Human Adipose-Derived Stem Cells Delay Muscular Atrophy after Peripheral Nerve Injury in Rats

Given that denervation atrophy often occurs in muscle after peripheral nerve injury, the effects of injections of human adipose-derived stem cells (hADSCs) and platelet-rich plasma (PRP) into muscle after peripheral nerve injury were examined. hADSCs were isolated from human subcutaneous fat tissue, and PRP was prepared from rat whole blood before injection into a rat sciatic nerve injury model. Muscle atrophy was evaluated by quantitating the gross musculature and muscle fiber area and walking footprint analysis. At 4 weeks post-surgery, there were significant differences in the sciatic functional index between experimental (injected with hADSCs, PRP, or combined hADSCs + PRP) and non-operated groups (p < 0.0001), but no significant differences were observed between the different treatment groups (p > 0.05). Post hoc Bonferroni tests also showed significant differences in the wet muscle weight ratios of hADSC, PRP, and combined groups compared to PBS group. The gastrocnemius muscle fiber area was larger in hADSC group and the combined group compared to PBS group at 4 weeks post-surgery. The injection of hADSCs delays muscular atrophy after sciatic nerve injury in rats; thus, hADSCs are a promising alternative for regenerating atrophied muscle.


Introduction
Denervation atrophy often occurs in muscle after peripheral nerve injury, making treating peripheral nerve injury more complicated. For instance, if atrophy of facial muscles occurs after facial nerve injury, nerve and muscle transplantation is required to reconstruct the facial expression muscle dynamics [1]. In clinical practice, delaying or even recovering the denervation atrophy in muscle will shorten the duration of peripheral nerve therapy and improve the therapeutic effect. It has been shown that isoflavones reduce apoptosisdependent signals and significantly alleviate muscle atrophy after denervation in mice [2], whereas apigenin inhibits denervation-induced muscle atrophy by inhibiting inflammatory processes in muscles [3]. Appropriate physical stimulation such as electric stimulation and electroacupuncture can alleviate the denervation of muscular atrophy [4,5]. However, the mechanism of muscular denervation atrophy is highly complex; thus, various treatment methods exist. New treatment and remission methods will be expected to be developed for muscular atrophy denervation.
Adipose-derived stem cells (ADSCs) are widely available and easy to obtain, and the donor age and collection site do not affect the therapeutic effect of the obtained stem cells [6]. ADSCs can differentiate into various cell lineages and have strong anti-inflammatory, anti-fibrosis, antiapoptosis, and pro-angiogenesis effects in vitro or in vivo. Therefore, ADSCs have become the first choice for preclinical studies [7], and they have been widely applied in the field of cell therapy and tissue engineering regeneration and repair [8][9][10]. Previous studies have revealed that ADSC transplantation can delay muscular atrophy caused by peripheral nerve injury by inhibiting the inflammatory responses [11]. Based on the low immunogenicity of stem cells in allotransplantation [12], human adipose-derived stem cells (hADSCs) were injected into the brain of adult rats [13], displaying that the survival rate was also high under heterogeneous conditions, meeting the ideal standard of cell therapy. Therefore, injecting hADSCs into the gastrocnemius muscle following peripheral nerve injury in rats was explored in this study to verify whether allogeneic adipose stem cells can also delay muscular atrophy caused by peripheral nerve injury.

Isolation and Culture of Human Adipose-Derived Stem Cells (hADSCs)
The hADSCs were harvested from the human subcutaneous fat tissue of healthy donors who underwent liposuction [14]. Using human adipose tissue was approved by the Ethics Committee of Plastic Surgery Hospital (No. 2150019022). The fresh fat tissue was washed thrice with PBS containing 1% penicillin/streptomycin, then digested with 0.25% collagenase type I (Sigma, USA) at 37°C for 0.5 h and centrifuged at 1000 rpm for 5 min. The cell pellet was resuspended and filtered through a 70-μm filter (Corning, USA), then centrifuged for 5 min before the cell pellet was resuspended in high glucose Human Mesenchymal Stem Cell Medium (MSCM, ScienCell, USA). The culture medium was replaced 48 h after seeding to remove nonadherent cells, then replenished every 2-3 days. The hADSCs were passaged three times for experiments, and hADSC immunophenotype was confirmed by flow cytometry using Human MSC Analysis Kit (BD, New Jersey, USA). The cell samples were labeled with each antibody separately, and after processing, concentrated cell populations were gated, and the percentage of cells labeled with the selected antibodies was identified as proposed by the International Federation for Adipose Therapeutics and Science (IFATS) and International Cell Therapy Society [15]. The isotype was used as a reference, and analysis was performed using FlowJo7.6.1 Software.

