Inhibiting glycolysis facilitated checkpoint blockade therapy for triple-negative breast cancer

Abstract

Cancer cells are characterized by their altered energy metabolism. A hallmark of cancer metabolism is aerobic glycolysis, also called the Warburg effect. Hexokinase 2 (HK2), a crucial glycolytic enzyme converting glucose to glucose-6-phosphate, has been identified as a central player in the Warburg effect. Deletion of HK2 decreases cancer cell proliferation in animal models without explicit side effects, suggesting that targeting HK2 is a promising strategy for cancer therapy. In this study, we discovered a correlation between HK2 and the tumor immune response in triple-negative breast cancer. Inhibition of HK2 led to a reduction in G-CSF expression in 4T1 cells and a decrease in the development of myeloid-derived suppressor cells which, in turn, enhanced T cell immunity and prolonged the survival of 4T1 tumor-bearing mice. Furthermore, the HK2 inhibitor 3-BrPA improved the therapeutic efficacy of anti-PD-L1 therapy in 4T1 tumor-bearing mouse models. This study highlights the potential of glycolysis-targeting interventions as a novel treatment strategy, which can be combined with immunotherapy for the treatment of triple-negative breast cancer.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-025-02320-w.

Introduction

Despite significant progress in breast cancer treatment over the past few decades, morbidity and mortality of breast cancer remain substantial [1]. Conventional chemotherapy agents such as vinorelbine, paclitaxel, and anthracyclines, while effective in inducing cancer cell apoptosis, also inflict considerable damage on normal cells, leading to severe side effects [2]. Therefore, it is imperative to explore alternative and innovative therapeutic strategies for comprehensive breast cancer treatment.

Altered energy metabolism stands as a recognized hallmark of cancer cells [3, 4]. Aerobic glycolysis, also known as the Warburg effect, is a prominent aspect of glucose metabolism in cancer cells. It describes the phenomenon where glucose is preferentially metabolized through fermentation rather than oxidative phosphorylation, characterized by elevated rates of lactate production and glucose uptake, even in the presence of oxygen [5, 6]. Hexokinase 2 (HK2) is a key enzyme in glycolysis, significantly involved in the metabolic reprogramming of cancer cells through the Warburg effect. HK2 is overexpressed in various cancers, including TNBC, and is associated with increased tumor growth, decreased survival, and therapy resistance [7–9]. This upregulation of HK2 in TNBC can contribute to the rapid growth and proliferation of tumor cells. HK2-mediated metabolic reprogramming in TNBC also affects the tumor microenvironment (TME) [10]. While the metabolic functions of HK2 are well-studied, its role in the tumor immune microenvironment, particularly in TNBC, remains underexplored. Therefore, gaining insights into the role of HK2 in regulating the tumor immune microenvironment is of particular importance in identifying novel therapeutic interventions.

The tumor immune microenvironment is a decisive factor in the success of immunotherapy, with a 'hot' immune microenvironment, characterized by a high level of immune cell infiltration and active anti-tumor immune responses, being conducive to positive responses to tumor immunotherapy [11–14]. Tumors manipulate the patient's immune system, converting immune cells within the TME from cancer-killing agents to protectors of the tumor, thereby resisting the immune system and immunotherapy. Tumor glycolysis has been shown to reshape the tumor immune microenvironment by influencing multiple immune cells, enabling cancer cells to evade effective anti-tumor immunity through various mechanisms [15, 16]. In triple-negative breast cancer (TNBC), the number of regulatory T cells (Tregs) is significantly higher than that in other breast cancer subtypes. Moreover, the proportion of exhausted CD8 + T cells increases significantly, resulting in impaired cytotoxic ability and making it difficult for these cells to effectively kill tumor cells. And the expression level of programmed death-ligand 1 (PD-L1) is higher than that in other breast cancer subtypes. The binding of PD-L1 on tumor cells to programmed cell death protein 1 (PD-1) on immune cells inhibits the activation of tumor-infiltrating lymphocytes (TILs) and induces tumor cell immune escape [17]. Myeloid-derived suppressor cells (MDSCs) are central to the immunosuppressive network within the TME, protecting the tumor from the patient’s immune response and limiting the efficacy of immunotherapy. MDSCs promote the generation of regulatory T cells (Tregs), which are pivotal for immune tolerance development, and tend to differentiate into tumor-associated macrophages (TAMs) and facilitate fibroblast differentiation into cancer-associated fibroblasts (CAFs) [18–22]. Patients harboring MDSCs face a nearly twofold increased risk of succumbing to cancer [23–25].

