J.S., B.J., Y.Q., J.G., M.J., X.C., M.B., D.P.K. and R.L. have declared no competing interests for this work. L.S. and L.C.B. are employees of Biogen-IDEC, and the authors confirm that this does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.
Contributed to data interpretation and scientific discussion: LCB DPK MJ. Conceived and designed the experiments: JS BJ MJ RL. Performed the experiments: JS JG MB YQ XC. Analyzed the data: JS BJ. Contributed reagents/materials/analysis tools: LS LCB TCL DPK. Wrote the paper: JS BJ MJ RL.
Inflammatory cytokines play an important role in the pathogenesis of heart failure. We have recently found the cytokine TWEAK (tumor necrosis factor (TNF)-like weak inducer of apoptosis), a member of the TNF superfamily, to be increased in patients with cardiomyopathy and result in the development of heart failure when overexpressed in mice. The molecular mechanisms underlying TWEAK-induced cardiac pathology, however, remain unknown.
Using mouse models of elevated circulating TWEAK levels, established through intravenous injection of adenovirus expressing TWEAK or recombinant TWEAK protein, we find that TWEAK induces a progressive dilated cardiomyopathy with impaired contractile function in mice. Moreover, TWEAK treatment is associated with decreased expression of peroxisome proliferator-activated receptor gamma coactivator-1α (PGC1α) and genes required for mitochondrial oxidative phosphorylation, which precede the onset of cardiac dysfunction. TWEAK-induced downregulation of PGC1α requires expression of its cell surface receptor, fibroblast growth factor-inducible 14 (Fn14). We further find that TWEAK downregulates PGC1α gene expression via the TNF receptor-associated factor 2 (TRAF2) and NFκB signaling pathways. Maintaining PGC1α levels through adenoviral-mediated gene expression is sufficient to protect against TWEAK-induced cardiomyocyte dysfunction.
Collectively, our data suggest that TWEAK induces cardiac dysfunction via downregulation of PGC1α, through FN14-TRAF2-NFκB-dependent signaling. Selective targeting of the FN14-TRAF2-NFκB-dependent signaling pathway or augmenting PGC1α levels may serve as novel therapeutic strategies for cardiomyopathy and heart failure.
Tumor necrosis factor (TNF)-like weak inducer of apoptosis (TWEAK) is a recently identified member of the TNF superfamily cytokines
In the present study, we find that TWEAK-induced cardiac pathology in mice is associated with mitochondrial dysfunction. We further identify peroxisome proliferator-activated receptor gamma coactivator 1α (PGC1α), an essential regulator of mitochondrial biogenesis and energy metabolism
C57BL/6J male mice (8 weeks old) were purchased from Jackson Laboratory. Fn14 knockout mice (FN14 KO) and wild-type (WT) counterparts have been previously reported and were generated on the 129 strain background and backcrossed onto the C57BL/6 strain, as described
Longitudinal cardiac function and chamber structure was assessed in conscious mice through serial non-invasive transthoracic echocardiography, using a Vevo2100 system (VisualSonics) as previously described
Adult rat ventricular cardiomyocytes were isolated from male Wistar rats (Charles River Laboratories) using collagenase perfusion and cell dissociation, as described previously
Adult rat cardiomyocytes were infected with adenovirus expressing PGC1α (Ad-PGC1α) or Ad-GFP
Cardiomyocytes were cultured in petri-dishes (35 mm×15 mm) or 24-well plates overnight and subsequently treated with 100 ng/ml IgG or rTWEAK for 24 hours, followed by incubation with 10 nM TMRE for 30 minutes. After washing with pre-warmed PBS, culture dishes were placed in a LSM700 confocal microscopy equipped with temperature-controlled chamber for live cell imaging. TMRE fluorescence was assessed by excitation at 555 nm. On average, 5–7 pictures were taken from each dish. Cardiomyocytes were hand-traced and quantified using SigmaPro software.
Membrane protein was isolated using a subcellular protein fractionation kit (Thermo Scientific). Cardiomyocytes were detached from culture dishes using a cell scraper and harvested into ice-cold PBS. Cells were centrifuged at 500 ×g for 5 minutes at 4°C, washed with ice-cold PBS, and centrifuged again at 500 ×g for 2 minutes. The cell pellet was suspended in cytoplasmic extraction buffer containing protease inhibitors and incubated at 4°C for 10 minutes with gentle mixing. After centrifugation at 500 ×g for 5 minutes, the separated supernatant contained the cytoplasmic fraction. The remaining pellet was resuspended in membrane extraction buffer containing protease inhibitors, incubated at 4°C for 10 minutes with gentle mixing, and then centrifuged at 3,000 ×g for 5 minutes. The resulting supernatant fraction provided the membrane proteins.
