SCREENING METHOD FOR ESTROGENIC AND ANTI-ESTROGENIC ACTIVITY

Abstract

The present invention relates to a cell as deposited under accession number DSM ACC332 and derivatives thereof. Furthermore, the present invention relates to methods for testing a compound of interest for potential estrogenic or anti-estrogenic activity as well as to methods for testing a cell for its potential to form discontinuous basolateral adherens junctions in presence of an anti-estrogenic compound. The application relates also to uses of a cell capable of forming discontinuous basolateral adherens junctions. The present invention relates furthermore to methods of analysing the morphology of adherens junctions in tissue of a subject and a respective kit.

Claims

1. A cell as deposited under accession number DSM ACC3321 or a cell deriving from the cell as deposited under DSM ACC3321 and forming discontinuous basolateral adherens junctions in presence of an anti-estrogenic compound.

2. A method for testing a compound of interest for potential estrogenic activity and/or level of estrogenic activity, the method comprising the following steps: a) growing a cell in presence of an anti-estrogenic compound and in in presence of the compound of interest, wherein growing the cell in presence of the anti-estrogenic compound may precede and/or occur in parallel to growing the cell in presence of the compound of interest, and b) determining the adherens junctions morphology of the cell, wherein the cell is a cell responding to the exposure to an anti-estrogenic compound by formation of discontinuous basolateral adherens junctions, and wherein prevention, reduction or reversion, respectively, of a discontinuous basolateral adherens junctions phenotype is indicative of estrogenic activity of the compound of interest.

3. The method of claim 2, wherein growing the cell in presence of the anti-estrogenic compound occurs in parallel to growing the cell in presence of the compound of interest.

4. The method of claim 2, wherein growing the cell in presence of the anti-estrogenic compound precedes growing the cell in presence of the compound of interest.

5. The method of claim 2, wherein the anti-estrogenic compound is selected from the group consisting of fulvestrant, tamoxifen, ZK 164015, MPP dihydrochloride, and (Z)-4-hydroxytamoxifen, in particular wherein the anti-estrogenic compound is fulvestrant.

6. A method for testing a compound of interest for potential anti-estrogenic activity and/or level of anti-estrogenic activity, the method comprising the following steps: a) growing a cell in presence of the compound of interest, and b) determining the adherens junctions morphology of the cell, wherein the cell is a cell responding to the exposure to an anti-estrogenic compound by formation of discontinuous basolateral adherens junctions, and wherein a formation of discontinuous basolateral adherens junctions is indicative of anti-estrogenic activity.

7. The method of claim 2, wherein the cell is a cell according to claim 1.

8. A method for testing a cell for its potential to form discontinuous basolateral adherens junctions in presence of an anti-estrogenic compound, the method comprising the following steps: a) growing the cell in presence of an anti-estrogenic compound, and b) determining the adherens junctions morphology of the cell.

9. The method of claim 8, wherein the anti-estrogenic compound is selected from the group consisting of fulvestrant, tamoxifen, ZK 164015, MPP dihydrochloride, and (Z)-4-hydroxytamoxifen, in particular wherein the anti-estrogenic compound is fulvestrant.

10. The method of claim 2, wherein the cell is i) a mammalian cell, preferably a human cell, and/or ii) an adenocarcinoma cell, preferably a breast adenocarcinoma cell.

11. An ex vivo method for assessing the basolateral adherens junction morphology of cells in a tissue sample of a subject, the method comprising determining in the tissue sample of the subject the fraction of cells exhibiting discontinuous basolateral adherens junctions and/or determining in the tissue sample of the subject the fraction of cells exhibiting continuous basolateral adherens junctions.

12. An ex vivo method for assessing the metastatic potential of a tumor in a subject, the method comprising determining in a tissue sample of the tumor the fraction of cells exhibiting discontinuous basolateral adherens junctions and/or determining in the tissue sample of the tumor the fraction of cells exhibiting continuous basolateral adherens junctions, wherein a reduced fraction of cells exhibiting discontinuous basolateral adherens junctions compared to a control and/or an increased fraction of cells exhibiting continuous basolateral adherens junctions compared to a control indicates an elevated metastatic potential of the humor.

