Acute melatonin administration improves exercise tolerance and the metabolic recovery after exhaustive effort

The present study investigated the effects of acute melatonin administration on the biomarkers of energy substrates, GLUT4, and FAT/CD36 of skeletal muscle and its performance in rats subjected to exhaustive swimming exercise at an intensity corresponding to the maximal aerobic capacity (tlim). The incremental test was performed to individually determine the exercise intensity prescription and 48 h after, the animals received melatonin (10 mg·kg−1) or vehicles 30 min prior to tlim. Afterwards, the animals were euthanized 1 or 3 h after the exhaustion for blood and muscles storage. The experiment 1 found that melatonin increased the content of glycogen and GLUT4 in skeletal muscles of the animals that were euthanized 1 (p < 0.05; 22.33% and 41.87%) and 3 h (p < 0.05; 37.62% and 57.87%) after the last procedures. In experiment 2, melatonin enhanced the tlim (p = 0.01; 49.42%), the glycogen content (p < 0.05; 40.03%), GLUT4 and FAT/CD36 in exercised skeletal muscles (F = 26.83 and F = 25.28, p < 0.01). In summary, melatonin increased energy substrate availability prior to exercise, improved the exercise tolerance, and accelerated the recovery of muscle energy substrates after the tlim, possibly through GLUT4 and FAT/CD36.


Methods
Animals and environmental conditions. Sixty-eight male Rattus norvegicus albinus (Wistar) rats that were 45 days old were housed in a bioterium and kept in polypropylene cages (length: 40 cm, width: 40 cm, height: 20 cm, and 5 animals per cage); they received feed and water ad libitum. Throughout the experiment, the environmental conditions were maintained, including the temperature (22 ± 2 ºC), relative humidity (45% and 55%), noise (< 85 decibels), and photoperiod (10:14 h light/dark cycle). Incandescent lamps (Philips's brand, soft model, 100 W, 2700 K; 565-590 nm; 60 lx, measured with a lux meter) were used during the 10-h light cycle. To carry out experimental interventions with the rats during the dark cycle (nighttime: 4:00 pm to 6:00 am), reflectors were installed in the bioterium and room and were surrounded by a red filter (ROSCO brand, model # fire19; > 600 nm; < 15 lx) 14,15,35 . The experimental procedures were conducted in accordance with Ethical Principles in Animal Research, adopted by the Brazilian College of Animal Experimentation (COBEA, Brazil) and was approved by the Ethics Committee on the Use of Animals (CEUA) of Federal University of São Carlos (São Paulo, Brazil) under protocol no. 9144181218. The experimental procedures were conducted in accordance with the Ethical Principles in Animal Research (ARRIVE guidelines 2.0). Experimental design. The animals (n = 68) were randomly split into 7 groups: a control (Ct: n = 10), rats treated with melatonin and euthanized 1 h (M1: n = 9) or 3 h after the last procedures (M3: n = 9), rats that exercised and were euthanized 1 h (Ex1: n = 10) or 3 h after the time to exhaustion (tlim) (Ex3: n = 10), and rats that were treated with melatonin, exercised, and were euthanized 1 h (ME1: n = 10) or 3 h after tlim (ME3: n = 10). The animals in the Ct, Ex1, and Ex3 groups received vehicles, while the animals in the M1, ME1, M3, and ME3 groups received melatonin (10 mg·kg −1 ); both were administered in the same volume. The M1, Ex1, and ME1 groups and the M3, Ex3, and ME3 groups were euthanized 1 and 3 h after the end of the experimental procedures for each group, respectively. The experiments were divided into experiment 1 (Ct, M1, and M3) and experiment 2 (Ex1, Ex3, ME1, and ME3).
After the housing familiarization period, when they were from 76 to 89 days old, all of the rats were adapted to aquatic environments and swimming according to a protocol from Lima et al. 36 that focused on the exposure time in water (5-20 min), water depth (10-80 cm), and load weight (0 or 3% of the body mass). The swimming protocol was individualized in cylindrical and opaque tanks that were 100 cm in height (80 cm in water depth) and 30 cm in diameter; the water temperature was maintained at 31 ± 1 °C in accordance with the guidelines of the American Physiological Society 37 . When they were 90 days old, all animals were submitted to an incremental test (IT) to determine the intensity of the effort corresponding to the individuals' maximal aerobic capacities.
