Chemical profiling by LC–MS/MS and GC–MS and biological activity assessment of different extracts of Portulaca oleracea through in-vitro and in-silico approaches

Abstract

In this study, the phenolic compound profile and essential oil composition of Portulaca oleracea L. collected from the Bulancak (Giresun) region of Türkiye were investigated in detail by LC–MS/MS and GC–MS techniques. The obtained data were then correlated with multifaceted biological activity tests and in-silico molecular modelling. In the scope of the study, the anti-diabetic (inhibition of α-amylase and α-glucosidase), anti-inflammatory (inhibition of protein denaturation) and anti-hemolytic effects were evaluated in conjunction with the chemical composition data. The results of the LC–MS/MS analysis indicate that p-coumaric acid is the predominant phenolic, with a concentration of 1228.10 mg/kg. The presence of other phenolics was also detected, including vanillin (5.12 mg/kg), caffeic acid (2.75 mg/kg), sesamol (1.87 mg/kg) and protocatechuic aldehyde (1.75 mg/kg). In the GC–MS analysis, hexahydrofarnesyl acetone was the most abundant component, accounting for 58.89% of the total. Subsequent analysis identified dillapiole (16.80%), myristicin (8.95%), and β-ionone (6.31%) as the active constituents. In the anti-diabetic activity test, the IC₅₀ values for α-amylase inhibition ranged from 352.24 to 623.45 mg/mL, and the strongest effect was observed in the ethanol extract. For the assessment of α-glucosidase inhibition, the IC₅₀ values were observed to range from 146.85 to 339.20 mg/mL, thus indicating a stronger inhibitory effect of α-glucosidase compared to α-amylase. The results of in-silico modelling provided support for the in-vitro data that had been obtained. It was shown that p-coumaric acid can interact strongly with both α-amylase (binding energy: −5.57 kcal/mol; inhibition constant: 83.05 µM) and α-glucosidase (binding energy: −5.84 kcal/mol; inhibition constant: 52.14 µM). Dillapiole demonstrated a noteworthy interaction with α-glucosidase, exhibiting a binding energy of -6.60 kcal/mol and an inhibition constant of 14.59 µM. In the protein denaturation inhibition test, which was used to investigate the anti-inflammatory effect, the highest effect was obtained in the chloroform extract (76.1% inhibition, IC₅₀ = 126 µg/mL). In the AAPH-induced hemolysis test, 200 µg/mL ethanol extract demonstrated a protection level of 65.17%, which was significantly higher than the equivalent dose of ascorbic acid (39.41%). In conclusion, this study demonstrates that the Giresun population of P. oleracea contains different type and levels of bioactive compounds, exhibits versatile biological activity, and could be an important resource for both functional food development and drug discovery. Furthermore, it was demonstrated that the extraction solvents employed (ethanol, chloroform, ethyl acetate, water) have a significant effect on the resulting compound profile and biological activity.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-24271-6.

Introduction

Portulaca oleracea L. is an annual plant belonging to the family Portulacaceae, which is distributed across the globe. The utilisation of P. oleracea in the domain of folk medicine can be attributed to its biological activities. It is listed as one of the most widely used medicinal plants by the World Health Organization and has been termed a “global panacea”^1–3. P. oleracea is predominantly an annual species, though it can exhibit perennial characteristics in tropical regions, attaining heights of up to 40 cm. The stems are hairless, fleshy and come in purple-red or green colours and often found hanging down. The cylindrical stem is up to 30 cm long, 2–3 mm in diameter, with swollen nodes, smooth, widely branched, and internodes 1.5–3.5 cm long. P. oleracea leaves are distinguished by their flat, fleshy texture and variable shape, ranging from 1–5 cm in length and 0.5–2 cm in diameter. The leaves exhibit a blunt or slightly notched apex, pointed base, and are either sessile or indistinctly petiolate. Their surface is characteristically glabrous, smooth, and waxy on the upper surface, with entire edges and small stipules. The leaves exhibit a length measurement of approximately 5–30 mm, and on occasion, they are distinguished by red edges⁴. The edibility of the plant under consideration is dual in nature, insofar as it may be consumed either raw or cooked. However, it has recently been defined as a “functional food”, a term used to describe foods that offer health benefits beyond simple nutrition. This is due to the high nutritional value of the plant, which is characterised by its high levels of vitamins A and C, and its low levels of carbohydrates^5,6.

P. oleracea demonstrates a high degree of variability in its chemical composition, a consequence of exposure to diverse environmental factors. The phytochemical content is subject to variation depending on the region of distribution. It has been documented that the nutritional composition of the plant, in terms of protein, lipid, carbohydrate/fibre, and ash, exhibits variability between different months of the year, even within the same geographic region. The findings of this study indicate that the conditions in which the plant is distributed and cultivated have a significant impact on its chemical composition. Environmental factors, including growing conditions (salinity, stress, etc.), harvest time, species origin, soil structure, and pollution, as well as pollution, have been demonstrated to cause significant changes in chemical composition⁷. A plethora of compounds have been isolated from P. oleracea, including flavonoids, alkaloids, fatty acids, terpenoids, polysaccharides, vitamins, sterols, proteins, and minerals¹. P. oleracea has a long history in folk medicine, where it is employed for its antipyretic and antiseptic properties in numerous countries. The pharmacological effects of the active compounds identified in the plant are extensive, including antibacterial, anti-ulcerogenic, anti-inflammatory, anti-oxidant, and wound-healing properties. These compounds have been shown to have numerous applications, including hemostatic, thermoregulatory, and detoxifying properties. The dried aerial parts of this plant have been found to be effective in the treatment of a range of ailments, including fever, dysentery, diarrhea, eczema, and hematochezia. The medicinal uses of P. oleracea include the relief of dry coughs and the treatment of wasp stings and snake bites.

