ween ORN Debio1347 site growth cones and the glial cells that become SZ 
glia occurs late in stage 3, and thus the signaling necessary for SZ glial migration may have occurred before the various drug treatments could take effect. Injecting drugs at earlier stages generally results in developmental arrest a short distance into the sorting zone. B: PD173074-treated animals exhibited unchanged fasciculation in traveling through the sorting zone, although they did show increased fasciculation on exiting the sorting zone. Projection depths = 35 mm in A, 45 mm in B. Another possibility is that redundancy in the signaling pathways that elicit SZ glial cell migration ensures formation of this critical region in the olfactory pathway. As for the continued migration of AN glia in PD173074-treated ” animals in the Manduca system, similar results have been reported in Drosophila antennal nerves in which glial cells express a dominant-negative form of Heartless. We have found AN glia to express EGFRs as well as FGFRs; it is possible that they depend on EGFR activation for migration and FGFR activation for survival. Survival. Activation of FGFRs is known to be essential for survival of many cell types, although this has been shown in vertebrates to depend on the particular FGF receptor activated. In M. sexta, when PD173074-treated animals were allowed to develop to stages later than stage 7, examination of the olfactory pathway revealed an extensive loss of NP, SZ and AN glial cells. This loss appears to be due to a combination of apoptosis and a reduction in proliferation. It is important to note that NP glial cells exhibit activated FGFRs at stage 3, before contact with ORNs, as well as in lobes chronically deprived of ORN innervation. This is consistent with other reports of a basal level of receptor tyrosine kinase activation ” in the absence of ligands and appears necessary, in the present case, to block apoptosis. Subsequent arrival of ORN axons could then trigger additional glial responses via upregulation of FGFR activation and subsequent, developmentally relevant, activation of various downstream pathways. We were not able to differentiate levels of FGFR activation at different stages by immunocytochemistry; future work will focus on questions of developmental regulation of FGFR activation and the relative localization of FGFRs to plasma membrane vs nucleus as well as possible shifts in activation of different second-messenger pathways. Blocking glial FGFR activation: effects on neurons In addition to the obvious effect of FGFR inactivation on NP glial cell migration, several, more subtle, effects were noted that suggest that loss of FGFR activation disrupts the effect of glial cells on the growth patterns of axons in the sorting zone and dendrites in the developing glomeruli. First, blockade of glial cell FGFR activation led to altered growth patterns of ORN axons as they navigated the sorting zone. Normally, ORN axons arrive at the sorting zone as a mixed population of MFas II-positive and MFas IInegative axons. On entering the sorting zone ORN axons reorient and refasciculate into MFas II-positive and MFas IInegative bundles. In PD173074-treated animals, the SZ glia had migrated Glial FGFRs in Glia-Neuron Signaling outward to form a sorting zone of normal length and glial density during the early stages of axon ingrowth, yet ORN axons did not exhibit fasciculation changes as they traversed the sorting zone. The unusual axonal phenotype is not due to reduced