Accepted Manuscript EDIL3 is a novel regulator of epithelial mesenchymal transition controlling early recurrence of hepatocellular carcinoma Hongping Xia, Chen Jianxiang, Ming Shi, Hengjun Gao, Sekar Karthik, Seshachalam Veerabrahma Pratap, London Lucien P.J. Ooi, Kam M. Hui PII: DOI: Reference:
S0168-8278(15)00327-X http://dx.doi.org/10.1016/j.jhep.2015.05.005 JHEPAT 5672
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
7 January 2015 30 March 2015 4 May 2015
Please cite this article as: Xia, H., Jianxiang, C., Shi, M., Gao, H., Karthik, S., Pratap, S.V., Ooi, L.L.P.J., Hui, K.M., EDIL3 is a novel regulator of epithelial mesenchymal transition controlling early recurrence of hepatocellular carcinoma, Journal of Hepatology (2015), doi: http://dx.doi.org/10.1016/j.jhep.2015.05.005
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
EDIL3 is a novel regulator of epithelial mesenchymal transition controlling early recurrence of hepatocellular carcinoma
Hongping Xia1, Chen Jianxiang1, Ming Shi2, Hengjun Gao2, Sekar Karthik1, Seshachalam Veerabrahma Pratap1, London Lucien P.J. Ooi3,4, and Kam M. Hui1,5-7,*
1
Laboratory of Cancer Genomics, Division of Cellular and Molecular Research, Humphrey Oei
Institute of Cancer Research, National Cancer Centre, Singapore; 2Department of Hepatobiliary Oncology, Cancer Center, Sun Yat-sen University, Guangzhou, 510060, P.R. China; 3
Department of Surgical Oncology, National Cancer Centre, Singapore; 4Department of General
Surgery, Singapore General Hospital, Singapore; 5Cancer and Stem Cell Biology Program, Duke-NUS Graduate Medical School, Singapore; 6Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore; 7Institute of Molecular and Cell Biology, A*STAR, Biopolis Drive Proteos, Singapore
Corresponding author*: Professor Kam M. Hui, Division of Cellular and Molecular Research, National Cancer Centre Singapore; Tel: (65) 6436-8338; Fax: (65) 6226-3843; E-mail: [email protected].
Electronic word count: 5955 Number of Figures: 8
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Abbreviations: HCC, hepatocellular carcinoma; EDIL3, EGF-like repeat and discoidin I-like domain-containing protein 3; DEL1, endothelial cell locus 1; IHC, immunohistochemistry; CTCs, circulating tumor cells; CFI, cancer-free interval; DMEM, Dulbecco’s Modified Eagle’s medium; HBV, hepatitis virus B; HCV, hepatitis virus C; RIPA radioimmunoprecipitation; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; ELISA, enzyme-linked immunosorbent assay; SCID, severe combined immunodeficiency; TUNEL, Terminal deoxynucleotidyl transferase dUTP nick end labelling; MVD, microvessel density; BrdU, 5-Bromo-2’deoxyuridine; IPA, Ingenuity Pathway Analysis.
Running title: EDIL3 in hepatocellular carcinoma
Key words: Hepatocellular carcinoma; Epithelial-mesenchymal transition; EDIL3; ERK; TGF-β; tumor angiogenesis.
Conflict of interest: None to declare.
Financial support: This work was supported by grants from the SingHealth Foundation and National Medical Research Council of Singapore.
Author's contributions: H.P.X. and K.M.H. designed the project and wrote the manuscript. H.P.X and J.X.C performed all biochemical experiments. M.S, H.G and L.L.O provide the clinical samples. S.K and S.V.P support for bioinformatics analysis. All authors discussed the results and implications and commented on the manuscript. 2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Abstract Background & Aims: Patients with advanced hepatocellular carcinoma (HCC) continue to have a dismal prognosis. Early recurrence, metastases and angiogenesis are the major obstacles to improve the outcome of HCC. Epithelial-mesenchymal transition (EMT) is a key contributor to cancer metastasis and recurrence, which are the major obstacles to improve prognosis of HCC. Methods: Combining gene expression profile of HCC samples with or without early recurrence and established cell lines with epithelial or mesenchymal phenotype, EDIL3 was identified as a novel regulator of EMT. The expression of EDIL3 was evaluated by quantitative PCR, western blotting or immunohistochemistry. The effects of EDIL3 on the angiogenesis and metastasis of HCC cells were examined by wound healing, Matrigel invasion and tube formation assay in vivo and orthotopic xenograft mouse model of HCC in vivo. The signalling pathways of EDIL3 mediated were investigated through microarray and western blotting analysis. Results: EDIL3 was identified as a novel regulator of EMT, which contributes to angiogenesis, metastasis and recurrence of HCC. EDIL3 induces EMT and promotes HCC migration, invasion and angiogenesis in vitro. Mechanistically, overexpression of EDIL3, which was regulated by downregulation of miR-137 in HCC, triggered the activation of ERK and TGF-β signaling through interactions with αvβ3 integrin. Blocking ERK and TGF-β signaling overcomes EDIL3 induced angiogenesis and invasion. Using the orthotopic xenograft mouse model of HCC, we demonstrated that EDIL3 enhanced the tumorigenic, metastatic and angiogenesis potential of HCC in vivo. Conclusions: EDIL3 mediated activation of TGF-β and ERK signalling could provide therapeutic implications for HCC.
3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Introduction Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer and one of the leading causes of cancer-related deaths worldwide [1]. Population-based studies have shown that the incidence rate continues to approximate the death rate of HCC, indicating that most patients who develop HCC die of the disease [1]. Some of the recognized risk factors associated with HCC include chronic infection with hepatitis B virus (HBV) or hepatitis C virus (HCV), excessive alcohol intake, exposure to aflatoxin and, more recently added, chronic lifestyle diseases such as diabetes and obesity [2]. Despite recent advances in our understanding of the genetic landscape of HCC, the molecular mechanisms underlying hepatocarcinogenesis remain unclear and the prognosis for patients with advanced HCC remain dismal [3, 4].
Early
recurrence, metastasis and angiogenesis are the major obstacles to improving the clinical outcome of these patients [5, 6]. The epithelial-mesenchymal transition (EMT) is a key step in cancer recurrence and metastasis by which epithelial cells lose their cell polarity and cell-cell adhesion, and gain migratory and invasive properties to become mesenchymal cells [7-9].
The EMT process
enables cancerous cells to depart from the primary tumour, invading surrounding stromal tissue and be disseminated to distant organs. EMT has also been shown to confer efficient tumourigenicity to murine breast cancer cells by up-regulating their expression of the proangiogenic factor VEGF-A and increasing tumour angiogenesis [10]. The expression of EMT markers, including vimentin, twist, ZEB1, ZEB2, snail, slug and E-cadherin, has been investigated in primary HCC tumours, adjacent non-tumoural liver tissues and circulating tumour cells (CTCs) in HCC patients [11, 12]. It has been reported that a decreased expression of Ecadherin in HCC patients significantly reduced the cancer-free interval (CFI) and induced a more 4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
aggressive phenotype [13]. Moreover, overexpression of Snail and Twist has been associated with the down-regulation of E-cadherin and reduced overall survival [11]. Tissue microarray studies also demonstrated that overexpression of vimentin was significantly associated with HCC metastasis [14]. These observations demonstrated that EMT regulatory molecules may play critical roles in HCC progression. To identify novel EMT regulatory molecules and investigate their correlation with angiogenesis, recurrence and metastasis in HCC, we have analysed and compared the gene expression profile between samples of HCC patients with early recurrent disease and liver cancer cell lines with epithelial or mesenchymal phenotype [15, 16] to identify potential novel EMT genes associated with the early recurrence of HCC.
Materials and Methods All the details of Materials and Methods are provided in Supplementary Materials and Methods (available online).
Results
EDIL3 is significantly up-regulated in in HCC samples of patients with early recurrent disease and poor survival We have previously established a global gene expression profile database of histologically normal liver tissues and tumour tissues of HCC patients with early recurrent disease using Affymetrix Human Genome U133 plus 2.0 arrays [16]. To facilitate the identification of novel early recurrence-related genes associated with EMT in HCC, we have analysed the expression 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
profiles of samples of HCC patients with early recurrence (“early recurrence” has been defined as recurrent disease detected within a 2-year time duration after curative hepatic resection) in conjunction with the expression profiles of liver cancer cell lines with demonstrated epithelial or mesenchymal phenotype based on the expression of E-cadherin and Vimentin [15, 16]. A group of 23 genes which expression was modulated were shortlisted (Fig. 1A, Supplementary Table S1). Among these, the expression of EDIL3 was up-regulated in samples of early HCC recurrence compared to histologically normal liver tissues. By qRT-PCR analysis, it was further demonstrated that EDIL3 expression was more markedly increased in samples of HCC patients with early recurrent disease compared to samples of HCC patients with non-recurrent disease (Fig. 1B). The patients’ clinicopathological features in HCC and survival univariate and multivariate analyses were shown in the Supplementary Table S2. Moreover, the expression of EDIL3 was shown to be significantly positive correlated with vimentin (VIM), a mesenchymal marker and negative correlated with E-cadherin (CDH1), an epithelial marker (Fig. 1C and 1D). The expression of EDIL3 was further studied in an independent cohort of 20 pairs of HCC tumour tissues (10 T-R and 10 T-NR) by IHC staining and EDIL3 protein expression was significantly increased in HCC tumour tissues (Fig. 1E and 1F). Since EDIL3 is known to be produced by endothelial cells, we also used the double-staining immunohistochemistry for the endothelial marker CD34 (brown staining) and EDIL3 (red staining). We observed that EDIL3 is also expressed in some intratumoral endothelial cells (Fig. S1A). Interestingly, the results also indicated that EDIL3 produced by tumoral cells promote adjacent endothelial cell growth (Fig.S1B). The correlation between EDIL3 and Vimentin/E-Cadherin expression was also indicated by immunohistochemistry analysis (Fig.S1C). When the median EDIL3 expression was calculated for all the fifty HCC samples studied by qRT-PCR and used as the cut-off for Fisher's 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
exact test and Kaplan-Meier analysis, it was demonstrated that high EDIL3 expression was significantly associated with a shorter overall survival (Fig. 1G). Consistent with EDIL3 being a secreted glycoprotein, EDIL3 protein expression was also shown to be significantly higher in the plasma of HCC patients compared to normal individuals by ELISA analysis. It is also showed that the EDIL3 level is significantly higher in the HCC patients with early recurrence than the HCC patients without early recurrence, suggesting EDIL3 can be a potential non-invasive diagnostic biomarker for HCC with early recurrence (Fig. 1H).
