The Evaluation of Citrus bergamia Phytochemicals as Potential Cholesterol-Lowering Agents against HMG-CoA Reductase: An In Silico Molecular Docking Study ^†

Abstract

Elevated cholesterol levels, or hypercholesterolemia, have been recognized as the underlying cause of various diseases, most notably cardiovascular diseases. Unfortunately, most cholesterol-lowering (or anti-hypercholesterolemic) drugs are associated with several adverse effects, emphasizing the need to identify new cholesterol-lowering strategies. Natural products, particularly bioactive phytochemicals, have gained significant attention for their safer profile, fewer side effects, and potential health benefits, including cholesterol-lowering properties. The citrus fruit bergamot (Citrus bergamia) is renowned for its diverse array of bioactive phytochemicals. In this study, an in silico approach was utilized to assess the cholesterol-lowering potential of phytochemicals derived from C. bergamia. Molecular docking using AutoDock Vina of the selected phytochemicals was performed against the HMG-CoA reductase (HMGR), an enzyme targeted for hypercholesterolemia. Our results indicated that among the selected 20 phytochemicals, 8, namely eriocitrin, narirutin, scolymoside, neodiosmin, brutieridin, neohesperidin, rhoifolin, and naringin, exhibited better binding affinities than the conventional HMGR inhibitor, atorvastatin (−9.2 kcal/mol). Notably, among these top eight phytochemicals, eriocitrin displayed the most favorable binding affinity of −10.0 kcal/mol. These findings strongly imply that C. bergamia possesses potential HMGR-inhibitory activity and anti-hypercholesterolemic activity, primarily due to the high binding affinities exhibited by its phytochemical constituents. Therefore, further studies must be considered to comprehensively explore the cholesterol-lowering properties of C. bergamia phytochemicals.

1. Introduction

Hypercholesterolemia, a metabolic condition characterized by high levels of cholesterol in the blood, is known to be a risk factor for atherosclerosis, heart attack, and coronary heart diseases [1]. HMG-CoA reductase (HMGR) is the rate-limiting enzyme that catalyzes the conversion of HMG-CoA to mevalonate, a precursor that is needed for cholesterol biosynthesis. The inhibition of HMGR at this committed step has proven effective in treating hypercholesterolemia [2].

Statins are a class of cholesterol-lowering drugs that act as competitive inhibitors of the HMGR enzyme. However, concerns have been expressed regarding the adverse effects of long-term statin use, highlighting the need to identify new cholesterol-lowering strategies with safer profiles and fewer side effects [3,4].

Citrus bergamia, commonly known as bergamot, is native to southern Italy and is a citrus fruit that is rich in bioactive phytochemicals, specifically flavonoids and polyphenolic compounds [5,6]. Previous studies have reported its cholesterol-lowering properties, with evidence showing that administration of C. bergamia can reduce total cholesterol and low-density lipoprotein cholesterol levels in both animal and human models [7,8,9]. However, to date, limited information is available regarding the activity of its phytochemicals in targeting the key enzymes that regulate cholesterol levels in the body. Moreover, there is a lack of in silico molecular docking studies that thoroughly explore the potential of its various phytochemicals to inhibit HMGR. To address this gap, we conducted molecular docking using AutoDock Vina to evaluate 20 selected polyphenolic compounds from C. bergamia as potential cholesterol-lowering agents.

2. Materials and Methods

2.1. Preparation of Protein

The 3D crystal structure of human HMG-CoA reductase (HMGR) in a complex with the inhibitor atorvastatin (see Figure 1) (PDB ID: 1HWK), with a resolution of 2.22 Å, was downloaded in the PDB file format from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) (www.rcsb.org, accessed on 23 January 2024). The PDB file was prepared using Discovery Studio Visualizer (DSV) software (version 21.1.0.20298). HMGR forms a tightly associated tetramer with four active sites: two located between polypeptide chains A and B, and the other two between chains C and D, symmetrically [10]. For the docking study, only chains A and B were used. All bound substances, including co-crystallized ligands, coenzymes, water molecules, and protein chains C and D, were removed from the HMGR 3D structure using DSV.

Figure 1. Three-dimensional crystal structure of HMG-CoA reductase (HMGR) with co-crystallized atorvastatin (encircled in yellow) bound at the active site.