13. An ex vivo method for assessing the metastatic potential of an estrogen receptor alpha positive tumor in a subject, the method comprising determining in a tissue sample of the subject the fraction of cells exhibiting essentially cytoplasmatic localization of estrogen receptor alpha and/or the step of determining in the tissue sample of the subject the fraction of cells exhibiting essentially nuclear localization of estrogen receptor alpha, wherein a reduced fraction of cells exhibiting essentially cytoplasmatic localization compared to a control and/or an increased fraction of cells exhibiting essentially nuclear localization indicates an elevated metastatic potential.

14. The method of claim 2, wherein determining adherens junction morphology comprises the use of optical means.

15. The method of claim 14, wherein the optical means comprise fluorescence microscopy, in particular optical sectioning fluorescence microscopy.

16. The method of claim 14, wherein determining the adherens junctions morphology comprises detecting the distribution and/or mean signal intensity of E-cadherin at the adherens junctions of the cell.

17. The method of claim 16, wherein E-cadherin is detected by means of immunostaining or by use of labelled E-cadherin, in particular wherein E-cadherin is detected by means of fluorescently labelled E-cadherin.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0099] In the following a brief description of the appended figures will be given. The figures are intended to illustrate the present invention in more detail. However, they are not intended to limit the scope of the invention to these specific examples only.

[0100] FIG. 1 illustrates by means of E-Cadherin staining the effect of fulvestrant on adherens junction morphology of MCF7/vBOS cells. A) Immunofluorescence images showing increased intercellular spacing and formation of discontinuous adherens junctions (honeycomb-like structure, resulting from reduced cell-cell contact) demarcated by immunofluorescence staining for E-Cadherin upon fulvestrant treatment for 48 hours (right panel) compared to the solvent control (left panel). White boxes indicate enlarged cell membrane areas shown in a-a. Scale bars, 10 m. B) Immunofluorescent staining of MCF7/vBOS cells treated with solvent control or increasing concentrations (10.sup.11 to 10.sup.7 M) of fulvestrant for 48 hours and stained for E-Cadherin demonstrate the appearance of discontinuous adherens junctions upon treatment with a minimal concentration of 10.sup.9 M fulvestrant for 48 hours. Scale bars, 10 m. C) Immunofluorescence staining for E-Cadherin showing adherens junction organization after fulvestrant treatment for 24 hours and subsequent cultivation for additional 24 or 48 hours in fulvestrant-free medium containing only the solvent ethanol revealed that discontinuous adherens junctions are stable for at least 48, independent of the presence of fulvestrant. Scale bars, 10 m.

[0101] FIG. 2 illustrates that the effect attained with fulvestrant in MCF7/vBOS cells can also be attained with other anti-estrogens such as A) tamoxifen, B) ZK 164015, C) MPP dihydrochloride, and D) (Z)-4-hydroxytamoxifen. MCF7/vBOS cells were incubated with 10.sup.8 and 10.sup.6 M tamoxifen for 144 hours (A), 10.sup.9 and 10.sup.7 M ZK164015 for 48 hours (B), 10.sup.8 and 10.sup.6 M MPP hydrochloride for 144 hours (C), or with 10.sup.9 and 10.sup.8 M (Z)-4-hydroxytamoxifen for 72 hours (D), and immunofluorescence staining for E-Cadherin performed. Discontinuous adherens junctions became apparent with all four anti-estrogens although at different concentrations and after different incubation times. From these data one can also conclude that fulvestrant is more active in this assay than tamoxifen, ZK164015, MPP or (Z)-4-hydroxytamoxifen. Scale bar, 10 m.

[0102] FIG. 3 illustrates by means of E-Cadherin staining that the MCF7/vBOS cell line exhibits unique properties. The figure shows that the MCF7/vBOS cell line (A) exhibits in absence of the anti-estrogenic compound fulvestrant a normal (continuous) basolateral adherens junction morphology (left panel), while said morphology changes in presence of 10.sup.8 M fulvestrant for 48 hours (right panel). In contrast, the parental MCF-7 subclone MCF-7/BOS (B) as well as the MCF-7/ACC115 cell line (C) do show the same continuous adherens junctions morphology in absence (left panels) and presence of fulvestrant (right panels). White boxes indicate enlarged cell membrane areas shown in a-c and a-c. Scale bars, 10 m.

[0103] FIG. 4 illustrates the prevention of the formation of discontinuous adherens junctions, if an estrogen, here 17-estradiol (E2), is additionally present in the cell culture medium along with fulvestrant. Staining of MCF7/vBOS cells for E-Cadherin by immunofluorescence demonstrates that the formation of discontinuous adherens junctions by treatment with 10.sup.8 M fulvestrant (left picture) can be prevented by co-incubation with increasing concentrations (10.sup.10 to 10.sup.7 M) of 17-estradiol (E2). Already co-treatment with 10.sup.9 M E2 suffices to preserve continuous adherens junctions that are indistinguishable to the one observed in untreated control cell (right picture). Scale bar, 10 m.