However, the link between glycolysis inhibition and immunotherapy outcomes in breast cancer remains largely unexplored. The objective of this study was to investigate the impact of HK2 intervention on the MDSCs development and immunotherapy in triple-negative breast cancer. Our research provides a comprehensive understanding that can potentially enhance therapeutic strategies for triple-negative breast cancer.

Materials and methods

Cell culture and reagents

The mouse triple-negative breast cancer cell line 4T1 and human triple-negative breast cancer cell line MDA-MB-231 were obtained from the National Collection of Authenticated Cell Cultures of China. GM-CSF (10 ng/mL) was added to media for MDSC culture. These cells were maintained in RPMI 1640 medium (Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco), and 1% penicillin/streptomycin, within a controlled environment of 5% CO2, 95% humidity, and at a temperature of 37 °C.

Silencing HK2 expression using short hairpin RNA (shRNA)

To shut down HK2 expression in cell lines, siRNA technology was carried out. Lentiviruses containing shRNA-HK2 were engineered by the Shanghai Genechem Co., Ltd. (Shanghai, China). Each virus, shHK2 and the control virus, with a titer of 1 × 10^9 was packaged separately. To maintain their stability, they were stored at − 80 °C to prevent repetitive freeze–thaw cycles. Before the actual infection process, a series of experiments were conducted in 96-well plates to determine the optimal virus dosage, following the provided instructions. Mouse triple-negative breast cancer cell line 4T1 cells in the logarithmic growth phase were selected for infection under specified conditions, and cultured for 12–16 h. Subsequently, the cells were incubated in culture medium containing 10% FBS. To establish stably transfected cell lines, cells were subjected to puromycin treatment at a concentration of 2 μg/ml. Stably transfected cell lines, identified by the presence of green fluorescent protein signals, were chosen for further experiments. The effectiveness of shHK2 inhibition was confirmed through western blot analysis.

HK2 siRNA design, synthesis and transfection

To shut down HK2 expression in cell lines, siRNA technology was used. HK2 siRNA and control siRNA were designed and synthesized by Gemma Biotechnology Co. The sequences are as follows: HK2 siRNA: CCUGCAACACACUUUAGGGCUU, control siRNA: TTCTCCGAACGTGTCACGT. Human triple-negative breast cancer cell line MDA-MB-231 in logarithmic growth phase was trypsinized, resuspended in complete medium, and seeded into six-well plates at a density of 2 × 10^6 cells per well. Transfection with HK2 siRNA was performed when cell confluence reached 70–90%.

To prepare the transfection complex, 20 pmol of HK2 siRNA was diluted in 50 μl of serum-free DMEM/F-12 medium and gently mixed. Separately, Lipofectamine™ 2000 reagent was diluted by mixing 1 μl of the reagent with 50 μl of serum-free DMEM/F-12 medium, followed by gentle mixing and incubation at room temperature for 5 min. The diluted HK2 siRNA and Lipofectamine™ 2000 reagent were then combined, gently mixed, and incubated at room temperature for 20 min to form the HK2 siRNA/Lipofectamine™ complex. Subsequently, 100 μl of the HK2 siRNA/Lipofectamine™ complex was added to each well of the culture plate containing cells and medium. The cells were then incubated at 37 °C in a CO2 incubator for 24–48 h for subsequent experiments.

