Abstract

Axenfeld–Rieger syndrome (ARS) patients with PITX2 point mutations exhibit a wide range of clinical features including mild craniofacial dysmorphism and dental anomalies. Identifying new PITX2 targets and transcriptional mechanisms are important to understand the molecular basis of these anomalies. Chromatin immunoprecipitation assays demonstrate PITX2 binding to the FoxJ1 promoter and PITX2C transgenic mouse fibroblasts and PITX2-transfected cells have increased endogenous FoxJ1 expression. FoxJ1 is expressed at embryonic day 14.5 (E14.5) in early tooth germs, then down-regulated from E15.5–E17.5 and re-expressed in the inner enamel epithelium, oral epithelium, tongue epithelium, sub-mandibular salivary gland and hair follicles during E18.5 and neonate day 1. FoxJ1 and Pitx2 exhibit overlapping expression patterns in the dental and oral epithelium. PITX2 activates the FoxJ1 promoter and, Lef-1 and β-catenin interact with PITX2 to synergistically regulate the FoxJ1 promoter. FoxJ1 physically interacts with the PITX2 homeodomain to synergistically regulate FoxJ1, providing a positive feedback mechanism for FoxJ1 expression. Furthermore, FoxJ1, PITX2, Lef-1 and β-catenin act in concert to activate the FoxJ1 promoter. The PITX2 T68P ARS mutant protein physically interacts with FoxJ1; however, it cannot activate the FoxJ1 promoter. These data indicate a mechanism for the activity of the ARS mutant proteins in specific cell types and provides a basis for craniofacial/ tooth anomalies observed in these patients. These data reveal novel transcriptional mechanisms of FoxJ1 and demonstrate a new role of FoxJ1 in oro-facial morphogenesis.

INTRODUCTION

Axenfeld–Rieger syndrome (ARS) is an autosomal dominant disorder discovered in 1935 (1). This syndrome presents an array of clinical features characterized by ocular anterior chamber anomalies, umbilical stump abnormalities, mild craniofacial dysmorphism and dental defects. Manifestations of dental defects include abnormally small teeth (microdontia), missing teeth (anodontia), and/or malformed teeth. Initial reports identified the first linkage of ARS to a locus on chromosome 4q25; subsequently point mutations discovered in the PITX2 gene provided the genetic basis of ARS (2). Since then PITX2 has provided insights into tooth development and morphogenesis.

PITX2, a paired-like homeobox transcription factor is the earliest transcription marker of tooth development and is specifically expressed in the dental epithelium (DE). It is also expressed in the brain, heart, pituitary, eye and umbilicus (2–4). PITX2A, PITX2B and PITX2C isoforms differ in the amino terminus and differentially regulate transcription (5). The Pitx family of genes have been implicated in human development, disease and evolution (2,3,6,7). Pitx2 also plays a central role in determining left–right asymmetry in vertebrates (4,8–10). However, unraveling the molecular mechanisms governing tooth morphogenesis and identifying new targets of PITX2 is mandatory to better understand the genetic basis of dental anomalies in ARS.

FoxJ1, also known as hepatocyte nuclear factor-3/forkhead homologue-4, was originally identified in a rat cDNA library screen of winged-helix/forkhead family of transcription factors (11) and contains a conserved 100 amino acid DNA binding motif. It plays a fundamental role in embryonic development and differentiation (12). FoxJ1 is expressed in upper and lower airway ciliated cells, choroids plexus, ependyma, testis, oviducts and embryonic node (13–16). In the immune system T- and B-cell activities are regulated by FoxJ1 (17,18). FoxJ1 plays a central role in determining left–right asymmetry (19). The wide range of significant roles played by FoxJ1 during development underscores the importance of identifying the regulatory factors, interacting factors and functional characterization of this gene.

Tooth morphogenesis is an excellent model to study molecular mechanisms involved in organogenesis. This process involves sequential and reciprocal interaction between the oral ectoderm and neural crest-derived mesenchyme. Tooth development processes take place through a series of stages starting with the initiation stage followed by bud, cap and bell stages with each stage showing distinct histo-morphometric changes under tight genetic control. In addition to PITX2, there are several other transcription factors including Lef-1 and β-catenin that are shown to be expressed and play crucial roles during different stages of tooth morphogenesis.

The role of Lef-1 in the development of teeth, whiskers and the mammary gland was identified through targeted inactivation of the Lef-1 gene in mice (20). β-catenin is expressed during the same time as Lef-1 in the developing tooth bud and similarly to Pitx2, (21). These transcription factors act in concert to regulate developmental processes. Lef-1 and β-catenin interact independently with PITX2 and synergistically regulate Lef-1 isoform expression (22,23). We also tested the hypothesis of how this complex of transcription factors would regulate FoxJ1 expression during tooth morphogenesis.

Bioinformatics and sequence analysis revealed several consensus PITX2 elements (5′-TAATCC-3′) in the FoxJ1 promoter. Chromatin immunoprecipitation (ChIP) assays identified FoxJ1 as a direct downstream target of PITX2. FoxJ1 is expressed in the oro-facial tissues during embryonic day 14.5 (E14.5), down-regulated from E16.5 to E17.5 and re-expressed at E18.5 and neonate day 1 (P1). We demonstrate that FoxJ1 physically interacts with PITX2 to synergistically activate the FoxJ1 promoter. Lef-1 and β-catenin interact together with PITX2 to regulate the FoxJ1 promoter. FoxJ1, PITX2, Lef-1 and β-catenin act in concert to activate the FoxJ1 promoter. Because PITX2 mutants and haploinsufficiency are causative of ARS, decreased FoxJ1 expression may play a role in the developmental anomalies associated with ARS.

RESULTS

FoxJ1 expression in maxilla, mandible, tongue and tooth

To identify FoxJ1 expression in the oro-facial region we excised the maxilla, mandible and tongue of an E14.5 mouse embryo and molar tooth germ from P1 pups. These tissues were lysed and resolved on a 10% sodium dodecyl sulfate (SDS)–polyacrylamide gel to probe for endogenous FoxJ1 expression using anti-FoxJ1 antibody. FoxJ1 pure protein was used as a positive Control. The FoxJ1 doublet band was seen in all tissues as previously reported (Fig. 1A) (24). The FoxJ1 protein has two consensus tyrosine kinase sites in the forkhead domain and preliminary experiments suggest that the doublet bands result from phosphorylation (unpublished data). FoxJ1 endogenous expression was also detected in LS-8 (oral epithelial cells) and Chinese Hamster ovary (CHO) cells used for transient transfection assays, however the expression was relatively higher in the LS-8 cell line (Fig. 1B).

Figure 1.

Endogenous FoxJ1 expression: (A) the mandible, maxilla, tongue of developing mouse embryo at E14.5 and the molar tooth germ at P1 were carefully excised. The excised tissues were lysed and resolved on 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was performed using FoxJ1 antibody; (B) LS-8 is a mouse oral epithelial cell line and CHO is a Chinese Hamster ovary cell line. LS-8 and CHO cell lysates were prepared. Cell lysates were resolved on a 10% polyacrylamide gel and western blot probed for endogenous FoxJ1 expression using FoxJ1 antibody.

Figure 1.

Endogenous FoxJ1 expression: (A) the mandible, maxilla, tongue of developing mouse embryo at E14.5 and the molar tooth germ at P1 were carefully excised. The excised tissues were lysed and resolved on 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was performed using FoxJ1 antibody; (B) LS-8 is a mouse oral epithelial cell line and CHO is a Chinese Hamster ovary cell line. LS-8 and CHO cell lysates were prepared. Cell lysates were resolved on a 10% polyacrylamide gel and western blot probed for endogenous FoxJ1 expression using FoxJ1 antibody.

FoxJ1 expression in oro-facial tissues

Because FoxJ1 expression was observed in excised oro-facial tissues, we performed immunohistochemistry (IHC) experiments to stage FoxJ1 expression in these tissues. E14.5, 16.5, 17.5, 18.5 and P1 mouse tissues were used for IHC and IHC was performed with and without FoxJ1 antibody.

