WO2000037681A1 - BIOASSAY FOR IDENTIFYING ESTROGEN RECEPTOR-β/α SELECTIVE MODULATORS - Google Patents

Abstract

The present invention provides novel assay methods for identifying compounds that selectively activate estrogen receptors (ER) of the alpha or beta subtype. In particular, the results from two assays, one measuring ER-β activity and the other measuring ER-α activity are interpreted. The assay measuring ER-β activity uses cells comprising endogenous metallothionein-II as well as a DNA plasmid comprising a polynucleotide encoding human ER-β. The assay monitors expression of metallothionein-II-mRNA in said cells, wherein the level of metallothionein-II expression is regulated when a potential ligand binds to ER-β. The assay measuring ER-α activity uses cells comprising ER-α as well as DNA plasmid comprising a reporter gene linked to an estrogen response element. The assay monitors expression of the reporter gene, wherein the level of reporter gene is regulated when a potential ligand binds to ER-α. Compounds which modulate activity in one assay but have little or no activity in the other assay are defined as estrogen receptor subtype selective.

Description

BIOASSAY FOR IDENTIFYING ESTROGEN RECEPTOR-β/α SELECTIVE

MODULATORS

Technical Field The present invention relates to hormone receptors, and, more particularly to methods for identifying compounds that selectively activate estrogen receptors (ER) of the alpha or beta subtype, as well as a test kit for use in the methods.

Background of the Invention

Proteins which regulate gene expression are essential for cell function. A well-studied family of gene regulatory proteins is the steroid hormone receptor superfamily. These receptor proteins enable cells to respond to various hormones by activating or repressing specific genes. One member of this family is the ER. Estrogens are classically known as important hormones in sexual development and reproductive function. It is well known that estrogens affect cell proliferation and differentiation in target tissues by binding to ERs in target cells. Estrogen replacement therapy is a well established treatment for prevention and/or amelioration of osteoporosis in postmenopausal women (Sagraves, 1995,; Lobo, 1995) because these compounds have been demonstrated to prevent bone loss and fractures in women. Additionally, estrogen replacement therapy has been associated with a decreased mortality from cardiovascular disease. Finally, ongoing studies suggest that estrogens may provide benefit to the central nervous system with regard to cognitive improvement and a decrease in Alzheimer's disease.

The classic ER, now designated ER-α, exhibits a modular structure

with distinct domains involved in ligand binding, receptor dimerization, DNA binding, and transcriptional activation (Green et al, 1 986). The nucleotide sequence of the DNA binding domain is conserved among steroid hormone receptors. Kuiper et al ( 1 996) exploited this similarity in an attempt to identify additional members of this family. The new member they discovered was recognized to be an ER because the amino acids in the DNA binding domain were almost identical to those of ER-α, ,

the in vitro translated protein specifically bound [³H]-estradiol with nanomolar affinity and was able to regulate transcription from a simple estrogen response element. This protein has been designated ER-β to

distinguish it from the previously known form (now called ER-α). Human

and mouse ER-β have also been cloned Bhat et.al. 1998; Mosselman,

1996; Pettersson et al, 1 997; Tremblay et al, 1 997). Following ER-β's

discovery, most work has focused on mapping the distribution of its mRNA in normal and neoplastic tissues, characterizing its binding affinity for a wide variety of ligands, and assessing its interaction with ER-α.

ER-β mRNA is detectable by RT-PCR and in situ hybridization in a

wide variety of tissues. While its distribution overlaps that of ER-α, some

tissues express only one receptor type. For example both receptors are found in the uterus, ovary and pituitary, whereas ER-β appears more prominent in the prostate, lung and bladder (Kuiper et al, 1 996). When examined by in situ hybridization, further distinctions between receptor distribution can be made. ER-β mRNA is expressed in the epithelial cells of the rat prostate; ER-α is found in the stromal compartment (Kuiper et al, 1996). In the rat ovary, ER-β appears in the granulosa cells; ER-α in the stroma (Shughrue et al, 1996; personal communication). In the rat hypothalamus, ER-β but not ER-α, message is expressed in the paraventricular region, whereas both messages are seen in the preoptic area (Shughrue et al, 1996). Interestingly, significant species differences may exist in the relative levels of ER-β and ER-α mRNA in certain organs. For example, ER-β mRNA is highly expressed in the rat prostate, whereas more modest levels are detected in the human prostate. ER-β predominates over ER-α in the rat prostate (Kuiper et al, 1 996; Lau et al, 1998; Enmark et al, 1997), but the levels are more equal in the mouse prostate (Couse et al, 1997).

Several groups have characterized ER expression in tumor samples and cancer cell lines. Most work has been done in the breast where one study reports ER-β mRNA using RT-PCR in 70% of forty breast biopsy samples (Dotzlaw et al, 1996). No correlation was noted between ER-α and ER-β expression. Another study also reports heterogeneity of ER-α/β mRNA expression in breast tumor samples, with some tumors having only ER-α mRNA and others expressing mRNA for both receptor subtypes (Enmark et al, 1997). Among breast cancer cell lines, MDA MB 231 has only ER-β mRNA, and conflicting reports exist about the ER status of

MCF-7 cells (Enmark, 1 997; Kuiper et al, 1 997; Vladusic et al, 1 998; R. Henderson, unpublished observations).

Although the ligand binding domain of human ER-α and ER-β are

59% identical at the amino acid level (Enmark et al, 1 997), the binding affinity of 1 7β-estradiol is quite similar. Some compounds show marked

selectivity for either ER-α or ER-β. Genistein is a phytoestrogen which

binds with approximately 10-25 fold higher affinity for ER-β (Kuiper 1 996,

H. Harris unpublished observations) . On the other hand raloxifene binds about 20 fold better to ER-α than ER-β (H. Harris, unpublished

observations).

When ER-α and ER-β are coexpressed , interaction can be

measured by several methods including a mammalian two-hybrid system, glutathione S transferase pulldown and gel shift/supershift assays (Cowley et al, 1 997; Pettersson et al, 1 997; Ogawa et al, 1 998). Since these receptors can heterodimerize, estrogens' effect on tissues containing both receptors may be mediated by a complex interaction between ER-α and ER-β.

One valuable tool in determining the function of ER-β is the

estrogen receptor-α knockout (ERKO) mouse (Lubahn et al, 1 993).

Because these mice lack functional ER-α, they can help define the

physiological roles of both ER-α and ER-β. One study has compared the

effectiveness of 1 7-β estradiol treatment in ameliorating consequences of artificially induced vascular injury to carotid arteries in wild type and ERKO mice (lafrati et al, 1997). In both types of mice, pharmacological doses of 1 7-β estradiol suppressed the increase in medial area and

smooth muscle cell proliferation seen in the vehicle treated d animals. It is thought that these responses to endothelial denudation may narrow the lumen of the vessel, thus restricting blood flow. Because 17-β estradiol

was equally effective in ERKO as wild type mice, one interpretation is that ER-α is not necessary for this response. Because ER-β mRNA is also expressed in these vessels, perhaps it mediates 17-β estradiol's action. However, direct evidence supporting this hypothesis is lacking. Another example of ERKO mice responding to 17-β estradiol replacement was

described in a recent poster and abstract (Pan et al, 1997, and personal communication). This study reports a loss of femoral bone mineral density and trabecular bone volume in ERKO mice after ovariectomy and an increase in these parameters upon treatment with 17-β estradiol (but not dihydrotestosterone). Again, based on indirect evidence, a suggestion is made that ER-β has a physiological role.

Thus, while certain aspects of ER-β have been characterized, the art fails to provide methods for identifying ligands which are functionally selective for ER-α or ER-β. Such an assay would be greatly advantageous to the pharmacological industry to uncover the possible therapeutic applications of the ER subtypes. Summary of the Invention

The present invention provides method for screening a test compound that binds to an ER in a receptor binding assay, wherein said method detects ER-β polypeptide-mediated transcription, said method

comprising the steps of:

(a) providing a cell which comprises at least one estrogen-induced DNA sequence encoding metallothionein (MT-II) and at least one DNA sequence encoding ER-β polypeptide, wherein said receptor is

transcriptionally active; (b) contacting said cell with either said test compound which binds

ER or a control; and

(c) detecting the expression of said MT-II, wherein enhanced expression of said MT-II relative to a control indicates that said test compound has estrogen agonist activity. This invention further provides a method for screening a test compound that binds to the ER in a receptor binding assay, wherein said method detects inhibition of ER-β )- polypeptide-mediated transcription,

said method comprising the steps of:

(a) providing a cell which comprises at least one estrogen-induced DNA sequence encoding metallothionein (MT-II) and at least one DNA sequence encoding ER-β polypeptide, wherein said receptor is

transcriptionally active; (b) contacting said cell with one or more estrogen(s) in the presence of the test compound known to bind ER; and

(c) and detecting the expression of said MT-II, wherein decreased expression of said MT-II relative to the addition of one or more estrogen(s) alone indicates that said test compound has estrogen antagonist activity A method of screening test compounds to identify drug candidates which mimic estrogens' effect on ER-β- or ER-α-mediated transcription is

further provided wherein said method comprises the steps of:

(a) contacting said test compound with a plurality of: (i) first cells comprising at least one DNA sequence encoding metallothionein (MT-II) and at least one DNA sequence encoding an ER-β polypeptide, wherein said

receptor is transcriptionally active and (ii) second cells comprising an ERE reporter gene construct, wherein said cells express ER-α polypeptide;

(b) identifying compounds which increase expression of MT-II in said first cells relative to control but have minimal effect on expression of said reporter gene in said second cells, wherein said compounds are considered ER-β selective; or

(c) identifying compounds which increase expression of the reporter gene in said second cells relative to control but have minimal effect on expression of MT-II in said first cells, wherein said compounds are considered ER-α selective. Lastly, the present invention provides a method of screening test compounds to identify drug candidates which inhibit estrogen's effect on ER-β- or ER-α-mediated transcription, said method comprising the steps

of: (a) contacting said test compound in the presence and absence of one or more estrogen(s) with a plurality of:

(i) first cells comprising at least one endogenous DNA encoding a metallothionein (MT-II) gene and DNA encoding a ER-β polypeptide, wherein said receptor is transcriptionally active and

(ii) second cells comprising an ERE reporter gene construct, wherein said cells express ER-α polypeptide; and

(b) identifying compounds which decrease expression of MT-II in said first cells relative to treatment with one or more estrogen(s) alone but have minimal effect on expression of said reporter gene in said second cells, wherein said compounds are considered ER-β selective; or

(c) identifying compounds which decrease expression of the reporter gene in said second cells relative to treatment with one or more estrogen(s) alone but have minimal effect on expression of

MT-II in said first cells, wherein said compounds are considered ER-α selective. Brief Description of the Figures

Figure 1 : RT-PCR amplification of ER-β and ER-α α from Saos -2 and

LNCaPLN3 cells.

Figure 2: Sequencing gel separating amplified cDNAs and 1 7-β estradiol

upregulation of fragment 6a.

Figure 3: (A) Nucleotide sequence of the regulated fragment and its alignment with human MT-II . (B) Translation of fragment sequence from first methionine to stop codon ( *)

Figure 4: Regulation of MT-II in two cell lines. To determine fold change in mRNA, MT-II signal was normalized to that of GAPDH, and compared to the control cells.

Figure 5: Dose response of 1 7-β estradiol regulation of MT-II in Saos-2

cells. Result are shown for four different experiments and the EC₅₀ shown for each.

Figure 6: Time course of metallothionein-II regulation in Saos-2 cells. Data is normalized with GAPDH and fold change calculated from the vehicle control at time = 0hr.

Figure 7: Receptor specificity of MT-II regulation in Saos-2 cells.

Figure 8: Regulation of MT-II in Saos-2 cells by various compounds.

Figure 9: Metallothionein-ll regulation in cycloheximide-treated Saos-2 cells. (A) Measurement of protein synthesis during treatment with cycloheximide. (B) Whole cell ER binding assay after 8 hours of cycloheximide treatment. (C,D) Metallothionein-ll regulation after 8 and 24 hours of treatment respectively.

Figure 10: MT is regulated by estrogens in the rat prostate.

Figure 1 1 : Induction of MT-II in the rat prostate requires at least two days of dosing.

Figure 1 2: Screening strategy for ligands which selectively activate ER-β and/or ER-α.

Detailed Description of the Invention The present invention provides an efficient way to screen large numbers of test compounds which selectively activate ERs of the α or β

subtype. These compounds may have desirable properties for either the treatment or the prevention of various diseases mediated by estrogens, including but not limited to cancers (e.g. breast, ovarian, endometrial, prostate), endometriosis, osteoporosis and cardiovascular and central nervous system diseases.

Definitions In the present description and claims, the following terms shall be defined as indicated below.

"hERβL" is defined as the human form of ER-β described in

Example 1 (the cDNA or its translated protein product).

"Estrogen" is defined as any ligand that can function as an estrogen agonist.

"Estrogen agonist" is defined as a compound that substantially mimics 17-β estradiol as measured in a standard assay for estrogenic

activity, for example, cellular assays as described in Webb et al. ( 1 992). "Estrogen antagonist" is defined as a compound that substantially inhibits the effect of estrogen agonists as measured in a standard assay for estrogenic activity, for example, cellular as described in Webb et al. ( 1 992) "A functional ER" is defined as a receptor capable of transcriptional activation of endogenous or transfected genes as measured by changes in RNA, protein and/or downstream biological events. "A nonfunctional ER" is defined as a receptor incapable of transcriptional activation of endogenous or transfected genes as measured by changes in RNA, protein and/or downstream biological events. A "test compound" includes but is not limited to any small molecule compound, peptide, polypeptide, natural product, toxin with potential biological activity. A "ligand" is intended to include any substance that interacts with a receptor. .

"Transfection" is defined as any method by which a foreign gene is inserted into a cultured cell.

A "reporter" is defined as any substance that can be readily measured and distinguished from other cellular components. The reporter may be the transfected receptor DNA, the transcribed receptor mRNA, an enzyme, a binding protein or an antigen.

A "cell" useful for the present purpose is one which has the ability to respond to signal transduction through a given receptor. "A "receptor binding assay" is an assay measuring the amount of ligand specifically interacting with a receptor. The ligand can be a radioligand (e.g. conjugated to ³H or ¹²⁵l), a fluorescinated ligand (either conjugated to a fluorochrome or possessing inherent fluorescence) or otherwise labeled so as to be detectable. Description of the Assay

The present invention relies on the discovery that the expression of MT-II is selectively regulated by the interaction of a ligand with ER-β. .

In the methods of the invention, the regulation of MT-II activity may be used to provide a screening system that selectively detects both estrogen agonist or antagonist functional activity of a ligand following its interaction with ER-β.

The methods typically comprise cultured cells that express functional human ER-β and no, or a diminished amount of ER-α.

Such cells include but are not limited to Saos-2 (ATCC HTB-85) and LNCaPLN3. Preferred cells for this purpose include cells which over- express ER-β, such as the cells described below which are recombinantly

manipulated to over-express ER-β. The ER-β receptor may be modified in

any way, such as in length; these modifications may result in increased ER-β selectivity and increased sensitivity of the assay.

One of skill will recognize that various recombinant constructs comprising ER-β can be used in combination with any cell or line which

lacks functional ER-α.

To screen a number of compounds for estrogen agonist activity, cells expressing ER-β are exposed to a test compound or a control

solution (which is used to dissolve the test compound). The cells can be exposed while either growing in separate wells of a multi-well culture dish or in a semi-solid nutrient matrix. After treatment for a suitable period of time, MT-II mRNA is measured. An estrogen agonist will increase MT-II mRNA when compared to treatment with the control solution alone. To screen a number of compounds for estrogen antagonist activity, cells expressing ER-β are exposed to one or more estrogens in the

presence or absence of a test compound. The cells can be exposed while either growing in separate wells of a multi-well culture dish or in a semi- solid nutrient matrix. After treatment for a suitable period of time, MT-II mRNA is measured. An estrogen antagonist will decrease MT-II mRNA when compared to treatment with the estrogen solution alone.

Estrogenic or antiestrogenic compounds identified in the assays of the invention can be used in standard pharmaceutical compositions for the treatment of cancer, as components of oral contraceptives, or any other application in which the modulation of estrogen activity is desired. The pharmaceutical compositions can be prepared and administered using methods well known in the art. The pharmaceutical compositions are generally intended for parenteral, topical, oral or local administration for prophylactic and/or therapeutic treatment. The pharmaceutical compositions can be administered in a variety of unit dosage forms depending upon the method of administration. For example, unit dosage forms suitable for oral administration include powder, tablets, pills, and capsules. Suitable pharmaceutical formulations for use in the present invention arefound in Remington's Pharmaceutical Sciences, Mack Publishing Company, Philadelphia, Pa., 1 7th ed. ( 1 985). A variety of pharmaceutical compositions comprising compounds of the present invention and pharmaceutically effective carriers can be prepared. Applications

The methods and compositions of the present invention can be used to identify compounds that interact with ER-β , either to stimulate or to

inhibit transcriptional activity . Such compounds include, without limitation, co-activator proteins, as well as estrogens and other steroids, steroid-like molecules, or non-steroid-like molecules that act as agonists or antagonists. Screening methods can also be used to identify tissue-specific estrogens. Identification of ER-β -interactive compounds can be achieved by

cell-free or cell-based assays. In one set of embodiments, purified ER-β is

contacted with a labeled ligand, such as, e.g., 1 7-β estradiol, in the

presence of test compounds to form test reactions, and in the absence of test compounds to form control reactions. The labeled moiety may comprise a radiolabel (such as, e.g., ³H or ¹²⁵l) or a fluorescent molecule. Incubation is allowed to proceed for a sufficient time and under appropriate conditions to achieve specific binding, after which binding of labeled ligand to ER-β is measured (by monitoring, e.g., radioactivity, fluorescence, or

fluorescence polarization). In one embodiment, the ligand binding domain of ER-β produced in E. coli is adsorbed to the wells of a microtiter dish and

incubated with [³H]-17β estradiol in the absence or presence of test compounds. Alternatively, soluble receptor is incubated with the labeled ligand in the absence or presence of test compounds, and bound ligand is separated from free ligand, either by filtration on glass fiber filters or by using dextran-coated charcoal Whole cell binding assays may also be used in which bound ligand is separated from free ligand by rinsing. Cells used in these assays may either contain endogenous receptor, or may overexpress the receptor subsequent to stable or transient transfection or infection of an ER-β cDNA. Non-

limiting examples of suitable cells include COS cells, HeLa cells, CHO cells, human umbilical vein endothelial cellsand yeast. Once a compound has been identified as an ER-β -interactive compound by its binding activity,

further in vivo and in vitro tests may be performed to determine the nature and extent of activity, i.e., as an agonist or antagonist (see below).

