SEARCH

SEARCH BY CITATION

Keywords:

  • Dlx5;
  • EphA7;
  • ephrinA5;
  • Msx2;
  • neural tube formation

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Homeodomain-containing transcription factors Dlx5 and Msx2 are able to form a heterodimer, and together can regulate embryonic development including skeletogenesis. Dlx5 functions as a transcriptional activator and Msx2 a transcriptional repressor, and they share common target genes. During mouse digit development, the expression domains of Dlx5 and Msx2 overlap at the distal region of the developing terminal phalange, although digit formation and regeneration are not altered in the Dlx5 and Msx2 null mutant embryos. Interestingly, we observed a high rate of defects in neural tube formation in Dlx5 and Msx2 double null mutants. In the absence of both Dlx5 and Msx2, a high occurrence of exencephaly and severe defects in craniofacial morphology are observed. Additionally, Dlx5 and Msx2 expression domain analysis showed overlap of the genes at the apex of the neural folds just prior to neural fold fusion. The expression patterns of ephrinA5 and two isoforms of EphA7 were tested as downstream targets of Dlx5 and Msx2. Results show that EphrinA5 and the truncated isoform of EphA7 are regulated by Dlx5 and Msx2 together, although the full length isoform of EphA7 expression is not altered. Overall, these data show that Dlx5 and Msx2 play a critical role in controlling cranial neural tube morphogenesis by regulating cell adhesion via the ephrinA5 and EphA7 pathway.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Neurulation is a fundamental process that establishes the neural system during embryonic development. The neural tube is the major structure that forms during neurulation, and represents the anlagen of the brain and spinal cord. The neural tube is formed by the progressive fusion of the neural folds, which are elevated tissue structures along the lateral edges of the neural plate at the dorsal midline of the embryo. Approximately 80 genes are involved in a mammalian neurulation, and neural tube defects (NTDs) result with disruption of any of these genes (Copp et al. 2003). Anencephaly is one such NTD that is caused by a neural patterning defect and results from disrupted tube closure in the cranial region of the embryo. Anencephaly is the resulting phenotype when exencephalic brain tissue gradually degenerates due to the exposure to amniotic fluid (Timor-Tritsch et al. 1996). In humans, the occurrence rate of the NTDs is approximately 1 out of 1000–2000 births in the United States (Copp et al. 2003).

Among the genes that are involved in neurulation, Dlx (distal-less homeobox) and Msx (msh-like homeobox) family genes have been reported for their roles in craniofacial embryogenesis, including anterior neural tube formation. Both the Dlx and Msx gene families encode for homeodomain-containing transcription factors. In mammals there are six members of the Dlx family (Dlx1-6) and three members of the Msx family (Msx1-3) (Depew et al. 2005; Ramos & Robert, 2005). The Dlx gene family has been implicated in embryonic development including brain, branchial arches, jaws and limb development (Depew et al. 2005). In a Dlx5 gene null mutant study, it was demonstrated that approximately 24% of mutant embryos showed an exencephaly phenotype at E13.5 (Depew et al. 1999) and this phenotype is more severe in the Dlx5/6 double mutant, suggesting that these genes play a redundant role in anterior neural tube patterning (Robledo et al. 2002). In the case of the Msx gene family, all three Msx genes are expressed during neurulation (Shimeld et al. 1996; Ramos & Robert 2005), and based on knockout studies, it has been demonstrated that Msx1 and Msx2 regulate craniofacial patterning processes (Satokata & Maas, 1994; Satokata et al. 2000; Bach et al. 2003). While neither Msx1 nor Msx2 single gene mutation cause anterior neural tube closure defect, the majority of Msx1/2 double null mutant embryos fail to close the anterior neural pore and exhibit severe craniofacial abnormalities including exencephaly (Han et al. 2007). These studies suggest that both the Dlx and Msx gene families are critical for proper neural tube formation, and in both cases there is evidence that their family members serve redundant roles.

Previous studies have demonstrated that Msx and Dlx homeoprotein families form homo- and heterodimeric complexes and suggests that protein–protein interactions can be an essential molecular event in particular regulation of gene expressions (Zhang et al. 1997). Dlx5 is known to function as a transcriptional activator and Msx2 as a transcriptional repressor, and these genes directly regulate a number of target genes including Runx2 (Shirakabe et al. 2001; Kim et al. 2004), Osteocalcin (Newberry et al. 1998), and BMP2 induced alkaline phosphatase (Kim et al. 2004).

