Intermolecular interactions between Remdesivir, EBDGp and SARS-CoV-2 RdRp
The detailed intermolecular interactions analysis is summarized in Table 2. The binding energy obtained for Remdesivir with SARS-CoV-2 RdRp is -19.36 kcal/mol. The analysis of intermolecular interaction between Remdesivir and SARS-CoV-2 RdRp is showing hydrogen bond interactions with the reported crucial residues Thr556 and Asp623 of SARS-CoV-2 RdRp [11, 66] with the bond distance of 2.22 Å and 2.08 Å respectively. As per the observation of docking interactions shown in Fig. 2a, the -NH group adjacent to the Phosphate group in the structure of Remdesivir is forming hydrogen bond interaction with the carboxylic oxygen atom of Asp623 amino acid of SARS-CoV-2 RdRp. The carboxylic oxygen atom of Thr556 is hydrogen-bonded with the hydroxyl group of Remdesivir. Hydrogen bonding interactions are also observed with Asp452, Cys622, and Arg624 of SARS-CoV-2 RdRp.
The binding energy obtained for EBDGp with SARS-CoV-2 RdRp is -23.32 kcal/mol. The intermolecular interaction of EBDGp with SARS-CoV-2 RdRp is showing hydrogen bond interactions with the same crucial amino acids Thr556 and Asp623 as observed with Remdesivir. The carboxylic oxygen atom of amino acids Thr556 and Asp623 is forming hydrogen bond interactions with the hydroxyl group of EBDGp as shown in Fig. 2b with the bond distance of 1.99 Å and 1.60 Å respectively. Thr556 of SARS-CoV-2 RdRp is forming similar molecular interactions with both Remdesivir and EBDGp. It can be deduced from the above results that EBDGp is showing a similar mode of interactions with amino acid residues Thr556 and Asp623 as Remdesivir despite Asp623 is forming interactions with Remdesivir and EBDGp with different functional groups but forming the similar type of intermolecular interactions i.e. Hydrogen bond. Additionally, Asp452, Arg555, Tyr619, Cys622, Thr687, and Asp760 are also involved in hydrogen bond interactions with EBDGp.
Table 2 Binding affinity of phytocompounds with the target SARS-CoV2 RdRp.
Sr.no.
|
Compound Name
|
Compounds CID
|
Binding energy
(kcal/mol)
|
Interacting Residues
|
Bond Type
|
Bond Distance (Å)
|
1.
|
|
121304016
|
-19.36
|
Thr556*
|
H bond
|
2.22
|
Remdesivir
|
Arg624
|
H bond
|
2.09, 1.92
|
Asp623*
|
H bond
|
2.08
|
Cys622*
|
H bond
|
2.07
|
Asp452
|
H bond
|
1.83
|
2.
|
|
10930068
|
-23.32
|
Thr556*
Arg555
|
H bond
π-cation
|
1.99
6.57
|
|
Asp623*
|
H bond
|
1.60
|
|
Asp452
|
H bond
|
1.76, 1.68
|
EBDGp
|
Cys622*
|
H bond
|
2.33
|
|
Tyr619
|
H bond
|
2.06
|
|
Asp760
|
H bond
|
1.49, 2.07
|
|
Thr687
|
H bond
|
2.09, 2.76
|
3.
|
|
10212
|
-10.30
|
Thr556*
|
H bond
|
2.02
|
Marmelide
|
Arg624
|
H bond
|
1.79
|
4.
|
|
68229
|
-10.1
|
Lys621
|
π-Cation
|
5.85
|
Seselin
|
Arg553
|
π-Cation
|
3.04
|
Cys622*
|
H bond
|
1.98
|
5.
|
Pedunculagin
|
442688
|
-1.57
|
Cys622*
|
H bond
|
1.91
|
Lys621
|
H bond
|
2.39
|
Arg553
|
H bond
|
1.62
|
Thr556*
|
H bond
|
1.57, 2.05
|
Asp760
|
H bond
|
1.78
|
6.
