One of the most fatal complications of diabetes is diabetic cardiomyopathy (DCM) [1]. DCM is defined as cardiac muscle dysfunction caused by diabetes, independent of hypertension and atherosclerosis [2]. It is estimated that 40 to 50 percent of diabetic patients suffer from heart disorders [3]. Although the exact induced mechanisms of DCM are unknown, the inflammation process, apoptosis, hypertrophy, and fibrosis contribute to the development of cardiac remodeling in diabetes [2]. Oxidative stress is caused by chronic hyperglycemia (the main complication of diabetes) and can lead to ventricular contractile dysfunction [4]. The elevation of reactive oxygen species (ROS) might activate the renin-angiotensin-aldosterone-system and signaling pathways like transforming growth factor beta (TGF)-$\beta$, which leads to hypertrophy and fibrosis in the diabetic heart [3, 5]. Cardiac fibrosis might reduce myocardial adjustment, resulting in failure of diastolic and systolic functions and, eventually, in cardiac contraction and pumping activity [6]. TGF-$\beta$1/Smad signaling has been shown to be an enhancer of collagen synthesis and interstitial fibrosis progression [6]. In addition, pathological hypertrophy of the heart following cardiac fibrosis is associated with the activation of fetal genes like brain natriuretic peptide (BNP) and atrial natriuretic peptide (ANP) and that enhances lengths or widths of cardiomyocytes and contributes to the development of DCM [3]. Mitochondrial dysfunction in various organs could induce cardiomyocyte energy insufficiency, reducing mitochondrial respiratory oxygen consumption [7]. Peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1$\alpha$) is pivotal in mitochondrial performance. The coactivator action of PGC-1$\alpha$ improves biogenesis, respiration, and transcriptional activation of mitochondria and reduces ROS generation and inflammatory processes in the vascular smooth muscle endothelial cells. PGC-1$\alpha$ levels depend on glycemic control and insulin sensitivity in skeletal muscle and myocardium [8]. AMP-activated protein kinase (AMPK) preserves ATP by retrieving the ratio of NAD+/NADH and excluding catabolic pathways, which will be turned on by reducing intracellular ATP levels [9]. Along the progression of DCM, the expression of some genes like the Ca${}^{\mathrm{2}\mathrm{+}}$ ATPase pump of the sarcoplasmic reticulum (SERCA) and ryanodine receptors (RyR), which recreate an influential contribution in the normal operation of the heart muscle, might be disrupted [10].

Physical activity as a thrifty and non-pharmacological co-treatment is recommended as part of the rehabilitation for diabetic patients with cardiac disorders. Exercise can improve cardiac function by decreasing cardiac risk factors and myocardial damage in diabetes [11]. Furthermore, exercise could enhance mitochondrial biogenesis, ATP production, and cardiomyocyte contractility [12]. Physical activity has been shown to promote angiogenesis and vascular performance by attenuating oxidative stress and inflammatory processes [13]. The impact of exercise on diabetes-induced cardiovascular disorder might be related to the modalities, duration, and intensity level of training programs. The evidence showed that low and moderate-intensity exercise had been indicated to improve the metabolism of glucose and cellular apoptosis in the heart, which results in enhanced cardiac function [11].

