Abstract
Wolfram syndrome (WS) is an ultra-rare progressive neurodegenerative disorder defined by early-onset diabetes mellitus and optic atrophy. The majority of patients harbour recessive mutations in the WFS1 gene, which encodes for Wolframin, a transmembrane endoplasmic reticulum protein. There is limited availability of human ocular and brain tissues, and there are few animal models for WS that replicate the neuropathology and clinical phenotype seen in this disorder. We, therefore, characterised two wfs1 zebrafish knockout models harbouring nonsense wfs1a and wfs1b mutations. Both homozygous mutant wfs1a-/- and wfs1b-/- embryos showed significant morphological abnormalities in early development. The wfs1b-/- zebrafish exhibited a more pronounced neurodegenerative phenotype with delayed neuronal development, progressive loss of retinal ganglion cells and clear evidence of visual dysfunction on functional testing. At 12 months of age, wfs1b-/- zebrafish had a significantly lower RGC density per 100 μm2 (mean ± standard deviation; 19 ± 1.7) compared with wild-type (WT) zebrafish (25 ± 2.3, p < 0.001). The optokinetic response for wfs1b-/- zebrafish was significantly reduced at 8 and 16 rpm testing speeds at both 4 and 12 months of age compared with WT zebrafish. An upregulation of the unfolded protein response was observed in mutant zebrafish indicative of increased endoplasmic reticulum stress. Mutant wfs1b-/- zebrafish exhibit some of the key features seen in patients with WS, providing a versatile and cost-effective in vivo model that can be used to further investigate the underlying pathophysiology of WS and potential therapeutic interventions.