We also examined the requirement of DLK in sensory axons that are the major component of the sciatic nerve. We deleted DLK expression in sensory neurons with Wnt1-Cre ( Danielian et al., 1998) ( Figure S1B), crushed sciatic nerves of KO and littermate controls, and assessed sensory axon regeneration by measuring the length of axons growing past the crush site. To label regenerating axons, we stained longitudinal nerve sections with a growth-associated neuronal protein, superior cervical ganglion 10 MEK activation (SCG10), which is highly expressed in developing and regenerating axons

( Mason et al., 2002). SCG10 levels in uninjured sciatic nerves were not significantly different between WT and DLK KO, though there was a tendency for higher levels of SCG10 in the DLK KO (26% ± 10% increase [mean ± SEM]; n = 5; p = 0.07). When WT sensory axons were allowed to regrow for 3 days after injury, they robustly regenerated beyond the site of lesion to a distance of approximately 4 mm, consistent with previous findings ( La Fleur et al., 1996). However, the length of regenerating axons is reduced in the absence of DLK ( Figure 1B). Immunolabeling with another marker of regenerating

axons, GAP43 ( Abe et al., 2010), shows similar results ( Figure S2). To quantify the regenerative deficit, we measured the distance from the crush site to the location where the SCG10 level is reduced to half of its level at the crush site and defined that as regeneration index ( Abe Protease Inhibitor Library mouse et al., 2010). Loss of DLK results in 2-fold reduction in the regeneration index (p < 0.05) ( Bay 11-7085 Figure 1B), demonstrating that DLK promotes sensory axon regeneration in vivo. In addition to sensory neurons, the Wnt1-Cre driver

line is also active in other neural crest lineages including Schwann cells ( Danielian et al., 1998). Since changes in myelination and the reaction of Schwann cells to injury may indirectly affect axonal structure and growth ( Jessen and Mirsky, 2008), we examined whether myelin is normally formed in Wnt1-Cre conditional DLK KO mice. No obvious abnormalities were noted in sciatic nerve axons from DLK KO mice by light and electron microscopic analysis ( Figure S1C). Additionally, we assessed a battery of myelination parameters such as cumulative g-ratio and fiber-width distribution by semiautomated nerve histomorphometry (see Supplemental Experimental Procedures) and found no alterations in the absence of DLK ( Figure S1C). This quantitative analysis also demonstrates that axon caliber distributions in DLK-deficient nerves are indistinguishable from control preparations. In addition, we studied the cellular reactions 3 days after nerve injury, when Schwann cells lose their myelin sheaths and dedifferentiate. Ultrastructural assessment shows normal dedifferentiation features of Schwann cells after sciatic nerve transection in DLK KO mice ( Figure S1D).