[0104] FIG. 5 illustrates again the prevention of the formation of discontinuous adherens junctions, if an estrogenic compound, here bisphenol A, is additionally present in the cell culture medium along with fulvestrant. Immunostaining for E-cadherin on cells treated with 10.sup.8 M fulvestrant and 10.sup.5 or 10.sup.4 M bisphenol A, respectively. Scale bar, 10 m.

[0105] FIG. 6 illustrates again the prevention of the formation of discontinuous adherens junctions, if an estrogenic compound, here nonylphenol, is additionally present in the cell culture medium along with fulvestrant. Immunostaining for E-cadherin on cells treated with 10.sup.8 M fulvestrant and 10.sup.5 nonylphenol. Scale bar, 10 m.

[0106] FIG. 7 illustrates the results of an automated quantification of morphology changes at adherens junctions. A) CellProfiler/CellProfiler-Analyst-based quantification of cells displaying continuous and discontinuous AJs. Bars represent the fraction of all cells classified as Continuous AJ (circles) and Discontinuous AJ (triangles) upon fulvestrant treatment for 48 hours (grey bars) compared to the solvent control (black bars). Biological replicates, n=3. Bars, mean+/s.d. Representative pictures of the analysed cells are shown in the left panel that also indicates cells that were classified as Continuous AJ (circles) or as Discontinuous AJ (triangles) in control or fulvestrant treated cells. B) CellProfiler/CellProfiler-Analyst-based quantification of the fraction of cells displaying continuous AJs (circle) and discontinuous AJs (triangle) (see FIG. 7 a) upon ER signalling inhibition (fulvestrant titration, left panel, see FIG. 1B), and restoration (fulvestrant at 10.sup.8 M in combination with 17-estradiol titration; right panel, see FIG. 4) for 48 hours (grey bars) compared to the solvent control (black bars). Bars, mean of three replicate images. C) Quantification of the mean membrane signal intensity of cells displaying continuous AJs (normalized signal intensity value 0.9-1.0) and discontinuous AJs (normalized signal intensity value >1.0) upon ER signalling inhibition (fulvestrant titration, left panel, see FIG. 1B), and restoration (fulvestrant at 10.sup.8 M in combination with 17-estradiol titration; right panel, see FIG. 4) for 48 hours normalized to the solvent control. Biological replicates, n=3. Bars, mean+/s.d.

[0107] FIG. 8 shows immunofluorescence images illustrating organization of adherens junctions including E-cadherin (E-Cad) and the cytoplasmic adaptor proteins -Catenin, -Catenin and p120-Catenin that connect E-Cad to the underlying actomyosin network (F-Actin staining) upon fulvestrant treatment for 48 hours compared to the solvent control. Scale bar, 10 m.

[0108] FIG. 9 provides a CellProfiler/CellProfiler-Analyst-based quantification (analogous to FIG. 7) showing the rate of cell rounding over the course of 120 minutes in MCF7/vBOS cells pre-treated with fulvestrant for 48 hours (dotted bars) compared to solvent control cells (black bars). Indicated time points depict time after application of EGTA-containing medium (EGTA, 810.sup.3 M) to disrupt adherens junctions. Biological replicates, n=3. Error bars, mean+/s.e.m. Statistical testing, one-way ANOVA with Sidak's correction for multiple comparisons; *, p<0.05; **, p<0.01; ***, p<0.005; ****, p<0.001. Scale bar, 50 m.

