Fly glial cells in the wing peripheral nervous system of Drosophila melanogaster originate from underlying epithelial cells. Two findings indicate that gliogenesis is closely associated with neurogenesis. First, it only occurs in regions that also give rise to sensory organs. Second, in mutants that induce the development of ectopic sensory organs glial cells develop at new positions. These findings prompted a genetic analysis to establish whether glial and sensory organ differentiation depend on the same genes. Loss of function mutations of the achaete-scute complex lead to a significant reduction of sensory bristles and glial cells. Genes within the complex affect gliogenesis with different strength and display some functional redundancy. Thus, neurogenesis and gliogenesis share the same genetic pathway. Despite these similarities, however, the mechanism of action of the achaete-scute complex seems to be different in the two processes. Neural precursors express products of the complex, therefore the role of these genes on neurogenesis is direct. However, markers specific to glial cells do not colocalize with products of the achaete-scute complex, showing that the complex affects gliogenesis indirectly. These observations lead to the hypothesis that glio-genesis is induced by the presence of sensory organ cells, either the precursor or its progeny.

Little is known about the genes inducing the first steps of gliogenesis. In the Drosophila wing, glial cells differentiate from epithelial cells, from regions that also give rise to sensory neurons (Giangrande, 1994). Moreover, mutations that induce ectopic sensory organs, also induce gliogenesis at these ectopic positions, indicating a strong association between gliogenesis and neurogenesis (Giangrande, 1994). Since several genes involved in sensory organ differentiation are known (for reviews, see Ghysen and Dambly-Chaudière, 1993; Jan and Jan, 1994), it was important to ask whether they also act on gliogenesis.

The so called proneural genes are necessary for the definition of the proneural cluster, a group of epithelial cells from which one is singled out to become the sensory organ precursor cell (SOP) (for reviews, see Ghysen et al., 1993; Jan and Jan, 1994). To this class of genes belong those of the achaete-scute complex (ASC): achaete (ac), scute (sc) and asense (ase) (Garcia-Bellido, 1978; for a review, see Campuzano and Modolell, 1992). ac and sc are initially expressed in the proneural cluster and then in the SOP cell (Cubas et al., 1991; Skeath and Carroll, 1991, 1992). Their expression fades after the SOP cell has been singled out, when ase starts being expressed (Brand et al., 1993; Dominguez and Campuzano, 1993; Jarman et al., 1993). These genes code for bHLH type transcription factors that form heterodimers and activate gene expression (Murre et al., 1989a,b; Cabrera and Alonso, 1991). ASC loss of function mutations lead to lack of sensory organs, because the proneural clusters and the SOP cells fail to form (Ghysen and O’Kane, 1989; Romani et al., 1989). Accordingly, ASC ectopic expression due to a heat shock promoter or to gain of function mutations such as Hairy-wing, (Hw), induce ectopic sensory organs, due to the formation of additional SOP cells (Campuzano et al., 1986; Balcells et al., 1988; Rodriguez et al., 1990; Blair et al., 1992). Other genes are needed for sensory organ differentiation. hairy (h) and extramacrochaete (emc) repress sensory organ formation at ectopic positions through inhibitory interactions with the ASC (Moscoso del Prado and Garcia-Bellido, 1984a,b; Skeath and Carroll, 1991; Van Doren et al., 1991, 1992; Cubas and Modolell, 1992).

Using different glial markers in ASC mutant backgrounds I show here that proneural genes as well as h are required for proper gliogenesis in the wing PNS. I also show that the products of the ASC are not expressed in the early steps of gli-ogenesis. These results strongly suggest that the mechanisms of action of ASC are different in gliogenesis and neurogenesis.

Stocks

The wild-type stocks used were Oregon R and Sevelen. The mutant stocks were the following: h1, ase (sc2 w s), Hw49c (In(1) Hw49c w s/FM7), ac sc (In(1) ac3 sc10-1 f36a/FM7), sc (In(1) sc8L-sc4R v f/In (1) dl49), ac (In(1)y3PL-sc8R), Hw1 (y Hw1m2g2/y2Y67g). The sc and ac stocks were gifts from J. Modolell, the Hw1 stock was a gift from C. Dambly-Chaudière. As glial markers, the following enhancer trap lines were used: AE2 and rA87 (gifts from V. Auld and C. Goodman; Auld and Goodman, 1992; Giangrande et al., 1993); 2206 (gift from M. Schubiger; Giangrande et al., 1993; Schubiger et al., 1994). The lines A409.1F3/TM3Sb and B72.1M3 (gifts from W. Gehring), which carry a P element on the third chromosome, were identified as being glial specific in a screen in adult flies (Giangrande, unpublished results). All lines label subsets of glial cells, the only exception being rA87, which most likely transiently labels all or nearly all glial nuclei (Giangrande, unpublished results). Mutant heterozygous females were crossed with males of the enhancer trap line. In the F1 generation, males of the appropriate genotype were selected on the basis of the y+ phenotype. For ase and ac, homozygous females were used in the cross and therefore all males were of the appropriate genotype. For Hw1, a stock homozygous mutant and heterozygous for the glial insert was obtained, in order to analyse homozygous females, which display a stronger phenotype than males. Glial markers were always analysed in heterozygotes, to avoid possible recessive phenotypes due to the insert. Control flies were obtained by crossing the enhancer trap line with a wild-type stock. The line ase-β-gal (Jarman et al., 1993) was a gift from A. Jarman.

