Lipophilicity, Pharmacokinetic Properties, and Molecular Docking Study on SARS-Cov-2 Target for Betulin Triazole Derivatives with Attached 1,4-Quinone

pharmaceutics

Article Lipophilicity, Pharmacokinetic Properties, and Molecular Docking Study on SARS-CoV-2 Target for Betulin Triazole Derivatives with Attached 1,4-Quinone

Monika Kadela-Tomanek 1,* , Maria Jastrz˛ebska 2, Krzysztof Marciniec 1 , Elwira Chrobak 1 , Ewa B˛ebenek 1 and Stanisław Boryczka 1

1 Department of Organic Chemistry, Faculty of Pharmaceutical Sciences in Sosnowiec, Medical University of Silesia, Katowice, 4 Jagiello´nskaStr., 41-200 Sosnowiec, Poland; [email protected] (K.M.); [email protected] (E.C.); [email protected] (E.B.); [email protected] (S.B.) 2 Silesian Center for Education and Interdisciplinary Research, Institute of Physics, University of Silesia, 75 Pułku Piechoty 1a, 41-500 Chorzów, Poland; [email protected] * Correspondence: [email protected]; Tel.: +48-32-3641666

Abstract: A key parameter in the design of new active compounds is lipophilicity, which inﬂuences the solubility and permeability through membranes. Lipophilicity affects the pharmacodynamic and toxicological proﬁles of compounds. These parameters can be determined experimentally or

by using different calculation methods. The aim of the research was to determine the lipophilicity of betulin triazole derivatives with attached 1,4-quinone using thin layer chromatography in a

Citation: Kadela-Tomanek, M.; reverse phase system and a computer program to calculate its theoretical model. The physiochemical Jastrz˛ebska,M.; Marciniec, K.; and pharmacokinetic properties were also determined by computer programs. For all obtained Chrobak, E.; B˛ebenek,E.; Boryczka, S. parameters, the similarity analysis and multilinear regression were determined. The analyses showed Lipophilicity, Pharmacokinetic that there is a relationship between structure and properties under study. The molecular docking Properties, and Molecular Docking study showed that betulin triazole derivatives with attached 1,4-quinone could inhibit selected Study on SARS-CoV-2 Target for SARS-CoV-2 proteins. The MLR regression showed that there is a correlation between afﬁnity scoring Betulin Triazole Derivatives with values (∆G) and the physicochemical properties of the tested compounds. Attached 1,4-Quinone. Pharmaceutics 2021, 13, 781. https://doi.org/ Keywords: 1,4-quinone; betulin; lipophilicity; molecular docking; SARS-CoV-2 proteins 10.3390/pharmaceutics13060781

Academic Editor: Sourav Bhattacharjee 1. Introduction

Received: 6 May 2021 Nowadays, the search for new drugs requires the application of computational chem- Accepted: 20 May 2021 istry, including experimental and in silico analysis of physicochemical properties, phar- Published: 23 May 2021 macokinetic features, ADMET analysis, and quantitative structure–activity relationship (QSAR) research [1,2]. The therapeutic potential of a drug depends on its distribution in Publisher’s Note: MDPI stays neutral the body. Often, substances with high biological activity show low bioavailability and with regard to jurisdictional claims in high toxicity to healthy tissues. Nanocarriers, including liposomes, micelles, and polymer published maps and institutional affil- nanoparticles, enable the improvement of the biodistribution of substances. Incorporation iations. of the hydrophobic core into hydrophilic nanotransporters can effectively influence the solubility of the drug, which contributes to the improvement of permeability through cell membranes [3,4]. A key parameter in the design of new active compounds is lipophilicity, which in- Copyright: © 2021 by the authors. fluences the solubility and permeability through membranes. Lipophilicity affects the Licensee MDPI, Basel, Switzerland. pharmacodynamic and toxicological profiles of compounds. These parameters can be deter- This article is an open access article mined experimentally or by using different calculation methods. Experimental lipophilicity distributed under the terms and is usually determined by chromatographic methods, such as reversed phase thin layer chro- conditions of the Creative Commons matography (RP-TLC), normal phase thin layer chromatography (NP-TLC), or reversed Attribution (CC BY) license (https:// phase high performance liquid chromatography (RP-HPLC). Theoretical methods are the creativecommons.org/licenses/by/ second way to determine lipophilicity. However, the calculated value of lipophilicity 4.0/).

Pharmaceutics 2021, 13, 781. https://doi.org/10.3390/pharmaceutics13060781 https://www.mdpi.com/journal/pharmaceutics Pharmaceutics 2021, 13, x 2 of 22

Pharmaceutics 2021, 13, 781 2 of 21

ods are the second way to determine lipophilicity. However, the calculated value of lipo- philicitydepends depends on the on algorithmthe algorithm employed employed for calculation,for calculation, and and for manyfor many compounds compounds it differs it differsfrom from the the experimental experimental ones ones [5– [5–7].7]. NaturalNatural substances, substances, especially especially secondary secondary metabolites, metabolites, are are often often an an inspiration inspiration to to ob- obtain tain semisyntheticsemisyntheticcompounds, compounds, which which exhibit exhibi hight high biological biological activity. activity. Microorganisms Microorganisms produce producemany many secondary secondary metabolites metabolites that affectthat affect antibacterial, antibacterial, antiviral, antiviral, and anticancer and anticancer activities. activities.The Streptomyces The Streptomycesspecies species produce produce compounds compounds containing containing the 5,8-quinolinedionethe 5,8-quinolinedione moiety, moiety,such such as streptonigrin as streptonigrin, lavendamycin, lavendamycin, and, andstreptonigron streptonigron, which, which exhibit exhibit a wide a wide spectrum spec- of trumbiological of biological activity activity (Figure (Figure1)[ 1)8, 9[8,9].]. Previous Previous studies studies dealt dealt with with thethe modiﬁcationmodification of 5,8- 5,8-quinolinedionequinolinedione moiety moiety at at the the C-2, C-2, C-6, C-6, and and C-7 C-7 positions positions and and they they showed showed that thatsuch such modificationmodiﬁcation can lead can leadto changes to changes in biological in biological activity activity [10–19]. [10– 19].

H3CO 6

2 N COOH 7 H2N N O H2N CH3 OH

H3CO OCH3 Streptonigrin FigureFigure 1. Chemical 1. Chemical structure structure of 7-amino-5,8-quinolinedione of 7-amino-5,8-quinolinedione antibiotics. antibiotics.

StudiesStudies on the on lipophilicity the lipophilicity of 5,8-quinolinedione of 5,8-quinolinedione derivatives derivatives were were carried carried out apply- out apply- ing theing RP-TLC the RP-TLC and andRP-HPLC RP-HPLC methods. methods. It was It wasfound found that thatthere there is a relationship is a relationship between between lipophilicitylipophilicity and andin silico in silico pharmacokinetic pharmacokinetic properties properties of ofsynthetic synthetic 5,8-quinolinedione 5,8-quinolinedione de- deriva- rivativestives [20–23]. [20–23]. As aa continuationcontinuation of of our our previous previous research, research, we we determined determined the the lipophilicity lipo- philicityexperimentally experimentally as well as aswell by as using by using computational computational methods methods for the for betulin the betulin triazole tria- deriva- zole tivesderivatives with attached with attached 1,4-quinone. 1,4-quinone. The physicochemical The physicochemical and pharmacokinetic and pharmacokinetic properties in- propertiesfluence influence the bioavailability the bioavailability and biological and bi activityological of activity compounds of compounds when they arewhen determined they are determinedby computer by programs.computer programs. The correlation The correlation between lipophilicitybetween lipophilicity and in silico and determinedin silico determinedstructural structural parameters parameters have been have analyzed been analyzed in this study. in this study. The Thetested tested compounds compounds contain contain two active two activemoieties, moieties, i.e., 1,4-quinone i.e., 1,4-quinone and betulin. and The betulin. mainThe mechanism main mechanism of activity of for activity 1,4-quinone for 1,4-quinone derivatives derivatives is the interaction is the interaction with NQO1 with pro- NQO1 tein,protein, for which for the which overexpression the overexpression is observed is observed in many in manytypes typesof cancer of cancer cell lines cell lines[8,10,24]. [8,10 ,24]. The betulin derivatives show a wide spectrum of activities including anticancer, antiviral, The betulin derivatives show a wide spectrum of activities including anticancer, antiviral, antibacterial, and anti-inflammatory effects. The use of betulin and its derivatives in antibacterial, and anti-inflammatory effects. The use of betulin and its derivatives in treat- treatment is limited due to low bioavailability and low water solubility. For these reasons, ment is limited due to low bioavailability and low water solubility. For these reasons, the the use of new drug nanoencapsulation procedures and/or nanoparticle-based delivery use of new drug nanoencapsulation procedures and/or nanoparticle-based delivery meth- methods are a key issue. These are issues of nanomedicine interest [25,26]. ods are a key issue. These are issues of nanomedicine interest [25,26]. Over the past decade, many betulin derivatives exhibiting high antiviral activity Over the past decade, many betulin derivatives exhibiting high antiviral activity against human immunodeficiency virus (HIV-1), herpes simplex (HSV-1), enteric cy- against human immunodeficiency virus (HIV-1), herpes simplex (HSV-1), enteric cyto- topathogenic human orphan virus (ECHO-6), and human cytomegalovirus (HCMV) were pathogenic human orphan virus (ECHO-6), and human cytomegalovirus (HCMV) were described [27–31]. The high antiviral activity of betulin derivatives has aroused interest as described [27–31]. The high antiviral activity of betulin derivatives has aroused interest as a possibility of using them in the treatment of COVID-19 [32–35]. The aim of the present a possibility of using them in the treatment of COVID-19 [32–35]. The aim of the present study was to characterize the triazole betulin derivatives with attached 1,4-quinone in studyterms was ofto their characterize lipophilicity, the pharmacokinetictriazole betulin derivatives properties, andwith molecular attached docking1,4-quinone with in SARS- termsCoV-2 of their proteins, lipophilicity, such as Mpro pharmacokinetic and PLpro. Moreover, properties, we and examined molecular the correlation docking with between SARS-CoV-2scoring values proteins, (∆G) such and as structural Mpro and properties. PLpro. Moreover, we examined the correlation between scoring values (ΔG) and structural properties. 2. Materials and Methods 2. Materials2.1. Data and Set Methods 2.1. Data SetIn this study, a series of triazole betulin derivatives with attached 1,4-quinone (1–16), asIn wellthis study, as triazole a series betulin of triazole derivatives betulin (17 derivatives–20), were used. with attached According 1,4-quinone to literature (1–16 data), [36], as wellthe as reaction triazole between betulin triazolederivatives betulin (17–20 derivatives), were used. (17–20 According) and 1,4-quinone to literature compounds data [36], in the

Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x 3 of 22 Pharmaceutics 2021, 13, x the reaction between triazole betulin derivatives (17–20) and 1,4-quinone compounds3 of in22 the reaction between triazole betulin derivatives (17–20) and 1,4-quinone compounds in Pharmaceutics 2021, 13, 781 the reactionpresence between of potassium triazole carbonate betulin derivatives in tetrahydrofuran (17–20) and lead 1,4-quinone to the hybrids compounds 1–163 of. The 21 in the reactionpresence between of potassium triazole carbonate betulin derivatives in tetrahydrofuran (17–20) and lead 1,4-quinone to the hybrids compounds 1–16. The in thechemical reactionpresence structures between of potassium of triazole compounds carbonate betulin 1–20 derivatives in are tetrahydrofuran presented (17–20 in) Tableand lead 1,4-quinone 1.to the hybrids compounds 1–16. The in thechemical reactionpresence structures between of potassium of triazole compounds carbonate betulin 1–20 derivatives in are tetrahydrofuran presented (17–20 in) Tableand lead 1,4-quinone 1.to the hybrids compounds 1–16. The in thechemical reactionpresence structures between of potassium of triazole compounds carbonate betulin 1–20 derivativesin are tetrahydrofuran presented (17–20 in) Tableand lead 1,4-quinone 1.to the hybrids compounds 1–16. The in Tablechemicalthe 1.presence Chemical structures of structure potassium of compounds and biologicalcarbonate 1–20 activity in are tetrahydrofuran presentedof compounds in Table1–20 lead. 1.to the hybrids 1–16. The Tablepresencechemicalthe 1.presence Chemical of structures potassium of structure potassium of carbonate compounds and biologicalcarbonate in tetrahydrofuran 1–20 activity in are tetrahydrofuran presentedof compounds lead in to Tablethe1–20 lead hybrids. 1.to the1– hybrids16. The chemical1–16. The Tablechemical 1. Chemical structures structure of compounds and biological 1–20 activity are ofpresented compounds in Table1–20. 1. Compound ChemicalTablestructureschemical 1. Chemical Structure ofstructures compounds structure of compounds and1–20 biologicalare Compound presented 1–20 activity are in ofpresented Tablecompounds1. in ChemicalTable1–20. 1. Structure Compound ChemicalTable 1. Chemical Structure structure and biological Compound activity of compounds Chemical1–20. Structure Compound ChemicalTable 1. Chemical Structure structure and biological Compound activity of compounds Chemical1–20. Structure Compound Table ChemicalTable 1. 1.Chemical Chemical Structure structure structure and and biological biological Compound activity activity of of compounds compounds1– Chemical1–2020. . Structure Compound Chemical Structure Compound Chemical Structure Compound Chemical Structure Compound Chemical Structure Compound Chemical Chemical Structure Compound Compound Chemical Chemical Structure 1 2 1 2 1 2 1 2 1 2 1 2 11 22

3 4 3 4 3 4 3 4 3 4 33 44 3 4

5 6 5 6 5 6 5 6 55 66 5 6 5 6 O Cl O Cl O N ClO O O N Cl O O N N Cl O 7 O O N 8 N O N ClO O 7 N N 8 7 O NN Cl O 8 7 O O N O N 8 O NN 77 O O 88 O O N N N O O NN O 7 O O N O 8 O O NN O 7 O O N 8 O O NN 7 O O N 8 O O N O O N O O O O O O 9 10 9 10 99 1010 9 10 9 10 9 10 9 10

11 12 1111 1212 11 12 11 12 11 12 11 12 11 12

1313 1414 13 14 13 14 13 14 13 14 13 14 13 14

Pharmaceutics 2021, 13, 781 4 of 21

Pharmaceutics 2021, 13, x 4 of 22 Pharmaceutics 2021, 13, x 4 of 22 Pharmaceutics 2021, 13, x 4 of 22 Table 1. Cont.

