Enhancing EGFR Inhibition: Monensin and Erlotinib Synergize to Eliminate Cancer Stem Cells in Canine Mammary Tumours

ABSTRACT

Canine mammary tumours (CMTs) and human breast cancer (HBC) share highly similar pathological characteristics. Cancer stem cells (CSCs) are critical to breast cancer invasion, metastasis, drug resistance, and recurrence. The experiment utilised CMT cell line CMT‐U27 and HBC cell line MDA‐MB‐231. Cancer stem cell spheres were isolated from the two cell lines by serum‐free culture respectively, named as CMT‐U27 microspheres (CMT‐U27S) and MDA‐MB‐231 microspheres (MDA‐MB‐231S), and the proportion of CSCs with the CD44⁺/CD24⁻ phenotype was identified by flow cytometry. Monensin (MON), one of Polyether ionophore antibiotics, has been demonstrated to effectively suppress various types of CSCs, while erlotinib (ERL), as a tyrosine kinase inhibitor targeting epidermal growth factor receptor (EGFR), effectively inhibits cancer cell growth. This study investigates the synergistic inhibitory effects of MON‐ERL combination on CSCs derived from CMTs. Cell viability was analysed by CCK‐8 assay, while cell invasion and mammosphere formation assays were conducted to evaluate changes. An orthotopic tumour model in nude mice using CMT‐U27S was established for in vivo validation. The results suggested that MON and ERL combination synergistically inhibited CSCs viability and significantly suppress their invasion and mammosphere formation abilities. The expression of key proteins of the EGFR pathway—p‐EGFR, PI3K, and p‐AKT was significantly reduced. In animal experiments, tumour volumes in the combination treatment group were markedly reduced compared to those in all other groups (p < 0.05), and no lung metastases were observed only in the combination group. Immunofluorescence and immunohistochemistry results indicated a marked reduction of CD44⁺/CD24⁻ cells in the combination group, along with suppressed expression of stem cell markers and proliferation/apoptosis proteins. In conclusion, MON combined with ERL effectively inhibits the proliferation, self‐renewal, and metastatic capabilities of CSCs from CMTs by synergistically downregulating the EGFR/PI3K pathway, demonstrating promising antitumor potential and offering a new strategy for CMTs treatment.

1. Introduction

Canine mammary tumours (CMTs) are common and severe malignancies, they are the most frequently diagnosed tumours in female dogs and therefore represent a significant challenge in veterinary oncology [1]. Approximately half of diagnosed CMTs are malignant, displaying aggressive behaviour and variable therapy responses [2]. Surgical resection is the first‐line treatment for CMTs, while adjuvant options include chemotherapy agents (e.g., doxorubicin, docetaxel, cyclophosphamide) and molecular‐targeted therapies (e.g., toceranib, firocoxib, piroxicam) [3, 4, 5, 6, 7]. CMTs are a leading cause of cancer‐related death in female dogs, especially when detected at an advanced stage, where survival is often limited to months despite surgery and other treatments [8]. Therefore, a novel treatment strategy is needed, as no treatment protocol has been established for CMTs.

CMTs share considerable similarities with human breast cancer (HBC) in terms of clinical features, genetic alterations, and tumour heterogeneity [9]. Studies have shown that gene expression profiles in CMTs closely resemble those of HBC; dysregulation of epidermal growth factor receptor (EGFR) activates key signalling pathways involved in tumour cell proliferation and metastasis [10, 11]. As a key driver of CMTs' cell behaviour, the EGFR pathway promotes both cell proliferation and survival, demonstrating that targeting this pathway holds promise as a novel therapeutic strategy against these aggressive tumours [12]. In HBC, EGFR‐targeted therapies have shown promise [13], and similar approaches could be adapted for CMTs. Although erlotinib (ERL), an EGFR‐tyrosine kinase inhibitor, was once considered a promising therapeutic target via the EGFR pathway for HBC, it actually has limited efficacy [14]. To enhance efficacy, direct EGFR inhibition should be combined with therapies that disrupt the signalling pathways crucial for cancer stem cells (CSCs) maintenance [15].