EDIL3 is a novel regulator of EMT in HCC Previously, we established that epithelial liver cancer cells such as HepG2, Hep3B, HuH7 and PLC/PRF/5 had high E-cadherin (CDH1) and low vimentin (VIM) expression. In comparison, liver cancer cells with a mesenchymal phenotype such as HLE, SK-HEP-1, SNU-449 and Mahlavu had low E-cadherin and high vimentin expression [17]. When the expression of EDIL3 was studied with the same panel of liver cancer cell lines by qRT-PCR and western blotting, we observed that EDIL3 expression was significantly higher in liver cancer cells with a mesenchymal phenotype than in the cells with an epithelial phenotype (Figs. 2A and 2B). Similar observations could be made using independent published microarray data for liver cancer cell lines [15].
EDIL3 expression was significantly correlated with expression of the
mesenchymal marker VIM and inversely correlated with the epithelial marker CDH1 (Figs. S2A and S2B). Next, epithelial HuH7 cells were stably transfected with either pLenti-EDIL3 or pLenticontrol vector. The stable cells were tentatively designated as HuH7-EDIL3 or HuH7-control, respectively. The expression of EDIL3 in these cells was confirmed by qRT-PCR (Fig. S2C). 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Compared to pLenti-control-transfected cells, the up-regulation of EDIL3 was associated with dramatic morphological changes observed in the HuH7-EDIL3 cells: from an epithelial cobblestone phenotype to an elongated fibroblastic phenotype, which is indicative of EMT (Fig. 2C). The induction of EMT in the HuH7-EDIL3 cells was also associated with reduced Ecadherin and elevated vimentin expression (Figs. 2E and 2G). Similarly, HLE cells with a mesenchymal phenotype and high EDIL expression were used as the recipient cells for the transfection of EDIL3 shRNA. Following the silencing of EDIL3 (Fig. S2D), striking morphological changes consistent with those of the mesenchymal-to-epithelial transition (MET) were observed (Fig. 2D). The up-regulation of E-cadherin and reduced vimentin expression were also observed (Fig. 2F and 2H).
Effect of EDIL3 expression on HCC cell migration and invasion in vitro EMT has been indicated as a key step in initiating cancer cell migration [17]. The migration potential of the HuH7-EDIL3 and HLE-shEDIL3 cells was studied using the wound healing assay. It was observed that stable overexpression of EDIL3 in the epithelial HuH7 cells significantly promoted cell migration (Fig. 3A) while stable knockdown of EDIL3 in the mesenchymal HLE cells significantly inhibited cell migration (Fig. 3B). Consistent with results of the wound healing assay, overexpression of EDIL3 in the HuH7-EDIL3 cells promoted invasion (Fig. 3C and 3D) and the repression of EDIL3 expression in the HLE-shEDIL3 cells reduced the number of cells invading the extracellular matrix gel (Fig. 3E and 3F). Meanwhile, we have investigated the effects of EDIL3 modulation on proliferation and apoptosis of HCC cells using soft agar colony formation assay and anoikis assay. Anoikis is a form of apoptosis that is induced by anchorage-dependent cells detaching from the surrounding extracellular matrix 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
(ECM)[18]. The results showed that overexpression of EDIL3 significantly promoted the anchorage-independent growth (Fig. S2E and S2F) and decreased the anoikis rate of HCC cells (Fig. S2G and S2H), suggesting that EDIL3 affects cell proliferation and apoptosis. Taken together, these data indicated that EDIL3 played an important role in regulating HCC cell migration and invasion.
EDIL3 enhances recruitment of endothelial cells and promotes capillary tube formation in vitro Angiogenesis is a key early step in the cancer invasion-metastasis cascade. EDIL3, being a glycoprotein secreted by endothelial cells, is likely to be involved in tumour angiogenesis. We have therefore studied the role of EDIL3 in HCC angiogenesis by in vitro endothelial recruitment and capillary tube formation assays using stably transfected HuH7 and HLE cells with different EDIL3 expression. In the presence of HLE cells with high EDIL3 expression, HUVECs migrated more efficiently through the transwell pores compared to those grown in the absence of tumour cells conditioned medium (NC) or in the presence of HLE-shEDIL3 cells with low EDIL3 expression. The knockdown of EDIL3 expression could significantly suppress the effects of HLE cells on HUVEC migration (Fig. 4A and 4C). Furthermore, cell-conditioned medium of EDIL3high HLE cells compared to the absence of tumour cells conditioned medium (NC) promoted the HUVECs to develop more capillary-like structures. Tube formation was reduced dramatically in HUVECs that were grown in conditioned medium from HLE-shEDIL3 cells (Fig. 4B and 4D). In contrast, the overexpression of EDIL3 in HuH7 cells enhanced HUVEC migration (Fig. 4E and 4G) and capillary tube formation (Fig. 4F and 4H). These data suggested that high EDIL3
9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
expression in HCC cells can promote endothelial cells migration and tumour angiogenesis in vitro.
miR-137 is a critical upstream regulator of EDIL3 To identify the potential upstream miRNA regulator of EDIL3, we employed the miRecords that has incorporated ten miRNA target prediction algorithms [19] to predict miRNAs that could be the potential regulators of EDIL3 (Table S3). Using the luciferase reporter assay, we validated that EDIL3 could potentially be regulated by miR-137. We then cloned a sequence containing the predicted 3′-UTR target site of the EDIL3 mRNA and its mutated sequence into the pGL3 luciferase reporter gene to generate pGL3-EDIL3-3’UTR-wt and pGL3-EDIL3-3’UTR-mut vector respectively (Fig. 5A). These vectors were co-transfected into the HLE cells together with either the p-miR-137 vector or the p-miR-control vector. A renilla luciferase vector (pRL-TK) was used to normalize the differences in transfection efficiency. The over-expression of miR-137 was validated by RT-qPCR (Fig. 5B). Luciferase activity in HLE cells co-transfected with either the p-miR-137 or the pGL3-EDIL3-3’UTR-wt vector was significantly decreased compared to the control (Fig. 5C). Over-expression of miR-137 decreased EDIL3 protein expression in HLE cells (Fig. 5D) and significantly inhibited HLE cell invasion and induced endothelial cell capillary tube formation (Fig. 5E). MiR-137 expression was significantly down-regulated in tumor tissues of HCC patients when compared to adjacent histologically normal liver tissues and miR-137 expression was significantly down-regulated in tumor samples of HCC patients with early recurrent disease compared to samples of patients with non-recurrent disease (Fig. 5F). The decreased expression of miR-137 also significantly correlated with increased expression of
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
EDIL3 in HCC samples (Fig. 5G). Therefore, overexpression of EDIL3 was correlated with downregulation of miR-137 in HCC and miR-137 is a critical upstream regulator of EDIL3.
EDIL3 regulates TGF-beta and ERK signalling through binding to αvβ3 integrin EDIL3 was initially reported to be a new ligand for the αvβ3 integrin receptor regulating embryonic vascular morphogenesis and remodelling [20, 21]. Co-immunoprecipitation experiments demonstrated that EDIL3 could be specifically co-precipitated with the αvβ3 integrin (Fig. 6A), demonstrating that EDIL3 and αvβ3 integrin physically interacted with each other. Next, immunofluorescence analysis demonstrated that EDIl3 and αvβ3 integrin co-localized in HLE cells (Fig. 6B). Recent studies showed that αvβ3 integrin promotes latent TGF-β-activation by human cardiac fibroblast contraction [22]. We then examined whether the over-expression of EDIL3 could promote the increase production of TGF-β1 in liver cancer cells and observed that the production of TGF-β1 was significantly increased in both HuH7 and PLC/PRF/5 cells stably transfected with EDIL3 (Fig. 6C). Interestingly, using data reported in the Cancer Cell Line Encyclopedia (CCLE) dataset (http://www.broadinstitute.org/ccle) [15], we compared the group of liver cancer cells with EDIL3 high expression (array signal intensity ≥ log26) and EDIL3 low expression (array signal intensity < log26) and identified many TGF-β-regulated genes that were also differentially expressed (>10 fold) between EDIL3 high expressing and EDIL3 low expressing liver cancer cells by Ingenuity Pathway Analysis (IPA) (Table S4 and Fig. S3A). Hierarchical clustering analysis further showed that these differentially expressed genes could also discriminate cells with the epithelial or mesenchymal phenotype (Fig. S3B). Among these differentially expressed genes, expression of TGFB1I1 and TGFB2 showed significantly correlation with the expression level of EDIL3 in the panel of liver cancer cells, Pearson r = 0.67, 11
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
r = 0.68, respectively (Figs. 6D and 6E). These data suggested that over-expression of EDIL3 can regulate TGF-β signaling through binding to αvβ3 integrin in liver cancer cells. To further explore the molecular mechanisms and downstream signal pathways of EDIL3 in HCC cells, we also performed microarray and bioinformatics analysis to identify the genes that are differentially expressed between HLE-shEDIL3 and HLE-shControl cells. Among the dysregulated genes, pseudopodium-enriched atypical kinase 1 (PEAK1) was significantly downregulated in the HLE-shEDIL3 cells. Furthermore, knockdown of EDIL3 in HLE and Mahlavu cells significantly decreased the expression of PEAK1, while overexpression of EDIL3 in HuH7 and PLC/PRF/5 cells increased the expression of PEAK1 (Figs. 6F and 6G). Moreover, EDIL3 expression was found to be significantly correlated with PEAK1 expression in HCC patient tumor samples (Fig. 6H) and was further validated by IHC staining with another 20 cases of HCC tissue samples (Fig. 6I). Next, we deciphered the PEAK1-associated regulatory signalling pathways involved with the over-expression of EDIL3 in HCC cells. IPA analysis showed that EDIL3 could interact with PEAK1 through the SRC family kinases (Figs. S4A and S4B). We subsequently demonstrated that the over-expression of EDIL3 not only significantly enhanced the expression of PEAK1 but also induced the phosphorylation of SRC, ERK and Smad2, suggesting the activation of ERK and TGF-β signaling (Fig. 6J). These data further suggest that over-expression of EDIL3 also activates ERK signalling which has been shown to be strongly associated with tumour angiogenesis and metastasis.
Blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated angiogenesis and invasion
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Since both TGF-beta and ERK signaling have been shown activated by overexpression of EDIL3, we next investigate the effects of TGF-beta and ERK signaling inhibitors on the established EDIL3 overexpression HCC cells. When HuH7-p-lenti-EDIL3 stable cells were treated with VX11e (1µg/ml, ChemieTek, Indianapolis, IN, USA), a potent, selective, and orally bioavailable inhibitor of ERK, or LY2109761 (1µg/ml, MedChemexpress, Shanghai, China), a selective TGFβ receptor type I/II dual inhibitor, strong inhibition of p-ERK and p-Smad2 expression in the HuH7-p-lenti-EDIL3 stable cells was detected (Fig. 7A). The MTS results showed that the cell viability was inhibited with the increasing dose of VX-11e but not LY2109761 and EDIL3 overexpression HuH7-p-lenti-EDIL3 cell is more sensitive to VX-11e treatment (Fig. S5A and S5B). . The MEK inhibitor also significantly inhibits proliferation (Fig.S5C and S5D) and induces apoptosis (Fig. S5E) of EDIL3 overexpression cells, while TGF-beta inhibitor is unable to blunt the apoptosis of EDIL3 overexpression cells (Fig. S5F). To determine if blocking the TGF-β pathway would sensitize the EDIL3 overexpression HuH7-p-lenti-EDIL3 cell to VX-11e, we examined the effect of the presence or absence of LY2109761 (2.5µg/ml) on the sensitivity of HuH7-p-lenti-EDIL3 cells to VX-11e. Interestingly, the results showed that the presence of LY2109761 (2.5µg/ml) did not significantly inhibit cell growth but was sufficient to sensitize the EDIL3 overexpression HuH7-p-lenti-EDIL3 cell to VX-11e (Fig. 7B). The colony formation assay showed that VX-11e treatment significantly decreased colony formation ability of HuH7p-lenti-EDIL3 stable cells and LY2109761 can sensitize this effect (Fig. 7C and 7D). Blocking the TGF-beta and ERK signalling was also shown significant inhibition effect on EDIL3 overexpression HuH7-p-lenti-EDIL3 cell invasion ability (Fig. 7E and 7F) and the HuH7-p-lentiEDIL3 cells conditional medium enhanced HUVEC capillary tube formation (Fig. 7G and 7H), suggesting that blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated 13
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
angiogenesis and invasion, which provide a promising therapeutic potential for the treatment of HCC with overexpression of EDIL3.
Overexpression of EDIL3 promotes tumour metastasis and angiogenesis in vivo To correlate our observations in vivo, we implanted HuH7-pLenti-EDIL3 and HuH7-pLenticontrol cells with luciferase orthotopically into nude mice. The results indicated that stable overexpression of EDIL3 in the HuH7 cells significantly promoted tumour growth (Fig. 8A and 8B). In addition, lung tissues harvested from mice at the end of the experiments inoculated with the HuH7-pLenti-EDIL3 cells gave prominent bioluminescent signals, indicating the presence of lung metastasis. On the other hand, no lung bioluminescent signal was detected in mice inoculated with HuH7-pLent-control cells (Fig. 8C). IHC staining further demonstrated the significant overexpression of EDIL3 in tumour tissues induced by HuH7-pLenti-EDIL3 cells compared to HuH7-pLenti-control cells (Fig. 8D).
Tumour tissues harvested from mice
implanted with the HuH7-pLenti-EDIL3 cells also showed a significantly higher CD34 expression compared to tumour tissues implanted with the HuH7-p-lenti-control cells, thus indicating that overexpression of EDIL3 promoted tumour angiogenesis of HCC in vivo (Fig. 8D). Increased expression of PEAK1 and phosphorylation of Smad2, SRC and ERK were also observed in tumour tissues harvested from mice implanted with HuH7-p-lenti-EDIL3 cells (Fig. 8D), further suggesting that overexpression of EDIL3 activated the TGF-beta and ERK signalling. A schematic diagram has been put forward to summarize the possible regulatory circuitry that governs EDIL3 contributing to EMT,metastasis/invasion, early recurrence and angiogenesis in HCC (Fig. 8E).
14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Discussion
High EDIL3 expression was shown to associate with early HCC recurrence and a mesenchymal cell phenotype. In vitro studies demonstrated that EDIL3 is a novel regulator of EMT, migration, invasion and tumor angiogenesis in HCC cells. In vivo studies further validated that over-expression of EDIL3 in HCC cells promotes tumour growth, metastasis and angiogenesis. MiR-137 is involved in the upstream regulation of EDIL3, while EDIL3 overexpression regulates the TGF-β and ERK signalling pathways downstream through binding to αvβ3 integrin in HCC cells. EDIL3 was cloned and characterized in 1998 [20]. The EDIL3 protein contains three EGF-like repeats homologous to those in Notch and related proteins, including an EGF-like repeat that contains an RGD motif and two discoidin I-like domains. EDIL3 has been shown to be a matrix protein, to promote adhesion of endothelial cells through interaction with the αvβ3 integrin receptor, and to inhibit the formation of vascular-like structures. Previous studies also showed that EDIL3 is an endogenous leukocyte-endothelial adhesion inhibitor and limits the recruitment of inflammatory cells [23] indicating that EDIL3 plays an important role in mediating angiogenesis and may be important in vessel wall remodelling and development [20]. Many human tumours have been reported to be dependent on angiogenesis for growth and development. Human osteosarcoma cells and murine Lewis lung carcinoma cells engineered to express EDIL3 resulted in two- to four-fold increases in tumour volume [24]. Down-regulated of EDIL3 expression has also been shown to inhibit the growth of colon cancer cells [25]. Overexpression of EDIL3 accelerates tumour growth by enhancing vascular formation, thus suggesting that EDIL3 is a potential target for anti-angiogenic agents. 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Tumour angiogenesis is a complex process that is regulated by many factors including vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and basic fibroblast growth factor (FGF) [26]. Inhibition of angiogenesis has been extensively studied as a potential therapeutic strategy for HCC. The anti-angiogenic multi-kinase inhibitor sorafenib is the first and only FDA approved target therapy for advanced HCC, but the median survival rate remains unsatisfactory [27]. In this study, we demonstrated that over-expression of EDIL3 increases the ability of HCC cells to promote migration and capillary tube formation by endothelial cells. Since HCC is usually regarded as a highly vascular tumour, thus EDIL3 can provide an attractive potential target for the development of novel anti-angiogenic interventions in HCC. Although previous studies indicated that EDIL3 is over-expressed in HCC and high expression level of EDIL3 in HCC is associated with poor prognosis, the molecular roles of EDIL3 in HCC has not been well investigated [28-30]. In this study, we found that EDIL3 overexpression triggers activate the TGF-β and ERK signaling pathway through interactions with αvβ3 integrin. These results provide important evidence for the development of novel treatment strategies targeting EDIL3 or downstream TGF-β and ERK signaling pathways for the management of HCC. TGF-β plays an important pro-tumorigenic role in HCC mainly by promoting angiogenesis and inducing EMT, invasion and metastasis. [31]. Inhibition of TGF-β receptor I kinase has been shown to block HCC growth and angiogenesis [32]. The ERK signalling has been shown to play important roles in the development of HCC and blocking ERK pathway has been demonstrated as a promising treatment for HCC [33]. Here, we shown that blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated angiogenesis
16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
and invasion. Therefore, the combination inhibition of TGF-β and ERK signaling pathways may provide a promising targeted therapy for HCC cells with high EDIL3 expression. Previously, we have identified a panel of early HCC recurrence-associated miRNAs, including miR-137, by comparing the miRNA expression in samples of recurrent and nonrecurrent human HCC tissue samples using microarrays [17]. Here, we further demonstrated that EDIL3 expression was regulated by miR-137. Its expression was shown to be significantly down-regulated in HCC tissues compared to adjacent histologically normal liver samples and in HCC samples of patients with early recurrence compared to samples of non-recurrent disease. Therefore, restoring miR-137 expression in human HCC such as the introduction of synthetic miRNA mimics could offer appealing miRNA-based therapeutic strategies.
Acknowledgements This work was supported by grants from the SingHealth Foundation and National Medical Research Council of Singapore. We thank Dr. Tony Lim (Pathologist, Singapore General Hospital) for helping on providing tissue slides.
