Background
Breast cancer is the most commonly diagnosed disease among women. Aggressive breast cancers have high potential to become metastatic, a transition that makes clinical intervention difficult. In recent years, many of the molecular mechanisms that facilitate a more invasive or metastatic state have been characterized [
1]–[
3]. It has been observed that the transcriptome of primary breast cancer cells with a predisposition for metastasis can be distinguished from non-metastatic breast cancer cells [
4]. This indicates that steady-state mRNA levels are altered in pre-metastatic tissue to promote metastasis. Therefore, it is critical to define the events leading to and maintaining a metastatic transcriptional landscape of metastatic breast cancer cells.
Assessing a cell’s transcriptome using RNA-seq or hybridization arrays provides measurement of the steady-state levels of cellular mRNAs. Cellular mRNA levels are controlled at multiple levels: transcriptional rate; mRNA processing and export; mRNA stability; and translational rate. miRNAs are small non-coding RNAs that post-transcriptionally regulate mRNA levels, many of which are critical for development [
5]. Similar to mRNAs, the steady-state levels of miRNAs are also altered in cancerous tissues as compared to normal cells. Further, just as for protein coding RNAs, many non-coding RNAs have the capacity to function as tumor suppressors and oncogenes (often referred to as onco-miRs) [
6],[
7]. miRNAs are also involved in cancer cell metastasis (metasta-miRs), and regulate key physiological steps in the metastatic process of many cancers, including breast [
2],[
8]–[
11]. Hence, characterizing members of this class of non-coding RNAs is important in our understanding of breast cancer development and progression.
We queried publicly available data from TCGA Breast Cancer project [
12] for aberrant miRNAs, and observed that hsa-mir-30c is elevated in expression level or copy number in approximately 4% or breast cancer patients. Importantly, these patients have significantly poorer survival than patients with normal hsa-mir-30c, which suggests that hsa-mir-30c may be a key factor in breast cancer progression. Using a cell-based model, we observed an association between cellular invasion and high levels hsa-mir-30c, which led us to hypothesize that the observed clinical mortality phenotype associated with aberrant hsa-mir-30c was as a result of hsa-mir-30c regulation of breast cancer cell invasion. To address this hypothesis, we examined factors targeted by hsa-mir-30c that regulate breast cancer cell invasion. We investigated potential cross talk between a factor known to be targeted by miR-30c, RUNX2 [
13], which promotes metastasis and osteolytic disease [
14]. Further, because there are still many unknown players promoting the metastatic phenotype of advanced breast cancer, we addressed other targets of hsa-mir-30c, first by
in silico analyses for targets linked to cellular invasion, then functional investigation of these potential targets.
Our key findings demonstrated that highly metastatic MDA-MB-231 breast cancer cells have robust levels of hsa-mir-30c compared to non-metastatic MCF-7 cells; and that hsa-mir-30c promotes breast cancer cellular invasion through targeting of NOV/CCN3, which we characterized as an inhibitor of invasion. We demonstrated the specificity of this pathway by showing: a) that only the canonical strand of hsa-mir-30c is detected and responsible for the invasive phenotype; and b) that hsa-mir-30c-NOV/CCN3-mediated invasiveness is completely independent of hsa-mir-30c targeting of RUNX2. Importantly, our cell-based experimental observations allow for mechanistic insight into the clinical observations of both hsa-mir-30c and NOV/CCN3, which suggests that the hsa-mir-30c-NOV pathway is an important target for future translational studies.
Discussion
Here we identify a novel pathway by which hsa-mir-30c promotes the invasiveness of the MDA-MB-231 cell line through targeting of NOV. The specificity of NOV’s involvement in the invasive phenotype observed by several experimental approaches including the knockdown by an siRNA targeting NOV, which resulted in significant increases in the invasiveness of MDA-MB-231, consistent with the high levels of hsa-mir-30c, and NOV cellular protein levels in these cells. These results demonstrate a novel pathway by which a miRNA (hsa-mir-30c) promotes the invasive phenotype of metastatic breast cancer cells.
