Diverse effects of coexpression of human SOD1 variants on motor neuron disease

Mutations in superoxide dismutase-1 (SOD1) are a common cause of amyotrophic lateral sclerosis (ALS). Inheritance is as a rule dominant, but in carriers of the most common mutation, D90A, disease can develop in both homozygous and, more rarely, in heterozygous individuals with unexplained differences in clinical presentation. There is mounting evidence that prion-like spread of SOD1 aggregation is the primary cause of the disease. Two different strains of aggregates have been found to arise in human SOD1 (hSOD1) transgenic mouse models of ALS. Strain A is formed by most mutants including hSOD1^G85R and hSOD1^WT, whereas hSOD1^D90A transgenic mice form a distinct strain B in addition to A. To explore the effects of aggregate strain propensities when hSOD1 variants are coexpressed, we generated digenic hSOD1^G85R/WT and hSOD1^G85R/D90A mice. Coexpression of hSOD1^WT considerably shortened the lifespan of hSOD1^G85R mice to the extent expected from the neurotoxicities of the variants alone. In contrast, coexpression of hSOD1^D90A had a minimal effect on survival, far smaller than expected. Moreover, time from onset to the end stage was markedly prolonged in the hSOD1^G85R/D90A mice. Aggregation of hSOD1 developed concomitantly with motor neuron disease, and the aggregates contained large amounts of both coexpressed variants in both digenic models. Our findings suggest that hSOD1^WT has high a capacity to coaggregate with mutants and enhance neurotoxicity. Such interactions may be restricted by differences in strain propensities, which may contribute to the primarily recessive inheritance associated with the hSOD1^D90A mutation.