Effects of EPHB receptors on facial morphology: non‐additive interactions during mouse facial development

Abstract

EPHRIN-B1 mutations underlie craniofrontonasal syndrome (CFNS), and signaling interactions between EPHRIN-B1 ligand and the EPHB1, EPHB2, and EPHB3 receptors are presumed to be critical for normal craniofacial development. However, our understanding of the relative and spatial contribution of each receptor to facial morphogenesis remains incomplete. Here, we expand upon previous work by providing a detailed morphometric analysis of an E14.5 allelic seriesof Ephb1, Ephb2, and Ephb3 null mouse mutants. We assess the impact of individual receptor loss on facial morphology and compare observed facial shapes with those predicted by an additive genetic model. We found that Ephb1, Ephb2, and Ephb3 as well as facial size contribute significantly to facial shape. Predicted Principal Components (PC) scores of facial shape variation suggest a minimal global effect of Ephb1, with moderate effects of Ephb2 and Ephb3 on facial morphology. Observed scores for Ephb2 and Ephb3 homozygous single nulls align closely with predicted additive genetic scores, intermediate between controls and triple homozygous nulls. Compared to controls, triple homozygous nulls have wider orbits, inferiorly displaced nasals, and rostrally placed ear pinnae, suggesting an overall wider but shorter face. Ephb2 and Ephb3 mutants exhibit different facial shape effects, with Ephb2 homozygous nulls having infero-laterally oriented orbits and less severe shape changes in the rest of the face. Predicted and observed PC scores for Ephb1/Ephb2 double nulls are similar to Ephb2 single nulls, and Ephb2/Ephb3 double nulls are similar to triple homozygous nulls. While facial shape variation across our sample is primarily explained by additive genetic effects, there is also a non-additive Ephb1 effect that may be influencing facial shape and size. Predicted additive genetic PC scores place Ephb1 homozygous nulls close to controls, and Ephb1/Ephb3 homozygous double nulls close to Ephb3 homozygous nulls. These genotypes instead deviate from additive genetic predictions by having a relatively wider and shorter face, and overall smaller facial size. In sum, we have identified major additive genetic effects of Ephb1, Ephb2, and Ephb3 on facial shape development, as well as a more subtle non-additive effect of Ephb1. We also describe specific morphologies associated with individual or compound loss of EPHB receptors and find overlapping and distinct facial shape effects of each. More broadly, this work helps identify signaling interactions that may occur between EPHRIN-B1 and EPHB1, EPHB2, and EPHB3 receptorsand the role each plays during normal craniofacial development and CFNS.