[0109] FIG. 10 illustrates that inhibition of ER signaling increases tissue stability by modulating cell mechanics and cell motility. (A) Quantification of cell stiffness (Apparent elastic (Young's) modulus) by atomic force microscopy (AFM) indentation measurements on cells treated with fulvestrant for 48 hours (E.sub.Repl.1=0.71410.1193 kPa; E.sub.Repl.2=0.6510.1407 kPa; E.sub.Repl.3=0.59380.1564 kPa; means.d.) compared to solvent control cells (, E.sub.Repl.1=0.48410.1194 kPa; E.sub.Repl.2=0.40880.1513 kPa; E.sub.Repl.3=0.46600.1189 kPa). Biological replicates, n=3; 50-60 cells per condition. Boxes, 25th, 50th (median), and 75th percentiles. Whiskers, 10th and 90th percentiles. Cross, mean for each group. Statistical testing, Mann-Whitney tests; **, p values<0.01. (B) Quantification of cell motility (velocity) by manual tracking of cells treated with fulvestrant for 48 hours (V.sub.Repl.1=0.06130.0176 m/min; V.sub.Repl.2=0.05530.0157 m/min; V.sub.Repl.3=0.0510.014 m/min; means.d.) compared to solvent control cells (V.sub.Repl.1=0.1130.0237 m/min; V.sub.Repl.2=0.07570.023 m/min; V.sub.Repl.3=0.08270.0174 m/min). Biological replicates, n=3; 3 fields with 10 cells each per replicate and condition were imaged every 10 minutes over the course of 16 h. Boxes, 25th, 50th (median), and 75th percentiles. Whiskers, 10th and 90th percentiles. Cross, mean for each group. Statistical testing, Mann-Whitney tests; **, p values<0.01.

[0110] FIG. 11 illustrates that human breast cancer sections can differ in terms of adherens junction organization and that ER signalling activity correlates with adherens junction organization. The figure shows CellProfiler/CellProfiler-Analyst-based quantification (analogous to FIG. 7) of the fraction of cells displaying continuous or discontinuous adherens junctions (black bars) and nuclear or cytoplasmic ER localization (white bars) in breast cancer tissue sections of 10 out of 54 sample images that comprised a greater number of detected cells (>60) than used for supervised training of the image analysis pipeline. Each sample represents an individual patient except of samples 1 and 2, which originate from the same patient, and samples 7 and 8, which originate from a further patient. Bars, fraction of cells per image.

EXAMPLES

[0111] In the following, specific examples illustrating various embodiments and aspects of the invention are presented. However, the present invention shall not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will become readily apparent to those skilled in the art from the foregoing description, accompanying figures and the examples below. All such modifications fall within the scope of the appended claims.

Example 1: Cell Line MCF7/vBOS

[0112] MCF7/vBOS (Michigan Cancer Foundation-7/variantBOS) cells have evolved from the original human breast adenocarcinoma MCF-7/BOS cell line by natural selection under laboratory conditions. Cell cultures from a cryopreserved master cell bank were routinely maintained in cell culture flasks (TPP) at 37 C. with 5% CO2 in low estrogen serum medium (Dulbecco's modified Eagle's medium (DMEM, Biochrom), 10% FBS (Biochrom, S 0615, LOT 1353W), 100 g/ml streptomycin/100 U/ml penicillin (Biochrom)). Cells were subcultured at 70-80% confluency over a maximum of 8-10 passages to minimize passage number-related effects, and regularly tested for bacterial contamination using a mycoplasma test service (GATC Biotech AG).

[0113] The cell line has been deposited at the Leibniz Institut DSMZ-Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH (DSZM) under accession number DSM ACC3321.

Example 2: ER Antagonist Fulvestrant Triggers Formation of Discontinuous Adherens Junctions

[0114] MCF7/vBOS cells were grown on glass coverslips in multi-well tissue culture plates (TPP) for 24 h and then treated for at least 48 h with 0.1% EtOH (solvent control) and different concentrations (10.sup.11 M, 10.sup.10 M, 10.sup.9 M, 10.sup.8 M, 10.sup.7 M) of the ER antagonist fulvestrant (Sigma-Aldrich, CAS-No. 129453-61-8). During experiments, medium was exchanged on a daily basis. All experiments were performed at 37 C. with 5% CO.sub.2 in reduced-serum medium (phenol red-free DMEM (Gibco), 5% FBS (Biochrom, S 0615, LOT1353 W), 1% glutamine, 100 g/ml streptomycin/100 U/ml penicillin (Biochrom)).

[0115] After treatment with fulvestrant the adherens junction morphology of the cells was analysed by means of immunofluorescence microscopy. Briefly, cells were fixed with 3.7% formaldehyde in PBS for 15 min, permeabilized with 0.2% Triton X-100 in PBS for 30 min, and blocked with 5% FBS in PBS for 60 min (all steps at room temperature). Incubation with primary and secondary antibodies in PBS containing 1.5% BSA was carried out over night at 4 C. and for 1 h at room temperature, respectively. The following antibodies/dyes were used: anti-E-Cadherin (H-108, Santa Cruz); DAPI (Roche); secondary antibody conjugated to Cy3 (alternatives are, e.g.: Cy2, Alexa Fluor 488/555/647; all Invitrogen). The samples were then mounted in Dako fluorescence mounting medium (Dako) or VectaShield (Vector Labs). All images were acquired with an Axio Observer.Z1 widefield microscope equipped with an Apotome.2 device for optical sectioning using structured illumination (Zeiss). After 48 hours the fraction of cells displaying discontinuous adherens junctions was determined by CellProfiler/CellProfer-analyst software (see example 7).

