Hedgehog Signalling Contributes to Trauma-Induced Tendon Heterotopic Ossification and Regulates Osteogenesis through Antioxidant Pathway in Tendon-Derived Stem Cells

Heterotopic ossification (HO) is defined as the generation of pathological ectopic bony structures in soft tissues, but the molecular mechanisms of tendon HO are not fully revealed. Hedgehog (Hh) signalling is reportedly critical in hereditary HO. Our study focuses on the role of Hh signalling in the formation of trauma-induced tendon ossification. In this study, samples of healthy tendons and injured tendons from C57BL/6J female mice at 1, 4, 7, and 10 weeks after Achilles tenotomy were collected for quantitative real-time polymerase chain reaction (qRT–PCR) and immunohistochemical analysis (IHC). At 1, 4, 7, and 10 weeks postinjury, tendon samples from the mice administered with vehicle, GANT58 (a GLI antagonist), or SAG (a smoothened agonist) were harvested for micro-CT, histological staining, qRT–PCR, and IHC. Rat tendon-derived stem cells (TDSCs) treated with vehicle, GANT58, or SAG were used to induce osteogenic and chondrogenic differentiation in vitro for qRT–PCR, alkaline phosphatase staining, Alcian blue staining, and reactive oxygen species (ROS) levels measurement. We found that Hh signalling is remarkably activated during the formation of trauma-induced tendon ossification in the model of Achilles tenotomy. The in vitro and in vivo assays both confirm that downregulation of Hh signalling significantly suppresses osteogenesis and chondrogenesis to inhibit tendon ossification, while upregulation of Hh signalling promotes this process. Under osteogenic induction, Hh signalling regulates antioxidant pathway and affects ROS generation of TDSCs. Collectively, Hh signalling contributes to trauma-induced tendon ossification and affects ROS generation through antioxidant pathway in osteogenic differentiation of TDSCs, indicating that targeting Hh signalling by GANT58 may be a potential treatment for trauma-induced tendon ossification.


Introduction
Tendon ossification is a common manifestation of chronic functional and structural disorders caused by tendon injury. In trauma-related tendon ossification, bony ectopicity frequently causes tendon rupture, restricted mobility, chronic pains, impaired tendon chondrogenesis of TDSCs and the antioxidant effect of Hh signalling on the osteogenesis of TDSCs.
In summary, we aimed to clarify the impact of Hh signalling on the pathogenesis of trauma-induced tendon HO and whether it can be a new therapeutic target for traumainduced tendon HO.

Animal Model and Grouping
Female C57BL/6J mice, aged 6-8 weeks, were used to build the Achilles tenotomy model. Firstly, a full transverse cutting was created at the midpoint of the right Achilles tendon in each mouse. Then, we sutured the skin and ensured that the ruptured tendon was inside the skin for self-healing. For sham surgery, a skin incision was made without Achilles tendon injury.

GANT58 and SAG Administration
For in vivo experiments, GANT58 (Abmole; M1812) dissolved in dimethyl sulfoxide (DMSO) was injected intraperitoneally (i.p.) into the mice every day (50 mg/kg, 100 µL/mouse) from day 1 to week 6 after Achilles tenotomy. SAG (Beyotime; SF6836) dissolved in DMSO was injected i.p. into the mice every day (10 mg/kg, 100 µL/mouse) from day 1 to week 6 postinjury. An equal volume of DMSO as the GANT58 or SAG solvent was administered to the vehicle groups. In vitro, cells were treated with gradient concentrations of GANT58 (0.25, 2.5, 25 mM) and SAG (0.02, 0.2, 2 mM) dissolved in DMSO separately. An equal amount of the vehicle was administered as a control ( Figure 1B).

