Differential expression of prostaglandin endoperoxide H synthase-2 ...

Carcinogenesis vol.20 no.4 pp.737–740, 1999

SHORT COMMUNICATION

Differential expression of prostaglandin endoperoxide H synthase-2 and formation of activated β-catenin–LEF-1 transcription complex in mouse colonic epithelial cells contrasting in Apc J.M.Mei*, N.G.Hord2,*, D.F.Winterstein1, S.P.Donald and J.M.Phang3 Laboratory of Nutritional and Molecular Regulation, Division of Basic Sciences, National Cancer Institute and 1Intramural Research Support Program, SAIC–Frederick, NCI–FCRDC, Frederick, MD 21702, USA 2Present address: Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, USA 3To whom correspondence should be addressed Email: [email protected]

Mutations in Apc underlie the intestinal lesions in familial adenomatous polyposis and are found in >85% of sporadic colon cancers. They are frequently associated with overexpression of prostaglandin endoperoxide H synthase-2 (PGHS-2) in colonic adenomas. It has been suggested that Apc mutations are linked mechanistically to increased PGHS-2 expression by elevated nuclear accumulation of β-catenin–Tcf-LEF transcription complex. In the present study, we show that PGHS-2 is differentially expressed in mouse colonic epithelial cells with distinct Apc status. Cells with a mutated Apc expressed markedly higher levels of PGHS-2 mRNA and protein and produced significantly more prostaglandin E2 than cells with normal Apc. Using electrophoretic mobility shift assays, we demonstrate that DNA–β-catenin–LEF-1 complex formation is differentially induced in these two cell lines in an Apc-dependent manner. Our data indicate that the differential induction of β-catenin–LEF-1 complex correlates closely with differential expression of PGHS-2. These findings support the hypothesis that the differential expression of PGHS2 is mediated through the proposed β-catenin/Tcf-LEF signaling pathway.

Epidemiological and intervention studies support the hypothesis that prostaglandin endoperoxide H synthase-2 (PGHS2) is associated with colorectal tumorigenesis (1,2). Moreover, high PGHS-2 expression in human colon cancer cells is closely associated with increased metastatic potential (3). Mutations in the tumor suppressor protein adenomatous polyposis coli (APC), on the other hand, have been demonstrated to play a role in the human inherited colorectal cancer susceptibility syndrome, familial adenomatous polyposis, and are found in >85% of colon tumors (4–7). Apc heterozygote mice, including transgenic Apc mice and Min (multiple intestinal neoplasia) mice, which carry germline Apc mutations, are highly susceptible to intestinal polyposis and adenocarcinoma formation (8,9). The biological consequences of the acquisition of Apc Abbreviations: APC, adenomatous polyposis coli; IFNγ, interferon-γ; LPS, lipopolysaccharide; PGE2, prostaglandin E2; PGHS-2, prostaglandin endoperoxide H synthase-2. © Oxford University Press

737

Downloaded from http://carcin.oxfordjournals.org/ by guest on May 30, 2013

*These two authors contributed equally to this study

heterozygosity in carcinogenesis have been linked to β-catenin localization and turnover (10–12). Loss of APC protein function results in dysregulation of β-catenin turnover (13–15) and nuclear accumulation of transcriptionally active β-catenin–TcfLEF complexes (16–18). While target genes for this Apc genotype-dependent transcription complex have not been fully identified, the PGHS-2 gene may be responsive to this transcription complex (19). PGHS-2 catalyzes the production of a host of bioactive arachidonate metabolites, including prostaglandin E2 (PGE2). This prostanoid has been shown to have anti-apoptotic activities (20,21) which may enhance neoplastic progression. Although a link between APC and β-catenin has been reported, their role in non-malignant, nontransformed colonic epithelial cells in response to external stimuli has not been demonstrated. Given the suggested connection between Apc genotype and the expression of PGHS-2, we sought a cultured cell model system consisting of colonic epithelial cells isolated from mice with similar genetic backgrounds but with different Apc status (a generous gift of Dr Robert Whitehead, Ludwig Institute for Cancer Research, Melbourne, Australia). This pair of nontransformed murine colon epithelial cell lines includes one that carries the ApcMin/1 mutation (22). These conditionally immortal cells are designated IMCE (ApcMin/1), derived from F1 hybrids resulting from the mating of ApcMin/1 and SV40LT antigen transgenic mice; YAMC (Apc1/1) are derived from an SV40LT antigen parental mouse (23). Since both YAMC and IMCE express the heat-labile SV40LT antigen that allows them to proliferate at 33°C, they revert to a non-transformed phenotype at the restrictive temperature of 39°C, at which proliferation of these cells ceases (22,23). The epithelial nature of these cells was demonstrated by staining with anti-keratin antisera. The genotype and expression of APC protein of these cell types have been confirmed by allele-specific PCR and by western immunoblotting, respectively (22–24). Furthermore, these cells have been used to demonstrate that the ApcMin/1 mutation in IMCE cells can cooperate with stably transfected oncogenic ras to produce the transformed, tumorigenic phenotype (e.g. growth in soft agar and tumor formation in athymic mice) (24). PGHS-2 expression is known to be responsive to inflammatory stimuli such as interferon-γ (IFNγ) and lipopolysaccharide (LPS). We first examined the expression of PGHS-2 induced by IFNγ and LPS in YAMC and IMCE cells monitored at both the mRNA and protein levels (Figure 1). After 12 h induction, PGHS-2 mRNA levels, assessed with a murine cDNA PGHS-2 probe (Oxford Biomedical Research, Oxford, MI), were markedly higher in IMCE cells than that in YAMC cells (Figure 1A). Interestingly, although PGHS-2 mRNA was not detected in unstimulated YAMC cells, it was consistently observed in unstimulated IMCE cells, indicating basal expression of PGHS-2 in IMCE cells in the absence of inflammatory stimuli. Protein levels of PGHS-2 were also examined using an

