Hindawi Evidence-Based Complementary and Alternative Medicine Volume 2020, Article ID 4932572, 14 pages https://doi.org/10.1155/2020/4932572 Research Article Investigating Potential Inhibitory Effect of Uncaria tomentosa (Cat’s Claw) against the Main Protease 3CLpro of SARS-CoV-2 by Molecular Modeling Andres F. Yepes-Pérez ,1 Oscar Herrera-Calderon ,2 José-Emilio Sánchez-Aparicio ,3 Laura Tiessler-Sala ,3 Jean-Didier Maréchal ,3 and Wilson Cardona-G 1 1 Chemistry of Colombian Plants, Institute of Chemistry, Faculty of Exact and Natural Sciences, University of Antioquia-UdeA, Calle 70 No. 52-21, A.A 1226, Medellin, Colombia 2 Academic Department of Pharmacology, Bromatology and Toxicology, Faculty of Pharmacy and Biochemistry, Universidad Nacional Mayor de San Marcos, Jr Puno 1002, Lima 15001, Peru 3 Insilichem, Departament de Quı́mica, Universitat Autònoma de Barcelona, Edifici C.n., 08193 Cerdanyola del Vallés, Barcelona, Spain Correspondence should be addressed to Andres F. Yepes-Pérez; andresf.yepes@udea.edu.co, Oscar Herrera-Calderon; oherreracalderon@gmail.com, and Jean-Didier Maréchal; jeandidier.marechal@uab.cat Received 7 May 2020; Accepted 25 July 2020; Published 1 October 2020 Guest Editor: Zheng Feei Ma Copyright © 2020 Andres F. Yepes-Pérez et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. COVID-19 is a disease caused by severe acute respiratory syndrome coronavirus 2. Presently, there is no effective treatment for COVID-19. As part of the worldwide efforts to find efficient therapies and preventions, it has been reported the crystalline structure of the SARS-CoV-2 main protease Mpro (also called 3CLpro) bound to a synthetic inhibitor, which represents a major druggable target. The druggability of Mpro could be used for discovering drugs to treat COVID-19. A multilevel computational study was carried out to evaluate the potential antiviral properties of the components of the medicinal herb Uncaria tomentosa (Cat’s claw), focusing on the inhibition of Mpro. The in silico approach starts with protein-ligand docking of 26 Cat’s claw key components, followed by ligand pathway calculations, molecular dynamics simulations, and MM-GBSA calculation of the free energy of binding for the best docked candidates. The structural bioinformatics approaches led to identification of three bioactive compounds of Uncaria tomentosa (speciophylline, cadambine, and proanthocyanidin B2) with potential therapeutic effects by strong interaction with 3CLpro. Additionally, in silico drug-likeness indices for these components were calculated and showed good predicted therapeutic profiles of these phytochemicals. Our findings suggest the potential effectiveness of Cat’s claw as complementary and/or alternative medicine for COVID-19 treatment. 1. Introduction (COVID-19)) caused by SARS-CoV-2 is considered a pandemic, affecting Asia and Europe with the highest death Severe acute respiratory syndrome coronavirus 2 (SARS- rate followed by America and other regions, causing serious CoV-2) is a part of coronavirus (CoV) family and was public health problems and considerable economic losses initially identified in Wuhan, China, at the end of December worldwide [2]. Respiratory viral infections are the frequent 2019. COVID-19 is highly contagious and is most frequently causes of morbidity and millions of hospital admissions in transmitted from human to human, spreading the virus developing countries every year [3]. For this reason, the easily to other countries in a very short time [1]. According pharmacotherapy based on natural products may be a to the last report of the World Health Organization (WHO), proper alternative for treating viral diseases. On the other the severe acute respiratory syndrome (coronavirus disease hand, traditional medicine is practiced by native South 2 Evidence-Based Complementary and Alternative Medicine American inhabitants who know the medicinal properties of attachment in the host cells was characterized as the main many plants from the rainforest [4]. Therefore, many of mechanism of its antiviral activity [27]. Furthermore, them are collected by ethnobotanists who investigate their other investigations mentioned immunomodulating ac- resources as antimicrobial [5] and antitumor agents and tivity which includes stimulation of phagocytosis, en- being the main source for target selection during scientific hancement of B- and T-lymphocytes, suppression of NF- investigation on compounds with antiviral activity [6]. The kappa B, and enhancement of IL-1 and IL-6 [28, 29]. In a biodiversity of South America countries [7] offers a series of Peruvian study on rats, the investigators found that medicinal plants which could combat the symptoms of phagocytosis was increased and might act as the potent infections such as coronavirus disease COVID-19. inhibitor of TNF-α [30]. In 2008, a study evidenced a Uncaria tomentosa (Willd. ex Schult.) DC. named Cat’s possible drug-drug interaction between Cat’s claw and claw (“uña de gato” in Spanish) is a woody vine indigenous protease inhibitors such as saquinavir, atazanavir, and to the Peruvian Amazon and other tropical areas of South ritonavir, increasing level of these drugs in plasma [31]. and Central America and belongs to Rubiaceae family [8]. In Individuals supplemented with a novel water-soluble ex- Peru, natives from the villages in the region of Chanchamayo and Nevati near Puerto Bermudez, Junin, boil approximately ® tract of Uncaria tomentosa (C-Med-100 ) showed in- creased effectiveness of pneumococcal vaccination as a 20 g of sliced root bark in 1 liter of water for 45 min result of an increase the lymphocyte/neutrophil ratios of commonly and use it in religious purposes and curatives [9]. peripheral blood and a reduced decay in the 12 serotype This species is marketed as dry and ground material re- antibody titer responses to pneumococcal vaccination markably in recent years, being widely studied in all aspects [32]. mainly by chemical (isolated chemical structures) and With neither drugs nor vaccines approved against SARS- bioassays in silico, in vitro, in vivo and some medical trials. CoV-2 yet, finding strategies to diminish the impact of the Currently, the raw material of U. tomentosa is dispensed in pandemic is fundamental. Medicinal herbs and, more Public Hospitals of the Social Health Insurance (EsSalud- particularly, those demonstrating antiviral activities are Peru) as Complementary Medicine Service (CMS) [10]. possible allies in this quest. Their use could slow down the Traditionally, extracts prepared by root and bark decoction spreading of the disease. Particularly in developing coun- are used against several diseases, such as allergies, arthritis, tries, in which the accessibility to these plants is easier and inflammations, rheumatism infections, and cancer [11]. more economically viable, adding these medicinal herbs to However, its popular use has been widely known worldwide, the general medical kit may be beneficial. In addition, tra- and analytical tests can be found in the United States ditional knowledge of these remedies may reduce possible Pharmacopeia and dosage forms have been authorized only side effects, allowing them to be implemented with fewer as dietary supplements. Currently, there are about 34 species medical risks [33]. of Uncaria, with Uncaria tomentosa being the most common On the basis of the aforementioned background of species [12]. Uncaria tomentosa (Cat’s claw), this work aims at compu- Bioactive constituents of U. tomentosa extracts include tationally identifying potential bioactive compounds against proanthocyanidins (proanthocyanidin B2 or epicatechin- COVID-19. It focuses on possible interactions and inhibi- (4β ⟶ 8)-epicatechin, the main component; proantho- tion of the 3CLpro protease (also called Mpro). 