CRC (colorectal cancer) is one of the most malignant tumours in both developing and developed countries. It is estimated that 60% of CRC patients have liver metastasis. In the present study, we show that miR-30b is an important regulator in human CRC migration and invasion, which are vital steps in CRC liver metastasis. miR-30b was significantly down-regulated in primary CRC specimens compared with normal tissues. Furthermore, miR-30b was much lower in liver metastasis tissues than in CRCs. We validated SIX1 (SIX homeobox 1), a member of the SIX homeodomain family of transcription factors and an EMT (epithelial–mesenchymal transition)-promoting gene, as the direct target of miR-30b. Forced expression of miR-30b inhibited CRC cell migration and invasion in vitro via its target gene SIX1. Furthermore, an inverse correlation between expression of SIX1 and miR-30b has been observed both in primary CRC specimens and liver metastasis. Taken together, miR-30b plays an important role in mediating metastatic related behaviour in CRC. miR-30b may serve as a potential diagnostic marker and therapeutic target for patients with CRC in the future.
CRC (colorectal cancer), commonly known as colon cancer or bowel cancer, is a kind of tumour that features uncontrolled cell growth in the colon or rectum (parts of the large intestine) or in the appendix. CRC is the third most commonly diagnosed cancer in the world and is a heterogeneous disease that is common in both men and women. Approximately 60% of cases were diagnosed in the developed world. In 2008, 1.23 million new cases of CRC patients were clinically diagnosed with 608 000 deaths from CRC worldwide . The recent data of the American Cancer Society showed 103170 new cases of colon cancer and 40 290 new cases of rectal cancer in the U.S.A. in 2012 . CRC is a very common malignant cancer and frequently manifests with liver metastases, often without other systemic diseases . Approximately 60% of patients with CRC developed liver metastases during the course of their disease. Chemotherapy improves survival outcome, converting some unresectable liver metastases into resectable sections . However, the factors of ideal chemotherapy including drug types and administration time during the course of treatment are uncertain. After liver resection, selected patients have a 5-year survival rate of 30% and a 10-year survival rate of 17–25% [5,6].
Many reported signalling pathways have been involved in LM (liver metastasis) from CRC [7,8]. MMP (matrix metalloproteinase) family overexpression is thought to enhance the liver metastatic potential [9–11]. The target gene p53 plays an important role in angiogenesis and metastasis in primary CRC [12,13]. Previous studies indicated that carriers of an EGF (epidermal growth factor) polymorphism might be at a higher risk of developing liver recurrences . EMT (epithelial–mesenchymal transition) is considered as an essential process in the metastatic cascade [15,16]. During EMT, epithelial cells lose their epithelial characteristics and acquire a mesenchymal phenotype , which increases the metastatic and invasive potential of cells. Previous studies have indicated the importance of EMT in CRC [18–20]. Following a comprehensive gene expression analysis using different types of CRC cell lines and a gene expression assay specialized for the analysis of EMT in CRC, the SIX1 (SIX homeobox 1) gene was identified as an EMT-related gene in CRC. The SIX1 gene is one of the groups of similar genes known as the SIX gene family of transcription factors which bind to DNA and control the activity of other genes. SIX1 overexpression represses E-cadherin (an epithelial marker) expression and promotes EMT in CRC through ZEB1 (zinc finger E-box-binding homeobox 1) activation . SIX1 has also been shown to target the lymphangiogenic factor VEGFC (vascular endothelial growth factor C) and increase lymphangiogenesis and metastasis in breast cancer . Therefore recent research has focused on the role of this gene regulator in tumour metastasis.
miRNAs are a short (18–22 nucleotides) evolutionarily conserved class of non-protein-coding RNA molecules, which negatively regulate gene expression at the post-transcriptional level by blocking translation through incomplete binding to the 3′-UTR of their target mRNAs and directing the degradation of target mRNAs . Previous studies have suggested that miRNAs can block or increase the evolution of malignant behaviours by regulating multiple targets [24–26]. The change in miRNA gene expression has been reported in a variety of cancers, and most miRNAs have the potential to affect the cell cycle and survival programmes. During the development of tumours, miRNAs may be useful predictive markers for the occurrence or recurrence of cancer in patients. miR-30b, a member of the miR-30 family, has been observed at a lower abundance in human pancreatic epithelial cells and is involved in the regulatory signalling events of pancreatic EMT [27,28]. The aim of the present study was to investigate the role of miR-30b on regulation of migration and invasion in CRC. We showed that miR-30b was reduced in CRC specimens and LM tissues compared with normal tissues.
