In silico study identifies peptide inhibitors that negate the effect of non-synonymous mutations in major drug targets of SARS-CoV-2 variants

Abstract Since its advent in December 2019, SARS-CoV-2 has diverged into multiple variants with differing levels of virulence owing to the accumulation of mutations in its genome. The structural changes induced by non-synonymous mutations in major drug targets of the virus are known to alter the binding of potential antagonistic inhibitors. Here, we analyzed the effects of non-synonymous mutations in major targets of SARS-CoV-2 in response to potential peptide inhibitors. We screened 12 peptides reported to have anti-viral properties against RBD and 5 peptides against Mpro of SARS-CoV-2 variants using molecular docking and simulation approaches. The mutational landscape of RBD among SARS-CoV-2 variants had 21 non-synonymous mutations across 18 distinct sites. Among these, 14 mutations were present in the RBM region directly interacting with the hACE2 receptor. However, Only 3 non-synonymous mutations were observed in Mpro. We found that LCB1 - a de novo-synthesized peptide has the highest binding affinity to RBD despite non-synonymous mutations in variants and engages key residues of RBD-hACE2 interaction such as K417, E484, N487, and N501. Similarly, an antimicrobial peptide; 2JOS, was identified against Mpro with high binding affinity as it interacts with key residues in dimerization sites such as E166 and F140 crucial for viral replication. MD simulations affirm the stability of RBD-LCB1 and Mpro-2JOS complexes with an average RMSD of 1.902 and 2.476 respectively. We ascertain that LCB1 and 2JOS peptides are promising inhibitors to combat emerging variants of SARS-CoV-2 and thus warrant further investigations using in-vitro and in-vivo analysis. Communicated by Ramaswamy H. Sarma


