Recombinant viral capsid protein VP1 suppresses migration and invasion of human cervical cancer by modulating phosphorylated prohibitin in lipid rafts

Cancer Letters 320 (2012) 205–214

Contents lists available at SciVerse ScienceDirect

Cancer Letters journal homepage: www.elsevier.com/locate/canlet

Recombinant viral capsid protein VP1 suppresses migration and invasion of human cervical cancer by modulating phosphorylated prohibitin in lipid rafts Ching-Feng Chiu a,b, Jei-Ming Peng c, Shao-Wen Hung b, Chi-Ming Liang c, Shu-Mei Liang a,b,c,⇑ a

Institute of Biotechnology, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan, Taiwan Agricultural Biotechnology Research Center, Academia Sinica, Taipei, Taiwan c Genomics Research Center, Academia Sinica, Taipei, Taiwan b

a r t i c l e

i n f o

Article history: Received 24 September 2011 Received in revised form 21 January 2012 Accepted 23 February 2012

Keywords: Prohibitin Raf-1 PIP3 rVP1 Cancer metastasis

a b s t r a c t Recombinant capsid protein VP1 (rVP1) of foot-and-mouth disease virus inhibits invasion/metastasis of cancer cells. Here we studied its mechanism of action on human cervical cancer cells. The inhibition of cell invasion by rVP1 was accompanied with reduction in phosphatidylinositol (3,4,5)-triphosphate (PIP3), phospho-Akt S473, phosphorylated prohibitin (phospho-PHB) T258 in lipid rafts, dissociation of phospho-PHB T258 with Raf-1 and the inactivation of Raf-1/ERK. Addition of PIP3 or overexpression of constitutively active Akt and raft-anchored PHB T258 but not PHB T258I mutant protein reversed the inhibitory effects of rVP1. rVP1 inhibited cervical tumor growth and metastasis, and prolonged survival in xenograft mouse models. These results suggest that rVP1 inhibits cancer metastasis via de-phosphorylation of Akt and PHB T258 in lipid rafts to downregulate Raf/ERK signaling. Ó 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Cervical cancer is one of the most common cancers in women. There are an estimated half million new cases worldwide per year and, in 2008, cervical cancer was responsible for approximately 275,000 deaths [1,2]. More than 85% of cervical cancer related deaths occur in developing countries where it is the second leading cause of cancer related mortality among women [1,2]. Cancer metastasis leads to around 90% of human cancer deaths [3,4]. Even though some progress has been made in the treatment of metastatic cervical cancer that has recurred outside the pelvis, nearly all patients still eventually succumb to their disease [5]. Due to the shortage of effective therapy for advanced metastatic cervical

Abbreviations: Akt, serine/threonine-specific protein kinase; DMEM, Dulbecco’s modified Eagle’s medium; ERK, extracellular signal-regulated kinases; FAK, focal adhesion kinase; FBS, fetal bovine serum; FMDV, foot-and-mouth disease virus; H&E, hematoxylin and eosin; IC50, half maximal inhibitory concentration; IP, immunoprecipitation; MEK, mitogen-activated protein kinase/ERK kinase; MMPs, matrix metalloproteinases; PBS, phosphate-buffered saline; PHB, prohibitin; PI3K, phosphatidylinositol 3-kinases; PIP3, phosphatidylinositol (3,4,5)-trisphosphate; PTEN, phosphatase and tensin homologue; RIPA, radioimmunoprecipitation assay; rVP1, recombinant viral capsid protein VP1; SCID, severe combined immunodeficiency; TBS, tris-buffered saline; WST-1, water soluble tetrazolium salt. ⇑ Corresponding author at: Agricultural Biotechnology Research Center, Academia Sinica, Taipei 11529, Taiwan. Tel.: +886 2 27872082; fax: +886 2 27872083. E-mail address: [email protected] (S.-M. Liang). 0304-3835/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2012.02.027

cancer, continued investigation of new therapeutic modalities and treatment strategies is clearly needed. Foot-and-mouth disease virus (FMDV) is a member of the Aphthovirus genus in the Picornaviridae family. It is composed of 60 copies of each of four capsid proteins termed VP1, VP2, VP3, and VP4 [6,7]. Among these, VP1 contains the Arg-Gly-Asp (RGD) tripeptide motif that binds to integrins to mediate FMDV infection [8]. Recombinant DNA-derived VP1 (rVP1) of FMDV has previously been found to induce apoptosis of human cancer cell lines MCF-7, PC-3, and 22Rv1 via modulation of the integrin/Akt signaling pathway [9]. Recently, rVP1 has also been shown to suppress progression of hepatocellular carcinoma [10] and invasion of SKOV3 ovarian adenocarcinoma cells [11] by decreasing MMP-2. However, how rVP1 suppresses MMP-2 and cancer metastasis remains largely unclear. Activation of Ras, an important proto-oncogene, may upregulate cellular migration and invasion via several important downstream effector signaling pathways, notably Raf-1/MAPK(ERK) and the PI3K/Akt cascade [12,13]. Elevation of PI3K activity has been correlated with clinicopathological significance in cervical cancer [14] and Raf/ERK has been proposed as a target for cancer treatment [15]. Although Ras may bind directly to Raf-1, full activation of Raf-1 requires prohibitin (PHB, also known as PHB1) [16–19] which localizes to the mitochondria, cytosol, nucleus, lipid rafts, and cell plasma membrane [20–23]. PHB contains multiple phosphorylation sites and has been reported to be phosphorylated

206

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

by Akt at T258 and by insulin at both Y114 and Y259 [24,25]. Since rVP1 has been shown to inhibit Akt signaling [9] which might lead to a decrease in phosphorylation of PHB at T258, we hypothesized that de-phosphorylation of PHB T258 might play an important role in the rVP1-mediated suppression of cancer metastasis. In this study we investigated whether rVP1 can suppress the migration/invasion and metastasis of human cervical cancer cell in vitro and in vivo. We evaluated the mechanism of rVP1 action and found that it involves modulation of PI3K, decrease in PIP3 and phospho-PHB T258 in the lipid rafts as well as dissociation of Raf-1 from PHB, resulting in a reduction in Raf-1/ERK signaling and MMP-2 activity, and a decrease in epithelial mesenchymal transition (EMT). By using an orthotopic xenograft mouse model of cervical cancer and analyzing phospho-PHB T258 and Raf-1 in the tumor, we further verified that the rVP1-mediated decrease in phospho-PHB T258 in the lipid raft domain is important for inhibition of tumor growth and metastasis of cervical cancer.

agene) according to the manufacturer’s instructions. The following primers were used to generate the PHB mutants: T258I: forward-50 CTCTCGGAACATCATCTACCTGCCAGCGG 30 and reverse-50 CCGCTGGCAGGTAGATGATGTTCCGAGAG 30 . All constructs were verified by DNA sequencing. The pUSEamp-Akt1 (wild type) and pUSEamp-myr-Akt1 (activated) were purchased from Upstate Biotechnology. Transfections were performed using FuGENE HD Transfection Reagent (Roche, Nutley, NJ) according to the manufacturer’s protocol. 2.4. Cell viability assay