MATERIALS AND METHODS
Cell culture, reagents and plasmid
CRC cell lines were from the A.T.C.C. (Manassas, VA, U.S.A.) and were cultured in RPMI 1640 medium. HEK (human embryonic kidney)-293T cells were cultured in DMEM (Dulbecco's modified Eagle's medium) supplemented with 10% FBS, 100 units/ml penicillin and 100 μg/ml streptomycin. Antibodies against E-cadherin and vimentin were from BD Biosciences and antibodies against ZEB1, SIX1 and β-actin were from Sigma. The miR-30b inhibitor (anti-miR-30b) and the corresponding controls were from Ambion. siRNA for human SIX1 was from Santa Cruz Biotechnology. The human SIX1 expression vector was from GeneCopoeia.
Tissue samples were obtained from the Affiliated Hospital of Oncology of Harbin Medical University with written informed consent from all of the patients, and were snap-frozen in liquid nitrogen for storage. All experimental procedures were approved by the Review Board of Shanghai Sixth People's Hospital.
RNA isolation and qPCR
Total RNA of cells or human tumour tissues was extracted using the TRIzol® reagent (Invitrogen) according to the manufacturer's instructions. To analyse gene expression levels, total RNAs were transcribed to cDNA using the PrimeScript RT Reagent kit (TaKaRa). The primers used are listed in Supplementary Table S1 (at http://www.biochemj.org/bj/460/bj4600117add.htm). For qPCR (quantitative real-time PCR), gene expression levels were normalized to that of GAPDH (glyceraldehyde-3-phosphate dehydrogenase). Expression of miR-30 was analysed using Taqman miRNA assays (Ambion). Expression of RNU6B (U6 control; Ambion) was used as an endogenous control.
Protein expression analysis
For Western blotting, cells were harvested and lysed on ice for 30 min in RIPA buffer supplemented with a protein kinase inhibitor cocktail. Lysates were centrifuged at 12000 g for 15 min at 4°C, and supernatants were collected. Total proteins were separated by SDS/PAGE (10% gel), transferred on to nitrocellulose membranes in transfer buffer, and detected with antibodies against E-cadherin, vimentin, ZEB1, SIX1 or β-actin.
Cell viability assay
Cell viability index was measured using the Cell Counting Kit-8 method (Dojindo) following the manufacturer's instructions.
Wound healing, cell migration and invasion assays
For the wound healing assay, cell monolayers were scratched with a 200 μl pipette tip and images taken. For migration and invasion assay, cells were subjected to a Boyden chamber assay (BD Bioscience) using 8-μm pore size membranes with (for invasion) and without (for migration) Matrigel.
Tissue samples were formalin-fixed and sectioned (5-μm thick), followed by heat-immobilization following the manufacturer's instructions for the specific antibody used. An antibody against SIX1 was used for the staining and detected through Dako Envision two-step IHC (immunohistochemistry) method.
After extensive washing with PBS, cells were permeabilized with 0.1% Triton X-100 in PBS for 10 min. Cells were then incubated overnight at 4°C with the anti-SIX1 antibody, washed three times with PBS and then treated for 1 h with FITC-conjugated goat-anti-(mouse IgG). Slides were prepared with mounting medium (Bio-Rad Laboratories), and images were taken using a Leica DM IRB fluorescence microscope equipped with a CF1 APO PLAN 60′/1.40 oil objective and a CoolPix digital camera.
Transfections and dual-luciferase reporter assay
For dual-luciferase assay, the 3′-UTR segment corresponding to SIX1 was amplified by PCR from whole cDNAs of SW620 cells and inserted into the pMIR-REPORTER vector (Ambion). The primers used are listed in Supplementary Table S1. Wild-type (SIX1-WT) and mutant (SIX1-MT) constructs were confirmed by plasmid DNA sequencing. SW620 cells in 24-well plates were co-transfected with luciferase vectors (empty luciferase vector and luciferase vectors containing either wild-type or mutant-type 3′-UTR of SIX1) and either miR-30b or a negative control using Lipofectamine™ 2000 (Invitrogen). Firefly and Renilla luciferase activities were measured 24 h after transfection using a dual-luciferase assay kit (Promega). Each experiment was repeated three times with three replicates.
