Generation and validation of vma21 mutant zebrafish.

(A) Schematic of the zebrafish (Danio rerio) vma21 gene. Mutations in exon 2 of vma21 were created using CRISPR-Cas9 gene editing. (B) Two mutant lines were generated: (1) a 1 base pair (bp) deletion and (2) a 14 bp deletion with a 21 bp insertion. A compound heterozygous line was also generated by crossing the two mutant lines together. (C) Representative western blot of Vma21 and b-actin (housekeeping protein) protein levels in both the vma21 1 bp deletion (vma21^Δ1/Δ1) and the 14 bp deletion with 21 bp insertion (vma21^{Δ14ins21/Δ14ins21}) homozygous mutants compared to their respective wild-type/heterozygous sibling controls at 5 days post fertilization. Vma21 is ~11 kDa in size and b-actin is ~42 kDa. (D) Quantitative analysis of Vma21 western blot normalized to b-actin loading controls shows that Vma21 protein levels are significantly reduced in both vma21 homozygous mutants (fivefold less) compared to their respective sibling controls. Each western blot had a total of three biological replicates. For each sample, n = 30 zebrafish were utilized. All values are shown as mean ± SEM. Significance at **P < 0.01; ***P < 0.001 (unpaired two-tailed t test). Source data are available online for this figure.

vma21 mutant zebrafish display overt morphological differences compared to controls.

Representative light microscopy images of (A) wild-type (WT)/heterozygous (HET) siblings (B) vma21^Δ1/Δ1 (homozygous for a 1 base pair (bp) deletion in vma21), (C) vma21^{Δ14ins21/Δ14ins21} (homozygous for a 14 bp deletion with a 21 bp insertion in vma21), and (D) vma21^{Δ1/Δ14ins21} (a compound heterozygous line that contains both aforementioned mutations in vma21) zebrafish mutants in both the ventral and lateral positions at 5 days post fertilization (dpf). The red arrows on the panel showing the larvae in the ventral position indicate the greater level of melanophores (darker pigmentation) in the siblings in comparison to all the mutant lines. There is a decrease of xanthophores (the yellowish pigment) in the mutant lines (purple arrows) in comparison to all the siblings. The blue arrows show reduced swim bladder size within the mutant lines. Scale bars: 400 mm. (E) Bar graph showing a significant difference in body length between WT and vma21^Δ1/Δ1 mutants but not between WT and HETs or HETs and vma21^Δ1/Δ1 mutants at 5 dpf. n = 9–13/group. Data represented as mean ± SEM. Significance at *P < 0.05 (one-way ANOVA with Tukey’s multiple comparisons test). (F) Stacked bar graph of the percentage of WT vs vma21^Δ1/Δ1 mutants with inflated swim bladders at 5 dpf. n = 10–13/group. Significance at ****P < 0.0001 (Fisher’s exact test). (G) Kaplan–Meier curve showing percent survival of WT and vma21 mutant zebrafish starting from 0 dpf. Sample sizes: siblings n = 194, vma21^Δ1/Δ1n = 74, vma21^{Δ14ins21/Δ14ins21}n = 60, and vma21^{Δ1/Δ14ins21}n = 60. As represented in the graph: WT siblings (dark blue), vma21^Δ1/Δ1 (light blue), vma21^{Δ14ins21/Δ14ins21} (magenta), and vma21^{Δ1/Δ14ins21} (green). Significance at ****P < 0.0001 (Mantel–Cox test). (H–J) Free swim movement tracked over a ten-minute interval and (K–M) swim assay with the photoactive compound optovin to stimulate motor activity in 6 dpf larvae. Graphs from left to right show total time spent moving in seconds (s), total distance traveled in millimeters (mm), and average total velocity (mm/s) for each type of vma21 mutant with their respective WT/HET sibling controls. n = 24/group with three replicates. Data represented as mean ± SEM. Significance at ****P < 0.0001 (unpaired two-tailed t test). As represented in the graphs: siblings (dark blue dots), vma21^Δ1/Δ1 (light blue dots), vma21^{Δ14ins21/Δ14ins21} (magenta dots), and vma21^{Δ1/Δ14ins21} (green dots). Source data are available online for this figure.

Lysosomal acidification is impaired in vma21 mutants.

Representative images of (A) wild-type (WT)/heterozygous sibling control and (B) vma21^{Δ14ins21/Δ14ins21} zebrafish larvae at 5 days post fertilization (dpf) stained with LysoTracker Red, a marker of lysosomal acidification. Scale bars: 45 mm. (C) Bar graph showing quantification of LysoTracker Red staining. Each dot represents the average of gray values (technical replicate) and three biological replicates were performed. Data are shown as mean ± SEM. Significance at *P < 0.05 (unpaired two-tailed t test). (D) Z-stack and (E) single Z plane confocal micrographs of whole-mount zebrafish embryos at 5 dpf labeled with Lamp1 (green), Dystrophin (magenta), and DAPI (blue). Lamp1 (left panels in (D, E)) shows a punctate staining pattern and localizes in the myofibers (yellow arrowheads) and at the myotendinous junctions (MTJs) (yellow arrows). Lamp1 puncta are less bright in vma21 mutants compared to WT. Scale bars: 20 mm (D) and 10 mm (E). (F) Single Z plane confocal micrographs of myofibers isolated from zebrafish embryos at 5 dpf and labeled with Lamp1 (green), Dystrophin (magenta), and DAPI (blue). Lamp1 puncta (left panels) are present throughout the myofibers, at the MTJs (yellow arrows), and in increased numbers around the nuclei (white arrowheads). Lamp1 puncta are less bright in vma21 mutant myofibers compared to WT myofibers. Scale bar: 10 mm. (G) Quantification of Lamp1 fluorescence intensity at the MTJs shows significantly reduced levels of Lamp1 in vma21 mutants. n = 16 WT and n = 8 vma21 mutants. Data represented as mean ± SEM. Significance at ***P < 0.001 (unpaired two-tailed t test). (H) Quantification of Lamp1 fluorescence intensity in the entire myofiber shows significantly reduced levels of Lamp1 in vma21 mutants. n = 11 WT and n = 15 vma21 mutants. Data represented as mean ± SEM. Significance at **P < 0.01 (unpaired two-tailed t test). Source data are available online for this figure.

