Background
Colorectal cancer (CRC) is the fourth leading cause of cancer-related mortality worldwide [
1], and tumor invasion and metastasis are the main causes of mortality in these patients [
2]. During invasion and metastasis, activation of oncogenes and inactivation of suppressors play significant roles [
3]. Although many mechanisms underlying these processes have been elucidated and treatment has improved, 50% of CRC patients still develop metastases following surgery [
4]. Therefore, identification of new predictive markers and the mechanisms involved in the invasion and metastasis of CRC is urgently needed.
MicroRNAs (miRNAs) are small non-coding RNAs (18–24 nucleotides) that can exist stably in serum and mediate biological processes such as tumor cell migration, invasion, proliferation and apoptosis [
5]. Increasing evidence has demonstrated that aberrant expression of miRNAs can directly promote or inhibit CRC progression by degrading the 3′ untranslated regions (3′UTRs) of target genes [
6]. For example, miR-34a-5p suppresses CRC metastasis via a p53-dependent pathway, and low miR-34a-5p expression was found to predict good prognosis in stage II/III CRC patients [
3]. Through transforming growth factor (TGFβ), miR-1269a directly targets Smad7 and HOXD10, forming a positive feedback loop that enhances the metastatic capacity of CRC cells [
7]. In addition, miR-4775 has been verified as a new onco-miRNA with a dual effect on the ERBB2/Her2 gene [
8]. However, there are no reports to date on the function of miR-4755 in CRC progression.
The Smad7 gene, also known as mothers against decapentaplegic homolog 7 (MADH7), is located on human chromosome 18q21.1 and encodes a protein of 426 amino acids [
9]. Smad7-dependent activation of TGFβ signaling is able to induce the epithelial to mesenchymal transition (EMT) in CRC, allowing CRC cells to leave the tissue parenchyma and enter systemic circulation, followed by tumor invasion and metastasis [
10]. Smad7 is an inhibitory Smad protein that interacts with TGFβ type I receptor, targeting it for proteasomal degradation and thereby inhibiting TGFβ-I-induced phosphorylation of Smad2/Smad3 [
11]. Smad7 usually acts as a tumor suppressor [
12]. Indeed, it has been reported that Smad7 inhibits breast cancer metastasis to the lung and liver, represses melanoma cell metastasis to bone, and inhibits hepatocellular carcinoma cell proliferation and invasion [
13‐
15]. According to gene target prediction, a miR-4775 recognition sequence is present in the Smad7 3′UTR, though it remains unclear whether miR-4775 promotes CRC metastasis through activation of the Smad7/TGFβ pathway.
In this study, we verified the following: 1) miR-4775 is more highly expressed in metastatic CRC tissues than in non-metastatic tumor tissues; 2) high miR-4775 expression predicts poor prognosis in CRC patients; 3) miR-4775 promotes CRC cell invasion, metastasis and EMT by activating Smad7-dependent TGFβ signaling; and 4) miR-4775-mediated Smad7/TGFβ activation predicts poor survival in CRC patients. These data clarify the role of miR-4775 in CRC progression and provide a new marker for predicting CRC metastasis.
Methods
Patient tissue samples and tissue microarray construction
A total of 544 patients with CRC diagnosed at the Department of General Surgery, Shanghai General Hospital, School of Medicine, Shanghai Jiaotong University from 2005 to 2011 were enrolled. None of the patients received anti-cancer treatment prior to tumor resection, and those with advanced-stage disease received standard postoperative 5-fluorouracil-based chemotherapy. Based on their tumor recurrence status, we divided the 544 patients into tumor recurrence and non-recurrence groups (Additional file
1: Table S1). Tissue samples were obtained during operations. This research was approved by the Ethics Committee of our hospital, and informed consent was obtained from all patients before being enrolled in the study. As previously reported, we constructed a tissue microarray that included samples from these 544 patients [
16].
