SLC31A1 promotes chemoresistance through inducing CPT1A-mediated fatty acid oxidation in ER-positive breast cancer

Highlights

• •SLC31A1 is associated with breast cancer progression
• •SLC31A1 expression levels are up-regulated in ER-positive breast cancer patients
• •SLC31A1 is associated with tumor recurrence in ER-positive breast cancer
• •SLC31A1 deficiency rescues chemosensitivity in ER-positive breast cancer cells
• •SLC31A1 mediates CPT1A-associated FAO in ER-positive breast cancer cells
• •SLC31A1 promotes FAO by regulating CPT1A in breast cancer cells

Abstract

Over 60% of breast cancer cases are diagnosed with estrogen-receptor (ER) positive. Tamoxifen (TAM), a commonly employed medication for ER-positive breast cancer, often yields suboptimal therapeutic outcomes due to the emergence of TAM resistance, leading to the recurrence and a poor prognosis. The copper transporter, solute carrier family 31 member 1 (SLC31A1), has been associated with tumor aggressiveness and unfavorable outcomes in various types of tumors. In our current study, we found high expression of SLC31A1 that predicted poor survival in patients with breast cancer. Significantly, ER-positive breast cancer tissues in patients with recurrence post-TAM treatment exhibited considerably stronger SLC31A1 expression levels. In vitro experiments verified that TAM-resistant ER-positive breast cancer cell lines expressed notably higher SLC31A1 levels compared to the parental cell lines. Of great significance, SLC31A1 depletion notably rescued TAM sensitivity in chemoresistant ER-positive breast cancer cells, as demonstrated by the attenuated cell proliferative and invasive capabilities. Conversely, promoting SLC31A1 significantly facilitated the proliferation and invasion of wild-type breast cancer cells. Subsequently, we detected reduced copper levels in TAM-resistant breast cancer cells with SLC31A1 depletion. Mechanistically, we observed that in chemoresistant breast cancer cell lines, SLC31A1 knockdown resulted in a substantial decrease in the expression of carnitine palmitoyltransferase 1A (CPT1A), a rate-limiting enzyme of fatty acid oxidation (FAO). RNA-Seq analysis indicated that FAO might be implicated in SLC31A1-mediated breast cancer progression. CPT1A was also overexpressed in TAM-resistant breast cancer cells, accompanied by enhanced FAO rates and ATP levels. Suppressing CPT1A significantly enhanced the chemosensitivity of TAM-resistant breast cancer cells in response to TAM treatments. Intriguingly, copper exposure dose-dependently increased CPT1A expression in chemoresistant breast cancer cells, but this could be abolished upon SLC31A1 knockdown, along with enhanced apoptosis, which elucidated that copper uptake contributed to CPT1A expression. Furthermore, SLC31A1 overexpression significantly augmented CPT1A expression in parental breast cancer cells, accompanied by facilitated copper levels, FAO rates, and ATP levels, while being notably diminished upon CPT1A suppression. Finally, our in vivo studies confirmed that SLC31A1 deficiency re-sensitized TAM-resistant breast cancer cells to TAM treatment and abolished tumor growth. Collectively, all our studies demonstrated that SLC31A1/copper suppression could enhance TAM responses for chemoresistant ER-positive breast cancer cells through constraining the CPT1A-mediated FAO process.

Graphical abstract

Introduction

Breast cancer is the most prevalent malignancy among females. As a highly heterogeneous disease, breast cancer possesses characteristics such as slow growth, favorable prognosis, high invasiveness, and poor prognosis and predictability [1]. Based on the expression status of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), breast cancer is typically classified into several subtypes: Luminal A, Luminal B, HER2-enriched, and triple-negative breast cancer (TNBC) [2,3]. Among these patients, approximately 60% present with aberrant ER expression. Tamoxifen (TAM), an antagonist of ERα66, is widely applied as a first-line endocrine in the treatment of ER-positive breast cancers to limit cancer recurrence and mortality. Regrettably, its therapeutic outcome is suboptimal due to the emergence of TAM resistance [4,5]. Consequently, TAM resistance has emerged as a predominant challenge in the treatment of ER-positive breast cancer patients. Nevertheless, the underlying molecular mechanisms responsible for TAM resistance and recurrence remain incompletely understood.

Copper serves as a catalytic cofactor for diverse physiological processes encompassing energy metabolism, mitochondrial respiration, signal transduction, and anti-oxidation [6,7]. Mounting studies have reported that dysregulated copper concentrations potentially contribute to various types of tumors, including gastric, breast, and colon cancers [[8], [9], [10]]. Copper-related therapeutics has been regarded as a novel opportunity for the management of chemotherapy-insensitive tumors, such as breast and lung cancers [11,12]. Accumulating evidence suggests that when copper levels in cells surpass the homeostatic threshold, it can induce cell death, which has recently been defined as cuproptosis [13]. However, copper is also a crucial cofactor for numerous metalloenzymes that contribute to tumor metastasis [14,15]. Copper plays an integral role in promoting tumor growth and angiogenesis, as it is recognized as a cofactor for certain pro-angiogenic molecules such as vascular endothelial growth factor (VEGF) [16]. Additionally, copper ions are implicated in the oncogenic BRAF signaling pathway, facilitating tumor cell proliferation and migration [17].

Solute carrier family 31 member 1 (SLC31A1), also known as copper transporter 1 (CTR1), is an indispensable copper transporter that influences dietary copper absorption across the cell membrane [18]. Elevated expression of the SLC31A1 gene is associated with an adverse prognosis and dysregulated immune cell infiltration in breast cancer, and overexpression of SLC31A1 leads to an increase in copper uptake in breast cancer cells and xenograft mouse models [19,20]. Recently, accumulating evidence indicates that SLC31A1 impacts patient prognosis and modulates the cisplatin (cDDP) concentration in various malignancies [21]. Furthermore, a robust inverse relationship between cDDP resistance and SLC31A1 expression has been established. Specifically, cDDP accumulates twice as much in cervical cancer cells when SLC31A1 is overexpressed [22]. Additionally, previous research has indicated that downregulation of zinc-finger protein 711 (ZNF711) inhibits SLC31A1 expression, subsequently promoting cDDP resistance in epithelial ovarian cancer [23]. Subsequent studies have revealed the association between SLC31A1 and cDDP chemoresistance in breast cancer, suggesting its potential as a prospective prognostic marker [24]. Inhibition of SLC31A1 degradation enhances cell sensitivity to cDDP treatment in osteosarcoma [25]. Conversely, a negative correlation between the expression of SLC31A1 and platinum resistance was previously reported in lung cancer [26]. Thus, the role of SLC31A1 in regulating chemoresistance is complex and may be related to the different types of tumor cells. Although the potential of SLC31A1 in breast cancer has been revealed, its impact on ER-positive breast cancer and recurrence remains largely unknown, particularly the underlying mechanisms.

In this study, we identified the copper transporter SLC31A1 as a molecular determinant implicated in ER-positive breast cancer progression and recurrence. Subsequently, by conducting in vitro and in vivo experiments, we explored how SLC31A1 influenced the proliferation, invasion, and sensitivity of TAM-resistant ER-positive breast cancer cells to uncover the underlying molecular mechanisms.

Materials and methods

Human samples

A total of 85 ER-positive breast cancer tissue samples with their paired normal tissues, accompanied by complete clinicopathological and follow-up data, were collected from patients who underwent radical mastectomy at the Department of Breast Medical Oncology, Shandong Cancer Hospital and Institute (Jinan City, Shandong Province, China) between January 2011 and December 2018. All these included breast cancer patients were pathologically diagnosed with ER-positive condition without radiotherapy, neoadjuvant chemotherapy or targeted therapy before surgical operation. All the patients were treated with standard tamoxifen therapy at least five years and stage appropriate chemotherapy. The Medical Ethics Committee of Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences approved this study, and the study was complied with the Declaration of Helsinki. Informed consents were obtained from all patients included or their family members. The information for clinical pathological characteristics of the tumor specimens was displayed in Table S1.

Public data acquisition and processing

The analysis of SLC31A1 mRNA expression in The Cancer Genome Atlas (TCGA) breast cancer cohort was carried out using online integrated tools (https://www.xiantao.love/). Subsequently, the correlation of SLC31A1 with clinical survival outcomes in the TCGA database was assessed using Kaplan-Meier survival analysis and the log-rank test. Overall survival (OS) and progression-free survival (PFS) and SLC31A1 were stratified into high and low expression groups based on the median value. The Tumor Immune Estimation Resource (TIMER) algorithm database (https://cistrome.shinyapps.io/timer/) was utilized to investigate the expression alterations of SLC31A1 in diverse tumor types, as well as the association between SLC31A1 and the expression levels of ESR1, ESR2, ERBB2, and PGR in BRCA-Basal, breast invasive carcinoma, BRCA-Luminal, and BRCA-Her2 samples. Images of immunohistochemical (IHC) staining for low/high expression of SLC31A1 in the breast cancer tissue microarray were retrieved from the online The Human Protein Atlas (https://www.proteinatlas.org/).

