Development process of traumatic heterotopic ossification of the temporomandibular joint in mice

a b s t r a c t
Purpose: The exact development process underlying traumatic heterotopic ossification of the tempo- romandibular joint (THO-TMJ) is largely unclear. In this study, we try to explore the histological devel- opment process of THO-TMJ. Materials and methods: Condylar cartilage of one-month-old male mice was partially removed from the left joint with small scissors to induce THO-TMJ. The phenotypes were observed using gross observation, microcomputed tomography (micro-CT) scans and histological examination from one month to six months after surgery. Results: The micro-CT examination results showed that the injured condyle integrated with ectopic bone tissue to form an osteophyte and that the volume and density of the osteophyte grew exponentially with time. Hematoxylin and eosin (H&E), safranin O and fast green staining of the THO-TMJ specimens revealed that the ectopic bone tissue was mainly nonmineralized fibrous tissue 1 month after surgery. This tissue gradually transformed into cartilage 3 months after surgery. Finally, the tissues transformed into mature bone tissue 6 months after surgery. Immunofluorescence staining showed VEGF-a expres- sion in the heterotopic tissue 1 month after surgery, and the expression of Sox9 in the heterotopic tissue was obvious 3 months after surgery. Furthermore, OCN expression was evident in most of the heterotopic tissue 6 months after surgery. The results also showed clear hypoxia-inducible factor 1-alpha (Hif-1a) expression in the injured chondrocytes of the condyle, especially in the articular proliferative zone and fibrocartilaginous zone. Conclusions: The THO-TMJ imaging characteristics indicated an exponential change with time. Histo- logically, the development process of THO-TMJ is an endochondral ossification process and includes three stages, fibroproliferative, chondrogenic and osteogenic stage. In addition, Hif-1a, which was expressed in some of the injured chondrocytes, may play an essential role in the initial THO-TMJ.

1.Introduction
Heterotopic ossification (HO) means the formation of ectopic bone in muscles, tendons, ligaments, and other soft tissues (Amar et al., 2015). HO is divided into acquired nongenetic HO and inherited genetic HO (Xu et al., 2018). For the rare genetic HO,fibrodysplasia ossificans progressiva (FOP) involves endochondral ossification, while progressive osseous heteroplasia (POH) and Albright hereditary osteodystrophy (AHO) leads to HO through intramembranous ossification (Shore and Kaplan, 2010). Neuro- genic trauma-induced HO is a common type of acquired nongenetic HO and occurs through both intramembranous and endochondral ossification (Huang et al., 2018). However, in trauma-induced myositis ossificans circumscripta, cartilage formation occasionally occurs through an intramembranous or endochondral process (Walczak et al., 2015). Trauma-induced HO is solely triggered by injury, while genetic HO has both injury-mediated and noninjury- mediated components.Trauma is the main cause of HO around the temporomandibular joint (TMJ) (Cetinkaya, 2012). When substantial ectopic bone for- mation occurs around the condyle, patients may suffer from the symptoms of TMJ ankylosis and may eventually experience limited mouth opening and facial deformity (Song and Yap, 2017). TMJ ankylosis is often divided into fibrous and bony; in addition, fibrous ankylosis can progress into bony ankylosis. Although a close rela- tionship exists between traumatic condyle and TMJ ankylosis, the pathogenesis of the disease remains indefinite.Currently, the main treatment for THO-TMJ is surgical resection (Zhu et al., 2015). However, there is currently no effective preventive method for THO-TMJ because the exact cellular and molecular mechanisms have not been completely elucidated (Mercuri and Saltzman, 2017). Fortunately, development of animal models which serve as powerful tools to study the unique traumatic injuries has greatly facilitated the mechanistic understanding of trauma-induced HO(Dey et al., 2017). We have successfully established a mouse model with traumatic HO of the TMJ (THO-TMJ) by partially removing condylar cartilage in our previous studies and showed that injured cartilage, but not injured bone and uninjured cartilage, plays a crucial role in the development of THO-TMJ (Dai et al., 2016). In this study, we mainly try to explore the histological development process of THO-TMJ, which will shed light on proper and effective treatments for THO-TMJ (Ohrbach and Dworkin, 2016).

