Eugenoplatin (a phyto-chemo drug conjugate) targets cancer stem cells through inhibition of the β-catenin signaling in triple-negative breast cancer cells

Abstract

Background

Breast cancer is the most prevalent and leading cause of cancer-related mortality among women worldwide. The triple-negative subtype, the most aggressive form of the disease, is associated with the poorest prognosis. In order to improve the treatment of these patients, we have synthesized a phyto-chemo drug conjugate composed of cisplatin and eugenol, two molecules with synergistic anti-cancer effects.

Methods

Eugenoplatin was synthesized and characterized by HPLC, LC/MS, and ¹H-NMR, while its physicochemical properties were predicted by in silico analysis using the Biovia Discovery Studio. The cytotoxic effects of eugenoplatin were first tested in vitro using the WST1 and the flow cytometry techniques. Cell proliferation, migration, and invasion abilities were assessed using the xCELLigence Real-Time Cell Analysis. The changes in gene expression were determined by proteomics analysis (LC/MS), immunoblotting, immunofluorescence, and quantitative RT-PCR. The effect of the drug on breast cancer stem cells was determined by the tumorsphere formation assay, and cell sorting was utilized to isolate a specific breast cancer sub-population of cells. Furthermore, orthotopic tumor xenografts were used to test the potential therapeutic value of eugenoplatin in vivo.

Results

The novel DNA-damaging molecule (eugenoplatin) has shown higher cytotoxicity through induction of apoptosis, and strong inhibition of cell growth via cell cycle delay at G2/M phase in different cell lines. In addition, eugenoplatin targeted cancer stem cells through inhibition of the β-catenin signaling pathway. These effects were confirmed in humanized orthotopic tumor xenografts, wherein the eugenoplatin-dependent tumor growth inhibition was more than 10-fold stronger than the cisplatin repressive effect, with a potent inhibitory effect on cancer stem cell biomarkers. This was associated with the suppression of various cancer-related signaling pathways, including STAT3/NF-κB, AKT/mTOR, and ERK. Furthermore, eugenoplatin down-regulated the pro-angiogenic factors VEGF-A, IL-6, and IL-8, and repressed angiogenesis both in vitro and in tumor xenografts. In silico analysis has suggested good permeability, good absorption, and higher bioavailability for eugenoplatin, which is also unlikely to be mutagenic, carcinogenic, or hepatotoxic.

Conclusion

These results indicate that eugenoplatin, a novel DNA-damaging agent that can also target cancer stem cells, could be of great therapeutic value for TNBC patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-08007-2.

Introduction

Breast cancer (BC) is the most common type of cancer and the leading cause of cancer-related deaths among women worldwide [1]. Triple-negative breast cancer (TNBC), the most aggressive subgroup of the disease, accounts for approximately 15%–20% of newly diagnosed breast cancer worldwide. Because TNBC lacks the targetable receptors that are expressed in other BC sub-types, treating this notoriously aggressive type of breast cancer is challenging, with conventional chemotherapy remaining the primary treatment option for patients with early and advanced-stage disease [2].

Cisplatin (cis-diamminedichloroplatinum II) is a well-known metal-based DNA-damaging chemotherapeutic drug, which has been used for the treatment of different types of cancer, including breast cancer [3]. However, the clinical application of cisplatin is often limited owing to severe side effects, especially nephrotoxicity. Indeed, even low and non-toxic concentrations of cisplatin in the serum could be toxic in the kidneys, and 30–40% of cisplatin-treated patients develop acute kidney injury [4]. Furthermore, the emergence of resistance to cisplatin is very frequent, which substantially handicaps treatment efficiency and enhances mortality in cancer patients. This chemoresistance is a multifactorial process mainly due to the enrichment of a small population of tumor-initiating cells with self-renewal and tumor regenerative capacities called cancer stem cells (CSCs), which are also believed to be responsible for tumor relapse and metastasis [5–7]. Therefore, strategies need to be developed to address cisplatin-related toxicities and drug resistance through the development of more effective and less toxic alternatives. To overcome these clinical limitations, different cisplatin analogues were synthesized and are currently in use for the treatment of various types of tumors. These analogues include carboplatin and oxaliplatin, which showed efficiency and less toxicity [8, 9]. Furthermore, several molecules, including eugenol has increased the anti-cancer effects of cisplatin. Indeed, we have recently shown that eugenol, a phenolic natural compound present essentially in clove oil with anti-cancer potential, can potentiate the effect of cisplatin against breast cancer and ovarian cancer cells both in vitro and in vivo [10, 11].

Based on these synergistic data and the wide anticancer effects of both molecules (cisplatin and eugenol), we have developed and synthesized a combined molecule eugenol-cisplatin, designated eugenoplatin, as a novel phyto-chemo drug conjugate (PCDC), and have shown its potent anti-cancer efficiency against various TNBC cell lines both in vitro and in orthotopic tumor xenografts.

Methods

Synthesis of the combined cisplatin-eugenol molecule (eugenoplatin)

Eugenol (0.304 mM) and NaOH (0.320 mM) were mixed in water and stirred at room temperature for 120 min. Subsequently, cisplatin (0.306 mM) was added to the eugenol solution, and the mixture was stirred at room temperature for 24 h. The next day, a yellow solution was formed in addition to a sticky solid. The yellow solution was removed by centrifugation, and the sticky solid was washed three times with 1.0 mL of methanol. At this stage, the sticky solid turned into a creamy precipitate, which was isolated by centrifugation and then dried under vacuum.

HPLC

The formed product (eugenoplatin) was first analyzed by HPLC using analytical C18 columns and dual detection UV Detector.

ADMET and drug likeness analysis

Drugs were screened for detailed analysis of physicochemical descriptors and drug likeness through the Lipinski rule of five and pharmacokinetics-associated variable i.e., absorption, distribution, metabolism, excretion, and toxicity (ADMET) [12]. The in silico prediction of mutagenicity and carcinogenicity of cisplatin and eugenoplatin was determined using the Ames test [13]. For the toxicity prediction of chemicals, the TOPKAT protocol was used to predict specific toxicological effects. TOPKAT toxicity prediction and ADMET rapidly access mutagenicity (Ames Test), carcinogenicity, and hepatotoxicity level of chemical compounds, based on Quantitative Structure-Toxicity Relationship (QSTR) model through Biovia Discovery Studio v24.1.0.321712.

Cells, cell culture, and reagents

MCF-10 A, MDA-MB-231, BT-20, and human umbilical vein endothelial cells (HUVEC) cell lines were purchased from ATCC, SUM149PT cell line was purchased from Asterand Bioscience. Cells were regularly screened for mycoplasma contamination using the Mycoplasma Detection Kit (ATCC). Cells were cultured as recommended and were grown at 37 °C in an incubator (humidified, 5% CO2). MCF-10 A was cultured in DMEM/F12 medium supplemented with 10% Fetal Bovine Serum (FBS), 1% Antibiotic-Antimycotic (AB), 1% HuMEC, and 0.5% Bovine Pituitary Extract. MDA-MB-231 and HUVEC cells were cultured in RPMI medium supplemented with 10% FBS and 1% AB. BT-20 was cultured in EMEM medium supplemented with 10% FBS and 1% AB. SUM149PT was cultured in RPMI medium supplemented with 10% FBS and 1% AB, 5 µg/ml insulin and 1 µg/ml hydrocortisone. HFSN1 (primary normal human skin fibroblast) was cultured in DMEM/F12 medium supplemented with 10% FBS and 1% Antibiotic-Antimycotic (AB). All supplements were obtained from Gibco except insulin and hydrocortisone from Sigma-Aldrich and EMEM from ATCC. Cisplatin and eugenol were purchased from Sigma-Aldrich. Carboplatin was purchased from Fresenius Kabi.

