Alkaloids of Abuta panurensis Eichler: In silico and in vitro study of acetylcholinesterase inhibition, cytotoxic and immunomodulatory activities

Rochelly da Silva Mesquita; Andrii Kyrylchuk; Regiane Costa de Oliveira; Ingrity Suelen Costa Sá; Gabriel Coutinho Borges Camargo; Gemilson Soares Pontes; Felipe Moura Araújo da Silva; Rita de Cássia Saraiva Nunomura; Andriy Grafov

doi:10.1371/journal.pone.0239364

Abstract

Natural products obtained from species of the genus Abuta (Menispermaceae) are known as ethnobotanicals that are attracting increasing attention due to a wide range of their pharmacological properties. In this study, the alkaloids stepharine and 5-N-methylmaytenine were first isolated from branches of Abuta panurensis Eichler, an endemic species from the Amazonian rainforest. Structure of the compounds was elucidated by a combination of 1D and 2D NMR spectroscopic and MS and HRMS spectrometric techniques. Interaction of the above-mentioned alkaloids with acetylcholinesterase enzyme and interleukins IL-6 and IL-8 was investigated in silico by molecular docking. The molecules under investigation were able to bind effectively with the active sites of the AChE enzyme, IL-6, and IL-8 showing affinity towards the proteins. Along with the theoretical study, acetylcholinesterase enzyme inhibition, cytotoxic, and immunomodulatory activity of the compounds were assessed by in vitro assays. The data obtained in silico corroborate the results of AChE enzyme inhibition, the IC₅₀ values of 61.24μM for stepharine and 19.55μM for 5-N-methylmaytenine were found. The compounds showed cytotoxic activity against two tumor cell lines (K562 and U937) with IC₅₀ values ranging from 11.77 μM to 28.48 μM. The in vitro assays revealed that both alkaloids were non-toxic to Vero and human PBMC cells. As for the immunomodulatory activity, both compounds inhibited the production of IL-6 at similar levels. Stepharine inhibited considerably the production of IL-8 in comparison to 5-N-methylmaytenine, which showed a dose dependent action (inhibitory at the IC₅₀ dose, and stimulatory at the twofold IC₅₀ one). Such a behavior may possibly be explained by different binding modes of the alkaloids to the interleukin structural fragments. Occurrence of the polyamine alkaloid 5-N-methylmaytenine was reported for the first time for the Menispermaceae family, as well as the presence of stepharine in A. panurensis.

Citation: da Silva Mesquita R, Kyrylchuk A, Costa de Oliveira R, Costa Sá IS, Coutinho Borges Camargo G, Soares Pontes G, et al. (2020) Alkaloids of Abuta panurensis Eichler: In silico and in vitro study of acetylcholinesterase inhibition, cytotoxic and immunomodulatory activities. PLoS ONE 15(9): e0239364. https://doi.org/10.1371/journal.pone.0239364

Editor: Alexandre G. de Brevern, UMR-S1134, INSERM, Université Paris Diderot, INTS, FRANCE

Received: March 10, 2020; Accepted: September 7, 2020; Published: September 29, 2020

Copyright: © 2020 da Silva Mesquita et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: European project Horizon 2020-MSCA-RISE-2016-734759, acronym VAHVISTUS; Finnish Center of Excellence in ALD; CAPES, PROCAD Amazônia, 88881.200581/201801; CNPq/MCT (CT-Amazônia Ed. No. 77/2013, proc. No. 408172/2013-4); FINEP; Foundation for Promotion of Research in the State of Amazonas - FAPEAM (Brazil). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Abbreviations: 1D NMR, One-dimensional NMR; 2D NMR, Two-dimensional NMR; AChE, acetylcholinesterase; AChI, acetylcholine iodide; C18, Octadecyl Carbon Chain; COSY, Correlated Spectroscopy; DAD, Diode Array Detection; DEPT-135, Distortionless Enhancement by Polarization Transfer; DMSO, dimethylsufoxide; DTNB, 5,5´-dithio-bis(2- nitrobenzoic)acid; ELISA, Enzyme-Linked Immunosorbent Assay; FBS, Fetal bovine serum; HMBC, Heteronuclear Multiple Bond Correlation; HPLC, High Performance Liquid Chromatography; HRMS, High Resolution Mass Spectrometry; HSQC, Heteronuclear Single Bond Correlation; IC₅₀, Half Minimal Inhibitory Concentration; IL, Interleukin; LC-APCI-MS, Liquid Chromatography—Atmospheric Pressure Chemical Ionization—Mass Spectrometry; m/z, Mass-to-charge ratio; MS/MS, Tandem Mass Spectrometry; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; NMR, Nuclear magnetic resonance; PBMC, peripheral blood mononuclear cells; PBS, Phosphate buffer saline; PFP, Pentafluorophenyl; RMSD, Root-mean-square deviation; TMS, Tetramethylsilane; UCSF, University of California, San Francisco; UV, Ultraviolet; XRD, X-ray diffraction; μM, micromolar