Cells and culture

Human breast cells such as T47D, MCF7, BT474, and MDA-MB231, along with the human non-cancerous mammary epithelium cell (MCF10A), were procured from the American Type Culture Collection (Manassas, VA, USA). Human breast cancer cell lines including ZR-75-1, HCC1954, and MDA-MB453 were acquired from Wuhan Pricella Biotechnology Co., Ltd. (Wuhan, China). All these cell lines were verified to be mycoplasma-free through morphological examination and had their identities confirmed by STR profiling. MDA-MB231 and MDA-MB453 were cultured in Discrete Macro-Element Modeling (DMEM) (#10313021, Gibco, NY, USA). MCF-10A cell line was cultured in DMEM/F-12 medium (#11320033, Gibco, USA). T47D, BT474, ZR-75-1 and HCC1954 cell lines were cultured in RPMI-1640 medium (#11875093, Gibco, USA). MCF7 cells were maintained in Minimum Essential Medium (#11095080, MEM, Gibco, USA) containing 10 μg/ml insulin (#11061-68-0, Sigma Aldrich, USA). Tamoxifen (TAM)-resistant MCF7 (MCF7/TR) and T47D (T47D/TR) cells were constructed by culturing the cells in the presence of 1 μM TAM for 6 months and were maintained in the culture medium containing 1 μM TAM consistently. All culture medium was supplemented with 10% fetal bovine serum (FBS) (#10091148, Gibco, USA) and 1% penicillin-streptomycin (#PB180120, Wuhan Pricella Biotechnology). All cells were incubated under a humidified atmosphere with 5% CO₂ at 37°C. The chemical reagents for cell treatments used in this study were obtained from the following providers: copper chelator tetrathiomolybdate (TTM) from MedChemExpress (#HY-128530, Shanghai, China), the irreversible inhibitor of CPT-1α etomoxir (ETO) (#HY-50202, Shanghai, China), TAM (#HY-13757A, Shanghai, China), palmitic acid (PA) (#ST3329, Beyotime Biotechnology, Shanghai, China), oleic acid (OA) (#60-33-3, Sigma Aldrich, USA) and CuSO₄ (#7758-98-7, Sigma Aldrich, USA).

In vitro transfection

Breast cancer cells lines that stably overexpressing SLC31A1 were established by lentiviral transduction using overexpression plasmids (GV219/CMV-MCS-SV40-Neomycin, Genechem Technologies, China). CPT1A siRNA (siCPT1A) and scramble control siRNA were synthesized from GenePharma Co. (Shanghai, China). Lipofectamine 3000 reagent was used for plasmids and siRNA transfections in Opti-MEM medium (#51985034, Gibco, USA) following the manufacturer's instructions (#L3000015, Invitrogen, USA). Lentiviral constructs of human SLC31A1 shRNAs (shSLC31A1) and the control shRNAs were purchased from Genechem Technologies (Shanghai, China). As for shRNA lentivirus infection, lentivirus was produced through transfecting the 293T cell line using the lentiviral vector and packing vector mix. After 48 h, lentivirus was collected and was used for cancer cell infection. Cells with stable knockdown were chosen using puromycin (3 μg/mL, Sigma Aldrich, USA) after 48 h infection.

RT-qPCR

Total RNA from cells and tissues was extracted using Trizol (Takara, Dalian, China) in accordance with the manufacturer's protocols. The PrimeScript™ RT Master Mix (Takara, Dalian, China) was used for cDNA synthesis. Then, RT-qPCR was performed to assess RNA expression levels using Hieff qPCR SYBR Green Master Mix (YEASEN, Shanghai, China). The relative targeting gene expression levels were quantified using the 2^–ΔΔCt method. GAPDH served as a housekeeping gene. The primers utilized for RT-qPCR are provided in Table S2.

Western blot

Total protein in tissues and cells was extracted using a RIPA lysis buffer (#P0013B, Beyotime Biotechnology, Shanghai, China) supplemented with Protease and phosphatase inhibitor cocktail (#P1049, Beyotime Biotechnology). The protein concentrations were subsequently quantified using the BCA Protein Assay Kit (#P0009, Beyotime Biotechnology) in accordance with the manufacturer's instructions. 20-40 μg of protein samples were separated by 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Merck Millipore, MA, USA). Subsequently, the membranes were blocked in 5% bovine serum albumin (BSA) (#ST023, Beyotime Biotechnology) in TBST for 1.5 h, followed by incubation with primary antibodies dissolved in TBST overnight at 4°C. After washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated IgG secondary antibodies in TBST for 1 hour. Finally, the immune complex was detected using an enhanced chemiluminescence (ECL) plus detection kit (#34095, Thermo Fisher Scientific). Bands were analyzed and quantified using ImageJ analysis software (National Institutes of Health, NIH). The expression levels of target proteins were normalized against β-actin. The primary antibodies against SLC31A1 (#ab129067, 1:1000, Abcam, Cambridge, UK), CPT1A (#66039-1-Ig, 1:1000, Proteintech Group Inc., Wuhan, China), ERα (#21244-1-AP, 1:1000, Proteintech Group Inc), β-actin (#4967, 1:1000, Cell Signaling Technology, USA), cleaved Caspase-3 (#abs132005, 1:1000, Absin, Shanghai, China) and cleaved PARP (#abs171752, 1:1000, Absin, Shanghai, China) were used for immunoblotting assay. Rabbit Anti-Mouse IgG H&L (HRP) (#ab6728, 1:5000) and Goat Anti-Rabbit IgG H&L (HRP) (#ab6721, 1:5000) were purchased from Abcam (Cambridge, UK) for secondary antibody incubation.

Immunofluorescence (IF) staining

After treatments, the cells were washed with PBS and fixed at room temperature with 4% paraformaldehyde for 20 min. Cells were then infiltrated using 0.5% Triton X-100 (#P0096, Beyotime Biotechnology) for 10 min at room temperature. Then, the cells were blocked in 5% goat serum (#C0265, Beyotime Biotechnology) in PBS at room temperature for 45 min and incubated overnight at 4°C with SLC31A1 (#67221-1-Ig, 1:200, Proteintech Group Inc.), ERα (#21244-1-AP, 1:200, Proteintech Group Inc.), Actin (#66009-1-Ig, 1:200, Proteintech Group Inc.) and cleaved Caspase-3 (#abs132005, 1:200, Absin, Shanghai, China) primary antibodies. After incubation with Alexa Fluor conjugated secondary antibodies in the dark for 1 h at 37°C, the cells were subjected to nuclear staining using DAPI (#P0131, Beyotime Biotechnology) for 5 min at room temperature. IF staining was finally captured under a confocal microscope.

For histological analysis, tumor tissues were fixed overnight in 4% paraformaldehyde. Then, the tissues were embedded in the tissue freezing medium and cut into 10-µm thick sections. Subsequently, the tissue sections were blocked in 5% goat serum containing 0.3% Triton X-100 in PBS for 45 min and then incubated with the primary antibodies against SLC31A1 (#67221-1-Ig, 1:200, Proteintech Group Inc.) and Cytokeratin (#26411-1-AP, 1:200, Proteintech Group Inc.) overnight at 4°C. After with PBS, the sections were incubated with the Alexa Fluor conjugated secondary antibodies in dark for 1 h at 37°C. Nuclei were stained with DAPI for 10 min at room temperature. Images were immediately captured using a confocal microscope. Goat Anti-Mouse IgG H&L (Alexa Fluor® 647) (#ab150115, 1:400) and Goat Anti-Rabbit IgG H&L (Alexa Fluor® 488) (#ab150077, 1:400) secondary antibodies were obtained from Abcam (Cambridge, UK).

Cell viability and proliferation calculation

A total of 1 × 10³ cells per well were seeded into a 96-well plate and cultivated at 37°C. After undergoing treatments as detailedly described in the figure legends, CCK8 reagent (#C0038, Beyotime Biotechnology) was added to each well and incubated for 2 hours at 37°C. Absorbance at a wavelength of 450 nm was measured using a microplate reader to assess the cell viability.

EdU uptake (#C0071L, Beyotime Biotechnology) was employed as the manufacturer's recommendation to assess the cell proliferation. Briefly, breast cancer cells were planted onto 96-well plates (1 × 10³/well). Following treatments, EdU (50 μM) was added to each well. Subsequently, the cells were fixed with 4% paraformaldehyde for 15 min before being permeabilized for 20 min using 0.5% Triton X-100 (Beyotime Biotechnology). After washing with PBS, click additive solution was added and incubated for an additional 30 min in the dark. Finally, the cells were captured using a fluorescent microscopy (Olympus, Japan).