Membrane-bound GPI (glycosyl phosphatidylinositol)-anchored ligands ephrins and its receptor Ephs play key roles in diverse biological processes. Receptor Ephs are a subgroup of tyrosine protein kinase receptor family (Campbell & Robbins 2008; Klein 2009). The Eph receptors and their ephrins ligands are divided into A- and B-subclass based on their molecular structure and binding affinities (Frisen et al. 1999). To date, five type-A ephrins (ephrinA1-A5), three type-B ephrins (ephrinB1-B3), nine type-A Ephs (EphA1-A8 and A10), and five type-B Ephs (EphB1-B4 and EphB6) have been identified in mammals (Klein 2009). It has been described that members of A-subclass, ephrinA5 and EphA7 function in the neural fold fusion through cellular adhesion/repulsion. A subpopulation of ephrinA5 (17%) mutant mice, as well as EphA7 (24%) mutant, exhibit cranial neural tube malformation due to a neural tube fusion defect (Frisen et al. 1999). The receptor EphA7 has three splice variants; one full length form (EphA7-FL) and two truncated forms both of which lack a cytoplasmic tyrosine kinase domain (EphA7-T1, EphA7-T2, Ciossek et al. 1995; Valenzuela et al. 1995; Frisen et al. 1999). When ephrinA5 ligand on one cell is bound to EphA7-FL homo-dimer on an adjacent cell, these two cells repulse each other. However, if ephrinA5 binds to EphA7-FL/EphA7-T1 or T2 heterodimer, the two cells will adhere together. This report suggests that regulation of cell adhesion and repulsion processes by ephrinA5/EphA7 plays a critical role in controlling cranial neural tube formation (Frisen et al. 1999).

To date, although many studies unveiled the molecular mechanisms of cranial neural tube patterning from a genetic-based approach, many aspects of the molecular networks of cranial neural tube patterning remain unknown. In this report, we demonstrated that Dlx5/Msx2 double mutants display an increased rate of malformation in cranial neural tube formation as compared to Dlx5 or Msx2 single mutants. Our studies show that the frequency of exencephaly increases incrementally in a Dlx5 mutant background with decreasing Msx2 gene dose. In addition, ephrin-A5 and EphA7-T1 expression, but not EphA7-FL, were downregulated in Dlx5/Msx2 double mutant embryos at the dorsal region of neural tube in association with the failure of neural fold fusion. Our report provides a novel molecular mechanism in which Dlx5 and Msx2 function reciprocally through the regulation of ephrin-A5/EphA7 expression in cranial neural tube closure.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Wildtype and Dlx5, Mxs2 mutant mice

Wildtype mouse embryos used in this study were either outbred CD#1 strain supplied by Charles River Laboratories or wildtype embryos from the breeding of Dlx5 and Msx2 mutant mice. Homozygous Dlx5 or Msx2 mutant embryos were obtained by mating of heterozygotes carrying a targeted deletion of either the Dlx5 gene (Depew et al. 1999) or the Msx2 gene (Satokata et al. 2000). Dlx5/Msx2 double mutant embryos were obtained by double heterozygotes mating. Embryos were collected at embryonic day (E) 9.5 and their genotype was verified by polymerase chain reaction (PCR) with genotype specific PCR primers. Procedures for the care and use of mice for this study were compliant with standard operating procedures (SOPs) approved by the Institutional Animal Care & Use Committee (IACUC) of Tulane University Health Science Center.

Fetal mouse digit amputation

To study regeneration in vivo, fetal mouse digit tips were amputated at E14.5. Timed-pregnant mice, which carry E14.5 embryos were anesthetized with sodium pentobarbital (60 μg/g body weight), fentanyl (1.6 μg/animal), and droperidol (80 μg/animal). The pregnant mouse abdomen was opened with a mid-ventral incision, and fetuses were exposed by incision of the anti-placental uterine wall. Access to the hindlimb was gained through an incision in the extraembryonic membranes and the hindlimb was teased out with a blunt probe. The three central hindlimb digits, digits 2, 3, and 4, were amputated at a distal level, approximately 75 mm from the digit tip. The uterus with attached fetuses was re-positioned within the abdominal cavity, and the abdominal wall of female mouse was closed. Operated fetuses were allowed to develop for 4 days exo utero (Muneoka et al. 1986), after which the hindlimbs were collected for analysis of the digits.

Whole mount skeletal staining

Differential whole mount bone staining of mouse embryos was performed according to the following process. Embryos were isolated at E18.5 and fixed with 95% ethanol (EtOH) overnight. Embryos were then skinned manually, delipidated in acetone, and stained with Alcian Blue 8XG/Alizarin Red S in 5% acetic acid, 95% EtOH. Stained embryos were treated in 1% KOH and cleared by glycerol.