2. Materials and methods

Cell viability was measured by water soluble tetrazolium salt (WST-1) assay according to the manufacturer’s instructions (Roche). In brief, 2  104 cells were added to 100 ll media per well on a 96-well plate and incubated at 37 °C in 5% CO2 overnight in a humidified incubator. The cells attached to the wells were incubated in medium supplemented with 0.5% FBS (0.5% FBS-medium) and treated with serial dilutions of rVP1. After incubation at 37 °C in 5% CO2 for 24 h to allow the drug to take effect, 10 ll WST-1 reagent was added to each well, and the plate was placed on a shaking table. After shaking at 150 rpm for 1 min, the cells were incubated at 37 °C in 5% CO2 for another 2 h to allow the WST-1 reagent to be metabolized, and the proportion of surviving cells were determined by optical density (450 nm test wavelength, 690 nm reference wavelength). The half maximal inhibitory concentration (IC50) is the concentration at which the reagent yields 50% inhibition of the cellular viability.

2.1. Materials

2.5. Clonogenic assay

The following cell lines were obtained from American Type Culture Collection (Rockville, MD): human cervical carcinoma SiHa and CaSki cell lines and human cervical adenocarcinoma HeLa cell lines. HeLa and SiHa cells were maintained in DMEM medium and CaSki cells in RPMI-1640 medium at 37 °C under 5% CO2. All media were supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 100 units/ml penicillin, and 100 lg/ml streptomycin. Rabbit anti-PHB (H-80), rabbit anti-E-cadherin (H-108) and mouse anti-b-actin (C-4) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse antiintegrin b1 (MAB1965) antibody was obtained from Millipore (Bedford, MA). Rabbit anti-FAK and mouse anti-phospho-FAKY397 antibodies were obtained from BD Biosciences (Bedford, MA). Mouse anti-vimentin antibody was obtained from Sigma– Aldrich (St. Louis, MO). Rabbit anti-phospho-PI3Kp85, rabbit anti-phospho-PI3Kp85Y458, rabbit anti-PTEN, rabbit anti-phospho-PTENS385, rabbit anti-Akt and rabbit anti-phospho-AktS473 antibodies were obtained from Cell Signaling Technology (Beverly, MA). Rabbit anti-Raf-1, mouse anti-phospho-Raf-1S338, rabbit anti-ERK1/ 2, rabbit anti-phospho-ERK1/2T185/Y187 and rabbit anti-caveolin-1 antibodies were obtained from Upstate Biotechnology (Charlottesville, VA). Rabbit anti-phosphoPHBT258 antibody was generated by Abnova Corporation, Taiwan. Phosphatidylinositol (3,4,5)-trisphosphate diC16 (PI(3,4,5)P3 diC16) was purchased from Echelon Biosciences Inc. (Salt Lake City, UT).

Human cervical cancer cells were seeded at 2000 cells/well in a 6-well plate. One day after seeding, the cells were treated with 0, 0.2 or 0.4 lM of rVP1 in 0.5% FBS-medium at 37 °C in 5% CO2 for 24 h. After washing, the cells were cultured in 0.5% FBS-medium for 11 days. The cells were subsequently rinsed with PBS, fixed in methanol and stained with 0.5% crystal violet. Colonies of >50 cells were counted. All the experiments were conducted in triplicate. 2.6. Boyden chamber assay Cell migration/invasion ability was determined by using the Boyden chamber assay [10]. Briefly, the upper membrane (8 lm pore size, Corning, Corning, NY) within a Boyden chamber was coated with 20 lg/ml fibronectin (Millipore) for cell migration, or 500 lg/ml Matrigel (BD Biosciences) for cell invasion. 2.7. Gelatin zymographic analysis Gelatinolytic activities of the gelatinases MMP-2 and MMP-9 were evaluated as previously described [11]. 2.8. Immunoprecipitation, pull-down assay, and western blot

2.2. Purification of recombinant VP1 protein Purification of recombinant VP1 protein was carried out as described previously [26]. In brief, the VP1 gene in the expression vector pET24a(+) (Novagen, Madison, WI) was expressed in Escherichia coli. After breaking up the bacteria with a microfluidizer in TEN buffer (50 mM Tris–HCl, pH 8.0, 1 mM EDTA, 0.1 M NaCl), the pellet was washed with 0.5% deoxycholate in TEN buffer, followed by rinsing with TEN buffer and resuspended in binding buffer (20 mM Tris–HCl, pH 8.0, 0.5 M NaCl, 8 M urea). The solution was then applied to a metal-chelating affinity column and eluted with a gradient of 0–0.2 M imidazole. SDS was then added to the protein solution to a final concentration of 1%. The protein solution was subsequently applied to a Superdex 200 column (GE Healthcare, Piscataway, NJ) and eluted with a buffer solution containing 25 mM Tris–HCl, pH 8.0, 1 mM EDTA, 0.1 M NaCl and 0.05% SDS. The fractions containing rVP1 protein were pooled, concentrated and dialyzed against PBS before use. 2.3. Construction of plasmids Human PHB cDNA was amplified by PCR from the pOTB7-PHB plasmid (Open Biosystems, Huntsville, AL) and subcloned into the pDisplay™ vector by Gateway cloning technology (Invitrogen, Carlsbad, CA) to generate plasmid pD-PHB, which fused the C-terminus of the PHB gene to the platelet-derived growth factor receptor (PDGFR) transmembrane domain and tagged it with the hemagglutinin (HA) epitope at the N-terminus. The fusion of expressed PHB with the platelet-derived growth factor receptor (PDGFR) transmembrane domain resulted in anchoring PHB to the plasma membrane and lipid rafts. PHB cDNA was also cloned into the pcDNA6/BioEase-DEST vector (Invitrogen) to generate the plasmid pBio-PHB which expressed biotin-tagged PHB. Human PHB cDNA was subcloned into the pEGFP-N1 vector (Clontech, Mountain View, CA) to generate the plasmid pPHB-GFP, with the GFP tagged to the C-terminus of PHB. The PHB-GFP or GFP gene was subcloned into the pDisplay™ vector to obtain the pD-PHB-GFP and pD-GFP plasmids. The PHB mutants were produced using the QuikChange Site-directed Mutagenesis Kit (Strat-

