Sage Journals: Discover world-class research

Abstract

Background: The prevalence of multi-drug-resistant E. coli isolates is on the increase around the world. This study was designed to evaluate the antimicrobial activity of Tetrapleura tetraptera (Uyayak) fruit bioactive compounds against E. coli using an in silico approach. Methods: Gas chromatography coupled to mass spectrophotometry (GC-MS) was used to identify the bioactive compounds in our sample T. tetraptera. The resulting bioactive compounds from GC-MS were converted into canonical strings and used for target and absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties predictions using the pkCSM and SWISSADME tools, respectively. Bioactive compounds that met Lipinski's rule of five (ROF) were docked against the dihydropteroate synthase of E. coli using the AutoDock Vina tool, and the resulting interactions were visualized in 2-D using Biovia Discovery Studio 21. Results: The GC-MS analysis returned a total of twenty-eight (28) bioactive compounds, out of which 13 did not violate Lipinski's ROF. ADMET analysis of the screened bioactive compounds showed better absorption (intestinal and water solubility) and toxicity (AMES and hepatoxicity) profiles than trimethoprim, the control drug. Molecular docking gave docking scores that ranged from −4.0 to −5.3 kcal/mol for the bioactive compounds while that of trimethoprim was −6.5 kcal/mol. Target prediction showed that all the bioactive compounds are capable of reaching various targets, with the nuclear receptor being the most abundant target. Conclusion: The bioactive compounds of T. tetraptera fruits examined showed favorable docking scores and pharmacokinetics, suggesting the need for further studies to validate their potential as antimetabolites.

Keywords

Docking ADMET Uyayak dihydropteroate synthase

Background

Esherichia coli is a Gram bacteria and a normal flora of the large intestines of human and other warm-blooded animals.¹ Unhygienic practices have made its isolation to be common among edible foods and vegetables.² Most strains are harmless, but some strains acquire bacteriophage or plasmid DNA-encoding enterotoxins or invasion factors that make them pathogenic or virulent.³ These virulent strains are responsible for diarrheal infections worldwide, and other infections such as neonatal meningitis, septicemia, and urinary tract infections (UTIs). The most common pathogenic strains of E. coli are enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), Shiga toxin-producing E. coli (STEC), enteroaggregative E. coli (EAEC), EIEC, enteropathogenic E. coli (EPIC), and diffusely adherent E. coli (DAEC). Their virulence is attributed to the virulent factors they produce, which include colonization and adhesion factors, the production of toxins, and effectors.^3,4

Added to these virulent factors is the ability of this pathogen to develop antimicrobial resistance to commonly used antibiotics in its management,⁵ making it to be enlisted as one of the priority pathogens by the World Health Organization (WHO).⁴ There are numerous reports of multi-drug resistance among E. coli isolates from various samples.^1,6‐8 The incidence of infections caused by multidrug-resistant strains of E. coli is on the rise.⁷Added to this public health challenge is the fact that they have an outstanding ability to act as donors or recipients of resistant genes.⁷ Commonly used antibiotics that are used to manage infections caused by species of the Enterobacteriaceae, such as extended-spectrum penicillins, cephalosporins, monobactams, carbapenems, fluoroquinolones (ciprofloxacin), and aminoglycosides (gentamicin), are not spared the multidrug resistance feat of E. coli.^9–12 Multidrug-resistant E. coli is on the WHO's list of priority and critical pathogens for which new antibiotics are urgently needed.^4,12 Members of the critical pathogens list, especially the multidrug pathogens, have been implicated in various hospital and nursing home infections.^4,12

Developing new antibiotics is a time-consuming and financially demanding process.¹³ To save time and money, researchers have turned to several strategies. In one study, researchers utilized machine learning, bioinformatics and explainable artificial intelligence to screen numerous compounds.¹⁴ Yet others have turned to screening bioactive compounds from medicinal plants using in silico approaches.^15,16 One of such plant is Tetrapluera tetraptera, locally called Uyayak in Cross River State, Southern Nigeria. The extracts of the fruit, leaves, stem, root, and essential oils of T. tetraptera have been shown to contain bioactive compounds that are responsible for its antimicrobial (bacteria and fungi) activities.^17–21 Lin et al¹⁷ revealed the mechanism of action of the root extract of T. tetraptera to be via eliciting severe damage to the integrity of the E. coli cell and its membrane permeability. These findings, put together, indicate that the various extracts and their bioactive compounds have excellent antimicrobial activities that need further evaluations. To the best of our knowledge, no study has utilized an in silico approach to evaluate the antimicrobial activity of the T. tetraptera bioactive compounds against any target protein of E. coli. Hence, the aim of this study was to conduct an in silico assessment (molecular docking) of the bioactive compounds revealed by Gas chromatography coupled to mass spectrophotometry (GC-MS) in addition to the evaluation of the drug-likeness via the prediction of their pharmacokinetic properties.

Methods

Collection of Fruit, Fruit Salad and T. tetraptera Samples

T. tetraptera (Uyayak) pods (Figure 1) were purchased from Marian and Watt Markets, respectively, in August 2022 and also stored in a Ziploc bag where they were transported without ice and within one hour of collection.

Figure 1.

Whole fruit pod (left) of the T. tetraptera sample used in the study and the 3D crystal structure of a binary complex of E. coli dihydropteroate synthase (right) retrieved from protein data bank.

Figure 2.

GC-MS chromatogram showing the peaks of the various bioactive compounds obtained from T. tetraptera pod in this study.