Moreover, a cardiotonic and protective role for high-intensity exercise training has been demonstrated [11]. Both moderate-intensity continuous training (MICT) and high-intensity interval training (HIIT) modalities of exercise have been shown to decrease the glycosylated form of hemoglobin and fasting blood glucose (FBS) in patients with diabetes [13, 14]. However, HIIT training showed better outcomes in reducing FBS and ameliorating hyposensitivity to insulin in adults with diabetes [15, 16]. Furthermore, HIIT attenuated glucose and fatty acid metabolism, the respiratory capacity of mitochondria, and ventricular mechano-energetic coupling in cardiac tissue. At the same time, the MICT exercise could not improve this parameter [17, 18]. Exercise training in diabetes improves myocardial fibrosis by decreasing myocardial collagens and restoring fibrosis-related gene expression of matrix metalloproteinases [6]. Significant protective effects of exercise training in DCM have been revealed by improving PGC-1$\alpha$ and Akt signaling pathways and mitochondrial performance in cardiomyocytes in diabetes [19]. Activation of AMPK and improvement of glucose metabolism following exercise training are mentioned in animal and clinical trial studies [9]. Regulation of RyR and SERCA, as the main contributors in the removal of intracellular calcium, improved with exercise training [10]. Due to its hypoglycemic effect, metformin is one of the most commonly used oral first-line drugs in managing diabetes-induced hyperglycemia. It has also been demonstrated to have cardioprotective effects by diminishing the heart complications of diabetes. For these reasons, it was considered a positive control and was used simultaneously with exercise training [3, 20]. Altogether, less is known about the effects of either of these two exercise modalities on diabetes-induced cardiac fibrosis and hypertrophy. We hypothesized that the HIIT and MICT exercise may prevent pathological cardiac remodeling in diabetic rat models induced by streptozotocin (STZ). Therefore, the expression of ANP and BNP genes (as markers of cardiac hypertrophy), TGF-$\beta$and collagen genes (as markers of cardiac fibrosis), RyR and SERCA genes (as an indicator of intracellular calcium homeostasis), PGC-1$\alpha$ and AMPK genes (as an indicator of mitochondrial function) and echocardiography parameters were assessed to clarify the preventive effects of exercise on DCM.

Although cardiomyopathy caused by diabetes does not have a precise and proven mechanism, pathological cardiac remodeling caused by inflammatory processes, fibrosis, and hypertrophy could be a principal cause of the development of this disorder [2]. The pathological cardiac remodeling due to fibrosis is induced by extracellular matrix accumulation and upregulation of TGF-$\beta$ as the main profibrotic factor. TGF-$\beta$1 could be activated by oxidative stress, following binding to its receptors and triggering collagen production through the smad signaling pathway [4]. Due to chronic hyperglycemia in diabetic patients, up-regulation of the fetal proteins related to hypertrophy, such as ANP and BNP in the cardiomyocytes, increases dramatically [24]. Moreover, the disturbed mitochondrial performance is involved in the DCM pathophysiological mechanisms, which could be regulated with PGC-1$\alpha$ [19]. Other signaling pathway activations like the AMPK in response to energetic demand and calcium-based myocardial performance mediated by RyR2 and SERCA2 are among the implicated parameters in DCM [10, 25]. The exercise protocols might alleviate cardiovascular complications and mitigate DCM incidence by preventing pathological cardiac remodeling.

Nevertheless, we studied the impact of two exercise modalities in preventing cardiac complications following diabetes. Metformin, a pharmacological diabetic treatment with a cardioprotective effect, was considered a positive control [24]. We observed that exercise with or without metformin and improving hyperglycemia inhibited fibrosis and hypertrophy by modifying the expression of target genes contributing to this process. Echocardiographic and histological data also confirmed this alleviation. Likewise, the improvement in the expression of genes that contributed to mitochondrial function and homeostasis of intracellular calcium after exercise treatment was in favor of the positive effect of exercise in inhibiting pathological cardiac remodeling in the course of diabetes.

As in previous similar experimental and clinical investigations, FBS levels in diabetic rats were significantly reduced following exercise training, although this reduction was more pronounced in the Met-HIIT group [23, 26, 27, 28, 29]. However, another study indicated the beneficial effect of MICT in lowering blood sugar in diabetic rats [13]. Exercise training might restore hyperglycemia by enhancing the muscle blood flow and ameliorating mitochondrial function. Additionally, its mechanism may be related to enhancing insulin sensitivity, which might improve glucose uptake [22, 29, 30].

