Modeling the Role of a Flexible Loop and Active Site Side Chains in Hydride Transfer Catalyzed by Glycerol-3-Phosphate Dehydrogenase

Abstract

<div> <div> <div> <p>Glycerol-3-phosphate dehydrogenase is a biomedically important enzyme that plays a crucial role in lipid biosynthesis. It is activated by a ligand-gated conformational change that is necessary for the enzyme to reach a catalytically competent conformation capable of efficient transition state stabilization. While the human form (hlGPDH) has been the subject of extensive structural and biochemical studies, corresponding computational studies to support and extend the experimental observations have been lacking. We perform here detailed empirical valence bond and Hamiltonian replica exchange molecular dynamics simulations of wild-type hlGPDH and its variants, as well as providing a novel crystal structure of the binary hlGPDH·NAD R269A variant where the enzyme is present in the open conformation. We estimated the activation free energies for the hydride transfer reaction in wild-type and substituted variants of hlGPDH and investigated the effect of mutations on the catalysis from a detailed structural study. Our structural data and simulations also illustrate the critical role of the R269 side chain in facilitating the closure of hlGPDH into a catalytically competent conformation, through modulating the flexibility of a key catalytic loop (292-LNGQKL-297), thus rationalizing a tremendous 41,000-fold decrease experimentally in the turnover number, kcat, upon truncating this residue. Taken together, our data highlight the importance of this ligand-gated conformational change in catalysis, a feature that can be exploited both for protein engineering and for the design of novel allosteric inhibitors targeting this biomedically important enzyme.</p></div></div></div>

Complex Loop Dynamics Underpin Activity, Specificity and Evolvability in the (βα)8 Barrel Enzymes of Histidine and Tryptophan Biosynthesis

“…However, it is unclear if this is an effect on kcat, KM or both, and it is plausible that the loss of activity is due to structural effects that prohibit productive substrate binding, that are not captured in our simulations. 38 This latter issue would be similar to our observations from a recent study of an analogous system activated by a ligand-gated conformational change, glycerol-3-phosphate dehydrogenase (GPDH). 38 In this system a substantial loss of activity upon truncation of a key catalytic arginine to alanine could only be explained by structural rearrangements (predominantly blocking of the closure of a key catalytic loop), that were observed upon crystallization of this variant.…”

“…38 This latter issue would be similar to our observations from a recent study of an analogous system activated by a ligand-gated conformational change, glycerol-3-phosphate dehydrogenase (GPDH). 38 In this system a substantial loss of activity upon truncation of a key catalytic arginine to alanine could only be explained by structural rearrangements (predominantly blocking of the closure of a key catalytic loop), that were observed upon crystallization of this variant. In contrast, this loss of activity could not be captured simply by performing a truncation of this side chain on the wild-type enzyme and considering only electrostatic effects which only accounted for a smaller part of the observed change in activity.…”

“…We note the similarity of these gripper interactions to corresponding interactions in enzymes such as TIM, OMPDC and GPDH, [34][35][36][37][38] where interactions with the non-reactive phosphodianion group of the substrate drives a ligand-gated conformational change. This in turn stabilizes otherwise energetically unfavorable but catalytically important closed conformations of key catalytic loops over the respective active sites of these enzymes.…”

“…This group forms hydrogen bonding interactions with the side chains of R83 from loop 3 and S103 from loop 4 of SeHisA, and with the corresponding side chains of R85 and T105 in MtPriA (Figure 2). Although neither of these residues are on the primary mobile loops of either enzyme (Figure 1), nevertheless, the interactions with the second phosphate group of ProFAR are very similar to analogous interactions in other enzymes activated by ligand-gated conformational changes, [34][35][36][37][38][39][40][41] suggesting that loops 3 and 4 in SeHisA and MtPriA may similarly act as "gripper loops" allowing these enzymes to attain relevant catalytically active conformations for the isomerization of the larger substrate. This effect would clearly not be present when the smaller substrate, PRA, is bound to the active site as this substrate lacks a non-reactive phosphodianion group to interact with these loops.…”

“…In this context, there have been extensive studies of a wide-range of enzymes, such as triosephosphate isomerase (TPI), 34,35 orotidine 5'-monophosphate decarboxylase, (OMPDC) 36 glycerol 3-phosphate dehydrogenase (GPDH), 37,38 1-deoxy-D-xylulose-5-phosphate reductoisomerase, 39,40 and β-phosphoglucomutase, 41 which have been demonstrated to be activated by ligand-gated conformational changes. Specifically, interactions between a key "gripper" loop decorating the active site and the non-reactive phosphodianion groups of the substrates of these enzymes trigger substantial conformational changes in the gripper loop, facilitating energetically unfavorable transitions from catalytically inactive open to catalytically active closed conformations, and these conformational transitions are central to the catalytic activities and high proficiencies of these enzymes.…”