This loss of either the hypaxial or epaxial trajectory was invariably accompanied by consistent increases in diameter of the nerve pathways remaining at affected segments ( Figures 2G–2H and Figures S2P–S2Q). Without preformed motor pathways, the organization of peripheral sensory projections thus appears to desintegrate into the randomized “all-or-nothing” formation of either epaxial or hypaxial sensory pathways at the expense

of the other ( Figures 2J–2K). These data therefore reveal an absolute requirement of preformed motor projections for establishing the overall division of sensory projections into epaxial and hypaxial nerve trajectories. These observations suggested that the determination of peripheral CP-868596 ic50 sensory trajectories involves signals provided by epaxial and/or hypaxial motor axons, which prompted us to address the identity or identities of the putative signals. We have previously shown that epaxial motor axons display markedly higher levels of

the receptor tyrosine kinases EphA3 and EphA4 compared to hypaxial motor axons (Gallarda et al., 2008). Moreover, contact-dependent activation of EphA3/4 on motor growth cones by their cognate ephrin-A proteins on sensory axons effectively repels developing epaxial motor axons from sensory pathways and DRGs (Gallarda et al., 2008). Since EphAs can also elicit “reverse” signaling by activating ephrin-As, selleck screening library we asked whether EphA3/4 could play additional roles in determining sensory projections (Egea and Klein, 2007 and Pasquale,

2008). We therefore traced Non-specific serine/threonine protein kinase sensory projections in mouse embryos lacking both EphA3 and EphA4 (Epha3/4null). Epha3/4null embryos displayed severely defective formation of epaxial sensory pathways, while hypaxial projections formed normally ( Figures S3E and data not shown). We next asked whether the selective failure to form epaxial sensory projection in Epha3/4null embryos involves EphA3/4 in motor neurons or in other cell types. To test this, we generated embryos in which the Epha4 gene was selectively inactivated via Cre/loxP-mediated recombination in the motor neuron lineage of EphA3-deficient (Epha3−/−) embryos ( Figure S3A–S3D) ( Herrmann et al., 2010). This strategy took advantage of the observation that any contribution of EphA3 to peripheral axon trajectories appears to be compensated as long as the functionally redundant EphA4 remains expressed ( Figure S3F) ( Gallarda et al., 2008 and Vaidya et al., 2003). We further ruled out that the “floxed” Epha4 allele (Epha4flox) affected EphA4 function in the absence of Cre expression, by confirming that Epha3−/−;Epha4flox/flox compound embryos did not show detectable peripheral projection defects ( Figure S3F).