[0116] The result of this experiment is shown in FIGS. 1 and 7. Fulvestrant triggers the formation of discontinuous adherence junctions between adjacent MCF7/vBOS cells throughout the epithelial monolayer within two days after treatment. Upon formation, discontinuous cell contacts are even preserved for at least two days without further fulvestrant addition. The discontinuity of cell contacts is restricted to the level of the basolateral adherence junctions marked by E-Cadherin staining.

Example 3: Other ER Antagonists Likewise Trigger Formation of Discontinuous Adherens Junctions

[0117] In view of the results obtained with MCF7/vBOS cells upon exposure to fulvestrant the inventors speculated that a similar phenotype might be obtainable with other ER antagonist as well. To test this hypothesis, the inventors used the ER antagonists tamoxifen (Sigma-Aldrich, CAS-No. 10540-29-1), MPP dihydrochloride (Sigma-Aldrich, CAS-No. 911295-24-4), ZK164015 (Tocris, CAS-No. 177583-70-9), and (Z)-4-hydroxytamoxifen (Tocris, CAS-No. 68047-06-3). The same protocols as in example 2 were used with the following deviations: [0118] Tamoxifen: Cells were grown for 24 h and then treated for 144 h with 10.sup.8 and 10.sup.6 M tamoxifen. [0119] MPP dihydrochloride: Cells were grown for 24 h and then treated for 144 h with 10.sup.8 and 10.sup.6 M MPP dihydrochloride. [0120] ZK164015: Cells were grown for 24 h and then treated for 48 h with 10.sup.9 and 10.sup.7 M ZK164015. [0121] (Z)-4-hydroxytamoxifen: Cells were grown for 24 h and then treated for 72 h with 10.sup.9 and 10.sup.8 M (Z)-4-hydroxytamoxifen.

[0122] The result of this experiment is shown in FIG. 2. In analogy to fulvestrant, all three tested ER antagonists trigger the formation of discontinuous adherence junctions between adjacent MCF7/vBOS cells throughout the epithelial monolayer within two to six days after treatment, as evidenced by E-Cadherin staining.

[0123] The MCF7/vBOS cell line is thus a suitable tool for testing a compound of interest for potential anti-estrogenic activity by exposing said cell line to the compound of interest and determining the adherens junction morphology of the cell, wherein the formation of discontinuous basolateral adherens junctions is indicative of anti-estrogenic activity.

Example 4: Other MCF-7 Cell Lines do not Show Changes in Basolateral Adherens Junctions Morphology Upon Contact with Anti-Estrogenic Compounds

[0124] Next, the inventors tried to reproduce the effects obtained with the cells as deposited under DSM ACC3321 with other conventional MCF-7 cells, namely MCF-7/BOS provided by Prof. Ana Soto (Tufts University School of Medicine, Boston, Mass., USA) and the MCF-7/ACC115 clone obtained from the Leibniz-Institut DSMZ-Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH (Braunschweig, Germany).

[0125] Briefly, the MCF-7 cells were seeded and grown on glass coverslips in multi-well tissue culture plates (TPP) for 24 h to reach near 100% confluence and then treated for at least 48 h with 0.1% EtOH (solvent control) and the ER antagonist fulvestrant (Sigma-Aldrich, CAS-No. 129453-61-8) at a concentration of 10.sup.8 M. During experiments, medium was exchanged on a daily basis. All experiments were performed at 37 C. with 5% CO.sub.2 in reduced-serum medium (phenol red-free DMEM (Gibco), 5% FBS (Biochrom, S 0615, LOT1353 W), 1% glutamine, 100 g/ml streptomycin/100 U/ml penicillin (Biochrom)).

[0126] The result of this experiment is shown in FIG. 3. None of the conventional MCF-7 cells showed a comparable phenotype as the MCF7/vBOS cells deposited under DSM ACC3321. The reasons for this discrepancy are not known to the inventors.

