INTRODUCTION
Asteraceae is a large family of flowering plants with worldwide distribution, especially in the Mediterranean, Eastern Europe, and Asia Minor. Several Asteraceae plants were reported to possess significant medicinal importance, being used for various ailments, and contain beneficial compounds like flavonoids and phenolic acids, contributing to their therapeutic properties (1). Asteraceae plants exhibit significant pharmacological activities, likely due to their rich content of flavonoids, phenolic acids, and other polyphenolic metabolites. These properties suggest their therapeutic potential in managing inflammatory conditions (2) and diabetes (3, 4). Aaronsohnia is a small genus under the family Asteraceae native to North Africa and the Middle East. Little research has previously documented only two Aaronsohnia species: A. factorovskyi and A. pubescens, for some medicinal and commercial applications. A. factorovskyi was reported for antimicrobial and anthelmintic efficacy as well as for the biosynthesis of silver nanoparticles (5, 6). Other studies have explored the phytochemical composition, antioxidant, antidiabetic, and antimicrobial activities of A. pubescens (7, 8). A. pubescens was also reported for its ability to inhibit steel corrosion in an acidic environment (9). Hence, Aaronsohnia looks like an interesting candidate for pharmaceutical research.
The LC-MS analysis enabled the phytochemical profiling of several Asteraceae plant extracts in which various metabolites, e.g., flavonoids, phenolic acids, terpenoids and saponins have been detected (10, 11).
Based upon the previous statements, the current research aims to explore the phytochemical composition of A. factorovskyi extract, using both qualitative and quantitative methods. It will screen the antioxidant, antidiabetic, and anti-inflammatory activity of A. factorovskyi using in vitro testing while also conducting in silico studies to correlate the observed bioactivity to the detected phytoconstituents.
EXPERIMENTAL
Plant material
A. factorovskyi (Asteraceae) was collected from the Qassim area during March 2023. A specimen labeled QPP-119 was kept at the College of Pharmacy, Qassim University, Saudi Arabia. The aerial parts of the plant were cleaned, dried, and then powdered. The powdered plant material (500 g) was extracted using aqueous methanol (80 %, 4 × 500 mL). The filtered extracts were pooled and dried under vacuum at a temperature not exceeding 50 ºC using a rotary evaporator.
Chemicals
Aluminum chloride, ammonium molybdate, ascorbic acid, 1,1-diphenyl-2-picrylhydrazyl (DPPH), Folin-Ciocalteu reagent, gallic acid, potassium acetate, quercetin, sodium carbonate, sodium phosphate, methanol, acetonitrile, formic acid, ascorbic acid, ibuprofen, and acarbose were purchased from Sigma-Aldrich Chemical Co. (USA). All used chemicals were of analytical grade.
Total phenolic content (TPC) and total flavonoid content (TFC)
TPC was evaluated using the Folin-Ciocalteu method according to the reported method (12). The results were expressed as mg gallic acid equivalents (GAE) per gram dry extract. TFC was assessed using aluminum chloride according to the reported method (12). The results were expressed as mg quercetin equivalents (QE) per gram dry extract.
UPLC-ESI-MS analysis
Metabolite profiling of the methanolic extract of A. factorovskyi aerial parts was performed using UPLC-ESI-MS analysis, following a previously established protocol (13). The extract was analyzed utilizing CQUITY UPLC, equipped with a BEH C18 column (1.7 µm, 2.1 × 50 mm) connected to an XEVO TQD triple quadruple mass spectrometer (Waters Corporation, USA). The sample was solubilized in methanol of HPLC grade and filtered using a PTFE (polytetrafluoroethylene) membrane with a 0.2-µm pore size. The solvent system is composed of 0.1 % formic acid in water (A) and 0.1 % formic acid in acetonitrile (B), with the flow rate of 0.2 mL min–1. Gradient elution was as follows: 90 % A (0–2 min), from 90 % A to 70 % A (2–5 min), from 70 % A to 30 % A (5–15 min), from 30 % A to 10 % A (15–22 min), then isocratic at 10 % A (22–25 min), from 10 % A to 0 % A (25–26 min), and finally, isocratic at 100 % B (26–29 min).