GC-MS Analysis of T. tetraptera

This was done as reported previously.^22–25 Exactly 10 g of the milled T. tetraptera pod was added to 20 mL of methanol, and the resulting mixture placed on an orbital shaker for 15 min. The mixture was allowed to stand for another 15 min after shaking. The mixture was then filtered, and the filtrate concentrated to 5 mL using a rotatory evaporator. From the concentrate, bioactive compounds were screened using an Agilent 5890N gas chromatograph fitted with an autosampler connected to an Agilent mass spectrophotometric detector. All the operating conditions were as previously reported.²⁶ The bioactive compounds in the sample were identified and interpreted using the database of the National Institute of Standards and Techniques (NIST).²⁶

Prediction of ADMET Properties of the Ligands and Target Prediction

Bioactive compounds were screened for their absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties using the pkCSM prediction tool. The pkCSM tool is a well-established tool for the prediction of ADMET properties of lead drug compounds.²⁷ First, the ligands were converted into their respective SMILE strings at PubChem (https://pubchem.ncbi.nlm.nih.gov/). The strings were then used individually to predict the ADMET properties of the various ligands, together with their molecular descriptors. Using the online SWISSADME tool, potential targets were predicted and the top ten targets displayed.²⁸

Lipinski Rule of Five (ROF) Screening

The ligands (bioactive) compounds obtained from the GC-MS analysis were further screened for Lipinski's rule of five (ROF) using the SWISSADME tool. The SWISSADME is a free-web tool used in the prediction of lead compounds or small molecules pharmacokinetics, drug-likeness, and medicinal chemistry friendliness.²⁶ Following Lipinski's ROF screening, ligands that meet the ROF were further prepared for molecular docking.

Ligand Preparation for Docking

The screened ligands were prepared for molecular docking using various tools. First, the structures of the ligands were retrieved from their unique SMILES strings using the Chem3d 15.1 tool. Following retrieval of their structures, the energies of the ligands were minimized and then saved in pdb files. The minimized ligands were further prepared using AutoDock 4.2 and saved in pdbqt files individually.²⁹

Preparation of Proteins for Molecular Docking

From the Research Collaboratory for Structural Bioinformatics (RCSC) protein database (https://www.rcsb.org/), the dihydropteroate synthase of E. coli was retrieved. The details of the protein as obtained from the MCULE tool and RCSC PDB were 1AJ2, DHPS_ECOLI, P0AC13, and 83333 for the PDB ID, Uniprot name, Uniprot accession ID, and Uniprot taxonomic ID, respectively. As a protein complexed with native ligands (n = 2), the binding or active site of the protein was retrieved using these native ligands as a guide. The definition of the binding site was done using Biovia Discovery Studio Client 21. The binding site was defined to be x = 41.244328, y = 4.789960, and z = 8.014888, with a radius of 15.434639. The protein was further prepared by removing water molecules, and the resulting protein structure saved in pdb format. The protein in pdb was converted into a pdbqt file using AutoDock 4.2 for docking.²⁹

Molecular Docking and Visualization

Molecular docking was performed by AutoDock Vina version 4.2.²⁹ First, the various AutoDock files were retrieved and configured accordingly to align with the various prepared ligands and protein. Using the docking command lines, docking was performed, and the resulting docking modes were split for each ligand. The various interactions in the various binding modes were then visualized in 2D using the Biovia Discovery Studio Client 21 software to display the amino acids, types of bonding, and bond lengths.

Results

Figure 2 shows the GC-MS chromatogram (abundance vs retention time) of the bioactive compounds from T. tetraptera. The bioactive compounds are represented by the various peaks. The peaks were 28, representing a total of 28 bioactive compounds. Table 1 shows the retention time (min) of the 28 bioactive compounds, their molecular weights, peak areas (%), and development (min). The retention time ranged from 8.274 to 24.710 min for the compounds, methone and beta-caryophyllene. On the other hand, the molecular weights of the various compounds ranged from 16.04 to 478.7 g/mol for methone to carpaine, respectively. Peak area shows that lupeol was the most abundant with a peak area of 21.71, while eugenol had the least peak area with a value of 1.13. This implies that lupeol was the concentrated biaoctive compound while the least was eugenol. Classification of the bioactive compounds indicates that the compounds were diverse with monoterpenes (n = 8) being the most abundant in term of number and this was followed by sesquiterpene (n = 3). Other classes of compounds were triterpene (n = 1), sesquiterpene alcohol (n = 1), monoterpenes alcohol (n = 1), monoterpenes ketone (n = 1), dihydrazine (n = 1), hydroxy fatty acid (n = 1), triterpenoid (n = 2), fatty alcohol (n = 1), alcohol (n = 1), phytosterol (n = 1) and phenylpropanoid (n = 1).

Table 1.

Names, Retention Times, Molecular Weights, Peak Area and Development Times of the Various Bioactive Compounds in Our Studied T. tetraptera Pod Revealed by GC-MS.

Compounds	R.T (min)	Molecular weight (g/mol)	Peak area (%)	Development (min)	Class of compound
Methone	8.274	16.04	7.86	36	-
Lemonene	8.839	136.24	4.21	51	Monoterpene
Octanol	9.421	130.228	2.3	74	Fatty alcohol
Methanol	9.718	156.27	9.42	62	Alcohol
Dihydrocarvone	10.352	152.23	2.74	42	Monoterpene
Carvone	10.894	150.22	1.82	69	Monoterpene
Beta-bourbonene	12.251	204.35	8.15	19	Sesquiterpene
Methyl-10-methylheptadecanoate	12.758	298.00	1.64	52	Fatty acid esther
Alpha-copaene	12.962	204.00	3.71	63	Sesquiterpene
Campesterol	14.313	401.01	2.53	28	Phytosterol
Piperitone	14.802	152.23	5.49	39	Monoterpene ketone
Beta-pinene	15.428	136.23	7.82	47	monoterpene
Lupenon	15.820	424.00	2.64	51	Triterpenoid ketone
Beta amyrene	16.326	410.00	1.22	58	Triterpene
Terpinen-4-ol	16.711	154.25	3.43	74	Monoterpene alcohol.
Lupeol	18.336	426.00	21.71	39	Triterpene
3-Carene	18.758	136.24	1.62	42	Monoterpene
Butanedioly dihydrazide	20.213	112.00	2.44	72	Dihydrazide
Alpha-pinene	20.801	136.23	8.79	68	Monoterpene
Eugenol	21.369	164.2	1.13	53	Phenylpropanoid
Sabinene	21.741	136.23	6.57	31	Monoterpene
15-Hydroxypentadecanoic	22.303	258.38	3.86	29	Hydroxy fatty acid
Carpaine	22.718	478.7	4.14	72	Tropane alkaloid
T-Cadinol	23.352	222.27	2.72	54	Sesquiterpene alcohol
Alpha-ocimene	23.694	136.24	6.63	18	Monoterpene
Olean-12-en-3-0ne	23.857	424.00	1.77	24	Triterpenoid
n-Hexadecanoic acid	24.293	256.00	3.42	36	Saturated fatty acid
Beta-caryophyllene	24.710	204.36	0.86	41	Sesquiterpene

Key = - no class.