Example 5: Changes to Adherens Junctions Morphology Induced by Fulvestrant can be Prevented by 17--Estradiol (E2)

[0127] In a next step, the inventors tried to elucidate whether the phenotype induced by exposure to anti-estrogenic compounds such as fulvestrant might be neutralized by stimulating estrogen signalling. For this purpose, the inventors used the estrogen 17--estradiol (E2) (Sigma-Aldrich, CAS-No. 50-28-2).

[0128] Briefly, MCF7/vBOS cells were grown on glass coverslips in multi-well tissue culture plates (TPP) for 24 h and then treated for 48 h with 0.1% EtOH (solvent control), the anti-estrogen fulvestrant in the fixed concentration of 10.sup.8 M and different concentrations (10.sup.10 M, 10.sup.9 M, 10.sup.8 M, 10.sup.7 M) of the ER agonist 17-estradiol. The solvent control was used as a positive control and fulvestrant at a concentration of 10.sup.8 M without 17-estradiol addition was used as a negative control. Sample preparation and immunofluorescence microscopy was performed as described in example 2 above. After 48 hours the fraction of cells displaying discontinuous adherens junctions was determined by CellProfiler/CellProfer-analyst software (see example 7).

[0129] The result of this experiment is shown in FIGS. 4 and 7. The analysis revealed that the fulvestrant-induced phenotype can be prevented by exposure to 17--estradiol (E2) at a minimal concentration of 10.sup.9 M.

Example 6: Changes to Adherens Junctions Morphology Induced by Fulvestrant can Also be Prevented by Estrogenic Industrial Chemicals

[0130] In view of the results obtained with MCF7/vBOS cells upon exposure to fulvestrant and 17--estradiol (E2) the inventors speculated that the phenotype inducible by anti-estrogens like fulvestrant could also be prevented with other estrogenic compounds. To test this hypothesis the inventors used the known estrogenic compounds bisphenol A (BPA) (Sigma-Aldrich, 239658) and nonylphenol (NP) (Sigma-Aldrich, 290858).

[0131] The same protocols as in example 2 were used with the following deviations: [0132] Bisphenol A: Cells were grown for 24 h and then treated for 48 h with 10.sup.8 M fulvestrant and 10.sup.5 or 10.sup.4 M bisphenol A, respectively, diluted in DSMO. [0133] Nonylphenol: Cells were grown for 24 h and then treated for 48 h with 10.sup.8 M fulvestrant and 10.sup.5 or 10.sup.4 nonylphenol, respectively, diluted in ethanol.

[0134] The result of this experiment is shown in FIGS. 5 and 6. The estrogenic compounds bisphenol A and nonylphenol were both able to prevent the fulvestrant-induced phenotype at a minimal concentration of 10.sup.5 M as at these concentrations the E-cadherin distribution was also similar to the ethanol treated solvent control.

[0135] Thus, the MCF7/vBOS cell line is also a tool suitable for testing a compound of interest for potential estrogenic activity, for example by exposing said cell line to fulvestrant or any other suitable anti-estrogen and to the compound of interest and determining the adherens junctions morphology of the cell, wherein a prevention of a discontinuous basolateral adherens junction phenotype is indicative of estrogenic activity.

Example 7: Automated Quantification of Morphology Changes at Adherens Junctions

[0136] To quantify the change in cell morphology upon fulvestrant treatment, the inventors implemented a software-based image analysis pipeline (software: CellProfiler; http://cellprofiler.org and CellProfiler Analyst, http://cellprofiler.org/cp-analyst/)) thereby greatly facilitating classification of cells into cells with continuous adherens junctions and discontinuous adherens junctions.

[0137] Briefly, cell culture, sample preparation and immunofluorescence microscopy was performed as described in example 2 (ER signalling inhibition (fulvestrant titration)) and example 5 (ER signalling restoration (fulvestrant at 10.sup.8 M in combination with 17-estradiol titration))..

