The mammalian secondary palate develops initially as two bilateral shelves that grow vertically down the sides of the developing tongue. At a precise developmental time, the bilateral palatal shelves elevate to a horizontal position above the dorsum of the tongue and fuse with each other at the midline to form the intact secondary palate. Any disturbance of the growth, elevation, and fusion of the palatal shelves causes cleft palate, which occurs frequently and affects approximately 1 in 1,500 births in humans. The molecular mechanisms underlying normal palate development and cleft palate pathogenesis are not well understood. Recently, mutations in the TBX22 gene have been identified in patients with X-linked cleft palate and ankyloglossia (Braybrook et al., 2001). TBX22 is a new member of an evolutionarily conserved gene family in which each gene encodes a transcription factor containing a conserved DNA binding domain, termed T-box (Papaioannou and Silver, 1998; Papaioannou, 2001). Mutations in three other members of the T-box gene family have been associated with human developmental disorders, the ulnar-mammary syndrome (TBX3), Holt-Oram syndrome (TBX5), and isolated adrenocorticotropic hormone deficiency (TBX19) (Bamshad et al., 1997; Basson et al., 1997; Lamolet et al., 2001). In addition, the TBX1 gene is deleted in the majority of the velo-cardio-facial/DiGeorge syndrome patients, and mice carrying targeted mutations in the Tbx1 gene show developmental abnormalities resembling defects in DiGeorge syndrome patients (Chieffo et al., 1997; Jerome and Papaioannou, 2001; Lindsay et al., 2001; Merscher et al., 2001). Many DiGeorge syndrome patients have a cleft palate phenotype, and mice homozygous for a targeted mutation in Tbx1 have cleft palate (Jerome and Papaioannou, 2001). Interestingly, recent studies show that Tbx1 is only expressed in the mesodermal core and the endoderm of the developing branchial arches but not in the developing neural crest derivatives (Garg et al., 2001; Vitelli et al., 2002), suggesting that the cleft palate phenotype of Tbx1 mutant mice is likely to be a secondary effect. It is not known whether TBX22 is expressed during palate development. Because the majority of the male patients with TBX22 mutations exhibit both cleft palate and ankyloglossia, it has been speculated that the cleft palate phenotype may be secondary to defective fetal tongue movement (Gorski et al., 1992; Stanier et al., 1993). To understand the role of TBX22 in disease pathogenesis and in normal development, it is necessary to carry out a detailed temporal and spatial expression analysis of the TBX22 gene. Of the only two reports on the TBX22 gene to date, one failed to detect TBX22 mRNA expression in any fetal and adult tissue, whereas the other showed a reverse transcriptase-polymerase chain reaction (RT-PCR) product of expected size in every fetal tissue examined, which did not include the palatal tissue (Laugier-Anfossi and Villard, 2000; Braybrook et al., 2001). To better characterize the developmental role of TBX22, we have isolated the mouse Tbx22 gene and analyzed its expression pattern during mouse embryonic development.
RESULTS AND DISCUSSION
We isolated the mouse Tbx22 cDNA through a combination of database mining and RT-PCR cloning (see Experimental Procedures section). The mouse Tbx22 gene encodes a protein of 517 amino acid residues that shares 72% overall amino acid sequence identity with the human TBX22 protein (Fig. 1). Because a previous study failed to detect the existence of a mouse homolog of TBX22 by using Southern and Northern hybridization methods (Laugier-Anfossi and Villard, 2000), we mapped the mouse Tbx22 gene by using the Jackson Laboratory BSS interspecific backcross panel to confirm orthology. The Tbx22 gene is localized to the central part of mouse chromosome X and cosegregates with the markers DXMit65 and DXMit214. This location is in a region of known synteny to human chromosome Xq21, where TBX22 resides (Oeltjen et al., 1997; Phippard et al., 2000; Braybrook et al., 2001). The chromosomal localization, together with the high degree of sequence identity in and outside of the T-box region (Fig. 1), indicates that Tbx22 is the mouse ortholog of human TBX22.