Cells were washed in ice-cold PBS and lysed in radioimmunoprecipitation assay (RIPA) buffer (Santa Cruz Biotechnology) containing proteinase and phosphatase inhibitor cocktail (Roche). The cell lysates (500 lg) were incubated with anti-PHB antibody and protein A/G agarose (Santa Cruz Biotechnology) for immunoprecipitation or streptavidin agarose (Sigma–Aldrich) for biotin-PHB pull-down on a rotating device at 4 °C overnight. Subsequently, pellets were washed sequentially with icecold RIPA and TBS buffer. Immunoprecipitated or pull-down proteins were separated by 12% SDS–PAGE and electrotransferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore). The membrane was blocked in 5% skim milk for 30 min and incubated with primary antibody (5–10 lg/ml) at 4 °C overnight. After washing with TBST buffer, the membrane was incubated with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology). The presence of antibody-protein complexes was detected by chemiluminescence using SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific, Rockford, IL) and exposure to Hyperfilm ECL film (GE Healthcare). 2.9. In vivo experimental metastasis assay Human cervical cancer cells were pretreated with or without 0.4 lM rVP1 in 0.5% FBS-medium at 37 °C for 24 h. The cells (1  106 cells/100 ll PBS/mouse) were injected into five female SCID mice per group via the tail vein. After 4 weeks, the mice were sacrificed and their lungs were collected, processed for hematoxylin and eosin (H&E) staining and the number of tumor colonies in each lung was counted. 2.10. Establishment of an orthotopic xenograft murine model of human cervical cancer An orthotopic human cervical cancer xenograft model was established in SCID mice. Briefly, 11 female SCID mice per group aged 8–10 weeks were anesthetized by 0.4% isofluorane inhalation. Following exposure of the uterus by an abdominal midline incision, the top and bottom sites of the cervix were stitched using

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214 4–0 bioabsorbable polyglycolic acid sutures. Cervical cancer cells (5  106 cells in 10 ll of PBS) were directly injected into the cervical cavity with a 30G Hamilton syringe. The abdomen was closed in two layers using 3–0 non-absorbable sutures. One week after orthotopic implantation, one group of mice was treated with vehicle and another group was treated with rVP1 (15 mg/kg body weight in 100 ll PBS) via tail-vein injection three times per week for 4 weeks. Three mice in each group were then sacrificed to evaluate the degree of tumor growth (size and weight) and metastasis to the organs such as the liver and lung. The remaining eight mice in each group were used to monitor survival time. Tumor volume (mm3) was calculated using the following formula: tumor volume = 1/2 (length  width2), in which length represents the greatest longitudinal diameter, and width the greatest transverse diameter. 2.11. Animal care Animal care and all in vivo experiments were performed in compliance with the guidelines of the Academia Sinica Institutional Animal Care and Utilization Committee. Female C.B17/Icr-scid mice (SCID mice, 8–10 weeks old) were purchased from the Laboratory Animal Center, College of Medicine, National Taiwan University, Taipei, Taiwan. Mice were provided a standard laboratory diet and distilled water and kept on a 12-h light/dark cycle at 25 ± 2 °C and 55 ± 5% relative humidity. 2.12. Extraction of tissue proteins Tissues were homogenized and extracted by using T-PER Tissue Protein Extraction Reagent (Thermo Scientific) supplemented with proteinase and phosphatase inhibitor cocktail (Roche). After centrifugation at 14,000 rpm for 10 min, the pellets were discarded and the supernatant containing proteins were stored at 80 °C. 2.13. Histopathology examination All organs were fixed in 10% neutral buffered formalin. The tissues were embedded in paraffin, cut into 4-lm-thick sections, and stained with H&E for light microscopy. 2.14. Extraction of membrane raft proteins Membrane raft proteins were extracted as described previously [27]. Briefly, cells (2  106) were washed in ice-cold PBS and lysed by incubation for 30 min in ice-cold lysis buffer (0.5% Triton X-100, 150 mM NaCl, 20 mM Tris–HCl, pH 7.5) containing proteinase and phosphatase inhibitor cocktail (Roche). After centrifugation at 14,000 rpm for 30 min at 4 °C, the supernatants (containing the Triton X-100 soluble fractions) were collected and referred to as cytosolic plus non-raft membrane (C + M) fraction. The insoluble pellets were resuspended in the same lysis buffer supplemented with 0.5% SDS and 2 mM DTT and sonicated for 10 min at 4 °C. After centrifugation at 14,000 rpm for 30 min at 4 °C, the supernatants consisting of membrane raft proteins were collected and analyzed by western blotting. 2.15. Measurement of PIP3 in cells with or without rVP1 treatment CaSki cells (1  108) were treated with or without rVP1 for 24 h in 0.5% FBS medium. The cells were washed twice with ice-cold PBS, and phospholipids were extracted and PIP3 levels were measured by ELISA assay using PIP3 Mass ELISA Kit (Echelon Biotechnology Inc.) according to the manufacturer’s instructions. 2.16. Statistical analysis All statistical comparisons were made with a two-tailed Student’s t test. The survival time was assessed using Kaplan–Meier curves and tested for significance by the log-rank test. Statistical evaluation was performed with GraphPad Prism 5.0 for Microsoft Windows (GraphPad Software, La Jolla, CA). Differences between groups were considered statistically significant at P values of <0.05 (⁄) and <0.01 (⁄⁄).

3. Results 3.1. rVP1 suppresses the migration/invasion of human cervical cancer cells To investigate the anti-cancer potential of rVP1 against cervical cancer cells, two human cervical cancer cell lines, SiHa and CaSki, were treated with rVP1. rVP1 inhibited cell viability of SiHa and CaSki in a concentration-dependent manner (Fig. 1A) with IC50 values of 0.75 and 0.93 lM, respectively. Interestingly, even at a con-

207

centration (0.4 lM) not sufficient to affect cell viability and clonogenic capability (Fig. 1B), rVP1 inhibited migration and invasion of SiHa and CaSki cells significantly (Fig. 1C). Since the reduced expression of epithelial cell marker E-cadherin and a concomitant increase in the expression of the mesenchymal cell marker vimentin are generally correlated with EMT and migration/invasion of cervical cancer [28], the effect of rVP1 on the expression level of E-cadherin and vimentin in cervical cancer cells was determined. Immunoblotting analysis revealed that treatment with 0.4 lM of rVP1 for 24 h resulted in enhanced Ecadherin and reduced vimentin expression levels in both SiHa and CaSki cervical cancer cell lines (Fig. 1D). As an increase in MMP activity is generally associated with enhanced cellular invasion [29], we next evaluated whether rVP1 inhibited MMP activity in cervical cancer cells. Gelatin zymographic analysis showed that the MMP-2 activity in SiHa and CaSki cells was reduced by 0.4 lM rVP1 (Fig. 1E). These results indicate that rVP1 not only enhances E-cadherin but also reduces vimentin expression and inhibits MMP-2 activity, resulting in suppression of the migration and invasion of cervical cancer cells in vitro.

