Enhanced Bioactive Myocardial Transforming Growth Factor-β in Advanced Human Heart Failure

Abstract

Background: Transforming growth factor (TGF)-β activation is known to play a central role in progressive ventricular remodeling in advanced heart failure in animal models, but there has been no direct evidence of increased TGF-β activity in the myocardium of patients with advanced human heart failure.

Methods and Results: Using a recently developed bioassay that measures TGF-β bioactivity rather than TGF-β abundance, we measured bioactive TGF-β in human myocardium from control non-failing donors (NF), and patients with ischemic cardiomyopathy (ICM) and dilated cardiomyopathy (DCM). Both free and total soluble TGF-β were significantly increased in ICM and DCM compared with NF. Free TGF-β had an excellent correlation with phosphorylated Smad2 (R²=0.55, P<0.0001), a downstream marker of TGF-β signaling. Collagen type I and type III were significantly upregulated in DCM compared with NF, consistent with histological evidence of myocardial fibrosis. Expression of fibulin-2, a positive modulator of TGF-β, was significantly increased in DCM compared with NF, and the free TGF-β level was correlated with fibulin-2 mRNA (R²=0.24, P<0.006).

Conclusions: Although both free and total soluble TGF-β are significantly increased in ICM and DCM compared with NF, the superior correlation of free TGF-β with downstream signaling suggests that this is the most functionally relevant form. The present findings suggest that sustained TGF-β activation in both ICM and DCM contributes to excess myocardial fibrosis. (Circ J 2014; 78: 2711–2718)

Transforming growth factor (TGF)-β is a multi-functional peptide growth factor that has a vital role in the regulation of cell proliferation, differentiation, inflammation, angiogenesis, and wound healing, as recently reviewed.^1,2 TGF-β is also known to mediate multiple pathological conditions including cancer progression, chronic inflammatory disorders, fibrosis, and heart failure.³ The role of TGF-β in pathogenesis of cardiac remodeling and heart failure has been studied extensively in the animal models,^4,5 but its functional relevance to human hearts remains unproven. Although studies have demonstrated increased TGF-β expression in hearts from patients with dilated cardiomyopathy (DCM),⁶ hypertrophic cardiomyopathy,⁷ and aortic stenosis,⁸ the significance of the findings remains unclear due to the absence of data demonstrating bioactive TGF-β in the remodeled heart.⁵ TGF-β signaling in human hearts has been indirectly implicated by studies correlating myocardial angiotensin (Ang) II and Smad protein isoforms that signal downstream of TGF-β, but direct measurement of TGF-β bioactivity in the human myocardium has not been reported to date.

Recently, we developed method of measuring bioactive TGF-β in solid organ tissues by a bioassay using mink lung epithelial cells (MLEC).⁹ The principle of this bioassay is based upon the dose-responsive activation of plasminogen activator inhibitor (PAI)-1 promoter by biologically active TGF-β.¹⁰ In the tissues, TGF-β exists either as an extracellular matrix (ECM)-bound insoluble form, soluble forms consisting of a carrier-bound form either as large latent complex or small latent complex, or a free TGF-β. Free TGF-β level was determined directly by PAI-1 promoter response to tissue protein lysate, whereas total soluble TGF-β level, a combination of free and carrier-bound soluble TGF-β, was measured after acid treatment by which free TGF-β was dissociated from its carrier proteins, latency-associated peptide (LAP) and/or latent TGF-β binding proteins (LTBP).¹¹ This method was reported to be highly effective in measuring all TGF-β isotypes, TGF-β1, 2, and 3.¹⁰ Although initially reported as a sensitive and specific assay for TGF-β in soluble fluids,¹² we have successfully adapted the method to allow quantification of TGF-β bioactivity in mouse tissues.⁹ The detected myocardial bioactive TGF-β level, of both free and total soluble forms, on bioassay was markedly higher than that obtained on ELISA.⁹

In the present study we utilized this methodology to demonstrate, for the first time, that myocardial TGF-β activity is significantly increased in severely failing human hearts compared with non-failing human control hearts. Building on recent studies indicating that the ECM protein fibulin-2 contributes to the progression of ventricular remodeling by modulating myocardial TGF-β signaling in mouse models,^13,14 we have also investigated the relationship of myocardial fibulin-2 in relation to myocardial TGF-β activity in the human heart.

