Transcriptome profiles, clinical information (including 1,023 BRCA samples and 113 adjacent non-tumor samples), and somatic mutation data (including 1,026 BRCA samples) from breast cancer patients were obtained from TCGA. Additional transcriptome signatures and clinical data were sourced from GSE42568 and GSE22219 datasets in the Gene Expression Omnibus (GEO) database. Protein expression analysis was conducted using data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC), accessed via the UALCAN portal.

The “Gene_DE” module of TIMER2.0 (Tumor Immune Estimation Resource, version 2) was used to evaluate GFPT1 expression levels across various TCGA tumor tissues compared to adjacent normal tissues (Li et al., 2020). For tumors lacking normal tissue samples, such as TCGA-ACC (adrenocortical carcinoma) and TCGA-DLBC (diffuse large B-cell lymphoma), the GEPIA2.0 web server was used to generate box plots to visualize GFPT1 expression differences (Tang et al., 2019). This analysis was performed using a significance threshold of p < 0.01 and a minimum log2 fold change of 1, with the expression data log-transformed as log2 (TPM + 1).

The Kaplan-Meier plotter, an online survival analysis tool incorporating various microarray datasets, was used to assess the correlation between GFPT1 expression and survival outcomes in cancer patients (Gyorffy, 2021). Multivariate Cox regression analysis was conducted to identify potential prognostic indicators. A nomogram was generated using the R rms package to visually represent prognostic significance. Statistical significance was set at p < 0.05, with p < 0.001 considered highly significant.

The Limma package was used to identify differentially expressed genes (DEGs) between low and high GFPT1 expression groups. DEGs were defined by an adjusted p-value <0.05 and log2 (fold change) > 0.5. The ClusterProfiler package was used for Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. The gseGO and gseKEGG functions were applied to explore GO terms, covering biological processes, cellular components, and molecular functions, as well as pathway enrichment. Gene Set Enrichment Analysis (GSEA) was performed using GSEA software to identify enriched pathways.

The ESTIMATE package was used to assess the composition of immune cells (immune score), the degree of stromal cell infiltration (stromal score), the combined stromal-immune score (ESTIMATE score), and the tumor purity of each sample. The CIBERSORT deconvolution algorithm was applied to estimate the proportions of 22 types of tumor-infiltrating immune cells (TIICs) in the TCGA-BRCA cohort. The Vioplot package was then employed to analyze the variations in proportions of these 22 types of TIICs across different groups.

The R package maftools was used to analyze mutations and identify frequently mutated genes within the TCGA-BRCA cohort. The plotmafSummary function was utilized to visualize variant counts per sample, including variant classification, type, SNV class, and overall mutation profiles, highlighting the top 10 most frequently mutated genes. Waterfall plots were generated using the oncoplot function to display gene mutation frequencies. The OncogenicPathways function was used to identify pathways enriched with mutated genes, while the tcgaCompare function compared mutational burdens across different cancer types. Chemosensitivity was predicted using the R package pRRophetic by comparing IC50 values between the GFPT1-High and GFPT1-Low groups.

The Hep3B2.1-7 human hepatocellular carcinoma cell line and breast cancer cell lines (MCF-10A, MCF-7, MDA-MB-231, and T47D) were purchased from Haixing Biosciences (Suzhou, Jiangsu, China). All cell lines were cultured at 37°C in a humidified incubator with 5% CO2 (Thermo Fisher, United States). Cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (Zeta Life, AUS), 100 units/mL penicillin, and 100 μg/mL streptomycin.

MDA-MB-231 and MCF-7 cell lines were transfected using Lipofectamine 2000 reagent (Thermo Fisher, United States) with specific siRNAs targeting GFPT1: siRNA-GFPT1#1 (5′-CAA​AGG​CUA​UGA​CUU​CGA​A-3′), siRNA-GFPT1#2 (5′-CAA​GUG​CUG​UCA​UAG​AAC​A-3′), and siRNA-GFPT1#3 (5′-GAA​UCA​UCA​CCA​ACU​ACA​A-3′). A scrambled siRNA was used as a negative control (NC) due to its lack of significant homology to human genome sequences. Final siRNA concentrations were 50 nM for MDA-MB-231 cells and 75 nM for MCF-7 cells, following the manufacturer’s instructions. For GFPT1 overexpression, the negative control (PcDNA3.1), PcDNA3.1 (+)/GFPT-1, were transfected into the cells utilizing Lipofectamine 3,000 (Thermo Fisher, United States) with a final concentration of 11 μg. After 48 h, cells were replenished with complete medium, and western blot analysis confirmed the reduction of GFPT1 protein levels post-transfection.

