Shenyang, China

KAI1 inhibits HGF-induced invasion of pancreatic cancer by sphingosine kinase activity

Xu Liu, Xiao-Zhong Guo, Wei-Wei Zhang, Zhuo-Zhuang Lu, Qun-Wei Zhang, Hai-Feng Duan and Li-Sheng Wang

Shenyang, China

BACKGROUND: KAI1/CD82 has been reported to attenuate the process of metastases in a variety of tumors; however, its mechanism of action in invasion has not been fully elucidated. The present study aimed to investigate the importance of KAI1 in invasion and its correlation with activation of sphingosine kinase (SPK) in human pancreatic cancer PANC1 and Miapaca-2 cell lines.

METHODS: The expression of KAI1 in PANC1 and Miapaca-2 cells, which was mediated by recombinant adenovirus (Ad-KAI1), was assessed by a flow cytometer and Western blotting. After successful infection was established,in vitrogrowth curve and invasive ability in Boyden Chamber assay were studied. The presence of KAI1 correlating with c-Met and SPK was detected by co-immunoprecipitation and [γ-32P] ATP incorporation.

RESULTS: KAI1 genes had no significant effects on the curve representing cell growth. After infection with the KAI1 gene, decreased invasive ability in the Boyden Chamber assay was observed in PANC1 and Miapaca-2 cells that were induced by hepatocyte growth factor. Over-expression of KAI1 in the cells led to the deactivation of SPK and a decreased level of intracellular sphingosine-1-phosphate. No correlation was observed between c-Met and KAI1 during coimmunoprecipitation.

CONCLUSION: The results of this study for the first time demonstrated a regulatory role for KAI1 in SPK activation, which leads to decreased invasive ability in disease progression of human pancreatic cancer.

(Hepatobiliary Pancreat Dis Int 2011; 10: 201-208)

adenovirus vector; KAI1 gene; sphingosine kinase; pancreatic cancer cell lines

Introduction

Pancreatic cancer remains one of the most difficult malignancies to diagnose and treat. In the United States, approximately 43 140 patients are diagnosed with pancreatic cancer annually, and nearly an equal number are expected to die of this disease in the same period. The risk for pancreatic cancer is nearly equal across both genders.[1]The majority of patients with pancreatic cancer are diagnosed when the disease has reached an advanced stage, and only 4% of patients survive up to 5 years post diagnosis due to the highly invasive and metastatic potential of this cancer.[1-3]At present, there is no effective treatment for this disease (radiation, chemotherapy, etc.), and complete surgical resection is the only therapeutic option available. These observations underscore the need for developing newer, improved therapies for this aggressive cancer.

Pancreatic cancer is a complex disease in which multiple subsets of genes undergo genetic activation or inactivation during cancer development, progression, and metastasis. Metastasis suppressor genes are characterised by their ability to prevent metastasis without affecting the primary tumor growth. Presently several metastasis suppressor genes, such as KAI1, MRP-1, mitogen activated protein kinase (MAPK) kinase 4, nm23-H, KiSS1 and BrMS1,[4]have been identified. Among them, it is believed that KAI1/CD82 may play a vital role in inhibiting metastasis. KAI1 that belongs to the transmembrane 4 superfamily (TM4SF) that regulates cell motility, morphology, fusion, signalling, fertilisation, and differentiation[5-8]was originally identified as a suppressor of metastatic spread of tumor cells in a rat prostate mode.[9]Decreased KAI1mRNA expression has been observed to correlate with the development of metastasis and poor prognosis in pancreatic cancer.[10-12]In other words, by increasing the expression of KAI1, the invasive ability of this cancer, bothin vitroandin vivo, can be decreased.

Sphingosine-1-phosphate (S1P) is a potent bioactive lipid produced by sphingosine kinase (SPK). It was demonstrated that sphingosine, in addition to being a structural component of cell membranes, plays a key role as a signalling molecule.[13]A large number of stimuli including cytokines and growth factors, for example, hepatocyte growth factor (HGF), stimulate SPK release, which leads to S1P production, which, in turn, by autocrine and paracrine mechanisms, activates S1PRs and their downstream signalling. This inside-out transactivation has been shown to be important for the migration and motility of cells.[14,15]

We developed a KAI1-expressing adenovirus vector and tested its anti-invasion effectin vitroon human pancreatic cancer cells. Based on the strong association between SPK/S1P and invasive behaviours of cancers, we examined the changes of SPK activity in pancreatic cancer cells before and after infection with Ad-KAI1. The interaction between KAI1 and c-Met, the tyrosine kinase receptor of HGF, was measured for a profound understanding of the mechanism KAI1 suppressing invasion by pancreatic cancer cells.

