Skip to main content
NCBI home page
Human Molecular Genetics logoLink to Human Molecular Genetics
. 2011 Dec 8;21(6):1374–1383. doi: 10.1093/hmg/ddr575

Cranial neural crest ablation of Jagged1 recapitulates the craniofacial phenotype of Alagille syndrome patients

Ryan Humphreys 1,†, Wei Zheng 5,†, Lawrence S Prince 2, Xianghu Qu 3, Christopher Brown 3, Kathleen Loomes 6, Stacey S Huppert 2, Scott Baldwin 2,3, Steven Goudy 4,*
PMCID: PMC3465692  PMID: 22156581

Abstract

JAGGED1 mutations cause Alagille syndrome, comprising a constellation of clinical findings, including biliary, cardiac and craniofacial anomalies. Jagged1, a ligand in the Notch signaling pathway, has been extensively studied during biliary and cardiac development. However, the role of JAGGED1 during craniofacial development is poorly understood. Patients with Alagille syndrome have midface hypoplasia giving them a characteristic ‘inverted V’ facial appearance. This study design determines the requirement of Jagged1 in the cranial neural crest (CNC) cells, which encompass the majority of mesenchyme present during craniofacial development. Furthermore, with this approach, we identify the autonomous and non-autonomous requirement of Jagged1 in a cell lineage-specific approach during midface development. Deleting Jagged1 in the CNC using Wnt1-cre; Jag1 Flox/Flox recapitulated the midfacial hypoplasia phenotype of Alagille syndrome. The Wnt1-cre; Jag1 Flox/Flox mice die at postnatal day 30 due to inability to masticate owing to jaw misalignment and poor occlusion. The etiology of midfacial hypoplasia in the Wnt1-cre; Jag1 Flox/Flox mice was a consequence of reduced cellular proliferation in the midface, aberrant vasculogenesis with decreased productive vessel branching and reduced extracellular matrix by hyaluronic acid staining, all of which are associated with midface anomalies and aberrant craniofacial growth. Deletion of Notch1 from the CNC using Wnt1-cre; Notch1 F/F mice did not recapitulate the midface hypoplasia of Alagille syndrome. These data demonstrate the requirement of Jagged1, but not Notch1, within the midfacial CNC population during development. Future studies will investigate the mechanism in which Jagged1 acts in a cell autonomous and cell non-autonomous manner.

INTRODUCTION

Alagille syndrome was first reported in 1969 in a group of patients who had idiopathic bile duct anomalies, cardiac disease and similar craniofacial features (1). Alagille syndrome is autosomally dominantly inherited and is predominantly due to mutations in the JAGGED1 gene (2,3). The bile duct anomalies include a paucity of bile ducts leading to cholestasis, the most recognizable characteristic of Alagille syndrome, occurring in >95% of patients (4). The cardiac manifestations of Alagille syndrome clinically present as stenosis in the pulmonary outflow tract or peripheral pulmonic vessels (5). Importantly, numerous defects in vascular development occur in Alagille syndrome patients, including renal artery stenosis, moyamoya and central nervous system vascular anomalies. The central nervous system vascular anomalies can lead to fatal intracranial bleeding, which is a major cause of morbidity and mortality in Alagille syndrome patients (6). Thus, there are multiple examples of Alagille syndrome patients having aberrant vasculogenesis, interestingly the expressivity of these traits varies greatly between patients.

The characteristic facies of Alagille syndrome include a prominent forehead, deep-set eyes, a straight nose with a flat tip, flattened midface and a prominent chin (1). This assemblage of midfacial findings gives the Alagille syndrome patients an appearance of having an ‘inverted V face’ in early childhood with some patients having a more severe phenotype, including cleft palate (7). The classic facial inverted V feature was found in 95% of Alagille syndrome patients who were diagnosed based on the intrahepatic bile duct phenotype (5). Kamath et al. (8) found that the characteristic facial features provided a 76% sensitivity, 82% specificity and 81% positive predicative value in the diagnosis of Alagille syndrome. Later in life, Alagille syndrome patients develop significant midfacial hypoplasia. The development of midfacial hypoplasia suggests that the midface, or maxilla, is not growing commensurately with the rest of the face. The palate, which makes up the majority of the maxilla, is permanently affected in Alagille syndrome patients, and cleft palate formation is reported in a minority of Alagille syndrome patients. This suggests that Jagged1 is involved early in midfacial and palatal morphogenesis. Palatogenesis, a key component of midface development, is a complex choreography of palatal shelf formation, elevation, elongation and fusion that requires active cellular proliferation and extracellular matrix production (9). It is unknown by what mechanism Jagged1 is involved in palatal and midfacial development.

Jagged1 signaling has been shown to be required for establishing the dorsal identity of the craniofacial skeleton in zebrafish (10). However, the majority of the zebrafish craniofacial skeleton is comprised of the mandibular arch, whereas the mammalian craniofacial structure is primarily derived from the frontonasal, maxillary and mandibular arches. To investigate the role of Jagged1 in the craniofacial abnormalities seen in Alagille syndrome, a mammalian model is required. The majority of the craniofacial mesenchyme is cranial neural crest (CNC) cells that migrate from the dorsal neural tube into the frontonasal process and branchial arches. We hypothesized that Jagged1 is a critical gene necessary in the CNC population for midfacial development in this study. We determined the role of Jagged1 in the midface using the mouse as a model system to identify the cell lineage(s) that required Jagged1. We conditionally removed Jagged1 specifically in the CNC lineage during development of the midface in mice and recapitulated the Alagille syndrome midface hypoplasia. Additionally, we demonstrated that Jagged1 acts on the midfacial tissue in both a cell-autonomous and non-cell-autonomous fashion.