Immunohistochemistry

X-gal labelling in adults was performed as described by Giangrande et al. (1993). Antibody labelling was as described by Giangrande (1994). In double labelling experiments with anti-β-gal and a mouse monoclonal antibody as primaries, polyclonal rabbit-anti-β-gal from Cappel was used at 1:4000, detected with Cy3 conjugated goat-anti-rabbit 1:600 (Jackson). The monoclonal antibody 22C10 (a gift from S. Benzer) was used at 1:100 and detected with FITC-conjugated). Rabbit-anti-asense (a gift from A. Jarman) was used at 1:8000; mouse-anti-achaete supernatant was prepared from the H990∑SF1 line (a gift from J. Skeath and S. Carroll) and used at 1:50. These two antibodies were revealed with Cy3-conjugated goat-anti-rabbit 1:300 and FITC goat-anti-mouse 1:300 (both from Jackson).

Confocal images were obtained using a Zeiss axiophot and a Leica DMRE.

ASC loss of function mutations affect gliogenesis

Glial cell organization in ac sc mutants

Adult wings carry two major nerves, L1 and L3, travelling along the L1 and the L3 veins respectively (Murray et al., 1984; Hartenstein and Posakony, 1989). These purely sensory nerves contain axons from the chemo-and mechanosensory bristle neurons located on the anterior margin and from the neurons of the large campaniform sensilla located on the L3 vein and on the margin (Figs 1A, 2A). Wing sensory organs differentiate from precursor cells that divide two or more times to give rise to the neuron(s) and to the sensory organ accessory cells (Hartenstein and Posakony, 1989). Glial cells present along the two nerves (Figs 1A, 2B) differentiate from wing epithelial cells, in the same regions in which sensory organs develop, and then migrate for short distances along the underlying nerves (Giangrande et al., 1993; Giangrande, 1994).

Fig. 1.

Glial labelling in wild-type and sc10-1 adults. The 2206 enhancer trap line was used as a glial marker. (A) 2206/+ and (B) sc10-1/Y; 2206/+ wings, labelled with X-gal. (A) Wild-type array of glial nuclei (arrowheads) and sensory organs. cb and mb indicate the chemo-and mechanosensory bristles on the anterior margin, respectively; acv, the anterior cross vein sensillum; L3, the sensilla present on L3; c and a, the costa and the alula; TSM, the position of the twin sensillum on the margin. L1-L3 shows the junction between veins L1 and L3. (B) Mutant wing shows dramatic defects at sensory organs and glial cells. Chemosensory bristles on the anterior margin and on the costa are absent (see asterisks). Other types of bristles are also partially missing and all sensilla on acv and L3 are absent. No glial nuclei can be detected. To facilitate X-gal penetration, wings were cut at several positions. Bar 100μm.

Fig. 1.

Glial labelling in wild-type and sc10-1 adults. The 2206 enhancer trap line was used as a glial marker. (A) 2206/+ and (B) sc10-1/Y; 2206/+ wings, labelled with X-gal. (A) Wild-type array of glial nuclei (arrowheads) and sensory organs. cb and mb indicate the chemo-and mechanosensory bristles on the anterior margin, respectively; acv, the anterior cross vein sensillum; L3, the sensilla present on L3; c and a, the costa and the alula; TSM, the position of the twin sensillum on the margin. L1-L3 shows the junction between veins L1 and L3. (B) Mutant wing shows dramatic defects at sensory organs and glial cells. Chemosensory bristles on the anterior margin and on the costa are absent (see asterisks). Other types of bristles are also partially missing and all sensilla on acv and L3 are absent. No glial nuclei can be detected. To facilitate X-gal penetration, wings were cut at several positions. Bar 100μm.

Three genes of the ASC are necessary for wing sensory organ differentiation: scute (sc), achaete (ac) and ásense (ase), in order of importance relative to their mutant phenotype. The sc10-1 mutation lacks functional scute and achaete products (Campuzano et al., 1985; Villares and Cabrera, 1987). In such wings, all chemosensory bristles and many mechanosensory bristles of the anterior margin as well as all the large campaniform sensilla are absent (Fig. 1B). I have crossed the 2206 glial-specific enhancer trap line (Giangrande et al., 1993; Schubiger et al., 1994) with sc10-1 flies and looked at the glial labelling in adult mutant wings. Since 2206, as well as most glial specific lines, mark subsets of glia, two other lines have also been used in similar crosses: B72.1M3 andA409.1F3 (data not shown). In no case can labelling be detected on the L3 and L1 nerves.

Glial cell development in ac sc mutants

A possible interpretation for the above results is that, in sc101 wings, glial cells differentiate and then degenerate before adulthood. For this reason, glial labelling has been analysed at different stages during pupal development. For the pupal analysis, another glial marker has also been used, rA87 (Giangrande et al., 1993). In rA87, the α-galactosidase (α-gal) labels most if not all glial cells during the early stages of wing development, however labelling progressively decreases and is absent in the adult. In wild-type wings at 24-38 hours after pupariation (AP), the glial cells and the sensory organs have differentiated and the axons have formed the L1 and L3 nerves (Fig. 2A,B). At similar stages, sc10-1/Y wings show some anti-HRP-positive cells on the anterior margin, which, by their position and the morphology, correspond to the mechanoreceptor neurons (Fig. 2C,E). These neurons differentiate much later than normal ones (data not shown) and send out axons that form a very thin, truncated nerve, which in most cases, does not display glial labelling (Fig. 2). In some cases, 5 out of 25 wings, a few glial labelled nuclei have been observed (Fig. 2 and Table 1).

Table 1.

Effects of ASC mutations on the number of labelled glial nuclei

Effects of ASC mutations on the number of labelled glial nuclei
Effects of ASC mutations on the number of labelled glial nuclei
Fig. 2.

