Abstract
Cleft palate, a common congenital deformity, can arise from disruptions in any stage of palatogenesis, including palatal shelf growth, elevation, adhesion, and fusion. Paired box gene 9 (Pax9) is recognized as a vital regulator of palatogenesis with great relevance to cleft palate in humans and mice. Pax9-deficient murine palatal shelves displayed deficient elongation, postponed elevation, failed contact, and fusion. Pax9 is expressed in epithelium and mesenchyme, exhibiting a dynamic expression pattern that changes according to the proceeding of palatogenesis. Recent studies highlighted the Pax9-related genetic interactions and their critical roles during palatogenesis. During palate growth, PAX9 interacts with numerous molecules and members of pathways (e.g., OSR2, FGF10, SHOS2, MSX1, BARX1, TGFβ3, LDB1, BMP, WNT β-catenin dependent, and EDA) in the mesenchyme and functions as a key mediator in epithelial-mesenchymal communications with FGF8, TBX1, and the SHH pathway. During palate elevation, PAX9 is hypothesized to mediate the time point of the elevation event in the anterior and posterior parts of the palatal shelves. The delayed elevation of Pax9 mutant palatal shelves probably results from abnormal expressions of a series of genes (Osr2 and Bmpr1a) leading to deficient palate growth, abnormal tongue morphology, and altered hyaluronic acid distribution. The interactions between PAX9 and genes encoding the OSR2, TGFβ3, and WNT β-catenin-dependent pathways provide evidence that PAX9 might participate in the regulation of palate fusion. This review summarizes the current understanding of PAX9’s functions and emphasizes the interactions between PAX9 and vital genes during palatogenesis. We hope to provide some clues for further exploration of the function and mechanism of PAX9, especially during palate elevation and fusion events.
Keywords
Introduction
Cleft lip and palate are recognized as common congenital maxillofacial deformities worldwide, with a global prevalence of 1:700 live births (Dixon et al. 2011). When cleft lip and palate were first described 100 y ago, they were regarded as the work of supernatural malefic forces. With the advancement in genetic and genomic analysis, a giant leap forward has been made in the etiology of cleft lip and palate. Like the embryonic development of other organs in the body, every tiny step of palatogenesis, from initiation to completion, is under strict control, as monitored by a gene network, tissue interaction, and environmental risk factors (Kaartinen et al. 1995; Rice et al. 2004; Lan and Jiang 2009; Burg et al. 2016; Li, Lan, and Jiang 2017).
Ever since the Pax (paired box) gene family was initially identified via a conserved DNA sequence element in the Drosophila segmentation genes paired (prd), gooseberry-proximal (gsb-p), and gooseberry-distal (gsb-d; Bopp et al. 1986), it has been studied to play paramount roles in ontogenesis for almost 3 decades. Paired box 9 (Pax9) mutant mice displayed cleft palate and tooth development arrest at bud stage and died shortly after birth (Peters et al. 1998; Kist et al. 2007). Paired box 9 (PAX9) hemizygosity in humans was first reported in 1999 (Schuffenhauer et al. 1999). Since then, researchers successively revealed the potential linkages between PAX9 and cleft palate in cases from America, Japan, Korea, and China (Slayton et al. 2003; Ichikawa et al. 2006; Lee et al. 2012; Song et al. 2013). Among these, the studies regarding PAX9 provided valuable insights into the developmental mechanisms involved in palatogenesis.
Herein, we review recent advances about the functions of PAX9 and its regulatory network in murine epithelial and mesenchymal cells and provide necessary elements for understanding and exploring the mechanism of palatogenesis.
