Unveiling Calotropis procera’s promising wound healing secrets by QSAR, molecular docking, DFT and ADMET profiling

Abstract

This research work comprises the computational analysis of natural compounds from Calotropis procera analyzed on the basis of their wound healing properties. 3D Quantitative structure activity relationship (QSAR) models were generated with statistical validation which was followed by leading compound candidates. This analytical technique was carried out for the evaluation of active compounds by correlation of structural parameters. Previously isolated natural compounds were selected on the basis of their in-vivo and in-vitro studies. After 3D-QSAR analysis, molecular docking procedure was performed on the active compounds against wound healing proteins (PDB: 6SMA, 5A8Y) and their binding scores were analyzed for the evaluation of binding interaction of Protein–Ligand complexes. Comparative analysis was also carried out by comparing the docking results of natural compounds and FDA approved antibiotics Cephalexin, Dicloxacillin, and Levofloxacin. After docking, Re-docking was also performed as an additional cross analysis technique for validation of the docking results. DFT study was carried out to reveal molecular activity by evaluation of quantum molecular properties. Hence, results revealed that Stigmasterol (12) is hit compound with better results than standard drugs Cephalexin and Levofloxacin. These derived results in this study provide significant information for wound healer drug-designing by using Stigmtasterol as natural compound having better wound healing property. The analysis suggested that the proposed hit compound exhibits the potential drug attribute, which indicates the suitability for future investigation to help treat the serious wounds or injuries.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-06822-z.

Introduction

Plants have been used as a medicinal source since ancient times, and these medicinal plants are still used for the development and production of modern drugs¹. Wounds are one of the most prevalent health problems globally². The wound is the damage or rupture of the skin tissues and wound healing is a gradual process of repairing the damaged tissues, clotting of blood, and killing of the infection. On the other hand, wound healing is a gradual dynamic process that is categorized into four stages: hemostasis, inflammation, proliferation and tissue maturation or remodeling^2–5. Wounds are categorized into two types on the bases of their causative agents and duration of the healing phases, such are acute and chronic wounds^6,7. Wounds may be caused due to some multiple reasons such as injuries, burns, cuts, surgery, some comorbid biological conditions like diabetes, skin necrosis and other vascular disorders⁷.

Human neutrophil elastase (HNE’s) is basically a serine proteinase polypeptide glycoprotein similar to chymotrypsin⁸. Elastases enzymes are produced and secreted by neutrophils and pancreas⁹. In the human body HNE’s are very crucial for the immunity and defense against invading microbial attack¹⁰. Neutrophil-derived elastases and proteinase 3 are the basic degeneration mediators of the cellular wound matrix, released at the wound site during inflammation phase. HNE’s play the destructive role by cleavage of the surface receptors and interfere with the cell signaling for the release of the leukocytes at the wound site, and influence the activation of the liberation of components that derive wound matrix for healing^11,12. Wound is healed and newly regenerated or repaired epithelial cells at the wound site protected from excessive hydrolysis or degeneration of the extracellular matrix by proteases and HNE inhibitors¹².

Medicinal plants after passing through the scientific investigation and experimental clinical trials reach the stage of drug development^13,14. According to the researchers about 15% of the medicinal plants have been studied for the analysis of essential phytochemical constituents and only 6% of these plants are assessed by scientific biological screening. No doubt use of medicinal plants is a traditional method of healing or treatment of various ailments and now many advancements in pharmaceutical industry have made clinical field a very sophisticated one with modern protocols of drug development¹⁴. The potential of the phytochemicals obtained from the medicinal plants is analyzed by formulating the plant extract containing that specific phytochemical constituents for wound healing and use of In vivo and in vitro assay techniques are used to evaluate efficacy of that phytochemical in clinical trials³.

Calotropis procera belongs to the plant family Asclepiadaceae also known as a giant milk weed has gained importance in in world health care system of developing countries due to its pharmacological aspect. It is an evergreen woody shrub with height approximately about 3–5 m. This plant is widely distributed in tropical regions and commonly in South West Asian regions¹⁵. Calotropis procera contains many phytochemical constituents that provide this plant the therapeutic potential to cure certain ailments. Presence of many phytochemicals such as calactin, amyrin uscharidin, calotropin, coroglaucigenin, amyrin esters, calotropageni, frugoside, voruscharine and calotoxin has made this plant very important for therapeutic purposes. Hence, Calotropis procera has proved to be a vital plant in health care management because of its therapeutic potential. Currently synthetic antimicrobial drugs are no doubt helpful to treat the ailment but it is expansive as compared to the herbal medicine and also has the certain side effects¹⁶. Extract of the Calotropis procera has the antimicrobial and insecticidal properties. Physico-chemical analysis of the alcoholic extract obtained from the leaves of the Calotropis procera identified the presence of mixtures of flavonoids. Isolated flavonoids from the extract of this plant are responsible for inhibition of bacterial growth and for protection from the insecticidal attack on grains and pulses¹⁷. The information about the FDA approval of these compounds is available on the data base https://www.fda.gov/ (Fig. 1).