For western blotting, cell lysates were prepared using radioimmunoprecipitation assay (RIPA) buffer (HUAYUNBIO, HB504A, China) containing protease and phosphatase inhibitors (Roche, United Kingdom). Total protein concentration was determined using the BCA protein assay kit (BEYOTIME BIOTECH INC., P0010S, China). Proteins were separated using 4%–12% Bis–Tris polyacrylamide gels and transferred to PVDF membranes (Millipore, United States). The membranes were washed with TBST, blocked with 5% bovine serum albumin in TBST for 1 h, and incubated with primary antibodies: anti-GFPT1 (1:1,000; 14132-1-AP, Proteintech Group), Anti-GSDMD (1:1,000; PU224937, Abmart) and anti-GAPDH (1:1,000; 60004-1-Ig, Proteintech Group). Membranes were then probed with LI-COR/IRDye 800CW goat anti-rabbit IgG (1:4,000, 926–68071, Proteintech Group) and LI-COR/IRDye 680CW goat anti-mouse IgG (1:4,000, 926–68070, Proteintech Group). Imaging was performed using a CCD camera (Bio-Rad ChemiDoc MP), and band intensities were quantified using a prestained protein ladder (MIKX, Co., Ltd., DB182-01, China).

Cell growth was assessed using the Cell Counting Kit-8 (MIKX, Co., Ltd., M0856-05, China) following the manufacturer’s procedures. Cells were seeded 8 × 10^3 cells per well in 96-well plates across different treatment groups. The CCK-8 solution was added at designated time intervals, followed by a 2-h incubation at 37°C, and the optical density (OD) was measured at a wavelength of 450 nm for each well using a microplate reader.

Cell proliferation was assessed using the EdU staining kit (Beyotime, Shanghai, China) following the manufacturer’s protocol. Cells from different treatment groups were seeded at a density of 3 × 10³ cells per well in 96-well plates and cultured for 48 h. EdU (20 mmol/L) was then added, and the cells were incubated for 2 h. After incubation, cells were fixed with 4% paraformaldehyde for 15 min at room temperature. EdU-positive cells were subsequently analyzed to evaluate the proliferation across treatment groups.

Apoptotic cells were identified using an Annexin V-APC/PI apoptosis kit, following the manufacturer’s protocol (MultiSciences, Hangzhou, China). Cells from different treatment groups were seeded at a density of 2 × 10^5 cells per well in 6-well plates and cultured for 48 h. After harvesting, the cells were stained with APC-conjugated Annexin V and PI for 5 min at room temperature in the dark. Following staining, binding buffer was added, and flow cytometry was used to analyze the cell suspensions. Data acquisition (10,000 events per sample) was performed on a fluorescence-activated cell sorting system (CytoFLEX; Beckman Coulter, Brea, CA, United States) using CytoFLEX software.

Cells from various treatment groups were seeded at a density of 2 × 10^4 cells per well in the upper chamber (Corning 353,097) with DMEM without fetal bovine serum (FBS). The lower chamber contained DMEM supplemented with 10% FBS. After 48 h of incubation, the cells were fixed with 4% paraformaldehyde, followed by staining with 0.1% crystal violet solution. The number of migrated cells was then manually counted under a microscope.

Continuous variables were compared using either Student’s t-test or the Wilcoxon rank-sum test. The Pearson correlation test was used to evaluate the relationships between different sample factors. Survival probabilities were calculated using the Kaplan-Meier method, with differences assessed by the log-rank test. All statistical tests were two-sided, and statistical significance was set at p < 0.05.

