Induction of Pro-Fibrotic CLIC4 in Dermal Fibroblasts by TGF-β/Wnt3a Is Mediated by GLI2 Upregulation

Abstract

Chloride intracellular channel 4 (CLIC4) is a recently discovered driver of fibroblast activation in Scleroderma (SSc) and cancer-associated fibroblasts (CAF). CLIC4 expression and activity are regulated by TGF-β signalling through the SMAD3 transcription factor. In view of the aberrant activation of canonical Wnt-3a and Hedgehog (Hh) signalling in fibrosis, we investigated their role in CLIC4 upregulation. Here, we show that TGF-β/SMAD3 co-operates with Wnt3a/β-catenin and Smoothened/GLI signalling to drive CLIC4 expression in normal dermal fibroblasts, and that the inhibition of β-catenin and GLI expression or activity abolishes TGF-β/SMAD3-dependent CLIC4 induction. We further show that the expression of the pro-fibrotic marker α-smooth muscle actin strongly correlates with CLIC4 expression in dermal fibroblasts. Further investigations revealed that the inhibition of CLIC4 reverses morphogen-dependent fibroblast activation. Our data highlights that CLIC4 is a common downstream target of TGF-β, Hh, and Wnt-3a through signalling crosstalk and we propose a potential therapeutic avenue using CLIC4 inhibitors.

Keywords: CLIC4; Ion channels; Scleroderma; fibrosis; morphogens.

Figure 1

CLIC4 expression is regulated by a number of morphogens in dermal fibroblasts. The healthy dermal fibroblasts were grown in serum-depleted media and stimulated with TGF-β, Wnt3a, and Smoothened agonist (SAG) for 48 h. CLIC4 transcript (A) and protein levels (B,C) were assessed. (C) Immunohistochemistry analysis for CLIC4 expression (Brown) in skin sections from wild-type and TβRIIΔK-fib transgenic mice. The scale bars represent 100 μm. Healthy dermal fibroblasts that were stimulated with TGF-β (D), Wnt3a (E), or SAG (F) for 48 h in the absence or presence of 1 μM SIS3 (D), 10 μM FH535 (E), or 10 μM GANT61 (F). CLIC4 and α-SMA protein levels were assessed. (G) Densitometry analysis of CLIC4 blots from (E) and (F). The graphs represent the means +/− standard error for three independent experiments. * p < 0.05.

Figure 2

Wnt/Hh/TGF-β cross talk is important for CLIC4 expression in dermal fibroblasts. Healthy dermal fibroblasts that were treated with SMAD3/scramble siRNA (A,B) or SIS3 (C), followed by transfection with the TOP-Flash-Firefly and CMV-Renilla luciferase reporters prior to treatment with Wnt3a. The relative TOP-Flash reporter activity was determined by the dual luciferase assay. (C) β-catenin and α-SMA protein levels from TGF-β-stimulated fibroblasts. (D) GLI2 transcript from Wnt3a/FH535-treated fibroblasts. GLI2 protein (E) and transcript (F) from TGF-β/FH535-treated fibroblasts. β-catenin protein (G) and transcript (H) from SAG/GANT61-treated fibroblasts. (I) CLIC4 and α-SMA protein from Wnt3a/SAG/SIS3-treated fibroblasts. (J) CLIC4 and α-SMA protein from TGF-β/SIS3/FH535/GANT61-treated fibroblasts (K) Schematic of the hierarchical relationship. The graphs represent the means +/− standard error for three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3

The regulation of CLIC4 expression by TGF-β and Hh signaling is conserved in other cell types. RNA and protein were extracted from Isogenic SW620 cells expressing endogenous WT PTCH1 or mutated PTCH1 (PTCH1^mut). RNA-seq analysis was performed on the RNA. (A) The table represents the fold change of genes relating to the TGF-β and Hh signalling pathways. (B) SMAD2, 3, 4, and CLIC4 protein levels were analysed by Western blot. WT PTCH1 or mutated PTCH1 (PTCH1^mut) were treated with 20 μM GANT61, 10 μM SD208, or DMSO for 24 h. GLI1 (C) and CLIC4 (D) transcript levels and protein levels (E) were assessed. The graphs represent the mean +/− standard error of three independent experiments. * p < 0.05, ** p < 0.01.

Figure 4

β-Catenin contributes to the enhanced CLIC4 expression in SSc fibroblasts. RNA and protein were extracted from healthy and SSc dermal fibroblasts that were treated with 10 μM FH535 or DMSO (vehicle) for 48 h. CLIC4 (A), α-SMA (B) transcript, CLIC4, and α-SMA protein levels (C) were assessed. The healthy and SSc fibroblasts were transfected with siRNA that was specific to β-catenin or scramble control siRNA. (D) CLIC4 and β-catenin protein levels were assessed by Western blot. β-actin served as a loading control. The graphs represent the mean +/− standard error for three independent experiments using three different patient cell lines. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

GLI2 is an important mediator of enhanced CLIC4 levels in SSc fibroblasts. CLIC4 transcript levels (A) and protein levels (B) from healthy and SSc fibroblasts that were treated or not with 10 μM GANT61. The healthy and SSc fibroblasts were transfected with siRNA that was specific to GLI2 or scramble control siRNA. (C) The GLI2 transcript levels were assessed by q-RT-PCR. (D) CLIC4 mRNA levels were quantified in the same samples from (C). (E) CLIC4, GLI2, and β-catenin protein levels from healthy and SSc fibroblasts that were treated with GLI2 or scrambled siRNA were assessed by Western blot. β-actin served as a loading control. The graphs represent the mean +/− standard error for three independent experiments using three different patient cell lines. * p < 0.05, ** p < 0.01.

Figure 6

The inhibition of CLIC4 blocks Wnt3a-/Hh-/TGF-β-mediated fibroblast activation. Healthy dermal fibroblasts were stimulated with TGF-β (A), Wnt3a (B), and SAG (C) in the presence and absence of the chloride channel inhibitors NPPB and IAA-94 for 48 h. The α-SMA, β-catenin, SMAD3, and GLI2 protein levels were analysed by Western blot. β-actin served as a loading control. (D) The healthy and SSc patient fibroblasts were treated with NPPB and IAA-94 for 48 h. The β-catenin and GLI2 protein levels were analysed by Western blot. β-actin served as a loading control. The graphs represent the mean +/− standard error for three independent experiments. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 7

The schematic of the role of CLIC4 in morphogen-mediated fibroblast activation.