GC-MS metabolite profiling and multi-target Docking analysis of Calotropis procera and Euphorbia tirucalli stem extracts for cytotoxicity and antioxidant activity.

Introduction
Cancer is a globally prevalent disease characterized by the rapid and uncontrolled proliferation of abnormal cells within the body1. In 2020, an estimated 19.3 million new cancer cases were reported worldwide, resulting in over 10 million deaths. Among women, breast cancer has become the most commonly diagnosed cancer, surpassing lung cancer2. In 2022 alone, approximately 2.3 million women were diagnosed with breast cancer globally, and the disease claimed the lives of around 670,000 individuals. Although the incidence of breast cancer is higher in more developed countries, mortality rates are disproportionately greater in less developed regions. In Africa, breast cancer accounts for 28% of all cancer cases and contributes to a mortality rate of about 20% among affected women3. In Ethiopia, cancer is responsible for 5.8% of all deaths, with an estimated 60,000 new cases and over 44,000 deaths each year4. The emergence of drug resistance mechanisms limits the long-term effectiveness of chemotherapy, emphasizing the need for alternative bioactive molecules. C. procera (Apocynaceae) (Fig. 1, A), commonly known as “Sodom apple” or “Milkweed,” is a shrub native to arid and semi-arid regions of Africa and Asia5. It thrives in prolonged dry seasons with annual rainfall of more than > 150 mm 6. C. procera is known for its drought resistance and high biomass production, making it a potential fodder option for livestooks in dry and harsh environments7. C. procera has traditionally been used to treat a wide range of human ailments, including colds, fever, leprosy, asthma, rheumatism, eczema, indigestion, diarrhea, elephantiasis, various skin diseases, and dysentery5. In Saudi Arabia, a decoction prepared from the aerial parts of the plant is commonly used to relieve fever, joint pain, muscular spasms, and constipation8. In Burkina Faso, the plant is also employed in the management of neuropsychiatric disorder9. Notably, in Ethiopia, C. procera is traditionally used for the treatment of cancer10. Pharmacologically, C. procera extracts have been reported to exhibit anticancer, anti-inflammatory, antidiabetic, gastroprotective, antioxidant, antifungal, anthelmintic, analgesic, and antibacterial activities11. E. tirucalli (Euphorbiaceae) (Fig. 1, B), also known as “Pencil tree” or “Firestick plant,” is a succulent shrub or small tree native to tropical and subtropical regions of Africa and Madagascar. This plant has adapted to arid conditions and is known for its ability to survive in harsh environments with minimal water availability12. E. tirucalli is traditionally used to induce emesis for snakebite treatment. Its latex has shown therapeutic potential in managing conditions such as sexual impotence, skin disorders, swollen glands, edema, hemorrhoids, arthritis, epilepsy, dental and ear pain, and tumors13. This study aimed to evaluate the cytotoxic activity of the stem extracts of Ethiopian C. procera and E. tirucalli against the MCF-7 breast cancer cell line, analyze their chemical composition using GC-MS, assess their antioxidant properties, and conduct multi-target molecular docking against selected breast cancer-associated receptors.

Materials and methods

Collection and identification of the plant materials
Fresh stem parts of E. tirucalli and C. procera (Fig. 1) were collected in October 2023 from Moyale, Dawa zone, Somali Region of Ethiopia (latitude: 4° 09’ 60.00” N, longitude: 40° 04’ 60.00” E), approximately 800 km southeast of Addis Ababa. The plants were identified by Mr. Melaku Wendafrash, a taxonomist at Addis Ababa University’s National Herbarium, Department of Biology. The voucher specimens GN009 for E. tirucalli and GN10 for C. procera were deposited at National Herbarium, Department of Biology, Addis Ababa University, Ethiopia.

Extraction process
Powdered stem of C. procera and E. tirucalli (10 g each) were soaked in 100 mL of methanol and n-hexane for 24 h. After maceration, the mixtures were filtered using Whatman No. 1 filter paper. The resulting filtrates were first concentrated under reduced pressure at 40 °C using a rotary evaporator, followed by complete drying using a water bath (40 °C). The dried methanol and n-hexane extracts were stored at 4 °C for further analysis.

