This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
Porphyromonas gingivalis is a Gram-negative, anaerobic bacilli found in the oral cavity and is commonly associated with oral and gingival diseases destructing the supportive structures of the tooth [1]. P. gingivalis is known to produce various virulence factors, viz., FimA and cysteine proteinases, toward colonization in mucosal tissues, host immune modulation [2]. Prolyl tripeptidyl peptidase coded by ptp-A gene is one such serine protease that plays a vital role in P. gingivalis virulence in the host tissues. ptp-A gene is known to encode a 82 kD protein possessing specific motif characteristic for S9 prolyl oligopeptidase family of serine proteases and belong to subtilisin family. ptp-A contributes its role in the P. gingivalis pathogenesis toward the periodontal tissue destruction with the underlying mutual interaction of ptp-A with the host and bacterial collagenases and dipeptidyl peptidases during the collagen degradation during the infection [3]. ptp-A working in a concert manner with its known similar collagen destructors inside the tissues would be the best target for novel drug candidates.
With the experimental proved fact that natural herbs could eliminate harmful proteinases from bacterial pathogens and could aid in reducing inflammation and irritation of periodontal ligaments, Rosmarinus officinalis is one such miraculous herb with potent bioactivities against chronic infections. R. officinalis is known to possess many phytocompounds such as diterpenes, carnosic acid, carnosol, and rosmanol. Ethanolic extracts of R. officinalis is known to possess vital antibacterial activity against Methicillin-resistant Staphylococcus aureus (MRSA) strains of Staphylococcus aureus [4]. The potent in vitro antibacterial activity was also studied and documented against Leuconostoc sp., Listeria sp., Streptococcus mutans, and Bacillus cereus and bacteriostatic action was reported against Penicillium and Botrytis sp. [5]. However, no proper documentation proves the role of the phytocompounds, viz., carnosic acid, rosmarinic acid, p-coumaric acid, and luteolin from R. officinalis against the ptp-A of P. gingivalis. The present investigation is thus aimed to assess the inhibitory effect of the selected phytocompounds from R. officinalis against ptp-A proteinase of P. gingivalis using bioinformatics tools and to evaluate the docking scores by AutoDock Tools (ADTs).
Methods
Retrieval of ptp-A and protein optimization
The crystal structure of prolyl tripeptidyl peptidase (ptp-A) from P. gingivalis was retrieved from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (http://www.rcsb.org/pdb). Hydrogen atoms, solvation parameters, and fragmental volumes to the protein were added, and Gasteiger charges were assigned to the protein atoms using Kollman united atoms force field by using ADT – 1.5.6.
Ligand preparation and optimization
Using ChemSketch software, the structures of the carnosic acid, rosmarinic acid, p-coumaric acid, and luteolin from R. officinalis L. (common name: Rosemary) and amoxicillin were drawn together with the generation of their three-dimensional (3D) structures and optimization. The selected ligands were retrieved in source database (SDB) format, which were further saved in .mol file followed by the subsequent conversion using Open Babel molecular converter program [6] and in the protein data bank (PDB) format.
Molinspiration assessment of the molecular properties of the selected compounds
The physiochemical and the pharmacological properties such as logP, hydrogen bond donor and acceptor characteristics, molecular size, and rotatable bonds were predicted by molinspiration server [7]. Based on the Lipinski’s rule of 5 [8], characterization of the absorption, distribution, metabolism, and elimination of the selected compounds with further assessments and estimations of the molecular properties of the selected ligands was assessed. Membrane permeability and bioavailability was also evaluated.
Docking simulations and interpretations
The docking analysis to interpret the affinity between carnosic acid, rosmarinic acid, p-coumaric acid, luteolin and the control amoxicillin against ptp-A of P. gingivalis was achieved by AutoDock tool with the intermediary steps such as pdb. qt files for the proteins and the ligands. Using graphical user interface program ADT, the grid box creation was completed. Prior preparation of the grid map using the grid box with a grid size of 126 × 126 × 126 xyz points was done. Further using Lamarckian genetic algorithm (LGA), docking simulation was achieved by setting the initial position, orientation, and torsions of the ligand molecules in a random position. 10 different runs set to terminate after a maximum of 250000 energy evaluations was used for each docking experiment with the population size set at 150. A translational step of 0.2 Å and quaternion and torsion steps of 5 were applied for each dock. The most favorable free energy of binding was achieved by clustering the results >1.0 Å in positional root-mean-square deviation [9].
Docking visualization
The protein–ligand interactions like hydrogen bonding and other nonbonded energies between the carnosic acid, rosmarinic acid, p-coumaric acid, luteolin, and the control amoxicillin against ptp-A of P. gingivalis were visualized using the Biovia Discovery Studio Visualizer software. The relative stabilities were evaluated using their molecular dynamics, binding affinities, and energy simulations with further docking score assessments.