ER-β -interactive compounds may also be identified using cell-based

assays that measure transcriptional activation or suppression of endogenous or transfected estrogen-responsive genes. For example, agonists (such as, e.g., 17β -estradiol) block interleukin-1 β induction of

endogenous E-selectin in primary human umbilical vein endothelial cells that express ER-β . Antagonists (such as, e.g., ICI-1 82780) block the agonist

activity of 17β -estradiol. Non-limiting examples of other suitable

endogenous estrogen-responsive promoter elements include those that regulate endothelin-1 (ET-1 ); HDL receptor (scavenger receptor type II); and enzymes involved in coagulation and fibrinolysis (such as, e.g., plasminogen activator inhibitor- 1 and complement C3). Any promoter element that responds to estrogen may be used as an appropriate target, including, e.g., the NFkB binding site or the apolipoprotein A1 gene enhancer sequence. In one set of embodiments, appropriate host cells are transfected with an expression vector encoding ER-β and the transfectants are incubated with or without one or more estrogens I in the presence or absence of test compounds. ER-β activity is assessed by measuring transcriptional activation. This may be achieved by detection of mRNA (using, e.g., Northern blot analysis, reverse transcriptase polymerase chain reaction, RNase protection assays) and/or by detection of the protein

(using, e.g., immunoassays or functional assays). If activation of the target sequence initiates a biochemical cascade, downstream biological events may also be measured to quantify ER-β activity. ER-β -interactive compounds are identified as those that positively or negatively influence target sequence activation.

In another set of embodiments, appropriate host cells (preferably, mammalian ) are co-transfected with an expression vector encoding ER-β and a reporter plasmid containing a reporter gene downstream of one or more ERE s. Transfected cells are incubated with or without an estrogen in the presence or absence of test compounds, after which ER-β activity is determined by measuring expression of the reporter gene. In a preferred embodiment, ER-β activity is monitored visually. Non-limiting examples of suitable reporter genes include luciferase, chloramphenicol acetyl transferase (CAT), and green fluorescence protein.

Preferably, the methods of the present invention are adapted to a high-throughput screen, allowing a multiplicity of compounds to be tested in a single assay. Candidate estrogens and estrogen-like compounds include without limitation diethylstilbesterol, genistein, and estrone. Other ER-β -

interactive compounds may be found in, for example, natural product libraries, fermentation libraries (encompassing plants and microorganisms), combinatorial libraries, compound files, and synthetic compound libraries. For example, synthetic compound libraries are commercially available from Maybridge Chemical Co. (Trevillet, Cornwall, UK), Comgenex (Princeton, NJ), Brandon Associates (Merrimack, NH), and Microsource (New Milford, CT). A rare chemical library is available from Aldrich Chemical Company, Inc. (Milwaukee, Wl). Alternatively, libraries of natural compounds in the form of bacterial, fungal, plant and animal extracts are available from, for example, Pan Laboratories (Bothell, WA) or MycoSearch (NC), or are readily produced . Additionally, natural and synthetically produced libraries and compounds are readily modified through conventional chemical, physical, and biochemical means (Blondelle et al., 1 996). ER-β binding assays

according to the present invention are advantageous in accommodating many different types of solvents and thus allowing the testing of compounds from many sources. Compounds identified as ER-β agonists or antagonists using the

methods of the present invention may be modified to enhance potency, efficacy, uptake, stability, and suitability for use in therapeutic applications, etc. These modifications are achieved and tested using methods well- known in the art.

Examples

The present invention is further described by the following examples. The examples are provided solely to illustrate the invention by reference to specific embodiments. These exemplifications, while illustrating certain specific aspects of the invention do not portray the limitations or circumscribe the scope of the invention.

EXAMPLE 1 : CONSTRUCTION OF ER-β RECOMBINANT ADENOVIRUS

The coding sequence of human ER-β was obtained as described in co-owned and co-assigned pending US Serial 08/906,365 entitled "Novel Human Estrogen Receptor β" filed August 5, 1997, and in Bhat et. al.,

(1998). Essentially, human testis Poly A + RNA (1 μg, Ciontech, Palo Alto CA) was mixed with 0.5 μg oligo dT primer (GIBCO-BRL, Gaithersburg MD) in a total volume of 10 μl. The mixture was heated at 70°C for 10 minutes, and, after cooling on ice, was supplemented with 500 μM of each deoxynucleoside triphosphate, 1 X cDNA synthesis buffer, and 10 mM DTT to a final reaction volume of 20 μl. The mixture was incubated at 42°C for

2.5 minutes and then supplemented with 1 -2 units reverse transcriptase (GIBCO-BRL, Gaithersburg MD), after which it was incubated at 45 °C for 30 minutes and 50°C for 5 minutes. One-tenth of this mixture (approximately 2 μl) containing the cDNA template was then used in PCR

amplification of ER-β using forward and reverse primers as described below.

The PCR primers designated in Serial No. 08/906,365 (supra) were used to amplify ER-β sequences in a reaction containing the following

components: 2 μl of the cDNA template described above; 1 X PCR buffer;

200 μM of each deoxynucleoside triphosphate, 2 units of Hot Tub DNA

polymerase (Amersham, Arlington Heights IL), and 1 μg of each of the

forward and reverse primers. The reaction mixture was heated to 95 °C for 2 minutes, annealed at 52 °C for 1 minute, and amplified using 36 cycles, followed by an incubation at 72°C for 1 .5 minutes.

A fragment of approximately 1 500 bp in length was produced. The fragment digested with Hindlll and Xbal (which cleave at sites present in the forward and reverse primer sequences, respectively, but not in the main body of the amplified cDNA sequence) and cloned into the corresponding sites of the pcDNA3 expression vector (Invitrogen, Carlsbad CA). This asymmetric cloning strategy places the 5' end of ER-β cDNA under the

control of the viral CMV promoter in pcDNA3. This clone was designated "truncated hERβ" or hERβT.

To verify the amino terminal and upstream sequence of human ERβ,

two independent approaches were taken, as described below. ( 1 ) 10 μl of a human ovary δ'-Stretch cDNA library

(Clontech, Palo Alto CA) was mixed with 50 μl of 1 X K solution (1 X PCR

Buffer (GIBCO-BRL, Gaithersburg MD), 2.5 mM MgCI₂, 0.5% Tween-20, 100 μg/ml Proteinase K), and the reaction mixture was incubated at 56°C

for 2 hours, then at 99°C for 10 minutes. Five μL of this reaction mixture

were then used as template in a nested PCR reaction using the PCR primers designated in Serial No. 08/906,365 (supra). The reaction contained 1 X Klentaq PCR reaction buffer (40 mM Tricine-KOH, 1 5 mM KOAc, 3.5 mM Mg(OAc)2, 75 μg/ml bovine serum albumin); 0.2 μM of each dNTP; 0.2 μM

of each of the above primers, and 1 unit of Klentaq Polymerase Mix

(Clontech, Palo Alto CA). Touchdown PCR conditions were as follows: 5 cycles of 94°C for 2 seconds and 72°C for 4 minutes, followed by 30 cycles of 94°C for 2 seconds and 67 °C for 3 minutes.

Excess nucleotides and primers were removed from first round PCR reactions by purification using Wizard PCR columns (Promega, Madison Wl). A second-round PCR reaction was then performed using 2 μl of the

purified first-round reaction mixture using the PCR primers designated in Serial No. 08/906,365 (supra). The PCR reaction and cycling conditions were identical to those employed in the first round. The products were cloned into pCR2.1 (Invitrogen, Carlsbad CA ) and three resulting clones were sequenced. All three clones (designated L1 , L2, and L3) contained ERβ inserts of different lengths, all of which were homologous to ERβ and

to each other. (2) A Marathon Ready thymus cDNA kit (Clontech, Palo Alto CA) for 5' rapid amplification of cDNA ends (RACE) was also used to isolated ERβ 5' clones. In the first round of a nested PCR reaction, 5 μl of

human thymus Marathon-ready cDNA (Clontech, Palo Alto CA) was used as template and using the primers designated in Serial No. 08/906,365 (supra). The PCR reaction and cycling conditions were identical to those described in (1 ) above.

