Cancer Cell Biology
IQGAP2 inactivation through aberrant promoter methylation and promotion of invasion in gastric cancer cells
Article first published online: 23 OCT 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 122, Issue 5, pages 1040–1046, 1 March 2008
How to Cite
Jin, S.-H., Akiyama, Y., Fukamachi, H., Yanagihara, K., Akashi, T. and Yuasa, Y. (2008), IQGAP2 inactivation through aberrant promoter methylation and promotion of invasion in gastric cancer cells. Int. J. Cancer, 122: 1040–1046. doi: 10.1002/ijc.23181
- Issue published online: 24 DEC 2007
- Article first published online: 23 OCT 2007
- Manuscript Accepted: 30 AUG 2007
- Manuscript Received: 25 MAY 2007
- Ministry of Education, Culture, Sports, Science, and Technology, Japan. Grant Number: 17015013
- gastric cancer;
Invasion and metastases of cancer cells are the main causes of treatment failure in cancer. IQ motif-containing GTPase activating protein 1 (IQGAP1), plays pivotal roles in intercellular adhesion, migration, invasion and metastases in various cancer cells. However, the role of another family member, IQGAP2, in carcinogenesis remains unknown. Here, we investigated IQGAP2 functions in gastric cancers. We found that IQGAP2 protein expression was lost in 5 of the 9 gastric cancer cell lines. Through analysis by the methylation-specific PCR, aberrant IQGAP2 methylation was detected in 3 gastric cancer cell lines. IQGAP2 mRNA was found to be activated after 5-aza-2′-deoxycytidine treatment of the methylation-positive cells. Moreover, IQGAP2 methylation was detected in 28 of the 59 (47%) primary gastric cancer tissues, but not in 12 normal gastric mucosa samples. Immunohistochemical staining revealed that 7 of the 8 (88%) gastric cancer tissues without methylation signals displayed IQGAP2 expression, whereas among 10 with methylation signals none expressed IQGAP2 (p = 0.0002), indicating that IQGAP2 methylation is highly associated with loss of the IQGAP2 expression in the primary gastric cancer tissues as well as gastric cancer cell lines. Furthermore, IQGAP2 methylation was also associated with tumor invasion and a poor prognosis. IQGAP2 knockdown with small interfering RNA increased the invasive capacity of a gastric cancer cell line. These results suggest that silencing of IQGAP2 by promoter methylation may contribute to gastric cancer development. © 2007 Wiley-Liss, Inc.
According to WHO, gastric cancer is the fourth most common malignancy worldwide, ∼870,000 new cases occurring yearly.1 Uncontrollable tumor invasion and dissemination of cancer cells around the primary organ comprise the neoplastic process responsible for most deaths from cancer because of inadequacy of surgical removal.2 Invasive and metastatic cancer cells undergo numerous genetic and epigenetic changes, manifested as adhesion molecules, motility factors, growth factors, and expression of proteolytic enzymes that degrade the basement membrane and migrate into the surrounding extracellular matrices of various tissues, including the stomach.3, 4, 5, 6
Inactivation of gene expression through abnormal methylation of CpG islands can act as a “hit” for tumor generation.7, 8 It has been recognized that aberrant hypermethylation events can occur early in tumorigenesis, predisposing cells to malignant transformation. Given the importance of gene expression and invasion, the determination of their relationship seems to be essential for a better understanding of tumor biology and for the development of new treatment strategies. However, the relationship between the two is poorly characterized.
The Rho-family small GTPases, especially Rac1 and Cdc42, regulate actin cytoskeletal dynamics by interacting with a number of effectors, including the IQGAPs.9, 10 IQGAPs are so named because they possess IQ domains, which are tandem repeats of 4 IQ motifs (tandem isoleucine and glutamine residues), and Ras GTPase-activating protein (GAP)-related domains.10 The IQGAP gene family consists of 3 members, IQGAP1, IQGAP2, and IQGAP3.11 Mammalian IQGAP1 is considered to be a scaffolding protein at the crossroads of several signaling pathways involved in the control of cell adhesion,12, 13 polarization14, 15 and directional migration.15, 16 IQGAP1 promotes invasion in a breast cancer cell line.17 Among gastric cancers, diffuse type gastric cancers exhibit a higher frequency of invasion with dissemination to the peritoneum and lymph node metastasis compared to intestinal type gastric ones.18, 19IQGAP1 was upregulated by gene amplification in 2 diffuse type gastric cancer cell lines.20 Furthermore, there seems to be a correlation between dysfunction of E-cadherin-mediated adhesion and increased membrane localization of IQGAP1 in gastric cancer.21 Similarly, IQGAP1 expression increases in human colorectal cancers, particularly at the invasion front.22 This observation raised the possibility that IQGAP1 promotes invasiveness of gastrointestinal tract cancer cells.
