Autophagy signaling in hypertrophied muscles of diabetic and control rats

Autophagy plays a vital role in cell homeostasis by eliminating nonfunctional components and promoting cell survival. Here, we examined the levels of autophagy signaling proteins after 7 days of overload hypertrophy in the extensor digitorum longus (EDL) and soleus muscles of control and diabetic rats. We compared control and 3‐day streptozotocin‐induced diabetic rats, an experimental model for type 1 diabetes mellitus (T1DM). EDL muscles showed increased levels of basal autophagy signaling proteins. The diabetic state did not affect the extent of overload‐induced hypertrophy or the levels of autophagy signaling proteins (p‐ULK1, Beclin‐1, Atg5, Atg12‐5, Atg7, Atg3, LC3‐I and II, and p62) in either muscle. The p‐ULK‐1, Beclin‐1, and p62 protein expression levels were higher in the EDL muscle than in the soleus before the hypertrophic stimulus. On the contrary, the soleus muscle exhibited increased autophagic signaling after overload‐induced hypertrophy, with increases in Beclin‐1, Atg5, Atg12‐5, Atg7, Atg3, and LC3‐I expression in the control and diabetic groups, in addition to p‐ULK‐1 in the control groups. After hypertrophy, Beclin‐1 and Atg5 levels increased in the EDL muscle of both groups, while p‐ULK1 and LC3‐I increased in the control group. In conclusion, the baseline EDL muscle exhibited higher autophagy than the soleus muscle. Although TDM1 promotes skeletal muscle mass loss and strength reduction, it did not significantly alter the extent of overload‐induced hypertrophy and autophagy signaling proteins in EDL and soleus muscles, with the two groups exhibiting different patterns of autophagy activation.

Autophagy plays a vital role in cell homeostasis by eliminating nonfunctional components and promoting cell survival. Here, we examined the levels of autophagy signaling proteins after 7 days of overload hypertrophy in the extensor digitorum longus (EDL) and soleus muscles of control and diabetic rats. We compared control and 3-day streptozotocin-induced diabetic rats, an experimental model for type 1 diabetes mellitus (T1DM). EDL muscles showed increased levels of basal autophagy signaling proteins. The diabetic state did not affect the extent of overload-induced hypertrophy or the levels of autophagy signaling proteins (p-ULK1, Beclin-1, Atg5, Atg12-5, Atg7, Atg3, LC3-I and II, and p62) in either muscle. The p-ULK-1, Beclin-1, and p62 protein expression levels were higher in the EDL muscle than in the soleus before the hypertrophic stimulus. On the contrary, the soleus muscle exhibited increased autophagic signaling after overload-induced hypertrophy, with increases in Beclin-1, Atg5, Atg12-5, Atg7, Atg3, and LC3-I expression in the control and diabetic groups, in addition to p-ULK-1 in the control groups. After hypertrophy, Beclin-1 and Atg5 levels increased in the EDL muscle of both groups, while p-ULK1 and LC3-I increased in the control group. In conclusion, the baseline EDL muscle exhibited higher autophagy than the soleus muscle. Although TDM1 promotes skeletal muscle mass loss and strength reduction, it did not significantly alter the extent of overload-induced hypertrophy and autophagy signaling proteins in EDL and soleus muscles, with the two groups exhibiting different patterns of autophagy activation.
Skeletal muscle accounts for around 40% of body weight in adults and is critical for maintaining blood glucose levels [1]. Type 1 diabetes (T1DM) causes muscle mass loss and weakness due to high proteolysis [2,3]. Although proteolytic activity threatens long-term muscle health, proteostasis is crucial in maintaining skeletal muscle functioning [4,5].
The two best-known proteolytic systems in skeletal muscle are the ubiquitin-proteasome (UPS) and autophagy [6,7]. During autophagy, misfolded/aggregated proteins or damaged organelles are engulfed and degraded in a double membrane that becomes an autophagosome that fuses with lysosomes forming the autolysosome, which exports amino acids and other byproducts to the cytoplasm [8,9]. Compared with other tissues, skeletal muscle is susceptible to defective autophagy [10,11]. Upregulation or downregulation of autophagy leads to muscle wasting and weakness [12][13][14][15] and may play a critical role in skeletal muscle mass loss and gain in people with diabetes. Interestingly, while proteolysis and muscle wasting are triggered at the onset of diabetes [16], rats in both the early stage of diabetes (3 days) and after chronic diabetes (30 days) exhibit the same relative response to short-term (7 days) and longterm (30 days) overload-induced skeletal muscle hypertrophy as normoglycemic controls [17,18].
Herein, we compared changes in autophagy signaling protein levels after 7 days of overload-induced hypertrophy in skeletal muscles with a predominance of glycolytic extensor digitorum longus (EDL) or oxidative (soleus) fibers in control and diabetic (3 days after streptozotocin induction) rats. We chose a 7-day experimental period because compensatory overload coincides with high activation of the protein synthesis signaling pathway, which returns to baseline levels after 30 days of hypertrophy [17].
The main objectives of the present study were: (1) to investigate whether seven-day compensatory overload induces hypertrophy of EDL and soleus muscles in control and diabetic rats in the same magnitude and (2) to explore whether there is a difference in autophagic signaling protein levels between EDL and soleus muscles in control and diabetic rats before or after overload-induced hypertrophy.

