Suppression of Intestinal Polyposis in Mdr1-Deficient Apcmin/ Mice1

[CANCER RESEARCH 63, 895–901, March 1, 2003] Advances in Brief

Suppression of Intestinal Polyposis in Mdr1-deficient ApcMin/؉ Mice1

Tesshi Yamada,2 Yasuharu Mori, Reiko Hayashi, Mizuho Takada, Yoshinori Ino, Yasuyoshi Naishiro, Tadashi Kondo, and Setsuo Hirohashi Cancer Proteomics Project, National Cancer Center Research Institute, Tokyo 104-0045, Japan

Abstract functional involvement of the MDR1 gene in intestinal tumorigenesis. We report suppression of intestinal tumorigenesis in ApcMin/ϩ mice ␤ Aberrant transactivation of a certain set of target genes by the -catenin lacking functional Mdr1 genes. and T-cell factor/lymphoid enhancer factor complex has been considered crucial for the initiation of intestinal tumorigenesis. The human multidrug Materials and Methods resistance (MDR)1 (ABCB1) gene contains multiple ␤-catenin–T-cell factor4- binding elements in its promoter and is one of the immediate targets of the Antibodies, Immunoblot Analyses, Immunohistochemistry, and Immu- complex. In the current study, we have further substantiated the biological nofluorescence Microscopy. Anti-␤-catenin monoclonal antibody (clone 14) involvement of MDR1 in intestinal tumorigenesis based on the following was purchased from BD Transduction Laboratories (San Diego, CA). Anti-␤- evidence: (a) aberrant induction of the Mdr1a (Abcb1a) gene product, P- catenin (C-18), anti-MDR (C-19 and H-241), and anti-actin (C-11) polyclonal glycoprotein, associated with nuclear accumulation of the ␤-catenin protein, antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, was observed even in nascent microscopic adenomas of Min mice; (b) Mdr1- CA). Immunoblot analyses and immunohistochemistry were performed essen- ؉ ؊ ؊ deficient Min (ApcMin/ Mdr1a/b / ) mice developed significantly fewer intes- tially as described previously (8). For immunofluorescence microscopy, cells ؉ ؉ ؉ tinal polyps than did ApcMin/ Mdr1a/b / mice; and (c) Inhibitors of P- were grown on type-I-collagen-coated coverslips (Asahi Technoglass, Tokyo, glycoprotein, verapamil, and cyclosporin A had a suppressive effect on the in Japan), fixed with 10% buffered formalin, and incubated with anti-␤-catenin vitro polypoid growth of IEC6 expressing stabilized (⌬N89) ␤-catenin protein. monoclonal antibody overnight. After incubation with Alexa Fluor 488-labeled Inhibitors of P-glycoprotein may be included in a novel class of chemopre- goat antimouse IgG (Molecular Probes, Eugene, OR), the coverslips were ventive agents against colorectal carcinogenesis. inspected with a laser scanning confocal microscope (Bio-Rad, Hercules, CA). Reverse Transcription-PCR. Total RNA was prepared from adrenal Introduction gland, normal small intestine, and polyp tissues of Min mice (C57BL/6J- ApcMin/ϩ) with TRIzol (Invitrogen, Carlsbad, CA). A 1-␮g sample of DNase- Mutational inactivation of the tumor suppressor gene APC3 is the I-treated total RNA was reverse transcribed and amplified by PCR. The earliest and most frequent genetic event in colorectal carcinogenesis following PCR primers were used: for Mdr1a, 5Ј-TGAAGCTTGTA- ␤ (1). Its inactivation causes IECs to accumulate cytoplasmic -catenin AATCTAAGGAT-3Ј and 5Ј-TGAAAGAAATGATCCCAAGGAT-3Ј; for protein and allows the formation of complexes between ␤-catenin and Mdr1b, 5Ј-TAATGCTTATGGATCCCAGAGT-3Ј and 5Ј-TTTCATGGTCA- TCF/LEF family transcription factors. Aberrant transactivation of a TCATCTCTTGA-3Ј; for Mdr3, 5Ј-ATTTGAAGTTGAGCTAAGTGAC-3Ј certain set of target genes by the ␤-catenin and TCF4/LEF complexes and 5Ј-ATAGCTATCATCTCAGACAGGA-3Ј; for cytokeratin 19, 5Ј-TT- has been considered crucial for the initiation of intestinal carcinogen- GAGATTGAGCTGCAGTCCCAGCT-3Ј and 5Ј-TTCCCAGGGGAGTCTC- esis, and several downstream targets of the complex have already been GCTGGTAGC-3Ј; for GAPDH (glyceraldehyde-3 phosphate dehydrogenase), identified, including c-myc, PPAR-␦, cyclin-D1, TCF-1, c-jun, fra-1, rodent GAPDH forward