Methods

Human Myocardium

Human left ventricular (LV) myocardium was obtained from the Heart Tissue Bank at the University of Pennsylvania, approved by the Institutional Review Board. Non-failing hearts (NF; n=10) were procured from brain-dead organ donors with no history of clinical heart failure, and failing heart tissue was obtained at the time of heart transplantation from subjects with ischemic cardiomyopathy (ICM; n=10) or DCM (n=10). In all cases, prospective informed consent was obtained for research use of donated heart tissue from patients or next of kin (for organ donors). All hearts were arrested in situ with antegrade perfusion-based cardioplegia and kept on wet ice during transportation and processing. Transmural samples of the LV free wall were placed in labeled cryovials and snap frozen in liquid nitrogen within 4 h of cardiectomy.

Real-Time Reverse Transcription (RT)-Polymerase Chain Reaction (PCR)

Total RNA from heart tissues was isolated using the Totally RNA kit (Ambion, Austin, TX, USA). Two μg of RNA was reverse transcribed to generate first strand cDNA using Bio-Rad RT Kit (Bio-Rad Life Science, Hercules, CA, USA). The second strand cDNA synthesis and PCR amplification were carried out using the Taqman Gene Expression Master Mix (Life Technologies, Carlsbad, CA, USA) with specific primers including TGF-β1, collagen types I and III (Col I and III), and fibulin-2 (Table S1). GAPDH was used as an internal control.

Histology

The myocardial tissues were fixed with formalin/phosphate-buffered saline for routine histology. Paraffin-embedded specimens were sectioned at 8 μm thickness for Masson’s trichrome staining. Microscopy images were obtained under 10× objective and analyzed using Image J (http://rsbweb.nih.gov/ij/) for myocardial fibrosis. Area of fibrosis was expressed as percentage over entire myocardial area.

Immunoblotting

Myocardial tissue was homogenized with RIPA lysis buffer (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and protein concentration of the supernatants was determined by Pierce BCA Assays (Pierce Biotechnology, Rockford, IL, USA). Equal amounts of protein (100 μg) were electrophoresed on a 10% sodium dodecyl sulfate-polyacrylamide gel, and then electrophoretically transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). After blocking, the membranes were incubated at 4℃ with primary antibodies on a rocking platform overnight. Antibodies used were phospho-Smad2 (pSmad2), Smad2 (Cell Signaling Technology, Danvers, MA, USA), TGF-β1, β-Actin (Santa Cruz Biotechnology) and fibulin-2.¹⁵ Blots were incubated with horseradish-labeled secondary antibodies (anti-mouse IgG or anti-rabbit IgG) for 1 h at room temperature, and signals were detected using the Pierce ECL Western Blotting Substrate (Pierce Biotechnology). Membranes were exposed to Hyperfilm (GE Health Care, Pewaukee, WI, USA) and signal intensities were analyzed with Image J.

TGF-β Bioassay

Detailed methods for the TGF-β bioassay have been described elsewhere.^9,10,16 Briefly, approximately 30–50 mg of human myocardial tissues was minced and sonicated in 500 μl of lysis buffer (50 mmol/L Tris-HCl pH 7.5), and the biologically active TGF-β was directly measured in the clear supernatant. Protein was quantified using the BCA method. To assess the amount of total TGF-β, acid activation was performed to isolate free TGF-β molecules from latent complex.¹⁶ MLEC (ATCC, Manassas, VA, USA) transfected with PAI-1 promoter/luciferase construct (generously provided by Dr D. Rifkin, New York University) by FuGENE-6 transfection reagent (Roche Diagnostic, Indianapolis, IN, USA) was used for the bioassay. MLEC (4×10⁴ cells) plated into 96-well plates were incubated with the protein lysate for 20 h at 37℃. Then, MLEC were harvested by adding 30 μl 1×passive lysis buffer (Promega, Madison, WI, USA). The luciferase assay reagent (100 μl: Promega) was added to each well by injector and the relative luciferase unit was read by a 2030 Multilabel Reader (Perkin Elmer, Waltham, MA, USA). The tissue activity of TGF-β was reported as ng/mg total protein.