Introduction
Severe Acute Respiratory Syndrome Coronavirus 2, the etiological agent of the unabating COVID-19 pandemic, was identified after an outbreak of viral pneumonia in the Wuhan region, Hubei province of China (Lu et al., 2020;Zhu et al., 2020).As of 20 th June 2022, the virus has led to 536,590,224 confirmed cases of infection and has caused 6,316,655 deaths globally (WHO, 2022).Clinical presentation of the disease in infected individuals can range from being asymptomatic to severe respiratory failure, cardinal symptoms include fever, fatigue, dry cough, myalgia, and dyspnea, there are also certain aberrant symptoms such as sputum production, hemoptysis, and diarrhoea (Alimohamadi et al., 2020;He et al., 2020;Park et al., 2020).
The viral genome of SARS-CoV-2 was reported to be mutating at a rate of �6.677 � 10 À 4 nucleotides per site per year as of January 2022 (Wang et al., 2022).As such, the virus has mutated into numerous variants since its outbreak, with WHO keeping track of prominent variants by classifying them as variants of concern and variants of interest (WHO, 2022); adaptive mutations may alter the pathogenic potential of a virus, even a single amino acid change can enhance the ability of a virus to evade the immune response (Giovanetti et al., 2021;Xu et al., 2020).Non-synonymous mutations in variants were observed to confer better affinity to host cell receptor and antigenic escape capability to the variant strains, as Andreano et al reported (Andreano et al., 2021), where the mutations E484K, N501Y, and K417N were able to generate an escape variant of SARS-CoV-2.Several anti-viral strategies have been reported against the wild-type i.e. the Wuhan strain of SARS-CoV-2, Since these mutations induce variations in the behaviour of viral variants (Janik et al., 2021;Sun et al., 2022), the therapeutic approaches targeting SARS-CoV-2 wild-type must also be validated against the emerging variants to determine their efficacy in curbing viral dissemination.SARS-CoV-2 genome size varies from 28.9 kb to 29.9 kb with a genome structure similar to that of orthodox coronaviruses i.e. the majority of the 5 0 end comprises ORF1ab, while the 3 0 end contains genes encoding structural proteins such as Spike, Envelope, Membrane, and Nucleocapsid proteins with 6 additional accessory proteins coded by ORF3a, ORF6, ORF7a, ORF7b, and ORF8 genes (Brant et al., 2021;Guo et al., 2020;Khailany et al., 2020;Naqvi et al., 2020).
The Spike glycoprotein as a whole protrudes from the surface of the viral membrane and consists of two domains; S1 and S2, wherein the S1 domain mediates viral cell entry by binding to the human Angiotensin Converting Enzyme 2 (hACE2) following which the protein is cleaved at the S1-S2 interface, enabling the S2 domain to fuse with the host cell membrane and achieve cell entry (Shang et al., 2020).The S1 subunit, being host to the RBD, is the likely target of action by neutralising antibodies and also inhibitory drugs (Gupta et al., 2021;Lan et al., 2020).Mutations in Spike protein have improved interaction with hACE2 and also a better ability to escape immune response (Souza et al., 2022).The mutated RBD in variants shows a higher affinity to the hACE2 receptor than wild-type (da Costa et al., 2022;Ortega et al., 2020).M pro cleaves the viral polyproteins pp1a and pp1ab, at 11 different definite sites, creating 12 functional proteins.It consists of three domains, Domains I, II, and III.Domain I (residues 8-100) and Domain II (residues 101-184) have chymotrypsin-like six-stranded antiparallel b sheets, due to which M pro is also known as 3-chymotrypsin-like protease (3CLpro) (Arya et al., 2021).M pro is relatively conserved across major variants with few mutations and none on the active site region used in drug repurposing approaches (Lee et al., 2022), thus making a desirable target to inhibit SARS-CoV-2 across major variants.Effective inhibition of these two targets can impede the viral life cycle and therefor is a promising strategy for antagonistic drug development against SARS-CoV-2 (Bafna et al., 2020;Cannalire et al., 2020;Jin et al., 2020).
Peptides often play crucial roles in human physiology, acting as hormones, neurotransmitters, growth factors, and antibacterial agents.Peptides have been used as drugs since 1921, when Insulin was isolated and recent advances have led to an increased potential for peptides to play a more crucial role in drug discovery (Henninot et al., 2018).Peptide inhibitors owing to their larger size, can interact with multiple active residues on the target protein and do not require any specific druggable pocket on the protein.They also disintegrate into amino acids and thus do not have the same risk of potential side effects as small molecules, making them an ideal choice for inhibiting protein-protein interactions (Tsomaia, 2015).Various peptide inhibitors have been reported with activity against Spike protein and M pro of SARS-CoV-2 wild-type.Here, we studied 17 such peptide inhibitors against major targets in SARS-CoV-2 variants; a, b, c, d, o, k, and m, to identify potential inhibitors that can act against their respective targets despite several non-synonymous mutations observed.To represent a spectrum of available anti-viral peptides we selected 5 peptides, based on the hACE2 receptor and 5 peptides from anti-viral peptide libraries, along with two de-novo-designed peptides against RBD and 5 antimicrobial peptides from peptide libraries against Mpro.

Sequence analysis and modelling of SARS-CoV-2 target proteins
The protein sequences of SARS-CoV-2 wild-type and variants of interest and variants of concern (as of March 1, 2022) were retrieved from National Centre for Biotechnology (NCBI) Virus database (Brister et al., 2015) and sequences were aligned using Multiple Alignment using Fast Fourier Transform (MAFFT) tool (Katoh et al., 2002).The crystal structure for SARS-CoV-2 RBD wild-type (PDB ID: 6LZG), Delta (PDB ID: 7WBQ), and Omicron (PDB ID: 7WBP) were retrieved from Protein Data Bank (PDB), along with the crystal structure for Main protease wild-type (PDB ID: 6Y2E).Homology modelling of variant proteins was performed using SWISS-MODEL (Schwede et al., 2003) to obtain their 3-Dimensional structure.

