Heterozygous RAB3A variants cause cerebellar ataxia by a partial loss-of-function mechanism.

Abstract

RAB3A encodes a small GTP-binding protein that is abundant in brain synaptic vesicles and crucial for the release of neurotransmitters and synaptic plasticity. Here we identified RAB3A as a candidate gene for autosomal dominant cerebellar ataxia by two independent approaches: linkage in a large dominant ataxia family and, in parallel, an untargeted computational genetic association approach, analyzing the 100,000 Genomes Project datasets. To further validate the role of RAB3A in ataxia, we next screened large rare disease databases for rare heterozygous RAB3A variants in probands with ataxia features. In total, we identified 18 individuals from 10 unrelated families all sharing a cerebellar ataxia phenotype. Notably, 9 out of 10 families carried a recurrent variant in RAB3A, p.Arg83Trp, including one de novo occurrence. In addition, our screening revealed three families with a neurodevelopmental phenotype and three unique RAB3A variants, which were either de novo or loss-of-function variants. In line with the different RAB3A variant types, protein domains, and predicted functional consequences, a comprehensive set of complementary methods was used to functionally characterize the identified variants. As expected, GTPase-activating protein (GAP)-dependent GTP hydrolysis was reduced for those two missense variants located in the GAP binding domain of RAB3A (Arg83Trp, Tyr91Cys). In a Drosophila Rab3 loss-of-function model, these two missense variants also failed to rescue a synaptic phenotype. Overexpression of Rab3 variants in Drosophila wildtype background did not cause an obvious phenotype, making a dominant negative effect of these variants unlikely. Lastly, exploring interactors of RAB3A variants by using co-immunoprecipitation and mass spectrometry showed differential changes in variant-specific interactions with known RAB3A key regulatory and effector proteins. In sum, our results establish RAB3A as a neurological disease gene. It represents an autosomal dominant gene for cerebellar ataxia with different variants associated with disease, including the frequent reoccurring variant p.Arg83Trp. Our study sheds light on the variant-specific interactome of RAB3A. Finally, we suggest an association of RAB3A with a neurodevelopmental phenotype, as reported for variants in several RAB3A interaction partners and as seen in Rab3A-deficent mice, although this possible association warrants further investigation by future studies.