2.Materials and methods
Male mice with a C57BL/6J genetic background were used in this study. All animal experiments complied with the ARRIVE guide- lines. Animal experimental procedures were performed in compliance with the Institutional Animal Care and Use Committees of the Shanghai Ninth People’s Hospital, Shanghai Jiao Tong Uni- versity School of Medicine. The procedures were reviewed and approved by the Ethics Committees of the Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, China (approval number 2016-45). In this study, condylar cartilage was partially removed from the left joint of one-month-old male mice with small scissors to induce THO-TMJ (Yan et al., 2014). The right condyle was used as the control. A total of 15 mice were killed at specific time points, 30, 90 and 180 days after surgery, and 20 mice were examined through microcomputed tomography (micro-CT) scans at 30, 60, 90, 120, 150, and 180 days after surgery.The THO-TMJ mouse model was established as previously described (Ouyang et al., 2018). Briefly, one-month-old male mice were anesthetized with 0.8% pentobarbital sodium intraperitone- ally. Surgery was performed on the left joint, whereas the right side was used as the control. The preauricular skin was sterilized with 75% alcohol, and a 1-cm linear preauricular skin incision was made. The preauricular fascia was cut open just above the superior border of the facial vessel to expose the zygomatic arch. Then, the joint capsule was cut open to expose the condyle, and some of the condylar cartilage was then removed using small scissors. Finally, the incision was closed using a 3-0 suture. The mice were given soft food for 2 weeks after surgery.

Total of twenty mice were collected and anesthetized with pentobarbital sodium intraperitoneally at 30, 60, 90, 120, 150 and 180 days after surgery (Bouxsein et al., 2010). Micro-CT scans wereperformed, and the density of ectopic bone was measured using an eXplore Locus Micro-CT scanner (GE Healthcare, Milwaukee, Wis- consin, USA). The slice thickness for micro-CT scans was 40 mm. Three-dimensional (3D) reconstruction of the skulls was conducted using GE MicroView software (GE Healthcare, Milwaukee, Wiscon- sin, USA). The ectopic bone volume was measured using Geomagic Studio software (Geomagic, North Carolina, USA) (Sousa et al., 2012).Whole heads of mice were freely dissected and collected 30, 90 and 180 days after surgery, and the samples were fixed in 4% paraformaldehyde (PFA) and demineralized in 0.5 M ethyl- enediaminetetraacetic acid (EDTA) for 2e3 weeks (Ballal et al., 2016). Subsequently, the samples were embedded in paraffin and cut to a thickness of 5 mm for regular hematoxylin and eosin (H&E) staining and to a thickness of 4 mm for safranin O (SigmaeAldrich, S2255) and fast green (SigmaeAldrich, F7252) staining. Briefly, slides were incubated in a safranin O solution for 5 min. Then, the slides were washed in differentiation solution and stained with afast green solution for 10 min at 37 ◦C (Wang et al., 2018). Subse-quently, the stained slides were washed with running tap water and cleared in xylene. Finally, the slides were mounted with resinous mounting medium for observation and image acquisition.Whole heads of mice were collected 30, 90 and 180 days after surgery and were embedded in paraffin. Five-micrometer thick sections were obtained, and the slides were then deparaffinized and hydrated (Donaldson, 2015). Next, antigen retrieval was per- formed using an antigen restoration liquid kit (Sunteam Biotech, China). The slides were subsequently blocked with 3% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) for 1 h and incubated with anti-VEGF-a (Abcam, 1/200), anti-sox-9 (Abcam, 1/ 100), anti-OCN (Santa Cruz Biotechnology, 1/300), and anti- hypoxia-inducible factor 1-alpha (Hif-1a; Abcam, 1/100) primaryantibodies overnight at 4 ◦C. Subsequently, secondary antibodies(AlexaFluor 568 donkey anti-mouse, AlexaFluor 488 donkey anti- rabbit, Jackson) were diluted in PBS (1/300) and incubated at room temperature for 1 h. Finally, the slides were mounted with Vectashield mounting medium for visualization under a fluores- cence microscope (Dey et al., 2016).The data are expressed as the mean ± standard deviation. Data analysis was performed using a Students T test using SPSS 18.0 software (International Business Machines, Armonk, NY, USA). P < 0.05 was considered to indicate a statistically significant difference.