and reverse primers (Applied Biosystems (Foster City, matrylisin (MMP-7), gastrin, ectodermal-neural cortex 1, laminin ␥2, CA). PCR products were analyzed by agarose gel electrophoresis and ethidium bromide staining. ITF-2, and axin2 (Axil) (2–7). We demonstrated previously that the ␤ Mdr1 and Apc Compound Mutant Mice. Animal experiments were per- human MDR1 (ABCB1) gene contains multiple -catenin–TCF4- formed in accordance with the guidelines of the National Cancer Center binding elements in its promoter and is an immediate target of the Research Institute (Tokyo, Japan). C57BL/6J male Min (ApcMin/ϩ) mice (12) complex (8). These target genes are likely to be important mediators were obtained from The Jackson Laboratory (Bar Harbor, ME), and FVB of intestinal carcinogenesis and good candidates for chemoprevention female Mdr1a/b double knockout mice (Mdr1aϪ/Ϫ/Mdr1bϪ/Ϫ; Ref. 13) were of colorectal cancer by molecular targeting. However, because with a obtained from Taconic Farms (Germantown, NY). The genotyping of the few exceptions (3, 9–11) the functional connection of these down- animals is described in Fig. 2 and its legends. N6F1 pups were housed under stream target genes with intestinal carcinogenesis has remained un- identical conditions and sacrificed at 15 weeks of age. The gut was filled with explored, in this study, we adopted a genetic approach to clarify the 10% buffered formalin via the anus immediately after sacrifice and then opened longitudinally. After overnight fixation, specimens were stained briefly with 0.5% methylene blue. The numbers and major diameters of polyps in the Received 10/24/02; accepted 1/20/03. small and large intestine were measured in the ϫ20 power field of a dissecting The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with microscope (Nikon, Tokyo, Japan). Formalin-fixed, paraffin-embedded sec- 18 U.S.C. Section 1734 solely to indicate this fact. tions were stained by standard H&E techniques. 1 Supported by grants from the Ministry of Health, Labor, and Welfare; the Ministry Cell Lines and Tetracycline-inducible Retroviral Expression. An im- of Education, Culture, Sports, Science and Technology, Japan; and the Program for mortalized rat small cell line, IEC6 (14), was obtained from the Riken Cell Promotion of Fundamental Studies in Health Sciences of the Organization for Pharma- ceutical Safety and Research of Japan. Y. M. and Y. N. are recipients of a Research Bank (Tsukuba, Japan) and cultivated in DMEM containing 5% Tet-system- Resident Fellowship from the Foundation for Promotion of Cancer Research. approved fetal bovine serum (BD Biosciences Clontech, Palo Alto, CA) and 4 2 To whom requests for reprints should be addressed, at Cancer Proteomics Project, ␮g/ml insulin (Invitrogen). A colorectal cancer cell line, DLD1, was obtained National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, from the Health Science Research Resources Bank (Osaka, Japan; Ref. 8). Japan. Phone: 81-3-3542-2511, extension 4054; Fax: 81-3-3547-5298; E-mail: [email protected]. We used the RevTet-Off gene expression system (BD Biosciences Clon- 3 Abbreviations used are: APC, adenomatous polyposis coli; MDR, multidrug resist- tech) to introduce the tetracycline-regulatory element, tTA, and subsequently ance; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IEC, intestinal epithelial cell; the Rev-TRE carrying AU1-tagged human ␤Ϫcatenin cDNA lacking the FAP, familial adenomatous polyposis; TCF, T-cell factor; LEF, lymphoid enhancer factor; NH2-terminal 89 amino acids (15), into IEC6 cells. Details of preparation of ABC, ATP-binding cassette; NSAID nonsteroidal anti-inflammatory drug; Min, multiple ␤ ⌬ intestinal neoplasia; Mom1, modifier of multiple intestinal neoplasia-1; Mom, modifier of the pRev-TRE -catenin N89 plasmid construct are available on request. multiple intestinal neoplasia; PAF, platelet-activating factor; Dox, doxycycline. Cells were infected for 24 h in the presence of 8 ␮g/ml Polybrene (Sigma, St. 895

