Epimorphin promotes human hepatocellular carcinoma invasion and metastasis through activation of focal adhesion kinase/extracellular signal-regulated kinase/matrix metalloproteinase-9 axis

Authors


  • Potential conflict of interest: Nothing to report.

  • This work was supported by grants from the Major State Basic Research Program of China (2009CB521704, 2010CB945504) and the National Nature Science Foundation of China (no. 30900857).

Abstract

The high incidence rate of hepatocellular carcinoma (HCC) is mainly the result of frequent metastasis and tumor recurrence. Unfortunately, the underlying molecular mechanisms driving HCC metastasis are still not fully understood. It has been demonstrated that tumor stroma cells contribute to primary tumor growth and metastasis. Within the HCC environment, activated hepatic stellate cells (HSCs) can release a number of molecules and enhance cancer cell proliferation and invasiveness in a paracrine manner. Here, for the first time, we demonstrate that epimorphin (EPM; also called syntaxin-2), an extracellular protein, is strongly elevated in activated HSCs within tumor stroma. We show that knockdown of EPM expression in HSCs substantially abolishes their effects on cancer cell invasion and metastasis. Ectopic expression of EPM in HCC cancer cells enhances their invasiveness; we demonstrate that the cells expressing EPM have markedly increased metastasis potential. Furthermore, EPM-mediated invasion and metastasis of cancer cells is found to require up-regulation of matrix metalloproteinase-9 (MMP-9) through the activation of focal adhesion kinase (FAK)/extracellular signal-regulated kinase (ERK) axis. Conclusion: Our results show that EPM, secreted by activated HSCs within HCC stroma, promotes invasion and metastasis of cancer cells by activating MMP-9 expression through the FAK-ERK pathway. (HEPATOLOGY 2011;)

Primary hepatocellular carcinoma (HCC) is one of the most deadly human cancers; it is the third leading cause of cancer-related deaths and the fifth most frequent neoplasm worldwide.1, 2 Long-term prognosis for HCC is still very poor, with most HCC patients dying from metastasis. Thus, it is urgent to advance our understanding of HCC metastasis mechanisms and improve the efficacy of the current treatment strategies. Recently, many studies have highlighted the importance of cross-talk between tumor cells and their microenvironment and the contribution of stroma cells to tumor progression. It has been demonstrated that various stroma cell types are recruited to neoplasms, where they are activated, and substantially promote the proliferation, invasiveness, and metastatic potential of cancer cells.3-6

Hepatic stellate cells (HSCs) belong to one of the most important stroma cell types in the liver tumor environment. They serve as mediators in the processes of inflammation, fibrosis, carcinoma formation, and tumor metastasis. In the tumor environment, HSCs undergo the transition from the “quiescent” to “activated” state and affect cancer cell proliferation and invasiveness.7 Although some evidence has demonstrated an important contribution of HSCs to HCC growth and progression, molecular interactions between those two cell types still remain to be identified.

It has been reported that activated HSCs can secrete a variety of protein molecules, such as growth factors, signaling molecules, and soluble mediators, which can affect HCC cancer cells in a paracrine manner.8 Epimorphin/syntaxin-2 (EPM) is an extracellular protein that functions as a key epithelial morphoregulator in various organs, such as in the mammary gland, lung, pancreas, intestine, hair follicles, and liver.9 EPM is expressed specifically by HSCs in the liver10 and is involved in liver morphogenesis and regeneration.11-14 Our previous work has shown that EPM regulates rat liver epithelial stem-like cell differentiation to bile duct via activation of the focal adhesion kinase–Ras homolog gene family, member A– extracellular signal-regulated kinase–CCAAT/enhancer binding protein beta (FAK-RhoA-ERK-C/EBPβ)-signaling pathway.15, 16 It has been reported that transgenic mice expressing the extracellular form of EPM in mammary epithelial cells develop alveolar hyperplasias and have a high incidence of mammary neoplasia,17 and that the lack of EPM inhibits chronic inflammation-associated colon carcinogenesis in mice.18 Although it is known to be involved in liver development and regeneration, its function in liver cancer has not been well characterized.

Here we report, for the first time, that HSCs promote HCC invasion and metastasis by enhancing EPM expression. The cancer cells promote the secretion of EPM in the HSCs, which then interact with cancer cells in a paracrine manner. We also found that an intricate multiprocessing cascade is involved in this process: matrix metalloproteinase-9 (MMP-9) and activation of the FAK-ERK pathway are required for EPM-mediated invasion and metastasis of cancer cells.