RESULTS

Cranial neural crest deletion of Jagged1 recapitulates midfacial hypoplasia phenotype

It is unknown which cells require Jagged1 and contribute to the defect in Alagille syndrome patients leading to the midfacial hypoplasia deformity. The CNC is a major source of midface tissue making the role of Jagged1 in the CNC an important question to be answered (11). To test whether Jagged1 was required in the CNC, we conditionally ablated Jagged1 (Jag1) in the CNC using Wnt1-cre; Jag1 Flox/Flox (Jag1 CKO) mice (12,13). The Jag1 CKO mice were viable at birth and grew to 30 days of age; however, they were smaller than their littermates. After 1 month, the Jag1 CKO mice were unable to eat properly requiring soft mouse chow and became progressively weaker due to the lack of maxillary growth. The dental occlusion of the Jag1 CKO mice was noticeably misaligned with severe midface hypoplasia requiring trimming of the lower incisors weekly.

To examine the skeletal deficiencies due to the loss of CNC Jag1 signaling, skeletal preparations of Jag1 CKO mice at 1 month were performed. These data demonstrated significantly shortened maxillary regions (Fig. 1A–B). The dimensions of the craniofacial structures of the Jag1 CKO mice were quantified using craniofacial morphometrics (14). The inframaxillary length and the posterior–anterior length were significantly reduced in the Jag1 CKO (Fig. 1C–D). The palatal dimensions of the Jag1 CKO mice were proportionally smaller; however, the structural components were intact, as demonstrated by the presence of a similar number of nine palatal rugae (Fig. 1E–F). The mandibular arch in the Jag1 CKO mice was similar to the control and the remaining craniofacial skeletal dimensions (superior maxillary, lateral maxillary, intramaxillary, superior mandibular, inferior mandibular, inferior coronoid) were unchanged in the Jag1 CKO mice. Anterior–posterior facial length was normal at P0, but over the ensuing weeks, the Jag1CKO mice had statistically smaller maxillary lengths (P < 0.05) up until the point of death at P30 (Fig. 1G). The loss of Jagged1 in the CNC tissues most greatly affects the palate and, therefore, the maxilla, similar to craniofacial growth in Alagille syndrome patients, and does not appear to affect the over all patterning or growth of other areas of the craniofacial skeleton.

Figure 1.

Figure 1.

Open in a new tab

Midfacial hypoplasia in the Jag1 CKO recapitulates Alagille syndrome facies. The midface of the mutant mice (B) is severely retruded (white arrow) with associated malocclusion compared with the control (A). Craniofacial measurements of the mutant mice in (C) reveal shortened anterior–posterior dimensions (red lines) of the maxillary segment with the maintenance of other dimensions (white lines). The morphometric analysis demonstrated that the inframaxillary (short red line (c) on maxilla) and the posterior-anterior (long red line (b) on skull) dimensions were significantly different (D) (all P< 0.05). The Jag1 CKO had smaller palates in the mutant mice (F) compared with the control (E). The developmental subunits of the palate are all intact in the mutant palates (H) with equal number of rugae (white arrows and black lines) when compared with the wild-types (G). At P7, there is reduction in midfacial growth in the Jag1CKO mice that persists until death at P30 (I) (all P < 0.05). *Statistically significant difference.

Jagged1 is necessary for palatal elongation

The palate is the most significant and substantial contributor to midface development and maxillary growth. Therefore, to identify the mechanism of the hyoplastic midfacial formation in the Jag1 CKO mice, we performed morphological analysis of embryonic day (E)13.5, E14.5 and E15.5 embryos during the critical time points of palate development. Expression of Jagged1 during this critical period of midfacial development was concentrated in the palate and tongue at E13.5 and in the palate and periocular regions at E14.5 and E 15.5 (Supplementary Material, Fig. S1). The Jag1CKO mice maintain epithelial Jagged1 expression at E13.5, E14.5 and E15.5, but minimal Jagged1 mesenchymal expression is seen. Jagged1 expression in the developing craniofacial region at E8.5 was not reduced in the Jag1CKO. Using Wnt1-Cre;Jag1F/F;R26R mice, the CNC midfacial contribution was compared between the wild-type (WT) and Jag1CKO and found to contain similar contributions of Lac-Z-positive cells at E13.5. Sox9 staining was similar between the WT and Jag1CKO mice in the palate and tongue region at E13.5 (Supplementary Material, Fig. S2). This evidence indirectly suggests that the palatal phenotype found in the Jag1CKO mice occurs due to events during palatogenesis, not prior to palatogenesis.

Palatal shelf formation was normal at E13.5 in the Jag1 CKO embryos. At E14.5, palatal shelf elevation above the tongue occurred normally but palatal elongation was significantly reduced anteriorly in the Jag1 CKO (Fig. 2C, D, J), whereas the posterior palate shelf length is unaffected. Palate shelf height is reduced in both the anterior and posterior regions of the Jag1CKO mice (Fig. 2K). Concomitantly, there is reduced anterior facial width in the Jag1CKO mice; however, the posterior area maintains a normal facial width (Fig. 2L). The reduction in anterior facial width likely narrows the face enough to allow palate fusion to occur. Some of the E15.5 Jag1 CKO embryos demonstrated delayed palate shelf apposition and the persistence of the epithelial seam; however, palate fusion occurred in all Jag1 CKO mice examined (Supplementary Material, Fig. S3).

Figure 2.

Figure 2.

