Quantum biochemistry, molecular docking, and dynamics simulation revealed synthetic peptides induced conformational changes affecting the topology of the catalytic site of SARS-CoV-2 main protease.

The recent outbreak caused by SARS-CoV-2 continues to threat and take many lives all over the world. The lack of an efficient pharmacological treatments are serious problems to be faced by scientists and medical staffs worldwide. In this work, an in silico approach based on the combination of molecular docking, dynamics simulations, and quantum biochemistry revealed that the synthetic peptides RcAlb-PepI, PepGAT, and PepKAA, strongly interact with the main protease (Mpro) a pivotal protein for SARS-CoV-2 replication. Although not binding to the proteolytic site of SARS-CoV-2 Mpro, RcAlb-PepI, PepGAT, and PepKAA interact with other protein domain and allosterically altered the protease topology. Indeed, such peptide-SARS-CoV-2 Mpro complexes provoked dramatic alterations in the three-dimensional structure of Mpro leading to area and volume shrinkage of the proteolytic site, which could affect the protease activity and thus the virus replication. Based on these findings, it is suggested that RcAlb-PepI, PepGAT, and PepKAA could interfere with SARS-CoV-2 Mpro role in vivo. Also, unlike other antiviral drugs, these peptides have no toxicity to human cells. This pioneering in silico investigation opens up opportunity for further in vivo research on these peptides, towards discovering new drugs and entirely new perspectives to treat COVID-19.Communicated by Ramaswamy H. Sarma.

Introduction

The current coronavirus outbreak caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the third epidemic event related to Coronaviruses in the last twenty years. The other two outbreaks that were caused by severe acute respiratory syndrome coronavirus 1 (SARS-CoV-1) and Middle East respiratory syndrome coronavirus (MERS-CoV) absolutely were not as severe and aggressive as SARS-CoV-2 (Song et al., 2019; WHO, 2020). To date, SARS-CoV-2 has infected 20,972,577 people in 215 countries with 750,377deaths (CDC, n.d; Hui et al., 2020; Practice, 2020) and transmission still continues to accelerate in different geographic region of the world. Often, the main problem caused by SARS-CoV-2 in infected patients is the immune response triggered against the virus infection. This immune reaction is characterized by overproduction of the proinflammatory cytokines tumor necrosis factors (TNFs) and interleukins IL-6 and 1, known as the “cytokine storm,” which is a hyperinflammatory state that led to multiple-organ damage and dysfunction, specially of lungs, heart, liver, and kidneys (Li et al., 2020; Ragab et al., 2020; Tang et al., 2020).

First identified in Wuhan, China, genomic data analysis revealed that the human coronavirus SARS-CoV-2 has 70% genetic similarity with SARS-CoV-1 (Almazán et al., 2014; Wu et al., 2020). One of the common features amongst these coronaviruses is that they entry in human cells by physical interaction of the viral spike glycoprotein (S protein) with the human angiotensin-converting enzyme 2 (ACE2) receptor, which is located at the cell membrane. Subsequently, conformational alteration of the S protein permits both the viral envelope to combine with the outer membrane and the transport of the virus’ genetic material inside the human cell. However, amino acid sequence data analysis showed the SARS-CoV-2 S protein possesses mutations in the receptor-binding domain (RBD), in relation to that of SARS-CoV-1 S protein. Such genetic alteration leads to 20-fold increased affinity of SARS-CoV-2 S protein to ACE2, compared to SARS-CoV-1 S protein, and enhanced infectivity and velocity of SARS-CoV-2 spreading in humans (Souza et al., 2021; Hoffmann et al., 2020; Yuan et al., 2017). Once inside the target cells, SARS-CoV-2 uses its main protease (Mpro), a pivotal protein for SARS-CoV-2 replication process, across the mechanism of viral replication (Ziebuhr et al., 2000). Therefore, besides the viral spike glycoprotein (S protein), SARS-CoV-2 Mpro has taken as other potential pharmacological target for drug action against SARS-CoV-2 infection (Estrada, 2020; Hall & Ji, 2020; Ngo et al., 2020; Ortega et al., 2020).

Although SARS-CoV-2 belongs to the genus coronavirus, as SARS-CoV-1 and MERS-CoV, its infection leads to a completely novel and unknown disease, coronavirus disease 2019 (COVID-19), which is ending human life and taking over the economy globally (Saxena, 2020). Indubitable, there is an urgent need to discover and develop effective drugs to combat the virus and discontinue the disastrous consequences it has brought to mankind. To speed up novel drug discovery and development for COVID-19 treatment, the integrated use of bioinformatics approaches, such as molecular docking, dynamics simulations, and quantum analysis (Liu et al., 2020; Morais et al., 2020; Ngo et al., 2020) allows to select potential antiviral molecules to be subsequently tested in vitro and in vivo against SARS-CoV-2 and other viruses.

