The root essential oil from the Tunisian endemic plant Ferula tunetana: Chemical composition, biological evaluation, molecular docking analysis and drug-likeness prediction

2.1 Plant material

The roots of F. tunetana were harvested in Kroussia (Sousse, Tunisia) at the end of March 2020. After authentication at the Bioresources: Integrative Biology and Valorization Laboratory, Higher Institute of Biotechnology of Monastir, University of Monastir, Tunisia, by the botanist, Professor Fethia Harzallah-Skhiri, a specimen of the collected sample was stored at the herbarium of the Laboratory of Heterocyclic Chemistry, Natural Products and Reactivity (LR11ES39), Faculty of Sciences of Monastir, Tunisia, with a voucher specimen number FT-20.

2.2 Hydrodistillation of essential oil

The fresh roots (1.5 kg) were divided into small pieces and submitted to hydrodistillation for 180 min on a Clevenger apparatus in a thermostatically controlled heating. The obtained REO was decanted, dried, weighted, and conserved in the dark into a freezer until further uses.

2.3 Gas Chromatography–Mass spectrometry analysis

As reported by Baccari et al. (2020), the hydrodistilled essential oil was diluted to 0.5% in HPLC-grade n-hexane and then injected into a GC–MS apparatus. Gas chromatography–electron impact mass spectrometry (GC–EIMS) analyses were performed with a Varian CP-3800 gas chromatograph (Agilent Technologies Inc., Santa Clara, CA, USA) equipped with an Agilent DB-5 capillary column (Agilent Technologies Inc., Santa Clara, CA, USA) capillary column (30 m × 0.25 mm; coating thickness 0.25 μm) and a Varian Saturn 2000 ion trap mass detector (Agilent Technologies Inc., Santa Clara, CA, USA). Analytical conditions were as follows: injector and transfer line temperatures 220 and 240 °C, respectively; oven temperature programmed to rise from 60 to 240 °C at 3 °C/min; carrier gas helium at 1 mL/min; injection of 1 μL (0.5% HPLC grade n-hexane solution); split ratio 1:25. Acquisition parameters were as follows: full scan; scan range: 30–300 m/z; scan time: 1.0 sec. The identification of the constituents was based on a comparison of the retention times with those of pure, authentic samples, comparing their linear retention indices relative to the series of n-hydrocarbons (C5-C25). Computer matching was also used against commercial (NIST, 2014) and laboratory-developed mass spectra library built up from pure substances and components of commercial essential oils of known composition and MS literature data (Adams, 2007).

2.4 Radical scavenging activity

2.5 Antimicrobial activity

2.6 Cytotoxic activity

The cytotoxic activity bioassay of REO was evaluated against five human cancer cell lines: two colorectal (HT-29 and HTC 116), cervical (HeLa), lung (A549) and leukemia (U937) cells. The antiproliferative potential was assessed by application of the method described by Jabrane et al. (2011). The cells used in this study were grown in entire RPMI medium complemented with 10% (v/v) fetal calf serum and antibiotics (penicillin and streptomycin). Culture was carried out in flasks placed in an incubator at 37 °C and maintained in a completely humidified atmosphere.

Cells were seeded at an initial concentration of 6,000 or 10,000 cells per well in 96-well plates and handled with growth medium for one day, so that the cells attached well and adhered, then treated with various concentrations of the tested REO. After 48 h, 20 μL of MTT solution (5 mg/mL in PBS solution) was added to the wells and the mixture was incubated for 4 h. All the mixtures were dissolved in 200 μL DMSO. At 570 nm, the optical density was spectrophotometrically determined. Doxorubicin was used as positive control. The tests were replicated twice and results were represented as IC₅₀ (µg/mL) values (concentration inhibiting 50% of cell growth). After plotting the percent inhibition versus the log of concentrations, the linear regression equation of the resulting trend line was displayed to determine the IC₅₀ value.

2.7 Molecular docking procedure

The chemical structures of the REO constituents of F. tunetana were generated and optimized using the software ACD (3D viewer) (https://www.filefacts.com/acd3d-viewer-freeware-info). The enzymatic crystal structure of the human DNA topoisomerase IIα in complex with etoposide (PDB: 5GWK) was downloaded from the protein data bank (https://www.rcsb.org). Before starting, water molecules and the complexed inhibitor ligand (etoposide) were removed. Then, the polar hydrogens were added to the protein structure. Hence, the grid box in the target enzyme has a spacing of 0.375 Å. The grid box, with dimensions of 40 × 40 × 40 points and centered with the coordinates x: 23.35, y: −38.58, and z: −59.57, was generated based on the etoposide binding position at the DNA topoisomerase IIα (PDB: 5GWK). The molecular docking analysis of the main REO components at the protein-binding site was performed using AutoDock Vina software (Trot and Olson, 2010). Ligand-enzyme interactions were construed with the Biovia Discovery Studio Visualizer (BIOVIA, D. S. (2017)).

