miR‐4732‐5p promotes breast cancer progression by targeting TSPAN13

Abstract MiR‐4732‐5p was previously found to be dysregulated in nipple discharge of breast cancer. However, the expression and function of miR‐4732‐5p in breast cancer remain largely unknown. Here, the expression of miR‐4732‐5p was detected using quantitative real‐time PCR in breast cancer tissues and cell lines. Cell proliferation, apoptosis, migration and invasion assays were performed to examine the effects of miR‐4732‐5p in breast cancer. In addition, mRNA sequencing, bioinformatics analysis, Western blot and luciferase assays were performed to identify the target of miR‐4732‐5p. Overall, miR‐4732‐5p was significantly down‐regulated in breast cancer tissues, especially in lymph node metastasis (LNM)‐negative tissues, compared with adjacent normal tissues. However, it was more highly expressed in LNM‐positive breast cancer tissues, compared with LNM‐negative ones. Expression of miR‐4732‐5p was positively correlated with lymph node metastasis, larger tumour size, advanced clinical stage, high Ki‐67 levels and poor prognosis. MiR‐4732‐5p promoted cell proliferation, migration and invasion in breast cancer. MiR‐4732‐5p directly targeted the 3′‐UTR of tetraspanin 13 (TSPAN13) and suppressed TSPAN13 expression at the mRNA and protein levels. These results suggested that miR‐4732‐5p may serve as a tumour suppressor in the initiation of breast cancer, but as a tumour promoter in breast cancer progression by targeting TSPAN13.

miR-4732-5p in breast cancer. In addition, mRNA sequencing, bioinformatics analysis, Western blot and luciferase assays were performed to identify the target of miR-4732-5p. Overall, miR-4732-5p was significantly down-regulated in breast cancer tissues, especially in lymph node metastasis (LNM)-negative tissues, compared with adjacent normal tissues. However, it was more highly expressed in LNM-positive breast cancer tissues, compared with LNM-negative ones. Expression of miR-4732-5p was positively correlated with lymph node metastasis, larger tumour size, advanced clinical stage, high Ki-67 levels and poor prognosis. MiR-4732-5p promoted cell proliferation, migration and invasion in breast cancer. MiR-4732-5p directly targeted the 3′-UTR of tetraspanin 13 (TSPAN13) and suppressed TSPAN13 expression at the mRNA and protein levels. These results suggested that miR-4732-5p may serve as a tumour suppressor in the initiation of breast cancer, but as a tumour promoter in breast cancer progression by targeting TSPAN13.

K E Y W O R D S
breast cancer, metastasis, miR-4732-5p, proliferation, TSPAN13 the processes of invasion and metastasis in several types of cancer. 7,8 MiR-10b and miR-335 were the first miRNAs to be reported as promoters or inhibitors of metastasis respectively. 9 In breast cancer, certain miRNAs have been associated with the invasionmetastasis cascade, including miR-10b, miR-21, miR-31, miR-373 and miR-520c. 10 MiR-4732-5p is a novel miRNA that has not been well studied in human cancers. In our previous study, miR-4732-5p was found to be down-regulated in breast cancerous nipple discharge compared with benign nipple discharge, supporting that it might serve as a potential tumour biomarker for breast cancer detection. 11 However, the expression and function of miR-4732-5p remain largely unknown.
Here, we provide evidence that miR-4732-5p was downregulated in lymph node metastasis (LNM)-negative breast cancer tissues; however, it was overexpressed in LNM-positive tissues.

| Celltransfection
The miR-4732-5p mimics and negative control were obtained from GenePharma (Shanghai, China). MiRNA mimics are small, chemically modified, double-stranded RNA molecules that mimic endogenous mature miRNA molecules and widely used for short-term gain-offunction miRNA experiments. 12 MiRNA mimics and negative control transfection was carried out with the X-tremeGENE transfection reagent (Roche) according to the manufacturer's instructions. For stable expression of miR-4732-5p, cells were transfected with lentivirus miR-4732-5p-expressing vector LV3-hsa-miR-4732-5p or a negative control LV3NC (GenePharma, Shanghai, China), which expressed GFP (green fluorescent protein) as a marker. The expression of GFP was monitored under a fluorescence microscope. The transfected cells were screened with puromycin for the ones stably expressing miR-4732-5p.

| Colonyformationassay
Cells (300-500 per well) were seeded in six-well plates. After incubation for 2 weeks, colonies were fixed and stained.

| Migrationandinvasionassay
Cell migration capability was detected by using Transwell chambers (Corning, NY, USA) as previously described. 4 Transwell chambers pre-coated with the Matrigel matrix (BD Biosciences) were used for cell invasion assay.

| mRNAsequencing
Total RNA from miR-4732-5p or negative control-transfected cells (n = 3 for each) was extracted using Trizol reagent (Invitrogen) and was subsequently used for removing the rRNAs using Ribo-Zero

| Luciferaseassay
The wild-type (WT) TSPAN13 3′-UTR fragments or the mutant (without the predicted miR-4732-5p target sequence), were in- Infinite M200 Pro instrument. The luminescence intensity of firefly luciferase was normalized to that of Renilla luciferase.