E14.5 sagital sections reveal FoxJ1 expression in the DE of both incisor and molar tooth buds (Fig. 2A–D). These data correlate with the Western blot data demonstrating FoxJ1 expression in the oro-facial region at E14.5. Pitx2 is expressed at the same stage and also confined to the DE (tooth bud is outlined) (4,25).

Figure 2.

Staging of FoxJ1 expression in the developing oro-facial/tooth structures: (A and B) E14.5 FoxJ1 staining of mouse incisors; (C and D) FoxJ1 expression in E14.5 molars, the tooth bud is outlined by the dotted line; (E and G) E16.5 and 17.5 mouse molar negative controls respectively; (F and H) neighboring sections of E16.5 and 17.5 molar tooth germs respectively, that are FoxJ1 antibody treated and detected by 3-3′-diaminobenzidine staining. FoxJ1 expression is not detected at these time points during embryogenesis (IEE, inner enamel epithelium; DM, dental mesenchyme; DE, dental epithelium).

Figure 2.

Staging of FoxJ1 expression in the developing oro-facial/tooth structures: (A and B) E14.5 FoxJ1 staining of mouse incisors; (C and D) FoxJ1 expression in E14.5 molars, the tooth bud is outlined by the dotted line; (E and G) E16.5 and 17.5 mouse molar negative controls respectively; (F and H) neighboring sections of E16.5 and 17.5 molar tooth germs respectively, that are FoxJ1 antibody treated and detected by 3-3′-diaminobenzidine staining. FoxJ1 expression is not detected at these time points during embryogenesis (IEE, inner enamel epithelium; DM, dental mesenchyme; DE, dental epithelium).

E16.5 and E17.5 sagital sections show bell stage molar tooth germ with the inner enamel epithelium (IEE) encircling the neural crest-derived ectomesenchyme (dental mesenchyme, DM) (Fig. 2F,H). At these stages FoxJ1 expression was not detected in the IEE or in the mesenchyme (Fig. 2F,H). FoxJ1 expression was observed in the nasal epithelium demonstrating that our IHC staining was working in these sections. Surprisingly, FoxJ1 expression was down-regulated at E16.5–17.5.

However, E18.5 late bell stage molar tooth germ demonstrated positive nuclear staining for FoxJ1 in the IEE (Fig. 3B,C). In P1 incisor and molar tooth germs, the IEE has differentiated into enamel secreting pre-ameloblasts and the mesenchyme differentiated into dentin secreting odontoblasts (Fig. 3D–I). At this stage FoxJ1 is expressed in both pre-ameloblasts and odontoblasts (Fig. 3F,I). FoxJ1 expression in the DM was not detected at E18.5, but the mesenchyme-derived odontoblasts showed positive expression at P1. To further confirm FoxJ1 expression in the odontoblasts we repeated the experiment on P1 molar tooth germ without a hematoxylin counter stain. We observed FoxJ1 expression in pre-ameloblasts as well as in the odontoblasts (Fig. 3K). In the dental and oral epithelium, Pitx2, Lef-1 and β-catenin have overlapping expression patterns with FoxJ1 (16,23). We demonstrate these overlapping expression patterns in P1 molar sections. Pitx2, Lef-1 and FoxJ1 are all expressed in the DE, whereas β-catenin is expressed in both the DE and mesenchyme (Fig. 3L–O). The oral and tongue epithelium revealed positive FoxJ1 expression at this stage (Fig. 3R,S,U). The expression pattern in the oral epithelium and tongue epithelium was similar at E18.5 (Fig. 3P,Q,T). Thus, FoxJ1 expression occurs during early and late stages of odontogenesis as well as in the oral and tongue epithelium. In addition to dental, oral, and tongue epithelium, FoxJ1 is also expressed in the sub-mandibular salivary gland and hair follicles (Fig. 4B,D).

Figure 3.

Staging of FoxJ1 expression in the developing oro-facial/tooth structures: (A, D and G) E18.5 mouse molar tooth germ, P1 molar tooth germ and P1 incisor tooth germ, negative Control, respectively; (B,E and H) E18.5 and P1 mouse molar and incisor FoxJ1 expression, respectively; (C,F and I) higher magnification of E18.5 and P1, showing expression in the IEE, PAB and OD, respectively; (J and K) P1 molar tooth germ negative Control and FoxJ1 antibody treated, respectively, without counter staining; (L-O) P1 molar sections demonstrating overlapping expression of endogenous Pitx2, Lef-1, β-catenin and FoxJ1, respectively; (P,Q and T) E18.5 proximal and distal oral epithelium, and tongue epithelium, respectively, that are FoxJ1 antibody treated; (R,S,U) P1 proximal and distal oral epithelium and tongue epithelium, respectively, that are FoxJ1 antibody treated (IEE, inner enamel epithelium; PAB, pre-ameloblasts; DM, dental mesenchyme; OD, odontoblasts; OE, oral epithelium; TE, tongue epithelium).

Figure 3.

Staging of FoxJ1 expression in the developing oro-facial/tooth structures: (A, D and G) E18.5 mouse molar tooth germ, P1 molar tooth germ and P1 incisor tooth germ, negative Control, respectively; (B,E and H) E18.5 and P1 mouse molar and incisor FoxJ1 expression, respectively; (C,F and I) higher magnification of E18.5 and P1, showing expression in the IEE, PAB and OD, respectively; (J and K) P1 molar tooth germ negative Control and FoxJ1 antibody treated, respectively, without counter staining; (L-O) P1 molar sections demonstrating overlapping expression of endogenous Pitx2, Lef-1, β-catenin and FoxJ1, respectively; (P,Q and T) E18.5 proximal and distal oral epithelium, and tongue epithelium, respectively, that are FoxJ1 antibody treated; (R,S,U) P1 proximal and distal oral epithelium and tongue epithelium, respectively, that are FoxJ1 antibody treated (IEE, inner enamel epithelium; PAB, pre-ameloblasts; DM, dental mesenchyme; OD, odontoblasts; OE, oral epithelium; TE, tongue epithelium).

Figure 4.

FoxJ1 expression in the sub-mandibular salivary gland and hair follicle: (A and C) P1 mouse sub-mandibular salivary gland and hair follicle negative control, respectively; (B and D) neighboring sections of sub-submandibular salivary gland and hair follicle, respectively, that are FoxJ1 antibody treated and detected by 3-3′-diaminobenzidine staining.

Figure 4.

FoxJ1 expression in the sub-mandibular salivary gland and hair follicle: (A and C) P1 mouse sub-mandibular salivary gland and hair follicle negative control, respectively; (B and D) neighboring sections of sub-submandibular salivary gland and hair follicle, respectively, that are FoxJ1 antibody treated and detected by 3-3′-diaminobenzidine staining.

PITX2 binds to the FoxJ1 promoter in vivo

Because sequence analysis revealed several cis-acting elements in the FoxJ1 promoter corresponding to PITX2 binding sites we asked if Pitx2 endogenously bound to the FoxJ1 promoter (Fig. 5A). ChIP assays were done in LS-8 cells to demonstrate Pitx2 binding to FoxJ1 promoter because the LS-8 cell line endogenously expresses Pitx2 and FoxJ1. PITX2 antibody was used to immunoprecipitate the Pitx2/chromatin complex. A sense and anti-sense primer, flanking the Pitx2 binding site located at −1046 to −1041 were designed and produced a 262 bp product (Fig. 5A). The sense primers anneal to the sequence located at −1126 to −1102 and anti-sense primers anneal to the sequence located at −865 to −889 of the FoxJ1 promoter. A polymerase chain reaction (PCR) reaction was performed with chromatin input derived from LS-8 cells and chromatin complex immunoprecipitated by Pitx2 antibody. The primer set amplified the FoxJ1 promoter from chromatin input, as well as from the Pitx2/chromatin complex demonstrating Pitx2 specifically bound to the TAATCC element in the FoxJ1 promoter (Fig. 5B, lanes 4 and 5). A PCR reaction was also performed without chromatin and primers only as a negative control (Fig. 5B, lane 2). Normal mouse immunoglobulin G (IgG) was used as a control and did not immunoprecipitate the FoxJ1 promoter (Fig. 5B, lane 3). DNA markers were used (PCR marker, Promega) to confirm sizes of the PCR product (Fig. 5B, lane 1). All PCR products were sequenced to confirm their identity.

