GDF11 Decreases Pressure Overload-Induced Hypertrophy, but Can Cause Severe Cachexia and Premature Death

Abstract

Rationale: Possible beneficial effects of GDF11 (growth differentiation factor 11) on the normal, diseased, and aging heart have been reported, including reversing aging-induced hypertrophy. These effects have not been well validated. High levels of GDF11 have also been shown to cause cardiac and skeletal muscle wasting. These controversies could be resolved if dose-dependent effects of GDF11 were defined in normal and aged animals as well as in pressure overload-induced pathological hypertrophy.

Objective: To determine dose-dependent effects of GDF11 on normal hearts and those with pressure overload-induced cardiac hypertrophy.

Methods and results: Twelve- to 13-week-old C57BL/6 mice underwent transverse aortic constriction (TAC) surgery. One-week post-TAC, these mice received rGDF11 (recombinant GDF11) at 1 of 3 doses: 0.5, 1.0, or 5.0 mg/kg for up to 14 days. Treatment with GDF11 increased plasma concentrations of GDF11 and p-SMAD2 in the heart. There were no significant differences in the peak pressure gradients across the aortic constriction between treatment groups at 1 week post-TAC. Two weeks of GDF11 treatment caused dose-dependent decreases in cardiac hypertrophy as measured by heart weight/tibia length ratio, myocyte cross-sectional area, and left ventricular mass. GDF11 improved cardiac pump function while preventing TAC-induced ventricular dilation and caused a dose-dependent decrease in interstitial fibrosis (in vivo), despite increasing markers of fibroblast activation and myofibroblast transdifferentiation (in vitro). Treatment with the highest dose (5.0 mg/kg) of GDF11 caused severe body weight loss, with significant decreases in both muscle and organ weights and death in both sham and TAC mice.

Conclusions: Although GDF11 treatment can reduce pathological cardiac hypertrophy and associated fibrosis while improving cardiac pump function in pressure overload, high doses of GDF11 cause severe cachexia and death. Use of GDF11 as a therapy could have potentially devastating actions on the heart and other tissues.

Keywords: cachexia; cardiomegaly; fibrosis; growth differentiation factors; hypertrophy.

Figure 1.. Injections of rGDF11 increase the circulating concentration of GDF11 and SMAD2 phosphorylation.

A: Plasma was collected 1–1.5 hours after injection of GDF11 or vehicle. B: Log transformations of ratio of phosphorylated to total forms of SMAD2 protein, as well as the ratio phosphorylated and total SMAD2 to GAPDH in the hearts of vehicle-treated and GDF11-treated TAC and sham mice was quantified by densitometry. (n=5–8 per group) C: Immunoblotting for phosphorylated and total forms of SMAD2 and GAPDH protein expression in the hearts of vehicle-treated and GDF11-treated TAC and sham mice. *p<0.05 vs vehicle-treated TAC mice. #p<0.05 vs vehicle-treated sham mice. T-tests: TAC vehicle vs sham vehicle, sham vehicle vs sham 5.0 mg/kg One-way ANOVA using Dunnet’s multiple comparison test: TAC animals.

Figure 2.. GDF11 reduces TAC-induced cardiac hypertrophy.

A: Peak pressure gradients across the aortic constriction were measured 1 week after TAC or sham surgery. B,C: Heart weight to body weight ratio and heart weight to tibia length ratio were determined 21 days after TAC surgery (16/17 days after TAC surgery for mice treated with 5.0 mg/kg). (n=8–19) D: Representative images of the mid-papillary transverse sections of hearts after 21 days of TAC and 14 injections of GDF11 or vehicle (16/17 days after TAC surgery and 9 or 10 injections for mice treated with 5.0 mg/kg of GDF11). Data are represented as mean ±SEM. *p<0.05 vs vehicle-treated TAC mice. #p<0.05 vs vehicle-treated sham mice. T-tests or Mann-Whitney test: vehicle TAC vs vehicle sham determined significance between sham and TAC mice with vehicle treatment. One-way ANOVA using Dunnet’s multiple comparison test or Kruskal-Wallis test with Dunn’s multiple comparison test: within sham animals, or within TAC animals.

Figure 3.. GDF11 decreases myocyte cross-sectional area and alters expression of markers of hypertrophy.