3.Results
During the experiment, 1 mouse in the experimental group died 3 days after the operation, and the other mice survived until the end of the experiment. There were two mice with a restricted mouth opening due to soft tissue adhesions 30 days after surgery. Micro-CT examination from different views showed that the deviation of the mandible and adhesion around the region with condylar trauma gradually changed over time. In addition, the shape and volume changed more obviously in the region with condylar trauma than at the side with the healthy condyle atdifferent time points (Fig. 1AeH). The ectopic bone tissues and injured condyle fused to become an osteophyte, which led to a gradual increase in the volume of the condyle (Fig. 1IeP), and the trend for the volume change between this region and the healthy condyle corresponded to exponential growth (Fig. 2B, C). Further- more, the density of the ectopic bone tissues also showed an exponential growth trend from 1 month to 5 months but exhibited a fast increase from 5 to 6 months after surgery (Fig. 2D).Safranin O and fast green staining showed that there was no obvious ectopic cartilage and bone tissue around the region with condylar trauma 30 days after surgery. The early HO may resemble reactive fibroblastic lesions (Fig. 3A, B). However, there was obvious ectopic cartilage in the periarticular region and glenoid fossa 90 days after surgery (Fig. 3C, D). Then, the ectopic cartilage becamemature bone tissue in the periarticular region 180 days after sur- gery (Fig. 3E, F). H&E staining showed that the area of ectopic cartilage/bone tissues gradually increased from 30, 90 and to 180 days after surgery (Fig. 4AeF).We used immunofluorescence staining to measure the expres- sion of factors related to angiopoiesis, chondrogenesis and osteo- genesis, and the results showed that there was obvious expression of VEGF-a in the soft tissues around the region with condylar trauma 30 days after surgery (Fig. 4G, J). In contrast, there was stronger expression of Sox9 in the periarticular region 90 days after surgery, as well as stronger expression of OCN in the periarticular region 180 days after surgery; these features were consistent with the endochondral ossification development process of THO-TMJ that was revealed by safranin O and fast green staining (Fig. 4H, I, K, L). At 30 days after surgery, immunofluorescence staining also showed evident Hif-1a expression in the injured chondrocytes of the condyle, especially in the fibrocartilaginous zone and prolifer- ative zone (Fig. 5A, B, C).

4.Discussion
The TMJ is a special and complex structure that is found only in mammals (Iwasaki et al., 2010). Due to the specific location and function of the TMJ, the incidence of trauma in the mandibular condyle is particularly high; this trauma willsometimes lead to THO-TMJ or TMJ ankylosis in some patients, followed by serious damage to the mouth function and facial appearance of these patients (Arakeri et al., 2012). For TMJ ankylosis, the new tissue formation is not physiological but pathological, because the continued ectopic tissue formation re- places the normal structure of TMJ and it seems that no remodeling takes place. The formation of HO is complex since it may involve many kinds of cells and specific osteogenesis process (Agarwal et al., 2016b).Endochondral ossification and intramembranous ossification are two distinct processes for bone development (Hayashi et al., 2014). For intramembranous ossification, mesenchymal cells directly differentiate into osteoblasts (Vieira et al., 2015). However, endochondral ossification refers to the process of bone formation wherein a cartilage intermediate is formed and gradually replaced by osteoblasts. Firstly, mesenchymal cells differentiate into chon- drocytes. Then, the chondrocytes undergo apoptosis, the cartilage stroma suffers degradation and finally the degraded cartilage is replaced by mature bone tissue (Gawlitta et al., 2010). The physi- ologic processes of osteogenesis and angiogenesis are highly coupled and interdependent. In this study, we found that the his- tological development process of THO-TMJ was an endochondral ossification process and included three stages, fibrosis with vascularization, chondrogenesis and osteogenesis. Especially, angiogenesis is an important stage in the formation of THO-TMJ. Thus, prevention of the occurrence and development of THO-TMJ through inhibiting angiogenesis may be a possible choice, and should be investigated in future.Trauma is the primary trigger of THO-TMJ, which will lead to changes in the local microenvironment around the injured TMJ.