Ethics approval
We used experimental procedures approved by the Ethics Committee for Animal Experimentation of the Institute of Biomedical Sciences at the University of São Paulo (ICB-USP). Experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Academy of Sciences, Washington DC) and the Brazilian College of Animal Experimentation (COBEA). The protocol is registered under No. 23, page 16, in Book 03 of ICB-USP for the experimental use of animals.

Animals
We used 48 eight-week-old male Wistar rats (200 AE 50 g) from the ICB-USP facility. Three rats were housed in each cage and maintained in a room with a 12-h/12-h light/dark cycle at 22°C. Throughout the protocol, the animals had free access to water and standard rodent chow (Nuvilab CR-1, Quimtia, S/A, Colombo, Brazil) containing 22.5% protein, 55% carbohydrates, and 4.5% fat. We subjected diabetic and control rats to the tibialis anterior muscle ablation for EDL hypertrophy or tenotomy of the gastrocnemius muscle for soleus hypertrophy [17,18]. After 7 days, the rats were euthanized in a CO 2 -filled chamber, and EDL and soleus muscles were collected for analysis (Fig. 1).

Type 1 diabetes mellitus induction
T1DM was induced by a single intravenous injection of streptozotocin (65 mgÁkg À1 body weight) dissolved in citrate buffer, pH 4.2 [19]. Control animals received an equivalent volume of citrate buffer by the same route [17,18]. Seventy-two hours after the streptozotocin injection, blood was drawn from the tail, and glucose concentration was measured with a glucometer (Roche Diagnostics Corporation, Indianapolis, IN, USA). Rats with blood glucose levels above 200 mgÁdL À1 (11.1 mmolÁL À1 ) were considered diabetic [20].

Synergistic muscle ablation and tenotomy surgeries
Three days after diabetes induction, rats were anesthetized with ketamine and xylazine (i.p. injection of 90 and 10 mgÁkg À1 body weight, respectively) for surgery. For the ablation of the tibialis anterior muscle, an incision was made in the anterior portion of the animal's left hind paw, exposing the tibialis muscle, which was isolated and removed completely [17,18,21,22].
The gastrocnemius muscle tenotomy was performed by making a longitudinal incision in the posterior portion of the animal's left hind paw, exposing the gastrocnemius and plantar muscles, which had the muscle fascia removed and the distal tendons isolated and sectioned [17,18,23,24].
In both surgeries, a sham operation was performed on the right paw, where the fascia was divulged, but the tendon was not sectioned [25][26][27][28]. The left paws in which the ablation or tenotomy surgeries were performed are called hypertrophied (H), and the right paws in which only the incision and divulsion of the fascia were made are referred to as contralateral (CL). The unilateral operations allow a pairwise comparison between the CL muscles subjected to sham operation and the overloaded muscles. This approach avoids inaccuracies due to using different animals and the systemic modifications this protocol may cause [29,30].
After 7 days of overload [18,31], the rats were euthanized by CO 2 inhalation, and the H and CL EDL or soleus muscles were collected (Fig. 1).
Commonly used housekeeping protein levels vary in cells and tissues depending on experimental conditions. For example, significant variations in the levels of five housekeeping proteins (GAPDH, b-actin, a-tubulin, ɣ-tubulin, and a-actinin) were found to be differentially expressed in STZ-induced diabetes and skeletal muscle hypertrophy models. In these situations, Ponceau S staining is more accurate for quantifying protein loading than the housekeeping proteins tested [33][34][35]. Therefore, our findings were normalized to the pool of samples and total protein content as determined by Ponceau S staining [17,18,33]. Results are expressed relative to the CL control muscle.

