The Protective Effect of Epimysium Preservation and FTY720 on Functional Gracilis Transfer: Experimental Rabbit Study


 Background Free functional gracilis transfer (FFGT) is one of the most important methods for the treatment of total brachial plexus avulsion. However, postoperative adhesion and fibrosis of the transferred muscle compromise the effectiveness. This study is to investigate whether intact epimysium and FTY720 could reduce adhesion and fibrosis following FFGT in rabbits.Methods A total of 60 New Zealand White rabbits were randomly divided into 4 groups: Group 1 was set as control, and orthotopic gracilis transfer models were established in Group 2-4, with different interventions: epimysium resection (Group 2), epimysium preservation (Group 3) and epimysium preservation combined with FTY720 treatment (Group 4). After that, at each time points (1, 3, 7, 14, and 56 days postoperatively), three rabbits in each group were selected to harvest their gracilis and adjacent muscles for histological analysis including hematoxylin & eosin (HE) staining, Masson staining, and anti-transforming growth factor beta 1 (TGF-β1) immuno-histochemical staining. Additionally, the thickness of extragracilis adhesion band by ultrasonography were also assessed at days 56.Results All of HE, Masson and TGF-β1 stains confirmed that there were least inflammatory cell infiltration, adhesion and fibrosis of the transferred gracilis in Group 4 than those in Groups 2 and 3. At the 56 days postoperatively, the thickness of the adhesion band of gracilis was lightest in group 4 than that in Groups 2 and 3. Conclusions For the functional gracilis transfer, epimysium preservation and FTY720 could relieve the inflammatory response between gracilis and adjacent muscles, reduce the fibrosis and adhesion.

adhesion and brosis of the transferred gracilis in Group 4 than those in Groups 2 and 3. At the 56 days postoperatively, the thickness of the adhesion band of gracilis was lightest in group 4 than that in Groups 2 and 3.
Conclusions For the functional gracilis transfer, epimysium preservation and FTY720 could relieve the in ammatory response between gracilis and adjacent muscles, reduce the brosis and adhesion.

Background
Free functional gracilis transfer (FFGT) is one of the most important methods for the treatment of total brachial plexus avulsion. Satisfactory outcomes following FFGT depend on many factors [1][2][3][4]: reliable blood supply, su cient axonal regeneration and target muscle reinnervation, ne muscle tension regulation, nerve matching between donor and recipient sites, prevention of adhesion and brosis of the transplanted muscle, postoperative rehabilitation and so on. In recent years, surgeons have paid more attentions to the prevention of adhesion and brosis of the transferred muscle, which affected by integrity of epimysium to a certain extent [5][6][7][8].
Previously, there was a high rate of adhesion of the gracilis and tendon following FFGT due to the broken epimysium of gracilis [9]. With technical advances in microsurgery and ongoing work in FFGT, Doi and his colleagues [10] mentioned "extra-muscular operation" in the report of gracilis harvesting technology. Seal et al [5] showed that free functional gracilis should be harvested with extended deep fascia and super cial vein included for venous out ow augmentation. To a certain extent, it could improve the postoperative function of the gracilis, but it was lack of animal experimental research.
Fibrosis is another important factor affecting the postoperative function of gracilis. Sphingosine-1phosphate (S1P) is a bioactive sphingolipid mediator involved in many physiological processes including angiogenesis and immune responses [11,12]. S1P signaling has been found to be essential for vascular development, neurogenesis, and lymphocyte tra cking [13][14][15], as well as a second messenger during in ammation [16,17]. Many of the actions of S1P in innate and adaptive immunity are mediated by its binding to ve speci c G protein-coupled receptors, S1P receptors (S1PRs) 1-5. To date, a number of S1P receptor modifying compounds have been developed [18]. FTY720 (Fingolimod), a functional antagonist of S1PR and originally discovered by chemical modi cation of a natural product (myriocin), has clari ed that S1P is important for the recruitment of various types of in ammatory cells together with other S1PR modifying compounds [19,20].
There is an in ammatory microenvironment between the transferred muscle and the recipient site.
Macrophages play an important role in the repair of injured muscles. Relevant studies have con rmed that S1P participates in the regulation of in ammation by binding to S1PR on macrophage membrane.
Moreover, we found that FTY720 could reduce the systemic and local in ammatory response after the muscle transfer, reduce the in ltration of macrophages, and ultimately reduce the brosis of the transplanted muscle in previous animal experiments.
In this experimental study, we aimed to investigate whether epimysium preservation and FTY720 could reduce adhesion and brosis following functional gracilis transfer in rabbits.

