Lung cancer (including non–small-cell lung cancer [NSCLC] and small-cell lung cancer) remains the leading cause of cancer-related deaths in the world.1 Although great efforts have been made toward improving early diagnosis and effective treatment, including findings of new biomarkers,2–4 establishment of modified operation, and development of specific drugs,5–9 the prognosis is still poor.
The most common features of malignancy are invasion and metastasis, which are characterized by the ability of cancer cells to invade into adjacent area, intravasate into blood or lymphatic vessels, and extravasate into a distant environment. Metastasis, which is primarily responsible for the low 5-year survival rates (approximately 10%–15%),10,11 is especially required for aggressive NSCLC, and once cancer cells lose contact with the neighboring cells, they become motile, invade the surrounding area, migrate, enter the circulation, extravasate into the target organ, proliferate, and metastasis. Therefore, the investigation of this area is important for patients with advanced NSCLC, but far from satisfaction to date.
The Wilms’ tumor gene 1 (WT1) encodes a 49 to 52 kDa protein with an N-terminal domain involved in RNA/protein interactions12 and is an important regulator of cell growth and development in the embryo kidney, adult urogenital system, and central nervous system.13 The C-terminal domain harbors four zinc fingers, which permit binding to target DNA sequences for its transcriptional regulatory function.14 Through two alternative splicing motifs of exon 5 (17 amino acid) and exon 9 (3 amino acid, KTS region), four main protein isoforms of WT1 are designated as A (-/-), B (+/-), C (-/+), and D (+/+).15 A large number of genes coding for growth factors (e.g., TGF-β, CSF-1), growth factor receptors (e.g., insulin R, IGF-IR, EGFR), transcription factors (e.g., EGR, WT1, cMyc, Pax2, Dax-1, and Sry), and other proteins (e.g., ODC, MDR1, Hsp70, p21, Bcl-2) have been identified as WT1 target genes.16 Although first recognized as a tumor suppressor in Wlims’ tumor (nephroblastoma),17 based on accumulating evidence, WT1 has been demonstrated to act as an oncogene in other sorts of malignancies including leukaemia,18–20 breast cancer,21,22 and lung cancer23,24 et al., which suggested it acts in a dichotomous manner. The reason why WT1, the chameleon gene,25 plays the very opposite role still remains unclear.
Studies have identified Sox9, Snail, and CDH1 as being deregulated by WT1 in WT1 conditional knockout mice, although available evidence for WT1 in direct transcriptional regulation is only available for Snail and CDH1, which was only reported in cardiovascular progenitor cells.26 The present study was designed to investigate the relationship between WT1 and its downstream molecule E-cadherin in NSCLC. We demonstrated that the expression of WT1 and E-cadherin (CDH1) were correlated and also associated with metastasis in NSCLC samples. We further showed that in vitro, WT1-D enhanced the invasiveness of H1568 and H1650 NSCLC cells via negative modulation of E-cadherin by directly bind to its promoter.
PATIENTS AND METHODS
Patients and Tissue Samples
The study was performed on a total of randomly selected 159 NSCLC patients, of whom 102 had adenocarcinomas and the remaining 57 had squamous cell carcinomas. The patients were treated at People’s Hospital of Jiangsu Province between September 2007 and August 2009, and all the specimens were collected under a protocol approved by the Human Ethics Committee of Nanjing Medical University. Each patient participated after providing informed consent. The median age of the patients was 64 years (range 49–85 years) and most of them were men (73%).
A total of 318 tissue specimens were included in the study with one tumor sample and one corresponding adjacent sample for each patient. All specimens were histologically classified by a professional pathologist according to the national NCCN guidelines for NSCLC version 3.201127 using blind method. Survival rate was calculated using 36 months as cutoff point.
DNA and RNA Preparation
Total RNA was extracted from tissue specimens using the TRIzol method (Invitrogen, Shanghai, China) and cDNA was synthesized using reverse transcriptase kit (TAKARA, Tokyo, Japan) according to the manufacturers’ protocol.