Introduction

Menispermaceae family has a wide geographic distribution, mainly in tropical and subtropical regions of the world; its name is related to a crescent moon shape of the seeds [1, 2]. The genus Abuta is native to tropical Central and South America, where it is represented by more than 30 species. Some of them have been used by indigenous people to prepare curare, alkaloid-containing arrow and dart head poisons that paralyze the prey [1, 3].

As a result, the Menispermaceae family have been in a focus of rising interest to study several classes of secondary plant metabolites, including alkaloids; which are rather abundant constituents [4–11]. Several classes of alkaloids were isolated from Abuta, those include: isoquinoline [12, 13], benzylisoquinoline [12, 14], benzyltetraisoquinoline [15], bisbenzyltetraisoquinoline [15–17], aporphine [18], and proaporphine [12, 15] derivatives; as well as other less frequent ones such as tropolone-isoquinoline, azafluoranthene, and benzazepine alkaloids [19, 20]. Those alkaloids reveal a wide range of pharmacological activities including muscle relaxant [1], antiplasmodial [14], inhibitory for acetylcholinesterase (AChE) and butyrylacetylcholinesterase enzymes [17, 21, 22], cytotoxic [19, 23, 24], and immunomodulatory [25].

A stepharine is one of the most representative proaporphine alkaloids in Menispermaceae family; it was first identified in the genus Stephania [26]. Proaporphine alkaloids are principally known for their potential to inhibit reversibly the acetylcholinesterase enzyme [27, 28]. Effectiveness and anti-AChE potential of the stepharine sulfate salt (stephaglabine) was reported for the treatment of traumatic and postoperative injuries of the peripheral nervous system and confirmed by a clinical study [29]. However, a pharmaceutical potential of stepharine is much wider according to the literature reports [30, 31]. It showed a cytotoxic activity against two human lung cancer cell lines [32], as well as a weak antifungal potential and DNA- damaging activity [30]. Therefore, both isolation of stepharine from plants cell cultures and several synthetic procedures were developed to satisfy a growing need for the medicinal use of the compound [31, 33].

Polyamine alkaloids (the derivatives of putrescine, spermidine, spermine, and cadaverine) are metabolites that occur widely in angiosperm plants [34–37], but are practically absent in sterile ones [35, 37–42]. Those compounds may also be isolated from other natural sources [43], particularly from fungi [44–48]. The polyamine alkaloids are not common for the Menispermaceae family, they were reported only for a Cissampelos genus [37, 49]. Cinnamoyl derivatives of polyamine alkaloids inhibit the AChE and α-glucosidase [38, 50, 51]. Alongside, they also inhibited cancer cell growth [46, 48, 52] and revealed an antiviral activity [53]. In the Amazon region, a polyamine alkaloid N,N’-di-E-cinnamoylspermidine or maytenine was isolated from Maytenus krukovii (Maytenus chuchuhuasha) trees (Celastraceae) [46, 54]. Several studies have stimulated the development of synthetic approach to that class of polyamides [46, 55–58] including the maytenine synthesis [59].

In the present study, 5-N-methylmaytenine (1) and the stepharine (2) were isolated for the first time from A. panurensis. This is also the first report on the occurrence of (1) in Menispermaceae family. Interactions of the alkaloids in question with acetylcholinesterase (AChE) enzyme and cytokines IL-6 and IL-8 were investigated in silico by molecular docking. The potential of (1) and (2) as AChE inhibitors, antitumor and immunomodulatory agents was demonstrated by in vitro studies.