Recently, our research group, by using bioinformatics approaches, found out that the synthetic peptides Mo-CBP3-PepII and PepKAA targeted SARS-CoV-2 spike protein and induced conformational alterations that disrupt its correct binding to the ACE2 receptor (Souza et al., 2020). In this current study, which was conducted to further exploit the peptide-based therapeutic approach to treat COVID-19, the integrated use of these above mentioned bioinformatics methods was employed to assess whether any out of the following eight synthetic small peptides, Mo-CBP3-PepI, Mo-CBP3-PepII, Mo-CBP3-PepIII (Oliveira et al., 2019), RcAlb-PepI, RcAlb-PepII, RcAlb-PepIII (Dias et al., 2020), PepGAT, and PepKAA (Souza et al., 2020), which were designed based on plant proteins purified by our research group, could physically interact with SARS-CoV-2 Mpro and inhibit both its binding to ACE2 cell membrane receptor and its crucial activity. Briefly, amongst the tested peptides, RcAlb-PepI, PepGAT, and, particularly, PepKAA interact with SARS-CoV-2 Mpro at a region far away from the active site, but induce conformational alterations of the protease that promote shrinkage of its catalytic domain. These results suggest that RcAlb-PepI, PepGAT, and PepKAA could allosterically inhibit the SARS-CoV-2 Mpro activity and theoretically prevent viral replication and, consequently, spread of SARS-CoV-2 infection. Certainly, this hypothesis must be tested in vitro before being tested in vivo.

Methodology

Three-dimensional structures

The three-dimensional (3D) structures of Mo-CBP3-PepI, Mo-CBP3-PepII, and Mo-CBP3-PepIII were identical to those used by Oliveira et al. (2019). The 3D structures of RcAlb-PepI, RcAlb-PepII, and RcAlb-PepIII were equal to those employed by Dias et al. (2020). The PepGAT and PepKAA 3D structures were identical to those used by Souza et al. (2020). The 3D structure files of SARS-CoV-2 main protease (Mpro) (PDB: 6M03) were downloaded from Protein Data Bank (PDB, https://www.rcsb.org/).

Molecular docking assays

Blind molecular docking analyses were carried out in FRODOCK 3.12 (Ramírez-Aportela et al., 2016) and ClusPro 2.0 (https://cluspro.org), using the synthetic peptides as ligands against SARS-CoV-2 Mpro. Docking scores were used to select the peptides with the highest potential for interaction with SARS-CoV-2 Mpro. A score of 1 was assigned to the peptide with the highest docking value (referred to as the top scored pose). The scores of the subsequent peptides were decreased successively by a factor of 0.125 as the docking values decreased in relation to that of the peptide that received score 1. The peptide score obtained in the FRODOCK was added to that generated by ClusPro. Based on the sum of scores, the three peptides with the highest potential were used to conduct the subsequent studies.

Molecular dynamics simulation

To stabilize the complexes formed between SARS-CoV-2 Mpro and the studied peptides before molecular dynamics simulation using Gromacs version 2018.4 (Van Der Spoel et al., 2005), they were minimized and equilibrated. OPLS-AA/L all-atom force field (Moal & Bates, 2010; Robertson et al., 2015) was used to build molecular topology. Then, a cubic water box of edge length 2 nm was created. Box solvation was done with the SPC/E water model, the systems were neutralized, and Na+ and Cl− added at 0.15 M concentration. Energy minimization of the complex structure was carried out until a negative potential energy and a maximum force below 103 kJ mol−1 nm−1 were attained. Next, temperature and pressure equilibration were performed during 100 ps simulation. Molecular dynamics simulations, in duplicate, were performed during 100 ns and the final structures generated were used for further analysis.

Interface analysis of the complexes formed between SARS-CoV-2 Mpro and the studied peptides

This analysis was carried out using the Protein Interactions Calculator (PIC) webserver (http://pic.mbu.iisc.ernet.in/). The 2D figures showing hydrogen bonds and hydrophobic interactions were built using the Ligplot software (Laskowski & Swindells, 2011). The Pymol software was used to generate 3D structures and to calculate RMSD (root mean square deviation) (Laskowski et al., 2018). CASTp 3.0 (Tian et al., 2018) was used to assess changes in the area and volume of the proteolytic site of the SARS-CoV-2 main protease (SARS-CoV-2 M’pro) when alone and complexed with the studied peptides.

Quantum biochemistry calculation

This was performed according to a protocol established previously by Zhang and Zhang (2003). Molecular fractionation with conjugate caps (MFCC) was carry out to calculate in silico the full quantum mechanical interaction energies between two pairs of specific amino acid residues (Ri and Rj) involving the studied peptides and SARS-CoV-2 Mpro, as follows, based on the work of Amaral et al. (2020):

Where E(Ci-1 Ri Ci+1 + Cj-1 Rj Cj+1), the first term of the equation, is the total energy of the system formed by the residues Ri and Rj correctly capped; E(Ci-1 Ri Ci+1 + Cj-1 Cj+1), the second term, is the total energy of the system formed by the capped residue Ri and the caps of the residue Rj; the third term, E(Ci-1 C +1 + Cj-1 Rj Cj+1), represents the total energy of the system formed by the capped residue Rj and the caps of the residue Ri; and the last term, E(Ci-1 Ci+1 + Cj-1 Cj+1), accounts for the total energy of the system formed by the caps of both residues Ri and Rj. The caps Ci-1(Ci+1) and Cj-1(Cj+1) are made from the residues covalently bound to the amine (carboxyl) groups of Ri and Rj. In the MFCC method used, all interaction between amino acid residues of the studied peptides and SARS-CoV-2 Mpro separated from each other within a 8 Å range were calculated, considering a dielectric function approaches of 40 (ɛ = 40) for all interactions. The structural files (PDB format) obtained after molecular dynamics simulation and MFCC were used as inputs for density functional theory calculations with DMOL3 (Delley, 2000).

