Original Article

Oncogene (2010) 29, 4297–4306; doi:10.1038/onc.2010.201; published online 31 May 2010

miR-200bc/429 cluster targets PLCγ1 and differentially regulates proliferation and EGF-driven invasion than miR-200a/141 in breast cancer

S Uhlmann1,3, J D Zhang1,3, A Schwäger1, H Mannsperger1, Y Riazalhosseini2, S Burmester1, A Ward1, U Korf1, S Wiemann1 and Ö Sahin1

  1. 1Division of Molecular Genome Analysis, German Cancer Research Center, Heidelberg, Germany
  2. 2Division of Functional Genome Analysis, German Cancer Research Center, Heidelberg, Germany

Correspondence: Dr Ö Sahin, Division of Molecular Genome Analysis, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 580, 69120, Heidelberg, Germany. E-mail: oe.sahin@dkfz-heidelberg.de

3These authors contributed equally to this work.

Received 28 September 2009; Revised 28 February 2010; Accepted 26 April 2010; Published online 31 May 2010.

Top

Abstract

The genes encoding microRNAs of the human miR-200 family map to fragile chromosomal regions and are frequently downregulated upon tumor progression. Although having been reported to regulate epithelial-to-mesenchymal transition and transforming growth factor-beta-driven cell invasion, the role of the miR-200 family in EGF-driven breast cancer cell invasion, viability, apoptosis and cell cycle progression is still unknown. In particular, there is no study comparing the roles of the two clusters of this miRNA family. In this study, we show for the first time that miR-200 family members differentially regulate EGF-driven invasion, viability, apoptosis and cell cycle progression of breast cancer cells. We showed that, all miR-200 family members regulate EGF-driven invasion, with the miR-200bc/429 cluster showing stronger effects than the miR-200a/141 cluster. Furthermore, expression of the miR-200a/141 cluster results in G1 arrest supported by increased p27/Kip1 and decreased cyclin dependent kinase 6 expression. In contrast, expression of the 200bc/429 cluster decreases G1 population and increases G2/M phase, in line with the observed reduction of p27/Kip1 and upregulation of the inhibitory phosphorylation of Cdc25C, respectively. To test the hypothesis that phenotypical differences observed between the two clusters are caused by differential targeting spectrums, we performed genome-wide microarray profiling in combination with gain-of-function studies. This identified phospholipase C gamma 1 (PLCG1), which was downregulated only by the miR-200bc/429 cluster, as a potential candidate contributing to these phenotypical differences. Luciferase reporter assays validated PLCG1 as a direct functional target of miR-200bc/429 cluster, but not of miR-200a/141 cluster. Finally, loss of PLCG1 in part mimicked the effect of miR-200bc/429 overexpression in viability, apoptosis and EGF-driven cell invasion of breast cancer cells. Our results suggest that the miR-200 family has a tumor-suppressor function by negatively regulating EGF-driven cell invasion, viability and cell cycle progression in breast cancer.

Keywords:

miR-200 family; PLCγ1; invasion; cell proliferation; tumor suppressor; breast cancer

Top

Introduction

Breast cancer ranks first among the causes of incidence of cancers among women worldwide with about one million new cases every year, and also the most frequent cause of cancer mortality in women (Parkin et al., 2005). It is becoming more and more evident that abnormal expression of microRNAs (miRNAs), a class of small non-coding RNAs of 20–22 nucleotides in length, is associated with a variety of human cancers including breast cancer, and that they have crucial roles in cell proliferation, differentiation and apoptosis (Esquela-Kerscher and Slack, 2006; Bartel, 2009). Several miRNAs have been shown to be involved in the regulation of the invasive and metastasizing phenotype of breast cancer cells. For instance, miR-10b initiates tumor invasion and distal metastases in vivo (Ma et al., 2007), whereas tumor suppressor miRNAs, such as miR-335, have also been reported, which inhibit breast cancer invasion and metastasis (Tavazoie et al., 2008). Similarly, several miRNAs have been reported to regulate cell proliferation by modulating cell cycle progression or apoptosis by targeting key intermediates. For example, the miR-15a/16 cluster was shown to induce cell cycle arrest at G1 phase by targeting cyclin D1/D3, cyclin E1 and cyclin-dependent kinase 6 (CDK6) (Liu et al., 2008). In another study, it was shown that transcriptional activation of miR-34a contributes to the p53-mediated apoptosis (Raver-Shapira et al., 2007).

The human miR-200 family consists of paralogs located on two chromosomal regions. Although miR-200b, miR-200a and miR-429 are located on chromosome 1p36, miR-200c and miR-141 are located on 12p13, and both regions are often deleted in several types of cancer (Sato et al., 2001; Bagchi and Mills, 2008). Alternatively, based on the seed region (nucleotides 2–7), they can be divided into two sequence clusters: miR-200b, 200c and 429 (miR-200bc/429), as well as miR-200a and 141 (miR-200a/141). The miR-200bc/429 cluster differs from the miR-200a/141 cluster with regard to only the fourth nucleotide (U to C) in the seed region. As it is commonly hypothesized that the seed region has a crucial role in miRNA direct targeting (Lewis et al., 2005), in this paper we focus on the common features and differences of miR-200bc/429 and miR-200a/141 in breast cancer.

