The role of ATP binding cassette (ABC) transporters in breast cancer: Evaluating prognosis, predicting immunity, and guiding treatment

ABSTRACT

Breast cancer is currently the most prevalent form of cancer worldwide. Nevertheless, there remains limited clarity regarding our understanding of the tumor microenvironment and metabolic characteristics associated with it. ATP-binding cassette (ABC) transporters are the predominant transmembrane transporters found in organisms. Therefore, it is essential to investigate the role of ABC transporters in breast cancer. Transcriptome data from breast cancer patients were downloaded from the TCGA database. ABC transporter-related genes were obtained from the Genecards database. By LASSO regression, ABC-associated prognostic signature was constructed in breast cancer. Subsequently, immune microenvironment analysis was performed. Finally, cell experiments were performed to verify the function of ABCB7 in the breast cancer cell lines MDA-MB-231 and MCF-7. Using the ABC transporter-associated signature, we calculated a risk score for each breast cancer patient. Patients with breast cancer were subsequently categorized into high-risk and low-risk groups, utilizing the median risk score as the threshold. Notably, patients in the high-risk group exhibited significantly worse prognosis (P<0.05). Additionally, differences were observed in terms of immune cell infiltration levels, immune correlations, and gene expression of immune checkpoints between the two groups. Functional experiments conducted on breast cancer cell lines MDA-MB-231 and MCF-7 demonstrated that ABCB7 knockdown significantly diminished cell activity, proliferation, invasion, and migration. These findings emphasize the significance of understanding ABC transporter-mediated metabolic and transport characteristics in breast cancer, offering promising directions for further research and potential therapeutic interventions.

KEYWORDS:

• Breast cancer
• ATP-binding cassette
• cancer metabolism
• signature
• immune

Introduction

Breast cancer has become one of the most common types of cancer in the world, seriously harming women’s physical and mental health [1,2]. It is estimated that more than 2.3 million new cases of breast cancer be diagnosed in 2020, placing a heavy burden on global health [3–5]. Although current clinical screening allows many breast cancer patients to be diagnosed at an early stage, a large proportion of patients in less developed regions are still diagnosed late [6–8]. In addition, for patients with ER(-), PR(-) and HER(-) breast cancer (triple negative breast cancer), the current treatment is still poor [9]. Although immunotherapy, which has achieved significant benefits in other solid tumors, has been initially used in the treatment of breast cancer, the occurrence of drug resistance and low response to treatment pose challenges for the treatment of breast cancer [10–13]. Therefore, it is necessary to deeply explore the tumor microenvironment of breast cancer, provide new biomarkers for its treatment, and provide experimental evidence for the study of drug resistance.

ATP binding cassette (ABC) transporters are the most common transmembrane transporters in organisms, mediating transmembrane transport of many substances, including ions, vitamins, sugars, drugs, etc [14,15]. Unlike Solute carrier (SLC) transporters, ABC transporters mediate transmembrane transport by ATP directly [16]. Multiple members of the ABC transporter are thought to be involved in multidrug resistance in cancer, particularly the ABCB, ABCC, and ABCG subgroups [17]. They are associated with cancer resistance to a variety of drugs, including doxorubicin, etoposide and vincristine, leading to poor treatment response and even death in many patients [18]. However, despite the challenges faced by clinical inhibitors targeting these receptors, it is imperative to delve into the role of ABC transporters in breast cancer to uncover potential alternative strategies or novel insights for therapeutic intervention.

In this study, we performed transcriptome analysis of breast cancer patient data from the TCGA database. The ABC-related prognostic signature of breast cancer was constructed. Subsequently, survival analysis and immune microenvironment analysis were carried out. Our study can provide reference for exploring the role of ABC transporters in breast cancer and provide novel prognostic biomarkers for breast cancer.

Methods

Transcriptome data download

Transcriptome data from breast cancer patients were downloaded from the TCGA database(https://www.tcga.org). A total of 802 breast cancer patients were enrolled in the study by excluding those with no clinical information. The RNA sequencing data of genes were transformed by log2 for subsequent analysis.

