Fig. 1: MooSEZ is a developmentally-conserved neuron eliciting backward locomotion at both larval and adult stages.

A Examples of single larva crawling trajectories for GH146^II-, NP225-, and NP5288-GAL4 driving UAS-CsChrimson (top) and their respective parental controls (bottom) before, during, and after optogenetic stimulation. 9-s red light pulse is designated by yellow, blue denotes backward crawling, and brown forward motion. The (0, 0) coordinate represents larvae’s trajectory onset at t = 0. A 9-s optogenetic activation of larvae expressing CsChrimson using GH146^II-, NP225- or NP5288-GAL4 driver lines elicits motor response of persistent backward retreat which is analogous to the motor pattern observed for adult flies of the same genotype. In contrast, light stimulation of the respective parental control larvae does not lead to backward crawling. B Top, mean translational velocity of larval crawling during a 9-s red light pulse for GH146^II-, NP225-, and NP5288-GAL4 driving UAS-CsChrimson. Pronounced backward locomotion is observed during the 9-s light pulse for PN driver lines expressing CsChrismon, but not for their respective parental controls (n for GH146^II>CsChrimson, GH146^II, NP225>CsChrimson, NP225, NP5288>CsChrimson, NP5288, CsChrimson: 12, 11, 13, 10, 13, 11, 11 respectively, ***p < 0.001, ****p < 0.0001 for pairwise comparisons with respective parental controls, Kruskal-Wallis test followed by Dunn’s post-hoc tests). Bottom, number of peristaltic backward waves during a 9-s red light pulse of larvae expressing UAS-CsChrimson under the control of GH146^II-, NP225-, and NP5288-GAL4. Number of peristaltic waves of muscle contractions traveling from anterior to posterior abdominal segments of the larva is significantly higher in GH146^II-, NP225-, and NP5288-GAL4 driving CsChrimson compared with their respective parental controls (n for GH146^II>CsChrimson, GH146^II, NP225>CsChrimson, NP225, NP5288>CsChrimson, NP5288, CsChrimson: 12, 11, 13, 10, 13, 11, 11 respectively, ***p < 0.001, ****p < 0.0001 for pairwise comparisons with respective parental controls, Kruskal-Wallis test followed by Dunn’s post-hoc tests). C Top, Expression pattern in larvae of GH146^II-, NP225-, and NP5288-GAL4. Maximum intensity projections of 95 confocal sections (1 µm) through the central brain and VNC are presented. Bottom, close-up of larval SEZ neurons labeled by GH146^II-, NP225-, and NP5288-GAL4. Maximum intensity projections of 48-69 confocal sections (1 µm) through the central brain and VNC are presented. UAS-CsChrimson.mVenus was used to label the cells. Larval SEZ neurons are designated with red arrowheads. Scale bars apply to all images within the same row. D Top, mean translational velocity of larval crawling during a 9-s red light pulse for GH146^II-GAL4 with subtracted GH146-QF. GH146-QF was used to drive QUAS-GAL80 to limit the expression pattern of GH146^II-GAL4 driving UAS-CsChrimson. Pronounced backward locomotion is observed during the 9-s light pulse for GH146^II-GAL4 intersection larvae, but not for their respective parental control larvae (n for GH146^II-GAL4 > UAS-CsChrimsonՈGH146-QF > QUAS-GAL80, UAS-CsChrimson, GH146^II-GAL4: 8, 11, 11 respectively, *p < 0.05, ****p < 0.0001 for pairwise comparisons with respective parental controls, Kruskal-Wallis test followed by Dunn’s post-hoc tests). Bottom, number of peristaltic backward waves during a 9-s red light pulse of larvae expressing CsChrimson under the control of GH146^II-GAL4 and in the presence of GH146-QF driving QUAS-GAL80. Number of peristaltic waves of muscle contractions traveling from anterior to posterior abdominal segments of the larva is significantly higher in GH146^II-GAL4 intersection larvae than in their respective parental control larvae (n for GH146^II-GAL4 > UAS-CsChrimsonՈGH146-QF > QUAS-GAL80, UAS-CsChrimson, GH146^II-GAL4: 8, 11, 11 respectively, **p < 0.01, ****p < 0.0001 for pairwise comparisons with respective parental controls, Kruskal-Wallis test followed by Dunn’s post-hoc tests). E Top, expression pattern of the intersection between GH146^II-GAL4 driving UAS-CsChrimson.mVenus, and GH146-QF driving the GAL4 inhibitor QUAS-GAL80 in larvae. Maximum intensity projection of 95 confocal sections (1 µm) through the central brain and VNC is presented. Bottom, close-up of larval SEZ neurons labeled in GH146^II-GAL4 intersection larvae. Larval SEZ neurons are designated with red arrowheads. Maximum intensity projection of 55 confocal sections (1 µm) through the central brain and VNC is presented. UAS-CsChrimson.mVenus was used to label the cells. Expression in the SEZ is maintained in GH146^II-GAL4 intersection larvae. F Top left, translational velocity ± SEM (shading) of larvae treated (dark blue) or untreated (light blue) with RU-486 following a 9-s optogenetic stimulation. RU-486 driven recombinase (hPR:FLP) permanently marks larval GH146^II-GAL4 neurons by flipping-out the FRT-flanked stop cassette in act5C-FRT-STOP-FRT-LexA. The 9-s light pulse is labeled by light red. Top right, mean translational velocity during the 9-s light pulse obtained from traces on the left. Optogenetic activation of RU-486 treated larvae induces pronounced larval backward locomotion which is not observed for RU-486 untreated larvae (n for -RU486, +RU486: 11, 14 respectively, ****p < 0.0001, Mann-Whitney test). Bottom left, translational velocity ± SEM (shading) of adult flies developed from larvae presented above either exposed (dark blue) or unexposed (light blue) to RU-486 during larval stages following a 1.5-s optogenetic stimulation. The 1.5-s light pulse is labeled by light red. Bottom right, mean translational velocity during the 1.5-s light pulse obtained from traces on the left. Larvae with immortalized GH146^II expression maintain their backward locomotion phenotype into adulthood. By contrast, larvae unexposed to RU-486 show a consistent lack of backward locomotion through development (n for -RU486, +RU486: 11, 14, respectively, ****p < 0.0001, Mann-Whitney test). G Top left, translational velocity ± SEM (shading) of mosaic larvae generated with the SPARC method expressing CsChrimson::tdTomato in stochastically distributed subsets of neurons within the GH146^II-GAL4 driver line during a 9-s red light stimulation. Larvae were grouped into backward crawlers (dark blue) and non-backward crawlers (light blue). The 9-s light pulse is labeled by light red. Top right, mean translational velocity during the 9-s light pulse obtained from traces on the left (n for NO BW, BW: 7, 13 respectively, ****p < 0.0001, Mann-Whitney test). Bottom left, translational velocity ± SEM (shading) of SPARC mosaic adult flies that were categorized as either backward crawlers (dark blue) or non-backward crawlers (light blue) during larval stages following a 2-s red light pulse. The 2-s light pulse is labeled in light red. Bottom right, mean translational velocity during the 2-s light pulse obtained from traces on the left. Adult flies classified as backward crawlers during larval stages walked backwards upon optogenetic stimulation, whereas adult flies classified as non-backward crawlers during larval stages did not (n for NO BW, BW: 7, 13 respectively, ***p < 0.001, Mann-Whitney test). For details of statistical analysis, see Supplementary Data 1.