A Natural Small Molecule Mitigates Kidney Fibrosis by Targeting Cdc42-mediated GSK-3β/β-catenin Signaling

Abstract

Kidney fibrosis is a common fate of chronic kidney diseases (CKDs), eventually leading to renal dysfunction. Yet, no effective treatment for this pathological process has been achieved. During the bioassay-guided chemical investigation of the medicinal plant Wikstroemia chamaedaphne, a daphne diterpenoid, daphnepedunin A (DA), is characterized as a promising anti-renal fibrotic lead. DA shows significant anti-kidney fibrosis effects in cultured renal fibroblasts and unilateral ureteral obstructed mice, being more potent than the clinical trial drug pirfenidone. Leveraging the thermal proteome profiling strategy, cell division cycle 42 (Cdc42) is identified as the direct target of DA. Mechanistically, DA targets to reduce Cdc42 activity and down-regulates its downstream phospho-protein kinase Cζ(p-PKCζ)/phospho-glycogen synthase kinase-3β (p-GSK-3β), thereby promoting β-catenin Ser33/37/Thr41 phosphorylation and ubiquitin-dependent proteolysis to block classical pro-fibrotic β-catenin signaling. These findings suggest that Cdc42 is a promising therapeutic target for kidney fibrosis, and highlight DA as a potent Cdc42 inhibitor for combating CKDs.

Keywords: Cdc42; GSK‐3β/β‐catenin; kidney fibrosis; natural small molecules.

Figure 1

Identification of DA as a potent antifibrotic compound in TGF‐β1‐activated renal fibroblasts. A) Sham or UUO surgery was performed on C57BL/6J mice. Vehicle (Veh) or crude extract of W. chamaedaphne (CE, 150 mg kg⁻¹) was intraperitoneally administered to mice daily for the consecutive 7 days after the surgery. Kidney homogenate samples were analyzed by western blotting (n = 6 per group) to quantify the protein levels of fibronectin, collagen I, and α‐SMA. B) Influence of CE on expression levels of fibrotic markers (fibronectin, collagen I, and α‐SMA). Immunoblotting analysis was carried out in TGF‐β1‐activated NRK‐49F cells treated with vehicle or indicated concentrations of CE for 48 h (n = 3 per group). C) The effects of diterpenoid 1–16 (1–16) on α‐SMA. Immunoblotting analysis was carried out in TGF‐β1‐activated NRK‐49F cells treated with vehicle or 10 µM compounds for 48 h (n = 3 per group). D) The inhibition of 1–16 on ECM (fibronectin and collagen I). The inhibition rate of each compound was calculated by the ratio of ECM levels in each group to Veh group [Compd/Veh (%)] (n = 3). E) Western blot analysis performed in TGF‐β1‐activated NRK‐49F cells treated with vehicle or indicated concentrations of DA for 48 h (n = 3 per group). F) The fibrotic protein levels were assessed in TGF‐β1‐activated NRK‐49F cells treated with vehicle or 5 µM DA at 12, 24, and 48 h (n = 3 per group). G) Immunostaining of α‐SMA (green) in NRK‐49F cells following 48 h of TGF‐β1 treatment. Scale bar = 20 µm. H, I) Immunoblot analysis of TGF‐β1‐stimulated NRK‐49F (H) and primary murine renal fibroblasts (pMRF) (I). Level of fibrotic proteins in DA or pirfenidone (PFD, 5 µM) treated cells (n = 3 per group). Control, CTL; vehicle, Veh. Data are presented as means ± SEM (A, B, C, D, E, F, H, and I). Data were analyzed using one‐way ANOVA followed by Bonferroni's multiple comparisons test. ^## p < 0.01, ^### p < 0.001, ^#### p < 0.0001 compared with CTL group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with Veh group or between groups under the line.