2.2. Preparation of Ligands

Ligands targeting HMGR were selected from a library comprising 20 polyphenolic phytochemicals, identified in studies by Russo et al. and Baron et al. [5,6]. This library specifically consisted of 18 flavonoids (12 flavones and 6 flavanones) and 2 coumarins. The 3D structures of the ligands were downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/, accessed on 23 January 2024). For the ligands that were not available in the PubChem database, their 2D structures were manually drawn in BIOVIA Draw software (version 21.1) and subsequently converted to 3D using DSV. We performed geometry optimization and energy minimization of the ligands using Avogadro software (version 1.2.0) using the universal force field and the steepest descent algorithm.

2.3. Molecular Docking

Prior to docking, the AutoDockTools (ADT) software (version 1.5.6) was used to further prepare the protein and the 20 ligands. Using ADT, polar hydrogens and Kollman charges were added to the protein structure, while the number of torsions for the ligands was kept at default. Both the protein and ligand structures were then saved in PDBQT file formats. The AutoDock Vina software (version 1.1.2) was used for the molecular docking simulations. In these simulations, the protein was kept rigid, while the ligands were flexible. The docking grid box (size: x = 30, y = 30, z = 30 at 1 Å spacing) was positioned on the coordinates of the co-crystallized atorvastatin (center: x = 2.894171, y = −9.553439, z = −12.241561) and set to cover its binding site, which is surrounded by the amino acid residues Glu 559A, Cys 561A, Ala 564A, Ser 565A, Lys 735A, Asn 755A, Leu 853A, Ala 856A, Arg 590B, Val 683B, Ser 684B, Asp 690B, Lys 691B, and Lys 692B, based on the crystal structure analysis in DSV. AutoDock Vina was run and executed using the Command Prompt program on a Windows computer. AutoDock Vina generated binding orientations for each ligand with corresponding binding affinity values. The binding affinities of all 20 ligands were recorded and compared to that of the reference inhibitor, atorvastatin. The molecular interactions of each protein–ligand complex were then visualized and analyzed using DSV.

3. Results

3.1. Validation of the Docking Protocol

To evaluate the accuracy of AutoDock Vina as an appropriate docking tool for this study, the co-crystallized ligand (atorvastatin) was redocked into the active site of HMG-CoA reductase (HMGR), and the docked orientation was compared to the crystal structure orientation by calculating the root-mean-square deviation (RMSD)—a measure of the average distance between the atoms of superimposed structures. Generally, RMSD values ≤ 2.0 Å indicate that the docking protocol can accurately reproduce the binding orientation of the co-crystallized ligand [11]. In this study, the superimposed structures (see Figure 2) resulted in an RMSD value of 0.85 Å, indicating that our docking results are good and valid and that AutoDock Vina is reliable for docking the 20 phytochemicals into the active site of HMGR.

Figure 2. Superimposed 3D structures of the co-crystallized (blue) and docked (red) orientations of atorvastatin.

3.2. Binding Affinities of the Phytochemicals

Table 1 summarizes the binding affinities of the 20 selected Citrus bergamia phytochemicals after docking into the active site of HMGR. The results reveal that eight phytochemicals, namely naringin, rhoifolin, neohesperidin, brutieridin, neodiosmin, scolymoside, narirutin, and eriocitrin, exhibited stronger (or more negative) binding affinities (−9.5 to −10.0 kcal/mol) compared to the conventional HMGR inhibitor, atorvastatin (−9.2 kcal/mol).

Table 1. Details and binding affinity values of the 20 studied ligands.

Ligand Name	PubChem ID	Phytochemical Class	Binding Affinity (kcal/mol)
Eriocitrin	83489	Flavanone	−10.0
Narirutin	442431	Flavanone	−9.9
Scolymoside	5282152	Flavone	−9.8
Neodiosmin	69964214	Flavone	−9.8
Brutieridin	45279710	Flavanone	−9.7
Neohesperidin	442439	Flavanone	−9.7
Rhoifolin	5282150	Flavone	−9.6
Naringin	442428	Flavone	−9.5
Melitidin	168011951	Flavanone	−8.7
Vitexin	5280441	Flavone	−8.7
Scoparin	20055255	Flavone	−8.6
Diosmetin-8-C-glucoside	—	Flavone	−8.6
Stellarin 2	44258167	Flavone	−8.3
Lucenin 2	44257937	Flavone	−8.2
Diosmetin-6,8-di-C-glucoside	—	Flavone	−8.1
Vicenin 2	442664	Flavone	−8.0
Isovitexin	162350	Flavone	−7.8
Bergamottin	5471349	Coumarin	−7.7
Bergapten	2355	Coumarin	−7.3
Sinensetin	145659	Flavone	−7.1
Atorvastatin *	—	—	−9.2

* Reference inhibitor.

