Estrogens and Colorectal Cancer

A. Di Leo*, C. Messa, A. Cavallini and M. Linsalata

Biochemistry Laboratory, I.R.C.C.S. “S. de Bellis”, Scientific Institute for Digestive Diseases, Via della Resistenza, I- 70013 Castellana G. (BA), Italy

*Address correspondence to this author at the Biochemistry Laboratory, I.R.C.C.S. “S. de Bellis”, Scientific Institute for Digestive Diseases, Via della Resistenza, I- 70013 Castellana G. (BA), Italy; e-mail: irccsbiochimica@libero.it

Abstract: In recent years, several lines of epidemiologic, clinical and experimental evidences have been reported showing that estrogen hormones may be involved in malignant colorectal tumors.

The sex differences in site-specific incidence, the increased incidence of colonic cancer in women with breast cancer, the protective effect of increasing parity and the reduced risk among women taking postmenopausal hormones, are all elements suggesting that sex hormones may play a role.

Male rats experimentally exposed to the carcinogen dimethylhydrazine, have twice the risk of developing colon cancer and significantly shorter survival times than their female counterparts. Along with the clinical, experimental and epidemiologic findings there are also biologic reasons why estrogen may be protective.

Most estrogen action appears to be exerted via the estrogen receptors (ERs) on target cells. ERs have been reported in several solid tumors including gastrointestinal neoplasms such as esophageal, gallbladder, gastric and colorectal cancer.

At the end of 1995, a second ER (ER-b) was cloned from the rat prostate cDNA library and subsequently, the human and mouse homologs. Its demonstration in normal and neoplastic human colorectal tissues and “in vitro” in colonic epithelial cells, has renewed interest in investigating the existence of two ER subtypes. The presence of two ERs could explain the selective actions of estrogens on different target tissues and, particularly, on the gastrointestinal tract.

Finally, our studies suggest that estrogens and their receptors play an important role in the growth and progression of colorectal tumors, by interacting with other molecules required for cell proliferation like growth factors and polyamines.

Epidemiological data

     Cancer of the colon and rectum is a major cause of cancer-associated morbidity and mortality in Europe and North America.

     Colorectal cancer is the fourth most common incident cause in United States and it is estimated that 130,200 new cases of cancer will occur in the coming year, with 56,300 deaths [1].

     The possible role of reproductive status stems from the observation by Fraumeni et al. [2] that nuns experience an excess of large bowel cancer as well as of the breast, uterus and ovary.

 

     In 1980, Mc Michael and Potter proposed that endogenous and exogenous steroid hormones might affect colon cancer risk [3]. Based on age-genders trend in colorectal incidence and mortality rates and other considerations, they hypothesised that nulliparity and the use of high-dose oral contraceptives might be protective against the development of colon cancer. Epidemiological data, however, on the relation of reproductive factors to colon cancer have been inconclusive. There has also been little to support an effect of oral contraceptives.

     In the United States in the past 30 years, colorectal cancer death rates have declined 7% for males but 30% for females [4]. During this period, estrogen replacement therapy (ERT) has become increasingly prevalent and may have contributed to the reduced incidence among females. Many epidemiological studies have examined this relationship, but the results were inconsistent.

     Seventeen studies reported that ERT was associated with a lower risk [relative risk (RR) < 0.9] of colon, rectal, or colorectal cancer, showing a significant reduction [5-22]. Three additional studies found that only recent or current use of ERT was associated with a lower risk of colorectal cancer [23-25]. However, other studies have found no association of ERT with colon cancer [26-32]. Indeed, one study found a significantly increased risk [10]. Two recent meta-analyses [33,34] calculated the overall RR of colorectal cancer among users of hormone replacement therapy. Each meta-analysis included slightly different studies. One found that the overall RR among hormone users compared with never users was 0.92 [95 percent confidence interval (CI), 0.79-1.08] and 0.97 (95 percent CI, 0.85-1.11) for colon and rectal cancer, respectively. The other analysis found an approximately 20 percent reduction in the risk of both colon and rectal cancer (RR = 0.81; 95 percent CI, 0.74-0.86 and RR= 0.81; 95 percent CI, 0.72-0.92). However, both meta-analyses found the reduced risk to be limited primarily to current or recent hormone users. The risk was significantly reduced by approximately one third.

Estrogen receptors

     Since 1970, when experimental data showed that hormones were involved in malignant intestinal tumors, a possible relationship between estrogens and human colorectal tumors has been generally accepted. Subsequently, the presence of estrogen receptors (ERs) was reported in these tumors [35]. At first, the data were related to the cytosolic ERs, but in 1987 we found the presence of both cytosolic and nuclear ERs in colorectal tumors and the surrounding normal mucosa [36]. The presence of nuclear ERs has given rise to experimental and clinical studies on the role of steroid hormones in the regulation of cell proliferation in colorectal cancer.

     In the beginning, estrogen action was believed to be mediated through a single ER, now called ER-alpha (ER-a). This contrasted with the situation of other members of the superfamily of steroid-thyroid-retinoic acid receptors in which multiple members have been found in each family. In this model, a tissue responded to hormone action when the receptors were activated by binding to ligands. Once activated, the receptors, as dimers, interacted with the control region of a target gene and regulated its expression. Thus, it was the binding affinity of an estrogen to its receptor that determined the biological activity of the estrogen in a target tissue. The discovery of a second ER [37,38] in 1996, called ER-beta (ER-b) to distinguish it from classical ER-a, began to explain the paradoxical behaviour of estrogens in different tissue types where there might be under-expression of the ER but increased estrogenic activity, for example. The cloning of ER-b provided an example of a steroid hormone receptor existing as two isoforms, each of which is encoded by a separate gene. Human ER-b is localized on chromosome 14q22-24 [39], in contrast to ER-a which sits on the long arm of chromosome 6.