The biological activities exhibited by P. oleracea, and the resulting medicinal uses to which it is attributed, are said to be a consequence of its active ingredients^5,6. Many studies have reported the presence of various flavonoids in P. oleracea, including kaempferol, myricetin, luteolin, apigenin, quercetin and genistin. It has been demonstrated that quercetin and kaempferol contribute to the therapeutic effects of P. oleracea on lung infections, immunological and allergic diseases^8,9. Myricetin, a polyphenol found in P. oleracea, has been reported to have therapeutic benefits in the treatment of steatohepatitis by inhibiting prostaglandin-endoperoxide synthase¹⁰. Miao et al.¹¹ reported that the presence of luteolin, kaempferol and quercetin in P. oleracea has been shown to provide anti-inflammatory effects. As demonstrated in the extant literature, a total of 29 different alkaloid contents, 26 of which are newly discovered, are reported in the phytochemistry of P. oleracea¹². It was determined that oleracamide A-C, isolated from P. oleracea, was not cytotoxic up to a concentration of 80 μM, while 50 μM oleracamide D concentration showed significant toxicity against SH-SY5Y cells^13–15. Oleraceins (A-E, X and Y), a class of alkaloids known as cyclo dopa-amides, are among the main alkaloids detected in P. oleracea. While oleracein A and oleracein B have been shown to possess antioxidant properties, oleracein E has been demonstrated to protect dopaminergic neurons against toxicity by reducing oxidative stress^15,16. Omega-3 fatty acids, which are generally regarded as a safeguard against cardiovascular diseases, are among the various organic acids found in P. oleracea¹². Also, Simopoulos et al.¹⁷ have demonstrated the presence of various types of omega-3 fatty acids, including alpha-linolenic acid, docosahexaenoic acid, eicosapentaenoic acid and docosapentaenoic acid. The presence of catecholamines, including dopamine and norepinephrine, has been detected in P. oleracea. Norepinephrine has been reported to be present at 0.074%, 0.029% and 0.054% in the leaves, stems and seeds, respectively¹⁸.

The present study aims to evaluate the comprehensive phytochemical and biological properties of P. oleracea samples collected from the Bulancak (Giresun-Türkiye). The total phenolic content (TPC) and total flavonoid content (TFC) of the plant were determined and the quantitative profile of phenolic compounds was analysed by LC–MS/MS, and the essential oil components were analysed by GC–MS to create a detailed chemical inventory. The anti-diabetic, anti-inflammatory, and anti-hemolytic activities of the plant material were evaluated within the scope of in-vitro biological tests conducted simultaneously with phytochemical analyses. In-silico molecular docking studies were conducted to elucidate the mechanistic contribution of major compounds contained in P. oleracea, which has ethnopharmacological importance, to enzyme inhibition. Particularly, the major compounds identified by LC–MS/MS and GC–MS were selected as ligands and their interactions with both α-amylase and α-glucosidase enzymes were studied in detail. Molecular docking provides a mechanistic framework for in-vitro enzyme inhibition results, allowing to predict which molecular interactions might contribute to the observed biological activities. This study aims to enrich the repertoire of bioactive compounds in local P. oleracea samples of Türkiye flora using LC–MS/MS and GC–MS based quantitative profiling, to use in-vitro biological activities together with in-silico binding models, to prioritize potential biological candidate compounds by converting observational/ethnopharmacological information into compound-activity relationships, to reveal the chemical variation among samples of the same species grown under different environmental conditions, and to provide a scientific basis for sustainable harvesting and bioprospecting strategies. In addition, the obtained findings provide initial candidate molecules for drug discovery and nutraceutical development processes. It should be emphasized that the study paves the way for further studies with comprehensive pharmacological evaluation (in-vivo activity, toxicology, ADME studies) and structural optimization of the isolated components. In this context, the study is expected to make significant contributions to both basic phytochemistry and applied pharmacology.

Materials and methods

Botanical authentication and comprehensive extraction workflow

P. oleracea samples were collected from the Bulancak district of Giresun (Türkiye) on 15 May 2023. The identification of the specimens was conducted by Professor Zafer Türkmen at Giresun University, Department of Botany and the specimens were subsequently included in the herbarium collection under the voucher number BIO-Pole-2023. Experimental research conducted with plant specimens, including the collection of plant material, complies with institutional, national, and international guidelines and legislation, and is permitted by local government. Subsequent to the extraction of roots and the removal of diseased and wilted parts, the plant material underwent a drying process in an oven set at 35 °C. Thereafter, it was ground to a fine powder using a grinder. The resulting 200 g of powder sample was extracted by maceration using the solvents water, ethanol, chloroform, and ethyl acetate. The maceration process was conducted for a duration of 24 h at ambient temperature in a shaking incubator. Subsequent to extraction, the solvent phase was separated from the solid particles by means of Whatman No. 4 filter paper. After the filtrate was centrifuged at 5000 rpm for 10 min, the upper layer obtained was evaporated using a rotary evaporator. The dry extracts obtained following the evaporation process were then collected and stored under appropriate conditions for use in biological activity analyses. The efficiency of extraction was calculated as the percentage ratio of the mass of dry extract (g) obtained to the mass of plant material used (g)¹⁹.

Quantitative assessment of total phenolic and flavonoid constituents

The total flavonoid content (TFC) of P. oleracea extract was determined spectrophotometrically. A mixture of 1 mL of 5% sodium nitrite and 10 mL of the extract was incubated for a period of 6 min. Thereafter, 1 mL of 10% aluminum nitrate was added. Following a 6 min. incubation period, 10 mL of NaOH was added, and the total volume was adjusted to 25 mL with dH₂O. The final mixture was then subjected to an incubation period of 15 min. after which the solution’s optical density was measured at a wavelength of 510 nm. Quercetin was utilised as the standard, and flavonoid quantities were expressed in terms of mg quercetin equivalents (QE)/g dry weight²⁰. The total phenolic content (TPC) of the P. oleracea extract was determined by means of the Folin–Ciocalteau method. A mixture containing 10 µL of H₂O, 10 µL of extract, 40 µL of Folin reagent, and 200 µL of 10% Na₂CO₃ was prepared and subsequently incubated for a period of 30 min. Subsequent to the process of incubation, the degree of absorption of the solution was measured at a wavelength of 725 nm, and the total phenol content was expressed as gallic acid equivalent (GAE) in mg GAE per gram of fresh weight²¹.