3.3. Protein–Ligand Interactions of the Top Ligands

Table 2 provides a summary of the interactions between the top eight ligands and the amino acid residues within the active site of HMGR. Notably, eriocitrin demonstrated the best binding affinity among these phytochemicals, suggesting a strong and favorable interaction with HMGR.

Table 2. Molecular interactions of the top 8 ligands with HMGR amino acid residues.

Ligand Name	Binding Affinity (kcal/mol)	Amino Acid Residues Involved in Interaction
		Hydrogen Bond	Hydrophobic Interactions	Electrostatic Interactions
Eriocitrin	−10.0	Gly 560A, Asn 755A, Arg 590B, Met 655B, Lys 691B, Lys 692B, Asp 690B, Ala 751A	Met 655B, Gly 806B	Glu 559A
Narirutin	−9.9	Gly 560A, Lys 735A, Asn 755A, Arg 590B, Lys 691B, Lys 692B, Asp 690B, Val 805B, Glu 559A	Met 655B, Gly 806B	Glu 559A
Scolymoside	−9.8	Gly 560A, Arg 590B, Gly 765B	Met 655B	Glu 559A, Asp 767B
Neodiosmin	−9.8	Gly 560A, Arg 590B, Lys 691B, Asp 690B	Met 655B, Gly 806B	Glu 559A
Brutieridin	−9.7	Asn 755A, Arg 590B, Met 655B, Gly 656B, Ser 684B, Lys 691B, Lys 692B, Asn 658B, Asp 690B	Met 655B, Ala 654B	None
Neohesperidin	−9.7	Gly 560A, Arg 590B	Met 655B	Glu 559A
Rhoifolin	−9.6	Gly 560A, Arg 590B, Glu 559A, Asp 690B	Met 655B, Gly 806B, Met 657B	Glu 559A
Naringin	−9.5	Asn 755A, Arg 590B, Lys 691B, Asp 767B, Gly 765B, Gly 560A	Met 655B, Gly 806B	Glu 559A, Asp 767B
Atorvastatin *	−9.2	Ser 565A, Ser 661B, Lys 735A, His 752A, Asn 755A, Arg 590B, Lys 692B, Asp 690B	Ala 856A, Leu 853A, Val 683B, Cys 561A, Ala 564A	None

* Reference inhibitor.

As seen in Figure 3, eriocitrin forms hydrogen bonds with amino acid residues, including Gly 560A, Asn 755A, Arg 590B, Met 655B, Lys 691B, Lys 692B, Asp 690B, and Ala 751A, within the active site of HMGR. It also forms hydrophobic interactions with Met 655B and Gly 806B and electrostatic interactions with Glu 559A. Atorvastatin, shown in Figure 4, forms hydrogen bonds with Ser 565A and Ser 661B via its bulky hydrophobic moiety, while its HMG-like moiety forms hydrogen bonds with Lys 735A, His 752A, Asn 755A, Arg 590B, Lys 692B, and Asp 690B. Additionally, atorvastatin forms hydrophobic interactions with Ala 856A, Leu 853A, Val 683B, Cys 561A, and Ala 564A. Notably, both eriocitrin and atorvastatin share common hydrogen bond interactions with Asn 755A, Arg 590B, Lys 692B, and Asp 690B, possibly indicating a similar binding pattern at the active site of HMGR.

Figure 3. Two-dimensional interactions of the top-ranked ligand, eriocitrin, with HMGR amino acid residues.

Figure 4. Two-dimensional interactions of atorvastatin with HMGR amino acid residues.

4. Discussion

The binding affinity measures the strength of interaction between a ligand and a target protein or enzyme [12]. More negative values indicate stronger binding affinities and greater stability [13,14]. In Table 1, the more negative binding affinities of the flavanones eriocitrin, narirutin, brutieridin, and neohesperidin, as well as the flavones scolymoside, neodiosmin, rhoifolin, and naringin, compared to atorvastatin—a known HMG-CoA reductase (HMGR) inhibitor—suggest that these bergamot phytochemicals may have greater potential for HMGR inhibition and could serve as more effective cholesterol-lowering agents than atorvastatin.

