The Effect of Disintegrin–Metalloproteinase ADAM9 in Gastric Cancer Progression

Published OnlineFirst October 24, 2014; DOI: 10.1158/1535-7163.MCT-13-1001

Molecular Cancer Cancer Biology and Signal Transduction Therapeutics

Jeong Min Kim1,2, Hei-Cheul Jeung1,3, Sun Young Rha1,2,3, Eun Jeong Yu1,4, Tae Soo Kim1, You Keun Shin1, Xianglan Zhang1, Kyu Hyun Park1, Seung Woo Park2,3, Hyun Cheol Chung1,2,3, and Garth Powis5

Abstract Advanced gastric cancer is one of the most aggressive gastrointestinal malignancies, and ADAM (A disintegrin and metalloproteinase)-9 is a cell-surface membrane glycoprotein with oncogenic properties that is overexpressed in several cancers. Herein, we investigated the biologic mechanism of ADAM9 in the progression, proliferation, and invasion of gastric cancer. First, we detected ADAM’s expression, processing, and protease activity in gastric cancer cells. Protease activity was moderately correlated with ADAM9 protein expression, but was better related to a processed smaller molecular weight (84 kDa) form of ADAM9. Knockdown of ADAM9 or speciﬁcally targeted monoclonal antibody (RAV-18) suppressed cancer cell proliferation and invasion in high ADAM9-expressing cells, not in low ADAM9-expressing cells. RAV- 18 showed in vivo antitumor activity in a gastric cancer xenograft model. Hypoxia (1% oxygen) induced ADAM9 expression and functional activity in low ADAM9-expressing gastric cancer cells that was inhibited by siRNA knockdown or RAV-18 antibody to levels in normoxic cells. Overall, our studies show that ADAM9 plays an important role in gastric cancer proliferation and invasion, and that while expressed in some gastric cancer cells at high levels that are responsive to functional inhibition and antitumor activity of a catalytic site–directed antibody, other gastric cancer cells have low levels of expression and only when exposed to hypoxia do ADAM9 levels increase and the cells become responsive to ADAM9 antibody inhibition. Therefore, our ﬁndings suggest that ADAM9 could be an effective therapeutic target for advanced gastric cancer. Mol Cancer Ther; 13(12); 3074–85. 2014 AACR.

Introduction matrix (ECM), and "invasion" is the first step by which Gastric cancer is the second leading cause of cancer- cancer cells break away from the primary tumor and related mortality worldwide with a median survival of infiltrate surrounding stoma (1). Proteolysis has only around a year after diagnosis of advanced disease. emerged as a key posttranslational regulator of inva- This dismal prognosis mainlyresultsfromearlyinva- sion in gastric cancer (2, 3). However, the mechanism sion and high metastatic activity with an unknown and major players in invasion have not yet been fully molecular basis. Cancer metastasis is a complex, mul- elucidated. tistep process involving the interplay between cancer A disintegrin and metalloproteases (ADAM) are cells and surrounding components of extracellular modular type I transmembrane proteins that contain a metalloprotease and a disintegrin-like domain, and recent studies have demonstrated a relationship between 1Cancer Metastasis Research Center, Institute for Cancer Research, increased ADAMs and cancer progression (4). Twenty- Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, one members of ADAM family have been identified in 2 Republic of Korea. Brain Korea 21 Projects for Medical Science, Yonsei humans, 13 of which have functional proteases. ADAM9 University College of Medicine, Seoul, Republic of Korea. 3Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Republic is shown to possess potent biologic activities and is of Korea. 4Department of Biology, Baylor University, Waco, Texas. 5San- highly expressed in cancers, including prostate, pancre- ford-Burnham Research Institute Cancer Center, La Jolla, California. as, breast, and colon, associated with cancer progression Note: Supplementary data for this article are available at Molecular Cancer and poor clinical outcome (5–9). In prostate cancer, Therapeutics Online (http://mct.aacrjournals.org/). microenvironmental stress on tumor cells leads to shed- Corresponding Author: Hei-Cheul Jeung, Department of Internal Medi- ding of proheparin-binding epidermal growth factor cine, Gangnam Severance Hospital, Yonsei University College of Medicine, 211 Eonju-Ro (146-92 Dogok-Dong), Gangnam-Gu, Seoul 135-720, Korea. (EGF), which is processed by ADAM9. The metallopro- Phone: 82-2-2019-1242; Fax: 82-2-362-5592; E-mail: tease domain of ADAM9 also cleaves the insulin b-chain, [email protected] tumor necrosis factor (TNF)-a, transforming growth doi: 10.1158/1535-7163.MCT-13-1001 factor (TGF)-a, gelatin, b-casein, and so on, and induces 2014 American Association for Cancer Research. the shedding of EGF, fibroblast-growth factor receptor

