momordin-Ic | AMPK signal

Toxicology in Vitro

Jing Wanga, Qian Liub, Haifang Xiaoc, Xingxing Luoa, Xuebo Liub,⁎

Suppressive eﬀects of Momordin Ic on HepG2 cell migration and invasion by regulating MMP-9 and adhesion molecules: Involvement of p38 and JNK pathways

a School of Food and Biological Engineering, Shaanxi University of Science & Technology, Xi’an 710021, China
b College of Food Science and Engineering, Northwest A&F University, Yangling 712100, China
c School of Agricultural and Food Engineering, Shandong University of Technology, Zibo 255049, China

A R T I C L E I N F O

Keywords:

Adhesion molecules Invasion
Metastasis MMP-9
Momordin Ic
A B S T R A C T

Momordin Ic was previously found to induce liver cancer cell apoptosis and autophagy. To further elucidate the anti-cancer activity of Momordin Ic, we analyzed the suppressive eﬀects of Momordin Ic on cell migration and invasion. We also investigated the mechanisms associated with MMP-9, adhesion molecules and signaling transductions. The results demonstrated that Momordin Ic eﬀectively prevented cell attachment, migration and invasion. E-cadherin, mediation of homotypic adhesion was induced while VCAM-1 and ICAM-1 and MMP-9 were inhibited. Momordin Ic inﬂuenced phosphorylations of p38, JNK and Erk with VEGF. p38 eﬀectively regulated expressions of E-cadherin, VCAM-1 and ICAM-1. JNK greatly contributed to E-cadherin alteration. Erk hardly modiﬁed E-cadherin, VCAM-1, ICAM-1 and MMP-9 although Erk phosphorylation decreased by Momordin Ic. These results revealed Momordin Ic prevent cell invasion by inhibiting VCAM-1, ICAM-1, MMP-9 but inducing E-cadherin expression via p38 and JNK pathways. Thus momordin Ic may be a promising candidate with anti-cancer bioactivity.

Introduction

Momordin Ic (oleanolic acid-3-O-β-D-xylopyranose(1–3) β-D- Pyranoid glucose) is a natural triterpenoid saponin isolated from the fruits of edible FructusKochiae (a traditional oriental plant). There are many reports regarding to the bioactivities of Momordin Ic, such as inhibition of ethanol-induced gastric mucosal lesionsin, prevention of blood sugar increase. Momordin Ic has also been shown to inhibit carbon tetrachloride-induced hepatotoxicity via enhancing the hepatic antioxidant defense system in rats (Kim et al., 2005). However, few ﬁndings were associated with anti-tumor merit for Momordin Ic. We previously found Momordin Ic could eﬀectively induce HepG2 liver cancer cell apoptosis and autophagy through MAPK-mediated mi- tochondrial pathways (Wang et al., 2013; Mi et al., 2016). These va- luable results inspired us to further explore the functional mechanism of Momordin Ic in suppressing liver cancer.
Cancer induced deaths are not only associated with abnormal cell proliferation but also closely related to metastasis (Fidler, 2011). Cancer cells initially lose cell-cell homotypic adhesion, increase adhe- sion to extracellular matrix (ECM) and heterogenic cells, then invade, migrate into adjacent tissue through systemic circulation, and ﬁnally colonize from microscopic growths into macroscopic secondary tumors (Gupta and Massagué, 2006). Many growth factors, including EGF, VEGF, and HGF play an important role in tumor development and metastasis. Vascular endothelial growth factor (VEGF) is one of the most important regulators of angiogenesis which is closely associated with the function of the endothelial cells and the pathogenesis of cancer. Metastasis represents the major problem in cancer manage- ment. It accounts for approximately 90% of deaths for patients with cancer (Cao et al., 2013; Ledford et al., 2011). Thus inhibition of cancer metastasis is greatly desirable for the development of eﬀective therapies and chemotherapeutic agents.

In the present study, the anti-adhesion, anti-migration and anti-in-
vasion eﬀects of Momordin Ic on liver cancer were examined in HepG2 cell line. Cancer metastasis is a complex process involving cell adhe- sion, migration and invasion aﬀected by series of factors and micro- environment (Gupta and Massagué, 2006; Ell and Kang, 2013; Lin et al., 2011).Thus, we also used VEGF as a cell migration or invasion promoter and further investigated some important proteins such as E-cadherin (epithelial cadherin) (Liu et al., 2014).VCAM-1 (vascular cell adhesion

Corresponding author.

E-mail address: [email protected] (X. Liu).

https://doi.org/10.1016/j.tiv.2019.01.007

Received 28 November 2018; Received in revised form 9 January 2019; Accepted 11 January 2019
Availableonline14January2019
0887-2333/©2019PublishedbyElsevierLtd.molecule-1), ICAM-1 (intercellular adhesion molecules-1) and MMP-9 (matrix metalloproteinase-9)that regulating cell adhesion and invasion (Ding et al., 2003; Bouillon et al., 1991; Bourboulia and Stetler- Stevenson, 2010). In addition, cell migration and invasion were also widely inﬂuenced by signaling pathways (Huang et al., 2015; Meng et al., 2013). MAP-kinase signaling pathways have been demonstrated to eﬀectively regulate Momordin Ic-induced cell apoptosis and autop- hagy (Wang et al., 2013; Mi et al., 2016). In the current work, we also studied these pathways to elucidate the possible mechanisms of in- hibitory eﬀect of Momordin Ic on liver cancer metastasis.

