|Home | About | Journals | Submit | Contact Us | Français|
The temporomandibular joint (TMJ) is a specialized synovial joint essential for the function of the mammalian jaw. The main components of the TMJ are the mandibular condyle, the glenoid fossa of the temporal bone, and a fibrocartilagenous disc interposed between them. The genetic program for the development of the TMJ remains poorly understood. Here we show the crucial role of sprouty (Spry) genes in TMJ development. Sprouty genes encode intracellular inhibitors of receptor tyrosine kinase (RTK) signaling pathways, including those triggered by fibroblast growth factors (Fgfs). Using in situ hybridization, we show that Spry1 and Spry2 are highly expressed in muscles attached to the TMJ, including the lateral pterygoid and temporalis muscles. The combined inactivation of Spry1 and Spry2 results in overgrowth of these muscles, which disrupts normal development of the glenoid fossa. Remarkably, condyle and disc formation are not affected in these mutants, demonstrating that the glenoid fossa is not required for development of these structures. Our findings demonstrate the importance of regulated RTK signaling during TMJ development and suggest multiple skeletal origins for the fossa. Notably, our work provides the evidence that the TMJ condyle and disc develop independently of the mandibular fossa.
The temporomandibular joint (TMJ) is a mammalian synovial joint essential for jaw function. The TMJ consists of multiple tissues, including the glenoid fossa of the temporal bone, the condylar head of the mandible, a fibrocartilaginous disc located between these two bones, and associated muscles and tendons (Avery, 2001). Although structural features of the TMJ are well-documented, little information is available regarding the genetic, cellular, and molecular mechanisms involved in TMJ morphogenesis.
TMJ development starts with the appearance of two distinct mesenchymal condensations, the temporal and condylar blastemas, at embryonic day (E) 13.5. The condylar blastema grows toward the temporal blastema, and at E15.5, the glenoid fossa, condyle, disc, and muscles are clearly visible. At E16.5, all TMJ components are well-formed, with the fossa and condyle in complementary shapes with the disc between them (Sperber, 1992). The condyle is endochondral in origin and an important growth site in the mandible. Proliferating and hypertrophic chondrocytes become arranged in columns forming a growth-plate-like zone found at the end of the expanding condylar cartilage (Sarnat, 1966; Silbermann and Frommer, 1972). The fossa forms by a combination of intramembranous and endochondral ossification (Silbermann and Frommer, 1972; Purcell et al., 2009; Wang et al., 2011), although this process is poorly understood. An important gene in TMJ development is Ihh (indian hedgehog), which is crucial for disc formation, cellular organization of the condyle, and maintenance of the jaw joint (Shibukawa et al., 2007; Purcell et al., 2009; Ochiai et al., 2010).
In our previous microarray studies, several components of the Fgf signaling pathway were discovered to be highly expressed in the TMJ at E16.5 (Purcell et al., 2009, GEO Series accession number GSE17473; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc_GSE17473). These included several Fgfs, their receptors (Fgfrs), and sprouty genes, which encode antagonists of RTK signaling including Fgf signaling. Specifically, Spry1 and Spry2 were expressed at high levels in the TMJ (Purcell et al., 2009). Components of the Fgf signaling pathway are highly conserved throughout evolution and are known to play crucial roles in development (reviewed in Dorey and Amaya, 2010; Hatch, 2010; Itoh and Ornitz, 2011). Here, we provide the first evidence that sprouty genes are essential in TMJ development and that the growth of the condyle and disc is independent of the fossa.
All animal procedures were performed according to guidelines approved by Harvard Medical Area and UCSF animal care committees. Embryos deficient in Spry1 and Spry2 were produced as reported previously (Petersen et al., 2011). Jaws from double-heterozygous embryos were indistinguishable from wild-type CD-1 embryos; such jaws were used as controls. The presence of a vaginal plug indicated embryonic day (E) 0.5. Osteoblast- and chondrocyte-specific inactivation of Spry1 (Basson et al., 2005) and Spry2 (Shim et al., 2005), alone and in combination, was achieved in the 2.3 kb Col1α1-Cre (Liu et al., 2004) and Col2α1-Cre (Ovchinnikov et al., 2000) mouse lines, respectively.