Figure 2

DA blocked the activation of renal fibroblasts and EMT of renal tubular cells. A, B) EdU assay was performed on TGF‐β1‐stimulated NRK‐49F cells. Representative flow cytometry (n = 3 per group) (A) and confocal images (B) showed that EdU positive ratio cells, scale bar, 20 µm. C) Microscopy images of the wound‐healing assay at different times. The solid lines showed the initial area without cells and the dashed lines indicated the margin of migrated cells. The scale bar represents 50 µm. D) Presentative microscopic images of pMRF cells migrating from the upper chamber to the lower chamber. Scale bar, 100 µm. E) Immunoblots showing the effects of DA on the fibrotic proteins in NRK‐49F cells incubated with AngII (10 µM), IL‐17 (50 ng mL⁻¹), LPS (25 ng mL⁻¹) or IL‐1β (10 ng mL⁻¹) (n = 3 per group). F) Immunoblot analysis of TGF‐β1‐stimulated NRK‐52E cells (fibronectin, E‐cadherin, vimentin, and α‐SMA) in DA or pirfenidone (PFD, 5 µM) treated group (n = 3 per group). Control, CTL; vehicle, Veh. Data are presented as means ± SEM (A, E, and F) and were analyzed using one‐way ANOVA followed by Bonferroni's multiple comparisons test. ^### p < 0.001, ^#### p < 0.0001 compared with CTL group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with Veh group or between groups under the line; ns, not significant.

Figure 3

DA protected against kidney fibrosis in UUO mice. A) The transcript levels of fibronectin (Fn1), collagen III (Col3a1), collagen I (Col1a1), and α‐SMA (Acta2) in DA or PFD‐treated mouse kidneys were examined by q‐PCR 7 days after sham or UUO surgery. β‐actin (Actb) was used as the internal reference (n = 6 per group). B) The protein levels of fibronectin, collagen I, and α‐SMA in DA or PFD‐treated mouse kidney homogenate samples analyzed by SDS‐PAGE 7 days after sham or UUO surgery (n = 6 per group). C) Representative images of picrosirius red staining and quantitative analyses of DA or PFD‐treated mouse kidneys 7 days after sham or UUO procedures, scale bar, 100 µm (n = 6 per group). D) The levels of serum AST and ALT of DA or PFD‐treated mice measured at day 7 after the surgery. E) Mice body weights in the 7 consecutive days after sham or UUO surgery (n = 6 per group). F) Kidney, liver, lung, heart, and spleen from mice treated with vehicle or 20 mg kg⁻¹ DA were collected. Tissue supernatants from their homogenates were analyzed by HPLC‐MS/MS. Vehicle, Veh; low dose, LD; high dose, HD; pirfenidone, PFD. Data are presented as means ± SEM (A, B, C, D, and E) and were analyzed using one‐way ANOVA (A, B, C, D) or two‐way ANOVA (E) followed by a Bonferroni's multiple comparisons test unless otherwise stated. ^#### p < 0.0001 compared with sham group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with UUO‐Veh group or between groups under the line; ns, not significant.