A: Wheat germ agglutinin (WGA) staining was performed to assess myocyte size. B: Quantification of cross sectional areas of cardiac myocytes in WGA stained hearts from 3–8 mice per group. At least 300 myocytes were analyzed when their nucleus was clearly present in each mouse. Scale bar = 50 microns. C-F: RT-PCR analysis of BNP, alpha skeletal actin (Acta1), alpha myosin heavy chain (Myh6), and beta myosin heavy chain (Myh7) mRNA expression transformed with log or square root (n=4–8). All data are represented as mean ±SEM *p<0.05 vs vehicle-treated TAC mice. #p<0.05 vs vehicle-treated sham mice. T-tests: vehicle TAC vs vehicle sham. One-way ANOVA using Dunnet’s multiple comparison test: within sham animals, within TAC animals.

Figure 4.. GDF11 altered cardiac remodeling after TAC.

Echocardiography was performed at baseline, 1, and 3 weeks after TAC surgery. A: Left ventricular anterior wall thickness (top: diastole bottom: systole) B: Left ventricular posterior wall thickness C: Left ventricular internal dimension (top: diastole, bottom: systole). D: Left ventricular mass. Data are represented as mean ±SEM *p<0.05 vs vehicle-treated TAC mice at same time point. #p<0.05 vs vehicle-treated sham mice at same time point. ^<0.05 vs same group at baseline. $p<0.05 vs same group at 1 week. Linear mixed models were used for statistical analysis.

Figure 5.. GDF11 improves cardiac function after TAC.

Echocardiography was performed at baseline, 1, and 3 weeks after TAC surgery. A: Ejection fraction B: Fractional shortening. C: End systolic volume. D: End diastolic volume. Data are represented as mean ±SEM *p<0.05 vs vehicle-treated TAC mice at same time point. #p<0.05 vs vehicle-treated sham mice at same time point. ^<0.05 vs same group at baseline. $p<0.05 vs same group at 1 week. Linear mixed models were used for statistical analysis.

Figure 6.. GDF11 reduces interstitial fibrosis and collagen mRNA expression.

GDF11 treatment began 1 week after mice underwent TAC surgery. A: Representative images of hearts stained with Masson’s trichrome. B: Quantification of interstitial fibrosis from Masson’s trichrome staining. C: Quantification of fibrosis from areas containing vessels with square root transformation. (n=3–8 mice, approximately 20 fields of view) D: RT-PCR analysis of collagen1a1 (col1a1) mRNA with square root transformation (n=4–8). Data are represented as mean ±SEM *p<0.05 vs vehicle-treated TAC mice. #p<0.05 vs vehicle-treated sham mice. T-tests: TAC vehicle vs Sham Vehicle. One-way ANOVA using Dunnet’s multiple comparison test: within sham animals, or within TAC animals.

Figure 7.. GDF11 increases markers of fibroblast activation and differentiation into myofibroblasts.

MEFs were treated with GDF11 or TGFβ for 72 hours for staining and confocal microscopy and 48 hours for PCR. A: Representative images of MEFs after 72 hours of treatment. Green-vimentin, Blue-Dapi, Red- αSMA. B: quantification of MEFs positive for both αSMA and vimentin. C-E: RT-PCR analysis of acta2, cola1a, and log of postn expression. (n=3) Data are represented as mean ±SEM #p<0.05 vs control. *p<0.05 vs 10 ng/mL TGFβ. One-way ANOVA using Tukey’s multiple comparison test or Kruskal-Wallis test with Dunn’s multiple comparison test: within sham animals, or within TAC animals.

Figure 8.. GDF11 induces severe cachexia at high dosage.

Mice were weighed daily beginning with the first dose of GDF11. A: percent body weight change was calculated using the weight at time of sacrifice and the weight at the time of the first injection of GDF11 (n=8–20). B: Tibialis anterior (TA) weight to tibia length, Quadriceps (Quad) weight to tibia length, and gastrocnemius (Gast) weight to tibia length ratios were analyzed 21 days after TAC surgery (16/17 days after TAC surgery and 9 for mice treated with 5.0 mg/kg of GDF11). (n=8–20) C: RT-PCR analysis of Bnip3, Bnip3L, Fbxo32, and Trim63 mRNA expression. (n=6) Data are represented as mean ±SEM *p<0.05 vs vehicle-treated TAC mice. #p<0.05 vs vehicle-treated sham mice. T-tests/Mann-Whitney test: One-way ANOVA using Dunnet’s multiple comparison test or Kruskal-Wallis test with Dunn’s multiple comparison test: within sham animals, or within TAC animals .