Metformin potentiates DSF/Cu-loaded pluronic nanoparticles to improve therapeutic efficacy in triple-negative breast cancer.

Introduction
Breast cancer is the most commonly diagnosed malignancy in women and the leading cause of cancer-related mortality among women worldwide [1]. The incidence of breast cancer has increased in recent decades, and alarmingly, current projections suggest that this upward trend will continue. It is estimated that annual global incidence and mortality rates will reach approximately 3 million and 1 million cases, respectively [2]. Breast cancers are often aggressive and capable of metastasizing to multiple tissues; notably, metastasis to vital organs is the primary cause of breast cancer–related mortality [3].
Breast tumors can be classified into subtypes based on the expression of hormone receptors and other molecular markers, including the estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). Triple-negative breast cancer (TNBC) is a subtype characterized by the absence of ER, PR, and HER2 expression. TNBC accounts for approximately 15–20% of all breast cancer cases and is associated with a poor prognosis, as it does not respond well to conventional targeted therapies [4–6].
New studies suggest that breast cancer stem cells (BCSCs) may be responsible for these complications. In general, cancer stem cells (CSCs) have been proposed as one of the determinants of tumor heterogeneity [7]. BCSCs, similarly, have the ability to initiate tumor formation and have self-renewal potential, and can cause therapeutic resistance, metastasis, and tumor recurrence. Biomarkers such as high aldehyde dehydrogenase 1 (ALDH1) activity, CD24, and CD44 are used to identify this type of tumor cells [8–10].
Furthermore, one of the most important changes that occurs in CSCs is the alteration and transformation of metabolic pathways to meet the increased needs of these cells for ATP and various metabolites [10, 11]. CSCs have the ability to survive in diverse microenvironments by obtaining energy from different sources, depending on the availability of substrates. Evidence suggests that these cells, depending on the conditions, can obtain their energy through glycolysis or oxidative phosphorylation (OXPHOS). Secondary pathways, such as fatty acid oxidation (FAO), serve as alternative strategies to fuel CSCs under conditions of excess energy demand [12].
Metformin is the first-line pharmacological treatment for type 2 diabetes and the most widely prescribed medication for this condition worldwide [13]. Beyond its role in glycemic control, metformin has demonstrated beneficial effects in a range of other diseases and conditions, including cardiovascular disorders, cancer, and aging. The diverse biological activities of metformin suggest that it may influence multiple cellular and molecular processes [14].
On the other hand, metabolic traits associated with type 2 diabetes, such as hyperglycemia, hyperinsulinemia, inflammation, oxidative stress, and obesity, are known risk factors for breast cancer, and 15% of patients with breast cancer also suffer from type 2 diabetes [15]. The anticancer effects of metformin can be divided into two distinct but related categories: an indirect or systemic effect by reducing blood glucose and insulin levels, and a direct effect on cancer cells by targeting various pathways, including tumor metabolism, inflammation, and angiogenesis [16].
Numerous studies have demonstrated that metformin exerts notable anticancer effects in breast cancer, particularly in triple-negative breast cancer (TNBC). In one study, metformin was shown to inhibit cell proliferation, induce partial S-phase cell cycle arrest, and suppress colony formation in TNBC cell lines. In addition, metformin induced apoptosis through activation of both intrinsic and extrinsic signaling pathways. At the molecular level, metformin treatment in TNBC cells was associated with increased phosphorylation of AMPK (p-AMPK) and reduced levels of phosphorylated EGFR (p-EGFR), total EGFR, phosphorylated MAPK (p-MAPK), phosphorylated Src (p-Src), cyclin D1, and cyclin E. Furthermore, metformin induced poly(ADP-ribose) polymerase (PARP) cleavage in a dose- and time-dependent manner [17].
Other studies have demonstrated that metformin can selectively target breast cancer stem cells. This effect is mediated through mechanisms such as the downregulation of the transcription factor KLF5. In addition, metformin suppresses the expression of KLF5 downstream target genes, including Nanog and FGF-BP1, leading to a reduction in the proportion of TNBC stem cells [18]. Metformin has also been shown to inhibit STAT3 activation by reducing phosphorylation at Tyr705 and Ser727, thereby attenuating downstream STAT3 signaling in TNBC cell lines [19].