3074 Mol Cancer Ther; 13(12) December 2014

ADAM9 in Gastric Cancer

2IIIB, and heparin-binding EGF-like growth factor RNA extraction and RT-PCR (10, 11). Therefore, the abilities of ADAM9 to degrade Total RNA was extracted using TRizol reagent specific ECM substrates, release active growth factors, (Qiazen Science) according to the manufacturer’s pro- interact with key regulatory factors, and its expres- tocol. Briefly, 500 mL of TRizol was added to 1 107 sion at the invasion edges of several cancer metastases cells, and after adding 200 mL chloroform, cell homo- suggest that ADAM9 regulates several cancer-related genates were centrifuged at 13,200 rpm for 25 minutes at processes—including cell growth and invasion. 4C,andtheaqueousphaseswerecollected.TotalRNA A previous report has shown that the transcription was precipitated with the same volume of isopropanol of ADAM9 is upregulated in gastric cancer (12); how- for 60 minutes at 20C and centrifuged for 30 minutes. ever, its biologic function and potential as a thera- The pellets were washed once with 70% ethanol and peutic target are largely unknown. In this study, we suspended in RNase-free water. The quantity of total investigated the biologic role of ADAM9 in gastric RNA was measured using a NanoDrop spectrophoto- cancer invasion, and explored a therapeutic approach meter (Thermo Scientific). of antimetastasis treatment for high ADAM9-ex- To synthesize the cDNA, 4 mg of total RNA from pressing cancers using an ADAM9-specific inhibitory each cell line was mixed with the oligo-dT primer and antibody. incubated at 65C for 10 minutes. After adding the Super- Script II, 5 first strand buffer, 100 mmol/L DTT, and 10 mmol/L dNTP mix to the RNA/oligo-dT mixture, a Materials and Methods reverse transcription process was performed at 42C Reagents and antibodies for 90 minutes. The remaining RNA was hydrolyzed Matrigel was supplied by BD Biosciences. Recombinant by incubation at 65C for 15 minutes in 0.1 N NaOH. The ADAM9 protein was from R&D Systems Inc. ADAM9 reaction was then neutralized by the addition of an equal antibodies were from Cell Signaling Technology for volume of 0.1 N HCl. immunoblot and GeneTex for immunohistochemistry. Amplification reactions were performed in an Eppen- Antibodies of hypoxia-inducible factor-1a (HIF1a) and dorf Mastercycler Grandient (Eppendorf). PCR was per- total/phospho forms of EGFR and ERK were obtained formed using 1 mg of cDNA, 2.5 mmol/L dNTPs, 1.5 from Santa Cruz Biotechnology. Antibodies against mmol/L 10 PCR buffer with MgCl2, and 5 U Taq poly- GAPDH and a-tubulin were purchased from Abcam and merase (Invitrogen) to total 50 mL of reaction volume. The Sigma-Aldrich. Total and phospho-Akt antibodies were primer sequences used were as follows: ADAM9 (for- purchased from Cell Signaling Technology. RAV-18, an ward: 50-AGT GGC GGG AAA AGT TTC TT-30; reverse: anti-ADAM9-specific blocking antibody, was kindly pro- 50-CCA GCG TCC ACC AAC TTA TT-30). The conditions vided by Macrogenics Inc. for PCR amplification were as follows: 94C for 2 minutes followed by 94C for 40 seconds, 60C for 50 seconds and Cell culture 72C for 1 minutes, then 72C for 2 minutes, and repeated The human gastric cancer cell lines AGS, NCI-N87, for 30 cycles. PCR products were separated on ethidium Hs746T, and KATO-III and the cervical cancer cell line bromide gels containing 1.2% agarose. HeLa, positive control of ADAM9, were purchased from the American Type Culture Collection (ATCC) in 2012. Exposure to hypoxic conditions SNU-1, -5, and -16 in 2006, SNU-216, -638, and -668 in When cells reached 80% confluence in a 25 cm2 flask, 2007, SNU-484 in 2008, MKN-1, -28, -45, and -74 were the flask was placed in a hypoxic chamber (Personal supplied from the Korean Cell Line Bank (KCLB). O2/CO2 Incubator; Aspect). Cells were exposed to 1% Above cell lines were authenticated by standard short O2,5%CO2,and37 C for 24 hours. After that, cells were tandem repeat (STR) DNA typing methodology before collected and transferred to a Matrigel-coated 24-well being purchased from the ATCC and KCLB. YCC-1, -2, plate for invasion assay. Invaded cells were stained and -3 were established in 1989, YCC-6 and -7 were using 0.5% crystal violet and counted after 24 hours. established in 1994, and YCC-9, -10, -11, and -16 were Protease activity assay, immunobloting, and prolife- established in 2000 by the Cancer Metastasis Research ration assay were conducted using cells exposed to Center at Yonsei University College of Medicine (Seoul, hypoxic conditions in the same way. Korea) from the ascites or peripheral blood (YCC-16) of patients with advanced gastric cancer. YCC cell lines Immunoblotting were authenticated by STR in July 20, 2012, that had Subconfluent cells were harvested and suspended in a been outsourced to the KCLB. All cells were maintained whole cell lysis buffer [50 mmol/L Tris (pH 8.0), 10% in RPMI-1640 (Nissui) and supplemented with 10% glycerol, 2 mmol/L EDTA, 5 mmol/L NaF, 1% NP-40, FBS (Omega Scientific, Inc.), 100 U/mL ampicillin, 150 mmol/L NaCl, 1 mmol/L sodium orthovanadate, 100 mg/mL streptomycin, and 2 mmol/L glutamine 1 mmol/L NaF, 10 mg/mL aprotinin, 10 mg/mL leupeptin, (Gibco) in 5% CO2 humidified atmosphere at 37 C. All and 1 mmol/L phenyl-methyl-sulfonyl fluoride]. The cell lines were expanded and cryopreserved in liquid lysates were subjected to SDS-PAGE and blotted onto nitrogen in our laboratory. the polyvinylidenedifluoride (PVDF) membrane. The

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3075

Kim et al.

membrane was then incubated with the appropriately violet at room temperature for 10 minutes. Invasion diluted primary antibody overnight at 4C, followed by activity was quantified by counting the cells on three a horseradish peroxidase (HRP)–conjugated secondary inserts. The data are expressed as the average number of antibody for 60 minutes at room temperature. The mem- cells per insert. brane was visualized with an ECL Western blotting reagent kit (Amersham Bioscience). Adhesion assay Harvested cells were suspended in media containing Measurement of ADAM9 protease activity 0.1% BSA, and 1 105 cells were seeded into 8 mg/mL Protease activity of ADAM9 was measured by a mod- Matrigel-precoated 96-well plates that were then incubat- ification of the method of Cisse and colleagues (13). ed for 2 hours with or without RAV-18. Media was Briefly, 5 104 cells were seeded in each well of a 96- removed and the attached cells were stained with 0.5% well plate (BD Falcon) and cultured for 1 day. The cells crystal violet at room temperature for 10 minutes. Stained were rinsed with PBS twice, and 100 mL PBS containing cells were dissolved in 0.1 mol/L sodium citrate and 1 mmol/L ADAM9 substrate (R&D Systems Inc.) was measured by using ELISA reader at 570 nm. added and incubated at 37C for 8 hours. Supernatant was transferred to a black 96-well plate and the fluores- Survival assay cence activity was measured at 320/405 nm using a Cells were seeded into a 96-well plate and incubated Spectrophotometer (PerkinElmer). To normalize the fluo- at 37C for 24 hours. RAV-18 was serially diluted with rescent activity to cell number, the cells remained in the media and added to each well. Following 72 hours of plate and were treated with a MTT (3-(4,5-dimethylthia- incubation, 50 mL(2mg/mL)oftheMTTsolutionwas zol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma) solu- added and incubated for an additional 4 hours. After tion. After incubation for 4 hours, the MTT solution was centrifugation at 400 g for 10 minutes, the media and discarded and the MTT-formazan crystals were dissolved MTT were removed from the wells and the remaining in 150-mL DMSO. The absorbance was measured at 570 nm MTT-formazan crystals were dissolved by addition of using a multiwell ELISA automatic Spectrometer (Behr- 150 mL of DMSO. Following 10 minutes of shaking ing ELISA Processor II). The protease activity was calcu- incubation, the absorbance at 570 nm was measured lated using the following equation: protease activity ¼ with multiwell ELISA automatic spectrometer. Results (fluorescence at 320/405 nm)/absorbance of MTT. were expressed as percentage cell survival, which was calculated using the following formula: %survival ¼ ADAM and HIF1a knockdown experiment [(mean absorbance of test wells standard absor- Small interfering RNA (siRNA) against ADAM9 was bance)/(mean absorbance of control wells standard purchased from Origene and was also independently absorbance)] 100. Control wells were treated with the designed using the Genescript siRNA Finder (https:// medium alone (without the drug). www.genscript.com/ssl-bin/app/rnai). HIF1a siRNA was designed by Genescript siRNA Finder. The ADAM9 Proliferation assay and HIF1a siRNA target sequence are as follows: Cells (5 104) were seeded in a 24-well plate with or ADAM9, 50-GACTGTCGGTTCCTTCCAG-30 and HIF1a, without treatment of RAV-18. Cells were counted every 50-GTCTGCAACATGGAAGGTA-30. ADAM-10 and -17 day for 7 days. Cellular doubling time (DT) was calculated siRNA were supplied from Bioneer Corporation. Cells by the following equation: DT ¼ (time after cell growth were transfected with Lipofectamine MAX (Invitrogen) time before cell growth)/log2 (number of cells after mixed with 50 nmol/L siRNA. After 48 hours, cells were growth/number of cells before growth). harvested for further analysis. In vivo xenograft model Matrigel Transwell assay BALB/c nu/nu mice (female, 7-weeks old; SLC Inc.) The lower side of an insert (diameter, 6.5 mm; pore were housed under specific pathogen-free conditions. size, 8 mm; Becton Dickinson) was coated with 8 mg/mL Experiments were performed according to the standard Matrigel. Filters were placed on a 24-well containing guidelines for animal experiments of Yonsei University media supplemented with 0.1% bovine serum albumin College of Medicine. The effect of RAV-18 on the xeno- (BSA). Cells were harvested with a cell dissociation graft model was examined as follows: 1 107 MKN-28 solution (Sigma) and suspended in a medium with cells were inoculated subcutaneously (s.c.) in the flank 3% BSA. Cells (1 105) were then placed in the upper of the mouse or injected intraperitoneally (IP). The part of a Transwell chamber and allowed to migrate for mice were divided into four groups: a control group 24 hours. of s.c. (PBS i.p.; n ¼ 7), a RAV-18–treated group of s.c. For ADAM9 inhibitor treatment, RAV-18 was added (50 mg/kg i.p., five times for 2 weeks; n ¼ 7), a control to the upper compartment in serial doses. After 24 groupofIP(PBSi.p.;n ¼ 7), and a RAV-18–treated hours, noninvaded cells were swabbed off, and invaded group of IP (50 mg/kg i.p., five times for 2 weeks; n ¼ 7). cells on the lower side of the membrane were fixed in 4% The treatment was started on day 21 after cell inocula- paraformaldehyde in PBS and stained with 0.5% crystal tionandmiceweresacrificedafter8weeks.Tumor