Quantification of phenolic constituents using LC–MS/MS

The quantitative determination of phenolic compounds in P. oleracea was performed using a liquid chromatography-tandem mass spectrometry (LC–MS/MS-Thermo Scientific/TSQ Quantum Access Max) analysis. For this purpose, 1 g of plant sample was extracted in a methanol–dichloromethane (4:1, v/v) solvent system, and the resulting extract was prepared for analysis by passing through a 0.45 µm pore diameter syringe filter. The analyses performed on a LC–MS/MS (Thermo Scientific) instrument involved the use of an ODS Hypersil column. The column oven temperature was optimised at 30 °C, the mobile phase flow rate was set at 0.7 mL/min, and the total analysis time was 34 min. Total ion chromatogram (Fig. S1), LC–MS program (Table S1) and ions in MS/MS (Table S2) are given in supplementary material file. The LC–MS/MS analysis was conducted at the HUBTUAM (Hitit University, Çorum-Türkiye).

Gas chromatography–mass spectrometry profiling of volatile oils

Essential oils obtained using a Clevenger-type hydrodistillation system was analyzed by GC–MS. After distillation, 1 mL of hexane was added to the essential oil samples, which were homogenized and transferred to glass vials for preparation for GC–MS analysis. Analyses were performed using a Thermo Scientific GC–MS system (TSQ Quantum GC Triple Quadrupole) on a TG–5MS (30 m × 0.25 mm, film thickness 0.25 µm) capillary column. Analysis conditions were optimized as follows: total analysis time 55 min, injection block temperature 250 °C, helium as carrier gas, and flow rate set at 1 mL/min. A detailed procedure for GC–MS analysis is provided in the supplementary material. Essential oil extraction and GC–MS analyses were conducted at the HUBTUAM (Hitit University, Çorum-Türkiye).

In-vitro α-amylase inhibition

The assay mixture contained 200 µL of sodium phosphate buffer (0.02 M), 20 µL of enzyme (0.25 mg/mL) and extract over a concentration range of 10–700 mg/mL. The mixture was incubated for 10 min. at room temperature, after which 200 µL of starch (5 mg/mL) was added to all test tubes. The addition of 400 µL of DNS reagent brought the reaction to a conclusion. The resultant mixture was then placed in a boiling water bath for 5 min., cooled, and diluted with 15 mL of distilled water. Absorbance of solutions were recorded at a wavelength of 540 nm. The control samples were prepared without the extract. The inhibition percentage was calculated using the Eq. 1. The IC₅₀ values were determined from plots of percent inhibition versus log inhibitor concentration, and were calculated by non-linear regression analysis from the mean inhibitory values. It is imperative to note that all tests were performed in triplicate²².

1

In-vitro α-glucosidase inhibition

Α-glucosidase inhibitory activity of P. oleracea extracts were determined by incubating 50 µL of sample with 100 µL of a solution containing α-glucosidase (1 U/mL) dissolved in 0.1 M phosphate buffer (pH 6.9) for a period of 10 min at 25 °C. Following the incubation period, 50 µL of 5 mM pNPG substrate solution prepared in 0.1 M phosphate buffer (pH 6.9) was subsequently added to each well. The reaction mixtures were subjected to an incubation period of five minutes at a temperature of 25 °C. Thereafter, the absorbances of the mixtures were measured at a wavelength of 405 nm. The inhibition test was repeated on three occasions for each extract. Acarbose was utilised as a positive control, and α-glucosidase inhibitory activity was calculated using Eq. 1. The percentage of inhibition was plotted as a function of inhibitor concentration, and the IC₅₀ value was expressed as the concentration at which 50% inhibition occurred²².

Docking-based interaction of major constituents with α-amylase

The α-amylase inhibitory activity of P. oleracea extract, as determined in-vitro, was further substantiated by an in-silico molecular docking approach. Three-dimensional crystal structure of the α-amylase was obtained from the Protein Data Bank (PDB) database (https://www.rcsb.org/). The α-amylase structure utilised in this study is the model with PDB ID:1SMD²³. The deficiencies in hydrogen atoms were rectified and the structural optimisation was accomplished by utilising the Maestro BioLuminate 5.0 software (Schrödinger Release, 2023). Dillapiole (PubChem ID: 10,231), hexahydrofarnesyl acetone (PubChem ID: 10,408), p-coumaric acid (PubChem ID: 637,542), and vanillin (PubChem ID: 1183), which are among the major compounds identified in P. oleracea, were selected as ligands. The three-dimensional structures of these compounds were obtained from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). The prepared macromolecules and ligand structures were then subjected to molecular docking using the AutoDock Vina (version 4.2.6) program²⁴. The resulting binding models and interaction details were then visualised using the Biovia Discovery Studio 2020 Client (BIOVIA, 2023) software.

The α-glucosidase inhibitory activity of P. oleracea extracts, as determined in-vitro, was further evaluated through an in-silico molecular docking approach. The three-dimensional crystal structure of α-glucosidase (PDB ID: 5NN8) was retrieved from the Protein Data Bank (https://www.rcsb.org/)²⁵. Deficiencies in hydrogen atoms were corrected, and structural optimisation was performed using the Maestro BioLuminate 5.0 software (Schrödinger Release 2023). Dillapiole (PubChem ID: 10,231), hexahydrofarnesyl acetone (PubChem ID: 10,408), p-coumaric acid (PubChem ID: 637,542), and vanillin (PubChem ID: 1183), identified as major phytoconstituents of P. oleracea were selected as ligands. The prepared macromolecule and ligand structures were subjected to molecular docking using AutoDock Vina version 4.2.6²⁴, and the resulting binding interactions were visualised and analysed using the Biovia Discovery Studio 2020 Client (BIOVIA, 2023).

Anti-inflammatory activity

The anti-inflammatory activity was determined using the serum albumin protein denaturation stabilization assay. A quantity of 1 mL of 1% bovine serum albumin, prepared in phosphate-buffered saline with a pH of 6.4, was added to 1 mL of extract at varying concentrations, ranging from 5 to 200 µg/mL. The mixture was left at room temperature for 20 min and then heated at 70 °C for 5 min. The resulting solution was then cooled at room temperature, after which the turbidity was measured at a wavelength of 660 nm. The procedure was repeated with double-distilled water and aspirin as the control and standard, respectively. The inhibition percentage of protein denaturation was calculated according to the Eq. 2.

2

Abs_control denotes the level of absorption in the absence of a sample, whereas Abs_sample represents the absorption of the sample (extract or standard). The IC₅₀ values were determined as the concentration corresponding to 50% inhibition²¹.

