The impact of hsa-miR-1972 on the expression of von Willebrand factor in breast cancer progression regulation

Data sources

The sequencing dataset for the breast invasive carcinoma project (TCGA-BRCA), which contained 113 normal samples and 1,118 tumor samples, were downloaded from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/). The annotation and clinical information of the samples were acquired based on the rowData and colData functions in the TCGAbiolinks package (Colaprico et al., 2016) (parameters: data.category = “transcriptome analysis”, data.type = “gene expression quantification”, and workflow.type = “STAR - Count”). Subsequently, annotations and clinical data related to the samples were extracted using the rowData and colData functions.

Cancer migration scores

In order to identify the genes associated with BC cell migration, the cancer migration scores were calculated by ssGSEA (van Roosmalen et al., 2015) based on the set of cancer migration-associated genes from collected from van Roosmalen et al. (2015) The expression matrix of for these gene sets in each sample in the TCGA-BRCA expression was transformed by log2 (FPKM+1) to calculate the enrichment scores.

Gene co-expression network development

For screening the critical modules related to BC cell migration, weighted gene co-expression networks were developed based on TCGA-BRCA gene expression data using the WGCNA package (Langfelder & Horvath, 2008). The expression matrix was preprocessed using the log2 (FPKM + 1) transform and then the outliers were removed by hierarchical clustering. The similarity matrix was then constructed using Pearson correlation and transformed to an adjacency matrix using a power function to ensure a scale-free nature of the network. The optimal soft-threshold power β was determined to satisfy the scale-free topology criterion (correlation of log (k) with p(k) ≥ 0.85) to convert adjacency matrix into a topological overlap matrix (TOM). Hierarchical clustering of highly similar gene modules was performed, followed by using Pearson correlation analysis to identify modules with the highest correlation to BC cell migration. The clusterProfiler package (version 4.8.2) was used to analyze the biological functions of genes within the target modules with the parameters of keyType = “ENTREZID”, pvalueCutoff = 0.05 and qvalueCutoff = 0.1 (Yu et al., 2012; Song et al., 2023).

Development of protein-protein interaction network and identification of hub genes

The key hub genes were identified by protein-protein interaction (PPI) network. After excluding disconnected nodes, genes from each module were entered into the STRING database (https://string-db.org/) (confidence > 0.4) to construct a PPI network (Szklarczyk et al., 2023). MCODE algorithm and cytoHubba in the Cytoscape (version 3.8.0) were used to for cluster identification and calculating the network connectivity. The top 10 genes showing the highest connectivity were defined as hub genes (Shannon et al., 2003).

Validation of hub genes

Association of hub genes with clinical outcomes and prognosis of BC was assessed by Kaplan-Meier (KM) survival analysis. Based on the optimal cutoff value determined by the surviviner package, patients were divided into high- and low-expression groups and survival curves were plotted using the survminer package (Kassambara et al., 2017).

Prediction of miRNAs regulating the hub genes

The miRNAs that targeted identified hub genes were predicted based on three databases, including miRDB (https://mirdb.org/), miRanda (http://www.microRNA.org), and TargetScan (https://www.targetscan.org/vert_80/) were used to predict miRNAs. The results from the three databases were integrated to determine the miRNA-target gene pairs.

Cell culture and transfection

DMEM (Gibco, Grand Island, NY, USA) supplemented with fetal bovine serum (Gibco, Grand Island, NY, USA) and penicillin/streptomycin was used to culture human BC cell lines (MDA-MB-231 and MCF-7) in 5% CO₂ at 37 °C. The two cell lines were purchased from Typical Culture Reserve Center of China (Shanghai, China). The si-RNA and oe-RNA of VWF (Sangon, China) and negative control as well as miR-1972 mimic or its control were transfected into the cells using Lipofectamine 2000 (Invitrogen, Waltham, MA, USA). See Table 1 for the sequences of small interfering RNAs.

