Structural and functional alterations in heart and skeletal muscle following severe TAC in mice: impact of moderate exercise training

Background Heart failure (HF) is the leading cause of death in western countries. Cardiac dysfunction is accompanied by skeletal alterations resulting in muscle weakness and fatigue. Exercise is an accepted interventional approach correcting cardiac and skeletal dysfunction, thereby improving mortality, re-hospitalization and quality of life. Animal models are used to characterize underpinning mechanisms. Transverse aortic constriction (TAC) results in cardiac pressure overload and finally HF. Whether exercise training improves cardiac remodeling and peripheral cachexia in the TAC mouse model was not analyzed yet. In this study, 2 weeks post TAC animals were randomized into two groups either performing a moderate exercise program (five times per week at 60% VO2 max for 40 min for a total of 8 weeks) or staying sedentary. Results In both TAC groups HF characteristics reduced ejection fraction (− 15% compared to sham, p < 0.001), cardiac remodeling (+ 22.5% cardiomyocyte cross sectional area compared to sham; p < 0.001) and coronary artery congestion (+ 34% diameter compared to sham; p = 0.008) were observed. Unexpectedly, peripheral cachexia was not detected. Furthermore, compared to sedentary group animals from the exercise group showed aggravated HF symptoms [heart area + 9% (p = 0.026), heart circumference + 7% (p = 0.002), right ventricular wall thickness − 30% (p = 0.003)] while muscle parameters were unchanged [Musculus soleus fiber diameter (p = 0.55), Musculus extensor digitorum longus contraction force (p = 0.90)]. Conclusion The severe TAC model is inappropriate to study moderate exercise effects in HF with respect to cardiac and skeletal muscle improvements. Further, the phenotype induced by different TAC procedures should be well documented and taken into account when planning experiments.


Introduction
Approximately 64.3 million people worldwide suffer from heart failure (HF). In European countries the prevalence is up to 1.3-4% [1]. Despite continuously improving therapies in HF the 5-and 10-year mortality still is estimated to be 53% and 65%, respectively [2]. Two thirds of HF patients suffer from skeletal muscle atrophy [3], fatty infiltration, decreased antioxidant capacity and show muscle fiber type shift from slow-twitching to fast-twitching muscle fibers resulting in a metabolic shift from aerobic, oxidative to a rather anaerobic, glycolytic metabolism [4]. These molecular alterations in skeletal muscle structure and