3076 Mol Cancer Ther; 13(12) December 2014 Molecular Cancer Therapeutics

ADAM9 in Gastric Cancer

volume and body weight were measured twice weekly. Effect of ADAM9 suppression in gastric cancer cells The tumor volume was calculated using the formula: First, we screened four siRNA against ADAM9 and volume ¼ length width width 0.5. At the end of selectedthemostefficientoneforfurtherexperiment the experiment, tumors and peritoneal nodules were (Supplementary Fig. S2). In cells with high ADAM9 collected. The weights of collected samples were mea- expression, protein levels were reduced, and its prote- sured and the peritoneal nodules were counted. ase activity was also decreased significantly by 61.8% 3.9% in SNU-638 and 42.7% 4.5% in MKN-28 (Fig. 2A Immunohistochemistry and B; P < 0.01). The knockdown of ADAM9 led to Tumor specimens were fixed in 10% formaldehyde decline in phospho-EGFR and phospho-ERK (Fig. 2A), and embedded in paraffin. All samples were cut into Then, we examined the invasion after siADAM9 trans- 5-mm-thick sections for immunohistochemistry (IHC). fection. Cells transfected with siADAM9 showed a The sections were stained with H&E and immunos- reduced invasion activity through Matrigel compared tained with anti-ADAM9 (1:100), anti-pEGFR (1:200), with control cells by 71.4% 2.2% in SNU-638 and by and anti-pERK (1:100) antibodies at room temperature 53.0% 3.9% in MKN-28 (Fig. 2C; P < 0.01), which for 90 minutes. The sections were reacted with an EnVi- suggests that the decreased invasion may be a conse- sion reagent (Dako Co.) for visualization. The results of quence of inhibited EGFR/ERK signaling. immunostaining were categorized as follows: staining in less than 10% of the tumor cells was scored as 0; staining Effect on the proliferation and invasion by inhibiting in more than 10% of the tumor cells was scored as 1þ; protease activity of ADAM9 weak to moderate staining in more than 10% of the ADAM9 is a proteolytically active enzyme and the tumor cells was scored as 2þ; and strong staining in protease activity is currently the well-defined function of more than 10% of the tumor cells was scored as 3þ. ADAMs, with most of the putative substrates being transmembrane proteins. Therefore, we focused on the pro- Statistical analysis teolytic domain of ADAM9 as a biomarker and therapeu- Quantitative data were represented as the mean tic target using RAV18, an ADAM9-specific blocking standard deviation (SD) of at least three independent antibody, for our studies. Dastric cancer cells with high experiments. Statistical comparison between groups was expression of ADAM9 treated with RAV-18 showed a done using the Student t test. Differences were regarded dose-dependent decrease in ADAM9 protease activity as statistically significant when the P < 0.05. (Fig. 3A; 68.7% 0.8% in SNU-638, 62.1% 2.6% in MKN-28, and 50.3% 2.1% in MKN-74 at 20 mg/mL RAV-18; P < 0.05); however, there was no decrease in Results ADAM9 protease activity in cells with low expression of Screening of ADAM9 expression and protease ADAM9. activity in the gastric cancer cell panel Next, we investigated the effect of RAV-18 on cell We first carried out immunohistochemistry on par- proliferation. For SNU-638 cells, whose baseline doubling affin-embedded tumor sections from 10 patients with time was 22.6 0.1 hours, after treatment with 20 mg/mL gastric cancer with tumor infiltrating beyond subserosa RAV-18, doubling time was prolonged to 32.3 0.5 hours (T3). Four of 10 (40%) cancers expressed ADAM9, (P < 0.01). Similarly, proliferation was significantly whereas no expression was found in adjacent noncan- delayed with other cells (YCC-1: 25.4 0.2 vs. 33.9 cerous tissue. ADAM9 expression was highest in cells 0.4 hours; MKN-28: 27.4 0.1 vs. 36.3 0.5 hours; MKN- along infiltrating margins bordering noncancerous epi- 74: 27.7 0.2 vs. 39.3 0.8 hours; P < 0.05). However, thelium, and was located at the membrane and in the RAV18 did not affect proliferation in cells with low cytoplasm (Fig. 1A). ADAM9 expression (Fig. 3B). Then, we screened for ADAM9 expression in 24 We investigated the effect of RAV-18 on cell adhe- gastric cancer cell lines. The mRNA levels varied (Sup- sion. At a dose of 20 mg/mL RAV-18, adhesion to plementary Fig. S1) and did not correlate with ADAM9- Matrigel was unchanged in any of cells tested (Supple- protein expression in the cell lines (R2 ¼ 0.106; P ¼ mentaryFig.S3).However,highADAM9-expressing 0.203). In immunoblots, ADAM9 was detected as two cells demonstrated reduced invasion by 60% in SNU- bands: a proform (110 kDa) and mature form (84 638, 47% in YCC-1, 31% in MKN-74, and 51% in MKN- kDa; Fig. 1B; ref. 14). For further experiments, we 28 (Fig. 3C; P < 0.05); however, there was no effect of selected four cells with high ADAM9 protein expression RAV-18 in the low ADAM9-expressing cells. These and protease activity (SNU-638, YCC-1, MKN-74, and findings can transfer that ADAM9 regulates cancer MKN-28) and three cells with low ADAM9 protein proliferation and invasion, which is dependent on its expression and protease activity (YCC-6, YCC-7, and protease activity. Hs746T) as being outside of 1-SD of the mean value of Phospho-EGFR and phospho-ERK were decreased in protease activity tested (Fig. 1C). Protease activity and cells that responded to RAV18, which accords to the protein expression of the mature form were correlated results of siADAM9 transfection. There was no effect of each other (Fig. 1D; R2 ¼ 0.493; P < 0.01). RAV-18 on phospho-EGFR and phospho-ERK signaling

