Insulin-like growth factor receptor levels are regulated by cell density and by long term estrogen deprivation in MCF7 human breast cancer cells.

This work describes a reciprocal relationship between cell density and levels of insulin-like growth factor receptors (IGFR) in MCF7 human breast cancer cells, which adds a new dimension to the mechanism of cross-talk between estrogen and insulin-like growth factors in the regulation of breast cancer cell growth. The reduced binding of both (125)I-IGF1 and alphaIR3 anti-IGFR antibody to whole cells showed that IGFR are lost from the surface of MCF7 cells as cell density increases, and this occurred irrespective of the presence or absence of estradiol. Western immunoblotting further confirmed loss of type I IGFR from MCF7 cells with increasing cell density. Long term estrogen deprivation was found to increase the levels of IGFR at all cell densities, such that after 96 weeks of estrogen deprivation, IGFR levels had become similar at the highest cell density in the absence of estradiol to the IGFR levels at the lowest cell density in the estrogen-maintained cells, and the levels of IGFR could be increased still further by estradiol. This overexpression of IGFR in the estrogen-deprived cells correlated with a reversal of response to exogenously added ligand, in that concentrations of insulin, IGFI, and IGFII that had stimulated growth of the estrogen-maintained cells became growth inhibitory to the estrogen-deprived cells. Blockade of the IGFIR with the alphaIR3 anti-IGFR antibody could partially inhibit the growth of the estrogen-deprived cells, suggesting that up-regulation of IGFR in these cells may contribute to the mechanism of adaptation to growth in steroid-deprived conditions which results in progression to estrogen independence of cell growth.

Estrogen regulation of breast cancer cell growth can be modulated by complex interactions with a variety of growth factors, particularly insulin-like growth factors (IGF) 1 (1). The extensive clinical literature provides evidence that, when estrogen receptors (ER) are present, estrogen is a main stimulus for growth of breast cancer cells both in vitro (2) and in vivo (3). IGFs have been implicated in growth regulation of breast can-cer cells (4) since they are mitogenic to such cells (5) and also because blockade of the IGF receptor can reduce growth of the cells in the absence of estrogen (6). Breast cancer cell lines possess IGFII (7), make several of the IGFBPs (8), and have functional IGF receptors (5), both type I (IGFIR) (9) and type II (IGFIIR) (10).
At a molecular level, there appear to be many cross-talk pathways between estrogen and IGF (11,12). Estrogen can regulate levels of IGFII mRNA (7), IGF-binding protein (8,13), IGFIR (14), and downstream signaling molecules insulin receptor substrates IRS-1 and IRS-2 (15,16). On the other hand, IGFs can influence expression of ER (17) and of estrogenregulated genes such as pS2 (18,19). Overexpression studies have shown that increasing limiting components of the IGF pathway can negate estrogen regulation (19 -21). Overexpression of IGFII alone could override the need for estrogen in MCF7 human breast cancer cells (20,21) where the limiting component was ligand (20) and not receptor (22), but in ZR-75-1 human breast cancer cells overexpression of IGFII could only override the estrogen requirement when IGFIR (the limiting component) were simultaneously overexpressed (19). The interrelationship between IGFR and ER signaling is shown also in vivo through the tendency for levels of IGFR to correlate positively with levels of ER in breast cancers (3,9), and it is thought that abnormally high levels of IGFRs may contribute to increase in tumor mass and/or aid tumor recurrence (9). Part of the molecular basis for these cross-talk pathways may relate to an ability of ER to interact with the AP1 pathway (23), but the full picture remains far from clear. Although some studies have shown that estrogen can alter IGFR levels (22), it is unknown whether this is a mechanism or a consequence of estrogen action on cell growth. Work in other systems has shown that growth factor receptor levels are regulated by cell density (24 -34), and estrogen acts to increase cell density including saturation density (35). Since regulation of cell density and contact inhibition are fundamental parameters in cell growth control, we have studied here the effects of cell density on IGFR levels in breast cancer cells grown in the presence and absence of estrogen.