Apoptosis analysis

Cells (2.5 × 10⁵) were planted on 6-well plates. After teratments, cells were washed with ice-cold PBS and stained with the Annexin V-YSFluorTM 647/PI Apoptosis Detection Kit (Yeasen Biotechnology, Shanghai, China) according to the manufacturer's instructions. The stained cells were analyzed by a Gallios Flow Cytometer (Beckman Coulter Inc., Brea, CA, USA). The data were processed using a FlowJo software (FlowJo LLC, Ashland, OR, USA).

Invasion analysis

For invasion calculation, the 8 µm porous membrane (Corning, NY, USA) of the transwell filter insert was coated with 0.5% Matrigel (#354230, Corning, USA). Cancer cells (5 × 10⁵ cells/ml) were seeded on the upper compartment containing serum-free medium, and medium containing 20% FBS was added to the lower chamber. After incubation for 24 h, the non-invasive cells adhered to the top of the membrane were removed by scrubbing. The invaded cells adhered to the opposite side of the membrane were fixed with methanol for 10 min and stained with 0.5% crystal violet for 5 min. Finally, five fields were randomly selected and captured under a light microscope.

Copper concentrations measurements

Breast cancer cells were digested in a mixture of nitric acid and hydrogen peroxide. All samples were incubated for 15 min at 25°C and subsequently heated at 120°C for a few hours for complete dissolution. Thereafter, the samples were diluted using ultrapure water. Copper contents were analyzed using an iCAP Q ICP-MS instrument (Thermo Fisher Scientific, Germany). The quantity was finally normalized to μg/10⁶ cells.

Fatty acid oxidation (FAO) rate assay

FAO was examined as previously described [27]. Briefly, cancer cells after treatments were collected and the mitochondria were extracted using a Cell Mitochondria Isolation Kit (#C3601, Beyotime Biotechnology) following the provider's instructions. Next, the mitochondria were used for determining the FAO rate according to the manufacturer's procedures of the Fatty Acid Oxidation Rate Assay Kit (#HL50679, Haling, Shanghai, China). The extracted mitochondria were then lysed and the oxidation rate was evaluated through assessing the reduction of palmitoyl carnitine oxidation-dependent ferricyanide.

ATP calculation and MitoTracker green staining

An ATP Microplate Assay Kit (#abs580117, Absin) was used to calculate the cellular ATP levels following the manufacturer's directions after treatments. In brief, cancer cells (5 × 10⁵ cells/well) were planted onto 100 mm dishes for 24 h. After each treatment, the cells were washed with cold PBS and collected into centrifuge tube. After centrifugation, 1 ml assay buffer for 5 × 10⁶ cell was added to the tube, followed by sonication and centrifuge. Then, the supernatants were put into a new tube and kept on ice for detection. Finally, ATP levels were measured at 660 nm using a microplate spectrophotometer.

Breast cancer cells post treatments were stained with MitoTracker Green (#C1048, Beyotime Biotechnology) to examine mitochondrial contents. In brief, cells after treatments were washed with PBS and were then exposed to 100 nM MitoTracker Green for 40 min at 37°C. Next, the cells were washed with PBS and incubated with Hoechst 33342 (#C1029, Beyotime Biotechnology) for nuclear staining for 5 min at 37°C. Images were captured using a fluorescent microscope.

RNA-seq and data analysis

Total RNA from cells or tissues was extracted using TRIzol (Invitrogen) and analyzed by an Agilent 2100 Bioanalyzer (Agilent Technologies, Beijing, China) and Qubit™ 3.0 Fluorometer (Invitrogen). RNA-seq libraries were produced and sequenced by the CapitalBio Technology (Beijing, China). Differentially expressed genes (DEGs) were analyzed by edgeR based on a negative binomial distribution. The Database for Annotation, Visualization and Integrated Discovery (DAVID) was performed for KEGG pathway analysis. GSEA was included for calculating that whether DEGs were associated with signaling pathways.

Lipid droplet staining

Intracellular lipid droplets were visualized with a fluorescent Bodipy dye. Breast cancer cells post culture were fixed in 4% paraformaldehyde for 15 min at room temperature, washed with PBS, permeabilized with 0.2% Triton X-100 (#P0096, Beyotime Biotechnology) in PBS for 10 min, and incubated with BODIPY 493/503 (#C2053S, Beyotime Biotechnology) diluted in PBS with a final concentration of 1 μg/ml overnight at 4°C. Cells were then washed with PBS and stained with DAPI (#P0131, Beyotime Biotechnology). Images were obtained under a confocal microscope.

Tumor inoculation and treatments

Animals were maintained under specific pathogen-free (SPF) conditions at the Animal Experimental Center of the Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences (Jinan City, Shandong PR, China) at 20-25°C with 50 ± 5% humidity and on a 12 h light/dark cycle. The female 6- to 8-week-old female BALB/c-nu/nu mice were purchased from the Vital River Laboratory Animal Technology Co. Ltd (Beijing, China). For in vivo animal studies, all mice were randomly divided into six groups with 4 in each as follows: the control group, TAM group, shSLC31A1 group, TAM+shSLC31A1 group; oe-Vector group and oe-SLC31A1 group. Briefly, 1 × 10⁶ MCF7-TR with stable SLC31A1 knockdown and MCF7 cells with SLC31A1 overexpression were suspended in 100 μl of media containing 50% Matrigel and subcutaneously injected into the left flank of female mice. The mice were checked for tumor weekly. After tumor cell injection for 7 days, 2 mg/kg TAM in corn oil was subjected to mice by oral gavage once a day. Tumor volumes were measured weekly using the following equation: volume = (length × width²) × 0.5. After 35 days, all mice were sacrificed and tumor samples were collected and weighed. Then, the tumor tissues were fixed in 4% paraformaldehyde for further studies. All procedures were conducted approved by the Principles of Laboratory Animal Care and according to the Ethics Committee of the Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences.

Histological staining

Tissues were fixed in 4% paraformaldehyde for 24 h and dehydrated with an ethanol gradient. The tissues were cleared with xylene, embedded in paraffin, and cut into 5-μm thick sections, which were then dewaxed, dehydrated, and stained with hematoxylin for 5 min and eosin for 3 min using a H&E staining kit (#C0105S, Beyotime Biotechnology). Regarding immunohistochemical (IHC) staining, tumor tissue sections (5 μm) were dewaxed in xylene and hydrated through an upgraded ethanol series. Antigen retrieval was accomplished by boiling the tissue sections in EDTA for 15 min in a microwave oven. Subsequently, endogenous peroxidase activity was quenched using 0.3% hydrogen peroxide solution for 10 min at room temperature. After washing with PBS, the sections were blocked in 10% goat serum (Beyotime Biotechnology) for 45 min. Thereafter, the samples were incubated with primary antibodies against KI-67 (#abs145260, Absin Biotechnology, 1:200 dilution), CPT1A (#66039-1-Ig, Proteintech Group Inc., 1:200 dilution), and N-Cadherin (#abs121341, Absin Biotechnology, 1:200 dilution) overnight at 4°C. Subsequently, antibody binding was detected by the use of Peroxidase/DAB (Gene Tech, Shanghai, China). Finally, tumor sections were counterstained with hematoxylin. Positive areas with KI-67, CPT1A and N-Cadherin staining were detected with a light microscope by two pathologists who were unknown of sample origins.

Statistical analysis

All data were expressed as the mean ± standard deviation (SD) from at least three independent experiments using GraphPad Prism® version 9.0 software (GraphPad Software, Inc., La Jolla, CA, USA). An unpaired two-tailed t-test was used for comparisons between two groups. One-way analysis of variance (ANOVA) followed by the post hoc Bonferroni test was used for multiple comparisons. The p values < 0.05 were considered statistically significant. No data were excluded from the study. Histological assays were analyzed through a blinded manner.

Results

SLC31A1 is associated with breast cancer progression

To investigate the clinically relevant factors associated with breast cancer progression and recurrence, RNA-seq analysis was conducted to identify the DEGs in tumor tissues with (recurrence, R) or without recurrence (non-recurrence, NR) and the normal samples. Among these DEGs, SLC31A1, a copper transporter, was highly expressed in the recurrent tumor specimens (Fig. 1A). To further explore the clinical and pathological relevance of SLC31A1 during breast cancer progression, its expression changes were evaluated using IF staining. The examination of SLC31A1 in our cohort by IF revealed that SLC31A1 expression was elevated in epithelial-derived tumor cells, and its expression was significantly higher in breast cancer tissues compared to that of the adjacent normal tissue samples. Meanwhile, a gradually increasing SLC31A1 was detected in advanced-stage tumor specimens compared to those in the early stage (Fig. 1B and C). To further explore the prognostic value and expression of SLC31A1, the TCGA database was analyzed, and the result demonstrated that SLC31A1 expression was significantly upregulated in breast invasive carcinoma (BRCA) samples compared to the paired normal samples (Fig. 1D). Additionally, higher SLC31A1 expression levels predicted poorer overall survival and progress-free interval survival rates than patients with lower SLC31A1 expression levels (Fig. 1E and F). We then investigated SLC31A1 protein expression levels using The Human Protein Atlas (HPA) database. The results in Fig. 1G showed that SLC31A1 expression was significantly up-regulated in breast tumor tissues with ≥25-75% stained cells, while its corresponding normal tissues were moderately stained or unstained. These aforementioned data suggested that SLC31A1 might be pathologically and clinically associated with breast cancer progression.