Compound Chemical Structure Compound Chemical Structure

15 16 15 16 1515 1616

17 18 1717 1818 17 18

1919 2020 19 20 19 20

2.2. Experimental Lipophilicity 2.2.2.2. Experimental Experimental Lipophilicity Lipophilicity 2.2. ExperimentalThe experimental Lipophilicity lipophilicity was determined using the RP-TLC method according TheThe experimental experimental lipophilicity lipophilicity was was determined determined using using the the RP-TLC RP-TLC method method according according to theThe literature experimental [20,21,37]. lipophilicity We used wasthe modified determined silica using gel theas a RP-TLC stationary method phase according and (tris- toto the the literature literature [[20,21,37].20,21,37]. We We us useded the the modified modiﬁed silica silica gel gel as asa stationary a stationary phase phase and and (tris- tohydroxymethyl)aminomethane the literature [20,21,37]. We us (0.2ed theM, pHmodified = 7.4) withsilica acetone gel as a as stationary a mobile phasephase. and The (tris- per- (tris-hydroxymethyl)aminomethanehydroxymethyl)aminomethane (0.2 (0.2 M, M,pH pH = 7.4) = 7.4) with with acetone acetone as a as mobile a mobile phase. phase. The The per- hydroxymethyl)aminomethanecent of organic solvent volume was(0.2 M,varied pH =wi 7.4)thin with the acetonerange of as 60–90% a mobile in 5%phase. increments. The per- percentcent of of organic organic solvent solvent volume volume was was varied varied wi withinthin the the range of 60–90% inin 5%5% increments.increments. cent ofThe organic compounds solvent 1–20 volume were was dissolved varied in wi chloroformthin the range (1.0 of mg/mL), 60–90% then in 5% 5µL increments. of sample TheThe compounds compounds1– 1–2020 were were dissolved dissolved in in chloroform chloroform (1.0 (1.0 mg/mL), mg/mL), then then 5 µ5µLL of of sample sample solutionThe wascompounds applied to1–20 the were chromatographic dissolved in chloroform plates with (1.0 a micropipette. mg/mL), then Spots 5µL were of sample visu- solutionsolution was was applied applied toto the chromatographic chromatographic plates plates with with a micropipette. a micropipette. Spots Spots were were visu- solutionalized by was spraying applied with to the 10% chromatographic ethanol solution plates of sulfuric with aacid micropipette. (VI) and then Spots heated were upvisu- to visualizedalized by byspraying spraying with with 10% 10% ethanol ethanol solution solution of of sulfuric sulfuric acid acid (VI) (VI) and and then then heated heated up up to alized110 °C.◦ by spraying with 10% ethanol solution of sulfuric acid (VI) and then heated up to to110 110 °C.C. 110 °C.The obtained values of retardation factor (Rf) were converted to RM parameters ac- TheThe obtained obtained values values ofof retardationretardation factorfactor (Rff)) were were converted converted to to R RMM parametersparameters ac- accordingcordingThe to toobtained Equation Equation values (1): (1): of retardation factor (Rf) were converted to RM parameters according to Equation (1): cording to Equation (1): 1 1 RM = log −1 1 (1) R =logR 1 −1 (1) R =logf R −1 (1) R =logR −1 (1) The R parameter was calculated for every concentrationR of acetone and extrapolated The MRM parameter was calculated for every concentration of acetone and extrapolated to zeroThe concentration RM parameter of was organic calculated component for every in the conc mobileentration phase. of acetone The chromatographic and extrapolated to zeroThe concentration RM parameter of was organic calculated component for every in conc the entrationmobile phase. of acetone The chromatographicand extrapolated parameterto zero concentration of lipophilicity of (Rorganic) was component calculated in using the Equationmobile phase. (2): The chromatographic toparameter zero concentration of lipophilicity of organic (RM0M0) was component calculated in using the mobile Equation phase. (2): The chromatographic parameter of lipophilicity (RM0) was calculated using Equation (2): parameter of lipophilicity (RM0) was calculated using Equation (2): R R= R=R+ b+𝑏CC (2)(2) M R =RM0 +𝑏C (2) R =R +𝑏C (2) where C is the concentration of acetone in the mobile phase, while b is the slope of the wherewhere C C is is the the concentration concentration of of acetone acetone in in the the mobile mobile phase, phase, while whileb bis is the the slope slope of of the the whereregression C is plot.the concentration of acetone in the mobile phase, while b is the slope of the regressionregression plot. plot. regressionThe hydrophobic plot. indexϕ (φ0) was determined according to Equation (3): TheThe hydrophobic hydrophobic index index ( (φ0)0) was was determined determined according according to to Equation Equation (3): (3): The hydrophobic index (φ0) was determinedR according to Equation (3): R φ =−RM0R (3) ϕ0 =φ−=−R𝑏 (3)(3) φ =−b 𝑏 (3) 𝑏 2.3.2.3. Theoretical Theoretical Lipophilicity Lipophilicity 2.3. Theoretical Lipophilicity 2.3.For TheoreticalFor compounds compounds Lipophilicity1–20, 1–20, the the calculated calculated lipophilicity lipophilicity was was determined determined using using various various For compounds 1–20, the calculated lipophilicity was determined using various onlineonlineFor tools tools compounds and and free free available available1–20, the software, software,calculated including: including: lipophilicity ALOGPs, ALOGPs, was determined AClogP,AClogP, AlogP, AlogP, using MLOGP, MLOGP, various online tools and free available software, including: ALOGPs, AClogP, AlogP, MLOGP, XLOGP2,onlineXLOGP2, tools XLOGP3 XLOGP3 and free (German (German available Research Research software, Center Center including: for for Environmental Environmental ALOGPs, AClogP, Health, Health, AlogP, Neuherberg, Neuherberg, MLOGP, XLOGP2, XLOGP3 (German Research Center for Environmental Health, Neuherberg, Germany),XLOGP2,Germany), milogPXLOGP3 milogP (Molinspiration (Molinspiration (German Research Cheminformatics, Cheminform Center foratics, Environmental Slovensky Slovensky Grob, Grob, Health, Slovak Slovak Neuherberg, Republic), Republic), Germany), milogP (Molinspiration Cheminformatics, Slovensky Grob, Slovak Republic), iLOGP,Germany),iLOGP, WLOGP, WLOGP, milogP and and (Molinspiration SILICOS-IT SILICOS-IT (Swiss (SwissCheminform Institute Instituteatics, of of Bioinformatics, Bioinformatics,Slovensky Grob, Lausanne, Lausanne, Slovak Republic), Switzer- Switzer- iLOGP, WLOGP, and SILICOS-IT (Swiss Institute of Bioinformatics, Lausanne, Switzer- land)iLOGP,land) [38 [38–41]. –WLOGP,41]. TheThe and pharmacokineticpharmacokinetic SILICOS-IT (Swiss and and physio physiochemical Institutechemical of Bioinformatics, parameters parameters were Lausanne, were determined determined Switzer- us- land) [38–41]. The pharmacokinetic and physiochemical parameters were determined us- land)ing pKCMS [38–41]. (Bio21 The pharmacokinetic Institute, Melbourne, and physio Australia)chemical and parametersSwissADME were software determined (Swiss us-In- ing pKCMS (Bio21 Institute, Melbourne, Australia) and SwissADME software (Swiss In- ingstitute pKCMS of Bioinformatics, (Bio21 Institute, Lausanne, Melbourne, Switzerland) Australia) [40–43]. and SwissADME software (Swiss In- stitute of Bioinformatics, Lausanne, Switzerland) [40–43]. stitute of Bioinformatics, Lausanne, Switzerland) [40–43].

Pharmaceutics 2021, 13, 781 5 of 21

using pKCMS (Bio21 Institute, Melbourne, Australia) and SwissADME software (Swiss Institute of Bioinformatics, Lausanne, Switzerland) [40–43].

2.4. Structure Optimization The optimized chemical structures of compounds 1–16 were calculated using the DFT (B3LYP/6-311G+(d.p)) method implemented in the Gaussian 16 program package [44]. All local minima of energy were conﬁrmed by the absence of an imaginary mode in the vibrational calculations. In our calculations, we applied the basis set of the diffuse function to heavy atoms (+) to obtain a better description of lone pair electrons with orbitals occupying a larger region of space. The obtained optimized structure is presented in Figure S1. The geometries of compounds 1–16 were used to determine the HOMO–LUMO energy, quantum chemical descriptor, the molecular electrostatic potential, and the molecular docking study [45]. All obtained results were visualized using the GaussView, Version 5 software package (Gaussian Inc., Wallingford, CT, USA) [46].

2.5. Molecular Docking Study The three-dimensional (3D) structures (in mol2 format) of the studied compounds were generated using the ChemOffice package (version 19.1, PerkinElmer, Waltham, MA, USA) [47]. Their low-energy conformations were calculated using Gaussian 16 (revision A.03) computer code [44] at the density functional theory (DFT, B3LYP) and 6-311+G(d,p) basis sets. Calculations were performed using the X-ray coordinates of chloroquine as the input structure obtained from the Cambridge Crystallographic Data Centre (CCDC ID: CDMQUI). Target macromolecule for molecular docking study was obtained from the Protein Data Bank (https://www.rcsb.org/, accessed on 1 April 2020). We used the 3D crystal structures of COVID-19 main protein and papain-like protease of SARS CoV-2 (PDB ID: 5R7Z and 6W9C, respectively). Ligands and proteins used in the calculations were prepared for docking using the AutoDockTools program (Molecular Graphics Laboratory, La Jolla, CA, USA) [48]. In this study, AutoDock Vina [49] tool compiled in PyRx [50] was employed to perform molecular docking. Implementing in AutoDock Vina scoring function was mostly inspired by X-score and combined an empirical free-energy force field with a Lamarckian Genetic Algorithm. The volume was set as 25 × 25 × 25 Å. The region of interest used for AutoDock Vina docking was fixed as X = −33,784, Y = 20,933, Z = 33,306 for papain-like protease and X = −11,631, Y = 2201, Z = 23,194 for COVID-19 main protein. After calculations, only the nine highest-scored poses were returned as a docking result for ligand–cavity configuration. The complexes obtained in the Vina program were visualized using the BIOVIA Discovery Studio virtual environment (v.17.2.0.16349, Dassault systems, Vélizy- Villacoublay, France) [51].

2.6. Correlation and Cluster Analysis Based on the experimental and theoretical values of lipophilicity and molecular descriptor values, the correlation and cluster analysis were performed. All data used for the cluster analysis were standardized and the cluster analysis was based on the Euclidean distance. The analysis was carried out using the Statistica 13.1 software (TIBCO Software Inc., Palo Alto, CA, USA).

3. Results and Discussion 3.1. Experimental and Theoretical Lipophilicity The hybrids 1–16 and substrates 17–20 were examined under the RP-TLC method. For each compound, the parameter RM was calculated from the retardation factor (Rf) according to Equation (1). Equation (2) was used to determine the RM0 and b values and the results are presented in Table2. The high correlation coefﬁcients (r = 0.976–0.997) for all compounds show good correlation between acetone concentration and RM value. Pharmaceutics 2021, 13, 781 6 of 21

Table 2. The experimental values of lipophilicity (RM0) for compounds 1–20.

