Interaction of quercetin and quinazoline with Interleukin 6 and Janus kinase 3 receptors, therapeutic targets in rheumatoid arthritis: An in silico approach

INTRODUCTION

Rheumatoid arthritis (RA) is a systemic inflammatory disease seen in many adults, caused by the malfunction of the autoimmune system. The disease is characterized by intense pain and swelling in the limb joints.¹ The disease sets in as a consequence of the body’s immune system going against and attacking its own cells, resulting in swellings and deformities. The inflammatory changes are mainly seen in the joints and synovial membrane in the form of cell proliferation leading to hyperplasia, made complicated by the penetration of inflammatory cells. Additionally, there may be inflammatory cytokines like tumor necrosis factor α, interleukin-1b, interleukin-6, etc. coming from macrophages. The severe impact of the disease, other than the pain and atrophy, is the gradual deformation of joints, distortion of muscles, and osteoporosis with erosion of bones. Those affected suffer from mobility-related issues, and eventually, there is an inferior quality of life with possible premature death.²

Various drugs have been used in the therapy of RA, most of which are chemically derived. Such chemical drugs always carry the problem of side effects. Moreover, there can be health issues associated with continuous usage over long periods of time like nausea, vomiting, gastrointestinal discomfort, and adverse reactions cardiovascular or central nervous system.³ Plant-based compounds can prove to be a suitable alternative to chemical drugs.

Bioinformatics is one of the recent and innovative approaches that also contribute to the design and development of novel drugs for various ailments. Bioinformatics facilitates the discovery of the mode of action of drugs in the body by elucidating the molecular interactions between drugs and proteins, the influence on metabolic pathways and functions in the body, and also suggesting genomic variants capable of changing drug response.⁴ Molecular docking, target point determination, simulation, and chemical stability studies are some of the major bioinformatics methods used in drug designing. Some of the drugs approved mainly by the interventions from bioinformatics include carbonic anhydrase inhibitors, dorzoalmide;⁵ captopril, the angiotensin converting enzyme inhibitor, as an antihypertensive drug,⁶ etc.

Molecular Docking is the prediction of the binding or interaction between a protein and a molecule (ligand) which in turn can also be a protein, in order to elucidate the mechanism of inhibition of the protein.⁷ Molecular docking has grown as one of the major methods used in structure-based drug design and has been used in the present study. The study also involved Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) analysis and BOILED egg plot of the selected ligands. ADMET analysis tells about the interaction within the system, retention in the system and also any toxicity associated with the molecule. The BOILED egg analysis mainly tells about the gastrointestinal absorption of the drug molecule and its possibility of penetrating the BBB.

The present investigation was conducted to analyze the interaction of two potential phytochemicals, quercetin and quinazoline, with two therapeutic targets of RA, interleukin 6 (IL-6) and Janus kinase 3 (JAK3), with the help of Bioinformatics. The two molecules have been reported to have some remedial effect in RA. However, the specific interaction of these two molecules with the receptors, IL6 and JAK3, has not been investigated so far. The study aims to shed more light on the mode of interaction of these two phytochemicals with the receptors and to explore the better molecule in this regard. This bioinformatics-based assessment of potential lead molecules can speed up the preliminary phase of drug discovery.

MATERIAL AND METHODS

RESULTS

Receptor and ligand structures

The ligands selected for docking were Quercetin and Quinazoline, which had their origin from plants. The targets or receptors selected for the present study were IL-6 and JAK3, which are associated with the pathogenesis of RA. The structure of the target proteins retrieved from PDB was PDB ID: 1P9M for IL-6 [Figure 1a] and PDB ID: 3LXK for JAK3 [Figure 1b]. The structure of the plant ligands obtained from PubChem was Quercetin- PubChem CID: 5280343 [Figure 2a] and Quinazoline- PubChem CID: 9210 [Figure 2b].

Close

Figure 1: Target proteins selected for docking, (a) IL-6 receptor, (b) JAK3 receptor.

Export to PPT

Close

Figure 2: Plant ligands selected for docking, (a) Quercetin, (b) Quinazoline.

Export to PPT

Docking analysis of ligands

The ligand Quercetin was bonded to the active site of the receptor IL-6 [Figure 3a] with a MolDock Score of -104.421kcal/mol, and the compound was involved in hydrogen bonding with amino acid residues SER226, TYR168, LYS118, SER165, GLY117, and GLN111 in the active site of IL-6 receptor. The ligand and the receptor share 6 linking hydrogen bonds. The strength of interaction between the ligand and the protein is dependent on the number of H-bonds existing between them. The ligand Quinazoline bonded with IL-6 [Figure 3b] with a MolDock Score of -56.246 kcal/mol, and the hydrogen bonding involved amino acids like GLY117, TYR168, GLN111, and TYR31. The ligand and the receptor were connected by 4 hydrogen bonds [Table 1].