J.M.Mei et al.

anti-PGHS-2 monoclonal antibody (Transduction Laboratories, Lexington, KY). Following 24 h treatment, PGHS-2 levels were markedly increased whereas both PGHS-1 and actin levels, used as controls, remained unchanged (Figure 1B). Similar to the mRNA findings, PGHS-2 protein also showed basal expression in unstimulated IMCE cells. The induction of PGHS-2 was much higher in IMCE cells than in YAMC cells. Although Figure 1 only shows data on PGHS-2 mRNA and protein expression after 12 and 24 h induction, respectively, we have observed differential expression in YAMC and IMCE cells of PGHS-2 mRNA and protein with various durations of stimulation (data not shown). In situ PGHS-2 enzymatic activity was assessed by measuring the amount of PGE2 (Cayman Chemical, Ann Arbor, MI) accumulated in the medium after cells were stimulated for 24 h with IFNγ and LPS. Although basal PGE2 levels in YAMC were barely detectable, the levels of PGE2 in medium from unstimulated IMCE cells were twice those from YAMC cells (Figure 2). More dramatically, the induced generation of PGE2 in either YAMC or IMCE cells was significantly higher than that of their unstimulated controls (P , 0.01). The difference in PGE2 generation between stimulated YAMC and IMCE cells was markedly widened, indicating that PGE2 generation in IMCE cells is more responsive to inflammatory stimulation. Sulindac sulfide, known to inhibit cyclooxygenase activity of both PGHS-2 and PGHS-1, blocked PGE2 production in both YAMC and IMCE cells (Figure 2). Significantly, neither cell type, under either transforming or non-transforming conditions, forms tumors in athymic nude 738

mice or grows in soft agar. Only when transfected with the oncogenic form of the ras gene do these cells exhibit tumorigenic characteristics (24). Therefore, the present study using non-malignant cells differing in Apc status and inflammatory stimuli, i.e. IFNγ and LPS, directly addresses the association between ApcMin/1 mutation(s), PGHS-2 expression and environmental factors. These data are consistent with the hypothesis that Apc mutation(s) affects PGHS-2 expression (19). Our results also show that PGHS-2 expression increased under the influence of proinflammatory stimuli. IMCE cells showed a strong propensity to overexpress PGHS-2 and overproduce PGE2 under environmental stress, i.e. stimulation with cytokine and endotoxin. To ascertain whether the induced expression of PGHS-2 in either YAMC or IMCE cells could be associated with the proposed β-catenin/Tcf-LEF pathway, we performed EMSA using a 32P-labeled murine LEF-1 consensus sequence oligonucleotide as probe (59-CACCCTTTGAAGCTC-39) (Gibco BRL, Grand Island, NY) (Figure 3). DNA–β-catenin–LEF-1 complex formation was increased in both YAMC and IMCE cells by inflammatory stimuli, but more so in IMCE cells (Figure 3A). The difference between these two cell lines in the amount of DNA–β-catenin–LEF-1 complex is striking, demonstrating its differential regulation in YAMC and IMCE cells. The specificity of the complex was shown by competition with unlabeled LEF-1 probe (Figure 3A) but not with nonhomologous probe (data not shown). The addition of anti-βcatenin polyclonal antibody (Sigma, St Louis, MO) to the reaction mixture caused an antibody-specific supershifting of the complex in both YAMC and IMCE cells (Figure 3B and C). This is direct evidence showing the participation of β-catenin in this DNA-binding transcription complex with LEF-1. We have repeatedly observed that this supershift increased with increasing anti-β-catenin antibody concentration (Figure 3B). Furthermore, failure of rabbit serum IgG (Sigma) to cause a supershift in either cell line demonstrates the specificity of this finding (Figure 3C). Anti-E-cadherin antibody was also used as a control and failed to cause any supershift (data not