Methods

Cell line and culture conditions

Miapaca-2 and PANC1 human pancreatic ductal adenocarcinoma cells were provided by Dr. Helmut Friess from University of Bern, Switzerland. The cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM), supplemented with streptomycin (100 mg/ml), penicillin (100 U/ml), and 10% foetal calf serum (FCS) at 37 ℃ in a humidified atmosphere containing 5%CO2.

Construction of recombinant adenovirus

The 850-bp KAI1 cDNA was extracted from pCMVKAI1, which was provided by Professor J. Carl Barrett and Dr. Dong from National Institute of Health, USA, by digestion of PvuII (Takara Bio Inc., Japan), and it was cloned into pshuttle-CMV (Stratagene, USA) using ligase. The pAdshuttle-CMV-KAI1 and pAdEasy-1 (Stratagene, USA) underwent homologous recombination inEscherichia coliBJ5183 (kindly provided by Beijing Institute of Radiation Medicine, China) in accordance with the ADEasyTMAdenoviral Vector System, and the newly formed recombinant plasmid was verified by restriction endonuclease digestion with ApaI and ScaI (Takara Bio Inc., Japan). This was then linearized with PacI (Biolab, China), and propagated in AD-293 cells (American Type Cell Collection). The virus was harvested, amplified and titrated according to the manufacturer's instructions (Stratagene, USA). The recombinant replication-defective adenoviruses encoding KAI1 (Ad-KAI1) were purified by the Adeno-XTMVirus Purification Kit (BD Biosciences, USA). The final plaqueforming units were quantified by titration on 293 cells under an agarose overlay. The virus was then aliquoted and stored at -80 ℃. A control vector with GFP (Ad-GFP, developed at our laboratory) was constituted, expanded, purified and titrated as described previously.

Adenovirus infectionin vitro

The cells were infected with the recombinant replication-defective adenovirus Ad-GFP or Ad-KAI1 at different multiplicities of infection (MOI) in the range of 10-200. The infective efficiency was determined by fluorescent microscopy.

Flow cytometric analysis

The cells were harvested after treatment with 0.25% trypsin in a six-well plate. After being washed three times with PBS, the cells were incubated for 30 minutes at 4 ℃ with antibodies conjugated to phycoerythrin (PE) against CD82/KAI1 (Abcam, HK SAR, China). Washing was repeated in the same manner, and cell-surface immunofluorescence was analyzed by flow cytometry (eBioscience Dickinson, Mountain View, USA).

Western blotting analysis

The cells were infected with 100 MOI of Ad-KAI1 or Ad-GFP for 48 hours. Cell lysates were prepared using 50 μl of ice-cold cell-lysis buffer (50 mmol/L NaCl, 30 mmol/L Sodium pyrophosphate, 50 mmol/L NaF, 5 mmol/L NaCl, 10 mmol/L Tris-HCl (pH 7.4), 1% triton X-100 and 1 mmol/L leupeptin). The supernatants separated by centrifugation constituted 12.6 mg of protein from the samples that were boiled for 5 minutes and subjected to 10% SDS-PAGE treatment under reducing conditions. The proteins were transferred to a polvvinylidene difluorid membrane (Bio-Rad, USA) and the membrane was blocked in 20 ml of T-TBS, 5% skimmed milk and 0.5% Tween 20 for 1 hour followed by incubation with a 1∶1200 dilution of mouse anti-KAI1 antibody (BD Biosciences, USA) at room temperature for 2 hours. After five cycles of washing with T-TBS, the membrane was incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG atroom temperature for 50 minutes. The signals were then evaluated using the enhanced chemiluminescence system (ECL, Amersham Life Science Ltd.).

