|Home | About | Journals | Submit | Contact Us | Français|
The MUC1 oncoprotein is overexpressed in most human breast cancers by mechanisms that are incompletely understood. The microRNA, miR-125b, is downregulated in breast cancer cells. The present studies demonstrate that the MUC1 3′UTR contains a site for binding of the miR-125b seed region. The results show that the MUC1 3′UTR suppresses luciferase expression and that this effect is abrogated by mutation or deletion of the miR-125b binding site. Expression of an antisense miR-125b in BT-549 breast cancer cells was associated with induction of MUC1 protein but not MUC1 mRNA levels. The antisense miR-125b also increased BT-549 cell growth by a MUC1-dependent mechanism. In addition, overexpression of exogenous miR-125b downregulated MUC1 protein and not MUC1 transcripts in ZR-75-1 breast cancer cells. Silencing of MUC1 in ZR-75-1 cells with a siRNA has been shown to promote DNA damage-induced apoptosis. In concert with these observations, miR-125b-induced decreases in MUC1 levels increased the apoptotic response of ZR-75-1 cells to cisplatin treatment. These findings indicate that miR-125b suppresses translation of the MUC1 oncoprotein and that miR-125b thereby functions as a tumor suppressor in breast cancer cells.
MicroRNAs (miRNAs) are a conserved class of small (~22 nucleotides) RNAs that posttranscriptionally regulate gene expression by interacting with the 3′ untranslated region (3′ UTR) and, in some settings, coding regions of target mRNAs.1,2 miRNAs suppress gene expression by promoting mRNA degradation or inhibiting translation.1 Early studies demonstrated that the lin-4 miRNA regulates lin-14 gene expression and pattern formation in Caenorhabditis elegans.3 Notably, expression of the mammalian homolog of lin-4, miR-125b, is aberrantly regulated in human cancers. The demonstration that miR-125b levels are decreased in breast cancers, as compared to normal breast tissue, provided support for function of miR-125b as a tumor suppressor.4-6 Moreover, the observation that expression of exogenous miR-125b in SKBR3 cells is associated with suppression of (1) ErbB2 and ErbB3 levels and (2) anchorage-dependent growth provided support for involvement of miR-125b in breast cancer progression.7 By contrast, miR-125b expression has been shown to be upregulated in pancreatic cancer8 and acute myeloid leukemia,9 suggesting that miR-125b promotes oncogenesis. In concert with these observations, overexpression of miR-125b represses p53 levels and the induction of apoptosis in human neuroblastoma cells and lung fibroblasts.10 These findings have collectively indicated that, depending on cell context, miR-125b contributes to oncogenesis or functions as a tumor suppressor.
The MUC1 oncoprotein is aberrantly overexpressed in more than 90% of human breast cancers.11 The MUC1 locus on chromosome 1q21 is frequently altered in breast carcinomas.12 In addition, transcription of the MUC1 gene is upregulated by the NF-κB and STAT pathways, which are often constitutively activated in breast cancer cells.13,14 MUC1 is translated as a single polypeptide that undergoes autocleavage into two subunits that in turn form a stable heterodimer.15 The MUC1 N-terminal subunit (MUC1-N) is the mucin component of dimer with glycosylated tandem repeats that extends extracellularly from the cell surface in a complex with the MUC1 C-terminal (MUC1-C) transmembrane subunit.15 Overexpression of MUC1 as found in breast and other carcinomas is associated with targeting of MUC1-C to the nucleus and mitochondria.15-17 Importantly, the MUC1-C cytoplasmic domain, which is sufficient to induce transformation, interacts directly with the Wnt effector β-catenin and with p53 and thereby contributes to oncogenesis.18,19 MUC1-C also binds to IKKs and RelA/p65 in activation of the NF-κB pathway and promotes transcription of the MUC1 gene in an auto-inductive loop.13,20 Other work has shown that MUC1-C interacts with STAT1/3 and, in similar auto-inductive loops, confers activation of the MUC1 gene.14 These findings have indicated that the overexpression of MUC1 in breast cancer cells is conferred, at least in part, by transcriptional activation. Of relevance to the present work, there is no available evidence to support the posttranscriptional regulation of MUC1 expression.