Anti-hemolytic activity

For anti-hemolytic activity determination, erythrocyte suspension (10% hematocrit) from a healthy donor was pretreated with different concentrations of the extract (50–200 μg/mL) in a 37 °C water bath for 30 min. Then, 500 μL of 2,2′-Azodiisobutyramidine dihydrochloride (AAPH) solution (200 mM) was added, and the reaction mixture was incubated under the same conditions for 4 h. The volume of all tubes was then adjusted to 4.5 mL with PBS and centrifuged at 2000 rpm for 10 min. The absorbance of the supernatant was read at 540 nm. The same procedure was repeated simultaneously with the standard substance ascorbic acid under the same conditions. The antihemolytic activity of the extracts was calculated using Eq. 3.

3

Statistical analysis

The analyses were conducted utilising the IBM SPSS Statistics 22 software program, and the resulting data were presented as the mean ± SD (standard deviation). Statistical significance between means was determined by Duncan’s test and One-way ANOVA, with a p value of less than 0.05 being considered statistically significant.

Results and discussion

Phytochemical yield and quantification of total phenolics and flavonoids

The highest extraction efficiency was obtained with ethanol solvent among the extractions using water, ethanol, chloroform, and ethyl acetate. The order of solvent performance in terms of extraction efficiency was as follows: ethanol (13.28%) > chloroform (13.15%) > ethyl acetate (9.96%) > water (8.11%). The ethanol extract was found to contain the highest TPC content, measuring 75.4 mgGAE/g. The quantity of TFC was found to be higher in the extracts than in the total phenolic compounds. The ethanol extract was found to contain the highest TFC content, with a reading of 99.6 mgGAE/g. Levels of TFC content obtained from the chloroform and ethyl acetate extracts were found to be similar. A number of studies have been conducted in the relevant literature, the aim of which was to investigate the TPC and TFC contents of P. oleracea. As reported by Lim et al.²⁶, the TFC contents of various P. oleracea cultivars obtained from Sungai Buluh exhibited a range from 127 to 478 mgGAE/100 g. In contrast, Uddin et al.^6,7 identified the highest concentrations of TPC and TFC in P. oleracea samples collected from Malaysia, specifically in the methanol extract, with a phenolic content of 360.3 ± 8.9 mgGAE/100 g and a flavonoid content of 49.2 ± 3.4 mgRE/g. The disparities between the levels documented in the extant literature and in the present study may be attributed to fluctuations in the levels of secondary metabolites produced by the plant during its adaptation to the prevailing conditions of its growth environment.

LC–MS/MS-based quantification of phenolics

The quantitative phenolic content of P. oleracea was determined by LC–MS/MS analysis, and the chromatogram and the presence rates of the main components are given in Fig. 1 and Table 1. The presence of p-coumaric acid was detected in major proportions in P. oleracea, with a level of 1228.10 mg/kg being determined. Vannilin (5.12 mg/kg), caffeic acid (2.75 mg/kg), sesamol (1.87 mg/kg) and protecatechuic aldehyde (1.75 mg/kg) were detected in minor quantities. It was determined that P. oleracea does not contain phenolic substances such as gallic acid, catechin, rutin, and resveratrol, which are commonly found in other plants. P. oleracea contains high levels of p-coumaric acid, a phenolic acid derivative that provides color, odor, and taste to plants and is derived from cinnamic acid. Depending on the position of the hydroxyl group, three different isomers can be distinguished: ortho, para, and meta. p-coumaric acid is a potent antioxidant that scavenges reactive oxygen species (ROS) and free radicals. In addition to these properties, it has been demonstrated to possess anti-diabetic, anti-hyperlipidemic, anti-cancer, anti-microbial, anti-inflammatory, anti-ulcer, anti-pyretic, analgesic, anti-arthritic, and neuroprotective effects²⁷. p-coumaric acid, a compound that has been demonstrated to possess a protective effect against stress, has been reported to exert a significant lethal effect on tumour cells. However, it is imperative to note that the ingestion of this acid in high quantities can result in deleterious effects. It has been established that p-coumaric acid exerts a significant beneficial effect in combating stomach cancer^28,29. Sesamol (3,4-methedioxyphenol) was detected in P. oleracea at a concentration of 1.87 mg/kg. Sesamol, a compound that has been shown to possess a range of biological activities including anti-microbial, anti-cancer, anti-angiogenic, immunomodulatory, cardioprotective and anti-oxidant properties, has also been observed to inhibit the growth of gram-positive bacteria (S. aureus and L. monocytogenes) and gram-negative bacteria (E. coli and Salmonella enterica)³⁰. Caffeic acid, which was detected at 2.75 mg/kg in P. oleracea, is also an important phenolic substance. Caffeic acid, also known as 3,4-hydroxy cinnamic acid, is a prevalent constituent of the human diet, found in numerous food sources. Caffeic acid has a wide range of uses in the pharmaceutical industry as an anti-inflammatory and antioxidant, as well as having a positive effect on the immune system^31,32. The presence of protocatechuic aldehyde, with a detection level of 1.75 mg/kg in P. oleracea, has been identified as a potent aldose reductase inhibitor. It has been reported that this has potential use in the prevention of diabetic complications^33,34.

Fig. 1.

Representative LC–MS/MS chromatogram of P. oleracea extract illustrating the separation and detection of bioactive compounds based on retention time and mass-to-charge ratio (m/z).

Table 1.

Amount of phenolics, Rt, R², LOD (mg/L), LOQ (mg/L) and RSD% values.