Figure 5.

Pitx2 endogenously binds to FoxJ1 promoter: (A) Schematic representation of FoxJ1 2.6 kb promoter with three Pitx2 binding sites (TAATCC) noted by ‘plus’ marks. The location of the sense primer (S) and the anti-sense primer (AS) are shown in the blowup of the −1126 bp and −865 bp region of the distal promoter used to amplify the immunoprecipitated chromatin; (B) ChIP assay was performed in LS-8 cells. Lane 1 contains the PCR marker and lane 2 is FoxJ1 primers only control. Lane 3 is the immunoprecipitation using normal mouse immunoglobulin G and FoxJ1 primers and lane 4 is the Pitx2 immunoprecipitated chromatin amplified using the specific FoxJ1 promoter primers and produced the correct size product of 262 bp. Lane 5 is chromatin input amplified using the FoxJ1 primers.

Figure 5.

Pitx2 endogenously binds to FoxJ1 promoter: (A) Schematic representation of FoxJ1 2.6 kb promoter with three Pitx2 binding sites (TAATCC) noted by ‘plus’ marks. The location of the sense primer (S) and the anti-sense primer (AS) are shown in the blowup of the −1126 bp and −865 bp region of the distal promoter used to amplify the immunoprecipitated chromatin; (B) ChIP assay was performed in LS-8 cells. Lane 1 contains the PCR marker and lane 2 is FoxJ1 primers only control. Lane 3 is the immunoprecipitation using normal mouse immunoglobulin G and FoxJ1 primers and lane 4 is the Pitx2 immunoprecipitated chromatin amplified using the specific FoxJ1 promoter primers and produced the correct size product of 262 bp. Lane 5 is chromatin input amplified using the FoxJ1 primers.

FoxJ1 expression is up-regulated in PITX2C transgenic mouse fibroblasts

PITX2C was cloned into a K14 promoter–green fluorescent protein (GFP) construct to make transgenic mice over-expressing PITX2C in the oral and DE as well as in the skin. Mouse fibroblasts were prepared from P1 wild-type and transgenic mice. PITX2C and GFP expression were confirmed in the transgenic fibroblasts and protein lysates were prepared for western blot. FoxJ1 is endogenously expressed in the mouse fibroblasts and increased expression was observed in the PITX2C transgenic cells (Fig. 6A). Furthermore, we observed an increase in FoxJ1 protein in PITX2-transfected CHO cells (Fig. 6B). Endogenous FoxJ1 is expressed at low levels in CHO cells (Fig. 1 and mock Fig. 6B), transfection of PITX2 increases endogenous FoxJ1 to levels seen when FoxJ1 plasmid was transfected into these cells. These data demonstrate that PITX2 specifically activates endogenous FoxJ1 expression.

Figure 6.

PITX2 activates endogenous FoxJ1 expression: (A) western blot of FoxJ1 protein from wild-type mouse and PITX2C transgenic mouse fibroblast lysates. Cells were lysed and 15 µg of protein lysates resolved on 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was performed using FoxJ1 antibody. As a control cell lysates were probed using the glyceraldehyde 3-phosphate dehydrogenase antibody to demonstrate equal loading of the lanes; (B) CHO cells were transfected with FoxJ1, mock (empty vector), PITX2 or PITX2 and FoxJ1 and cell lysates probed for FoxJ1 expression. Transfected PITX2 cell lysates demonstrate an increase in FoxJ1 expression compared with mock transfected and similar to transfected FoxJ1 only.

Figure 6.

PITX2 activates endogenous FoxJ1 expression: (A) western blot of FoxJ1 protein from wild-type mouse and PITX2C transgenic mouse fibroblast lysates. Cells were lysed and 15 µg of protein lysates resolved on 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was performed using FoxJ1 antibody. As a control cell lysates were probed using the glyceraldehyde 3-phosphate dehydrogenase antibody to demonstrate equal loading of the lanes; (B) CHO cells were transfected with FoxJ1, mock (empty vector), PITX2 or PITX2 and FoxJ1 and cell lysates probed for FoxJ1 expression. Transfected PITX2 cell lysates demonstrate an increase in FoxJ1 expression compared with mock transfected and similar to transfected FoxJ1 only.

PITX2 isoforms activate the FoxJ1 promoter

The FoxJ1 promoter was cloned to test the hypothesis that PITX2 directly regulates FoxJ1 expression. There are three PITX2 binding sites in the FoxJ1 2.6 promoter (Fig. 7A). Transient transfections were performed in CHO cells, as they do not endogenously express Pitx2. The PITX2A, PITX2B and PITX2C expression plasmids and FoxJ1 2.6 promoter were co-transfected in CHO cells. PITX2A and PITX2C activated the FoxJ1 2.6 promoter at ∼20 and ∼18-fold respectively (Fig. 7B; P < 0.05), whereas PITX2B activated at only 5-fold (P < 0.05) and the PITX2 T68P ARS mutant protein showed no activation (Fig. 7B). This cell line was compared with LS-8 cells and the transcriptional activity is low in LS-8 cells due to high background caused by the endogenous tooth specific transcription factors activating the promoters. However the relative activity by PITX2 was similar in LS-8 cells compared with CHO cells (data not shown).

Figure 7.

PITX2 isoforms activate the FoxJ1 promoter: (A) FoxJ1 2.6 contains 2.6 kb of 5′ sequence flanking the FoxJ1 gene; (B) CHO cells were transfected with 2.5 µg of PITX2 T68P, PITX2A, PITX2B and PITX2C expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmid. All DNA was double CsCl banded for purity and cells were transfected by electroporation. To control for transfection efficiency all transfections included the SV-40 β-galactosidase reporter (0.5 µg). The activities are shown as mean-fold activation compared with the FoxJ1 2.6 promoter plasmid without expression plasmids and normalized to β-galactosidase activity [±SEM (standard error of the mean) from five independent experiments].

Figure 7.

PITX2 isoforms activate the FoxJ1 promoter: (A) FoxJ1 2.6 contains 2.6 kb of 5′ sequence flanking the FoxJ1 gene; (B) CHO cells were transfected with 2.5 µg of PITX2 T68P, PITX2A, PITX2B and PITX2C expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmid. All DNA was double CsCl banded for purity and cells were transfected by electroporation. To control for transfection efficiency all transfections included the SV-40 β-galactosidase reporter (0.5 µg). The activities are shown as mean-fold activation compared with the FoxJ1 2.6 promoter plasmid without expression plasmids and normalized to β-galactosidase activity [±SEM (standard error of the mean) from five independent experiments].

PITX2A, Lef-1 FL and β-catenin activate the FoxJ1 promoter

Because FoxJ1 exhibits overlapping expression patterns with the transcription factors PITX2, Lef-1 and β-catenin, we asked if these transcription factors regulated FoxJ1 expression. PITX2A, Lef-1 FL, Lef-1 ΔN113, β-catenin and the FoxJ1 2.6 promoter were co-transfected in CHO cells. Lef-1 ΔN113 is a short isoform and does not contain β-catenin binding domain. PITX2A activated the FoxJ1 2.6 promoter at ∼21-fold whereas β-catenin, Lef-1 FL and Lef-1 ΔN113 showed no significant activation (Fig. 8A; P < 0.05). However, co-transfection of PITX2A and Lef-1 FL, or PITX2A and Lef-1 ΔN113, synergistically activated the FoxJ1 2.6 promoter at ∼50 and 49-fold respectively (Fig. 8A; P < 0.05). PITX2A and β-catenin synergistically activate the FoxJ1 2.6 promoter at ∼38-fold (Fig. 8A; P < 0.05).

Figure 8.

PIXT2A, Lef-1 FL and β-catenin activate the FoxJ1 promoter: (A) CHO cells were transfected with 2.5 µg of PITX2A, PITX2A T68P, Lef-1 FL, Lef-1 ΔN113 and β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmid. Transfections were done as in Figure 7 [±SEM (standard error of the mean) from six independent experiments]; (B) Western blot of transfected proteins with specific antibodies visualized using enhanced chemiluminescence reagents.

