The shape of a neuron facilitates its functionality within neural circuits. Dendrites integrate incoming signals from axons, receiving excitatory input onto small protrusions called dendritic spines. Therefore, understanding dendritic growth and development is fundamental for discerning neural function. We previously demonstrated that EphA7 receptor signaling during cortical development impacts dendrites in two ways: EphA7 restricts dendritic growth early and promotes dendritic spine formation later. Here, the molecular basis for this shift in EphA7 function is defined. Expression analyses reveal that EphA7 full-length (EphA7-FL) and truncated (EphA7-T1; lacking kinase domain) isoforms are dynamically expressed in the developing cortex. Peak expression of EphA7-FL overlaps with dendritic elaboration around birth, while highest expression of EphA7-T1 coincides with dendritic spine formation in early postnatal life. Overexpression studies in cultured neurons demonstrate that EphA7-FL inhibits both dendritic growth and spine formation, while EphA7-T1 increases spine density. Furthermore, signaling downstream of EphA7 shifts during development, such that in vivo inhibition of mTOR by rapamycin in EphA7-mutant neurons ameliorates dendritic branching, but not dendritic spine phenotypes. Finally, direct interaction between EphA7-FL and EphA7-T1 is demonstrated in cultured cells, which results in reduction of EphA7-FL phosphorylation. In cortex, both isoforms are colocalized to synaptic fractions and both transcripts are expressed together within individual neurons, supporting a model where EphA7-T1 modulates EphA7-FL repulsive signaling during development. Thus, the divergent functions of EphA7 during cortical dendrite development are explained by the presence of two variants of the receptor.

Intercellular signaling via the Eph receptor tyrosine kinases and their ligands, the ephrins, acts to shape many regions of the developing brain. One intriguing consequence of Eph signaling is the control of mixing between discrete cell populations in the developing hindbrain, contributing to the formation of segregated rhombomeres. Since the thalamus is also a parcellated structure comprised of discrete nuclei, might Eph signaling play a parallel role in cell segregation in this brain structure? Analyses of expression reveal that several Eph family members are expressed in the forming thalamus and that cells expressing particular receptors form cellular groupings as development proceeds. Specifically, expression of receptors EphA4 or EphA7 and ligand ephrin-A5 is localized to distinct thalamic domains. EphA4 and EphA7 are often coexpressed in regions of the forming thalamus, with each receptor marking discrete thalamic domains. In contrast, ephrin-A5 is expressed by a limited group of thalamic cells. Within the ventral thalamus, EphA4 is present broadly, occasionally overlapping with ephrin-A5 expression. EphA7 is more restricted in its expression and is largely nonoverlapping with ephrin-A5. In mutant mice lacking one or both receptors or ephrin-A5, the appearance of the venteroposterolateral (VPL) and venteroposteromedial (VPM) nuclear complex is altered compared to wild type mice. These in vivo results support a role for Eph family members in the definition of the thalamic nuclei. In parallel, in vitro analysis reveals a hierarchy of mixing among cells expressing ephrin-A5 with cells expressing EphA4 alone, EphA4 and EphA7 together, or EphA7 alone. Together, these data support a model in which EphA molecules promote the parcellation of discrete thalamic nuclei by limiting the extent of cell mixing.