Fig. 1.

Structural representation of the FDA approved compounds from C.procera having antimicrobial potential.

In-vitro or In-vivo experimental techniques for the Discovery of new drugs are expensive and time-intensive¹⁸. Computational drug analysis approach is used for the purpose of optimization of the target compounds leading towards the discovery of new drugs that is designated as Computer-aided drug design (CADD)¹⁹. Although it are further for computer-assisted molecular design (CAMD)²⁰. Various analytical approaches such as ligand-based drug design approach andstructure-based drug design approach are included in CADD. Virtual screening is performed by both of these approaches in molecular docking to determine and optimize the compounds of desired activity²¹. An advanced branch of computer based analytical techniqueis ‘Machine learning’ is used for the improvement in data mining tools. Drug discovery is one of the leading and exceptional analytical approach in CADD²². In-Silico analysis of the target ligands for drug designing includes some computational techniques such as:virtual screening to generate the pharmacophore models, conformation generation, 2D/3D quantitative structure–activity relationship (2D/3D-QSAR) quantitative free-energy calculation, ab-initio design, molecular docking and²³.

Herein this research paper, virtual screening of the natural compounds from the C. procera has been performed for the predictive analysis of the target compounds having improved wound healing properties. Standard drugs Dicloxacin, Levofloxacin, and Cephalexin are clinically used antimicrobial and wound healing medicines. In this current research we are going to analyze new natural compounds having efficient wound healing activity. IC₅₀ and PIC₅₀ of the compounds are used for QSAR model generation of the active antimicrobial natural compounds showing prominent wound healing property as well. After QSAR modeling active compounds were selected and analyzed with molecular docking that identified the target compounds with highest wound healing activity on the basis of docking score. Then Re-docking was also performed as an analytical tool for the validation of the accuracy of the docking results. DFT study was performed as an additional computational accuracy tool for the scrutiny of the atomic properties and quantum structure.

Materials and methods

Preparation of the database

Selected liberary of the natural compounds was downloaded from the Pubchem database https://pubchem.ncbi.nlm.nih.gov/.1290 compounds were extracted from the large liberary of compoundson PubChem, 110 natural compounds (Table S-1: supplementary material) from C. procera plant were selected having various bilogical activites. Then 50 compounds were selected having antimicrobial IC₅₀. Firslty, canonical smiles were downloaded from Pubchem and their molecular structures werecollectedfrom ChemOffice 2014 software. The receptorproteins having resolution larger than 1.5 Å were downloaded from Protein Data Bank database (RCSB)(http://www.rcsb.org/pdb). 3D-QSAR studies was performed by FLARE V5 software Package, CRESSET U.K. software.

3D-QSAR model generation

FLARE V5 software Package, CRESSET U.K. was used to generate the 3D-QSAR model on the basis of ligand based and structure-based technique. Fifty natural compounds having antimicrobial IC₅₀ valuesfrom C. procera were nominated using previous literature and PI_C50 were calculated as per activity by formula PIC₅₀ =  − log IC₅₀. With the random selection of compounds thirty-eight (76%)were selected as training set and remaining twelve (24%) were taken as test set and ratio was prepared for the 3D-QSAR modeling (Table S-1: Supplementary material). To investigate the reliability and effectiveness of the generated models, external evaluation was carries out on the developed 3D-QSAR models with the help of test set compounds. 3D-QSAR model was generated on the basis of PIC₅₀ values of the test set and training set of compounds. Antimicrobial protein (PDB: ID 5NNM) was selected for the 3D-QSAR study of natural compounds. Partial least squares (PLS) regression analysis was carried out for generating 3D-QSAR Field-based models providing the justification of the predictive activity. Statistical parameters were also computed such as RMSE_pred cross-validated correlation coefficient (q²), and cross-validated correlation coefficient (r²). Based on high values of r² and q² should fall under the following range of (r² > 0.6) and (q² > 0.5)²⁴.

Molecular alignment

In flare V5software, distance-dependent dielectric (DDD) with aligned field points on ligands were used in calculations and generation of model rather than random field points. Alignment of the field points of ligandsis carried out for the calculations of the similarity score using distance-dependent dielectric. 3D-QSAR Field-based models were generated according to the ligand-based alignment by analyzing and generating the possible poses for active sites of the protein molecule. Ligands were lined on the basis of the molecular structures using fields of the ligands.