Previous study showed that the traumatic cartilage plays an important role in the development of HO (Hinton et al., 2017), which may be due to the injured chondrocytes secreting some essential bioactive factors that could initiate and contribute to the ectopic tissue formation (Coimbra et al., 2004). Thus, it is important to explore the exact role and molecular mechanism of traumatic condylar cartilage on the formation of THO-TMJ, which may help us to properly preserve the residual condylar cartilage in younger in- dividuals to simultaneously preserve the function and growth ability, and avoid the promotion effect of injured cartilage on the formation of THO-TMJ.Previous studies showed that some mesenchymal stem cells inthe soft tissues may contribute to traumatic HO, including local mesenchymal stem cells or some other progenitors recruited to the lesions from bone marrow (Agarwal et al., 2017). Previous study has shown that vascular endothelial cells have emerged as the chief candidate for the cellular origin of HO in FOP, which was induced by a combination of genetic mutation and acute inflammatory re- sponses. Especially, endothelial-mesenchymal transition (EndMT) of some cells may act as an important role in the process of ectopic bone formation (Medici et al., 2010). These cell lineage tracing studies provide new insight into the cellular pathophysiology of heterotopic ossification (Lounev et al., 2009). Thus, identifying the definite precursor cells which contribute to THO-TMJ through cell tracking and animal models is necessary in future.Local factors, such as oxygen tension, pH, micronutrients, andmechanical stimuli, may play an essential role in the development of HO (Ranganathan et al., 2015).

A large amount of various in- flammatory cytokines, such as IL-1b, IL-6 and TNF-a, could inducesome mesenchymal stem cells to initiate ectopic endochondral ossification (Huang et al., 2015). Nonsteroidal anti-inflammatory drugs (NSAIDs) prevent HO by inhibiting the osteogenic differen- tiation of progenitor cells, but lower oxygen tension will facilitate this process (Joice et al., 2018). HIF-1a is a key transcriptional regulator for cellular response to ischemia through stimulation of vascular endothelial cell precursors, and also plays a crucial role in the development of HO (Agarwal et al., 2016a). Therefore, local inflammatory and lower oxygen tension microenvironment may mediate the differentiation of mesenchymal stem cells into chon- drocytes and osteoblasts, and finally resulted in THO-TMJ through endochondral ossification (Medici and Olsen, 2012) (Fig. 6). These findings implied that anti-inflammation or application of small- molecule drugs for intervention of these factors may suitable methods for inhibition of HO formation around the traumatic TMJ, which also are research hotspots for the future.However, the exact progenitors, as well as the exact molecule and mechanism that triggers this endochondral ossification in THO-TMJ, have not been determined clearly. Moreover, the role of mechanical force also cannot be overlooked in the formation of THO-TMJ due to the specific location and function of TMJ, and should be explored in future (Ruggiero et al., 2015).

5.Conclusion
In this study, we first revealed that the histological development process of THO-TMJ is an endochondral ossification process, which included three stages: fibrosis and vascular ingrowth, cartilage formation and bone formation. It provided useful information for advanced understanding of the molecular and cellular pathogen- esis of THO-TMJ (Downey et al., 2015). Future therapeutic BAY 87-2243 strategies may focus on targeted inhibition of local factors and signaling pathways to inhibit the endochondral ossification process in THO- TMJ (Juarez et al., 2018).