Close

Figure 3: Docking interaction at receptor IL-6, (a) Quercetin, (b) Quinazoline.

Export to PPT

Table 1: Docking score of plant ligands quercetin and quinazoline bound to the receptor IL-6.

Name	MolDock scorea	Rerank scoreb	Interactionc	Internald	H-bonde	MWf	LE1g	LE3h
[00]5280343 Quercetin	-104.421	-51.9427	-116.772	-123.542	-7.330-2	302.236	-5.52995	-2.36103
[01]9210 Quinazoline	-56.246	-52.5773	-56.2669	-59.8868	-2.25067	130.147	-5.62669	-5.25773

^a MolDock score is represented in kcal/mol, ^bThe rerank score is a linear combination of E-inter (steric, Van der Waals, hydrogen bonding, electrostatic) between the ligand and the protein, and E-intra (torsion, sp2-sp2, hydrogen bonding, Van der Waals, electrostatic) of the ligand weighted by pre-defined coefficients, ^cThe total interaction energy between the pose and the protein (kJ/mol), ^dThe internal energy of the pose, ^eHydrogen bonding energy (kJ/mol), ^fMolecular weight, ^gLigand efficiency 1: MolDock score divided by heavy atoms count, ^hLigand efficiency 3: Rerank score divided by heavy atoms count.

The MolDock Score of Quercetin on bonding to JAK3 receptor [Figure 4a] was -118.601kcal/mol, and the bonding comprised amino acids like ASP949, LEU828, and ARG911 that were bonded with the ligand by 3 hydrogen bonds. The ligand Quinazoline bonded with the active site of JAK3 [Figure 4b] with a MolDock Score of -57.9824 kcal/mol, and the binding site contained amino acid residues like GLY831, VAL836, and LYS830 bonded by hydrogen bonds. The hydrogen bonds connecting the ligands (quercetin and quinazoline) and the JAK3 receptor were 3 [Table 2].

Close

Figure 4: Docking interaction at receptor JAK3 (a) Quercetin, (b) Quinazoline.

Export to PPT

Table 2: Docking score of plant ligands quercetin and quinazoline bound to the receptor JAK3.

Name	Rerank scoreb	MolDock scorea	Interactionc	Internald	H-bonde	MWf	LE1g	LE3h
[00]5280343 Quercetin	-79.6518	-118.601	-126.689	-118.007	-10.3565	302.236	-5.18359	-3.62054
[02]9210 Quinazoline	-51.4406	-57.9824	-72.4798	-58.5053	0	130.147	-5.79834	-5.14406

^aMolDock score is represented in kcal/mol, ^bThe rerank score in kcal/mol is a linear combination of E-inter (steric, Van der Waals, hydrogen bonding, electrostatic) between the ligand and the protein, and E-intra (torsion, sp2-sp2, hydrogen bonding, Van der Waals, electrostatic) of the ligand weighted by pre-defined coefficients, ^c The total interaction energy between the pose and the protein (kJ/mol), ^d The internal energy of the pose, ^eHydrogen bonding energy (kJ/mol), ^fMolecular weight, ^g Ligand efficiency 1: MolDock score divided by heavy atoms count, ^h Ligand efficiency 3: Rerank score divided by heavy atoms count.

The Re-rank score of quercetin with the receptor IL-6 was -51.9427 kcal/mol, and that of Quinazoline was -52.5773 kcal/mol. The Re-rank score of quercetin bound to the receptor JAK3 was -79.6518 kcal/mol, and that of Quinazoline was -51.4406 kcal/mol.

The interaction energy of Quercetin was found to be -116.77 kJ/mol, and that of Quinazoline was -56.2669 kJ/mol when docked in cavity 1 of the receptor IL-6. When ligands were docked in cavity 2 of the JAK3 protein, the interaction energy of Quercetin was found to be -126.689 kJ/mol, and that of Quinazoline was -72.4798 kJ/mol.