3.2. rVP1 suppresses cervical cancer cell migration/invasion via downregulation of integrin/Akt/PHB pathway rVP1 has recently been shown to suppress invasion of ovarian adenocarcinoma cells by downregulating the Akt signaling pathway through integrin a5b1/FAK/Akt [11]. Here, we examined whether rVP1 suppresses invasion of human cervical cancer cells through a similar pathway. We found that the phosphorylation of Akt S473 was decreased in SiHa or CaSki cells treated with 0.4 lM rVP1 for 24 h (Fig. 2A). The rVP-1-mediated reduction of Akt phosphorylation in CaSki cells could be reversed by anti-integrin b1 antibody, but not by the control, immunoglobulin G (IgG) (Fig. 2B). Next, we searched for Akt downstream molecule(s) involved in rVP1-reduced cell migration/invasion in cervical cancer cells. Recent study has identified PHB as a downstream substrate of active Akt which phosphorylates PHB at T258 residue [24]. As shown in Fig. 2A, the level of phospho-PHB T258 was decreased in SiHa and CaSki cells treated with 0.4 lM rVP1. The inhibitory effect of rVP1 on phospho-PHB T258 was also blocked by pretreating cells with anti-integrin b1 antibody but not control IgG (Fig. 2B). Knockdown of PHB in HeLa cervical cancer cells has led to the suggestion that PHB plays a critical role in epithelial cell migration and adhesion of cells to the extracellular matrix [16,30]. We therefore next examined whether rVP1 might inhibit the activation of Akt and phospho-PHB T258 to affect cell migration/invasive ability using HeLa cells. HeLa cells transfected with the plasmid pAkt-DA which constitutively expressed active Akt exhibited increased levels of phospho-Akt S473 and phospho-PHB T258 as compared with those cells transfected with empty vector or plasmid pAkt-wt which expressed wild-type Akt (Fig. 2C). We also constructed a pBio-PHB plasmid that expressed PHB tagged with biotin at the N-terminus and a T258 phosphorylation site mutant pBio-PHB T258I plasmid that expressed PHB mutant protein in which residue T258 was substituted with isoleucine. We found that HeLa cells cotransfected with the pAkt-DA and pBio-PHB plasmids increased cell invasiveness more than threefold as compared with those cotransfected with the pAkt-wt and pBio-PHB plasmids (Fig. 2D). On the other hand, HeLa cells co-transfected with pAkt-DA and pBioPHB T258I did not significantly increase cell invasiveness (Fig. 2D). Taken together, these results suggest that active Akt phosphorylates PHB T258 and increases cell invasion, whereas, rVP1 reduces phospho-PHB T258 through inhibition of the integrin/Akt pathway to suppress cell migration/invasion.

208

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

OD 450nm

A

2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

SiHa CaSki

0

0.2 0.4 0.6 0.8 1

1.2 1.4 1.6 1.8

2

rVP1 (μM) CaSki ns ns

2000 1500 1000 500 0

0

C Migration/invasion Number of cells/field

Colony formation

Colony formation

SiHa

0.2 0.4 rVP1 (μM)

SiHa 125

Migration Invasion

100

*

75

**

50 25 0

0

0.2

0.4

ns ns

1500 1000

Migration/invasion Number of cells/field

B

500 0

0.2 0.4 rVP1 (μM)

0

CaSki

150

Migration

125

**

75 50 25 0

0

0.2

+

+

rVP1

0.4

rVP1 (μM)

rVP1 (μM)

D

Invasion

**

100

E

+

+

α-E-cadherin

rVP1 MMP-9

α-vimentin MMP-2

α-β-actin SiHa

CaSki

SiHa

CaSki

Fig. 1. rVP1 suppresses the migration/invasion of cervical cancer cells in vitro. (A) rVP1 decreases cervical cancer survival. SiHa and CaSki cells were treated with various concentrations of rVP1 in 0.5% FBS medium for 24 h and assayed for viability. Data represent means ± SD of three independent experiments. (B) Treatment with low concentrations of rVP1 for 24 h did not affect clonogenic capability of cervical cancer cells in vitro. SiHa and CaSki cells were treated with 0, 0.2 or 0.4 lM of rVP1 in medium supplemented with 0.5% FBS for 24 h and then cultured for 11 days to determine their clonogenic capability as described in Section 2. Data represent means ± SD of three independent experiments; ns, not significant. (C) rVP1 inhibits cell migration and invasion of SiHa and CaSki cells in a concentration-dependent manner. The migration and invasion capability of the cells were measured by Boyden chamber assay. Data represent means ± SD of three independent experiments; ⁄P < 0.05; ⁄⁄P < 0.01. (D) rVP1 decreases EMT. SiHa and CaSki cells were treated with 0 or 0.4 lM rVP1 in 0.5% FBS medium for 24 h. The expression level of epithelial cell marker E-cadherin and mesenchymal cell marker vimentin were analyzed by western blot. b-actin was used as a loading control. Blots are representative of three independent experiments. (E) rVP1 decreases MMP-2 activity. SiHa and CaSki cells were treated with 0 or 0.4 lM rVP1 in 0.5% FBS medium for 24 h. The MMP-2 and MMP-9 enzyme activity in the cell-cultured medium was analyzed by a gelatinolytic zymography assay. Data are representative of three independent experiments.

3.3. rVP1 decreases the association of PHB T258 with Raf-1, activation of Raf-1 and cancer cell metastasis It is known that the C-terminal region of PHB (residues 243– 272) interacts with Raf-1 [17,31] and Akt phosphorylates PHB at T258 which is located in the same C-terminal region of PHB

[24,25]. To explore whether the phosphorylation of PHB T258 is involved in the association with Raf-1, HeLa cells were transfected with either the plasmid pBio-PHB or pBio-PHB T258I and the cell lysates were then incubated with streptavidin-agarose to pulldown Biotin-PHB and its associated proteins. Our results showed that significantly more Raf-1 was associated with Biotin-PHB than

209

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

A

IgG

B +

+

rVP1

α-integrin β1

+

+

α-phospho-AktS473

rVP1 α-phospho-AktS473

α-Akt α-Akt α-phospho-PHBT258 α-phospho-PHBT258 α-PHB α-PHB

SiHa

CaSki

vector

pAkt-wt pAkt-DA α-phospho-PHBT258

α-PHB α-phospho-AktS473 α-Akt

Invasion Number of cells/field

D

C

250 200

pBio-PHB w t

**

pBio-PHB T258I

150 100 50 0 ve ctor

pAk t-w t

pAk t-DA

α-β-actin Fig. 2. Treatment of cervical cancer cell lines with rVP1 inhibits PHB T258 phosphorylation and cell invasion. (A) rVP1 decreases phosphorylation of Akt and PHB. SiHa and CaSki cells were treated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium. The expression level of phosphorylated Akt S473, total Akt, phospho-PHB T258 and total PHB were analyzed by western blot. b-actin was used as a loading control. Blots are representative of three independent experiments. (B) Anti-integrin b1 antibodies reversed the inhibitory effects of rVP1 on phosphorylation of Akt and PHB. CaSki cells were pretreated with control IgG or anti-integrin b1 antibodies (2 lg/ml) for 30 min followed by rVP1 treatment for 24 h in 0.5% FBS medium. The expression levels of phospho-AktS473, total Akt, phospho-PHBT258 and total PHB were determined by western blot. b-actin was used as a loading control. Blots are representative of three independent experiments. (C) Overexpression of active Akt increases phosphorylation of PHB. HeLa cells were transfected with empty vector, plasmid encoding wild-type Akt (pAkt-wt) or constitutively active Akt (pAkt-DA) for 48 h. The expression levels of phosphorylated AktS473, total Akt, phospho-PHBT258, total PHB and b-actin were determined by western blot. Blots are representative of two independent experiments. (D) Increase in cell invasion requires phospho-PHBT258. HeLa cells were transfected with pBio-PHB wild-type (wt) or pBio-PHB T258I and co-transfected with control vector, pAkt-wt or pAkt-DA for 48 h respectively. Cell invasion was measured by Boyden chamber assay. Data represent means ± SD of three independent experiments; ⁄⁄P < 0.01.