[0138] Subsequently, quantitative image analysis was performed using a pipeline that builds on the CellProfiler software (Carpenter et al., Genome Biol., 2006; 7(10):R100.) for segmentation and extraction of cellular parameters, and on the CellProfiler Analyst software (Jones et al., BMC Bioinformatics, 2008 Nov. 15; 9:482) for parameter-based classification of cells. In the segmentation process, primary objects (nuclei) were identified from DAPI staining by global thresholding using Otsu's method. Next, primary objects served as seeding points for identification of secondary objects (cell membranes) at a defined distance. Finally, tertiary objects (cytoplasm) were defined by subtracting primary from secondary objects. The resulting object masks were then applied to extract cellular parameters including area, shape, distribution and variation of pixel intensities, and mean intensity from the corresponding greyscale images using the CellProfiler MeasureObjectSizeShape, MeasureObjectRadialDistribution, MeasureTexture, and MeasureObjectIntensity modules, and stored in a database file. This database file was then imported into the Classifier module of the CellProfiler Analyst software for supervised training with a subset of cells, and subsequent parameter-based automatic classification of all cells within the entire image dataset. For immunofluorescence images, supervised training based on cell membrane features (E-Cad channel) was conducted with two classes representing continuous adherens junctions (Continuous AJs') and discontinuous adherens junctions (Discontinuous AJs) (about 30 cells for each class). After the supervised training process had been completed, the entire image dataset from a specific experiment was evaluated and all identified cells were automatically classified into corresponding classes according to their individual parameters. The fraction of cells representing Continuous AJs and Discontinuous AJs classes were finally graphically represented in stacked bar plots.

[0139] In addition to the CellProfiler Analyst-based classification of cells, the mean intensity parameter that is alongside extracted from images using CellProfiler can also be used to distinguish between continuous AJs and discontinuous AJs. The quantified mean membrane signal intensity of cells was finally graphically represented.

[0140] All graphical representations and statistical analyses of data were performed using the GraphPad Prism software (GraphPad Software, Inc.).

[0141] The result of this experiment is shown in FIG. 7. In accordance with the visual inspection of immunofluorescence images, upon treatment with different concentrations of fulvestrant, the fraction of cells with continuous adherens junctions decreased while the fraction of cells with discontinuous adherens junctions increased. In turn, fulvestrant-treatment in combination with different concentrations of 17--estradiol (E2) restores the fraction of cells with continuous adherens junctions. Furthermore, the appearance of discontinuous adherens junctions also correlated with an increase in the mean signal intensity of cell membranes.

[0142] Thus, morphological changes at adherens junctions can easily be monitored and evaluated even in an automated manner. This renders the methods of the present invention particularly suitable for high-throughput screening and automated image analysis.

Example 8: Fulvestrant Induced Changes to Adherens Junction Morphology can Also be Assessed by Imaging of Adherens Junctions Components Other than E-Cadherin

[0143] In a further set of experiments, the inventors assessed the impact of fulvestrant treatment on other components of the adherens junctions, namely -Catenin, -Catenin, p120-Catenin, and the cortical actin cytoskeleton. Briefly, the MCF-7/vBOS cells were seeded and grown on glass coverslips in multi-well tissue culture plates (TPP) for 24 h to reach near 100% confluence and then treated for at least 48 h with 0.1% EtOH (solvent control) and the ER antagonist fulvestrant (Sigma-Aldrich, CAS-No. 129453-61-8) at a concentration of 10.sup.8 M. During experiments, medium was exchanged on a daily basis. All experiments were performed at 37 C. with 5% CO.sub.2 in reduced-serum medium (phenol red-free DMEM (Gibco), 5% FBS (Biochrom, S 0615, LOT1353 W), 1% glutamine, 100 g/ml streptomycin/100 U/ml penicillin (Biochrom)). As a result, the change in adherens junctions morphology was also observable when staining any of -Catenin, -Catenin, p120-Catenin, or the cortical actin cytoskeleton.

Example 9: Anti-Estrogens Elevate Adherens Junction Stability

[0144] In a further set of experiments, the inventors tested whether the anti-estrogen-mediated reorganization of adherens junctions likewise increased adherens junction stability in MCF-7/vBOS cells. Therefore, EGTA-containing medium (810.sup.3 M) was applied to both control cells and cells pre-treated for 48 hours with fulvestrant to disrupt homophilic E-Cadherin binding, and the fraction of rounded cells was monitored over the course of 120 minutes by live imaging. Quantification of rounded cells vs. non-rounded cells was done in analogous manner as described above in example 7 for quantification of continuous and discontinuous adherens junctions. Notably, cells pre-treated with fulvestrant showed a significantly reduced rate of cell rounding (FIG. 9) indicating that reorganized adherens junctions were stabilized and more resilient against disruption.