Statistical Analysis

Data are reported as mean±SD. Multiple comparisons were performed with 1-way analysis of variance (ANOVA), followed by Dunnett’s multiple comparison test to assess the significance. For correlation analysis, linear regression analysis was used. P<0.05 was regarded as statistically significant.

Results

Patients

The demographic patient features are listed in Table 1. Mean patient age was similar among the 3 groups. The patients were predominantly male in the ICM and DCM groups, and the male and female sexes were balanced in the NF group. The patients with end-stage heart failure, both ICM and DCM, were more overweight than NF patients. The hearts of both ICM and DCM were significantly larger in weight and heart/body surface area (BSA) ratio. Echocardiogram showed increased chamber size (LV end diastolic diameter) and significantly diminished LV systolic performance (% ejection fraction) in ICM and DCM compared with NF.

Table 1. Patient Characteristics

Sex	Age (years)	Race	Weight (kg)	Height (cm)	Heart weight (g)	Heart/BSA (g/m2)	CHF (months)	LVEDD (cm)	LVEF (%)
Non-failure heart
M	66	L	65	163	446	260	NA	4.9	65
M	18	C	58	183	275	160	NA		70
F	54	C	70	167	297	165	NA		65
M	61	L	84	178	397	195	NA		58
M	33	C	77	178	360	185	NA		65
F	60	C	83	163	465	240	NA		60
M	44	L	92	163	434	213	NA		65
F	80	C	73	168	367	199	NA	4.5	77
F	20	AA	57	155	314	200	NA	2.8	78
F	55	C	57	168	343	210	NA		70
Mean±SD	49.1±20.2		71.6±12.4	169±9	370±65	203±31		4.1±1.1	67.3±6.5
Ischemic cardiomyopathy
M	58	C	100	180	484	216	33	7.8	15
M	58	C	93	180	967	448	180		15
M	49	C	83	183	600	292	9	7.2	25
M	58	C	104	180	585	266	84	7.5	10
M	67	C	90	163	867	429	72	9.3	12
M	58	C	102	188	551	239	96
M	54	C	72	174	617	331	4	8	10
M	51	C	127	193	680	261	60	5	55
M	64	C	116	178	612	256	25	7.8	25
M	56	AA	106	185	610	261	96	7.3	25
Mean±SD	57.3±5.4		99.3±15.8***	180±8*	657±148***	299±80**	66±53	7.5±1.2*	21.3±14.1***
Dilated cardiomyopathy
F	58	C	64	150	592	363	88	7.1	10
M	67	C	99	178	614	278	254	6.7	18
M	53	AA	123	180	697	281			12
M	57	C	95	175	584	272	192	6.5	15
M	30	AA	108	182	710	304	98		10
M	48	C	82	180	703	347	69	8.5	15
M	51	C	97	173	517	239	84	7.7	10
M	44	AA			491			6.1	20
M	51	C	66		529		180	7.8	5
M	62	C	97	183	654	295	65	5.8	25
Mean±SD	52.1±10.3		92.3±19.0*	175±11	609±81***	297±40***	129±70†	7.0±0.9**	14±5.8***

*P<0.05, **P<0.005, and ***P<0.0005 compared with NF. ^†P<0.05 compared with ischemic cardiomyopathy. AA, African-American; BSA, body surface area; C, Caucasian; CHF, congestive heart failure; L, Latino; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; NA, not applicable.

Myocardium Profile

The mRNA levels of TGF-β1, Col I and Col III, LTBP-1, -2, -3, and -4, and fibulin-2 were examined on quantitative RT-PCR (qRT-PCR; Figure 1). Col I, Col III, Col I/Col III ratio, all LTBP, and fibulin-2 were all significantly upregulated in DCM compared with NF. Typical histological features of myocardial fibrosis are presented in Figure 2A. Both ICM and DCM had markedly increased myocardial fibrosis compared with NF. Neither TGF-β1 nor fibulin-2 protein levels were significantly different among NF, ICM, and DCM on immunoblot (Figure 3). Although there was no significant change in TGF-β1 protein level among the 3 groups, there was a significant increase in Smad2 activation represented by pSmad2/Smad2 in both ICM and DCM (Figure 3C), suggesting upregulation of myocardial TGF-β signaling in advanced heart failure.