with Biotin-PHB T258I mutant protein (Fig. 3A). We then tested whether rVP1 which decreased phospho-PHB T258 could also decrease the association of PHB with Raf-1. Immunopreciptitation analysis confirmed that PHB associated with Raf-1 in CaSki cells and that this association was blocked by treating CaSki cells with rVP1 (Fig. 3B). These results suggest that phosphorylation of PHB T258 is essential for association of PHB with Raf1. As association of PHB with Raf-1 is required for Raf-1/ERK activation [16,17], we further investigated the effect of rVP1 on the Raf-1/ERK signaling pathway. rVP1 treatment reduced Raf-1 and ERK phosphorylation in both SiHa and CaSki cells (Fig. 3C), suggesting that rVP1 diminishes the interaction of phospho-PHB T258 with Raf-1 and inactivates the Raf-1/ ERK pathway. To test whether rVP1 inhibits cervical cancer metastasis in vivo, an experimental metastasis murine model was used. CaSki cells were pretreated with 0.4 lM rVP1 for 24 h and then implanted into SCID mice via tail-vein injection. Four weeks after implantation histopathologic examination showed that there was less CaSki cell metastasis to the lung in the rVP1-pretreated group than in the control group (Fig. 3D). Taken together, these results suggest that rVP1 may suppress the metastatic capability of human cervical cancer cells via decreasing phospho-PHB T258 leading to dissociation of PHB with Raf-1 and inactivation of the Raf-1/ERK pathway.

3.4. rVP1 suppresses cervical tumor growth, metastasis and prolongs survival of orthotopic xenograft mice To further investigate the in vivo effect of rVP1, we established an orthotopic xenograft murine model of human cervical cancer. One week after orthotopic implantation of CaSki cells in SCID mice, 15 mg/kg body weight of rVP1 was administered intravenously via the tail vein three times per week for 4 weeks. A significant decrease in the size of the cervix and the number of tumor foci in the cervix was observed in mice treated with rVP1 (Fig. 4A) as compared to those treated with vehicle. Further analysis of tumor lysates obtained from the vehicle- and rVP1-treated mice showed that the levels of phospho-PHB T258 and phospho-Raf-1 in tumors from the rVP1-treated mice were lower than those from tumors in the vehicle-treated mice (Fig. 4B). In addition, immunoprecipitation of PHB with anti-Raf-1 antibodies revealed that there was less interaction of PHB with Raf-1 in tumor lysates from rVP1-treated mice than in those from vehicle-treated mice (Fig. 4C). We also found that treatment with rVP1 significantly suppressed cancer metastasis to the lung and liver in mice (Fig. 4D). Histological analysis of the lung and liver revealed that dissemination of tumor cells was absent in tissue sections from rVP1-treated mice whereas an abundance of cancer cells were found in vehicle-treated mice (Fig. 4D). Comparison of the survival curve of rVP1-treated mice

210

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

Biotin-PHB

A vector

wt

B

T258I

IP: PHB IgG

+

rVP1

Pull-down

α-Raf-1 α-Raf-1 Biotin-PHB

Lysates

α-PHB Biotin-PHB

IgG

Endo-PHB

C

+

+

rVP1 α-phospho-Raf-1S338

α-Raf-1 α-phospho-ERK1/2T185/Y187

α-ERK1/2 α-β-actin

SiHa

CaSki

D

H&E

Lung

CaSki

T

0.4 μM rVP1, 24 h

Tail vein injection

rVP1-pretreated CaSki

No. of tumor foci in lung

CaSki cells

6 5 4 3 2 1 0

**

CaSki

rVP1-pretreated CaSki

Fig. 3. rVP1 inhibits PHB T258 phosphorylation, Raf-1 activation and cancer cell metastasis. (A) Association of Raf-1 with PHB requires phosphorylation of PHB at T258. HeLa cells were transfected with pBio-PHB wild-type (wt) or pBio-PHB T258I for 48 h respectively. Biotin-PHB in total cell lysate was pulled down with streptavidin agarose and the amounts of Raf-1 and biotin-PHB were analyzed by western blot. Blots are representative of three independent experiments. (B) rVP1 inhibits interaction between PHB and Raf-1. CaSki cells were treated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium. PHB was immunoprecipitated with anti-PHB antibody and the level of Raf-1 and PHB were detected by western blotting. IgG was used as an immunoprecipitation control. Blots are representative of three independent experiments. (C) rVP1 decreases phosphorylation of Raf-1 and ERK. SiHa and CaSki cells were treated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium. The expression levels of phosphorylated Raf-1 at Ser338, total Raf-1, phosphorylated ERK1/2 at Thr185/Tyr187, total ERK1/2 and b-actin were determined by western blot. b-actin was used as a loading control. Blots are representative of three independent experiments. (D) rVP1 decreases metastatic capability of cervical cancer cells. CaSki cells were pretreated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium and 1  106 cells were then implanted into SCID mice (n = 5) via tail-vein injection. Four weeks after implantation, mice were sacrificed and lungs were collected, processed for H&E staining and examined at 100 magnification. The number of tumor foci in the lungs of sacrificed mice was counted. Arrows show tumor foci; T, tumor. Data represent means ± SD ⁄⁄P < 0.01.

with that of vehicle-treated mice showed that rVP1 treatment significantly prolonged the survival of tumor-bearing mice (Fig. 4E;

median survival time 61.5 days versus 38 days, P = 0.0002). Collectively, these results suggest that rVP1 effectively inhibits cervical

211

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

rVP1

Cervix

Tumor volume (mm3 )

vehicle

T

vehicle 1

2

**

600 400 200 0

vehicle

rVP1 3

1

2

rVP1

C

3

α-PHB α-phospho-Raf-1S338 α-Raf-1

vehicle

IP: Raf-1

IgG

α-phospho-PHBT258

Tumor lysates

B

800

Tumor lysates

A

e cl hi e v

vehicle 1 2

rVP1 1

2 PHB IgG α-Raf-1

rVP1

E 8 6

**

Survival (%)

Lung

Lung nodules (numbers)

D

4 2 0 vehicle

T

rVP1

vehicle rVP1

100 90 80 70 60 50 40 30 20 10 0

rVP1

0

10

20

30

40

50

60

70

80

T

Liver nodules (numbers)