To comprehensively analyze the expression and distribution of GFPT1 in various tumor tissues and adjacent normal tissues, we initially used the TIMER 2.0 tool to examine GFPT1 mRNA expression across multiple cancer types in the TCGA dataset. As shown in Figure 1A, GFPT1 expression was significantly higher in tumor tissues compared to adjacent normal tissues in several cancer types, including BLCA (bladder urothelial carcinoma, p = 5.63E-03), BRCA (breast invasive carcinoma, p = 2.56E-2), CHOL (cholangiocarcinoma, p = 4.07E-06), ESCA (esophageal carcinoma, p = 4.99E-02), LIHC (liver hepatocellular carcinoma, p = 9.61E-10), LUAD (lung adenocarcinoma, p = 5.31E-25), LUSC (lung squamous cell carcinoma, p = 3.15E-05), PAAD (pancreatic adenocarcinoma, p = 2.50E-2), PRAD (prostate adenocarcinoma, p = 8.55E-3), STAD (stomach adenocarcinoma, p = 1.80E-08), and UCEC (uterine corpus endometrial carcinoma, p = 1.83E-03). In contrast, GFPT1 expression was lower in GBM (glioblastoma multiforme, p = 0.020), KIRC (kidney renal clear cell carcinoma, p = 3.79E-06), KIRP (kidney renal papillary cell carcinoma, p = 0.016), and THCA (thyroid carcinoma, p = 7.99E-05) tumor tissues compared to normal tissues (Figure 1A).

To expand this analysis, we incorporated normal tissue samples from the GTEx dataset as controls, comparing GFPT1 mRNA expression between cancerous and normal tissues across additional cancer types, including ACC, DLBC, LAML, LGG, OV, SARC, SKCM, TGCT, THYM, and UCS. Significant increases in GFPT1 expression were observed in DLBC, SKCM, and THYM tumor tissues, while no notable differences were found in ACC, LGG, OV, SARC, TGCT, or UCS samples (Figure 1B). Additionally, GFPT1 expression was lower in LAML tumor tissues compared to normal controls.

To further assess GFPT1 protein expression across different cancer types, we analyzed data from the CPTAC dataset. As shown in Figure 1C, GFPT1 protein levels were elevated in breast cancer, colon cancer, ovarian cancer, clear cell RCC, uterine corpus endometrial carcinoma, lung cancer, glioblastoma, and liver cancer tissues compared to normal tissues. In contrast, pancreatic adenocarcinoma tissues showed lower GFPT1 protein expression compared to normal tissues.

Overall, our pan-cancer analysis revealed increased GFPT1 mRNA and protein levels in BRCA, LIHC, and UCEC tumor tissues compared to normal tissues. Further analysis using the CPTAC database demonstrated a positive correlation between GFPT1 protein levels and advanced tumor stages in BRCA and UCEC, suggesting a potential link between GFPT1 and tumor progression (Supplementary Figure S1A, B). Additionally, higher GFPT1 protein expression was significantly associated with higher tumor grades in UCEC, and GFPT1 levels varied across different molecular subtypes of BRCA (Supplementary Figure S1A, B).

We then assessed the correlation between GFPT1 expression and patient survival in BRCA, LIHC, and UCEC using Kaplan-Meier analysis and the log-rank test. An online survival tool (https://kmplot.com/analysis/), which integrates published RNA-seq datasets, revealed a significant association between high GFPT1 expression and reduced overall survival (OS) and progression-free survival (PFS) specifically in breast cancer patients (Figure 2A) (Lanczky and Gyorffy, 2021). However, GFPT1 mRNA levels were not significantly predictive of OS in UCEC patients or relapse-free survival (RFS) in LIHC and UCEC patients (Supplementary Figure S2A, B). These findings suggest that GFPT1 is highly expressed in breast cancer tissues and is associated with a poorer prognosis in breast cancer patients.

Next, we utilized data from TCGA and GEO to further explore the relationship between GFPT1 expression and survival prognosis in breast cancer patients. In the TCGA cohort, elevated GFPT1 expression in tumor tissues was correlated with poorer survival outcomes in breast cancer patients (Figure 2B). Similarly, in the GSE42568 dataset, GFPT1 expression was a significant predictor of OS and RFS in breast cancer patients (Figures 2C, D). Comparable predictive trends for RFS were observed in the GSE22219 dataset (Figure 2E).

To evaluate the prognostic significance of GFPT1 and other clinicopathological factors for OS, we conducted a univariate Cox regression analysis. Significant risk factors for OS in breast cancer patients from the TCGA cohort included age (p = 1.2E-03), cancer stage (p = 1.6E-05), lymph node stage (p = 6.6E-07), distant metastasis stage (p = 0.005), and GFPT1 expression (p = 1.5E-04) (Supplementary Figure S3). Multivariate Cox analysis, adjusting for covariates, confirmed that both age (p = 0.018) and GFPT1 expression (p = 0.019) were independent predictors of OS in TCGA-BRCA patients (Figure 2F). In conclusion, GFPT1 serves as an independent prognostic marker in breast cancer patients, significantly associated with poorer survival outcomes.