TNBCs are highly dependent on glucose metabolism and lipids, which are metabolized for energy and cellular building blocks, to maintain their high proliferation rate. Metformin has been shown to inhibit lipid metabolism, specifically by targeting fatty acid synthase (FASN), cholesterol biosynthesis, and GM1 lipid rafts in TNBCs [20]. Metformin has also been shown to downregulate or inhibit the activity of several key enzymes required for glycolysis and glucose metabolism in TNBC. These enzymes include glucose-6-phosphate dehydrogenase (G6PD), triose phosphate isomerase (TPI), phosphoglycerate kinase 1 (PGK), phosphoglucomutase 1 (PGM), enolase 1 (ENO), muscle pyruvate kinase 2 (PKM2), and lactate dehydrogenase A (LDHA) and pyruvate dehydrogenase kinase (PDK). LDHA catalyzes the conversion of pyruvate to lactate, and its inhibition likely reduces proliferation rates by switching cellular mitochondria to oxidative phosphorylation. PDK also facilitates the conversion of pyruvate to acetyl Co-A, and this process is also inhibited by metformin. In addition to these enzymes, metformin has also been reported to downregulate some glucose transporter proteins, including GLUT1, GLUT10, GLUT12, GLUT14, and glucose-6-phosphate transporter, in TNBC cells [21, 22].
Disulfiram (tetraethylthioram disulfide) or DSF is an electrophilic quaternary ammonium compound. This drug is commonly used to treat alcoholism; however, it is shown to have a significant ability to inhibit a diverse set of metabolic enzymes. Disulfiram and its metabolites act as potent chelators and can thereby disrupt the function of metal-containing enzymes, including aldehyde dehydrogenase (ALDH), carboxylesterase, cholinesterase, superoxide dismutase, and dopamine β-hydroxylase (DBH), by chelating their essential cofactors [23].
It is shown that DSF in combination with copper ions induces intracellular ROS accumulation, inhibits or suppresses NF-κB signaling, proteasome activity, and ALDH activity, and ultimately induces apoptosis. Also, DSF/Cu induces endoplasmic reticulum (ER) stress and autophagy, and induces autophagy-dependent apoptosis in cancer cells. In addition, the effects of DSF/Cu on cell cycle progression by inhibiting DNA synthesis, reducing cyclin-dependent kinases 1 (CDK1), and blocking entry into mitosis have also been revealed. This inhibitory ability of DSF/Cu combination is also applicable to cancer stem cells [24–27].
Pluronics, also known as poloxamers, are synthetic, nonionic triblock copolymers composed of polypropylene oxide (PPO) and polyethylene oxide (PEO) blocks and exhibit amphiphilic properties. In their triblock structure, a hydrophobic PPO segment is flanked by two hydrophilic PEO segments, forming the general configuration (PEO)n–(PPO)m–(PEO)n These copolymers are soluble in both polar and nonpolar solvents. Moreover, Pluronics demonstrate high biocompatibility and favorable physicochemical properties, and no significant toxicity has been reported [28].
In addition to these properties and their role as drug carriers, Pluronic copolymers themselves can enhance tumor cell responses to anticancer drugs and increase their therapeutic efficacy. They achieve this by altering the pharmacokinetic properties of chemotherapeutic agents and improving their bioavailability. These copolymers can enter cells by integrating with the cell membrane and sensitize cancer cells—particularly multidrug-resistant tumors—to chemotherapeutic agents. This sensitization occurs through inhibition of drug efflux pumps, such as multidrug resistance proteins and P-glycoprotein, as well as through modulation of various cellular functions, including mitochondrial respiration, apoptotic signaling pathways, gene expression, and ATP production [29, 30].
Based on these considerations, the present study was designed to investigate whether metabolic priming of TNBC cells by sub-cytotoxic metformin could potentiate the anticancer efficacy of DSF/Cu delivered via pluronic-based nanoparticles. We hypothesized that metformin-induced energetic stress sensitizes TNBC cells to DSF/Cu-mediated oxidative and proteotoxic stress, resulting in enhanced apoptosis and suppression of stemness-associated molecular signatures. To test this hypothesis, we evaluated nanoparticle characteristics, cytotoxicity, apoptosis, and expression of key stemness-related genes and STAT3 in MDA-MB-231 cells.