Materials And Methods
Experimental animals 60 male New Zealand White rabbits (8-12 months old, weight 2.5±0.3kg) were purchased from the Experimental Animal Center of the First A liated Hospital of Sun Yat-sen University (Guangzhou, China).
The rabbits were housed and acclimatized to a standard laboratory diet and tap water under climatecontrolled conditions (25℃, 55% humidity, 12-h light/dark cycle). The study was approved by the animal ethics committee of the First A liated Hospital of Sun Yat-sen University [Approval No. (2019)377] and was performed according to guidelines for the use of laboratory animals.

Surgical procedure
After fasting for 12 hours, rabbits were anesthetized with 3% pentobarbital sodium 1 ml/kg by ear vein injection. The medial thighs were shaved and each animal was positioned supine on the operating platform with legs taped laterally (to facilitate access to the gracilis muscle). The gracilis muscle is large in the rabbit and a longitudinal incision parallel to the femuris was made through the skin in the medial thigh. The gracilis muscle was then dissected from the tibial part and elevated, and exposed associated branch of the obturator nerve and deep femoral vessels with preservation of the pubic attachments. The epimysium around the gracilis was preserved in group 3 ( Fig 1A) while the epimysium was removed in group 2 ( Fig 1B). Then deep femoral vessels were clipped for half hour and the obturator nerve was cut off. Under a standard operating microscope, obturator nerve coaptations were performed using 10-0 nylon sutures. The gracilis was sutured back into its original position and the skin was closed using interrupted 4-0 sutures.

Experimental groups and drug delivery method and dose
The rabbits were divided randomly into 4 groups (n = 15 each) using a block randomization method, including control group as group 1, none epimysium group as group 2, epimysium preservation group as group 3 and epimysium preservation + FTY720 group as group 4. FTY720 (Fingolimod) was purchased from Selleck Chemicals (USA), dissolved in sterile saline, and stored at -20°C. For group 4, 10 mg/kg FTY720 was administered intraperitoneally daily for 3 days after operation.

Cryosectioning and Histological Analysis
Gracilis and adjacent muscles were harvested at 1 day, 3 days, 7 days, 14 days, and 56 days for histological analysis after the rabbits were anaesthetised. The histologic block xed in 4 % paraformaldehyde and embedded in para n. Three sagittal sections for each block were made perpendicularly to the line of the postsurgical insertion of the gracilis, adductor longus and adductor magnus muscle. Tissues in each group were examined microscopically using hematoxylin & eosin (H&E) staining to general histologic observation. In ammation was quanti ed by averaging the number of in ammatory cells in three high powered eld views (400x) on light microscopy of intermuscular space.

Ultrasonography examination
A portable ultrasonic equipment VENUE 40 (GE, American) with an 8.2-MHz linear array transducer was applied to evaluate tissue proliferation between the gracilis muscle and adductor longus and adductor magnus muscle at 56 days. The hypertrophic scar tissue was usually characterized by a band and irregular middle echo in the image. The proximal, middle and distal part of the gracilis muscle of each experimental animal were measured respectively, and the average value was taken.

Statistical analysis
All data were presented as the means ± SD and were analyzed by ANOVA to evaluate the differences between groups by Statistical Package for Social Sciences 25.0 software (IBM Corp, NY, USA). A P value < 0.05 was considered statistically signi cant.

Results
Histology The in ammatory cell in ltration in the cross-sections between gracilis muscle and adductor magnus and adductor longus muscle in different groups is presented in Fig. 2. HE-stained sections ( Fig. 2A) showed different characteristics of in ammatory cell in ltration among groups: no in ammatory cell in ltration was found in group 1, severe in ammatory cell in ltration was found from day1 to day 3 in group 2 (especially in the gracilis), obvious in ammatory cell in ltration continued from day 1 to day 3 in group 3, and only mild in ltration was found from day1 to day 3 in group 4. Multifocal calcium deposits formed and muscle space disappeared from day 7 to day 14 in group 2. Compared with group 4, muscle space became narrow and brous tissue proliferated obviously on day 7, day 14 and day 56 in group 3.
Additionally, Masson staining of the gracilis muscle in each group on day 56 showed no obvious new collagen deposition in group 1, severe collagen deposition in group 2, moderate collagen deposition ingroup 3, and a few collagen depositions in group 4 (Fig. 3).
HE and TGF-β1 stains con rmed the most severe in ammatory cell in ltration and adhesion in group 2.