Reduced expression of the miR-200 family was recently shown to be associated with tumor progression (Baffa et al., 2009). Epithelial-to-mesenchymal transition (EMT), a crucial process in the transformation of tumor cells into aggressive metastatic cancer cells, also seems to be tightly regulated by this family (Gregory et al., 2008). In particular, all its members regulate transforming growth factor-beta (TGF-β)-induced EMT, as TGF-β stimulation reduces miR-200 levels and thus allows upregulation of the E-cadherin transcriptional repressors, zing finger E-box binding homeobox 1 (ZEB1) and Smad Interacting Protein 1, both of which are established targets of miR-200. However, there is no study examining the function of the miR-200 family in EGF-driven invasion, which has a central role in both normal breast development and cancer (Citri and Yarden, 2006), and it is also not clear yet whether the two clusters have different regulatory effects.

Little is known about the role of the miR-200 family in cell growth and proliferation in breast cancer. Recently, Saydam et al. (2009) showed that miR-200a downregulation in meningioma subtypes of brain tumors promotes tumor growth by increasing Cyclin D1 and β-catenin in vitro and in vivo. Similarly, it was shown that the overexpression of miR-200a inhibits nasopharyngeal carcinoma cell growth, migration and invasion by targeting β-catenin and Smad interacting protein 1, whereas its knockdown stimulates these processes in these cells (Xia et al., 2010). In contrast, it was reported that the human miR-200 family promotes cell growth when transfected into several cancer cell lines. In addition, they have shown that the miR-8-null fly, which is the Drosophila homolog of human miR-200 family, showed increased apoptosis in the brain and smaller body size compared with wild types (Hyun et al., 2009). Controversial results among tumor types and the lack of breast cancer studies drove us to compare the roles of the two miR-200 clusters in breast cancer cell proliferation.

In this study, we show that the miR-200bc/429 cluster regulates cell viability, apoptosis, cell cycle progression and EGF-driven invasion in breast cancer cells with distinct patterns from the miR-200a/141 cluster. With miRNA target prediction algorithms and transcriptome profiling, we could identify 21 potential exclusive direct targets of miR-200bc/429, one of which is phospholipase C gamma 1 (PLCG1), the enzymatic activity of which is required for cell motility induced by EGF (Chen et al., 1994), and for cell proliferation and survival (Markova et al., 2010).

Top

Results

Different targeting spectrums of two different clusters of the miR-200 family

The human miR-200 family can be divided into two clusters on the basis of seed–sequence similarity (miR-200bc/429 cluster and miR-200a/141 cluster, Figure 1a). We examined whether the single-nucleotide change between miR-200bc/429 and miR-200a/141 (U to C) is reflected to the targeting spectrum. Queries with three widely used miRNA target prediction tools (Figure 1b) supported our hypothesis that although the two clusters have common predictions, there are far more mutually exclusive predicted targets for each cluster. However, it is unclear whether this difference will be reflected in the cancer cell phenotypes when the two clusters are expressed.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Two sequence clusters of miR-200 family members have only one single nucleotide difference in the seed region but excessive differences in the predicted targeting spectrums. (a) Human miR-200 family is located in two fragile chromosomal regions on 1p36.33 (200b, 200a and 429) and 12p13.31 (200c and 141), respectively. It consists of two clusters based on seed sequence similarity: miR-200bc/429 (red) and 200a/141 (blue), distinguished by a single nucleotide change (U to C). (b) Numbers of predicted targeting genes of two clusters by three miRNA targeting prediction algorithms. Despite the single nucleotide change, the two clusters have distinct potential targeting spectrum (64~87% exclusive predicted targets for miR-200bc/429 and 65~81% for miR-200a/141).

Full figure and legend (101K)

The miR-200bc/429 cluster reduces EGF-driven cell invasion stronger than the miR-200a/141 cluster in breast cancer

To examine the role of miR-200 family in EGF-dependent breast cancer cell invasion, we transfected the highly invasive MDA-MB-231 cell line with control mimic or miR-200 family mimics. Cells were then seeded in the presence of EGF, whereas fetal bovine serum was used as a chemoattractant in the trans-well assays. Overexpression of miR-200 family members reduced the invasion of cells compared with the mimic control (Figure 2), comparable to the effects in the TGF-β-induced cell invasion (Gregory et al., 2008). Interestingly, the effect of the miR-200bc/429 cluster was much more predominant than that of miR-200a/141 (P=8.7e–5). We hypothesize that the different regulatory roles of two clusters are caused by exclusively regulated genes.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Upon EGF stimulation, the overexpression of miR-200bc/429 has a significantly stronger inhibitory effect on cell invasion compared with miR-200a/141 (P=8.7e–5, Student's t-test), although all five members regulate cell invasion negatively. The bar height indicates the number of invaded cells transfected with miR-200 family-member mimics in invasion assay after EGF stimulation (biological replicate number N=3), and the error bars, if not otherwise specified, indicate the 95% confidence interval of the estimated mean value. Three asterisks stand for P<0.001.

Full figure and legend (59K)

miR-200bc/429 cluster, but not miR-200a/141 cluster, reduces cell viability and increases apoptosis in breast cancer

