Endogenous grn regulates unc-5. (A-D) unc-5 mRNA expression was examined in aCC and RP2 motoneurons of Stage 15 grn mutant embryos (B,D) and their heterozygous siblings (A,C). Anterior is up and dorsal is right in all panels. The top panels (A,B) show unc-5 in situ signal in magenta and myc antibody in green to label tau-myc expressed in aCC and RP2 by the RN2-Gal4 driver. The bottom panels (C,D) show RNA signals in magenta. Anterior is up in all panels. White marks indicate the positions of the xz and yz sections displayed below and to the right of the main xy panels, respectively. Two separate xz sections are shown, one for aCC (a) and one for RP2 (R). (A) In grn heterozygous embryo, clear expression of unc-5 mRNA in both aCC and RP2 is observed, and the cells are easily identified even in the absence of the anti-myc labelling (C). (B) grn mutants show less than half the level of fluorescence in both aCC and RP2 neurons. The reduced unc-5 signal is obvious in the bottom panel in the absence of anti-myc labelling (D). (E,F) Quantification of unc-5 expression in aCC and RP2 neurons (E) or glial cells (F) of st-15 grn heterozygous and grn mutant embryos. Genotypes for each cell analysed are indicated on the x axis, and fluorescence intensity is indicated on the y axis. unc-5 mRNA expression in glia cells is not affected in grn mutants, and is within the same range in both grn/+ and grn mutant embryos (69±21 s.e.m. and 90±15 s.e.m. fluorescence units, respectively) (F); however, it is drastically reduced in aCC from 23.6±4.8 to 10.7±1.5 (P<0.02) and RP2 from 30.6±3.8 to 14.3±1.9 (P<0.005) compared with the heterozygous siblings (arbitrary fluorescence units ±s.e.m.) (E) ^**P<0.05, ^***P<0.005. Neuronal bodies for aCC (a) and RP2 (R) are indicated.

unc-5 partially rescues grn mutants. (A-C) Motor axon projections of flat mounted late st-16 to st-17 embryos visualized with anti-Fas2. Anterior is left and dorsal is up in all panels. Partial genotypes are indicated below each panel. Wild-type embryos (A,B) or animals expressing HA-unc-5 in aCC and RP2 motoneurons through RN2Gal4 (C) present a normal ISN branch. A magnification of the region in the rectangle is shown in C′ (anti-Fas2), C″ (anti-HA) and C″′ (merged). The most dorsal muscle 1 (top arrowhead in A, top arrows in B,C) is innervated by the aCC motor neuron. Muscle 2 is innervated by RP2 motor neuron (2nd arrowhead in A, 2nd arrows in B,C) and a 3rd branch from the ISN innervates muscle 3 (bottom arrowhead in A, bottom arrows in B,C). (D-F) In grn mutants (D,E), 85% of ISNs show defective muscle 1/9 innervations (arrowheads with asterisks in E), but axonal projections to muscles 2/10 are normal. Expression of HA-unc-5 in grn mutants partially restores projections beyond muscle 2 (F, 2nd arrow), and decreases the percentage of failed ISN projections from 85% to 72%. A magnification of the region in the rectangle is shown in F′ (anti-Fas2), F″ (anti-HA) and F″′ (merged). Red asterisks indicate defective projections beyond muscle 2. (G) Quantification of ISN branches projecting beyond muscle 2 in different genetic backgrounds, RN2Gal4; grn/grn; wild-type hemisegments=16.4%±1.8 s.e.m., n=250 and RN2Gal4, UAS-HA-unc-5; grn/grn wild-type hemisegments=26.6%±2.9 s.e.m., n=240. ^***P<0.006.

unc-5 levels are further reduced in eve and grn double mutants. (A-F) unc-5 mRNA expression was examined in aCC and RP2 motoneurons of Stage 15 eve (E,F) or double eve, grn (C,F) mutant embryos and in eve heterozygous embryos (A,D). Anterior is up, dorsal is right and in situ signal shows unc-5 mRNA in magenta in all panels. The top panels (A-C) are also labelled with a myc antibody in green to reveal tau-myc expressed in aCC and RP2 by the RN2-Gal4 driver. The bottom panels (D-F) show RNA signals in magenta. In eve heterozygous embryos, clear expression of unc-5 mRNA in both aCC and RP2 is observed, and the cells are easily identified even in the absence of the anti-myc labelling (A,D). eve mutants show very low levels of fluorescence in most aCC and RP2 neurons although some of them (asterisk, 11%) still express significant amount of unc-5 mRNA (B,E). In double eve, grn mutants there is reduction of the number of cells in which unc-5 is still present (2%) (C,F). Number of cells scored and percentage of cells expressing unc-5 is also indicated on each panel. neuronal cell bodies are outlined in D-F.