Preparation of Platelet-Rich Plasma
Rats were placed under general anesthesia by intraperitoneal injection of sodium pentobarbital (50 mg/kg), and whole blood was drawn through cardiac puncture and transferred into a 15 mL centrifuge tube containing 1.5 mL sodium citrate. After centrifugation at 2500 rpm for 10 min, the whole blood was separated into three layers, including the plasma (upper layer), buffy coat (intermediate layer), and red blood cell (bottom layer) (Fig. 1A). The upper layer and intermediate thin layer were transferred to an empty sterile tube and centrifuged at 1000 rpm for 10 min at room temperature (Fig. 1B). The top two-thirds of the supernatant consisting of platelet-poor plasma (PPP) was removed to leave PRP layer (1/3) [14]. The number of platelets in PRP was counted and was about three times the concentration of platelets in the whole blood.

Sciatic Nerve Injury Model Construction
All animal experiments conducted in this study were reviewed and approved by the Local Animal Ethics Fig. 1 A After the first centrifugation, the blood sample was divided into three layers. The upper layer was plasma, the middle was the white blood cell layer, and the lower layer was red blood cell. B After the second centrifugation, the bottom layer was the red blood cells, the white layer was accumulated platelets, and the supernatant was platelet-poor plasma. The top two-thirds of the supernatant which consisted of platelet-poor plasma (PPP) was removed. The remaining layer (1/3) was considered as PRP Committee (No. 202003003). Surgeries were performed by an experienced surgeon under a neurosurgical microscope (M400-E, Leica, Germany). Rats were placed under general anesthesia by intraperitoneal injection of sodium pentobarbital (50 mg/kg). At about 0.5 cm below the midpoint of the femur in rats, the skin was cut parallel to the femur, and then the biceps femoris muscle was bluntly separated to expose the sciatic nerve trunk, which was cut with straight microscissors. The severed ends retracted to produce a gap of about 0.5 cm ( Fig. 2A, B). The surgical wound was closed with 4-0 nylon sutures, and the rats were returned to their cages. The whole process was performed under aseptic conditions. Rats were randomly divided into the five groups (n = 10 each): injury without ischiatic nerve (A: non-operated group); ischiatic nerve was transected, and the gastrocnemius muscle was injected with phosphate-buffered saline (PBS) (PBS group) (Fig. 2C); ischiatic nerve was transected, and the gastrocnemius muscle was injected with 0.5 mL of 10 6 /mL hADSCs in PBS (hADSC group); ischiatic nerve was transected, and the gastrocnemius muscle was injected with 0.5 mL PRP (PRP group); ischiatic nerve was transected, and the gastrocnemius muscle was injected with a mixture of 0.25 mL PRP and 0.25 mL hADSCs (combined group). Before experimentation, rats were injected with methylene blue, and complete muscular dissection revealed diffuse injection throughout the gastrocnemius muscle (Fig. 2D). The experimental groups were injected a week thrice after the operation. All experiments were approved by the Local Animal Ethics Committee.

Walking Footprint Analysis
Before the ischiatic nerve was transected, 1, 2, 3, and 4 weeks post-operation, walking track assessments of all animals sacrificed at each time point were performed. Briefly, the walking footprint was analyzed using Digigait System (Mouse Specifics, Framingham, MA, USA). The rats were trained at a speed of 25 cm/s for testing. For each test, at least 2 s of continuous walking were recorded, and the footprints were captured and analyzed using Digigait analysis software (Digigait 12.4). The sciatic functional index (SFI) for each time point was calculated as follows: where EPL is the experimental print length, NPL is the normal print length, ETS is the experimental toe spread, and NTS is the normal toe spread [16].

Gastrocnemius Muscle Analysis
The rats were randomly selected and sacrificed by cervical dislocation 2 and 4 weeks after surgery. The double lower Fig. 2 A The sciatic nerve trunk was exposed. B Two severed ends of the sciatic nerve retracted and a gap of about 0.5cm appeared. C The gastrocnemius muscle was injected with phosphate buffered saline. D The pros and cons of complete muscular dissection revealed diffuse injection throughout the gastrocnemius muscle limbs were cut open, and the intact gastrocnemius was cut off. The bilateral gastrocnemius muscles were blotted with absorbent paper before weighing. The remaining rate of wet muscle weight (WMW) was calculated as follows: WMW on the operation side/WMW on the healthy side ×100% [17].
Next, the muscles were fixed in 4% paraformaldehyde, dehydrated with a serial gradient of ethanol, and cleared in xylene. The specimens were then embedded in paraffin and sectioned into 5-μm slices before staining with Masson's trichrome (Solibor, Beijing, China) and observed under a microscope (Leica). Digital images were captured, and muscle fiber areas were calculated using Image-pro plus 6.0 software. Each sample counted 50 muscle fibers, and the area was expressed as mean ± SD per group.