Materials and methods

Materials
Pluronic® F127,1,1′-carbonyldiimidazole (CDI), dialysis membrane with MWCO of 3.5 kDa and RPMI-1640 culture medium were purchased from Sigma, USA. Dichloromethane (DCM), dimethyl sulfoxide (DMSO), triethyleneamine (TEA) and ethanol were purchased from Merck, Germany. Trypsin-EDTA and fetal bovine serum (FBS) were purchased from Biosera, France. MTT [3(4,5-dimethylthiazolyl-2)2,5-diphenyltetrazolium bromide] was purchased from Sigma, USA. Agarose was purchased from UltraPure Agarose, USA. Trizol was purchased from CinnaGen, Iran. Also, cDNA synthesis kit was purchased from Zist Virayesh, Iran and Real-time PCR kit was purchased from Parstous, Iran. Annexin V/PI assay kit was purchased from Mab Tag, Germany. Also, metformin and disulfiram were purchased from Sigma, USA. MDA-MB-231 cell line was also purchased from the Pasteur Institute, Iran.

Nanoparticle synthesis
This section was performed based on our previous study [31]. Briefly, to synthesize hyaluronic acid (HA)-conjugated Pluronic F127 (PF127) nanoparticles (HA-PF127), 200 mg of Na-HA salt was dissolved in 20 mL of distilled water with a pH of approximately 3.5. Then, this solution was dialyzed into a solution containing HCl with a pH of approximately 3.5, using a 3.5 kDa dialysis bag, for 24 h. Then, the carboxylic acid group of HA was activated using CDI (81.8 mg, 0.5 mM). ¹H-NMR spectroscopy was performed at 25 °C using a Bruker 400 MHz UltRAShield spectrometer (Germany).
Next, 270 mg of HA-PF127 was dissolved in 3 mL of DCM and 3 mL of ethanol and mixed thoroughly with a 2 mg of DSF and 1 mg CuCl2 at room temperature for 30 min at 200 rpm in a rotary device. To create a thin and uniform layer of HA-PF127 and drugs, the solvents DCM and ethanol were slowly evaporated under vacuum at 35 °C for 30 min. A thin layer of HA-PF127 and DSF/Cu was formed and then completely dried using a desiccator. Then, this thin layer was hydrated in PBS solution with a pH of about 7.4 at 35 °C.

Evaluation of physicochemical and morphological properties of nanoparticles
To examine the size and surface charge of nanoparticles, the DLS (dynamic light scattering) technique was used using a Malvern device (UK). In this end, the samples were placed on aluminum disks and their surfaces were coated with a thin layer of gold in the presence of argon plasma. In addition, scanning electron microscopy (SEM, MIRA3 FEG-SEM, Tescan, Czech Republic) and transmission electron microscopy (TEM, Zeiss LEO 906, Germany) were also used to study the morphology of DSF/Cu-loaded nanoparticles.
To evaluate the structure of HA-PF127 nanoparticles and their interaction with drugs, FTIR (Fourier transform infrared spectroscopy, Bruker Optik GmbH, Germany) was used. This technique was used to obtain the absorption spectra of HA-PF127 nanoparticles and their combination with DSF, DSF/Cu, and also to evaluate the binding of hyaluronic acid to pluronic in the absorption range of 400–4000 cm− 1.
In order to determine the loading rate and entrapment efficiency of DSF and DSF/Cu in HA-PF127 nanoparticles, different concentrations of each of compounds DSF and DSF/Cu were prepared and analyzed using the spectrophotometery technique. The optical absorption of DSF and DSF/Cu samples was measured at wavelengths of 314.5 and 314 nm (λ max), respectively, using a spectrophotometer. Then, the encapsulation efficiency and drug loading capacity were calculated using the relevant formulas.

