Aspirin inhibits inflammation and scar formation in the injury tendon healing through regulating JNK/STAT‐3 signalling pathway

Abstract Objectively Tendinopathy is a common problem in sports medicine which can lead to severe morbidity. Aspirin, as the classical representative of non‐steroidal anti‐inflammatory drugs (NSAIDs) for its anti‐inflammatory and analgesic actions, has been commonly used in treating tendinopathy. While its treatment effects on injury tendon healing are lacking, illuminating the underlying mechanism may provide scientific basis for clinical treatment. Materials and methods Firstly, we used immunohistochemistry and qRT‐PCR to detect changes in CD14, CD206, iNOS, IL‐6, IL‐10, MMP‐3, TIMP‐3, Col‐1a1, biglycan, Comp, Fibronectin, TGF‐β1，ACAN，EGR‐1 and FMOD. Next, Western blot was used to measure the protein levels (IL‐6, IL‐10, TGF‐β1, COMP, TIMP‐3, STAT‐3/P‐STAT‐3 and JNK/P‐JNK) in TSCs. Then, migration and proliferation of TSCs were measured through wound healing test and BrdU staining. Finally, the mechanical properties of injury tendon were detected. Results After aspirin treatment, the inflammation and scar formation in injury tendon were significantly inhibited by aspirin. Still, tendon's ECM was positively balanced. Increasing migration and proliferation ability of TSCs induced by IL‐1β were significantly reversed. JNK/STAT‐3 signalling pathway participated in the process above. In addition, biomechanical properties of injury tendon were significantly improved. Conclusions Taken together, the findings suggested that aspirin inhibited inflammation and scar formation via regulation of JNK/STAT‐3 signalling and decreased rerupture risk of injury tendon. Aspirin could be an ideal therapeutic strategy in tendon injury healing.


| INTRODUC TI ON
Tendons are connective tissues which attach muscle to bone and carry mechanical force, permitting joint and whole body movement. 1 Repeated over mechanical force loading leads to tendinopathy, which is characterized as imbalance between micro-tear and repair in tendon. Studies have reported that tendinopathy covers about 30% of sports injuries and its pathogenesis is highly related to inflammation and degradation of connective tissue. 2,3 Inflammation is a starter and necessary for injury tissue healing process, which may generate scar formation in the injury site. At the same time, scar formation may increase the fragility of tendon, which may induce risk of tendon rupture. Based on this, it is important to treat tendinopathy in anti-inflammation and inhibiting scar healing process.
NSAIDs inhibit the inflammation and relieve pain through inhibiting prostaglandin. Aspirin, one of the most commonly used NSAIDs for over 150 years, has been recently used to display multiple effects such as antipyretic, analgesic in cardiovascular system, 4 central nervous system and some cancers. 5,6 Though aspirin has been used to treat tendinopathy, evidence for this treatment is lacking, especially its effect on tendon regeneration healing. 7 In response to micro-damage of tendinopathy, tissues adjacent to the injury area initiate a cascade of inflammation and repair events that are necessary to restore tissue integrity and function.
Some reports have pointed out that inflammation reaction has run through the whole process of tissue repair and has played the role like a double sword. 8 On the one hand, inflammation can induce wound healing and closing through fibrosis and scar formation and prevent wound infection; on the other hand, inflammation-induced scar healing can hamper tissue regeneration and further may reduce the function of tissue or organ. Inflammation is bound to affect tendon extracellular matrix (ECM). Anomalies in the ECM composition of the scar tissue after tendon injury may contribute to a poor and delayed regeneration healing process resulting in compromised tissue quality 9 ; for example, dexamethasone-induced spontaneous tendon rupture and dexamethasone-decreased self-repair capability are very common in clinical practice.
TSCs or progenitor cells were reported firstly in human and mouse tendons in 2007 and were confirmed subsequently in rat and rabbit tendons. [10][11][12] TSCs differed from tenocytes in terms of their colony-forming ability, self-renewal ability and multi-differentiation potential, which enable them to differentiate into tenocytes, adipocytes, chondrocytes and osteocytes, 13 so the viability and tenogenic differentiation of TSCs are tightly associated with the maintenance of the tendon microenvironment and the development of tendinopathy.
The present study aimed to make clear that the effect of aspirin on TSCs viability in vitro, inflammation and regeneration healing process of tendinopathy and the mechanical properties of the injury tendon, and to provide new therapeutic knowledge of aspirin treatment on tendon injury.