|
|
442674
|
1.64
|
Arg555
|
H-bond
|
2.26
|
Chebulagic acid
|
Arg553
|
π-cation
Salt bridge
|
4.30
2.87
|
Arg624
|
H bond
Salt bridge
|
2.10
3.08
|
Asp623*
|
H bond
|
0.59
|
Cys621
|
H bond
|
2.76
|
Asp760
|
H bond
|
1.69
|
Tyr619
|
H bond
|
2.60
|
*Common Interacting Residues with SARS-CoV-2 RdRp
Intermolecular interactions between Marmelide, Seselin and SARS-CoV-2 RdRp
The Marmelide shows the binding energy of -10.30 kcal/mol which is observed to be lower than that of Remdesivir and EBDGp. Interactions of Marmelide with SARS-CoV-2 RdRp is shown in Fig. 3a. Hydroxyl group of Marmelide shows to form hydrogen bond interaction with an amino group of Thr556 (2.02 Å) and the secondary amino group of Arg624 with a bond distance of 1.79 Å. Six carbon aromatic ring of Seselin is forming a π-cation bond with the secondary amino group of Arg553 (3.04 Å) and the side-chain amino group of Lys621 (5.85 Å) and backbone amino group of Cys622 (1.98 Å) forms a hydrogen bond with the carbonyl oxygen atom of the Seselin (Fig. 3b).
Intermolecular interactions between Pedunculagin, Chebulagic acid and SARS-CoV-2 RdRp
Pedunculagin and Chebulagic acid show the lowest binding energies (-1.57 and 1.64 kcal/mol respectively) among all the docked compounds. Interactions of Pedunculagin and Chebulagic acid with SARS-CoV-2 RdRp is shown in Fig. 4a and 4b respectively. Hydroxyl groups of Pedunculagin forms two hydrogen bonds with a carboxylic group of crucial amino acid i.e. Thr556 with a bond distance of 1.57 Å and 2.05 Å. Carbonyl oxygen of Pedunculagin forming a hydrogen bond with an amino group of Cys622 with a bond distance of 1.91 Å. Besides, Arg553 and Lys621 are also forming hydrogen bond interactions with Pedunculagin. Chebulagic acid shows hydrogen bond interactions with amino acids viz. Arg553, Arg555, Lys621, Arg624, Thr619, Asp623, and Asp760. The carboxylic group of crucial amino acid Asp623 makes hydrogen bond interaction with the hydroxyl group of Chebulagic acid.
Dynamic behavior of Remdesivir and lead compounds with SARS-CoV-2 RdRp
To evaluate the dynamic behavior, 100ns simulation runs for the docked compounds including Remdesivir in the defined binding pocket of SARS-CoV-2-RdRp was carried out. The information about the structural stability of the protein-ligand complex could be analyzed by RMSD. RMSD calculations were performed using changes in C-alpha atoms of SARS-CoV-2-RdRp in complex with docked phytocompounds.
In each complex, it appears that stable equilibrium was reached after 5ns. The RMSD was observed within 4.5 Å RMSD for the receptor in complex with Seselin, Marmelide, Pedunculagin, and Chebulagic acid throughout the simulation. These compounds have been shown to have lower RMSD values as compared to Remdesivir. Among these compounds, EBDGp was observed to have the lowest RMSD value below 1.5 Å where Remdesivir deviates within the RMSD range of 1.5 Å to 2.0 Å suggesting higher stability as compared to other phytocompounds and Remdesivir (Fig. 5b). On the other hand, it can be seen that the receptor is least stable when in complex with Remdesivir, as shown by its highest RMSD (Fig. 5a). The Ligand RMSD analysis shows that there is a less deviation of EBDGp (RMSD-1.25 Å) from the binding pocket of the receptor thus showing its role in overall higher stability of SARS-CoV-2 RdRp as compared to Remdesivir (RMSD-2.25 Å). To further understand the dynamics of the backbone atoms, the root mean square fluctuation (RMSF) values were calculated for backbone atoms at each point of the trajectories. Higher RMSF values indicate greater flexibility during the MD simulation [86].
Low RMSF values (< 1.5 Å) of active site residues for all SARS-CoV-2 RdRp-complexes indicate their higher stabilities during the entire MD simulation (Fig. 6), signifying that there are no major conformational changes seen in the binding pocket of the SARS-CoV-2 RdRp in complex with all the compounds. Also, RMSF values for the binding pocket residues (C-α atoms) were summarized in Table3. The RMSF of the SARS-CoV-2 RdRp binding site residues, upon binding of lead compounds i.e. EBDGp, Marmelide Seselin Pedunculagin, and Chebulagic acid is lower than 1 Å, which suggest that the SARS-CoV-2 RdRp binding pocket is more stable with minimum fluctuation during the 100ns MD simulation.
Table 3 RMSF values of the amino acids (C-α atoms) which are involved in the SARS-CoV-2 RdRp binding pocket after binding of lead compounds and Remdesivir.