Gas chromatography-mass spectrometry (GC-MS) analysis
The chemical compositions of stem extracts of C. procera and E. tirucalli were analyzed using gas chromatography-mass spectrometry (GC-MS), with slight modification of a previously reported method14. The samples were analyzed using a GC-MS instrument from Agilent Technologies (Santa Clara, CA, USA) with a 6890 N network GC system, 5975 inert mass selective detector, 7683B series auto sampler injector (10 µL in size), HP5MS column (30 m length × 0.25 mm internal diameter × 0.25 μm film thickness), coated with 5% phenyl 95% methyl poly siloxane, and G1701DA GC/MSD Chem Station. 2 µL methanol and n-hexane solutions were injected through an auto sampler and analyzed using an HP5MS column. For the n-hexane extract of E. tirucalli, the temperature started at 60 °C, increased to 260 °C at a rate of 10 °C/min, then to 280 °C at 2 °C/min, and finally to 325 °C at 20 °C/min. For the n-hexane extract of C. procera, the temperature was initially set to 60 °C, then increased to 150 °C at 4 °C/min, to 260 °C at 6 °C/min, to 280 °C at 5 °C/min, and finally to 325 °C at 10 °C/min. For the methanol extracts of both plants, the temperature started at 70 °C, increased to 100 °C at 3 °C/min, then to 120 °C at 10 °C/min, and finally to 220 °C at 10 °C/min. A 3-minute solvent delay was applied. The mass spectra transfer line temperatures were set at 280 °C for the n-hexane extract and 220 °C for the methanol extract. The carrier gas was helium (1 mL/min) with a 100:1 split ratio. The mass spectra were acquired in electron ionization mode at 70 eV, with scanning from 50 to 500 amu (atomic mass unit) at 0.5 s with the mass source set to 230 °C. Phytochemical identification was conducted by comparing peak spectrum data to standard mass spectra from the spectrometer databases (NIST MS Search Program v. 2.0 software).

Cytotoxicity assay

Cell lines
To investigate the cytotoxic effect of the methanol extract from the stem of C. procera and E.tirucalli, the MCF-7 breast cancer cell line was used. The cells were cultured in DMEM media supplemented with 10% FBS and 1X Penicillin/Streptomycin solution. The cells were cultured at 37 °C with 5% CO2.

MTT assay
The in vitro cytotoxic activity of the methanol extract from the stem of C.procera and E. tirucalli was evaluated using the MTT assay. A stock solution of 1 mg/mL of the sample was prepared in DMSO. Treatment media were prepared with concentrations (50 µg/mL, 100 µg/mL, and 200 µg/mL) of the plant extract in media containing DMEM supplemented with 1X Penicillin/Streptomycin solution. The MCF-7 cells were seeded in triplicate in 96-well plates at a density of 10,000 cells/well in DMEM media supplemented with 10% FBS and 1X Penicillin/Streptomycin solution. Once the cells reached approximately 70% confluence, they were treated with the prepared treatment media. The MTT assay was performed 24 h after treatment. The spent media was removed, and 5 mg/mL stock of MTT reagent was added to each well. The plates were incubated for 4 h at 37 °C. The MTT reagent was then replaced with DMSO to dissolve the formed formazan crystals. Absorbance was measured at 570 nm using an ELISA plate reader. The results are presented as the mean of triplicate experiments.
The negative control consisted of non-treated cell cultures containing only the cells in growth medium, while doxorubicin was used as the positive control at concentrations equivalent to those of the tested extracts.

DPPH radical-scavenging assay
The antioxidant activity was evaluated using the DPPH free-radical scavenging assay15,16. The methanol extract were serially diluted in methanol in the concentration range from 100, 200, 400, 600, 800, and 1000 µg/mL from 1 mg/mL stock solution. To each of these diluted samples, 2 mL of DPPH solution (dissolved in methanol) was added, and the solutions were incubated for 30 min in the dark. Then, absorbance was read at 517 nm using a UV-Vis spectrophotometer.

ABTS radical-scavenging assay
A second in vitro method was conducted to assess the antioxidant potential of the extracts using the ABTS assay, following a procedure similar to that of a previous study15,16. Initially, a 7 mM ABTS solution of ABTS and 2.45 mM potassium persulfate solution were mixed and allowed in the dark at room temperature for 16 h until the reaction was complete and the absorbance stabilized. The resulting radical cation was further diluted with ethanol to adjust the absorbance value to 0.700 ± 0.02 at 734 nm using a UV-Vis Spectrophotometer. Test samples were separately dissolved in methanol to obtain a test solution of 1 mg/mL. A series of solutions with varying concentrations (100, 200, 400, 600, 800, and 1000 µg/mL) were prepared by diluting with methanol. To each well containing 1 mL of sample, 2 mL of ABTS •+ solution was added. The mixture was gently shaken and allowed to stand for 2 h in the dark at room temperature. The absorbance was then measured spectrophotometrically at 734 nm. The reactivity of the various concentrations of each extract was compared to that of ascorbic acid. All measurements were carried out in triplicate. The percent scavenging of ABTS •+ radical was calculated for different concentrations (100 to 1000 µg/mL) of extracts and standard using the following equation:
Where, A control is the absorbance of ABTS •+ (= 0.700 ± 0.02); A sample is the absorbance of sample + ABTS •+.