Figure 8.

PIXT2A, Lef-1 FL and β-catenin activate the FoxJ1 promoter: (A) CHO cells were transfected with 2.5 µg of PITX2A, PITX2A T68P, Lef-1 FL, Lef-1 ΔN113 and β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmid. Transfections were done as in Figure 7 [±SEM (standard error of the mean) from six independent experiments]; (B) Western blot of transfected proteins with specific antibodies visualized using enhanced chemiluminescence reagents.

When we co-transfected PITX2A, Lef-1 FL and β-catenin together, the complex activated the FoxJ1 2.6 promoter at ∼64-fold (Fig. 8A; P < 0.05). PITX2A, Lef-1 ΔN113 and β-catenin co-transfection activated the promoter at ∼63-fold (Fig. 8A; P < 0.05). The difference in activation between the two groups (PITX2A, β-catenin, Lef-1 FL, and PITX2A, β-catenin, Lef-1 ΔN113) is insignificant supporting the previous studies that β-catenin and Lef-1 interact independently with PITX2A (23) (Fig. 8A).

The ARS mutant PITX2 T68P can bind DNA, but is transactivation deficient (6). This mutant contains a proline residue at position 68 within the homeodomain (HD) (26). PITX2 T68P did not activate the FoxJ1 promoter or in combination with Lef-1 isoforms or β-catenin (Fig. 8A). Our data demonstrate a role of these factors in regulating FoxJ1 expression. Western blot analyses reveal expression of PITX2, PITX2 T68P, FoxJ1, β-catenin, Lef-1 ΔN113 and Lef-1 FL proteins (Fig. 8B). β-catenin and Lef-1 isoform expression have been previously reported (23).

FoxJ1 and PITX2A synergistically activate the FoxJ1 2.6 promoter

FoxJ1 and/or PITX2A, PITX2 T68P, Lef-1 FL and β-catenin were co-transfected with FoxJ1 2.6 promoter to determine if FoxJ1 regulates the FoxJ1 promoter (Fig. 9A). FoxJ1 did not significantly activate its own promoter (Fig. 9A). However, FoxJ1 and PITX2A synergistically activated the FoxJ1 2.6 promoter at ∼75-fold, whereas FoxJ1 and PITX2A T68P showed no activation (Fig. 9A; P < 0.05). FoxJ1 transfected with Lef-1 FL or β-catenin activated the FoxJ1 promoter by ∼5 and ∼8-fold, respectively (Fig. 9A). When FoxJ1 was co-transfected with PITX2A and Lef-1 FL, PITX2A and Lef-1 ΔN113 or PITX2A and β-catenin, the synergistic activity increased to ∼190, ∼200, ∼250-fold, respectively (Fig. 9B; P < 0.05). FoxJ1 transfected with PITX2A, Lef-1 FL and β-catenin or PITX2A, Lef-1 ΔN113 and β-catenin, the synergistic activity increased by ∼300 and ∼220-fold, respectively (Fig. 9B; P < 0.05). These data suggest FoxJ1 may regulate its own promoter in concert with other specific transcription factors in a positive feedback fashion.

Figure 9.

FoxJ1 and PITX2A synergistically activate the FoxJ1 2.6 promoter: (A) 2.5 µg of FoxJ1 and/or PITX2A, PITX2 T68P, Lef-1 FL, β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids were transfected into CHO cells. Transfections were done as in Figure 6 [±SEM (standard error of the mean) from five independent experiments]; (B) CHO cells were transfected with 2.5 µg of FoxJ1, PITX2A, PITX2A T68P, Lef-1 FL, Lef-1 ΔN113 and β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids. Transfections were done as in Figure 7 (±SEM from six independent experiments).

Figure 9.

FoxJ1 and PITX2A synergistically activate the FoxJ1 2.6 promoter: (A) 2.5 µg of FoxJ1 and/or PITX2A, PITX2 T68P, Lef-1 FL, β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids were transfected into CHO cells. Transfections were done as in Figure 6 [±SEM (standard error of the mean) from five independent experiments]; (B) CHO cells were transfected with 2.5 µg of FoxJ1, PITX2A, PITX2A T68P, Lef-1 FL, Lef-1 ΔN113 and β-catenin expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids. Transfections were done as in Figure 7 (±SEM from six independent experiments).

PITX2 T68P was unable to increase activation of the FoxJ1 promoter in concert with FoxJ1 (Fig. 9B). Furthermore, co-expression of PITX2 T68P and FoxJ1 with Lef-1 isoforms, β-catenin or combinations of all factors did not activate the FoxJ1 promoter (Fig. 9B).

FoxJ1 physically interacts with PITX2

Because FoxJ1 and PITX2 synergistically activated the FoxJ1 promoter, we performed a co-immunoprecipitation assay to test if they physically interact. FoxJ1 antibody was used to immunoprecipitate the FoxJ1/PITX2 complex in transfected CHO cell lysates (Fig. 10). Mock and FoxJ1-transfected cell lysates do not immunoprecipitate a FoxJ1/PITX2 complex (lanes 1 and 2, Fig. 10). However, as CHO cells endogenously express FoxJ1, transfected PITX2 formed a complex with endogenous FoxJ1 and was immunoprecipitated by FoxJ1 antibody (lane 3, Fig. 10). To demonstrate expression of PITX2, the cell lysates were directly analyzed on a gel (Input, Fig. 10).

Figure 10.

FoxJ1 physically interacts with PITX2: FoxJ1 and/or PITX2, and PITX2 T68P expression plasmids (10 µg) were transfected in CHO cells and incubated for 24 h. Cells were harvested and lysed and FoxJ1/PITX2 protein complex was immunoprecipitated using the FoxJ1 antibody. The immunoprecipitated complex was resolved on a 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was done using PITX2 antibody. Glyceraldehyde 3-phosphate dehydrogenase was analyzed on the blot as a loading Control.

Figure 10.

FoxJ1 physically interacts with PITX2: FoxJ1 and/or PITX2, and PITX2 T68P expression plasmids (10 µg) were transfected in CHO cells and incubated for 24 h. Cells were harvested and lysed and FoxJ1/PITX2 protein complex was immunoprecipitated using the FoxJ1 antibody. The immunoprecipitated complex was resolved on a 10% polyacrylamide gel and transferred to a polyvinylidenefluride filter and western blotting was done using PITX2 antibody. Glyceraldehyde 3-phosphate dehydrogenase was analyzed on the blot as a loading Control.

The transfected PITX2 T68P protein is not immunoprecipitated by low levels of endogenous FoxJ1 expression in CHO cells (lane 4, Fig. 10). A western blot demonstrates decreased PITX2 T68P expression from the input sample (lane 4, Fig. 10). When FoxJ1 and wild-type PITX2 were co-transfected, FoxJ1 antibody immunoprecipitated the PITX2 protein (lane 5, Fig. 10). FoxJ1 co-expression with PITX2 T68P demonstrates that FoxJ1 was able to immunoprecipitate the mutant protein (lane 6, Fig. 10). PITX2 does not activate FoxJ1 expression from the transfected plasmid nor does FoxJ1 activate PITX2 expression from the PITX2-transfected plasmid (unpublished data). Previous results indicated that the PITX2 T68P protein was stable when expressed in COS-7 cells (6,27). Furthermore, FoxJ1, β-catenin, Lef-1 and PITX2 did not alter the expression of PITX2 or PITX2 T68P from its plasmid (Fig. 10, and data not shown).