Another similarity between CMTs and HBC is the notable presence of CSCs, which are strongly implicated in recurrence and metastasis due to their self‐renewal capacity and resistance to conventional therapies [16, 17]. The commonly evaluated CD44⁺/CD24⁻ CSCs phenotype has been associated with poor prognosis in both CMTs and HBC [18, 19]. Therefore, targeting CSCs may be an effective strategy for treating CMTs. CSCs are a subset of cells in solid mammary tumours that possess self‐renewal capacity and express stemness markers, including CD44⁺/CD24⁻, ALDH, OCT4, and EpCAM (CD326) [17]. Previous studies reported that CSCs from CF41.Mg and REM134 canine mammary carcinoma cells demonstrated both sphere formation ability in vitro and resistance to various chemotherapeutic agents, including doxorubicin [20, 21]. However, CSC‐targeted therapy in CMTs will be a challenge because of the lack of specific CSCs markers and the unclear molecular links between the EGFR pathway and CSCs phenotype maintenance. The polyether ionophore antibiotic monensin (MON) and salinomycin have demonstrated selective elimination of CSCs populations [22, 23]. In addition, the combination of MON and ERL exhibits good efficacy against triple‐negative breast cancer (TNBC) and reduces their CSCs populations, which has been effectively validated in our previous research [24].

On this basis, we hypothesise that the therapeutic strategy combining MON and ERL is also applicable to CSCs derived from CMTs. Although previous studies have demonstrated aberrant activation of the EGFR pathway in CMTs [25, 26], the direct link between this pathway and the CSCs phenotype, as well as the specific regulatory mechanisms underlying CMTs development and progression, remain unclear. Accordingly, this study aims to evaluate the efficacy of this combination therapy and to establish optimised treatment protocols for CMTs.

2. Materials and Methods

2.1. Cell Culture

The CMT‐U27 (canine mammary tumour) cell line was obtained from the American Type Culture Collection (ATCC CRL‐3456, Manassas, VA, USA). Cells were cultured in RPMI 1640 (with Glutamax, 25 mM Hepes) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin at 37°C in 5% CO₂. The MDA‐MB‐231 (human breast carcinoma) cell line is well characterised for its high invasiveness and migratory capacity, which share considerable similarities with CMTs in terms of clinical features, genetic alterations, tumour heterogeneity and the PI3K pathway activation [27]. Therefore, we chose the MDA‐MB‐231 cell line from Cell Bank, Chinese Academy of Sciences (Serial TCHu227) for the cell experiment section. The cell line was cultured in DMEM (HyClone, Logan, USA), added with 10% FBS and 1% penicillin–streptomycin at 37°C in 5% CO2. All experiments were conducted using cells after thawing. No further authentication was performed on the two cell lines beyond the certification provided by the supplier. All experiments were conducted with cells at low passages to ensure genetic stability.

CSCs were enriched through the sphere formation assay. Briefly, CMT‐U27 and MDA‐MB‐231 cells harvested during logarithmic growth were seeded at a density (5 × 10⁵ cells per well). Additionally, the cells were brought into suspension and cultured in DMEM/F‐12 cell culture medium, added with B27, 10 ng/mL FGF, and 20 ng/mL EGF. Cell growth was monitored every 2 days. After 7 days of culture, tumour spheres with diameters ≥ 60 μm were collected for subsequent experiments, named as CMT‐U27 microspheres (CMT‐U27S) and MDA‐MB‐231 microspheres (MDA‐MB‐231S).

2.2. Chemicals Information

For cell experiments, MON solid (Solarbio, Beijing, China; Cat. M8670) was dissolved in ethanol. For animal experiments, MON was prepared as a suspension, and solid ERL (MCE, NJ, USA, HY‐50896) was dissolved both in 0.9% NaCl.

