Overexpression of Snail is associated with lymph node metastasis and poor prognosis in patients with gastric cancer
verfasst von:
Na Ri Shin, Eun Hui Jeong, Chang In Choi, Hyun Jung Moon, Chae Hwa Kwon, In Sun Chu, Gwang Ha Kim, Tae Yong Jeon, Dae Hwan Kim, Jae Hyuk Lee, Do Youn Park
Epithelial–mesenchymal transition (EMT) plays a significant role in tumor progression and invasion. Snail is a known regulator of EMT in various malignant tumors. This study investigated the role of Snail in gastric cancer.
Methods
We examined the effects of silenced or overexpressed Snail using lenti-viral constructs in gastric cancer cells. Immunohistochemical analysis of tissue microarrays from 314 patients with gastric adenocarcinoma (GC) was used to determine Snail’s clinicopathological and prognostic significance. Differential gene expression in 45 GC specimens with Snail overexpression was investigated using cDNA microarray analysis.
Results
Silencing of Snail by shRNA decreased invasion and migration in GC cell lines. Conversely, Snail overexpression increased invasion and migration of gastric cancer cells, in line with increased VEGF and MMP11. Snail overexpression (≥75% positive nuclear staining) was also significantly associated with tumor progression (P < 0.001), lymph node metastases (P = 0.002), lymphovascular invasion (P = 0.002), and perineural invasion (P = 0.002) in the 314 GC patients, and with shorter survival (P = 0.023). cDNA microarray analysis revealed 213 differentially expressed genes in GC tissues with Snail overexpression, including genes related to metastasis and invasion.
Conclusion
Snail significantly affects invasiveness/migratory ability of GCs, and may also be used as a predictive biomarker for prognosis or aggressiveness of GCs.
The online version of this article (doi:10.1186/1471-2407-12-521) contains supplementary material, which is available to authorized users.
Na Ri Shin, Eun Hui Jeong, Chang In Choi contributed equally to this work.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
NRS, EHJ, CIC and DYP were involved in the design of the study, collected the clinical data, performed the immunohistochemical analysis and drafted the manuscript. HJM performed in vitro study. CHK performed the analysis of microarray data and helped to draft the manuscript. ISC provided general support and helped to analyze the microarray data. GHK, TYJ, DHK and JHL provided the study materials or patients. DYP supervised the study. All authors read and approved the final manuscript.
Background
Epithelial–mesenchymal transition (EMT), a developmental process whereby epithelial cells reduce intercellular adhesion and acquire myofibroblastic features, is critical to tumor progression [1‐3]. During EMT, significant changes occur, including downregulation of epithelial markers such as E-cadherin, translocation of β-catenin (i.e., dissociation of membranous β-catenin and translocation into the nuclear compartment), and upregulation of mesenchymal markers such as vimentin and N-cadherin [3‐6]. EMT is induced by repression of E-cadherin expression by EMT regulators such as Snail, Slug, and Twist. The Snail family of zinc-finger transcriptional repressors directly represses E-cadherin in vitro and in vivo via an interaction between their COOH-terminal region and the 5′-CACCTG-3′ sequence in the E-cadherin promoter [7‐9]. Snail is reportedly important in several carcinomas, including non-small cell lung carcinomas, ovarian carcinomas, urothelial carcinomas, and hepatocellular carcinoma [10‐13]. Studies have also used immunohistochemical analyses to show the clinical significance of Snail overexpression in gastric adenocarcinoma (GC) [14, 15]. However, few reports on the roles of Snail in GC have included clinicopathological, prognostic, and functional in vitro analyses as well as gene expression results. We therefore evaluated Snail’s effect on invasiveness/migratory ability in gastric cancer cell lines, and also investigated the possibility of Snail being used as a predictive marker for evaluating poor prognosis or tumor aggressiveness in GC patients. We also evaluated the gene expression pattern in 45 GC tissues with Snail overexpression, using cDNA microarrays.