References [1]
Maluccio M, Covey A. Recent progress in understanding, diagnosing, and treating
hepatocellular carcinoma. CA: A Cancer Journal for Clinicians 2012;62:394-399. [2]
Sherman M, Llovet JM. Smoking, hepatitis B virus infection, and development of
hepatocellular carcinoma. Journal of the National Cancer Institute 2011;103:1642-1643. [3]
Villanueva A, Llovet JM. Liver Cancer in 2013: Mutational landscape of HCC [mdash]
the end of the beginning. Nature Reviews Clinical Oncology 2014. 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[4]
Nault J-C, Zucman-Rossi J. Genetics of hepatocellular carcinoma: The next generation. J
Hepatol 2014;60:224-226. [5]
Bruix J, Gores GJ, Mazzaferro V. Hepatocellular carcinoma: clinical frontiers and
perspectives. Gut 2014:gutjnl-2013-306627. [6]
Zhu AX, Duda DG, Sahani DV, Jain RK. HCC and angiogenesis: possible targets and
future directions. Nature Reviews Clinical Oncology 2011;8:292-301. [7]
Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelial–mesenchymal
transition. Nature Reviews Molecular Cell Biology 2014;15:178-196. [8]
Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. The Journal of
clinical investigation 2009;119:1420-1428. [9]
Craene BD, Berx G. Regulatory networks defining EMT during cancer initiation and
progression. Nat Rev Cancer 2013;13:97-110. [10]
Fantozzi A, Gruber DC, Pisarsky L, Heck C, Kunita A, Yilmaz M, et al. VEGF-Mediated
Angiogenesis Links EMT-Induced Cancer Stemness to Tumor Initiation. Cancer Research 2014. [11]
Yang MH, Chen CL, Chau GY, Chiou SH, Su CW, Chou TY, et al. Comprehensive
analysis of the independent effect of twist and snail in promoting metastasis of hepatocellular carcinoma. Hepatology 2009;50:1464-1474. [12]
Li Y, Xu S, Li J, Han K, Pi H, Zheng L, et al. Epithelial–mesenchymal transition markers
expressed in circulating tumor cells in hepatocellular carcinoma patients with different stages of disease. Cell death & disease 2013;4:e831. [13]
Giannelli G. The epithelial‐mesenchymal transition: Fact or fiction in cancer?
Hepatology 2009;50:1344-1346.
18
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[14]
Hu L, Lau SH, Tzang C-H, Wen J-M, Wang W, Xie D, et al. Association of Vimentin
overexpression and hepatocellular carcinoma metastasis. Oncogene 2004;23:298-302. [15]
Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, et al. The
Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 2012;483:603-607. [16]
Wang SM, Ooi LLP, Hui KM. Identification and validation of a novel gene signature
associated with the recurrence of human hepatocellular carcinoma. Clinical Cancer Research 2007;13:6275-6283. [17]
Xia H, Ooi LLPJ, Hui KM. MicroRNA-216a/217-induced epithelial-mesenchymal
transition targets PTEN and SMAD7 to promote drug resistance and recurrence of liver cancer. Hepatology 2013;58:629-641. [18]
Paoli P, Giannoni E, Chiarugi P. Anoikis molecular pathways and its role in cancer
progression. Biochimica et Biophysica Acta (BBA)-Molecular Cell Research 2013;1833:34813498. [19]
Xiao F, Zuo Z, Cai G, Kang S, Gao X, Li T. miRecords: an integrated resource for
microRNA–target interactions. Nucleic Acids Res 2009;37:D105-D110. [20]
Hidai C, Zupancic T, Penta K, Mikhail A, Kawana M, Quertermous EE, et al. Cloning
and characterization of developmental endothelial locus-1: an embryonic endothelial cell protein that binds the αvβ3 integrin receptor. Genes & development 1998;12:21-33. [21]
Rezaee M, Penta K, Quertermous T. Del1 mediates VSMC adhesion, migration, and
proliferation through interaction with integrin αvβ3. American Journal of Physiology-Heart and Circulatory Physiology 2002;282:H1924-H1932.
19
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[22]
Sarrazy V, Koehler A, Chow ML, Zimina E, Li CX, Kato H, et al. Integrins αvβ5 and
αvβ3 promote latent TGF-β1 activation by human cardiac fibroblast contraction. Cardiovascular research 2014;102:407-417. [23]
Choi EY, Chavakis E, Czabanka MA, Langer HF, Fraemohs L, Economopoulou M, et al.
Del-1, an endogenous leukocyte-endothelial adhesion inhibitor, limits inflammatory cell recruitment. Science 2008;322:1101-1104. [24]
Aoka Y, Johnson FL, Penta K, Hirata K-i, Hidai C, Schatzman R, et al. The embryonic
angiogenic factor Del1 accelerates tumor growth by enhancing vascular formation. Microvascular research 2002;64:148-161. [25]
Zou X, Qiao H, Jiang X, Dong X, Jiang H, Sun X. Downregulation of developmentally
regulated endothelial cell locus-1 inhibits the growth of colon cancer. J Biomed Sci 2008;16:33. [26]
Carmeliet P, Jain RK. Molecular mechanisms and clinical applications of angiogenesis.
Nature 2011;473:298-307. [27]
Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc J-F, et al. Sorafenib in
advanced hepatocellular carcinoma. N Engl J Med 2008;359:378-390. [28]
Luo JH, Ren B, Keryanov S, Tseng GC, Rao UN, Monga SP, et al. Transcriptomic and
genomic analysis of human hepatocellular carcinomas and hepatoblastomas. Hepatology 2006;44:1012-1024. [29]
Sun J-C, Liang X-T, Pan K, Wang H, Zhao J-J, Li J-J, et al. High expression level of
EDIL3 in HCC predicts poor prognosis of HCC patients. World journal of gastroenterology: WJG 2010;16:4611.
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[30]
Feng M-X, Ma M-Z, Fu Y, Li J, Wang T, Xue F, et al. Elevated autocrine EDIL3 protects
hepatocellular carcinoma from anoikis through RGD-mediated integrin activation. Molecular cancer 2014;13:226. [31]
Neuzillet C, de Gramont A, Tijeras-Raballand A, de Mestier L, Cros J, Faivre S, et al.
Perspectives of TGF-β inhibition in pancreatic and hepatocellular carcinomas. Oncotarget 2014;5:78. [32]
Mazzocca A, Fransvea E, Lavezzari G, Antonaci S, Giannelli G. Inhibition of
transforming growth factor β receptor I kinase blocks hepatocellular carcinoma growth through neo‐angiogenesis regulation. Hepatology 2009;50:1140-1151. [33]
Liu L, Cao Y, Chen C, Zhang X, McNabola A, Wilkie D, et al. Sorafenib blocks the
RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell apoptosis in hepatocellular carcinoma model PLC/PRF/5. Cancer research 2006;66:11851-11858.
Figure legends
Figure 1. EDIL3 is significantly upregulated in HCC and associated with recurrence and patient survival. (A) The venn diagram of significantly different expressed genes among HCC tissues (T) vs matched normal tissues (MN), HCC with recurrence (R) vs non-recurrence (NR) and liver cancer cells with epithelial phenotype vs cells with mesenchymal phenotype. (B) Validation of EDIL3 expression in a cohort of patient samples (T-R: Tumor Recurrence, T-NR: Tumor Non-Recurrence, MN: Matched Normal and NN: histologically normal liver tissues) by RT-qPCR. (C and D) The expression of EDIL3 was significantly correlated with the expression of the mesenchymal marker vimentin (VIM) and epithelial marker E-cadherin (CDH1) in a 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
group of HCC samples. (E) Representative imaging of IHC staining for validation of the expression of EDIL3 in another panel of HCC tumour tissues (100 ×). Up panel: H&E staining for the tumorous and non-tumorous parts. Down panel: IHC staning for the EDIL3 expression. (F) Imaging analysis and quantification of EDIL3 IHC staining. The IHC quantification was evaluated according to the percentage of cells with positive nuclei. (G) The expression of EDIL3 was associated with the disease-free survival of patients with HCC. The median expression value of EDIL3 was chosen as the cut-off point for survival analysis using the Kaplan-Meier method (*P=0.028). (H) The expression of EDIL3 in the plasma of HCC patients (n=40, 16R, 24NR) compared with normal individuals (n=20).
Figure 2. EDIL3 is a novel regulator of EMT in HCC. (A) Expression of EDIL3 was studied in a panel of liver cancer cell lines by qRT-PCR. (B and C) Expression of EDIL3 was studied in a panel of liver cancer cell lines by western blotting and immunofluorescence analysis showed the location of EDIL3 in HLE cells (C). (D and E) The represent images of morphological change of HuH7-EDIL3 cells from an epithelial cobblestone phenotype to an elongated fibroblastic phenotype, indicative of EMT (D), or HLE-shEDIL3 cells from an elongated fibroblastic phenotype to an epithelial cobblestone phenotype, indicative of MET (E). (F) The expression of EDIL3 and the EMT markers E-cadherin and vimentin in HuH7 cells stably transfected with pLenti-hEDIL3. (G) The expression of EDIL3 and the EMT markers E-cadherin and vimentin in HLE cells stably transfected with shEDIL3. (H and I) The expression of the epithelial marker E-cadherin and the mesenchymal marker vimentin was further analysed in HuH7 and HLE transfected cells by confocal microscopy. The red signal represents staining for
22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
vimentin (upper panel) or E-cadherin (lower panel). Nuclear DNA was detected by staining with Hoechst 33342. Scale bar represents 50 µm.
Figure 3. The effects of EDIL3 on HCC cell migration and invasion in vitro. (A and B) Representative images of cell migration ability in stably transfected HuH7 and HLE cells evaluated by wound healing assay. Twenty four hours after wounding, cells with extended membrane protrusions moved into the wounded areas. The distance of migrated cells travelled in each treatment group is shown. (C and E) Representative images of stably transfected HuH7 (C) and HLE (E) cell invasion. (D and F) The invading cells were quantified by plotting them as the average number of cells per field of view from three different experiments as described.
Figure 4. EDIL3 promotes tumour angiogenesis of HCC in vitro. (A and E) Representative images of endothelial cell migration after incubation with conditioned media (CM) from stably transfected HuH7 and HLE cells using the endothelial recruitment assay. (B and F) Representative tube formation by ECs after incubation with conditioned media (CM) from stably transfected HuH7 and HLE cells using the tube formation assay. (C and G) Quantification of the numbers of migrating endothelial cells in different groups. (D and H) Quantification of the numbers of branchs in different groups, showing their tube forming ability.
