Towards Accurate Computations of Cofactor-Containing Biosystems
Suliman Adam · Universitätsbibliothek der FU Berlin Hochschulschriftenstelle u. Dokumentenserver · 2018
In the cell, cofactor proteins take on a wide range of tasks. Photosystem II, for example, generates oxygen during photosynthesis and channelrhodopsin from the alga Chlamydomonas reinhardtii makes the organism move in response to light. For these proteins to function properly, the presence of their cofactor molecules is indispensable. Theoretical attempts to studying cofactor- containing systems like photosystem II might struggle with the lack of reliable force field parameters. For a retinal cofactor, on the other hand, quantum mechanical description of its electrostatic interactions is necessary to correctly characterize its behaviour inside of the protein. Here, we used two different theoretical approaches to study cofactors of two proteins: classical mechanics and force field optimization to describe the iron- containing cofactors of photosystem II, and combined quantum and classical mechanics to analyse the environmental effects on proton transfer energetics in channelrhodopsin. We successfully derived new CHARMM force field parameters for the hæm and non-hæm iron complexes of photosystem II. By comparing to quantum mechanical data and by using test simulations, we showed that these new parameters provide a greatly improved description of intramolecular and non-bonded interactions. The parameters presented here will facilitate reliable all-atom simulations of proteins that contain hæm and non-hæm iron complexes. In channelrhodopsin, deprotonation of the Schiff base only occurs after the light-induced isomerization of the retinal. We used the Weighted Histogram Approach Method to generate proton transfer pathways to understand what prevents retinal deprotonation in the dark state of the protein and identified three important determinants that ensure the chromophore’s continued protonation. These determinants were the presence of a positively charged lysine (K132), water molecules in the active site and the protonation state of the Schiff base counterion glutamate-162. Ultimately, we demonstrated that a combined quantum and classical mechanical approach can be applied to proteins embedded in a hydrated lipid membrane, and we created a protocol that can be transferred to other proteins to study their proton transfer pathways and energetics while taking the protein and lipid environment into account.