Effects of Low-Intensity and Long-Term Aerobic Exercise on the Psoas Muscle of mdx Mice: An Experimental Model of Duchenne Muscular Dystrophy

Duchenne muscular dystrophy (DMD) is a muscle disease characterized by the absence of the protein dystrophin, which causes a loss of sarcolemma integrity, determining recurrent muscle injuries, decrease in muscle function, and progressive degeneration. Currently, there is a need for therapeutic treatments to improve the quality of life of DMD patients. Here, we investigated the effects of a low-intensity aerobic training (37 sessions) on satellite cells, peroxisome proliferator-activated receptor-gamma coactivator (PGC)-1α protein (PGC-1α), and different types of fibers of the psoas muscle from mdx mice (DMD experimental model). Wildtype and mdx mice were randomly divided into sedentary and trained groups (n = 24). Trained animals were subjected to 37 sessions of low-intensity running on a motorized treadmill. Subsequently, the psoas muscle was excised and analyzed by immunofluorescence for dystrophin, satellite cells, myosin heavy chain (MHC), and PGC-1α content. The minimal Feret’s diameters of the fibers were measured, and light microscopy was applied to observe general morphological features of the muscles. The training (37 sessions) improved morphological features in muscles from mdx mice and caused an increase in the number of quiescent/activated satellite cells. It also increased the content of PGC-1α in the mdx group. We concluded that low-intensity aerobic exercise (37 sessions) was able to reverse deleterious changes determined by DMD.


Introduction
Duchenne muscular dystrophy (DMD) is the most common and severe muscle disorder caused by a mutation in chromosome X (region Xp 21.2) that leads to an absence of the protein dystrophin [1]. DMD affects one in every 5000 boys, and clinical manifestations appear in childhood, causing muscle weakness and the loss of the ability to walk, wheelchair confinement, and respiratory and cardiac failure [1,2]. Dystrophin is part of a protein complex and provides stability to muscle fibers during muscle contractions. With the absence of dystrophin, lesions in the sarcolemma occur, resulting in weakness and loss of muscle function [3,4].
The psoas is a glycolytic muscle, predominantly composed of type II fibers (FTII), and one of the most susceptible to injuries in DMD [5,6]. The preferential involvement

Morphological Characteristics of the Dystrophic Muscle Improved after Low-Intensity Exercise
We used an immunofluorescence assay to confirm the absence of the dystrophin protein in mdx animals ( Figure 1). We observed the presence of dystrophin in the sarcolemma of psoas fibers in wildtype mice (green color, Figure 1A) and the absence of the protein in the muscle of mdx mice ( Figure 1B), as expected.
A semiquantitative morphological analysis of the psoas muscle stained with hematoxylin and eosin (HE) showed a healthy muscle structure, with polyhedral fibers, peripheral nuclei, and no pathological changes in the WT-SED and WT-TR groups (Figure 2A,B). However, we observed variation in fiber size, centralized nuclei (an indicative of muscle degeneration and regeneration), basophilic fibers, splitting, necrosis, increased connective tissue, and inflammatory infiltrate in the mdx-SED group ( Figure 2C). The mdx-TR group presented similar alterations, but they were less expressive, which resulted in an improvement in the muscle cytoarchitecture ( Figure 2D, Table 1). A semiquantitative morphological analysis of the psoas muscle stained with hematoxylin and eosin (HE) showed a healthy muscle structure, with polyhedral fibers, peripheral nuclei, and no pathological changes in the WT-SED and WT-TR groups (Figure 2A,B). However, we observed variation in fiber size, centralized nuclei (an indicative of muscle degeneration and regeneration), basophilic fibers, splitting, necrosis, increased connective tissue, and inflammatory infiltrate in the mdx-SED group ( Figure 2C). The mdx-TR group presented similar alterations, but they were less expressive, which resulted in an improvement in the muscle cytoarchitecture ( Figure 2D, Table 1).   Wildtype mice (control); dystrophin present in green. (B) mdx mice (dystrophic); dystrophin absent. Cell nuclei were stained with DAPI (blue). Images were obtained with the ImageXpress XLS System microscope at 20× magnification. Scale bar = 100 µm.
A semiquantitative morphological analysis of the psoas muscle stained with hematoxylin and eosin (HE) showed a healthy muscle structure, with polyhedral fibers, peripheral nuclei, and no pathological changes in the WT-SED and WT-TR groups (Figure 2A,B). However, we observed variation in fiber size, centralized nuclei (an indicative of muscle degeneration and regeneration), basophilic fibers, splitting, necrosis, increased connective tissue, and inflammatory infiltrate in the mdx-SED group ( Figure 2C). The mdx-TR group presented similar alterations, but they were less expressive, which resulted in an improvement in the muscle cytoarchitecture ( Figure 2D, Table 1).