Glial labelling in wild-type and sc10-1 wings at 25-38 hours AP. 2206 and rA87 enhancer trap lines were used as glial markers. Anti-HRP detects neurons and nerves (A,C,E); anti-β-gal detects glial nuclei (B,D,F). G is a double exposure of E and F. (A,B) rA87/+ wing: arrows indicate glial nuclei along the L1(L1) and L3 (L3) nerves. Labelled spots outside of L1 and L3 are artefacts as defined by their morphology and their focal plane. (C,D) sc10-1/Y; 2206/+ wing: asterisks indicate the absence of axons on the distal and proximal parts of L1 nerve and the absence of L3 axons. The L1 nerve, which has lost its polarity and goes proximally and distally, does not carry glial nuclei (open arrowheads). (E-G) sc10-1/Y; rA87/+ wing, one of the rare cases in which some glial nuclei are present (see also Table 1). (A,B) Bar 50 gm; (C-G) bar 25μm.

Fig. 2.

Glial labelling in wild-type and sc10-1 wings at 25-38 hours AP. 2206 and rA87 enhancer trap lines were used as glial markers. Anti-HRP detects neurons and nerves (A,C,E); anti-β-gal detects glial nuclei (B,D,F). G is a double exposure of E and F. (A,B) rA87/+ wing: arrows indicate glial nuclei along the L1(L1) and L3 (L3) nerves. Labelled spots outside of L1 and L3 are artefacts as defined by their morphology and their focal plane. (C,D) sc10-1/Y; 2206/+ wing: asterisks indicate the absence of axons on the distal and proximal parts of L1 nerve and the absence of L3 axons. The L1 nerve, which has lost its polarity and goes proximally and distally, does not carry glial nuclei (open arrowheads). (E-G) sc10-1/Y; rA87/+ wing, one of the rare cases in which some glial nuclei are present (see also Table 1). (A,B) Bar 50 gm; (C-G) bar 25μm.

Glial and neuronal organization has also been analysed earlier in development to determine whether glial labelling ever appears in the mutant background. The earliest stage at which glial labelling can be observed on the anterior margin is at 910 hours AP, when the mechanosensory neurons have not yet differentiated whereas some chemosensory neurons are already detectable (Giangrande et. al, 1993). 10 hour AP sc10–1 wings do not show neuronal labelling (data not shown), in agreement with the observation that chemosensory bristles are completely absent in adult wings (see Table 1 and Fig. 1B), nor do they show glial labelling (data not shown). These results indicate that glial labelling is absent throughout pupal development.

The relative roles of sc and ac

Each member of the ASC has a specific role, even though one gene can, at least partially, replace the activity of the others (Rodriguez et al., 1990; Brand et al., 1993; Dominguez and Campuzano, 1993; Hinz et al., 1994). In the wing, sc plays a more important role for sensory organ differentiation than ac. The sc product is necessary and sufficient for the differentiation of most campaniform sensilla (Leyns et al., 1989). On the margin, the effects of sc are stronger than those observed in ac, especially with respect to the chemosensory bristles (Table 1 and data not shown).

To assess the relative roles of the ASC products on gliogen-esis, I have analysed the glial pattern in pupae mutant for only one of the genes. 50% and 25% of the glial labelling is absent in sc and ac mutants, respectively (see Table 1). As expected from the adult phenotype, the number of chemo-and mechanosensory neurons is smaller in sc than in wild-type wings (Table 1 and data not shown) and all neurons from the large campaniform sensilla on L3 and L1 are absent. Neuronal defects are less drastic in ac wings. The number of chemosen-sory organs is less reduced compared to sc and all campaniform sensilla neurons are present (see Table 1 and data not shown). Overall, in both ac and sc, the strength of the effects on glial cells parallels that of the effects on neurons.

ac and sc loss of function mutations behave as recessive mutations: the number of neurons and glial nuclei is not or only slightly affected in heterozygotes (data not shown).

The role of ase

Compared to sc and ac, ase loss of function mutations have weaker effects on the campaniform sensilla and on the mechanoreceptors, but remove as many chemosensory bristles on the anterior margin as sc (Dominguez and Campuzano, 1993). The effects of ase on sensory organ differentiation also differ qualitatively. In sc and ac, the lack of sensory organs is due to the absence of their precursors (Romani et al., 1989; Boulianne et al., 1991; Cubas et al., 1991). On the other side, ase mutant third larval instar discs do carry chemosensory SOP cells even though their number is slightly decreased compared to wild type (Dominguez and Campuzano, 1993). Later on, some SOP cells do not divide or their progeny are abnormally distributed. In the adult, defects in the sensory organ lineage (duplicated or triplicated shafts) have also been observed (Dominguez and Campuzano, 1993). Therefore, in ase, the dramatic reduction of chemosensory bristles along the anterior margin is due in part to a defective singling out of the SOP cell and in part to defects in later steps, either in the SOP divisions or in the differentiation of the SOP progeny.

As for gliogenesis, while in ac and sc the effects on anterior margin glial cells are comparable to those observed on the early appearing sensory organs (chemosensory bristles), in ase there seems to be no such correlation. The number of glial nuclei in ase pupal wings is only reduced by about 15% (Table 1 and data not shown). In addition, while sc10’1 wings do not show glial labelling at early stages of development, ase wings do. 13 hour AP wings show labelling with neuronal markers (Fig. 3). The labelling is weaker and neurogenesis seems less advanced than in the wild type, but there are no gross rearrangements in the sensory organ organization. Although counting is difficult because the labelling is weak, several glial nuclei have been observed associated with the chemosensory cells (Fig. 3) and a few nuclei have also been observed at earlier stages (data not shown).

Fig. 3.

Neuronal and glial labelling in early ase wings. ase/Y; rA87/+ wings at 13 hours AP, distal region. (A) Wing labelled with mAb22C10, to label neurons and (B) wing showing double labelling with 22C10 and anti-β-gal to see the distribution of glial nuclei along the growing nerve. (C,D) A different wing, labelled with anti-β-gal (C) or (D) showing the double labelling. AM indicates the anterior margin. Neuronal somata of the chemosensory neurons are shown by triangles, glial nuclei by arrows. Bar 25μm.

Fig. 3.