In vivo murine 4T1 breast tumor model

To explore the role of HK2 in immune response and immunotherapy, the 4T1 breast cancer model was established. The in vivo assays were approved by the Ethics Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. For the shHK2 4T1 cell-bearing mice model, shHK2 or vector control 4T1 cells (1 × 10^6) were mixed with 200 μl of phosphate-buffered saline (PBS), 4 °C. The cells were subcutaneously inoculated into the right flank of BALB/c nude mice (purchased from the Experimental Animal Center at Huazhong University of Science and Technology). The body weight and tumor volume were measured every 2 days, and the tumor volume was calculated using a simple formula [(length × width × width)/2]. Mice were euthanized on reaching an endpoint as per the Institutional Animal Care and Use Committee (IACUC) guidelines after 23days. For the breast cancer syngeneic model, 1 × 10^6 4T1 mouse breast cancer cells were subcutaneously injected into the right flank of BALB/c mice (purchased from the Experimental Animal Center at Huazhong University of Science and Technology). Following the injection, the mice were closely monitored for 7 days before being randomly divided into four groups, each consisting of five animals. Drug therapy commenced on day 7 with the following treatment regimens: (1) Control Group: Saline, 10 mL/kg, administered daily. (2) 3-BrPA-Only Group: 3-BrPA, 10 mg/kg, administered three times a week. (3) Anti-PDL1-Only Group: PD-L1 antibody was purchased from BioXCell and administered via intraperitoneal injection at a dosage of 5 mg/kg, administered three times a week. (4) Co-Treatment Group: 3-BrPA, 5 mg/kg, and PD-L1 antibody, 10 mg/kg, both administered three times a week. All injections were freshly prepared and delivered intraperitoneally (IP). Tumor volumes were calculated using the formula (length × width × width)/2. Body weight and tumor size were monitored every 2 days throughout the experimental period. After 19 days of treatment, mice were euthanized on reaching an endpoint as per the Institutional Animal Care and Use Committee (IACUC) guidelines. Immediately upon euthanasia, tumors were excised and weighed. Sections of the tissues were promptly collected for flow cytometry analysis, while the remaining portions were snap-frozen in liquid nitrogen and prepared for subsequent experiments.

Western blotting analysis

To detect the protein levels of HK2 and G-CSF, immunoblot assays were used. Harvested cellular samples were washed three times with precooled PBS. Total protein was extracted using RIPA lysis buffer (#R0278, Sigma), and its concentration was quantified with a BCA protein assay kit (#P0012, Beyotime). Extracted proteins were separated by SDS-PAGE and electrotransferred to a PVDF membrane (Millipore). The PVDF membrane was blocked with 5% skim milk for 1 h at room temperature, then incubated overnight at 4 °C with primary antibodies: anti-HK2 (Abcam, ab209847, 1:1000), anti-G-CSF (Proteintech, 17185-1-AP, 1:1000), and anti-β-actin (Protinetech, 66009-1-Ig, 1:1000). Next, it was incubated with HRP-conjugated goat anti-rabbit secondary antibody (#SA00001-15, Proteintech, 1:3000) for 1 h at room temperature. Then the membrane was visualized using ECL reagents (#6883, Cell Signaling Technology) with the Invitrogen iBright CL1000 imaging system (Thermo Fisher Scientific).

Quantitative reverse transcription PCR (qRT-PCR)

To detect the mRNA expression levels of HK2 and G-CSF, quantitative reverse transcription PCR was performed. Total RNA was isolated with TRIzol reagent (Life Technologies) and reverse-transcribed into cDNA using PrimeScript RT Master Mix (#RR036A, Takara). Diluted cDNA from each sample was used for qRT-PCR analysis with TB Green Premix Ex Taq II (#RR820A, Takara) on a StepOne Plus Real-Time PCR System (Thermo Fisher Scientific) to quantify target gene mRNA levels. β-actin served as the endogenous control for normalization, and relative gene expression fold-change was calculated by the 2 − ΔΔCt method. PCR primer sequences were designed based on NCBI database sequences: β-actin: F 5'-CTGGAACGGTGAAGGTGAC-3', R 5'-AAGGGACTTCCTGTAACAATGCA-3';

HK2: F 5'-CAAAGTGACAGTGGGTGTGG-3', R 5'-GCCAGGTCCTTCACTGTCTC-3';

Mouse G-CSF: F 5'-CCTGGATGATAGAACCTAACGGG-3', R 5'-CTCTCCAGCGAAGGTGTAGACA-3'.

