Cdk5 selectively affects the migration of different populations of neurons in the developing spinal cord.

It has been shown that cyclin-dependent kinase 5 (Cdk5) is crucial for neuronal migration and survival in the brain. However, the role of Cdk5 in neuronal migration in the spinal cord has never been investigated. The present study is the first to show that Cdk5 affects the migration of different populations of neurons in the developing spinal cord. In the absence of Cdk5, at least four neuronal populations failed to migrate to their final destinations: sympathetic and parasympathetic preganglionic neurons, as well as dorsally originating and ventrally originating (U-shaped group) diaphorase-positive dorsal horn interneurons. In contrast, the migration of somatic motor neurons and various types of ventral and dorsal interneurons was unaffected by the absence of Cdk5. Moreover, our results suggest that Cdk5-dependent migration in the developing spinal cord is axon- or glial fiber-mediated. Finally, our results show that sympathetic preganglionic neurons and somatic motor neurons in Cdk5-deficient mice continue to extend processes and project toward their normal target areas, suggesting that Cdk5 has no obvious effects on axonal outgrowth and guidance mechanisms of these two neuronal populations in spinal cord development.

Migration of sympathetic preganglionic neurons in the spinal cord is regulated by Reelin-dependent Dab1 tyrosine phosphorylation and CrkL.

The actions of Reelin in neuronal positioning in the developing cortex and cerebellum are relayed by Src-family kinase (SFK)-mediated phosphorylation of Dab1. Biochemical studies show that after phosphorylation Dab1 binds to an adaptor protein, CrkL. Whether CrkL is important for Reelin signaling in vivo is unknown, because crkl(-/-) embryos die before cortical development is complete. In the developing spinal cord, Reelin and components of its signaling pathway, VLDLR, ApoER2, and Dab1, control the positioning of sympathetic preganglionic neurons (SPN); however, it is not known whether SFKs or Dab1 tyrosine phosphorylation is required. In the present study, we asked whether Reelin-controlled SPN migration depends on tyrosine phosphorylation of Dab1 by SFKs and whether CrkL is involved in SPN migration. To answer these questions, we examined the location of SPN in various mutant mouse embryos. Results showed that, in dab1(5F/5F) embryos, which express a nonphosphorylated mutant of Dab1, and in src(-/-)fyn(-/-) double knockout embryos, the location of SPN is identical to that of reeler. These results show that tyrosine phosphorylation of Dab1 by SFKs is required for Reelin-regulated SPN positioning. In addition, we found that SPN migration in crkl(-/-) showed a partial reeler phenotype, suggesting a partial loss of response of SPN to Reelin signaling. These results suggest a role for CrkL in the Reelin signaling pathway to control neuronal migration.

Location of preganglionic neurons is independent of birthdate but is correlated to reelin-producing cells in the spinal cord.

Many studies suggest that during neuronal development the birthdate of a neuron appears to have significant consequences for its ultimate location and identity. Our past study shows that sympathetic preganglionic neurons (SPN) in mice lacking the reelin gene settle in abnormal positions in the spinal cord. In the present study we determined that birthdate is not a factor contributing to the abnormal position of SPN in reeler. In both normal and reeler mice the period of neurogenesis of SPN was similar, and the final location of SPN in the spinal cord was independent of birthdate. Additionally, we have identified at least two types of ventral interneurons, V1 and V2, that are involved in the production of Reelin and the positioning of SPN in the spinal cord.

Components of the reelin signaling pathway are expressed in the spinal cord.

The Reelin signaling pathway in the brain involves the binding of Reelin to very-low-density lipoprotein receptors (VLDLR) and apolipoprotein E receptor 2 (ApoER2). After Reelin binds the lipoprotein receptors on migrating neurons, the intracellular adaptor protein Disabled-1 (Dab1) becomes phosphorylated, ultimately resulting in the proper positioning of cortical neurons. Previous work showed that Reelin also affects the positioning of sympathetic preganglionic neurons (SPN) in the spinal cord (Yip et al. [2000] Proc Natl Acad Sci USA 97:8612-8616). We asked in the present study whether components of the Reelin signaling pathway in the brain also function to control SPN migration in developing spinal cord. Results showed that Reelin and reelin mRNA are found adjacent to migrating SPN. In addition, dab1 mRNA and protein are expressed by migrating SPN, and dab1-null mice show abnormal SPN migration similar to that seen in reeler. Finally, vldlr and apoER2 are also expressed in migrating SPN, and mice lacking both vldlr and apoER2 show aberrant SPN location that is identical to that of reeler and dab1-null mice. Because molecules known to be involved in Reelin signaling in the brain are present in the developing spinal cord, it is likely that the Reelin signaling pathways in the brain and spinal cord function similarly. The relative simplicity of the organization of the spinal cord makes it a potentially useful model system with which to study the molecular and cellular function of the Reelin signaling pathway in control of neuronal migration.

Migratory pathway of sympathetic preganglionic neurons in normal and reeler mutant mice.

Our previous study showed that the migration of sympathetic preganglionic neurons (SPN) in the spinal cord is affected in the reeler mutant. The present study, using morphometric analysis to describe and compare the location of SPN at progressive developmental stages, provides detailed information on how SPN migrate in the presence or absence of the reelin gene. We found that the initial migration (prior to E11.5) of SPN from the neuroepithelium to the ventrolateral spinal cord is similar in both control (wild-type and heterozygous) and reeler mice. However, as development progressed (E12.5-E15.5), SPN in control mice migrated dorsally toward the intermediate lateral spinal cord region, where 80% settled to form the intermediolateral column (IML); the rest migrated medially to locations between the IML and the central canal. In reeler, 80% of SPN migrated dorsomedially to cluster around the central canal, with the rest distributed between the central canal and the intermediate lateral spinal cord region. The present study also examined the relationship among SPN, Reelin, and radial glial fibers in control and reeler mice. Confocal microscopic studies showed that during their initial migration, SPN in both control and reeler mice were closely apposed to radial glial fibers in the ventrolateral spinal cord. The majority of SPN in control mice then migrated dorsolaterally, in a direction perpendicular to radial glial fibers, to form the IML. In contrast, the majority of SPN in reeler migrated in the same orientation as radial glial fibers back toward the central canal, instead of migrating dorsolaterally to form the IML. A possible explanation for these results is that Reelin acts to prevent SPN from back-migration on radial glial fibers toward the central canal.