Chemical Composition, Preliminary Toxicity, and Antioxidant Potential of Piper marginatum Sensu Lato Essential Oils and Molecular Modeling Study

The essential oils (OEs) of the leaves, stems, and spikes of P. marginatum were obtained by hydrodistillation, steam distillation, and simultaneous extraction. The chemical constituents were identified and quantified by GC/MS and GC-FID. The preliminary biological activity was determined by assessing the toxicity of the samples to Artemia salina Leach larvae and calculating the mortality rate and lethal concentration (LC50). The antioxidant activity of the EOs was determined by the DPPH radical scavenging method. Molecular modeling was performed using molecular docking and molecular dynamics, with acetylcholinesterase being the molecular target. The OES yields ranged from 1.49% to 1.83%. The EOs and aromatic constituents of P. marginatum are characterized by the high contents of (E)-isoosmorhizole (19.4–32.9%), 2-methoxy-4,5-methylenedioxypropiophenone (9.0–19.9%), isoosmorhizole (1.6–24.5%), and 2-methoxy-4,5-methylenedioxypropiophenone isomer (1.6–14.3%). The antioxidant potential was significant in the OE of the leaves and stems of P. marginatum extracted by SD in November (84.9 ± 4.0 mg TE·mL−1) and the OEs of the leaves extracted by HD in March (126.8 ± 12.3 mg TE·mL−1). Regarding the preliminary toxicity, the OEs of Pm-SD-L-St-Nov and Pm-HD-L-St-Nov had mortality higher than 80% in concentrations of 25 µg·mL−1. This in silico study on essential oils elucidated the potential mechanism of interaction of the main compounds, which may serve as a basis for advances in this line of research.


Introduction
Aromatic plants are those that possess a pleasant fragrance due to the presence of essential oils in their leaves, flowers, stems, or other parts. These plants are often cultivated for their aromatic properties and are used in various applications, including the culinary, medicinal, perfumery, and food industries. Aromatic plants contain essential oils, which are complex mixtures of volatile compounds. These essential oils (EO) give the plants their characteristic aroma and are extracted through methods such as steam distillation or hydrodistillation. Essential oils possess a wide range of biological properties, including antimicrobial, antifungal, antiviral, and insecticidal activities [1].
Essential oils are composed of a complex mixture of various chemical compounds that give them their characteristic aroma and potential therapeutic properties. While there are hundreds of different compounds found in essential oils, there are some common groups of compounds, such as terpenes, terpenoids, phenylpropanoids, aldehydes, and phenols, among other classes; these compounds are responsible for the potential biological activity of essential oils. In addition, they are responsible for the defense of plants against herbivorous insects and phytopathogens and in the attraction of insects and other pollinators [2][3][4].
Furthermore, the chemical composition of essential oils can be altered by environmental stimuli that can redirect the plant metabolic pathway, causing the biosynthesis of different compounds. These factors prominently include plant/insect, plant/plant, and plant/microorganism interactions; age and development stage; abiotic factors such as temperature, luminosity, rainfall, collection season, and time; and harvest and postharvest techniques. It is worth noting that these factors may correlate with each other and do not act in isolation, though they may exert a joint influence on secondary metabolism [5][6][7]. The chemical characteristics of an essential oil (EO) may vary according to the extraction method, such as hydrodistillation (HD), maceration, solvent extraction, enfleurage, supercritical gas treatment, and microwave-assisted extraction. The heat and pressure used during extraction can, for example, affect the final quality of the EO because the sensitive molecules of a valuable active ingredient can be broken down and oxidized into less effective or sometimes even toxic products [8,9].
Toxicity tests are designed to evaluate or predict the toxic effects on biological systems and measure the relative toxicity of substances [27]. Artemia salina is often used in preliminary toxicity assays due to its sensitivity to various chemical substances [28], including compounds present in essential oils [29]. While the A. salina lethality test is commonly used for preliminary toxicity screening, it is important to note that the results obtained from this test may not directly translate to the potential toxicity of essential oils in humans or other animals [30][31][32]; however, they can bring a potential toxicity perspective to natural products. Furthermore, the toxicity in A. salina may be related to the inhibition of acetylcholinesterase (AChE) [33], which is an enzyme that plays a crucial role in the termination of nerve impulse transmission by catalyzing the hydrolysis of the acetylcholine neurotransmitter; several studies reported AChE as a potential molecular target to cause the death of A. salina using molecular modeling, which can be a useful tool to analyze the potential interactions between the molecules present in essential oils and AChE [34].
Molecular modeling refers to the computational techniques used to study and predict the behavior and properties of molecules. In the context of studying acetylcholinesterase (AChE) in A. salina, molecular modeling can be used to gain insights into the structure and function of the enzyme. As the experimentally determined structure of AChE in Artemia salina is not available, molecular modeling techniques can be used to predict its structure. Molecular docking simulations can be performed to study the interactions between AChE and ligands. Molecular dynamics simulations can provide insights into the dynamic behavior of AChE, supporting experimental studies [35][36][37][38].
Additionally, antioxidant analysis methods are important because antioxidants can protect the biological system against the harmful effect of processes or reactions that can cause excessive oxidation [39]. The growing epidemiological evidence regarding the role of antioxidant foods in the prevention of certain diseases has led to the development of a large number of methods to determine antioxidant capacity [40]. Thus, the objective of this study was to study the chemical composition, antioxidant potential, and preliminary toxicity and perform an in silico study to elucidate the potential mechanism of molecular interaction of the major compounds of EOs and volatile concentrates from the leaves, stems, and spikes of P. marginatum sensu lato.