Open Access
Biological Research *Correspondence: Julia.Boettner@medizin.uni-leipzig.de 1 Department of Cardiology, Heart Center Leipzig at University of Leipzig, Strümpellstr. 39, 04289 Leipzig, Germany Full list of author information is available at the end of the article composition result in a loss of lean body mass, muscle weakness and fatigue [3]. Nevertheless, HF can be prevented or decelerated by modifying behavioral patterns. One easy-to-address and establish cardioprotective habit is physical activity, because it has the ability to correct the structural and metabolic alterations in the peripheral muscle [4][5][6][7][8]. Regular exercise is beneficial for cardiac health by reducing risk factors [6,9]. Furthermore, exercise improves quality of life in HF patients by increasing exercise tolerance and oxygen uptake [7,10,11]. Exercise-based cardiac training is known to effectively reduce total and cardiovascular mortality as well as hospital admission [12].
Despite the accepted protective effect of exercise on HF, the molecular basis is still not completely understood. This is hindering the development of individualized exercise regimens and the identification of underpinning regulatory pathways that may be targeted pharmaceutically.
Different animal models, that imitate human pathologies, are accessible for HF induction. In patients with HF due to myocardial infarction exercise improves cardiac remodeling and function [13]. Thus, ligation of the left anterior descending artery (LAD) is a common method to induce myocardial infarction and subsequently HF in animal models. A drawback of this model is high mortality and low reproducibility [14].
To overcome these obstacles transverse aortic constriction (TAC) is used to induce HF whereas low mortality and high reproducibility are reported for this method [15]. TAC limits the left ventricular (LV) outflow resulting in LV pressure overload and fast evolving HF characterized by cardiac hypertrophy and remodeling as well as decreasing cardiovascular performance [16,17]. To standardize the procedure a needle is placed next to the aorta, a full constriction is placed and after the needle is removed, the final aortic diameter is determined. Lower needle diameters determine more severe hypertrophy as well as accelerated time course of HF development [15,16]. Mild TAC introduced by aortic banding using a 25 gauge (G) needle represents a mild model of hypertrophy and diastolic dysfunction without systolic dysfunction or cardiac fibrosis, consistent with the human condition of hypertensive heart disease [16]. Moderate 26G TAC results in systolic dysfunction with perivascular cardiac fibrosis.
Severe TAC applied by using a 27G needle eventuates in systolic and diastolic dysfunction, distinct cardiac fibrosis, hypertrophy and secondary increased lung mass, consistent with pulmonary edema [16].
The effect of exercise on HF development and associated muscle remodeling using the TAC model in mice was studied by Wang et al. and Tian et al. [8,18]. They observed improved cardiac performance following exercise training with increased intensity [8]. Most existing studies on TAC in combination with exercise intervention focus on cardiac effects and lack information about TAC-and exercise-induced structural, functional and metabolic alterations of the skeletal muscles. On that account, we applied severe cardiac pressure overload by banding the aorta using a 27G needle to analyze the effects of moderate training in HF on cardiac and skeletal muscle structure and function. To our best knowledge, this is the first study to address both, cardiac and skeletal muscle performance in HF mice after exercise intervention.

Evaluation of TAC phenotype
In the 48 h post-operative period six mice died following 27G severe TAC, two mice died following sham surgery. In TAC SED one mouse died 22 d post-surgery from unknown cause, which led to final groups of n = 13 in sham, n = 11 in sedentary (TAC SED ) and n = 12 in exercising TAC (TAC EX ) groups.
Echocardiographic measurements (Fig. 1A) confirmed establishment of TAC 2 weeks after surgery. The velocity in the left common carotid artery (LCCA) decreased after TAC (p < 0.001), whereas the right common carotid artery (RCCA) velocity only numerically increased after TAC compared to sham-operated animals (p = 0.098).
Ten weeks post-surgery body weights (BW) and tibial length (TL) in TAC SED and TAC EX were comparable to sham. All results are summarized in Table 1.

Severe TAC alters cardiac structure and function in sedentary animals
Cardiac remodeling after TAC operation was evaluated by analysis of heart metrics, development of interstitial and perivascular fibrosis as well as coronary artery remodeling 10 weeks post TAC (see Fig. 2 and Table 1). The area and circumference in TAC SED were comparable to the sham operated control group. LV thickness of heart cross sections was slightly decreased in TAC SED . Interestingly, the right ventricular (RV) wall thickness in TAC SED was significantly reduced by 22% in comparison to the sham group (p = 0.049 see Fig. 2A). Cardiomyocyte cross sectional area (CSA) in TAC SED was significantly increased by 22% in contrast to sham (see Fig. 2B). The amount of interstitial and perivascular fibrosis in TAC SED was comparable to sham (see Fig. 2D). The coronary artery diameter in TAC SED was significantly increased by 9 µm (p = 0.008 vs. sham; Fig. 2E). The heart weight of the TAC SED was increased up to 1.4-fold in contrast to sham. Echocardiographic analyses revealed an increase in diastolic LV volume by 50% (p = 0.002), but no significant alterations in LV posterior wall thickness (p = 0.27) as well as a significant reduction in left ventricular ejection fraction (LVEF) of 15% compared to sham in TAC SED . Additionally, systolic LV volume was significantly increased by 28 µL (p < 0.001), although LV wall thickness in the systolic phase was unaltered in TAC SED (p = 0.64).

