The reactions of horse heart cytochrome c with Fe(ethylenediaminetetraacetate)r, Co(1, It has been suggested previously that electron transfer between horse heart cytochrome c and certain inorganic redox agents takes place by an outer-sphere mechanism at the heme edge that is partially exposed at the protein surface (1-3). In a particularly striking example, the calculated protein self-exchange rate constant (k1) based on oxidation of ferrocytochrome c by Co(phen)33+ (phen = 1,10-phenanthroline) has been shown to be in good agreement with the measured value, suggesting that the crossreaction involves a very similar mechanism (2). It should not necessarily be expected, however, that all reagents will have equal access to the heme edge in solution, as examination of structural models reveals that hydrophobic residues surround it. Indeed, the degree of access to the heme edge needs to be determined for a variety of substrates of various sizes, charges, and surface properties.The purpose of this paper is to present a detailed analysis of the kinetics of the electron transfer reactions of cytochrome c with Fe(CN)O3, Ru(NH3)(j+, Co(phen)33+, and Fe(EDTA)2-(EDTA = ethylenediaminetetraacetate). We shall employ the Marcus theory of outer-sphere electron transfer reactions to compensate for the variation in driving force and inherent reactivity of the reagents. Special attention will be directed to the evaluation of the electrostatic interactions between the protein and each reagent, thereby allowing an estimate to be made of the magnitudes of nonelectrostatic contributions to the activation free energies for the reactions.
THEORYThe Marcus theory correlates the crossreaction rate constant (k12) with the electron exchange rate constants for the two reactants (kil and k22) and the equilibrium constant (K) through the expressions k12 = (klk22Kf)/2 [1] phen, 1,10-log f = (log K)2/[4 log (kilk22/Z2)] [2] where the factor f is quite near 1 for the reactions to be considered here because they have rather small equilibrium constants (Z is the collision frequency) (4). With the inclusion of adiabaticity factors p, Eq. 1 becomes k12 = pI2(k11k22Kf/ p1Ip22)/2 (4). For the purposes of this treatment, it will be assumed that the reactions are adiabatic (Plu = P22 = P12 = 1), or at least uniformly nonadiabatic (p122 = PlIP22). Eq. 1 may.be applied as written, and the predicted kl2 value may be calculated from known values of ki1, k22, and K. In this approach, any deviation of the calculated from the observed values can be attributed to either the protein's or the reagent's undergoing a different activation process than in the self-exchange reaction, or to interaction energies between the reagent and protein which are not cancelled by the interactions in the exchange processes. Alternatively, it may be assumed that the activation process for reagent electron transfer, characterized by the exchange rate k22, is approximately a constant. Under this assumption Eq. 1 can be solved for kI, (the subscript 1 refers to the protein in ...