Statistical Analysis
All data are presented as the mean ± standard error of the mean (SEM). GraphPad Prism 6.0 (USA) was used for statistical analysis, and the data were analyzed by two-way ANOVA with factors of treatment and time. Intergroup differences were analyzed by performing post hoc Bonferroni tests. P ≤ 0.05 were considered statistically significant (### or ***P ≤ 0.001; **P ≤ 0.01; *P ≤ 0.05).

Walking Footprint Analysis
The ischiatic nerve functional recovery was evaluated by a walking track test. The SFI was evaluated weekly using Digigait, and two-way ANOVA revealed no significant interactions between different treatments and times for SFI (F (12, 32) = 0.2193; p > 0.05). Post hoc Bonferroni tests indicated significant differences between experimental and non-operated groups for SFI (p < 0.0001), but there were no significant differences in SFI between the different treatment groups (p > 0.05) (Fig. 4).

Gastrocnemius Muscle Analysis
Recovery of the gastrocnemius muscle from atrophy was assessed by the wet weight of the gastrocnemius muscle ratio and calculating the muscle fiber area. According to the macroscopic appearance of the gastrocnemius muscles (Fig.  5A) and Masson's trichrome-stained sections (Fig. 6A) of 2 and 4 weeks post-surgery, there was a different extent of atrophy in the four treatment groups. Two weeks post-surgery, the gastrocnemius muscle wet weight ratios significantly decreased in the four nerve injury treatment groups (P < 0.0001), but there was no significant difference between the four nerve injury treatment groups. Four weeks post-surgery, post hoc Bonferroni tests showed significant differences in the wet weight ratios of hADSC group (37.01 ± 1.88, p < 0.0001), PRP group (33.45 ± 1.61, p < 0.01), and the combined group (38.590 ± 0.61, p < 0.0001) compared to PBS group (25.20 ± 0.36) (Fig.  5B). The gastrocnemius muscle fiber area in the four nerve injury treatment groups was also reduced (p < 0.0001), with a larger area of muscle fibers in hADSC group (936.83 ± 56.74 μm 2 , p < 0.05) and the combined group (912.01 ± 45.07 μm 2 , p < 0.05) compared to PBS group (642.26 ± 4.31 μm 2 ) at 4 weeks post-surgery (Fig. 6B).

