Journal Logo


Mechanical and Structural Remodeling of Cardiac Muscle after Aerobic and Resistance Exercise Training in Rats


Author Information
Medicine & Science in Sports & Exercise: August 2021 - Volume 53 - Issue 8 - p 1583-1594
doi: 10.1249/MSS.0000000000002625
  • Free


Adaptation of the heart in response to chronic exercise has been an area of considerable research interest (1–3). Participation in exercise and the associated improvements in fitness have been shown to lower the risk for cardiac diseases and are important factors for enhancing performance in many sports (4). However, the detailed cellular mechanisms underlying increased cardiac function after participation in chronic exercise training remain unclear.

Chronic aerobic and resistance exercise training have been shown to result in distinct cardiac phenotypes. Aerobic training is characterized by eccentric hypertrophy of the heart (proportional increases in wall thickness and ventricular volume), and resistance training is characterized by concentric hypertrophy (increased wall thickness and preserved ventricular volume) (5,6). Although there is some controversy regarding concentric hypertrophy, these data emerge from systematic reviews and meta-analyses (3,7) and are generally well accepted (2,5,6,8).

Aerobic training in rats results in greater unloaded/maximal shortening velocities and increased calcium sensitivity of the left ventricle (LV) compared with the corresponding values obtained in sedentary control rats (9–12). Diffee and Chung (10) observed greater rates of loaded shortening and higher peak power output in skinned cardiac myocytes from rats trained aerobically for 11 wk compared with sedentary control rats, and Diffee and Nagle (11) observed increases in calcium sensitivity of the myocardium in aerobically trained compared with sedentary control group rats. However, it is not known how the myocardium adapts to resistance exercise training. It has been suggested that active stress production may be greater in isolated intact fibers from hearts after 8 wk of resistance training (13), but it is unknown how resistance training affects mechanical properties of the myocardium, such as passive force, shortening velocity, and calcium sensitivity. Also, it is unknown how a combination of aerobic and resistance exercise affects the structure and function of the heart. Understanding the cellular adaptations of the heart to different exercise regimes may provide useful insights into the prevention of and rehabilitation from cardiac disease.

Therefore, the purpose of this study was to compare the potential mechanical adaptations of skinned cardiac fiber bundles to chronic resistance training, aerobic training, and combined resistance/aerobic exercise training in rats. We hypothesized that differences in cardiac function and structure with exercise are supported by corresponding adaptations at the cellular level.


Exercise Training

Twelve-week-old, male Sprague-Dawley rats were randomized into four groups: (i) aerobic exercise (n = 6), (ii) resistance exercise (n = 6), (iii) combination exercise (n = 6), and (iv) a sedentary control group (n = 5). Rats in the aerobic group participated in a progressive treadmill program (14) consisting of 60 min of treadmill running, 5 d·wk−1 at 25 m·min−1. Resistance-trained rats performed weighted ladder climbing 3 d·wk−1 as described in reference (15). Briefly, in each session, rats climbed a 1-m ladder inclined 80° carrying 50%, 75%, 90%, and 100% of their previous maximal training load. After matching their previous maximal load, during each subsequent climb, an additional 30 g was added until the rat could not overcome the load or lost their grip, and their new maximal training load was reached. Similar to Duncan et al. (16), the only stimulation necessary to encourage the rats to climb was an occasional hand prod to the base of the tail. Rats were given a 2-min rest between climbs. The combination group performed a hybrid program in which they completed 60 min of treadmill running 3 d·wk−1 at a speed of 25 m·min−1 and weighted ladder climbing 2 d·wk−1. The sedentary group rats were placed on the stationary treadmill for 10 min at 4 d·wk−1 and walked for 15 min at 10 m·min−1 once a week. The resistance-trained animals also received treadmill familiarization consisting of 10 min on the stationary treadmill once a week and walking for 15 min at 10 m·min−1 once a week. All training lasted 12 wk. Animals were housed individually at 21°C on a 12:12 light–dark cycle and had access to standard rat chow and water ad libitum. One animal from the control group was found dead during the intervention and was excluded. All other animals completed training. The study protocol was approved by The University of Calgary Animal Care Committee and conformed to the Guide for the Care and Use of Laboratory Animals.

Body Mass and Composition

Body mass was recorded weekly, and body composition was measured at the end of the 12-wk intervention protocol using dual-energy x-ray absorptiometry.

Fitness Measures

At the end of the 12-wk intervention protocol, aerobic fitness was determined for all rats using a graded treadmill test to exhaustion (17). For this test, animals were placed on the treadmill running at a speed of 12 m·min−1. The speed was increased by 1 m·min−1 every 2 min for the first 16 min, and then was increased by 2 m·min−1 every 3 min until failure. The maximal strength for rats in the resistance training group and the combination group was defined as the greatest load they could carry up the ladder for each week (15).

Cardiac Structure and Function

Echocardiographic evaluation was performed at the end of the intervention using an Esaote MyLab30 Gold Cardiovascular Ultrasound system (Canadian Veterinary Imaging, Georgetown, Ontario, Canada). Rats were anesthetized with isoflurane, placed in the dorsal decubitus position, and the ventral thoracic area was shaved. Two-dimensional images of the orthogonal long-axis four- and two-chamber views were obtained (18). End-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) were calculated from three consecutive cardiac cycles, by tracing the endocardial border of the LV in end-diastole and end-systole using the Simpson biplane method (18). LV wall thickness was determined with M-mode imaging measuring the anterior and posterior walls during systole (LVAWs, LVPWs) and diastole (LVAWd, LVPWd) (18). Structural measures were indexed to body surface area calculated using the Meeh’s equation (19) to control for body size.