Excess nucleotides and primers were removed from the first round PCR reactions by purification over Wizard PCR columns (Promega, Palo Alto CA). A second round PCR reaction was performed using 2 μl of the purified

first round reaction using primers designated in Serial No. 08/906,365 (supra). The second round PCR reaction and cycling conditions were identical to those employed in the first round. The products were cloned into the pCR2.1 vector (Invitrogen, Carlsbad CA) and two clones were sequenced. The two clones contain insert sequences of different lengths that are homologous to ERβ, to each other, and to the sequences isolated

from a human ovary cDNA library as described above.

All of the ERβ sequences isolated by methods (1 ) and (2) above

contained 1 10 nucleotides corresponding to hERβT sequences, as well as

228 additional nucleotides at the 5' end.

The hERβT and 5' end sequences were joined and the resulting cDNA was

cloned into pCDNA3 (Invitrogen, Carlsbad CA ) under the control of the cytomegalovirus IE promoter; this expression vector was designated "long hER-β" or hERβL . Full length hERβL cDNA sequence encodes a polypeptide

having 530 amino acid residues as seen in SEQ ID No.1 . The hERβL sequence contained an optimal translation initiation sequence

CCACC immediately upstream to the initiation codon and the sequence was under the control of the cytomegalovirus IE promoter. The coding sequence of hERβL was then transferred into an Ad5ΔE1 a vector plasmid

containing adenovirus sequences from map unit 0-1 7 with an deletion of E1 a region between map units 1 .4 -9.1 (Davis AR et. al., 1 985, GluzmanN. et. al., 1 982). The hERβL transcription unit in Ad5ΔE1 a

plasmid contained cytomegalovirus 1 E promoter, Ad5 tripartite leader, coding sequence of hERβ_L and SV40 late polyadenylation signal

sequences. hERβL in Ad5 ΔE1 a plasmid was then linearized with BstEII

enzyme and transfected along with Clal A fragment of Ad5 virus with E3 region deletion (80-88 map units) into 293 cells (transformed primary human embryonic kidney, ATCC CRL 1 573). Viral plaques generated by homologous recombination were isolated, amplified and characterized by restriction DNA analysis and cell lysis assay in A549 cells (human lung carcinoma, ATCC CCL 185). Confirmatory tests indicated that the recombinant Ad5 hERβL virus contained the expected DNA fragments and

was replication defective. The virus was further purified by re-plaquing. The isolated plaques were amplified, tested and used as a seed stock to generate large amounts of the virus in 293 cells. The virus was titered in 293 cells by plaque assay and the stock contained 1 .28 X10 ⁹ PFU/ml. EXAMPLE 2: ASSESSMENT OF ENDOGENOUS LEVELS OF ER MRNA IN SAOS-2 AND LNCAPLN3 CELLS

Unless otherwise noted, cell culture reagents were obtained from Gibco BRL (Gaithersburg MD). LNCaPLN3 cells were grown in RPMI 1 640 medium supplemented with 10% FBS, 2 mM GlutaMAX-1 , 100 U/ml penicillin g, and 100 //g/ml streptomycin sulfate. Saos-2 cells (ATCC, Manassas VA) were maintained in monolayer culture using McCoy's 5A medium supplemented with 1 0% fetal bovine serum (FBS), 2 mM GlutaMAX-1 , 100 U/ml penicillin g, 100 g/ml streptomycin sulfate. RT-PCR

Total RNA was isolated from LNCaPLN3 and Saos-2 cells using TRIzol (Gibco BRL, Gaithersburg MD) according to the manufacturer's directions. The samples were then treated with RNase-free DNase I (Gibco BRL, Gaithersburg MD) at 1 unit//vg for 30 minutes at 37°C. RNA

was purified from the reaction using RNeasy columns (Qiagen, Hilden Germany) and amount recovered estimated by UV spectrophotometry.

Reverse transcription reactions were performed on 0.5 μg of RNA in a 20 μ\ reaction. For ER-α the reaction contained 1 x PCR Buffer (Gibco

BRL, Gaithersburg MD), 5 mM MgCl2, 1 .25 μM ERα-specific reverse

primer (5'-CCAGCAGCATGTCGAAGATC-3'), 0.5 μM GAPDH-specific reverse primer (5'-CACCCTGTTGCTGTAGCCAAATTC-3'), 0.5 mM dNTPs, 20 units of RNasin (Promega, Madison Wl) and 200 units of Superscript II Reverse Transcriptase (Gibco BRL, Gaithersburg MD). The ER-β reaction contained the same components as ER-α with the following

exceptions: 2.5 mM MgCl2 and 1 .25 μM ERβ-specific reverse primer (5'-

GCAGAAGTGAGCATCCCTCTTTG-3'). A duplicate reaction which was identical in all reagents except that it did not contain Superscript II Reverse Transcriptase was performed for each sample as a negative control to ensure that the RNA samples were not contaminated by DNA. Reactions were incubated at 42°C for 1 5 minutes followed by 5 minutes

at 99°C and 5 minutes on ice prior to amplification.

PCR was initiated by adding 80/71 of master mix containing ER-α- specific forward primer (5'-GGAGACATGAGAGCTGCCAAC-3') or ER-β- specific forward primer (5'-CAGCATTCCCAGCAATGTCAC-3') and GAPDH-specific forward primer (5'-GACATCAAGAAGGTGGTGAAGCAG- 3') directly to the 20 /I reverse transcriptase reaction. Final concentration of reagents in the ER-α 100μl PCR reaction was as follows:

0.25 μM each ER-specific primer, 0.1 /M each GAPDH primer, 1 x PCR Buffer (Gibco BRL, Gaithersburg MD), 0.2 mM dNTPs, 2 mM MgCl2 and

0.5 units of Taq DNA Polymerase (Gibco BRL, Gaithersburg MD). The ER- β reaction contained the same amount of reagents except for 1 mM MgCI₂. Two step PCR was carried out in a PE 9600 for 25 cycles as follows: 95°C for 30 sec, 64°C for 1 .5 min. Samples were incubated at 64°C for 10 min after amplification.

Twenty microliters of each sample were separated using a 1 .5% agarose gel and transferred to Hybond-N + (Amersham Pharmacia, Piscataway NJ) by capillary alkali Southern Blotting in 0.4 N NaOH, 0.6M NaCI. Blots were pre-hybridized at 42°C for 30 min in Rapid-Hyb buffer

(Amersham Pharmacia, Piscataway NJ). Oligonucleotide probes specific for ER-α (5'-TGAACCAGCTCCCTGTCTGCCAGGTTGGT-3'), ER-β (5'-

CCGCATACAGATGTGATAACTGGCGATGGA-3' ) and GAPDH (5'- GCTGTTGAAGTCACAGGAGACAACCTGGT-3') fragments were end-

32 labeled with P-γ ATP using Polynucleotide Kinase (Gibco BRL,

Gaithersburg MD). Probes were added to the blot at 3.0 x 10⁶ CPM/ml and incubated at 42°C for 1 hour. ER and GAPDH hybridizations were

done independently. Blots were washed once in 2x SSC, 0.1 % SDS at room temperature for 1 5 min then twice in 0.2x SSC, 0.1 % SDS at 42°C

for 1 5 min. Blots were then exposed to film.

Saos-2 cells expressed endogenous ER-β but not ER-α mRNA when

assessed by PCR (Figure 1 ). As a positive control, GAPDH mRNA was coamplified in these reactions. Whereas both cell lines contained GAPDH mRNA, only the Saos-2 cells contained ER-β. Similar results were

obtained for the LNCaPLN3 cells (data not shown).

EXAMPLE 3: DISCOVERY THAT ER-β UPREGULATES MT-II- IN SAOS-2 AND LNCAPLN3 CELLS

Cell culture and infection

Because message levels for ER-β were low, Saos-2 and LNCaPLN3

cells were engineered to overexpress hERβL by transient infection with a recombinant adenovirus (see Example 1 ) prior to differential display.

LNCaPLN3 and Saos-2 cells were cultured as described above. Sixteen hours prior to infection, the cells were plated in phenol red-free RPMI 1 640 medium supplemented with 10% charcoal/dextran-treated ("stripped") FBS (HyClone, Logan UT), 2 mM GlutaMAX-1 , 100 U/ml penicillin g, 100 μg/ml streptomycin sulfate. This medium was used for the remainder of the experiment.

Cells were infected with a 1 /20 dilution of an Ad5 hERβL virus (see

Example 1 ) using 2% stripped FBS phenol red-free medium with antibiotics and GlutaMAX-1 for 2 hours at 37° C. Medium containing virus was aspirated and the cells were washed with medium. Fresh medium was added and the cells allowed to recover overnight at 37°C.