Rt	Phenolic substance	mg/kg plant	LOD	LOQ	RSD%	R2
9,68	Gallic acid	NF	0.061	0.203	4.5	0.9850
11,77	Protocatechuic acid	NF	0.049	0.162	4.9	0.9983
12,64	Protocatechuic aldehyde	1.75	0.026	0.087	2.2	0.9984
12,65	Sesamol	1.87	0.031	0.103	2.1	0.9961
12,91	Catechin	NF	0.010	0.032	3.1	0.9999
14,38	Epicatechin	NF	0.003	0.006	2.5	0.9988
14,76	Caffeic acid	2.75	0.042	0.058	2.5	0.9991
15,27	Vannilin	5.12	0.023	0.076	3.7	0.9994
15,77	Syringic asit	NF	0.194	0.647	4.6	0.9987
16,27	Taxifolin	NF	0.001	0.005	3.8	0.9994
16,42	P_Coumaric acid	1228.1	0.069	0.109	3.5	0.9985
17,19	Salicylic acid	NF	0.030	0.099	2.3	0.9989
17,31	Ferulic acid	NF	0.063	0.118	4.5	0.9985
17,34	4_OH benzoic acid	NF	0.243	0.809	4.1	0.9993
17,41	Rosmarinic acid	NF	0.011	0.036	2.6	0.9999
17,51	Oleuropein	NF	0.003	0.005	2.0	0.9991
18,03	Rutin	NF	0.022	0.073	1.6	0.9950
18,05	Resveratrol	NF	0.019	0.062	4.1	0.9986
19,16	Ellagic acid	NF	0.087	0.289	4.9	0.9962
20,29	Quercetin	NF	0.001	0.005	2.9	0.9983
21,55	Kaempherol	NF	0.092	0.306	3.9	0.9948
23,74	Flavone	NF	0.037	0.124	1.7	0.9995

NF: not found.

There are studies in the literature investigating the phenolic substances contained in P. oleracea. Xu et al.³⁵ detected the presence of kaempferol, apigenin, myricetin, quercetin, and luteolin in P. oleracea obtained from Fujian (China). Sicari et al.³⁶ detected the presence of caffeic acid, p-coumaric acid, ferulic acid, apigenin, kaempferol, luteolin, quercetin, isorhamnetin, kaempferol-3-O-glucoside, and rutin in P. oleracea grown in Reggio Calabria (Italy). Literature studies have reported the presence of similar phenolic compounds as well as different phenolic compounds in P. oleracea collected from different regions. There are also differences in the levels detected in studies reporting the presence of the same phenolic compound. This diversity, reported in literature studies investigating the phenolic compounds of P. oleracea, is closely related to the environment in which the plant grows. Plants produce different types and levels of phenolics to adapt to their environmental conditions. This causes variations in the phytochemical content of plants of the same species growing in different geographic regions^37,38.

Comprehensive GC–MS profiling of P. oleracea constituents

The GC–MS system facilitates the identification of substances contained within structures comprising diverse components, in addition to the mass chromatographic analysis of organic samples amenable to gasification or present in the gas phase. The essential oils obtained using the clevenger apparatus were analyzed by the GC–MS system, and the GC–MS chromatogram of P. oleracea, retention time (Rt) and yield% of the detected compounds are given in Fig. 2. Hexahydrofarnesyl acetone (58.89%) was detected at significant levels in P. oleracea by GC–MS analysis. Dillapiole (16.80%), myristicine (8.95%), and β-Ionone (6.31%) were the compounds detected at moderate levels. The levels of heptacosane, dotriacontane, and safranal detected by GC–MS were determined to be in the range of 1–2%. The presence of active compounds such as β-cyclocitral, β-damascenone, dodecane, cedrenol, and verrucarol was also determined at low doses (0.30%-0.84%). The combined effect of these compounds on P. oleracea has been shown to result in a variety of biological and pharmacological properties being exhibited by the plant. Hexahydrofarnesyl acetone, a sesquiterpene, is an essential oil that is detected in the major proportion. Hexahydrofarnesyl acetone has been shown to possess anti-bacterial, anti-nociceptive, and anti-inflammatory properties. The substance exhibits a broad-spectrum cidal effect by acting against fungi, gram-positive, and gram-negative bacteria. Hexahydrofarnesyl acetone, a compound that has been demonstrated to possess both anti-inflammatory properties and significant potential in the field of pain relief research, has also been shown to possess insecticidal activity^39,40. Dillapiole, detected at 16.80% in P. oleracea, is an important essential oil. Although dillapiole can also be found in other plants such as fennel root, it is an organic chemical compound that is most commonly obtained from P. oleracea. The methoxy group positioned on the benzene ring is of particular significance, as evidenced by the anti-inflammatory effects observed in numerous studies^41,42. Myristicin, detected at 8.95%, has been shown to be an effective chemopreventive agent that activates detoxification enzymes. It has been demonstrated that herbal extracts containing myristicin are capable of preventing the proliferation of the human leukemia cell line (Molt 4B). Myristicin has also been demonstrated to induce apoptosis via mitochondrial pathways⁴³. The active substances detected in P. oleracea by GC–MS have been shown to exhibit a range of biological activities. The resultant effect of these activities on P. oleracea is significant, with the capacity to manifest biological and pharmacological effects. A plethora of studies in the extant literature have reported disparate components as a consequence of GC–MS analysis of P. oleracea collected from a variety of geographical locations. Ma et al.⁴⁴ detected a total of 12 active ingredients in essential oils extracted from P. oleracea and the highest content was linalool (17.51%), followed by limonene (16.55%) and borneol (16.53%). Tleubayeva et al.⁴⁵ reported the presence of volatile and semi-volatile compounds terpinen-4-ol, cis-β-farnesene, spathulenol and hexahydrofarnesyl acetone extracted from the aboveground parts of P. oleracea collected from Kazakhstan.

Fig. 2.

Detailed GC–MS chromatogram of P. oleracea showing the separation and detection of volatile and semi-volatile compounds based on retention time (Rt) and yield (%).

