Moderate Endurance Exercise Increases Arrhythmia Susceptibility and Modulates Cardiac Structure and Function in a Sexually Dimorphic Manner

Abstract

Background: Although moderate endurance exercise has been reported to improve cardiovascular health, its effects on cardiac structure and function are not fully characterized, especially with respect to sexual dimorphism. We aimed to assess the effects of moderate endurance exercise on cardiac physiology in male versus female mice.

Methods and results: C57BL/6J mice of both sexes were run on a treadmill for 6 weeks. ECG and echocardiography were performed every 2 weeks. After 6 weeks of exercise, mice were euthanized, and triple parametric optical mapping was performed on Langendorff perfused hearts to assess cardiac electrophysiology. Arrhythmia inducibility was tested by programmed electrical stimulation. Left ventricular tissue was fixed, and RNA sequencing was performed to determine exercise-induced transcriptional changes. Exercise-induced left ventricular dilatation was observed in female mice alone, as evidenced by increased left ventricular diameter and reduced left ventricular wall thickness. Increased cardiac output was also observed in female exercised mice but not males. Optical mapping revealed further sexual dimorphism in exercise-induced modulation of cardiac electrophysiology. In female mice, exercise prolonged action potential duration and reduced voltage-calcium influx delay. In male mice, exercise reduced the calcium decay constant, suggesting faster calcium reuptake. Exercise increased arrhythmia inducibility in both male and female mice; however, arrhythmia duration was increased only in females. Lastly, exercise-induced transcriptional changes were sex dependent: females and males exhibited the most significant changes in contractile versus metabolism-related genes, respectively.

Conclusions: Our data suggest that moderate endurance exercise can significantly alter multiple aspects of cardiac physiology in a sex-dependent manner. Although some of these effects are beneficial, like improved cardiac mechanical function, others are potentially proarrhythmic.

Keywords: aerobic; arrhythmias; electrical remodeling; exercise; sex differences.

Figure 1. Experimental protocol.

A, Different experimental protocols and timelines applied to Sed and EX mice over 6 weeks (W0 to W6) are illustrated. B, Changes in the total body weight of mice in the 4 experimental groups were recorded biweekly and are plotted over 6 weeks. Three‐way ANOVA tests were performed to determine significant differences in body weight data presented in (B), with time, sex, and treatment (Sed, EX) as variables. For post hoc analysis, to compare changes in body weight in each mouse over the duration of the experimental protocol, paired Student's t tests were performed, and significant P values are indicated in the figure. Benjamini–Hochberg correction was applied for multiple comparisons correction. P<0.05 denotes significance. The sample size for each group is as follows: male sedentary=11, male exercised=9, female sedentary=11, female exercised=9. EX indicates exercised; and Sed, sedentary.

Figure 2. Exercise causes LV dilatation and improves mechanical function in females.

A, Echocardiography was performed biweekly and representative echocardiograms from male and female, Sed and EX mice at baseline (week 0, top panels) and after the 6‐week protocol (week 6, bottom panels) are illustrated. Echocardiograms in each column's top and bottom panels are from the same mouse. B, Summary data of structural (top) and mechanical (bottom) parameters measured from the echocardiograms, normalized to each mouse's own baseline value before the start of experimental protocol are plotted over time. Three‐way ANOVA tests were performed to determine significant differences in echocardiographic data presented in (B), with time, sex, and treatment as variables. Paired, 2‐tailed Student's t tests were performed for post hoc analysis to determine significant changes induced by experimental protocol in the same mouse over time, and significant P values are indicated in the figure. Benjamini–Hochberg correction was applied for multiple comparisons correction. P<0.05 denotes significance. The sample size for each group is as follows: male sedentary=11, male exercised=9, female sedentary=11, female exercised=9. CO indicates cardiac output; EF, ejection fraction; EX, exercised; FS, fractional shortening; HR, heart rate; LV mass, left ventricular mass; LVEDD, left ventricular end diastolic volume; LVESD, left ventricular end systolic volume; LVPWd, left ventricular posterior wall thickness during diastole; LVPWs, left ventricular posterior wall thickness during systole; Sed, sedentary; and SV, stroke volume.

Figure 3. Electrocardiographic assessment of cardiac response to exercise.

A, Electrocardiography was performed biweekly and representative ECGs from the male and female, Sed and EX mice at baseline (week 0, black) and after the 6‐week protocol (week 6, blue: male, pink: female). Overlapping ECG traces are from the same mouse. B, Summary data of ECG parameters measured and normalized to each mouse's own baseline value before start of experimental protocol are plotted over time. Three‐way ANOVA tests were performed to determine significant differences in ECG data presented in (B), with time, sex, and treatment as variables. Paired, 2‐tailed Student's t tests were performed for post hoc analysis to determine significant changes induced by experimental protocol in the same mouse over time and significant P values are indicated in the figure. Benjamini–Hochberg correction was applied for multiple comparisons correction. P<0.05 denotes significance. The sample size for each group is as follows: male sedentary=11, male exercised=9, female sedentary=11, female exercised=9. EX indictes exercised; and Sed, sedentary.

Figure 4. Exercise modulates cardiac electrophysiology in a sex‐specific manner.