Low-Intensity Exercise Promotes an Increase in the Number of Satellite Cells in the Dystrophic Muscle
Quiescent/activated SCs were identified using the Pax-7 marker ( Figure 3A). The WT-SED group had few quiescent/activated SCs when compared to the mdx-SED group (WT-SED × mdx-SED, p < 0.05) ( Figure 3B). After training, we observed an increase in the SC number in mdx mice when compared to the respective sedentary group (mdx-SED × mdx-TR, p < 0.05) ( Figure 3B).   Additionally, SCs in differentiation/fusion were identified using F5D, a myogenin marker ( Figure 4A). As expected, the WT-SED group had fewer SCs in differentiation/fusion when compared to the mdx-SED group (WT-SED × mdx-SED, p < 0.05) ( Figure 4B).

PGC-1α Is Increased in Dystrophic Muscles
Fluorescence measurements for the PGC-1α protein showed that DMD increased the oxidative characteristics of the psoas muscle, as we observed a higher amount of the protein in the mdx-SED group when compared to the WT-SED group (WT-SED × mdx-SED, p < 0.05) ( Figure 5A,B). As expected, the low-intensity aerobic training, applied for a long period, increased the amount of PGC-1α in both trained groups (WT-TR and mdx-TR) when compared to their respective sedentary groups ( Figure 5A,B).

A Long Period of Low-Intensity Aerobic Exercise Increases the Trophism of Fibers in mdx Mice
Fiber diameter analyses showed a significant reduction in the trophism of FTIIA, FTIID, and FTIIB fibers of the psoas muscle of the mdx-SED group when compared to the animals in the WT-SED group (WT-SED × mdx-SED, p < 0.05) ( Table 2). However, after 37 training sessions, it was possible to observe an increase in FTIIB and FTIID diameters in the mdx-TR group when compared to the sedentary one (mdx-SED × mdx-TR, p < 0.05) ( Table 2). On the other hand, in WT animals, training reduced the diameter of FTIIB and FTIIDB/BD when compared to the WT-SED group (WT-SED × WT-TR, p < 0.05) ( Table 2).

Low-Intensity Exercise Did Not Affect the Fiber Type Distribution
As expected, we observed a predominance of FTIIB fibers in all studied groups, as the psoas is a glycolytic muscle characterized by the dominance of this fiber type (Figures 6 and 7). The presence of hybrid fibers (FTIIAD/DA and FTIIDB/BD) was also verified (Figures 6 and 7). However, 37 training sessions were not able to change the number of different fiber types.
As expected, we observed a predominance of FTIIB fibers in all studied groups, as the psoas is a glycolytic muscle characterized by the dominance of this fiber type (Figures 6 and  7). The presence of hybrid fibers (FTIIAD/DA and FTIIDB/BD) was also verified (Figures 6  and 7). However, 37 training sessions were not able to change the number of different fiber types.