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3077

Kim et al.

×400 ×1,000 B

YCC-1 YCC-2 YCC-3 YCC-16 YCC-7 YCC-9 YCC-10 YCC-11 YCC-6 AGS NCI-N87 KATO III SNU-1 SNU-5 SNU-16 SNU-216 SNU-484 SNU-638 SNU-668 MKN-1 MKN-28 MKN-45 MKN-74 Hs746T HeLa IB:

P, 110 kDa M, 84 kDa ADAM9

36 kDa GAPDH

C D 2.5 1,000 R 2 = 0.49, P < 0.01

2 800

1.5 600

400 1

200 0.5 Fluorescence, arbitrary units Fluorescence, arbitrary 0

III 0 Expression (mature form/GAPDH density) AGS

HeLa 0 200 400 600 800 1,000 YCC-9 YCC-1 YCC-3 YCC-2 YCC-6 YCC-7 SNU-1 SNU-5 MKN-1 YCC-11 YCC-16 YCC-10 SNU-16 MKN-74 MKN-28 MKN-45 NCI-N87 KATO HS-746T SNU-216 SNU-668 SNU-638 SNU-484 Protease activity (arbitary unit) Cell lines

Figure 1. ADAM9 protease activity and expression in gastric cancer. A, ADAM9 expression is shown in the cell membrane and cytoplasm in gastric cancer tissues. B, protein levels of ADAM9 varied in gastric cancer cell lines(P,proform;M,matureform).HeLa cell lysates were used in positive control. C, ADAM9 protease activities varied in gastric cancer cell lines. D, ADAM9 protease activities were correlated with protein expressions of ADAM9 mature form in gastric cancer cell lines (R2 ¼ 0.49; P < 0.01). High and low activities were distinguished as being outside one SD of the mean of all cell lines' protease activity.

in the low-ADAM9 cells that did not respond to RAV-18 treatment significantly reduced the invasiveness in hyp- (Fig. 3D). oxia (by 30% in YCC-7 and 14% in Hs746T at 20 mg/mL RAV-18; P < 0.05; Fig. 4B). Effect of hypoxia on ADAM9 protease activity and When these cells were incubated under hypoxic con- cancer invasion ditions, phospho- and total EGFR and ERK expressions Our hypothesis is that the effect of RAV-18 depends were, as expected, induced. In addition, treatment of on its inhibition of ADAM9 proteolytic activity and if so, RAV-18 abrogated the EGFR and ERK phosphorylation cells with low ADAM9 should respond to RAV-18 if (Fig. 4C). Taken together, these findings suggest that ADAM9 is induced. We investigated the influence of hypoxia, a feature present in many solid tumors, includ- hypoxia on the constitutively low ADAM9-expressing ing gastric cancer, increases ADAM9 expression and cells and found that exposure to 1% O2 for 24 hours activity, together with an increased invasion, that may increased ADAM9 protease activity in YCC-6, YCC-7, contribute to the growth and spread of gastric cancer. and Hs746T cells, and this was returned to nonhypoxia levels by RAV-18 treatment in a dose-dependent man- Relationship with ADAM9 and hypoxia ner (Fig. 4A). To obtain further information on the relationship We also found that the invasiveness of cells with low between ADAM9 and hypoxia, we treated low ADAM9 ADAM9 cells was increased under hypoxic conditions (by cells with HIF1a siRNA under hypoxic condition. 24% in YCC-7 and 13% in Hs746T; P < 0.05). RAV-18 ADAM9 protease activity was significantly decreased

3078 Mol Cancer Ther; 13(12) December 2014 Molecular Cancer Therapeutics

ADAM9 in Gastric Cancer

ABSNU-638 MKN-28 120

C N R C N R IB: 100 P; 110 kDa M; 84 kDa ADAM9 80 173 kDa pEGFR Control * 60 si_Scramble 173 kDa EGFR si_ADAM9 * 40

44/42 kDa pERK Protease activity (%)