Real-Time Polymerase Chain Reaction for WT1 and CDH1 RNA Expression
WT1 and E-cadherin (CDH1) mRNA levels were measured by real-time polymerase chain reaction (RT-PCR) using SYBR Premix Ex Taq (TAKARA). WT1 and CDH1 transcription values were normalized against the expression of β-actin. Amplification conditions, primers, and probes sequences for the WT1 and β-actin were from the work by Sitaram et al.28 and for CDH1 were the same as those in the work by Martínez-Estrada et al.26
H1568 (CRL-5876), H1650 (CRL-5883), and 293T (CRL-11268) cell lines from ATCC (Manassas, VA) were used for the present study. H1568 and H1650 were cultured in RPMI 1640 medium whereas 293T was cultured in Dulbecco’s Modified Eagle Medium (DMEM) high-glucose medium, both supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA). All cells were maintained in a humidified 37°C incubator with 5% CO2.
Lentivirus Production and Transduction
WT1 gene was synthesized (purchased from GenScript, Nanjing, China) with restrictive digestion site of MluI in both ends, subcloned into pLV-GFP plasmid (gifted by Prof. Beicheng Sun, University of Nanjing Medical University, China), and named pLV-GFP-WT1. To generate plasmid-expressing WT1-shRNA, double-stranded oligonucleotides were cloned into pLL3.7 vector (gifted by D. Yun Chen, University of Nanjing Medical University, China) and named pLL3.7-WT1-shRNA. The sequences of WT1-shRNA used are aac TCAGGGTTACAGCACGGTC ttcaagaga GACCGTGCTGTAACCCTGA tttttt c. The uppercase letters represent WT1-specific sequence, and lowercase letters represent hairpin sequences. Recombinant lentivirus was generated from 293T cells using calcium phosphate precipitation. H1568 and H1650 were transfected with lentivirus using polybrene (8 µg/ml).
Western Blotting Assay
Proteins were extracted from cultured cells, quantitated using a protein assay (bicinchoninic acid [BCA] method; Beyotime, Shanghai, China). Proteins were fractionated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride (PVDF) membrane, blocked in 4% dry milk at room temperature for 1 hour, and immunostained with primary antibodies at 4°C overnight using anti-WT1 (1:1000, 6F-H2; Millipore, Billerica, MA), anti-E-cadherin (1:1000; Abcam, Cambridge, United Kingdom), and β-actin (1: 20,000; Dako, Glostrup, Denmark). The results were visualized via a chemiluminescent detection system (Pierce ECL Substrate Western blot detection system; Thermo, Rockford, IL) and exposed in Molecular Imager ChemiDoc XRS System (Bio-Rad, Hercules, CA).
Cell-Proliferation Assay (Cell-Counting Kit-8)
Cells were seeded into 96-well plates (6.0 × 103 cells per well). Cell viability was assessed by cell-counting kit-8 assay (Beyotime Institute of Biotechnology, Shanghai, China). The absorbance of each well was read on a spectrophotometer (Thermo) at 450 nm (A450). Three independent experiments were performed in quintuplicate.
Cells were seeded in six-well plates and cultured to confluence. Wounds of 2-mm width were created with a plastic scriber and the floating cells were washed away thrice with phosphate buffered saline (PBS). After incubation in a serum-free medium for 48 hours, cultures were observed and photos were taken under a microscope. A minimum of five randomly chosen areas was measured.
Transwell Invasion Assay
The invasive ability of the cells was investigated using Transwells (8-µm pore size; Corning Costar Corp, Bedford, MA) put into the 24-well plates. First, 50 µl Matrigel (50 µg/ml; BD Biosciences, San Jose, CA) was added onto each surface of the chamber, incubated for 2 hours for solidification, then the supernatant was washed away with warm PBS. H1568 and H1650 were suspended in RPMI 1640 containing 2% fetal bovine serum. A total of 100 µl of the cell suspension (5 × 104 cells) was added to the upper chamber coated with Matrigel, and 400 µl of RPMI 1640 containing 10% fetal bovine serum was added to the lower compartment. After incubation for 48 hours at 37°C in a 5% CO2 humidified incubator, the Matrigel and cells on the upper surface of the filter were removed with cotton swabs and the cells that invaded into the lower surface were fixed with 2% paraformaldehyde, stained with crystal violet. Then the filters were removed from the chambers, air-dried on the precleaned slides and applied with cover-slides using resina. Images were taken under an inverted microscope (Olympus Corp, Tokyo, Japan) at ×100 magnification over three random fields in each well. ImageJ 1.45s software (National Insititutes of Health, Bethesda, MD) was used for integrated optical density analysis. Each experiment was performed in triplicate.