Evaluation of drug release rate from synthesized nanoparticles
To this, 50 mg of DSF/Cu and metformin nanoparticles were weighed and dissolved in 5 mL of PBS at pH 7.4 and 8.5 to simulate physiological and tumor conditions. The solutions were placed in 3.5 kDa dialysis bags, sealed, and submerged in 50 mL of matching PBS solution. Samples were incubated at 37 °C with shaking at 100 rpm. Every hour for the first day, 2 mL samples were withdrawn and replaced with fresh solution, continuing at 24-hour intervals for four days. Absorbance readings were taken at 314 nm for DSF/Cu and 233 nm for metformin using a spectrophotometer. Concentrations were determined using a calibration curve, and the release profile was plotted.

Evaluation of nanoparticle uptake by cells
The cellular accumulation efficiency of nanoparticles containing a fluorescent agent in cancer cells was evaluated and compared with that of cells treated with the fluorescent agent alone (doxorubicin (Dox)) [32–35]. For this, MDA-MB-231 cells were seeded in 24-well plates at a density of 30,000 cells per well. After 24 h of incubation, the cells were treated with HA-PF127@DOX, PF127@DOX nanoparticles, and free Dox at a concentration of 2 µg/mL, and incubated for 2 and 4 h. The cells were then trypsinized and collected in 15-mL Falcon tubes. Following centrifugation at 1300 rpm for 5 min, the supernatant was removed. To completely eliminate extra doxorubicin, the cell pellet was washed twice with phosphate-buffered saline (PBS) (1300 rpm for 5 min). Finally, the intracellular fluorescence intensity of doxorubicin in each group was measured at a wavelength of 570 nm using a BD FACSCalibur flow cytometer.

Cell viability assessments
MTT assay was performed based on our previous works [36, 37]. In brief, The MTT assay was used to evaluate the toxicity of drugs and nanoparticles on MDA-MB-231 cells and to determine the IC50 of the respective drugs. For this purpose, a 96-well plate was used, with 5,000 cells per well. After treatment with different formulations, the cells were incubated for 48 h. The incubation conditions were 37 °C and 5% CO2. The absorbance of the samples in each well at a wavelength of 570 nm was measured by an ELISA reader, and the percentage of cell viability was calculated compared to the control group.

Assessment of apoptosis/necrosis
Flow cytometry was used to evaluate the apoptosis rate of MDA-MB-231 cells treated with different formulations of nanoparticles and drugs [38–40]. After reaching the desired confluency, cells were trypsinized, collected, and centrifuged at 160 × g for 5 min. A total of 200,000 cells were seeded into each well of 6-well plates and incubated for 24 h, followed by drug treatments for 48 h. The DSF/Cu concentration (free or nanoparticle-loaded, with or without metformin) was equal to the IC₅₀ value obtained from the MTT assay, and the same metformin concentration was used accordingly. Cells were then harvested, washed, and stained with FITC-Annexin V and propidium iodide. Apoptotic and necrotic cells were quantified by flow cytometry at 488 nm.

Gene expression assessments
In line with our previous works, SYBR Green-based Real Time PCR was used to evaluate the mRNA change in the expression level of target genes [31, 40, 41]. MDA-MB-231 cancer cells were treated with different drug formulations for 48 h. Then, total RNA was isolated from the samples using RNAX Plus reagent and reverse-transcribed into complementary DNA (cDNA) with an RT reagent kit. RNA concentration and purity were assessed by spectrophotometry using a NanoDrop 2000 (Thermo Scientific, USA). Quantitative real-time PCR was performed using SYBR Green dye. The amplification protocol consisted of an initial denaturation at 95 °C for 30 s, followed by 45 cycles of denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 30 s. using Mic qPCR cycler (Bio Molecular Systems, Australia). The 2−ΔΔCT method was used to quantify the results. The sequence of primers is provided in S1.

Statistical analysis
All statistical analyses were performed with GraphPad Prism version 8. For comparisons between two groups, an unpaired Student’s t-test was applied, while one-way ANOVA followed by Tukey post hoc tests was used for comparisons among three or more groups. Data are expressed as mean ± standard deviation (SD), and a p-value < 0.05 was considered statistically significant.

