Available online at www.sciencedirect.com
Digestive and Liver Disease 40 (2008) 260–266
Alimentary Tract
ER- expression in large bowel adenomas: Implications in colon carcinogenesis A. Di Leo a,∗,1 , M. Barone a,1 , E. Maiorano b , S. Tanzi a , D. Piscitelli b , S. Marangi a , K. Lofano a , E. Ierardi c , M. Principi a , A. Francavilla a a
Section of Gastroenterology, Department of Emergency and Organ Transplantation (D.E.T.O.), Bari, Italy b Department of Pathologic Anatomy, University of Bari, Bari, Italy c Section of Gastroenterology, Department of Medical Science, University of Foggia, Italy Received 17 July 2007; accepted 29 October 2007 Available online 18 December 2007
Abstract Background. A pivotal role of oestrogen receptor-beta has been suggested in colon carcinogenesis in humans. However, few data are available on oestrogen receptor-beta in colorectal pre-cancerous lesions. Aim. In the present study, we evaluated oestrogen receptor-beta expression and its possible correlation with proliferative activity and apoptosis in colorectal adenomas and normal colon tissue. Patients/methods. Adenomatous tissue from 25 patients with colonic polyps, and normal tissue from 25 controls were used. Oestrogen receptor-beta expression, colonocyte proliferation (expressed as PCNA positivity) and apoptosis were evaluated. Results. In adenomatous tissue, a significant reduction of oestrogen receptor-beta was observed compared to normal mucosa (10.1 ± 5.5% vs. 44.2 ± 13.7; p < 0.03), while the expression of oestrogen receptor-alpha remained unvaried. Cell proliferative activity significantly increased in adenomatous tissue compared to normal mucosa (59.3 ± 7.1 vs. 18.5 ± 8.8; p < 0.0001), doubling the PCNA/apoptosis ratio. An inverse correlation was found between oestrogen receptor-beta and PCNA expression in adenomas (r = −0.81), a datum confirmed by confocal microscopy evaluation. Conclusions. Our data demonstrate, for the first time, a significant reduction of oestrogen receptor-beta expression already in the precancerous phase of colon carcinogenesis. This suggests a role of selective oestrogen receptor-beta agonists in the prevention of colorectal cancer. © 2007 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: Colorectal cancer; Oestrogens receptors; Polyps intestinal
1. Introduction There is evidence of the ability of oestrogens to modulate proliferative activity not only in the classic oestrogenresponsive tissues [1] but also in organs and/or cells of other apparatuses [2–5]. They exert their biological activity by binding with two type of receptors: oestrogen receptor-alpha ∗ Corresponding author at: Sezione di Gastroenterologia, D.E.T.O., Universit`a degli Studi di Bari, Policlinico, piazza Giulio Cesare, 11, 70124 Bari, Italy. Tel.: +39 080 5478611; fax: +39 080 5592494. E-mail address:
[email protected] (A. Di Leo). 1 Both authors equally contributed to the realization of this paper.
(ER-␣), the prevalent form in the breast, bone, cardiovascular tissue, urogenital tract and central nervous system, and oestrogen receptor-beta (ER-), the prevalent form in the gut [6,7]. In 1987, using radiolabelled oestradiol, we demonstrated, the presence of ERs in colorectal cancer showing that well differentiated forms express higher levels of ERs as compared to undifferentiated colorectal cancers [8]. Subsequently, we demonstrated by immunoenzymatic assay that ER levels are lower in neoplastic mucosa than in normal surrounding tissues and that polyamine (polycationic compounds normally implicated in cell proliferation) reach higher levels in ERnegative colorectal carcinomas as compared to ER-positive
1590-8658/$30 © 2007 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2007.10.018
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266
colorectal carcinomas [9]. More recently, a pivotal role of ER has been suggested in preventing malignant transformation of colon epithelial cells in humans. In fact, ER- expression is significantly lower in colon adenocarcinoma cells than in normal colon epithelial cells and this reduction is correlated with the degree of tumour dedifferentiation [10]. In addition, decreased ER- mRNA levels have been associated with human carcino tumorigenesis in women [11]. Experimental studies have also supported the role of ER- in preventing malignant transformation of colon epithelial cells. In fact, in germinal adenomatous polyposis coli (APC) mutated mice (C57BL/6J-Min/+mice), ovariectomy increased the development of intestinal adenomas by 77%, while, 17-oestradiol replacement abolished this increase in the same animals. In addition, 17-oestradiol replacement was associated to a dramatic increase of ER- protein and mRNA levels [12]. Although ER expression has been widely investigated in colorectal cancers [10,11,13], few data are available about colorectal pre-cancerous lesions. The importance of this aspect comes from the fact that the development of adenocarcinoma mostly involves polyp formation [14]. Therefore, the demonstration of low ER- in the polyps would further support the role of this receptor in the pathogenesis of colorectal cancer, and reduced ER- expression might be considered as a true tumour-promoting condition. In the present study, we evaluated ER- expression and studied its possible correlation with proliferative activity and/or apoptosis in colorectal adenomas and normal colon tissue epithelial cells.