Materials and methods

Chemicals

Reagents and HPLC-grade solvents were purchased from Tedia Company (Fairfield, OH, USA) and Sigma-Aldrich and used as supplied. P.A. (Nuclear) grade solvents were purified by standard procedures used in natural products chemistry. An ultrahigh-purity water was obtained by Milli Q system (Millipore, Bedford, MA, USA).

Plant material

The authors declare that a specific permission from the National Institute of Amazonian Research (INPA) was required to collect plant material. The authors got the permission No. 35/12 of 02.12.2017 and confirm that the study did not involve endangered or protected species. A. panurensis plant material was collected at the Adolpho Ducke Forest Reserve, 26 km along the AM-010 highway from the city of Manaus, the State of Amazonas, Brazil. The species under investigation had been identified by the taxonomist L.S.Mergulhão. The voucher specimens were deposited in the Herbarium of the National Institute of Amazonian Research (INPA) under the voucher no 279373. The access to genetic heritage was registered at the National System of Genetic Heritage and Associated Traditional Knowledge Management (SisGen, Brazil) under the code number A9CC956.

The branches collected were dried at room temperature (ca. 20°C) for 10 days. Subsequently, the vegetal material (1.4 kg of branches) was crushed in a knife mill and stored in a refrigerator until use.

Extraction

Dried and crushed plant material was subjected to an acid-base extraction [60]. The crushed branches (300g) were macerated with a 10% solution of NH₄OH (2L) and CH₂Cl₂ (2L) at room temperature (20°C) for 72h, the material was stirred every 24 hours. The organic phase (1.5L) was transferred to a separatory funnel with a 10% solution of acetic acid (2L) and stirred manually. Then, the acidic aqueous phase was transferred to another vessel and the pH was adjusted to 10 using NH₄OH and extracted with CH₂Cl₂ (2 × 300mL). The CH₂Cl₂ phase was separated, concentrated on a rotary evaporator under reduced pressure, and dried with anhydrous sodium sulfate, resulting in the alkaloid fraction (280mg).

LC-APCI-MS analysis

LC-APCI-MS analyzes were performed on an Acella chromatograph (Thermo Scientific); coupled to a triple-quadrupole mass spectrometer model TSQ Quantum Acess^® (Thermo Scientific), equipped with an Atmospheric Pressure Chemical Ionization (APCI) source, operated in positive mode with monitoring in the range of m/z 100–800. The mass spectrometer was equipped with Surveyor LC Pump Plus, Surveyor Autosampler Plus, Rheodyne injection valve (25μL), Luna C18 column (150 × 4.60mm, 5μm) (Phenomenex–Torrance, CA, USA), operating simultaneously with Surveyor PDA Plus diode array detector (DAD). The mobile phase was composed of B (methanol) and A (formic acid 1% v/v in H₂O) with a linear elution gradient: 0–20 min 20–80% B, 20–35 min 80% B, 35–45 min 20–80% B. The flow rate of the mobile phase was 1 mL/min and the injection volume was 10μL. The DAD detector was set up for monitoring between 200-400nm. The spectra were processed using an Xcalibur software (version 2.2).

Semi-preparative HPLC analysis

Isolation of the alkaloids was performed on a semi-preparative scale on a Shimadzu chromatograph composed of a CBM-20A communication module, SPD-20AV UV detector, DGU-20A5 degasser, LC-6AD pump, 200 μL Rheodyne injection valve, and Luna C18 column (250 x 15.00mm, 5μm) (Phenomenex–Torrance, CA, USA) with a flow rate of 3 mL/min. The mobile phase was composed of B (methanol) and A (formic acid 1% v/v in H₂O), with a linear elution gradient: 0–20 min 20–80% B, 20–35 min 80% B, 35–45 min 20–80% B. The UV detector was set to monitoring at 260nm and 280nm. Fractions containing 5-N-methylmaytenine (11.2mg—1) and stepharine (22.1mg—2) were collected and analyzed by high-resolution mass spectrometry (HRMS) and NMR spectroscopy.