Fig. 1. Differential expression of PGHS-2 in YAMC and IMCE cells in response to IFNγ and LPS. Cells were cultured in 75 cm2 culture flasks coated with type I collagen at 5 µg/cm2 (Collaborative Biomedical Products, Bedford, MA) in RPMI 1640 medium supplemented with 5% neonatal calf serum, 1% ITS1 (6.25 mg/l insulin, 6.25 mg/l transferrin, 6.25 µg/l selenous acid, 5.35 mg/l linoleic acid and 1.25 g/l bovine serum albumin), 5 U/ml murine IFNγ at 33°C (22). Upon reaching confluence, they were transferred to serum-free medium for 48 h at 39°C before each experiment. (A) Induction of PGHS-2 mRNA by IFNγ (100 U/ml) and LPS (1 µg/ml) for 12 h. This is one representative northern blot from at least three experiments. (B) Expression of PGHS-2 protein induced by IFNγ and LPS for 24 h. Constitutively expressed PGHS-1 and actin were used as controls. The same blot was reprobed for both PGHS-1 and actin after stripping of anti-PGHS-2 antibody. This is one representative blot from at least five different experiments.

Fig. 2. Differential generation of PGE2 by YAMC and IMCE cells in response to IFNγ and LPS. Conditioned culture medium was collected after stimulation with IFNγ (100 U/ml) and LPS (1 µg/ml) for 24 h. Sulindac sulfide (10 µM) was used to block PGE2 production in both YAMC and IMCE cells by co-incubation. Data are means 6 SD of triplicate determinations and were analyzed by one-way ANOVA with the Duncan multiple comparison test. *P , 0.01, basal levels compared with induced levels of PGE2 within the same genotype group; **P , 0.01, induced levels of PGE2 compared between the two genotype groups.

Colonic epithelial cells contrasting in Apc

shown). Based on the data from Figure 3A, the DNA–βcatenin–LEF-1 complex in IMCE cells in response to IFNγ/ LPS is ~5-fold of that in YAMC cells. In order to achieve

Fig. 3. Analysis of differential DNA–β-catenin–LEF-1 complex formation and supershift in YAMC and IMCE cells by EMSA. YAMC and IMCE cells were stimulated with IFNγ (100 U/ml) and LPS (1 µg/ml) for 4 h. Nuclear extracts were collected and analyzed by EMSA. (A) Competition of DNA–β-catenin–LEF-1 complex formation by unlabeled probe. For EMSA 5 µg of nuclear proteins from each treatment of YAMC and IMCE cells was loaded onto the gel. This is one representative blot from at least four experiments. (B) Increasing anti-β-catenin antibody concentration caused increased amounts of supershift. For analysis, nuclear extract (5 µg protein) was loaded in each lane. Polyclonal anti-β-catenin antibody was used in the following dilutions from stock: 1:12.5, 1:25 and 1:50. This is one representative blot from at least three experiments. (C) Supershift by anti-β-catenin antibody but not by rabbit serum IgG. For EMSA nuclear extract (see text) from IFNγ/LPS-stimulated YAMC and IMCE cells (5 and 1 µg protein, respectively) were analyzed. Anti-β-catenin antibody and rabbit serum IgG were both used at 1:30 dilution from respective stock to achieve the same protein concentration. This is one representative blot from at least three experiments.