RNA was extracted using Trizol reagent (Invitrogen). Genomic DNA was removed by using Turbo-DNA free kit (Ambion). iScriptTM cDNA Synthesis Kit (Bio-Rad) was used for cDNA synthesis and quantitative RT-PCR was performed using a CFX96 real-time PCR system (Bio-Rad). Primers used for qPCR include: PGC1α forward
For protein isolation, heart tissues were homogenized or cultured cardiomyocytes were harvested in cell lysis buffer (Cell Signaling). Equal amounts of proteins were used for SDS/PAGE and electrotransferred to a PVDF membrane (Millipore). The membranes were treated with Odyssey Blocking buffer (Li-Cor) for 1 hour and incubated with appropriate primary antibodies overnight at 4°C. After washing, blots were incubated with corresponding secondary antibodies conjugated with IRDye 800CW or IRDye 680LT. Blots were then scanned and analyzed using the Odyssey infrared scanner (Li-Cor).
Low glucose DMEM and Laminin were purchased from Invitrogen. SC-514 and antibody against PGC1α were obtained from Calbiochem. Antibodies against p-p65, p-IκBα and IκBα were obtained from Cell Signaling Technology. Antibody against GAPDH was obtained from R&D Systems. Ad-GFP and Ad-TWEAK were provided by Biogen Idec Inc.
All data are expressed as mean ± SEM. Student’s
To elucidate the underlying mechanisms by which TWEAK induces cardiac dysfunction, we utilized our established mouse model of TWEAK induced cardiomyopathy. Increased circulating level of TWEAK was achieved via adenovirus-mediated gene delivery as we described previously
(A) Fractional shortening (FS%, calculated as the difference in chamber dimension between diastole and systole over the chamber dimension in diastole) and (B) LV ventricular diastolic chamber dimension (LVID) were determined in mice by transthoracic echocardiography 1-day prior as well as 1, 2, and 3-week after intravenously delivery of control, Ad-GFP or Ad-TWEAK. (C) Representative M-Mode echocardiographic images at one week and three weeks post Ad-GFP and Ad-TWEAK injection in mice. Real time PCR analysis for expression of (D) PGClα and (E) OXPHOS genes in cardiac samples at 3-weeks post-Ad-GFP or Ad-TWEAK injection. Real time PCR analysis for (F) PGClα and (G) OXPHOS genes in cardiac samples at 1 week (prior to the development of heart failure) post-Ad-GFP or Ad-TWEAK injection. All real time PCR data were normalized to β-actin and presented relative to the Ad-GFP group. * p<0.05 vs. Ad-GFP, # p<0.05 vs. Ad-TWEAK-injected mice at 1 week time point, N = 5 for each group.
To exclude the potential effects of adenoviral delivery and gene expression in the Ad-TWEAK mouse model, we developed a novel mouse model with tail vein delivery of recombinant TWEAK (rTWEAK). As shown in
(A) Fractional shortening (FS%) and (B) Left ventricular diastolic chamber dimension (LVID), as determined by transthoracic echocardiography in mice one week after intravenous delivery of IgG or rTWEAK. (C) Representative M-Mode echocardiographic images before and following IgG or rTWEAK injection in mice. Real time PCR analysis of heart samples for expression of (D) PGClα and (E) OXPHOS genes, normalized to β-actin and presented relative to the IgG group. (F) Mitochondria membrane potential in IgG and rTWEAK treated cardiomyocytes. The membrane potential was measured by TMRE staining. The fluorescent intensities of TMRE were normalized to the total number of cardiomyocytes in the respective fields. * p<0.05 vs. IgG, N = 3 for each group.