In-vitro α-amylase and α-glucosidase inhibition

The IC₅₀ values of P. oleracea extracts for enzyme inhibiton are presented in Fig. 3, and the IC₅₀ value for the α-amylase enzyme of the tested extracts was determined to be within the range of 352.24–623.45 mg/mL. The lowest IC₅₀ value for α-amylase was obtained with the ethanol extract, followed by chloroform, ethyl acetate, and water, respectively. The IC₅₀ value is defined as the inhibitor concentration that results in a 50% reduction in enzyme activity. The finding that the lowest IC₅₀ value was obtained with the ethanol extract suggests that this particular extract is capable of providing stronger α-amylase inhibition and may possess stronger anti-hyperglycemic activity. This discrepancy in the solvents utilised during extraction processes may be attributable to the divergent solubilities of phytochemicals in the plant material when exposed to different solvents. It is evident that there is a discrepancy in the levels of enzyme inhibitory activity exhibited by P. oleracea plants cultivated in disparate regions. Sicari et al.³⁶ reported that the methanol extract of P. oleracea leaves obtained from Reggio Calabria (Italy) exhibited an IC₅₀ value of 640.01 mg/mL against α-amylase. The inhibitory effects of P. oleracea extracts on α-glucosidase enzyme are illustrated in Fig. 3. The inhibition of α-glucosidase activity was determined based on the IC₅₀ values, as 339.2 mg/mL, 245.67 mg/mL, 146.85 mg/mL, and 239.15 mg/mL for water, chloroform, ethanol, and ethyl acetate extracts, respectively. Decreased IC₅₀ values are indicative of increased anti-diabetic activity. As with α-amylase inhibition, the most effective activity in α-glucosidase inhibition was obtained with the ethanol extract, which had the lowest IC₅₀ value. A notable finding of the study was the observation that P. oleracea extracts exhibited superior inhibitory activity against α-glucosidase compared to α-amylase. The highest inhibition was observed in the ethanol extract, which inhibited α-glucosidase by 58.3% compared to α-amylase. In addition, studies in the relevant literature have reported that P. oleracea extract is capable of inhibiting α-amylase and α-glucosidase enzymes, with the inhibition of α-glucosidase being particularly effective. Sicari et al.³⁶ reported that the ethanol extract of P. oleracea leaves obtained from Reggio Calabria (Italy) exhibited an IC₅₀ value of 138.51 mg/mL against α-glucosidase and was more effective than α-amylase.

Fig. 3.

IC₅₀ value for the α-amylase and α-glucosidase extracts and main components.

The IC₅₀ value for α-amylase was determined to be 56.2 ± 3.9 mg/mL for kaempferol, one of the main components tested, while that for dillapiole was 71.9 ± 4.5 mg/mL. For α-glucosidase, kaempferol and dillapiole have IC₅₀ values of 69.7 ± 4.1 and 99.8 ± 5.7 mg/mL, respectively. This finding suggests that the primary constituents of the extract are more efficacious in enzyme inhibition in isolation, and that their potency is diminished in the presence of other components within the extract. It is evident that a multitude of factors may be responsible for the reduced activity observed in the extract in comparison to the main components. It is noteworthy that other components present in the extract may exert antagonism, impede access to the enzyme through competitive binding, or destabilize the active compounds. A number of compounds are known to exert an effect on particular enzymes. Consequently, the efficacy of compounds directed against α-glucosidase may not necessarily be indicative of their effectiveness against α-amylase. In this case, the overall inhibitory effect of the extract may vary depending on the enzyme type^46,47. While no studies have been found in the literature evaluating the effects of both the extract and the main components in terms of α-amylase and α-glucosidase enzyme inhibition, there are some studies examining the inhibitory effects of the main components alone. Hua et al.⁴⁸ reported that kaempferol monoglycoside inhibited α-glucosidase with an IC₅₀ of 40.02 ± 4.61 μM, and kaempferol diglycoside inhibited α-amylase with an IC₅₀ of 0.09 ± 0.02 μM.

In-silico inhibition of α-amylase and α-glucosidase

In order to corroborate the outcomes derived from in-vitro enzyme inhibition studies, in-silico molecular docking analysis was conducted. In the literature, many studies support in-vitro and in-vivo biological activities using in-silico methods using molecular docking. In this study, anti-diabetic activity was supported by examining the interaction of active compounds identified in P. oleracea with α-amylase and α-glucosidase^49–51. In this study, major compounds identified by GC–MS and LC–MS/MS analyses were selected as ligands, and their interactions with the α-amylase enzyme were investigated. The utilisation of GC–MS analysis enabled the identification of dillapiole and hexahydrofarnesyl acetone in significant concentrations. Among the major phenolic compounds identified by LC–MS/MS analysis, vannilin and p-coumaric acid were selected as ligands. Subsequent investigation focused on the interaction of these main compounds with α-amylase. The interactions of the selected ligands with the α-amylase enzyme were investigated, and the results are given in Fig. 4 and Table 2. The inhibition constants calculated for the interaction of vannilin and p-coumaric acid with α-amylase are 532.09 µM and 83.05 µM, respectively. The compounds selected as ligands interact with α-amylase via different amino acids. These interactions are facilitated by hydrogen bonds, pi-lone pairs, and pi-alkyl bonds. Furthermore, dillapiole and hexahydrofarnesyl acetone have been observed to interact with α-amylase through different amino acids. These interactions occur through various bonds, including hydrogen bonds, carbon-hydrogen bonds, and pi-alkyl bonds. The binding energies of dillapiole and hexahydrofarnesyl acetone to α-amylase were calculated as −4.67 kcal/mol and −4.44 kcal/mol, respectively, and the inhibition constants were calculated as 375.09 µM and 551.84 µM. The tested phenolic substances and essential oil ligands have been found to interact with different amino acids in the polypeptide chain of α-amylase, thereby causing deterioration in the enzyme structure and decreased activity. α-amylase catalyses the breakdown of starch into smaller polysaccharides and disaccharides, such as maltose. This process is pivotal in facilitating the absorption of carbohydrates into the bloodstream. Inhibition of enzyme activity has been demonstrated to slow the rate of carbohydrate digestion and reduce the amount of glucose released into the bloodstream, thereby exerting anti-hyperglycemic activity. A plethora of studies have been published which report in-silico interactions between phenolic compounds and α-amylase. Huang et al.⁵² determined that α-amylase and p-coumaric acid interacted, resulting in binding energies ranging from -6.30 to -4.50 kcal/mol.

Fig. 4.

Interaction of α-amylase with a. vanillin, b. coumaric acid, c. Dillapiole, d. Hexahydrofarnesyl acetone Hyrogen bond, Pi-Aklyl, Carbon Hydrogen bond, Amide-Pi stacked.

Table 2.

Data on the interaction of P. oleracea phytochemicals with α-amylase.

Ligand	Binding energy	Inhibiton constant	Hydrogen interactions	Other interactions
Dillapiole	−4.67	375.09	ARG346	ARG303, HIS305, ILE312, PHE348, GLY351, LYS352, ASP353, ASP356
Hexahydrofarnesyl acetone	−4.44	551.84	ASN216	LYS208, LEU211, HIS215, ASN216, SER226, LYS227
Vanillin	−4.48	532.09	VAL 129, HIS 185	TYR 67, LYS 68, LYS 178
p−Coumaric acid	−5.57	83.05	ARG252,SER289ASP402, RG421	−

Binding energy: kcal/mol; Inhibiton constant: µM, other inreactions: Pi-Aklyl, Carbon Hydrogen bond, Amide-Pi stacked.

