Seasonal variation in the phenolic compounds of Algerian Cistus creticus leaf extracts: an in silico and in vitro study

Abstract

This study investigates the seasonal influence on the phenolic and flavonoid composition of Cistus creticus leaf extracts and evaluates their antioxidant and antiproliferative properties through experimental and computational methods. The results showed that samples collected in September exhibited significant antioxidant activity, with IC₅₀ values of 127.44 µg/mL and 108.62 µg/mL in the ABTS and molybdenum assays, respectively. In comparison, samples collected in March displayed higher IC₅₀ values of 139.67 µg/mL (ABTS) and 166.51 µg/mL (molybdenum), indicating reduced antioxidant potency. Additionally, the extracts demonstrated strong antiproliferative effects against two cancer cell lines, with the March sample exhibiting the highest total polyphenol content (113.42 mg GAE/g) and the greatest efficacy against A549 cells (IC₅₀ = 11.9 µg/mL). Molecular docking results indicated that the myricetin-EGFR complex was stabilized by five hydrogen bonds and had a binding energy of -9 kcal/mol. Similarly, the myricetin-NADPH oxidase complex showed seven hydrogen bonds and an identical binding energy of -9 kcal/mol. The stability of these complexes was further validated through 100-ns molecular dynamics simulations. Overall, these findings underscore the therapeutic potential of C. creticus leaf extracts as natural antioxidant and anticancer agents.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-07975-7.

Introduction

For millennia, humans have turned to medicinal plants for healing purposes, and these natural remedies continue to play a critical role in contemporary healthcare and industry. Early uses—such as teas, tinctures, and poultices—have evolved into sophisticated formulations with diverse applications. There is growing global interest in utilizing medicinal and aromatic plants as natural resources across the pharmaceutical, food, and cosmetic industries. The bioactive compounds in these plants, recognized for their antioxidant, antibacterial, and anticancer properties, hold immense potential for applications in botanical medicines, dietary supplements, functional foods, and sustainable food packaging solutions^1,2.

The genus Cistus, commonly known as Rockroses, belongs to the family Cistaceae and comprises between sixteen and twenty-eight species of perennial shrubs. Cistus species are native to the Mediterranean, western Africa, and Asian regions^3,4. They are characterized by wooden stems, leathery leaves, and showy flowers that typically range in color from white to pink and purple⁵.

Cistus species are renowned for their rich phytochemical composition, which includes essential oils, flavonoids, tannins, and phenolic compounds^6,7. These constituents are responsible for various biological activities attributed to the genus, including antimicrobial^2,8,9anti-inflammatory agents¹⁰antitumor¹¹antiviral¹²and cytotoxic activities^13,14.

Recent studies have demonstrated that extracts from Cistus creticus species exhibit significant antioxidant activity, helping to neutralize free radicals and mitigate oxidative stress¹⁵. Additionally, research has highlighted their antiproliferative action against various cancer cell lines, suggesting potential applications in cancer prevention and therapy^13,16. Numerous studies have demonstrated that various parts of C. creticus are rich in bioactive compounds, including rutin, diterpenes, monomeric and polymeric flavanols, and gallic acid^17,18. EGFR, a receptor tyrosine kinase, is frequently overexpressed in various cancers, driving tumor growth and survival, and is a well-established target for anticancer therapies¹⁹. NADPH oxidase, a key enzyme in reactive oxygen species (ROS) production, contributes to oxidative stress, which is critical in cancer initiation and progression²⁰. Targeting these proteins allows for a dual approach: EGFR inhibition disrupts cancer cell proliferation, while NADPH oxidase modulation reduces ROS levels and oxidative damage. Furthermore, flavonoids and phenolic compounds in plants are known to interact with key molecular targets, making them promising candidates for the assessment of chemopreventive potential²¹.

This study aims to investigate the seasonal variation in the phenolic and flavonoid profiles of C. creticus leaf extracts and evaluate their potential as antioxidant and antiproliferative agents. By employing advanced analytical techniques such as LC-MS/MS, the composition of the extracts was characterized in detail. Furthermore, the chemopreventive potential of the extracts was assessed through biological screening and in silico studies, including molecular docking, PASS prediction, and molecular dynamics simulations. These analyses focused on the interaction and stability of the most abundant phenolic compounds with EGFR and NADPH oxidase, two critical targets in cancer progression and oxidative stress. This work seeks to provide new insights into the therapeutic potential of C. creticus and its bioactive metabolites in the context of cancer prevention and treatment.

Materials and methods

Plant material

Freshly harvested leaves of Cistus creticus (locally named “Qaçça”) were harvested during the four seasons: summer (June 2022), autumn (September 2022), winter (December 2022), and spring (March 2023) in the Aflou zone (Latitude: 34°06’50’’N Longitude: 2°05’50’’E, Altitude: 1400 m). It is situated 100 km northwest of Laghouat city in southern Algeria. The humidity, precipitation, and maximum and minimum temperature data obtained from the NASA power are shown in Table 1. Professor Mohamed Kouidri identified this plant at the Agricultural Department of the Faculty of Sciences at Laghouat University in Algeria. The voucher specimens are stored at the Laboratory of Biological and Agricultural Sciences (LBSA) of the same university. with the numbers LBAS Cc/06/22, LBAS Cc/09/22, LBAS Cc/12/22 and LBAS Cc/03/23 respectively. The plant material was shaded and dried (at room temperature) for 15 days, then ground and stored in paper sacks until needed.

Table 1.

Seasonal environmental data for aflou, Algeria (2022).

