Exercise Training Stabilizes RyR2-Dependent Ca2+ Release in Post-infarction Heart Failure

Figure 1. Aerobic capacity and echocardiographic parameters. High-intensity exercise training improved aerobic capacity in sham-ET and HF-ET rats. (A) Exercise training protocol. (B) Mean VO2max measurements from sham and heart failure (HF) rats. Number of rats in each group (SED/ET): sham 15/14, HF 13/10. Echocardiographic parameters were used for stratification of rats to the sham and HF groups. Representative M-mode images from the left ventricle in the parasternal long axis are shown for sham-SED (C) (left, n = 9), sham-ET (middle left, n = 8), HF-SED (middle right, n = 12), and HF-ET rats (left, n = 8). Bar graphs of the (D) left atrium diameter, (E) left ventricular diameter in diastole, and (F) left ventricle fractional shortening represent mean ± SEM. P < 0.05: $sham vs. HF; *SED vs. ET; #HF-ET vs. sham-ET; §week 1 vs. week 6; ¤HF SED vs. sham SED (Paired Student's t-test).

Figure 2. Effect of exercise training on spontaneous Ca2+ release in heart failure (HF) cardiomyocytes. (A) Representative Ca2+ wave recordings and (D) confocal images of Ca2+ sparks are shown after field stimulation of HF-SED and HF-ET cardiomyocytes that were exposed to isoprenaline sulfate (ISO). Pacing at 1, 2, and 4 Hz was provided, and ISO was added to the experimental solution. Bar graphs of (B) Ca2+ wave frequency, (C) Ca2+ wave frequency across all pacing frequencies, and (E) Ca2+ spark frequency represent mean ± SEM. Number of rats in each group (SED/ET): Ca2+ wave frequency: n = 6/5, Ca2+ spark frequency: n = 10/8. Number of cells used for measurements of Ca2+ spark frequency (SED/ET): without ISO 101/81, with ISO 88/75. P < 0.05: *HF-SED vs. HF-ET (unpaired nested ANOVA); *(Poisson test for Ca2+ spark frequency, Bonferroni); #1 Hz vs. 2 and 4 Hz (Unpaired Student's t-test); ¤ +ISO vs. –ISO at same frequency (Unpaired Student's t-test), $least square mean for HF-SED vs. HF-ET (two-way ANOVA, Holm-Sidak).

Figure 3. Effects of exercise training on Ca2+ handling in heart failure (HF) cardiomyocytes. (A) Representative pacing protocol and (B) Ca2+ transients for HF-SED and HF-ET cardiomyocytes during 1 Hz stimulation and exposure to isoprenaline sulfate (ISO). (C) Ca2+ transient amplitude was reduced after exercise training at high stimulation frequency and ISO exposure and (D) across all frequencies with and without ISO exposure. (E,F) Ca2+ decay rate was lower across all frequencies with ISO exposure. (G) Representative tracings of SR Ca2+ content, measured by peak F/F0 in response to a rapid application of 10 mM caffeine. Exercise training (H) did not affect SR Ca2+ content and (I) did not alter Ca2+ removal by SERCA2a or sodium–calcium exchanger (NCX) and plasma membrane Ca2+ ATPase (PMCA), as (J) measured by the decay rate of the caffeine-induced Ca2+ release. (K) Exercise training did not alter diastolic Ca2+; (L) however, fractional release was higher in HF-ET compared to HF-SED when exposed to ISO. Number of rats (SED/ET): 6/5. Number of cells in C–F, K (SED/ET): –ISO 34/26, +ISO 35/36. Number of cells in H–J, L (ET/SED): HF 26/22, ISO 22/28. Bar graphs represent mean ± SEM. P < 0.05:…

