Midline Epithelial Seam Research Articles

Lithium inhibits palatal fusion and osteogenic differentiation in palatal shelves in vitro

ObjectiveGlycogen synthase kinase-3β (Gsk-3β)/β-catenin signaling regulates development of the secondary palate. It has been unclear about the effects of Gsk-3β/β-catenin signaling on palatal fusion and osteogenic differentiation in palatal shelves. DesignIn this study, palatal shelves from mouse embryonic day 13 (E13) were cultured in vitro with or without lithium chloride (LiCl). Palatal fusion was evaluated by haematoxylin-eosin staining. The expression of osteogenic markers in palatal shelves was measured by quantitative PCR, and immunohistochemical staining. Cell proliferation and apoptosis were examined by Ki-67 immunohistochemical and TUNEL staining, respectively. Gsk-3β expression was evaluated by quantitative PCR and Western blotting. β-catenin protein expression was evaluated by Western blotting. ResultsAfter the treatment with 10mM LiCl, palatal shelves failed to fuse, and the mRNA and protein levels of osteogenic markers were reduced compared with controls. The number of Ki67-positive cell in the palatal osteoid was significantly higher in the LiCl group than in the controls. The apoptotic cells in the midline epithelial seam were reduced by LiCl. Gsk-3β mRNA and protein expression levels decreased and β-catenin protein expression levels increased by treatment of LiCl. ConclusionOur findings show that LiCl-mediated GSK3β inhibition prevents palatal fusion and osteogenic differentiation in palatal shelves by increased β-catenin signaling. It indicated that overactivation of canonical Wnt signaling might impair the fusion of the secondary palate.

Ephrin reverse signaling induces mouse palatal fusion without TGFβ3 signaling (921.4)

Orofacial clefts are the most prevalent craniofacial birth defects. The secondary palate forms from mesenchymal shelves covered with epithelium which adheres to form the midline epithelial seam (MES). MES cells then proceed through epithelial to mesenchymal transition (EMT) forming a fused palate. TGF‐β3 is essential for the disintegration of the palatal MES. Eph receptor tyrosine kinases and their ephrin ligands are responsible for many developmental processes. Binding of ephrins causes receptor activation in Eph‐bearing cells (forward signaling), and intracellular signaling inside ephrin‐bearing cells (reverse signaling). We asked if activation of ephrin reverse signaling is sufficient to fuse mouse palates if we blocked TGFβ receptors. We cultured E13.5 mouse palatal shelves in different conditions: Control IgG Fc; anti‐TGF‐β3 with or without clustered recombinant EphB2/Fc protein; TGF‐β RI kinase inhibitor (SB 431542) with or without EphB2/Fc. Palatal fusion was quantified and analyzed with a mean fusion score (MFS) in anterior, medial and posterior regions. Palates treated with anti‐TGF‐β3 or (SB 431542) did not fuse while those treated with SB 431542 and EphB2/Fc and those treated with anti‐TGF‐β3 and EphB2/Fc were partially fused. Palates from the control groups fused. These results demonstrate that ephrin reverse signaling induces fusion of the mammalian palate without TGF‐β3 signaling.Grant Funding Source: Biomedical Sciences Department, baylor College of Dentistry, Baylor Oral Health Foundation

Palatal adhesion is dependent on Src family kinases and p38MAPK.