Total RNA was extracted from H1568, H1568-WT1, and H1568-shWT1 and reverse transcribed to cDNA. Subsequently, cDNA was amplified by PCR using 23 Super Array PCR master mix (SuperArray Bioscience, Frederick, MD) and then RT-PCR was carried out using the Human Tumor Metastasis RT2 Profiler PCR array (SuperArray Bioscience) in an ABI PRISM7900 system (Applied Biosystems, Foster City, CA), according to the manufacturer’s instructions.
Luciferase Reporter Assay
A dual-luciferase reporter assay was performed to investigate the transcriptional regulation of CDH1 by WT1. The promoter of CDH1-containing WT1 binding site (cgcccccac) was synthesized artificially and named control. The binding site is italicized and underlined. The promoter with the mutated binding site (gcgaaccga) was also synthesized and named mutant, as shown in Figure 3C. Products were validated via agarose gel electrophoresis, and cloned into the luciferase reporter vector (Promega, Madison, WI). WT1 gene was cloned into pcDNA3.1 plasmid (gifted by D. Beicheng Sun) and named pcDNA3.1-WT1. H1568 cells were seeded into 24-well plates and cotransfected with WT1-siRNA (pooled siGENOME SMART pool WT1 siRNA [50 nM per well]; Dhamacon, Chicago, IL) or pcDNA3.1-WT1, control siRNA (Dhamacon), and empty pcDNA3.1 as control, respectively, and luciferase reporter vectors using Lipofectamine 2000 (Invitrogen), following the instructions. Zero, 6, 12, and 24 hours after incubation, cells were collected and firefly and Renilla luciferase activity was measured with the dual-luciferase reporter assay system (Promega). All results were gained through three independent experiments.
Statistical analysis was performed using GraphPad Prism (version 5.01; GraphPad Software, Inc, La Jolla, CA) statistical software. The Student’s t test and paired t test were used to analyze significance between independent groups and paired materials, respectively. The correlation test was used to analyze the correlation between WT1 and CDH1. The χ2 test was used to test the significance of observed differences in proportions except when the cells size was less than 5 (Fisher’s exact tests). The significance was accepted as p value was less than 0.05.
WT1 and CDH1 RNA levels were analyzed in NSCLC specimens and corresponding adjacent samples cut off from all 159 patients. Significantly higher WT1 (p < 0.0001; Fig. 1A) and lower CDH1 (p < 0.0001; Fig. 1B) RNA levels were detected in tumor specimens than in adjacent areas. Moreover, a negative correlation was observed between WT1 and CDH1 (R = −0.697; p < 0.0001; Fig. 1C) in tumor samples. The results suggested up-regulated expression of WT1 and down-regulation of CDH1 in NSCLC.
A total of 159 NSCLC were divided into two groups using median RNA level of WT1 (median = 3.0013) and CDH1 (median = 0.4031) as threshold, respectively. The relationship of WT1 (CDH1) expression levels and clinical (pathological) characteristics were shown in Table 1. Significant higher pathological stage (p < 0.0001), metastasis rate (p < 0.0001), and lower survival rate (p < 0.0001) were observed in both high-WT1 and low-CDH1 expression groups, which supported the result of correlation analysis, indicating a potential functional link of WT1 and CDH1. Age, sex, and histologic diagnosis were not associated with WT1 or CDH1, as shown in Table 1.
In vitro, both H1568 and H1650 cells exhibited an invasive phenotype (Fig. 2A, E) that could be significantly enhanced when WT1 was overexpressed (Fig. 2B, F) and evidently suppressed when WT1 was down-regulated by shRNA (Fig. 2C, G), which was confirmed by integrated optical density value testing (Fig. 2D, H; p < 0.005). The results of wound-healing assay were consistent with those of the transwell assay, which proved that WT1 overexpression could enhance the migration and that down-regulation could suppress the process in both H1568 (Fig. 2I) and H1650 (Fig. 2J). WT1 overexpression or silencing could not affect cell proliferation (Fig. 2K, L), which, taken together, proved that the differences between the penetrated cells after WT1 plasmid or shRNA transfection were caused by variety of invasive rather than proliferative abilities.