Fig. 1. Increased expression of P-glycoprotein in intestinal adenoma of Min mice. Immunoperoxi- dase staining was performed on serial sections ex- posed to anti-␤-catenin (C-18; A, B, and E) and anti-P-glycoprotein (C-19; C, D, and F) polyclonal antibodies. A and B, expression of ␤-catenin protein in an adenomatous polyp of a Min mouse (C57BL/6J-ApcMin/ϩ). Original magnification: ϫ40 (A) and ϫ200 (B). C and D, expression of P-glycoprotein in the same polyp as in A and B. Original magnification: ϫ40 (C) and ϫ200 (D). E and F, ␤-catenin (E) and P-glycoprotein (F) expression in a univillous adenoma. Original magnification: ϫ100. G, mRNA expression of Mdr1a, Mdr1b, Mdr3, cytokeratin 19 (Cy19), and GAPDH in adrenal gland (Ad.), normal small intestinal (N), and adenomatous (T) tissue of Min mice. Cyto- keratin 19 was included to monitor the endodermal contribution (6). GAPDH was included to monitor the integrity of the cDNA templates.

Louis, MO) with conditioned medium containing ecotropic retroviral particles, Dual Luciferase Reporter Assay. We used a pair of luciferase reporter as described previously (16) and cloned by limiting dilution with uninfected constructs, TOP-FLASH and FOP-FLASH (Upstate, Placid, NY), to evaluate cells as feeder layers. TCF/LEF transcriptional activity (16). Cells were transiently transfected in To visualize cross-sections of polypoid foci, cells were cultured on type-I- triplicate with one of these luciferase reporters and phRL-TK (Promega, collagen (Koken, Tokyo, Japan)-coated PetriPERM dishes (Vivascience, Han- Madison, WI) by using FuGENE 6 transfection reagent (Roche, Mannheim, over, Germany) for 4 weeks. Glutaraldehyde and osmic acid-fixed cells were Germany). Luciferase activity was measured with the Dual-luciferase reporter dehydrated in a graded ethanol series and then embedded in LR-white resin. assay system (Promega), 72 h after transfection, with Renilla luciferase activity ␮ Thin sections (4–6 m) were cut and stained by standard toluidine blue as an internal control. techniques. Chemicals. Dox (Sigma) was dissolved in deionized water to a stock concentration of 1 mg/ml. Dox was added to the culture medium to a final Results and Discussion concentration of 0.01 ␮g/ml for maintenance and to 0.1 ␮g/ml for suppression Increased Expression of P-Glycoprotein in Intestinal Adenoma of induction. R(ϩ)-verapamil and cyclosporin A (Sigma) were dissolved in ethanol to obtain a stock concentration of 20 and 5 mM, respectively, and added of Min Mice. To substantiate the involvement of the MDR1 gene in to the culture medium in the range of 5–100 ␮M and 0.5–10 ␮M, respectively. intestinal tumorigenesis, we first examined the expression of Mdr1 in ϩ The culture media containing these drugs were replaced every 2 or 3 days. intestinal polyps of Min (C57BL/6J-ApcMin/ ) mice. Min mice have a 896

Fig. 2. Genotyping of compound mutant mice. A, genotyping of Mdr1a and Mdr1b genes by PCR. Genomic DNA was extracted from mouse tails. PCR was carried out using primer pairs, as described by Smit et al. (33). PCR products were analyzed by agarose gel electrophoresis and ethidium bromide staining. B, confirmation of genotypes of the Mdr1b gene. Long PCR was carried out with a pair of primers, 5Ј-TCTGAGGTTC- CGCTCAACTCAGAGCTACTTCC-3Ј and 5Ј- ACATGGCCATCTCTTCCTCCAGACTGCT- GTTGC-3Ј, flanking the targeted site of the Mdr1b gene. Targeted and wild-type alleles were distinguished by digestion with a restriction enzyme, XbaI. C, protein truncation test for genotyping of the Apc gene. A portion of the Apc gene containing codon 850 was amplified by PCR with a pair of primers, 5Ј-GGATCCTAATACGACTCACTAT- AGGGACAGACCACCATGGAAGACCAGG- AAAGCCTTGTGG-3Ј and 5Ј-CTGAACTTG- GACGCAGCTG-3Ј. The resulting fragments containing T7 RNA polymerase priming sites were transcribed and translated in vitro by using the TNT quick-coupled reticulocyte lysate system from Pro- 35 mega in the presence of L-[ S]methionine (Amer- sham, Amersham, England). Products were analyzed by SDS-PAGE and autoradiography. The wild-type and Min alleles were expected to yield calculated masses of Mr 42,400 and 17,700, respectively. D, confirmation of genotypes of the Apc gene by sequencing. PCR fragments amplified for protein truncation tests were directly sequenced with a primer, 5Ј-CACAAGCAGAATCTTTAT- GG-3Ј, as described previously (16). E, genotyping of Mom1 (the Pla2g2a gene) by PCR. PCR products were amplified with a pair of primers, 5Ј-GAGAGCTGACAGCATGAAGG-3Ј and 5Ј- CCGTTTCTGACAGAGGTTCTGGTT-3Ј. Mutant and wild-type alleles were distinguished by digestion with a restriction enzyme, BamHI. F, confirmation of genotypes of the Pla2g2a gene by sequencing. The above PCR fragments were directly sequenced by using a primer, 5Ј-GCG- CAGTTTGGGGAAAT-3Ј.