739


optimal separation and supershifting in IMCE cells, the amount of nuclear protein loaded onto the gel was adjusted to a 5:1 ratio between YAMC and IMCE cells (5 and 1 µg, respectively) (Figure 3C). In the face of similar amounts of complexes formed, the same amount of anti-β-catenin antibody caused a similar amount of supershift in both YAMC and IMCE cells (Figure 3C). Overall, these data clearly associate a β-catenin-containing DNA-binding protein or protein complex residing in the nuclear extracts of IFNγ/LPS-treated cells with the conditions known to produce PGHS-2 overexpression. This finding suggests that Apc mutations play a permissive and perhaps amplifying role for inducers of PGHS-2, such as IFNγ and LPS, frequently produced under inflammatory conditions. Furthermore, chronic infection and inflammation in diseases affecting the colon, such as chronic ulcerative colitis, predispose the subjects to higher cancer risk and increase the frequency of colon malignancy (25). This supports the notion that β-catenin–LEF1 complex acts as a crucial regulator of the differential expression of PGHS-2 mRNA, protein and PGE2 generation in response to inflammatory stimuli in non-malignant, nontransformed colonic epithelial cells contrasting in Apc status, such as YAMC and IMCE cells. The identification of β-catenin, often found to be bound with cadherins at the adherens junctions, as part of an important transcription complex with the DNA-binding proteins Tcf-LEF indicates an essential role for this protein in colon carcinogenesis (12,16–18). Although the target genes controlled by this transcription pathway have not been fully identified, evidence supports PGHS-2, frequently overexpressed in colon cancer cells, as being one of these genes (19). We found that the PGHS-2 59 regulatory region does, in fact, contain consensus LEF-1-binding motifs (data not shown). The data presented in this study support the hypothesis that ApcMin/1 mutation(s) causes PGHS-2 overexpression by increasing the level of β-catenin available to form DNAbinding complexes capable of increasing the expression of PGHS-2 mRNA. However, the regulatory mechanism(s) through which β-catenin plays a central role has not been completely elucidated. It has been suggested recently that superstabilization of β-catenin, an E-cadherin-binding protein involved in cell–cell adhesion and communication, by Apc mutation(s) is among the main factors causing enhanced PGHS-2 expression mediated through the β-catenin–Tcf-LEF heterodimeric tran-

J.M.Mei et al.

scription complex (16,17,19). It is now known that β-catenin is constantly degraded by GSK-3β and APC protein in the presence of other important mediators, such as axin, in normal mammalian cells (26). Therefore, nuclear β-catenin levels remain relatively low and β-catenin is normally concentrated and evenly distributed in the cytoplasm in healthy cells (10–13). Mutations of Apc, GSK-3β and/or β-catenin itself would enable β-catenin to elude the destruction process (16,17). Thus, up-regulation of β-catenin levels occurs, resulting in increased transcriptional activation, in concert with the DNAbinding proteins Tcf-LEF, of certain genes, such as PGHS-2. Our findings presented in this study provide experimental evidence for the connection between Apc mutation(s) and PGHS-2 inducibility in non-malignant and non-transformed cells in response to inflammatory stimuli. These data support the hypothesis that Apc is associated with PGHS-2 induction, through affecting either the nuclear β-catenin levels and/or the amount of β-catenin-containing DNA-binding complex formation.

References 1. DuBois,R.N., Radhika,A., Reddy,B.S. and Entingh,A.J. (1996) Increased cyclooxygenase-2 levels in carcinogen-induced rat colonic tumors. Gastroenterology, 110, 1259–1262. 2. Williams,C.S., Luongo,C., Radhika,A., Zhang,T., Lamps,L.W., Nanney,L.B., Beachamp,R.D. and DuBois,R.N. (1996) Elevated cyclooxygenase-2 levels in Min mouse adenomas. Gastroenterology, 111, 1134–1140. 3. Tsujii,M., Kawano,S. and DuBois,R.N. (1997) Cyclooxygenase-2 expression in human colon cancer cells increase metastatic potential. Proc. Natl Acad. Sci. USA, 94, 3336–3340. 4. Kinzler,K.W. and Vogelstein,B. (1996) Lessons from hereditary colorectal cancer. Cell, 87, 159–170. 5. Moser,A.R., Pitot,H.C. and Dove,W.F. (1989) A dominant mutation that predisposes to multiple intestinal neoplasia in the mouse. Science, 247, 322–324. 6. Jacoby,R.F., Marshall,D.J., Kailas,S., Schlack,S., Harms,B. and Love,R. (1995) Genetic instability associated with adenoma to carcinoma progression in hereditary nonpolyposis colon cancer. Gastroenterology, 109, 73–82. 7. Groden,J., Thliveris,A., Samowitz,W. et al. (1991) Identification and characterization of the familial adenomatous polyposis coli gene. Cell, 66, 589–600. 8. Oshima,M., Oshima,H., Kitagawa,K., Kobayashi,M., Itakura,C. and Taketo,M. (1995) Loss of APC heterozygosity and abnormal tissue building in nascent intestinal polyps in mice carrying a truncated Apc gene. Proc. Natl Acad. Sci. USA, 92, 4482–4486.

