Dysregulation of E-cadherin by oncogenic ras in intestinal epithelial cells is blocked by inhibiting MAP kinase

The American Journal of Surgery 186 (2003) 426 – 430

Scientific paper

Dysregulation of E-cadherin by oncogenic Ras in intestinal epithelial cells is blocked by inhibiting MAP kinase Carl R. Schmidt, M.D.a, M. Kay Washington, M.D.a, Y. J. Gi, M.S.a, Robert J. Coffey, M.D.a, R. Daniel Beauchamp, M.D.a, A. Scott Pearson, M.D.a,b,* a

Departments of Surgery, Pathology, Cell and Developmental Biology, and Cancer Biology, Vanderbilt University Medical Center, and Department of Veterans Affairs, Tennessee Valley Healthcare System, Nashville, TN, USA b Surgical Oncology, Vanderbilt-Ingram Cancer Center, 597 Preston Research Bldg., 2220 Pierce Ave., Nashville, TN 37232, USA Manuscript received June 2, 2003; revised manuscript July 4, 2003

Abstract Background: Mutations in oncogenic Ras contribute to colorectal tumorigenesis. Loss of the cell adhesion protein E-cadherin is associated with tumor invasion and metastasis. Methods: Expression of oncogenic Ras was induced in intestinal epithelial cells. Changes in cell morphology, E-cadherin protein expression, and E-cadherin localization were examined by light microscopy, Western blot, and immunofluorescence respectively. Expression of E-cadherin in human colorectal tumors was examined by immunohistochemistry. Results: Induction of oncogenic Ras results in an epithelial to mesenchymal transformation with loss of membranous E-cadherin expression and mis-localization to the cytoplasm. Removal of Ras stimulus or blockade of the MAP kinase pathway allowed reversion to a normal cellular phenotype and return of E-cadherin to the cell membrane. Loss of or decreased expression of E-cadherin was observed in seven of eight colorectal tumors. Conclusions: Oncogenic Ras contributes to malignant transformation and altered E-cadherin expression in intestinal epithelial cells. Similar dysregulation of E-cadherin is found in human colorectal tumors. Ras effects on E-cadherin are critical to malignant transformation in our in-vitro model and may be an important event in human colorectal tumors. © 2003 Excerpta Medica, Inc. All rights reserved. Keywords: Colorectal carcinoma; E-cadherin; Transformation; Ras

Colorectal cancer is a national and global healthcare problem. It is the second leading cause of cancer death and the fourth most commonly diagnosed cancer in the United States [1]. In the year 2000, nearly 1 million incident cases of colorectal cancer were diagnosed world-wide [2]. Men and women are at an equal lifetime risk of 6% [3] for developing the disease and the incidence of new cases has risen each year since 1975 [4]. The estimated annual cost to the nation for treatment of colorectal cancer is 4.8 billion dollars [5]. However, mortality rates of colorectal cancer have declined in recent decades, and improvements in screening and treatment have likely contributed to this encouraging trend [1]. The adenoma-carcinoma sequence of gene mutations in

* Corresponding author. Tel.: ⫹1-615-343-9090; fax: ⫹1-615-3537622. E-mail address: [email protected]

colorectal cancer has been well-characterized [6,7]. However, a more clear understanding of the role and significance of the oncogenes and tumor suppressor genes involved is needed. Mutations of the K-Ras gene are present in half of colorectal malignancies in which the oncogenic form promotes activation of downstream signaling targets [8]. Disruption of intercellular junctions allows loss of cellcell contact and is important for tumor invasion and metastasis [9]. Previous studies have demonstrated altered regulation of critical proteins at both the adherens junction and tight junction in colorectal cancer [10 –12]. E-cadherin is the major protein located at the adherens junction, and altered regulation of E-cadherin is associated with tumor cell invasion, metastasis, and a worse prognosis [13–17]. We have previously demonstrated an important contribution of oncogenic Ras to altered regulation of E-cadherin and ␤-catenin [12]. In this study, we demonstrate that overactive mutated Ras promotes transformation of intestinal

0002-9610/03/$ – see front matter © 2003 Excerpta Medica, Inc. All rights reserved. doi:10.1016/j.amjsurg.2003.07.004

C.R. Schmidt et al. / The American Journal of Surgery 186 (2003) 426 – 430

epithelial cells associated with altered regulation of E-cadherin. These effects are reversible upon removal of overactive Ras and can be prevented with blockade of the MAP kinase signaling pathway downstream of Ras. Lastly, we demonstrate abnormal E-cadherin expression in the majority of human colorectal tumors examined.