Overview of Pax9 Expression during Murine Palatogenesis
The Pax9 gene, encoding the transcription factor in mesenchymal and epithelial cells, exhibits a dynamic expression pattern during palatogenesis (Fig. 1). Pax9 was initially detected at embryonic day 8.5 (E8.5) in the epithelium of murine foregut (Neubuser et al. 1995). At E12.5, during initial outgrowth, Pax9 was expressed throughout palatal shelves following oral-to-nasal and buccal-to-lingual gradient expressions in the coronal plane (Fig. 1D). As palatal shelves elongate vertically, Pax9 presented a more pronounced buccal-to-lingual gradient expression by E13.0 (Lan et al. 2004). Additionally, Pax9 in the horizontal plane displayed high-level expression in the posterior mesenchyme and along the medial side while exhibiting a contrast low-level expression in the anterior region (Lan et al. 2004; Fig. 1C). At E13.5, Pax9 remained a high-level expression in the buccal region throughout the shelves with an upregulation in the lingual region of the anterior coronal plane (Fig. 1F). It was strongly detectable in the medial region throughout the anterior-posterior axis in the horizontal plane (Lan et al. 2004; Zhou et al. 2013; Fig. 1E). Epithelial Pax9 was observed in the posterior region at E13.5 by Zhou et al. (2013) but failed to be detected in Jia and colleagues’ study (Jia, Zhou, Fanelli, et al. 2017). At E14.5, the amount of mesenchymal Pax9 was reduced while confining to the medial edge epithelium (MEE; Sasaki et al. 2007; Fig. 1G, H). From E14.5 to E15.5, epithelial Pax9 gradually faded with the disappearance of the midline epithelium seam (MES) and fusion of the shelves (Sasaki et al. 2007; Fig. 1I, J). The variation of Pax9’s distribution correlated precisely with the proceeding of developmental events.
Development of murine palate and the dynamic expression patterns of paired box gene 9 (Pax9). The coronal and horizontal planes of the palate during the process of palatogenesis are ranged according to the timeline. (
Pax9 mutant palatal shelves appeared an abnormally broadened shape in the nasal side and with deficient elongation resulting from the decreased mesenchymal proliferation, especially in the middle and posterior regions (Peters et al. 1998; Zhou et al. 2013; Jia, Zhou, Fanelli, et al. 2017). At E14.5, delayed elevation occurred to one or both sides of Pax9 mutant shelves (Peters et al. 1998; Zhou et al. 2013; Li, Lan, Krumlauf, and Jiang 2017). Eventually, separated palatal shelves with irregular and poorly developed palatal rugae were observed in Pax9 mutants (Peters et al. 1998; Zhou et al. 2013).
Subsequent studies demonstrated that PAX9 participated in a massive network during palatogenesis, including epithelial-mesenchymal communications of palate growth (initiate outgrowth and growth after elevation), palatal shelf elevation, and fusion (Peters et al. 1998; Sasaki et al. 2007; Zhou et al. 2013). This regulatory landscape involved multiple pathways (e.g., SHH, EDA, WNT, BMP) and numerous genes (e.g., Osr2, Msx1, Fgf10, Fgf8, Tbx1, Tgfβ3, Barx1, Ldb1; Tables 1 and 2).
Locations and Functions of Pax9-related Genes in Palatogenesis.
MEE, medial edge epithelium; Pax9, paired box gene 9.
Time expressed in embryonic day (E13.5 to E15.5).
Locations and Functions of Pax9-Related Pathways in Palatogenesis.
Time expressed in embryonic day (E13.5 to E15.5).
Pax9, paired box gene 9.
Regulatory Network of PAX9 in Palatal Growth
The Regulatory Network Involving PAX9 within the Palatal Mesenchyme
Mesenchymal PAX9 coordinates with plenty of mesenchymal molecules for promoting palate cell proliferation, especially in the middle and posterior regions (Fig. 2).
Paired box gene 9 (Pax9)-involved regulatory networks of anterior and posterior palatal shelves: palate growth (
Regulations between OSR2 and PAX9 Varies from the Anterior to Posterior Regions of Palate Shelves
Osr2 is expressed throughout the anterior-posterior mesenchyme, and the mutation of which exhibited deficient mesenchymal cell proliferation and delayed shelf elevation (Lan et al. 2004; Fig. 3A).
The expression domains of Pax9-related genes and crucial members of the Pax9-related pathway. Expressions in the coronal planes of the anterior and posterior palatal shelves: Osr2 (
OSR2 was reported to interplay with some molecules during palatogenesis and odontogenesis, including PAX9, MSX1, FGF10, and the members of the SHH, BMP, and WNT pathways (Lan et al. 2004; Zhou et al. 2013; Jia et al. 2016; Fu et al. 2017). Osr2 acts downstream of Pax9 for activating mesenchymal odontogenic potential during tooth development (Zhou et al. 2011). In addition, Osr2 and Pax9 display a more intricate interrelationship in palate growth.