2.3. Cell Viability and Combination Index Analysis

Cell viability was tested via the CCK‐8 assay. Absorbance was measured at 450 nm with a FLUOstar Omega reader. Cell viability calculation formula:

(1)

All cells and their stem cells were treated with gradient concentrations (0–1.00 μM) of MON to assess its cytotoxicity. Then, all cell lines were exposed to gradient concentrations (0–10.00 μM) of ERL to evaluate its cytotoxicity.

CMT‐U27S and MDA‐MB‐231S were treated as follows: MON alone (0.25 μM or 0.50 μM), ERL alone (0–8 μM), combinations of MON (0.25/0.50 μM) and ERL (0–8 μM). Combination Index (CI) Analysis was employed to analyse drug interactions. A CI value < 1.0 defines synergy, CI = 1.0 defines additivity, and CI > 1.0 defines antagonism. The CI₅₀ was subsequently calculated by the following equation:

$${\mathrm{CI}}_{50}={\mathrm{OD}}_{\mathrm{Mon}}/{\mathrm{OD}}_{50\mathrm{Mon}}+{\mathrm{OD}}_{\mathrm{Erl}}/D_{50\mathrm{Erl}}$$ (2)

2.4. Sphere Formation Analysis

CMT‐U27S and MDA‐MB‐231S were treated for 24 h with one of the following: Cells were exposed to solvent control (ethanol), MON (0.25 μM), ERL (4.00 μM), or combination (MON 0.25 μM + ERL 4.00 μM), harvested, rinsed with PBS, and then transferred to DMEM/F12 complete medium for further culture. We seeded 5000 cells/well in low attachment (6‐well) plates. After 7 days of cell culture, spheres were harvested, and then suspended into single cells, passaged for next generation of sphere formation. Sphere numbers were quantified on day 7 after re‐seeding using microscopy.

Spheres with a diameter ≥ 60 μm were counted as valid. Three fields of view were selected for each cell sample.

2.5. Cell Invasion Analysis

Pretreated cells with the same drug treatments as described in Sphere Formation Analysis for 24 h before seeding. Subsequently, cells were trypsinized, washed with PBS, and stained with trypan blue for counting.

Matrix Coating: Matrigel matrix was diluted in DMEM/F12 medium at 1:8. Then, 50 μL of diluted Matrigel was coated into the upper chamber of each insert and allowed to solidify overnight at room temperature. All Matrigel handling steps were performed on ice. The remaining procedures followed the previously described cell migration assay protocol.

2.6. Cell Apoptosis Analysis

Apoptotic cells were detected by Annexin V‐FITC/PI staining. Cells were harvested by completely digesting, resuspended, incubated with 6‐well plates, and then cultured for 48 h. The cells were treated as described in Sphere Formation Analysis. Cells were harvested post‐treatment, washed twice, and subjected to Annexin V‐FITC/PI staining. Stained cells were detected by Flow cytometric analysis (BD Biosciences), and the acquired data were evaluated using FlowJo software.

2.7. CD44 ⁺/CD24 ⁻ Stem Cell Population Analysis

BCSCs were treated with drugs as described in Sphere Formation Analysis for 24 h. Following incubation, cells were resuspended in staining solution for immunostaining with PE‐anti‐CD24 and FITC‐anti‐CD44 antibodies. The CD44⁺/CD24⁻ population was isolated by FACS, and its proportion was determined and analysed.

2.8. Protein Expression Analysis

We extracted total protein from cancer cells and tumour tissues that had been treated with MON and ERL. BCA protein kit was used to test protein concentration. Primary antibodies in Table 1 were used:

TABLE 1.

Primary antibody information for Western blotting.

Primary antibody	IHC/WB	Host species	Antibody dilution	Manufacturer	Code
β‐actin	WB	Rabbit	1:5000	Servicebio	GB11001
EGFR	WB	Rabbit	1:5000	Abcam	ab52894
p‐EGFR	WB	Rabbit	1:1000	CST	3777T
PI3K	WB	Rabbit	1:2000	CST	4249
p‐AKT	WB	Rabbit	1:1000	Proteintech	80,455‐RR
AKT	WB	Mouse	1:5000	Proteintech	60,203‐Ig

2.9. Xenograft Tumour Model in Nude Mice

Twenty‐four female BalB/C nude mice (6 weeks old, 20–22 g) were obtained from Weitonglihua Company. During the experiment, mice were kept under specific pathogen‐free (SPF) conditions and had free access to food and water. The study adhered to international ethical standards for animal research.