Yields of Essential Oils
The essential oil (EO) yields of P. marginatum are presented in Table 1. The yields of P. marginatum EOs obtained from the leaves and stems by hydrodistillation (HD) and steam distillation (SD) in the months of November and March ranged from 1.66% to 1.83%. The EO yields showed significantly different results; this difference may be related to the seasonality of collection, as described in seasonal studies of the EO of P. marginatum [41].

Chemical Composition
In addition to the great qualitative variability in the secondary metabolites among the EOs and volatile concentrates of P. marginatum (Table 2), variations in these metabolites were found with respect to the part being studied (leaves, stems, and spikes) and the extraction technique (HD, SD, and simultaneous distillation and extraction (SDE)), as shown in Table 2. Circadian rhythm, humidity, atmospheric air composition, herbivory, and pathogen attack, altitude, ultraviolet and visible radiation, rainfall index, availability of macro-and micronutrients, seasonality, plant age, and temperature were shown to be key factors explaining the quantitative and even qualitative variation in the production of secondary metabolites in the same species [41].

Multivariate Analysis
Heatmap clustering shows the closeness of the samples on the basis of their chemical composition ( Figure 1). In the heatmap, only compounds >1.0% are considered. In the color gradient, yellow represents the lowest constituent percentage, and green represents the highest constituent percentage. HCA showed three major clusters. In the first cluster,  A multivariate statistical approach, PCA, was performed to distinguish the studied samples based on the class of compounds and major chemical constituents (>1.0%). The first two principal components (PCs) explained over 97% of the total variance. SDE-L-Stnov was separated by phenylalkaloids ( Figure 2A). Among the chemical constituents, the samples were mainly separated by 2-methoxy-4,5-(methylenedioxy)-propiophenone, (E)isoosmorhizole, 2-methoxy-4.5-methylenedioxypropiophenone isomer, and isoosmorhizole ( Figure 2B). The HCA results also supported the PCA results.

Antioxidant Activity
The Trolox antioxidant standard was similarly used to test the samples. A concentration versus inhibition curve was prepared to directly compare the standard with the samples. The inhibition curve of Trolox was prepared at concentrations of 10.0-1 µg·mL −1 , and the inhibition varied from 84.6% to 12.2%, as observed in Table 3. The reaction was quite fast, approximately 10 min. The dose-response correlation was highly linear (R 2 = 0.97), and the obtained linear equation (y = 0.108x) was used to express the antioxidant activity results in mg of Trolox equivalents per mL of oil (mg TE·mL −1 ). The EOs of P. marginatum were evaluated at a single concentration; the end point of the reaction was determined after 120 min; the absorbance was measured at 517 nm; and the results are expressed in terms of Trolox equivalents, as shown in Table 4.  The tested oils exhibited DPPH inhibition ranging from 31.2 ± 1.5 to 49.8 ± 3.0%. The EOs of the leaves and stems of P. marginatum extracted by SD in November and the EOs of the leaves extracted by HD in March showed the highest antioxidant capacity. Regarding antioxidant activity, the EOs of the leaves and stems obtained by HD in November had the highest value (135.3 mg TE·mL −1 ), followed by the EOs of the leaves and stems extracted by HD in March (126.8 mg TE·mL −1 ). The EOs of the leaves and stems of P. marginatum obtained by SD in November exhibited lower potential (84.9 mg TE·mL −1 ) than the other EO fractions studied. However, the percentage of antioxidant activities was significantly higher than that found in the EO of the roots of P. marginatum collected in the state of Rondônia, except for the EO extracted from the leaves and stems in March [44].