2. Patients and methods 2.1. Patients This study was conducted on patients admitted to our hospital to undergo colonoscopy. Adenomatous tissue samples were obtained from 25 patients (M/F: 13/12, aged 55–75 years, mean 65 years) found to be affected by colonic polyps and treated with endoscopic polypectomy. None of these patients with polyps had a familial history of colon cancer. Normal tissue samples, used as controls, were obtained from 25 subjects (M/F: 11/14, aged 52–73 years, mean 60 years) that underwent colonoscopy and biopsy, subsequently found to be affected by irritable bowel syndrome or haemorrhoids. Only adenomatous lesions from the left-sided colon were selected in order to make as homogeneous as possible our selection. In fact, there is evidence of a different pathogenetic mechanism in left- and right-sided colon cancer [15]. All histopathological samples were independently evaluated by two pathologists. This study was approved by the local ethical committee and informed consent was given by all patients.
261
2.2. Methods All colonic tissue samples were fixed in 10% buffered formalin and embedded in paraffin and serial sections were used for haematoxylin–eosin (H–E) and immunohistochemical evaluations or confocal microscopy studies. Tissue sections of 4 m were used for H–E and immunohistochemical preparation, and 7 m thick sections for confocal microscopy. 2.2.1. Immunohistochemistry 2.2.1.1. ER expression. ER expression was detected using a modification of the method described by Kostantinopoulos et al. [16]. Briefly, for our study we used anti-ER-␣ primary rabbit polyclonal antibody (Ab-16 Neomarkers, Fremont, CA) and anti-ER- primary mouse monoclonal antibody (anti-ER- 14C8 antibody, Novus Biologicals, California, USA) that recognized residues 1-153 of the human ER- N-terminal. Both antibodies were diluted 1:50 with phosphate-buffered saline (PBS) before use. Endogenous peroxidase was blocked by 0.3% H2 O2 for 15 min, followed by incubation in 5% normal goat serum in PBS for 30 min, at room temperature. After antigen retrieval by microwave irradiation in citric buffer at pH 6.0, slides were incubated with primary antibodies overnight at 4 ◦ C. ERs expression was assessed using a polymer-based visualization kit (EnVision, DAKO A/S, Glostrup, Denmark), according to the manufacturer’s instructions, using 3,3 -diaminobenzidine tetrahydrochloride (DAB, Sigma) as the chromogen and Harris haematoxylin for nuclear contrast. Sections from human breast cancer tissue were used as positive controls. Negative controls were processed by replacing the primary antibody with the incubation medium. 2.2.1.2. Colonocyte proliferative activity. Colonocyte proliferative activity was evaluated by the determination of proliferating cell nuclear antigen (PCNA) expression with anti-PCNA mouse monoclonal antibody (PCNA clone PC10, DAKO) as previously described [17]. 2.2.1.3. Apoptosis. Apoptosis was studied using an antihuman PARP rabbit polyclonal antibody (poly-ADP-ribose polymerase p85 fragment, Promega) as previously described [18]. The number of immunolabelled cells over the total of cells counted, i.e. the percent of labelled cells (labelling index = LI) for ER-␣ (ER-␣ LI), ER- (ER- LI), PCNA (PCNA LI) and PARP (PARP LI) in normal and adenomatous tissue were evaluated in a blinded protocol by two independent observers in 10 well-oriented crypts, in the case of normal mucosa, or in fields containing at least 1000 cells, in the case of adenomatous polyps. 2.2.2. Immunofluorescence For the study of tissue colocalization of ER- and PCNA in adenomas and normal tissues, tissue sections were treated as follows. After permeabilization with 5% Tween 20, followed
262
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266
by microwave irradiation at 750 W in citrate buffer (pH 6.0) and incubation for 30 min in blocking buffer (PBS pH 7.4, 5% goat serum), the slides were incubated overnight at 4 ◦ C with the anti-human ER- antibody (1:30 in blocking buffer) and the anti-PCNA antibody (FL-261, sc-7907, Santa Cruz, California, USA) (1:100 in blocking buffer). The slides were then washed for 2 × 5 min in PBS, incubated for 2 h with a mixture containing the two secondary antibodies diluted 1:100 (Alexa Fluor 488 goat anti-mouse antibody, and Cy5 goat anti-rabbit, Invitrogen-Molecular Probes, The Netherlands). Slides were washed with PBS and counterstained with TO-PRO 3 (Invitrogen-Molecular Probes) diluted 1:10,000 in PBS, mounted in Vectashield (Vector Laboratories, California, USA) and finally sealed with nail varnish. Negative control sections were prepared by omitting primary antibodies. The reactivity of these negative controls was consistent with the expected results. All sections were viewed under the Leica TCS SP2 (Leica, Germany) confocal laser scanning microscope using 40× and 63× objective lenses, with a sequential scan procedure during acquisition of the two double immunolabelling fluorophores. Confocal images were taken at 430-nm intervals through the z-axis of the section, covering a total of 7 m in depth. Images from individual optical planes and multiple serial optical sections were analysed, digitally recorded and stored as TIFF files in Adobe Photoshop software (Adobe Systems, California, USA).