Acquisition of genes associated with ABC transporter

In order to study the correlation between different genes and ABC transporters, and to extract the gene sets highly correlated with ABC transporters, we conducted correlation analysis in the Genecards database. Genecards database is a web-based tool containing genomic, transcriptome, proteome, genetic, clinical and functional information from 150 databases. We used the genecards website to download the ABC transporter-related genes. Input “ABC transporter” as the keyword yielded 1653 results, and the first 100 genes were selected as ABC transporter-related genes.

Construction of the prognostic signature

After obtaining the gene set associated with ABC transporters, it is necessary to construct a prognostic signature in order to further explore their prognostic value and guide patient prognosis assessment. The Least Absolute And Selection Operator (LASSO) regressions were performed on the above 100 genes obtained from genecards database. By constructing a penalty function and compressing some regression coefficients, the optimal prognostic signature was obtained. A risk score can be calculated for each breast cancer patient. And with the median risk score value as a cutoff, breast cancer patients can be divided into high-risk and low-risk groups. Survival analysis was performed between the two groups using the “survival” R package. The “pROC” package was used to build signature’s ROC curve to evaluate the model’s accuracy.

Evaluation of the signature

After this prognostic signature was constructed, further evaluation of its clinical value and accuracy is warranted. Multivariate COX regression was performed to evaluate the independent prognostic value of each gene in the signature. The “survival” R package and the “forestPlot” package were used for multivariate COX regression. The “RMS” R package was used to construct a Nomogram to assess patient survival.

Immune cell infiltration analysis

The immune microenvironment plays an important role in influencing tumorigenesis, progression and prognosis. Therefore, after building this prognostic signature related to ABC transporters, it is necessary to explore its significance in the immune microenvironment. RNAseq data (level3) and corresponding clinical information of breast cancer were obtained from the cancer genome atlas (TCGA) dataset (https://portal.gdc.com). To provide a reliable immune score assessment, we used Immunedeconv, an R software package that integrates six of the latest algorithms, including TIMER, xCell, McP-counter, CIBERSORT, EPIC and quanTIseq. We used “CIBERSORT” to calculate the levels of immune cell infiltration in the high-risk and low-risk groups.

Correlation analysis of immune score

Further, we conducted an analysis of tumor immune scores to investigate variations in immune scores among different risk groups. RNAseq data (level3) and corresponding clinical information of breast cancer were obtained from the cancer genome atlas (TCGA) database (https://portal.gdc.com). The single correlation map was realized by R software package GGStatsPlot, and the multi-gene correlation map was displayed by R software package PheatMap. Spearman’s correlation analysis was used to describe the correlation between quantitative variables without normal distribution. A P value of less than 0.05 was considered statistically significant. All the above analysis methods were implemented by R V4.0.3.

Expression analysis of immune checkpoint related genes

Currently, aberrant activity of immune checkpoint genes has been widely recognized as a key mechanism in tumorigenesis and disease progression, and corresponding inhibitory therapy represents one of the critical approaches to enhance tumor prognosis. Therefore, it is necessary to identify the expression of immune checkpoint related genes between different risk groups. RNA-sequencing expression (level 3) profiles and corresponding clinical information for breast cancer were downloaded from the TCGA dataset(https://portal.gdc.com). SIGLEC15,TIGIT,CD274,HAVCR2,PDCD1,CTLA4,LAG3 and PDCD1LG2 were selected to be immune-checkpoint – relevant transcripts and the expression values of these eight genes were extracted. All the above analysis methods and R package were implemented by R foundation for statistical computing (2020) version 4.0.3. Using the ggplot2 R package and pheatmap R package.