2. Materials and methods

2.1. Chemicals and materials

Momordin Ic (98% purity, Chengdu purechem-standard co., ltd, China). Modiﬁed RPMI 1640 medium and fetal bovine serum (FBS) were purchased from Thermo Fisher Scientiﬁc. Inhibitors (SB203580, SP600125, U0126), MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenylte- trazolium bromide) were obtained from Sigma–Aldrich. DAPI (4′,6-
diamidino-2-phenylindole) were from Beyotime Institute of Biotechnology (China). Rabbit polyclonal antibodies for Jun N-terminal kinase (JNK) (9252), p38 (9212), Erk1/2 (9102), and phospho-Jun N- terminal kinase (p-JNK) (9251), phospho-p38 (p-p38) (9211) and phospho-Erk1/2 (p-Erk1/2) (9101) were from Cell Signaling Technology, Inc. GAPDH (sc-25,778), α-tubulin (sc-5286), E-cadherin (sc-8426) and ICAM-1 (sc-8439), VCAM-1 (sc-13160) and MMP-9(sc- 13520), as well as goat anti-rabbit IgG-HRP (sc-2004), goat anti-mouse IgG-HRP (sc-2005) were from Santa Cruz. Biotech. Matrigel (BD Bioscience), Bovine serum albumin BSA (Sigma-Aldrich), Human VEGF (PeproTech, Rocky Hill, NJ USA), Membrane transwell (24-well, 8 μm pore size, 6.5-mm diameter, Corning 174, Costar Corporation). All other reagents were of analytical reagent grade.

2.2. Cell culture

HepG2 cell line (human hepatocyte carcinoma cell) was obtained from Collection of Cell Cultures of the Fourth Military Medical University (Shaanxi, China). Cells were cultured in RPMI medium supplemented with10% FBS, benzylpenicillin (100 kU/L)and strepto- mycin (100 mg/L)in a humidiﬁed incubator (5% CO2).

2.3. Cell attachment assay

Matrigel stored at minus 20 °C was dissolved overnight at 4 °C. Cell culture plates (96-well) were pre-coated with 50 μl of Matrigel (50 μg/ ml) overnight at 4 °C and then blocked with 2% BSA for 2 h at room temperature. Previously, we found that exposure to Momordin Ic (1–10 μM) for up to 24 h exerted little eﬀect on cell viability, and the IC50 values of momordin Ic in HepG2 cell line were 22.5 μΜ (4 h) and
12.8 μΜ (24 h), respectively. For this assay, cells were pre-treated with
diﬀerent concentrations of Momordin Ic (1–10 μM) or VEGF (10 ng/ml) for 24 h. Cells were then harvested and re-suspended in RPMI medium at a density of 2.0 × 105 cells/ml. Approximately 100 μl of cells were added to the coated wells and incubated for 2 h at 37 °C, then plates were washed 3 times with PBS to remove unattached cells. Solvent control and no-wash group were set. The number of remaining attached cells was determined via MTT assay (Lu et al., 2009). The rate of cell attachment was calculated as the ratio of attaching cells to that of no- wash group. All independent experiments were run in triplicate.

2.4. Cell migration and invasion assay

Cell migration and invasion were performed with a transwell cul- ture system. For migration assay, about 3 × 105 cells in 0.5% BSA medium (200 μl) were plated in the upper chamber of a non-coated transwell insert. In the lower chamber, medium with 10% FBS (600 μl) was added. For the Matrigel invasion assay, the upper chamber of the transwell inserts were pre-coated with 50 μl of Matrigel (2.5 mg/ml). Those cells attached to the lower surface were calculated by MTT assay. The rate of cell invasion or migration was determined by comparing to control group. Each independent experiment was repeated 3 times.

2.5. Cell spreading assay

Cells treated with diﬀerent concentrations of Momordin Ic for 24 h were harvested and re-suspended. Cells were then plated in 6-well plates pre-coated with Matrigel. Plates were incubated for 2 h at 37 °C and cell morphology images were captured by microscope (Olympus, TH4-200, Japan).

2.6. Wound-healing assay

Wound-healing assay was performed with some modiﬁcations (Balekar et al., 2012). HepG2 cells were growing in 6-well plates until they reached 90% conﬂuence, then a wound was scratched on the cell surface. Diﬀerent concentrations of VEGF or Momordin Ic were added or not. After incubation for 24 h at 37 °C, the width of the wound was measured under the microscope. Wound closure rate (%) = (wound distance in 0 h – the remaining distance) / wound distance in 0 h × 100%. The experiments were performed in triplicate.

