Clinicopathological Significance of E-cadherin and PCNA Expression in Hunman Non-small Cell Lung Cancer ======================================================================================================= * Jianwei Ma * Kai Fan * Yanli Zhang * Dawei Song * Jianmei Ma ## Abstract **OBJECTIVE** This study was designed to assess E-cadherin (E-cad) and proliferating cell nuclear antigen (PCNA) expression as well as their clinicopathological significance in hunman nonsmall cell lung cancers (NSCLCs). Possible molecular mechanisms of differentiation and metastasis of NSCLCs are discussed. **METHODS** Immunohistochemical and immunofluorescence double staining were performed to examine the expression of E-cad and PCNA in 68 primary NSCLCs cases. **RESULTS** The E-cad expression in squamous cell carcinomas and adenocarcinomas showed no significant difference. E-cad expression had a positive correlation with the histological-differentiated grade. A significant difference of E-cad expression was found between metastatic and non-metastatic groups. PCNA expression in squamous cell carcinomas and adenocarcinomas showed no significant difference. The PCNA expression had a reverse correlation with the histological-differentiated grade. A significant difference of PCNA expression was found between metastatic and non-metastatic groups. The E-cad and PCNA expression presented a reverse correlation. **CONCLUSION** E-cad expression is not associated with the histological type of NSCLC, but is associated with differentiation and metastasis of the cancer. Down-regulation of E-cad expression affects the proliferation of cancer cells. Conjoint analysis of E-cad and PCNA expression is a good way to evaluate tumor biological behavior. KEY WORDS: * NSCLC * E-cadherin * PCNA * immunohistochemistry ## Introduction Lung cancer is the leading cause of cancer deaths in both women and men, and its incidence has increased markedly in the past decade. Epithelial tumors, which present the majority of lung tumors, are classified primarily into two subgroups: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC is composed of squamous cell, adeno-, large cell and adeno-squamous carcinoma. As the 5-year survival for lung cancer patients is currently 13%, mainly due to metastasis, the devastation caused by this single cancer type deserves special attention. Cell adhesion is a key process because it is directly related to the differentiation, architecture and normal tissue development. It has been postulated that changes in cell-cell and cell-matrix interactions are responsible for cancer cell transgression through normal tissue boundaries and their dispersion to distant sites[1]. The interactions among cells or among cells and their substrate are mediated by adhesion molecules, a large number of which are involved in intercellular adhesion of epithelial and tumor cells. Among cadherins, E-cadherin (E-cad) is the most important protein. E-cad, a 120-kDa transmembrane glycoprotein, is a calcium-dependent cell adhesion molecule that is expressed in epithelial tissues involved in formation and maintenance of the histoarchitecture. Reduced/absent expression of E-cad has been found in a variety of human carcinomas[2]. Even though recent immunohistochemical studies have shown that down-regulation of E-cad is associated with lung cancer, the relationship between the expression of E-cad and the histological type, differentiation, as well as proliferation degree of NSCLC is less understood. This study was designed to determine the relationship between E-cad expression and the histological type, the differentiated level as well as metastasis of NSCLC. In addition, conjoint detection of the expression of proliferating cell nuclear antigen (PCNA) was performed to elucidate the possible molecular mechanism of how E-cad affects NSCLCs differentiation and metastasis. ## Patients and Methods ### Patients Sixty-eight primary NSCLCs patients underwent surgical resection between 2004 and 2006 at the Hospital affiliated with Dalian University. The clinicopathological characteristics are shown in Table 1. View this table: [Table 1.](http://www.cancerbiomed.org/content/5/2/87/T1) Table 1. Summary of the clinicopathological characteristics of the lung cancer patients. ### Immunohistochemical analysis and image analysis Paraffin sections, 4 μm thick, were deparaffinized and rehydrated before being washed thoroughly in phosphate buffered saline (PBS, pH 7.2~7.6). Microwave heating in 10 mmol/L sodium citrate buffer (pH 6.0) for 6 min was used to retrieve the antigen followed by thorough PBS washing. Endogenous peroxidase was