Microcomputed Tomography (Micro-CT)
The distal hind limbs harvested from the mice processed with GANT58 (n = 10), SAG (n = 10), or vehicle (n = 10) at 10 weeks after Achilles tenotomy were fixed in 4% paraformaldehyde (PFA) at 4 • C for at least 24 h and subjected to a micro-CT scanner (Skyscan 1176; Bruker, Billerica, MA, USA) at 50 kVp and 200 µA. Data were analysed at a threshold of 255 for the detection of ectopic mineralized bones.

Gene Expression Analysis
Total RNA of the samples from animals and cells was extracted by TRIzol and reversetranscribed into complementary DNA (cDNA) by PrimeScript TM RT Master Mix (both Takara, San Jose, CA, USA). Quantitative real-time polymerase chain reaction (qRT-PCR) was conducted using SYBR Premix Ex Taq TM II (Takara) on an ABI QuantStudio5 (Applied Biosystems, Foster City, CA, USA). Data were standardized to glyceraldehyde-3-phosphate dehydrogenase. Relative expression of each gene was computed using 2 −∆∆Ct . The primer sequences of relevant genes are shown in Supplementary Table S1.

Cells Culture
TDSCs were isolated from the Achilles tendons of 6-to 8-week-old SD female rats as described previously [25][26][27]. After the tendon sheath was stripped off, the tendon samples were cut into small pieces and digested with 3 mg/mL collagenase I (Sigma, C0130, Burlington, MA, USA) for 3 h at 37 • C. After the digestion solution containing TDSCs was passed through a 70 µm cell strainer, the released TDSCs were resuspended into single-cell suspensions and cultivated in plates with low-glucose Dulbecco's modified Eagle medium (DMEM) added with 15% foetal bovine serum (FBS) (both Gibco, Vacaville, CA, USA) and 1% penicillin/streptomycin. Then cells were collected with 0.25% Trypsin-EDTA (Gibco) for passage. The third to fifth passage cells were chosen for in vitro experiments. The medium was changed every 3 days. For osteogenic induction, 1.0 × 10 5 cells/well in 6-well plates and 5.0 × 10 4 cells/well in 12-well plates were grown for gene detection and alkaline phosphatase (ALP) staining, respectively. After reaching 80% confluence, the medium was changed to an osteogenic inducing medium composed of high-glucose DMEM (Gibco) containing 10 −8 M dexamethasone, 50 µg/mL vitamin C, 10 mM β-glycerol phosphate (all Sigma-Aldrich, St. Louis, MO, USA), 10% FBS and 1% penicillin/streptomycin. The medium was renewed every 2 days. On day 7 of culture, the ALP activity of TDSCs seeded into 12-well plates was tested using ALP staining. On day 7 and 14, cells seeded into 6-well plates were used for RNA extraction.
To evaluate chondrogenesis in a micromass culture system [28], 2.0 × 10 5 cells in a volume of 10 µL were carefully seeded into the centre of each well of a 24-well plate. After the cells adhered at 37 • C for 3-5 h, each well was added with a chondrogenic induction medium composed of high-glucose DMEM containing 1 mM sodium pyruvate (Gibco), 1% insulin-transferrin-selenium and 40 µg/mL proline (both Sigma-Aldrich, Burlington, MA, USA), 50 µg/mL vitamin C, 10 −7 M dexamethasone, 10 ng/mL TGF-β3 (Peprotech, Westlake Village, CA, USA), and 1% penicillin/streptomycin. The medium was renewed every other day. On day 7, micromass cultures were used for RNA extraction and stained with Alcian blue.

ALP Staining
TDSCs seeded into 12-well plates were induced with an osteogenic medium using different concentrations of GANT58 and SAG for 7 days, fixed in 4% PFA for 30 min, and then stained with a BCIP/NBT ALP colour development kit (Beyotime, C3206) as instructed. The positive rates of ALP staining were quantified on ImageJ.

Alcian Blue Staining
TDSCs seeded in the micromass culture system were induced using a chondrogenic medium with different concentrations of GANT58 and SAG for 7 days, fixed in 4% PFA for 30 min, and then stained using an Alcian blue stain kit (Solarbio, G1563) at pH 1.0. The positive rates of Alcian blue staining were quantified using ImageJ.