Results

Molecular docking parameters of the interaction between the synthetic peptides and SARS-CoV-2 Mpro

Before molecular dynamics and quantum biochemistry analyses, molecular docking simulations were run on the FRODOCK 3.12 and ClusPro 2.0 webservers to find out which peptides best interact with SARS-CoV-2 Mpro. Both servers calculated that the eight studied peptides bind strongly to SARS-CoV-2 Mpro with score values that varied from 1700.10 to 2182.11 (Table 1). ClusPro 2.0 was used to calculate the free energies (ΔG) of interactions, which were low and varied from −579.1 to −736.5 kcal.mol−1 (Table 1). The sum of the punctuations provided by FRODOCK and ClusPro 2.0 was used to select RcAlb-pepII, PepGAT, and PepKAA as the best peptides, in relation to their capacity of binding to SARS-CoV-2 Mpro, with punctuation values of 1.500, 1.375, and 1.375, respectively (Table 1). Molecular docking analysis also revealed that these peptides do not bind to the proteolytic site of SARS-CoV-2 Mpro (His41 and Cys145) (Figure 1). Nevertheless, RcAlb-pepII, PepGAT, and PepKAA were further analyzed using molecular dynamics and quantum biochemistry.

Table 1.

Molecular docking parameters of the interactions between the studied synthetic peptides and SARS-CoV-2 Mpro.

Peptide	FRODOCKa	Punctuationb	ClusProc	Punctuationd	Sume
Mo-CBP3-PepI	1816.15	0.375	−603.5	0.250	0.625
Mo-CBP3-PepII	1809.39	0.250	−736.5	1.000	1.250
Mo-CBP3-PepIII	1700.10	0.125	−654.0	0.625	0.750
RcAlb-PepI	2021.32	0.750	−579.1	0.125	0.875
RcAlb-PepII	2002.75	0.625	−694.5	0.875	1.500
RcAlb-PepIII	1871.59	0.500	−654.7	0.750	1.250
PepGAT	2061.69	0.875	−642.2	0.500	1.375
PepKAA	2182.11	1.000	−632.7	0.375	1.375

Calculated using the FRODOCK v.3.12 server.

b,dPunctuation is associated with both the FRODOCK score and the ClusPro lowest binding free energy (kcal mol−1), respectively. A score of 1 was assigned to the peptide with the highest docking value (referred to as the top scored pose). The scores of the subsequent peptides were decreased successively by a factor of 0.125 as the docking values decreased in relation to that of the peptide that received score 1.

The lowest binding free energy (ΔG) calculated using ClusPro 2.0.

Represents the sum of the scores (punctuation) in b and d.

Figure 1.

Complexes formed by docking of the peptides RcAlb-PepII (A), PepGAT (B), and PepKAA (C) on SARS-CoV-2 Mpro.

Molecular dynamics simulations were done with SARS-CoV-2 Mpro alone and complexed with the peptides RcAlb-pepII, PepGAT, and PepKAA. RMSD of SARS-CoV-2 Mpro alone and complexed with the peptides increased gradually up to 1.5 Å within 50 ns and then remained stable until 100 ns, with RMSD fluctuation below 1 Å (Figure 2). The stable conformation obtained from each molecular dynamics simulation was used to perform further analyzes.

Figure 2.

RMSD plot of the conformation stability of SARS-CoV-2 Mpro alone and of the complexes formed by docking of RcAlb-PepII, PepGAT, and PepKAA on SARS-CoV-2 Mpro as a function of time (ns).

Interaction between SARS-CoV-2 Mpro with RcAlb-PepII

The key amino acid residues responsible for the multi-point binding interactions between SARS-CoV-2 Mpro and RcAlb-PepII were Lys137, Asn274, Met276, Tyr237, and Asn277 with Leu9, Lys2, Lys2, Pro5, and Lys2 (Table 2; Figure 3A, B and D) with interaction energy of −12.13, −8.30, −4.82, −4.60, and −3.75 kcal.mol−1, respectively (Table 2). Repulsive interactions occur between the amino acid residues Asp197, Lys137, and Glu290 of SARS-CoV-2 Mpro with Leu9, Ala8, and Leu9 of RcAlb-PepII, with interaction energy of +1.64, +1.66, and +1.92 kcal mol−1 and distance of 1.64, 1.66, and 1.92 Å, respectively (Table 2). shows the quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of RcAlb-PepII at distances up to 8 Å. The complex formed between SARS-CoV-2 Mpro and RcAlb-PepII is molded through hydrophobic interactions established by the amino acid residues Tyr237, Tyr239, Tyr239, Leu271, Leu272, Leu272, Met276, Leu286, Leu286, Leu286, Leu287, and Leu287 of SARS-CoV-2 Mpro with Pro5, Ile4, Leu9, Ile4, Ile4, Pro5, Ile4, Ile4, Ile7, Leu9, Ile4, and Leu9 of RcAlb-PepII (Figure 3A, B and D). Hydrogen bonds are established between the amino acid residues Met276 and Asn274 of SARS-CoV-2 Mpro with Lys2 of RcAlb-PepII (Figure 3A, B and D). The individual interaction energy of the residues Ala1, Lys2, Leu3, Ile4, Pro5, Thr6, Ile7, Ala8, Leu9 was −5.28, −20.44, −3.55, −10.60, −5.86, −0.74, −11.58, −5.98, and −10.83 kcal.mol−1, respectively (Figure 3C).