A, At the end of the experimental protocol, hearts were explanted, and triple parametric optical mapping was performed. Simultaneously recorded NADH intensity maps and voltage and calcium activation maps generated from triple‐parametric optical mapping of male and female, Sed and EX hearts paced at 150 ms basic cycle lengths are illustrated. All 3 maps in a given column are simultaneous recordings of the same field of view. B, Representative voltage and calcium optical traces from the optical mapping of hearts in the 4 experimental groups. Sedentary: solid line; exercised: dashed lines. C, Transmembrane potential, calcium handling, and metabolism‐related parameters in exercised mice were normalized to sedentary controls and are illustrated as 10 parameter panels and radial graph. Orange lines on the radial graph correspond with 20% and −20% change in any given parameter. Two‐way ANOVA was performed to determine significant differences in data presented in (C), with sex and treatment as the 2 variables. Unpaired, 2‐tailed Student's t tests were performed to determine significant changes in optical mapping parameters presented in (C) and significant P values are indicated. Benjamini–Hochberg correction was applied for multiple comparisons correction. P<0.05 denotes significance. The sample size for each group is as follows: male sedentary=5, male exercised=6, female sedentary=6, female exercised=6. APD₈₀ indicates action potential duration at 80% repolarization; AR, anisotropic ratio; Ca τ, decay constant of calcium transient; CaRT, rise time of upstroke of calcium transient; CaTD₈₀, calcium transient duration at 80% reuptake; CV_L, longitudinal conduction velocity; CV_T, transverse conduction velocity; EX, exercised; NADH, normalized NADH intensity; Sed, sedentary; V _m RT, rise time of the upstroke of action potential; V _m‐Ca delay, delay between activation of voltage and calcium signals.

Figure 5. Increased arrhythmia susceptibility in hearts due to exercise.

A, Representative ECG traces illustrating incidences of sinus rhythm, VT, and VF in ex vivo mouse hearts. B, Pseudo‐ECG traces illustrating burst pacing (indicated by black pulses above ECG traces) followed by either an arrhythmia episode or sinus rhythm in the 4 experimental groups (left) is illustrated. Quantification of VT and VF incidences and total arrhythmia duration in these 4 groups is plotted on the right. C, Phase maps were generated from optical recordings of arrhythmia episodes in exercised male and female ex vivo hearts. These maps generated 10 seconds apart illustrate the propagation of the excitation wavefront (red regions) during the arrhythmia. White arrows indicate direction of propagation of excitation wavefront. In the male heart, a clear circular movement of the wavefront is evident, indicating reentrant arrhythmia, whereas in the representative female heart, a more complex arrhythmia pattern is observed. Two‐way ANOVA was performed to determine differences in data presented in (B), with sex and treatment as the 2 variables. Unpaired, 2‐tailed Student's t tests were performed as post hoc analysis, to determine significant changes in arrhythmia parameters presented in (B). Benjamini–Hochberg correction was applied for multiple comparisons correction. P<0.05 denotes significance. The sample size for each group is as follows: male sedentary=5, male exercised=6, female sedentary=6, female exercised=6. EX indicates exercised; Sed, sedentary; VF, ventricular fibrillation; and VT, ventricular tachycardia.

Figure 6. Differential cardiac gene expression in response to exercise.

A, Principal component analysis plot illustrating the separation of sedentary and exercised groups in male and female gene expression profiles. B, Volcano plots showing DEGs in exercised vs sedentary controls. C and D, Top 10 DEGs in exercised vs sedentary mice of either sex. In instances where fewer DEGs were identified between groups, only those genes are shown. E and F, Top enriched pathways in exercised vs sedentary mice of either sex. DESeq2 package in R was used to determine significant DEGs. clusterProfiler package in R was used to determine enriched pathways. The significance level was set to <0.05 and log2 fold change of >0 was reported as upregulated and <0 as downregulated. The sample size for each group is as follows: male sedentary=3, male exercised=3, female sedentary=3, female exercised=3. DEG indicates differentially expressed gene; ECM, extracellular matrix; EX, exercised; EXF, exercised female; EXM, exercised male; Sed, sedentary; Sed F, sedentary female; and Sed M, sedentary male.

Figure 7. Sex differences in cardiac gene expression in exercised and sedentary mice.

A, Volcano plots showing sex differences in DEGs of exercised and sedentary mice. B and C, Top 10 DEGs in female vs male mice of exercised and sedentary groups. D and E, Top enriched pathways in female vs male mice of exercised and sedentary groups. DESeq2 package in R was used to determine significant DEGs. clusterProfiler package in R was used to determine enriched pathways. The significance level was set to <0.05 and log2 fold change of >0 was reported as upregulated and <0 as downregulated. The sample size for each group is as follows: male sedentary=3, male exercised=3, female sedentary=3, female exercised=3. AGE indicates advanced glycation end product; DEG, differentially expressed gene; EX, exercised; EXF, exercised female; EXM, exercised male; NF kappa B, nuclear factor kappa B; RAGE, receptor for advanced glycation end product; Sed, sedentary; Sed F, sedentary female; and Sed M, sedentary male.

Figure 8. Summary of the effects of exercise on cardiac structure and function.

An overview of the findings of this study is illustrated, indicating the effects of exercise on cardiac gene expression, structure, mechanics, and electrophysiology in female and male hearts. APD indicates action potential duration; EP indicates electrophysiology; and VT, ventricular tachycardia.