2.3. Statistical analysis Statistical analysis was performed using Student’s t-test for unpaired data and the Pearson test was used to correlate ER- expression to epithelial proliferation.
3. Results The general characteristics of the polyps examined can be summarized as follows: most of them (80%) showed the presence of low-grade dysplasia [among these the prevalent form was the tubular type (85%)] and their dimension ranged from 7 to 22 mm (median value 14). Distribution of ER- in normal colonic mucosa and adenomatous tissue is shown in Fig. 1A and B, respectively. ER- immunoreactivity was considered positive only when a strong dark brown discoloration was detected within the nuclei of the epithelial cell. In normal mucosal samples, the majority of epithelial cells showed consistent nuclear immunoreactivity but occasional and weaker intra-cytoplasmic positivity was detectable as well (Fig. 1A). In tubular adenomas, a reduced number of epithelial cells with mild dysplasia exhibited nuclear ER- immunoreactivity. ER--positive epithelial cells were both isolated and grouped in small clusters, and were particularly evident at the bottom of glandular structures (Fig. 1B). As shown in Fig. 2, in normal mucosa, ER- immunoreactivity, expressed as LI, was significantly higher than ER-␣ LI (44.2 ± 13.7 vs. 8.1 ± 3.2; p < 0.001). In adenomatous tissue, a significant reduction of ER- LI was observed as compared to normal colonic mucosa (10.1 ± 5.5% vs. 44.2 ± 13.7; p < 0.03), while the expression of ER-␣ remained unchanged. When we evaluated the proliferative activity by PCNA LI expression, our findings demonstrated a significantly higher cell proliferative activity in adenomatous tissue as compared to normal mucosa (59.3 ± 7.1 vs. 18.5 ± 8.8, respectively; p < 0.0001) (Fig. 3). A different behaviour was observed when PARP LI was evaluated in the two types of tis-
Fig. 1. Immunohistochemical evaluation of ER- expression in normal colonic mucosa and adenomas. (A) Normal colonic mucosa in which most enterocytes, with the exception of some (black arrows), show consistent immunoreactivity for ER-. (B) Adenomatous tissue in which only cells clusters (black arrow) show nuclear positivity for ER- (avidin-biotin peroxidase anti-ER-; ×200).
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266
263
tion that an inverse relationship exists between ER- and PCNA expression. In fact, colonocytes showed elevated ER (Fig. 5A) and low PCNA immunolabelling (Fig. 5B) in normal mucosa as compared to adenomatous polyps (Fig. 5D and E, respectively). Moreover, superimposing these images of ER- and PCNA signals (A + B and D + E), the signals referring to ER- and PCNA were rarely observed in the same cells (Fig. 5C and F).
4. Discussion Fig. 2. Evaluation of ER-␣ and ER- labelling index in normal colonic mucosa and in adenomatous polyps. Values represent the mean ± S.D. obtained from all samples. *p < 0.001 by t-test. **p < 0.03 by t-test.