Example 10: Anti-Estrogens Promote Cell Rigidity and Reduce Cell Motility

[0145] To test whether anti-estrogens also impact on the biomechanical properties of MCF-7/vBOS breast cancer cells, the inventors performed atomic force microscopy (AFM) indentation measurements on confluent monolayers. For this purpose, MCF7/vBos cells were grown and treated in 35 mm FluoroDish cell culture dishes (WPI) as described above. AFM indentation measurements were performed in CO.sub.2 Independent Medium (Gibco) at 37 C. using the NanoWizard 1 or 4 (JPK Instruments). Prior to measurements, tip-less Arrow-TLlsilicone cantilevers (Nanoworld) were equipped with polystyrene beads of 5 m in diameter (microParticles GmbH, Germany) using epoxy glue, and calibrated using built-in procedures of the SPM software (JPK Instruments). The cantilever was positioned above the confluent epithelial monolayer and lowered with a speed of 10 m/s. In each experiment, force-distance curves (force setpoint 2.5 nN) from 3-4 different positions per cell were collected. Force-distance curves were transformed into force-versus-tip sample separation curves according to Radmacher (Methods Cell Biol 2002; 68:67-90), and fitted with the Hertz/Sneddon model (27) for a spherical indenter using the JPK Data Processing software (JPK Instruments). A Poisson ratio of 0.5 was assumed for the calculation of the apparent elastic (Young's) modulus. To map the elastic modulus distribution, cells were probed with a spatial resolution of 1 m using a MLCT cantilever (Bruker). Apparent elastic (Young's) moduli were determined using the Hertz model for a quadratic pyramid using the JPK Data Processing software (JPK Instruments).

[0146] Intriguingly, cell rigidity was significantly elevated throughout the cell surface and at junctional regions in fulvestrant-treated cells (FIG. 10A). Given that anti-estrogens increase adherens junction stability and cell rigidity, the inventors asked whether cell motility was likewise affected. Hence, the inventors recorded the motility of MCF-7/vBOS breast cancer cells over the course of 48 hours in a confluent monolayer after introducing a scratch nearby the imaged region. Notably, both the velocity of cells (FIG. 10B) and the length of cell trajectories (data not shown) were significantly reduced upon fulvestrant treatment. Together, the increased cell rigidity and reduced cell motility of MCF-7/vBOS breast cancer cells suggest that anti-estrogens can promote tissue stability and indicate a previously unknown mechanism mediating the metastasis-protective effect of anti-estrogens in endocrine therapy.

Example 11: Er Signalling Activity and Adherens Junction Organization Correlate in Human Breast Cancer Sections

[0147] The inventors tested whether the ER signalling-dependent organization of adherens junctions can also be observed in breast cancer patients. Due to the poor availability of specimens from patients undergoing anti-estrogen-based endocrine therapy, the inventors utilized tissue microarrays (TMA) containing breast tissue sections from patients with diagnosed invasive ductal carcinoma (IDC). Using nuclear or cytoplasmic ER localization as a surrogate for high or low ER signalling activity, respectively, the inventors assessed the state of ER signalling activity and the organization of adherens junctions by immunofluorescence microscopy.

[0148] Briefly, tissue microarrays containing 54 formalin-fixed paraffin-embedded breast cancer tissue sections (1.5 mm diameter) from 28 human patients with diagnosed invasive ductal carcinoma (IDC) (provided by Carsten Denkert; Institute for Pathology, Charit-Universittsmedizin Berlin, Germany) were prepared for immunofluorescence microscopy according to Junghans et al. (Dev Dyn 2005; 233:528-39). Tissue sections were subjected to immunofluorescence staining, imaging, and image analysis as described for the cell culture experiments.

[0149] As expected for IDC cells, all tumour samples expressed E-Cadherin as well as ER, however, the organization of adherens junctions and the localization of ER varied between patients. Notably, discontinuous adherens junction morphologies, that were reminiscent to the ones observed in MCF-7/vBOS breast cancer cells, were detectable in cells displaying low ER signalling activity. Hence, the inventors applied the CellProfiler/CellProfiler-Analyst-based image analysis pipeline on all patient samples and quantified the fraction of IDC cells displaying continuous or discontinuous adherens junctions (FIG. 11, black bars), and nuclear or cytoplasmic ER localization (FIG. 11, white bars). Intriguingly, the inventors found that the appearance of discontinuous adherens junctions correlated with primarily cytoplasmic ER localization/low ER signalling activity, indicating that the organization of adherens junctions in vivo also seems to be dependent on ER signalling activity.