2.7. Preparation of total cell extracts and western blot analyses

After treatment, cells were lysed with lysis buﬀer (20 mM Tris, pH 7.5; 150 mM NaCl; 1% Triton X-100; sodium pyrophosphate; β- glycerophosphate; EDTA;Na3VO4; leupeptin, pH 7.5) with 1% (100 μg/ mL) PMSF on ice for 10 min. After centrifugation for 15 min (15,000g) at 4 °C, the supernatant was harvested and quantiﬁed via bicinchoninic acid (BCA) protein assay. Proteins were separated by sodium dode- cylsulfate–polyacrylamide gel electrophoresis (SDS–PAGE), and trans-
ferred onto PVDF membrane. The membranes were incubated with the
primary antibodies (1:500) at 4 °C overnight followed by secondary antibody (12000) incubation. The bands were visualized by the en- hanced chemiluminescence (ECL, Bio-RAD ChemiDoc XRS) method according to the manufacturer’s instructions.

2.8. Statistical analysis

Unless otherwise indicated, all data are expressed as means ± standard deviation (SD). All the protein bands were quantiﬁed by Quantity One 4.6.2 software. Statistical diﬀerences were evaluated using ANOVA, and signiﬁcant diﬀerence between the tested groups was determined using Duncan’s multiple-range test (DPS 9.50). Values of p < .05 were considered statistically signiﬁcant and p < .01 were considered extremely signiﬁcant. *p < .05 and **p < .01 vs. control. #p < .05 and ##p < .01 vs. Momordin Ic treatment.

3. Results

3.1. Momordin Ic inhibited cell attachment

Cell attachment (Cell-matrix adhesion) is important for tumor cell invasion and migration (Yue et al., 2012; Watson et al., 2012; Korb et al., 2004). HepG2 cells attached to the Matrigel-coated plates were calculated (Fig. 1). After treatment with 5 or 10 μM of Momordin Ic, the number of cells attached to the plates reduced (Fig. 1A-b,c), and the rate of cell attachment decreased by 10%–20%, compared with control
group (Fig. 1B). Meanwhile, the attached cells increased with VEGF
stimulation but decreased after Momordin Ic intervention (Fig. 1A-d,e,f and Fig. 1B). Whereas no signiﬁcant diﬀerence was observed comparing VEGF group and VEGF + Momordin Ic group (p > .05).

+ VEGF
d
e
f
cont
Momordin Ic/ 5 μM Momordin Ic/ 10 μM
Fig. 1. HepG2 cell attachment was inhibited by Momordin Ic in a dose-dependent manner. Cells were treated with Momordin Ic (1–10 μM) or VEGF (10 ng/ml) for 24 h, har-
vested and re-suspended. Approximately 100 μl of cells were
added to Matrigel-coated wells and incubated for 2 h at 37 °C, and then plates were washed 3 times with PBS to remove unattached cells. (A) The ﬂuorescent images of attached cells stained by DAPI (×100) (Olympus, TH4-200, Japan). (B) MTT assay was performed to determine the number of at- tached cells. Control and no-wash group were set. The rate of cell attachment was the remaining attached cells to that of no-wash group. The results are presented as mean ± S.D. (n = 3). *p < .05 and**p < .01 vs. control (cont).

3.2. Momordin Ic prevented cell migration

Cell migration plays an important role in gastrulation embryonic morphogenesis, placental development, vascular sprouting, wound healing, etc. (Yamaguchi et al., 2005; Doyle et al., 2013). The eﬀect of Momordin Ic on HepG2 cell migration was evaluated by transwell mi- gration assay in vitro. The ﬁndings showed that 5 μM of Momordin Ic
inhibited about 50% of cell migration (Fig. 2A). VEGF can stimulate
cancer cell migration while Momordin Ic suppressed VEGF-promoted cell migration. There was no signiﬁcant diﬀerence when comparing the VEGF and VEGF + Momordin Ic groups (p > .05) (Fig. 2B). The wound-healing assay was also performed to determine cell migration ability. As presented in Fig. 2C, cells treated with Momordin Ic scarcely migrated into the wound area during 24 h (wound closure rate
15.7 ± 5.2%), whereas control group cells had more potential to
migrate to the wound area (wound closure rate 63.7 ± 2.5%). Wound closure rate of Momordin Ic intervention group (VEGF + 10 μM) de- creased to 49.8 ± 3.4%, compared to 85.9 ± 3.6% of VEGF stimu- lated-group (Fig. 2C). These results further suggested that cell migra- tion was suppressed by Momordin Ic.