Table 2 shows the predicted ADMET properties for ligands that had no violation of Lipinski ROF. Adsorption was evaluated using water solubility, intestinal absorption, skin permeability, P-glycoprotein substrate, and P-glycoprotein I and II inhibitors. The water solubility values of the study bioactive compounds ranged from −2.25 to 5.477, while those of trimethoprim was −2.744, which was within the range of our study ligands. Intestinal absorption for the various ligands ranged from 91.485 to 95.257 and was higher than that of trimethoprim, which was 71.787. Skin permeability for the ligands ranged from −1.063 to −2.715 while that of trimethoprim was −2.735. A total of 11 ligands were not substrates for P-glycoprotein while ligands II and X as well as trimethoprim were substrates. All the ligands, as well as trimethoprim, were not inhibitors of P-glycoprotein I and II. Distribution was evaluated using VDss (human), fraction unbound (human), BBB, and CNS permeabilities. From the predicted result presented in Table 2, the VDss values ranged from −0.574 to 0.812 for the ligands and 0.954 for the control drug. Fraction unbound values ranged from 0.0104 to 0.514 for the ligands and 0.554 for trimethoprim. For BBB and CNS permeabilities, the values for the ligands ranged from 0.084 to 0.792 and −2.977 to −1.763, respectively, while for trimethoprim, the respective values were −1.225 and −3.273 for BBB and CNS permeabilities. The predicted parameters for metabolism were CYP2D6 and CYP3A4 substrates and CYP1A2, CYP2C19, CYP2C9, CYP2D6, and CYP3A4 inhibitors. From the predicted result presented in Table 2, none of the ligands nor trimethoprim were substrates for CYP2D6 while for CYP3A4, 11 ligands were not substrates, but ligands 10, 13, and trimethoprim were substrates. All the ligands as well as the trimethoprim were not inhibitors of CYP2C19, CYP2C9, CYP2D6, and CYP3A4; however, ligand VIII was predicted to inhibit CYP1A2. The predicted excretion parameter showed that total clearance ranged from 0.191 to 1.86 for the ligands, while it was 0.98 for trimethoprim. All the ligands and trimethoprim were not substrates for renal OCT2. Ames's toxicity profiling of the ligands and trimethoprim showed that all the ligands, apart from VIII and trimethoprim, could be carcinogenic. On the other hand, all the ligands did not indicate hepatoxicity concerns, while trimethoprim was predicted to be hepatotoxic. Furthermore, all the ligands, as well as trimethoprim, were not inhibitors of hERG I and II.

Table 2.

Predicted ADMET Properties for Ligands That Meet Lipinski's Rule of Five.

ADMET	Predicted values
	I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII	XIII	XIV
Absorption
Water solubility (log mol/L)	−2.668	−3.568	−2.556	−2.217	−2.385	−2.586	−2.296	−2.25	−4.31	−4.724	−4.073	−4.472	−5.477	−2.744
Caco2 permeability (log Papp in 10⁻⁶ cm/s)	1.229	1.401	1.471	1.376	1.412	1.134	1.502	1.559	1.438	0.849	1.479	1.411	1.565	0.038
Intestinal absorption (human)	97.324	95.898	92.552	95.257	97.555	97.957	94.014	92.041	91.485	91.891	94.296	95.155	92.51	71.787
Skin Permeability (log Kp)	−1.872	−1.721	−1.531	−1.919	−2.131	−1.915	−2.182	−2.207	−2.714	−2.782	−1.923	−1.063	−2.715	−2.735
P-glycoprotein substrate	No	Yes	No	No	No	No	No	No	No	Yes	No	No	No	Yes
P-glycoprotein I inhibitor	No	No	No	No	No	No	No	No	No	No	No	No	No	No
P-glycoprotein II inhibitor	No	No	No	No	No	No	No	No	No	No	No	No	No	No
Distribution
VDss (human) (log L/kg)	0.174	0.396	0.183	0.137	0.182	0.176	0.21	0.24	−0.771	0.812	0.42	0.364	−0.574	0.954
Fraction unbound (human)	0.458	0.48	0.488	0.453	0.526	0.47	0.514	0.251	0.237	0.378	0.28	0.389	0.104	0.554
BBB permeability	0.607	0.732	0.648	0.584	0.585	0.6	0.563	0.374	−0.318	−0.351	0.596	0.792	−0.084	−1.225
CNS permeability	−2.155	−2.37	−2.174	−2.119	−2.478	−2.169	−2.473	−2.007	−2.977	−2.948	−2.151	−1.912	−1.763	−3.273
Metabolism
CYP2D6 substrate	No	No	No	No	No	No	No	No	No	No	No	No	No	No
CYP3A4 substrate	No	No	No	No	No	No	No	No	No	Yes	No	No	Yes	No
CYP1A2 inhibitior	No	No	No	No	No	No	No	Yes	No	No	No	No	No	No
CYP2C19 inhibitior	No	No	No	No	No	No	No	No	No	No	No	No	No	No
CYP2C9 inhibitior	No	No	No	No	No	No	No	No	No	No	No	No	No	No
CYP2D6 inhibitior	No	No	No	No	No	No	No	No	No	No	No	No	No	No
CYP3A4 inhibitior	No	No	No	No	No	No	No	No	No	No	No	No	No	No
Excretion
Total Clearance	0.244	0.213	1.558	1.182	0.273	0.191	1.269	0.282	1.752	0.856	1.085	0.462	1.86	0.98
Renal OCT2 substrate	No	No	No	No	No	No	No	No	No	No	No	No	No	No
Toxicity
AMES toxicity	No	No	No	No	No	No	No	Yes	No	No	No	No	No	Yes
Max. tolerated dose (human)	0.821	0.777	0.723	0.94	0.773	0.843	0.857	1.024	−0.922	−0.685	0.343	0.736	−0.713	0.01
hERG I inhibitor	No	No	No	No	No	No	No	No	No	No	No	No	No	No
hERG II inhibitor	No	No	No	No	No	No	No	No	No	No	No	No	No	No
Oral Rat Acute Toxicity (LD50)	1.796	1.88	1.679	1.946	1.87	1.777	1.811	2.118	1.69	2.968	1.918	1.702	1.449	2.936
Oral Rat Chronic Toxicity (LOAEL)	2.028	2.336	1.981	2.017	1.98	2.068	2.02	2.049	2.939	−1.167	1.475	2.414	3.109	1.135
Hepatotoxicity	No	No	No	No	No	No	No	No	No	No	No	No	No	Yes
Skin sensitization	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes	No
T.Pyriformis toxicity	0.42	0.579	0.207	0.333	0.419	0.439	0.189	0.3	0.907	0.285	1.49	0.924	0.865	0.301
Minnow toxicity	1.205	1.203	1.193	1.29	1.402	1.23	1.545	1.702	0.085	1.127	0.743	0.676	−0.956	2.499