blocked by 0.3% hydrogen peroxidase in methanol for 30 min. The sections were then washed in PBS for 3~5 min. After blocking of nonspecific antibody-binding sites, the sections were incubated at 4°C overnight with mouse E-cad antibody (1:1000, BD Transduction Laboratories, US)as the primary antibody. After PBS washing, the sections were incubated with the secondary antibody, biotinylated anti-mouse immunoglobulin for 1 h, and then with avidin-biotin complex (ABC) for 1 h at room temperatures[3]. Antibody binding was identified with diaminobenzidine tetrahydrochloride (DAB) as the chromogenic substrate. Finally, the nuclei were counterstained with hematoxylin and dehydrated with an ascending series of alcohols, cleared through xylene and mounted. Immunostaining for the PCNA antigen was performed in a similar manner. Mouse PCNA antibody (1:1000; Vectorlaboratory, US) was used. For fluorescence immunohistochemistry double staining, rabbit E-cad antibody (1:200, Dako, US) and mouse PCNA antibody (1:1000, Vectorlaboratory, US) were mixed and used as the primary antibodies, then mixed secondary antibodies(antimouse and anti-rabbit) labeled with Alexa594 (red) and Alexa488 (green) (Invitrogen) were applied. Finally the sections were mounted with glycerin. After the immunohistochemical reaction, images were captured using identical exposure times and cam-era settings. We measured the sum of staining intensity in 5 fields of view for each sample for each antibody using on Image J Plus system, and the values used for statistical analysis. The independent *t*-test, Spearman correlation and Person’s correlation were determined by SPSS11.5 software package. Results were considered statistically significant at a *P* value of 0.05 or less. ## Results ### E-cad expression in NSCLC The expression of E-cad was mainly located on the cellular membrane and partly in the cytoplasm near the membrane. The positive signal of E-cad expression in adenocarcinomas was stronger than that in squamous cell carcinomas. However, the difference between these two types was not significant (*P* = 0.34). On the other hand, it was observed that E-cad expression had a positive correlation with pathological differentiation of NSCLC. In other words, the expression of E-cad in well-differentiated tumors was significantly stronger compared to poorly differentiated (*P* = 0.003, Fig.1). The E-cad expression in the non-metastastic group was significantly stronger than that in the metastatic group (*P* = 0.002, Table 2). ![Fig.1.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/5/2/87/F1.medium.gif) [Fig.1.](http://www.cancerbiomed.org/content/5/2/87/F1) Fig.1. Immunohistochemical distribution of E-cadherin in hunman lung cancers. The A and B sections are from squamous cell carcinomas, C and D are from adenocarcinomas. The A and B sections are both from well-differentiated squamous cell carcinomas. The expression of E-cadherin was very weak in A section which was associated with lymph node metastasis, and very strong in B from a patient without any metastasis. The C section is from well-differentiated adenocarcinoma and D is from a poorly differentiated one. The expression intensity of E-cadherin in section C is stronger compared with poorly differentiated (D). Bar is 20 μm. View this table: [Table 2.](http://www.cancerbiomed.org/content/5/2/87/T2) Table 2. The relationships between E-cad expression and clinicopathological factors. ### PCNA expression in NSCLC The results showed no significant difference in PCNA expression between NSCLCS squamous cell carcinomas and adenocarcinomas. Furthermore, it was found that the expression of PCNA in well-differentiated tumors was weaker, but stronger in poorly differentiated tumors (*P* = 0.018, Fig.2). It also was found that the PCNA expression in the metastastic group was significantly stronger compared to the non-metastastic group (*P* = 0.001, Table 3). The E-cad and the PCNA expressions showed an inverse correlation (*r* = -0.51, *P* = 0.002, Fig.3). ![Fig.2.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/5/2/87/F2.medium.gif) [Fig.2.](http://www.cancerbiomed.org/content/5/2/87/F2) Fig.2. Immunohistochemical distribution of PCNA in various differentiated human lung cancers. Section A is from a well-differentiated and B from poorly differentiated squamous cell cacinoma. In poorly differentiated squamous cell cacinomas (B), the expression of PCNA was more intense compared to well-differentiated tumors (A). Section C is from a well-differentiated and B from poorly differentiated adenocarcinoma. In poorly differentiated adenocarcinoma (D), the expression intensity of PCNA was more common and stronger compared to well-differentiated tumors (C). Bar is 20 μm. View this table: [Table 3.](http://www.cancerbiomed.org/content/5/2/87/T3) Table 3. The relationships between expression of PCNA and clinicopathological factors. ![Fig.3.