Deletion of Fn14 receptor protects from right heart fibrosis and dysfunction

Abstract

Pulmonary arterial hypertension (PAH) is a fatal disease for which no cure is yet available. The leading cause of death in PAH is right ventricular (RV) failure. Previously, the TNF receptor superfamily member fibroblast growth factor-inducible molecule 14 (Fn14) has been associated with different fibrotic diseases. However, so far there is no study demonstrating a causal role for endogenous Fn14 signaling in RV or LV heart disease. The purpose of this study was to determine whether global ablation of Fn14 prevents RV fibrosis and remodeling improving heart function. Here, we provide evidence for a causative role of Fn14 in pulmonary artery banding (PAB)-induced RV fibrosis and dysfunction in mice. Fn14 expression was increased in the RV after PAB. Mice lacking Fn14 (Fn14(-/-)) displayed substantially reduced RV fibrosis and dysfunction following PAB compared to wild-type littermates. Cell culture experiments demonstrated that activation of Fn14 induces collagen expression via RhoA-dependent nuclear translocation of myocardin-related transcription factor-A (MRTF-A)/MAL. Furthermore, activation of Fn14 in vitro caused fibroblast proliferation and myofibroblast differentiation, which corresponds to suppression of PAB-induced RV fibrosis in Fn14(-/-) mice. Moreover, our findings suggest that Fn14 expression is regulated by endothelin-1 (ET-1) in cardiac fibroblasts. We conclude that Fn14 is an endogenous key regulator in cardiac fibrosis and suggest this receptor as potential new target for therapeutic interventions in heart failure.

Fig. 1

Fn14 expression in RV heart disease. a Real-Time PCR analyses of Fn14 mRNA expression from RV demonstrated elevated expression following PAB. No significant changes were observed in LVs (*p < 0.05, n = 4). Loading control: gapdh. b Western blot analysis of RV extracts of WT, SHAM- and PAB-operated mice showing a prevalent expression of Fn14 after PAB in RV. No significant difference was observed for TWEAK. Loading control: Pan-actin. c Quantitative analysis of b. Fn14 levels (normalized to pan-actin) of four individual hearts per condition were expressed as arbitrary units ± SEM. d Immunohistochemistry: Fn14 protein expression is elevated after PAB in RV of Fn14^+/+ animals. Nuclei were counterstained with methyl green. Sections from Fn14^−/− mice served as control for the specificity of the used anti-Fn14 antibody. e and f Co-staining experiments with fibroblast-markers (e, DDR2; f, P4HB: prolyl 4-hydroxylase, beta polypeptide) indicating that Fn14 expression is upregulated in fibroblasts after PAB in RV of Fn14^+/+ animals. Arrows: Fn14-positive fibroblasts. Stars: cardiomyocytes. g Western blot analysis: Fn14 is upregulated in isolated Fn14^+/+ CFs from RVs after PAB. Scale bars: 65 μm. RV right ventricle, LV left ventricle, PAB pulmonary artery banding

Fig. 2

Fn14^−/− mice show improved heart function after PAB. MRI imaging. a RV ejection fraction (EF) before PAB was not significantly different between wild-type and knockout mice (SHAM: Fn14^+/+: n = 4; Fn14^−/−: n = 6; PAB-operated mice: n = 11 both for Fn14^+/+ and Fn14^−/−). b RV EF was markedly decreased 3 weeks after PAB in Fn14^+/+ mice (n = 11, ***p < 0.0001). Reduction of RV EF was significantly inhibited in Fn14^−/− mice (n ≥ 4 for SHAM-operated mice; n = 11 for PAB-operated mice, *p < 0.05). c End-diastolic (ESV) and EDVs (*p < 0.05, n = 11 for PAB-operated mice). d Representative MRI images of hearts from SHAM- and PAB-operated mice. ED end-diastole, ES end-systole. e Peak of RV systolic pressure. RV right ventricle, PAB pulmonary artery banding. LVs are indicated by arrows, RVs by arrowheads