Open in a new tab

Palatal elongation defects occur in the Jag1 CKO mice due to alteration in proliferation and HA staining. At E14.5, the mutant palate shelves have elevated above the tongue but are significantly shortened and are unable to oppose anteriorly (C) and posteriorly (D) when compared with the control (A) and (B), respectively. At E14.5, the shortened anterior palate shelves have reduced phosphohistone H3 (PH3) staining (black arrows) in the Jag1 CKO mice (G) compared with control mice (E) which is statistically significant (P < 0.05) (I). HA staining is greatly reduced at E14.5 in the anterior palate shelves of Jag1 CKO palate shelves (black arrow) (H) compared with control palate shelves (F). Length of the anterior palatal shelves at E14.5 is significantly shorter (P < 0.05) in the Jag1CKO (J) and palatal height is significantly reduced in both the anterior and posterior palatal shelves in the Jag1CKO (all P < 0.05) (K). Total facial width at E14.5 was significantly reduced (P < 0.05) in the anterior palatal region in the Jag1CKO; however, the posterior total facial width was normal (L). PH3 staining in the P14 Jag1CKO palates (O) is qualitatively similar to controls (M) in the mesenchyme (black arrows), but PH3 is reduced in the Jag1CKO epithelium compared with controls (grey arrows). At P14, there continues to be a qualitative reduction in HA staining in the Jag1CKO (P) compared with control (N). *Statistical significance.

To determine the cause of the delayed palatal shelf elongation in the Jag1 CKO embryos, we investigated whether cellular proliferation was altered. The E14.5 Jag1 CKO embryo palate shelves were stained for phosphohistone H3 (PH3) to evaluate proliferation. There was markedly reduced proliferation in the Jag1 CKO embryo palate shelves compared with controls (Fig. 2E–G) that were significant (Fig. 2I). At P14, mesenchymal proliferation was qualitatively similar in the control compared with the Jag1CKO (black arrows) (Fig. 2M–O). In the adjacent palatal epithelium, there was qualitatively reduced proliferation in the Jag1CKO compared with the controls (grey arrows).

The shortening of the E14.5 Jag1 CKO palate shelves may also be due to decreased extracellular matrix production. Hyaluronic acid (HA) staining was used to study the extracellular matrix. There was reduced HA present in the Jag1 CKO embryo palate shelves (Fig. 2F–H), suggesting that Jagged1 in the CNC of the palate is required for cell cycle progression and extracellular matrix production. Reduction in the extracellular matrix in the Jag1 CKO leads to shortening of the mutant palate shelves. Continued reduction in HA production was found in the palate of the P14 Jag1CKO mice (Fig. 2N–P).

Aberrant vascular patterning in the palate in the absence of Jagged1

Due to cranial vascular patterning defects in Jag1−/− mice, we investigated whether vascular patterning was aberrant in the midface of the Jag1 CKO mice (15). The E13.5, E14.5 and E15.5 palate shelves in Jag1 CKO embryos were stained for Jagged1 and platelet endothelial cell adhesion molecule (PECAM) to correlate Jagged1 staining with vascular development. In the Jag1 CKO embryos, there was reduced palatal mesenchyme expression of Jag1 as expected (Fig. 3D, E, F). At E13.5, the palate shelves in Jag1 CKO embryos had similar PECAM staining to the control palate shelves (Fig. 3A–D). However, at E14.5, Jag1 CKO embryos have reduced PECAM staining and reduced vascular branching compared with control palate shelves (Fig. 3B–E). The significant surge of Jagged1 expression in the E14.5 palate (Fig. 3B) suggests a narrow temporal requirement during palate formation. The reduced vascular branching pattern persists in the E15.5 Jag1 CKO mice palate shelves compared with controls (Fig. 3C–F). Of note, there is concentrated Jagged 1 expression at E15.5 adjacent to the palatal artery (Fig. 3C) where vascular smooth muscle development occurs. The amount of branching in the E14.5 Jag1 CKO was significantly less (Fig. 3G). Evaluation of P14 Jag1 CKO mice demonstrated reduced vasculature organization, poor vascular smooth muscle investment and irregular vessel formation (Fig. 3I and K).

Figure 3.

Figure 3.

Open in a new tab

Jagged1 is necessary for vascular branching in the palate. Expression of Jagged1 staining (red staining) in the control palate is greatest at E14.5 (B) compared with E13.5 (A) and E15.5 (C). At E15.5, the expression of Jagged1 is concentrated around the central palate artery (red arrow C). In the Jag1 CKO mutant mice, there is minimal mesenchymal Jagged1 staining, but there is a small amount in the midline palate fusion plane at E15.5 (F), where epithelial breakdown is occurring. The vascular development evaluated by PECAM staining (green) is unorganized at E13.5 in both the control (A) and mutant palate shelf (D). However, at E14.5, vascular branching and organization is taking place in the control (B, white arrows), but the Jag1 CKO have comparably reduced vascular branching (E, white arrows) that is statistically significant (P < 0.05) (G). At E15.5, the control vasculature is branching further (C), whereas the mutant vascular branching is irregular and less dense (F). At P14, there continues to be reduced vascular organization and vessel size (white arrows) in the Jag1 CKO (J and K) compared with control (H and I). There are fewer and less organized vascular structures with poor smooth muscle investment in the Jag1CKO (K) compared with controls (I). *Statistical significance.

Cell-autonomous and non-cell-autonomous Jagged1 signaling in the midface

To determine whether Notch signaling is required cell autonomously in the CNC lineage, we disrupted the reception of Notch signaling within the CNC population. Wnt1-cre; Notch1 Flox/Flox (N1CKO) mice were generated. The palate formation of the N1CKO embryos was found to be normal on morphologic analysis. At E14.5, the N1CKO embryonic palate shelves elevated normally and were in the process of palate fusion similarly to the timing in control embryos (Fig. 4C–D). This result is in contrast to the delay observed in the Jag1CKO embryos (Fig. 2C–D). The amount of cellular proliferation in the N1CKO palate is similar to that of the wild-type mice (Fig. 4E–H) (data not shown). Measurements of palatal length and facial width in the anterior and posterior regions of the N1CKO mice did not reveal any differences compared with controls (P> 0.5) (data not shown). These mice were not viable for unknown reasons; therefore, we could not examine the long-term craniofacial formation compared with the Jag1CKO mice.

Figure 4.

Figure 4.

