Copper-zinc superoxide dismutase (CuZnSOD, SOD1 protein) is an abundant copper- and zinc-containing protein that is present in the cytosol, nucleus, peroxisomes, and mitochondrial intermembrane space of human cells. Its primary function is to act as an antioxidant enzyme, lowering the steady-state concentration of superoxide, but when mutated, it can also cause disease. Over 100 different mutations have been identified in the sod1 genes of patients diagnosed with the familial form of amyotrophic lateral sclerosis (fALS). These mutations result in a highly diverse group of mutant proteins, some of them very similar to and others enormously different from wild-type SOD1. Despite their differences in properties, each member of this diverse set of mutant proteins causes the same clinical disease, presenting a challenge in formulating hypotheses as to what causes SOD1-associated fALS. In this review, we draw together and summarize information from many laboratories about the characteristics of the individual mutant SOD1 proteins in vivo and in vitro in the hope that it will aid investigators in their search for the cause(s) of SOD1-associated fALS.

The thermodynamics of zinc binding to metal-free (apo) human and bovine copper-zinc superoxide dismutases (SOD1) were measured using isothermal titration calorimetry. The apparent thermodynamics of zinc binding to the apoproteins were favorable (Ka > 108 M-1), with an observed stoichiometry of one zinc per homodimer. The change in heat capacity for the one-zinc binding event was large and negative (approximately -650 cal mol-1 K-1), suggestive of significant structural changes to the protein upon zinc binding. We further characterized the one-zinc derivative by circular dichroism and determined that this derivative had nearly the same secondary structure as the two-zinc derivative and that both are structurally distinct from the metal-free protein. In addition, we monitored the effect of zinc binding on hydrogen-deuterium exchange and accessibility of histidyl residues to modification by diethyl pyrocarbonate and observed that more than 50% protection was afforded by the binding of one zinc in both assays. Differential scanning calorimetry on the human SOD1 zinc derivatives also showed increased thermostability of the protein due to zinc binding. Further, the melting transitions observed for the one-zinc derivative closely resembled those of the two-zinc derivative. Finally, we observed that the quaternary structure of the protein is stabilized upon binding of one and two zinc ions in analytical ultracentrifugation experiments. Combined, these results suggest communication between the two monomers of SOD1 such that the binding of one zinc ion per homodimer has a more profound effect on the homodimeric protein structure than the binding of subsequent metal ions. The relevance of these findings to amyotrophic lateral sclerosis is discussed.

The existence of a link between some cases of familial amyotrophic lateral sclerosis (FALS) and copper-zinc superoxide dismutase (CuZnSOD) has been understood for almost a decade. However, beyond the fact that mutations in CuZnSOD cause FALS by a toxic gain of function, the mechanism whereby specific mutations in the protein structure result in development of the disease has remained almost a complete mystery to date. We have undertaken a critical survey of in vitro characteristics of over 30 of the 90 different CuZnSOD mutant proteins that are known to cause FALS in order to determine the differences that exist between mutant and wild-type properties. As-isolated metal content analysis, SOD activity assays, and thermal stability determinations of a significant fraction of the mutants show that the FALS mutant SOD proteins can be classified distinctly into one of two groups. Members of the first group, termed wild-type-like, have physical properties and enzymatic activities that are strikingly similar to those of wild-type CuZnSOD. The second group, however, show aberrant metal content in the as-isolated forms, compromised SOD activities, and unusual DSC thermoscans. All mutations in the members of this second group occur in or near the metal binding sites of the protein and thus they are termed metal binding region mutants. We have also compared the relative rates of self-inactivation caused by reaction of the wild-type protein and several FALS-linked CuZnSOD mutants with hydrogen peroxide, as a measure of relative peroxidative activities. Results and implications of the role of CuZnSOD in FALS are discussed.

No abstract