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/5/2/87/F3.medium.gif) [Fig.3.](http://www.cancerbiomed.org/content/5/2/87/F3) Fig.3. The expression of E-cad (green) and PCNA (red) in well- and poorly differentiated hunman squamous cell cacinoma and adenocarcinoma of the lung by immunofl uorescent double staining. Sections A and B are from squamous cell carcinomas, C and D are from adenocarcinomas. A and D are from well-differentiaed, B and C are from poorly differentiated tumors. Both squamous cell cacinomas and adenocarcinoma, in well-differentiated tumors (A and D) show a strong expression of E-cad, but less expression of PCNA. Conversely, poorly differentiated tumors (B and C) showed strong and more PCNA expression and poor E-cad expression. Bar is 10 μm. ## Discussion E-cad expression has been widely investigated in many kinds of human cancers, such as those of the esophagus[4], stomach[5], colon[6], liver[7], and others[8,9]. In the past decade, several reports have dealt with the relationships between E-cad expression and proliferation, differentiation, metastasis, prognosis as well as other clinicopathological characteristics of lung cancers. However, correlations between E-cad expression and invasion/metastasis or survival are still controversial topics[10-12]. These inconsistent findings are thought to be attributable to the following reasons[13]: *i*) a majority of studies on lung cancer investigated NSCLC as a whole, but these included squamous cell carcinomas and adenocarcinomas in which average E-cad expression levels are different; *ii*) the expression of E-cad on cancer cell membranes has been evaluated using different criteria by different authors; *iii*) different antibodies against E-cad and different dilutions were used; *iv*) different ways of antigen retrieval or other staining procedures were employed, etc. To circumvent the first factor, for example, we analyzed the relationship between the E-cad expression and histological type. After making sure no significant difference was found between these two types, we compared the difference in E-cad expression between metastastic and non-metastastic cancers. To circumvent the second factor and avoid subjectivity, we measured the intensity of each sample for each antibody by using an Image J Plus system based on impersonal criteria. Furthermore, we used two types of E-cad antibodies in order to get more exact results. We found that there was a significant correlation between E-cad expression and the degree of cancer differentiation as well as metastasis. E-cad expression was significantly lower in the poorly differentiated and metastastic group compared to the well- and moderately differentiated and non-metastastic groups. Therefore, E-cad expression is a reliable indicator that can reflect the level of differentiation and metastasis of lung cancers. It is known that tumor growth and progression is closely related to activity of cellular proliferation. PCNA is a nuclear protein widely used for evaluation of tumor cell proliferation. The expression of PCNA has been found significantly correlated with metastasis and relapse in many kinds of cancers, such as those of colon, thyroid gland and stomach[14]. As for lung cancer, it is widely thought that PCNA has high expression in poorly differentiated cancers and with those cancers associated with metastasis[15-18]. Our results supported these conclusions. However, up to now, it has been unclear whether the expression of PCNA correlates with the pathological lung cancer type. Some researchers have suggested that PCNA expression is different in various phathological types[19,20], others proposed that PCNA had no relationship with the histological type of lung cancer, but only reflected the condition of cellular proliferation[21-23], Our results were consistent with the latter concept. In the present study, we also found that E-cad expression had an inverse correlation with the expression of PCNA. From Fig.3, we can understand that regardless of whether the tumor was a squamous cell carcinoma or adenocarcinoma, the samples with higher expression of E-cad had lower expression of PCNA and vice versa. These results indicated that low expression of E-cad is associated with tumor cell proliferation. St Croix et al.