FGF/FGFR signaling coordinates skull development by modulating magnitude of morphological integration: evidence from Apert syndrome mouse models

Abstract

The fibroblast growth factor and receptor system (FGF/FGFR) mediates cell communication and pattern formation in many tissue types (e.g., osseous, nervous, vascular). In those craniosynostosis syndromes caused by FGFR1-3 mutations, alteration of signaling in the FGF/FGFR system leads to dysmorphology of the skull, brain and limbs, among other organs. Since this molecular pathway is widely expressed throughout head development, we explore whether and how two specific mutations on Fgfr2 causing Apert syndrome in humans affect the pattern and level of integration between the facial skeleton and the neurocranium using inbred Apert syndrome mouse models Fgfr2(+/S252W) and Fgfr2(+/P253R) and their non-mutant littermates at P0. Skull morphological integration (MI), which can reflect developmental interactions among traits by measuring the intensity of statistical associations among them, was assessed using data from microCT images of the skull of Apert syndrome mouse models and 3D geometric morphometric methods. Our results show that mutant Apert syndrome mice share the general pattern of MI with their non-mutant littermates, but the magnitude of integration between and within the facial skeleton and the neurocranium is increased, especially in Fgfr2(+/S252W) mice. This indicates that although Fgfr2 mutations do not disrupt skull MI, FGF/FGFR signaling is a covariance-generating process in skull development that acts as a global factor modulating the intensity of MI. As this pathway evolved early in vertebrate evolution, it may have played a significant role in establishing the patterns of skull MI and coordinating proper skull development.

Figure 1. Landmarks collected from μCT reconstructions of P0 mouse skulls.

Top: Left lateral view, superior endocranial view with vault removed and inferior view with mandible removed. Bottom: Wireframes used in Fig. 3 to display shape changes of the facial configuration of landmarks (blue) and the neurocranial configuration of landmarks (red). Codes and landmark definitions can be found in Table S1 and at our website http://getahead.psu.edu/LandmarkNewVersion/P0mouseskull_updated_applet.html.

Figure 2. Scatterplots of PLS1 scores of the facial and the neurocranial skeleton using varying subsets of individuals.

A) PLS analysis including Apert syndrome mouse models and non-mutant littermates before adjusting for allometry; B) PLS analysis including Apert syndrome mouse models and non-mutant littermates after adjusting for allometry; C) PLS analysis of non-mutant littermates after adjusting for allometry; D) PLS analysis of mutant Apert syndrome mouse models after adjusting for allometry.

Figure 3. PLS analyses among each Apert syndrome mouse model and their non-mutant littermates after removing allometry.

A) Fgfr2^+/S252W and non-mutant littermates; B) Fgfr2^+/P253R and non-mutant littermates. Associated facial and neurocranial shape changes corresponding to the first pair of PLS1 axes show similar skull MI patterns between the two models. Orange wireframes display face and neurocranium shape changes associated with positive and negative values of PLS1 in comparison to mean shape PLS1 values (grey dashed wireframe). For anatomical correspondence see Fig. 1. Note that all landmarks cannot be seen from a single skull view and we chose to display the inferior view of the skull because main shape changes occur in the palate and the anterior aspect of the neurocranium.

Figure 4. Comparison of MI within and between facial skeleton and neurocranium across Apert syndrome mouse models and non-mutant littermates.

Bar graphs with standard deviation error bars comparing the distribution of the integration index (EV, Eigenvalue variance standardized by group variance*10⁵) within the face (white bars), within the neurocranium (dark grey bars) and between the face and the neurocranium (light grey bars).