Cell viability

The effect of hsa-miR-1972 mimic on the growth of BC cells (MDA-MB-231 and MCF-7) was assessed by cell viability assay. The cells (1 × 10³ cells) with different treatments were cultured in each well of 96-well plates. Cell Counting Kit-8 (CCK-8) solution (Beyotime, Jiangsu, China) was applied at specific time points. After cell incubation at 37 °C for 2 h (h), the OD₄₅₀ value was detected using a microplate reader (Thermo Fisher, Waltham, MA, USA).

Western blot

Western blot assay was applied to evaluate the protein expression of VWF in BC cells. Total protein was isolated using RIPA buffer and then quantified. Protein sample was separated by 10% SDS-PAGE, then moved onto a PVDF membrane (0.45-μm) and blocked with 5% skimmed milk for 2 h. The corresponding primary antibodies (anti-VWF (11778-1-AP, 1/1,000) and anti-GAPDH (ab8245, 1/5,000)) were incubated with the membrane at 4 °C overnight. The secondary antibody was incubated with the protein at room temperature for 1 h. Immunoblots were developed using an ECL fluorescence detection kit (Beyotime, Shanghai, Jiangsu, China) and visualized with a Tanon 4,600 system (Tanon Science and Technology Co., Ltd., Shanghai, China).

QRT-PCR

Total RNA from MCF-7 and MDA-MB-231 cell lines was extracted applying TRIzol (Thermo Fisher, Waltham, MA, USA) reagent. The cDNA was extracted from 500 ng of RNA with HiScript II SuperMix (Vazyme, Nanjing, China). PCR amplification was performed under conditions of 10 min at 94 °C, 10 seconds (s) at 94 °C and 45 s at 60 °C. GAPDH served as an internal reference. Table 2 displayed the sequence list of primer pairs for the targeted genes. The relative expression levels were calculated with the 2^−ΔΔCT method (Sindhuja, Amuthalakshmi & Nalini, 2022).

Dual-luciferase reporter assay

The specificity of the interaction between hsa-miR-1972 mimic and VWF was detected by the dual-luciferase reporter assay. The coding sequence (CDS) of VWF containing mutant or wild-type (wt) was ligated into the firefly luciferase expression vector pMIR-REPORT (Sangon, Shanghai, China). Transfection of VWF-mut and VWF-wt was conducted applying Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) together with miR-NC with miR-mimic (Sangon, Shanghai, China). The pTK-Renilla luciferase reporter (Genomeditech, Inc., Shanghai, China) containing the full-length Renilla luciferase (which was used to normalize the luciferase activity) was co-transfected into the cells and the fluorescence intensity was quantified 48 h after transfection with Dual-Luciferase Assay System (Promega, Inc., Madison, Wisconsin, US).

Transwell

Transwell assay was performed to evaluate the cell migration ability of MDA-MB-231 and MCF-7. The cells (5 × 10⁴) were inoculated into the upper chamber of the Transwell added with serum-free medium, while complete DMEM medium were added to the lower layer. After incubation for 24 h, the cells were fixed using paraformaldehyde and dyed by 0.1% crystal violet.

Enzyme-linked immunosorbent assay

The concentrations of angiogenic factors, VEGF, bFGF, TGF-β, and HIF-1α, were detected using enzyme-linked immunosorbent assay (ELISA) kits following the manufacturer’s instructions (Elabscience). The total protein concentration of each sample was then determined using the Bradford assay (Cat. ab119216; Abcam, Cambridge, UK) and concentrations were shown in the form of pg/mL.

Statistical analysis

All measurement data were represented as mean ± standard deviation and all experiments were performed in independent triplicates. Student’s t-test was used to compare differences in continuous variables between two groups, and all the computational analyses were conducted in R (version 4.3.1). GraphPad Prism (version 8.0.2) was applied for the statistical analysis. A p < 0.05 was considered statistically significant.

