Skip Navigation


Human Molecular Genetics Advance Access originally published online on January 31, 2006
Human Molecular Genetics 2006 15(6):905-919; doi:10.1093/hmg/ddl008
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
15/6/905    most recent
ddl008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Berry, F. B.
Right arrow Articles by Walter, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Berry, F. B.
Right arrow Articles by Walter, M. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Functional interactions between FOXC1 and PITX2 underlie the sensitivity to FOXC1 gene dose in Axenfeld–Rieger syndrome and anterior segment dysgenesis

Fred B. Berry1,*,{dagger}, Matthew A. Lines2,{dagger}, J. Martin Oas3, Tim Footz1, D. Alan Underhill2, Philip J. Gage3 and Michael A. Walter1,2

1Department of Ophthalmology and 2Department of Medical Genetics, University of Alberta, Edmonton, Alberta, Canada, T6G 2H7 and 3Department of Ophthalmology and Visual Sciences, University of Michigan, Ann Arbor, MI 48105, USA

* To whom correspondence be addressed. Tel: +1 7804923028; Fax: +1 7804926934; Email: fberry{at}ualberta.ca

Received December 19, 2005; Accepted January 27, 2006


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Axenfeld–Rieger ocular dysgenesis is associated with mutations of the human PITX2 and FOXC1 genes, which encode transcription factors of the homeodomain and forkhead types, respectively. We have identified a functional link between FOXC1 and PITX2 which we propose underpins the similar Axenfeld–Rieger phenotype caused by mutations of these genes. FOXC1 and PITX2A physically interact, and this interaction requires crucial functional domains on both proteins: the C-terminal activation domain of FOXC1 and the homeodomain of PITX2. Immunofluorescence further shows PITX2A and FOXC1 to be colocalized within a common nuclear subcompartment. Furthermore, PITX2A can function as a negative regulator of FOXC1 transactivity. This work ties both proteins into a common pathway and offers an explanation of why increased FOXC1 gene dosage produces a phenotype resembling that of PITX2 deletions and mutations. Ocular phenotypes arise despite the deregulated expression of FOXC1-target genes through mutations in FOXC1 or PITX2. Ultimately, PITX2 loss of function mutations have a compound effect: the reduced expression of PITX2-target genes coupled with the extensive activation of FOXC1-regulated targets. Our findings indicate that the functional interaction between FOXC1 and PITX2A underlies the sensitivity to FOXC1 gene dosage in Axenfeld–Rieger syndrome and related anterior segment dysgeneses.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Axenfeld–Rieger syndrome (ARS) describes a group of genetically and phenotypically heterogeneous disorders that primarily affects the anterior segment of the eye (1Go,2Go). Affected individuals display a characteristic spectrum of ocular anomalies, which can include iris hypoplasia, corectopia, iris strands connecting the iridocorneal angle to the trabecular meshwork (TM) and a prominent, anteriorly displaced Schwalbe's line. Systemic features of this disorder, although more variable in their presentation, can include maxillary hypoplasia, telecanthus, hypertelorism, hypodontia, adontia and redundant periumbilical skin. In addition, cardiac anomalies have been found in some patients, although whether this presentation reflects a comorbidity remains to be determined (3Go–5Go). The most disabling complication of this disorder is a heightened propensity for developing glaucoma, with approximately 50% of affected individuals acquiring this progressively blinding disease.

ARS is transmitted in an autosomal dominant manner and is highly penetrant. There are four known genetic loci for ARS, situated on human chromosomes 4q25 (6Go–8Go), 6p25 (3Go,9Go), 13q14 (10Go) and 16q (11Go). Disease-causing mutations have been identified in two transcription factor genes, PITX2 and FOXC1, which are responsible for ARS mapping to chromosomes 4q25 and 6p25, respectively (3Go,8Go,9Go). The causative genes at the 13q14 and 16q loci have yet to be identified.

PITX2 is a homeobox transcription factor related to the paired class of homeodomain (HD) proteins. PITX2 produces four mRNA transcripts, with three (PITX2A, PITX2B, PITX2C) representing distinct combinations of the four 5' exons but sharing the majority of the open-reading frame. A fourth isoform, PITX2D, has been described and suggested to be translated from an internal methionine, but a canonical Kozak consensus is absent at this position in PITX2 (12Go). The translation product of this isoform has never been detected and would, in any case, lack a complete HD. The PITX2A isoform encodes the smallest polypeptide (32 kDa) and has been the best studied within the context of AR malformations. Several missense mutations resulting in single amino acid substitution in the PITX2 coding region have been identified and characterized. These PITX2 mutant alleles alter protein function to varying degrees. For example, R84W is a hypomorphic allele that reduces DNA-binding and transcriptional activation (13Go). The T68P, K88E and R91P mutations produce loss of function alleles that cannot bind DNA, and K88E is believed to act in a dominant negative fashion by binding to, and interfering with, the wild-type (WT) PITX2 allele (13Go,14Go). Finally, the V84L PITX2 mutation displays potent transcriptional activation and is a hypermorphic allele (15Go).

FOXC1 is a member of the forkhead box or FOX class of transcription factors. These proteins are characterized by a conserved 110 amino acid DNA-binding domain, known as a forkhead domain (FHD). FOX proteins are key regulators of diverse cellular functions including the development of many organ systems, energy homeostasis and oncogenesis (16Go–18Go). The mutation spectrum for FOXC1 largely comprises frameshift and nonsense mutations occurring upstream of the FHD that encodes truncated proteins, as well as missense mutations occurring within the FHD itself (19Go). Morever, interstitial duplications of the FOXC1 gene can also lead to anterior segment dysgenesis and glaucoma, indicating the necessity of a stringent control of FOXC1 levels and activities (20Go–22Go). The analyses of AR-causing FOXC1 missense mutations reveal defects in protein stability, nuclear localization, DNA-binding specificity and transcriptional activation. Ultimately, these mutations disrupt the proper activation of FOXC1-target genes (23Go–25Go).

Expression of both Pitx2 and Foxc1 mRNAs during mouse development correlates well with organs affected in ARS. Expression of Pitx2 mRNA is detectable in the embryonic periocular mesenchyme, dental epithelium, first branchial arch, umbilicus, pituitary primordium and limb buds (8Go,26Go–28Go). During ocular development, Pitx2 mRNA is expressed in mesenchyme surrounding the optic eminence as early as E8.5, in anticipation of the formation of the optic cup and lens vesicle (26Go). By E11.5, when the latter two structures have formed, Pitx2 is highly expressed in the now migrating periocular mesenchyme (8Go,27Go,29Go). Expression is retained in mesenchymal derivatives such as the corneal endothelium and TM thereafter, gradually becoming restricted to the angle by E18.5. In the developing murine eye, the expression pattern of Foxc1 is very similar to that of Pitx2, initiating in the periocular mesenchyme prior to the onset of mesenchymal cell migration at E11.5 (30Go,31Go). Foxc1 is detectable shortly thereafter in the precorneal mesenchyme, the hyaloid plexus at E12.5 and the TM and conjunctiva by E16.5. Expression in the TM and conjunctiva is seen to persist at least until parturition and potentially thereafter.