Molecular docking and binding free energy calculations
The 12 Selected peptide inhibitors were docked with wildtype and 7 variant structures of Spike-RBD, and 5 inhibitors were docked against M pro wild-type and the 3 variants with mutations (Beta, Omicron and Lambda) using the HADDOCK web server (Dominguez et al., 2003;Honorato et al., 2021).
The peptide-bound complex of RBD was also docked with the host cell receptor to study the obstruction of viral cell entry.The interactions were analyzed using LigPlot (Laskowski & Swindells, 2011) and all the visualizations were done in Pymol (DeLano, 2002).

Molecular dynamic simulations
MD simulation was performed for the selected complexes in the Desmond package of Schr€ odinger suite version 2021-1 for 100 ns (Bowers et al., 2006).The simulation systems were prepared by soaking in an orthorhombic box with SPC water molecules and then neutralized by adding sodium or chlorine ions.All parameters were assigned for proteins, solvents, and ions using the OPLS4 force field (Lu et al., 2021).The simulation systems were then equilibrated under NVT and NPT conditions to stabilize the temperature and pressure at 300 K and 1.01325 bars respectively.To assess the binding affinity of the selected complexes from MD simulation, MM/ GBSA binding free energy was calculated using the prime package in Schrodinger suite version 2021-1 (Jacobson et al., 2002;2004).Principal component analysis (PCA) was performed to analyze the essential dynamics from the simulated trajectories using Schrodinger.

Mutational profile of RBD and M pro in seven SARS-CoV-2 variants
Reconstruction of the mutational landscape of SARS-CoV-2 RBD reveals 21 non-synonymous mutations in a total of 18 distinct sites across 7 SARS-CoV-2 variants (a, b, c, d, o, k, and m).RBD of these variants had unique mutations at 13 different sites, wherein 11 mutations were present in the omicron variant (G339D, N440K, G446S, S371L, S373P, S375F, S477N, Q493R, G496S, Q498R, and Y505H) and one each in lambda (F490S) and Mu (R346K) (Figure 1A and B).However recurrent substitutions were observed at K417, L452, and E484 among variants, wherein Lysine at 417 was replaced by Asparagine in Beta and Threonine in Gamma and Omicron variants, Similarly, Leucine at 452 was substituted with Arginine and Glutamate in Delta and Lambda variants and glutamic acid at 484 was replaced by Lysine in Beta, Gamma, Mu and Alanine in the Omicron variant, these mutations (K417N/T, L452R/Q, E484K/A) are known to expressively increase the affinity of spike protein to the host cell receptor (Sanches et al., 2021).Among the variants, omicron has the highest number of mutations in the spike protein, more than twice the number of mutations in the delta variant (Tian et al., 2022), and the Omicron variant is reported to be resistant to RBD-targeted  neutralizing antibodies (Cao et al., 2022) because it includes 15 mutations in the RBD region alone.Several antibodies have their epitopes centred around K417 and mutations K417N/T thus leading to evasion of the immune response (Wibmer et al., 2021).E484K/A along with N501Y are two of the most common mutations observed in the variants, these mutations have been previously reported as increasing the binding affinity between RBD and hACE2 (Barton et al., 2021), also the E484K substitution is known to facilitate evasion from neutralizing antibodies (Andreano et al., 2021;Greaney et al., 2021,Weisblum et al., 2020).The significant number of mutations observed indicates that they could potentially affect the binding of potential inhibitors to RBD, thus highlighting the need to study the effect of these mutations on the binding of any potential inhibitor against SARS-CoV-2 variants.In contrast to RBD, the main protease of SARS-CoV-2 variants are highly conserved as only single mutations were seen in