Molecular docking study

Preparation of ligand

3D structures of the natural compounds were downloaded from https://pubchem.ncbi.nlm.nih.gov/ in simple document format (SDF).

Preparation of protein

Protein of wound healing with resolution less than 3 Å and organism Homo sapiens searched from the literature and then downloaded by an online protein database RCSB.PDB (https://www.rcsb.org/) (Table 1).

Table 1.

Characteristics of selected proteins.

Sr.no	PDB ID	Resolution (Å)	Organism	Reference
1	518Y	1.90	Homo sapiens	https://www.rcsb.org/structure/5a8y
2	6SMA	2.59	Homo sapiens	https://www.rcsb.org/structure/6SMA

Molecular docking of the selected compounds with predictive activity was carried out with 2 proteins (PDB ID: 5A8Y, and 6SMA). Moldock score was considered as the docking score function. Compounds showing higher than -70 Moldock score were considered to be the active phytochemical with strong hydrogen bond interactions and binding affinity with docked protein molecule. Binding affinity of the ligands with the amino acid residues on the active sites of the proteins as the docking results were visualized and analyzed in Discovery Studio Visualizer 2020. After completing the procedure of docking, MolDock score and hydrogen binding energies were generated which indicated the binding effectiveness. The resultant docked compounds were saved in PDB format, and their interactions with target protein was visualized through Discovery Studio software. The crystallized ligand was repositioned within the targeted receptor protein to investigate the precision of the docking process. The original crystallized ligand was compared with the resultant docking configuration, and the root mean square deviation (RMSD) value was calculated to evaluate the docking results under the predetermined threshold range of ≤ 2.0 Å²⁵.

ADMET

The absorption, distribution, metabolism, excretion, and toxicity of the lead drug checked by an online web tool ADMET-AI (https://admet.ai.greenstonebio.com/) on the bases of its percentile ratio.

DFT analysis

Optimize molecular geometry and analyze electronic properties, DFT calculations were carried out using the B3LYP functional with the 6-311G basis set of the reference drug and the lead compound. For the purpose, Gaussian 09 and Gauss view 6 employed. The B3LYP/6-31G level is perfect for medium-sized molecules because it strikes a compromise between precision and computational efficiency. The 6-311G basis set offers enough accuracy for reactivity and stability investigations, whereas B3LYP predicts geometries and electrical properties with reliability. For organic and biomolecular systems, this combination has received extensive validation (Hehre, Ditchfield et al. 1972). Analysis was done on electronic characteristics such as molecule electrostatic potential (MEP), HOMO–LUMO, and energy gap.

Results and discussion

D QSAR model

Active ligands were selected on the basis of non-covalent bond interactions with protein moleculessuch as H-bonding, π-π interactions, hydrophobic interactions, salt bridge and van der Waals forces. Van der Waals forces, electrostatic and hydrophobic interactions play anessential rolein determining the binding of active ligand on the active site of the target receptor protein. XED (eXtended Electron Distribution) force directs the distribution of elctrons in the ligand molecule for generating of 3D QSAR model²⁶.

3D-QSAR model gives the visual discription of the permediated field points of binding interactions of ligand on binding sites of the protein. Field points of the target compounds determine thebinding interactions such as Van der Waals forces, electrostatic and hydrophobic interactions. Common field points after the alignment of the ligand fields determine the strenght of the field points of the extended pharmacophores. 3D QSAR model represents various colored regions which displaythe 3D field points shown in figure (A) yellow, red, orange/golden and blue colors represent Van der Waals field, Electrostatic fields (Positive and Negative), and aromatic ring character respectively.

Red and blue colored field points represented the hydrogen bonding and ionic bond interactions and yellow and golden/orange colorsdisplayed Van der Waals and hydrophobic field points²⁷. Unsaturation in the ligands represents the pi–pi interaction on the active site of the protein molecule. The graphs generated in the 3D QSAR studies describes the best fit model on the basis of r²_, q² values and root mean square error (RMSE) which results in the reliable descriptive information of hit compounds.

enerated 3D-QSAR model and field points

3-dimentional models were analyzed on the basis of coeffiecient and variance of the field points that indicatethe biological activity of the target compounds. A reference compound was used as a standard ligand to generate 3D-QSAR modeland for the better alignment and distribution of the field points²⁸. In this research paper, β-sitosterol was used as a refrence ligand for the aligning the field points of target compounds (Fig. 2).

Fig. 2.

Representation of the electrostatic potential regions and molecular field points alignment of active ligands according to the reference ligand. Here, blue color represents negative electrostatic field points, red color displays the positive electrostatic field points, golden/orange color represents hydrophobic field points, and field points of yellow color representsVan der Waals forces.

