Fibroblast growth factor-inducible 14 mediates macrophage infiltration in heart to promote pressure overload-induced cardiac dysfunction

Abstract

Aims:

Heart failure (HF) is characterized by compromised cardiac structure and function. Previous work has identified a link between upregulation of pro-inflammatory cytokines and HF. Tumor necrosis factor (TNF)-like weak inducer of apoptosis (TWEAK) is a pro-inflammatory cytokine, which binds to fibroblast growth factor inducible 14 (Fn14), a ubiquitously expressed cell-surface receptor. The objective of this study was to investigate the role of TWEAK/Fn14 pathway in promoting cardiac inflammation under non ischemic stress conditions.

Main Methods:

Wild type (WT) and Fn14 knock out (Fn14^−/−) mice were subjected to pressure overload [transaortic constriction (TAC)] for 1 or 6 weeks. A subset of WT TAC animals were treated with the Fn14 antagonist L524–0366. Cardiac function was measured by echocardiography. Cardiac fibrosis and macrophage infiltration were quantified using immunohistochemistry and flow cytometry, respectively. Cardiac fibroblasts were isolated for quantifying TWEAK-induced chemokine release.

Key Findings:

Fn14^−/− mice displayed improved cardiac function, reduced fibrosis and lower macrophage infiltration in heart compared to WT following TAC. L524–0366 mitigated maladaptive remodeling with TAC. TWEAK induced secretion of the pro-inflammatory chemokine, monocyte chemoattractant protein 1 from WT but not Fn14^−/− fibroblasts in vitro, in part through activation of non-canonical NF-κB signaling. Finally, Fn14 expression was increased in mouse following TAC and in human failing hearts.

Significance:

Our findings support an important role for the TWEAK/Fn14 promoting macrophage infiltration and fibrosis in heart under non-ischemic stress, with potential for therapeutic intervention to improve cardiac function in the setting of HF.

Introduction

Heart failure (HF) is a debilitating disease characterized by extensive cardiac remodeling (e.g. hypertrophy, fibrosis) and compromised cardiac function, affecting nearly 6 million people in the US with high morbidity and mortality [1]. Although the etiology of HF is complex and multifactorial, there is growing appreciation for the role that chronic inflammation plays in HF progression. While inflammation is a critical defense mechanism for resolving pathogenic stimuli and initiating repair following injury, dysregulation of this process promotes maladaptive remodeling and cardiac dysfunction [2–5]. In the setting of ischemic injury (myocardial infarction), inflammation plays an important role in reparative fibrosis required for scar formation, in the face of significant myocyte death. The inflammatory response to ischemic injury has been well characterized with necrotic cell death thought to release damage signals [damage-associated molecular patterns (DAMPs)], which activate the innate immune system, thereby launching a pro-inflammatory cascade [6, 7]. Inflammation is also triggered in the setting of non-ischemic injury (e.g. pressure overload) to promote interstitial and perivascular fibrosis [8, 9]. While release of chemokines from resident cardiac cells (fibroblasts, myocytes) likely plays an important role in activation of the innate immune system in non-ischemic injury [10], a precise understanding of how pro-inflammatory signals are elicited in non-ischemic disease remains elusive.

In previous studies, we identified increased expression of fibroblast induced growth factor 14 (Fn14), a tumor necrosis factor (TNF) receptor superfamily member, in mouse heart subjected to chronic pressure overload to induce HF (transaortic constriction, TAC) [11]. Increased Fn14 expression occurred downstream of stress-induced loss of the cytoskeletal protein β_IV-spectrin and subsequent dysregulation of gene programs controlled by the spectrin-associated transcription factor signal transducer and activator of transcription 3 (STAT3). Here, we test the hypothesis that Fn14 plays a role in generating a damage signal important for activation and infiltration of immune cells following TAC. Our findings demonstrate that Fn14 mediates secretion of monocyte chemoattractant protein 1 (MCP-1), a major chemokine for infiltration of monocytes/macrophages, from cardiac fibroblasts, which depends in part on activation of non-canonical NF-κB signaling. Furthermore, we show that genetic or pharmacological inhibition of Fn14 abrogates macrophage infiltration, maladaptive remodeling and cardiac dysfunction following TAC. Finally, we report increased Fn14 expression in mouse following TAC and in human failing hearts, suggesting that the Fn14 pathway serves as a potential therapeutic target to treat HF.

Methods

Experimental animals –

Adult (8–12 weeks) C57/BL6J male and female wildtype (WT, control) and global Fn14 knock out (Fn14^−/−) animals were used. Fn14 ^−/− mice were generously provided by Biogen, Inc. Mice were sacrificed using carbon dioxide followed by cervical dislocation, before tissue collection and downstream processing of tissues for cell isolation.

