MDA-MB-231, the breast cancer cell line, was procured from Wuhan Servicebio Technology Co., Ltd. These cells were maintained in DMEM medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS. The cell cultures were incubated in a humidified 5% CO₂ environment at a constant temperature of 37˚C.

A 6-well plate was prepared by plating 5x10⁵ MDA-MB-231 breast cancer cells per well, followed by a 24 h stabilization period in the incubator. The siRNA sequences used in the present study were as follows: HNRNPAB-siRNA sense strand: 5'-GGAGAGGUCGUUGACUGUAdTdT-3', antisense strand: 5'-UACAGUCAACGACCUCUCCdAdA-3'; negative control (NC)-siRNA sense strand: 5'-UUCUCCGAACGUGUCACGUdTdT-3', antisense strand: 5'-ACGUGACACGUUCGGAGAAdTdT-3'. Subsequently, NC-siRNA and HNRNPAB-siRNA (50 nm) (Tianjin Sheweisi Biotechnology Co., Ltd.) were individually introduced into the MDA-MB-231 cells at 37˚C for 24-72 h using Lipofectamine™ 3000 reagent (Thermo Fisher Scientific, Inc.) in strict accordance with the manufacturer's guidelines. At 8 h post-transfection, the transfection medium was replaced with fresh DMEM containing 10% FBS. The efficiency of siRNA transfection was quantified and the results are presented with corresponding SD values.

For this assay, MDA-MB-231 cells following siRNA transfection were selected. The culture medium was aspirated and lysis buffer (Wuhan Servicebio Technology Co., Ltd.) was added to the cells. Total RNA was isolated using a commercial RNA extraction kit (Tiangen Biotech Co., Ltd.). After quantifying the RNA concentration, reverse transcription into cDNA was performed using the ABScript II RT Master Mix for qPCR with gDNA Remover (ABclonal Biotech Co., Ltd.) as per the manufacturers' instructions. The synthesized cDNA was stored long-term at -80˚C. Fluorescence qPCR was conducted on a LightCycler480 real-time fluorescence quantitative PCR instrument (Roche Diagnostics GmbH), with cDNA loaded according to the protocol of the Genious 2X SYBR Green Fast qPCR Mix kit (ABclonal Biotech Co., Ltd.). The standardized thermal cycling conditions were set as follows: Initial pre-denaturation at 95˚C for 3 min; followed by 40 cycles of denaturation at 95˚C for 15 sec, annealing at 60˚C for 20 sec and extension at 72˚C for 20 sec. Melting curve analysis was subsequently performed from 65˚C to 95˚C to verify the specificity of amplification products. The primer sequences used were as follows: HNRNPAB forward, 5'-TTTGGCGAGTTTGGGGAGATT-3' and reverse, 5'-GCCATACTGCTGCTGATAGAC-3'; GAPDH forward, 5'-GGAGCGAGATCCCTCCAAAAT-3' and reverse, 5'-GGCTGTTGTCATACTTCTCATGG-3'. GAPDH served as the internal reference gene and the 2^-ΔΔCq method was applied to calculate the relative mRNA expression levels (18). All experimental reactions were conducted in triplicate to ensure reproducibility.

Logarithmic-phase MDA-MB-231 cells transfected with either siRNA or NC-siRNA were digested with trypsin for 2 min and then seeded into a 96-well plate at a density of 8x10⁴ cells per ml. After incubation for 24, 48 and 72 h in a 5% CO₂ incubator at 37˚C, the 96-well plate was removed. The culture medium was discarded and a mixture of DMEM and CCK-8 solution (Jiangsu Kaiji Biotechnology Co., Ltd.) at a 9:1 ratio was added to each well. After 1 h, the optical density value of the cells was measured at a wavelength of 450 nm. The cell proliferation rate was calculated, with the blank knockdown group designated as the control.

Firstly, the upper chamber of a Transwell insert (Corning, Inc.) was coated at 37˚C for 2 h with 100 µl 10% Matrigel (BD Biosciences) diluted in serum-free DMEM, followed by a 2 h incubation in a cell culture incubator. Cells were then seeded into the upper chamber of the 8 µm pore-sized Transwell at a density of 5x10⁴ cells per well. The lower chamber was filled with 750 µl culture medium containing 10% FBS and the cells were incubated for 48 h at 37˚C in a 5% CO₂ environment. After incubation, the medium in the upper chamber was discarded. Cells remaining in the chamber were fixed at room temperature with 4% paraformaldehyde for 30 min and stained with 0.1% crystal violet at room temperature for 15 min. The number of cells was counted in three randomly selected fields of view per chamber using an inverted light microscope.