3D-QSAR model displayed the regions of field points having significant impact on the activity of the target compounds. Number of the field points around the ligands and target protein indicate the strength of electrostatic fields. These field points are the characteristic features having considerable impact on the biological activity of the compounds. Higher number of field points around the protein–ligand complex depict the higher binding affinity values with higher association between the steric fields. Generated field points indicate the extended pharmacophores. Negative field points represented by blue color indicate the regions with favourable interactions due to H-bond donors of target protein receptor. Hydrophobic field points were represented by orange or golden colored field points indicating hydrophobic interactions. Regions with red color represent positive electrostatic field on aligned molecular structure. Positive electrostatic field indicates H-bond acceptors on the target protein. Van der Waals field points in yellow color displayed possible surface of active site ontarget protein (Fig. 3).

Fig. 3.

Visualization of molecular field of reference ligand with the field points representing electrostatic variance and coefficients that indicating the bioactivity of gererated 3D-QSAR model. (A) Red color represents positive electrostatics field points, and negative electrostatic coefficients represented by cyan color. (B) Electrostatic field points indicating region of high changes. (C) Field points displaying regions of high steric variance.

Field points contribution towards the predicted activity

The activity of the ligands is predicted by evaluating the field points in fieldregions of structural 3D-QSAR.Favourable electrostaticcontributions field of green and blue color square shapesrepresented the increase in predicted activity. On the other hand unfavourable electrostatic contributions were represented byorange color square shaped field points, that contributed to decrease in predicted activity (Fig. 4).

Fig. 4.

Electrostatic contributions of the reference compound, displaying region contribution to the predicted activity. Square shaped cubes in green and cyan color represents the favorable electrostatic and steric contributions respectively, both these contributions seem to have positive impact on the predicted activity by increasing activity. Field points of orange cube shapes indicates unfavorable electrostatic contributions regulating decrease in predicted activity.

Activity atlas model

3D-QSAR ligand based virtual screening by Flare provides 3-dimensional visualization of the field based activity of the generated models having considerable imact on the biologivcal activity of the target compounds. The refrence drug was superimposed on these generated field points to comprehend and compare the active field points in 3D-view. Analysis of structure activity relationship (SAR) was carried out for evaluation of the activity of the target ligands, as compared to the reference ligand. Analysis of activity atlas model was done by analysis of regions present in activity cliff summary and average of actives. Activity atlas model displayselectrostatic interactions, structural similarities of the field points, and SAR of the active ligands²⁸. The higher regions of electronegitive, electropositive field points and hydrophobic surface contours of active compounds indicates stronger biological activity²⁹. Larger regions of red and cyan colors in activity cliffs summary of 3D-SAR analysis depicts the higher activity as compared to the repurposed refrence drug (Fig. 5). Analysis of activity cliff summary of hydrophobics indicated the presence of larger region of favourable hydrophobic interactions represented by green and magenta color that indicated the higher activity of the target compounds according to reference compound (Fig. 6). The area under the red color represents the higher biologically active part of the ligands while the area under the cyan color represents the less biologically active area of the ligands according to the standard of QSAR.

Fig. 5.

3D field view of the activity cliffs summary. (A) Activity cliffs summary of the reference ligand. (B) View of activity cliffs summary of the active aligned ligands. Positive electrostatics displayed by red color regions and negative electrostatics represented by cyan color regions. Larger region in red and cyan color indicates the higher activity.

Fig. 6.

Activity cliff summary of hydrophobic of SAR model with 3D molecular contour plots. (A) Activity cliff summary of hydrophobic of reference ligsnd. (B) Activity cliff summary showing favorable and unfavorable hydrophobic interactions of aligned ligands. Cliffs model’s regions in green and magenta color indicate the favorable hydrophobic interactions and unfavorable hydrophobic interactions, respectively.

Activity cliff summary analysis

3D molecular contour view of cliff summary models explored the forecast activity ofactive ligands. Positive field regions are represented by red color. Results suggested thatred color regions of positive filed in the SAR contour structures, regulate the biological activity (Fig. 7a). These activity clif summary diagrams visualize the active parts of target compounds compared to the reference drug. Hydrophobics regions of the aligned contourfield of ligands also explored the activity. larger region of yellow color in hydrophobics indicated the favourable hydrophobic interactions of the hit ligands,predicting biological activity (Fig. 7b). Shape explored in white color regionsindicates high activity and presence of either positive or negative field points in this region contributed to the higher pharmachological activty. SAR model is mainly consisted of white region which indicated higher bioactivity (Fig. 7c). Average electrostatics of actives revealed acrucialimpact on pharmacological activity. Negavtive electrostatic field regionin cyan color regulates the activity that assisted in concluding theactivity of pharmacophore (Fig. 7d).