E14.5-E18.5 embryo heads were fixed in 4% paraformaldehyde and embedded in paraffin or OCT compound (Tissue-Tek, Torrance, CA, USA). Ten-micrometer sections were stained with hematoxylin and eosin according to standard procedures. Area measurements were performed with Adobe Photoshop software.
Pregnant mice at E14.5 and E17.5 were injected with 1 mg BrdU (Invitrogen, Carlsbad, CA, USA) for 2 hrs at E14.5 and E17.5. Ten-micrometer cryosections were stained with anti-BrdU antibody (Abcam, Cambridge, MA, USA) for analysis of proliferation and Caspase-3 (Cell Signaling, Danvers, MA, USA) for measurement of apoptosis. Immunohistochemistry was performed as described in Petersen et al. (2011).
In situ hybridization was performed on 10-µm cryo or paraffin sections with digoxigenin-labeled probes as described (Purcell et al., 2009). RNA probe information is available upon request.
MicroCT scans were taken with the Siemens MicroCAT System (Malvern, PA, USA) and data analyzed by Dolphin Imaging V.11.5 software (Chatsworth, CA, USA).
Components of the Fgf signaling pathway, in particular Spry1 and Spry2, were observed to be enriched by microarray in the TMJ at E16.5 (Purcell et al., 2009); therefore, we analyzed their expression by in situ hybridization. Spry1, Spry2, and Spry4 were expressed in the lateral pterygoid and temporalis muscles that surround the TMJ (Figs. 1A, ,1B,1B, ,1D),1D), whereas Spry3 was not detected (Fig. 1C). Expression of Fgfrs was examined to determine co-localization with sprouty genes. Fgfr1 was expressed in the periosteum and Fgfr2 in the perichondrium of the fossa and the condyle; Fgfr3 in the immature chondrocytes of the condyle (Figs. 1E--1G),1G), consistent with previous observations (Purcell et al., 2009); Fgfr4 was expressed in the lateral pterygoid and temporalis muscles (Fig. 1H), consistent with its role in myogenesis (Lagha et al., 2008). Numerous candidate Fgf genes were also analyzed by in situ hybridization, including Fgf3, Fgf4, Fgf6, Fgf7, Fgf8, and Fgf18; Fgf6 was the only gene in this group to show strong expression during the examined stages of TMJ development (Fig. 1I). Notably, Spry1, Spry2, Spry4, Fgfr4, and Fgf6 were co-expressed in the lateral pterygoid and temporalis muscles surrounding the TMJ, suggesting the importance of Fgf signaling in these tissues (Figs. 1A, ,1B,1B, ,1D,1D, ,1H,1H, ,1I1I).
To define the role of sprouty genes in TMJ development, we examined mouse lines carrying null alleles of Spry1 and/or Spry2. Mice null for either Spry1 or Spry2 did not show any TMJ abnormalities (data not shown). However, there was an absence of the glenoid fossa in Spry1−/−;Spry2−/−mice, herein also referred to as mutant mice (Fig. 2). The temporalis muscle, which is normally located superior and lateral to the fossa, was enlarged in mutant mice, expanding into the space that would normally be occupied by the fossa (Figs. 2G, ,2H2H).
To determine whether sprouty genes are required for glenoid fossa development or its maintenance, we examined the developing TMJ between E14.5 and E18.5 in control and mutant embryos (Figs. 2A--2H).2H). At E14.5, the TMJ had not yet formed, but the condyle and fossa were clearly visible as mesenchymal condensations in controls (Fig. 2A). However, in mutant embryos, the fossa condensation was not detected, and the temporalis muscle appeared enlarged (Fig. 2B). At E15.5 in control embryos, the fossa began to ossify and assume its complementary shape with respect to the adjacent head of the condyle. The temporalis muscle was situated lateral to the fossa, and the disc had become more condensed (Fig. 2C). In mutants, the temporalis muscle was dramatically enlarged, such that it filled the space normally occupied by the fossa (Fig. 2D). Remarkably, the condyle and disc appeared normal (Figs. 2C, ,2D).2D). At E16.5 and E18.5 in mutants, the condyle and disc continued to develop normally, but the fossa was absent, with its usual location occupied by the enlarged temporalis muscle (Figs. 2E--2H).2H). Notably, a small lateral distal tip of the fossa, a part of the zygomatic arch, was present in mutant mice (Figs. 2B, ,2D,2D, ,2F,2F, ,2H).2H). As expected, the lateral pterygoid muscle was also enlarged in mutant embryos compared with control littermates (Figs. 2A--2H2H).