Murine HF model with proximal aortic banding –

TAC was performed to induce pressure overload conditions in adult mice, as described [11, 12]. Mice were anesthetized using isoflurane (3%), intubated, and placed on a respirator (120 breaths per minute − 1; 0.1 ml tidal volume). Aorta was exposed after performing a midline sternotomy and a 7.0 Prolene suture was tied around the aorta distal to the brachiocephalic artery using a blunted 27-gauge needle next to the aorta as a guide for the degree of aortic constraint. After tightening, the needle was removed, and the chest was closed. The Fn14 antagonist L524–0366 (9mg/kg in corn oil, intraperitoneal daily) was administered to a subset of mice 3 days after TAC surgery until 6 weeks post TAC.

Mouse echocardiography –

To evaluate cardiac structure and function in vivo, transthoracic echocardiography was performed on conscious mice using a VisualSonics Vevo 2100 ultrasound system (VisualSonics, Toronto, ON, Canada) before surgery and at 6 weeks following surgery to assess cardiac function, as described [11]. The MS-400 transducer was used in the short-axis M-mode to assess cardiac function.

Histology –

Mice were heparinized, and excised hearts were perfused with 10 ml of PBS followed by 20 ml of 4% paraformaldehyde (PFA)-PBS and then placed in 4% PFA-PBS for 24 h. Hearts were sectioned and stained using standard methods (Comparative Pathology and Mouse Phenotyping Shared Resource Research Core at Ohio State University). Heart sections were examined for fibrosis by Masson’s trichrome staining and automated quantitation using custom, publically available software[13]. For immunohistochemistry, paraffin-embedded heart sections were deparaffinized with xylene followed by rehydration with graded alcohols (100%, 95%, 70% and 50%). A heat-induced epitope retrieval procedure was carried out by heating the tissue slides in citrate buffer (pH 6.0) at 95°C for 10 min. Tissue sections were rehydrated with PBS, blocked (3% goat serum and 1.5% BSA) for 3 h at room temperature and used for immunohistochemistry. A subset of sections were incubated with anti F4/80 antibody (Serotec, catalog number: MCA497G). After 30 min incubation, sections were washed and incubated with biotinylated rabbit anti-rat antibody for 30 min at room temperature. After incubation, slides were rinsed, incubated with chromagen (DAB) for 5 minutes, counterstained with hematoxylin for 30 seconds and again rinsed with water to remove excess hematoxylin. Slides were dehydrated in ethanol, cleared in xylene and imaged. To assess cardiomyocyte cross-sectional area, a subset of tissue sections were incubated with Alexa Fluor 488 coupled Wheat Germ Agglutinin (WGA) (Invitrogen, Cat # W11261) and stained according to manufacturer’s manual. For immunofluorescence experiments, slides were blocked with 10% goat serum (Invitrogen, Cat # 50197Z) for 60 min, followed by incubation with primary antibodies against Rel A (1:100, Cell signaling, Catalog: D14E12) and Rel B (1:100, Invitrogen, Catalog: PA599541) overnight at 4°C. Sections were washed three times with blocking buffer and incubated with labeled secondary antibodies (1:500, Invitrogen, Cat # A32740). After two hours, sections were washed thoroughly with blocking buffer and imaged using Zeiss LSM 700 microscope

Quantitative RT-PCR –

Total RNA from WT and Fn14^−/− hearts and isolated fibroblasts treated with DNase I, was used for first-strand complementary DNA synthesis with the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). Quantitative PCR (qPCR) reactions were performed in triplicate on cDNA samples in 96-well optical plates with PowerUp SYBR Green Master Mix (Thermo Fisher Scientific) to maximize PCR precision and uniformity. PCR was performed at 95°C for 3 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute on an Applied Biosystems QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific). PCR data were analyzed using the relative standard curve method, and the ΔΔC_T method was used to calculate fold changes in relative gene expression. PCR products were confirmed by melt-curve analysis, amplicon length, and DNA sequencing. Rpl7 levels were used as a normalization control.

Target genes and primer sequences used to amplify the target genes were as follows: mouse Mcp-1: forward 5′-CTCTCTTCCTCCACCACCATG-3′ and reverse 5′-AGTGGGGCGTTAACTGCAT-3′; mouse Mip1α: forward 5′-TGCTTCTCCTACAGCCGGAA-3′ and reverse 5′-TTTGGAGTCAGCGCAGATCT-3′; mouse Mip1β: forward 5′- TGTCTGCCCTCTCTCTCCTC −3′ and reverse 5′- AGGAAGTGGGAGGGTCAGAG −3; mouse Mip2: forward 5′- GCCTTGACCCTGAAGCC −3′ and reverse 5′- TTCCCGGGTGCTGTTTG −3; mouse Rpl7: forward 5′- TGGAACCATGGAGGCTGT −3′ and reverse 5′- CACAGCGGGAACCTTTTTC −3′