A further issue in estrogen growth control of breast cancer cells is the ability of these cells to escape from regulation by estrogen. Loss of response to endocrine therapy is a major problem in the clinical management of breast cancer (3), and such loss of response can be modeled in vitro (36). In cell culture systems, breast cancer cells possess a remarkable ability to adapt to prevailing growth conditions and to escape from any imposed growth inhibition. While maintained in the presence of estradiol, estrogen-sensitive breast cancer cell lines remain growth-regulated by estrogen, in that growth is severely reduced by removal of estrogen or addition of anti-estrogen (37). However, upon long term removal of estrogen from the culture medium, the cells gradually adapt, developing an ability to grow without estrogen until they can eventually grow at the same rate as they had done previously only in the presence of estrogen (38 -40). Similar adaptive processes enable the cells to escape from growth inhibition imposed by the anti-estrogens tamoxifen (41) and ICI 182,780 (42) or by retinoic acid (43). The molecular basis for this growth adaptation remains unknown, but recent work has shown that development of estrogen hypersensitivity resulting from up-regulation of ER and increased estrogen-regulated gene expression may be involved (44). In this work, we have investigated whether there is any involvement of an IGF cross-talk pathway through development of hypersensitivity to IGF and increased IGFR levels.

Stock Culture of MCF7 Human Breast Cancer Cells-Stock MCF7
McGrath human breast cancer cells were kindly provided by Dr. K. Osborne at passage number 390 (45) and were dependent on estrogen for growth in monolayer culture as described previously (35). Cells were maintained routinely as monolayer cultures in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 5% fetal calf serum (FCS) (Life Technologies, Inc.), 10 Ϫ8 M 17␤-estradiol (Steraloids, Croydon, UK), and 10 g/ml insulin (Sigma) in a humidified atmosphere of 10% carbon dioxide in air at 37°C. Estradiol was dissolved in ethanol and diluted 1/10,000 in culture medium. All cell stocks were subcultured at weekly intervals by suspension with 0.06% trypsin, 0.02% EDTA, pH 7.3.
Long Term Estrogen Deprivation of Cells-A new vial of cells was thawed from liquid nitrogen at the start of each experiment, which ensured that control cells of the starting passage number were available for comparison at any time. Freshly thawed cells were grown for 2 weeks as stock cultures (see above) and then suspended with phenol red-free 0.06% trypsin, 0.02% EDTA, pH 7.3, washed with phenol red-free RPMI 1640 medium (Life Technologies, Inc.) and replated in phenol red-free RPMI 1640 medium containing 5% dextran-charcoal stripped FCS (DCFCS) (46). Cells were routinely maintained in phenol red-free RPMI 1640 medium with 5% DCFCS, subculturing every 2-3 weeks during the initial period of slow growth, and increasing to subculturing at weekly intervals as the growth rate increased.
Cell Growth Experiments-Cells were suspended from stock plates (either estrogen-maintained or estrogen-deprived) by treatment with phenol red-free 0.06% trypsin, 0.02% EDTA, pH 7.3, added to an equal volume of phenol red-free RPMI 1640 medium (Life Technologies, Inc.) containing 5% DCFCS (36), and counted on a hemocytometer. Cells were then added to the required volume of phenol red-free RPMI 1640 medium containing 5% DCFCS at a concentration of 0.2 ϫ 10 5 cells/ml and plated in monolayer in 0.5-ml aliquots into 24-well plastic tissue culture dishes (Nunc). After 24 h, the medium was changed to phenol red-free RPMI 1640 medium supplemented with 5% DCFCS and the required concentration of 17␤-estradiol, insulin, IGFI (Roche Molecular Biochemicals), IGFII (Bachem), ␣IR3 antibody (Oncogene Science), or mouse IgG (Sigma). Culture medium was changed routinely every 3-4 days in all experiments. Cell counts were performed by counting released nuclei on a model ZBI Coulter Counter, as described previously (46).