Although the similar expression patterns and phenotypes of PITX2 and FOXC1 imply that both genes fulfil a common biological role during ocular development, no definitive in vivo confirmation of coexpression has been presented, and the precise molecular nature of this functional relationship remains obscure. PITX2 and FOXC1 might be subunits of a common macromolecular complex within the cell nucleus. We have examined this hypothesis via biochemical and microscopic methods. We show coexpression of the two proteins in periocular mesenchyme and report that recombinant PITX2A and FOXC1 can indeed form a physical complex and have mapped this activity to essential domains within both proteins. Moreover, PITX2A and FOXC1 are colocalized within a restricted subset of chromatin. Our findings suggest that the activities of PITX2A and FOXC1 are functionally interconnected by the presence of both proteins in a common complex on chromatin. Finally, we demonstrate that PITX2 can negatively regulate the transcriptional activation potential of FOXC1, and such regulation appears to underlie sensitivity to FOXC1 dosage in the eye.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
PITX2 and FOXC1 colocalize in the developing mouse eye
To assess the possibility that FOXC1 and PITX2 might be capable of regulating common regulatory networks in the same cells during embryonic eye development, we immunostained sections taken from E11.5, E12.5 and E16.5 embryos to detect the presence of both FOXC1 (Fig. 1A, D, G and J) and PITX2 (Fig. 1B, E, H and K) proteins. We found compelling evidence for coexpression of the two proteins within specific populations of periocular mesenchyme cells at each time point examined. At E11.5, the two transcription factors are coexpressed within cells located in the presumptive anterior segment that are fated to contribute to the corneal endothelium or stroma, as well as within cells located at the periphery of the optic cup (Fig. 1C). The latter cells are fated to become either the choroidal capillary system or the sclera. Coexpression of PITX2 and FOXC1 continues within the anterior segment at E12.5 and has spread to include cells at the posterior of the optic cup (Fig. 1F). By E16.5, coexpression has become localized to cells within specific ocular structures, including the corneal endothelium, presumptive iridocorneal angle and the choroid (Fig. 1I and L). Interestingly, the degree of coexpression appears to be variable across the population of cells at each time point, with some cells exhibiting higher levels of FOXC1 or PITX2 staining and others exhibiting approximately equal staining levels. This suggests that there may be quantitative differences in the relative expression levels of PITX2 and FOXC1 between individual cells. These observations demonstrate that both FOXC1 and PITX2 are present during anterior segment development consistent with the hypothesis that both proteins may interact in the regulation of common downstream target genes within specific cell lineages.


Figure 0081
View larger version (128K):
[in this window]
[in a new window]
  		Figure 1. Expression of FOXC1 and PITX2 during eye development. Transverse views of whole eye primordia from E11.5 (A–C) and E12.5 (D–F) as well as E16.5 iridocorneal angle (G–I) and choriod/sclera (J–L). Sections were coimmunostained to detect FOXC1 (green) and PITX2 (Red). L, Lens; OC, optic cup; OS, optic stalk; EM, extraocular muscle; CS, corneal stroma; and S, sclera. Arrowheads indicate presumptive cornea; open arrowheads, corneal endothelium; arrows, iridocorneal angle; and open arrows, choroid.

 
We also noted significant sites within the periocular mesenchyme where FOXC1 and PITX2 are not coexpressed. PITX2 is expressed throughout the extraocular muscle, whereas FOCX1 is absent in all but a few cells (Fig. 1C and F). Within the anterior segment at E11.5 and E12.5, cells expressing only PITX2 are intermingled with cells coexpressing both proteins (Fig. 1C and F). In contrast, some cells at the posterior optic cup and surrounding the optic stalk express FOXC1 but not PITX2 at the same time points (Fig. 1C and F). Later in development, cells of the differentiated corneal endothelium express both proteins, whereas cells of the corneal stroma express only PITX2 (Fig. 1I). These observations are consistent with the idea that certain cell populations of the periocular mesenchyme may rely independently on the function of FOXC1 or PITX2, or alternatively that relative levels of both proteins fluctuate on a cell-autonomous basis within the periocular mesenchyme.

PITX2A and FOXC1 physically interact
PITX2 and FOXC1 may function cooperatively during ocular ontogeny, as evidenced by their similar loss of function ocular phenotypes in both humans and mice and by the pattern of coexpression described previously. Nonetheless, essentially, all functional studies of PITX2 and FOXC1 have focused on either protein individually, and a model relating the two AR-associated factors has therefore remained wholly undefined. One possible link between PITX2 and FOXC1 activities might be the coordination of both proteins in a shared physical complex. To investigate whether PITX2A and FOXC1 can physically interact, immunoprecipitation (IP) experiments were performed using COS-7 cell extracts expressing haemagglutinin (HA)-tagged PITX2A and Xpress-tagged FOXC1 (Xp-FOXC1). Xp-FOXC1 was recovered in {alpha}-HA IP isolates in an HA-PITX2A-dependent fashion and HA-PITX2A was recovered by {alpha}-Xpress in an Xp-FOXC1-dependent manner (Fig. 2). PITX2A and FOXC1 can therefore specifically associate in a physical complex.


Figure 0082
View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. PITX2 and FOXC1 physically interact. Coimmunoprecipitation experiments from COS-7 cell extracts transfected with HA-PITX2A and Xpress-FOXC1 (Xp-FOXC1) along with the appropriate empty expression vectors pCI-HA or pcDNA4. (A) Extracts were immunoprecipitated with anti-HA antibodies and immunoblotted with anti-Xpress antibodies to detect FOXC1. (B) Reciprocal IP experiments in which extracts were immunoprecipitated with anti-Xpress antibodies and immunoblotted with anti-HA antibodies to detect PITX2A. (C) Input fractions for IP experiments demonstrating expression of HA-PITX2 and Xp-FOXC1. Asterisks indicate a PITX2A-dependent 65 kDa anti-HA immunoreactive band.

 
Having established that PITX2A and FOXC1 can physically interact, we mapped the regions of FOXC1 responsible for PITX2A binding. COS-7 cell extracts cotransfected with HA-PITX2A and a panel of Xp-FOXC1 deletion constructs were subjected to immunoprecipitation experiments with {alpha}-HA antibodies. These experiments, as shown in Figure 3, indicated that the N-terminal activation domain (AD1; residues 1–50) and regulatory domain (TRD; residues 215–366) of FOXC1 were dispensable for PITX2A binding. Conversely, C-terminal truncations of FOXC1 at residues 366, 435 or 475 completely ablated the interaction with PITX2A. These data are consistent with a PITX2-binding domain (PITX2BD) situated within the C-terminal third (residues 475–553) of FOXC1 that overlaps with the C-terminal activation domain of FOXC1 (AD2; residues 436–553). Thus, these results imply that PITX2 binding may be functionally involved in target gene activation by FOXC1.


Figure 0083
View larger version (43K):
[in this window]
[in a new window]
  		Figure 3. Mapping of the FOXC1 regions necessary for PITX2A interactions. (A and B) COS-7 cells were transfected with a panel of FOXC1 deletion constructs (39Go), along with HA-PITX2A. Extracts were immunoprecipitated with anti-HA antibodies and immunoblot analysis was performed with anti-Xpress antibodies. (C) Summary of the domain mapping experiments indicating the PITX2A interaction domain on FOXC1.