3.3. Momordin Ic inhibited cell invasion

Tumor cell invasion is highly signiﬁcant for metastasis since it is the ﬁrst rate limiting step in metastasis (Perlikos et al., 2013; Carey et al., 2013). In terms of invasion (Fig. 3A, C), a signiﬁcant diﬀerence was observed between the control and 10 μM of Momordin Ic group (inva- sion rate reduced by 32%). There was no signiﬁcant diﬀerence com- paring the VEGF and VEGF + Momordin Ic groups (p > .05) (Fig. 3B). Moreover, cell spreading was signiﬁcantly suppressed as shown in Fig. 3D. Control cells were spreading out with large and ﬂat lamelli- podia while Momordin Ic treated cells were round with loose adhesion

Momordin Ic suppressed HepG2 cell migration. (A, B) Cell migration was performed with 24-well transwell culture system. Cells were pre-treated with Momordin Ic, VEGF or not (control group) for 24 h and then harvested. About 3 × 105 cells in 0.5% BSA medium (200 μl) were plated in the upper chamber of a non-coated transwell insert. Medium with 10% FBS (600 μl) was used as a chemo-attractant in the lower chamber. After incubation for 24 h, the number of cells attached to the lower surface was calculated by MTT assay. The rate of cell migration was compared to that of the control group. (C) A wound was scratched on the monolayer cell surface of 6-well plates. After incubation with diﬀerent concentrations of VEGF or Momordin Ic for 24 h at 37 °C, the width of the wound was measured under the microscope. Wound closure rate (%) = (wound distance in 0 h– the remaining distance) × 100%/ wound dis- tance in 0 h. The experiments were performed in triplicate. Values are means ± S.D. for at least three independent experiments performed in tripli- cate. *p < .05 and **p < .01 vs. control. #p < .05 and ##p < .01 vs. VEGF alone treatment.

These results indicated that Momordin Ic prevented HepG2 cell invasion.

3.4. MomordinIc up-regulated E-cadherin and decreased VCAM-1, ICAM-1 and MMP-9

Cadherin is an essential class of transmembrane proteins that form the link of intercellular adhesions (Liu et al., 2014). VCAM-1, ICAM-1 and MMP-9 positively promote tumor metastasis (Ding et al., 2003; Bouillon et al., 1991; Bourboulia and Stetler-Stevenson, 2010). Mo- mordin Ic signiﬁcantly stimulated E-cadherin expression in a dose- and time-dependent manner, whereas protein levels of VCAM-1, ICAM-1 and MMP-9 greatly decreased, as shown in Fig. 4A, B. VEGF which is eﬀective in stimulating cell migration and invasion was also used to further investigate these proteins expression. The results showed that VEGF increased VCAM-1 and ICAM-1 expression. However, Momordin
10uM +VEGF
5uM +VEGF
Ic (10 μM) could signiﬁcantly counteract the eﬀects of VEGF (Fig. 4C).
These results indicated the potential role of E-cadherin, VCAM-1, ICAM- 1 and MMP-9 in regulating the anti-metastatic eﬀect of Momordin Ic.

3.5. Momordin Ic-suppressed cell invasion was regulated by MAPK pathways

MAPK pathways play an important role in regulating cell growth, diﬀerentiation, apoptosis as well as cell adhesion and metastasis (Huang et al., 2015; Meng et al., 2013; Braccini et al., 2012; Plotnikov et al., 2011). We detected that Momordin Ic down-regulated Erk phosphorylation and increased p38, JNK phosphorylation (Fig. 5A). VEGF alone activated Erk and p38 while VEGF co-treatment with Mo-
mordin Ic (10 μM) remarkably reversed this eﬀect. JNK phosphoryla-
tions were also inﬂuenced by VEGF and Momordin Ic co-treatment (Fig. 5B). In terms of MAPK pathways, SB203580 (p38 inhibitor) pre- treatment decreased E-cadherin and increased expressions of VCAM-1 and ICAM-1 without much inﬂuence on MMP-9 level, compared with Momordin Ic-treated group. There was no signiﬁcant diﬀerence be- tween Momordin Ic treatment and Erk inhibitor (U0126) pre-treatment (Fig. 5C). In Fig. 5D (JNK inhibitor, SP600125), the protein levels of ICAM-1, VCAM-1 and MMP-9 were hardly aﬀected except induction of E-cadherin. These data elucidated Momordin Ic-inhibited cell adhesion and invasion was mediated by MAPK pathways from diﬀerent aspects.

4. Discussion

Metastasis occurs through a multistep process triggered and fa- cilitated by transcription factors, growth factors and other enzymes or chemokines (Fidler, 2011; Gupta and Massagué, 2006; Bertin et al., 2010; Yadav et al., 2011). Cell attachment or adhesion is associated with cell growth, diﬀerentiation, migration as well as metabolism. Cell- matrix adhesion promotes cells to invade and degrade extracellular Cancer cell adhesion to endothelial cells and lymphocytes assist cells to invade and metastasize through microcirculation (Yue et al., 2012; Watson et al., 2012; Korb et al., 2004). We found that cell at- tachment rate somehow decreased with Momordin Ic treatment (Fig. 1). In addition to adhesion, subsequent cell invasion into sur- rounding tissues and the vasculature is essential. This requires che- motactic migration of cancer cells. Generally, processes of cell migra- tion play an important role in nervous system development, gastrulation embryonic morphogenesis, placental development, vascu- larsprouting, wound healing as well as immune cell traﬃcking. How- ever, cell movement is deregulated when pathological situations occur. These abnormal regulations in tumor cell migration and invasion is highly signiﬁcant since invasion is the ﬁrst rate limiting step in me- tastasis (Yamaguchi et al., 2005; Doyle et al., 2013; Perlikos et al., 2013; Carey et al., 2013). We observed that cell migration and invasion were suppressed by Momordin Ic (Figs. 2, 3). Wounding-healing test and cell spreading assay provided further evidence. Therefore, Mo- mordinIc has a potential property to suppress cell invasion and mi- gration.