The following day, cells were treated with 10 nM 17β-estradiol or vehicle for 24 hours and total RNA prepared for differential display using the TRIzol reagent (Gibco BRL, Gaithersburg MD). To remove residual DNA, samples were treated with RNase-free DNase I (Gibco BRL, Gaithersburg MD) at 1 unit/μg for 30 minutes at 37°C. RNA was purified from the reaction using RNeasy columns (Qiagen, Hilden Germany) and amount recovered estimated by UV spectrophotometry. Rapid Analysis of Differential Expression (RAPE) Reverse transcription (RT)

Following DNase I treatment, six micrograms total RNA was incubated with 1 x RT buffer (25 mM Tris-CI, pH 8.3, 37.6 mM KCI, 3 mM MgCI₂ and 5 mM DTT, from Genhunter, Nashville TN), 20 μM dNTP's (dA, C, G and TTP 2'- deoxynucleoside 5' triphosphates, From Gibco BRL (Gaithersburg MD), 0.2 μM HT„C (oligonucleotide AAGCTTTTTTTTTTTC) in a final volume of 600 μL. This reaction mixture was incubated at 65°C

for five minutes to denature secondary structures, followed by a ten minute incubation at 37°C. At this time 30 μ\ Superscript II reverse

transcriptase (200U/ /I, Gibco BRL, Gaithersburg MD) was added to the reaction and incubation proceeded for 1 hr at 37°C. The enzyme was

inactivated by heating at 75°C for five minutes. An aliquot of this reaction

was then used for the second strand synthesis by PCR. Polymerase chain reaction (PCR)

To 2μ\ of the RT reaction was added, 1 x PCR buffer (10 mM Tris- CI, pH 8.4, 100 mM KCI, 1 .5 mM MgCI₂ and 0.001 % gelatin), 2 μM dNTP's, 1 5 nM ³³P dATP (NEN, Boston MA), 1 unit AmpliTaq DNA polymerase (Perkin-Elmer, Norwalk CT) and ^μM arbitrary primer 5ΑAGCTTGCCATGG-3' for a total reaction volume of 20μl. This reaction mixture was then thermocycled using the following parameters: 92°C for 2 min, 1 cycle 92°C for 1 5 sec, 40°C for 2 min, 72°C for 30 sec, 40

cycles

72°C for 5 min Gel electrophoresis Duplicate samples of PCR products were separated by gel electrophoresis on a 6% denaturing polyacrylamide gel (5.7% acrylamide, 0.3% bisacrylamide, 42% urea and 51 % H₂0) in 1 x TBE buffer (0.1 M Tris, 0.09 M Boric Acid, 1 mM EDTA) for three hours at 2000 volts. The gel was then transferred to filter paper (Schleicher & Schuell, Keene NH), dried under vacuum at 80°C for one hour and exposed to X-ray film for

twenty four hours.

One band, designated 6a, was apparently upregulated in both the Saos-2 and LNCaPLN3 cells by 17-β estradiol (Figure 2). The developed film was then superimposed over the dried gel and the band's corners were marked using a 22 gauge syringe needle. The gel slice within these boundaries excised with a razor blade and immersed in 100 μl H₂0. This sample was boiled in a water bath for fifteen minutes, centrifuged for two minutes and the supernatant solution transferred to a new tube. Added to this sample was 5 μl of 10 mg/ml glycogen, 10μl of 3 M sodium acetate and 450 μl of 100% ethanol. The sample was mixed, allowed to precipitate overnight at -20°C and centrifuged for ten minutes at 10 000

g. The supernatant solution was removed, the pellet washed with 200 μl of 85% ethanol, dried and resuspended in 10 μl H₂0. A 3 μl aliquot was used in a reamplification PCR reaction in the presence of 1 x PCR buffer, 20 μM dNTP's, 0.2 μM arbitrary primer and 0.2 μM oligonucleotide HTnC and 2 units AmpliTaq polymerase using the same cycling parameters as the PCR reaction above. The resulting product was then used as probe in a Northern hybridization assay to confirm regulation and cloned into a bacterial plasmid for sequence analysis. Fragment cloning

Band 6a was cloned using the TA cloning kit (Invitrogen, Carlsbad CA). After lysis, colonies were screened by PCR for the correct insert size. Colonies were lysed in 20 mM Tris-HCI, pH 8, 50 mM KCI, 2.5 mM MgCI₂, 0.5% Tween 20 and 100 μg/ml Proteinase K by incubating for 30 min at 56°C then 10 min at 99°C to inactivate the Proteinase K. Of this

reaction 2μl was used in a PCR reaction of 20 mM Tris-HCI, pH 8, 50 mM KCI, 2.5 mM MgCI₂, 75 μM dNTPs, 375 nM M-1 3 forward and reverse primers and 2.5 U of Taq polymerase (Gibco BRL, Gaithersburg MD). The reactions were cycled as follows: 95°C for 30 sec; 64°C for 30 sec;

72°C for 45 sec for 30 cycles. One clone, designated 6a.2, was chosen

for sequencing. Fragment seqυencing

Clone 6a.2 was sequenced according to the ABI PRISM™ Dye

Terminator Cycle Sequencing Ready Reaction Kit with AmpliTaq DNA Polymerase using the recommended protocol from Applied Biosystems (Foster City CA). Spin columns (AGTC) were employed to remove unincorporated dye-labeled nucleotides after cycle sequencing. Automated DNA sequencing grade 4.75 % polyacrylamide gels were run for all the DNA sequencing samples using ABI 373 DNA sequencers. Sequencing data was edited using Sequence Navigator and assembled using DNAStar (DNAStar, Madison Wl). The sequence was trimmed of RADE primers and used in a BLAST search using Millennium Software (Boston MA). A nucleotide homology search of clone 6a.2 sequence revealed a 98% identity with human MT-II-. Over the coding sequence, however, there were no amino acid mismatches (Figure 3) Confirmation of regulation using Northern blots and Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR):

To confirm the results obtained with the PCR amplification of cellular mRNAs, Northern blots and qRT-PCR were used to assess the effect of 17-β estradiol or other compounds on message levels. Unless

otherwise noted, all experiments use cells transiently overexpressing hERβL in response to adenovirus infection as described above, and were

treated with compound for 24 hours. In some cases cells transiently overexpressing ER-α were used for comparison. Methods for adenovirus-

mediated overexpression of ER-α are identical to those described for ER-β.

Northern blots

Poly A + RNA was isolated from total RNA using Oligotex mRNA Isolation kit (Qiagen, Hilden Germany) according to manufacturers instructions. Six micrograms of mRNA or 10 μgs of total RNA were separated on a 1 .5% agarose, 0.22 M formaldehyde, 1 0 mM HEPES, 1 mM EDTA gel. RNA was transferred to Hybond-N (Amersham Pharmacia, Piscataway NJ) by capillary action in 20X SSC (Gibco BRL, Gaithersburg MD ) overnight. After transfer, the membrane was UV-cross-linked and dried at 80°C for 10 min. Northern Blots were pre-hybridized in Rapid

Hyb solution (Amersham Pharmacia, Piscataway NJ) for 30 minutes.

The insert from clone 6a.2 was isolated from the plasmid using PCR as described above for colony screen. The PCR product was purified from an agarose gel using Wizard Preps (Promega, Madison Wl). Re- amplified RADE fragments for band 6a or fragment from clone 6a.2 were random primer labeled using Redi-prime kit (Amersham Pharmacia, Piscataway NJ) according to manufacturers instructions. Unincorporated nucleotides were removed using a Nap-5 column (Amersham Pharmacia, Piscataway NJ) and incorporation of [³²P]-dCTP measured by liquid scintillation counting.

The probes were denatured at 100°C for 10 min and 1 .5 x 10⁶

CPM/ml of Rapid-Hyb hybridization solution was added to the membrane. The blots were hybridized at 65°C for 5 hours and were washed as

follows: Once in 2X SSC, 0. 1 % SDS at 65° C for 1 5 min; twice in 0.2X

SSC, 0.1 % SDS at 65°C for 1 5-30 min. Blots were exposed to film and

to a Phosphorlmager screen (Molecular Dynamics, Sunnyvale CA). After probing with 6a RADE fragment or 6a.2 cloned fragment, blots were probed with a cDNA homologous to GAPDH as above. Hybridization signal from 6a.2 fragment was normalized to that of GAPDH on a Phosphorlmager (Molecular Dynamics, Sunnyvale CA) to determine fold induction. QRT-PCR A MT-II fragment identical to clone 6a.2 except that it contained a

63 bp deletion was subcloned into pcDNA.3 (Invitrogen, Carlsbad CA). RNA was transcribed using the T7 Promoter Large Scale Transcription Kit (Novagen, Madison Wl). After phenol-chloroform extraction and ethanol precipitation the synthesized RNA was quantified and analyzed using UV spectrophotometry and gel electrophoresis.

Reverse transcription reactions were performed on 200 ng and 300 ng of DNased Saos-2 total RNA plus a known amount of MT-II standard RNA in a 20 μl reaction. The reaction contained 1 x PCR Buffer (Gibco BRL, Gaithersburg MD), 3.75 mM MgCl2, 1 .25 μM MT-ll-specific reverse

primer (5'-GGAATATAGCAAACGGTCAGGGTC-3'), 0.5 mM dNTPs, 1 mM DTT, 20 units of RNasin (Promega) and 200 units of Superscript II Reverse Transcriptase (Gibco BRL, Gaithersburg MD). Reactions were incubated at 42°C for 1 5 minutes followed by 5 minutes at 99°C and 5 minutes on ice prior to amplification PCR was initiated by adding 80μl of master mix containing MT-II specific forward primer (5'-GGCTCCTGCAAATGCAAAGAG-3') directly to the 20μl reverse transcriptase reaction. Final concentration of reagents in the 100μl PCR reaction was as follows: 0.25 μM each MT-II -specific primer, 1 x PCR Buffer (Gibco BRL, Gaithersburg MD), 0.1 mM dNTPs, 1 .5 mM MgCl2 and 0.5 units of Taq DNA Polymerase (Gibco BRL,

Gaithersburg MD). Two step PCR was carried out in a PE 9600 (Perkin- Elmer, Norwalk CT) for 40 cycles as follows: 95°C for 30 sec, 64°C for

1 .5 min. Samples were incubated at 64°C for 10 min after amplification.