[24] considered that E-cad is not only a suppressor of metastasis, but also a kind of cell contact growth suppressor. Its ability to inhibit proliferation involves up-regulation of the cyclin-dependent kinase inhibitor p27. Therefore conjoint analysis of E-cad and PCNA expression is a reliable means to judge tumor biological behavior. In summary, the results of our present study indicate that the expression of E-cad has no significant correlation with the NSCLC histological type, but has a significant relationship with the differentiated grade and lymph node metastasis. Low expression of E-cad is associated with tumor cell proliferation. Conjoint analysis of E-cad and PCNA expression is a good way to evaluate tumor biological behavior. Further studies are necessary to elucidate the mechanism by which the down-regulation of E-cad expression is induced in lung cancer. * Received September 10, 2007. * Accepted January 9, 2008. * Copyright © 2008 by Chinese Anti–Cancer Association ## References 1. Dorudi S, Johuson JP, Leiti F, et al. E-cadherin expression in colorectal cancer. An immunocytochemical and in situ hybridization study. Am J Pathol 1993; 142: 981-988. [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7682766&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1993KY01400004&link_type=ISI) 2. Hirohashis S. Inactivation of E-cadherin-mediated cell adhesion system in human cancers. Am J Pathol 1998;153: 333-339. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/S0002-9440(10)65575-7&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9708792&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000075268900001&link_type=ISI) 3. Ma J, Tanaka KF, Yamada G, et al. Induced expression of cathepsins and cystatin C in a murine model of demyelination. Neurochem Res 2007; 32: 311-20. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/s11064-006-9183-y&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=17086443&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000244094900019&link_type=ISI) 4. Nakanishi Y, Ochiai A, Akimoto S, et al. Expression of E-cadherin, alpha-catenin, beta-catenin and plako-globin in esophageal carcinomas and its prognostic significance: immunohistochemical analysis of 96 lesions. Oncology 1997; 54: 158–165. [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9075789&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1997WN27000013&link_type=ISI) 5. Jawhari A, Jordan S, Poole S, et al. Abnormal immunoreactivity of the Ecadherin-catenin complex in gastric carcinoma: relationship with patient survival. Gastroenterology 1997; 112: 46–54. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/S0016-5085(97)70218-X&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8978342&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1997WA90500010&link_type=ISI) 6. Gofuku J, Shiozaki H, Tsujinaka T, et al. Expression of E-cadherin and alpha-catenin in patients with colorectal carcinoma. Correlation with cancer invasion and metastasis. Am J Clin Pathol 1999; 111: 29–37. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1093/ajcp/111.1.29&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9894451&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) 7. Garcia S, Martini F, De Micco C, et al. Immunoexpression of E-cadherin and beta-catenin correlates to survival of patients with hepatocellular carcinomas. Int J Oncol 1998; 12: 443–447. [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9458373&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) 8. Gunji N, Oda T, Todoroki T, et al. Pancreatic carcinoma:correlation between E-cadherin and alpha-catenin expression status and liver metastasis. Cancer 1998; 82: 1649–1656. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/(SICI)1097-0142(19980501)82:9<1649::AID-CNCR9>3.0.CO;2-5&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9576284&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000073232400009&link_type=ISI) 9. Kudo Y, Kitajima S, Ogawa I, et al. Invasion and metastasis of oral cancer cells require methylation of E-cadherin and/or degradation of membranous betacatenin. Clin Cancer Res 2004; 10: 5455–5463. [Abstract/FREE Full Text](http://www.cancerbiomed.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MTA6ImNsaW5jYW5yZXMiO3M6NToicmVzaWQiO3M6MTA6IjEwLzE2LzU0NTUiO3M6NDoiYXRvbSI7czoxNjoiL2NibS81LzIvODcuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 10. Bremnes RM, Veve R, Gabrielson E, et al. High-throughput tissue microarray analysis used to evaluate biology and prognostic significance of the E-cadherin pathway in non-small-cell lung cancer. J Clin Oncol 2002; 10: 2417–2428. 