Results

Conjugation of hyaluronic acid to pluronic F127
The carboxyl groups of the repeating disaccharide units in HA were identified as suitable reactive sites for conjugation with Pluronic F127 (PF127). During the coupling process, CDI reacted with the carboxyl groups of HA to generate an activated imidazolyl intermediate. This activated species subsequently underwent nucleophilic substitution with the terminal hydroxyl groups of PF127, leading to the successful formation of the HA–PF127 conjugate. The reaction proceeded efficiently, indicating that CDI-mediated activation provided a straightforward and effective route for covalent linkage between HA and PF127. Based on our previous study, the yield of conjugation was achived 51.7–61.8%, depending on the assumed PPO block length (56–67 units) [31].
The ¹H-NMR spectrum of the HA–PF127 conjugate revealed a distinct resonance at approximately 2 ppm, corresponding to the residual N-acetyl groups of hyaluronic acid. Multiple signals observed in the 3–4 ppm region were attributed to the protons of β-N-acetyl-D-glucosamine units within the HA backbone. In addition, a characteristic peak at around 1 ppm indicated the presence of unreacted polypropylene oxide (PPO) groups from the PF127 structure. Collectively, these spectral features confirmed the successful conjugation of HA to PF127, consistent with the expected chemical structure (Fig. 1A-C).

Uniform nanoscale dimensions of nanoparticles confirmed by DLS
DLS was employed to determine the hydrodynamic diameters of the synthesized nanoparticles, including HA–PF127, HA–PF127@DSF, HA–PF127@DSF/Cu. The measured average sizes were 76.8 nm, 70.9 nm and 155.5 nm, respectively. Also, it was observed that adding Cu increases the size of the nanoparticles (Fig. 2A-C). In addition, the surface charge (zeta potential) of HA–PF127@DSF/Cu nanoparticles was determined to be − 3.08 mV, indicating a slightly negative surface potential. These findings confirm the successful formation of nanoscale structures with distinct size distributions depending on the incorporated drug or complex.

Spherical and uniform morphology of HA–PF127 nanoparticles confirmed by electron microscopy
The morphology and size of HA–PF127@DSF and HA–PF127@DSF/Cu nanoparticles were examined using both SEM (Fig. 3A, B) and TEM (Fig. 3C, D). The obtained images revealed that the nanoparticles exhibited a spherical and uniform morphology. The particle sizes were estimated to be within the range of 50–100 nm for HA–PF127@DSF and 90–160 nm for HA–PF127@DSF/Cu, which were consistent with the hydrodynamic diameters measured by DLS. These findings confirm the successful synthesis of well-defined nanoscale structures with morphology and size distributions in agreement across complementary characterization techniques.

Confirmation of chemical bonding and structural features of HA–PF127 nanoparticles by FTIR
The chemical bonding of HA to the PF127 copolymer, as well as the structural characteristics of pure PF127, HA–PF127@DSF and HA–PF127@DSF/Cu nanoparticles, were investigated by FTIR spectroscopy in the range of 400–4000 cm⁻¹(Fig. 4).
The appearance of a new absorption band at 1625.03 cm⁻¹ in the HA–PF127 spectrum, compared with pure PF127(Fig. 4A), confirmed the successful conjugation of HA to Pluronic (Fig. 4B). This band corresponds to the C = O stretching vibration of the carboxyl groups present in HA. For HA–PF127@DSF nanoparticles, distinct peaks at 758.21, 1147.84, 1629.12, and 2972.50 cm⁻¹, corresponding to pure DSF, were detected, along with a synergistic band at 2889 cm⁻¹. Similarly, the characteristic DSF peak at 1635.56 cm⁻¹ was also present in the HA–PF127@DSF/Cu spectrum. Collectively, these findings confirm the successful incorporation of HA, DSF and DSF/Cu into the PF127-based nanostructures, validating the expected chemical interactions.