Discussion
Traumatic brachial plexus injuries are devastating events that frequently result in severe chronic functional impairment. At present, the main treatment methods are multiple nerve transfer and nerve transfer combined with free functional gracilis transfer. The indications for FFGT surgery include presentation beyond 9 months from injury, unavailable or failed prior local tendon or nerve grafts or transfers, and as the primary procedure for reconstruction of elbow exion [21]. Functional gracilis transplantation can lead to muscle brosis due to ischemia-reperfusion injury. At the same time, there is in ammatory response in the recipient site of the transplanted muscle, leading to tissue adhesion. In clinic, some surgeons have proposed to improve operative gracilis function by using modi ed gracilis muscle harvesting technique (preserving the epimysium), but it is lack of animal experiments.
The preservation or recreation of a plane of differential tissue movement or gliding is vital for the maintenance of function, particularly where intricate movements are required such as in the hand and upper limb. Fascia has been used as a free graft or vascularized ap in an attempt to reconstitute a gliding surface. However, when the epimysium is disrupted the fascia adopts the pattern of a scar which spans the tissue planes and this will limit movement between these planes [22].
As we known, the endomysium, perimysium, epimysium and deep fasciae have not just a role of containment, limiting the expansion of the muscle with the disposition in concentric layers of the collagen tissue, but are fundamental elements for the transmission of muscular force, each one with a speci c role [23,24]. Where the epimysium of two muscles connects, there is frequently a connective tissue membrane that transports the vessels and the nerves destined to reach these muscles. Primarily the epimysium is subjected to mechanical tension and to forces that act orthogonally on its internal and outer surface, so the epimysium is very important for muscle contraction. There is an in ammatory microenvironment between the transferred muscle and the recipient site. Ischemia or reperfusion (I/R) injury in free muscle transfer in reconstructive surgery leads to a signi cant up-regulation of in ammatory parameters, in ltration of in ammatory cells, and angiogenesis [25]. Our study demonstrated that the epimysium could be used as a physical barrier to isolate the transplanted muscle from the external environment. It also could reduce the in ltration of in ammatory cells into the gracilis, reduce the formation of postoperative adhesion, and improve the gracilis function.
Moreover, FTY720, a S1PR1 agonist, has been widely used in clinical and experimental research of brosis. Numerous studies have shown that FTY720 can reduce lung, liver and kidney brosis, and scar formation [26][27][28][29]. In the muscle injury model of rats, FTY720 can effectively reduce the local in ammatory cell in ltration, and make macrophages enter M2 phase ahead of time, reduce the severity of muscle injury, improve muscle regeneration and reduce muscle brosis [30,31]. Some studies showed that the application of an FTY720-loaded lm increased M2 monocyte/macrophage recruitment, as well as arteriogenesis in ischemic skeletal muscle [32][33][34]. FTY720 can signi cantly reduce the degree of myocardial brosis and reduce the mortality of experimental animals in the model of myocardial ischemia-reperfusion [35]. In this study, FTY720 was used to further intervene the experimental animals with preserved epimysium. It was found that the in ltration of in ammatory cells in the early stage decreased between the interface between the gracilis muscle and adductor longus muscle and adductor magnus muscle in the group 3 and group 4, compared with group 2, as well as the collagen deposition, the thickness of adhesive band and the positive expression of TGF-β1.This is similar to the above results. Therefore, S1P / S1PR1 axis maybe take part in the in ammatory process of gracilis and recipient site, but it needed further study by cell experiment in vivo.
There are several limitations of this study. First, the study is based on the short-term observation of gracilis function without long-term follow-up to assess functional recovery of gracilis. Second, we use a the gracilis muscle orthotopic transplantation model instead of free gracilis transfer model in this study.

Conclusions
As a physical barrier, epimysium can reduce the in ltration of in ammatory cells into gracilis and protect gracilis. FTY720 (S1PR1 agonist) can further improve the in ammatory response between gracilis and adjacent muscles, reduce the brosis of gracilis and adhesion with adjacent muscles.
Declarations Acknowledgements Not applicable.
Authors' contributions CJP, LZP and JA contributed signi cantly to performing the experiments, analysis, and manuscript preparation. GFB and FJT were involved in the data collection and statistical analysis. QBG and WHG performed the analysis with constructive discussion. GLQ and YJT were responsible for the project coordination and approved the nal version. All authors read and approved the nal manuscript.

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request, taking into account any con dentiality.

Ethics approval and consent to participate
All procedures performed in this study were in accordance with the ethical standards of the institution (Animal Care and Use Committee of the First A liated Hospital of Sun Yet-Sen University).

Consent for publication
Not applicable.

Competing interests
All authors declare that they have no competing interests.     TGF-β1 staining results: (a, e) no brownish yellow particle deposition in group 1, (b, f) severe adhesion with a lot of brownish yellow particle deposition (red arrow) in group 2, (c, g) moderate adhesion with numerous brownish yellow particle deposition (red arrow) in group 3, and (d, h) mild adhesion with a few brownish yellow particle (red arrow) deposition in group 4. (B) Different TGF-β1 positively stained cell distributions in different groups; it is the highest in group 2 than other groups. *** P< 0.001; scale bar=50μm