We later examined whether miRNAs of the two clusters would act differently on breast cancer cell viability and apoptosis. Cell viability assay after transfection of miRNA mimics showed that the miR-200a/141 cluster did not affect viability of the MDA-MB-231 breast cancer cell line, whereas expression of the miR-200bc/429 cluster resulted in the reduction of cell viability at 48 and 72h after transfection (Figure 3a). In line with the viability assay results, although transfection of the cells with miRNA mimics of the miR-200a/141 cluster led to no significant change in the apoptosis indicator caspase 3/7 activity, the population of apoptotic cells was increased by miR-200bc/429 after both 48 and 72h of transfection (Figure 3b). Again we hypothesize that exclusively regulated targets of miR-200bc/429 cause the difference in cell viability and apoptosis.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

miR-200bc/429, but not miR-200a/141, reduces cell viability and promotes apoptosis. (a) ATP-production-based viability assay (N=6) showed no significant reduction by overexpressing miR-200a/141 with mimics (P=0.45 for miR-200a and 0.12 for miR-141 at 72h, Bonferroni corrected), whereas miR-200bc/429 cluster significantly reduced cell viability 72h after transfection (P=4.6e–4 for miR-200b, 7.6e–4 for miR-200c and 1.1e–4 for miR-429). (b) The miR200bc/429 cluster increases the CASP3/7 activity, suggesting that its members promote the Caspase-dependent apoptosis (N=6). No significant changes of CASP3/7 activity were detected in the cells overexpressing miR-200a/141 (P=0.68 and P=0.11 for miR-200a at 48h and 72h, respectively), whereas the miR-200bc/429 increases the CASP3/7 activity (P=5.3e–8 and 3.4e–16 for miR-200c at two different time points). Two asterisks stand for P<0.01, three for P<0.001.

Full figure and legend (144K)

Although 200bc/429 cluster decreases G1 and increases G2/M populations, miR-200a/141 cluster results in G1 arrest in breast cancer

Having identified apoptosis as one of the potential causes of reduced viability for the miR-200bc/429 cluster, we examined whether the two clusters would have any effect on cell cycle progression or retardation. Therefore, we first performed the 7-amino-actinomycin D (7-AAD) cell cycle assay with miRNA mimics in MDA-MB-231 cells. As seen in Figure 4a, the 200a/141 cluster resulted in a higher G1 population compared with control, whereas the miR-200bc/429 cluster decreased the G1 population and increased the G2 population. In order to examine the differential effects of the two clusters on S-phase of the cell cycle in detail, we performed bromodeoxyuridine staining with pulse-labeling, and observed that although the miR-200bc/429 cluster did not significantly differ from the control (P=0.52), the 200a/141 cluster resulted in a reduction in S-phase population of the cells (P=0.027).

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

miR-200bc/429 causes G2/M cell cycle arrest, whereas miR-200a/141 causes G0/1 arrest. (a) Two clusters showed distinct profiles in the cell cycle assay on the basis of the 7-AAD-staining: miR-200a/141 caused G0/1 phase accumulation, whereas miR-200bc/429 reduced G0/1 phase proportion and increased G2/M phase cells at the same time (N=3). (b) bromodeoxyuridine staining combined with 7-AAD staining validates the observations in (a). Each panel shows one of three biological replicates with representative miRNAs (miR-200a and miR-200c) from each cluster, the proportions of S- (red gate) and G2/M (blue gate) phase cells are presented in the form of mean±s.d. (c) Protein expression quantification of cell cycle proteins with the reverse phase protein arrays after overexpressing miR-200a or miR-200c (N=5). (d) Western blotting of cell cycle regulators of G2/M transition, after the transfection of miRNA mimics of all five miR-200 family members along with control (N=3). One asterisk stands for P<0.05, two for P<0.01, three for P<0.001. A full colour version of this figure is available at the Oncogene journal online.

Full figure and legend (183K)

In order to elucidate the molecular changes responsible for the different effects of the two clusters on the cell cycle, we examined the changes in several cell cycle components either with reverse phase protein arrays (Sahin et al., 2007) (Figure 4c) or with western blotting (Figure 4d). The phosphorylation of retinoblastoma protein (Ser 807/811), which is a G1/S transition marker (Sahin et al., 2009), was significantly reduced for the samples transfected with miR-200a mimic (representative of miR-200a/141 cluster), but not affected for the ones transfected with miR-200c mimic (representative of the miR-200bc/429 cluster) (Figure 4c). Overexpression of miR-200a resulted in a strong reduction of the CDK6 level and an increase of the CDK inhibitor, p27/Kip1 (Figure 4c). This may have resulted in a lower phosphorylation of retinoblastoma protein and finally in G1 arrest. On the other hand, miR-200c overexpression reduced the CDK inhibitor p27/Kip1 and increased CDK6 levels (Figure 4c), which could potentially explain the reduction of G1-phase after miR-200c mimic. Furthermore, we observed a significant decrease in CDK2 and increase in pCdc2 (T14) and pCdc25c (Ser 216), which refrain cells from transition from G2 to M phase, although we observed a strong increase in cyclin B1 and cyclin A levels in the miR-200c-transfected samples (Figures 4c and d). In conclusion, the distinct patterns of the cell cycle progression when two clusters are expressed are in line with the differential regulation of cell cycle components. These results may indicate that differential targeting of the two clusters is reflected to different cell cycle progression profiles.

Target prediction algorithms and genome-wide expression profiling identify potential exclusive direct targets of the miR-200bc/429 cluster

We propose that the distinct regulatory roles of miR-200bc/429 and miR-200a/141 in breast cancer invasion, proliferation and cell cycle are caused by differential targets of the two clusters, especially those exclusive direct targets. As the miR-200bc/429 shows significant changes in all these phenotypes, we focused only on the genes that targeted by miR-200bc/429. First, we analyzed the predicted targets of the miR-200bc/429 cluster from the three algorithms, which identified 321 genes in common (Figure 5a). Subsequently, we performed genome-wide mRNA microarray profiling with MDA-MB-231 cells transfected with all miR-200 family members. This identified 240 genes downregulated by miR-200bc/429 compared with control or miR-200a/141 (Figure 5b). Combining the prediction and profiling, 21 genes are predicted to be exclusively direct targets of miR-200bc/429 and are downregulated at the mRNA level (Figure 5c, Supplementary Table 1). The potential regulatory effects of miR-200bc/429 on these genes, therefore, could be, at least in part, accomplished by the mRNA cleavage mechanism.

