Hydrogen Enhanced the Myogenic Differentiation of Adipose Mesenchymal Stem Cells Through P38 MAPK Signaling Pathway

Purpose: This study aims to clarify the systems underlying regulation and regulatory roles of hydrogen in the myogenic differentiation of adipose mesenchymal stem cells (ADSCs). Materials and methods: In this study, ADSCs acted as an in vitro myogenic differentiating mode. First, the Alamar blue Staining and mitochondrial tracer technique were used to verify whether hydrogen could promote cell proliferation. In addition, this study assessed myogenic differentiating markers (e.g., Myogenin, Mhc and Myod protein expressions) based on the Western blotting assay, analysis on cellular morphological characteristics (e.g., Myotube number, length, diameter and maturation index), RT-PCR (Mhc and Myod mRNA expression) and Immunouorescence analysis (Desmin, Myosin and β-actin protein expression). Lastly, to verify the myogenic differentiating system of hydrogen, Western blotting assay was performed to detect p38 and p-p38 proteins expressions. Results: Hydrogen can remarkably enhance the proliferation of ADSCs in vitro by increasing the number of single-cell mitochondria and by up-regulating the expression of myogenic biomarkers (e.g., Myod, Mhc and myotube formation). The expressions of both p38 and p-p38 were up-regulated by hydrogen. The differentiating ability was suppressed when the cells were cultivated in combination with SB203580 (p38 MAPK signal pathway inhibitor). Conclusions: The present study initially indicated that hydrogen can promote myogenic differentiation via the p38 MAPK pathway. Thus, the mentioned results present insights into myogenic differentiation and are likely to generate one potential alternative strategy for skeletal muscle related diseases.

Besides, as suggested by ZuK and Mizuno et al., ADSCs could achieve myogenic differentiation. However, studies have shown that only 15% of the stem cells harvested from the processed liposuction (PLA) can fully differentiation [12], and it is di cult for stem cells to return to the injured site after direct transplantation to the target organ, and the cell survival rate is low, and the e ciency of local differentiation is not high [13]. Thus, the clinical myogenic differentiation potential of ADSCs is limited. At present, researchers try to promote myogenic differentiation of stem cells in different ways, such as drug induction, physical induction, cell scaffold induction and so on. Among them, drugs as a more mainstream way of induction, but there are some problems, such as low induction e ciency, obvious inhibition of stem cell self-renewal ability and so on [14]. Therefore, how to make stem cells get e cient myogenic differentiation in the repair area is the main challenge in the treatment of muscle loss.
Hydrogen(H 2 ), one kind of endogenous gas, was suggested to be one vital energy origin and to critically impact physiologically-related regulating process [15]. Hydrogen molecules can enter tissue as well as performing anti-in ammation-related, antioxidant and anti-apoptotic effects [16]. It is noteworthy that H 2 exhibits great e cacy and safety to cases clinically [17]. For example, Aoki et al found that hydrogen reduced the blood lactic acid level of male football players [18] and improved the decline of muscle function caused by exercise. H 2 can also reduce exercise-induced muscle injury and delayed muscle atrophy, but has no effect on peripheral neutrophil count and neutrophil dynamics and function [19]. In the mouse hindlimb Istroke R injury model, H 2 inhalation signi cantly reduced the infarct area and tissue structure loss area, alleviated muscle injury, and enhanced functional recovery [20]. However, how H 2 exerts the above effects remains to be further studied.
As indicated from the existing research, ERK1/2, NF-kB, Notch and Wnt signal pathways are capable of regulating the forming and reconstructing processes of general skeletal muscle [21][22][23][24]. Moreover, the p38 MAPK class refers to the signal transducing elements that can facilitate the muscle differentiation in vitro, impact the in vivo growth and repair of skeletal muscle, and directly affect the myogenic transcribing elements pertaining to the Myod family [25,26]. To clarify the action mechanism of H 2 , the GEO database was searched, and the H 2 -cultured cells were found to be obtained through the highthroughput sequencing of the data set GSE62434. The data were compared with the transcriptional group of normal cultured cells. On the whole, 351 differential genes were obtained by conducting a differential gene analysis. Meantime, as indicated from the analysis of KEGG (Kyoto gene and encyclopedia) enrichment pathway, the differential genes were largely enriched in mitogen-activated protein kinase (MAPK signal pathway) (Fig. 1A,B). Through the further analysis of MAPK signal pathway, we found that the gene differential expression and signi cance of p38-MAPK signal pathway were the most signi cant. (Fig. 1C). Besides, according to the results of GSEA enrichment analysis, a signi cant difference was identi ed in the expression of p38-MAPK pathway between H 2 culture and normal culture (Fig. 1D,E). As suggested from a GSEA analysis, the function of mitochondria in H 2 culture group was signi cantly more effective than that in normal culture group (Fig. 1F). As obviously indicated from the mentioned analysis, p38 MAPK signal pathway and mitochondrial function may impact H 2 -induced myogenic differentiation of ADSCs to a certain extent, whereas its speci c molecular mechanism remains unclear. For this reason, H 2 was speculated to be capable of regulating the adipose stem cell myogenic differentiation by impacting the p38 MAPK signaling pathway.
The hypothesis of the present study was that H 2 might enhance the myogenic differentiation of adipose mesenchymal stem cells through p38 MAPK signaling pathway. To test this hypothesis, we evaluated the effect of H 2 on myogenic differentiation of adipose stem cells and its molecular mechanism by molecular biology and reverse transcriptase polymerase.