3CLpro is cyanidin B4, proanthocyanidin C1, an epicatechin trimer, responsible for 100% of the proteolytic mechanism of the epiafzelechin-4β ⟶ 8-epicatechin, and an epicatechin tet- virus and is involved in virulence, infectivity, transcription, ramer) [13], oxindole alkaloids (isopteropodine, pter- and replication cycle of the virus [34, 35]. It has been opodine, rhynchophylline, mytraphylline, speciophylline, identified as the main druggable target of SARS-CoV-2 for uncarine F, and uncarine E) [14], indole alkaloidal gluco- new antiviral discovery. Moreover, its X-ray structure has sides (cadambine, 3-dihydrocadambine, and 3-iso- been recently released, hence allowing possible computa- dihydrocadambine) [15, 16], quinovic acid glycosides, tional analysis. In fact, several computational studies have tannins, polyphenols, catechins, beta-sitosterol, and proteins already been undertaken on this system including a long which individually or synergistically contribute to their 20 μs molecular dynamics (MD) study and virtual screening therapeutic properties [8, 17–24]. of several databases [36, 37]. In regard to the antiviral properties of U. tomentosa, Here, our study stands on a multilevel computational the alkaloid fraction has been demonstrated to be the most strategy reminiscent to those applied at the early stage of effective on human monocytes infected with dengue virus- current state-of-the-art drug discovery pipelines and in- 2 (DENV) in vitro [25]. Another study revealed that only cludes (1) protein-ligand docking of all bioactive compounds the alkaloidal fraction has inhibitory activity on dengue of Cat’s claw against 3CLpro structure, (2) simulations of the virus, and the negative effect was observed with the ligand pathway of the best predicted compounds from step 1 nonalkaloidal fraction [26]. In another study, the anti- to evaluate the convenient entrance mechanism of the herpetic activity of U. tomentosa seems to be associated compounds to the binding site, (3) molecular dynamics with polyphenols or their synergistic effect with penta- simulation to assess the stability of the best protein-ligand cyclic oxindole alkaloids or quinovic acid glycosides [27]. complexes from step 2, (4) calculation of the free energy of U. tomentosa hydroethanolic extracts demonstrated a binding based on MD postprocessing (MM-GBSA), and (5) significant in vitro inhibitory effect on the replication of calculation of pharmacokinetics parameters for the most herpes simplex virus type 1, and the inhibition of viral qualified compounds resulting from the previous parts of the Evidence-Based Complementary and Alternative Medicine 3 protocol. The study leads to identification of at least three Table 1: Best binding energy (kcal/mol) based on AutoDock compounds with potential antiviral activity in Cat’s claw- scoring of the main constituents of Cat’s claw into the cleavage site based products to propose the extract of Uncaria tomentosa of the novel SARS-CoV-2 main protease (PDB ID: 6LU7). as a rapid phytotherapeutic option for COVID-19. Main constituents of Cat’s claw Best binding energy (kcal/mol) Spiroxindole alkaloids 2. Materials and Methods Speciophylline −8.1 Isopteropodine −6.6 2.1. Protein Structure and Setup. Calculated binding affinity Isomitraphylline −7.6 of the main constituents of the Cat’s claw extracts (Table 1) Uncarine F −8.2 was explored against the main protease 3CLpro of SARS- Mitraphylline −7.0 CoV-2 findings, a facile therapeutic option for anti-coro- Pteropodine −7.0 naviral therapy; the crystal structure of the protease 3CLpro Rhynchophylline −5.9 was downloaded from the Protein Data Bank (PDB entry Isorhynchophylline −6.1 code 6LU7) [38]; and all bounded ligands, ions, and solvent Indole glycoside alkaloids molecules were manually removed using the DS Visualizer Cadambine −8.6 2.5 program. For docking studies, the structure of the se- 3-Dihydrocadambine −8.0 lected protein was parameterized using AutoDock Tools 3-Isodihydrocadambine −8.0 [38]. Gasteiger partial charges were calculated, and polar Polyhydroxylated triterpenes hydrogens to facilitate the formation of hydrogen bonds PHT-1 −6.8 Uncaric acid −7.0 were added. Floridic acid −7.6 Quinovic acid glycosides 2.2. Ligand Dataset Preparation and Optimization. QAG-1 −7.8 Ligands used in this study are major components of the Cat’s QAG-2 −7.4 claw extracts, a potent irreversible inhibitor recently re- QAG-3 −7.2 ported for COVID-19 virus 3CLpro (namely, N3) [39], and QAG-4 −7.9 QAG-5 −7.8 three well-known FDA-approved viral protease inhibitors QAG-6 −7.8 that may be repurposed to treat COVID-19 [40–46]. The 2D structures of 26 Cat’s claw constituents were obtained as Proanthocyanidins Chlorogenic acid −6.8 mol.2 files from the ZINC database [47]. The resultant Epicatechin −7.2 compounds were submitted to MarvinSketch 8.3 [48] to Proanthocyanidin B2 −9.2 correct the protonation states of the ligands at physiological Epiafzelechin-4β-8-epicatechin −8.9 pH 7.4 and its structures were parameterized using Auto- Proanthocyanidin B4 −9.2 dockTools to add full hydrogens to the ligands, to assign Proanthocyanidin C1 −8.8 rotatable bonds, and to compute Gasteiger charges and save References the resulting structure in the required format for use with N3a −8.1 (−8.1)b, (−8.3)c, (−7.9)d AutoDock. All possible flexible torsions of the ligand Remdesivirh −8.5 molecules were defined using AUTOTORS in PDBAuto- Ritonavirh −8.1 (−7.7)e, (−7.5)f, (−8.9)g DockTools [49] to promote the calculated binding with the Lopinavirh −8.0 (−8.4)e, (−7.4)f, (−9.4)g a SARS-CoV-2 protease structure. Potent irreversible inhibitor of SARS-CoV-2 virus 3CLpro. bBinding affinity reported by [47]. cBinding affinity reported by [77]. dBinding affinity re- ported by [78]. eBinding affinity reported by [79]. fBinding affinity reported 2.3. Docking-Based Virtual Screening. Our docking protocol by [80]. gBinding affinity reported by [3]. hFDA-approved antiviral drugs was performed using AutoDock 4.2 with the Lamarkian promising to treat SARS-CoV-2 [81–83]. genetic algorithm and default procedures to dock a flexible ligand to a rigid protein. Docking simulation was carried out Affinity scores (in kcal/mol) given by AutoDock Vina for all on the main protease 3CLpro of the SARS-CoV-2 cleavage compounds were obtained and ranked based on the free site (PDB code: 6LU7), where the enzyme residues are in energy binding theory (more negative value means greater proximity to the recently reported potent inhibitor, known binding affinity). The resulting structures and the binding as N3, which was co-crystallized in complex with the main docking poses were graphically inspected to check the in- SARS-CoV-2 protease. Once a potential binding site was teractions using DS Visualizer 2.5 (http://3dsbiovia.com/ identified, 26 compounds which are the major components products/) or PyMOL Molecular Graphics System 2.0 of the Cat’s claw extracts were docked to this enzyme site to programs [50]. determine the most probable and the most energetically favorable binding conformations. To accomplish rigorous docking simulations involving a grid box to the identified 2.4. GPathFinder Calculations. To assess the feasibility of the catalytic site, Autodock Vina 1.1.2 was used [49]. The ex- binding route for the ligand, ten runs of GPathFinder [51] haustiveness was 20 for each protein-ligand pair (number of calculations for each ligand-protease complex were per- internal independent runs). The active site was surrounded formed. The ligand was placed at the position obtained from by a docking grid of 42 Å3 with a grid spacing of 0.375 Å. docking calculations and required to search for a possible 4 Evidence-Based Complementary and Alternative Medicine unbinding route outside the protease. Full flexibility was considered as convergence indicators [47]. Considering the allowed for the ligand along the pathway including rota- alpha carbons of the protease backbone, RMSD from the meric states of amino acid side chain and backbone motions minimized structure, all-to-all frames RMSD, and cluster based on some previous MD simulations of 10 µs. A clus- counting analysis were performed. Moreover, to ensure that tering was carried out by means of the quality threshold dynamic transitions occur between different conformations, methodology [52] to obtain a pool of 110 representative a principal component analysis (PCA) was carried out by frames from a 10 µs simulation of the apo structure [53]. This plotting the two principal modes relative to each other. pool of representative frames was used as the possible The distance from the geometric center of the ligand to conformations that the protease could adopt during the the center of the binding site was computed along all the ligand transport. Vina score and steric clashes were mini- trajectories. Alpha carbons of the residues GLY143, CYS145, mized to obtain a total of 120 solutions for each run. A HIS163, HIS164, GLU166, GLN189, and THR190 were complete input file to use with version 1.2.1 of the software is considered to calculate the center of the binding site, and the provided in Supplementary Materials (available here). distance from the crystallographic inhibitor N3 to that binding site center (3.947 Å) was taken as reference for comparison. 