Fig. 1.

SLC31A1 is associated with breast cancer progression. (A) DEGs for the normal, paired primary and recurrent breast cancer tissues (n = 6 per group). (B) IF staining for SLC31A1 in tissues from our breast cancer cohort (n = 6 per group). Scale bar, 50 μm. (C) Quantification for SLC31A1 fluorescent intensity was shown. (D) The expression of SLC31A1 was elevated in breast cancer tissues compared to normal tissues based on TCGA database. (E,F) Kaplan-Meier plotter analysis showing that high SLC31A1 expression levels predict poor overall survival rate (OS) and progress free interval (PFI) in breast cancer patients based on TCGA dataset. (G) IHC staining results for SLC31A1 in patients with breast cancer from HPA database (https://www.proteinatlas.org/). Data were presented as means ± SD. ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

SLC31A1 expression levels are up-regulated in ER-positive breast cancer patients

In this part, we verified higher expression of SLC31A1 in BRCA tissues compared to normal tissues through TIMER analysis (Fig. 2A). Furthermore, we discovered a potential positive correlation between SLC31A1 expression levels and ESR1 (ERα) and ERS2 (ERβ) in BRCA-Basal, breast invasive carcinoma, and BRCA-Luminal samples. Despite detecting a positive correlation between SLC31A1 and ERBB2 (HER2) in BRCA samples, there was no significant difference observed between SLC31A1 and HER2 in BRCA-Basal and -Her2 specimens (Fig. 2B-E). Additionally, no significant difference was detected between SLC31A1 expression and PGR expression in BRCA-Basal and BRCA samples (Fig. 2B and C). In vitro studies further demonstrated that SLCA31A mRNA and protein expression levels were higher in ER-positive breast cancer cell lines (T47D, MCF7, BT474, ZR-75-1), while being undetectable or slightly expressed in ER-negative breast cancer cell lines (MB231, HCC1954, MB453) and the normal epithelial cell line (MCF10A) (Fig. 2F-H). Collectively, these data indicated that SLC31A1 expression might be related to ER-positive breast cancer growth.

Fig. 2.

SLC31A1 expression levels are up-regulated in ER-positive breast cancer patients. (A) Expression of SLC31A1 in varying tumor tissues and normal tissues from the TIMER database (https://cistrome.shinyapps.io/timer/). (B) Correlation between SLC31A1 gene expression and the expression of ESR1 (ERα), ESR2 (ERβ), ERBB2 (HER2) and PGR in BRCA-basal patients (n = 139) in TIMER. (C) Correlation between SLC31A1 gene expression and the expression of ESR1, ESR2, ERBB2 and PGR in breast invasive carcinoma (n = 1093) in TIMER. (D) Correlation between SLC31A1 gene expression and the expression of ESR1 and ESR2 in BRCA-luminal samples (n = 611) in TIMER. (E) Correlation between SLC31A1 gene expression and the expression of ERBB2 in BRCA-HER2 samples (n = 67) in TIMER. (F) RT-qPCR, (G) western blot and (H) IF staining assays for SLC31A1 mRNA and protein expression calculation in ER-positive (T47D, MCF7, BT474 and ZR-75-1) and -negative (MB231, MCF10A, HCC1954 and MB453) breast cancer cell lines (n = 3 in each group). Scale bar, 15 μm. SLCA31A, red; ER, green. Data were presented as means ± SD. *P<0.05, ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001.

SLC31A1 is associated with tumor recurrence in ER-positive breast cancer

To validate the potential of SLC31A1 expression levels in ER-positive breast cancer progression, its expression changes were subsequently explored in our cohort. RT-qPCR results demonstrated that SLC31A1 was highly expressed in ER-positive breast cancer specimens in comparison to the matched adjacent normal tissues (Fig. 3A and B). Notably, much stronger SLC31A1 gene expression levels were detected in ER-positive breast cancer samples from patients with recurrence (R) than those of the patients without recurrence (NR) (Fig. 3C). The IHC results revealed significantly stronger SLC31A1 protein expression levels in tumor sections collected from ER-positive breast cancer tissues with recurrence compared to the non-recurrent samples (Fig. 3D). The IF staining analysis indicated that post-treatment tumors from the NR group exhibited similar SLC31A1 expression levels to their pre-treatment counterparts (Fig. 3E and F). However, post-treatment tumors from non-responders with recurrence showed a marked increase in SLC31A1 expression levels compared to their pre-treatment counterparts (Fig. 3G and H). We further examined SLC31A1 expression in our cohort of breast cancer patients with metastasis. IF staining results indicated that SLC31A1 expression was highly elevated in metastatic tissues compared to the primary tumor tissues (Fig. 3I and J). Collectively, these results illustrated that SLC31A1 up-regulation predicted chemotherapy outcomes in ER-positive breast cancer patients.

Fig. 3.

SLC31A1 is associated with tumor recurrence in ER-positive breast cancer. (A,B) RT-qPCR analysis for SLC31A1 in tumor tissues from ER-positive breast cancer patients and the paired normal tissues from our cohort (n = 85 in each group). (C) RT-qPCR analysis for SLC31A1 mRNA expression levels in tumor tissues from ER-positive breast cancer patients received tamoxifen treatments with (n = 18) or without recurrence (n = 67). (D) IHC staining for SLC31A1 in tumor tissues from ER-positive breast cancer patients received tamoxifen treatments with (n = 6) or without recurrence (n = 6). Scale bar, 50 μm. (E,F) IF staining images of SLC31A1 and Cytokeratin in ER-positive breast cancer tissues obtained pre- and post-chemotherapy responders without recurrence (n = 6), and SLC31A1 positive fluorescent intensity was quantified. Scale bar, 50 μm. (G,H) IF staining images of SLC31A1 and Cytokeratin in ER-positive breast cancer tissues obtained from pre- and post-chemotherapy with recurrence (n = 6), and SLC31A1 positive fluorescent intensity was quantified. Scale bar, 50 μm. (I,J) IF staining images of SLC31A1 and Cytokeratin in primary ER-positive breast cancer tissues and metastatic tumor tissues (n = 6), and SLC31A1 positive expression fluorescence was quantified. Scale bar, 50 μm. Data were presented as means ± SD. ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001; ns, no significant difference.

SLC31A1 deficiency rescues chemosensitivity in ER-positive breast cancer cells

Subsequently, in vitro experiments were carried out to further investigate the impact of SLC31A1 expression changes on breast cancer progression and the underlying mechanisms using the TAM-sensitive breast cancer cell lines T47D and MCF7, and the constructed TAM-resistant (TR) ER-positive breast cancer cell lines (Fig. 4A). Initially, RT-qPCR results indicated that T47D/TR and MCF7/TR cells exhibited significantly elevated expression of SLC31A1 at the mRNA level compared to the corresponding sensitive groups (Fig. 4B). Western blot and IF staining confirmed significantly stronger SLC31A1 protein expression levels in T47D/TR and MCF7/TR cells than in the chemosensitive cell lines (Fig. 4C and D, Supplementary Fig. 1A and B). Thereafter, SLC31A1 expression levels were knocked down in T47D, MCF7, T47D/TR, and MCF7/TR cell lines through shRNA lentivirus infection to explore the regulatory potential of SLC31A1 on breast cancer growth. RT-qPCR and western blot results confirmed the successful knockdown of SLC31A1 in breast cancer cells (Fig. 4E and F). Subsequently, it was found that the absence of SLC31A1 further reduced the cell viability of T47D and MCF7 under TAM exposure, initially clarifying the suppressive effects of SLC31A1 inhibition on breast cancer proliferation. More importantly, shSLC31A1 significantly enhanced the sensitivity of T47D/TR and MCF7/TR cells to TAM treatment, as indicated by the decreased cell viability through CCK-8 analysis (Fig. 4G). Similarly, EdU staining confirmed that SLC31A1 depletion significantly reduced the proliferative capacity of T47D/TR and MCF7/TR cells in response to TAM incubation, which was comparable to the TAM single-treatment group (Fig. 4H). Transwell analysis demonstrated that TAM had no significant effect on the invasion of T47D/TR and MCF7/TR cells, while shSLC31A1 infection strongly inhibited the invasive ability of TAM-resistant breast cancer cells (Fig. 4I), accompanied by a marked decrease in the expression of invasive markers, including N-Cadherin, VEFG, and matrix metalloproteinase-9 (MMP9) (Supplementary Fig. 2A-C). Collectively, the above data elucidated that downregulation of SLC31A1 could enhance the sensitivity of T47D/TR and MCF7/TR cells to drug treatment.