2.8 Drug-likeness prediction procedure

Physicochemical properties and Lipinski’s “rule of five” parameters of main detected compounds were predicted by the online computational tool ADMETlab 2.0 (https://admetmesh.scbdd.com/). Pharmacokinetic properties and bioavailability scores were estimated using SwissADME online server (https://www.swissadme.ch/). Bioactivity scores were predicted through the online Chemoinformatics tools of Molinspiration software (https://www.molinspiration.com). On the other hand, the toxicity risk prediction of the picked compounds was assessed using the OSIRIS Property Explorer (https://www.organic-chemistry.org/prog/peo/).

2.9 Statistical analysis

Statistical analysis was performed using GraphPad Prism 7.0. Data are presented as the mean ± Standard Deviation (SD). The difference between two groups was evaluated using Student's t-test. The difference among three groups was determined by one-way ANOVA with a post hoc analysis (Tukey test).

3.1 Chemical composition

The pale yellow and strong-smelling REO was obtained with a 0.06% (w/w) hydrodistillation yield. The identified constituents, their retention indices, and their relative percentages are illustrated in Table 1. In total, nine volatile compounds were identified, accounting for 94.5% of the total composition. They all belonged to the sesquiterpene chemical class, among which the oxygenated ones (82.8%) dominated the REO composition, while sesquiterpene hydrocarbons were far less represented (11.7%). β-Caryophyllene (1,9.3%) β-copaene (2) (1.1%), and α-humulene (3) (1.3%) were the only sesquiterpene hydrocarbons identified in this essential oil. On the other hand, REO was chiefly composed of caryophyllene oxide (4, 33.9%), α-cyperone (9, 13.9%), 14-hydroxy-9-epi-(E)-caryophyllene (8, 12.3%), T-muurolol (6, 8.2%), α-cadinol (7, 7.6%) and humulene epoxide II (5, 6.9%) (Fig. 1).

Close

Fig. 1

Chemical structures of the major compounds identified in Ferula tunetana root essential oil (REO).

Export to PPT

It is rather interesting to mention that caryophyllene oxide (4) and α-cyperone (9), which acquires elevated biological potentials, were detected in higher proportions in the REO of the F. tunetana. Caryophyllene oxide (4) is a well-known oxygenated sesquiterpene, which has enormous applications in medications, nutrition and cosmetics as a protective. It has also a several pharmacological and biological activities, such as anti-inflammatory, antifungal, analgesic and anticholinesterase (Savelev et al., 2003; Chavan et al., 2010; Karakaya et al., 2020; Gyrdymova and Rubtsova, 2021; Jassal et al., 2021; Schofs et al., 2021; Moghrovyan et al., 2022). α-Cyperone (9), has been denoted as the main constituent of various essential oils from aromatic plants and acquires diverse pharmacological activities like anti-inflammatory, antifungal and anti-tumor, so it influences a large range of biological pathways implicated in cancer treatment (Huang et al., 2018; Pei et al., 2020; Horn and Vediyappan, 2021). To the best of our knowledge, there is no previous data on the detailed chemical composition of this species root EO. On the contrary, many previous reports regarding the chemical composition of Ferula genus root EO of different geographical origins are available. Arjmand and Dastan (2020) reported that camphene (19.05%) and α-pinene (16.9%) were the major compounds of F. haussknechtii REO from Iran. In the same region, δ-cadinene (18.27%) was the main component of F. aucheri (Ahmadi et al., 2020). Another report by Shatar (2005) revealed that guaiol (58.8%) represents the major compound of Mongolian F. ferulaoides. Al-Ja’fari et al., 2011 also stated α‐pinene as the major component, representing 43.3% of the total composition, in the EO of F. hermonis from Jordan. Nevertheless, numerous studies on the Ferula genus reported the presence of caryophyllene oxide (4), the prevalent compound of our REO, as a remarkable constituent such as in F. glauca L. (14.3%) (Maggi et al., 2009), F. communis L. (8.0%) (Nguir et al., 2016), F. vesceritensis Coss. & Dur. (7.6%) (Labed-Zouad et al., 2015), and F. szovitsiana D.C. (4.1%) (Dehghan et al., 2007). However, the variation of the chemical profile is affected by diverse abiotic factors, such the climate, soil nature, luminosity, harvest time, and temperature (Silva et al., 2020; Yakhlef et al., 2020). Moreover, the chemical composition of other organs of F. tunetana species has been studied during the last decade in our laboratory. It was observed that the essential oil extracted from the roots of F. tunetana (REO) presented a very limited number of constituents compared to the EO of the flowers of the same species studied before in our laboratory. Only five common compounds have been identified (β-caryophyllene, α-humulene, caryophyllene oxide, humulene epoxide II, and α-cadinol), with the common predominance of oxygenated sesquiterpenes (Baccari et al., 2020). On the other hand, a different chemical composition was noticed in the seed EO of this species, also previously studied in our laboratory, consisting of a high amount of monoterpene hydrocarbons (77.3%) with α-pinene (39.8%) and β-pinene (11.5%) being the main compounds (Znati et al., 2017). This may be due to the fact that various plant organs acquire different enzymes, which adjust the biosynthesis of the secondary metabolites (Tian et al., 2019). Meanwhile, the results of the present work seem to propose that its roots could be considered as a new natural source of sesquiterpenes-rich EO.