Extraction and cultivation of ADSCs
All animal experiments were performed under the approval of the Institutional Animal Care and Use Committee of Jiangsu University. Five two weeks old male SD rats were euthanized with excessive ether asphyxiation. The rats were routinely disinfected by 75% ethyl alcohol, and their blood vessels, fascia and other tissues were removed. The harvested adipose tissue was cut into 1mm 3 pieces and then placed into a 15 ml centrifuge tube. 0.1% type I collagenase was used to digest tissue for 40-60 min. After lter to remove nondigested tissue, ltrate was collected and seed into tissue culture plastic in culture medium with 10% fetal bovine serum. The incubation was conducted at 37℃ with a volume fraction of 5% CO 2 , and the liquid was altered after 24 h, 48 h and 72 h. Under the cell concentration of 80%, 0.25% trypsin-EDTA was employed for digestion and subculture. The 3rd generation cells were employed for the immunocytochemical identi cation and subsequent tests.

Identi cation of ADSCs
ADSCs were cultured and then identi ed when 90% of the cells were covered. The cells were digested by trypsin at ambient temperature for 10 min, and it was collected after being centrifuged. Subsequently, the cells were washed with PBS and then prepared into 1*10^5/ml cell number suspension. 100µL of 1*10^5 cell suspension was added with 10µL of uorescently labeled phenotypic antibody (CD45, CD44, CD90 and CD31) respectively for incubation. After the washing process, the expressions of CD45, CD44, CD90 and CD31 surface antigens were detected by performing the ow cytometry.

Identifying process for three-line differentiation of ADSCs
The multilineage differentiating potentials of ADSCs were characterized by osteogenic, adipogenic, and chondrogenic differentiation. Brie y, according to 5*10 4 cells/cm 2 cell density, cells received the collecting and inoculating processes inside the ori ce plate. The cells received the culturing process with 5%CO 2 at 37℃. Under the cell con uence of 90%, The cells were cultured in osteogenic, lipogenic and chondrogenic differentiating medium, and identi ed and analyzed by alizarin red (Solebo G8550), oil red O (Solebo G1262), toluidine blue staining (Solebo G3661). Stained areas were observed under a uorescence microscope (OLYMPUS, IX71).

Groups and cell proliferation of ADSCs
Cells were collected after being inducted with control, H 2 , 5-Aza, H 2 +5-Aza for 1, 2 and 3 days. The cell proliferation ability of H 2 and 5-Aza was compared by MTT and Alamar blue staining respectively. Cells were stained with MTT (20µL; 5 mg/ml) for 4 h at 37 ℃. Dimethyl sulfoxide was added to samples and incubated for 30 min at 37 ℃. The absorbance was measured using a microplate reader (Thermo MK3 type) at 492 nM.
Cell samples were taken, and 10µL Alamar blue staining was introduced to 100µL cell suspension at 1and 3-day induction time. Incubating in an incubator away from light for 6 h, and detecting with microplate reader with a wavelength of 600nm. The diluted AM/PI kit was added and incubated for 30 min before washing. Observed by means of uorescence microscope (OLYMPUS, IX71). Red and green denote dead cells and live cells, respectively. With ImageJ software (National Institutes of Health, Bethesda), this study determined the uorescence intensity.