Exercise effects on TAC induced cardiac dysfunction
We analyzed the impact of moderate exercise training (TAC EX ) on the cardiac alterations observed in TAC SED ( Fig. 2 and Table 1). Cardiac CSA was increased by 8% (p = 0.026) and heart circumference was increased by 6% (p = 0.002), while RV wall thickness was decreased by 30% (p = 0.003) in relation to sham. RV wall thickness reduction was observed in TAC SED (23%) and TAC EX (24%) without differences between these two groups (p = 0.56). In TAC EX cardiomyocyte CSA was increased by 18% in contrast to sham (p < 0.001). In comparison to TAC SED , cardiomyocyte CSA remained similar in TAC EX (p = 0.56). TAC EX showed a 1.44-fold rise in cardiac interstitial fibrosis compared to sham (p = 0.01).
No difference in the amount of interstitial fibrosis in TAC SED and TAC EX was observed. The amount of perivascular fibrosis in TAC EX was comparable to sham and to TAC SED . Numerically, exercise intervention in the TAC operated group reduced the coronary artery diameter in the myocardium by 6 µm in TAC EX compared to TAC SED . Heart weight in TAC EX was significantly increased in contrast to sham (p < 0.001). No difference in heart weight between TAC SED and TAC EX was found. Functional cardiac parameters of TAC EX are shown in Fig. 1. Diastolic and systolic LV volume in TAC SED and in TAC EX remained comparable. The diastolic and systolic LV wall thickness in TAC SED were not improved in TAC EX . LV volume in diastolic and systolic phase remained 1.5-and twofold increased and the LVEF was reduced by 24% (p < 0.001) compared to sham. Additionally, no differences of LV output were found in TAC SED and in TAC EX .

Functional and structural alterations in skeletal muscle in sedentary and trained TAC mice
Effects of moderate exercise on HF were analyzed focusing on cardiac characteristics as well as peripheral muscle alterations. Extensor digitorum longus (EDL) contraction force, muscle fiber diameter and the proportion of slow-twitch and fast-twitch fibers in the EDL and the soleus muscle were analyzed to study the effects of moderate exercise in TAC induced HF 10 weeks after TAC surgery (see Table 1 and Fig. 3). EDL contraction force was significantly increased in TAC SED compared to sham (p = 0.017), but not in TAC EX . The EDL fiber diameter was significantly lower in TAC SED (p = 0.044) but not in TAC EX (p = 0.16), whereas there was no difference between both TAC groups.
Soleus muscle fiber diameter in TAC SED remained comparable to sham. Interestingly, soleus muscle fiber diameter was significantly decreased (p = 0.002) in TAC EX compared to sham. No alterations in the amount of slow and fast muscle fibers in EDL as well as in slow muscle fibers in soleus were observed.

Discussion
We applied 27G severe TAC to induce a severe HF phenotype but chose a moderate exercise regimen to avoid overcharging TAC mice by exercise. Resulting cardiac dysfunction and peripheral cachexia as well as the potential capability of moderate exercise to improve them was examined. The results of the study can be summarized as follows: 1. additional aggravation of cardiac remodeling after exercise intervention 2. similar cardiac performance independently from exercise intervention 3. comparable structure and function of skeletal muscles in both, TAC SED and TAC EX 4. EDL cross-sectional area was improved by exercise intervention, whereas the Soleus muscle showed further fiber atrophy following exercise intervention