PCR products were separated and analyzed on a reverse-phase ion- pair high performance liquid chromatography DNASep column (Sarasep, San Jose CA). The elution system was a gradient of acetonitrile in 0.1 M triethylammonium acetate (Fluka, Ronkonkoma NY) at a flow rate of 0.7 ml/min. The acetonitrile gradient increased from 14.6% to 1 6.6% over 5 minutes. The amount of product from the standard and the native RNA was determined by UV absorbance detection at 254 nm and signal was analyzed by an on-line integrator. From the chromatograms, the ratio of the area under each peak was used to determine the ratio of the amount of input MT-II standard RNA to the amount of native MT-II message in the Saos-2 RNA.

In LNCaPLN3 cells, the magnitude of MT-II induction by 1 7-β

estradiol is approximately 6 fold (Figure 4). In Saos-2 cells, 10 nM 17β-

estradiol upregulates MT-II mRNA as much as 14 fold (Figure 4). This upregulation in Saos-2 cells has been repeated in at least 20 experiments and the range of induction is 3.5-14 fold. The EC₅₀ of this regulation is 4.6 ± 2.7 nM as assessed by qRT-PCR (Figure 5). If Saos-2 cells are

treated with 10 nM 17β-estradiol and RNA prepared at different times post treatment, the first discernible increase in MT-II message occurs after 8 hours and the response peaks by 24 hours (Figure 6).

Since the samples prepared for differential display overexpressed hERβL, , MT-II regulation was assessed in native Saos-2 cells and those

overexpressing ERα. As illustrated in Figure 7 , endogenous levels of ER-

β or overexpressed levels of ER-α were ineffective at mediating 17β-

estradiol's regulation of MT-II.

Other compounds were tested for their ability to upregulate MT-II in Saos-2 cells. Ten nanomolar diethylstilbesterol (Sigma, St. Louis MO) and 0.1 μM genistein (RBI, Natick MA) both increased MT-II mRNA. One

micromolar of the antiestrogen, ICI-1 82780 (Zeneca, Wilmington DE), completely blocked induction by 1 7β-estradiol and genistein but had no

effect when given alone (Figure 8). Co-treatment with 1 μM of the

antiprogesterone/antiglucocorticoid, RU486 (Ligand Pharmaceuticals, La Jolla CA), did not block regulation by 1 7β-estradiol (data not shown).

EXAMPLE 4: TREATMENT OF SAOS-2 CELLS WITH CYCLOHEXIMIDE

To determine if the increase in MT-II mRNA requires new protein synthesis, cycloheximide was used after hERβL overexpression and

before treatment with 1 7-β estradiol to arrest translation.

Verification of cycloheximide's effect on protein synthesis

Cells were plated and infected with hERβL virus as described

above, then pre-treated with 10 μg/mL of cycloheximide or vehicle for one hour at 37°C. After this preincubation, 10 μg/mL of cycloheximide, 10

nM 17-β estradiol and 50 μCi/ml ³⁵S-Methionine (NEN, Boston MA) in

methionine-deficient media was added and incubation continued at 37°C

for 8 or 24 hours. Cells were washed 3x in cold PBS and then scraped \ off plates in 500 μl of PBS. . Cells were pelleted and resuspended in 200

μl of RIPA buffer (150 mM NaCI, 1 % NP-40, 0.5% DOC, 0.1 % SDS, and

50 mM Tris, pH 8.0). Methionine incorporation was measured by TCA precipitation. Five and ten microliters of each sample were spotted on individual Whatman filters. Filters were boiled in 10% trichloroacetic acid for 10 minutes, then washed 3 times for 10 minutes in deionized water, 3 times for 10 minutes in 95 % ethanol and once for 10 minutes in acetone. The dried filters were placed in scintillation fluid and counted for 1 minute. Treatment of cells with 1 7-β estradiol

Cells were plated and infected with hERβL virus as described

above. Cells were pretreated with 10 μg/mL of cycloheximide or vehicle

for one hour at 37°C. After this preincubation, 10 nM 1 7-β estradiol or

vehicle was added and incubation continued at 37°C for 8 or 24 hours.

RNA was isolated from the cells as described above and gene expression evaluated by Northern blot analysisor qRT-PCR as described in Example 3.

Verification of functional ER-β protein after cycloheximide treatment Duplicate cultures were prepared and treated as described in the previous section. After the 8 hour incubation, cells were washed four times with DMEM to remove 1 7-β estradiol. Cycloheximide (10 μg/mL)

was added to all samples as was 1 nM [³H]-estradiol. Some cells were co-treated with 0.3 μM diethylstilbesterol to estimate nonspecific binding.

After incubation for 2.5 hr at 37°C, cells were washed with DMEM and

lysed with 0.1 % sodium dodecyl sulfate. DPM were measured by liquid scintillation counting.

The MT promoter contains glucocorticoid and metal response elements, but EREs have not been described. It is possible the effect of 17-β estradiol on MT-II mRNA expression is indirect. After

overexpression of ER-β, Saos-2 cells were treated with cycloheximide to

severely limit new protein synthesis (Figure 9A). Although comparable amounts of receptor protein were expressed with and without cycloheximide treatment as measured by a whole cell binding assay (Figure 9B), MT-II induction did not occur (Figure 9C,D).

EXAMPLE 5 : TREATMENT OF RATS WITH 17-β ESTRADIOL

Because 1 7-β estradiol upregulates MT-II in LNCaPLN3 cells, a

prostate cancer cell line, we looked for a similar response in the rat prostate.

All animals were treated according to institutional guidelines using approved protocols. Adult (1 5-1 9 week, ~375g) castrated Sprague- Dawley rats were purchased from Taconic Farms (Germantown NY). Ten days after castration, rats were injected with vehicle, 16 μg 17-β

estradiol, 16 μg diethylstilbesterol (Sigma, St. Louis MO), or 16 μg 17-β estradiol plus 1.6 mg raloxifene (synthesized in-house) subcutaneously once per day for three days. Approximately 24 hr after the last dose, rats were euthanized by C0₂ asphyxiation and the prostate excised. Total RNA was prepared and analyzed for MT-II mRNA as described above. In another experiment, rats were dosed for 1 , 2 or 3 days with 16 μg 17-β estradiol before prostate tissue was collected. Metallothionein-ll mRNA increased five-fold after 17-β estradiol

treatment. A similar response occurred when rats were dosed with diethylstilbesterol, but this upregulation was blocked by coadministration with a 100 fold excess of raloxifene hydrochloride, an estrogen antagonist (Figure 10). Although the initial experiments used tissue from rats dosed for three days, the upregulation of MT-II in the rat prostate is first seen after two days of dosing with 17-β estradiol (Figure 1 1 ). EXAMPLE 6 : ERE-LUCIFERASE REPORTER ASSAY USING MCF-7 CELLS

Because MT-II regulation in Saos-2 cells is mediated by ER-β and

not ER-α, another assay was needed to compare selectivity of compounds for these two receptors in cell based assays. MCF-7 is an estrogen- responsive breast cancer cell line and, in our hands, expresses only ER-α

as measured by RT-PCR (data not shown). When MCF-7 cells are transiently infected with an ERE-reporter gene construct and treated with 1 7-β estradiol, reporter gene activity can be measured.

MCF-7 HTB 22, ATCC , Manassas VA) cells are passaged twice a week with growth medium [D-MEM/F-1 2 medium containing 10% (v/v) heat-inactivated fetal bovine serum 1 00 U/ml penicillin g, 1 00 μg/ml

streptomycin sulfate, 2 mM GlutaMAX-1 ]. The cells are maintained in

vented flasks at 37°C inside a 5% Cθ2/95% humidified air incubator.

One day prior to treatment, the cells are plated with growth medium at

25,000/well into 96 well plates and incubated at 37°C overnight.

The cells are infected for 2 hr at 37°C with 50 μl/well of a 1 : 10

dilution of adenovirus 5-ERE-TK-luciferase (Bodine et. al., 1 997) in experimental medium [phenol red-free DMEM/F-1 2 medium containing 10% (v/v) heat-inactived charcoal-stripped fetal bovine serum, 100 U/ml penicillin g, 100 μg/ml streptomycin sulfate, 2 mM GlutaMAX-1 , 1 mM

sodium pyruvate]. The wells are then washed once with 1 50 μl of

experimental medium. Finally, the cells are treated for 24 hr at 37°C in replicates of 8 wells/treatment with 1 50 μl/well of vehicle ( <_ 0.1 % v/v

DMSO) or compound that is diluted _> 1 000-fold into experimental medium. After treatment, the cells are lysed on a shaker for 1 5 min with 25 μl/well of 1 X cell culture lysis reagent (Promega, Madison Wl). The cell

lysates (20 μl) are transferred to a 96 well luminometer plate, and luciferase activity is measured in a MicroLumat LB 96 P luminometer (EG & G Berthold, Bad Wildbad Germany) using 100 μl/well of luciferase substrate (Promega, Madison Wl). Prior to the injection of substrate, a 1 second background measurement is made for each well. Following the injection of substrate, luciferase activity is measured for 10 seconds after a 1 second delay. After subtracting background subtracts the mean and standard deviation are calculated.