Figure 4

DA promoted the phosphorylation of β‐catenin at S33/37/45/T41 and targeted β‐catenin for ubiquitin‐proteasome degradation. A) Volcano plots depicting the differentially expressed genes between vehicle‐treated (veh) and DA‐treated groups. B) Barplot showing all the down‐regulated KEGG pathways termed signaling pathway. C) Immunoblot analysis of β‐catenin in TGF‐β1‐activated NRK‐49F treated with vehicle or DA at instructed times (n = 3 per group). D, E) The protein levels of β‐catenin in the nucleus and cytoplasm of NRK‐49F treated with vehicle or DA. Subcellular cytosolic and nuclear fractions of β‐catenin were analyzed by immunofluorescence (scale bar, 20 µm) (D) and western blot (n = 3 per group) (E). Green fluorescence signals represent β‐catenin proteins. Red signals represent the nucleus, scale bar: 20 µm. F) Immunoblot showing the protein levels of β‐catenin in cells treated with vehicle, DA, or DA+bortezomib (BTZM, 1 nM), a proteasome inhibitor (n = 3 per group). G) Effects of DA on β‐catenin ubiquitination. HA‐Ub was co‐transfected into NRK‐49F cells with Flag‐β‐catenin constructs in the presence of BTZM. Immunoprecipitation (IP) was performed by anti‐Flag beads or IgG beads and immunoblotting (IB) with anti‐HA antibodies. H, I) Protein levels of phosphorylated β‐catenin (p‐β‐catenin) at different amino acid residues including serine (S) 33, 37, 45, 552, 675, and threonine (T) 41 were assessed at 12 and 24 h in NRK‐49F cells. NRK‐49F cells were activated by TGF‐β1 and treated with vehicle or DA (n = 3 per group). J) Western blots performed on TGF‐β1‐activated NRK‐49F following the treatment of DA (5 µM), ICG001 (ICG, 5 µM), XAV939 (XAV, 5 µM) respectively (n = 3 per group). Control, CTL; vehicle, Veh; Ser, S; Thr, T. Data are presented as means ± SEM (C, E, F, H, I, and J) and were analyzed using one‐way ANOVA followed by a Bonferroni's multiple comparisons test. ^# p < 0.05, ^## p < 0.01, ^### p < 0.001, ^#### p < 0.0001 compared with CTL group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with Veh group or between groups under the line; ns, not significant.

Figure 5

DA increased p‐β‐catenin (S33/37/45/T41) and reduced total β‐catenin in UUO kidneys. A, B) Kidney homogenate samples were analyzed by SDS‐PAGE (n = 6 per group). Protein levels of total β‐catenin (A), p‐β‐catenin (S33/37/45/552/675/T41) (B) were detected in each group 7 days after sham or UUO surgery (n = 6 per group). C) Representative images and quantitative analyses of the immunohistology results of β‐catenin in kidney tissue (n = 6 per group). Blue dashed lines outline the glomerulus and red dashed lines for the tubule. Scale bar, 100 µm. D) Representative images of immunofluorescent staining of UUO mice kidney samples. Expression and colocalization of α‐SMA (red) with β‐catenin (green) are shown. Grey signals represent the nucleus. Scale bar, 20 µm. Vehicle, Veh; low dose, LD; high dose, HD; pirfenidone, PFD; Ser, S; Thr, T. Data are presented as means ± SEM (A, B, and C) and were analyzed using one‐way ANOVA followed by a Bonferroni's multiple comparisons test. ^## p < 0.01, ^### p < 0.001, ^#### p < 0.0001 compared with sham group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with UUO‐Veh or between groups under the line group; ns, not significant.

Figure 6

DA directly bound to Cdc42 and inhibited its activity. A) Heatmap indicating ‐log₂(FC) of ΔT _m at each temperature for Uba5, Igf2bp1, Cdc42, Dctn1 and Plod2. Results are shown from 2 replicate experiments. B) Effects of the knockdown of Cdc42, Uba5, and Dctn1 on the protein levels of α‐SMA measured by western blotting (n = 3 per group). C) The HPCL/MS/MS analysis of NRK‐49F treated with vehicle or DA. D) CETSA analysis of intracellular binding between DA and Cdc42. Protein levels were investigated at different temperatures under the treatment of DA (5 µM) in NRK‐49F cells for 1 h. The graph shows the quantification of Cdc42 protein versus temperature points based on Western analyses (n = 3 per group). E) Surface plasmon resonance (SPR) analysis of interactions between DA or ZCL278 and Cdc42. F) Immunoblots of Cdc42 in NRK‐49F cells treated with vehicle or DA (n = 3 per group). G) Analysis of the active and total Cdc42 in NRK‐49F cells treated with vehicle or DA by pull‐down assay. Quantitative analysis of the grey value for active Cdc42/total Cdc42 ratio using ImageJ software (n = 3 per group). H) Representative immunofluorescence microscopy of active Cdc42. Serum‐starved NRK‐49F cells were treated with vehicle, 5 µM DA, or 5 µM ZCL278. Cells were probed with an active Cdc42 antibody (green). Nuclei were visualized by DAPI (blue). Scale bar, 20 µm. I) Analysis of the active and total Cdc42 in kidneys from mice treated with vehicle, low‐dose, or high‐dose DA by pull‐down assay (n = 6 per group). Results were quantified by ImageJ software. J) Molecular docking of Cdc42 and DA. K) CETSA analysis of intracellular binding between DA and wild‐type or mutant Cdc42 (n = 3 per group). Data were analyzed by two‐way ANOVA followed by Bonferroni's multiple comparisons test. Control, CTL; vehicle, Veh; WT, wild‐type; F, Phe; K, Lys; L, Leu. Data are presented as means ± SEM (B, D, F, G, I, and K) and were analyzed using one‐way ANOVA followed by a Bonferroni's multiple comparisons test unless otherwise stated. ^## p < 0.01, ^### p < 0.001, ^#### p < 0.0001 compared with CTL or sham group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with Veh group or between groups indicated by a line; ns, not significant.