Immunoblotting –

Whole tissue or cell lysates were produced in PhosphoSafe™ Extraction Reagent (Millipore Sigma, Burlington, MA) supplemented with protease inhibitor cocktail (Roche Diagnostics). Subsequently, samples were incubated in bicarbonate sample buffer, including 2% β-mercaptoethanol, for 10 min at 95°C. SDS-PAGE and immunoblot analysis were performed using the 10% Bio- Rad Gel System (Bio-Rad Laboratories). Membranes were blocked in 5% BSA in TBS-T and incubated in primary antibody overnight at 4°C and then secondary horseradish peroxidase (HRP)-conjugated antibodies for 1 h at room temperature. Primary antibodies include anti TWEAK Receptor/Fn14 antibody (Cell signaling, catalog number: 4403), anti GAPDH (Fitzgerald, catalog number: 10R-G109A) Images were generated using Pierce ECL Western Blotting Detection Reagent (Thermo Fisher Scientific) and the Biorad Universal Hood II Gel Doc System (Bio-Rad Laboratories).

Cytokine/Chemokine Analysis –

The level of cytokines/chemokines was measured in cell culture supernatant of cardiac fibroblasts isolated from both WT and Fn14^−/− hearts using the proteome profiler array kit (ARY006, R&D Systems, MN, USA) according to the manufacturer’s manual. Cardiac fibroblasts were cultured in DMEM with 10% FBS until 80% confluency in 100mm² culture dish. Cells were serum starved for 24 hours prior to incubation with TWEAK. After 24 hours of serum starvation, cells were incubated in100μg/ml TWEAK for 24 hours in serum free media. Cell culture media was collected and centrifuged at 2000 rpm for 5 min and supernatant was stored at −80°C, until samples were analyzed

Flow cytometry Analysis -

Cardiac mononuclear cells were isolated as described previously[14]. Briefly, whole hearts were flushed with PBS, finely-minced using a single-edged blade, and digested in RPMI medium with 1 mg/ml collagenase-2 (Worthington Biochemical, Lakewood, New Jersey) at 37°C for 20 min. Cell suspensions were filtered through a 40 μm cell strainer into 1mg/ml BSA in PBS with 2mM EDTA on ice. Undigested tissue chunks were pulverized using the blunt end of a 3ml syringe. Cells were collected, washed with PBS, and fixed with 1% paraformaldehyde. The cell pellet was reconstituted in 500 μl of staining buffer and 200μl of cell-suspension was aliquoted into a 5-mL round bottom polystyrene FACS tube. An appropriate amount of Fc blocker was added (BioLegend catalog no. 101302), and the samples were incubated on ice for 10 min to block nonspecific binding sites. Total amount of antibodies needed for all the samples was calculated and added together to prepare a cocktail in a 0.5-mL microfuge tube. The antibodies CD45-PE/Cy7, clone I3/2.3 (catalog no. 147704); CD11b-APC/Cy7 (catalog no. 101226); Ly6G-PE, clone 1A8 (catalog no. 127608), CD64 BV605 clone X54–5/7.1 (catalog no. 139323), CCR2 FITC clone SA203G11 (catalog no. 150608), MHCII APC (clone M5/114.15.2 (catalog no. 107614) were obtained from BioLegend (San Diego, CA). The antibody CD4-PerCP, clone RM4-5 (catalog no. 45-0042-82) and Ly6C-eFluor 450 were obtained from Invitrogen (Carlsbad, CA). Samples were incubated for 30–45 min on ice, followed by the addition of 4 mL cold PBS to wash excess antibody, and the cells were pelleted by centrifugation at 500 g for 10 min. The supernatant was discarded by decanting, and cells were resuspended in the left-over volume by mild to moderate vortexing. Cell counting beads (2 μL; Spherotech, ACBP-100–10, 1 × 10⁶ beads/mL) were added to each sample. Data was collected using BD Fortessa flow cytometer and was analyzed using Flowjo software (version 10.2).

Human heart samples –

Left ventricular tissue was obtained from explanted hearts from patients undergoing heart transplantation at The Ohio State University. Tissue from non-failing donor hearts not suitable for transplantation was obtained through Lifeline of Ohio.

Statistics –

Comparisons were made in Sigma Plot (San Diego, CA) using an unpaired t-test for single comparisons or one-way ANOVA with Sidak Holm’s post hoc analysis. Two-way Anova with Tukey’s post hoc test was used to compare differences between immune cell infiltration between the different treatment for multiple comparisons. The null hypothesis was rejected for P value < 0.05. Summary data are presented as mean ± SEM.