Preliminary Toxicity
In the control group, no mortality was observed. This demonstrates that it is feasible to use dimethyl sulfoxide (DMSO) as a solvent for bioassays with A. salina larvae. The LC 50 values were calculated by converting the percentage of larval mortality into probit values [51], which were used to draw a linear equation on a semilogarithmic scale. The results of the present work can be observed in Table 5; in all analyzed cases, the concentrations that presented mortality were those superior to 10 µg·mL −1 . The analyzed samples Pm-SD-L-St-Nov and Pm-HD-L-St-Nov showed mortality above 80% at concentrations of 25 µg·mL −1 , and the LC 50 was 17.47 ± 0.33 µg·mL −1 and 17.33 ± 0.53 µg·mL −1 , respectively. Values above the positive control lapachol had an LC 50 of 21.2 ± 2.2 µg·mL −1 , that is, presenting a superior bioactivity. According to Meyer et al. [52], an extract can be considered toxic if its LC 50 value is ≤30 µg·mL −1 . In the present results, this may be associated with the presence of a class of compounds such as phenylpropanoids and phenylalkanoids [53][54][55] or, more specifically, may be related to the presence of the compounds (E)-Anethole, Isoosmorhizole, (E)-Isoosmorhizole, iso-mer-2-methoxy-4.5-Methylenedioxypropiophenone, and 2-Methoxy-4,5-(methylenedioxy)-propiophenone, as shown in Table 2.

Molecular Docking Analysis
The interaction between the molecules of natural origin and the molecular targets of pharmacological interest was effectively assessed using in silico methods [34]. The present research employed molecular docking to analyze the interaction of the principal compounds found in specific plant species with the binding pocket of AChE, which is a molecular target that is associated with toxicity and was previously investigated in A. salina models [34]. Thus, we tested five major components such as (E)-Anethole, Isoosmorhizole, (E)-Isoosmorhizole, isomer-2-methoxy-4.5-Methylenedioxypropiophenone, and 2-Methoxy-4,5-(methylenedioxy)-propiophenone against the binding cavity of AChE ( Figure 3). Our molecular docking results suggested that the compound (E)-Isoosmorhizole exhibited the best docking score, −9.76 kcal/mol, compared to the other essential oil (EO) components. The compound (E)-Anethole remained interacting with the active binding site amino acid residues via π-Sigma, π-π stacking, van der Waals force, and alkyl or via π-alkyl interactions (docking score of −8.11 kcal/mol). The key amino acid residues involved during docking were Tyr

Molecular Dynamics Analysis (MDA)
The stability and convergence of analyses for acetylcholinesterase (AChE target) against various EO components ((E)-Anethole, Isoosmorhizole, (E)-Isoosmorhizole, isomer-2-methoxy-4.5-Methylenedioxypropiophenone, and 2-Methoxy-4,5-(methylenedioxy)propiophenone) were analyzed using extended molecular dynamics simulations over the period of 150 ns using the "Desmond, Schrodinger, LLC, NY, 2022" tool [6]. The results of the 150 ns simulation indicated a consistent conformation, as evidenced by the root-mean-square deviation (RMSD) analysis. From Figure 4A-I, it is very clear that complexes with EO components had a stable RMSD value. The RMSD of the backbone of the (E)-Isoosmorhizole-AChE C-RMSD complex remained under 3.2 Å, while the ligand RMSD of (E)-Isoosmorhizole was 6.4 Å, indicating good convergence and stable conformations throughout the simulation ( Figure 4E). This conclusion is further corroborated by the stable graphs of the root-mean-square deviation (RMSD). The ligand's high binding affinity implies a stable complex with AChE, as evidenced by Figure 2B-J. While there were some minor fluctuations in the the root-mean-square (RMSF) fluctuations plot for AChE, most of the residues remained relatively constant throughout the simulation. This suggests that the protein structure is stiff in the ligand-bound conformations and that the residues may be more flexible, as indicated by the RMSF plots. The RMSF values of the EO components, including (E)-Anethole, Isoosmorhizole, (E)-Isoosmorhizole, isomer-2-methoxy-4.5-Methylenedioxypropiophenone, and 2-Methoxy-4,5-(methylenedioxy)-propiophenone, showed noticeable but minimal fluctuations. These fluctuations indicate that these ligands exhibit significant internal atomic fluctuations during their interaction with AChE, which can be attributed to their flexibility properties. The flexibility of these small molecule ligands allows them to adopt various conformations and interaction patterns within the receptor protein cavity, resulting in the observed fluctuations in the RMSF values. Overall, the results of the simulation suggest that the complex formed between (E)-Isoosmorhizole and AChE is stable and that the amino acid conformations are also stable. The considerable decrease in gyration (Rg) signifies that the protein assumes a tightly aligned configuration upon binding to the ligand. Additionally, the existence of hydrogen bonds between the protein and ligand supports the notion that the complex is both stable and has a strong interaction. During the 150 ns simulation, the compound (E)-Isoosmorhizole and AChE were observed to form significant hydrogen bonds (Figure 5e), and oppositely charged residues also exhibited salt bridges, which significantly contributed to the protein's stability. Collectively, the analysis of Rg suggests that the protein structure becomes more condensed and less pliant following ligand binding.