     The alpha and beta forms of ER have an identical number of exons (eight) and have six functional domains that, in accordance with other members of the steroid receptor superfamily, are denoted A-F domains [39]. The area or domain where the ERs bind the DNA (DNA-binding domain) is virtually identical for the two ER subtypes (96% amino acid similarity). This suggests that ER-a and ER-b interact with, and activate, the same genes. Instead, the ligand-binding domain where the ERs bind the ligand has only approximately 58% similarity. It appears today that the ligand binding affinity of the two receptors may be quite different, suggesting that each may have a distinct biological role [40]. The tissue distribution, at least at the RNA level, also appears different. The uterus, pituitary, epididymus, and mammary gland, for example, show a great predominance of ER-a, whereas other tissues (the ovary, prostate and brain) show equal or greater levels of ER-b [41-45]. This may explain why some tissues are more responsive to estrogens than others. Although there is no direct correlation between the expression of ER-a and ER-b, some tissues co-express both receptors, and their ratio has been associated with cancer progression [46-48].

     All these studies have proposed two models of estrogen receptor signaling: (a) via the classical estrogen responsive element (ERE) or (b) through an activation protein 1 element (AP 1). When signaling is mediated by the former, ligand-bound ER undergoes a conformational change and binds to the ERE either as a homodimer (a/a or b/b) or as a heterodimer (a/b) [49, 50]. In the latter case, for transcriptional activation the ligand-bound homodimeric receptor requires the co-operation of the transcription factors Fos and Jun [51].

     Although there are many studies on the expression of ER-b in human reproductive tissues, there is very limited knowledge on the distribution of this receptor in the human gastrointestinal mucosa. For the first time, the presence of ER-b mRNA in human colic epithelium was reported by Enmark et al. [39]. However, in this study, the ER-b mRNA was detectable only in a single normal human colon specimen using in situ hybridization method. More recently, ER-b was studied in normal and malignant colon samples by Foley et al. [52]. Western blot analysis revealed very low levels of ER-a protein in tumor and normal colon tissue and a selective loss of ER-b protein expression was observed in malignant when compared to normal colon tissue. The authors hypothesized a posttranslational regulatory mechanism in the expression of this ER subtype. In our recent experiments, we confirmed the results obtained by Foley et al. and, by a semiquantitative Western blot analysis, we estimated that the decrease of the ER-b protein levels is about 1.5 fold in tumor tissue as compared to the adjacent normal mucosa and that in pathological samples the ER-b are twice the ER-a proteins present either in tumor or in adjacent normal mucosa compartments (personal communications).

     ER-b expression has also been shown in colorectal cancer cell lines, as well as in colon tissue,. Fiorelli et al. investigated either mRNA or protein expression for both ER subtypes in four different colon cancer cell lines (HCT8, HCT116, DLD-1, and LoVo cells) and demonstrated that ER-a was present, at mRNA and protein level, in these cells, whereas ER-a was always absent [53]. Subsequently, the same results were obtained by Arai et al. with other colorectal cancer cell lines [54].

     Comparing the distribution of the two ER subtypes in vivo and in vitro, ER-b, the most important ER subtype in colic cells, is present either in colorectal cancer cell lines or in colorectal mucosa. On the contrary, ER-a is present at mRNA and protein level in colorectal tissue only. However, there are conflicting results on the presence or absence of ER-a in colon cancer cell lines because Di Domenico et al. have shown the presence of a band of approximately 67 kDa, corresponding to the ER-a protein, in Caco-2 cells by immunoblot [55]. Therefore, further studies need to shed light on this issue and also to determine the role of the ER-a proteins, present at very low levels, in the colorectal mucosa.

     As well as wild-type ER, several variant forms of ER, generated by alternative mRNA splicing, have been identified in different tissues. These variants have been shown in both normal and neoplastic human tissues [56-58]. Most of these variants contain a deletion of one or more exons of the wild-type mRNA and their putative proteins, lacking some functional domains, might interfere with wild-type ER signalling pathways. A dominant negative activity has been observed. For example, an ER-a exon 5-deleted protein (ER-a D5), encoded by a messenger RNA that is deleted in the exon 5 sequence, lacks a part of the hormone binding domain of the wild-type molecule [59]. Interestingly, a constitutive hormone independent activity [60] and a wild-type enhancing activity [61] have also been attributed to the ER-a D5 protein in different systems.

     ER-b variant forms have also been described [55,56], which have further complicated the issue of estrogen action.

     The expression of an ER-b variant with a 139 bp deletion, similar to ER-a D5, could result in defective ligand binding with possible implications on acquisition of antiestrogen resistance [62,63]. More recently, five different isoforms of human ER-b (ER-b 1-5) have been described which differ in their C-terminal sequences [64]. It has also been proposed that changes in the relative expression of ER-b1 (wild-type form), ER-b2 and ER-b5 mRNAs could occur during tumorigenesis and the relative expression of ER-b1 is inversely related to the tumor grade [65]. At present, however, it is not known whether these variant mRNAs are translated in vitro and whether they contribute to endocrine sensitivity or resistance. Routine measurement of the variant forms of both ER subtypes in specific tissues, such as the human reproductive tissues, is likely to enhance the ability to individualize therapy and to offer a more selective approach in defining the most appropriate initial hormone treatment.