Effect of hydrogen on mitochondria by mito-tracker green uorescent staining
To investigating the in uence exerted by H 2 for mitochondria. First, ADSCs received the inoculating process on confocal petri dishes and the culturing process inside normal medium for 12 h. Next, the cells received the seven-day incubation with control, H 2 , 5-Aza, H 2 +5-Aza. Then, the medium received the removal and rinsing process by applying PBS, and then it was incubated with the mito-tracker (1:5000-1:50000) at 37℃ for 15-45 min. Lastly, under a Laster confocal uorescence microscope, the observation of cells was conducted.

Immuno uorescence detecting process of the expressions of Desmin, Myosin and β-actin in cells
After incubation with control, H 2, 5-Aza, H 2 +5-Aza for 7 days, ADSCs were xed in 4% paraformaldehyde for 15min. Then, 0.1% Triton was used for stabilization at ambient temperature for 15 min. The cells were sealed with 5% FBS for 15 min under the temperature of the ambient. Subsequently, the cells were incubated by using primary antibody at 4℃ throughout the night. After washing, it was incubated at 37℃ for 1h by using secondary antibody. Hochest received the incubating process under ambient temperature and the isolating process away from light for 15min. Lastly, with the use of a uorescence microscope (OLYMPUS, IX71), the samples were observed.

Real-time PCR analysis of Myod and Mhc
Under the p38 MAPK signal pathway inhibitor (SB203580), the cells were collected after being inducted with control, H 2 , 5-Aza, H 2 +5-Aza for 7 days. Overall RNA was extracted from the cells under the transfection with the Trizol extraction tool in accordance with the guidelines of the producer. According to the guidelines of the producer, cDNA received the synthesis based on reversely transcribing process with the rst strand cDNA synthesis tool. Primers for reverse transcription PCR received the designing and synthesizing processes with Primer Premier 5.0 software (Shanghai Biotechnology, China) based on internal reference of housekeeper gene GAPDH. Table 1 lists primer sequences. After PCR ampli cation, the results were automatically analyzed by using the uorescence quantitative PCR tool in the real time, the baseline and threshold were regulated in accordance with the negative control to determine the Ct value of the respective specimen, as well as whether the Ct value was effective based on the fusion curve.
For the export, the 2 -△△CT approach inside gene expressing state differences of control and the concentration groups is: △Ct≡ Ct gene-Ct inside. Subsequently the control group △Ct remember was obtained for △Ct contrast, △Ct contrast average was achieved, in which the respective group of △Ct minus △Ct contrast average, calculated by △△Ct value, i.e., the △△Ct = △Ct sample-△Ct contrast, next, the respective group 2 -△△CT value was calculated, which indicated the comparative expressions of genes.

Observation and analysis of myotube
For investigating the myogenic differentiating system of H 2 -induced ADSCs, the authors introduced the SB203580(5uM) to the culture medium (DMEM). Under the p38 MAPK signal pathway inhibitor (SB203580), cells were collected after being inducted with control, H 2 , 5-Aza, H 2 +5-Aza for 7 days. First, ADSCs received the inoculating process on 24-well plates under a density of 8000 cells /cm for 12 h. Subsequently, the cells received further culturing process with a concentration of H 2 , and the medium received the renewal every two days. With the immuno uorescence staining, the myotube formation of ADSCs impacted by hydrogen was determined. ImageJ software was used to obtain the index maturation of myotube (National Institutes of Health, Bethesda, USA).

Statistical analyses
All experiments were performed in triplicate. For the data recording and the statistical analyses, SPSS 24.0 software in terms of Windows (SPSS Inc., Chicago, IL, USA) was used. The information has the expression to be the mean±standard deviation (SD) pertaining to a range of measuring processes. The authors performed student's t testing process for comparing one individual experiment-related mean with the control mean. P<0.05 exhibited statistical signi cance.