Cell culture and transfection

Breast cancer cell lines MDA-MB-231 and MCF-7 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were grown on DMEM supplemented with 10% fetal bovine serum (Gibco). Lipofectamine 3000 (Thermo Fisher Scientific, Waltham, MA, USA) was used with a small interfering RNA of a previously synthesized targeted gene ABCB7 (GenePharma Inc., Shanghai, China) transfected cells. The siRNA sequence of ABCB7 gene is as follows :si-ABCB7–1: CCTCATAGTATCTATTCAGAA; si-ABCB7–2: GCACAGAGATATGATGGATTT.

CCK-8 assay

Cell proliferative activity can be measured using the CCK-8 reagent. The breast cancer cells with ABCB7 knocked down and the cells in the control group were implanted in 96-well plates, respectively. 24 hours later, CCK-8 reagent was added, and the absorbance was measured after incubation for 2 h in the dark. The proliferation of the cells in the two groups was compared.

Edu assay

Edu is a cell proliferation marker that detects DNA synthesis during the S phase of the cell cycle. The breast cancer cells with ABCB7 knocked down and the cells of the control group were seeded in 96-well plates. Edu reagent was added 24 hours later and incubated for 2 h. Staining and microscopic observation were performed to compare the DNA synthesis of the cells in the two groups.

Scratch tests

Scratch tests can assess the ability of cells to migrate. A straight line was drawn on the confluent monolayer of breast cancer cells with low ABCB7 expression and the cells in the control group using a 200ul sterile pipette tip, and cell migration was observed 48 hours later.

Transwell assay

The Transwell assay assesses cell migration and invasion. Breast cancer cells with ABCB7 knocked down and cells of the control group were added into the upper compartment respectively. Transwell plates with holes were separated between the upper compartment and the lower compartment. The invasion of the lower compartment cells was detected after incubation for 24 hours with a 20% serum medium.

Results

Construction of the prognostic signature associated with ABC transporters

First, the prognostic signatures related to ABC transporters were constructed by Lasso regression, and we found that the final signature was composed of 4 genes (Figure 1a,b). Riskscore=(−0.0735)*ABCA10+(−1e-04)*TAP2+(0.3215)*ABCB7+(0.2097)*ABCC2. A risk score was calculated for each patient, and their risk score distribution, clinical survival status, and gene expression heat map were shown in Figure 1c. Based on the median risk score value as the cutoff value, TCGA breast cancer patients could be divided into high-risk and low-risk groups. Survival analysis showed a significantly poorer prognosis in the high-risk group (Figure 1d, HR = 1.579, P = 0.0237). By calculating the area under ROC curve (AUC) of the signature, it can be seen that the AUC values of the signature in 1, 3 and 5 years are 0.671, 0.633 and 0.662, respectively(Figure 1e). This proves that the signature has some accuracy in assessing the prognosis of breast cancer patients at 1, 3, and 5 years.

Figure 1. Construction of the prognostic signature associated with ABC transporters. (a, b) LASSO regression. Through Lasso regression, the first 100 genes associated with ABC transporters were screened to narrow down the range of variables and obtain the corresponding coefficients. It was found that signature was constructed when the number of genes was 4. In other words, the patient’s prognosis can be evaluated by the signature of these four genes. (C)risk score, survival status and model gene expression of breast cancer patients in TCGA database. (D)the prognosis was significantly worse in the high-risk group (P = 0.0237). (E)the ROC curve of this signature.

Evaluation of the signature

To evaluate the independent prognostic value of each gene in this signature, multivariate COX regression was performed(Figure 2a). The results showed that all four genes were independent prognostic indicators of breast cancer, and the HR values of TAP2 and ABCA10 were less than 1, indicating that these two genes were better prognostic indicators of breast cancer.And HR values of ABCC2 and ABCB7 were greater than 1 and they are indicators of poor prognosis of breast cancer. Then we construct a nomogram based on this signature, and the score calculated by this nomogram can be used to evaluate the survival rate of breast cancer patients in 1, 2, 3 and 5 years(Figure 2b).