| Ethics Statement
Eight-week-old Sprague Dawley rats weighing 200-250 g were used and housed under a 12 hours light/dark cycle in a pathogen-free area with free access to water and food. All animals were treated according to institutional guidelines for laboratory animal treatment and care. All experimental procedures were approved by the Animal Research Ethics Committee of the Third Military Medical University, China.

| Animal model establishment
A total of 24 male Sprague Dawley rats (8 weeks old, 200-250 g) were divided into three groups: the intact group (C), the injury group and aspirin treatment group after injury model establishment (ASA group). In injury and ASA groups, rats were injected with 30 μL type I collagenase solution (10 mg/mL) into both Achilles tendons to establish micro-damaged Achilles tendon model. In ASA group, aspirin (30 mg/d) was given to each rat through gavage administration. One week after establishment of injury model, aspirin was given to each rat in ASA group.

| Isolation and culture of rat TSCs
Isolation of rat TSCs was performed as our previously described. 14 Briefly, the Achilles tendons from both hind feet were dissected after euthanasia. Only the mid-substance tissue was collected, and peritendinous connective tissue was carefully removed. Mid-substance tissue was minced in sterile phosphate-buffered saline (PBS) and digested in 3 mg/mL of type I collagenase (Sigma-Aldrich, St. Louis, MO, USA), 2.5 hours at 37℃. A 70 mm cell strainer (Becton Dickinson, Franklin Lakes, NJ, USA) was used to yield a single-cell suspension. The released cells were washed in PBS, centrifuged at 300 g for 5 minutes and then resuspended in Dulbecco's modified Eagle's medium (DMEM) (Gibco, Carlsbad, CA, USA) with 10% foetal bovine serum, 100 U/mL penicillin, 100 mg/mL streptomycin and 2 mmol/L L-glutamine (all from Invitrogen, Carlsbad, CA, USA).
Isolated cells were plated and grown for 2 days at 37℃ in 5% CO2 and then washed twice in PBS to remove non-adherent cells. On day 7 of culture, the cells were trypsinized with trypsin-EDTA solution (Sigma-Aldrich), mixed together and cultured as passage 0 cells. Cells from passages 3 (P3) were used in the subsequent experiments. TSCs were seeded onto 6-well plates for real-time quantitative PCR (qRT-PCR) and scratch assays, and 10 cm diameter Petri dishes for protein extraction.
TSCs were seeded into a six-well plate at a density of 6 × 10 4 cells/well for each experiment. We added IL-1β (10 ng/mL) as proinflammatory stimulation to establish inflammatory cell model in vitro. To evaluate effect of aspirin on TSCs, we applied IL-1β alone, aspirin (2 mmol/L) alone and IL-1β + aspirin (2 mmol/L) for 24 hours in addition to no-treatment control group. In inhibition studies, TSCs were pre-incubated with JNK inhibitor SP600125 (10 μmol/L) or STAT-3 inhibitor S3I-201 (100 μmol/L) for 1 hour before addition of IL-1β + ASA for 24 hours. Culture medium with or without agents was changed every 3 days throughout the experiments.
Nine sections were cut at a thickness of 4 mm and stained with haematoxylin and eosin. To analyse the changes of total histological scores on HE-stained slides after treatment, we used the established histological scoring system by Stoll 15 et al The score of the intact group was defined as 20 points.