To summarize, we demonstrate that EPM, a morphogen expressed by HSCs, does not affect cancer cell proliferation, but promotes HCC cell invasion and metastasis via the FAK-ERK-MMP-9 signaling pathway.

Abbreviations

α-SMA, alpha smooth muscle actin; CCL5, chemokine (C-C motif) ligand 5; C/EBPβ, CCAAT/enhancer binding protein beta; CK18, cytokeratin 18; ECM, extracellular matrix; EMT, epithelial-mesenchymal transition; EPM, epimorphin/syntaxin-2; ERK, extracellular signal-regulated kinase; FAK, focal adhesion kinase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; H&E, hematoxylin and eosin; HCC, hepatocellular carcinoma; HGF, hepatocyte growth factor; HSC, hepatic stellate cell; 97H-EPM, MHCC97H-epimorphin; 97L-EPM, MHCC97L-epimorphin; MMPs, matrix metalloproteinases; MTT, (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; OPN, osteopontin; RhoA, Ras homolog gene family, member A; RT-PCR, reverse-transcriptase polymerase chain reaction; SD, standard deviation; SDF-1, stromal cell-derived factor-1; shRNA, short-hairpin RNA; TGF-β, transforming growth factor beta; TIMPs, tissue inhibitors of metalloproteinases.

Materials and Methods

For a description of the materials and methods used in this study, see the Supporting Information.

Results

HSCs Promoted HCC Cell Proliferation, Migration, and Invasion.

To examine the effect of HSCs on HCC cell biology, we incubated HCC cell lines (HepG2, MHCC97H, and MHCC97L) with conditioned medium collected from activated stellate cell culture (LX-2 cells), as described in the previous study.24 The results of proliferation analysis demonstrated that HSC-conditioned medium promoted HCC growth in vitro (Fig. 1A). We used scratch assays and Matrigel assays to investigate the effect of HSCs on carcinoma cell migration and invasion. Results showed that, in comparison with the control medium, HSC-conditioned medium enhanced migration and invasion potential of HepG2, MHCC97H, and MHCC97L cells significantly (Fig. 1C,D). To confirm the effect of HSCs on HCC proliferation, we mixed MHCC97H and LX-2 cells and injected them subcutaneously into nude mice. Growth kinetics of tumors containing HSCs were compared to those formed by HCCs injected alone. Four weeks later, we found that the presence of LX-2 cells accelerated the growth of MHCC97H tumors significantly (Fig. 1B).

Figure 1.

HSCs promoted HCC cell proliferation, migration, and invasion. HepG2, MHCC97H, and MHCC97L cells were cultured in medium preconditioned with LX-2 cells for 72 hours (CM). Results were the average of three independent experiments (A,C,D). (A) MTT assays showed that LX-2 CM significantly promoted cancer cell growth in vitro. (C) Representative images of monolayer scratch assays indicated that LX-2 CM increased the migratory potential of HepG2, MHCC97H, and MHCC97L cells. (D) Matrigel invasion assay was used to analyze the invasion capability of HepG2, MHCC97H, and MHCC97L cells with and without LX-2 CM treatment. Results demonstrated that LX-2 CM increased cancer cell invasion. (B) Tumor volume measurements (mean ± standard deviation [SD]) for MHCC97L cells injected subcutaneously into nude mice with or without LX-2 cells. LX-2 cells accelerate the growth of MHCC97H tumors (n = 6). *P < 0.05.

EPM Was Up-regulated in Carcinoma Cell-Activated HSCs.

Previous studies have suggested that stroma cells function as enhancers or suppressors through a paracrine fashion.25 To understand better how HSCs affect cancer cell proliferation and invasion, we cocultured them with HCC cancer cells and measured their expression levels for various molecules, which have been suggested to be involved in the cross-talk between stroma cells and cancer cells, such as cytokines, chemokines, and growth factors. We found that several genes, including hepatocyte growth factor (HGF), chemokine (C-C motif) ligand 5 (CCL5), cytokeratin 18 (CK18), osteopontin (OPN), EPM, stromal cell-derived factor-1 (SDF-1), and transforming growth factor beta (TGF-β), were up-regulated in HSCs after coculture with HCC cancer cells (Fig. 2A; Supporting Fig. 1).