Discussion
Denervation of muscle results in a rapid and programmed loss of muscle and function known as muscular atrophy. It is generally considered that muscle function is irretrievable after prolonged denervation (6-12 months) despite re-acquisition of innervation. This prolonged denervation decreases the number of muscle stem cells, which is detrimental to muscle regeneration after innervation. Previous studies suggested that muscular atrophy is the result of protein homeostasis deficiency, and this process is related to the apoptosis of muscle cells. The apoptotic mechanism remains unclear, and the molecular mechanism controlling the imbalance of protein synthesis and degradation pathways in denervation muscle atrophy has not been fully elucidated. Many studies have explored the inhibition of muscle atrophy: resveratrol has antiaging effects and can relieve metabolic disorders, and it can significantly prevent muscle atrophy after denervation in mice [18]; NF-κB targeted drugs have been used in to delay muscle atrophy, inhibiting apoptosis by inhibiting the classical NF-κB signaling pathway [19]; denervated muscular atrophy treated by injection of the allogenic ADSC showed a decrease in inflammatory factors and delayed muscular atrophy [11,20].
This study explored the use of hADSCs to delay muscle atrophy by promoting the regeneration of muscle stem cells and reducing inflammation [11]. Human adipose stem cells have low immunogenicity and have been widely used in animal experiments to expand their clinical adaptability [21,22]. hADSC injection delayed muscle atrophy as evidenced by changes in the muscle wet weight and muscle fiber area 4 weeks after surgery. This suggests that hADSCs require time to act in the muscle, consistent with Schilling's view that ADSCs need to overcome the trauma of injection to exert their regenerative effect [11]. The present study also showed similar effects in hADSC and combined groups, suggesting that hADSCs might be the key factor in the treatment as PRP injection alone had little effect.  A General observation at 4 weeks after surgery. B Wet weight ratio were mearsured to evaluate atrophy of the gastrocnemius muscle. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, vs. Non-operated group. Data are expressed as the mean ± SD (n = 3; two-way ANOVA analysis followed post hoc multiple comparisons with a Bonferroni correction) Fig. 6 Masson's trichrome staining of gastrocnemius muscle from the experimental side at 2 and 4 weeks postsurgery, and assessment of the muscle fiber area at 2 and 4 weeks post-surgery. A Crosssections with Masson's trichrome staining at 2 and 4 weeks after surgery. ×100, Scale bar: 100 μm. 400X, Scale bar: 25 μm. B Gastrocnemius muscle fiber areas were measured to evaluate atrophy of the gastrocnemius muscle. *P < 0.05, ****P < 0.0001, vs. Non-operated group. Data are expressed as the mean ± SD (n = 3; two-way ANOVA analysis followed post hoc multiple comparisons with a Bonferroni correction) PRP is a useful treatment method used in orthopedics, oral surgery, plastic surgery, dermatology, and other medical fields [23][24][25][26]. PRP contains a high concentration of platelets and growth factors [26], with α particles in the platelets responsible for promoting stem cell regeneration and soft tissue remodeling. PRP particles contain many basic growth factors, such as platelet-derived growth factor, vascular endothelial growth factor, epithelial growth factor, transforming growth factor, insulin-like growth factor, etc. [27,28]. These growth factors are also thought to induce cell proliferation, angiogenesis, and chemotaxis and contain serotonin, dopamine, histamine, adenosine, and calcium, all of which increase cell membrane permeability [29]. Studies have revealed that PRP can promote the recovery of arthritis models induced by proinflammatory cytokines with the assistance of collagen protein, displaying certain anti-inflammatory abilities [30]. Bendinelli et al. exhibited that PRP exerts an antiinflammatory effect on human chondrocytes by inhibiting NF-κB via HGF [30]. Previous studies have found that the immunomodulatory and anti-inflammatory effects of ADSCs contributed to reduced recruitment of inflammatory cells and acceleration of the conversion from the inflammatory phase to the repair phase, and ADSCs can inhibit denervation muscular atrophy [11,20]. Therefore, we sought to inhibit denervation atrophy by injecting PRP to inhibit inflammation and promote cell regeneration and the proliferation of stem cells. However, the walking analysis disclosed that sciatic nerve function was not restored, eliminating the impact of nerve regeneration on muscle. The wet weight of gastrocnemius muscle in PRP injection group increased 4 weeks after operation compared to that in PBS group, and the difference was statistically significant. However, there was no increase in the area of single muscle fibers, indicating that PRP injection did not delay the atrophy of muscle fibers to achieve weight increase but may increase the weight of other tissues, or some substances may not be fully metabolized after PRP injection.
Our study has several limitations. Since this was a 4-week study using a rat model, the long-term prognosis remains unclear. Future studies should extend the trial period and use different clinical models to further explore the mechanism. Finally, we only carried out animal experiments and did not explore the mechanism further. Future studies are needed to unravel these limitations.
In conclusion, hADSC injection delays muscular atrophy after sciatic nerve injury in rats; thus, hADSCs are a promising alternative for regenerating atrophied muscle and have potential application in allogeneic therapy.

Data Availability
All data generated or analyzed during this study are included in this published article and its supplementary information files.
Author Contributions Q.S. performed the experiments, analyzed the data, and wrote the manuscript. M.N., W.W., C.S., W.Q., and L.Y. designed the research, analyzed the data, and contributed to the writing of the manuscript. Y.Z. supervised the study.
Funding This study was supported by the Discipline Construction Project of Peking Union Medical College (201920200401) and CAMS Innovation Fund for Medical Sciences (2021-I2M-1-068).

Compliance with Ethical Standards
Conflict of Interest The authors declare no competing interests.
Ethical approval and consent to participate All experiments conducted in this study were reviewed and approved by the Local Animal Ethics Committee (No. 202003003). Using human adipose tissue was approved by the Ethics Committee of Plastic Surgery Hospital (No. 2150019022). All methods in this study were conducted following relevant guidelines and regulations. All methods are reported in this study following ARRIVE guidelines for the reporting of animal experiments.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.