Month	T2M (°C)	T2M_MAX (°C)	T2M_MIN (°C)	RH2M (%)	Precipitation (mm/day)
Jan	4.32	18.49	-4.35	61.25	0.1
Feb	8.35	22.16	-1.87	55.88	0.5
Mar	8.78	22.80	-3.63	60.06	1.9
Apr	12.07	25.03	1.98	56.38	2.09
May	18.84	34.50	1.92	45.38	0.34
Jun	27.73	39.09	15.69	24.75	0.07
Jul	29.27	39.08	16.29	25.50	0.04
Aug	28.30	38.37	15.63	25.44	0.06
Sep	24.37	38.51	7.01	36.81	0.86
Oct	17.87	28.52	8.70	51.12	0.91
Nov	11.35	24.22	-0.76	58.50	0.22
Dec	9.03	21.48	-0.77	66.69	0.29
Annual	16.73	39.09	-4.35	47.25	0.61

T2M (°C) Average monthly temperature at 2 m.T2M_MAX (°C) Maximum monthly temperature at 2 m.T2M_MIN (°C) Minimum monthly temperature at 2 m. RH2M (%) Monthly average relative humidity at 2 m. Precipitation (mm/day): Corrected precipitation from MERRA-2.

Preparation of extracts

We soaked 40 g of each sample in 400 mL of an aqueous methanol solution (2 parts methanol to 8 parts water) for five days at room temperature in the dark. This duration was chosen to allow adequate diffusion and solubilization of phenolic and flavonoid compounds while minimizing the risk of light or temperature-induced degradation. After soaking, we filtered the mixture and repeated this process twice. Once we finished filtering each solution, we used a rotary evaporator (Rotavapor IKA VB 10, Germany) to evaporate the mixture under reduced pressure at a temperature below 45 °C. This kept the plant compounds from being lost or damaged. Finally, we lyophilized the crude aqueous extract and stored it at −4 °C²².

Total phenolic content (TPC)

The total phenolic content of the extracts was quantified using the Folin-Ciocalteu’s colorimetric method, following the protocol described by Boulanouar et al. (2013)²³with slight modifications for improved reproducibility. Briefly, 200µL of diluted extract was mixed with 1 mL of Folin-Ciocalteu’s reagent (10%). After 5 min, 800µL of sodium carbonate solution Na₂CO₃ (7.5%) was added. The reaction mixture was vortexed and then incubated at room temperature (25 ± 2 °C) for 30 min in the dark. After incubation, the absorbance of the resulting blue complex was measured at 765 nm using an OPTIZEN 2120 UV spectrophotometer. A calibration curve was prepared using gallic acid (0–200 µg/mL) as a standard, and results were expressed as mg of gallic acid equivalents (GAE) per gram of dry extract. All determinations were performed in triplicate.

Total flavonoid content (TFC)

Total flavonoid content was determined using the aluminium chloride colorimetric method described by Boulanouar et al. (2013)²³. To 1 mL of extract or standard solution, 1 mL of 2% (w/v) aluminium chloride in ethanol was added. The reaction mixture was allowed to stand for 30 min at room temperature (25 ± 2 °C) in the dark. The absorbance was then recorded at 420 nm using an OPTIZEN 2120 UV spectrophotometer. Quercetin (0–200 µg/mL) was used as the reference standard, and the results were expressed as mg quercetin equivalents (QE) per gram of dry extract. All measurements were performed in triplicate.

Antioxidant activity assessment

Free radical scavenging activity (DPPH)

The free radical scavenging activity of the C. creticus extracts was determined using the DPPH (2,2-diphenyl-1-picrylhydrazyl) assay²³.Briefly, 100 µL of each extract at various concentrations was mixed with 1900 µL of a 60 µM methanolic DPPH solution in a quartz cuvette. The reaction mixture was incubated for 30 min at room temperature in the dark. The decrease in absorbance was measured at 517 nm using an OPTIZEN 2120 UV-visible spectrophotometer. The percentage of radical scavenging activity was calculated using the following formula:

Where A0 is the absorbance of the control (without sample) and A1 is the absorbance in the presence of the extract. We plotted the percentage inhibition against the sample concentrations and determined the IC₅₀ in µg/mL. We used butylated hydroxytoluene (BHT) and ascorbic acid as positive controls.

ABTS free radical-scavenging activity

For the determination of ABTS^•+ (2,2’-azino-bis-3-ethylbenzothiazoline-6-sulfonic acid) radical scavenging, we followed the method outlined by Boulanouar et al. (2013)²³with slight modifications. ABTS^•+ was generated by reacting 7 mM ABTS stock solution with 2.45 mM potassium persulfate, incubated in the dark at room temperature for 12–16 h before use. The resulting ABTS^•+ solution was diluted with ethanol to an absorbance of 0.70 at 734 nm. Then, 100 µL of extract at various concentrations was mixed with 1.9 mL of ABTS^•+ solution. The mixture was incubated for 30 min at room temperature, and absorbance was measured at 734 nm. We measured the IC₅₀ values (in µg/mL) as described earlier. Each test was done three times to ensure accuracy. We used butylated hydroxytoluene (BHT) and ascorbic acid as positive controls.

Where A0 is the absorbance of the control and A1 is the absorbance of the sample.

Total antioxidant capacity (TAC)

The total antioxidant capacity (TAC) of the extracts is determined by the phosphomolybdenum method according to the procedure described by Rezzoug et al. (2019)²². Briefly, 0.3 mL of each extract was combined with 3 mL of reagent solution containing 0.6 M sulfuric acid, 28 mM sodium phosphate, and 4 mM ammonium molybdate. The tubes were incubated at 95 °C for 90 min. The absorbance was recorded at 695 nm using an OPTIZEN 2120 UV spectrophotometer.

The antioxidant capacity was expressed as EC₅₀, defined as the concentration of extract (in µg/mL) that gives an absorbance of 0.5, calculated using linear regression analysis. Butylated hydroxytoluene (BHT) and ascorbic acid were used as positive controls. All experiments were performed in triplicate.

Antiproliferative activity.

The antiproliferative activities of the investigated extracts were assessed using the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT) colorimetric assay²⁴. Human lung carcinoma A549 and human ductal breast carcinoma T47D cell lines were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/mL penicillin, and 0.1 mg/mL streptomycin at 37 °C in a humidified incubator with 5% CO₂. Cells were seeded at a fixed density of 1 × 10⁴ cells per well in 96-well plates, and then set to adhere for 24 h. The tested extracts were applied in triplicate at various dilutions, with a total of nine replicates evaluated on three separate occasions. The MTT assay was carried out to identify the number of viable cells based on the formazan product quantity, which was measured by absorbance at 490 nm.