Neuronal and glial labelling in early ase wings. ase/Y; rA87/+ wings at 13 hours AP, distal region. (A) Wing labelled with mAb22C10, to label neurons and (B) wing showing double labelling with 22C10 and anti-β-gal to see the distribution of glial nuclei along the growing nerve. (C,D) A different wing, labelled with anti-β-gal (C) or (D) showing the double labelling. AM indicates the anterior margin. Neuronal somata of the chemosensory neurons are shown by triangles, glial nuclei by arrows. Bar 25μm.

ASC gain of function and h mutations induce ectopic glial cells

Hw mutations induce sc and ac expression at positions at which ectopic sensory organs will later differentiate (Campuzano et al., 1986; Balcells et al., 1988; Blair et al., 1992). Ectopic bristles and campaniform sensilla are present on the entire wing blade except on the anterior margin (Palka et al., 1983). In the wing disc, ac ectopic expression appears later than in the normal set of proneural clusters. This is also reflected in the delayed development of the ectopic sensilla (Blair et al.,1992). The ectopic axons join the L1 or L3 nerves or extend in the intervein epithelium without reaching the base of the wing (Palka et al., 1983; Blair et al., 1992). The two mutations, Hw1 and Hw49c (Campuzano et al., 1986; Balcells et al., 1988), display qualitatively similar phenotypes, even though Hw49c is stronger than Hw1 and ectopic ac expression appears earlier and at more numerous positions than in Hw1 (Blair et al., 1992).

Hw flies have been analysed with the same glial-specific markers used to analyse loss of function mutations: a normal pattern of glial labelling is present along the anterior margin nerve (Table 1), in addition, labelled nuclei are present where the ectopic nerves have developed, even though not all ectopic nerves carry glial nuclei (Fig. 4). I have found no ectopic glial cells using markers for late pupal stages, thus, like the sensory organs, ectopic glial cells develop later than normal glial cells (data not shown). Both mutants display ectopic glial cells (Figs 4, 5), although slightly more have been found in Hw49c (data not shown).

Fig. 4.

Glial labelling in wildtype and Hw adults. Nomarsky view of wild-type (A) and mutant (B) wings, heterozygous for the B72.1M3 insertion, labelled with X-gal. (A) Glial nuclei (arrowheads) line the radius (r), L1 (L1) and L3 (L3) veins. (B) Hw1Hw1 wing carrying ectopic sensory organs (asterisks) on the wing blade (see bristles on the L5 (L5) vein). Glial nuclei (arrowheads) line the ectopic L5 nerve (arrows).

Fig. 4.

Glial labelling in wildtype and Hw adults. Nomarsky view of wild-type (A) and mutant (B) wings, heterozygous for the B72.1M3 insertion, labelled with X-gal. (A) Glial nuclei (arrowheads) line the radius (r), L1 (L1) and L3 (L3) veins. (B) Hw1Hw1 wing carrying ectopic sensory organs (asterisks) on the wing blade (see bristles on the L5 (L5) vein). Glial nuclei (arrowheads) line the ectopic L5 nerve (arrows).

Fig. 5.

Glial labelling in Hw pupae. Hw49c/Y; rA87/+ wings at around 25 hours AP labelled with (A,C) anti-HRP, (B,D) anti-β-gal.(E) Double exposure of C and D. (A,B) The distal part of the wing carrying the L1 nerve (L1) and an ectopic nerve on the L2 vein (L2). Ectopic glial nuclei are indicated by arrows. (C,D,E) The posterior and proximal part of a mutant wing. a and PM indicate the alula and the posterior margin, respectively. Neurons and axons are present along the PM, forming a nerve wrapped by glial nuclei (arrows). Bar 25μm.

Fig. 5.

Glial labelling in Hw pupae. Hw49c/Y; rA87/+ wings at around 25 hours AP labelled with (A,C) anti-HRP, (B,D) anti-β-gal.(E) Double exposure of C and D. (A,B) The distal part of the wing carrying the L1 nerve (L1) and an ectopic nerve on the L2 vein (L2). Ectopic glial nuclei are indicated by arrows. (C,D,E) The posterior and proximal part of a mutant wing. a and PM indicate the alula and the posterior margin, respectively. Neurons and axons are present along the PM, forming a nerve wrapped by glial nuclei (arrows). Bar 25μm.

The pair rule segmentation gene h (Nüsslein-Volhard and Wieschaus, 1980; Ingham et al., 1985), which codes for an altered bHLH protein (Rushlow et al., 1989; Skeath and Carroll, 1991; Van Doren et al., 1992; Wainwright and Ish-Horowitz, 1992), also affects sensory organ development in the wing (Palka et al., 1983; Skeath and Carroll, 1991; Blair et al., 1992). h represses ASC expression at ectopic positions and ensures that only the normal set of SOP cells differentiates. In h wings, ac is initially expressed as in the wild type, but is later activated at positions at which ectopic sensory organs arise (Skeath and Carroll, 1991; Blair et al., 1992). h wings carry the normal set of sensory organs and, as in Hw wings, ectopic sensory organs develop later (Palka et al., 1983; Blair et al., 1992). The glial pattern has been analysed in h pupae. Ectopic neurons are mostly found along the longitudinal veins, sometime sending out axons with reversed polarity (Palka et al., 1983 and Fig. 6). Ectopic glial cells are present along the ectopic nerves, even on those with reversed polarity (Fig. 6).

Fig. 6.

Ectopic glial labelling in h wings. 39 hours AP h1/h1 wing labelled with (A) anti-HRP and (B) RK2, a glial-specific antibody (Campbell et al., 1994; Giangrande, 1994; Xiong et al., 1994). C is a double exposure of A and B. L1, L2, L3 indicate the nerves present along the longitudinal veins L1, L2, L3. (A). The L2 nerve navigates distally, meets the L1 nerve and changes direction. The directions taken by the different nerves are indicated by large arrows.(B,C) Glial nuclei (small arrows) are present along the ectopic nerve. Bar 25μm.

