Giant Axonal Neuropathy (GAN) is a rare neurodegenerative disorder affecting axons in the peripheral and central nervous system, accompanied by axonal swellings and axon loss. Disease onset is around 3 years of age, and is characterized by muscle weakness, impaired sensation, gait disturbance, mental retardation, and seizures. While GAN is caused by presumptive loss-of-function mutations in the human GAN gene, the etiology of the disease is poorly understood. We aim to establish a zebrafish model of GAN to gain insight into the cellular and biochemical mechanisms underlying the disease phenotype. GAN encodes a 468 amino-acid protein, Gigaxonin, a member of the BTB-Kelch superfamily involved in the ubiquitin-proteasome pathway. In mice, gan is expressed extensively in multiple tissues, including the brain and spinal cord. Similarly, zebrafish gan is expressed from 18 hpf to 4 dpf, with high expression in the brain at 4 dpf.
To investigate roles of gan in disease, we generated loss-of-function mutations in gan using the CRISPR/Cas9 system. Using guide RNA targeted to the 2nd exon (AA 84-90 of Gigaxonin), we identified two alleles with frameshift mutations resulting in premature stop codons. Since sensory-motor functions are compromised in GAN patients, we examined two behaviors in zebrafish gan mutants involving sensorimotor circuits: touch-evoked escape response and food intake. Wild type and mutant larvae exhibited normal escape responses when touched in the trunk. In contrast, a significantly larger proportion of mutants responded when touched on the head, consistent with extensive gan expression in the hindbrain, where touch responses are processed. Feeding is a complex behavior requiring the coordination of sensory (vision, olfaction) and motor (jaw movement, locomotion) activities. At 7 dpf, mutant larvae exhibited a significant decrease in food intake compared to wild type siblings, consistent with gan expression in the hindbrain, where the motor neurons driving jaw movements are located. These data suggest that zebrafish gan mutants have defects in sensorimotor functions consistent with the defects seen in the mouse GAN model and in human patients, and may represent a good model for GAN in a simple vertebrate.