A comparison of the structure and functions of IQGAP2 with those of IQGAP1 revealed some similarities and several differences. The human IQGAP2 protein exhibits 59% identity to IQGAP1, and their domain structures are well conserved.11, 23 Although IQGAP2 expression is widely distributed in various tissues and cellular processes,24, 25, 26 the roles of IQGAP2 in gastric carcinogenesis remain unclear. In this study, we found that IQGAP2 was frequently inactivated in gastric cancers and that the major mechanism was aberrant methylation of IQGAP2. Furthermore, we showed that inactivation of IQGAP2 was correlated with invasion of gastric cancer cells and a poor prognosis.
Material and methods
Cell lines and tissue samples
The human gastric cancer cell lines, except for HSC-44, HSC-39 and HSC-59,27i.e., MKN28, NUGC-4, MKN7, MKN45, TGBC11TKB and KATOIII, were obtained from the Human Science Research Resources Bank (Osaka, Japan) or Riken Cell Bank (Tsukuba, Japan). Surgically resected specimens from 59 primary gastric cancer patients were randomly obtained from the Affiliated Hospital of School of Medicine, Tokyo Medical and Dental University. Informed consent was obtained from all patients, and the study was approved by the appropriate institutional review committee. Histological classification was performed according to the general rules established by the Japanese Gastric Cancer Association.28
The gastric cancer cell lines were grown in Dulbecco's modified Eagle's medium or RPMI1640 supplemented with 10% fetal bovine serum (FBS) and kanamycin (50 μg/ml). For demethylation studies, cells were daily treated with 5 μM 5-aza-2′ deoxycytidine (DAC; Sigma, St Louis, MO) for 48 hr.29
Western blot analysis
Cultured cells were lysed in SDS sample buffer. Proteins were run on a 8% SDS-PAGE gel and then transferred to a nitrocellulose membrane, which was subsequently blotted using anti-IQGAP2 mouse monoclonal antibodies (1:1,000) (Upstate Biochemicals, Lake Placid, NY) and revealed with adequate secondary antibodies coupled to peroxidase. Blots were visualized with a Immun-Star™ Chemiluminescent Protein Detection System according to the manufacturer's instructions (Bio-Rad Laboratories, Hercules, CA).
Immunofluorescence cytochemistry and immunohistochemistry
Cells were fixed with 2% formaldehyde in phosphate-buffered saline (PBS) for 10 min at room temperature and then treated with PBS containing 0.1% Triton X-100. After blocking with 1% bovine serum albumin for 60 min at room temperature, the cells were incubated with anti-IQGAP2 antibodies (1:50) at 4°C overnight. After incubation with Alexa Fluor 594 anti-mouse IgG antibodies (Molecular Probes, Eugene, OR), specimens were observed under a confocal laser scanning microscope (LSM510; Carl Zeiss, Oberkochene, Germany).
As for immunohistochemistry, samples were sectioned, deparaffinized, and then pretreated by microwaving in 10 mM citric acid buffer for 30 min to retrieve antigenicity. After the peroxidase activity had been blocked with 3% H2O2-methanol for 15 min, the sections were incubated with 10% normal goat serum in PBS to block nonspecific protein binding, followed by incubation with primary antibodies at 4°C overnight. Then, the sections were incubated with horseradish peroxidase-labeled goat anti-mouse antibodies (Dako, Carpinteria, CA) for 1 hr at room temperature, and the signal was amplified and visualized with diaminobenzidine-chromogen, followed by counterstaining with 0.1% hematoxylin. Expression was considered to be “positive” when 10% or more cancer cells were stained.