The same ligands were utilised in the support of α-glucosidase inhibition by molecular docking, with the results presented in Table 3 and Fig. 5. Vanillin and p-coumaric acid, which were chosen as ligands, interact with the enzyme by establishing hydrogen bonds through amino acids such as phenylalanine, tyrosine, serine. It has been established that proline, valine, alanine, elements of the polypeptide, also contribute to the establishment of other bonds. The inhibition constants calculated for the interaction of vannilin and p-coumaric acid with α-glucosidase were found to be 201.43 µM and 52.14 µM, respectively. Dillapiole has been observed to interact with amino acids such as phenylalanine, valine, proline, alanine, and tyrosine, which are components of the α-glucosidase polypeptide. Hexahydrofarnesyl acetone, on the other hand, has been found to interact with amino acids including valine, arginine, phenylalanine, proline, and valine. The inhibition constants calculated for the interaction of dillapiole and hexahydrofarnesyl acetone with α-glucosidase were 14.59 µM and 26.64 µM, respectively, and the binding energies were calculated as −6.60 kcal/mol and −6.24 kcal/mol. The binding of both ligands to α-glucosidase, exhibiting disparate interaction kinetics, serves to corroborate the enzyme inhibition that was detected in-vitro. α-glucosidase is an enzyme that functions by the breakdown of large carbohydrate molecules to release glucose. In this manner, it guarantees the presence of glucose in the blood following a period of dieting. α-glucosidase inhibitors impede elevations in blood glucose levels by constraining the activity of enzymes. Interactions determined by inhibition and molecular docking in in-vitro studies have been shown to cause disruptions and delays in enzyme function, resulting in anti-hyperglycemic activity. The anti-diabetic activity tested in-vitro in this study was also supported by molecular docking, and there are many studies in the literature where in silico interactions are used to support the data obtained under laboratory conditions^53–58.

Table 3.

Detailed interaction profiles of P. oleracea phytochemicals with α-glucosidase.

Ligand	Binding energy	Inhibiton constant	Hydrogen interactions	Other interactions
Dillapiole	−6.60	14.59	PHE129	VAL84, PHE90, PRO131,PRO217, THR235,VAL236, ALA237, PRO238
Hexahydrofarnesyl acetone	−6.24	26.64	–	VAL84, ARG89, PHE90, PHE128, VAL236, PRO238
Vanillin	−5.04	201.43	SER 88, PHE 90	VAL 84, SER 88, PRO 131, ALA 237
p−Coumaric acid	−5.84	52.14	PHE90,PHE129, TYR133	PRO238

Binding energy: kcal/mol; Inhibiton constant: µM, other inreactions: Pi-Aklyl, Carbon Hydrogen bond, Amide-Pi stacked.

Fig. 5.

Interaction of α-glucosidase with a. vanillin, b. p-coumaric acid, c. dillapiole, d. hexahydrofarnesyl acetone Hyrogen bond, Pi-Aklyl, Carbon Hydrogen bond, Amide-Pi stacked.

Anti-inflammatory activity

Assays targeting protein denaturation inhibition are emerging as a valuable in-vitro method for identifying potential anti-inflammatory compounds without the use of living subjects. Protein denaturation has been identified as a critical step in the inflammatory process, playing a significant role in the onset of inflammatory and arthritic complications associated with autoantigen formation. Presently, the anti-inflammatory potential of herbal extracts is being subjected to rigorous scrutiny as a prospective alternative to synthetic anti-inflammatory medications. Herbal extracts offer a biologically diverse array of safe and readily available sources, thereby presenting a compelling proposition as a substitute for conventional pharmaceuticals, which are associated with limitations in terms of toxicity and adverse side effects²¹. BSA denaturation inhibition data of P. oleracea extract obtained with different solvents are given in Fig. 6. The highest inhibition was obtained with chloroform extract, while the lowest inhibition was obtained with water extract. The chloroform extract showed 76.1% inhibition of protein denaturation at a concentration of 200 µg/mL, and the IC₅₀ value was calculated as 126 µg/mL. Measurements performed at the same concentration revealed 63.9% inhibition for the ethanol extract, 75.0% for the ethyl acetate extract, and 59.9% for the water extract. These results suggest that different solvents produce significant differences in the anti-inflammatory potential of the resulting extracts. The IC₅₀ value of aspirin tested as standard was calculated as 229 µg/mL. The mechanisms by which a plant-derived extract inhibits protein denaturation can be explained at the molecular level by several main pathways. These vary depending on the structural and chemical properties of the phytochemicals contained in the extract. Polyphenols, including phenolic compounds, flavonoids and tannins, have been shown to interact with polar and aromatic amino acids on the protein’s surface, thereby preserving its folding structure. Protein denaturation has been shown to expose hydrophobic regions, with the interaction of these regions with water having the potential to disrupt the protein’s structure. The active components of extract have been shown to impede aggregation through the coating or surrounding of these hydrophobic regions. It has been established that certain high-molecular-weight polyphenols or polysaccharides bind to the surface of proteins, thereby forming a protective layer. This layer has the capacity to retard denaturation caused by factors such as heat, pH changes or chemical stress. The inhibition of protein denaturation is typically achieved through a combination of direct protein-compound interactions and indirect protective mechanisms. Swathigadevi et al.⁵⁹ reported that the protein denaturation inhibition activity of P. oleracea extract obtained from Sudarakottai, Mannargudi was 16.4% for 100 μg/mL and they related the presence of this activity with steroids.

Fig. 6.

Protein denaturation inhibition of P. oleracea extracts obtained with different solvents and their images in the test tube.