Study approval –

Animal studies were conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the NIH using protocols that were approved by the Institutional Animal Care and Use Committee (IACUC) at The Ohio State University. The local Institutional Review Board approved the use of human tissue and the investigation conforms to the principles outlined in the Declaration of Helsinki.

Results

Fn14 receptor mediates pressure overload induced pathological remodeling in vivo.

To understand the in vivo role of Fn14 in pressure overload induced heart failure, Fn14^−/− and WT mice were subjected to transaortic constriction (TAC). Cardiac function was assessed by 2D echocardiography at baseline and following 6 weeks of TAC. Fn14^−/− animals displayed a trend towards reduced ejection fraction (EF) and fractional shortening (FS) at baseline compared to WT but the differences did not reach statistical significance (Table 1). Interestingly, Fn14^−/− mice experienced a less dramatic decrease in EF and FS following TAC compared to WT (Table 1 and Figure 1A–C). Fn14^−/− mice also showed a trend towards less hypertrophy following TAC as assessed by left ventricular wall thickness (Table 1) together with a slight but significant reduction in cardiomyocyte cross-sectional area, compared to WT TAC (Figure S1). Left ventricular chamber diameter was not different between Fn14^−/− and WT at baseline or following TAC (Table 1). Histological analysis revealed no significant difference in the degree of fibrosis between WT and Fn14^−/− hearts at baseline, as measured by Masson’s trichrome staining. However, TAC induced a greater degree of fibrosis in WT hearts compared to Fn14^−/− (Figure 1D–E). Quantitative PCR analysis of select fibrotic genes showed no significant differences between WT and Fn14^−/− hearts at baseline, however, TAC induced a significant increase in Acta2 and Col1a1 in WT compared to Fn14^−/− mice (Figure 1F). Together, these data indicate that genetic inhibition of Fn14 abrogates pressure overload-induced fibrosis, remodeling and cardiac dysfunction.

Table 1.

Echocardiographic parameters in wildtype (WT) and Fn14^−/− mice at baseline and WT mice subjected to 6 weeks of transaortic constriction (TAC)

Parameters	WT baseline (N=9)	WT TAC (N=9)	Fn14−/− baseline (N=9)	Fn14−/− TAC (N=9)
EF (%)	67.6± 2.9	40.9± 3.6*	58.6±1.5	45.5±2.9*
FS (%)	35.0±2.3	18.5±2.1*	29.7±1.2	21.9±1.8*
LVID,d (mm)	3.72±0.08	3.87±0.12	3.81 ±0.14	3.93±0.18
LVPW,d (mm)	0.76±0.08	1.27±0.12*	0.72±0.03	1.12±0.10*
HR (bpm)	412	458	422	474

EF = ejection fraction;

FS = fractional shortening;

LVID,d = left ventricular inner chamber diameter in diastole;

LVPW,d = LV posterior wall thickness in diastole;

HR = heart rate;

^*P<0.05 vs. Baseline

Figure 1. Echocardiographic assessment of LV function in WT and Fn14^−/− animals.

(A) Representative echocardiograms from WT and Fn14 knock out (Fn14^−/−) animals at baseline and following 6 weeks of transaortic constriction (TAC). (B-C) Summary data (mean±SEM) of changes in ejection fraction and fractional shortening in WT and Fn14^−/− animals, 6 weeks after TAC. *P < 0.05 versus WT TAC by 2- tailed t test. N = 9 for both groups, where N = number of animals. Echocardiographic parameters: ΔEF – change in ejection fraction from baseline to 6 weeks after TAC, ΔFS – change in fractional shortening. (D) Masson’s trichrome–stained heart sections (collagen is labeled in blue, scale bar = 50μm); and (E) Summary data (mean±SEM) showing fibrosis as a percentage of the tissue area at baseline and following 6 weeks of TAC. *P < 0.05 versus baseline. #P<0.05 versus Fn14 TAC by 1-way ANOVA with the Holm-Sidak post hoc test. N = 3 for baseline and N= 6 for TAC for both groups, where N=number of animals. (F) Expression of select fibrotic genes (Acta2, Col1a1,Pdgfra and Tcf21) as quantified by quantitative PCR in whole hearts either at baseline or after 6 weeks of pressure overload (TAC) in WT and Fn14^−/− mice. Rpl7 was used as a control. *P < 0.05 vs. Fn14^−/− TAC by 1-way ANOVA with the Holm-Sidak post hoc test. N = 3 for baseline and N=5 for TAC each group, where N = number of independent preparations.

