Adult eng^mu130 fish display cutaneous and retinal vascular malformations similar to those in patients with hereditary haemorrhagic telangiectasia and an enlarged heart. (A) Adult 5- to 6-month-old eng^mu130 zebrafish display cutaneous vascular malformations (black arrows). Scale bars: 1 cm. (B) Representative eng^mu130 and wild-type (WT) hearts (n=5/group). Scale bars: 500 μm. (C) Body weight of adult WT and eng^mu130 zebrafish. (D) Cardiac ventricle weight/body weight of WT and eng^mu130 zebrafish. (E) Fluorescent micrographs of whole-mount retinae from Tg(kdrl:Hsa.HRAS-mCherry)^s916 zebrafish, showing abnormal vessel communications in eng^mu130 (white arrowheads). Scale bars: 500 μm. (F) Schematic diagram of a 5- to 6-month-old zebrafish retinal vasculature. Figure created with BioRender. (G) Number of retinal vascular branches in WT and eng^mu130 homozygous siblings. (H) Optic artery diameter in WT and eng^mu130 homozygous siblings. (I) Inter-capillary distance in WT and eng^mu130 homozygous siblings. **P<0.01, ****P<0.0001 (unpaired two-tailed Student's t-test, 8-10 animals/group).

eng^mu130 mutant embryos have increased basilar artery diameter and increased number of endothelial kugeln. (A) Representative maximum-intensity projection of Tg(kdrl:Hsa.HRAS-mCherry)^s916 zebrafish brain vasculature at 72 hpf. Black arrowheads indicate individual kugeln. Scale bars: 150 μm/50 μm (inset). Three-dimensional reconstruction of the same image shown in green is on the right. (B) Representative maximum-intensity projection of the basilar artery (BA) at 72 hpf. Black arrowheads indicate individual kugeln. White bars indicate where diameter of the BA was measured. Scale bar: 50 μm. Three-dimensional reconstructions of the same images shown in green are on the right. (C) BA diameter in eng^mu130 mutants and WT siblings (unpaired two-tailed Student's t-test, 10-12 animals/group). (D) eng mutation did not alter endothelial cell (EC) number in the BA (unpaired two-tailed Student's t-test, 10-12 animals/group). (E) Kugel number per animal was increased in eng^mu130 mutants at 52 and 74 h post fertilisation (hpf) but not at 100 hpf. *P<0.05; ns, not significant (Mann–Whitney U-test, 8-27 animals/group).

Vegf inhibition between 2 and 3 dpf rescues the abnormal trunk and cerebral vessel phenotypes of eng^mu130 mutants. (A) Experimental plan for rescuing the eng mutant phenotype in zebrafish using AV951. dpf, days post fertilisation. (B) Representative maximum-intensity projection of the trunk vasculature of eng^mu130 and WT embryos±AV951 (25 nM) treatment for 24 h. DA, dorsal aorta; PCV, posterior cardinal vein. Scale bar: 150 μm. (C) Representative maximum-intensity projection of the cerebral vasculature of eng^mu130 and WT embryos±AV951 (25 nM) treatment for 24 h. Scale bar: 100 μm. (D) DA diameter in eng^mu130 and WT embryos±AV951 treatment. (E) PCV diameter in eng^mu130 and WT embryos±AV951 treatment. (F) Intersegmental blood vessel (ISV) diameters in eng^mu130 and WT embryos±AV951 treatment. (G) Percentage of open ISVs in eng^mu130 and WT embryos±AV951 treatment. (H) Number of kugeln in eng^mu130 and WT embryos±AV951 treatment. (I) BA diameter in eng^mu130 and WT embryos±25 nM AV951 treatment. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significant (two-way ANOVA with Tukey post-hoc test, 10 animals/group).

MEK or TOR inhibition normalises the vascular phenotype of eng^mu130 embryos. WT and eng^mu130 mutant embryos were treated with the MEK inhibitor PD0325901 (10 µM) or with the TOR inhibitor rapamycin (2.5 µM) for 24 h between 2 and 3 dpf, or with DMSO vehicle. (A) Representative maximum-intensity projections of the trunk vessels of eng^mu130 and WT embryos±PD0325901 (10 µM) or rapamycin (2.5 µM) treatment for 24 h. Scale bar: 150 μm. (B) Representative maximum-intensity projections of the cerebral vessels of eng^mu130 and WT embryos±PD0325901 (10 µM) or rapamycin (2.5 µM) treatment for 24 h. Scale bar: 100 μm. (C,D) eng^mu130 mutant embryos show reduced DA (C) and PCV (D) diameter following PD0325901 treatment, with PCV diameter normalised to WT values. (E,F) eng^mu130 mutant embryos show reduced DA diameter (E) and PCV diameter (F) following rapamycin treatment, with PCV diameter normalised to WT values. (G) eng^mu130 mutant embryos show normalised kugel formation following PD0325901 treatment. (H) eng^mu130 mutant embryos show normalised kugel formation following rapamycin treatment. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significant (two-way ANOVA with Tukey post-hoc test, 10-14 animals/group).

Combined low-dose mTOR and MEK inhibition prevents abnormal trunk vasculature and excess kugel formation in eng^mu130 zebrafish embryos. (A-C) Neither 2 µM rapamycin (RAP) nor 7.5 µM PD0325901 (PD0) alters the vascular phenotype of eng^mu130 mutants, but combined treatment resulted in normalised DA and PCV diameters, as well as reduced kugel formation in eng^mu130 zebrafish embryos. *P<0.05, ***P<0.001, ****P<0.0001; ns, not significant (two-way ANOVA with Tukey post-hoc test, 10-14 animals/group).