Figure 2. Evaluation of the model. (a)multivariate regression showed that the four model genes were independent prognostic indicators of breast cancer. (b)nomogram based on this signature. With four genes in signature, the patient’s 1, 3, and 5-year survival rate can be predicted. The nomogram showed that this signature has a C-index of 0.656.

Immune cell infiltration analysis

Immune microenvironment plays an important role in the pathogenesis of breast cancer, and in-depth analysis of it can provide reference for immunotherapy. As shown in Figure 3(a,b), many immune cells showed different levels of infiltration between the high-risk group (G1) and the low-risk group (G2). These differential infiltrated immune cells include B cell naive, T cell CD8+, T cell CD4+ naive, T cell CD4+ memory resting, T cell CD4+ memory activated, T cell Tregs, Macrophage M1, Macrophage M2, and Neutrophil (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 3. Immune cell infiltration analysis. (a,b)These differential infiltrated immune cells include B cell naive, T cell CD8+, T cell CD4+ naive, T cell CD4+ memory resting, T cell CD4+ memory activated, T cell tregs, Macrophage M1, Macrophage M2, and neutrophil (*P<.05, **P<.01, ***P<.001).

Correlation analysis of immune score

Spearman test was used to analyze the correlation between risk score and immune cell score. The results showed that risk score was significantly correlated with B cell, T cell CD4+, T cell CD8+, Endothelial cell expression, macrophages and NK cells(Figure 4). The risk score was positively correlated with CD4+T cells and CD8+T cells, and negatively correlated with B cells, endothelial cells, macrophages and NK cells.

Figure 4. Correlation analysis of immune score. risk score was significantly correlated with B cell, T cell CD4+, T cell CD8+, endothelial cell expression, macrophages and NK cells.

Expression of immune checkpoint related genes in high-risk and low-risk groups

Immune checkpoint plays an important role in immunotherapy. Exploring the expression of genes related to immune checkpoint can provide reference for immunotherapy. Our analysis revealed a higher expression trend of immune checkpoint related genes in the high-risk group (G1), including CD274, CTLA4, HAVCR2, LAG3, PDCD1, PDCD1LG2, TIGIT, and SIGLEC15 (Figure 5, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 5. Expression of immune checkpoint related genes in high-risk and low-risk groups. the expression trend of immune checkpoint related genes was higher in the high-risk group. *P<.05, **P<.01, ***P<.001.

Cell experiments to verify the function of ABCB7 in breast cancer cell lines

We further selected cell experiments to verify the function of ABCB7. These experiments were performed in the MDA-MB-231 and MCF-7 cell lines. First, EdU experiments showed that the activity of breast cancer decreased significantly after ABCB7 knockdown (Figures 6a,b). The same result was found in the CCK-8 experiment(Figures 6c,d). Scratch experiments showed that the migration ability of breast cancer cell lines decreased significantly after ABCB7 knockdown (Figures 6e,f). Transwell experiment showed that the migration and invasion ability of breast cancer cell lines decreased significantly after ABCB7 knockdown (Figures 6g,h).

Figure 6. Cell experiments to verify the function of ABCB7 in breast cancer cell lines. (a, b)EdU experiments showed that the activity of breast cancer decreased significantly after ABCB7 knockdown. (c, d)CCK8 assay of MDA-MB-231 cell line and MCF-7 cell line. Result showed that the activity of breast cancer decreased significantly after ABCB7 knockdown. (e, f)Scratch experiments. The migration ability of breast cancer cell lines decreased significantly after ABCB7 knockdown. (g, h)Transwell experiment. The migration and invasion ability of breast cancer cell lines decreased significantly after ABCB7 knockdown.