2.5. Molecular Dynamics Simulations. All the molecular dynamics (MD) simulations were carried out using the dimeric structure of the COVID main protease available in 2.6. Prediction of Drug-Likeness Properties for the Most the PDB [53] (code 6lu7) as a receptor. The best-scored Docking Promissory Compounds. Drug-likeness prediction docking positions were used as starting conformations for along with further ADME properties presents a wide range the ligands. Two ligands (one for each monomer) were of opportunities for a rapid new antiviral drug discovery. placed at symmetric positions of the dimeric structure. The The drug-like and ADME properties for the most active protein was prepared by removing waters and crystallo- components of the Cat’s claw extracts (constituents having graphic small molecules from the PDB structure to have the the highest binding affinity) were screened using open-ac- protease residues free of interaction with the ligands during cess cheminformatic platforms such as Molinspiration (for the simulation. Finally, protons were then added through the molecular weight (MW), rotatable bonds, and polar surface algorithm implemented in UCSF Chimera Software [54]. area (PSA) descriptors), ALOGPS 2.1 (for log Po/w de- MD simulations were set up with the LEaP, which was scriptor), and Pre-ADMET 2.0 to predict four pharma- instructed to solvate the protein with a cubic box of pre- ceutically relevant properties such as intestinal permeability equilibrated TIP3P water molecules and balance the total (App. Caco-2), albumin-binding proteins (KHSA), Madin- charge with Na + ions (ions94.lib library). The AMBER14SB Darby Canine Kidney (MDCK Line) cell permeation, and force field [55] was used for the standard residues, while the intestinal absorption (%HIA). These parameters establish GAFF force field was adopted for the remaining atoms. The movement, permeability, absorption, and action of potential geometries of the ligands were optimized in water solvent drugs [61–64]. The interpretation of both MDCK and Caco- (SMD continuum model) at DFT level of theory using 2 permeability using PreADMET [65] is as follows: Gaussian09 [56]. Geometry and frequency calculations were (1) Permeability lower than 25: low permeability performed using the B3LYP hybrid functional with 6- 31G(d,p) as basis set and included Grimme’s dispersion [57]. (2) Permeability between 25 and 500: medium Atomic charges were computed using the RESP protocol permeability (restrained electrostatic potential) at the same level of (3) Permeability higher than 500: high permeability theory, and the atom types and force constants and equi- librium parameters were assigned using antechamber and 3. Results and Discussion parmchk2 from AmberTools18 [58]. For all the MDs, the solvent and the whole system were 3.1. Database of Cat’s Claw Bioactive Compounds. This study sequentially submitted to 3000 energy minimization steps to was performed to identify if certain components of the Cat’s relax possible steric clashes. Then, thermalization of the claw extracts have potential therapeutic effects against system was achieved by increasing the temperature from COVID-19. To do so, a database of 26 compounds that have 100 K up to 300 K. MD simulations under periodic boundary shown prevalence on the herbal therapeutic activity has been conditions were carried out for 250 ns with the OpenMM generated (Figure 1) [8, 9, 13, 23, 25, 29, 66–72]. Our initial engine [59] using OMMProtocol [60]. Three simulations hypothesis is that Cat’s claw should contain molecules with were run in total, considering the following protease-ligand highest therapeutic profiles against SARS-CoV-2, because of systems: speciophylline, cadambine, and proanthocyanidin their interaction with 3CLpro main protease. B2. Currently, some of this chemical components have been Analysis of the trajectories was carried out by means of isolated and synthesized, such as cadambine [73], spe- CPPTraj implemented in AmberTools18. The MD trajectory ciophylline [74], and proanthocyanidin B2 [75], and several was considered converged when a stable exploration of the bioassays were performed for their derived products. conformational space was achieved. In particular, a stable However, pure drugs that are industrially produced or conformation or a pool of relative stable conformations isolated from plants may be an option for their high bio- visited for a statistically consistent number of times were logical activity, but they could have some disadvantages. Evidence-Based Complementary and Alternative Medicine 5 H H O N H O N H N H O N N N H N N O O O HN O HN H NH OH O HN H H NH OH OMe NH OH OMe O H H H H H NH O H O H H O O O NH O O O O O O O O O O O O O Speciophylline Isopteropodine Isomitraphylline Uncarine F Mitraphylline Pteropodine Rhynchophylline Isorhynchophylline (a) OH N OH N N H OH H OH O H OH O O O NH H NH H OH NH OH H O OH H O H O H3CO O H3CO O H3CO O OH OH O OH O O OH OH OH Cadambine 3-Dihydrocadambine 3-Isodihydrocadambine (b) H3C H3C H3C HO HO HO COOH COOH COOH H HO HO HO OH CHOOH CH2 OH Uncaric acid PHT-1 Floridic acid (c) OH OH HO O OH OH OH O OH O HO O OH OH HO O HO OH O OH OH HO O OH O O OH OH OH OH OH HO OH OH OH OH OH OH OH OH O HO OH O HO O HO OH OH OH OH HO OH OH HO OH OH O OH OH HO OH OH Chlorogenic acid Epicatechin OH OH OH OH Proanthocyanidin B2 Epiafzelechin-4β-8-epicatechin Proanthocyanidin B4 OH Proanthocyanidin C1 (d) OH OH H OH OH OH OH O HO H O H O OH OH O HO O H O O O O OH OH O O O H OH HO O HO HO O OH OH O O H OH HO O O O HO O OH OH OH HO O O HO OH OH QAG-1 QAG-3 QAG-2 OH OH OH HO OH O OH OH H OH HO OH OH O HO O HO O HO O O H OH H H O O O O O HO O OH O H O OH H HO OH HO OH HO O OH HO O HO O O QAG-6 HO QAG-4 OH QAG-5 (e) Figure 1: 2D structures for the major bioactive constituents of Cat’s claw studied as ligands against the SARS-CoV-2 main protease. (a) Spiroxindole alkaloids; (b) indole glycosides alkaloids; (c) polyhydroxylated triterpenes; (d) quinovic acid glycosides. Pure drugs rarely have the same degree of activity as a The structure of 3CLpro shows the protein to be a totalextract at comparable concentrations or doses of the homodimer with a subunit of 306 amino acids long activecomponent [76]. This phenomenon is attributed to (Figure 2). The catalytic site is mainly constituted by loops theabsence of interacting substances present in the extrac- and hairpins that suggests a flexible site, a feature common t.Furthermore, many plants contain chemical con- in many proteases. The co-crystallized N3 drug stands in the stituentsthat could inhibit drug resistance in viral diseases. catalytic site that defines the subsites of cleavage site S1 and Synthesizeddrugs are often more expensive to produce, is characterized by the following amino acids: THR190, beingunavailable to the poorest populations. In contrast, GLN189, GLU166, HIS164, CYS145, GLY143, GLU166, herbalmedicines used as infusion, decoction, and macera- HIS163, PHE140, ASN142, MET49, HIS41, MET165, tion cansometimes be grown and produced locally, at lower ARG188, and ASP187. cost,taking into account its ethnopharmacological use. This site was defined as the binding pocket for the docking runs. Because of the symmetry in the X-ray structure between the two dimers of 3CLpro, dockings were 3.2. Docking Results. Despite limitations in terms of ener- performed on only one monomer. Molecular docking getic functions (scoring) and conformational sampling studies were performed with Autodock for all compounds. (limited to ligand rotational bound mainly and restricted All compounds show docked structures that fit well into local motion of the protein at the most), protein-ligand the S1 cavity of 3CLpro (Figures 3(a) and 3(b)) and with good dockings are still today the main computational strategy to predicted docking scores that range from −5.9 to −9.2 kcal/ identify potential binders to a given protein target. Protein- mol (Table 1 and Figure S1). In general, calculations revealed ligand dockings of the 26 constituents of Cat’s claw were that most of the 12 amino acids involved in the N3-3CLpro performed against the three-dimensional structure of the interactions identified in the X-ray structures also constitute SARS-Cov-2 main protease 3CLpro (PDB: 6LU7) [37] to the binding pocket of the docking solutions. In particular, study the potential of ethanolic extracts of Cat’s claw for most predicted complexes have the interaction fingerprint COVID-19 treatment based on its majority of components. with CYS145, GLU166, MET165, GLN189, and GLY143, the 