Results
Structure retrieval of the ptp-A protein from P. gingivalis
The crystal structure of prolyl tripeptidyl peptidase from P. gingivalis (strain SC5314) was downloaded from the PDB database and its structure id was documented as 2D5LA-Chain. Removal of the water molecules and final-stage merging of hydrogen atoms to the receptor molecule were successful. The 3D structure of ptp was visualized using RasMol with the analysis of pink color indicating the alpha helix, yellow arrow indicating the beta sheets, and white color indicating the turns (Figure 1).
Figure 1
3D structure of ptp-A as visualized by RasMol (helix, pink color; beta sheet, yellow; turns, white). 3D, three dimensional
Structure retrieval of the R. officinalis compounds (the ligands)
The ligand optimization was achieved using ACD ChemSketch and retrieved in a compatible format using Open Babel molecular converter tool. The retrieved structures of the ligands are shown in Figure 2.
Figure 2
Structure of the selected biocompounds from R. officinalis
Molinspiration estimation toward drug likeliness
The prediction of bioactivity scores of carnosic acid, rosmarinic acid, p-coumaric acid, luteolin, and the control amoxicillin against ptp-A of P. gingivalis based on the calculation toward drug likeliness is scored and tabulated in Table 1. Molecular properties were calculated based on the Lipinski’s rule of 5 and its components. Topological polar surface area (TPSA) was <140 Å for all the compounds, thus indicating its higher absorption and promising oral bioavailability (Table 1).
Molinspiration calculations of Rosmarinus officinalis biocompounds and drug likeliness of the R. officinalis against ptp-A of Porphyromonas gingivalis
Compounds
Molecular weight
Molecular formula
Hydrogen bond donor
Hydrogen bond acceptor
LogP
Rotatable bonds
TPSA
Volume
N atoms
G protein-coupled receptors ligand
Ion channel modulator
Kinase inhibitor
Nuclear receptor ligand
Protease inhibitor
Enzyme inhibitor
Carnosic acid
332.44
C20H28O4
3
4
4.60
2
77.75
322.31
24
0.41
0.24
−0.24
0.72
−0.01
0.32
Rosmarinic acid
360.318
C18H16O8
5
8
1.63
7
144.52
303.54
26
0.17
−0.08
−0.18
0.57
0.15
0.24
p-Coumaric acid
164.16
C9H8O3
2
3
1.43
2
57.53
146.48
12
−0.56
−0.26
−0.91
−0.12
−0.87
−0.15
Luteolin
286.239
C15H10O6
4
6
1.97
1
111.12
232.07
21
−0.02
−0.07
0.26
0.39
−0.22
0.28
Amoxicillin
365.404
C16H19N3O5S
5
8
−1.32
4
132.96
306.89
25
0.07
−0.42
−0.65
−0.47
0.84
0.27
TPSA, topological polar surface area.
Docking analysis of the R. officinalis compounds against ptp-A protein from P. gingivalis
The best conformers were selected using LGA based on the best ligand–receptor structure from the docked structure based on the lowest energy and minimal solvent accessibility. Biovia discovery studio visualizing tool of the hydrogen bond interactions in stick model between the carnosic acid, rosmarinic acid, p-coumaric acid, luteolin and the control amoxicillin against ptp-A of P. gingivalis is given in Figure 1. The amino acids of ptp-A binding with the bioactive compounds rosmarinic acid and luteolin showed the lowest estimated free binding energy of −9.81 kcal/mol with 10 hydrogen bond interactions and −9.99 kcal/mol with 7 hydrogen bond interactions, respectively, followed by carnosic acid and p-coumaric acid that showed a binding energy of −7.14 kcal/mol with 5 hydrogen bond interactions and −6.34 kcal/mol with 5 hydrogen bond interactions, respectively. Control amoxicillin showed a binding energy of −7.18 kcal/mol with only 5 hydrogen bond interactions. The torsional energy and the docking scores between the drug and ligands are given in Table 2. The other interactions such as van der Waals, π–σ interactions/π–π T-shaped interactions/amide–π stacked interactions, alkyl/π–alkyl interactions, and salt bridge/attractive charge are shown in Table 3 and Figure 3.