Figure 5. miR-137 is identified as an upstream regulator of EDIL3 and is downregulated in HCC samples. (A) The target sequences predicted by RNAhybrid 2.2 or TargetScan and mutations generated in the 3’-UTR of the EDIL3 mRNA are shown. (B) The overexpression of miR-137 was examined by qRT-PCR. (C) Effects of co-transfection of P-miR-137 with wild23
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
type (wt) or mutant (mut) pGL3-EDIL3 constructs into HLE cells on luciferase reporter assays. Data were normalized by the ratio of Firefly and Renilla luciferase activities measured at 48 h post-transfection. The bar graph shows the mean±SD in three independent transfection experiments. *P
S0168-8278(15)00327-X http://dx.doi.org/10.1016/j.jhep.2015.05.005 JHEPAT 5672
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
7 January 2015 30 March 2015 4 May 2015
Please cite this article as: Xia, H., Jianxiang, C., Shi, M., Gao, H., Karthik, S., Pratap, S.V., Ooi, L.L.P.J., Hui, K.M., EDIL3 is a novel regulator of epithelial mesenchymal transition controlling early recurrence of hepatocellular carcinoma, Journal of Hepatology (2015), doi: http://dx.doi.org/10.1016/j.jhep.2015.05.005
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
EDIL3 is a novel regulator of epithelial mesenchymal transition controlling early recurrence of hepatocellular carcinoma
Hongping Xia1, Chen Jianxiang1, Ming Shi2, Hengjun Gao2, Sekar Karthik1, Seshachalam Veerabrahma Pratap1, London Lucien P.J. Ooi3,4, and Kam M. Hui1,5-7,*
1
Laboratory of Cancer Genomics, Division of Cellular and Molecular Research, Humphrey Oei
Institute of Cancer Research, National Cancer Centre, Singapore; 2Department of Hepatobiliary Oncology, Cancer Center, Sun Yat-sen University, Guangzhou, 510060, P.R. China; 3
Department of Surgical Oncology, National Cancer Centre, Singapore; 4Department of General
Surgery, Singapore General Hospital, Singapore; 5Cancer and Stem Cell Biology Program, Duke-NUS Graduate Medical School, Singapore; 6Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore; 7Institute of Molecular and Cell Biology, A*STAR, Biopolis Drive Proteos, Singapore
Corresponding author*: Professor Kam M. Hui, Division of Cellular and Molecular Research, National Cancer Centre Singapore; Tel: (65) 6436-8338; Fax: (65) 6226-3843; E-mail: [email protected].
Electronic word count: 5955 Number of Figures: 8
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Abbreviations: HCC, hepatocellular carcinoma; EDIL3, EGF-like repeat and discoidin I-like domain-containing protein 3; DEL1, endothelial cell locus 1; IHC, immunohistochemistry; CTCs, circulating tumor cells; CFI, cancer-free interval; DMEM, Dulbecco’s Modified Eagle’s medium; HBV, hepatitis virus B; HCV, hepatitis virus C; RIPA radioimmunoprecipitation; RT-PCR, Reverse Transcription-Polymerase Chain Reaction; ELISA, enzyme-linked immunosorbent assay; SCID, severe combined immunodeficiency; TUNEL, Terminal deoxynucleotidyl transferase dUTP nick end labelling; MVD, microvessel density; BrdU, 5-Bromo-2’deoxyuridine; IPA, Ingenuity Pathway Analysis.
Running title: EDIL3 in hepatocellular carcinoma
Key words: Hepatocellular carcinoma; Epithelial-mesenchymal transition; EDIL3; ERK; TGF-β; tumor angiogenesis.
Conflict of interest: None to declare.
Financial support: This work was supported by grants from the SingHealth Foundation and National Medical Research Council of Singapore.
Author's contributions: H.P.X. and K.M.H. designed the project and wrote the manuscript. H.P.X and J.X.C performed all biochemical experiments. M.S, H.G and L.L.O provide the clinical samples. S.K and S.V.P support for bioinformatics analysis. All authors discussed the results and implications and commented on the manuscript. 2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Abstract Background & Aims: Patients with advanced hepatocellular carcinoma (HCC) continue to have a dismal prognosis. Early recurrence, metastases and angiogenesis are the major obstacles to improve the outcome of HCC. Epithelial-mesenchymal transition (EMT) is a key contributor to cancer metastasis and recurrence, which are the major obstacles to improve prognosis of HCC. Methods: Combining gene expression profile of HCC samples with or without early recurrence and established cell lines with epithelial or mesenchymal phenotype, EDIL3 was identified as a novel regulator of EMT. The expression of EDIL3 was evaluated by quantitative PCR, western blotting or immunohistochemistry. The effects of EDIL3 on the angiogenesis and metastasis of HCC cells were examined by wound healing, Matrigel invasion and tube formation assay in vivo and orthotopic xenograft mouse model of HCC in vivo. The signalling pathways of EDIL3 mediated were investigated through microarray and western blotting analysis. Results: EDIL3 was identified as a novel regulator of EMT, which contributes to angiogenesis, metastasis and recurrence of HCC. EDIL3 induces EMT and promotes HCC migration, invasion and angiogenesis in vitro. Mechanistically, overexpression of EDIL3, which was regulated by downregulation of miR-137 in HCC, triggered the activation of ERK and TGF-β signaling through interactions with αvβ3 integrin. Blocking ERK and TGF-β signaling overcomes EDIL3 induced angiogenesis and invasion. Using the orthotopic xenograft mouse model of HCC, we demonstrated that EDIL3 enhanced the tumorigenic, metastatic and angiogenesis potential of HCC in vivo. Conclusions: EDIL3 mediated activation of TGF-β and ERK signalling could provide therapeutic implications for HCC.
3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Introduction Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer and one of the leading causes of cancer-related deaths worldwide [1]. Population-based studies have shown that the incidence rate continues to approximate the death rate of HCC, indicating that most patients who develop HCC die of the disease [1]. Some of the recognized risk factors associated with HCC include chronic infection with hepatitis B virus (HBV) or hepatitis C virus (HCV), excessive alcohol intake, exposure to aflatoxin and, more recently added, chronic lifestyle diseases such as diabetes and obesity [2]. Despite recent advances in our understanding of the genetic landscape of HCC, the molecular mechanisms underlying hepatocarcinogenesis remain unclear and the prognosis for patients with advanced HCC remain dismal [3, 4].
Early
recurrence, metastasis and angiogenesis are the major obstacles to improving the clinical outcome of these patients [5, 6]. The epithelial-mesenchymal transition (EMT) is a key step in cancer recurrence and metastasis by which epithelial cells lose their cell polarity and cell-cell adhesion, and gain migratory and invasive properties to become mesenchymal cells [7-9].
The EMT process
enables cancerous cells to depart from the primary tumour, invading surrounding stromal tissue and be disseminated to distant organs. EMT has also been shown to confer efficient tumourigenicity to murine breast cancer cells by up-regulating their expression of the proangiogenic factor VEGF-A and increasing tumour angiogenesis [10]. The expression of EMT markers, including vimentin, twist, ZEB1, ZEB2, snail, slug and E-cadherin, has been investigated in primary HCC tumours, adjacent non-tumoural liver tissues and circulating tumour cells (CTCs) in HCC patients [11, 12]. It has been reported that a decreased expression of Ecadherin in HCC patients significantly reduced the cancer-free interval (CFI) and induced a more 4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
aggressive phenotype [13]. Moreover, overexpression of Snail and Twist has been associated with the down-regulation of E-cadherin and reduced overall survival [11]. Tissue microarray studies also demonstrated that overexpression of vimentin was significantly associated with HCC metastasis [14]. These observations demonstrated that EMT regulatory molecules may play critical roles in HCC progression. To identify novel EMT regulatory molecules and investigate their correlation with angiogenesis, recurrence and metastasis in HCC, we have analysed and compared the gene expression profile between samples of HCC patients with early recurrent disease and liver cancer cell lines with epithelial or mesenchymal phenotype [15, 16] to identify potential novel EMT genes associated with the early recurrence of HCC.
Materials and Methods All the details of Materials and Methods are provided in Supplementary Materials and Methods (available online).
Results
EDIL3 is significantly up-regulated in in HCC samples of patients with early recurrent disease and poor survival We have previously established a global gene expression profile database of histologically normal liver tissues and tumour tissues of HCC patients with early recurrent disease using Affymetrix Human Genome U133 plus 2.0 arrays [16]. To facilitate the identification of novel early recurrence-related genes associated with EMT in HCC, we have analysed the expression 5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
profiles of samples of HCC patients with early recurrence (“early recurrence” has been defined as recurrent disease detected within a 2-year time duration after curative hepatic resection) in conjunction with the expression profiles of liver cancer cell lines with demonstrated epithelial or mesenchymal phenotype based on the expression of E-cadherin and Vimentin [15, 16]. A group of 23 genes which expression was modulated were shortlisted (Fig. 1A, Supplementary Table S1). Among these, the expression of EDIL3 was up-regulated in samples of early HCC recurrence compared to histologically normal liver tissues. By qRT-PCR analysis, it was further demonstrated that EDIL3 expression was more markedly increased in samples of HCC patients with early recurrent disease compared to samples of HCC patients with non-recurrent disease (Fig. 1B). The patients’ clinicopathological features in HCC and survival univariate and multivariate analyses were shown in the Supplementary Table S2. Moreover, the expression of EDIL3 was shown to be significantly positive correlated with vimentin (VIM), a mesenchymal marker and negative correlated with E-cadherin (CDH1), an epithelial marker (Fig. 1C and 1D). The expression of EDIL3 was further studied in an independent cohort of 20 pairs of HCC tumour tissues (10 T-R and 10 T-NR) by IHC staining and EDIL3 protein expression was significantly increased in HCC tumour tissues (Fig. 1E and 1F). Since EDIL3 is known to be produced by endothelial cells, we also used the double-staining immunohistochemistry for the endothelial marker CD34 (brown staining) and EDIL3 (red staining). We observed that EDIL3 is also expressed in some intratumoral endothelial cells (Fig. S1A). Interestingly, the results also indicated that EDIL3 produced by tumoral cells promote adjacent endothelial cell growth (Fig.S1B). The correlation between EDIL3 and Vimentin/E-Cadherin expression was also indicated by immunohistochemistry analysis (Fig.S1C). When the median EDIL3 expression was calculated for all the fifty HCC samples studied by qRT-PCR and used as the cut-off for Fisher's 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
exact test and Kaplan-Meier analysis, it was demonstrated that high EDIL3 expression was significantly associated with a shorter overall survival (Fig. 1G). Consistent with EDIL3 being a secreted glycoprotein, EDIL3 protein expression was also shown to be significantly higher in the plasma of HCC patients compared to normal individuals by ELISA analysis. It is also showed that the EDIL3 level is significantly higher in the HCC patients with early recurrence than the HCC patients without early recurrence, suggesting EDIL3 can be a potential non-invasive diagnostic biomarker for HCC with early recurrence (Fig. 1H).