Heart Isolation and Tissue Preparation

During tissue harvest, rats were anesthetized under 5% isoflurane, and their chest cavities opened. Animals were euthanized by severing the aorta and vena cava and excising the heart. Hearts were immediately flushed with rigor solution, and the vessels were removed. Heart mass was measured and normalized to the left tibia length to control for body size, independent of body composition (20). The LV was cut open, and 3–4 thin strips of trabeculae were sliced along the LV wall and placed in rigor solution on ice. After 2 h in the rigor solution, samples were transferred to a 50/50 rigor/glycerol solution and left overnight on ice in the fridge. The following morning, the muscle strips were transferred to a fresh 50/50 rigor/glycerol solution and chemically skinned at −20°C for 3 wk before mechanical testing. All samples were tested between 3 and 4 wk after harvest. An additional strip of trabeculae was dissected and immediately flash-frozen in liquid nitrogen and stored at −80°C for biochemical analysis.

Mechanical Testing

Once skinned, a strip of muscle was removed and placed in the relaxing solution where fiber bundles of approximately 100–300 μm in width and 1000–2000 μm in length were isolated manually under a Nikon SMZ1500 microscope. One end of the sample was pierced and glued with cellulose acetate to a hook connected to a length controller (Model 308; Aurora Scientific Inc., Ontario, Canada). The other end was similarly attached to the hook of a force transducer (Model 400A; Aurora Scientific Inc.), thereby allowing for simultaneous control of myocyte length and measurement of force. All experiments were performed at ~15°C. The sample length was coarsely adjusted from slack until tension first developed, and a He-Ne laser beam was then used to finely adjust average sarcomere length to a resting length of 2.2 μm. After a 3-min rest period to allow for stress relaxation, the length and width of the sample, and the passive (resting) force were recorded. Cross-sectional area of the sample was calculated from its diameter, assuming that the samples were cylindrical in shape. Two skinned samples were tested from each heart, and data were pooled for statistical analysis. Solutions were as previously described (21).

Maximal active stress

Maximal active isometric force was determined using a maximal activating solution (pCa = −log[Ca2+] = 4.2). Once peak force was reached, the sample was deactivated. Maximal active force was calculated as the difference between the total (measured) force and the resting passive force measured immediately preceding the contraction. Force was normalized to the sample’s cross-sectional area to obtain stress.

Passive stress

Samples were stretched passively from a resting average sarcomere length of 2.2 to 2.42 μm (10% of the sample’s total resting length) at a rate of 5% fiber length per second. Once stretched to 2.42 μm, sample length was held constant for 20 s to allow for stress relaxation, before being returned to the initial length (2.2 μm). The peak passive force was taken as the maximal value at the end of the stretch, whereas the steady-state passive force was determined as the mean value of the last second during stress relaxation when force had reached a constant value. Peak and steady-state passive forces were then also measured at an average sarcomere length of 2.53 μm using the same protocol, as just described previously. Passive force was normalized by each sample’s cross-sectional area to obtain the passive stress.

After the passive stretches, samples were activated maximally to ensure the sample did not sustain damage. If the maximal active force decreased by more than 15% from the initial contraction before stretch, the data from that fiber were excluded from analysis.

Unloaded shortening velocity

Maximal unloaded shortening velocity was determined at a sarcomere length of 2.2 μm using the slack test protocol (22). Briefly, samples were maximally activated and rapidly shortened (within 2 ms) by 10% (ΔL) of the sample’s length. By doing so, samples became slack and force dropped to zero. The time from the onset of the rapid shortening until force redeveloped (Δt) was measured. Samples were then relengthened, relaxed, and allowed to rest for 3 min. This rapid shortening test was repeated for ΔL of 11%, 12%, and 13% of the sample’s length. The slope of the linear relationship between ΔL and Δt was used to determine the unloaded (maximal) shortening velocity.

Calcium sensitivity

Calcium sensitivity was determined by establishing the force–pCa curve between pCa 7.0 and 4.2 (23). The baseline resting force was measured, before transferring the sample to the first solution (pCa 7.0). Once force stabilized, the sample was moved to the next solution containing a higher calcium concentration (pCa 6.8). Samples were exposed to solutions of continuously increasing calcium concentrations of pCa 6.6, 6.4, 6.2, 6.0, 5.8, 5.4, until reaching pCa 4.2 (maximal activating solution). Relative force at each calcium concentration was calculated by dividing the difference between the maximal and the baseline force at each calcium concentration by the difference between the maximal force produced at pCa 4.2 and the baseline force. The force–pCa relationship was then calculated by approximating the data using a least-squares regression for the Hill equation in SigmaPlot 13. Calcium sensitivity was defined and quantitatively compared between samples by determining the pCa value that yielded half of the maximal tension (pCa50) and the corresponding slope of the force–pCa relationship: cross-bridge cooperativity.

Biochemical Testing

Myofibrillar protein content

Muscle strips were ground into a powder and dissolved in an extraction buffer (50 mM Tris, 0.5 M NaCl, 20 mM NaPPI, 1 M EDTA, and 1 mM DTT, pH 7.4) at a concentration of 1 mg of muscle powder per 10 μL of buffer. Samples were vortexed, sonicated for 1 h, and centrifuged for 15 min at 13,000 rpm. Total myofibrillar protein content in the supernatant was then quantified using a BCA assay kit.