Mass spectrometric analyses were performed in both positive and negative electrospray ionization (ESI) modes. The source temperature was maintained at 150 °C, with a desolvation temperature of 400 °C. Nitrogen was used as desolvation and cone gas at (600 and 50 L h–1) flow rates, resp. The spray voltage used was 4.48 kV, the tube lens voltage was 10.00 V, cone voltage was 30.00 V, while the capillary voltage was 3.0 kV. Spectra were acquired over an m/z range of 100–1200.
Metabolites were tentatively identified by comparing their obtained mass and spectral data with previously reported data for the genus and family, as well as with online public databases (see Table SI). Data acquisition and analysis were managed using MetFrag (https://msbi ipb-halle de/MetFrag/).
Antioxidant activity
DPPH assay(12)and phosphomolybdenum assay (14) were carried out according to the reported methods. The total antioxidant capacity (TAC) was expressed as mg ascorbic acid equivalents (AAE) per gram dry extract.
Anti-inflammatory activity
The inhibitory potentials of A. factorovskyi crude extract, as well as the reference compound ibuprofen, against cyclooxygenase enzymes were assessed utilizing an enzyme immunoassay (EIA) human COX-2 and COX-1 inhibitor screening kits (Cayman Chemical, USA).
Antidiabetic activity
The antidiabetic potential of A. factorovskyi extract was evaluated in vitro by assessing the extract's inhibitory activity against digestive enzymes, including α-glucosidase and α-amylase.
The α-glucosidase inhibitory assay was carried out in compliance with the kit’s protocol instructions (Biovision, USA). For the assay, 10 µL of the extract (at various concentrations), or the reference drug (acarbose), was combined with 10 µL of a diluted α-glucosidase enzyme solution (2 µL enzyme with 38 µL α-glucosidase assay buffer) and 10 µL of assay buffer in a 96-well clear plate. The total volume was adjusted to 80 µL with the buffer, mixed thoroughly, and incubated at room temperature for 15–20 minutes. Finally, 20 µL of the enzyme substrate (p-nitrophenyl-α-D-glucopyranoside, PNPG) was mixed, and the absorbance was measured at 410 nm; the concentration for 50 % inhibition for the extract and acarbose was calculated.
The α-amylase inhibitory capacity of the extract was determined in vitro according to the directions in the kit protocol (Biovision). The extract or the standard (50 µL each) was mixed with 50 µL of the enzyme solution, mixed well, and incubated for ten minutes at room temperature. Then the starch solution (50 µL) was added and incubated for 3 min, followed by the addition of 50 µL 3,5-dinitrosalicylic acid (DNS) reagent to terminate the reaction. Then the reaction mixture was boiled at 85–90 °C in a water bath for ten minutes, cooled to room temperature, and the absorbance was read at 405 nm.
Molecular docking
Virtual screening software Pyrx (version 0.8), https://pyrx sourceforge io (15) that uses Autodock Vina, has been utilized for the molecular docking study (Molecular Graphics Laboratory, The Scripps Research Institute, USA). Metabolite structures were downloaded from PubChem (16) [February, 2024], and Chem Bio 3D (Chem Bio Office Ultra 12.0 suite) was used to minimize their energies. PDB entries for human lysosomal acid α-glucosidase enzyme (5NN8) (17) and human pancreatic α-amylase enzyme (4W93)18() were retrieved from Protein Data Bank (https://www rcsb org/). Grid centers were set to be as follows: X = –14.650, Y = –28.868, Z = 94.112 for α-glucosidase, and X = –9.872, Y = 8.881 and Z = –21.291 for α-amylase. The binding scores for the best poses were tabulated, and the interactions were visualized. Pymol software (https://pymol.org/) was used to convert pdbqt files to pdb format, followed by visualization of the interactions in 3D and 2D views using Biovia Discovery Studio visualizer v21.1.0.20298 (Dassault Systèmes Biovia Corp., USA).