The present studies have identified a binding motif for miR-125b in the MUC1 3′UTR. The results demonstrate that miR-125b suppresses MUC1 expression by inhibiting translation. In concert with the downregulation of MUC1, the results show that miR-125b inhibits breast cancer cell growth and promotes their sensitivity to genotoxic anticancer agents.
The human MUC1 3′UTR of 321 base-pairs (bp) was searched for miRNA binding sites using RNAhybrid.21 The search identified a putative binding motif for miR-125b at bp 46 to 67 (Fig. 1A). The sequence 5′-UCAGGG-3′ in the MUC1 3′UTR was identified as being complementary to the core segment or “seed” region of miR-125b bases 2 through 7 (5′-CCCUGA-3′) (Fig. 1A). RT-PCR analysis of RNA from human BT-549, MDA-MB-231, ZR-75-1, and MCF-7 breast cancer cell lines confirmed detectable levels of miR-125b expression (Fig. 1B). miR-125b was also detectable in nonmalignant MCF-10A mammary epithelial cells. MUC1 mRNA was detectable in the 4 breast cancer cell lines and at somewhat lower levels in MCF-10A cells (Fig. 1C). In contrast to these results, MUC1-C protein was relatively high in ZR-75-1 and MCF-7 cells as compared to that in MCF-10A cells (Fig. 1D). Moreover, MUC1-C protein was low to undetectable in BT-549 and MDA-MB-231 cells (Fig. 1D), indicating that MUC1-C expression may be regulated at the translational level.
To determine if there is an interaction with miR-125b, the wild-type MUC1 3′UTR was cloned into the pMIR-LUC reporter (pMIR-LUC/MUC1-WT3′UTR). In addition, the MUC1 3′UTR 5′-GGGAC-3′ sequence was mutated to 5′-UUUGA-3′ (pMIR-LUC/MUC1-MUT3′UTR) or deleted (pMIR-LUC/MUC1-DEL3′UTR) (Fig. 2A). Expression of pMIR-LUC/MUC1-WT3′UTR in BT-549 cells was substantially decreased compared to that obtained with the pMIR-LUC vector (Fig. 2B). By contrast, expression of pMIR-LUC/MUC1-MUT3′UTR and pMIR-LUC/MUC1-DEL3′UTR was comparable to that found with pMIR-LUC (Fig. 2B). Similar results were obtained when MCF-10A cells were transfected to express the vectors (Fig. 2C). These findings thus indicated that the predicted miR-125b binding motif is responsible, at least in part, for MUC1 3′UTR-mediated suppression of gene expression.
BT-549 cells were infected with a control lentivirus or pMIRZIP-125b, which expresses an antisense miR-125b. Analysis of the transduced cells by quantitative RT-PCR (qRT-PCR) demonstrated little if any effect on MUC1 mRNA levels (Fig. 3A, left). The pMIRZIP-125b-infected BT-549 cells were, however, found to have an increased level of MUC1-C protein as compared to that in control cells (Fig. 3A, right). Upregulation of MUC1-C expression was associated with an increase in growth rate (Fig. 3B). Moreover, colony formation was substantially increased in the BT-549 cells expressing the antisense miR-125b (Fig. 3C). To confirm that this response is related to the upregulation in MUC1 expression, we silenced MUC1 in the BT-549/ZIP-125b cells with a MUC1siRNA (Fig. 3D, left). Consistent with a MUC1-dependent response, downregulation of MUC1-C levels with the MUC1siRNA was associated with slowing of BT-549/ZIP-125b proliferation (Fig. 3D, right). These findings indicate that the antisense miR-125b increases MUC1 expression and thereby BT-549 cell growth.