3.2 Antioxidant activity

The root EO of F. tunetana showed IC₅₀ values at various levels against four in vitro tests (DPPH, ABTS⁺ radicals, O₂^• and H₂O₂ assays). Table 2 presents the results of REO as well as the standards used as reference: ascorbic acid, eugenol, and BHT. As shown, the EO displayed a remarkable potential. The highest effect was seen with DPPH test, it was able to reduce the deep purple hued, stable radical 2,2-diphenyl-1-picrylhydrazyl (DPPH^•) to the yellow tinted diphenylpicrylhydrazine, with an IC₅₀ value of 30.13 ± 0.28 µg/mL, followed by the the superoxide anion assay (IC₅₀ = 42.87 ± 0.81 µg/mL). On the other hand, the studied REO reduces H₂O₂ with an IC₅₀ value of 48.03 ± 1.21 µg/mL, whereas it reduces ABTS^•+ with an IC₅₀ value of 61.40 ± 1.67 µg/mL. The antioxidant properties were probably related to the chemical profile, particularly to the high relative percentage of oxygenated sesquiterpenes (82.8%), such as caryophyllene oxide (4), α-cyperone (9) and 14-hydroxy-9-epi-(E)-caryophyllene (8) (Lemouchi et al., 2017; Abd-ElGawad et al., 2021). More precisely, the obtained antioxidant potency of REO can be assigned chiefly to the presence of caryophyllene oxide in high proportions (Coté et al., 2017; Figueiredo et al., 2019; Nafis et al., 2019; Cascaes et al., 2022). In fact, caryophyllene oxide, the predominant constituent of REO, has formerly been reported as a compound with remarkable antioxidant potential in DPPH assay, displaying an IC₅₀ value of 102.5 μg/mL (Wan Salleh et al., 2015; Sarikurkcu et al., 2018). Other less abundant components, such as, α-humulene (3) and β-copaene (2), can also synergistically participate to the reached antioxidative activity (Mischko et al., 2018; Abd-ElGawad et al., 2020). The aforementioned constituents can be the accountable for the predestined antioxidant activity; however, the synergistic effect of other compounds cannot be undervalued, in agreement with that previously stated for numerous other Ferula species. Indeed, among many examined natural antioxidants, those originate from Ferula species have attracted a quite interest. Firstly, worth noting that the antioxidant potential of EOs is related to the geographic location as well as the species types and the harvesting time (Kavoosi and Rowshan, 2013). Study of Sharopov et al., (2019) on the measure of antioxidant potential of F. tadshikorum M. Pimen EO using DPPH, ABTS and FRAP (ferric reducing antioxidant power) tests suggested weak radical scavenging (IC₅₀ = 17.8 and 8.2 mg/mL for DPPH and ABTS, respectively), while the FRAP value was 1072.4 µM Fe(II)/mg of the oil sample. Topdas et al. (2020) reported the antioxidant activity of F. orientalis L. by means of DPPH and ABTS radical scavenging assays. Results gave IC₅₀ values of 50.04 ± 0.12 μg/mL and 80.78 ± 0.10 μg/mL for DPPH and ABTS tests, respectively. The chloroform extract of F. caspica M. Bieb. displayed a weak ABTS radical scavenging activity (88.57 ± 1.24 mg Trolox/g extract), compared to the ethyl acetate extract of the same species (268.28 ± 1.84 mg Trolox/g extract) (Kahraman et al., 2019b). In a similar way, the methanol extract of F. rigidula Fisch. ex DC. also showed low antioxidant activity in the DPPH radical scavenging capacity (19.91 ± 0.55 mg TE/g extract) compared to the scavenging ABTS radical result which was more potent (49.25 ± 3.34 mg TE/g extract). This radical scavenging effect can be attributed, at least partially, to the presence of phenols and flavonoids in the extracts (Zengin et al., 2020). Although the cited studies report results on the antioxidant capacity of other Ferula species, this study can be deemed as the first itemized statement on the evaluation of antioxidant activity of F. tunetana root EO. On the contrary, there is a previous report on the antioxidant properties of the EO extracted from the seeds of the same species, which showed a moderate potential against H₂O₂ and O₂^• with IC₅₀ values of 78.2 ± 2.98 and 89.2 ± 3.82 µg/mL, respectively (Znati et al., 2017).