 
Next, we mapped the region of PITX2 responsible for the interaction with FOXC1. Full-length FOXC1 was expressed in bacteria as a hexahistadine fusion protein, purified and bound to Ni2+-NTA agarose. COS-7 cell extracts expressing HA-PITX2A deletion constructs were incubated with Ni2+-NTA agarose alone or with Ni2+-NTA agarose conjugated with FOXC1. Full-length PITX2A bound to Ni2+-FOXC1 in these assays (Fig. 4A), corroborating our previous IP experiments. The analysis of cell extracts expressing PITX2A deletions indicated that the HD of PITX2A (residues 39–98) was necessary for interactions with Ni2+-FOXC1. Reciprocal interaction experiments were performed with HA-PITX2A deletions expressed in bacteria, conjugated to Ni2+-NTA agarose and incubated with cell extracts expressing V5-tagged FOXC1. As indicated in Figure 4B, deletion of the PITX2A HD ({Delta}39-98) ablated FOXC1 interactions. In addition, the deletion of PITX2A residues 99–159 and 231–271 largely disrupted the interaction with FOXC1.


Figure 0084
View larger version (33K):
[in this window]
[in a new window]
  		Figure 4. The PITX2A HD mediates the interaction with FOXC1. (A) Ni2+-pulldown experiments were performed by incubating bacterially expressed 6xHIS-FOXC1 conjugated to Ni2+-NTA agarose with COS-7 cell extracts expressing HA-tagged deletions of the PITX2A protein and by immunoblotting with anti-HA antibodies. Arrows indicate the expressed PITX2A deletions. Non-specific bands are indicated with asterisks. (B and C) Pulldown experiments were performed with bacterially expressed PITX2A deletion constructs and incubated with COS-7 cell extracts expressing V5-tagged FOXC1. (D) Summary of the FOXC1–PITX2A interaction experiments.

 
One logical question stemming from the preceding work was whether AR patient alleles of PITX2A that harbour mutations in the HD might display altered FOXC1-binding properties. To evaluate this hypothesis, we performed IP experiments in which Xpress-FOXC1 was coexpressed along with WT, T68P, R91P, V83L or K88E alleles of HA-PITX2A. As observed in Figure 5, no gross differences in FOXC1 interaction were noted between WT PITX2A and the various patient mutants. Therefore, qualitatively disrupted PITX2A–FOXC1 interactions were not an apparent feature of the patient PITX2A alleles tested in vitro.


Figure 0085
View larger version (44K):
[in this window]
[in a new window]
  		Figure 5. FOXC1 interactions with PITX2A mutant alleles. IP experiments of COS-7 cell extracts transfected with Xpress-FOXC1 along with WT and mutant HA-PITX2 alleles. Extracts were immunoprecipitated with anti-Xpress antibodies followed by immunoblotting with anti-HA antibodies.

 
Subnuclear localization of PITX2 and FOXC1
The subnuclear localization of both WT and mutant PITX2A proteins was analysed by indirect immunofluorescence to determine whether any of the mutants might compromise the PITX2A–FOXC1 interaction via an altered protein targeting in vivo. The localization patterns of Xpress-PITX2A, FOXC1-GFP fusion and chromatin (stained with Hoechst 33342) were compared in COS-7 cells. These analyses demonstrated that FOXC1-GFP and WT PITX2A were indeed highly colocalized within the cell nucleus (Fig. 6). Both proteins are confined to a subset of Hoechst-stained chromatin and are excluded from the largest, densest Hoechst foci. The high degree of colocalization observed for PITX2 and FOXC1 confirmed that both proteins are indeed available to each other for interaction in the context of the cell nucleus.


Figure 0086
View larger version (57K):
[in this window]
[in a new window]
  		Figure 6. Colocalization of FOXC1 and PITX2 within a subdomain of the cell nucleus. (A) COS-7 cells seeded onto glass coverslips were transfected with pGFP-N1 (empty vector or WT FOXC1) along with Xpress-PITX2A (WT or indicated mutant construct). Immunofluorescence was performed following a 3 day expression period. Xpress-tagged PITX2 proteins were visualized with mouse anti-Xpress primary antibodies followed by Cy3-conjugated anti-Mouse IgG secondary antibodies. The cells were stained with Hoescht 33342 to visualize the nucleus. Two nuclei are shown to illustrate the continuum of cell phenotypes found in the R84W and K88E cultures. Scale bar: 5 mm. (B) Line scan plots present the density of each channel along the white diagonal line across the cell nucleus.

 
Immunofluorescence also revealed that mutations found in AR patients of the PITX2 HD may interfere with the normal targeting of PITX2A to FOXC1-containing chromatin. With the exception of the hyperactive V83L allele, each of the AR-associated PITX2A mutations reduced colocalization with both FOXC1-GFP and bulk chromatin to varying extent (Fig. 6A). In particular, the non-DNA-binding T68P and R91P mutants displayed clearly reduced colocalization with FOXC1-GFP and tended to occupy the area that did not stain with Hoechst in the nucleus. The same was observed, although less uniformly, for the dominant negative K88E and mild hypomorph R84W proteins. The latter two proteins varied somewhat in appearance between nuclei and formed Hoechst-excluding aggregates only in a subset of cells. In other nuclei, K88E and R84W proteins remained perfectly colocalized with FOXC1-GFP. The two nuclei expressing K88E and R84W shown in Figure 6A therefore represent the extremes of the continuum of distributions observed for these two mutant proteins. Line scan data graphically present the distribution of FOXC1-GFP and PITX2 immunoreactivity throughout the cell nucleus and indicate the extent of colocalization between FOXC1 and PITX2 proteins (Fig. 6B).

As PITX2 mutations are genetically dominant, AR patients retain one WT allele of the gene, and heterodimeric interactions between mutant and WT PITX2A are possible. The in vivo localization defect of mutant PITX2A may therefore be mitigated in trans by interaction with the WT protein. Similarly, mutant PITX2A might interfere with the activity of the WT protein by altering its normal subnuclear targeting. To test these two hypotheses, we simultaneously examined the subnuclear distribution of WT HA-PITX2A and FOXC1-GFP when cotransfected with cDNA constructs encoding either WT or mutant Xp-PITX2A in COS-7 cells (Fig. 7). In general, this experiment produced observations similar to the simple PITX2A versus FOXC1-GFP immunofluorescence experiment discussed previously. The T68P and R91P loss of function mutant proteins colocalized poorly with the WT HA-PITX2A or FOXC1-GFP, whereas the hypermorphic V83L allele was indistinguishable from WT. The K88E and R84W proteins, when coexpressed with WT PITX2A, were more uniformly colocalized with FOXC1-GFP than when expressed alone, suggesting that some degree of rescue of the altered subnuclear localization may have occurred. In no case was mutant PITX2A observed to reduce the colocalization of WT HA-PITX2A with FOXC1-GFP, suggesting the absence of a dominant negative effect.


Figure 0087
View larger version (81K):
[in this window]
[in a new window]
  		Figure 7. Mutant PITX2A does not affect the subnuclear localization of WT PITX2A or FOXC1. Indirect immunofluorescence of COS-7 cultures transfected with pGFP-N1 (FOXC1 or empty vector), Xpress-PITX2A (WT or indicated mutant construct) and WT HA-PITX2. Xpress-tagged PITX2 proteins were visualized with mouse anti-Xpress primary antibodies followed by Cy3-conjugated anti-mouse IgG secondary antibodies. HA-tagged proteins were detected by rabbit anti-HA primary antibodies followed by incubation with AMCA-conjugated anti-rabbit IgG secondary antibodies. Scale bar: 5 mm.