Methods
shRNA lentivirus-mediated silencing and overexpression of Snail in gastric cancer cells
Human gastric cancer cell lines SNU216 and SNU484 were obtained from Korean Cell Line Bank (KCLB) and were authenticated by DNA profiling. These cells cultured in RPMI1640 medium with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin (hyClone, Ogden, UT). All cells were maintained at 37°C in 5% CO2. Lentiviral-based RNA knockdown and overexpression were used for silencing and overexpression of Snail. Lentiviruses expressing either non-target or Snail-targeted shRNAs were used for silencing; a PLKO lentiviral vector targeting Snail or an empty PLKO vector were used for overexpression of Snail in the SNU216 and SNU484 cells. Lentivirus stocks were produced using the Virapower™ lentiviral packaging mix using the 293FT cell line according to the manufacturer’s protocol (Invitrogen, Carlsbad, CA). SNU216 and SNU484 cells grown to 50% confluence were incubated for 24 h in a 1:1 dilution of virus:media with 5 μg/ml Polybrene. After a 24-h recovery period in complete media without virus, polyclonal stable cell lines were selected and maintained in media containing 5 μg/ml puromycin. Silencing or overexpression of Snail was determined by RT-PCR and western blotting.
Anzeige
Real time RT-PCR analysis of VEGF, MMP11, and Snailin gastric cancer cells
Total cellular RNA was extracted using the TRIzol method (Sigma-Aldrich, St Louis, MO, USA). For RT-PCR analysis, 2-μg aliquots of RNA were subjected to cDNA synthesis with 200 U of MMLV reverse transcriptase and 0.5 μg of oligo(dT)-15 primer (Promega, Madison, WI, USA). Quantitative real-time PCR was performed with the Rotor-Gene™ System (QIAGEN, Hilden, Germany) using AccuPower 2× Greenstar qPCR Master Mix (Bioneer, Daejeon, Korea). cDNA in 1 μl of the reaction mixture was amplified with 0.5 U of GoTaq DNA polymerase (Promega) and 10 pmol each of the following sense and antisense primers: GAPDH 5′-TCCATGACAACTTTGGTATCG-3′, 5′-TGTAGCCAAATTCGTTGTCA-3′; Snail 5′-CTTCCTCTCCATACCTG-3′, 5′-CATAGTTAGTCACACCTCGT-3′; VEGF 5′-TTGCTGCTCTACCTCCACCA-3′, 5′-GCACACAGGATGGCTTGAA-3′; MMP11 5′-CTTGGCTGCTGTTGTGTGCT-3′, 5-AGGTATGGAGCGATGTGACG-3′. The thermal cycling profile was: denaturation for 30 s at 95°C, annealing for 30 s at 52°C (depending on the primers used), and extension for 30 s at 72°C. For semi-quantitative assessment of expression levels, 30 cycles were used for each PCR reaction. PCR products were size-fractionated on 1.0% ethidium bromide/agarose gels and quantified under UV transillumination. The threshold cycle (CT) is defined as the fractional cycle number at which the fluorescence passes a fixed threshold above baseline. Relative gene expression was quantified using the average CT value for each triplicate sample minus the average triplicate CT value for GAPDH. Differences between the control (empty vector) and experiment groups (infected with the lentivirus) were calculated using the formula 2 – ([△CT Lenti] – [△CT control]) and expressed as a fold change in expression according to the comparative threshold cycle method (2–△△CT) [16].
Western blotting
Cells were harvested and disrupted in lysis buffer (1% Triton X-100, 1mM EGTA, 1mM EDTA, 10mM Tris–HCl, pH 7.4 and protease inhibitors). Cell debris was removed by centrifugation at 10,000 × g for 10 min at 4°C. The resulting supernatants were resolved on a 12% SDS-PAGE under denatured reducing conditions and transferred to nitrocellulose membranes. The membranes were blocked with 5% non-fat dried milk at room temperature for 30 min and incubated with primary antibodies. The membranes were washed and incubated with horseradish peroxidase-conjugated secondary antibody. The signal was visualized using an enhanced chemiluminescence (Amersham, Buckinghamshire, UK).