Fig. 7.

3D view of regions (a) positive electrostatics, (b) hydrophobics, (c) shape, (d) negative electrostatics of active reference ligand.

Validation of the 3D-QSAR models for activity prediction

In-Silico studies includes the alalysisof 3D-QSAR models that revealed the prediction of acitvity of target ligands. Initially, predicted activity of the test and training set lignads was evaluated by analyzing QSAR models. Virtual screening was further carried out after analyzing the binding sites of target compounds through ligand field analysis.

Virtual screening

Ligand-based virtual screenig was carried out to predict the hit ligands having prominentwound healing activity. After 3D-QSAR analysis ten hit ligands were selected having higher activity on the basis of compariveligand field analysis criteria of trainging set and hence predicted activities were considered to be reliable. Evaluation of activity-based descriptors was carried outby QSAR models and activity of pharmachophores was predicted. Ligands showing (50% inhibitory concentration) IC₅₀ value less than 16.2 µM were nominated for 3D-QSAR analysis. According to the ‘Average of actives’ ligands with PIC₅₀(activity) greater than 4.8 were considered as hit ligands having prominentantimicrobial and wound healing activity. Therefore, β-sitosterol, nystatin, alpha-copaene, gallic acid, caryophyllene oxide, alpha-amyrin, ciproflocacin, ursolic acid, streptomycin, neryl acetone, and stigmasterol were nominatedas hit compounds. Further analysis was performedby virtual screeningof hit ligands for bioavailability (oral) usingLipinski’s rule of five. Virtual screening was carried out by ADME analysis and toxicity risks were evaluated through drug-likness.

Molecular docking study for predictive analysis of wound healing activity

Virtual screening is an optimization process carried out by molecular docking, for evaluation of binding pose of the active and stable ligand with the receptor protein molecule. In this computational analytical technique evaluation of binding affinity of target ligand on receptor protein molecule was carried out. Highest binding score value was exhibited by most stable protein–ligand complex³⁰. Molecular docking of active ligands was performed against 2 wound healing proteins. Stigmasterol was considered as hit compound because it resulted in highest Moldock scores with both the docked proteins and it’s docking results were better than standard drugs Cephalexin and Levofloxacin. Stigmasterol exhibited best docking results with both the wound healing proteins − 127.002 and − 141.903 with PDB ID: 6SMA and 5A8Y respectively. With Pdb id 6SMA Stigmaterol shows two hydrophobic interactions LEU100 and ARG178, The type of both the interactions are alkyl-alkyl and their respective distances are 4.15611 and 3.86091. With pdb id 5A8Y stigmasterol shows 13 hydrophobic interactions with amino acid residues CYS168, ARG178, ARG178, ALA232, VAL181, ARG178, LEU130, VAL162, VAL181, ARG129, LEU130, VAL181, and HIS210. All the interactions shows alkyl type of interaction except the HIS210 that shows pi-alkyl interaction and their respective distances are 4.74133, 4.03932, 4.67793, 5.04492, 3.68713, 2.86113, 3.73891, 2.95271, 3.50983, 3.96706, 4.6856, 4.46745, and 4.6889.

On the other hand, Beta-sitosterolalso showed good Moldock scores with both the proteins -133.002 and -138.892 respectively, with 6SMA beta-sitosterol shows 7 hydrophobic interactions with amino acid residues LEU100 and their respective distances are 4.9902, 5.37818, 4.71146, 3.93525, 3.25319, 4.29845, 4.165. While with 5A8Y beta-sitosterol shows 12 hydrophobic interactions with amino acid residues ARG129, ARG129, ARG178, ARG178, ARG178, VAL181, ALA232, ALA232, CYS168, VAL181, and PRO230. All of these amino acid residues show alkyl type on interactions along their respective distances that are 4.4601, 4.69737, 3.79438, 4.95414, 4.76323, 4.42319, 3.3602, 4.01622, 4.66932, 4.80969, and 4.72841. Alpha-amyrin showed good moldock score − 106.527 with wound healing protein (PDB ID: 6SMA) (Table. 2).

Table 2.

Docking results of hit ligands with ‘Wound Healinng Proteins’.