Compound RM0 b r SD 1 3.51 −0.04 0.995 0.051 2 3.64 −0.04 0.992 0.067 3 3.97 −0.05 0.996 0.047 4 4.08 −0.05 0.992 0.072 5 3.56 −0.04 0.983 0.100 6 3.59 −0.04 0.997 0.041 7 4.07 −0.05 0.985 0.101 8 4.24 −0.05 0.993 0.065 9 3.65 −0.04 0.996 0.047 10 3.77 −0.04 0.996 0.048 11 4.12 −0.05 0.995 0.054 12 4.39 −0.05 0.981 0.115 13 3.78 −0.04 0.998 0.028 14 3.84 −0.04 0.992 0.067 15 4.21 −0.05 0.979 0.120 16 4.58 −0.05 0.960 0.183 17 4.40 −0.05 0.995 0.065 18 4.54 −0.05 0.989 0.098 19 5.07 −0.06 0.995 0.073 20 5.26 −0.06 0.976 0.161

b is the slope, r is the correlation coefficient, and SD is the standard deviation for the linear relationship RM = RM0 + bC.

The calibration curve is required for the conversion of RM0 to logPTLC. The following compounds were used as standard substances: acetanilide, prednisone, 4-bromoacetophenone, benzophenone, anthracene, dibenzyl, 9-phenylanthracene, and dichlorodiphenyltrichloroethane (DDT), for which the literature values of logPlit are in the range 1.21–6.38 [52,53]. The RM0 for the standard substance was determined in the same conditions as for compounds 1–20 (Table3).

Table 3. The experimental (RM0) and literature (logPlit) lipophilicity values of standard substance.

Compound logPlit RM0 b r SD logPTLC Acetanilide 1.21 0.55 −0.01 0.954 0.051 1.19 Prednisone 1.63 0.73 −0.02 0.932 0.077 1.39 4-Bromoacetophenone 2.43 1.89 −0.03 0.993 0.037 2.72 Benzophenone 3.18 2.37 −0.03 0.993 0.046 3.26 Anthracene 4.45 3.48 −0.04 0.994 0.053 4.52 Dibenzyl 4.79 3.65 −0.04 0.997 0.043 4.72 DDT 6.01 4.87 −0.06 0.985 0.041 6.17 9-Phenylanthracene 6.38 4.93 −0.06 0.993 0.080 6.11

b is the slope, r is the correlation coefﬁcient and SD is the standard deviation for the linear relationship RM = RM0 + bC.

The calibration curve Equation (4) was determined by linear correlation between the logPlit and RM.

logPTLC = 1.139 RM0 + 0.561 (r = 0.996; SD = 0.188) (4)

For the standard substances the logPTLC was calculated according to the calibration curve (Equation (4)). Figure S2 shows a good agreement between experimental and literature values of lipophilicity (r = 0.996). Equations (3) and (4) were used to determine the logPTLC and the hydrophobic index (ϕ0), respectively. The results are presented in Figure2 and Table S1. Pharmaceutics 2021, 13, x 7 of 22

Pharmaceutics 2021, 13, 781 7 of 21 Equations (3) and (4) were used to determine the logPTLC and the hydrophobic index (φ0), respectively. The results are presented in Figure 2 and Table S1.

(a) (b)

Figure 2. TheThe experimental experimental values values of of (a ()a )lipophilicity lipophilicity (logP (logPTLCTLC) and) and (b) ( bhydrophobic) hydrophobic index index (φ0) ( ϕfor0) 5,8-quinolinedione for 5,8-quinolinedione 1–4 1(yellow),–4 (yellow), 5,8-isoquinoline 5,8-isoquinoline 5–8 (violet),5–8 (violet), 2-methylo-5,8-quinolinedione 2-methylo-5,8-quinolinedione 9–12 (blue),9–12 (blue), 1,4-naphthoquinone 1,4-naphthoquinone 13–16 (green)13–16 (green) derivatives,derivatives, and substrate and substrate 17–2017 (grey).–20 (grey).

ForFor compounds 1–20,1–20, lipophilicitylipophilicity is is in in the the range range of of 4.56–6.55. 4.56–6.55. For For a a series series of of com- com- pounds 17–2017–20,, the the lipophilicity lipophilicity depends depends on on the the type type of group at the C-3 position of the betulinbetulin moiety. moiety. Derivatives with with hydroxyl ( (1717)) or or oxo ( 18) groups at this position exhibit comparablecomparable values values of of logP logPTLCTLC. Replacement. Replacement of ofthe the hydroxyl hydroxyl group group by acyl by acyl substituents substituents (19 and(19 and 20) causes20) causes an increase an increase in lipophilicity. in lipophilicity. ComparingComparing the the logP logPTLCTLC forfor hybrids hybrids 1–161– 16andand derivatives derivatives 17–2017– shows20 shows that thatintroduc- intro- tionduction of the of 1,4-quinone the 1,4-quinone fragment fragment to the to betulin the betulin moiety moiety reduces reduces the lipophilicity. the lipophilicity. The lipo- The philicitylipophilicity depends depends on the on type the typeof 1,4-quinone of 1,4-quinone moiety moiety and the and order the order is as follows: is as follows: 5,8-quin- 5,8- olinedionequinolinedione (1–4) ( 1>– 5,8-isoquinolinedione4) > 5,8-isoquinolinedione (5–8) ( 5>– 2-methyl-5,8-quinolinedione8) > 2-methyl-5,8-quinolinedione (9–12 (9) –>12 1,4-) > naphthoquinone1,4-naphthoquinone (13–16 (13).– 16). ComparingComparing the lipophilicity of compounds 1–161–16 shows that the type of substituent at thethe C-3 C-3 position position of of betulin betulin moiety moiety influences inﬂuences the the lipophilicity, lipophilicity, and and the the order order is is as as follows: follows: hydroxyhydroxy ( (11,, 5,, 9,, and 13) > oxo oxo ( 2,, 6, 10, and 14) >> ethanoyl ( 3, 7, 11, andand 15)) >> propanoyl ((4,, 8,, 12,, and and 16). TheThe hydrophobic indexes indexes for for hybrids hybrids 1–161–16 and compounds 17–2017–20 are in the ranges ϕ ofof 79.81–88.21 79.81–88.21 and 82.62–87.13, re respectively.spectively. A higher value of φ0 means that compounds 1–201–20 areare less less soluble soluble in in water. water. Comparing Comparing the the hydrophobicity hydrophobicity of derivatives of derivatives 1–161 –and16 and17– 2017 –shows20 shows that that the theintroduction introduction of 1,4-quinone of 1,4-quinone moiety moiety slightly slightly affects affects the solubility the solubility in wa- in ter.water. In the In theseries series of 1–16, of 1–16, no relationshipno relationship between between the thetype type of 1,4-qinone of 1,4-qinone moiety moiety and andhy- drophobicityhydrophobicity index index has has been been observed. observed. TheoreticalTheoretical valuesvalues ofof lipophilicitylipophilicity were were determined determined with with the the available available programs programs[38 –[38–41]. 41].The The calculated calculated values values of logPof logP cover cover a widea wide range range from from 4.11 4.11 to to 12.19 12.19 dependingdepending on the mathematicalmathematical module used by programs. programs. The The th theoreticaleoretical results results are presented in Figure 33 andand Table Table S2. For 1–20, all programs show that lipophilicity depends on the type of substituent at the C-3 position of the betulin moiety, and this correlation is consistent with the experimental results. As shown in Figure3, compounds 17–20 have lower lipophilicity than hybrids 1–16, which is not in agreement with the experimental results. The exception is the MLOGP program, for which the theoretical lipophilicity for derivatives 17–20 are similar to the experimental values (Table S2). Comparing the logP values for 1–16 shows that the theoretical lipophilicity slightly depends on the type of 1,4-quinone moiety. Moreover, hybrids with 5,8-quinolinedione (1–4) and 5,8-isoquinolinedione (5–8) moiety have the same molecular formula. In this case, atomistic (WLOGP) and topological (MOLGP and SILCOS- IT) methods exhibit the same value of lipophilicity, but the LogPTLC is different. For this reason, to better predict the lipophilicity, it is important to ﬁnd a correlation between its experimental and theoretical values. Due to the structural differences of hybrids 1–16 and betulin derivatives 17–20, separate equations were developed for these two groups of compounds (Table4).

Pharmaceutics 2021, 13, 781 8 of 21 Pharmaceutics 2021, 13, x 8 of 22

FigureFigure 3.3. TheThe proﬁleprofile ofof changeschanges forfor theoreticaltheoretical lipophilicitylipophilicity forfor compoundscompounds1 1–20–20..

Table 4. Correlation equations forFor experimental 1–20, all (LogPprogramsTLC) and show theoretical that lipophilicity (LogPcalc) values depends of lipophilicity on the type for of compounds substituent at 1–16 and 17–20. the C-3 position of the betulin moiety, and this correlation is consistent with the experimental results. As shown in Figure 3, compounds 17–20 have lower lipophilicity than hy- Program Correlation Equation r SD brids 1–16, which is not in agreement with the experimental results. The exception is the MLOGP program, forCompounds which the 1theoretical–16 lipophilicity for derivatives 17–20 are similar − ALOGPsto logP theTLC experimental= 1.094 logP valuesCALC (Table2.416 S2). Comparing 0.932 the logP values for 1–16 0.137 shows that the AClogP logP = 0.540 logP + 1.104 0.854 0.196 theoreticalTLC lipophilicityCALC slightly depends on the type of 1,4-quinone moiety. Moreover, AlogP logPTLC = 0.492 logPCALC + 0.987 0.843 0.203 hybrids with 5,8-quinolinedione (1–4) and 5,8-isoquinolinedione (5–8) moiety have the XLOGP2 logPTLC = 0.378 logPCALC + 1.268 0.854 0.196 XLOGP3same logP TLCmolecular= 0.472 logPformula.CALC − In0.121 this case, atomistic 0.887 (WLOGP) and topological 0.174 (MOLGP and milogPSILCOS-IT) logPTLC = 0.527 methods logPCALC exhibit+ 0.549 the same value of 0.729 lipophilicity, but the LogP 0.258TLC is different. iLOGPFor logP thisTLC reason,= 0.758 to logP betterCALC predict+ 0.620 the lipophilicity, 0.887 it is important to find a 0.174 correlation be- − WLOGPtween logPTLC its =experimental 0.755 logPCALC and 2.091theoretical values. Due 0.953 to the structural differences 0.114 of hybrids MLOGP logP = 0.582 logP + 2.353 0.723 0.260 1–16 andTLC betulin derivativesCALC 17–20, separate equations were developed for these two SILICOS-IT logPTLC = 0.634 logPCALC − 0.443 0.798 0.227 groups of compoundsCompounds (Table 4). 17–20 ALOGPs logPTLC = 1.313 logPCALC − 1.485 0.980 0.114 AClogPTable logP TLC4. Correlation= 1.078 logP equationsCALC − 0.058 for experimental (LogP 0.962TLC) and theoretical (LogPcalc 0.157) values of lipo- AlogPphilicity logPTLC for= compounds 0.847 logPCALC 1–16+ and 0.251 17–20. 0.962 0.232 XLOGP2 logPTLC = 0.567 logPCALC + 1.462 0.880 0.273 XLOGP3 logPProgramTLC = 0.718 logPCALC − 0.097Correlation Equation 0.929 r 0.213 SD milogP logPTLC = 0.766 logPCALC + 0.345Compounds 0.960 1–16 0.160 iLOGP logP = 0.994 logP + 0.916 0.924 0.220 ALOGPsTLC CALC logPTLC = 1.094 logPCALC − 2.416 0.932 0.137 WLOGP logPTLC = 1.083 logPCALC − 1.828 0.979 0.116 AClogP logPTLC = 0.540 logPCALC + 1.104 0.854 0.196 MLOGP logPTLC = 1.754 logPCALC − 3.543 0.955 0.170 TLC CALC SILICOS-IT logPTLCAlogP= 0.965 logPCALC −logP0.092 = 0.492 logP 0.815 + 0.987 0.843 0.332 0.203 XLOGP2 logPTLC = 0.378 logPCALC + 1.268 0.854 0.196 XLOGP3 logPTLC = 0.472 logPCALC − 0.121 0.887 0.174 The relationship between the lipophilicity and structure of compounds 1–20 was analyzedmilogP by cluster analysislogP (FigureTLC = 0.5274). logPCALC + 0.549 0.729 0.258 AsiLOGP shown in Figure4,logP compoundsTLC = 0.7581 –logP20 areCALC arranged + 0.620 in two main 0.887 clusters. The 0.174 ﬁrst clusterWLOGP consists of betulin derivativeslogPTLC = 0.75517–20 logPandCALC the − second2.091 one of hybrids 0.9531 –16. The 0.114 high valueMLOGP of Euclidean distancelogP betweenTLC = 0.582 two logP clustersCALC suggests+ 2.353 that there 0.723 is a low correlation 0.260 betweenSILICOS-IT structures of theselogP twoTLC groups = 0.634 of logP compounds.CALC − 0.443 0.798 0.227 The second cluster can be dividedCompounds into four 17–20 subclusters. In the ﬁrst subcluster, the hybridsALOGPs are arranged accordinglogPTLC to = the 1.313 type logP of betulinCALC − 1.485 moiety, which means 0.980 that they 0.114 have oxo groupAClogP at the C-3 positionlogPTLC of the= 1.078 betulin logP moiety.CALC − 0.058 The second and 0.962 third subclusters 0.157 consistAlogP of 1,4-naphthoquinone logP (TLC13 ,=15 0.847, and logP16)CALC and + 5,8-quinolinedione 0.251 0.962 moiety (5, 7, 0.232 and 8), respectively. Compounds containing the 5,8-quinolinedione (1, 3, and 4) and 2-methyl-5,8- XLOGP2 logPTLC = 0.567 logPCALC + 1.462 0.880 0.273

Pharmaceutics 2021, 13, x 9 of 22

XLOGP3 logPTLC = 0.718 logPCALC − 0.097 0.929 0.213 milogP logPTLC = 0.766 logPCALC + 0.345 0.960 0.160 iLOGP logPTLC = 0.994 logPCALC + 0.916 0.924 0.220 WLOGP logPTLC = 1.083 logPCALC − 1.828 0.979 0.116 Pharmaceutics 2021, 13, 781 9 of 21 MLOGP logPTLC = 1.754 logPCALC − 3.543 0.955 0.170 SILICOS-IT logPTLC = 0.965 logPCALC − 0.092 0.815 0.332

quinolinedioneThe relationship (9, 11, between and 12) arethe includedlipophilicity into and the subclusterstructure of four. compounds The similarity 1–20 analysiswas an- alyzedshows by a strong cluster correlation analysis (Figure between 4). lipophilicity and structure of hybrids 1–16.