6 Evidence-Based Complementary and Alternative Medicine Cleavage site, S1 Subunit B Subunit A N3 (a) N3 inhibitor ∆G = −8.1 kcal.mol−1 (b) (c) Interactions Van der waals Alkyl Carbon hydrogen bond Pi-Alkyl Conventional hydrogen bond Figure 2: (a) Representation of the 3CLpro protease with N3 inhibitor. Overall structure of the dimer (left) and cleavage site S1. (b) The best conformation of the potent 3CLpro inhibitor N3 into 3CLpro (c) 2D ligand-protease interaction plot between the inhibitor N3 with 3CLpro. Dashed lines indicate interactions of N3 with 3CLpro. Figure 3: (a) N3 docked into the 3CLpro cleavage pocket. (b) Superposition of the best conformation of the most active components: speciophylline (yellow), uncarine F (blue), cadambine (red), 3-dihydrocadambine (orange), 3-isodihydrocadambine (black), proantho- cyanidin B2 (purple), epiafzelechin-4β-8-epicatechin (brown), proanthocyanidin B4 (magenta), and N3 (cyan), a potent 3CLpro inhibitor. most important residues in the reported active pocket for Notably, nine components have calculated affinities that SARS-CoV-2 (Figures 3(b) and 3(c)). range from −8.1 to −9.2 kcal·mol−1. In this set of nine Importantly, some compounds have similar or even molecules, we found that most of the proanthocyanidins of lower binding free energy values than those of N3 and other the Cat’s claw extracts show the higher binding affinities to FDA-approved viral protease inhibitors identified as 3CLpro 3CLpro with regard to any other family of compounds. This binders which have energies ranging from −8.0 to −8.5 kcal/ result is consistent with several in vitro and in vivo studies mol (see the references section of Table 1). At this point, it is that show that proanthocyanidins display potent virucidal worth mentioning that the docking calculations involving activity for herpes simplex virus (HSV), human immuno- these drugs give values in very good agreement with ex- deficiency virus (HIV), influenza A and B, hepatitis B and D, perimental ones and around −8.1 kcal/mol, hence providing human Norovirus, and Aichi virus (AiV) [68, 84]. The five a certain amount of confidence regarding the Autodock remaining compounds of high affinity for 3CLpro are from scoring function of this project. the two alkaloidal fractions, the spiroxindole, and the indole Evidence-Based Complementary and Alternative Medicine 7 glycoside ones. These findings are also consistent with openMM software [86] and the Amber force field [55] (for previous reports demonstrating the antiviral properties of details, readers can refer to Section 2). It is important to the alkaloid fractions from Cat’s claw [25]. Amongst those notice that the simulations were carried out on the dimeric compounds, speciophylline and cadambine are particularly (hence complete) structure of the protease-ligand complexes interesting since they are part of the phytochemical fin- after their reconstruction by symmetric operation. For each gerprint for Uncaria tomentosa (Cat’s claw). system, then the stoichiometry of ligand: protein is 2 : 1 (2 From this part of the study, nine components are ligands for one dimer). computationally estimated to have similar or even better We assessed the quality of our MD experiment by using binding affinities for the 3CLpro cleavage site S1 than the four different criteria: (1) root mean square deviation of the current protease inhibitor N3 and other antiviral drugs. backbone (RMSD), (2) all-to-all RMSD, (3) principal Speciophylline, cadambine, and proanthocyanidin B2 were component analysis (PCA), and (4) cluster counting. This selected for further investigation since they are the best combination of analytical tools has previously proven to be a representative of three different families of compounds of reliable assessment methodology of protein structural the Cat’s claw fractions (docking scores of −8.1, −8.6, and convergence [87]. Moreover, ligand and protein flexibility −9.2 kcal/mol, respectively). were assessed by calculation of RMSF analysis. By taking all these parameters into account, all simulations were run until 3.3. Ligand Pathway Analysis. A first series of refinements 250 ns each (see Figure 5 for speciophylline as an illustrative consisted in the simulation of the ligand pathways of spe- case of analysis, and ESI Figures S3 and S4 for cadambine ciophylline, cadambine, and proanthocyanidin B2 to the and proanthocyanidin B2). catalytic site S1 of 3CLpro. Indeed, protein-ligand docking Based on the analysis of these different variables and does not account with the protein motions associated to the visual inspection, a clear behavior appears along the mo- transition of the ligands from the solvent to the binding site, lecular dynamics, which is summarized as follows: a phenomenon that could lead to false positives in screening (1) The overall trajectories of all systems are stable with steps of drug discovery processes and to avoid for further no modification in terms of the secondary structure. stages. To ascertain the accessibility of the compounds to the Only slight transitions on the tertiary structure cleavage site, calculations were performed using the (intra- and intermonomeric interfaces) are observed GPathFinder software [51]. For each compound, 120 though of different magnitudes depending on which channels were calculated (10 runs segmented in finding the the Cat’s claw component is bound and their motion 12 lowest energy paths), and their results were averaged. The during the course of the MD. sampling of the protein structure was based on 110 snap- shots from the 10 μs simulation of the unbound structure (2) For speciophylline, both monomers behave in a very recently released [85], hence allowing that large-scale mo- symmetric manner with the ligands remaining well tions (domain motions) could be accounted in the sited into the cavity during the entire course of the simulation. simulation and at the same location of the initial For the three compounds, the calculations tend to docking solution (Figures 5(b) and 5(c)). However, converge to the same entrance pathway that mainly involve two different orientations are observed. The orien- rearrangements of the loop and hairpin motives that con- tation with higher time of residence is very close to stitute the S1 cleavage site as well as some breathing motions the initial docked complex with most of the inter- involving the alpha helix domain of the dimer (Figure 4). For actions of the cleavage site maintained through all ligands, low energy barriers are observed and none additional hydrogen bonds that appear with Gln189 overcome the 5.3 kcal/mol in average (±3.2 kcal/mol) (see and Gln192 (Figure 5(d), Table S2). The second one Table S1). This value is consistent with the relatively solvent shows speciophylline displaying a slight rotation exposed binding site of the protease and suggests the absence around its main axis of inertia (Figure 5(b)). This of restrictive motions for the binding of the ligands. It was second orientation appears related to the loss of the noticed that the lowest barrier is observed for speciophylline hydrogen bond of the Gln189 as well as a slight with a value of 2.4 kcal/mol. rearrangement of the α domain with respect to the β To this point, our study shows that (1) speciophylline, domain. These changes in the ternary structure are cadambine, and proanthocyanidin B2 (epicatechin-(4β-8)- also related to modifications of the interfaces be- epicatechin) have good predicted binding affinity for S1 tween both monomers. It is to note that these mo- cleavage site and (2) they can naturally access it without tions are consistent with those observed for ligand noticeable energetic cost. binding simulation of the previous GPathFinder calculations (see Section 3.3 and Figure 4). 3.4. Molecular Dynamics (MD) Simulations and Calculations (3) For cadambine and proanthocyanidin B2, the overall of the Free Energy of Binding. With higher confidence on the fold is also very stable. However, a conformational viability of our docking predictions for speciophylline, change occurs at about 200 ns, which essentially cadambine, and proanthocyanidin B2, we further evaluated affects the tertiary and quaternary nature of the the stability of the docked complexes throughout molecular system (Figures S3 and S4). Further inspection shows dynamics simulations. Calculations were performed with that both monomers have also distinct behaviors. 