Introduction

Food is a complex and heterogenous system containing different major and minor constituents. The major macromolecular components include proteins, polysaccharides, and lipids while minor food components are organic acids, pigments, aroma compounds, vitamins, and minerals, among others, which also provide the essential nutrients (Samant et al. 1993). These constituents individually or as complexes play critical functional roles in foods and in different bioactivities. Most scientific literature has documented the nutrient composition of foods, identification of novel compounds and their functional properties, and bioactivities. In the past decades, most studies had focused on determining the content of bioactive compounds and bioactivity of vegetables, legumes, grains, fruits, herbs, and seafood. At present, the attention has been drawn to the fundamental understanding of the behavior of food macromolecules to interpret bioactivities at the molecular level.

In this contribution, an overview of the main interactions between phenolics and proteins in foods including their effects on biological activities and food properties are presented. In addition, particular emphasis is given to understanding and predicting these interactions and outcomes in terms of molecular docking simulations with model systems containing selected phenolic compounds and proteins. The insights obtained at the molecular level will provide a fundamental framework for in-vitro studies from nutritional, functional and pharmaceutical perspectives.

Proteins are one of the primary macronutrients and functional components of the diet that determine food’s textural, sensorial, and nutritional properties. According to the chemical structure, proteins are made up of 20 different amino acids that are linked together by peptide bonds in different combinations (Małecki et al. 2021). The linear sequence of amino acids within a protein is considered as the primary structure of the protein which drives the formation of the secondary structure through stable folding patterns such as alpha helices and beta sheets. The folding is further driven by non-specific hydrophobic interactions that ultimately determine the protein’s unique three-dimensional shape or tertiary structure and its function. The quaternary structure is a result of association of several protein chains into a specific spatial arrangement. Proteins undergo a wide range of structural and conformational changes through various complex interactions with other food components which provides numerous beneficial effects. Some of proteins have specific binding sites for other molecules that result in stable conformations upon binding and modulate the functioning of the system (Chen et al. 2021). In relation to the interactions between protein and phenolic compounds in food, proteins can form complexes with them, leading to changes in their structural, functional and nutritional properties which can broaden the range of functionalities achieved (Alu’datt et al. 2020; Ozdal et al. 2013).

Phenolic compounds are a major class of secondary metabolites in plants that are mainly derived from phenylalanine and tyrosine (Shahidi & Chandrasekara 2017). In this category, phenolic acids, namely hydroxybenzoic acid and hydroxycinnamic acid derivatives may be noted that are regarded as simple phenols consisting of a benzene ring substituted with hydroxyl groups. Meanwhile, polyphenols contain multiple rings with more than one phenolic group. Polyphenols can be classified as flavonoids and non-flavonoids according to their structures. The flavonoids include flavanones, flavanonols, flavones, flavonols, isoflavones, flavanols and tannins (hydrolysable and condensed) that are the most abundant and widespread dietary polyphenols (Fig. 1). The chemical structures of phenolic acids are given in Figs. 2 and 3. The different flavonoids differ greatly in their molecular structure based on the degree and pattern of glycosylation, hydroxylation, methoxylation and/or prenylation as shown in Figs. 4 and 5 (Guan et al. 2021). Most of them possess antioxidant activity, and hence they are frequently used as additives in functional dietary products (Li, He, et al. 2020). The nature of interactions and formation of thermodynamically favorable conformation were studied by docking and MD simulations of ferulic acid with the dimer and monomer forms of βLG. The results showed that the preferred binding site in the dimer form lies at the interface of the two monomers whereas it lies within the calyx shaped β-barrel of monomer form of βLG which exhibits the highest binding affinity. In two cases, the complexes were stabilized by hydrogen bonding and hydrophobic interactions. The binding of ferulic acid in monomer form reflects that the ligand entering at the calyx region of βLG following non-covalent interactions with the surrounding residues is essential in stabilizing the ligand within the receptor. This kind of information could be highly relevant to the food industry to enhance associative interactions for the development of novel bioactive compounds (Abdollahi et al. 2020).