In situ hybridization

Digoxigenin-labeled antisense RNA probes for Dlx5, Msx2, EphrinA5, EphA7-FL, and EphA7-T1 were used to perform in situ hybridization. EphrinA5, EphA7-FL, and EphA7-T1 containing DNA plasmids were kindly provided by Dr Jonas Frisén. Embryos were collected at E 9.5 and fixed in 4% paraformaldehyde at 4°C overnight. Fixed mouse embryos were dehydrated with an ascending series of ethanol (25%, 50%, 75% and 100%), infiltrated in xylene, and embedded in paraffin. Paraffin sections were cut at 5 μm thickness. In situ hybridization was performed according to previous method (Han et al., 2003).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Dlx5 and Msx2 in mouse fetal digit regeneration

The transcriptional repressor Msx1 and cell signaling molecule Bmp4 are co-expressed at the apex of the forming fetal mouse digit and both have been implicated in the control of digit tip regeneration (Han et al. 2003). DLX and MSX proteins can form heterodimers that can regulate gene transcriptions (Zhang et al. 1997). Particularly, the DLX5 and MSX2 have been shown to form a heterodimeric complex that regulates differentiation during skeletogenesis (Newberry et al. 1998). To study the role of Dlx5 and Msx2 in digit regeneration we began by analyzing the expression of Dlx5 and Msx2 on mouse digit at E14.5. Expression of Dlx5 was detected in ectoderm and mesenchymal tissue between the epidermis and condensed cartilage of digit (Fig. 1A). The Msx2 expression pattern is similar to the expression of Dlx5 at the digit tip, but extended proximally (Fig. 1B). Since Msx2 and Dlx5 are co-expressed at the apex of the forming digit, we were interested in whether these two genes were involved in the control of fetal digit regeneration. The Msx2 mutant digit had previously been tested and was found to regenerate normally, thus suggesting that Msx2 was not required for digit tip regeneration (Han et al. 2003). Here we tested the regenerative capacity of Dlx5 mutant digits at E14.5, and as well, we re-tested the Msx2 mutant. We found that the Dlx5 and Msx2 mutant digits possessed a regenerative capacity similar to wildtype digits, thus indicating that neither gene was required for fetal digit tip regeneration (Table 1). To test the role of both Dlx5 and Msx2 in fetal digit regeneration, we generated Dlx5−/−;Msx2−/− double mutant embryos. E14.5 digits from this double mutant were tested for regenerative ability and their regenerative response was undistinguishable from wildtype control digits (Table 1). These studies demonstrate that despite their co-expression at the apex of the forming digit and their known interactions in regulating skeletogenesis, Dlx5 and Msx2 do not appear to play a functional role in fetal digit tip regeneration.

Table 1. Regeneration response of fetal digit tips
 Genotype
WTDlx5+/−Dlx5−/−Msx2−/−Dlx5+/−;Msx2−/−Dlx5−/−;Msx2−/−
  1. WT, wildtype.

Number of regenerated digit25/2746/4817/186/627/3030/30
(92.6%)(95.8%)(94.4%)(100%)(90.0%)(100%)
image

Figure 1. Expression domains of Dlx5 and Msx2 overlap in fetal digit tip. (A–B) Expression of Dlx5 (A) and Msx2 (B) are detected in ectoderm and mesenchymal tissue of the E14.5 fetal mouse digit tip. The expression domain of Dlx5 is restricted to the distal region of the digit (A), but Msx2 expression domain is extended more proximally (B).

Download figure to PowerPoint

Exencephaly phenotype in Dlx5 and Msx2 double mutant mouse embryo

During our studies on the role of Dlx5 and Msx2 in fetal digit tip regeneration, we noted that the most dramatic phenotype associated with the double mutant embryos was that many embryos displayed exencephaly. It has been already reported that 12% or 28% of Dlx5 null mutant embryos displayed an exencephalic phenotype (Acampora et al. 1999; Depew et al. 1999), but that exencephaly was not noted in Msx2 null embryos (Satokata et al. 2000). To characterize the exencephalic phenotype among the different genotypes of embryos developing from a double heterozygote cross, we collected embryos from stages ranging from E9.5 to E18.5 to establish the frequency of exencephaly. In this analysis we obtained and scored embryos with multiple genotypes, including wildtype, Dlx5−/−, Msx2−/−, Dlx5+/−;Msx2+/−, Dlx5+/−;Msx2−/−, Dlx5−/−;Msx2+/−, and Dlx5−/−;Msx2−/− (Table 2). Similar to previous studies, we found that 19% of Dlx5 null mutant embryos displayed exencephaly while Msx2 null mutants had no exencephalic embryos. Dlx5 heterozygote embryos display no exencephaly (Acampora et al. 1999; Depew et al. 1999); however, if the embryos lack either one or both copies of Msx2 they display a low level of exencephaly (Dlx5+/−;Msx2+/−: 7%; Dlx5+/−;Msx2−/−: 9%). Interestingly, in a Dlx5 mutant background, the frequency of exencephaly increased to 39% when one copy of Msx2 is absent, and when both copies are absent the exencephalic frequency jumps to 73%. These studies clearly show that the exencephaly phenotype associated with the Dlx5 mutant is influenced by Msx2 in a synergistic manner.

Table 2. Exencephalic phenotype ratio in different mouse genotypes
StageGenotype
Dlx5−/− Msx2−/− Dlx5+/−;Msx2+/− Dlx5+/−;Msx2−/− Dlx5−/−;Msx2+/− Dlx5−/−;Msx2−/−
  1. N/A, data not available.