Transfections

CTNNB1-ORF and its corresponding control (1 µg) (Origene) were used to transfect cells using Lipofectamine 2000 following the manufacturer’s instructions (Invitrogen). Transfected cells were selected using Neomycin sulfate (4 mg/ml) (Sigma-Aldrich). CTNNB1-siRNA and its corresponding control (30 nM) (Origene) were used to transfect cells using Lipofectamine RNAiMAX following the manufacturer’s instructions (Invitrogen).

Cytotoxicity assay

This has been performed as previously described [14]. Briefly, cells (5 × 10³) were seeded in 96-well plates with appropriate culture media. After cell treatment for 72 h, WST1 reagent (Roche) was added to each well according to the manufacturer’s instructions. The absorbance of the samples was measured using a microplate (ELISA) reader at 450 nm (Bio-Rad). IC₅₀ was determined by GraphPad Prism. These experiments were performed in triplicate and repeated several times.

RNA purification and quantitative RT-PCR

RNA was isolated using mRNeasy mini kit (Qiagen) following the manufacturer’s instructions. RNA (1 µg) was used for reverse transcription with Advantage RT-PCR kit (Clontech Laboratories). FastStart Essential DNA Green Master (Roche) was used for quantitative real-time PCR on the LightCycler^® 96 Real-time PCR detection system (Roche). Primer sequences are as follows:

Primers	Sequence
IL-6	Forward	5’-AGACAG CCA CTC ACC TCT TCA G -3’
	Reverse	5’- TTC TGC CAG TGC CTC TTT GCT G -3’
ALDH1A1	Forward	5’- TTT GGA AGA TAG GGC CTG CA -3’
	Reverse	5’- AGG CCC ATA ACC AGG AAC AA -3’
CD44	Forward	5′- TGGTGAACAAGGAGTCGTCA-3’
	Reverse	5′- GTTAAGTGTCCCAGCTCCCT-3’
AXIN2	Forward	5’-ATGCAAAAGCCACTCCAAGG − 3’
	Reverse	5’-CTCACTCTCCAGCATCCACT − 3’
C-MYC	Forward	5’-CCTGGTGCTCCATGAGGAGAC-3’
	Reverse	5’- CAGACTCTGACCTTTTGCCAGG-3’
SOX2	Forward	5’-GCTACAGCATGATGCAGGACCA-3’
	Reverse	5’-TCTGCGAGCTGGTCATGGAGTT-3’
CCND1	Forward	5’- CAC GCG CAG ACC TTC GTT − 3’
	Reverse	5’-CGA TGG AGG GCG GAT TG-3’
GAPDH	Forward	5’-GAGTCCACTGGCGTCTTC-3’
	Reverse	5’-GGGGTGCTAAGCAGTTGGT-3’

Cellular lysate preparation and immunoblotting

This has been performed as previously described [14]. Antibodies directed against cleaved-PARP (Asp214), cleaved caspase-3 (Asp175), cleaved caspase-9 (Asp315), Sox2, β-Catenin, Bmi1, Cyclin D1, STAT3, p-STAT3 (Tyr705), pNF-κB p65 (Ser536), NF-κB p65, p-mTOR (Ser2448), mTOR, p-AKT (Thr308), AKT, p-ERK1/2 (Thr202/Tyr204), ERK1/2, p-P53 (Ser15) and α-tubulin were purchased from Cell Signaling Technology. IL-6, IL-8, VEGF-A, Twist1, Ki-67, and DNMT1 were purchased from Abcam. GAPDH, Cyclin A, CDK2, CD24, p21, and E2F1 were purchased from Santa Cruz Biotechnology. CD44 and HDAC were purchased from Sigma-Aldrich. ALDH1 was purchased from BD Biosciences.

Immunofluorescence

This has been performed as previously described [14]. Briefly, cells (2 × 10⁵) were fixed in formaldehyde (4%) and blocked with goat serum (5%), Triton X (0.3%), and sodium azide (0.05%) for 1 h. Cells were then stained overnight at 4 °C with γH2AX antibody (Novus Biologicals). Subsequently, incubated with Alexa Fluor 594-conjugated goat anti-rabbit IgG secondary antibody and DAPI (Thermo Fisher Scientific) for 1 h. Images were acquired using a fluorescence microscope and ZEN Microscopy Software (ZEISS).

Cell proliferation, migration, and invasion assays

These assays were carried out using xCELLigence Real-Time Cell Analysis (RTCA) (Agilent Technologies) as previously described [15]. In brief, for cell migration and invasion, (5 × 10³) cells were seeded with serum-free medium (SFM) in the upper chamber wells of the CIM-plate, non-coated (migration) or pre-coated (invasion) with a thin layer of Matrigel (1:40) (BD Biosciences). Serum-containing medium (10%) was used as a chemoattractant in the lower chamber wells. The plate was incubated in the RTCA system for 24 h. For the proliferation assay, cells (5 × 10³) were seeded in E-plates with serum-containing medium (10%), and then incubated for 72 h. The RTCA pro software was used to analyze the obtained results expressed as Cell Index (CI) values.

Apoptosis and cell cycle analysis

Cells (3 × 10⁵) were harvested, centrifuged, and stained using either the dead cell apoptosis kit (Invitrogen) for cell apoptosis analysis or with propidium iodide (PI) (Invitrogen) for cell cycle analysis following the manufacturer’s instructions. Cells were then analysed using the LSR II flow cytometer and the BD FACSDiva operating software (BD Biosciences).

Flow cytometry for cell surface marker analysis and cell sorting

Cells (3 × 10⁵) were double-stained with CD44 APC (Miltenyi Biotec) and CD24 PerCP (BioLegend) antibodies for 30 min on ice. For CD44^high/CD24^low population analysis, cells were analyzed using the LSR II flow cytometer and the BD FACSDiva operating software. For CD44^high/CD24^low cell sorting, cells were sorted using a BD FACSMelody flow cytometer and the BD FACShorus operating software (BD biosciences).

Immunohistochemistry staining on FFPE tissues

Formalin-fixed paraffin-embedded tissue section slides were stained with CD31, CD34, or Ki-67 antibodies (Abcam) overnight at a dilution of 1:100, or with Hematoxylin and Eosin using an automated staining platform (Ventana). Envision + Dual Link System-HRP (DAB+) (Dako) was used as a secondary antibody. Images were acquired using an inverted microscope and CellSense Standard Microscopy Software (Olympus).

Spheroid formation

Cells (1 × 10³) were seeded in 96-well ultra-low attachment plates with stem cell-specific medium (DMEM/F12 medium supplemented with 1% AB, 2% B-27, 20 ng/mL EGF, 20 ng/ml bFGF, 500 ng/ml hydrocortisone, 2 U/ml Heparin, 4% FBS, and 5 µg/ml insulin). After 10 days, spheroids with a diameter of ≥ 100 μm were counted and photographed using Floid™ Cell Imaging Station (Life technologies).

In vitro angiogenesis assay

Serum-free conditioned media (SFCM) were collected after culturing cells in SFM for 24 h, and then were mixed with HUVEC cells (4 × 10⁴) and were seeded into pre-coated 96-well plates with Geltrex (Gibco). After 16 h, the formed tube-networks were counted and photographed using an inverted light microscope (OLYMPUS).

Orthotopic tumor xenografts

Breast cancer orthotopic xenografts were created by implantation of MDA-MB‐231 cells (5 × 10⁶) under the nipple of female nude mice (Mus musculus, Nu/J, 6 weeks). When the average tumor volume reached 100 mm³, mice were randomized into the following treatment conditions: Dimethyl sulfoxide (DMSO), 0.2 mg/kg eugenoplatin, and 2 mg/kg cisplatin administered weekly via intraperitoneal injection. Tumor size was measured twice a week. After 4 weeks of treatment, tumors were surgically retrieved and processed for both immunoblotting and immunohistochemistry. For tumor Fragment Implantation, a small piece of tumor (2–3 mm³) was cut and subcutaneously implanted into the flank of 3 female nude mice, left side: DMSO, right side: eugenoplatin. After tumor growth, tumors were measured. Tumor volume was calculated using this formula: (width² × length/2). Animal experiments were approved by the KFSH&RC Institutional Animal Care and Use Committee (ACUC) under RAC# 2,220,020 and were conducted according to relevant national and international guidelines.