The large mir-30 family shares a conserved seed sequence. However, our
in silico research suggests that seed sequence differences may give rise to selectivity in targeting aggressive compared to modestly invading cells among mir-30 members. It is appreciated from the literature that the miR-30 family has numerous targets, and others have been identified in breast [
57]–[
59] and other cancers [
30],[
60]–[
64]. For example, a recent study showed that has-mir-30a marker targets the 3’-UTR of Vimentin (VIM) which was observed to cause reductions in both VIM protein levels and invasiveness of MDA-MB-231 cells [
63], and here we demonstrate that miR-30c also promotes cellular invasion by targeting NOV. It is also quite interesting that these mir-30 miRNAs appear on many chromosomes rather than co-regulated in a cluster, and are involved in the regulation of a myriad of pathways such as tumor suppression (p53) [
63], apoptosis (BCL) [
64], and epithelial to mesenchymal transition (VIM) [
59]. Individual members of the mir-30 family have been implicated in both tumor suppression and oncogenesis; it is therefore difficult to define the family as a “tumor suppressive” or “oncogenic”. This ambiguity makes studying the functions of the mir30 family members on a case-by-case basis critical for understanding the basis of post-transcriptional molecular mechanisms of disease. The relative levels of the mir-30 family members and their temporal expression may play a critical role in disease progression.
Also, to be considered is the context in which our experiments were conducted. We investigated the functions of hsa-mir-30c in MCF-7, a less invasive breast cancer cell line and the more invasive MDA-MB-231 cells, which is a cell model of advanced metastatic breast cancer [
65]. We find a positive in correlation in levels of hsa-mir-30c and invasion properties
in vitro assays and high expression of hsa-mir-30c has been observed in MDA-MB-231cells, as well as in other cellular models of breast cancer. However detection of hsa-mir-30c did not appear to be predictive of breast cancer subtypes [
66]. Our cell-based observations are similar to observations made in
in vivo studies of miRNA expression patterns in breast cancer patient samples in which hsa-mir-30c expression was not significantly associated with any clinical or pathological features [
12],[
67]–[
70]. However, elevated expression of hsa-mir-30c was observed to be significantly associated with estrogen receptor (ER) positive breast cancers that were beneficially responsive to tamoxifen treatment [
58]. In a separate study, ER-positive breast cancer patients who relapsed after tamoxifen treatment had significantly increased levels of NOV [
49], which further underscores the inverse relationship between NOV and hsa-mir-30c in breast cancer, as well as in our observations of low hsa-mir-30c in the ER positive cell line (MCF-7) and higher levels in the MDA-MB-231 cells which are ER negative.. Thus, we have observed that hsa-mir-30c was important for the invasive properties which is discordant with the
in vivo observations that hsa-mir-30c expression was not positively associated with metastatic disease, but rather, with positive outcome in response to hormone therapy. We therefore suggest that the cohort of potential mRNA targets available to hsa-mir-30c in patients may be quite different from those in MDA-MB-231 cells, which could explain the differences between
in vivo tumors with a heterogenous tumor cell population and individual cell lines. The significance of our findings is that there are likely to be tumor cells during progression of breast cancer
in vivo that could benefit from targeted hsa-mir-30c or NOV therapy in reducing invasion and later stage metastasis.
We have demonstrated that the functional consequence of increased invasiveness of MDA-MB-231 cells following the reduction of NOV through either hsa-mir-30c or NOV-siRNA is directly sensitive to NOV levels. NOV appears to play a context-sensitive role in oncogenesis/tumor suppression [
71]. In several cancers that develop from mesenchymal tissues, NOV has been shown to promote tumor growth and metastasis [
72]–[
74]. However, in the context of brain cancer, NOV appears to inhibit disease progression [
53],[
75]–[
77], but in the context of alveolar rhabdmyosarcoma NOV/CCN3 levels are increased and contribute to survival of oppressive behaviors of the cancerous cells promotes motility. While there are conflicting observations of NOV functions in breast cancer cell lines [
28],[
49], NOV protein levels in human tissue samples of breast cancer were observed to be negatively correlated with late-stage and metastatic disease [
50]. These histological observations suggest that in breast cancer, NOV functions to inhibit disease progression.