We have previously identified Fn14 as a critical receptor that mediates TWEAK-induced cardiac dilatation and dysfunction
Cardiomyocytes isolated from WT and Fn14 KO mice were treated with 100 ng/ml IgG or rTWEAK, and then harvested for determination of PGC1α expression and TRAF2 membrane translocation. (A) Real time PCR analysis for PGClα expression post IgG and rTWEAK treatment, normalized to β-actin and presented relative to the IgG treated WT group. (B) Real time PCR analysis for gene expression profiling of TRAF family proteins, expressed as the mRNA ratio of TRAF/β-actin. (C) Immunoblot analysis for TRAF2 membrane translocation after IgG and rTWEAK treatment for 10 min. Lower panel shows Ponceau S staining of PVDF membranes that was used for evaluation of protein loading amount. (D) Bar-graph shows the relative levels of TRAF2 by densitomitric analysis with IgG treated WT group as 1-fold. * p<0.05 vs. IgG treatment in WT group, # p<0.05 vs. rTWEAK treatment in WT group. (E) Silencing of TRAF2 using shRNA (sh-TRAF2) prevents TWEAK-induced downregulation of PGC1α. * p<0.05 vs. IgG in sh-Scramble group, and # p<0.05 vs. rTWEAK in sh-Scramble group.
Activation of Fn14 may result in membrane translocation of TRAF (tumor necrosis factor receptor associated factor) leading to activation of downstream signaling pathways
Treatment of cardiomyocytes with rTWEAK, but not IgG, further induced downstream mediators, including IκBα phosphorylation, degradation, and resynthesis, accompanied by the phosphorylation of NFκB p65, which occurred as early as 10 minutes and persisted for several hours (
(A) Immunoblots of phospho-p65, phospho-IκBα, total-IκBα and GAPDH in isolated cardiomyocytes incubated with 100 ng/ml IgG or rTWEAK at designated time points. (B) Inhibition of NFκB activation with SC-514 (25µM) abolished TWEAK-mediated downregulation of PGC1α expression. * p<0.05 vs. IgG and # p<0.05 vs. rTWEAK in the absence of SC-514.
To test our hypothesis that the downregulation of PGC1α plays a causal role in TWEAK-induced cardiac dysfunction, adenoviral-mediated PGC1α expression was used in cultured cardiomyocytes for 24 hours prior to treatment with IgG or rTWEAK. Upon rTWEAK treatment, PGC1α expression was downregulated in Ad-GFP-infected cells but maintained at normal levels in Ad-PGC1α-infected cells (
Isolated cardiomyocytes were infected with adenovirus (MOI of 10) expressing GFP or PGC1α for 24 hours prior to treatment with IgG or rTWEAK for 48 hours. (A) Western blot analysis of PGC1α expression, normalized with GAPDH. N = 3 independent biological replicates group. (B) Isolated cardiomyocyte function (%CS) was determined using edge detection method. (C) Representative tracings of single cell shortening at indicated conditions. (D) Time to 90% relaxation in isolated cardiomyocytes. Cellular function was assessed in three independent biological replicates and data from 8–12 cells was averaged as N = 1 for a given experiment. * p<0.05 vs. IgG and # p<0.05 vs. rTWEAK in Ad-GFP group.
In the present study, we reveal that the cytokine TWEAK downregulates PGC1α and mitochondrial OXPHOS gene expression in cardiomyocytes, which contributes to TWEAK-induced cardiac dysfunction. Moreover, we find that TWEAK regulates PGC1α expression via Fn14/TRAF2/NFκB-dependent signaling pathways (
Normal PGC1 levels maintain OXPHOS gene expression and mitochodrial function to support cardiomyocyte contractility (left panel). TWEAK downregulates PGC1 levels through FN14-TRAF2-NFκB-dependent signaling, leading to impaired OXPHOS gene expression and mitochondrial dysfunction, and reduced cardiomyocyte contractility (right panel).
PGC1α is a transcriptional coactivator that is preferentially expressed in tissues with high-energy demand and greater mitochondrial abundance, including the heart
We have previously found that TWEAK-induced cardiac dysfunction requires expression of Fn14
A universal characteristic of TWEAK signaling through Fn14 is the activation of NFκB signaling pathway with modulation of numerous downstream target genes
In summary, our data suggest that Fn14-TRAF2-NFκB-dependent suppression of PGC1α expression plays a crucial role in TWEAK-induced cardiac dysfunction. Modulation of PGC1α expression or antagonism of Fn14-TRAF2-NFκB may serve as candidate therapeutic targets in preventing TWEAK-induced heart failure.
The authors wish to thank Mr. Soeun Ngoy at Brigham and Women’s Hospital Cardiovascular Physiology Core for his assistance in animal work.