Discussion

Breast cancer is a highly heterogeneous tumor type [19]. At present, the classification is mainly based on ER, PR, HER2 receptor status, Ki-67 protein histochemical staining and BRCA gene mutation [20–22]. The existing clinical classification has greatly prompted the development of breast cancer treatment. Mammography, MRI and other imaging examinations have promoted the early detection of breast cancer [23]. This has led to a gradual decline in the death rate from breast cancer [24]. However, the death rate from breast cancer remains high in countries with poor health care. Moreover, the response to triple negative breast cancer is still far from ideal [25]. Therefore, it is necessary to explore the tumor microenvironment of breast cancer to identify novel biomarkers.

Currently, cancer metabolism is regarded as a promising direction in cancer therapy [26]. The research on cancer metabolism has a long history, dating back to the early 20th century, when Otto Warburg discovered the active aerobic glycolysis process in cancer cells, which started the research on cancer metabolism [27–29]. Up to now, changes in cancer metabolism have been considered as one of the important features of cancer, and have been applied to the diagnosis and treatment of cancer, such as PET/CT based on cancer glucose metabolism, and tumor chemotherapy drugs targeting cancer metabolism [30]. ATP binding cassette (ABC) transporters are the largest membrane transporters family in organisms [31]. In the human body, they are not only involved in the regulation of normal homeostasis, but also involved in the pathophysiological processes of many diseases [32]. Because they transport so many substrates, including ions, peptides, amino acids, vitamins, etc., their role in cancer metabolism is thought to be critical [33]. In particular, ABC transporters are thought to play an important role in multidrug resistance [34]. Multidrug resistance is the cause of death in many cancer patients, and understanding the mechanism of multidrug resistance can improve patient survival and treatment effectiveness [35]. However, previous attempts to target the ABC transporter have failed. To address this issue, it is necessary to continue to understand the significance of ABC transporters in the cancer microenvironment.

In this study, we performed an extensive bioinformatics analysis that revealed the significance of ABC transporters in breast cancer. By Lasso regression, the ABC transporter-associated prognostic signature was constructed in breast cancer. This model allows the risk of each breast cancer patient to be quantified and prognostic assessment to be carried out accurately. Among them, the prognosis of breast cancer patients in the high-risk group is significantly worse. For these patients, it is necessary to carry out early intervention. In addition, we also explored the value of the signature in the microenvironment of breast cancer. Significant differences in levels of immune cell infiltration were observed between the high-risk and low-risk groups, which may account for the difference in prognosis between the two groups. We then explored the expression of immune checkpoints between high-risk and low-risk groups. Higher levels of immune checkpoint gene expression were observed in the high-risk group, which may contribute to immune escape in the high-risk group and may be a factor in their poor prognosis. It also suggests that the high-risk group may be more likely to benefit from treatment with immune checkpoint inhibitors.

Immunotherapy is a landmark development in the history of cancer therapy and has led to improved treatment and survival in many tumor types [36]. However, the benefits of immunotherapy in breast cancer are unclear [37]. Breast cancer is also known as a “cold” tumor [38]. It is necessary to explore the tumor microenvironment of breast cancer to identify differences in levels of immune cell infiltration and to identify the expression of immune checkpoint genes [39]. Our study analyzed the immune microenvironment of high-risk and low-risk breast cancer patients, which is valuable for us to carry out immunotherapy based on this signature.

However, there are some limitations to our study. We lack in vivo experiments to confirm our conclusions. We will improve it in the future.

Conclusions

We explored the significance of ABC transporters in breast cancer by bioinformatics analysis. This ABC transporter-related signature can play a certain role in evaluating the prognosis and immune level of breast cancer patients.

Abbreviations

ATP-binding cassette (ABC); Least Absolute And Selection Operator (LASSO); the cancer genome atlas (TCGA);

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available in TCGA database[https://www.tcga.org].

Additional information

Funding

This study was supported by Talents Training Program of the Seventh People´s Hospital Shanghai University of Traditional Chinese Medicine (Grant No.XX2021-05, Grant No. QMX2022-05) and Health Committee of Pudong New Area (PWGw2020-02) and Pudong New Area Science and Technology Development Fund (grant no.PKJ2023- Y02).