Figure 1.

Myocardial mRNA transforming growth factor (TGF)-β1, collagen (Col) I, Col III, Fibulin-2, and latent TGF-β binding proteins (LTBP)-1–4 on quantitative reverse transcription-polymerase chain reaction. Data given as mean±SD. Although TGF-β1 mRNA was not significantly different among the 3 groups, both Col I and Col III mRNA were significantly increased in DCM compared with NF. Col I/Col III ratio, LTBP-1–4, and fibulin-2 were significantly increased in DCM compared with NF. DCM, dilated cardiomyopathy (n=10); ICM, ischemic cardiomyopathy (n=10); NF, non-failing donor (n=10). *P<0.05, **P<0.01 compared with NF. ^#P<0.05 compared with ICM.

Figure 2.

Myocardial fibrosis. (A) Masson’s trichrome staining of myocardium in NF, ICM, and DCM at low magnification Not only was interstitial fibrosis (blue) increased in ICM and DCM, but also myocardial cross-sectional areas were markedly increased in ICM and DCM compared with NF. Magnification bar, 100 μm. (B) Area of fibrosis was measured morphometrically in selected cases (n=4 from each group) as a percentage of the entire myocardium. Data given as mean±SD. Both ICM and DCM had significant increase in myocardial fibrosis compared with NF (*P<0.05 compared with NF). DCM, dilated cardiomyopathy; ICM, ischemic cardiomyopathy; NF, non-failing donor.

Figure 3.

Myocardial protein expression and Smad2 signaling. (A) Myocardial protein levels of transforming growth factor (TGF)-β1, pSmad2, Smad2, fibulin-2, and β-actin on immunoblot. There was a significant increase of pSmad2 in ICM and DCM compared with in NF. (C) pSmad2/Smad2 ratio was significantly increased in ICM and DCM compared with NF (**P<0.001 compared with NF). (B,D) There was no obvious increase in protein levels of TGF-β1 or fibulin-2 on immunoblot. DCM, dilated cardiomyopathy (n=10); ICM, ischemic cardiomyopathy (n=10); NF, non-failing donor (n=10). Data given as mean±SD.

Myocardial Bioactive TGF-β

Free and total soluble TGF-β in human myocardium, as measured by the PAI-1/luciferase bioassay using MLEC, is presented in Figure 4. Compared with NF myocardium, free myocardial TGF-β activity in ICM and DCM was increased 4-fold and 5-fold, respectively. Likewise, total myocardial TGF-β activity was increased in ICM and DCM by more than 2-fold and 3-fold, respectively.

Figure 4.

Bioactive transforming growth factor (TGF)-β, free (A) and total (B), in human myocardium, measured on mink lung epithelial cell bioassay with plasminogen activator inhibitor-1 promoter/luciferase construct. Free myocardial TGF-β was significantly increased in ICM and DCM compared with NF (NF, 0.33±0.24 ng/mg; ICM, 1.31±0.73 ng/mg; and DCM, 1.67±0.85 ng/mg). Total soluble myocardial TGF-β was also upregulated in ICM and DCM (NF, 1.15±0.44 ng/mg; ICM, 2.62±1.13 ng/mg; DCM, 3.62±1.46 ng/mg). *P<0.01 compared with NF, **P<0.001 compared with NF. DCM, dilated cardiomyopathy; ICM, ischemic cardiomyopathy; NF, non-failing donor.

Further supporting the functional validity of the TGF-β activity assay is the correlation of Smad2 activation (pSmad2/Smad2 ratio; Figures 5A,B). These data further indicate that free rather than total soluble TGF-β is the best indicator of TGF-β function at the tissue level (P<0.0001). Likewise, the correlations in Figures 5C,D indicate a functional link between fibulin-2 mRNA and TGF-β activity, with the correlation again stronger for free rather than total TGF-β activity (P=0.0055). In this case, however, the lack of significant correlation among NF cases suggests there are other important contributors to the regulation of TGF-β activity.

Figure 5.