Reverse transcription-polymerase chain reaction (RT-PCR)
Total RNA was isolated with Trizol reagent (Invitrogen Corp., Carlsbad, CA). For RT-PCR, 2 μg of total RNA was reverse transcribed using a Superscript kit (Invitrogen). The thermal profile consisted of initial denaturation at 94°C for 4 min followed by cycles of 94°C for 1 min, 59°C for 1 min, and 72°C for 1 min, with a final extension step of 72°C for 10 min. We amplified the IQGAP2 gene from 50 ng of cDNA template with multiple cycles (24–35 cycles) to determine the appropriate conditions for observing semiquantitative differences in expression levels. Oligonucleotide primers were designed according to the published sequence for IQGAP2 (GenBank accession No. NM_006633), 5′-TGCATGAGAAAGGTG TCCTG-3′ (sense) and 5′-GGCACCTGGGTACATTCTTC-3′ (antisense). The PCR products were confirmed by sequencing.
Methylation-specific PCR (MSP)
Bisulfite-treated DNA was amplified with either a methylation-specific or unmethylated-specific primer set for IQGAP2. The primers for methylated IQGAP2 were 5′-GGAGTGGGTCGTA GATTTTCGGGC-3′ (sense) and 5′-CTACCCTCGCTAACC AAACTCGCG-3′ (antisense), and those specific for unmethylated IQGAP2 were 5′-AGGGAGTGGGTTGTAGATTTTTGGGT-3′ (sense) and 5′-CAACTACCCTCACTAACCAAACTCACA-3′ (antisense). The PCR reaction was performed for 35 cycles of 95°C for 30 sec, 58°C for 30 sec, and 72°C for 30 sec.
PCR amplification of bisulfite-treated DNA for sequencing
DNA extraction, bisulfite treatment and DNA sequencing were performed as described previously.30 PCR amplification was performed as follows: a 5-min 95°C incubation step followed by 40 cycles of 30 sec at 95°C, 30 sec at 53°C and 30 sec at 72°C. A 10-min elongation step at 72°C completed the PCR amplification program. The primer sequences were 5′-GTTATTTTTAGGTTAT TAGTTGTG-3′ (sense) and 5′-ACAACTCTTCRTATAACATC CTAC-3′ (antisense).
IQGAP2 siRNA transfection
For analyses with small interfering RNA (siRNA) for IQGAP2, RNA oligomers containing 21 nucleotides corresponding to human IQGAP2 at exons 27 and 28 (5′-GGAAUUCAG GAAAUAUUUC-3′) with dTdT overhangs at each 3′ terminus were synthesized (Ambion, Austin, TX). Scrambled siRNA (5′-UUCUCCGAACGUGUCACGU-3′) with dTdT overhangs at each 3′ terminus (Qiagen, Valencia, CA) was used as a nonspecific siRNA control. Transfection was performed using HiPerFect Transfection Reagent (Qiagen). The siRNA experiment was performed according to the manufacturer's protocol. MKN45 cells were incubated with 30 nM IQGAP2 siRNA or scrambled siRNA for at least 48 hr before biochemical experiments and/or functional assays were conducted.
The invasive potential of the tumor cells was examined using a BioCoat Matrigel Invasion Chamber kit (Becton Dickinson Biosciences, Bedford, MA) according to the manufacturer's instructions. Briefly, before seeding into the transwell inserts, cells were released using trypsin-EDTA and sequentially rinsed with RPMI1640 containing 10% FBS. The rinsed cells were resuspended in serum-free RPMI1640 (1 × 105/ml), and 500 μl of the cell suspension (5 × 104) was added to the transwell insert chamber with a filter that was coated with Matrigel. In the lower compartment, 750 μl RPMI1640 containing 10% FBS as a chemoattractant was added to 24-well culture dishes. As a control, an equal number of uncoated Matrigel companion inserts were seeded with cells in parallel. Both inserts were incubated at 37°C under a 5%CO2/95% air atmosphere for 24 hr. After incubation, the cells on the upper surface of the inserts were removed by gentle scraping with a sterile cotton swab. Invasive cells that migrated to the lower side of inserts were fixed with methanol and stained with 4% Giemsa's solution (Merck, Darmstadt, Germany), and then microscopically observed and counted in 10 random fields of view at 400× magnification. Data were expressed as the percentages of cells that invaded through the Matrigel matrix-coated membrane relative to the cells that migrated through the control membrane. All assays were performed in triplicate.