Fig. 4.

SLC31A1 deficiency rescues chemosensitivity in ER-positive breast cancer cells. (A) CCK-8 analysis of T47D, MCF7, T47D/TR and MCF7/TR cells treated with varying concentrations of tamoxifen ranging from 0 to 20 μM for 24 h to examine the proliferation of tumor cells (n = 4 in each group). (B) RT-qPCR, (C) western blot and (D) IF staining assays for SLC31A1 mRNA and protein expression levels in parental T47D and MCF7 cells and TAM-resistant T47D (T47D/TR) and MCF7 (MCF7/TR) cell lines (n = 3 or 4 in each group). Scale bar, 15 μm. (E) RT-qPCR and (F) western blot analysis for SLC31A1 mRNA and protein expression levels in T47D, MCF7, T47D/TR and MCF7/TR cells with stable knockdown of SLC31A1 (n = 3 or 4 in each group). (G-I) T47D, MCF7, T47D/TR and MCF7/TR cells with or without SLC31A1 knockdown were exposed to TAM (2.5 μM) incubation for 24 h. Then, all cells were collected for subsequent assays. (G) Cell viability was assessed using CCK-8 analysis (n = 4 in each group). (H) EdU staining was used to calculate the cell proliferative capacity (n = 4 in each group). Scale bar, 50 μm. (I) Transwell analysis was conducted to examine the invasion of chemoresistant breast cancer cell lines (n = 4 in each group). Scale bar, 100 μm. Data were presented as means ± SD. ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001; ⁺⁺⁺P<0.001, ⁺⁺⁺⁺P<0.0001 vs the TAM group; ns, no significant difference.

SLC31A1 promotes the proliferation and invasion of ER-positive breast cancer cells in vitro

To further confirm the association between SLC31A1 and TAM resistance in ER-positive breast cancer cells, we overexpressed SLC31A1 in parental T47D and MCF7 cells, which were identified by RT-qPCR and western blot assays (Fig. 5A and B). As anticipated, CCK-8 analysis indicated that SLC31A1 overexpression significantly restored the proliferation of T47D and MCF7 cells compared with the oe-Vec control group (Fig. 5C). EdU staining revealed that SLC31A1 could significantly enhance the proliferative capacity of the wild-type breast cancer cells, as evidenced by the upregulated EdU-positive cells. Notably, the TAM-restricted proliferation of T47D and MCF7 cells was notably abolished upon SLC31A1 overexpression (Fig. 5D). Similarly, oe-SLC31A1 not only strongly promoted the invasion of T47D and MCF7 cells but also rescued the invasive ability of breast cancer cells (Fig. 5E), as evidenced by the increased number of invaded cells. Consistent with this, the suppressed expression levels of N-Cadherin, VEGF, and MMP9 in T47D and MCF7 cells by TAM were significantly reduced when SLC31A1 was overexpressed (Fig. 5F). These in vitro data presented above illustrated that the upregulation of SLC31A1 contributed to breast cancer proliferation and invasion.

Fig. 5.

SLC31A1 promotes the proliferation and invasion of ER-positive breast cancer cells in vitro. (A) RT-qPCR and (B) western blot analysis for SLC31A1 mRNA and protein expression levels in T47D and MCF7 cells transfected with SLC31A1 over-expression plasmids (n = 3 or 4 in each group). (C) T47D and MCF7 cells with or without SLC31A1 over-expression were subjected to increasing concentrations of TAM incubation for 24 h, followed by CCK-8 analysis to evaluate the cell proliferation (n = 4 in each group). (D-F) T47D and MCF7 cells post SLC31A1 over-expression plasmids transfection were incubated with or without TAM (2.5 μM) for 24 h. Subsequently, all cells were collected for studies as follows. (D) Cell proliferation was examined using EdU staining (n = 4 in each group). Scale bar, 50 μm. (E) The number of invasive cells was quantified using transwell assay (n = 4 in each group). Scale bar, 100 μm. (F) RT-qPCR analysis for invasion markers including N-Cadherin, VEGF and MMP9 in cells (n = 4 in each group). Data were presented as means ± SD. *P<0.05, ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

SLC31A1 mediates CPT1A-associated FAO in ER-positive breast cancer cells

Since SLC31A1 is a key molecule for copper transportation into cells, we then examined the copper levels in TAM-resistant breast cancer cells. As shown in Fig. 6A, we found higher copper concentrations in T47D/TR and MCF7/TR cells than that of their parental cell lines. Notably, both in the wild type and TAM-resistant ER-positive breast cancer cells, SLC31A1 knockdown significantly reduced the intracellular copper levels compared with the corresponding control groups. ER^- breast cancer cells are also reported to contain high copper levels [28]. To investigate whether SLC31A1 could mediate copper uptake in ER^- breast cancer cells, copper concentrations in MB231, HCC1954, and MB453 were examined. Meanwhile, the non-tumor cell line MCF10A was used as a control. As shown in Supplementary fig. 4A, ER^- breast cancer cells also contained higher copper levels, which were similar to those in ER⁺ breast cancer cells. We further knocked down SLC31A1 in MB231 and HCC1954 cells. Surprisingly, SLC31A1 depletion had no significant effects on the cellular copper levels (Supplementary fig. 4B and C). The expression of SLC31A1 was very low in ER^- breast cancer cells. Therefore, we hypothesized that although there was also a high level of copper in ER^- breast cancer cells, the regulation of cell proliferative and migratory behaviors through the copper uptake process might be independent of SLC31A1 in ER^- breast cancer cells.

Fig. 6.

SLC31A1 mediates CPT1A-associated FAO in ER-positive breast cancer cells. (A) Copper concentrations in T47D, MCF7, T47D/TR and MCF7/TR cells with or without SLC31A1 knockdown (n = 4 in each group). (B) Volcano plot of DEGs in T47D/TR cells with SLC31A1 deletion versus the control group (n = 4 in each). Genes with p value <0.05 and gene fold change ≥1.5 were marked in purple (down-regulated) and orange (up-regulated) with genes not meeting the cut-off shown in gray. (C) KEGG enrichment scatter plot for DEGs genes in shSLC31A1 T47D/TR cells compared with the control group. (D) GSEA enrichment of the signaling pathways for breast cancer- and FAO-related genes. (E) Heatmap showing the FAO-associated gene expression levels by RT-qPCR analysis from T47D/TR cells with or without SLC31A1 knockdown (n = 3 in each). (F) RT-qPCR and (G) western blot analysis for CPT1A mRNA and protein expression levels in T47D, MCF7, T47D/TR and MCF7/TR cells as shown (n = 3 or 4 in each group). (H) FAO rates were measured in T47D, MCF7, T47D/TR and MCF7/TR cells (n = 4 in each group). (I) ATP levels in T47D, MCF7, T47D/TR and MCF7/TR cells were examined (n = 4 in each group). (J) Cell viability was examined in T47D/TR and MCF7/TR cells with CPT1A silence after varying concentrations of TAM incubation for 24 h (n = 4 per group). (K) BODIPY493/503 staining of T47D/TR and MCF7/TR cells with or without CPT1A deletion after PO stimulation (200 μM) for 24 h (n = 4 per group). Scale bar, 15 μm. (L) Representative images for MitoTracker (green) and actin (red) of T47D/TR and MCF7/TR cells with or without siCPT1A transfection after PO exposure (200 μM) for 24 h. Relative MitoTracker fluorescence was quantified (n = 4 per group). Scale bar, 15 μm. Data were presented as means ± SD. *P<0.05, ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