Discussion
In the present study, we confirmed that the absence of dystrophin in mdx mice causes

Discussion
In the present study, we confirmed that the absence of dystrophin in mdx mice causes negative alterations in the psoas muscle. Most importantly, after 37 sessions of low-intensity aerobic training on a treadmill, we observed an improvement in several characteristics of the dystrophic muscle, an indicative that physical exercise, when applied correctly, can be an important and promising noninvasive therapeutic tool to treat DMD. We observed a decrease of basophilic cells, a decrease in inflammatory markers, and few necrotic areas in the muscle of trained mdx mice, which indicates that the exercise improved the general characteristics of the muscle cytoarchitecture. Our data corroborate several studies that also observed an improvement in the morphology of mdx mice muscles after a low-intensity exercise training [24,25].
A better understanding of the regenerative capacity of the mdx mice muscle can provide important information about the disease; thus, it is important to investigate the role of satellite cells in muscle regeneration, from the quiescence and activation phases to the differentiation and fusion phases, where the process of repair and formation of new myofibers takes place [9,10]. We analyzed SCs through Pax7 (quiescence-activation) and myogenin (differentiation-fusion) fluorescent markers. As expected, we observed a large number of SCs in the psoas of the sedentary mdx animals (Pax-7 + ; myogenin + ), a consequence of the chronic muscle damage that leads to numerous degeneration-regeneration cycles, activating SCs for repair [26,27]. This high number of SCs suggests that these cells are still able to participate in muscle repair, although they are dysfunctional due to the absence of dystrophin [28]. Ribeiro Jr and colleagues (2019) demonstrated that dystrophic gastrocnemius muscles retain a pool of proliferating SCs that rapidly respond to regenerating stimuli and that the muscle repair is abnormal, but preserved by upregulation of myogenic factors such as myogenin, a result that corroborates our observations [29].
It is known that aerobic exercise activates SCs and improves their function in healthy muscles, both in humans and in rodents [21,30,31]. Our results showed that dystrophic mice, after low-intensity training for 37 sessions, had an increase in SC activation (Pax7 + ) when compared to the sedentary animals. Interestingly, we did not observe an increase in differentiated SCs (Myog) after the applied exercise protocol, which suggests that some fibers were already repaired after training (low expression of myogenin by SCs). Myogenin is a muscle regulatory factor, temporarily expressed during differentiation; thus, data interpretation is temporally limited [32,33]. Yablonka-Reuveni and Anderson (2006) also reported an acceleration in the differentiation of SC in mdx mice when compared to normal muscles, a result that was supported by the significant improvement in the cytoarchitecture of the muscle tissue [34], which corroborates our findings.
We also demonstrated that the disease increased the oxidative characteristics of dystrophic muscles, as we observed high amounts of PGC-1α in the sedentary mdx mice when compared to the wildtype animals. This was an unexpected result, considering that PGC-1α regulates mitochondria biogenesis, which helps to compensate for the metabolic dysregulation present in dystrophic muscles [35]. Moreover, skeletal muscles of DMD individuals have a reduced capacity for oxidative phosphorylation, in addition to mitochondrial dysfunction [22,36]. It is still unclear why we found an increased PGC-1α content in the psoas muscle of the sedentary mdx mice, and future studies are needed to elucidate this finding.