Fig. 3. Evaluation of PCNA and PARP labelling indexes in normal colonic mucosa and adenomas. Values represent the mean ± S.D. obtained from all samples. *p < 0.0001 by t-test.
sue (Fig. 3). Interestingly, the PCNA/PARP ratio increased from approximately 1, in the normal mucosa, to 2, in the adenomatous tissue. Finally, when we correlated ER- immunoreactivity, expressed as percentage of labelled cells, and proliferative activity, expressed as PCNA LI, an inverse correlation was found in adenomas (r = −0.81, p < 0.02) (Fig. 4). Fig. 5 shows the images obtained with confocal microscopy using double immunolabelling for ER- and PCNA. In this case, our findings not only confirmed the immunohistochemical data but provided a clear demonstra-
Fig. 4. Correlation between ER- expression and PCNA labelling index in adenomas. Values were obtained from all samples. p < 0.05.
Epidemiological studies on oral contraceptive and oestrogen replacement therapy have revealed a possible protective role of these ER-ligands for colon cancer development [18–20]. This beneficial effect seems to be specifically mediated by ER- [11–13], that is the prevalent ER form in the gut [6,7]. The data obtained in this study strongly support the hypothesis of a protective effect of ER--mediated signal transduction in the pathogenesis of colorectal cancer, demonstrating a significant reduction of ER- expression already in the pre-cancerous phase. In fact, ER- expression was significantly reduced in adenomatous tissues as compared to normal mucosa (Figs. 1, 2 and 5D). The reduction specifically regarded this type of oestrogen receptor without involving ER-␣ that remained unmodified (Fig. 2). Moreover, ER- expression was inversely correlated with the significant increase of proliferative activity observed in the adenomatous tissue (Fig. 4). Another interesting aspect revealed by our study is that while apoptosis remained unchanged, proliferative activity significantly increased in the adenomatous tissue (Fig. 3), a finding that explains the progressive increase in size of adenomas, a biological characteristic that we have already pointed out in other pre-cancerous gastrointestinal lesions [21]. Finally, the use of confocal microscopy clearly demonstrated two points: ER- expression inversely correlates with PCNA immunoreactivity. In fact, in colon adenomas not only did we find a remarkable reduction of ER- expression as compared to normal mucosa (Fig. 5D and A, respectively), but this signal was located in cells with low/absent expression of PCNA (Fig. 5F). All these data suggest that in polyps, i.e. pre-cancerous lesions at elevated risk of colorectal cancer development, the reduction of ER- levels could have a promoting effect on cancer development, through an increase of proliferation which is not counterbalanced by apoptosis. In addition, a persistently elevated proliferative activity is not only responsible for hyperplastic phenomena (polyp growth) but could facilitate tumour development also by enhancing the probability of genomic mutations [22]. What mechanism mediates the influence of ER- on colonocyte proliferative activity? Studies on breast and
264
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266
Fig. 5. Evaluation by confocal microscopy of ER- and PCNA immunolabelling in normal colonic mucosa and adenomas. Normal mucosa shows: (A) diffuse ER- distribution, (B) PCNA expression mostly in lower part of the glands. (C) Superimposed image of ER- and PCNA signals (×40 magnification). Adenomatous tissue shows: (D) lower ER- expression of cells; (E) diffuse immunolabelling for PCNA. (F) Superimposed image of ER- and PCNA signals (×40 magnification).
prostate carcinogenesis suggest an opposite role of ER-␣ and ER- in the proliferation and differentiation of target tissues [23,24], a hypothesis described by Weihua et al. [25] as the yin/yang relationship. ER- can form a heterodimer with ER-␣ to inhibit the ER-␣ mediated transcriptional activation of classic oestrogen-responsive elements. In fact, in MCF-7 cells, adenovirus-mediated expression of ER inhibits ER-␣-induced proliferation by repressing the transcription of some oncogenes (c-myc, cyclin D1 and cyclin A) and increasing the expression of tumour suppressor genes (p21 and p27), a combination of events that leads to a cell cycle arrest in the G1-G2 transition [26]. However, the fact that ER- does not inhibit some genes induced by ER-␣ implies that ER- may also inhibit cell proliferation by other mechanisms independent of ER-␣ [26]. There is evidence of other possible mechanisms that could mediate the protective effect of ER- on tumour development. It is known that the mechanism preserving the functional and structural integrity of the intestine is the
continuous renewal of the epithelium by cells that differentiate and migrate from the crypt to the luminal surface. Enterocyte migration is decreased by 25% in preneoplastic small intestinal mucosa of ApcMin/+ (Min/+) mouse, an animal model of adenomatous polyposis coli [27], while it is restored by chemopreventive doses of sulindac, oestradiol and coumestrol, a phyto-oestrogen with a potent -selective agonist activity [28,29]. In addition, we have recently demonstrated that pharmacological doses of silymarine, another ER- selective agonist [30], increase colonic epithelial cell migration in normal mice [31]. All these data suggest that this biological effect could also contribute to the anti tumour effect of ERs. In conclusion, our data not only confirm the involvement of ERs- in colorectal cancer but suggest a possible explanation for the protective effect of oestrogens on cancer development [29,30]. They also provided further support of the role of vegetable-rich diets in the prevention of bowel cancer, thanks to their high content of phyto-oestrogens (ER- selective agonists).
