a Schematic representation of the technique used to silence C9orf72 in zebrafish. The transgene is designed to express four different micro-RNAs targeting C9orf72’s 3’UTR and triggering knockdown by both repressing C9orf72 translation and affecting its stability. b Images demonstrating proper eGFP expression in the crystallin of the transgenic fish, a marker used to identify carrier/knockdown larvae. Scale bar = 100 μm. c Bar graph shows the relative expression of the endogenous C9orf72 gene. mRNA was normalized to elf1α mRNA levels (N = 4, ***p < 0.0001, Student’s t test). d Immunoblotting of the zebrafish protein C9orf72 and beta-actin as control. e Bar graph shows the relative expression of the C9orf72 protein compared with actin between C9orf72 mutants and control line (N = 3; **p = 0.0034; Student’s t test). f Gross morphological analyses of wild-type control and C9orf72-LOF fish (C9-miR). g Kaplan–Meier survival plot over 17 days after fertilization (dpf) showing low survival of C9-miR compared to controls after 10 dpf (N = 3, n = 25). Data are presented as mean ± SEM. n represents the number of fish, N represents the number of experimental repeats.

a Representative swimming tracks of control, C9-miR and C9-rescue fish at 6 dpf. Scale bar = 0.2 cm. b C9-miR larvae (N = 3, n = 65) displayed impaired swimming compared to controls (N = 3, n = 60; ***p < 0.0001; one-way ANOVA). Expression of the human C9orf72 long transcript in C9-miR fish (C9-rescue; N = 3, n = 53) significantly rescued the motor behavioural defects (***p < 0.0001; one-way ANOVA). c Representative images of co-immunostaining of zebrafish neuromuscular junctions with presynaptic (SV2; green) and postsynaptic (α-bungarotoxin; red) markers in 6 dpf zebrafish. Scale bar = 100 μm. d Quantification of the colocalizing presynaptic and postsynaptic markers per somite showed a significant reduction in the number of puncta in C9-miR fish at 6 dpf that can be rescued with the expression of human C9orf72 mRNA (C9-rescue) (n = 8–10; ***p < 0.0001; one-way ANOVA). Data are presented as mean ± SEM. n represents the number of fish, N represents the number of experimental repeats.

a Recordings of mEPCs, which result from spontaneous release of a quantum, were recorded in 6 dpf control and C9-miR fish (n = 6). b Representative mEPCs. Animals with reduced C9orf72 (C9-miR) displayed mEPCs with reduced frequency (c) (n = 6–7; ***p < 0.0001; Student’s t test) and amplitude (d) (n = 7; **p < 0.001; Student’s t test). Rise time (e) (n = 7; p = 0.3947; Student’s t test) and decay time (f) constant kinetics of mEPC (n = 7; p = 0.8385; Student’s t test) were not found to be significantly different between controls and C9-miR. Data are presented as mean ± SEM. n represents the number of fish.

a Illustration of 6 dpf zebrafish skeletal muscle cells with large nuclei labelled with muscle marker, phalloidin, and nucleus marker, Hoechst. b Representative images of 6 dpf zebrafish skeletal muscle cells for TDP-43. Compared to control fish, we observed cytoplasmic clustering of TDP-43 expression in the C9-miR skeletal muscles that can be rescued in C9-rescue fish. Arrows indicate clusters of TDP-43 expression. c Quantification of the nucleus-to-cytoplasmic ratio for TDP-43. A significant reduction in N-to-C for TDP-43 was observed in C9-miR zebrafish (n = 7; **p < 0.005; one-way ANOVA) that can be significantly rescued upon the expression of the human C9orf72 mRNA (n = 5; *p < 0.05; one-way ANOVA). Scale bar = 50 μm. Data are presented as mean ± SEM. n represents the number of fish.

a Representative traces of swimming activity of five adult controls and C9-miR fishes (12-month old) during 30 s (left panel). b C9-miR fish exhibit behavioural deficits (right panel) (n = 5, ***p < 0.0001, Student’s t test). c Adult 12-month-old NMJs were examined in trunk section by co-immunostaining SV2 (green) and α-bungarotoxin (red). d Quantification of the colocalizing presynaptic and postsynaptic clusters at NMJ in adult (12-month old) wild-type control (n = 6) and C9-miR fish (n = 7) (***p < 0.0001, Student’s t test). e ChAT staining in adult zebrafish spinal cord . f Large (mature) motor neurons (inset in e; scale bar = 10 μm) are reduced in size in C9-miR compared to controls (n = 4; ***p < 0.0001, Student’s t test). g Examination of adult zebrafish muscle myotomes by haematoxylin and eosin staining. h C9-miR fish display a smaller diameter of muscle fibres compared to controls (n = 10; ***p < 0.0001, Student’s t test). Data are presented as mean ± SEM. Scale bar = 50 μm. n represents the number of sections from three adult fish per genotype.

Representative fluorescence images of spinal motor neurons immunostained with antibodies against ChAT (green), TDP-43 (red) and labelled with DAPI (blue) in 14–16-month-old adult (a) control (wild type) or (b) C9-miR spinal cord sections. Scale bar = 20 μm. Arrows and arrowheads illustrate TDP-43 mislocalization. c Quantification of colocalization (%) between the total cellular TDP-43 antibody signal and the nuclear DAPI stain in ChAT-positive motor neurons in C9-miR fish normalized to colocalization percentage in control fish (n = 3; *p = 0.0251, Student’s t test). Data are presented as mean ± SEM. n represents the number of fish.

a Volcano plot showing the log2 fold change against the −log10 p value. Proteins in blue are upregulated while proteins in red are downregulated (log2 fold change of −1.5 and 1.5, delineated by vertical lines) and are significantly dysregulated (−log10 (p value) > 1.301, delineated by horizontal dotted line) between control and C9-miR fish (N = 4). b Differentially expressed proteins with p < 0.05 in C9-miR 6 dpf larvae (N = 4). c Immunoblot analysis of RFP-immunoprecipitated proteins and lysate of HEK293T cells co-expressing RFP-tagged C9orf72 or RFP in combination with Myc-tagged SV2a or Myc-tagged SMRC8 (a known interactor of C9orf72; positive control). Myc-SV2a and Myc-SMCR8 were extracted out of solution in samples where RFP-C9orf72 was co-transfected, but not in samples where RFP was co-transfected (N = 3).

a FM1-43 loading of NMJ boutons in 6 dpf fish. C9-miR fish displayed decreased FM1-43 loading compared to controls. b Quantification of FM1-43 fluorescent intensity in control (n = 14) and C9-miR fish (n = 16), showing a reduced FM1-43 fluorescent intensity in C9-miR fish (p < 0.0001; Student’s t test). c Putative synapses (arrows) were visualized with Rab3a immunostaining at 6 dpf (inset; scale bar = 10 μm). Rab3a+ synaptic puncta were reduced in number (d) and area (e) in 6 dpf C9-miR larvae when compared with wild-type controls (n = 5; ***p < 0.0001; Student’s t test). Scale bar = 50 μm. Data are presented as mean ± SEM. n represents the number of fish.