Docking results of R. officinalis biocompounds with ptp-A of P. gingivalis (energy in kcal/mol)
ptp-A docking with compounds
Number of hydrogen bonds
Binding energy
Ligand efficiency
Intermolecular energy
vdW + Hbond + desolv energy
Electrostatic energy
Torsional energy
Total internal unbound
Carnosic acid
5
−7.19
−0.3
−8.68
−7.56
−1.11
1.49
−1.38
Rosmarinic acid
10
−9.81
−0.38
−12.2
−11.88
−0.31
2.39
−2.23
p-Coumaric acid
5
−6.34
−0.53
−7.53
−6.72
−0.81
1.19
−0.07
Luteolin
7
−9.99
−0.48
−11.48
−11.47
−0.02
1.49
−1.37
Amoxicillin
4
−7.18
−0.29
−9.26
−7.3
−1.97
2.09
−2.12
Interactions of R. officinalis biocompounds with ptp-A of P. gingivalis
Biovia discovery studio visualization of the hydrogen interactions between ptp-A and (A) carnosic acid, (B) rosmarinic acid, (C) p-coumaric acid, (D) luteolin, and (E) amoxicillin
Discussion
Prolyl tripeptidyl peptidase (ptpA) in P. gingivalis is indicated as 1 among the 7 different enzymatic enzymes playing a vital role in the pathogenesis of periodontitis. ptp-A falls under the cysteine and serine catalytic classes of peptidases playing a vital role in periodontal infections [10]. In the control of periodontal infection by P. gingivalis, ptp-A can be considered as the best target for novel natural bioactive compounds. R. officinalis and its biocompounds, being phytochemically rich in tannins, flavonoids, and alkaloids, are best known for their antibacterial potential [11]. However, the best fit of the biocompounds with the ptp-A of P. gingivalis was efficiently achieved in the present investigation by molecular docking analysis, which is a molecular modeling technique. ptp-A was retrieved from the PDB database as a desirable target based on the data recorded in database and was freely accessible. In the present study, carnosic acid, rosmarinic acid, p-coumaric acid, luteolin, and the control amoxicillin docked against ptp-A of P. gingivalis resulted in a promising receptor–ligand complex. Docking analysis involves 2 major steps where the prediction of the exact orientation of the conformers into the best active site pocket called pose and the strength of target–ligand binding interactions is achieved by scoring [12]. Analysis using Biovia Discovery Studio Visualizer to predict hydrogen bond interactions between ptp-A of P. gingivalis and the ligands yielded promising results with hydrogen bonds and bonding energies. The number of hydrogen bonds together with the enthalpic gain due to the water molecules determines the best fit [13]. In this context, rosmarinic acid scores to be the best inhibitory agent of ptp-A of P. gingivalis with a docking score of −9.81 kcal/mol with 10 hydrogen bonds, albeit amoxicillin scored only 4 hydrogen bonds with a score of −7.18 kcal/mol.
We performed molinspirational calculations in the present study to assess and evaluate the drug likeliness of the selected ligands. This is due to the fact that molecular properties such as membrane permeability, hydrophobicity, and bioavailability are associated with some basic molecular descriptors such as log P (partition coefficient), log S (solubility), molecular weight, number of hydrogen bond acceptors and donors in a molecule highly attribute in designing novel drugs [14]. In the present study, molinspirational results were very promising for carnosic acid, rosmarinic acid, p-coumaric acid, and luteolin in comparison with the control amoxicillin suggesting the promising ptp-A inhibitory activity of the selected compounds from R. officinalis.
In molinspiration analysis, TPSA of a molecule is considered as a useful descriptor to characterize the drug absorption and bioavailability and the values of TPSA and OH–NH interactions indicate that the selected ligands, viz, cinnamaldehyde, cinnamyl acetate, and eugenol, possess a smooth and efficient binding to the target proteins. However, the drug molecules with TPSA values of >140 Å or higher have low absorption with the lipophilicity (miLogP) playing their vital role in the prediction of the oral bioavailability for the drugs. In this context, all the ligands score high absorption with high membrane penetration with a TPSA score of <140 Å.
We used ADT, which is considered as a suite of automated docking tool with software for modeling flexible small molecules such as drug molecule binding to receptor proteins of known 3D structure. ATD 2.0 newer version generates 10 conformations (poses) of the protein–ligand complex, which are displayed from the lowest to the highest binding free energy (ΔG). A computational docking algorithm was used to predict the relative binding affinities for ptp-A with carnosic acid, rosmarinic acid, p-coumaric acid, luteolin from R. officinalis, and control amoxicillin, and observation of structure–inhibitory action relationships was achieved. In this study, the LGA was used to explore the binding conformational landscape of carnosic acid, rosmarinic acid, p-coumaric acid, luteolin, and the control amoxicillin docked against ptp-A of P. gingivalis. The docking scores on ptp-A indicated that there is a direct relationship between the energy of the binding affinity, referring to the lowest docking scores and the stability. In accordance with this, apart from the binding energy, the intermolecular energy, vanderwaal’s energy, and torsional energy were also at a higher end for rosmarinic acid and luteolin followed by carnosic acid (Table 2). Thus, theoretically, it was evident that rosmarinic acid, luteolin, carnosic acid, and p-coumaric acid from R. officinalis exhibited the highest ptpA inhibition activity.
Conclusion
Novel selection of inhibitors against specific target protein ptp-A in P. gingivalis is the need of the hour, which was best achieved by computational assessments. The docking calculations in this study suggest rosmarinic acid, luteolin, carnosic acid, and p-coumaric acid as promising candidates in targeting periodontitis infection caused by P. gingivalis. The preliminary clue obtained from the present investigation alarms for further target-based experimental screening of the R. officinalis biocompounds for better selectivity and mechanism of action.