Methods Cell culture Rat intestinal epithelial (RIE-iRas) cells were constructed with an inducible mutated Ha-RasVal12 cDNA under control of the LacSwitch eukaryotic expression system (Stratagene, La Jolla, California) as previously described [18]. Cells were grown in a monolayer and maintained with Dulbecco’s modified Eagle medium (Invitrogen, Carlsbad, California) supplemented with 10% fetal bovine serum, penicillin/streptomycin, L-glutamine, 150 ␮g/mL Hygromycin B (Invitrogen), and 400 ␮g/mL G418 (Invitrogen). Expression of mutated Ha-Ras cDNA was induced by adding IPTG (isopropyl-1-thio-B-D-galactopyranoside; Fisher Scientific, Fair Lawn, New Jersey) at 5 mM concentration to cell media. Cell morphology was examined with standard light microscopy. The MAPK inhibitor U0126 (Promega, Madison, Wisconsin) was added to cell media at a concentration of 10 ␮M 1 hour prior to IPTG administration. Similarly, the PI3-kinase inhibitor LY294002 (Cell Signaling, Beverly, Massachusetts) was added at 10 ␮M concentration 1 hour prior to IPTG treatment. Western blot Cells were washed with PBS or PBS containing 0.1 mM Na3VO4 and 0.2 mM H2O2 (pMAPK group) for 2 minutes. Total protein was obtained using RIPA lysis buffer. Protein levels were quantified using a Bradford reagent protein assay (BioRad, Hercules, California) with bovine serum albumin standards. Equal protein amounts were denatured with SDS-PAGE and loaded onto SDS-polyacrylamide gels. Proteins were then transferred to PVDF (polyvinylidene difluoride; BioRad) membranes and probed with the following primary antibodies: mouse anti-E-cadherin (Transduction Labs, Lexington, Kentucky) at 1:1,000, mouse anti-pan-Ras (Oncogene, Boston, Massachusetts) at 1:1,000, mouse anti-␤-actin (Sigma, St. Louis, Missouri) at 1:5,000, rabbit anti-pMAPK (Promega) at 1:1,000, or rabbit anti-MAPK at 1:1,000 (Cell Signaling). Goat anti-mouseIgG-HRP (Santa Cruz, Santa Cruz, California) at 1:5,000 was used as the secondary antibody in all cases and membranes were developed using the Enhanced Chemiluminescence System (Amersham Biosciences, Piscataway, New Jersey).

427

Immunofluorescence Five ⫻ 104 RIE-iRas cells were placed for each condition on a cover slip and treated with IPTG for 0, 24, 48, or 72 hours. The cells were washed again with PBS and fixed with methanol for 20 minutes at ⫺20°C. Next, the cells were made permeable with 0.5% Triton X-100 in PBS for 10 minutes at room temperature. Nonspecific binding was blocked with 2% BSA in PBS for 30 minutes. The cells were treated with anti-mouse E-Cadherin (Transduction Labs) primary antibody at 1:200. Cells were washed three times with PBS for 10 minutes, and then bound primary antibodies were detected with FITC-labeled goat antimouse antibody (Jackson ImmunoResearch, West Grove, Pennsylvania) at 1:500 for 1 hour at room temperature. Cover slips were mounted with Vectashield H-1000 (Vector Labs, Burlingame, California) and cells visualized by standard fluorescent microscopy. Immunohistochemistry Human colorectal tumor and matched normal mucosa samples were obtained from the Vanderbilt Medical Center Pathology Department and stored at ⫺80°C. Consent was obtained from all patients and the study has been approved by the Vanderbilt Institutional Review Board. Ten-micron thick sections were placed on glass slides and stored at ⫺80°C. Prior to staining, slides were dried for 30 minutes at room temperature and then placed in acetone for 2 minutes for tissue fixation. Borders were marked around each sample using a hydrophobic pen and slides were placed in 0.3% hydrogen peroxide and methanol for 30 minutes for peroxidase quenching. Tissues were stained using the ABC Elite Mouse (Vector Labs) kit per the manufacturer’s instructions. The primary antibody, mouse anti-E-cadherin (Zymed, San Francisco, California), was used at 1:1,000 in PBS/1% BSA for 1 hour at room temperature. The chromagen DAB was applied and slides were counterstained with hematoxylin (Vector Labs) per manufacturer’s instructions. All histology and protein expression interpretation was confirmed by a board-certified pathologist, Dr. M. Kay Washington.

