Mutations in Cu/Zn superoxide dismutase (SOD), a hallmark of familial amyotrophic lateral sclerosis (FALS) in humans, are shown here to confer striking neuropathology in Drosophila. Heterozygotes with one wild-type and one deleted SOD allele retain the expected 50Y% of normal activity for this dimeric enzyme. However, heterozygotes with one wild-type and one missense SOD allele show lesser SOD activities, ranging from 37% for a heterozygote carrying a missense mutation predicted from structural models to destabilize the dimer interface, to an average of 13% for several heterozygotes carrying missense mutations predicted to destabilize the subunit fold. Genetic and biochemical evidence suggests a model of dimer dysequilibrium whereby SOD activity in missense heterozygotes is reduced through entrapment of wild-type subunits into unstable or enzymatically inactive heterodimers. This dramatic impairment of the activity ofwild-type subunits in vivo has implications for our understanding of FALS and for possible therapeutic strategies.Mutations in Cu/Zn superoxide dismutase (SOD, EC 1.15.1.1) have been identified in the etiology of familial amyotrophic lateral sclerosis (FALS), a syndrome commonly known as Lou Gehrig's disease (1, 2). Some FALS-affected individuals are carriers of missense mutations that appear to alter the stability (3) and/or the function of this critical oxygen radicalmetabolizing enzyme (2). The dominant phenotype associated with FALS mutations in SOD could result from loss of enzyme function, consistent with the reduced SOD activity in FALS patients (2, 4-7), or from the gain of a deleterious function, such as enhanced reactivity with peroxynitrite leading to elevated protein-tyrosine nitration (8, 9), consistent with the motor neuron damage found in transgenic mice overexpressing FALS-type mutant SOD (10, 11).A direct test of the relationship between SOD subunit function and neurodegenerative disease would be to induce and select mutations in SOD in an organism amenable to genetic analysis and then characterize the biochemical and neurological consequences conferred by these mutations. We have previously shown that null mutation for SOD in Drosophila melanogaster confers a syndrome including reduced adult lifespan, infertility, toxic hypersensitivity to a variety of oxygen stress conditions, and lethality in combination with mutations conferring defects in other oxygen radical scavengers (12-16). Overt symptoms of the SOD-null syndrome have been described only in homozygotes and are ameliorated by P element-vectored SOD transgenes (ref. 17; T. Parkes, A.J.H., and J.P.P., unpublished data).Here we show that mutations in SOD in Drosophila, as in humans, cause neuropathology. We also show that missense mutations expected to destabilize SOD subunits and dimer assembly significantly impair the activity of normal subunits in heterozygotes for a wild-type allele, resulting in significantly lower enzyme activity than expected in heterozygotes. These findings are important to our understanding of...