ZR-75-1 cells were infected with a control lentivirus (ZR-75-1/vector) or one expressing miR-125b (ZR-75-1/miR-125b). Analysis of the ZR-75-1/miR-125b cells demonstrated an increase in miR-125b expression (Fig. 4A). Notably, as determined by qRT-PCR, there was no apparent effect of the overexpressed miR-125b on MUC1 mRNA levels (Fig. 4B, left). By contrast, MUC1-C protein levels were decreased in response to exogenous miR-125b, confirming that miR-125b downregulates MUC1 expression by inhibiting translation (Fig. 4B, right). Previous work has demonstrated that silencing MUC1 in ZR-75-1 cells with a MUC1siRNA increases their sensitivity to cisplatin (CDDP)–induced apoptosis.17 In concert with these observations and as determined by the percentage of ZR-75-1 cells with sub-G1 DNA content, downregulation of MUC1 expression with miR-125b was associated with an increased apoptotic response to treatment with 25 µM CDDP (Fig. 4C). These findings were confirmed in repetitive experiments and in the response to 50 µM CDDP (Fig. 4D).
The overexpression of MUC1 in breast and other cancers is conferred in part by MUC1-C auto-inductive loops involving NF-κB RelA- and STAT-mediated activation of the MUC1 promoter.13,14 The present studies provide evidence that MUC1 overexpression is also regulated posttranscriptionally by miR-125b. Expression of miR-125b is downregulated in primary breast cancers and particularly in ER+ and metastatic tumors.4,6,22 miR-125b is also downregulated to a greater extent in ErbB2-overexpressing as compared to ErbB2-negative breast cancers.5 In this regard, MUC1 is also overexpressed in ER+/ErbB2+ breast cancers.23 Our results demonstrate that miR-125b suppresses MUC1 expression in breast cancer cells by interacting with a 6-bp binding motif (5′-UCAGGG-3′) in the MUC1 3′UTR that is complementary to the miR-125b seed region. Introduction of an antisense miR-125b was associated with increases in MUC1-C protein in BT-549 breast cancer cells. In addition, overexpression of miR-125b decreased MUC1-C protein in ZR-75-1 cells. The downregulation of MUC1-C protein occurred in the absence of changes in MUC1 mRNA levels, indicating that miR-125b suppresses MUC1 translation. The ErbB2 and ErbB3 3′UTRs each contain a 5′-CUCAGGG-3′ motif that functions as miR-125b binding sites.7 However, in contrast to the present results with MUC1, introduction of exogenous miR-125b in SKBR3 breast cancer cells was associated with suppression of ErbB2 and ErbB3 expression at both the mRNA and protein levels.7 The mechanism(s) responsible for the downregulation of miR-125b in breast cancer cells is not known. In other cell types, miR-125b expression is downregulated in the inflammatory response,24 a setting linked to upregulation of MUC1 expression.13,14 In prostate cancer cells, miR-125b expression is upregulated in the response to androgen stimulation.25 Perhaps consistent with those findings, MUC1 is downregulated in androgen receptor positive as compared to negative prostate cancer cells.26
Overexpression of MUC1, as found in human breast cancers, promotes growth and protects against stress-induced cell death.15 Previous studies have shown that introduction of exogenous miR-125b in SKBR3 cells suppresses growth.7 In the present work, an antisense miR-125b was expressed in BT-549 cells, which express MUC1 mRNA but not protein. The antisense miR-125b increased both MUC1-C protein and BT-549 cell growth rate. Consistent with an effect mediated, at least in part, by increased MUC1 expression, silencing MUC1 with a siRNA partially reversed the miR-125b-induced increases in growth. In contrast to BT-549 cells, expression of endogenous MUC1-C protein is markedly high in ZR-75-1 cells. Partial silencing of MUC1 in ZR-75-1 cells with a siRNA had little effect on growth but increased the induction of apoptosis in response to CDDP and other DNA-damaging agents.17 Notably, downregulation of MUC1 in ZR-75-1 cells with miR-125b also had little if any effect on growth. Moreover, as found previously,17 miR-125b-induced decreases in MUC1 expression were associated with a significant increase in CDDP-induced apoptosis. These results are in apparent contrast to the observations that overexpression of miR-125b suppresses stress-induced apoptosis in human neuroblastoma cells and lung fibroblasts.10 However, one possible explanation for these differences could reside in the dependence of these different cell types on MUC1 for blocking the apoptotic response to stress. In this regard, ZR-75-1 cells overexpress MUC1 to promote survival,17 whereas overexpression of MUC1 has not been reported for neuroblastoma cells or lung fibroblasts. Alternatively, miR-125b functions as a tumor suppressor by downregulating MUC1 in mammary epithelial cells and not other cell types. In summary, our findings indicate that overexpression of MUC1 in human breast cancers can be attributed, at least in part, to dysregulation of miR-125b and that MUC1 represents a potential target for miRNA-directed therapies.