Methods
Cell Properties
Cell lines
MCF-7 cells were grown in DMEM in media and MDA-MD-231 cells in alpha MEM and maintained as previously described [
53]. These cells have been maintained in our laboratory and used in seven publications since 2005. In 2012 the cells were revalidated by satellite phenotyping. These cell lines were grown and maintained as previously described [
53]. Constructs and stable cell lines were generated as previously described [
56]. MDA-MB-231 cells which were lentivirus-transduced stable cell lines to express either empty vector, wild-type
Mus musculus Runx2 a subnuclear-targeting-deficient mutant form of
Mus musculus Runx2 (R398A/Y428A), which inhibits the invasiveness of MDA-MB-231 [
54]–[
56] were used in these studies.
RNA isolation
RNA was isolated using Qiagen miRNAeasy Mini Kit (217004) following the manufacturer’s recommended protocol with optional in-column DNAse I digestion of genomic DNA (Qiagen RNase-Free DNase Set 79254).
miRNA amplification and detection
Complimentary miRNA-specific cDNA was amplified and detected using Applied Biosystems TaqMan MicroRNA Assays for hsa-mir-30c (#4427975) hsa-mir-30c-2* (#4427975) and RNU6B (#4427975).
cDNA amplification and detection
cDNA was amplified from equal quantities of total cellular RNA for each treatment or cell line. cDNA was amplified using the Invitrogen SuperScript First-Strand Synthesis System for RT-PCR (#11904-018) according to the manufacturer’s protocol. Reactions were volumetrically diluted, and reaction products were used as templates for Real Time qPCR using Bio-Rad iQ SYBR Green Supermix (#170-8880).
cDNA qPCR primers
Real Time qPCR primers were designed using FoxPrimer (
http://www.foxprimer.org) and validated for efficiency by standard curve using cDNA amplified from untreated MDA-MB-231 cells.
Protein isolation and Western blotting
Cells grown on tissue culture plates were placed directly on ice, and washed twice with PBS supplemented with Roche cOmplete, EDTA-free Protease Inhibitor Cocktail (#11873580001) and 25 μM MG132 (Calbiochem (EMD Millipore) CAS 133407-82-6). Cells were scraped into screw-top microcentrifuge tubes, gently spun down to pellet cells and excess PBS was aspirated and discarded. Cells were snap-frozen in liquid nitrogen. Protein lysates were prepared by the addition of RIPA buffer (50 mM Tris pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% v/v NP-40, 0.1% w/v SDS, 1× Roche cOmplete, EDTA-free Protease Inhibitor Cocktail and 25 μM MG132) and placing tubes on a 100 °C heat block for 10 min. Protein lysates were quantified using Pierce BCA Protein Assay Kit (#23225) according to manufacturer’s instructions. Fifty μg protein per sample was loaded onto an SDS-PAGE gel. SDS-PAGE was performed as described [
53]. Briefly, lysates were run through an 8.5% acrylamide gel, and then transferred to a PVDF Transfer Membrane (Thermo Scientific #88518).
Membranes were blocked with 5% (w/v) milk (BioRad #170-6404XTU) in PBS and then subjected to immunodetection using the following primary antibodies and dilution factors in 1% (w/v) milk in PBS: NOV (Santa Cruz Biotechnology H-71 sc-50304 1:1000), Lamin A/C (Santa Cruz Biotechnology N-18 sc-6215 1:5000), α-Tubulin (Santa Cruz Biotechnology H-300 sc-5546 1:2000), RUNX2 (mouse hybridoma clone 8G5 1:1000) [
53], FOXG1 (Santa Cruz Biotechnology N-15 sc-1858), SATB2 (Santa Cruz Biotechnology H-118 sc-98677), SMAD2 (Santa Cruz Biotechnology S-20 sc-6200), MTSS1 (Santa Cruz M7-P3A7 sc-101390), WWP1 (Santa Cruz Biotechnology N-20 sc-11892), RUNX1 (Active Motif #39000), Cdk2 (Santa Cruz Biotechnology M2 sc-163). Secondary antibodies used were from Santa Cruz Biotechnology and were diluted 1:5000 in 1% (w/v) milk in PBS: donkey anti-goat IgG-HRP (sc-2020), goat anti-mouse IgG-HRP (sc-2005), and goat anti-rabbit IgG-HRP (sc-2004). After incubation with primary and secondary antibodies, the membranes were washed three times for 30 min each with 0.1% (v/v) Tween-20 in PBS. HRP reaction was achieved by one minute incubation with Perkin Elmer Western Lightning ECL (NEL102001EA). Membranes were exposed to Kodak BioMax Light File for Chemiluminescent Imaging (#868-9358) in serial exposure times to empirically determine the exposure time at which the signal was most linear.