Drug-likeness analysis

According to Lipinski’s rule of five, H-bond donors, H-bond acceptors, molecular weight, Mol Log S, log P Octanol/water partition coefficient, and their positions have been presented in Table 3. Log P values of Quercetin and Quinazoline were observed to be less than 5. The highest lipophilicity was expected for Quercetin, having a log P value of 1.19, whereas Quinazoline showed a log P value of 1.09 [Table 3]. The molecular weight of either compound was found to be <500. The Lipinski’s rule of five permits less than 10 hydrogen bond acceptors (O and N atoms) and less than 5 hydrogen bond donors (NH and OH), which was obeyed well in the tested compounds [Table 3]. Figures 5 and 6 show the drug likeness plot for quercetin and quinazoline, respectively. Compounds with drug potential should have a value higher than zero. Quercetin has a value of 0.52 (positive value) as a drug-likeness score, while Quinazoline has -1.60 (negative value) as a drug-likeness score.

Table 3: Principal descriptors calculated by Lipinski’s rule of five.

Lead molecules	Molecular weighta	Number of HBAb	Number of HBDc	Mol Log Pd	Mol Log Se
Quercetin	302.04	7	5	1.19	-2.19
Quinazoline	130.05	2	0	1.09	-1.07

^a Molecular weight of the molecule (160 to 500); ^b Estimated number of hydrogen bonds that would be accepted by the solute from water molecules in an aqueous solution (not more than 10); ^c Estimated number of hydrogen bonds that would be donated by the solute to water molecules in an aqueous solution (not more than 5); ^d Log P for octanol/water (−2.0 – 6.5); ^e Predicted aqueous solubility, log S. S in moldm–3 is the concentration of the solute in a saturated solution that is in equilibrium with the crystalline solid (−6.5 – 0.5).

Close

Figure 5: Plotting of drug-likeness score of compound Quercetin using MolSoft.

Export to PPT

Close

Figure 6: Plotting of drug-likeness score of compound Quinazoline using MolSoft.

Export to PPT

ADMET analysis

The SwissADME tool was used to analyze the ADME properties, while the toxicity was calculated using the online tool PreADMET. The drug profiles of the plant compounds were evaluated by estimating various ADME properties, i.e., lipophilicity, water solubility, physicochemical parameters, pharmacokinetics, medicinal chemistry, etc. [Figure 7 and Table 4]. Quercetin, as well as quinazoline, showed high absorption from the gastrointestinal tract and a bioavailability value of 0.55. However, they differed in terms of synthetic accessibility, for which the former showed a value of 3.23 while the latter had a value of 1.00. Additionally, quercetin showed leadlikeness, which, in the case of quinazoline, showed violation. TPSA of Quercetin was found to be higher than Quinazoline and is below the 160 Å limit. The TPSA for quercetin was higher than quinazoline. The BBB permeation was estimated by the BOILED-Egg model (Brain Or IntestinaL EstimateD permeation predictive model), In the BOILED Egg plot, the compounds capable of penetrating the BBB lie within the yellow color, while those outside the yellow color are not permeable to the brain [Figure 8]. The drug likeness filters included herein are Ghose’s, Veber’s, Lipinski’s, Muegge’s, and Egan’s. The compound Quercetin showed no violations, whereas Quinazoline violated the rules of Muegge’s and Ghose’s drug likeness filters.

Close

Figure 7: Bioavailability radar analysis, (a) Quercetin, (b) Quinazoline.

Export to PPT

Table 4: Drug profile and ADME analysis of the compounds quercetin and quinazoline.

Parameter		Quercetin	Quinazoline
Physicochemical parameters	Molar refractivity	78.04	39.54
Lipophilicity	TPSA Log Po/w (iLOGP) Log Po/w (XLOGP3) Log Po/w (WLOGP) Log Po/w (MLOGP) Log Po/w (SILICOS-IT) Consensus Log Po/w	131.36Å2 1.63 1.54 1.99 -0.56 1.54 1.23	25.78Å2 1.62 1.00 1.63 1.01 2.05 1.46
Pharmacokinetics	GI absorption BBB permeant P-gp substrate CYP1A2 inhibitor CYP2C19 inhibitor CYP2C9 inhibitor CYP2D6 inhibitor CYP3A4 inhibitor Log Kp (skin permeation)	High No No Yes No No Yes Yes -7.05 cm/s	High Yes No Yes No No No No -6.38 cm/s
Druglikeness	Lipinski Ghose Veber Egan Muegge Bioavailability Score	Yes Yes Yes Yes Yes 0.55	Yes No; 3 violations: MW<160, MR<40,atoms<20 Yes Yes No; 1 violation: MW<200 0.55
Water Solubility	Log S (ESOL) solubility class	-3.16 2.11e-01 mg/mL; 6.98e-04 mol/L Soluble	-2.02 1.25e+00 mg/mL; 9.62e-03 mol/L Soluble
	Log S (SILICOS-IT) solubility class	-3.14 1.73e-01 mg/mL; 5.73e-04 mol/L Soluble	-3.29 6.67e-02 mg/ml; 5.13e-04 mol/L Soluble
Medicinal chemistry	Leadlikeness PAINS synthetic accessibility	Yes 1 alert: catechol_A 3.23	No; 1 violation: MW<250 0 alert 1.00

TPSA: Topological polar surface area

Close

Figure 8: BOILED-Egg analysis of Quercetin (molecule 1) and Quinazoline (molecule 2).