8 Evidence-Based Complementary and Alternative Medicine Monomer B Monomer A Speciophylline Cleavage site β domain S1 α domain Figure 4: Schematic view of the entrance pathway for speciophylline into monomer B of the 3CLpro protease of SARS-CoV-2. The ligand is shown in yellow and the protein in orange (monomer A) and blue (monomer B). Snapshots of the lowest energy pathway are presented with the ligand represented in thin sticks and the final pose in ball and stick. While the ligand remains in monomer B, all along 3.5. Calculation of Drug-Likeness Indices and Scoring. the simulation with poses close to the predicted Calculated human pharmacokinetics profiles play a critical docked complex and strong interaction with the role in assessing the quality of novel antiviral candidates. catalytic binding site, monomer A displaces its ligand Early predictions of pharmacokinetic behavior of the at about 200 ns, which then remains closer to the promising antiviral compounds based on their structure entrance of the monomer-binding site. This dis- could help finding safer and effective leads for preclinical placement toward more solvent exposed location of antiviral testing. Herein, we calculated and analyzed various monomer A is more pronounced for proanthocya- drug-likeness indices for the most qualified Cat’s claw nidin B2 than cadambine (Figures S3 and S4). components (Table 3). Ten pharmacokinetics parameters Therefore, cadambine and proanthocyanidin B2 were calculated as a drug-likeness filter for speciophylline, seem to lead to one very stable binding with asso- cadambine, and proanthocyanidin B2 and compared to N3 ciated conformational changes that relate, in an and three selected antiviral drugs. Results obtained dem- apparent allosteric manner, on the second one. onstrate the feasibility of the selected components from Cat’s claw exhibiting suitable drug-like characteristics. These structural observations were further quantified by MM-GBSA analysis (Table 2). Results show that ΔGbinding of In addition, lipophilicity (calculated as log Po/w) is the the species in the different monomers range from ca. key physicochemical property for rational drug design. This −52 kcal/mol to ca. −15 kcal/mol. Taking average values parameter provides valuable information about transport between S1A and S1B sites, the affinity of the different through lipid bilayers. Compounds that display high log phytochemical compounds to 3CLpro can be classified as Po/w tend to have good permeability across the cell wall cadambine (approx. −40 kcal/mol) > speciophylline (approx. [66, 67]. In this study, the selected compounds exhibited −35 kcal/mol) > proanthocyanidin (approx. −20 kcal/mol). optimal log Po/w values < 6.5; notably speciophylline has a These values suggest the two former molecules to be good log Po/w of 1.709 (optimal value compared to 95% of binders to 3CLpro and the latest to be somehow of lesser current drugs) implying good permeability across the cell quality. When analyzing each monomer individually, the membrane of infected cells. Furthermore, the in silico ΔGbinding values are consistent with the structural obser- passive transmembrane permeation was calculated for all vation with speciophylline having almost symmetric values compounds using Caco-2 cell monolayers or MDCK cells (ca. −33 kcal/mol) and cadambine and proanthocyanidin as a model. Cadambine and proanthocyanidin B2 exhibit having different magnitudes. The closer energy values for low values of permeability (<27 nm/s), clearly suggesting cadambine and speciophylline are consistent with the poor bioavailability [88]. However, an optimal value for the structural similarities between both alkaloids with respect to alkaloid speciophylline was predicted (307 nm/s for Caco-2 proanthocyanidin. model) close to ritonavir and lopinavir drugs [89–91] and In the light of the molecular dynamics analysis and the was much better than the inhibitor N3 (6 nm/s for Caco-2 calculation of the free energy of binding, at least cadambine model). and speciophylline are predicted to present very good in- Finally, binding to serum albumin (expressed as log hibition to the SARS-CoV-2 main protease. Because these KHSA) is the most important parameter for distribution and components are found in the ethanolic extract of Cat’s claw, transport of antiviral drugs in the systemic circulation. Early it may position itself as possible therapeutic herbal for prediction of this parameter reduces the amount of wasted COVID-19. time and resources for drug development candidates in the Evidence-Based Complementary and Alternative Medicine 9 4.0 250 4.0 250 30 24 3.5 3.5 225 21 200 20 200 3.0 3.0 18 175 2.5 2.5 10 RMSD (A) RMSD (A) Time (ns) Time (ns) 150 150 15 # clusters Dimer PC2 2.0 2.0 0 125 12 1.5 100 1.5 100 –10 9 75 1.0 1.0 –20 6 50 50 0.5 0.5 25 3 –30 0.0 0 0.0 0 0 0 50 100 150 200 250 0 50 100 150 200 250 –30 –20 –10 0 10 20 30 0 50 100 150 200 250 Time (ns) Time (ns) PC1 Time (ns) (a) 14 Monomer A Monomer B 12 10 Distance (A) 8 6 4 2 0 0 50 100 150 200 250 Time (ns) (b) (c) (d) 4.0 250 4.0 250 30 24 3.5 3.5 225 21 200 20 200 3.0 3.0 Monomer A 175 18 2.5 10 RMSD (A) RMSD (A) 2.5 Time (ns) Time (ns) 150 15 # clusters 150 PC2 2.0 2.0 0 125 12 1.5 100 1.5 100 –10 9 75 1.0 1.0 6 50 –20 50 0.5 0.5 25 3 –30 0.0 0 0.0 0 0 0 50 100 150 200 250 0 50 100 150 200 250 –30 –20 –10 0 10 20 30 0 50 100 150 200 250 Time (ns) Time (ns) PC1 Time (ns) (e) 4.0 250 4.0 250 30 24 3.5 3.5 225 21 200 20 200 3.0 3.0 18 175 Monomer B 2.5 2.5 10 RMSD (A) RMSD (A) Time (ns) Time (ns) 150 150 15 # clusters PC2 2.0 2.0 0 125 12 1.5 100 1.5 100 –10 9 75 1.0 1.0 –20 6 50 50 0.5 0.5 25 3 –30 0.0 0 0.0 0 0 0 50 100 150 200 250 0 50 100 150 200 250 –30 –20 –10 0 10 20 30 0 50 100 150 200 250 Time (ns) Time (ns) PC1 Time (ns) (f ) Figure 5: Complete analysis of the 250 ns trajectory of 3CLpro of SARS-CoV-2 bound with speciophylline. Monomer A is shown in orange and monomer B in blue. (a), (e), and (f ) shows, from left to right, RMSD, all-to-all RMSD, PCA, and cluster counting (cutoff 1.5 Å). (b) shows the relative position of speciophylline molecules versus the center of mass of the S1 cavity along the simulation (black dashed line indicates the distance of the N3 inhibitor as reference). In (c), size of the ribbon is proportional to the flexibility of the protein backbone and the size on the ligand center is proportional to the amount of structural deviation of the ligand during the 250 ns. (d) reproduces the binding site S1 with the most important residues involved in the interactions. antiviral therapy and management. Behavior of the selected compared to the reference value taken from 95% of currently compounds with human plasma protein (log KHSA) is within known drugs (log KHSA from −1.5 to 2.0). Optimal physi- recommended values (ranging from −0.30 to −0.044) cochemical properties obtained for the active components of 10 Evidence-Based Complementary and Alternative Medicine Table 2: Ligand-protease ΔGbinding in kcal/mol. Cadambine Proanthocyanidin B2 Speciophylline Monomer B −51.92 ± 6.03 −21.20 ± 6.22 −33.49 ± 4.88 Monomer A −32.25 ± 5.84 −15.25 ± 7.31 −33.34 ± 6.15 The average obtained from 250 frames for each MD trajectory (one per nanosecond) using the MM-GBSA method in AmberTools18. The following parameters were used: igb � 2, saltcon � 0.100. Table 3: Calculated drug-likeness properties of the most qualified Cat’s claw components. n-Rot App. n-ON Log Log Caco-2g Lipinski rule of Compound MWa PSAb Bond n-OHNHd MDCK % HIAi (<10)c Po/we KHSAf (nm/s) h five (≤1) (0-10) (nm/s) Speciophylline 368.432 82.804 1 8 1 1.709 −0.044 307 153 81 0 Cadambine 544.557 158.806 8 11 5 0.037 −0.592 27 11 27 0 Proanthocyanidin B2 578.528 209.177 10 12 10 0.505 −0.300 1 1 <25 1 N3 680.800 221.219 17 14 3 2.578 −0.497 6 11 85 2 Remdesivirj 602.583 196.086 16 16 5 1.135 −0.685 37 14 36 2 Ritonavirj 720.943 139.542 18 11 3 6.335 0.638 647 1014 75 2 Lopinavirj 628.810 124.690 16 9 4 5.751 0.554 510 598 83 2 a Molecular weight of the hybrid (150–500). bPolar surface area (PSA) (7.0–200 Å2). cn-ON number of hydrogen bond acceptors <10. dn-OHNH number of hydrogen bond donors ≤5. eOctanol water partition coefficient (log Po/w) (–2.0 to 6.5). fBinding-serum albumin (KHSA) (−1.5 to 1.5). gHuman intestinal permeation (<25 poor, >500 great). hMadin-Darby canine kidney (MDCK) cell permeation. iHuman intestinal absorption (% HIA) (>80% is high, <25% is poor). jFDA-approved antiviral drugs used as references. Cat’s claw indicated that this herb should be considered for good predicted inhibitory profiles. To anticipate the thera- the rapid progression in the antiviral treatment. peutic behavior