Sr.#	Compound	Protein(6SMA) Moldock score	Protein(5A8Y) Moldock score
1	Alpha-amyrin	− 106.527	− 99.941
2	Gallic acid	− 72.1948	− 86.8584
3	Neryl acetone	− 107.119	− 109.43
4	Nystatin	946.373	950.671
5	Alpha-copaene	− 61.3187	− 83.7074
6	Caryophyllene Oxide	− 85.6932	− 108.039
7	Ciprofloxacin	− 113.705	− 122.145
8	Ursolic acid	− 104.827	− 95.2392
9	Streptomycin	− 138.035	− 147.142
10	Stigmasterol	− 127.002	− 141.903
11	Beta-sitosterol	− 133.002	− 138.892

Caryophyllene oxide also showed good moldock score of − 108.0389 with protein PDB ID: 5A8Y and shows 8 hydrogen bond interactions and 10 hydrophobic interactions. The amino acid residues of the hydrogen interactions are ARG129, ARG129, GLN233, HIS210, THR164, ASN180, ARG128, and HIS210 along their respective distances that are 2.26167, 2.93146, 2.64611, 2.89524, 2.40603, 3.03425, 2.33327, and 2.64794. It also shows 10 hydrophobic interactions with the amino acid residues VAL162, CYS168, VAL181, ALA232, LEU130, VAL181, HIS210, ARG178, ARG129, and VAL181 along their respective distances that are 3.89267, 5.28491, 4.26513, 4.1283, 4.86539, 4.33209, 5.31062, 5.1867, 4.44382, and 5.05271. 3D and 2D images of the structural representations showing hydrophobicity and hydrogen bonding of docking poses were also visualized through Discovery Studio to observe the docking results. These results of stigmasterol displayed its interactions on binding sites of wound healing protein (PDB ID: 5A8Y) (Fig. 8). Hydrophobic region represented the available surface for favorable interactions (Fig. 8).

Fig. 8.

3D and 2D structural view showing hydrophobicity and hydrogen bonding of the dockedProtein-Ligand complex displaying stigmasterol interactions with wound healing proteins (PDB ID:5A8Y).

Molecular docking of standard drugs

Comparative analysis of standard drugs with active ligands was performed by molecular docking study on the basis of best moldock score values of the hit compounds with wound healing proteins as compared to standard drugs (Tables 2 and 3). 3 antibiotics having antimicrobial and wound healing potentialwere selected as standard drugs and their molecular docking was performed with 2 wound healing proteins (Table 3). 2 neutrophil elastrase proteins PDB ID: 5A8Y and 6SMA with3-Oxo-beta-Sultam inhibitor LMC249 and dihydropyrimidone inhibitor respectively were selected as wound healing receptor proteins for molecular docking analysis.

Table 3.

Molecular docking results of standard drugs.

Sr.#	PDB ID	Inhibitor	Cephalexin (Moldock score)	Dicloxacillin (Moldock score)	Levofloxacin (Moldock score)
1	5A8Y	Dihydropyrimidone	− 136.651	− 153.127	− 116.362
2	6SMA	3-Oxo-beta-Sultam inhibitor LMC249	− 105.103	− 150.612	− 99.2132

Comparative analysis of the active ligands with standard drugs

Hit compound Stigmasterol exhibited best Moldock score of − 127.002 and − 141.903 with both the wound healing proteins PDB ID: 6SMA and 5A8Y respectively, comparatively higher than the binding scores of standard drugs, Levofloxacin and Cephalexin (Tables 2 and 3). Therefore, on the basis of docking results we can say that stigmasterol shows better wound healing properties as compared to Cephalexin and Levofloxacin.

Docking validation

Validation of the docking (Re-docking) procedure was performed for the accuracy and evaluation of the ability to rejuvinate the original co-crystallized poses of the ligands. Re-docking was carried outwithout optimization procedures of the ligands because ligands were removed from the protein molecule and redocked into the original receptor. Root Mean Square Deviation (RMSD) was designated as re-docking parameter (Fig. 9).

Fig. 9.

Structural view of the superimposed protein (PDB:5A8Y) with ligand after re-docking.

ADMET

The ADMET study of the lead ligand Stigmasterol carried out by using an online tool ADMET-AI which showed that stigmasterol has good physicochemical properties as the molecular weight of the compound is 412.70 which is leass than 500 has good TPSA value less than 140 and other parameters also has good drug bank percentile as shown in the Table 4:

Table 4.

ADME analysis of compounds with parameters.