Proteomics analysis using label-free quantitative liquid chromatography mass spectrometry (LC/MS)

Proteomics analysis was performed using label-free quantitative liquid chromatography–mass spectrometry (LC/MS). Equal amounts of protein (100 µg), extracted in the presence of protease inhibitors, were subjected to solution tryptic digestion as previously described [16, 17]. The resulting peptides were analyzed by label-free quantitative nano-LC/MS using a NanoAcquity liquid chromatography system coupled to a Synapt G2 mass spectrometer equipped with a Trizaic nano-flow source (Waters Scientific, Manchester, UK).

Mass spectrometry data were acquired over an m/z range of 50–2000 Da using a 120-min gradient and data-independent acquisition with ion mobility separation (MS^E/HDMS^E). Each biological sample was analyzed in triplicate technical runs using the MassLynx platform (version 4.1, SCN833).

Raw MS data were processed using Progenesis QI for Proteomics (QIfP, version 3.0; Nonlinear Dynamics, Newcastle, UK) for chromatographic alignment, feature detection, normalization, and relative quantification. MS/MS spectra were searched within Progenesis QI using its integrated database-search workflow, supporting standard search engines (SEQUEST, Mascot, and PLGS), against UniProt/Swiss-Prot Homo sapiens protein sequences. A target–decoy strategy was applied during database searching to control false discovery rates (FDR), and peptide-spectrum matches and protein identifications were filtered to an FDR of ≤ 1%.

Protein abundances were calculated from normalized peptide ion intensities. Differential protein abundance between experimental groups was assessed using analysis of variance (ANOVA) as implemented in Progenesis QI for Proteomics. To account for multiple hypothesis testing across the quantified proteome, p-values were adjusted using the Benjamini–Hochberg false discovery rate correction. Proteins with an adjusted p-value (q-value) ≤ 0.05 were considered statistically significant and were included in downstream pathway and functional enrichment analyses.

Functional pathway and network analyses

Ingenuity Pathways Analysis (IPA) (QIAGEN, https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis) was used for functional, canonical pathway, and protein-protein interaction network analyses. Differentially expressed proteins (DEPs) were mapped to their corresponding objects in the Ingenuity Pathway Knowledge Base and gene interaction networks. A right-tailed Fisher’s exact test was used to calculate p-values. All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant.

Statistical analysis

Statistical analysis was performed using a two-tailed unpaired Student’s t-test. P values of 0.05 and less were considered statistically significant.

Results

Synthesis and characterization of eugenoplatin

The conjugate molecule (eugenol-cisplatin: eugenoplatin) was first analyzed by HPLC. Figure 1A shows the presence of a one 100% single peak with no other peaks or starting material. The novel pick (molecule) had a retention time of 5 min, while the retention times of eugenol and cisplatin are 3 min and 16 min, respectively (Fig. 1A). This suggests the formation of a third molecule with new physical features and an expected molecular weight of 427. 79 Da with the most possible structure as shown in Fig. 1A. The LC/MS analysis of the molecule showed one main peak of MS + 1 = 452, which corresponds to the molecule plus sodium (428.0 + 23=451) (Fig. 1B). The structure of the molecule was further confirmed by ¹H-NMR, showing the presence of a strong metal-oxygen chemisorption bond (Fig. 1C). The compound name (4-allyl-2-methoxyphenoxy)diaminoplatinum(IV)chloride, has an empirical formula (C10H15C1N2O2Pt) and a neutral molecular weight of 426 Da.

Fig. 1.

HPLC, LC/MS, ¹H-NMR, and in silico analysis of eugenoplatin. A, Eugenol, cisplatin, and their combination (eugenoplatin) were analyzed by HPLC using analytical C18 columns and a dual detection UV detector. B, LC/MS analysis of eugenoplatin. The mass spectrum is a graphical representation of the mass-to-charge ratios (m/z) of ions obtained using a Synapt G2 HD mass spectrometer. The x-axis represents the m/z values, while the y-axis indicates the relative intensity/abundance of each ion. C, ¹H-NMR spectrum of eugenoplatin (300 MHz, DMSO-d6) δ 6.51–6.73 (m, 3 H), 5.92 (s, 1H), 5.00 (m, 1H), 4.75 (m, 3 H), 3.99 (s, 3 H), 3.70–3.74 (d, J = 13.5 Hz, 2 H), 3.26 (s, 2 H). D, The ADMET-based prediction of drug absorption for cisplatin and eugenoplatin compounds through Biovia Discovery Studio v24.1.0.321712. ADMET Descriptors, 2D polar surface area (PSA 2D) in Å 2 for each compound is plotted against their corresponding calculated atom-type partition coefficient (ALogP98)

Drug-likeness potential of eugenoplatin and cisplatin

For drug-likeness screening, we utilized Lipinski’s Rule of Five (LRo5) [16], which showed that cisplatin and eugenoplatin fall under the acceptable scores of LRo5. Particularly, the chemical structure of eugenoplatin can donate 5 hydrogen bonds (HB) and accepts 4 HB, Molecular weight (M.W) 427.79 Da, AlogP value is -0.321, and thus, violating no rule of LRo5 (Table 1). In contrast, cisplatin (used as control) exhibited 2 HBA, 6 HBD, M.W. 300.40 Da, AlogP value is -2.53, violating one rule of LRo5 (Table 1). The ALogP analysis revealed that eugenoplatin and the standard cisplatin have LogP values below 5, which is within the desirable threshold (LogP ≤ 5). According to the Veber rule [17], eugenoplatin exhibited a molecular polar surface area (MPSA) value of 72.5 Ų, which is in the acceptable range (70–140 Ų) and below the desired threshold (140 Ų). However, the MPSA value for cisplatin was 53.6 Ų. Veber’s criteria are upper-threshold rules (PSA ≤ 140 Ų; rotatable bonds ≤ 10) used to flag potential poor oral bioavailability, so a PSA of 53.6 Ų exhibited by cisplatin is also not violating the Veber rule (Table 1). The number of rotatable bonds and the number of HBD and HBA from both cisplatin and eugenoplatin did not violate the Veber rule (Table 1).

Table 1.

Drug-likeness properties of cisplatin and eugenoplatin

Lipinski Rule					Veber Rule
Compound	No. H Acceptor (HBA ≤ 10)	No. H Donor (HBD ≤ 5)	Mol. Wt. (M.W. ≤ 500 Da)	ALogP (LogP ≤ 5)	No. of Rotatable bonds (≤ 10)	MPSA (≤ 140 Ų)	No. H Acceptor (HBA ≤ 12)	No. H Donor (HBD ≤ 12)
*Cisplatin	2	6	300.405	-2.53	0	53.6	2	2
Eugenoplatin	4	6	427.79	-0.32	5	72.06	2	2

* Represents the reference standard chemical. MW: Molecular Weight; HBD: Hydrogen bond doner; HBA: Hydrogen bond acceptors; MPSA: Molecular polar surface area

ADMET variables and toxicity prediction analysis of cisplatin and eugenoplatin

Standard levels of ADMET descriptors were calculated using the Biovia Discovery Studio v24.1.0.321712 (Table S1). The calculated ADMET descriptors for cisplatin and eugenoplatin compounds are shown in Table 2. The obtained results indicate that while cisplatin has no aqueous solubility, eugenoplatin exhibited optimal or low aqueous solubility (Table 2). Both compounds can act as non-inhibitors of CYP2D6 and can be assimilated in Phase-I metabolism [18] (Table 2). Cisplatin and eugenoplatin have shown less plasma protein binding parameters (Table 2). The binding of drugs to plasma proteins may reduce their bioavailability. Therefore, the PPB for eugenoplatin and cisplatin reflects the lower binding affinity of the compounds to plasma proteins (Table 2). A low or weak binding indicates that a greater proportion of the drug will be available in its free (active) form in circulation. Therefore, the ADMET test indicates that both compounds can reach the desired targets. The hepatotoxicity prediction through ADMET suggested cisplatin as hepatotoxic and eugenoplatin as a non-hepatotoxic compound (Table 2).