A variety of genes were screened out after WT1 overexpression and down-regulation via the PCR array of human metastasis pathway (Supplementary Table 1, Supplemental Digital Content 1, http://links.lww.com/JTO/A459) and nine of the genes whose variation larger than twofolds were picked and displayed in Figure 3A. CDH1 increased (4.09-fold) the most in H1568-shWT1 cells and NR4A3 decreased the most (−3.57-fold) among these genes in H1568-WT1 cells. The Western blot assay confirmed the negative regulatory effect of WT1 on CDH1 (E-cadherin) in both H1568 and H1650 cell lines (Fig. 3B).
To demonstrate the functional link between WT1 and CDH1, we performed luciferase reporter assay in H1568 cell lines with pcDNA3.1-WT1 plasmid and WT1 siRNA for transient transfection experiments. The luciferase activity representing CDH1 transcription was weakened after WT1 overexpression and, in contrast, enhanced after WT1 silencing, both with time increases (Fig. 3D, E). Moreover, no differences of luciferase activity were observed with mutated CDH1 promoter (Fig. 3F), suggesting a regulation function of WT1 on CDH1 through direct binding to its promoter.
Lung cancer is an aggressive malignancy, causing the most cancer-related deaths in the world.1 The best treatment for early-stage patients who are defined by the tumor, node, metastasis (TNM) classification continues to be surgical resection, however, a considerable number of these patients will develop metastasis or relapse even when the surgeries are successful.29,30 The elucidation of the mechanisms of metastatis is crucial for improving surgery and treatment outcome, especially for selecting and customizing chemotherapy. Recently, numerous regulatory functions of WT1 have been reported and, most notable among these was that WT1 was found to be involved in inducing and mediating cardiovascular progenitor cell formation.26 In our study, we discovered a relationship of WT1 RNA level and pathological stage, metastasis and survival rate, found the association between WT1 and CDH1, and demonstrated that WT1 could promote the metastasis of NSCLC cells by direct regulation of CDH1. Furthermore, as reported, CDH1 may play a role in inducing epithelial–mesenchymal transition (EMT) and mesenchymal–epithelial transition (MET), which are considered to be involved in migration, tumor invasion, and dissemination.31
Metastasis is a very complicated, polygene-involved, and multistage cascade. A variety of studies have revealed that EMT and MET processes were essential for metastasis.32–35 The balance of EMT and MET may be the decisive factor for original tumor to possess metastatic ability.36,37 Very recently, Martínez-Estrada et al.26 reported that WT1 was required for cardiovascular progenitor cell formation in which process EMT played important role. Miller-Hodges et al.38 summarized the role of WT1 in proper control of the mesenchymal–epithelial balance of cells. The role of Cdh1 in regulating metastasis and invasion of cancer cells have been demonstrated in many studies. Geiger et al.39 systemically concluded the structure, function, and signal transduction of Cadherin and explained that it played a very important role in cell adhesion. Takeichi et al.40 proved that the perturbation of cadherin function causes temporary or permanent disaggregation of tumor cells and may thus promote their invasion and metastasis. Pinheiro et al.41 expounded that Cdh1 may initiate a variety of transcriptional events to regulate invasion and angiogenesis. According to our study, WT1’s functions on EMT, metastasis and invasion, are probably through Cdh1, the very classical invasion suppressor. However, although WT1 has a dichotomous function in either tissue development or tumor formation, the current studies could only illustrate the importance of WT1 in a tissue-dependent manner. Similarly, although we tried to elucidate the mechanism of WT1 on metastasis and invasion within limits of NSCLC, the biological and pathological functions of WT1 in general, however, remain an unsolved mystery.
A plethora of its target genes have been found since Knudson et al.42 first discovered, located, and named WT1 in 1972. As a transcriptional factor, via direct and indirect regulation, WT1 participates in nearly all types of biological processes. As summarized in Supplementary Table 1 (Supplemental Digital Content 1, http://links.lww.com/JTO/A459) and illustrated in Figure 3A, genes that up-regulated in H1568-shWT1 cells were almost tumor metastasis inhibitors (e.g., CDH1) or suppressors (e.g., CD82 and MCAM) but those that were down-regulated were promoters of metastasis (e.g., FLT4 and MTA1) or proliferation (e.g., IGF1); accordingly, very contradictory results of these gene-expression levels were received in WT1 overexpressed H1568 cells, suggesting that changes in metastasis should not be attributed to alternations of one or a small quantity of genes.