Cell culture and transfection
We purchased CRC (SW480, HT-29, DLD-1, HCT116, RKO, LoVo, HCT8, and SW62) and normal colon epithelial (FHC) cell lines from the American Type Culture Collection. All cell lines were tested by short tandem repeat analysis and used within 6 months; the last time of authentication was February 2016. All cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco BRL, Grand Island, NY, USA) containing 10% fetal bovine serum (FBS, Invitrogen, Camarillo, CA, USA). Cells in which miR-4775 was either overexpressed or knocked down were generated by transfecting miR-4775 mimics and anti-miR-4775 mimics, respectively. Smad7 was overexpressed or knocked down by transfecting with lv-Smad7 or shRNA Smad7 plasmids, respectively. The empty vector and negative control (NC) vector were used as controls for miR-4775 mimics and anti-miR-4775 mimics, respectively. NC and a scrambled vector were used as controls for the lv-Smad7 and sh-Smad7 vector, respectively. All overexpression and knockdown sequences used in the study were the same as those in a previous report [
17].
qRT-PCR, in situ hybridization and western blotting
We performed a quantitative reverse transcription polymerase chain reaction (qRT-PCR) assay, in situ hybridization, and western blotting as described previously [
13,
14,
18,
19]. U6 and GAPDH were used as internal controls. Antibody data are summarized in Additional file
2: Table S2.
In vitro wound healing, migration and invasion assays
For the wound healing assay, we plated cells into six-well plates and scraped the cells when they grew to cover 90% of the surface. The cells were washed after 48 h, and the closed wound widths were then measured and analyzed.
Matrigel invasion chambers (8.0 μm pore size, BD Biosciences, Franklin Lakes, NJ, USA) and Transwell cell migration plates were used for migration and invasion assays. In total, 5 × 105 cells/ml were seeded in serum-free medium in the upper chamber; 10% FBS was added to the lower chamber. For the invasion assay, the Matrigel coating was overlaid in the upper chamber. The cells were incubated for 24 h and fixed, and 0.1% crystal violet was used to stain the cells that adhered to the underside of the membrane, which were counted under a light microscope (Olympus Corporation, Center Valley, PA).
Immunofluorescence assay
Cells were incubated overnight in glass-bottom dishes, washed three times with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min and washed three times with PBS. The cells were then permeabilized with 0.1% Triton X-100 for 10 min and incubated overnight at 4 °C with primary antibodies against E-cadherin, N-cadherin, and vimentin. After washing, Alexa Fluor 488-conjugated secondary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) were added for 2 h at room temperature. Nuclei were stained for 5 min with 4′,6-diamidino-2-phenylindole (DAPI; Roche, Basel, Switzerland). A confocal laser-scanning microscope (TCS SP8; Leica, Wetzlar, Germany) was used to collect fluorescence images.
In vivo tumorigenesis in nude mice
To explore metastasis ability in vivo, cells overexpressing or knocked down for miR-4775 and control cancer cells (105/ml, 200 μl) were injected into the tail veins of nude mice. All mice were sacrificed after 4 weeks. The tumor colonies formed in the lungs and livers were observed and counted under a microscope using hematoxylin-eosin (HE) staining.
Statistical analysis
For categorical variables, either Fisher’s exact test or a Chi-square test was used to compare differences. For continuous variables, either Student’s t-test or one-way analysis of variance was used to analyze differences. Kaplan–Meier analyses with log-rank tests were used to evaluate disease-free survival (DFS) and overall survival (OS). The Cox proportional hazard model was applied to estimate the hazard ratio and 95% confidence intervals for DFS and OS. SPSS 22.0 statistical software (SPSS Inc., Chicago, IL) was used for the data analysis. A p value < 0.05 indicates a significant difference.
Discussion
We found that miR-4775 was upregulated in CRC tissues, predicting poor prognosis for CRC patients. Indeed, miR-4775 overexpression promoted CRC cell invasion, metastasis and EMT in vitro and in vivo, highlighting the important role of miR-4775 in tumorigenesis and the development of CRC. Furthermore, the present study verified that miR-4775 regulates Smad7/TGFβ pathway activation. The high risk of tumor recurrence and metastasis suggests that CRC patients with high miR-4775 expression and Smad7/TGFβ pathway activation should receive individualized and more aggressive treatment after tumor resection.
