The solution structures of the reduced and oxidized forms of the cytochrome c are used to reevaluate the reorganization energy for oxidation of cytochrome c. This is achieved by using the linear response approximation in concert with the NMR structures as pseudo energy constraints. Alternative estimates, obtained using a free energy perturbation approach employing umbrella sampling and a continuum dielectric approach, are also provided. The reorganization energy obtained is larger than that previously estimated using crystal structures of the protein. Nevertheless, the present estimate remains significantly smaller than the corresponding reorganization energy in water (9−15 kcal mol-1 as compared to ≈37 kcal mol-1 in water) and the protein contribution to the reorganization energy is only 8−10 kcal mol-1. This provides further support for the proposal that proteins assist in electron transfer reactions by reducing the relevant reorganization energies. The solution structures are also used to estimate the redox potential of cytochrome c. Several strategies are employed including a newly formulated scaled linear response approximation. The calculations agree reasonably well with the observed redox potential. Analysis of the group contributions to the reorganization energy and redox potential reveals a clear energetic linkage between these fundamental parameters of electron transfer and a redox-dependent surface feature likely to influence recognition of cytochrome c by its redox partners. Specifically, the rearrangement of Ile81 and other residues at the heme edge upon a change in oxidation state gives rise to a large contribution to both the redox potential and the reorganization energy. Finally this work is used to explore and illustrate the meaning of macroscopic dielectric models. It is shown that the “proper” dielectric constant depends strongly on the model used since it basically represents the implicit contributions of the given model rather than a fundamental physics. Thus we obtain different effective dielectric constants for different treatments of redox potential and reorganization energy.

No abstract

No abstract

A model for the solution structure of horse heart ferricytochrome c has been determined by nuclear magnetic resonance spectroscopy combined with hybrid distance geometry-simulated annealing calculations. Forty-four highly refined structures were obtained using a total of 1671 distance constraints based on the observed magnitude of nuclear Overhauser effects and 58 torsion angle restrains based on the magnitude of determined J-coupling constants. The model incorporates six long-lived water molecules detected by pseudo-two-dimensional NOESY-TOCSY spectra. The all-residue root mean square deviation about the average structure is 0.33 +/- 0.04 A for the backbone N, C alpha, and C' atoms and 0.83 +/- 0.05 A for all heavy atoms. The overall topology of the model for solution structure is very similar to that seen in previously reported models for crystal structures of homologous c-type cytochromes though there are a number of significant differences in detailed aspects of the structure. Two of the three main helices display localized irregularities in helical hydrogen bonding resulting in bifurcation of main chain hydrogen bond acceptor carbonyls. The N- and C-terminal helices are tightly packed and display several interhelical interactions not seen in reported crystal models. To provide an independent measure of the accuracy of the model for the oxidized protein, the expected pseudocontact shifts induced by the spin 1/2 iron were compared to the observed redox-dependent chemical shift changes. These comparisons confirm the general accuracy of the model for the oxidized protein and its observed differences with the structure of the reduced protein. The structures of the reduced and oxidized states of the protein provide a template to explain a range of physical and biological data spanning the redox properties, folding, molecular recognition, and stability of the cytochrome c molecule. For example, a redox-dependent reorganization of surface residues at the heme edge can be directly related to the redox behavior of the protein and thereby provides a previously undocumented linkage between structural change potentially associated with molecular recognition of redox partners and the fundamental parameters governing electron transfer.

Submillisecond burst phase signals measured in kinetic protein folding experiments have been widely interpreted in terms of the fast formation of productive folding intermediates. Experimental comparisons with non-folding polypeptide chains show that, for ribonuclease A and cytochrome c, these signals in fact reflect a shift from one biased ensemble of the unfolded state to another as a function of change in denaturant concentration.

A model for the solution structure of horse heart ferrocytochrome c has been determined by nuclear magnetic resonance spectroscopy combined with hybrid distance geometry-simulated annealing calculations. Forty-four highly refined structures were obtained using a total of 1940 distance constraints based on the observed magnitude of nuclear Overhauser effects and 85 torsional angle restraints based on the magnitude of determined J-coupling constants. The all-residue root mean square deviation about the average structure is 0.47 +/- 0.09 A for the backbone N, C alpha, and C' atoms and 0.91 +/- 0.07 A for all heavy atoms. The overall topology of the model for solution structure is very similar to that seen in previously reported models for crystal structures of homologous c-type cytochromes. However, a detailed comparison between the model for the solution structure and the available model for the crystal structure of tuna ferrocytochrome c indicates significant differences in a number of secondary and tertiary structural features. For example, two of the three main helices display 3(10) to alpha-helical transitions resulting in bifurcation of main-chain hydrogen bond acceptor carbonyls. The N- and C-terminal helices are tightly packed and display several interhelical interactions not seen in previously reported models. The geometry of heme ligation is well-defined and completely consistent with the crystal structures of homologous cytochromes c as are the locations of four of six structural water molecules. Though the total solvent-accessible surface area of the protoporphyrin ring is similar to that seen in crystal studies of tuna ferrocytochrome c, the distribution is somewhat different. This is mainly due to a difference in packing of residues Phe-82 and Ile-81 such that Ile-81 crosses the edge of the heme in the solution structure. These and other observations help to explain a range of physical and biological data spanning the redox properties, folding, molecular recognition, and stability of the protein.

No abstract

He is a graduate student at the University of Pennsylvania. Phoebe X. Qi received a Ph.D. degree in biochemistry at the University of Illinois in Urbana. She is a postdoctoral researcher at the University of Pennsylvania.Yawen Bai received M.S. and Ph.D. degrees in chemistry and biophysics at the University of Pennsylvania and did postdoctoral work at the Scripps Research Institute. He is a staff scientist at the NIH.