Phycobilins as Potent Food Bioactive Broad-Spectrum Inhibitors Against Proteases of SARS-CoV-2 and Other Coronaviruses: A Preliminary Study

In the 21st century, we have witnessed three coronavirus outbreaks: SARS in 2003, MERS in 2012, and the ongoing pandemic coronavirus disease 2019 (COVID-19). The search for efficient vaccines and development and repurposing of therapeutic drugs are the major approaches in the COVID-19 pandemic research area. There are concerns about the evolution of mutant strains (e.g., VUI – 202012/01, a mutant coronavirus in the United Kingdom), which can potentially reduce the impact of the current vaccine and therapeutic drug development trials. One promising approach to counter the mutant strains is the “development of effective broad-spectrum antiviral drugs” against coronaviruses. This study scientifically investigates potent food bioactive broad-spectrum antiviral compounds by targeting main protease (Mpro) and papain-like protease (PLpro) proteases of coronaviruses (CoVs) using in silico and in vitro approaches. The results reveal that phycocyanobilin (PCB) shows potential inhibitor activity against both proteases. PCB had the best binding affinity to Mpro and PLpro with IC50 values of 71 and 62 μm, respectively. Also, in silico studies with Mpro and PLpro enzymes of other human and animal CoVs indicate broad-spectrum inhibitor activity of the PCB. As with PCB, other phycobilins, such as phycourobilin (PUB), phycoerythrobilin (PEB), and phycoviolobilin (PVB) show similar binding affinity to SARS-CoV-2 Mpro and PLpro.

Till now, there are no approved vaccines and therapeutic drugs for COVID-19 or other human coronavirus infections and a lack of enough clinical trial data to make treatment decisions. Although vaccines have been developed against animal viruses IBV, canine CoV, and TGEV to help prevent serious diseases (Park et al., 1998;Carmichael, 1999;Liu and Kong, 2004), there are some potential problems, such as recombination events between field and vaccine strains, the emergence of novel serotypes, and antibody-dependent enhancement remain. The rapid development of vaccines and repurposing of approved antivirals drugs (e.g., remdesivir) are major clinical approaches of a pandemic preparedness plan. The development of broadspectrum antiviral agents that are effective against a wide range of CoVs and other classes of viruses, including emerging ones, could be a promising strategy (Bekerman and Einav, 2015;Fauci and Morens, 2016;Cho and Glenn, 2020).
Broad-spectrum antiviral targeting strategies can be classified into two categories: (i) entry inhibitors that interact with existing virus particles outside of cells and prevent infection (Hangartner et al., 2006) and (ii) replication inhibitors aimed at stopping viral genome replication to curtail production of new virus particles (De Clercq, 2004). The S glycoprotein of coronaviruses, the main determinant of host cell attachment and viral entry, is not well conserved between HCoVs (Totura and Bavari, 2019). On the other hand, CoV non-structural proteins (nsps) are highly conserved components of the coronavirus life cycle that mediate viral replication (Totura and Bavari, 2019). Literature studies report the following SARS-CoV-2 nsp targets; main protease (M pro ), papain-like protease (PL pro ), Nsp3, RdRp, Helicase, Nsp14, Nsp15, Nsp16, N protein to inhibit virus replication (Wu C. et al., 2020). Proteolytic processing of viral polyproteins into functional nsps by two viral proteases, the M pro and PL pro , is an important event of the CoV life cycle. The M pro acts on minimum 11 cleavage sites of replicase 1ab, ∼790 kDa; at recognition sequence Leu-Gln↓ (Ser, Ala, Gly) (↓ indicates cleavage site), most cleavage sites block viral replication . PL Pro enzyme hydrolyses the peptide bond at the carboxyl side of glycine (P1 position) and releases nsp1, nsp2, and nsp3 functional proteins, which play a key role in viral replication (Rut et al., 2020). Therefore, these proteases would be potential targets for the development of broad-spectrum antiviral drugs. CoVs M pro and PL pro crystal structures are available for public access in the protein data bank (PDB).
Natural food bioactive compounds are gaining importance as supplementary antiviral therapeutic compounds in the modern healthcare sector because of their lower toxicity and fewer side effects, additional health benefits (antioxidant, antiinflammatory, and immunomodulation activities), and potential use in conjunction with preexisting therapies. Several literature studies report antiviral properties of food bioactive compounds against CoVs and other viruses ( Table 1; Ghildiyal et al., 2020;Mani et al., 2020). In view of the issues posed above, identifying natural food bioactive broad-spectrum antiviral agents against the CoVs is a more reasonable and attractive prospect and could provide an effective first line of defense against future emerging CoVs related diseases. Herein, we report the phycobilins as potent food bioactive broad-spectrum inhibitor compounds against M pro and PL pro of SARS-CoV-2 and other CoVs via in silico and in vitro approaches.