In silico pharmacokinetic and toxicity analysis
Phytochemicals identified in the GC-MS analysis with a relative abundance of ≥ 1% were selected for in silico pharmacokinetic and toxicity evaluation. Their pharmacokinetic properties were assessed using the SwissADME tool (http://www.swissadme.ch/) to determine drug-likeness based on Lipinski’s Rule of Five. Furthermore, toxicological endpoints and LD₅₀ values were predicted using the ProTox-II online platform (https://tox.charite.de).

Molecular Docking analysis
This study aimed to complement in vitro results by performing multi-target in silico molecular docking to predict the binding orientation and affinity of isolated compounds toward key proteins involved in antiproliferative and antioxidant activities.

Selection of protein
Four key target proteins were selected for molecular docking based on their critical roles in breast cancer and established therapeutic relevance. Estrogen receptor alpha (ERα) regulates nuclear DNA transcription and is central to estrogen-sensitive breast cancers17; tamoxifen, a selective estrogen receptor modulator, targets ERα. Cyclin-dependent kinases 4/6 (CDK4/6) control the G1 to S phase cell cycle transition, with overactivation driving proliferation in hormone receptor-positive, HER2-negative breast cancers18; palbociclib is an FDA-approved CDK4/6 inhibitor improving progression-free survival. Epidermal growth factor receptor (EGFR) activates growth signaling pathways; its overactivation promotes aggressive tumors, and gefitinib is the standard EGFR inhibitor19. Topoisomerase IIα (TOP2A) manages DNA topology during replication; inhibition causes DNA breaks triggering cancer cell apoptosis, targeted by doxorubicin20. These proteins represent diverse mechanisms of breast cancer progression, making them suitable docking targets.
For antioxidant activity, Human Peroxiredoxin 5 (Prdx5) and Human Myeloperoxidase (MPO) were selected, with ascorbic acid as the reference. Prdx5, an antioxidant enzyme, reduces reactive oxygen species (ROS) like hydrogen peroxide to water and alcohols via thiol-dependent catalysis, maintaining redox balance and protecting biomolecules from oxidative damage implicated in cancer and other diseases21. Human Myeloperoxidase (MPO), found in neutrophils and monocytes, generates reactive oxidants such as hypochlorous acid during inflammation; while essential for pathogen defense, excessive MPO activity contributes to oxidative stress and tissue injury. MPO inhibition can mitigate this damage22. Docking against Prdx5 and MPO aims to identify compounds that enhance antioxidant defense by supporting Prdx5 activity or inhibiting MPO’s pro-oxidant effects, benchmarked against ascorbic acid.

Selection of ligands
Ligands were selected from GC-MS identified compounds with a relative abundance of ≥ 1%. This cutoff prioritized major constituents likely to exert biologically relevant effects while minimizing computational complexity and enhancing the interpretability of the docking outcomes.

Protein and ligand preparation
Protein crystal structures were obtained from the RCSB Protein Data Bank (https://www.rcsb.org/), including Estrogen Receptor Alpha (PDB ID: 3ERT), Cyclin-Dependent Kinases 4/6 (5L2S), Epidermal Growth Factor Receptor (4LQM), Topoisomerase IIα (4FM9), Human Peroxiredoxin 5 (1HD2), and Human Myeloperoxidase (1DNU). Structures were imported into BIOVIA Discovery Studio Visualizer 2021, where water molecules, heteroatoms, and co-crystallized ligands were removed. Proteins were prepared for docking by adding Kollman and Gasteiger charges and AD4-type atoms using MGL Tools. Binding site coordinates (x, y, z) for each protein are listed in Table S1. 2D structures of selected phytochemicals and standard drugs were retrieved from PubChem in SDF format and converted to PDB via Open Babel. For compounds unavailable in PubChem, structures were manually drawn in ChemDraw 16.0 and 3D conformations generated. Energy minimization and optimization were performed using MMFF94 in ChemBio3D Ultra 13.0 until the RMSD gradient reached 0.01 kcal/mol, ensuring stable, low-energy conformations. Optimized ligands were saved in PDB, converted to PDBQT format with torsion adjustments via AutoDock Tools, and subjected to molecular docking via command-line interface. The best-scoring poses, with lowest binding affinities and RMSD, were analyzed using BIOVIA Discovery Studio Visualizer 2021 for binding energies, hydrogen bonds, and key amino acid interactions.