Protein-Ligand Interactions
Figure 3a-j illustrate the interactions that occurred between the amino acid residues of AChE and the ligands during the simulation time (0.00-150.00 ns). The "protein-ligand contact" plots indicate the time fractions of protein-ligand interactions that were maintained throughout the simulation. The "Timeline of interactions and contacts" diagram shows the timelines of interactions and contacts, including H-bonds and hydrophobic, ionic, and water bridges. The top panel displays the total number of specific contacts between the protein and the ligand during the simulation time. The bottom panel provides a detailed list of the residues that interacted with the ligand in each frame of the MD simulation course. Residues with multiple contacts with the ligand are represented by darker shades of orange, as indicated on the right side of the diagram.
From Figure 5a, we can notice that compound (E)-Anethole in the AChE complex is stabilized via H-bond, hydrophobic, and water-bridge-like interactions.

Collection of Botanical Material
The specimens (P. marginatum sensu lato) were collected on the campus of the Federal Rural University of Amazonia in the city of Belém, Pará, in the morning on days between November 2018 and March 2019. Botanical identification was performed by comparison with materials identified by Elsie Franklin Guimarães, a specialist in Piperaceae, and samples were incorporated into the "João Murça Pires" Herbarium of the Emílio Goeldi Museum of Pará (P. marginatum sensu lato MG184836) in Belém, Pará.

Determination of Residual Moisture
Before moisture analysis, the sample was dried in an air circulation oven at approximately 35 • C, for a period of 5 days. The moisture present in the samples was determined with the aid of a Gehaka infrared moisture analyzer (IV2500).

Hydrodistillation
For the EO extraction process, 40 g of fresh botanical material were dried in an air circulation oven and then subjected to HD. Equal proportions of water and plant material were used, according to the methodology described by Ferreira et al. [56]. Essential oils from leaves and stem were not separately extracted, and the process was carried out for 3 h, at a temperature of approximately 100 • C.

Distillation and Simultaneous Extraction
Distillation-simultaneous extraction was performed in a Chrompack Nickerson and Likens extractor coupled to a refrigeration system (5-10 • C) and connected to two roundbottomed flasks. Then, 10 g of botanical material and 125 mL of distilled water were added to a 250 mL flask with a heating mantle, from which the vapors passed to the condenser. Two milliliters of n-pentane were added to a 5 mL flask, which was kept in a water bath at 53-56 • C for evaporation and extraction (condensation) of the volatile concentrate. The extraction time was 2 h [57].

Steam Distillation
The EO isolation process using steam distillation (SD) was carried out using a modified Clevenger glass system apparatus coupled to a refrigeration system to maintain the condensation water between 10 and 15 • C for 3 h, as described by [38].

Identification of Chemical Constituents
The chemical compositions of the P. marginatum EO samples were analyzed using a single quadrupole gas chromatography/mass spectrometry (GC/MS) system (Thermo DSQ-II, Waltham, MA, USA) equipped with a DB-5MS silica capillary column (30 m × 0.25 mm, 0.25 mm; Agilent Technologies, Stevens Creek Blvd., Santa Clara, CA, USA). Aqueous 2:1000 n-hexane was injected in one step (0.1 mL); the temperature of the ion source and other parts was set at 200 • C. The operational conditions of injection and identification were previously described by our research group [57]. The components were identified by comparison of (i) the experimental mass spectra with those compiled in libraries (reference) and (ii) the experimental retention indices with those found in the literature [49,50]. The volatile constituents were quantified by peak-area normalization using a FOCUS GC/flame ionization detector (FID), which was operated under the same conditions as the GC-MS instrument.