Osr2tm1Jian/tm1Jian mutants presented a lower expression of Pax9 in the middle plane at E13.5 (Lan et al. 2004). Osr2 expression decreased in the middle and posterior parts of palatal shelves, whereas it increased in the very anterior part of the mesenchyme in Pax9del/del mutants (Zhou et al. 2013). To further investigate the relationship between Pax9 and Osr2, Pax9Osr2KI/Osr2KI embryo—in which Osr2 was expressed from the Pax9 locus—was created and exhibited partially restored cell proliferation and fused shelves in the posterior region; nevertheless, it showed a cleft palate and increased cell proliferation in the anterior region as compared with Pax9del/del embryo (Zhou et al. 2013). Thus, in the middle and posterior regions, PAX9 mediates cell proliferation partially via regulating Osr2 expression. However, in the anterior region, the regulatory pattern of Osr2 and Pax9 needs to be extensively investigated.
OSR2 cooperates with PAX9 to maintain the expression of Fgf10 for normal palatal morphology in the anterior region (Zhou et al. 2013). SHOX2 is the positive regulator of Fgf10 (Yu et al. 2005). Shox2 expression, occupying the anterior mesenchyme of palatal shelves, displayed a defused border and an anterior shift in Pax9del/del and Pax9Osr2KI/Osr2KI mutants (Yu et al. 2005; Sun et al. 2013; Zhou et al. 2013; Fig. 3B), which indicated that Osr2 functions parallel to Shox2. From the aforementioned results, we concluded that OSR2 and PAX9-SHOX2 independently mediate Fgf10 expression during the growth of the anterior region.
Members of the BMP Pathway Interact Distinctively with PAX9
The BMP pathway expressed in the buccal region of the palatal shelves plays a vital role during palatal growth. BMP4 and BMP2, ligands of the BMP pathway, recruit and translocate SMAD1/5/8 to the nucleus via binding to BMPR1A receptors (Shi and Massagué 2003; Massagué 2012; Li et al. 2013; Figs. 3C, 4C).
Pathways that embrace a regulatory relationship with paired box gene 9 (Pax9): WNT (
Researchers revealed that the BMP pathway acted as an upstream regulator of Pax9 in tooth development and inhibited Pax9 expression through FGF proteins in the mandibular mesenchyme (Neubuser et al. 1997; Feng et al. 2017). As for palatogenesis, members of the BMP pathway interact differently with Pax9 during palate growth. Pax9del/del palatal shelves, displaying aberrant morphology, exhibited a reduction of Bmp4 and Smad1/5/8 expressions throughout the mesenchyme, particularly in the posterior region at E13.5 (Zhou et al. 2013; Li, Lan, Krumlauf, and Jiang 2017). The results confirmed that PAX9 functions as a positive regulator of the BMP pathway during palate growth.
Bmpr1a appears specifically in the primary palate and anterior part of palatal shelves (Baek et al. 2011; Fig. 3C). Tissue-specific elevation of Bmpr1a in the mesenchyme (Wnt1-Cre;pMes-caBmprIa) led to reduced cell proliferation in the anterior region, delayed shelf elevation, and subsequent complete cleft of the secondary palate (Li et al. 2013). Inactivation of mesenchymal Bmpr1a (Osr-IresCre Bmpr1afl/fl) presented reduced cell proliferation and submucous cleft with a comparable Pax9 expression in wild type at E13.5 (Baek et al. 2011). The inactivation of Bmpr1a (Nestin-Cre Bmpr1afl/fl) gave rise to an expanded expression area of Pax9 in the anterior region at E12.5, reduced cell proliferation, and partially cleft anterior palate (Liu et al. 2005). The results imply potential connections between Pax9 and Bmpr1a during palate growth. However, researchers are encouraged to dig into the questions of when the regulation happens and whether Pax9 is regulated by mesenchymal or epithelial BMPR1A.
WNT Agonists Uncover the Connections between PAX9 and the WNT β-catenin-Dependent Pathway
The WNT β-catenin-dependent pathway was strongly relevant to embryonic development and tumorigenesis in mice and humans (Clevers 2006; Liu et al. 2007; Chen et al. 2009), and researchers revealed its potential function during palate growth, elevation, and fusion (He et al. 2011; Feng et al. 2013).