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Prediction algorithms and transcriptome profiling suggests that PLCG1, together with other 20 genes, might be differentially regulated by miR-200bc/429 than by miR-200a/141 by direct targeting. (a) 321 genes are predicted to be targets of miR-200bc/429, but not of miR-200a/141, by three algorithms. (b) Genome-wide mRNA profiling identifies 240 genes downregulated exclusively by miR-200bc/429 cluster members, out of which 21 genes intersect with the commonly predicted set defined in the subfigure (a). (c) The mRNA expression pattern of these 21 genes after overexpressing the miR-200 family members or miRNA mimic control. (d) The differential regulation of PLCG1 by two clusters is validated on the protein level by the western blotting.

Full figure and legend (130K)

Among these genes, PLCG1 is highly relevant in the breast cancer tumorigenesis and cancer development as it was reported to modulate EGF receptor-driven invasion (Li et al., 2009) and recently cell proliferation and apoptosis (Markova et al., 2010). Western blot analysis validated that only the miR-200bc/429 cluster, but not the miR-200a/141 cluster, resulted in a reduction of PLCγ1 protein expression (Figure 5d). Therefore, we hypothesize that the potential regulation by direct targeting of miR-200bc/429 on PLCG1 contributes to the different regulatory effects of the two clusters on the phenotypes we examined.

PLCG1 is a direct target of miR-200bc/429, but not of miR-200a/141, in breast carcinoma cell lines

We analyzed the 3′-UTR of PLCG1 (NM_002660) mRNA and identified one highly conserved target site of miR-200bc/429 (position 4915–4921) (Figures 6a and b), which was not predicted for miR-200a/141. The forced expression of miR-200bc/429 in MDA-MB-231 cells (comparable with the endogenous expression levels of the family in MCF-7 cells, which have a high expression level of miR-200 family, Supplementary Figure 1) resulted in a reduction at both the protein and mRNA level of PLCG1 (Figures 6c and d). As a complementary experiment, we transfected the miRNA hairpin inhibitors in MCF-7 cells and observed a significant increase in both mRNA and protein levels (Figures 6c and d). The mRNA level regulation validates our hypothesis that the regulation of PLCG1 exclusively by miR-200bc/429 may, at least in part, be by an mRNA cleavage mechanism.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

PLCG1 is a direct target of miR-200bc/429, but not of miR-200a/141. (a) Gene structure of PLCG1 with the predicted target site for miR-200bc/429. (b) Evolutionary conservation of the target site across five species, asterisks incidate conserved nucleotides. (c) The expression level of PLCγ1 protein was reduced by overexpressing 200c in the MDA-MB-231 cell line, and increased by inhibiting 200bc/429 in MCF-7 cells (N=3, showing one representative blot). (d) The expression level of PLCG1 mRNA was also reduced by overexpressing miR-200c in MDA-MB-231 cells and increased by inhibiting miR-200bc/429 in MCF-7 cells (N=3), suggesting that the mRNA cleavage may contribute to the targeting. (e) Luciferase assay with the 3′-UTR of PLCG1 mRNA suggested that miR-200bc/429, but not 200a/141, directly targets PLCG1. Overexpressing miR-200bc/429 with mimics in MDA-MB-231 cells reduced the luciferase activity (N=8, P=1.6e–12). Complementarily inhibiting miR-200bc/429 in MCF-7 cells increased the luciferase activity (N=8, P=2.1e–14). Two asterisks stand for P<0.01, three for P<0.001.

Full figure and legend (134K)

To determine whether the reduced expression of PLCγ1 was because of the direct targeting of miR-200bc/429, we cloned reporter constructs containing the 3′-UTR region of the PLCG1 gene downstream of the luciferase open reading frame. The miR-200bc/429 cluster reduced the relative luciferase activity by targeting the 3′-UTR of PLCG1 (Figure 6e, left). To further prove that this predicted target site is indeed responsible for the reduced luciferase activity, we cloned only the target site downstream of the luciferase open reading frame and measured the luciferase activity, where we also observed a significant reduction of luciferase activity, confirming the targeting of PLCG1 mRNA (Supplementary Figure 2a). Furthermore, we observed an increase in the luciferase activity when the miR-200bc/429 was inhibited in MCF-7 cells (Figure 6e, right). Finally, we introduced mutations into the target site by changing four nucleotides within the seed-matching sequence. Disrupting the target site by mutagenesis resulted in a relative luciferase activity that was not significantly different from that obtained with control mimic (P=0.11, Supplementary Figure 2b). To conclude, these results confirm our hypothesis that PLCG1 is a direct target of the miR-200bc/429 cluster, but not of miR-200a/141.

Targeting PLCG1 by miR-200bc/429 cluster contributes to the observed differential effects of two miR-200 family clusters

Having established PLCG1 as a direct functional target of the miR-200bc/429 cluster, we examined the effect of loss of PLCG1 on cell viability, apoptosis and EGF-driven invasion using siRNAs. As can be seen from Figure 7a, knockdown of PLCG1 with a pool of four siRNAs resulted in reduced cell viability of MDA-MD-231 cells, which is in line with the effect of the miR-200bc/429 cluster (Figure 7a). Similarly, loss of PLCG1 moderately increased the caspase 3/7 activity, which reflects more apoptosis (Figure 7b). Finally, we showed that loss of the PLCG1 gene with siRNAs would lead to the reduction of EGF-driven invasion in MDA-MB-231 cells in a ZEB1-pathway-independent manner (Figure 7c). Furthermore, quantitative proteomics data indicated that ZEB1, known to be targeted by all five members and to have role in the reduction of TGF-β-driven invasion, was similarly targeted by all five members of the miR-200 family under EGF stimulus (Supplementary Figure 3). These results indicate that the regulation by direct targeting of miR-200bc/429 contributes to the differential effects of the two clusters on breast cancer cell invasion, proliferation and apoptosis.