Fig. 3

Fn14 ablation reduces myocardial fibrosis after PAB. a Sirius Red staining demonstrating attenuation of RV fibrosis 3 weeks after PAB in Fn14^−/− mice. Scale bars: 300 μm. b Quantitative analysis (SHAM: Fn14^+/+ and Fn14^−/−: n = 4, PAB: Fn14^+/+: n = 5, Fn14^−/−: n = 7, *p < 0.05). c–f Real-Time PCR demonstrating that upregulation of Col1a1, Col1a2, Col3a1 and Col4a1 expression in RVs after PAB was diminished in Fn14^−/− mice. Expression levels were normalized to gapdh (**p < 0.005, *p < 0.05, n = 4). RV right ventricle, PAB pulmonary artery banding

Fig. 4

Fn14 signaling regulates collagen expression via the RhoA-Mal axis. (a and b) Luciferase reporter assays reveal that TWEAK stimulation activates Col1a1 and Col1a2 promoter (n = 7, **p < 0.01). HEK293T cells were transfected with indicated promoter constructs and 24 h later stimulated with TWEAK. c Western blot analysis of TWEAK-stimulated Rat2 fibroblasts showing markedly upregulated Collagen I expression. d TWEAK-induced production of collagens in CFs depends on Fn14 (n = 4, TWEAK versus serum-free, Fn14^+/+: *p < 0.05; Fn14^−/−: not significant). e TWEAK-induced nuclear translocation of MAL in cardiac fibroblasts depends on Fn14. Arrows indicate MAL-negative nuclei. Arrowheads indicate MAL-positive nuclei. Scale bars: 50 μm. f Promoter assays in HEK293T revealed enhancement of Col1a2 promoter activation after MAL and TWEAK overexpression. pEGFP-N1 (pEGFP) was used as control plasmid. g Amount of activated RhoA determined by immunoprecipitation with Rho-binding domain of Rhotekin (GST-RBD) in Fn14^+/+ and Fn14^−/− CFs. h Pre-incubation of NIH3T3 cells with ROCK kinase inhibitor Y27632 (20 μM) [*p < 0.05 vs. DMSO (control)] and CCG-1432 (5 μM) [**p < 0.005 vs. DMSO (control)] inhibited TWEAK-induced nuclear MAL translocation (n = 3). RV right ventricle, PAB pulmonary artery banding. (I) Real-Time PCR revealed a significant increase in ET-1 levels in RVs following PAB in wildtype Fn14^+/+ mice. Loading control: gapdh (n > 3, *p < 0.05). j Western blot analysis: ET-1 (100 nM) induces Fn14 expression in Fn14^+/+ CFs (n = 3). k Co-stimulation with TWEAK/ET-1 markedly induced MAL translocation in Fn14^+/+ CFs compared to ET-1 (*p < 0.01) and TWEAK stimulation alone (*p < 0.05) (four mice per each group, n = 4)

Fig. 5

Fn14 regulates myofibroblast differentiation. (a and b) Real-Time PCR revealed a significant reduction of SM22 and SMA induction in RVs of Fn14^−/− PAB-operated mice. Loading control: gapdh (n = 4, *p < 0.05). c Quantification of SMA-positive myofibroblasts (four mice per group, *p < 0.05). d TWEAK stimulation resulted in the enrichment of actin fibers (Rhodamine-phalloidin, red) in Fn14^+/+, but not Fn14^−/− CFs. Scale bars: 50 μm

Fig. 6

Fn14 regulates fibroblasts proliferation. a Quantitative analysis of PCNA-positive interstitial cells after PAB (Fn14^+/+: 6 and, Fn14^−/−: seven mice per group). b Representative examples of stained heart sections used to determine interstitial nuclear density. Scale bars: 100 μm. c Quantification of b (Fn14^+/+: six mice and Fn14^−/−: seven mice per group, **p < 0.005, ***p < 0.0001). d Immunofluorescence analyses: the majority of cells in the clusters are fibroblasts (DDR2). Red arrow: fibroblasts. Yellow arrows: non-fibroblast interstitial cells. Scale bars: 25 μm. e TWEAK increased the number of Fn14^+/+, but not Fn14^−/− CFs in a dosage-dependent manner (n = 5, **p < 0.001). f Proposed model of TWEAK/Fn14 axis activity during RV fibrosis. RV right ventricle, PAB pulmonary artery banding