WGCNA analysis and identification of the hub genes

The ssGSEA and WGCNA were conducted based on the cancer migration-related gene set in the samples in TCGA-BRCA, noticeably, the MEblue module was positively correlated with cell migration, the MEmagenta and MEpink modules were correlated with cell migration enhancement and facilitation, and the MEfloralwhite module was negatively correlated impaired cell migration (Fig. S1). To further explore the functional roles of these modules in BC development, enrichment analysis was performed to analyze their biological processes. It was found that the genes in the MEblue module were mainly involved in mRNA processing, autophagy and TOR signaling (Fig. 1A), whereas genes in the MEmagenta and MEpink modules were mainly enriched in biological processes involving the migration of multiple cell types (Figs. 1B, 1C). In addition, a PPI network was developed based on the genes in each module to further screen key hub genes. Specifically, MRPL20, TUFM, MRPL12, MRPL43, MRPL21, MRPL41, AURKAIP1NDUFB7, MRPS34, and ATP5F1D were the hub genes in the MEblue module (Fig. 1D); MRPL13, POLR2K, COPS5, ELOC, NUDCD1, TATDN1, ENY2, DCAF13, MTERF3, and UBE2V2 were the hub genes in the MEfloralwhite module (Fig. 1E); and COL5A1, COL4A1, COL6A2, COL6A1, BGN, COL6A3, FBN1, POSTN, DCN, and LUM were the hub genes in the MEmagenta module (Fig. 1F); KDR, PECAM1, CDH5, FLT4, VWF, TIE1, GAPDH, FGF7, SOX18, and GJA4 were the hub genes in the MEpink module (Fig. 1G). Taken together, our study revealed the expression patterns of the set of genes related to BC cell migration and their potential functions in the cancer progression, especially in cell migration and signaling pathways.

Hub genes associated with BC prognosis

To further explore the impact of the hub genes on the prognosis of BC samples in the TCGA-BRCA cohort, we performed survival analysis for all the hub genes. KM survival curves showed that low-expressed MRPL20, MRPL12, AURKAIP1, NDUFB7, ATP5F1D, and BGN were in BC patients were correlated with a poor survival trend, whereas low-expressed COL4A1 and VWF were related to a significant survival advantage (Figs. 2A–2H).

Prediction of miRNAs regulating the hub genes

The miRDB, miRanda, and TargetScan databases were used to predict the miRNAs that targeted the key hub genes MRPL20, MRPL12, AURKAIP1, NDUFB7, ATP5F1D, BGN, COL4A1, and VWF. By combining the prediction results from the three databases, we obtained 51 intersections of miRNAs with target gene pairs (Table S1, Fig. 3A). Detailed information of these miRNA-target gene pairs was shown in Fig. 3B. Among these hub genes, COL4A1 is responsible for encoding the α1 chain of type IV collagen, which is a key component in the formation of basement membrane. By interacting with other components in the basement membrane, type IV collagen α1 chain promotes stable adhesion between tumor cells and the basement membrane, thereby facilitating tumor cell migration and invasion in vivo (Zhang, Wang & Ding, 2021). In addition, VWF encoded by the VWF gene is an important plasma protein synthesized by endothelial cells and megakaryocytes, which promote vascular endothelial cell proliferation and migration to favor neointima formation and also increase platelet adhesion and aggregation by binding to the GPIb-IX-V complex on the surface of platelets (Fan et al., 2015). MRPL20 contributes to tumor cell migration and invasion by affecting the energy metabolism and proliferation ability of tumor cells through regulating the function of mitochondria (Zietzer et al., 2022). BGN encodes acetylheparin sulfate proteoglycan as a cell surface glycoprotein in the processes of cell adhesion, migration, and proliferation, and it also forms a stable adhesion structure through the interactions with other components in the basement membrane, thereby promoting the adhesion of tumor cells to the basement membrane and facilitating the migration and invasion of tumor cells in vivo (Haupt et al., 2009). It was found that hsa-miR-1972 was one of the important miRNAs that targeted VWF (Fig. 3B), indicating that there may be a potential association between hsa-miR-1972 and VWF regulation in BC progression.