     However, no comprehensive study describing the variant forms of the two ER subtypes in the colorectal mucosa has been reported. The existence of ER-b variant forms was shown in only two studies in vitro. In the first study, multiple ER-b isoforms (ER-b 1-5) were documented in colorectal cancer cell lines [53]. ER-b1 was present in all cell lines examined, whereas the ER-b2-5 isoforms were present only in two of four cell lines of colic adenocarcinoma. More recently, the search for an ER-b variant with an insertion of 54 nucleotides, resulting in an 18 amino acid insertion in the ligand binding domain observed in other human cell lines [66], was negatively found in five colon cancer cell lines [54]. To date, there is one report only on ER-a mRNA variant expression in human colorectal mucosa [67]. The authors describe the presence of an ER-a exon 5 deletion splicing variant in only one colic adenocarcinoma tissue sample.

     In a recent study, we investigated mRNA and protein variant expression of both ER subtypes in colorectal cancer and in adjacent normal mucosa samples, to probe the presence and distribution of ER variant forms in this tissue [68]. Particularly, our study focused on the mutation of exons 3 and 5 of both ER subtypes. RT-PCR analysis showed only the exon 5-deleted form of ER-a mRNA (ER-a D5). Since ER-a mRNA D5 was expressed with similar frequency in the tumor and in surrounding normal mucosa, this mutated form, regardless of tumorigenetic factors, probably has a physiological role in the colorectal tract. The messenger of the D5 form was not translated into protein unlike the mRNA wild-type. The lack of protein expression of the ER-a mRNA D5, in about fifty per cent of the total ER-a mRNA examined in each tissue sample, may be due to an alternative splicing mechanism that down-regulates ER-a protein expression in colon cells.

Of five mRNA isoforms of ER-b, we found ER-b1 (wild-type), ER-b2 and ER-b5, the last form being present only in few colon samples. Comparing the amounts of each isoform, higher ER-b1 and -b2 levels than ER-b5 levels were observed, ER-b2 having the highest levels of all. We also studied the distribution of these isoforms in colorectal tumors and the surrounding normal mucosa. The expression of both ER-b1 and -b2 mRNAs was higher in colorectal tumor tissue vs. adjacent normal mucosa, the ER-b2 form having the highest relative magnitude . These data suggest the possibility of a direct association between the relative expression of ER-b mRNA isoforms and tumor disease. However, it remains to be determined whether increased ER-b isoform expression per se is responsible for tumorigenesis or is simply a consequence of inflammation of the pathological tissue, because lymphocytes also express the ER-b isoforms. This should be the focus of further studies.

     In view of the above data, ER-b wild-type and its variant forms may have diagnostic and/or prognostic implications in the management of colorectal carcinoma. Moreover, the presence of ER-b and its variant forms may help to explain the different effects of serum estrogens or of xenoestrogens, such as phytoestrogens present in the diet, on colonocyte proliferation.

Estrogens and polyamines

     Estrogens regulate growth, differentiation, and functioning of various target tissues, both within and outside the reproductive system. It is known that these substances also play an important role in the induction and progression of human cancers.

     Among the substances actively involved in cell proliferation and differentiation, polyamines (putrescine, spermidine and spermine) are a group of polycations found in high concentrations both in normal and in neoplastic rapidly proliferating cells [69].

     Polyamine synthesis is one of the early events occurring during the G1 phase of the cell cycle and before cell division. The metabolism of polyamines begins with the decarboxylation of ornithine to form putrescine. This step is governed by ornithine decarboxylase (ODC), which is the first and the rate-limiting enzyme in the polyamine biosynthetic pathway [70,71]. The structural characteristics and charge specificity of polyamines enable them to enter into hydrophobic and electrostatic interactions with macromolecules, DNA, RNA, and proteins, and to alter their three-dimensional structure and biological function [72]. Consequently, alterations in ODC activity and polyamine levels exert multiple, coordinated effects on cell proliferation and differentiation.

     ODC has been implicated as an essential promoter of cellular proliferation. It has recently been postulated that the ODC gene may act as an oncogene, because overexpression of this gene is essential for cell transformation. Overexpression of ODC may also lead to increased tumor invasiveness and angiogenesis [73,74]. Many different studies have shown that polyamines accumulate in cancer cells and that the use of inhibitors of polyamine biosynthesis or polyamine analogues has a remarkable ability to block tumor growth and prevent metastasis [75,76].

     In human gastrointestinal carcinoma, we previously reported higher polyamine levels and altered polyamine metabolism in neoplastic tissue than in the surrounding uninvolved mucosa [77-80]. High polyamine levels have also been found in pathological conditions characterized by increased cell proliferation and a high risk of neoplastic transformation [81,82]. Besides, our studies revealed the presence of higher polyamine levels in erythrocytes, which are their privileged carriers, from colorectal adenocarcinoma patients than from non neoplastic patients [83,84].

     The molecular mechanism underlying the effect of polyamine levels on cancer cell growth remains to be established, although several studies in target tissues have suggested that the polyamine pathway is interlinked with estrogenic regulation of cell growth [85-92].

     However, few data are currently available on the polyamine pathway and estrogenic modulation in colorectal cancer.