20 44/42 kDa ERK

0 36kDa GAPDH SNU-638 MKN-28

C Control si_Scramble si_ADAM9 120

100 SNU-638 80 Control 60 si_Scramble * si_ADAM9 40 * Invasiveness (%) 20 MKN-28

0 SNU-638 MKN-28 (*, P < 0.01)

Figure 2. Effects of ADAM9 expression modulation by transient ADAM9 inhibition to protease activity and cell invasiveness. A, when siADAM9 was transfected to high ADAM9 cell lines, SNU-638 and MKN-28, ADAM9 expression was decreased. Phospho-EGFR and phospho-ERK expression also decreased when siADAM9 was transfected to high ADAM9 cell lines (C, control; N, scrambled siRNA; and R, siADAM9). B, protease activity was signiﬁcantly decreased in siADAM9-transfected cells (62.9% 3.9%; SNU-638 and 43.7% 0.5%; MKN-28 at siADAM9 transfection; P < 0.001). C, when high ADAM9 cells were transfected with siRNA to ADAM9, invasiveness decreased (71% 2.2%; SNU-638 and 53% 3.9%; MKN-28 at siADAM9 transfection; P < 0.01). when HIF1a was suppressed, and on double siRNA xenograft study using BALB/c nu/nu mice. Mice were inhibition of HIF1a and ADAM9, there was an additional inoculated s.c. with MKN-28 (high ADAM9 protease effect in suppressing ADAM9 activity to normoxic levels activity) cells. At 3 weeks after inoculating the cells, when (Fig. 5A; P < 0.05). Under hypoxic conditions, cell prolif- the tumor volumes reached around 100 mm3, mice were eration was also increased and this effect was prevented split into two groups and treated with either saline or with the treatment RAV-18 and HIF1a siRNA (Fig. 5B). 50 mg/kg RAV-18 intraperitoneally for 2 weeks. The In immunoblots, we detected that ADAM9 and HIF1a RAV-18–treated group showed suppressed tumor growth expression in the cells were increased under hypoxic compared with the saline-treated group (mean tumor condition. When HIF1a siRNA were transfected, ADAM9 volume: 692 219 vs. 1,568 557 mm3; P < 0.01; Fig. 6A). expression and EGFR and ERK activity was decreased, Subsequently, MKN-28 cells were inoculated intra- while AKT activity was unchanged. On double inhibition peritoneally into mice. After 3 weeks, saline and RAV- of HIF1a and ADAM9, these effects were pronounced, 18 were administered as above. At 5 weeks of treatment, while HIF1a expression was not different compared with tumor nodules were signiﬁcantly smaller in the RAV-18 HIF1a siRNA alone (Fig. 5C). Thus, hypoxia and HIF1a group compared with the control group; the average induction may be a factor that regulates the ADAM9 tumor nodules in RAV-18–treated group were 58.3 expression and its downstream signaling that lead to 17.8, while those of the control were 96.3 19.6, indi- cancer progression. cating a 40% reduction of tumor formation (Supplemen- tary Fig. S4; P ¼ 0.0001). Immunoblot showed a decrease In vivo experiment of RAV-18 for therapeutic effect of phospho-EGFR and phospho-ERK expression fol- To investigate whether ADAM9 protease activity was lowing RAV-18 treatment as had been seen in vitro related to cancer progression in vivo, we conducted a studies. The average p-EGFR histoscore of the control

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3079

Downloaded from mct.aacrjournals.org on September 28, 2021. © 2014 American Association for Cancer Research. 3080 o acrTe;1(2 eebr2014 December 13(12) Ther; Cancer Mol rtaeatvt hwdn hnei rtaeatvte fo C- oH76) ,cl ie ihhg AD high with lines cell ADAM9 B, low Hs746T). with to lines prolife cell YCC-6 the However, However, (from MKN-28). proliferation. activities to of SNU-638 protease inhibition (from in seen dose-dependent change was showed activity no protease showed in activity decrease dose-dependent protease a RAV-18, with treated were ,we ellnswr rae ihRV1,pGR n EK expressi pERK, and pEGFR, lines RAV-18, cell with ADAM9-expressing treated low were lines cell when D, iue3. Figure al. et Kim oedpnetyrdcdivsvns ncel in invasiveness reduced dose-dependently Downloaded from C B A Protease activity (%)

RAV-18 conc. (mg/mL) 100 120 140 20 40 60 80 0 hne nAA9poes ciiis nain n rlfrto wit proliferation and invasion, activities, protease ADAM9 in Changes 0.2 20 2 0 N-3 C- K-4MN2 C- C- HS-746T YCC-7 YCC-6 MKN-28 MKN-74 YCC-1 SNU-638 No. of cells SNU-638 YCC-1 MKN-74 MKN-28 YCC-7 Hs746T SNU-638 YCC-7 MKN-28 MKN-74 YCC-1 * * 100 150 200 250 10 20 30 40 50 60 70 50 0 0 24561234561234561234 123456123456 Published OnlineFirstOctober24,2014;DOI:10.1158/1535-7163.MCT-13-1001 mct.aacrjournals.org * * High ADAM-9 YCC-6 SNU-638 * * Cell lines * i o eosrt change. a demonstrate not did * * * * * * * ie ihhg DM ciiy u i o in not did but activity, ADAM9 high with lines l 100 120 140 100 120 140 160 20 40 60 80 20 40 60 80 0 0 on September 28, 2021. © 2014American Association for Cancer Research. (*, < 0.05) P < Low ADAM-9Low YCC-7 YCC-1 * 20 μ 2 μ 0.2 μ 0 μ g/mL g/mL g/mL g/mL * * * aino ellnswt o DM act ADAM9 low with lines cell of ration (42/44 kDa) (42/44 kDa) (42/44 nwsds-eednl erae ncl ie xrsighg ADAM9; high expressing lines cell in decreased dose-dependently was on D (170 kDa) (170 kDa) * (36 kDa) Days * * GAPDH pEGFR EGFR pERK ERK ramn fRV1.A hnclswt ihAA9poes activity protease ADAM9 high with cells when A, RAV-18. of treatment h 100 120 140 20 40 60 80 10 20 30 40 50 60 70 0 0 123456 SNU-638 Invasiveness (%) 0 0.2 220 100 120 140 20 40 60 80 0 N-3 C- K-4MN2 C- HS-746T YCC-7 MKN-28 MKN-74 YCC-1 SNU-638 MKN-74 0 0.2 220 High ADAM9 * HS-746T YCC-1 * ﬂ * ec naieeso o ciiycls( cells activity low of invasiveness uence * 0 0.2 220 * MKN-28 * * * * * Cell lines * * 0 0.2 220 MKN-74 oeua acrTherapeutics Cancer Molecular vt a o niie.C A-8al RAV-18 C, inhibited. not was ivity 100 120 M ciiytetdwt RAV-18 with treated activity AM9 20 40 60 80 0 123456 * * 0 0.2 220 YCC-6 MKN-28 Low ADAM9 (*, P<0.05) 20 μ 2 μ 0.2 μ 0 μ 0 0.2 220 * YCC-7 g/mL g/mL g/mL g/mL * * 0 0.2 220 Hs746T * * 20 μ 2 μ 0.2 μ 0 μ , P g/mL g/mL g/mL (µg/mL) < RAV-18 g/mL 0.05). so Published OnlineFirst October 24, 2014; DOI: 10.1158/1535-7163.MCT-13-1001

ADAM9 in Gastric Cancer

AB* * * * 400 * * 200 * * 180 350 * * * 160 300 * * 140 250 Control 120 200 NMX μ HPX + 0 g/mL 100 HPX 150 HPX + 0.2 μg/mL HPX + 0.2 μg/mL μ μ (arbitrary unit) HPX + 2 g/mL 80 HPX + 2 g/mL 100 μ HPX + 20 μg/mL HPX + 20 g/mL Invasiveness (%) 60 50 ADAM-9 protease activity 40 0 YCC-6 YCC-7 HS-746T 20 Cell lines 0 YCC-7 HS-746T Cell lines C YCC-6 YCC-7 Hs746T

Hypoxia –++++ –++++ –++++ RAV-18 (mg/mL) 0 0 0.2 2 20 0 0 0.2 2 20 0 0 0.2 2 20 pEGFR (170 kDa) EGFR (170 kDa)

pERK (42/44 kDa)

