Dysregulated miR-183 inhibits migration in breast cancer cells

Breast cancer is a heterogeneous disease and represents a number of phenotypically diverse tumours[1]. The complex nature of breast cancer makes progression difficult to predict and challenging to manage. Up to forty percent of breast cancer patients still present with regional or distant disease progression at the time of diagnosis[2], and the overall survival rates for these patients has changed very little in the past decade despite advances in systemic therapy.

The key to optimising breast cancer management lies in early detection, effective prognostication and individualised therapy, all of which demand a comprehensive understanding of the complex molecular interactions that underlie breast cancer development and progression. Gene expression profiling has gone some way towards unlocking the heterogeneity of breast cancer; the advent of such high throughout technologies as microarray expression profiling facilitated the extensive investigation of breast cancer related genes [3–5]. Breast tumors can now be classified into major subtypes on the basis of gene expression: luminal, HER2/neu overexpressing and basal-like, and further analysis has identified additional subtypes within the original sub-groups[6].

Multigene prognostic and predictive tests have been developed, commercialized and have become established as tools in breast cancer diagnostics[7], however as yet there is little knowledge regarding the precise regulation of these genes and receptors, and the precise manner in which they drive breast cancer progression and metastasis.

Mi(cro)RNAs represent a fascinating new division of biomolecules. The analysis of their expression will help decipher complex patterns in disease states such as malignancy, and promises to improve classification of tumors such as breast cancer [8] in addition to uncovering pathogenic pathways and therapeutic avenues.

The aim of this study was to investigate the potential role of a specific miRNA miR-183 in breast cancer progression. MiR-183 is a member of a miRNA family comprised of three homologous miRNAs (miR-183, miR-182 and miR-96) that are clustered within 2-4 kb at chromosome 7q32. MiRNAs from this locus are dysregulated in diverse cancers including leukaemia, hepatic, colorectal and breast cancer[9–13]. Furthermore, downregulation of miR-183 has been shown to be associated with metastasis in lung cancer, and its' ectopic expression inhibits the invasiveness of cancer cells [14], suggesting that miR-183 plays a role in carcinogenesis or the metastatic cascade, possibly having a tumour suppressor role.

In this study we investigated the expression pattern and functional role of miR-183 in breast cancer. This was undertaken by quantitation of miR-183 in breast cancer specimens in relation to known clinicopathological parameters of breast cancer. Functional analysis was then performed by transfection of miR-183 precursors into a breast cancer cell line with low endogenous miR-183 levels. The effects of transfection were subsequently assessed on cell viability patterns, cell migration and alterations in gene and protein expression by real-time quantitative PCR and immunocytochemistry respectively.

The findings of this study support a role for miR-183 in breast cancer, particularly in the metastatic process, and identify this miRNA as a novel therapeutic target in breast cancer.

MiRNA expression patterns are altered in breast cancer as previously described[13, 21–24], and expression profiling of miRNA to classify tumors according to clinicopathological variables appears to have superior accuracy to mRNA profiling[8]. This evidence underlines the role of miRNAs in breast cancer molecular biology and provides the starting point for exploration of novel molecular mechanisms.

Quantitation of miR-183 by RQ-PCR in primary breast tumors revealed that miR-183 expression is significantly lower in ER and PR positive compared to ER and PR negative breast tumors, while miR-183 expression was significantly higher in HER2/neu positive compared to HER2/neu negative tumors. Luminal A tumors exhibited the lowest expression of miR-183. This is the first report of miR-183 dysregulation in association with hormone and HER2/neu receptor status in clinical breast tumour samples. Differential expression of miR-183 in breast cancer cell lines has been demonstrated by Sempere et al [25] using northern blotting, with decreased expression in basal-type compared to luminal-type cell lines. Interestingly, the expression of miR-182 which is co-ordinately expressed from the same genetic locus as miR-183 has been shown to be downregulated in ER positive tumors[13], this supports our finding of miR-183 dysregulation in association with ER status. The identification of dysregulated miRNAs in relation to ER and HER2/neu receptor status in breast cancer is of great clinical significance as the specific combination of receptor status has implications for the current classification and targeted therapy of breast cancer. The past decade has seen the development of a molecular taxonomy of breast cancer which has been divided into subtypes based on the expression of specific genes, including ER and HER2/neu. However, this taxonomy remains limited because clinical outcome may be variable within each subtype, suggesting the existence of further divisions that are yet to be characterised; for example, subgroups have been identified within the Luminal A and ER positive breast tumours which have distinct survival and prognostic profiles[26, 27]. One obvious explanation for the lack of homogeneity within the breast cancer subtypes is the potential dysregulation of miRNAs between and within subgroups. MiRNAs represent a new layer of molecular complexity in breast cancer and miRNA expression profiles are thought to be more accurate than mRNA at classifying tumours[8]. Their unique ability to requlate gene expression confers functionality on these molecules, and when dysregulated in breast cancer they may be responsible for altering the phenotype of breast tumours. Thus, it is important to functionally characterise miRNAs, such as miR-183 that are dysregulated in breast cancer and establish what effect this dysregulation may have on tumour behaviour.

