Fig. 1

TJ⁺ neurons are required for proper larval crawling. a, b 3D reconstruction of whole VNC of first (a) and third (b) instar larvae expressing nuclear GFP under the control TJ-Gal4. Color scale is the z-axis scale; most dorsal cells are red, and most ventral are blue. White scale is the x/y-axis scale. c–g Staining of a second instar larva VNC for TJ (magenta) and TJ-Gal4 expression reported by nuclear H2AGFP (green). Totality of TJ-expressing cells are shown in dorsal (c) to ventral (g) panels. Dashed lines on the right-hand side of the panels indicate segment boundaries and the full line the midline. A unique hemisegment is shown in each panel. Anterior of the VNC is up. Right panels are schematic representations of one hemisegment showing stereotyped ventral–dorsal and medial–lateral cell position of TJ-expressing cells. h, i Number of peristaltic waves per 30 s at (23 °C) and (31 °C). Note that larvae naturally increase the numbers of peristaltic waves at 31 °C compared to 23 °C. Silencing of the entire TJ⁺ population (second beige dot plot, h) causes a slight decrease in the number of peristaltic waves. Activation of the entire TJ⁺ population (second red dot plot) causes a drastic decrease in the number of peristaltic waves (i). This decrease is no longer visible upon activation of the TJ⁺ neurons exclusively in the brain (second salmon pink dot plot, i). For h and i, each single point represents recording of a single first instar larva. Error bars indicate the mean ± SD and n the number of larvae tested. Statistical analysis: Graph i: one-way ANOVA. ***p ≤ 0.001, ns = not significantly different. Graph h: Kruskall–Wallis. **p ≤ 0.01, ns = not significant. j–l Superimposition of three consecutive time frames (0, 2, and 4 s) showing the postures of 5-h-old larvae: control larvae (j), upon silencing of the TJ⁺ neuronal population (k) and upon activation of the TJ⁺ population (l)