Figure 2.

EPM was up-regulated in HCC cancer-associated HSCs. (A) RT-PCR analysis of candidate genes in LX-2 cells cocultured with HCC cells (HepG2, MHCC97H, and MHCC97L). EPM, HGF, SDF-1, CK-18, and TGF-β were up-regulated after 72-hour coculture with HCC cells. Gene-expression results were normalized to internal glyceraldehyde 3-phosphate dehydrogenase (GAPDH). (B) Immunofluorescence and H&E for EPM protein in 23 pairs of clinical liver tumors and matched normal adjacent tissues. EPM (green) was expressed at a much higher level in the stroma tissue surrounding tumors than in normal liver parenchyma tissues; strong staining of EPM was seen in the α-SMA (red)-positive cells. Bars: 100 μm. (C) Western blotting analysis of EPM expression in LX-2 cells cultured with or without HCC cells. EPM and α-SMA levels were elevated after 72-hour coculture with HCC cells. (D) Western blotting analysis showed that EPM expression was markedly higher in tumor adjacent tissues (within 1 cm of the discrete tumor margin) than in matched normal tissues. β-actin was used as a loading control.

Using reverse-transcriptase polymerase chain reaction (RT-PCR) and western blotting assay, we confirmed that EPM, an HSC-specific gene product, accumulated to a high level after coculture with HCC cancer cells (Fig. 2A,C). EPM expression was further analyzed by immunofluorescence and hematoxylin and eosin (H&E) in 23 paired sets of samples from clinical liver tumors and normal adjacent tissues (Fig. 2B). As in a previous study, we found that in normal liver tissue, EPM was present mainly along the sinusoidal lining of hepatocytes (position of quiescent HSCs) and around blood vessels.26 EPM expression was much higher in the stroma tissue surrounding tumors than in normal liver parenchyma tissues. Further analysis demonstrated that an EPM-positive signal appeared mainly in alpha smooth muscle actin (α-SMA)-positive cells, suggesting that EPM was predominantly expressed by activated HSCs (Fig. 2B,C). Western blotting results showed that EPM expression was markedly higher in the tissues adjacent to the tumors (within 1 cm of the discrete tumor margin) than that of the matched normal tissues (Fig. 2D). These results show that EPM is up-regulated in activated HSCs in hepatocellular tumor stroma.

HSCs Promoted HCC Cell Proliferation in an EPM-Independent Way In Vitro and In Vivo.

To investigate the biological function of EPM in hepatocellular tumor, we established two stable EPM-overexpressing cell lines, named 97H-EPM (MHCC97H-epimorphin) and 97L-EPM (MHCC97L-epimorphin) (Fig. 3A), and two different EPM short-hairpin RNA (shRNA)-LX-2 cell lines (Fig. 3E). We also detected the secretion of 30 kD of EPM in the supernatant of these EPM-expression cells (Supporting Fig. 2). We performed (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell-proliferation and colony-formation assays and xenograft experiments to examine whether EPM could affect cell proliferation. Results indicated that both overexpression of EPM in HCC cells and knockdown of EPM in HSCs did not affect cancer cell proliferation in vitro or tumor growth in vivo to a significant extent (Fig. 3B-F; Supporting Fig. 3). These results suggest that HSCs promote HCC cell proliferation in an EPM-independent manner.

Figure 3.

HSCs did not affect cell proliferation through EPM expression. (A) MHCC97H and MHCC97L cancer cells were transfected with EPM construct. RT-PCR and western blotting analysis were used to detect EPM expression in parental cells transfected with control vector or in stable EPM-expressing cell lines. For proliferation analysis, MTT assay (B), colony-formation assay (C), and xenografts (D) were performed to explore the effect of stable EPM-expressing on HCC cell proliferation. (E) Upper panel: Inhibition efficiency of the two different EPM shRNAs in LX-2 cells (∼95%) was demonstrated by western blotting analysis with β-actin as the internal control. Lower panel: MTT assay showed that HCC cells stimulated by medium (CM) conditioned with EPM shRNA LX-2 cells had a similar proliferative ability to cells from scrambled shRNA groups. (F) Xenografts were performed to explore the effect of EPM shRNA LX-2 cells on HCC cells. All the data showed that EPM did not significantly affect cancer cell proliferation in vitro or tumor growth in vivo. Results were expressed as mean ± SD of at least three independent experiments (P > 0.05).