Preparation of Protein and Ligand for Docking
The crystal structures of M pro (PDB ID -6LU7) and PL pro of SARS-CoV-2 (PDB ID -6WUU) and other CoVs used in this study were obtained from the RCSB PDB. Ligand structures were obtained from Pubchem and Chemical Entities of Biological Interest (ChEBI) as SDF format, Open Babel was used for format transformation or 3-D coordinate generation for the uploaded files (O'Boyle et al., 2011). The MGLTools were used to delete other chains, and heteroatoms (included water), adding missing atoms, hydrogens, and charges. Further, the pdbqt files were prepared for proteins and ligands binding.

Molecular Docking and Molecular Simulation Studies
Autodock Vina was used as a docking engine. It is critical to define the docking grid box appropriately due to the small molecule docking procedure (Trott and Olson, 2010). The docking box is defined as the center of native ligand coordinates with 40 Å × 40 Å × 40 Å in length to include the residues of the entire cavity, and the exhaustiveness level was set on 12 with number of modes 10. For visualization, the docking results PDBQT files were exported, and docked protein-ligand complex structures were visualized using Pymol. Active site residues within 3 or 3.5 Å of ligand and polar contacts were determined with this same tool. The ligand docking procedure was validated by redocking of the native Thymol Exhibited significant antiviral activity with an IC50 of 7 µM against herpes simplex virus type I. Lai et al., 2012 ligand with the same protocol and the grid parameters as used for food bioactive compounds. The redocked ligand was then superimposed onto the reference co-crystallized ligand complex using Pymol, and the root mean square deviation (RMSD) was analyzed. Molecular dynamic simulations were performed using NAMD (Phillips et al., 2005). The parameters, structure, and topology files for the ligand were generated using the CHARMM-GUI Web server (Jo et al., 2008). Visual molecular dynamics (VMD) was used to generate protein structure (PSF) files (Humphrey et al., 1996). Each protein-ligand docked complex was solvated and ionized with 0.15 M ions (Na+ and Cl−) to neutralize the charge and electrostatic screening. The systems were subjected to 2000 steps of steepest descent energy minimization before a production run at the NPT of 0.5 ns (250,000 steps). The temperature (310 K) and pressure (1 atm) were controlled by the Langevin and Langevin piston methods (Feller et al., 1995). VMD software was used to visualize simulations and to analyze average ligand-RMSD and protein-RMSD and hydrogen bonds.

M pro Assay
Fluorescence resonance energy transfer (FRET)-based cleavage assay (Zhu et al., 2011) was used for in vitro enzyme inhibition study. SARS-CoV-2 M pro or 3CL Protease, GenBank Accession No. YP_009725301, amino acids 1-306 (full length), with an N-terminal MBP-tag, expressed in an Escherichia coli and its fluorescent substrate with cleavage site (indicated by ↓) DABCYL-KTSAVLQ↓SGFRKME-EDANS, inhibitor control (GC376), and the assay buffer composed of 20 mM Tris, 100 mM NaCl, 1 mM EDTA, 1 mM DTT, pH 7.3 were used. Initially, 15 µL of the SARS-CoV-2 M pro in reaction buffer at the final concentration of 10 ng/µL and 5 µL of inhibitor control (GC376, final concentration 50 µM)/test inhibitor (10-600 µM)/inhibitor solvent (positive control) was pipetted into a 384-well plate. Stock solutions of the compounds were prepared with 100% DMSO. Afterward, the plate was preincubated for 30 min at room temperature with slow shaking. The enzymatic reaction was then initiated by adding of 5 µL of the substrate dissolved in the reaction buffer to 25 µL final volume (final concentration 50 µM) and incubated at room temperature for 4 h. The fluorescence signal of the Edans generated due to the cleavage of the substrate by the M pro was monitored at excitation at 360 nm with an emission wavelength of 460 nm, using a spectrophotometric microplate reader (Synergy H1 Hybrid Multi-Mode Reader; BioTek Instruments, Inc., Winooski, VT, United States).

