Fig. 7

a Experimental strategy in zebrafish models. Injected with WT and mutant ARF3-encoding mRNAs at 1 cell stage and phenotyped at different stages. b–b’ Images and close-ups of ARF3^WT and ARF3^K127E expressing embryos at 24 hpf, co-injected with GFP-CAAX-encoding mRNA and Phenol-Red (dashed circle depicts cephalic region). c Bright-field images of embryos expressing WT and mutant ARF3. The images are representative of embryos from two (b, b’) and five (c) independent batches. c’ Embryo survival, no. of embryos = 246, 114, 53, 86, 85, 161, 114 (not injected, WT, K127E, *p = 0.03, L12V/D67V, P47S, D93N, T32N) from pooled batches. c” Incidence of gross phenotypes at 24 hpf (classes: I, II = yellow arrows, III, IV = gray and black arrows, respectively), no. of embryos = 132 (not injected); 69 (WT); 21 (K127E, ****p < 0.0001); 58 (L12V/D67V, *p = 0.02); 45 (P47S, *p = 0.03); 86 (D93N, **p = 0.0018); 64 (T32N, ****p < 0.0001). Data are expressed as mean ± SEM of four (not injected, WT), three (D93N), and two (K127E, L12V/D67V, P47S, T32N) independent batches. d–d” Bright-field images (d) and phenotype incidence at 24 and 48 hpf of arf3a/arf3b MO-injected embryos (d’, d”). Respectively, in d’ and d” no. of embryos = 50, 48 (not injected); 25, 22 (MO 0.4 mM, ***p = 0.0002, ****p < 0.0001); 31, 27 (MO 0.6 mM **** < 0.0001); 21, 17 (MO 0.8 mM, ****p < 0.0001) of one batch. e–e” Bright-field images (e) and phenotype incidence at 24 and 48 hpf (e’, e”) of arf3a/arf3b MO-injected embryos (0.6 mM)−/+ARF3^WT-encoding mRNA. The images in d and e are representative of embryos of one batch. Respectively in e’ and e”, no. of embryos = 47 (not injected); 22, 18 (MO 0.6 mM, ****p < 0.0001); 17,15 (MO 0.6 mM + ARF3, **p = 0.0091 in e”) of one batch. f Phenotype worsening index at 48 hpf (fold-change) for ARF3 mutants (severe + deceased) compared to controls (co-injected with arf3 MO). In the scatter plot the values < =0 (green) are found in the “alleviation window” depicted with green shading. Dots represent the “worsening index” for each experiment, calculated by dividing the percentage of severely diseased fish (class IV–V) in “MO+” condition by the same percentage obtained in “MO−”condition of two (K127E, L12V/D67V, P47S, T32N) or three (D93N) independent batches. The mean effect of MO for each mutation is also shown as bar graph. No. of embryos = 21 and 36 (K127E – and + MO *p = 0.0307); 58 and 47 (L12V/D67V – and + MO, **p = 0.0068); 45 and 54 (P47S – and +MO); 86 (D93N+ and − MO, ***p = 0.0004); 64 and 35 (T32N – and + MO, *p = 0.0370). Data in the bar graphs are expressed as a mean ± SEM of two independent batches. Survival is assessed by Log-Rank (Mantel–Cox) test (c’), Two-sided Chi-square’s test in a 2 × 2 contingency table (class II, III and IV vs. I in c”, d’, d”, e’, e”) or Two-sided One sample t-test (class III/ IV/V vs. I in f) testing null hypothesis H₀, represented by the expected mean value of the control population, are used to assess statistical significance. Source data are provided as a Source Data file.