ofCat’s claw components, several physico- chemical and ADME-score indices were calculated using themore activemolecules and compared to anti- 4. Conclusions viralmarketed drugs. The returned values show optimal The coronavirus pandemic is a serious public health crisis drug-like properties for speciophylline, cadambine, and due to high mortality, high basic reproduction numbers, and proanthocyanidin B2. neither approved drugs nor vaccines. The recent publication Due to the remarkable presence of these compounds in of the crystal structure of the SARS-CoV-2 main protease the Cat’s claw extracts, we believe that this in silico study at has provided the community with critical structural infor- least points at Uncaria tomentosa as a whole as an interesting mation. Potential inhibitors of this enzyme could have a herb opening novel therapeutically horizons for COVID-19 major contribution in the reduction, prevention, or eradi- treatment. Based on our findings, we believe that Cat’s claw cation of the viral load of patients. should be taken into consideration in looking for COVID-19 This study aimed at computationally explore if Uncaria treatments. tomentosa (Cat’s claw), an indigenous medicinal herb known for its antiviral properties against other high mor- Abbreviations tality viruses, contains phytochemicals potentially able to inhibit the SARS-CoV-2 main protease, 3CLpro. PCA: Principal component analysis After screening 26 key compounds of Uncaria tomentosa ADME: Absorption, distribution, metabolism, and against the 3CLpro cleavage site, nine compounds displayed excretion lower binding energies than those of known inhibitors of MD: Molecular dynamics this enzyme (N3 as well as three FDA-approved antiviral FDA: Food and Drug Administration drugs). Amongst those nine compounds, one of each family TNFα: Tumor necrosis factor alpha of well-characterized active phytochemical fractions of Cat’s EsSalud: Seguro Social de Salud (in Spanish) claw, namely, speciophylline, cadambine, and proantho- PDB: Protein Data Bank. cyanidin B2, attracted our attention because of their strong docking scores (−8.1, −8.6, and −9.2 kcal/mol, respectively). The potential inhibitory effects of those molecules were Data Availability further analyzed by means of ligand pathway simulations All data used to support the findings of this study can be (which show very low barriers for binding), molecular dy- made available from the corresponding author upon request. namics of the docked complexes (until convergence at aprox. 250 ns), and MM-GBSA free energy binding calculations (with values ranging from ca. −50 kcal/mol to ca. −15 kcal/ Conflicts of Interest mol). Altogether, these results confirm that the three compounds, and more particularly the alkaloid ones, have The authors declare no conflicts of interest. Evidence-Based Complementary and Alternative Medicine 11 Authors’ Contributions ligand during the 250 ns. Panel D reproduces the binding site S1 with the most important residues involved in the in- O.H-C. and A.F.Y-P. conceptualized the study; A.F.Y-P. and teractions. Table S1: GPathFinder vina barriers for the ligand J.-D.M. developed the methodology; A.F.Y-P., J.-E.S-A., and route. 120 solutions were obtained for each ligand. Average L.T-S were responsible for software; W.C-G. and J-D.M. and standard deviation are reported. Table S2: amino acids validated the results; J.-D.M. performed the formal analysis; of the 3CLpro protease that participate in hydrogen bonding O.H.-C. and A.F.Y.-P. were involved in investigation; W.C.-G. interactions with cadambine, proanthocyanidin B2, and was responsible for data curation; O.H.-C., A.F.Y.-P., J.D.M., speciophylline. Input file (.yaml format) used in Gpath- and J.-E.S.-A. were involved in writing and editing and original Finder calculations. (Supplementary Materials) draft preparation; A.F.Y.-P., L.T.-S., and J.D.-M. were involved in review and editing of the manuscript; W.C.-G. was involved in visualization; A.F.Y.-P. supervised the study. All authors References have read and agreed to the published version of the [1] T. Singhal, “A review of coronavirus disease-2019 (COVID- manuscript. 19),” Indian Journal of Pediatrics, vol. 87, no. 4, pp. 281–286, 2020. Acknowledgments [2] Y. W. Chen, C. P. B. Yiu, and K. Y. Wong, “Prediction of the SARS-CoV-2 (2019-nCoV) 3C-like protease (3CLpro) J.-E.S.-A., L.T.-S., and J.-D.M. gratefully acknowledge the structure: virtual screening reveals velpatasvir, ledipasvir, and financial support from the Spanish MINECO (CTQ2017- other drug repurposing candidates,” F1000 Research, vol. 9, 87889-P). The authors would like to thank Universidad de p. 129, 2020. Antioquia, Universidad Nacional Mayor de San Marcos, and [3] C. Wang, P. W. Horby, F. G. Hayden, and G. F. Gao, “A novel Universitat Autònoma de Barcelona. coronavirus outbreak of global health concern,” The Lancet, vol. 395, no. 10223, pp. 470–473, 2020. [4] J. Williamson, R. Ramirez, and T. Wingfield, “Health, Supplementary Materials healthcare access, and use of traditional versus modern medicine in remote Peruvian Amazon communities: a de- Figure S1: histogram showing molecular docking results scriptive study of knowledge, attitudes, and practices,” The between the novelSARS-CoV-2 main protease and nine American Journal of Tropical Medicine and Hygiene, vol. 92, components from Cat’s claw which showed the highest no. 4, pp. 857–864, 2015. binding affinity. Figure S2: (A) the best conformation of the [5] R. W. Bussmann, G. Malca-Garcı́a, A. Glenn et al., “Minimum nine active components inside the catalytic pocket of the inhibitory concentrations of medicinal plants used in SARS-CoV-2 main protease 3CLpro (PDB: 6LU7). 2D in- Northern Peru as antibacterial remedies,” Journal of Ethno- teraction mode plots of the selected compounds with 3CLpro. pharmacology, vol. 132, no. 1, pp. 101–108, 2010. Interactions between each component and amino acid [6] S. Ben-Shabat, L. Yarmolinsky, D. Porat, and A. Dahan, residues inside the 3CLpro cleavage domain are indicated by “Antiviral effect of phytochemicals from medicinal plants: the dashed lines. Figure S3: complete analysis of the 250 ns applications and drug delivery strategies,” Drug Delivery and Translational Research, vol. 10, no. 2, pp. 354–367, 2020. trajectory of 3CLpro ofSARS-CoV-2bound with cadambine. [7] O. Herrera-Calderon, J. L. Arroyo-Acevedo, J. Rojas-Armas Monomer A is shown in orange and monomer B in blue. et al., “Phytochemical screening, total phenolic content, an- Panels A, E, and F show, from left to right, RMSD, all-to-all tioxidant and cytotoxic activity of chromolaena laevigata on RMSD, PCA, and cluster counting (cutoff 1.5 Å). Panel B human tumor cell lines,” Annual Research & Review in Bi- shows the relative position of cadambine molecules versus ology, vol. 21, no. 1, pp. 1–9, 2017. the center of mass of the S1 cavity along the simulation [8] M. E. Heitzman, C. C. Neto, E. Winiarz, A. J. Vaisberg, and (black dashed line indicates the distance of the N3 inhibitor G. B. Hammond, “Ethnobotany, phytochemistry and phar- as reference). For panel C, size of the ribbon is proportional macology of uncaria (Rubiaceae),” ChemInform, vol. 36, to the flexibility of the protein backbone and the size on the no. 17, 2005. ligand center is proportional to the amount of structural [9] K. Keplinger, G. Laus, M. Wurm, M. P. Dierich, and deviation of the ligand during the 250 ns. Panel D repro- H. Teppner, “Uncaria tomentosa (Willd.) duces the binding site S1 with the most important residues DC.