Myosin heavy-chain and actin content

Myosin heavy-chain (MHC) and actin content was assessed using SDS-PAGE gel electrophoresis on 4.5% and 12% acrylamide stacking and separating gels, respectively. Myofibrillar protein extracts were dissolved in an SDS solubilization buffer (62.5 mM Tris HCl, 10% glycerol, 2% SDS, 5% β-mercaptoethanol, 0.02% bromophenol blue, pH 6.8) to a final concentration of 2.2 mg·μL−1. Samples were boiled for 3 min and immediately spun at 5000 rpm for 15 min at 4°C. Solubilized samples were loaded (0.5 μL per well) into 0.75-mm-thick acrylamide gels, and the gels were run in a Biorad Mini-Protean III unit at a constant voltage of 22 V at room temperature for 18 h. Following electrophoresis, the gels were stained with Coomassie Blue for 60 min, then destained with a 50% ethanol, 7% acetic acid solution for 5 min, and a 5% ethanol, 7% acetic acid solution for at least 60 min. After destaining, the gels were scanned with a Biorad scanner and analyzed with Image J to compare the optical density of MHC and actin between groups.

MHC isoforms

MHC isoforms were determined using SDS-PAGE gel electrophoresis on 4.5% and 7.5% acrylamide stacking and separating gels, respectively, using the same procedure described previously to determine MHC and actin content. The gels were run at 4°C for 10 h at a constant voltage of 72 V and for 25 h at a constant current of 1 mA per gel. After electrophoresis, the gels were stained with Coomassie Blue, and ImageJ was used to determine the relative composition of α-MHC and β-MHC.


Kruskal-Wallis tests were used to determine statistical significance with pairwise comparisons to assess differences between groups when appropriate. Pearson correlation was used to assess the relationship between MHC isoforms and shortening velocity. Significance was accepted for P < 0.05. Group size was based on observed differences between groups in previous work (21,24) for a power of >80% for mechanical variables.


Body Mass and Composition

At the end of the 12-wk intervention period, there were no statistical differences in body mass between the groups (Table 1; P = 0.337). Animals from the aerobic group had significantly lower body fat content than did the control (P = 0.008) and resistance-trained animals (P = 0.016). There were no differences in body fat percentage between the other groups (P = 0.070–0.709; Table 1).

TABLE 1 - Descriptive end-point data.
Body Mass, g Body Fat, % Heart Mass
Absolute Heart Mass, g Heart Mass/Body Mass, mg·g−1 Heart Mass/Tibia Length, g·cm−1
Control 735 ± 27 21.5 ± 2.3 1.90 ± 0.06 2.6 ± 0.09 0.43 ± 0.02
Aerobic 655 ± 39 11.9 ± 0.8*,** 2.07 ± 0.09 3.2 ± 0.08*,*** 0.47 ± 0.02
Resistance 679 ± 31 19.9 ± 2.4 2.01 ± 0.07 2.9 ± 0.03* 0.47 ± 0.02
Combination 680 ± 26 14.9 ± 2.3 1.88 ± 0.09 2.8 ± 0.10 0.43 ± 0.02
Data displayed are means ± 1 SEM.
*Significant difference from the control group animals with P < 0.05.
**Significant difference from the resistance-trained animals with P < 0.05.
***Significant difference from the combination exercise–trained animals with P < 0.05.

Fitness Measures

Animals in the aerobic- and combination exercise–trained groups ran significantly longer on the graded treadmill fitness test than did the control (P = 0.001, P = 0.001) and resistance-trained (P = 0.010, P = 0.011) animals (Fig. 1A). The aerobic-trained animals had similar running times to animals from the combination exercise–trained group.

A, Individual data for time to exhaustion during graded treadmill test for each group. *Significant difference compared with the control group (P < 0.05). †Significant difference from the resistance-trained animals with P < 0.05. B, Mean ± SEM maximal weekly load lifted for resistance and combination training groups.

On average, the initial loads carried by the resistance- and combination exercise–trained animals were 98% and 104% of the animal’s body mass, respectively (Fig. 1B). The average maximal loads carried in the final week of training were 258% and 230% greater than the animals’ initial loads, respectively. There were no differences in the initial or the final load carrying capacity between the resistance- and combination exercise–trained animals (P = 0.522).

Heart Mass

There were no statistical differences for absolute heart mass (P = 0.455) or heart mass normalized to tibial length (P = 0.331; Table 1). When normalized to body mass, the aerobic-trained animals had larger hearts compared with the controls (P = 0.001) and combination exercise–trained animals (P = 0.011), and the resistance-trained animals had a larger relative heart mass than did the control group animals (P = 0.026).