In summary, we identified the negative correlation of WT1 and CDH1 in our 159 NSCLC cases, both of which were associated with pathological stage, metastasis, and survival rate and proved the regulatory function of WT1 on CDH1 through direct binding of its promoter. As the limit on the number of NSCLC samples and cell types, more elaborate studies will be necessary for further exploration of the link between WT1 and CDH1 and its potential role in tumorigenesis and metastasis.
The authors thank Prof. Beicheng Sun and D. Yun Chen from University of Nanjing Medical University for providing plv-GFP, pcDNA 3.1, and pLL-3.7 plasmids.
1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012. CA Cancer J Clin. 2012; 62:10–29
2. Kuang P, Zhou C, Li X, et al. Proteomics-based identification of secreted protein dihydrodiol dehydrogenase 2 as a potential biomarker for predicting cisplatin efficacy in advanced NSCLC patients. Lung Cancer. 2012; 77:427–432
3. Rosser CJ, Goodison S. CD24, a promising biomarker in NSCLC. Biomark Med. 2010; 4:495
4. Tsao AS, Liu S, Lee JJ, et al. Clinical outcomes and biomarker profiles of elderly pretreated NSCLC patients from the BATTLE trial. J Thorac Oncol. 2012; 7:1645–1652
5. Bradbury PA, Shepherd FA. Chemotherapy and surgery for operable NSCLC. Lancet. 2007; 369:1903–1904
6. Tsao AS, Roth JA, Herbst RS. Surgery: future directions in multimodality therapy for NSCLC. Nat Rev Clin Oncol. 2010; 7:10–12
7. van Meerbeeck JP. The controversial role of surgery in stage III NSCLC. Lancet Oncol. 2008; 9:607–608
8. Dubey S, Schiller JH. Three emerging new drugs for NSCLC: pemetrexed, bortezomib, and cetuximab. Oncologist. 2005; 10:282–291
9. Maione P, Gridelli C, Troiani T, Ciardiello F. Combining targeted therapies and drugs with multiple targets in the treatment of NSCLC. Oncologist. 2006; 11:274–284
10. Sanchez de Cos J, Sojo Gonzalez MA, Montero MV, et al. Non-small cell lung cancer and silent brain metastasis. Survival and prognostic factors. Lung Cancer. 2009; 63:140–145
11. Furák J, Troján I, Szöke T, et al. Lung cancer and its operable brain metastasis: survival rate and staging problems. Ann Thorac Surg. 2005; 79:241–7; discussion 241
12. Call KM, Glaser T, Ito CY, et al. Isolation and characterization of a zinc finger polypeptide gene at the human chromosome 11 Wilms’ tumor locus. Cell. 1990; 60:509–520
13. Scharnhorst V, van der Eb AJ, Jochemsen AG. WT1 proteins: functions in growth and differentiation. Gene. 2001; 273:141–161
14. Bruening W, Moffett P, Chia S, Heinrich G, Pelletier J. Identification of nuclear localization signals within the zinc fingers of the WT1 tumor suppressor gene product. FEBS Lett. 1996; 393:41–47
15. Haber DA, Sohn RL, Buckler AJ, Pelletier J, Call KM, Housman DE. Alternative splicing and genomic structure of the Wilms tumor gene WT1. Proc Natl Acad Sci U S A. 1991; 88:9618–9622
16. Ariyaratana S, Loeb DM. The role of the Wilms tumour gene (WT1) in normal and malignant haematopoiesis. Expert Rev Mol Med. 2007; 9:1–17
17. Haber DA, Buckler AJ, Glaser T, et al. An internal deletion within an 11p13 zinc finger gene contributes to the development of Wilms’ tumor. Cell. 1990; 61:1257–1269
18. Ogawa H, Tamaki H, Ikegame K, et al. The usefulness of monitoring WT1 gene transcripts for the prediction and management of relapse following allogeneic stem cell transplantation in acute type leukemia. Blood. 2003; 101:1698–1704
19. Scheibenbogen C, Letsch A, Thiel E, et al. CD8 T-cell responses to Wilms tumor gene product WT1 and proteinase 3 in patients with acute myeloid leukemia. Blood. 2002; 100:2132–2137