HPLC-LC-MS/MS analysis

The phytochemicals flavonoids and phenolic compounds analysis was performed using a SciEx UPLC (Exion-UPLC, USA) equipped with the LC-ESI-MS/MS-4500-QTRAP system (AB Sciex Instrument, Framingham, MA, USA). Moreover, data analysis used Analyst 1.7 software for data analysis²⁵.

The samples were separated at a temperature of 50 ± 1 °C with an ODS column (100 × 2.1 mm, 5 μm). The initial mobile phase was made up of 80% water (with 1% formic acid, labeled as A) and 20% methanol (with 1% formic acid, labeled as B) (Table 2). We achieved the separation using the gradient program outlined below^25,26.

Table 2.

The elution system of the column.

Time	% A (Methanol containing 1% formic acid)	% B (Water containing 1% formic acid)
0	80	20
1	80	20
12	0	100
18	0	100
19	80	20
22	80	20

The solvent flowed at a rate of 0.35 mL per minute, and the injection volume was 5 µl. We conducted MS/MS analyses using negative ion mode. Nitrogen gas, at a pressure of 60 psi, served as both the nebulizing and drying gas. The mass spectra were obtained over an m/z range of 100–900 amu. The calibration curves were prepared using the standard samples of the phenolic compound and phytochemicals and the components of the extract were calculated accordingly. Table 2 represents the phenolic profiling (ng/mg of compound identified) of the extract of C. creticus using HPLC-MS/MS analysis.

In silico analysis

Compounds and bioinformatics tools

We investigated the phenolic compounds in C. creticus using the RP-HPLC-PDA technique. The molecular structures of these compounds were sourced from the PubChem database. We prepared the optimized structures in .pdb format and submitted them in. pdbqt format using AutoDock software. Furthermore, we employed the RCSB Protein Data Bank, PyMOL, and Discovery Studio for our research.

Protein preparation

The receptors utilized in this study include the Epidermal Growth Factor Receptor tyrosine kinase domain complexed with the 4-anilinoquinazoline inhibitor erlotinib (PDB ID: 1M17)²⁷ and the crystal structure of NAD(P)H oxidase (PDB ID: 2CDU)²⁸both obtained from the Protein Data Bank (PDB). We prepared the receptors by removing water molecules and any small ligands bound to the target receptor using PyMOL software. Subsequently, we employed AutoDock software to identify the active site in the processed receptor and converted the structure to .pdbqt format using AutoDock Vina for docking analysis.

Virtual screening

For high-speed virtual screening, we utilized iGEMDOCK (Generic Evolution Method for Docking) version 2.1²⁹. In silico screening of the 21 compounds identified in the C. creticus extract was performed using the PDB codes: (ID: 1M17) for the Epidermal Growth Factor Receptor tyrosine kinase domain with the 4-anilinoquinazoline inhibitor erlotinib, and (ID: 2CDU) for the crystal structure of NAD(P)H oxidase. The screening score was derived from total energy calculations, which encompass van der Waals forces, hydrogen bonding, and electrostatic interactions (total energy = VdW + HBond + electrostatic). We conducted this analysis with iGEMDOCK version 2.1.11, using screening parameters that included a population size of 300, 70 generations, and 2 solutions. After evaluating the energy-based results, we selected potential inhibitors based on their stability for further detailed analysis.

Docking approach

The anchoring protocol was employed for the co-crystallized reference ligand and water molecules from the crystal structure. We added polar hydrogen atoms during this process, maintaining the protein in a rigid state while allowing the ligand to remain flexible. The ligands (Salvianolic acid-A, Rutin, Quercetin, Hesperetin, Myricetin, Catechin, Rosmarinic acid, CAPE, Carnosic acid) that demonstrated the best virtual screening results with the tyrosine kinase domain of the Epidermal Growth Factor Receptor were anchored using PDB code (ID: 1M17). Additionally, Luteolin 7-O-glucoside, Myricetin, Hesperetin, Catechin, Quercetin, Rutin, Rosmarinic acid, CAPE, and Carnosic acid were anchored with NAD(P)H oxidase (ID: 2CDU). We utilized AutoDock Vina to determine the binding positions of the bioactive ligands within the active sites of both targets³⁰. Following docking, we assessed the minimum binding energy based on the ligand placements. The results were visualized using Discovery Studio and PyMOL, and we compared the types of interactions that each molecule formed within the active sites.

Biological activity prediction (PASS)

We evaluated the ten most notable secondary metabolites identified in C. creticus post-docking using the Prediction of Activity Spectra for Substances (PASS) approach (https://www.way2drug.com/passonline/). We generated the chemical structures of these compounds, which exhibit potential antioxidant and antineoplastic activities (notably for breast and lung cancer), from their SMILES representations sourced from the PubChem database. This research yielded predictions for probability activity (Pa) and probable inactivity (Pi) values, stipulating that Pa must surpass Pi, with a range from 0.000 to 1.000. A PA number over 0.7 signifies a high degree of biological activity, while values ranging from 0.5 to 0.7 imply moderate bioactivity; values below 0.5 denote poor bioactivity^31,32.

MD simulation

Molecular dynamics simulations were conducted to assess the stability and dynamic behavior of the two most promising protein–ligand complexes. These are myricetin-EGFR and myricetin-NAD(P)H oxidase. They were found using molecular docking. Schrödinger LLC’s Desmond software³³suite performed molecular dynamics simulations for 100 nanoseconds. Molecular docking studies facilitated the prediction of a ligand’s binding state under static conditions. We performed computational simulations to predict the nature of ligand binding in the natural environment. We preprocessed the protein-ligand complexes using either Protein Preparation Wizard or Maestro. This phase entailed optimizing and reducing the compounds. We configured all systems using the System Builder tool. We employed the TIP3P solvent model within an orthorhombic box and used the OPLS_2005 force field for the simulation³⁴. We neutralized the models by strategically introducing counter ions at suitable sites. We simulated physiological circumstances using a 0.15 M concentration of sodium chloride (NaCl). We selected the NPT ensemble for the extensive simulation, maintaining the number of moles (N), pressure (P), and temperature (T) at a temperature of 300 K and a pressure of 1 atm. Before the simulation, the models experienced a relaxing procedure. We recorded trajectories at 100 ps intervals for further analysis and evaluated the stability of the simulations by computing the root mean square deviation (RMSD), root mean square fluctuation (RMSF), protein-ligand interactions, protein secondary structure, and the characteristics of both the protein and ligand over time.