FigureFigure 4. 4. SimilaritySimilarity analysis analysis of ofthe the experimental experimental and and theoretical theoretical values values of lipophilicity of lipophilicity for com- for com- poundspounds 1–201–20..

3.2. PhysicochemicalAs shown in Figure and Pharmacokinetic 4, compounds Properties1–20 are arranged in two main clusters. The first clusterIn consists the early of drugbetulin discovery derivatives process, 17–20 one and of the the second important one of stages hybrids is to 1–16 optimize. The high the valuephysicochemical of Euclidean properties distance between of a potential two cluste drug.rsBased suggests on thethat observation there is a low of correlation the physic- betweenochemical structures properties of these of oral two drugs, groups Lipinski of compounds. formulated the Rule of Five. According to this rule,The second the lipophilicity cluster can (logP), be divided molecular into weight four subclusters. (MW), number In the of acceptorsfirst subcluster, (HA), andthe hybridsdonors (HD)are arranged of hydrogen according bond shouldto the type be a of multiple betulin ofmoiety, ﬁve [54 which]. In an means attempt that to they improve have oxothe predictiongroup at the of C-3 bioavailability, position of the the betulin rules were moiety. expanded The second by Veber and tothird the subclusters number of rotat-con- sistable of bonds 1,4-naphthoquinone (RB) and topological (13, polar15, and surface 16) and area 5,8-quinolinedione (TPSA) are also determined moiety (5 [,55 7, 56and]. The 8), respectively.molecular descriptors Compounds for testedcontaining hybrids the1 –5,8-quinolinedione16 are presented in (1 Table, 3, and5. 4) and 2-methyl- 5,8-quinolinedioneThe tested hybrids (9, 11,1 and–16 have12) are a molecularincluded into mass the above subcluster 500 g/mol, four. The which similarity means analysisthey do shows not meet a strong the mass correlation criterion. between All compounds lipophilicity have and less structure than ﬁve of hydrogenhybrids 1–16 bond. donors (HD = 0–1). The 5,8-quinolinedione and 5,8-isoquinolinedione derivatives with 3.2.hydroxyl Physicochemical and oxo groupsand Pharmacokinetic at C-3 position Properties of betulin (1–2, 5–6, and 9–10) have less than 10 hydrogen bond acceptors (HA = 9–10) and the logP less than 5, which means that In the early drug discovery process, one of the importantTLC stages is to optimize the these compounds meet three of four Lipinski rules. Only compounds with propanoyl physicochemical properties of a potential drug. Based on the observation of the physico- substituents at the C-3 position of the betulin moiety (4, 8, 12, and 16) do not meet the chemical properties of oral drugs, Lipinski formulated the Rule of Five. According to this Veber’s rules concerning the number of rotatable bonds (RB < 10). The TPSA for 1–16 is less than 140 Å, which determines the oral bioavailability. For compounds 1–16, the relationship between the type of 1,4-quinone moiety, the physicochemical parameters and experimental lipophilicity was analyzed by means of a dendrogram (Figure5). Pharmaceutics 2021, 13, 781 10 of 21

Table 5. Molecular descriptors for hybrids 1–16.

Compound MW (g/mol) HA HD RB TPSA (Å) 1 787.42 10 1 9 133.50 2 785.41 10 0 9 130.34 3 829.46 11 0 10 139.57 4 843.49 11 0 11 139.57 5 787.42 10 1 9 133.50 6 785.41 10 0 9 130.45 7 829.46 11 0 10 139.57 8 843.39 11 0 11 139.57 9 801.45 10 1 9 133.50 10 799.43 10 0 9 130.34 11 843.49 11 0 10 139.57 12 857.51 11 0 11 139.57 13 786.44 9 1 9 120.61 14 784.42 9 0 9 117.45 Pharmaceutics 2021, 13, x 15 828.47 10 0 10 126.6811 of 22

16 842.50 10 0 11 126.68

FigureFigure 5 5.. SimilaritySimilarity analysis analysis of of the the physicochemical physicochemical parameters parameters and and the theexperimental experimental lipophilicity lipophilicity forfor compounds compounds 1–161–16. .

TheThe similarity similarity analysis analysis shows shows two two main main cluste clusters.rs. The first The cluster first clusterconsists consistsof 1,4-naph- of 1,4- thoquinonenaphthoquinone derivatives derivatives (13 and (13 14and) and14) the and second the second one oneis divided is divided into into four four subclusters. subclusters. AnalysisAnalysis of of subclusters subclusters show show that that compounds compounds with with the the 5,8-quinolinedione 5,8-quinolinedione (1–4 (1) –and4) and 2- 2- methyl-5,8-quinolinedionemethyl-5,8-quinolinedione (5–8 (5–)8 exhibit) exhibit similar similar properties. properties. Comparing Comparing the properties the properties of 5,8-quinolinedioneof 5,8-quinolinedione (1–4 ()1 and–4) and5,8-isoquinolinedione 5,8-isoquinolinedione (9–11 ()9 shows–11) shows that the that position the position of ni- of trogennitrogen atom atom in heterocyclic in heterocyclic ring ringinfluences influences their physicochemical their physicochemical parameters. parameters. The simi- The laritysimilarity analysis analysis shows shows that nitrogen that nitrogen atom atomsignificantly significantly affects affects properties properties of the tested of the testedde- rivatives.derivatives. TheThe multilinear multilinear regression (MLR) Equation (5)(5) expressesexpresses thethe relationshiprelationshipbetween between the theexperimental experimental lipophilicity lipophilicity (logP (logPTLCTLC) and) and physicochemical physicochemical parameters, parameters, suchsuch asas molecularmolec- ularmass mass (MW), (MW), topological topological polar polar surface surface (TPSA), (TPSA), and and number number of rotatableof rotatable bonds bonds (RB). (RB). logPTLC = 0.992MW − 0.488TPSA + 0.213RB − 3.362 logP = 0.992MW − 0.488TPSA + 0.213RB − 3.362 TLC (5) (5) (r(r = = 0.985, 0.985, rr22 == 0.969, SD = = 1.11, 1.11, VIF VIF = = 3.12, 3.12, F F= =126.7) 126.7) The correlation coefficient shows good agreement between physicochemical parameters and experimental lipophilicity. A comparison of the experimental and calculated parameters for the investigated compounds are shown in Table S3. The result shows that lipophilicity could be determined by in silico parameters. The physicochemical properties were used to in silico calculate the pharmacokinetic parameters, which determine the absorption of the potential drug [57]. The prediction of oral and transdermal absorption was performed in silico using the Caco-2 cell (logPapp), human intestinal absorption (HIA), and skin permeability (logKp) models. Moreover, the neurotoxicity of the compounds can be designated by blood–brain barrier permeability (logBB) and central nervous system (logPS) penetration [12,58–60]. The pharmacokinetic parameters of compounds 1–16 have been determined in silico by pkCSM software and they are presented in Table 6.

Pharmaceutics 2021, 13, 781 11 of 21

The correlation coefﬁcient shows good agreement between physicochemical parameters and experimental lipophilicity. A comparison of the experimental and calculated parameters for the investigated compounds are shown in Table S3. The result shows that lipophilicity could be determined by in silico parameters. The physicochemical properties were used to in silico calculate the pharmacokinetic parameters, which determine the absorption of the potential drug [57]. The prediction of oral and transdermal absorption was performed in silico using the Caco-2 cell (logPapp), human intestinal absorption (HIA), and skin permeability (logKp) models. Moreover, the neurotoxicity of the compounds can be designated by blood–brain barrier permeability (logBB) and central nervous system (logPS) penetration [12,58–60]. The pharmacokinetic parameters of compounds 1–16 have been determined in silico by pkCSM software and they are presented in Table6.

Table 6. Pharmacokinetic parameters of hybrids 1–16.

Compound logBB logPS logKp logPapp HIA 1 −1.459 −2.481 −2.734 0.641 97.226 2 −1.581 −2.513 −2.735 0.639 98.193 3 −1.804 −2.442 −2.735 0.195 98.510 4 −1.831 −2.434 −2.735 0.215 98.358 5 −1.457 −2.466 −2.735 0.672 97.226 6 −1.579 −2.499 −2.735 0.671 98.193 7 −1.802 −2.428 −2.735 0.158 98.510 8 −1.830 −2.420 −2.735 0.178 98.358 9 −1.447 −2.443 −2.734 0.598 97.235 10 −1.569 −2.475 −2.735 0.597 98.090 11 −1.792 −2.404 −2.735 0.181 98.358 12 −1.820 −2.396 −2.735 0.200 98.231 13 −1.219 −2.215 −2.734 0.747 97.410 14 −1.341 −2.248 −2.735 0.745 97.838 15 −1.564 −2.177 −2.735 0.733 97.885 16 −1.592 −2.169 −2.735 0.728 97.879

According to the model used in the pkCMS software, if logPapp is more than 0.9, it means that the compound expresses the high Caco-2 permeability [42]. For the tested hybrids 1–12, the logPapp value is in the range 0.200–0.672, which means moderate Caco-2 permeability. In the series of 5,8-quinolinedione hybrids (1–12), compounds with hydroxyl or oxo groups at the C-3 position of betulin moiety (1–2, 5–6, 9–10) have higher values of logPapp than hybrids with acyl group at this position (3–4, 7–8, 11–12). In the group of 1,4-naphthoquinone compounds (13–16), the Caco-2 permeability does not depend on the type of substituent at the C-3 position of the betulin moiety. Comparing the logPapp values shows that the permeability depends on the type of 1,4-quinone moiety and the following order is fulﬁlled: 1,4-naphthoquinone > 5,8-quinolinedione > 2-metyl-5,8-quionolinedione > 5,8-isoquinolnedione. For 1–16 the human intestinal absorption (HIA) are in the range of (97.226–98.510), which are considered to be high. The HIA index slightly depends on the type of 1,4-quinonemoiety. Hybrids 1–16 have a moderate skin permeability because they have the logKp lower than −2.5. For 1–16, the logBB are lower than −1, which means these hybrids poorly cross the brain–blood barrier. Compounds do not pass into the central nervous system when the logPS is less than −2. This result shows that the tested derivatives 1–16 are not neurotoxic. The similarity analysis showed no correlation between pharmacokinetic parameters and the structure of compounds 1–16 (Figure S3). However, the cluster analysis showed that only betulin moiety inﬂuences the pharmacokinetic parameters of hybrids. Pharmaceutics 2021, 13, 781 12 of 21

3.3. Molecular Properties The energy of HOMO and LUMO orbitals determines the ability of the molecule to donated or receive an electron. The energy of orbitals can be used to calculate the global reactivity descriptors, such as: ionization potential (I), electron afﬁnity (A), hardness (η), chemical potential (µ), electronegativity (χ), and electrophilicity index (ω)[61]. The HOMO and LUMO orbitals of 1–16 are presented in Figure S4. Table7 shows the HOMO and LUMO energy, the energy gap (∆E = EHOMO − ELUMO), the global reactivity descriptors, and dipole moment (DM).

Table 7. Electrostatic descriptor of hybrids 1–16.

