Neuron-glia interactions are necessary for the formation of the longitudinal axon trajectories in the Drosophila central nervous system. Longitudinal glial (LG) cells are required for axon guidance and fasciculation, and pioneer neurons for trophic support of the glia. Neuregulin is a neuronal molecule that controls glial survival in the vertebrate nervous system. The Drosophila protein Vein has structural similarities with Neuregulin. Vein functions like a Neuregulin to maintain glial cell survival. Direct in vivo evidence is presented at single-cell resolution that Vein is produced by pioneer neurons and maintains the survival of neighboring LG. This mechanism links axon guidance to control of glial cell number and may contribute to plasticity during the establishment of normal axonal trajectories (Hidalgo, 2001).

LG are overproduced in wild-type embryos and Vn is produced by pioneer neurons to promote LG survival. Several lines of evidence support the functional homology of Vn to Neuregulin. First, an increase in LG apoptosis occurs in vn mutant embryos. Second, knocking out Vein function in neurons only by targeted vnRNAi expression induces glial apoptosis. In cocultures of oligodendrocytes and dorsal route ganglion neurons, blocking Neuregulin activity with anti-Neuregulin antibodies promotes oligodendrocyte death. Third, blocking Vn signaling in the LG by targeting the expression of dominant-negative forms of its receptor DER or of Ras (signaling downstream) induces glial apoptosis. Expression of an ErbB4-Fc fusion of the CNS Neuregulin receptor to cocultures of oligodendrocytes and dorsal route ganglion induces oligodendrocyte death. Fourth, targeted neuronal expression of Vn rescues glial apoptosis in vn null mutants. Delivery of Neuregulin rescues Schawnn cell and oligodendrocyte number in mutant explants and in normal and transected nerves. From these data, it is concluded that neuronal Vn functions like Neuregulin to promote glial survival (Hidalgo, 2001).

These findings contrast with the notion that insect nervous system is hard-wired and does not require trophic factors. The functional homology between Vn and Neuregulin implies that the same set of trophic neuron-glia interactions -- and therefore perhaps also a similar degree of plasticity -- operates in the insect and vertebrate CNS (Hidalgo, 2001).

Vn is produced by the MP2 pioneer neurons during axonogenesis and binds DER in some of the LG, where it activates the Ras/MAPkinase pathway to promote survival. Two results provide direct evidence of the mechanism of Vn function: (1) in the absence of vn glial die of apoptosis, which is prevented when LG express p35. Lack of Neuregulin signaling leads to a reduction in glial cell numbers, but it has not been directly shown that changes in glial numbers are due to apoptosis. (2) vn functions nonautonomously from the MP2 pioneer neurons to maintain glial survival. Knocking out vn function from only the MP2 pioneer neurons is sufficient to induce glial apoptosis, and delivering Vn to the MP2 neurons in vn null mutant embryos is sufficient to prevent glial apoptosis (Hidalgo, 2001).

Vn is most likely not the only neuronal factor controlling LG survival. In fact, targeted ablation of neurons causes more severe apoptosis of glia than that seen in vn mutants. LG apoptosis and loss increase in vn-spi double-mutant embryos. Thus, Spi appears to be one factor cooperating with Vn to maintain glial survival. However, Spi is also a ligand for DER, and DER is only expressed in a few LG per hemisegment in a narrow time window. Thus, only a subset of the LG respond to DER (Hidalgo, 2001).

Interestingly, dominant-negative expression of Ras has a stronger effect than dominant-negative expression of DER. This suggests that other signaling pathways function in parallel to control glial survival. Ras functions also downstream of the FGFR, which is also expressed in the LG. Hence, Ras may integrate signaling from both DER and FGFR. Nevertheless, the MAPKinase pathway is activated only in subsets of LG at a given time. Perhaps it is active in different cells at different times, or other signaling pathways are involved in the control of glial survival (Hidalgo, 2001).

The requirement for multiple factors to control glial survival is reminiscent of the situation in vertebrates. Oligodendrocyte survival in culture is only rescued by a combination of multiple survival factors whose receptors are expressed in the oligodendrocyte lineage. Such signaling molecules are expressed in the CNS in limiting amounts and this restriction may be vital to balance how much oligodendrocyte precursors divide and how many of their progeny cells die (Hidalgo, 2001).

Vn is involved in the neuron-glia interactions required during axon guidance. The possibility that vn plays a cell-autonomous role in the MP2 neurons cannot be ruled out. However, MP2 determination is unaffected in vn mutants, and targeted expression of dominant-negative DER to the MP2 pioneer neurons has no consequence on axonal patterns. Vn does not function like a guidance molecule either, as pan-neural expression of vn does not alter axonal patterns (Hidalgo, 2001).