germ-line mutation at codon 850 of the Apc gene and spontaneously Mdr3 (Abcb4, P-glycoprotein 2) genes. Reverse transcription-PCR develop numerous adenomatous polyps in the small and large intes- using primers specific for Mdr1a, Mdr1b, and Mdr3, respectively tine (12), and for that reason, they are recognized as an appropriate (Fig. 1G), revealed exclusive expression of Mdr1a in the intestine of animal model of human FAP syndrome. Immunohistochemical anal- Min mice. Consistent with immunohistochemical analyses, Mdr1a ysis of intestinal adenomas of Min mice revealed that all of the mRNA was found to be up-regulated in tumor tissue when compared adenoma cells examined contained higher amounts of P-glycoprotein, with normal intestine. the Mdr gene product, than neighboring normal IECs (Fig. 1, C and Suppression of Intestinal Tumorigenesis in Mdr1a/b-deficient -D). The increased expression of P-glycoprotein was always associated ApcMin/؉ Mice. Next, we generated compound mutant mice by in with nuclear accumulation of ␤-catenin protein as determined by troducing a homozygous deletion mutation of the Mdr1a and Mdr1b staining serial sections with anti-␤-catenin antibody (Fig. 1, A and B) genes into the APCMin/ϩ heterozygotes to validate the functional and anti-P-glycoprotein antibody (Fig. 1, C and D). Aberrant expres- involvement of Mdr1 in ␤-catenin-mediated colorectal carcinogene- sion of P-glycoprotein was evident even in a nascent microadenoma sis. Although Mdr1b, a sister gene of Mdr1a in rodents, is not formed within a single villus (Fig. 1F), supporting the hypothesis that expressed in intestinal adenoma of Min mice (Fig. 1G), both the induction of the Mdr gene is an early event in adenoma development. murine Mdr1a and Mdr1b genes contain two consensus sequences of These results are consistent with our previous observations in human TCF/LEF binding, CTTTGA/TA/T, in their promoters (17, 18). To specimens (8). avoid compensation by unexpected transactivation of the Mdr1b gene, The anti-P-glycoprotein polyclonal antibody used for immunohis- we used Mdr1a and Mdr1b double-knockout mice, instead of Mdr1a tochemistry is known to react with the products of the Mdr1a single-knockout mice. No clear phenotypical abnormalities have been (Abcb1a, P-glycoprotein 3), Mdr1b (Abcb1b, P-glycoprotein 1), and reported in mutant mice lacking the Mdr1a and Mdr1b genes (13). 897

Fig. 3. Suppression of intestinal tumorigenesis in Mdr1a/b-deficient ApcMin/ϩ mice. A, representative macroscopic view of the ileum of ApcMin/ϩMdr1a/bϩ/ϩ and ApcMin/ϩMdr1a/bϪ/Ϫ N6F1 mice. Specimens were stained with 0.5% methylene blue. B, number of polyps in the small intestine of N6F1 mice. Each column represents the mean number of polyps per mouse, and each bar represents the SD (n ϭ 15 for ApcMin/ϩMdr1a/bϩ/ϩ males, n ϭ 12 for ApcMin/ϩMdr1a/bϪ/Ϫ males, n ϭ 9 for ApcMin/ϩMdr1a/bϩ/ϩ females, n ϭ 11 for ApcMin/ϩMdr1a/bϪ/Ϫ females). C, representative dissecting microscopic view of polyps in the small intestine of ApcMin/ϩMdr1a/bϩ/ϩ (left) and ApcMin/ϩMdr1a/bϪ/Ϫ (right) N6F1 mice. D, histological appearance of polyps in the small intestine of ApcMin/ϩMdr1a/bϩ/ϩ (left) and ApcMin/ϩMdr1a/bϪ/Ϫ (right) N6F1 mice (H&E staining). Note that both polyps have almost the same diameter (0.8 mm) and that the photographs were taken at the same magnification (ϫ100).

Mdr1a/b-knockout mice were generated in the FVB strain, as dinated N6 male ApcMin/ϩPla2g2aϪ/ϪMdr1a/bϩ/Ϫ mice and N6 fe- described previously (13). Different genetic backgrounds are thought male Apcϩ/ϩPla2g2aϪ/ϪMdr1a/bϩ/Ϫ mice were mated to produce to be capable of altering the severity of the Min phenotype, and one N6F1 pups. ApcMin/ϩ N6F1 mice were assessed for intestinal polyp- known modifier of the Min phenotype is the Mom1 locus, which osis according to their Mdr1a/b genotypes. Because of their close contains the secretory phospholipase A2 (Pla2g2a) gene (19). The physical distance, we never observed dissociation between the FVB strain carried only dominant wild-type alleles for the Pla2g2a Mdr1a and Mdr1b loci in any of the experiments.4 gene (ϩ/ϩ, considered to be resistant to the Min phenotype; Fig. 2, E The number of polyps in the small intestine of ApcMin/ϩMdr1a/bϪ/Ϫ and F), whereas Min (C57BL/6J-ApcMin/ϩ) mice carry only recessive mice was reduced to almost half the number in the control mutated alleles (Ϫ/Ϫ, sensitive to the Min phenotype). To replace ApcMin/ϩMdr1a/bϩ/ϩ mice (Fig.3, A and B; average 102.1–55.7/male Mom1 and the other possible unknown modifiers in the FVB strain mouse, P ϭ 0.0021; 147.1–74.4/female mouse, P ϭ 0.0041). The and unify the genetic background, female FVB-Mdr1a/bϪ/Ϫ mice ApcMin/ϩ female N6F1 mice tended to have more polyps than the males, were backcrossed repeatedly to male C57BL/6J-ApcMin/ϩ or C57BL/6J mice to the N6 generation. The resulting genetically coor- 4 A. H. Schinkel, personal communication. 898