MiR-183 is located on chromosome 7q32.3 and is part of a miRNA family comprised of three homologous miRNAs (miR-183, miR-182 and miR-96) that are co-ordinately expressed from this locus. MiR-183 family members have been shown to be upregulated in leukaemia, hepatic and colorectal cancer [9–12]. Conversely, miR-183 has been shown to be downregulated and inversely associated with metastasis in lung cancer[14]. This evidence indicates that the expression patterns of miR-183 are tissue specific, and that this miRNA may have divergent functions depending on the tissue or cell type. Such context dependence of miRNA function has previously been reported in the results of biochemical studies indicating that miRNA repression of mRNA is dependent on the specific cellular conditions[28].

There has been limited functional analysis of miR-183; it has been postulated that miR-183 may downregulate anti-apoptotic factors including forkhead transcription factor (FOX) proteins, suggesting a pro-apoptotic function for this miRNA[9, 29], however this was based on computational analysis and no experimental verification was provided.

During the course of these experiments, the first functional study of miR-183 in malignancy was reported; Wang et al[14] identified miR-183 as a potential metastasis inhibitor in lung cancer and demonstrated that over expression of miR-183 inhibited migration of lung cancer cells. They reported miR-183 induced dysregulation of expression of genes involved in migration and invasion, including VIL2. Overexpression of miR-183 in HeLa cells has also been shown to inhibit migration and invasion, however this was shown to be mediated through direct targeting of integrin β1 (ITGB1)[30], indicating that miR-183 is likely to have a number of mRNA targets through which it mediates biological effects in cancer cells.

The functional role of miR-183 in breast cancer was investigated in these experiments using a gain-of-function approach. The lowest endogenous expression of miR-183 was observed in T47D cells, which are ER and PR positive and HER2/neu negative, making these cells a suitable in-vitro representation of the Luminal A subtype of breast cancer.

The effect of miR-183 modulation on cellular functions which are frequently dysregulated in malignancy was assessed in T47D cells (ER⁺, PR⁺, HER2/neu ^-), SKBR3 cells (ER^-, PR^-, HER2/neu ⁺), and MDA-MB-231(ER^-, PR^-, HER2/neu ^-), cells which exhibit distinct hormone and HER2/neu receptor profiles. There was no effect on apoptosis or proliferation, which refutes the hypothesis of miR-183 having a pro-apoptotic function. Induced overexpression of miR-183 resulted in decreased migration of T47D cells in transwell assays, but there was no significant change in migration of SKBR3 or MDA-MB-231 cells (figure 4).

The difference in effect noted between the cell lines is not completely surprising for a number of reasons; firstly T47D cells were shown to express the lowest endogenous levels of miR-183, thus the effect of overexpression in these cells would be expected to result in a more dramatic or measurable phenotypic effect than in cells where miR-183 is expressed at higher levels endogenously. Furthermore, although the three cell lines are epithelial breast cancer cell lines, they are unrelated and are phenotypically distinct; differences in cell behaviour would thus be expected that relate to the previously mentioned cell-type specific regulatory mechanisms and actions of miRNA. As miR-183 was shown to be dysregulated in relation to ER and HER2/neu status it is not unexpected that this miRNA would behave differently depending on the cell type and environment, particularly in relation to the expression of these receptors.