Statistical analysis

We presented the calculations’ findings as the mean and standard deviation from three independent measurements. We conducted an analysis of variance (ANOVA) and Tukey’s post hoc test for mean comparison, using GraphPad Prism software version 8.4 (Software, San Diego, CA, USA). We deemed the differences significant statistically when the p-value was below 0.05.

Results

The effect of seasonal variation on the phenolic content and antioxidant activities

This study investigated the seasonal variation in phenolic content, secondary metabolite production, and antioxidant activities of Algerian C. creticus leaf extracts. Antioxidant activities were assessed using DPPH, ABTS, and molybdate assays. As shown in Fig. 1, the highest extraction yield was observed in samples harvested in June (32.95%), followed by September (28.65%), March (19.60%), and December (15.57%). The Folin-Ciocalteu’s assay revealed significant variations in total phenolic content (TPC), with March samples showing the highest TPC (113.42 mg GAE/g extract), followed by September (108.14 mg GAE/g extract). In contrast, June and December samples exhibited lower TPC values (83.49 and 82.35 mg GAE/g extract, respectively). A similar trend was observed for total flavonoid content (TFC), with March samples recording the highest value (5.84 mg EQ/g extract), followed by September (5.10 mg EQ/g extract), June (4.47 mg EQ/g extract), and December (4.11 mg EQ/g extract) (Fig. 1).

Fig. 1.

Total phenolic and flavonoid content: CCJ Cistus creticus June, CCS Cistus creticus September, CCD Cistus creticus December, CCM Cistus creticus March, TPC Total Phenolic Content, TFC Total Flavonoid Content, AGE gallic acid equivalents, QE quercetin equivalents, ns no significant difference statistically.

Figure 2 highlights the noticeable seasonal variation in the antioxidant activities of C. creticus extracts. In the DPPH assay, the December sample exhibited the highest activity with the lowest IC₅₀ value (86.92 µg/mL), followed by June (110.62 µg/mL), September (122.40 µg/mL), and March (124.15 µg/mL). Conversely, in the ABTS assay, the September sample was the most effective with the lowest IC₅₀ value (127.44 µg/mL), followed by March (139.67 µg/mL). The June and December samples displayed similar IC₅₀ values (200.86 µg/mL and 200.80 µg/mL, respectively). In the molybdate assay, the September sample showed the highest antioxidant activity with an EC₅₀ value of 108.62 µg/mL, followed by December (161.84 µg/mL), March (166.51 µg/mL), and June (231.24 µg/mL). These findings demonstrate significant seasonal variations in the total phenolic content, total flavonoid content, and antioxidant activities of C. creticus extracts. Previous studies have shown that phenolic content and antioxidant activity are influenced by factors such as seasonal variation, ecological conditions, and extraction methods^35–38.

Fig. 2.

Antioxidant activity: (A) DPPH: 1,1-Diphenyl-2-picrylhydrazyl; (B) ABTS: 2,2-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid); (C) Molybdate assay.

Our data suggests that the harvesting season significantly influences the phenolic content and antioxidant activity of the extracts. The highest total phenolic and flavonoid content was observed in samples harvested in late winter (March), while the maximum antioxidant activity was found in samples collected in late summer and early fall (September). These findings are consistent with previous studies, which have also reported seasonal variations in the bioactive properties of plant extracts^39,40.

Seasonal variations significantly influence the growth and metabolism of C. creticus, affecting the accumulation of bioactive compounds and their biological activities. High temperatures and increased light intensity in summer enhance photosynthesis and activate the phenylpropanoid pathway. This leads to elevated flavonoid and phenolic acid synthesis, particularly quercetin, myricetin, and caffeic acid, contributing to antioxidant activity. Conversely, lower temperatures and higher humidity in winter may slow metabolic rates but promote the accumulation of anthocyanins and tannins, known for their protective roles. Water stress in dry seasons further triggers stress-response metabolites, enhancing the plant’s defense mechanisms. These variations directly impact the biological properties of the extracts, as seen in the March sample, which exhibited the highest polyphenol content and potent antiproliferative effects against cancer cells. The synergistic antioxidant and cytotoxic activities observed are likely due to seasonal metabolic adaptations regulating bioactive compound biosynthesis. To strengthen these findings, further metabolomic and transcriptomic analyses are needed to elucidate the molecular mechanisms driving seasonal shifts in secondary metabolite production^41,42.

Antiproliferative activity

The IC₅₀ values of the natural extracts against lung and breast cancer cells varied with the seasons (Fig. 3), suggesting that seasonal changes may influence the anticancer properties of these extracts. The IC₅₀ values represent the concentration of extract needed to inhibit 50% of the cancer cell population, with lower values indicating higher potency. In June (summer), the CCJ extract exhibited IC₅₀ values of 11.9 µg/mL and 14.6 µg/mL against breast and lung cancer cells, respectively. In September (autumn), the CCS extract showed higher IC₅₀ values of 17.8 µg/mL and 18.7 µg/mL against the same cancer cell lines. The CCD extract, collected in December (winter), had IC₅₀ values of 12.3 µg/mL and 13.9 µg/mL. In contrast, the CCM extract, harvested in March (spring), showed IC₅₀ values of 13.2 µg/mL for breast cancer cells and 11.9 µg/mL for lung cancer cells.

Fig. 3.

Antiproliferative activity of the tested samples.

Overall, the data suggest that seasonal variations significantly affect the anticancer properties of natural extracts. Extracts harvested in summer and spring exhibited higher potency, which may be attributed to more favorable growing conditions and higher concentrations of key bioactive compounds. These findings emphasize the importance of considering seasonal factors when assessing the therapeutic potential of natural extracts against cancer cells. Further research is needed to identify the specific compounds responsible for the observed seasonal variations in anticancer activity.