Correlations between bioactive transforming growth factor (TGF)-β and (A,B) Smad2 Phosphorylation and (C,D) Fibulin-2 mRNA. (Upper) Correlation between (A) free TGF-β and pSmad2/Smad2 and between (B) total soluble TGF-β and pSmad2/Smad2 in NF, ICM, and DCM. Both free and total TGF-β had an excellent correlation with Smad2 activation in all 3 groups (solid line: P<0.0001 for free TGF-β and P=0.011 for total TGF-β). (Lower) Correlation between (C) free TGF-β and fibulin-2 mRNA and between (D) total TGF-β and fibulin-2 mRNA in NF, ICM, and DCM. Both free and total TGF-β had good correlation with fibuiln-2 mRNA in all 3 groups (solid line: P=0.0055 for free TGF-β and P=0.0252 for total TGF-β). DCM, dilated cardiomyopathy; ICM, ischemic cardiomyopathy; NF, non-failing donor.

Discussion

Although it has been postulated by multiple investigators that excessive TGF-β mediates the development of cardiac remodeling and heart failure,^4,5,17 there has been no direct evidence of increased TGF-β activity in the myocardium from patients with heart failure. In this study, we have demonstrated that both active and total soluble TGF-β are indeed significantly elevated in end-stage heart failure (ICM and DCM) compared with NF. Both Col I and Col III, the most abundant types of collagen in the cardiac ECM, were significantly upregulated in DCM compared with NF. The Col I/Col III ratio was also significantly increased in DCM (Figure 1), as previously reported.^6,18 All LTBP were significantly upregulated in DCM, consistent with the recent study in which LTBP-2 was upregulated in human heart failure in both mRNA and protein levels.¹⁹ In the context of increased myocardial fibrosis (Figure 2), the correlation of increased TGF-β activity with increased pSmad2/Smad2 ratio (Figure 3) supports a pathogenic role for sustained increase in TGF-β activity in the failing human myocardium. Preclinical studies demonstrating the beneficial effects of sustained TGF-β inhibition in rodents^20–24 further support a pathogenic role and its potential therapeutic significance.

Here we demonstrate that sustained TGF-β activation was noted in advanced human heart failure regardless of the etiology, suggesting a common pathological path for this chronically progressive disorder. Interestingly, neither TGF-β expression based on total mRNA (via qRT-PCR) nor protein (via immunoblot) indicated upregulation of TGF-β in groups ultimately found to have increased TGF-β activity and signaling. This suggests that post-translational activation of TGF-β is an important pathological mechanism in the human myocardium. This finding also suggests that direct measurement of tissue TGF-β bioactivity by the method we used may have a wider role for elucidating TGF-β biology in pathological conditions.

Previously, TGF-β levels, both mRNA and protein, were shown to be upregulated in human hypertrophic cardiomyopathy,^25,26 and activation of myocardial TGF-β signaling was demonstrated as a pivotal mechanism for increased fibrosis in hypertrophic cardiomyopathy in the mouse model.²⁷ Sustained TGF-β activation was shown to exacerbate myocardial fibrosis, ventricular dysfunction, and ventricular dilatation following myocardial infarction in the mouse model,²⁸ but direct involvement of TGF-β in human ICM has not been reported. Similarly in the mouse model, TGF-β played a principal role in inducing ventricular remodeling following pressure overload.²² In human DCM, in contrast to the present findings of no significant increase in TGF-β mRNA or protein abundance in severely failing hearts, 1 previous study reported increased TGF-β1 and TGF-β2 expression based on qRT-PCR.⁶ That prior study, however, included patients with relatively mild LV dysfunction, obtained samples from the right ventricle, and measured neither protein abundance nor activity of TGF-β. Thus, the present measurement of myocardial TGF-β activity dissociated from measures of mRNA and protein abundance is both novel and not in conflict with any prior studies.

The present study indicates 3 different levels of TGF-β latency: One is the inactive TGF-β stored abundantly in the insoluble ECM reservoir,²⁹ and the others are soluble TGF-β bound to LAP and/or LTBP (=carrier-bound soluble TGF-β) that have been cleaved from insoluble ECM. These carrier-bound TGF-β forms are dissociated by acid treatment. The combination of this dissociated carrier-bound soluble TGF-β and free TGF-β corresponds to the present measurement of total TGF-β activity that can be assayed only after acid activation. Although we observed an increase in this total TGF-β activity in both ICM and DCM, the proportional increase in activity was greater for the free TGF-β, and free TGF-β activity had much better correlation with Smad2 phosphorylation than total TGF-β activity (Figure 5). The fact that we did not detect any statistically significant differences in TGF-β level among the 3 groups using a commercially available ELISA kit (Table 2) attests to the importance of using the PAI-1/luciferase bioassay with MLEC to determine tissue TGF-β activity.