     In human colorectal tissue we have shown significantly higher ER concentrations in normal surrounding mucosa than in neoplastic tissue, whereas the polyamine content was significantly higher in the neoplastic tissue than in the surrounding mucosa “Fig. (1,2)”. Moreover, we found lower polyamine levels in ER-positive than in ER-negative tumors “Fig. (3)” [93,94]. We observed the same behavior in gastric cancer as in colon cancer [95].

     The correlation found between ERs and polyamine levels strengthens the hypothesis that estrogens-via their receptors-modulate the cellular growth of the gastrointestinal mucosa. Cellular activity of “normal” surrounding colonic mucosa may be partly controlled by estrogens, and this hormonal activity may not take place in neoplastic tissue.

 

    

However, few data have been reported on in vitro interactions among steroid hormones, polyamine metabolism, and cell proliferation in human gastrointestinal cancer cell lines. Cintron et al. [96] have shown that dihydrotestosterone stimulates ODC messenger RNA expression in the HT-29 colonic cell line, suggesting that androgens may play an important role in the growth of these cells. Recently, together with an inhibitory effect of 17b-Estradiol on the growth of the human gastric cancer cell line (HGC-27), we found that the hormone also inhibited ODC activity and polyamine synthesis [97]. It is possible that one of the mechanisms underlying growth inhibition of HGC-27 cells by 17b-Estradiol involves a decrease in ODC enzyme activity and polyamine synthesis to lower levels than those observed in the absence of estradiol.

     However, early studies on the action of estrogens in target tissues showed that this hormone stimulates ODC activity [98-100]. Besides, it has been shown that polyamines are essential for the growth of MCF-7 cells and that estradiol stimulates the polyamine cascade by inducing ODC mRNA [101]. Moreover, the growth inhibitory effects of an antiestrogen, tamoxifen, can be reversed by the addition of polyamines in several tamoxifen/estrogen-sensitive cell culture systems. Tamoxifen was found to reduce the level of ODC mRNA, ODC activity, and polyamine concentrations in MCF-7 breast cancer cells [102-106].

     In any case, all these studies suggest a potential role for polyamines as “second messengers” in the estrogen action of facilitating transcription, DNA replication, and cell division.

 

    

 Therefore, it may be supposed that estrogenic modulation of gastrointestinal cell growth may occur at the transcriptional level, consequent to an interaction between the hormone-receptor complex and its cognate ERE. These sequences are able to regulate the transcription of mRNA for the synthesis of proteins, hormones, and growth factors [107].

     On the other hand, it has been demonstrated that polyamines are able to alter the structural organization and DNA binding of estrogen receptors in a dose-dependent manner [108]. At low concentrations, polyamines appear to facilitate ER binding to its ERE, and this may occur either through electrostatic shielding of negatively charged ER from the negatively charged DNA phosphate backbone, or by directly altering the ERE docking structure by promoting a Z-DNA conformational change. In addition to affecting the DNA structure, stabilization of ER by polyamines might also contribute to enhanced ER-DNA binding. For example, estradiol binding to the receptor and the affinity of the ligand-receptor interactions were increased by polyamines [109].

 

    

At higher concentrations polyamines may impair ER function by interfering with its DNA-binding. For example, an excess of polyamine production leads to loss of ER DNA-binding in mouse uterine tissue, and this abnormality can be reversed by treatment with a polyamine biosynthesis inhibitor. Moreover, inhibition of ER binding to DNA at high polyamine concentrations might be a consequence of competitive interactions of polyamines and ER for available sites on DNA. Furthermore, polyamine-induced condensation and aggregation of DNA might also dislodge ER from its preferential binding sites on DNA [110,111].

     In any case, the modulation of ER-ERE interaction may be one mechanism through which polyamines influence gene regulation.

Estrogens and colon cell growth in vitro and in vivo

     The presence of appreciable amounts of ERs and their mRNA in colorectal carcinoma and normal colonic epithelium implies possible direct effects of steroid hormones on proliferation of intestinal epithelial cells.

     In fact, significant growth effects of sex steroid hormones on human gastric and colorectal cancer cell lines have been reported in vivo and in vitro.

     The role of female gonadal hormones in the growth of normal colonic mucosa became evident on the early studies by Hoff et al. [112], who investigated the effect of estrogen on cell proliferation in the descending colon of the mouse. In this study ovariectomized mice were given single or multiple injections of 10ng/g body weight of 17 beta-estradiol and were killed 1h after a ³H-thymidine injection. Estrogen treatments decreased incorporation of ³H-thymidine into the DNA of colonic mucosa most markedly at 4h after the single injection or the last multiple injections. A similar inhibitory effect was observed in the colonic mucosa of male mice as well as in the mucosa of mice in which colonic epithelial cell proliferation was enhanced by refeeding after 48h of fasting.

     More recently, Narayan et al. [113] clearly indicated a significant effect of female gonadal steroids, particularly estradiol, on the growth of mouse colon cancer (MC-26) in vivo. Ovariectomized mice demonstrated a significant attenuation of the growth of MC-26 tumors, which was reversed by treatment with an optimally effective dose of estradiol over and above the effect observed in intact sham-operated animals. Moreover, human colorectal carcinomas xenografted into nude mice, have similarly been reported to respond to the growth effects of estradiol.

     While some studies [113,114] do not support the view that estradiol has direct effects on colon cancer cells, other investigators have shown significant growth effects of sex steroid hormones on gastric and colorectal cancer cell lines.