Our previous study profiled the chemical composition of Algerian C. creticus leaf extracts and evaluated their antioxidant and antiproliferative activities using in vitro and in silico methods. We found a wide variety of phenolic compounds in different extracts showing the strongest antioxidant and antiproliferative effects, particularly against T47D and A549 cancer cell lines⁴³. A recent study focused on extracting bioactive and antioxidant compounds from C. creticus leaves and using these extracts to enrich yogurt desserts. The optimal extract improved the antioxidant capacity and physicochemical properties of the yogurt⁴⁴. Palaiogiannis and his collaborators compared successive and single-solvent extraction methods for polyphenols and flavonoids from C. creticus leaves. They found that both ethanol and 50% ethanol-water mixtures were highly effective for extracting antioxidant compounds, and that using a range of solvents in succession increased the yield and diversity of extracted bioactives⁴⁵. Another recent study examined the effects of different solvents on the chemical composition, antioxidant, and antimicrobial activity of C. creticus extracts from Turkey. Methanol extracts showed higher antioxidant activity and richer polyphenol profiles than water extracts, although antimicrobial effects were weak⁴⁶. Also, researchers compared freeze drying and spray drying techniques for producing C. creticus powders, using maltodextrin and inulin as carriers. Both carriers preserved high levels of bioactive polyphenols and antioxidant capacity, with inulin proving as effective as maltodextrin for powder formulation⁴⁷.

LC-MS/MS analysis

The C. creticus extract’s LC-MS/MS analysis reveals the diversity of flavonoids and phenolic compounds (Table 3) ‎In June, the extract had high concentrations of gallic acid (25.35 ng/mg), catechin (10.9 ng/mg), rutin (4.815 ng/mg), apigenin (3.505 ng/mg), hesperidin (3.35 ng/mg), quercetin (1.86 ng/mg), vanillic acid (0.815 ng/mg), and minor components such as rosmarinic acid (0.4685 ng/mg), ferulic acid (0.3205 ng/mg), syringic acid (0.316 ng/mg), salvianolic acid (0.2795 ng/mg), luteolin 7-glucoside (0.251 ng/mg), apigenin-7-glucoside (0.251 ng/mg), 4-hydroxy-cinnamic acid (0.185 ng/mg), caffeic acid (0.0805 ng/mg), and luteolin (0.067 ng/ mg).

Table 3.

Phenolic compounds and flavonoids in the sample (ng/mg, extract).

Compounds	CCJ	CCS	CCD	CCM
Gallic acid	25.35	17.2	16.25	6.05
Catechin	10.9	4.335	9.55	19.65
Rutin	4.815	1.645	4.18	3.51
Apigenin	3.505	0.705	1.635	2.695
Hesperidin	3.35	1.335	2.985	2.86
Quercetin	1.86	4.61	1.67	3.28
Vanillic acid	0.815	0.2845	0.4325	0.595
Rosmarinic acid	0.4685	-	-	0.995
Ferulic acid	0.3205	0.259	0.28	0.3025
Syringic acid	0.316	0.2075	0.293	0.3155
Salvianolic acid-A	0.2795		-	0.2855
Luteolin 7-glucoside	0.251	-	0.294	0.2715
Apigenin 7-glucoside	0.251	-	0.1855	0.214
4-Hydroxy-cinnamic acid	0.185	0.01965	0.121	0.088
Caffeic acid	0.0805	-	-	-
Luteolin	0.067	-	0.0765	0.089
CAPE	-	13.75	-	15.65
Carnosic acid	-	-	-	2.625
Resveratrol	-	-	-	0.2435
Hesperetin	-	1.305	0.03155	-
Myricetin	-	2.185	-	-

CCJ Cistus creticus June, CCS Cistus creticus September, CCD Cistus creticus December, CCM Cistus creticus March, -: undetected.

Three months later, in September, the sample that was harvested produced a high concentration of gallic acid (17.2 ng/mg of extract), quercetin (4.61 ng/mg), catechin (4.335 ng/mg), rutin (1.645 ng/mg), and hesperidin (1.335 ng/mg), along with a few new compounds like myricetin (2.185 ng/mg), caffeic acid phenethyl ester (13.75 ng/mg), and hesperetin (1.305 ng/mg). In addition to the compounds, vanillic acid (0.2845 ng/mg); ferulic acid (0.259 ng/mg); syringic acid (0.2075 ng/mg); and 4-hydroxy-cinnamic acid (0.01965 ng/mg) were also detected in the extract.

With a minor content fluctuation, the primary chemical constituents found in the samples during December were similar to those found in June. Gallic acid (16.25 ng/mg), catechin (9.55), rutin (4.18), hesperidin (2.985), quercetin (1.67) and apigenin (1.635) were the main constituents, with trace amounts of vanillic (0.4325), luteolin-7-glucoside (0.294), syringic acid (0.293), ferulic acid (0.28), apigenin-7-glucoside (0.1855), 4-hydroxy-cinnamic acid (0.121), luteolin (0.0765), and hesperetin (0.03155).

In March, the extract was rich in catechin (19.65 ng/mg) and caffeic acid phenethyl ester (15.65). The amount of rutin and hesperidin was comparable to December. The quercetin and apigenin were slightly increased as compared to December. Other compounds were carnosic acid (2.625) and rosmarinic acid (0.995). Minor compounds were vanillic (0.595); syringic acid (0.3155); ferulic acid (0.3025); salvianolic acid (0.2855); luteolin-7-glucoside (0.2715); resveratrol (0.2435); apigenin-7-glucoside (0.214); luteolin (0.089), and 4-hydroxy-cinnamic acid (0.088).

The extract in March had high concentrations of caffeic acid phenethyl ester (15.7 ng/mg of extract) and catechin (19.7ng/mg). Hesperidin and rutin concentrations were similar to those of December. In comparison to December, there was a modest rise in quercetin and apigenin. Carnosic acid (2.63) and rosmarinic acid (0.995) were the other chemicals. The following were minor compounds: apigenin-7-glucoside (0.214), resveratrol (0.2435), vanillic (0.595), ferulic acid (0.3025), salvianolic acid (0.2855), luteolin-7-glucoside (0.2715), luteolin (0.089), and 4-hydroxy-cinnamic acid (0.088).