FoxJ1 interacts with the PITX2 homeodomain

Two PITX2A deletion constructs were tested for FoxJ1 interactions in transfection experiments (Fig. 11A). PITX2A NΔ38 with a complete N-terminal 38 amino acid deletion and PITX2A CΔ173 with a complete C-terminal 173 amino acid deletion were used. All deletion constructs retain the HD. The deletion constructs were co-transfected with FoxJ1 in CHO cells (Fig. 11B). PITX2 NΔ38 and CΔ173 minimally activate the FoxJ1 promoter. FoxJ1 and PITX2A NΔ38 and FoxJ1 and PITX2A CΔ173 synergistically activated the FoxJ1 2.6 promoter at ∼12 and ∼26-fold, respectively (Fig. 11B; P < 0.05). However, the highest synergism was seen with FoxJ1 and the PITX2A CΔ173 deletion construct (Fig. 11B). These data correlate to our previous report that the C-terminal tail of PITX2 has an inhibitory effect on PITX2 activity, which when completely deleted shows increased activation (28). Synergistic activation of the FoxJ1 2.6 promoter by FoxJ1 and PITX2 N or C-terminal deletion constructs (Fig. 11B) suggests that FoxJ1 may interact with PITX2 through its HD.

Figure 11.

FoxJ1 interacts with the PITX2 HD: (A) PITX2 deletion constructs used to map the FoxJ1 interaction; (B) CHO cells were transfected with 2.5 µg of FoxJ1, PITX2A, PITX2A NΔ38 and PITX2A CΔ173 expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids. The activities are compared with the FoxJ1 2.6 promoter plasmids without expression plasmids and normalized to β-galactosidase activity [±SEM (standard error of the mean) from five independent experiments]; (C) GST-PITX2 deletion constructs used to map the FoxJ1 interaction. The location of the FoxJ1 binding domain is noted; (D) GST-PITX2, GST-PITX2 NΔ38 (N-terminal deletion of PITX2), GST-PITX2 CΔ173 (C-terminal deletion of PITX2), GST-PITX2 HD (homeodomain only), GST-PITX2 C173 (C-terminal tail only), GST-PITX2 C 39 (C-terminal 39 amino acids only) pulldown assay with bacterial-expressed and purified FoxJ1 protein (500 ng). FoxJ1 bound to PITX2, PITX2 NΔ38, PITX2 CΔ173 and PITX2 HD. The bound protein was detected by Western blot using the FoxJ1 antibody. As a control GST-beads were incubated with purified FoxJ1 to show the specificity of binding to PITX2A.

Figure 11.

FoxJ1 interacts with the PITX2 HD: (A) PITX2 deletion constructs used to map the FoxJ1 interaction; (B) CHO cells were transfected with 2.5 µg of FoxJ1, PITX2A, PITX2A NΔ38 and PITX2A CΔ173 expression plasmids and 5 µg of FoxJ1 2.6 promoter plasmids. The activities are compared with the FoxJ1 2.6 promoter plasmids without expression plasmids and normalized to β-galactosidase activity [±SEM (standard error of the mean) from five independent experiments]; (C) GST-PITX2 deletion constructs used to map the FoxJ1 interaction. The location of the FoxJ1 binding domain is noted; (D) GST-PITX2, GST-PITX2 NΔ38 (N-terminal deletion of PITX2), GST-PITX2 CΔ173 (C-terminal deletion of PITX2), GST-PITX2 HD (homeodomain only), GST-PITX2 C173 (C-terminal tail only), GST-PITX2 C 39 (C-terminal 39 amino acids only) pulldown assay with bacterial-expressed and purified FoxJ1 protein (500 ng). FoxJ1 bound to PITX2, PITX2 NΔ38, PITX2 CΔ173 and PITX2 HD. The bound protein was detected by Western blot using the FoxJ1 antibody. As a control GST-beads were incubated with purified FoxJ1 to show the specificity of binding to PITX2A.

A glutathione-S-transferase (GST) pulldown assay was performed to map the FoxJ1 interaction domain of PITX2. Immobilized GST-PITX2, GST-PITX2 NΔ38 (N-terminal deletion of PITX2), GST-PITX2 CΔ173 (C-terminal deletion of PITX2), GST-PITX2 HD (HD only), GST-PITX2 C173 (C-terminal tail only), GST-PITX2 C39 (C-terminal 39 amino acids only) (Fig. 11C) were incubated with FoxJ1 (50–500 ng) in one reaction (protein concentrations were adjusted to equal similar amounts of protein molecules). After incubation and extensive washing, each aliquot was resolved separately on a polyacrylamide gel (Fig. 11D). Western blot was probed using the FoxJ1 antibody (Fig. 11D). As a control, GST-beads were used. FoxJ1 bound to PITX2, PITX2 NΔ38, PITX2 CΔ173 and PITX2 HD (Fig. 11D) demonstrates that FoxJ1 interacts with PITX2 through its HD. FoxJ1 did not interact with PITX2 C173 and PITX2 C39 (Fig. 11D).

DISCUSSION

The forkhead family of transcription factors containing a conserved DNA binding motif plays a significant role during development in cell fate determination (12). FoxJ1 is expressed in the conducting airway epithelium (11), choroids plexus epithelium (15,16) and also in the oviduct and developing spermatids (15). During late-stage ciliogenesis, the programs promoting basal body docking and axoneme formation is regulated by FoxJ1 (24). In the immune system T-cell activation is regulated by FoxJ1 and it also prevents autoimmunity mediated via the antagonistic regulation of the proinflammatory transcriptional activity (17). FoxJ1 also causes restrained B-cell activation by antagonizing NF-κB and IL-6 (18). FoxJ1−/− mice exhibit reversal of laterality (19). In this report we demonstrate FoxJ1 expression in the oro-facial tissues and a new molecular mechanism regulating FoxJ1. FoxJ1 expression is observed at E14.5 in the molar and incisor DE; however its expression is down-regulated from E15.5 to E17.5. We show FoxJ1 expression in molar IEE, oral epithelium and tongue epithelium at E18.5. FoxJ1 expression continued in P1 mouse incisor and molar tooth germ with a similar expression pattern to E18.5 embryo. However, at P1 FoxJ1 expression was seen in both epithelial and mesenchymal cell types of developing tooth germ. FoxJ1 is also expressed in the developing sub-mandibular salivary gland. Onset of FoxJ1 expression at early stages, reduced at middle stages and expressed again at late stages of tooth development implicates a new role of this factor in tooth morphogenesis. The differential regulation of FoxJ1 reveals a novel mechanism of this factor during craniofacial/tooth development.

ARS patients with point mutations in PITX2 gene exhibit a spectrum of clinical features including embryotoxon, corectopia, polycoria, signs of congenital glaucoma, microdontia, hypodontia, anodontia, midface dysmorphism and umbilical stump abnormalities (26). Pitx2 is required for normal cardiogenesis, development of maxillary and mandibular prominences, tooth, and pituitary development (10,29,30). Transcription factors act in concert with other transcription factors in regulating development. Protein–protein interactions are required for DNA binding, chromatin remodeling and initiation of transcription. Functional interactions between FOX and HD proteins have been elucidated previously, however if the interactions to be cooperative or inhibitory is dictated by the local cellular environment. For example, FoxC1 a member of the forkhead domain transcription factor family interacts with PITX2 and its activity is inhibited by PITX2 (31), whereas HOXA5 and FOXO1A synergistically activate the IGFBP promoter in HuF cells (32). We demonstrate that PITX2 activates FoxJ1 expression through activation of its promoter. Furthermore, FoxJ1 physically interacts with the PITX2 HD and this interaction synergistically activates the FoxJ1 promoter.

The PITX2 T68P HD mutant variant of ARS did not activate the FoxJ1 promoter with or without FoxJ1. Another report has shown using transfected COS-7 whole cell extracts expressing the PITX2 T68P ARS mutant protein (PITX2 T30P) that it did not bind DNA and was transactivation deficient (27). We have previously shown using purified protein that this ARS mutant protein binds DNA albeit at lower levels compared with wild-type PITX2, but is transactivation defective (6,33). Although the PITX2 T68P protein can physically interact with FoxJ1 this interaction does not result in an active transcriptional complex. These results demonstrate that this PITX2 ARS mutant protein can bind FoxJ1, which may sequester it from binding to wild-type PITX2 and lead to defective gene expression. We contrast this ARS mutant protein to another PITX2 ΔT1261 ARS mutant protein that binds DNA normally but is transactivation deficient due to a deletion of the C-terminus that disrupts protein interactions (34). The PITX2 ΔT1261 protein demonstrates the critical role of protein interactions in regulating the transcriptional activity of PITX2. However, here we demonstrate that protein interactions with the PITX2 T68P protein cannot rescue its defective transactivation activity. Thus, the HD mutation in PITX2 T68P does not affect protein interactions, can bind DNA but cannot activate transcription. We have previously shown that this mutant protein is hyper-phosphorylated, which may be the cause for defective transactivation (34). Overall these data demonstrate that FoxJ1 is a direct downstream target of PITX2 and its decreased expression in ARS due to mutant PITX2 proteins may lead to abnormal cranio-facial/tooth phenotypes.