Fig. 6.

Ectopic glial labelling in h wings. 39 hours AP h1/h1 wing labelled with (A) anti-HRP and (B) RK2, a glial-specific antibody (Campbell et al., 1994; Giangrande, 1994; Xiong et al., 1994). C is a double exposure of A and B. L1, L2, L3 indicate the nerves present along the longitudinal veins L1, L2, L3. (A). The L2 nerve navigates distally, meets the L1 nerve and changes direction. The directions taken by the different nerves are indicated by large arrows.(B,C) Glial nuclei (small arrows) are present along the ectopic nerve. Bar 25μm.

ASC pattern of expression and glial labelling

The earliest stage at which glial labelling can be observed in the wing disc is during the wandering third larval instar. Initially, one nucleus is present in the presumptive radial region. Later, the number of labelled nuclei in the region increases (Fig. 7E,F). At this stage, a large cluster of achaete-positive cells is located in the presumptive position of the dorsal radius, where many campaniform sensilla will develop (Cubas et al., 1991 and Fig. 7D,E,F,I). The labelled glial nuclei are located near to this cluster but do not express ac. Wandering larval discs have also been labelled with anti-asense (Brand et al., 1993) and, similarly, no colocalization of the asense product and the glial marker are observed (Fig. 7G,H).

Fig. 7.

ASC and glial labellings in pupal and larval wings. (A,B,C) 10 hour AP wing from the ase-lacZ line labelled with (A) mAb22C10 or (B) with anti-β-gal to detect ase expression. C was obtained by superimposing A and B. L1 indicates the anterior margin. The 22C10 antigen is membrane associated whereas the β-gal is localized in the nucleus. All the β-gal is contained within 22C10-positive cells, therefore, ase is not expressed in glial nuclei. (A,B,C) Bar, 10 pm. (D,E,F) rA87/+ wing disc from wandering larva labelled with anti-achaete (D) or anti-β-gal (E) to detect glial nuclei. F is obtained by superimposing D and E. None of the glial nuclei colocalizes with achaete-positive nuclei. The images have been obtained by analysing and superimposing optical sections taken in the region of the dorsal radius (see G,H,I) through 15 pm with a 0.5 pm step. (G,H) Optical sections through a rA87/+ disc labelled with anti-asense (green) and anti-β-gal (red). The step between each image is of 0.5 pm. (G) acv, dr and ANWP indicate, respectively, the precursors of the sensillum on the anterior cross vein, the sensilla on the dorsal radius and the sensilla campaniformia on the anterior notal wing process. A glial nucleus is present in the tegula (t) region, the SOP cell close to this nucleus is not in the same focus plane. (H) dHCV indicates the precursor of the dorsal large campaniform sensillum on the humeral cross vein. (I) Schematic drawing of the fate map of a third instar wing disc. Modified from Campuzano and Modolell (1992). Filled and open dots indicate the positions of the SOP cells. WM, TG, indicate the wing margin and the tegula region, respectively. DR, dHCV as in (G) and (H). The stippled area indicates the wing region shown in G and H.

Fig. 7.

ASC and glial labellings in pupal and larval wings. (A,B,C) 10 hour AP wing from the ase-lacZ line labelled with (A) mAb22C10 or (B) with anti-β-gal to detect ase expression. C was obtained by superimposing A and B. L1 indicates the anterior margin. The 22C10 antigen is membrane associated whereas the β-gal is localized in the nucleus. All the β-gal is contained within 22C10-positive cells, therefore, ase is not expressed in glial nuclei. (A,B,C) Bar, 10 pm. (D,E,F) rA87/+ wing disc from wandering larva labelled with anti-achaete (D) or anti-β-gal (E) to detect glial nuclei. F is obtained by superimposing D and E. None of the glial nuclei colocalizes with achaete-positive nuclei. The images have been obtained by analysing and superimposing optical sections taken in the region of the dorsal radius (see G,H,I) through 15 pm with a 0.5 pm step. (G,H) Optical sections through a rA87/+ disc labelled with anti-asense (green) and anti-β-gal (red). The step between each image is of 0.5 pm. (G) acv, dr and ANWP indicate, respectively, the precursors of the sensillum on the anterior cross vein, the sensilla on the dorsal radius and the sensilla campaniformia on the anterior notal wing process. A glial nucleus is present in the tegula (t) region, the SOP cell close to this nucleus is not in the same focus plane. (H) dHCV indicates the precursor of the dorsal large campaniform sensillum on the humeral cross vein. (I) Schematic drawing of the fate map of a third instar wing disc. Modified from Campuzano and Modolell (1992). Filled and open dots indicate the positions of the SOP cells. WM, TG, indicate the wing margin and the tegula region, respectively. DR, dHCV as in (G) and (H). The stippled area indicates the wing region shown in G and H.

In the anterior margin, glial cells are first labelled by 10 hours AP. However, at this stage little or no ac or ase expression is observed in the wing. Therefore I have used a transformant line carrying the ase promoter fused to the lacZ gene (Jarman et al., 1993). Since the P-gal product is more stable than asense protein, one can use this line to trace the history of the cells that have expressed ase. Indeed, in 10 hour AP wings, anti-β-gal and the neuronal specific antibody 22C10 display perfect colocalization (Fig. 7A-C). In contrast, β-gal is not found in glial cells. This shows that ase is only expressed in the neuronal lineage.

In the present study I show that gliogenesis in the fly PNS is affected by genes that are also required for sensory organ differentiation. Three genes of the ASC play a role in wing gli-ogenesis, ac, sc and ase, the same ones required for wing neurogenesis. The relative requirement of the three products in gli-ogenesis is different, as is the case for sensory organs. The similarity between neurogenesis and gliogenesis is not limited to the effects of the ASC mutations, but extends to genes such as h, normally required to repress ASC expression at ectopic positions. While the genes of the ASC are expressed in the cells that give rise to sensory organs, ASC products and glial markers do not colocalize, even at early stages of glial development.