Open in a new tab

Absence of Notch1 in the CNC does not disrupt palatal elongation. The elongation and apposition of the Wnt1-cre;Notch1 F/F (N1CKO) palatal shelves is similar to the control mice in the anterior (A) and posterior (B) regions when compared with the N1CKO mice anterior (C) and posterior (D) palate shelves. There is no qualitative difference in phosphohistone H3 staining in the control anterior (E) and posterior (F) palate shelves compared with the N1CKO mice anterior (G) and posterior (H) palate shelves.

To evaluate the Notch receptor response to the absence of Jagged1 in the CNC, we performed quantitative PCR on isolated maxillary tissue of E12.5, E13.5, E14.5 and E15.5 Jag1CKO during midfacial development. The expression of Jagged1 and its receptors, Notch 1-4, in the isolated samples of the midface was examined. At E12.5, a reduction in Jag1 and Notch 1-4 expression was already apparent (Fig. 5). At day E13.5 when palate shelves are undergoing active proliferation and extracellular matrix production, Jag1 and Notch3 expression was reduced compared with control levels, but Notch 1, Notch2 and Notch 4 were unchanged. At day E14.5, when palate shelves elevate and appose to begin fusion, Jagged1 and Notch2 were reduced, but Notch1, Notch3 and Notch4 were up-regulated. These data suggest that Notch1, Notch3 and Notch4 are up-regulated in the Jag1 CKO mice in response to reduced Jagged1 in the CNC; however, Notch2 does not appear to be up-regulated. At day E15.5, during the final stage of palate fusion, there is no difference in Jagged1 or Notch 1-4. The normalization of Jagged1 and Notch RNA expression at E15.5 in Jag1CKO is similar to the equivalent Jagged1 staining in the E15.5 Jag1CKO and control palate shelves (Fig. 3C–F), suggesting that Jagged1 is temporally required primarily in the E14.5 palate shelves.

Figure 5.

Figure 5.

Open in a new tab

Loss of Jagged1 in the CNC leads to increased Notch1 and Notch3 expression in the midface. At E12.5, loss of Jagged1 in the CNC results in a global reduction in all Notch receptors by quantitative PCR of isolated midfacial tissues. At E13.5, Notch , 2, and 4 are unchanged in expression, but Notch 3 is still reduced. At E14.5, Notch1, 3 and 4 are significantly up-regulated and Notch 2 is down-regulated. At E15.5, there is normalization of Jagged1 expression and all Notch receptors (all P-values < 0.05). *Statistical significance. Diagram in right upper corner depicts area of tissue dissection.

DISCUSSION

Jagged1 is required in the CNC for midfacial development

In this study, we set out to study the role of Jagged1 in the midface of mice and aDDRess the specific cell lineage requirement that leads to the midfacial hypoplasia phenotype observed in Alagille syndrome patients. We recreated the midfacial hypoplasia phenotype in mice by conditionally deleting Jagged1 specifically in the cranial neural crest tissue (Fig. 1B and C), resulting in a foreshortened maxillary projection and marked midface retrusion, similar to the human Alagille syndrome facial phenotype. The resulting malocclusion in the Jag1 CKO led to poor feeding and demise at 1 month of age.

The mutant palate shelves were capable of elevation but they were unable to elongate, suggesting a primary palatal defect. During normal palatogenesis, there is a distinct surge of Jagged1 expression just prior to the fusion of the palate shelves (Fig. 3B). The images in Supplementary Material, Fig. S1B show still active, yet diminished, Jagged1 during the fusion of palate shelves. The difference noted between the two images illustrates the temporally dynamic expression of Jagged1 pre- and post-palate apposition and fusion. The palatal anomaly in Jag1CKO does not lead to cleft palate formation, although cleft palate is observed in a subset of Alagille syndrome patients (7). Despite reduced palatal elongation, palate fusion occurs due to reduced anterior facial width, allowing palate fusion to occur. The presence of equal number of palatal rugae suggests that patterning of the midface in the Jag1CKO is occurring normally. Additionally, there is sufficient CNC investment in the Jag1CKO midface as demonstrated using Wnt1-Cre; Jag1 F/F; R26R to avoid the severe craniofacial deformities seen with aberrant CNC migration (i.e. Wnt1-Cre; Rac1 F/F) (16). The midface of the Jag1CKO is stunted due to a reduced cell population and the deficient extracellular matrix, leading to a deficient maxilla that ultimately leads to poor feeding and death. The mandible of the Jag1CKO mice is unaffected, similar to what is observed in Alagille syndrome patients. Similarly, as demonstrated in zebrafish, Endothelin1 guides mandibular development whereas Jagged1 is not required (10).

Hypoplasia of the midface in the Jag1CKO is associated with shortened palate shelves, decreased cellular proliferation and reduced extracellular matrix. Notch signaling plays a critical role in the relationship between the extracellular matrix and mesenchyme during embryogenesis and perturbation of their interaction leads to aberrant development (17,18). The mesenchymal cells' interaction with the extracellular matrix generates the scaffolding for many processes, including providing the foundation for proper bone and vascular formation (18). The production of extracellular matrix, especially HA, is necessary for palate elevation and elongation, and we identified a similar lack of HA in the Jag1 CKO palate shelves, leading to a defect in palatal elongation (19,20).

Jagged1 signaling in the palate: cell autonomous and non-cell autonomous?