During secondary palate development, palatal shelves adhere to each other in the midline to form a midline epithelial seam leading to palatal closure. Cell-cell and cell-extracellular matrix adhesions, which are mediated by cell adhesion receptors, E-cadherin and integrins, are implicated in the process of adhesion of the palatal shelves. Src family kinases (SFK) function downstream of both receptors. In this study, we focused on the role of SFK in the process of palatal adhesion. During palatal adhesion, the expression of SFK mRNA, as well as localization and quantitation of the protein in the activated form, were examined by real-time qPCR and immunofluorescence. Palatal organ cultures were performed to identify the effect of pharmacological inhibition of SFK on palatal adhesion. Activated SFKs were found to be co-localized with adhesion receptors, E-cadherin and integrins in the palatal medial edge epithelium. Src, Fyn and Yes subfamily members were expressed in the palatal tissue. The expression of SFK mRNA and the quantity of the activated form of the protein were upregulated during palatal adhesion. An SFK inhibitor, PP2, blocked palatal adhesion, but another SFK inhibitor, SU6656 was not inhibitory. However, the combination of SU6656 and either of the p38MAPK inhibitors, SB203580 or BIRB0796, showed similar inhibitory effects on palatal adhesion compared to PP2 alone. The p38MAPK inhibitors alone did not alter palatal adhesion. Real-time qPCR revealed that p38MAPK alpha and delta were elevated during palatal adhesion. This study indicates that palatal cell adhesion is dependent on signaling from integrin receptors and E-cadherin through SFK and p38MAPK.

Ephrin Reverse Signaling Induces Mouse Palatal Fusion and Epithelial To Mesenchymal Transition In Cultured Medial Edge Epithelia

Understanding the molecular events in palate development is a prerequisite to more effective treatment of cleft palate. The secondary palate in humans and mice forms from shelves of mesenchyme covered with epithelium. These shelves adhere to form the midline epithelial seam (MES). MES cells then proceed through epithelial to mesenchymal transition (EMT) and apoptosis to yield a fused palate. This fusion requires Transforming Growth Factor β3 (Tgfβ3). Eph receptor tyrosine kinases and their ephrin ligands are responsible for many contact‐mediated developmental processes. Binding of ephrins causes receptor activation in Eph‐bearing cells (forward signaling), and intracellular signaling inside ephrin‐bearing cells (reverse signaling). Previously, we established that ephrin‐B reverse signaling is required for palate fusion in mice and is sufficient to cause fusion in chicken palates without Tgfβ3. In this study, we asked if activation of ephrin reverse signaling is sufficient to fuse mouse palatal shelves in vitro in the absence of Tgfβ3. We cultured embryonic day 13.5 mouse palatal shelves with a blocking antibody against Tgfβ3 that prevents fusion and clustered recombinant EphB2/Fc protein. EphB2/Fc treatment restored fusion. We also found that treatment with EphB2/Fc of cultured cells isolated from MES induced phenotypic and gene expression changes indicative of EMT. These results demonstrate that ephrin reverse signaling is sufficient to cause fusion of the mammalian palate.Grant Funding Source: RO3DE020119 (NIDCR)

Microtubule disassembly prevents palatal fusion and alters regulation of the E-cadherin/catenin complex

During palatal fusion, the midline epithelial seam (MES) degrades to achieve mesenchymal confluence. Epithelial mesenchymal transition (EMT) is one mechanism which is active in MES degradation. TGF-β induces EMT in medial edge epithelium (MEE) by down-regulation of an epithelial marker, E-cadherin. Microtubule disassembly impaired palatal fusion leading to a multi-layered MES in the mid-region. In this study, we investigated the effect of microtubule disruption on the regulation of the E-cadherin/catenin adhesion complex. Nocodazole (NDZ) enhanced the accumulation of the adhesion complex at cell-cell contacts in MEE, while loss of the adhesion complex was observed in the control. NDZ caused aberrant regulation of the E-cadherin transcriptional repressors (Snail and Zeb) and the activator (c-MYC) through inhibition of the TGF-β/SMAD2 signaling pathway, which led to a failure in EMT. These results suggest that the microtubule cytoskeleton plays an important role in mediating TGF-β/SMAD2 signals to control E-cadherin gene expression in MEE during palatal fusion.