11. Ramasami S, Kerr KM, Chapman AD, et al. Expression of CD44v6 but not E-cadherin or beta-catenin influences prognosis in primary pulmonary adenocarcinoma. J Pathol 2000; 192: 427–432. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1096-9896(2000)9999:9999<::AID-PATH741>3.0.CO;2-Z&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=11113858&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000165700400002&link_type=ISI) 12. Smythe WR, Williams JP, Wheelock MJ, et al. Cadherin and catenin expression in normal bronchial epithelium and non-small cell lung cancer. Lung Cancer 1999; 24: 157–168. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/S0169-5002(99)00032-X&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=10460003&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000081710500002&link_type=ISI) 13. Liu D, Huang C, Kameyama K, et al. E-cadherin expression associated with differentiation and prognosis in patients with non-small cell lung cancer. Ann Thorac Surg 2001; 1: 949–954. 14. Ng IO, Lai ECS, Fan ST, et al. Prognostic significance of proliferating cell nuclear antigen expression in hepatocellular carcinoma. Cancer 1994; 73: 2268-2274. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(19940501)73:9<2268::AID-CNCR2820730906>3.0.CO;2-O&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7513246&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) 15. He J, Wang WY, Yao M. Significance and expression of CerB-2, p21 and PCNA in lung cancer. Zhejiang Cancer 1998; 4: 8-10 (Chinese). 16. Zhao T, Zhang YL, Li CD. A comparison of proliferating cell nuclear antigen immunostaining, nucleolar organizer region staining in lung cancer. Chin J Tuber Respir Dis 1993; 16: 342-429 (Chinese). 17. Wang EH, Liu GN, He MC. PCNA expression and Ag-NORs image analysis as factors in judging the prognosis of lung cancer. Chin J Pathol 1995; 24: 143-145. 18. Ogawa J, Tsurumi T, Yamada S, et al. Blood vessel invasion and expression of sialyl Lewis-x and proliferating cell nuclear antigen in stage I non-small-cell lung cancer. Relation to postoperative recurrence. Cancer 1994; 73: 1177-1183. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(19940215)73:4<1177::AID-CNCR2820730409>3.0.CO;2-0&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7906193&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) 19. Oyama T, Mitsudomi T, Mizoue T, et al. Proliferating cell nuclear antigen may be superior to argyrophilic nucleolar organizer regions in predicting shortened survival of patients with non-small-cell lung cancer. Surg Oncol 1995; 4: 83-89. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/S0960-7404(10)80011-8&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7551263&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1995RH51900003&link_type=ISI) 20. Kawai T, Suzuki M, Kono S, et al. Proliferating cell nuclear antigen and Ki-67 in lung carcinoma. Correlation with DNA flow cytometric analysis. Cancer 1994; 74: 2468-2475. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(19941101)74:9<2468::AID-CNCR2820740913>3.0.CO;2-X&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7923002&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1994PP48500012&link_type=ISI) 21. Castellano VM, Sotelo T, Ballestin C, et al. Analysis of proliferating cell nuclear antigen (PCNA) expression in 24 cases of primary non-small cell pulmonary carcinomas and correlation with survival. Arch Pronconeumol 1996; 33: 127-131. 22. Fontanini G, Macchiarini P, Pepe S, et al. The expression of proliferating cell nuclear antigen in paraffin sections of peripheral, node-negative non-small-cell lung cancer. Cancer 1992; 70: 1520-1527. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(19920915)70:6<1520::AID-CNCR2820700613>3.0.CO;2-K&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=1355398&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1992JN87500012&link_type=ISI) 23. Fujii M, Motoi M, Saeki H, et al. Prognostic significance of proliferating cell nuclear antigen (PCNA) expression in non-small celllung cancer. Acta Med Okayama 1993; 47: 103-108. [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8099466&link_type=MED&atom=%2Fcbm%2F5%2F2%2F87.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1993LA45200006&link_type=ISI) 24. St Croix B, Sheehan C, Rak JW, et al. E-cadherin-dependent growth suppression is mediated by the cylin-dependent kinase inhibitor p27KIPI. J Cell Biol 1998; 142: 557-571. [Abstract/FREE Full Text](http://www.cancerbiomed.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiamNiIjtzOjU6InJlc2lkIjtzOjk6IjE0Mi8yLzU1NyI7czo0OiJhdG9tIjtzOjE2OiIvY2JtLzUvMi84Ny5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=)