| Westernblot
Cells were harvested and protein extracts were obtained with lysis buffer. Equal amounts of protein were electrophoresed on 10% SDS-PAGE gels, and then transferred to PVDF membrane. After blocking, the membranes were incubated with anti-TSPAN13 (

| Statisticalanalysis
Differences between two groups were analysed by Student's t test, and ANOVA was used to find differences among three or more groups. Two-sided P < 0.05 was considered significant. The webtools miRpower, 14 BioProfiling 15, 16 and PrognoScan 17 were utilized to determine the prognostic value of miR-4732-5p and TSPAN13, respectively, in breast cancer using publicly available data.
As breast cancer is rather heterogeneous, the relationship be- The prognostic role of miR-4732-5p was determined by the miRpower database. 14 Our results showed that patients with high was also associated with worse prognosis in ER+, PR+ and HER2+ patients ( Figure 3B-D), suggesting that miR-4732-5p may be a risk prognostic factor independent of breast cancer subtypes.

| Tetraspanin13(TSPAN13)wasidentifiedasa directtargetofmiR-4732-5p
To identify genes targeted by miR-4732-5p, we performed genomewide mRNA sequencing analysis using MDA-MB-231 cell line and selected down-regulated genes (fold change > 2 and P < 0.05) by miR-4732-5p transfection compared with negative control transfection ( Figure 4A). Among 30 putative candidate genes, two genes (TSPAN13 and CDK18) contained putative binding sites of miR-4732-5p in their 3′-UTR regions, as revealed by the TargetScan database ( Figure 4B). TSPAN13 was of particular interest in light of its reported roles in suppressing cancer cells progression. 18,19 qRT-PCR and Western blot were performed to investigate whether miR-4732-5p altered TSPAN13 gene and protein expression. The Next, luciferase assay was performed to determine whether TSPAN13 mRNA carried a target site for miR-4732-5p. The TargetScan database predicted that one putative miR-4732-5p binding site existed in the 3′-UTR of the gene (position 151-157). Data showed that the luminescence intensity was significantly reduced by co-transfection with miR-4732-5p and the vector carrying the wild-type 3′-UTR of TSPAN13, rather than the mutant 3′-UTR with deleted target sites ( Figure 4F).
Furthermore, the prognostic value of TSPAN13 was determined using publicly available data. As shown in BioProfiling, 15,16 low expression of TSPAN13 predicted poor survival of breast cancer patients (GEO database: GSE250665, n = 507, P = 0.000546, Figure 4G). Results from the PrognoScan database 17 showed that breast cancer patients with low expression of TSPAN13 had shorter survival time (GSE12276 dataset, n = 204, P = 0.001299, Figure 4H) than those with high TSPAN13 expression.

| D ISCUSS I ON
MiRNAs act as key regulators in cancer initiation and progression, including cell proliferation, apoptosis, differentiation, invasion, metastasis and drug resistance. 3,4,7,10,20 MiRNAs play roles as either tu- In this study, miR-4732-5p was found to be down-regulated in breast cancer tissues compared with adjacent normal tissues.
However, increased expression of miR-4732-5p was found in cancer tissues with lymph node metastasis, larger tumour size, high Ki-67 index and advanced clinical stage, independent of cancer subtypes.
High expression of miR-4732-5p was associated with poor prognosis in breast cancer. MiR-4732-5p significantly promoted cell proliferation, colony formation, migration and invasion in breast cancer cells.
Mechanistically, TSPAN13 was identified as a direct target of miR-4732-5p and low expression of TSPAN13 was associated with poor survival of breast cancer patients.
Till now, reports on miR-4732-5p in human cancers were scarce. P38ɑ suppresses inflammation-associated epithelial damage and tumourigenesis but contributes to the proliferation and survival of colon cancer cells. 25 In breast cancer, TGFβ inhibits cell growth at early stages of carcinogenesis, but acts as an aggressive oncogene in more advanced malignant stages. 26 We comment that the "dual role" of miR-4732-5p in breast cancer needs further investigation.
TSPAN13 (also known as NET-6) belongs to the tetraspanin superfamily, which has been implicated in a range of biological processes, including differentiation, motility, proliferation and metastasis. 27 The TSPAN13 gene is located on chromosome 7p21.1 and encodes a 204 amino acid protein with a predicted molecular weight of 24 kDa. 28 Geradts et al have found that expression level of TSPAN13 was lowest in breast carcinomas with aggressive characteristics 18 and TSPAN13 inhibited growth and invasion, as well as increased apoptosis of breast cancer cells in vitro and in vivo. 19 The downstream effectors of TSPAN13 in breast cancer have been reported to include MMP-1, MMP-3, P53, Bax, Bak and Caspase 3. 19 However, the upstream regulators of TSPAN13 have never been determined. Thus, our current findings provide evidence to improve characterization of the miR-4732-5p/TSPAN13 regulatory axis in breast cancer.
In summary, we demonstrated that miR-4732-5p was down-