Molecular modelling

Virtual screening

The Epidermal Growth Factor Receptor (EGFR) is present in excess in multiple types of cancer. When a ligand binds to EGFR, it activates the tyrosine kinase (TK) domain. This activation starts a process that helps cells grow and survive cancer. TK inhibitors are used to block the activity of EGFR and other receptors as a treatment strategy for cancer. Over 40 TK inhibitors have been approved by the FDA for various types of cancer⁴⁸. The present study aimed to identify the active C. creticus phenolic compounds that inhibits EGFR-TK against lung cancer cells and also healthy cancer. The receptor tyrosine kinase domain of the Epidermal Growth Factor Receptor together with the 4-anilinoquinazoline inhibitor erlotinib (PDB ID: 1M17) were used as model systems for both types of cancer. Oxidants are important for keeping redox homeostasis for the body cells. However, too many oxidants can disrupt this balance. When levels of reactive oxygen species (ROS) increase, they can cause oxidative stress, which is linked to several diseases. Because of this, researchers are looking for antioxidants—agents that help restore the balance of oxidants and reduce ROS. Some enzymes, like cytochrome P450, myeloperoxidase (MP), lipoxygenase (LO), NADPH oxidase (NO), and xanthine oxidase (XO), produce ROS during the breakdown of arachidonic acid. By inhibiting these enzymes, we can lower ROS levels⁴⁹. This study aims to explore the antioxidant properties of C. creticus.

In this study, the compounds salvianolic acid-A, rutin, rosmarinic acid, quercetin, myricetin, hesperetin, catechin, CAPE, and carnosic acid were identified as active against the Epidermal Growth Factor Receptor tyrosine kinase (PDB: 1M17). Additionally, CAPE, carnosic acid, catechin, hesperetin, myricetin, quercetin, rosmarinic acid, rutin, and luteolin 7-O-glucoside were evaluated for the second target (ID: 2CDU), which corresponds to the NAD(P)H oxidase crystal structure. Table 4 presents the results of the virtual screening of phenolic compounds extracted from the leaves of C. creticus.

Table 4.

High-throughput screening results between the two targets and the selected molecules of C. creticus potentials.

Total energy (kcal/mol)
PDB: 1M17 Epidermal Growth Factor Receptor tyrosine kinase domain
Salvianolic acid- A	-89.19
Rutin	-77.64
Quercetin	-85.3171
Hesperetin	-85.4518
Myricetin	-84.7225
Catechin	-76.74
Rosmarinic acid	-75.21
CAPE	-72.0732
Carnosic acid	-70.5667
PDB: 2CDU for the Crystal Structure of Nad (P)H Oxidase
Luteolin 7-O-glucoside	-99.2067
Myricetin	-97.2955
Hesperetin	-92.0272
Catechin	-91.3151
Quercetin	-90.3901
Rutin	-86.6341
Rosmarinic acid	-79.17
CAPE	-78.4288
Carnosic acid	-73.8101

Molecular docking

Molecular docking assays are commonly employed to illustrate how ligands bind to receptors. To validate the relevance of the molecular docking, the target ligand is docked to the active site, confirming its accuracy. Compounds for both targets were docked to the same active sites based on the best virtual screening scores. Drugs with screening scores ranging from − 70.5667 kcal/mol to -89.19 kcal/mol were identified as promising candidates for the tyrosine kinase domain of the Epidermal Growth Factor Receptor, while compounds with scores between − 61.7025 kcal/mol and − 99.2067 kcal/mol were selected for NAD(P)H oxidase. The total energy of hydrogen bonds (H-bonds) and other interactions for the nine selected compounds is presented in Table 4. We chose the nine molecules with the best scores from the virtual screening and docked them into the active sites of both the tyrosine kinase domain of the Epidermal Growth Factor Receptor and NAD(P)H oxidase. We identified promising compounds with docking scores between − 9.9 kcal/mol and − 6 kcal/mol for the Epidermal Growth Factor Receptor, and between − 6 kcal/mol and − 9 kcal/mol for NAD(P)H oxidase. Table 5 summarizes the total energy and hydrogen bonds (H-bonds) of the nine top-selected compounds.

Table 5.

Docking results showing binding affinities of C. creticus molecules and hydrogen interactions with active site amino acids for both targets.

PDB: 1M17 Epidermal Growth Factor Receptor tyrosine kinase domain
	Binding Affinity kcal/mol	Hydrogen Bonds
Quercetin	-9.1	Met(769), Met(742)
Myricetin	-9	Met(769), Gln(767), Thr(766), Glu(738), Asp(831),
Hesperetin	-9	Asp(831), Met(769)
Carnosic acid	-8.3	Asp(831), Lys(721)
Catechin	-8.3	Thr(766), Glu(738), Asp(831), Met(742), Thr(830)
Salvianolic acid- A	-8.2	Met(769), Lys(721), Ala(719)
Rutin	-8.2	Met(769), Thr(830), Asp(831), Arg(817)
Rosmarinic acid	-8.2	Met(769), Asp(831), Glu(738), Gly(772), Arg(817)
CAPE	-7.1	Met(769), Asp(831), Thr(830)

PDB: 2CDU for the Crystal Structure of Nad (P)H Oxidase
Hesperetin	-9	Asp(282), Csx(42), Lys(134)
Myricetin	-9	Csx(42), Lys(134), Asp(282), Glu(32), Thr(9), Ser(115), Thr(113)
Quercetin	-8.5	Glu(32), Asp(282)
Carnosic acid	-8.3	Thr(112), Thr(9)
Catechin	-8.3	Glu(32), His(10)
Rutin	-8.3	Thr(112), Glu(32), Thr(9), Lys(134)
Rosmarinic acid	-8.2	Asp(282), Glu(32), Thr(9), Asn(36)
CAPE	-7.1	Thr(112)
Luteolin 7-O-glucoside	-6.9	Thr(113), Thr(9), Thr(112), Gly(12), Ala(11)

Post docking analysis

The structures derived from the molecular docking results were analyzed individually. Compounds were ranked based on the binding energy of their highest-rated conformations, and the top nine components were visually examined for their interactions with the active site. Subsequently, we utilized Biovia Discovery Studio Visualizer version 2016 to inspect the docking poses and analyze the interactions between the protein and the ligands. Visualization images were created using PyMOL.