Measurement of IGF Receptors-Cells were plated onto 24-well plastic tissue culture dishes in 0.5-ml aliquots in RPMI 1640 medium supplemented with 5% DCFCS at densities from 0.05-0.5 ϫ 10 5 cells per dish and grown for 5 days. Cells were then washed in RPMI 1640 medium, transferred into serum-free medium (RPMI 1640 medium supplemented with 15 mM HEPES buffer, 0.25% w/v bovine serum albumin (fraction V, Sigma)), and left overnight. After 24 h, three dishes of each treatment were used to estimate cell numbers by Coulter counting as above, and three dishes were used to assay IGFR by radioligand binding assay. For the binding assay, cells were washed twice with ice-cold phenol red-free RPMI 1640 medium and incubated with 80,000 cpm of 125 I-IGF-I (specific activity 200 -250 Ci/mmol made by the IODO-GEN method (47)) with or without 30 M insulin in 0.2 ml of phenol red-free RPMI 1640 medium containing 1 mg/ml bovine serum albumin for 3 h at 4°C. After incubation, cells were washed 3 times with ice-cold isotonic saline; the cell layer was solubilized in ice-cold 0.5 M sodium hydroxide, and incorporated radioactivity was measured in a gamma counter. Since insulin at 30 M suppresses only receptor binding and does not bind to binding proteins (48), specific receptor binding was assessed by subtraction of non-receptor binding in the presence of insulin from total binding in the absence of insulin. Results are ex-pressed as fmol 125 I-IGFI bound per 10 6 cells.
Scatchard analysis was carried out as above but by adding increasing amounts of 125 I-IGFI and by using 125 I-IGFI radiolabeled to a specific activity of 2000 Ci/mmol (Amersham Pharmacia Biotech). Results are expressed on the y axis as ratios of bound (specific) IGFI to free IGFI and on the x axis as pmol of IGFI-specific binding per 1,000,000 cells per liter (pM). Values for the dissociation constant (K d ) were calculated from the slope of the best-fit line.
Immunocytochemistry-Stock MCF7 cells were plated at varying cell densities onto 22-mm diameter glass coverslips inside 3.5-cm plastic tissue culture dishes in phenol red-free RPMI 1640 medium containing 5% DCFCS either with or without 10 Ϫ8 M 17␤-estradiol for 3 days. Cells were fixed on each coverslip in methanol at room temperature for 10 min and then washed in PBS. Each coverslip was incubated for 10 min with 20 l of normal rabbit serum/80 l of 0.5 M Tris-buffered saline, pH 7.6 (TBS), and then washed in TBS. Cells were incubated in 100 l of TBS containing ␣IR3 antibody at 1:1000 dilution overnight at 4°C and then washed in 400 ml TBS for 10 min. Cells were incubated in 100 l of TBS containing biotinylated anti-chicken antibody (Dako) at 1:500 dilution at room temperature for 4 h and then washed in 400 ml of TBS for 10 min. Cells were finally incubated with 100 ml of PBS with streptavidin-fluorescein isothiocyanate conjugate (Dako) at 1:100 dilution at room temperature for 1 h, washed in tap water for 10 min, and wet-mounted using vectoshield sealed with cosmetic nail varnish. Microscopy was performed at once using a Leitz confocal microscope.
