Newborn cerebral cortical neurons migrate along radial glia to the cortical plate. Experiments using a collagen gel assay revealed that the choroid plexus repelled cerebral cortical neurons and olfactory interneuron precursors, which were mimicked by Neuro-2A cells. Fractionation of Neuro-2A-conditioned medium identified a protein of 190 kDa, equivalent to full-length Slit proteins. Indeed, it cross-reacted with an antibody against Slit2, suggesting that it is either Slit2 or another Slit protein. Further, Slit2, expressed in COS cells, repelled cerebral cortical neurons and olfactory interneuron precursors. Thus, Slit2, which is expressed by the choroid plexus and the septum, acts as a chemorepulsive factor for neuronal migration. These results suggest chemorepulsion as a guidance mechanism for neuronal migration in the developing forebrain.

Slit proteins are a family of secreted guidance proteins that can repel neuronal migration and axon growth via interaction with their cellular roundabout receptors (Robos). Here it was shown that Slit2-Robo-1 interactions were enhanced by cell-surface heparan sulfate. Removal of heparan sulfate decreased the affinity of Slit for Robo by about threefold. In addition, removal of cell-surface heparan sulfate by heparinase III abolished the chemorepulsive response to Slit2 normally shown by both the migrating neurons and growing axons. These results indicate essential roles for cell-surface heparan sulfate in the repulsive activities of Slit2.

Slit3 along with Slit1 and Slit2 comprise the Slit family of proteins. The latter two proteins are known to be involved in axon guidance and cell migration during animal development. However, little is know about the functions of Slit3. We created a Slit3-deficient mouse model from an OmniBank ES cell line with a Slit3 allele trapped by insertional mutagenesis to analyze the in vivo functions of this protein. In this model, congenital diaphragmatic hernia is the most obvious phenotype. Herniation was found to be caused by a defective central tendon (CT) of the diaphragm that remained fused with the liver. Electron microscopic analyses of the defective CT revealed disorganized collagen fibrils that failed to form tight collagen bundles. The hearts of Slit3-deficient mice have an enlarged right ventricle. In addition, 20% of homozygous mice also showed a range of kidney defects that include unilateral or bilateral agenesis of the kidney and ureter, or varying degrees of renal hypoplasia. Thus, we concluded that Slit3 is involved in the development of multiple organ systems that include the diaphragm and the kidney. Slit3-deficient mice represent a genetic animal model for physiological and pathological studies of congenital diaphragmatic hernia.