Enzyme-linked immunosorbent assay (ELISA) for mouse G-CSF

To detect G-CSF released by tumor cells, Enzyme-linked immunosorbent assay was conducted. G-CSF ELISA was carried out using a commercial kit (R&D Systems, MCS00) following the manufacturer's protocol. 100 μL of cytokine standard or coculture supernatant was added to the precoated ELISA plate wells. Then, 50 μL of biotinylated detection antibody was added to each well, and the plate was incubated at 37 °C for 90 min. After washing the wells, 100 μL of streptavidin-HRP conjugate was added per well and incubated at 37 °C for 30 min. The wells were washed again, and 100 μL of TMB substrate solution was added. After a 15-min incubation at 37 °C, the reaction was stopped with stop solution, and absorbance at 450 nm was measured using a microplate reader.

Flow cytometry

To measure the immune cell number and activity, flow cytometry was performed. Single-cell suspensions from various tumor tissues were prepared and washed with FACS buffer (0.5% BSA and 0.05% sodium azide in PBS). MDSCs were generated from bone-marrow cells co-cultured with G-CSF or supernatants of scrambled and shHK2-transfected 4T1 cells. The single-cell suspensions were stained in the dark at 4 °C for at least 30 min in FACS buffer with fluorescence-conjugated antibodies against CD45, CD3, CD8, TNF-α, IFN-γ, Gr-1, Ly6G, and CD11b (BD, USA). The stained cells were analyzed by an LSR II flow cytometer (BD Biosciences). The acquired data were processed and analyzed using FlowJo v10.1 software (Treestar).

Bioinformatics analysis

The correlation between HK2 expression and survival in breast cancer patients was analyzed using the Kaplan–Meier Plotter (https://kmplot.com/analysis/), which combines gene expression data from NCBI GEO (https://www.ncbi.nlm.nih.gov/geo/) and GDC Data Portal (https://portal.gdc.cancer.gov/). Clinical data were sourced from the TCGA open-access database (https://portal.gdc.cancer.gov), specifically the breast invasive carcinoma dataset with vital status, tumor status, and overall survival info. TNBC patients were stratified into two subgroups based on HK2 expression. Tumors with above-mean HK2 expression were in the 'high HK2 expression' subgroup, and those below the mean in the 'low HK2 expression' subgroup. Statistical analyses were done using R software (v4.0.3), and p < 0.05 was considered significant.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 7.0 software (GraphPad Software, Inc.). Data are presented as the mean ± standard deviation (SD) with 'n' representing the number of replicates (n = 3). A Student's unpaired t-test was utilized for comparing two groups, while comparisons involving multiple groups were analyzed using one-way analysis of variance (ANOVA) followed by Dunnett's multiple comparison test. A significance level of P < 0.05 was considered statistically significant, and P < 0.01 was deemed highly statistically significant.

Results

Elevated HK2 expression detrimental to clinical prognosis and T cell immune response in breast cancer

To assess the significance of HK2 in the progression of breast cancer and its impact on the tumor immune microenvironment, we conducted an analysis of RNA expression data obtained from TCGA. Analyses of the association between HK2 expression and breast cancer patient survival using the Kaplan-Meier plotter database (https://kmplot.com) revealed that HK2 expression levels were inversely correlated with the survival time of breast cancer patients (Fig. 1A). In addition, HK2 expression is significantly elevated in stage IV and T4 tumors compared with stage I and T1 patients respectively (Fig. 1B, C), although no statistically significant differences are observed in stage II/III and T2/T3. When TNBC patients were categorized based on median HK2 expression levels, those with high HK2 expression showed an enrichment of glycolytic gene signatures (Fig. 1D). In contrast, patients with low HK2 expression displayed enrichment in CD8+ T cell activation, the IFN-γ signaling pathway, and the TNF-α signaling pathway (Fig. 1E–G). Collectively, these findings suggest that HK2 may serve as a potential prognostic biomarker in the context of triple-negative breast cancer.