20. Cilloni D, Gottardi E, De Micheli D, et al. Quantitative assessment of WT1 expression by real time quantitative PCR may be a useful tool for monitoring minimal residual disease in acute leukemia patients. Leukemia. 2002; 16:2115–2121
21. Tuna M, Chavez-Reyes A, Tari AM. HER2/neu increases the expression of Wilms’ Tumor 1 (WT1) protein to stimulate S-phase proliferation and inhibit apoptosis in breast cancer cells. Oncogene. 2005; 24:1648–1652
22. Oji Y, Miyoshi Y, Kiyotoh E, et al. Absence of mutations in the Wilms’ tumor gene WT1 in primary breast cancer. Jpn J Clin Oncol. 2004; 34:74–77
23. Menssen HD, Bertelmann E, Bartelt S, et al. Wilms’ tumor gene (WT1) expression in lung cancer, colon cancer and glioblastoma cell lines compared to freshly isolated tumor specimens. J Cancer Res Clin Oncol. 2000; 126:226–232
24. Oji Y, Miyoshi S, Takahashi E, et al. Absence of mutations in the Wilms’ tumor gene wt1 in de novo non-small cell lung cancers. Neoplasma. 2004; 51:17–20
25. Huff V. Wilms’ tumours: about tumour suppressor genes, an oncogene and a chameleon gene. Nat Rev Cancer. 2011; 11:111–121
26. Martínez-Estrada OM, Lettice LA, Essafi A, et al. Wt1 is required for cardiovascular progenitor cell formation through transcriptional control of Snail and E-cadherin. Nat Genet. 2010; 42:89–93
27. Ward JH. NCCN Guidelines and the International Community. J Natl Compr Canc Netw. 2011; 9:133–134
28. Sitaram RT, Degerman S, Ljungberg B, et al. Wilms’ tumour 1 can suppress hTERT gene expression and telomerase activity in clear cell renal cell carcinoma via multiple pathways. Br J Cancer. 2010; 103:1255–1262
29. Johnston MR. The limits of surgical resection alone for non-small cell lung cancer. Lung Cancer. 1997; 17:(Suppl 1)S99–102
30. Van Schil PE. Surgery for non-small cell lung cancer. Lung Cancer. 2001; 34:(Suppl 2)S127–S132
31. Ye J, Wu D, Shen J, et al. Enrichment of colorectal cancer stem cells through epithelial-mesenchymal transition via CDH1 knockdown. Mol Med Rep. 2012; 6:507–512
32. Tsai JH, Donaher JL, Murphy DA, Chau S, Yang J. Spatiotemporal regulation of epithelial-mesenchymal transition is essential for squamous cell carcinoma metastasis. Cancer Cell. 2012; 22:725–736
33. Yan W, Cao QJ, Arenas RB, Bentley B, Shao R. GATA3 inhibits breast cancer metastasis through the reversal of epithelial-mesenchymal transition. J Biol Chem. 2010; 285:14042–14051
34. Yang MH, Chang SY, Chiou SH, et al. Overexpression of NBS1 induces epithelial-mesenchymal transition and co-expression of NBS1 and Snail predicts metastasis of head and neck cancer. Oncogene. 2007; 26:1459–1467
35. Zhang H, Meng F, Liu G, et al. Forkhead transcription factor foxq1 promotes epithelial-mesenchymal transition and breast cancer metastasis. Cancer Res. 2011; 71:1292–1301
36. Thompson EW, Williams ED. EMT and MET in carcinoma—clinical observations, regulatory pathways and new models. Clin Exp Metastasis. 2008; 25:591–592
37. Chen J, Han Q, Pei D. EMT and MET as paradigms for cell fate switching. J Mol Cell Biol. 2012; 4:66–69
38. Miller-Hodges E, Hohenstein P. WT1 in disease: shifting the epithelial-mesenchymal balance. J Pathol. 2012; 226:229–240
39. Geiger B, Ayalon O. Cadherins. Annu Rev Cell Biol. 1992; 8:307–332
40. Takeichi M. Cadherins in cancer: implications for invasion and metastasis. Curr Opin Cell Biol. 1993; 5:806–811
41. Pinheiro H, Carvalho J, Oliveira P, et al. Transcription initiation arising from E-cadherin/CDH1 intron2: a novel protein isoform that increases gastric cancer cell invasion and angiogenesis. Hum Mol Genet. 2012; 21:4253–4269
42. Knudson AG Jr, Strong LC. Mutation and cancer: a model for Wilms’ tumor of the kidney. J Natl Cancer Inst. 1972; 48:313–324