EPM Promoted HCC Cell Invasion and Metastasis.

Matrigel invasion assays demonstrated that the invasive capacities were dramatically elevated in the EPM-expressing cancer cells (Fig. 4A; Supporting Fig. 4). To investigate the role of EPM secreted by HSCs on cancer cell invasion, we performed Matrigel invasion assays, and the results suggested that knockdown of EPM expression in HSCs significantly reduced their effect on cancer cell invasion (Fig. 4B). These results demonstrated that EPM produced by activated HSCs worked as an invasion enhancer for HCC cells cultured in vitro.

Figure 4.

EPM increased HCC cell invasion in vitro. (A) Matrigel invasion assay was used to analyze the invasive capacity of MHCC97H and MHCC97L control cells and stable EPM-expressing cells. EPM-expressing cells had dramatically higher invasive capacities than control cells. (B) Knockdown of EPM in HSCs inhibited their effect on cancer cell invasion. Matrigel invasion assay showed that HCC cells stimulated by medium (CM) conditioned with EPM shRNA LX-2 cells possessed much lower invasive ability than cells from scrambled shRNA groups. Results were the average of three independent experiments. Statistical significance was assessed by the Student's t-test. *P < 0.05. Magnification: ×200.

To further explore whether EPM could promote tumor metastasis in vivo, we performed orthotopical liver implantation for intrahepatic metastasis assay with stable EPM-expressing MHCC97L cells, as described in the previous study.27 Parental and EPM-expressing MHCC97L cells, with a similar growth rate, were injected into the left hepatic lobe of nude mice, respectively. Mice were sacrificed after 6-7 weeks (see Fig. 5A for details). We found a dramatic increase in number of metastatic nodules on the liver surface in mice injected with EPM-overexpressing cells, compared with those injected with parental cells (Fig. 5B). Further histological analysis also demonstrated that mice transplanted with EPM-overexpressing cells had much more intrahepatic metastatic nodules (Fig. 5C) and displayed more definite pulmonary metastasis sites than the control group (Fig. 5D; Supporting Fig. 5). Similarly, we injected HCC cells together with EPM shRNA HSCs or control HSCs into mice, and found that the metastatic ability of HCC cells with EPM shRNA HSCs was much weaker than in control groups (Supporting Fig. 6). Taken together, these data support an important role for EPM in hepatocellular tumor invasion and metastasis in vivo.

Figure 5.

EPM promoted HCC cell metastasis in vivo. Orthotopical liver implantation method was used for intrahepatic and pulmonary metastasis analysis. The mice were sacrificed after 6-7 weeks. (A) Tables showed details of experimental mice and status of intrahepatic metastasis. (B) Metastatic nodules on the liver surface were quantified (green arrows: orthotopic tumor; yellow arrows: metastatic nodules). The difference between the groups with and without EPM expression was highly significant. (C) Intrahepatic metastasis of the two groups was demonstrated by H&E staining and immunohistochemistry. In both groups, orthotopic tumors presented malignancy features and green fluorescent protein–positive in the two groups; distal part of the EPM group showed remarkable intrahepatic metastatic nodules with invasive edges, whereas the distal part of the control group mainly presented healthy liver features. (D) Pulmonary metastasis in the two groups was demonstrated by H&E staining. A total of 20 random visual fields were chosen from different lung sections of each group, and pulmonary metastatic foci were quantified as the average number across the 20 visual fields per group. The EPM group had much more definite pulmonary metastasis sites than the control group. Statistical significances were assessed by the Student's t-test. *P < 0.05.

EPM Increased HCC Invasion via Activation of FAK-ERK-MMP-9-Signaling Pathway.

To determine how EPM facilitates HCC cell invasion, the expression of several proteolytic enzymes involved in degrading basement membrane, including four matrix metalloproteinases (MMPs) and three tissue inhibitors of metalloproteinases (TIMPs), was analyzed by RT-PCR in MHCC97H and MHCC97L cells after EPM transfection. Results showed that only MMP-9 was dramatically up-regulated after EPM transfection (Fig. 6A), which was confirmed by real-time PCR, western blotting, and immunohistochemistry results (Fig. 6B). To address whether EPM would promote cell invasion through MMP-9 expression, RNA interference was used to knock down MMP-9 expression in EPM-expressing MHCC97H and MHCC97L cancer cells. We found that knockdown of MMP-9 was sufficient to abolish the increase in cell invasion (Fig. 6C), suggesting that EPM facilitated the process by up-regulating MMP-9.