     In gastric cancer cell lines contrasting results have been obtained on the effects of estrogens on cell growth i.e., stimulation, inhibition and no change [115-118]. However, the different estradiol concentrations used in the various experiments could also account for the different proliferative responses among the human gastric cancer cell lines.

     In this connection, Messa et al. [97,119] showed that two human gastric carcinoma cell lines, HGC-27 and AGS, display ERs and that in vitro cell proliferation is affected by the administration of sex hormones, being inhibited by 17b-estradiol.

     It has been suggested that estrogens and their receptors may play a role in the growth of gastrointestinal tumors by interacting with other molecules required for cell proliferation, such as the growth factors [120]. Exposure of AGS cell to combined treatment with 17b-estradiol and epidermal growth factor (EGF) induces changes in the cell proliferation rate [119]. Specifically, the addition of EGF significantly enhances AGS cell proliferation in the presence of the lowest estradiol concentrations. Conversely, at the highest 17b-estradiol concentrations the proliferative effect of EGF administration is suppressed., and a decrease in proliferation rates occurs. Thus, these findings provide evidence of the existence of a functional cross-talk between 17b-estradiol and EGF in regulating the growth of human gastric cancer cell lines.

     Disparate results were obtained in several studies examining the action of estradiol upon the colon cells.

     Harrison et al. [117], studied the effect of sex hormones on the growth of human gastric and colorectal cancer cell lines. Significant growth stimulation of the gastric and 2 colorectal cell lines (LoVo and ST16) occurred at physiologic concentrations of estradiol whereas the other colorectal lines (C170 and C277 ) showed no response to estradiol.

     Moreover, Lointier et al. [121], observed that synthetic estrogen, moxestrol and ethinyl estradiol, produced inhibition of the malignant colonic cell line LoVo, but only at supraphysiological levels. In medium containing 10% charcoal-treated fetal bovine serum, the inhibitory effects of estrogens were not observed, and lower concentrations of moxestrol and ethinyl estradiol facilitated cell growth.

     Differential growth response to estrogens in premalignant and malignant colonic cell lines was investigated by Singh et al. [122].

     This study demonstrated that 17b-estradiol, at near physiological concentration, promote growth of a colonic epithelial cell line derived from an adenomatous polyp, AA/C1. The chemically transformed derivative of this cell line, AA/C1/SB10, did not respond to estradiol even though it expressed a similar level of estrogen receptors. The authors hypothesize that the lack of effect of estradiol on growth of transformed cell line may be related to the constitutive growth rates of these cells. The faster growing colonic cell lines, such as AA/C1/SB 10, may be more autonomous in their growth, and so insensitive to the influence of estradiol.

     Furthermore, the human colon cancer line Caco-2 was reported to grow at a very slow rate in the absence of estradiol and to respond to physiological concentrations of estradiol by active proliferation [55]. In this study the authors observed that estradiol stimulates c-src and c-yes of Caco-2 cells. These enzymes are involved in signal transduction processes initiated by growth factors [123], and activation of human c-src has been observed in a large proportion of human breast and colon carcinoma [124,125].

     Therefore, it is likely that estradiol activation of c-src and c-yes kinases is crucial in triggering cell proliferation and in carcinogenesis.

     Xu et al. [126] employed antisense oligodeoxynucleotides to ER mRNA to block ER synthesis in MC-26 cells, a mouse colon cancer cell line, to demonstrate ER-mediated direct stimulation of MC-26 cell growth in vitro. The growth- stimulatory effect of estrogen was abolished by antisense oligo treatment, demonstrating that the ER is directly involved in regulation of colon cancer cell growth.

     In four different colon cancer cell lines (HCT8, HCT116, DLD-1, and LoVo cells), cell proliferation has been shown to be affected by 17-b estradiol in a cell-specific fashion [53].

     Specifically, estradiol induced a significant increase in HCT8 cell growth at physiological concentrations (1 and 100 pM ), but was inert at higher concentrations. Conversely, the hormone significantly reduced HCT116 and DLD-1 cell growth at concentrations from 10 nM to 1mM and inhibited LoVo cell proliferation at concentrations from 1pM to 1mM.

     In other experiments [54], results from a cell growth assay demonstrated that five human colon cancer cells ( HT29, Colo320, LoVo, SW480, and HCT116 ) were not influenced by estrogen at either lower ( 1 and 10nM ) or higher (100 and 500nM ) concentrations.

     Lastly, the effects of estradiol, progesterone, DHEA, testosterone, 17 epiestriol and quercetin were tested on the growth of the colon cancer cell line DLD-1 [127]. When these agents were given singly, they did not show any significant effects on cellular growth. Conversely, in the same study combinations of tamoxifen and steroids inhibited cell growth with a tumor-static effect on colon cancer cells.

     Overall, these data suggest that estrogens may play an important role in the growth of colon carcinoma cells.

     Furthermore, the in vitro cell-specific responsiveness to estrogen could partially reflect differential tumor biology in different individuals

Abbreviations

ER         =    Estrogen Receptor

ERT      =    Estrogen Replacement Therapy

RR        =    Relative Risk

ER-a    =    Estrogen Receptor alpha

ER-b     =    Estrogen Receptor beta

ERE      =    Estrogen Responsive Element

ODC     =    Ornithine-Decarboxylase

EGF      =    Epidermal Growth Factor

REFERENCES

[1]        Greenle, R.T., Murray, T., Bolden, S., Wingop, A. Cancers Statistics, 2000 CA Cancer J. Cli. 2000, 50, 7.

