Electron-transfer reactions between flavodoxin semiquinone and c-type cytochromes: comparisons between various flavodoxins

Abstract
As an extension of previous work from this laboratory using Clostridium pasteurianum flavodoxin [Tollin, G., Cheddar, G., Watkins, J. A., Meyer, T. E., and Cusanovich, M. A. (1984) Biochemistry 23, 6345-6349], we have measured the rate constants as a function of ionic strength for electron transfer from the semiquinones of Clostridium MP, Anacystis nidulans, and Azotobacter vinelandii flavodoxins to the following oxidants: cytochrome c from tuna and horse, Paracoccus denitrificans cytochrome c2, Pseudomonas aeruginosa cytochrome c-551, and ferricyanide. The rate constants extrapolated to infinite ionic strength (kINF) for the C. MP flavodoxin are all slightly smaller than for the C. pasteurianum flavodoxin, as would be predicted on the basis of the higher redox potential of the C. MP protein. This indicates that there is a close similarity between the surface topographies of the two proteins in the vicinity of the coenzyme binding site. Moreover, the electrostatic interactions between the two flavodoxins and the various oxidants are also approxmiately the same. These studies justify our previous use of the crystallographic structure of the C. MP flavodoxin to interpret kinetic results obtained with the structurally uncharacterized C. pasteurianum flavodoxin. Despite their lower redox potentials, both Anacystis and Azotobacter flavodoxins are appreciably less reactive toward all of these oxidants (as much as 2 orders of magnitude in some cases) than are the Clostridium flavodoxins. This consistent with crystallographic evidence, which indicates that the accesssibility of the flavin mononucleotide (FMN) prosthetic group of Anacystis flavodoxin is significantly smaller than that of C. MP flavodoxin and suggests that a similar situation exists for Azotobacter flavodoxin as well, for which the crystal structure is not known. Whereas tuna and horse cytochromes are less reactive than are Paracoccus and Pseudomonas cytochromes toward three of the flavodoxin semiquinones, they are more reactive with Azotobacter flavodoxin. This indicates that there are appreciable differences in the surface topographies between the Anacystis and the Azotobacter flavodoxins. The directions and approximate magnitudes of the ionic strength dependencies of the Anacystis and Azotobacter flavodoxin reaction rate constants are the same as for those of the Clostridium flavodoxins for all of the oxidants except Pseudomonas cytochrome c-551, for which the directions are opposite in sign. This demonstrates that the Pseudomonas cytochrome which has a relatively weak positive electrostatic potential in the vicinity of the heme prosthetic group, interacts differently with these two groups of flavodoxins, most likely as a consequence of the differences in the orientations of the FMN cofactor and in the steric properties of the flavodoxin surfaces. Computer modeling experiments are in agreement with this. These studies support our previous contention that the difference in redox potential between reacting electron-transfer proteins, the location and magnitude of electrostatic potential on the protein surfaces, the relative exposure of the prosthetic groups, and surface topography are all involved in controlling the specificity of a particular protein for its reaction partner. Optimum reactivity is achieved when the reactant proteins have regions of opposite electrostatic potential adjacent to the site of electron transfer. Furthermore, this work provides evidence that relatively small differences in surface topography can produce large effects on the kinetics of electron transfer by altering the mutual orientations of the two proteins within an intermediate complex.