Cardiac Structure and Function

Absolute structural values are presented in Table 2 and indexed values in Figure 2. Absolute LV EDV was greater in hearts from aerobic- and combination exercise–trained animals as compared with hearts from the resistance-trained (P = 0.004, P = 0.003) and sedentary control (P = 0.022, P = 0.016) animals, which were not different from each other (P = 0.673; Table 2). When indexed to body surface area, EDV was greater in hearts from aerobic- and combination exercise–trained animals as compared with hearts from the resistance-trained (P = 0.017, P = 0.005) and sedentary control (P = 0.006, P = 0.010) animals, which were not different from each other (P = 0.865; Fig. 2). There were no differences in absolute LV ESV between groups (P = 0.141). Indexed ESV was greater in the aerobic-trained animals compared with the controls (P = 0.004) but was not different between other groups (P = 0.101–0.865). The resistance-trained animals had thicker absolute anterior walls in diastole (LVAWd) than control (P = 0.001) and combination (P = 0.045) group animals. Indexed LVAWd was greater in the resistance- and aerobic-trained animals than the control animals (P = 0.001, P = 0.038), but was not different between the combination exercise–trained and resistance-trained (P = 0.115) or the combination exercise–trained and control (P = 0.051) group animals. The resistance-trained animals also had thicker absolute anterior walls at systole (LVAWs) than did the control (P = 0.004) and aerobic-trained (P = 0.022) animals. Indexed LVAWs was greater in the combination exercise–trained and resistance-trained animals compared with the controls (P = 0.037, P = 0.002). Resistance- and aerobic-trained animals had thicker absolute ventricular walls (LVPWd) compared with the control group animals (P = 0.007, P = 0.031), but not compared with the combination exercise–trained animals. Indexed LVPWd was greater in all exercised groups compared with the controls (P = 0.005–0.038). There were no differences for absolute LVPW at systole (LVPWs; P = 0.071), but when indexed, LVPWs was greater in the resistance-trained animals compared with the control (P = 0.036) and aerobic (P = 0.01) group animals.

TABLE 2 - Cardiac structure and functional parameters measured via echocardiography.
Control Aerobic Resistance Combination
EF, % 72.8 ± 2.4 77.5 ± 5.3 66.4 ± 2.3 72.5 ± 4.6
EDV, μL 296 ± 21 391 ± 23*,** 270 ± 24 387 ± 15*,**
ESV, μL 78 ± 6 116 ± 13 92 ± 14 90 ± 9
LVAWd, mm 1.88 ± 0.06 2.53 ± 0.43 3.00 ± 0.20*,*** 2.20 ± 0.08
LVAWs, mm 3.14 ± 0.27 3.35 ± 0.18** 4.00 ± 0.11* 3.60 ± 0.15
LVPWd, mm 1.88 ± 0.09 2.43 ± 0.20* 2.55 ± 0.08* 2.27 ± 0.19
LVPWs, mm 2.40 ± 0.20 3.48 ± 0.25 3.20 ± 0.20 2.90 ± 0.25
Data displayed are means ± 1 SEM.
*Significant difference from the control group animals with P < 0.05.
**Significant difference from the resistance-trained animals with P < 0.05.
***Significant difference from the combination exercise–trained animals with P < 0.05.

Indexed cardiac structural data from echocardiogram. A, EDV. B, ESV. C, LVAWd. D, LVAWs. E, LVPWd. F, LVPWs. *Significant difference from the control group animals with P < 0.05. †Significant difference from the resistance-trained animals with P < 0.05. ‡Significant difference from the combined exercise–rained animals with P < 0.05.

Mechanical Testing

Maximal active and passive stress

Maximal active stress was greater in samples from animals in the resistance-trained (P = 0.016) and combination exercise–trained (P = 0.008) groups than those from control animals (Fig. 3A). Active stress for the control and aerobic-trained groups was similar (P = 0.209). There were no differences between any groups for peak passive stresses at the end of stretching or passive stresses after stress relaxation for the 10% and 15% stretch magnitudes (P = 0.297–0.550; Table 3).

A, Active stress. B, Sample trace for slack test. Vertical lines indicate Δt for unloaded shortening velocity with shortening. C, Maximal shortening velocity. Individual data. *Significant difference from the control group animals with P < 0.05. ‡Significant difference from the combined exercise–trained animals with P < 0.05.
TABLE 3 - Passive stress data from skinned trabeculae preparations (means ± 1 SEM).
Passive 10% Peak Stress, kN·m−2 Passive 10% Steady State Stress, kN·m−2 Passive 15% Peak Stress, kN·m−2 Passive 15% Steady-State Stress, kN·m−2
Control 8.3 ± 2.7 5.1 ± 1.6 14.9 ± 4.2 8.7 ± 2.6
Aerobic 8.4 ± 1.4 5.4 ± 0.9 16.0 ± 2.1 10.0 ± 1.3
Resistance 10.1 ± 1.9 8.4 ± 0.9 18.3 ± 3.2 11.3 ± 1.9
Combination 11.6 ± 1.6 7.2 ± 1.1 16.2 ± 2.3 9.7 ± 1.4
There were no significant differences in passive stress between groups.

Unloaded shortening velocity

Maximal unloaded cardiac tissue shortening velocity was significantly higher in aerobic-, resistance-, and combination exercise–trained animals compared with the sedentary control animals (P = 0.001, P = 0.009, P = 0.016, respectively; Figs. 3B, C). Unloaded shortening velocity was also greater for the aerobic-trained than the combination exercise–trained (P = 0.016) animals.

Calcium sensitivity

Animals in the aerobic and resistance exercise groups had greater calcium sensitivity than did the control and the combination exercise–trained animals (Figs. 4A, B), with pCa50 values significantly greater in the aerobic-trained (P = 0.006, P = 0.001) and resistance-trained (P = 0.017, P = 0.004) animals compared with the control and combination exercise–trained animals (Fig. 4C). Cross-bridge cooperativity was significantly greater in the combination exercise–trained animals compared with the control (P = 0.022), aerobic-trained (P = 0.017), and resistance-trained (P = 0.021) animals (Fig. 4D). There were no differences in cross-bridge cooperativity between control, aerobic-trained, and resistance-trained animals (P = 0.964–0.993).