CMT‐U27S (1 × 10⁵ cells) were injected into the mammary pads of nude mice (n = 24, female, 6‐week‐old), which were housed in a SPF condition. When tumours grew to approximately 75 mm³, the mice were randomly assigned to four groups (n = 6 per group): (1) Control group: Intraperitoneal injection (i.p.) of saline (volume‐matched to Group 4), (2) 2.5 mg/kg MON. (3) 20 mg/kg ERL. (4) MON 1.25 mg/kg + ERL 10 mg/kg. Treatments were administered every 2 days for 4 weeks. Tumours were measured to calculate volume via the following equation:

$$\text{Tumour volume}\left({\mathrm{mm}}^3\right)=\text{Length}\times {\text{Width}}^2$$ (3)

The mice body weights and tumour volumes were recorded weekly. At the experimental endpoint, mice were anaesthetised, and then euthanized by cervical dislocation. Blood samples, tumour tissues and organs were harvested, weighed, and stored appropriately for subsequent analysis.

2.10. Haematoxylin and Eosin Staining Analysis

Haematoxylin and Eosin Staining (H & E) of the tumour and organ specimens was carried out according to a standard five‐step protocol. (1) Fixed in 10% formalin. (2) Embedded into paraffin. (3) Sectioned into slices (4‐μm‐thick). (4) Stained. (5) Examined by a light microscope, and images were taken for analysis to assess cellular morphology and tissue architecture.

2.11. Immunohistochemical Analysis

Immunohistochemical (IHC) staining was performed in accordance with the antibody manufacturer's instructions by an ultrasensitive two‐step method. The embedded tissue sections underwent deparaffinisation, rehydration, and IHC staining. Primary antibody (Ki 67, Cleaved Caspase‐3, p‐EGFR, PI3K, p‐AKT and SOX2, details regarding the antibodies are available in Table 2) incubation: Tissue sections underwent serum blocking (ab7481, Abcam, USA) and were placed in a 4°C refrigerator overnight. Secondary antibody (Goat Anti‐Rabbit IgG‐HRP, G1267, Servicebio, China) incubation: Secondary antibodies were applied at 25°C for 10 min. The rabbit IgG (1:1000) was used as the negative control for IHC. Signals were visualised by a DAB kit; the sections were stained with haematoxylin, followed by dehydration, clearing, and mounting. The positive cells were recognised by the appearance of brown staining. Six tumour tissues were collected from each group, and the six fields from the maximum cross‐sectional areas were analysed. Expression levels in these six fields were quantified using ImageJ 1.47v software (NIH, Bethesda, MD, USA).

TABLE 2.

Primary antibody information for immunohistochemistry.

Primary antibody	IHC/WB	Host species	Antibody dilution	Manufacturer	Code
Ki 67	IHC	Rabbit	1:1000	Abcam	ab15580
Cleaved casepase 3	IHC	Rabbit	1:100	Abcam	ab32042
SOX2	IHC	Rabbit	1:200	Abcam	ab92494
p‐EGFR	IHC	Rabbit	1:200	CST	3777T
PI3K	IHC	Rabbit	1:200	CST	4249
p‐AKT	IHC	Rabbit	1:500	Proteintech	80455‐RR

2.12. Immunofluorescence Analysis

Tumour tissue sections were prepared for immunofluorescence (IF) staining, then incubated with the primary antibodies: FITC‐CD44 antibody (103021, BioLegend, USA) and PE‐CD24 antibody (101 807, BioLegend, USA). After washing, the tumour tissue sections were further incubated with goat anti‐mouse IgG (ab6789, Abcam, USA). Nuclei were counterstained with DAPI medium (C0060, Solarbio, China). Negative control sections were included for every single IF staining experiment. All tissue sections were accessed by a Zeiss scanning microscope.