Cell migration and Matrigel invasion assay
Gastric cancer cells were harvested with 0.05% trypsin containing 0.02% EDTA (Sigma-Aldrich), and suspended in RPMI at a concentration of 3 × 103 cells/well. Membrane filters (pore size: 8 μm) in disposable 96-well chemotaxis chambers (Neuro Probe, Gaithersburg, MD) were pre-coated for 4 h with 5 mg/ml fibronectin at room temperature. Aliquots (50 μl/well) of the cell suspension were loaded into the upper chambers, and 1% FBS was loaded into the lower chamber. After 24-h incubation, non-migrating cells were removed from the upper chamber with a cotton swab; cells present on the lower surface of the insert were stained with Hoechst33342 (Sigma-Aldrich). Invasive cells were counted under a fluorescence microscope at × 10 magnification.
For the Matrigel invasion assay, 3 × 104 cells/well were seeded in the upper chamber, which was coated with Matrigel (5 mg/ml in cold medium, BD Transduction Laboratories, Franklin Lakes, NJ, USA), and serum-free medium containing 1% FBS or control vehicle was added to the lower chamber. After 24-h incubation, non-migrating cells were removed from the upper chamber with a cotton swab, and cells present on the lower surface of the insert were stained with Hoechst33342 (Sigma-Aldrich). Invasive cells were then counted under a fluorescence microscope at × 10 magnification.
Anzeige
Tissue microarrays, immunohistochemistry, and interpretation of results
A semi-automated tissue arrayer (Beecher Instruments, WI, USA) was used to construct the tissue microarrays. We obtained 3 tissue cores, each 0.6 mm in diameter, from tumor blocks taken from GC patients. Cores were not collected from the more invasive frontal or central areas of the tumors. Slides were baked at 60°C for 30 min, deparaffinized with xylene, and then rehydrated. The sections were subsequently submerged in citrate antigen retrieval buffer, microwaved for antigen retrieval, treated with 3% hydrogen peroxide in methanol to quench endogenous peroxidase activity, and then incubated with 1% bovine serum albumin to block non-specific binding. Thereafter, the sections were incubated with rabbit anti-Snail (Abcam, UK) overnight at 4°C. Normal rabbit serum was used as a negative control. After washing, tissue sections were treated with secondary antibody, counterstained with hematoxylin, dehydrated, and mounted. At least 500 tumor cells were counted. The percentage of cells with Snail+ nuclei was expressed relative to the total number of tumor cells counted. Nuclear expression of Snail was graded by classifying the extent of positive nuclear staining as ≤50%, 50–75%, or ≥75%.
Clinicopathological and survival analysis of gastric cancer patients
We studied a cohort of 314 GC patients who each underwent a gastrostomy with lymph node dissection at Pusan National University Hospital (PNUH) between 2005 and 2007. The group comprised 218 men and 96 women with a mean age of 58.3 years (range, 25–83 years). Standard formalin-fixed and paraffin-embedded sections were obtained from the Department of Pathology, PNUH, and the National Biobank of Korea, PNUH. The study was approved by the Institutional Review Board. None of the patients received preoperative radiotherapy and/or chemotherapy. Adjuvant chemotherapy based on 5-FU was administered on patients with stages II, III and IV after curative resection. We assessed several clinicopathological factors according to the Korean Standardized Pathology Report for Gastric Cancer, the Japanese Classification of Gastric Carcinoma (3rd English edition), and the American Joint Committee on Cancer Staging Manual (7th edition), including tumor site, gross appearance and size, depth of invasion, histological classification (i.e., intestinal or diffuse), and lymphovascular invasion [17‐19]. Clinical outcome for each patient was followed from the date of surgery to the date of death or March 1, 2012. Follow-up periods ranged from approximately 1 to 81.5 months (average, 51.4 months). Cases lost to follow-up or death from any cause other than gastric cancer were censored from the survival rate analysis. Clinicopathological features were analyzed using Student’s t-test, the χ2 test, or Fisher’s exact test to test for differences in Snail expression. Cumulative survival plots were obtained using the Kaplan–Meier method, and significance was compared using the log-rank test. Prognostic factors were identified using the Cox regression stepwise method (proportional hazard model), adjusted for the patients’ age, gender, tumor site, morphologic type (intestinal versus diffuse). Statistical significance was set at P < 0.05. Statistical calculations were performed with SPSS version 10.0 for Windows (SPSS Inc., Chicago, IL, USA).
