key: cord-0749053-4vac7kv1
authors: Çakır, Bilal; Okuyan, Betül; Şener, Göksel; Tunalı-Akbay, Tuğba
title: Investigation of beta-lactoglobulin derived bioactive peptides against SARS-CoV-2 (COVID-19): in silico analysis
date: 2020-12-01
journal: Eur J Pharmacol
DOI: 10.1016/j.ejphar.2020.173781
sha: 11cc17b069770922296cf94a9d00652ab68ab67d
doc_id: 749053
cord_uid: 4vac7kv1

The coronavirus disease of 2019 (COVID-19) outbreak caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which started in late 2019 in Wuhan, China spread to the whole world in a short period of time, and thousands of people have died due to this epidemic. Although scientists have been searching for methods to manage SARS-CoV-2, there is no specific medication against COVID-19 as of yet. Two main approaches should be followed in the treatment of SARS-CoV-2; one of which is to neutralize the virus, and the other is to inhibit the host cell membrane receptors, where SARS-CoV-2 will bind. In this study, peptides derived from beta-lactoglobulin, which inactivates both the virus and its receptors in the host cell, were identified using computer-based in silico analysis. The beta-lactoglobulin derived peptides used in this study were obtained by the treatment of goat milk whey fraction with trypsin. The structure of the peptides was characterized by the liquid chromatography quadrupole time-of-flight mass spectrometry (LC-Q-TOF/MS), and six beta-lactoglobulin derived peptides were selected as candidate peptides. Subsequently, the effects of peptides on SARS-CoV-2 and host cells were identified using virtual screening. According to the results of this in silico analysis, Ala-Leu-Pro-Met-His-Ile-Arg (ALMPHIR) and Ile-Pro-Ala-Val-Phe-Lys (IPAVFK) peptides were evaluated as potential candidates to be used in the treatment of SARS-CoV-2 after the future in vitro and in vivo studies.

Viral diseases are still serious threats to public health all over the world. The major viral outbreaks (severe acute respiratory syndrome coronavirus [SARS-CoV] [MERS-CoV] , 2012) occurring in the last 20 years have caused serious deaths and unfortunately, the treatment is still unclear (Bradley and Bryan, 2019) . World Health Organization (WHO) declared coronavirus disease of 2019 COVID-19 infection, which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a pandemic on March 11, 2020 (WHO, 2020 . To date, neither specific therapy nor a vaccine has been invented against COVID-19.

The current management of COVID-19 is aimed to prevent and control the infection via empirical treatment methods and supportive care.

Once the structure of SARS-CoV-2 has been identified (Kong et al., 2020) , researchers have focused on SARS-CoV-2 spike proteins and its main protease using in silico or experimental methods for neutralizing the virus or inhibiting its cellular entrance. The spike protein (S) of SARS-CoV-2, a trimeric transmembrane glycoprotein, recognizes the host cell and mediates its entry. It contains a receptor-binding domain (S1) and a domain that mediates the integration of the virus and host cell membranes (S2). For SARS-CoV-2 to enter cells, host cell proteases, furin, a calcium-dependent serine endopeptidase found in the host lung cell membrane, affects the S1/S2 cleavage site of the SARS- CoV-2 (Coutard et al., 2020) . Due to this interaction, cleavage of S protein leaves the S2 domain exposed, and the transmission of After the SARS-CoV-2 pandemic has appeared and the virus structure has been determined, new potential drugs that can interact with ACE2-related genes or prevent the virus from entering the host cell have been studied by in silico analysis (Cava et al., 2020) . Peptide-like inhibitors, which are used in the management of various diseases, could also be potential agents against COVID-19. Pant et al. (2020) evaluated 300 peptide-like structures and determined sixty-six FDA-approved drugs by using a medication re-purposing approach (such as lopinavir, ritonavir, darunavir) against COVID-19 in silico. Another in silico study investigated an α-ketoamide inhibitor, a peptide-like inhibitor against the crystal structure of SARS-CoV-2 main protease and its complex (Zhang et al., 2020) .