Liver

Days 5 4 3 2 1 0

**

vehicle

rVP1

Fig. 4. rVP1 suppresses cervical cancer metastasis in an orthotopic xenograft mouse model through decreasing the interaction of PHB with Raf-1. CaSki cells were orthotopically injected into the cervices of mice as described in Section 2. One week after implantation, rVP1 (15 mg/kg body weight) was administered via tail-vein injection three times per week for 4 weeks. (A) rVP1 inhibits the growth of cervical cancer. After rVP1 treatment for 4 weeks, the cervices of three mice were collected and processed for H&E staining and examined at 100 magnification. The tumor volumes of the cervices of the sacrificed mice were measured. T, tumor. Data represent means ± SD (n = 3); ⁄⁄ P < 0.01 (vehicle group versus rVP1-treated group). (B) rVP1 treated cervical cancer expresses a low level of phospho-PHB T258 and phosphorylated Raf-1. Total cervical tumor lysates were obtained from CaSki-xenograft mice that were treated with or without rVP1 and the expression levels of phospho-PHB at T258, total PHB, phosphorylated Raf-1 at Ser338 and total Raf-1 were determined by western blot. Data shown are representative of three tumor samples of each group. (C) rVP1 treatment decreases the association of PHB with Raf-1. Total tumor lysates were immunoprecipitated with anti-Raf-1 antibody and the levels of Raf-1 and PHB were detected by western blot. IgG was used as an immunoprecipitation control. Data shown are representative of two tumor samples from each group. (D) rVP1 treatment decreases the metastasis of cervical cancer to the lung and liver. After rVP1 treatment for 4 weeks, the lungs and livers of mice were collected, processed for H&E staining and examined at 100 magnification. The number of tumor foci in the lungs and livers of sacrificed mice were measured. T, tumor. Data represent means ± SD (n = 3); ⁄⁄P < 0.01 (vehicle group versus rVP1-treated group). (E) rVP1 prolongs the survival of cervical cancer bearing mice. Comparison of survival rates between the vehicle- and rVP1-treated CaSki-xenograft mice. Vehicle treatment t50 = 38 day, rVP1 treatment t50 = 61.5 day; n = 8; P = 0.0002.

tumor growth and suppresses metastasis of cervical cancer cells to the lung and liver through decreasing PHB T258 phosphorylation that reduces PHB/Raf-1 interaction and Raf-1 phosphorylation. 3.5. rVP1 decreases phospho-PHB T258 in rafts via modulation of PIP3 Phosphatidylinositol 3-kinase (PI3K) phosphorylates phosphatidylinositol (4,5)-biphosphate (PIP2) to generate phosphatidylinositol (3,4,5)-triphosphate (PIP3) which binds to the pleckstrinhomology (PH) domains of Akt to increase the attachment of Akt

to lipid rafts and stimulate Akt signaling pathways [32–35]. To further understand the mechanism of rVP1 action, we examined whether PIP3 plays any role in the inhibitory effect of rVP1 on phosphorylated Akt and PHB T258. Our results showed that phosphoPHB T258 existed mainly in the rafts and was modulated by rVP1 treatment (Fig. 5A). rVP1 decreased phosphorylation of FAK, PI3Kp85 Y458, Akt S473 and increased phospho-PTEN S385 (Fig. 5B). It also decreased the cellular level of PIP3 (Fig. 5C). Addition of PIP3 reversed the inhibitory effect of rVP1 on phospho-PHB T258 in the rafts (Fig. 5D). Overexpression of exogenous raft-anchored PHB

212

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

A -

Raft

+

-

PI(3,4,5)-P3 (pmol)/108 cells

C Total

C+M

+

-

+

rVP1 α-PHB α-phospho-PHBT258 α-caveolin-1

B-

20 16

*

12 8 4

**

0 0

0.1

0.2

0.4

rVP1 (μM)

+

rVP1 α-phospho-FAKY397

D Total

α-FAK

-

α-phospho-PI3Kp85Y458

+ -

+

C+M

Raft

+ +

-

+ -

+

+ +

-

+ -

+

+ +

rVP1 PIP3 α-PHB

α-p85 α-phospho-AktS473

α-phosphoPHBT258

α-Akt

α-caveolin-1

α-phospho-PTENS385 α-PTEN

vector

-

+

F

pD-PHB wt pD-PHB T258I

-

+

-

+

rVP1 α-phospho-Raf-1S338 α-Raf-1 exo-phospho-PHBT258 endo-phospho-PHBT258

Invasion Number of cells/field

E

ns

200 150

**

**

non rVP1

100 50 0 vector

pD-PHB w t

pD-PHB T258I

exo-PHB endo-PHB α-phospho-AktS473 α-Akt Fig. 5. rVP1 decreases PHB phosphorylation at T258 through reducing the expression of PIP3. (A) rVP1 decreases phospho-PHB T258 in the lipid rafts. Western blot analysis of the expression level of PHB and phospho-PHB at T258 in the membrane raft or cytosolic plus non-raft membrane (C + M) fraction of CaSki cells after treatment of 0.4 lM rVP1 for 24 h in 0.5% FBS medium. Caveolin-1 served as a membrane raft marker. (B) rVP1 decreases phosphorylation of FAK, PI3K, and Akt and increases phosphorylation of PTEN. CaSki cells were treated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium. The expression levels of phospho-FAKY397, phosphorylated PI3K-p85Y458, phospho-AktS473 and phospho-PTENS385 were determined by western blot. (C) rVP1 decreases the cellular level of PIP3. CaSki cells were treated with 0.4 lM rVP1 for 24 h in 0.5% FBS medium. Cells were then harvested, phospholipids were extracted, and PIP3 levels determined by ELISA. Data are plotted as PIP3 content (pmol/108 cells; means ± SD); ⁄P < 0.05; ⁄⁄P < 0.01. (D) Addition of PIP3 reverses the inhibitory effect of rVP1 on phospho-PHB T258 in the rafts. CaSki cells were treated with 0.4 lM rVP1, 5 lM PIP3 or rVP1 plus PIP3 as indicated for 24 h in 0.5% FBS medium. The expression level of PHB and phospho-PHB at T258 in the membrane raft or cytosolic plus non-raft membrane (C + M) fraction were determined by western blot. Caveolin-1 served as a membrane raft marker. (E) Overexpression of plasmid pD-PHB but not pD-PHB T258I reverses the inhibitory effects of rVP1 on phosphorylation of Raf-1 and PHB but not phosphorylation of Akt. CaSki cells were transfected with empty vector, pD-PHB wild-type (wt) or pD-PHB T258I for 48 h and then treated with 0.4 lM rVP1 for 24 h. The expression levels of phospho-Raf-1S338, phospho-PHBT258 and phospho-AktS473 were determined by western blot. Exo-PHB, exogenous PHB; endo-PHB, endogenous PHB. (F) Overexpressing raft-PHB but not raft-PHB T258I mutant protein reverses the rVP1-mediated inhibition of cancer invasive capability. CaSki cells were transfected with empty vector, pD-PHB wild-type (wt) or pD-PHB T258I for 48 h and then treated with 0.4 lM rVP1 for 24 h. The cell invasiveness was measured by Boyden chamber assay. Data represent means ± SD of three independent experiments; ns, not significant; ⁄P < 0.05; ⁄⁄P < 0.01.

but not raft-anchored PHB T258I mutant protein blocked the inhibitory effect of rVP1 on Raf-1 phosphorylation and cancer cell invasion (Fig. 5E and F). Interestingly, overexpression of raft-anchored PHB did not affect the phosphorylation of Akt S473, suggesting that

phospho-PHB is downstream of Akt. Taken together, these results indicate that rVP1 inhibits integrin/FAK/PI3K to decrease the level of PIP3 which in turn modulates the level of phosphorylated Akt and PHB T258 in the rafts to regulate cell invasive capability.