Loss of function mutants of the proneural genes affect neurogenesis and gliogenesis

Previous results have shown that glial cells of the wing only form in regions that give rise to sensory organs. In engrailed and fused mutants, which have sensory organs at ectopic positions in the wing epithelium, glial cells accompany the ectopic nerves (Giangrande, 1994). In addition, the neurogenic genes, which interact with the proneural genes and are necessary for sensory organ differentiation, also affect glio-genesis in the embryo (Hartenstein et al., 1992) and in the adult (Giangrande, unpublished results). These data suggest a link between neurogenesis and gliogenesis. Therefore it is important to establish whether both processes are controlled by the proneural genes. The fact that the sc10-1 deficiency, a loss of function of ac and sc, removes most sensory organs and glial cells indicates that neurogenesis and gliogenesis do indeed share a common genetic pathway. The possibility that glial cells are present but do not express a given antigen has been eliminated by using different types of markers. Since in sc10-1 glial labelling is absent throughout development, it is very unlikely that glial cells differentiate and subsequently degenerate. Rather, it strongly suggests that they never form.

If the effects of either ac or sc mutations alone are analysed, it becomes evident that sc10-1, which lacks both products, displays more than an additive effect on glial cells (see Table 1). This means that, in the single mutants, the remaining gene can in part substitute functionally for the missing one, as has already been observed for the sensory organs (Rodriguez et al., 1990; Brand et al., 1993; Dominguez and Campuzano, 1993; Hinz et al., 1994).

ASC expression at ectopic positions induces gliogenesis

In Hw, ASC misexpression induces the formation of ectopic nerves that are lined by glial nuclei. Since glial cells migrate along nerves and in many instances ectopic bundles are not connected with the normal ones, it is unlikely that glial cells originating at normal positions migrate to the ectopic nerves. Moreover, the number of glial cells on the normal nerves is not reduced, as one might expect if glial cells had migrated out of the normal positions.

The expression of the ASC only at the correct positions depends on h (Rushlow et al., 1989; Skeath and Carroll, 1991; Van Doren et al., 1992; Wainwright and Ish-Horowitz, 1992). The present study shows that glial cell development, like neuronal development, is repressed at ectopic positions due to the activity of the h gene. Thus, mutations acting in cis or in trans to misactivate the ASC always result in ectopic gliogen-esis.

Does sensory organ differentiation induce gliogenesis?

ASC mutations have a direct effect on neurogenesis. By analogy, the simplest hypothesis is that glial cells arise from a cell that expresses the ASC, either the SOP cell, or one of the other achaete-positive cells in the proneural cluster. The present study however shows that none of the genes in the ASC are expressed in the glial lineage, unless they are present at such low levels that they are not detectable with the antibodies used. It is also formally possible that these genes are expressed in the glial lineage and switched off before rA87 labelling becomes evident. However, rA87 is a very early marker, which labels glial cells at the same time as or slightly before RK2, a glial-specific antibody that recognises a homeobox-containing protein (Campbell et al., 1994; Xiong et al., 1994; Giangrande, unpublished results). Studies on the embryonic longitudinal glial cells have shown that RK2 recognises glial precursors (Campbell et al., 1994; Xiong et al., 1994) which suggests that the initial rA87 labelling is found in glial precursors. These observations make it unlikely that the genes of the ASC are expressed during gliogenesis.

In light of these results, I propose that the mechanisms of action of the ASC are different in neurogenesis and gliogenesis. The effects of ASC mutations on glial cells seem to be indirect, due to the defects in sensory organ differentiation, which means that glial cells require normal sensory organ development to differentiate. Several findings support the hypothesis of an inductive process in glial differentiation. First, the effects of Hw on glial cells parallel perfectly those previously observed for sensory organ: both cell types develop at ectopic positions, later than the normal populations. The effects of Hw49c on glial and neuronal cells are stronger and earlier than in Hw1. Although Hw and h both result in ectopic sensory organs, regional differences exist between them. These same differences are observed for the ectopic glial cells. Second, the effects of ac and sc loss of function on glial cells also parallel those found for sensory organs. In sc10-1 wings, most sensory organs do not differentiate and the few that are still there differentiate abnormally, since they appear much later than in wild type and send axons that never get to the central nervous system. These major defects in precursor singling out and in sensory organ differentiation result in the absence of glial cells in most mutant wings. Finally, previous data have demonstrated that ac, sc and ase mutations display different strengths with respect to their neuronal phenotype. The present results show that this is also true for gliogenesis and that, for each mutation, the glial phenotype parallels perfectly the neuronal phenotype. For example, sc has a stronger effect than ac on chemosensory organ and on glial cell development. In the case of ase, in which chemosensory neurons and glial cells are differently affected, it is likely that the initial steps of sensory organ differentiation, rather normal in this mutant, are sufficient to promote gliogen-esis. Alternatively, since sc and ac, which affect both mechano-and chemosensory organs, affect glial cells more than ase, which affects preferentially chemosensory organs, it is possible that mechano-but not chemosensory organs are mainly required for glial cell differentiation. Further studies will be necessary to establish at what point downstream of ASC glial cell induction happens and whether this process depends on postmitotic cells such as neurons or on premitotic cells, either the SOP or its daughter cells. They will also be important to determine whether glial induction depends on short range cellcell interactions and whether glial precursors are always induced at reproducible positions.