Fig. 4. Suppressive effect of P-glycoprotein inhibitors on in vitro polypoid growth. A, immunoblot analysis of IEC6-TetOFF ␤-catenin ⌬N89 (left) and IEC6-TetOFF control (right). Cells were cultured for 24, 48, or 72 h in the presence of Dox [Dox (ϩ)] or in the absence of Dox [Dox (Ϫ)]. Endogenous native (␤-cat) and induced truncated (⌬N89) ␤-catenin proteins were detected with anti-␤-catenin monoclonal antibody. B, IEC6-TetOFF ␤-catenin ⌬N89 (left) and IEC6-TetOFF control (right) cells were transiently transfected with reporter constructs containing oligomerized optimal TCF/LEF motifs (TOP) or mutant motifs (FOP). Luciferase activity was measured after 72-h incubation with (blue) or without Dox (red). Columns represent mean arbitrary relative luciferase units, and bars represent the SD. C, immunofluorescence microscopy. ␤-catenin protein was detected in IEC6-TetOFF ␤-catenin ⌬N89 cells cultured for 4 days with (left) or without (right) Dox. D, phase-contrast microscopy showing the morphology of IEC6-TetOFF ␤-catenin ⌬N89 cells cultured for 4 weeks with (left) or without (right) Dox. Original magnification: ϫ100. E, vertical cross-sections of IEC6-TetOFF ␤-catenin ⌬N89 cells cultured for 4 weeks with (top) or without (bottom) Dox. F, immunoblot analyses of IEC6-TetOFF ␤-catenin ⌬N89 (⌬N89; left) and IEC6-TetOFF control (mock; right) cells cultured for 4 days with or without Dox. Blots were detected with anti-P-glycoprotein (H-241; top), anti-␤-catenin (clone 14; middle), and anti-actin (bottom) antibodies. G, phase-contrast microscopy showing the morphology of IEC6-TetOFF ␤-catenin ⌬N89 cells cultured for 4 weeks without Dox (with induction of the stabilized ␤-catenin protein) in the presence of 1.5 ␮M cyclosporin A (top)or40␮M verapamil (bottom). Original magnification: ϫ100. H, phase-contrast microscopy showing the morphology of DLD1 cells cultured for 2 weeks without (left) or with (right)60␮M verapamil. Original magnification: ϫ40.

implying the presence of a gender-related weak modifier(s), probably the MDR/TAP subfamily (21). The incomplete inhibition of intestinal derived from the FVB strain. There were no significant differences polyposis in the absence of the Mdr1a/b genes may be attributable between the Mdr1a/bϪ/Ϫ and Mdr1a/bϩ/ϩ mice in the major diameters or to compensation by other member(s) of the ABC transporter super- locations of the polyps, a finding consistent with the recently proposed family. “hit-and-run” theory of the role of mutated ␤-catenin, in which the In addition to the reduction of polyp numbers, we noted mor- ␤-catenin–TCF4 complex plays a major role only in the initiation of phological alterations in the polyps of Mdr1a/b-deficient mice turmorigenesis (20). A similar reduction in polyp number was observed (Fig. 3, C and D). The intestinal polyps of Mdr1a/bϪ/Ϫ mice in the colon of male N6F1 ApcMin/ϩMdr1a/bϪ/Ϫ mice (Mdr1a/bϩ/ϩ, 4.3 tended to be flat, and their top surface was centrally depressed and and Mdr1a/bϪ/Ϫ, 2.3/mouse on the average) but not in female mice often covered with necrotic material (Fig. 3C). Histological exam- (Mdr1a/bϩ/ϩ, 3.6 and Mdr1a/bϪ/Ϫ, 3.2). However, because of the gen- ination of the polyps (Fig. 3D) revealed detachment of the covering eral paucity of polyps in the colon, it would be premature to draw non-neoplastic epithelium. Similar morphology has been reported conclusions. in the polyps of Apc⌬716 mice deficient in prostaglandin receptor P-glycoprotein, encoded by the MDR1 gene, is a member of the EP2 or cyclooxygenase-2 (22, 23). An association between MDR1 superfamily of ABC transporters that transfer various molecules and expression of cyclooxygenase-2 has been documented recently across membranes. Human ABC genes are divided into eight distinct (24, 25). Cytosolic phospholipase A2 releases lyso-PAF, a precur- subfamilies: (a) ABC1;(b) MDR/TAP;(c) MRP;(d) ALD;(e) OABP; sor of PAF, from plasma membrane lipid layers as a coproduct of (f) GCN20;(g) White; and (h) ANSA, and the MDR1 gene belongs to arachidonic acid. PAF is one of the substrates of P-glycoprotein 899