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Direct targeting and regulation of PLCG1 by miR-200bc/429 contributes to the phenotypes observed by miRNAs’ expression. (a) Knock down of PLCG1 with siRNA inhibits cell viability (N=6, P=0.024 and 2.5e–4 for 48 and 72h respectively), coherent with the effects of miR-200bc/429 overexpression. (b) Knock down of PLCG1 also promotes apoptosis (N=6, P=0.017 and 1.1e–5 for two time points), in the same manner as that of the miR-200bc/429 cluster. (c) Neither ZEB1 nor PLCG1 knock down affects the expression of each other (validated with the western blotting), yet both cause significant reduction of cell invasion upon EGF stimulation (N=3). This suggests that the downregulation of PLCG1 decreases the invasion capacity of breast cancer cell line in a ZEB1-independent pathway, indicating a novel mechanism of miR-200bc/429 regulating EGF-dependent cell invasion by directly targeting PLCG1. One asterisk stands for P<0.05, three for P<0.001.

Full figure and legend (114K)

Top

Discussion

In this study, we showed that two sequence clusters of human miR-200 family, miR-200bc/429 and miR-200a/141, have distinct profiles in EGF-dependent breast cancer cell invasion, proliferation and cell cycle progression. We have shown for the first time that all miR-200 family members regulate EGF-driven invasion, with the miR-200bc/429 cluster having stronger effects than the miR-200a/141 cluster. We have also shown that, although the miR-200a/141 cluster had no significant effect, miR-200bc/429 cluster results in reduced viability and increased apoptosis of breast cancer cells. Similarly, although the miR-200a/141 cluster results in G1 arrest, the 200bc/429 cluster decreases the G1 population and increases the G2/M phase. Genome-wide microarray profiling and bioinformatic analysis identified 21 potential exclusive direct targets of the miR-200bc/429 cluster, one of which was PLCG1, as candidates contributing to these phenotypical differences. Luciferase reporter assay validated PLCG1 as a direct functional target of the miR-200bc/429 cluster, but not that of the miR-200a/141 cluster. Finally, loss of PLCG1 in part mimicks the effects of miR-200bc/429 expression in viability, apoptosis and EGF-driven cell invasion of breast cancer cells.

The role of miR-200 family in TGF-β-driven invasion by targeting ZEB1 and smad interacting protein 1 has been established (Gregory et al., 2008). In this study, we showed that the expression of all miR-200 family members in MDA-MB-231 breast cancer cells also resulted in reduced EGF-driven invasion. However, the inhibitory effect of miR-200bc/429 is stronger than that of miR-200a/141 (Figure 2). In line with our study, Baffa et al. (2009) could show that miR-200c expression was significantly higher in primary tumors compared with metastases. Another study found that ductal mammary tumors expressed miR-200c at higher levels than sarcomatoid metaplastic tumors, where a significant correlation between expression of miR-200c and E-cadherin was detected for the two tumor types (Gregory et al., 2008). We also showed that the PLCG1, which was suggested to be involved in EGF-induced cell motility (Chen et al., 1994), is a direct target of the miR-200bc/429 cluster (Figure 6). Our results show the knockdown of PLCG1, which was reported to be increased in metastatic patients in comparison with primary breast tumors (Sala et al., 2008), reduced EGF-driven cell invasion in a ZEB1-independent manner (Figure 7).

Little is known about the involvement of miR-200 family in tumor progression in the context of cell proliferation. Even less is known about how different family members can affect tumor proliferation and/or whether there is any difference between their effects. In this study, we report that the miR-200bc/429 cluster, but not miR-200a/141, resulted in reduced viability of cells, as well as higher apoptosis induction (Figure 3). The exclusive targeting of PLCG1 by miR-200bc/429 contributes to the observed distinct effect of this cluster on these phenotypes as it was quite recently shown that, in line with our study, suppression of PLCG1 resulted in inhibition of cell proliferation and enhanced apoptosis potentially by interaction of PLCG1 with the mTOR/p70S6K pathway in chronic myeloid leukemia (Markova et al., 2010). We observed that miR-200bc/429 decreased the G1 population of cells and increased the G2/M population compared with control (Figures 4a and b). A strong increase in the phosphorylation of Cdc25C at the Ser-216 site, which prevents Cdc25C dual-specificity phosphatase from dephosphorylating and activating cyclin B1/Cdc2 (Peng et al., 1997) for the miR-200bc/429 cluster-transfected cells may explain the observed differential effects on the G2/M phase. Targeting of PLCG1 by miR-200bc/429 may further explain the observed effect of this cluster on G2/M phase as it was recently shown that PLCG1 interaction with AKT regulates G2/M transition in the MDA-MB-231 cell line (Browaeys-Poly et al., 2009).

