- Computersimulation (1) (remove)
- Proton Transfer Networks and the Mechanism of Long Range Proton Transfer in Proteins (2010)
- The main energy providing reaction systems in living cells, for example the photosynthesis or the respiratory chain, are based on long range proton transfer (LRPT) reactions. Even since these LRPT reactions have been heavily investigated in the last decades, the mechanism of these reactions is still not completely understood. The reaction kinetics of the LRPT are under heavy discussion and it is not clear, whether the reorientation of the hydrogen bond network (HBN)or the electrostatic barrier for the charge transfer is rate limiting. The main purpose of this work is to investigate the dynamics of chemical reactions inside of proteins, focused on long range proton transfer reactions. Electron transfer reactions, rotations of water molecules or conformational changes of the protein are also considered. The developed sequential dynamical Monte Carlo (SDMC) method is applicable to almost all kinds of chemical reactions. For all proton transfer reactions, the HBN of a protein plays a major role. Protons are transferred along such hydrogen bonds. Therefore, knowledge about the hydrogen bond network of a protein is crucial for the simulation of LRPT systems. The HBN can be calculated from the protein structure and the rotational state of the amino acid side chains. The reaction rate can be calculated from the electrostatic energies of the participating proton donor and acceptor groups. These two criteria are combined for the decision if a proton transfer between two molecules is possible and how fast this transfer would happen. While the calculation of electrostatic energies of protonatable amino acid side chains or relevant cofactors in proteins (among them also water molecules) is already solved - implemented in various programs - the remaining tasks - calculating the hydrogen bond network followed by calculating the reaction rates - were solved during this work. Before the hydrogen bond network and the electrostatic energies could be calculated, the lack of water positions in many available crystallographically resolved protein structures made it necessary to develop an algorithm to detect internal cavities in proteins and fill these cavities with water molecules. The derived water positions could be included in the electrostatic calculations as well as in the calculation of the HBN. The simulation of the LRPT in Gramicidin A (gA) compared to experimental data of the proton transfer in this polypeptide showed the possibilities of the simulation of the LRPT by the SDMC algorithm. The promising results encouraged us to investigate the mechanism of the LRPT, especially, if the reorientation of the HBN or the electrostatic energy barrier of the charge transfer is rate limiting for the LRPT. The results indicate, that both effects influence the LRPT and none of them is exclusively responsible for the LRPT rate. Further analysis of the hydrogen bond network topology showed that graph algorithms can be used to analyze these networks. Hydrogen bond networks can be clustered into regions which are close connected to each other. On the other hand, residues connecting two or more of these densely connected regions might play an important role for proton transfer pathways since a loss of such residues cuts a proton transfer pathway. A comparison of an analysis of the HBN topology of the photosynthetic reaction center with mutation studies of the same system showed, that residues identified as important for proton transfer by the mutation studies are identified as connection points between clusters by the network analysis. The developed algorithms together with the introduction of a new method for the simulation of the LRPT process (SDMC) improved the picture of the proton transfer processes in proteins. Starting from the protein structure, the developed algorithms cover all steps from the detection of protein cavities, the placement of water molecules in these cavities, the calculation and analysis of the hydrogen bond network, the simulation of the LRPT and the investigation of the reaction kinetics. The analysis of the HBN by graph theoretical methods gives further insight into the HBN topology and identifies residues important for proton transfer pathways and therefore important for the protein activity.