All the data were expressed as mean ± standard deviation (SD) from three experiments. The data of TPC, TFC, and antioxidant activity were calculated from linear calibration curves.
RESULTS AND DISCUSSION
Chemical composition of A. factorovskyi aerial parts
As shown in Table I, the total phenolic content was determined to be 52.46 ± 5.93 mg GAE g–1, while the total flavonoid content was calculated as 19.01 ± 2.50 mg QE g–1. Previous investigations of TPC and TFC in A. pubescens aerial parts reported higher phenolic and flavonoid contents (121.5 ± 2.0 GAE g–1 and 79.3 ± 5.5 QE g–1, resp.) in the methanolic extract (7).
Table I. Total phenolic and flavonoid contents of A. factorovskyi aerial parts
Mean ± SD, n = 3.
a mg gallic acid equivalent in 1 g of dry sample
b mg quercetin equivalent in 1 g of dry sample
TFC – total flavonoid content, TPC – total phenolic content
Analysis of the UPLC-ESI-MS positive and negative modes spectra allowed tentative identification of 19 compounds from 14 chromatographic peaks (Table II, Figs. 1 and 2). The identified compounds were classified as eight flavonoids, four sesquiterpenes, one lignan, two fatty acids, two saponins, and two phenolics.
Table II. Tentative metabolite profiling of A. factorovskyi aerial parts methanolic extract using UPLC-ESI-MS
N – negative mode of ionization, NI – not identified, P – positive mode of ionization
Fig. 1. Total ion chromatograms of the methanolic extract of the A. factorovskyi aerial parts using UPLC-ESI-MS: a) positive mode; b) negative mode.
The eight flavonoids were identified from the mass ion peaks at m/z 431.2256, 447.2209, 447.1292, 317.3847, and 287.9257. The mass ion peak at 431.2256 [M-H]– was in agreement with the molecular formula C21H20O10 and could be annotated as kaempferol-7-O-rhamnoside, vitexin, or isovitexin. As for Asteraceae, kaempferol-7-O-rhamnoside19() was previously isolated from Carduus nutans L., while vitexin and isovitexin20() from North American Trugopogon species. The mass ion peak at 447.2209 [M-H]– is in agreement with the molecular formula C21H20O11 and was annotated as quercitrin, whereas the peaks at 447.1292 [M+H]+ and 445.1909 [M-H]– matched with C21H18O11 that could be annotated as baicalin. Quercitrin (21) was previously detected in Solidago chilensis Meyen (Asteraceae), and baicalin (22) was isolated from Scutellaria baicalensis (Lamiaceae) using high-speed counter current chromatography. Isorhamnetin was annotated from the mass ion peak at 317.3847 [M+H]+ (C16H12O7), isorhamentin (23, 24) was previously isolated from Cichorium species and marigold flowers (Asteraceae). Finally, the mass ion peak at 287.9257 [M+H]+ matched with the molecular formula C15H10O6 which could be annotated as kaempferol or luteolin. Kaempferol (25, 26) was previously characterized in Parthenium and Clibadium species (Asteraceae), while luteolin (27) was formerly detected using LC-MS in the flower of Chrysanthemum morifolium (Asteraceae).
Four sesquiterpene compounds were annotated from the signals at m/z: 251.1711 [M+H]+ (C15H22O3), which matched with arbusculin A (28), earlier detected in Artemisia arbuscula (Asteraceae), or warburganal (29) that was previously identified in the ethyl acetate fraction of Warburgia stuhlmannii root (Canellaceae). The mass ion peaks at 235.1724 [M+H]+ matched with C15H22O2, identified as valerenic acid (30) previously also reported in Valeriana officinalis L. (Caprifoliaceae), and 293.1759 [M-H]– in agreement with C17H26O4 recognized as phytuberin (31) that was formerly isolated from the leaves of Nicotiana tabacum (Solanaceae).