Analysis of Snail1 function and regulation by Twist1 in palatal fusion

Palatal fusion is a tightly controlled process which comprises multiple cellular events, including cell movement and differentiation. Midline epithelial seam (MES) degradation is essential to palatal fusion. In this study, we analyzed the function of Snail1 during the degradation of the MES. We also analyzed the mechanism regulating the expression of the Snail1 gene in palatal shelves. Palatal explants treated with Snail1 siRNA did not degrade the MES and E-cadherin was not repressed leading to failure of palatal fusion. Transforming growth factor beta 3 (Tgfβ3) regulated Snail1 mRNA, as Snail1 expression decreased in response to Tgfβ3 neutralizing antibody and a PI-3 kinase (PI3K) inhibitor. Twist1, in collaboration with E2A factors, regulated the expression of Snail1. Twist1/E47 dimers bond to the Snail1 promoter to activate expression. Without E47, Twist1 repressed Snail1 expression. These results support the hypothesis that Tgfβ3 may signal through Twist1 and then Snail1 to downregulate E-cadherin expression during palatal fusion.

Rgs19 regulates mouse palatal fusion by modulating cell proliferation and apoptosis in the MEE

Palatal development is one of the critical events in craniofacial morphogenesis. During fusion of the palatal shelves, removal of the midline epithelial seam (MES) is a fundamental process for achieving proper morphogenesis of the palate. The reported mechanisms for removing the MES are the processes of apoptosis, migration or general epithelial-to-mesenchymal transition (EMT) through modulations of various signaling molecules including Wnt signaling. RGS19, a regulator of the G protein signaling (RGS) family, interacts selectively with the specific α subunits of the G proteins (Gαi, Gαq) and enhances their GTPase activity. Rgs19 was reported to be a modulator of the Wnt signaling pathway. In mouse palatogenesis, the restricted epithelial expression pattern of Rgs19 was examined in the palatal shelves, where expression of Wnt11 was observed. Based on these specific expression patterns of Rgs19 in the palatal shelves, the present study examined the detailed developmental function of Rgs19 using AS-ODN treatments during in vitro palate organ cultivations as a loss-of-function study. After the knockdown of Rgs19, the morphological changes in the palatal shelves was examined carefully using a computer-aided three dimensional reconstruction method and the altered expression patterns of related signaling molecules were evaluated using genome wide screening methods. RT-qPCR and in situ hybridization methods were also used to confirm these array results. These morphological and molecular examinations suggested that Rgs19 plays important roles in palatal fusion through the degradation of MES via activation of the palatal fusion related and apoptotic related genes. Overall, inhibition of the proliferation related and Wnt responsive genes by Rgs19 are required for proper palatal fusion.

Transient Appearance of Tyrosine Hydroxylase Immunoreactive Cells in the Midline Epithelial Seam of the Human Fetal Secondary Palate

Transient immunoreactivity for tyrosine hydroxylase, which mediates the conversion of the amino acid L-tyrosine to dihydroxyphenylalanine, in the midline epithelial seam between the bilateral palatal shelves was investigated in human fetuses. Horizontal or frontal paraffin sections of two human fetuses at 9 and 15 weeks of gestation were used to examine the distribution of tyrosine hydroxylase-immunoreactive cells in regions of the entire head other than the brain. Immunohistochemical staining for S100 protein, calretinin, cytokeratin 14, and vimentin was examined using adjacent or near sections. Tyrosine hydroxylase-immunoreactive cells were large and densely distributed in the midline epithelial seam at the site of palatal fusion in fetuses at 9 weeks but not in fetuses at 15 weeks, in which the midline epithelial seam had already disappeared. No expression of S100 protein, calretinin, or vimentin was detected, but the midline epithelial seam was positive for cytokeratin 14. Tyrosine hydroxylase immunoreactivity was not detected in epithelia during the process of palatal fusion in mice from E 14.0 to 15.0. These findings indicate that tyrosine hydroxylase-immunoreactive cells in the midline epithelial seams are nonneural epithelial cells and suggest that the tyrosine hydroxylase is a novel factor involved in normal palatal formation, especially the fate of the midline epithelial seam in humans.