A, Mean force–pCa relationships for each group. B, Sample trace for calcium sensitivity test. Noise reflects changing of solutions through pCa solutions (pCa 7.0–5.4) with maximal activation (pCa 4.2) in the last well. C, Individual data points for pCa 50 values for each group. D, Individual data for Hill slope values for each group. *Significant difference from the control group animals with P < 0.05. †Significant difference from the resistance-trained animals with P < 0.05. ‡Significant difference from the combined exercise–trained animals with P < 0.05.

Biochemical Testing

There were no significant differences between groups for MHC and actin content or myofibrillar protein content (Figs. 5A, B). Hearts from the control group animals had a significantly lower proportion of α-MHC than did those from the aerobic-trained (P = 0.003) and resistance-trained (P = 0.013) groups, but there were no differences between the other groups (P = 0.240–0.593; Fig. 5C). There was a significant positive correlation (r = 0.66; P = 0.001) between the relative α-MHC composition and the maximal unloaded shortening velocity (Fig. 5D).

Biochemical testing. A, MHC and actin content expressed relative to the values in the control animals, B, Myofibrillar protein content. C. Relative expression of α-MHC isoform. D, Correlation between maximal shortening velocity and the relative expression of α-MHC. Individual data points. *Significant difference from the control group animals with P < 0.05.


The purpose of this study was to compare the mechanical adaptations of skinned cardiac muscle fiber bundles after chronic resistance, aerobic, and combined exercise training in rats. The main findings were that, compared with sedentary control rats, hearts from animals that participated in aerobic training had greater unloaded shortening velocity and increased calcium sensitivity; resistance-trained hearts had greater active stress, greater unloaded shortening velocity, and increased calcium sensitivity; and hearts from animals performing both aerobic and resistance training had greater active stress, greater shortening velocity, and greater cross-bridge cooperativity. In addition, both eccentric and concentric hypertrophies were observed with aerobic and resistance training, as predicted by the literature.

Fitness Indices

Based on the treadmill and load carriage findings, we conclude that the rat models of exercise used were effective in increasing aerobic fitness in those training on the treadmill and in increasing strength in those climbing the ladder, as has been shown previously (14,15). It is interesting to note that animals in the combination exercise–trained group had similar average time to exhaustion to that of the aerobic group animals, and similar maximal loads to that of the resistance group animals, despite performing 2 d less of aerobic exercise and 1 d less of resistance exercise per week compared with the singe-modality animals (Fig. 1). A major strength of this study is that the animal’s activity was strictly limited to the assigned training modality, as the lack of such control has been a limitation in long-term human studies (3). Animals were housed individually without enrichment to control for additional cage activity. Although we do not have measures for strength in the aerobic-trained animals, the lack of difference between the resistance-trained and control animals for treadmill time to failure indicates that the ladder climbing exercise was distinctly strength based and did not affect aerobic fitness, thereby supporting the distinction between aerobic only and resistance only.

Unsurprisingly, exercise training led to differences in body composition. These were the most defined in the aerobic-trained animals, which were exposed to a greater metabolic load and greater effect on energy balance than the other animals, and therefore had the most significant changes in body composition (25).

Cardiac Structural Remodeling

In previous work, unique differences between the structures of hearts from those participating in aerobic and resistance exercise training were identified (5,7,26,27), although the results have been conflicting. There is agreement regarding eccentric hypertrophy after aerobic training, but concentric hypertrophy with resistance training is less clear (3,28). Spence et al. (6) used magnetic resonance imaging to compare cardiac dimensions before and after 6 months of strictly controlled aerobic or resistance training in initially untrained humans. They observed the expected increase in LV volume and mass for the aerobic group but did not observe the expected increase in LV wall thickness in the resistance-trained group. Conversely, Baggish et al. (8) used echocardiography to compare cardiac structure before and after 90 d of either endurance or resistance training in healthy young athletes. They observed increased LV mass and volume with aerobic training and an increase in LV mass only with resistance training. In the present study, we observed hypertrophy in both aerobically trained animals (increased LVPW, LVAW, EDV, and ESV) and in resistance-trained animals (increased LVAW and LVPW, but no change in EDV and ESV). The combination training group presented with features similar to both the aerobic and resistance-trained groups (EDV, LVAW, LVPW). Outside the tight controls of laboratory animal husbandry, it is likely that structural adaptation in human athletes more closely reflects that of the combination exercise–trained group, whereby adaption is characterized by features of both concentric and eccentric hypertrophies, which may explain the conflicting results previously reported with resistance training.

Cardiac Mechanical Adaptations

Through each cardiac cycle, the muscular wall of the LV is required to produce an isometric contraction, a forceful concentric contraction, and then relax for passive stretching as the chamber refills (29). Exercise has been shown to reduce cardiac stiffness, resulting in more compliant ventricles during diastole (30,31). However, we did not observe differences in passive stresses in the present study. We attribute this to the skinning process where the connective tissues are partially disrupted (32) and therefore less able to contribute to passive properties. Had we tested intact fibers, we would have expected to see decreased passive stresses with exercise training. Given that there were no observed differences in this study for the inherent passive force properties of the muscle, this discussion will focus primarily on the active properties. An increase in contractility during the active phase could be accomplished through either an increase in maximal and submaximal stress production or a greater speed of shortening (10,29).