PL pro Assay
Severe acute respiratory syndrome coronavirus-2 PL pro (papainlike protease), GenBank Accession No. QHD43415, amino acids 1564-1882, with N-terminal His-tag, expressed in an E. coli and its fluorescent substrate Z-Arg-Leu-Arg-Gly-Gly-AMC, inhibitor control (GRL0617) and the assay buffer (40 mM Tris pH 8, 110 mM NaCl, 1 mM DTT) was used for the inhibition assay. Briefly, 30 µL of the SARS-CoV-2 PL pro in reaction buffer at the final concentration of 0.44 ng/µL and 10 µL of inhibitor control (GRL0617, final concentration 100 µM)/test inhibitor (10-600 µM)/inhibitor solvent (positive control) was pipetted into a 96-well plate. Afterward, the plate was preincubated for 30 min at room temperature with slow shaking. The enzymatic reaction was then initiated by the addition of 10 µL of the substrate dissolved in the reaction buffer to 50 µL final volume (final concentration 25 µM), incubated at room temperature for 40-60 min. The fluorescence signal of the substrate after the enzymatic reaction was monitored at an excitation at 360 nm with an emission wavelength of 460 nm, using a spectrophotometric microplate reader (Synergy H1 Hybrid Multi-Mode Reader; BioTek Instruments, Inc., Winooski, VT, United States). Triplicate experiments (N = 3) were performed for both M pro and PL pro assays, and the mean value was presented with ± standard deviation (SD).

In silico Binding Interaction Studies of Selected Phytochemical Compounds With SARS-CoV-2 M pro and PL pro
The 16 selected phytochemicals were docked into the active site pocket of SARS-CoV-2 M pro and PL pro . Table 2 depicts the source, docking score, and polar contacts of selected phytochemical bioactive compounds with binding site amino acid residues of SARS-CoV-2 proteases. For M pro , the results show PCB docked with the best score or binding energy of −8.6 Kcal/mol followed by Riboflavin (−7.9 Kcal/mol), Cyanidin (−7.9 Kcal/mol), Daidzein (−7.8 Kcal/mol), and Genistein (−7.6 Kcal/mol). Twelve key active-site amino acid residues (Tyr 54, Gly 143, His 163, Asp 187, Gln 189, Glu 166, Cys 145, Leu 141, Ser 144, Thr 26, Gln 192,and Thr 190) of SARS-CoV-2 M pro involved in polar interactions at a distance of ≤3 Å with ligand phytochemical compounds. Specific polar contacts of each phytochemical compound are shown in Table 2. In the case of PL pro , as the reported peptide inhibitor VIR250 is bound to the dimer interface in the crystal structure of 6WUU (Rut et al., 2020), the docking studies were performed with dimer form. Similarly, PCB docked with the best score or binding energy of −9.8 Kcal/mol followed by Astaxanthin, (−9.3 Kcal/mol), β-carotene (−9.  Eggs, meat, fruits, and vegetables Riboflavin −7.9 E166, C145, H163, L141, S144  PL pro are involved in polar interactions at a distance of ≤3 Å with ligand phytochemical compounds. Table 2 illustrates the specific polar contacts between phytochemical compounds and proteases. Figure 1 shows a 3-D representation of the binding pocket of M pro and PL pro with top score model pose of PCB. The co-crystalized structure of native 6LU7-N3 and 6WUU-VIR250 complexes and polar contacts are represented in Supplementary  Figures 1a, 2a. The docking validation studies reveal that both N3 peptide inhibitor and VIR250 bound exactly to the active site of 6LU7 and 6WUU, respectively. Superimposed redocked N3 on to the native co-crystallized N3 show a low RMSD of 1.82 Å was observed, whereas redocked VIR250 had a RMSD of 2.096 Å (Supplementary Figures 1b, 2b). These results show less variation in comparison with the native binding pose of ligands in co-crystallized form.
To evaluate the reliability of the molecular docking and stability of docked complexes, we conducted molecular dynamics simulations with a traditional force field (CHARMM36m), starting with the docking-generated pose of M pro -ligand complexes with binding energy cutoff values of −7.6 Kcal/mol, PL pro -ligand complexes with binding energy cut-off values of −8.0 Kcal/mol accompanying with a higher binding affinity toward M pro . Root mean square deviation (RMSD) was measured to measure the structural conformation differences between the initial and final positions of proteins and ligands. Table 3