High encapsulation efficiency and drug-loading capacity of HA–PF127 nanoparticles
The EE% and DL% of the synthesized nanoparticles were quantitatively evaluated. The results demonstrated high encapsulation efficiencies of 89.5% for DSF and 73.0% for DSF/Cu. Correspondingly, the drug-loading capacities were 7.0% and 6.5%, respectively. These findings confirm the successful incorporation of DSF and DSF/Cu into the HA–PF127 nanocarrier system.

pH-responsive drug release behavior of HA–PF127 nanoparticles
The in vitro release profile of DSF/Cu from HA–PF127@DSF/Cu nanoparticles was evaluated in phosphate buffer solutions at pH 5.8 and pH 7.4 over 96 h (Fig. 5). DSF/Cu release was significantly faster and more efficient at pH 5.8, which mimics the acidic tumor microenvironment and endosomal compartments, compared with pH 7.4. A relatively higher fraction of the drug was released during the initial hours, and after 96 h, the cumulative release reached approximately 50% under both acidic and neutral conditions. The observed release plateau (~ 50%) likely reflects strong hydrophobic interactions between DSF/Cu and the PPO core of PF127, resulting in a sustained-release profile rather than incomplete formulation.

Efficient cellular uptake of DOX-loaded nanoparticles compared with free DOX
The cellular uptake of DOX-loaded nanoparticles (PF127@DOX and HA–PF127@DOX) by MDA-MB-231 breast cancer cells was evaluated at 2 h and 4 h and compared with the internalization of free DOX under identical conditions (Fig. 6). Quantitative analysis revealed that intracellular accumulation of free DOX reached 53.51% and 77.39% after 2 h and 4 h, respectively. In contrast, uptake of PF127@DOX and HA–PF127@DOX nanoparticles was markedly higher, increasing from 93.03% to 94.23% at 2 h to 98.04% and 97.45% at 4 h, respectively. These results demonstrate that both nanoparticle formulations facilitated faster and more efficient intracellular delivery of DOX compared with the free drug. However, no statistically significant difference in uptake efficiency was observed between PF127@DOX and HA–PF127@DOX at either time point (p-value > 0.05). HA modification did not significantly enhance cellular uptake under in vitro conditions.

Enhanced cytotoxicity of DSF/Cu-loaded nanoparticles and synergistic-like potentiating effect with metformin
The cytotoxicity of free drugs, blank nanoparticles, and drug-loaded nanoparticles against MDA-MB-231 breast cancer cells was assessed using the MTT assay (Fig. 7). Treatment groups and IC₅₀ values of each formulation are summarized in Table 1.
As shown in Fig. 8, after 48 h of incubation, blank HA–PF127 nanoparticles exhibited no significant cytotoxicity; in fact, at certain concentrations they slightly promoted cell proliferation compared with the control group. Even at the highest tested concentration (21.6 mg/mL), cell viability remained above 75%, confirming their biocompatibility. Free metformin also displayed relatively low cytotoxicity, with ~ 78% cell viability at the highest tested concentration (12.388 mM), suggesting that its IC₅₀ value is considerably higher than the tested range.
In contrast, free DSF/Cu and HA–PF127@DSF/Cu nanoparticles exhibited potent cytotoxic effects, with IC₅₀ values of 65.3 µM and 48.95 µM, respectively. These results indicate that encapsulation of DSF/Cu within HA–PF127 nanoparticles enhanced its cytotoxicity, thereby lowering the IC₅₀. Furthermore, co-treatment with a subtoxic concentration of free metformin (1.548 mM), which alone had negligible effects on cell viability, significantly potentiated the cytotoxicity of HA–PF127@DSF/Cu, reducing its IC₅₀ to 28.95 µM.

Synergistic induction of apoptosis by HA–PF127@DSF/Cu combined with metformin
To evaluate the apoptotic effects of the formulations and to determine potential synergistic interactions, flow cytometry analysis was performed using Annexin V/PI staining on MDA-MB-231 breast cancer cells. The concentration of DSF/Cu was fixed at 29 µM (approximately equal to the IC₅₀ obtained for HA-PF127@DSF/Cu + free metformin in the MTT assay), while free metformin was applied at 1.548 mM in all relevant groups (Fig. 8). Consistent with the MTT findings, the lowest cell viability (34.72%) was observed in cells treated with the combination of HA–PF127@DSF/Cu and free metformin. The highest proportion of necrotic cells (22.37%) was detected in the free DSF/Cu group. In contrast, the strongest apoptotic response was induced by HA–PF127@DSF/Cu + free metformin (62.67%), followed by HA–PF127@DSF/Cu alone (52.77%).