Table 2. Myocardial TGF-β Protein in Selected Cases (n=6)

	TGF-β (ng/mg)
	Free	Total
Bioassay
NF	0.41±0.18	1.17±0.43
ICM	1.74±0.50**	2.91±1.30*
DCM	2.09±0.47**	3.85±1.03**
ELISA
NF	0.48±0.44	0.74±0.22
ICM	0.41±0.62	1.35±1.13
DCM	0.41±0.30	1.85±0.86

Data given as mean±SD. *P<0.05 compared with NF, **P<0.001 compared with NF. ELISA levels were generally significantly lower than those obtained on bioassay. No statistically significant differences were seen among the 3 groups on ELISA. DCM, dilated cardiomyopathy; ICM, ischemic cardiomyopathy; NF, non-failing donor; TGF, transforming growth factor.

We recently showed that the ECM protein fibulin-2 modulated TGF-β activation and ventricular remodeling after myocardial infarction¹³ and continuous Ang II infusion¹⁴ in mice. Fibulin-2 null mice had significantly better survival after myocardial infarction than wild-type mice due to preserved LV function in conjunction with attenuated TGF-β activation,¹³ and failed to show Ang II-induced myocardial hypertrophy and fibrosis secondary to attenuation of TGF-β activation.¹⁴ Both studies suggested that upregulation of fibulin-2 and subsequent TGF-β activation are critical processes in the pathogenesis of ventricular remodeling in the mouse model and that the decrease in fibulin-2 is likely to prevent progression of ventricular remodeling and heart failure. Separately, fibulin-2 was recently identified as a principal downstream target mediator in microRNA (miR-1) treatment that prevented pressure overload-induced pathological remodeling in conjunction with attenuated TGF-β signaling.³⁰ They also showed upregulation of fibulin-2 protein in a small number of patients with advanced heart failure. In the present study, free TGF-β activity was correlated with the abundance of mRNA for fibulin-2, particularly in the hearts from patients with DCM. The present study showed statistically significant correlation between free TGF-β level and fibulin-2 mRNA expression (Figure 5C; P=0.0055). Collectively, it is plausible that fibulin-2 contributes, in part, to the pathogenesis of ventricular remodeling also in human heart.

Study Limitations

First, although myocardial TGF-β bioactivity and activation of its downstream Smad signaling pathways were significantly increased and correlated in ICM and DCM, profiling of human myocardium cannot confirm the causal relationship between the increased myocardial TGF-β activity and the pathogenesis of heart failure. The previous animal studies, however, strongly support the notion that TGF-β plays a central role in the pathogenesis of progressive ventricular remodeling.^20–24 Second, it is important to understand that this bioassay exclusively measures the amount of biologically active TGF-β in the tissue that is in a soluble form. Thus, abundant insoluble TGF-β incorporated in the ECM reservoir, a major source of soluble TGF-β, is not included by this assay. It is also not clearly understood how much of carrier-bound TGF-β can be dissociated by acid treatment. Last, the number of tissue samples in this study was relatively small (n=10 for each group) and hearts from brain-dead organ donors are not truly normal healthy myocardium. Therefore, it is possible that a larger sample size or different control hearts would enable identification of additional differences between NF and ICM/DCM hearts that were not detected here.

Conclusions

By direct measurement using the MLEC bioassay, we have demonstrated that tissue TGF-β activity is significantly higher in severely failing human hearts than in NF hearts. The fact that increased free TGF-β activity is well-correlated with increased phosphorylation of Smad2 supports, but cannot prove, a pathogenic role for sustained increase in TGF-β activity in the failing human myocardium. Combined with preclinical studies demonstrating the beneficial effects of sustained TGF-β inhibition and a large body of literature linking TGF-β and myocardial fibrosis, these findings should motivate proof-of-concept trials examining whether TGF-β antagonists have a useful role in clinical settings prone to pathological myocardial fibrosis.