The finding that eriocitrin exhibited the most negative binding affinity among the top eight phytochemicals suggests a potentially superior inhibitory effect on HMGR compared to the other seven phytochemicals. Essentially, eriocitrin forms the most stable complex with HMGR among the studied phytochemicals, suggesting its potential as a potent inhibitor of HMGR activity.

The HMG-like moiety of atorvastatin mimics the natural substrate (HMG-CoA) of the enzyme, enabling it to competitively inhibit HMG-CoA reductase (HMGR) by binding to its active site with high affinity. This prevents the enzyme from catalyzing the conversion of HMG-CoA to mevalonate, a crucial step in cholesterol biosynthesis. By inhibiting HMG-CoA reductase, statins reduce low-density lipoprotein cholesterol (LDLc) levels, which in turn decreases the hepatic cholesterol content. This reduction triggers an upregulation of hepatic LDL receptors, enhancing the clearance of LDLc from the bloodstream [15]. As seen in the results, eriocitrin and the HMG-like moiety of atorvastatin share common hydrogen bond interactions with Asn 755A, Arg 590B, Lys 692B, and Asp 690B. This similarity in hydrogen bond interactions between them may suggest a similar competitive inhibition mechanism.

Furthermore, previous studies have reported the beneficial effects of orally administered bergamot juice in reducing total cholesterol (TC), low-density lipoprotein cholesterol (LDLc), and triglycerides (TGs), while increasing high-density lipoprotein cholesterol (HDLc) levels. This effect has been observed both in animal models and in dyslipidemic patients, with reductions of up to 40% in TC, LDLc, and TG [7,8,9,16]. These findings support the potential of bergamot phytochemicals in contributing to cholesterol-lowering effects and underscore their relevance as effective agents for managing cholesterol levels.

While the current molecular docking results highlight the promising inhibitory potential of these phytochemicals, the clinical relevance of these findings must be considered in the context of pharmacokinetics. ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) properties play a critical role in determining the drug-like behavior of these compounds. Although detailed ADMET data for the studied phytochemicals are not available in the current study, future research should explore these parameters to assess their bioavailability, metabolism, and potential toxicity. For example, compounds with strong binding affinities may still face limitations due to poor absorption or rapid metabolism, which could reduce their effectiveness as therapeutic agents. Eriocitrin, despite showing the strongest binding affinity, would require an in-depth ADMET analysis to determine if it has favorable pharmacokinetic properties for clinical use, such as sufficient oral bioavailability and minimal toxicity.

Another important factor to consider is the typical amount of these phytochemicals that is consumed in a regular diet compared to the dose of atorvastatin administered therapeutically. Eriocitrin and the other flavonoids are naturally present in citrus fruits, but the concentration of these compounds in a typical diet is likely much lower than the pharmacologically active doses of atorvastatin. This raises questions about whether dietary intake alone would be sufficient to achieve comparable cholesterol-lowering effects, or if higher, more concentrated forms (e.g., supplements) would be necessary. In a previous study, Gliozzi et al. (2013) reported that the bergamot polyphenol fraction effectively reduced total cholesterol, LDL, and urinary mevalonate and was safe when used alone or in combination with rosuvastatin (another type of statin) over a 30-day period [17]. While these results suggest that bergamot can be safely combined with statins, it is essential to consider the potential for food–drug interactions between the individual flavonoids and statins. Consuming these phytochemicals alongside atorvastatin or other statins might either enhance or interfere with the drug’s effects due to competition for the same target enzyme, HMG-CoA reductase.

5. Conclusions

Citrus bergamia, commonly known as bergamot, shows potential as an HMG-CoA reductase inhibitor and an anti-hypercholesterolemic agent based on molecular docking analysis. This could be due to the strong binding affinities of its phytochemical constituents observed in silico. While the molecular docking results provide strong evidence for the cholesterol-lowering potential of eriocitrin and related phytochemicals, future studies should incorporate ADMET analysis to evaluate their pharmacokinetic properties. Additionally, the typical dietary intake of these compounds and the potential for food–drug interactions should be further explored to assess their clinical relevance and applicability as natural cholesterol-lowering agents. Therefore, further in vitro and in vivo studies are needed to confirm these in silico findings and to thoroughly investigate the cholesterol-lowering effects of C. bergamia phytochemicals.