To accomplish tumor metastasis, malignant epithelial cells have to
integrate environment cues with intracellular machinery to dissociate from the primary tumor and invade neighboring sites resulting in dis- tant lesion. This capacity arises when cells lose the intercellular adhe- sion among cancer cells and increase the potential to invade. This is a process highly associated with loss of E-cadherin (epithelial cadherin) expression. In fact, functional loss of E-cadherin contributes to onco- genic potential and is frequently demonstrated during tumor progres- sion (Carneiro et al., 2012). Thus E-cadherin is considered as a broad- acting tumor suppressor essentially for management of tumor pro- gression and invasion in epithelial cancer, namely hepatocellular car- cinomas. Matrix metalloproteinases (MMPs) play an important role in tissue destruction to facilitate tumor invasion and metastasis. Extensive evidence indicated that MMP levels were abnormally up-regulated in Momordin Ic /10 μM + VEGF various cancer tissues, but the members of the enzyme varied in dif- ferent cell types. Comparing liver cancer cells and normal liver cells, the members of MMPs remain the same but the levels of them diﬀer dra- matically. Particularly, MMP-9 and MMP-2 (two major members of the MMP family) were aberrantly over-expressed in liver cancer cells (Bourboulia and Stetler-Stevenson, 2010; Lee et al., 2013; van Deventer et al., 2008). However, MMP-2 is usually constitutively expressed while MMP-9 is sensitively inductive by a series of cytokines and other sti- mulatory signals (Yan and Boyd, 2007).Thus, regulation of MMP-9 has been considered as a feasible approach for developing anti-metastatic drugs (Zhang et al., 2006; Ko et al., 2012). Momordin Ic inhibited he- patocellular carcinoma cancer cells metastasis at least partly by down- regulation of MMP-9 expression (Fig. 4). After cells crossed the local extracellular matrix in the tumor, they penetrated the vascular en- dothelium, survived and transit across the vessel wall into the tissue. Consistent traﬃcking through the tissues and vessel walls is necessary in the course of cell migration. These processes are mainly dependent on a number of cell surface molecules mediating cell attachment. ICAM- 1, expressed in various cell surface (such as lymphocytes, endothelial.

Momordin Ic prevented HepG2 cell invasion. (A, B) Cell invasion was performed with 24-well transwell culture system. Cells were pre-treated with Momordin Ic, VEGF or not (control group) for 24 h and then harvested. About
3× 105 cells in 0.5% BSA medium (200 μl) were plated in the upper chamber coated with Matrigel. Medium with 10% FBS (600 μl) was added to the lower
chamber. Cells that did not invade were removed after 24 h of incubation. The number of cells invaded was determined by MTT assay. Cell invasion rate (%) = invaded cells × 100%/control cells. (C) The invaded cells stained with MTT and captured by a microscope (×100). (D) Cell spreading was inhibited by Momordin Ic. Cells treated with or without MomordinIc for 24 h and re- suspended. Cells were then plated in 6-well plates pre-coated with Matrigel for another 2 h of incubation at 37 °C and cell morphology image was captured by microscope (×200). Arrow marked non-spreading cells (D-c). Values are means ± S.D. for at least three independent experiments performed in tripli- cate. *p < .05 and **p < .01 vs. control. cells, cancer stem cells, etc.), mediates cell-ECM or cell-cell adhesion (Bouillon et al., 1991; Liu et al., 2013). VCAM-1, also named inducible cell adhesion molecule, is highly sensitive to cytokines (e.g. IL-1, TNF- α) induction. VCAM-1 is rarely detected on the surface of normal tissues and cells, whereas it is expressed in activated endothelial cells, smooth muscle cells, macrophage, ﬁbroblasts as well as variety of cancer cells (Ding et al., 2003; dos Santos et al., 1996). On one hand, tumor cells selectively recognize and bind to ICAM-1 and VCAM-1 on the surface of endothelial cells, accelerating cells attachment to basement membrane and extracellular matrix. On the other hand, ICAM-1 and VCAM-1 on the surface of tumor cells speciﬁcally combine with receptor molecules embedded in neutrophil cells and then neutrophil cells bind to en- dothelial cells to adhere and invade. It is also reported that tumor cells could bind to lymphocytes or peripheral blood lymphocytes by ICAM-1

Momordin Ic stimulated E-cadherin expression and inhibits VCAM-1, ICAM-1 and MMP-9 expression in a time- and concentration-dependent manner. (A) Cells were treated with diﬀerent concentrations of Momordin Ic for 24 h. (B) Cells were treated with 10 μM of Momordin Ic for 4 h, 12 h and 24 h, respectively. (C) Cells were co-treated with or without diﬀerent concentrations of Momordin Ic and VEGF (10 ng/ml) for 24 h. Protein expressions of E-cadherin, VCAM-1, ICAM-1 and MMP-9 were analyzed by western blot. *p < .05 and **p < .01 vs. control. #p < .05 and##p < .01 vs. Momordin Ic treatment.