LV remodeling and dysfunction in severe TAC
In the present study 27G severe TAC induced HF characteristics in the mouse model. Increased diastolic and systolic LV filling accompanied by a significantly reduced EF were found, indicating LV dilatation and dysfunction as hallmarks of HF. The observed interstitial fibrotic and coronary artery remodeling is also in accordance to other studies [16,19]. Cardiac remodeling in the TAC mouse model is accompanied by a congestion into coronary vessels [19] resulting in an increase in coronary artery diameter. We found an 1.  fibrosis [16]. Yang et al. found interstitial and perivascular fibrosis 5 weeks after TAC in male mice (no needle diameter specified) [19]. Wang et al. [8] reported significant fibrosis 10 weeks after TAC (no needle diameter specified, presumably 27G [20]). Taken together, the development and pattern of fibrosis seems to depend strongly on the used needle diameter and the progressed time post TAC [7,8]. Notably, our data suggest that the development of fibrosis in TAC SED is highly individual, which is contrary to the concept that TAC results in an uniform and reproducible development of HF [15].
Based on the current results, future studies should include the specific TAC phenotype in their hypothesis generation including a detailed description of the performed TAC procedure as well as mouse strain, sex, age and origin. Furthermore, the high inter-individual variability in cardiac remodeling observed in this study should be taken into account when planning TAC experiments with regards to animal numbers and sample size calculation.

RV remodeling in severe pressure overload
While we found no differences in LV wall thickness following TAC surgery, the RV wall was significantly thinner in both TAC groups. The echocardiographic analysis of the lateral RV wall thickness confirmed diastolic RV dilation, maybe resulting in RV dysfunction. RV dilatation as a consequence of LV pressure overload was addressed in only a few studies [21,22] and thereby remains elusive. In pulmonary hypertension (PH) RV dilatation was found to represent an adjustment mechanism to reduce RV load [23]. Our data are in line with recent findings in TAC operated mice that showed distinct RV remodeling and diastolic dysfunction [22]. Platt et al. applied 26G TAC and monitored structural and functional RV and LV remodeling over 18 weeks. They postulated diastolic RV dysfunction to occur secondary to LV dysfunction. Finally, they showed establishment of PH secondary to RV dysfunction [22]. Interestingly, Platt et al. [22] found RV and LV peak pressure 4 weeks post TAC. While LV pressure normalized thereafter RV pressure remained elevated. This is well in line with our observations 10 weeks post TAC.
Right heart failure due to chronic RV overload is the major cause of death among patients with PH [24]. PH is associated with high mortality and its treatment is still challenging [25]. Furthermore, there is an unfortunate lack of sufficient animal models for PH. Therefore, we suggest the characterization of the TAC mouse model as a suitable model for PH type 2 [22], which is secondary to left heart disease [26].