Figure 4. Protein phosphorylation levels in heart failure (HF) left ventricles. Exercise did not decrease phosphorylation at sites that are known to be involved in destabilization of RyR2. Phosphoprotein levels were normalized to total protein levels. RyR2 phosphorylated at (A) serine 2814 (p-RyR2 ser 2814) and (B) serine 2808 (p-RyR2 ser 2808) were normalized to (C) total RyR2. Plb phosphorylated at (D) threonine 17 (p-Plb thr 17) and (E) serine 16 (p-Plb ser 16) were normalized to (F) total Plb and CaMKII phosphorylated at (G) threonine 286 (p-CaMKII thr 286) was normalized to (H) total CaMKII. (I) SERCA2a level was unaltered by exercise training. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as loading control, except for Plb, and the membrane was reprobed with the specified antibodies with stripping of the membrane between each antibody, shown at 37 kDa. (C,F,H,I) Total protein levels were normalized to GAPDH levels. Number of rats (SED/ET): 6/5.

Figure 5. Effect of exercise training on spontaneous Ca2+ release in sham cardiomyocytes. (A) Representative Ca2+ wave recordings, bar graphs of (B) Ca2+ wave frequencies at different pacing frequencies in the absence and presence of isoprenaline sulfate (ISO) and (C) two-way ANOVA comparing HF-ET with HF-SED. (D) Confocal images of Ca2+ sparks, and (E) bar graphs of Ca2+ spark frequency after different pacing frequencies. Number of rats (SED/ET): Ca2+ waves 4/3; Ca2+ sparks 7/6; number of cells (SED/ET): Ca2+ sparks: –ISO (60/53), +ISO (46/48); Ca2+ handling: –ISO (23/17), +ISO (10/9). P < 0.05: $least square mean for sham-SED vs. sham-ET (two-way ANOVA, Holm-Sidak).

Figure 6. Effect of exercise training on Ca2+ handling in sham cardiomyocytes. (A) Representative Ca2+ transient recordings is shown after field-stimulation of sham-SED and sham-ET cardiomyocytes. Stimulation of 1, 2, and 4 Hz was provided, and isoprenaline sulfate (ISO) was added to the experimental solution. Bar graphs of (B,C) Ca2+ transient amplitude, (D,E) Ca2+ removal and (F) diastolic Ca2+ represent mean ± SEM. (G) Representative tracings of SR Ca2+ content, measured by peak F/F0 in response to a rapid application of 10 mM caffeine at 1 Hz. (H) Bar graphs of SR Ca2+ load, represents mean ± SEM. Phosphoprotein levels were normalized to total protein levels. RyR2 phosphorylated at (I) serine 2814 (p-RyR2 ser 2814) and (J) serine 2808 (p-RyR2 ser 2808). (I,J) Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as loading control, and the membrane was reprobed with the specified antibodies with stripping of the membrane between each antibody, shown at 37 kDa. Number of rats (SED/ET): Ca2+ handling (4/3); proteins (5/5). Number of cells (SED/ET): Ca2+ handling: –ISO (23/17), +ISO (10/9). P < 0.05: $least square mean for sham-SED vs. sham-ET (two-way ANOVA, Holm-Sidak).

Figure 7. Beta-adrenoceptor density in heart failure (HF) and sham left ventricles. Competitive binding assays between [125I]-(–)iodocyanopindolol and the (A) beta1-adrenoceptor selective ligand CGP20712A or the (B) beta2-adrenoceptor selective ligand ICI118551 in membranes from the left ventricle from animals with the indicated treatment. The (C) total beta-adrenoceptor density and (D) percentage of beta1-adrenoceptor was determined by a two-site binding model, where the average of high-affinity CGP20712A and low-affinity of ICI118551 is presented. (E) Beta1-adrenoceptor density was determined as the average of high affinity of CGP20712ACFP20712A displacement and low affinity of ICI118551 displacement. (F) Beta2-adrenoceptor density was determined as the average of high affinity of ICI118551 displacement and low affinity of CGP20712ACFP20712A displacement. The data shown are mean ± SEM of five (sham SED), four (sham ET), six (HF SED), and six (HF ET) animals. P < 0.05: *SED vs. ET, ¤HF SED vs. sham SED (one-way ANOVA, Tukey's correction).