—ethnomedicinal use and new pharmacological, toxi- involved in the interactions. Figure S4: complete analysis of cological and botanical results,” Journal of Ethno- the 250 ns trajectory of 3CLpro of SARS-CoV-2bound with pharmacology, vol. 64, no. 1, pp. 23–34, 1998. proanthocyanidin B2. Monomer A is shown in orange and [10] G. F. Gonzales, J. Aguilar, and M. Villar, “The world summit of harmonization on traditional, alternative and comple- monomer B in blue. Panels A, E, and F shows, from left to mentary medicine (TACM) in Lima, Peru,” Evidence-based right, RMSD, all-to-all RMSD, PCA, and cluster counting Complementary and Alternative Medicine, vol. 7, no. 2, (cutoff 1.5 Å). Panel B shows the relative position of pp. 271–275, 2010. proanthocyanidin B2 molecules versus the center of mass of [11] A. Kośmider, E. Czepielewska, M. Kuraś et al., “Uncaria the S1 cavity along the simulation (black dashed line in- tomentosa leaves decoction modulates differently ROS pro- dicates the distance of the N3 inhibitor as reference). For duction in cancer and normal cells, and effects cisplatin cy- panel C, size of the ribbon is proportional to the flexibility of totoxicity,” Molecules, vol. 22, no. 4, p. 620, 2017. the protein backbone and the size on the ligand center is [12] S. Zhu, Q. Li, S. Chen et al., “Phylogenetic analysis of Uncaria proportional to the amount of structural deviation of the species based on internal transcribed spacer (ITS) region and 12 Evidence-Based Complementary and Alternative Medicine ITS2 secondary structure,” Pharmaceutical Biology, vol. 56, [28] L. Allen-Hall, J. T. Arnason, P. Cano, and R. M. Lafrenie, no. 1, pp. 548–558, 2018. “Uncaria tomentosa acts as a potent TNF-α inhibitor through [13] A. D. Snow, G. M. Castillo, B. P. Nguyen et al., “The Amazon NF-κB,” Journal of Ethnopharmacology, vol. 127, no. 3, rain forest plant Uncaria tomentosa (cat’s claw) and its specific pp. 685–693, 2010. proanthocyanidin constituents are potent inhibitors and re- [29] J. E. Williams, “Review of antiviral and immunomodulating ducers of both brain plaques and tangles,” Scientific Reports, properties of plants of the Peruvian rainforest with a par- vol. 9, no. 1, 2019. ticular emphasis on uña de gato and sangre de grado,” Al- [14] G. Laus, D. Brössner, and K. Keplinger, “Alkaloids of Peruvian ternative Medicine Review, vol. 6, no. 6, pp. 567–579, 2001. Uncaria tomentosa,” Phytochemistry, vol. 45, no. 4, [30] M. Sandoval, R. M. Charbonnet, N. N. Okuhama et al., “Cat’s pp. 855–860, 1997. claw inhibits TNFα production and scavenges free radicals: [15] I. Vera-Reyes, A. A. Huerta-Heredia, T. Ponce-Noyola et al., role in cytoprotection,” Free Radical Biology and Medicine, “Monoterpenoid indole alkaloids and phenols are required vol. 29, no. 1, pp. 71–78, 2000. antioxidants in glutathione depleted Uncaria tomentosa root [31] R. M. López Galera, E. Ribera Pascuet, J. I. Esteban Mur, cultures,” Frontiers in Environmental Science, vol. 3, 2015. J. B. Montoro Ronsano, and J. C. Juárez Giménez, “Interaction [16] P. Montoro, V. Carbone, J. de Dioz Zuniga Quiroz, between cat’s claw and protease inhibitors atazanavir, rito- F. De Simone, and C. Pizza, “Identification and quantification navir and saquinavir,” European Journal of Clinical Phar- of components in extracts of Uncaria tomentosa by HPLC-ES/ macology, vol. 64, no. 12, pp. 1235-1236, 2008. MS,” Phytochemical Analysis, vol. 15, no. 1, pp. 55–64, 2004. [17] R. Aquino, N. De Tommasi, F. De Simone, and C. Pizza, ® [32] C. Åkesson, R. W. Pero, and F. Ivars, “C-Med 100 , a hot water extract of Uncaria tomentosa, prolongs lymphocyte “Triterpenes and quinovic acid glycosides from Uncaria survival in vivo,” Phytomedicine, vol. 10, no. 1, pp. 23–33, tomentosa,” Phytochemistry, vol. 45, no. 5, pp. 1035–1040, 2003. 1997. [33] S. L. Chen, H. Yu, H. M. Luo et al., “Conservation and [18] R. Cerri, R. Aquino, F. De Simone, and C. Pizza, “New sustainable use of medicinal plants: problems, progress, and quinovic acid glycosides from Uncaria tomentosa,” Journal of prospects,” Chinese Medicine (United Kingdom), vol. 11, no. 1, Natural Products, vol. 51, no. 2, pp. 257–261, 1988. 2016. [19] E. M. C. Peñaloza, S. Kaiser, P. E. De Resende et al., “Chemical [34] M. Prajapat, P. Sarma, N. Shekhar et al., “Drug targets for composition variability in the Uncaria tomentosa (cat’s claw) corona virus: a systematic review,” Scholars International wild population,” Quı́mica Nova, vol. 38, no. 3, 2015. Journal of Biochemistry, vol. 3, no. 6, pp. 143–153, 2020. [20] M. N. Hoyos, F. Sánchez-Patán, R. M. Masis et al., “Phenolic [35] C. Wu, Y. Liu, Y. Yang et al., “Analysis of therapeutic targets assessment of Uncaria tomentosa L. (cat’s claw): leaves, stem, for SARS-CoV-2 and discovery of potential drugs by com- bark and wood extracts,” Molecules, vol. 20, no. 12, putational methods,” Acta pharmaceutica Sinica. B.vol. 10, pp. 22703–22717, 2015. no. 5, pp. 766–788, 2020. [21] M. Kitajima, K. I. Hashimoto, M. Yokoya, H. Takayama, and [36] A. D. Elmezayen, A. Al-Obaidi, A. T. Şahin, and K. Yelekçi, N. Aimi, “Two new 19-hydroxyursolic acid-type triterpenes “Drug repurposing for coronavirus (COVID-19): in silico from Peruvian ‘Una de Gato’ (Uncaria tomentosa),” Tetra- hedron, vol. 56, no. 4, pp. 547–552, 2000. screening of known drugs against coronavirus 3CL hydrolase [22] A. M. P. Yépez, O. L. de Ugaz, C. M. P. Alvarez et al., and protease enzymes,” Journal of Biomolecular Structure and “Quinovic acid glycosides from Uncaria guianensis,” Phyto- Dynamics, vol. 1-13, 2020. chemistry, vol. 30, no. 5, pp. 1635–1637, 1991. [37] K. A. Peele, C. Potla Durthi, T. Srihansa et al., “Molecular [23] R. Aquino, F. De Simone, C. Pizza, C. Conti, and M. L. Stein, docking and dynamic simulations for antiviral compounds “Plant metabolites. Structure and in vitro antiviral activity of against SARS-CoV-2: a computational study,” Informatics in quinovic acid glycosides from Uncaria tomentosa and Medicine Unlocked, vol. 19, Article ID 100345, 2020. Guettarda platypoda,” Journal of Natural Products, vol. 52, [38] G. M. Morris, D. S. Goodsell, R. S. Halliday et al., “Automated no. 4, pp. 679–685, 1989. docking using a Lamarckian genetic algorithm and an em- [24] J. Yunis Aguinaga, G. S. Claudiano, P. F. Marcusso et al., pirical binding free energy function,” Journal of Computa- “Acute toxicity and determination of the active constituents of tional Chemistry, vol. 19, no. 14, pp. 1639–1662, 1998. aqueous extract of Uncaria tomentosa bark in Hyphessobrycon [39] Q. Wang, Y. Zhao, X. Chen, and A. Hong, “Virtual screening eques,” Journal of Toxicology, vol. 2014, Article ID 412437, of approved clinic drugs with main protease (3CLpro) reveals 5 pages, 2014. potential inhibitory effects on SARS-CoV-2,” Journal of [25] S. R. I. N. Reis, L. M. M. Valente, A. L. Sampaio et al., Biomolecular Structure and Dynamics, pp. 1–11, 2020. “Immunomodulating and antiviral activities of Uncaria [40] L. Zhang, D. Lin, X. Sun et al., “Crystal structure of SARS- tomentosa on human monocytes infected with dengue virus- CoV-2 main protease provides a basis for design of improved 2,” International Immunopharmacology, vol. 8, no. 3, α-ketoamide inhibitors,” Science, vol. 368, no. 6489, pp. 468–476, 2008. pp. 409–412, 2020. [26] R. S. Lima-Junior, C. Da Silva Mello, C. F. Kubelka, [41] Z. Jin, X. Du, Y. Xu et al., “Structure of Mpro from SARS- A. C. Siani, and L. M. M. Valente, “Uncaria tomentosa al- CoV-2 and discovery of its inhibitors,” Nature, vol. 582, kaloidal fraction reduces paracellular permeability, IL-8 and no. 7811, pp. 289–293, 2020. NS1 production on human microvascular endothelial cells [42] N. Fintelman-Rodrigues, C. Q. Sacramento, C. R. Lima et al., infected with dengue virus,” Natural Product Communica- “Atazanavir inhibits SARS-CoV-2 replication and pro-in- tions, vol. 8, no. 11, Article ID 1934578X1300801, 2013. flammatory cytokine production,” 2020. [27] T. Caon, S. Kaiser, C. Feltrin et al., “Antimutagenic and [43] D. L. McKee, A. Sternberg, U. Stange, S. Laufer, and antiherpetic activities of different preparations from Uncaria C. Naujokat, “Candidate drugs against SARS-CoV-2 and tomentosa (cat’s claw),” Food and Chemical Toxicology, COVID-19,” Pharmacological Research, vol. 157, Article ID vol. 66, pp. 30–35, 2014. 