Figure 6.

EPM promoted HCC cell invasion by up-regulating MMP-9. (A) RT-PCR analysis of MMP and TIMP expression in MHCC97H and MHCC97L cells 36 hours after transfection with EPM or control vectors. MMP-9 was dramatically up-regulated. (B) MMP-9 expression was confirmed in vitro by real-time RT-PCR and western blotting, with GAPDH or β-actin as internal controls. In vivo, MMP-9 expression was up-regulated in 97L-EPM tumors of the orthotopic mouse model, compared with 97L-control tumors. (C) Knockdown of MMP-9 in cells with high EPM expression inhibited their effect on cancer cell invasion. Left panel: RT-PCR and western blotting analysis results showed the inhibition efficiency of the two MMP-9 shRNAs (60%-80%). Right panel: invasion of 97H-EPM and 97L-EPM cells was analyzed by Matrigel assay 36 hours after transfection. Statistical significance was assessed by the Student's t-test. *P < 0.05. Magnification: ×200.

It has been reported that EPM binds to αv-integrin-containing receptors, leading to activation of the FAK-ERK-signaling pathway and induction of epithelial morphogenesis.28 Our previous studies confirmed that FAK and ERK phosphorylation is a key factor in EPM-induced bile duct formation of liver epithelial stem-like cells.16 To elucidate the mechanism by which EPM activates MMP-9 expression, two important kinases, FAK and ERK, known to mediate integrin signaling were analyzed. We found that phosphorylation of FAK and ERK1/2 was up-regulated in EPM-transfected HCC cells (Fig. 7A). ERK inhibitor PD98059 could blunt MMP-9 expression both in 97H-EPM and 97L-EPM cells and reduced their invasive ability (Fig. 7B,C). Furthermore, knockdown of FAK in EPM-expressing cells significantly inhibited MMP-9 expression and reduced the invasive ability of HCC cells (Fig. 7D,E). These results demonstrate that EPM activates the FAK- and ERK-signaling pathway to facilitate MMP-9 expression and stimulate cancer cell invasion.

Figure 7.

EPM induced MMP-9 expression via FAK-ERK activation. (A) Western blotting assay for FAK and ERK1/2 in MHCC97H, MHCC97L, HepG2, and Sk-hep1 cells with or without EPM. Phosphorylation of FAK and ERK1/2 was up-regulated in four cell lines 36 hours after transfection with EPM or control vectors. (B) 97H-EPM and 97L-EPM cells were treated with the ERK inhibitor, PD98059 (50 μM) for 12 or 24 hours, then ERK activation and MMP-9 expression were examined using western blotting. (C) Inhibition of ERK in cells with high EPM expression inhibited their effect on cancer cell invasion. Matrigel assay for invasion of 97H-EPM and 97L-EPM cells 24 hours after treating with or withour PD98059 (50 μM). (D) EPM-expressing cell lines, 97H-EPM and 97L-EPM, were transfected with FAK shRNA or control shRNA. FAK and MMP-9 were analyzed by western blotting 48 hours after transfection. Knockdown of FAK (∼90%) caused a decline in MMP-9 levels. (E) Knockdown of FAK in cells with high EPM expression inhibited their effect on cancer cell invasion. Matrigel assay for invasion of 97H-EPM and 97L-EPM cells 36 hours after transfection with FAK shRNA or scrambled shRNA. Statistical significance was assessed by the Student's t-test. *P < 0.05. Magnification: ×200.

Discussion

Metastasis are responsible for more than 90% of cancer-related mortality.29 These secondary growths arise through a multistep process that begins when cancer cells within primary tumors break away from neighboring cells and invade the basement membrane. This involves the loss of cell-to-cell adhesion, increased cancer cell migration, intravasation into blood and/or lymph vessels, transport through the circulatory system, extravasation, and, finally, seeding at a distant site.

Invasion and metastasis are driven by genetic and epigenetic alterations of cancer cells. Besides this, increasing number of studies suggest that exposure of cancer cells into paracrine signals from mesenchymal cells within tumor stroma also play an important role in tumor metastasis. In the tumor microenvironment, a variety of stromal cells are recruited and localize to tumors, which not only enhance the growth of the primary cancer, but also facilitate its metastatic dissemination to distant organs.30 HCC invasive and metastatic capabilities are tightly linked to its microenvironment. To better understand the role of stroma cells in liver tumorigenesis, especially their roles in tumor metastasis, we set out to explore the effects and molecular mechanisms of HSCs in the progression of liver metastasis.