There is increasing evidence that EMT may contribute to the resistance to different therapies in cancer (Thomson et al., 2005; Yauch et al., 2005; Arumugam et al., 2009). The miR-200 family, which regulates EMT, has also been reported in the sensitization of cancer cells to different drugs, for example, to the EGFR-targeting molecule gefitinib in bladder carcinoma (Adam et al., 2009) or to paclitaxel in breast, ovarian and endometrial cancers (Cochrane et al., 2010). A current focus in drug discovery is to develop agents that target the cell cycle checkpoints, aiming at increasing or decreasing the amount of arrested cells at these points (reviewed in DiPaola, 2002). We showed that miR-200bc/429 and miR-200a/141 members can lead to G2/M- or G1-arrest of the cell cycle, and therefore act as a tumor supressor. This observation could potentially be of therapeutic use as the overexpression of miR-200bc/429 in breast cancer may lead to higher apoptosis and less viability of the tumor cells. At the same time, our findings suggest that the drug sensitization role of miR-200 family may be explained not only by controlling the EMT, but also by leading to cell cycle arrest followed by apoptosis.

In conclusion, we have shown that two clusters of the miR-200 family differentially regulate EGF-driven breast cancer cell invasion, viability, apoptosis and cell cycle progression. Direct targeting of the key molecule PLCG1 in breast cancer by miR-200bc/429 contributes to this difference, and more comprehensive studies should be performed to elucidate the contribution of all mutually exclusive targets (or regulations by secondary effects of direct targeting) to the phenotypes caused by miR-200bc/429 and miR-200a/141 expression. We believe that the modulation of expression of miR-200 family members expression could be a potential approach for breast cancer treatment, highlighting the necessity for further studies.

Top

Materials and methods

MiRNA target prediction

MiRNA-targeting predictions were queried in three target prediction databases: TargetScan (http://www.targetscan.org), PITA (http://genie.weizmann.ac.il/pubs/mir07) and DIANA-microT (http://diana.pcbi.upenn.edu/cgi-bin/micro_t.cgi) with default parameters (for PITA the no-flanking top table was used, see the website for details).

Cell culture and serum/growth factor stimulation

Two human breast cancer cell lines (MDA-MB-231 and MCF-7) were obtained from American Type Culture Collection (Manassas, VA, USA). Culturing media and supplements for the two cancer cell lines were described previously (Sahin et al., 2009).

Transfection with siRNAs and miRNA mimics

Cells were transfected with siRNAs targeting ZEB1 or PLCG1, purchased from Dharmacon (Lafayette, CO, USA). For each gene, four individual siRNAs were pooled (Supplementary Table 2). Allstar siRNA (Qiagen, Hilden, Germany) was used as non-silencing control. miRIDIAN miRNA mimics for 200 family, as well as the negative control, were from Dharmacon. siRNA transfection using Lipofectamine 2000 (Invitrogen, CA, USA) was carried out as previously described (Sahin et al., 2009). siRNAs, miRNA mimics and miRNA hairpin inhibitors were transfected at final concentrations of 20, 25 and 100nM, respectively.

Cell lysis and western blotting

Preparation of protein lysates and western blotting was previously described (Sahin et al., 2009). In all, 10μg of protein was separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and exposed to primary antibodies (Supplementary Table 3). Horseradish peroxidase-conjugated secondary antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Reverse-phase protein array

Lysates were prepared as previously described (Sahin et al., 2009). Five technical replicate spots were printed for every sample on nitrocellulose-coated glass slides (Grace Biolabs, Bent, OR, USA) using an Aushon 2470 contact printer (Aushon, Boston, MA, USA). Target protein-specific primary antibodies (Supplementary Table 4) were diluted to 1:300. Alexa680-labeled secondary antibodies (Invitrogen, Karlsruhe, Germany) were used and visualized using an Odyssey scanner (LI-COR, Lincoln, NE, USA). Signal intensities were quantified using Gene Pix Pro 5.0 (Molecular Device, Sunnyvale, CA, USA) and background corrected. Differences between miR-200a, miR-200c and control were identified for each antibody using Tukey's method applied to the ANOVA model. All the P-values reported have been adjusted for multiple testing using the Bonferroni method.

RNA isolation and mRNA/miRNA expression profiling

Total RNA was isolated according to the manufacturer's protocol using the RNeasy Mini Kit (Qiagen). Sentrix HumanWG-6 arrays (Illumina, San Diego, CA, USA) were used for mRNA profiling. Quality control of total RNA, as well as labeling and hybridization, was performed at the DKFZ microarray core facility. Expression profiling data were normalized with the variance stabilization transformation algorithm (Huber et al., 2002) and analyzed using the Bioconductor limma package (Smyth, 2004).

Quantitative RT–PCR of mRNA and miRNAs

cDNA synthesis was carried out with the Revert Aid H Minus First Strand cDNA Synthesis Kit (Fermentas St Leon-Rot, Germany), and using 10ng of total RNA. The quantitative reverse transcriptase–PCR reactions for target genes and housekeeping genes ACTB, HPRT and TFRC were performed with an ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Weiterstadt, Germany) using probes from the Universal Probe Library (Roche, Penzberg, Germany). Oligonucleotide primers were synthesized at MWG (Ebersberg, Germany). Sequences of primers and the respective UPL probe numbers are given (Supplementary Table 5). Reverse transcription of miRNAs was performed using the TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems) with 4ng total RNA and thereafter, using the miRNA TaqMan Assay (Applied Biosystems). RNU44 and RNU48 were used as housekeeping controls. Data were analyzed using the Delta-Delta-Ct algorithm (Livak and Schmittgen, 2001) (Bioconductor ddCt package).