The protein expressions of E-cad- herin, VCAM-1, ICAM-1 and MMP-9 were mediated by MAPK pathways. (A) Cells were treated with diﬀerent concentrations of Momordin Ic for 24 h. (B) Cells were co- treated with or without Momordin Ic (10 μM) and VEGF (10 ng/ml) for 24 h. JNK, p38, Erk, p-JNK, p-p38 and p-Erk were analyzed by western blot. (C, D) Cells were pretreated with 10 μM of SB203580, 20 μM of U0126(C), 10 μM of SP600125(D) for 30 min, respectively, followed by exposure to Momordin Ic (10 μM) for 24 h. Values are means ± S.D. for at least three independent experiments performed in triplicate.Schematic of the proposed mechanism for the inhibitory eﬀect of Momordin Ic on HepG2 cell adhesion, migration and invasion (see text for details). → indicates activation and ⊥indicates inhibition and VCAM-1 to escape immune killing and migrate with lymph circu- lation (Song et al., 2012; Zecchinon et al., 2006). Therefore, adhesion molecules, especially ICAM-1 and VCAM-1, are closely associated with tumor metastasis and become an attractive research point. We found that Momordin Ic not only greatly decreased the levels of ICAM-1 and VCAM-1 but also inhibited VEGF-stimulated ICAM-1 and VCAM-1 ex- pressions, preventing heterogeneity adhesion and migration (Fig. 4).

Metastasis together with other physiological and pathologic pro- cesses will not proceed without complicated signal transductions. We previously reported MAPK pathways mediated Momordin Ic-induced HepG2 cell apoptosis and autophagy (Wang et al., 2013; Mi et al., 2016). We are also interested in their eﬀects on invasion and metastasis. Sustained kinase activation or phosphorylation not only induce AIG (anchorage-independent growth) and EMT (Epithelial-mesenchymal transitions) but also stimulate matrix degradation and angiogenesis. MAPK pathways mediate tumor cell adhesion by adhesion molecules, and aﬀect angiogenesis by tumor angiogenesis factors (Haga et al., 2008; Infusino and Jacobson, 2012). Momordin Ic activated p38, JNK and inactivated Erk1/2. VEGF-altered MAPK expression was also in- tervened by Momordin Ic treatment. Interestingly, VEGF alone induced p38 activation and this needs further elucidation considering the complex regulation and signal transduction mechanisms. We concluded that p38-MAPK eﬀectively regulated the expressions of E-cadherin, VCAM-1 and ICAM-1. JNK greatly contributed to E-cadherin alteration. Erk1/2 pathway hardly modiﬁed E-cadherin, VCAM-1, ICAM-1 and MMP-9 expression although Erk1/2 phosphorylation decreased by Momordin Ic (Fig. 5). It is possible that Erk1/2 might mediate cell in- vasion by regulating other signal molecules. The decrease of MMP-9, which was not due to MAPK pathways, may associated with other pathways. Further study is need to elucidate these possibilities.
In summary, we demonstrated that Momordin Ic eﬀectively pre-
vents cell adhesion, migration and invasion. E-cadherin was highly induced whereas heterogenic adhesion molecules VCAM-1, ICAM-1, as well as MMP-9 were greatly inhibited. Momordin Ic inhibited VCAM-1, ICAM-1, MMP-9 and induced E-cadherin expression by p38 and JNK rather than Erk pathways, preventing cell adhesion, migration and in- vasion, as proposed in Fig. 6. Our ﬁndings provided new evidence for
the anti-cancer mechanism of Momordin Ic.

Acknowledgements

This work was supported by the National Natural Science Foundation Item of China (grant no. 31271810 and 21707086) and Shandong Provincial Natural Science Foundation (no.ZR2014CQ002).

Notes

The authors declare that there is no conﬂict of interest.