Although we have achieved a better understanding of the genetic alterations in CRC in recent years, the prognosis of CRC patients remains poor because of cancer cell invasion and distant metastasis [
24], which result from the inherent heterogeneity and complex gene interactions of CRC [
25]. Accordingly, discovering new biological mechanisms and novel regulators of CRC will provide a better approach for controlling CRC development [
4]. In the present, we identified miR-4775 as a driver of CRC progression. These findings are consistent with a previous report that miR-4775 acts as an onco-miRNA and promotes breast cancer development via a dual effect on the ERBB2/Her2 gene [
8]. It is widely accepted that tumor invasion and metastasis are the main driving forces of relapse [
2], and herein, high miR-4775 expression was positively correlated with TNM stage. Furthermore, the level of miR-4775 expression was clearly higher in the tumor recurrence group than in the non-recurrence group. Most importantly, high expression of miR-4775 predicted poor OS in the tumor recurrence group, though no such significant correlation was found in the non-recurrent group. Overexpressing miR-4775 increased CRC cell wound healing, migration and invasion ability and enhanced CRC cell lung and liver metastasis in nude mice. EMT has been widely reported as a critical step in tumor invasion and metastasis [
26]. For example, LncRNA-ATB was shown to induce hepatocellular carcinoma EMT and invasion by competitively binding to and inhibiting miR-200 family members and then upregulating ZEB1 and ZEB2 [
27]. We also verified that overexpressing miR-4775 induces EMT in CRC cells, with changes in cell morphology, upregulation of the mesenchyme markers N-cadherin and vimentin, and downregulation of the epithelial cell marker E-cadherin. These findings further support our previous hypothesis that miR-4775 promotes the progression and recurrence of CRC in patients by increasing the invasive and metastatic abilities of CRC cells.
TGFβ signaling is widely reported to be involved in several biological processes, such as tumor cell growth, apoptosis, adhesion, migration, and differentiation [
28]. More importantly, in advanced stages of several epithelial cancers, TGFβ functions as a potential activator of cancer progression and metastasis by promoting EMT [
29]. TGFβ signaling is activated by the interaction of heterodimeric transmembrane serine/threonine kinase complexes, including type I (TGFβ RI) and type II (TGFβ RII) receptors, with cytokines [
30]. In the inactivation process, Smad7 blocks R-Smad phosphorylation by occupying the catalytic domain of TGFβ RI. Inhibiting Smad7 leads to activation of R-Smads via phosphorylation through TGFβ RI. Activated R-Smads (p-Smad2 and p-Samd3) form a complex with Smad4, which translocates to the nucleus and regulates the transcription of specific target genes [
31]. In the present study, the promoting function of high miR-4775 expression was weakened by overexpression of Smad7, and high miR-4775 expression was generally found to be concomitant with low Smad7 and high p-Smad2 and p-Smad3. Smad7 levels decreased and p-Smad2 and p-Smad3 levels increased gradually with enhanced miR-4775 expression. These findings indicate that the underlying mechanism by which miR-4775 promotes CRC progression is related to canonical activation of the Smad7/TGFβ pathway. In addition, other miRNAs have been described as affecting EMT in CRC by targeting the Smad7/TGFβ pathway, including miR1269, miR224 and miR200c [
7,
19,
26]. Nonetheless, it needs to be further explored whether miR-4775 has synergistic effects with these reported miRNAs on EMT in CRC and whether the impact of miR-4775 is more important than that of these miRNAs.
Conclusions
In summary, our present study demonstrates the critical roles of miR-4775 and the Smad7/TGFβ pathway in the progression of CRC. Moreover, we identified a novel activation mechanism of the TGFβ pathway in CRC, providing a new treatment strategy for improving the prognosis of these patients.
Acknowledgements
Not applicable.