| Immunostaining
At week 4 after aspirin treatment, eight rats each group were sacrificed. The Achilles tendons were harvested for immunohistochemistry. Serial sagittal paraffin sections were prepared from the Achilles tendons as previously described. 16 Briefly, the sec-

| Real-time quantitative PCR
The mRNA expression levels of tendon-related genes were determined using qPCR. Total RNA was extracted from cells using TRIzol reagent, according to the protocol provided by the manufacturer (Takara, Dalian, China). cDNA was synthesized from total RNA using a Superscript III First-Strand Synthesis Kit (TaKaRa). qPCR was performed using a SYBR Green RT-PCR kit (TaKaRa) and an ABI Prism

| Migration analysis
Migration analysis was performed similarly to the previous studies. 17,18 For random migration, 6.0 × 10 4 cells of TSCs were seeded on 6-well plates. Marker lines were made behind the plates, and at least five lines were on the every well. On the next day, the tips were used to scratch the lines which are perpendicular to the marker lines on the wells; then, serum-free medium was added into the wells. Time-lapse photography was performed at 0, 3, 6, 16, 24 and 36 hours. Scratch coverage area was analysed with ImageJ at every time point. Results of random TSCs migration were measured by three independent analyses.

| Biomechanical analysis
We followed the procedures as described in the previous study. 19 The Achilles tendon with bony end was first isolated. The two bony ends of tendon were fixed on a custom-made testing jig with two clamps. The lower one was used to fix the calcaneus end, while the upper one was used to fix the tibia end. The whole construct was then mounted onto the biomechanical testing machine.

| Statistical analysis
All values were expressed as mean ± standard deviation (SD). Student's t test was used to compare between two groups. Multiple comparisons were made using a one-way analysis of variance followed by Fisher's tests. A P-value of <0.05 was considered to be statistically significant.

| Aspirin improves tendon healing and changes abnormal macrophage profile in tendinopathy
Haematoxylin and eosin staining revealed that tendon in ASA treatment group showed more continuous and integrated phenotype compared with discontinuous tendon in injury group

| Aspirin inhibits IL-6 and promotes IL-10 expression of injury tendon
Next, we explored the effect of aspirin on pro-inflammatory factor IL-6 and anti-inflammatory factor IL-10. The results of immunohistochemistry showed that the expression of IL-6 and IL-10 in injury group increased significantly compared with intact tendon control group (Figure 2A-D). After ASA treatment, the expression level of IL-6 decreased and that of IL-10 increased ( Figure 2E-F). Similarly, gene expression in vitro showed that ASA treatment significantly decreased the expression of IL-6 induced by IL-1β, and IL-10 expression was elevated by ASA treatment (Figure 2I-J).

| In vivo aspirin improves tendon healing through changing expression of MMP-3, TIMP-3 and Col-1a1
Then, we discussed the effect of aspirin on tendon matrix-related factors, matrix metalloproteinase-3(MMP-3), tissue inhibitors

| In vitro aspirin reduces scar formation-related gene expression and reduces formation of scar tissue in vivo
After that, we discussed the changes in scar formation after ASA

| In vitro aspirin reduces cell migration and proliferation of TSCs stimulated by IL-1β
Migration and proliferation were tested through scratch assay and BrdU staining. Firstly, we found that TSCs motility increased significantly after IL-1β stimulation. While, to our surprise, ASA+ IL-1β delayed the migration of TSCs compared with that in IL-1β group ( Figure 5A-U). Secondly, IL-1β stimulated TSCs proliferation and ASA + IL-1β reversed this promotion effect on TSCs ( Figure 5V-Z).

| JNK/STAT-3 signalling is engaged in aspirininduced anti-inflammation and tendon healing
Western blotting showed that ASA + IL-1β group increased P-STAT-3 and P-JNK compared with IL-1βonly group, and expression of STAT-3 and JNK had no differences between IL-1β group and IL-1β + ASA group ( Figure 6A). After adding the STAT-3 inhibitor S3I-201 and JNK inhibitor SP600125, the increase trend of P-STAT-3 and P-JNK was reversed by two inhibitors. At the same time, IL-10 and TIMP-3 expressions in TSCs induced by IL-1β + ASA were significantly F I G U R E 1 Aspirin changes abnormal macrophage profile in rat tendinopathy model and improve tendon healing. More continuous and integrated tendons occurred in ASA treated group (I) compared with injury group (E). iNOS + M1 Mφs and CD14+ monocytes were more abundant in injury (F,G) compared with ASA treated tendon (J,K). CD206+ M2 Mφs showed reverse tendency compared with iNOS + M1 Mφs and CD14 + monocytes(H,L). The tendons in ASA treatment group were thinner compared with injury model group(P-R). Histology score showed that score of ASA treatment group was higher than those of injury group(S). The data are presented as the means ± SD. *: vs control, #: vs injury group, N = 6 diminished by STAT3 and JNK signalling inhibitors, and IL-6 was significantly promoted by the two inhibitors, while decreasing expression of scar formation marker COMP was significantly reversed by S3I201 only ( Figure 6B).