The Notch signaling pathway is necessary for cell cycle progression and extracellular matrix production during embryogenesis (21,22). The reduction in cellular proliferation and extracellular matrix in the Jag1 CKO mice used in this study suggests that Jagged1 acts on the CNC in a cell-autonomous fashion during midface development. The deletion of Jagged1 in the CNC did not appreciably affect CNC migration as shown in Supplementary Material, Fig. S2, suggesting that Jagged1 is required in a temporal fashion during palate formation. During CNC migration, the expression of Jagged1 in the cranial region is qualitatively increased. This finding suggests that Jagged1 expression in the cranial region is produced by cells adjacent to the CNC at E8.5 (Supplementary Material, Fig. S2B); however, we did not quantitate these findings. We are currently working to further elucidate the cell lineage responsible for expression of Jagged1 at E8 in the cranial region, as these data suggest non-autonomous cranial production of Jagged1. Despite the fact that Jagged1 is required in a temporal fashion during palate development, the reduction in HA at P14 demonstrates a permanent change in the ability of palatal cells to secrete HA. It remains unknown as to whether Jagged1 signals to the CNC via Notch receptors in the CNC progenitors comprising the competence zone to affect the cell cycle and extracellular matrix production in a cell-autonomous fashion and which Notch receptors are necessary in the CNC. Conversely, it is also possible that Jagged1 could signal to Notch receptors on adjacent non-CNC cells, which then in turn indirectly influence the CNC to guide cell cycle and extracellular matrix production. Midfacial tissues have abundant expression of Notch1-3, and Notch1 was found to be a necessary receptor during palatal epithelium development (23). The cardiac and biliary phenotype of Alagille syndrome was phenocopied in the Jag1 +/−; Notch2 +/− mice; however, they lacked typical craniofacial and vertebral defects, including 50% who survived past birth, suggesting that Jagged1 may signal through a different receptor during facial formation (24). Notch 3−/− and Notch 4−/− mice are viable and do not have a reported facial phenotype, all of which suggests that Notch1 is the best candidate for the reception of Jagged1 signaling in the CNC (23–26). To determine whether Jagged1 signaled via Notch1 in the CNC, we conditionally deleted Notch1 in the CNC. N1CKO mice do not display alterations in cellular proliferation, lead to defects in palatal elongation or have alterations in facial width. Unfortunately, they die at birth for unknown reasons thus preventing us from fully analyzing their midface development (Fig. 4). Constitutively, active Notch signaling within the neural crest lineage, using the Prx1-Cre Rosa-NICDF/+ mouse line, results in defects during dorsal cranial development, but the midface was normal (22). The conditional deletion of the protein O-fucosyltransferase 1(Pofut1), a modifier of all Notch receptors that is required for Notch receptor signaling, in the CNC using Wnt1-Cre did not lead to obvious craniofacial defects. Taken together, these data suggest that Notch receptors may not be required in the CNC for midface development (27). Therefore, we conclude that Jagged1 may be acting in a non-cell-autonomous fashion signaling to Notch receptors on the adjacent non-CNC mesenchymal cells influencing vascular branching, indirectly mispatterning the midface (Fig. 3).

Vascular formation in the embryo occurs very early during development and is dependent on Notch signaling (28). Vasculogenesis involves angioblasts and non-CNC mesodermally derived endothelial cells assembling a primary vascular plexus. The primary vascular plexus is then dynamically remodeled during angiogenesis that involves sprouting of new vessels to form vascular channels and specification of each vessel's phenotype (artery, vein, capillary). Craniofacial vasculature formation occurs with CNC investing around non-CNC mesenchymal endothelial cells that give rise to the vasculature (29). During vasculogenesis stalk, endothelial cells support the adjacent tip cells where vascular sprouting and branching occurs. Many of the Notch genes participate during formation of the vascular network, including Jagged1. Loss of Jagged1 and Notch1 signaling is embryonic lethal by E10.5 in mice, each having a similar cranial vascular hemorrhage pattern, suggesting that they may signal in a common pathway during facial development (15,26). Conditional removal of Jagged1 from vasculature using Tie2-cre;Jag1 Flox/Flox mice led to cranial hemorrhage similar to that found in the Jag1−/− and Notch1−/− mice (30). The reduction in vascular branching of the Jag1 CKO mice in this study becomes apparent at E14.5 and is pronounced at E15.5 with reduced vascular branching and irregularly shaped vessels in the Jag1 CKO (Fig. 3). At P14, there is persistence of reduced vascular branching in the Jag1 CKO mice and reduced vascular smooth muscle investment of palatal vessels also occurred in the Jag1 CKO mice (Fig. 3). The obvious vascular branching differences at E14.5 and P14 likely lead to poor palatal elongation and reduced anterior facial width, which allows palate fusion to occur, but results in stunted maxillary growth. The reduction in maxillary growth results in the midface hypoplasia synonymous with the Alagille phenotype of a flattened midface. These data further suggest that CNC Jagged1 is required in a non-cell-autonomous role during vasculogenesis in the adjacent non-CNC cells. It remains unknown whether the reduction in vascular smooth muscle staining in the P14 Jag1 CKO mice is directly related to the loss of Jagged1 in the CNC or is due to the reduced number and diameter of vessels in the palate.

To test the ability of the Notch pathway to compensate for the loss of Jagged1 in the CNC, the mRNA of isolated midfacial tissues from the Jag1 CKO mice were analyzed. The expression of the Jag1 and the Notch receptors (1–4) was reduced in the Jag1 CKO mice midface early in development at E12.5, but by E14.5 there was up-regulation of Notch1, Notch3 and Notch4, which could represent a compensatory response for the early loss of Jagged1; however, Notch2 was not increased (Fig. 5). The approach we took to analyze expression levels of Notch pathway components was to isolate whole embryonic midfacial tissues, containing cells from neural crest, epithelial and para-axial mesenchymal lineages. Therefore, the extent of Jagged1 expression changes within the neural crest lineage may be masked at E13.5 and E14.5 by elevation of Jagged1 in the epithelial and para-axial cell lineages, as suggested in Supplementary Material, Figure S1D and E. By E15.5, there is normalization of Jagged1 and Notch1-4 receptor mRNA, suggesting that either the non-CNC Jagged1 (epithelium or non-CNC mesenchyme) levels mask the loss of Jagged1 in the CNC or that Jagged1-Notch expression is not normally expressed at this time point as illustrated in Figure 3C–F.