Result and discussion

Chemical constituents from the stem of C. procera
The methanol stem extract of C. procera revealed a diverse profile with 93.24% (22 components) of compounds identified through GC-MS analysis. Fatty acids were the predominant compounds, comprising 48% of the total composition, followed by terpenes (14%). Linoleic acid (23.13%), palmitic acid (13.5%), oleic acid (13.49%), β-amyrin (13.07%), and α-amyrin (7.76%) were among the principal constituents (Table 1; Figs. 2, and 4). The greater percentage of fatty acids found in C. procera suggests they play a role in anticancer activity23. Compared to the Pakistan ecotype, the chemical composition of the Ethiopian C. procera ecotype shows significant differences in both the number of identified compounds and the main constituents24. The GC-MS analysis of C. procera’s n-hexane extracts, presented in (Table 2; Figs. 3, and 4), demonstrated a diverse profile with 97.04% (23 components) identified. The major constituents identified were lupeol (26.05%), α-amyrin (19.5%), β-amyrin (16.2%), moretenone (6.27%), and lanosterol (4.84%). Lupeol, a major pentacyclic lupane-type triterpene identified in the extracts, is widely present in edible plants and medicinal herbs25. It exhibits diverse pharmacological activities, including antioxidant, anti-inflammatory, anti-hyperglycemic, antimicrobial, cardioprotective, and anticancer properties26,27. Similarly, α-amyrin and β-amyrin, commonly occurring triterpenes, have well-documented effects such as antitumor, anti-inflammatory, anxiolytic, hepatoprotective, gastroprotective, and antibiofilm activities28,29. GC-MS analysis confirmed that the n-hexane extract of C. procera stem predominantly contains triterpene derivatives30, consistent with findings from the chloroform root extract31. The ethyl acetate fraction of the root bark methanol extract also revealed lupeol and α-/β-amyrin using Soxhlet extraction32. Additionally, ethanolic leaf extracts identified 31 compounds, with α-amyrin (39.36%) as the major constituent33. However, chemotypic variations have been reported in C. procera from Iraq and Saudi Arabia, differing in composition except for phytol5,34. Such variability is influenced by environmental and physiological factors including habitat, salinity, temperature, altitude, seasonality, plant age, extraction method, and water availability35–37.

Chemical constituents from the stem of E. tirucalli
The chemical composition of E. tirucalli stem extracts (methanol and n-hexane) was analyzed by GC-MS, as detailed in Tables 3 and 4, with major peaks and components illustrated in Figs. 5, 6 and 7. The methanol extract contained 24 compounds representing 99.97% of the total mass, while the n-hexane extract included 21 compounds accounting for 98.13%. Terpenes were major ones constituting 58% and 52% of the methanol and n-hexane extracts, respectively. Previous studies have identified terpenoids as major bioactive components in E. tirucalli essential oils and solvent extracts36,38. Compared to the Malaysian E. tirucalli ecospecies39, the Ethiopian E. tirucalli showed notable differences in compound profiles, likely due to extraction methods and environmental factors. The dominance of triterpenes aligns with reports from other Euphorbia species such as E. hyssopifolia40 and E. ingens41. In the n-hexane extract, major triterpenes included lanosterol (36.65%), germanicol (12.24%), and lupenone (9.54%), while the methanol extract’s main triterpenes were lanosterol (30.42%), olean-18-ene (12.49%), lupenone (9.13%), and lanosta-8,24-dien-3-one (7.06%). Lanosterol, a major compound in several Euphorbia ecospecies from Yemen41 and Kenya42, was also dominant here, consistent with a GC-MS study on Egyptian E. tirucalli latex methanol extract43. Lanosterol exhibits chemopreventive and cytotoxic activities against colon cancer and other human cancer cell lines44. Germanicol has reported anti-inflammatory and antibacterial effects against Klebsiella pneumoniae and Pseudomonas aeruginosa45,46. Lupenone, common in medicinal plants, shows diverse pharmacological activities including anti-inflammatory, anti-diabetic, anticancer, and antiviral effects, enhancing the medicinal value of E. tirucalli47,48. Hentriacontane contributes additional anti-inflammatory, antitumor, and antimicrobial properties49. Olean-18-ene, a triterpenoid, demonstrates anti-HIV activity, further supporting the plant’s pharmacological potential50,51. The methanol extract also contained diterpenes 2-ketomanool and phytol, whereas the n-hexane extract included only phytol, reflecting the limited occurrence of diterpenes in Euphorbiaceae52.