In vitro Enzymatic Assay Studies to Screen Potent Phytochemical Inhibitor Compounds Against SARS-CoV-2 M pro and PL pro
To further validate the molecular docking and molecular dynamics studies, in vitro enzymatic studies were conducted. A positive control without the inhibitor compound in the reaction mixture, an inhibitor control that contains authentic inhibitors GC376 (for M pro ), GRL0617 (for PL pro ) were used in this study. The enzyme's relative activity in the presence of inhibitors was estimated by considering positive control activity as 100%. Based on in silico studies, we selected the top six phytochemicals (PCB, quercetin, riboflavin, cyanidin, daidzein, and genistein) with a binding energy cutoff value of −7.6 Kcal/mol for M pro enzymatic assay studies. Initial screening results revealed that PCB had higher inhibitor activity followed by quercetin, genistein, cyanidin, and riboflavin (p < 0.05) (Figure 2). Further, the IC 50 value of top two compounds, PCB and quercetin, was determined, and the results show an effective IC 50 value of 71 µM for PCB (Figure 3) than quercetin (145 µM) (Supplementary Figure 3). For PL pro , four compounds (Phycocyanobilin, Riboflavin, Genistein, and Quercetin) with a binding energy cutoff value of −8.0 Kcal/mol, accompanying higher inhibitor activity toward M pro , were selected for the in vitro inhibitor activity assay (Figure 2). It was envisaged that PCB showed potent inhibitor activity compared to other  compounds (Figure 2), with an IC 50 value of 62 µM (Figure 3). Overall, in silico docking and in vitro enzyme inhibitor activity data show PCB as a potent inhibitor against SARS-CoV-2 M pro and PL pro .
In silico Insights of Broad-Spectrum Inhibitor Activity of PCB Against M pro and PL pro The broad-spectrum efficacy of PCB against CoVs was evaluated by molecular docking studies with available crystal PDB structure of various human and animal CoVs. Table 4 shows the PDB identification code and top docking scores of PCB with M pro and PL pro enzymes of human and animal CoVs. Due to the limitation on the availability of crystal PDB structures of PL pro , both dimer and monomeric forms were used in docking studies. For M pro , docking scores are in the range of −8.3 to −9.3 Kcal/mol. PCB showed higher binding affinity with docking score (−9.3 Kcal/mol) for MERS M pro followed by HCoV NL63 (−9.0 Kcal/mol) and IBV (−8.9 Kcal/mol). For PL pro , docking scores were in the range of −8.9 to −7.6 Kcal/mol. The results reveal that PCB had a higher binding affinity to the dimer form of PL pro enzymes than monomeric forms. When compared monomers only, PCB had best docking score for MERS-CoV (−8.5 Kcal/mol) followed by TGEV (−8.1 Kcal/mol) and SARS-CoV-2 (−8.0 Kcal/mol). Supplementary Figures 4, 5 show polar contacts of PCB with binding pocket key amino acid residues of M pro and PL pro enzymes of human and animal CoVs. Surprisingly, the docking results suggest PCB as a promising broad-spectrum food bioactive inhibitor compound against CoVs proteases.

In silico Insights Into Inhibitor Activities of Other Phycobilins
Phycobilins are linear tetrapyrrole chromophore compounds found in certain photosynthetic organisms (cyanobacteria, red algae, glaucophytes, and some cryptomonads) and covalently linked to phycobiliproteins (Beale, 1993). Four types of phycobilins are identified: (i) phycoerythrobilin (PEB), (ii) phycourobilin (PUB), (iii) phycoviolobilin (PVB), and (iv) PCB. Figure 4 represents the molecular structures of phycobilins. Based on the PCB results, the other phycobilin inhibitor activity against SARS-CoV-2 proteases via molecular docking approach was demonstrated and docking scores, polar contacts are given in Table 5.