Fig. 1.

Elevated HK2 expression is detrimental to clinical prognosis and T cell immune response in breast cancer. A Kaplan-Meier survival analysis of HK2 expression in breast cancer patients (n = 4929, data retrieved from the Kaplan–Meier plotter database). B, C Expression of HK2 in tumors stage and T grade in breast cancer (n = 1098, data retrieved from TCGA). (D-G) Gene signatures were assessed and compared between triple-negative breast cancers (TNBCs) with high and low HK2 expression levels (n = 117). The normalized enrichment score (NES) is represented by the green line, indicating the extent of over-representation at the top or bottom of the ranked gene list. A positive NES value reflects a stronger association with "high HK2-expressing tumors," while a negative value reflects a stronger association with "low HK2-expressing tumors."

HK2 inhibition suppresses G-CSF expression in triple-negative breast cancer cells

Given that HK2 plays a central role in the glycolytic pathway and glycolysis governs the expression of granulocyte colony-stimulating factor (G-CSF) in breast cancer cells[26], it is reasonable to hypothesize that HK2 may exert regulatory influence on G-CSF expression in triple-negative breast cancer. To test this hypothesis, shHK2 was transfected into 4T1 cells and MDA-MB-231 cells respectively, and the efficacy of transfection was confirmed by western blotting (Fig. 2A) and RT-qPCR (Fig. 2B, C), showing successful silencing of HK2 (p < 0.01).

Fig. 2.

HK2 inhibition suppresses G-CSF expression in triple-negative breast cancer cells. A, B Efficiency of shHK2 in knocking down HK2 expression in 4T1 cells demonstrated by western blotting (A) and RT-qPCR (B). C Knockdown of HK2 via siRNA in MDA-MB-231 cells shown by RT-qPCR. D–G Effect of shHK2 in 4T1 and MDA-MB-231 on G-CSF transcription assessed by real-time PCR (D, E), and G-CSF protein levels, measured by ELISA (F–G). H, I Impact of 3-BrPA on HK2 mRNA expression in 4T1 and MDA-MB-231 cells. Significance levels are indicated as follows: *p < 0.05, **p < 0.01, ***p < 0.001. Error bars represent the mean ± SD

Subsequently, the expression of G-CSF in shHK2-transfected 4T1 cells and MDA-MB-231 cells were evaluated and found to be significantly reduced compared to the control group, as determined by RT-qPCR (Fig. 2D, E, p < 0.05) and ELISA (Fig. 2F, G). As an HK2 inhibitor, 3-BrPA effectively suppressed glycolysis in both cells. Our results demonstrate that the expression of G-CSF in 4T1 cells and MDA-MB-231 cells was markedly attenuated by 3-BrPA (Fig. 2H, I, p < 0.01). In summary, the inhibition of HK2 function led to a down-regulation of G-CSF expression in breast cancer cells.

Knockdown of HK2 suppresses MDSC development in the tumor microenvironment

G-CSF plays a pivotal role in modulating MDSCs and inducing their development under both in vitro and in vivo conditions. In our in vitro experiment, we observed a significant reduction in the induction of MDSCs when exposed to the supernatant from shHK2 4T1 cells in comparison to the tumor-cultured medium from control 4T1 cells (Fig. 3A, B). In vivo, tumor growth was notably diminished in the shHK2 group as compared to the control group (Fig. 3C, D) and mice in the shHK2 group exhibited improved survival outcomes (Fig. 3E). However, in BALB/c-nu mice, tumor growth volume was only moderately decreased in the shHK2 group compared to the scrambled control group (Supplementary Fig. 3A, B). Furthermore, shHK2 4T1 tumors exhibited markedly diminished expression of G-CSF compared to scrambled tumor tissues (Fig. 3F). As anticipated, the infiltration of MDSCs within shHK2 4T1 tumors was lower than that in control tumor tissues (Fig. 3G, H). These observations collectively indicate that HK2 inhibition effectively reduces the development of MDSCs within the tumor microenvironment.