High resolution mass spectrometry

HRMS analyses were performed on a Shimadzu chromatograph composed of a CBM-20A communication module, a SPD-20AV UV detector, a LC-20AD pump, a SIL-20A HT autosampler (200μL), a CTO-20A oven, and a Luna PFP column (150 x 2mm, 100A); coupled to a Bruker microTOF-QII mass spectrometer, equipped with an Atmospheric Pressure Chemical Ionization (APCI) source, operated in a positive mode. The instrument parameters were as follows: capillary voltage, 4500V; nebulizer pressure (N₂), 4.0 bar; dry gas flow (N₂), 8L/min; dry heater temperature, 200°C; with a monitoring in the range of m/z 100–800 Da. The mobile phase was composed of B (formic acid 0.1% v/v in methanol) and A (formic acid 0.1% v/v in H₂O) with a linear elution gradient as follows: 0–2 min 20–80% B; 2–42 min 100% B. The flow rate of the mobile phase was set to 0.2 mL/min and the injection volume was 10μL. The UV detector was set up for monitoring between 254nm and 330nm. The spectra were processed using a Bruker Compass Data Analysis software (version 4.2).

1D and 2D NMR spectroscopy

NMR spectra were recorded on a Bruker Avance IIIHD spectrometer, 500.13 MHz for ¹H and 125.0 MHz for ¹³C operated at a magnetic field strength of 11.7 Tesla, equipped with a 5 mm direct detection PA BBO BBF HD-05-Z SP Intelligent probe incorporating Z-axis gradient coil, capable of providing gradient amplitudes up to 50 G/cm. Shigemi’s 5.0 mm NMR tubes were used. For structural elucidation, the samples of 5-N-methylmaytenine and stepharine were solubilized in 600 μL of DMSO-d₆ (δ_H 2.50, δ_C 39.9) and CD₃OD (δ_H 3.34, δ_C 49.8), respectively. The acquisition of ¹H, ¹³C, DEPT 135, COSY, HSQC, and HMBC spectra was performed using standard Bruker pulse sequences. The analysis based on ¹H NMR data was performed by solubilizing 10.0 mg of the 5-N-methylmaytenine in 550 μL of DMSO-d₆ with 50 μL of TMS (0.5 mM, 98%, Tokyo Chemical Industry) and 15.0 mg of the stepharine in 550 μL of CD₃OD with 50 μL of TMS (0.5 mM, 98%, Tokyo Chemical Industry) at 25 °C. Acquisition of 5-N-methylmaytenine and stepharine spectra was performed using the zg30 pulse sequence with water signal suppression, data points of the 64 kB time domain, 10 kHz spectral width, 1.00 second relaxation delay (D1), 3.27 second acquisition time (AQ), 32 scan numbers with DS of 2, decomposition resolution of 0.31 Hz, a constant receiver gain at 161 (5-N-methylmaytenine) and 181 (stepharine) with displacement frequency set at 2,425.23 Hz, PLW1 of 20.3 W. The calibration pulse (P1 9.400 μs to 5-N-methylmaytenine and P1 10.300 μs to stepharine) with PLW9 were of 7.183×10⁻⁵ W (5-N-methylmaytenine) and 8.6243×10⁻⁵ W (stepharine). Data were processed using Bruker^® Topspin 4.0.6 software.

Molecular docking calculations

Ligand structures preparation.

Ligand structures were built manually and preliminary optimized in a classic molecular mechanics software. In order to use equilibrium ligand structures, their geometry optimizations were conducted using a semi-empirical PM7 level [61] within the MOPAC program [62]. Ligand structure files for docking were prepared using AutoDock Tools [63, 64]. Default settings for the detection of rotatable bonds were used.

Protein files preparation.

X-ray structures of the proteins under investigation (AChE, PDB ID: 6H12; IL-6, PDB ID: 4NI7 and IL-8, PDB ID: 3IL8) were obtained from the RCSB Protein Data Bank. Missing residues. Modeling of missing residues in the protein structures was performed using Modeller web-service [65]. Obtained structures that do not possess serious structural issues were selected for docking studies. Clashes or contacts between the side chains of the produced structures were resolved by minimization routine implemented in UCSF Chimera [66]. Water molecules are commonly found in the XRD structures of proteins and can have substantial impact on the docking affinities. Selection of water molecules that could be important in docking was performed according to the distance criterion. All water molecules farther than 3.3 Å from the H-bond donors and acceptors in the protein structures were removed. Protonation of the oxygen atoms in water molecules was made in UCSF Chimera followed by minimization of the hydrogen positions.