Results:
3.1 The evaluation of ADSCs 3.1.1 ADSCs were identi ed by ow cytometry ADSCs have no speci c surface antigens. Accordingly, the analysis of multiple surface antigens simultaneously to determine the characteristics of adipose stem cells. By ow cytometry, the ADSCs surface markers was tested. We examined the expression of CD31, CD44, CD45 and CD90 in ADSCs at passage 3. Flow cytometry results showed that 1.6% expressed CD31, 95.2% expressed CD44, 1.1% expressed CD45, and 97.4% expressed CD90 (Fig. 1G). CD44 (95.2%) and CD90 (97.4%) express stem cells. The low expressions of CD31 (1.6%) and CD45 (1.1%) excluded epidermal cells and vascular endothelial cells, respectively. For the mentioned results, almost all the experimental cells expressed stem cell characteristics.

ADSCs were identi ed by three-line differentiation
The multilineage differentiating potentials of ADSCs were characterized through the osteogenic, adipogenic and chondrogenic differentiation. Alizarin red staining under the induction by ADSCs osteogenesis are presented in Fig. 1H(a), demonstrating the appearance of considerable osteoblasts when the induction was achieved. Fig. 1H(b) shows the staining of toluidine blue when the chondroblast induction was achieved, and many chondrocytes were stained. Fig. 1H(c) shows the staining process of oil red O when the lipogenic induction was achieved, and the stained cells were observed as well. In brief, the mentioned cell could achieve a three-line differentiation, which demonstrated that it acts as a rat ADSCs.

The optimal concentration of hydrogen
Whether H 2 could promote cell viability and proliferation was evaluated by MTT assay (Fig. 2F)

Assessment of cell proliferation by hydrogen
First, we evaluated whether H 2 could promote cell viability and proliferation by MTT and Alamar blue Staining ( Fig. 2A-E). In H 2 group, when cultivated for 1 and 3 days, the cells were primarily alive (green), while few had death (red) ( Fig. 2A,C). In addition, the cell proliferation characteristic received the quantitative analysis of Live-Dead Cell Staining. As compared with the 5-Aza group, cells displayed a remarkably high proliferation in H 2 group(p<0.01) on day 1. There was no signi cant difference between the other three groups (Control, H 2 , 5-Aza+H 2 ) (p>0.05). On day 3, compared with 5-Aza and 5-Aza+H 2 groups, H 2 group displayed the highest expression(p<0.01). Compared with the 5-Aza group, the survival rate of cells in the 5-Aza+H 2 group was still higher(p<0.01).
Furthermore, the cell proliferation behavior was quantitatively analyzed by Alamar blue assay. On day 1, the cell viability in H 2 group was signi cantly higher than other groups (5-Aza, 5-Aza+H 2 ) (p<0.01). And, compared with the control group, only the H 2 group showed high proliferation(p<0.05). Interestingly, compared with the 5-Aza group, the 5-Aza+H 2 group also showed signi cantly higher proliferation(p<0.01). On day 3, the cell viability in H 2 , 5-Aza and 5-Aza+H 2 groups maintained the same trend as day 1 (Fig. 2B). Based on MTT analysis (Fig. 2D,E), compared with the 5-Aza group, the H 2 group signi cantly promoted cell proliferation in the rst three days(p<0.01). At the same time, compared to the 5-Aza group, the 5-Aza+H 2 group also increased cell proliferation(p<0.05). As demonstrated by all the mentioned results, H 2 remarkably enhanced the ADSCs proliferation.

Mito-tracker green uorescent staining
The green mito-tracker staining was performed for explaining the effect of H 2 on the mitochondria function of ADSCs. (Fig. 2G,H). Obviously, compared with the other three groups, H 2 group expressed signi cant single-cell mitochondrial uorescence intensity(p<0.01). Compared with the control group, the intensity of green uorescence was signi cantly decreased in the 5-Aza group(p<0.01), and the uorescence expression of single-cell mitochondria was also signi cantly increased in the 5-Aza+H 2 group compared with the 5-Aza group(p<0.01).