The dynamic activity of the WNT β-catenin-dependent pathway, indicated by BATGAL, is observed primarily in the mesenchyme of the anterior region of the palatal shelves at E12.5, suggesting its critical role in the early stage of palate growth (Liu et al. 2015). At E13.5, BATGAL is detected in MEE along the anterior-posterior axis of the shelves with a scattered expression in the mesenchyme (He et al. 2008; He et al. 2011; Liu et al. 2015; Fig. 3D). The BATGAL activity expands to palatal rugae epithelium at E14.5 (He et al. 2011; Liu et al. 2015).
The initiation of the β-catenin-dependent pathway requires the binding of WNT ligands to FRIZZLED and LRP5/6 receptors to maintain the expression of β-catenin. Afterward, the nuclear translocation of β-catenin triggers the expression of Axin2 (Clevers 2006; Liu et al. 2015; Table 2; Fig. 4A).
PAX9 and the WNT pathway embrace an extensive regulatory relationship. During early tooth bud formation, PAX9 acts in parallel to the WNT β-catenin-dependent pathway (Chen et al. 2009). Regarding palatogenesis, the active β-catenin and mesenchymal Axin2 expression in the posterior palate were dramatically downregulated in Pax9del/del mice at E13.5, elucidating that PAX9 is upstream of the WNT β-catenin-dependent pathway (Li, Lan, Krumlauf, and Jiang 2017).
The inactivation of Wise, serving as the negative mediator of the WNT pathway, was also found to rescue palatal morphogenesis without rescuing the malformed rugae in Pax9del/del mice, demonstrating that PAX9 is upstream of the WISE/WNT pathway (Li, Lan, Krumlauf, and Jiang 2017). However, Pax9del/del showed a decreased Wise expression in epithelium and mesenchyme at E13.5 (Li, Lan, Krumlauf, and Jiang 2017). This expression pattern of Wise poses obstacles to figure out how WISE interacts with PAX9 before mediating the WNT β-catenin-dependent pathway.
DKK1 and DKK2, the inhibitors of the WNT pathway, significantly induced in Pax9-deficient palatal mesenchyme at E13.5 in the posterior region (Jia, Zhou, Fanelli, et al. 2017; Li, Lan, Krumlauf, and Jiang 2017). The in utero delivery of a WNT agonist named WAY-262611, the inhibitor of Dkk1, rescued cell proliferation and palatine bone formation and resulted in an intact palate as compared with Pax9-/- littermates (Jia, Zhou, Fanelli, et al. 2017). Pax9-/- mutant mice treated with in utero delivery of Dkk2’s inhibitor, named IIIC3a, rescued palate morphogenesis and exhibited partially fused palate in the middle and posterior regions, which further confirmed that the WNT β-catenin-dependent pathway is positively monitored by PAX9, especially in the posterior region during palate growth (Li, Lan, Krumlauf, and Jiang 2017). As mentioned, Pax9-/- palate presented different phenotypes while being delivered different inhibitors of Dkk1, Dkk2, and Wise. This phenomenon might be due to the different expression patterns of these genes. However, more experiments should be carried out for further investigations.
Apart from the WNT β-catenin-dependent pathway which plays an extraordinary role in posterior palatal growth, there is a unique mesenchymal expression of Barx1 in the posterior region controlling posterior mesenchymal cell proliferation (Welsh et al. 2007; Fig. 3E). Barx1 expression shifted anteriorly in Pax9 mutant palatal shelves at E13.5 (Smith et al. 2012; Zhou et al. 2013), pointing out the possibility that PAX9 modulates Barx1 expression during posterior mesenchymal proliferation.
Anti-EDAR monoclonal antibody 1, known as AbEDAR, was proven to regain the expression level of the EDA pathway and result in an intact palate in Pax9-/- embryo with abnormal shapes of palatal rugae. It indicated that PAX9 regulates anterior and posterior palatal growth partially through the EDA pathway (Jia, Zhou, Wee, et al. 2017). The EDA pathway interacts complicatedly with the WNT pathway in developing a series of murine ectodermal organs, including skin appendages and tooth (Laurikkala et al. 2001; Kowalczyk-Quintas and Schneider 2014). The WNT pathway is known to be rather conservative, from which the hypothesis could be made that this intimate relationship might also exist in palatogenesis.