References

Balekar, N., Katkam, N.G., Nakpheng, T., Jehtae, K., Srichana, T., 2012. Evaluation of the wound momordin-Ic healing potential of Wedeliatrilobata (L.) leaves. J. Ethnopharmacol. 141, 817–824.
Bertin, S., Mohsen-Kanson, T., Baqué, P., Gavelli, A., Momier, D., Anjuere, F., Carle, G.F.,
Pierreﬁte-Carle, V., 2010. Tumor microenvironment modiﬁcations induced by so- luble VEGF receptor expression in a rat liver metastasis model. Cancer Lett. 298, 264–272.
Bouillon, M., Tessier, P., Boulianne, R., Destrempe, R., Audette, M., 1991. Regulation by
retinoic acid of ICAM-1 expression on human tumor cell lines. Biochim. Biophys. Acta 1097, 95–102.
Bourboulia, D., Stetler-Stevenson, W.G., 2010. Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs): positive and negative regulators in tumor cell adhesion. Semin. Cancer Biol. 20, 161–168.
Braccini, L., Ciraolo, E., Martini, M., Pirali, T., Germena, G., Rolfo, K., Hirsch, E., 2012.
PI3K keeps the balance between metabolism and cancer. Adv. Biol. Regul. 52, 389–405.
Cao, Z., Shang, B., Zhang, G., Miele, L., Sarkar, F.H., Wang, Z., Zhou, Q., 2013. Tumor cell-mediated neovascularization and lymphangiogenesis contrive tumor progression and cancer metastasis. Biochim. Biophys. Acta 1836, 273–286.
Carey, S.P., D’Alfonso, T.M., Shin, S.J., Reinhart-King, C.A., 2013. Mechanobiology of
tumor invasion: engineering meets oncology. Crit. Rev. Oncol. Hematol. 83, 170–183. Carneiro, P., Fernandes, M.S., Figueiredo, J., Caldeira, J., Carvalho, J., Pinheiro, H., Leite, M., Melo, S., Oliveira, P., Simões-Correia, J., Oliveira, M.J., Carneiro, F., Figueiredo,
C., Paredes, J., Oliveira, C., Seruca, R., 2012. E-cadherin dysfunction in gastric cancer—cellular consequences, clinical applications and open questions. FEBS Lett. 586, 2981–2989.
Ding, Y.B., Chen, G.Y., Xia, J.G., Zang, X.W., Yang, H.Y., Yang, L., 2003. Association of VCAM-1 overexpression with oncogenesis, tumor angiogenesis and metastasis of gastric carcinoma. World J. Gastroenterol. 9, 1409–1414.
Doyle, A.D., Petrie, R.J., Kutys, M.L., Yamada, K.M., 2013. Dimensions in cell migration.