3.3 Antimicrobial activity

The antimicrobial activity of F. tunetana REO was evaluated by the measurement of its IZD (inhibition zone diameter), MIC (minimum inhibitory concentration), and MBC (minimum bactericidal concentration). Gentamicin and Amphotericin B were used as positive controls. As shown in Table 3, the development of the tested Gram-negative bacteria was inhibited by the root EO, with IZD ranging from 14.6 ± 0.5 to 16.7 ± 0.4 mm, with the exception of P. aeruginosa ATCC 27853 (8.1 ± 0.0 mm). On the other hand, among the tested bacteria, S. typhimurium ATCC 13311 was the most sensitive strain against the REO, with IZD value of 16.7 ± 0.4 mm. However, the Gram-positive bacteria were rather more resistant to the EO, with IZD varying from 12.1 ± 0.2 to 15.9 ± 0.4 mm. In particular, the tested EO was efficient against B. subtilis ATCC 6633, B. cereus ATCC 14579, M. luteus NCIMB 8166, and S. aureus ATCC 25923, with IZD values of 15.9 ± 0.4, 15.2 ± 0.2, 15.2 ± 1.8, and 15.2 ± 2.2 mm, respectively. With regard to Candida yeasts, the F. tunetana REO inhibited the growth of C. albicans ATCC 90028 and C. glabrata ATCC 90030, with IZD values of 13.8 ± 1.5 mm and 14.8 ± 0.5 mm, respectively.

Based on the scale cited by Radünz et al. (2019) (strong antimicrobial potential: MIC ≤ 0.5 mg/mL; moderate potential: 0.6 ≤ MIC ≤ 1.5 mg/mL; weak potential: 1.6 mg/mL ≤ MIC), it appears reasonable to presume that the REO exhibited a prominent antimicrobial activity, with MIC values ranging from 0.039 to 0.625 mg/mL. Considering these values, we noticed that E. coli ATCC 25922 from the Gram-negative rods and S. epidermidis CIP 106510 from the Gram-positive strains were found to be the most sensitive to REO, with the same MIC value (0.039 mg/mL). Afterwards, the MBC determination showed that REO exerted a bactericidal action (MBC/MIC < 4) against all the tested bacteria, with the exception of S. epidermidis CIP 106510 and M. luteus NCIMB 8166, towards which REO showed only a bacteriostatic efficacy (MBC/MIC > 4), and C. albicans ATCC 90028, against which it showed a fungistatic effect. Our results were found to be in agreement with those of Maggi et al. (2009), reporting that F. glauca L. leaf EO, a sesquiterpene-rich oil (73.1%), had notable antibacterial activity against B. subtilis ATCC 6633, with MIC value of 0.078 mg/mL. It is important to mention that caryophyllene oxide (14.3%) is one of the main constituents of this EO, like the composition of the present study. Another study on the antimicrobial potential of F. haussknechtii root EO reported its significant antibacterial activity against S. aureous, S. epidermidis, and B. pumilus (Arjmand and Dastan, 2020). Karakaya et al. (2019), too, reported the importance of caryophyllene oxide and β-caryophyllene in the EO of the aerial parts of F. orientalis and H. microcarpumen in significant proportions (23.1 and 31.4%, respectively) as antimicrobial agents against C. albicans and S. aureus, in accordance with the present results. Moreover, according to Coté et al. (2017), caryophyllene oxide separately tested from Tanacetum vulgare L. EO was among the main responsible for the antimicrobial activity against E. coli (IC₅₀ = 97 ± 2 µg/mL) and S. aureus (IC₅₀ = 10.4 ± 0.9 µg/mL), while the crude EO was less effective against the same two microorganisms with IC₅₀ values of 241 ± 13 and 59 ± 5 µg/mL, respectively. Therefore, the antimicrobial activity of REO could be attributed to the high concentration of caryophyllene oxide (33.9%). other studies have shown the importance of β-caryophyllene and its isomer, α-humulene, as antibacterial agents (Sabulal et al., 2006; Ullah et al., 2021). Thus, the presence of β-caryophyllene (9.3%) and α-humulene (1.3%) in REO may also explain the antimicrobial potential against the studied bacteria. Usually, it is difficult to match the antimicrobial potential to a single specific compound because of the great number of compounds constituent the EOs. Nevertheless, the mechanism of cell death has been explained in different studies: the antimicrobial operation was linked to the dissolution of EO compounds in the lipids present in the membrane of microbial cell and mitochondria by dint of their lipophilic nature, which leads to a cell damage by disturbing the cell membrane and causing ionic leakage (Ksouda et al., 2019). Eventually, our results were added to those obtained by Znati et al. (2017), which revealed an interesting antimicrobial activity of F. tunetana seed EO rich in monoterpene hydrocarbons (77.3%) against the same bacterial and fungal strains.

3.4 Cytotoxic activity

In this investigation, the cytotoxic activity of the extracted EO was estimated against the human cervical cancer HeLa, human lung cancer A549, human leukemia U937, and human colon cancer HCT-116 and HT-29 cell lines. Results of REO cytotoxicity potential are given in Table 4.