Downregulation of stemness-related genes and STAT3 by HA–PF127@DSF/Cu and combination therapy
The expression of stemness-associated genes (ALDH, CD44, and CD24) was evaluated in MDA-MB-231 cells, as a reliable cell line enriched for CSCs [42, 43] following 48 h of treatment with free metformin, HA–PF127@DSF/Cu, or the combination of HA–PF127@DSF/Cu with free metformin, using real-time PCR analysis. Also, expression level of transcription factor STAT3 was evaluated for HA–PF127@DSF/Cu and HA–PF127@DSF/Cu + free metformin formulations.
The concentrations applied were identical to those used in the apoptosis assay (1.548 mM for free metformin and 29 µM for DSF/Cu). The results demonstrated that ALDH expression was downregulated by 1.36-, 2.46-, and 3.65-fold in cells treated with free metformin, HA–PF127@DSF/Cu, and the combination therapy, respectively, compared with the control. Similarly, CD44 expression decreased by 3.14-, 3.31-, and 6.25-fold in the same groups. In contrast, CD24 expression was markedly upregulated, showing 6.47-, 16.11-, and 2.33-fold increases, respectively. Furthermore, STAT3 expression was significantly reduced by 5.07-fold in the HA–PF127@DSF/Cu group and by 6.76-fold in the combination group relative to the control (Fig. 9).

Discussion and conclusion
Breast cancer is currently the most prevalent malignancy and the leading cause of cancer-related mortality among women worldwide. Triple-negative breast cancer (TNBC) is an aggressive and heterogeneous subtype characterized by the absence of estrogen and progesterone receptors and human epidermal growth factor receptor 2 (HER2). TNBC accounts for approximately 15–20% of all breast cancer cases and is associated with the poorest prognosis in terms of tumor recurrence and patient survival compared with other breast cancer subtypes. Owing to the lack of validated molecular targets, no targeted therapies are currently available for TNBC, and treatment relies primarily on non-specific systemic chemotherapy in combination with radiotherapy and surgical resection. However, the clinical efficacy of chemotherapy is limited by severe systemic toxicity and the development of intrinsic or acquired drug resistance. These challenges highlight the need for combination therapeutic strategies that minimize drug dosage while reducing the risk of chemotherapy resistance [44, 45].
On the other hand, one of the interesting strategies in the field of drug therapy is drug repurposing. Drug repurposing actually means discovering or identifying a new use for existing drugs. This means that we can use existing drugs approved by the Food and Drug Administration (FDA) to treat a pathological condition other than the drug’s original use [46, 47]. Compared to developing new drugs, using this strategy requires less expenditure and saves time and resources.
In this context, DSF, an approved anti-alcoholism drug and a clinically available treatment for alcohol dependence, has been shown to have tumor-targeting potential. This anti-tumor effect (especially in combination with Cu) is exerted through targeting the STAT3 signaling pathway and inhibiting ALDH1 activity, and can suppress stem-like properties in TNBC tumors [5, 25].
On the other hand, we and some studies have shown that metformin (a drug approved for the treatment of type 2 diabetes) has the ability to inhibit cancer cell proliferation, induce apoptosis, and reduce the number of stem-like cells in breast tumors, especially TNBC tumors [40]. These effects of metformin on tumors occur through the influence on various pathways and mechanisms, including inhibition of respiratory complex I of the electron transport chain in mitochondria (reducing ATP production and energy requirements of the cell), inhibition of protein synthesis and cell growth through activation of LKB1 and AMPK (and consequently inhibition of mTOR), inhibition of several STAT3-related signaling pathways involved in breast cancer, including the IL-6/JAK2/STAT3 pathway, etc. In addition, it has been shown that in TNBC cells, metformin can reduce the activation (phosphorylation) of STAT3 [48, 49]. Our results indicated significant downregulation of Notch-1 and HIF-1a signaling by metformin loaded-chitosan NPs alone or in combination with Digoxin in MCF-7 breast cancer spheroids as well as blocking angiogenesis confirmed by CAM-assay comparable to anti-angiogenesis effects of Avastin [40].
The present study aimed to investigate the synergistic-like effects of two FDA-approved drugs, metformin and DSF (as the DSF/Cu complex), formulated in Pluronic nanoparticles, on suppressing TNBC cells, particularly through modulation of cellular energy pathways. DSF and DSF/Cu were efficiently loaded into Pluronic PF127 nanoparticles, identifying them as a suitable carrier. Co-treatment with low, sub-cytotoxic concentrations of metformin and DSF/Cu-loaded nanoparticles potentiated anticancer activity, as evidenced by increased cell death, decreased IC₅₀, and approximately two-fold downregulation of stemness-related genes. These findings indicate that HA–PF127@DSF/Cu nanoparticles effectively suppress stemness-associated gene expression, which is further enhanced by metformin, while the functional role of HA is expected primarily in vivo rather than in 2D culture.
Pluronic nanoparticles also improved cellular uptake of DSF/Cu, enhancing sensitivity of MDA-MB-231 cells and reducing the required drug dose, thereby potentially decreasing side effects. Additionally, nanoparticle delivery shifted cell death from necrosis, observed with free DSF/Cu, toward apoptosis, which is more desirable therapeutically. Compared with previously reported DSF-based nanocarriers, HA–PF127 nanoparticles demonstrate several advantages: high encapsulation efficiency, nanoscale size suitable for cellular internalization, and a simple, biocompatible, FDA-familiar Pluronic platform. The sustained-release behavior, together with the apoptosis-favoring effect and potentiation by low-dose metformin, highlights the translational potential of this nanomicellar strategy. Here, “low-dose metformin” refers to a sub-cytotoxic in vitro concentration used as a metabolic sensitizer rather than a clinically equivalent plasma dose.