Measurement of Intracellular ROS Levels
The levels of ROS were detected by a reactive oxygen species assay kit (BestBio, BB-4705) according to the manufacturer's protocol. The results were evaluated by flow cytometer and fluorescent microscopy.

Statistical Methods
Data obtained from at least three independent experiments are shown as mean ± standard deviation (SD) of the mean. Inter-group differences were computed using two-tailed Student's t test and one-way analysis of variance. The significance levels are * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001.

Hh Signalling Is Activated during the Progression of Heterotopic Ossification in the Tendon after Achilles Tenotomy
To determine whether Hh signalling is essential in the formation of trauma-induced tendon ossification, we collected the injured tendon specimens from mice at 1, 4, 7, and 10 weeks after Achilles tenotomy for further analyses. At the gene level, qRT-PCR implied the specific gene expressions of Hh signalling-related molecules (e.g., SHH, IHH, SMO, PTCH1, GLI1, HHIP) significantly increased at all time points (Figure 2A-D). At the protein level, IHC staining showed that SHH, IHH, SMO, and GLI1 were significantly upregulated during HO development in the injured tendons at 4 and 10 weeks, compared to the sham groups ( Figure 2E,F). These results support that Hh signalling is distinctly upregulated during ossification formation in the Achilles tenotomy model, suggesting that Hh signalling is very likely involved in trauma-induced tendon HO.

Inhibition of Hh Signalling Restricts Posttraumatic Tendon Endochondral Ossification
GANT58, a GLI1 antagonist, was used to treat the mice from day 1 to week 6 after Achilles tenotomy, and tendons were sampled at 1, 4, 7, and 10 weeks. At 10 weeks, micro-CT uncovered obvious mineralized bone formation at the proximal and distal ends of the Achilles tendons in the vehicle-treated group, while GANT58 strongly protected the injured tendons from forming ectopic mineralized bones ( Figure 3A,B).
Histologically, the H&E staining demonstrated that GANT58 significantly reduced the areas of ectopic bones in injured tendon sections at 10 weeks (0.24 ± 0.09-fold above vehicle group; p < 0.01) ( Figure 3C). Because endochondral ossification is a significant part of bone formation [6], we stained the tendon sections from the vehicle and GANT58 groups with Alcian blue at 4 weeks and detected positive cartilage nodules in both groups,

Inhibition of Hh Signalling Restricts Posttraumatic Tendon Endochondral Ossification
GANT58, a GLI1 antagonist, was used to treat the mice from day 1 to week 6 after Achilles tenotomy, and tendons were sampled at 1, 4, 7, and 10 weeks. At 10 weeks, micro-CT uncovered obvious mineralized bone formation at the proximal and distal ends of the Achilles tendons in the vehicle-treated group, while GANT58 strongly protected the injured tendons from forming ectopic mineralized bones ( Figure 3A,B). mation. Gene transcript values were normalized to healthy tendons. Results revealed that GANT58 significantly downregulated the expressions of bone-related genes such as OCN and RUNX2 at 7 and 10 weeks when mature ectopic mineralized bone was gradually formed ( Figure 3E,F). Additionally, GANT58 restricted the expression of cartilagic genes (e.g., AGG, COL2A1, SOX9) at 4 weeks when cartilage tissues were formed and matured. Moreover, expression levels of chondrogenesis marker genes decreased during the osteogenic phase at 7 and 10 weeks ( Figure 3G-I).  Histologically, the H&E staining demonstrated that GANT58 significantly reduced the areas of ectopic bones in injured tendon sections at 10 weeks (0.24 ± 0.09-fold above vehicle group; p < 0.01) ( Figure 3C). Because endochondral ossification is a significant part of bone formation [6], we stained the tendon sections from the vehicle and GANT58 groups with Alcian blue at 4 weeks and detected positive cartilage nodules in both groups, indicating that tendon ossification occurred through endochondral ossification. More importantly, formation of cartilage nodules was heavily suppressed in the GANT58 group compared to the vehicle group (30.85 ± 3.06% vs. 8.15 ± 5.44%; p < 0.001) ( Figure 3D).
Cartilage is generated and matures to bone during the development of endochondral ossification. In our specimens, the expressions of osteogenic-and chondrogenic-specific genes were confirmed by qRT-PCR at all time points during tendon ossification formation. Gene transcript values were normalized to healthy tendons. Results revealed that GANT58 significantly downregulated the expressions of bone-related genes such as OCN and RUNX2 at 7 and 10 weeks when mature ectopic mineralized bone was gradually formed ( Figure 3E,F). Additionally, GANT58 restricted the expression of cartilagic genes (e.g., AGG, COL2A1, SOX9) at 4 weeks when cartilage tissues were formed and matured. Moreover, expression levels of chondrogenesis marker genes decreased during the osteogenic phase at 7 and 10 weeks ( Figure 3G-I).