[2]        Fraumeni, J.F. Jr., Lloyd, W., Smith, E.M., Wagoner, S.K. J. Natl. Cancer Inst., 1969, 42,455

[3]        Mc Michael, A.J., Potter, S.D. J. Natl. Cancer Inst., 1980, 65, 1201.

[4]        American Cancer Society. Cancer fact figures. Atlanta: American Cancer Society 1995.

[5]        Burch, J.C., Byrd, B.F., Vaughn, W.K. Front. Horm. Res., 1975, 3, 208.

[6]        Hammond, C.B., Jelovsek, F.R., Lee, K.L., Creasman, W.T., Parker, R.T. Am. J. Obstet. Gynecol., 1979, 133, 537.

[7]        Potter, J. D., Michael, A. J. J. Natl. Cancer Inst, 1983, 71, 703.

[8]        Furner, S.E., Davis, F.G., Nelson, R.L., Haenszel, W. Cancer Res., 1989, 49, 4936.

[9]        Chute, C.G., Willett, W.C., Colditz, G.A., Stampfer, M.J., Rosner, B., Speizer, F.E. Epidemiology, 1991, 2, 201.

[10]      Wu- Williams, A.H., Lee, M., Whittemore, A.S., Gallagher, R.P., Jiao, D.A., Zheng, S., Zhou, L., Wang, X.H., Chen, K., Jung, D. Cancer Res., 1991, 51, 2307.

[11]      Gerhardsson de Verdier, M., London, S. Cancer Cause Control, 1992, 3, 355.

[12]      Jacobs, E.J., White, E., Weiss, N.S. Cancer Causes Control, 1994, 5, 359.

[13]      Risch, H.A., Howe, G.R. Cancer Epidemiol. Biomarkers Prev., 1995, 4, 21.

[14]      Calle, E.E., Miracle-McMahill, H.T., Thun, M.J., Heath, C.W. Jr. J. Natl. Cancer Inst., 1995, 87, 517.

[15]      Marcus, P.M., Newcomb, P.A., Young, T., Storer, B.E. Ann. Epidemiol., 1995, 5,303.

[16]      Newcomb, P.A., Storer, B.E. J. Natl. Cancer Inst., 1995, 87, 1067.

[17]      Folsom, A.R., Mink, P.J., Sellers, T.A., Hong, C.P., Zheng, W., Potter, J.D. Am. J. Public Health, 1995, 85, 1128.

[18]      Sturgeon, S.R., Schairer, C., Brinton, L.A., Pearson, T., Hoover, R.N. Epidemiology, 1995, 6, 227.

[19]      Persson, I., Yuen, J., Bergkvist, L., Schairer, C. Int. J. Cancer, 1996, 67, 327.

[20]      Fernandez, E., La Vecchia, C., D’Avanzo, B., Franceschi, S., Negri, E., Parazzini, F. Br.J. Cancer, 1996, 73, 1431.

[21]      Kampman, E., Potter, J.D., Slattery, M. L., Caan, B.J., Edwards, S. Cancer Causes Control, 1997, 8, 146.

[22]      Fernandez, E., La Vecchia, C., Braga, C., Talamini, R., Negri, E., Parazzini, F., Franceschi, S. Cancer Epidemiol. Biomarkers Prev., 1998, 7, 329.

[23]      Troisi, R., Schairer, C., Chow, W.H., Schatzking, A., Brinton, L.A., Fraumeni, J.F. Jr. Cancer Causes Control, 1997, 8, 146.

[24]      Grodstein, F., Martinez, M.E., Platz, E.A., Giovannucci, E., Colditz, G.A., Kautzky, M., Fuch, C., Stampfer, M.J. Ann. Intern. Med., 1998, 128, 705.

[25]      Paganini-Hill, A. Dis. Colon Rectum, 1999, 42, 1300.

[26]      Nachtigall, L.E., Nachtigall, R.H., Nachtigall, R.D., Beckman, E.M. Obstet. Gynecol., 1979, 54, 74.

[27]      Weiss, N.S., Daling, J.R., Chow, W.H. J. Natl. Cancer Inst., 1981, 67, 57.

[28]      Wu, A.H., Paganini-Hill, A., Ross, R.K., Henderson, B.E. Br. J. Cancer, 1987, 55, 687.

[29]      Davids, F.G., Furner, S.E., Persky, V., Koch, M. Int. J. Cancer, 1989, 43, 587.

[30]      Adami, H.O., Persson, I., Hoover, R., Schairer, C., Bergkvist, L. Int. J. Cancer, 1989, 44, 833.

[31]      Peters, R.K., Pike, M.C., Chang, W.W., Mack, T.M. Br. J. Cancer, 1990, 61, 741.

[32]      Bostick, R.M., Potter, J.D., Kushi L.H., Sellers, T.A., Steinmetz, K.A., McKenzei, D.R., Gapstur, S.M., Folsom, A.L. Cancer Causes Control, 1994, 5, 38.

[33]      Nanda, K., Bastian, L.A., Hasselblad, V., Simel, D. L. Obstet. Gynecol., 1999, 93, 880.

[34]      Grodstein, F., Newcomb, P.A., Stampfer, M.J. Am. J. Med., 1999, 93, 880.