These data suggest that the normalization of mRNA at E15.5 represents a reduction in Jagged1–Notch1-4 expression that is also demonstrated by the noticeable reduction in Jagged1 signal in the E15.5 palate or that increased epithelial expression of Jagged1–Notch1-4 is compensating for the loss of Jagged1 in the CNC (Fig. 3C–F and Supplementary Material, Fig. S1E and F). This underscores the temporal requirement of Jagged1 at E14.5 in the palate shelves for cell cycle progression, extracellular matrix production and vasculogenesis. These data suggest that Jagged1 signaling during midfacial development may occur in a: (i) Notch1-independent manner within the CNC between adjacent cells expressing other Notch receptors; (ii) cell-autonomous Jagged1-dependent, but Notch-independent, manner in the CNC; (iii) non-cell-autonomous manner to adjacent Notch expressing non-CNC cells to guide vascular development, indirectly influencing proliferation and extracellular matrix production of the CNC. These possibilities are currently being tested.

The necessity of Jagged1 in the CNC during midfacial formation is clear from the identical craniofacial malformations present in Alagille syndrome patients and Jag1CKO mice. The presence of a palatal anomaly in the Jag1 CKO mice is likely the reason that some Alagille syndrome patients have been reported to have cleft palate highlighting the highly variable expressivity of the Alagille syndrome phenotype. Our findings suggest a temporal requirement of Jagged1 signaling at E14.5 during palate formation for proper cellular division and extracellular matrix production contributing to palatal elongation that persists through P14. The long-term midfacial skeletal abnormalities in the Jag1 CKO mice are likely related to the attenuated vascular network generated by Jagged1 deficiency in the CNC, reduced CNC cellular proliferation and the deficient extracellular matrix. The reduced vascular branching formation may also explain why Alagille syndrome patients are at much greater long-term risk from cerebral vascular bleeding in up to 15% of patients (4). Further advances need to be made in the understanding of the role of Jagged1 during craniofacial formation, the role of Jagged1 in craniofacial vasculogenesis and extracellular matrix formation and the mechanism by which Jagged1 signals to the CNC.

MATERIALS AND METHODS

Murine model

To test the requirement of Jagged1 and Notch1 in the CNC, we conditionally deleted them in the CNC using Wnt1-Cre. Wnt1-Cre and Notch1 F/F mice were obtained from Jax labs and Jagged1 F/F were a gift from Dr Kathleen Loomes (12,13,31). Four successive generations were bred to generate Wnt1-Cre Jagged1 F/F and Wnt1-Cre Notch 1 F/F mice on a C57B6 background. Once the lethal craniofacial phenotype was established at ∼1 month in the Wnt1-Cre Jagged1 F/F mice, developmental stages of the midface were evaluated at E13.5, 14.5 and 15.5 via interval breeding. All procedures and protocols were done in accordance with a Vanderbilt IACUC approved protocol.

Tissue preparation

Autopsy was performed on the first two successful generations of Wnt1- Cre Jagged1 F/F mice. To further assess craniofacial development, embryonic dissection was undertaken. Fixation in 4% paraformaldehyde was followed by tissue prepared for paraffin embedding via ethanol dehydration, or frozen optimal cutting temperature media (OCT) embedding after sucrose dehydration.

H + E staining

Paraffin tissues underwent rehydration prior to a 5 min staining in Meyer's hematoxylin. They were subsequently re-dehydrated and counterstained with eosin via standard procedures. Palatal and facial measurements were performed using Nikon E800 camera with Spot software as previously described by Goudy et al. (19). Palatal elongation was defined as a change in the length from ‘hinge’ of palate defined in Rice et al. to tip of palate at the medial edge epithelium (32).

Skeletal preparations and craniometrics

Ethanol fixation of the eviscerated and denuded Wnt1-Cre Jagged1 F/F carcass was followed by acetic acid blocking and cartilaginous staining via Alcian Blue. The soft tissues were digested from the carcass using KOH 2% solution, and the osseous structure was counterstained with Alizarin Red S. The remaining soft tissue was cleared with glycerol, and the skeletal preparation was permanently fixed in a glycerol–ethanol solution. Specific measurements were taken using digital calipers in three separate biological samples, using Richtsmeier as a reference (14).

Immunoflorescence

Florescence staining was performed on serial coronal sectioning of OCT embedded tissue. Eight micrometer thick sections were thawed and rehydrated in phosphate buffered saline. The sections were treated with 0.1% Tween and blocked with 10% donkey serum. The following antibodies were used: 1° = Jagged 1 (R+D Systems, AF599, Lot-BHP02), 1° = PECAM/CD31 Antimouse/Rat IgG (550274, BD Pharmingen), 2° = Cy3 AntiGoat/Donkey, Alexa 488 AntiRat/Goat Polyclonal. Hard Mount with Dapi® was used to counterstain (Vectastain). Imaging was performed on a Nikon E800 microscope; and images were obtained with SPOT® software. Branching of the blood vessels on PECAM staining was quantified on the E14.5 palate shelves in three separate biological samples. A branch point was defined as a PECAM-positive vascular channel that was attached to, but divergent from, an adjacent channel.

PhosphoHistone-3 immunohistochemistry

Seven micrometer paraffin coronal sections were dewaxed and rehydrated prior to unmasking with a boil in 10 mm Tris, pH 10.0. The sections were quenched in 3% hydrogen peroxide, and then blocked with 5% goat serum. The sections were incubated with 1° antibody—Phospho Histone 3 (9701L, Cell Signaling Technology) and stained using ABC Kit® according to the standard protocol (Vector). Diaminobenzidine reagent was carefully added to the sections and watched under Nikon E800 for development. The sections were immediately washed and counterstained with hematoxylin. Brightfield microscopy (Nikon E800) was used for cell counting in the palate shelves, followed by statistical analysis using Microsoft Excel®.

HA histochemistry

Standard paraffin sections were dewaxed and rehydrated prior to unmasking with dilute bovine testicular hyaluronidase. The sections were washed and then blocked with 3% acetic acid. Alcian Blue (pH 2.5) was utilized for staining. The sections were washed prior to counterstaining with nuclear fast red.