Western Immunoblotting-Stock MCF7 cells were plated at varying densities onto 24-well plastic tissue culture dishes in RPMI 1640 medium supplemented with 5% DCFCS and 10 Ϫ8 M 17␤-estradiol. After 5-7 days, 3 wells were counted (see above) to provide the cell density at the time of harvest. Other wells were harvested directly into 26 mM Tris-HCl, pH 6.8, 2% w/v SDS, 10% glycerol, 0.01% w/v bromphenol blue. Samples were sonicated and boiled before loading onto SDS-6% polyacrylamide gel electrophoresis. Protein concentrations were determined using the BCA reagent according to manufacturer's instructions (Pierce), and 150 g of protein was loaded per gel track. Rainbowcolored high molecular weight protein markers in the 14 -220-kDa range were loaded as standards to each gel (Amersham Pharmacia Biotech). Proteins were transferred onto Hybond-ECL membranes (Amersham Pharmacia Biotech) by semi-dry Western blotting in 48 mM Tris, 39 mM glycine, 1.3 mM SDS, 20% v/v methanol. The membrane was blocked with 0.1% v/v Tween 20, 5% w/v dried milk in PBS at 4°C for 18 h, and immunoblotted with a 1:100 dilution of monoclonal mouse anti-human IGFIR antibody (Ab-3, Oncogene Research) in 0.1% v/v Tween 20, 1% w/v dried milk in PBS (PBS-TM) for 30 min at room temperature. The epitope for this antibody is the extracellular domain of the type I IGFR. After washing 6 times for 5 min with 0.1% v/v Tween 20 in PBS (PBS-T), the membrane was incubated with a 1:1000 dilution of biotinylated rabbit anti-mouse antibodies (Dako, Copenhagen, Denmark) in PBS-TM for 30 min at room temperature. The membrane was washed again 6 times for 5 min in PBS-T and then incubated with a 1:1500 dilution of streptavidin-horseradish peroxidase conjugate (Amersham Pharmacia Biotech) in PBS-TM for 30 min at room temperature. Enhanced chemiluminescence was performed according to the manufacturer's instructions, and the signal was detected on ECL-Hyperfilm (Amersham Pharmacia Biotech).

RESULTS
Effect of Cell Density on IGFR Levels-MCF7 human breast cancer cells are maintained routinely in the presence of 17␤estradiol in order to maintain sensitivity to estrogen (37,40,43). Fig. 1 shows that these estrogen-maintained MCF7 cells possess receptors capable of binding 125 I-IGF1. However, following a series of experiments in which the cells were plated at different cell densities, the levels of 125 I-IGF1 binding were found to drop substantially as the cell density increased (Fig.  1). Levels per cell dropped by around one-fifth as cell numbers increased 10-fold from low density (50,000 cells per well of a 24-well plate) to high density (500,000 cells per well of a 24-well plate). Short term growth of the cells (up to 14 days) in the absence of estradiol resulted in a markedly reduced growth rate ( Fig. 2A), but again the levels of 125 I-IGF1 binding were found to decrease as cell density increased (Fig. 1). When comparing between cells at equivalent densities, short term removal of estradiol did not appear to have any effect on levels of 125 I-IGF1 binding (Fig. 1).
Immunofluorescent confocal microscopy was used to confirm and to visualize these effects of cell density on IGFR levels. Fig.  3 shows the immunofluorescent staining of IGFR localized on the cell membrane. The fluorescent staining was much more intense for cells at low density in small clusters on a culture dish (Fig. 3A) than for cells at high density (Fig. 3B). When the cells formed an even monolayer on a culture dish, almost no fluorescent staining could be seen anymore (Fig. 3B). When cells were in large monolayer groups but at subconfluence, staining tended to be more intense around the edges of those cells not in contact with adjacent cells.
The effects of cell density on IGFR levels were further investigated using Western immunoblotting. Antibody Ab-3 was raised to the extracellular domain of the type I IGFR and accordingly detected a band on Western immunoblotting of 135 kDa in size corresponding to the ␣-subunits of the IGFIR in MCF7 cells (Fig. 4). This band was more intense in cells at low density (Fig. 4, track 1) than at high density (Fig. 4, track 3).