Compounds which induce MT-II in Saos-2 cells and do not stimulate reporter gene activity in MCF-7 cells are thus selective for ER-β. In addition, results from these two assays can also be used to select a compound that is selective for ER-α as shown in Figure 12.

Discussion

Using differential display, it was unexpectedly discovered that ER-β increases MT-II mRNA in two cell lines treated with 17-β estradiol. To our knowledge, this is the first gene discovered to be regulated by this new form of the ER. This response has been extensively characterized in the human osteosarcoma cell line Saos-2. A variety of estrogens can upregulate MT-II, and this response is blocked by cotreatment with the estrogen antagonist ICI-182780. The EC₅₀ for 17-β estradiol is approximately 5 nM and this response is mediated by ER-β acting through as yet unknown proteins. The action of ER-β on MT-II is not a general phenomenon as a

survey of over a dozen cell lines failed to reveal this induction in samples other than Saos-2 and LNCaPLN3 cells. However, the fact that MT-II is regulated by 1 7-β estradiol in the rat prostate strengthens the possibility

that this is a physiologically relevant response and not a artifact of receptor overexpression in cancer cell lines. We are confident that this prostate response is mediated by the ER for two reasons: A nonsteroidal estrogen (diethylstilbesterol) upregulates MT-II and raloxifene, an estrogen agonist/antagonist, blocks 1 7-β estradiol's action. However, at this time

it is not clear if this induction reflects activity of ER-β and/or ER-α.

Although ER-β is the predominant form of the ER in the rat prostate, ER-α

is also present. Current studies are underway to define the receptor type responsible for this in vivo response.

Metallothioneins are low molecular weight, cysteine rich proteins that bind metals such as cadmium, copper and zinc. Although the first metallothionein was discovered over forty years ago (Vallee, 1 957), debate continues as to their function. Several proposals have been made and these include protection form metal toxicity, regulation of zinc and copper homeostasis and defense against oxidative stress. Regulation of energy balance has also been implicated because, after reaching sexual maturity, MT (-I and -II) knockout mice become obese (Beattie et al, 1 998). Recently, studies have detailed how MT may act to regulate zinc homeostasis in cells. Using purified zinc-dependent enzymes such as E. coli alkaline phosphatase, bovine carboxypeptidase A and sheep sorbitol dehydrogenase two recent publications show how agents in the cellular milieu can facilitate exchange between zinc complexed with MT and (metal depleted) apoenzymes (Jacob et al, 1 998; Jiang et al, 1 998). Citrate and glutathione can influence the direction of zinc transfer and thus regulate enzyme activity depending on the redox state of the cell. Although not an enzyme, the ER also requires zinc for activity, and ER from MCF-7 cells can reversibly exchange zinc with purified MT in vitro (Cano-Gauci et al, 1 996).

A myriad of agents can regulate MT levels, including glucocorticoids and metals such as cadmium (for review see Hamer, 1986). Estrogens are not a classical regulators of MT, but two intriguing papers appear in the literature. First, a two week treatment of female rats with 1 7-β estradiol upregulated a copper binding protein in intestinal

mucosa which reduced the amount of copper absorbed into the plasma. Since the molecular weight of this protein was about 10K, the authors suggest it is MT (Cohen et al, 1 979). Recently, MT was identified in a subtractive hybridization screen of uterine mRNAs regulated after a single injection of 1 7α ethinyl estradiol Rivera-Gonzalez et al, 1 998). Although

the isoform is not identified, a MT transcript increased three fold between four and eight hours after 1 7-α ethinyl estradiol stimulation. In addition, it is not clear which receptor type effects this regulation because ER-α,

not ER-β, is the most abundant ER in the uterus (Kuiper at al, 1 996;

Couse et al, 1997).

Since the function of MT is only beginning to be elucidated and is completely unknown for ER-β, understanding the significance of their

association is impossible at this time. However, even if the connection between these two proteins is unclear, the observation of regulation can still be exploited to help design an ER-β or ER-α specific ligand. The fact

that genistein upregulates MT-II- in Saos-2 cells as well or better than 17β-estradiol indicates an ER-β-selective compound can effectively direct

transcription. In addition, several structurally diverse compounds which bind to ER-β can also upregulate this message in Saos-2 cells. When data

on a compound's ability to regulate MT-II in Saos-2 cells are coupled with information about how it regulates reporter gene activity in an MCF-7 cell, ER-β and ER-α selectivity can be assessed. Thus, when used together,

these two assays are tools to help design selective compounds for either ER type. Finally, if the regulation in the prostate can also be shown to be mediated by ER-β, then the in vivo activity of compounds can be

assessed, providing another valuable tool for drug discovery. REFERENCES

Beattie JH, Wood AM, Newman AM, Bremner I, Choo KHA, Michalska AE, Duncan JS, Trayhurn P. 1998. Obesity and hyperleptinemia in metallothionein (-I and -II) null mice. Proc Natl Acad Sci (USA) 95:358- 363.

Bhat RA, Harnish DC, Stevis PE, Lyttle CR, Komm BS. 1998. A novel human estrogen receptor β: Identification and functional analysis of additional N-terminal amino acids. J Steroid Molec Biol, in press.

Blondelle SE, Houghten RA. 1 996. Novel antimicrobial compounds identified using synthetic combinatorial library technology. Trends Biotechnol 14(2):60-65.

Bodine PVN, Green J, Harris HA, Bhat R, Stein GS, Lian JB, Komm B. 1997. Functional properties of a conditionally phenotypic, estrogen- responsive, human osteoblast cell line. J Cell Biochem 65:368-387.

Cano-Gauci DF, Sarkar B. 1 996. Reversible exchange between metallothionein and the estrogen receptor zinc finger. FEBS Letters 386:1-4. Cohen Dl, lllowsky B, Linder MC. 1 979. Altered copper absorption in tumor-bearing and estrogen-treated rats. Am J Physiol 236(3):E309-1 5.

Couse JF, Lindzey J, Grandien K, Gustafsson J, Korach, KS. Tissue distribution and quantitative analysis of estrogen receptor-α (ERα) and

estrogen receptor-β (ERβ) messenger ribonucleic acid in the wild type and

ERα-knockout mouse. 1 997. Endocrinology 1 38(1 1 ):461 3-4621 .

Cowley SM, Hoare S, Mosselman S, Parker MG. 1997. Estrogen receptors α and β form heterodimers on DNA. J Biol Chem

272(32): 1 9858-1 9862.

Davis AR, Kostek B, Mason B, Hsiao CL, Morin J, Dheer SK, Hung P. 1985. Expression of hepatitis B surface antigen with a recombinant adenovirus. Proc Natl Acad Sci (USA) 82:7560-7564.

Dotzlaw H, Leygue E, Watson PH, Murphy LC. 1 996. Expression of estrogen receptor-beta in human breast tumors. J Clin Endocrinol Metab 82(7):2371 -2374.

Enmark E, Pelto-Huikko M, Grandien K, Lagercrantz S, Lagercrantz J, Fried G, Nordenskjold M, Gustafsson J. 1 997. Human estrogen receptor- β gene structure, chromosomal localization, and expression pattern. J Clin

Endocrinol Metab 82(1 2):4258-4265

Green S, Kumar V, Walter KP, Chambon P. 1986. Structural and functional domains of the estrogen receptor. Cold Spring Harbor Symposia on Quantitative Biology Ll:751 -758.

Gluzman Y, Reichl H, Solnick D. 1 982. Helper-free adenovirus type-5 vectors. In Eukaryotic Viral Vectors, Y Gluzman ed., (Cold Spring Harbor Laboratory), pp 187, 192.

Hamer DH. 1986. Metallothionein. Ann Rev Biochem 55:913-951 .

lafrati MD, Karas RH, Aronovitz M, Kim S, Sullivan TR, Lubahn DB, O'Donnell Jr. TF, Korach KS, Mendelsohn ME. 1997. Estrogen inhibits the vascular injury response in estrogen receptor-α-deficient mice. Nature

Med 3(5):545-548.

Jacob C, Maret W, Vallee BL. 1998. Control of zinc transfer between thionein, metallothionein and zinc proteins. Proc Natl Acad Sci (USA) 95:3489-3494. Jiang L, Maret W, Valee BL. 1 998. The glutathione redox couple modulates zinc transfer from metallothionein to zinc-depleted sorbitol dehydrogenase. Proc Natl Acad Sci (USA) 95:3483-3488.

Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA.

1 996. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci (USA) 93:5925-5930. Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson, JA. 1 997. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors α and β. Endocrinology 1 38(3):863-

870.