Differences in frequency as to methylation status were examined using Fisher's exact test, and differences in mean values were examined using Student's t test. Association between the methylation status and depth of tumor invasion was also analyzed using the Mann Whitney's U test. We further studied the association between DNA methylation of IQGAP2 and the clinicopathological characteristics by multiple regression analysis. Survival curves were obtained using the Kaplan-Meier product-limit method, and the significance of differences between the survival curves was determined using the log-rank test. Differences were considered significant at p < 0.05. The statistical software used was Statview 5.0 for Macintosh (SAS Institute, Cary, NC).
Expression of IQGAP2 in gastric cancer cell lines
We studied the expression of IQGAP2 in 9 gastric cancer cell lines by Western blot analysis. The anti-IQGAP2 antibody used recognizes IQGAP2, Mr 180 kDa, but not IQGAP1. The IQGAP2 protein was detected in the HSC-59, NUGC-4, MKN45 and TGBC11TKB cell lines, but not in the HSC-44, HSC-39, MKN28, MKN7 and KATOIII ones (Fig. 1a).
To determine where the IQGAP2 protein is localized in the gastric cancer cell lines, we performed immunofluorescence cytochemistry. Representative results of IQGAP2 immunostaining of cancer cell lines are shown in Figure 1b. IQGAP2 was observed at the cell membrane in gastric cancer cell lines. These results are consistent with those of Western blot analysis.
Methylation of IQGAP2 in gastric cancer cell lines
Silencing of genes through DNA hypermethylation could be the sole mechanism of gene inactivation, or cooperate with genetic mechanisms to inactivate putative tumor suppressor genes in tumors. Therefore, to determine whether IQGAP2 methylation results in loss of IQGAP2 expression, we used a demethylation agent, DAC, to study the epigenetic status of IQGAP2 in gastric cancer cell lines. The basally silent or weakly expressed IQGAP2 gene was upregulated with this treatment in 3 cell lines, HSC-44, MKN28 and KATOIII (Fig. 2).
MSP analyses revealed that the HSC-44 cell line, which did not basally express IQGAP2, exhibited a methylation signal, and the MKN28 and KATOIII cell lines expressing low mRNA levels exhibited methylated and unmethylated bands. On the contrary, the HSC-59, NUGC-4, MKN45 and TGBC11TKB cell lines, which expressed high IQGAP2 mRNA as well as IQGAP2 proteins, exhibited no signal for methylation. The HSC-39 and MKN7 cell lines exhibiting IQGAP2 mRNA but no protein did not show any methylation either (Fig. 2). Data representing the relationship between IQGAP2 expression and the IQGAP2 methylation status in gastric cancer cell lines are summarized in Table I.
|Cancer cell line||IQGAP2 expression||Methylation status3|
|Protein level1||mRNA level2|
Methylation of IQGAP2 in gastric carcinoma tissues
To determine whether IQGAP2 inactivation through methylation is characteristic of gastric cancers, we analyzed the methylation status of IQGAP2 by MSP in 59 gastric carcinoma specimens. Typical results are shown in Figure 3a. IQGAP2 in 12 normal gastric mucosa samples was completely methylation free, but methylation-positive patterns were observed in 47% (28 of 59) of gastric cancer tissues (p = 0.0022).
In representative samples, we verified the MSP results by sequencing (Fig. 3b). IQGAP2 expression-negative cultured cancer cells, HSC-44, MKN28 and KATOIII, and a methylation-positive primary gastric cancer sample, TG34Ca, showed dense methylation of the IQGAP2 CpG island. On the contrary, IQGAP2 expression-positive cells, MKN45, and two normal stomach mucosa samples, S800 and TG34N, showed little or no methylation within the examined regions.