One lignan was detected at m/z 381.1144 [M+H]+ and 379.1395 [M-H]–, which are in agreement with C21H16O7 and identified as diphyllin. Diphyllin (32, 33) was previously isolated from Justicia procumbens (Acanthaceae).
Fig. 2. Chemical structures of possible metabolites in A. factorovskyi aerial parts.
Two phenolic compounds were annotated in the negative mode from m/z at 181.0433, which matched with C9H10O4 and were identified as syringaldehyde or homovanillic acid. Syringaldehyde and homovanillic acid (34) were both identified in the Asteraceae plant Picris rhagadioloides (L.) Desf.
Also, two saponins, yamogenin 3-O-neohesperidoside and convallasaponin A, were identified from the signals at m/z [M-H]– 721.4780 (C39H62O12) and 579.3938 (C32H52O9), resp. Yamogenin 3-O-neohesperidoside (35) was previously identified in Borassus flabellifer (Arecaceae), and convallasaponin A (36) was identified in Convallaria majalis (Liliaceae).
The two signals at m/z: 277.2374 [M+H] + and 399.3374 [M-H]– were in agreement with C18H28O2 and C22H44O2 and were annotated as fatty acids, stearidonic acid and docosanoic acid, resp. Stearidonic acid (37) and docosanoic acid (38) were previously detected in Ibicella lutea and Proboscidea louisiana (Martyniaceae).
It is noteworthy that all the above-mentioned metabolites have been identified for the first time from A. factorovskyi extract, thereby enriching the phytochemical library of the genus Aaronsohnia. Still, this preliminary data needs to be confirmed.
Biological activity of A. factorovskyi aerial parts
The antioxidant potential of A. factorovskyi extract evaluated by DPPH and phosphomolybdenum assay indicated TAC of 12.31 ± 2.33 and 17.40 ± 0.96 mg AAE g–1, resp. (Table III). It is worth mentioning that this is the first report for the assessment of the antioxidant capacity of A. factorovskyi extract. In the previous research on A. pubescens, antioxidant activity was attributed to the high content of phenolics and flavonoids (7).
Table III. Antioxidant activity of A. factorovskyi aerial parts using DPPH and phosphomolybdenum assay
Mean ± SD, n = 3.
a mg ascorbic acid equivalent in 1 g of dry extract.
DPPH – 2,2-diphenyl-1-picrylhydrazyl, TAC – total antioxidant capacity
COX enzymes play a key role in prostanoid production, where COX-1 is constitutively expressed to support gastric protection and platelet aggregation, and COX-2, produced in response to inflammation. The current study assessed the anti-inflammatory potential of A. factorovskyi extract utilizing the COX inhibition assay. Results in Table IV indicate characteristic activity of the extract against COX-1 and COX-2, with concentration for 50 % inhibition of 20.85 ± 0.73 and 8.25 ± 0.29 µg mL–1, resp. (respective IC50 for ibuprofen is 6.195 ± 0.23 and 2.398 ± 0.09 µg mL–1). Metabolic profiling of the extract revealed flavonoids as prevailing metabolites in A. factorovskyi aerial parts. Flavonoids are known for their anti-inflammatory activity that is attributed to variable mechanisms, including COX enzyme inhibition (39). Consequently, the observed anti-inflammatory effect of the tested extract may likely be attributed to its flavonoid content.