Statistical analysis
Data are presented as the mean AE standard error of the mean (SEM) and were analyzed first with the Shapiro-Wilk normality test and then with Student's t-test (for comparison between two groups) or two-way ANOVA (for comparison between three or more groups). The Bonferroni post-test was used to compare contralateral and hypertrophied muscles of the same group and contralateral and contralateral muscles and hypertrophied and hypertrophied muscles of different groups. As indicated in the text and figure legends, ANOVA was used only for comparing diabetic and control rats, considering both contralateral and hypertrophied muscles and the comparisons between diabetic and control hypertrophied muscles and diabetic and control contralateral muscles. Grubb's test was used to exclude outliers. Differences between results were considered significant for p values < 0.05. All results were analyzed using GRAPHPAD PRISM 5.0 statistical software (GraphPad Software, San Diego, CA, USA).

Results
During the 7-day experimental period, the body mass of control animals increased by 42.8 AE 7.1 g but remained unchanged in the diabetic rats (Fig. 2, panel  A). The blood glucose levels of diabetic rats increased fourfold (Fig. 2, panel B).
The wet and dry weights of EDL and soleus muscles of diabetic rats were significantly lower than those of control animals (Figs 3 and 4, panels A and D), even after normalization to tibial length (Figs 3 and 4, panels B and E). Diabetes decreased the dry weight of contralateral EDL muscles by 22%. After 7 days of overload-induced hypertrophy, EDL muscle dry weight increased by 7% in control and 10% in diabetic rats (Fig. 3, panel D). Under the same conditions, the dry weight of the soleus muscle increased by 28% and 31% in control and diabetic rats (Fig. 4, panel D). The increase in hypertrophied muscle wet and dry weights, normalized to tibial length, compared with the contralateral muscle, was similar in both groups (Figs 3 and 4, panels C and F). Thus, both groups' EDL and soleus muscles displayed similar hypertrophic responses.
The autophagy signaling protein levels were also measured after 7 days of overload in the EDL (Fig. 6) and soleus (Fig. 7) muscles of control and diabetic rats.
Overload had marked effects on autophagy signaling protein levels of EDL muscle. For example, p-ULK1 content increased by 71% in the control group compared with the contralateral muscle, but it did not change in diabetic rats (Fig. 6, panel A). Beclin-1 content increased twofold in the control group and 79% in the diabetic group (Fig. 6, panel B). Additionally, Atg5 content was increased by 67% and 68% in control and diabetic rats, respectively (Fig. 6, panel C). By contrast, Atg12-5 (Fig. 6, panel D), Atg7 (Fig. 6, panel E), Atg3 (Fig. 6, panel F), and p62 (Fig. 6, panel I) were not significantly changed in the EDL muscle of either group. LC3-I content increased twofold in control but remained unchanged in diabetic rats (Fig. 6, panel G). Moreover, LC3-II content in the hypertrophied muscle of control rats was 39% higher than in diabetic ones (Fig. 6, panel H). However, none of the LC3-I and LC3-II alterations significantly altered the ratios of the two proteins (Fig. 6, panel J).
Overload induced more pronounced changes in autophagy signaling protein levels in the soleus than detected in the EDL. As shown in Fig. 7, the p-ULK1 content was increased by twofold in control rats (Fig. 7, panel A), Beclin-1 increased by 58% in the control group and twofold in the diabetic one (Fig. 7, panel B), Atg5 increased by twofold in the control and 94% in the diabetic group (Fig. 7, panel C), Atg12-5 increased by twofold in control and diabetic rats (Fig. 7, panel D), Atg7 increased by 70% in control and 96% in diabetic rats (Fig. 7, panel E), Atg3 increased by 78% in the control and 70% in the diabetic groups (Fig. 7, panel F), LC3-I increased by 78% in the control and twofold in the diabetic groups (Fig. 7, panel G), and LC3-II increased by 86% in the diabetic group but did not change in control rats (Fig. 7, panel H). Like the EDL muscle results, p62 (Fig. 7, panel I) and the LC3-II/LC3-I ratio (Fig. 7, panel J) were not altered. A summary of the results from Figs 6 and 7 is presented in Fig. 8.