Table 2.

Predicted ADMET properties of cisplatin and eugenoplatin

S. No	Compound	ADMET Solubility_Level	ADMET BBB Level	ADMET CYP2D6	ADMET Hepatoxicity	ADMET PPB_Level	ADMET Absorption_Level
1	*Cisplatin	No	Low	Non-Inhibitor	Toxic	> 95%	Very poor
2	Eugenoplatin	Yes-Low	Very Low	Non-Inhibitor	Non-Toxic	> 95%	Good

* Represents the reference standard compound. BBB: blood brain barrier; PPB: Plasma protein binding; PSA_2D: Fast polar surface area_2D; CPA: Cisplatin analogue; CC.: Chemical compound; V/L: value/level; V/P: value/prediction

In the ADMET method of Biovia Discovery Studio, blood brain barrier absorption model contains a quantitative linear regression model for the prediction of absorption of a compound as 95% and 99% confidence ellipses in the ADMET_PSA_2D and ADMET_AlogP98 plane [19]. Using ADMET_PSA_2D and ADMET_A logp98 properties, blood-brain barrier (BBB) and Human Intestinal Absorption (HIA) ADME plots for cisplatin and eugenoplatin are shown in Fig. 1D. Based on the ADMET plot (Fig. 1D) and the physicochemical descriptors summarized in the ADMET prediction (Table 2 and Table S1), clear differences were observed between cisplatin and eugenoplatin. According to ADMET_PSA_2D and ADMET_AlogP98, eugenoplatin showed a good absorption level as compared to cisplatin, which showed poor absorption (Fig. 1D A and Table 2). The efficiency of a particular drug relies on its intestinal absorption and distribution to target organs. As shown in the BBB-PLOT, cisplatin falls outside the 95% and 99% BBB confidence ellipses, indicating very low BBB permeability (BBB Level 4–5; very low/undefined intensity) (Fig. 1D). This suggests that cisplatin is unlikely to cross the blood–brain barrier, reducing the likelihood of CNS-related toxicity. In contrast, eugenoplatin appears inside, but not in the center of the 95% and 99% BBB ellipses, corresponding to low to very low BBB permeability (BBB Level 3–4) (Fig. 1D), implying that eugenoplatin may or may not penetrate the CNS. Whether BBB penetration is beneficial or harmful depends on the molecule’s therapeutic target and toxicity profile. These compounds, cisplatin and eugenoplatin, also passed AMES mutagenicity and carcinogenicity prediction tests (Table 3). The Ames test prediction for eugenoplatin and cisplatin indicates that these compounds are unlikely to cause mutations and do not pose a significant genetic risk in terms of mutagenicity (Table 3). However, TOPKAT carcinogenicity prediction suggested cisplatin as a carcinogen but not eugenoplatin (Table 3).

Table 3.

Toxicity prediction of chemical compounds using TOPKAT toxicity prediction model

S. No.	Compound	TOPKAT Ames Prediction (Mutagenicity)	TOPKAT-Carcinogenicity Prediction
1	Cisplatin	Non-Mutagen	Carcinogenic
2	Eugenoplatin	Non-Mutagen	Non-Carcinogen

* Represents the reference standard compound; CC: chemical compound

Eugenoplatin promotes apoptosis through the mitochondrial pathway in TNBC cells

Next, we sought to study the cytotoxic effect of eugenoplatin and its widely used analogues, cisplatin and carboplatin, on TNBC cells. To this end, three TNBC cell lines (MDA-MB-231, BT-20, and SUM149PT) were treated with increasing concentrations of eugenoplatin and either cisplatin or carboplatin for 72 h, and then the cytotoxicity was analyzed using the WST1 assay, and the IC₅₀ was determined for each treatment.

As compared to cisplatin, the cytotoxic effect of eugenoplatin, as determined by IC₅₀, was 30.8-fold higher in MDA-MB-231 (Fig. 2A). Similarly, eugenoplatin was more potent than carboplatin, with IC₅₀ 44.3- and 39.3-fold higher in BT-20 and SUM149PT, respectively (Fig. 2A).

Fig. 2.

Eugenoplatin promotes apoptosis through the mitochondrial pathway in TNBC cells. A, Cells were either sham-treated (DMSO) or challenged with the indicated drugs and concentrations for 3 days. Cell viability was determined using the WST1 assay. Dotted lines indicate IC_₅₀. B, Cells were treated as indicated for 3 days. Representative flow cytometry plots of normal, apoptotic, and necrotic cells stained with Annexin V/PI. C, Bar charts showing the percentages of apoptotic and necrotic cells determined by flow cytometry. D, MDA-MB-231, BT-20, and SUM149PT cells were either sham-treated (DMSO) or challenged with eugenoplatin (3 µM) and cisplatin (50 µM), eugenoplatin (8 µM) and carboplatin (100 µM), and eugenoplatin (0.3 µM) and carboplatin (3 µM), respectively, for 3 days. Whole-cell lysates were used for immunoblotting of the indicated apoptotic markers. GAPDH serves as a loading control. The numbers underneath each band represent fold changes relative to the control (DMSO) after correction against the corresponding internal controls. E, Bar charts showing the Bax/Bcl-2 ratio determined by immunoblotting in (D). Each experiment was repeated 3 times. Error bars represent mean ± SEM

To test the specificity of eugenoplatin for TNBC cells, non-cancerous mammary epithelial cells (MCF-10 A) were challenged with eugenoplatin. Figure S1A shows that the cytotoxicity of eugenoplatin was minimal on MCF-10 A cells, indicating a high selectivity of the drug for cancer cells.

Later, we decided to elucidate the eugenoplatin-related killing mechanism. To this end, MDA-MB-231, BT-20, and SUM149PT cells were treated with eugenoplatin, cisplatin, or carboplatin for 72 h, and then the cells were stained with annexin V/PI and analyzed by flow cytometry. The obtained results show that eugenoplatin induced cell death through apoptosis in the three TNBC cell lines, with only marginal necrosis (Fig. 2B). The selectivity for cancer cells was confirmed by showing that eugenoplatin has only a slight effect on MCF-10 A cells (Figure S1B). Interestingly, eugenoplatin was much more effective compared to cisplatin and carboplatin (Fig. 2C).

To confirm eugenoplatin-dependent induction of apoptosis, we have performed immunoblotting analysis for the major apoptotic markers, namely cleaved PARP, cleaved caspase 3, and cleaved caspase 9. Similar to cisplatin and carboplatin, eugenoplatin promoted the intrinsic apoptotic pathway in TNBC cells by increasing the expression of these key apoptotic markers (Fig. 2D). The induction of the mitochondrial pathway was further confirmed by showing an increase in the Bax/Bcl-2 ratio in response to eugenoplatin, cisplatin, and carboplatin in the 3 cell lines (Fig. 2E). Together, these results indicate that eugenoplatin has a remarkable anti-cancer effect in TNBC cells compared to cisplatin and carboplatin, and induces cell death through the mitochondrial apoptotic pathway.

Cisplatin-resistant TNBC cells are sensitive to eugenoplatin

One of the major hurdles of the chemotherapy-based treatments is the formation of resistant cells with the potential to grow and spread. To address this problem, we developed cisplatin-resistant MDA-MB-231 cells following treatment with 50 µM. Indeed, while the IC₅₀ for the parental cells (MDA^P) is 30 µM, it’s more than 100 µM for cisplatin-resistant cells (MDA^R) (Figure S2A). When cells were treated with eugenoplatin, the IC₅₀ of MDA^P was 1.8 µM and reached 4.7 µM for MDA^R cells (Figure S2B). This shows the high sensitivity of these cisplatin-resistant cells to eugenoplatin. To confirm this, we studied the effect of eugenoplatin on the invasive capacity of MDA^P and MDA^R cells using the noninvasive and real-time technology (RTCA-DP xCELLigence System). Figure S2C shows that while both cisplatin (50 µM) and eugenoplatin (3 µM) inhibited the invasive capacity of MDA^P cells, only eugenoplatin inhibited the invasive potential of MDA^R cells. This shows that eugenoplatin can suppress the metastatic potential of these cisplatin-resistant cells.