Key: I = Methone, II = Lemonene, III = Octanol, IV = Menthol, V = Dihydrocarvone, VI = Piperitone, VII = Terpinene-4-ol, VIII = Eugenol, IX = 15-Hydroxypentadecanoic acid, X = Carpaine, XI = T-Cadinol, XII = Alpha-ocimene and XIII = hexadecanoic acid.

Figure 3A and B show the summary of the various predicted targets for the various ligands and trimethoprim. The result showed various targets that ranged from 6 to 10 targets for the various bioactive compounds, while for trimethoprim, the number of targets was 7. Figure 4 shows a summary of the top predicted targets for the various ligands and trimethoprim. The top targets were family A G protein-coupled receptors, the fatty acid binding family, oxidoreductase, enzyme, and lyase for the ligands, while for trimethoprim, it was kinase. The most frequent target was the nuclear receptor, which was the most frequent target in six of the 13 bioactive compounds.

Figure 3.

(A) A summary of target prediction results for ligands I to VI. Key: I = Methone, II = Lemonene, III = Octanol, IV = Menthol, V = Dihydrocarvone, VI = Piperitone. (B) A summary of target prediction results for ligands VII to XII and trimethoprim. Key: VII = Terpinene-4-ol, VIII = Eugenol, IX = 15-hydroxypentadecanoic acid, X = Carpaine, XI = T-cadinol, XII = Alpha-ocimene and XIII = Hexadecanoic acid and XIV = Trimethoprim.

Figure 4.

A bar chart presentations of the top targets for each of the ligand and control. Key: M1 to M14 = Methone, Lemonene, Octanol, Menthol, Dihydrocarvone, Piperitone, Terpinene-4-ol, Eugenol, 15-hydroxypentadecanoic acid, Carpaine, T-cadinol, Alpha-ocimene, Hexadecanoic acid, and Trimethoprim, respectively. FAG = Family A G protein coupled receptor, FABF = fatty acid binding family, OX = Oxidoreductase and NR = Nuclear receptor.

Figure 5 shows the various interactions between the various ligands and trimethoprim against the dihydropteroate synthase of E. coli. The various docking poses showed the various bond lengths and amino acid residues that were involved in the interactions between the ligands and the enzymes. Table 3 shows the various amino acids involved in the interactions and their docking scores. The docking scores of the various ligands ranged from −4.0 to −5.3 kcal/mol, and this was lower than that of trimethoprim, which returned a docking score of −6.5 kcal/mol. However, one of the ligands, carpaine, returned a positive docking score of +44.3 kcal/mol, indicating unfavorable docking. The interacting amino acid residues varied among the various amino acids (Figure 5). Furthermore, the most common amino acid residue among the bioactive compounds was isoleucine 117, followed by threonine 62, as both residues were found to be involved in eight and six complexes, respectively. The amino acids also varied in terms of number. Trimethoprim complexed with 9 amino acids, and this was the highest, while the bioactive compounds or test ligands complexed with 2-7 amino acid residues. However, carpaine complexed with a total of 8 amino acids.

Figure 5.

Docking outputs for the various ligands and trimethoprim visualized in 2-D using Biovia Discovery Studio 21. Key: I = Methone, II = Lemonene, III = Octanol, IV = Menthol, V = Dihydrocarvone, VI = Piperitone, VII = terpinene-4-ol, VIII = eugenol, IX = 15-hydroxypentadecanoic acid, X = carpaine, XI = T-cadinol, XII = alpha-ocimene and XIII = hexadecanoic acid and XIV = Trimethoprim.

Table 3.

Amino Residues Involved in Binding of the Various Bioactive Compounds Against Dihydropteroate Synthase and Their Docking Scores.

Bioactive compounds	Amino acid residues	Docking/Vina scores (kcal/mole)
1	Lys 221 and Ile 117	−4.8
2	Arg63 and Ile 117	−4.4
3	Gly217, Phe190, Ile117 and Asp 185	−4.0
4	Asn22 and Ile117	−4.5
5	Arg225, Thr62 and Met 139	−4.7
6	Thr 62, Ile117 and Met 139	−4.8
7	Thr62, His257 and Arg63	−4.8
8	Thr 62, Arg 225, Ile117, Phe190 and Lys221	−5.1
9	Thr62, Asn115	−5.3
10	Ser61, Lys221, Asn22, Met139, Asp105, Asn115, Ile117 and Asp96	44.3
11	Lys221 and Phe190	−4.4
12	His257, Ile20, Met139 and Phe190	−4.4
13	Thr62, Met139, Ile117, Phe190, Lys221, Arg63 and His63	−4.8
Trimethoprim (2,4-diamino-5- (3′,4′,5′-trimethoxybenzyl) pyrimidine)	Asn22, Asp96, Met139, Arg225, Glu 60, Ile20, His257, Lys221 and Asn 115.	−6.5

Key: I = Methone, II = Lemonene, III = Octanol, IV = Menthol, V = Dihydrocarvone, VI = Piperitone, VII = Terpinene-4-ol, VIII = Eugenol, IX = 15-hydroxypentadecanoic acid, X = Carpaine, XI = T-cadinol, XII = Alpha-ocimene and XIII = Hexadecanoic acid and XIV = Trimethoprim.