MTT assay

Cell growth was measured using a modified 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) method. The cells were seeded into 96-well plates at a concentration of 1×104cells per well in culture medium (100 ml). At different, specified time points, the medium was replaced with 100 μl of 2% MTT and the plate was incubated at 37 ℃ for 30 minutes. A volume of 100 μl of 0.2 mol/L Tris, pH 7.7, 4% formalin was then added to each well. After incubation at room temperature for 5 minutes, the liquid was removed and the wells were allowed to dry. Each well was rinsed with 200 μl of water followed by the addition of 100 μl dimethyl sulfoxide (containing 6.35% 0.1 mol/L NaOH) to each well. The optical absorption of each well at 562 nm was measured using a spectrophotometer.

Cell cycles

Cell-cycle analyses were performed by flow cytometry, and each group was examined for 3 times.

Cell-invasion assay

In vitroinvasive ability was tested by the Boyden Chamber assay according to the kit directions. The Invasion Chamber inserts (BD Biosciences, USA) with 8-mm pores in their PET membrane had been coated by matrigel. Prior to the invasion assay, the invasion chambers were rehydrated with DMEM (serum-free) for 2 hours in a humidified tissue culture incubator at 37 ℃ with 50 ml/L of CO2atmosphere. Six hundred microlitres of DMEM containing 1% acid-free bovine serum albumin with 100 ng/ml of HGF (Peprotech EC, UK) was added to the lower compartment of the invasion chamber, and 1.5×105cancer cells that were infected with Ad-KAI1 or Ad-GFP and control cells in serum-free DMEM were added to the upper compartment of the chamber. Each cell group was plated in three duplicate wells. After 24-hour incubation, the matrigel was removed, the filter was washed, and the cells were fixed and stained in Giemsa solution. Then, the cells, having migrated to the lower sides of the PET membrane in five random-visual fields (×200), were counted under a light microscope.

SPK-activity assay

After infection with Ad-KAI1 or Ad-GFP for 48 hours, the cells were resuspended in ice-cold 0.1 mol/L phosphate buffer (pH 7.4) containing 20% glycerol, 1 mmol/L mercaptoethanol, 1 mmol/L EDTA, phosphate inhibitors (1 mmol/L sodium orthovanadate, 15 mmol/L sodium fluoride), protease inhibitors (10 μg/ml each of leupeptin, aprotinin, trypsin, chymotrypsin, and 1 mmol/L phenylmethylsulfonyl fluoride), and 0.5 mmol/L 4-deoxypyridoxine. The cells were lysed by freeze-thaw cycles three times and lysates were separated by centrifugation at 12 000×g for 30 minutes at 4 ℃. The protein concentration of the cell lysates was determined using bicinchoninic acid, and the lysates were aliquoted and frozen at -80 ℃. Samples were assayed for SPK activity by incubation with sphingosine (Sigma, USA) and [γ-32P] ATP (5 μCi, 50 μmol/L final) for 30 minutes at 37 ℃, the products were separated by thin-layer chromatography on Silica Gel G60 (Merck KgaA, Germany) using 1-butanol/methanol/acetic acid/water (80∶20∶10∶20) and then visualized by autoradiography. Paternal cells and cells cultured with HGF for 24 hours were used as controls.

Co-immunoprecipitation

Cells were infected with Ad-KAI1 or Ad-GFP for 48 hours. The cell lysates were prepared using ice-cold celllysis buffer. For co-immunoprecipitation, 300 μg of cell lysate was incubated with an antibody-protein-A/G bead complex, made by incubating anti-c-Met antibody with protein A/G beads in 0.5 ml PBS at room temperature for 2 hours. The beads were subjected to three cycles of washing with lysis buffer, separated by 10% SDSPAGE, and transferred onto a polvvinylidene difluoride membrane (BioRad, USA). c-Met complexed with KAI1 was detected by Western blotting (1∶1200) with anti-KAI1.

Presentation of data and statistical analysis

In SPK-assay experiments, a blot representative of at least three similar results was presented. Densitometric analysis was performed using the Image-Pro Plus 6.0 software. The results of the study were expressed as mean±SD. Statistical significance between mean values was determined by Student'sttest, and aPvalue less than 0.05 was considered statistically significant.