HNRNPAB expression levels in BRCA tissues from The Cancer Genome Atlas (TCGA; https://portal.gdc.cancer.gov/) and Genotype-Tissue Expression (GTEx) databases were retrieved through the Gene Expression Profiling Interactive Analysis 2 platform (19-21). TCGA pan-cancer (PANCAN) dataset was acquired from the UCSC Xena browser (https://xenabrowser.net/), which offers a standardized and comprehensive compilation of pan-cancer datasets. Immunohistochemistry data for HNRNPAB was obtained from The Human Protein Atlas (HPA) database (https://www.proteinatlas.com) (22) which characterizes the expression pattern of HNRNPAB protein in both BRCA tissues and normal breast tissues. Prognostic survival analysis of BRCA sample data was conducted using the Kaplan-Meier Plotter online tool (http://kmplot.com/analysis/) (23) to explore the association between HNRNPAB and the prognosis of patients with BRCA. The cBioportal (https://www.cbioportal.org) (24) and TCGA databases were utilized to investigate HNRNPAB-associated gene alterations in patients with breast cancer. Immunotherapy outcomes and immune cell infiltration profiles of patients with breast cancer were collected from The Cancer Immunome Database (TCIA; https://tcia.at) and the influence of HNRNPAB on immune checkpoint inhibitor efficacy was predicted using the Immune Phenotype Score (IPS) (25). In addition, the StromalScore, ImmuneScore and ESTIMATEScore for BRCA samples in TCGA database was computed using the ‘estimate’ package in R software (26) and the immune cell types associated with HNRNPAB expression levels were summarized.

To further characterize differentially expressed genes and elucidate the functions of potential target genes, gene function enrichment analyses were performed using HALLMARK (https://www.gsea-msigdb.org/gsea/msigdb/index.jsp), REACTOME (https://reactome.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG; https://www.kegg.jp/) databases. The ‘ClusterProfiler’ package (version 3.18.0) in R software was employed to analyze pathway enrichment, thereby enhancing the understanding of mRNA-associated carcinogenic mechanisms (27-29). Gene matrix transposed files were downloaded from the Molecular Signatures Database (https://www.gsea-msigdb.org/gsea/msigdb/) and GSEA was applied to conduct enrichment analyses for HALLMARK, REACTOME and KEGG gene sets (30-32). Genes with |log₂ fold change| >1 and adjusted P<0.05 were defined as differentially expressed genes, and enriched terms with adjusted P<0.05 were considered statistically significant.

A total of three independent experiments were performed for each assay (n=3) to guarantee the reliability and reproducibility of the experimental findings. Experimental data are presented as the mean ± SD, which is used to illustrate the central tendency and degree of dispersion of the results. For comparisons among multiple groups, one-way ANOVA was employed and Tukey's post hoc test was subsequently conducted to identify the specific differences between individual groups. The paired Student's t-test was utilized for comparisons between two groups. Pearson's correlation analysis was applied to evaluate the linear correlation between relevant indicators. All statistical analyses were carried out with SPSS (version 26.0; IBM Corp.) software with P<0.05 considered to indicate a statistically significant difference.

To comprehensively characterize HNRNPAB expression patterns across diverse tissues of patients with BRCA and its association with clinicopathological features, the standardized pan-cancer dataset TCGA TARGET GTEx was retrieved (PANCAN; n=19,131; genes=60,499) from the UCSC Xena browser. HNRNPAB was found to be significantly upregulated in 24 tumor types, whereas downregulation was detected in 2 tumor types (Fig. 1A). Differential analysis of normal and tumor tissues from TCGA and GTEx databases further demonstrated that HNRNPAB expression was significantly higher in BRCA tissues compared with normal breast tissues (Fig. 1B). Immunohistochemical data of HNRNPAB from the HPA database demonstrated that at the protein level, HNRNPAB expression was elevated in BRCA tissues compared with normal breast tissues (Fig. 1C and D). Overall survival analysis revealed that high HNRNPAB mRNA expression was closely associated with unfavorable prognosis in patients with BRCA (Fig. 1E). As breast cancer progressed to advanced stages, the median expression level of HNRNPAB exhibited a declining trend (Fig. 1F). The number of lymph node metastases in patients with breast cancer showed a positive association with increased median HNRNPAB expression (Fig. 1G). Patients with BRCA aged 21-40 years displayed a marked increase in median HNRNPAB expression, with elevated levels also observed in other age groups compared with normal tissue (Fig. 1H). HNRNPAB expression was found to be notable higher in invasive ductal carcinoma and medullary carcinoma compared with normal tissues, while a mild elevation was noted in other histological subtypes (Fig. 1I). In addition, Pearson's correlation was calculated between HNRNPAB expression and ploidy across a number of tumors and thus identified a significant positive correlation in BRCA. The expression of HNRNPAB positively correlated with tumor ploidy in breast cancer, suggesting that further investigation is warranted (Fig. 1J).

Pre-transfection with siRNA targeting HNRNPAB effectively reduced HNRNPAB levels in MDA-MB-231 cells, leading to a significant decrease in intracellular HNRNPAB content (Fig. 2A). Proliferation, migration and invasion assays were performed to compare two groups, namely the MDA-MB-231 siRNA control group and the MDA-MB-231 HNRNPAB-siRNA group. The CCK-8 assay results indicated that knockdown of HNRNPAB expression significantly inhibited the proliferation of breast cancer cells (Fig. 2B). Transwell migration and invasion experiments further demonstrated that HNRNPAB promoted a significant decrease in cancer cell migration and invasion (Fig. 2C and D).