Molecular docking analysis of 17 reported peptide inhibitors against major drug targets in SARS-CoV-2
In this study, 12 peptides from three distinct classes -hACE2 receptor-based (Pep13, Mod13AApi, Inhibitor 1, ABP D25Y , SPB25 F8N ), de-novo synthesized (LCB1, LCB3), and Antimicrobial peptides (2JPK-Mutant, 2KUY-Mutant, DBP6, S2P25, S2P26) (Supplementary Table 1) reported to have good binding affinity to the RBD region of SARS-CoV-2 wild-type were selected.Similarly, 5 antimicrobial peptides (2JOS, 1PXQ, 2KET, 2JPK-Mutant, 2KUY-Mutant) reported against wild-type M pro were also selected for our analysis.However, their affinities against RBD and M pro of SARS-CoV-2 variants remain shrouded since non-synonymous mutations in viral variants are known to alter the binding of potential inhibitors (Arya et al., 2021;Sun et al., 2022).Hence we screened 12 peptide inhibitors against the RBD of wild-type and 7 variants (a, b, c, d, o, k, and m), and 5 peptides against the wild-type and mutated variants of M pro (b, o, and k) to determine their binding affinities across variants and identify peptide inhibitors that negate the effect of non-synonymous mutations in the major drug targets of SARS-CoV-2 variants.
The Haddock scores of the hACE2 receptor with RBD was used as reference while evaluating the binding of inhibitors against RBD of wild-type and variants.4 peptide inhibitors, LCB1 (-160.6 kcal/mol), LCB3 (-150.4),2JPK-Mutant (-144.8kcal/mol), and Pep13 (-141.6 kcal/mol) had better binding affinities to wild-type RBD than the hACE2 receptor (-136 kcal/mol), and all of these peptides except 2JPK-mutant maintained their higher affinity across variants of SARS-CoV-2 in comparison to wild-type (Figure 2A).2JPK-Mutant has docking scores in the range of À 91.4 kcal/mol to À 163.4 kcal/mol and shows lower binding affinity than the hACE2 receptor to RBD of Beta, Gamma, Omicron, and Mu variants, However since 2JPK-mutant is an antimicrobial peptide obtained from lactococcus lactis (Rogne et al., 2008) and not specifically designed to target the RBM region, it remains an impressive candidate.The de novo synthesized peptide LCB1 had docking scores in the range of À 143 kcal/mol to À 155.7 kcal/mol in the 7 variants compared to that of the hACE2 receptor with variant RBDs that ranged from À 125.4 kcal/mol to À 145.4 kcal/mol.LCB1 forms hydrogen bonds with 13 residues of wild-type RBD (Figure 2B) and an average of 11 hydrogen bonds are maintained by LCB1 with the 7 variants (Supplementary Table 3).LCB1, like all other peptides screened against RBD, binds to the RBM region and engages key residues such as 417, 449, 453, 487, and 493 for the binding of hACE2 to RBD (Nelson-Sathi et al., 2022;Wang et al., 2020) and these interactions are maintained across variants (Figure 2C).Mutations at K417N, E484K, and N501Y are known to generate escape variants (Andreano et al., 2021) however LCB1 forms hydrogen bonds at these sites in at least 5 variants.Thus as LCB1 obstructs key residues required for RBD-hACE2 interaction across SARS-CoV-2 variants, it should be able to impede the attachment of RBD to the host cell receptor.
All 5 antimicrobial peptides that were screened against M pro showed consistent binding affinities to SARS-CoV-2 wildtype and variants as expected since the active sites were not mutated (Figure 3A).An antimicrobial peptide 2JOS derived from the mast cells of striped bass (Morone saxatilis) (Kawulka et al., 2004) was identified as having the highest affinity to wild-type and variants with their docking scores ranging from À 102.3 kcal/mol to À 112.5 kcal/mol, followed by another antimicrobial peptide 1PXQ with docking scores in the range of À 98.8 kcal/mol in wild-type to À 104.1 kcal/mol in the Omicron variant (Figure 3A).All of the screened peptides except 2KUYmutant interact with at least one dimerization site of M pro (Goyal & Goyal, 2020) in all variants (Figure 3B and C) and may obstruct the formation of the functional homodimeric form of Main protease (Ferreira et al., 2021;Su� arez & D� ıaz, 2020).The peptide inhibitor 2JOS interacts with E166 across all variants and F140 in wild-type and omicron variant (Supplementary Table 4) of M pro , which are the key residues involved in dimerization by directly interacting with other monomer and may be targeted to prevent dimerization and thus halt viral replication (Anand et al., 2003;Sharma et al., 2022), as such 2JOS chosen for further analysis.