Export to PPT

Radar analysis was performed using SwissADME for both the compounds. The plot indicates both plant compounds to be lying outside the pink area with regard to the saturation parameter [Figure 7].

Toxicity analysis was done using the web server pkCSM-pharmacokinetics. Both Quercetin and Quinazoline showed no AMES toxicity, hepatotoxicity, hERG I and hERG II toxicity. Furthermore, the compounds showed widely varied values for minnow toxicity as shown in Table 5. The pkCSM software predicted the LD50 values of Quercetin and Quinazoline to be 2.471 mol/kg and 2.099 mol/kg, respectively, which fell completely within the safe range. The predicted values for Skin Sensitization were negative for Quercetin and positive for Quinazoline.

Table 5: The predicted toxicity of phytochemicals quercetin and quinazoline.

Parameters	Quercetin	Quinazoline
AMES toxicity	No	No
hERG I inhibitor	No	No
hERG II inhibitor	No	No
Hepatotoxicity	No	No
Skin sensitisation	No	Yes
Human max. tolerated dose (mg/kg/day)	0.499	0.622
Oral rat acute toxicity (mol/kg)	2.471	2.099
Oral rat chronic toxicity (mg/kg_bw/day)	2.612	2.424
T. pyriformis toxicity	0.288	0.231
Minnow toxicity	3.721	1.048

DISCUSSION

The present work involved a comprehensive computational investigation of the selected plant compounds, quercetin and quinazoline in the context of RA using molecular docking and ADMET analysis. The potential binding energies of the studied compounds was elucidated through the analysis. Additionally, the stability of the ligand–protein complex and the inflammatory protein target’s maximum affinity were also revealed. Finally, the ADMET analysis revealed a range of pharmacokinetic and toxicokinetic properties of the selected compounds.

The capability of phytoconstituents namely Quercetin and Quinazoline to form complex with the targets was expressed as MolDock Score. The parameters used for docking analysis included MolDock Score and the Re-rank score, of which the MolDock score was used to rank the phytoconstituents. A stable bond is indicated by negative values and higher the negativity more stable the bond shall be. The Re-rank score suggests the most promising docking solution among the solutions given by the docking algorithm (MolDock). Furthermore, docking experiments have showed that high-quality binding modes can be identified simply by using a simple docking scoring function followed by a re-ranking procedure.²¹

The interaction energy scores indicated that Quercetin showed energetically better interaction than Quinazoline at the binding site 1 of IL-6 protein. Quercetin also had a stronger interaction than Quinazoline at the binding cavity 2 of JAK3 protein also. Docking result of both the ligands (Quercetin and Quinazoline) and receptors (IL-6 and JAK3) revealed absence of electrostatic interactions but confirmed hydrogen bonding and steric interaction between the ligand and receptor. Among the various interactions, electrostatic forces contribute the most to the binding energy.²² However, no differences were observed in the studied molecules in this regard since only hydrogen bond and steric forces were observed.

With the MolDock Score of -104.421kcal/mol, the active compound Quercetin bonded with the receptor IL-6 more strongly than Quinazoline showing a MolDock Score of -56.246kcal/mol. Exhibiting a MolDock Score of -118.601kcal/mol, Quercetin bonded with the receptor JAK3 more efficiently than Quinazoline which had a MolDock Score of -57.9824kcal/mol. The results were interpreted based on the assumption that higher negative values of binding energy meant better thermodynamic stability.²³ Therefore Quercetin showed better docking results when compared to Quinazoline at both the receptors IL-6 and JAK3. Inhibition of the JAK receptors has been suggested as an effective strategy for dealing with the sufferings associated with rheumatoid arthritis and quercetin can be a promising lead.

Molsoft prediction tool was used for analysing the pharmaceutically significant properties and physically significant descriptors of the lead compounds. Log P is the partition coefficient which measures molecular hydrophobicity, a crucial characteristic in rational drug design. The result implied that the compounds were lipophilic in nature with good permeability through the cell membrane. Owing to the lipophilic nature and small size, the plant compounds seemed to be capable of getting easily transported and absorbed by the system. Drug likeness determines whether a lead molecule is having potency as a drug and whether it is bioavailable and safe within the system. This parameter is difficul to quantify but has been based on various molecular properties and structural characteristics like Hydrogen bonding features, hydrophobicity, electronic distribution, molecule size etc. Only quercetin seemed to qualify as a prospective lead and not quinazoline which showed one violation for drug-likeness.

