Figure 2. Identification of host circuits and genetic components involved in summiting behavior.

(A) Regions and pathways targeted in the candidate screen. AMMC = antennal mechanosensory and motor center; CX = central complex; SOG = subesophageal ganglion; MB = mushroom body; NM & NT = neuromodulator or neurotransmitter; NP = neuropeptide; PI = pars intercerebralis. (B and C) Effects of neuronal disruption (B; 12<N<111, median N=35) or gene knockdown or mutagenesis (C; 10<N<182, median N=46) on summiting. Above: Summiting effect size estimate distributions as estimated by bootstrapping. Experimental groups are ordered by mean effect (negative to positive). Below: gene function and brain region annotations associated with each screened reagent. See Supplementary file 1 for genotype and annotation details. Solid gray line indicates an effect size of zero. Dashed vertical lines separate ranked data into quintiles. (D and E) Frequency of annotations by quintile for (B) and (C), respectively. The number of lines screened (N) is indicated for each annotation. Dashed line indicates the frequency of annotation expected from a null, uniform distribution. Black arrowheads highlight annotations that are overrepresented in the first quintile. For (D), pink overlays indicate the portion of line annotations that are co-annotated for expression in the PI. (F and G) Summiting effect size estimate distributions of disrupting specific circadian genes (F; 19<N<182, median N=62) or circadian and/or PI neurons (G; 11<N<111, median N=46) compared to genotype-matched controls. Lines are ordered by effect size. Pink indicates Gal4 expression in the PI, lime circadian Gal4 lines and genes, and black outlines expression only in DN1ps. Asterisks indicate statistically significant effects on summiting behavior by a two-tailed t-test (*=p<0.05; **=p<0.01; *** p<0.001). R19G10 is highlighted in pink to emphasize its subsequent use as the main PI reagent. See Supplementary file 2 for genotypes and matched controls. (H) Maximum z-projections of brains showing pre- (synaptotagmin; syt-eGFP) and post- (DenMark) synaptic compartments of R19G10 neurons. Bruchpilot (nc-82) staining (blue) visualizes neuropil. Above: brain imaged from anterior. Below: another brain, imaged from the posterior. (I and J) Mean speed of unexposed flies vs time for Clk4.1>TrpA1 and R19G10>TrpA1 genotypes and sibling controls, respectively. Shaded regions are +/− 1 standard error of the mean. Bars along the x-axis indicate the state of visible illumination (above) and temperature (below). (K) Red light onset-triggered mean speed across flies of unexposed R19G10>CsChrimson flies versus time. All trans retinal (ATR) indicates control flies not fed CsChrimson cofactor. Shaded regions are +/− 1 standard error of the mean. Bar along the x-axis indicates lighting conditions (black: darkness, red: red-light illumination).

Figure 2—figure supplement 1. Additional experiments assessing summiting after clock neuron and R19G10 disruption.

(A–C) Plots visualizing summiting for manipulations targeting R19G10 and DN1p neurons. Top: Mean zombie baseline-corrected speed versus time. Shaded area is +/− 1 standard error. Bottom: Individual SM (circles) and median SM (line). *=p<0.05; **=p<0.01; ***p<0.001 by two-tailed t-test. Clk4.1 drives expression in 8–10 and R18H11 drives expression in 7–8 of DN1p neurons per hemisphere, respectively (Kunst et al., 2014; Zhang et al., 2010). (D) Left: Summiting effect size estimate distributions for Gal4/effector combinations targeting clock neurons. Sample size of experimental animals is in black, and controls in gray. Right: Diagrams depicting target cells of tested reagents following Shafer et al., 2022. Violin plot outline colors match diagrams. *=p<0.05; **=p<0.01; ***p<0.001 by two-tailed t-test. (E) Confocal micrographs demonstrating ablation of Pdf-expressing neurons in Pdf-Gal4>hid animals. (F) Summiting effect size estimate distributions when targeting components of the circadian locomotor output pathway identified by Cavanaugh et al., 2014. *=p<0.05; **=p<0.01; ***p<0.001 by two-tailed t-test.

Figure 2—figure supplement 2. Additional experiments assessing the sufficiency of DN1p and R19G10 neuron activation for increased locomotion.

(A–D) Mean speed of unexposed flies with thermogenetically activated DN1p or R19G10 neurons (solid line) versus sibling controls (dashed line) separated by sex. The presence or absence of light is indicated below the horizontal axis with a yellow or black bar, respectively. The measured temperature of the room is shown as a heat map below the light cue indicator. Shaded regions are +/− 1 standard error. For A and B, Clk4.1-Gal4 was used to drive expression in DN1ps. (E and F) Kernel density estimates of the distribution of log-speed for R19G10 >CsChrimson flies with (F) or without (E) of dietary all trans retinal (ATR) in the presence of absence of 5 Hz pulsed red light stimulus (OFF and ON, respectively). A bimodal distribution of speeds is observed regardless of ATR and light treatment. The lower peak (centered around 0) corresponds to non-walking and the higher peak (centered around 6mm/s) corresponds to walking flies. (G) Mean speed across flies of unexposed R19G10>CsChrimson flies before and after the second presentation of red light in the experiment described in Figure 2K. ATR+ flies show a smaller increase in locomotion upon red light stimulus, consistent with CsChrimson depolarization block or adaptation within the stimulated circuit.