Binding sites of the proteins were identified using Discovery Studio Visualizer [67]. AutoDock Tools [63,64] were used for the preparation of Structure files for docking.

Docking runs.

Docking studies were performed with AutoDock Vina program [68]. The protein structures stayed rigid during the docking in all cases. Each run generated nine binding poses. Exhaustiveness parameter of 50 was used for the medium search space sizes, e.g. for the binding pocket in AChE. In the case of larger search spaces (IL-6, IL-8) exhaustiveness of 500 was used. Since the success of a docking run depends on a random seed, which is defined at the beginning of the run and does not change during it; we have performed three docking runs for each protein-ligand pair and search space. The best docking poses found were similar between the runs in most cases, showing that the chosen parameters provided exhaustive search of the conformational space. Missing residues. Selected protein structures from Modeller service were used for the estimation of binding to the regions that were missing in the initial XRD structures. Search spaces that include the modeled parts of the protein were chosen. Water molecules. It is known that inclusion of all water molecules in the docking run can lead to erroneous results [69]. Therefore, trial docking runs were performed with an inclusion of one of the crystallization water molecules found in the XRD structures at a time. The water molecules were fixed during the docking run. The search space either covered the defined binding site of the protein or was centered on the water molecule and had a size of 20×20×20 Å.

Protein structure images were obtained using Discovery Studio Visualizer [67], PyMOL [70], and UCSF Chimera [66]. Detailed docking parameters are collected in the S1 File.

Acetylcholinesterase inhibition assay

Acetylcholinesterase enzyme from an electric ray Tetronarce californica (Sigma-Aldrich, USA) was used for the experiments. The in vitro AChE inhibition assay was performed in 96-well microplates according to the methodology proposed by Ellman et al. [71] (1961) and Senol et al. [72] (2015) with some modifications. The alkaloids were tested at concentrations of 2.8; 5.6; 11.2; 22.5; 45.0 and 90.0 μg/mL. Initially, 20 μL of each sample from the stock solution (1 mg/mL) were added and serial dilutions were performed. Then, 150 μL of a sodium phosphate buffer pH = 8 (0.1 mM), 20 μL of 5,5´-dithio-bis(2- nitrobenzoic)acid (DTNB, 0.0025 M), and 20 μL of the acetylcholinesterase enzyme (1 U/mL) were added subsequently to each well at 25 °C and left for 15 minutes. The reaction was initiated by addition of 10 μL of acetylcholine iodide (AChI) (0.1 M). Neostigmine (0.28–9.0 μg/mL) was used as a positive control.

A thiocholine formed by the enzymatic hydrolysis of the AChI, reacts with the DTNB giving rise to yellow 5-mercapto-2-nitrobenzoate anion. Concentration of the latter in each well was measured as absorbance at 405 nm using a 96-well microplate reader spectrophotometer (Biotek model ELX800). The transformation was monitored for 30 min at 5 min intervals. Inhibition curve was plotted as the inhibition percent vs concentration. All assays were performed in triplicate.