E9.51/60/133/201/94/118/10
E10.51/20/50/170/73/82/5
E11.50/20/22/122/53/71/2
E12.50/20/50/150/73/51/1
E14.5N/A0/120/82/221/513/18
E18.51/40/10/30/5N/A5/5
Total3/160/385/755/5514/3630/41
(18.8%)(0%)(6.7%)(9.1%)(38.9%)(73.2%)

Since Dlx5 and Msx2 are both transcriptional regulators and are known to interact during skeletogenesis, we next analyzed skull formation of neonates in whole mount skeletal preparations of Msx2 mutants, and of Dlx5−/− and Dlx5−/−;Msx2−/− mutants displaying exencephaly at E18.5. Ossifying frontal, parietal, interparietal, and suppraoccipital bones are shown in the calvarium of wildtype embryos at E18.5 (Fig. 2A,B). Skull morphology of the Msx2 null mutant shows that skull size was slightly reduced in comparison to wildtype controls, and that ossification of interparietal and supraoccipital bones is delayed (arrows in Fig. 2E,F). On the other hand the frontal and parietal bones are not affected in the Msx2 mutant. The size of the calvarium of the Dlx5−/− exencephaly phenotype is grossly reduced, and all five fontanelle bones as well as the supraoccipital bone do not form (Fig. 2C,D). Similarly, in the Dlx5 and Msx2 double null mutant embryo the calvarium displays an identical morphology to the Dlx5 mutant (Fig. 2G,H). We also examined the whole mount skull sample of non-exencephaly embryos of Dlx5−/−;Msx2−/− mutants. Although the epidermis of the cranium is intact, frontal, parietal and interparitetal bones were only partially developed, and the suppraoccipital bones were missing (Fig. 2I,J). While the cranial phenotypes vary depending on genotype, gross morphologies of embryos are not significantly altered from wildtype (Fig. S1). These data show that the primary effect of removing copies of the Msx2 gene on the Dlx5 mutant is associated with the frequency of exencephaly and not the severity of phenotype. This finding, combined with evidence that the expression domains of Dlx5 and Msx2 do not overlap during the embryogenesis of the skull (Kim et al. 1998; Holleville et al. 2003) suggests that the exencephalic defect caused by the double mutation is linked to developmental events that precede skeletogenesis.

image

Figure 2. Gross skull morphology of wildtype (WT), Dlx5−/−, Msx2−/−, and Dlx5;Msx2−/− mutant mouse embryos. (A–J) At E18.5, differential skeletal staining images of embryo calvaria with lateral view (A, C, E, G, I) and dorsal view (B, D, F, H, J). (A, B) Wildtype. (C, D) Dlx5−/−. Fontanelle and supraoccipital bones are missing. (E, F) Msx2−/−. Ossification of interparietal and supraoccipital bones are delayed (arrows). (G, H) Exencephaly embryo of Dlx5−/−;Msx2−/−. The skull morphology is similar to Dlx5 single mutant. (I, J) Non-exencephaly embryo of Dlx5−/−;Msx2−/−. The frontal, parietal and interparietal bones are only partially developed, and the supraoccipital bone is missing. Black dot lines indicate the outline of the wildtype embryo skull. fr, frontal bone; ip, interparietal bone; pa, parietal bone; so, supraoccipital bone. Scale bars: 3 mm.

Download figure to PowerPoint

Dlx5 and Msx2 expression patterns during cranial neural tube formation

The cause of the exencephaly phenotype is generally linked to a failure of the anterior neural tube to close properly during neurulation early in embryogenesis (Copp et al. 2003). Double mutant embryos analyzed for exencephaly confirm that the phenotype is present in the early embryo (Table 2), and is associated with the failure of the anterior neural tube to close. To investigate the role of Dlx5 and Msx2 in neural tube closure we first carried out a detailed analysis of gene expression during neurulation. The neurulation process can be divided into four stages: (i) formation of neural plate; (ii) folding of the neural plate to form the neural groove; (iii) elevation of the neural folds; and (iv) closure of the neural folds to form the neural tube (Gilbert 2003).The cranial region of the mouse embryo undergoes its incipient neural groove stage at E8.5. At this time point, Dlx5 transcripts were not detected anywhere in the neural folds (Fig. 3A), whereas Msx2 expression was detected at the edges of the neural folds (black arrowheads, Fig. 3D). At E9.5, neural tube formation is completed in the cranial region of the embryo and both the neural fold elevation stage and neural tube stage can be observed in the same embryo analyzed at different cranial-caudal levels. Dlx5 is transiently expressed at the apex of the neural folds (black arrowheads, Fig. 3B); however, after closure of the neural tube Dlx5 expression is downregulated and is no longer detected in the neural tube (Fig. 3C). During this stage Msx2 transcripts are detected in the apex of neural folds in a region that overlaps the expression domain of Dlx5 (Fig. 3E). After closure of the neural tube Msx2 remains expressed at the point of fusion along the dorsal midline (black arrowheads, Fig. 3E,F). Summarizing, Msx2 is expressed in the apex of the neural folds and at the dorsal midline of the neural tube during and after fusion to close the anterior neural tube, whereas Dlx5 is transiently upregulated in the apex of the neural folds immediately prior to neural fold fusion and downregulated after fusion (Fig. 3G). In terms of expression domain, Dlx5 and Msx2 expression sites overlap in the apex of the neural folds just prior to fusion (Fig. 3B,E).