At 92 days old, the animals (body mass: 398.06 ± 3.80 g at the end of the experiment) received melatonin or vehicles from 30 min before the swimming exercise until exhaustion at the intensity corresponding to the maximal aerobic capacity, which was called the time to exhaustion (tlim). The criteria for identifying the animals' exhaustion were standardized according to Beck and Gobatto 35 ; an analysis of the swimming behaviors of the animals was performed to observe the execution of vigorous efforts in returning to the surface without success for a period of 15 s. The achievement of exhaustion was accepted upon the agreement of two experienced observers using the above criteria. Then, the animals were euthanized 1 or 3 h after the end of the experiment via decapitation, a method that is allowed by the American Veterinary Medical Association 38 . The experimental design is shown in Fig. 1.
Incremental swimming test. The IT consisted of proportional increases in the load over time in order to identify a disproportionate increase in the concentration of blood lactate at a given moment 39 , which was called the maximal aerobic capacity. Therefore, the animals were subjected to five-minute stages with overloads corresponding to 4.0, 4.5, 5, 5.5, 6, 6.5, and 7.0% of their body mass (% BM); these overloads were attached to the animals' chests with an elastic strap. After each stage, blood samples (25 µL) were collected from the distal part of the animals' tails and then stored (4 °C) in order to determine the lactate concentration. After analyzing the lactate concentration with the enzymatic method, the intensity of the exercise in relation to the blood lactate concentration was plotted on a scatter plot, and any changes in the blood lactate concentration were identified through visual inspection, as previously described by Matsumoto et al. 40 . Then, two linear regressions were constructed after the breaking point. The intersection of these linear regressions was interpolated to the X-axis and then used to define the intensity corresponding to the anaerobic lactacidemic threshold 39 . The interpolation for line y corresponded to the blood lactate concentration at the intensity of the maximal aerobic capacity. M-5250, > 98%) was dissolved in ethanol (< 0.1%) and diluted in saline (0.9% NaCl) for administration at 10 mg·kg −114,15 . The preparation was carried out just prior to its use, and it was stored in an amber bottle that was wrapped in aluminum foil. Administration was intraperitoneal and took place 30 min prior to the tlim.
Analytical procedures on biological materials. Plasma and serum parameters. During the IT, blood samples (25 μL) were collected from the animals' tails in heparinized and calibrated glass capillaries. These samples were immediately transferred to 1.5 mL tubes containing 400 μL of trichloroacetic acid (4%), which were then agitated and stored at 4 °C. After stirring and centrifuging (3000 rpm for 3 min), 50 µL of supernatant was extracted and transferred to a 96-well microplate, where added 250 µL of reactive solution that was prepared for immediate use (glycine/EDTA and hydrazine hydrate stock), NAD (β-nicotinamide adenine dinucleotide), and LDH (L-lactic dehydrogenase bovine heart) were added; the pH was properly adjusted to 9.45 before the added of NAD and LDH. The samples and reagent were incubated (20 min, 37 °C) and the absorbance was determined  www.nature.com/scientificreports/ in a spectrophotometer (Spectramax i3, Molecular Devices; San José, CA, USA) at 340 nm. The blood lactate concentration was determined in relation to the standard curve constructed from the serial dilution of L-Lactate 1-15 mmol/L. After euthanasia, an aliquot of approximately 2.0 mL of blood was obtained and allowed to rest for 20 min (4 °C) before a subsequent centrifugation (15 min, 3000 rpm, 10 °C). These samples were stored at − 20 °C for further analysis.
Skeletal muscle glycogen. The procedure was performed according to the method presented by Dubois et al. 41 . Firstly, skeletal muscle tissue (200-250 mg; gluteus maximus) was digested in potassium hydroxide (KOH 30%). Then, a saturated sodium sulfate solution (20 µL, Na 2 SO 4 ) and ethanol (3 mL, CH 3 CH 2 OH 70%) were added for the precipitation of glycogen. The samples were submitted to the colorimetric phenol (10 µL, C 6 H 6 O) and sulfuric (2.0 mL, H 2 SO 4 ) method and measured via spectrophotometry (Hach Company, Loveland, Colo, USA; 490 nm) against a standard glucose curve.