We developed a nomogram that integrates predictive factors such as age, cancer stage, T stage, N stage, and GFPT1 expression to quantitatively assess overall survival (OS) probabilities at 1, 3, and 5 years for BRCA patients (Figure 3A). Each patient’s covariates were assigned specific point values, with higher total scores corresponding to lower predicted survival rates. Calibration plots demonstrated that the nomogram’s predictions closely aligned with the ideal model (Figure 3B), indicating strong predictive accuracy.

We performed differential expression analysis between the GFPT1-Low and GFPT1-High groups in the BRCA cohort, identifying 354 differentially expressed genes (DEGs). Among these, 111 genes were upregulated and 243 were downregulated in the GFPT1-High group compared to the GFPT1-Low group (Supplementary Table S1). To explore how GFPT1 influences clinical outcomes in BRCA patients, we conducted GO and KEGG pathway enrichment analyses using the clusterProfiler package.

As shown in Figure 4A, the GO enrichment analysis for biological processes (BP) revealed that the GFPT1-High group was enriched in processes such as “cytoplasmic translation,” “endoplasmic reticulum to Golgi vesicle-mediated transport,” “Golgi organization,” “Golgi vesicle transport,” and “complement activation.” The top three enriched BP GO terms and their associated genes were visualized using correlation Circos plots (Figure 4B). Additionally, the top four enriched cellular component (CC) GO terms were all related to ribosomes, including “cytosolic large ribosomal subunit,” “cytosolic ribosome,” “large ribosomal subunit,” and “ribosomal subunit.”

Gene Set Enrichment Analysis (GSEA) revealed significant associations between oncogenic signaling pathways regulating translation, such as the EIF4 pathway and HALLMARK_PI3K_AKT_MTOR_SIGNALING gene sets, and the GFPT1-High expression group (Figure 4C). Positive correlations were observed between GFPT1 and EIF4E (r = 0.31, p = 2.61E-24), as well as GFPT1 and PIK3C2A (r = 0.47, p = 6.27E-58) (Figure 4D). These findings suggest that tumors with higher GFPT1 expression are more likely to exhibit enhanced translational activity, promoting mRNA translation crucial for cancer progression.

Moreover, KEGG pathway enrichment analysis identified several oncogenic pathways enriched in the GFPT1-High group, including the cell cycle, autophagy, nucleotide sugar biosynthesis, AMPK signaling pathway, and Fanconi anemia pathway, involving genes linked to breast cancer susceptibility (Fang et al., 2020) (Figure 4E). GSEA further showed significant enrichment of cell cycle-related genes, such as HALLMARK_E2F_TARGETS and HALLMARK_G2M_CHECKPOINT, in the GFPT1-High group (Figure 4F). These results highlight the role of elevated GFPT1 expression in promoting cell cycle progression and driving cancer development.

Previous studies have identified somatic mutations as critical drivers of cancer development (Oh and Sung, 2021). To investigate differences in somatic mutations between the GFPT1-High and GFPT1-Low groups in the BRCA cohort, we analyzed the TCGA-BRCA dataset using the Maftools package. The mutational landscape revealed that missense mutations were the most common, with single nucleotide polymorphisms (SNPs) being the predominant variant type in both groups. Among single nucleotide variants (SNVs), C > T transitions were the most frequently observed (Supplementary Figure S4A, B).

Compared to the GFPT1-Low group, the GFPT1-High group exhibited a higher median mutational burden per sample (median: 41), with TP53 mutations present in 37% of samples—a significantly higher proportion than in the GFPT1-Low group (Figures 5A, B). The Oncoplot function visualized the top 20 mutated genes in both groups, displayed in waterfall plots for the GFPT1-High (Figure 5C) and GFPT1-Low groups (Supplementary Figure S5C). Notably, genes such as TP53, DNAH17, USH2A, LRP1, and TTN were more frequently mutated in the GFPT1-High group compared to the GFPT1-Low group (Figure 5D).

Pathway analysis revealed that oncogenic pathways, particularly the PI3K pathway, were significantly enriched among the mutated genes in the GFPT1-High group (Figure 5E). Figure 5F highlights the specific mutated genes within the PI3K pathway. Additionally, when comparing mutation rates between the GFPT1-High BRCA cohort and tumor mutation burdens (TMBs) across various cancer types in the TCGA dataset, BRCA exhibited a median mutation rate, ranking between cholangiocarcinoma and uterine carcinosarcoma (Figure 5G).