cDNA microarray analysis of GC tissues based on Snail overexpression
A total of 45 fresh GC tissues were obtained from the National Biobank of Korea, PNUH, and CNUH; approval was obtained from their institutional review boards. Total RNA was extracted from the fresh-frozen tissues using a mirVana RNA Isolation kit (Ambion Inc., Austin, TX). Five hundred nanograms of total RNA was used for cDNA synthesis, followed by an amplification/labeling step (in vitro transcription) using the Illumina TotalPrep RNA Amplification kit (Ambion) to synthesize biotin-labeled cRNA. cRNA concentrations were measured by the RiboGreen method (Quant-iT RiboGreen RNA assay kit; Invitrogen-Molecular Probes, ON, Canada) using a Victor3 spectrophotometer (PerkinElmer, CT), and cRNA quality was determined on a 1% agarose gel. Labeled, amplified material (1500 ng per array) was hybridized to Illumina HumanHT-12 BeadChips v4.0, according to manufacturer’s instructions (Illumina, San Diego, CA). Array signals were developed by streptavidin-Cy3. Arrays were scanned with an Illumina iScan system. The microarray data were normalized using the quantile normalization method in Illumina BeadStudio software. The expression level of each gene was transformed into a log2 base before further analysis. Excel was primarily used for statistical analyses. Gene expression differences were considered statistically significant if P < 0.05; all tests were 2-tailed. Cluster analyses were performed using Cluster and Treeview [20]. The gene ontology (GO) program (http://david.abcc.ncifcrf.gov/) was used to categorize genes into subgroups based on biological function. Fisher’s exact test was used to determine whether the proportions of genes in each category differed by group. GC tissues were further divided into those with higher (≥75%) and lower (<75%) levels of Snail expression; differential gene expression between the groups was identified. Primary microarray data are available in NCBI’s GEO (Gene Expression Omnibus) database (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE38024).
Results
Regulation of migration and invasion of gastric cancer cells by Snail
Lentiviral-based RNA knockdown and overexpression approaches were used to determine Snail’s role in invasion and migration of gastric cancer cell lines. SNU216 and SNU484 cells used in this study are established gastric adenocarcinoma cell lines from Korean patients. These cells were infected with a lentivirus expressing either non-target or Snail-targeted shRNAs for silencing. A PLKO lentiviral vector that targeted Snail and an empty PLKO vector were used to induce Snail overexpression in SNU216 and SNU484 cells. Polyclonal stable cell lines were selected using puromycin. Snail expression was determined by RT-PCR and western blotting; stable Snail knockdown (sh-Snail) and Snail overexpression cell lines (OE-Snail) were obtained (Figure 1).
Figure 1
Role of Snail in invasion and migration of gastric cancer cell lines. A. SNU216 (upper panel) and SNU484 (lower panel) cells were infected with lentiviruses expressing either non-target shRNA (shNT) or Snail shRNA on day 0, and then harvested on day 7 post-infection. Snail knockdown was determined by RT-PCR and western blotting; stable cell lines were generated for each of the cell lines (sh-Snail). Silencing of Snail in SNU216 and SNU484 cells induced decreased migration and invasion. B. SNU216 (upper panel) and SNU484 (lower panel) cells were infected with lentiviruses expressing either a lentiviral PLKO vector targeting Snail or an empty PLKO vector (EV) on day 0, and then harvested on day 7 post-infection. The overexpression of Snail was determined by RT-PCR and western blotting; stable cell line was generated for each of the cell lines (O/E-snail). Snail overexpression in SNU216 and SNU484 cells induced increased migration and invasion. C. Snail overexpression induced increased mRNA expression of VEGF and MMP11 in SNU216 and SNU484 cells in real-time RT-PCR analysis. Lower panel indicates representative RT-PCR figures for VEGF, MMP11, Snail, and GAPDH. Data show the mean ± SE of at least 3 independent experiments. * indicates P < 0.05 by Student’s t-test.
×
To determine Snail’s roles in gastric cancer cell invasion, we measured chemotactic invasion by the cells using the Transwell system with filters pre-coated with Matrigel. To measure migration of gastric cancer cells, we assayed cell migration using a Boyden chamber apparatus. Silencing of Snail by shRNA induced decreased migration and invasion of SNU216 and SNU484 cells, as shown in Figure 1A. In contrast to the Snail silencing results, overexpression of Snail induced increased migration and invasion of SNU216 and SNU484 cells, as shown in Figure 1B. Overexpression of Snail was also associated with increased VEGF and MMP11 (Figure 1C).
