Unveiling polyphenol-protein interactions: a comprehensive computational analysis

Abstract

Our study investigates polyphenol-protein interactions, analyzing their structural diversity and dynamic behavior. Analysis of the entire Protein Data Bank reveals diverse polyphenolic structures, engaging in various noncovalent interactions with proteins. Interactions observed across crystal structures among diverse polyphenolic classes reveal similarities, underscoring consistent patterns across a spectrum of structural motifs. On the other hand, molecular dynamics (MD) simulations of polyphenol-protein complexes unveil dynamic binding patterns, highlighting the influx of water molecules into the binding site and underscoring limitations of static crystal structures. Water-mediated interactions emerge as crucial in polyphenol-protein binding, leading to variable binding patterns observed in MD simulations. Comparison of high- and low-resolution crystal structures as starting points for MD simulations demonstrates their robustness, exhibiting consistent dynamics regardless of the quality of the initial structural data. Additionally, the impact of glycosylation on polyphenol binding is explored, revealing its role in modulating interactions with proteins. In contrast to synthetic drugs, polyphenol binding seems to exhibit heightened flexibility, driven by dynamic water-mediated interactions, which may also facilitate their promiscuous binding. Comprehensive dynamic studies are, therefore essential to understand polyphenol-protein recognition mechanisms. Overall, our study provides novel insights into polyphenol-protein interactions, informing future research for harnessing polyphenolic therapeutic potential through rational drug design.Scientific contribution: In this study, we present an analysis of (natural) polyphenol-protein binding conformations, leveraging the entirety of the Protein Data Bank structural data on polyphenols, while extending the binding conformation sampling through molecular dynamics simulations. For the first time, we introduce experimentally supported large-scale systematization of polyphenol binding patterns. Moreover, our insight into the significance of explicit water molecules and hydrogen-bond bridging rationalizes the polyphenol promiscuity paradigm, advocating for a deeper understanding of polyphenol recognition mechanisms crucial for informed natural compound-based drug design.

Keywords: Dynamic behavior; Glycosylation; Molecular dynamics simulations; Noncovalent interactions; Polyphenol-protein complexes; Polyphenols; Water-mediated interactions.

Fig. 1

Polyphenols form water-mediated H-bonds in static as well as dynamic structures. a A prototypical snapshot from an MD simulation where rosmarinic acid (ROA, carbons denoted as dark-green sticks) binds to the factor X enzyme [16]. We observe several water-mediated H-bond bridges (red lines) within the active binding site that stabilize ROA binding. b The crystal structure (PDB ID: 7B3E) of flavonoid myricetin (MYC, carbons denoted as light-green sticks) covalently bound to the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease. MYC forms several water-mediated H-bond bridges (red lines) within the binding site. MYC also forms direct H-bonds, which are shown as purple lines

Fig. 2

Our classification of polyphenols known to bind to proteins sourced from the PDB, with each class represented by a prototypical compound

Fig. 3

The main properties of the constructed protein-polyphenol database. a Classification of polyphenol-protein complexes into biological kingdoms. b The most represented polyphenols within the database. c Number of polyphenols in each class. d Classification of proteins containing polyphenolic ligands. e Distribution of polyphenol-protein complexes based on the resolution of the solved crystal structures. In all cases, alternative conformations are not counted separately

Fig. 4

Radial pair distribution functions (RDFs) illustrating the spatial relationships between selected pairs of protein and polyphenolic atoms. The RDFs depict the distance distributions between: a O.3$_{\text{sc}}^{}$ and O.3, b O.co2$_{\text{sc}}^{}$ and O.3, c N.ar$_{\text{sc}}^{}$ and O3, d N.pl3$_{\text{sc}}^{}$ and O.3, e O.3$_{\text{w}}^{}$ and O3, f N.pl3$_{\text{sc}}^{}$ and O.co2, g C.ar$_{\text{sc}}^{}$ and Car, h O.2$_{\text{bk}}^{}$ and O.3, (i) metal ions (M) and C.ar, and (j) metal ions (M) and O.3 atom types. In all cases, the left-hand atom-type corresponds to a protein atom, while the right-hand atom corresponds to a polyphenolic atom. Each pair is present more than 1000 times, except for the M - O.3 pair, which is present 575 times. RDFs of other atom pairs that are present more than 1000 times are displayed in Fig. S1