Results Induction of oncogenic Ras results in cellular transformation Examination by light microscopy confirmed previous findings that induction of oncogenic Ras in RIE-iRas cells results in epithelial to mesenchymal transformation (EMT; Fig. 1A). This is consistent with malignant transformation as previously demonstrated [12]. Induction of Ras mislocalizes E-cadherin from the cell membrane to the cytoplasm as observed by immunofluorescence (Fig. 1B) and downregu-

428

C.R. Schmidt et al. / The American Journal of Surgery 186 (2003) 426 – 430

Fig. 1. Dysregulation of E-cadherin by oncogenic Ras. A. Induction of oncogenic Ras in intestinal epithelial cells by treatment with IPTG over 72 hours results in an epithelilal to mesenchymal transformation by light microscopy. B. Induction of Ras causes mislocalization of E-cadherin from the cell membrane to the cytoplasm by immunofluorescence. C. Induction of Ras causes downregulation of E-cadherin expression by Western blot.

lates total E-cadherin protein as measured by Western blot (Fig 1C). Transformation by oncogenic Ras is reversible Induction of Ras for 72 hours in RIE-iRas cells results in maximal phenotypic transformation and mislocalization of E-cadherin. Removal of Ras stimulus over a subsequent seventy-two hour period results in reversion of cells to the original epithelial phenotype (Fig. 2A), relocalization of E-cadherin to the cell membrane from the cytoplasm (Fig. 2B), and return of E-cadherin protein levels to baseline (Fig. 2C). Transformation by oncogenic Ras is blocked by mitogen activated protein (MAP) kinase inhibition Because the effects of oncogenic Ras are reversible in this model, we hypothesized that inhibition of downstream Ras signaling blocks phenotypic conversion and alterations in E-cadherin. UO126, a pharmacologic inhibitor of the MAP kinase pathway, blocks the EMT induced by Ras (Fig. 3A). In addition, UO126 prevents mislocalization (Fig. 3B) and downregulation (Fig. 3C) of E-cadherin. Treatment with the PI3-Kinase inhibitor LY294002 failed to block E-cadherin downregulation (Fig. 3C) and could not reverse transformation nor E-cadherin mislocalization (data not shown).

Fig. 2. Dysregulation of E-cadherin by oncogenic Ras is reversible. A. Withdrawal of oncogenic Ras in intestinal epithelial cells by removal of IPTG over 72 hours results in a return to the normal epithelial phenotype. B. Withdrawal of Ras results in return of E-cadherin from the cytoplasm to the cell membrane. C. Withdrawal of Ras results in return of E-cadherin expression to baseline.

E-cadherin expression is lost in human colorectal carcinoma Normal mucosa from 8 patients with adjacent colorectal carcinoma displayed strong E-cadherin staining at epithelial cell membranes by immunohistochemistry (Fig. 4A). Complete loss of E-cadherin protein or mislocalization from the cell membrane was observed in 7 of 8 colorectal carcinomas in these patients (Fig. 4B). In addition, we examined 2 liver metastases and 1 lymph node metastasis and all displayed complete loss of E-cadherin from metastatic tumor cell membranes (Fig. 4C).

Comments Expression of oncogenic Ras is present in half of colorectal carcinomas. Alterations in cell junction proteins and loss of cell-cell contact likely play a key role in tumor cell invasion and metastasis. E-cadherin is a major protein at epithelial cell junctions, and altered E-cadherin function is present in colorectal carcinomas and is associated with late tumor stage. Kanazawa et al [16] demonstrated altered regulation and abnormal expression of E-cadherin protein in mucinous and poorly differentiated colorectal tumors, both of which carry a poor prognosis. This study demonstrates the importance of oncogenic Ras in malignant transformation and altered E-cadherin

C.R. Schmidt et al. / The American Journal of Surgery 186 (2003) 426 – 430

429

Fig. 4. E-cadherin immunostaining in human colorectal carcinomas. A. Normal mucosa from a patient with an adenocarcinoma in the sigmoid colon demonstrates strong E-cadherin staining at epithelial cell membranes by immunohistochemistry (arrows). B. Adenocarcinoma tissue from the same patient demonstrates loss of E-cadherin staining at the cell membrane of tumor cells. C. A liver metastasis from the same patient demonstrates loss of E-cadherin staining at metastatic tumor cell membranes. Fig. 3. Inhibition of MAP kinase blocks dysregulation of E-cadherin. A. Induction of oncogenic Ras in intestinal epithelial cells by treatment with IPTG over 72 hours results in morphologic conversion from an epithelial to mesenchymal phenotype. Treatment with the MAP kinase inhibitor UO126 prevents this effect of Ras. B. Induction of Ras results in mislocalization of E-cadherin from cell membrane to cytoplasm and this effect is also prevented by UO126. C. Induction of oncogenic Ras results in downregulation of E-cadherin expression by Western blot. Concomitant treatment with the MAP kinase inhibitor UO126 prevents E-cadherin downregulation by Ras, and this effect is not observed with the PI3 Kinase inhibitor LY294002. As expected, phosphorylation of MAPK is prevented by treatment with UO126.