Figure 7

DA activated p‐PCKζ/p‐GSK‐3β‐mediated β‐catenin phosphorylation at S33/37/T41 by targeting Cdc42. A, B) Immunoblots showing the inhibitory effects of DA and ZCL278 on fibrotic markers in A) TGF‐β1‐activated NRK‐49F cells (n = 3 per group) and B) UUO mice (n = 6 per group). C, E) Influence of Cdc42 knockdown on (C) fibrosis‐related proteins and (E) p‐PKCζ/p‐GSK‐3β/p‐β‐catenin protein levels in NRK‐49F cells (n = 3 per group). NRK‐49F cells were transfected with si‐Cdc42 or negative control siRNA (si‐nc) and incubated with TGF‐β1+Veh or TGF‐β1+DA for C) 48 h or E) 12 h. D, F) Immunoblot analysis of D) fibrotic indicators and F) p‐PKCζ/p‐GSK‐3β/p‐β‐catenin axis in NRK‐49F cells transfected with constitutively active Cdc42 plasmid (Cdc42^Q61L) (n = 3 per group). The post‐transfected NRK‐49F were treated with TGF‐β1+Veh or TGF‐β1+DA for D) 48 h or F) 12 h. G) Schematic diagram summarizing how active Cdc42 modulates p‐PKCζ/p‐GSK‐3β/p‐β‐catenin axis. H) Immunoblot analysis of p‐PKCζ and p‐GSK‐3β in mice kidneys of each group on the seventh days after sham or UUO surgery (n = 6 per group). Control, CTL; vehicle, Veh; NC, negative control; Ser, S; Thr, T. Data are presented as means ± SEM (A, B, C, D, E, F, and H) and were analyzed using one‐way ANOVA followed by a Bonferroni's multiple comparisons test. ^# p < 0.05, ^## p < 0.01, ^### p < 0.001, ^#### p < 0.0001 compared with CTL group; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 compared with Veh group or between groups indicated by a line; ns, not significant.

Figure 8

Higher expressions of Cdc42 were seen in both UUO mice and CKD patients. A) Immunoblot analysis of Cdc42 in mice kidneys of each group on the seventh day after sham or UUO surgery (n = 3 per group). B) Cdc42 transcript expression analysis for human kidneys from healthy controls (HC) and CKD patients in two public RNA‐sequencing datasets, GSE66494 and GSE115857. C) Correlation of the mRNA levels between Cdc42 and α‐SMA (Acta2), collagen I (Col1a1), collagen III (Col3a1), and fibronectin (Fn1) were measured by Pearson's correlation coefficient (r) in GSE115857. The data is shown from 7 HC and 23 CKD patients. D) Immunohistochemical staining of Cdc42 in kidney sections from HC (n = 5) and CKD (n = 12). Scale bar = 200 µm. Blue dashed lines outline the glomerulus and red dashed lines for tubule. Data are presented as means ± SEM (A, B, and D) and were analyzed by students’ t‐test. ^### p < 0.001, ^#### p < 0.0001 compared with sham or HC group.