show less enzymatic activities, many retain full activity (4). Several lines of transgenic mice that express mutant human SOD1 linked with FALS developed progressive neurodegeneration and a phenotype that clearly resembles human FALS (5, 6) despite having higher SOD activity. In contrast, SOD1-knock out mice do not exhibit motor neuron dysfunction (7). These findings suggest that this disease is the result of a toxic gain of function but not a loss in SOD activity. Several different hypotheses have been proposed to explain the toxic gain of function, including the production of ROS, mitochondrial defects, glutamate-induced excitotoxicity, neurofilament inclusions, and the formation of intracellular toxic protein aggregates (reviewed in Ref. 8). However, the mechanism by which FALS mutant SODs causes motor neuron degeneration is not completely understood.The tertiary structure of Cu/Zn-SOD is characterized by the presence of a Greek key -barrel containing an internal disulfide bond between Cys-57 and Cys-146 (9, 10), both of which contribute to its high stability. On the other hand, various FALS mutant SODs show a decreased stability and a lower level of metallation (4,11,12). Several reports, including ours ( Fig. 1), 2 reported that FALS mutant SODs exhibit an accelerated turnover or an increased susceptibility to degradation in proteasome (13,14) and suggest that unfolding or conformational perturbations of mutant SOD proteins occur in vivo. Niwa et al. (15) reported that Dorfin, a RING finger-type ubiquitin-protein isopeptide ligase (E3) ubiqutinates the FALS mutant SODs but not wild-type Cu/Zn-SOD, suggesting that FALS mutant SODs have a unique structure that can be recognized by Dorfin. We reported previously that FALS mutant SODs are more susceptible to glycation or fructation (16) and that they form aggregates at a higher rate than the wild-type SOD, when incubated in the presence of copper ions (4).