Plasmid construction, mutagenesis and luciferase assay

The 3′-UTR of PLCG1 gene was amplified by PCR using human genomic DNA (Supplementary Table 6) and cloned downstream of the luciferase open reading frame using the NheI and XhoI sites in the pLuc vector. Four point mutations were generated within the predicted target site of PLCG1 3′-UTR using the QuikChange XL site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA) according to the manufacturer's instruction. For the luciferase assay, cells were transfected with 25μM of miRNAs and the luciferase vector together with the pMIR-REPORT β-gal vector (Ambion, Austin, TX, USA). Luciferase and β-gal activities were measured using the Dual-Light System (Applied Biosystems, Foster City, CA, USA), following the manufacturer's instructions. Relative luciferase activity was determined by the ratio of luciferase signal intensity to that of β-gal for normalization. The average and standard deviations of the ratio were estimated by the Delta-method (Bioconductor ratioAssay, under review).

Cell viability and Caspase 3/7 assays

Cell proliferation and apoptosis were measured using Cell TiterGlo Luminescent Cell Viability and Caspase-Glo 3/7 assays (Promega, Fitchburg, WI, USA) respectively following the manufacturer's instructions. Transfections were carried out in a 96-well plate (8 × 103cells per well) in a final miRNA mimic and siRNA concentrations of 25 and 20nM per well in six replicates.

Cell cycle analysis

7-AAD/bromodeoxyuridine cell cycle assay was applied according to the manufacturer's manual (BD Pharmingen San Diego, CA, USA) and analyzed by flow cytometry using Cell Quest Pro software (BD Bioscience, Heidelberg, Germany).

Invasion assay

Cells were transfected with siRNAs or miRNAs as described above and seeded in the invasion plates (Millipore, Billerica, MA, USA). EGF (25ng/ml) was added to cells in the upper chamber and after 30h the number of invaded cells was determined by flow cytometry.

Top

Conflict of interest

The authors declare no conflict of interest.