Real-time quantitative polymerase chain reaction

Palate shelves were surgically removed from three separate biological specimens E12.5, E13.5, E14.5, E15.5 murine embryos, respectively. The tissue was fresh processed and RNA prepared with Trizol Reagent® following the standard protocol (Invitrogen). RNA was quantified on a Nanodrop® spectrometer. We used 5 μg of RNA to generate cDNA using the SuperScript III Reverse Transcriptase kit (Invitrogen). Quantitative PCR was performed on iQ5 Multicolor Real-Time PCR Detection System® using iQTM SYBR Green Supermix kit (BioRad). Sequences for the mouse Notch 1-4 and Jagged 1 primers have been published (33). All samples were run in triplicate. The sample concentrations were normalized with mouse Gapdh control.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This work was supported by K08 DE017953 (S.G.), HL097195, HL086324 (L.S.P.), DK078640 (S.S.H.), HL086964 (S.B.), HL092551 (S.B.) and HL105334 (S.B.).

Supplementary Material

Supplementary Data
supp_21_6_1374__index.html (1.1KB, html)

ACKNOWLEDGEMENTS

We would like to thank the expert technical assistance of Yan Zhao.

Conflict of Interest statement. None declared.

REFERENCES

  • 1.Alagille D., Odievre M., Gautier M., Dommergues J.P. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental, and sexual development, and cardiac murmur. J. Pediatr. 1975;86:63–71. doi: 10.1016/s0022-3476(75)80706-2. [DOI] [PubMed] [Google Scholar]
  • 2.Oda T., Elkahloun A.G., Pike B.L., Okajima K., Krantz I.D., Genin A., Piccoli D.A., Meltzer P.S., Spinner N.B., Collins F.S., et al. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat. Genet. 1997;16:235–242. doi: 10.1038/ng0797-235. [DOI] [PubMed] [Google Scholar]
  • 3.Li L., Krantz I.D., Deng Y., Genin A., Banta A.B., Collins C.C., Qi M., Trask B.J., Kuo W.L., Cochran J., et al. Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nat. Genet. 1997;16:243–251. doi: 10.1038/ng0797-243. [DOI] [PubMed] [Google Scholar]
  • 4.Emerick K.M., Rand E.B., Goldmuntz E., Krantz I.D., Spinner N.B., Piccoli D.A. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology. 1999;29:822–829. doi: 10.1002/hep.510290331. [DOI] [PubMed] [Google Scholar]
  • 5.Piccoli D.A., Spinner N.B. Alagille syndrome and the Jagged1 gene. Semin. Liver Dis. 2001;21:525–534. doi: 10.1055/s-2001-19036. [DOI] [PubMed] [Google Scholar]
  • 6.Hoffenberg E.J., Narkewicz M.R., Sondheimer J.M., Smith D.J., Silverman A., Sokol R.J. Outcome of syndromic paucity of interlobular bile ducts (Alagille syndrome) with onset of cholestasis in infancy. J. Pediatr. 1995;127:220–224. doi: 10.1016/s0022-3476(95)70298-9. [DOI] [PubMed] [Google Scholar]
  • 7.Legius E., Fryns J.P., Eyskens B., Eggermont E., Desmet V., de Bethune G., Van den Berghe H. Alagille syndrome (arteriohepatic dysplasia) and del(20)(p11.2) Am. J. Med. Genet. 1990;35:532–535. doi: 10.1002/ajmg.1320350419. [DOI] [PubMed] [Google Scholar]
  • 8.Kamath B.M., Loomes K.M., Oakey R.J., Emerick K.E., Conversano T., Spinner N.B., Piccoli D.A., Krantz I.D. Facial features in Alagille syndrome: specific or cholestasis facies? Am. J. Med. Genet. 2002;112:163–170. doi: 10.1002/ajmg.10579. [DOI] [PubMed] [Google Scholar]
  • 9.Gritli-Linde A. Molecular control of secondary palate development. Dev. Biol. 2007;301:309–326. doi: 10.1016/j.ydbio.2006.07.042. [DOI] [PubMed] [Google Scholar]
  • 10.Zuniga E., Stellabotte F., Crump J.G. Jagged-Notch signaling ensures dorsal skeletal identity in the vertebrate face. Development. 2010;137:1843–1852. doi: 10.1242/dev.049056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chai Y., Jiang X., Ito Y., Bringas P., Jr, Han J., Rowitch D.H., Soriano P., McMahon A.P., Sucov H.M. Fate of the mammalian cranial neural crest during tooth and mandibular morphogenesis. Development. 2000;127:1671–1679. doi: 10.1242/dev.127.8.1671. [DOI] [PubMed] [Google Scholar]
  • 12.Danielian P.S., Muccino D., Rowitch D.H., Michael S.K., McMahon A.P. Modification of gene activity in mouse embryos in utero by a tamoxifen-inducible form of Cre recombinase. Curr. Biol. 1998;8:1323–1326. doi: 10.1016/s0960-9822(07)00562-3. [DOI] [PubMed] [Google Scholar]
  • 13.High F.A., Jain R., Stoller J.Z., Antonucci N.B., Lu M.M., Loomes K.M., Kaestner K.H., Pear W.S., Epstein J.A. Murine Jagged1/Notch signaling in the second heart field orchestrates Fgf8 expression and tissue-tissue interactions during outflow tract development. J. Clin. Invest. 2009;119:1986–1996. doi: 10.1172/JCI38922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Richtsmeier J.T., Zumwalt A., Carlson E.J., Epstein C.J., Reeves R.H. Craniofacial phenotypes in segmentally trisomic mouse models for Down syndrome. Am. J. Med. Genet. 2002;107:317–324. doi: 10.1002/ajmg.10175. [DOI] [PubMed] [Google Scholar]
  • 15.Xue Y., Gao X., Lindsell C.E., Norton C.R., Chang B., Hicks C., Gendron-Maguire M., Rand E.B., Weinmaster G., Gridley T. Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1. Hum. Mol. Genet. 1999;8:723–730. doi: 10.1093/hmg/8.5.723. [DOI] [PubMed] [Google Scholar]