(26). Overexpression of P-glycoprotein may directly or indirectly Acknowledgments influence the release of arachidonic acid and subsequent produc- tion of prostanoids. The precise mechanism of the suppression of We thank Dr. A. H. Schinkel (Netherlands Cancer Institute) for providing a protocol for PCR genotyping of the murine Mdr1a/b genes and Dr. K. Wak- intestinal polyposis in Mdr1-deficient mice, however, remains to abayashi (National Cancer Center Research Institute) for technical advice on be elucidated. polyp numbering. We also thank the contributions of Dr. K. Yanagihara for Establishment of a Rat IEC Clone Capable of Inducing Stabi- breeding the mice; S. Tamura for photographic techniques; F. Hasegawa, A. lized ␤-Catenin. Lastly, to clarify the role of Mdr1 in early stage Miura, and F. Kaiya for the preparation of microscopic specimens; and Y. intestinal tumorigenesis, we used a tetracycline-regulatory system Ishiyama in regard to secretarial assistance. to engineer an immortalized rat IEC line, IEC6. Native IEC6 cells maintain contact inhibition and anchorage dependency and do not References display tumorigenicity in nude mice (27). IEC6 cells are pheno- 1. Kinzler, K. W., and Vogelstein, B. Lessons from hereditary colorectal Cancer. Cell, typically similar to undifferentiated normal intestinal cryptic cells 87: 159–170, 1996. 2. Wong, N. A., and Pignatelli, M. Beta-catenin: a linchpin in colorectal carcinogenesis? and retain the potential to differentiate into various cell lineages, Am. J. Pathol., 160: 389–401, 2002. including absorptive enterocytes, goblet cells, Paneth cells, and 3. Koh, T. J., Bulitta, C. J., Fleming, J. V., Dockray, G. J., Varro, A., and Wang, T. C. endocrine cells (28). We used retroviral gene transfer and limiting Gastrin is a target of the beta-catenin/TCF-4 growth-signaling pathway in a model of ␤ ⌬ intestinal polyposis. J. Clin. Investig., 106: 533–539, 2000. dilution to establish a stable clone, IEC6-TetOFF -catenin N89, 4. Fujita, M., Furukawa, Y., Tsunoda, T., Tanaka, T., Ogawa, M., and Nakamura, Y. that is capable of inducing stabilized ␤-catenin protein (15) Up-regulation of the ectodermal-neural cortex 1 (ENC1) gene, a downstream target of the ␤-catenin/T-cell factor complex, in colorectal carcinomas. Cancer Res., 61: on removal of Dox from the culture medium (Fig. 4A, left) and a 7722–7726, 2001. mock clone, IEC6-TetOFF control (Fig. 4A, right). Induction of 5. Hlubek, F., Jung, A., Kotzor, N., Kirchner, T., and Brabletz, T. Expression of the ␤-catenin ⌬N89 caused TCF/LEF-specific gene trasactivation, as invasion factor laminin ␥2 in colorectal carcinomas is regulated by ␤-catenin. Cancer Res., 61: 8089–8093, 2001. revealed by luciferase reporter assays (Fig. 4B), and nuclear and 6. Kolligs, F. T., et al. ITF-2, a downstream target of the Wnt/TCF pathway, is activated cytoplasmic accumulation of the ␤-catenin protein, as revealed by in human cancers with beta-catenin defects and promotes neoplastic transformation. immunofluorescence microscopy (Fig. 4C). Cancer Cell, 1: 145–155, 2001. 7. Jho, E. H., Zhang, T., Domon, C., Joo, C. K., Freund, J. N., and Costantini, F. IEC6-TetOFF ␤-catenin ⌬N89 cells developed numerous pol- Wnt/beta-catenin/Tcf signaling induces the transcription of Axin2, a negative regu- ypoid foci when maintained confluent for 2– 4 weeks in the ab- lator of the signaling pathway. Mol. Cell. Biol., 22: 1172–1183, 2001. ␤ ⌬ 8. Yamada, T., Takaoka, A. S., Naishiro, Y., Hayashi, R., Maruyama, K., Maesawa, C., sence of Dox (with induction of the -catenin N89 protein; Ochiai, A., and Hirohashi, S. Transactivation of the multidrug resistance 1 gene by Fig. 4D, right). IEC6-TetOFF ␤-catenin ⌬N89 cells maintained a T-cell factor 4/␤-catenin complex in early colorectal carcinogenesis. Cancer Res., 60: flat monolayer in the presence of Dox (without induction of the 4761–4766, 2000. 9. Wilson, C. L., Heppner, K. J., Labosky, P. A., Hogan, B. L., and Matrisian, L. M. ␤-catenin ⌬N89 protein; Fig. 4D, left), similar to native IEC6 cells. Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrily- We reported previously that induction of a dominant-negative form sin. Proc. Natl. Acad. Sci. USA, 94: 1402–1407, 1997. 