Maximal active stress

According to Laplace’s law and assuming equal pressure (which was not measured in the present study) with an increase in chamber volume, a greater amount of force must be produced to maintain wall stress and function of the ventricle (33). With aerobic training, maintenance of wall stress has been shown to originate from a proportional thickening of the muscular wall rather than by an increase in the intrinsic force production of the muscle (1,3). Active force production (normalized to cross-sectional area) has consistently been shown to be unchanged after aerobic training in rats ranging from 30 to 60–240 min·d−1 and for periods of up to 12 wk (9–11,14).

A limitation of this study is that we used echocardiography to measure cardiac structure but did not take any hemodynamic or pressure measurements. However, it has been shown previously that during resistance training, greater afterload due to elevated blood pressure requires a greater force of ejection (7,28). In response to resistance training, the heart has been shown to accomplish this through thickening of the muscular walls (2,7,8). In addition to observing thicker walls, in the present study, we also identified an increase in the intrinsic force production of the heart, as has been suggested by Pinter et al. (13). Of note, greater stress production was observed in both the resistance-only and the combination exercise–trained animals. This increased stress production augments the potential maximal ejection pressure of the heart. However, it remains unclear how this increase in intrinsic force is achieved. Structural analysis did not reveal a difference in the amount or proportion of contractile proteins between the experimental groups. Pinter et al. (13) observed greater stress values in isolated intact papillary muscle after resistance training and speculated that this greater stress is partly due to a greater rate of myosin ATPase activity. Therefore, we speculate that the increase in intrinsic force observed in this study for the resistance-trained animals is primarily associated with an increase in the proportion of bound cross-bridges contributing to total stress after chronic resistance training. However, further work is required to identify the mechanisms for the increase in specific cardiac force in resistance-trained rats.

Calcium sensitivity

Changes in calcium sensitivity of cardiac tissue have been associated with functional implications for cardiac function. For example, an increased calcium sensitivity with stretch is thought to be a key factor in the increased contractility described by the Frank–Starling law of the heart (34,35). An overall increase in calcium sensitivity from exercise training would increase contractility even at low ventricular volumes and thus less cardiac tissue stretch. In addition, the ability to produce the same force with less intracellular calcium results in less calcium handling, thereby providing for more efficient contractions (36).

It was previously unknown how resistance training affects calcium sensitivity. However, in agreement with our findings from the aerobic-trained animals, increased pCa50 values with no change in cross-bridge cooperativity have been reported after 12 wk of chronic aerobic exercise training (11,12). The authors attribute these changes to alterations in myosin light-chain (MLC) and troponin C expression (37). For example, Diffee and Nagle (11) observed greater pCa50 in myocytes from aerobically trained rats and reported a significant correlation between pCa50 and the relative expression of MLC-1. The specific mechanisms for the observed changes in calcium sensitivity were not explored in our study but should be the focus of further research.

It is of particular interest that, although aerobic and resistance training independently led to increases in pCa50, the combination of the two did not. Maximal stress and cross-bridge cooperativity were greater in the combination group but not in the aerobic group. We speculate that the greater maximal stress output and cooperativity (slope of the force–pCa curve) may have compensated for no change in pCa50 in the combination exercise–trained animals. Although speculative, the absence of an increase in pCa50 in the combination group animals might be explained by the absence of an increase in the relative expression of α-MHC. The MLC and regulatory proteins normally expressed by a muscle cell are characteristic of its MHC isoforms. It may be speculated that the increase in α-MHC expression observed in the aerobic and resistance training groups was accompanied by changes in MLC and regulatory protein isoforms, causing an increase in calcium sensitivity. The combination group did not result in an increase in the expression of α-MHC, and therefore, we speculate that its MLC and regulatory protein isoforms may also not have been altered. Cooperativity, on the other hand, has been shown to depend on the kinetics of cross-bridges (38) and therefore may be altered in the absence of changes in MHC, MLC, and regulatory protein expression. It is also possible that, although either 5 d·wk−1 of treadmill training or 3 d of ladder climbing was sufficient to stimulate adaptation, the combination of 3 d of aerobic training and 2 d of resistance training was insufficient to stimulate the adaptation of pCa50. This highlights the need for further work to determine the relative proportions of aerobic and combination training that leads to optimal adaptations of heart function.

Shortening velocity

The isometric phase of the cardiac cycle is short, and much of the cardiac contraction cycle is concentric (29). Thus, the combination of force production and rate of shortening contributes to the ejection power. A limitation of this study is that maximal stress was measured isometrically, and shortening velocity was only measured under zero load. For both these conditions, muscular power is zero by definition. However, because unloaded shortening velocity is thought to reflect the cross-bridge cycling rates within a preparation (39), these rates would be expected to be elevated as well during loaded shortening conditions (22,40) and would be a reflection of increased contractility during ejection.

Our results are supported in part by the literature, in which shortening velocity has been shown to be greater in aerobically trained animals compared with sedentary controls (9). For example, after 11 wk of aerobic exercise training, Diffee and Chung (10) tested the isotonic force–velocity relationship and observed greater rates of shortening for given relative force outputs and therefore a greater power output under these conditions. Although previously untested, in the current study, resistance and combined training also resulted in higher cardiac shortening velocities relative to the values observed for the sedentary control group animals. However, the effects did not seem to be additive as shortening velocity for the combination exercise–trained animals increased to a similar degree to that of the aerobic- and resistance-only group animals.

