Evidence for Nonequivalent Binding Sites in Human Methemoglobin*

On the basis of nuclear magnetic resonance spectra, Davis, Charachc, and Ho (1) have recently suggested that the heme groups of the a! and /3 methemoglobin chains are not equivalent. Gibson, Parkhurst, and Geraci (2) have reported biphasic kinetics for the reaction of methemoglobin with a variety of iron-coordinating ligands. These results were also explained assuming nonequivalent heme groups. In contrast, Anusiem, Beetlestone, and Irvine (3) ; Beetlestone and Irvine (4) ; and Beetlestonc, Epega, and Irvine (5) have reported hyperbolic curves for the binding to methemoglobin of all the ligands they have investigated. This latter observation suggests that the four heme irons bind ligands with identical, intrinsic affinities. There is, therefore, an apparent contradiction between the thermodynamic results, and the kinetic and nuclear magnetic resonance observations. Because of the large difference in time scales, it is possible that the binding sites may appear equivalent by one technique, but not by another. On the other hand, if the difference in binding energies is small, nonequivalent binding would be difficult to detect. To test for this latter possibility, we undertook a careful analysis of the methemoglobin binding curves. We have found that the binding of imidazole to human methemoglobin A is best interpreted on the basis of two nonequivalent binding sites.