Table 2.

Quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of RcAlb-PepII.

Amino acid residues		Distance (Å)	Free energy of interaction (kcal mol−1)	Residue charge		Residue atom
SARS-CoV-2 Mpro	RcAlb-PepII			SARS-CoV-2 Mpro	RcAlb-PepII	SARS-CoV-2 Mpro	RcAlb-PepII
LYS137	LEU9	1.80	−12.13	1	−1	HZ3	O2
ASN274	LYS2	2.79	−8.30	0	1	O	HZ1
MET276	LYS2	2.17	−4.82	0	1	H	O
TYR237	PRO5	2.57	−4.60	0	0	CE1	HG1
ASN277	LYS2	2.39	−3.75	0	1	OD1	HD2
LEU287	ILE4	2.33	−2.99	0	0	HD13	HD1
LEU287	LEU9	2.33	−2.86	0	−1	HB2	HD11
LEU286	LEU9	2.42	−2.82	0	−1	HD12	HB2
LEU272	ILE4	2.44	−2.77	0	0	HA	HB
GLY275	LYS2	2.44	−2.75	0	1	HA1	O
LEU286	ILE7	2.82	−2.75	0	0	HD21	HG22
LEU272	PRO5	2.50	−2.63	0	0	HA	HD2
ARG131	LEU9	5.02	−2.08	1	−1	HH12	O1
THR199	LEU9	2.63	−1.94	0	−1	HG23	HD12
LEU271	ILE4	2.68	−1.88	0	0	HB2	HG12
ASP197	LEU9	6.44	1.64	−1	−1	OD1	O2
LYS137	ALA8	5.44	1.66	1	0	HZ3	C
GLU290	LEU9	5.70	1.92	−1	−1	OE2	O1

Figure 3.

The 3D and 2D structure representations of the complex formed by docking of RcAlb-PepII (Ala1-Lys-Leu-Ile-Pro-Thr-Ile-Ala-Leu9) on SARS-CoV-2 Mpro are depicted in A and B, respectively. The interaction energies of the amino acid residues of RcAlb-PepII with those of SARS-CoV-2 Mpro, and the amino acid residues that participate in the hydrophobic interactions and hydrogen bonds between RcAlb-PepII and SARS-CoV-2 Mpro are shown in C and D, respectively.

Interaction between SARS-CoV-2 Mpro with PepGAT

Distances (Å), total free energy of interaction (kcal mol−1), amino acid residue charge and amino acid residue atom closest (up to 8 Å) to the interactions between SARS-CoV-2 Mpro and PepGAT are shown in the . Interaction between SARS-CoV-2 Mpro and PepGAT was establish predominantly by Gln306, Gln306, Asp153, Asp248, Phe294, Gln110, Ile249, Phe294, and Asp245 with Gly1, Arg5, Gly1, Arg10, Arg5, Thr3, Arg10, Ala2, and Arg10, with interaction energy of −19.62, −12.35, −11.12, −10.39, −6.28, −4.73, −4.51, −3.96, and −3.89 kcal.mol−1, and distances of 1.57, 2.32, 2.50, 2.80, 2.31, 2.23, 2.10, 4.12, and 2.54 Å, respectively, to form stable complex (Table 3; Figure 4A and B). Repulsive interactions occur mainly between the amino acid residues Arg298, Lys102, and Phe305 of the protease with Gly1 of the peptide, with interaction energy of +1.47, +2.26, and +2.79 kcal.mol−1 and distances of 6.14, 5.39, and 3.49 Å, respectively (Table 3).

Table 3.

Quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of PepGAT.

Amino acid residues		Distance (Å)	Free energy of interaction (kcal mol−1)	Residue charge		Residue atom
SARS-CoV-2 Mpro	PepGAT			SARS-CoV-2 Mpro	PepGAT	SARS-CoV-2 Mpro	PepGAT
GLN306	GLY1	1.57	−19.62	−1	1	O1	H3
GLN306	ARG5	2.32	−12.35	−1	1	O1	HH21
ASP153	GLY1	2.50	−11.12	−1	1	OD2	H2
ASP248	ARG10	2.80	−10.39	−1	0	OD1	HH12
PHE294	ARG5	2.31	−6.28	0	1	HD1	HD1
GLN110	THR3	2.23	−4.73	0	0	HE22	O
ILE249	ARG10	2.10	−4.51	0	0	HG21	HH22
PHE294	ALA2	4.12	−3.96	0	0	HB1	O
ASP245	ARG10	2.54	−3.89	−1	0	OD1	HH22
PHE294	ARG10	3.02	−2.94	0	0	HE1	HG2
GLN107	VAL7	2.76	−2.32	0	0	HG2	HG11
ILE106	THR3	2.86	−2.15	0	0	HG21	HB
ASP295	ALA2	2.41	−2.05	−1	0	HB1	HB2
ASP295	GLY1	1.92	−1.87	−1	1	HB1	C
PHE294	ALA6	2.92	−1.87	0	0	HE2	HA
ILE249	ALA6	3.83	−1.82	0	0	HG21	HA
GLN107	ALA6	6.92	−1.79	0	0	OE1	HB1
ARG298	GLY1	6.14	1.47	1	1	HD2	HA1
LYS102	GLY1	5.39	2.26	1	1	HZ3	H2
PHE305	GLY1	3.49	2.79	0	1	HD2	HA1

Figure 4.