104859, 2020. Evidence-Based Complementary and Alternative Medicine 13 [44] M. A. Martinez, “Compounds with therapeutic potential screening,” Journal of Pharmaceutical Sciences, vol. 88, no. 1, against novel respiratory 2019 coronavirus,” Antimicrobial pp. 28–33, 1999. Agents and Chemotherapy, vol. 64, no. 5, Article ID e00399, [62] Calculation of Molecular Properties and Bioactivity Score 2020. Software, Molinspiration Cheminformatics, 2020, http:// [45] A. T. Ton, F. Gentile, M. Hsing, F. Ban, and A. Cherkasov, www.molinspiration.com/cgi-bin/properties. “Rapid identification of potential inhibitors of SARS-CoV-2 [63] I. V. Tetko and G. I. Poda, “Application of ALOGPS 2.1 to main protease by deep docking of 1.3 billion compounds,” predict log D distribution coefficient for Pfizer proprietary Molecular Informatics, vol. 39, no. 8, Article ID e2000028, compounds,” Journal of Medicinal Chemistry, vol. 47, no. 23, 2020. pp. 5601–5604, 2004. [46] S. Jo, S. Kim, D. H. Shin, and M. S. Kim, “Inhibition of SARS- [64] F. Broccatelli, L. Salphati, E. Plise et al., “Predicting passive CoV 3CL protease by flavonoids,” Journal of Enzyme Inhi- permeability of drug-like molecules from chemical structure: bition and Medicinal Chemistry, vol. 35, no. 1, pp. 145–151, where are we?” Molecular Pharmaceutics, vol. 13, no. 12, 2020. pp. 4199–4208, 2016. [47] Z. Haider, M. M. Subhani, M. A. Farooq et al., “In silico [65] H. Yang, C. Lou, L. Sun et al., “AdmetSAR 2.0: web-service for discovery of novel inhibitors against main protease (Mpro) of prediction and optimization of chemical ADMET properties,” SARS-CoV-2 using pharmacophore and molecular docking Bioinformatics, vol. 35, no. 6, pp. 1067–1069, 2019. based virtual screening from ZINC database,” 2020. [66] I. J. Flores-Sánchez, J. Ortega-López, M. D. C. Montes- [48] ChemAxon, ChemAxon—Software Solutions and Services for Horcasitas, and A. C. Ramos-Valdivia, “Biosynthesis of sterols Chemistry and Biology (Version 16.10.31), ChemAxon, and triterpenes in cell suspension cultures of Uncaria Budapest, Hungary, 2016. tomentosa,” Plant and Cell Physiology, vol. 43, no. 12, [49] O. Trott and A. J. Olson, “AutoDock Vina: improving the pp. 1502–1509, 2002. speed and accuracy of docking with a new scoring function, [67] M. Navarro, W. Zamora, S. Quesada et al., “Fractioning of efficient optimization, and multithreading,” Journal of proanthocyanidins of Uncaria tomentosa. Composition and Computational Chemistry, vol. 31, no. 2, pp. 455–461, 2010. structure-bioactivity relationship,” Antioxidants, vol. 6, no. 3, [50] Schrödinger, “PyMOL, molecular visualization system,” 2018, p. 60, 2017. https://pymol.org. [68] M. Navarro, E. Arnaez, I. Moreira et al., “Polyphenolic [51] J. E. Sánchez-Aparicio, G. Sciortino, D. V. Herrmannsdoerfer composition and antioxidant activity of Uncaria tomentosa et al., “Gpathfinder: identification of ligand-binding pathways commercial bark products,” Antioxidants, vol. 8, no. 9, p. 339, by a multi-objective genetic algorithm,” International Journal 2019. of Molecular Sciences, vol. 20, no. 13, p. 3155, 2019. [69] M. D. C. Santos Araújo, I. L. Farias, J. Gutierres et al., “Uncaria [52] R. González-Alemán, D. Hernández-Castillo, J. Caballero, tomentosa—adjuvant treatment for breast cancer: clinical and L. A. Montero-Cabrera, “Quality threshold clustering of trial,” Evidence-Based Complementary and Alternative Med- molecular dynamics: a word of caution,” Journal of Chemical icine, vol. 2012, Article ID 676984, 8 pages, 2012. Information and Modeling, vol. 60, no. 2, pp. 467–472, 2020. [70] L. C. L. De Paula, F. Fonseca, F. Perazzo et al., “Uncaria [53] H. M. Berman, J. Westbrook, Z. Feng et al., “The Protein Data tomentosa (Cat’s claw) improves quality of life in patients with Bank (www.rcsb.org),” Nucleic Acids Research, vol. 28, no. 1, advanced solid tumors,” The Journal of Alternative and pp. 235–242, 2000. Complementary Medicine, vol. 21, no. 1, pp. 22–30, 2015. [54] E. F. Pettersen, T. D. Goddard, C. C. Huang et al., “UCSF [71] C. Gonçalves, T. Dinis, and M. T. Batista, “Antioxidant Chimera—a visualization system for exploratory research and properties of proanthocyanidins of Uncaria tomentosa bark analysis,” Journal of Computational Chemistry, vol. 25, no. 13, decoction: a mechanism for anti-inflammatory activity,” pp. 1605–1612, 2004. Phytochemistry, vol. 66, no. 1, pp. 89–98, 2005. [55] L. F. Song, T. S. Lee, C. Zhu, D. M. York, and K. M. Merz, [72] O. Lock, E. Perez, M. Villar, D. Flores, and R. Rojas, “Bioactive “Using AMBER18 for relative free energy calculations,” compounds from plants used in Peruvian traditional medi- Journal of Chemical Information and Modeling, vol. 59, no. 7, cine,” Natural Product Communications, vol. 11, no. 3, pp. 3128–3135, 2019. pp. 315–337, 2016. [56] M. J. Frisch, G. W. Trucks, H. B. Schlegel et al., Gaussian 09, [73] R. T. Brown, D. M. Duckworth, and C. A. M. Santos, “Bio- Revision B.01, Gaussian, Inc., Wallingford, CT, USA, 2009. genetically patterned synthesis of cadambine,” Tetrahedron [57] C. I. Bayly, P. Cieplak, W. D. Cornell, and P. A. Kollman, “A Letters, vol. 32, no. 17, pp. 1987–1990, 1991. well-behaved electrostatic potential based method using [74] D. Pierrot, V. Sinou, S. S. Bun et al., “Design and synthesis of charge restraints for deriving atomic charges: the RESP simplified speciophylline analogues and β-carbolines as active model,” The Journal of Physical Chemistry, vol. 97, no. 40, molecules against Plasmodium falciparum,” Drug Develop- pp. 10269–10280, 1993. ment Research, vol. 80, no. 1, pp. 133–137, 2019. [58] D. A. Case, I. Y. Ben-Shalom, S. R. Brozell et al., “Amber [75] W. Fujii, K. Toda, K. Matsumoto et al., “Syntheses of pro- 2018,” 2018. delphinidin B1, B2, and B4 and their antitumor activities [59] P. Eastman, J. Swails, J. D. Chodera et al., “OpenMM 7: rapid against human PC-3 prostate cancer cell lines,” Tetrahedron development of high performance algorithms for molecular Letters, vol. 54, no. 52, pp. 7188–7192, 2013. dynamics,” PLoS Computational Biology, vol. 13, no. 7, Article [76] P. Rasoanaivo, C. W. Wright, M. L. Willcox, and B. Gilbert, ID e1005659, 2017. “Whole plant extracts versus single compounds for the [60] J. Rodrı́guez-Guerra Pedregal, L. Alonso-Cotchico, treatment of malaria: synergy and positive interactions,” L. Velasco-Carneros, and J.-D. Maréchal, “OMMProtocol: a Malaria Journal, vol. 10, no. S1, 2011. command line application to launch molecular dynamics [77] S. Khaerunnisa, H. Kurniawan, R. Awaluddin, and simulations with OpenMM,” 2018. S. Suhartati, “Potential inhibitor of COVID-19 main protease [61] J. D. Irvine, L. Takahashi, K. Lockhart et al., “MDCK (Madin- (Mpro) from several medicinal plant compounds by molecular Darby canine kidney) cells: a tool for membrane permeability docking study,” 2020. 14 Evidence-Based Complementary and Alternative Medicine [78] C. Navan, “Possible drug candidates for COVID-19,” 2020. [79] K. Gao, D. D. Nguyen, R. Wang, and G.-W. Wei, “Machine intelligence design of 2019-nCoV drugs,” 2020. [80] L. Mittal, A. Kumari, M. Srivastava, M. Singh, and S. Asthana, “Identification of potential molecules against COVID-19 main protease through structure-guided virtual screening approach,” Journal of Biomolecular Structure and Dynamics, pp. 1–19, 2020. [81] Y. Kumar, H. Singh, and C. N. Patel, “In silico prediction of potential inhibitors for the main protease of SARS-CoV-2 using molecular docking and dynamics simulation based drug-repurposing,” Journal of Infection and Public Health, vol. 13, no. 9, pp. 1210–1223, 2020. [82] X. Wang, R. S. Dhindsa, G. Povysil et al., “Transcriptional inhibition of host viral entry proteins as a therapeutic strategy for SARS-CoV-2,” 2020. [83] H. Amawi, G. I. Abu Deiab, A. A. A. Aljabali, K. Dua, and M. M. Tambuwala, “COVID-19 pandemic: an overview of epidemiology, pathogenesis, diagnostics and potential vac- cines and therapeutics,” Therapeutic Delivery, vol. 11, no. 4, pp. 245–268, 2020. [84] S. M. Lipson, G. Karalis, L. Karthikeyan et al., “Mechanism of anti-rotavirus synergistic activity by epigallocatechin gallate and a proanthocyanidin-containing nutraceutical,” Food and Environmental Virology, vol. 9, no. 4, pp. 434–443, 2017. [85] C. S. Lupala, X. Li, J. Lei et al., “Computational simulations reveal the binding dynamics between human ACE2 and the receptor binding domain of SARS-CoV-2 spike protein,” 2020. [86] P. Eastman and V. S. Pande, “OpenMM: a hardware-inde- pendent framework for molecular simulations,” Computing in Science & Engineering, vol. 12, no. 4, pp. 34–39, 2010. [87] L. Alonso-Cotchico, J. R. G. Pedregal, A. Lledós, and J. D. Maréchal, “The effect of cofactor binding on the con- formational plasticity of the biological receptors in Artificial Metalloenzymes: the case study of LmrR,” Frontiers in Chemistry, vol. 7, 2019. [88] C. A. Lipinski, F. Lombardo, B. W. Dominy, and P. J. Feeney, “Experimental and computational approaches to estimate solubility and permeability in drug discovery and develop- ment settings,” Advanced Drug Delivery Reviews, vol. 64, pp. 4–17, 2012. [89] J. T. Ortega, M. L. Serrano, F. H. Pujol, and H. R. Rangel, “Unrevealing sequence and structural features of novel coronavirus using in silico approaches: the main protease as molecular target,” EXCLI Journal, vol. 19, 2020. [90] G. A. Gyebi, O. B. Ogunro, A. P. Adegunloye, O. M. Ogunyemi, and S. O. Afolabi, “Potential inhibitors of coronavirus 3-chymotrypsin-like protease (3CLpro): an in silico screening of alkaloids and terpenoids from African medicinal plants,” Journal of Biomolecular Structure and Dynamics, pp. 1–13, 2020. [91] C. J. Dennis, L. Bruno, D. Rodrigo et al., “Antiretroviral drug activity and potential for pre-exposure prophylaxis against COVID-19 and HIV infection,” 2020.