Researchers have provided various explanations regarding the mechanisms underlying the increase in maximal unloaded shortening velocity observed as a result of exercise training. The primary hypothesis has been that the increase in shortening velocity is associated with specific changes in the relative expression of the two cardiac-specific MHC isoforms: α-MHC and β-MHC (41). Exercise has been shown to lead to a greater relative proportions of the fast α-MHC compared with the slow β-MHC (26,42). However, some research groups (10) compared the unloaded and loaded shortening velocities and MHC expression after 11 wk of treadmill running and did not observe a relationship between the two. Although the unloaded and loaded shortening velocities were higher in the trained compared with the nontrained rats at the same resistances, there was no change in the MHC isoform composition; thus, there was no relationship between velocity of shortening and the α-MHC versus β-MHC ratio (8). Instead, they attributed the differences in shortening speeds to increases in MLC-1 composition. We did not evaluate MLC expression in the present study. However, we observed a significant correlation (r = 0.658) between α-MHC expression and shortening velocity, suggesting that MHC may partly explain the observed differences in unloaded cardiac tissue shortening velocity. However, given the strength of the correlation, it is clear that there may be other factors (such as MLC) influencing shortening velocity after exercise, and more work needs to be done to explore these mechanisms.

Consequences of Training

In this study, all forms of exercise led to structural and functional adaptations of the heart muscle. Aerobic and resistance training increased contractility of the isolated cardiac fiber bundles via different mechanisms. Except for the calcium sensitivity, but with the addition of increased cross-bridge cooperativity, the combination exercise–trained animals had the same adaptations as those observed in the aerobic- and resistance-trained animals. Thus, combining the two modalities may result in greater overall adaptation, or the same adaptation with less overall training. More research should be done to identify how combining aerobic and resistance training can further enhance these adaptations to result in overall greater cardiac health and performance.


We conclude from the results of this study that exercise affects the structure and function of the heart and its cellular components in a manner that is specific to the exercise protocol. There may be additional benefits of combining aerobic and resistance exercise training, but more work is required to fully elucidate these mechanisms.

This work was supported by the Vanier Canada Graduate Scholarship Program, Alberta Innovates–Health Solutions, Canada Research Chair Programme, the Canadian Institutes for Health Research, and the Killam Foundation.

The authors declare no conflicts of interest.

The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of the present study do not constitute endorsement by the American College of Sports Medicine.