ERK (42/44 kDa)

α-Tubulin (52 kDa)

Figure 4. Modulation of ADAM9 protease activity, invasion, and downward signaling in gastric cancer cells under hypoxic condition. A, ADAM9 protease activities were increased in hypoxic conditions. When treated with RAV-18, ADAM9 protease activities were dose-dependently decreased. B, invasiveness also increased in hypoxia and was decreased following RAV-18 treatment (, P < 0.05). C, after exposure to hypoxia, EGFR and ERK expression was increased in all low ADAM9 cell lines. In addition, phospho-EGFR and phospho-ERK expressions were dose-dependently decreased when cells received RAV-18 treatment. group was 188.1 68.4, while that of the RAV-18 group Taken together, these findings suggest that inhibition of was 98.6 34.9 (P ¼ 0.018). The average p-ERK histo- ADAM9 activity is a rational therapeutic approach to score was 139.3 35.7 in the control group and 63.4 suppress gastric cancer progression and metastasis. 21.2 in the RAV-18 group (P ¼ 0.002; Fig. 6B and C). There has been an earlier report demonstrating that These findings suggest that ADAM9 protease activity ADAM9 transcription is increased in gastric cancer (12). regulates gastric cancer tumorigenesis, including in a Although little is known about the regulatory mecha- model for intraperitoneal tumor spread, and may be a nism of ADAM9, some recent evidences suggest that a therapeutic target in gastric cancer. role for posttranslational modification is more important. In one prostate cancer model, induction of ADAM9 is regulated by stress through the accumulation of reac- Discussion tive oxygen species (ROS) as a common mediator, In this study, we have demonstrated that ADAM9 is and removing ROS, or the use of antioxidants resulted expressed in gastric cancer cells, which is related to cancer in a marked reduction of ADAM9 (11). These findings proliferation and invasion. In addition, using the protease suggest a link between stress-induced signaling and domain-specific blocking antibody RAV-18, we showed ADAM9, with ROS serving as a common mediator that this effect depends on the protease activity of (15). We have shown that another stress factor, hypoxia, ADAM9. The level of the ADAM9 enzymatic activity in a common feature found in many solid tumors, leads to cells with low ADAM9 expression is induced by hypoxia, an increase in the expression of ADAM9 protein along- and inhibited by RAV-18. We further showed that side enzymatic activity in HIF1-dependent way. We ADAM9 protease activity regulates gastric cancer tumor- found that posttranslational modification and proces- igenesis, including intraperitoneal metastatic spread. sing to a "mature" cleaved (84 kDa) form, is correlated

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3081

Kim et al.

* * A 400 * * * * 350 * * 300 * 250 NMX Figure 5. HIF1a regulates the HPX ADAM9 expression under hypoxic 200 condition. A, hypoxia-dependent HPX + negative siRNA 150 ADAM9 protease activity was HPX + siHIF (arbitrary unit) decreased when cells were treated 100 HPX + siHIF + RAV with HIF1a siRNA under hypoxic

ADAM9 protease activity 50 condition. HIF1a siRNA and RAV- 18 combined to inhibit ADAM9 0 protease activity more than HIF1a YCC-6 YCC-7 HS-746T siRNA treatment alone (, P < 0.05). Cell lines NMX, normoxia; HPX, hypoxia; B siHIF, siRNA against HIF1a; and 0.8 RAV, RAV-18. B, cell proliferation 1 YCC-6 YCC-7 0.9 Hs746T 0.9 0.7 0.8 was increased in low ADAM9 cells 0.8 0.7 * under hypoxic condition and * 0.6 * 0.7 * *0.6 N-C decreased by HIF1a siRNA 0.6 0.5 * H-C *0.5 transfection of cells ( , P < 0.05). N- 0.5 0.4 H-N 0.4 C, control on normoxia; H-C, 0.4 0.3 H-H 0.3 control on hypoxia; H-N, negative 0.3 0.2 H-R siRNA on hypoxia; H-H, siRNA 0.2 0.2 H-H-R a 0.1 0.1 0.1 against HIF1 on hypoxia; H-R, Absorbance (at 570 nm) 0 0 0 RAV-18 on hypoxia; and H-H-R, 123412 3 41234 combination of siHIF1a and RAV- Days 18 on hypoxia. C, increased expression of HIF1a as well as C YCC-6 YCC-7 Hs746T ADAM9 and its downstream - - N H H - - N H H - - N H H signaling was inhibited by HIF1a RAV-18 (20 mg/mL) - - - - + - - - - + - - - - + siRNA transfection. When HIF1a Hypoxia - + + + + - + + + + - + + + + siRNA and RAV-18 were used in HIF1α combination, HIF1a expression 1 3.6 2.7 0.7 0.7 1 5.8 5.3 1.3 1.2 1 2.6 2.6 0.7 0.7 was not changed compared with ADAM9 HIF1a siRNA treatment alone. 1 1.8 0.7 0.2 0.2 1 4.1 4.1 1.2 1.1 1 7.8 8.4 7.5 7.6 However, phospho-EGFR and pEGFR phospho-ERK expression was 1 2.3 2.1 1.5 0.4 1 3.7 2.4 1.9 1.2 1 3.1 3.0 2.3 1.1 decreased more than HIF1a siRNA EGFR treatment alone. Phospho-AKT 1 1.8 1.9 2.0 2.0 1 2.9 3.0 2.7 2.8 1 1.2 1.2 1.2 1.2 expression was not changed when pAKT HIF1a siRNA, with or without RAV- 1 1.6 1.7 1.7 1.5 1 1.4 1.4 1.3 1.3 1 1.4 1.5 1.3 1.3 18, was used. N. negative siRNA; AKT H, siRNA against HIF1a. 1 1.3 1.3 1.3 1.3 1 1.1 1.1 1.1 1.0 1 1.2 1.1 1.2 1.2 pERK 1 1.9 1.9 0.4 0.2 1 3.8 3.9 1.0 0.0 1 1.4 1.3 1.2 1.1 ERK 1 7.7 7.8 7.8 5.9 1 2.2 1.7 1.8 1.6 1 37.6 41.9 41.9 37.2 α-Tubulin

with its enzymatic activity. Although all ADAMs pos- ADAM, and (ii) when cotreated with RAV-18, additive sess the matrix metalloproteinase (MMP)–like domain effect is prominent only in high ADAM9-expressing adjacent to the prodomain, only 50% exhibit protease cells, implying that RAV-18 works specifically to ADAM9. activity. Currently, protease activity is the best-defined Several explanations are possible about how ADAM9 function of ADAMs, with most putative substrates cur- plays a role in cancer progression. There is evidence that rently identified as being transmembrane proteins (16). ADAM9 mediates the release of growth factor ligands, One of the limitations of our study is the substrate which regulate EGFR signaling. In the stomach, trans- we used for ADAM9 protease activity is not only specific activation of the EGFR by peptides, such as gastrin, to ADAM9, but also reactive to other ADAMs—especially angiotensin II, bradykinin, bombesin, or substance P, ADAM10 and ADAM17 (17–21). As in Supplementary is mediated by shedding of EGFR ligands by ADAMs Fig. S5, (i) changes of protease activity after siRNA knock- (22, 23). ADAM9 also releases some transmembrane down is dependent on basal expression level of each protein-derived ligands, such as heparin-binding EGF,