Discussion

Our study was designed to evaluate the bioactive compounds in Tetrapleura tetraptera (Uyayak) and in-silico assessment of the resulting bioactive compounds. The GC-MS result in our study revealed a total of 28 bioactive compounds that belong to various classes of biochemical compounds. The most abundant compound was monoterpene. In an earlier study, oxygenated monoterpenes were shown to exhibited in vitro antimicrobial and antioxidant effects against food pathogens that included E. coli.³⁰ This findings is in line with a review on the benefits of T. tetraptera where its benefits were highlighted to include antibacterial and antioxidant properties.³¹ The bioactive compounds were further screened for their compliance with the Lipinski rule of five. Out of the twenty-eight (28) bioactive compounds, only 13 met Lipinski's ROF without any violations. According to Lipinski's ROF, an orally active drug should not violate more than one of its criteria (molecular weight < 500; ≤5 H bond donors; ≥10 H bond acceptors; and log P less than 5).³² This implies that all 13 bioactive compounds are potential oral drug candidates.

In addition to Lipinski's ROF, we also evaluated their ADMET properties. ADMET properties are important properties that must be taken into consideration during drug discovery process.³³ Absorption of a drug is a very important consideration in drug discovery, especially for drugs intended for oral consumptio.³⁴ The water solubility of the ligands was higher than that of trimethoprim. Similarly, the intestinal absorption of the bioactive compounds ranged from 91% to 95% compared to that of trimethoprim, which was 71.7%. The ligands further showed variance for the evaluated distribution properties. The result of the BBB permeation indicates that all the ligands could cross the BBB compared to trimethoprim, which is predicted to poorly cross the BBB.³⁵ The efficacy of a lead compound depends a lot on its distribution in various tissues, which explains the lack of relationship between plasma levels and their end biological effects.³⁶ The metabolism of the ligands showed some consistency, as shown by the various cytochrome systems in the liver. None of the ligands, including trimethoprim, were substrates for CYP2D6, while two out of the 13 ligands were substrates for CYP3A4, implying that the majority will not be metabolized easily. However, the bulk of ligands were not inhibitors of the various cytochrome inhibitors, including CYP3A4, which are known to clear certain drugs from the system³⁷ further affecting their bioavailabilities. Toxicity profiling of the ligands indicates an absence of cancer risk as all the ligands were negative for Ames's test except for ligand 8 that was positive,³⁸ Furthermore, an absence of hepatotoxicity for all the lead compounds except the control drug was recorded. The absence of toxicity except for eugenol further indicates that these bioactive compounds can be utilized in making drugs that are safer to humans and also friendly to the environment. This is in line with Díaz et al³⁹ who profiled dihydrocarvone-hybrid derivatives and found them to be nontoxic. In addition to the ADMET properties, the ligands were docked against the dihydropteroate synthase of E. coli.

Molecular docking has emerged as the most utilized in silico or computational aided approach for studying interactions between proteins and other macromolecules with small lead drug compounds.^{15,16,38–41}We docked a total of 13 ligands and trimethoprim against the dihydropteroate synthase of E. coli, and the docking scores for the various ligands ranged from −4.0 to −5.3 kcal/mol while trimethoprim gave a docking score of −6.5 kcal/mol. The higher docking score shown by trimethoprim indicates better binding with dihydropteroate synthase, even though it showed poor absorption and was predicted to be toxic. The scores shown by the ligands in our study are similar to those obtained in an earlier study with bioactive compounds from honey that were docked against the dihydropteroate synthase of S. aureus.¹⁶ The low binding affinities observed by the ligands compared to the standard drug is in contrast to the report of Islam et al⁴² who reported that limonene and its derivatives to have higher binding energies than their utilized control standard drug that ranged from −7.1 to −7.4 kcal/mol against Herpes virus indicating excellent antiviral property. However, it was further observed that the primary limonene had poor binding affinity which improved upon on the addition of functional groups⁴² This implies that the low binding activities observed by the ligands in our study can be improved upon by the addition of suitable functional groups as reported by Islam et al.⁴³ Noumi et al,⁴⁴ in their study further showed excellent binding scores against enzymes involved in biofilm formation that were within range of our reported bioactive compounds. In addition to the docking and predicted pharmacokinetic properties of the ligands, there are reports of biological activities by these bioactive compounds.

Methone have been reported to possess antimicrobial activity against methicillin resistant Staphylococcus aureus (MRSA) via significant membrane disruption,⁴⁵ and antibacterial properties.⁴⁶ Limonene have been reported to possess excellent antibacterial activity against Listeria monocytogenes mediated via the disruption of cell wall integrity, membrane permeability and ATP synthesis.⁴⁷ Similarly, octanol derivatives such as 3,7-dimethyl-1-octanol have been shown to have possess antifungal, anthelmintic, and antioxidant activities but not cytotoxic properties.⁴⁸ Menthol have been shown to possess broad spectrum activity against Gram positive and negative bacteria (E. coli and S. aureus) via the alteration of the membrane integrity.⁴⁹ This implies that following the disruption of the membrane, a potential drug can penetrate the cell further and in the process hinder important cellular activities such as the activity of key enzymes such as dihydropteroate synthase targeted in this study and its important function of synthesizing folic acid.⁴⁹ Furthermore, Díaz et al³⁹ in their study established the antifungal properties of dihydrocarvone-hybrid derivatives while Noumi et al⁴⁴ in their study identified limonene and cis-dihydrocarvone with antibiofilm activity. Barrientos Ramírez et al⁵⁰ showed various derivatives of piperitone possess antibacterial activity against S. aureus and E. coli. Cordeiro et al⁵¹ in their study revealed Terpinen-4-ol to possess remarkable antibiofilm activity even at sub-inhibitory concentrations. Eugenol have been shown to inhibit the growth of both Gram positive and negative bacteria and their biofilms.⁵² On the other hand, Carpaine have been shown to have anti-plasmodial and anti-dengue activities⁵³ while cadinol have been shown to possess antimicrobial, antioxidant and insecticidal properties.⁵⁴ In their study, Hshemi et al showed ocimene to in-vitro antibacterial activity.⁵⁵ These findings further validate the favorable in-silico results of the bioactive compounds. Their favorable binding to dihydropteroate synthase is an indication that these bioactive compounds have the capacity to act as an antimetabolite.