Discussion
Our study is the first to evaluate autophagy signaling protein expression 7 days post-overload-induced hypertrophy in the EDL and soleus muscles of control and T1DM rats. It was previously reported that 7 days of the diabetic state attenuates the EDL skeletal muscle mass [18,36,37], an observation consistent with the present study. On the contrary, the short diabetic state duration did not affect the soleus muscle mass. It has been proposed that this differential response is due to oxidative muscle fibers (soleus) being more resistant to functional impairment and muscle mass loss in the diabetic state than glycolytic ones (EDL) [18,38,39].
Despite the impairment of muscle mass by T1DM, streptozotocin-induced diabetes concomitant with EDL or soleus muscle overload did not alter hypertrophy after 7 days. In our previous study, the diabetic rats showed a similar hypertrophic response in the same muscles as the control group [17,18]. It has been proposed that autophagy plays a role in maintaining skeletal muscle homeostasis [11,40], and the EDL muscle had higher basal levels of p-ULK1, Beclin-1, and p62/SQSTM1 compared with the soleus before hypertrophy induction. Interestingly, Par e et al. [41] reported that autophagic signaling protein levels are lower in a glycolytic muscle than in an oxidative one; however, the glycolytic muscle's basal autophagic flux was augmented.
Several studies demonstrated that T1DM exposes skeletal muscle to signals that could alter autophagy activity [42][43][44][45]. However, in the present study, the diabetic state per se did not affect autophagy signaling protein levels in either muscle compared with control animals. In contrast, others reported a diabetesinduced increase in autophagy signaling proteins in the gastrocnemius [46,47], soleus, and EDL muscles of mice after 9-10 weeks of diabetes [48,49]. In this sense, the duration of the diabetic state may influence autophagy-related protein expression.
Under basal conditions, specific autophagy signaling protein levels were higher in the EDL muscle. However, the soleus muscle had a more substantial increase in autophagic signaling protein levels after overload-induced hypertrophy, which were also elevated in diabetic animals (Figs 6 and 7). Previous   studies showed that overload-induced skeletal muscle hypertrophy stimulates autophagy [50][51][52]. Indeed, muscle contraction increases protein synthesis and generates reactive oxygen species, enhancing the need for autophagic clearance of damaged cellular components to maintain the working muscle mass and optimal muscle protein content [53]. Moreover, autophagy provides energy for the cells to sustain muscle cell integrity and function [4,54].
In the baseline condition (without diabetes and hypertrophy), our results showed that EDL muscle exhibited greater autophagy, as indicated by the upregulation of phospho-ULK1, Beclin-1, and p62 proteins. After the compensatory hypertrophy protocol, Fig. 5. Representative protein blots (A) and levels of autophagy signaling proteins (B) in extensor digitorum longus (EDL) and Soleus muscles of control rats (CTRL) before overload (CL), measured by western blotting. The following proteins were measured: p-ULK1, Beclin-1, Atg5, Atg12-5, Atg7, Atg3 LC3-I, LC3-II, p62/SQSTM1, and LC3-II /LC3-I ratio. Each band presented in panel A was extracted from the original gel, and all their intensities were normalized by the respective Ponceau S. All raw data are exhibited in Appendix S1 (Attachment 1 for EDL and 2 for Soleus). Statistical analysis was performed using the Student's t-test; *P < 0.05; **P < 0.01. Values are expressed as the mean AE SD. Six animals were used per group. CTRL CL EDL = Contralateral EDL muscle of control animal; CTRL CL Soleus = Contralateral Soleus muscle of control animal. The number of animals used in each group was 6. there was an increase in autophagic markers associated with increased protein turnover, ultimately leading to a net positive effect. The mentioned increase was observed in the EDL muscle of control animals (increases in p-ULK1, Beclin-1, Atg5, and LC3-I) and diabetic animals (increases in Beclin-1 and Atg5), and in the soleus muscle of the control group (increase in p-ULK1, Beclin-1, Atg5, Atg12-5, Atg7, Atg3, and LC3-I) and diabetic animals (increase in Beclin-1, Atg5, Atg12-5, Atg7, Atg3, and LC3-I and LC3-II).