Fn14 regulates macrophage infiltration in hearts subjected to pressure-overload

Previous reports demonstrated a significant increase in macrophage density and an upregulation of pro-inflammatory mediators in heart following 7 days of TAC [15]. To determine whether the difference in cardiac remodeling response in Fn14^−/− mice was associated with changes in cardiac infiltration of immune cells, cardiac monocytes/macrophages (F4/80⁺ cells) were quantified in Fn14^−/− and WT ventricular sections following 7 days of TAC, before decline in cardiac function was apparent. While the numbers of CD11b⁺F4/80⁺ monocytes/macrophages were similar in WT and Fn14^−/− hearts at baseline (Figure 2A–C), 7 days of TAC induced a dramatic increase in F4/80⁺ monocytes/macrophages in perivascular regions of WT hearts compared to Fn14^−/−.

Figure 2. Effect of cardiac pressure overload stress on macrophage infiltration in hearts.

Representative F4/80 stained paraffin embedded sections of (A) WT and (B) Fn14 knock out (Fn14^−/−) hearts following 7 days of TAC. Black arrows highlight perivascular infiltration of macrophages. Scale bar = 50μm. (C) Quantitative estimation of macrophages in perivascular regions (12.6 mm²) of WT and Fn14^−/− TAC hearts. *P < 0.05 versus WT, #P < 0.05 versus TAC by 1-way ANOVA with the Holm-Sidak post hoc test. N = 3 for each group, where N = number of animals.(D) Representative flow cytometry gate plots of MHCII^high/low macrophages pre-gated on CD45⁺CD11b⁺Ly6G⁻CD64⁺ cells from TAC mice. (E-F) Quantification of infiltrating (CCR2⁺MHCII^high) and resident (CCR2⁻MHCII^high) macrophage subsets.

Flow cytometry analysis was performed using a gating strategy to evaluate the immune cell composition in heart. (Figure S2). Consistent with previous reports[16], flow cytometry analysis confirmed increased infiltration of CD45⁺ immune cells (Figure S3) with significantly increased inflammatory CCR2⁺MHCII^lowLy6C^high and patrolling CCR2⁺MHCII^lowLy6C^low monocytes (Figure S3) in TAC hearts as compared to baseline. While baseline hearts were predominantly enriched with resident macrophages (CD11b⁺CD64⁺CCR2⁻MHC^high) and with a small fraction of infiltrating monocyte-derived macrophages (CD11b⁺CD64⁺CCR2⁺MHC^high); we observed a significant increase in infiltrating monocyte-derived macrophages and a significant decrease in resident macrophages at 7d post-TAC (Figure S4) supporting previously published results[16]. Consistent with immunohistochemistry data, the frequency of resident and infiltrating monocyte-derived macrophages was significantly decreased in Fn14^−/− hearts following 7 days of TAC, as compared to WT (Figure 2D–F). In addition, when compared with WT TAC, increased frequency of patrolling CD11b⁺CD64⁺CCR2⁺MHC^low Ly6C^low monocytes was observed in Fn14^−/−TAC hearts, whereas no difference was observed in pro-inflammatory CD11b⁺CD64⁺CCR2⁺MHC^lowLy6C^high monocytes (Figure S3). Since, Ly6C^low monocytes are also known to promote wound-healing and augment reparative processes, these data are consistent with the cardiac protection observed in Fn14^−/− TAC mice

TWEAK-Fn14 pathway promotes MCP-1 secretion in cardiac fibroblasts

We hypothesized that cardiac fibroblasts are a major source for inflammatory chemokines activated by the TWEAK-Fn14 pathway to promote macrophage infiltration. To test this hypothesis, expression of the gene for MCP-1 (Mcp1) and other pro-inflammatory chemokines (Ccl11, Mip1a and Mip1b) were assessed by qPCR in isolated WT and Fn14^−/− cardiac fibroblasts at baseline, and following treatment with the Fn14 agonist TNF-like weak inducer of apoptosis (TWEAK, 100ng/ml). TWEAK increased the expression of Ccl11 (non-significant trend) and Mcp-1 expression in WT fibroblasts (Figure 3A). Fn14^−/− fibroblasts showed a similar change in Ccl11 but not in Mcp-1 with TWEAK, compared to WT. Levels of select chemokines were also determined at the protein level using a dot blot assay (panel of 40 different chemokines). Consistent with qPCR findings, MCP-1 levels were elevated in cell culture supernatant collected from WT fibroblasts compared to that from Fn14^−/− fibroblasts (Figure 3B and C). Together, these data indicate that TWEAK/Fn14 pathway specifically regulates MCP-1 expression and secretion from cardiac fibroblasts.

Figure 3. TWEAK regulates MCP-1 expression in cardiac fibroblasts.