Milk proteins are composed of approximately 80% caseins and 20% whey proteins. The hydrolysis of these fractions yields various peptides. Although there are many studies related to the health benefits of casein-based peptides, the peptides of whey proteins have not fully been characterized yet. Whey bioactive peptides are beneficial to the cardiovascular system through in vitro ACE-inhibitory activity and in vivo antihypertensive effects. Furthermore, a DPP-4 inhibitor derived from goat milk casein has been shown to have an effect on the nervous system, immune system, and endocrine system (Sharma et al., 2017; Kitts and Weiler, 2003; Möller et al., 2008; Atanasova and Ivanova 2010; Zhang et al., 2015;  2. Materials and Methods

Goat milk samples were obtained from Saanen-Maltız crossbred goats of the same age that were grown at a private farm in the Marmara Region of Turkey. Goat milk was kept at +4 o C after morning milking.

The milk was defatted by centrifugation at 3000 xg for 15 min at 4 °C followed by precipitation of caseins by adjusting the medium pH to 4.6 using 1 M HCl. Subsequently, the precipitated caseins were removed by centrifugation at 19000 xg for 1 h at 4 °C, and the resulting whey fraction was collected in the form of supernatant. Trypsin was used to digest the whey fraction. Whey fraction proteins were mixed with trypsin at 37°C in 20:1 (wt/wt) ratio at pH 8. After digestion, each sample was heated at 98°C for 10 min to inactivate the protease (Abubakar et al., 1998) .

For protein extraction, the trypsin digested whey fraction was mixed with Universal Protein Extraction (UPX) Kit (Expedeon-44101) and protease inhibitor cocktail (Thermo Sci.-87785).

Samples were sonicated at 200 g and then boiled at 95°C for 10 min. After the boiling procedure, samples were centrifuged for 10 min at 20000 g. Peptide production was performed utilizing FASP Protein Digestion Kit (Expedeon-44250) and trypsin treatments (Pierce-90057) on the supernatant. Samples were diluted to 200 ng/µl with 0.1% formic acid.

Detector and calibration settings were made with the MassLynx program (V4.1-Waters) that is specific to Xevo G2-XS Q-TOF-MS (Waters Corp., Milford, MA, USA) device where the analysis was carried out. The tryptic peptides were fractionated with an acetonitrile gradient (5-35%) in HSS T3 (Waters-186008818) column and analyzed by mass spectrometry upon J o u r n a l P r e -p r o o f electrospray ionization. Peptide data was identified in an m/z range of 50-1950. MS analysis was performed for 0.7 seconds and the data was collected from the entire peptide. UniProt protein database and Progenesis software were used for peptide identification.

The BIOPEP-UWM database (available at http://www.uwm.edu.pl/biochemia/index.php/ pl/biopep ) (Minkiewicz et al., 2008) was used to predict the relevant inhibitory activities of the obtained peptides identified from LC-Q-TOF-MS analysis on the ACE and DPP-4 enzymes, the receptors to which SARS-CoV-2 binds on the cell surface. IPP (Isoleucine-Proline-Proline) peptide was used as a positive control (Watanabe et al., 2015) for inhibition of ACE and IPI (Isoleucine-Proline-Isoleucine; Diprotin A) peptide was used as a positive control (Song et al., 2017) for inhibition of DPP-4.

PepSite2 available at http://pepsite2.russelllab.org) (Trabuco et al., 2012) was used to predict the peptide-binding sites of in silico identified ACE and DPP-4 inhibitor peptides to ACE and DPP-4 enzymes.

ToxinPred (available at https://webs.iiitd.edu.in/raghava/toxinpred/index.html) (Gupta et al., 2013) was used to assess characteristic properties of beta-lactoglobulin derived peptides and predict its toxicity.

The beta-lactoglobulin derived peptides were docked with SARS-CoV2 proteases (PDB:6LU7 and PDB:6M03), spike proteins (PDB:6VSB and PDB:6VXX) and host cell furin (PDB:1P8J), angiotensin-converting enzyme (ACE, PDB: 2YDM), and dipeptidyl peptidase-https://bioserv.rpbs.univ-paris-diderot.fr/ services/ pepATTRACT/) (Schindler et al., 2015) and GalaxyPepDock (available at http://galaxy.seoklab.org/ pepdock ) (Lee et al., 2015) servers. The HPEPDOCK is a server that investigates the protein-peptide docking based on the hierarchical algorithm. The Galaxy webserver was also used for confirming the output of the HPEPDOCK server. These docking scores in the HPEPDOCK server reflect the prediction of protein-ligand binding affinity by free energy in kcal/mol units (Zhou et al., 2018; Tao et al., 2020) . GalaxyPepDock gives the similarity-based docking results by running energy-based optimization that allows structural flexibility. pepATTRACT is another peptide docking server that globally searches the entire protein surface, identifies the binding site, and predicts the bound peptide conformation.