The 3D and 2D structure representations of the complex formed by docking of PepGAT (Gly1-Ala-Thr-Ileu-Arg-Ala-Val-Asn-Ser-Arg10) on SARS-CoV-2 Mpro are depicted in A and B, respectively. The interaction energies of the amino acid residues of PepGAT with those of SARS-CoV-2 Mpro, and the amino acid residues that participate in the formation of hydrogen bond, hydrophobic, cation-pi, and ionic interactions between PepGAT and SARS-CoV-2 Mpro are shown in D.

shows the quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of PepGAT at distances up to 8 Å. SARS-CoV-2 Mpro establishes hydrophobic, ionic, cation-pi and hydrogen bond interactions with PepGAT. Hydrogen bonds are established between the amino acid residues Gln110, Asp153, and Asp248 of SARS-CoV-2 Mpro with Thr3, Gly1, and Arg10 of PepGAT (Figure 4A, B, and D). The ionic interactions occur between the amino acid residues Asp245 and Asp248 of Mpro with Arg10 of PepGAT (Figure 4A, B and D). A cation-pi interaction is formed by Phe294 of SARS-CoV-2 Mpro with Arg10 of PepGAT. Hydrophobic interactions are established by the amino acid residues Val104, Ile249, Phe294, and Phe294 of SARS-CoV-2 Mpro with Ile4, Ala6, Ala2, and Ala6 of PepGAT (Figure 4B and D). The individual interaction energy of the residues Gly1, Ala2, Thr3, Ileu4, Arg5, Ala6, Val7, Asn8, Ser9, and Arg10 of PepGAT was −19.73, −6.59, −12.20, −12.10, −16.45, −3.56, −6.26, −0.80, −0.74, and −22.80 kcal.mol−1, respectively (Figure 4C).

Interaction between SARS-CoV-2 Mpro with PepKAA

The key amino acid residues responsible for the multi-point binding interactions between SARS-CoV-2 Mpro and PepKAA are Asp245, Asp248, Asp248, Gln256, Phe294, Asp153, Ile249, Pro252, and Val297 with Arg5, Arg5, Lys1, Phe9, Tyr8, Lys7, Arg5, Tyr8, and Tyr8 with interaction energy of −11.93, −11.32, −10.36, −6.55, −6.08, −4.93, −4.03, −3.60, and −3.29 kcal.mol−1 and distances of 1.68, 1.76, 3.71, 2.70, 2.31, 3.79, 2.60, 2.14, and 2.14 Å, respectively (Table 4). The repulsive interactions occur between Val247 and Lys102 of SARS-CoV-2 Mpro with Arg5 and Lys7 of PepKAA, with interaction energy of +0.58 and +1.34 kcal.mol−1 and distances of 6.08 and 7.05 Å, respectively (Table 4). shows the quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of PepKAA at distances up to 8 Å. SARS-CoV-2 Mpro binds to PepKAA through hydrophobic, ionic, aromatic, and hydrogen bond interactions. Hydrogen bonds are formed by Asp245 and Asp248 of SARS-CoV-2 Mpro with Arg5 of PepKAA (Figure 5A, B and D). Ionic interactions occur between Asp153, Asp245, Asp248, and Asp248 of SARS-CoV-2 Mpro with Lys7, Arg5, Lys1, and Arg5 of PepKAA, respectively (Figure 5B and D). Hydrophobic interactions are established by Ile213, Pro252, Pro252, Pro252, Leu253, Leu253, Pro293, Phe294, Val296, and Val297 of SARS-CoV-2 Mpro with Phe9, Ile6, Tyr8, Phe9, Tyr8, Phe9, Tyr8, Tyr8, Phe9, and Tyr8 of PepKAA. An aromatic interaction is formed by Phe294 of SARS-CoV-2 Mpro with Tyr8 of PepKAA (Figure 5A and D). The individual interaction energy of the residues Lys1, Ala2, Ala3, Asn4, Arg5, Ile6, Lys7, Tyr8, Phe9, Gln10 of PepKAA was −6.27, −1.06, +0.05, −0.80, −25.54, −2.34, −11.91, −22.90, −13.16, and −8.42 kcal.mol−1, respectively (Figure 5C).

Table 4.

Quantum parameters of the multi-point binding interactions between the amino acid residues of SARS-CoV-2 Mpro and those of PepKAA.

Amino acid residues		Distance (Å)	Free energy of interaction (kcal mol−1)	Residue charge		Residue atom
SARS-CoV-2 Mpro	PepKAA			SARS-CoV-2 Mpro	PepKAA	SARS-CoV-2 Mpro	PepKAA
ASP245	ARG5	1.68	−11.93	−1	1	OD2	HH22
ASP248	ARG5	1.76	−11.32	−1	1	OD2	HH11
ASP248	LYS1	3.71	−10.36	−1	2	OD1	HB1
GLN256	PHE9	2.70	−6.55	0	0	HB1	HB1
PHE294	TYR8	2.31	−6.08	0	0	HD1	HE1
ASP153	LYS7	3.79	−4.93	−1	1	OD1	HZ3
ILE249	ARG5	2.60	−4.03	0	1	HD2	HD2
PRO252	TYR8	2.14	−3.60	0	0	HG1	HB1
VAL297	TYR8	2.14	−3.29	0	0	HG23	HA
ASP245	LYS1	7.09	−2.66	−1	2	OD1	HA
PRO252	ILE6	2.26	−2.60	0	0	HB2	HA
LEU253	PHE9	2.32	−2.50	0	0	HD21	HD2
TYR154	LYS7	3.17	−2.41	0	1	HE2	HE2
LEU253	TYR8	2.41	−2.40	0	0	HD22	HD2
PRO252	PHE9	1.90	−2.31	0	0	HB1	H
CYS300	PHE9	2.70	−1.77	0	0	O	HZ
ASP248	ALA2	4.12	−1.74	−1	0	OD2	H
PRO293	TYR8	2.06	−1.73	0	0	HB1	HE2
VAL247	ARG5	6.08	0.58	0	1	C	HH11
LYS102	LYS7	7.05	1.34	1	1	HZ1	HZ3

Figure 5.