740

Received August 25, 1998; revised November 3, 1998; accepted December 1, 1998


Acknowledgements The authors acknowledge the National Cancer Institute for allocation of computing time and staff support at the Frederick Biomedical Supercomputing Center of the Frederick Cancer Research and Development Center.

9. Moser,A.R., Luongo,C., Gould,K.A., McNeley,M.K., Shoemaker,A.R. and Dove,W.F. (1995) Apc Min: a mouse model for intestinal and mammary tumorigenesis. Eur. J. Cancer, 31A, 1061–1064. 10. Aberle,H., Schwarts,H. and Kemler,R. (1996) Cadherin–catenin complex: protein interactions and their implications for cadherin function. J. Cell. Biol. Chem., 61, 514–523. 11. Behrens,J., Von Kries,J.P., Kuhl,M., Bruhn,L., Wedlich,D., Grosschedl,R. and Birchmeier,W. (1996) Functional interaction of β-catenin with the transcription factor LEF-1. Nature, 382, 638–642. 12. Huber,O., Korn,R., McLaughlin,J., Ohsugi,M., Herrmann,B.G. and Kemler,R. (1996) Nuclear localization of β-catenin by interaction with transcription factor LEF-1. Mech. Dev., 59, 3–10. 13. Munemitsu,S., Albert,I., Souza,B., Rubinfeld,B. and Polakis,P. (1995) Regulation of intracellular β-catenin levels by the adenomatous polyposiscoli (APC) tumor-suppressor protein. Proc. Natl Acad. Sci. USA, 92, 3046–3050. 14. Peifer,M. (1997) β-Catenin as oncogene: the smoking gun. Science, 275, 1752–1792. 15. Rubinfeld,B., Souza,B., Albert,I., Muller,O., Chamberlain,S.H., Masiarz,F.R., Munemitsu,S. and Polakis,P. (1993) Association of the Apc gene product with β-catenin. Science, 262, 1731–1734. 16. Korinek,V., Barker,N., Morin,P.J., Van Wichen,D., De Weger,R., Kinzler,K.W., Vogelstein,B. and Clevers,H. (1997) Constitutive transcriptional activation by a β-catenin–Tcf complex in Apc–/– colon carcinoma. Science, 275, 1784–1787. 17. Morin,P.J., Sparks,A.B., Krinek,V., Barker,N., Clevers,H., Vogelstein,B. and Kinzler,K.W. (1997) Activation of β-catenin-Tcf signaling in colon cancer by mutations in β-catenin or APC. Science, 275, 1787–1790. 18. Porfiri,E., Rubinfeld,B., Albert,I., Hovanes,K., Waterman,M. and Polakis,P. (1997) Induction of a β-catenin-LEF-1 complex by wnt-1 and transforming mutants of β-catenin. Oncogene, 15, 2833–2839. 19. Prescott,S.M. and White,R.L. (1996) Self-promotion? Intimate connections between APC and prostaglandin H synthase-2. Cell, 87, 783–786. 20. Tsujii,M. and DuBois,R.N. (1995) Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell, 83, 493–501. 21. Sheng,H., Shao,J., Morrow,J.D., Beauchamp,R.D. and DuBois,R.N. (1998) Modulation of apoptosis and Bcl-2 expression by prostaglandin E2 in human colon cancer cells. Cancer Res., 58, 362–366. 22. Whitehead,R.H. and Joseph,J.L. (1994) Derivation of conditionally immortalized cell lines containing the Min mutation from the normal colonic mucosa and other tissues of an ‘Immortalized’ Min hybrid. Epithelial Cell Biol., 3, 119–125. 23. Whitehead,R.H., VanEeden,P.E., Nobel,M.D., Ataliotis,P. and Jat,P.S. (1993) Establishment of conditionally immortalized epithelial cell lines from both colon and small intestine of adult H-2Kb-tsA58 transgenic mice. Proc. Natl Acad. Sci. USA, 90, 587–591. 24. D’Abaco,G.M., Whitehead,R.H. and Burgess,A.W. (1996) Synergy between Apc min and an activated ras mutation is sufficient to induce colon carcinomas. Mol. Cell. Biol., 3, 884–891. 25. Goldman,H. (1996) Significance and detection of dysplasia in chronic colitis. Cancer, 78, 2261–2263. 26. Sakanaka,C., Weiss,J.B. and Williams,L.T. (1998) Bridging of β-catenin and glycogen synthase kinase-3β by axin and inhibition of β-cateninmediated transcription. Proc. Natl Acad. Sci. USA, 95, 3020–3023.