Deprivation of estrogen causes breast tumors in women to adapt and develop enhanced sensitivity to this steroid. Accordingly, women relapsing after treatment with oophorectomy, which substantially lowers estradiol for a prolonged period, respond secondarily to aromatase inhibitors with tumor regression. We have utilized in vitro and in vivo model systems to examine the biologic processes whereby long-term estradiol deprivation (LTED) causes cells to adapt and develop hypersensitivity to estradiol. Several mechanisms are associated with this response, including up-regulation of estrogen receptor-a (ERa) and the MAP kinase, phosphoinositol 3 kinase (PI3-K) and mammalian target of rapamycin (mTOR) growth factor pathways. ERa is four-to tenfold up-regulated and co-opts a classical growth factor pathway using Shc, Grb-2 and Sos. This induces rapid non-genomic effects which are enhanced in LTED cells. The molecules involved in the non-genomic signaling process have been identified. Estradiol binds to cell membrane-associated ERa, which physically associates with the adaptor protein Shc, and induces its phosphorylation. In turn, Shc binds Grb-2 and Sos, which result in the rapid activation of MAP kinase. These non-genomic effects of estradiol produce biologic effects as evidenced by Elk-1 activation and by morphologic changes in cell membranes. Additional effects include activation of the PI3-K and mTOR pathways through estradiolinduced binding of ERa to the IGF-I and epidermal growth factor receptors. A major question is how ERa locates in the plasma membrane since it does not contain an inherent membrane localization signal. We have provided evidence that the IGF-I receptor serves as an anchor for ERa in the plasma membrane. Estradiol causes phosphorylation of the adaptor protein, Shc and the IGF-I receptor itself. Shc, after binding to ERa, serves as the 'bus' which carries ERa to Shc-binding sites on the activated IGF-I receptors. Use of small inhibitor (si) RNA methodology to knockdown Shc allows the conclusion that Shc is needed for ERa to localize in the plasma membrane. In order to abrogate growth factorinduced hypersensitivity, we have utilized a drug, farnesylthiosalicylic acid, which blocks the binding of GTP-Ras to its membrane acceptor protein, galectin 1, and reduces the activation of MAP kinase. We have also shown that this drug is a potent inhibitor of mTOR as an additional mechanism of inhibition of cell proliferation. The concept of 'adaptive hypersensitivity' and the mechanisms responsible for this phenomenon have important clinical implications. The efficacy of aromatase inhibitors in patients relapsing on tamoxifen could be explained by this mechanism and inhibitors of growth factor pathways should reverse the hypersensitivity phenomenon and result in prolongation of the efficacy of hormonal therapy for breast cancer.

Mutations in the Wnt signalling cascade are believed to cause aberrant proliferation of colorectal cells through T-cell factor-4 (TCF4) and its downstream growth-modulating factors. HOXB13 is exclusively expressed in prostate and colorectum. In prostate cancers, HOXB13 negatively regulates b-catenin/TCF4-mediated transactivation and subsequently inhibits cell growth. To study the role of HOXB13 in colorectal tumorigenesis, we evaluated the expression of HOXB13 in 53 colorectal tumours originated from the distal left colon to rectum with their matching normal tissues using quantitative RT -PCR analysis. Expression of HOXB13 is either lost or diminished in 26 out of 42 valid tumours (62%), while expression of TCF4 RNA is not correlated with HOXB13 expression. TCF4 promoter analysis showed that HOXB13 does not regulate TCF4 at the transcriptional level. However, HOXB13 downregulated the expression of TCF4 and its target gene, c-myc, at the protein level and consequently inhibited b-catenin/TCF-mediated signalling. Functionally, forced expression of HOXB13 drove colorectal cancer (CRC) cells into growth suppression. This is the first description of the downregulation of HOXB13 in CRC and its mechanism of action is mediated through the regulation of TCF4 protein stability. Our results suggest that loss of HOXB13 may be an important event for colorectal cell transformation, considering that over 90% of colorectal tumours retain mutations in the APC/b-catenin pathway.

