Hydroxylamine oxidoreductase (HAO) of Nitrosomonas europaea catalyzes the four-electron oxidation of NH2OH to NO2-. Each subunit of the trimeric enzyme contains seven c-hemes and one heme P460. In previous work [Hendrich, M. P., et al. (1994) J. Am. Chem. Soc. 116, 11961-11968], an integer-spin EPR signal at g = 7.7 was discovered from the active site of the resting enzyme. This new signal was assigned to an exchange-coupled cluster containing ferric heme P460 and a ferric c-heme. An electrochemical titration of HAO is presented here in which EPR signals and optical bands, believed to be associated with the P460 heme, are monitored. In the EPR titration, as a redox center with Em8 = -140 mV becomes reduced, the integer-spin signal disappears. Then, upon reduction of a redox center with Em8 = -190 mV, a g = 6 type signal, which has been previously assigned to a high-spin form of the ferric P460 heme of HAO, appears. However, in the -140 to -190 mV range, we have been unable to identify an additional EPR signal attributable to the P460 center. Thus, the electronic environment of oxidized P460 heme of HAO appears to pass through three states before reduction in a titration experiment, with an intermediate state that is not readily detectable by X-band EPR. The best candidate for the c-heme partner of the P460 heme is the heme at -190 mV, which would correspond to heme 6 of the crystal structure. A possible function of the exchange-coupled heme cluster is to facilitate two-electron oxidation steps of the substrate. An earlier spectropotentiometric titration of HAO [Collins, M. J., et al. (1993) J. Biol. Chem. 268, 14655-14662] identified a broad, weak optical band, centered near 740 nm, that was tentatively assigned to the oxidized P460 heme. This assignment has been strengthened by additional spectropotentiometric titrations at several values of pH and also by rapid kinetic experiments following the reduction of HAO by dithionite. The 740 nm band is not observed in fully oxidized HAO. In the spectropotentiometric titrations, its appearance cannot be correlated with reduction of a specific c-heme nor modeled to a Nernstian one-electron redox center. Instead, the range of potential in which the 740 nm band is present depends on whether the titration is carried out in an oxidative or reductive direction. One possible interpretation is that the 740 nm band is a property of the oxidized high-spin P460 heme but not of the low-spin state, and that the transition between the two spin states occurs at different potentials depending on the direction of the electrochemical titration.