Effect of Estrogen Deprivation on Cell Growth and IGFR Levels-When MCF7 cells are maintained in the continuous presence of estradiol, the cells remain dependent on estradiol for growth. Short term removal of estradiol results in a much reduced growth rate ( Fig. 2A). However, following long term estrogen deprivation, short term growth of the cells is no longer influenced by estradiol (Fig. 2B). This occurs by a gradual adaptive process in which the cells gradually develop an ability over time to increase their growth rate in the absence of estrogen. A representative time course is shown in Fig. 5. The increase in basal growth rate in the absence of estrogen occurred gradually over a period of 12 weeks such that at the end growth in the absence of estrogen had increased to the same rate as that at which the cells had originally grown only in the presence of estradiol (Fig. 5). This time course is reproducible since the results shown in Fig. 5 have been repeated 3 times for this cell line.
Levels of IGFR were found to also increase with increasing periods of estrogen deprivation (Fig. 6). At all time points, the levels of IGFR continued to decrease as cell density increased, but the curves were shifted to higher values as length of estro-gen deprivation increased (Fig. 6). When the effect of cell density was taken into account, short term readdition of estradiol now increased levels of IGFR (Fig. 6). Thus, although levels of IGFR had been unaffected in the short term by estradiol in the estrogen-maintained cells (Fig. 1), IGFR levels were now increased by estradiol following estrogen deprivation (Fig. 6).
Scatchard analysis was performed in order to ascertain whether there were any alterations in binding affinity following long term steroid deprivation (Fig. 7). However, K d values were similar for estrogen-maintained cells (Fig. 7A, K d 0.11 nM) after 1 week of estrogen deprivation (Fig. 7B, K d 0.11 nM), after 108 weeks of estrogen deprivation (Fig. 7C, K d 0.13 nM), and after 108 weeks of estrogen deprivation followed by 1 week of estrogen readdition (Fig. 7D, K d 0.12 nM). The K d values are in line with previously published (49) values for MCF7 cells assayed using similar methodology.

Effect of IGFR Blockade on Growth of Estrogen-deprived
Cells-In order to determine whether the increased IGFR levels play any role in the increased basal growth rate of estrogendeprived cells, we used the ␣IR3 antibody that has biological blocking activity on IGFIR activity (50). Addition of the ␣IR3 antibody was able to reduce growth of the estrogen-deprived cells in both the absence of estrogen (Fig. 8 p ϭ 0.002 for ϪEϩIgG (track 3) versus ϪEϩ␣IR3 (track 4)) and the presence of estrogen (Fig. 8 p ϭ 0.006 for ϩEϩIgG (track 6) versus ϩEϩ␣IR3 (track 7)), although only to a limited extent. This was reproducible over 3 separate experiments using two different batches of ␣IR3 antibody.
Sensitivity of Estrogen-deprived Cells to Insulin and IGFII-Since receptor levels are important in determining cellular sensitivity to ligands, we studied the effects of exogenous insulin, IGFI, and IGFII on the estrogen-deprived cells that had raised IGFR levels. Interestingly, whereas all these ligands stimulated growth of estrogen-maintained cells, all ligands showed a dose-dependent inhibition of growth of the estrogendeprived cells (Fig. 9). Although insulin increased growth of estrogen-maintained cells at 1 and 10 g/ml (Fig. 9A, tracks 3 and 4), growth of the estrogen-deprived cells was inhibited by 10 g/ml insulin (Fig. 9B, track 3 (p Ͻ 0.001)). Whereas 10 and 100 ng/ml IGFI stimulated growth of the estrogen-maintained cells (Fig. 9A, tracks 5 and 6), 100 ng/ml IGFI became growthinhibitory to the estrogen-deprived cells (Fig. 9B, track 5 (p Ͻ 0.001)). Whereas 10 and 100 ng/ml IGFII stimulated growth of the estrogen-maintained cells (Fig. 9A, tracks 7 and 8), 100 ng/ml IGFII became growth-inhibitory to the estrogen-deprived cells (Fig. 9B, track 7 (p Ͻ 0.001)). The inhibitory action of insulin has been reproducible over at least 4 separate experiments. The inhibitory action of IGFI and IGFII has been repeated twice with similar results.