Effect of Snail overexpression on tumor aggressiveness and GC patient survival
Positive nuclear staining for Snail at levels of ≤50%, 50–75%, and ≥75% was observed in 13.4% (42/314), 52.2% (164/314), and 34.4% (108/314), respectively, of the 314 GC patients in immunohistochemical analysis. Snail expression was noted in intestinal and diffuse type of GCs (Figure 2A, B). Snail overexpression (≥75% positivity) significantly correlated with tumor size, gross type, depth of invasion, lymphovascular invasion, perineural invasion, and lymph node metastasis (Table 1). Snail overexpression was also associated with increased tumor size (P = 0.028) and excavated gross type (P< 0.001); and increased tumor invasiveness, i.e., higher T stage (P< 0.001) and the presence of perineural invasion (P< 0.001) and lymphovascular tumor emboli (P = 0.002). Increased lymph node metastasis was also related to Snail overexpression (P = 0.002).In accordance with the above data showing the positive relationship between Snail overexpression and GC aggressiveness, Snail overexpression significantly correlated with overall survival among GC patients (P = 0.023) (Figure 2C). A linear relationship was observed between increased nuclear expression of Snail and shortened survival (≤50%: 76.6 ± 2.7 months; 50–75%: 68.5 ± 2.0 months; ≥75%: 63.3 ± 2.8 months). Snail overexpression (≥75% positivity) was identified as an independent predictor of poor prognosis in 314 patients with GC, adjusted for age, sex, histologic classification, and tumor location, using a Cox regression proportional hazard model (P = 0.033; Table 2).
Figure 2
Snail expression in gastric adenocarcinoma (GC) tissue samples and Kaplan–Meir plots of overall survival of 314 GC patients. Snail was mostly expressed in nuclei of GC cells (intestinal type (A), and diffuse type (signet ring) cells (B)) included in tissue array specimens. Some reactive fibroblasts also showed Snail nuclear expression (magnification: ×400). C. Kaplan–Meier analysis of overall survival of GC patients based on Snail expression. A linear relationship between increased Snail nuclear expression and shorter survival was seen among GC patients (P = 0.023). Log-rank test was used to calculate P values.
×
Table 1
Relationship between Snail expression and clinicopathological characteristics in 314 patients with gastric cancer
Number of patients (N = 314)
Snail Positivity
Pvalue
<75%
≥75%
Age (years)
58.5 ± 10.6
59.1 ± 11.9
0.695
Sex
Male
218
143
75
0.996
Female
96
63
33
Tumor size
≤4.0 cm
192
135
57
0.028
>4.0 cm
122
71
51
Location
Upper/Middle
167
112
55
0.561
Lower
147
94
53
Invasion depth
T1
160
127
33
< 0.001
T2
41
26
15
T3
68
33
35
T4
43
19
24
Gross type
Elevated
77
51
26
< 0.001
Flat/depressed
131
105
26
Excavated
106
50
56
Histological type
Intestinal
182
123
59
0.609
Diffuse
122
76
46
Mixed
10
7
3
Perineural invasion
Negative
202
150
52
< 0.001
Positive
111
55
56
Lymphovascular emboli
Negative
193
139
54
0.002
Positive
120
66
54
Lymph node metastasis
N0, N1
270
186
84
0.002
N2, N3
44
20
24
Table 2
Multivariate survival analysis with Cox regression model in 314 gastric cancers
Identification of gene expression patterns based on Snail overexpression using cDNA microarrays
cDNA microarrays were used to compare gene expression profiles of 45 GC specimens. We identified 213 genes that were differentially expressed at significant levels (P < 0.05) between GC specimens with higher (≥75%) and lower levels (<75%) of Snail expression (Table 3). Of these 213 genes, 82 were upregulated and 131 were downregulated in the GC specimens with higher levels (≥75%) of Snail expression. We used hierarchical clustering analysis to assess the 213 genes and 45 GC specimens; supervised clustering analysis gave patterns for samples with higher and lower levels of Snail expression clustered into 2 distinct groups, except for one sample with higher levels of Snail expression (Figure 3). To investigate the biological functions involved in discriminating genes, we performed a GO category analysis. Eleven genes were associated with regulating cancer cell–ECM adhesion (P < 0.021) and ECM protein regulation (P < 0.028, Table 4). Most have been implicated in cancer. ONECUT1, ADAMTS, IFNAR2, MSR1, and SORL1 affect migration or metastasis, a process that involves attachment of tumor cells to the basement membrane, degradation of local connective tissue, and penetration and migration of tumor cells through stroma [21‐25].