E E ∆E I A η µ χ ω Hybrid HOMO LUMO DM (D) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) 1 −6.227 −3.918 −2.310 6.227 3.918 1.155 −5.073 5.073 11.140 5.6507 2 −6.510 −3.989 −2.521 6.510 3.989 1.261 −5.250 5.250 10.930 9.563 3 −6.423 −3.980 −2.443 6.423 3.980 1.222 −5.202 5.202 11.076 6.3035 4 −6.422 −3.984 −2.438 6.422 3.984 1.219 −5.203 5.203 11.106 6.5282 5 −6.503 −3.816 −2.687 6.503 3.816 1.344 −5.159 5.159 9.906 5.8305 6 −6.228 −3.869 −2.360 6.228 3.869 1.180 −5.049 5.049 10.801 8.9646 7 −6.421 −3.867 −2.554 6.421 3.867 1.277 −5.144 5.144 10.359 6.548 8 −6.426 −3.866 −2.560 6.426 3.866 1.280 −5.146 5.146 10.345 6.4041 9 −6.499 −4.151 −2.348 6.499 4.151 1.174 −5.325 5.325 12.078 6.7295 10 −6.248 −4.183 −2.065 6.248 4.183 1.032 −5.216 5.216 13.176 7.2985 11 −6.450 −4.162 −2.288 6.450 4.162 1.144 −5.306 5.306 12.306 4.5204 12 −6.450 −4.165 −2.285 6.450 4.165 1.142 −5.307 5.307 12.328 4.6037 13 −6.485 −3.786 −2.699 6.485 3.786 1.349 −5.135 5.135 9.772 7.0345 14 −6.239 −3.816 −2.424 6.239 3.816 1.212 −5.027 5.027 10.428 8.3989 15 −6.428 −3.792 −2.637 6.428 3.792 1.318 −5.110 5.110 9.903 5.0970 16 −6.442 −3.793 −2.649 6.442 3.793 1.325 −5.117 5.117 9.885 5.6097

For all compounds 1–16, the LUMO orbitals are mainly delocalized at the 1,4-quinone moiety. Localization of HOMO orbitals depends on the type of substituent at the C-3 position of the betulin moiety. The HOMO orbitals of hybrids with hydroxy and acyl groups (1, 3–5, 7–9, 11–13, and 15–16) at the C-3 position are mainly delocalized at the isopropenyl group and A ring of betulin. For hybrids with the oxo group (2, 6, 10, and 14), these orbitals are delocalized at the D ring and oxo group (Figure S4). The molecular properties influence the reactivity and stability of molecules. The high value of EHOMO and low value of ELUMO show that the compounds are highly reactive with nucleophilic molecules. The energy gap ∆E depends on the van der Waals interaction between molecules and its value could be correlated with the biological activity of the compounds [62,63]. The hybrids 1, 5, 9, and 13 are characterized by the lowest chemical hardness and the highest negative value of chemical potential, which means they are comparatively soft with high polarizability compared with other compounds in this series. The arrangement of nucleophilic and electrophilic regions of a molecule influences its interaction with a biological target through hydrogen bonds and hydrophobic interactions. The distribution of the positive and negative charges of molecule is described by the molecular electrostatic potential map (MEP) [64,65]. The different charges are repre- sented by different colors, which means that the red, blue, and green areas are negative, positive, or neutral, respectively [66]. The MEP’s are sketched for an order of (−47,935) to 47,935 kcal/mol. The maps for hybrids 1, 5, 9, and 13 present in Figure6, while for the rest of the compounds, they are shown in Figure S5. For all hybrids 1–16, the nucleophilic regions (red color) are localized in four main area. The first and second areas are localized on 1,4-quinone moiety. The first area contains the carbonyl atom at C-5 and the second area the carbonyl group at C-8, and the nitrogen atom. The third area includes the triazole linker, and the fourth, the substituent at the C-3 position of the betulin moiety. The electrophilic regions (blue color) are localized near the methine group at 1,4-quinone moiety and the methylene group at the triazole linker, while the betulin regions are neutral (Figure6 and Figure S5). Pharmaceutics 2021, 13, x 13 of 22

Table 7. Electrostatic descriptor of hybrids 1–16.

EHOMO ELUMO ΔE A η µ ω Hybrid I (kcal/mol) χ (kcal/mol) DM (D) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) (kcal/mol) 1 −6.227 −3.918 −2.310 6.227 3.918 1.155 −5.073 5.073 11.140 5.6507 2 −6.510 −3.989 −2.521 6.510 3.989 1.261 −5.250 5.250 10.930 9.563 3 −6.423 −3.980 −2.443 6.423 3.980 1.222 −5.202 5.202 11.076 6.3035 4 −6.422 −3.984 −2.438 6.422 3.984 1.219 −5.203 5.203 11.106 6.5282 5 −6.503 −3.816 −2.687 6.503 3.816 1.344 −5.159 5.159 9.906 5.8305 6 −6.228 −3.869 −2.360 6.228 3.869 1.180 −5.049 5.049 10.801 8.9646 7 −6.421 −3.867 −2.554 6.421 3.867 1.277 −5.144 5.144 10.359 6.548 8 −6.426 −3.866 −2.560 6.426 3.866 1.280 −5.146 5.146 10.345 6.4041 9 −6.499 −4.151 −2.348 6.499 4.151 1.174 −5.325 5.325 12.078 6.7295 10 −6.248 −4.183 −2.065 6.248 4.183 1.032 −5.216 5.216 13.176 7.2985 11 −6.450 −4.162 −2.288 6.450 4.162 1.144 −5.306 5.306 12.306 4.5204 12 −6.450 −4.165 −2.285 6.450 4.165 1.142 −5.307 5.307 12.328 4.6037 13 −6.485 −3.786 −2.699 6.485 3.786 1.349 −5.135 5.135 9.772 7.0345 14 −6.239 −3.816 −2.424 6.239 3.816 1.212 −5.027 5.027 10.428 8.3989 15 −6.428 −3.792 −2.637 6.428 3.792 1.318 −5.110 5.110 9.903 5.0970 16 −6.442 −3.793 −2.649 6.442 3.793 1.325 −5.117 5.117 9.885 5.6097

For all compounds 1–16, the LUMO orbitals are mainly delocalized at the 1,4-quinone moiety. Localization of HOMO orbitals depends on the type of substituent at the C-3 position of the betulin moiety. The HOMO orbitals of hybrids with hydroxy and acyl groups (1, 3–5, 7–9, 11–13, and 15–16) at the C-3 position are mainly delocalized at the isopropenyl group and A ring of betulin. For hybrids with the oxo group (2, 6, 10, and 14), these orbitals are delocalized at the D ring and oxo group (Figure S4). The molecular properties influence the reactivity and stability of molecules. The high value of EHOMO and low value of ELUMO show that the compounds are highly reactive with nucleophilic molecules. The energy gap ΔE depends on the van der Waals interaction between molecules and its value could be correlated with the biological activity of the compounds [62,63]. The hybrids 1, 5, 9, and 13 are characterized by the lowest chemical hardness and the highest negative value of chemical potential, which means they are comparatively soft with high polarizability compared with other compounds in this series. The arrangement of nucleophilic and electrophilic regions of a molecule influences its interaction with a biological target through hydrogen bonds and hydrophobic interactions. The distribution of the positive and negative charges of molecule is described by the molecular electrostatic potential map (MEP) [64,65]. The different charges are repre- Pharmaceutics 2021, 13, x sented by different colors, which means that the red, blue, and green areas are14 negative,of 22

positive, or neutral, respectively [66]. The MEP’s are sketched for an order of (−47,935) to Pharmaceutics 2021, 13, 781 13 of 21 47,935 kcal/mol. The maps for hybrids 1, 5, 9, and 13 present in Figure 6, while for the rest of the compounds, they are shown in Figure S5.

Pharmaceutics 2021, 13, x 14 of 22

For all hybrids 1–16, the nucleophilic regions (red color) are localized in four main area. The first and second areas are localized on 1,4-quinone moiety. The first area contains the carbonyl atom at C-5 and the second area the carbonyl group at C-8, and the nitrogen atom. The third area includes the triazole linker, and the fourth, the substituent

at the C-3 position of the betulin moiety. The electrophilic regions (blue color) are localized near the methine group at 1,4-quinone moiety and the methylene group at the triazole linker, while the betulin regions are neutral (Figures 6 and S5).

In each area, the local minima have been determined for 1–16, and they are collected in Table(c )S4. The analysis of the local minima in the first, third,(d )and fourth areas show that the arrangement of charges does not depend on the type of 1,4-quinone moiety. However, Figure 6.in Molecular the second electrostatic area, the potentialamount plottedof potential for hybrids: minima (a) 1depends,(, (b) 5,(, (c) on9, and the ( dtype) 13.. of 1,4-quinone moiety, i.e., the hybrids with 5,8-quinolinedione or 5,8-isoquinolinedione moiety have In each area, the local minima have been determined for 1–16, and they are collected two potentialFor all minima, hybrids but 1–16 compounds, the nucleophilic with 1,4-naphthoquinone regions (red color) moietyare localized have onlyin four one main in Table S4. The analysis of the local minima in the ﬁrst, third, and fourth areas show that potentialarea. Theminima first in and this second area (Table areas S4).are localized on 1,4-quinone moiety. The first area con- tainsthe arrangement the carbonyl of atom charges at C-5 does and not the depend second on thearea type the ofcarbonyl 1,4-quinone group moiety. at C-8, However, and the in the second area, the amount of potential minima depends on the type of 1,4-quinone 3.4. Molecularnitrogen atom. Docking The Study third area includes the triazole linker, and the fourth, the substituent atmoiety, the C-3 i.e., position the hybrids of the betulin with 5,8-quinolinedione moiety. The electrophilic or 5,8-isoquinolinedione regions (blue color) moietyare localized have Since the end of 2019, coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2, neartwo potentialthe methine minima, group but at compounds1,4-quinone withmoiety 1,4-naphthoquinone and the methylene moiety group haveat the only triazole one has infected many people around the word. The World Health Organization (WHO) esti- linker,potential while minima the betulin in this regions area (Table are S4).neutral (Figures 6 and S5). mated that up to April 2021 more than 142 million people tested positive for COVID-19 In each area, the local minima have been determined for 1–16, and they are collected and 3.4.more Molecular than 3 million Docking people Study died from this virus [67]. At the beginning of the pan- in Table S4. The analysis of the local minima in the first, third, and fourth areas show that demic, attempts were made to treat with chloroquine, but this treatment was halted and the arrangementSince the end of ofcharges 2019, coronavirusdoes not depend disease on the 2019 type (COVID-19), of 1,4-quinone caused moiety. by SARS-CoV- However, now chloroquine is not recommended (Figure 6) [68]. Contemporary reports indicate the in2, hasthe second infected area, many the people amount around of potential the word. minima The depends World Health on the Organization type of 1,4-quinone (WHO) effective use of amantadine and remdesivir in the treatment of patients with SARS-CoV- moiety,estimated i.e., that the up hybrids to April with 2021 5,8-quinolinedione more than 142 million or 5,8-isoquinolinedione people tested positive moiety for COVID- have 2 (Figure 7) [69,70]. two19 and potential more thanminima, 3 million but compounds people died with from 1,4-naphthoquinone this virus [67]. At moiety the beginning have only of one the Continuing our research on molecular docking to SARS-CoV-2 targets, we examined potentialpandemic, minima attempts in werethis area made (Table to treat S4). with chloroquine, but this treatment was halted and the interactionnow chloroquine between is nothybrids recommended 1–16 and Mpro (Figure and6)[ PLpro68]. Contemporary proteins using reports the AutoDock indicate the Vina program (referred to as Vina). As reference ligands, chloroquine, amantadine, and 3.4.effective Molecular use of Docking amantadine Study and remdesivir in the treatment of patients with SARS-CoV-2 remdesivir(Figure 7were)[69 ,used70]. (Figure 7). Since the end of 2019, coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2, has infected many people around the word. The World Health Organization (WHO) estimated that up to April 2021 more than 142 million people tested positive for COVID-19 NH2 and more than 3 million people died from this virusO [67]. At the beginning of the pan- N O HN demic, attempts were madeN to treat with chloroquine, but this treatment was halted and N O P O now chloroquine is not recommendedO (FigureO 6) [68]. Contemporary reports indicate the effective use of amantadineN and remdesivir in the treatment of patients with SARS-CoV- 2 (Figure 7) [69,70]. HO OH remdesivir Continuing our research on molecular docking to SARS-CoV-2 targets, we examined FiguretheFigure 7. interaction Chemical 7. Chemical structure between structure of hybridsdrugs of drugs used 1–16 used in COVID-19and inCOVID-19 Mpro treatment. and treatment. PLpro proteins using the AutoDock Vina program (referred to as Vina). As reference ligands, chloroquine, amantadine, and remdesivirContinuing were ourused research (Figure on7). molecular docking to SARS-CoV-2 targets, we examined the interaction between hybrids 1–16 and Mpro and PLpro proteins using the AutoDock Vina program (referred to as Vina). As reference ligands, chloroquine, amantadine, and remdesivir were used (Figure7). NH2 O O N HN N N O P O O O N HO OH remdesivir Figure 7. Chemical structure of drugs used in COVID-19 treatment.

Pharmaceutics 2021, 13, 781 14 of 21

The results obtained with the use of the Vina program are presented in Table8. The lower the ∆G energy, the better the afﬁnity of the tested ligands for the receptor. Calcu- lations of the Ki from the binding energy of the pose generated by Vina were performed using Equations (6) and (7).

∆G kcal = lnK (for T = 298 K and R = 1.987 [ ]), (6) R·T i K·mol ∆G K = exp . (7) i R·T

Table 8. Vina afﬁnity scoring values (∆G) (kcal/mol) and pKI for tested compounds.