Table IV. Antidiabetic and anti-inflammatory activity of A. factorovskyi aerial parts
Mean ± SD, n = 3.
a Selectivity index (SI) = conc. 50 % inhibitionCOX-1/ conc. 50 % inhibitionCOX-2
b Selectivity index (SI) = IC50 COX-1/IC50 COX-2
The extract was tested for its inhibition of the metabolic enzymes (α-glucosidase and α-amylase) to assess its antidiabetic potency. The results of α-glucosidase and α-amylase inhibition (Table IV) indicated extract concentration for 50 % inhibition as 0.243 ± 0.009 mg mL–1 (IC50 for acarbose is 0.418 ± 0.015 mg mL–1) and 0.275 ± 0.01 mg mL–1 (IC50 for acarbose is 0.168 ± 0.006 mg mL–1), resp.
Molecular docking study
To highlight the possible potential of antidiabetic phytoconstituent(s) in A. factorovskyi extract, a molecular docking study was conducted, where the detected metabolites were docked against α-glucosidase and α-amylase enzymes. The comparative values for binding scores of the docked compounds against both enzymes are given in Table V. Results have been compared with those for the co-crystallized ligands (acarbose for α-glucosidase and montbretin A for α-amylase).
Table V. The docking scores of tentatively identified metabolites towards α-glucosidase and α-amylase
a Tentative identification.
ND – not docked into the active site
The binding score is an estimate of the free energy of binding in kcal mol–1 between the ligand and the receptor; the more negative the score, the stronger and more favorable the binding is considered. It is a combination of various terms into a scoring function, such as steric interactions, hydrophobic interactions, and hydrogen bonding, where all contributions are summed to give a final relative score (40). Among the docked compounds yamogenin 3-O-neohesperidoside, convallasaponin A and baicalin unveiled notably strong binding affinities for both α-glucosidase and α-amylase binding sites with binding energies of –8.2, –8.0 and –8.0 kcal mol–1, resp., for α-glucosidase, and –9.9, –9.9 and –9.1 kcal mol–1, resp., for α-amylase enzyme, compared to the co-crystallized ligand of α-glucosidase acarbose, –7.0 kcal mol–1, and that of α-amylase montbretin A, –9.3 kcal mol–1.
The sugar moiety in yamogenin 3-O-neohesperidoside was accommodated in the active site of α-glucosidase and its hydroxyl groups interacted via hydrogen bonds with the acid-base residue Asp616 and the nucleophile Asp518 and via sulfur bond with Met519, while its triterpenoid nucleus established hydrophobic interactions with Trp376 and with Leu677 and Leu678 positioned at the broad groove that leads to the active site (Fig. 3a). Hydroxyl groups in sugar moiety of the steroidal saponin convallasaponin A were involved in hydrogen bond interactions with Leu283 and Ala284, while the acid-base residue Asp616 and Arg411 formed hydrogen bonds with OH group at C-5 and oxygen of the lactone ring in the steroid nucleus, and further stabilization has been accomplished via hydrophobic interactions between Trp376, Trp481 and Leu650, and the steroidal saponin backbone (Fig. 3b). On the other hand, the ring B of the flavonoid baicalin has been positioned close to the entry of the active site via π-π stacking with Trp376 and π-alkyl interactions with Leu677 and Leu678 while the sugar moiety was accommodated in the active site and interacted via hydrogen bonds with Trp481, Asp518, Asp616 and His674, in addition to π-sulfur bond between ring A and Met519 (Fig. 3c).
Fig. 3. 2D (right) and 3D (left) depictions of docked poses and interactions of: a) yamogenin 3-O-neohesperidoside; b) convallasaponin A; c) baicalin, in the active site of α-glucosidase enzyme. Ligands are depicted in green stick model, labeled residues in grey line models, representing hydrogen bonds by green dashed lines, hydrophobic interactions by pink and purple dashed lines, and sulfur bonds by yellow dashed lines.