Results

Construction of the recombinant adenovirus encoding the KAI1 gene

We constructed a recombinant adenovirus encoding the KAI1 gene, which was extracted from pCMV-KAI1. The lane 3 as the correct direction for obtaining pshuttle-CMV-KAI1 was verified to contain five fragments consistent with the theory value after SacI digestion (Fig. 1A). Then, we choose this particular clone to recombine with pAdEasy-1 inEscherichia coliBJ5183, and harvested the newly recombinant adenovirus plasmid, designated Ad-KAI1, which was tested by restriction endonuclease digestion with ApaI, ScaI and ScaIto for verification (Fig. 1B).

Adenoviral transfection efficiency of PANC1 and Miapaca-2 cells

PANC1 and Miapaca-2 cells were infected with Ad-GFP and the infection efficiency was evaluated by GFP fluorescence after 48 hours. A highly efficient transfection of nearly 83.2% and 91.6% was obtained at MOI of 100 pfu per cell in two cell lines, respectively (Fig. 2A). The cells were then infected with Ad-KAI1 at MOI of 100 pfu per cell. After 48 hours, the transfection efficiency of Ad-KAI1 was analyzed by flow cytometry (Fig. 2B). There were about 86.33% and 83.46% positive cells, in PANC1 and Miapaca-2 cells infected by Ad-KAI1, respectively, and these were renamed as PANC1-K and Miapaca-2-K. KAI1 protein was detected by Western blotting (Fig. 2C); the major specific KAI1 protein was clearly visible at 29-kDa in PANC1 and Miapaca-2 cells infected with Ad-KAI1, but almost undetectable in cells infected with Ad-GFP.

Growth curve and cell cycles

Fig. 1. Construction of recombinant adenovirus carrying the KAI1 gene. A: Correct directional fragments were released on lane 3 (454bp) and, compared with it, the reverse direction fragment and no fragment were released on lanes 2, 4 and lane 5, 6, 7 with self-looped of PAdtrack-CMV-KAI1, respectively. Lanes 1 and 8 were HindIII marker and DL2000 marker. B: Correct direction fragments of pAd-KAI1 were released in lanes 2 and 3 after ApaI and ScaI digestion. Lanes 1 and 4 were determined for HindIII marker and DL2000 marker, respectively.

Thein vitrocell growth rates of both infected with Ad-GFP and Ad-KAI1 and control clones were measured by MTT assay in PANC1 and Miapaca-2 and plotted (Fig. 3). No significant differences in growth curve and cell cycles were observed in the two cell lines in 96 consecutive hours (Table 1). The results demonstrate that the KAI1 gene had no obvious effect on the proliferation ability of the cells.

Fig. 2. Efficiency of infection and expression of KAI1 in PANC1 and Miapaca-2 cells. A: After infection with Ad-GFP for 48 hours at 100 MOI, the fluorescence images of PANC1 and Miapaca-2 cells were obtained using fluorescence microscopy; B: PANC1 and Miapaca-2 cells were infected with Ad-KAI1 or Ad-GFP, harvested after 48 hours, and assayed for efficiency of infection with Ad-KAI1 by FACS. In the histogram, expression of KAI1 in cells is depicted by the black-colored area. About 86.33% and 83.46% positive cells in Ad-KAI1-infected PANC1 and Miapaca-2 cells respectively were revealed and they were much higher than those of cells infected with Ad-GFP and the controls; C: Cells infected as in B were harvested and assayed for expression of KAI1 by Western blotting. Total protein was isolated from the cell lines and subjected to Western blotting analysis with an antibody specific for KAI1/CD82. PANC1 and Miapaca-2 cells infected with Ad-KAI1 manifested a high level of KAI1 expression at 29-kDa band, whereas KAI1 was undetectable or present at low levels in control or Ad-GFP-infected cells.

Fig. 3. Growth curves of PANC1 and Miapaca-2 cells infected with Ad-KAI1 or Ad-GFP and controls. PANC1 and Miapaca-2 cells were infected with Ad-KAI1 or Ad-GFP, harvested after 48 hours, and transferred to 96-well plates (1×104cells per well). The number of viable cells was determined at various times thereafter and the optical absorption of each well at 562 nm was measured using a spectrophotometer.