Thanks to V. Auld and C. Goodman, C. Dambly-Chaudière, W. Gehring, A. Jarman, J. Modolell, M. Schubiger for sending me fly stocks. Thanks also to S. Benzer, J. Skeath and S. Carroll, A. Tomlinson for providing me with antibodies or hybridoma cells. I am grateful to J. Palka, in whose laboratory this work was started, and to P. Heitzler, P. Sassone-Corsi, M. Schubiger and P. Simpson who gave me helpful advice. Thanks to P. Lawrence, J. Modolell, M. Schubiger and P. Simpson for many thoughtful comments on the manuscript. I thank C. Carteret and C. Ackerman for excellent technical assistance and Color System, B. Boulay and S. Metz for help with the figures. J. L. Vonesch helped me with the confocal images. This work was supported by CNRS and INSERM.

Auld
,
V. J.
and
Goodman
,
C. S.
(
1992
).
Molecular characterization of embryonic glia in Drosophila
.
Society for Neuroscience
,
22nd Annual Meeting
,
Anaheim, USA
.
Balcells
,
L.
,
Modolell
,
J.
and
Ruiz-Gomez
,
M.
(
1988
).
A unitary basis for different Hairy-wing mutations in Drosophila melanogaster
.
EMBO J
.
7
,
3899
3906
.
Blair
,
S. S.
,
Giangrande
,
A.
,
Skeath
,
J. B.
and
Palka
,
J.
(
1992
).
The development of normal and ectopic sensilla in the wing of hairy and Hairy wing mutants in Drosophila
.
MOD
38
,
3
16
.
Boulianne
,
G. L.
,
de la Concha
,
A.
,
Campos-Ortega
,
J. A.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1991
).
The Drosophila neurogenic gene neuralized encodes a novel protein and is expressed in precursors of larval and adult neurons
.
EMBO J
10
,
2975
2983
.
Brand
,
M.
,
Jarman
,
A. P.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1993
).
asense is a Drosophila neural precursor gene and is capable of initiating sense organ formation
.
Development
119
,
1
17
.
Cabrera
,
C. V.
and
Alonso
,
M.
(
1991
).
Transcriptional activation by heterodimers of the achaete-scute and daughterless gene products of Drosophila
.
EMBO J
.
10
,
2965
2973
.
Campbell
,
G.
,
Goring
,
H.
,
Lin
,
T.
,
Spana
,
E.
,
Anderson
,
S.
,
Doe
,
C.Q.
and
Tomlinson
,
A.
(
1994
).
RK2, a glial-specific homeodomain protein required for embryonic nerve cord condensation and viability in Drosophila
.
Development
120
,
2957
2966
.
Campuzano
,
S.
,
Balcells
,
L.
,
Villares
,
R.
,
Carramolino
,
L.
,
Garcia-Alonso
,
L
and
Modolell
,
J.
(
1986
).
Excess function Hairy-wing mutations caused by gypsy and copia insertions within structural genes of the achaete-scute locus of Drosophila
.
Cell
44
,
303
312
.
Campuzano
,
S.
,
Carramolino
,
L.
,
Cabrera
,
C. V.
,
Ruiz-Gomez
,
M.
,
Villares
,
R.
,
Boronat
,
A.
and
Modolell
,
J.
(
1985
).
Molecular genetics ofthe achaete-scute gene complex of D. melanogaster
.
Cell
40
,
327
338
.
Campuzano
,
S.
and
Modolell
,
J.
(
1992
).
Patterning of the Drosophila nervous system: the achaete-scute gene complex
.
Trends genet
.
8
,
202
208
.
Cubas
,
P.
,
de Celis
,
J.-F.
,
Campuzano
,
S.
and
Modolell
,
J.
(
1991
).
Proneural clusters of achaete-scute expression and the generation of sensory organs in the Drosophila imaginal wing disc
.
Genes Dev
.
5
,
996
1008
.
Cubas
,
P.
and
Modolell
,
J.
(
1992
).
The extramacrochaetae gene provides informations for sensory organ patterning
.
EMBO J
.
11
,
3385
3393
.
Dominguez
,
M.
and
Campuzano
,
S.
(
1993
).
asense, a member of the Drosophila achaete-scute complex, is a proneural and neural differentiation gene
.
The EMBO J
.
12
,
2049
2060
.
Garcia-Bellido
,
A.
(
1978
).
Genetic analysis of the achaete-scute system of Drosophila melanogaster
.
Genetics
91
,
491
520
.
Ghysen
,
A.
and
Dambly-Chaudière
,
C.
(
1993
).
The specification of sensory neuron identity in Drosophila
.
BioEssays
15
,
293
298
.
Ghysen
,
A.
and
O’Kane
,
C. J.
(
1989
).
Neural enhancer-like elements as specific cell markers in Drosophila
.
Development
105
,
35
52
.
Giangrande
,
A.
,
Murray
,
M. A.
and
Palka
,
P.
(
1993
).
Development and organization of glial cells in the peripheral nervous system of Drosophila melanogaster
.
Development
117
,
895
904
.
Giangrande
,
A.
(
1994
).
Glia in the fly wing are clonally related to epithelial cells and use the nerve as a pathway for migration
.
Development
120
,
523
534
.
Hartenstein
,
A. Y.
,
Rugendorff
,
A.
,
Tepass
,
U.
and
Hartenstein
,
V.
(
1992
).
The function of the neurogenic genes during epithelial development in the Drosophila embryo
.
Development
116
,
1203
1220
.
Hartenstein
,
V.
and
Posakony
,
J. W.
(
1989
).
Development of adult sensilla on the wing and notum of Drosophila melanogaster
.
Development
107
,
389
405
.
Hinz
,
U.
,
Giebel
,
B.
, and
Campos-Ortega
,
J. A.
(
1994
).
