Tissue-specific responses to aberrant FGF signaling in complex head phenotypes

Abstract

Background: The role of fibroblast growth factor and receptor (FGF/FGFR) signaling in bone development is well studied, partly because mutations in FGFRs cause human diseases of achondroplasia and FGFR-related craniosynostosis syndromes including Crouzon syndrome. The FGFR2c C342Y mutation is a frequent cause of Crouzon syndrome, characterized by premature cranial vault suture closure, midfacial deficiency, and neurocranial dysmorphology. Here, using newborn Fgfr2c(C342Y/+) Crouzon syndrome mice, we tested whether the phenotypic effects of this mutation go beyond the skeletal tissues of the skull, altering the development of other non-skeletal head tissues including the brain, the eyes, the nasopharynx, and the inner ears.

Results: Quantitative analysis of 3D multimodal imaging (high-resolution micro-computed tomography and magnetic resonance microscopy) revealed local differences in skull morphology and coronal suture patency between Fgfr2c(C342Y/+) mice and unaffected littermates, as well as changes in brain shape but not brain size, significant reductions in nasopharyngeal and eye volumes, and no difference in inner ear volume in Fgfr2c(C342Y/+) mice.

Conclusions: These findings provide an expanded catalogue of clinical phenotypes in Crouzon syndrome caused by aberrant FGF/FGFR signaling and evidence of the broad role for FGF/FGFR signaling in development and evolution of the vertebrate head.

Figure 1

Combined results of PCA of skull based on Procrustes coordinates and of EDMA of landmark coordinates. A) Scatter plots of individual scores based on PCA from global skull morphology of Fgfr2c^C342Y/+ mutant mice and unaffected littermates along first and second Principal Components axes (PC1 and PC2). B) Lateral (top) and inferior (bottom) views of 3D isosurfaces of µCT reconstruction of unaffected P0 mouse skull with mandible removed displaying the 39 landmarks used in global skull shape analysis. Facial landmarks are shown in red, cranial base landmarks in green, cranial vault landmarks in blue. C-F) Results of PCA analyses based on Procrustes coordinates of regional configurations of skull landmarks on left and EDMA analysis of regional configuration of points on right. Scatter plots of individual morphologies of the facial skeleton (C), cranial base (D), cranial vault (E) and palate (F) along first two principal axes (PC1 and PC2). Since size-related differences in shape (allometry) were significant (P ≤0.05) in the analysis of the face and palate, Figs 1C and 1F represent shape variation after mathematically adjusting for correlations among shape variables due to allometric effects following (Drake and Klingenberg, 2008). Allometry did not have a significant effect on cranial base or cranial vault. EDMA results show linear distances within each regional configuration of points that show at least a 5% difference in length and are significantly different between groups by confidence intervals. Blue lines are significantly larger in mutant mice relative to unaffected littermates; fuchsia lines are significantly smaller in mutant mice.

Figure 2

Linear distances (measured between 3D landmarks) that are significantly different in Fgfr2c^C342Y/+ Crouzon syndrome mutant mice relative to unaffected littermates when landmarks from global skull are analyzed by EDMA (Lele and Richtsmeier, 2001). Blue lines are significantly larger in mutant mice relative to unaffected littermates. Fuchsia lines are significantly smaller in mutant mice. Significant differences are shown on facial skeleton and palate (lateral (A), and superior (B) views) and on cranial vault and cranial base (lateral (C) and superior view (D)). Fuller 3D views are available as videos (Videos S1 and S2).

Figure 3

Patterns of suture patency in Fgfr2c^C342Y/+ Crouzon syndrome mice and unaffected littermates at P0. Sutures are shown on µCT reconstructions of an unaffected neonatal mouse skull showing lateral view of complete skull (A) and inferior view of palate (B) with incisors at top: 1-frontal-nasal; 2-frontal-premaxilla; 3-frontal-maxilla; 4-coronal; 5-premaxilla-maxilla; 6-zygomatic-maxilla; 7-inter-premaxillary; 8: maxillary-palatine; 9: inter-palatine. C) Patterns of suture patency in Fgfr2c^C342Y/+ Crouzon syndrome mice and unaffected littermates at P0. ^a a sutural state where only the most anterior section remains patent (see Table 3). ^b a sutural state where none of the suture appears patent (see Table 3). See Table 3 for supporting quantitative information relating to sample size and patency patterns for individual sutures.

Figure 4

Results of PCA of brain based on Procrustes coordinates. Scatter plots of individual brain morphologies along PC1 and PC2 of PCA of whole brain (A), surface brain landmarks (B), and subsurface brain landmarks (C). Size related differences in shape (allometry) did not have a significant effect on shape difference of the whole brain or surface landmarks, but there was a significant allometric effect of size (P ≤ 0.05) in the analysis of the subsurface landmarks. Brain landmarks used in analysis visualized on 3D MRM reconstructions of unaffected mouse brain. Views are: D: superior; E: lateral; F : sagittal section. Landmark numbering corresponds with Table 4. For additional views and definitions of these landmarks see: http://www.getahead.psu.edu/LandmarkNewVersion/Mousebrain.html

Figure 5

Significant differences between Fgfr2c^C342Y/+ Crouzon syndrome mutant mice relative to unaffected littermates as analyzed by EDMA shown on 3D reconstruction of MRM images (A superior view, B lateral view) and a sagittal slice (C). Blue lines represent neural dimensions that are significantly larger in mutant mice relative to unaffected littermates. Fuschia lines are significantly smaller in mutant mice. Landmarks are identified in Table 4 and Fig 4.

Figure 6

Nasopharynx, vestibular canal and cochlea of Fgfr2c^C342Y/+ Crouzon mice (A) and unaffected littermates (B) segmented from MRM images showing relative locations and magnified reconstructions of soft tissue structures (C, D). Nasopharynx volumes were significantly restricted in Fgfr2c^C342Y/+ Crouzon mice (C) compared to unaffected littermates (D), while vitreous volumes of the eye were significantly larger in Fgfr2c^C342Y/+ Crouzon mice. No significant differences between Fgfr2c^C342Y/+ Crouzon mice and unaffected littermates were detected for inner ear volume.

Figure 7

Relative position of brain, globe of the eye, nasopharynx, skull, and inner ear in Fgfr2c^C342Y/+ mutant mouse (A) and unaffected littermate (B). In each panel the top view shows the superimposition of skull as visualized by µCT and soft tissues segmented from MRM; the bottom view represents MRM data only. Scales are internally consistent for each imaging modality. Skull=yellow, brain=purple, vitreous humor=green, nasopharynx=blue, cochea and vestibular canals=blue. Additional soft tissues visualized on MRM shown in blue/green.