Topological polar surface area (TPSA) is a parameter which suggests the transport properties of the drug such as intestinal absorption, penetration of the blood brain barrier etc. or in other words it is connected to the bioavailability of the drug and it is based on the hydrogen bonding potential of the molecule. The BOILED- egg plot revealed no BBB permeant in Quercetin but BBB permeant nature was observed in Quinazoline. Therefore Quercetin posed to be a better drug candidate than Quinazoline.

Radar analysis is a computational tool that gives idea about parameters such as flexibility, lipophilicity, polarity, size, solubility and saturation.²⁴ For a molecule to be considered suitable as a drug, its plot should remain within the pink region in the radar analysis.

The toxicity analysis proved Quercetin as safe since it showed no mutagenic or carcinogenic properties. Thus Quercetin shows less toxicity than the ligand Quinazoline. Moreover, the acute toxicity values [Table 5] remained within the acceptable limit indicating safety of the compound.²⁵ However, clinical studies are warranted for the final confirmation regarding toxicity. Quercetin from plants is well capable of acting against inflammation and it possesses gastro-intestinal cytoprotective and mast cell stabilizing activity.²⁶ Quercetin has been shown to somewhat relieve the inflammation associated with RA and prevent pannus formation, and hence it has prospects of becoming an adjuvant drug in RA treatment.²⁷ Quercetin also regulates gene expression associated with transcription factors (NF-kB), thus suppressing the secretion of inflammatory cytokines.^28,29 In a recent study, quercetin was shown to down regulate NF-kB in stimulated human mast cells in vitro.³⁰ The therapeutic potential of quercetin against RA has been tested in animal models and small clinical studies.^31-33 Plants such as Allium cepa, Hypericum perforatu, Camellia sinensis and Podophyllum peltatum possess quercetin in high concentration. It has been shown to act by promoting a significant reduction in edema volume in both acute and chronic models.³⁴

Quercetin’s effects in therapeutic procedures for RA have been evaluated in a rat model of adjuvant arthritis, and it was found to inhibit the enzymatic activity of 12/15-lipoxygenase.³⁵ The positives of quercetin as an adjuvant natural drug for treating RA include its minimal side effects and well-characterized pharmacological activities.³⁶ Excitingly, with the advent of nanoformulations, quercetin has improved not only in terms of oral bioavailability but has also opened the option for external transdermal use, which is a totally novel reference for the treatment of RA.³ The present analysis of the interaction of the phytoconstituents quercetin and quinazoline with the receptors IL6 and JAK3 also reveals the better and promising drug potential of quercetin.

The results of this study demonstrate that Quercetin has the potential to be used for the development of effective medication for use in the therapy of RA. It showed promising values with regard to the docking parameters, especially with the IL6 receptor. The results of the ADMET screening were also promising in this regard. It also shows great potential as a drug cited by other researchers for the treatment of RA. Previous findings have consistently shown that the long-term usage of synthetic drugs produces undesirable side effects, due to which there is a need to replace them with plant-derived drugs having minimal or no side effects.

The identification of new lead compounds from natural therapeutic products has been made extremely fast and cost-effective by the use of computational methods, which have grown as smart alternatives to experimental screening of large compound libraries. This has allowed a reduction in the number of experiments needed to determine the molecular mechanisms of action of therapeutic molecules.^37,38 However, the quercetin molecule needs to be subjected to further extensive screening through dry and wet lab experimentation, along with the inclusion of more receptors, in order to understand its efficiency. After ADME studies, it can be subjected to wet lab studies and finally designed as a plant-derived drug for the treatment of RA.

CONCLUSION

The results of this study demonstrated the interaction of two phytoconstituents, quercetin and quinazoline, with two receptors, IL6 and JAK3, relevant in RA. The phytochemicals have been implicated in the treatment of RA. The results of the study point towards the better possibility of Quercetin for use as a potential lead compound for drug development for RA. Quercetin showed promising values in the docking analysis and was found to be satisfactorily safe in the ADMET screening. Since the continuous use of synthetic drugs over long periods produces undesirable side effects, turning to plant-derived drugs with little or no side effects will be a more viable option. This is the first report of the analysis of the interaction between these two phytoconstituents with receptors IL6 and JAK3. Future expansion after ADME studies can be checked by subjecting quercetin to wet lab studies and finally designing a plant-based drug for the treatment of RA.