Activation of Hh Signalling Promotes Posttraumatic Tendon Endochondral Ossification
Next, we used SAG, a SMO agonist, to treat the mice after Achilles tenotomy. At 10 weeks, while micro-CT showed obvious ectopic bone generation in the Achilles tendons of the vehicle group, SAG more significantly accelerated the formation of ectopic bone tissues in the injured tendons ( Figure 4A,B).
The qRT-PCR showed that SAG significantly upregulated the expressions of bonerelated genes (e.g., OCN and RUNX2) at 4 and 7 weeks ( Figure 4E,F) as well as cartilagerelated genes such as AGG and SOX9 at 7 weeks and COL2A1 at 4 weeks ( Figure 4G-I), indicating that SAG accelerates chondrogenesis and osteogenesis of injured tendons during endochondral ossification.

Regulation of Hh Signalling Affects Osteogenesis and Chondrogenesis of TDSCs In Vitro
The TDSCs isolated from SD rats were incubated and used to detect the impacts of GANT58 and SAG on osteogenic and chondrogenic differentiation potentials.
As for the effect on osteogenic differentiation, TDSCs were processed with GANT58 and SAG under osteogenic induction. After 7 and 14 days, the gene expressions of osteogenic markers (e.g., OCN, RUNX2, ALP) were suppressed after treatment with GANT58, but promoted after treatment with SAG ( Figure 5A-D). After 7 days, GANT58 reduced ALP staining, indicating osteogenic differentiation was suppressed with downregulation of Hh signalling, while SAG increased ALP staining, suggesting osteogenic differentiation was promoted with upregulation of Hh signalling ( Figure 5E,F).

Activation of Hh Signalling Promotes Posttraumatic Tendon Endochondral Ossification
Next, we used SAG, a SMO agonist, to treat the mice after Achilles tenotomy. At 10 weeks, while micro-CT showed obvious ectopic bone generation in the Achilles tendons of the vehicle group, SAG more significantly accelerated the formation of ectopic bone tissues in the injured tendons ( Figure 4A,B).  Scale bar = 20 µm. * p < 0.05, ** p < 0.01. A micromass culture system was used for chondrogenic induction. After 7 days of culture, qRT-PCR demonstrated gene expressions of chondrogenic markers, including AGG, SOX9, and COL2A1, were suppressed in TDSCs in response to GANT58 treatment, but were promoted in response to SAG treatment ( Figure 5G,H). At the same time, the GANT58 group exhibited less intense Alcian blue staining, suggesting chondrogenic differentiation was suppressed with downregulation of Hh signalling, while the SAG group displayed more intense staining, indicating chondrogenic differentiation was promoted with upregulation of Hh signalling ( Figure 5I,J).
In summary, osteogenesis and chondrogenesis during endochondral ossification of tendons are greatly regulated by Hh signalling at the cellular level. Thus, TDSCs are likely to be one of the targets for GANT58 and SAG and are in charge of tendon ossification involving Hh signalling in this model.