RNA-Seq analysis was subsequently carried out to delve into the underlying molecular mechanisms mediated by SLC31A1 during breast cancer progression using T47D/TR cells with or without shSLC31A1 knockdown. Volcano plots depicted DEGs comparing shSLC31A1 T47D/TR cells to the control group. We discovered that 481 genes were downregulated in T47D/TR cells relative to the control ones, encompassing CPT1A (Fig. 6B), a crucial gene in mediating the FAO process [29]. Further functional enrichment analysis of these DEGs by KEGG indicated that these signals were primarily implicated in the PPAR signaling pathway, fatty acid metabolism, gap junction, endocrine resistance, and fatty acid degradation (Fig. 6C). GSEA based on the acquired DEGs consistently disclosed that SLC31A1 was affiliated with fatty acid β-oxidation and metabolism (Fig. 6D). Additionally, we discerned a positive correlation between SLC31A1 and the expression levels of CPT1A, CPT2, acyl-CoA oxidase 1 (ACOX1), and ACOX2 in BRCA samples from the TCGA database (Supplementary fig. 3A-D). RT-qPCR results further demonstrated that the deletion of SLC31A1 conspicuously alleviated the expression of genes specifically implicated in FAO, including CPT1A, CPT1B, ACOX1, and 3-hydroxymethylglutaryl-CoA synthase 2 (HMGCS2), but augmented the fatty acid synthesis-related genes fatty acid synthase (FASN), acetyl-CoA carboxylase (ACC), and stearoyl-CoA desaturase (SCD) (Fig. 6E). Given the specific effect of SLC31A1 on CPT1A and its role in FAO, we postulated that SLC31A1-mediated breast cancer progression and recurrence might be correlated with the FAO process. Subsequently, we endeavored to assess CPT1A expression in TAM-resistant breast cancer cells. RT-qPCR and western blot assays indicated that T47D/TR and MCF7/TR cells manifested stronger CPT1A mRNA and protein expression levels than those of the parental breast cancer cell lines (Fig. 6F and G). As anticipated, higher FAO and ATP levels were detected in T47D/TR and MCF7/TR compared to the chemosensitive breast cancer cell lines (Fig. 6H and I). To explore the function of CPT1A on breast cancer recurrence and FAO, its expression was subsequently knocked down in breast cancer cell lines by transfecting with its siRNA. Transfection efficiency was verified by RT-qPCR and western blot analysis (Supplementary fig. 5A and B). Notably, CPT1A deficiency significantly enhanced the chemosensitivity of T47D/TR and MCF7/TR cells to TAM incubation by CCK-8 analysis (Fig. 6J). Subsequently, we further stimulated T47D/TR and MCF7/TR cells to generate droplets by using PO (PA+OA, 200 μM). In accordance with expectation, CPT1A deletion resulted in significantly higher lipid deposition in TAM-resistant T47D and MCF7 cells (Fig. 6K), as indicated by the larger size of lipid droplets, revealing the limited FAO process. Furthermore, PO-induced mitochondrial injury was exacerbated in T47D/TR and MCF7/TR cells transfected with siCPT1A (Fig. 6L). Mitochondrial damage mediated by FAO suppression due to CPT1A repression induces apoptotic cell death and exerts anti-cancer effects [30,31]. We further discovered that PO stimulation significantly led to an increase in the apoptosis rates in T47D/TR and MCF7/TR cells, while being notably promoted upon CPT1A deficiency, accompanied by downregulated ATP levels in the cells (Supplementary fig. 6A and B). RT-qPCR results demonstrated that PO treatment significantly increased the expression of the pro-apoptotic signal Bax, while decreasing the expression of the anti-apoptotic molecule Bcl-2. However, these events were further accelerated when CPT1A was depleted in PO-treated T47D/TR and MCF7/TR cells (Supplementary fig. 6C). As expected, the expression levels of cleaved Caspase-3 and PARP, well-known apoptosis markers, were also found to be markedly upregulated by PO stimuli, and were further enhanced in PO-exposed ER⁺ breast cancer cells with CPT1A silencing (Supplementary fig. 6D). These results suggested that CPT1A suppression contributed to ATP reduction and apoptosis induction in ER⁺ breast cancer cells. Together, these results above demonstrated that SLC31A1-mediated recurrence of ER-positive breast cancer might be associated with FAO process through controlling CPT1A expression.

Copper/SLC31A1 enhances CPT1A expression levels in chemoresistant breast cancer cells

To explore the physiological effects of copper on CPT1A expression in TAM-resistant breast cancer cells, we discovered that copper supplementation led to CPT1A expression levels in a dose-dependent manner, and importantly, SLC31A1 deletion could significantly abrogate copper-triggered increase of CPT1A in T47D/TR and MCF7/TR cell lines (Fig. 7A). To uncover the impacts of SLC31A1 on copper uptake, CPT1A signaling, and associated cell death, apoptosis was then examined in T47D/TR and MCF7/TR cells under CuSO4 coculture. As depicted in Supplementary fig. 7A, we found that copper overload further decreased the apoptotic rates; however, SLC31A1 knockdown potently induced apoptosis in ER+ breast cancer cells, accompanied by a marked upregulation of cleaved Caspase-3 expression (Supplementary fig. 7B). These data further suggest that copper loading-induced tumor cell proliferation might be partially attributed to the suppression of apoptotic cell death, which was abolished upon SLC31A1 deletion, particularly highlighting the significance of SLC31A1 in mediating copper uptake and associated apoptosis in ER⁺ breast cancer cells with chemoresistance. Moreover, CPT1A suppression by its specific inhibitor ETO or siCPT1A markedly mitigated CPT1A expression levels in T47D/TR and MCF7/TR cells under copper exposure (Fig. 7B). Subsequently, we found that CuSO₄-impaired apoptosis and cleaved Caspase-3 protein expression levels were significantly abolished and induced when CPT1A signaling was suppressed by ETO or siCPT1A (Supplementary fig. 7C and D), confirming the role of copper/CPT1A axis blockade in triggering apoptosis and inhibiting the progression of ER⁺ breast cancer. Subsequently, we found that copper-induced CPT1A up-regulation was potently antagonized by copper chelator TTM in chemoresistant breast cancer cell lines (Fig. 7C). TTM, a potent copper chelator, has been demonstrated to exert efficient anti-cancer effects in various types of tumors, such as breast cancer, lung cancer, and liver cancer, by mediating DNA damage repair, apoptosis, and metastasis processes [[32], [33], [34]]. We then confirmed that copper chelation by TTM significantly induced apoptosis and Caspase-3 cleavage in T47D/TR and MCF7/TR cells under CuSO₄ stimulation (Supplementary fig. 7E and F). We then examined the influence of TTM single treatment on CPT1A expression changes in chemoresistant ER⁺ breast cancer cells. We found that TTM strongly inhibited CPT1A expression levels in T47D/TR and MCF7/TR cells in a dose-dependent manner, accompanied by decreased cellular copper levels (Supplementary fig. 8A and B). Consistent with the expression changes of CPT1A, significantly downregulated FAO rates and ATP levels were detected in TTM-incubated T47D/TR and MCF7/TR cell lines (Supplementary fig. 8C and D). These results suggested that disrupting copper accessibility by TTM also exhibited anti-cancer effects in ER⁺ breast cancer cells with chemoresistance by interrupting the FAO process and CPT1A signaling.

Fig. 7.

Copper/SLC31A1 enhances CPT1A expression levels in chemoresistant breast cancer cells. (A) T47D/TR and MCF7/TR cells in the absence or presence of stable SLC31A1 deletion were incubated with varying concentrations of CuSO₄ (1, 2.5, 5.0 and 10.0 μM) for 24 h. Then, the cells were collected for western blot analysis of CPT1A (n = 3 in each group). (B) T47D/TR and MCF7/TR cells with or without CPT1A silence were treated with copper for 24 h in the absence or presence of CPT1A inhibitor ETO (10 μM). Subsequently, all cells were harvested for CPT1A protein expression calculation using western blot assay (n = 3 per group). (C) T47D/TR and MCF7/TR cells were exposed to 10.0 μM of copper combined with or without increasing concentrations of copper chelator TTM. After treatments for 24 h, western blot analysis was used for CPT1A protein expression level calculation (n = 3 in each). Data were presented as means ± SD. ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

Considering the crucial effect of SLC31A1 on the regulation of cuproptosis, we next examined the expression changes of several other cuproptosis hallmarks. As shown in Supplementary fig. 9A and B, RT-qPCR results indicated that SLC31A1 knockdown and overexpression had no significant impact on the expression alterations of dihydrolipoamide branched-chain transacylase E2 (DBT), dihydrolipoamide S-acetyltransferase (DLAT), pyruvate dehydrogenase E1 subunit alpha 1 (PDHA1), dihydrolipoamide dehydrogenase (DLD), and ferredoxin 1 (FDX1) in chemoresistant and chemosensitive breast cancer cell lines. These data above revealed that the SLC31A1/copper axis might play a key role in inducing CPT1A expression levels, possibly contributing to the progression and recurrence of breast cancer independently of cuproptosis.

SLC31A1 promotes FAO by regulating CPT1A in breast cancer cells

Next, we aimed to further explore the potential of the SLC31A1/CPT1A axis in modulating FAO and breast cancer by overexpressing SLC31A1 in wild-type breast cancer cell lines. As expected, promoting SLC31A1 expression significantly enhanced CPT1A in T47D and MCF7 cells (Fig. 8A), along with higher copper concentrations (Fig. 8B). Moreover, FAO rates and ATP levels were markedly upregulated by SLC31A1 overexpression in parental breast cancer cells (Fig. 8C and D). Surprisingly, CCK8-analysis revealed that SLC31A1-facilitated proliferation of T47D and MCF7 cells was significantly abolished when CPT1A expression was inhibited by siCPT1A and ETO (Fig. 8E), which was validated by EdU staining (Fig. 8F and G), evidenced by the strongly decreased EdU-positive cells. Transwell analysis further demonstrated that CPT1A expression suppression abolished the promotive effect of SLC31A1 on the invasion of breast cancer cells (Fig. 8H and I). In addition, in T47D and MCF7 cells overexpressing SLC31A1, CPT1A inhibition significantly reduced the FAO rates and ATP levels (Fig. 8J and K), along with diminished expression levels of CPT1A (Fig. 8L). These data above elucidated that SLC31A1-promoted FAO was at least in part via CPT1A up-regulation, contributing to breast cancer proliferation and invasion.