Eugenoplatin affects several physiological processes related to cancer progression in TNBC cells

In order to delineate the effect of eugenoplatin on protein expression in TNBC cells, MDA-MB-231, BT-20, and SUM149PT cells were either sham-treated or challenged with 3, 8 or 0.3 µM eugenoplatin for 24 h, respectively. Whole cell lysates were prepared and subjected to one-dimensional label-free quantitative global proteomic analysis. Unsupervised hierarchical clustering of differentially expressed proteins (DEPs) revealed 242, 82 and 742 DEPs that were significantly dysregulated (p < 0.05, fold change > 2) in MDA-MB-231, BT-20 and SUM149PT cells, respectively (Figure S3A, B). The Venn Diagram depicted in Figure S3B shows 14 proteins commonly differentially expressed across all three cell lines. Figure 3A shows that treatment with eugenoplatin affected several important physiological pathways with well-known roles in cancer onset and spread, such as DNA synthesis and cell cycle checkpoints. The graphical summary illustrates overview of the core analysis of major biological diversities, interdependence and inferred relationships of the regulation of different cellular processes and related biological activities of the 742 dataset of DEPs between pairs of eugenoplatin-treated and control SUM149PT cells (Fig. 3B). The key aspects of directly or indirectly regulated cellular processes by the cancer stem cells biomarker CTNNB1 includes: cell proliferation, cell cycle, migration and invasion of breast cancer cells among others (Fig. 3B).

Fig. 3.

Eugenoplatin affects various cancer-related physiological pathways. A, Comparison Analysis of Pathways illustrated by a heat map for several canonical pathways showing trends and clusters of pathway scores among datasets of differentially expressed proteins (DEPs) between pairs of control and eugenoplatin-treated cells (MDA-MB-231 (M), BT-20 (B), and SUM149PT (S)). Arrows indicate pathways of particular interest. A gradient of colors is used to represent the scaling range of expression values, with orange and blue indicate degree of up- and down representation, respectively, while dots represent absence. The image was generated in IPA comparison Analysis (Qiagen). B, Graphical Summary illustrating overview of the core analysis of major biological diversities, interdependence, and inferred relationships of the regulation of different cellular processes and related biological activities of the 742 dataset of differentially expressed proteins (DEPs) between pairs of control and eugenoplatin-treated SUM149PT cells. The color-coded molecules and processes in solid orange are predicted activity activation, and blue represents predicted inhibition. Orange and blue lines represent activation and inhibition of relationships between nodes, respectively. While solid and broken grey lines represent direct and indirect interaction, respectively, the faint broken lines represent inferred relationships in the connections. (The image was generated in IPA core Analysis (Qiagen))

Eugenoplatin delays the cell cycle at G2/M, and inhibits the migratory/invasive capacities of TNBC cells

Based on the results depicted in Fig. 3, we have decided to test the effect of eugenoplatin on the proliferative and the migratory/invasive capacities of TNBC cells using the real-time RTCA-DP xCELLigence System. Therefore, MDA-MB-231, BT-20, and SUM149PT cells were treated with eugenoplatin as described above. Figure 4A shows clear eugenoplatin-related proliferation inhibition for the 3 cell lines. Similar inhibitory effects were also observed for the migratory and the invasive capacities of cells (Fig. 4A).

Fig. 4.

Eugenoplatin inhibits the proliferation, migration, and invasion abilities of TNBC cells and causes cell cycle arrest. MDA-MB-231, BT-20, and SUM149PT cells were either sham-treated (DMSO) or challenged with eugenol (3, 8, and 0.3 µM, respectively) for the indicated periods of time. A, Representative charts for cell proliferation, migration, and invasion abilities, which were assessed using the RTCA-DP xCELLigence System. B, Representative flow cytometry plots for the DNA contents of cells that were stained with PI. The proportions of cells in the different cell cycle phases are indicated. C, Immunoblotting for the indicated cell cycle markers upon treatment for 24 h. GAPDH serves as a loading control. Each experiment was repeated 3 times

The eugenoplatin-dependent inhibition of cell proliferation prompted us to study the effect of the drug on the cell cycle. Therefore, MDA-MB-231, BT-20, and SUM149PT were treated as described above for different periods of time (0, 24, and 48 h), and then cellular DNA content was quantified by PI/flow cytometry. Figure 4B shows a clear eugenoplatin-dependent delay in G2/M phase of the cell cycle in the 3 cell lines. Indeed, the proportion of cells in G2/M reached 57%, 33.5%, and 55.2% in MDA-MB-231, BT-20, and SUM149PT cells, respectively (Fig. 4B). These results indicate that eugenoplatin delays the cell cycle in G2/M phase in TNBC cells. The eugenoplatin-dependent delay in G2/M phase was confirmed at the molecular level by showing that eugenoplatin downregulates two important genes, cyclin A and CDK2, in the 3 tested cell lines (Fig. 4C).

Eugenoplatin inhibits stemness features in breast cancer cells

Next, we tested the ability of eugenoplatin to target breast cancer stem cells in the 3 cell lines MDA-MB-231, BT-20, and SUM149PT, which were treated as described above for 24 h. Immunoblotting analysis shows marked downregulation of the major BCSC markers ALDH1, CD44, and Sox2, whereas CD24 was upregulated in the 3 cell lines compared with their corresponding controls (Fig. 5A). This indicates that eugenoplatin suppresses stemness features in TNBC cells. To confirm this, the proportion of CD44^high/CD24^low subpopulations was determined by flow cytometry. Figure 5B shows a decrease in the proportion of CD44^high/CD24^low subpopulations in eugenoplatin-treated cells compared with their respective controls. Furthermore, we examined ALDH activity and have shown eugenoplatin-dependent reduction in the ALDH activity in the 3 cell lines relative to their respective controls (Fig. 5C). In addition, we examined the effect of eugenoplatin on the self-renewal potential of breast cancer cells using an in vitro spheroid formation assay. Figure 5D shows that eugenoplatin treatment significantly decreased the number of spheroids in MDA-MB-231, BT-20 and SUM149PT cells compared with their respective controls. This suggests eugenoplatin-dependent targeting of cancer stem cells in TNBC cells. To confirm this, the CD44^high/CD24^low subpopulation was sorted from MDA-MB-231 cells, and then was either sham-treated (DMSO) or challenged with eugenoplatin (3 µM). This treatment was highly cytotoxic for these cells (Fig. 5F), and also suppressed their migratory capacity (Fig. 5G). Together, these findings show that eugenoplatin can efficiently target CSCs sub-populations in TNBC cells.

Fig. 5.