Cytotoxicity assay

The cytotoxicity of the alkaloids to different cell lines was evaluated by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) cell proliferation assay. The following cell lines were used for the evaluation: K562 (human chronic myelogenous leukemia, ATCC^® CCL-243^™), U937 (human histiocytic lymphoma, ATCC^® CRL-1593.2^™), HL60 (human acute promyelocytic leukemia, ATCC^® CCL-240^™), Vero (kidney epithelial cells extracted from an African green monkey Chlorocebus sp., ATCC^® CCL-81^™), and human peripheral blood mononuclear cells (PBMC) from healthy blood donors. The cell lines K562, HL60, and U937 were kindly donated by Prof. S.O. Saad (Hematology and Hemotherapy Center at the National Institute of Blood Science and Technology of the University of Campinas, Campinas, São Paulo, Brazil). The Vero cells were kindly donated by Prof. F.Naveca (Laboratory of Infectious Disease Ecology in the Amazon, L. and M.Deane Institute, FIOCRUZ, Manaus, Amazonas, Brazil). The cells were cultured in 96-well plates. An amount of 2×10⁴ cells was seeded into each well containing 0.2 mL of a RPMI medium supplemented with 10% FBS, penicillin-streptomycin and fungizone, in an atmosphere of 5% CO₂ at 37° C for 24 hours. After a formation of sub-confluent monolayer, the cells were treated with different concentrations of the alkaloids (12 … 100 μg/mL) and incubated again at the same conditions for 24, 48, and 72 hours. Sterile PBS and DMSO 100% were used as negative and positive controls, respectively. Subsequently, the medium was removed from all wells and 10 μL of the MTT (5 mg/mL in sterile PBS) diluted in 100 μL of a DMEM medium (without phenol red to avoid misinterpretation) was added into the wells and incubated for 4 hours at the same conditions mentioned above. After that, the MTT was removed and 50 μL of the MTT lysis buffer were added to each well. The plate was homogenized gently to dissolve the formazan crystals and incubated for 10 minutes at 37 °C. Optical densities of the samples at wavelength of 570 nm were measured using a microplate reader. The relative viability of the cells was estimated using the following equation: where A570 is the absorbance at 570nm. All assays were done in triplicate.

Immunological assay

To analyze the immunomodulatory potential, human peripheral blood mononuclear cells (PBMCs) obtained from healthy blood donor candidates were cultured in RPMI-1640 medium in a 96-well plate and incubated for 24h with 5-N-methylmaytenine (12.5 μM and 25 μM) or stepharine (28.48 μM and 57.0 μM). After the incubation, the supernatants were collected for cytokine assays. The supernatant of untreated PBMC cells was used as control. The concentrations of IL-6 and IL-8 cytokines were evaluated by commercially available enzyme-linked immunosorbent assay (ELISA) kits (Boster, Pleasanton, CA, USA). The results were normalized for protein levels contained in each sample and were expressed in pg/mg of total protein. The assays were repeated in triplicate for each individual sample using untreated cells as negative control. This study was approved by the Committee for Ethics in Research on Human Beings of the HEMOAM (approval number: 3.138.343).

Statistical analysis

Statistical analysis was performed with GraphPad Prism 7.0^® software using Student’s t-test and ANOVA. A probability value of less than 0.05 was chosen as a statistical significance criterion. Throughout the text, the asterisks correspond to the following probability values: * means p<0.01; ** means p< 0.001; and *** means p<0.0001, when compared to the negative control (untreated cells). The half maximum inhibitory concentration IC₅₀ values were calculated using nonlinear regression.

Results

Spectral data

5-N-methylmaytenine (1) was isolated as a light yellow amorphous solid (11.2 mg). ¹H NMR (500 MHz, DMSO d₆, TMS): δ 8.12 (t; 2H; J = 5.5 Hz, 1 and 10-NH), δ 7.56 (m, 2H, 5´-H and 5´´-H or 9´-H and 9´´-H), δ 7.54 (m, 2H, 9´-H and 9´´-H or 5´-H and 5´´-H), δ 7.55 (m, 2H, 6´-H, 6´´-H, 8´-H, 8´´-H), δ 7.40 (m, 2H, 7´-H and 7´´-H), δ 7.42 (m, 1H, 3´-H or 3´´-H), δ 7.38 (m, 1H, 3´´-H or 3´-H), δ 6.63 (d, 1H, 2.4 Hz, 2´-H or 2´´-H), δ 6.60 (d, 1H, 2.4 Hz, 2´´-H or 2´-H), δ 3.20 (m, 2H, 2-H), δ 3.18 (m, 2H, 9-H), δ 2.35 (m, 2H, 4-H), δ 2.32 (m, 2H, 6-H), δ 2.16 (s, 3H, 5-NCH₃), δ 1.61 (m, 2H, 3-H), δ 1.47 (m, 2H, 8-H), δ 1.45 (m, 2H, 7-H). ¹³C NMR (125 MHz, DMSO d₆, TMS): δ 165.32; 165.27; 135.43; 127.98; 127.94, 129.39; 129.83; 138.85; 138.88; 122.79; 122.83; 55.26; 57.12; 42.09; 37.58; 39.07; 27.28; 24.53; 27.48 ppm. (S1 Table in S1 File). MS (APCI+) m/z 420 [M+H]⁺: 202, 188, and 131. HRMS m/z 420.2669 (calc. for C₂₆H₃₄N₃O₂ m/z 420.2646, Δ_{m/z theor.} = -5.6 ppm).