Hh Signalling Regulates Antioxidant Pathway in Osteogenic Differentiation of TDSCs
Since ROS has been reported to be involved in the formation of tendon HO [22], we next examined whether Hh signalling regulates the antioxidant pathway and ROS generation during osteogenic differentiation of TDSCs. After weeklong osteogenic induction, most of the gene expressions of antioxidant related markers (e.g., glutathione S-transferase pi 1 (Gstp1), catalase, superoxide dismutase 1 (SOD1), superoxide dismutase 2 (SOD2), glutathione peroxidase 1 (Gpx1), glutathione peroxidase 2 (Gpx2)) of TDSCs were promoted after treatment with GANT58, but suppressed after treatment with SAG ( Figure 6A,B). Flow cytometry and fluorescent microscopy images showed that GANT58 reduced the levels of ROS of TDSCs under osteogenic induction for 48 h (0.84 ± 0.04-fold above vehicle group in flow cytometry; p < 0.01) (19.77 ± 4.03 AU vs. 6.43 ± 0.91 AU in mean fluorescence intensity; p < 0.0001), while SAG promoted ROS generation under the same conditions (1.47 ± 0.21-fold above vehicle group in flow cytometry; p < 0.05) (19.43 ± 2.79 AU vs. 34.53 ± 6.88 AU in mean fluorescence intensity; p < 0.01) ( Figure 6C-F). These data suggest Hh signalling can regulate antioxidant pathway and affect ROS generation in osteogenic differentiation of TDSCs, and the antioxidant effect of Hh signalling may be partially involved in the regulation of tendon HO.

Hh Signalling Regulates Antioxidant Pathway in Osteogenic Differentiation of TDSCs
Since ROS has been reported to be involved in the formation of tendon HO [22], we next examined whether Hh signalling regulates the antioxidant pathway and ROS generation during osteogenic differentiation of TDSCs. After weeklong osteogenic induction, most of the gene expressions of antioxidant related markers (e.g., glutathione S-transferase pi 1 (Gstp1), catalase, superoxide dismutase 1 (SOD1), superoxide dismutase 2 (SOD2), glutathione peroxidase 1 (Gpx1), glutathione peroxidase 2 (Gpx2)) of TDSCs were promoted after treatment with GANT58, but suppressed after treatment with SAG ( Figure  6A,B). Flow cytometry and fluorescent microscopy images showed that GANT58 reduced the levels of ROS of TDSCs under osteogenic induction for 48 h (0.84 ± 0.04-fold above vehicle group in flow cytometry; p < 0.01) (19.77 ± 4.03 AU vs. 6.43 ± 0.91 AU in mean fluorescence intensity; p < 0.0001), while SAG promoted ROS generation under the same conditions (1.47 ± 0.21-fold above vehicle group in flow cytometry; p < 0.05) (19.43 ± 2.79 AU vs. 34.53 ± 6.88 AU in mean fluorescence intensity; p < 0.01) ( Figure 6C-F). These data suggest Hh signalling can regulate antioxidant pathway and affect ROS generation in osteogenic differentiation of TDSCs, and the antioxidant effect of Hh signalling may be partially involved in the regulation of tendon HO.