Acknowledgments

This work was supported by the U.S. National Institutes of Health (P20RR020173-01 and P20GM103446-12 to T.T. and R01HL105993 and R01HL089847 to K.B.M.).

Disclosures

Conflict of Interest: None.

Supplementary Files

Supplementary File 1

Table S1. Primers for quantitative reverse transcription-polymerase chain reaction

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-14-0511

References

• 1.    Verrecchia  F,  Mauviel  A. Control of connective tissue gene expression by TGF beta: Role of Smad proteins in fibrosis. Curr Rheumatol Rep 2002; 4: 143–149.
• 2.    Leask  A,  Abraham  DJ. TGF-beta signaling and the fibrotic response. FASEB J 2004; 18: 816–827.
• 3.    Blobe  GC,  Schiemann  WP,  Lodish  HF. Role of transforming growth factor beta in human disease. N Engl J Med 2000; 342: 1350–1358.
• 4.    Khan  R,  Sheppard  R. Fibrosis in heart disease: Understanding the role of transforming growth factor-beta in cardiomyopathy, valvular disease and arrhythmia. Immunology 2006; 118: 10–24.
• 5.    Dobaczewski  M,  Chen  W,  Frangogiannis  NG. Transforming growth factor (TGF)-beta signaling in cardiac remodeling. J Mol Cell Cardiol 2011; 51: 600–606.
• 6.    Pauschinger  M,  Knopf  D,  Petschauer  S,  Doerner  A,  Poller  W,  Schwimmbeck  PL, et al. Dilated cardiomyopathy is associated with significant changes in collagen type I/III ratio. Circulation 1999; 99: 2750–2756.
• 7.    Li  RK,  Li  G,  Mickle  DA,  Weisel  RD,  Merante  F,  Luss  H, et al. Overexpression of transforming growth factor-beta1 and insulin-like growth factor-I in patients with idiopathic hypertrophic cardiomyopathy. Circulation 1997; 96: 874–881.
• 8.    Fielitz  J,  Hein  S,  Mitrovic  V,  Pregla  R,  Zurbrugg  HR,  Warnecke  C, et al. Activation of the cardiac renin-angiotensin system and increased myocardial collagen expression in human aortic valve disease. J Am Coll Cardiol 2001; 37: 1443–1449.
• 9.    Khan  SA,  Joyce  J,  Tsuda  T. Quantification of active and total transforming growth factor-beta levels in serum and solid organ tissues by bioassay. BMC Res Notes 2012; 5: 636.
• 10.    Abe  M,  Harpel  JG,  Metz  CN,  Nunes  I,  Loskutoff  DJ,  Rifkin  DB. An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal Biochem 1994; 216: 276–284.
• 11.    Annes  JP,  Munger  JS,  Rifkin  DB. Making sense of latent TGFbeta activation. J Cell Sci 2003; 116(Pt2): 217–224.
• 12.    van Waarde  MA,  van Assen  AJ,  Kampinga  HH,  Konings  AW,  Vujaskovic  Z. Quantification of transforming growth factor-beta in biological material using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal Biochem 1997; 247: 45–51.
• 13.    Tsuda  T,  Wu  J,  Gao  E,  Joyce  J,  Markova  D,  Dong  H, et al. Loss of fibulin-2 protects against progressive ventricular dysfunction after myocardial infarction. J Mol Cell Cardiol 2012; 52: 273–282.
• 14.    Zhang  H,  Wu  J,  Dong  H,  Khan  SA,  Chu  ML,  Tsuda  T. Fibulin-2 deficiency attenuates angiotensin II-induced cardiac hypertrophy by reducing transforming growth factor-beta signalling. Clin Sci (Lond) 2014; 126: 275–288.
• 15.    Pan  TC,  Zhang  RZ,  Fassler  R,  Timpl  R,  Chu  ML. Structures and expression of fibulin-2, a novel extracellular matrix protein with multiple EGF-like repeats and consensus motif for calcium binding. J Cell Biol 1993; 123: 1269–1277.
• 16.    Jurukovski  V,  Dabovic  B,  Todorovic  V,  Chen  Y,  Rifkin  DB. Methods for measuring TGF-beta using antibodies, cells, and mice. Methods Mol Med 2005; 117: 161–175.