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

213

4. Discussion

Acknowledgements

Metastatic cervical cancers are not usually amenable to curative treatment even after surgical removal of cervical tumors combined with radiation and chemotherapy [5]. We previously found that rVP1 suppressed invasion and metastasis of SKOV3 ovarian adenocarcinoma cells [11] and progression of hepatocellular carcinoma in vivo [10]. Here, we showed that rVP1 treatment of human cervical cancer cells (SiHa and CaSki) at concentrations that do not cause significant cytotoxic and clonogenic effects (0.4 lM for 24 h) reduced their EMT, MMP-2 activity, and migration/invasive capacity in vitro (Fig. 1B–D). Treatment with rVP1 suppressed metastasis of cervical cancer to the lung and liver not only in a murine xenograft model of metastasis but also in an orthotopic xenograft model in vivo (Figs. 3D and 4D). To our knowledge, this is the first report of the effectiveness of a recombinant viral protein in inhibiting migration/invasion and metastasis of cervical cancer. rVP1 inhibits ovarian cancer cell metastasis in an integrin-dependent manner that correlates with downregulation of FAK, Akt, and MMP-2 as well as activation of GSK-3b, PTEN, caspase-3, and PARP cleavage [11]. Here, we showed that rVP1 also decreased membrane bound PIP3 leading to modulation of phospho-PHB T258 in the lipid rafts of the cervical cancer cells (Fig. 5A). As an increase in PIP3 or phospho-PHB T258 was able to reverse the inhibitory effects of rVP1 on invasion/migration (Fig. 5D and F), our results revealed for the first time that inhibition of both PIP3 and phospho-PHB T258 in the raft are essential for the inhibitory effect of rVP1 on the invasion/migration of cervical cancer cells. Raf/MEK/ERK signaling is a central signaling pathway for regulating MMPs, cell motility and cancer metastasis [36–39]. Although PHB has been suggested to directly interact with Raf-1 and subsequently carry Raf-1 to membrane Ras resulting in the continuous activation of the Raf-1/ERK pathway [16,17,31], our data showed that a decrease in phosphorylated T258 rather than unphosphorylated PHB in the raft is critical for the inhibitory effect of rVP1 on Raf-1/ERK (Figs. 3–5). These results thus further our understanding of the role of PHB in the Raf-1/ERK signaling pathway and cancer invasion/migration/metastasis. It has been proposed that PI3K/Akt and Raf-1/ERK pathways are interdependent and may be involved in crosstalk to regulate the synthesis and expression of MMP-2 [40]. Here, we found that rVP1 inhibited not only PI3K/Akt but also Raf-1/MEK/ERK, resulting in downregulation of MMP-2 and a decrease in cell migration/invasion/metastasis (Figs. 2–5). Our results showed clearly that these rVP1-mediated inhibitory effects could be reversed by the PI3K phosphorylated product PIP3 and phospho-PHB T258 (Fig. 5D and F), suggesting that the inhibition of both PI3K/Akt and PHB/Raf-1/ MEK/ERK is essential for the inhibitory effect of rVP1 on MMP-2 activity and cell migration/invasion/metastasis. Of note, although rVP1-mediated inhibition of phospho-PHB T258 resulted in downregulation of phosphorylated Raf-1 and ERK (Figs. 3C and 5E) which may regulate MMPs [39], whether signaling pathways other than Raf/MEK/ERK are also involved in the effect of rVP1 on MMP-2, EMT and cell migration/invasion/metastasis remains to be further elucidated. Moreover, it would be interesting to search and evaluate whether agents that selectively target phospho-PHB T258 would be more effective than rVP1 in inhibiting migration/invasion and metastasis of cervical cancer. In summary, our study demonstrated that rVP1 suppressed the migration and invasion of cervical cancer cells by modulating integrin/FAK, PIP3, Akt and phospho-PHB T258 in the rafts, dissociation of interaction between PHB and Raf-1, as well as inactivation of Raf-1/ERK signaling in vitro and in vivo. These results suggest that agents that are selective and effective in modulating raft-PHB T258 may have great potential for development as novel therapeutic agents for metastatic cancer.

We thank Ms. Chiao-Li Chu for preparing rVP1 and Miranda Loney (Institute of Molecular Biology, Academia Sinica, Taiwan) for English editorial assistance. Grant support: National Science Council, Taiwan (NSC 99-2313-B-001-004-MY3 to S.-M. L) and Academia Sinica (to S.-M. L. and C.-M. L).

