Abnormal fucosylation but normal neuromast morphology in gmds^−/−. Aleuria aurantia lectin staining for fucosylated proteins (A,B) is present on control (n = 20), but not gmds^−/− neuromasts (n = 14). Scanning electron microscope (SEM) image of the lateral line neuromasts at 5 dpf under homeostatic conditions in wild-type (n = 9) and gmds^−/− (n = 9) (C,D) with a higher magnification image of a supraorbital neuromast in (E,F).

Increased neuromast hair cell number in gmds^−/− mutants under homeostatic conditions. The SO3 and O1 neuromasts show a statistically significant increase in hair cell number at 5 dpf in gmds^−/− mutants when compared to control siblings, as evidenced by YO-PRO-1 staining (A–D). While the other quantified neuromasts (SO1, SO2, O2) show a trend of increased hair cell number under homeostatic conditions, these data were not statistically significant (E). Data from two experiments (WT n = 11, Mut n = 7). Data presented as mean ± SEM, significance testing using a one-factor ANOVA with Sidak’s multiple comparisons test * p < 0.05.

No change in neuromast genetic markers or neural innervation in gmds^−/− mutants. Expression level of the gmds gene at 3 dpf shows downregulation in homozygous mutants when compared to control siblings (A,B). No changes are observed in the expression of slc17a8 (C,D), the neuromast-specific fucosyltransferase fut9b (E,F) or the support cell specific sox2 (G,H) between gmds^−/− and control siblings at 3 dpf. In situ hybridization experiment was repeated three times for each genetic marker (WT n = 15, gmds^−/− n = 15), with pictures at 40X. Neuromasts are innervated, with both gmds^−/− and control siblings staining positive for Hnk-1 (I,J, pictures taken at 200X) and acetylated tubulin (K,L, pictures taken at 630X) using immunohistochemistry from two independent experiments (WT n = 6, gmds^−/− n = 6). Arrows point to nerve innervation in K and L.

Increased neuromast regeneration capacity of gmds^−/− mutants with DASPEI Staining. WT and gmds^−/− mutant unablated neuromasts at 6 dpf from four independent experiments (WT n = 17, gmds^−/− n = 18) (A,B). Wild-type and gmds^−/− mutants show a statistically significant difference in the number of labelled neuromasts 24 h post neomycin ablation. For four experiments, WT n = 17, gmds^−/− n = 19 (C,D,I). A further increase is observed at 48 hpa WT n = 23, gmds^−/− n = 24 (E,F,I) and 72 hpa WT n = 18, gmds^−/− n = 17 (G–I). Control siblings quantified at the same time points show no difference in neuromast number (J). Images taken at 40X. Data presented as mean ± SEM, significance testing using a one-factor ANOVA with Sidak’s multiple comparisons test * p < 0.05, ** p < 0.01, *** p < 0.001. ns = not significant.

Increased rate of regeneration of larval head neuromasts after neomycin ablation. Neuromasts are morphologically similar in control and gmds^−/− mutants at 5 dpf without neomycin ablation (A,E). Little difference is seen in neuromast morphology at 24 hpa (B,F), but by 48 hpa, neuromast morphology is normal in gmds^−/−, implying complete regeneration (G), while control siblings have not yet fully regenerated (C). By 72 hpa, both control and gmds^−/− neuromasts appear fully regenerated (D,H). Data shown from three experiments for each condition or time post-treatment, WT n = 9, gmds^−/− n = 9. Scalebar in H is 10 μm.

Increased rate of regeneration of larval tail neuromasts after neomycin ablation. Neuromasts are morphologically similar in control and gmds^−/− mutants at 5 dpf without neomycin ablation (A,E). Little difference is seen in neuromast morphology at 24 hpa (B,F), but by 48 hpa, neuromast morphology is normal in gmds^−/−, implying complete regeneration (G), while control siblings have not yet fully regenerated (C). By 72 hpa, both control and gmds^−/− neuromasts appear fully regenerated (D,H). Data shown from three experiments for each condition or time post-treatment, WT n = 9, gmds^−/− n = 9. Scalebar in H is 5 μm.

Quantitative assessment of increased hair cell regeneration of gmds^−/− mutants with YO-PRO-1 staining. Data shown were from two independent experiments. At 48 hpa, a trend of increased hair cell number is observed in gmds^−/− mutants, which is statistically significant for the SO3 neuromast (WT SO1 n = 6, Mut SO1 n = 8, WT SO2 n = 7, Mut SO2 n = 7, WT SO3 n = 7, Mut SO3 n = 9, WT O1 n = 9, Mut O1 n = 7, WT O2 n = 7, Mut O2 n = 6) (A,C). By 72 hpa, all assayed neuromasts show a statistically significant increase in hair cell number in gmds^−/− mutants when compared to control siblings (WT SO1 n = 10, Mut SO1 n = 5, WT SO2 n = 12, Mut SO2 n = 13, WT SO3 n = 13, Mut SO3 n = 13, WT O1 n = 13, Mut O1 n = 13, WT O2 n = 12, Mut O2 n = 9) (B,D). Data presented as mean ± SEM, significance testing using a one-factor ANOVA with Sidak’s multiple comparisons test * p < 0.05, ** p < 0.01, **** p < 0.0001.

Inhibition of Notch signalling increases hair cell number after ablation in mutants and control groups (wild-type and gmds +/- siblings). Hair cell YO-PRO-1 counts 48 h post neomycin treatment, with DMSO control group (WT n = 17, +/− n = 27, −/− n = 31) or Notch inhibition with 10 µm DAPT group following ablation (WT n = 6, +/− n = 29, −/− n = 16) (A). Hair cell YO-PRO-1 counts 72 h post neomycin treatment, with DMSO control group (WT n = 18, +/− n = 20, −/− n = 17) or Notch inhibition with 10 µm DAPT treatment following ablation (WT n = 18, +/− n = 34, −/− n = 7) (B). Data presented as mean ± SEM, significance testing using a one-factor ANOVA with Sidak’s multiple comparisons test * p < 0.05, *** p < 0.001, **** p < 0.0001.