The 3D and 2D structure representations of the complex formed by docking of PepKAA (Lys1-Ala-Ala-Asn-Arg-Ile-Lys-Tyr-Phe-Gln10) on SARS-CoV-2 Mpro are depicted in A and B, respectively. The interaction energies of the amino acid residues of PepKAA with those of SARS-CoV-2 Mpro, and the amino acid residues that participate in the formation of hydrogen bond, hydrophobic, ionic, and aromatic interactions between PepKAA and SARS-CoV-2 Mpro are shown in D.

Quantum biochemistry description

The most important amino acid residues of SARS-CoV-2 Mpro that interact with RcAlb-PepII are Leu287, Leu286, Asn274, Gly275, Met276, Lys137, Tyr237, Leu272, and Asn277 with interaction energy of −9.36, −9.03, −8.78, −6.47, −6.46, −6.41, −5.97, −5.80, and −5.60 kcal.mol−1, respectively. The Asp197, Asp289, and Glu290 residues showed repulsive energies (Figure 6A). The sum of all free energies of interaction between SARS-CoV-2 Mpro and RcAlb-PepII (Et) was ‒74.85 kcal.mol−1 (Figure 7).

Figure 6.

Binding site, interaction energy, and residue domain (BIRD) panel showing the MFCC interaction energies between the main amino acid residues of SARS-CoV-2 Mpro with those of RcAlb-PepII (A), PepGAT (B), and PepKAA (C). The minimal distance (Å) between each residue that participates in the interaction is indicates at the right side of the panel. The amino acid residues at the left side of the panel are from SARS-CoV-2 Mpro.

Figure 7.

Total interaction energy between SARS-CoV-2 Mpro and the Peptides as a function of the interaction distance. Orange, Magenta, and Black squares represent RcAlb-PepII, PepGAT, and PepKAA, respectively. Et accounts for the sum of the interaction energies up to 8 Å.

The key amino acid residues involved in the interaction of SARS-CoV-2 Mpro with the PepGAT are Asp153, Gln306, Phe294, Asp248, and Asp245 with interaction energy of −12.47, −11.27, −8.36, −6.67, and −4.72 kcal.mol−1, respectively. Residues of Arg298, Phe305, Lys102 showed small repulsive energies (Figure 6B). The sum of all free energies of interaction between SARS-CoV-2 Mpro with PepGAT was ‒101.2 kcal.mol−1 (Figure 7).

The major amino acid residues of the interaction between SARS-CoV-2 Mpro and PepKAA are Asp248, Pro252, Phe294, Asp153, and Asp245, with interaction energy of −20.14, −11.27, −10.15, −9.57, and −8.17 kcal. mol−1, respectively. Lys102 showed a small repulsive energy (Figure 6C). The sum of all free energies of interaction between SARS-CoV-2 Mpro and PepKAA was ‒92.35 kcal.mol−1 (Figure 7).

Assessment of the conformational changes induced by RcAlb-PepII, PepGAT, and PepKAA in the area and volume of the proteolytic site of SARS-CoV-2 Mpro

RMSD analyses revealed that RcAlb-PepII, PepGAT, and PepKAA induced alterations in the 3D structure of SARS-CoV-2 Mpro (Figure 8). The control structure presented RMSD value of 0, which indicates a typical and functional 3D structure (Figure 8A). However, the structural alignment of SARS-CoV-2 Mpro complexed with RcAlb-PepI, PepGAT, or PepKAA revealed several alterations in the SARS-CoV-2 Mpro structure, which was confirmed by the RMSD values of 3.118, 3.164, 3.054 Å, respectively (Figure 8B–D).

Figure 8.

3D visualization of the molecular complexes formed between SARS-CoV-2 Mpro and the peptides with the RMSD values of SARS-CoV-2 Mpro alone (in cyan, A) and when conjugated with RcAlb-PepII (B), PepGAT (C), and PepKAA (D). The merged structures suggest conformational alteration induced by the peptides.

These structural changes in SARS-CoV-2 Mpro induced by interactions with the studied peptides lead to alterations in the area and volume of the SARS-CoV-2 Mpro proteolytic site. SARS-CoV-2 Mpro itself, not complexed with the studied peptides, presented an area and volume of 279.7 Å2 and 298.1 Å3, respectively (Figure 9A).

Figure 9.

Conformational structure of the proteolytic site of SARS-CoV-2 Mpro alone and complexed with the peptides. The proteolytic site of SARS-CoV-2 Mpro alone (A) presents a volume and area higher than when SARS-CoV-2 Mpro is complexed with RcAlb-PepII (B), PepGAT (C), and PepKAA (D).