image

Figure 3. Expression patterns of Dlx5 and Msx2 during cranial neural tube formation. (A–C) Expression of Dlx5 at E8.5 (A) and E9.5 (B, C). Dlx5 transcripts are detected transiently prior to neural tube closure at E9.5 (arrowheads in b), and the transcripts are no longer observed after neural tube closure (c). (D–F) Expression of Msx2 at E8.5 (D) and E9.5 (E, F). Expression of Msx2 is detected before neural tube closure (arrowhead in D), and its expression is detected continuously at the point of fusion (arrowheads in E and F). (G) Schematic diagram of temporal expression pattern of Dlx5 (Orange line) and Msx2 (yellow line) during cranial neural folds fusion. The cranial neural tube formation images are modified from the e-Mouse Atlas (www.emouseatlas.org). Scale bars: 100 μm.

Download figure to PowerPoint

EphrinA5 and EphA7 are regulated by Dlx5 and Msx2 together

The developmental expression of Dlx5 and Msx2 suggests that the exencephaly phenotype associated with the double mutation may be linked to a synergistic interaction between these two genes during a transient period when they are both expressed at the apex of the neural folds during neural fold fusion. Since both Dlx5 and Msx2 are transcriptional regulators that must regulate morphogenetic events by affecting the expression of structural genes, we explored potential downstream target genes that might be linked to the exencephaly phenotype. In a previous report, it has been demonstrated that ephrinA5 and its receptor EphA7 participate in cranial neural tube morphogenesis via cell attraction and cell repulsion, and that embryos that possess a defect in this signaling pathway can display exencephaly (Holmberg et al. 2000). To explore whether Dlx5 and Msx2 have a regulatory role on the expression of ephrinA5, EphA7-FL, and EphA7-T1, we performed gene expression pattern analysis of ephrinA5, EphA7-FL, and EphA7-T1 on WT, Dlx5−/−, Msx2−/−, and Dlx5−/−;Msx2−/− mouse embryos at E9.5 (two embryos were used in each genotype for the gene expression pattern analysis). In wildtype embryos, the expression domain of ephrinA5 and EphA7-FL are almost identical to each other in that expression is restricted to the outer layer of the neural tube (Fig. 4A,E). The expression domain of EphA7-T1 is in the dorsal two-thirds of the neural tube (Fig. 4I, the ventral margin of EphA7-T1 expression domain indicated by arrows), and the domain is broader than ephrinA5 and EphA7-FL. In Dlx5 null mutant embryos, expression of ephrinA5 and EphA7-FL appeared similar to wildtype embryos (Fig. 4B,F). However, EphA7-T1 expression domain was expanded to the ventral region of the neural tube in Dlx5−/− mutants, although the intensity of expression did not appear to be significantly changed (Fig. 4J). In Msx2 null mutant embryos, expression of EphA7-FL was not changed (Fig. 4G), whereas ephrinA5 expression was slightly decreased (Fig. 4C). The expression domain of EphA7-T1 in Msx2 mutant was expanded to the ventral region, but the expression level was not altered (Fig. 4K). In Dlx5 and Msx2 double null mutant embryos displaying exencephaly the expression of ephrinA5 was decreased all over the neural tissue, particularly in the region of Dlx5/Msx2 co-expression at the apex of the neural folds (Fig. 4D). EphA7-FL expression was not modified in the apex of neural folds in Dlx5−/−;Msx2−/− embryos (Fig. 4H), whereas EphA7-T1 transcripts were largely absent throughout the neural tissue including the apex of the neural folds (Fig. 4D,L). These results show that the expression of the ligand, ephrinA5, and one of its receptors, EphA7-T1, is regulated by the combined activity of Dlx5 and Msx2 during anterior neural tube closure.

image

Figure 4. Expression pattern of ephrinA5, EphA7-FL, and EphA7-T1 in WT, Dlx5−/−, Msx2−/−, and Dlx5−/−;Msx2−/− mice embryos at E9.5. (A–D) Expression of ephrinA5 in wildtype (A), Dlx5−/− (B), Msx2−/− (C), and Dlx5−/−;Msx2−/− (D). Transcripts of ephrinA5 are detected outer layer of the neural tube. This expression is not altered in Dlx5, or slightly decreased in Msx2 mutant. In contrast, the double mutant embryo shows significantly decreased expression of ephrinA5. (E–H) Expression of EphA7-FL in wildtype (E), Dlx5−/− (F), Msx2−/− (G), and Dlx5−/−;Msx2−/− (H). Expression pattern of EphA7-FL is not modified in Dlx5, Msx2, or the double mutant. (I–L) Expression of EphA7-T1 in wildtype (I), Dlx5−/− (J), Msx2−/− (K), and Dlx5−/−;Msx2−/− (L). Expression domain of EphA7-T1 in the neural tube is expanded toward the ventral side in Dlx5 and Msx2 single mutant embryo. In Dlx5/Msx2 double mutant, EphA7-T1 expression is significantly decreased. Scale bars: 100 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