Immuno uorescence detecting process of the expressions of Desmin, Myosin and β-actin in cells
To investigate the effect of H 2 on myogenic differentiation at the myoblast stage, ADSCs were employed as an in vitro model of myogenic differentiation. The cells were respectively cultured with 5-Aza, H 2 , H 2 +5-Aza for 7 days. Fig. 3A shows the results of immuno uorescence staining assay after incubation for 7 days. The immuno uorescence results showed that remarkably higher immuno uorescence expressing states of β-actin and desmin in the H 2 group and 5-Aza group than in the control(p<0.01). In addition, compared with the control group, the H 2 group showed increased myosin expression(p<0.05).
Additionally, the expressing states of all test antibody were remarkably higher in the H 2 +5-Aza group than in the group H 2 (p<0.01). It is noteworthy that the expressing states of the β-actin in 5-Aza and H 2 group were found, whereas noticeable difference was identi ed between the two groups (p<0.01).

Observation and analysis of myotube
Mhc protein immuno uorescence staining assay was performed to visualize the morphology of the formed myotubes in ADSCs. Fig. 3B shows the results of immuno uorescence staining assay after incubation for 7 days. The myotube number, length, diameter and maturation index, were quanti ed by morphological analysis. Through the determination of the ratio of myotube number with more than 2 nuclei to the total myotube number (myotube maturation index), the myotube maturation was quanti ed. Compared with control group, 5-Aza+SB203580 and H 2+ SB203580 groups promoted the high expression of Mhc in the number and length of myotubes(p<0.01), and there were statistical differences in the diameter and fusion index of myotubes(p<0.05). It is noteworthy that this study signi cantly high positive Mhc staining was found in 5-Aza+H2+SB203580 group(p<0.01). As compared with 5-Aza+SB203580 and H 2 +SB203580, 5-Aza+H 2 +SB203580 demonstrated the signi cantly high myotube number, myotube length, myotube diameter and myotube maturation index (Fig. 3B).

Real-time PCR analysis of Myod and Mhc
We further investigated the in uence of the H 2 on myogenic genes expression of ADSCs in vitro. Myod and Mhc, the early and late markers of myogenesis, were used to determine the myogenic differentiation at mRNA level (Fig. 3C). After being cultured for 7 days, the quantitative RT-PCR results showed remarkably higher mRNA expressing states of MyoD and Mhc in the H 2 +5-Aza+SB203580 group than other four group(p<0.01). The Myod and Mhc gene expression of 5-Aza+SB203580 and H 2 +SB203580 was almost the same as control group(p>0.05). However, compared with the SB203580 group, the expression of MyoD and MHC genes was signi cantly higher in the H 2 group(p<0.01). No noticeable difference was identi ed in Myod and Mhc RNA expressions among the H 2 +SB203580 and 5-Aza+SB203580 groups(p>0.05). As impacted by SB203580, the expressions of Myod and Mhc of the inhibitor groups was also a little less than control group(p<0.01).

Western blotting assay of Myogenin and Myod
In addition, under the expressing states of Myogenin and Myod (early myogenic marker) and the differentiated myotube marker myosin heavy chain (Mhc), we conducted the western blotting assay for assessing myogenic differentiation (Fig. 4A c-e). As revealed from the results of the western blotting assay, remarkably higher stripe expressing states of Myod, Mhc and Myogenin were found in the H 2 group, 5-Aza group and 5-Aza+H 2 group as compared with those in the control (p< 0.01). Besides, the expressing states of Mhc and Myogenin were noticeably higher in the H 2 +5-Aza group than those in the H 2 group (p< 0.01). It is worth noting that the expressing states of Myod was no noticeable difference was identi ed in expressing states between the 5-Aza+H 2 group and the H 2 group(p> 0.05).