Molecular dynamic simulations of RBD-LCB1 and m pro -2JOS complexes
The structural stability of RBD-LCB1 complexes of SARS-CoV-2 wild-type and 7 variants were analyzed using MD simulations of 100 ns each.The Root mean square Deviation (RMSD) graph indicates LCB1 remains stable when bound with RBD of wild-type and all 7 variants with an average RMSD of 1.902 Å (Figure 4A).The Root Mean Square Fluctuation (RMSF) graph shows that the fluctuation of individual residues is within the limit of 2 Å, wherein all variant RBDs showed an average fluctuation of 0.992 Å except for a higher fluctuation ranging from 1.629 Å to 5.375 Å in the extended loop region of Receptor Binding Motif in the beta variant (Figure 4B).It was observed that N487 in that region forms hydrogen bond interactions with Q7 of the peptide LCB1.This interaction was maintained for an average of 89% of the simulation period in wild-type and all variants except beta, where this interaction was maintained for only 40% of the simulation period (Figure 4C), which may cause the observed fluctuation.LCB1 forms and maintains 12 hydrogen bonds and 1 pi-pi stacking interaction with the wild-type RBD throughout the 100 ns simulation where 417 and 487 form multiple interactions with the peptide (Figure 4D) and 9 of these interactions are maintained across all 7 variants.LCB1 is able to maintain stability despite mutations in interacting residues, Mutations K417N/T in Beta, Gamma, and Omicron variants led to LCB1 losing interactions at S29 and D30 that were maintained with wild-type and non-mutated variants.LCB1 is known to be a potent inhibitor of SARS-CoV-2 wild-type with an IC 50 value of 23.54 (Cao et al., 2020), we expect a similar potency of viral neutralization to be maintained across the 7 considered variants of SARS-CoV-2.
Molecular dynamic simulations of 2JOS peptide with M pro wild-type, Beta, Omicron, and lambda variants affirm the structural stability of the complexes with an average RMSD of 2.4, 2.1, 2.4, and 2.9 Å respectively (Figure 5A).Root mean square fluctuation of M pro variants shows the flexibility of all individual residues within the range of 2 Å, thus reaffirming the structural stability of the protein-peptide complexes (Figure 5B).2JOS forms and maintains 3 hydrogen bonds and a salt bridge with wild-type M pro (Figure 5D), throughout the simulation period.Two of these interactions; E166-R7 and N142 are maintained in Beta and Omicron variants (Figure 5C), wherein E166 is a key dimerization site of M pro and 2JOS can thus obstruct the formation of the functional homodimeric form of M pro by forming a salt bridge with E166 in all variants except lambda, where 2JOS forms 3 hydrogen bonds with different residues of lambda , 2 of which are with the catalytic dyads; H41 and C145, by binding to which 2JOS can impede substrate binding (Bharadwaj et al., 2021;Du et al., 2004) and maintain its inhibition of M pro across all 3 variants.

