Thermodynamic integration network study of electron transfer: from proteins to aggregates

Abstract

We describe electron transfer through the NrfHA nitrite reductase heterodimer using a thermodynamic integration scheme based upon molecular dynamics simulations. From the simulation data, we estimate two of the characteristic energies of electron transfer, the thermodynamic driving forces, ΔG, and the reorganization energies, λ. Using a thermodynamic network analysis, the statistical accuracy of the ΔG values can be enhanced significantly. Although the reaction free energies and activation barriers are hardly affected by protein aggregation, the complete reaction mechanism only emerges from the simulations of the dimer rather than focussing on the individual protein chains: it involves an equienergetic transprotein element of electron storage and conductivity.

Thermodynamic integration network approach to ion transport through protein channels: Perspectives and limits

“…Recently, we have presented that the thermodynamic network analysis based on the least squares problem has noticeably improved the statistical accuracy of the simulated free energies gained with TI methods . Although, the focus of our recent studies was lying on electron transfer reactions.…”

“…The Figure shows representative examples of ∂V (Λ)/∂Λ‐plots obtained from the TI‐calculations: the transport reaction of the cesium ion from the center to the mouth of the DH dimer, the alchemical transformation reaction from Cs + to K + at the center of the H2H dimer and the solvation of sodium ion (the direction of the simulated reaction is from water to vacuum). In contrast to our investigations of electron transfer, the data obtained from reactions involving ions cannot be described by a linear function. The shape of all ∂V (Λ)/∂Λ‐ plots is roughly hyperbolic.…”

“…Our main perspective is a methodological one, we are interested in the question to which degree the thermodynamic integration approach can be applied to systems in which the thermodynamic states correspond to transient centers of alkali metal cation localization, that is, particles that show a high degree of mobility. We are motivated by a recent extension of the TI methodology in the context of electron transfer, the construction of thermodynamic integration networks . They enable a least‐squares calculation within a net of free energy computations, and the estimate of a corresponding error.…”

Monte Carlo simulation and thermodynamic integration applied to protein charge transfer

“…Scaled, the driving forces for biologically relevant charge transfer reactions range up to 13 kcal mol −1 , the range is similar to that observed for molecular dynamics simulations or dielectric theory. Again, we can sort sites of electron localization into a high and low energy class . We find most of the H subunit hemes to be low‐energy sites (H2, H3, H4), with only the entry site, H1, showing a significantly higher energy of about 13 kcal mol −1 .…”

“…For NrfHA, we will only briefly comment on the protein geometry and the preparation of the Monte Carlo simulation input, as they are very similar to the molecular dynamics simulations and the Poisson‐Boltzmann approach recently published. The protein geometries are based on the X‐ray structure of NrfH 2 A 4 from Desulfovibrio vulgaris by Rodrigues et al, from which we have selected the structure in the absence of an inhibitor (protein data bank ID 2J7A).…”