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266
Practice points • Clinical and experimental evidences support a protective role of oestrogen receptor-beta (ER-) in colon carcinogenesis. • The demonstration of ER- involvement also in pre-malignant conditions, such as colonic polyps formation, would further sustain the use of ER- selective agonists such as phyto-oestrogens (highly represented in vegetable-rich diet) in colon cancer prevention. • We demonstrated a reduction of ER- expression in colonic adenomas together with the presence of an inverse correlation between ER- and proliferative activity (also confirmed by the use of confocal microscopy). This suggests a protective role of ER- against tumour promoting mechanisms in pre-malignant stages of colon carcinogenesis.
Research agenda • Studies on tumour cell lines from breast cancer suggest that ER- would act by antagonizing the stimulatory effect of ER-␣ on cell proliferation. • However, ER- may inhibit cell proliferation also by ER-␣ independent mechanisms. • The exact molecular mechanisms that link ER- to colonocyte proliferative activity remain to be elucidated.
Conflict of interest statement None declared.
List of abbreviations APC, adenomatous polyposis coli; ER, oestrogen receptor; LI, labelling index; PARP, poly-ADPribose-polymerase; PCNA, proliferating cell nuclear antigen.
References [1] Yager JD, Liehr JG. Molecular mechanisms of estrogen carcinogenesis. Annu Rev Pharmacol Toxicol 1996;36:203–32.
265
[2] Fisher B, Gunduz N, Saffer EA, Zheng S. Relation of estrogen and its receptor to rat liver growth and regeneration. Cancer Res 1984;44:2410–5. [3] Francavilla A, Polimeno L, DiLeo A, Barone M, Ove P, Coetzee M, et al. The effect of estrogen and tamoxifen on hepatocyte proliferation in vivo and in vitro. Hepatology 1989;9:614–20. [4] Messa C, Russo F, Pricci M, Di Leo A. Epidermal growth factor and 17-estradiol effects on proliferation of a human gastric cancer cell line (AGS). Scand J Gastroenterol 2000;35:753–8. [5] Barone M, Ladisa R, Di Leo A, Spano D, Francioso D, Aglio V, et al. Estrogen-induced proliferation in cultured hepatocytes involves cyclin D1, p21(Cip1) and p27(Kip1). Dig Dis Sci 2006;51:580–6. [6] Gustafsson JA. Estrogen receptor : a new dimension in estrogen mechanism of action. J Endocrinol 1999;163:379–83. [7] Dechering K, Boersma C, Mosselman S. Estrogen receptors alpha and beta: two receptors of a kind? Current Med Chem 2000;7:561–76. [8] Francavilla A, Di Leo A, Polimeno L, Conte D, Barone M, Fanizza G, et al. Nuclear and cytosolic estrogen receptors in human colon carcinoma and in surrounding non cancerous colonic tissue. Gastroenterology 1987;93:1301–6. [9] Linsalata M, Russo F, Cavallini A, Berloco P, Di Leo A. Polyamines, diamine oxidase, and ornithine decarboxylase activity in colorectal cancer and in normal surrounding mucosa. Dis Colon Rectum 1993;36:662–7. [10] Foley EF, Jazaeri AA, Shupnik MA, Jazaeri O, Rice LW. Selective loss of estrogen receptor beta in malignant human colon. Cancer Res 2000;60:245–8. [11] Campbell-Thompson M, Lynch IJ, Bhardwaj B. Expression of estrogen receptor (ER) subtypes and ERbeta isoforms in colon cancer. Cancer Res 2001;61:632–40. [12] Weyant MJ, Carothers AM, Mahmoud NN, Bradlow L, Remoti H, Bilinski RT, et al. Reciprocal expression of ER␣ and ER is associated with estrogen-mediated modulation of intestinal tumorigenesis. Cancer Res 2001;61:2547–51. [13] Di Leo A, Messa C, Cavallini A, Linsalata M. Estrogens and colorectal cancer. Curr Drug Targets Immune Endocr Metabol Disord 2001:1– 12. [14] Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319:525–32. [15] Nasir A, Kaiser HE, Boulware D, Hakam A, Zhao H, Yeatman T, et