The choice of dihydropteroate synthase was predicated on the fact that only prokaryotes such as bacteria elaborate this important enzyme that is involved in folic acid synthesis, while eukaryotes do not and have to rely on an external supply of folic acid since they lack this enzyme, making it an important drug target.⁵⁶ For a drug to be effective, it has to be able to get to its target. In this case, the target dihydropteroate synthase is an enzyme that belongs to the ligase class of enzymes. From the predicted targets, all the ligands had enzymes as one of their predicted targets. The bioactive compounds, despite their lower docking scores compared to trimethoprim, showed better pharmacokinetics, especially in terms of absorption and toxicity profiles.

Conclusion

Our findings indicate that T. tetraptera is rich with bioactive compounds with diverse monoterpenes and its derivatives that formed the majority of detected compounds. The next most abundant bioactive compound was sesquiterpenes while the most concentrated was lupeol which is triterpenoid ketone. Out of the 28 compounds, 13 passed passed Lipinski's ROF without any violation. Furthermore, ADMET analysis reveals favorable pharmacokinetic properties, especially the absorption, metabolism, and toxicity profiles of the bioactive compounds. The predicted target analysis, showed that the bioactive compounds could target the enzyme dihydropteroate synthase. Docking scores returned favorable binding scores and interactions that were comparable to those of trimethoprim. Put together, the findings indicate that the bioactive compounds from this edible pod, T. tetraptera, could be exploited further as potential lead compounds capable of interfering with the function of dihydropteroate synthase in E. coli.

Footnotes

Acknowledgements

We appreciate the laboratory technologists at Mifor Consult for carrying out the GC-MS.

Author's Contribution

The study was conceptualized and designed by GPB and UOE. All the microbiological analyses were handled by GPB, UOE, JCU, AAOE, and AYO (isolation, identification, and sensitivity testing). UOE, GPB, and ENM handled SMILES string generation, ADMET prediction, ligand and protein preparations, and molecular docking jointly. All the results were jointly validated by all the authors. The initial manuscript draft was done by UOE, GPB, and AYO, while the editing of the manuscript draft was carried out by all the authors. All authors took turns reading the manuscript and approved it for publication.

Availability of Data

All the data generated during the course of this study and are not captured herein will be made available on request.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethics Approval and Consent to Participate

Not applicable.

Statement of Animal Rights

Not applicable as no animal subjects were utilized in this study.

Statement of Informed Consent

Not applicable as there were no human subjects.

Consent to Publication

Not applicable.

ORCID iD

Uwem Okon Edet

References

Waturangi

Hudiono

Aliwarga

. Prevalence of pathogenic Escherichia coli from salad vegetable and fruits sold in Jakarta. BMC Res Notes. 2019;12(1):247. doi:https://doi.org/10.1186/s13104-019-4284-2

Obaji

Oli

Enweani

Udigwe

Okoyeh

JNN

Ekejindu

. Public health challenges associated with street-vended foods and medicines in a developing country: a mini-review. J Med Res. 2018;4(6):283‐287.

Kaper

Nataro

Mobley

HLT

. Pathogenic Escherichia coli. Nat Rev Microbiol. 2004;2(2):123‐140. doi:https://doi.org/10.1038/nrmicro818

WHO. E. coli ; 2018. Accessed February 2, 2023. https://www.who.int/news-room/fact-sheets/detail/e-coli

Poirel

Madec

Lupo

, et al. Antimicrobial resistance in Escherichia coli. Microbiol Spectr. 2018;6(4):10.1128/microbiolspec.ARBA-0026-2017. doi:https://doi.org/10.1128/microbiolspec.ARBA-0026-2017

Boerlin

Travis

Gyles

, et al. Antimicrobial resistance and virulence genes of Escherichia coli isolates from swine in Ontario. Appl Environ Microbiol. 2005;71(11):6753‐6761. doi:https://doi.org/10.1128/AEM.71.11.6753-6761.2005

Hamza

Omran

. Multidrug drug resistance of Escherichia coli and Klebsiella isolated from Iraqi patients and microbiota. J Biosci Med. 2022;10(11):240‐252. doi:https://doi.org/10.4236/jbm.2022.1011019

Kourtis

Sheriff

Weiner-Lastinger

, et al. Antibiotic multidrug resistance of Escherichia coli causing device- and procedure-related infections in the United States reported to the National Healthcare Safety Network, 2013–2017. Clin Infect Dis. 2020;73(11):e4552‐ee4559. doi:https://doi.org/10.1093/cid/ciaa1031

Oyedele

Kuzamani

Adetunji

, et al. Bacteriological assessment of tropical retail fresh-cut, ready-to-eat fruits in south-western Nigeria. Sci Afr. 2020;9:e00505.

10.

Ojewole

Adewunmi

. Antiinflammatory and hypoglycaemic effects of Tetrapleura tetraptera (Taub) [Fabaceae] fruit aqueous extract in rats. J Ethnopharmacol. 2004;95(2-3):177‐182.

11.

Teklu

Negeri

Legese

Bedada

Woldemariam

Tullu

. Extended-spectrum beta-lactamase production and multi-drug resistance among Enterobacteriaceae isolated in Addis Ababa, Ethiopia. Antimicrob Resist Infect Control. 2019;8(1):1‐12.

12.