Studies using the ␣IR3 antibody confirmed previously published work (5) that ␣IR3 inhibits growth stimulation by IGFI and IGFII but not by insulin in the estrogen-maintained cells (data not shown). However, addition of ␣IR3 could reverse the inhibitory action by all three ligands (10 g/ml insulin, 100 ng/ml IGFI, and 100 ng/ml IGFII) in the estrogen-deprived cells (Fig. 9B, tracks 3-8). Addition of ␣IR3 removed the inhibitory action of 10 g/ml insulin (p ϭ 0.55 for Fig. 9B, track 4 versus  track 2) and of 100 ng/ml IGFII (p ϭ 0.10 for Fig. 9B, track 8  versus track 2). Addition of ␣IR3 reduced the inhibitory effect of IGFI significantly (p ϭ 0.001) but did not abolish the effect (p Ͻ 0.001 Fig. 9B, track 6 versus track 2). DISCUSSION The results presented here demonstrate a reciprocal relationship between levels of IGFR and cell density during the growth of MCF7 human breast cancer cells in monolayer culture. IGFR levels were found to decrease with increasing cell density irrespective of the presence or absence of estradiol and irrespective of the estrogen responsiveness of the cells. Regulation of cell density by contact inhibition and anchorage dependence is fundamental to cell growth control, and downregulation of receptor for a growth stimulator with increasing cell density would provide a logical contributory mechanism. The reduced binding of both 125 I-IGF1 and ␣IR3 anti-IGFR antibody to whole cells suggests that IGFR are indeed lost from the MCF7 cell surface at increasing density. Western immunoblotting showed that cellular levels of type I IGFR were reduced with increasing density.
Although alteration of IGFR levels with cell density has not been described previously in human breast cancer cells, the phenomenon has been well documented for other receptors in other cell lines. Levels of epidermal growth factor (EGF) receptors (26, 28 -30), transforming growth factor-␤ receptors (26), platelet-derived growth factor receptors (26), fibroblast growth factor receptors (26,27), IGF type II receptor (25), hepatocyte growth factor receptor (31), tumor necrosis factor receptor (32), ciliary neurotrophic factor receptor (33), and vascular endothe- lial growth factor receptor (KDR) (34) have all been shown to be reduced as cell density increases. Furthermore, a similar trend has been shown for growth factor-binding proteins (IGF-binding protein (51); heparan sulfate proteoglycans (52)) and for downstream signaling components (from EGF receptors (53); from transforming growth factor-␤ receptors (54); from interleukin 1␤ receptor (55)), suggesting that whole signal transduction pathways can be down-regulated with increasing cell density. Density-dependent effects on EGF receptors have been shown to operate through extensive cell-cell adhesion that is mediated by excess E-cadherin (56), and in this respect it is interesting that overexpression of IGFIR in MCF7 cells has been reported to induce E-cadherin-mediated cell-cell adhesion (49). It is also interesting to note that regulation of the ErbB-2 protein is different, showing increasing activity in confluent cultures (57), and overexpression of this EGF receptor family member has been correlated with a worse prognosis in breast cancer (58). This regulation by cell density may explain the differing levels of IGFR reported in breast cancer cells in the literature (59) since the IGFR measurements may simply have been made on cells at different densities. Cell density will now need to be taken into account in any future assessments of cellular IGFR levels and particularly in studies of regulation where the regulator might alter cell density through effects on proliferation or apoptosis.