The in-silico elucidation of the antidiabetic mechanism of EGCG with α-glucosidase indicated the binding of EGCG at the site close to the active site pocket of α-glucosidase and forms a stable complex with five intermolecular hydrogen bonds, which are the main forces between EGCG and α-glucosidase. It was identified that the binding site between α-glucosidase and EGCG was different from acarbose, a competitive inhibitor against α-glucosidase and there was no interaction between EGCG and the catalytic residues of Glu-277 and Asp-352. The docking studies illustrated that EGCG is a non-competitive inhibitor for α-glucosidase, providing the sense that EGCG might act as a promising antidiabetic agent (Xu et al. 2019).

Docking validation of potential human protein targets of apple polyphenols and possible mechanisms of chemoprevention in colorectal cancer revealed that some selected antioxidants can form stable complexes at cavities different from the active site. For instance, the binding of chlorogenic acid (CGA) to GTPase H-ras occurs at the allosteric site of the enzyme instead of the canonical binding site. In the presence and absence of ions/cofactors, direct docking analysis has been carried out to understand if the ligand is bound to a functionally active molecule and identify potential and stable protein-ligand interactions with selected targets. In this study, the lowest binding energy or the most stable complex corresponds to the interaction between quercitrin and hypoxanthine-guanine phosphoribosyltransferase enzyme. In addition, it was found that KDM1A gene coding lysine demethylase 1A and hyperin prefers binding whereas gene GAMT coding guanidinoacetate N-methyltransferase is the preferred target for hyperin, isoquercitrin, phloridzin and rutin present in apple. These observations clearly support the hypothesis that polyphenolic compounds act synergistically with proteins, providing cumulative effects on nucleotide metabolism and methyltransferase enzymes similar to the action of anti-cancer drugs (Scafuri et al. 2016).

Youn and Jun (2019) investigated the potential inhibitory action of cardamonin, pinocembrin, and pinostrobin on beta-site amyloid precursor protein cleaving enzyme1 (BACE1) as natural products-based therapy for Alzheimer’s disease (AD). Molecular docking analysis illustrated a non-competitive inhibitory activity for all these three compounds with reference to resveratrol used as the positive control. According to the binding energies, the most stable conjugates among proposed complexes were cadamonin-BACE1 which showed the strongest and effective inhibition. However, it was suggested that hydrophobic interaction between this compound involving the stabilization as cardamonin does not form any hydrogen bonds with BACE1. Stable conformations formed between P-glycoprotein and the flavonoids were analyzed through docking simulations to predict their possibility to pass the blood brain barrier. The computational analysis indicated that these flavonoids from B. rotunda may be promising AD preventative agents, and extensive examination needs to be carried out through in-vitro assays to support these predictions.

Identification of flavonoids that can regulate sirtuin 6 (SIRT6) activities is considered promising therapeutics for age-related diseases including cancer, diabetes, neurodegenerative diseases and metabolic disorders. Therefore, molecular docking has been carried out to discover their binding sites on SIRT6, and to identify principle interactions occurring on the enzyme active site with inhibitors and activators. The results have shown that the cleft present in enzyme forms the pocket for acetylated sirtuin substrate and for NAD+ cofactor, which requires the activation of the enzyme. Furthermore, they reported that activator compounds bind to a site outside of the cleft by forming interactions with a loop near the acetylated peptide substrate binding site. The activators may induce conformational changes in the loop upon binding to the putative activator site, improving acetylated substrates’ binding. Contrarily, the binding site of the majority of inhibitors was situated close to the binding site of known sirtuin inhibitors which can restrict NAD+ binding. Among different flavonoids, catechins with galloyl moiety exhibit a greater inhibitory activity by partially occupying the acetylated sirtuin substrate’s binding site and allowing more interactions with the binding pocket. It was also identified that kaempferol can act as a potent dual modulator due to its ability to bind multiple sites and form similar interactions with the same amino acid residues where the known activators and inhibitors interact with. An in-silico based mutation analysis was performed to examine the impact of activator site’s residues Gly-156, Asp-185, Trp-186, Glu-187 and Asp-188 on the action and/ structure of SIRT6. Extensive docking analysis revealed that different flavonoids could alter SIRT6 activity in a structure-dependent manner (Guan et al. 2021; Rahnasto-rilla et al. 2018).