Clinical observations suggest that human breast tumors can adapt to endocrine therapy by developing hypersensitivity to estradiol (E 2 ). To understand the mechanisms responsible, we examined estrogenic stimulation of cell proliferation in a model system and provided in vitro and in vivo evidence that long-term E 2 deprivation (LTED) causes 'adaptive hypersensitivity'. The enhanced responses to E 2 do not involve mechanisms acting at the level of transcription of estrogen-regulated genes. We found no evidence of hypersensitivity when examining the effects of E 2 on regulation of c-myc, pS2, progesterone receptor, several estrogen receptor (ER) reporter genes, or c-myb in hypersensitive cells. Estrogen deprivation of breast cells long-term does up-regulate both the MAP kinase and phosphatidyl-inositol 3-kinase pathways. As a potential explanation for up-regulation of these signaling pathways, we found that ERα is 4-to 10-fold up-regulated and co-opts a classic growth factor pathway using Shc, Grb-2 and Sos. This induces rapid non-genomic effects which are enhanced in LTED cells. E 2 binds to cell membrane-associated ERα, physically associates with the adapter protein SHC, and induces its phosphorylation. In turn, Shc binds Grb-2 and Sos, which results in the rapid activation of MAP kinase. These non-genomic effects of E 2 produce biological effects as evidenced by Elk activation and by morphological changes in cell membranes. Further proof of the non-genomic effects of E 2 involved use of cells which selectively expressed ERα in the nucleus, cytosol and cell membrane. We created these COS-1 'designer cells' by transfecting ERα lacking a nuclear localization signal and containing a membrane localizing signal.The concept of 'adaptive hypersensitivity' and the mechanisms responsible for this phenomenon have important clinical implications. Adaptive hypersensitivity would explain the superiority of aromatase inhibitors over the selective ER modulators (SERMs) for treatment of breast cancer. The development of highly potent third-generation aromatase inhibitors allows reduction of breast tissue E 2 to very low levels and circumvents the enhanced sensitivity of these cells to the proliferative effects of E 2 . Clinical trials in the adjuvant, neoadjuvant and advanced disease settings demonstrate the greater clinical efficacy of the aromatase inhibitors over the SERMs. More recent observations indicate that the aromatase inhibitors are superior for the prevention of breast cancer as well. These observations may be explained by the hypothesis that estrogens induce breast cancer both by stimulating cell proliferation and by their metabolism to genotoxic products. The SERMs block ER-mediated proliferation only, whereas the aromatase inhibitors exert dual effects on proliferation and genotoxic metabolite formation.

Background. Oral contraceptives (OC) contain an orally active estrogen in combination with an orally active synthetic progestin derived from 19‐nortestosterone. OC have had an enormous positive impact on public health for the past three decades, and in the main, there has been a remarkably low incidence of troublesome side effects. Although estrogens are implicated in an increased incidence of breast and endometrial cancer, epidemiologic studies have not provided convincing evidence to support a direct correlation between OC use and an increase in breast cancer incidence. By contrast, OC do cause a decrease in the incidence of endometrial and ovarian carcinoma. During the past decade, several isolated reports have linked an increased incidence of breast cancer with the use of synthetic progestins. No mechanism for the proliferative potential of progestins has been offered. Therefore, the authors investigated this problem to formulate a hypothesis, based on laboratory data, that might be evaluated in populations at risk. Methods. The synthetic progestins (19‐nortestosterone derivatives) chosen for the study were norethynodrel, norethindrone, norgestrel (levonorgestrel), and gestodene. These were compared with the actions of medroxyprogesterone acetate (MPA). To determine whether the progestins produced their effects via the ER, the cells were transfected with a chloramphenicol acetyl transferase (CAT) reporter gene containing an estrogen response element only activated by ER. Results. The 19‐nortestosterone derivatives all stimulated the growth of estrogen receptor (ER)‐positive but not ER‐negative breast cancer cells in culture. Antiestrogens, but not the antiprogestin mifepristone (also known as RU 486), inhibited progestin‐stimulated cell proliferation. MPA did not stimulate cell proliferation. All the synthetic progestins that increased replication also activated CAT. Activation was blocked by antiestrogens but not by mifepristone; the synthetic progestin MPA was inactive. Conclusions. These studies provided direct evidence that some synthetic progestins exert estrogenic effects through the ER. The results demonstrated that progestins can have a dual effect on estrogen target tissues either to stimulate or differentiate cells. The results suggest that some beneficial estrogen‐like effects could be produced by synthetic progestins (e.g., bone preservation), but epidemiologic studies of OC use should focus of the “total estrogen” content to establish whether some formulations place some groups of women at greater risk of having breast cancer.

AIB1 expression correlates with HER-2 expression in breast cancer and shows a trend of association with loss of PR expression in ER+ tumours. Our study supports the postulated role of AIB1 in ER-growth factor interactions.