Six beta-lactoglobulin (BLG) derived peptides (i.e., ALPMHIR (Ala-Leu-Pro-Met-His-Ile-Arg; BLG 160-166 ), GLDIQK(Gly-Leu-Asp-Ile-Gln-Lys; BLG 27-32 ), TPEVDK(Thr-Pro-Glu-Val-Asp-Lys; BLG 143-148 ), IPAVFK(Ile-Pro-Ala-Val-Phe-Lys; BLG 96-101 ), EALEK(Glu-Ala-Leu-Glu-Lys; BLG 149-153 ), and IIAEK(Ile-Ile-Ala-Glu-Lys; BLG 89-93 )) were characterized from the goat milk whey fraction. The characteristics of these beta-lactoglobulin derived peptides were presented in Table 1 . MS spectrum that corresponds to the ionization of ALPMHIR, IPAVFK, and GLDIQK peptides was demonstrated in Fig 2. 

The ACE and DPP-4 inhibitory activity scores of the beta-lactoglobulin derived peptides are presented in Table 2 . According to the data obtained using the BIOPEP server, the frequency of a peptide in a protein sequence and potential biological activity of the peptide demonstrates the potential of the peptide to become a bioactive peptide.

Although all 6 peptides have been found to demonstrate both ACE and DPP-4 inhibitory activity; IPAVFK, IIAEK, and ALPMHIR displayed the higher ACE and DPP-4 inhibitory activities when compared to the other three peptides (Table 2) .

The amino acids of the beta-lactoglobulin derived peptides that potentially could interact with ACE (PDB:2YDM) and DPP-4 (3WQH) were listed in Table 3 and Table 4 . Table 3 and Table 4 also present the binding points of each peptide to the ACE and DPP-4 enzymes, respectively.

For ACE and DPP-4 inhibition, 3 different peptides display the best protein-peptide interaction (considering p values and potentially bound amino acids) were presented. Binding potential of ALPMHIR (P <0.0001), IPAVFK (P <0.001), IIAEK (P <0.001), and TPEVDK (P <0.001) peptides to the ACE enzyme was found to be statistically significant according to data obtained using the PepSite2 database (Table 3) . ALPMHIR, IPAVFK, IIAEK, and TEPVDK appear to interact with all the amino acids that IPP bind on ACE enzymes. Furthermore, the ALPMHIR, IPAVFK, and TEPVDK peptides appeared to bind other than these amino acids that IPP binds on the ACE (Table 3) .

Binding potential of ALPMHIR (P <0.001), IPAVFK (P <0.05), and TPEVDK (P <0.01) peptides to the DPP-4 enzyme was also found to be statistically significant according to data obtained using the PepSite2 database (Table 4 ). As a result of protein-peptide interactions obtained concerning the DPP-4 inhibitor IPI peptide, ALPMHIR peptide was observed to bind to the same amino acids as the IPI peptide (Table 4 ). IPAVFK and TEPVDK peptides appear to bind all amino acids that IPI bind. However, these peptides bind to 4 different amino acids on DPP-4, where IPI did not bind on DPP-4 (Table 4 ). Table 5 presents the docking scores based on the minimum energy of the beta-lactoglobulin derived peptides and main protease (apo and holo forms) and spike protein of SARS-CoV-2 and host cell furin analysis with the HPEPDOCK.

According to Table 5 , among the 6 peptides obtained from the beta-lactoglobulin fraction, the best docking score value was obtained from ALPMHIR (Fig 3) . The second-best score was obtained with IPAVFK.

According to the data obtained from the pepATTRACT program, the peptide molecules with the lowest binding energy level were ALPMHIR, IPAVFK, and GLDIQK. All betalactoglobulin derived peptides had lesser binding energy levels than that obtained from remdesivir and hydroxychloroquine. The interfacial propensity of peptides to interact with the main protease of SARS-CoV-12 and the energy values of these interactions were demonstrated in Table 6 .