EDIL3 is a novel regulator of EMT in HCC Previously, we established that epithelial liver cancer cells such as HepG2, Hep3B, HuH7 and PLC/PRF/5 had high E-cadherin (CDH1) and low vimentin (VIM) expression. In comparison, liver cancer cells with a mesenchymal phenotype such as HLE, SK-HEP-1, SNU-449 and Mahlavu had low E-cadherin and high vimentin expression [17]. When the expression of EDIL3 was studied with the same panel of liver cancer cell lines by qRT-PCR and western blotting, we observed that EDIL3 expression was significantly higher in liver cancer cells with a mesenchymal phenotype than in the cells with an epithelial phenotype (Figs. 2A and 2B). Similar observations could be made using independent published microarray data for liver cancer cell lines [15].
EDIL3 expression was significantly correlated with expression of the
mesenchymal marker VIM and inversely correlated with the epithelial marker CDH1 (Figs. S2A and S2B). Next, epithelial HuH7 cells were stably transfected with either pLenti-EDIL3 or pLenticontrol vector. The stable cells were tentatively designated as HuH7-EDIL3 or HuH7-control, respectively. The expression of EDIL3 in these cells was confirmed by qRT-PCR (Fig. S2C). 7
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Compared to pLenti-control-transfected cells, the up-regulation of EDIL3 was associated with dramatic morphological changes observed in the HuH7-EDIL3 cells: from an epithelial cobblestone phenotype to an elongated fibroblastic phenotype, which is indicative of EMT (Fig. 2C). The induction of EMT in the HuH7-EDIL3 cells was also associated with reduced Ecadherin and elevated vimentin expression (Figs. 2E and 2G). Similarly, HLE cells with a mesenchymal phenotype and high EDIL expression were used as the recipient cells for the transfection of EDIL3 shRNA. Following the silencing of EDIL3 (Fig. S2D), striking morphological changes consistent with those of the mesenchymal-to-epithelial transition (MET) were observed (Fig. 2D). The up-regulation of E-cadherin and reduced vimentin expression were also observed (Fig. 2F and 2H).
Effect of EDIL3 expression on HCC cell migration and invasion in vitro EMT has been indicated as a key step in initiating cancer cell migration [17]. The migration potential of the HuH7-EDIL3 and HLE-shEDIL3 cells was studied using the wound healing assay. It was observed that stable overexpression of EDIL3 in the epithelial HuH7 cells significantly promoted cell migration (Fig. 3A) while stable knockdown of EDIL3 in the mesenchymal HLE cells significantly inhibited cell migration (Fig. 3B). Consistent with results of the wound healing assay, overexpression of EDIL3 in the HuH7-EDIL3 cells promoted invasion (Fig. 3C and 3D) and the repression of EDIL3 expression in the HLE-shEDIL3 cells reduced the number of cells invading the extracellular matrix gel (Fig. 3E and 3F). Meanwhile, we have investigated the effects of EDIL3 modulation on proliferation and apoptosis of HCC cells using soft agar colony formation assay and anoikis assay. Anoikis is a form of apoptosis that is induced by anchorage-dependent cells detaching from the surrounding extracellular matrix 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
(ECM)[18]. The results showed that overexpression of EDIL3 significantly promoted the anchorage-independent growth (Fig. S2E and S2F) and decreased the anoikis rate of HCC cells (Fig. S2G and S2H), suggesting that EDIL3 affects cell proliferation and apoptosis. Taken together, these data indicated that EDIL3 played an important role in regulating HCC cell migration and invasion.
EDIL3 enhances recruitment of endothelial cells and promotes capillary tube formation in vitro Angiogenesis is a key early step in the cancer invasion-metastasis cascade. EDIL3, being a glycoprotein secreted by endothelial cells, is likely to be involved in tumour angiogenesis. We have therefore studied the role of EDIL3 in HCC angiogenesis by in vitro endothelial recruitment and capillary tube formation assays using stably transfected HuH7 and HLE cells with different EDIL3 expression. In the presence of HLE cells with high EDIL3 expression, HUVECs migrated more efficiently through the transwell pores compared to those grown in the absence of tumour cells conditioned medium (NC) or in the presence of HLE-shEDIL3 cells with low EDIL3 expression. The knockdown of EDIL3 expression could significantly suppress the effects of HLE cells on HUVEC migration (Fig. 4A and 4C). Furthermore, cell-conditioned medium of EDIL3high HLE cells compared to the absence of tumour cells conditioned medium (NC) promoted the HUVECs to develop more capillary-like structures. Tube formation was reduced dramatically in HUVECs that were grown in conditioned medium from HLE-shEDIL3 cells (Fig. 4B and 4D). In contrast, the overexpression of EDIL3 in HuH7 cells enhanced HUVEC migration (Fig. 4E and 4G) and capillary tube formation (Fig. 4F and 4H). These data suggested that high EDIL3
9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
expression in HCC cells can promote endothelial cells migration and tumour angiogenesis in vitro.
miR-137 is a critical upstream regulator of EDIL3 To identify the potential upstream miRNA regulator of EDIL3, we employed the miRecords that has incorporated ten miRNA target prediction algorithms [19] to predict miRNAs that could be the potential regulators of EDIL3 (Table S3). Using the luciferase reporter assay, we validated that EDIL3 could potentially be regulated by miR-137. We then cloned a sequence containing the predicted 3′-UTR target site of the EDIL3 mRNA and its mutated sequence into the pGL3 luciferase reporter gene to generate pGL3-EDIL3-3’UTR-wt and pGL3-EDIL3-3’UTR-mut vector respectively (Fig. 5A). These vectors were co-transfected into the HLE cells together with either the p-miR-137 vector or the p-miR-control vector. A renilla luciferase vector (pRL-TK) was used to normalize the differences in transfection efficiency. The over-expression of miR-137 was validated by RT-qPCR (Fig. 5B). Luciferase activity in HLE cells co-transfected with either the p-miR-137 or the pGL3-EDIL3-3’UTR-wt vector was significantly decreased compared to the control (Fig. 5C). Over-expression of miR-137 decreased EDIL3 protein expression in HLE cells (Fig. 5D) and significantly inhibited HLE cell invasion and induced endothelial cell capillary tube formation (Fig. 5E). MiR-137 expression was significantly down-regulated in tumor tissues of HCC patients when compared to adjacent histologically normal liver tissues and miR-137 expression was significantly down-regulated in tumor samples of HCC patients with early recurrent disease compared to samples of patients with non-recurrent disease (Fig. 5F). The decreased expression of miR-137 also significantly correlated with increased expression of
10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
EDIL3 in HCC samples (Fig. 5G). Therefore, overexpression of EDIL3 was correlated with downregulation of miR-137 in HCC and miR-137 is a critical upstream regulator of EDIL3.
EDIL3 regulates TGF-beta and ERK signalling through binding to αvβ3 integrin EDIL3 was initially reported to be a new ligand for the αvβ3 integrin receptor regulating embryonic vascular morphogenesis and remodelling [20, 21]. Co-immunoprecipitation experiments demonstrated that EDIL3 could be specifically co-precipitated with the αvβ3 integrin (Fig. 6A), demonstrating that EDIL3 and αvβ3 integrin physically interacted with each other. Next, immunofluorescence analysis demonstrated that EDIl3 and αvβ3 integrin co-localized in HLE cells (Fig. 6B). Recent studies showed that αvβ3 integrin promotes latent TGF-β-activation by human cardiac fibroblast contraction [22]. We then examined whether the over-expression of EDIL3 could promote the increase production of TGF-β1 in liver cancer cells and observed that the production of TGF-β1 was significantly increased in both HuH7 and PLC/PRF/5 cells stably transfected with EDIL3 (Fig. 6C). Interestingly, using data reported in the Cancer Cell Line Encyclopedia (CCLE) dataset (http://www.broadinstitute.org/ccle) [15], we compared the group of liver cancer cells with EDIL3 high expression (array signal intensity ≥ log26) and EDIL3 low expression (array signal intensity < log26) and identified many TGF-β-regulated genes that were also differentially expressed (>10 fold) between EDIL3 high expressing and EDIL3 low expressing liver cancer cells by Ingenuity Pathway Analysis (IPA) (Table S4 and Fig. S3A). Hierarchical clustering analysis further showed that these differentially expressed genes could also discriminate cells with the epithelial or mesenchymal phenotype (Fig. S3B). Among these differentially expressed genes, expression of TGFB1I1 and TGFB2 showed significantly correlation with the expression level of EDIL3 in the panel of liver cancer cells, Pearson r = 0.67, 11
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
r = 0.68, respectively (Figs. 6D and 6E). These data suggested that over-expression of EDIL3 can regulate TGF-β signaling through binding to αvβ3 integrin in liver cancer cells. To further explore the molecular mechanisms and downstream signal pathways of EDIL3 in HCC cells, we also performed microarray and bioinformatics analysis to identify the genes that are differentially expressed between HLE-shEDIL3 and HLE-shControl cells. Among the dysregulated genes, pseudopodium-enriched atypical kinase 1 (PEAK1) was significantly downregulated in the HLE-shEDIL3 cells. Furthermore, knockdown of EDIL3 in HLE and Mahlavu cells significantly decreased the expression of PEAK1, while overexpression of EDIL3 in HuH7 and PLC/PRF/5 cells increased the expression of PEAK1 (Figs. 6F and 6G). Moreover, EDIL3 expression was found to be significantly correlated with PEAK1 expression in HCC patient tumor samples (Fig. 6H) and was further validated by IHC staining with another 20 cases of HCC tissue samples (Fig. 6I). Next, we deciphered the PEAK1-associated regulatory signalling pathways involved with the over-expression of EDIL3 in HCC cells. IPA analysis showed that EDIL3 could interact with PEAK1 through the SRC family kinases (Figs. S4A and S4B). We subsequently demonstrated that the over-expression of EDIL3 not only significantly enhanced the expression of PEAK1 but also induced the phosphorylation of SRC, ERK and Smad2, suggesting the activation of ERK and TGF-β signaling (Fig. 6J). These data further suggest that over-expression of EDIL3 also activates ERK signalling which has been shown to be strongly associated with tumour angiogenesis and metastasis.
Blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated angiogenesis and invasion
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Since both TGF-beta and ERK signaling have been shown activated by overexpression of EDIL3, we next investigate the effects of TGF-beta and ERK signaling inhibitors on the established EDIL3 overexpression HCC cells. When HuH7-p-lenti-EDIL3 stable cells were treated with VX11e (1µg/ml, ChemieTek, Indianapolis, IN, USA), a potent, selective, and orally bioavailable inhibitor of ERK, or LY2109761 (1µg/ml, MedChemexpress, Shanghai, China), a selective TGFβ receptor type I/II dual inhibitor, strong inhibition of p-ERK and p-Smad2 expression in the HuH7-p-lenti-EDIL3 stable cells was detected (Fig. 7A). The MTS results showed that the cell viability was inhibited with the increasing dose of VX-11e but not LY2109761 and EDIL3 overexpression HuH7-p-lenti-EDIL3 cell is more sensitive to VX-11e treatment (Fig. S5A and S5B). . The MEK inhibitor also significantly inhibits proliferation (Fig.S5C and S5D) and induces apoptosis (Fig. S5E) of EDIL3 overexpression cells, while TGF-beta inhibitor is unable to blunt the apoptosis of EDIL3 overexpression cells (Fig. S5F). To determine if blocking the TGF-β pathway would sensitize the EDIL3 overexpression HuH7-p-lenti-EDIL3 cell to VX-11e, we examined the effect of the presence or absence of LY2109761 (2.5µg/ml) on the sensitivity of HuH7-p-lenti-EDIL3 cells to VX-11e. Interestingly, the results showed that the presence of LY2109761 (2.5µg/ml) did not significantly inhibit cell growth but was sufficient to sensitize the EDIL3 overexpression HuH7-p-lenti-EDIL3 cell to VX-11e (Fig. 7B). The colony formation assay showed that VX-11e treatment significantly decreased colony formation ability of HuH7p-lenti-EDIL3 stable cells and LY2109761 can sensitize this effect (Fig. 7C and 7D). Blocking the TGF-beta and ERK signalling was also shown significant inhibition effect on EDIL3 overexpression HuH7-p-lenti-EDIL3 cell invasion ability (Fig. 7E and 7F) and the HuH7-p-lentiEDIL3 cells conditional medium enhanced HUVEC capillary tube formation (Fig. 7G and 7H), suggesting that blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated 13
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
angiogenesis and invasion, which provide a promising therapeutic potential for the treatment of HCC with overexpression of EDIL3.
Overexpression of EDIL3 promotes tumour metastasis and angiogenesis in vivo To correlate our observations in vivo, we implanted HuH7-pLenti-EDIL3 and HuH7-pLenticontrol cells with luciferase orthotopically into nude mice. The results indicated that stable overexpression of EDIL3 in the HuH7 cells significantly promoted tumour growth (Fig. 8A and 8B). In addition, lung tissues harvested from mice at the end of the experiments inoculated with the HuH7-pLenti-EDIL3 cells gave prominent bioluminescent signals, indicating the presence of lung metastasis. On the other hand, no lung bioluminescent signal was detected in mice inoculated with HuH7-pLent-control cells (Fig. 8C). IHC staining further demonstrated the significant overexpression of EDIL3 in tumour tissues induced by HuH7-pLenti-EDIL3 cells compared to HuH7-pLenti-control cells (Fig. 8D).
Tumour tissues harvested from mice
implanted with the HuH7-pLenti-EDIL3 cells also showed a significantly higher CD34 expression compared to tumour tissues implanted with the HuH7-p-lenti-control cells, thus indicating that overexpression of EDIL3 promoted tumour angiogenesis of HCC in vivo (Fig. 8D). Increased expression of PEAK1 and phosphorylation of Smad2, SRC and ERK were also observed in tumour tissues harvested from mice implanted with HuH7-p-lenti-EDIL3 cells (Fig. 8D), further suggesting that overexpression of EDIL3 activated the TGF-beta and ERK signalling. A schematic diagram has been put forward to summarize the possible regulatory circuitry that governs EDIL3 contributing to EMT,metastasis/invasion, early recurrence and angiogenesis in HCC (Fig. 8E).
14
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Discussion
High EDIL3 expression was shown to associate with early HCC recurrence and a mesenchymal cell phenotype. In vitro studies demonstrated that EDIL3 is a novel regulator of EMT, migration, invasion and tumor angiogenesis in HCC cells. In vivo studies further validated that over-expression of EDIL3 in HCC cells promotes tumour growth, metastasis and angiogenesis. MiR-137 is involved in the upstream regulation of EDIL3, while EDIL3 overexpression regulates the TGF-β and ERK signalling pathways downstream through binding to αvβ3 integrin in HCC cells. EDIL3 was cloned and characterized in 1998 [20]. The EDIL3 protein contains three EGF-like repeats homologous to those in Notch and related proteins, including an EGF-like repeat that contains an RGD motif and two discoidin I-like domains. EDIL3 has been shown to be a matrix protein, to promote adhesion of endothelial cells through interaction with the αvβ3 integrin receptor, and to inhibit the formation of vascular-like structures. Previous studies also showed that EDIL3 is an endogenous leukocyte-endothelial adhesion inhibitor and limits the recruitment of inflammatory cells [23] indicating that EDIL3 plays an important role in mediating angiogenesis and may be important in vessel wall remodelling and development [20]. Many human tumours have been reported to be dependent on angiogenesis for growth and development. Human osteosarcoma cells and murine Lewis lung carcinoma cells engineered to express EDIL3 resulted in two- to four-fold increases in tumour volume [24]. Down-regulated of EDIL3 expression has also been shown to inhibit the growth of colon cancer cells [25]. Overexpression of EDIL3 accelerates tumour growth by enhancing vascular formation, thus suggesting that EDIL3 is a potential target for anti-angiogenic agents. 15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Tumour angiogenesis is a complex process that is regulated by many factors including vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and basic fibroblast growth factor (FGF) [26]. Inhibition of angiogenesis has been extensively studied as a potential therapeutic strategy for HCC. The anti-angiogenic multi-kinase inhibitor sorafenib is the first and only FDA approved target therapy for advanced HCC, but the median survival rate remains unsatisfactory [27]. In this study, we demonstrated that over-expression of EDIL3 increases the ability of HCC cells to promote migration and capillary tube formation by endothelial cells. Since HCC is usually regarded as a highly vascular tumour, thus EDIL3 can provide an attractive potential target for the development of novel anti-angiogenic interventions in HCC. Although previous studies indicated that EDIL3 is over-expressed in HCC and high expression level of EDIL3 in HCC is associated with poor prognosis, the molecular roles of EDIL3 in HCC has not been well investigated [28-30]. In this study, we found that EDIL3 overexpression triggers activate the TGF-β and ERK signaling pathway through interactions with αvβ3 integrin. These results provide important evidence for the development of novel treatment strategies targeting EDIL3 or downstream TGF-β and ERK signaling pathways for the management of HCC. TGF-β plays an important pro-tumorigenic role in HCC mainly by promoting angiogenesis and inducing EMT, invasion and metastasis. [31]. Inhibition of TGF-β receptor I kinase has been shown to block HCC growth and angiogenesis [32]. The ERK signalling has been shown to play important roles in the development of HCC and blocking ERK pathway has been demonstrated as a promising treatment for HCC [33]. Here, we shown that blocking the TGF-beta and ERK signalling effectively inhibits EDIL3 mediated angiogenesis
16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
and invasion. Therefore, the combination inhibition of TGF-β and ERK signaling pathways may provide a promising targeted therapy for HCC cells with high EDIL3 expression. Previously, we have identified a panel of early HCC recurrence-associated miRNAs, including miR-137, by comparing the miRNA expression in samples of recurrent and nonrecurrent human HCC tissue samples using microarrays [17]. Here, we further demonstrated that EDIL3 expression was regulated by miR-137. Its expression was shown to be significantly down-regulated in HCC tissues compared to adjacent histologically normal liver samples and in HCC samples of patients with early recurrence compared to samples of non-recurrent disease. Therefore, restoring miR-137 expression in human HCC such as the introduction of synthetic miRNA mimics could offer appealing miRNA-based therapeutic strategies.
Acknowledgements This work was supported by grants from the SingHealth Foundation and National Medical Research Council of Singapore. We thank Dr. Tony Lim (Pathologist, Singapore General Hospital) for helping on providing tissue slides.
References [1]
Maluccio M, Covey A. Recent progress in understanding, diagnosing, and treating
hepatocellular carcinoma. CA: A Cancer Journal for Clinicians 2012;62:394-399. [2]
Sherman M, Llovet JM. Smoking, hepatitis B virus infection, and development of
hepatocellular carcinoma. Journal of the National Cancer Institute 2011;103:1642-1643. [3]
Villanueva A, Llovet JM. Liver Cancer in 2013: Mutational landscape of HCC [mdash]
the end of the beginning. Nature Reviews Clinical Oncology 2014. 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[4]
Nault J-C, Zucman-Rossi J. Genetics of hepatocellular carcinoma: The next generation. J
Hepatol 2014;60:224-226. [5]
Bruix J, Gores GJ, Mazzaferro V. Hepatocellular carcinoma: clinical frontiers and
perspectives. Gut 2014:gutjnl-2013-306627. [6]
Zhu AX, Duda DG, Sahani DV, Jain RK. HCC and angiogenesis: possible targets and
future directions. Nature Reviews Clinical Oncology 2011;8:292-301. [7]
Lamouille S, Xu J, Derynck R. Molecular mechanisms of epithelial–mesenchymal
transition. Nature Reviews Molecular Cell Biology 2014;15:178-196. [8]
Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. The Journal of
clinical investigation 2009;119:1420-1428. [9]
Craene BD, Berx G. Regulatory networks defining EMT during cancer initiation and
progression. Nat Rev Cancer 2013;13:97-110. [10]
Fantozzi A, Gruber DC, Pisarsky L, Heck C, Kunita A, Yilmaz M, et al. VEGF-Mediated
Angiogenesis Links EMT-Induced Cancer Stemness to Tumor Initiation. Cancer Research 2014. [11]
Yang MH, Chen CL, Chau GY, Chiou SH, Su CW, Chou TY, et al. Comprehensive
analysis of the independent effect of twist and snail in promoting metastasis of hepatocellular carcinoma. Hepatology 2009;50:1464-1474. [12]
Li Y, Xu S, Li J, Han K, Pi H, Zheng L, et al. Epithelial–mesenchymal transition markers
expressed in circulating tumor cells in hepatocellular carcinoma patients with different stages of disease. Cell death & disease 2013;4:e831. [13]
Giannelli G. The epithelial‐mesenchymal transition: Fact or fiction in cancer?
