br Results br Coordinated expression of pri miR
3.1. Coordinated expression of pri-miR-3614 and its host gene TRIM25 in breast cancer
By conducting The Cancer Genome Atlas (TCGA) database search, we determined that TRIM25 mRNA expression is markedly increased in breast tumor tissues (P b 0.01, n = 389) compared to that in hyper-plastic (non-tumor) tissues (n = 61) (Fig. S1a). Next, we quantified the TRIM25 mRNA expression levels in BC samples by qRT-PCR analy-sis. Consistent with the data from the TCGA analysis, our results indi-cated that TRIM25 mRNA was upregulated in BC tissues (n = 30) compared to that of non-tumor tissues (P b 0.05, n = 30) (Fig. 1a). We also investigated TRIM25 protein expression levels by
immunohistochemistry. We observed that TRIM25 protein was local-ized in the nucleus and Tunicamycin of BC cells, and found that its expression was upregulated by 83.3% in cancer tissues compared to its expression in hyperplastic tissues (Fig. 1b). Western blot analysis also showed that TRIM25 expression was enriched in BC cell lines compared to the control mammary epithelial cell line, HBL-100 (Fig. 1c). Moreover, the expression of TRIM25 mRNA and protein levels in estrogen-receptor-positive (ER+) BC samples was higher than that of the ER− BC samples (Fig. 1b and c).
As pri-miR-3614 and its host gene TRIM25 are both located on chro-mosome 17q22 and share the same promoter, we suspected that they display similar expression patterns. First, we examined the expression levels of pri-miR-3614 in BC tissues and cell lines by qRT-PCR. In BC and hyperplastic (non-tumor) tissues obtained from 60 BC patients, pri-miR-3614 expression was upregulated (median fold change: 1.5-fold) in 23/30 of the tumor samples, but no significant difference was observed between the ER+ BC and ER− BC samples (Fig. 1d). This result was further validated in BC cell lines. Pri-miR-3614 expression was higher in the MCF-7 and MDA-MB-231 cancer cell lines compared to the HBL-100 control cell line (Fig. 1e). Next, we examined the levels of mature miR-3614-3p and miR-3614-5p by qRT-PCR. The expression of miR-3614-5p was almost undetectable, likely because miR-3614-3p maturation is prevalent in breast cancer. Unexpectedly, there was no significant difference between the levels of mature miR-3614-3p in BC tissues compared to that of non-tumor tissues, whereas it was found to be significantly lower in BC cell lines (Fig. 1f and g). r> To further investigate the roles of TRIM25, pri-miR-3614, and miR-3614-3p expression in BC, clinicopathologic factors were analyzed in relation to their expression levels in BC samples (Table S1). However, the expression of TRIM25, pri-miR-3614 and miR-3614-3p was not as-sociated with lymph node metastasis, ER, PR (progestrone receptor), AR (androgen receptor), p53, E-cadherin, Cerb-2, or Ki67 expression.
Altogether, these results indicate that pri-miR-3614 and its host gene TRIM25 exhibit high levels of coordinated expression in BC, with the exception of mature miR-3614-3p.
Estrogens can promote breast cancer progression by increasing the transcription of proto-oncogenes and growth factors . TRIM25 can act as a primary response gene to mediate estrogen-regulated func-tions including cellular growth in human breast cancer . There-fore, we hypothesized that miR-3614 transcription may be similarly regulated by estrogen. To test this hypothesis, we treated the ER-positive MCF-7 BC cells and the ERα-negative MDA-MB-231 BC cells with 0–100 nM of estrogen (E2) for 6 h [27,28], and analyzed the levels of TRIM25, pri-miR-3614, and miR-3614-3p using qRT-PCR and western blot. As expected, we observed a significant up-regulation of TRIM25 and pri-miR-3614 expression in MCF-7 cells following the E2 (100 nM) treatment. Similarly, E2 produced the same effect on TRIM25 and pri-miR-3614 expression in MDA-MB-231 cells. However, the levels of miR-3614-3p decreased in both BC cell lines following the E2 treatment (Fig. 2a–d). Subsequently, we demonstrated that estrogen-induced gene expression was completely attenuated by an estrogen antagonist tamoxifen (TAM) in both BC cell lines (Fig. 2e–h).
These data suggest that pri-miR-3614 and its host gene TRIM25 are simultaneously induced by E2 in MCF-7 and MDA-MB-231 cells, which is partially dependent on the estrogen receptor. Pri-miR-3614
transcription in BC is driven by its host gene. However, miR-3614-3p did not occur simultaneously with TRIM25 transcription. Thus, there appears to be a mechanism that suppresses its maturation.
Next, we investigated the mechanism underlying the inhibition of miR-3614 maturation. Upon analyzing the TRIM25 mRNA sequence, we identified putative RBP binding sites, including HuR- and IGF2BP3-binding motifs. Importantly, there are two binding sites for IGF2BP3 on the TRIM25 gene that are located within the pri-miR-3614 sequence (Fig. 3a). We also analyzed the TCGA data bank and found that the IGF2BP3 level was higher in BC tissues compared to that of non-tumor tissues (P b 0.01, n = 1645), and identified a significant positive corre-lation between IGF2BP3 and TRIM25 expression (Fig. S1b and c). Next, we investigated whether IGF2BP3 participates in TRIM25-mediated post-transcriptional regulation of miR-3614. As shown in Fig. 3b–e, siRNA-mediated depletion of IGF2BP3 in MCF-7 and MDA-MB-231 cells reduced the levels of IGF2BP3 and TRIM25 mRNA to 30% and ~50–70%, respectively. A similar trend was observed for pri-miR-3614, but these differences were not statistically significant, whereas the miR-3614-3p levels increased by ~1.5–2-fold compared to that of the scrambled controls. Western blot experiments confirmed that E2 treat-ment and IGF2BP3 silencing competitively regulated TRIM25 expres-sion (Fig. 3f). We used actinomycin D to inhibit transcription and measure the decay rate of TRIM25 and found that TRIM25 mRNA has a shorter half-life in IGF2BP3-depleted cells relative to control cells