| Electron transfer in a protein is a complicated process to model as it takes place in a strongly disordered, insulating environment, yet each protein has a very specific structure determined by evolution. We have developed a simple model to calculate protein electron transfer rates that examines the structure to a first level of approximation, with results matching the data to within the data's error bars, .8 log units. We then used that model to analyze the crystal structures of 31 natural electron transfer proteins to statistically determine what factors (angular orientation, thermodynamic properties, protein density, distance, secondary structure) have been utilized by natural selection to obtain sufficiently fast electron transfer. The edge-to-edge distances between redox centers are 14 A or less, due to the strong exponential dependency of rate on distance. (A 1.6 A shift gives an order of magnitude change in rate.) Longer distances are crossed by chains of redox centers with individual distances of 14 A or less. None of the other factors had statistically significant patterns indicating they were optimized by evolution to modulate electron transfer rates.;This makes setting the distance not just necessary but sufficient for the design of electron transfer systems, removing the need for specific pathways or detailed optimized structures. For redox chains, short distances can compensate for endergonic steps, removing the need for superexchange. At very short distances, the high-energy radical states of substrates can be reached without any special mechanisms, allowing the creation of catalytic sites just through the use of proximity. All this makes for a robust electron transfer engineering that is less likely to be upset by random evolutionary mutations. These concepts are also applicable for those trying to understand how natural proteins work, or trying to build artificial electron transfer proteins. |
PDF Full Text Request