The basic-helix-loop-helix domain of Drosophila lethal of scute protein is sufficient for proneural function and activates neurogenic genes
.
Cell
76
,
77
87
.
Ingham
,
P.
,
Pinchin
,
S. M.
,
Howard
,
K. R.
, and
Ish-Horowitz
,
D.
(
1985
).
Genetic analysis of the hairy gene in Drosophila
.
Genetics
111
,
463
486
.
Jan
,
Y. N.
and
Jan
L. Y.
(
1994
).
Neuronal cell fate specification in Drosophila
.
Curr. Opin.Neurobiol
.
4
,
8
13
.
Jarman
,
A. P.
,
Brand
,
M.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1993
).
The regulation and function of the helix-loop-helix gene, asense, in Drosophila neural precursors
.
Development
119
,
19
29
.
Leyns
,
L.
,
Dambly-Chaudière
,
C.
and
Ghysen
,
A.
(
1989
).
Two different sets of cis elements regulate scute to establish two different sensory patterns
.
Roux’s Arch. Dev. Biol
.
198
,
227
232
.
Moscoso del Prado
,
J.
and
Garcia-Bellido
,
A.
(
1984a
).
Genetic regulation of the achaete-scute complex of Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
193
,
242
245
.
Moscoso del Prado
,
J.
and
Garcia-Bellido
,
A.
(
1984b
).
Cell interactions in the generation of the chetae pattern in Drosophila
.
Roux’s Arch. Dev. Biol
.
193
,
246
251
.
Murray
,
M. A.
,
Schubiger
,
M.
and
Palka
,
J.
(
1984
).
Neuron differentiation and axon growth in the developing wing of Drosophila melanogaster
.
Dev. Biol
.
104
,
259
273
.
Murre
,
C.
,
Schonleber Mc Caw
,
P.
, and
Baltimore
,
D.
(
1989a
).
A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins
.
Cell
56
,
777
783
.
Murre
,
C.
,
Schonleber Mc Caw
,
P.
Vaessin
,
H.
,
Caudy
,
M.
,
Jan
,
L. Y.
,
Jan
,
Y. N.
,
Cabrera
,
C. V.
,
Buskin
,
J. N.
,
Hauschka
,
S. D.
,
Lassar
,
A. B.
,
Weintraub
,
H.
and
Baltimore
,
D.
(
1989b
).
Interactions between heterologous Helix-Loop-Helix proteins generate complexes that bind specifically to a common DNA sequence
.
Cell
58
,
537
544
.
Nusslein-Volhard
,
C.
and
Wieschaus
,
E.
(
1980
).
Mutations affecting segment number and polarity in Drosophila
.
Nature
287
,
795
801
.
Palka
,
J.
,
Schubiger
,
M.
and
Ellison
,
R. L.
(
1983
).
The polarity of axon growth in the wings of Drosophila melanogaster
.
Dev. Biol
.
98
,
481
492
.
Palka
,
J.
,
Schubiger
,
M.
and
Schwaninger
,
H.
(
1990
).
Neurogenic and antineurogenic effects from modifications at the Notch locus
.
Development
109
,
167
175
.
Rodriguez
,
I.
,
Hernandez
,
R.
,
Modolell
,
J.
, and
Ruiz-Gomez
,
M.
(
1990
).
Competence to develop sensory organs is temporally and spatially regulated in Drosophila epidermis primordia
.
EMBO J
.
9
,
3583
3592
.
Romani
,
S.
,
Campuzano
,
S.
,
Macagno
,
E. R.
and
Modolell
,
J.
(
1989
).
Expression of achaete and scute genes in Drosophila imaginal discs and their function in sensory organ development
.
Genes Dev
.
3
,
997
1007
.
Rushlow
,
C. A.
,
Hogan
,
A.
,
Pinchin
,
S. M.
,
Howe
,
K. M.
,
Lardelli
,
M.
, and
Ish-Horowitz
,
D.
(
1989
).
The Drosophila hairy protein acts both in segmentation and bristle patterning and shows homology to N-myc
.
EMBO J
8
,
3095
3103
.
Schubiger
,
M.
,
Feng
,
Y.
,
Fambrough
,
D. M.
, and
Palka
,
J.
(
1994
).
A mutation of the Drosophila sodium pump a subunit gene results in bangsensitive paralysis
.
Neuron
12
,
373
381
.
Skeath
,
J. B.
and
Carroll
,
S. B.
(
1991
).
Regulation of achaete-scute gene expression and sensory organ pattern formation in the Drosophila wing
.
Genes Dev
.
5
,
984
995
.
Skeath
,
J. B.
and
Carroll
,
S. B.
(
1992
).
Regulation of proneural gene expression and cell fate during neuroblast segregation in the Drosophila embryo
.
Development
114
,
939
946
.
Van Doren
,
M.
,
Ellis
,
H. M.
, and
Posakony
,
J. W.
(
1991
).
The Drosophila extramacrochaetae protein antagonizes sequence-specific DNA binding by daughterless/achaetae-scute protein complexes
.
Development
113
,
245
255
.
Van Doren
,
M.
,
Powell
,
P. A.
,
Pasternak
,
D.
,
Singson
,
A.
and
Posakony
,
J. W.
(
1992
).
Spatial regulation of proneural gene activity: auto- and crossactivation of achaetae is antagonized by extramacrochaetae
.
Genes Dev
.
6
,
2592
2605
.
Villares
,
R.
and
Cabrera
,
C. V.
(
1987
).
The achaete-scute gene complex of D
.
melanogaster: conserved domains in a subset of genes required for neurogenesis and their homology to myc. Cell
50
,
415
424
.
Wainwright
,
S. M.
and
Ish-Horowitz
,
D.
(
1992
).
Point mutations in the Drosophila hairy gene demonstrate in vivo requirements for basic, helixloop-helix, and WRPW domains
.
Mol Cell. Biol
12
,
2475
2483
.
Xiong
,
W.-C.
,
Okano
,
H.
,
Patel
,
N. H.
,
Blendy
,
J. A.
,
Montell
,
C.
(
1994
).
repo codes a glial-specific homeo domain protein required in the Drosophila nervous system
.
Genes Dev
.
8
,
981
994
.