We performed immunohistochemistry to analyze IQGAP2 protein expression in 20 normal gastric mucosae and 18 primary gastric carcinoma tissues. In normal gastric mucosae, IQGAP2 expression was observed in gastric glands, particularly in parietal cells (Fig. 4). Representative results of IQGAP2 immunostaining of gastric carcinomas are shown in Figure 4. Expression of IQGAP2 was lost in 11/18 (61%) gastric carcinoma cases, while IQGAP2 was detected in all 20 normal gastric mucosa cases.
To determine whether IQGAP2 methylation results in loss of IQGAP2 protein expression, we analyzed the association between IQGAP2 expression on immunohistochemistry and the methylation status of IQGAP2 in 18 gastric cancer cases. Among the 10 gastric cancer samples exhibiting methylation signals on MSP, none was positive for IQGAP2 protein expression, while 7 of the 8 (88%) without methylation signals exhibited IQGAP2 protein (p = 0.0002) (Table II).
|Expression1||IQGAP2 methylation status||p2|
The relationship between the methylation frequencies of IQGAP2 and clinicopathological factors
The clinicopathological factors of the patients with the IQGAP2 methylation status are shown in Table III. IQGAP2 methylation was associated with tumor size (p = 0.004), depth of tumor invasion (p < 0.0001), and lymph node metastasis (p = 0.006). However, other clinicopathological factors, such as age, sex and histological type, were not found to correlate with methylation of IQGAP2. On the basis of the results of these statistical tests, we conducted multiple regression analysis (Table III). As a result, only the depth of tumor invasion was found to be independently associated with methylation of IQGAP2 (p < 0.0001). These results suggest that methylation of IQGAP2 is closely correlated with the inactivation of IQGAP2 expression and tumor invasion in gastric cancer tissues.
|Clinicopathological factors||Methylation status||p||p1|
|(n = 28)||(n = 31)|
|Age (y), mean ± SD||62.8 ± 2.3||63.6 ± 1.9||0.7952||0.557|
|Size (cm), mean ± SD||7.8 ± 0.7||4.9 ± 0.6||0.0042||0.939|
|Depth of tumor invasion|
|Lymph node metastasis|
To determine whether IQGAP2 methylation is a significant prognostic factor for the survival of patients with surgically resected advanced gastric carcinomas, we performed a log-rank test with Kaplan-Meier estimates. Among the 36 patients analyzed, the 5-year survival rate was significantly lower in 23 patients with IQGAP2 methylation-positive gastric cancers than in those with 13 IQGAP2 methylation-negative ones (p = 0.029, Fig. 5). These results clearly indicate that IQGAP2 methylation is associated with the 5-year survival of gastric cancer cases and suggest that IQGAP2 inactivation is a valuable biomarker of the prognosis.
Suppression of IQGAP2 with siRNA increases cell invasion in IQGAP2-expressing gastric cancer cells
We examined whether the loss of IQGAP2 expression is involved in the invasion activity of gastric cancer cells. Among the 4 IQGAP2 protein expression-positive gastric cancer cell lines, MKN45 cells have reportedly been used for the detection of invasiveness.31, 32 To determine whether IQGAP2 influences the invasiveness of MKN45 cells, experiments were conducted in which it was knocked down using siRNA. The efficiencies of siRNA introduction that were determined by use of a fluorescein-labeled scrambled control siRNA were 70–80% in MKN45 cells. With this approach, IQGAP2 protein expression in MKN45 transfectants was reduced by >70%, 48 (data not shown) and 72 hr subsequent to siRNA introduction (Fig. 6a). MKN45 cells transfected with scrambled control or IQGAP2 siRNA for 48 hr were plated on Matrigel or control inserts for an invasion assay. IQGAP2 knockdown significantly increased the invasiveness of the siRNA-treated cells (Fig. 6b).
We found here, using MSP and bisulfite sequencing methods, that 3 gastric cancer cell lines without or with less IQGAP2 mRNA expression exhibited methylation of the IQGAP2 CpG island, whereas 4 cell lines with high IQGAP2 expression did not. Moreover, treatment of 3 methylation-positive gastric cancer cell lines with demethylating agent DAC induced reactivation of IQGAP2 expression. Therefore, the loss of IQGAP2 expression in some gastric cancer cell lines may be strongly associated with IQGAP2 gene promoter methylation. However, 2 gastric cancer cell lines without methylation revealed IQGAP2 mRNA but no protein. It is possible that the loss of IQGAP2 protein in these cell lines might be caused by other mechanisms, such as aberrant posttranscriptional control.