Aberrant autophagy in vma21 mutant zebrafish.

Representative electron micrographs from (A) wild-type/heterozygous sibling control and (B) vma21^Δ1/Δ1 zebrafish larvae skeletal muscle at 6 days post fertilization (dpf), showing the presence of double-membrane vacuoles in vma21 mutants but not controls. Scale bars: 2 mm. (C) Representative western blot of LC3I, LC3II, and b-actin (housekeeping protein) protein levels in the vma21 1 bp deletion (vma21Δ1/Δ1) and the 14 bp deletion with 21 bp insertion (vma21Δ14ins21/Δ14ins21) homozygous mutants as well as the compound heterozygous line (vma21Δ1/Δ 14ins21) compared to their respective wild-type/heterozygous sibling controls at 5 days post fertilization. LC3I protein is ~19 kDa, LC3II protein is approximately 17 kDa, and b-actin is approximately 42 kDa. Each lane represents n = 10 zebrafish (50 mg of total protein). Each western blot had a total of three biological replicates. (D) LC3I and LC3II protein levels were quantified by normalization to b-actin loading controls. The LC3II/LC3I ratio was also determined. For each sample, a total of n = 30 zebrafish were utilized. Data shown as mean ± SEM. Significance at *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA). Representative fluorescent images of (E, F) wild-type/heterozygous sibling control and (G, H) vma21^Δ1/Δ1 zebrafish larvae in GFP and RFP channels at 5 dpf following injection of 1-cell stage embryos with the pTol2 (Ubbi: GFP-LC3-RFP-LC3ΔG) construct used to assess autophagic flux. Scale bar: 500 mm. (I) GFP/RFP ratio as a measure of autophagic flux in 5-dpf sibling control and vma21^Δ1/Δ1. n = 15–17/group. Data represented as mean ± SEM. Significance at ***P < 0.001 (unpaired two-tailed t test). Source data are available online for this figure.

vma21 mutants have hepatic lipid accumulation, smaller liver size, and disrupted bile flux.

Representative H&E staining of (A) WT and (B) ma21^Δ1/Δ1 mutants at 6 dpf. The liver is outlined in white and the black arrow highlights a lipid droplet. n = 6 per group. Representative fluorescent images of the liver of (C) Cas9-only and (D) vma21 crispants injected into the Tg(Fabp:mCherry) transgenic line at 5 dpf. The fluorescent red liver is outlined in white. (E) Quantification of liver size at 6 dpf. n = 27–30/group. Data represented as mean ± SEM. Significance at ****P < 0.0001 (unpaired two-tailed t test). Representative fluorescent images of (F) wild-type (WT) and (G) vma21^Δ1/Δ1 mutants (homozygous for a 1 base pair (bp) deletion in vma21) at 6 days post fertilization (dpf) fed a BODIPY diet. The gallbladder is outline by the white dotted line. ×32 Magnification, scale bar: 200 mm. SB swim bladder, i intestine. (H) Quantification of the percentage of larvae with fluorescence in the gallbladder at 6 dpf. n = 26-29/group. Data represented as mean ± SEM. Significance at ****P < 0.0001 (Fisher’s exact test). Source data are available online for this figure.

Autophagy antagonist drug library screening identifies corrective molecules in the vma21 mutant zebrafish.

(A) Results from birefringence assessment of vma21^Δ14ins21 heterozygote in-cross matings with zebrafish larvae incubated in autophagy library drug compounds. Dorsal muscle birefringence assessments of the drug-treated vma21^{Δ14ins21/Δ14ins21} homozygote zebrafish larvae offspring. Zebrafish were scored as dead (gray bars; no detectable heartbeat), affected (red bars; showing at least one area of poor birefringence), or unaffected (green bars; no birefringence). Dashed lines shown at 25%, 18%, and 12.5% target range for lead compound advancement. (B) Long-term survivability assessment of eight autophagy antagonists tested in the vma21^{Δ14ins21/Δ14ins21} homozygote mutant zebrafish advanced from the first pass short-term screen, LY294002 (positive control), and vehicle and untreated (negative controls). (C) Touch-evoke response measured with % of fish shown and groupings of high responders (green bars), medium responders (yellow bars), and low/non-responders (red bars). Compounds are administered into vma21^{Δ14ins21/Δ14ins21} homozygote mutant zebrafish at 1 dpf and drug water was changed every other day over a 21-day time period. The compounds include: ROC-325, GSK2578215A, DC661, MRT68921 HCl, Lucanthone, CZC-25146, GNE7915, edaravone, and LY294002 administered at 2.5 µM. Vehicle and untreated controls are also shown. One hundred (n = 100) vma21^{Δ14ins21/Δ14ins21} mutant fish were assessed over 3 independent trials. Compounds were later validated in the vma21D1/D1 mutant zebrafish. (D) Percentage of larvae with fluorescence in the gallbladder at 6 dpf following feeding with BODIPY as a measure of bile acid flux. n = 10–22/group. Data represented as mean ± SEM. Not significant (ns; P > 0.05, Fisher’s exact test). Source data are available online for this figure.