2.13. Data Analysis

Results are shown as mean ± SEM. The statistical significance of the difference was analysed using one‐way ANOVA, followed by Tukey's multiple comparison tests. All statistical analyses were performed with GraphPad Prism 9.0. *p < 0.05 was considered as statistically significant, **p < 0.01, and ***p < 0.001 were considered as extremely significant, compared to control group. ^# p < 0.05 was considered as statistically significant, ^## p < 0.01, and ^### p < 0.001 were considered as extremely significant means significant difference, compared to combination group.

3. Results

3.1. Screening, Isolation, and Characterization of CSCs

As shown in Figure 1A, both CMT‐U27 and MDA‐MB‐231 cells grew adherently in medium added with 10% FBS. CMT‐U27 cells displayed an elongated spindle‐like morphology, whereas MDA‐MB‐231 cells exhibited irregular polygonal shapes with nearly circular contours. And the microspheres displaying breast cancer stem cell characteristics, named as CMT‐U27 Sphere (CMT‐U27S) and MDA‐MB‐231 Sphere (MDA‐MB‐231S) cells. As illustrated in Figure 1B, flow cytometric phenotyping revealed that both CMT‐U27S and MDA‐MB‐231S predominantly exhibited the CD44⁺/CD24⁻ molecular phenotype. After enrichment by serum‐free culture, the proportions of CD44⁺/CD24⁻ cells reached 75.79% ± 6.78% and 73.21% ± 8.78%. Figure 1C shows the comparative analysis results.

FIGURE 1.

Characterization of cancer stem cells (CSCs) from CMT‐U27 and MDA‐MB‐231 cancer cell lines: Morphology, Surface Marker Identification, and Tumorigenic Activity. (A) Morphology of CMT‐U27, MDA‐MB‐231 cancer cells and their CSCs, (B, C) Percentage of cells identified as CD44⁺/CD24⁻ molecular phenotype, (D) Effect of MON on the activity of CMT‐U27, MDA‐MB‐231 cancer cells and their CSCs, (E) Effect of ERL on the activity of CMT‐U27, MDA‐MB‐231 cancer cells and their CSCs. n = 3; Compared with the control group, **p < 0.01, ***p < 0.001.

As shown in Figure 1D, MON significantly reduced MDA‐MB‐231 and CMT‐U27 cell viability at 0.50 μM, but suppressed CSCs (CMT‐U27S and MDA‐MB‐231S) proliferation at 0.25 μM (p < 0.01). As shown in Figure 1E, ERL inhibited MDA‐MB‐231 proliferation at 4 μM and CMT‐U27 viability at 6 μM, yet an 8 μM concentration was required to exert a pronounced inhibitory effect on CSCs (CMT‐U27S and MDA‐MB‐231S) (p < 0.01).

3.2. Effects of MON Combined With ERL on CSCs Activity and Apoptosis

Combined treatment with MON (0.25 or 0.50 μM) and ERL (2–8 μM) significantly increased the anti‐proliferative effect of ERL alone in both CMT‐U27S and MDA‐MB‐231S (p < 0.01; Figure 2A,B), indicating MON‐induced sensitization to the EGFR inhibitor. Synergy was confirmed by combination index (CI) values below 1.0 for all tested ratios (1:2 to 1:32), with the 1:16 ratio showing the strongest effect against CSCs (Figure 2C).

FIGURE 2.

Effects of MON and ERL combination on the activity and apoptosis of CSCs from CMT‐U27 and MDA‐MB‐231 cancer cell lines. (A and B) Effects of MON combined with ERL on the activity of CSCs, (C) Combination index of MON combined with ERL on CSCs, (D, E and F) Effects of MON combined with ERL on CSCs apoptosis. n = 3. Compared with the control group, **p < 0.01, ***p < 0.001. Compared with combination group, ^## p < 0.01, ^### p < 0.001.