The Wnt family of secreted glycoproteins has been implicated in numerous events in development and disease. In the absence of Wnt signaling, β-catenin is targeted for destruction by the APC, Axin, Gsk3β complex that phosphorylates β-catenin and directs it to a destruction pathway (35). In the presence of Wnt signaling, β-catenin is stabilized and enters the nucleus where it interacts with T-cell factors, such as Lef-1, to regulate gene expression. PITX2 and β-catenin transcripts are expressed in the DE from initiation stage through the bell stage, whereas Lef-1 transcripts are expressed from the bud through the bell stage of tooth morphogenesis (20,21). Our lab has previously shown that β-catenin and Lef-1 interact independently with PITX2 through the HD and C-terminal tail of PITX2, respectively, and synergistically to regulate the LEF-1 promoter (23). Here we demonstrate that PITX2, β-catenin, and LEF-1 transcription factors together can synergistically activate the FoxJ1 promoter. However, FoxJ1 increases the transcriptional activation in concert with these transcription factors. We propose a mechanism that PITX2, Lef-1 and β-catenin transcription factors that are available from the early stages through the bell stage form a complex (23) and activate the FoxJ1 gene (Fig. 12). Once the FoxJ1 transcription factor is expressed, it can interact with the PITX2, Lef-1 and β-catenin complex through the HD of PITX2 and regulate its own promoter (Fig. 12). Our data implicate a mechanism where FoxJ1 is regulated in a positive feedback loop fashion in concert with other specific transcription factors (Fig. 12).

Figure 12.

Molecular mechanism showing regulation of the FoxJ1 promoter by PITX2, Lef-1, β-catenin and FoxJ1 interactions.

Figure 12.

Molecular mechanism showing regulation of the FoxJ1 promoter by PITX2, Lef-1, β-catenin and FoxJ1 interactions.

In summary, we show that FoxJ1 is a new downstream target of PITX2 during oro-facial morphogenesis. We also demonstrate how transcription factors act in concert in regulating FoxJ1 expression, which emphasizes the significance of studying protein–protein interactions in explaining the complex nature of genetic defects. This report also implicates a novel molecular mechanism regulating FoxJ1 expression in the oro-facial region. Further functional characterization of FoxJ1 in knockout mice will be published elsewhere.

MATERIALS AND METHODS

Immunohistochemistry

Murine C57BL/6 embryos were used for IHC. Embryos were treated with 4% paraformaldehyde and dehydrated with sequential concentration of alcohol and finally with xylene. The embryos were embedded in paraffin and sections were made at 7 µm thickness. Sections were deparaffinized and treated with 0.1 M sodium citrate buffer for 7.5 min at 100% power in a revolving 800 W microwave and additional two cycles of 5 min at 50% power. Subsequently sections were treated with 4% hydrogen peroxide to block the endogenous peroxidase activity. The slides were incubated with 2% goat serum for 30 min, followed by overnight incubation with anti-FoxJ1 antibody at 1:400 dilution (kindly provided by Dr Steve Brody). After antibody incubation, the slides were treated with biotinylated anti-mouse secondary antibody (Vector Laboratories, Burlingame, CA, USA) at a concentration of 1:200 for 30 min. Avidin biotin complex and 3-3′-diaminobenzidine substrate (Vector Laboratories) were used as our reporting system. Negative controls were treated exactly the same except phosphate buffered saline (PBS) with 2% goat serum was substituted for the primary antibody.

Expression and reporter constructs

Expression plasmids containing the cytomegalovirus (CMV) promoter linked to the PITX2 cDNA were constructed in pcDNA 3.1 MycHisC (Invitrogen, Carlsbad, CA, USA) (5,6,28). Lef-1, FoxJ1 and β-catenin S37A expression plasmids have been previously described (13,23,36,37). To construct the FoxJ1 2.6 luciferase reporter ∼2.6 kb of 5′ flanking sequence from the FoxJ1 gene was cloned into our previously described luciferase vector (28). All constructs were confirmed by DNA sequencing. A SV-40 or CMV β-galactosidase reporter plasmid was co-transfected in all experiments as a control for transfection efficiency. All plasmids were double-banded CsCl purified.

Cell culture, transient transfection, luciferase and β-galactosidase assays

CHO cells were cultured in Dulbecco's modification of eagle's medium (DMEM) supplemented with 5% fetal bovine serum (FBS) and penicillin/streptomycin and transfected by electroporation. Cultures were fed 24 h prior to transfection, cells were resuspended in PBS and mixed with 2.5 µg of expression plasmids, 5 µg of reporter plasmid and 0.5 µg of SV-40 β-galactosidase plasmid. Electroporation of CHO cells were performed at 380 V and 950 microfarads (µF) (Gene Pulser XL; Bio-Rad Laboratories, Hercules, CA, USA). Transfected cells were incubated for 24 h in 60 mm culture dishes and fed with 5% FBS and DMEM and then lysed and assayed for reporter activities and protein content by Bradford assay (Bio-Rad). Luciferase was measured using reagents from Promega (Madison, WI, USA). β-galactosidase was measured using the Galacto-Light Plus reagents (Tropix Inc., Bedford, MA, USA). All luciferase activities were normalized to β-galactosidase activity.

Generation of the PITX2C transgenic mouse and fibroblast cultures

The PITX2C cDNA was cloned into the K14 promoter construct (38). We have placed the hrGFP (humanized Renilla GFP) gene in the cassette to observe expression in live cells and have observed good expression of PITX2C in transgenic mice by PCR. The PITX2C GFP DNA was excised from the plasmid and used for pronuclear injection. Donor female mice (FVB/NCr), stud male (FVB/NCr), vasectomized male (ICR) and recipient female (ICR) were used in the experiments. Multiple founders were analyzed for transgene expression and crossed to BL6 mice and re-evaluated for expression.

Fibroblasts were prepared by mincing the embryo and allowing the cells to proliferate from the tissue in 10% FBS, DMEM. PITX2C transgenic and wild-type mouse cells were cultured and lysates prepared for western blotting.

Immunoprecipitation assay

Approximately 24 h after cell transfection with PITX2 and FoxJ1, CHO cells were rinsed with 1 ml of PBS and then incubated with 1 ml ice-cold radio immunoprecipitation assay buffer for 15 min at 4°C. Cells were harvested and disrupted by repeated aspiration through a 25-gauge needle attached to a 1 ml syringe. The lysates were then incubated on ice for 30 min. Cellular debris was pelleted by centrifugation at 10 000g for 10 min at 4°C. An aliquot of lysate was saved for analysis as input control. Supernatant was transferred to a fresh 1.5 ml microcentrifuge tube on ice and precleared using the mouse Exactacruz C IP matrix (Santa Cruz Biotechnology, Santa Cruz, CA) for 30 min at 4°C. The matrix was removed by brief centrifugation, and the pellet was transferred to a new tube. The IP antibody-IP matrix was prepared according to the manufacturer’s instructions, using primary anti-FoxJ1 (5 µl) antibody (Upstate). The IP antibody-IP matrix complex was incubated with the precleared cell lysate at 4°C for 12 h. After incubation, the lysate was centrifuged to pellet the IP matrix. The matrix was washed two times with PBS and resuspended in 15 µl of double distilled water and 3 µl 6X SDS loading dye. Samples were boiled for 5 min and resolved on a 10% polyacrylamide gel. Western blotting was performed with anti-PITX2 antibody and HRP (horse radish peroxidase) conjugated reagent to detect immunoprecipitated proteins.