We analyzed Tbx22 gene expression during mouse embryonic development by using in situ hybridization of both whole-mount and paraffin sections. Tbx22 mRNA expression is first detected at embryonic day (E) 9 during mouse embryogenesis in the developing somites (Fig. 2A,B). At E10.5, in addition to the somites, Tbx22 mRNA is detected in the frontonasal processes and mandibular processes of the first branchial arches (Fig. 2C). In situ hybridization of paraffin sections of E10.5 and E11.5 embryos shows that the mandibular Tbx22 expression domain is in the primordial tissue of the developing tongue (Fig. 2C,D). Tbx22 expression in embryos older than E11.5 is only detected in the craniofacial regions by in situ hybridization of both whole-mount and sections. At E11.5, Tbx22 is expressed in the developing tongue primordia as well as in the oral sides of the maxillary and nasal processes (Fig. 3A,B). By E12.0, the palatal shelves have emerged from the oral sides of the maxillary processes and started growing down the sides of the developing tongue. Tbx22 expression in the maxillary processes becomes restricted to the palatal shelves (Fig. 3C). Tbx22 expression in the mandibular processes is detected at the base of the developing tongue, in the mesenchymal precursors of Meckel's cartilage, as well as in two subsets of lateral mesenchymal cells (Fig. 3C,D). High levels of Tbx22 expression persist in mesenchyme of the palatal shelves and at the base of the tongue through E13.5 (Fig. 3E–G). Tbx22 expression is also detected in the mesenchyme surrounding the developing eyes at E13.5 (Fig. 3H). By E14.5, the palatal shelves have elevated to a horizontal position above the dorsum of the tongue and fused with each other at the midline. At this stage, Tbx22 expression is turned off in the palatal shelves and is significantly down-regulated in the mesenchyme at the base of the tongue, whereas high levels of Tbx22 mRNA persist in the mesenchyme surrounding the developing eyes (Fig. 3I). The temporally and spatially highly restricted pattern of expression during palate and tongue development, together with the cleft palate and ankyloglossia phenotypes in patients with mutations in the TBX22 gene, indicate a primary role for Tbx22 in both palate and tongue development.
Isolation of Tbx22
Searching the mouse Expressed Sequence Tags (EST) database (www.ncbi.nlm.nih.gov) by using the human TBX22 protein sequence identified four mouse ESTs with high sequence similarities (Genbank accession nos. BB606355, BB614519, BB657920, and BB664980). Comparison of the assembled cDNA sequence with the human genomic and mouse genomic trace sequences shows that the assembled Tbx22 cDNA sequence is missing the 3′ part of the last coding exon (exon 8). RT-PCR using a 5′ primer corresponding to sequences in exon 5 and a 3′ primer corresponding to sequences downstream of the putative inframe stop codon amplified a product of approximately 650 bp from E12.5 mouse head RNA. Sequencing of the RT-PCR product confirmed the predicted open reading frame. Sequence analyses were performed by using MacVector (Oxford Molecular, Ltd.).
A microsatellite sequence containing 24 (GA) repeats was identified in the genomic sequence of the last exon of the Tbx22 gene. PCR primers (primer 1, 5′-GACATGCTATCAGTGATAATTGAGG-3′ and primer 2, 5′-GCCAATCTTCATAGATAAGAGTG-3′) flanking the (GA) repeat region were tested for polymorphisms between C57BL/6J and SPRET/Ei mouse strains and were found to amplify unique products of 213 bp and approximately 190 bp, respectively, from the two strains. This microsatellite polymorphism was used to map the chromosomal location of the Tbx22 gene by using the Jackson Laboratory BSS interspecific backcross panel, which consists of genomic DNA from 94 N2 animals [(C57BL/6Jei × SPRET/Ei)F1 × SPRET/Ei]. No recombinants were detected between Tbx22 and the markers DXMit65 and DXMit214, which places Tbx22 at approximately 49 centimorgans from the centromere on mouse chromosome X. Raw mapping data for this gene as well as thousands of markers typed by using the Jackson Laboratory BSS backcross panel are available on the World Wide Web (www.jax.org/resources/documents/cmdata).
In Situ Hybridization
In situ hybridization of whole-mount and sectioned embryos was carried out as previously described (Jiang et al., 1998; Lan et al., 2001). Digoxigenin- and 33P-labeled antisense RNA probes were made by using the subcloned 650-bp cDNA fragment as described above.
We thank Paul Kingsley for critical reading of the manuscript and discussions. This work was supported by a NIH grant (DE13681) to R.J. J.O.B. was supported by a NIH training grant (J32 DE07202).