This work is a theoretical analysis in four stages of association between the blue copper protein plastocyanin and the heme protein cytochrome f, which are physiological partners in the photosynthetic electron-transfer chain. In the first stage, 32 000 trajectories of approach by plastocyanin to cytochrome f were generated with implicit consideration of hydration and with gradual cooling of the system from 300 to 0 K. Approximately 2000 trajectories resulted in local minima of energy, i.e., in docking. The molecular configurations having relatively low energies were grouped, by structural similarity, into six families. In the second stage, six configurations having the lowest energies, one from each family, were subjected to thorough molecular dynamics simulation, for 260 ps. Extensive hydration of the proteins was treated explicitly. The whole plastocyanin molecule and the relevant parts of the cytochrome f molecule were given conformational freedom. In the third stage, the following three contributions to the energy of binding were calculated:  polarization of water by the proteins, determined from numerical solutions of the Poisson−Boltzmann equation; nonelectrostatic (van der Waals and other) interactions involving the proteins and water; and the Coulombic interactions within and between the protein molecules. Total energy of association was calculated with a thermodynamic cycle; several realistic sets of parameters gave consistent results. The configuration having the most favorable Coulombic interactions turned out to have the second highest total energy. This finding exemplifies the importance of allowing for hydration and for conformational flexibility in docking calculations and perils of neglecting these factors. In the fourth stage, electronic coupling between the copper and heme sites in the six configurations was analyzed and compared by the Pathways method. The configuration providing the most efficient path for electron tunneling turned out to be different from the most stable configuration. There are indications that the evident interaction between Lys65 in cytochrome f and Tyr83 in plastocyanin may involve the ammonium group of the former and the aromatic ring of the latter. These surprisingly strong noncovalent interactions, so-called charge-π interactions, have recently been discovered and are important for molecular recognition. Modeling and structural optimization of these interactions are beyond the state of the art in molecular mechanics, but these studies should become possible with improved force fields. The electron-transfer reaction between cupriplastocyanin and ferrocytochrome f is fast in the noncovalent complex and undetectably slow in the covalent complex. This contrast is explained in terms of our theoretical analysis.

Silica hydrogel (glass) was doped with native (iron-containing) cytochrome c and with its zinc derivative. Ultraviolet−visible, circular dichroism, and resonance Raman spectra of both proteins and the lifetime of the triplet state of the zinc protein show that encapsulation in the sol-gel glass only slightly perturbs the polypeptide backbone and does not detectably perturb the heme group. Because thermal (ground-state) redox reactions of the encapsulated native cytochrome c are very slow, we take advantage of the transparency of the silica to study, by laser flash spectrometry, photoinduced (excited-state) redox reactions of zinc cytochrome c, which occur in milliseconds. The triplet state, 3Zncyt, is oxidatively quenched by [Fe(CN)6]3-, dioxygen, and p-benzoquinone. These reactions are monophasic in bulk solutions but biphasic in solutions confined in glass. Changes in ionic strength and pH differently affect the kinetics in these two environments. Adsorption of cytochrome c, which is positively charged, to the pore walls, which are negatively charged at pH 7.0, affects the kinetics in the doped glass. Exclusion of the [Fe(CN)6]3- anions from the glass interior also affects the kinetics. Even at equilibrium the anion concentration is lower inside the glass than in the external solution. This exclusion can be lessened or eliminated by raising ionic strength and lowering the pH value. The electroneutral quenchers are not excluded from the glass. Diffusion of all three quenchers is slower in the confined solution than in the bulk solution, as expected. The smaller the molecule, the lesser this hindrance by the glass matrix. In light of these findings, the assumption that porosity of sol-gel glasses ensures uniform penetration of relatively small molecules into the pores must be taken skeptically and tested for each solute (or analyte) of interest, especially for the charged ones. These considerations are important in the design of sensors.

No abstract

No abstract