• 17.    Lim  H,  Zhu  YZ. Role of transforming growth factor-beta in the progression of heart failure. Cell Mol Life Sci 2006; 63: 2584–2596.
• 18.    Marijianowski  MM,  Teeling  P,  Mann  J,  Becker  AE. Dilated cardiomyopathy is associated with an increase in the type I/type III collagen ratio: A quantitative assessment. J Am Coll Cardiol 1995; 25: 1263–1272.
• 19.    Bai  Y,  Zhang  P,  Zhang  X,  Huang  J,  Hu  S,  Wei  Y. LTBP-2 acts as a novel marker in human heart failure: A preliminary study. Biomarkers 2012; 17: 407–415.
• 20.    Okada  H,  Takemura  G,  Kosai  K,  Li  Y,  Takahashi  T,  Esaki  M, et al. Postinfarction gene therapy against transforming growth factor-beta signal modulates infarct tissue dynamics and attenuates left ventricular remodeling and heart failure. Circulation 2005; 111: 2430–2437.
• 21.    Kuwahara  F,  Kai  H,  Tokuda  K,  Kai  M,  Takeshita  A,  Egashira  K, et al. Transforming growth factor-beta function blocking prevents myocardial fibrosis and diastolic dysfunction in pressure-overloaded rats. Circulation 2002; 106: 130–135.
• 22.    Koitabashi  N,  Danner  T,  Zaiman  AL,  Pinto  YM,  Rowell  J,  Mankowski  J, et al. Pivotal role of cardiomyocyte TGF-beta signaling in the murine pathological response to sustained pressure overload. J Clin Invest 2011; 121: 2301–2312.
• 23.    Ellmers  LJ,  Scott  NJ,  Medicherla  S,  Pilbrow  AP,  Bridgman  PG,  Yandle  TG, et al. Transforming growth factor-beta blockade down-regulates the renin-angiotensin system and modifies cardiac remodeling after myocardial infarction. Endocrinology 2008; 149: 5828–5834.
• 24.    Kapur  NK,  Wilson  S,  Yunis  AA,  Qiao  X,  Mackey  E,  Paruchuri  V, et al. Reduced endoglin activity limits cardiac fibrosis and improves survival in heart failure. Circulation 2012; 125: 2728–2738.
• 25.    Li  G,  Li  RK,  Mickle  DA,  Weisel  RD,  Merante  F,  Ball  WT, et al. Elevated insulin-like growth factor-I and transforming growth factor-beta 1 and their receptors in patients with idiopathic hypertrophic obstructive cardiomyopathy: A possible mechanism. Circulation 1998; 98: II144–II149; discussion II9–II50.
• 26.    Li  G,  Borger  MA,  Williams  WG,  Weisel  RD,  Mickle  DA,  Wigle  ED, et al. Regional overexpression of insulin-like growth factor-I and transforming growth factor-beta1 in the myocardium of patients with hypertrophic obstructive cardiomyopathy. J Thorac Cardiovasc Surg 2002; 123: 89–95.
• 27.    Teekakirikul  P,  Eminaga  S,  Toka  O,  Alcalai  R,  Wang  L,  Wakimoto  H, et al. Cardiac fibrosis in mice with hypertrophic cardiomyopathy is mediated by non-myocyte proliferation and requires TGF-beta. J Clin Invest 2010; 120: 3520–3529.
• 28.    Ikeuchi  M,  Tsutsui  H,  Shiomi  T,  Matsusaka  H,  Matsushima  S,  Wen  J, et al. Inhibition of TGF-beta signaling exacerbates early cardiac dysfunction but prevents late remodeling after infarction. Cardiovasc Res 2004; 64: 526–535.
• 29.    ten Dijke  P,  Arthur  HM. Extracellular control of TGFbeta signalling in vascular development and disease. Nat Rev Mol Cell Biol 2007; 8: 857–869.
• 30.    Karakikes  I,  Chaanine  AH,  Kang  S,  Mukete  BN,  Jeong  D,  Zhang  S, et al. Therapeutic cardiac-targeted delivery of miR-1 reverses pressure overload-induced cardiac hypertrophy and attenuates pathological remodeling. J Am Heart Assoc 2013; 2: e000078, doi:10.1161/JAHA.113.000078.