IQGAP2 methylaton was also found in 28 of the 59 (47%) primary gastric carcinomas. The methylation frequency in cancers without IQGAP2 protein expression was significantly higher than that in ones with positive expression, and the 12 normal gastric tissues exhibited no IQGAP2 methylation. These results indicate that the IQGAP2 gene may be silenced through methylation in primary gastric carcinomas as well as in gastric cancer cell lines, and that the loss of expression of this gene may contribute to gastric carcinogenesis. To our knowledge, this is the first report of silencing through methylation being found in the IQGAP family.
IQGAP2 methylation in gastric carcinomas was significantly correlated with 3 clinicopathological factors, size, depth of invasion, and lymph node metastasis, as shown in Table III. However, only the depth of invasion was found to be independently and significantly associated with IQGAP2 methylation, as revealed on multivariate logistic regression analysis. To determine whether IQGAP2 influences the invasiveness of gastric carcinoma cells, we performed an in vitro invasion assay. Knockdown of IQGAP2 using siRNA in MKN45 cells significantly upregulated the invasiveness, suggesting that IQGAP2 suppresses the invasive capacity of cells. The antiinvasion capacity of IQGAP2 is consistent with our data that the survival of patients with IQGAP2 methylation was significantly shorter than that without methylation.
In contrast to IQGAP2, IQGAP1 promotes cell invasion in a mammary cancer cell line.17 It is unknown why IQGAP1 and IQGAP2 act oppositely as to cell invasion. Although IQGAP1 and IQGAP2 share domain structures and exhibit amino acid homology, there are small differences in their structures, that is, IQGAP2 lacks 1 IQGAP-specific repeat and 1 IQ motif.33 Though the function of IQGAP-specific repeats remains unknown, the IQ motif binds to calmodulin and shows a discrete association with the myosin essential light chain.34, 35
The expression patterns of the 2 IQGAPs are different in the gastric epithelium. IQGAP2 is uniformly expressed in oxyntic mucosal cells, predominantly at cell–cell contacts and in nuclei. On the contrary, IQGAP1 expression is largely restricted to the lateral cortical regions of nonparietal cells.36 The differences in phosphorylation consensus sites, with a preponderance of PKA sites in IQGAP2 and a preponderance of PKC sites in IQGAP1, suggest that these proteins may also be differently regulated through phosphorylation-dependent mechanisms.36 In neuronal cells, IQGAP2 and IQGAP3 are required for axon outgrowth in hippocampal neurons, but IQGAP1 is dispensable.33 In IQGAP1−/− mice, the expression pattern of IQGAP2 was not obviously changed, suggesting that functional redundancy with IQGAP2 is unlikely to mask essential roles of IQGAP1 in adult mouse tissues.37 These data indicate that IQGAP1 and IQGAP2 modulate distinct cellular activities.
Among gastric carcinomas, the IQGAP1 protein was expressed in all of the 47 primary cancers, although its localization varied.21 Moreover, IQGAP1 gene amplification was detected in 2 gastric cancer cell lines.20 In contrast, we found that IQGAP2 expression was lost in 61% of primary gastric carcinomas. Therefore, IQGAP1 and IQGAP2 may play different roles in gastric carcinogenesis, including cell invasion. Further investigations are necessary to clarify the role of IQGAP2 in cell invasion.
The IQGAP2 methylation was correlated with the poor prognosis of gastric cancers. Considering that silencing of IQGAP2 through methylation in primary gastric carcinomas or knockdown of IQGAP2 by siRNA influenced the invasiveness of gastric cancer cells, IQGAP2 may not only be a good prognostic indicator of gastric cancer but also a candidate molecular target of therapy for gastric cancer.
Authors thank Drs. H. Tanaka and A. Jinawath for advice regarding cloning and sequencing.
- 1Stomach cancer. Lyon: IARC Press, 2003. 197., .