DISCUSSION
Several SARS-CoV-2 mutants are spreading globally; most notably, mutants emerged in the United Kingdom (B.1.1.7), South Africa (B.1.351), and Brazil (P.1) (CDC, 2021). Hence, the efficacy of currently developed vaccines against these variants is questionable. The development of broad-spectrum antiviral drugs against existing viruses is an attractive approach that could provide first-line defense against emerging viral variants. The selection of highly conserved drug targets is an important step to develop broad-spectrum antiviral drugs. Sequence alignment analysis of SARS-CoV-2 and other CoVs (SARS-CoV and MERS) reveals that M pro and PL pro are highly conserved, especially in the functional regions, which makes them potential targets for COVID-19 drug targets (Wu C. et al., 2020). Several target-based in silico screening approaches were performed to find promising protease inhibitors among repurposed drugs, natural phytochemicals, and herbal medicinal compounds against COVID-19 (Amin et al., 2020;Wu C. et al., 2020;Xian et al., 2020). In this study, we focused on screening natural bioactive compounds for potential inhibitor activity against SARS-CoV-2 proteases. Amin et al. (2020) report the list of key amino acids involved in catalysis and substrate binding for M pro (H41, C145, M49, G143, S144, H163, H164, M165, E166, L167, D187, R188, Q189, T190,  A191, and Q192) and PL pro (C111, H272, D286, Y268, M208, P247, P248, T301, P248, Y264, N267, Q269, L162, C270, G271, and Y273). Our molecular docking results with natural compounds also show polar interactions in these specific substrate-binding regions for both proteases. Further in vitro validation studies found a good correlation between docking and enzymatic assay results. Both in silico and in vitro approaches show that PCB has strong inhibitor activity against both SARS-CoV-2 proteases. Shih et al. (2003) report direct antiviral activity of allophycocyanin against enterovirus 71 in human rhabdomyosarcoma cells and African green monkey kidney cells. In another study, El-Morsi et al. (2016) demonstrate the reduction of X174 and MS2 by extracted phycobiliproteins from cyanobacterium Synechococcus cedrorum Sauvageau.
Though numerous research studies report potential inhibitors with major emphasis on SARS-CoV-2, limited literature is available on broad-spectrum inhibitors against a wide range of CoVs, including human and animal CoVs. For instance, Sheahan et al. (2020) report broad-spectrum antiviral activity of ribonucleoside analog β-D-N4-hydroxycytidine (NHC; EIDD-1931) against SARS-CoV-2, SARS-CoV, MERS-CoV, and bat-CoVs. This study reports the broad-spectrum activity of natural phytochemical compound PCB against 11 CoVs (seven human CoVs and four animal CoVs). The computed physical properties of phycocyanobilin show a rotatable bond count of 10, hydrogen bond donor count of five, and hydrogen bond acceptor count of seven (National Center for Biotechnology Information (NCBI), 2020), which makes multiple hydrogen bond interactions  between the compound and specific amino acid residues located at the active site of the pocket of the wide range of protease enzymes. Molecular docking studies indicated that propionic carboxyl and lactam ring carbonyl oxygens of PCB are involved in polar interactions with proteases' amino acid residues.
To investigate other structurally similar phycobilin (PEB, PUB, and PVB) inhibitor activity, we attempted molecular docking with SARS-CoV-2 proteases. The results reveal that, like PCB, all phycobilins show similar binding affinity toward M pro and PL pro of SARS-CoV-2. Besides this, potent therapeutic properties, such as peroxy radical scavenging, inhibition of cancer cell proliferation, and platelet aggregation are reported for phycobilins (Watanabe et al., 2014). Phycobilin compounds can be directly administered orally as phycobiliproteins Frontiers in Microbiology | www.frontiersin.org (a complex of phycobilins and protein). For instance, when phycocyanin is administered orally to humans, it can be digested and free phycocyanobilin released in the gastrointestinal tract (Watanabe et al., 2014). Thus, noticed therapeutic properties of phycobiliproteins might reflect the effects of their phycobilins (chromophores).
In conclusion, by using in silico (molecular docking and MD simulations), in vitro enzymatic assay screenings, we discovered PCB as potent phytochemical inhibitors to M pro and PL pro proteases of SARS-CoV-2. Phycocyanobilin had IC 50 values of 71 and 62 µM for SARS-CoV-2 M pro and PL pro , respectively. Further PCB docking studies with other CoVs Mpro and PLpro proteases revealed its broad-spectrum inhibitor activity. A similar binding affinity of other phycobilins (PEB, PUB, and PVB) to these proteases were observed. However, in vitro enzymatic studies with M pro and PL pro of other CoVs and in vivo studies on the inhibition of CoVs infectivity using human cells and animal models are needed. Further structure-guided development of phycobilin lead compounds could rapidly lead to discovering a single agent with clinical potential against existing and possible future emerging CoV-associated diseases.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.