When complexed with RcAlb-PepII, the area and volume of SARS-CoV-2 Mpro proteolytic site reduced to 120.6 Å2 and 93.3 Å3, which represent 68.7% and 56.9% decrease, respectively (Figure 9B). The structural changes caused by PepGAT also reflected in the area and volume of the SARS-CoV-2 Mpro proteolytic site, which decreased from 298.1 Å2 and 279.7 Å3 to 183.7 Å2 and 179.0 Å3, resulting in a 38.4% and 36.0% decrease, respectively (Figure 9C). Regarding to PepKAA, its interaction with the protease promoted a 13.0% (279.7 Å2 to 243.3 Å2) and 23.1% (298.1 Å3 to 229.2 Å3) decrease, respectively, in the area and volume of the SARS-CoV-2 Mpro proteolytic site (Figure 9D).

Discussion

Currently, there are two main ways to find out and develop medicines against SARS-CoV-2. The first one is through drug repositioning in which the antiviral activity of several pharmacological classes of medicines already approved for use in humans is investigated as a possibility for generating new treatments against COVID-19 (Diamond & Pierson, 2020; Serafin et al., 2020; Tay et al., 2020; Zhang et al., 2020). To date, such attempts have been tried unsuccessfully. The second approach is still a promise, but it is based on systematic investigations conducted to identify novel small molecules designed to target key SARS-CoV-2 factors such as the Spike glycoprotein and RNA polymerase (Souza et al., 2020). Such an approach involves the in silico molecular docking and molecular dynamics simulation, which are structure-based method employed to predict the binding affinity of a ligand molecule to a target (Dai et al., 2020; Hall & Ji, 2020; Ngo et al., 2020; Ortega et al., 2020; Zhang et al., 2020).

So, molecular docking and molecular dynamics simulation constitute useful tools to facilitate the drug development process as they allow to predict in silico the binding affinity of a designed ligand molecule to a target as SARS-CoV-2 Mpro (Zhang et al., 2020). Mpro has three-domain (domains I to III) cysteine protease essential for coronavirus replication as it plays a crucial role in the processing of the viral polyproteins into mature proteins (Ziebuhr et al., 2000). SARS-CoV-2 Mpro and SARS-CoV-1 Mpro primary structures differ by only 12 amino acid residues, which means they have 96% similarity (Estrada, 2020). However, a considerable difference between the catalytic site of SARS-CoV-1 and that of SARS-CoV-2 is observed. In SARS-CoV-1 Mpro, the active site cavity is a well-defined pocket with the area and volume of 256.8 Å2 and 191.24 Å3, respectively, compared to 352.1 Å2 and 323.73 Å3 of that of SARS-CoV-2 Mpro (Ortega et al., 2020). These differences increase by 15% the proteolytic effectiveness of SARS-CoV-2 Mpro compared to that of SARS-CoV-1 Mpro (Estrada, 2020; Ortega et al., 2020; Zhang et al., 2020). Such higher catalytic efficiency of SARS-CoV-2 Mpro could enhance the virus replication process and formation of new virus particles, which could lead to higher infectivity. Therefore, due to its crucial involvement in maturation of most of the nonstructural proteins that are translated from SARS-CoV-2 RNA, SARS-CoV-2 Mpro is an attractive target to design antiviral drugs (Sharma et al., 2020).