Molecular docking and simulation analysis of RBD-LCB1 with hACE2 receptor
The RBD-LCB1 complex was docked to the hACE2 receptor, The stability of the resultant RBD-LCB1-hACE2 complex was assessed using Molecular dynamics simulation and further compared to the stability of RBD-LCB1 and RBD-hACE2 structures.The RBD-LCB1-hACE2 complex exhibited the least binding affinity (-96.4 kcal/mol) compared to that of the RBD-hACE2 (-136 kcal/mol) and RBD-LCB1 (-160.6 kcal/mol).Molecular dynamics simulations were performed to assess the stability of these complexes and we found that RBD-LCB1 in complex with hACE2 is rather unstable and that peptide-bound RBD withdraws from the hACE2 receptor interface (Figure 5A-C), similar to that observed by Awad et al.where a silodosin bound RBD was easily withdrawn from the hACE2 receptor (Awad et al., 2022).This is evident in the average RMSD value of 11.717 Å for the RBD-LCB1-hACE2 complex (Figure 5D), which is a 6-fold increase in RMSD from that of the RBD-hACE2 complex (1.9184 Å), hence we observe that RBD is unable to form a favourable complex with hACE2 in presence of LCB1, we also found that LCB1 forms a more favourable complex with RBD in terms of RMSD (1.853 Å) than the RBD-hACE2 complex.The binding energy of the RBD-LCB1, RBD-hACE2 and RBD-LCB1-hACE2 complexes were calculated using MM/GBSA method from the MD Simulation trajectories, wherein the bindings were thermodynamically favourable and entities such as Van der Waals energy, Solvation energy and Electrostatic interactions along with H-bonds contributed to the complex formation.We observed that the RBD-LCB1 complex has a slightly higher binding energy (-120.70kcal/mol) than the RBD-Ace2 complex (-120.02kcal/mol) and that the RBD-LCB1-Ace2 complex was rather unfavourable with a binding energy value of just À 18.275 kcal/mol, (Table 1) Indicating that LCB1 forms a more favourable complex with hACE2 and in the presence of LCB1, interactions between RBD and hACE2 are unfavourable.
To study the structural and conformational changes in the RBD-LCB1, RBD-hACE2 and RBD-LCB1-hACE2 complexes, simulation trajectories of these complexes were analyzed using Principal Component Analysis (Aier et al., 2016).From the 10 Principal components generated, the PC1 and PC2 were plotted since the first two principal components are adequate to analyze more than 60% of motion of the complexes (Rathi et al., 2022).The scatter plot of PCA shows RBD-LCB1-hACE2 complex occupies most of the space and scattered in the conformational space reveals its highly unstable nature (Figure 6E).Due to the lower flexibility of RBD-LCB1, It occupied less conformational space than RBD-hACE2, showing that the RBD-LCB1 complex is more stable than RBD-hACE2.

Conclusions
In this study, we have considered 17 known peptide inhibitors against the SARS-CoV-2 Wuhan strain and studied them against major drug targets such as Spike glycoprotein and Main protease to identify potential inhibitors that can combat emerging variants of SARS-CoV-2.Reconstruction of the mutational landscape revealed that the Spike-RBD harboured a significant number of non-synonymous mutations in its active sites which could affect the binding of potential inhibitors in mutated variants, In contrast, Main-protease was relatively conserved with only 3 non-synonymous mutations across major variants and none in its active sites making it a prime target in therapeutic approaches.We screened 12 previously reported peptide inhibitors against RBD and 5 inhibitors against M pro in major variants of b,c,d,o,k,and m).Molecular docking and simulation studies revealed a de-novo designed peptide; LCB1 and an antimicrobial peptide; 2JOS with high binding affinities to their respective targets; RBD and M pro across major SARS-CoV-2 variants as potential inhibitors.LCB1 was able to prevent viral cell entry by binding to the RBM region on the RBD that interacts with the hACE2 receptor and thus impeding RBD-hACE2 interaction.2JOS binds to key residues of M pro involved in dimerization or by interacting with the catalytic dyad to prevent substrate binding in major variants of SARS-CoV-2, Molecular dynamic simulation studies validated these findings by affirming the stability of RBD-LCB1 and M pro -2JOS complexes across variants.Also, we observed that Spike-RBD forms a more favourable complex with LCB1 rather than hACE2 in terms of stability and that LCB1-bound RBD is unlikely to form a stable complex with the hACE2 receptor and may impede viral cell entry.We believe that the peptide inhibitors LCB1 and 2JOS are promising candidates against emerging variants of SARS-CoV-2 and warrant further in-vitro and in-vivo analysis to develop them into potential therapeutic agents.Since LCB1 and 2JOS have been analyzed against multiple prominent variants of SARS-CoV-2 and were found to be effective in binding to their targets despite mutations, We believe that these peptides, apart from being effective against existing variants of SARS-CoV-2, could still effectively bind to emerging variants of SARS-CoV-2 and thus be a potential drug against the whole spectrum of SARS-CoV-2 variants for the foreseeable future and Further development of these drugs can yeild a high-affinity, neutralizing agent against future outbreaks of SARS-CoV-2 variants.