References [1] J. Ferlay, H.R. Shin, F. Bray, D. Forman, C. Mathers, D.M. Parkin, Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008, Int. J. Cancer 127 (2011) 2893–2917. [2] A. Jemal, F. Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancer statistics, CA Cancer J. Clin. 61 (2011) 69–90. [3] J. Sleeman, P.S. Steeg, Cancer metastasis as a therapeutic target, Eur. J. Cancer 46 (2010) 1177–1180. [4] G.P. Gupta, J. Massague, Cancer metastasis: building a framework, Cell 127 (2006) 679–695. [5] P.A. Argenta, R. Ghebre, K.E. Dusenbery, M.D. Chen, P.L. Judson, L.S. Downs Jr., L.F. Carson, Radiation therapy with concomitant and adjuvant cisplatin and paclitaxel in high-risk cervical cancer: long-term follow-up, Eur. J. Gynaecol. Oncol. 27 (2006) 231–235. [6] M.H. van Regenmortel, M.A. Mayo, C.M. Fauquet, J. Maniloff, Virus nomenclature: consensus versus chaos, Arch. Virol. 145 (2000) 2227–2232. [7] G.J. Belsham, Distinctive features of foot-and-mouth disease virus, a member of the picornavirus family; aspects of virus protein synthesis, protein processing and structure, Prog. Biophys. Mol. Biol. 60 (1993) 241–260. [8] J. Ruiz-Saenz, Y. Goez, W. Tabares, A. Lopez-Herrera, Cellular receptors for foot and mouth disease virus, Intervirology 52 (2009) 201–212. [9] J.M. Peng, S.M. Liang, C.M. Liang, VP1 of foot-and-mouth disease virus induces apoptosis via the Akt signaling pathway, J. Biol. Chem. 279 (2004) 52168– 52174. [10] T.A. Chen, J.L. Wang, S.W. Hung, C.L. Chu, Y.C. Cheng, S.M. Liang, Recombinant VP1, an Akt inhibitor, suppresses progression of hepatocellular carcinoma by inducing apoptosis and modulation of CCL2 production, PLoS ONE 6 (2011) e23317. [11] J.M. Peng, Y.H. Chen, S.W. Hung, C.F. Chiu, M.Y. Ho, Y.J. Lee, T.C. Lai, M. Hsiao, C.M. Liang, S.M. Liang, Recombinant viral protein promotes apoptosis and suppresses invasion of ovarian adenocarcinoma cells by targeting alpha5beta1 integrin to down-regulate Akt and MMP-2, Br. J. Pharmacol. 165 (2012) 479– 493. [12] A.E. Karnoub, R.A. Weinberg, Ras oncogenes: split personalities, Nat. Rev. Mol. Cell Biol. 9 (2008) 517–531. [13] N.L. Nock, N.A. Berger, Obesity and cancer: overview of mechanisms, in: N.A. Berger (Ed.), Cancer and Energy Balance, Epidemiology and Overview, Springer Science+Business Media, New York, USA, 2010, pp. 129–179. [14] X.Y. Zhang, H.Y. Zhang, P.N. Zhang, X. Lu, H. Sun, Elevated phosphatidylinositol 3-kinase activation and its clinicopathological significance in cervical cancer, Eur. J. Obstet. Gynecol. Reprod. Biol. 139 (2008) 237–244. [15] G. Maurer, B. Tarkowski, M. Baccarini, Raf kinases in cancer-roles and therapeutic opportunities, Oncogene 30 (2011) 3477–3488. [16] K. Rajalingam, C. Wunder, V. Brinkmann, Y. Churin, M. Hekman, C. Sievers, U.R. Rapp, T. Rudel, Prohibitin is required for Ras-induced Raf-MEK-ERK activation and epithelial cell migration, Nat. Cell Biol. 7 (2005) 837–843. [17] K. Rajalingam, T. Rudel, Ras-Raf signaling needs prohibitin, Cell Cycle 4 (2005) 1503–1505. [18] A.S. Dhillon, S. Meikle, Z. Yazici, M. Eulitz, W. Kolch, Regulation of Raf-1 activation and signalling by dephosphorylation, EMBO J. 21 (2002) 64–71. [19] S. Zimmermann, K. Moelling, Phosphorylation and regulation of Raf by Akt (protein kinase B), Science 286 (1999) 1741–1744. [20] N. Patel, S.K. Chatterjee, V. Vrbanac, I. Chung, C.J. Mu, R.R. Olsen, C. Waghorne, B.R. Zetter, Rescue of paclitaxel sensitivity by repression of Prohibitin1 in drug-resistant cancer cells, Proc. Natl. Acad. Sci. USA 107 (2010) 2503–2508. [21] S. Mishra, L.C. Murphy, L.J. Murphy, The Prohibitins: emerging roles in diverse functions, J. Cell Mol. Med. 10 (2006) 353–363. [22] M.G. Kolonin, P.K. Saha, L. Chan, R. Pasqualini, W. Arap, Reversal of obesity by targeted ablation of adipose tissue, Nat. Med. 10 (2004) 625–632. [23] M. Terashima, K.M. Kim, T. Adachi, P.J. Nielsen, M. Reth, G. Kohler, M.C. Lamers, The IgM antigen receptor of B lymphocytes is associated with prohibitin and a prohibitin-related protein, EMBO J. 13 (1994) 3782–3792. [24] E.K. Han, T. McGonigal, C. Butler, V.L. Giranda, Y. Luo, Characterization of Akt overexpression in MiaPaCa-2 cells: prohibitin is an Akt substrate both in vitro and in cells, Anticancer Res. 28 (2008) 957–963. [25] S.R. Ande, Y. Gu, B.L. Nyomba, S. Mishra, Insulin induced phosphorylation of prohibitin at tyrosine 114 recruits Shp1, Biochim. Biophys. Acta 1793 (2009) 1372–1378. [26] J.H. Wang, C.M. Liang, J.M. Peng, J.J. Shieh, M.H. Jong, Y.L. Lin, M. Sieber, S.M. Liang, Induction of immunity in swine by purified recombinant VP1 of footand-mouth disease virus, Vaccine 21 (2003) 3721–3729.

214

C.-F. Chiu et al. / Cancer Letters 320 (2012) 205–214

[27] Q. Huang, H.M. Shen, G. Shui, M.R. Wenk, C.N. Ong, Emodin inhibits tumor cell adhesion through disruption of the membrane lipid Raft-associated integrin signaling pathway, Cancer Res. 66 (2006) 5807–5815. [28] M.Y. Lee, M.R. Shen, Epithelial-mesenchymal transition in cervical carcinoma, Am. J. Trans. Res. 4 (2012) 1–13. [29] L.A. Liotta, W.G. Stetler-Stevenson, Tumor invasion and metastasis: an imbalance of positive and negative regulation, Cancer Res. 51 (1991) 5054s– 5059s. [30] C. Sievers, G. Billig, K. Gottschalk, T. Rudel, Prohibitins are required for cancer cell proliferation and adhesion, PLoS ONE 5 (2010) e12735. [31] S. Wang, N. Nath, G. Fusaro, S. Chellappan, Rb and prohibitin target distinct regions of E2F1 for repression and respond to different upstream signals, Mol. Cell. Biol. 19 (1999) 7447–7460. [32] L.C. Cantley, The phosphoinositide 3-kinase pathway, Science 296 (2002) 1655–1657. [33] B.D. Manning, L.C. Cantley, AKT/PKB signaling: navigating downstream, Cell 129 (2007) 1261–1274. [34] T.L. Yuan, L.C. Cantley, PI3K pathway alterations in cancer: variations on a theme, Oncogene 27 (2008) 5497–5510.

[35] D. Calay, D. Vind-Kezunovic, A. Frankart, S. Lambert, Y. Poumay, R. Gniadecki, Inhibition of Akt signaling by exclusion from lipid rafts in normal and transformed epidermal keratinocytes, J. Invest. Dermatol. 130 (2010) 1136– 1145. [36] B.B. Friday, A.A. Adjei, Advances in targeting the Ras/Raf/MEK/Erk mitogenactivated protein kinase cascade with MEK inhibitors for cancer therapy, Clin. Cancer Res. 14 (2008) 342–346. [37] D.T. Leicht, V. Balan, A. Kaplun, V. Singh-Gupta, L. Kaplun, M. Dobson, G. Tzivion, Raf kinases: function, regulation and role in human cancer, Biochim. Biophys. Acta 1773 (2007) 1196–1212. [38] P.J. Roberts, C.J. Der, Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer, Oncogene 26 (2007) 3291–3310. [39] C.M. Overall, C. Lopez-Otin, Strategies for MMP inhibition in cancer: innovations for the post-trial era, Nat. Rev. Cancer 2 (2002) 657–672. [40] D. Zhang, M. Bar-Eli, S. Meloche, P. Brodt, Dual regulation of MMP-2 expression by the type 1 insulin-like growth factor receptor: the phosphatidylinositol 3kinase/Akt and Raf/ERK pathways transmit opposing signals, J. Biol. Chem. 279 (2004) 19683–19690.