot and quantitative analysis of renal GPX4, Col-1, and FN protein expression. (E) Real-time PCR for renal GPX4, α-SMA, Col-1, and FN mRNA expression. Data are presented as mean ± SD for groups of 6 mice. *P < 0.05, **P < 0.01, ***P < 0.001. Scale bar, 50 μm.

Figure 7

Silencing GPX4 restores TGF-β1-induced fibrosis in Smad3 KO MEFs in vitro. Western blot analysis shows that silencing GPX4 restores TGF-β1 (5ng/ml)-induced fibrosis such as Col-1, FN in Smad3 KO MEFs to the comparable levels of Smad3 WT MEFs. (A) Wester blots; (B) quantitative analysis. Data are presented as mean ± SD for 3 independent experiments. **P < 0.01, ***P < 0.001.