A chemokine, SDF -1, promotes embryonic retinal ganglion cell survival and reduces the effectiveness of multiple axonal repellents

The development of the nervous system requires that neuronal processes (axons) connect precisely with other neurons or muscle cells. Some axons travel large distances through various environments in search of their targets. Extracellular signaling molecules have been identified that guide these axons. Many of these molecules are members of four families. The slits, ephrins and semaphorins function primarily as repellents. The netrins, function primarily as attractants. We have identified a new class of guidance cues that neither attract nor repel axons on their own, but instead, modulate axonal responsiveness to more traditional repellent guidance cues. The chemokine SDF-1 reduces the repellant activities of slit-2 on retinal ganglion cell axons, semaphorin 3A on DRG axons, and semaphorin 3C on sympathetic axons. This effect is mediated through an elevation of CAMP and the inactivation of rho. The spinal cords of mice missing the SDF-1 receptor CXCR4 contain aberrantly projecting axons. Axons within the gray matter of CXCR4 mutant E13.5 spinal cords are abnormally hyperfasciculated. TrkA positive cutaneous sensory axons enter the cord prematurely and have highly abnormal trajectories within the dorsal gray matter. We also found that the neurotransmitter glutamate can reduce the repellant effects of Slit-2 on retinal ganglion cells, Sema3A on dorsal root ganglion cells and Sema3C on sympathetic neurons. We show that DHPG, a specific Class I agonist for metabotropic glutamate receptors can mimic this effect of glutamate, while Class II specific agonist APDC and the Class III specific agonist AP-4 have no such modulatory effect. The Class I mGluRs are comprised of the mGluR1 and mGluR5 receptors. We find that specific mGluR1 antagonists can block the glutamate effect, while the specific mGluR5 antagonists do not. These results identify glutamate as another potential modulatory axonal guidance cue through its activation of mGluR1 receptors. Our finding that a transmitter can modulate responses to a broad range of repellents suggests a mechanism by which neural activity could affect the morphology of neurite processes. SDF-1 also promotes the survival of cultured embryonic retinal ganglion cell neurons even in the absence of other neurotrophic factors. This survival effect is largely mediated through a cAMP dependent pathway that acts through protein kinase A and MAP kinase. Mouse embryos lacking the CXCR4 receptor have a reduced number of retinal ganglion cells. We have thus identified SDF-1 and glutamate as modulators affecting the strength of axonal repellents through a cyclic nucleotide dependant signaling pathway. The SDF-1 signaling pathway is essential for normal axonal pathfinding in the spinal cord. SDF-1 may also provide generalized trophic support to neurons during their development and maturation