Result classification in the context of TAC induced HF and exercise intervention in mice
Numerous clinical and experimental studies showed that exercise improves cardiac structure and performance in ischemic cardiomyopathy [27][28][29], in systemic hypertension models [30,31], in myocardial infarction [7] and reduces the rate of re-infarction and mortality in patients with HF [5]. Exercise-induced cardioprotection is underpinned by various molecular mechanisms, such as activation of heat shock and ER stress proteins as well as cyclooxygenase, enhancing the antioxidant capacity and autophagy processes [32]. Several studies investigated cardiac and skeletal muscle performance using mild to severe TAC with or without exercise in mice [8,18]. Studies on TAC and exercise intervention always applied exercise as secondary intervention. Independently of the different starting times for exercise intervention (0-14 days post TAC) in the different studies, TAC was established at any time post-surgery, but no HF characteristics were verifiable at the exercise starting time.
Nine weeks of swimming exercise with an increasing duration were reported to induce beneficial cardiac effects in severe TAC operated mice [8]. LV mass was lower in comparison to TAC SED while EF and FS were improved, indicating at least a partial recovery of cardiac output. Further, cardiomyocyte CSA, amount of fibrosis, catecholamines, nitric oxide and reactive oxygen species levels were improved in this study [8]. Correspondingly, improved LV mass, EF, FS, cardiomyocyte CSA, atrial and brain natriuretic peptide levels were reported following 8 weeks of treadmill running with increasing intensity in mice following severe TAC [18]. Due to these studies, we expected an improvement of HF-induced cardiac and skeletal muscle function following exercise training.
Instead, we observed more pronounced cardiac hypertrophy and thinning of the RV wall in TAC EX . Diastolic as well as systolic LV volume and wall thickness in TAC EX were similar to TAC SED , indicating no beneficial effects of moderate exercise on cardiac remodeling or output. Exclusively, coronary artery remodeling induced by pressure overload may be at least partially reversed or prevented by exercise training.
Therefore, we comprehensively compared our data with the current literature on TAC-induced HF with a focus on exercise intervention or muscle function (    Left ventricular (LV) hypertrophy normalized to sham operated mice-simplified overview to allow for comparisons as some publications reported LV mass, LV mass/bodyweight, heart mass/tibial length or total heart weight (own data) n.d not declared, EF ejection fraction, FS fractional shortening, CSA cross-sectional area, ETT exercise tolerance test, ANP Atrial Natriuretic Peptide, BNP Brain Natriuretic Peptide a Value estimated from figure as not indicated otherwise patients [6]. This is in strong contrast to the increasing intensity exercise regimes that were done by other groups [8,18]. Van [16,33]. Both found higher LV mass and increased cardiomyocyte CSA four and up to 18 weeks post TAC [16], overall indicating a deterioration of cardiac performance in the course of time. Richards et al. performed exercise tolerance test at 3, 7, 11 and 15 weeks post 27G TAC and detected increasing exercise intolerance accompanied by decreasing EF in a time dependent aggravation [34]. Sung et al. observed decreased treadmill running distances 3 and 5 weeks post 27G severe TAC indicating exercise intolerance in HF mice [35]. They observed deteriorated LV mass, EF and FS [35]. Additionally, van Deel et al. examined the effects of mild exercise training on 27G severe TAC, observing a further increased LV mass as well as decreased FS down to 14% [6]. This is in accordance to our data, as we observed further aggravated cardiac performance following exercise intervention [6,7,34]. Neither LV mass nor EF, FS or fibrosis development were improved by exercise intervention, but rather additionally aggravated. Interestingly, Wang and Tian found intense exercise intervention to have beneficial effects on 27G severe TAC [8,18], although van Deel et al. found mild exercise in 25G mild TAC to already deteriorate cardiac function [6]. The underlying cause for these contrasting results remain unclear. One limitation to the studies by Wang and Tian is that they didn't test for maximum VO 2 consumption before designing their exercise protocol [8,18]. Instead they increased exercise intensity and duration without adapting the protocol according to individual capability of the HF mice in the first place. Wang et al. declare no mice were floating during swimming exercise [8], but did not discuss fear-evoking, psychological effects [36] of swimming exercise to the animals. Tian et al. did not mention, if mice dropped out during treadmill running [18]. We think careful observation of the HF mice during exercise training is essential, because progressing HF results in increased mice dropouts due to exhaustion. This is in accordance with EDL contraction force of TAC EX showing no significant difference from Sham or TAC SED . One reason for this observation may be the severely diminished perfusion of the skeletal muscle due to increased peripheral vasoconstriction in HF. This in turn is caused by the impaired cardiac output [37].
One pitfall in the studies of Richards [34] and van Deel [6] is the lack of histological or molecular analyses of skeletal muscle alterations of TAC operated mice with and without exercise training. Our data show no beneficial influence of moderate exercise training on skeletal muscle structure or function. To our knowledge, the study by Sung et al. is the only one to assess metabolic alterations in skeletal muscles 5 weeks post TAC. They found a reduced activity in the insulin pathway in skeletal muscles resulting in reduced whole body glucose consumption, increased fatty acid oxidation in EDL and soleus, but no alterations in force production of tibialis anterior and soleus muscle [35]. Especially the absence of HF-induced peripheral muscle strength loss [35] is in accordance to our study [6,33,34].
All other studies point out that increasing TAC severity is associated with worsening cardiac performance defining the HF phenotype. While in humans HF induces skeletal muscle atrophy [3], we could not observe this effect in mice. Additionally, we and others observed that moderate exercise further deteriorated cardiac function and promoted fiber atrophy in the TAC mouse model.
The data of the current study suggest that TAC in mice in combination with moderate exercise training is not suitable to ameliorate HF associated cardiac dysfunction.
Exercise itself leads to temporarily increased blood pressure, which physiologically can be regulated via peripheral impedance in healthy training individuals. In active muscles, resistance vessels relax to increase blood flow in order to meet the muscles metabolic requirements [38]. Outside the active muscle systemic vascular resistance decreases, which finally leads to an increased heart rate. The LV output is increased due to ensure proper nourishment of the active muscle [38]. In TAC operated mice systolic and diastolic LV filling pressure is significantly increased alongside with an increased vascular stiffness, vascular resistance as well as increased blood pressure and LV filling pressure [39]. These factors fix systemic vascular and LV load at a high level [39]. This condition restricts the capability to regulate blood pressure during exercise [6].
Finally, since physical activity is capable of correcting structural and metabolic alterations in the skeletal muscle of HF patients [4][5][6][7][8], the TAC mouse model especially in combination with exercise intervention does not mirror this.