3082 Mol Cancer Ther; 13(12) December 2014 Molecular Cancer Therapeutics

ADAM9 in Gastric Cancer

A 2,100

Control ) 3 1,800 1,500 * 1,200 Control RAV-18 900 RAV-18 treat. 600

Tumor volume (mm 300

0 D1 D4 D7 D11 D14 D18 D21 D25 D28 D32 D35 B Days after RAV-18 treatment (*, P < 0.01) Control RAV-18 treat. C H&E 300 P = 0.018 Control RAV-18 treat. C Control C RAV-18 treat. 250 pEGFR P = 0.002 (170 kDa) ADAM9 200

EGFR 150 (170 kDa) Histoscore 100 pERK pEGFR (42/44 kDa)

50 ERK (42/44 kDa)

0 GAPDH pERK p-EGFR p-ERK (36 kDa)

Figure 6. In vivo xenograft model of ADAM9. A, photographs of tumors from control and RAV-18–treated groups. Tumor volumes increased in both control and RAV-18–treated groups; however, the tumor volumes of the control group were larger than in the RAV-18–treated group (P < 0.01). B, representative phospho-EGFR and phospho-ERK immunostaining (original images were captured at 400 magniﬁcation) and H&E staining. Average of phospho-EGFR histoscores in the control group was approximately two times higher than in the RAV-18–treated group (P ¼ 0.018). Furthermore, the average phospho-ERK histoscores in controls was approximately three times higher than in RAV-18–treated groups (P ¼ 0.002). C, Western blot analysis showed that phospho-EGFR and phospho-ERK expressions were lower in the RAV-18–treated group. C, MKN-28 cell whole lysate. amphiregulin, TGFa,andTNFa (24, 25). We herein downstream signaling was not altered by HIF1a knock- found that ADAM9 knockdown blocks EGFR phos- down or RAV-18 treatment. It is possible that Hs746T phorylation and downstream ERK activation, and that is more dependent on other ADAM proteins—such as RAV-18 inhibited tumor growth, in line with EGFR and ADAM10 and ADMA17—rather than ADAM9 for its ERK phosphorylation in vivo. Thus, inhibition of EGFR biologic activity, as shown in Supplementary Fig. S2. ligand shedding by ADAM9 with subsequent decreased We have shown that proliferation and invasion is EGFR activation and downstream signaling could be a affected by RAV18, but not adhesion. The disintegrin mechanism for the antitumor activity of RAV-18. domain is found in all ADAMs and is located down- It is intriguing that hypoxia induces ADAM9 activity, stream of the metalloproteinase domain. We believe relating to cancer cell proliferation and invasion, which that adhesion proteins are important for cellular adhe- is reversed by RAV-18 although not beyond the levels in sion, migration, and signal transduction (26). However, normoxia. This does not appear to be related to the most information on the binding of disintegrin to integ- processing of ADAM9 to the cleaved form as this was rin comes from in vitro studies using recombinant dis- detected in both low and high ADAM9-expressing cell integrin domains (16). Thus, the biologic relevance of lines. We speculate that there are two forms of ADAM9, our ﬁnding that RAV-18 does not alter adhesion is one not responsive to inhibition by RAV-18, which is unclear at this time. perhaps intracellular located or masked, and another Our in vivo studies demonstrate that ADAM9-targeting induced by hypoxia that is accessible to inhibition by antibody is a reasonable option for treatment of gastric RAV-18, that is responsible for the growth stimulatory cancer. One limitation is lack of information on the actual and invasive effects of ADAM9. Another point is that incidence of ADAM9 expression in large-scale clinical ADAM9 was induced by hypoxia in Hs746T cells but its samples. Our results suggest that only high-level

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3083

Kim et al.

ADAM9-expressing tumors, and those where ADAM9 is Analysis and interpretation of data (e.g., statistical analysis, bio- statistics, computational analysis): J.M. Kim, H.-C. Jeung, S.Y. Rha, induced in response to hypoxia, will be responsive to T.S. Kim, Y.K. Shin, X. Zhang, H.C. Chung, G. Powis ADAM9 antibody. The subcellular location of ADAM9 Writing, review, and/or revision of the manuscript: J.M. Kim, H.-C. Jeung, G. Powis could also play a role in the response to antibody Administrative, technical, or material support (i.e., reporting or orga- treatment. nizing data, constructing databases): J.M. Kim, H.-C. Jeung, S.Y. Rha Collectively, our work shows that ADAM9 plays an Study supervision: H.-C. Jeung, S.Y. Rha, S.W. Park, H.C. Chung important role in gastric cancer proliferation and invasion, and that while expressed at high levels in some cancer cells that are responsive to functional inhibition Acknowledgments and antitumor activity of a catalytic site–directed anti- The authors thank Macrogenics Incorporation for generously support- ing the ADAM9 inhibitor RAV-18. body, other gastric cancer cells have low levels of expression and only when exposed to hypoxia do ADAM9 levels increase and the cells become responsive to anti-ADAM9 treatment. These finding suggest that ADAM9 could be Grant Support For this study, H.C. Jeung received a grant from the National R&D an effective therapeutic target for therapy of metastatic Program for Cancer Control, Ministry of Health & Welfare, Republic gastric cancer. of Korea (1420060). G. Powis was supported by grants CA163541 and CA172670 from the National Cancer Institute. J.M. Kim, H.C. Jeung, fl S.Y. Rha, T.S. Kim, K.H. Park, and H.C. Chung received the Public Welfare Disclosure of Potential Con icts of Interest and Safety Research Program through the National Research Foundation No potential conflicts of interest were disclosed. of Korea (NRF) funded by the Ministry of Education, Science and Tech- nology (2010-0020841). Authors' Contributions The costs of publication of this article were defrayed in part by the Conception and design: J.M. Kim, H.-C. Jeung, S.Y. Rha, E.J. Yu, K.H. Park, payment of page charges. This article must therefore be hereby marked H.C. Chung advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate Development of methodology: J.M. Kim, T.S. Kim, K.H. Park this fact. Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.M.Kim,H.-C.Jeung,S.Y.Rha, Received December 13, 2013; revised September 16, 2014; accepted Y.K. Shin, K.H. Park September 21, 2014; published OnlineFirst October 24, 2014.