HSCs account for 5%-8% of total cells in the healthy liver and play important roles in liver development, differentiation, and regeneration.31 Some existing studies report that stromal HSCs undergo phenotypic transformation from the “quiescent” to the “activated” state during liver injury and regeneration. This is accompanied by the up-regulation of cytoskeletal protein expression (e.g., expression of α-SMA32). HSCs produce and secrete a wide variety of proteins, especially after their activation under disease conditions, and are thought to be involved in various liver pathologies, including liver fibrosis and the progression of liver carcinogenesis.

Activated HSCs can accumulate around dysplastic nodules of liver tumors.33 Although it has been suggested that HSCs might enhance the tumorigenicity and invasiveness of cancer cells by producing a multitude of cytokines, chemokines, and growth factors,34 the molecular mechanisms underlying their interaction with cancer cells are still poorly understood.

In the present study, we found that besides several representative growth factors (including HGF, SDF-1, CCL5, OPN, and TGF-β) that have already been reported to be up-regulated in tumor-associated-stroma cells, EPM was also strongly elevated in cancer cell-activated HSCs as well as in HSCs of stroma surrounding HCC in clinical specimens. EPM has been reported to be expressed exclusively in HSCs in the liver10 and plays an essential role in liver morphogenesis and regeneration.11-14 The up-regulation of EPM in activated HSCs suggested that it may function in tumor progression. Our further study demonstrated that EPM enhanced cancer cell motility, invasion capability, and metastasis, whereas it had no effect on cancer cell proliferation.

A critical step in tumor metastasis is the degradation of basement membrane, which is catalyzed by proteolytic enzymes, such as MMPs and TIMPs.35, 36 We found that up-regulation of MMP-9 via FAK and ERK activation was required in the EPM-mediated invasion of cancer cells. Hirai et al. uncovered that EPM binds to αv-integrin-containing receptors, leading to activation of the FAK-ERK signaling pathway and induction of epithelial morphogenesis, for the first time.28 Our previous study suggested a biophysical role for EPM in regulating mitosis orientation during the duct formation of liver epithelial stem-like cells.15 We also confirmed the contribution of FAK and ERK phosphorylation in EPM-induced bile duct formation.16, 37 Accumulating evidence suggests that integrins, activated by extracellular matrix (ECM) proteins, regulate MMP and TIMP expression through intracellular signaling pathways.38 It seems that EPM functions as a metastasis enhancer through binding to integrins and activating FAK and ERK.

In recent years, epithelial-mesenchymal transition (EMT) has been described as a relevant process in tumor progression not only implicated in tumor invasion, but also in other stages during the metastatic process.39 Here, we further explored some EMT-associated markers (including E-cadherin, N-cadherin, Snail, and Twist 1), and found that in EPM-expressing HCC cells, E-cadherin was down-regulated, whereas Twist 1 was dramatically up-regulated (Supporting Fig. 7). These findings, together with up-regulation of MMP-9 expression, demonstrated that EPM, to some extent, could induce EMT in HCC cells and promoted HCC to initiate metastatic invasion.

Two independent reports demonstrated that transgenic mice expressing the extracellular form of EPM in mammary epithelial cells develop alveolar hyperplasias and show a high incidence of mammary neoplasia,17 and that EPM deletion inhibits chronic inflammation-associated colon carcinogenesis in mice.18 In present study, we found that overexpression of EPM had no marked impact on HCC cancer cell proliferation, whereas significant effects on HCC cell invasion and metastasis were observed. We provide here the first evidence suggesting that EPM was involved in HCC progression.

To summarize, our results demonstrate that EPM secreted by HSCs in HCC stroma promotes cancer cell invasion and metastasis by up-regulating MMP-9 expression; knockdown of EPM expression in HSCs abolishes these effects. These findings will support EPM as an important biomarker in activated HSCs for HCC diagnosis; these results may lay a foundation for further uncovering the precise mechanisms of EPM during tumorigenesis and providing insights for the development of novel anticancer therapies. We propose that the interruption of the EPM-FAK-ERK-MMP-9 pathway may be a useful therapeutic approach for controlling HCC metastasis.

Ancillary

Advertisement