When taking cell density into account, we could not find any short term (5 days) regulation of IGFR levels by estradiol in the stock estrogen-maintained cells (Fig. 1). It was only following long term estrogen deprivation when we observed any short term estrogen regulation of IGFR levels (Fig. 6). Although IGFR levels have been reported to be dependent on estrogen (9,14), this has not been a universal finding in all reports (41). Although results presented here suggest that discrepancies may be explained by lack of consideration to cell density in assays, another explanation could lie with the different culture conditions used. Our MCF7 cells are maintained routinely in the presence of estradiol as well as insulin (see "Experimental Procedures"), whereas other studies use only insulin, and memory effects from exposure to estrogen are known to last in cells for up to 14 days (37). Such memory effects have been interpreted to relate to the time taken to lose patterns of estrogeninduced gene expression and specifically to the half-lives of estrogen-regulated mRNAs and proteins in the cells. While some mRNAs such as c-Myc can be up-and down-regulated in minutes by estradiol (60), other estrogen-regulated mRNAs such as pS2 can take several days to down-regulate after estrogen removal (61). These studies suggest that the cells need to be deprived of estrogen for several weeks in order to measure any dependence of IGFR on estrogen.
However, long term estrogen deprivation had the effect not only of revealing an estrogen regulation of IGFR but also of  , track 9, and B, track 9). The effect of addition of ␣IR3 antibody at 5 g/ml was assayed in tracks B4, B6, and B8. The plating density of the experiments is indicated (PE) (A, track 1, and B,  track 1). Bars indicate the standard error of triplicate dishes, and where not seen, the error was too small for visual display.
generally increasing IGFR levels in the cells. As length of estrogen deprivation increased from 8 to 52 to 96 weeks, the general levels of IGFR increased in the cells, and the short term estrogen regulation of the IGFR became more marked. It is interesting to note that in the 96-week deprived cells, the IGFR levels had become similar at the highest cell density in the absence of estradiol to the IGFR levels at the lowest cell density in the estrogen-maintained cells. This up-regulation of IGFR may be a further part of multiple processes of adaptation to growth without estrogen described by others (44) to involve up-regulation of ER and increased ER target gene activation. Such raised levels of ER in the cells have been suggested to be important in enabling the cells to respond to reduced environmental levels of estrogen (44). In an analogous way, raised IGFR levels may enable the cells to respond to the reduced IGF levels of the DC-stripped FCS or low levels of IGFII produced endogenously (20). Transfection experiments have already demonstrated the importance of adequate IGFR levels in the response of breast cancer cells to IGFII in cell culture (19) and that increased levels of transfected IGFR can alter sensitivity of breast cancer cells to exogenously added IGFs (22). Furthermore, overexpression of fibroblast growth factor receptor in breast cancer cells has been shown to reverse responses to added fibroblast growth factor from growth stimulatory to growth inhibitory (62). Interestingly, overexpression of IGFR in the estrogen-deprived cells here also correlated with a reversal of responses to exogenously added ligand, in that insulin, IGFI, and IGFII also became growth inhibitory rather than growth stimulatory, and this growth inhibition could be reduced with the ␣IR3 antibody that has biological blocking activity on the IGFIR function (50).
The fact that blockade of the IGFIR with the ␣IR3 antibody could, at least partially, inhibit the growth of the estrogendeprived cells suggests that this up-regulation of IGFR must serve some function in the loss of dependence on estrogen for growth. However, it is clearly not the whole story since the ␣IR3 did not block all the growth. Likewise, previous authors (63) have felt that increase in androgen receptor in LNCaP prostate cancer cells and increase in ER in MCF7 human breast cancer cells (44) could only partially explain increased androgenic and estrogenic activities, respectively. It is therefore possible that overall mechanisms of adaptation to growth in steroid-deprived conditions involve multiple cross-talk pathways rather than a single pathway. Specifically, up-regulation of ER (44) and IGFR (reported here) simultaneously in MCF7 cells may interact in the adaptive processes during estrogen deprivation, through the ability of ER to increase components of IGFR signaling (7,8,(13)(14)(15)(16) and to influence AP1-mediated pathways (23), and the ability of IGFR to influence levels of ER (17) and of estrogen-regulated gene expression (18,19).