 
Lastly we explored the influence of FOXC1 naturally occurring patient mutations that disrupt its proper nuclear and subnuclear localization on the relative distribution of PITX2A. For these experiments, we cotransfected WT HA-PITX2A along with WT, R127H or S131L alleles of FOXC1 fused to GFP. These two mutant FOXC1 alleles display impaired nuclear and subnuclear localization when compared with WT FOXC1 [(24Go); F. Berry and M. Walter, unpublished data]. As evident in Figure 8, both HA-PITX2A and WT FOXC1-GFP displayed a uniform colocalization, whereas the degree of colocalization between HA-PITX2A and R127H or S131L FOXC1-GFP is largely disrupted.


Figure 0088
View larger version (40K):
[in this window]
[in a new window]
  		Figure 8. Mutant FOXC1 alleles do not disrupt PITX2A subnuclear localization. HA-PITX2A transfected into COS-7 cells, along with WT FOXC1, R127H and S131L mutant alleles fused to GFP, was visualized by indirect immunofluorescence. Scale bar: 5 mm.

 
PITX2A impairs FOXC1 transcriptional activation potential
Our findings to this point indicate that PITX2A and FOXC1 are common components of a higher order transcription factor complex; however, the functional significance of this interaction is not known. Therefore, we tested the transcriptional regulatory potential of PITX2A and FOXC1 complexes, both WT and mutant, to assay the biological ramification of a FOXC1–PITX2A interaction. First, we examined the influence of PITX2A on FOXC1 transcriptional activation of a FOXC1-responsive reporter gene. As indicated in Figure 9A, FOXC1 activated this reporter gene over 100-fold compared with an empty expression vector. PITX2A displayed a slight activation of the FOXC1-reporter, but less than 10% of the activity possessed by FOXC1. However, when PITX2A was cotransfected along with FOXC1, activation of the reporter gene by FOXC1 was markedly impaired. Similar results were obtained with both RL-TK and CMV-ßGal control vectors, indicating that differences in transfection efficiencies did not account for the reduced FOXC1 transactivation response. Furthermore, levels of FOXC1 protein and its ability to bind DNA were not affected by PITX2A expression (Fig. 9A; data not shown). Next, we examined whether PITX2A mutations influenced the suppression of FOXC1 transactivation. Again, WT FOXC1 was cotransfected into HeLa cells along with WT and mutant PITX2A alleles and the FOXC1-reporter gene activity was assessed (Fig. 9B). Only the hypermorphic V83L PITX2A mutation impaired FOXC1 activity to levels observed with WT PITX2A. The remainder of PITX2A mutants, all of which are defective in transcriptional activity relative to WT, either had no effect on FOXC1 activity (K88E and R91P) or elevated the level of FOXC1 transcriptional regulatory activity (T68P). Finally, we assessed what impact FOXC1 and its mutated alleles may have on PITX2A transcriptional potential of a PITX2-responsive reporter gene. As indicated in Figure 9C, the WT, R127H and S131L alleles of FOXC1 had little effect on the PITX2A transcriptional response.


Figure 0089
View larger version (15K):
[in this window]
[in a new window]
  		Figure 9. PITX2A attenuates transcriptional activation by FOXC1. (A) Reporter gene assays from HeLa cells transfected with empty pCI-HA (vector), Xp-FOXC1 or HA-PITX2 along with a FOXC1-BS-luciferase reporter. To control for transfection efficiencies, cells were cotransfected with RL-TK or CMV-ß-Gal vectors. Luciferase activities were normalized to activities of transfection control and data are presented relative to WT FOXC1 activity. Error bars represent the standard error of the mean (SEM). Relative expression levels for Xp-FOXC1 and HA-PITX2A are presented subsequently. (B) FOXC1 transactivation activity was examined in the presence of WT and mutant HA-PITX2A alleles in HeLa cells. An RL-TK vector was included as a transfection control. Luciferase activity is presented relative to FOXC1 alone. Data represent the average from three independent experiments transfected in triplicate. Error bars indicate the SEM. (C) PITX2A transcriptional activation assays. HeLa cells were cotransfected with HA-PITX2 along with WT FOXC1, R127H and S131L alleles and the activity of a luciferase reporter containing a single PITX2 binding site upstream of the TK promoter was examined. Luciferase activity is expressed relative to PITX2A alone.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
PITX2 and FOXC1 physically interact
Transcription factors do not act alone. Often, these proteins rely on protein–protein interactions for their binding to their target sites on DNA, for the remodelling or modification of the chromatin backbone and for the engagement of the RNA polymerase II complex and the initiation of transcription. In this article, we have demonstrated that the forkhead box FOXC1 and the HD PITX2 transcription factors are coexpressed in discrete cell types during eye development, that they can physically interact and that they share a common subnuclear compartment. We have also demonstrated a functional consequence of this interaction as PITX2A inhibits transcriptional regulatory activity by FOXC1. Mutations in FOXC1 or PITX2 produce a similar human phenotype, AR, that includes malformations of tissues derived from the periocular mesenchyme. Together, these results suggest that FOXC1 and PITX2 are components of a higher order transcription factor complex that is crucial for the proper development and function of the anterior segment of the eye. The FOXC1–PITX2A interaction is mediated by important functional domains on each protein. The HD of PITX2A was deemed necessary for a FOXC1 interaction. Two other regions in the C-terminal of PITX2A influenced FOXC1–PITX2A interactions. We hypothesize that these domains may be required to maintain the proper PITX2A structure to allow an interaction with FOXC1. The FOXC1-related forkhead protein, FOXA2, has been demonstrated to interact with a number of HD transcription factors. Similar to the interaction we observed between PITX2 and FOXC1, the interaction of HD proteins with FOXA2 is maintained through the HD, but additional regions in the HD proteins modulate the interaction with FOXA2 (32Go).

Functional interactions between other FOX and HD proteins have been documented, and depending on the cellular context, these interactions may be cooperative or inhibitory. For example, HOXA5 can repress FOXO1A activation of the insulin growth factor-binding protein promoter in HepG2 cells, but in HuF cells, HOXA5 and FOXO1A synergistically activate the IGFBP promoter (33Go). Oct4 can similarly repress the FOXD3-dependent activation of the FOXA2 promoter (34Go). In our studies, PITX2A suppressed the ability of FOXC1 to regulate a reporter gene containing FOXC1 consensus-binding sites in HeLa cells. PITX2A did not affect the binding of FOXC1 to its cognate DNA target in electrophoretic mobility shift assays, nor did PITX2A bind to this sequence (data not shown). It is possible that PITX2 binding to the C-terminal activation domain of FOXC1 simply masked the recruitment of coactivator complexes to the region of FOXC1. However, many of the mutant PITX2A alleles that we tested did not impair FOXC1 activity but were able to bind to FOXC1. Transcriptionally inactive alleles of PITX2A (T68P, K88E and R91P) did not affect FOXC1 transcriptional activity, whereas WT or a hypermorphic (V83L) alleles did. These data suggest a transcriptionally competent PITX2A allele is required to impair FOXC1 activity. Alternatively, as mutations in PITX2 abrogated the subnuclear colocalization with FOXC1, it is possible that the mutant proteins are simply unavailable at sufficient quantities to bind and impair FOXC1 activity.