Tumors are known to elicit immune responses by presenting antigens, which lead to infiltration of lymphocytes into both the tumor and its surrounding stroma (Rohan et al., 2021). Previous studies have shown that breast cancer tissues exhibit a suppression of the adaptive immune system and an activation of the innate immune system (DeNardo and Coussens, 2007). To explore the immune landscape in breast cancer, we used the ESTIMATE algorithm to assess immune scores, stromal scores, and tumor purity between the GFPT1-Low and GFPT1-High groups in BRCA (Newman et al., 2015). Our analysis revealed that the GFPT1-High group exhibited significantly higher tumor purity, along with markedly lower stromal, immune, and ESTIMATE scores, suggesting reduced immune cell infiltration in this group. This immunosuppressive environment likely supports tumor progression (Figure 6A). Figures 6B,C highlight the distribution and percentages of immune cells in the GFPT1-High group, showing that macrophages and T lymphocytes were the dominant immune cells infiltrating the breast cancer tissues. Figure 6D illustrates weak to moderate correlations between representative immune cells infiltrating the tumor microenvironment in the GFPT1-High BRCA group.

To further investigate the differences in tumor-infiltrating immune cells (TIICs) between the GFPT1-Low and GFPT1-High groups, we applied the CIBERSORT algorithm to breast cancer patients from the TCGA cohort. The violin plot revealed significant reductions in CD8⁺ T cells (p < 0.001), Tregs (p < 0.001), resting NK cells (p < 0.001), monocytes (p = 0.01), and resting dendritic cells (p = 0.048) in the GFPT1-High group (Figure 6E). Research has shown that M2 macrophage polarization is strongly associated with breast cancer aggressiveness, larger tumor size, advanced stages, and angiogenesis (Jayasingam et al., 2019; Jeong et al., 2019). In our study, elevated GFPT1 expression was linked to a significantly increased proportion of M2 macrophages, suggesting a direct correlation between GFPT1 expression and the anti-inflammatory infiltration of M2 macrophages in breast cancer tissues (Figure 6E).

Further analysis using TIMER2.0 demonstrated a robust association between GFPT1 expression and M2 macrophage infiltration in breast cancer tissues (Figure 6F). Additionally, TIMER2.0 revealed that patients with low GFPT1 expression and low M2 macrophage infiltration had the most favorable prognosis, while those with high GFPT1 expression and high M2 macrophage infiltration had the poorest survival outcomes (Figure 6G). These findings underscore the critical role of GFPT1 expression, M2 macrophage infiltration, and their impact on clinical outcomes in BRCA patients.

To assess the clinical significance of GFPT1 expression, we applied the pRRophetic algorithm and package to analyze the half-maximal inhibitory concentration (IC50) values of eight commonly used chemotherapeutic agents in breast cancer treatment, including Cisplatin, Doxorubicin, Docetaxel, Gemcitabine, Imatinib, Paclitaxel, Vinblastine, and Vinorelbine, across GFPT1-High and GFPT1-Low groups in BRCA. The results demonstrated significantly lower IC50 values in the GFPT1-Low group for most drugs: Cisplatin (p = 0.039), Docetaxel (p = 2.22E-16), Gemcitabine (p = 1.8E-04), Imatinib (p = 1.3E-05), Paclitaxel (p = 2.22E-16), Vinblastine (p = 2.22E-16), and Vinorelbine (p = 2.7E-03), with the exception of Doxorubicin (p = 0.530) (Figures 7A–H).

We further evaluated the IC50 values of immunomodulatory drugs, including Bortezomib (Figure 7I), Lenalidomide (Figure 7J), Rapamycin (Figure 7K), and Vorinostat (Figure 7L), based on GFPT1 expression. The data revealed that GFPT1 could serve as a potential biomarker for predicting the efficacy of immune-related drugs such as Bortezomib (p = 2.22E-16), Rapamycin (p = 2.22E-16), and Vorinostat (p = 6.1E-15). These findings highlight the potential of GFPT1 expression levels as a valuable predictor of chemotherapy response and immunotherapy outcomes in invasive breast carcinoma.

We next assessed GFPT1 protein levels and compared its expression in MCF10A cells, derived from normal human mammary epithelial cells, with breast cancer cell lines MCF-7, MB-MDA-231, and T47D. Our results demonstrated a decrease in GFPT1 protein expression in the breast carcinoma cell lines (MCF-7, MB-MDA-231, and T47D) compared to the MCF-10A cell line (Figure 8A). Hep3B2.1-7 cells, a human hepatocellular carcinoma cell line, were used as a positive control for GFPT1 protein expression.

