Anti-remodeling effects of rapamycin in experimental heart failure: dose response and interaction with angiotensin receptor blockade

Abstract

While neurohumoral antagonists improve outcomes in heart failure (HF), cardiac remodeling and dysfunction progress and outcomes remain poor. Therapies superior or additive to standard HF therapy are needed. Pharmacologic mTOR inhibition by rapamycin attenuated adverse cardiac remodeling and dysfunction in experimental heart failure (HF). However, these studies used rapamycin doses that produced blood drug levels targeted for primary immunosuppression in human transplantation and therefore the immunosuppressive effects may limit clinical translation. Further, the relative or incremental effect of rapamycin combined with standard HF therapies targeting upstream regulators of cardiac remodeling (neurohumoral antagonists) has not been defined. Our objectives were to determine if anti-remodeling effects of rapamycin were preserved at lower doses and whether rapamycin effects were similar or additive to a standard HF therapy (angiotensin receptor blocker (losartan)). Experimental murine HF was produced by transverse aortic constriction (TAC). At three weeks post-TAC, male mice with established HF were treated with placebo, rapamycin at a dose producing immunosuppressive drug levels (target dose), low dose (50% target dose) rapamycin, losartan or rapamycin + losartan for six weeks. Cardiac structure and function (echocardiography, catheterization, pathology, hypertrophic and fibrotic gene expression profiles) were assessed. Downstream mTOR signaling pathways regulating protein synthesis (S6K1 and S6) and autophagy (LC3B-II) were characterized. TAC-HF mice displayed eccentric hypertrophy, systolic dysfunction and pulmonary congestion. These perturbations were attenuated to a similar degree by oral rapamycin doses achieving target (13.3±2.1 ng/dL) or low (6.7±2.5 ng/dL) blood levels. Rapamycin treatment decreased mTOR mediated regulators of protein synthesis and increased mTOR mediated regulators of autophagy. Losartan monotherapy did not attenuate remodeling, whereas Losartan added to rapamycin provided no incremental benefit over rapamycin alone. These data lend support to investigation of low dose rapamycin as a novel therapy in human HF.

Figure 1. Serum rapamycin levels in response to dosing and administration routes.

Mice were dosed with 2, 4, 5 or 8/kg/day rapamycin through intraperitoneal (IP) or oral (PO) administration routes. Serum levels from 2 (n = 5), 5 (n = 6) and 8 (n = 8) mg/kg/day PO administration were dose dependent, with the latter approaching serum levels achieved with 2 mg/kg/day IP administration (n = 5). Serum rapamycin levels for HF mice given 8 mg/kg/day PO (n = 13) dosing were not different than that for normal mice, whereas 4 mg/kg/day PO (n = 6) showed reduced serum levels as expected. Data are mean ± SEM. †: P<0.05 vs. 2 mg/kg/day IP.

Figure 2. Effect of rapamycin on myocardial mTOR signaling in normal mice.

Western blots and grouped data are shown for S6K1 and Thr389 phosphorylated S6K1 (A), or S6 and Ser235/236 phosphorylated S6 (B) in rapamycin (8 mg/kg/day PO, n = 4) versus placebo treated normal mice (n = 4). Rapamycin treatment showed a strong trend towards decreasing total S6K1 expression, whereas Thr389 phosphorylated S6K1 and total and Ser235/236 phosphorylated S6 were all significantly decreased. Data are mean ± SEM. †: P<0.05 vs. placebo.

Figure 3. Effect of rapamycin on autophagy and autophagic flux in normal mice.

Using an antibody that identified both LC3B-I (top band) and -II (bottom band), representative Western blots and grouped data are presented for myocardial LC3B-II levels in normal mice treated with placebo (n = 4), 8 mg/kg/day PO rapamycin (n = 5) or rapamycin plus chloroquine (10 mg/kg IP, n = 4) to assess autophagic flux. Data are mean ± SEM. †: P<0.05 vs placebo, ‡: P<0.05 vs. rapamycin.

Figure 4. Effect of rapamycin on cardiomyocyte cross-sectional area in HF.

Representative left ventricular myocardial sections and grouped data for cardiomyocyte area in SHAM operated (1±0.04, n = 5), placebo treated HF (1.40±0.05, n = 5) and 8 mg/kg/day PO rapamycin treated HF mice (1.21±0.05, n = 6). Data are mean ± SEM. †: P<0.05 vs SHAM, ‡: P<0.05 vs. placebo treated HF mice.

Figure 5. Effect of rapamycin on myocardial fibrosis in HF.

Representative examples of picrosirius red stained left ventricular sections (A) and grouped data for interstitial fibrosis score (B) and relative collagen type I (C) and III (D) transcript levels in SHAM operated, placebo treated HF and 8 mg/kg/day PO rapamycin treated HF mice. Interstitial fibrosis scores were significantly higher in HF (n = 13) and HF + rapamycin mice (n = 11) when compared to SHAM (n = 8). For collagen type I and III transcripts, expression levels were significantly higher in HF (n = 5) and HF + rapamycin mice (n = 4) when compared to SHAM (n = 4). Data are mean ± SEM. †: P<0.05 vs SHAM.

Figure 6. Effect of rapamycin on myocardial mTOR signaling in HF.

Representative Western blots and grouped data for total or Thr389 phosphorylated S6K1 (A) and total or Ser235/236 phosphorylated S6 (B) in SHAM operated (n = 6), placebo treated HF (n = 6) and 8 mg/kg/day PO rapamycin treated HF mice (n = 6). Std: A standard LV homogenate sample for reliable comparison across gels. Data are mean ± SEM. †: P<0.05 vs SHAM, ‡: P<0.05 vs. placebo treated HF mice.

Figure 7. Effect of rapamycin on autophagy in HF.

Representative Western blots and grouped data for LC3B-II expression (bottom band) in myocardium from SHAM operated (n = 6), placebo treated HF (n = 6) and 8 mg/kg/day PO rapamycin treated HF mice (n = 6). Std: A standard LV homogenate sample for reliable comparison across gels. Data are mean ± SEM. †: P<0.05 vs SHAM, ‡: P<0.05 vs. placebo treated HF mice.