As stated by the American National Cancer Institute, samples with IC₅₀ values up to 30 µg/mL against the target cells can be considered as effective anticancer agent for furthermore therapies improvement (Zardi-Bergaoui et al., 2020). The obtained results revealed that REO displayed a significant cytotoxicity against HeLa, A549 and U937 cell lines with IC₅₀ values of 3.37 ± 0.02, 8.17 ± 0.29 and 13.82 ± 0.16 μg/mL, respectively, but it was less active towards HT-29 (IC₅₀ = 21.13 ± 0.16 μg/mL) and HCT-116 (IC₅₀ = 46.66 ± 1.22 μg/mL) colon cells. The HeLa cancer cell line seems particularly sensitive to REO treatments. This finding may be explained by its high caryophyllene oxide (4) content (33.9%). Several reports were found regarding the anticancer activities of this compound, as well as its derivatives against various cancer cell lines, including A549, HeLa, HCT-116 and HT-29 cells (Wang et al., 2002; Fidyt et al., 2016; Hanušová et al., 2017; Zardi-Bergaoui et al., 2019; Delgado et al., 2021). Moreover, its distinctive chemical profile is wealthy in natural known anticancer chemical components such as β-caryophyllene (1, 9.3%), α-cadinol (7, 7.6%), and α-humulene (3, 1.3%). It has been reported that these compounds restrained the cell growth of OEC-M1 and A549 cell lines with IC₅₀ values varying from 9.9 to 31.3 μg/mL (Su and Ho, 2016). More generally, sesquiterpenes have proven their effectiveness as anticancer agents against human lung A549 and colon DLD-1 cell lines (Sylvestre et al., 2005), and different mechanisms of sesquiterpenes anticancer potential were investigated. Their ability to intensify the oxidative stress in damaged cells is one of them (Ambrož et al., 2019). The actual results are in compliance with those of other researchers who revealed that Ferula genus suggests a considerable potency as natural chemotherapeutic agents. The root EO of F. lutea showed good cytotoxic potential (IC₅₀ = 26.39 ± 3.98 μg/mL) against the HT-29 cell line (Ben Salem et al., 2016). Likewise, breast MCF7, cervical HeLa and liver HePG2 cell lines were sensitive to F. tingitana L. EOs with IC₅₀ values ranging from 4.3 and 10.9 μg/mL (Elghwaji et al., 2017).

3.5 Molecular docking studies

DNA topoisomerase enzymes are important enzymes that play essential roles in cellular processes such as transcription, replication, recombination, repair, and catalysis of DNA cleavage. In particular, topoisomerase IIα is preferentially secreted in proliferating cells and considered as the target enzyme for killing cancer cells, on which they depend on this enzyme more than healthy cells, since they divide more rapidly (Lynch et al., 1997; Cowell et al. 2012; Arencibia et al., 2020; Toan et al., 2020). DNA topoisomerase IIα is found in all organisms, from bacteria to human, and even in some viruses (Liu et al. 2018). This enzyme cleaves and relegates DNA to regulate DNA topology and constitutes a major class of antibacterial and anticancer drug targets (Bax et al. 2010). Topoisomerase IIα inhibitors represent a targeted approach to highly proliferative cells and therefore are of particular interest (Lynch et al., 1997; Cowell et al. 2012; Arencibia et al., 2020; Toan et al., 2020). The crystal structure of the human topoisomerase IIα is composed of two chains, A and B, with a sequence length of 806 amino acids (https://www.rcsb.org/structure/5GWK).

Extensive molecular docking studies were executed on the main REO components (1, 4–9), to explore the binding mode and rationalize the observed in vitro antimicrobial and cytotoxic activities of the EO of F. tunetana roots. The structures of the major EO constituents (1, 4–9) and etoposide (the standard ligand) were docked within the hydrophobic binding site of the etoposide (EVP) of the protein crystal structure of the human DNA topoisomerase IIα enzyme (PDB: 5GWK) (https://www.rcsb.org/structure/5GWK).

Interestingly, compounds 1, 4–9 behaved as expected. They are perfectly localized in the active pocket of the target enzyme. The binding positions of etoposide (the standard reference) and the major REO components in the structure of the human DNA Topoisomerase IIα enzyme are shown in Fig. 2.

Close

Fig. 2

3D representations of the binding positions of the docked molecules into the hydrophobic binding pocket of EVP in PDB: 5GWK. Compounds 1, 4–9 and Etoposide are indicated by red sticks, double strand DNA is displayed in pink and Topoisomerase IIα enzyme is shown as a grey ribbon.

Export to PPT

The binding modes of the etoposide and compounds (1, 4–9) within the human DNA Topoisomerase IIα enzyme (Fig. 3) are compared below.

Close

Fig. 3

3D representations of the binding modes of compounds (1, 4–9) and etoposide into the hydrophobic binding pocket of EVP in PDB: 5GWK.