Table 1. Cell-cycle distribution of different cells revealed by flow cytometry (%, mean±SD)

Effect of over-expressed KAI1 on cell invasionin vitro

The effect of Ad-KAI1 transduction was assessed by evaluating the invasion capacity of PANC1 and Miapaca-2 cell lines by the Matrigel transwell system (Fig. 4). The number of cells migrating through the Matrigel matrix was counted, and the results are presented in Table 2. PANC1-K and Miapaca-2-K showed significantly reduced invasiveness as compared with control or Ad-GFP-infected cells (P＜0.05). These data indicate that the enhanced expression of KAI1 in pancreatic cancer cells is associated with reduced invasive capability.

Analysis of SPK activity

Fig. 4. Effect of invasion in PANC1 and Miapaca-2 cells infected with Ad-KAI1 or Ad-GFP and controls. PANC1 and Miapaca-2 cells were infected with Ad-KAI1 or Ad-GFP, harvested after 48 hours, and assayed for cell invasiveness with a Boyden Chamber, respectively, as described in detail in the methods section. The cells that had migrated through the membrane after incubation for 24 hours were stained and assessed under a light microscope.

Table 2. The number of cells penetrating the filter membrane (mean±SD)

Fig. 5. S1P was measuredin vitrofrom PANC1 and Miapaca-2 cells. Total protein was isolated from PANC1 and Miapaca-2 cells, cultured with HGF for 24 hours and infected with Ad-KAI1 or Ad-GFP for 48 hours, and was separated by thin-layer chromatography on Silica Gel G60. A low level of S1P expression was detected in PANC1-K and Miapaca-2-K cells. The highest level of S1P expression was detected in PANC1 and Miapaca-2 cells cultured with HGF.

The sphingosine kinase activity in PANC1 and Miapaca-2 cell lysates was determined as described in the methods section. The production of S1P, which was catalyzed by SPK, was significantly decreased in PANC1-K and Miapaca-2-K cells, compared with cellsinfected with Ad-GFP or control and cultured with HGF (Fig. 5).Analysis of c-Met interaction with KAI1

Fig. 6. The interaction between KAI1 and c-Met was measured by co-immunoprecipitation. After immunoprecipitated with anti-c-Met antibody, the lysates isolated from PANC1-K and Miapaca-2-K cells were subjected to Western blotting analysis with an antibody specific for KAI1. There was a moderate level of c-Met expression at 170-kDa band, whereas KAI1 was undetectable at 29-kDa band.

The proteins extracted from PANC1-K and Miapaca-2-K were used for co-immunoprecipitation by using antibodies directed to c-Met and KAI1. c-Met was observed in two kinds of cells; however, the evidence of complexes containing KAI1 was not observed (Fig. 6).

Discussion

KAI1 expression has been reported to be inversely correlated with the metastatic potential of some cancer cells both at messenger and protein levels. To further elucidate the effect of KAI1 on pancreatic cancer metastasis, in this study we inserted Ad-KAI1 cDNA into two highly malignant pancreatic cancer cell lines, PANC1 and Miapaca-2, both of which have low levels of endogeneous KAI1 expression. Our results showed that cells infected with Ad-KAI1 have over-expression of the KAI1 gene, and this finding is similar to that reported previously.[16]

A technique for production of replication-deficient adenovirus in high titres has proven to have excellent safety in the human body.[17]As a useful tool it has been widely used in gene delivery bothin vitroandin vivoconditions.[18,19]However, the expression of a transgene delivered adenovirally involves several steps of vectorhost-cell interaction,[20-22]and there are several factors such as different infection intensities in different cell lines, that limit the effect of adenovirus-mediated gene transfer into cancer cells. In the present study, we selected the appropriate infection intensity by Ad-GFP estimation. Fluorescence microscopy demonstrated that the infection efficiency was up to 83.2% and 91.6% in PANC1 and Miapaca-2 cells at MOI of 100 pfu per cell infected with Ad-GFP. Therefore, high over-expression of KAI1 was achieved in PANC1 and Miapaca-2 cells infected with Ad-KAI1 at MOI of 100 pfu per cell, which was verified by Western blotting and flow cytometry. These observations could form the basis for further investigation of the anti-invasion mechanism that is exerted by KAI1 in pancreatic carcinoma.