Top

References

  1. Adam L, Zhong M, Choi W, Qi W, Nicoloso M, Arora A et al. (2009). miR-200 expression regulates epithelial-to-mesenchymal transition in bladder cancer cells and reverses resistance to epidermal growth factor receptor therapy. Clin Cancer Res 15: 5060–5072. | Article | PubMed | ChemPort |
  2. Arumugam T, Ramachandran V, Fournier KF, Wang H, Marquis L, Abbruzzese JL et al. (2009). Epithelial to mesenchymal transition contributes to drug resistance in pancreatic cancer. Cancer Res 69: 5820–5828. | Article | PubMed | ChemPort |
  3. Baffa R, Fassan M, Volinia S, O'Hara B, Liu CG, Palazzo JP et al. (2009). MicroRNA expression profiling of human metastatic cancers identifies cancer gene targets. J Pathol 219: 214–221. | Article | PubMed | ChemPort |
  4. Bagchi A, Mills AA. (2008). The quest for the 1p36 tumor suppressor. Cancer Res 68: 2551–2556. | Article | PubMed | ChemPort |
  5. Bartel DP. (2009). MicroRNAs: target recognition and regulatory functions. Cell 136: 215–233. | Article | PubMed | ChemPort |
  6. Browaeys-Poly E, Perdereau D, Lescuyer A, Burnol AF, Cailliau K. (2009). Akt interaction with PLC (gamma) regulates the G(2)/M transition triggered by FGF receptors from MDA-MB-231 breast cancer cells. Anticancer Res 29: 4965–4969. | PubMed | ChemPort |
  7. Chen P, Xie H, Sekar MC, Gupta K, Wells A. (1994). Epidermal growth factor receptor-mediated cell motility: phospholipase C activity is required, but mitogen-activated protein kinase activity is not sufficient for induced cell movement. J Cell Biol 127: 847–857. | Article | PubMed | ISI | ChemPort |
  8. Citri A, Yarden Y. (2006). EGF-ERBB signalling: towards the systems level. Nat Rev Mol Cell Biol 7: 505–516. | Article | PubMed | ISI | ChemPort |
  9. Cochrane DR, Howe EN, Spoelstra NS, Richer JK. (2010). Loss of miR-200c: a marker of aggressiveness and chemoresistance in female reproductive cancers. J Oncol 2010 (doi:10.1155/2010/821717). | Article
  10. DiPaola RS. (2002). To arrest or not to G(2)-M Cell-cycle arrest: commentary re: A K. Tyagi et al., Silibinin strongly synergizes human prostate carcinoma DU 145 cells to doxorubicin-induced growth inhibition, G(2)-M arrest, and apoptosis; 2002,Clin Cancer Res, 8: 3512-3519. Clin Cancer Res 8: 3311–3314. | PubMed | ChemPort |
  11. Esquela-Kerscher A, Slack FJ. (2006). Oncomirs—microRNAs with a role in cancer. Nat Rev Cancer 6: 259–269. | Article | PubMed | ISI | ChemPort |
  12. Gregory PA, Bert AG, Paterson EL, Barry SC, Tsykin A, Farshid G et al. (2008). The miR-200 family and miR-205 regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol 10: 593–601. | Article | PubMed | ChemPort |
  13. Huber W, von Heydebreck A, Sultmann H, Poustka A, Vingron M. (2002). Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics 18(Suppl 1): S96–S104. | PubMed |
  14. Hyun S, Lee JH, Jin H, Nam J, Namkoong B, Lee G et al. (2009). Conserved MicroRNA miR-8/miR-200 and its target USH/FOG2 control growth by regulating PI3K. Cell 139: 1096–1108. | Article | PubMed | ChemPort |
  15. Lewis BP, Burge CB, Bartel DP. (2005). Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120: 15–20. | Article | PubMed | ISI | ChemPort |
  16. Li S, Wang Q, Wang Y, Chen X, Wang Z. (2009). PLC-gamma1 and Rac1 coregulate EGF-induced cytoskeleton remodeling and cell migration. Mol Endocrinol 23: 901–913. | Article | PubMed | ChemPort |
  17. Liu Q, Fu H, Sun F, Zhang H, Tie Y, Zhu J et al. (2008). miR-16 family induces cell cycle arrest by regulating multiple cell cycle genes. Nucleic Acids Res 36: 5391–5404. | Article | PubMed | ChemPort |
  18. Livak KJ, Schmittgen TD. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25: 402–408. | Article | PubMed | ISI | ChemPort |
  19. Ma L, Teruya-Feldstein J, Weinberg RA. (2007). Tumour invasion and metastasis initiated by microRNA-10b in breast cancer. Nature 449: 682–688. | Article | PubMed | ChemPort |
  20. Markova B, Albers C, Breitenbuecher F, Melo JV, Brummendorf TH, Heidel F et al. (2010). Novel pathway in Bcr-Abl signal transduction involves Akt-independent, PLC-gamma1-driven activation of mTOR/p70S6-kinase pathway. Oncogene 29: 739–751. | Article | PubMed | ChemPort |
  21. Parkin DM, Bray F, Ferlay J, Pisani P. (2005). Global cancer statistics, 2002. CA Cancer J Clin 55: 74–108. | Article | PubMed | ISI
  22. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS, Piwnica-Worms H. (1997). Mitotic and G2 checkpoint control: regulation of 14–3–3 protein binding by phosphorylation of Cdc25C on serine-216. Science 277: 1501–1505. | Article | PubMed | ISI | ChemPort |
  23. Raver-Shapira N, Marciano E, Meiri E, Spector Y, Rosenfeld N, Moskovits N et al. (2007). Transcriptional activation of miR-34a contributes to p53-mediated apoptosis. Mol Cell 26: 731–743. | Article | PubMed | ISI | ChemPort |
  24. Sahin O, Frohlich H, Lobke C, Korf U, Burmester S, Majety M et al. (2009). Modeling ERBB receptor-regulated G1/S transition to find novel targets for de novo trastuzumab resistance. BMC Syst Biol 3: 1. | Article | PubMed | ChemPort |
  25. Sahin O, Lobke C, Korf U, Appelhans H, Sultmann H, Poustka A et al. (2007). Combinatorial RNAi for quantitative protein network analysis. Proc Natl Acad Sci USA 104: 6579–6584. | Article | PubMed | ChemPort |
  26. Sala G, Dituri F, Raimondi C, Previdi S, Maffucci T, Mazzoletti M et al. (2008). Phospholipase Cgamma1 is required for metastasis development and progression. Cancer Res 68: 10187–10196. | Article | PubMed | ChemPort |
  27. Sato Y, Kobayashi H, Suto Y, Olney HJ, Davis EM, Super HG et al. (2001). Chromosomal instability in chromosome band 12p13: multiple breaks leading to complex rearrangements including cytogenetically undetectable sub-clones. Leukemia 15: 1193–1202. | Article | PubMed | ISI | ChemPort |
  28. Saydam O, Shen Y, Wurdinger T, Senol O, Boke E, James MF et al. (2009). Downregulated microRNA-200a in meningiomas promotes tumor growth by reducing E-cadherin and activating the Wnt/beta-catenin signaling pathway. Mol Cell Biol 29: 5923–5940. | Article | PubMed | ChemPort |
  29. Smyth GK. (2004). Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 3: Article 3.
  30. Tavazoie SF, Alarcon C, Oskarsson T, Padua D, Wang Q, Bos PD et al. (2008). Endogenous human microRNAs that suppress breast cancer metastasis. Nature 451: 147–152. | Article | PubMed | ChemPort |
  31. Thomson S, Buck E, Petti F, Griffin G, Brown E, Ramnarine N et al. (2005). Epithelial to mesenchymal transition is a determinant of sensitivity of non-small-cell lung carcinoma cell lines and xenografts to epidermal growth factor receptor inhibition. Cancer Res 65: 9455–9462. | Article | PubMed | ISI | ChemPort |
  32. Xia H, Ng SS, Jiang S, Cheung WK, Sze J, Bian XW et al. (2010). miR-200a-mediated downregulation of ZEB2 and CTNNB1 differentially inhibits nasopharyngeal carcinoma cell growth, migration and invasion. Biochem Biophys Res Commun 391: 535–541. | Article | PubMed | ChemPort |
  33. Yauch RL, Januario T, Eberhard DA, Cavet G, Zhu W, Fu L et al. (2005). Epithelial versus mesenchymal phenotype determines in vitro sensitivity and predicts clinical activity of erlotinib in lung cancer patients. Clin Cancer Res 11: 8686–8698. | Article | PubMed | ISI | ChemPort |
Top

Acknowledgements

This work was supported in part by the National Genome Research Network (Contract No. 01GS0864) of the Federal Ministry of Education and Research (BMBF) and by Wilhelm-Sander Stiftung (Contract No. 2009.051.1). JDZ is supported by the DKFZ International PhD Program. We thank Moritz Küblbeck, Christian Schmidt and Ute Ernst for their excellent technical assistance.

Supplementary Information accompanies the paper on the Oncogene website

Top

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated

NEWS AND VIEWS

MicroRNAs ????? keeping cells in formation

Nature Cell Biology News and Views (01 May 2008)

Cancer Micromanagement of metastasis

Nature News and Views (11 Oct 2007)