Skeletal muscle triglyceride. Initially, skeletal muscle tissue (100-200 mg; gluteus maximus) and Triton X-100 (1%) were mixed at the same proportions (200 mg of tissue to 1 mL of Triton). Next, the samples were homogenized with magnet bars (5 × 3 mm) overnight (2-8 °C) and centrifuged (10 min, 4000 rpm). After this period, 10 µL of the supernatant was extracted, pipetted into a 96-well microplate in a mixture with the kit reagent Histological and immunofluorescence procedures. Immediately after euthanasia, the soleus muscle was dusted in talc, frozen in liquid nitrogen, and stored (− 80 °C). Transversal histological frozen Sections (6 μm) were obtained from a cryostat (− 25 °C; Leica CM 1850 UV) and collected on glass slides (26 × 76 mm). Prior to the immunofluorescence protocol, the slides were stained with Hematoxylin-Eosin (HE) in order to identify morphological changes in the tissue that could compromise the analysis with a light microscope. The slides were photographed with an automated high-resolution epifluorescent microscopy system (Imag-eXpress® Micro, Molecular Devices; San José, CA, USA) using an objective lens with a magnification of 20×, with specific filters for GLUT4 (FITC: 1000-1200 ms exposure), FAT/CD36 (Cy5: 1800-2200 ms exposure), and laminin (FITC and Cy5: 200 ms exposure). The images were saved with an identical size and resolution.
The integrated density of the fluorescence intensity of GLUT4 and FAT/CD36 was quantified in five distinct and random fields (height: 220 and width: 220) by the ImageJ 1.52a software (National Institutes of Health, USA), and the images were individually analyzed. The mean values of the proteins in each sample were calculated and plotted in a graph.
Statistical data analysis and processing. The data were presented as mean ± standard error. Normality was verified with the Shapiro-Wilk test (p > 0.05). The time to exhaustion was analyzed with the t-test for independent samples by using pooled data from all exercised animals that were treated with melatonin (ME1 and ME3) versus exercised animals that were treated with a vehicle (Ex1 and Ex3). A one-way analysis of variance was performed for all the parameters of experiment 1 (experiment 1: Ct, M1, and M3) and for lactacidemia and % BM in experiment 2. A two-way analysis of variance was performed for the other parameters in experiment 2-the effects of melatonin (melatonin or vehicle) and the time of euthanasia (1 or 3 h) (experiment 2: Ex1, Ex3, ME1, and ME3). When appropriate, we used the Newman-Keuls post hoc test. A significance level of 5% Energy substrates in the muscles and blood. The glycogen content increased in the gluteus maximus at 3 h compared to that at 1 h (F = 15.57, p < 0.01; 3 h > 1 h), while the groups that exercised and received melatonin did not experience a difference in glycogen content compared to the animals that received the vehicle (F = 1.12, p = 0.72). For the muscle triglyceride, the time and treatment did not promote an effect on the gluteus maximus (F = 0.70, p = 0.40 and F = 0.05, p = 0.80, respectively) ( The blood triglyceride concentration was higher at 1 h than that at 3 h (F = 54.39, p < 0.01; 3 h < 1 h), while the blood glucose remained unchanged between the animals that were euthanized 1 or 3 h after tlim (F = 0.82, p = 0.36). Moreover, melatonin decreased the serum triglyceride concentration (F = 9.50, p < 0.01), but did not cause a change in the blood glucose compared to animals that received the vehicle (F = 1.16, p = 0.28) ( Table 2) Table 1. Data on the muscle glycogen and triglyceride content and blood glucose and triglyceride concentrations in the groups that were treated with the vehicle (Ct) or melatonin (M1 and M3) and euthanized 1 (M1) or 3 (M3) hours after the last procedures. Control group (Ct); rats treated with melatonin and euthanized 1 h (M1) or 3 h after the last procedures (M3). Values are expressed as mean and standard error. a p < 0.05 with respect to Ct for the same parameter. g: grams; mg: milligrams; dL: deciliters.

