Exercise training reverses cardiac aging phenotypes associated with heart failure with preserved ejection fraction in male mice

FIGURE 1. Age‐related changes in resting cardiac function in C57BL/6 male mice are similar to cardiac functional phenotypes in human HFpEF. Nonsedated transthoracic echocardiography in C57BL/6 male mice at 3–4 months (young (Y), n = 12), 24–26 months (old (O), n = 17), and 28–30 months (very old (VO), n = 14). (a) Fractional shortening, (b) systolic strain, (c) early diastolic strain rate (SR). Data shown as mean ± SEM, with all individual data points plotted. One‐way ANOVA with post hoc Tukey's test used for analyses. *p < .05, **p < .01, ***p < .001

FIGURE 2. Progressive age‐related decline in exercise capacity and cardiac reserves in C57BL/6 male mice recapitulates exercise intolerance phenotypes in human HFpEF. Stress echocardiography testing in C57BL/6 male mice at 3–4 months (Y, n = 12), 24–26 months (O, n = 17), and 28–30 months (VO, n = 13). (a) Exercise capacity measured by total distance run and total work achieved (adjusted for body weight). (b) Chronotropic and contractile reserves measured at peak exercise. (c) Pearson correlation of exercise capacity (work) with chronotropic or contractile reserves. In panels a and b, data shown as mean ± SEM, with all individual data points plotted, and one‐way ANOVA with post hoc Tukey's test used for analyses. *p < .05, **p < .01, ***p < .001

FIGURE 3. Pathologic cardiac hypertrophy in aged C57BL/6 male mice recapitulates histopathologic cardiac phenotypes in human HFpEF. (a) Cardiac histopathologic assessments of 4‐month (Y, n = 6–12) and 25‐ to 26‐month (O, n = 6–9) mice, including representative photomicrographs and quantification of cardiomyocyte cross‐sectional area (CM CSA), myocardial fibrosis, and capillary density. Scale bar = 100 µm. (b) Relative cardiac mRNA expression of BNP. n = 5/group. (c) Systemic mean arterial pressure (MAP) in Y (n = 11) versus O (n = 8) mice. For all panels, data shown as mean ± SEM, with all individual data points plotted. Unpaired Student's t test used for analyses. *p < .05, **p < .01, ***p < .001

FIGURE 4. Aerobic exercise training reverses some HFpEF phenotypes in aged male mice. (a) Experimental design of exercise training substudy. Prior to study completion, 3 mice died in the sedentary group (SED) and 2 mice died in the exercise‐trained group (ExT). (b) Resting cardiac function by strain echocardiography. (c) Exercise capacity. (d) Chronotropic and contractile cardiac reserves. (e) Lung weight (LW) normalized to tibial length (TL). (f) Heart weight (HW) normalized to TL. (g) % Fibrosis. (h) Capillary density. For fibrosis and capillary density measurements, histologic sections from a SED animal that died one day prior to the final timepoint functional testing were included in the final analyses. Data shown as mean ± SEM, with all individual data points plotted. Unpaired Student's t test used for analyses. *p < .05, **p < .01, ***p < .001

FIGURE 5. Reversal of cardiac aging pathways with exercise training. RNAseq analyses on cardiac tissue from young (4 months) sedentary, very old (30 months) sedentary, and very old (30 months) exercise‐trained (ExT) C57BL/6 male mice. ExT mice completed an 8‐week treadmill running protocol from 28 to 30 months prior to sacrifice. n = 3/group. (a) Volcano plot of differentially expressed genes between young and very old sedentary mice. (b) Volcano plot of differentially expressed genes between sedentary and ExT very old mice. (c) Most highly significant pathways that are differentially regulated in opposing directions by age (red, young sedentary vs. very old sedentary) and ExT (blue, very old sedentary vs. very old ExT). Only pathways with an FDR < 0.05 in both the aging and exercise‐trained cohorts with normalized enrichment scores (NES) in opposing directions are displayed. For volcano plots, black = nonsignificant; orange = log2(FC) ≥1; red = p adj < .05; green = log2(FC) ≥1 + p adj < .05