Cytotoxic activity
The cytotoxic activity of methanolic extracts from C. procera and E. tirucalli was evaluated against MCF-7 breast cancer cells using standard in vitro assays. The results demonstrated a clear dose-dependent response for both extracts, with cell viability decreasing as extract concentration increased (Table 5; Fig. 8). The MCF-7 cells treated with E. tirucalli extract exhibited survival rates of 100 ± 0.53%, 46 ± 0.78%, and 36 ± 0.31% at concentrations of 50, 100, and 200 µg/mL, respectively. These findings are in line with a previous study conducted in South Africa, where butanol, hexane, and methanol extracts of E. tirucalli significantly inhibited MCF-7 cell proliferation by inducing G₀/G₁ phase cell cycle arrest (38%, 42%, and 67%, respectively) and reducing the S-phase population. Additionally, the hexane extract was shown to reduce caspase-3/7 activity, suggesting a non-apoptotic mechanism of growth inhibition53. Similarly, treatment with C. procera methanol extract resulted in MCF-7 cell survival rates of 68 ± 0.73%, 50 ± 0.66%, and 31 ± 0.98% at 50, 100, and 200 µg/mL, respectively. These observations are supported by a study from Riyadh, Saudi Arabia, which reported that C. procera latex reduced MCF-7 cell viability in a concentration-dependent manner, with cell viability decreasing to 31% at 500 µg/mL and 79% at 50 µg/mL 54. Additionally, study from Dubai, UAE, showed that the ethanol extract of C. procera leaves inhibited MCF-7 cell proliferation, with an IC₅₀ value of 50 µg/mL 11. Future studies should include additional cancer and normal cell lines to determine the selectivity index and further assess the extracts’ safety and therapeutic potential.

Antioxidant activity by DPPH assay
The antioxidant activity of extracts from the stem of C. procera and E. tirucalli was investigated through their capacity to reduce DPPH (2,2-diphenyl-1-picrylhydrazyl) radical. The antioxidant activity of extracts from the stem of C. procera and E. tirucalli, as well as ascorbic acid, was tested against the DPPH assay using concentrations ranging from 100 to 1000 µg/mL. Table 6; Fig. 9 present the findings regarding the scavenging potential of extracts from C. procera and E. tirucalli. Our results confirm that the free radical scavenging activity of both C. procera and E. tirucalli increased as the concentration of the methanol and n-hexane extracts solution reaching a maximum at 1000 µg/mL (around 50% inhibition). The IC50 values for C. procera were 18.52 and 10.79 µg/mL for n-hexane and methanol extracts, respectively. Studies have reported that different extracts of C. procera including the ethyl acetate fraction of the root bark methanol extract (IC₅₀ = 369.87 µg/mL), ethanolic bark extract (IC₅₀ = 28.57 µg/mL), and leaf extract (IC₅₀ = 366.33 µg/mL) exhibit varying antioxidant activities in the DPPH assay, likely due to differences in extract concentration, environmental conditions, and geographical factors affecting plant growth55–57. The IC50 values for E. tirucalli were 33.76 and 12.37 µg/mL, respectively. A study reported that the hydromethanolic extract of the aerial parts of E. tirucalli exhibited an IC₅₀ value of 12.15 µg/mL, while another study found that the phenolic-rich latex extract had an IC₅₀ value of 6 µg GAE/mL, both of which are comparable to our findings43,58. The ascorbic acid exhibited an IC50 of 4.48 µg/mL. The observed antioxidant activity of the C. procera and E. tirucalli extracts may result from the combined effects of multiple constituents, potentially involving synergistic or additive interactions, as suggested by GC-MS analysis59–61.

Antioxidant activity of extracts by ABTS assay
The oxidation of ABTS with potassium persulfate generates an ABTS radical cation. This radical cation is reduced in the presence of hydrogen-donating antioxidants. During the reaction, the blue ABTS radical cation was decolorized62. Table 7; Fig. 10 show the absorbance and percent inhibition of ABTS⁺ by extracts from the aerial parts of C. procera and E. tirucalli. A dose-dependent increase in scavenging activity was observed, with the highest inhibition at 1000 µg/mL for both extracts. The IC₅₀ values were as follows: for C. procera, n-hexane extract 11.98 µg/mL and methanol extract 7.24 µg/mL; for E. tirucalli, n-hexane extract 23.52 µg/mL and methanol extract 8.97 µg/mL. Ascorbic acid, used as the standard, had an IC₅₀ of 3.85 µg/mL.

