Abstract
Neurodegenerative diseases, including polyglutamine diseases, remain a clinical challenge, partly because of limited animal models that recapitulate human disease. Here, we describe a second-generation transgenic marmoset model of spinocerebellar ataxia 3 (SCA3), a polyglutamine disease, which stably expresses expanded CAG repeats in ataxin 3 (ATXN3). All five offspring of the founder marmoset harbored the transgene with reduced transgene integration sites compared with the founder and without repeat instability or genetic mosaicism, offering improved construct validity. Three of the five marmosets developed progressive motor impairments that segregated into two distinct phenotypes - early onset with rapid progression and late onset with mild progression - accompanied by corresponding patterns in body weight gain and grip strength. Pathological analysis revealed cerebellar Purkinje cell loss, spinal cord neurodegeneration and widespread intranuclear inclusions. The severity of motor phenotypes correlated with transgene expression levels in disease-relevant brain regions, including the cerebellum and spinal cord. By overcoming the translational limitations of rodent systems, our second-generation model offers a powerful platform for investigating disease mechanisms and testing potential therapeutics, advancing the utility of transgenic marmosets as clinically relevant models of neurodegenerative diseases.