Curr. Opin. Cell Biol. 25, 642–649.
Ell, B., Kang, Y., 2013. Transcriptional control of cancer metastasis. Trends Cell Biol. 23, 603–611.
Fidler, I.J., 2011. The biology of cancer metastasis. Semin. Cancer Biol. 21, 71.
Gupta, G.P., Massagué, J., 2006. Cancer metastasis: building a framework. Cell 127, 679–695.
Haga, A., Funasaka, T., Deyashiki, Y., Raz, A., 2008. Autocrine motility factor stimulates the invasiveness of malignant cells as well as up-regulation of matrix metallopro- teinase-3 expression via a MAPK pathway. FEBS Lett. 582, 1877–1882.
Huang, Y.L., Chu, Y.L., Ho, C.T., Chung, J.G., Lai, C.I., Su, Y.C., Kuo, Y.H., Sheen, L.Y.,
2015. Antcin K, an active triterpenoid from the fruiting bodies of basswood-culti- vated Antrodia cinnamomea, inhibits metastasis via suppression of integrin-mediated adhesion, migration, and invasion in human hepatoma cells. J. Agric. Food Chem. 63, 4561–4569.
Infusino, G.A., Jacobson, J.R., 2012. Endothelial FAK as a therapeutic target in disease.
Microvasc. Res. 83, 89–96.
Kim, N.Y., Lee, M.K., Park, M.J., Kim, S.J., Park, H.J., Choi, J.W., Kim, S.H., Cho, S.Y.,
Lee, J.S., 2005. Momordin Ic and oleanolic acid from Kochiae Fructus reduce carbon tetrachloride-induced hepatotoxicity in rats. J. Med. Food 8, 177–183.
Ko, H.S., Lee, H.J., Kim, S.H., Lee, E.O., 2012. Piceatannol suppresses breast cancer cell invasion through the inhibition of MMP-9: involvement of PI3K/AKT and NF-κB pathways. J. Agric. Food Chem. 60, 4083–4089.
Korb, T., Schlüter, K., Enns, A., Spiegel, H.U., Senninger, N., Nicolson, G.L., Haier, J., 2004. Integrity of actin ﬁbers and microtubules inﬂuences metastatic tumor cell adhesion. Exp. Cell Res. 299, 236–247.
Ledford, H., 2011. Cancer theory faces doubts. Nature 472, 273.
Lee, H., Chang, K.W., Yang, H.Y., Lin, P.W., Chen, S.U., Huang, Y.L., 2013. MT1-MMP
regulates MMP-2 expression and angiogenesis-related functions in human umbilical vein endothelial cells. Biochem. Biophys. Res. Commun. 437, 232–238.
Lin, M.C., Wang, F.Y., Kuo, Y.H., Tang, F.Y., 2011. Cancer chemopreventive eﬀects of lycopene: suppression of MMP-7 expression and cell invasion in human colon cancer cells. J. Agric. Food Chem. 59, 11304–11318.
Liu, S., Li, N., Yu, X., Xiao, X., Cheng, K., Hu, J., Wang, J., Zhang, D., Cheng, S., Liu, S.,
2013. Expression of intercellular adhesion molecule 1 by hepatocellular carcinoma stem cells and circulating tumor cells. Gastroenterology 144, 1031–1041.
Liu, G.L., Yang, H.J., Liu, T., Lin, Y.Z., 2014. Expression and signiﬁcance of E-cadherin, N- cadherin, transforming growth factor-β1 and twist in prostate cancer. Asian Pac J Trop Med 7, 76–82.
Lu, Z.J., Ren, Y.Q., Wang, G.P., Song, Q., Li, M., Jiang, S.S., Ning, T., Guan, Y.S., Yang, J.L., Luo, F., 2009. Biological behaviors and proteomics analysis of hybrid cell line EAhy926 and its parent cell line A549. J. Exp. Clin. Cancer Res. 28, 16.
Meng, X., Cai, C., Wu, J., Cai, S., Ye, C., Chen, H., Yang, Z., Zeng, H., Shen, Q., Zou, F., 2013. TRPM7 mediates breast cancer cell migration and invasion through the MAPK
pathway. Cancer Lett. 333, 96–102.
Mi, Y., Xiao, C., Du, Q., Wu, W., Qi, G., Liu, X., 2016. Momordin Ic couples apoptosis with autophagy in human hepatoblastoma cancer cells by reactive oxygen species (ROS)- mediated PI3K/Akt and MAPK signaling pathways. Free Radic. Biol. Med. 90,
230–242.
Perlikos, F., Harrington, K.J., Syrigos, K.N., 2013. Key molecular mechanisms in lung cancer invasion and metastasis: a comprehensive review. Crit. Rev. Oncol. Hematol. 87, 1–11.
Plotnikov, A., Zehorai, E., Procaccia, S., Seger, R., 2011. The MAPK cascades: signaling
components, nuclear roles and mechanisms of nuclear translocation. BBA-Mol. Cell Res. 1813, 1619–1633.
dos Santos, W.L., Rahman, J., Klein, N., Male, D.K., 1996. Control of lymphocyte adhesion to brain and aortic endothelium: ICAM-1, VCAM-1 and negative charge. J. Neuroimmunol. 66, 125–134.
Song, K., Zhu, F., Zhang, H.Z., Shang, Z.J., 2012. Tumor necrosis factor-α enhanced fu-
sions between oral squamous cell carcinoma cells and endothelial cells via VCAM-1/ VLA-4 pathway. Exp. Cell Res. 318, 1707–1715.
van Deventer, H.W., Wu, Q.P., Bergstralh, D.T., Davis, B.K., O’Connor, B.P., Ting, J.P.,
Serody, J.S., 2008. C-C chemokine receptor 5 on pulmonary ﬁbrocytes facilitates migration and promotes metastasis via matrix metalloproteinase 9. Am. J. Pathol. 173, 253–264.
Wang, J., Yuan, L., Xiao, H., Xiao, C., Wang, Y., Liu, X., 2013. Momordin Ic induces
HepG2 cell apoptosis through MAPK and PI3K/Akt-mediated mitochondrial path- ways. Apoptosis 18, 751–765.
Watson, K., Koumangoye, R., Thompson, P., Sakwe, A.M., Patel, T., Pratap, S., Ochieng, J., 2012. Fetuin-A triggers the secretion of a novel set of exosomes in detached tumor cells that mediate their adhesion and spreading. FEBS Lett. 586, 3458–3463.
Yadav, A., Kumar, B., Datta, J., Teknos, T.N., Kumar, P., 2011. IL-6 promotes head and
neck tumor metastasis by inducing epithelial-mesenchymal transition via the JAK- STAT3-SNAIL signaling pathway. Mol. Cancer Res. 9, 1658–1667.
Yamaguchi, H., Wyckoﬀ, J., Condeelis, J., 2005. Cell migration in tumors. Curr. Opin. Cell Biol. 17, 559–564.
Yan, C., Boyd, D.D., 2007. Regulation of matrix metalloproteinase gene expression. J.
Cell. Physiol. 211, 19–26.
Yue, L., Cui, H., Li, C., Lin, Y., Sun, Y., Niu, Y., Wen, X., Liu, J., 2012. A polysaccharide from Agaricusblazei attenuates tumor cell adhesion via inhibiting E-selectin expres- sion. Carbohydr. Polym. 88, 1326–1333.
Zecchinon, L., Fett, T., Vanden Bergh, P., Desmecht, D., 2006. Bind another day: the LFA-
1/ICAM-1 interaction as therapeutic target. Clin. Appl. Immunol. Rev. 6, 173–189.
Zhang, S., Li, Z., Wu, X., Huang, Q., Shen, H.M., Ong, C.N., 2006. Methyl-3-indolylacetate inhibits cancer cell invasion by targeting the MEK1/2-ERK1/2 signaling pathway. Mol. Cancer Ther. 5, 3285–3293.