Drug likeness and ADMET properties
Lipinski’s Rule of Five and Veber’s rule were applied to evaluate the drug-likeness of compounds identified by GC-MS. Lipinski’s criteria define favorable drug-like properties as MW < 500 Da, NHA < 10, NHD < 5, and LogP < 5. Veber’s rule predicts oral bioavailability based on TPSA ≤ 140 Ų and NRB ≤ 10 63. All phytochemicals from C. procera and E. tirucalli met Lipinski’s rule, while 15/20 compounds from C. procera and 15/24 from E. tirucalli complied with Veber’s rule, indicating good oral bioavailability (Tables S2-S5). None were predicted to interact with P-glycoprotein, supporting effective intestinal absorption. Only 6/24 E. tirucalli compounds, and none from C. procera, potentially inhibited CYP1A2, with no inhibition predicted for CYP2C19, CYP2D6, or CYP3A4, suggesting low risk of CYP-mediated drug interactions and strong therapeutic potential.

Toxicity prediction of phytochemicals
The major compounds identified in the methanol and n-hexane stem extracts of C. procera were diethyl phthalate, linoleic acid, α-amyrin, and β-amyrin which showed predicted LD₅₀ values > 5000 mg/kg, placing them in toxicity class 6 (non-toxic) based on acute oral toxicity models. Similarly, major constituents from E. tirucalli stem extracts, lanosterol and lupeol, had predicted LD₅₀ values of 2000 mg/kg (toxicity class 4), while olean-18-ene and germanicol exhibited LD₅₀ values of 5000 and 70,000 mg/kg, corresponding to classes 5 and 6, respectively. In silico toxicity profiling also indicated that all tested phytochemicals from both plants were non-hepatotoxic, non-carcinogenic, non-mutagenic, non-immunotoxic, and non-cytotoxic (Tables S6 and S7).

Molecular Docking
The present study aimed to complement the in vitro findings with in silico molecular docking analysis by predicting the binding orientation and affinity of phytochemicals identified through GC-MS towards selected anticancer and antioxidant target proteins. Compounds with a relative abundance of ≥ 1% were chosen for docking studies.

Binding mode analysis of phytochemicals targeting human myeloperoxidase (PDB ID: 1DNU)
GC-MS analysis of C. procera identified various phytochemicals in both methanol and n-hexane extracts. Molecular docking showed that 14 compounds exhibited lower binding energies toward human myeloperoxidase (MPO) compared to ascorbic acid (-5.8 kcal/mol) (Table S8). The triterpenoids α-amyrin (-10.8 kcal/mol), lupeol (-10.1 kcal/mol), and β-amyrin (-9.7 kcal/mol) emerged as promissing candidates, consistent with their known antioxidant, anti-inflammatory, and cytoprotective activities64,65. α-Amyrin formed one hydrogen bond with GLU-116 and eighteen van der Waals interactions with residues such as ASP-98, ARG-333, PHE-366, and LEU-420 (Table S10, Fig. 11). Lupeol established two hydrogen bonds (HIS-95, ARG-239), one Pi-alkyl interaction (PHE-147), and seventeen van der Waals interactions. β-Amyrin formed one hydrogen bond with GLU-102 and seventeen van der Waals contacts. Similarly, 15 compounds from E. tirucalli extracts showed stronger MPO binding compared ascorbic acid. Lanosterol (-10.3 kcal/mol), germanicol (-10.1 kcal/mol), and olean-18-ene (-10.4 kcal/mol) were among the top ones (Table S9). Lanosterol formed two hydrogen bonds (ARG-424, GLU-102), three Pi-alkyl interactions (HIS-95, PHE-332, HIS-336), and fourteen van der Waals interactions (Table S10, Fig. 11).