Conclusion
TAC is a solid model to study the effects of LV pressure overload in C57BL/6J mice. Our data clearly indicate the suitability of the 27G TAC technique to introduce HF characteristics in mice. Cardiac and skeletal muscle remodeling and function is not ameliorated but rather enhanced by exercise intervention in the 27G TAC mouse model. Other TAC phenotypes (mild/25G or moderate/26G) may affect skeletal muscle structure and performance to a different degree, but this remains unclear.
We observed that TAC alters RV structure in addition to LV remodeling, which indicates development of a biventricular dysfunction. Thus, TAC in mice may be considered as an appropriate model for type 2 PH.
In summary, moderate exercise training intervention in the 27G TAC mouse model is not suitable to study beneficial exercise effects on cardiac remodeling and peripheral cachexia in HF.

Animals, TAC surgery, and sacrifice
All experiments in this study were approved by the local Animal Research Authority, in accordance with the European Directive (TVV 51/18). A total of 45 male C57BL/6J mice (Janvier Labs, Le Genest-Saint-Isle, France) entered the study (see Fig. 4 for study course). The 6 weeks old mice were randomized into two groups: sham-operated control (n = 15) and TACoperated group (n = 30). A second randomization of the TAC-operated group was done 2 weeks after TAC surgery: TAC + exercise (n = 15), TAC + sedentary (n = 15).
Eight-week old mice weighing 19-24 g, were subcutaneously administered with 5.0 mg/kg carprofene for preoperative analgesia. The mice were sedated by intraperitoneal injection of 0.5 mg/kg Medetomidin, 5 mg/ kg Midazolam und 0.05 mg/kg Fentanyl. The chest was opened through a lateral thoracic window from the throat to the second rip. A 27G needle was placed on the transverse aorta and secured using 6.0 prolene to reduce aortic diameter to 0.4 mm. The needle was removed immediately and the chest was closed using 6.0 prolene. The sham-operated control group underwent the same procedure but without binding of the aorta. Post-operative analgesia comprised of 0.1 mg/ kb buprenorphine 3× per day, 5.0 mg/kg carprofene 1× per day and additional buprenorphine supply in the drinking water for 2-3 days in total. Mice were weighted and sacrificed 1 day after the last training unit using intraperitoneal injection of 5 mg/kg xylazinhydrochloride, 100 mg/kg ketaminhydrochloride and 0.1 mg/kg atropinsulfate. The hearts were separated, weighted and immediately stored in liquid nitrogen whereas the medial portion of the heart was fixed in 4% paraformaldehyde. Left extensor digitorum longus was removed after ligating it's proximal tendon using silk braid black, followed by dissecting the muscle without pressure or dragging and finally binding the distal EDL tendon for muscle functional measurement. Muscle preparation was conducted in carbogen (95% O 2 , 5% CO 2 )-flushed Krebs-Henseleit-Buffer (120.5 mM NaCl, 4.8 mM KCl, 1.2 mM MgSO 4 , 1.2 mM NaH 2 PO 4 , 20.4 mM NaHCO 3 , 1.6 mM CaCl 2 , 10 mM dextrose, 1 mM pyruvate, pH = 7.4).