10. Roose, J., Huls, G., van Beest, M., Moerer, P., van der Horn, K., Goldschmeding, R., of TCF4 or sulindac, a clinically proven chemopreventive agent Logtenberg, T., and Clevers, H. Synergy between tumor suppressor APC and the against polyposis in FAP patients, suppressed the similar polypoid beta-catenin-Tcf4 target Tcf1. Science (Wash. DC), 285: 1923–1926, 1999. growth of a colorectal cancer cell line, DLD1 (16). The formation 11. Wilding, J., Straub, J., Bee, J., Churchman, M., Bodmer, W., Dickson, C., Tomlinson, I., and Ilyas, M. Cyclin D1 is not an essential target of ␤-catenin signaling during of polypoid foci in vitro is thus reconfirmed to reflect the adenom- intestinal tumorigenesis, but it may act as a modifier of disease severity in multiple atous proliferation of IECs. intestinal neoplasia (Min) mice. Cancer Res., 62: 4562–4565, 2002. 12. Su, L. K., Kinzler, K. W., Vogelstein, B., Preisinger, A. C., Moser, A. R., Luongo, Vertical cross-sections revealed that the polypoid foci were shaped C., Gouldmm, K. A., and Dove, W. F. Multiple intestinal neoplasia caused by a by cells migrating underneath the flat cell monolayer [Fig. 4E, bottom; mutation in the murine homolog of the APC gene. Science (Wash. DC), 256: Dox (Ϫ)], and those cells exhibited enlarged nuclei and nucleoli. 668– 670, 1992. 13. Schinkel, A. H., Smit, J. J., van Tellingen, O., Beijnen, J. H., Wagenaar, E., van Abnormal migration of IEC6 cells expressing the stabilized ␤-catenin Deemter, L., Mol, CA., van der Valk. M. A., Robanus-Maandag, E. C., and te Riele, protein may imitate the inward migration of IECs, which is believed H. P. Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc. Natl. Acad. Sci. USA, 94: 4028–4033, to give a rise to nascent microadenoma (29). 1997. Suppressive Effects of P-Glycoprotein Inhibitors on in Vitro 14. Quaroni, A., Wands, J., Trelstad, R. L., and Isselbacher, K. J. Epithelioid cell cultures Polypoid Growth. On induction of ␤-catenin ⌬N89 protein, IEC6 from rat small intestine. Characterization by morphologic and immunologic criteria. J. Cell Biol., 80: 248–265, 1979. cells increased their expression of P-glycoprotein (Fig. 4F), and we 15. Munemitsu, S., Albert, I., Rubinfeld, B., and Polakis, P. Deletion of an amino- examined the effect of two well-characterized inhibitors of P- terminal sequence beta-catenin in vivo and promotes hyperphosporylation of the adenomatous polyposis coli tumor suppressor protein. Mol. Cell. Biol., 16: 4088– glycoprotein, verapamil and cyclosporin A (30), on the polypoid 4094, 1996. growth of IEC6 cells expressing ␤-catenin ⌬N89 protein. The 16. Naishiro, Y., Yamada, T., Takaoka, A. S., Hayashi, R., Hasegawa, F., Imai, K., and Hirohashi, S. Restoration of epithelial cell polarity in a colorectal cancer cell line by addition of a minimum of 40 ␮M verapamil or 1.5 ␮M cyclosporin suppression of beta-catenin/T-cell factor 4-mediated gene transactivation. Cancer A to the culture medium almost completely suppressed polypoid Res., 61: 2751–2758, 2001. growth (Fig. 4G). Cyclosporin A (data not shown) and verapamil 17. Raymond, M., and Gros, P. Cell-specific activity of cis-acting regulatory elements in the promoter of the mouse multidrug resistance gene mdr1. Mol. Cell. Biol., 10: exerted similar suppressive effects on the polypoid growth of 6036–6040, 1990. DLD1 cells (Fig. 4H). 18. Hsu, S. I., Cohen, D., Kirschner, L. S., Lothstein, L., Hartstein, M., and Horwitz, S. B. Current chemoprevention by NSAIDs alone does not seem to be a Structural analysis of the mouse mdr1a (P-glycoprotein) promoter reveals the basis for differential transcript heterogeneity in mutidrug-resistant J774.2 cells. Mol. Cell. satisfactory means of suppressing tumorigenesis in FAP patients and Biol., 10: 3596–3606, 1990. groups at high risk of colorectal cancer (31). Several newer generation 19. MacPhee, M., Chepenik, K. P., Liddell, R. A., Nelson, K. K., Siracusa, L. D., and Buchberg, A. M. The secretory phospholipase A2 gene is a candidate for the Mom1 P-glycoprotein antagonists, such as PSC-833 (Valspodar) and MS- locus, a major modifier of ApcMin-induced intestinal neoplasia. Cell, 81: 957–966, 209, have been developed with the intention of reversing MDR in 1995. cancer cells (32), and these drugs may be diverted to chemopreventive 20. Chan, T. A., Wang, Z., Dang, L. H., Vogelstein, B., and Kinzler, K. W. Targeted inactivation of CTNNB1 reveals unexpected effects of beta-catenin mutation. Proc. uses against colorectal cancer. The addition of these P-glycoprotein Natl. Acad. Sci. USA, 99: 8265–8270, 2002. inhibitors may improve the chemopreventive efficacy of the current 21. Klein, I., Sarkadi, B., and Varadi, A. An inventory of the human ABC proteins. Biochim. Biophys. Acta Biochim. Biophys. Acta, 1461: 237–262, 1999. protocol using NSAIDs alone by mechanisms different from those of 22. Sonoshita, M., Takaku, K., Sasaki, N., Sugimoto, Y., Ushikubi, F., Narumiya, S., NSAIDs. Oshima, M., and Taketo, M. M. Acceleration of intestinal polyposis through pros- 900