Mpro PLpro Mpro PLpro Compound Compound ∆G pKI ∆G pKI ∆G pKI ∆G pKI 1 −8.8 6.46 −8.3 6.09 11 −9.3 6.82 −6.4 4.70 2 −8.9 6.53 −6.9 5.06 12 −9.1 6.68 −7.0 5.14 3 −8.6 6.31 −6.5 4.77 13 −8.6 6.31 −7.6 5.58 4 −7.8 5.72 −6.5 4.77 14 −8.6 6.31 −8.2 6.02 5 −8.8 6.46 −7.9 5.80 15 −8.6 6.31 −6.4 4.70 6 −8.9 6.53 −8.1 5.94 16 −8.4 6.16 −7.0 5.14 7 −8.0 5.87 −6.9 5.06 Chloroquine −5.7 4.18 −5.2 3.82 8 −8.5 6.24 −6.4 4.70 Remdesivir −7.4 5.43 −5.7 4.18 9 −9.1 6.68 −8.1 5.94 Amantadine −4.5 3.30 −4.1 3.01 10 −9.2 6.75 −6.8 4.99

Based on the obtained results, it can be concluded that the tested derivatives show docking scores in the range of −9.3 to −7.8 and −8.3 to −6.4 for Mpro and PLpro, respectively. The obtained ∆G are lower compared to the reference compounds (−7.4 to −4.5 and −5.7 to −4.1 for Mpro and PLpro, respectively). This indicates that in preliminary in silico studies, hybrids 1–16 show a higher afﬁnity for the proteins used than the reference drugs. Comparing the score values of hybrids 1–16 with betulin-5,8-quinolinedione derivatives [18] showed that the introduction of triazole linker between betulin and 1,4-quinone increased the ∆G value for Mpro and PL pro proteins. The main protease Mpro, also known as 3CLpro, is one of the coronavirus nonstructural proteins (Nsp5). Inhibition of Mpro would prevent the virus from replication and as a consequence, Mpro is one of the coronavirus proteins designated as potential targets for drug development. According to the crystallographic data, amino acids His41, Met49, and residues 164–168 of the long strand play an important role in stabilizing the ligand-Mpro complexes [71]. As seen in Figure8a, the hybrids are localized in the hydrophobic matrix of the protein, but the arrangement depends on the type of the 1,4-quinone moiety. Comparing the score values for Mpro protein shows that in the series of 5,8-quinolinedione (1–4) and 5,8-isoquinolinedione (5–8), the highest ∆G values exhibit derivatives with oxo group at the C-3 position of betulin moiety. In the group of 2-methyl-5,8-quinolinedione, better docking scores are obtained for hybrids 10 and 11. However, in the series of 1,4-naphthoquinone derivatives (13–16), the group at the C-3 position of the betulin moiety does not inﬂuence the ∆G. Ligands 2, 6, and 14, which have the best score values, contain the same betulin moiety but different 1,4-quinone fragments. In the group of 2-methyl-5,8-quinolinedione hybrids, the lowest ∆G has hybrid 11. For detailed analysis of molecular docking, complexes of the Mpro with 2, 6, 10, 11, and 14 ligands have been chosen. Complete models of the possible interaction in 3D and 2D views are presented in Figure9a,e and Figure S6A–E. The detailed data about the type and length of the binding interactions between these ligands and the protein residues are summarized in Table S5. Pharmaceutics 2021, 13, x 16 of 22

Pharmaceutics 2021, 13, x Pharmaceutics 2021, 13, 781 16 of 22 15 of 21

(a) (b) Figure 8. The superposition of docked ligands: 2 (violet), 6 (yellow), 10 (blue), 11 (red), and 14 (black) in the binding site of (a) Mpro and (b) PLpro.

Ligands 2, 6, and 14, which have the best score values, contain the same betulin moiety but different 1,4-quinone fragments. In the group of 2-methyl-5,8-quinolinedione hybrids, the lowest ΔG has hybrid 11. For detailed analysis of molecular docking, complexes of the Mpro with 2, 6, 10, 11, and 14 ligands have been chosen. Complete models of the possible interaction (ina) 3D and 2D views are presented in Figures(b) 9a,e and S6A–E. The FigureFiguredetailed 8. The 8. superposition Thedata superposition about of the docked type of ligands:docked and length2 ligands:(violet), of 62the(yellow), (violet), binding 106 (yellow),(blue), interactions11 (red),10 (blue), and between14 11(black) (red), these inand the ligands 14 binding site of (a) Mpro(black)and and the in (b protein )the PLpro. binding residues site of are (a) Mprosummarized and (b) PLpro. in Table S5.

(a) (b)

Figure 9. Cont.

Pharmaceutics 2021, 13, x 17 of 22 Pharmaceutics 2021, 13, 781 16 of 21

(e)

Figure 9. Docking pose of COVID-19Figure 9. MproDocking protein pose complex of COVID-19 with hybrids Mpro (a protein) 2,(b) 6 ,(complexc) 10,(d )with11, and hybrids (e) 14. (a) 2, (b) 6, (c) 10, (d) 11, and (e) 14. Comparing the arrangement of hybrids 2, 6, 10 in the active site of the protein shows that the positionComparing of nitrogen the arrangement atoms in the 5,8-quinolinedioneof hybrids 2, 6, 10 moiety in the influencesactive site the of the inter- protein shows action between the ligand and the hydrophobic matrix of protein. As seen in Figure9a,c that the position of nitrogen atoms in the 5,8-quinolinedione moiety influences the inter- and Table S5, ligands 2 and 10 create the hydrogen bond and hydrophobic interaction betweenaction 1,4-quinone between moiety the ligand and His41, and the Asn142, hydrophobic Gly143, and ma Cys145.trix of Inprotein. these complexes, As seen in Figure 9a,c the betulinand Table moiety S5, is ligands stabilized 2 byand hydrophobic 10 create the interaction hydrogen with bond Pro168. and Changing hydrophobic the interaction 5,8-quinolinedionebetween 1,4-quinone ring to 5,8-isoquinolinedione moiety and His41, moietyAsn142, causes Gly143, different and Cys145. arrangement In these of complexes, the ligandthe betulin in the hydrophobic moiety is stabilized matrix. In by complex hydrophobic Mpro-hybrid interaction6, the nitrogen with Pro168. atom and Changing tri- the 5,8- azolequinolinedione ring create the hydrogen ring to bond5,8-isoquinolinedione and hydrophobic interaction moiety causes with Pro168 different and Asn142,arrangement of the Glu166,ligand Leu141, in the respectively hydrophobic (Figure matrix.9b and In Figure complex S7B). Mpro-hybrid In this case, betulin 6, theis nitrogen bound to atom and tria- Met49,zole Met165, ring create and His41 the hydrogen by hydrophobic bond and interaction. hydrophobic The 1,4-naphthoquione interaction with Pro168 moiety and Asn142, (hybrid 14) interacts with His41, Cys145, Met165, and Gln189 by hydrophobic interaction. Glu166, Leu141, respectively (Figures 9b and S76B). In this case, betulin is bound to Met49, Moreover, the triazole unit and betulin moiety are stabilized by hydrophobic interaction with Cys145Met165, and and Pro168, His41 respectively by hydrophobic (Figure interactio9e and Figuren. The S6E). 1,4-naphthoquione moiety (hybrid 14) Figureinteracts9d and with Figure His41, S6D Cys145, present theMet165, possible and interaction Gln189 by of thehydrophobic best fitted compound interaction. Moreover, 11 insidethe triazole the binding unit pocket and betulin of Mpro. moiety Corresponding are stabilized amino by acidshydrophobic that are significantly interaction with Cys145 involvedand in Pro168, the hydrophobic respectively interactions (Figures between 9e and ligand S6E). and the betulin unit are as follows: His41, Met49,Figures and Met165.9d and S6D Moreover, present the the whole possible complex interaction is additionally of the stabilizedbest fitted by compound 11 the hydrophobicinside the binding interactions pocket of the of newly Mpro. introduced Corresponding substituents: amino the acids triazole that unit are with significantly in- Glu166volved and the in 1,4-quinonethe hydrophobic moiety interactions with Leu167 between and Pro168. ligand and the betulin unit are as follows: It should be emphasized that the 1,4-quinone moiety plays an important role in the His41, Met49, and Met165. Moreover, the whole complex is additionally stabilized by the stabilization of Mpro hybrid complexes. Onehydrophobic of the attractive interactions antiviral of drug the targets newly is introd the SARS-CoV-2uced substituents: papain-like the protease triazole unit with PLproGlu166 which isand responsible the 1,4-quinone for processing moiety three with cleavage Leu167 sites and of Pro168. the viral polyprotein to release matureIt should nonstructural be emphasized proteins 1, that 2, and the 3. 1,4-quinone In the case of moiety PLpro, theplays drug an molecules important role in the bind tostabilization S3/S4 domains. of Mpro The S3/S4hybrid pocket complexes. contains the residues Asp164, Val165, Arg166, Glu167, Met208,One of Ala246, the attractive Pro247, Pro248, antiviral Tyr264, drug Gly266, target Asn267,s is the Tyr268,SARS-CoV-2 Gln269, papain-like Cys217, protease Gly271,PLpro Tyr273, which Thr301, is responsible and Asp302 for [72 ].processing three cleavage sites of the viral polyprotein to Asrelease seenin mature Figure 8nonstructuralb, the hybrids are proteins localized 1, in2, theand hydrophobic 3. In the case matrix of PLpro, of the protein, the drug molecules but thebind type to of S3/S4 1,4-quinone domains. slightly The influences S3/S4 pocket the arrangement contains the of theresidues hybrid Asp164, in the pocket Val165, Arg166, site of proteins. Comparing the score values for PLpro protein shows that in the group Glu167, Met208, Ala246, Pro247, Pro248, Tyr264, Gly266, Asn267, Tyr268, Gln269, Cys217, of 5,8-quinolinedione hybrids (1–4 and 9–12) the highest values have derivatives with hydroxyGly271, group Tyr273, at the C-3 Thr301, position and of Asp302 betulin moiety,[72]. while in the series of 5–8 and 13–16 the best resultsAs seen are obtainedin Figure for 8b, hybrids the hybrids with the are oxo localized group at in this the position. hydrophobic matrix of the protein, but the type of 1,4-quinone slightly influences the arrangement of the hybrid in the pocket site of proteins. Comparing the score values for PLpro protein shows that in the group of 5,8-quinolinedione hybrids (1–4 and 9–12) the highest values have derivatives with hydroxy group at the C-3 position of betulin moiety, while in the series of 5–8 and 13–16 the best results are obtained for hybrids with the oxo group at this position.

Pharmaceutics 2021, 13, x 18 of 22

Pharmaceutics 2021, 13, 781 17 of 21

In each group of 1,4-quinone, we chose one of the best ΔG, and performed detailed analysis for it. CompleteIn each models group of of the 1,4-quinone, possible interaction we chose one in of 3D the and best 2D∆G, views and performed are pre- detailed sented in Figuresanalysis 10a–d and for it.S7A–D. Complete The modelsdetailed of data the possibleabout the interaction type and in length 3D and of 2Dthe views are presented in Figure 10a–d and Figure S7A–D. The detailed data about the type and length binding interactions between these ligands and the protein residues are summarized in of the binding interactions between these ligands and the protein residues are summarized Table S5. in Table S5.

(a) (b)

FigureFigure 10. 10. DockingDocking pose pose of COVID-19 PLpro PLpro protein protein complex complex with with hybrids hybrids (a) 1,( (ab)) 16,(, (cb))9 6,, and (c) (9d, )and14. (d) 14. Comparing the optimal docking post of compounds 1, 6, and 9 shows a general Comparing thetrend optimal that the docking betulin moietypost of iscompounds bound to Pro248 1, 6, and and 9 Lue162 shows by a general hydrophobic trend interaction (Figure 10a–c and Figure S7A–C). The hydroxyl or carbonyl group at the C-3 position that the betulin moiety is bound to Pro248 and Lue162 by hydrophobic interaction (Fig- creates a hydrogen bond with Thr301 (1) or Arg166 (6 and 9). The 1,4-quinone moiety and ures 10a–c and S7A–C).triazole The ring hydr are involvedoxyl or incarbonyl hydrophobic group interaction at the C-3 with position Pro248 andcreates Tyr268, a hy- respectively. drogen bond with Thr301Hybrid (1 with) or Arg166 1,4-naphtoquinone (6 and 9). 14Thehas 1,4-quinone a differentarrangement moiety and intriazole the active site of ring are involvedPLpro in hydrophobic than these with interaction 5,8-quinolinedione with Pro248 or5,8-isoquinolinedione and Tyr268, respectively. ligands (Figure 10d and Hybrid withFigure 1,4-naphtoquinone S7D). The 1,4-quinone 14 has moietya different creates arrangement two hydrogen in the bonds active with site Lys157 of and one PLpro than thesewith with Gly163. 5,8-quinolinedione The complex is or stabilized 5,8-isoquinolinedione by hydrophobic ligands interaction (Figures with Leu162, 10d Pro248, and S7D). The 1,4-quinoneMet208, and moiety Tyr264. creates In this tw case,o hydrogen the triazole bonds ring does with not Lys157 interact and with one the with hydrophobic Gly163. The complexmatrix is of enzyme.stabilized by hydrophobic interaction with Leu162, Pro248, As indicated by literature date, the molecular, physicochemical, and pharmacokinetic Met208, and Tyr264. In this case, the triazole ring does not interact with the hydrophobic properties can be used to determine the biological activity of the compounds [73,74]. The matrix of enzyme.MLR analysis allows to ﬁnd correlation between score values (∆G), experimental lipophilic- As indicatedity by and literature in silico date, determined the mole parameters.cular, physicochemica The MLR equationl, and (8) pharmacoki- shows the correlation netic properties canbetween be used the to score determine value for the Mpro, biological experimental activity lipophilicity of the compounds (logPTLC ),[73,74]. number of rotat- The MLR analysis allows to find correlation between score values (∆G), experimental lipophilicity and in silico determined parameters. The MLR equation (8) shows the correlation between the score value for Mpro, experimental lipophilicity (logPTLC), number of

Pharmaceutics 2021, 13, 781 18 of 21

able bonds (RB), and the energy of LUMO (ELUMO) orbitals. The ∆G for PLpro correlates with molecular mass (M), number of acceptors of hydrogen bond (HA), and energy of HOMO (EHOMO) orbitals (Equation (9)).