Anti-hemolytic activity

The ability of P. oleracea extract to protect red blood cells from AAPH-induced hemolysis is presented in Table 4. The AAPH-induced hemolysis model is a widely utilised method for the estimation of the anti-oxidant and anti-hemolytic capacity of plant extracts. At 37 °C, AAPH generates free radicals that generate a low but constant flux of peroxyl radicals, causing erythrocytes to undergo chain oxidation of membrane lipids and proteins. It has been demonstrated that erythrocyte membrane disruption, occasioned by an excess of free radicals, can result in hemolysis or the demise of healthy erythrocyte cells⁶⁰. As the dose increased, there was a concomitant increase in antihemolytic activity (p < 0.05), with the highest activity being detected at a concentration of 200 µg/mL. The ethanol extract demonstrated higher activity in comparison to other extracts. 200 µg/mL ethanol extract demonstrated 65.17% anti-hemolytic activity, while ascorbic acid at the equivalent dose exhibited 39.41% activity (p < 0.05). This anti-hemolytic activity is likely attributable to the various phenolic compounds, known as phenolic hydrogen atom donors, that are present in both extracts. Through their anti-oxidant activity, they help to stabilise free radicals, thereby increasing the resistance of erythrocytes to oxidative stress. Furthermore, the phenolic compounds present in extracts have been shown to donate one or more electrons to neutralise the AAPH radical, thereby preventing hemolysis. In addition, it has been demonstrated that polyphenols possess the capacity to interact with the hydrophilic portion of the lipid membrane, thereby inducing alterations in the packing pattern of the polar heads of lipids⁶¹. No studies have reported in-vitro anti-hemolytic activity of P. oleracea extracts. However, anti-oxidant activity, specifically indicating a radical-scavenging effect, has been reported. Aoudeh et al.⁶² reported that P. oleracea exhibited high activity in neutralizing DPPH, a free radical similar to AAPH, with an IC₅₀ value of 57.23 µg/mL. Additionally, studies have reported that flavonoids found in plants have anti-hemolytic activity due to their ability to bind to the membranes of erythrocytes and protect them from hypotonic lysis and lipid peroxidation⁶³. Many activities exhibited by plants, including anti-hemolytic activity, are related to the active components they contain, and these components are the main elements that enable plants to exhibit various biological activities and to exhibit a protective role against toxicity^64–70.

Table 4.

Protective effect of P. oleracea extract against AAPH-induced hemolysis.

Dose	Ethanol	Chloroform	Ethyl acetate	Water	Ascorbic acid
5 µg/mL	25.41 ± 1.90 h	19.45 ± 1.23 h	18.79 ± 1.19 h	19.56 ± 1.95f.	13.56 ± 2.17 h
10 µg/mL	31.65 ± 2.15 g	21.78 ± 1.86 g	20.34 ± 2.24 g	21.20 ± 2.77 g	18.45 ± 2.83 g
25 µg/mL	33.46 ± 3.21f.	24.56 ± 1.75f.	26.78 ± 2.76f.	19.47 ± 2.36f.	21.76 ± 3.13f.
50 µg/mL	36.10 ± 3.90e	29.45 ± 2.13e	29.56 ± 1.97e	24.50 ± 1.95e	25.81 ± 2.56e
75 µg/mL	47.80 ± 3.59d	34.79 ± 1.76d	32.45 ± 3.12d	29.87 ± 2.90d	28.66 ± 4.11d
100 µg/mL	51.23 ± 4.15c	40.56 ± 2.81c	35.71 ± 3.17c	32.46 ± 4.62c	31.75 ± 3.16c
150 µg/mL	59.64 ± 3.80b	42.78 ± 3.09b	40.12 ± 4.19b	38.75 ± 3.40b	36.91 ± 3.90b
200 µg/mL	65.17 ± 4.26a	55.79 ± 3.65a	43.17 ± 3.67a	41.37 ± 4.13a	39.41 ± 4.22a
Images of hemolysis

Conclusion

P. oleracea is an edible plant with high nutritional value that is widely consumed both raw and cooked throughout the world. However, the phytochemical composition and biological activities of the species may show significant differences depending on geographical origin, growing conditions and environmental factors. Research on P. oleracea, a species that is extensively distributed throughout the Türkiye flora, remains limited. There is a paucity of integrative studies that comprehensively examine the essential oil and phenolic compound profiles of regional samples, and furthermore, there is a dearth of studies that elucidate the multifaceted biological activities of these components. This study was conducted on samples of P. oleracea collected from Bulancak (Türkiye). The investigation involved the comprehensive profiling of phenolic and essential oil components by employing both LC–MS/MS and GC–MS techniques. Notably, this study stands as one of the inaugural comprehensive analyses to correlate the obtained phytochemical data with in-vitro anti-diabetic (inhibition of α-amylase and α-glucosidase), anti-inflammatory (inhibition of protein denaturation) and anti-hemolytic activities. The findings of this study are further substantiated by in-silico molecular docking analyses. In consequence of the LC–MS/MS analysis, p-coumaric acid was identified as the predominant compound, while vanillin, caffeic acid, sesamol and protocatechuic aldehyde were detected at negligible levels. The results of the GC–MS analysis revealed that the predominant compound in the essential oil fraction was hexahydrofarnesyl acetone, with dillapiole, myristicine and β-ionone detected at moderate levels. The IC₅₀ values for the α-amylase enzyme were found to range from 352.24 to 623.45 mg/mL. However, a stronger activity was observed in the inhibition of α-glucosidase. The most significant effect in terms of anti-hemolytic activity was detected in the 200 µg/mL ethanol extract, which was considerably higher than the equivalent concentration of ascorbic acid. In anti-inflammatory tests, 200 µg/mL chloroform extract demonstrated 76.1% inhibition, with an IC₅₀ value determined as 126 µg/mL. The in-silico molecular docking results support the in-vitro enzyme inhibition data, demonstrating that ligand-enzyme interactions can lead to structural changes in the active sites of α-amylase and α-glucosidase, resulting in loss of activity. Although single-site biological activity assessments of P. oleracea are commonplace in the extant literature, comprehensive studies are lacking in which the phenolic and essential oil compositions of this species are determined using high-sensitivity analytical methods within the same study and integrated with multiple biological activity tests and molecular modelling data. This study will serve as a model for elucidating the relationships between the phytochemical composition and biological activities of plant structures consumed as food, thereby providing a scientific basis for functional food development studies.

Author contributions

Ö.Y.G: investigation; methodology; visualization; writing—review and editing. E.Y: conceptualization; methodology; data curation; software; visualization; K.Ç: conceptualization; data curation; investigation; methodology; writing—review. B.Ö: investigation; methodology, visualization; writing—review.

Data availability

The datasets used and/or analyzed in the current investigation will be made available by the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.