The present study indicate that the KAI1 gene had no obvious effects onin vitrocell growth and proliferation of pancreatic cancer PANC1 and Miapaca-2 cells. These findings were consistent with those reported by previous studies in pancreatic cancer and other cancer cells.[7]Our results also demonstrated that the KAI1 gene significantly suppressed the invasion potential. Invasion has been found to be one of the important and necessary properties for tumor metastasis formation.[23-28]The Boyden Chamber assay, which imitates the invasion process ofin vivotumor cells, has also been found to be an ideal method for evaluating the invasive and metastatic abilities of tumor cells.[29]A report indicated that KAI1 inhibits the invasion of cancer cells.[12]However, little is known about its molecular mechanism as a negative regulator of pancreatic cancer invasion.[5,30-32]

In our study the over-expression of KAI1 gene in pancreatic cancer PANC1 and Miapaca-2 cells and S1P was decreased. Many studies have shown that external stimuli-activated SPK results in an increase of intracellular S1P, which is an intracellular second messenger contributing to the migration of different types of cells.[14,15]Therefore, it is possible that KAI1 regulates SPK activity, thus leading to decreased invasion of PANC1 and Miapaca-2 cells.

A key enzyme may also catalyze the formation of S1P in response to diverse stimuli, such as platelet derived growth factor, nerve growth factor (NGF),[33]epithelial growth factor (EGF),[34]TNF-α,[35]vitamin D3,[36]phorbol ester, serum, oxidised LDL and HGF.[37]HGF, also known as a scatter factor and hepatoprotein A, is a multifunctional factor regulating cell growth, motility, and migrationin vitro. Our data prove that HGF promotes the expression of S1P and the invasion ability of pancreatic cancer cells; in addition, this action of promotion is less in Ad-KAI1 group of cells than in Ad-GFP and control group cells. These diverse biological functions of HGF are mediated through the activation of its tyrosine kinase receptor, the c-Met proto-oncogene.[38]It has previously been reported that c-Met is highly expressed in pancreatic cancer cell lines[39]and the activated c-Met receptor recruits several signalling proteins, such as PI3K, PKB/Akt, and ERK/MAPK,[40,41]leading to activation of invasion. Ourinvestigation shows that c-Met is expressed in PANC1-K and Miapaca-2-K cells; however there is no direct correlation between KAI1 and c-Met, proved by the results of co-immunoprecipitation. This is consistent with previous results indicating that KAI1 affects c-Met activation through an indirect mechanism in prostate cancer.[42]Another report suggested that KAI1 decreased the metastatic phenotype of H1299 lung carcinoma cells through the PI3K/Akt/mTOR pathway.[43]We, therefore, hypothesize that SPK is inactivated, presumably, but not directly by KAI1/c-Met to decrease its invasive ability in pancreatic cancer induced by HGF rather by c-Met receptor-mediated means such as PI3K, PKB/Akt, and ERK/MAPK.

Invasion/metastases are devastating events in pancreatic cancer patients. Regional inhibition of these events has been proposed as a target for current and future cancer treatment. The present findings indicate that over-expression of the KAI1 gene and lower SPK activity might inhibit the migration and invasion of pancreatic cancer cells, suggesting that a regional delivery system with Ad-KAI1 might be a reasonable strategy in the treatment of pancreatic cancer for preventing metastasis.

Funding: The study was supported by grants from the National Nature Science Foundation of China (39970334 and 30470798).

Ethical approval: Not needed.

Contributors: LX proposed the study. LX and GXZ wrote the first draft. All authors contributed to the design and interpretation of the study and to further drafts. GXZ is the guarantor.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

1 Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277-300.

2 Chames P, Kerfelec B, Baty D. Therapeutic antibodies for the treatment of pancreatic cancer. ScientificWorldJournal 2010; 10:1107-1120.

3 Klapman J, Malafa MP. Early detection of pancreatic cancer: why, who, and how to screen. Cancer Control 2008;15:280-287.

4 Kauffman EC, Robinson VL, Stadler WM, Sokoloff MH, Rinker-Schaeffer CW. Metastasis suppression: the evolving role of metastasis suppressor genes for regulating cancer cell growth at the secondary site. J Urol 2003;169:1122-1133.

5 Malik FA, Sanders AJ, Jiang WG. KAI-1/CD82, the molecule and clinical implication in cancer and cancer metastasis. Histol Histopathol 2009;24:519-530.

6 Yá?ez-Mó M, Barreiro O, Gordon-Alonso M, Sala-Valdés M, Sánchez-Madrid F. Tetraspanin-enriched microdomains: a functional unit in cell plasma membranes. Trends Cell Biol 2009;19:434-446.