Residues
|
Remdesivir
|
EBDGp
|
Marmelide
|
Seselin
|
Pedunculagin
|
Chebulagic acid
|
Phe441
|
1.573
|
0.728
|
0.548
|
0.615
|
0.619
|
0.468
|
Asp452*
|
1.115
|
0.756
|
0.535
|
0.652
|
0.832
|
0.907
|
Tyr455
|
0.898
|
0.777
|
0.481
|
0.593
|
0.503
|
0.981
|
Tyr456
|
0.766
|
0.776
|
0.660
|
0.572
|
0.558
|
1.947
|
Lys545
|
0.944
|
0.714
|
0.427
|
0.579
|
0.587
|
0.801
|
Ala547
|
1.132
|
0.637
|
0.487
|
0.496
|
0.496
|
0.693
|
Arg553*
|
1.240
|
0.845
|
0.874
|
0.658
|
0.595
|
0.958
|
Arg555*
|
0.880
|
0.713
|
0.674
|
0.503
|
0.475
|
0.615
|
Thr556*
|
0.867
|
0.794
|
0.556
|
0.728
|
0.503
|
0.630
|
Ala558
|
0.605
|
0.676
|
0.602
|
0.450
|
0.590
|
0.974
|
Tyr619*
|
0.794
|
0.635
|
0.625
|
0.897
|
0.541
|
0.498
|
Lys621*
|
0.676
|
0.982
|
0.575
|
0.908
|
0.626
|
0.664
|
Cys622*
|
0.720
|
0.735
|
0.546
|
0.784
|
0.520
|
0.825
|
Asp623*
|
0.943
|
0.655
|
0.465
|
0.705
|
0.431
|
0.714
|
Arg624*
|
0.634
|
0.973
|
0.501
|
0.618
|
0.771
|
0.551
|
Thr680
|
0.508
|
0.859
|
0.437
|
0.620
|
0.706
|
1.030
|
Thr687*
|
0.683
|
0.498
|
0.554
|
0.536
|
0.445
|
0.762
|
Asp760*
|
0.962
|
0.759
|
0.514
|
1.536
|
0.429
|
0.482
|
*indicates interacting residues of SARS-CoV-2 RdRp with lead compounds and Remdesivir.
Intermolecular interaction profile between Remdesivir, lead compounds, and SARS-CoV-2 RdRp.
To understand the binding pocket stability, the MD trajectories captured for all systems were superimposed and analyzed using the Simulation Event Analysis tool of Desmond. Fig. 7 and 8 shows, interacting residues of SARS-CoV-2 RdRp with the Remdesivir and lead compound during 100ns MD simulation. Some residues make more than one specific contact with the ligand, which is represented by a darker shade of orange, according to the scale to the right of the plot. Remdesivir, Pedunculagin, EBDGp, and Chebulagic acid are showing good interactions with some active site residues.
Hydrogen bond interaction analysis of the Remdesivir and phytocompounds with SARS-CoV-2 RdRp.
To reveal the binding stability between SARS-CoV-2 RdRp and phytocompounds like EBDGp, Marmelide, Seselin, Pedunculagin, and Chebulagic acid; hydrogen bond monitoring were done using the resulting MD trajectories from 100ns simulation via Simulation Event Analysis module of Maestro. The plots of the hydrogen bonding profile are presented in Fig. 9. In Fig. 9a, it is observed that the Remdesivir is making 8 hydrogen bonds with the active site residues of SARS-CoV-2 RdRp such as Asp623 and Asp760 throughout the 100 ns MD simulation. Interestingly, EBDGp is making the highest contacts i.e. about 10 hydrogen bonds as shown in Fig. 9b till 25ns and is observed to maintain 8-10 hydrogen bond contacts with the active site residues such as Asp452, Cys622, Asp623, and Thr680 throughout the 100 ns MD simulation, which is observed to be the highest as compared to a reference compound, Remdesivir. Whereas, other phytocompounds such as Marmelide, Seselin, Pedunculagin, and Chebulagic acid are shown to have the least hydrogen bond contacts as compared to all the other phytocompounds as can be seen in Fig. 9c, 9d, 9e, and 9f respectively. From these results, we can conclude that the screened lead compound EBDGp forms a stable complex with SARS-CoV-2 RdRp and thus obtain the complex stability during the 100 ns MD simulation. It is observed that EBDGp is making the highest hydrogen bond contacts for a longer period with most of the active site residues as compared to Remdesivir and other phytocompounds which shows its greater potential in inhibiting SARS-CoV-2 RdRp as compared to Remdesivir.