Data in the present study are means±S.E.M. of at least three independent experiments except when indicated otherwise. The significances between two matched groups or two independent groups were analysed using Student's t test or Mann–Whitney U test respectively. For three groups, the Kruskal–Wallis test was used. Pearson's correlation coefficient was used to measure the correlation between expression levels of miR-30b and SIX1. Values were considered significantly different at P<0.05.
miR-30b is correlated with EMT features in CRC cells and human tissue specimens
To evaluate the levels of miR-30 family in the human CRC cell lines SW480 and SW620, we measured the expression levels of each miR-30 family member by qPCR. miR-30b was significantly reduced in mesenchymal SW620 cells compared with epithelial SW480 cells (Figure 1A). In other human CRC cell lines, the expression levels of miR-30b were between those in SW480 and SW620 cells (Figure 1B). To further understand the expression profiles of miR-30b in EMT-related cells, the expression levels of E-cadherin (an epithelial marker) and vimentin (a mesenchymal marker) were determined by qPCR and Western blotting (Figures 1C and 1D). SIX1 and ZEB1, EMT-related genes, were also validated. SW480 cells showed reduced E-cadherin expression, which is consistent with an epithelial phenotype. In contrast, SW620 cells, derived from the lymph node of a patient used as the source of SW480 cells, primarily expressed SIX1, ZEB1 and vimentin. To confirm our findings from CRC cell lines, we tested miR-30b in tissue specimens. We analysed the expression levels of miR-30b in 50 primary CRC tissues, 30 LM tissues and 20 normal colorectal tissues (normal control) by qPCR analysis. Consistent with the findings in cells, miR-30b expression was markedly decreased in CRCs and LM, especially in LM, indicating that miR-30b may be correlated with colorectal LM (Figure 1E). Next, we analysed whether miR-30b expression was correlated with the survival probability of CRC patients. Of all of the 50 CRC specimens, the 30% of patients with the lowest miR-30b expression and the 30% of patients with the highest miR-30b expression were included for survival analysis. As shown in Figure 1(F), low expression of miR-30b in CRCs was significantly associated with poor overall survival rates.
Expression of miR-30b in human CRC cells and tissue specimens
Restoration of miR-30b inhibits the migration and invasion of CRC cells
To explore the potential roles of miR-30b in LM, SW620 CRC cells were transfected with either pre-miR-30b or a negative control. The forced expression level of miR-30b was confirmed by qPCR (Supplementary Figure S1 at http://www.biochemj.org/bj/460/bj4600117add.htm). Ectopic forced expression of miR-30b in SW620 cells did not lead to a reduction in cell viability (Figure 2A). Furthermore, we investigated the roles of miR-30b in the migration and invasion of CRC cells. In the wound healing assay, overexpression of miR-30b markedly reduced the migrating ability of CRC cells in monolayer cultured cells following scratching (Figure 2B). In a Boyden chamber assay, the inhibition of migration in CRC cells was also observed in response to overexpression of miR-30b (Figure 2C). Similarly, miR-30b significantly reduced the ability of CRC cells to invade through Matrigel-coated transwell membranes (Figure 2C). To further confirm whether miR-30b actually inhibits an EMT phenotype, we examined the epithelial marker E-caderin, the mesenchymal marker vimentin and the EMT-promoting protein ZEB1 in SW620 cells using immunoblotting (Figure 2D). Forced expression of miR-30b increased the level of E-cadherin and reduced the expression of vimentin and ZEB1, indicating an inverse correlation of EMT features and miR-30b expression.
Function of miR-30b in CRC cells
miR-30b directly targets the SIX1 gene
It has been reported that SIX1 is a putative mesenchymal marker and promotes EMT in CRC cells. Using miRNA target prediction algorithms, such as TargetScan (http://www.targetscan.org) and microRNA.org (http://www.microrna.org/microrna/home.do), we found a putative seed-matching binding site between the seed sequences of miR-30b and the 3′-UTR of SIX1 (Figure 3A), which is highly conserved in different species (Supplementary Table S2 at http://www.biochemj.org/bj/460/bj4600117add.htm). To clarify whether miR-30b directly targets SIX1 through binding to its 3′-UTR, we constructed luciferase reporter vectors containing putative wild-type (SIX1-WT) or mutant (SIX1-MT) binding sites to perform reporter assays. The luciferase activity of the wild-type reporter was markedly suppressed by miR-30b overexpression and prompted by the miR-30b inhibitor (anti-miR-30b), whereas the mutant reporter remained unchanged (Figure 3B). Furthermore, overexpression of miR-30b in CRC SW620 cells reduced both the mRNA and protein levels of SIX1, whereas inhibition of miR-30b had the opposite effect (Figures 3C–3E), indicating that SIX1 is the direct target of miR-30b. In SW480 cells (with epithelial features), we found, as expected, that miR-30b regulated the expression of SIX1 (Supplementary Figure S2 at http://www.biochemj.org/bj/460/bj4600117add.htm). Many reports have provided conclusive evidence that SIX1 is a metastatic regulator and induces metastasis via its ability to induce EMT, a process that is often correlated with increased metastases . As shown in Figure 3(F), overexpression of miR-30b reversed the SIX1-induced changes in EMT-related genes. Suppression of SIX1 by shRNA has been shown to decrease the cell motility and invasiveness of heptocellular carcinoma . We suppressed SIX1 expression using siRNA (Supplementary Figure S3A at http://www.biochemj.org/bj/460/bj4600117add.htm) and found that reduced expression of SIX1 led to reduced migration and invasion of CRC cells (Supplementary Figure S3B), consistent with the results for miR-30b overexpression.