Inhibition of Smad signaling is implicated in cleft palate induced by all-trans retinoic acid

The effect of all-trans retinoic acid (atRA) on palatal fusion and the underlying mechanisms were investigated using organ culture. Compared with control group, the atRA-treated group (1 μM and 5 μM) had more medial edge epithelium (ME) remaining within the midline epithelial seam (MES). At 10 μM atRA, the opposing shelves were not in contact at the culture end (72 h). Cell death detection by TUNEL and laminin immunohistochemistry demonstrated that atRA (5 μM) induced apoptosis in mesenchyme and inhibited degradation of basal lamina within MES. Notably, migration and apoptosis of ME cells and degradation of basal lamina within MES markedly represented vehicle control palatal shelves in culture. Additionally, apoptosis was not detected in mesenchyme of control palatal shelves. Immunoblotting analysis revealed that Smad2 and Smad3 were endogenously activated and expression of Smad7 was inhibited during the fusion process. In contrast, atRA treatment abrogated phosphorylation of Smad2 and Smad3 and inducible expression of Smad7 in ME. From these data, it is assumed that inhibition of Smad pathway by atRA in ME may play a critical role in abrogation of the ME cell apoptosis and degradation of the basal laminin, which might contribute to failure of palatal fusion.

Tgf-β-mediated FasL-Fas-Caspase Pathway Is Crucial during Palatogenesis

Programmed cell death, or apoptosis, is one of the fates of the medial edge epithelium (MEE) during palatal fusion. Transforming growth factor β (Tgf-β) signaling (such as Tgf-β3) is required for the disappearance of the MEE, but the relationship between Tgf-β3 and apoptosis remains unclear. Here we show that the Fas ligand (FasL)-Fas-Caspase extrinsic apoptosis pathway functions during palatal fusion in wild-type mice, but is not detectable in mice lacking Tgf-β3 (Tgf-β3−/− ) or Tgfβr2 in the MEE (K14-Cre;Tgfbr2fl/fl). Inhibition of the FasL-Fas system results in persistence of the midline epithelial seam (MES) and inhibition of caspase activity during palatal organ culture. Moreover, ectopic FasL protein induces apoptosis in MES of K14-Cre;Tgfbr2fl/fl mice. Thus, we conclude that the FasL-Fas-caspase extrinsic apoptosis pathway is regulated by the Tgf-β3 signaling cascade and is essential for palatal fusion during craniofacial development.

Mechanisms of transforming growth factor β induced cell cycle arrest in palate development

Immaculate and complete palatal seam disintegration, which takes place at the last phase of palate development, is essential for normal palate development. And in absence of palatal midline epithelial seam (MES) disintegration, cleft palate may arise. It has been established that transforming growth factor (TGF) β induces both epithelial mesenchymal transition (EMT) and/or apoptosis during MES disintegration. It is likely that MES might cease cell cycle to facilitate cellular changes prior to undergoing transformation or apoptosis, which has never been studied before. This study was designed to explore whether TGFβ, which is crucial for palatal MES disintegration, is capable of inducing cell cycle arrest. We studied the effects of TGFβ1 and TGFβ3, potent negative regulators of the cell cycle, on p15ink4b activity in MES cells. We surprisingly found that TGFβ1, but not TGFβ3, plays a major role in activation of the p15ink4b gene. In contrast, following successful cell cycle arrest by TGFβ1, it is TGFβ3 but not TGFβ1 that causes later cellular morphogenesis, such as EMT and apoptosis. Since TGFβ signaling activates Smads, we analyzed the roles of three Smad binding elements (SBEs) on the p15ink4b mouse promoter by site specific mutagenesis and found that these binding sites are functional. The ChIP assay demonstrated that TGFβ1, not TGFβ3, promotes Smad4 binding to two 5' terminal SBEs but not the 3' terminal site. Thus, TGFβ1 and TGFβ3 play separate yet complimentary roles in achieving cell cycle arrest and EMT/apoptosis and cell cycle arrest is a prerequisite for later cellular changes.