Binding free energies for Remdesivir and EBDGp with SARS-CoV-2 RdRp
To compute binding free energies (ΔG Bind) of protein-ligand complexes, MM/GBSA calculations were performed, which gives the output in the context of VDW, hydrophobic, and solvation components. The phytocompounds and reference compound, Remdesivir within the binding cavity of SARS-CoV-2 RdRp was subjected to ensemble-averaged Prime MM/GBSA calculations. The resulting free binding energies of each complex by taking ensemble-average MM/GBSA are summarized in Table 4.
Based on the MM/GBSA values obtained, as reported in Table 4, theEBDGp is expected to have a strong binding affinity (-66.498 kcal/mol) as compared to other phytocompounds. On the other hand, Remdesivir shows binding free energy of -49.492 kcal/mol with SARS-CoV-2 RdRp. The binding free energy calculation signifies that EBDGp has the most favorable binding free energy (-66.498 kcal/mol) closely followed by Marmelide (-57.145 kcal/mol) as shown in Fig. 10. Binding free energy calculations of the compounds reveal that EBDGp forms a stronger and highly stable complex with the SARS-CoV-2 RdRp and all computed energies are found to be thermodynamically favorable as compared to other both phytocompounds and Remdesivir.
Table 4 The ensemble-average Prime binding free energies (kcal/mol) of docked complexes during the 100ns MD simulation.
Lead Compounds complexed with SARS-CoV-2 S RdRp
|
ΔG Binda
(kcal/mol)
|
ΔG
Bind
Coulombb
(kcal/mol)
|
ΔG
Bind
Lipoc
(kcal/mol)
|
ΔG
Bind
Solv GBd
(kcal/mol)
|
ΔG
Bind
vdWe
(kcal/mol)
|
Remdesivir
|
-49.492
|
-17.452
|
-9.452
|
30.854
|
-52.830
|
±4.700
|
±12.150
|
±1.502
|
±8.552
|
±5.803
|
EBDGp
|
-66.498
|
-42.547
|
-9.074
|
49.440
|
-48.418
|
±2.619
|
±14.407
|
±1.482
|
±7.150
|
±4.748
|
Marmelide
|
-57.145
|
-33.670
|
-7.569
|
32.040
|
-39.696
|
±4.506
|
±24.894
|
±1.280
|
±7.538
|
±4.260
|
Seselin
|
-32.399
|
-41.354
|
-10.269
|
30.923
|
-34.461
|
±2.859
|
±5.746
|
±1.024
|
±4.850
|
±2.266
|
Pedunculagin
|
-25.681
|
-44.190
|
-10.053
|
30.622
|
-35.925
|
±3.175
|
±4.459
|
±1.134
|
±5.082
|
±4.689
|
Chebulagic acid
|
-24.464
|
-45.851
|
-9.413
|
29.695
|
-35.777
|
±2.804
|
±4.141
|
±1.110
|
±3.717
|
±5.252
|
aMM/GBSA binding free energy
bCoulomb energy
cLipophilic energy
dGeneralized born solvation energy
eVan der Waal energy
Molecular field-based similarity analysis
To evaluate the importance of features involved in the strong binding affinity of EBDGp towards SARS-CoV-2 RdRp, we have performed molecular field-based similarity analysis using FieldTemplater software. It provides the necessary 3D-molecular field properties of the EBDGp in alignment with the Reference molecule, Remdesivir. FieldTemplater took two compounds EBDGp and Remdesivir, optimally aligned their conformer fields and yielded 89 templates ranked as per incorporated score (Structural similarity, Field similarity and Shape similarity). The top-ranked molecular field template is presented in Fig. 11. In this study, we have explored only common fields with the aligned templates of Remdesivir and EBDGp describing electrostatic (positive and negative), Hydrophobic, and van der Waals properties.
Large points indicating strong interactions as observed in field point patterns [87], abundant in positive and negative electrostatic fields were observed in Remdesivir and EBDGp. Positive electrostatic fields are seen along with the hydroxyl group of EBDGp and amino group of Remdesivir, but interestingly the positive electrostatic fields are seen in large points along with hydroxyl groups of EBDGp. Large points of negative electrostatic fields are observed along with the carbonyl group and hydrophobic field along the benzoyl aromatic ring of both Remdesivir and EBDGp. Moreover, Van der Waals fields are also abundant along both Remdesivir and EBDGp equally. These results of molecular field-based similarity analysis show that positive electrostatic fields are largely observed along with hydroxyl groups of the EBDGp which indicates the importance of hydroxyl groups in an efficient binding with SARS-CoV-2 RdRp. This study demonstrates that the presence of the hydroxyl group can be assessed further for lead optimization and design a more potent lead candidate.