Validation of SIX1 as a direct target gene of miR-30b
miR-30b is inversely correlated with SIX1 in tumours
To explore whether SIX1 levels in primary CRC cells were correlated with those in LM tissues, we collected a cohort of samples containing 50 primary CRC tissues, 30 LM tissues and 20 normal colorectal tissues. Of all these samples, 12 pairs of adjacent normal colorectal tissues, primary CRCs and corresponding secondary LM tissues were included. In normal tissues, IHC staining indicated that the expression of SIX1 was very low (Supplementary Figure S4A at http://www.biochemj.org/bj/460/bj4600117add.htm). In primary CRC tissues, SIX1 expression was universally higher than in normal tissues and, in some CRC tissues, no IHC signal was detected (Supplementary Figure S4B). In contrast, in secondary LM tissues SIX1 was highly expressed (Supplementary Figure S4C). To better investigate the expression profile of SIX1 in tissue samples, we analysed SIX1 protein expression by Western blotting. As shown in Figure 4(A), none of the 12 secondary LM tissues had higher SIX1 expression than the corresponding primary CRCs. SIX1 expression in both the LM and CRC tissues was usually more highly expressed than in the normal tissues (Supplementary Figure S5 at http://www.biochemj.org/bj/460/bj4600117add.htm). Furthermore, after measuring the immunoblotting signals of all of the tissue samples, we found that SIX1 was up-regulated in primary CRCs compared with normal colorectal controls. More excitingly, the levels of SIX1 in the LM specimens were much higher than those in primary CRC tissues (Figure 4B), indicating that SIX1 expression levels maybe correlated with LM in patients with CRC. Finally, we analysed the correlation of miR-30b and SIX1 in primary CRC and LM tissues. The expression levels of miR-30b had an inverse correlation with SIX1 expression in primary CRC tissues (Figure 4C). The same inverse trend was also observed between the expression of miR-30b and SIX1 in LM tissues (Figure 4D). Taken together, our results indicate that SIX1, as an EMT-promoting gene, may be correlated with LM in patients with CRC, and that SIX1 and miR-30b may be together considered as a biomarker of CRC LM.
Correlation of miR-30b and SIX1 in tissue specimens
miR-30b regulates the migration and invasion of CRC via SIX1
To further identify whether miR-30b regulated the migration and invasion of CRC cells via its target gene SIX1, we employed a vector containing a SIX1 expression sequence without the binding site for miR-30b at the 3′-UTR. Expression of SIX1 was validated both in HEK-293T and SW620 cells (Supplementary Figure S6A at http://www.biochemj.org/bj/460/bj4600117add.htm). Forced expression of SIX1 in CRC cells rescued expression of SIX1 which was suppressed by miR-30b (Supplementary Figure S6B). SIX1 has been shown to be involved in regulating metastasis in numerous human cancers [30,31]. In CRC cells, overexpression of SIX1 enhanced both the cell migration and invasion in vitro (Supplementary Figure S6C) . Next, we investigated whether miR-30b regulated cell migration and invasion via SIX1. First, overexpression of miR-30b reduced the migration ability of SW620 cells, whereas simultaneously expression of SIX1 with miR-30b increased miR-30b-inhibited cell migration (Figure 5A). Secondly, using a Boyden chamber assay we showed that SIX1 significantly restored miR-30b-inhibited cell migration and invasion of SW620 cells (Figure 5B). All of the above data show that miR-30b regulates the migration and invasion of CRC through its target gene SIX1.