Hepatology 2009;50:1344-1346.
18
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[14]
Hu L, Lau SH, Tzang C-H, Wen J-M, Wang W, Xie D, et al. Association of Vimentin
overexpression and hepatocellular carcinoma metastasis. Oncogene 2004;23:298-302. [15]
Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, et al. The
Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 2012;483:603-607. [16]
Wang SM, Ooi LLP, Hui KM. Identification and validation of a novel gene signature
associated with the recurrence of human hepatocellular carcinoma. Clinical Cancer Research 2007;13:6275-6283. [17]
Xia H, Ooi LLPJ, Hui KM. MicroRNA-216a/217-induced epithelial-mesenchymal
transition targets PTEN and SMAD7 to promote drug resistance and recurrence of liver cancer. Hepatology 2013;58:629-641. [18]
Paoli P, Giannoni E, Chiarugi P. Anoikis molecular pathways and its role in cancer
progression. Biochimica et Biophysica Acta (BBA)-Molecular Cell Research 2013;1833:34813498. [19]
Xiao F, Zuo Z, Cai G, Kang S, Gao X, Li T. miRecords: an integrated resource for
microRNA–target interactions. Nucleic Acids Res 2009;37:D105-D110. [20]
Hidai C, Zupancic T, Penta K, Mikhail A, Kawana M, Quertermous EE, et al. Cloning
and characterization of developmental endothelial locus-1: an embryonic endothelial cell protein that binds the αvβ3 integrin receptor. Genes & development 1998;12:21-33. [21]
Rezaee M, Penta K, Quertermous T. Del1 mediates VSMC adhesion, migration, and
proliferation through interaction with integrin αvβ3. American Journal of Physiology-Heart and Circulatory Physiology 2002;282:H1924-H1932.
19
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[22]
Sarrazy V, Koehler A, Chow ML, Zimina E, Li CX, Kato H, et al. Integrins αvβ5 and
αvβ3 promote latent TGF-β1 activation by human cardiac fibroblast contraction. Cardiovascular research 2014;102:407-417. [23]
Choi EY, Chavakis E, Czabanka MA, Langer HF, Fraemohs L, Economopoulou M, et al.
Del-1, an endogenous leukocyte-endothelial adhesion inhibitor, limits inflammatory cell recruitment. Science 2008;322:1101-1104. [24]
Aoka Y, Johnson FL, Penta K, Hirata K-i, Hidai C, Schatzman R, et al. The embryonic
angiogenic factor Del1 accelerates tumor growth by enhancing vascular formation. Microvascular research 2002;64:148-161. [25]
Zou X, Qiao H, Jiang X, Dong X, Jiang H, Sun X. Downregulation of developmentally
regulated endothelial cell locus-1 inhibits the growth of colon cancer. J Biomed Sci 2008;16:33. [26]
Carmeliet P, Jain RK. Molecular mechanisms and clinical applications of angiogenesis.
Nature 2011;473:298-307. [27]
Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc J-F, et al. Sorafenib in
advanced hepatocellular carcinoma. N Engl J Med 2008;359:378-390. [28]
Luo JH, Ren B, Keryanov S, Tseng GC, Rao UN, Monga SP, et al. Transcriptomic and
genomic analysis of human hepatocellular carcinomas and hepatoblastomas. Hepatology 2006;44:1012-1024. [29]
Sun J-C, Liang X-T, Pan K, Wang H, Zhao J-J, Li J-J, et al. High expression level of
EDIL3 in HCC predicts poor prognosis of HCC patients. World journal of gastroenterology: WJG 2010;16:4611.
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
[30]
Feng M-X, Ma M-Z, Fu Y, Li J, Wang T, Xue F, et al. Elevated autocrine EDIL3 protects
hepatocellular carcinoma from anoikis through RGD-mediated integrin activation. Molecular cancer 2014;13:226. [31]
Neuzillet C, de Gramont A, Tijeras-Raballand A, de Mestier L, Cros J, Faivre S, et al.
Perspectives of TGF-β inhibition in pancreatic and hepatocellular carcinomas. Oncotarget 2014;5:78. [32]
Mazzocca A, Fransvea E, Lavezzari G, Antonaci S, Giannelli G. Inhibition of
transforming growth factor β receptor I kinase blocks hepatocellular carcinoma growth through neo‐angiogenesis regulation. Hepatology 2009;50:1140-1151. [33]
Liu L, Cao Y, Chen C, Zhang X, McNabola A, Wilkie D, et al. Sorafenib blocks the
RAF/MEK/ERK pathway, inhibits tumor angiogenesis, and induces tumor cell apoptosis in hepatocellular carcinoma model PLC/PRF/5. Cancer research 2006;66:11851-11858.
Figure legends
Figure 1. EDIL3 is significantly upregulated in HCC and associated with recurrence and patient survival. (A) The venn diagram of significantly different expressed genes among HCC tissues (T) vs matched normal tissues (MN), HCC with recurrence (R) vs non-recurrence (NR) and liver cancer cells with epithelial phenotype vs cells with mesenchymal phenotype. (B) Validation of EDIL3 expression in a cohort of patient samples (T-R: Tumor Recurrence, T-NR: Tumor Non-Recurrence, MN: Matched Normal and NN: histologically normal liver tissues) by RT-qPCR. (C and D) The expression of EDIL3 was significantly correlated with the expression of the mesenchymal marker vimentin (VIM) and epithelial marker E-cadherin (CDH1) in a 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
group of HCC samples. (E) Representative imaging of IHC staining for validation of the expression of EDIL3 in another panel of HCC tumour tissues (100 ×). Up panel: H&E staining for the tumorous and non-tumorous parts. Down panel: IHC staning for the EDIL3 expression. (F) Imaging analysis and quantification of EDIL3 IHC staining. The IHC quantification was evaluated according to the percentage of cells with positive nuclei. (G) The expression of EDIL3 was associated with the disease-free survival of patients with HCC. The median expression value of EDIL3 was chosen as the cut-off point for survival analysis using the Kaplan-Meier method (*P=0.028). (H) The expression of EDIL3 in the plasma of HCC patients (n=40, 16R, 24NR) compared with normal individuals (n=20).
Figure 2. EDIL3 is a novel regulator of EMT in HCC. (A) Expression of EDIL3 was studied in a panel of liver cancer cell lines by qRT-PCR. (B and C) Expression of EDIL3 was studied in a panel of liver cancer cell lines by western blotting and immunofluorescence analysis showed the location of EDIL3 in HLE cells (C). (D and E) The represent images of morphological change of HuH7-EDIL3 cells from an epithelial cobblestone phenotype to an elongated fibroblastic phenotype, indicative of EMT (D), or HLE-shEDIL3 cells from an elongated fibroblastic phenotype to an epithelial cobblestone phenotype, indicative of MET (E). (F) The expression of EDIL3 and the EMT markers E-cadherin and vimentin in HuH7 cells stably transfected with pLenti-hEDIL3. (G) The expression of EDIL3 and the EMT markers E-cadherin and vimentin in HLE cells stably transfected with shEDIL3. (H and I) The expression of the epithelial marker E-cadherin and the mesenchymal marker vimentin was further analysed in HuH7 and HLE transfected cells by confocal microscopy. The red signal represents staining for
22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
vimentin (upper panel) or E-cadherin (lower panel). Nuclear DNA was detected by staining with Hoechst 33342. Scale bar represents 50 µm.
Figure 3. The effects of EDIL3 on HCC cell migration and invasion in vitro. (A and B) Representative images of cell migration ability in stably transfected HuH7 and HLE cells evaluated by wound healing assay. Twenty four hours after wounding, cells with extended membrane protrusions moved into the wounded areas. The distance of migrated cells travelled in each treatment group is shown. (C and E) Representative images of stably transfected HuH7 (C) and HLE (E) cell invasion. (D and F) The invading cells were quantified by plotting them as the average number of cells per field of view from three different experiments as described.
Figure 4. EDIL3 promotes tumour angiogenesis of HCC in vitro. (A and E) Representative images of endothelial cell migration after incubation with conditioned media (CM) from stably transfected HuH7 and HLE cells using the endothelial recruitment assay. (B and F) Representative tube formation by ECs after incubation with conditioned media (CM) from stably transfected HuH7 and HLE cells using the tube formation assay. (C and G) Quantification of the numbers of migrating endothelial cells in different groups. (D and H) Quantification of the numbers of branchs in different groups, showing their tube forming ability.
Figure 5. miR-137 is identified as an upstream regulator of EDIL3 and is downregulated in HCC samples. (A) The target sequences predicted by RNAhybrid 2.2 or TargetScan and mutations generated in the 3’-UTR of the EDIL3 mRNA are shown. (B) The overexpression of miR-137 was examined by qRT-PCR. (C) Effects of co-transfection of P-miR-137 with wild23
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
type (wt) or mutant (mut) pGL3-EDIL3 constructs into HLE cells on luciferase reporter assays. Data were normalized by the ratio of Firefly and Renilla luciferase activities measured at 48 h post-transfection. The bar graph shows the mean±SD in three independent transfection experiments. *P