To further confirm the absence of the glenoid fossa in Spry1−/−;Spry2−/−mice, we examined the TMJ at E18.5 using microCT analysis. In controls, the fossa was clearly distinguished by its characteristic deep concave shape, which complements the configuration of the condylar head for articulation (Fig. 2I). In contrast, mutants did not exhibit the depression of the temporal bone that forms the concave fossa, but instead possessed a flat temporal bone (Fig. 2J). In addition, even though the composition and shape of the mutant condyle appeared normal in the histological studies, the CT scan showed that the size of the condyle was markedly reduced (Figs. 2I, ,2J).2J). Together, our results substantiate the lack of glenoid fossa development when Spry1 and Spry2 are absent.
The increase in size of the temporalis and lateral pterygoid muscles was quantified in control and mutant embryos (Fig. 2K). In mutants, the temporalis and lateral pterygoid muscles were 48% and 69% larger relative to control embryos. No significant size difference in Meckel’s cartilage was observed, indicating that the effects of Spry1 and Spry2 deletion were specific to muscle. To better understand the mechanism responsible for muscle enlargement, we analyzed cell proliferation and apoptosis in the temporalis and lateral pterygoid muscles of control and mutant littermates at E14.5 and E17.5. High cell proliferation activity was observed at E14.5 throughout the head, and no apparent difference between controls and mutants was detected (data not shown). At E17.5, we detected a 33.1% and 46.2% increase in proliferating muscle progenitors in mutant pterygoid and temporalis muscles, respectively (Fig. 2L). No significant apoptosis at E14.5 or E17.5 was observed in the TMJ and no differences between controls and mutants (data not shown). These results suggest that the increase in muscle size in the mutants is likely due to an increase in proliferation of muscle progenitors.
To investigate whether Spry1 and Spry2 act cell-autonomously in muscle or whether they affect bone or cartilage, we generated mice harboring bone- or cartilage-specific inactivation of Spry1 and/or Spry2. The resultant mice did not show any phenotype in the TMJ (data not shown). Although we did not delete sprouty genes specifically in the muscle, the comparison of tissue-specific vs. global Spry1−/−;Spry2−/− mice suggests a cell-autonomous role for Spry1 and Spry2 in regulating the sizes of cranial muscles.
To confirm the cellular and molecular integrity of the condyle and disc formed in the absence of glenoid fossa in mutant mice, we analyzed the expression of key genes involved in cartilage and bone formation (Fig. 3). Sox9 and Acan (aggrecan), markers of proliferating and mature chondrocytes, respectively, showed similar expression patterns in control and mutant embryos (Figs. 3A--3D).3D). The expression of ColX (collagen type X), a marker for hypertrophic chondrocytes, was maintained in mutants, although it appeared to be reduced relative to controls (Figs. 3E, ,3F).3F). This difference is likely due to the smaller size of the condyle in mutant embryos. The expression of ColI (collagen type I), a marker for osteoblasts, remained unchanged in the condyle of control and mutant littermates (Figs. 3G, ,3H).3H). However, the strong expression of ColI in control fossa was not detected in mutants, consistent with the lack of a fossa in these embryos. To test the integrity of the disc and the attachment sites of the muscles to the bones in the absence of Spry1 and Spry2, we studied the expression pattern of Scx (scleraxis), a marker for tendons and ligaments, described to be regulated by Fgf signaling (Brent and Tabin, 2004). We observed that Scx was strongly expressed in the disc and the attachment points of muscle to bone in mutant and control mice (Figs. 3I, ,3J).3J). Thus, molecular analysis of the TMJ in control and mutant embryos confirmed normal development of the condyle and disc, which remarkably were not affected by the absence of the fossa.
The TMJ consists of multiple interacting tissues that are prone to injury- and disease-related degeneration. According to the National Institutes of Health (NIH), an estimated 3% to 5% of Americans suffer from a TMJ disorder. The lack of understanding of the development and function of the TMJ at the molecular level has hampered progress toward the diagnosis and treatment of TMJ disorders.