function in intestinal epithelial cells. This effect is both reversible upon withdrawal of oncogenic Ras and preventable by blockade of the MAP kinase pathway. Similar dysregulation of E-cadherin occurs in human colorectal carcinoma as evidenced by mislocalization or loss of the protein in the majority of human colorectal tumors examined. We conclude that dysregulation of E-cadherin by oncogenic Ras is a central process in cellular transformation in our in-vitro model. In addition, altered regulation of Ecadherin is present in the majority of human colorectal carcinomas. The contribution of Ras to altered E-cadherin function may be important in colorectal carcinogenesis. As suggested by this study, pharmacologic blockade of Ecadherin loss in colorectal cancer warrants further investigation. Acknowledgments Supported by grants from the Vanderbilt-Ingram Cancer Center Development Fund and Vanderbilt Physician Scien-

tist Development Award (ASP); National Institutes of Health Grants CA 69457 and DK 52334, and VanderbiltIngram Cancer Center Support Grant CA 68485 (R.D.B.). References [1] Hawk ET, Limburg PJ, Viner JL. Epidemiology and prevention of colorectal cancer. Surg Clin North Am 2002;82:905– 41. [2] Boyle P, Leon ME. Epidemiology of colorectal cancer. Br Med Bull 2002;64:1–25. [3] Osias GL, Osias KB, Srinivasan R. Colorectal cancer in women: an equal opportunity disease. J Am Osteopath Assoc 2001;101:S7–12. [4] Boyle P, Langman JS. ABC of colorectal cancer: Epidemiology. BMJ 2000;321:805– 8. [5] Sandler RS, Everhart JE, Donowitz M, et al. The burden of selected digestive diseases in the United States. Gastroenterology 2002;122: 1500 –11. [6] Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319:525–32. [7] Fearon ER. Molecular genetic studies of the adenoma-carcinoma sequence. Adv Intern Med 1994;39:123– 47. [8] Grady WM, Markowitz SD. Genetic and epigenetic alterations in colon cancer. Annu Rev Genomics Hum Genet 2002;3:101–28. [9] Cavallaro U, Christofori G. Cell adhesion in tumor invasion and metastasis: loss of the glue is not enough. Biochim Biophys Acta 2001;1552:39 – 45. [10] Irby RB, Yeatman TJ. Increased Src activity disrupts cadherin/catenin-mediated homotypic adhesion in human colon cancer and transformed rodent cells. Cancer Res 2002;62:2669 –74. [11] Chiang JM, Chou YH, Chen TC, et al. Nuclear beta-catenin expression is closely related to ulcerative growth of colorectal carcinoma. Br J Cancer 2002;86:1124 –9. [12] Fujimoto K, Sheng H, Shao J, Beauchamp RD. Transforming growth factor-beta1 promotes invasiveness after cellular transformation with activated Ras in intestinal epithelial cells. Exp Cell Res 2001;266: 239 – 49.

430

C.R. Schmidt et al. / The American Journal of Surgery 186 (2003) 426 – 430

[13] Ikeguchi M, Taniguchi T, Makino M, Kaibara N. Reduced E-cadherin expression and enlargement of cancer nuclei strongly correlate with hematogenic metastasis in colorectal adenocarcinoma. Scand J Gastroenterol 2000;35:839 – 46. [14] Ikeguchi M, Makino M, Kaibara N. Clinical significance of E-cadherin-catenin complex expression in metastatic foci of colorectal carcinoma. J Surg Oncol 2001;77:201–7. [15] Kimura T, Tanaka S, Haruma K, et al. Clinical significance of MUC1 and E-cadherin expression, cellular proliferation, and angiogenesis at the deepest invasive portion of colorectal cancer. Int J Oncol 2000;16:55– 64.

[16] Kanazawa N, Oda T, Gunji N, et al. E-cadherin expression in the primary tumors and metastatic lymph nodes of poorly differentiated types of rectal cancer. Surg Today 2002;32:123– 8. [17] Brabletz T, Jung A, Reu S, et al. Variable beta-catenin expression in colorectal cancers indicates tumor progression driven by the tumor environment. Proc Natl Acad Sci USA 2001;98:10356 – 61. [18] Sheng H, Shao J, Dixon DA, et al. Transforming growth factor-beta1 enhances Ha-ras-induced expression of cyclooxygenase-2 in intestinal epithelial cells via stabilization of mRNA. J Biol Chem 2000; 275:6628 –35.