Specific Deletion of Mesenchymal Ldb1 Points Out That PAX9 Might Be the Downstream Target of LDB1 during Palate Morphology
Ldb1 is expressed in epithelium and mesenchyme of palatal shelves (Almaidhan et al. 2014; Fig. 3F). Murine palatal shelves with specific deleted mesenchymal Ldb1 (Wnt1-Cre;Ldb1fl/-) exhibited downregulated Pax9 expression throughout the palatal mesenchyme at E13.5 and appeared in an abnormally broadened in the nasal side and failed to elevate (Almaidhan et al. 2014). Although the relationship between these genes has not been fully studied, a comparison of the impairments pointed out that Wnt1-Cre;Ldb1fl/- and Pax9-deficient mutants shared a similar misshaped palate morphology (Peters et al. 1998; Almaidhan et al. 2014), indicating that PAX9 might harbor a regulatory relationship with LDB1 during palate morphology.
Mesenchymal PAX9 Participates in Epithelial-Mesenchymal Communications
The SHH pathway is recognized as a key early regulator in reciprocal epithelial-mesenchymal communications controlling epithelial and mesenchymal proliferation (Lan and Jiang 2009). Shh was detected in the MEE of anterior palatal shelves at E12.5 and E13.5 (Rice et al. 2006; Han et al. 2009; Xu et al. 2016), and palatal rugae gradually exhibited intense expression of Shh from E12.5 to E15.5 (Han et al. 2009; Lan and Jiang 2009; Xu et al. 2016; Fig. 3G). Shh mutation contributed to retarded palatal shelves and abnormal palatal rugae (Lan and Jiang 2009; Welsh and O’Brien 2009).
The SHH pathway tightly links to FGFs (e.g., epithelial FGF8 and mesenchymal FGF10) which have a regulatory relationship with PAX9 during palatogenesis (Rice et al. 2004; Stanier and Pauws 2012; Zhou et al. 2013). The ectopic activation of Fgf8 in the mesenchyme brought about the impaired elevation of the whole palate and enlarged posterior palatal mesenchyme, revealing its role as an upstream regulator of Shh (Wu et al. 2015). Shh and Fgf10 expressed in the mesenchyme function in a positive feedback loop during epithelial-mesenchymal communications and together stimulated cell proliferation in palatal tissue (Rice et al. 2004; Lan and Jiang 2009; Wu et al. 2015; Fig. 3H). Pax9 was shown to be induced by FGF8 in the explants of posterior palatal mesenchyme (Hilliard et al. 2005). PAX9 maintained Fgf10 expression in anterior palatal shelves with the help of OSR2 (Zhou et al. 2013). Pax9del/del mutants presented reduced Shh expression in the epithelium from E12 to E14.5, which led to aberrant shelf elongation and disorganized rugae morphology (Zhou et al. 2013). These findings elucidate the great possibility that PAX9 participates in the epithelial-mesenchymal communications by interacting with SHH and FGFs.
Recent research demonstrated that Tbx1 mutants displayed a significantly reduced palatal mesenchymal cell proliferation, which might be the cause of slightly downregulated Pax9 expression (Funato et al. 2012; Zoupa et al. 2018). With the discoveries that Tbx1 mutants exhibited downregulation of Fgf8 expression and FGF8 increased Pax9 expression in the explants of the posterior palatal mesenchyme, TBX1 might indirectly mediate Pax9 expression via regulating Fgf8.
In addition to FGF8, FGF10, and TBX1, a considerable number of molecules correlate with PAX9 and SHH, further verifying that PAX9 takes an active part in epithelial-mesenchymal communications. MSX1, confining to anterior mesenchyme of palatal shelves, modulates in a positive feedback loop with BMP4, which is essential for maintaining Shh expression in the anterior region (Zhang et al. 2002; Alappat et al. 2003; Smith et al. 2012; Fig. 3I). MSX1 orchestrates with PAX9 in promoting odontogenesis and palatogenesis (Nakatomi et al. 2010; Zhou et al. 2013). MSX1 and PAX9 interact synergistically throughout lower incisor development (Nakatomi et al. 2010). Msx1 expression was reduced in the anterior region of palatal shelves in Pax9del/del embryos (Zhou et al. 2013). These results demonstrate that mesenchymal PAX9 indirectly monitors Shh expression by altering the expressions of Msx1 and Bmp4.