The low-intensity aerobic training (37 sessions) increased PGC-1α protein levels in both trained groups (wildtype and mdx trained animals) when compared to their respective sedentary groups. Studies have shown that physical exercise induces the expression of PGC-1α, mainly in chronically exercised muscles, and that overexpression of the protein increases mitochondrial biogenesis, improving the oxidative capacity in muscles [8,37,38]. In the case of mdx mice, the overexpression of PGC-1α can improve muscle function, reducing the negative effects of the pathology [21,[39][40][41]. It is known that PGC-1α induces utrophin expression, a protein homologous to dystrophin in dystrophic muscles, which compensates for its absence, being a promising strategy for the treatment of DMD [42,43]. Therefore, the increased PGC-1α content found in the psoas muscles of trained mdx animals in this study is an important and promising result that indicates the role of PGC-1α as a protective factor on dystrophic muscles.
DMD also led to a significant reduction in the diameter of pure fibers (FTIIA, FTIID, and FTIIB) in the psoas muscles of the sedentary mdx group (mdx-SED), suggesting the occurrence of an atrophic process during the disease progression, as reported in the literature [44,45]. After low-intensity training (37 sessions), we observed an increase in the diameter of FTIID and FTIIB fibers in the psoas muscles of the mdx animals (mdx-TR). This result has clinical relevance, helping to prevent the progression of muscle injuries, since glycolytic fibers (FTIIB, FTIID) can support mechanical loads imposed during muscle contraction [46]. It is well known that PGC-1α protects myofibers, and its overexpression increases the mitochondrial content and resistance to fatigue, reducing muscle atrophy in cases of denervation [47,48]. Although we did not perform an analysis to elucidate the association among fiber types, trophism, and PGC-1α protein expression, our data suggest that the increase in PGC-1α content had a positive effect on the trophism of glycolytic fibers. Consequently, we hypothesize that the augmented diameter of glycolytic fibers characterizes muscle protection in the psoas of the trained mdx mice. In addition, aerobic exercise also plays a role in preventing skeletal muscle atrophy and may serve as a prior treatment for atrophic situations, such as in DMD disorder, which presents intense atrophy over time [49].
The psoas of mice contains type IIA, IID, and IIB fibers unevenly distributed throughout the muscle, with a region containing only FTIIA, FTIID, and FTIIB fibers and a region composed almost exclusively of FTIIB [50][51][52]. Our results showed this predominance of FTIIB in all groups, as well as an unequal distribution of fibers. We also identified the presence of hybrid fibers (IIAD/DA and IIDB/BD), suggesting a combination of myosin heavy chains (MHC) that can undergo transformation from one fiber type to another (shifting) [53]. Glycolytic fibers are the most affected by DMD, undergoing more degenerative processes than oxidative fibers [54,55]. Lindsay and colleagues (2019) proposed that the resistance of oxidative fibers occurs due to the increased expression of utrophin [56], and Selsby and colleagues (2012) observed that overexpression of PGC-1α induces a shift from glycolytic (fast-twitch) to oxidative (slow-twitch) fibers in dystrophic muscles [57]. We know that physical exercise can increase the oxidative capacity of all myofibers, promoting a transition from more glycolytic to more oxidative fibers [58]. This shift has already been observed in different muscles of mdx mice after physical training [24,59,60]. However, studies using the psoas muscle of mdx mice are scarce. Although our results did not show statistical differences in the proportion of fibers between the trained and sedentary mdx animals, the decrease in the number of FTIIB fibers concomitantly with an increase in FTIIA suggests a positive adaptation of the psoas muscle. It is tempting to speculate that glycolytic muscles need mechanical stimulation of low intensity for a longer period than 37 sessions of training to present a significant shift in the proportion of muscle fibers.
One limitation of the present study is that we did not characterize the number of SCs in different fiber types, as it is known that oxidative fibers have more satellite cells than glycolytic cells [7,54,61]. However, given the high proportion of type II fibers in the psoas, we believe that our results are representative of what happens with DMD before and after training of the whole muscle.