The finding that miR-183 reduces migration in breast cancer cells indicates that this miRNA may play an important role in the suppression of metastases, as cell motility and migration are key steps in the metastatic cascade. These findings are in concordance with the findings of Wang et al[14] who demonstrated a similar effect in lung cancer cells. Increasing evidence supports a role for multiple miRNAs in the regulation of metastases in breast cancer[31–35]. This is hugely significant as metastasis is the overwhelming cause of mortality in breast cancer patients, but a complete understanding of its molecular and cellular determinants remains elusive[36]. Gene expression profiling studies have provided gene-sets that are predictive of breast cancer progression and prognosis[37], however the regulatory networks that establish such altered gene expression states are only now coming into focus. It is important that the targets and regulators of metastasis involved miRNAs are elucidated to improve our understanding of the specific roles played by miRNAs in the complex, multistep process of primary tumor invasion and metastasis.

To further elucidate the mechanism by which miR-183 regulates metastasis we analysed the expression of 96 breast cancer related genes using a low density array and identified candidate genes which were dysregulated following transfection of T47D cells with miR-183 (table 2). T47D cells were chosen for this part of the experiment as they had shown the greatest phenotypic change in response to miR-183 modulation. The finding of genes that are both downregulated and upregulated is in keeping with recent reports that miRNAs are differential modulators of gene expression and can activate mRNA translation in addition repression [38]. Functions of the changed genes were analysed using NetAffx Gene Ontology Mining Tool available at the Affymetrix website http://www.affymetrix.com/ and a number of the genes were found to have roles in migration and maintenance of the actin cytoskeleton integrity. These data support the hypothesis that regulation of metastasis requires coordinated expression of multiple genes. Computational prediction algorithms were used to identify which of these miRNAs may be direct targets of miR-183 via complementary binding at the 3'UTR. All three of the algorithms used (miRBase, PicTar and TargetScan) predicted VIL2/Ezrin a member of the ezrin/radixin/moesin (ERM) family of proteins that mediate cytoskeletal-membrane interactions and cell signalling[39] to be a target of miR-183 via perfect complementarity at a binding site at position 366-373 of Ezrin 3'UTR. Downregulation of VIL2 as a result of miR-183 overexpression was validated using RQ-PCR, indicating that regulation occurred at the mRNA level, this is in contrast to the findings of Wang et al[14] who found regulation of Ezrin to be at the posttranscriptional level in lung cancer. We also found changes in Ezrin protein expression in breast cancer cells following transfection with miR-183.

It is likely that much of the migration inhibition induced by miR-183 in breast cancer cells is modulated via Ezrin downregulation. Ezrin is a membrane cytoskeleton crosslinker which plays a role in cell functions including control of the actin cytoskeleton, adhesion and motility. The actin cytoskeleton is a highly dynamic network of actin polymers and proteins and is required for cell motility, cortical organisation and cell division. Ezrin has been shown to be a key component in tumor metastasis [40, 41] and is associated with metastatic potential and poor outcome in breast carcinomas[42, 43].

The assessment of Ezrin expression in breast cancer cells can be both cytoplasmic and membranous, however the differential localization, rather than total Ezrin protein levels is associated with the malignant potential of the cell. In an immunohistochemical analysis of Ezrin in primary breast tumors and lymph node metastases, expression levels of Ezrin protein were significantly higher in primary cancer tissues than lymph node metastases but the Ezrin staining pattern was seen mainly in the membrane region of primary cancer cells, whereas mainly in cytosols of metastatic cancer cells, indicating that the subcellular location of Ezrin protein in breast cancer is integral to the metastatic potential of the cell [44]. Interestingly, the immunocytochemical analysis of T47D cells for Ezrin expression following transfection with miR-183 demonstrated more membranous staining in the transfected cells, which is in keeping with less metastatic potential (figure 6a). These were the cells that demonstrated decreased migration on functional analysis. In the SKBR-3 cells which did not exhibit decreased migration, there appeared to be decreased membranous and cytoplasmic expression of Ezrin in the transfected cells (figure 6b), this suggests that the overexpression of miR-183 in these cells had disrupted Ezrin expression but this had not translated to a measurable functional effect.

These results imply that there is an important regulatory role of miR-183 in breast cancer, centering on cell migration and metastatic potential.

The identification of the molecular mechanisms responsible for aggressive tumor behaviour and metastasis is important for understanding disease progression, guiding disease management, and developing potentially novel cancer treatment strategies. The elucidation of the functional role of miR-183 in breast cancer promises to provide an important avenue for further analysis in in-vivo systems with an aim to develop a new diagnostic and therapeutic target in the management of metastatic breast cancer.