In GalaxyPepDock server, the interaction of the beta-lactoglobulin derived peptides with SARS-CoV-2 (COVID-19) 6M03 (Apo) and 6LU7 (Holo) enzyme, and lung 1P8J (Furin) were investigated in terms of protein structure similarity (TM-score), interaction similarity score, and estimated accuracy.

The structure similarity scores (TM) for the protein-peptide interaction, which were above 0.98, were shown in Table 7 . According to the protein-peptide docking results with 6LU7 and J o u r n a l P r e -p r o o f 6M03 proteins, the peptides with an interaction similarity score greater than 50 and an estimated accuracy score greater than 0.70 are IPAVFK, EALEK, ALPMHIR, IIAEK, and GLDIQK peptides. For the 1P8J protein, only ALPMHIR is the peptide that interacts in this score range (Table 7) . Fig 4 shows the docking of SARS-CoV-2 6LU7 protein with the ALPMHIR peptide.

The global epidemic caused by the new type of human coronavirus, SARS-CoV-2, is a concern for all humanity. In order to find an effective treatment method for COVID-19, all the aspects of SARS-CoV-2 are still being investigated all over the world.

Since peptides have recently been considered as promising agents against various pathologic conditions (Baig et al., 2018) , the design of peptide-derived drugs often requires structural characterization of the underlying protein-peptide interactions. The present in silico study aims to evaluate whether the beta-lactoglobulin derived peptides could be effective against the main protease enzyme and spike protein of SARS-CoV-2 and host cell furin. Betalactoglobulin derived peptides demonstrated various biological effects that include the inhibition of ACE and DPP-4 (Pérez and Calvo, 1995; Pihlanto-Leppälä, 2000; Walzem et al., 2002) . However, there is no study related to their potential antiviral effects in the literature.

Thus, the present hypothesis-driven peptidomics study focused on the antiviral effects of goat milk beta-lactoglobulin derived peptides. The LC-Q-TOF/MS method was used to determine peptide sequences hidden in goat milk-based beta-lactoglobulin and ALPMHIR, IPAVFK, GLDIQK, IIAEK, EALEK, and TPEVDK were selected which demonstrated inhibitory effects on ACE and DPP-4 activities based on BIOPEP calculations.

The potential therapeutic target would be the disruption of binding of SARS-CoV-2 to host cell protease by beta-lactoglobulin derived peptides. This present in silico docking analysis was performed to find a probable ligand-receptor confirmation between the host cell furin, J o u r n a l P r e -p r o o f SARS-CoV-2 main protease, and spike protein and beta-lactoglobulin derived peptides by using pepATTRACT, HPEPDOCK, and Galaxy Pep DOCK databases.

In many viruses, proteases play essential roles in viral replication. Therefore, proteases are often used as targets during the investigation of new antiviral agents. When beta-lactoglobulin derived peptides and both holo and apo forms of SARS-CoV-2 main protease were docked, ALPMHIR, IPAVFK, and GLDIQK were found to demonstrate the lowest docking scores.

Since the main protease is essential for the maturation of the SARS-CoV-2 and has been examined as a potential target protein, ALPMHIR, IPAVFK, and GLDIQK showed a predicted inhibitory effect on SARS-CoV-2 main protease.

Furthermore, when spike protein of SARS-CoV-2 and beta-lactoglobulin derived peptides were docked, our results have shown that these beta-lactoglobulin derived peptides could bind to the SARS-CoV-2 spike protein (PDB:6VXX) and its receptor binding site (PDB:6VSB).

ALPMHIR peptide had the lowest docking score among the other peptides, indicating better interactions with spike proteins of SARS-Co-2. Therefore, ALPMHIR might prevent the binding of SARS-CoV-2 to the host cell.

The spike protein of SARS-CoV-2 contains a potential cleavage site for furin. Furin is a proprotein convertase that proteolytically cleaves SARS-CoV-2 spike protein at the S1-S2 boundary by inserting 12 nucleotides to that site (Shang et al., 2020) . According to the virtual screening data obtained in this study, beta-lactoglobulin derived these six peptides bound to the furin with low docking scores. Among them, ALPMHIR, IPAVFK, and GLDIQK peptides, which demonstrated the lowest score, should be evaluated as candidate agents in terms of furin inhibition during COVID-19.