Fig. 5

Representative noncovalent interactions commonly formed by standard polyphenolic compounds

Fig. 6

Overview of the average interaction distances and their standard deviations based on polyphenol classes. PA phenolic acids, F flavonoids, HAD hydroxycinnamic acid derivatives, HB hydroxybenzenes, S stilbenes, C coumarins

Fig. 7

Relative frequency distributions of non-covalent interactions identified with PLIP, categorized by polyphenol classes with high representation in the PDB

Fig. 8

Interaction contact maps between glutamate dehydrogenase residues and XEG for the main a and replica b simulations, depicting the presence of hydrophobic (yellow), hydrogen bonds (green), and rare π-cation (purple) interactions. c–k Dynamic binding of epicatechin-3-gallate (XEG) to glutamate dehydrogenase (GDH). Analysis of the water-mediated binding for the first cluster (snapshots 0-598) simulation, showing the c entire medoid (409) snapshot, d zoomed-in binding site, and e the Bridge2 output of water-mediated H-bonding interactions. f–h panels corresponding to the second cluster (snapshots 599-745) and i–k panels corresponding to the third cluster (snapshots 746-1000). Blue cartoons in panels represent the backbone of GDH and sticks with grey carbons the amino-acid residues forming H-bonding interactions (cyan lines) with XEG. Sticks with green carbons represent the XEG ligand.The numbers on the edges represent the average number of water molecules bridging the H-bonding interaction

Fig. 9

Water-mediated H-bonds were observed during the MD simulations of a TTR-resveratrol structure. a, b The lower-resolution crystal structure (1DVS) contains two water molecules within the binding site. One forms a water-mediated H-bond to Glu54A. c, d During MD simulations, an extensive water-mediated H-bond network was formed within the binding site, including residues from chains A and C. A frame from a 930$_{}^{\mathit{th}}$ ns is chosen for visualization. e Bridge2 output of the resveratrol water-mediated H-bonding. Values on the edges represent the average number of bridging water molecules during the main MD simulation of the low resolution structure. Corresponding figures from remaining simulations are deposited in Supporting Information Fig. S11i-k. Blue cartoons represent the backbone of H-bonding amino-acid residues, and sticks with grey carbons their side-chains. Resveratrol (STL) is presented by sticks of green carbons, and waters by balls-and-sticks representation (red oxygens, white hydrogens). Direct H-bonds are shown with purple dotted lines, and water mediated ones with cyan dotted lines. f 7Q9O in yellow-colored cartoon model with green stick model ligand is overlaid with 2fo-fc electron density map in light-blue mesh. Crystal-modelled waters that were fitted to the electron density are emphasized by dark-blue spheres. Our MD snapshot is superposed (rose-colored cartoon model with light-pink stick model ligand) with MD TIP3 waters of the inspected snapshot in element-colored stick model (red oxygen and white hydrogens). It can clearly be observed that all modelled crystal water locations are also occupied by MD TIP3 waters

Fig. 10

Water molecules mediate a large number of interactions between flavonoids and SIRT6 as observed during MD simulations. a Water-mediated H-bonds formed with quercetin (QUE) in the main MD simulation. b Water-mediated H-bonds formed with QUE in the replica MD simulation. c Water-mediated H-bonds formed with isoquercetin (HW2) in the main MD simulation. d Water-mediated H-bonds formed with HW2 in the replica MD simulation. e, g 3D view of the formation of a frequent (occupancy $\ge$ 50%) water-mediated H-bond network involving the sugar moiety of HW2 as observed in the main MD simulation. f, h 3D view of the formation of a frequent (occupancy $\ge$ 50%) water-mediated H-bond network involving the sugar moiety of HW2 as observed in the replica MD simulation