Export to PPT

The binding mode of the etoposide, the standard inhibitor ligand of the DNA topoisomerase IIα enzyme, showed that it is involved in fifteen non-covalent interactions with seven amino acids. It formed a conventional hydrogen bond with the residue ASP-463, four carbon hydrogen bonds with CD-8, CT-9 and GLY-462, six π-π stacked bonds with DC-8, DT-9 and DG-13, an amide-π stacked binding with ARG-487, an alkyl bond with MET-766 and two π-alkyl bonds with ARG-487 and DT-9 residues (Fig. 3).

The β-caryophyllene (1) interacted with five amino acids (PRO-439, LYS-440, PHE-484, PRO-485 and ARG-487). The β-caryophyllene and the standard ligand, etoposide, share interactions with the residue ARG-487, which indicated that it is in the correct binding pose. It has been implicated just in a π-alkyl and seven alkyl bonds (Fig. 3).

The major compound, caryophyllene oxide (4, 33.9%), an oxygenated sesquiterpene, was implicated in eleven non-covalent interactions with five amino acids. Therefore, its binding mode shows that it is properly in the bonding pose and it interacts by three conventional hydrogen bonds with DC-8 and DG-13, four alkyl bonds with ARG-478, and four π-alkyl interactions with DT-9, DA-12 and DG-13 (Fig. 3).

The last five compounds (5–9) are also oxygenated sesquiterpenes, which constitute a large fraction of the REO (48.9%). Therefore, the molecular insight of docking analysis suggested that all of them exhibited a remarkable conventional hydrogen bond with the residue ASP-463 (Fig. 3). Moreover, the binding mode of humulene epoxide II (5), into the target hydrophobic binding site, showed that it is implicated in nine other non-covalent interactions with five residues. Therefore, it formed a π-σ bond with DG-13, three alkyl bonds with ARG-487, and five π-alkyl bonds with DC-8, DT-9 and DG-13 (Fig. 3). More of the conventional hydrogen binding with ASP-463, T-muurolol (6) was involved in three alkyl interactions with the ARG-487 amino acid and eight π-alkyl bonds with DC-8, DA-12, and DG-13. Also, the binding mode of α-cadinol (7) showed that it is involved in conventional hydrogen binding with ASP-463, two alkyl bonds with ARG-487 and six π-alkyl bindings with DC-8, DT-9 and DG-13 residues. The molecular insight of docking analysis suggested that, other than the conventional hydrogen binding with ASP-463, 14-hydroxy-9-epi-(E)-caryophyllene (8) interacts with two alkyl bonds with ARG-487 and five π-alkyl bindings with DA-12 and DG-13. The binding analysis suggests that α-cyperone (9) forms a conventional hydrogen bond with the ASP-463, a carbon hydrogen bond with GLY-462, two alkyl bonds with ARG-487 and seven π-alkyl bindings with DC-8, DA-12 and DG-13 (Fig. 3).

What is very important is that the conventional hydrogen interaction with ASP-463 is consistent with the single hydrogen bond established by the standard ligand (EVP) with the same residue (ASP-463). The major compound (4) interacts in particular via three conventional hydrogen bonds, which makes it the most interesting. In addition, the standard ligand (EVP) and the seven docked sesquiterpenes (1, 4–9) share an important number of interactions with the two residues ARG-487 and DG-13 (Fig. 3). Therefore, compounds 1, 4–9 are in the perfect bonding position (Figs. 2 and 3). All the aforementioned findings indicate that the binding modes of the docked sesquiterpenes (1, 4–9) are very similar to that of the DNA topoisomerase IIα standard inhibitor (EVP) and clearly rationalize the in vitro cellular toxicities of the root EO.

3.6 Drug-likeness prediction

3.6.1

3.6.1 Physicochemical properties and Lipinski's “rule of five”

The evaluation of ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) properties is considered as key step in drug development processes. Hence, a drug-like product is absorbed within a desired time frame and distributed throughout the body for an efficient metabolism (Bruna et al., 2022). On the other hand, ADMET assessment can be concluded after predicting physicochemical properties and obeying Lipinski's “rule of five” (Ye et al., 2018; Horchani et al., 2022). The physicochemical parameters of a molecular structure, namely molecular weight (MW), molecular volume (MV), Log of aqueous solubility (logS), Log of octanol/water partition coefficient (logP), number of hydrogen bond acceptors (nHA), number of hydrogen bond donors (nHD), number of rings (nRing), number of rotatable bonds (nRB), topological polar surface area (TPSA), percentage of absorption (%Abs) and number of Violations according to Lipinski's “rule of five” (nVLip5R), are important in predicting the efficacy, the safety, and the drug-likeness properties of a selected compound (Znati et al., 2019; Saidi et al., 2022). Physicochemical parameters and drug-likeness scores are associated with aqueous solubility, intestinal permeability, oral bioavailability and therefore pharmaceutical properties (Saidi et al., 2022).