In the present work, we evaluated the role of sprouty genes during TMJ development. Fgf and Spry genes have previously been shown to have important roles during the development of various organs, including ear, tooth, lens, mandible, palate, and muscle (Shim et al., 2005; Boros et al., 2006; Klein et al., 2006; Goodnough et al., 2007; Mina et al., 2007; Welsh et al., 2007; Yang et al., 2010; Matsumura et al., 2011). We observed strong expression of Spry1 and Spry2 in the lateral pterygoid and temporalis muscles. We discovered that the combined inactivation of Spry1 and Spry2 resulted in overgrowth of these muscles, leading to the disruption of normal glenoid fossa development. Surprisingly, Spry1−/−;Spry2−/−embryos formed a complete condyle and joint disc, providing the first evidence that the condyle and disc form independently of the fossa.
The failure of glenoid fossa formation in Spry1−/−; Spry2−/−embryos may be due to one of three possibilities. First, the absence of the fossa could result from increased Fgf signaling in the absence of Spry1 and Spry2 in the muscle, which may inhibit the ossification of the forming bridge that ultimately gives rise to the temporal bone. Consequently, an overgrowth of the muscle could fill the space that otherwise would have been occupied by the fossa. Second, the lack of fossa may be due to a physical impediment to bone formation, perhaps due to excessive growth of the temporalis muscle that occupies the space between the two individual cartilages, preventing their fusion for fossa formation. This hypothesis is supported by the fact that cranial bones form by the fusion of many individual ossification centers (McBratney-Owen et al., 2008). We observed the formation of an isolated lateral bony fragment as early as E14.5. At E15.5 in control animals, it becomes part of the fossa. By contrast, in Spry1−/−;Spry2−/−embryos, the small bony fragment remains isolated in a position equivalent to the lateral distal tip of the fossa. Because this fragment expresses ColI and ColX, it is likely that bone has formed via endochondral ossification. In fact, only this portion of the fossa is missing in a Sox9 conditional knockout (Wang et al., 2011), suggesting that more than one mesenchymal condensation may give rise to the fossa, and that the fossa possesses multiple skeletal origins. Third, hyperactivated Fgf signaling may alter the fate of mesenchymal cells. The specific fate of an individual neural crest cell is determined by the signals they receive from the surrounding tissues (for review, see Trainor, 2010). The glenoid fossa and cranial muscles derive partly from cranial neural crest cells (Gu et al., 2008; Tzahor, 2009), and hyperactivation of Fgfr4 might favor muscle rather than bone differentiation, thereby generating temporalis muscle instead of fossa.
In summary, we showed that Fgfr4 expression is restricted to the cranial muscles and mimics the expression of Spry1, Spry2, and Spry4, suggesting that sprouty genes modulate signaling downstream of FGFR4 in the TMJ. Moreover, FGF6 is a key ligand of FGFR4 and has been reported to play a crucial role in myogenesis (reviewed in Armand et al., 2006). In the absence of Spry1 and Spry2, Fgf signaling via FGFR4 may be hyperactivated in the muscle, increasing myoblast proliferation, evidenced by increased cell proliferation in the mutants. Therefore, we suggest that condensations of the temporal bone that give rise to the glenoid fossa are able to form in the absence of Spry1 and Spry2, but the overgrown temporalis muscle impedes the fusion of these two elements to form the fossa (Fig. 4).
Future studies will need to investigate the relationship between sprouty genes and other signals involved in the growth and differentiation of muscle cells, as well as how these affect the formation of the fossa. Further molecular understanding of TMJ organogenesis is essential to improve diagnoses and develop novel therapeutic approaches for TMJ disorders.
We are grateful to Drs. Matthew Warman, Cliff Tabin, Richard Maas, Bjorn Olsen, Pamela Tran, Bryan Macdonald, Philip Stashenko, and Saunders Ching for technical assistance, suggestions, and discussion. We thank Drs. Gail Martin, Albert Basson, and Wenhan Chang for providing mouse lines.
This work was supported by NIDCR award K22DE-016309 to P.P., The Boston Plastic and Oral Surgery Foundation, Inc., a K99-DE022059 to A.H.J, and by the National Institutes of Health through the NIH Director’s New Innovator Award Program, 1-DP2- OD007191, to O.D.K.
The authors declare that there is no conflict of interest.