Fish maintenance and breeding. Zebrafish (Danio rerio) larvae used for the behavioral analysis were of the TLF (Tubingen long fin) strain. For calcium imaging, siblings from crossings of male and female adults (TLF or AB) were divided into control and treatment groups. Adult fish were maintained in stand-alone, self-circulating systems (Aquatic Ecosystems and TECNIPLAST) in the animal facility at the National Institute on Alcohol Abuse and Alcoholism (NIAAA) following National Institutes of Health (NIH) Animal Care and Use Committee guidelines (Permit number: LMP-FO-11). Embryos were collected in the morning and thereafter maintained at 28°C. Experiments were conducted at 6 days post-fertilization (dpf), unless otherwise indicated. All larvae were used for experiments at stages before their sex was determined.
Behavior recording and kinematic analysis. Video recording and locomotion kinematic analysis was performed as described previously 
. Briefly, images at 512×512 resolution were collected with a Photron high speed camera at 1000 frames/s. Experiments were carried out at 25–28°C with the experimental setup isolated by a black shroud. Larvae were illuminated using a custom built array of infrared (880 nm peak) LEDs (Stackley Devices). Larvae were studied in groups of 20–30 in 6 cm Petri dishes mounted on a mini-shaker (Bruel and Kjaer). Vibration of the mini-shaker was controlled by computer-generated waveforms for startle response experiments 
. Startle responses (C-starts) were identified by changes in body orientation >16°C over a 3 ms window 
. Individual responses were shown as latency histograms. Alternatively the percentage of larvae displaying the startle response was calculated for each Petri dish. Automated tracking software written in the Interactive Data Language (IDL Visual Information Systems) was used for locomotion detection and analysis.
Calcium imaging of reticulospinal neurons. Mauthner, vestibulospinal and MiD3 neurons were labeled by backfilling, following previous studies with modifications 
. Briefly, a fluorescent Ca2+
indicator, Calcium Green dextran (10,000 molecular weight; Invitrogen) was injected into the spinal cord of 5 dpf larvae. After the injection, larvae were allowed to recover in bath solution (NaCl 110 mM, HEPES 5 mM, CaCl2
2 mM, glucose 3 mM, KCl 2 mM and MgCl2
0.5 mM, pH 7.4) for >15 h. Larvae were mounted in an upright position with the dorsal side facing the objective. This was accomplished using low-melting point agarose (2.5%; Nacalai Tesque) in glass-bottom 35 mm plastic dishes. For the hemisected preparation, mounted larvae were cut at the level of the 10th
body segment and the caudal portion removed. Intact sibling larvae were used as controls. The dish was fixed with dental wax to a hole drilled in a plastic plate (165×260×2 mm) to which an audio speaker (SC5.9, Art.No. 8006; VISATON) was attached with screws. A silicon gel sheet (V30Z62MCH100230; Advanced Antivibration Components) supported the four corners of the plate. Imaging was performed on a LSM510 Meta confocal microscope (Carl Zeiss Microscopy) with a 40×1.2 NA water-immersion objective at 256×256 resolution. Mauthner, vestibulospinal, and MiD3 neurons were illuminated with a 488 nm argon laser line and sequential confocal images (at 250 ms intervals) were acquired for ROIs encircling the neurons. Vestibulospinal neurons constitute a compact group and the ROI was chosen so that it encompasses all labeled neurons. For MiD3 neurons, the ROI encompassed MiD3cm and MiD3i neurons 
. The pinhole was fully open which provided an optical slice of 11.8 μm. A TTL signal generated by the Zeiss Zen software was used to trigger the sound/vibration stimulation. A stimulation waveform (150 ms, 500 Hz) generated by a custom made oscillator (Stackley Devices) was routed to an audio amplifier prior to driving a speaker. To ensure that the increase in fluorescence intensity was not caused by a focal plane shift, we chose the focal plane with the highest intensity before each trial. In experiments with multiple cells in the field, we picked a focal plane where the cells of interest were at or near their maximum intensity. All quantification was performed in Zeiss Zen software. Fluorescence intensities of cell bodies were measured and the relative changes in fluorescence from the resting intensity (ΔF/F) were calculated and plotted. The frame during the vibration was excluded from the analysis due to mechanical artifacts, as in other zebrafish studies of calcium imaging using mechanical stimuli 
. To image the cranial vasculature, larvae in the anterior segment preparation or intact larvae were incubated in bath solution containing 0.6 mg/ml Alexa 488 dextran (10,000 molecular weight; Invitrogen) for 30 min before confocal microscopy.
Histology. Larvae were anesthetized with MS-222 and fixed with 4% PFA at 4°C for 12–16 hrs. After washing with PBS containing 20% sucrose at 4°C overnight, the fixed larvae were embedded in a 1
1 mixture of Tissue Tek OCT compound (Sakura Finetech) and 20% sucrose. Samples were then frozen and sectioned (16 μm) on a cryostat. Sections were mounted on Superfrost Plus slides (VWR) and dried. After rinsing in tap water, the Nova Ultra H&E Stain Kit (IHC world) was used for hematoxylin and eosin stain.
Acridine orange (AO) staining. Apoptotic cells in larvae were detected using AO 
. After treatment with 0 or 300 mM ethanol for 1 h, larvae were washed twice in bath solution, and transferred to bath solution containing 5 μg/ml AO. After staining for 1 h at room temperature followed by eight 5-minute washes in bath solution, the larvae were anesthetized with 0.01% MS-222, mounted in 2% low-melting point agarose and examined by confocal microscopy.
Statistical analysis. Statistical analyses were performed using Excel (Microsoft) and Prism 5 (GraphPad Software). After analysis of variance (ANOVA), significant results followed by Bonferroni-corrected t tests between groups are indicated in the text. Graphs show average and standard error of the mean.