al. Cyclooxygenase-2 expression in right- and left-sided colon cancer: a rationale for optimization of cyclooxygenase-2 inhibitor therapy. Clin Colorectal Cancer 2004;3:243–7. [16] Konstantinopoulos PA, Kominea A, Vandoros G, Sykiotis GP, Andricopoulos P, Varakis I, et al. Oestrogen receptor beta (ER) is abundantly expressed in normal colonic mucosa, but declines in colon adenocarcinoma paralleling the tumor’s dedifferentiation. Eur J Cancer 2003;39:1251–8. [17] Barone M, Berloco P, Ladisa R, Ierardi E, Caruso ML, Valentini AM, et al. Demonstration of a direct stimulatory effect of bile salts on rat colonic epithelial cell proliferation. Scand J Gastroenterol 2002;37:88–94. [18] Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risk and benefit of estrogen plus progestin in health postmenopausal women: principals results from the Women’s Health Initiative randomized controlled trial. J Am Med Assoc 2002;288:321–33. [19] Nelson H, Humphrey L, Nygren P, Teutsch S, Allan J. Postmenopausal hormone replacement therapy: a scientific review. J Am Med Assoc 2002;288:872–81. [20] Giardiello FM, Hylind LM, Trimbath JD, Hamilton SR, Romans KE, Cruz-Correa M, et al. Oral contraceptives and polyp regression in familial adenomatous polyposis. Gastroenterology 2005;128:1077–80. [21] Ierardi E, Di Leo A, Barone M, Marangi S, Burattini O, Panarese A, et al. Tumor necrosis factor alpha and apoptosis in Helicobacter pylori related progressive gastric damage a possible mechanism of immune
266
[22]
[23]
[24] [25] [26]
[27]
A. Di Leo et al. / Digestive and Liver Disease 40 (2008) 260–266 system involvement in epithelial turnover regulation. Immunopharmacol Immunotoxical 2003;25:203–11. Barone M, Maiorano E, Ladisa R, Cuomo R, Pece A, Berloco P, et al. Influence of ursodeoxycholate-enriched diet on liver tumor growth in HBV transgenic mice. Hepatology 2003;37:880–6. Roger P, Sahla ME, Makela S, Gustafsson JA, Baldet P, Rochefort H. Decreased expression of estrogen receptor beta protein in proliferative preinvasive mammary tumors. Cancer Res 2001;61:2537–41. Horvath LG, Henshall SM, Lee CS. Frequent loss of estrogen receptorbeta expression in prostate cancer. Cancer Res 2001;61:5331–5. Weihua Z, Andersson S, Cheng G, Simpson ER, Warner M, Gustafsson JA. Update on estrogen signaling. FEBS Lett 2003;546:17–24. Paruthiyil S, Parmar H, Kerekatte V, Cunha GR, Firestone GL, Leitman DC. Estrogen receptor beta inhibits human breast cancer cell proliferation and tumor formation by causing a G2 cell cycle arrest. Cancer Res 2004;64:423–8. Mahmoud N, Boolbol S, Bilinski R, Martucci C, Chadburn A, Bertagnolli M. Apc gene mutation is associated with a dominant negative effect upon intestinal cell migration. Cancer Res 1997;57:5045–50.
[28] Mahmoud N, Kucherlapati R, Bilinski R, Churchill M, Chadburn A, Bertagnolli M. Genotype-phenotype correlation in murine Apc mutation: differences in enterocyte migration and response to sulindac. Cancer Res 1999;59:353–9. [29] Javid SH, Moran AE, Carothers AM, Redston M, Bertagnolli M. Modulation of tumor formation and intestinal cell migration by estrogens in the Apc(Min/+) mouse model of colorectal cancer. Carcinogenesis 2005;26:587–95. [30] Seidlova-Wuttke D, Becker T, Christoffel V, Jarry H, Wuttke W. Silymarin is a selective estrogen receptor beta (ERbeta) agonist and has estrogenic effects in the metaphysis of the femur but no or antiestrogenic effects in the uterus of ovariectomized (ovx) rats. J Steroid Biochem Mol Biol 2003;86:179–88. [31] Di Leo A, Barone M, Margiotta M, Castellaneta A, Troiani L, Demarinis L, et al. Effect of selective agonist for the estrogen receptor  (Silymarin) upon colonic cell migration in normal mice. Dig Liv Dis 2006;38:S105.