Development of secondary palate requires strict regulation of ECM remodeling: sequential distribution of RECK, MMP-2, MMP-3, and MMP-9

We have evaluated RECK (reversion-inducing-cysteine-rich protein with Kazal motifs), MMP-2 (matrix metalloproteinase-2), MMP-3, and MMP-9 involvement during palate development in mice by using various techniques. Immunohistochemical features revealed the distribution of RECK, MMP-2, and MMP-3 in the mesenchymal tissue and in the midline epithelial seam at embryonic day 13 (E13), MMPs-2, -3, and -9 being particularly expressed at E14 and E14.5. In contrast, RECK was weakly immunostained at these times. Involvement of MMPs was validated by measuring not only their protein expression, but also their activity (zymograms). In situ hybridization signal (ISH) for RECK transcript was distributed in mesenchymal and epithelial regions within palatal shelves at all periods evaluated. Importantly, the results from ISH analysis were in accord with those obtained by real-time polymerase chain reaction. The expression of RECK was found to be temporally regulated, which suggested possible roles in palatal ontogeny. Taken together, our results clearly show that remodeling of the extracellular matrix is finely modulated during secondary palate development and occurs in a sequential manner.

False positive reactivity of a substance P-antibody in the ectodermal/epithelial plug of the nose, ear, eye and perineum of the human and mouse fetuses

Epithelial/ectodermal plug formation in the developing nose, ear, and eye regions is followed by canalization/recanalization mediated by cell death. However, the mechanism is not well understood. Recently, substance P (SP)-mediated cell death, rather than cell apoptosis, has been reported in neuronal and non-neuronal cells. Horizontal paraffin sections of 5 human fetuses at 15-16 weeks of gestation were used to examine the entire area of the nose, ear, eye and perineum with immunohistochemistry for SP and its receptor neurokinin-1 (NK-1), and protein gene product (PGP) 9.5 and S100 protein to identify whether the positive cells had neural origins. The deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end-labeling (TUNEL) method was also conducted to identify apoptosis. Four SP antibodies were commercially obtained and compared the results. In addition, using the same antibodies for SP, those results were compared with fetal mouse heads (E14-17). Substance P immunoreactivity of one of the 4 antibodies (sc9758) was clearly found in the nasal plug, the epithelium of the anterior nasal cavity, the entire excretory tear duct, the marginal palpebral conjunctiva, the auditory meatal plug, the parotid duct, the external urethral orifice and, the preputial lamella along the future prepuce. Immunoreactivity was usually seen in enlarged round cells in humans. In fetal mouse heads, in spite of negative reaction in all these sites, the midline epithelial seam at the palate fusion and the oral epithelium especially at and near the tooth germ specifically reacted with the sc9758. Nevertheless, the other 3 antibodies did not react at any of those sites both in human and mouse fetuses. NK-1 receptor-positive cells were seen in the nose and meatal plugs and preputial lamella, but not in the tear duct. S100 protein, PGP 9.5, and TUNEL method all demonstrated negative reactivity at any sc9758-positive sites. Consequently, the present immunoreactivity of the sc9758 antibody seemed to be false positive, but it was likely to react with a specific substance in the epithelial or ectodermal cell because of the clearly restricted staining. Which substance it crossed remains unknown.

Regulation of Epithelial-Mesenchymal Transition in Palatal Fusion

During palatal fusion, the midline epithelial seam between the palatal shelves degrades to achieve mesenchymal confluence. Morphological and molecular evidence support the theory that the epithelial-mesenchymal transition is one mechanism that regulates palatal fusion. It appears that transforming growth factor (TGF)-beta signaling plays a role in palatal EMT. TGFbeta3 is the main inducer in palatal fusion and activates both Smad-dependent and -independent signaling pathways, including the key EMT transcription factors, Lef1, Twist, and Snail1, in the MEE prior to the palatal EMT program. The roles and interactions among these transcription factors will be discussed.