Troxerutin (TX), a bioflavonoid, has been shown to exhibit anti-neoplastic and anti-cancer activities. Assessing the modulatory action of TX on transcription factors such as IKKβ, Nrf2 and Keap-1 through docking studies illustrated that TX interacts with the active sites of proteins by forming hydrogen bonds and π-cation interaction. TX with IKKβ conjugate has shown the most stable complex with the lowest binding energy compared to binding with other transcription factors, implying that high binding affinities are accompanied with hydrogen bond interactions. Based on the combination of docking analysis with immunoblotting and immunocytochemistry analysis, it was predicted that the formation of stable conjugates might result in a conformation change of transcription factors, followed by inactivating the expression of oncogenes and hence can be deemed as a potent drug in anti-cancer therapy (Thomas et al. 2018).

The structure-based virtual screening of flavonoids indicated that most flavonoids can bind to the acetylated lysine (KAc) binding site of BD1 of Brd4 receptor molecule and can act as novel natural bromodomain inhibitors. Flavonoids were found to occupy the active site, forming hydrogen bonds between the acetyl carbonyl oxygen and the amino group of the receptor’s conserved Asn-140 amino acid residue. Quercetin binds in a similar manner and shows the highest binding affinity while illustrating that these ligands may prevent the binding of Brd4 to acetylated lysines on nucleosome histones and inhibit RNA Polymerase II mediated transcription elongation. Moreover, the blocking of KAc site may inhibit the binding to the Myc gene resulting in a low level expression of the Myc gene, leading to low proliferation of cancer cells (Raj et al. 2017).

Investigation of inhibitory effects of anthraquinones on tyrosinase shows that they enter the active site of tyrosinase in the form of one molecule and competitively inhibit the activity of tyrosinase. The majority of interactions were electrostatic forces and hydrophobic interactions compared to hydrogen bonds due to the presence of aromatic rings in the structure of anthraquinones. Further analysis revealed that the inhibitory effect on tyrosinase activity was accomplished by acting on histidine residues bound to copper ions rather than chelating them. It was suggested that the binding of anthraquinones at the active site of tyrosinase results in conformational changes of secondary structure and prevented the entry of substrates, inhibiting the tyrosinase activity and in turn regulating the melanogenesis (Zeng et al. 2020). Similar molecular docking analyses were carried out to evaluate the synergetic effect of quercetin, cinnamic acid and ferulic acid on tyrosinase enzyme. Quercetin located at the hydrophobic pocket of the enzyme formed strong hydrophobic interactions and showed slightly higher binding affinity compared to binding with cinnamic and ferulic acids. Furthermore, computational docking simulations showed that quercetin, cinnamic acid and ferulic acid bind with different sites on tyrosinase in a non-competitive manner, exhibiting their ability to express synergistic action in inhibiting the tyrosinase activity (Yu et al. 2019).