(A) Expression of select genes of cardiac pro-inflammatory chemokines (Ccl-11, Mcp-1, Mip-1a, Mip-1b) were quantified by quantitative PCR. Rpl7 was used as a control. *P < 0.05 vs. control, ^#P < 0.05 vs. Fn14^−/−, by 1-way ANOVA with the Holm-Sidak post hoc test. N = 3 for WT and Fn14^−/− control, and N=9 for WT and Fn14^−/− fibroblasts incubated with TWEAK, where N = number of independent preparations. Analysis of chemokines/cytokines in cell culture supernatant of (B) WT and Fn14^−/− fibroblasts incubated with 100μg/ml TWEAK for 24 hours in serum free media, as analyzed by proteome profiler mouse cytokine array(1:TIMP-1, 2:M-CSF, 3:MCP-1, 4:SDF-1). (C) Semi quantitative estimation of chemokines/cytokines in cell culture supernatant from WT and Fn14^−/− fibroblasts incubated with 100μg/ml TWEAK for 24 hours.

TWEAK-Fn14 signaling activates the non-canonical NF-κB pathway to promote cardiac inflammation

The NF-κB pathway is a central mediator of inflammation in heart. Previous studies have reported that TNF super family members activate NF-κB signaling to promote inflammatory cytokine secretion [17]. To test whether TWEAK/Fn14 acts through the NF-κB pathway in cardiac fibroblasts, subcellular localization of RelA and RelB (markers of canonical and non-canonical NF-κB signaling, respectively[18]) were assessed in isolated fibroblasts following treatment with TWEAK in the presence or absence of the Fn14 antagonist L524–0366 (25 μM). Interestingly, TWEAK induced a clear nuclear translocation of RelB, which was abrogated by L524–0366 (Figure 4A–B), indicating engagement of the non-canonical NF-κB pathway. In contrast, TWEAK had no significant effect on RelA subcellular localization in mouse cardiac fibroblasts (not shown). To determine whether cross-talk between TWEAK/Fn14 and NF-κB modulated Mcp-1 expression in cardiac fibroblasts, qPCR was performed on isolated WT cardiac fibroblasts incubated under control conditions (normal media) or treated with TWEAK in the presence or absence of pretreatment with the NF-κB inhibitor BMS-345541 (10μM). NF-κB inhibition partially abrogated the TWEAK-induced increase in Mcp-1 observed in WT fibroblasts (Figure 4C), suggesting that NF-κB pathway participates in TWEAK-Fn14-dependent regulation of Mcp-1 production in cardiac fibroblasts. Interestingly, cardiac fibroblasts treated with TWEAK in the presence of BMS-345541 and L524–0366 (25 μM), almost completely normalized Mcp-1 expression (Figure 4C).

Figure 4. Cross-talk between FN14/TWEAK and NF-κB signaling pathways in cardiac fibroblasts.

(A) Representative confocal images of isolated, fixed WT fibroblasts immunostained for RelB (red) following 24 hr treatment with DMSO (0.1% in PBS, control), TWEAK (100 ng/ml), and TWEAK (100 ng/ml) in the presence of the Fn14 antagonist L524–0366 (25μM). Nuclei are indicated by DAPI staining (blue in merged images). Scale bar = 20μm. (B) Summary data showing nuclear RelB immunoreactive signal as a percentage of total RelB for control and TWEAK in the presence or absence of L524–0366. *P< 0.001 vs. control; #P<0.001 vs. L524–0366; N = 3 animals with 30 cells analyzed per animal. (C) Expression of Mcp-1 (encodes monocyte chemoattractant protein 1) as quantified by quantitative PCR in wildtype cardiac fibroblasts under control conditions (0.1%DMSO) and following treatment with TWEAK alone, TWEAK in the presence of the NF-κB inhibitor BMS-345541 (10 mM, TWEAK+BMS), or TWEAK with BMS-345541 and L524–0366 (TWEAK+BMS+L524). Rpl7 was used as a control. *P < 0.05 vs. WT control, # P < 0.05 vs. WT TWEAK by 1-way ANOVA with the Holm-Sidak post hoc test. N = 6 for each group, where N = number of independent preparations.

Fn14 expression is upregulated in human and murine HF samples

Previous reports demonstrated that soluble TWEAK levels are elevated in patients with idiopathic dilated cardiomyopathy, and overexpression of soluble TWEAK in mice resulted in heart failure and early mortality [19]. At the same time, we have previously shown that Fn14 mRNA is increased in mouse heart following TAC downstream of loss of the cytoskeleton protein β_IV-spectrin [11]. To determine whether Fn14 expression (at the protein level) is altered by pressure overload in the mouse, Fn14 levels were evaluated by immunoblot in WT mice at baseline and after 6 weeks of TAC. A significant increase in Fn14 expression was observed in WT TAC compared to control (Figure 5A). As a first step in determining relevance to human, Fn14 expression was assessed in ventricular lysates from normal (non-failing) and failing human hearts. Consistent with findings in the mouse, Fn14 expression was significantly increased in human HF samples compared to non-failing controls (Figure 5B).