Table 3
Genes differentially expressed in GC specimens with higher levels of Snail expression
PROBE_ID
SYMBOL
NAME
Genes upregulated in specimens with higher levels (≥75%) of Snail expression (P< 0.05)
ILMN_2374449
SPP1
Secreted phosphoprotein 1
ILMN_2337923
TPD52L1
Tumor protein D52-like 1
ILMN_1679838
WBP5
WW domain binding protein 5
ILMN_2078592
C6orf105
Androgen-dependent TFPI-regulating protein
ILMN_1714383
TPD52L1
Tumor protein D52-like 1
ILMN_1674817
C1orf115
Chromosome 1 open reading frame 115
ILMN_1813561
SCIN
Scinderin
ILMN_1759818
SORL1
Sortilin-related receptor, L(DLR class) A repeats containing
Similar to dehydrogenase/reductase (SDR family) member 4 like 2
ILMN_1893728
Hs.377660
Homo sapienscDNA FLJ26242 fis, clone DMC00770
ILMN_1667005
LOC652676
Similar to similar to hypothetical protein FLJ36144
ILMN_3241607
LOC100132106
Hypothetical LOC100132106
ILMN_1797503
GOLGA8G
Golgin A8 family, member G
ILMN_1828034
Hs.154513
ik89c11.z1 Human insulinomaHomo sapienscDNA clone IMAGE:6027645 3, mRNA sequence
ILMN_1886816
Hs.544491
qq31a07.x1 Soraes_NhHMPu_S1 Homo sapienscDNA clone IMAGE:1934100 3, mRNA sequence
ILMN_1847950
Hs.505398
wq87c02.x1 NCI_CGAP_GC6 Homo sapienscDNA clone IMAGE:2479010 3, mRNA sequence
ILMN_1734479
ACCN3
Acid-sensing (proton-gated) ion channel 3
ILMN_1675025
H2BFM
H2B histone family, member M
ILMN_2073279
SIM1
Single-minded homolog 1 (Drosophila)
ILMN_1910185
Hs.98563
zw57h03.s1 Soares_total_fetus_Nb2HF8_9w Homo sapienscDNA clone IMAGE:774197 3, mRNA sequence
ILMN_3251491
UQCRB
Ubiquinol-cytochrome c reductase binding protein
ILMN_2180315
ATG4D
ATG4 autophagy related 4 homolog D (S. cerevisiae)
ILMN_1885583
Hs.542934
Homo sapienscDNA FLJ26431 fis, clone KDN01390
ILMN_1743301
MSR1
Macrophage scavenger receptor 1
ILMN_1809820
LOC648963
Similar to retinitis pigmentosa 1-like 1
ILMN_1869348
Hs.460114
UI-E-EJ0-ahv-d-07-0-UI.s1 UI-E-EJ0 Homo sapienscDNA clone UI-E-EJ0-ahv-d-07-0-UI 3, mRNA sequence
ILMN_1711332
TFEC
Transcription factor EC
ILMN_2228538
IRAK1BP1
Interleukin-1 receptor-associated kinase 1 binding protein 1
ILMN_1756455
IL5RA
Interleukin 5 receptor, alpha
ILMN_1719202
ZNF174
Zinc finger protein 174
ILMN_1847029
Hs.553290
HESC3_84_D06.g1_A036 Human embryonic stem cells Homo sapienscDNA clone IMAGE:7483454 5, mRNA sequence
ILMN_1740217
HACE1
HECT domain and ankyrin repeat containing E3 ubiquitin protein ligase 1
ILMN_1787464
LOC651296
Similar to RAB, member of RAS oncogene family-like 2B isoform 1
ILMN_1734096
DCLRE1A
DNA cross-link repair 1A
ILMN_2391333
CYP20A1
Cytochrome P450, family 20, subfamily A, polypeptide 1
ILMN_2226314
DBR1
Debranching enzyme homolog 1 (S. cerevisiae)
ILMN_2379560
CDC14B
CDC14 cell division cycle 14 homolog B (S. cerevisiae)
ILMN_2078466
DZIP1L
DAZ interacting protein 1-like
ILMN_1653039
LOC642934
Hypothetical protein LOC642934 (LOC642934), mRNA
ILMN_2044293
KBTBD7
Kelch repeat and BTB (POZ) domain containing 7
ILMN_1809951
ZNF200
Zinc finger protein 200
ILMN_1760280
NXT1
NTF2-like export factor 1
ILMN_1657796
STMN1
Stathmin 1
ILMN_1793578
ZFP37
Zinc finger protein 37 homolog (mouse)