Fig. 3.

Knockdown of HK2 suppresses MDSC development in the tumor microenvironment. A, B MDSCs derived from bone marrow cells co-cultured with G-CSF or supernatants from scrambled and shHK2 4T1 cells were analyzed by flow cytometry (n = 3/group). Mice (n = 9/group) were implanted with 4T1 or shHK2 4T1 cells. C Tumor images from scrambled and shHK2 4T1-bearing mice after 23 days (n = 9/group). D Tumor growth curves of indicated groups (n = 9/group). E Survival analysis of mice bearing scrambled and shHK2 4T1 tumors. F G-CSF expression in tumor tissues analyzed by western blotting. G, H MDSCs in tumor tissues of mice bearing scrambled and shHK2 4T1 cells, shown as percentages of Gr1+CD11b+cells within CD45+ cells by flow cytometry. Significance: *p < 0.05, **p < 0.01, ***p < 0.001. Data presented as mean ± SD

HK2 inhibition augments immunotherapeutic efficacy of PD-L1 antibody in triple-negative breast cancer

Normalization of the tumor immune microenvironment is a crucial step in immunotherapy for tumors. MDSCs, as immunosuppressive cells, negatively regulate cytotoxic-T cell activity within the tumor. To evaluate the synergistic effects of HK2 inhibition and anti-PD-L1 therapy in triple-negative breast cancer (TNBC), we administered a combination of 3-BrPA and PD-L1 antibody to tumor-bearing mice. The results clearly demonstrated that the combined treatment of 3-BrPA and PD-L1 antibody resulted in the most significant tumor growth reduction (Fig. 4A, B) and the highest overall survival rates (Fig. 4C) in mice bearing 4T1 tumors.

Fig. 4.

HK2 inhibition augments immunotherapeutic efficacy of PD-L1 antibody in triple-negative breast cancer. 4T1-bearing mice were treated with control, 3-BrPA, PD-L1 antibody, or combination therapy (n = 5/group). A Tumor images of 4T1 breast cancer mice. B Tumor growth curves for each group; tumor volume data are shown as mean ± SD (n = 5/group). C Survival analysis of mice treated with PD-L1 antibody, 3-BrPA, or their combination (n = 5/group). D, E Flow cytometry dot plots showing Gr1+ CD11b+ cells within CD45+ cells in tumor tissues from treated mice; percentages are given (n = 5/group). F, G Scatter plots of CD8+ T lymphocytes within CD45 + cells for each treatment group (n = 5/group). H–J Flow cytometry dot plots and percentages of TNF-α and IFN-γ expression in CD8+ T cells in 4T1 tumor tissue (n = 5/group). Significance: ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001. Error bars indicate mean ± SD

Furthermore, combined treatment group exhibited the most pronounced reduction in intratumoral MDSC infiltration (Fig. 4D, E). Consequently, CD8 + T cell infiltration was significantly increased (Fig. 4F, G) accompanied by elevated levels of TNF-α (Fig. 4H, I) and IFN-γ (Fig. 4H, J). Compared with the control group, both PD-L1 antibody and 3-BrPA alone could reduce the percentage of MDSCs to a certain extent, while the combination of the two has a more significant effect on reducing the percentage of MDSC cells, suggesting that PD-L1 antibody and 3-BrPA may have a synergistic effect. To further confirm these findings, we performed similar in vivo experiments using shHK2 4T1 cells. Tumors derived from shHK2 4T1 cells treated with PD-L1 antibody showed minimized growth (Supplementary Fig. 4A, B). Additionally, MDSC infiltration was most prominently reduced in this treatment group (Supplementary Fig. 4C, D), while CD8 + T cell infiltration (Supplementary Fig. 4E, F) and the levels of TNF-α (Supplementary Fig. 4G, I) and IFN-γ (Supplementary Fig. 4G, H) were significantly increased. These findings collectively demonstrate that HK2 inhibition restores the tumor immune microenvironment, thereby improving the efficacy of anti-PD-L1 therapy for TNBC.