Epidermal growth factor receptor tyrosine kinase domain-ligand complexes

Based on the 9 potential compounds molecular docking results, we observed that most of the compounds found share common hydrogen interactions with residues Met(769), Lys(721), Ala(719) Thr(830), Asp(831), Arg(817), Gly(772), Gln(767), Thr(766), Glu(738), Met(742) for the tyrosine kinase domain of the EGFR (Fig. 4). This is consistent with recent studies⁴⁸.

Fig. 4.

The various interactions between the active site the tyrosine kinase domain of the epidermal growth factor receptor and 9 molecule condensates: (a) Quercetin; (b) Myricetin; (c) Hesperetin; (d) Carnosic acid; (e) Catechin; (f) Salvianolic acid-A; (g) Rutin; (h) Rosmarinic acid; (i) CAPE.

The NAD(P)H oxidase-ligand complexes

For NAD(P)H oxidase (Fig. 5), we observed that most of the identified compounds exhibit common hydrogen interactions with residues Csx(42), Thr(112), Thr(9), Glu(32), His(10), Asp(282), Lys(134), Ser(115), Thr(113), Asn(36), Gly(12), and Ala(11). This finding is consistent with recent studies⁵⁰.

Fig. 5.

The various interactions between the active site of the NAD(P)H oxidase receptor and 9 molecule condensates: (a) Hesperetin; (b) Myricetin; (c) Quercetin; (d) Carnosic acid; (e) Catechin; (f) Rutin; (g) Rosmarinic acid; (h) CAPE; (i) Luteolin 7-O-glucoside.

Prediction of biological activity

Table 6 presents the PASS predictions for the ten substances in C. creticus, along with their various biological activities, including antioxidant properties and anticancer effects against lung and breast cancer. Our analysis revealed that myricetin possesses 1456 projected biological activities (PBAs), of which 35 PBAs have a Pa value reaching 0.9 following molecular docking and the execution of the PASS server for biological activity prediction. Myricetin and rutin exhibit significant potential for antioxidant action; Hesperetin displays elevated Pa values for antineoplastic effects in breast cancer. Furthermore, myricetin and rutin exhibit a significant likelihood of efficacy against lung cancer. The anticipated outcomes from the PASS server corroborate the acquired molecular docking data.

Table 6.

PASS prediction of the best compounds found in C. creticus.

Compounds	Biological activities
	Antioxidant		Antineoplastic (breast cancer)		Antineoplastic (lung cancer)
	Pa	Pi	Pa	Pi	Pa	Pi
Luteolin 7-O-glucoside	0.841	0.003	0.495	0.019	0.446	0.018
Salvianolic acid- A	0.563	0.005	0.159	0.128	0.127	0.126
Rutin	0.923	0.003	0.536	0.016	0.443	0.018
Quercetin	0.872	0.003	0.577	0.012	0.355	0.030
Hesperetin	0.746	0.004	0.656	0.007	0.276	0.045
Myricetin	0.924	0.003	0.578	0.012	0.417	0.021
Catechin	0.810	0.003	0.486	0.020	0.193	0.073
Rosmarinic acid	0.539	0.005	0.185	0.104	0.127	0.125
CAPE	0.512	0.006	0.300	0.055	0.159	0.093
Carnosic acid	0.423	0.010	-	-	0.288	0.042

Pa, probability ‘to be active’; Pi, probability ‘to be inactive.

MD simulation

RMSD and RMSF

The RMSD study shows that myricetin binds to EGFR and NAD(P)H oxidase in a way that is different from other molecules. This suggests that the dynamics and stability of the interactions may be different. Figure 6a illustrates that the Myricetin-EGFR complex reached stabilization after 10 ns, exhibiting significant structural flexibility in the protein, which has an RMSD of around 10 Å, whereas the ligand retains a very stable conformation with an RMSD of about 2.5 Å. This indicates that myricetin facilitates or permits structural modifications in EGFR while preserving strong binding affinity. Conversely, the myricetin-NAD(P)H oxidase complex (Fig. 6b) exhibits a more rigid protein conformation, with the protein RMSD stabilizing between 1.8 and 2.4 Å, while the ligand RMSD experiences minor fluctuations but remains stable between 1.0 and 2.0 Å. The data indicate that myricetin establishes a persistent and particular association with both targets, exhibiting a greater binding affinity to EGFR, despite its dynamic nature, and a more rigid connection with NAD(P)H oxidase, presumably signifying distinct mechanisms of action. This dual stability suggests that myricetin may function as a multifaceted therapeutic drug that addresses both signaling pathways and oxidative stress mechanisms.

Fig. 6.

The RMSD and RMSF plot of myricetin with EGFR and NAD(P)H oxidase receptor: (a) RMSD of myricetin-EGFR, (b) RMSD of myricetin- NAD(P)H oxidase, (c) RMSF of myricetin-EGFR, (d) RMSF of myricetin-NAD(P)H oxidase.

The myricetin-EGFR and myricetin-NAD(P)H oxidase complexes have different patterns of structural flexibility shown by their root mean square fluctuation (RMSF) plots. The Myricetin-EGFR complex (Fig. 6c) has a prominent peak at the C-terminal region (12Å), suggesting considerable localized flexibility, maybe resulting from the intrinsic dynamics of this region or conformational alterations caused by myricetin. The myricetin-NAD(P)H oxidase complex shown in Fig. 6d, on the other hand, has a lot less overall fluctuation. Its RMSF values are mostly between 0.5 and 2.5 Å, and the only peaks (3.5 Å) are found at residues chain A: His 451 (3.422 Å) and chain B: Thr 315 (3.082 Å), which are likely connected to flexible loops or termini. This comparative research indicates that myricetin binding stabilizes NAD(P)H oxidase, leading to decreased structural flexibility, although its interaction with EGFR may enhance localized flexibility, especially in the C-terminal region.