SIX1 reversed the inhibition of miR-30b on CRC migration and invasion
Distant metastasis is the major cause of death in patients with CRC [32,33]. Despite new diagnostic and therapeutic approaches, increasing numbers of patients with metastatic CRC have been observed worldwide. Consequently, the molecular mechanisms or potential biomarkers for cancer diagnosis are required to develop effective therapeutic strategies for patients with metastatic CRC. EMT is thought of as one of the key molecular steps in the process of distant metastasis, which permits the invasion and migration in various cancers and is associated with poor prognosis in CRC [18,19]. The present study has shown that miR-30b directly targets the EMT-related gene SIX1 which promotes CRC metastasis. Moreover, overexpression of miR-30b changed the level of EMT markers, and inhibited the migration and invasion of the CRC cell line SW620. In clinical samples, miR-30b had an inverse correlation with SIX1 expression in primary CRC and LM tissues, conferring a potential diagnostic and prognostic value for this miRNA as a biomarker. These results may also have implications for the clinical management of patients with metastatic CRC.
SIX1 is a member of the six-homeodomain family of transcription factors. SIX1 is highly expressed in multiple tumour cells and plays important roles in the proliferation, metastasis and survival of cancers [34–36]. Recent studies have indicated that SIX1 promoted EMT in CRC through ZEB1 activation . Transgenic mice expressing the homeoprotein SIX1 in epithelial cells had an increase in subsequent tumour development and metastasis via EMT [37,38]. These findings suggest that SIX1 possibly is a central mediator of tumour behaviour. The results of the present study indicate that miR-30b directly targets SIX1 through binding to its 3′-UTR. Although we could not explore the effect of SIX1-induced EMT and CRC metastasis in in vivo models, overexpression of SIX1 in CRC SW620 cells improved EMT-related gene expression (Figure 3). Excitingly, miR-30b significantly reversed the effect of SIX1 on cell migration and invasion. In addition, CRC cell lines contained different levels of miR-30b (Figure 1B), which may have resulted from a different EMT phenotype or SIX expression. Furthermore, we analysed the correlation of miR-30b and SIX1 expression in tissue specimens. As shown in Figure 4(B), the levels of SIX1 were increased in CRC metastasis tissues (LM>CRC>normal). miR-30b was also shown to be inversely correlated with SIX1 expression in primary CRC and LM tissues of patients (Figures 4C and 4D).
It has been reported that miR-30 maintained or regulated an epithelial phenotype in pancreatic islet cells , the first time the relationship between miR-30 and EMT was shown. The miR-30 family has been confirmed to be involved in tumour-related behaviours and multiple signal pathways are involved in the regulation of tumorigenesis by the miR-30 family. Overexpression of the miR-30 family in prostate cancer cells suppressed EMT phenotypes and inhibited cell migration and invasion though EGF/Src signalling . miR-30a has been shown to reduce PIK3CD (phosphoinositide 3-kinase catalytic subunit δ) and suppress cell migration and invasion in colorectal carcinoma , and inhibit the progression and development of breast tumours . miR-30c has been shown to control lung cancer invasion via MTA1 (metastasis-associated protein 1)  and hypoxia-induced EMT in human renal cell carcinoma ; however, the effect of miR-30b on CRC metastasis remains unclear. According to expression analysis of the miR-30 family in CRC cell lines, we suggest that lower levels of miR-30b disrupt the regulatory balance and increase the risk of the development and progression of CRC. Our data show that miR-30b possibly plays an essential role in the regulation of EMT and CRC metastasis. Previous studies also have described the role of miRNAs in the progression of CRC developing from a primary process to metastatic disease, such as the miR-200 family . Likewise, we suggest that miR-30b may be an integral marker of CRC progression and, furthermore, a novel target for CRC prognosis and therapy.
miR-30b may control the migration and invasion of the CRC cell line SW620 by negatively regulating the expression of mesenchymal gene transcripts of SIX1. miR-30b may serve as a potential diagnostic marker and therapeutic target for patients with CRC.
Hui Zhao and Zifeng Xu designed and performed experiments. Huanlong Qin contributed to the experimental work including maintaining the animals and performing IHC and qPCR. Zhuo Gao and Lu Gao directed the study. Hui Zhao wrote the paper.
This work was supported by the National Natural Science Foundation of China [grant number 81201628] and the Young Scientist Fund of Heilongjiang province [grant number QC2012C099].
These authors contributed equally to this work.