• Figure 3.
    Download figureOpen in new tabFigure 3. dbh mutants and bopindolol-treated WT larave exhibit lower arousal threshold.

    Stimulus-response curves generated by a tapping assay for dbh−/ and sibling control larvae (A), and bopindolol and DMSO treated WT larvae (B). Thirty trials were performed at each stimulus intensity, with a 1 min inter-trial interval. Each data point indicates mean ± s.e.m. Stimulus-response curves were constructed using the non-linear variable slope module in Prism and fitted using ordinary least squares. Dashed lines mark the ETP50 values for different genotypes and drug treatments. dbh−/ have an ETP50 value of 2.1 vs 3.1 for sibling controls (32% decrease, p < 0.0001 by extra sum-of-squares F test) (A), and bopindolol treated animals have an ETP50 value of 2.6 vs 5.7 for sibling controls (54% decrease, p < 0.0001 by extra sum-of-squares F test) (B). n indicates number of larvae.

    DOI: http://dx.doi.org/10.7554/eLife.07000.010

    Figure 7.
    Download figureOpen in new tabFigure 7. Optogenetic stimulation of Hcrt neurons activates LC neurons.

    (A) Schematic representation of areas stimulated (Hcrt, yellow) and imaged (LC, green). a = anterior, p = posterior. (B, C) Representative images of a LC cell expressing GCaMP6s in a Tg(hcrt:EGFP) (B) or Tg(hcrt:ChR2-EYFP) (C) larva before (B, C) and immediately after (B′, C′) stimulation of Hcrt neuron region. (D) GCaMP6s fluorescence intensity for a representative LC neuron in a Tg(hcrt:EGFP) (top) and a Tg(hcrt:ChR2-EYFP) (bottom) larva. Single-trial (gray) and average (black) responses are shown. (E) Mean ± s.e.m. ∆F/F (%) values averaged for all trials. n indicates number of trials for 5 Tg(hcrt:EGFP) and 4 Tg(hcrt:ChR2-EYFP) larvae. ***, p < 0.001 by one-way ANOVA. See Videos 1 and 2 for examples of GCaMP6s imaging.

    DOI: http://dx.doi.org/10.7554/eLife.07000.020

    Figure 8.
    Download figureOpen in new tabFigure 8. dbh is required for Hcrt neuron-induced increased locomotor activity.

    (A, C) Representative locomotor activity traces for dbh−/ and sibling control larvae without (A) and with (C) the hcrt:ChR2-EYFP transgene during blue light exposure. (B, D) Average locomotor activity relative to sibling controls. Data is pooled from 2 (B) or 3 (D) experiments. Data is represented as mean ± s.e.m. and is plotted as percentage of dbh+/+ (B) or hcrt:ChR2-EYFP; dbh+/+ (D). n indicates number of larvae. *, p < 0.05 and ***, p < 0.0001 by one-way ANOVA followed by Tukey's test to correct for multiple comparisons.

    DOI: http://dx.doi.org/10.7554/eLife.07000.023

  • Video 1. Pulsed illumination of hcrt:EGFP neurons does not affect GCaMP6s fluorescence in LC neurons.

    Two trials for a representative LC neuron are shown. Frames labeled ‘Stimulation’ indicate periods during which the soma of Hcrt neurons were illuminated by ten 0.3 s pulses of 488 nm light, during which time GCaMP6s was not imaged in the LC. See Figure 7 for quantification.

    DOI: http://dx.doi.org/10.7554/eLife.07000.021

  • Video 2. Pulsed illumination of hcrt:ChR2-EYFP neurons increases GCaMP6s fluorescence in LC neurons.

    Two trials for a representative LC neuron are shown. Frames labeled ‘Stimulation’ indicate periods during which the soma of Hcrt neurons were illuminated by ten 0.3 s pulses of 488 nm light, during which time GCaMP6s was not imaged in the LC. See Figure 7 for quantification.

    DOI: http://dx.doi.org/10.7554/eLife.07000.022

  • Source code 1.

    Scripts used for analysis of behavioral data. sort_fish_sttime_192.pl is a Perl script (Prober et al., 2006) that converts data acquired by the Viewpoint videotracker system to a format that is useful for analysis using Matlab and removes notations that are not relevant to behavioral analysis. perl_batch_192well.m is a Matlab script that allows the sort_fish_sttime_192.pl script to run on the Matlab platform. TapAnalysis.m is a Matlab script that analyzes tapping assay data and generates a table that lists the number of larvae that moved during each tapping event. VT_analysis.m is a Matlab script (modified from Prober et al., 2006) that analyzes locomotion data collected by the Viewpoint videotracker system to quantify several metrics, including activity, waking activity, sleep, sleep architecture and sleep latency. Detailed instructions on the use of these scripts will be provided upon request.

    DOI: http://dx.doi.org/10.7554/eLife.07000.024

    Download source data [source-code-1.media-1.zip]