7 Miranti CK. Controlling cell surface dynamics and signaling: how CD82/KAI1 suppresses metastasis. Cell Signal 2009;21: 196-211.

8 Lazo PA. Functional implications of tetraspanin proteins in cancer biology. Cancer Sci 2007;98:1666-1677.

9 Dong JT, Lamb PW, Rinker-Schaeffer CW, Vukanovic J, Ichikawa T, Isaacs JT, et al. KAI1, a metastasis suppressor gene for prostate cancer on human chromosome 11p11.2. Science 1995;268:884-886.

10 Friess H, Guo XZ, Berberat P, Graber HU, Zimmermann A, Korc M, et al. Reduced KAI1 expression in pancreatic cancer is associated with lymph node and distant metastases. Int J Cancer 1998;79:349-355.

11 Guo XZ, Friess H, Maurer C, Berberat P, Tang WH, Zimmermann A, et al. KAI1 is unchanged in metastatic and nonmetastatic esophageal and gastric cancers. Cancer Res 1998;58:753-758.

12 Guo XZ, Cui ZM. Current diagnosis and treatment of pancreatic cancer in China. Pancreas 2005;31:13-22.

13 Pyne NJ, Pyne S. Sphingosine 1-phosphate and cancer. Nat Rev Cancer 2010;10:489-503.

14 Leong WI, Saba JD. S1P metabolism in cancer and other pathological conditions. Biochimie 2010;92:716-723.

15 Takabe K, Paugh SW, Milstien S, Spiegel S. "Inside-out" signaling of sphingosine-1-phosphate: therapeutic targets. Pharmacol Rev 2008;60:181-195.

16 Mu Z, Wang H, Zhang J, Li Q, Wang L, Guo X. KAI1/CD82 suppresses hepatocyte growth factor-induced migration of hepatoma cells via upregulation of Sprouty2. Sci China C Life Sci 2008;51:648-654.

17 Pipiya T, Sauthoff H, Huang YQ, Chang B, Cheng J, Heitner S, et al. Hypoxia reduces adenoviral replication in cancer cells by downregulation of viral protein expression. Gene Ther 2005;12:911-917.

18 Gregory LG, Harbottle RP, Lawrence L, Knapton HJ, Themis M, Coutelle C. Enhancement of adenovirus-mediated gene transfer to the airways by DEAE dextran and sodium capratein vivo. Mol Ther 2003;7:19-26.

19 Heilbronn R, Weger S. Viral vectors for gene transfer: current status of gene therapeutics. Handb Exp Pharmacol 2010;(197):143-170.

20 Bergelson JM, Cunningham JA, Droguett G, Kurt-Jones EA, Krithivas A, Hong JS, et al. Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 1997;275:1320-1323.

21 Pearson AS, Koch PE, Atkinson N, Xiong M, Finberg RW, Roth JA, et al. Factors limiting adenovirus-mediated gene transfer into human lung and pancreatic cancer cell lines. Clin Cancer Res 1999;5:4208-4213.

22 Alemany R. Designing adenoviral vectors for tumor-specific targeting. Methods Mol Biol 2009;542:57-74.

23 Kurokawa H, Katsube K, Podyma KA, Ikuta M, Iseki H, Nakajima M, et al. Heparanase and tumor invasion patterns in human oral squamous cell carcinoma xenografts. Cancer Sci 2003;94:277-285.

24 Geiger TR, Peeper DS. Metastasis mechanisms. Biochim Biophys Acta 2009;1796:293-308.

25 Katayama M, Sanzen N, Funakoshi A, Sekiguchi K. Laminin gamma2-chain fragment in the circulation: a prognostic indicator of epithelial tumor invasion. Cancer Res 2003;63:222-229.

26 Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression and metastasis. Cell 2010;141:39-51.

27 Patarroyo M, Tryggvason K, Virtanen I. Laminin isoforms in tumor invasion, angiogenesis and metastasis. Semin Cancer Biol 2002;12:197-207.

28 Mareel M, Oliveira MJ, Madani I. Cancer invasion and metastasis: interacting ecosystems. Virchows Arch 2009;454: 599-622.