Discussion
Tendon ossification, one of the most common and serious complications after tendon injury, is detrimental to the normal repair of tendons and seriously affects tendon function. Therefore, it is of both basic scientific value and clinical significance to clarify the intervention targets and regulatory mechanisms of trauma-induced tendon ossification and explore effective drugs or new methods for targeting post-trauma tendon ossification.
Hh signalling is a highly conserved signal transducing pathway that is critically involved in embryonic growth and maturation, maintenance of the stable internal environment of organs, postinjury tissue repair and regeneration, and tumour occurrence, development, differentiation, and invasion [29]. Dysregulation of Hh signalling can disrupt musculoskeletal tissue homeostasis [30]. At present, research about the function of Hh signalling in tendon development, homeostasis maintenance, or postinjury tendon repair and HO is rare. Reportedly, Hh signalling is upregulated in tendon sheath tissues after tendon injury and promotes Bglap + cell migration into the tendons, but the origin of Bglap + cells or their specific role in repair was not clarified [31]. Additionally, spontaneous tendon ossification depends on the upregulation of Hh signalling in the Achilles tendons of MKX −/− mice, but the role of Hh signalling in tendon ossification of healthy mice is unknown [32]. Moreover, intensified Hh signalling pathologically drives ectopic cartilage and bone generation in soft tissues [33] via heterotopic chondrogenesis and HO [34]. In animal models, gene-modulated ectopic Hh signalling is enough to initiate HO, while pharmacologic or genetic restriction of Hh signalling heavily weakens the severity of this condition [3]. Likewise, synovial chondromatosis, an induction of synovial tissue ossification, is linked with intensified canonical Hh signalling [35]. In contrast to cartilage and bone gain, higher Hh signalling is also related to cartilage deterioration and loss [36,37]. Thus, suitable Hh signalling normally participates in the ectopic inhibition and the maintenance and genesis of normtopic bone/cartilage. Notably, IHH, which guides Hh signalling within the developing limbs at birth, is expressed in an area of postmitotic, prehypertrophic chondrocytes that are just near the zone of proliferating chondrocytes [38][39][40] and is pivotal in endochondral ossification, and initiates osteoblast differentiation in the perichondrium [41]. However, the effect of IHH on endochondral ossification in tendons has not been investigated. Recently, a self-amplifying and self-propagating loop of YAP and SHH is reportedly a common mechanism of ectopic bone formation and expansion in mouse models of genetic HO and injury-induced HO. Upregulation of YAP and SHH are also found in samples from patients with HO [42]. These findings indicate that SHH may drive the process of formation of HO.
As reported extensively, Hh signalling directly acts on mesenchymal stem cells (MSCs) and regulates the direction of lineage differentiation [43][44][45]. A relatively clear conclusion holds that Hh signalling promotes the osteogenic and chondrogenic differentiation of MSCs [44] and inhibits adipogenic differentiation [44]. Current evidence supports that TDSC motion is necessary for tendon injury curing. During tendon restoration and reproduction, TDSCs can move to the injury site, differentiate into tenocytes, and substitute the abnormal tenocytes participating in the pathophysiological process [46], but due to postinjury changes in the local microenvironment, local stem cells can mis-differentiate into bones and cartilage [47], which is one of the reasons for the occurrence and development of tendon ossification. The pathological procedure of chronic tendinopathy generates some dysfunctional matrix parts (e.g., hypervascularity, acquisition of chondrocyte phenotypes, calcium formation). Erroneous differentiation of TDSCs is indicated to cause TDSC pool consumption and ectopic chondro-ossification [10]. However, for tendon tissues, it is unclear whether Hh signalling directly acts on TDSCs and regulates the postinjury differentiation direction of lineages. Our results provide evidence that Hh signalling may directly modulate the osteogenic and chondrogenic differentiation of TDSCs during the pathogenesis of tendon endochondral ossification. The maintenance of normal physiological function requires a balance between oxidation and antioxidation. Antioxidant systems such as Gstp1, catalase, SOD, and Gpx sever the purpose of elimination of exces-sive ROS [22]. When overwhelming ROS break the balance between oxidative stress and antioxidant systems, oxidative stress contributes to pathophysiological processes [48]. For MSCs, upregulation of ROS impairs osteogenic differentiation and stimulates chondrogenesis [49]. For tendon stem/progenitor cells, high level of ROS promote osteogenesis and chondrogenesis differentiation and participate in the pathogenesis of tendon HO [22]. These results imply that ROS may have different effects on osteogenesis and chondrogenesis in different cells and disease models. Hh signalling has been reported to regulate osteogenesis in MSCs through modulation of ROS [50]. Our study finds that Hh signalling regulates antioxidant pathway and affects ROS generation in osteogenic differentiation of TDSCs, suggesting the antioxidant effect of Hh signalling may be partially involved in the regulation of tendon HO.