Impact of hsa-miR-1972 mimic on the gene expression and BC cell behaviors

Based on the above findings, the effects of hsa-miR-1972 mimic on the expression of several genes involved in the pathogenesis of BC and on the proliferation of MDA-MB-231 and MCF-7 cells were explored. Expression analysis showed that in MCF-7 cells, the mRNA expression of MAP2K1, MMP9, NEK1 and SHC1 was decreased following the transfection of miR-1972 mimic (Fig. 4A), while miR-1972 mimic caused the down-regulated expression of CDH1, ITGB8, MMP9, NEK1 and SHC1 in MDA-MB-231 cells (Fig. 4B). Hsa-miR-1972 had the ability to target and regulate critical genes involved in BC progression and metastasis. In addition, migration and proliferation assays demonstrated that hsa-miR-1972 mimics noticeably suppressed the growth and migration of MDA-MB-231 and MCF-7 cell lines over time, demonstrating a potential therapeutic value in suppressing BC cell proliferation (Figs. 5A–5F).

Effects of hsa-miR-1972 mimic on the expression of angiogenic factors and VWF regulation in BC cells

It was observed that hsa-miR-1972 mimics effectively downregulated the expression of angiogenic markers, including VEGF, bFGF, TGF-β, and HIF-1α, suggesting a potent inhibitory effect on angiogenesis-related pathways (Figs. 6A–6H). In addition, Western blot confirmed a lowered protein expression of VWF in BC cells when treated with hsa-miR-1972 mimics, highlighting a direct regulatory effect of hsa-miR-1972 on VWF (Figs. 7A, 7B). The specificity of the interaction was verified by luciferase reporter gene assay as luciferase activity was significantly reduced in the presence of the hsa-miR-1972 mimic in the VWF-WT construct but not in the mutant construct (Fig. 7C). Furthermore, the significant reduction of VWF expression in MDA-MB-231 and MCF-7 cell lines suggested the potential of miRNAs as therapeutic agents to inhibit VWF-mediated pathways in BC (Figs. 7D, 7E). Collectively, these findings elucidated the role of hsa-miR-1972 mimics in regulating angiogenesis and VWF, providing novel insights for their application in BC therapeutic strategies.

Modulation of the gene expression and the levels of angiogenic factors by manipulating VWF in BC cells treated with miRNA mimic

The effect of VWF interacting with hsa-miR-1972 on BC progression was investigated by constructing plasmids to inhibit and overexpress VWF and transfecting them into BC cell lines overexpressing hsa-miR-1972. The results showed that in cells overexpressing hsa-miR-1972 with inhibited VWF expression, the expression of tumor malignancy indicators was significantly suppressed. Specifically, all the six indicators of malignancy in the two BC cell lines were significantly downregulated after inhibiting VWF expression. However, five malignancy indicators were re-expressed when VWF was also overexpressed after overexpressing hsa-miR-1972 in the MCF-7 cell line, while four of them were upregulated in the MDA-MB-231 cell line (Figs. 8A, 8B).

Similarly, changes in the levels of pro-angiogenic factors were consistent with changes in malignant indicators. Under the conditions of overexpression of hsa-miR-1972 and inhibition of VWF expression, the concentration of pro-angiogenic factors was significantly reduced, whereas the expression of these pro-angiogenic factors was significantly elevated when VWF expression was enhanced (Figs. 9A, 9B). The results from Western blot further verified that the VWF expression levels were successfully regulated (Figs. 9C, 9D). Moreover, the results of Transwell assay also supported the above findings as inhibition of VWF expression significantly suppressed the migratory ability of BC cell lines and overexpression of VWF restored the cellular migratory ability inhibited by hsa-miR-1972 (Figs. 9E–9G). In conclusion, these data showed that the interaction between VWF and hsa-miR-1972 played a critical part in BC development, and that its regulation may provide new targets for BC therapy.