As shown in Figure 2D, MON alone and combined with ERL significantly induced apoptosis in both CSCs, unlike ERL monotherapy which was ineffective. The statistical analyses of these findings are shown in Figure 2E,F.

3.3. Effects of MON Combined With ERL on CSCs Sphere Formation and Invasion

Figure 3A showed the combination treatment exerted the most potent suppression on microsphere formation in both CSC lines. It significantly reduced (p < 0.001) the microsphere count. MON alone also showed a significant inhibitory effect, while ERL alone had a minimal impact. A similar trend was observed in microsphere size (Figure 3B). The corresponding statistical analyses are presented in Figure 3C,D.

FIGURE 3.

Effects of MON combined with ERL on CSCs from CMT‐U27 and MDA‐MB‐231 cancer cell lines Sphere Formation and Invasion. (A) Effect of MON combined with ERL on sphere formation, (B) effect of MON combined with ERL on invasion of CSCs, (C–F) Data analysis results. n = 3. **p < 0.01, ***p < 0.001 compared to control group. ^## p < 0.01, ^### p < 0.001 compared to combination group.

As shown in Figure 3B, co‐treatment with MON (0.25 μM) and ERL (4.00 μM) most strongly inhibited the invasion of both CMT‐U27S and MDA‐MB‐231S. The invasive cell count was drastically reduced (p < 0.001). MON alone also showed a significant inhibitory effect, whereas the combination regimen demonstrated superior efficacy. Data analysis results are shown in Figure 3E,F.

3.4. Effects of MON Combined With ERL on CSCs Population

As shown in Figure 4A, combination treatment with MON (0.25 μM) and ERL (4.00 μM) most effectively reduced the CD44⁺/CD24⁻ stem cell subpopulation in both CSC lines from CMTs and HBC. In CMT‐U27S cells, the percentage was drastically lowered (p < 0.001). MON monotherapy also showed a significant inhibitory effect, yet the combination regimen demonstrated superior efficacy. Data analysis results are shown in Figure 4B,C.

FIGURE 4.

Effects of MON combined with ERL on CSCs Population. (A) Effect of MON combined with ERL on CSCs population in CMT‐U27S and MDA‐MB‐231S, (B and C) Data analysis results. n = 3. Compared with control group, *p < 0.05, **p < 0.01, ***p < 0.001. Compared to combination group, ^## p < 0.01.

3.5. Effects of Combination With MON and ERL on EGFR Pathway in CSCs

As shown in Figure 5A,B, the combination treatment most potently suppressed EGFR phosphorylation in both CMT‐U27S and MDA‐MB‐231S (p < 0.001), outperforming either agent alone (p < 0.05). Additionally, MON specifically targeted the PI3K/AKT pathway, significantly downregulating PI3K and p‐AKT protein levels in both cell lines (p < 0.01), but did not affect the EGFR/ERK signalling axis. The corresponding quantitative data analyses are presented in Figure 5C,D.

FIGURE 5.

Effects of MON combined with ERL on EGFR pathway in CSCs from CMT‐U27 and MDA‐MB‐231 cancer cell lines. (A and B) Effect of MON combined with ERL on EGFR pathway in CMT‐U27S and MDA‐MB‐231S, (C and D) Data analysis results. n = 3. Compared with control group *p < 0.05, **p < 0.01, ***p < 0.001. Compared with combination group, ^# p < 0.05, ^## p < 0.01 (Full‐length blots are available upon request or as review‐only data.).

3.6. Effects of MON Combined With ERL on CSCs Growth in Vivo

Following the experimental scheme (Figure 6A), in vivo studies demonstrated that the MON and ERL combination most effectively suppressed tumour growth, as evidenced by significantly reduced tumour weights (Figure 6B) and volumes (Figure 6D), with no severe systemic toxicity indicated by body weight monitoring. In contrast to the highly vascularized and viable tumours in the control group, histopathology (Figure 6C) showed extensive necrosis in all treatment groups. Lung metastasis incidence was highest in control groups (4/6), reduced by MON (1/6) or ERL (2/6). No significant differences were found in splenomegaly among the groups, and no other organ abnormalities were detected. Immunofluorescence analysis (Figure 6E,F) revealed a significant reduction of CSCs from CMTs in tumour tissues following either MON or combination with ERL (p < 0.01).