29 Kleinman HK, Jacob K. Invasion assays. Curr Protoc Cell Biol 2001;Chapter 12:Unit 12.2.

30 Jackson P, Marreiros A, Russell PJ. KAI1 tetraspanin and metastasis suppressor. Int J Biochem Cell Biol 2005;37:530-534.

31 Tonoli H, Barrett JC. CD82 metastasis suppressor gene: a potential target for new therapeutics? Trends Mol Med 2005; 11:563-570.

32 Liu WM, Zhang XA. KAI1/CD82, a tumor metastasis suppressor. Cancer Lett 2006;240:183-194.

33 Zhang YH, Vasko MR, Nicol GD. Intracellular sphingosine 1-phosphate mediates the increased excitability produced by nerve growth factor in rat sensory neurons. J Physiol 2006; 575:101-113.

34 Johnstone ED, Speake PF, Sibley CP. Epidermal growth factor and sphingosine-1-phosphate stimulate Na+/H+ exchanger activity in the human placental syncytiotrophoblast. Am J Physiol Regul Integr Comp Physiol 2007;293:R2290-2294.

35 Chandru H, Boggaram V. The role of sphingosine 1-phosphate in the TNF-alpha induction of IL-8 gene expression in lung epithelial cells. Gene 2007;391:150-160.

36 Buehrer BM, Bardes ES, Bell RM. Protein kinase C-dependent regulation of human erythroleukemia (HEL) cell sphingosine kinase activity. Biochim Biophys Acta 1996;1303:233-242.

37 Duan HF, Qu CK, Zhang QW, Yu WM, Wang H, Wu CT, et al. Shp-2 tyrosine phosphatase is required for hepatocyte growth factor-induced activation of sphingosine kinase and migration in embryonic fibroblasts. Cell Signal 2006;18:2049-2055.

38 Bottaro DP, Rubin JS, Faletto DL, Chan AM, Kmiecik TE, Vande Woude GF, et al. Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene product. Science 1991;251:802-804.

39 Ide T, Kitajima Y, Miyoshi A, Ohtsuka T, Mitsuno M, Ohtaka K, et al. The hypoxic environment in tumor-stromal cells accelerates pancreatic cancer progression via the activation of paracrine hepatocyte growth factor/c-Met signaling. Ann Surg Oncol 2007;14:2600-2607.

40 Tanimura S, Nomura K, Ozaki K, Tsujimoto M, Kondo T, Kohno M. Prolonged nuclear retention of activated extracellular signal-regulated kinase 1/2 is required for hepatocyte growth factor-induced cell motility. J Biol Chem 2002;277:28256-28264.

41 Dong G, Chen Z, Li ZY, Yeh NT, Bancroft CC, Van Waes C. Hepatocyte growth factor/scatter factor-induced activation of MEK and PI3K signal pathways contributes to expression of proangiogenic cytokines interleukin-8 and vascular endothelial growth factor in head and neck squamous cell carcinoma. Cancer Res 2001;61:5911-5918.

42 Sridhar SC, Miranti CK. Tetraspanin KAI1/CD82 suppresses invasion by inhibiting integrin-dependent crosstalk with c-Met receptor and Src kinases. Oncogene 2006;25:2367-2378.

43 Choi UJ, Jee BK, Lim Y, Lee KH. KAI1/CD82 decreases Rac1 expression and cell proliferation through PI3K/Akt/mTOR pathway in H1299 lung carcinoma cells. Cell Biochem Funct 2009;27:40-47.

Received November 5, 2010

Accepted after revision December 21, 2010

Author Affiliations: State Key Laboratory of Cancer Biology and Institute of Digestive Diseases, Xijing Hospital of Digestive Disease, Fourth Military Medical University, Xi'an 710032, China (Liu X); Department of Gastroenterology, Shenyang General Hospital of PLA, Shenyang 110016, China (Liu X, Guo XZ and Zhang WW); Department of Experimental Hematology, Beijing Institute of Radiation Medicine, Beijing 100850, China (Lu ZZ, Zhang QW, Duan HF and Wang LS)

Xiao-Zhong Guo, MD, Department of Gastroenterology, Shenyang General Hospital of PLA, Shenyang 110016, China (Tel: 86-24-28856230; Fax: 86-24-23848453; Email: guoxiaozhong1962@163. com)

? 2011, Hepatobiliary Pancreat Dis Int. All rights reserved.