To explore the genetic architecture and transcriptional variations of HNRNPAB, the cBioportal database was utilized. With regard to genetic alterations of HNRNPAB, 1.3% of patients exhibited modifications, with amplification being the predominant type of alteration in cancer (Fig. 3A). Among varying breast cancer subtypes, invasive ductal carcinoma exhibited the highest frequency of HNRNPAB mutations, followed by invasive lobular carcinoma. By contrast, invasive mucinous breast carcinoma and breast invasive carcinoma (not otherwise specified) displayed minimal HNRNPAB mutation rates (Fig. 3B). Analysis of the breast tumor cohort landscape revealed that distinct HNRNPAB expression levels were associated with specific genetic alterations. The high HNRNPAB expression group exhibited significant differences in the mutation status of PIK3CA, titin (TTN) and tumor protein p53 (TP53), whereas the low expression group exhibited higher mutation frequencies of these three genes. Therefore, mutations in PIK3CA, TP53 and TTN may affect the progression of breast cancer regulated by HNRNPAB (Fig. 3C and D). In the high HNRNPAB expression group, C>G mutations ranked second and C>A mutations ranked third. Conversely, in the low expression group, C>A mutations were the second most common, surpassing C>G mutations (Fig. 3E). Differences were also observed in variation classifications between the two groups: In the high HNRNPAB group, ‘Nonstop_Mutation’ ranked fifth (surpassing ‘Translation_Start_Site’ mutations at sixth), while in the low expression group, ‘Translation_Start_Site’ mutations occupied the fifth position (surpassing ‘Nonstop_Mutations’ at sixth; Fig. 3F). These discrepancies may be attributed to altered HNRNPAB expression; however, further studies are required to clarify their impact on breast cancer progression, which will enhance the general understanding of HNRNPAB function.

HALLMARK enrichment analysis suggested that HNRNPAB may participate in ‘OXIDATIVE_PHOSPHORYLATION’, ‘G2M_CHECKPOINT’, ‘PI3K_AKT_MTOR_SIGNALING’, ‘FATTY_ACID_METABOLISM’ and ‘WNT_BETA_CATENIN_SIGNALING’ (Fig. 4A). KEGG enrichment analysis indicated that HNRNPAB is primarily involved in ‘CELL_CYCLE’, ‘DNA_REPLICATION’, ‘OXIDATIVE_PHOSPHORYLATION’, ‘GLUTATHIONE_METABOLISM’ and the ‘ERBB_SIGNALING_PATHWAY’ (Fig. 4B). REACTOME enrichment analysis results demonstrated that HNRNPAB may be involved in ‘CELL_CYCLE’, ‘ABC_TRANSPORTER_DISORDERS’, ‘SIGNALING_BY_WNT’, ‘HSP90_CHAPERONE_CYCLE_FOR_STEROID_HORMONE_RECEPTORS_SHR’ and ‘REGULATION_OF_CHOLESTEROL_BIOSYNTHESIS_BY_SREBP_SREBF’ (Fig. 4C).

ESTIMATE algorithm analysis results showed that the StromalScore, ImmuneScore and ESTIMATEScore of the low HNRNPAB expression group was significantly higher compared with those of the high expression group (Fig. 5A). Although reduced immune cell infiltration was observed, the present findings revealed a phenomenon whereby the abundance of ‘Monocytic_lineage’ cells increased with elevated HNRNPAB expression (Fig. 5B). Correlation analysis between HNRNPAB and immune checkpoint gene expression levels indicated that HNRNPAB was significantly positively correlated with 11 immune checkpoint genes and negatively correlated with 3 immune checkpoint genes (Fig. 5C). IPS data of patients with BRCA from TCIA database demonstrated that the low HNRNPAB expression group exhibited a higher IPS and improved responsiveness to immune checkpoint inhibitors (Fig. 5D).

This result suggests that HNRNPAB downregulation is associated with enhanced immunotherapeutic potential, highlighting an important prognostic and predictive implication for BRCA immunotherapy.

Based on the aforementioned enrichment results, it was hypothesized that HNRNPAB participates in biological processes such as cell cycle regulation, DNA replication and metabolic control, all of which are associated with stem cell functions. Therefore, the standardized TCGA pan-cancer dataset was retrieved from the UCSC Xena browser. Specifically, HNRNPAB expression data was extracted from each sample and the RNA stemness score for each tumor was calculated based on mRNA signatures. Pearson correlation coefficients were computed for each tumor type, revealing significant correlations in 26 tumors. Notably, 24 tumor types (including breast cancer) exhibited a positive correlation, while 2 tumor types exhibited a negative correlation (Fig. 6A). In addition, the correlation between HNRNPAB and common breast cancer stem cell markers was explored, identifying significant positive correlations with SOX2, integrin subunit β-1 (ITGB1), epithelial cell adhesion molecule (EPCAM) (Fig. 6B-D). Collectively, these preliminary results suggested that HNRNPAB may positively regulate breast cancer stem cells.