WHO. WHO publishes list of bacteria for which new antibiotics are urgently needed; 2017. Accessed February 20, 2023. https://www.who.int/news/item/27-02-2017-who-publishes-list-of-bacteria-for-which-new-antibiotics-are-urgently-needed

13.

Chen

Wang

, et al. Drug screening identifies gemcitabine inhibiting rotavirus through alteration of pyrimidine nucleotide synthesis pathway. Antiviral Res. 2020;180:104823. doi:https://doi.org/10.1016/j.antiviral.2020.104823

14.

Rudrapal

Kirboga

Abdalla

Maji

. Explainable artificial intelligence-assisted virtual screening and bioinformatics approaches for effective bioactivity prediction of phenolic cyclooxygenase-2 (COX-2) inhibitors using PubChem molecular fingerprints. Mol Diversity. 2024:1‐20. doi:https://doi.org/10.1007/s11030-023-10782-9

15.

Edet

Nwaokorie

Mbim

, et al. Evaluation of Annona muricata extract against Staphylococcus aureus isolate and in-silico activity of bioactive compounds against Capsular protein (Cap5O). BMC Complement Med Ther. 2022;22(1):1‐9.

16.

Edet

Mbim

Ezeani

, et al. Antimicrobial analysis of honey against Staphylococcus aureus isolates from wound, ADMET properties of its bioactive compounds and in-silico evaluation against dihydropteroate synthase. BMC Complement Med Ther. 2023;23(1):39. doi:https://doi.org/10.1186/s12906-023-03841-z

17.

Lin

Agyemang

Abdel-Samie

MAS

Cui

. Antibacterial mechanism of Tetrapleura tetraptera extract against Escherichia coli and Staphylococcus aureus and its application in pork. J Food Saf. 2019;39(6):e12693.

18.

Kamdem

MHK

Melacheu

GLF

Silihe

, et al. Cytotoxic and antibacterial activities of compounds isolated from the fruits and stem-bark of Tetrapleura tetraptera (Schumach. & Thonn.) Taub.(Fabaceae). Chem Res. 2022;31(11):1948‐1958.

19.

Larbie

Mills-Roberston

Quaicoe

Opoku

Kabiri

ABrokwah

. Tetrapleura tetraptera of Ghanaian origin: phytochemistry, antioxidant and antimicrobial activity of extracts of plant parts. J Pharm Res Int. 2020;32(35):78‐96.

20.

Udobre

Etim

Udoh

Johnson

. Chemical constituents of Tetra Pleura Tetraptera (Fabaceae) and their antimicrobial activities. Niger J Pharm Appl Sci Res. 2018;7(1):52‐55.

21.

Adeoye-Isijola

Naidoo

Coopoosamy

Olajuyigbe

. Antistaphylococcal effects of alcoholic extracts of Tetrapleura tetraptera (Schum and Thonn.)(Taub.) against multidrug methicillin resistant Staphylococcus aureus. J Med Plants Econ Dev. 2022;6(1):122.

22.

Edet

Ebana

Etok

Udoidiong

. Nutrient profile and phytochemical analysis of commercially cultivated oyster mushroom in Calabar, South-South Nigeria. Adv Res. 2016;7(3):1‐6.

23.

Edet

Ebana

Ekanemesang

Ikon

Edem

Mbim

. Phytochemical screening, nutrient analysis, anti-termite and antimicrobial activity of Citrus paradis peel powder. J Appl Life Sci Int. 2016;9(4):1‐9.

24.

Ikon

Udobre

Etang

Ekanemesang

Ebana

Edet

. Phytochemical screening, proximate composition and antibacterial activity of oyster mushroom, Pleurotus ostreatus collected from Etim Ekpo in Akwa Ibom state, Nigeria. Asian Food Sci J. 2018;6(2):1‐10.

25.

Ebana

RUB

Edet

Ekanemesang

Ikon

Etok

Edet

. Antimicrobial activity, phytochemical screening and nutrient analysis of Tetrapleura tetraptera and Piper guineense. Asian J Med Health. 2016;1(3):1‐8.

26.

Ebana

RUB

Andy

Edet

Benjamin

Mbim

Anosike

. Nutritional studies and antimicrobial activities of Jatropha tanjorensis leaves extracts against Escherichia coli isolates. Int J Innov Sci Res Technol. 2019;4(8):945‐955.

27.

Pires

Blundell

Ascher

. pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem. 2015;58(9):4066‐4072. doi:https://doi.org/10.1021/acs.jmedchem.5b00104

28.

Daina

Michielin

Zoete

. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017;7(1):42717. doi:https://doi.org/10.1038/srep42717

29.

Trott

Olson

. Autodock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010;31(2):455‐461.

30.

Badawy

Marei

GIK

Rabea

Taktak

. Antimicrobial and antioxidant activities of hydrocarbon and oxygenated monoterpenes against some foodborne pathogens through in vitro and in silico studies. Pestic Biochem Physiol. 2019;158:185‐200.

31.

Mensah

Adusei

Azupio

Kwakye

. Nutritive value, biological properties, health benefits and applications of Tetrapleura tetraptera: an updated comprehensive review. Heliyon. 2024;10(6):e27834.

32.

Lipinski

. Lead-and drug-like compounds: the rule-of-five revolution. Drug Discov Today: Technol. 2004;1(4):337‐341.

33.

Cheng

Zhou

, et al. admetSAR: a comprehensive source and free tool for assessment of chemical ADMET properties. J Chem Inf Model. 2012;52(11):3099‐3105. doi:https://doi.org/10.1021/ci300367a

34.

Subramanian

Kitchen

. Computational approaches for modeling human intestinal absorption and permeability. J Mol Model. 2006;12:577‐589. doi:https://doi.org/10.1007/s00894-005-0065

35.

Kunwittaya

Nantasenamat

Treeratanapiboon

Srisarin

Isarankura-Na-Ayudhya

Prachayasittikul

. Influence of logBB cut-off on the prediction of blood-brain barrier permeability. Biomed Appl Technol J. 2013;1:16‐34.

36.

Onetto

Sharif

. Drug distribution. In: StatPearls. StatPearls Publishing; 2021.

37.