An enhancer element for dMNs is regulated by Eve. (A) Mapping the location of the unc-5 neuronal transcriptional enhancer. A schematic of the unc-5 locus is shown at the top and the DNA fragments used in enhancer-reporter transgene expression analysis are depicted below. All constructs were site-directed to attP2 on 3L. A 1.6 kB fragment that spans the 5th intron was identified as a neuronal-specific enhancer element being the only one driving expression in aCC and RP2 (highlighted in green). There is a conserved GATA binding site present within the neuronal element in Drosophila species with a divergence time of >30 million years (from D. melanogaster, DMEL to D. willistoni, DWIL) that might be bound by Grn directly. The aligned sequence as well as the approximate position is highlighted under the neuronal element. (B-D) Unc-5 neuronal enhancer is expressed in Eve-positive aCC and RP2 neurons. Flat mounted st-17 embryos expressing UAS-NLS-GFP under the control of Unc-5-Gal4 were visualized with anti-Eve (B) and anti-GFP (C). Colocalization of Eve and the neuronal enhancer is observed in aCC and RP2 neurons (D). (E-H) unc-5 neuronal enhancer expression was examined in aCC and RP2 neurons of stage-13 eve mutant embryos (G,H) and their heterozygous siblings (E,F). The cell bodies and axonal projections of aCC and RP2 neurons are visualized with anti-22C10 and the unc-5 neuronal enhancer expression was visualized with anti-GFP. Enhancer expression is absent in aCC and RP2 cells in eve mutant embryos. Anterior is up in all panels. a, aCC; p, pCC; R, RP2; SN, segmental nerve.

Individual or combinatorial misexpression of eve and grn in dMP2 cells induce unc-5 expression. (A-D) unc-5 mRNA expression was examined in dMP2 neurons of stage-15 embryos. unc-5 signal is in magenta, and myc antibody to localize UAS-tau-myc under the control of dMP2Gal4 is in green. Two identical xz and yz sections are displayed below or at the right of each panel, top and left sections are the merged image of both channels and bottom or right section is the in situ signal alone. There is no evidence of unc-5 expression in wild-type dMP2 neurons (A). However, ectopic expression of UAS-grn (B), UAS-eve, UAS-grn (C), or UAS-eve (D) in dMP2 neurons, activates expression of the unc-5 gene in these cells. The percentage and number of cells expressing unc-5 mRNA is shown in each panel.

eve and grn do not cross-regulate each other in dMP2 neurons. Anti-myc antibody was used to visualize dMP2 neurons expressing UAS-tau-myc through dMP2Gal4 in green. (A) Eve protein expression was examined in St-15 embryo misexpressing UAS-grn in dMP2 cells. Anti-eve antibody in magenta and anti-myc in green. (B) The same image as in A without myc staining. Ectopic grn expression in dMP2 cells does not induce eve expression (encircled in white). (C) grn mRNA expression was also analysed in St-15 embryo misexpressing UAS-eve in dMP2 cells. grn in situ signal is in magenta and anti-myc antibody is in green. (D) The same image as in C but without myc staining. As in wild-type dMP2 neurons, there is no evidence of grn expression in dMP2 cells misexpressing eve (encircled in white), suggesting that in these cells eve does not regulate grn expression.

Lateral axon exit of dMP2 neurons promoted by eve and grn is partially dependent on unc-5. (A-E) Axonal projections of dMP2 neurons on flat mounted st-15 embryos visualized with anti-GFP. Partial genotypes are indicated below each panel. Axons in dMP2 neurons expressing EGFP through dMP2Gal4 fasciculate and project longitudinally through the ventral nerve cord (A). In dMP2 neurons misexpressing grn, axons do not significantly project into the lateral muscle field (B). Combinatorial misexpression of uas-grn with uas-eve, results in lateral exit of 34% of dMP2 axons (C). Ectopic expression of HA-unc-5 in dMP2 neurons redirects 43% of dMP2 axons laterally (D). In an unc-5 mutant embryo percentage of dMP2 lateral axonal exit triggered by UAS-grn, UAS-eve significantly decreases from 34%±5.1 s.e.m. to 21%±2.4 s.e.m. in an unc-5/+ background and to 6%±2.1 s.e.m. exit in an unc-5 null background (E). (F) Quantification of lateral projection of dMP2 axons in UAS-grn, UAS-eve overexpressing flies in a wild-type background (e.g. GOF), unc-5/+ (e.g. GOF, XTE-18/+) or unc-5/unc-5 (e.g. GOF, XTE-18/+) genetic backgrounds (XTE-18 is a deficiency that removes unc-5) (Labrador et al., 2005). ^**P<0.02, ^***P<0.0001.