Discussion
The main finding of this study was the ability of melatonin to increase GLUT4, FAT/CD36, and the metabolic recovery process in exercised skeletal muscle favoring cellular environment for future efforts, which corroborated our hypothesis. In addition, this is the first study to highlight the acute effect of melatonin administration on energy substrate transporters, as well as melatonin's role in the metabolic recovery of rats that were submitted to an individualized exhaustive exercise session with an intensity corresponding to maximal aerobic capacity.
In experiment 1, we observed that the acute administration of melatonin increased the muscular glycogen content (p < 0.05; 22.33% and 37.62%, M1 and M3 > Ct, respectively). This was associated with an increase in the GLUT4 presented by the animals that were treated with melatonin compared to the animals treated with the vehicle (p < 0.05; 41.87% and 57.87%, M1 and M3 > Ct, respectively). It is well known that melatonin acts www.nature.com/scientificreports/ by binding to membrane receptors that are coupled to G proteins (MTNR1A or MT1 and MTNR1B or MT2) 43 , which are present in the membranes of skeletal muscles (BioGPS (http:// biogps. gnf. org)); this causes an increase in the activity of IRS-1 and PI3K 44 . These upstream signals are responsible for raising the activity and content of GLUT4 in a way that is similar to insulin signaling.
Considering the robust effect of melatonin on muscular glycogen content (as observed in experiment 1) and the importance of the oxidation of carbohydrates during exercise, we investigated the effects of melatonin in animals that exercised at their maximal aerobic capacity and were euthanized at different times after the exercise. Therefore, the intensity of the effort was individually determined by using an incremental test; no differences were demonstrated between the groups (p > 0.05) for lactacidemia or % BM before the tlim. Then, the ergogenic capacity of melatonin was confirmed by the time to exhaustion (tlim) (Ex1 and Ex3, 52.40 ± 19.66 min; ME1 and ME3, 78.30 ± 32.43 min; p = 0.01). These findings corroborate those of previous studies published by our group, which demonstrated high performance in tlim by animals treated with melatonin during periods of wakefulness (10 mg·kg −1 ) 14,15 .
According to Bergstrom et al. 45 , an increased glycogen content is one of the main determinants for performance in moderate and prolonged exercise. In addition, the dependence on carbohydrates in high-intensity and long-term physical exercises is well recognized 46 . To confirm this, previous studies by our group demonstrated that, when submitted to endurance exercise until exhaustion at individualized intensities of effort corresponding to the maximal aerobic capacity, Wistar rats (92 days old) showed a depletion of the glycogen content in the gluteus maximus immediately after swimming exercise, among other effects (p < 0.05) 47 . In addition, Matsui et al. 48 demonstrated that as the duration of the exercise increased, the glycogen content was observed to decrease after the effort. Based on these assumptions, animals treated with melatonin were expected to show lower values for the muscular glycogen content, as they swam longer than animals treated with the vehicle (p = 0.01, 49.42%). However, in the analysis of the muscular glycogen content, despite the ME1 group showing a reduction with respect to the Ex1 group (16.7%), the values were found to be statistically equal (p > 0.05). This possibly occurred due to the ability of melatonin to increase the glycogen content (p < 0.05; M1 and M3 > Ct), thus improving the rats' performance in the exercise. Therefore, the data indicated that melatonin is one of the factors responsible for the better performance due to the increase in the pre-effort glycogen stores (as seen in the M1 and M3 groups), thus consequently increasing the time until exhaustion during the swimming exercise at an intensity corresponding to the maximal aerobic capacity (as seen in the ME1 and ME3 groups).
Regarding the metabolic recovery, in the presence of melatonin, the animals euthanized 3 h after tlim (ME3) showed an increase in glycogen content with respect to the animals euthanized 1 h after tlim (ME1) (p < 0.05; 40.03%). However, in the absence of melatonin, no differences in glycogen content were demonstrated when comparing the Ex3 and Ex1 groups (p > 0.05). Moreover, no statistical differences were observed between the ME3 and Ex3 groups (p > 0.05); however, the ME3 group swam longer than the Ex3 group (p < 0.05). Based on these results, in the presence of melatonin, the metabolic recovery after exercise until exhaustion was improved. The enhancement of the glycogen content possibly occurred due to the increase in the GLUT4 demonstrated by the animals treated with melatonin (ME1 and ME3) in comparison to the animals treated with the vehicle (Ex1 and Ex3, p < 0.05), thus increasing the uptake of glucose in the skeletal muscles after exercise. These data are consistent with the findings of Mendes et al. 49 , who demonstrated an increase in the content of PI3K, GLUT4, and glycogen stores in the skeletal muscles of rats that were submitted to treadmill training (20 m min -1 , 5 days week −1 , 16 weeks) and treated with melatonin (10 mg·kg day −1 , 8 weeks).