∆G = −0.65 logP + 1.04 RB + 0.624 E − 3.052 Mpro TLC LUMO (8) (r = 0.788, r2 = 0.621, SD = 0.281, F = 6.559, p = 0.007)

∆G = 0.280 M − 0.57 HA + 0.38 E − 30.306 PLpro HOMO (9) (r = 0.844, r2 = 0.712, SD = 0.135, F = 9.894, p = 0.001) The obtained results show that the physicochemical and molecular properties of tested hybrids inﬂuence their interaction with the SARS-CoV-2 proteins.

4. Conclusions In the presented study, the lipophilicity of betulin triazole derivatives with attached 1,4- quinone moiety was determined and analyzed in terms of structure, physicochemical, and pharmacokinetic parameters. Comparing the lipophilicity of betulin triazole derivatives and hybrids with attached 1,4-quinone, the introduction of 1,4-quinone moiety reduces the lipophilicity. The cluster analysis showed a correlation between the molecular structure of hybrids and their experimental and calculated lipophilicity that can be used to predict these values when designing new compounds. The bioavailability of the tested compound was described by pharmacokinetic and physicochemical properties using the Lipinski and Veber rules. The obtained in silico parameters showed that most of the hybrids could be applied orally and they did not exhibit the neurotoxic activity. For compounds with attached 1,4-quinone, the LUMO orbitals were mainly delocalized at the 1,4-quinone moiety. Localization of HOMO orbitals depended on the type of substituent at the C-3 position of the betulin moiety. The molecular properties depended on the type of 1,4-quinone moiety. The molecular electrostatic potential showed that the negative potential sites are present in the nucleophilic atoms, i.e., nitrogen and oxygen atoms. The arrangement of the charge depends on the type of 1,4-quinone moiety. The molecular docking study showed that betulin triazole derivatives with attached 1,4-quinone can inhibit selected SARS-CoV-2 proteins. The high afﬁnity score values of Mpro and PLpro proteins resulted from the introduction of the triazole ring as a linker connecting the betulin moiety with the 1,4-quinone fragment. The multilinear regression equation showed the correlation between the score values for Mpro and PLpro proteins, and experimental lipophilicity (logPTLC), molecular descriptors, and global properties. The study showed that the determination of these parameters allows for the prediction of the interactions between the ligand and SARS-CoV-2 proteins.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/pharmaceutics13060781/s1, Table S1. The experimental of lipophilicity (logPTLC) and hydrophobic index (ϕ0) for compounds 1–20, Table S2. The calculated lipophilicity for compounds 1–20, Table S3. The experimental and predicted logP, Table S4. The local minima of hybrids 1–16, Table S5. Interaction of hybrids 1, 2, 6, 9, 10, 11, and 14 with active site of Mpro and PL protein, Figure S1. The optimized structure of hybrids 1–16, Figure S2. The linear regression between the experimental and literature lipophilicity for standard substance, Figure S3. The similarity analysis of pharmacokinetic parameters of hybrids 1–16, Figure S4. The HOMO and LUMO orbitals for hybrids 1–16, Figure S5. The MEP for hybrids 1–16, Figure S6. Docking pose of COVID-19 Mpro protein complex with hybrids 2 (A), 6 (B), 10 (C), 11 (D), and 14 (E), Figure S7. Docking pose of COVID-19 PLpro protein complex with hybrids 1 (A), 6 (B), 9 (C), and 14 (D). Author Contributions: M.K.-T. and S.B. developed the concept of this work; M.J., E.B., and E.C. par- ticipated in evaluating the experimental and theoretical lipophilicity; K.M. developed the molecular docking study. M.K.-T. wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Pharmaceutics 2021, 13, 781 19 of 21

Funding: This work was supported by the Medical University of Silesia in Katowice, Poland. Grant no. PCN-2-006/K/0/F. Calculations were carried out using resources provided by Wroclaw Centre for 818 Networking and Supercomputing (http://wcss.pl), grant no. 382. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References 1. Arnott, J.; Planey, S. The influence of lipophilicity in drug discovery and design. Expert Opin. Drug Discov. 2021, 7, 863–875. [CrossRef] 2. Hammoudi, N.-E.-H.; Sobhi, W.; Attoui, A.; Lemaoui, T.; Erto, A.; Benguerba, Y. In silico drug discovery of Acetylcholinesterase and Butyrylcholinesterase enzymes inhibitors based on Quantitative Structure-Activity Relationship (QSAR) and drug-likeness evaluation. J. Mol. Struct. 2021, 1229, 129845. [CrossRef] 3. Mougin, J.; Bourgaux, C.; Couvreur, P. Elongated self-assembled nanocarriers: From molecular organization to therapeutic applications. Adv. Drug Deliv. Rev. 2021, 172, 127–147. [CrossRef][PubMed] 4. Duggal, D. Role of nanotechnology in new drug delivery systems. Int. J. Drug Dev. Res. 2011, 4, 4–8. 5. Kovaˇcević,S.; Banjac, M.; Milošević,N.; Curˇcić,J.;´ Marjanović,D.; Todorović,N.; Krmar, J.; Podunavac-Kuzmanović,S.; Banjac, N.; Ušćumlić,G. Comparative chemometric and quantitative structure-retention relationship analysis of aniso-tropic lipophilicity of 1-arylsuccinimide derivatives determined in high-performance thin-layer chromatography system with aprotic solvents. J. Chromatogr. A 2020, 1628, 461439. [CrossRef][PubMed] 6. Tshepelevitsh, S.; Kadam, S.A.; Darnell, A.; Bobacka, J.; Rüütel, A.; Haljasorg, T.; Leito, I. LogP determination for highly lipophilic hydrogen-bonding anion receptor molecules. Anal. Chim. Acta 2020, 1132, 123–133. [CrossRef][PubMed] 7. Kulig, K.; Malawska, B. RP-TLC determination of the lipophilicity of 1-substituted pyrrolidin-2-one derivatives. Correlation of lipophilicity with affinity for α-adrenoceptors. JPC J. Planar Chromatogr. Mod. TLC 2009, 22, 141–144. [CrossRef] 8. Bolzán, A.D.; Bianchi, M.S. Genotoxicity of streptonigrin: A review. Mutat. Res. Rev. Mutat. Res. 2001, 488, 25–37. [CrossRef] 9. Boger, D.L.; Yasuda, M.; Mitscher, L.A.; Drake, S.D.; Kitos, P.A.; Thompson, S.C. Streptonigrin and lavendamycin partial structures. Probes for the minimum, potent pharmacophore of streptonigrin, lavendamycin, and synthetic quinoline-5,8-diones. J. Med. Chem. 1987, 30, 1918–1928. [CrossRef] 10. Kadela-Tomanek, M.; B˛ebenek,E.; Chrobak, E.; Boryczka, S. 5,8-Quinolinedione Scaffold as a Promising Moiety of Bioactive Agents. Molecules 2019, 24, 4115. [CrossRef][PubMed] 11. Kadela-Tomanek, M.; Jastrz˛ebska,M.; Chrobak, E.; B˛ebenek,E.; Latocha, M.; Kusz, J.; Boryczka, S. Structural and spectral characterisation of 2-amino-2H-[1,2,3]triazolo[4,5-g]quinoline-4,9-dione polymorphs. Cytotoxic activity and molecu-lardocking study with NQO1 enzyme. Spectrochim. Acta A Mol. Biomol. Spectrosc. 2020, 230, 118038–118050. [CrossRef][PubMed] 12. Kadela-Tomanek, M.; B˛ebenek, E.; Chrobak, E.; Latocha, M.; Boryczka, S. Alkoxy and enediyne eerivatives containing 1,4- benzoquinone subunits-synthesis and antitumor activity. Molecules 2017, 22, 447. [CrossRef][PubMed] 13. Kadela-Tomanek, M.; Jastrz˛ebska,M.; B˛ebenek,E.; Chrobak, E.; Latocha, M.; Kusz, J.; Tarnawska, D.; Boryczka, S. New Acetylenic Amine Derivatives of 5,8-Quinolinediones: Synthesis, Crystal Structure and Antiproliferative Activity. Crystals 2017, 7, 15. [CrossRef] 14. Kadela-Tomanek, M.; B˛ebenek, E.; Chrobak, E.; Marciniec, K.; Latocha, M.; Ku´smierz,D.; Jastrz˛ebska,M.; Boryczka, S. Betulin-1,4- quinone hybrids: Synthesis, anticancer activity and molecular docking study with NQO1 enzyme. Eur. J. Med. Chem. 2019, 177, 302–315. [CrossRef] 15. Bayrak, N. Novel azanaphtoquinone compounds with aromatic amino moiety: Synthesis, structural characterization, and antimicrobial features. J. Mol. Struct. 2019, 1195, 411–416. [CrossRef] 16. Gholampour, M.; Sakhteman, A.; Pirhadi, S.; Seradj, H. In silico design of novel diamino-quinoline-5,8-dione derivatives as putative inhibitors of NAD(P)H:Quinone oxidoreductase 1 based on docking studies and molecular dynamics simulations. J. Mol. Struct. 2021, 1230, 129906. [CrossRef] 17. Ling, Y.; Yang, Q.-X.; Teng, Y.-N.; Chen, S.; Gao, W.-J.; Guo, J.; Hsu, P.-L.; Liu, Y.; Morris-Natschke, S.L.; Hung, C.-C.; et al. Development of novel amino-quinoline-5,8-dione derivatives as NAD(P)H:quinone oxidoreductase 1 (NQO1) inhibitors with potent antiproliferative activities. Eur. J. Med. Chem. 2018, 154, 199–209. [CrossRef][PubMed] 18. Kadela-Tomanek, M.; Marciniec, K.; Jastrz˛ebska,M.; B˛ebenek,E.; Chrobak, E.; Boryczka, S. Spectroscopic investigations, computational analysis and molecular docking to SARS-Co-2 targets studies of 5,8-quinolinedione attached to betulin de-rivatives. Crystals 2021, 11, 76. [CrossRef] 19. Kadela-Tomanek, M.; Pawełczak, B.; Jastrzebska, M.; B˛ebenek,E.; Chrobak, E.; Latocha, M.; Kusz, J.; Ksi ˛azek,˙ M.; Boryczka, S. Structural, vibrational and quantum chemical investigations for 6,7-dichloro-2-methyl-5,8-quinolinedione. Cytotoxic and molecular docking studies. J. Mol. Struct. 2018, 1168, 73–83. [CrossRef] Pharmaceutics 2021, 13, 781 20 of 21