Stepharine (2) was isolated as a light brownish amorphous solid (22.1mg). ¹H NMR (500 MHz, CD₃OD, TMS): δ 7.01 (dd, 1H, 3 and 10Hz, 12-H), δ 7.16 (dd, 1H, 3 and 10Hz, 8-H), δ 6.89 (s, 1H, 3-H), δ 6.41(dd, 1H, 1.8 and 10Hz, 11-H), δ 6.29 (dd, 1H, 1.8 and 10Hz, 9-H), δ 4.72 (m, 1H, 6a-H), δ 3.82 (s, 3H, 2-OCH₃), δ 3.70 (ddd, 1H, 1.5, 6.3 and 13Hz, 5-H), δ 3.61 (s, 3H, 1-OCH₃), δ 2.52 (dd, 1H, 6.6 and 12 Hz, 7-H), δ 2.42 (dd, 1H, 10.5 and 12Hz, 7`-H), δ 3.44 (ddd, 1H, 6.3, 11 and 13Hz, 5-H), δ 3,02 and δ 3,00 (m, 2H, 4-H), δ 1.95 (s, 1H, NH) ppm. ¹³C NMR by HSQC (125 MHz, CD₃OD, TMS): δ 153.63; 150.28; 112.15; 127.81; 126.74; 56.48; 55.37; 43.71; 59.95, 44.90; 43.71; 23.5 ppm. (S2 Table in S1 File). MS (APCI+) m/z 298 [M+H]⁺: 281, 266, 250, 235, 161. HRMS m/z 298.1461 (calc. for C₁₈H₂₀NO₃ m/z 298.1438, Δ_{m/z theor.} = -7.9 ppm).

Binding with acetylcholinesterase of Tetronarce californica (TcAChE)

X-ray structure of the AChE from the electric ray T. californica was found in the RCSB Protein Data Bank under 6H12 code [73]. The crystal structure retrieved from the Data Bank represents the AChE adduct with a functionalized urea ligand. In order to verify robustness of the method, preliminary docking studies were conducted using the ligand structures obtained from the experimental XRD data. The best affinity of the ligand without the inclusion of water molecules reached -15.3 kcal/mol with a RMSD value of 4.03 Å compared to the ligand pose found in the crystal structure. Relatively high RMSD value can be explained by substantial internal mobility of the ligand molecule. Inclusion of 18 crystallization water molecules increased the affinity value to -14.0 kcal/mol with comparable RMSD. In the case of fully hydrated pocket, many binding poses were found with similar affinity values, ranging from -13 to -14 kcal/mol. Some of those conformations showed better RMSD values reaching 2.2 Å. Screening of the water molecules failed to achieve interaction energies lower than those obtained for the empty binding pocket. Therefore, water molecules do not seem to play a substantial role in the binding of the urea ligand. Still, there was a possibility that hydration could be important for interactions with stepharine or 5-N-methylmaytenine. Therefore, it was considered for those ligands as well.

Two major binding sites were identified in the AChE structure–the main active site and the region near the modeled C-terminal loop (Fig 1). Consequently, we have conducted docking studies of these sites and a rear region that can possess smaller pockets (S32–S37 Figs in S1 File and the accompanying information).

Download:

Fig 1. Two major binding sites found for AChE.

Other, smaller sites are not shown.

https://doi.org/10.1371/journal.pone.0239364.g001

Neostigmine is known as an efficient AChE inhibitor that was used as positive control in this study. According to the neostigmine docking results, the best affinity value was achieved for the main cavity of the enzyme and reaches -7.6 kcal/mol. The binding energies for the other sites were weaker by ca. 2 kcal/mol. Inclusion of water molecules did not produce any pronounced effect on the affinities and changed them by ca. 0.5 kcal/mol (S31 Fig in S1 File and the accompanying information).