Discussion

Chemotherapy remains the primary mode of adjuvant therapy, particularly for triple-negative breast cancer (TNBC), which is defined by the absence of estrogen, progesterone, and HER2 receptors. Since TNBC does not respond to hormone therapy, chemotherapy is often the only viable option [27]. However, its clinical utility is significantly limited by severe adverse effects such as gastrointestinal toxicity, bone marrow suppression, and cardiotoxicity [28]. Additionally, chemotherapy resistance and early relapse pose major clinical challenges [29]. In recent years, immunotherapy has emerged as a promising treatment modality, with breakthroughs in immune checkpoint blockade therapies for several solid tumors. As a result, immunotherapy has gained increasing attention as a potential option for TNBC treatment.

The Warburg effect, defined by a shift from oxidative phosphorylation to aerobic glycolysis, is a hallmark of cancer that confers a growth advantage to tumor cells. This heightened glycolytic activity is driven by the overexpression of key enzymes such as hexokinase (HK), phosphofructokinase (PFK), pyruvate kinase (PK), and lactate dehydrogenase (LDH) [30]. HK is regarded as the first important rate-limiting enzyme. However, the expression of HK isoforms differs between normal tissues and tumors. HK1 is widely expressed across various organs, whereas HK2 is markedly upregulated in cancer cells. The increased hexokinase activity in cancer cells is primarily driven by HK2 induction [7, 31]. Consequently, inhibiting HK2 in tumor cells may potentially play a more crucial role in cancer therapy. In our study, we observed that high expression of hexokinase 2 (HK2) was inversely associated with patient survival, with the benefits of low HK2 expression becoming evident earlier. These findings underscore the pivotal role of HK2 in tumor progression.

HK2 can influence tumor progression through multiple pathways. Inhibition of HK2 not only reduces lactate production but also affects overall glycolytic flux, which induces broader metabolic reprogramming in cancer cells that may provide additional benefits beyond lactate modulation. Inhibition of HK2 can induce the opening of the mitochondrial permeability transition pore (MPTP), thereby increasing the permeability of the mitochondrial membrane and subsequently promoting apoptosis in tumor cells [32]. Additionally, inhibiting HK2 can also promote gluconeogenesis, thereby reducing the energy supply to tumor cells and consequently suppressing their proliferation [33]. All these suggest that a combined approach targeting both metabolic reprogramming and tumor growth could be more effective. However, the specific role of HK2 in regulating immune responses TNBC remains largely uncharacterized.

TNBC, which accounts for approximately 10–20% of invasive breast cancers, is characterized by its heterogeneity and complex molecular pathways. G-CSF plays a key role in promoting the development of MDSCs and is highly expressed in TNBC, contributing to immune evasion and tumor progression [34]. In our study, we found that HK2 inhibition in TNBC cells significantly reduced G-CSF expression, thereby creating a microenvironment less conducive to MDSC development. In immune-competent mice, tumors with HK2 knockdown exhibited slower progression compared to controls. In contrast, in immunodeficient mice, tumors in the HK2-knockdown group showed only a moderate reduction compared to the scrambled control group. These findings suggest that HK2 promotes tumor progression by suppressing anti-tumor immunity within the TME. Given the established role of MDSCs in facilitating immune evasion [35], our results indicate that targeting HK2 may help restore anti-tumor immunity and represents a promising therapeutic strategy for TNBC.