Human coronaviruses have been shown to bind to host cell receptors that include furin, ACE2, and DPP-4 and to use them to invade target cells (Millet and Whittaker, 2015; Bosch et al., 2014) . Iacobellis (2020) stated that DPP-4 might be a potential target to prevent multi-J o u r n a l P r e -p r o o f organ damage and reduce the risk of acute respiratory complications in patients with type 2 diabetes during COVID-19 infection.

Dilemma on the risk of using ACE inhibitors during the COVID-19 outbreak in patients having the concurrent cardiovascular disease has been discussed due to potential enhancement of the entry of SARS-CoV-2 to host cells via the upregulation of ACE2. However, it is well known that Ang 1-7 produced by ACE2 has beneficial effects in the renin-angiotensinaldosterone system and could prevent multi-organ damage by counteracting the harmful effects of AII (Li et al., 2020; Warner et al., 2004; South et al., 2020) . Similarly, Liu et al.

(2020) demonstrated that viral load and lung injury were associated with an increase of serum angiotensin II levels in patients with COVID-19.

In conclusion, this in silico study used a hypothesis-driven peptidomics strategy for selecting the beta-lactoglobulin derived peptides that inhibit the ACE, DPP-4, and furin enzymes in the host cell and SARS-CoV-2 main protease and that bind to the spike protein of SARS-CoV-2. J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f Table 3 . Predicted binding of the beta-lactoglobulin derived peptides on the protein surface from ACE from Homo sapiens. Ile-3  0.0025  TYR48, VAL546, TYR547, TRP627,   GLY628, TRP629, SER630,   TYR631, TYR666, GLY741, HIS748, TYR752 *: P value presents the statistical significance. DPP-4 inhibitor IPI (Diprotin A) peptide was used as positive control (Song et al. 2017) J o u r n a l P r e -p r o o f Residue: The number of amino acid that the peptide bound on the protein, Contacts:

Interface propensity, PDB codes; 6M03: The crystal structure of COVID-19 main protease in apo form 6LU7: The crystal structure of COVID-19 main protease in complex with an inhibitor N3

J o u r n a l P r e -p r o o f 

Structural analysis of new antihypertensive peptides derived from cheese whey protein by proteinase K digestion

Antibacterial peptides from goat and sheep milk proteins

Peptide based therapeutics and their use for the treatment of neurodegenerative and other diseases

Membrane ectopeptidases targeted by human coronaviruses

Emerging Respiratory Infections: The infectious disease pathology of SARS, MERS, pandemic influenza, and Legionella

The ACE2/angiotensin-(1-7)/MAS axis of the reninangiotensin system: focus on angiotensin

Fully blind peptide-protein docking with pepATTRACT

Cell entry mechanisms of SARS-CoV-2

Identification of Bioactive Peptides in Goat Milk and Their Health Application

Identification of dipeptidyl peptidase-IV inhibitory peptides from mare whey protein hydrolysates

COVID-19, ACE2, and the cardiovascular consequences

PepSite: prediction of peptidebinding sites from protein surfaces

Emerging WuHan (COVID-19) coronavirus: glycan shield and structure prediction of spike glycoprotein and its interaction with human CD26

Whey components: millennia of evolution create functionalities for mammalian nutrition: what we know and what we may be overlooking

What's new in the renin-angiotensin system?: Angiotensin-converting enzyme-2: a molecular and cellular perspective

The influence of dietary peptide inhibitors of angiotensin-converting enzyme on the hypotensive effects of enalapril

The role of furin cleavage site in SARS-CoV-2 spike protein-mediated membrane fusion in the presence or absence of trypsin

Post-proline cleaving enzyme and post-proline dipeptidyl aminopeptidase. Comparison of two peptidases with high specificity for proline residues

Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved α-ketoamide inhibitors

Isolation and Identification of Dipeptidyl Peptidase IV-Inhibitory Peptides from Trypsin/Chymotrypsin-Treated Goat Milk Casein Hydrolysates by 2D-TLC and LC-MS/MS

HPEPDOCK: a web server for blind peptideprotein docking based on a hierarchical algorithm