taglandin receptor EP2 in Apc(Delta 716) knockout mice. Nat. Med., 9: 1048–1051, transformation and tumorigenesis of intestinal epithelial cells. Oncogene, 20: 4180– 2001. 4187, 2001. 23. Oshima, M., Dinchuk, J. E., Kargman, S. L., Oshima, H., Hancock, B., Kwong, E., 28. Kedinger, M., Simon-Assmann, P. M., Lacroix, B., Marxer. A., Hauri, H. P., and Trzaskos, J. M., Evans, J. F., and Taketo, M. M. Suppression of intestinal polyposis Haffen, K. Fetal gut mesenchyme induces differentiation of cultured intestinal in Apc delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell, 87: endodermal and crypt cells. Dev. Biol., 113: 474–483, 1986. 803–809, 1997. 29. Oshima, H., Oshima, M., Kobayashi, M., Tsutsumi, M., and Taketo, M. M. Morpho- ⌬ 24. Fantappie, O., Masini, E., Sardi, I., Raimondi, L., Bani, D., Solazzo, M., Vannacci, logical and molecular processes of polyp formation in Apc( 716) knockout mice. A., and Mazzanti, R. The MDR phenotype is associated with the expression of COX-2 Cancer Res., 57: 1644–1649, 1997. 30. Ling, V. Multidrug resistance: molecular mechanisms and clinical relevance. Cancer and iNOS in a human hepatocellular carcinoma cell line. Hepatology, 35: 843–852, Chemother. Pharmacol., 40: S3–S8, 1997. 2002. 31. Thun, M. J., Henley, S. J., and Patrono, C. Nonsteroidal anti-inflammatory drugs as 25. Patel, V. A., Dunn, M. J., and Sorokin, A. Regulation of MDR-1 (P-glycoprotein) by anticancer agents: mechanistic, pharmacologic, and clinical issues. J. Natl. Cancer cyclooxygenase-2. J. Biol. Chem., in press. Inst. (Bethesda), 94: 252–266, 2002. 26. Raggers, R. J., Vogels, I., and van Meer, G. Upregulation of the expression of 32. Naito, M., and Tsuruo, T. New multidrug-resistance-reversing drugs, MS-209 and endogenous Mdr1 P-glycoprotein enhances lipid translocation in MDCK cells trans- SDZ PSC 833. Cancer Chemother. Pharmacol., 40: S20–S24, 1997. fected with human MRP2. Biochem. J., 357: 859–865, 2002. 33. Smit, J. W., Huisman, M. T, van Tellingen, O., Wiltshire, H. R., and Schinkel, A. H. 27. Soubeyran, P., Haglund, K., Garcia, S., Barth, B. U., Iovanna, J., and Dikic, I. Absence of pharmacological blocking of placental P-glycoprotein profoundly in- Homeobox gene Cdx1 regulates Ras, Rho and PI3 kinase pathways leading to creases fetal drug exposure. J. Clin. Investig., 104: 1441–1447, 1999.

901

Tesshi Yamada, Yasuharu Mori, Reiko Hayashi, et al.

Cancer Res 2003;63:895-901.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/63/5/895

Cited articles This article cites 25 articles, 15 of which you can access for free at: http://cancerres.aacrjournals.org/content/63/5/895.full#ref-list-1

Citing articles This article has been cited by 10 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/63/5/895.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/63/5/895. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.