Cardiac fibrosis might disturb myocardial adjustment and cardiac contraction, which leads to cardiomyopathy development and heart failure [6]. Here, we revealed that the exercise alleviated the expression of TGF-$\beta$ and collagen genes in the diabetic group; the best response belonged to the Met-HIIT-treated rats. Likewise, the enhancement in these fibrotic factors has been reported in different study models of diabetic hearts and cardiac infarction [4, 6, 13, 31]. The TGF-$\beta$1/Smad signaling pathway is related to collagen synthesis and interstitial fibrosis progression. Nonetheless, alleviated TGF-$\beta$ and collagen gene expression in treated animals might reflect that exercise training may mitigate fibrosis in cardiac tissue by inhibiting the molecular mechanism such as renin-angiotensin-aldosterone-system that leads to the prevention of myocardial collagen deposition and consequently cardiomyopathy in diabetes [4, 6]. The modulation of TGF-$\beta$ and collagen gene expression was compatible with changes in the histopathological pattern associated with fibrosis. According to our histopathological evaluation and consistent with previous evidence, it can be supposed that exercise training intervention in diabetic cases can act as an anti-fibrotic factor by reducing collagen accumulation [6, 27, 32].

Pathological hypertrophy following cardiac fibrosis is associated with the over-expression of fetal hypertrophic genes that enhance the lengths or widths of cardiomyocytes and develop DCM [3]. Improvement of cardiac hypertrophy by reversing ANP and BNP gene expression has been indicated after exercise conditioning in diabetic animals [33]. In this regard, we observed that in the Met-HIIT group, the expression of these genes decreased more and became closer to the control group. The changes in the echocardiographic data mirrored this. Lowered expression of the hypertrophy hallmark genes, in addition to restoring the echocardiographic indices of LVEDD and LVESD in exercise groups, indicate that exercise training might prevent DCM by improving cardiac hypertrophy in the course of diabetes.

Nevertheless, exercise conditioning has been demonstrated to fail to boost the cardiac hypertrophy of diabetic db/db mice in a study, but metformin administration reversed cardiac hypertrophy. Consequently, cardiac hypertrophy in the early stages of cardiomyopathy is more related to glucose metabolism abnormality [34]. It is worth noting that in this work, exercise training prevented weight loss and decreased HW/BW index in diabetic animals. It has been shown that exercise might improve glucose metabolism and prevent diabetic ketoacidosis and weight loss by increasing the skeletal muscle and liver response to insulin and reducing fat oxidation [16, 29]. The effectiveness of exercise training may improve cardiac performance by attenuating fibrosis and hypertrophy in the heart, which results in the prohibition of cardiac disturbance induced by diabetes. It also can be a reasonable justification for these changes, as mentioned by previous studies [6, 13, 16].

Mitochondrial dysfunction in myocardial cells is another complication caused by diabetes. The level of PGC-1$\alpha$ gene expression, as a pivotal element contributing to mitochondrial function, has been decreased in heart failure [8, 19]. The ability of exercise training as an ameliorative factor for this cellular function by upregulating the expression of the PGC-1$\alpha$ has been previously reported [8, 19, 35, 36]. We also observed over-expression of this gene in exercise-treated groups. Meanwhile, the expression was higher in the combined group of Met-HIIT. PGC-1$\alpha$, as a primary modifier in the biogenesis of mitochondria, has a potential role in the energy metabolism of the myocardium. The researchers indicated that exercise training improves DCM through ameliorating cardiac performance associated with restoring mitochondrial biogenesis, accompanied by PGC-1$\alpha$ activation and Akt signaling [19]. AMPK is an essential cellular pathway turned on due to bio-energetic demand and physical activity [25]. Aerobic exercise might increase AMPK gene expression and phosphorylation in diabetic animals [33]. The expression level of the AMPK gene was considerably increased in exercise groups compared to diabetic animals, especially in the Met-HIIT group. The preventive role of exercise against DCM in diabetescould be increasing AMPK associated with downregulating forkhead box transcription factors 1 (FOXO1) as a downstream effector of AMPK [33]. These changes in gene expression related to mitochondrial function, which improves cellular energy supply, have been accompanied by increased EF and FS performance indicators. During the progression of diabetes, heart diastolic dysfunction in the absence of hypertension can inevitably confirm the disorder of intracellular calcium homeostasis [34]. Severe systolic dysfunction was also demonstrated in diabetic patients, which might be related to the alteration of gene expression involved in adjusting intracellular calcium homeostasis. The changeover of calcium-based activities such as RyR2 and SERCA2a sensitivity change in diabetesis somewhat responsible for the myocardial contraction disturbance [10].