Limitations of the study
This study has several limitations that should be acknowledged. First, the primary objective of this work was to evaluate a drug repurposing strategy by combining a sub-cytotoxic dose of metformin with DSF/Cu-loaded pluronic nanoparticles, rather than to provide an exhaustive mechanistic dissection of their molecular interactions. Notably, metformin alone did not exhibit a measurable IC₅₀ in TNBC cells under the tested conditions; however, when administered at a low, non-cytotoxic concentration, it markedly potentiated the anticancer efficacy of DSF/Cu nanoparticles. This finding highlights the therapeutic relevance of the combination despite the limited standalone activity of metformin. Second, although the study demonstrates coordinated modulation of apoptosis, oxidative stress, and stemness-associated markers, detailed mechanistic validation—such as pathway-specific inhibition studies or phospho-protein analyses (e.g., phospho-STAT3)—was beyond the scope of the current work. While STAT3 regulation is primarily controlled at the phosphorylation level, the present study focused on transcriptional changes, which are supported by extensive literature linking sustained STAT3 pathway inhibition to downstream gene suppression. Future studies employing Western blotting, pathway inhibitors, and rescue experiments will be necessary to fully delineate causal signaling hierarchies. Third, the suppression of stemness was inferred from molecular markers (ALDH, CD44, CD24, and STAT3) rather than functional cancer stem cell assays such as mammosphere formation. Although these markers are well-established regulators of TNBC stemness and MDA-MB-231 cells are known to possess a high stem-like population, functional 3D assays will be important in future investigations to directly confirm effects on self-renewal capacity.
In addition, the study did not include formal synergy quantification using combination index models (e.g., Chou–Talalay). Accordingly, the interaction between metformin and DSF/Cu nanoparticles is described as potentiating or synergistic-like, rather than strictly synergistic. Quantitative synergy analyses will be addressed in future studies. Moreover, the targeting potential of hyaluronic acid (HA) requires validation in in vivo models, as no significant difference in cellular uptake was observed between HA-PF127 and PF127 nanoparticles under in vitro conditions. Finally, the sustained drug release profile observed in vitro, with approximately 50% release over 96 h, may limit rapid drug availability but likely contributes to improved safety and controlled intracellular delivery. Further in vivo pharmacokinetic and biodistribution studies are required to fully assess release behavior under physiological and tumor-specific conditions.

Conclusion
Despite these limitations, the study provides strong proof-of-concept evidence that repurposed low-dose metformin can substantially enhance the therapeutic efficacy of DSF/Cu nanoformulations, supporting further mechanistic and translational investigations.

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