TNBC is widely recognized as the most immunogenic subtype of breast cancer, suggesting its potential responsiveness to immunotherapy (IO) [36]. IM offers a promising therapeutic strategy for TNBC by harnessing the immune system to identify and eliminate tumor cells. Several clinical trials have demonstrated the potential benefits of IM for TNBC patients, particularly when combined with chemotherapy [37, 38]. In July 2021, the U.S. Food and Drug Administration (FDA) approved a neoadjuvant regimen of pembrolizumab plus chemotherapy, followed by adjuvant pembrolizumab, for the treatment of high-risk, early-stage TNBC. This approval was based on statistically significant and clinically meaningful improvements in pathological complete response (PCR) and event-free survival (EFS) observed in a phase III randomized controlled trial (RCT) [38, 39]. However, PD-L1/PD-1 and CTLA-4 blockade therapies for TNBC have not achieved the expected levels of clinical success [39]. Since tumor metabolism plays a key role in reshaping the tumor immune microenvironment [40, 41], combining metabolic interventions with immune checkpoint inhibitors may offer a strategy to overcome resistance to immunotherapy.

In our study, combining 3-bromopyruvate (3-BrPA), an HK2 inhibitor, with a PD-L1 antibody produced a significant antitumor effect in a mouse model of breast cancer, greatly surpassing the efficacy of either treatment alone. This finding highlights a novel therapeutic approach for TNBC that integrates metabolic reprogramming with immunotherapy to enhance antitumor immunity and reduce immune suppression. 3-BrPA depletes ATP in cancer cells by targeting the Warburg effect [30], without significantly affecting ATP utilization in normal tissues.

Although 3-BrPA is an effective inhibitor of HK2 and glycolysis, it has known off-target effects that should be considered when interpreting its therapeutic impact. 3-BrPA can disrupt mitochondrial function, induce oxidative stress, and inhibit other glycolytic enzymes such as GAPDH, potentially affecting cellular metabolism beyond glycolysis [42, 43]. Additionally, its alkylating properties may lead to non-specific protein modifications, influencing various signaling pathways [44]. These effects could contribute to the observed enhancement of anti-PD-L1 therapy, but they also underscore the need for careful evaluation when using 3-BrPA in preclinical and clinical settings due to its high toxicity and side effects, including risks of hepatotoxicity and nephrotoxicity [47–49 ]. Several similar drugs that can inhibit HK2, such as LN and 2-DG, have been studied for decades and have shown significant inhibition in various cancers. However, due to side effects, they still failed to be applied in clinical practice [45, 46]. In the future, other HK2 inhibitors with fewer side effects could potentially be used to enhance anti-tumor immunity in TNBC.

In summary, our findings underscore the critical role of HK2 in regulating tumor immunity within TNBC. Inhibition of HK2 by 3-BrPA effectively reduces G-CSF expression, inhibits MDSC development, and enhances T cell-mediated immunity in the tumor microenvironment. The combination of 3-BrPA with anti-PD-L1 therapy represents a promising therapeutic approach for TNBC, as demonstrated both in vitro and in a mouse model. However, further research is needed to assess the feasibility and clinical efficacy of this combination therapy in human settings.

Author contributions

All authors contributed to the study conception and design. CL.: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing-Original Draft, Visualization, Project administration. YT.: Conceptualization, Formal analysis, Investigation, Resources, Data Curation, Project administration. RZ.: Investigation, Validation. LS.: Visualization, Project administration. JC.: Data curation, Investigation. PZ.: Supervision. NZ.: Visualization, Investigation. WL.: Conceptualization, Writing- Reviewing and Editing, Supervision, Funding acquisition. All authors contributed to the review of the first draft and provided valuable input for revisions. All authors read and approved the final manuscript. All authors contributed to the article and approved the final version for submission.

Funding

This work was supported by National Natural Science Foundation of China (82072736, 81902703, 82322052 and 82273319).

Data availability

The data supporting the findings of this study are included within the paper and its Supplementary Information files. Additional raw data files can be provided by the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval for this study was obtained from the Institutional Animal Care and Use Committee at Tongji Medical College, Huazhong University of Science and Technology (Approval Number: S2524). All methods were carried out in accordance with relevant guidelines and regulations. The Ethics committee allows tumor burden of 2 cm or about 10% of the body weight in mice. The maximum tumor burden recorded in the current study did not exceed 10% of the body weight.

Competing interests

The authors declare no competing interests.