| Aspirin improves tendon healing and biomechanical character in injury tendon
Four weeks after ASA treatment on tendinopathy, the tendon samples were collected for tendon healing analysis and biomechanical testing. Col-1a1/Col-III which was representative of tendon healthy conditions increased significantly in ASA treatment group comparing with injury group (Figure 7A-M). The biomechanical testing results showed that the ultimate stress and Young's modulus were significantly higher in ASA treatment group compared with those in injury group ( Figure 7N-P).

| D ISCUSS I ON
cytokines aspects, aspirin may execute its anti-inflammation through inducing IL-10 and decreasing IL-6.
MMP-3 is a member of MMP family, which is involved in the breakdown of extracellular matrix proteins and tissue remodelling in normal physiological processes or disease processes, such as arthritis, and tumour metastasis. 28,29 TIMP-3 is member of tissue inhibitor of metalloproteinases family, which is a group of peptidases involved in degradation of the ECM. TIMP-3 and MMP-3 are a pair of mutual antagonism to balance the ECM's metabolism. 30 The present study showed that ASA treatment balanced tendon ECM synthesis and degradation through modulating the metabolic balance between MMP-3 and TIMP-3.
The ECM of tendon tissue consists of collagen I, collagen III, elastin and various proteoglycans and mucopolysaccharides. 16 Erroneous ECM deposition in scar tissue after tendon injury may lead to a poor and delayed healing process resulting in compromised tissue quality. 9 Promoted by the observation, we observed that ECM-related proteins were changed in ASA treated group. Biglycan is a kind of proteoglycan which is found in a variety of extracellular matrix tissues.
F I G U R E 5 Aspirin significantly reduced TSCs migration and proliferation. (A-U) In vitro wound healing assays showed that scratch closure of ASA + IL-1β group was significantly slower compared with IL-1β treated TSCs. (V-Z) BrdU assay showed that ASA + IL-1β group diminished TSCs proliferation induced by IL-1β. The data are presented as the means ± SD. *: vs. control, #: vs. injury group, N = 3 COMP, an ECM protein primarily present in cartilage, is present in high quantities in fibrotic scars and systemic sclerosis. 31  production. 37,38 Similarly, STAT3 signalling has been identified in cancer inflammation, neuronal differentiation 39 and in tendon healing or pathology. 40 Our signalling data presented a novel role of STAT3 signalling axis in ASA-induced IL-10 and TIMP-3 expression of TSCs, and that were related to ASA-inhibited tendinopathy inflammation and scar formation.
Inflammation is a pivotal process in both normal and pathologic tendon healing. There is substantial evidence in different model organisms that the immune system is of primary importance in determining the quality of the repair response, including the extent of scarring, and the restoration of organ structure and function. 41  In summary, the present study demonstrated that aspirin inhibited the inflammation and scar formation in tendinopathy tendon, and the process was related to JNK/STAT-3 signalling pathway. On the other hand, aspirin decreased migration and proliferation ability, which explained that aspirin decreased the excess ECM's deposition. In our previous study, we demonstrated that high concentration of aspirin can induce TSCs apoptosis via inhibiting Wnt/β-Catenin pathway. 42 The present study demonstrated that the proper concentration of aspirin regulated inflammation and scar formation in the injury tendon healing process, providing new therapeutic evidence of aspirin on tendon injury.

ACK N OWLED G EM ENTS
The

CO M PE TI N G I NTER E S TS
No competing financial interests exist among any authors in relation to this submission.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
All experimental procedures were performed in accordance with the National Institutes of Health (NIH) Guidelines for Care and Use of Laboratory Animals and approved by the animal research ethics committee of Third Military Medical University.

CO N S E NT FO R PU B LI C ATI O N
Not applicable.

DATA AVA I L A B I L I T Y
The authors declare that the data supporting the findings of this study are available within the article and its supplementary information files.