Regarding the metabolism of lipids, the muscular triglyceride content of the ME1 group was higher than that of the Ex1 group (p < 0.05), which was possibly due to the increase in the FAT/CD36 shown by the ME1 group compared to Ex1 group (p < 0.05). Interestingly, in the absence of melatonin, such an increase with respect to the Ex1 group (p < 0.05) occurred only 3 h after the tlim (Ex3). Due to the increase in FAT/CD36 in comparison to Ex1 (p < 0.05), it also occurs only 3 h after the tlim (Ex3). Therefore, melatonin enhances the triglyceride content 1 h after exercise, which possibly improves the metabolic recovery process. Assuming that the activation pathway of FAT/CD36 is similar to that of GLUT4 50,51 and given the presence of MTNR1A/MTNR1B in the skeletal muscles of rats 44 (BioGPS (http:// biogps. gnf. org)), we believe that the enhancement of FAT/CD36 was possibly influenced by melatonin through its binding to the MTNR1B receptor and, consequently, its activation of PI3K, IRS 44 , DAG, IP3, PLC, and 43 Ca 2+ . However, there are no studies concerning the effects of acute melatonin Table 2. Data on the muscular glycogen and triglyceride content, as well as the blood glucose and triglyceride concentrations, in the exercised groups (Ex1, Ex3, ME1, and ME3), those treated with the vehicle (Ex1 and Ex3) or melatonin (ME1 and ME3), and those euthanized 1 (Ex1 and ME1) or 3 (Ex3 and ME3) hours after tlim. Rats that exercised and were euthanized 1 h (Ex1) or 3 h after tlim (Ex3); rats that were treated with melatonin, exercised, and were euthanized 1 h (ME1) or 3 h after tlim (ME3). Values are expressed as the mean and standard error. a p < 0.05 with respect to Ex1; b p < 0.05 with respect to Ex3; c p < 0.05 with respect to ME1 for the same parameter. g: grams; mg: milligrams; dL: deciliters. www.nature.com/scientificreports/ www.nature.com/scientificreports/ administration on the content of FAT/CD36 in exercised skeletal muscles. Furthermore, the ME3 group showed a reduction in the muscle triglyceride content in comparison to the ME1 group (p < 0.05), which was possibly due to the greater use of fat while resting, which is considered an optimal muscle environment for fat oxidation and the consequent supply of ATPs for the post-exercise recovery. In addition, the reduction demonstrated by the ME3 group possibly occurred due to the increase in FAT/CD36 in the ME3 group (p < 0.05), which consequently increased the transport of triglyceride from the blood to the skeletal muscles to be oxidized. Thus, it would be plausible to affirm the ability of melatonin to accelerate the metabolic recovery processes related to carbohydrate metabolism and to modulate the supply of lipids after exhaustive exercise. Some limitations in this manuscript must be addressed. First, other dosages should be tested in order to demonstrate the lowest concentration of melatonin that would make it possible to achieve similar effects. Finally, we focused on transporters and their respective substrates; however, evaluating the activation of upstream signals would be quite enlightening. However, our findings make clear that future studies must be conducted in order to deepen the knowledge on this relevant area.
In conclusion, the present study demonstrated that melatonin increased the availability of glycidic substrates and GLUT4 in skeletal muscles and consequently provided a greater tolerance to physical exercise. In addition, melatonin improved the efficiency of the recompositing of energetic substrates and enhanced GLUT4 and FAT/ CD36 in the exercised skeletal muscles, thus improving the cellular environment for future efforts, at least from the bioenergetic point of view. www.nature.com/scientificreports/