Eugenoplatin inhibits the stemness features of breast cancer cells. MDA-MB-231 and BT-20 and SUM149PT cells were either sham-treated (DMSO) or challenged with eugenoplatin (3, 8 and 0.3 µM, respectively) for 24 h. A, cell lysates were prepared and used for immunoblotting analysis, with GAPDH utilized as an internal control. The numbers underneath each band represent fold changes relative to the control (DMSO) after normalization to GAPDH. B, Cells (2 × 10⁵) were double-stained for both CD24 and CD44, and the proportions of the subpopulation CD44^high/CD24^low were determined by flow cytometry and are shown in the boxes. C, ALDH activity was assessed by the Aldefluor assay and flow cytometry. The numbers in the boxes represent the proportion of ALDH ^high cells. D, Cells (1 × 10³) were seeded in an ultra-low attachment 96‐well plate containing stem cell‐specific medium (100 µl). The formed spheroids (> 100 μm) were counted. Representative photographs of spheroids (left panels), number of formed spheroids (right panels). Error bars represent mean ± SEM (n = 3). *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. E, Cells (1 × 10³) were seeded in an ultra-low attachment 96‐well plate containing stem cell‐specific medium (100 µl). When spheroids were formed, they were either sham-treated (DMSO) or challenged with eugenol or cisplatin using the indicated concentrations for 3 days. Cell viability was determined using the WST1 assay and plotted relative to control cells (DMSO). Error bars represent mean ± SEM (n = 3). *P ≤ 0.05; ***P ≤ 0.001; ^$P ≤ 0.0001. F and G, CD44^high/CD24^low MDA-MB-231cells were sorted by flow cytometry, and then were either sham-treated (DMSO) or challenged with eugenoplatin (3 µM). F, Cell viability was determined using the WST1 assay. Error bars represent mean ± SEM (n = 3), *P ≤ 0.05. G, The migration ability was determined using the RTCA-DP xCELLigence System

Eugenoplatin suppresses stemness in breast cancer cells through inhibition of the Wnt/β-catenin pathway

In order to delineate the molecular pathway that mediates eugenoplatin-dependent inhibition of stemness in BC cells, we made use of the IPA analysis shown in Fig. 3B, which indicates a central role of the CTNNB1 and its target C-MYC gene in the response of TNBC cells to eugenoplatin. Therefore, we have first tested the effect of eugenoplatin on the important stemness-regulator pathway β-catenin. Figure 5A shows that eugenoplatin reduced the expression of β-catenin in the 3 cell lines. Since β-catenin is a nuclear protein, we investigated the effect of eugenoplatin on the translocation of β-catenin to the nucleus. Therefore, the level of this transcription factor was assessed in the nuclear and cytoplasmic fractions prepared from eugenoplatin-treated MDA-MB-231 and BT-20 cells and their corresponding controls. The immunoblots showed eugenoplatin-related decrease in the nuclear β‐catenin level in both cell lines compared with their respective controls, whereas the cytoplasmic levels of β‐catenin were only slightly affected by the treatment (Fig. 6A). Therefore, eugenoplatin treatment reduces the accumulation of β‐catenin into the nucleus of breast cancer cells. This was confirmed by showing that the eugenoplatin treatment downregulates the β‐catenin target genes CD44, AXIN2, ALDH1A1, SOX-2, and IL-6 (Fig. 6B). Interestingly, the important β‐catenin downstream oncogenic effectors C-MYC and CCND1 were also downregulated in response to eugenoplatin at both the mRNA and protein levels (Figs. 5A and 6B). These results indicate that eugenoplatin is a potent inhibitor of the Wnt/β‐catenin signaling pathway, and therefore could suppress the stemness features through this CSC-regulatory pathway.

Fig. 6.

Eugenoplatin suppresses stemness in breast cancer cells through inhibition of the Wnt/β-catenin pathway. MDA-MB-231 and BT-20 cells were either sham-treated (DMSO) or challenged with eugenoplatin (3 and 8 µM, respectively) for 24 h. A, Cytoplasmic and nuclear lysates were prepared for immunoblotting analysis, and α-tubulin and HDAC1 were utilized as respective internal controls. The numbers underneath each band represent fold changes relative to the control (DMSO) after correction against specific internal controls. B, qRT-PCR, Error bars represent mean ± SEM (n = 3). *P ≤ 0.05; **P ≤ 0.01. C and D, MDA-MB-231 cells were transfected with CTNNB1-ORF or an empty vector (CTNNB1-ORF and CT-ORF, respectively), and the generated cells were either sham-treated (DMSO) or challenged with eugenoplatin (3 µM) for 24 h. (C) Whole cell lysates were prepared and used for immunoblotting analysis. The numbers underneath each band represent fold changes relative to the control (DMSO) after normalization to GAPDH. D, Cells (1 × 10³) were seeded in an ultra-low attachment 96‐well plate containing stem cell‐specific medium (100 µl). The formed spheroids (> 100 μm) were counted. Representative photographs of spheroids (left panel), number of formed spheroids (right panel). Error bars represent mean ± SEM (n = 3). **P ≤ 0.01; ns: not significant. E, MDA-MB-231 cells were transfected with CTNNB1-specific siRNA (CTNNB1-si) while a scrambled sequence was used as control (CTNNB1-CT), and then the cells were challenged with the indicated concentrations of eugenoplatin for 3 days. Cell viability was determined using the WST1 assay. Dotted lines indicate IC_₅₀. Error bars represent mean ± SEM (n = 3). *P ≤ 0.05; ***P ≤ 0.001

To confirm the implication of the β-catenin pathway in eugenoplatin-dependent inhibition of stemness in TNBC cells, we introduced the CTNNB1-ORF or an empty vector into MDA-MB-231 cells (CTNNB1-ORF and CTNNB1-CT, respectively). In DMSO-treated cells, the level of the β-catenin protein was increased in ORF-expressing cells as compared to control cells (Fig. 6C). As expected, eugenoplatin downregulated β-catenin in both CTNNB1-ORF and CTNNB1-CT cells. On the other hand, while eugenoplatin affected the expression of the stemness markers CD24, CD44 and ALDH in CT-ORF cells, it didn’t affect the expression of these genes in CTNNB1-ORF cells (Fig. 6C). These results were confirmed by showing eugenoplatin-dependent inhibition of the self-renewal capacity of CT-ORF cells but not those expressing the β-catenin coding gene CTNNB1-ORF (Fig. 6D). These results show that eugenoplatin suppresses the stemness-characteristics of BC cells through inhibition of the Wnt/β-catenin pathway. Indeed, CTNNB1 knockdown with specific siRNA sensitized MDA-MB-231 cells to eugenoplatin as compared to their control cells (Fig. 6E). On the other hand, ectopic expression of the CTNNB1 gene enhanced the resistance of MDA-MB-231 cells to eugenoplatin compared to controls (Fig. 6E). This shows that eugenoplatin targets CSCs through the Wnt/β-catenin signaling pathway.

In addition to the canonical β-catenin pathway, eugenoplatin suppressed 2 other stemness-related pathways STAT3 and NF-κB, in the 3 cell lines tested (Fig. 5A). Indeed, eugenoplatin reduced the level of the active/phosphorylated forms of both STAT3 and NF-κB with only marginal effects on their basal levels (Fig. 5A). This shows that eugenoplatin can suppress the formation of CSCs through the inhibition of these 3 major stemness-related pathways, namely Wnt/β-catenin, STAT3, and NF-κB.

Eugenoplatin inhibits tumor growth and targets cancer stem cells in orthotopic tumor xenografts

To confirm the anti-cancer effect of eugenoplatin, we decided to test its effect in vivo in humanized orthotopic breast tumor xenografts. To this end, breast cancer xenografts were created by injecting MDA-MB‐231 cells (5 × 10⁶) orthotopically into the mammary fat pad of nude mice (n = 15). Upon tumor growth, mice were randomized into 3 groups (n = 5 each), which were treated with DMSO (100 µL), cisplatin (2 mg/kg in 100 µL corresponding to 60 mM), or eugenoplatin (0.2 mg/kg in 100 µL corresponding to 4.67 mM) once weekly via intraperitoneal injection. Figure 7A shows significant inhibition of tumor growth in response to both cisplatin and eugenoplatin as compared to the control (DMSO). Importantly, the effect of eugenoplatin was significantly higher than that of cisplatin, even though the used concentration was 10-fold lower than that of cisplatin (Fig. 7A). To confirm these results, tumors were extracted, and whole cell lysates were prepared, to test the effect on relevant cancer-related proteins. Figure 7B shows that cisplatin and eugenoplatin induced the cleavage of the pro-apoptotic proteins PARP, caspase 3, and caspase 9, showing the induction of apoptosis through the mitochondrial pathway in vivo as well. The proliferation marker cyclin D1 was also strongly downregulated in response to cisplatin and eugenoplatin (Fig. 7B). Figure 7C shows a strong reduction in the staining intensity of the proliferation marker Ki-67 in response to both drugs, with a stronger effect of eugenoplatin. Additionally, eugenoplatin had a strong effect on the BCSCs markers in tumor xenografts as well. Indeed, eugenoplatin downregulated CD44 and ALDH1, and upregulated CD24 (Fig. 7B). Like in vitro, eugenoplatin downregulated β-catenin and its downstream effector BMI1 (Fig. 7B). These results indicate that eugenoplatin inhibits the formation of cancer BCSCs in humanized tumor xenografts as well. To confirm this, we performed a post-treatment reimplantation experiment using nude mice. Therefore, BC tumor xenografts were created as described above, and when tumors became palpable, animals were randomized into 2 groups (n = 5), which were treated with DMSO or eugenoplatin (0.2 mg/kg) once weekly via intraperitoneal injection. After treatment, tumors were excised, cut, and then reimplanted under the skin of nude mice. Figure 7D shows significant eugenoplatin-dependent inhibition of tumor growth. When reimplanted, DMSO-treated tumors generated significantly bigger tumors than eugenoplatin-treated tumors (Fig. 7E). This shows that eugenoplatin targets BCSCs in orthotopic tumor xenografts.