Developmental ramifications for PITX2–FOXC1 interactions
The genetic inactivation of Pitx2 or Foxc1 in mice disrupts the formation of certain common structures in the anterior segment of the eye. Therefore, it is not surprising that FOXC1 and PITX2 proteins are colocalized to regions of the periocular mesenchyme that will eventually form these structures. This is the first study to report the colocalization of two transcription factors, PITX2 and FOXC1, that are essential for anterior segment development and human disease. The fact that both FOXC1 and PITX2 display a general but not complete overlap in expression patterns in the developing eye has important functional implications for these proteins. First, the formation of certain structures of the anterior segment may be dependent on either FOXC1 or PITX2 acting independently. Secondly, the degree of coexpression appears to be variable across the population of cells at each time point, with some cells exhibiting greater or lesser degrees of immunostaining for PITX2 versus FOXC1. This is reminiscent of variable expression of Pitx2 and the related Pitx1 gene in the pituitary gland (35Go) and suggests that there may be quantitative differences in the relative expression levels of both proteins in individual cells. Given the inhibitory influence of PITX2 on FOXC1 transcriptional activity, this may imply that context-specific regulation of FOXC1 target genes is occurring in the developing anterior segment. For instance, increased levels of PITX2 relative to FOXC1 may convert FOXC1-containing complexes from activator to repressor, providing a model for why rigorous quantitative control of the activities of both proteins is required for a correct FOXC1–PITX2 transcriptional response, which is entirely consistent with the genetics of ARS. Upstream elements and transcription factors that drive the temporal-spatial expression patterns of FOXC1 and PITX2 in the presumptive anterior segment are not known. Their identity will be important in the elucidation of mechanisms that can give rise to the discrete patterns of FOXC1 and PITX2 activity and allow for precise regulation of FOXC1 and PITX2 target genes by controlling the stoichiometry of this regulatory complex. Interestingly, retinoic acid has been recently implicated in playing a role upstream of both Foxc1 and Pitx2 during early eye development (36Go), consistent with the coordinated expression of both genes in the developing eye.

PITX2 and FOXC1 may potentiate mesenchymal responses to signals required for normal migration and condensation into endothelia such as those of the cornea and TM. Elaboration of the anterior chamber angle is likely to involve the action of both soluble chemotactic agents and mechanical messages transduced by the cytoskeleton and extracellular matrix during migration. As PITX2 and FOXC1 have been shown to respond to distinct types of extracellular signals, interaction of these two proteins may allow the cell to coordinate distinct suites of developmentally important stimuli. Because PITX2 and FOXC1 are coexpressed in the periocular mesenchyme prior to migration and formation of the anterior segment and because mutations of either gene produce equivalent AR spectrum phenotypes, it is reasonable to presume that these genes together govern the response of the presumptive anterior chamber angle to morphogenetic signals. Finally, as Pitx2 is crucial for the proper development of a number of non-ocular tissues including the heart, pituitary, branchial arches and ventral body wall (37Go,38Go), our findings that PITX2 can negatively modulate the activity of transcription factors through protein–protein interactions may reflect a novel paradigm for PITX2 function in the development of these tissues.

Disease ramifications of PITX2–FOXC1 interactions
The existence of a functional PITX2–FOXC1 complex has important implications for the molecular biology of AR malformations in patients. Most obvious of these is that coordinated function in a common complex may be a root cause of the clinically indistinguishable phenotypes of human PITX2 and FOXC1 mutations. Hypo- and hypermorphic mutations, deletions and duplications of both PITX2 and FOXC1 genes demonstrate that both genes are required at a strictly enforced level during normal development of the anterior segment (3Go,8Go,9Go,29Go–31Go,37Go). Simple haploinsufficiency is a plausible explanation for the pathogenicity of PITX2 microdeletions and loss of function mutations such as T68P and R91P. However, this term is of limited value in the construction of a model for dosage sensitivity and offers no insight into the essentially identical phenotype caused by duplications or hyperactive mutations of FOXC1 and PITX2, respectively. Our understanding of the molecular basis of AR malformations necessarily involves a model that explains the phenotypes of both increased and decreased PITX2 and FOXC1 activity in terms of a common molecular pathology. In this regard, protein interactions offer an explanation for the strict dosage sensitivity of PITX2 and FOXC1, given that PITX2–PITX2 and PITX2–FOXC1 complexes may form in a concentration- and/or cofactor-dependent fashion. Variations in the expression of either gene may alter the relative abundance of both types of complex. If PITX2–PITX2 and PITX2–FOXC1 complexes are mutually exclusive, they may interact with distinct cofactors and/or activate differing suites of target genes. Alternatively, PITX2–PITX2 and PITX2–FOXC1 complexes may compete for a common pool of transcriptional cofactors and/or binding sites.

Given the complex regulatory capacity of the FOXC1–PITX2 interaction, the molecular consequence of FOXC1 or PITX2 mutations is the transcriptional dysregulation of a subset of target genes rather than a simple loss of target gene expression by either PITX2 or FOXC1 alone. Thus, in a particular cellular context, mutations in FOXC1 or PITX2 could result in a loss of transcriptional activation or a loss of transcriptional repression of target genes. For example, in a tissue expressing both FOXC1 and PITX2, the consequence of a PITX2 loss of function mutation would be the impaired activation of PITX2-target genes coupled with the dysregulation of FOXC1-target genes in response to a mutant PITX2–FOXC1 complex (Fig. 10). As both loss of function mutations and gene duplications of FOXC1 result in a similar anterior segment dysgenesis phenotypes, there are stringent upper and lower critical thresholds for FOXC1 activity. Therefore, an ancillary function of PITX2 is the strict control of FOXC1 activities, and the dichotomous role of PITX2 is a likely explanation of the observation that mutations to PITX2 often result in more severe AR phenotypes. Mice heterozygous for Pitx2 mutations display more severe ocular developmental defects compared with Foxc1 heterozygotes (30Go,31Go,37Go). Moreover, patients with PITX2 mutations display a higher frequency of corectopia and polycoria compared with patients with FOXC1 mutations (Strungaru and Walter, unpublished data).


Figure 00810
View larger version (27K):
[in this window]
[in a new window]
  		Figure 10. Summary of the functional significance of PITX2A–FOXC1 interactions. (A and B) In cells expressing either PITX2A or FOXC1 alone, the respective target genes will be activated, whereas the respective counterpart FOXC1 or PITX2A target will not be expressed. (C) In cells expressing both WT FOXC1 and PITX2A proteins, PITX2-targets continue to be activated, whereas FOXC1-dependent targets are inhibited by the PITX2A–FOXC1 complexes. (D) Cell expressing loss of function PITX2A alleles will exhibit inhibited expression of PITX2-target genes and the deregulated expression of FOXC1 targets normally inhibited by the FOXC1–PITX2A complex.