PAX9-Involved Regulatory Network in Elevation
Loss of Pax9 expression contributed to the postponement of approximately 1 embryonic day in anterior and posterior palatal shelf elevation in mice (Peters et al. 1998; Zhou et al. 2013). Three following potential factors could attribute to this phenomenon: deficient shelf growth, malformed tongue, and decreased mesenchymal hyaluronic acid (HA). HA was hypothesized to play an essential role in processing elevation (Ferguson 1988; Zhou et al. 2013; Almaidhan et al. 2014; Lan et al. 2016; Li, Lan, and Jiang 2017; Li, Lan, Krumlauf, and Jiang 2017).
Pax9 is expressed in epithelial and mesenchymal cells in tongues (Jonker et al. 2004; Lan et al. 2004; Almaidhan et al. 2014). Malformed tongues were observed in Pax9 mutants. However, conclusions have not been made concerning the Pax9-related malformed tongue in recent reports (Zhou et al. 2011; Zhou et al. 2013). Although the malformed tongue has been observed in delayed or failed elevation mouse models and was presumed as one of the factors disturbing abnormal elevation by many researchers, there is a possibility that the tongue morphology is not critical for shelf elevation (Ferguson 1988; Goudy et al. 2010; Jin et al. 2010; Yu and Ornitz 2011; Almaidhan et al. 2014). Therefore, further explorations are suggested to figure out the underlying connection between tongue morphology and shelf elevation, which could provide the basic knowledge for understanding the function of PAX9 during the palate elevation period.
Accumulation of HA was detected in mesenchymal cells throughout the palate and was demonstrated to play an essential role in elevation (Li, Lan, and Jiang 2017). Pax9del/del mutants exhibited decreased HA at E13.5, which might be the cause of delayed elevation (Li, Lan, Krumlauf, and Jiang 2017). Intriguingly, inactivation of Wise in Pax9del/del mutants could partially restore HA at E13.5 and lead to punctual elevation (Li, Lan, Krumlauf, and Jiang 2017), suggesting that the WNT pathway might be downstream of Pax9 in triggering HA expression in mesenchymal cells.
During E13.5 to E14.5, the critical period of elevation, Pax9 expression expanded from the buccal and lingual mesenchyme to MEE (Sasaki et al. 2007; Zhou et al. 2013). How PAX9 mediates the elevation is largely unknown. Nevertheless, relationships between PAX9 and several molecules provide some trails for exploring the PAX9-related mechanism of elevation (Fig. 2).
Osr2 and Bmpr1a Mutants Exhibited Delayed Palate Elevation
A previous study revealed postponed elevation in anterior and posterior regions of Osr2tm1jian/tm1jian palatal shelves (Lan et al. 2004). Osr2tm1jian/tm1jian palate exhibited uniformly low expression of mesenchymal Pax9 throughout the palatal shelves from E13.5 to E14.5 (Lan et al. 2004). The researchers showed the 4 of 48 Pax9Osr2KI/Osr2KI mutants elevated palatal shelves timely at E14.5. However, these mice still exhibited unfused anterior palatal shelves (Zhou et al. 2013). The results revealed complicated interactions between OSR2 and PAX9 during palate elevation.
Mutant mice with specifically inactivated Bmpr1a in the palate mesenchyme (Osr2-IresCre;Bmpr1af/f) showed delayed palatal shelf elevation, postponing to E15.5 (Baek et al. 2011), which followed an identical pattern with Pax9del/del littermates (Zhou et al. 2013). The results pointed out a potential relationship between Pax9 and Bmpr1a during palatal shelf elevation. However, in the same mutants, Pax9 was illustrated to remain in a comparable expression pattern with wild type (Baek et al. 2011); hence, more experiments need to be included for better understanding of the regulatory relationship between Bmpr1a and Pax9 during elevation.