Figure 1 .
Figure 1.Mutational landscape of Spike RBD and M pro in SARS-CoV-2 variants.(A) Conservation of SARS-CoV-2 RBD region across seven variants.The Receptor Binding Motif (RBM) region is highlighted in blue, mutations are marked in red, and interacting residues of the RBD-hACE2 complex are represented with an asterisk.(B) 3-Dimensional structure of RBD with mutations marked.(C) Conservation of M pro across variants.Domain I, Domain II, Loop, and Domain III were colourcoded as brown, green, purple, and blue respectively with active site residues marked with an asterisk.Catalytic dyad residues His41 and Cys145 are highlighted in grey and amino acid residues involved in dimerization were also highlighted (D) 3-Dimensional structure of M pro with mutations marked.

Figure 2 .
Figure 2. Molecular docking analysis of 12 peptide inhibitors with SARS-CoV-2 Spike RBD.(A) Docking scores of peptides bound to RBD wild-type and variants.(B) Interacting residues of RBD with LCB1 peptide across variants.RBD and LCB1 are colour coded as green and blue respectively and the interacting residues of RBD with LCB1 are shown in stick representation.(C) Interactions of top scoring peptides with RBD across variants.Asterisk represents residues forming a hydrogen bond with the hACE2 receptor.

Figure 3 .
Figure 3. Molecular docking analysis of 5 peptide inhibitors with M pro (A) Docking scores of peptides bound to M pro wild-type and 3 variants having mutations in M pro .(B) Interacting residues of M pro with 2JOS peptide.M pro and 2JOS are colour coded as green and blue respectively and the interacting residues of M pro with 2JOS are shown in stick representation.(C) Interactions of the peptides with M pro across variants and dimerization sites are marked in red.
three variants (b, o, and k) that reside in Domain I (K90R-Beta, G15S-Lambda) and Domain II (P132H-Omicron) (Figure 1C and D).None of the residues in dimerization sites, catalytic dyad, or the active sites of M pro variants are mutated, and hence the effect of potential inhibitors should be maintained across variants as observed by Ulrich et al. (Ullrich et al., 2022), and Vangeel et al.(Vangeel et al., 2022)

Figure 4 .
Figure 4. Molecular dynamics simulation summary of RBD-LCB1 complex.(A) RMSD graph of RBD-LCB1 Complex for SARS-CoV-2 Variants and wild-type.(B) RMSF graph of RBD for SARS-CoV-2 Variants and wild-type.(C) Interactions between RBD and LCB1 are depicted wherein hydrogen bonds, Salt bridges, and Pi-Pi stacking interactions are colour coded in black, blue, and green respectively.Mutated residues were also coloured in red (D) Cartoon representation of LCB1 in complex with RBD, Interacting residues of wild-type RBD with LCB1 were enlarged.

Figure 5 .
Figure 5. Molecular dynamics simulation summary of M pro -2JOS complexes.(A) RMSD graph of M Pro -2JOS Complex for SARS-CoV-2 wild-type and variants.(B) RMSF graph of M pro for SARS-CoV-2 wild-type and variants.(C) Interactions between M pro and 2JOS are depicted wherein hydrogen bonds, salt bridges are listed in black and blue respectively.(D) Cartoon structure of 2JOS bound to M pro .Interacting residues of wild-type M pro were zoomed.

Table 1 .
MM/GBSA based free energies calculated from MD simulation trajectories.