The drug-likeness of a selected chemical compound can be predicted with the Lipinski's “rule of five”. Lipinski's “rule of five” specifies that most “drug-like” molecules have molecular weight ≤ 500, logP ≤ 5, number of hydrogen bond donors ≤ 5, and number of hydrogen bond acceptors ≤ 10. Indeed, the molecular weight describes the size, the permeability, the transport and the absorption of molecules (https://www.molinspiration.com). LogP value is the best parameter to characterize the molecular hydrophobicity. It is very important in the Quantitative Structure-Activity Relationship (QSAR) study because it is included in the distribution of molecules between fatty and aqueous phases in biological systems (Znati et al., 2019; Saidi et al., 2022). Besides, the ability to give and accept hydrogen bonds indicates the quantification and efficiency of the compound under study to form hydrogen bonds, which improves its interactions and its stability within the active pocket of the target enzyme. Furthermore, bioavailability problems can be reached when more than one of the “rule of five” parameters is violated (https://www.molinspiration.com).

In addition, the analysis of the physicochemical parameters, calculated online with ADMETlab 2.0 (https://admetmesh.scbdd.com/), revealed that all main compounds of the studied REO had scores within the norms and obeyed the Lipinski's “rule of five” (Table 5). As shown in Fig. 4, the bioavailability radar chart, obtained online with SwissADME (https://www.swissadme.ch/), of the major sesquiterpenes identified in REO exhibited significant scores, better than the used standard references. The selected components are entirely in the pink zone, indicating their higher oral bioavailability and drug-likeness. All the predicted sesquiterpenes have molecular weight < 500, therefore, they can be easily absorbed diffused and transported (Znati et al., 2019; Saidi et al., 2022). On the other hand, the octanol/water partition coefficient (logP) of all studied compounds is less than five. Thus, they have a good binding into hydrophobic pocket of enzymes and a best distribution and excretion by metabolism (Saidi et al., 2022). Moreover, all the main components of REO exhibited no violations of the Lipinski's rules, whereas, the used standard references, except for ascorbic acid, violated at least two Lipinski's rules (Table 5 and Fig. 4).

Table 5 Predicted Physicochemical Properties and Lipinski Parameters of the major compounds identified in Ferula tunetana root essential oil (REO) and the used standard references.

N°	Acceptable value	MW a	MV b	logS c	logP (o/w) d	nHA e	nHD f	nRing g	nRB h	TPSA i	%Abs j	nVLip5R k
	Compound name	(100 ∼ 600)	(5 0 0)	(-4 ∼ 0.5)	(0 ∼ 3)	(≤12)	(≤7)	(≤6)	(≤11)	(≤140)	(100%)	(≤1)
1	β -caryophyllene	204.190	245.610	−5.810	5.341	0	0	2	0	0.000	109	1
4	caryophyllene oxide	220.180	248.481	−5.296	4.536	1	0	3	0	12.530	104.677	0
5	humulene epoxide II	220.180	254.401	−5.580	4.564	1	0	2	0	12.530	104.677	0
6	T-muurolol	222.200	257.037	−3.718	4.723	1	1	2	1	20.230	102.021	0
7	α -cadinol	222.200	257.037	−4.998	4.619	1	1	2	1	20.230	102.021	0
8	14-hydroxy-9-epi-(E)-caryophyllene	220.180	254.401	−4.064	3.565	1	1	2	1	20.230	102.021	0
9	α -cyperone	218.170	251.764	−4.083	3.929	1	0	2	1	17.070	103.111	0
Ascorbic acid (SR)		176.030	151.244	−0.613	−1.420	6	5	1	2	114.290	69.570	0
Gentamicin (SR)		477.320	462.618	−0.970	−1.878	12	11	3	7	199.730	40.093	2
Amphotericin B (SR)		923.490	932.501	−2.932	−0.249	18	13	3	3	319.610	−1.265	3
Doxorubicin (SR)		543.170	516.727	−2.458	2.325	12	7	5	5	212.390	35.725	3
Etoposide (SR)		588.180	546.063	−3.573	1.687	13	3	7	5	160.830	53.514	2

(SR) Standard reference, ^a Molecular weight, ^b Molecular volume, ^c Log of the aqueous solubility, ^d Log of the octanol/water partition coefficient, ^e Number of hydrogen bond acceptors, ^f Number of hydrogen bond donors, ^g Number of rings, ^h Number of rotatable bonds, ⁱ Topological polar surface area, ^j Percentage of absorption (%Abs = 109 − [0.345 × TPSA]), ^k Number of violations according to the Lipinski “rule of five”.

Close

Fig. 4

Bioavailability radar chart of compounds (1, 4–9) identified in Ferula tunetana root essential oil (REO) and the used standard references. The pink zone is the suitable physicochemical space for oral bioavailability. LIPO (lipophilicity): −0.7 < XLOGP3 < +5.0, SIZE: 150 g/mol < MV < 500 g/mol, POLAR (polarity): 20 Å² < TPSA < 130 Å², INSOLU (insolubility): −6 < Log S (ESOL) < 0, INSATU (insaturation): 0.25 < Fraction Csp³ < 1, FLEX (flexibility): 0 < Num. rotatable bonds < 9.