[35]      Alford, T.C., Do, H.M., Ceelhoed, C.W., Tsangaris, N.T., Lippman, M.E. Cancer, 1979, 43, 980.

[36]      Francavilla, A., Di Leo, A., Polimeno, L., Conte, D., Barone, M., Fanizza, G., Chiumarulo, C., Rizzo, G., Rubino, M. Gastroenterology, 1987, 93,1301.

[37]      Mosselman, S., Polman, J., Dijkema, R.. FEBS Lett., 1996, 392, 49.

[38]      Kuiper, G.G.J.M., Enmark, E., Pelto-Huikko, M., Nilsson, S., Gustafsson, J.A. Proc. Natl. Acad. Sci. USA, 1996, 93, 5925.

[39]      Enmark, E., Pelto-Huikko, M., Grandien, K., Lagercrantz, S., Lagercrantz, J., Fried, G., Nordenskjold, M., Gustafsson, J.A. J. Clin. Endocrinol. Metab., 1997, 82, 4258.

[40]      Kuiper, G.G.J.M., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., Gustafsson, J.A. Endocrinology, 1997, 138, 863.

[41]      Ostrelund, M.K., Gustafsson, J.A., Keller, E., Hurd, Y.L. J. Clin. Endocrinol. Metab., 2000, 85, 3840.

[42]      Chiang, C-H., Cheng, K.W., Igarashi, S., Nathwani, P.S., Leung, P.C.K. J. Clin. Endocrinol. Metab., 2000, 85, 3828.

[43]      Brandenberger, A.W., Tee, M.K., Jaffe, R.B. J. Clin. Endocrinol. Metab., 1998, 83, 1025.

[44]      Rutherford, T., Brown, W.D., Sapi, E., Aschkenazi, S., Munoz, A., Mor, G. Obstet. Gynecol., 2000, 96, 417.

[45]      Sasano, H., Suzuki, T., Matsuzaki, Y., Fukaya, T., Endoh, M., Nagura, H., Kimura, M. Clin. Endocrinol. Metab., 1999, 84, 781.

[46]      Pujol, P., Rey, J-M., Nirde, P., Roger, P., Gastaldi, M., Laffargue, F., Rochefort, H., Maudelonde, T. Cancer Res., 1998, 58, 3197.

[47]      Leygue, E., Dotzlaw, H., Watson, P.H., Murphay, L.C. Cancer Res., 1998, 58, 3197.

[48]      Speirs, V., Parkes, A.T., Kerin, M..J., Alton, D.S., Carleton, P.J., Fox, J.N., Atkin, S.L. Cancer Res., 1999, 59, 525.

[49]      Cowley, S.M., Hoare, S., Mosselman, S., Parker, M.G. J. Biol. Chem., 1997, 272, 19858.

[50]      Pace, P., Taylor, J., Suntharalingam, S., Coombes, R.C., Ali, S. J. Biol. Chem., 1997, 272, 25832.

[51]      Gaub, M-P., Bellard, M., Scheuer, I., Chambon, P., Sassone-Corsi, P. Cell, 1990, 63, 1267.

[52]      Foley, E.F., Jazaeri, A.A., Shupnik, M.A., Jazaeri, O., Rice, L.W. Cancer Res., 2000, 60, 245.

[53]      Fiorelli, G., Picariello, L., Martineti. V., Tonelli, F., Brandi, M.L. Biochem. Biophys. Res. Commun., 1999, 261, 521.

[54]      Arai, N., Strom, A., Rafter, J.J., Gustafsson J-A. Biochem. Biophys. Res. Commun., 2000, 270, 425.

[55]      Di Domenico, M., Castoria, G., Bilancio, A., Migliaccio, A., Auricchio, F. Cancer Res., 1996, 56, 4516.

[56]      Yang, J., Liu, A., Chiou, S.K., Guzman, R., Nandi, S. Int. J. Mol. Med., 2000, 5, 223.

[57]      Leygue, E.R., Watson, P.H., Murphy, L.C. J. Natl. Cancer Inst., 1996, 88, 284.

[58]      van Dijk, M.A., Hart, A.A., van’t Verr, L.J. Cancer Res., 2000, 60, 530.

[59]      Desai, A., Luqmani Y., Walters, J.E., Coope, R.E., Dagg, B., Gomm, J.J., Pace, P.E., Ress, C.N., Thirunavukkarasu, V., Shousha, S., Groome, N.P., Coombes, R., Ali, S. Br. J. Cancer, 1997, 75, 1173.

[60]      Fuqua, S.A.W., Fitzgerald, S.D., Chamness, G.C., Tandon, A.K., Mc Donnell, D.P., Nawaz, Z., O’Malley, B.W., Mc Guire, W.L. Cancer Res., 1991, 51, 105.

[61]      Chaidarun, S. S., Alexander, J. M. Mol. Endocrinol., 1998, 12,1355.

[62]      Speirs, V., Adams, I.P., Walton, D.S., Atkin, S.L. J. Clin. Endocrinol. Metab., 2000, 85, 1601.

[63]      Vladusic, E.A., Horby, A.E., Guerra-Vladusic, K; Lupu, R. Cancer Res., 1998, 58, 210.

[64]      Moore, J.T., McKee, D.D., Slentz-Kesler, K., Moore, L.B., Jones, S.A., Horne, E.L., Su, J.L., Kliewer, S.A., Lehmann, J.M., Willson, T.M. Biochem Biophys. Res. Commun., 1998, 247, 75.