Effect of hydrogen on the phosphorylation of p38 in myogenic differentiation
According to the mentioned results, H 2 could noticeable improve ADSCs proliferation and myogenic differentiation, whereas the molecular system was unclear. Through the analysis of bioinformatics and literature reports, it is known that p38 MAPK was indicated to participate in a range of biological processes of myogenic differentiation. Accordingly, if the phosphorylation levels of p38 MAPK were affected by H 2 regulation in the myogenic differentiation was assessed. ADSCs were incubated in DMEM for 7 days. According to Fig. 4A, the level of p38 and p-p38 phosphorylation in the 5-Aza and H 2 cotreatment group was signi cantly greater than that in the control (p<0.01), which indicated an enhancement in the active form of p38 that promotes differentiation. This result suggests that 5-Aza and H 2 cotreatment during myogenic differentiation could improve p38 activity. It is noteworthy that the levels of p38 and p-p38 in the 5-Aza and H 2 single-treatment groups also signi cantly increased compared with those in the control, which indicated that either 5-Aza or H 2 treatment during myogenic differentiation increases p38 activity 3.5 The myogenic differentiation facilitated by hydrogen is damaged by the pharmacological inhibition of p38 H 2 regulation improved p-p38 in ADSCs. The mentioned effect, however, was abolished by the presence of speci c concentrations of SB203580 (5 nM), a pharmacological inhibitor of p38 that has been established and applied extensively. In the continued presence of SB203580, rare myotubes were detected based on immuno uorescence staining in the control (Fig. 3B). However, major myotubes with Mhc protein staining received the testing process in H 2 , 5-Aza and H 2 +5-Aza groups (Fig. 3B) However, both groups were remarkably higher than the SB203580 group (p<0.05). As impacted by SB 203580, the expressions of Myod and Mhc of the inhibitor groups were slightly lower than those of the normal group (p<0.05) (Fig. 3C). For a broader analysis of the variations in protein, the Western blotting assay was conducted on Myod and Mhc proteins after 7 days of incubation in DMEM with or without SB203580. In addition, as compared with SB203580 and H 2 +SB203580 groups, control and H 2 groups had stronger bands, a sharp decline was identi ed when SB203580 was added. SB203580 groups showed the slight Myod and Mhc protein bands, whereas the rest of the three groups (i.e., H 2 +SB203580, 5-Aza+SB203580 and H 2 +5-Aza+SB203580) obviously revealed the protein bands. Subsequently, the protein expression of signaling pathway during the myoblast differentiation was detected. There was decline levels of p38 during the introduction of the inhibitor. However, the clear protein band was still found in four groups (i.e., control, H 2 +SB203580, 5-Aza+SB203580 and H 2 +5-Aza+SB203580). As revealed from the comparison of the four groups, the band of H 2 +5-Aza+ SB203580 was the clearest and strongest. The protein bands were similar in expression for H 2 +SB203580 and 5-Aza+SB203580, and much clearer and stronger than ADSCs group. For this reason, SB203580 prevents muscle fusion, whereas three group except for SB203580 could still promote myogenic differentiation as impacted by SB203580. In brief, H 2 can activate p38 MAPK signaling pathway by elevating the level of p-p38 protein and facilitate myogenic differentiation of ADSCs.

