Let There Be Sight

levels. This fits well with the observation that the repul-1998). It remains to be seen, however, whether these or 1133.102, 363–375. tions be in mediating guidance decisions at choice points? Another possible site for this mechanism to operate is at the ventral midline of the CNS. In vertebrates, commissural axon crossing of the ventral midline of the spinal cord is regulated by a combination of long-range and short-range, attractive and repulsive guidance cues. Axons are initially attracted to the midline by the long-range chemoattractant netrin-1. Once at the floor plate, local, short-range cues are thought to play a role in Neurogenesis begins when some members of a popula-directing these axons to cross the ventral midline. Inter-tion of proliferative cells withdraw from the mitotic cycle estingly, it has been a long-standing observation that and begin to differentiate as neurons. The withdrawal L1 is specifically upregulated by axons after crossing and differentiation are both readily detectable, and nu-of the floor plate (Dodd et al., 1988). This upregulation merous reports have implicated the expression of of L1 surface expression occurs concomittant with " proneural " genes in the decision to follow that pathway, changes in the axonal responsiveness to cues at the but information about the control of proneural gene ex-floor plate. For instance, postcrossing axons lose the pression is not so plentiful. In this issue of Neuron, Masai ability to be attracted to netrin-1 (Shirasaki et al., 1998). et al. (2000) provide new insights on this subject and on Furthermore, in a recent paper in Cell, Zou and col-the conservation of developmental signaling mecha-leagues (2000) showed that upon crossing, axons also nisms across phyla. Their paper shows that the onset alter their responsiveness to chemorepulsive cues. In of retinal neurogenesis in zebrafish is triggered by the particular, while uncrossed axons are unresponsive to expression of ath5, a homolog of atonal. This proneural the Sema3B and 3F, after crossing these axons are gene is involved in neurogenesis in the fly eye, and has repelled by Sema3B and 3F, presumably via activation known homologs in other vertebrates (Xenopus: Xath5, of the NP-2 receptor. Interestingly, commissural axons Kanekar et al., 1997; and mouse: Math5, Brown et al., express NP-2 along the length of their trajectory, raising 1998), where it is also preferentially expressed in the the question of how this altered responsiveness to retina. The major contribution of Masai et al. is to identify Sema3B …