Figure 5. Expression of Fn14 in mouse model and human heart failure.

(A) Representative immunoblots of Fn14 in WT mouse hearts at baseline (control) and following 6 weeks of TAC (*P<0.05 vs. control, N=3 for each group). (B) Representative immunoblots of Fn14 in ventricular lysates from nonfailing (control) and failing human hearts (black line separates noncontiguous lanes from the same immunoblot). Quantitative analysis demonstrated significantly higher Fn14 expression in WT TAC and in human HF samples. *P<0.05 vs. control, N=3 for non-failing controls and N=6 for HF, where N is the number of hearts.

Fn14 Inhibition Offers Protection from Pressure Overload Induced Heart Failure

Together, our findings suggest that Fn14 may serve as a therapeutic target in the setting of chronic pressure overload. As proof of concept for targeting of Fn14 as a therapeutic strategy, WT mice were subjected to TAC in the presence of an Fn14 antagonist (L524–0366, 9 mg/kg, intraperitoneal daily) or vehicle control. Echocardiography analysis revealed that L524–0366 treated animals showed a blunted decrease in cardiac function without affecting hypertrophy following 6 weeks of TAC compared to vehicle control, (Figure 6). Together, these data indicate that Fn14 inhibition is effective at mitigating maladaptive remodeling in the setting of chronic pressure overload.

Figure 6. Fn14 antagonist ameliorates cardiac remodeling and dysfunction in chronic pressure overload.

(A) Representative echocardiograms from WT mice injected with vehicle (corn oil, IP, daily) or the Fn14 antagonist L524–0366 (9mg/kg) following 6 weeks of TAC. (B–E) Summary data (mean±SEM) of changes in echocardiographic features, 6 weeks post TAC with L524–0366 or vehicle. *P < 0.05 versus baseline, ^#P < 0.05 versus vehicle by 1-way ANOVA with the Holm-Sidak post hoc test. N = 3 for placebo and N = 5 for antagonist, where N is the number of animals.

Discussion

Our current work demonstrates that the TNF receptor family member Fn14 regulates macrophage infiltration, fibrosis and maladaptive remodeling response under chronic pressure overload conditions. In addition, our studies suggest that Fn14/TWEAK signaling activates downstream non-canonical NF-κB signaling to promote MCP-1 secretion in mouse cardiac fibroblasts, which may provide a critical signal for infiltrating macrophages. Importantly, we report upregulation of Fn14 in both mouse and human non ischemic HF conditions. Finally, we show that genetic or pharmacological inhibition of Fn14 ameliorates maladaptive remodeling and cardiac dysfunction induced by pressure overload. Together, our new data identify the TWEAK-Fn14 signaling axis as a promising target for modulating the cardiac remodeling response to chronic non-ischemic stress.

The TWEAK/Fn14 pathway has been previously implicated in maladaptive cardiac remodeling and HF [19–21]. Previous studies have reported elevated levels of circulating TWEAK in serum samples from patients with dilated cardiomyopathy, and an overexpression of soluble TWEAK led to cardiac hypertrophy and HF in mice [19]. Activation of Fn14 signaling by TWEAK upregulates expression of Fn14 through a positive feedback loop that promotes sustained Fn14 expression [22]. In addition, previous reports have demonstrated that Fn14 expression is promoted by hypertrophic agonists including angiotensin II (Ang II), phenylephrine (PE), and endothelin-1 (ET-1)[23]. Fn14 deletion offered protection against right ventricular (RV) hypertrophy and fibrosis, caused by pulmonary artery banding (PAB). In vitro experiments also demonstrated that TWEAK/Fn14 signaling promotes fibroblast proliferation, collagen synthesis and myofibroblast differentiation [21]. Our findings provide additional mechanistic insights into the role of TWEAK-Fn14 pathway in promoting macrophage infiltration, fibrosis and remodeling in the setting of non-ischemic HF.