Figure 3
Supervised clustering analysis of 45 gastric adenocarcinoma (GC) specimens and 172 genes. Hierarchical clustering was used for 45 GC specimens and 213 genes. Data are shown in a matrix format, with rows representing individual genes and columns representing tissues. Each cell in the matrix represents the expression level of a gene featured in an individual tissue. Red and green cells reflect GCs with higher (≥75%) and lower (<75%) levels of Snail expression, respectively. Matrix patterns for specimens clustered into 2 distinct groups, except for one sample with higher levels of Snail expression.
×
Table 4
Cellular functions of selected genes that are differentially expressed in GC specimens that overexpress Snail
Probe ID
Gene acronym
Gene name
Accession No.
Pvalue
Cancer cell–ECM adhesion
ILMN_1759487
EGFLAM
EGF-like, fibronectin type III, and laminin G domains (↑)
NM_182801
0.005
ILMN_2114812
ONECUT1
One cut homeobox 1 (↓)
NM_004498
0.002
ILMN_2374449
SPP1
Secreted phosphoprotein 1 (↑)
NM_000582
0.004
ECM protein regulation
ILMN_1676452
ADAMTS14
ADAM metallopeptidase with thrombospondin type 1 motif, 14 (↑)
NM_080722
0.005
ILMN_1759487
EGFLAM
EGF-like, fibronectin type III, and laminin G domains (↑)
NM_182801
0.005
ILMN_1683453
FRAS1
Fraser syndrome 1 (↓)
NM_020875
0.003
ILMN_1791057
IFNAR2
Interferon (alpha, beta, and omega) receptor 2 (↑)
NM_207585
0.001
ILMN_1756455
IL5RA
Interleukin 5 receptor, alpha (↓)
NM_000564
0.004
ILMN_1747850
CRIM2
Kielin/chordin-like protein (↑)
NM_199349
0.005
ILMN_1743301
MSR1
macrophage scavenger receptor 1 (↓)
NM_002445
0.002
ILMN_2374449
SPP1
secreted phosphoprotein 1 (↑)
NM_000582
0.004
ILMN_2256050
SERPINA1
Serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1 (↑)
NM_000295
0.002
ILMN_2060115, ILMN_1759818
SORL1
Sortilin-related receptor, L(DLR class) A repeats-containing (↑)
NM_003105
0.003 <0.001
NOTE: ↑, upregulation; ↓, downregulation.
Discussion
Snail is reportedly a key regulator of tumor progression and metastasis via increased MMP expression and tumor invasion [26, 27]. Similarly, we found that upregulated Snail expression increased gastric cancer cell invasion/migration, whereas downregulated Snail expression decreased gastric cancer cell invasion/migration. Yang et al. reported that Snail overexpression in hepatocellular carcinoma cell lines induced increased invasiveness/metastasis [13]. In addition, Kosaka et al. reported that Snail knockdown was associated with decreased invasive capacity of a urothelial carcinoma cell line, supporting our results [12]. We also found that Snail overexpression induced increased expression of VEGF and MMP11, which are known markers of tumor invasion and metastasis. Jin et al. also reported that Snail knockdown by antisense Snail was associated with inhibited MMP activity, demonstrating the importance of regulating MMP activity in cancer metastasis.10 Furthermore, Peinado et al. reported that I MDCK cells with Snail overexpression had increased angiogenesis and VEGF [28]. We also observed increased VEGF in gastric cancer cells with Snail overexpression.