In the mouse, neural tube closure initiates at three different points along the cranial-caudal axis. The primary initiation point (closure 1) is located at the hindbrain/cervical boundary, and closure then proceeds in both cranial and caudal directions. The second neural tube closure (closure 2) initiation point is located at the forebrain/midbrain boundary. The last neural tube closure (closure 3) initiation point is located at the extreme rostral end of the embryo, and closure proceeds in the caudal direction (Copp et al. 2003). In this study we focused on the role of Dlx5 and Msx2 in cranial neural tube closure (closure 2) and found that Dlx5 and Msx2 are co-expressed at the apex of the neural folds during neural tube formation. Using Dlx5 and Msx2 genetically disrupted mice, we confirm a low frequency of exencephaly in the Dlx5 mutant (Depew et al. 1999), and found that the frequency of exencephaly becomes more severe with the sequential removal of Msx2 alleles. Since Dlx5 is a known transcriptional activator and Msx2 is a transcriptional repressor known to heterodimerize and/or compete with Dlx5 for DNA binding to antagonize Dlx5 activity (Zhang et al. 1997), the observed genetic interaction between Dlx5 and Msx2 in the double mutant poses a bit of a conundrum that requires further investigation. A similar synergistic interaction between Dlx5 and Msx1 has been reported in association with frontal bone development (Chung et al. 2010). Based on these results the simple interpretation that the loss of a transcriptional repressor would phenotypically cancel the loss of a transcriptional activator seems unlikely.

Functional redundancy among members of the Dlx or Msx families has been previously demonstrated (Robledo et al. 2002; Lallemand et al. 2005). Within the Dlx family, Dlx5 and Dlx6 are very similar in their expression pattern, homology of the amino acid sequences (Merlo et al. 2000; Zerucha et al. 2000), and they have redundant function in limb development and cranial neural tube formation (Robledo et al. 2002). Similarly, in the Msx family, redundancy has also been demonstrated in the regulatory function of Msx1 and Msx2 in limb development and cranial neural tube formation (Lallemand et al. 2005; Han et al. 2007). Therefore, the functional redundancy among the Dlx transcriptional activators and the Msx transcriptional repressors can account for the dose-related effects of Msx2 in a Dlx5 mutant background. Thus, the functional redundancy within Dlx and Msx family members, the established functional antagonism between Dlx5 and Msx2, and the co-expression of Dlx5 and Msx2 at the tips of the neural fold prior to fusion suggest that the interaction between Dlx and Msx proteins play a primary regulatory role in controlling neural tube closure. This conclusion is further supported by gain of function studies in which Msx2 overexpression also results in a low frequency of exencephalic embryos (Winograd et al. 1997).

Our Dlx5/Msx2 mutant studies identified that the expression of the ephrinA5 gene and transcripts of the truncated form of its receptor gene, EphA7-T1, are downregulated during neural tube formation. These data provide evidence of molecular networking between Dlx5/Msx2 and ephrinA5/EphA7 in cranial neural tube morphogenesis (Fig. 5A,B); however, the detailed molecular mechanisms of this network are not clear at this time. The expression domains of Dlx5 and Msx2 are restricted to the very tip of the neural folds, whereas transcripts of ephrinA5, EphA7-FL, and EphA7-T1 are detected broadly in the dorsal side of neural folds and the neural tube. These non-overlapping expression domains suggest an indirect regulatory interaction possibly involves downstream signaling between cells of the neural fold. However, it is clear that the ephrinA5 gene is a downstream target in this pathway, whereas the EphA7 gene is not. It is also important to note that the frequency of neural tube closure defect in the Dlx5/Msx2 double mutants is four times greater than the ephrinA5 null mouse, thus indicating additional downstream targets modulating neural tube closure.

image

Figure 5. Dlx5/Msx2 regulates ephrinA5/EphA7-T1 during cranial neural fold fusion. (A) This model demonstrates that Dlx5/Msx2 regulates expression of ephrinA5, and pre-mRNA alternative splicing process of EphA7 transcripts to the truncated isoform (EphA7-T1). (B) Schematic diagram demonstrates a model that ephrinA5/EphA7 combination in wildtype and Dlx5 and Msx2 double mutant results either in cell adhesion or cell repulsion.