Binding mode analysis of phytochemicals targeting human Peroxiredoxin 5 (PDB ID: 1HD2)
Molecular docking analysis showed that 15 compounds from the methanol and n-hexane stem extracts of C. procera exhibited lower binding energies to human peroxiredoxin 5 (PDB ID: 1HD2) than ascorbic acid, suggesting stronger antioxidant potential. Notably, the triterpenoids β-amyrin (-6.3 kcal/mol), α-amyrin (-6.4 kcal/mol), and lupeol (-6.9 kcal/mol) displayed significantly stronger binding affinities compared to ascorbic acid (-4.5 kcal/mol) (Table S8). β-Amyrin primarily formed hydrophobic alkyl interactions with VAL-80 and van der Waals contacts with ILE-119, ARG-124, ASN-76, ALA-42, PHE-43, PHE-120, THR-44, and PRO-45. α-Amyrin established one hydrogen bond with ASN-76, one alkyl interaction with ILE-119, and van der Waals interactions with PHE-120, PRO-45, THR-44, ARG-124, ALA-42, PHE-43, and VAL-80 (Table S10, Fig. 12). Lupeol interacted via one alkyl bond with LEU-149 and thirteen van der Waals contacts, including residues PRO-40, LEU-116, PHE-120, ILE-119, THR-117, ARG-127, GLY-148, BR-303, THR-44, PRO-45, GLY-46, LYS-49, and THR-50. Similarly, 16 compounds from E. tirucalli extracts demonstrated stronger binding than ascorbic acid. Of these, lanosterol (-6.7 kcal/mol), germanicol (-6.8 kcal/mol), and olean-18-ene (-6.4 kcal/mol) showed higher binding affinity (Table S9). Lanosterol formed one hydrogen bond with ASN-76, hydrophobic interactions with PRO-45, LEU-116, and ILE-119, and a Pi-alkyl interaction with PHE-120, along with five van der Waals interactions involving THR-44, GLY-41, ALA-42, PHE-43, and VAL-80 (Table S10, Fig. 12).

Binding mode analysis of phytochemicals targeting cyclin-dependent kinases 4/6 (PDB ID: 5L2S)
Molecular docking analysis showed that β-amyrin from C. procera, and β-amyrin and olean-18-ene from E. tirucalli methanol stem extracts exhibited stronger binding affinities to the target protein than the reference drug palbociclib (-9.6 kcal/mol), with binding energies of -9.8 to -10.0 kcal/mol. Other major compounds, α-amyrin (-9.1 kcal/mol) and lanosterol (-9.0 kcal/mol), also showed comparable binding affinities (Table S8 and S9). β-Amyrin interacted with the target protein primarily through hydrophobic interactions, including Pi-sigma interaction with PHE-98, and alkyl interactions with VAL-27 and ALA-162. Additionally, van der Waals interactions were observed with residues VAL-77, LYS-43, ASP-163, TYR-24, LYS-147, LEU-152, ILE-19, ASN-150, GLY-20, and GLN-149 (Table S11, Fig. 13). Lanosterol formed seven hydrophobic interactions with the target protein, including Alkyl interactions with ILE-19, VAL-27, ALA-41, ALA-162 (twice), VAL-77, and LEU-152, as well as a Pi-alkyl interaction with PHE-98. Furthermore, it established twelve van der Waals interactions involving GLY-20, GLN-149, LYS-43, ASP-104, ASP-163, ASN-150, TYR-24, THR-182, LYS-147, ALA-23, GLY-22, and VAL-181 (Table S11, Fig. 13). Olean-18-ene primarily interacted with the target protein through three alkyl hydrophobic interactions with VAL-27, ALA-41, and ALA-162, along with eleven van der Waals interactions.

Binding mode analysis of phytochemicals targeting epidermal growth factor receptor (PDB ID: 4LQM)
Molecular docking analysis revealed that 14 compounds from C. procera and 13 from E. tirucalli methanol and n-hexane stem extracts exhibited lower binding energies than the reference drug gefitinib (Table S8 and S9). Among the major constituents, α-amyrin (-9.9 kcal/mol) and lupeol (-9.1 kcal/mol) from C. procera showed stronger binding affinities. In contrast, lanosterol from E. tirucalli exhibited a binding energy of -7.5 kcal/mol, comparable to gefitinib (-7.5 kcal/mol). β-Amyrin primarily engaged the target protein through hydrophobic interactions, including a Pi-Sigma interaction with PHE-723 and twelve van der Waals contacts (Table S11, Fig. 14). Lanosterol formed two alkyl interactions with ALA-755 and ILE-759, along with twelve van der Waals interactions involving key residues such as ARG-841, PHE-723, GLU-758, and ASP-855 (Table S11, Fig. 14).