Modulating RyR and SERCA mRNA expression and their protein levels following endurance exercise protocols, which were reduced in diabetic rats, may also improve systolic and diastolic dysfunction in diabetic cardiomyopathy [10, 37]. Our data did not show a significant change in the expression levels of the RyR gene in all of the treated groups; meanwhile, in the Met-HIIT treated animals, there was an increase in the expression level of the SERCA gene. This upregulation could be attributed to the impact of exercise conditioning in enhancing cardiac performance in terms of echocardiographic parameters EF and FS.

Moreover, we showed a significant correlation between the gene expression level of ANP as an index of hypertrophy in cardiac tissue and TGF-$\beta$ as a fibrotic index within the echocardiographic indices of EF and FS, which indicated a close association between the expression level of the mentioned genes and cardiac function. This significant relationship emphasizes the positive effect of simultaneous administration of metformin and HIIT exercise treatment on improving the prevention of pathological cardiac remodeling.

It should be mentioned that our findings cannot suggest which type of exercise is more effective for diabetes management. According to the effectiveness of exercise type, some parameters, including EF and FS in the echocardiography and the expression of the AMPK gene, were improved more in the HIIT-exercised rats than the MICT one, at the same time, reduced the percentage of collagen content in the MICT-exercised animals than the HIIT-exercised group. Controversial findings were also mentioned in previous studies [29, 38].

Considering that the progression of fibrosis and hypertrophy are the most critical mechanisms in the development of cardiomyopathy and diastolic dysfunction in diabetic patients, it seems that recommending physical exercises in addition to common diabetes treatments can prevent or delay pathological remodeling of the heart.

The data showed that exercise training, especially in combination with metformin, in addition to improving hyperglycemia, prevents cardiomyopathy in diabetic rats by attenuating cardiac hypertrophy and fibrosis and maintaining mitochondrial function and intracellular calcium homeostasis (Fig. 7). These effects were observed more prominently in the Met-HIIT group, related to the type of exercise training protocols. Our findings increase our understanding of the benefits of physical activity on diabetes-induced cardiovascular disease and provide a practical target for DCM prevention. These results showed that exercise, especially with an anti-diabetic drug such as metformin, can be included in the rehabilitation therapy of diabetic patients with cardiovascular complications.

Data will be made available on request, and the corresponding author can be contacted if needed.

SS and MF performed the experiments, and MM designed the study, contributed to data acquisition, analysed and interpreted the data, revised the manuscript and supervised the project advancement. ZME, ZG, and SN assisted in data collecting and interpreting. SS, MF and ZME made the manuscript draft, and MM, ZG and SN critically reviewed it. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The Mashhad University of Medical Sciences Ethics Committee approved the experiments’ protocols. The experimental process was promoted according to the caring standard for laboratory animals (Approval No. IR.MUMS.MEDICAL.REC.1399.224).

The authors would like to express their acknowledgment to the Mashhad University of Medical Sciences Research Affairs for their support. This manuscript is extracted from Sadegh Shabab’s Ph.D. thesis in Medical Physiology. Dr. Narges Kasraie made edits and revisions.

Mashhad University of Medical Sciences Research Affairs; Grant No. 981872.

The authors declare no conflict of interest.