Echocardiographic measurement of cardiac function in the developing HF
For echocardiography (Vevo 770, Visual Sonics, Canada) mice were anesthetized with 5 vol% isoflurane followed by a 2 vol% isoflurane flow for maintaining anesthesia under continuous ventilation. Baseline measurements were performed at 7 weeks old mice (t = − 1 week) and again at an age of 10 weeks (t = 2 weeks) and 18 weeks (t = 10 weeks). Systolic and diastolic LV thickness, volume, ejection fraction and fractional shortening in control, HF and HF training groups were measured at the indicated time points (see Fig. 4).

Exercise protocol
The TAC EX group started exercise intervention 2 weeks after TAC operation five times a week for a total of 8 weeks. Maximum VO 2 max was determined using an exercise tolerance test and a moderate training regime was defined at 60%. According to the previously performed exercise tolerance test to estimate VO 2 max, a training schedule was applied (see Table 3). Treadmill speed was adapted to the group's slowest individuum regarding the exercise tolerance test. Moderate training was performed in a treadmill at an inclination of 20° including an initial 10 min warm up and final cool down at 40% of the maximum performance and a 40 min training phase at 60% of the maximum performance in between.

Histological evaluation of TAC-induced hypertrophy and fibrosis
Cardiac hypertrophy was evaluated by analyzing crosssections (3 µm) were stained with hematoxylin and eosin. Area, circumference, as well as the thickness of the left and right ventricular wall were determined using ImageJ (v1.8, NIH, USA). Data are shown in relation to sham. Sirius Red staining was applied for identification and quantification of collagenous fibrotic tissue. For analysis of heart metrics ObjectJ PlugIn (Vischer, N., The Netherlands) was used to measure each ventricle wall 5 times at 5 positions and a polygonal region drawn around the cross section to calculate the heart area and circumference. For fibrosis estimation a threshold for Sirius Red staining intensity was applied and the proportion of interstitial and perivascular fibrosis was calculated in relation to the area of the whole section.

Muscle function
Left EDL muscle function was determined as described [40]. In brief, EDL was fixed on a force transducer (Aurora Scientific, Aurora, Canada) and equilibrated in an organ bath (Aurora Scientific, Aurora, Canada) at 37 °C in carbogen-flushed Krebs-Henseleit-Buffer for 15 min. To measure EDL force production the muscle was stimulated by an impulse of 700 mA and 0.25 ms with increasing frequencies 1, 15, 30, 50, 80, 120, 150, 300 Hz followed by a one minute rest period using a bipolar high performance stimulator (Aurora Scientific, Aurora, Canada). The tension generated by the EDL muscle contraction is recorded in gram (g) as a function of frequency.

Slow-fast muscle fiber staining
Serial formalin-fixed and paraffin-embedded EDL and Soleus slices (3 µm) were stained for histological analyses of cross-sectional area. First muscle tissue slices were deparaffinized in 3 × 10 min xylol followed by a descending ethanol series hydration. Antigen retrieval was conducted in target retrieval solution (Dako, Santa Clara, USA). After cooling to room temperature, the slices were washed with deionized water and treated with (

Data processing and statistics
Heart weight was normalized to body weight, or to tibial length to compensate potential differences in BW. Statistical analyses were performed using Graph-PadPrism6 (San Diego, CA, USA). Data are presented as mean ± standard error of the mean (SEM). Shapiro-Wilks-Test was performed to check for normal distribution. Group differences were analyzed using one-way ANOVA or Dunnett's multiple comparison. Muscle function (force-time-relationship) was assessed by two-way repeated measures ANOVA. Bonferroni correction was used as posttest. All p-values < 0.05 were considered statistically significant.