Fig. 8.

SLC31A1 promotes FAO by regulating CPT1A in breast cancer cells. (A) Western blot analysis of CPT1A protein expression levels in T47D and MCF7 cells over-expressing SLC31A1 (n = 3 in each group). (B) Copper concentrations, (C) FAO rates and (D) ATP levels were examined in T47D and MCF7 cells transfected with oe-SLC31A1 or oe-Vec plasmids (n = 3 in each group). (E-L) T47D and MCF7 cells were transfected with oe-SLC31A1 and/or si-CPT1A in the presence or absence of ETO (10 μM) for 24 h. Then, all cells were collected for subsequent assays. (E) Cell viability was determined using CCK-8 analysis (n = 4 in each group). (F,G) EdU staining of T47D and MCF7 cells to assess cell proliferation (n = 4 in each group). Scale bar, 50 μm. (H,I) Invasive status of breast cancer cells was examined by transwell assay (n=4 in each group). Scale bar, 100 μm. (J) FAO rates and (K) ATP levels were examined (n=4 in each group). (L) CPT1A protein expression levels were measured by western blot analysis (n = 3 in each group). Data were presented as means ± SD. 7^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

SLC31A1 knockdown improves ER-positive breast cancer cells to tamoxifen treatment in vivo

To further evaluate the regulatory effects of SLC31A1 on the growth of TAM-resistant breast cancer in vivo, we subsequently implanted MCF7/TR cells subcutaneously into nude mice with or without TAM administration. As shown in Fig. 9A-C, TAM treatment had no significant effect on tumor growth compared to the control group; however, stable SLC31A1 knockdown not only significantly reduced the tumor growth rates and tumor weight, but also enhanced the response of these mice to TAM treatment, as indicated by the much lower tumor volume and weight. H&E staining indicated that the tumor cell density was significantly reduced in mice receiving TAM and shSLC31A1 combinatorial treatments (Fig. 9D). IHC staining further demonstrated that SLC31A1 deletion markedly decreased the positive expression of KI-67, CPT1A, and N-Cadherin in tumor sections, and enhanced the inhibitory effect of TAM on these molecules in mice bearing MCF7/TR tumors (Fig. 9D and E). Furthermore, IF staining identified that SLC31A1 depletion induced higher Caspase-3 cleavage levels in tumor tissues, which were further enhanced in these mice receiving TAM plus shSLC31A1 (Fig. 9F). Conversely, in xenograft mouse models bearing MCF7 tumors, we found that SLC31A1 overexpression further increased the tumor growth rates and weights compared to the oe-Vec group (Supplementary fig. 10A-C). The density of tumor cells was significantly enhanced in the oe-SLC31A1 group of mice. In addition, mice bearing tumor tissues overexpressing SLC31A1 exhibited significantly facilitated KI-67, CPT1A, and N-Cadherin expression levels (Supplementary fig. 10D and 6E). These in vivo studies above demonstrated that SLC31A1 could promote breast cancer progression, and its suppression could re-sensitize TAM-resistant breast cancer cells to chemotherapy and restrain tumor growth.

Fig. 9.

SLC31A1 knockdown improves ER-positive breast cancer cells to tamoxifen treatment in vivo. (A) Tumor samples collected from each group were imaged (n = 4 per group). (B) Tumor growth rates were measured (n = 4 per group). (C) Tumor weights were recorded (n = 4 per group). (D) H&E staining, and IHC staining of KI-67, CPT1A and N-Cadherin expression levels in tumor sections were performed (n = 4 per group). Scale bar, 50 μm. (E) Quantification for KI-67, CPT1A and N-Cadherin positive staining was exhibited. (F) IF staining for cleaved Caspase-3 in tumor sections (n = 4 per group). Scale bar, 50 μm. (G) Working model indicating the effects of SLC31A1 on chemoresistance in ER-positive breast cancer by inducing CPT1A-mediated FAO process. Data were presented as means ± SD. *P<0.05, ^⁎⁎P<0.01, ^⁎⁎⁎P<0.001, ^⁎⁎⁎⁎P<0.0001.

Discussion

TAM serves as a selective modulator of the ER and has been utilized for numerous years as an adjuvant medication to treat and prevent breast cancer in patients, particularly those with ERα-positive tumors [4,5]. TAM operates by competitively binding to ERα, thereby suppressing the proliferation of ER-positive breast cancer cells. Additionally, by regulating multiple signaling targets such as protein kinase C, transforming growth factor β (TGF-β), calmodulin, mitogen-activated protein kinase p38 (MAPK-p38), and c-Jun terminal kinase, TAM also induces apoptosis in breast cancer cells through an ERα-independent mechanism [[35], [36], [37]]. Despite the fact that the mortality rate from breast cancer can be significantly reduced with TAM treatment, approximately 50% of patients still exhibit a poor response to the drug and recurrence of cancers that are resistant to TAM therapy [38]. Herein, identifying the molecular mechanisms underlying TAM resistance in ER-positive BC cells is crucial for discovering and developing effective therapeutic strategies to improve the quality of treatment outcomes.

In our present study, we discovered that high expression of SLC31A1, a crucial copper transporter [18,21,22], was essentially correlated with poor survival outcomes for breast cancer patients and associated with metastasis and TAM resistance among ER-positive breast cancer patients undergoing TAM therapy. A positive correlation between SLC31A1 and ERs expression levels was detected in BRCA-Basal and -Luminal patients. Significantly enhanced SLC31A1 expression at both the mRNA and protein levels was observed in several ER-positive breast cancer cell lines in comparison to the ER-negative cell lines. We then discovered that SLC31A1 knockdown significantly enhanced the sensitivity of chemoresistant breast cancer cells in response to TAM treatment, evidenced by the decreased cell proliferation and invasion, while SLC31A1 overexpression accelerated the proliferative and invasive capabilities of wild type breast cancer cells. Mechanistically, FAO mediated by CPT1A might be a potential underlying mechanism through which SLC31A1 exerts its regulatory effect on TAM-resistant breast cancer, accompanied by significantly reduced copper concentrations. Further analysis revealed that TAM-resistant breast cancer cells exhibited stronger CPT1A expression levels, FAO rates, and ATP levels than those of the parental wild-type breast cancer cells. Copper addition dose-dependently upregulated CPT1A expression levels in TAM-resistant breast cancer cells, whereas being eliminated in cells treated with shSLC31A1 or copper chelating agent TTM, accompanied by strongly enhanced apoptosis and cleaved Caspase-3. We also detected high levels of copper in ER^- breast cancer cells. However, shSLC31A1 had no significant effect on the changes in cellular copper content in ER^- breast cancer cells. Given the low expression of SLC31A1 detected in ER^- breast cancer cells, we hypothesized that the regulation of cell proliferative and migratory behaviors through the copper uptake process might be independent of SLC31A1 in ER^- breast cancer cells. CPT1A silencing robustly enhanced the sensitivity of chemoresistant breast cancer cell lines in response to TAM incubation. Furthermore, CPT1A knockdown conspicuously constrained FAO and impaired mitochondria in TAM-resistant breast cancer cells, accompanied by high apoptosis and decreased ATO levels. Significantly, suppressing SLC31A1 substantially downregulated CPT1A expression levels in breast cancer cells with chemoresistance. In contrast, SLC31A1 overexpression potently promoted CPT1A expression, copper concentrations, FAO rates, and ATP levels, which, however, were attenuated upon CPT1A suppression, resulting in the reduced proliferation and invasion of breast cancer cells. Collectively, instead of inhibiting breast cancer cells, reducing copper uptake through SLC31A1 suppression significantly restrained the expression of CPT1A, ultimately restricting the proliferative and invasive abilities of breast cancer cells. Animal studies confirmed that SLC31A1 deficiency could markedly enhance the sensitivity of TAM-resistant breast cancer to TAM treatment partially through repressing CPT1A, as evidenced by the reduced tumor growth rates and weights. Taken together, we concluded that SLC31A1 upregulation enhanced copper transportation into breast cancer cells, which drove CPT1A expression, followed by FAO elevation, thereby facilitating the proliferation, EMT process, and TAM resistance, and suppressing apoptosis in ER-positive breast cancer cells (Fig. 9G).

