No abstract

The accepted mechanism of Fe-containing superoxide dismutase (Fe-SOD) activity and inhibition by anions implies the existence of a group with a pK of 8.6-9.0 in the active site of reduced Fe-SOD [Bull, C. & Fee, J. A. (1985) J. Am. Chem. Soc. 107, 3295-3304]. We have performed pH titrations of reduced Fe-SOD by NMR spectroscopy and observe a pK of 8.5 at 30°C which is the only pK affecting the active site between pH 5.5 and 10.5. Thus, we present the first spectroscopic evidence of the predicted pK. Although the pK is associated with chemical shift changes for almost all of the resonances of the active site, resonance line widths and the T 1 of a ligand proton are not significantly affected by the pK, indicating that there is no significant conformational change and only relatively minor effects on the electronic spin properties of Fe 2+ . The changes in chemical shift are probably caused by changes in hydrogen bonding to a ligand and attendant subtle perturbation of the Fe 2+ paramagnetism upon loss of the proton with the pK of 8.5. The pK is also associated with a dramatic restriction of the exchange of at least one ligand proton. Thus, active site accessibility to solvent and OH -decreases by more than 2 orders of magnitude upon loss of the proton with the pK of 8.5. Since OH -is a competitive inhibitor of Fe-SOD, and thus a substrate analog, this dramatic and unusual decrease in accessibility to OH -is consistent with the increase in the K M for O 2•-that is associated with a pK near 9.

We have compared the magnetic resonance properties and pH dependence of wild-type and mutant Fe-containing superoxide dismutase (Fe-SOD) in which the conserved active site tyrosine (Tyr 34) is replaced by phenylalanine. The EPR spectrum of the oxidized state and the NMR spectrum of the paramagnetically shifted resonances of the reduced state indicate that in both states the active site is relatively unperturbed by the mutation. Similarly, the mutant Fe-SOD retains approximately 41% of wild-type catalytic activity on a per Fe basis. However, replacement of Tyr 34 by Phe abolishes both NMR spectroscopic signatures of the active site pK of 8.5 of (reduced) Fe2+-SOD. Neither accessibility to base-catalyzed exchange nor the chemical shifts of active site residues are affected by pH in the range of 6.5-10.5 in Y34F Fe2+-SOD. Thus, the active site pK of 8.5 of Fe2+-SOD most likely corresponds to deprotonation of Tyr 34. The widespread chemical shift changes associated with the pK could reflect Tyr 34's participation in the active site hydrogen bonding network and the network's propagation of the effects of deprotonating Tyr 34 to the Fe2+. Deprotonation of Tyr 34 can also explain the dramatic decrease in active site accessibility to base-catalyzed exchange as the result of electrostatic repulsion between the exchange catalyst OH- and the (Tyr 34)- ion formed at high pH. Similar electrostatic repulsion between (Tyr 34)- and the substrate O2.- is also consistent with the observed increase in KM above pH 9.

We report the first spectroscopic observation of substrate analogue binding to the reduced state of iron superoxide dismutase from Escherichia coli (Fe 2+ SOD) and demonstrate that the pH dependence reflects inhibition of anion binding by ionized Tyr34, not loss of a positive contribution on the part of Tyr34's labile proton. This can also explain the pH dependence of the K M of Fe 2+ SOD. Thus, it appears that substrate binding to Fe 2+ SOD occurs in the second sphere and is not strongly coupled to hydrogen bond donation. Parallel investigations of substrate analogue binding to the oxidized state (Fe 3+ SOD) confirm formation of a six-coordinate complex and resolve the apparent conflict with earlier nuclear magnetic relaxation dispersion (NMRD) results. Thus, we propose that two F -ions can bind to the oxidized Fe 3+ SOD active site, either displacing the coordinated solvent or lowering its exchange rate with bulk solvent. We show that neutral Tyr34's unfavorable effect on binding of the substrate analogue N 3 -can be ascribed to steric interference, as it does not apply to the smaller substrate analogues F -and OH -. Finally, we report the first demonstration that HS -can act as a substrate analogue with regard both to redox reactivity with FeSOD and to ability to coordinate to the active site Fe 3+ . Indeed, it forms a novel green complex. Thus, we have begun to evaluate the relative importance of different contributions that Tyr34 may make to substrate binding, and we have identified a novel, redox active substrate analogue that offers new possibilities for elucidating the mechanism of FeSOD.