Download figure to PowerPoint

Transcripts for the EphA7 gene encode for a full-length receptor (EphA7-FL), and two tyrosine kinase domain truncated isoforms, EphA7-T1 and EphA7-T2, which are the products of alternative splicing (Ciossek et al. 1995; Valenzuela et al. 1995). Since Dlx5/Msx2 regulates expression of a truncated isoform of EphA7, but not the full length isoform, Dlx5/Msx2 must regulate the mechanism by which EphA7 RNA is differentially spliced. To date, there is no evidence for regulation of pre-mRNA alternative splicing by Dlx or Msx transcription factors, However, modulation of the 5′-splice site by the transcription factor c-Myb has been reported (Orvain et al. 2008). Further investigations into this novel molecular network in which Dlx5/Msx2 regulates neural fold morphogenesis by controlling differential cell adhesion via ephrinA5/EphA7 interactions is necessary for our understanding of cranial neural fold fusion.

With respect to the developing digit we find that despite overlapping expression domains at the digit tip, an interaction between Dlx5 and Msx2 is not functionally linked to either digit tip formation or regeneration. Since other Dlx and Msx family members are co-expressed in similar domains (Robledo et al. 2002; Han et al. 2003) it is reasonable to conclude that functional redundancy may be masking any phenotypic defect. However, the discovery of a Dlx5/Msx2 link to the control of ephrinA5/EphA7 activity during neural tube closure is suggestive that this signaling network may be conserved during digit formation and regeneration. The role of ephrin/Eph signaling in mammalian digit regeneration has not been explored in detail; however, differential cell adhesion is known to play a critical role both during limb development and limb regeneration (see Wada 2011) making further exploration of this molecular network primed for future studies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