Discussion:
The aim of this study was to investigate the promotive effects of H 2 in proliferation and myogenic differentiation processes of ADSCs as well as the possible signaling pathways involved. H 2 has been proven to be used in multiple biological systems, including those in the Cardiovascular, Digestive and Motor System [28]. In the present study, the H 2 effect can result from the following: 1) reinforcing the single-cell mitochondrial number; 2) stimulating the myogenic biomarking genes' expressing state pertaining to Mhc and Myod, etc.; 3) promoting p38 phosphorylating process inside MAP kinases (MAPK) signaling pathway, by leading the promoted myoblast differentiation.
Over the past few years, uses of H 2 have been largely anticipated as novel medical treatments [29]. H 2 has been employed in different forms to various disease models, and research on its curative effects has progressed rapidly [30,31]. In the present study, H 2 -induced ADSCs were con rmed to exhibit a high biocompatibility in vitro based on MTT and Live-Dead Cell Staining (Fig. 2). It is noteworthy that H 2 can still alleviate the cytotoxicity of 5-Aza, thereby enhancing the cell viability. Accordingly, H 2 can be signi cantly ensured to promote the myogenic differentiation of ADSCs. Mitochondria, one of the vital intracellular organelles, signi cantly impacts various biological processes of eukaryotic cells (e.g., energy generation, calcium balance, intracellular substance metabolism, reactive oxygen production, cell signal transduction and apoptosis] [32][33][34]. As indicated from the recent advances, adequate mitochondrial function in stem cells is essential to maintain proliferation and differentiation abilities [35,36]. Accordingly, green mito-tracker staining was adopted to explain the effect of H 2 on the mitochondria of ADSCs. The uorescence intensity of single cell mitochondria signi cantly increased after H 2 induction.
For this reason, the promoted proliferation of ADSCs was probably because H 2 could increase the mitochondrial number.
Myod and Mhc, the early and late markers of myogenesis [37,38], were used to determine the myogenic differentiation at mRNA, protein and myotube formation. In the present study, levels of Myod and Mhc both increased signi cantly when H 2 induction. Notably, H 2 and 5-Aza can synergistically promote the myogenic differentiation of ADSCs. In ADSCs was evaluated using desmin immuno uorescence. Desmin, a muscle-speci c member of the family of intermediate laments, is one of the earliest appearing myogenic markers in both skeletal and heart muscles [39]. Our immuno uorescence results showed that Desmin was signi cantly overexpressed in H 2 -induced ADSCs. Interestingly, in myotubule observation, H 2 and 5-Aza still promoted myotubule maturation under the continuous action of SB203580 (Fig. 3B).
Therefore, this result indicates that the rst is that 5-Aza and H 2 can compete with the p38 MAPK signaling pathway to promote myobgenic differentiation. The second is that H 2 promotes myogenic differentiation via the p38 MAPK signaling pathway, but not only via the p38 MAPK signaling pathway (Fig. 3B). It has been reported that the myogenic differentiation and myoblast's myotube formation noticeably relied upon cell proliferation [40]. The balance between myoblast proliferation and differentiation is important during muscle development [41]. In the present study, the results of the above cell proliferation and myogenic differentiation show a certain correlation. The process of myogenic differentiation into myotube formation is often accompanied by changes in mitochondrial energy metabolism and ROS production. Previous studies have shown that Reactive oxygen species (ROS) is essential mediators of muscle differentiation [42], and it has long been associated with skeletal muscle physiology [43,44]. However, with the accumulation of ROS, due to its strong oxidation, it can cause irreversible damage to proteins, nucleic acids, sugars, lipids, etc., which signi cantly inhibits cell activity and leads to cell apoptosis [45,46]. In the process of myogenic differentiation of stem cells, intracellular ROS level is signi cantly increased, and the expression of apoptotic proteins such as p53 and other genes is also signi cantly increased, and cell activity is signi cantly inhibited [47,48].
Previous studies have shown that H 2 can reduce ROS level in radiation-injured mice, reduce liver damage, and inhibit radiation-induced apoptosis, thus proving that H 2 can play a protective role on radiationinduced immune system injury by eliminating ROS [49]. The observation that H 2 treatment signi cantly improved the level of SH-SY5Y ATP and Δψm in neuroblastoma [50] is an indication that H 2 treatment can elevate energy metabolism in mitochondria by activating oxidative phosphorylation. In conclusion, H 2 can promote mitochondrial oxidative phosphorylation and maintain ROS dynamic balance to effectively protect the cell damage in the differentiation stage, and further promote the myogenic differentiation of stem cells. However, how mitochondrial function changes in the process of H 2 -induced myogenic differentiation of stem cells remains to be studied.
As revealed from the mentioned results, H 2 could remarkably enhance ADSCs proliferation and myogenic differentiation, whereas the molecular system was unclear. Here, based on the p38 MAPK classes refer to signal transducing elements promoting myogenic differentiation in vitro and in uencing muscle growing and repairing in vivo, whereas only p38 MAPK has a direct effect on myogenic transcribing elements of the Myod class [51,52]. And through bioinformatics analysis infer H 2 was speculated to be involved in the p38 MAPK signaling pathway, probably affecting myogenic differentiation. During the differentiation, skeletal muscle cell can proliferate, migrate, subsequently seed from the cell cycle associated with an improvement in p38 MAPK signaling activity and then fuse to form multinucleated myotubes [53,54]. To verify the probability of whether p38 signaling activity is required for myogenic differentiation induced by Availability of data and materials The data used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing Interest
The authors declare that they have no competing interests.

Funding
This work was supported by the funds from the young medical key talent project of Jiangsu province (contract number QNRC2016458). Jiangsu Provincial Medical Innovation Team (Grant#CXTDB2017004).
Erkai Pang made the equal contribution to the article and should be considered co-rst author.
Lei Hou, Jihang Dai, Mingsheng Liu, Xuanqi Wang, Bin Xie,:analysis and interpretation of data and drafted the manuscript.
All authors read and approved the nal manuscript.    Figure(2D, E). Cell viability was assessed by MTT. Figure(2F). Selection of the optimum concentration of hydrogen. All experiments were performed in triplicate (*p< 0.05, **p< 0.01).