The protein ULK1 is activated by AMPK (protein kinase activated by AMP) due to a negative ATP balance in the absence of nutrients and is inhibited by mTOR activation [55]. After 7 days of overloadinduced hypertrophy, p-ULK1 content increased in the control rats and remained unchanged in the diabetic ones. Nevertheless, the p-ULK1 discrepancy had a negligible effect on the upregulation of other downstream autophagy signaling protein levels caused by overload-induced hypertrophy in both groups.
Beclin-1, which increased after hypertrophy in control and diabetic animals, acts in autophagosome formation at the beginning of the process and can be activated via the ULK1-dependent pathway [56] or through phosphorylation and subsequent inhibition of Bcl-2 (B-cell lymphoma 2), which uncouples from Beclin-1 and promotes its activation under conditions of muscle activity [4,57].
The Atg12-5-16L complex is required for membrane transport to its target and the lipidation of LC3-I to form LC3-II [58]. Raben et al. [15] reported that , and LC3-II /LC3-I ratio using the results in panels G and H (J). Statistical analysis was performed using two-way ANOVA followed by Bonferroni post-test; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Values are expressed as the mean AE SD. Each group was composed of 6 animals. CL, contralateral muscle; CLC, control group contralateral muscle; CLD, diabetic group contralateral muscle; H, hypertrophied muscle; HC, control group hypertrophied muscle; HD, diabetic group hypertrophied muscle. silencing the Atg5 gene results in skeletal muscle mass loss, protein aggregation, abnormal membrane structure accumulation, and impaired muscle strength.
Additionally, Atg5 transgenic mice moderately overexpressing this protein show enhanced autophagy and prolonged mean lifespan [59]. We found that Atg5 alone significantly increased after hypertrophic stimulation in both groups. On the contrary, only Atg12-5 increased in the soleus muscle following overload. It is known that Atg7 plays a role in holding the Atg12-5-16L complex together [60]. Knockdown of Atg7 leads to myopathy, misalignment of the A and Z bands, and increased numbers of mitochondria with membranous structures and protein aggregates. The absence of Atg7 also leads to a 20-40% reduction in skeletal muscle cross-sectional area [14]. At the same time, overexpression of Atg7 in aged mice restores the loss of neuromuscular function and autophagy activity due to aging. Herein, Atg7 levels increased after soleus muscle hypertrophy in both groups.
In the intermediate phase of the autophagic process, Atg7, together with Atg3, conjugates phosphatidylethanolamine in the LC3-I molecule [60], producing LC3-II, which is the actual effector molecule of the autophagic process. Atg3 knockout mice exhibit disrupted autophagosome formation due to a defect in this conjugation step, which prevents the closure of the isolation membrane [61]. We observed an increased Atg3 response in hypertrophied soleus muscle of diabetic and control rats. Furthermore, we observed upregulation of LC3-I due to hypertrophy in the EDL and soleus muscles of the control group. The soleus muscle of diabetic rats exhibited increased LC3-I and LC3-II content only after hypertrophy. It is important to point out that the LC3-II/LC3-I ratio is used to represent autophagic flux [62], and no changes in this parameter were detected in the present study.
In the final steps of autophagy, LC3-II is coupled to the autophagosome membrane, and the p62/SQSTM1 acts as a bridge between ubiquitinated substrates bound to the cargo destined to be degraded and LC3-II, finally promoting the closure of the isolation membrane and degradation. We did not observe changes in p62/SQSTM1 levels in either muscle or experimental group after the overload. A previous study demonstrated that once the autophagosome encloses and fuses with the lysosome, p62/SQSTM1 levels are attenuated due to autolysosomal degradation, thus making it difficult to detect this protein [63].
A limitation of this study is the lack of a paired feeding DM group. However, the aim of this study was achieved since we compared a well-established diabetic state with a control condition, and novel discoveries were reported.

Conclusions
The EDL muscle had higher autophagy signaling protein levels at the baseline. By contrast, the soleus muscle exhibited more elevated autophagy signaling protein levels in control and diabetic rats after hypertrophy induction. The magnitude of a 7-day compensatory overload-induced hypertrophy of EDL and soleus muscles of control and three-day diabetic rats was not different. Diabetes did not alter autophagy signaling protein levels of EDL and soleus muscles before or after functional load hypertrophy.