FIGURE 6.

Effects of MON combined with ERL on CSCs from CMT‐U27 cancer cell lines Growth in vivo. (A) Experimental Protocol, (B) Tumour‐bearing nude mice body weight, (C) Histopathology of tumours, lungs and spleens, (D) Tumour photographs, (E and F) Immunofluorescence analysis of CSCs population, n = 6, Compared with control group *p < 0.05, **p < 0.01, ***p < 0.001. Compared with combination group, ^## p < 0.01.

As shown in Figure 7A, the results showed that positive cells of tumour proliferation (Ki 67) and CSC marker (SOX2) were reduced after treating with MON, ERL or combination of them. According to the data analysis results, the combination of MON and ERL significantly inhibited the expression of Ki 67 and SOX2, while enhancing the expression of the apoptotic protein cleaved caspase‐3 in tumour tissues (Figure 7B–D). Combination of MON and ERL also synergistically suppressed EGFR phosphorylation and the PI3K/AKT pathway, as evidenced by the significant reductions in p‐EGFR, PI3K, and p‐AKT levels (Figure 7E–G).

FIGURE 7.

Effects of MON combined with ERL suppressed tumour proliferation, apoptosis and the EGFR/PI3K pathways in CSCs from CMT‐U27 cell lines. (A) IHC staining of Ki 67, Cleaved caspase‐3, CSC marker SOX2, p‐EGFR, PI3K, p‐AKT, (B–G) Data analysis results. Compared with control group *p < 0.05, **p < 0.01, ***p < 0.001. n = 6. Scale bars = 50 μm.

In summary, we found that ERL, as an EGFR inhibitor, targeted the EGFR/ERK pathway, while MON inhibited the PI3K/AKT signalling pathway, enhancing EGFR inhibition. Combination of the MON and ERL synergistically downregulated the EGFR/PI3K/AKT pathways, thereby effectively suppressing the proliferation, invasion, and metastasis of CMT‐U27 cells and their stem cells, CMT‐U27S, as shown in Figure 8.

FIGURE 8.

Schematic diagram illustrates that the combined application of MON and ERL treats CMT‐U27 cancer cell lines and their CSCs by inhibiting the EGFR/PI3K/AKT pathways.

4. Discussion

In this study, cancer stem cell‐like spheres were isolated from canine mammary tumour cell line (CMT‐U27) and human breast cancer cell line (MDA‐MB‐231) via serum‐free culture and identified by flow cytometry as CSCs, and named as CMT‐U27S and MDA‐MB‐231S, respectively. In our prior study, we found that MON and ERL combination downregulated CSCs markers expression, such as SOX2 and OCT4, while also inhibiting cell proliferation and colony formation [24]. This led us to investigate whether this combination treatment could modulate the behaviour of CSCs derived from CMTs and HBC. We systematically examined its role in inhibiting cancer stem cells activity, metastasis, and CSCs populations in vitro and in vivo. We found that combination with MON and ERL significantly inhibited CSCs activity, induced cells' apoptosis, reduced cancer stem cells sphere formation, invasion and CSCs population, downregulated EGFR, PI3K and p‐AKT protein levels in CMTs or HBC.

As mentioned in the introduction, CMTs share considerable similarities with HBC in terms of dysregulation of EGFR, which activates key signalling pathways involved in tumour cell proliferation and metastasis [10, 11]. This pathway can be targeted to inhibit tumour growth via two distinct actions: (i) Since mTOR serves as a key controller of autophagic initiation, its inhibition induces autophagy, (ii) suppression of AKT phosphorylation, coupled with the activation of caspase‐3 and caspase‐9, leads eventually to apoptosis [25, 26]. Hence, drugs targeting EGFR have shown significant antitumor activity over the past decades [27, 28]. Among them, EGFR tyrosine kinase inhibitors have suggested efficacy not only in breast cancer but also in various other malignancies, including non‐small cell lung cancer and pancreatic cancer [29, 30, 31]. These EGFR inhibitors typically function by competitively binding to the ATP‐binding site, thereby blocking tyrosine kinase phosphorylation [32, 33]. However, currently available EGFR inhibitors, such as ERL, have shown limited response rates in TNBC, lung cancer, and other cancers [34, 35, 36].