Matrigel invasion and migration assays
Proliferating MDA-MB-231 cells were trypsinized and counted using Cellometer Auto T4 Cell Counter. A cell suspension of 100,000 cells/mL in growth medium was prepared and 100 μL of the suspension was loaded into each BD Matrigel 24-well 8.0 μm PET Membrane Invasion Chamber (#354483). Matrigel coated plates, and control insert plates had 500 μL NIH3T3-conditioned medium loaded in the bottom as the chemoattractant. Plates and chemoattractant medium were incubated at 37 °C for 3–4 h prior to loading MDA-MB-231 cells. Cells were incubated for 16 h at 37°C in 5% CO2 and then fixed and stained using the Fisher HealthCare PROTOCOL Hema 3 Manual Staining System (#22-122-911) according to the manufacturer’s instructions. Cotton swabs were used to eliminate cells which did not migrate/invade as well as Matrigel. Cells were counted using an inverted light microscope. To control for proliferation effects, rates of cellular invasion through Matrigel were normalized by rates of cellular migration through control plastic-only insert wells.
Transient transfection
Proliferating MDA-MB-231 cells were transfected with 50nM of siRNA/miRNA using Oligofectamine (Invitrogen #12252-011) according to the Oligofectamine protocol.
siRNAs
Dharmacon SMARTpool: ON-TARGETplus RUNX2 siRNA (L-012665-00-0005); Dharmacon SMARTpool: ON-TARGETplus NOV siRNA (L-010527-00-0005); Dharmacon ON-TARGETplus Non-targeting Pool (D-001810-10-05).
miRNAs and anti-miRNAs
Dharmacon miRIDIAN microRNA hsa-mir-30c-1 mimic (C-300542-03-0005); Dharmacon miRIDIAN microRNA hsa-mir-30c-1* mimic (C-301199-01-0005); Dharmacon miRIDIAN microRNA hsa-mir-30c-1 haripin inhibitor (IH-300542-07-0005); Dharmacon miRIDIAN microRNA Mimic Negative Control #1 (CN-001000-01-05); Dharmacon miRIDIAN microRNA Hairpin Inhibitor Negative Control #1 (IN-001005-01-05).
Screen for hsa-mir-30c targets
The top 300 targets of hsa-mir-30c based on mirSVR were downloaded from
http://www.microrna.org in January 2011. Gene symbols were used to access gene ontology (GO) terms from DAVID (
http://david.abcc.ncifcrf.gov) and gene reference into function (GeneRIF) from NCBI (
http://www.ncbi.nlm.nih.gov/gene/about-generif). Genes whose GO terms or GeneRIFs were associated with invasion, migration, extracellular matrix, or transcription factors were selected and qPCR primers were designed. After 48 h of transfection, RNA was isolated, cDNA was amplified and Real Time qPCR was carried out to detect the relative levels of mRNAs following transfection with hsa-mir-30c.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
qPCR for miRNAs performed by JRD and HT. qPCR for RUNX2 and NOV performed and analyzed by JRD. siRNA and miRNA transfections performed by JRD. Invasion assays performed by JRD and HT. Western blotting performed by JRD. MDA-MB-231 cells stably overexpressing Runx2 and Runx2-RY engineered by DH. Initial cancer-centric screens for hsa-mir-30c targets in MDA-MB-231 cells performed by YH and JP. Ontological and qPCR screen was designed and executed by JRD. Study was executed under the guidance of JBL, AJVW, JLS, GSS, and JP, who contributed to the writing of this manuscript. All authors read and approved the final manuscript.