To investigate the role of GFPT1 in breast cancer cell growth, we first employed RNA interference to suppress GFPT1 expression in MCF-7 and MDA-MB-231 cells. Western blot analysis confirmed the successful downregulation of GFPT1 protein levels by GFPT1 siRNA, with siRNA-GFPT1#3 showing the most significant reduction in both cell lines (Figures 8B, C). Subsequent analysis of cell viability following GFPT1 inhibition revealed a marked decrease in survival rates in both MCF-7 and MDA-MB-231 cells transfected with siRNA-GFPT1#2 and siRNA-GFPT1#3 (Supplementary Figure S5A, B). To further explore the function of GFPT1, we overexpressed GFPT1 by transfecting MCF-7 and MDA-MB-231 cells with PcDNA3.1 (+)/GFPT1 plasmids, with overexpression efficiency confirmed via western blot (Supplementary Figure S6A, B). Overexpression of GFPT1 significantly increased cell viability in both breast cancer cell lines (Figure 8D).

Additionally, we assessed cell proliferation using the EdU staining assay. The results demonstrated that GFPT1 knockdown with siRNA-GFPT1#3 significantly inhibited cell proliferation, while overexpression of GFPT1 enhanced proliferation in both MCF-7 and MDA-MB-231 cells (Figures 8E, F). These findings suggest that GFPT1 is critical for the growth and proliferation of breast cancer cells in vitro.

To investigate the influence of GFPT1 on apoptosis in breast cancer cells, we employed APC/Annexin V and PI dual-staining flow cytometry to analyze apoptosis and necrosis in human MCF-7 and MDA-MB-231 cells. This analysis included early apoptotic (APC+/PI-) and late apoptotic/necrotic cells (APC+/PI+). As shown in Figure 9A, the percentages of apoptotic MCF-7 cells significantly increased with various GFPT1 siRNAs: siRNA-GFPT1#1 resulted in approximately 6.80% early apoptosis and 3.00% late apoptosis/necrosis; siRNA-GFPT1#2 showed around 6.21% early apoptosis and 2.09% late apoptosis/necrosis; and siRNA-GFPT1#3 led to approximately 9.61% early apoptosis and 2.54% late apoptosis/necrosis. Similar trends were observed in MDA-MB-231 cells, where siRNA-GFPT1#1 induced about 17.21% early apoptosis and 2.82% late apoptosis/necrosis; siRNA-GFPT1#2 resulted in approximately 20.32% early apoptosis and 2.58% late apoptosis/necrosis; and siRNA-GFPT1#3 showed around 19.06% early apoptosis and 4.19% late apoptosis/necrosis following GFPT1 siRNA transfection (Figure 9A). Overall, these results suggest that suppressing GFPT1 expression promotes apoptosis and necrosis, thereby hindering the growth of breast cancer cells. We also investigated the impact of GFPT1 overexpression on apoptosis in breast cancer. As illustrated in Figure 9C, the proportion of early and late apoptotic cells was significantly reduced in MCF-7 and MDA-MB-231 cells overexpressing GFPT1 compared to mock control cells. These results clearly indicate that GFPT1 overexpression attenuates apoptotic cell death in both MCF-7 and MDA-MB-231 cell lines.

We also investigated the potential effects of GFPT1 on other types of cell death. The expression of GSDMD, a key effector of pyroptosis, was evaluated in MCF-7 and MDA-MB-231 breast cancer cell lines following transfection with GFPT1-targeting siRNA (siRNA-GFPT1#3), an empty vector control (PcDNA3.1), and a GFPT1 overexpression vector (PcDNA3.1 (+)/GFPT1). However, we did not observe any significant differences in the expression of cleaved GSDMD protein, the active form associated with pyroptosis, among the various treatment groups (Supplementary Figure S7A, B). These findings indicate that GFPT1 does not play a regulatory role in modulating pyroptosis in breast cancer cells.

Given that tumor metastasis involves the migration and invasion of cancer cells, we conducted a transwell migration assay to determine the impact of GFPT1 siRNA and GFPT1 overexpression on breast cancer cell migration. The findings indicated that reducing GFPT1 expression significantly impaired the migration ability of MDA-MB-231 cells while the migration ability of the cells was significantly increased after GFPT1 overexpression, suggesting that GFPT1 plays a role in promoting breast cancer cell migration and metastasis (Figures 9B, D).