 
Delayed onset of glaucoma
Approximately 50% of AR patients will develop elevated IOP and glaucoma. Because variable developmental defects of the angle of the anterior segment of the eye are seen in AR patients, physical occlusion of the angle is a potential mechanism for glaucoma pathogenesis. However, AR-associated glaucoma is an early onset rather than a congenital form; dysfunction of the mature TM, rather than mere physical occlusion, may be a contributing factor. Because both development and mature function of the TM are likely to invoke a combination of mechanical and soluble signals, PITX2/FOXC1 complexes might reasonably be expected to mediate either or both of anterior chamber development and aqueous drainage in the mature eye. The variability of anterior angle malformations and progression to glaucoma exhibited in individuals with FOXC1 or PITX2 mutations may also be explained, in part, by stochastic events. For instance, the chance of the formation of PITX2/FOXC1 complexes with mutant and/or WT subunits might influence the developmental behaviour of individual cells within the periocular mesenchyme, leading to sectors of the mature TM that form less or more markedly abnormally. We propose that a hierarchy of target genes for transcriptional regulation by the FOXC1–PITX2 complex exists and that the correct spatio-temporal regulation of these targets is required for normal formation and function of the anterior segment of the eye. The elucidation of these jointly regulated targets will aid in our understanding of glaucoma pathogenesis.

In summary, we have identified a functional link between two transcription factors, FOXC1 and PITX2, which we propose underpins the similar AR phenotype caused by mutations of these genes. FOXC1 and PITX2A physically interact and share a common subnuclear compartment. Furthermore, PITX2A can function as a negative regulator of FOXC1 transactivity. This work ties both proteins into a common pathway and offers an explanation for why increased FOXC1 gene dosage produces a phenotype resembling that of PITX2 deletions and mutations. Ocular phenotypes arise despite the deregulated expression of FOXC1-target genes through mutations in FOXC1 or PITX2. Ultimately, PITX2 loss of function mutations have a compound effect: the reduced expression of PITX2 target genes coupled with the extensive activation of FOXC1-regulated targets. This novel activity of PITX2 contributes in the stringent regulation of FOXC1 dosage in the eye and underlies gene dosage effects of these ocular transcription factors.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Plasmids and reagents
The WT N-terminal HA-tagged PITX2A protein isoform was expressed from a cDNA carried in the pCI plasmid (Promega, Madison, WI, USA) and was constructed by subcloning the EcoRI/XbaI fragment from the pcDNA4:PITX2 vector (13Go) downstream from the HA-epitope sequence (5'-ATG GCT TCT AGC TAT CCT TAT GAC GTG CCT GAC TAT GCC AGC CTG GGA GGA CCT TCT-3') between the NheI/XbaI sites in pCI. The deletions of the PITX2A cDNA were generated using the QuickChange® Site-Directed Mutagenesis Kit (Stratagene). Sequence of the mutagenesis primers are available upon request. All vectors were sequenced to verify correct open-reading frames. The Xpress-tagged FOXC1, GFP-FOXC1 and the pET-FOXC1 expression constructs were described previously (23Go,39Go,40Go).

Timed pregnancies and immunohistochemistry
Timed pregnancies were produced by mating C57BL/6J male and female mice from our existing colony. We designated the morning in which a plug was detected as embryonic day (E0.5). Embryos were collected by C-section after euthanasia of the mother. All procedures using mice were approved by the University of Michigan Committee on Use and Care of Animals and were conducted in accordance with the principles and procedures outlined in the NIH Guidelines for the Care and Use of Experimental Animals.

Embryos were fixed for 2–4 h with 4% paraformaldehyde in phosphate-buffered saline (PBS), washed and dehydrated and embedded into paraffin. Mounted sections were deparaffinized and treated for antigen retrieval by boiling for 10 min in the citrate buffer (pH 6.0). Immunostaining was performed using standard methods. Briefly, sections were incubated with antibodies directed against FOXC1 (Abcam, Inc.) and PITX2 (29Go), followed by biotinylated species-specific secondary antibodies (Jackson Immuno Research). Signals were detected using tyramide signal amplification kits (Perkin Elmer).

Cell culture and transfections
COS-7 and HeLa cells were maintained in Dulbecco's modified Eagle's medium (Invitrogen) containing 10% (v/v) fetal bovine serum and 1% (v/v) antimycotic (Invitrogen) at 37°C. Transfections were performed on cells subcultured the previous day using Fugene6 (Roche) following the manufacturer's protocol. Three microlitres of Fugene6 reagent was used for each microgram of DNA transfected.

Immunoprecipitation
Transfected cells were harvested 3 days after transfection to allow optimal expression of recombinant proteins. Plates were washed in PBS at 25°C prior to harvest. All subsequent steps were performed at 4°C in the presence of the mammalian protease inhibitor cocktail (5 µl/ml; Sigma Aldrich). Briefly, cells were scraped from 10 cm dishes in 1 ml of PBS with sterile cell lifters (Corning) and pelleted via microcentrifugation at 3000g for 10 min. The pellet supernatant was removed and a whole-cell lysate obtained by suspending the pellet in 200 µl RIPA buffer [1x PBS, 1% (v/v) IGEPAL CA-630, 0.5% (w/v) sodium deoxycholate, 0.1% (w/v) SDS]. Samples were vortexed briefly and incubated on ice for 45 min to allow maximal nuclear lysis. Lysates were cleared of cell debris by microcentrifugation at 18 000g for 10 min and total protein concentration was calculated (Bio-Rad). Lysate concentrations were equalized in a fixed volume of RIPA prior to immunoblot analysis of immunoprecipitation inputs. Immunoprecipitation experiments were performed as previously described (40Go), but all washes were performed in the RIPA buffer.

Nickel agarose-pull downs
COS-7 extracts (80 or 150 g) expressing HA-tagged PITX2A constructs were precleared with 50 l of Ni-NTA agarose suspension (Qiagen) at 4EC for 1 h, then incubated with purified 6xHis-tagged protein-bead suspensions (~1–4 g of full-length protein per reaction; unused beads were added when required for volumetric equalization of different samples) at 4EC for 1 h, followed by four washes with wash buffer. The beads were resuspended with SDS–PAGE loading buffer and loaded onto 15% SDS–PAGE gels. The samples were analysed by immunoblot analysis with antibodies directed against the mammalian-expressed recombinant proteins. COS-7 extracts were equilibrated by immunoblot analysis so that equivalent amounts of recombinant protein could be used in the pulldown reactions.

Immnunofluoresence
Immunofluorescence was performed 3 days following the transfection of COS-7 cells plated onto sterile coverslips (1.67x105 cells per 25 mm well). Briefly, the cells were fixed for 10 min in PBS containing 4% w/v paraformaldehyde and subsequently permeated in PBS containing 0.1% (v/v) Triton X-100. Coverslips were blocked in PBS containing 5% (w/v) bovine serum albumin (BSA) for 1 h. Primary and secondary antibody incubations as well as washes were conducted in PBS and 1% (w/v) BSA for 1 h at room temperature. Bisbenzimide 33342 (Hoechst) staining, if performed, was accomplished by adding 50 g/ml Hoechst to the secondary antibody dilution. Coverslips were mounted onto precleaned glass slides in freshly prepared mountant (9:1 glycerol:PBS, 5 mM p-phenylenediamine) and examined by indirect epifluorescence using a Leica DMRE microscope with PL-FLUOTAR 100x oil objective (Leica). Pixel intensities were obtained using ImageJ (http://rsb.info.nih.gov/ij) to create line scan plots for FOXC1-GFP and PITX2 immunofluorescence.