ChIP assay

The ChIP assays were performed as previously described using the ChIP Assay Kit (Upstate) with the following modifications (23,39). LS-8 cells were fed for 24 h, harvested and plated in 60 mm dishes. Cells were cross-linked with 1% formaldehyde for 10 min at 37°C the next day. All PCR reactions were done under an annealing temperature of 58°C. Two primers for amplifying the Pitx2 binding site in the FoxJ1 promoter are as follows: sense –5′GTACTGCCCAAGAGTTAGGATCACA3′ and anti-sense – 5′GTTGTTTCTGTTCTGGTTTTTGAGG3′. All the PCR products were evaluated on a 2% agarose gel in 1x TBE for appropriate size (261 bp) and confirmed by sequencing. As Controls the FoxJ1 primers were used without chromatin, normal mouse IgG was used replacing the PITX2 antibody to reveal non-specific immunoprecipitation of the chromatin.

Expression and purification of GST-FoxJ1 fusion proteins

FoxJ1 was PCR amplified from cDNA clones and ligated into pGEX6P-2 GST vector using BamHI and Not1 restriction enzyme sites engineered into the primers. The plasmid was transformed into BL21 cells. Protein was isolated as described (28). FoxJ1 protein was cleaved from the GST moiety using 80 units of Prescission protease per milliliter of glutathione Sepharose. Cleaved proteins were stored in 10% glycerol. Protein concentration was quantitated with Bradford Reagent (Bio-Rad). Proteins were examined by electrophoresis on denaturing SDS–polyacrylamide gels, followed by Coomassie Blue staining (50% methanol, 10% acetic acid and 0.5% Coomassie brilliant blue stain).

GST pulldown assays

Immobilized GST-PITX2, GST-PITX2 ΔN38, GST-PITX2 ΔC173, GST-PITX2 HD, GST-PITX2 C173, GST-PITX2 C 39 fusion proteins were suspended in binding buffer [20 mm HEPES pH 7.5, 5% glycerol, 50 mm NaCl, 1 mm EDTA (ethylenedinitriotetraacetic acid), 1 mm DTT (dithiothreitol), with or without 1% milk and 400 µg/ml of ethidium bromide]. Purified bacteria-expressed FoxJ1 proteins (50–500 ng) were added to 10–30 µg immobilized GST-PITX2 fusion proteins or GST in a total volume of 100 µl and incubated for 30 min at 4°C. The beads were pelleted and washed five times with 200 µl binding buffer. The bound proteins were eluted by boiling in SDS-sample buffer and separated on a 10% SDS–polyacrylamide gel. Approximately 100 ng of purified FoxJ1 proteins were analyzed in separate western blots. Following SDS gel electrophoresis, the proteins were transferred to polyvinylidenefluride filters (Millipore, Temecula, CA, USA), immunoblotted and detected using FoxJ1 antibody (Active Motif, Carlsbad, CA) and enhanced chemiluminescence reagents from GE Healthcare.

FUNDING

Support for this research was provided by grant DE 13941 from the National Institute of Dental and Craniofacial Research and ES09106 from the National Institute of Environmental Health Sciences to B.A.A. and from the Swedish Research Council to T.A.H.

ACKNOWLEDGEMENTS

We thank Chuzhi Yin, HuoJun Cao (IBT, Texas A&M Health Science center, Houston, TX, USA), and Alfredo Davalos, Jose Balderas (ITESM, Mexico) for their advice, technical support and helpful discussions.

Conflict of Interest statement. None declared.