Export to PPT

The physicochemical and drug-likeness properties, predicted from the structures of the main compounds of the studied REO, confirm the significant antimicrobial and cytotoxic activities of the tested EO. Therefore, the obtained results suggest that the REO of the Ferula tunetana plant may offer therapeutic uses in the future.

3.6.2

3.6.2 Prediction of pharmacokinetic properties

Pharmacokinetic properties are from the best virtual screening filters of bioactive chemical compounds (Bruna et al., 2022; Zafar et al., 2020). To investigate the viability of the major sesquiterpenes (1, 4–9) detected in the studied REO, the online tool SwissADME (https://www.swissadme.ch/) was used to predict these pharmacokinetic descriptors. Through the analysis of the Boiled-Egg graph (Fig. 5), it is evident that all oxygenated sesquiterpenes (4–9), exhibit high gastrointestinal absorption (GI_a) and blood–brain barrier permeability (BBB_p). Since the P-glycoprotein (P-gp) is a membrane protein that removes compounds from cells, it is critical that drug-like compounds are not P-gp substrates. In this respect and according to the estimates made, the tested sesquiterpenes 1, 4–9 fulfilled this condition. The studied compounds (1, 4–9) have good skin permeability coefficients (Log Kp) from −5.53 to −4.44 (cm/s), which gives them an effective permeability (Liu et al., 2020a; Kirk et al., 2022) to treat skin diseases, better than the used standards. Therefore, and as small hydrophobic chemicals, compounds 4–9 have high membrane permeability and they cannot be cleared by P-gp (Bruna et al., 2022).

Close

Fig. 5

Boiled-egg graph of compounds (1, 4–9) identified in Ferula tunetana root essential oil (REO) and the used standard references. Points located in Boiled-Egg’s yolk are molecules predicted to passively permeate through the blood–brain barrier. Points located in Boiled-Egg’s white are molecules predicted to be passively absorbed by the gastrointestinal tract. Blue dots are for molecules predicted as substrates of P-glycoprotein. In contrast, the red dots correspond to molecules that not predicted to be P-glycoprotein substrates.

Export to PPT

On the other hand, cytochromes P450 (CYP), also derived from the pharmacokinetic-related proteins, are involved in the oxidative metabolism of compounds, where the oxidized molecules become polar and can thus be excreted, especially by the kidneys within a given time. Noticeably, the studied components interacted with at most two isoenzymes CYP2C19 and CYP2C9 (Table 6), confirming their effectiveness with some secondary toxicities (Alminderej et al., 2020; Bruna et al., 2022). Thus, this finding is consistent with the cellular toxicity obtained with the in vitro assays. Moreover, the bioavailability scores of 0.55 value indicate their more drug-like properties and the high feasibility of use as drugs (Table 6).

Table 6 Predicted Pharmacokinetic Properties and Bioavailability Scores of the major compounds identified in Ferula tunetana root essential oil (REO) and the used standard references.

N°	Compound name	GIa	BBBp	P-gps	CYP1A2i	CYP2C19i	CYP2C9i	CYP2D6i	CYP3A4i	Log Kp	Bioavailability Score
1	β -caryophyllene	Low	No	No	No	Yes	Yes	No	No	−4.44	0.55
4	caryophyllene oxide	High	Yes	No	No	Yes	Yes	No	No	−5.12	0.55
5	humulene epoxide II	High	Yes	No	No	No	Yes	No	No	−4.91	0.55
6	T-muurolol	High	Yes	No	No	No	No	No	No	−5.29	0.55
7	α -cadinol	High	Yes	No	No	Yes	No	No	No	−5.29	0.55
8	14-hydroxy-9-epi-(E)-caryophyllene	High	Yes	No	No	Yes	No	No	No	−5.53	0.55
9	α -cyperone	High	Yes	No	No	Yes	Yes	No	No	−4.93	0.55
Ascorbic acid (SR)		High	No	No	No	No	No	No	No	−8.54	0.56
Gentamicin (SR)		Low	No	Yes	No	No	No	No	No	−12.12	0.17
Doxorubicin (SR)		Low	No	Yes	No	No	No	No	No	−8.71	0.17
Etoposide (SR)		Low	No	Yes	No	No	No	Yes	No	−9.46	0.17

(SR) Standard reference, GI_a: Gastrointestinal absorption, BBB_p: Blood-brain barrier permeant, P-gp_s: P-glycoprotein substrate, CYP1A2_i: Cytochrome P450 1A2 inhibitor, CYP2C19_i: Cytochrome P450 2C19 inhibitor, CYP2C9_i: Cytochrome P450 2C9 inhibitor, CYP2D6_i: Cytochrome P450 2D6 inhibitor, CYP3A4_i: Cytochrome P450 3A4 inhibitor, Log Kp: skin permeation (cm/s).