Tissue injury can cause local inflammation and macrophage gathering, which further result in an assembly of osteogenic factors, including BMPs. This probably disrupts local stem/progenitor cells and drives them to follow osteogenic routes with HO development, implying that such disruption can be a major cellular mechanism of HO [7]. Kan et al. demonstrate that niche-dwelling progenitor/stem cells and niche supportive cells (including mast cells, neurites, vasculature, and macrophages) constitute injury-induced local microenvironment (MSC niche) of HO. Through feedback and non-cell autonomous mechanisms, BMP and Hh signalling co-regulate the formation of the MSC niche, which may initiate the pathological osteogenic cascade [51]. BMS-833923, a SMO antagonist, has been reportedly shown to remarkably inhibit osteoblast differentiation of human skeletal (mesenchymal) stem cells (hMSCs). Global gene expression profiling of BMS-833923-treated identifies many genes with significant changes, which markedly impact multiple signalling including inflammatory response signalling [43]. These studies suggest that our treatments may also have effects on immune cells such as monocytes and macrophages in the inflammatory stage. Other cells such as immune or vascular cells may also contribute to Hh family ligands. Further research is needed to determine whether formation of tendon HO in the Achilles tenotomy model is driven by a cell-autonomous Hh signalling activation of TDSCs or by factors from other cells.
GANT58, a small molecule inhibitor of GLI1, prevents GLI1-dependent transcription through the suppression of its post-translational modification [52]. As an Hh inhibitor, it shows therapeutic potential for age-related bone loss and tumour-induced bone disease [52,53]. SAG, a specific Hh agonist, has been shown to promote osteogenic differentiation and bone regeneration [54][55][56]. In this study, we consider the Hh as a whole signalling pathway and show that Hh is responsible for ectopic mineralized bone formation in injured tendons. The use of GANT58 to downregulate Hh signalling suppresses, while SAG as an upregulator promotes, osteogenesis and chondrogenesis during the formation of trauma-induced tendon ossification both in vivo and in vitro. However, there are some limitations to our findings: (1) This study does not further explore the regulatory mechanism of Hh signalling-related molecules during the trauma-induced tendon ossification, which is another interesting issue that deserves further investigation. (2) We do not go further to test whether there is any side effect on the musculoskeletal system after drug treatment. The development of more precisely targeted drugs requires further exploration. (3) All experiments are limited to animals and cells, which requires that the effect of GANT58 on human tendons shall be investigated in clinical trials. All the same, we confirm that Hh signalling can regulate the formation of trauma-induced tendon ossification. This study offers direct proof that Hh signalling may be a potential clinical therapeutic target for trauma-induced tendon ossification.

Conclusions
In conclusion, Hh signalling is activated in post-trauma tendon ossification and the regulation of Hh signalling affects the formation of tendon ossification by modulating the differentiation direction of TDSCs. Hh signalling can regulate antioxidant pathway and affect ROS generation in osteogenic differentiation of TDSCs. Thus, we demonstrate that a classic signalling pathway directly and remarkably regulates ossification formation in injured tendons, which may be an extremely meaningful functional study. Both in vivo and in vitro results suggest that GANT58 inhibits trauma-induced tendon ossification by restricting chondrogenesis and osteogenesis of tendons. Therefore, targeting Hh signalling by GANT58 may be a potential therapeutic strategy for useful prevention of traumainduced tendon ossification.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/antiox11112265/s1, Supplementary Table S1 showing the primer sequences for quantitative real-time polymerase chain reaction analysis of gene expression conducted in this study.