As mentioned, OSR2 and BMPR1A were suspected to function with PAX9 in cell proliferation during palate growth, which is a vital element for driving normal elevation (Almaidhan et al. 2014; Li, Lan, and Jiang 2017). Adding to the fact that Pax9, Osr2, and Bmpr1a mutants were suffering from delay rather than failure in elevation, the postponed elevation may be caused by the defected palatal growth. Thus, PAX9 is deduced to mediate the time point of palate elevation through orchestrating with a series of molecules in palate growth, tongue morphology, and HA distribution.
PAX9-Involved Regulatory Network in Palatal Fusion
Pax9 mutant shelves failed to contact each other, arising from aberrant cell proliferation during palate growth and delayed elevation (Peters et al. 1998; Zhou et al. 2013). The phenotype suggests that there is no chance for Pax9 mutant shelves to fuse since the shelf contact does not occur under this circumstance. Nevertheless, PAX9 might have chances to perform in palate fusion for the reason that Pax9 appears to have a sharp expression along MEE cells at E14.5 before fading with the disappearance of the MES at the end of fusion (Hamachi et al. 2003; Sasaki et al. 2007). Moreover, intricate interactions between Pax9 and key genes of fusion offered clues that PAX9 probably participates in palate fusion (Fig. 2).
The Relationship between Epithelial TGFβ3 and PAX9
TGFβ3 is required for removing the periderm layer and promoting the disappearance of MES during palate fusion (Proetzel et al. 1995; Jin et al. 2014; Ke et al. 2015; Hammond et al. 2017; Fig. 3J). Tgfβ3-/- palatal shelves exhibited low expression of Pax9 throughout the anterior-posterior axis during the critical timing of fusion (Sasaki et al. 2007). Therefore, PAX9 may be involved in TGFβ3 regulation during palatal fusion.
Activations of the WNT β-catenin-Dependent Pathway and Osr2 Resulted in the Fused Pax9 Mutant Palate
Mesenchymal OSR2 and the WNT β-catenin-dependent pathway embraced regulatory relationships with TGFβ3 and exhibited unfused palatal shelves (Lan et al. 2004; He et al. 2011; Barrio et al. 2014; Hu et al. 2015).
A recent study clarified that activating the WNT pathway could rescue cleft palate in Pax9-/- mutants (Jia, Zhou, Fanelli, et al. 2017). Pax9Osr2KI/Osr2KI showed a fused posterior region and a separated anterior region (Zhou et al. 2013). However, whether the fusion arose from normal mesenchymal proliferation and/or shelf adhesion and fusion is unclear.
The previous paragraphs point out that although the expression pattern and regulatory relationship of PAX9 imply the possibility of its role in palatal fusion, in vitro palate fusion assay is suggested to determine if normal adhesion and fusion could occur to Pax9 mutants.
Conclusion and Outlook
In brief, numerous genes have been proven to participate in palatogenesis according to previous studies. They shed light on the importance of the genetic regulatory networks and matched the networks to certain stages of palatogenesis via tracking the dynamic expression patterns and phenotypes presented by the mutations of the genes involved (Li, Lan, and Jiang 2017).
Since cleft palate was observed in Pax9-deficient mice by Peters et al. (1998), great progress has been obtained in the function and regulation of PAX9 in mice. The essential role of PAX9 in promoting palatal mesenchyme proliferation has been extensively studied and well confirmed (Zhou et al. 2013; Jia, Zhou, Fanelli, et al. 2017). Nevertheless, the specific function of Pax9 in palate elevation and fusion has not yet been figured out. We summarize the linkages between Pax9 and critical genes in palate elevation and fusion, and thus a hypothesis regarding whether and how Pax9 participates in elevation and fusion events has been put forward. Additionally, future investigations are encouraged for a better understanding of the functions of PAX9 and the mechanisms of palatogenesis.
Author Contributions
R. Li, contributed to conception, design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript; Z. Chen, contributed to conception, design, data analysis, and interpretation, drafted and critically revised the manuscript; Q. Yu, contributed to conception and data interpretation, critically revised the manuscript; M. Weng, contributed to data interpretation, critically revised the manuscript; Z. Chen, contributed to conception, design, and data interpretation, critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work.