Binding mode analysis of phytochemicals targeting Estrogen receptor alpha (PDB ID: 3ERT)
Molecular docking analysis revealed that three compounds α-amyrin, β-amyrone, and moretenone from C. procera, and four compounds olean-18-ene, β-amyrone, germanicol, and friedelan-3-one from the methanol and n-hexane stem extracts of E. tirucalli exhibited lower binding energies than the reference drug tamoxifen (-9.7 kcal/mol) (Tables S8 and S9). Among the major constituents, α-amyrin from C. procera showed the strongest affinity (-9.9 kcal/mol), while germanicol and lanosterol from E. tirucalli showed  - 9.8 and - 8.7 kcal/mol, respectively compared to tamoxifen (-9.7 kcal/mol). α-Amyrin formed one hydrogen bond with ASP-351, a Pi-alkyl interaction with TYR-526, and twelve van der Waals interactions involving LEU-536, TRP-383, THR-347, and others (Table S11). Lanosterol engaged in one hydrogen bond with ASP-351, two alkyl interactions with MET-522, two Pi-alkyl interactions with TYR-526, and nine van der Waals interactions with residues including MET-528, LEU-525, and CYS-530 (Table S11, Fig. 15). Germanicol formed a hydrogen bond with GLU-380 and several van der Waals interactions with key residues such as TYR-526, LYS-529, and TRP-383.

Binding mode analysis of phytochemicals targeting topoisomerase IIα (PDB ID: 4FM9)
Molecular docking analysis showed that four compounds from C. procera and five from E. tirucalli methanol and n-hexane stem extracts exhibited stronger binding affinities than doxorubicin (-10.1 kcal/mol) (Table S8 and S9). Among major constituents, α-amyrin and β-amyrin from C. procera showed binding energies of -10.8 and - 10.7 kcal/mol, respectively, while germanicol from E. tirucalli showed - 10.3 kcal/mol. β-Amyrin formed one hydrogen bond with LYS-863 and fifteen van der Waals interactions involving residues such as GLN-938, TRP-931, and DA-12 (Table S11, Fig. 16). α-Amyrin similarly formed a hydrogen bond with LYS-863 and fourteen van der Waals interactions, including DC-11, THR-930, and LEU-995. Germanicol also formed one hydrogen bond with LYS-863 and fourteen van der Waals interactions with key residues.

Conclusion
Methanolic extracts from the stem aerial parts of C. procera and E. tirucalli showed dose-dependent growth inhibition of MCF-7 breast cancer cells, with inhibitory percentage of 31 ± 0.98% and 36 ± 0.31%, respectively, suggesting notable anticancer potential. Both the methanol and n-hexane stem extracts of C. procera and E. tirucalli demonstrated significant antioxidant activity in the DPPH and ABTS assays. In both assays, a dose-dependent increase in radical scavenging activity was observed, with the highest inhibition recorded at 1000 µg/mL for all extracts. In the DPPH assay, the IC₅₀ values for C. procera were 10.79 µg/mL (methanol extract) and 18.52 µg/mL (n-hexane extract), while for E. tirucalli, the values were 33.76 µg/mL and 12.37 µg/mL, respectively. In the ABTS assay, the IC₅₀ values were 7.24 µg/mL (methanol) and 11.98 µg/mL (n-hexane) for C. procera, and 8.97 µg/mL (methanol) and 23.52 µg/mL (n-hexane) for E. tirucalli. GC-MS analysis identified various compounds in both methanol and n-hexane stem extracts of C. procera and E. tirucalli, revealing key triterpenoid constituents such as α-amyrin, β-amyrin, lanosterol, germanicol, and lupeol, as well as fatty acids like linoleic acid and oleic acid, which are believed to contribute to the observed biological activities. The pharmacokinetic properties of the identified compounds with a composition of ≥ 1% were evaluated using SwissADME, and molecular docking studies to predict their binding affinities to multiple breast cancer and antioxidant related protein targets , aiming to identify potential drug candidates. All selected phytochemicals from both stem extracts of C. procera and E. tirucalli complied with lipinski’s rule of five, and their toxicity profiles revealed no hepatotoxic, carcinogenic, mutagenic, or cytotoxic effects. Molecular docking analysis showed that lupeol, β-amyrin, α-amyrin, lanosterol, olean-18-ene, and germanicol had strong binding affinities with human myeloperoxidase (-10.1 to -10.8 kcal/mol) and estrogen receptor alpha (-7.5 to -9.9 kcal/mol), outperforming reference compounds ascorbic acid (-5.8 kcal/mol) and gefitinib (-7.5 kcal/mol). Based on the combined results of cytotoxicity, antioxidant, molecular docking, and ADMET analyses, C. procera and E. tirucalli show significant potential as sources of bioactive compounds, warranting further isolation of active constituents, along with in vivo and mechanistic studies to fully elucidate their therapeutic potential.

Supplementary Information
Below is the link to the electronic supplementary material.