We thank the Muneoka lab for discussions and Dr Robert Kosher for kindly providing the Dlx5 mutant. Also, we thank Dr Jonas Frisén for kindly providing the EphrinA5, EphA7-FL, and EphA7-T1 containing DNA plasmids. Research funded by P01HD022610 from the NIH, W911NF-06-1-0161 from DARPA, W911NF-09-1-0305 from the US Army Research Center, and the John L. and Mary Wright Ebaugh endowment fund at Tulane University.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Acampora, D., Merlo, G. R., Paleari, L., Zerega, B., Postiglione, M. P., Mantero, S., Bober, E., Barbieri, O., Simeone, A. & Levi, G. 1999. Craniofacial, vestibular and bone defects in mice lacking the distal-less-related gene Dlx5. Development 126, 37953809.
  • Bach, A., Lallemand, Y., Nicola, M. A., Ramos, C., Mathis, L., Maufras, M. & Robert, B. 2003. Msx1 is required for dorsal diencephalon patterning. Development 130, 40254036.
  • Campbell, T. N. & Robbins, S. M. 2008. The eph receptor/ephrin system: an emerging player in the invasion game. Curr. Issues Mol. Biol. 10, 6166.
  • Chung, I., Han, J., Iwata, J. & Chai, Y. 2010. Msx1 and Dlx5 function synergistically to regulate frontal bone development. Genesis 48, 645655.
  • Ciossek, T., Lerch, M. M. & Ullrich, A. 1995. Cloning, characterization, and differential expression of MDK2 and MDK5, two novel receptor tyrosine kinases of the eck/eph family. Oncogene 11, 20852095.
  • Copp, A. J., Greene, N. D. & Murdoch, J. N. 2003. The genetic basis of mammalian neurulation. Nat. Rev. Genet. 4, 784793.
  • Depew, M. J., Liu, J. K., Long, J. E., Presley, R., Meneses, J. J., Pedersen, R. A. & Rubenstein, J. L. 1999. Dlx5 regulates regional development of the branchial arches and sensory capsules. Development 126, 38313846.
  • Depew, M. J., Simpson, C. A., Morasso, M. & Rubenstein, J. L. 2005. Reassessing the Dlx code: the genetic regulation of branchial arch skeletal pattern and development. J. Anat. 207, 501561.
  • Frisen, J., Holmberg, J. & Barbacid, M. 1999. Ephrins and their eph receptors: multitalented directors of embryonic development. EMBO J. 18, 51595165.
  • Gilbert, S. F. 2003. Developmental Biology. Sinauer, USA.
  • Han, M., Yang, X., Farrington, J. E. & Muneoka, K. 2003. Digit regeneration is regulated by Msx1 and BMP4 in fetal mice. Development 130, 51235132.
  • Han, J., Ishii, M., Bringas, P. Jr, Maas, R. L., Maxson, R. E. Jr & Chai, Y. 2007. Concerted action of Msx1 and Msx2 in regulating cranial neural crest cell differentiation during frontal bone development. Mech. Dev. 124, 729745.
  • Holleville, N., Quilhac, A., Bontoux, M. & Monsoro-Burq, A. H. 2003. BMP signals regulate Dlx5 during early avian skull development. Dev. Biol. 257, 177189.
  • Holmberg, J., Clarke, D. L. & Frisen, J. 2000. Regulation of repulsion versus adhesion by different splice forms of an eph receptor. Nature 408, 203206.
  • Kim, H. J., Rice, D. P., Kettunen, P. J. & Thesleff, I. 1998. FGF-, BMP- and Shh-mediated signalling pathways in the regulation of cranial suture morphogenesis and calvarial bone development. Development 125, 12411251.
  • Kim, Y. J., Lee, M. H., Wozney, J. M., Cho, J. Y. & Ryoo, H. M. 2004. bone morphogenetic protein-2-induced alkaline phosphatase expression is stimulated by Dlx5 and repressed by Msx2. J. Biol. Chem. 279, 5077350780.
  • Klein, R. 2009. bidirectional modulation of synaptic functions by eph/ephrin signaling. Nat. Neurosci. 12, 1520.
  • Lallemand, Y., Nicola, M. A., Ramos, C., Bach, A., Cloment, C. S. & Robert, B. 2005. Analysis of Msx1; Msx2 double mutants reveals multiple roles for msx genes in limb development. Development 132, 30033014.
  • Merlo, G. R., Zerega, B., Paleari, L., Trombino, S., Mantero, S. & Levi, G. 2000. multiple functions of Dlx genes. Int. J. Dev. Biol. 44, 619626.
  • Muneoka, K., Wanek, N. & Bryant, S. V. 1986. Mouse embryos develop normally exo utero. J. Exp. Zool. 239, 289293.
  • Newberry, E. P., Latifi, T. & Towler, D. A. 1998. Reciprocal regulation of osteocalcin transcription by the homeodomain proteins Msx2 and Dlx5. Biochemistry 37, 1636016368.
  • Orvain, C., Matre, V. & Gabrielsen, O. S. 2008. The transcription factor c-Myb affects pre-mRNA splicing. Biochem. Biophys. Res. Commun. 372, 309313.
  • Ramos, C. & Robert, B. 2005. Msh/Msx gene family in neural development. Trends Genet. 21, 624632.
  • Robledo, R. F., Rajan, L., Li, X. & Lufkin, T. 2002. The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development. Genes Dev. 16, 10891101.
  • Satokata, I., Ma, L., Ohshima, H., Bei, M., Woo, I., Nishizawa, K., Maeda, T., Takano, Y., Uchiyama, M., Heaney, S., Peters, H., Tang, Z., Maxson, R. & Maas, R. 2000. Msx2 deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation. Nat. Genet. 24, 391395.
  • Satokata, I. & Maas, R. 1994. Msx1 deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat. Genet. 6, 348356.
  • Shimeld, S. M., McKay, I. J. & Sharpe, P. T. 1996. The murine homeobox gene Msx-3 shows highly restricted expression in the developing neural tube. Mech. Dev. 55, 201210.
  • Shirakabe, K., Terasawa, K., Miyama, K., Shibuya, H. & Nishida, E. 2001. Regulation of the activity of the transcription factor Runx2 by two homeobox proteins, Msx2 and Dlx5. Genes Cells 6, 851856.
  • Timor-Tritsch, I. E., Greenebaum, E., Monteagudo, A. & Baxi, L. 1996. Exencephaly-anencephaly sequence: proof by ultrasound imaging and amniotic fluid cytology. J. Matern. Fetal. Med. 5, 182185.
    Direct Link:
  • Valenzuela, D. M., Rojas, E., Griffiths, J. A., Compton, D. L., Gisser, M., Ip, N. Y., Goldfarb, M. & Yancopoulos, G. D. 1995. Identification of full-length and truncated forms of Ehk-3, a novel member of the eph receptor tyrosine kinase family. Oncogene 10, 15731580.
  • Wada, N. 2011. Spatiotemporal changes in cell adhesiveness during vertebrate limb morphogenesis. Dev. Bio. 240(5), 969978.
  • Winograd, J., Reilly, M. P., Roe, R., Lutz, J., Laughner, E., Xu, X., Hu, L., Asakura, T., vander Kolk, C. & Strandberg, J. D.1997. Perinatal lethality and multiple craniofacial malformations in MSX2 transgenic mice. Hum. Mol. Genet. 6, 369379.
  • Zerucha, T., Stuhmer, T., Hatch, G., Park, B. K., Long, Q., Yu, G., Gambarotta, A., Schultz, J. R., Rubenstein, J. L. & Ekker, M. 2000. A highly conserved enhancer in the Dlx5/Dlx6 intergenic region is the site of cross-regulatory interactions between Dlx genes in the embryonic forebrain. J. Neurosci. 20, 709721.
  • Zhang, H., Hu, G., Wang, H., Sciavolino, P., Iler, N., Shen, M. M. & Abate-Shen, C. 1997. Heterodimerization of Msx and Dlx homeoproteins results in functional antagonism. Mol. Cell. Biol. 17, 29202932.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
dgd12044-sup-0001-FigS1.pdfapplication/PDF98KFig. S1. Gross morphology of WT, Dlx5−/−, Msx2−/−, and Dlx5;Msx2−/− mouse embryos at E18.5. Note that, while cranial phenotypes vary depending on genotype, gross morphologies of embryos are not significantly altered from wildtype. Scale bar; 1 cm.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.