SLC31A1 is a major copper transporter, which plays crucial roles in modulating intracellular copper homeostasis [18,21,22,25]. Copper is regarded as a double-edged sword. Stable intracellular copper levels are essential for the actions of specific enzymes to trigger pro-angiogenic responses and enhance the metabolism and tumor cell proliferation [39,40]. Copper removal by the oral copper chelating agent TTM suppresses TNBC metastasis by inactivating Complex IV and reducing mitochondrial oxidative adenosine triphosphate production [32]. However, excessive copper can cause cuproptosis. SLC31A1 is recognized as an important molecule that facilitates cuproptosis in kidney and lung cancer cell lines primarily through forcing intracellular free copper accumulation [41,42]. SLC31A1 plays an essential role in regulating tumor progression and is differentially expressed in tumor and normal tissues [43]. As reported, the expression of SLC31A1 was significantly higher in pancreatic cancer and correlated with shorter survival times for the patients. Its expression suppression induced autophagy to restrain pancreatic cancer growth, but did not affect apoptosis [44]. The expression of SLC31A1 is also increased in human glioma samples compared to normal samples, and its high expression is associated with poor outcomes and an immunosuppressive tumor microenvironment; however, SLC31A1 knockdown significantly induces apoptosis in glioma cells to suppress its progression [45]. Considering the different effects of SLC31A1 on apoptosis, we supposed that the regulatory role of SLC31A1 in mediating apoptotic cell death induction might be associated with the types of tumor cells. Increasing studies suggest that SLC31A1 is also accountable for the active transport of platinum drugs in tumor cells [21]. Aberrant SLC31A1 expression in tumor cells can impact the sensitivity of cells to platinum drugs and induce chemoresistance in tumor cells [46]. Additionally, SLC31A1 is implicated in ZNF711-Regulated suppression of cDDP resistance in epithelial ovarian cancer [23]. SLC31A1 is also related to PTBP1-induced chemoresistance in response to cDDP in osteosarcoma [21]. Consistent with previous studies, TCGA data indicated that high SLC31A1 expression was associated with poor overall survival and progression-free survival. We unexpectedly discovered here that SLC31A1 expression levels were significantly up-regulated in clinical ER-positive breast cancer tissues, especially in those with metastasis and recurrence after TAM treatments. Furthermore, higher SLC31A1 was noticed in advanced-stage tumors compared to early-stage tumors. Post-treatment tumors from recurrent patients displayed elevated SLC31A1 expression compared to their pre-treatment counterparts, but we did not observe this tendency in non-recurrent patients. These aforementioned findings demonstrated that SLC31A1 was a marker for ER-positive breast cancer progression and recurrence. In our acquired TAM-resistant cell model established in vitro, we verified that SLC31A1 was highly expressed when compared with the parental breast cancer cell lines. Of note, SLC31A1 knockdown significantly restored TAM sensitivity in chemoresistant ER-positive breast cancer cells, as evidenced by the attenuated proliferative and invasive capabilities of cells, which was accompanied by decreased expression levels of invasion hallmarks, including N-Cadherin, VEGF, and MMP9 [47,48]. Furthermore, in copper-overloaded ER⁺ breast cancer cells with chemoresistance, SLC31A1 depletion significantly induced apoptosis and increased cleaved Caspase-3 expression levels, contributing to the inhibition of breast cancer. Animal studies confirmed that stable SLC31A1 knockdown not only restricted tumor growth but also enhanced the sensitivity of TAM-resistant breast cancer to TAM treatment. In contrast, overexpression of SLC31A1 accelerated the proliferation and invasion of breast cancer cells, which was validated by the xenograft mouse model, as indicated by the increased tumor volume and weights. At this point, we initially concluded that SLC31A1 was a key driver promoting the development of TAM resistance and played a predominantly pro-carcinogenic role in ER-positive breast cancer.

Through RNA-Seq analysis, we discovered that FAO mediated by CPT1A might be a potential mechanism by which SLC31A1 regulated the progression of ER-positive breast cancer, as indicated by the reduced CPT1A expression levels and the enrichment of fatty acid metabolism and β-oxidation by DEGs. An increasing number of studies have reported that fatty acid metabolism provides a significant amount of energy in tumor cells. De novo synthesis of fatty acids promotes membrane synthesis, which contributes to cell proliferation. Meanwhile, fatty acid catabolism through the FAO process generates more ATP to enable the survival of tumor cells [49]. Aberrant FAO activation participates in stemness maintenance, cell growth, metastasis initiation, chemoresistance development, and immune response evasion, thereby facilitating tumor progression [[50], [51], [52]]. FAO inhibition can impair NADPH generation and elevate reactive oxygen species (ROS), leading to ATP reduction and cell death in various types of tumor cells, including glioma and breast cancer [53,54]. Additionally, FAO inhibition was considered to induce apoptosis to subsequently restrain the tumorigenesis and chemoresistance events [55,56]. CPT1A is a key rate-limiting enzyme for FAO, and overexpressed in various cancers [57]. Targeting CPT1A expression has been identified as a promising therapeutic strategy for suppressing the FAO rate to enhance tumor therapeutic efficacy [58,59]. Recent studies also reported that pharmacological suppression of CPT1A and FAO exerted inhibitory effects on ER-positive breast cancer progression and cell proliferation [60]. In our present study, we found higher CPT1A expression levels in TAM-resistant breast cancer cells compared to the parental wild-type cells, accompanied by enhanced FAO and ATP contents. Notably, CPT1A Knockdown markedly enhanced the sensitivity of chemoresistant breast cancer cells to TAM treatments, as indicated by the significantly reduced cell viability. Additionally, FAO oxidation and mitochondrial integrity were impaired in chemoresistant breast cancer cells upon CPT1A depletion, which was accompanied by increased apoptosis and cleaved Caspase-3 protein expression levels. However, CPT1A overexpression along with aggravated FAO process and ATP levels was observed in parental breast cancer cells, revealing an oncogenic role of CPT1A-Regulated FAO in the breast cancer context.

Copper is a unique transition metal and is necessary for a variety of important biological processes, including redox balance, mitochondrial function and FAO [[6], [7], [8],61,62]. Recently, copper uptake was shown to promote FAO and mitochondrial biogenesis, contributing to the improvement of fatty liver disease [63]. Conventional copper chelators including TTM are reported to be anti-angiogenesis agents through mediating DNA damage repair, apoptosis and metastasis processes [6,[32], [33], [34],[64], [65], [66]]. Copper depletion was also shown to result in an elevation of oxidative stress and mitochondrial membrane rupture, which led to the cell death of TNBC cells [67]. Here in our study, we found that copper addition significantly promoted CPT1A expression levels in chemoresistant breast cancer cells via a dose-dependent manner, but were diminished upon SLC31A1 deficiency, CPT1A inhibition and copper chelator TTM treatment, which were accompanied by significant apoptotic cell death. These results elucidated that copper uptake status could influence CPT1A expression, and importantly such process was strongly influenced by SLC31A1 in TAM-resistant breast cancer cells. We examined the influence of TTM single treatment on CPT1A expression changes in chemoresistant ER⁺ breast cancer cells. TTM-incubated T47D/TR and MCF7/TR cell lines exhibited lower copper contents, CPT1A expression levels, FAO rates and ATP levels. These results on the one provided further evidence that disruption of copper accessibility by TTM indeed exerted anti-cancer effects in ER⁺ breast cancer cells with chemoresistance via interrupting FAO process and CPT1A signaling. Notably, consistent with the suppression of CPT1A expression and FAO rates, TTM alone treatment dose-dependently reduced the copper concentrations in chemoresistant breast cancer cells, which were similarly with previous studies [68,69]. As for this, we speculated that TTM might inhibit the absorption of copper ions by cells through competitively binding copper ions and then preventing copper ions from entering the cell. This further corroborated and explained that inducing copper deprivation by TTM or other methods such as SLC31A1 knockdown may exert anti-tumor potential in chemoresistant breast cancer cells. Growing evidence has identified that TTM could restrain tumor growth partially via reducing copper ion absorption [44,70,71]. We then found that the enhanced proliferation and invasion of ER-positive breast cancer cells induced by SLC31A1 overexpression were significantly ameliorated upon CPT1A expression suppression, accompanied by limited FAO rates and ATP levels. Considering the crucial role of SLC31A1 in regulating cuproptosis, we also examined the expression levels of several cuproptosis-related markers, including DBT, DLAT, PDHA1, DLD, and FDX1 [72,73]; however, no significant difference was detected in the expression changes of these cuproptosis hallmarks in both chemo-resistant and -sensitive ER-positive breast cancer cell lines. Therefore, we hypothesized that the suppressive effect of SLC31A1 on breast cancer progression and chemoresistance might be independent of cuproptosis, but rather through inhibiting the CPT1A-mediated FAO process. Nevertheless, further exploration is still required to investigate the role of SLC31A1/copper signaling in regulating CPT1A and other signaling pathways in more detail.

In summary, our findings uncovered that SLC31A1/copper axis was a significant drug-resistant factor, which could promote the proliferation, EMT and chemoresistance but restrain apoptosis in ER-positive breast cancer through promoting the expression of CPT1A and associated FAO process (Fig. 9G). Targeting SLC31A1 for its inhibition would be a potential strategy to overcome chemoresistance and improve TAM sensitivity, thereby improving the clinical therapeutic outcomes of ER-positive breast cancer patients.

CRediT authorship contribution statement

Fei Pan: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization

Declaration of competing interest

The authors declare that they have no competing interests.