Antioxidant Potential
The essential oil samples (10 µL) were mixed with 900 µL of 100 mM Tris-HCl buffer (pH = 7.4), 40 µL of ethanol, and 50 µL of a 0.5% Tween 20 solution (m/m), and then 1.5 mL of 0.5 mM DPPH in ethanol (250 µM in the reaction mixture) was added. Tween 20 was used as an emulsifier for oil-water mixing [58,59]. The mixture was vigorously stirred and kept in a dark environment at room temperature for 30 min. The absorbance reading was performed in the UV-visible at 517 nm in an 800XI spectrophotometer (Femto; São Paulo/SP, Brazil). The control reaction was performed by replacing the sample with 50 µL of Trolox 1 mM in ethanol (the final concentration in the reaction was 25 µM). Calculation of inhibition percentage-IDPPH (%). The percentage of inhibition of DPPH radicals (IDPPH) was performed according to what is described in the literature [59]. The percentages of inhibition of the oils were compared with the inhibition induced by the 1 mM Trolox solution. The total antioxidant capacity expressed in mg ET/mL of oil was calculated according to the equation proposed by [60,61]. Essential oils were tested without dilution.

Determination of Preliminary Toxicity in Artemia salina Leach
An artificial brine was prepared with 46 g of NaCl, 22 g of MgCl2·6H2O, 8 g of Na 2 SO 4 , 2.6 g of CaCl 2 · 2 H 2 O or CaCl 2 · 6 H 2 O, and 1.4 g of KCl dissolved in 2000 mL of distilled water. The brine pH was adjusted to 9.0 using Na 2 CO 3 to avoid the risk of larval death due to a pH decrease during the incubation period [62].
A. salina cysts (25 mg) were incubated in artificial brine at 25 • C in a glass container with a capacity of 10.6 dm3 and an oxygenation system consisting of an aeration pump. The container consisted of two parts, one containing the eggs, which was protected from light, and one that was illuminated by artificial light generated by a 40 W lamp.
This division was performed because the larvae have positive phototropism, i.e., affinity for light, and, consequently, after hatching, the larvae migrated through the partition to the illuminated portion of the glass container and were separated [62,63].
Twenty-four hours after hatching, approximately 10 larvae were added to the sample test tubes using an automatic micropipette. The tubes were filled to a total volume of 5 mL with brine water. The control group was prepared using 5 mL of 5% DMSO brine and 10 A. salina larvae. The experiments were performed in triplicate (n = 3).
After 24 h of contact between the A. salina larvae and the sample solutions, the percent mortality was calculated. The LC 50 value was calculated using semilogarithmic interpolation by converting mortality percentages into probits [62]. For the control, a naphthoquinone extracted from the bark of several species of plants of the genus Tabebuia (Bignoniaceae), lapachol, which has wide biological activity against different organisms, was used as a positive standard [64].
Next, molecular dynamics simulations were carried out on the docked complexes using the Desmond 2022 software from Schrodinger, LLC. The OPLS-2005 force field [68,69] and TIP3P water molecules were used in an explicit solvent model [70], and Na + ions were added to balance the charge. NaCl solutions were also included to mimic physiological conditions. The system was equilibrated using an NVT ensemble for 10 ns and an NPT ensemble for 12 ns. The Nose-Hoover chain coupling approach was used to set up the NPT ensemble and the variable temperature. During the simulations, a time step of 2fs was used, and the Martyna-Tuckerman-Klein chain coupling technique was used to manage pressure [71]. The final production run lasted for 100 ns, and the stability of the MD simulations was monitored using root-mean-square deviation (RMSD), radius of gyration (Rg), root-mean-square fluctuation (RMSF), H-bonds, salt bridges, and SASA calculations [72].

Statistical Analysis
Chemometric analysis was performed on the basis of two multivariate analyses, i.e., heatmap clustering consisting of hierarchical cluster analysis (HCA) and principal component analysis (PCA). Multivariate analyses were performed using OriginPro 2023 version 10.0.0.154 (Learning Edition).

Conclusions
In the present work, it was observed that the class of phenylpropanoid compounds, and the major substances isoosmorhizole, (E)-Isoosmorhizole, isomer-2-Methoxy-4.5methylenedioxypropiophenone, and 2-Methoxy-4,5-(methylenedioxy)-propiophenone, may be responsible for the potential toxicity of antioxidants. In general, natural antioxidants may be promising inhibitors of oxidative stress; however, it is important that the toxicity of these molecules be low even in small concentrations. Furthermore, the results of in silico studies demonstrated that the main constituents present in the essential oil fractions interact with the molecular target (catalytic site) of AChE. This is the first report on the molecular interaction of the compounds present in the essential oils of P. marginatum in AChE; in addition, this P. marginatum chemotype may be an important source of bioactive compounds, so further studies need to be carried out to explore the biological potential of this species. We also recommend studies in human cells to measure the toxicity of this essential oil.