Lau K, Leav I, Ho S. 1 998. Rat estrogen receptor-α and -β, and

progesterone receptor mRNA expression in various prostatic lobes and microdissected normal and dysplastic epithelial tissues of the Noble rats. Endocrinology 139(1 ):424-427.

Lobo R. 1 995. Benefits and risks of estrogen replacement therapy. Am J Obstet Gynecol 1 73(3):982-989.

Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach, KS. 1 993. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci (USA) 90: 1 1 1 62-1 1 1 66.

Mosselman S, Polman J, Dijkema R. 1 996. ERβ: identification and

characterization of a novel human estrogen receptor. FEBS Letters 392:49-53.

Ogawa S, Inoue S, Watanabe T, Hiroi H, Orimo A, Hosoi T, Ouchi Y, Muramatsu M. 1 998. The complete primary structure of human estrogen receptor β (hERβ) and its heterodimerization with ER α in vivo and in vitro.

Biochem Biophys Res Comm 243: 1 22-1 26.

Pan LC, Ke HZ, Simmons HA, Crawford DT, Chidsey-Fink KL, McCurdy SP, Schafer JR, Kimbro KS, Taki M, Korach KS, Thompson DD. Estrogen receptor-alpha knockout (ERKO) mice lose trabecular and cortical bone following ovariectomy. 1 997. J Bone Mineral Res 1 2 (Supplement 1 ):S134.

Pettersson K, Grandien K, Kuiper GGJM, Gustafsson J. 1 997. Mouse estrogen receptor β forms estrogen response element-binding

heterodimers with estrogen receptor-α. Mol Endocrinol 1 1 : 1486-1496. Rivera-Gonzalez R, Peterson DN, Tkalcevic G, Thompson DD, Brown TA. 1 998. Estrogen-induced genes in the uterus of ovariectomized rats and their regulation by droloxifene and tamoxifen. J Steroid Biochem Molec Biol 64(1 /2):13-24.

Sagraves R. 1995. Estrogen therapy for postmenopausal symptoms and prevention of osteoporosis. J Clin Pharmacol 35:2s-10s.

Shughrue PJ, Komm B, Merchenthaler I. 1996. The distribution of estrogen receptor-β mRNA in the rat hypothalamus. Steroids 61 :678- 681 .

Trembley GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, Giguere V. 1997. Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor β. Mol Endocrinol 1 1 :353-365.

Vallee BL. 1957. A cadmium protein from equine kidney cortex. J Am Chem Soc 79:4813.

Vladusic EA, Hornby AE, Guerra-Vladusic FK, Lupu R. 1 998. Expression of estrogen receptor β messenger RNA variant in breast cancer. Cancer

Res 58:210-214. Webb P, Lopez GN, Greene GL,. Baxter JD, Kushner PJ. 1 992. The limits of the cellular capacity to mediate an estrogen response. Mol Endocrinology. 6(2): 1 57-67.

Claims

What is claimed is:

1 . A method for screening a test compound that binds to an ER in a receptor binding assay, wherein said method detects ER-β polypeptide- mediated transcription, said method comprising the steps of:

(a) providing a cell which comprises at least one estrogen-regulated DNA sequence encoding MT-II and at least one DNA sequence encoding ER-β polypeptide, wherein said receptor is transcriptionally active;

(b) contacting said cell with either said test compound which binds ER or a control; and

(c) detecting the expression of said MT-II, wherein enhanced expression of said MT-II relative to a control indicates that said test compound has estrogen agonist activity.

2. The method of claim 1 , wherein the DNA sequence encoding ER-β

polypeptide is incorporated into an adenovirus.

3. The method of claim 2 wherein the adenovirus is a replication- defective Ad5 virus.

4. The method of claim 1 , wherein said cells endogenously express ER-β mRNA at a higher rate than ER-α mRNA.

5. The method of claim 4, wherein said cells are transformed with a recombinant DNA plasmid comprising a polynucleotide encoding a human ER-β operably linked to a suitable promoter, wherein said transformed

cells express human ER-β at higher levels than the cells which have not

been transformed.

6. The method of claim 1 wherein said cells are Saos-2 or LNCaPLN3 cells.

7. The method of claim 1 , wherein said cells do not have a functional

ER-α.

8. A method for screening a test compound that binds to an ER in a receptor binding assay, wherein said method detects inhibition of ER-β

polypeptide-mediated transcription, said method comprising the steps of:

(a) providing a cell which comprises at least one estrogen DNA sequence encoding MT-II and at least one DNA sequence encoding ER-β

polypeptide, wherein said receptor is transcriptionally active;

(b) contacting said cell with one or more estrogens in the presence of the test compound known to bind ER; and

(c) detecting the expression of said MT-II, wherein decreased expression of said MT-II relative to the addition of one or more estrogens alone indicates that said test compound has estrogen antagonist activity.

9. The method of claim 8, wherein the DNA encoding ER-β

polypeptide is incorporated into an adenovirus.

10. The method of claim 9 wherein the adenovirus is a replication- defective Ad5 virus.

1 1 . The method of claim 8, wherein said cells endogenously express ER-β mRNA at a higher rate than ER-α mRNA.

12. The method of claim 1 1 , wherein said cells are transformed with a recombinant DNA plasmid comprising a polynucleotide encoding a human ER-β operably linked to a suitable promoter, wherein said transformed cells express human ER-β at higher levels than the cells which have not been transformed.

13. The method of claim 1 2 wherein said cells are Saos-2 or LNCaPLN3 cells.

14. The method of claim 13, wherein said cells do not have a functional ER-α.

1 5. A method of screening test compounds to identify drug candidates which mimic estrogen's effect on ER-β- or ER-α-mediated transcription,

said method comprising the steps of:

(a), contacting said test compound with a plurality of: (i) first cells comprising at least one endogenous DNA sequence encoding MT-II and at least one DNA sequence encoding a ER-β polypeptide, wherein said

receptor is transcriptionally active and (ii) second cells comprising an ERE reporter gene construct, wherein said cells express ER-α polypeptide;

(b) identifying compounds which increase expression of MT-II in said first cells relative to control but have minimal effect on expression of said reporter gene in said second cells, wherein said compounds are considered ER-β selective; or

(c) identifying compounds which increase expression of the reporter gene in said second cells relative to control but have minimal effect on expression of MT-II in said first cells, wherein said compounds are considered ER-α selective.

1 6. The method of claim 1 5, wherein the DNA encoding ER-β

polypeptide of said first cells is incorporated into an adenovirus.

17. The method of claim 16 wherein the adenovirus is a replication- defective Ad5 virus.

18. The method of claim 1 5, wherein said first cells endogenously express ER-β mRNA at a higher rate than ER-α mRNA.

1 9. The method of claim 1 5, wherein said first cells are transformed with a recombinant DNA plasmid comprising a polynucleotide encoding a human ER-β operably linked to a suitable promoter, wherein said transformed first cells express human ER-β at higher levels than said second cells.

20. The method of claim 1 5 wherein said first cells are Saos-2 or LNCaPLN3 cells.

21 . The method of claim 1 5, wherein said first cells do not have a functional ER-α and said second cells do not have a functional ER-β.

22. The method of claim 15, wherein said second cells were transformed with a DNA polynucleotide comprising an ERE operably linked to a reporter gene.

23. The method of claim 22, wherein said reporter gene is a luciferase gene.

24. A method of screening test compounds to identify drug candidates which inhibit estrogen's effect on ER-β- or ER-α-mediated transcription, said method comprising the steps of:

(a), contacting said test compound in the presence and absence of one or more estrogens with a plurality of:

(i) first cells comprising at least one endogenous DNA sequence encoding a metallothionein (MT-II) gene and at least one DNA sequence encoding a ER-β polypeptide, wherein said receptor is transcriptionally active; and (ii) second cells comprising an ERE-reporter gene construct, wherein said cells express ER-α polypeptide; and

(b) identifying compounds which decrease expression of MT-II in said first cells relative to treatment with one or more estrogens but have minimal effect on expression of said reporter gene in said second cells, wherein said compounds are considered ER-β selective; or

(c) identifying compounds which decrease expression of the reporter gene in said second cells relative to treatment with one or more estrogens but have minimal effect on expression of MT-II in said first cells, wherein said compounds are considered ER-α selective.

25. The method of claim 24, wherein the DNA encoding ER-β polypeptide of said first cells is incorporated into an adenovirus.

26. The method of claim 25 wherein the adenovirus is a replication- defective Ad5 virus.

27. The method of claim 24, wherein said first cells endogenously express ER-β mRNA at a higher rate than ER-α mRNA.

28. The method of claim 24, wherein said first cells are transformed with a recombinant DNA plasmid comprising a polynucleotide encoding a human ER-β operably linked to a suitable promoter, wherein said transformed first cells express human ER-β at higher levels than said second cells.

29. The method of claim 24 wherein said first cells are Saos-2 or LNCaPLN3 cells.

30. The method of claim 24, wherein said first cells do not have a functional ER-α and said second cells do not have a functional ER-β.

31 . The method of claim 24, wherein said second cells were transformed with a DNA polynucleotide comprising an ERE operably linked to a reporter gene.

32. The method of claim 31 , wherein said reporter gene is a luciferase gene.