Property	Value	Drug bank percentile (%)	Units	Property	Prediction	Drug bank percentile (%)	Units
Physicochemical properties				Absorption
Molecular weight	412.70	68.20	Dalton	Human intestinal absorption	1.00	68.28	–
LogP	7.80	97.98	log-ratio	Oral bioavailability	0.72	38.93	–
Hydrogen bond acceptors	1.00	6.65	#	Aqueous solubility	-6.82	1.74	log(mol/L)
Hydrogen bond donors	1.00	36.68	#	Lipophilicity	4.79	99.26	log-ratio
Lipinski rule of 5	3.00	20.92	# of 4	Hydration free energy	− 2.06	97.36	kcal/mol
Quantitative estimate of druglikeness (QED)	0.46	41.68	–	Cell effective permeability	− 5.07	44.44	log(10−6 cm/s)
Stereo xenters	9.00	93.95	#	PAMPA permeability	0.97	89.34	–
Distribution				Excretion
Blood–Brain barrier penetration	0.70	46.34	–	Half life	15.00	61.23	hr
Plasma Protein binding rate	100.00	97.79	%	Drug clearance (hepatocyte)	74.92	79.99	uL/min/106 cells
Volume of distribution at steady state	7.17	84.30	L/kg	Drug clearance (Microsome)	53.76	79.57	uL/min/mg
Metabolism				Toxicity
CYP1A2 inhibition	0.01	31.68	–	Estrogen receptor (full length)	0.28	87.59	–
CYP2C19 inhibition	0.09	48.39	-	Estrogen Receptor (ligand binding domain)	0.11	88.87	–
CYP2C9 inhibition	0.05	52.46	–	Peroxisome proliferator-activated receptor gamma	0.01	60.53	–
CYP2D6 inhibition	0.03	42.34	–	Nuclear factor (erythroid-derived 2)-like 2/antioxidant responsive element	0.42	78.83	–
CYP3A4 inhibition	0.07	51.34	–	ATPase family AAA domain-containing protein 5 (ATAD5)	8.43e−04	21.52	–
CYP2C9 substrate	0.15	50.25	–	Heat Shock factor response element	0.19	91.62	–
CYP2D6 substrate	0.19	64.99	–	Mitochondrial membrane potential	0.35	79.99	–

DFT analysis

Molecular descriptor’s calculation

After docking, the ligand-receptor complex may still suffer from some geometrical inaccuracies, as docking algorithms generally involve rigid body docking or simplistic treatment of flexibility. DFT can help by optimizing the geometry of the ligand-receptor complex and making more realistic corrections for the conformational changes that occur upon binding. While molecular docking may identify a potential binding site and a plausible binding mode, DFT allows for the detailed analysis of the interactions (e.g., hydrogen bonds, van der Waals forces, electrostatic interactions, and π–π stacking interactions)^31,32. Stigmasterol as active wound healing natural compound with best docking results was analyzed by Gaussian 09 supported by GaussView 5.1 interface which gave the visual description of optimized molecular structures³³. Basic DFT configuration set 6311-G with hybrid B3LYP function was performed with geometrical optimization of molecular structure, which provided essential information about molecular descriptors such as net charge, orbial geometries (HOMO and LUMO), energy gap (ΔE), and dipole moment³⁴. Quantum parameters obtained by DFT study are shown in Table 5.

Table 5.

Calculation of quantum parameters of Stigmasterol with 5A8Y protein on the basis of DFT study.

Parameters	Calculated values of quantum chemical parameters
EHOMO (eV)	− 0.23659
ELUMO (eV)	0.01654
Energy gap “ΔE = ELUMO − EHOMO”	0.25313
Dipole moment “µ(Debye)”	1.79249
Chemical hardness “η	0.12656
Chemical softness “S”	7.90108
Electronegativity “χ”	0.11003
Electrophilicity Index “(ω)’’	0.04782
Ionization potential “I = − EHOMO”	0.23659
Electronic energy “E” (Hartree)	− 1209.148
Chemical potential “CP”	− 0.110025
Nucleophilicity index “N”	20.91044

Frontier molecular orbial analysis (FMO)

According to Frontier molecular orbial theory, molecular orbitals HOMO and LUMO aresignificantdescriptors of reactivity of target compounds. LUMO accepts electrons and HOMO donates electrons. Optimized values of HOMO and LUMO calculations describe the comparative stability of HOMO than LUMO. Ionization potential and electron affinity values depend upon HOMO and LUMO respectively³⁴. Energy gap is calculated by difference between HOMO and LUMO and smaller the value of energy gap softer is the target ligand with greater polarizability, higher pharmachological activity with less kinetic stability³⁵. Stigmasterol displayed small energy gaps (ΔE) such as 0.25313 (Fig. 10). Smaller value of energy gap indicates that nucleophile is soft and eaily gives reaction. Small value of electrophilicity index (ω) 0.04782, determined less electrophilic character. HSAB (hard-soft-acid–base) concept is that hard acids tend to react with hard bases and soft acids react with soft bases.

Fig. 10.

View of Frontier molecular orbitals of Stigmasterol showing energy gap between HOMO and LUMO.