[65]      Leygue, E., Dotzlaw, H., Watson, P.H., Murphy, L.C. Cancer Res. 1999, 59, 1175.

[66]      Hanstein, B., Liu, H., Yancisin, M.C., Brown, M. Mol. Endocrinol., 1999, 13, 129.

[67]      Campbeil-Thompson, M., Valentine, J.F., Bui, A., MacKay, S.L. Gastroenterology, 1999, 116, A595.

[68]      Cavallini, A., Messa, C., Di Leo, A. Int. J. Cancer, 2001 (submitted).

[69]      Tabor, C.W., Tabor, H. Ann. Rev. Biochem., 1981, 53, 749

[70]      Pegg, A.E., McCann, P.P. Am. J. Physiol., 1982, 243, C212.

[71]      Seiler, N. Digestion 1990, 46, 319.

[72]      Thomas, T., Kiang, D.T. Cancer Res., 1987, 47, 1799.

[73]      Auvinen, M. J. Natl. Cancer Inst., 1997, 89, 533.

[74]      Kubota, S., Kiyosawa, H., Nomura, Y., Yamada, T., Seyama, Y. J. Natl. Cancer Inst., 1997, 89, 567.

[75]      McCloskey, D. E., Woster, P. M., Casero, R. A. Jr., Davidson, N.E. Clin. Cancer Res. 2000, 6, 17.

[76]      Takahashi, Y., Mai, M., Nishioka, K. Int. J. Cancer 2000, 85, 243.

[77]      Linsalata, M., Russo, F., Cavallini, A., Berloco, P., Di Leo, A. Dis. Colon Rectum, 1993, 36, 662.

[78]      Notarnicola, M., Linsalata, M., Caruso, M.G., Cavallini, A., Di Leo A. J. Gastroenterol., 1995, 30, 705

[79]      Russo, F., Linsalata, M., Giorgio, I., Caruso, M.L., Armentano, R., Di Leo A. Dig. Dis. Sci., 1997, 42, 576.

[80]      Linsalata, M., Cavallini, A., Di Leo A. Anticancer Res., 1997, 17, 3757.

[81]      Linsalata, M., Russo, F., Notarnicola, M., Berloco, P., Di Leo A. Ital. J. Gastroenterol. Hepatol., 1998, 30, 484.

[82]      Lorusso, D., Linsalata, M., Pezzolla, F., Berloco, P., Osella, R.O., Guerra, V., Di Leo, A., Demma, I. Anticancer Res., 2000, 20, 2197.

[83]      Linsalata, M., Amati, L., Caradonna, L., Caccavo, D., Jirillo, E., Di Leo A. Oncol. Rep., 1999, 6, 1411.

[84]      Linsalata, M., Leo, S., Guerra, V., Di Leo A. Anticancer Res. 2000, 20, 2113.

[85]      Manni, A., Wright, C. J. Natl. Cancer Inst., 1984, 73, 511.

[86]      Kremmer, T., Palyi, I., Daubner, D., Boldizsar, M., Vincze, B., Paulik, E., Sugar, J., Pokorny, E., Tury E. Anticancer Res., 1991, 11, 1807.

[87]      Lee, H.L., Kim, S.O., Lee, H.D., Chung, B.C. Cancer Letters, 1998, 133, 47.

[88]      Gomez-Fabre de Pedro, E., Medina, M.A., Nunez de Castro, I., Marquez, J. Cancer Letters, 1997, 113, 141.

[89]      Huber, M., Poulin, R. J. Clin. Endocrin. Metabol., 1996, 81, 113.

[90]      Thomas, T., Thomas, T.J. Cancer Res., 1994, 54, 1077.

[91]      Manni, A., Khin, S., Biser, N., English, H., Badger, B., Martel, J.L. Cancer Res., 1992, 52, 5720.

[92]      Leveque, J., Foucher, F., Havouis, R., Desury, D., Grall, J.Y., Moulinoux, J.P. Anticancer Res., 2000, 20, 97.

[93]      Di Leo, A., Linsalata, M., Cavallini, A., Messa, C., Russo, F. Dis. Colon Rectum, 1992, 35, 305.

[94]      Russo, F., Linsalata, M., Messa, C., Cavallini, A., Notarnicola, M., Guerra, V., Lantone, G., Lacatena, M., Di Leo, A. Ital. J. Gastroenterol., 1992, 24, 8.

[95]      Linsalata, M., Messa, C., Russo, F., Cavallini, A., Di Leo, A. Scand. J. Gastroenterol., 1994, 29, 67.

[96]      Cintron, J.R., Asadi, F.K., Malakouti, S., Abcarian, A. Dis. Colon Rectum, 1996, 39, 406.

[97]      Messa, C., Pricci, M., Linsalata, M., Russo, F., Di Leo A. Scand. J. Gastroenterol., 1999, 34, 79.

[98]      Cohen, S., O’Malley, B.W., Stastny, M. Science, 1970, 170, 336.

[99]      Huber, M., Poulin R. Mol. Cell. Endocrinol., 1996, 117, 211.

[100]    Kawarada, Y., Mizumoto, R. Carcinogenesis, 1995, 16, 2965.

[101]    Thomas, T., Thomas, T.J. Breast Cancer Res. Treat., 1993, 29, 189.

[102]    Lima, G., Shiu, R.P.C. Cancer Res., 1985, 45, 2466.

[103]