References 1. Albelda SM. Role of integrins and other cell adhesion molecules in 13. Cisse MA, Gandreuil C, Hernandez J, Martinez J, Checler F, Vincent B. tumor progression and metastasis. Lab Invest 1993;68:4–17. Design and characterization of a novel cellular prion-derived quenched 2. Nicolson GL. Tumor and host molecules important in the organ ﬂuorimetric substrate of alpha-secretase. Biochem Biophys Res Com- preference of metastasis. Semin Cancer Biol 1991;2:143–54. mun 2006;347:254–60. 3. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100: 14. Roghani M, Becherer JD, Moss ML, Atherton RE, Erdjument-Bromage 57–70. H, Arribas J, et al. Metalloprotease–disintegrin MDC9: intracellular 4. Becherer JD, Blobel CP. Biochemical properties and functions of maturation and catalytic activity. J Biol Chem 1999;274:3531–40. membrane-anchored metalloprotease-disintegrin proteins (ADAMs). 15. Shigemura K, Sung SY, Kubo H, Arnold RS, Fujisawa M, Gotoh A, et al. Curr Top Dev Biol 2003;54:101–23. Reactive oxygen species mediate androgen receptor- and serum 5. Grutzmann€ R, Luttges€ J, Sipos B, Ammerpohl O, Dobrowolski F, starvation-elicited downstream signaling of ADAM9 expression in Alldinger I, et al. ADAM9 expression in pancreatic cancer is associated human prostate cancer cells. Prostate 2007;67:722–31. with tumour type and is a prognostic factor in ductal adenocarcinoma. 16. Duffy MJ, Mullooly M, O'Donovan N, Sukor S, Crown J, Pierce A, et al. Br J Cancer 2004;90:1053–8. The ADAMs family of proteases: new biomarkers and therapeutic 6. Hirao T, Nanba D, Tanaka M, Ishiguro H, Kinugasa Y, Doki Y, et al. targets for cancer? Clin Proteomics 2011;8:9. Overexpression of ADAM9 enhances growth factor-mediated recy- 17. Canault M, Leroyer AS, Peiretti F, LesAche~ G, Tedgui A, Bonardo B, cling of E-cadherin in human colon cancer cell line HT29 cells. Exp Cell et al. Microparticles of human atherosclerotic plaques enhance the Res 2006;312:331–9. shedding of the tumor necrosis factor-alpha converting enzyme/ 7. Lendeckel U, Kohl J, Arndt M, Carl McGrath S, Donat H, Rocken€ C. ADAM17 substrates, tumor necrosis factor and tumor necrosis factor Increased expression of ADAM family members in human breast receptor-1. Am J Pathol 2007;171:1713–23. cancer and breast cancer cell lines. J Cancer Res Clin Oncol 18. Reiss K, Cornelsen I, Husmann M, Gimpl G, Bhakdi S. Unsaturated 2005;131:41–8. fatty acids drive disintegrin and metalloproteinase (ADAM)-dependent 8. Yamada D, Ohuchida K, Mizumoto K, Ohhashi S, Yu J, Egami T, et al. cell adhesion, proliferation, and migration by modulating membrane Increased expression of ADAM 9 and ADAM 15 mRNA in pancreatic ﬂuidity. J Biol Chem 2011;286:26931–42. cancer. Anticancer Res 2007;27:793–9. 19. Barsoum IB, Hamilton TK, Li X, Cotechini T, Miles EA, Siemens DR, 9. Edwards DR, Handsley M, Pennington CJ. The ADAM metalloprotei- et al. Hypoxia induces escape from innate immunity in cancer cells via nases. Mol Aspects Med 2008;29:258–89. increased expression of ADAM10: role of nitric oxide. Cancer Res 10. Fischer OM, Hart S, Gschwind A, Prenzel N, Ullrich A. Oxidative and 2011;71:7433–41. osmotic stress signaling in tumor cells is mediated by ADAM proteases 20. Guo J, He L, Yuan P, Wang P, Lu Y, Tong F, et al. ADAM10 over- and heparin-binding epidermal growth factor. Mol Cell Biol 2004;24: expression in human non-small cell lung cancer correlates with cell 5172–83. migration and invasion through the activation of the Notch1 signaling 11. Sung SY, Kubo H, Shigemura K, Arnold RS, Logani S, Wang R, et al. pathway. Oncol Rep 2012;28:1709–18. Oxidative stress induces ADAM9 protein expression in human prostate 21. Shou Z, Jin X, Zhao Z. Upregulated expression of ADAM17 is a cancer cells. Cancer Res 2006;66:9519–26. prognostic marker for patients with gastric cancer. Ann Surg 2012; 12. Carl-McGrath S, Lendeckel U, Ebert M, Roessner A, Rcken C. The 256:1014–22. disintegrin-metalloproteinases ADAM9, ADAM12, and ADAM15 are 22. Miyazaki Y, Shinomura Y, Tsutsui S, Zushi S, Higashimoto Y, upregulated in gastric cancer. Int J Oncol 2005;26:17–24. Kanayama S, et al. Gastrin induces heparin-binding epidermal growth

3084 Mol Cancer Ther; 13(12) December 2014 Molecular Cancer Therapeutics

ADAM9 in Gastric Cancer

factor-like growth factor in rat gastric epithelial cells transfected with PKCdelta are involved in TPA-induced ectodomain shedding of mem- gastrin receptor. Gastroenterology 1999;116:78–89. brane-anchored heparin-binding EGF-like growth factor. EMBO J 23. Park JW, Baek NS, Lee SC, Oh SJ, Jang SH, Kim IH, et al. Preclinical 1998;17:7260–72. efﬁcacy testing for stomach and liver cancers. Cancer Res Treat 25. Schfer B, Gschwind A, Ullrich A. Multiple G-protein–coupled receptor 2014;46:186–93. signals converge on the epidermal growth factor receptor to promote 24. Izumi Y, Hirata M, Hasuwa H, Iwamoto R, Umata T, Miyado K, et al. migration and invasion. Oncogene 2004;23:991–9. A metalloprotease-disintegrin, MDC9/meltrin-gamma/ADAM9 and 26. Stupack DG. The biology of integrins. Oncology 2007;21:6–12.

www.aacrjournals.org Mol Cancer Ther; 13(12) December 2014 3085

The Effect of Disintegrin−Metalloproteinase ADAM9 in Gastric Cancer Progression

Jeong Min Kim, Hei-Cheul Jeung, Sun Young Rha, et al.

Mol Cancer Ther 2014;13:3074-3085. Published OnlineFirst October 24, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-13-1001

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2014/10/24/1535-7163.MCT-13-1001.DC1

Cited articles This article cites 26 articles, 7 of which you can access for free at: http://mct.aacrjournals.org/content/13/12/3074.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/13/12/3074.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/13/12/3074. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.