The preferred binding sites of phenolics on scallop gonad protein isolates (SGPIs) were identified and visualized by molecular docking simulation studies. Vitellogenin and β-actin, as the main crystal structures of SGPIs have been analyzed with EGCG to evaluate their ability to form stable complexes. It was found that EGCG inserts into the hydrophobic central cavity of vitellogenin or β-actin and stabilizes the conjugate via hydrogen bonds, van der Waals and hydrophobic interactions. Besides, interaction of these proteins with EGC and ECG resulted in decreasing order of SGPIs-binding capacity of phenolics as EGCG > ECG > EGC, indicating that binding affinities mainly rely on the number of –OH groups present in the ligand molecule. In the process of interactions, hydrogen bond and van der Waals forces dominated between SGPIs and EGCG while hydrophobic interaction forces were dominant in SGPIs-ECG complex. Therefore, a comprehensive theoretical understanding of the induced effect of these phenolic compounds on the structure and function of SGPIs could be used to accomplish desired practical aspects (Han et al. 2021).

Investigation on binding sites and binding affinities of catechin derivatives for bovine serum albumin (BSA) showed that ester catechins (EGCG and ECG) possess high binding affinities than non-ester catechins (EGC and EC) as they can form more hydrogen bonds (Yu et al. 2020). Further, interaction of EGCG with BSA by molecular operating environment (MOE) suite docking simulation program has indicated that EGCG interact with both Trp − 134 residue; at drug-binding site I by π–π stacking and Trp-213 residue; on the molecular surface of BSA (Ikeda et al. 2017). These results suggest that substitution at the C-3 position or galloyl moiety of catechins determines their binding affinities against serum albumin. Together, these simulation studies support the idea of formation of tea cream and controlling the performance of tea beverage products by introducing tea polyphenols (Yu et al. 2020). In studying the molecular nature of interactions between β-casein and p-coumaric acid, computational docking has been done to identify the location of their specific binding site and to obtain the thermodynamically stable conformation of β-casein with p-coumaric acid. Modeling outcomes have shown that the ligand molecule interacts with amino residues within the core of β-casein receptor molecule forming a hydrogen bond between hydroxyl group of p-coumaric acid and the carbonyl group of the peptide backbone of Ile-27 (Kaur et al. 2018).

Several studies have shown that docking algorithms are capable of identifying putative phenolic-protein binding sites of novel conjugates with their binding affinities, analyzing interactions and conformations that can be used to recognize and interpret their potential activities from practical perspectives. Moreover, it demonstrated that polyphenols with galloyl moiety bind most strongly to extended proteins with a high proline content to form thermodynamically favorable complexes. The degree of hydroxylation, methoxylation and steric hindrance of the polyphenol mainly determine its binding affinity for the protein molecule and structural conformation of the compound.

Conclusion

Complexation of phenolic compounds with proteins via covalent and/or non-covalent bonding entails changes in proteins that could result in favorable or unfavorable properties. The phenolic-protein interactions affect the compound’s activity and exert synergistic or antagonistic effects depending on the type and structure of compounds, molecular weight, concentration, pH, temperature, cofactors, method and food processing conditions, and physiological status. These conjugates are able to improve biological activities such as antioxidant, anti-inflammatory, anti-allergic and anti-cancer activities and bioavailability of polyphenols compared to individual components. In addition, they can also act as food preservation agents against microorganisms and lipid oxidation, emulsions for delivery of nutraceuticals, edible films for food packaging and drug releasing modulators and may also act as natural food color, flavor and texture modifier. It has been demonstrated that the underlying mechanism of these actions and interactions at the molecular level can be extensively investigated, specifically using molecular docking approaches with various analytical techniques. It can be concluded that polyphenols bind most strongly to extended proteins with a high proline content and polyphenols with galloyl moiety to form thermodynamically favorable complexes, by modulating their activities. As shown in earlier studies, molecular docking programs are able to successfully predict the binding modes between protein and polyphenols in a large scale. Therefore, the docking studies provide scientific basis to determine putative binding sites, stable conformations of conjugates and identifying promising compounds that can be developed further, to accomplish desired functional and health aspects. However, more in-depth studies are needed for a comprehensive understanding of the interaction between phenolic compounds and proteins and factors that affect their binding affinities for enhancing and expanding the application potential of different phenolic compounds in the food and pharmaceutical industries.