Susa

Preuss

. Drug metabolism. In: StatPearls. StatPearls Publishing; 2022.

38.

Cheng

Chen

, et al. In silico prediction of chemical Ames mutagenicity. J Chem Inf Model. 2012;52(11):2840‐2847.

39.

Díaz

Werner

Besoain

, et al. In vitro antifungal activity and toxicity of dihydrocarvone-hybrid derivatives against Monilinia fructicola. Antibiotics (Basel, Switzerland). 2021;10(7):818. doi:https://doi.org/10.3390/antibiotics10070818

40.

Mboto

Edet

Mbim

, et al. In-silico evaluation of bioactive compounds from selected medicinal plants from Southern Nigeria against hepatitis C virus genotype 1 RNA-directed RNA polymerase. Sci Afr. 2023;22:e01919.

41.

Iyam

Ogbodo

Okafor

, et al. Elucidating the antibacterial efficacy of thiadiazol derivative against carbapenem-resistant Klebsiella pneumoniae and Pseudomonas aeruginosa: an in-silico perspective. Chem Phys Impact. 2024;8:100466.

42.

Yuriev

Ramsland

. Latest developments in molecular docking: 2010-2011 in review. J Mol Recognit. 2013;26(5):215‐239. doi:https://doi.org/10.1002/jmr.2266

43.

Islam

Islam Sovon

Amena

, et al. Ligand-based drug design against Herpes Simplex Virus-1 capsid protein by modification of limonene through in silico approaches. Sci Rep. 2024;14(1):9828. doi:https://doi.org/10.1038/s41598-024-59577-4

44.

Noumi

Ahmad

Adnan

, et al. GC/MS profiling, antibacterial, anti-quorum sensing, and antibiofilm properties of Anethum graveolens L. essential oil: molecular docking study and in-silico ADME profiling. Plants (Basel, Switzerland). 2023;12(10):1997. doi:https://doi.org/10.3390/plants12101997

45.

Zhao

Yang

Zhang

, et al. Menthone exerts its antimicrobial activity against methicillin resistant Staphylococcus aureus by affecting cell membrane properties and lipid profile. Drug Des Devel Ther. 2023;17:219‐236. doi:https://doi.org/10.2147/DDDT.S384716

46.

Zaia

Cagnazzo

Feitosa

, et al. Anti-inflammatory properties of menthol and menthone in Schistosoma mansoni infection. Front Pharmacol. 2016;7:170. doi:https://doi.org/10.3389/fphar.2016.00170

47.

Han

Sun

Chen

. Antimicrobial susceptibility and antibacterial mechanism of limonene against Listeria monocytogenes. Molecules (Basel, Switzerland). 2019;25(1):33. doi:https://doi.org/10.3390/molecules25010033

48.

Coêlho

Islam

da Silva Oliveira

, et al. Cytotoxic and antioxidant properties of natural bioactive monoterpenes nerol, estragole, and 3, 7-dimethyl-1-octanol. Adv Pharmacol Pharm Sci. 2022;2022:1‐11.

49.

Trombetta

Castelli

Sarpietro

, et al. Mechanisms of antibacterial action of three monoterpenes. Antimicrob Agents Chemother. 2005;49(6):2474‐2478. doi:https://doi.org/10.1128/AAC.49.6.2474-2478.2005

50.

Barrientos Ramírez

Silva Guzmán

Osorio Muñoz

, et al. Chemical composition, antioxidant properties, and antibacterial activity of essential oils of Satureja macrostema (Moc. and Sessé ex Benth.) Briq. Molecules. 2023;28(12):4719. doi:https://doi.org/10.3390/molecules28124719

51.

Cordeiro

Figueiredo

Souza

, et al. Terpinen-4-ol as an antibacterial and antibiofilm agent against Staphylococcus aureus. Int J Mol Sci. 2020;21(12):4531. doi:https://doi.org/10.3390/ijms21124531

52.

Bai

Chen

, et al. Antibacterial activity and mechanism of clove essential oil against foodborne pathogens. Lwt. 2023;173:114249. doi:https://doi.org/10.1016/j.lwt.2022.114249

53.

Shrivastava

Alagarasu

Cherian

Parashar

. Antiviral & platelet-protective properties of Carica papaya in dengue. Indian J Med Res. 2022;156(3):459‐463. doi:https://doi.org/10.4103/ijmr.ijmr_2406_21

54.

Tourabi

Nouioura

Touijer

, et al. Antioxidant, antimicrobial, and insecticidal properties of chemically characterized essential oils extracted from Mentha longifolia: in vitro and in silico analysis. Plants (Basel, Switzerland). 2023;12(21):3783. doi:https://doi.org/10.3390/plants12213783

55.

Hashemi

Ehsani

Hosseini Jazani

Aliakbarlu

Mahmoudi

. Chemical composition and in vitro antibacterial activity of essential oil and methanol extract of Echinophora platyloba D.C against some of food-borne pathogenic bacteria. Vet Res Forum: Int Q J. 2013;4(2):123‐127.

56.

Hevener

Zhao

Ball

, et al. Validation of molecular docking programs for virtual screening against dihydropteroate synthase. J Chem Inf Model. 2009;49(2):444‐460. doi:https://doi.org/10.1021/ci800293n

Evaluation of Bioactive Compounds in Tetrapleura tetraptera (Uyayak) and Its Activity Against Dihydropteroate Synthase (1AJ2) of E. coli : In-Sight from ADMET Profiling and Molecular Docking

Abstract

Keywords

Background

Methods

Collection of Fruit, Fruit Salad and T. tetraptera Samples

GC-MS Analysis of T. tetraptera

Prediction of ADMET Properties of the Ligands and Target Prediction

Lipinski Rule of Five (ROF) Screening

Ligand Preparation for Docking

Preparation of Proteins for Molecular Docking

Molecular Docking and Visualization

Results

Discussion

Conclusion

Footnotes

Acknowledgements

Author's Contribution

Availability of Data

Declaration of Conflicting Interests

Funding

Ethics Approval and Consent to Participate

Statement of Animal Rights

Statement of Informed Consent

Consent to Publication

ORCID iD

References