Transactivation assays
HeLa cells were plated into 24-well tissue culture plates at a density of 2x104 cells per well. Cells were cotransfected with 100 ng of WT or mutant expression vectors, Xpress-FOXC1 and HA-PITX2A or empty expression vectors along with 20 ng reporter (6XFOXC1-BS-Luc or 1XBicoid-Luc, for monitoring FOXC1 and PITX2 activities, respectively) and 0.2 ng pRL-TK per well. Total amounts of DNA transfected were equalized with empty pcDNA or pCI-HA. Cells were harvested and assayed for luciferase activity 40 h after transfection. Unless otherwise stated, all transfections were performed in triplicate and repeated at least three times.


   ACKNOWLEDGEMENTS
 
We would like to thank Drs O. Lehmann and R. Wevrick and the members of the Ocular Genetics Laboratory at the University of Alberta for the critical reading of this manuscript. We also wish to thank May Yu for tissue culture assistance. This work was supported by operating grants awarded to M.A.W. from the Canadian Institutes for Health Research (CIHR) and to P.J.G. from the National Institutes of Health (EY014126, EY07003) and Research to Prevent Blindness.

Conflict of Interest statement. None declared.


   FOOTNOTES
 
{dagger} The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Alward, W.L. (2000) Axenfeld–Rieger syndrome in the age of molecular genetics. Am. J. Ophthalmol., 130, 107–115.[CrossRef][ISI][Medline]

  2. Lines, M.A., Kozlowski, K. and Walter, M.A. (2002) Molecular genetics of Axenfeld–Rieger malformations. Hum. Mol. Genet., 11, 1177–1184.[Abstract/Free Full Text]

  3. Mears, A.J., Jordan, T., Mirzayans, F., Dubois, S., Kume, T., Parlee, M., Ritch, R., Koop, B., Kuo, W.L., Collins, C. et al. (1998) Mutations of the forkhead/winged-helix gene, FKHL7, in patients with Axenfeld–Rieger anomaly. Am. J. Hum. Genet., 63, 1316–1328.[CrossRef][ISI][Medline]

  4. Winnier, G.E., Kume, T., Deng, K., Rogers, R., Bundy, J., Raines, C., Walter, M.A., Hogan, B.L. and Conway, S.J. (1999) Roles for the winged helix transcription factors MF1 and MFH1 in cardiovascular development revealed by nonallelic noncomplementation of null alleles. Dev. Biol., 213, 418–431.[CrossRef][ISI][Medline]

  5. Swiderski, R.E., Reiter, R.S., Nishimura, D.Y., Alward, W.L., Kalenak, J.W., Searby, C.S., Stone, E.M., Sheffield, V.C. and Lin, J.J. (1999) Expression of the Mf1 gene in developing mouse hearts: implication in the development of human congenital heart defects. Dev. Dyn., 216, 16–27.[CrossRef][ISI][Medline]

  6. Vaux, C., Sheffield, L., Keith, C.G. and Voullaire, L. (1992) Evidence that Rieger syndrome maps to 4q25 or 4q27. J. Med. Genet., 29, 256–258.[Abstract]

  7. Heon, E., Sheth, B.P., Kalenak, J.W., Sunden, S.L., Streb, L.M., Taylor, C.M., Alward, W.L., Sheffield, V.C. and Stone, E.M. (1995) Linkage of autosomal dominant iris hypoplasia to the region of the Rieger syndrome locus (4q25). Hum. Mol. Genet., 4, 1435–1439.[Abstract/Free Full Text]

  8. Semina, E.V., Reiter, R., Leysens, N.J., Alward, W.L., Small, K.W., Datson, N.A., Siegel-Bartelt, J., Bierke-Nelson, D., Bitoun, P., Zabel, B.U. et al. (1996) Cloning and characterization of a novel bicoid-related homeobox transcription factor gene, RIEG, involved in Rieger syndrome. Nat. Genet., 14, 392–399.[CrossRef][ISI][Medline]

  9. Nishimura, D.Y., Swiderski, R.E., Alward, W.L., Searby, C.C., Patil, S.R., Bennet, S.R., Kanis, A.B., Gastier, J.M., Stone, E.M. and Sheffield, V.C. (1998) The forkhead transcription factor gene FKHL7 is responsible for glaucoma phenotypes which map to 6p25. Nat. Genet., 19, 140–147.[CrossRef][ISI][Medline]

  10. Phillips, J.C., del Bono, E.A., Haines, J.L., Pralea, A.M., Cohen, J.S., Greff, L.J. and Wiggs, J.L. (1996) A second locus for Rieger syndrome maps to chromosome 13q14. Am. J. Hum. Genet., 59, 613–619.[ISI][Medline]

  11. Ferguson, J.G., Jr. and Hicks, E.L. (1987) Rieger's anomaly and glaucoma associated with partial trisomy 16q. Case report. Arch. Ophthalmol., 105, 323.[ISI][Medline]

  12. Cox, C.J., Espinoza, H.M., McWilliams, B., Chappell, K., Morton, L., Hjalt, T.A., Semina, E.V. and Amendt, B.A. (2002) Differential regulation of gene expression by PITX2 isoforms. J. Biol. Chem., 277, 25001–25010.[Abstract/Free Full Text]

  13. Kozlowski, K. and Walter, M.A. (2000) Variation in residual PITX2 activity underlies the phenotypic spectrum of anterior segment developmental disorders. Hum. Mol. Genet., 9, 2131–2139.[Abstract/Free Full Text]

  14. Saadi, I., Semina, E.V., Amendt, B.A., Harris, D.J., Murphy, K.P., Murray, J.C. and Russo, A.F. (2001) Identification of a dominant negative homeodomain mutation in Rieger syndrome. J. Biol. Chem., 276, 23034–23041.[Abstract/Free Full Text]

  15. Priston, M., Kozlowski, K., Gill, D., Letwin, K., Buys, Y., Levin, A.V., Walter, M.A. and Heon, E. (2001) Functional analyses of two newly identified PITX2 mutants reveal a novel molecular mechanism for Axenfeld–Rieger syndrome. Hum. Mol. Genet., 10, 1631–1638.[Abstract/Free Full Text]

  16. Erickson, R.P. (2001) Forkhead genes and human disease. J. Appl. Genet., 42, 211–221.[Medline]

  17. Carlsson, P. and Mahlapuu, M. (2002) Forkhead transcription factors: key players in development and metabolism. Dev. Biol., 250, 1–23.[CrossRef][ISI][Medline]

  18. Lehmann, O.J., Sowden, J.C., Carlsson, P., Jordan, T. and Bhattacharya, S.S. (2003) Fox's in development and disease. Trends Genet., 19, 339–344.[CrossRef][ISI][Medline]

  19. Berry, F.B., Tamimi, Y., Carle, M.V., Lehmann, O.J. and Walter, M.A. (2005) The establishment of a predictive mutational model of the forkhead domain through the analyses of FOXC2 missense mutations identified in patients with hereditary lymphedema with distichiasis. Hum. Mol. Genet., 14, 2619–2627.[Abstract/Free Full Text]