5-N-methylmaytenine (1) preferably binds to the AChE active site with the affinity of -10.5 kcal/mol. Stepharine (2) fits well inside the AChE active site with the predicted binding affinity of -10.3 kcal/mol. Affinities of both ligands to the other sites were weaker by 2 …4 kcal/mol. Explicit inclusion of water molecules into the binding cavity had little effect on the binding strengths (S31 Fig in S1 File and the accompanying information).

Binding with interleukin-6

X-ray structure of human interleukin-6 (PDB ID: 4NI7) was used for the docking studies [74]. The structure contains a co-crystallized nucleic acid moiety that was removed prior to docking. During the signaling event, the IL-6 associates with the IL-6 receptor (IL-6r) and forms a complex. Then, the second receptor protein, viz. the gp130 glycoprotein, binds to the complex giving rise to a dimer; and the signaling is initiated [74–76]. X-ray structure of the interleukin-receptor complex (Fig 2B) was retrieved from the RCSB PDB (ID: 1P9M) [76].

Download:

Fig 2. Binding sites of the IL-6 according to [74] (A) and IL-6 receptor complex 1P9M (B).

In Fig 2B, the IL-6 structures are beige, the gp130 is marked red, and the IL-6r is green.

https://doi.org/10.1371/journal.pone.0239364.g002

According to the original paper [74], the IL-6 molecule possesses three binding sites, where the interactions with the receptor proteins and the second IL-6 molecule occur (Fig 2A). IL-6 –IL-6r interaction takes place at the Site 1, whereas the IL-6 interacts with the gp130 at the Sites 2 and 3. The IL-6 molecule does not have an apparent binding pocket. Four small binding pockets at the Sites 1 and 2 and several smaller pockets at the upper rim of the protein were identified. Since the pocket volume has never exceed 42 Å³, the entire protein surface was considered for the docking calculations.

The strongest affinity of -7.9 kcal/mol was found for 1, when the ligand interacted with two helices and a modeled loop at the IL-6 upper rim. Several other binding modes employed an interaction with a side surface of the two helices, including a π-stacking with both 5-N-methylmaytenine phenyl rings as well. Water molecules had also been screened for possible interactions with the docked ligand, but no significant effect was found (p 45 of the S1 File).

Stepharine (2) binds to the IL-6 with the affinity of -6.9 kcal/mol at the same site as 5-N-methylmaytenine. Again, water molecules did not provide any additional binding strength (p 45 of the S1 File).

Interleukin-8

The crystal structure of the interleukin-8 was retrieved from the RCSB protein data bank (PDB ID: 3IL8) [77]. It is known that IL-8 interaction with the appropriate receptors (CXCR1 and CXCR2) involves two sites close to the N-terminus (Fig 3) [78–80]. The N-loop designated as Site I includes the residues from Ser14 to Lys20. The Site II is comprised of Glu4-Leu5-Arg6 (‘‘ELR”) residue sequence. During the signaling event, the Site I interacts with the receptor N-terminal residues while the Site II is involved in the interaction with the receptor extracellular residues [78]. Two other important regions of the protein structure comprise lateral sides of the α-helix and the β-strand involved in the IL-8 dimer formation.

Download:

Fig 3. Binding sites of the IL-8 monomer according to [78–80] (A) and IL-8 dimer structure (B).

https://doi.org/10.1371/journal.pone.0239364.g003

No binding sites were identified by the Discovery Studio routines. Both the IL-8 monomer and dimer were employed in the docking studies. Due to a moderate size of the interleukin molecule, the entire surface of the protein was included into the search space. 5-N-methylmaytenine (1) interacts with the IL-8 monomer and dimer with the affinities of -6.9 and -7.0 kcal/mol, respectively. Stepharine (2) interacts with the monomeric and dimeric forms of the IL-8 with the similar affinity of -5.9 kcal/mol. Screening of the water molecules did not improve the binding strengths (S38 and S39 Figs in S1 File and accompanying information).