The human MCF-7 and MDA-MB-231 breast cancer cells (ATCC) and the 293T kidney epithelial cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat-inactivated fetal bovine serum (FBS), 100 units/mL penicillin, 100 µg/mL streptomycin, and 2 mM L-glutamine. Human ZR-75-1 and BT-549 breast cancer cells were grown in RPMI-1640 medium with 10% FBS, antibiotics, and L-glutamine. Human MCF-10A mammary epithelial cells were grown in mammary epithelial growth medium (MEGM) complete medium (Lonza). Transfection of cells with small interfering RNA (siRNA) pools (Dharmacon) was performed in the presence of Lipofectamine 2000 (Invitrogen) as described.13
The MUC1 3′UTR was cloned into the pMIR-LUC vector (Applied Biosystems). The wild-type MUC1 3′UTR in pMIR-LUC (pMIR-LUC/MUC1-WT3′UTR) was altered at the miR-125b binding site by (1) mutating 5 nucleotides (CAGGG→AGUUU) that are complementary to the miR-125b seed sequence (pMIR-LUC/MUC1-MUT3′UTR) or (2) deleting the CAGGG sequence (pMIR-LUC/MUC1-DEL3′UTR). Cells were cotransfected with the pMIR-LUC vectors and the Renilla luciferase reporter (Promega) in the presence of Lipofectamine (Invitrogen). At 72 h, the cells were lysed with passive lysis buffer and assayed for luciferase activity using the dual luciferase reporter kit (Promega).
Lentiviruses expressing miR-125b (pMIRNA-125b), an antisense miR-125b (pMIRZIP-125b), or a control scrambled hairpin vector (all from System Biosciences) were used to infect cells in the presence of 8 µg/mL polybrene for 3 to 4 h, followed by replacement with complete cell culture medium and selection in 3 µg/mL puromycin.
RNA was isolated using the RNeasy total RNA isolation kit (Qiagen). Expression of miR-125b was assessed using a small RNA-specific RT-PCR detection kit (Quantimir RT; Systems Biosciences) with the universal U6 reverse primer and a miR-125b specific (5′-AGGGACTCTGGGATTGAACACT-3′) forward primer. MUC1 expression was determined by RT-PCR as described.14 For qRT-PCR, MUC1 expression was analyzed with the SYBR GreenER qPCR SuperMix Universal kit (Invitrogen). HPRT mRNA was used as an internal control. The results are expressed as ΔCt values as determined by Ct(MUC1 mRNA)–Ct(HPRT mRNA). Expression of MUC1-C protein was determined by immunoblot analysis of cell lysates with anti-MUC1-C (Ab5; Neomarkers) and, as a control for loading, anti-β-actin (Sigma-Aldrich).
The authors acknowledge Dr. Isidore Rigoutos, IBM Thomas J. Watson Research Center, for his critical reading of the manuscript and his insightful suggestions.
The authors declared no potential conflicts of interest with respect to the authorship and/or publication of this article.
This work was supported by grants CA97098 and CA42802 awarded by the National Cancer Institute.