20. Kadela-Tomanek, M.; Bober, K.; B˛ebenek,E.; Chrobak, E.; Boryczka, S. Application of thin-layer chromatography to evaluate the lipophilicity of 5,8-quinolinedione compounds. J. Planar Chromatogr. Mod. TLC 2017, 30, 219–224. [CrossRef] 21. Kadela-Tomanek, M.; Jastrz˛ebska,M.; Chrobak, E.; B˛ebenek,E.; Boryczka, S. Chromatographic and Computational Screening of Lipophilicity and Pharmacokinetics of Newly Synthesized Betulin-1,4-quinone Hybrids. Processes 2021, 9, 376. [CrossRef] 22. Ezeokonkwo, M.A.; Ibeanu, F.; Eze, C.; Ibezim, A.; Ezeokoye, C.; Ezenwa, O.; Ezeoka, T.; Vincent, O. Synthesis, antimi-crobial activity and molecular docking studies of 7-bromoquinoline-5,8-dione containing aryl sulphonamides. Int. J. Appl. Chem. 2019, 15, 99–112. 23. Musioł, R.; Jampiledek, J.; Podeszwa, B.; Finster, J.; Tabak, D.; Dohnal, J.; Polański,J. RP-HPLC determination of lipo-philicity in series of quinoline derivatives. Cent. Eur. J. Chem. 2009, 7, 586–597. 24. Atia, A.; Alrawaiq, N.; Abdullah, A. A review of NAD(P)H: Quinone oxidoreductase 1 (NQO1); a multifunctional anti-oxidant enzyme. J. Appl. Pharm. Sci. 2014, 4, 118–122. 25. Valdés, K.; Morales, J.; Rodríguez, L.; Günther, G. Potential use of nanocarriers with pentacyclic triterpenes in cancer treatments. Nanomedicine 2016, 11, 3139–3156. [CrossRef][PubMed] 26. Mukherjee, B.; Al Hoque, A.; Dutta, D.; Paul, B.; Mukherjee, A.; Mallick, S. Nanoformulated Drug Delivery of Potential Betulinic Acid Derivatives: A Promising Approach Toward Cancer Therapy. In Nanomedicine for Bioactives; Springer: New York, NY, USA, 2020; pp. 127–153. 27. Kazakova, O.B.; Smirnova, I.E.; Baltina, L.A.; Boreko, E.I.; Savinova, O.V.; Pokrovskii, A.G. Antiviral Activity of Acyl Derivatives of Betulin and Betulinic and Dihydroquinopimaric Acids. Russ. J. Bioorgan. Chem. 2018, 44, 740–744. [CrossRef] 28. Zhao, Y.; Chen, C.-H.; Morris-Natschke, S.L.; Lee, K.-H. Design, synthesis, and structure activity relationship analysis of new betulinic acid derivatives as potent HIV inhibitors. Eur. J. Med. Chem. 2021, 215, 113287. [CrossRef][PubMed] 29. Karagöz, A.Ç.; Leidenberger, M.; Hahn, F.; Hampel, F.; Friedrich, O.; Marschall, M.; Kappes, B.; Tsogoeva, S.B. Synthesis of new betulinic acid/betulin-derived dimers and hybrids with potent antimalarial and antiviral activities. Bioorgan. Med. Chem. 2019, 27, 110–115. [CrossRef][PubMed] 30. B˛ebenek,E.; Jastrz˛ebska,M.; Kadela-Tomanek, M.; Chrobak, E.; Orzechowska, B.; Zwolińska,K.; Latocha, M.; Mertas, A.; Czuba, Z.; Boryczka, S. Novel Triazole Hybrids of Betulin: Synthesis and Biological Activity Profile. Molecules 2017, 22, 1876. [CrossRef][PubMed] 31. Amiri, S.; Dastghaib, S.; Ahmadi, M.; Mehrbod, P.; Khadem, F.; Behrouj, H.; Aghanoori, M.; Machaj, F.; Ghamsari, M.; Rosik, J.; et al. Betulin and its de-rivatives as novel compounds with different pharmacological effects. Biotechnol. Adv. 2020, 38, 107409. [CrossRef] 32. Zhang, D.-H.; Wu, K.-L.; Zhang, X.; Deng, S.-Q.; Peng, B. In silico screening of Chinese herbal medicines with the potential to directly inhibit 2019 novel coronavirus. J. Integr. Med. 2020, 18, 152–158. [CrossRef] 33. Wen, C.-C.; Kuo, Y.-H.; Jan, J.-T.; Liang, P.-H.; Wang, S.-Y.; Liu, H.-G.; Lee, C.-K.; Chang, S.-T.; Kuo, C.-J.; Lee, S.-S.; et al. Specific Plant Terpenoids and Lignoids Possess Potent Antiviral Activities against Severe Acute Respiratory Syndrome Coronavirus. J. Med. Chem. 2007, 50, 4087–4095. [CrossRef][PubMed] 34. Kadioglu, O.; Saeed, M.; Greten, H.J.; Efferth, T. Identification of novel compounds against three targets of SARS CoV-2 coronavirus by combined virtual screening and supervised machine learning. Comput. Biol. Med. 2021, 133, 104359. [CrossRef] [PubMed] 35. Marciniec, K.; Chrobak, E.; D ˛abrowska, A.; B˛ebenek,E.; Kadela-Tomanek, M.; P˛ecak,P.; Boryczka, S. Phosphate deriva-tives of 3-carboxyacylbetulin: Synthesis, in vitro anti-HIV and molecular docking study. Biomoleclues 2020, 10, 1148. [CrossRef] 36. Kadela-Tomanek, M.; Jastrz˛ebska,M.; Marciniec, K.; Chrobak, E.; B˛ebenek,E.; Latocha, M.; Ku´smierz,D.; Boryczka, S. Design, synthesis and biological activity of 1,4-quinone moiety attached to betulin derivatives as potent DT-diaphorase substrate. Bioorganic Chem. 2021, 106, 104478. [CrossRef][PubMed] 37. B˛ebenek,E.; Bober-Majnusz, K.; Siudak, S.; Chrobak, E.; Kadela-Tomanek, M.; Wietrzyk, J.; Boryczka, S. Application of TLC to Evaluate the Lipophilicity of Newly Synthesized Betulin Derivatives. J. Chromatogr. Sci. 2020, 58, 323–333. [CrossRef] [PubMed] 38. Virtual Computational Chemistry Laboratory. Available online: http://www.vcclab.org/lab/alogps/ (accessed on 24 April 2021). 39. Molinspiration. Available online: https://www.molinspiration.com/services/logp.html (accessed on 24 April 2021). 40. SwissADME. Available online: http://www.swissadme.ch/index.php (accessed on 24 April 2021). 41. Daina, A.; Michielin, O.; Zoete, V. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci. Rep. 2017, 7, 42717. [CrossRef][PubMed] 42. pkCSM. Available online: http://biosig.unimelb.edu.au/pkcsm/prediction# (accessed on 24 April 2021). 43. Pires, D.E.V.; Blundell, T.L.; Ascher, D.B. pkCSM: Predicting Small-Molecule Pharmacokinetic and Toxicity Properties Using Graph-Based Signatures. J. Med. Chem. 2015, 58, 4066–4072. [CrossRef] 44. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G.A.; et al. Gaussian 16, Revision A. 03. Wallingford 2016, 1, 3. 45. Politzer, P.; Laurence, P.; Jayasuriya, K. Molecular electrostatic potentials: An effective tool for the elucidation of bio-chemical phenomena. Environ. Health Perspect. 1985, 61, 191–202. [CrossRef][PubMed] 46. Dennington, R.; Keith, T.; Millam, J. GaussView Version 5; Semichem Inc.: Shawnee Mission, KS, USA, 2009. 47. PerkinElmer ChemOffice Suite. Available online: https//www.perkinelmer.com (accessed on 24 April 2021). Pharmaceutics 2021, 13, 781 21 of 21

48. Morris, G.M.; Huey, R.; Lindstrom, W.; Sanner, M.F.; Belew, R.K.; Goodsell, D.S.; Olson, A.J. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 2009, 30, 2785–2791. [CrossRef][PubMed] 49. Trott, O.; Olson, A.J. AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 2010, 31, 455–461. [CrossRef][PubMed] 50. Dallakyan, S.; Olson, A.J. Small-molecule library screening by docking with PyRx. Methods Mol. Biol. 2015, 1263, 243–250. [PubMed] 51. Dessault Systemes BIOVIA. Available online: https://www.3dsbiovia.com/products/collaborative-science/biovia-discovery- studio/ (accessed on 24 April 2021). 52. Mannhold, R.; Cruciani, G.; Dross, K.; Rekker, R. Multivariate analysis of experimental and computational descriptors of molecular lipophilicity. J. Comput. Mol. Des. 1998, 12, 573–581. [CrossRef][PubMed] 53. Bodor, N.; Gabanyi, Z.; Wong, C.K. A new method for the estimation of partition coefficient. J. Am. Chem. Soc. 1989, 11, 3783–3786. [CrossRef] 54. Lipinski, C.A.; Lombardo, F.; Dominy, B.W.; Feeney, P.J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 2001, 46, 3–26. [CrossRef] 55. Veber, D.F.; Johnson, S.R.; Cheng, H.-Y.; Smith, B.R.; Ward, K.W.; Kopple, K.D. Molecular Properties That Influence the Oral Bioavailability of Drug Candidates. J. Med. Chem. 2002, 45, 2615–2623. [CrossRef][PubMed] 56. Šegan, S.; Penjišević,J.; Šukalović,V.; Andrić,D.; Milojković-Opsenica,D.; Kostić-Rajaˇcić,S. Investigation of lipophilicity and pharmacokinetic properties of 2-(methoxy)phenylpiperazine dopamine D2 ligands. J. Chromatogr. B 2019, 1124, 146–153. [CrossRef][PubMed] 57. Johnson, T.W.; Dress, K.R.; Edwards, M. Using the Golden Triangle to optimize clearance and oral absorption. Bioorganic Med. Chem. Lett. 2009, 19, 5560–5564. [CrossRef] 58. Kulkarni, A.; Han, Y.; Hopfinger, A. Predicting Caco-2 cell permeation coefficients of organic molecules using mem-brane- interaction QSAR analysis. J. Chem. Inf. Comput. Sci. 2002, 42, 331–342. [CrossRef] 59. Feher, M.; Schmidt, J.M. Property Distributions: Differences between Drugs, Natural Products, and Molecules from Combinatorial Chemistry. J. Chem. Inf. Comput. Sci. 2002, 43, 218–227. [CrossRef][PubMed] 60. Hutter, M.C. Prediction of blood–brain barrier permeation using quantum chemically derived information. J. Comput. Mol. Des. 2003, 17, 415–443. [CrossRef][PubMed] 61. Govindarajan, M.; Karabacak, M.; Periandy, S.; Tanuja, D. Spectroscopic (FT-IR, FT-Raman, UV and NMR) investigation and NLO, HOMO–LUMO, NBO analysis of organic 2,4,5-trichloroaniline. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 2012, 97, 231–245. [CrossRef][PubMed] 62. Kumar, S.; Saini, V.; Maurya, I.K.; Sindhu, J.; Kumari, M.; Kataria, R.; Kumar, V. Design, synthesis, DFT, docking studies and ADME prediction of some new coumarinyl linked pyrazolylthiazoles: Potential standalone or adjuvant antimicrobial agents. PLoS ONE 2018, 13, e0196016. [CrossRef][PubMed] 63. Singha, P.; Islama, S.; Ahmada, H.; Prabaharan, A. Spectroscopic investigation (FT-IR, FT-Raman), HOMO-LUMO, NBO, and molecular docking analysis of N-ethyl-N-nitrosourea, a potential anticancer agent. J. Mol. Struct. 2018, 1154, 39–50. [CrossRef] 64. Zhou, Z.; Liu, Y.; Ren, Q.; Yu, D.; Lu, H. Synthesis, crystal structure and DFT study of a novel compound N-(4-(2,4- dimorpholinopyrido[2,3-d]pyrimidin-6-yl)phenyl)pyrrolidine-1-carboxamide. J. Mol. Struct. 2021, 1235, 130261. [CrossRef] 65. Mary, Y.S.; Mary, Y.S.; Krátký, M.; Vinsova, J.; Baraldi, C.; Gamberin, M.C. DFT, molecular docking and SERS (concen-tration and solvent dependant) investigations of a methylisoxazole derivative with potential antimicrobial activity. J. Mol. Struct. 2021, 1232, 130034. [CrossRef] 66. Chandrasekaran, K.; Kumar, R.T. Structural, spectral, thermodynamical, NLO, HOMO, LUMO and NBO analysis of fluconazole. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 2015, 150, 974–991. [CrossRef] 67. World Health Organization. WHO Coronavirus Disease (COVID-19) Dashboard. 2020. Available online: https://covid19.who.int/ (accessed on 3 March 2021). 68. Oscanoa, T.; Romero-Ortuno, R.; Carvajal, A.; Savarino, A. A pharmacological perspective of chloroquine in SARS-CoV-2 infection: An old drug for the fight against a new coronavirus? Int. J. Antimicrob. Agents 2020, 56, 106078. [CrossRef] 69. Ko, W.; Rolain, J.; Lee, N.; Chen, P.; Huang, C.; Lee, P.; Hsueh, P. Arguments in favour of remdesivir for treating SARS-CoV-2 infections. Int. J. Antimicrob. Agents 2020, 55, 105933. [CrossRef] 70. Borra, A. Does amantadine have a protective effect against COVID-19? Neurol. Neurochir. Pol. 2020, 54, 284–285. [CrossRef] [PubMed] 71. Jin, Z.; Du, X.; Xu, Y.; Deng, Y.; Liu, M.; Zhao, Y.; Zhang, B.; Li, X.; Zhang, L.; Peng, C.; et al. Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors. Nature 2020, 582, 289–293. [CrossRef][PubMed] 72. Arya, R.; Das, A.; Prashar, V.; Kumar, M. Potential inhibitors against papain-like protease of novel coronavirus (SARS-CoV-2) from FDA approved drugs. ChemRxiv 2020, 1–10. [CrossRef] 73. Oashi, T.; Ringer, A.L.; Raman, E.P.; MacKerell, A.D. Automated Selection of Compounds with Physicochemical Properties to Maximize Bioavailability and Druglikeness. J. Chem. Inf. Model. 2010, 51, 148–158. [CrossRef][PubMed] 74. Lagorce, D.; Douguet, D.; Miteva, M.A.; Villoutreix, B.O. Computational analysis of calculated physicochemical and ADMET properties of protein-protein interaction inhibitors. Sci. Rep. 2017, 7, srep46277. [CrossRef][PubMed]