Animals
This study was approved by the Ethics Committee on Animal Use of the Federal University of São Carlos-UFSCar (protocol number-CEUA n • 4740230518). Twelve wildtype mice (C57BL-10) and 12 mdx mice (C57BL-10; experimental model of DMD) were purchased from the Multidisciplinary Center for Biological Investigation on Laboratory Animal Science (CEMIB, UNICAMP, Campinas, Brazil). All animals were maintained in cages supplied with water and food ad libitum in an environment with a light-dark cycle (12 h/12 h) at 22 • C. Twenty-four male mice were used in this study. Twelve mdx mice (C57BL/10-Dmdmdx/PASUnib; body weight 18.33 ± 1.49 g) were randomly divided into two experimental groups: mdx-SED (sedentary mdx; n = 6) and mdx-TR (trained mdx; n = 6), and 12 wild-type mice (background: C57BL/10; body weight 19 g ± 0.0 g) were randomly divided into two experimental groups: WT-SED (sedentary wildtype; n = 6) and WT-TR (trained wildtype; n = 6).

Low-Intensity Training
Training sessions started when the mice completed 6 weeks. Before each session, all animals underwent a 2 min warm-up period, at a speed of 7 m/min. During the training period, they exercised at a speed of 9-10 m/min for 30 min. Current studies show that these values correspond to low-intensity exercise for mdx mice [18,62]. In total, 37 training sessions were carried out on a motorized treadmill (EP 132C; Insight, Ribeirão Preto, São Paulo, Brazil). The training was performed 3 days per week (Monday, Wednesday and Friday), until completing 37 training sessions (12 weeks and 1 day) to investigate muscle adaptations after a long period of training. Pedrazzani and colleagues (2021) showed that low-intensity training proved to be more effective when applied over longer periods [24]. The animals were sacrificed after the end of the 37th session (WT-SED, WT-TR, mdx-SED and mdx-TR) at 18 weeks old and 1 day, and the psoas muscle was excised and frozen in liquid nitrogen for histological and immunofluorescence observation.

Histology
Histological sections of frozen psoas muscles (6 µm of thickness) were obtained with a cryostat (Leica CM 1850 UV, Wetzlar, Hessen, Germany) at −25 • C. Hematoxylin and eosin (HE) stain was used to evaluate the morphological alterations of the muscle fibers of wildtype and mdx mice (centralized nucleus, basophilic cells, necrosis, and presence of inflammatory infiltrate). Images were captured using a light microscope Zeiss Vert.A1, software AxioVision Rel 4.8 (40x lens) (Zeiss, Jena, Thuringia, Germany), and a semiquantitative analysis was performed.

Immunofluorescence
Immunofluorescence was applied to quantify satellite cells during activation and differentiation, as well as PGC-1α protein, different isoforms of myosin heavy chain (MHC), and dystrophin protein.  [24]. The minimal Feret's diameters were measured with the minimum distance of the parallel tangents in opposing borders of all muscle fibers in each field. The proportions of fibers were analyzed in the same four fields where all fiber types were counted. The PGC-1α content was quantified from the fluorescence unit, and all satellite cells, located below the laminin staining underlying the nucleus, were counted. Using the same images, all muscle fibers were counted for calculation of the SC/fiber ratio. All analysis was done using the Image J software (version 1.52a, Bethesda, MA, USA) [63].

Statistical Analysis
Quantitative comparisons for immunofluorescence of Pax-7, myogenin, and PGC-1α were made through a one-way analysis of variance (ANOVA), followed by post hoc analysis using Bonferroni or Brown-Forsythe and Welch tests. The analyses were performed with GraphPad Prism (version 5.01 for Windows, San Diego, CA, USA). Data for the minimal Feret's diameter and the proportion of different fiber types were analyzed using mixedeffects linear models, and multiple comparisons with contrasts were performed using the SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA). The proportion of fibers was analyzed after a natural logarithmic transformation of the data to obtain homogeneous variability, consistent with a normal distribution [64]. Values of p < 0.05 were considered statistically significant for all analyses.

Conclusions
Low-intensity aerobic exercise applied for 37 sessions reversed some of the deleterious alterations determined by DMD, improving tissue morphology and fiber trophism through SC activation and increased content of PGC-1α protein in the psoas muscle of mdx mice. Our results suggest that a chronically applied low-intensity aerobic exercise can be a noninvasive therapeutic modality to improve or delay the degeneration of dystrophic muscles in individuals with DMD without side-effects.