REFERENCES

1
Rieger
H.
Dysgenesis mesodermalis coreneal et iridis
Z. Augenheilk
 , 
1935
, vol. 
86
 pg. 
333
 
2
Semina
E.V.
Reiter
R.
Leysens
N.J.
Alward
L.M.
Small
K.W.
Datson
N.A.
Siegel-Bartelt
J.
Bierke-Nelson
D.
Bitoun
P.
Zabel
B.U.
, et al.  . 
Cloning and characterization of a novel bicoid-related homeobox transcription factor gene, RIEG, involved in Rieger syndrome
Nature Genet.
 , 
1996
, vol. 
14
 (pg. 
392
-
399
)
3
Mucchielli
M.-L.
Mitsiadis
T.A.
Raffo
S.
Brunet
J.-F.
Proust
J.-P.
Goridis
C.
Mouse Otlx2/RIEG expression in the odontogenic epithelium precedes tooth initiation and requires mesenchyme-derived signals for its maintenance
Dev. Biol.
 , 
1997
, vol. 
189
 (pg. 
275
-
284
)
4
Hjalt
T.A.
Semina
E.V.
Amendt
B.A.
Murray
J.C.
The Pitx2 protein in mouse development
Dev. Dyn.
 , 
2000
, vol. 
218
 (pg. 
195
-
200
)
5
Cox
C.J.
Espinoza
H.M.
McWilliams
B.
Chappell
K.
Morton
L.
Hjalt
T.A.
Semina
E.V.
Amendt
B.A.
Differential regulation of gene expression by PITX2 isoforms
J. Biol. Chem.
 , 
2002
, vol. 
277
 (pg. 
25001
-
25010
)
6
Amendt
B.A.
Sutherland
L.B.
Semina
E.
Russo
A.F.
The molecular basis of rieger syndrome: analysis of Pitx2 homeodomain protein activities
J. Biol. Chem.
 , 
1998
, vol. 
273
 (pg. 
20066
-
20072
)
7
Shapiro
M.
Marks
M.
Peichel
C.
Blackman
B.
Nereng
K.
Jonsson
B.
Schluter
D.
Kingsley
D.
Genetic and developmental basis of evolutionary pelvic reduction in threespine sticklebacks
Nature
 , 
2004
, vol. 
428
 (pg. 
717
-
723
)
8
Liu
C.
Liu
W.
Lu
M.
Brown
N.A.
Martin
J.F.
Regulation of left–right asymmetry by thresholds of Pitx2c activity
Development
 , 
2001
, vol. 
128
 (pg. 
2039
-
2048
)
9
Liu
C.
Liu
W.
Palie
J.
Lu
M.F.
Brown
N.A.
Martin
J.F.
Pitx2c patterns anterior myocardium and aortic arch vessels and is required for local cell movement into atrioventricular cushions
Development
 , 
2002
, vol. 
129
 (pg. 
5081
-
5091
)
10
Lin
C.R.
Kioussi
C.
O’Connell
S.
Briata
P.
Szeto
D.
Liu
F.
Izpisua-Belmonte
J.C.
Rosenfeld
M.G.
Pitx2 regulates lung asymmetry, cardiac positioning and pituitary and tooth morphogenesis
Nature
 , 
1999
, vol. 
401
 (pg. 
279
-
282
)
11
Clevidence
D.E.
Overider
D.G.
Tao
W.
Qian
X.
Pani
L.
Lai
E.
Costa
R.H.
Identification of nine tissue-specific transcription factors of the hepatocyte nuclear factor 3/forkhead DNA-binding-domain family
Proc. Natl Acad. Sci. USA
 , 
1993
, vol. 
90
 (pg. 
3948
-
3952
)
12
Carlsson
P.
Mahlapuu
M.
Forkhead transcription factors: key players in development and metabolism
Dev. Biol.
 , 
2002
, vol. 
250
 (pg. 
1
-
23
)
13
Blatt
E.N.
Yan
X.H.
Wuerffel
M.K.
Hamilos
D.L.
Brody
S.L.
Forkhead transcription factor HFH-4 expression is temporally related to ciliogenesis
Am. J. Respir. Cell. Mol. Biol.
 , 
1999
, vol. 
21
 (pg. 
168
-
176
)
14
Brody
S.L.
Yan
X.H.
Wuerffel
M.K.
Song
S.-K.
Shapiro
S.D.
Ciliogenesis and left–right axis defects in forkhead factor HFH-4-null mice
Am. J. Repir. Cell. Mol. Biol.
 , 
2000
, vol. 
23
 (pg. 
45
-
51
)
15
Hackett
B.P.
Brody
S.L.
Liang
M.
Zeitz
I.D.
Bruns
L.A.
Gitlin
J.D.
Primary structure of hepatocyte nuclear factor/forkhead homologue 4 and characterization of gene expression in the developing respiratory and reproductive epithelium
Proc. Natl Acad. Sci. USA
 , 
1995
, vol. 
92
 (pg. 
4249
-
4253
)
16
Lim
L.
Zhou
H.
Costa
R.H.
The winged helix transcription factor HFH-4 is expressed during choroid plexus epithelial development in the mouse embryo
Proc. Natl Acad. Sci. USA
 , 
1997
, vol. 
94
 (pg. 
3094
-
3099
)
17
Lin
L.
Spoor
M.S.
Gerth
A.J.
Brody
S.L.
Peng
S.L.
Modulation of Th1 activation and inflammation by the NF-KB repressor Foxj1
Science
 , 
2004
, vol. 
303
 (pg. 
1017
-
1020
)
18
Lin
L.
Brody
S.L.
Peng
S.L.
Restraint of B Cell activation by Foxj1-mediated antagonism of NF-kB and IL-6
J. Immunol.
 , 
2005
, vol. 
175
 (pg. 
951
-
958
)
19
Tamakoshi
T.
Itakura
T.
Chandra
A.
Uezato
T.
Yang
Z.
Xiao-Dong
X.
Wang
B.
Hackett
B.P.
Yokoyama
T.
Miura
N.
Roles of the Foxj1 and Inv genes in the left–right determination of internal organs in mice
Biophys. Biochem. Res. Comm.
 , 
2006
, vol. 
339
 (pg. 
932
-
938
)
20
van Genderen
C.
Okamura
R.M.
Farinas
I.
Quo
R.-G.
Parslow
T.G.
Bruhn
L.
Grosschedl
R.
Development of several organs that require inductive epithelial-mesenchymal interactions is impared in LEF-1-deficient mice
Genes Dev.
 , 
1994
, vol. 
8
 (pg. 
2691
-
2703
)
21
Fausser
J.L.
Schlepp
O.
Aberdam
D.
Meneguzzi
G.
Ruch
J.V.
Lesot
H.
Localization of antigens associated with adherens junctions, desmosomes, and hemidesmosomes during murine molar morphogenesis
Differentiation
 , 
1998
, vol. 
63
 (pg. 
1
-
11
)
22
Vadlamudi
U.
Espinoza
H.M.
Ganga
M.
Martin
D.M.
Liu
X.
Engelhardt
J.F.
Amendt
B.A.
PITX2, β-catenin, and LEF-1 interact to synergistically regulate the LEF-1 promoter
J. Cell Sci.
 , 
2005
, vol. 
118
 (pg. 
1129
-
1137
)
23
Amen
M.
Liu
X.
Vadlamudi
U.
Elizondo
G.
Diamond
E.
Engelhardt
J.F.
Amendt
B.A.
PITX2 and β-catenin interactions regulate Lef-1 isoform expression
Mol. Cell. Biol.
 , 
2007
, vol. 
27
 (pg. 
7560
-
7573
)
24
You
Y.
Huang
T.
Richer
E.J.
Schmidt
J.-E.H.
Zabner
J.
Borok
Z.
Brody
S.L.
Role of f-box factor foxj1 in differentiation of ciliated airway epithelial cells
Am. J. Physiol. Lung Cell. Mol. Physiol.
 , 
2004
, vol. 
286
 (pg. 
L650
-
L657
)
25
Mucchielli
M.
Martinez
S.
Pattyn
A.
Goridis
C.
Brunet
J.
Otlx2, an Otx-related homeobox gene expressed in the pituitary gland and in a restricted pattern in the forebrain
Mol. Cell. Neurosci.
 , 
1996
, vol. 
8
 (pg. 
258
-
271
)
26
Amendt
B.A.
Semina
E.V.
Alward
W.L.M.
Rieger syndrome: a clinical, molecular and biochemical analysis
Cell. Mol. Life Sci.
 , 
2000
, vol. 
57
 (pg. 
1652
-
1666
)
27
Kozlowski
K.
Walter
M.A.
Variation in residual PITX2 activity underlies the phenotypic spectrum of anterior segment developmental disorders
Hum. Mol. Genet.
 , 
2000
, vol. 
9
 (pg. 
2131
-
2139
)
28
Amendt
B.A.
Sutherland
L.B.
Russo
A.F.
Multifunctional role of the Pitx2 homeodomain protein C-terminal tail
Mol. Cel. Biol.
 , 
1999
, vol. 
19
 (pg. 
7001
-
7010
)
29
Gage
P.J.
Suh
H.
Camper
S.A.
The bicoid-related Pitx gene family in development
Mamm. Genome
 , 
1999
, vol. 
10
 (pg. 
197
-
200
)
30
Lu
M.
Pressman
C.
Dyer
R.
Johnson
R.L.
Martin
J.F.
Function of Rieger syndrome gene in left–right asymmetry and craniofacial development
Nature
 , 
1999
, vol. 
401
 (pg. 
276
-
278
)
31
Berry
F.B.
Lines
M.A.
Oas
J.M.
Footz
T.
Underhill
D.A.
Gage
P.J.
Walter
M.A.
Functional interactions between FOXC1 and PITX2 underlie the sensitivity to FOXC1 gene dose in Axenfeld–Rieger syndrome and anterior segment dysgenesis
Hum. Mol. Genet.
 , 
2006
, vol. 
15
 (pg. 
905
-
919
)
32
Foucher
I.
Volovitch
M.
Frain
M.
Kim
J.J.
Souberbielle
J.-C.
Gan
L.
Unterman
T.G.
Prochiantz
A.
Trembleau
A.
Hoxa5 overexpression correlates with IGFBP1 upregulation and postnatal dwarfism: evidence for an interaction between Hoxa5 and Forkhead box transcription factors
Development
 , 
2002
, vol. 
129
 (pg. 
4065
-
4074
)
33
Espinoza
H.M.
Cox
C.J.
Semina
E.V.
Amendt
B.A.
A molecular basis for differential developmental anomalies in Axenfeld–Rieger syndrome
Hum. Mol. Genet.
 , 
2002
, vol. 
11
 (pg. 
743
-
753
)
34
Espinoza
H.M.
Ganga
M.
Vadlamudi
U.
Martin
D.M.
Brooks
B.P.
Semina
E.V.
Murray
J.C.
Amendt
B.A.
Protein kinase C phosphorylation modulates N- and C-terminal regulatory activities of the PITX2 homeodomain protein
Biochemistry
 , 
2005
, vol. 
44
 (pg. 
3942
-
3954
)
35
Huelsken
J.
Birchmeier
W.
New aspects of Wnt signaling pathways in higher vertebrates
Curr. Opin. Genet. Dev.
 , 
2001
, vol. 
11
 (pg. 
547
-
553
)
36
Filali
M.
Cheng
N.
Abbott
D.
Leontiev
V.
Engelhardt
J.F.
Wnt-3A/β-catenin signaling induces transcription from the LEF-1 promoter
J. Biol. Chem.
 , 
2002
, vol. 
277
 (pg. 
33398
-
33410
)
37
Brody
S.L.
Hackett
B.P.
White
R.A.
Structural characrterization of the mouse Hfh4 gene, a developmentally regulated forkhead family member
Genomics
 , 
1997
, vol. 
45
 (pg. 
509
-
518
)
38
Cui
X.-M.
Shiomi
N.
Chen
J.
Saito
T.
Yamamoto
T.
Ito
Y.
Bringas
P.
Chai
Y.
Shuler
C.F.
Overexpression of Smad2 in Tgf-β3-null mutant mice rescues cleft palate
Dev. Biol.
 , 
2005
, vol. 
278
 (pg. 
193
-
202
)
39
Diamond
E.
Amen
M.
Hu
Q.
Espinoza
H.M.
Amendt
B.A.
Functional interactions between Dlx2 and lymphoid enhancer factor regulate Msx2
Nucleic Acids Res.
 , 
2006
, vol. 
34
 (pg. 
5951
-
5965
)