Protein-ligand contacts

During a 100 ns simulation, 20 amino acid residues of EGFR in the inter-domain binding site came into contact with myricetin. They made hydrogen bonds, water bridges, and hydrophobic bonds, with hydrogen bonds being the most common type of interaction. Myricetin formed significant interactions with 30 amino acid residues of NAD(P)H oxidase, which included hydrogen bonds, water bridges, and hydrophobic bonds, with water bridges and hydrogen bonds accounting for the majority of these interactions. The four pictures together (Fig. 7) show in great detail how myricetin interacts with NAD(P)H oxidase and EGFR as a protein ligand. In Fig. 7a, the histograms show the percentage of interactions for different residues. These show important contact residues like GLU-738, GLN-767, MET-769, LEU-820, and ASP-831 in EGFR and GLY-12, GLU-32, MET-33, THR-112, LYS-134, and ASP-282 in NAD(P)H oxidase (Fig. 7c). These residues exhibit high interaction fractions, indicating their significant role in stabilizing the ligand’s binding through hydrogen bonds, hydrophobic contacts, and possibly ionic interactions. The timeline representations complement this by illustrating the temporal persistence of these interactions over a 100 ns simulation. Some residues in EGFR (Fig. 7b), like GLU-738 and MET-769, don’t have as many highly stable, continuous contacts. This suggests a tighter and more focused binding mode. On the other hand, interactions in NAD(P)H oxidase (Fig. 7d) are more fluid and spread out across a larger group of residues, with GLY-12, GLU-32, and ASP-282 consistently binding. These images highlight differences in binding dynamics: NAD(P)H oxidase demonstrates a more stable, residue-specific interaction profile, whereas EGFR shows a broader and more flexible interaction landscape.

Fig. 7.

Protein-ligand contacts histogram and timeline of myricetin with EGFR and NAD(P)H oxidase receptor : (a) PL contacts histogram of myricetin-EGFR, (b) PL contacts timeline of myricetin-EGFR, (c) PL contacts histogram of myricetin-NAD(P)H oxidase, (d) PL contacts timeline of myricetin-NAD(P)H oxidase.

Protein secondary structure

The secondary structure elements simulation (SSE) of Myricetin-EGFR shows that residues 60–70, 100–130, and 200–290 are mostly made up of α-helices, while residues 1–50, 75–100, and 125–175 are mostly made up of β-strands (Fig. 8a). For the myricetin-EGFR and myricetin-NAD(P)H oxidase systems (Fig. 8a and b), the percentages of α-helices are 25.41% and 20.40%, and the percentages of β-strands are 15.67% and 25.41%. In these systems, the total percentages of secondary structures are 41.08% and 45.81%, respectively.

Fig. 8.

Protein secondary structure elements (SSE) like alpha-helices and beta-strands of myricetin with: (a) EGFR and (b) NAD(P)H oxidase receptor.

Ligand properties

We analyzed the ligand properties of myricetin-EGFR and myricetin-NADPH oxidase complexes over time using six metrics: RMSD, radius of gyration (Rg), intramolecular hydrogen bonds (IntraHB), molecular solvent-accessible surface area (MolSASA), SASA, and polar surface area (PSA). In the EGFR complex (Fig. 9a), the RMSD varied between 0.25 and 0.75°, while the Rg values indicated stable compactness, ranging from 3.80 to 3.95 Å. IntraHB exhibited frequent transitions, whereas MolSASA and SASA values ranged from 258 to 264 Ų and 25 to 75 Ų, respectively, suggesting moderate solvent exposure. The PSA value ranged from 315 to 325 Ų. The NADPH oxidase complex (Fig. 9b) had an RMSD range of 0.3° to 0.9° and a slightly lower Rg value range of 3.80° to 3.92 Å. This showed that the structure was similar but a little more flexible. IntraHB exhibited increased bonding transitions, while MolSASA and SASA indicated greater solvent exposure, measuring 258–267 Ų and 25–100 Ų, respectively. PSA values for NADPH oxidase were elevated (312–330 Ų), suggesting increased polar surface exposure. The findings indicate that both complexes demonstrate stable structural behaviors; however, the NADPH oxidase complex exhibits greater flexibility and solvent accessibility than the EGFR complex.

Fig. 9.

Ligand properties of myricetin with: (a) EGFR and (b) NAD(P)H oxidase receptor.

Conclusion

The present study highlights the strong antioxidant and antiproliferative potential of C. creticus leaf extracts, particularly those harvested in spring and autumn, when phenolic and flavonoid contents are highest. Seasonal variation markedly influences both the phytochemical composition and biological activities of the extracts. In silico analyses further support the interaction of key phytochemicals with therapeutic targets such as EGFR and NAD(P)H oxidase, offering a mechanistic insight into their bioactivity. These findings support the potential of C. creticus as a natural source for antioxidant and anticancer agents. Future work should explore clinical relevance and optimize extraction protocols for therapeutic development.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Yacine AOUIFFAT: Performed the plant extraction, chemical experiments, data curation, and formal analysis; Boulanouar BAKCHICHE: Designed and performed the experiments, interpreted the results, drafted the manuscript, and revised it; Farouk BENACEUR: Helped in experimental work; Mounir M. Bekhit: performed supervision and investigation; Mohamed M. Salem: manuscript revision; Mohamed Abbas Ibrahim: Resources and revision; Soumia MOUJANE: Software and docking; Aziz BOUYMAJANE: Software and docking; Imadeddine KADI: Resources and revision; Fathi BERRABAH: Supervision; Hicham GOUZI: Funding acquisition; Sanaa K. BARDAWEEL: Determined and discussed the anticancer activity, wrote the manuscript and revised it; Ashok K. SHAKYA: Conducted the chemical composition analysis; Mosad A. GHAREEB: Designed and conceived the study, participated in chemical profiling, wrote the manuscript, and revised it. All authors have read and agreed to the published version of the manuscript.

Funding

The authors would like to thank Amar Telidji University and the Ministry of Higher Education and Scientific Research for financial support.

Data availability

The authors confirm that the data supporting the findings of this study is available within the article. Raw data is accessible from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.