Fig. 7.

Eugenoplatin inhibits tumor growth, EMT, and stemness in orthotopic breast tumor xenografts. A, Breast cancer xenografts were created by injecting MDA-MB‐231 cells (5 × 10⁶) orthotopically into the mammary fat pad of nude mice (n = 15). When tumors reached a volume of 100 mm³ in 12 mice, animals were randomized into 3 groups (n = 4 each), which were treated as follows: DMSO, cisplatin (2 mg/kg), and eugenoplatin (0.2 mg/kg) once weekly via intraperitoneal injection. Graph depicting tumor volume fold change over time, and error bars indicate means ± SD (n = 5). Representative photographs showing the size of the formed tumors in each group. B, Following the treatments, tumors from each group were excised, and protein extracts were prepared and used for immunoblotting analysis. The numbers under the bands represent fold change relative to the control (DMSO). C, Paraffin-embedded tissues were immunostained with the indicated antibodies. D, Breast cancer xenografts were created by injecting MDA‐MB‐231 cells (5 × 10⁶) orthotopically into the mammary fat pad of nude mice (n = 10). When tumors reached a volume of 100 mm³, mice were randomized into 2 groups (n = 5 each), which were treated with DMSO or eugenoplatin (0.2 mg/kg) once weekly via intraperitoneal injection. Error bars indicate means ± SD (n = 5). ^$P ≤ 0.0001. E, After 4 weeks of treatment, tumors were excised, cut into small pieces, and then were reinjected under the skin of 3 nude mice: DMSO-treated pieces in the left sides and eugenoplatin-treated pieces in the right sides. When tumors grew, their volumes were measured and depicted in the graph. Error bars indicate means ± SD (n = 3), *P ≤ 0.05. F, Cells were treated as indicated, and then the media were changed and replaced with drug-free SM for 24 h. The resulting SFCM were used to treat HUVEC cells, and then the formed cavities were photographed and counted under a microscope in 5 representative regions. Error bars represent mean ± SEM (n = 3). *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Cells were treated as indicated, and then whole cell lysates were used for immunoblotting analysis

Eugenoplatin has a potent anti-angiogenic effect

During the analysis of the effect of eugenoplatin on the various cancer-related pathways in tumor xenografts, we have found a strong inhibitory effect on the pro-angiogenic factors VEGF-A, IL-6 and IL-8 (Fig. 7B). This prompted us to investigate the potential anti-angiogenic effect of eugenoplatin. Therefore, we made use of the paraffin-embedded tissues to assess the density of blood vessels in the treated tissues using the endothelial cell biomarkers CD31 and CD34. Figure 7C shows that eugenoplatin had a potent inhibitory effect on blood vessel density, stronger than that of cisplatin, despite the fact that cisplatin was 10 times more concentrated than eugenoplatin. This anti-angiogenic effect of eugenoplatin was confirmed in vitro by showing a strong eugenoplatin-dependent inhibitory effect on the differentiation of endothelial cells (HUVEC) into cavities when exposed to SFCM from the 3 eugenoplatin-treated TNBC cells as compared to their corresponding controls (Fig. 7F). These results were confirmed by showing that eugenoplatin downregulates the expression of 3 important pro-angiogenic factors VEGF-A, IL-8, and HIF-1α, in the three TNBC cell lines MDA-MB-231, BT-20, and SUM149PT (Figure S4). Together, these findings show that eugenoplatin is a strong inhibitor of angiogenesis.

Eugenoplatin targets the PI₃K/AKT and ERK1/2 cancer-related signaling pathways

Ingenuity pathway analysis (IPA) of DEPs in the TNBC cell lines showed that eugenoplatin can target several proteins in the important pro-carcinogenic ERK1/2 and AKT signaling pathways (Fig. 8A, B). These proteins play major roles in the prognosis and response to therapy of breast cancer (Fig. 8A, B). Eugenoplatin-dependent targeting of ERK1/2 and AKT signaling pathways was confirmed by immunoblotting. Indeed, Fig. 8C shows eugenoplatin-dependent inhibition of AKT and ERK1/2 in the 3 cell lines and in orthotopic tumor xenografts. While the basal expression of both proteins was only marginally affected, the level of the active/phosphorylated forms was reduced upon eugenoplatin treatment compared to controls (Fig. 8C). Furthermore, eugenoplatin inhibited mTOR, a downstream effector of AKT (Fig. 8C).

Fig. 8.

Eugenoplatin can target ERK1/2 and PI₃K/AKT signaling pathways in TNBC cells. A and B, Ingenuity pathway merged annotations of some of the DEPs in (A) MDA-MB-231 and SUM149PT cells with some of the molecules that were associated with cell death and survival, cellular function and maintenance, and protein synthesis, and mapped in ERK½ signaling network. B, MDA-MB-231, BT-20, SUM149PT cells with some molecules associated with cancer, cardiovascular disease, and developmental disorder, and mapped in the AKT signaling network. The networks display an interactive graphical representation of the interrelationships between different molecules. The implicated proteins interacting with other genes and enzymes are represented in various shapes and colors. The molecules from the analyzed dataset in red/pink colors with fold change and p values of expression changes displayed underneath represent up-regulation, while those in green color are down-regulated. Other molecules are those that were added from the Ingenuity Knowledge Base that formed the network together. Overlays in the network are molecules and genes involved with biomarkers (BM) and canonical pathways (CP). The images were generated in IPA Core Analysis (Qiagen). IPA networks were visually simplified to highlight key hub molecules and principal pathways relevant to the study objectives. Edge labels and non-essential overlays were removed, and node labels were enlarged to improve readability. C, Cells were treated as indicated, and then whole cell lysates were prepared and utilized for immunoblotting analysis using specific antibodies against the indicated proteins. The numbers underneath each band represent fold changes relative to the control (DMSO) after normalization to GAPDH

Eugenoplatin is a potent DNA damaging agent

Cisplatin is a well-known DNA-damaging agent; we sought to test the DNA-damaging potential of eugenoplatin on normal skin fibroblast cells. To this end, exponentially growing HFSN1 cells were either sham-treated or challenged with cisplatin (50 µM) or eugenoplatin (3 µM) for 24 h, and then cells were immunostained with γH2AX, a DNA-damage binding protein, and DAPI. Figure 9A shows clear and strong nuclear γH2AX immunostaining in cells treated with cisplatin and eugenoplatin, while no staining was observed in control cells. This eugenoplatin-dependent induction of DNA-damage was confirmed by showing strong upregulation of the active/phosphorylated form of p53 as well as its downstream target p21 (Fig. 9B). This shows that, like cisplatin, eugenoplatin is also an efficient DNA-damaging agent.

Fig. 9.