Regarding α-amylase, yamogenin 3-O-neohesperidoside bound to α-amylase and the interactions involved hydrogen bonding with the catalytic residue Gln63, with further stabilization via a number of hydrophobic interactions with Tyr151, Ile235, Leu237, and Ala307 (Fig. 4a). The steroidal saponin convallasaponin A exhibited multiple interactions with the enzyme. Its hydroxyl groups on the nucleus formed hydrogen bonds with Glu233, a critical active site residue involved in catalytic activity, and with Lys200. Additionally, carbon-hydrogen bonding was observed with Glu240. Furthermore, the steroidal saponin backbone was engaged in hydrophobic interactions with a surface comprising Tyr151, Leu162, His201, Ile235, Leu237 and Ala307 (Fig. 4b). The flavonoid baicalin revealed five hydrogen bonds with Tyr151, Arg195, His201, Glu233 and Ile235, two π-anion bonds with Asp300, two π-π stackings with Tyr59 and Tyr62, and π-alkyl interaction with Leu165 (Fig. 4c). It is noteworthy that many natural triterpenoids have been reported to act against diabetes and various in vitro investigations unveiled the potential of triterpenoids from plant sources to inhibit α-glucosidase and α-amylase and to form hydrogen bonds with important residues in the active sites of both enzymes (41). Furthermore, plenty of investigations have reported the impact of saponins and flavonoids in the management of diabetes by different mechanisms (42–44).
Fig. 4. 2D (right) and 3D (left) depictions of docked poses and interactions of: a) yamogenin 3-O-neohesperidoside; b) convallasaponin A; c) baicalin in the active site of α-amylase. Ligands are depicted in green stick model, labeled residues in grey line models, representing hydrogen bonds by green dashed lines, hydrophobic interactions by pink and purple dashed lines, and ionic bonds by orange dashed lines.
CONCLUSIONS
The present research provides the introductory phytochemical and biological exploration of the aerial parts of A. factorovskyi, adding a contribution to the phytochemistry of this less frequently studied species. Preliminary insight into complex composition and metabolite profiling using UPLC-ESI-MS analysis highlighted some phytoconstituents. In vitro testing indicated antioxidant, anti-inflammatory, and antidiabetic activities of the plant extract. Molecular docking further supported these findings by recording strong interactions between the active sites of key enzymes and the three metabolites: yamogenin 3-O-neohesperidoside, convallasaponin A, and baicalin. These findings highlight the potential medicinal value of this plant and recommend further detailed phytochemical and in vivo testing.
Abbreviations, acronyms, symbols. – AAE – ascorbic acid equivalent, COX – cyclooxygenase enzymes, DPPH – 1,1-diphenyl-2-picrylhydrazyl, GAE – gallic acid equivalent>, IC50 – half maximal inhibitory concentration, QE – quercetin equivalent, TAC – total antioxidant capacity, TFC – total flavonoid content, TPC – total phenolic content, UPLC-ESI-MS – ultra-performance liquid chromatography-electrospray ionization mass spectrometry.
Availability of data and materials. – Supplementary material is available upon request.
Acknowledgments. – The authors would like to thank Dr. Esam Rashwan, Head of the confirmatory diagnostic unit VACSERA-EGYPT, for carrying out the in vitro α-glucosidase, α-amylase, and COX inhibitory assays. Additionally, we would like to thank Mr. Ibrahim Aldubayan for plant authentication. Additionally, the authors would like to acknowledge the Centre of Drug Discovery Research and Development, Department of Pharmacognosy, Faculty of Pharmacy, Ain Shams University, Egypt, for carrying out UPLC-ESI-MS analysis.
Conflict of interest. – The authors declare that they have no competing interests.
Funding. – Not applicable.
Authors contributions. – Conceptualization, E.A., A.H.E., and M.H.A.H.; investigation, R.A., N.H.A., R.A., G.L.A., D.F.A., A.H.E., M.H.A.H., and E.A.; methodology, A.H.E., M.H.A.H., and E.A.; software, A.H.E., M.H.A.H., and E.I.A.M.; original draft preparation, A.H.E., M.H.A.H., E.I.A.M., and E.A.; writing, review, and editing, A.H.E., M.H.A.H., and E.A. All authors have read and agreed on the final version of the manuscript.