The persistence of CSCs often limits the efficacy of conventional therapy and leads to tumour relapse [37]. Research indicates CSCs in CMTs as key drivers of post‐treatment recurrence and metastasis, owing to their high tumorigenic potential, multidrug resistance, and invasiveness [38]. CSCs are identified using various molecular markers, including CD44 and CD24 [39, 40]. The CD44⁺/CD24⁻ phenotype molecular is frequently used as CSCs identification and has been associated with high‐grade tumours [40]. Growing evidence confirms the presence of CSCs, indicated by CD44⁺/CD24⁻ phenotypic cells, microsphere‐forming ability, high ALDH activity, elevated stemness markers (such as SOX2 and OCT4) expression, and abnormal activation of stemness‐related signalling pathways such as Wnt/β‐catenin and PI3K/mTOR [41, 42].

To improve the therapeutic efficacy of targeting EGFR and eliminate CSCs, we adopted a combination therapy consisting of the ionophore antibiotic monensin (MON) and an EGFR inhibitor. MON has been shown to suppress EGFR in human head and neck squamous cell carcinoma (HNSCC), acting through a mechanism analogous to that of statins, which block EGFR activation and downstream signalling [43]. According to previous findings, co‐administration of MON and rapamycin inhibits neuroblastoma xenograft growth, promotes apoptosis, and reduces PI3K/AKT/mTOR expression [44]. Our earlier findings revealed that MON alone or in combination therapy lowered the expression of PI3K (PI3Kp110/PI3Kp85), p‐AKT, and p‐mTOR in both TNBC cells and nude mouse tumour tissues [21]. In addition, MON served as both a potent inhibitor of EMT‐like CSCs [22] and a suppressor of gastric CSCs [45]. The potent antitumor effects of combination with MON and ERL, achieved at low doses and after short exposure times, position them as promising therapeutic candidates for breast cancer, especially in the metastatic setting. In conclusion, the combination of MON and ERL suppresses CMT‐U27 cells and their stem cells CMT‐U27S through simultaneous modulation of the EGFR and PI3K signalling pathways.

Comparing our findings with previous studies (including our own prior work), the novelty of this study lies in the first demonstration of synergy between MON and ERL in CSC‐enriched populations derived from CMTs and HBC. Due to the difficulty in obtaining CSCs derived from CMTs and HBC, the experimental progress has been slow, and there are many limitations in this study. Therefore, the shortcomings of the experiments are elaborated in the following sections in order to accumulate some experience for future research.

5. Limitations

Several limitations should be acknowledged in this study. First, the study relied on specific breast cancer cell lines (e.g., CMT‐U27 and MDA‐MB‐231); therefore, the findings need to be validated in additional models, such as other cell lines and primary patient‐derived samples. Second, cancer stem cells were defined primarily using the CD44⁺/CD24⁻ marker, and future studies should incorporate additional stemness markers such as Sox2, Oct4, and Nanog to further support the cancer stem cell phenotype. Third, the current study lacks a detailed in vivo toxicity assessment of the MON and ERL combination therapy, which is essential for evaluating its translational potential. Addressing these limitations in future work will strengthen the conclusions of this study.

Funding

This work was supported by the Fundamental Research Funds for the Central Universities (KYCXJC2025003).

Ethics Statement

All animal experiments were conducted according to the guidelines of the Experimental animal Ethics Committee of JinLing Hospital (Approval No.: 2020 JLHFTDWLS‐0010).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

All data generated or analysed during this study are included in this published article.