1. Diffee GM. Adaptation of cardiac myocyte contractile properties to exercise training. Exerc Sport Sci Rev. 2004;32:112–9.
2. Naylor LH, George K, O’Driscoll G, Green DJ. The athlete’s heart: a contemporary appraisal of the ‘Morganroth hypothesis’. Sports Med. 2008;38(1):69–90.
3. Utomi V, Oxborough D, Whyte GP, et al. Systematic review and meta-analysis of training mode, imaging modality and body size influences on the morphology and function of the male athlete’s heart. Heart. 2013;99(23):1727–33.
4. Nayor M, Vasan RS. Preventing heart failure: the role of physical activity. Curr Opin Cardiol. 2015;30(5):543–50.
5. Morganroth J, Maron BJ, Henry WL, Epstein SE. Comparative left ventricular dimensions in trained athletes. Ann Intern Med. 1975;82:521–4.
6. Spence AL, Naylor LH, Carter HH, et al. A prospective randomised longitudinal MRI study of left ventricular adaptation to endurance and resistance exercise training in humans. J Physiol. 2011;589:5443–52.
7. Pluim BM, Zwinderman AH, van der Laarse A, van der Wall EE. The athlete’s heart. A meta-analysis of cardiac structure and function. Circulation. 2000;101:336–44.
8. Baggish AL, Wang F, Weiner RB, et al. Training-specific changes in cardiac structure and function: a prospective and longitudinal assessment of competitive athletes. J Appl Physiol. 2008;104(4):1121–8.
9. Chung E, Diffee GM. Moderate intensity, but not high intensity, treadmill exercise training alters power output properties in myocardium from age rats. J Gerontol A Biol Sci Med Sci. 2012;67:1178–87.
10. Diffee GM, Chung E. Altered single cell force–velocity and power properties in exercise-trained rat myocardium. J Appl Physiol. 2003;94:1941–8.
11. Diffee GM, Nagle DF. Exercise training alters length dependence of contractile properties in rat myocardium. J Appl Physiol. 2003;94:1137–44.
12. Diffee GM, Seversen EA, Titus MM. Exercise training increases the Ca2+ sensitivity of tension in rat cardiac myocytes. J Appl Physiol. 2001;91:309–15.
13. Pinter RCCE, Padilha AS, De Oliveira EM, Vassallo DV, de Fúcio Lizardo JH. Cardiovascular adaptive responses in rats submitted to moderate resistance training. Eur J Appl Physiol. 2008;103(5):605–13.
14. Boldt KR, Rios JL, Joumaa V, Herzog W. Force properties of skinned cardiac muscle following increasing volumes of aerobic exercise in rats. J Appl Physiol. 2018;125(2):495–503.
15. Hornberger TA Jr, Farrar RP. Physiological hypertrophy of the FHL muscle following 8 weeks of progressive resistance exercise in the rat. Can J Appl Physiol. 2004;29:16–31.
16. Duncan ND, Williams DA, Lynch GS. Adaptations in rat skeletal muscle following long-term resistance exercise training. Eur J Appl Physiol Occup Physiol. 1998;77(4):372–8.
17. Hohl R, Ferraresso RL, De Oliveira RB, Lucco R, Brenzikofer R, De Macedo DV. Development and characterization of an overtraining animal model. Med Sci Sports Exerc. 2009;41:1155–63.
18. Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. J Am Soc Echocardiogr. 1989;2(5):358–67.
19. Gouma E, Simos Y, Verginadis I, Lykoudis E, Evangelou A, Karkabounas S. A simple procedure for estimation of total body surface area and determination of a new value of Meeh’s constant in rats. Lab Anim. 2012;46(1):40–5.
20. Yin FC, Spurgeon HA, Rakusan K, Weisfeldt ML, Lakatta EG. Use of tibial length to quantify cardiac hypertrophy: application in the aging rat. Am J Physiol. 1982;243(6):H941–7.
21. Boldt K, MacDonald GZ, Joumaa V, Herzog W. Mechanical adaptations of skinned cardiac muscle in response to dietary-induced obesity during adolescence in rats. Appl Physiol Nutr Metab. 2020;45(8):893–901.
22. Edman KA. The velocity of unloaded shortening and its relation to sarcomere length and isometric force in vertebrate muscle fibres. J Physiol. 1979;291:143–59.
23. Danieli-Betto D, Betto R, Midrio M. Calcium sensitivity and myofibrillar protein isoforms of rat skinned skeletal muscle fibres. Pflugers Arch. 1990;417(3):303–8.
24. Boldt K, Rios JL, Joumaa V, Herzog W. Mechanical function of cardiac fibre bundles is partly protected by exercise in response to diet-induced obesity in rats. Appl Physiol Nutr Metab. 2021;46(1):46–54.
25. Drenowatz C, Grieve GL, DeMello MM. Change in energy expenditure and physical activity in response to aerobic and resistance exercise programs. Springerplus. 2015;4(1):798.
26. Fernandes T, Baraúna VG, Negrão CE, Phillips MI, Oliveira EM. Aerobic exercise training promotes physiological cardiac remodeling involving a set of microRNAs. Am J Physiol Heart Circ Physiol. 2015;309(4):H543–52.
27. duManoir GR, Haykowsky MJ, Syrotuik DG, Taylor DA, Bell GJ. The effect of high-intensity rowing and combined strength and endurance training on left ventricular systolic function and morphology. Int J Sports Med. 2007;28(6):488–94.
28. Haykowsky MJ, Dressendorfer R, Taylor D, Mandic S, Humen D. Resistance training and cardiac hypertrophy. Sports Med. 2002;32:837–49.
29. Helmes M, Najafi A, Palmer BM, et al. Mimicking the cardiac cycle in intact cardiomyocytes using diastolic and systolic force clamps; measuring power output. Cardiovasc Res. 2016;111(1):66–73.
30. Chung CS, Hiske MA, Chadha A, Mueller PJ. Compliant titin isoform content is reduced in left ventricles of sedentary versus active rats. Front Physiol. 2020;11:15.
31. Kwak HB, Kim JH, Joshi K, Yeh A, Martinez DA, Lawler JM. Exercise training reduces fibrosis and matrix metalloproteinase dysregulation in the aging rat heart. FASEB J. 2011;25(3):1106–17.
32. Kentish JC, Ter Keurs HE, Ricciardi L, Bucx JJ, Noble MI. Comparison between the sarcomere length–force relations of intact and skinned trabeculae from rat right ventricle. Influence of calcium concentrations on these relations. Circ Res. 1986;58:755–68.
33. Lee LC, Genet M, Dang AB, Ge L, Guccione JM, Ratcliffe MB. Applications of computational modeling in cardiac surgery. J Card Surg. 2014;29(3):293–302.
34. Fuchs F, Smith SH. Calcium, cross-bridges, and the Frank–Starling relationship. News Physiol Sci. 2001;16:5–10.
35. Shiels HA, White E. The Frank–Starling mechanism in vertebrate cardiac myocytes. J Exp Biol. 2008;211:2005–13.
36. Chung JH, Biesiadecki BJ, Ziolo MT, Davis JP, Janssen PM. Myofilament calcium sensitivity: role in regulation of in vivo cardiac contraction and relaxation. Front Physiol. 2016;7:562.
37. Willott RH, Gomes AV, Chang AN, Parvatiyar MS, Pinto JR, Potter JD. Mutations in troponin that cause HCM, DCM AND RCM: what can we learn about thin filament function?J Mol Cell Cardiol. 2010;48(5):882–92.
38. Moss RL, Razumova M, Fitzsimons DP. Myosin crossbridge activation of cardiac thin filaments: implications for myocardial function in health and disease. Circ Res. 2004;94(10):1290–300.
39. Claflin DR, Faulkner JA. Shortening velocity extrapolated to zero load and unloaded shortening velocity of whole rat skeletal muscle. J Physiol. 1985;359:357–63.
40. Reggiani C. When fibres go slack and cross bridges are free to run: a brilliant method to study kinetic properties of acto-myosin interaction. J Physiol. 2007;583(Pt 1):5–7.
41. Herron TJ, Korte FS, McDonald KS. Loaded shortening and power output in cardiac myocytes are dependent on myosin heavy chain isoform expression. Am J Physiol Heart Circ Physiol. 2001;281:H1217–22.
42. Hashimoto T, Kambara N, Nohara R, Yazawa M, Taguchi S. Expression of MHC-β and MCT1 in cardiac muscle after exercise training in myocardial-infarcted rats. J Appl Physiol. 2004;97(3):843–51.


Copyright © 2021 by the American College of Sports Medicine