TWEAK/Fn14 signaling has been shown to promote cardiac fibroblast proliferation, extracellular matrix gene expression and collagen production via activation of NF-κB pathways [19, 21, 24]. TWEAK/Fn14 axis has been shown to activate multiple downstream signaling pathways in heart, however, it signals predominantly through the NF-κB pathway [24–28]. In the unstimulated state, NF-κB associates with an inhibitory protein, IκB (inhibitor of κB) and is sequestered in the cytoplasm. Upon activation through the canonical pathway, IKK (IκB kinase) phosphorylates the inhibitory protein IκB which causes its degradation and releases NF-κB into the nucleus, activating the gene transcription of multiple cytokines such as IL-6, IL-8 and RANTES[29]. TWEAK/Fn14 complex is also known to activate non-canonical NF-κB signaling pathway signal via association of Fn14 with TNF receptor associated factor (TRAF) proteins such as TRAF2 and TRAF5[30]. Das et al reported that cytoplasmic adapter molecule TRAF3 Interacting Protein 2 (TRAF3IP2) mediates TWEAK-induced p38 MAPK, NF-κB and AP-1 activation in cardiac fibroblasts to regulate their proliferation, migration, collagen expression and proinflammatory response [31] Consistent with these earlier reports, our work demonstrates that TWEAK/Fn14 signaling acts through non-canonical NF-κB pathway to enhance MCP-1 expression and secretion from cardiac fibroblasts. Pretreatment of cardiac fibroblasts with both the NF-κB inhibitor BMS-345541and Fn14 antagonist L524–0366 had an additive effect in suppressing Mcp-1 expression. Previously, BMS-345541 has been shown to associate with both IKKα (involved in both canonical and non-canonical signaling) and IKKβ (involved only in canonical signaling).However, it has an approximately 10-fold greater inhibitory effect on IKKβ than on IKKα, therefore it is more effective in blocking canonical over non-canonical NF-κB pathway. Immunofluorescence data (Figure 4) suggests that Fn14 antagonist predominantly blocks the non-canonical pathway. Therefore, the combination of both BMS-345541 and L524–0366 could be more effective in blocking both, canonical and non-canonical NF-κB pathways, to achieve greater suppression of MCP-1 production. In addition, Fn14 antagonist could be effective in blocking other TWEAK-Fn14 activated signaling pathways such as c-Jun N-terminal kinase pathway [32] which were shown to activate MCP-1 production[33].

While these studies identify a role for TWEAK/Fn14 in cardiac remodeling, the upstream signals for dysregulation of this axis in chronic stress remain to be elucidated. Previous studies have shown increased TWEAK levels in HF, although the mechanism is unclear [19]. At the same time, we have identified a link between adrenergic stress, stability of the spectrin-based cytoskeleton and activity of the pro-fibrotic transcription factor STAT3 in cardiomyocytes and fibroblasts [11, 34]. Specifically, we report that TAC induces β_IV spectrin degradation, thereby disrupting the spectrin-STAT3 complex and promoting STAT3 nuclear translocation and thereby promoting fibroblast activation. While STAT3 directly regulates fibrosis, cell proliferation and ECM synthesis, we (27) and others (7) have shown that STAT3 regulates Fn14 expression. Based on these data, it will be interesting to explore whether β_IV spectrin/STAT3/Fn14 links chronic adrenergic stress to activation of cardiac fibroblasts and the downstream inflammatory response. Although our studies focus on the role of TWEAK/Fn14 signaling in cardiac fibroblasts in promoting macrophage infiltration, fibrosis and cardiac dysfunction under pressure overload conditions, we cannot rule out involvement of other cell types in observed phenotypes associated with global Fn14 inhibition. The Fn14 receptor is expressed systemically, and is found in multiple cell types including neutrophils, macrophages, monocytes and dendritic cells[35, 36]. Similarly, Fn14 signaling in endothelial cells has been shown to regulate cell expansion and angiogenesis with potential implications for cardiac remodeling response to chronic stress[37]. Future studies using tissue-specific Fn14 knockout mice will help address the important question of whether cardiac protection afforded by Fn14 inhibition is due specifically to its activity in fibroblasts and/or other cell types. Finally, while our studies show a reduction in fibrosis and improvement in cardiac function with Fn14 inhibition following TAC, we did not track the remodeling process beyond 6 weeks of injury and long term studies might offer greater insights into long term protection offered, if any, by Fn14 inhibition

Conclusion

In summary, we report the role of TWEAK-Fn14 pathway in pressure overload induced cardiac dysfunction. Specifically, we showed that TWEAK/Fn14 signaling activates non-canonical NF-κB pathway to promote MCP-1 secretion in cardiac fibroblasts. Secretion of MCP-1 results in excessive infiltration of macrophages in myocardium and eventually leads to increased cardiac fibrosis and decreased cardiac function. We also demonstrate that genetic and pharmacological deletion of Fn14 offers protection from pressure overload induced stress and therefore, blocking TWEAK-Fn14 signaling could serve as a potential therapeutic intervention in setting of non-ischemic HF.

Highlights:

• Genetic deletion of the TNF receptor family member Fn14 (Fn14^−/−) reduces cardiac fibrosis and improves cardiac function in response to chronic pressure overload compared to WT.

• Fn14^−/− mice show a reduction in resident and infiltrating macrophages in myocardium in response to chronic pressure overload compared to WT

• TWEAK-Fn14 signaling activates downstream non-canonical NF-kB pathway to promote MCP-1 secretion in cardiac fibroblasts

• TWEAK-Fn14 pathway mediates pressure overload cardiac dysfunction and serves as a potential therapeutic target in non-ischemic HF