Files
Abstract
Proton transport, the movement of hydrogen ions across membranes, is a pervasive biological process which appears in many contexts and is central to cellular energy transduction, hence life itself. Because it plays a central role in many enzymatic mechanisms, understanding what protein features cause defective or, conversely, facile proton transport can produce insights into how to treat diseases linked to those enzymes, as well as suppress or enhance their function. Proton transport processes in enzymes, though they enable such important events as the synthesis of ATP and co- or anti-transport of small molecules, are often not mechanistically well-understood. In particular, existing proton transport studies have often assumed that transport will only occur readily when coherent, stable hydration structures can be resolved. More recent work has cast doubt on this assumption and produced evidence of relatively facile proton transport even through only transiently hydrated structures, illuminating the need to account for the coupling of proton motion to local water structure when studying this important process. In this thesis, I present model parameters for reactive molecular dynamics simulations of glutamate, aspartate, and histidine, three titratable amino acids that often mediate proton transport in channels. I show that these models produce physically accurate results and are transferrable to aqueous environments despite being derived from gas-phase data. I then apply some of these models to a real system with important biological implications, mitochondrial respiratory Complex
I, a key component of the electron transport chain and one of a few proteins responsible for maintaining the mitochondrial proton gradient that powers ATP synthase. First, I present a study detailing the employment of the fitRMD parameterization procedure to produce a physically accurate reactive molecular dynamics model of aspartate, as well as improvements to the existing methodology that result in a more accurate glutamate model than previously existed. I am able to show that these models strongly agree with available data for both bulk and protein environments. Then, I present similar work for histidine, discussing the modeling differences from the carboxylates due to the two non-equivalent protonation sites of histidine, and validating the model produced against experimental data. Finally, I use some of these models in a real system, Complex I. This pathway is not well-understood, and the lateral proton transport between several critical titratable carboxylate residues has been hypothesized to enable delivery of the driving force produced by ubiquinone reduction elsewhere in the enzyme to each of the sites of proton translocation. This enzyme is a case study in the coupling of local hydration dynamics to proton transport energetics, as well as the more general phenomenon of proton transport being coupled to processes with much longer and larger time- and length-scales. I calculate the potential of mean force of proton transport with respect to local water connectivity and proton progress through the channel to shed light on how transport through this channel might interact with sequential transport events. These results represent a step forward toward more accurate and versatile models of reactive residues in protein proton transport processes.