Vn is expressed in the MP2 pioneer neurons, which play a particularly relevant role in the formation of the longitudinal pathways. Since MP2 axons require glia for guidance and fasciculation, by maintaining the survival of neighboring glia, Vn anchors glial cells in proximity to the pioneer axons, thus enabling pathfinding (Hidalgo, 2001).

It is important to appreciate how the control of cell survival affects final cell number. The extent of glial cell apoptosis observed in vn mutants is likely to be an underestimate of the extent of glial cell death actually taking place. In fact, apoptotic cells are cleared very quickly, and it has been estimated that only about 4% of apoptotic cells are seen at one given time. Nevertheless, the final glial phenotype of vn null mutants can be mild. Remarkably, although targeted expression of dominant-negative Ras or dominant-negative DER increase LG apoptosis, it does not necessarily decrease glial number and superficially these embryos can look normal. This could be due to the fact that the driver (htlGAL4) used is not expressed early enough, and, therefore, relatively mild effects occurring late in the lineage are being assessed. Thus, expression of dominant-negative DER or Ras at this stage may be sufficient to induce apoptosis in some cells, but other cells may not longer be sensitive to lack of trophic support. This is supported by the fact that DER is expressed in LG only for a narrow time window, which suggests that glial cell survival is under tight temporal control (Hidalgo, 2001).

It is also possible that glial cell number regulation involves a balance of apoptosis and cell proliferation. Thus, expression of dominant-negative Ras in the LG may lead to glial overproliferation as well as apoptosis, resulting in final normal glial number. Oligodendrocyte precursor proliferation is also controlled by interactions with axons. The link between the nonautonomous control of cell number and axon guidance and fasciculation may provide developmental plasticity to the CNS (Hidalgo, 2001).

Regulation of Drosophila EGF receptor (Egfr) activity plays a central role in propagating the evenly spaced array of ommatidia across the developing Drosophila retina. Egfr activity is essential for establishing the first ommatidial cell fate, the R8 photoreceptor neuron. In turn, R8s appear to signal through Rhomboid (Rho) and Vein to create a patterned array of ‘proneural clusters’ that contain high levels of phosphorylated ERKA and the bHLH protein Atonal. Secretion by the proneural clusters of Argos represses Egfr activity in less mature regions to create a new pattern of R8s. Propagation of this process anteriorly results in a retina with a precise array of maturing ommatidia (Spencer, 1998).

Vein is a Neuregulin ortholog postulated to bind to and activate Egfr. Consistent with this view, removal of a single copy of vein in a DER^Elp mutant background strongly enhances the rough eye phenotype observed with DER^Elp /+ alone. Vein mRNA is present at high levels throughout the anterior of second instar eye discs where Egfr is thought to play a role in cell proliferation. By the third larval instar, however, vein is restricted in the MF to single cells within the R8 equivalence group. Thus, at least one cell of the R8 equivalence group contains two potential activators of Egfr: Vein and Rhomboid. To assess the role of Vein in R8 formation, early clonal patches homozygous for a vein null mutation were created. Few such patches are observed, although commonly observed are ‘twin spots’ (groups of cells containing two copies of the GFP marker and homozygous wild type for Vein, which are formed when mitotic recombination occurs). This suggests that Vein may be required early for cell proliferation or survival, similar to the requirement previously observed for Egfr. Within the small mutant patches that do survive, Boss expression is normal; thus loss of Vein alone, as with loss of Rhomboid, does not prevent R8 formation. These results suggest that neither Rhomboid nor Vein alone is essential for R8 differentiation. This is similar to what has been observed in the embryonic CNS, where neuroblast formation requires Egfr activity, but is only strongly affected if both rhomboid and vein activity are removed together. To determine if Rhomboid and Vein also act in parallel to specify R8 in the retina, rho-;vn- double mutant clonal patches were created by mitotic recombination. Patches were created later in second and third instar larvae to circumvent the requirement for Vein in early cell survival, and many of the resulting clonal patches (and their corresponding ‘twin spots’) contained only 4-8 cells. R8 specification is never observed in the interior of these patches, although R8 cells are able to form along the periphery. In addition, often the pattern of ommatidia surrounding and anterior to the patch is altered. In rare rho-;vn- patches that cross the MF, Atonal expression in the proneural clusters also appears to be reduced; these large clones do not distinguish whether this loss is due to a direct requirement for rho vn function in proneural clusters or is a secondary consequence of a loss of more posterior, differentiated R8s. These experiments suggest that Rhomboid/Vein-mediated Egfr activation has two roles: specification of the R8 fate, and setting the pattern of proneural clusters (Spencer, 1998).