Interactions between TGF- β1 and TGF- β3 and their role in medial edge epithelium cell death and palatal fusion in vitro

In recent decades, studies have shown that both TGF- β 1 and TGF- β 3 play an important role in the induction of medial edge epithelium (MEE) cell death and palatal fusion. Many of these experiments involved the addition or blockage of one of these growth factors in wild-type (WT) mouse palate cultures, where both TGF- β 1 and TGF- β 3 are present. Few studies have addressed the existence of interactions between TGF- β 1 and TGF- β 3, which could modify their individual roles in MEE cell death during palatal fusion. We carried out several experiments to test this possibility, and to investigate how this could influence TGF- β 1 and TGF- β 3 actions on MEE cell death and palatal shelf fusion. We double-immunolabelled developing mouse palates with anti-TGF- β 1 or anti-TGF- β 3 antibodies and TUNEL, added rhTGF- β 1 or rhTGF- β 3 or blocked the TGF- β 1 and TGF- β 3 action at different concentrations to WT or Tgf-β 3 null mutant palate cultures, performed in situ hybridizations with Tgf-β 1 or Tgf-β 3 riboprobes, and measured the presence of TUNEL-positive midline epithelial seam (MES) cells and MES disappearance (palatal shelf fusion) in the different in vitro conditions. By combining all these experiments, we demonstrate great interaction between TGF- β 1 and TGF- β 3 in the developing palate and confirm that TGF- β 3 has a more active role in MES cell death than TGF- β 1, although both are major inductors of MES disappearance. Finally, the co-localization of TGF- β 1, but not TGF- β 3, with TUNEL in the MES allows us to suggest a possible role for TGF- β 1 in MES apoptotic clearance.

Wnt11/ Fgfr1b cross-talk modulates the fate of cells in palate development

Various cellular and molecular events underlie the elevation and fusion of the developing palate that occurs during embryonic development. This includes convergent extension, where the medial edge epithelium is intercalated into the midline epithelial seam. We examined the expression patterns of Wnt11 and Fgfr1b – which are believed to be key factors in convergent extension – in mouse palate development. Wnt-11 overexpression and beads soaked in SU5402 (an Fgfr1 inhibitor) were employed in in vitro organ cultures. The results suggested that interactions between Wnt11 and Fgfr1b are important in modulating cellular events such as cell proliferation for growth and apoptosis for fusion. Moreover, the Wnt11 siRNA results showed that Wnt11-induced apoptosis was necessary for palatal fusion. In summary, Fgfr1b induces cell proliferation in the developing palate mesenchyme so that the palate grows and contacts each palatal shelf, with negative feedback of Fgfs triggered by excessive cell proliferation then inhibiting the expression of Fgfr1b and activating the expression of Wnt11 to fuse each palate by activating apoptosis.

Fate-mapping of the epithelial seam during palatal fusion rules out epithelial–mesenchymal transformation

During palatogenesis, fusion of the palatine shelves is a crucial event, the failure of which results in the birth defect, cleft palate. The fate of the midline epithelial seam (MES), which develops transiently upon contact of the two palatine shelves, is still strongly debated. Three major mechanisms underlying the regression of the MES upon palatal fusion have been proposed: (1) apoptosis has been evidenced by morphological and molecular criteria; (2) epithelial–mesenchymal transformation has been suggested based on ultrastructural and lipophilic dye cell labeling observations; and (3) migration of MES cells toward the oral and nasal areas has been proposed following lipophilic dye cell labeling. To verify whether epithelial–mesenchymal transformation of MES cells takes place during murine palatal fusion, we used the Cre/lox system to genetically mark Sonic hedgehog- and Keratin-14-expressing palatal epithelial cells and to identify their fate in vivo. Our analyses provide conclusive evidence that rules out the occurrence of epithelial–mesenchymal transformation of MES cells.