  • 16.Thomas P.S., Kim J., Nunez S., Glogauer M., Kaartinen V. Neural crest cell-specific deletion of Rac1 results in defective cell-matrix interactions and severe craniofacial and cardiovascular malformations. Dev. Biol. 2010;340:613–625. doi: 10.1016/j.ydbio.2010.02.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nehring L.C., Miyamoto A., Hein P.W., Weinmaster G., Shipley J.M. The extracellular matrix protein MAGP-2 interacts with Jagged1 and induces its shedding from the cell surface. J. Biol. Chem. 2005;280:20349–20355. doi: 10.1074/jbc.M500273200. [DOI] [PubMed] [Google Scholar]
  • 18.Chen X.D. Extracellular matrix provides an optimal niche for the maintenance and propagation of mesenchymal stem cells. Birth Defects Res. C. Embryo Today. 2010;90:45–54. doi: 10.1002/bdrc.20171. [DOI] [PubMed] [Google Scholar]
  • 19.Goudy S., Law A., Sanchez G., Baldwin H.S., Brown C. Tbx1 is necessary for palatal elongation and elevation. Mech. Dev. 2010;127:292–300. doi: 10.1016/j.mod.2010.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Snyder-Warwick A.K., Perlyn C.A., Pan J., Yu K., Zhang L., Ornitz D.M. Analysis of a gain-of-function FGFR2 Crouzon mutation provides evidence of loss of function activity in the etiology of cleft palate. Proc. Natl Acad. Sci. USA. 2010;107:2515–2520. doi: 10.1073/pnas.0913985107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lai E.C. Notch signaling: control of cell communication and cell fate. Development. 2004;131:965–973. doi: 10.1242/dev.01074. [DOI] [PubMed] [Google Scholar]
  • 22.Dong Y., Jesse A.M., Kohn A., Gunnell L.M., Honjo T., Zuscik M.J., O'Keefe R.J., Hilton M.J. RBPjkappa-dependent Notch signaling regulates mesenchymal progenitor cell proliferation and differentiation during skeletal development. Development. 0000;137:1461–1471. doi: 10.1242/dev.042911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Casey L.M., Lan Y., Cho E.S., Maltby K.M., Gridley T., Jiang R. Jag2-Notch1 signaling regulates oral epithelial differentiation and palate development. Dev. Dyn. 2006;235:1830–1844. doi: 10.1002/dvdy.20821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.McCright B., Lozier J., Gridley T. A mouse model of Alagille syndrome: Notch2 as a genetic modifier of Jag1 haploinsufficiency. Development. 2002;129:1075–1082. doi: 10.1242/dev.129.4.1075. [DOI] [PubMed] [Google Scholar]
  • 25.Krebs L.T., Xue Y., Norton C.R., Sundberg J.P., Beatus P., Lendahl U., Joutel A., Gridley T. Characterization of Notch3-deficient mice: normal embryonic development and absence of genetic interactions with a Notch1 mutation. Genesis. 2003;37:139–143. doi: 10.1002/gene.10241. [DOI] [PubMed] [Google Scholar]
  • 26.Krebs L.T., Xue Y., Norton C.R., Shutter J.R., Maguire M., Sundberg J.P., Gallahan D., Closson V., Kitajewski J., Callahan R., et al. Notch signaling is essential for vascular morphogenesis in mice. Genes Dev. 2000;14:1343–1352. [PMC free article] [PubMed] [Google Scholar]
  • 27.Okamura Y., Saga Y. Notch signaling is required for the maintenance of enteric neural crest progenitors. Development. 2008;135:3555–3565. doi: 10.1242/dev.022319. [DOI] [PubMed] [Google Scholar]
  • 28.Iso T., Hamamori Y., Kedes L. Notch signaling in vascular development. Arterioscler. Thromb. Vasc. Biol. 2003;23:543–553. doi: 10.1161/01.ATV.0000060892.81529.8F. [DOI] [PubMed] [Google Scholar]
  • 29.Yoshida T., Vivatbutsiri P., Morriss-Kay G., Saga Y., Iseki S. Cell lineage in mammalian craniofacial mesenchyme. Mech. Dev. 2008;125:797–808. doi: 10.1016/j.mod.2008.06.007. [DOI] [PubMed] [Google Scholar]
  • 30.High F.A., Lu M.M., Pear W.S., Loomes K.M., Kaestner K.H., Epstein J.A. Endothelial expression of the Notch ligand Jagged1 is required for vascular smooth muscle development. Proc. Natl Acad. Sci. USA. 2008;105:1955–1959. doi: 10.1073/pnas.0709663105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kiernan A.E., Xu J., Gridley T. The Notch ligand JAG1 is required for sensory progenitor development in the mammalian inner ear. PLoS Genet. 2006;2:e4. doi: 10.1371/journal.pgen.0020004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Rice R., Spencer-Dene B., Connor E.C., Gritli-Linde A., McMahon A.P., Dickson C., Thesleff I., Rice D.P. Disruption of Fgf10/Fgfr2b-coordinated epithelial-mesenchymal interactions causes cleft palate. J. Clin. Invest. 2004;113:1692–1700. doi: 10.1172/JCI20384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Sparks E.E., Huppert K.A., Brown M.A., Washington M.K., Huppert S.S. Notch signaling regulates formation of the three-dimensional architecture of intrahepatic bile ducts in mice. Hepatology. 0000;51:1391–1400. doi: 10.1002/hep.23431. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_21_6_1374__index.html (1.1KB, html)
supp_ddr575_ddr575supp.docx (15.7MB, docx)

Articles from Human Molecular Genetics are provided here courtesy of Oxford University Press

RESOURCES