Eugenoplatin induces DNA-damage. A, Human fibroblasts (HFSN1) were either sham-treated or challenged with cisplatin (50 µM) or eugenoplatin (3 µM) for 24 h, and then the cells were immunostained with anti-γH2AX antibody. Left panel: Immunofluorescence, scale bars = 25 µM, right panel: quantification of the γH2AX immunofluorescence intensity. Error bars represent mean ± SEM (n = 3). **P ≤ 0.01; ***P ≤ 0.001. B, cells were treated with eugenoplatin (3 µM) for the indicated periods of time, and then whole cell lysates were prepared and used for immunoblotting analysis. The numbers underneath each band represent fold changes relative to the control (0) after normalization to GAPDH

Discussion

Phytochemicals have recently received increased attention as anti-cancer molecules due to their safety and capacity to affect different cancer-related signaling pathways in various types of cancer. However, their clinical use has been limited so far owing to their poor bioavailability and biodistribution [20]. We present here the synthesis and characterization of a new phyto-chemo drug conjugate composed of covalently linked eugenol and cisplatin. The compound has an MW of 427. 79 determined by both LC/MS and in silico analysis. Using Biovia Discovery Studio, we have also found that eugenoplatin does not violate LRo5, has good permeability properties, good absorption, and enhanced bioavailability. Furthermore, eugenoplatin is unlikely to be mutagenic, carcinogenic, or hepatotoxic. This indicates that eugenoplatin has features of a good anti-cancer drug. In fact, eugenoplatin has shown high stability in plasma and potent and specific cytotoxicity against TNBC cells. This cytotoxicity was tested against 4 different cell lines and was stronger than that of cisplatin and carboplatin, which are widely used for the treatment of TNBC patients. Eugenoplatin-related cytotoxicity was mediated through the induction of DNA-damage-promoting apoptosis via the mitochondrial pathway. The emergence of resistance to platinum-based regimens during the treatment of TNBC patients is very frequent [21]. This prompted us to test the effect of eugenoplatin against cisplatin-resistant MDA-MB-231 cells. Interestingly, these resistant cells showed sensitivity to the killing effect of eugenoplatin. This may suggest the possible use of this novel molecule for the treatment of cisplatin-resistant TNBC tumors. The resistance of TNBC to cisplatin is mediated through multiple mechanisms [22], including promoting the formation of cancer stem cells, known to be resistant to various forms of therapies, including platinum-based regimens [6, 23, 24]. Thereby, targeting these metastasis-mediating cells was a major objective in the present study. To this end, we have shown that eugenoplatin can efficiently target CSCs and inhibit their self-renewal capacity. This was mediated mainly through inhibition of the CSC pathway WNT/β-catenin. Indeed, eugenoplatin downregulated β-catenin and its downstream targets, and limited its translocation to the nucleus. On the other hand, ectopic expression of the β-catenin coding gene in TNBC cells inhibited these processes, showing the importance of inhibiting this important pathway for eugenoplatin-targeting CSC in TNBC cells. In fact, the WNT/β-catenin signaling pathway plays a critical role in breast carcinogenesis and stemness, especially in the TNBC subtype of the disease [25, 26]. β-catenin knockdown cells exhibited higher sensitivity to cisplatin [26], and targeting this pathway and BCSCs has shown great therapeutic potential for BC [27]. Moreover, inhibition of the WNT/β-catenin signaling sensitized TNBC cells to the cytotoxic effects of carboplatin [28]. This shows the importance of eugenoplatin-dependent targeting of the WNT/β-catenin signaling pathway for the treatment of TNBC tumors.

Eugenoplatin-treatment of TNBC cells also inhibited STAT3 and NF-κB, 2 other key CSC pathways known to promote breast malignancy and chemoresistance [29–32]. This provides clear evidence that eugenoplatin possesses strong anti-CSC effects through inhibiting various breast CSC-related pathways [33]. These findings were confirmed in vivo in orthotopic humanized tumor xenografts by showing eugenoplatin-dependent inhibition of tumor growth and modulation in the expression of breast CSC-related biomarkers. Eugenoplatin exhibited greater tumor growth inhibitory effects compared to cisplatin. When reimplanted, eugenoplatin-treated tumors generated significantly smaller tumors than controls in nude mice. Together, these results clearly show that eugenoplatin can effectively target the resistant CSCs both in vitro and in a preclinical animal model. This indicates that eugenoplatin may have the potential to overcome one of the most significant challenges in breast cancer treatment: resistance mediated by cancer stem cells, and therefore could represent a promising novel therapeutic solution for TNBC patients.

LC/MS proteomics analysis has shown that eugenoplatin can modulate the expression of hundreds of proteins belonging to different physiological pathways. This was confirmed by showing that eugenoplatin is a potent inhibitor of BC cells migration and invasion, two important processes for tumor spread and metastasis. This indicates that eugenoplatin has the capacity to suppress the potential of the aggressive TNBC cells to metastasize. In fact, eugenoplatin also inhibited angiogenesis, another essential process in the metastatic cascade. Indeed, eugenoplatin downregulated the pro-angiogenic factors VEGF-A, IL-6, and IL-8 both in vitro and in orthotopic tumors, and also suppressed the formation of new blood vessels. It has been previously shown that increased micro-vascular density is needed for tumor growth and spread, and also correlates well with poor breast cancer prognosis [34]. Targeting angiogenesis is an appealing anti-cancer approach for solid tumors. Thereby, several anti-angiogenic factors were developed and tested in preclinical and clinical settings, aiming at normalizing tumor vasculature and inhibiting tumor growth [35].

LC/MS proteomics analysis has also shown that eugenoplatin can target two important cancer-promoting signaling pathways: PI3K/AKT and RAS/ERK1/2, which are the most frequently dysregulated or mutated oncogenic pathways in breast cancers, especially TNBC [36, 37]. This was confirmed by immunoblotting analysis, showing that eugenoplatin is indeed a potent inhibitor of both pathways, both in vitro and in orthotopic tumor xenografts. This further shows that eugenoplatin is a promising therapeutic solution for TNBC patients. In fact, dual inhibition of Akt and ERK signaling pathways induced cell cycle arrest and senescence in TNBC cells [38]. Similarly, we have shown here that eugenoplatin suppresses the proliferation of TNBC cells by delaying the cell cycle at G2/M phase. This was also confirmed by showing eugenoplatin-dependent inhibition of CDK2 and its regulatory subunit cyclin A, 2 master regulators of cell cycle progression and oncogenesis [39]. You et al. have shown that the dual blocking of the AKT and MEK pathways sensitized TNBC cells to the EGFR inhibitor (gefitinib) [40]. Furthermore, 2 phase II clinical trials have shown improved disease-free survival in response to AKT inhibitors, especially in patients with tumors bearing an AKT altered pathway [41].

Conclusion

Together, the present findings indicate that eugenoplatin, a novel phyto-chemo drug conjugate, has therapeutic potential as a superior alternative to the widely used platinum-based drugs. While cisplatin is a DNA cross-linking agent, eugenol is a potent anti-inflammatory molecule that can target the chemo-resistant BCSCs. Thereby, eugenoplatin is expected to bring novel solutions and add value to the TNBC therapeutic regimen.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CSCs

Cancer stem cells

BC

Breast cancer

BCSCs

Breast cancer stem cells

TNBC

Triple-negative breast cancer

DEPs

Differentially expressed proteins

Author contributions

AA, TT contributed to the project conception, design, interpretation, and initial drafting of the manuscript. NNA, AAA contributed to data collection and curation, and initial drafting of the manuscript. BA, IA contributed to data collection and curation (drug synthesis and analysis). AAl contributed to data collection and curation (flow cytometry), and initial drafting of the manuscript. FA contributed to animal studies. MYK, MC contributed to data collection and curation (In silico analysis), and initial drafting of the manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit the manuscript.

Funding

There was no funding source for this study.

Data availability

Further aggregated data relating to this study might be available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Animal experiments were approved by the KFSH&RC Institutional Animal Care and Use Committee (ACUC) under RAC# 2220020 and were conducted according to relevant national and international guidelines.

Consent for publication

Not applicable.

Competing interests

We declare no competing interests.