Many studies have tried to either analyze new drugs or already known drugs that could interact with the catalytic site of SARS-CoV-2 Mpro and inhibit its activity (Hall & Ji, 2020; Ngo et al., 2020; Ortega et al., 2020; Sharma et al., 2020; Zhang et al., 2020). Moreover, drugs that have been employed to treat HIV by inhibiting the viral protease were also tested (Hall & Ji, 2020). In this current in silico study we used molecular docking and dynamics simulations to predict and show that the synthetic peptides RcAlb-PepII, PepGAT, and PepKAA, out of eight peptides tested, interact most efficiently with SARS-CoV-2 Mpro (Table 1, Figures 1–7). These peptides interact with SARS-CoV-2 Mpro at a region far away from the protease catalytic site, which suggests that the peptides may modulate the SARS-CoV-2 Mpro activity allosterically (Figures 8 and 9). Actually, this is a pioneer study in which quantum biochemistry is employed to analyze the interaction of peptides against SARS-CoV-2 Mpro (Tables 2–4). Quantum biochemistry calculations (Morais et al., 2020; Sousa et al., 2016) allowed to deduce the individual energies of interactions of each amino acid residue of the studied peptides and those of SARS-CoV-2 Mpro. Therefore, it was possible to predict the hydrogen bonds, ionic, aromatic, and hydrophobic interactions that are important to establish attractive or repulsive interactions between the peptides and SARS-CoV-2 Mpro (Figures 3–6). Figure 7 shows that, by using quantum biochemistry calculations, the total interaction energy between RcAlb-PepII, PepGAT, and PepKAA with SARS-CoV-2 Mpro was −74.85, −101.2, and −92.35 kcal.mol−1, respectively. As this is the first work that analyzes, through quantum biochemistry, interactions of peptides with SARS-CoV-2 Mpro, the results presented herein are compared with those reported by Campos et al. (2020), in which the interaction of two peptides against the Zika virus protease was analyzed. The authors demonstrated, by quantum biochemistry, that the interaction energies of the peptides cn-716 and acyl-KR-aldehyde with the protease NS2B–NS3 were −63.35 kcal.mol−1 and −71.4 kcal.mol−1, respectively. For instance, RcAlb-PepII, PepGAT, and PepKAA interacts with SARS-CoV-2 Mpro even more strongly than cn-716 and acyl-KR-aldehyde to the protease NS2B–NS3. Nevertheless, interaction of other non-peptide-like antiviral drugs have been tested with the same purpose of inhibiting the proteolytic activity of SARS-CoV-2 Mpro. For example, Ortega et al. (2020) tested by molecular docking the interaction of the clinically proven anti-HIV drugs Saquinavir, Lopinavir, and Tripranavir, which act as inhibitors of the HIV protease, with the proteolytic site of SARS-CoV-2 Mpro. Additionally, Hall and Ji (2020) reported that Remdesivir, a nucleotide analogue initially developed to treat hepatitis C and later Ebola and Marburg virus, also binds to the proteolytic site of SARS-CoV-2 Mpro. These two later studies are referred as drug reposition or drug repurposing, which is an approach to speed up the drug discovery process through identification of a novel clinical use for an existing drug approved for a different indication (Serafin et al., 2020). Moreover, Ngo et al. (2020) docked some natural compounds and found that cannabisin A and isoacteoside interacted with the proteolytic site of SARS-CoV-2 Mpro. Currently, the most reliable inhibitor of SARS-CoV-2 Mpro, as seen by molecular docking and proven in vitro, is 13b, an ɑ-ketoamide that is a protease inhibitor of coronavirus that strongly binds to the proteolytic site of SARS-CoV-2 Mpro (Zhang et al., 2020). RcAlb-PepII, PepGAT, and PepKAA interact with SARS-CoV-2 Mpro at a region far away from the protease active site (Figures 1, 8, and 9), contrary to these above-mentioned antiviral drugs tested. However, Bzówka et al. (2020) showed that the proteolytic site of SARS-CoV-2 Mpro has high flexibility and plasticity, which means it is highly susceptible to genetic mutational change. Such possibility constitutes a serious problem for designing drugs targeting the proteolytic site because a simple genetic mutation could prevent the action of these antiviral drugs. Actually, SARS-CoV-2 is an RNA virus and has high mutational rates (Bzówka et al., 2020) and, for example, the antiviral 13 b, which strongly binds to the proteolytic site of SARS-CoV-2 Mpro (Zhang et al., 2020), could quickly become ineffective against coronavirus. Prediction that RcAlb-PepII, PepGAT, and PepKAA interact with SARS-CoV-2 Mpro at a region far away from the protease active site (Figures 1, 8 and 9) is an advantage because besides their interactions being unaffected by mutation of the active site of SARS-CoV-2 Mpro, they induce conformational alterations in the 3D structure of this protease (Figure 8), as shown by the RMSD values of SARS-CoV-2 Mpro complexed with the peptides, compared to SARS-CoV-2 Mpro alone, that could disrupt the proteolytic activity of the viral enzyme, leading to a severe reduction in SARS-CoV-2 replication. Indeed, interaction of RcAlb-PepII, PepGAT, and PepKAA reduce the volume and area of SARS-CoV-2 Mpro proteolytic site, respectively, in 56.9% and 68.7% (Figure 9B), 38.4% and 36.0% (Figure 9C), and 23.1 and 13.0% (Figure 9D). In a recent report by our research group, it was shown that PepKAA also induced conformational changes in the SARS-CoV-2 Spike glycoprotein, disrupting its interaction with the ACE2 receptor located at the human cell membrane (Souza et al., 2020).

As PepKAA shows to interact with two different targets (SARS-CoV-2 Spike glycoprotein (Souza et al., 2020) and SARS-CoV-2 Mpro of SARS-CoV-2, this particular synthetic peptide is a potential small molecule candidate to be further tested in vitro against SARS-CoV-2, either alone or combined with other peptides like RcAlb-PepII, PepGAT. Of utmost importance is that RcAlb-PepII, PepGAT, and PepKAA were not toxic to mammalian cells like rabbit erythrocytes and human ABO type red blood cells (Dias et al., 2020; Oliveira et al., 2019; Souza et al., 2020), neither to fibroblast (lines L929 and MRC-5) and keratinocytes (Data not shown, manuscript in preparation).

Conclusion

Quantum biochemistry and molecular dynamics simulations allow to predicted that the peptides RcAlb-PepII, PepGAT, and PepKAA interact physically with SARS-CoV-2 Mpro and alter its 3D structure and provoke shrinkage of the active site of the protease. These findings suggest that these peptides are likely antiviral small molecules that could potentially inhibit SARS-CoV-2 replication in vivo, a hypothesis that requires further investigation in the near future to be proven true. Importantly, besides to be apparently harmless to mammalian cells, RcAlb-PepII, PepGAT, and PepKAA do not interact with the active site of SARS-CoV-2 Mpro, unlike most antiviral drugs, and, thus, mutation of this protease domain will not affect the peptide effectiveness. In conclusion, this pioneering in silico investigation opens up opportunity for further in vivo investigations on these peptides, towards discovering new drugs and entirely new perspectives to treat COVID-19. For instance, peptide-based therapeutics have various advantages in relation to traditional small-molecule drugs, since peptides have higher specificity to selected targets, low toxicity because the possibility for accumulation in the body is improbable, and their synthesis is not a complex, costly, and time-consuming technique (VanPatten et al., 2020).

Click here for additional data file.