IN UTERO PREDNISONE EXPOSURE AFFECTS GENITAL DEVELOPMENT

The etiology of the worldwide incidence of hypospadias remains unexplained. Data from the Center for Disease Control suggest that maternal asthma exposure may be a risk factor. The development of the mouse and human urethra shows significant similarities, such as fusion of 2 epithelial edges, closure of a midline epithelial seam and subsequent cellular remodeling, to justify the use of the mouse as an experimental animal model. Prednisone may act as a weak androgen directly or alternatively lowering testosterone by hypothalamic suppression of gonadotropins. We describe the effects of prednisone on urethral formation in the mouse. The model comprised 10, 20 and 20 timed pregnant C57/6 mice exposed to 1,000, 200 and 100 mg/kg prednisone, respectively, on gestational days 12 through 18. The morphology of the genital tubercles of both sexes were examined on gestational day 19 using histological techniques and 3-dimensional computer reconstruction. Specific attention was focused on the developing urethral seam. Microscopic serial analysis confirmed the presence of an arrest in seam formation in approximately 25% of male fetuses given supraphysiological doses of prednisone (1,000 mg/kg). In contrast, acceleration of urethral fold fusion and a longer urethral tube were observed in those males treated with 100 and 200 mg/kg prednisone. The female fetuses treated with the same dosages of prednisone did not show any change at the level of the urethral seam area or in the remainder of the genital tubercles. Supraphysiological doses of prednisone treatment has an inhibitory effect on urethral fold fusion leading to hypospadias whereas low doses accentuate urethra formation in male mouse fetuses. In mice lower doses do not cause hypospadias in the males or affect urethral formation in females.

Gene expression patterns during palatal shelf fusion

Abstract Introduction: Cleft palate is a common disorder that occurs when there is failure of the palatal shelves to fuse during normal development. The genes mediating normal palatal fusion are largely unknown. Methods: We harvested mRNA and performed microarray (∼40,000 element) analysis of mice palatal shelves at timepoints before, during, and after palatal fusion (E13.5, E14.5, and E15.5). IHC was performed using standard techniques. Results: Replicates clustered together by time point indicating consistency between harvests. Hox-7A, a gene implicated in both mice and humans as important to cleft lip and palate formation was upregulated during palatal fusion. Several retinoic acid responsive genes such as Stra6 were also upregulated during fusion. Furthermore, we found the homeobox gene Lim-1 to be upregulated during palatal fusion. To confirm our microarray findings of differential expression of LIM-1 during palate formation, we performed immunohistochemistry. During early secondary palate formation (E12.5), as the shelves are pointing inferiorly, LIM-1 expression was maximal at the lateral wall and tip of the palate epithelia. Interestingly, minimal expression was noted in the medial palatal wall epithelia. LIM-1 expression regressed with epithelial to mesenchymal transition at the midline epithelial seam as the approaching palates fuse on E14.5. LIM-1 expression was completely abolished within the palate mesenchyme by E16.5 after completion of fusion. Conclusions: We have identified novel genes that may be involved in normal palatogenesis using large-scale microarray analysis. These data may be a valuable resource for other researchers attempting to elucidate the complex dysregulation of expression cascades involved in cleft palate formation.

Terminal differentiation of palatal medial edge epithelial cells in vitro is not necessarily dependent on palatal shelf contact and midline epithelial seam formation.

During fusion of the mammalian secondary palate, it has been suggested that palatal medial edge epithelial (MEE) cells disappear by means of apoptosis, epithelial-mesenchymal transformation (EMT) and epithelial cell migration. However, it is widely believed that MEE cells never differentiate unless palatal shelves make contact and the midline epithelial seam is formed. In order to clarify the potential of MEE cells to differentiate, we cultured single (unpaired) palatal shelves of ICR mouse fetuses by using suspension and static culture methods with two kinds of gas-mixtures. We thereby found that MEE cells can disappear throughout the medial edge even without contact and adhesion to the opposing MEE in suspension culture with 95% O2/5% CO2. Careful examination of MEE cell behavior in the culture revealed that apoptosis, EMT, and epithelial cell migration all occurred at various stages of MEE cell disappearance, including the transient formation and disappearance of epithelial triangles and islets. In contrast, MEE cells showed poor differentiation in static culture in a CO2 incubator. Furthermore, mouse and human amniotic fluids were found to prevent MEE cell differentiation in the cultured single palatal shelf, although paired palatal shelves fused successfully even in the presence of amniotic fluid. We therefore conclude that terminal differentiation of MEE cells is not necessarily dependent on palatal shelf contact and midline epithelial seam formation, but such MEE cell differentiation appears to be prevented in utero by amniotic fluid unless palatal shelves make close contact and the midline epithelial seam is formed.