Molecular reactivity discriptors

In DFT study various significant reactivity parameters of molecular structure are discussed such as electronegativity (χ), chemical (global) hardness (η), chemical (global) softness (S), chemical potenial (µ), electrophilicity index (ω),dipole moment (µ) and electronic energy (E) (Table 5). Calculated values of electrophilicity and nucleophilicity indexes are also also significant parameters. Pharmachologically active compound exhibits less value of electrophilicity index (ω) and higher value of nucleophilicity index (N). Optimized molecular structure of Stigmasterol showed high value of nucleophilicity index 0.76844 eV and very less value of electrophilicity index 20.91044 (Fig. 11). Low value of electronegativity (χ) 0.04782 (Table 5), indicated that Stigmasterol is a good electrophile as it can easily accept electron pairs³⁶. Charge distribution in a molecule is determined by its dople moment. Larger value of dipole moment is due to larger difference in electronegativity values of the bonded atoms³⁷. Large calculated value of dipole moment leads to best charge distribution, higher conductivity and low electrophilicity³⁸. Stigmasterol showed higher value of dipole moment 1.179249 Debye (Table 5). If value of electronic energy (E) is more negative then compounds is more stable due to strong electrostatic forces, dipole–dipole interactions, and hydrogen bond with lower value of IC₅₀, higher solute–solvent interactions and higher binding affinity³⁹. Calculated values of DFT parameters discussed above indicated pharmachological activity of Stigmasterol.

Fig. 11.

View of optimized geometrical structure of Stigmasterol with vector of dipole moment.

Molecular electrostatic potential (MEP)

Molecular electrostatic potential (MEP) is a significantanalytical tool for observation ofmolecular interaction with the target protein, electrophilic, nucleophilic sites, electron density, shape, and molecular size, relative polarity^40,41. Active sites of the molecules, size, shape, and charge density are determined by electrostatic potential surface and electron density map surface⁴⁰. MEP map of Stigmasterol was found to be in range − 5.904e−2 (extreme negative) to 5.904e−2 (extreme positive) respectively. Region in green color displayed zero electrostatic potential, red–orange colored region on MEP scale showed negative electrostatic potential and region in blue color displayed positive electrostatic potential. Positive electrostatic potential (blue color region) on MEP scale indicated the available site for nucleophilic reaction,negative electrostatic potential (red color region)represented the site for electrophilic reactions andred or yellow color on the scale represents site for electrophilic attack.Hence we can conclude that,oxygen atom bonded in the molecule allows electrophilic attack (Fig. 12). Green color indicated the neutral regions, having no contribution in molecular reactivity (Figs. 10 and 12).

Fig. 12.

Molecular electrostatic potential (MEP) scale of Stigmasterol.

Conclusion

This research paper deals with generating 3D-QSAR field-based model of natural compounds from C. procera to investigate wound healing potential. Compounds having good antimicrobial IC₅₀ value were selected for further analysis of wound healing potential. Furthermore, molecular docking was performed as a predictive analytical technique, in which target ligands were docked against wound healing receptor proteins. Results of molecular docking revealed that all the natural compounds exhibited good binding score but Stigmasterol displayed best and exceptional Moldock score against wound healing protein. Stigmasterol showed greater Moldock scores as compared to already available standard drugs Cephalexin and Levofloxacin used for wound healing purpose. On the other hand, Nystatin showed higher positive Moldock score that was out of applicable range. Ciprofloxacin and streptomycin also showed good docking results but these compounds were not considered as hit compounds because these drugs are already available in market as antibiotics. Hence Stigmasterol was considered as hit ligand that gave best binding interaction with wound healing protein PDB:5A8Y. Re-docking was performed as an additional analytical tool for accuracy of docking results. DFT study was also carried out for evaluation of quantum molecular properties to reveal Stigmasterol’s molecular activity. Hence, after evaluation of 3D-QSAR models, statistical results of molecular docking and quantum mechanical calculations obtained by DFT analysis, it was concluded that Stigmasterol is the hit compounds having better wound healing potential as compared to standard antibiotics Cephalexin and Levofloxacin. These outcomes might open up the possibility for Stigmasterol clinical trials that could result in the discovery of a new drug for wound healing.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Maryam Rashid, Musaab Dauelbait: proofreading, English check, Figure quality, and resolution; Edit Drafting, evaluation of results; Neha Sajjad, Erum Chaudhry: Library preparation, 3D-QSAR, Pharmacophore generation, Molecular Docking, validation of study, article writing, DFT study; Nusrat Shafiq, Pravin Badhe: Conceptualization, writing the original draft, supervision, Molecular docking study, Funding; Shagufta Parveen, Mohammed Bourhia: Data analysis, Literature evaluation; Rashid Ahmed Khan, Mohammed H AL Mughram: Library data analysis, proofreading, English check.

Funding

This work was funded by the Higher Education Commission of Pakistan under Grant# TDF03-172.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.