The Protein’s Role in Substrate Positioning and Reactivity for Biosynthetic Enzyme Complexes: the Case of SyrB2/SyrB1

Abstract

Biosynthetic enzyme complexes selectively catalyze challenging chemical transformations, including alkane functionalization (e.g., halogenation of threonine, Thr, by non-heme iron SyrB2). However, the role of complex formation in enabling reactivity and guiding selectivity is poorly understood, owing to the challenges associated with obtaining detailed structural information of the dynamically associating protein complexes. Combining over 10 ms of classical molecular dynamics of SyrB2 and the acyl carrier protein SyrB1 with large-scale QM/MM simulation, we investigate the substrate–protein and protein–protein dynamics that give rise to experimentally observed substrate positioning and reactivity trends. We confirm the presence of a hypothesized substrate-delivery channel in SyrB2 through free energy simulations that show channel opening with a low free energy barrier. We identify stabilizing interactions at the SyrB2/SyrB1 interface that are compatible with phosphopantatheine (PPant) delivery of substrate to SyrB2. By sampling metal–substrate distances observed in experimental spectroscopy of native SyrB2/SyrB1-PPant-<i>S</i>-Thr and non-native substrates, we characterize essential protein–substrate interactions that are responsible for substrate positioning, and thus, reactivity. We observe the hydroxyl sidechain and terminal amine of the native Thr substrate to form cooperative hydrogen bonds with a single N123 residue in SyrB2. In comparison, non-native substrates that lack the hydroxyl interact more flexibly with the protein and therefore can orient closer to the Fe center, explaining their preferential hydroxylation and higher turnover frequencies.

Evaluating Unexpectedly Short Non-covalent Distances in X-ray Crystal Structures of Proteins with Electronic Structure Analysis

“…5a. Although MM force fields predict both the double HB and standard O-type interactions to be favorable, the distinction between the two orientations is underestimated, consistent with our recent observations of more favorable double HB configurations in QM over MM simulation 102 . The weaker N-type CC energetics arise due to reduced electrostatic and induction stabilization with comparable exchange repulsion penalties (Supporting Information Table S9).…”

Both Configuration and QM Region Size Matter: Zinc Stability in QM/MM Models of DNA Methyltransferase

“…[30][31] When small QM regions are used in QM/MM simulation, they are sometimes adequate to predict physically reasonable mechanisms 32 , but in other cases are missing critical residues needed to describe the essential enzyme action [33][34] , and distinguishing these two scenarios is challenging. Despite the many successes of force fields essential for describing globular protein structure and dynamics, qualitatively incorrect descriptions in other cases (e.g., disordered structures [35][36] and dynamic non-covalent interactions 37 ) can limit the ability of predominantly MM-based models in small-QM region QM/MM simulation to reveal enzyme mechanism. Recent advances 30,35,[38][39][40][41][42][43][44] in algorithmic efficiency and hardware have enabled quantum chemical simulation of polypeptides 35,45 and large-scale QM (ca.…”

“…Metalloproteins present unique challenges for modeling 37,53,59,68 due to the unique trade-off between high levels of theory needed to adequately describe the metal-organic bond and the need to establish the dependence of properties on adequate sampling of variations in the greater protein environment. For this study, we select DNA methyltransferase as a prototypical metalloprotein in which Zn 2+ plays an essential structural role [76][77][78] .…”

Protein Dynamics and Substrate Protonation State Mediate the Catalytic Action of Trans-4-Hydroxy-L-Proline Dehydratase

“…To account for the sensitivity of these results to potential inaccuracies in the force field parameters, 32,35,60 we also optimized active-site structures for the cysteine radical using QM cluster calculations. These large-scale QM calculations incorporate Hyp ZW , one conserved water molecule, and 14 nearby HypD residues that were selected based on their proximity to the substrate and proposed roles 33 from mutagenesis experiments (Supporting Information Table S1 and Figure S9).…”