Anzeige
The clinical importance of Snail in various carcinomas, including non-small cell lung carcinomas, ovarian carcinomas, urothelial carcinomas, hepatocellular carcinoma, and breast cancer, is well known, as is the poor prognosis associated with Snail overexpression [10‐13, 29]. However, only limited immunohistochemical data have been available on Snail expression in GC, with no comprehensive clinical and functional analysis of Snail expression in GC patients. Kim et al. reported immunohistochemical data indicating that Snail expression was an independent indicator of prognosis in tissue microarray specimens [14]. Rye et al. reported that the combination of Snail, vimentin, E-cadherin, and CD44 was also significantly associated with poor prognosis in gastric cancer [15]. In contrast, no significant correlation between tumor stage and Snail expression was noted in upper gastrointestinal tract adenocarcinoma, including cancers of the esophagus, cardia, and stomach [30]. In our study, overexpression of Snail (≥75% nuclear Snail expression) was significantly associated with tumor progression, lymph node metastases, lymphovascular invasion, perineural invasion, and poor prognosis in GC patients. Recently, He et al. reported Snail to be an independent prognostic predictor of patient survival among gastric cancer patients; this is in agreement with our data [31]. Although 5-FU based adjuvant chemotherapy for advanced or metastatic gastric adenocarcinoma was usually performed in our cohort, further work is required to reveal exact significance of Snail expresssion as predictor of chemotherapy response in gastric adenocarcinoma. For the practical use of Snail as a tissue biomarker in predicting lymph node metastasis and poor prognosis, we defined a cut-off value of 75% positive nuclear expression for Snail overexpression. There are wide variations in cut-off values for Snail overexpression in different types of cancer; for example, 75% is used in non-small cell lung carcinoma [11], 100 (score of mean percentage × intensity, range 0–300) is used in urothelial carcinomas [12], and 50% is used in hepatocellular carcinoma [13]. For gastric cancers, cut-off values of 10% [14] and 5% [15] positive nuclear expression of Snail have been reported. Further work is required to determine a practical consensus cut-off value for Snail overexpression.
A total of 213 genes that were differentially expressed among GC samples with higher (≥75%) and lower levels of Snail expression were clustered into 2 distinct groups: those associated with regulation of cancer cell–ECM adhesion, and those associated with ECM protein regulation, such as ONECUT1[21], ADAMTS[22], IFNAR2[23], MSR1[24], and SORL1[25]. These functions indicate that Snail greatly affects cancer cell migration and metastasis by regulating attachment of tumor cells to basement membranes, degradation of local connective tissue, and penetration and migration of tumor cells through stroma.
Conclusions
In this study, we showed that Snail overexpression induced increased migration and invasion in gastric cancer cell lines. Snail overexpression was also significantly associated with tumor progression, lymph node metastases, lymphovascular invasion, perineural invasion, and poor prognosis in GC patients. We identified 213 genes that were differentially expressed in GC tissues that overexpressed Snail, including genes related to metastasis and invasion by tumor cells. Our results indicate that Snail is crucial in controlling progression and metastasis of gastric cancer. Thus Snail may be used as a predictive biomarker for evaluating prognosis or aggressiveness of GCs.
Acknowledgements
This study was supported by grant 0920050 from the National R&D Program for Cancer Control, Ministry for Health, Welfare and Family Affairs, Republic of Korea.
Anzeige
Open Access
This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License (
https://creativecommons.org/licenses/by/2.0
), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
NRS, EHJ, CIC and DYP were involved in the design of the study, collected the clinical data, performed the immunohistochemical analysis and drafted the manuscript. HJM performed in vitro study. CHK performed the analysis of microarray data and helped to draft the manuscript. ISC provided general support and helped to analyze the microarray data. GHK, TYJ, DHK and JHL provided the study materials or patients. DYP supervised the study. All authors read and approved the final manuscript.