  20. Lehmann, O.J., Ebenezer, N.D., Jordan, T., Fox, M., Ocaka, L., Payne, A., Leroy, B.P., Clark, B.J., Hitchings, R.A., Povey, S. et al. (2000) Chromosomal duplication involving the forkhead transcription factor gene FOXC1 causes iris hypoplasia and glaucoma. Am. J. Hum. Genet., 67, 1129–1135.[ISI][Medline]

  21. Nishimura, D.Y., Searby, C.C., Alward, W.L., Walton, D., Craig, J.E., Mackey, D.A., Kawase, K., Kanis, A.B., Patil, S.R., Stone, E.M. et al. (2001) A spectrum of FOXC1 mutations suggests gene dosage as a mechanism for developmental defects of the anterior chamber of the eye. Am. J. Hum. Genet., 68, 364–372.[CrossRef][ISI][Medline]

  22. Lehmann, O.J., Ebenezer, N.D., Ekong, R., Ocaka, L., Mungall, A.J., Fraser, S., McGill, J.I., Hitchings, R.A., Khaw, P.T., Sowden, J.C. et al. (2002) Ocular developmental abnormalities and glaucoma associated with interstitial 6p25 duplications and deletions. Invest. Ophthalmol. Vis. Sci., 43, 1843–1849.[Abstract/Free Full Text]

  23. Saleem, R.A., Banerjee-Basu, S., Berry, F.B., Baxevanis, A.D. and Walter, M.A. (2001) Analyses of the effects that disease-causing missense mutations have on the structure and function of the winged-helix protein FOXC1. Am. J. Hum. Genet., 68, 627–641.[CrossRef][ISI][Medline]

  24. Saleem, R.A., Banerjee-Basu, S., Berry, F.B., Baxevanis, A.D. and Walter, M.A. (2003) Structural and functional analyses of disease-causing missense mutations in the forkhead domain of FOXC1. Hum. Mol. Genet., 12, 2993–3005.[Abstract/Free Full Text]

  25. Murphy, T.C., Saleem, R.A., Footz, T., Ritch, R., McGillivray, B. and Walter, M.A. (2004) The wing 2 region of the FOXC1 forkhead domain is necessary for normal DNA-binding and transactivation functions. Invest. Ophthalmol. Vis. Sci, 45, 2531–2538.[Abstract/Free Full Text]

  26. Mucchielli, M.L., Mitsiadis, T.A., Raffo, S., Brunet, J.F., Proust, J.P. and Goridis, C. (1997) Mouse Otlx2/RIEG expression in the odontogenic epithelium precedes tooth initiation and requires mesenchyme-derived signals for its maintenance. Dev. Biol., 189, 275–284.[CrossRef][ISI][Medline]

  27. Gage, P.J. and Camper, S.A. (1997) Pituitary homeobox 2, a novel member of the bicoid-related family of homeobox genes, is a potential regulator of anterior structure formation. Hum. Mol. Genet., 6, 457–464.[Abstract/Free Full Text]

  28. Kitamura, K., Miura, H., Yanazawa, M., Miyashita, T. and Kato, K. (1997) Expression patterns of Brx1 (Rieg gene), Sonic hedgehog, Nkx2.2, Dlx1 and Arx during zona limitans intrathalamica and embryonic ventral lateral geniculate nuclear formation. Mech. Dev., 67, 83–96.[CrossRef][ISI][Medline]

  29. Hjalt, T.A., Semina, E.V., Amendt, B.A. and Murray, J.C. (2000) The Pitx2 protein in mouse development. Dev. Dyn., 218, 195–200.[CrossRef][ISI][Medline]

  30. Kume, T., Deng, K.Y., Winfrey, V., Gould, D.B., Walter, M.A. and Hogan, B.L. (1998) The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus. Cell, 93, 985–996.[CrossRef][ISI][Medline]

  31. Kidson, S.H., Kume, T., Deng, K., Winfrey, V. and Hogan, B.L. (1999) The forkhead/winged-helix gene, Mf1, is necessary for the normal development of the cornea and formation of the anterior chamber in the mouse eye. Dev. Biol., 211, 306–322.[CrossRef][ISI][Medline]

  32. Foucher, I., Montesinos, M.L., Volovitch, M., Prochiantz, A. and Trembleau, A. (2003) Joint regulation of the MAP1B promoter by HNF3beta/Foxa2 and Engrailed is the result of a highly conserved mechanism for direct interaction of homeoproteins and Fox transcription factors. Development, 130, 1867–1876.[Abstract/Free Full Text]

  33. Foucher, I., Volovitch, M., Frain, M., Kim, J.J., Souberbielle, J.C., Gan, L., Unterman, T.G., Prochiantz, A. and Trembleau, A. (2002) Hoxa5 overexpression correlates with IGFBP1 upregulation and postnatal dwarfism: evidence for an interaction between Hoxa5 and Forkhead box transcription factors. Development, 129, 4065–4074.[Abstract/Free Full Text]

  34. Guo, Y., Costa, R., Ramsey, H., Starnes, T., Vance, G., Robertson, K., Kelley, M., Reinbold, R., Scholer, H. and Hromas, R. (2002) The embryonic stem cell transcription factors Oct-4 and FoxD3 interact to regulate endodermal-specific promoter expression. Proc. Natl Acad. Sci. USA, 99, 3663–3667.[Abstract/Free Full Text]

  35. Charles, M.A., Suh, H., Hjalt, T.A., Drouin, J., Camper, S.A. and Gage, P.J. (2005) PITX genes are required for cell survival and Lhx3 activation. Mol. Endocrinol., 19, 1893–1903.[Abstract/Free Full Text]

  36. Matt, N., Dupe, V., Garnier, J.M., Dennefeld, C., Chambon, P., Mark, M. and Ghyselinck, N.B. (2005) Retinoic acid-dependent eye morphogenesis is orchestrated by neural crest cells. Development, 132, 4789–4800.[Abstract/Free Full Text]

  37. Gage, P.J., Suh, H. and Camper, S.A. (1999) Dosage requirement of Pitx2 for development of multiple organs. Development, 126, 4643–4651.[Abstract]

  38. Kitamura, K., Miura, H., Miyagawa-Tomita, S., Yanazawa, M., Katoh-Fukui, Y., Suzuki, R., Ohuchi, H., Suehiro, A., Motegi, Y., Nakahara, Y. et al. (1999) Mouse Pitx2 deficiency leads to anomalies of the ventral body wall, heart, extra- and periocular mesoderm and right pulmonary isomerism. Development, 126, 5749–5758.[Abstract]

  39. Berry, F.B., Saleem, R.A. and Walter, M.A. (2002) FOXC1 transcriptional regulation is mediated by N- and C-terminal activation domains and contains a phosphorylated transcriptional inhibitory domain. J. Biol. Chem., 277, 10292–10297.[Abstract/Free Full Text]

  40. Berry, F.B., O'Neill, M.A., Coca-Prados, M. and Walter, M.A. (2005) FOXC1 transcriptional regulatory activity is impaired by PBX1 in a filamin A-mediated manner. Mol. Cell. Biol., 25, 1415–1424.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles: