WO2000025800A9 - Steroid receptor rna activator - Google Patents

Abstract

The invention relates to a Steroid Receptor RNA Activator (termed SRA) which acts as a steroid hormone receptor coactivator and to the detection of RNA transcripts capable of transactivation of a DNA sequence. The use of SRA to detect protein/RNA interactions including the specific situation of RNA/protein interactions for transactivation of heterologous promoters is described. Methods to evaluate or modulate type I or 'classical' nuclear receptors are also described. Furthermore, SRA and its expression levels can facilitate design and evaluation of agents capable of modulating its levels or interactions with other proteins. It also relates to the levels of SRA expression facilitating detection, prognosis or treatment of steroid-dependent tumors or to the detection of prenatal or postnatal cortisol resistance. Finally, SRA expression patterns can be utilized to detect steroid receptor expression patterns.

Description

Steroid Receptor RNA Activator

This invention was made with government support. The government has certain rights in the invention.

FIELD OF THE INVENTION The invention relates to the technical fields of biotechnology and medicine.

The invention more specifically relates to an RNA transcript which acts as a steroid hormone receptor coactivator (termed SRA for Steroid Receptor RNA Activator) and its use for detection of RNA/protein interaction, transcription of heterologous promoters, modulation and activation of target genes, detection and treatment of hormone dependent tumors, and prenatal and postnatal detection of cortisol resistence.

BACKGROUND OF THE INVENTION

Nuclear receptors are members of a structurally and functionally related family of ligand-activated and sequence-specific eukaryotic transcription factors. By modulating the transcription of target genes in response to their own ligands and other afferent signals, they play key physiological roles in the regulation of development, metabolism and reproduction. Receptor activation involves a multifaceted cascade of events which results in the binding of the receptor to specific regulatory DNA- sequences and which culminate in the modulation of target gene expression. (Tsai and O'Malley, 1994; Mangelsdorf and Evans, 1995). The pleiotropic functions of nuclear receptors are reflected in the tripartite structure of their functional domains. A highly conserved DNA-binding domain (DBD) mediates receptor binding to cz^'s-acting hormone response elements (HREs). Common to nearly all nuclear receptors is the activation function AF-2 in the distal carboxyl-terminus of the ligand binding domain (LBD). A highly conserved amphipathic helix in AF-2 has been shown to be important for ligand binding and hormone-dependent transactivation (Danielian et al., 1992; Negeto et al, 1992; Lanz and Rusconi, 1994). (Transactivation is used herein to indicate the process by which a protein binds sequence specifically to DΝA and acts in trans to increase transcription of one or more genes.) The variable amino-terminal domain of nuclear receptors is extended in the Type I or "classical" receptor subclass comprising the receptors for androgens (AR), estrogens (ER), glucocorticoids (GR), mineralocorticoids (MR) and progestins (PR). This modulatory domain contains a strong and autonomous transactivation function (AF- 1 ), which has been suggested to be a critical determinant for target gene specificity (Tora et al., 1988).

The role of activated nuclear receptors is to direct the assembly and stabilization of a preinitiation complex in a transcriptionally permissive environment at the promoter of a target gene. This involves the functional interaction of the receptor with factors contained in the transcription preinitiation complex (Tsai et al. ,

1987; Beato and Sanchez-Pacheco, 1996) and with other DΝA-bound transcription activators (Jonat et al., 1990; Yang- Yen et al., 1990). Such interactions are necessary but not sufficient for accurate regulation of transcription. The initial finding that in cotransfection assays distinct receptors interfere with or "squelch" each other's transcriptional functions (Meyer et al., 1989) indicated that common limiting factors were involved in transactivation by different receptors. Several biochemical and genetic screens have since identified a number of proteins with which activated receptors undergo ligand-dependent interaction. These proteins fulfill a number of functional criteria to suggest they are coactivators for nuclear receptors. These criteria include significant enhancement transactivation by nuclear receptors without altering basal transcriptional activity; when overexpressed, they specifically reverse squelching between different receptors; and they contain autonomous, transferable activation domains (Horwitz et al., 1996). Coactivators that have recently received considerable attention are members of the SRC-gene family and the 'cointegrators' p300 and CBP. SRC-1 has been cloned in our laboratory as a general coactivator for nuclear receptors (Onate et al., 1995) and has since been termed variously as pl60/NCoA-l (Kamei et al., 1996) or ERAP-160 (Halachmi et al., 1994). SRC-1 contains autonomous activation domains which may physically contact basal transcription factors (Gill et al., 1998; Onate et al., 1998). Highlighting the critical physiological role of coactivators, the targeted deletion of SRC-1 in vivo causes partial hormone insensitivity (Xu et al., 1998). Other coactivators have been subsequently identified and characterized that are structurally and functionally related to SRC-1, including transcription intermediary factor-2 (TIF-2/GRIP-1/SRC-2) (Hong et al., 1996; Noegel et al., 1996), and the p300/CBP cointegrator-interacting protein (p/Cπ>/ACTR/RAC-3/AIB-l/TRAM-l/SRC-3) (Takeshitaet al., 1996; Anzick et al, 1997; Chen et al, 1997; Li etal, 1997; Torchia etal, 1997). Since the three members of the SRC family exhibit striking conservation of a recurrent LXXLL motif, which is thought to mediate the interaction of SRC coactivators with the AF-2 domain of ligand receptors (Heery et al, 1997; Torchia et al, 1997), the SRC family members are referred to as AF-2 coactivators. Another subclass of nuclear receptor coactivators, the cointegrators, includes the CREB binding protein (CBP) (Chrivia et al, 1993) and the closely related adenovirus ElA-associated p300 (Eckner et al,

1994). p300 and CBP are general coactivators that interact not only with the LBD of multiple nuclear receptors but also with a wide variety of transcriptional activators (Eckner et al, 1994; Goldman et al, 1997).

Recent evidence has linked transcriptional regulation by nuclear receptors to processes that modify the chromatin structure of their target genes (for review see (Kadonaga, 1998). Acetylation of core nucleosomal histones by histone acetyltransferase (HAT) proteins has been closely linked to transcriptional activation and is believed to facilitate access of activators to their cognate ci-s-acting sequences. HAT activity has been identified in several coactivators that interact with nuclear receptors, including P/CAF (Yang et al, 1996) p300/CBP (Bannister and Kouzarides, 1996; Ogryzko et al, 1996), SRC-1 (Spencer et al, 1997) and SRC-3/ACTR (Chen et al, 1997). In addition, nuclear receptors have been shown to interact with proteins that couple ATPase activity to chromatin modification, including BRG-1 (Fryer and Archer, 1998; Korzus et al, 1998). These observations suggest that rearrangement of chromatin is involved in transcriptional regulation by nuclear receptors. Abundant functional and in vitro interaction studies suggest that receptor/coregulator and coregulator/coregulator interaction result in the assembly of large complexes at the promoters of target genes (Klein-Hitpass et al, 1990). Recently, it has emerged that these interactions are not of uniform affinity, but that a hierarchy exists. As a consequence, different coregulator subclasses are arranged into largely distinct preformed sub-complexes, probably facilitating their assembly into promoter-specific configurations (Korzus et al, 1998; McKenna et al, 1998). These findings emphasize the fact that transcriptional regulation by nuclear receptors is a convoluted process that involves the recruitment of heterogeneous, multi-functional enzymatic complex to the DNA template.

To date the majority of receptor interacting proteins have been identified by genetic screens, such as the yeast two-hybrid system using the LBD of a nuclear receptor as a bait. This approach has led to the identification of numerous AF-2 coactivators with common structural and functional features. The "classical" type I steroid receptors, however, also exert transactivation by their amino-terminal transcription activation function, AF-1. For some steroid receptors, AF-1 and AF-2 have a distinct pattern of cell and promoter specificity (Bocquel et al, 1989; Tasset et al, 1990). Reasoning that specificity in steroid receptor-mediated transactivation might be provided by factors that associate with the poorly conserved AF-1, the present invention found coregulators that interact with the amino-terminal AF-1 domain of hPR. The present invention provides functional and mechanistic evidence that SRA acts as an RNA transcript in distinct ribonucleoprotein complexes, one of which contains the AF-2 coactivator SRC-1.

SUMMARY OF THE INVENTION

An obj ect of the present invention is the provision of an RNA Transcript which acts as a steroid receptor co-activator.

A further object of the present invention is a method for detection of RNA Transcripts which are capable of transactivating a gene or DNA construct.

An additional object of the present invention is a method for determining the presence of specific RNA transcripts.

A further object of the present invention is a method for detecting RNA/protein interaction based on the ability of steroid receptor RNA activators to transactivate heterologous promoter.

A further object of the present invention is an SRA product for detecting RNA/protein interaction.

An additional object of the present invention is a kit for selective isolation of specific RNA transcripts. A further object of the present invention is a method for evaluation of type I or "classical" nuclear receptors.

A further object of the present invention is a method for modulating type one or "classical" nuclear receptor activation.

An additional object of the present invention is a method for treatment of hormone dependant tumors.

An additional object of the present invention is a method for the detection of hormone dependant tumors.

A further object of the present invention is a method for evaluation of agents to modulate SRA levels.

A further object of the present invention is a method for diagnosing prenatal or postnatal detection of cortisol resistance. An additional object of the present invention is a method of designing a therapeutic agent.

Another object of the present invention to provide RNA transcripts which act as steroid hormone coactivators.

Another object of the invention is to provide methods for detection of SRAs or other RNA transcripts which are capable of transactivating a gene or DNA construct.

An object of the invention is to provide a method and a kit for verifying the identity of an RNA transcript.

Another object of the invention is to provide a method and a product for the detection of RNA/protein interactions.

Another object of the invention is to provide a method and a product for the selective isolation of specific RNA transcripts.

A further object of the invention is to provide a method for the diagnosis of hormone-dependent tumors. A further object of the invention is to provide a method for the detection of expression patterns of steroid receptors.

Thus in accomplishing the foregoing objects, there is provided in accordance with one aspect of the present invention an RNA transcript having the ability to selectively enhance transcriptional activation of steroid receptors without the requirement of translation of said RNA transcript into protein in eukaryotic cells.

In specific embodiments the RNA transcript is selected from the group consisting of SRA- 1 , SRA-2 and SRA-3

In another specific embodiment, the RNA transcript is the core sequence. Another embodiment includes an RNA having the ability to transactivate a DNA response element wherein said activation indicates RNA/protein binding or interaction. A further embodiment is a method for the detection of an RNA/protein interaction between an RNA sequence and a protein. This method comprises (a) inserting a first DNA sequence coding for an RNA transcript sequence into a first expression vector transcriptionally controlled by an inducer, wherein said inserted DNA sequence is inserted in series with vector DNA coding for a linker sequence and SRA and wherein expression of this first construct produces a first product containing a RNA sequence consisting of said RNA transcript sequence linked in series to said linker sequence and said SRA; (b) inserting a second DNA sequence coding for a first protein into a second expression vector wherein said second DNA sequence is in frame with DNA coding for a GAL4 DNA binding domain and wherein expression and translation of this second construct produces a second product containing a fusion protein of first protein and GAL4 binding domain, said fusion protein capable of binding to GAL4-binding DNA element; (c) transfecting cells with the vector constructed in step (a), the vector constructed in step (b), an expression vector for the activation domain 2 of SRC-1 and a reporter construct; (d) inducing the vector constructed in step (a) with an appropriate inducing agent to promote transcription of the DNA coding for the R A transcript; (e) transcribing and translating said vector constructed in step (b), the expression vector and the reporter construct; and (f) measuring the reporter product or activity.

In specific embodiments of this method the DNA sequence coding for RNA transcript contains at least one in-frame stop codon. Preferably the at least one in- frame stop codon is in the 5' region of the sequence. In another embodiment of this method the reporter construct codes for a protein conferring resistance to or improved survivability in an otherwise toxic environment to the host cell and the expression vector for the activation domain 2 SRC-1 is replaced with hPRΔDBD.

An additional embodiment comprises the additional step of contacting the transferred cells of step (c) with sufficient cycloheximide to substantially reduce or eliminate de novo protein synthesis.

A further embodiment of the invention includes a viral vector under the transcriptional control of an exogenous inducer in which a heterologous DNA sequence has been inserted in series with a linker sequence and an RNA coactivator sequence, wherein transfection and induction of transcription with said inducer produces an RNA transcript comprising the RNA sequences for the inserted DNA, the linker and the RNA coactivator in series.

In the preferred embodiment of the viral vector the RNA coactivator is SRA. In specific uses the viral vector is selected from a group consisting of retroviral, adenoviral, and vaccinia viral vectors.

Another specific embodiment includes a method for the detection of RNA transcript that is capable of transactivating a gene or DNA construct. This method comprises the steps of (a) inserting a first DNA sequence coding for said RNA transcript into a first expression vector transcriptionally controlled by an inducer, wherein said inserted DNA sequence is in series with vector DNA coding for a linker sequence and SRA and wherein expression of the first construct produces an RNA sequence consisting of RNA transcript linked in series to a linker sequence and a SRA; (b) inserting a second DNA into a second expression vector sequence coding for an SRC-1 mutant, wherein said mutant comprises the SRA-binding domain of SRC-1 but lacks the ability to transactivate a heterologous promoter, wherein said DNA sequence is in frame with DNA coding for the GAL4 DNA binding domain and a reporter and wherein expression and translation of this second construct produces a fusion protein comprised of the SRC- 1 mutant and the GAL4 DNA binding domain and wherein said fusion protein is capable of binding to a GAL4-binding DNA element; (c) transfecting cells with the vector constructed in step (a) and the vector constructed in step (b); (d) transcribing and translating the vector constructed in step (b); (e) inducing the vector constructed in step (a) with an appropriate inducing agent to promote transcription of the RNA transcript; and (f) measuring the reporter message.

In a specific embodiment the further comprises the step of contacting the transfected cells in step (c) with sufficient cycloheximide to substantially reduce or eliminate de novo protein synthesis. An additional embodiment is a method for determining the presence of a specific RNA transcript. This method comprises the steps of (a) incubating a cell extract independently with each of the following: (1) no additional agents, (2) RNase Tl at a concentration sufficient to substantially reduce or eliminate single-stranded RNA, (3) RNase H at a concentration sufficient to substantially cleave RNA in RNA.DNA hybrids, (4) multiple concentrations of deoxynucleotides antisense to said

RNA transcript followed by RNase H at a concentration sufficient to substantially cleave RNA in RNA.DNA hybrids, (5) deoxyoligonucleotides sense to such RNA transcript followed by RNase H at a concentration sufficient to substantially cleave RNA in RNA.DNA hybrids; (b) isolating total RNA from each of said extracts in step (a); (c) generating cDNA from each of said total RNA; (d) amplifying selected cDNA products using RT-PCR with PCR primers specific to a region of said RNA transcript that contains the region which hybridizes to said antisense deoxynucleotides; and (e) measuring the selected amplified PCR products.

In specific embodiments the amplified PCR products are detected by size fractionation.

Another embodiment of the invention is the provision of a kit for the detection of SRA transcripts. The kit includes 1) RNase Tl; 2) RNase H; 3) antisense deoxyoligonucleotides to SRA; 4) sense deoxyoligonucleotides to SRA; and 5) sense and antisense primers to SRA for RT-PCR amplification.

In a specific embodiment the method for modulating type I or "classical" nuclear receptor activation of target genes comprising the step of increasing or decreasing the availability of SRA, wherein increasing SRA results in increased nuclear receptor activity and decreasing SRA results in decreased nuclear receptor activity.

Another embodiment includes a method for the treatment of hormone- dependent tumors comprising the step of selective reduction of RNA coactivator availability.

In specific embodiments of the method the RNA coactivators are selected from the group of sequences consisting of SRA- 1, SRA-2 and SRA-3.

In other embodiments of the method the reduction is by: introduction of deoxyoligonucleotides complementary to the RNA coactivator, or transfection of cells of said tumor with a vector encoding an RNA transcript which is complementary to the RNA coactivator, or contacting of said tumor cells with a chemical agent which inhibits the transcription of RNA coactivators, or contacting of said tumor cells with a chemical agent which promotes the degradation of RNA coactivators, or introduction of an SRA dominant negative. Another embodiment of the invention is a method for the creation of a non- human animal wherein SRA is overexpressed for the purpose of: increasing responsiveness of a hormone receptor to a hormone, or increasing responsiveness of a hormone receptor to a hormone.

Another embodiment is a method for the detection of hormone-dependent tumors comprising the measurement of RNA coactivator in a tissue sample, wherein an elevated level of SRA or an isoform thereof is predictive of a hormone-dependent tumor.

Another embodiment is a method for the prenatal or postnatal detection of cortisol resistance comprising the step of measuring RNA coactivator in a tissue sample, wherein decreased RNA coactivator expression is predictive of cortisol resistance. Afurther embodiment is a kit for the prenatal or postnatal detection of cortisol resistance comprising the following elements: a) RT-PCR primers for SRA of SEQ. ID. No. 6 and SEQ. ID. No. 7; b) normal tissue sample standard; and c) confirmed cortisol resistance tissue sample standard.

Another embodiment is a method for the evaluation of an agent for the ability to modulate RN coactivator levels, comprising contacting a cell with said agent and subsequently measuring RNA coactivator transcript levels.

Another embodiment is a method for the evaluation of an agent for the ability to affect the half-life of SRA transcript, comprising contacting said transcript cell with said agent and subsequently measuring SRA. Another embodiment is a method for the selective isolation of specific RNA transcripts expressed in an expression vector comprising the steps of: (a) inserting the cDNA coding for the desired RNA transcript into a vector in any reading frame with an SRA insert such that expression of said vector produces an RNA transcript comprising said desired RNA transcript and SRA in series, (b) transfecting cells with said vector; (c) harvesting total RNA from said cells; (d) incubating said total RNA with SRC-1 protein; and (e) contacting said total RN A/SRC- 1 protein mixture with conjugated antibodies to SRC- 1 such that the RN A SRC- 1 /antibody complex is bound to a substance which facilitates isolation of said complex.

In a specific embodiment the RNA of interest and SRA can be cleaved by RNase H by incorporation of a specific nucleotide sequence into said vector such that expression of said vector produces an RNA transcript comprising said desired RNA transcript, said specific nuleotide sequence, and SRA in series.

Another embodiment is a kit for the expression and isolation of a specific RNA transcript comprising the following components: (a) a vector comprising a restriction site for the insertion of a selected cDNA such that expression of said vector produces a single R A transcript comprising said desired RNA transcript and SRA in series; (b) an SRC family protein; and (c) conjugated antibody to SRC-1.

Another embodiment is a method of drug design whereby the SRA/SRC-1 interaction is used as a model to evaluate agents for the ability to modulate hormone receptor activity.

Another embodiment is a method of designing a therapeutic agent wherein said therapeutic agent alters activation of a promoter, said method comprising: selectively mutating an RNA transcript; contacting said RNA transcript with a mixture comprising a chimeric protein consisting of a coactivator protein fused with the GAL4 DNA binding domain and a reporter construct; and measuring the product of said reporter construct. Another embodiment of the present invention is a cell line which lacks endogenous production of wildtype SRA.

Another embodiment of the present invention is an animal which has been genetically engineered such that said animal lacks the ability to produce wildtype SRA.

Another embodiment of the present invention is a method of designing a therapeutic agent wherein said therapeutic agent alters coactivation of a steroid hormone receptor, said method comprising: selectively mutating wildtype SRA; introducing a reporter gene construct responsive to said steroid hormone receptor into a cell lacking the ability to produce wildtype SRA; contacting one fraction containing said cell with said SRA mutant transcript; contacting a second fraction containing said cell with SRA wildtype transcript; contacting said cell fractions with a specific steroid receptor ligand; and measuring the production of product from said reporter construct, wherein production of said reporter construct product over basal values indicates coactivation and where said coactivation by SRA mutant can be compared to that of SRA wildtype.

Another embodiment of the present invention is a method of designing a therapeutic agent wherein said therapeutic agent reverses interference by transcriptional activators with common coregulators, said method comprising: selectively mutating an RNA transcript; introducing said RNA transcript into a cell in which a first steroid receptor, a second steroid receptor and a responsive reporter gene construct have been transfected; contacting said transfected cell with receptor specific ligands; and, measuring the product of said reporter construct. Another embodiment of the present invention is a method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to interact with a protein target, said method comprising: selectively mutating and transcribing said wildtype SRA; contacting said protein target with labeled wildtype SRA; washing away unbound labeled wildtype SRA; contacting one fraction containing said SRA-bound protein target with unlabeled said SRA mutant transcript; contacting a second fraction containing said SRA-bound protein target with unlabeled wildtype SRA; and measuring displaced labeled wildtype SRA in each fraction, wherein a greater displacement of labeled SRA wildtype by said SRA mutant compared with unlabeled SRA wildtype indicates a greater binding affinity to said protein target.

Another embodiment of the present invention is a method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to interact with a protein target, said method comprising: selectively mutating wildtype SRA (SRA mutant) and generating labeled transcript; contacting one fraction containing a target protein with said SRA mutant; contacting a second fraction containing said target protein with labeled SRA wildtype; isolating said RNA/target protein complexes; and measuring RNA transcript label in each of said complexes, wherein an increase in retained label in mutant SRA complex compared to wildtype SRA complex indicates a greater binding affinity to said protein target. Another embodiment of the present invention is a method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to coactivate individual steroid hormone receptors, said method comprising: selectively mutating wildtype SRA (mutant SRA); introducing a responsive reporter construct for steroid a hormone receptor into a group of cells; contacting one fraction containing said cells with said mutant SRA; contacting a second fraction containing said cells with wildtype SRA; contacting said cells with a ligand specific for said steroid hormone receptor; and measuring the production of product from said reporter construct, wherein production of said reporter construct product over basal values indicates coactivation and where said coactivation by SRA mutant can be compared to that of SRA wildtype.

Another embodiment is a method for the expression of a protein wherein the RNA transcript acts as a coactivator, said method comprising the steps of: inserting a gene of interest into a vector, wherein expression of said vector results in an RNA transcript comprising SRA, an internal ribosomal entry site, and the mRNA for said protein in series; and inducing expression of said vector.

Another embodiment is a method to diagnose hormone-dependent tumors wherein RNA coactivator is measured for elevated levels in a tumor tissue sample.

A further embodiment is a method for the detection of expression patterns of SRA levels wherein said patterns are predictive of expression patterns of steroid receptors.

Other and further objects, features and advantages would be apparent and eventually more readily understood by reading the following specification and by reference to the company drawing forming a part thereof, or any examples of the presently preferred embodiments of the invention are given for the purpose of the disclosure.

Brief Description of the Drawings Figure 1. Characterization of SRA Genes

(A) Structure of three SRA isoforms (I - III) deduced from screening of different cDNA and genomic DNA libraries from human and mouse. The sequences are identical in a 'core' region of 687 bp (no shadow), but are divergent in their 5' and 3' sequences (distinct shadings). The vertical lines indicate the location of the proposed termination codon of the putative open reading frame ORF1.

(B) Northern analysis of human SRA gene expression. (Left panel) Multiple tissue Northern blot, containing 2μg of human poly(A) RNA from each of the tissues indicated at top, was hybridized with a cDNA probe corresponding to the core sequence of human SRA (Figure 1 A). A tissue specific expression pattern with predominant transcripts of about 0.7-0.85kb (double arrows) and less abundant transcripts of 1.3-1.5kb is apparent. The blot was stripped and re- probed with β-actin to correct for RNA loading (bottom).

(Right panel) Northern analysis of human tissue-culture RNA probed with the longest cD A sequence of SRA (Figure IN isoform III) indicates a cell line specific expression of SRA isoforms. MCF-7 and T-47D cells have significantly higher levels of the smaller SRA-transcript (open arrow) compared to other tissues. Total RΝA was isolated from different human cell lines and 15μg electrophoresed and blotted.

The membrane was subsequently hybridized with β-actin probe as an internal control for loading (bottom). Size markers are indicated on the right (kb). Abbreviations: A549, lung carcinoma; HeLa, epithelioid cervix; HepG2, hepatoblastoma; LNCaP, metastatic prostate adenocarcinoma; MCF-7, breast adenocarcinoma; T-47D, breast ductal carcinoma; 293, transformed primary embryonal kidney. Figure 2. Primary Nucleotide Sequence Alignment of SRA Isoform I cDNA from Human and Mouse

Primary nucleotide sequence alignment of SRA Isoform I cDNA from human and mouse.

The nucleotides of mouse cDNA are indicated where they differ from the human cDNA sequence. Brackets [ ] represent the boundaries of the SRA core sequence. Arrows illustrate the location and orientation of the primer set used for SRA-specific

RT-PCR (shown in Figures 6 and 7). The Kozak consensus sequence is marked in bold; Circles indicate the putative translation initiation codons (ATG) targeted for mutation analysis (Figure 4); a consensus polyadenylation signal (AAT AAA) is boxed. Figure 3. SRA is a Steroid Hormone Receptor-Specific Coactivator (A) SRA enhances transcription mediated by steroid receptors.

Thin-layer chromatograph showing the chloramphenicol-acetyltransferase (CAT) reporter activity as response to steroid receptor-mediated transactivation in the presence or absence of SRA. HeLa cells were transiently transfected with plasmids encoding the human receptors for progesterone (PR), glucocorticoid (GR), androgen (AR) and estrogen (ER) and their cognate hormone response element coupled to a

TATA-CAT reporter gene along with SRA (+) or empty vector (-) and induced with their appropriate ligands or the PR antagonist RU486.

(B) SRA enhances transcription via the N-terminal activation domain AF-1 of the receptors. (upper panel) PR deletion mutants lacking either the ligand binding domain (PR-

ΔLBD: lanes 12-15) or the amino terminal activation function (PR-ΔAF-1 : lanes 16- 19) were assayed as in (A). Similar experiments with rat GR or domains thereof excluded the DNA binding domain (DBD) as a possible target for SRA (lower panel). The GR truncations were fused to the activation domain of GAL4 to monitor transactivation. SRC-1 was used as an assay control. MMTV-driven luciferase gene expression is shown as relative light units (RLU). (C) Multiple SRA isoforms coactivate endogenous PR.

Recombinant SRA sequences from different sources were transiently expressed in T- 47D breast carcinoma cells and assayed for transcription of cotransfected MMTV- luciferase reporter in the presence of 50nM PR agonist R5020. Fold transactivation mediated by endogenous PR is indicated in relation to PR-transcription in the absence of SRA (empty expression vector: v). SRA-sequences in CMV-driven mammalian expression vector are indicated: Core, core-region common to all SRA isoforms; inv, SRA expressed in 3*-5* orientation; B9, 3'-RACE product; C21/C10/C13/C5, SRA clones of different isoforms isolated from a skeletal muscle cDNA library; E6, SRA isoform I isolated from heart muscle cDNA library; F5, partial human genomic clone; HI 0, mouse cDNA clone. Fold activity is given as the mean (±SD) of triplicate values.

(D) Over expression of SRA reverses squelching of PR by ER in a dose-dependent manner. HeLa cells were transfected with PR and ER expression plasmids (50ng), MMTV-Luciferase reporter (2.5μg) and different amounts of SRA (0-4μg), and supplemented with 50nM receptor-specific ligands R5020 or E2, or both, where appropriate. Luciferase activities were determined and plotted as relative light units

(RLU) per μg of protein assayed as the mean (±SD) of triplicate values. Figure 4. Mutated SRA Constructs Enhance PR Transactivation (A) Schematic presentation of SRA mutants. Top: SRA core cDNA and deduced ORF-map; selected restriction sites and presumptive termination codon for ORFl (TAA) are indicated; "Y", location of peptide sequence used to generate SRA-mAb; vertical lines in ORF-map, stop codons; red lines, putative initiation codons. Lower: SRA mutants used in the transactivation studies presented and described in (B) and (C). Triple numbers indicate total of stop codons in each ORF; asterisk, point mutation(s); open arrows, translation initiation region of the tymidine kinase promoter (tk); colors indicate reading frames of presumptive translation products: light brown, unconstrained; yellow, ORFl (recognized by mAb); green, ORF2; blue, ORF3.

(B) Immunodetection and coactivation of transfected SRA-mutants revealed that SRA coactivates PR-transactivation in an ORF-independent manner. SRA mutants along with MMTV-Luc reporter and PR expression plasmid were transfected into COS cells and analyzed for: immunoreactivity to SRA antibody raised against a peptide sequence deduced from the C-terminus of ORF 1 (left panel); and for coactivation (right panel).

Mutants are as follows: (i) Truncation at the intrinsic Kozak sequence (ΔATG), and (ii) fusion with translation initiation region of the tymidine kinase promoter (tk) in two different open reading frames (tk-ORFl and tk-ORF2). The constructs are illustrated in (C). Protein size markers are indicated on the left. Fold coactivation in relation to PR-transcription as the mean (±SD) of triplicate values is indicated on the right.

(C) Enhancement of PR-mediated transactivation of an MMTN-Luc reporter by various mutated SRA constructs shown schematically in (A). Fold coactivation is indicated relative to expression of empty vector (v) and shown as the mean (±SD) of triplicate values. The mutations are described in Experimental Procedures. Abbreviations: SRN wild type SRA; SRA inv, cDΝA of SRA expressed in 3'-5' orientation; ORFl, ORF2, ORF3, nonsense mutations at the BamHI site obliterating the Kozak sequence of two reading frames and permitting only one putative translation product; Ylle, mutant ORF2 with an additional point mutation altering an ATG and generating a Mfel site; B, frame-shift mutation at the Bbsl site of SRA; S, frame-shift mutation at the SgrAI site; MS, frame-shift mutations at the Mfel and

SgrAI sites; YMS, mutant MS with additional frame shift mutation at the BamHI site; YMBS; mutant ORF2 with additional frame shift mutations at Mfel, Bbsl and SgrAI; ORF1ΔB/ ORF2ΔB/ORF3ΔB, 3' deletion at Bbsl of the mutants ORFl, ORF2, and

ORF3, respectively.

Figure 5. SRA is an RNA Coactivator

Two separate groups of HeLa cells were transiently transfected with reporter MMTN- Luc (2 μg) along with different amounts of CMV-driven expression plasmids for SRA

(3, 2, 1, 0.5μg), SRC-1 and CBP (3, 2, lμg), or empty vector (v, 3μg), and treated with EtOH control (-) or dexamethasone. One set of transfected cells was assayed for luciferase protein expression (upper panel), the other set of cells was incubated in medium containing 50μM cycloheximide and subjected to RΝA isolation, DΝasel digestion, followed by Northern analysis for luciferase RNA expression (lower panel) .

The Northern blot was hybridized with probes specific for cyclophilin (F) and luciferase RNA (Luc). Lane numbers (bottom) are common to both assays. 10,000 fold dilution of R A from cotransfected CMN-luciferase plasmid expression is shown as control in the Northern analysis (lane 17). A longer exposure of the blot revealed low levels of luciferase transcripts in all samples that were treated with dexamethasone. Figure 6. SRA is Present in Multiprotein Complexes

(A) Specific SRA RT-PCR assay.

Agarose gel analysis of SRA-specific RT-PCR products. T-47D cell extract ( ~20μg) was preincubated with different concentrations of amixture of SRA-specific antisense deoxyoligonucleotides (lanes 4-6, corresponding to lμg, 10ng, O.lng oligonucleotides) or with a sense oligonucleotide (lane 7, lμg) and subsequently digested with RNase H (100U). Untreated lysate (lane 1) and RNase Tl digested lysate (lane 2) were used as controls. RNA was isolated and subjected to SRA-specific RT-PCR (see Figure 2 for primers). RT-PCR controls (C) of assay buffer without cell extract (lane 8) and SRA-cDNA as template (lane 9) are indicated.

(B) Co-purification of SRA and SRC-1 complexes by gel filtration chromatography T-47D lysate (-400 μg) was fractionated on a Superose 6 column and the fractions subsequently analyzed by immunoblot with SRC-1 -specific antibody (left panels) and by SRA-specific RT-PCR (right panels). Fractionation was carried out either in the absence (-, top panels) or in the presence of preincubation of cell lysate with SRC-1 antibody (aSRC-1 Ab, lower panels). Numbers indicate fractions; elution peaks of molecular size markers are given for mammalian SWI/SNF complex (~2 MDa) and thyroglobulin (670 kDa). The void volume (4 MDa for globular proteins) was determined at fraction 20 by silver staining . SRC-1 commonly appears as doublet in Western analysis. The lower band of 120 kD in the lower left panel is most likely due to proteolytic degradation.

(C) Co-immunoprecipitation of SRA with AR or SRC-1 in Xenopus oocyte extracts. Ethidiumbromide-stained agarose gel of SRA-specific RT-PCR products from cell lysate and immunoprecipitates (top) and SDS-PAGE of precipitated proteins

(bottom). In vitro transcribed RNAs for SRA, AR and SRC-1 were injected along with L- 35 S-methionine into Xenopus laevis oocyte, and the translation products targeted for immunoprecipitationby monoclonal antibodies, as indicated. Lanes 1 and

6 represent assay controls of 100 fold diluted input control (lane 1) and PCR plasmid control (laneό).

Figure 7. Coexpression of SRA, PR, and GR in brain tissue Photomicrographs of in situ hybridization analysis illustrating RNA expression in adjacent coronal sections taken from brains of adult 129/SvEvBrd male mice and hybridized with antisense riboprobes for either SRN PR, or GR. Top: Olfactory bulb, Middle: Hippocampus, Bottom: Hypothalamus. The expression of SRA is shown in white, PR in yellow and GR in red. Hybridizations with the sense probes indicated low background and are not shown.

Abbreviations: 3V, 3rd ventricle; AO, anterior olfactory nucleus; ArcLP, arcuate hypothalamic nucleus, lateropost; ArcMP, arcuate hypothalamic nucleus, mediopost; CN cornu ammonis; DG, dentate gyrus; dlo, dorsolateral olfactory tract; Gl, glomerular layer of olfactory bulb GrO, granule layer of olfactory bulb; lo, lateral olfactory tract. Scale bar applies to all figures and corresponds to 500μm. Figure 8. The Sequence for the Exons and Isoforms of Human SRA SRA sequence in Homo Sapiens chromosome 5, BAC clone 319C17 is shown.

Sequences were found experimentally or predicted based on computer-aided retrotranscription of EST clones.

Figure 9. A Three-Dimensional Illustration of a Multiple Tissue Expression Array The Multiple Tissue Array was (Clontech) hybridized with a random-primed cDNA probe for human SRA at 68 °C in ExpressHyb hybridization solution (Clontech). Variation exists in the hybridization signals between different tissues. X-ray film was exposed for 1 day, scanned and computed using NIH Image 1.62. The quantitative Northern blot was generated with 53-780 ng of poly A(+) RNA from different human tissues and cancer cell lines (columns 1-11) and controls (column 12) that have been normalized to the expression levels of 8 different "housekeeping genes". The poly A(+) RNA was spotted to charged nylon membrane and fixed by UV irradiation. Selected tissue that expresses relatively high levels of SRA (adrenal gland, pituitary gland) or low levels of SRA (mammary gland, ovary, uterus) are indicated. Detailed specification of all tissues which were tested and an average of the quantification of four differently exposed X-ray films are shown in Figures 16-2 IB. Figure 10. SRA Overexpression in Steroid-Dependent Tumors Northern Territory-Human Tumor Panel Blot IV (Invitrogen) hybridized with a random-primed cDNA probe encompassing human SRA2 (hSRA) at 68 °C in ExpressHyb hybridization solution (Clontech). X-ray film was exposed for 1 day.

Approximately 20 μg of total RNA isolated from four different human tumor and normal tissues were run on a 1% denaturing agarose/formaldehyde gel, vacuum blotted to a positively charged nylon membrane and fixed by UN irradiation and baking. Single strand R A size markers are indicated on the left margin. Tumor tissue (t), normal tissue (n), and total RΝA (tRΝA) are indicated. The bottom picture is a polaroid picture of an ethidium bromide-stained gel after electrophoresis. Lane 1 is tissue from a breast tumor of a 51 year-old female with invasive ductal carcinoma of the breast. Lane 2 is tissue from a normal breast. Lane 3 is tissue of a uterine tumor from a 55 year-old female with well differentiated adenocarcinoma of the endometrium. Lane 4 is normal uterine tissue. Lane 5 is tissue of a fallopian tube tumor from a 46 year-old female with adenocarcinoma of the fallopian tube. Lane 6 is normal fallopian tube tissue. Lane 7 is ovarian tumor tissue from a 67 year-old female with mucinous cystandenocarcinoma of the ovary. Lane 8 is tissue from a normal ovary.

Figure 11. A Computational Three-Dimensional Illustration of the SRA Overexpression in Steroid-Dependent Tumors The blot shown in Figure 10 was used to generate a three-dimensional illustration utilizing ΝIH Image 1.62.

Figure 12. Quantification of Relative Expression of SRA in Tumor Tissues Compared to Normal Tissue (=100) Calculations using NIH Image 1.62 were based on the scanned X-ray image shown in Figure 10.

Figure 13. SRA Overexpression in Many Ovarian Tumors Human ovary tumor multi-sample total RNA northern blot (BioChain) hybridized with a randomly primed cDNA probe of human SRA2 (hSRA) for 17 hours at 42 °C in SSC Denhardt solution. X-ray film was exposed for 20 hours. Approximately 20 μg of total RNA isolated from four different donor's ovarian tumor and normal tissues were run on a 1% denaturing agarose/formaldehyde gel, transferred to a charged- modified nylon membrane and fixed by UV cross-linking and baking. Single strand RN A size markers are indicated on the left margin. Tumor tissue (t) and normal tissue

(n) are indicated. Lane 1 is tissue from an ovarian tumor characterized as poorly differentiated malignant mesodermal mixed tumor from a 40 year-old female. Lanes

2, 4, 6, and 8 are normal ovarian tissue. Lane 3 is tissue from the ovary of a 71 year- old female, said tumor characterized as moderately-poorly differentiated transitional cell carcinoma. Lane 5 is poorly differentiated transitional cell carcinoma from an ovary tumor of a 50 year-old female. Lane 7 is borderline papillary serous cystadenoma from an ovarian tumor of a 37 year-old female.

Figure 14. A Computational Three-Dimensional Illustration of the SRA Overexpression in Steroid-Dependent Tumors

The blot shown in Figure 13 was used to generate a three-dimensional illustration utilizing NTH Image 1.62.

Figure 15. A Quantification of Relative Expression of SRA in Tumor Tissues

Compared to Normal Tissue (=100) Calculations using NIH Image 1.62 were based on the scanned X-ray image shown in Figure 13.

The drawings and figures are not necessarily to scale and certain features mentioned may be exaggerated in scale or shown in schematic form in the interest of clarity and conciseness.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS It is readily apparent to one skilled in the art that various substitutions and modifications may be made to the invention disclosed in this Application without departing from the scope and spirit of the invention. As used herein the terms "SRA" or "steroid receptor RNA activator" refer to an RNA transcript which acts as a steroid hormone co-activator. As used herein the term "RNA transcript" refers to an RNA product of DNA transcription.

As used herein the word "transactivate" refers to the ability to increase transcription of a gene or DNA construct. As used herein, the term "core sequence refers to sequence produced by sequentially linking exon II, exon III, exon IV and exon V.

As used herein the sequence "SRA-1" refers to the sequence made by sequentially linking the sequences in SEQ. ID. No. 1 at locations 36358 through 36381, the core sequence and sequences in SEQ. ID. No. 1 at locations 43146 through 43306.

As used herein sequence "SRA-2" refers to the sequence made by sequentially linking the sequences in SEQ. ID. No. 1 at locations 36414 through 36536, the core sequence and sequences in SEQ. ID. No. 1 at locations 43330 through 43559.

As used herein the term sequence "SRA-3" refers to the sequence made by sequentially linking the core sequence and sequences in SEQ. ID. No. 1 at locations

43546 through 43695.

As used herein the term "exon II" refers to the sequences in SEQ. ID. No. 1 at locations 36537 through 36662.

As used herein the term "exon IIT refers to sequences in SEQ. ID. No. 1 at locations 41686 through 41827.

As used herein the term "exon IV" refers to sequences in SEQ. ID. No. 1 at locations 42640 through 42749.

As used herein the term "exon V" refers to sequences in SEQ. ID. No. 1 at locations 42889 through 43145. Three isoforms of SRA have thus far been reported (SRA-1, SRA-2 and

SRA-3). The highly conserved core sequence appears to be necessary and sufficient for coactivator activity, while highly variable 5' and 3' regions which flank the core sequence appear to confer specificity.

The invention provides an RNA transcript (SRA) and multiple isoforms thereof which act as coactivators for the steroid hormone receptors. The term SRA is herein intended to include, independently or as a group, the SRA 'core' sequence, isolated SRA isoforms from human or animal, and predicted SRA isoforms from human or animal. Human isoforms I- VI and the 'core' sequence as well as mouse, rat, and goat sequences are presented in Figure 8.

The ability of R A molecules to perform many functions that were commonly attributed to proteins has been well documented. RNA molecules perform enzymatic reactions such as trans-esterification (Jaeger, 1997) or catalysis of peptide bond formation (Zhang and Cech, 1997). The unusual stability of globin mRNA has been shown to play a role in the determination of its expression (Liebhaber, 1997). Similarly, the stable RNA stem-loop structure at the 5' end of nascent transcripts is a c/s-acting transactivation response element for the HIV-Tat activator (Jones and

Peterlin, 1994), demonstrating that gene expression can be regulated in a post- transcriptional manner by the structure and the stability of primary RNA transcripts. Noncoding RNAmolecules and 3'-untranslated regions of mRNAs have been reported to function as trαns-acting regulators via antisense RNA-RNA interactions in C. elegans and in plants (Lee et al, 1993; Crespi et al, 1994). Recently, Sit et al. reported the tπ s-activation of viral RNA transcription by the interaction of two genomic-sense RNAs (Sit et al, 1998). However, SRA is unique among eukaryotic transcriptional coactivators in its ability to function as an RNA transcript to specifically regulate the activity of a family of transcriptional activators. In the method of detecting SRA or RNA transcripts capable of transactivating a gene or DNA, an RNA transcript is transcribed from a vector also coding for SRA (or a portion thereof) such that a single RNA transcript results with RNA transcript of interest in series with SRA. The SRA portion of the transcript binds to a mutant SRC-1 protein which possesses an inactive transactivation domain and is fused to the DNA-binding domain of GALA A reporter construct consisting essentially of UAS- reporter indicates the ability of RNA transcript of interest to activate a promoter. The reporter can be a measurable RNA message or protein or can be a protein which confers selectivity to the cell.

The method of verifying the identity of an RNA transcript comprises incubating total RNA with specific oligonucleotides prior to incubating the total RNA with RNase Tl or RNase H. Subsequently, total RNA is converted to cDNA and amplified with RT-PCR. The RT-PCR products are then size fractionated and the presence or absence of specific sized products for the various incubation conditions determines if a specific RNA transcript was present.

The prior art provides methods for the verification of specific RNA transcripts. For instance, Northern hybridizations are routinely used to identify messages based on size and affinity for specific single-stranded probes. This process typically involves the steps of size fractionating the RNA (usually in an agarose gel), transferring the RNA to a membrane, hybridizing the R A to a specific labeled probe, and evaluation by autoradiography. Alternatively, RNA transcripts can be amplified by RT-PCR with the resulting products used as probes on a Northern blot or the products transferred to a membrane and analyzed by Southern analysis. Additionally, there are several variations of the above methods which rely essentially on the same criteria of size fractionation and hybridization.

The above prior art methods are subject to a number of possible false identifications. For instance, the above methods may not easily distinguish between similar isoforms when the isoforms are of a similar message size or when RT-PCR produces dissimilar products of a similar size. Additionally, small regions of divergence in a message are unlikely to significantly alter the ability of a message to hybridize to a probe.

In order to verify the identity of a message with certainty, the message must be sequenced. Depending on the method of sequencing and specific resources available, sequencing may involve screening for positive clones, subcloning, isolating

PCR products etc. These processes can be quite time consuming and are therefore not ideal for a number of applications such as for the rapid screening of a particular message.

The present invention provides significant advantages over the prior art by providing verification of a particular message at the step of size fractionation. Such a feature saves considerable time and expense related to transferring RNA to a membrane, hybridizing to a labeled probe, and visualizing by any of several methods. In addition, the present invention allows for the evaluation of a message based on several regions of interest and has a built in control to prevent false positives due to genomic DNA contamination. RT-PCR alone is also capable of providing information regarding the identity of a message at the point of size fractionation when multiple primers are used. In this situation, primers are selected based on a number of strategies including selecting primers which hybridize to highly divergent regions of a message or regions which are known to possess a mutation. However, unlike the present invention which results in the simple presence or absence of PCR product, the

RT-PCR alone method often involves the interpretation of multiple product bands which lends itself to the possibility of misinterpretation. Furthermore, when the selected PCR primers do not span an intron, it is difficult to impossible to rule out the possibility of genomic DNA contamination. Detection of RNA/Protein interaction is based on the ability of SRAs to transactivate a heterologous promoter. In this method, an interaction between an R A transcript (RNA transcript of interest) and a protein (protein of interest) is determined by generating an RNA transcript containing RNA transcript of interest in series with SRA, generating a chimeric protein containing protein of interest fused to the GAL4 DBD, and measuring activation of a reporter construct. Interactions between RNA transcript of interest and protein of interest bring the SRA/RNA transcript of interest/SRC- 1 complex in proximity to the reporter construct promoter and drives transcription. A suitable reporter can be an measurable RNA message or protein or can be a protein which confers selectivity to the cell.

In the method of selective isolation of specific RNA transcript, an RNA of interest is inserted into a vector which codes for SRA (or a portion thereof) such that expression of the vector results in the production of RNA transcripts consisting of the

RNA of interest and SRA in series. Total RNA is isolated and the RNA transcript of interest/SRA transcript is separated through interaction with an SRA-binding protein. SRA binding proteins include, but are not limited to, SRC- 1 , SRC-2, and SRC-3. The SRA binding protein may be modified in such a way as to facilitate isolation. The RNA of interest can then be separated from SRA by adding an oligonucleotide antisense to an RNA sequence between the RNA of interest and SRA and then adding RNase H. RNase H then cleaves at the point of the oligonucleotide binding, thus freeing RNA of interest from SRA.

In the method of evaluation of type 1 or "classical" nuclear receptor activity, SRA coactivator transcript is measured as an indication of the activity of the nuclear receptors, where higher levels of SRA indicate greater activity. Such a method is valuable because it gives information as to the functional state of the receptor, not simply an indication of the presence or concentration of the receptor.

Type I or "classical" nuclear receptor activity can be modulated by the availability of SRA. In this method, SRA coactivator transcript levels are increased for the purpose of increasing receptor activity and decreased for the purpose of decreasing receptor activity. Such a method will result in the control of genes directly under the transcriptional control of type I nuclear receptors as well as some genes downstream. Thus, genes which are not directly regulated by type I nuclear receptors may also be modulated by this method.

The treatment of hormone-dependent tumors can be achieved by reducing SRA coactivator in the tumor. By "reducing" it is meant that the SRA concentration is reduced or the effective concentration is reduced. For instance, the effective concentration of SRA can be reduced by rendering the transcript non-functional by binding to a specific oligonucleotide or to a protein ("false substrate") which lacks the ability to enhance transcription. Furthermore, another RNA transcript which lacks the ability to coactivate can be introduced which binds to the SRA-binding site of a protein ("dominant negative"), thus preventing the interaction with native SRA. The above mentioned examples demonstrating how the effective concentration can be reduced are not intended to limit the invention; one skilled in the art will recognize that other methods and variations of the above methods can achieve essentially the same outcome. Reduction of the SRA in the hormone-dependent tumor cell will have the effect of reducing the responsiveness of that cell to steroid hormones. The degree to which SRA is reduced in a cell will determine the degree to which hormone responsiveness is reduced.

The method for the detection of hormone-dependent tumors is achieved by measuring SRA is a tissue sample. In this method, a tissue biopsy sample is evaluated for SRA levels as an indication of type I nuclear receptor activity. High levels of SRA are indicative of high type I nuclear receptor activity. High SRA levels are predictive for the presence of hormone-dependent tumors. Furthermore, high SRA levels indicate that a therapy in which SRA levels are reduced may be effective. Another method is the evaluation of an agent for the ability to modulate SRA levels. In this method, a cell is contacted with an putative SRA modulating agent and SRA levels are subsequently measured. This method further provides methods by which putative SRA modulating agents affect SRA transcript half-life in vitro.

In the method for the prenatal or postnatal detection of cortisol resistance, a tissue sample is evaluated for SRA levels, where significantly reduced SRA levels are predictive of cortisol resistance. Tissue samples can be obtained as a result of, but not limited to, amniocentesis, chorionic villi sampling, direct biopsy, or isolation of fetal cells in the mother's blood.

In the method of designing a therapeutic agent SRA is mutated and the functional result of such a mutation is evaluated. The object of the method is to identify SRA mutants which possess greater, similar, or lesser activity as coactivators to type I nuclear receptors. The invention provides several means of evaluating the functional result of the mutations.

A non-human animal genetically altered such that SRA is overexpressed is produced. One objective of this component of the invention is to increase the responsiveness of a hormone receptor to a hormone. Such an increased hormone receptor responsiveness will allow hormone-mediated responses to be potentiated without the requirement for high levels of added hormone. This feature of the invention is particularly attractive for animals which are intended for consumption or where a product of the animal is intended for consumption. In this method, the effect of adding hormones to the animals is achieved without passing the hormones on to the consumer. Since, RNA transcripts (such as SRA) are extremely labile, their consumption is expected to be without significant consequence. It is to be understood that genes which are directly responsive to type I hormone receptors can be modulated as well as genes which are indirectly modulated by the hormone receptors. The invention also provides a method for the detection of SRAs or other RNA transcripts which are capable of transactivating a gene. In this method, the DNA sequence which codes for an RNA transcript of interest is ligated into a first expression vector which contains 'linker' DNA and DNA encoding SRA. In one embodiment, the linker DNA is 1 to 1000 base pairs. In another embodiment, no linker DNA is used. In a preferred embodiment, the first expression vector is under the transcriptional control of an exogenous inducer. Such exogenous inducers include, but are not limited to, hormones and metal ions. The expression of the first expression vector results in a sense RNA transcript comprising RNA transcript of interest, linker RNN and SRA in series. In a preferred embodiment, the linker RNA and SRA are minimized to the smallest functional size in order to reduce the total message size. Any of a number of methods well known in the art will suffice for determining the smallest functional message size. In another preferred embodiment, multiple stop codons are placed throughout the inserted gene with special emphasis on the 5' region. In a second vector, DNA coding for a chimeric protein consisting essentially of an SRC mutant fused to the GAL4DBD. The SRC mutant may be generated from, but not limited by, SRC-1, SRC-2, and SRC-3. The SRC mutant possess a substantially reduced ability to activate a promoter while the ability to bind SRA is preserved.

Expression of the second vector results in the production of a chimeric protein (herein referred to as SRC_mu/GAL4DBD) which is capable of I) binding to the GAL4 DNA binding domain (herein referred to as UAS) and 2) binding to an RNA transcript containing SRA or a portion thereof, thus bringing the RNA transcript of interest/SRA containing RNA transcript in close proximity to UAS . The second vector also contains a reporter construct consisting essentially of one of more UAS sequences in close proximity to a promoter which controls expression of a reporter gene. The promoter may consist of, but is not limited to, a TATA box. The reporter gene may contain, but is not limited to, luciferase (LUC) and chloramphenicol acetyltransferase (CAT). The reporter gene may alternatively contain, but is not limited to, any gene which confers enhanced survivability to the host cell. In a preferred embodiment, the first expression vector and the second vector are derived from a group consisting of retroviral, adenoviral, and vaccinia viral vectors. One skilled in the art will recognize the inherent advantages of one viral vector over another for a given cell of interest.

Mammalian cells are transfected with the first expression vector and the second vector. Transcription of the second vector is induced with an appropriate inducer and an appropriate time is allowed for translation of SRC_mu/GAL4DBD into protein. In one embodiment (method A), cycloheximide is then added to the cells to substantially inhibit de novo protein synthesis. In another embodiment (method B), cycloheximide is omitted from the protocol. The first expression vector is then induced (methods A and B) with an appropriate inducer, resulting in transcription of the RNA transcript of interest/SRA containing message. The RNA transcript of interest/SRA containing message is then recruited to the reporter construct through the SRA/SRC protein interaction. RNA transcript of interest may act directly to activate the promoter or may act with one or more endogenous factors to activate the promoter. For method A, reporter message is measured as an indication of RNA transcript of interest-dependent promoter activation. For methodB, reporter construct product, activity, or effect is measured as an indication of RNA transcript of interest- dependent promoter activation.

The invention also provides a method for the verifying the identity of an RNA transcript of interest. In this method, cell extract containing the RNA of interest is preincubated with different concentrations of a mixture of RNA transcript of interest specific antisense deoxyoligonucleotides or with sense oligonucleotides and then subjected to digestion with RNase H. Control samples include untreated cell extract and cell extract treated with RNase Tl . Total RNA is then isolated and subjected to RT-PCR using primers specific for RNA transcript of interest. It is important that the selected primers bracket the regions of the cDNA generated from RNA transcript of interest to which the oligos bind. Size fractionation of the RT-PCR products is performed which reveals the presence or absence of amplified product. In this context, absence of product is intended to indicate no detectable product or product of substantially reduced concentration compared to a standard. Verification of RNA transcript of interest is made by RT-PCR product from the untreated cell extract sample, the RNase sample, the RNase plus sense oligos, and a control RNA transcript of interest sample while no product is detectable from the RNase T 1 sample, and assay buffer control. Importantly, a variable amount of RT-PCR product is detected from the RNase H plus antisense oligos samples, where the product varies from no detectable product to a detectable product for high to low concentrations of antisense oligos, respectively. In a preferred embodiment, RNA transcript of interest is SRA and corresponding sense and antisense oligos are used. In one embodiment, the amount of cell extract is about a range of 0.1 to 100 μg and sense/antisense oligos are about a range of 0.01 ng to 100 μg. In a preferred embodiment, the amount of cell extract is about 20 μg and sense/antisense oligos are about a range of 0.1 ng to 1 μg. In another embodiment, the amount of RNase H and RNase Tl used are 100 and 50 units, respectively. Size fractionation may be achieved by any of several methods well known in the art.

The invention further provides a kit for the verification of SRA based on the above method. Such a kit consists essentially of RNase H, RNase Tl, antisense deoxyoligonucleotides to SRA(5'-CTTACCGAGATGACCACGTTCCTTGATTC- 3 *) (SEQ . ID .No .4) , sense oligonucleotides to SRA ( 5 ' -

GCCGACGCCTGGCACTGCTGCAGG-3'XSEQ.ID.No. 5), (5' and antisense PCR primers for SRA (sense: 5'-CGCGGCTGGAACGACCCGCCGC-3\ SEQ.ID.No. 6; antisense: 5'- CAGACTCACCGGACACCATCTCCTA-3', SEQ.ID.No. 7), and reaction buffer. Isolation of total RNA, RT-PCR, and size fractionation of PCR products is left up to the user of the kit. It is to be understood that the above mentioned primers and oligos are not intended to limit the invention in any way. Other primers and oligos will be readily apparent to one of skill in the art. The invention also provides a method for the detection of RNA protein interactions between an RNA of interest and a protein of interest. In this method, the DNA sequence which codes for RNA transcript of interest is ligated into a first expression vector which contains 'linker' DNA and DNA encoding SRA. In one embodiment, the linker DNA is 1 to 1000 base pairs. In another embodiment, no linker DNA is used. In a prefeoed embodiment, the first expression vector is under the transcriptional control of an exogenous inducer. Such exogenous inducers include, but are not limited to, hormones and metal ions. The expression of the first expression vector results in a sense RNA transcript comprising RNA transcript of interest, linker RNA, and SRA in series. In a preferred embodiment, the linker RNA and SRA are minimized to the smallest functional size in order to reduce the total message size. Any of a number of methods well known in the art will suffice for determining the smallest functional message size. In another prefeoed embodiment, multiple stop codons are placed throughout the inserted gene with special emphasis on the 5' region. In a second vector, DNA is inserted coding for a chimeric protein consisting essentially of protein of interest fused to the GAL4DBD. In one embodiment, the cDN A coding for protein of interest, RNA transcript of interest, or both come from a cDNA library. In another embodiment, the interactions of a specific RNA and protein are evaluated. In addition, DNA is inserted which codes for the activation domain 2 of SRC- 1. In another embodiment, DNA coding for PRΔDBD is substituted for the activation domain 2 of SRC-1. Expression of the second vector results in the production of a chimeric protein (herein refeoed to as protein of interest/GAL4DBD) which is capable of binding to the GAL4 DNA binding domain (UAS). In addition, either the activation domain 2 of SRC-1 or PRΔDBD is produced. A reporter construct consisting essentially of one of more UAS sequences in close proximity to a promoter which controls expression of a reporter gene is also provided. The promoter may consist of, but is not limited to, a TATA box. The reporter gene may contain, but is not limited to, luciferase (LUC) and chloramphenicol acetyltransferase (CAT). The reporter gene may alternatively contain, but is not limited to, any gene which confers enhanced survivability to the host cell. In one embodiment, the reporter construct is contained within the first expression vector. In a preferred embodiment, the first expression vector and the second vector are derived from a group consisting of retroviral, adenoviral, and vaccinia viral vectors. One skilled in the art will recognize the inherent advantages of one viral vector over another for a given cell of interest.

Suitable mammalian cells are transfected with the first expression vector and the second vector. Mammalian cells may include, but are not limited to, HeLa cells,

COS cells, A549 cells, HepG2 cells, LNCaP cells, MCF-7 cells, t-47D cells, and 293 cells. Transcription is induced and sufficient time is allowed for translation of protein products. In one embodiment, the first expression vector is induced independently some time after the second vector has been induced. In yet another embodiment, cycloheximide is used to prevent translation of the first expression vector products into protein. The RNA transcript of interest/SRA containing message is then recruited to the reporter construct through the RNA transcript of interest/protein of interest interaction. SRA recruits either the activation domain 2 of SRC-1 or PRΔDBD. This SRA complex is then able to activate the promoter of the reporter construct. For the embodiment in which cycloheximide is used, reporter message is measured as an indication of RNA transcript of interest/protein of interest interaction. For the embodiment in which cycloheximide is not used, reporter construct product, activity, or effect is measured as an indication of RNA transcript of interest dependent promoter activation. A product is provided consisting essentially of the above mentioned the first expression vector. The first expression vector additionally contains features for selection common in the art. In addition, an RNA transcript which results from transcription of the first expression vector is provided. Such a transcript contains an RNA of interest in series with a linker sequence of variable length and SRA. In one embodiment, the DNA coding for the RNA of interest is obtained from a library. In another embodiment, multiple stop codons are placed within the transcript, with particular emphasis in the 5' region. In yet another embodiment, the length of the linker sequence and SRA sequence are minimized to the smallest functional size.

The invention further provides a method for the selective isolation of a specific RNA transcript of interest by expressing RNA transcript of interest in series with SRA. In this method, DNA coding for RNA transcript of interest is ligated into an expression vector which contains DNA coding for SRA, such that expression of the vector results in an RNA transcript containing essentially RNA transcript of interest and SRA in series. Total RNA is isolated by any of a number of methods well known in the art. The RNA transcript of interest/SRA transcript is then separated from the total RNA through the SRA interaction with SRC proteins. SRC proteins include, but are not limited to, SRC- 1 , SRC-2, and SRC-3. It is to be understood that any SRA- binding protein may be used for this purpose. In one embodiment, the SRA-binding protein is modified such as to facilitate separation. Modifications may include, but are not limited to, addition of an epitope recognizable by an antibody, incorporation into a fusion protein which allows conjugation such as to beads coated beads, or any of a number of methods well known in the art. In another embodiment, antibodies to SRC proteins are conjugated to a surface for the purpose of isolating the RNA/SRC complex. In one embodiment of this invention, a minimal portion of SRA is utilized which retains the ability to bind to SRC proteins. In a preferred embodiment, an oligonucleotide to a region of the RNA transcript between RNA transcript of interest and SRA is administered with RNase H in order to cleave RNA transcript of interest and SRA from each other. In a more prefeoed embodiment, RNA transcript of interest is cleaved from the RNA protein complex such that RNA transcript of interest can be easily isolated.

The invention provides a kit for the selective isolation of a specific RNA transcript of interest consisting essentially of I) an expression vector which allows for insertion of DNA coding for RNA transcript of interest such that expression of the vector produces RNA transcript of interest in series with SRA and 2) SRC protein which is amenable to isolation. SRC protein is made amenable to isolation by any of a number of means well known in the art including, but not limited to, expressing as a fusion protein. In another embodiment, specific antibodies to SRC proteins are provided which are amenable to conjugation to any of a number of substances including, but not limited to, beads, magnetic beads, and plastic. In one embodiment of the invention, the size of SRA is minimized while retaining significant binding to SRC proteins.

In another embodiment of the invention, an animal is produced which lacks the ability to produce wildtype SRA (herein referred to as an SRA knockout) . Techniques for the generation of a knockout animal are well known in the art. In one embodiment, homologous recombination results in the complete removal of the SRA gene. In another embodiment, homologous recombination results in the loss of the portion of the SRA gene containing the core sequence. Suitable animals include, but are not limited to, mouse and rat.

In a further embodiment of the invention, a cell line is generated from the above mentioned SRA knockout. Any of a number of procedures well known in the art may be used for the immortalization of cells. In a preferred embodiment, a cell line is generated by fusing an SRA knockout cell with a tumor cell, thus producing a hybridoma.

In yet another embodiment, methods are provided for the creation of a transgenic animal in which SRA is overexpressed. Techniques for the generation of transgenic animals are well known in the art. In one embodiment, multiple copies of SRA are introduced. In another embodiment, copies of SRA are introduced which are under the control of a constitutively active promoter. In still another embodiment, copies of SRA are introduced which are under the control of a promoter which confers tissue specific expression. It is to be understood that other methods of over expressing SRA in a transgenic animal may be used without altering the scope of the invention.

In still another embodiment of the invention, a non-human animal is provided which has been genetically altered such that SRA is overexpressed. Suitable animals for SRA Over expression include, but are not limited to, mice, rats, cats, dogs, goats, sheep, cows, and horses.

In another embodiment of the invention, SRA transcript is measured in a tissue sample as an indicator of steroid hormone receptor activity. In this embodiment, SRA transcript can be measured by any of several methods well known in the art. Non- limiting examples include I in situ hybridization, Northern hybridization, and semi- quantitative PCR. Higher levels of SRA in a sample are indicative of greater steroid hormone activity. In another embodiment of the invention, SRA levels are measured in a tissue sample as an indicator of the prevalence of steroid hormone receptors in the tissue. In still another embodiment of the invention, steroid hormone receptor activity is altered by modulating the availability of SRA. In this method, SRA transcript levels are increased to achieve greater steroid hormone receptor activity and decreased to achieve decreased activity. In a preferred method, SRA transcript availability is increased by transfection with an expression vector possessing SRA. In another preferred embodiment, SRA transcript availability is increased by stabilizing native

SRA. In yet another prefeoed embodiment, SRA transcript availability is increased by increasing SRA transcription from the native gene. SRA transcript availability is decreased by one or a combination of several means. In one embodiment, oligonucleotides antisense or complementary to SRA are introduced into a cell. Oligonucleotides may be introduced by any of several methods well known in the art including, but not limited to, endocytosis, encapsulation in a lipid vesicle, direct injection, and electroporation. In another embodiment, SRA antisense RNA is introduced into a cell. It is preferred that the antisense RNA be introduced via transfection with an expression vector. In yet another embodiment, an SRA mutant which binds to the SRC proteins but which does not act as a coactivator (herein referred to as an SRA dominant negative) is introduced into a cell. In still another embodiment, SRA availability is decreased by contacting the cell with an agent which decreases SRA transcription. In another embodiment, SRA availability is decreased by contacting the cell with an agent which promotes degradation of SRA.

In another embodiment of the invention, steroid-hormone dependent tumors are treated by decreasing steroid hormone receptor activity via reduced availability of SRA. In this method, SRA availability is decreased by one or a combination of several means. In one embodiment, oligonucleotides antisense or complementary to SRA are introduced into a tumor cell. Oligonucleotides may be introduced by any of several methods including, but not limited to, endocytosis, encapsulation in a lipid vesicle, and direct injection. In another embodiment, SRA antisense RNA is introduced into a tumor cell. It is preferred that the antisense RNA be introduced via transfection with an expression vector. In yet another embodiment, an SRA dominant negative is introduced into a tumor cell. In still another embodiment, SRA availability is decreased by contacting a tumor cell with an agent which decreases SRA transcription. In another embodiment, SRA availability is decreased by contacting a tumor cell with an agent which promotes degradation of SRA. This method may be used either alone or in combination with other treatment methods. In a preferred embodiment, estrogen-dependent hormone tumors are treated by this method. Yet another embodiment of the invention provides a method for detecting steroid hormone-dependent tumors. In this method, SRA levels are measured from a tissue sample. Above normal levels of SRA are predictive of a steroid hormone- dependent tumor. In a prefeoed embodiment, SRA levels are compared to those of a non-tumor tissue standard. SRA levels may be measured by any of a number of methods well known in the art including, but not limited to, in situ hybridization, semi- quantitative PCR, and Northern hybridization.

Still another embodiment of the invention provides a method for the evaluation of an agent for the ability to modulate SRA levels. In this method, a group of cells are contacted with the putative SRA modulating agent and SRA levels are subsequently measured. It is preferred that agents which are being evaluated for the ability to increase SRA levels be administered to cells which possess low to moderate levels of SRA. It is also preferred that agents which are being evaluated for the ability to decrease SRA levels be administered to cells which possess moderate to high levels of SRA. Suitable cells for this method may include, but are not limited to, fibroblasts, epithelial cells, blood cells, HeLa cells, COS-7 cells, T-47D cells, and MCF-7 cells. In another embodiment, SRA levels are evaluated by the ability to activate a reporter construct. In another embodiment of the invention, methods are provided by which the half-life of SRA can be evaluated following contact with an agent. In this method, parallel studies are performed where an agent is administered to one fraction possessing SRA transcript and a control solution is administered to a second fraction. At different time intervals, aliquots from each fraction are taken and assayed for SRA. In a prefeoed embodiment, aliquots are taken, RNA extracted, and the SRA is used to generate first strand cDNA. Subsequent amplification by PCR allows for semi- quantification and comparison between control and agent groups. In another embodiment, aliquots are taken and subjected immediately to first strand cDNA synthesis. Another embodiment of the invention provides a method for the prenatal or postnatal detection of cortisol resistance. In this method, a tissue sample from the fetus or person for which testing is needed is evaluated for SRA levels. Low or unmeasurable levels of SRA are predictive of cortisol resistance. It is preferred that the tested tissue sample be relatively easily accessible. Relatively easily accessible fetal cells include, but are not limited to, cells from amniocentesis, cells from chorionic villi sampling, and isolated fetal cells found in the mother's blood. Relatively easily accessible cells from a person following delivery include, but are not limited to, blood cells, epithelial cells, and fibroblasts. SRA levels may be measured by any of a number of methods well known in the art including, but not limited to, in situ hybridization, semi-quantitative PCR, and Northern hybridization. In a prefeoed embodiment, SRA levels are compared with normal and confirmed cortisol resistance standards. It is to be understood that this method may be used in conjunction with other methods in order to increase the reliability of the diagnosis. Still another embodiment of the invention provides a kit for the prenatal or postnatal detection of cortisol resistance. In one embodiment, the kit comprises PCR primers for semi-quantitation by RT-PCR, in addition to normal and confirmed cortisol resistance tissue samples. In a prefeoed embodiment, the tissue sample standards comprise immortalized cells of human origin. The invention also provides methods for designing a therapeutic agent where the therapeutic agent possesses the ability to modulate steroid hormone receptor activity. Methods consist essentially of mutating wildtype SRA and evaluating the functional result of such a mutation. 'Wildtype' SRA is herein intended to indicate SRA of a naturally occurring isoform or an isoform which is predicted based on SRA gene sequence. Mutations of SRA may include deletions, additions, or substitutions of nucleotides through any of a number of methods well known in the art. In one embodiment, mutations are made in SRA which correspond to one or more stem regions. In another embodiment, mutations are made in SRA which cooespond to one or more loop regions. In yet another embodiment, mutations are made which result in a shortening of one or more stem regions. In yet one more embodiment, mutations are made which result in a lengthening of one or more stem regions. In a further embodiment, mutations are made which result in a diminished loop size in one or more loop regions. In still another embodiment, mutations are made which result in an increased loop size in one or more loop regions. In another embodiment, mutations are made which result in a modification of the tertiary (3 -dimensional) structure of the folded transcript. One skilled in the art will also be able to predict changes in folding and structure for a given SRA mutant using any of several computer routines cuoently available.

In one embodiment of the invention, the ability of an SRA mutant is evaluated for the ability to activate a promoter. In this method, SRA is mutated and then contacted with an appropriate reporter construct. In a preferred embodiment, the mutant SRA is expressed from an expression vector which is used to transfect a cell.

The cell is cotransfected with a second vector which possesses the above reporter construct and DNA coding for a chimeric protein consisting of a coactivator protein fused to the GAL4 DBD. In a more preferred embodiment, the coactivator protein is selected from a group including, but not limited to, SRC-1, SRC-2, and SRC-3. In another preferred embodiment, a portion of the coactivator protein which interacts with SRA is fused with GAL4 DBD. Reporter construct product, product activity, or product effect is measured as an indication of promoter activation by the SRA mutant. In a preferred embodiment, activation of the same reporter construct by wildtype SRA is evaluated for comparison. In another embodiment of the invention, the ability of an SRA mutant is evaluated for the ability to alter coactivation of a steroid hormone receptor. In this method, an expression vector possessing a reporter construct and DNA coding for the SRA mutant is used to transfect one fraction of a group of cells which lack the ability to produce wildtype SRA. Such cells may include, but are not limited to, freshly isolated SRA knockout cells, cultured SRA knockout cells, and immortalized SRA knockout cells. A second fraction of the same group of cells is transfected with an expression vector possessing the same reporter construct and DNA coding for wildtype SRA. The reporter construct possesses a steroid receptor binding domain in close proximity to a promoter and reporter gene. Steroid receptor binding domains for any steroid receptor may be used, including but not limited to, AR, ER, GR, MR, and PR. Promoters may include, but are not limited to, TATA and TK promoters. Reporter genes may include, but are not limited to, LUC, CAT, Ampicillin, and

Neomycin. Cells are then contacted with an appropriate steroid hormone receptor ligand. The appropriate ligand is one which stimulates the steroid hormone for which the reporter construct is designed. Reporter construct product, product activity, or product effect is measured for SRA and SRA mutant and compared for the relative ability to coactivate the steroid hormone receptor.

In yet another embodiment of the invention, the ability of an SRA mutant to reverse interference by transcriptional activators with common coregulators (herein referred to as squelching) is evaluated. In this embodiment, an expression vector possessing a reporter construct and DNA coding for the SRA mutant is used to transfect one fraction of a group of cells. A second group of cells is transfected with essentially the same expression vector, except that it lacks SRA coding DNA. In another embodiment, the reporter construct and SRA mutant are on separate expression vectors such that the amount of SRA transcribed can be modulated by the amount of vector transfected. The reporter construct consists essentially of a steroid receptor binding domain in close proximity to a promoter and reporter gene. Steroid receptor binding domains for any steroid receptor may be used, including but not limited to, AR, ER, GR, MR, and PR. Promoters may include, but are not limited to, T AT A and TK promoters. Reporter genes may include, but are not limited to, LUC, CAT, Ampicillin, and Neomycin. In a prefeoed embodiment, the reporter construct possesses SRBD-TATA-LUC, where SRBD is defined as a steroid receptor binding domain. Each group of cells is cotransfected with an expression vector possessing DNA coding for two different steroid hormone receptors. In another embodiment, steroid hormone receptors are contained within separate expression vectors. The steroid hormone receptors may include, but are not limited to, AR, ER, GR, MR, and PR. In a preferred embodiment, Hela cells are transfected with PR and ER expression vectors (50ng), MMTV-Luciferase reporter (2.5 μg) and different amounts of SRA (0- 4μg), and supplemented with 50nM receptor-specific ligands R5020 or E2, or both, where appropriate. Luciferase activities are determined and plotted as relative light units (RLU) per μg of protein assayed. In another embodiment, the ability of SRA to reverse squelching is evaluated for relative comparison to that of SRA mutant.

In yet another embodiment of the invention, the ability of an SRA mutant is evaluated for the ability to interact with a protein target. In this embodiment, a protein target is bound with an excess of labeled SRA. Unbound SRA is washed away. It is prefeoed that the protein target is immobilized to a substrate. Immobilization may be achieved by any of a number of means well known in the art. Labeling of SRA may be achieved by any of a number of methods well known in the art, including but not limited to, incorporation of radioactive nucleotides and addition of fluorescent tags.

Variable amounts of unlabeled SRA mutant are added to the SRA/protein complex and the displacement of SRA is measured. In one embodiment, unlabeled SRA is added in place of SRA mutant in variable amounts for comparison. In another embodiment of the invention, labeled SRA mutant is first bound to the protein target and unlabeled SRA is used to compete for binding. Suitable protein targets include, but are not limited to, SRC family proteins such as SRC-1, SRC-2, and SRC-3 and mutants thereof. (T20}In another embodiment of the invention, the ability of SRA mutant to bind to a protein target is compared to that of wildtype SRA by measuring bound transcript. Labeled SRA and SRA mutant are added independently to separate solutions containing a target protein. Following an incubation period, the RNA/protein complexes are isolated. In one embodiment, the incubation occurs at any of a range of temperatures from 1 to 99 ° C . In another embodiment, the incubation occurs at any of a range of pH values from 1 to 13. In yet another embodiment, a mixture of SRA and SRA mutant transcripts are added to a first solution containing the protein target where the SRA transcript is labeled and a second solution containing the protein target where the SRA mutant is labeled. RNA/protein complexes are isolated by any of a number of methods well known in the art. In a preferred embodiment, the target protein is amenable to rapid isolation. Such methods may include, but are not limited to, conjugation to beads, conjugation to magnetic beads, isolation with an antibody, fusing the protein with an enzyme, and fusing the protein with an enzyme substrate. Labeled RNA transcript is measured for the isolated complexes. In a prefeoed embodiment, the protein in the complex is also quantitated for standardization. Protein may be quantitated by any of several methods well known in the art. Suitable protein targets include, but are not limited to, SRC family proteins such as SRC-1, SRC-2, and SRC-3 and mutants thereof.

In an additional embodiment of the invention, SRA mutants are evaluated for the ability to coactivate specific steroid hormone receptors. In this method, cells are transfected with an expression vector possessing a reporter construct and DNA coding for the SRA mutant. The cells are cotransfected with another expression vector possessing a steroid hormone receptor (herein refeoed to as steroid receptor A). In an alternative embodiment, the reporter construct is placed within the vector possessing steroid receptor A. An appropriate ligand for steroid receptor A is then administered and reporter construct product, product activity, or effect is measured as an indication of the ability of SRA mutant to coactivate steroid receptor A. In a preferred embodiment, parallel experiments are performed where SRA is substituted for SRA mutant above for the purpose of determining coactivation activity of SRA mutant relative to wildtype SRA. Steroid receptors may include, but are not limited to, AR, ER, GR, MR, and PR. Specific ligands for the steroid receptors are well known in the art. It is prefeoed that the reporter construct consist essentially of steroid hormone A binding domain in close proximity to a promoter which drives expression of a reporter gene. Specific binding domain sequences for the steroid hormone receptors are well known in the art. Promoters within the reporter construct may include, but are not limited to, TATA and TK promoters. Reporter genes may include, but are not limited to, LUC, CAT, A picillin, and Neomycin.

The following examples serve to illustrate specific embodiments of this invention, but should not be considered as limiting the scope of the invention in any way.

Example 1 DNA Library Screening

The coding sequence of the AF- 1 domain of the human PRA (cooesponding to amino acids 165 - 567 of hPRβ) was subcloned into the pASl yeast expression plasmid in frame with the amino acid sequence of the GAL4-DBD (1-147). The yeast-two hybrid screen was performed as previously described in (Onate et al, 1995). Transformants of a human β-lymphocyte cDNA expression library were tested in Y 190 strain for interaction with progesterone-induced h PRA RACE was performed using the Marathon cDN A Amplification Kit (Clontech) with skeletal muscle mR A (Clontech) and the following primers cooesponding to a presumptive ORF from a PR-interacting yeast clone: antisense 5'-CTGGGGGATCCATCCTGGGGTGCG-3' (Onl), antisense 5'-CCTGCAGCAGTGCCAGGCGTCGG-3' (On5) and sense 5'-

CGCGGCTGGAACGACCCGCCGC-3' (On3). SRA clones were isolated by homology screening of human lgtll cDNA libraries from skeletal muscle, heart and HeLa S3 cells (Clontech), human genomic library EMBL3 SP6/T7 (Clontech), mouse heart cDNA library 1ZAP cDNA (Strategene) and 129SVJ mouse genomic library lgtFIX II (Stratagene) using recommended bacteria strains and protocols as provided by the library manufacturers. Both strands of SRA clones were sequenced using Sequenase (Amersham) or Thermal Cycle DNA Sequencing (New England Biolabs).

Example 2

Northern and Southern Analysis

Human Tissue Northern (MTN) blot (Clontech) was hybridized with a probe corresponding to the Nael-HincII fragment of SRA and processed as recommended by the manufacturer. Tissue cell blots were prepared by isolation of total RNA using

TRIzol Reagent (Life Technologies) followed by electrophoresis in agarose (FMC)/TAE gels, alkaline transfer onto Zeta-Probe (BioRad), UV-crosslinking and hybridization overnight at 60°Cin0.5MNaHPO4pH7.2, 7% SDS, ImMEDTA. The human tissue-culture RNA blot was hybridized with a 1.5kb probe cooesponding to SRA-isoform III (Figure 1A). The HeLa cell blot was hybridized with a probe encompassing the entire luciferase cDNA. The blots were stripped and subsequently hybridized with a probe specific for β-actin (MTN and tissue cell blots) or cyclophilin (HeLa cell blot), respectively. RT-PCR products were electrophoresed, blotted and hybridized with a probe cooesponding to isoform I of SRA. Probes were generated using random DNA labeling Kit (Life Technologies) and 50μCi of [³²P]dCTP,

300Ci/mmol (ICN) followed by EtOH-precipitation or G-50 (Boehringer Mannheim) column purification.

Example 3 Plasmids The reporter constructs (PRE)2-TATA-CAT (CAT, chloramphenicol acetyltransferase) and (ERE)2-T AT A-C AT have been described (Vegeto et al, 1992), the MMTV-Luc (MMTV, mouse mammary tumor virus LTR; Luc, Luciferase) was -re¬

generated by subcloning the Acc65I-XbaI fragment from pGLBasic3 (Promega) into the blunt-ended EcoRI site ofMMT V-KCR (Baylor College ofMedicine). The human cytomegalo virus (CMV)-driven mammalian expression vectors p STC for human PRβ , GR, AR and ER were generated by fusion of the cDNAs to the HSV-TK leader sequence containing a Kozak consensus sequence (Lanz et al, 1995); rat GR,

GRΔAF1 and GR-DBD have been described (Rusconi and Yamamoto, 1987). PRΔAFl is an N-terminal truncation of pSTC- hPRβ at the Accl-site and re-ligation to the blunt-ended BamHI-site of the TK leader, PRΔLBD is a C-terminal truncation of pSTC- hPRB at Dral. TRβ, RARa, RARg, RXRg, Gall 47, SP1, E2F, E47D and CREB and corresponding reporter constructs were from S.AO and M.-J.T. and published elsewhere (Cooney et al, 1992; Baniahmad et al, 1993; Leng et al, 1994; Onate et al, 1995), PPARg (Glaxo Research Institute), SRC-1 (Baylor College of Medicine) and CBP (Vollum Institute, Oregon Health Sciences University). SRA constructs were generated by subcloning the cDNAs into a modified linker of the CMV-driven pSCT-1 vector (Rusconi et al, 1990). Excision of BamHI fragment of pSCT-SRA and re-ligation generated ΔATG, and the fusion of the BamHI-or Nael- restricted SRA to the HSV-TK leader sequence generated tk-ORFl and tk-ORF2, respectively. The reading frame mutations ORFl, ORF2, ORF3 were generated by PCR-cloning using the sense primers 5'-TGGGGGATCCTACCTCAGGTGCGG-3', 5 ' - T GGGAGAT C T AT C C T AG GGT G C G G - 3 ' an d 5 ' - T G-

GGGGATCCTACCTAGGGTGCGG-3' and antisense 5'-CAGACTCACCGGAC- ACCATCTCCTA-3' (On8), followed by restriction subcloning into the pSCT-SRA vector. Ylle used the primer 5'-ATAGCAATTGGGCCTCCACCTCCTTCAAG-3' to destroy an ATG and to introduce a Mfel site in mutant ORF2. Frame shift mutations were generated by restriction of SRA or mutant ORF2 with selected enzymes, filled-in with Klenow DNA polymerase, and re-ligated at following sites: Bbsl (generated mutant B), SgrAI (S), Mfel and SrgAI (MS, YMS), and Mfel, Bbsl and SgrAI (YMBS). 3' deletions at Bbsl of ORFl, ORF2, ORF3 generated ΔORF1, ΔORF2 and ΔORF3. pP(A)LiSK vectors for in vitro transcription were generated by subcloning the cDNAs for SRA, AR and SRC-1 into a modified version of pSP64 Poly A (Promega) containing an additional poly linker 3' of the polyA-sequence for linearization of the plasmid.

Example 4

Cell Culture and Transient Transfection Assays

Cell lines were routinely maintained at 37°C / 5% CO2 in Dulbecco's modified

Eagles' medium (HeLa, COS) or RPMI medium 1640 (T-47D) supplemented with 5- 10% charcoal- stripped fetal calf serum. 10⁵ cells were plated out per well in 12-well dishes for luciferase assays, 5x10⁵ cells per dish in 6-well dishes for CAT-assays and 10^s cells per 10cm dish for assays that involved cell culture in the presence of cycloheximide. Medium was replaced 3h prior transfection with medium containing 50μM cycloheximide and maintained until cell harvesting. Cells were transfected with the indicated DNAs using lipofectin (Life Technologies) or SuperFect (QIAGEN) and treated according to the manufacturers guidelines. In all transfection experiments, reporter plasmids were abundant (2.5μg per 10 cells), whereas nuclear receptors were transfected in limiting amounts (20-100ng per IO⁶ cells). Upon DNA addition, cells were cultured for 36-42h for CAT-assays, 20-24h for Luciferase assays and 11- 14h in the presence of cycloheximide. Ligand stimulation involved incubation of cells with progesterone (lOnM), RU486 (50nM), dexamethasone (50nM), R1881 (lOnM), or estradiol E2 ( 1 OnM) for six hours prior to cell harvesting. Cell lysates were assayed for CAT activity with lOOμCi of [ 14 Cjchloramphenicol and 5mM acetyl coenzyme

A (Sigma) as substrate and separated by thin-layer chromatography. Luciferase activity was determined using the luciferase assay system (Promega) and an analytical luminescence detector. Values were cooected for protein concentration. Data are presented as the mean (±SD) of triplicate values obtained from a representative experiment that was independently repeated at least three times.

Example 5 Western Analysis Mouse monoclonal antibodies (mAb) against SRA, SRC-1 and AR were prepared at the University of Colorado Health Science Center. SRA-mAb was raised against the peptide sequence T AEKNHTIPGFQQAS cooesponding to the C-teoninus of the presumptive ORFl of human SRA. The mAb was purified from hybridoma culture supernatants using a mAb TRAP Gil column (Pharmacia) . SRC- 1 -mAb was described previously (Spencer et al, 1997), AR-mAb (Baylor College of Medicine), CBP-Ab was obtained from Upstate Biotechnology, NY. Protein blots were blocked in 150mM NaCl, 50mMHEPES, 5mMEDTN 3% BSA, 0.25% Gelatin, 0.05% Triton X-100 for 2h at room temperature (rt), then denaturated in 4M urea for 3h at RT followed by washing and incubation overnight at 4°C with primary mAb in 150mM NaCl, 50mM HEPES, 5mM EDTA, 1% BSA, 0.25% Gelatin, 0.1% Triton X-100.

Alternatively, protein blots were blocked for one hour at rt in 5% milk in NTT (137mM NaCl, 50mM Tris-HCl pH 7.5, 0.05% Tween-20) and incubation with the appropriate primary antibody for 2h at RT . Specifically bound antibody was visualized by one hour incubation with goat anti-mouse secondary antibodies conjugated to alkaline phosphatase or to horseradish peroxidase (BioRad) followed by chemiluminescence detection with ECL, as recommended by the manufacturer (Amersham Life Science). Blots were stripped for reprobing by incubation in 65mM Tris-HCl pH 6.7, 2% SDS, 50mM β-mercaptoethanol for 30 min at 50°C.

Example 6 Gel Filtration

Biochemical fractionation of cell lysate was carried out as described (McKenna et al, 1998). Two subconfluent 15 cm plates of T-47D or HeLa cells were washed and harvested in phosphate-buffered saline and thoroughly lysed with a motor pestle homogenizer in 50mM NaCl, 5mM KC1, 20mM HEPES pH 7.2, ImM EDTA, 10% glycerol, l-2U/μl RNasin ribonuclease inhibitor (Promega), ImM DTT, ImM phenylmethylsulfonylfluoride (PMSF) and 1 μg/ml of a protease inhibitor cocktail (Sigma). After centrifugation at 60,000 rpm at 4°C for 20 min, 200 μl of supernatant was loaded on a Superose 6 gel filtration column (Pharmacia) pre-equilibrated with 150mM NaCl, 50mM sodium phosphate pH 7.0, lU/μl RNasin and controlled by an FPLC system (Pharmacia). For antibody shift experiments, clarified lysates were rocked for 45 min. at 4°C with 2μg of SRC-1 mAb and a four-fold excess of rabbit anti-mouse IgG (Zymed). Half of each column fraction (400 μl) was processed for

RNA isolation and RT-PCR analysis, the other half precipitated with BSA/trichloroacetic acid, separated on 7.5% polyacrylamide gels and transferred overnight to nitrocellulose membrane (BioRad) at 60V and 0-4°C for Western analysis. Example 7

SRA-specific RT-PCR 20-3 Oμl cell extracts or 200 μl column fractions were supplemented with 5mM MgSO4 and incubated for 25min at 37°C with 20-40U RNase-free DNasel (Boehringer Mannheim) and 2U RNasin ribonuclease inhibitor (Promega). Alternatively, cell extracts were incubated for 20 min. on ice with different concentrations (lμg, lOng, O.lng) of a mixture of SRA-specific antisense deoxyoligonucleotide of Onl, On5, On7 (5'-CTTACCGAGATGACCACGTT- CCTTGATTC-3') or 1 μg of sense On4 (5*-GCCGACGCCTGGCACTGCTGCAGG- 3') and subsequently treated with 100U ribonuclease RNase H (Life Technologies) in lOOμl 20mM HEPES pH8.0, 50mM KC1, 4mM MgC12, ImM DTT, 50mg/ml BSA for 20 min. at 37°C. As a control, cell extract was digested with 50U RNase Tl (Boehringer Mannheim) for 20min. at 37°C. Total RNA was extracted using 1ml TRIzol Reagent and processed according the manufacture protocol (Life Technologies). EtOH-washed RNA was resuspended in 12μl H2O, 2pmole SRA- specific primers (sense On3: 5'-CGCGGCTGGAACGACCCGCCGC-3' and antisense On8: 5'-CAGACTCACCGGACACCATCTCCTA-3\ see Figure 2). First strand cDNA synthesis was generated using Moloney reverse transcriptase and reagents supplied with the Superscript II Kit (Life Technologies). 20% (4μl) of the reaction was used in a 50μl PCR- amplification using 5U of Taq-DN A Polymerase (Promega), 2mM MgCl2, 150μM dNTPs, lμM of primers (On3/On8). PCR was performed as follows: 3min. denaturation at 95°C, 25-40cycles of 30s at 95°C, 45s at 58°C, 40s at 71°C, and 5 min. extension at 72°C. PCR products were visualized on 1.2% agarose gel, blotted to Zeta-Probe GT membrane (BioRad) by alkaline transfer and Southern analysis performed as described above.

Example 8 Immunoprecipitation in Xenopus leavis oocytes pP(A)LiSK-cDNA constructs were transcribed in vitro with SP6 RNA polymerase and the mMessage mMachine kit (Ambion) to generate 200-400μg/μl specific mRNA. Xenopus laevis oocytes were injected with 27.6nl specific mRNA and L- 35 S-

Methionine and cultured for 12-16h at 18°C inMBSH [lOπiM HEPES ρH7.6, 88mM

NaCl, ImM KCl, 2.4mM NaHCO3, 0.82mM MgSO4, 0.41mM CaCl2, 0.33mM Ca(NO3)2]. Oocytes were lysed in extract buffer [20mMHEPES ρH7.6, 70mMKCl,

2mMDTT, 0.1%NP-40, 8% Glycerol, ImM PMSF and lU/μl RNasin (Promega)] in a ratio of lOμl extract buffer per oocyte and repeatedly centrifuged to remove cell debris. Clear lysates were incubated with 4μg of SRC-1-mAb or 2μg of AR-mAb and a four-fold excess of rabbit anti-mouse IgG (Zymed) for 45min. at 4°C, followed by 45min. incubation at 4°C with protein- A Sepharose (Pharmacia) that was washed and equilibrated in extract buffer. Subsequently, beads were washed four times with extract buffer and bound material was analyzed by RT-PCR as described above and by SDS-PAGE.

Example 9 Specimen Preparation and RNA In-Situ Hybridization Adult 129SvEvBrd male mice were sacrificed by cervical dislocation, the brains removed and fixed in ice-cold 4% paraformaldehyde for 20h. Tissue was dehydrated, embedded in parafilm, and sectioned at a thickness of 7μm. In situ hybridization was carried out as described (Albrecht et al, 1997). Antisense and sense riboprobes were synthesized with T3 or T7 RNA polymerase in the presence of a 35 S -UTP (1250Ci μmol, Du Pont NEN). The entire mouse SRA cDNA (0.9kb) was used to generate the SRA probe. The PR probe was made from a lkb mouse genomic DNA containing 700bp of 5'UTR-sequence (Lydon et al, 1995). The GR probe was generated from a cDNA encompassing 1038bp of the DNA- and ligand-binding domain of mouse GR (Cole et al, 1995). Hybridization was done overnight at 55°C. Stringency washes were performed at 65°C. Slides were dipped in NTB-2 emulsion and exposed for 6-9 days. Tissue was visualized by fluorescence of Hoechst dye- stained nuclei (blue color in figures). Silver grains (white = SRA, yellow = PR, red = GR) were visualized by dark-field illumination.

Example 10 Cofactors

This invention provides methods and products based substantially on the unexpected discovery of novel RNA transcripts which function as steroid receptor coactivators without the requirement of translation into protein. Efforts which lead to the invention arose from attempts to find cofactors that interact with steroid hormone receptors. These efforts are chronicled below.

In an attempt to find cofactors that interact with steroid hormone receptors, different functional domains of the human progesterone receptor (PR) were used as baits in a yeast two hybrid screening system. Previous reports have described the isolation and characterization of a protein, SRC-1, that interacts with the ligand binding domain of PR (Onate et al, 1995). In addition, a similar screen of a human B- lymphocyte library with the amino terminus of PR_A (cooesponding to amino acids 165-567 of PR_B) was performed. Primary sequence analysis of two positive clones from this screen revealed a rather short open reading frame (ORF). The 3' extension of reverse-transcribed skeletal muscle poly-A⁺ RNA resulted in an extended ORF with sequence identical to the 5' ORF from the lymphocyte library. Given that this cDNA enhanced PR transactivation when subcloned into a mammalian expression vector and transfected into cultured cells, this activity was further investigated and later termed

SRA.

Example 11 Full Length SRA cDNA In order to retrieve full length SRA cDNA, conventional screening of three different human cDNA libraries from skeletal muscle, heart and HeLa S3 was performed. Thirteen positive clones were obtained which possessed DNA sequences that were identical in a central region. Three variants of SRA were predicted, all containing unique 5' and 3' extensions beyond an identical 687bp long core sequence (Figure 1 A and 2). A human genomic DNA library was also screened and two clones were found possessing partial sequence identity to the original SRA clones.

Additionally, screening of a mouse genomic DNA library identified five positive clones, and screening of a mouse cDNA library found 14 positive clones, of which two revealed 75% identity to the human SRA cDNA (Figure 2). Primary sequence analysis of the full length clones from different human and mouse cDNA and genomic DNA suggested that SRA represented a family of clones that are highly homologous in a core sequence but are divergent in their 5' and 3' regions. Sequence comparison using the BLAST algorithm indicated no homologs but identified partial SRA sequences isolated as HepG2-3 'UTR (accession number D 16861 ), expressed sequence tag (EST) clones (H30345, W82929, W49707), and chromosome 5 BAC clone 319C17 (AC005214), although no functions for these sequences were described.

Example 12 Length of Transcript

To determine the length of the corresponding transcripts and to study the expression profile of SRA Northern analysis was performed using a cDNA probe corresponding to the core sequence of human SRA. Major transcripts of 0.7-0.85 kilobases (kb) in length and less abundant transcripts of 1.3-1.5 kb were detected in a human Multiple Tissue Northernblot (MTN) (Figure IB), indicating that the cloned cDNAs were likely to be full length. The levels of expression of SRA were tissue specific; SRA was predominantly found in liver and skeletal muscle, but was expressed at a low level in brain. Interestingly, the expression of the two messages in the 0.7- 0.85 kb doublet appeared to be tissue specific in the MTN. A cell line-specific expression of the isoform ratios was also seen in a Northern analysis of poly-A⁺ selected mRNA from different human tissue-culture cell lines. The 1.3-1.5kb transcripts were not detected in the cell lines tested. In addition, all the cell lines tested expressed the -0.85 kb doublet species, whereas the smaller -0.7 kb species was expressed at significantly higher levels in the breast cancer cell lines MCF7 and T-47D compared to the other cell lines investigated (Figure IB). This isoform-specific expression was conserved in mouse tissues . It was concluded that SRA isoforms are expressed in a tissue- and cell type- specific manner.

Example 13 Functional Characteristics

To investigate the functional consequences of the isolation of SRA in the two hybrid screen with PR-AF1, diverse cDNAs encoding human SRA were subcloned into mammalian expression vectors and assayed for the effect of SRA on PR-dependent transactivation. HeLa cells were cotransfected with the CMV-hPR and CMV-SRA along with (PRE)₂-TATA-CAT reporter and induced with progesterone (R5020). It was observed that SRA enhanced PR transactivation (Figure 3 A, compare lanes 2 and 3) and that SRA did not alter the activity of PR in the presence of its antagonist

RU486 (lane 4). Furthermore, SRA did not significantly elevate the basal activity of the minimal promoter (lane 5). Similar transfection experiments with the human receptors for glucocorticoid (GR), androgen (AR) or estrogen (ER) and CAT reporters containing cognate HREs revealed that SRA enhanced steroid receptor- mediated transactivation (Figure 3 A).

It was next verified that SRA enhanced transactivation through the N-terminal AF1 portion of steroid receptors. Truncation of the A/B-domain of the PR (PR- ΔAF1) abolished coactivation by SRA (Figure 3B lanes 16 and 17), whereas activation of transcription by PR lacking the LBD (PR-ΔLBD) was fully responsive to SRA in a hormone-independent manner (lanes 12 -15). In order to exclude the

DNA binding domain as a mediator for SRA coactivation, different domains of rat GR were tested as fusion proteins with the activation domain of GAL4. As expected, neither the N-terminal truncated GR-ΔAF1 nor the DNA binding domain of rat GR (GR-DBD) responded to SRA to enhance luciferase reporter activity. As a control, GR-ΔAF1 enhanced reporter activity in the presence of SRC-1. These results indicated that SRA functionally interacts with the amino terminal AF 1 of the receptors to enhance transcription.

Example 14 Coactivation

Having established that SRA was an AF 1 -specific coactivator its specificity of coactivation was next tested. SRA did not enhance transactivation induced by thyroid hormone receptor (TR), all-trans retinoic acid receptor (RAR), 9-cis retinoic receptor (RXR) or peroxisome-proliferator activated receptor (PPAR). Furthermore, SRA also did not alter the transcription activity of other activators including GAL4, SP1, E2F, E47 and forskolin-stimulated CREB. In addition, the possibility that an intramolecular 'crosstalk' between the AF-1 and AF-2 of the receptor is mediated by

SRA and SRC-1 was tested, but found that coexpression of both coactivators had only an additive effect on the coactivation of PR-mediated transactivation. To better determine the coactivation potential of SRA in vivo different SRA clones were transfected into T-47D cells and tested for the ability of SRA to enhance endogenous PR. All three isoforms of SRA cDN A as well as a portion of the human genomic SRA enhanced transactivation mediated by the endogenous PR by 8-12 fold; the core domain of SRA was found to be necessary and sufficient for this coactivation (Figure 3C).

Example 15 Reverse Interference

Another criteria for classification as a coactivator is the ability to reverse interference by transcriptional activators with common coregulators. In order to ask if SRA is a limiting factor that can be sequestered by an excess of another receptor in vivo, SRA was overexpressed in a PR-regulated transcription reaction in the presence of ER (Figure 3D). Ligand activated ER reduced the transcription activity of ligand- bound PR by 50%. Full PR transactivation was re-established by addition of SRA, confirming that SRA regulates the transactivation of both PR and ER in a dose- dependent manner and that SRA has a similar affinity for both receptors. This indicated that SRA was a limiting cellular factor for steroid receptors. Together with the ability of SRA to enhance transactivation without alteringbasal transcription, these results clearly characterized SRA as a bonafide coactivator, specific for the AF-1 domain of steroid receptors. As indicated earlier, sequence analysis of the SRA clones revealed an open reading frame (ORF) that is terminated at the 3 '-end of the core sequence (see Figure 1A. A detailed ORF map is shown in Figure 4 A. For clarity this ORF is denoted ORFl). A second ORF, denoted ORF2, contains a consensus Kozak sequence (Kozak, 1989) in the 5' portion of the SRA cDNA. This ORF2 cooesponds to the presumed receptor-interacting reading frame of the yeast hybrid clones. Unexpectedly however, an in-frame stop codon terminated GAL/SRA fusion products prematurely at the 5' end of the core sequence. It was concluded that the interaction of the original yeast-two hybrid SRA clones with the AF- 1 of PR was unlikely to have been mediated by a protein product encoded by ORF2 of SRA.

Example 16 Activation Domain In order to define the activation domain of SRA, various cDNAs were fused in different reading frames to the GAL4 DNA binding domain. The resulting fusion proteins were tested in cultured cells with 4xUAS (upstream activation sequence) linked to a luciferase reporter gene. Interestingly, all SRA constructs failed to activate the UAS heterologous promoter while the control construct consisting of the activation domain 2 of SRC-1 (Onate et al, 1998) fused to the GAL-DBD significantly enhanced reporter gene activity. These results indicated that SRA did not posses an intrinsic activation function. Attempts were then made to characterize the translated SRA "protein" product. Surprisingly, all efforts to generate SRA protein were unsuccessful. In vitro translation of different clones did not result in detectable levels of protein, whereas carboxyl terminal fusions with GAL4 or GST produced the expected translation products. A monoclonal antibody (mAb) against the peptide sequence encoded by the 3'-end of the SRA core (ORFl) was then generated. In

Western analysis, GAL/SRA and GST/SRA ORFl -fusions were immunoreactive, whereas no translation products of expressed SRA cDNAs were detectable in cell extracts. It was concluded that the SRA-cDNA sequence did not result in a viable translation product.

Example 17 Correlation Coactivation with Function Attempts were next made to cooelate the coactivation function of SRA with its expression. Mutated SRA constructs were transfected into cultured cells and analyzed in a side-by side comparison for SRA "immunoreactivity" and for coactivation of PR-mediated transactivation. The constructs tested were a 5'- truncation at the BamHI-site, eradicating the consensus Kozak sequence, and a fusion of this truncated cDN A to the HS V-thymidine kinase initiation sequence (tk) in two distinct reading frames producing tk-ORFl and tk-ORF2 (Figure 4A). Figure 4B shows that all SRA mutants enhanced PR-mediated transactivation (right panel), whereas only one construct - the reading frame of which cooesponded to ORFl - was recognized by the monoclonal antibody (left panel) . No endogenous SRA protein was detected which cooesponded to the constrained translation of tk-ORFl . Systematic screening of a panel of tissue-culture cell lines by matrix-bound SRA-mAb confirmed the absence of endogenous SRA 'protein' in tissue-culture cells . Taken together, these results suggested that coactivation by SRA was unlikely to be mediated by its presumptive protein product. In order to substantiate these results, various SRA mutants were generated and tested in cell cultures for their ability to coactivate PR-dependent transcription Figure 4 A illustrates the sequence of the SRA mutants relative to the original SRA clone. Several of the mutants lacked the ATGs in ORFl and ORF2; others contained mutations within the Kozak sequence, allowing a presumptive translation of only one given reading frame (see legend to Figure 4 A for details). Other mutants contained single or multiple frame shifts along the core sequence, resulting in a 'mosaic' organization of reading frames each containing -6 stop codons on average. A representative assay of in vivo expressed SRA mutants (Figure 4C) clearly demonstrates that most of the SRA mutants retained the ability to enhance PR- transcription by 8-12 fold. Only SRA expressed in 3'-5' orientation (SRA inv) or 3' half-truncated versions of the ORF-exclusion mutations were inactive (Figure 4C). Similar results were obtained with mutants of other SRA isoforms . These results further suggested that the coactivation exerted by SRA on steroid receptor transcription was unlikely to be mediated by a translation product of the SRA gene(s). Motivated by the fact that extensive internal mutagenesis did not alter SRA coactivity, investigation was next focused on the transcription products of SRA. An assay for RNA-mediated transactivation was designed by targeting endogenous GR in cells that were cultured in the presence of cycloheximide and it was determined whether SRA retained the ability to coactivate GR-mediated transcription. As controls, the coregulators SRC-1 and CBP were used, both of which interact with nuclear receptors as proteins. Two separate sets of HeLa cells were transiently transfected with an identical mixture of MMTV-Luciferase reporter along with CMV- driven expression plasmids for SRA, SRC- 1 , CBP, or empty vector, and treated with EtOH or dexamethasone. One set of transfected cells was subjected to a conventional luciferase protein assay for GR-mediated transactivation. The second set of cells was incubated in medium containing cycloheximide from 3h prior to transfection until harvesting. After harvesting, these cells were subjected to RNA isolation followed by

Northern analysis for luciferase RNA expression. Figure 5 shows a representative side-by side comparison of luciferase expression as protein (upper half) and RNA (lower half). As expected, a hormone- and dose-dependent enhancement of transactivation was observed for all coregulators in the absence of cycloheximide, as measured by luciferase protein activity (upper panel) . The relatively low coactivity for all coactivators resulted from lower protein expression levels due to the necessarily shorter incubation time for cycloheximide-treated cells. In contrast the Northern analysis for luciferase expression of the set of cycloheximide treated cells (lower panel) revealed that SRA (lanes 5-8) but not SRC-1 (10-12) or CBP (14-16) was able to enhance transcription under these conditions. As a control, ³⁵S-methionine incorporation was assayed in the two groups of cells; it was found that the amount of cycloheximide used in our assays (50μM) abolished >99% of the total cellular translation products . Moreover, a third control set of transfected cells treated with cycloheximide and analyzed for luciferase reporter activity showed relative light units below 1000, cooesponding to basal activity . The fact that only SRA and not SRC-1 and CBP was capable of potentiating GR-mediated transcription in the absence of de novo protein synthesis was clear evidence that the functionality of SRA was not contingent upon translation of the primary SRA transcript.

Example 18 RNA/Protein Association Given that functional RNAs are known to associate with proteins as ribonucleoprotein complexes, it was next asked if SRA might function as a component of similar complexes. For this purpose, a specific RT-PCR assay for SRA was developed. Figure 6A shows that PCR amplified SRA in RNA preparations of untreated extract (lane 1) but did not generate a signal in extracts treated with RNase Tl, which is an endoribonuclease that cleaves single-stranded RNA (lane 2). Cell extract incubated with antisense deoxyoligonucleotides and subsequently digested with the endoribonuclease RNase H, which specifically cleaves RNA in RNA:DNA hybrids, destroyed the SRA signal in an oligonucleotide- and dose-dependent manner (lanes 4-6). These results demonstrated that the RT-PCR assay was specific for SRA- RNA and that signals were not due to DNA contamination. The above RT-PCR detection method was employed to investigate protein/SRA interaction in a steady-state situation in vivo. Human T-47D cells were lysed and fractionated on a Superose 6 column as previously described (McKenna et al, 1998). One half of the fractions were processed for Western analysis using specific antibodies against transcriptional coregulators, and the remainder of each fraction was subjected to RNA isolation for the RT-PCR assay. Unexpectedly it was found that SRA co-purified with fractions containing the general nuclear receptor coactivator SRC-1 in complexes of 600-700 kD (Figure 6B, compare fractions 34 and 35 in the top panels). In addition, these fractions contained the steroid receptor coactivator TIF2 . The co-localization of SRC-1 and SRA suggested that they may be part of a common complex in vivo. Based on this presumption, it was predicted that addition of antibodies specific for SRC-1 would result in a shift of the elution pattern of this complex. Consequently, cell lysates were incubated with anti-SRC- 1 antibody and rabbit anti-mouse IgG prior to fractionation. As predicted, this resulted in a clear shift of both the SRA signal and SRC-1 immunoreactivity from fractions 34-35 to fractions 29-30, which contain significantly larger protein complexes. To exclude the possibility that the shifted SRC-1 was due to non-specific antibody binding, the Western blots were stripped and re-probed with anti-CBP antibody. The elution profile of CBP was the same irrespective ofpreincubationofcell lysate with SRC-1 antibody . To test the possibility that SRA might have a structural role in the SRC-1 complex, cell extract was treated with RNase prior to fractionation. These extracts did not produce SRA- signals in the RT-PCR analysis, but still revealed SRC-1 in fractions 35-36, suggesting that SRA does not have a vital structural role in these complexes. SRA was also detected in larger complexes in the ~2Mda range, none of which migrated upon treatment with SRC-1 antibodies (fractions 22, 24, 26). The human homologs of the SWI/SNF proteins BRG-1 and BAF-57 have previously been identified in these fractions (McKenna et al, 1998), indicating that SRA may be a component of distinct complexes involved in transcription and chromatin remodeling. Interestingly, SRA was not detected in fractions containing p300/CBP (fractions 28-31). Taken together, biochemical fractionation experiments indicated that SRA is a component of distinct ribonucleoprotein complexes, one of which contains the nuclear receptor coactivator SRC-1.

Example 19 SRA Hormone Interaction Having established that SRA was present in SRC-1 -containing complexes, it was next important to determine if SRA interacts with steroid receptors as a component of a ribonucleoprotein complex. McKenna et al (1998) showed that ligand hPR interacted stably with complexes containing SRC-1 and TIF2. To address the possibility that SRA might interact with steroid receptors as part of an SRC-1 containing complex, co-immunoprecipitation experiments were performed using a previously described expression system in Xenopus oocytes (Wong et al, 1995). In vitro generated RNA encoding SRA, SRC-1 and AR along withL-³⁵S-Methionine was injected into oocytes as indicated in Figure 6C and the cell extracts subjected to co- immunoprecipitation with antibodies against AR and SRC-1. Figure 6C shows the cDNA products generated by SRA-specific RT-PCR of the various immunoprecipitates along with an autoradiograph of SDS-PAGE analysis of the precipitates. SRA was undetectable after immunoprecipitation using a non specific antibody from cell lysates programmed with SRA (lane 2). Similarly, SRA was not detected after immunoprecipitation with an AR antibody from cell lysates programmed with AR, although AR was detectable in this precipitate (lane 3). In contrast, the AR antibody precipitated SRA in extracts from oocytes injected with RN As for SRA and AR (lane 4). In addition, immunoprecipitation using a monoclonal antibody against SRC-1 from oocytes programmed with RNAs for SRA and SRC-1 clearly co- precipitated SRA and SRC-1 (lane 5), verifying that SRA is in a stable association with SRC-1. It was concluded that SRA exists in ribonucleoprotein complexes containing SRC-1 and that this complex is recruited by a steroid receptor. Given the selectivity of SRA for steroid receptors it was speculated that the spatial and temporal expression of SRA might coincide with the expression of type I receptors, in contrast to the rather ubiquitous distribution of SRC-1 and other AF-2 coactivators. The mouse brain was selected for in situ hybridization analysis because it was anticipated that the low expression of SRA in brain (Figure IB) may allow comparison with the reportedly subtle and distinct distribution of the steroid receptors in this tissue (Simerly et al, 1990; Shughrue et al, 1997). Based on their disparate functions, PR and GR were selected for the comparison and mouse orthologs were used to generate specific riboprobes. Figure 7 shows a typical in situ hybridization analysis on adjacent coronal sections of three different regions in mouse brain, revealing cell-type specific expression which cooelates with PR and GR expression. An identical distribution is most evident in the olfactory bulb (top panels), where the distribution of SRA, PR and GR was restricted to the lateral and dorsolateral olfactory tract (lo and dlo), the glomerular layer of olfactory bulb (Gl) and the anterior olfactory nucleus (AO). In the hippocampus (middle panels), SRA was highly expressed in all the fields of the cornu ammonis (C A), whereas GR was restricted to C Al and PR was absent. However, AR has been shown to be expressed throughout the cornu ammonis (Simerly et al, 1990; Keo et al, 1996), indicating that the expression of Type I receptors overlaps with the distribution of SRA in the hippocampus. In the hypothalamus, the expression of SRA, PR and GR was ubiquitous (Figure 7, lower panels); PR was expressed predominantly in the lateropost arcuate hypothalamic nuclei (ArcLP), whereas GR was also highly expressed in the mediapost arcuate hypothalamic nuclei (ArcMP). The distribution of SRA in the hypothalamus cooelates with the expression of the steroid receptors, since it has been previously shown that AR and ER are also expressed in this part of the brain (Simerly et al, 1990; Shughrue et al, 1997). Taken together, these results demonstrate a cell-type specific expression of SRA that correlates with the expression of several steroid receptors. Given the ubiquitous expression of SRC-1 in similar hybridization experiments (unpublished data), these results suggest that SRA is not an obligatory component of all SRC-1- containing complexes, but rather associates with SRC-1 in tissues in which they are coexpressed, possibly to confer transcriptional specificity on these complexes. The above description of the isolation and functional characterization of a novel transcriptional coactivator (SRA) appears to be the first evidence that RNA transcripts (without need of translation) modulate steroid receptor function. SRA is the first endogenous RNA transcript identified to function as a eukaryotic transcriptional regulator. Several unique features of SRA have been defined: (i) a bonafide transcriptional coactivator, (ii) SRA is selective for the AF-1 of steroid hormone receptors, (iii) SRA isoforms are expressed in a cell- and tissue-specific manner, and (iv) SRA is present in preformed coregulator complexes.

Example 20 SRA is Overexpressed in Steroid-Dependent Human Tumor Tissue Northern analysis of different tissue culture cell lines revealed a significantly higher expression of SRA in breast adenocarcinoma MCF-7 cells and in breast ductal carcinoma T-47D cells when compared to other cell lines tested. Based on these results analysis of several commercially available human RNA blots was performed. Hybridization of an SRA-cDNA probe to a human tumor panel blot from Invitrogen showed that SRA is significantly overexpressed in breast, uterine and ovarian tumor tissue but was not overexpressed in an adenocarcinoma of the fallopian tube (Figures 10, 11 and 12). Likewise, Northern analysis ofa human ovary tumor multi-sample blot (BioChain) showed that SRA was much more abundant in three of the four ovarian tumors tested (Figures 13, 14 and 15). Example 21 Correlation of SRA Expression Levels

The expression levels of SRA can be cooelated, to a certain extent, with the response of the tissue to steroids. The tissues from breast, ovaries, and uterus respond very sensitively to progesterone and estrogen (and other ligands), whereas the fallopian tubes are not steroidal. This correlation may be valid for tumor tissue but does not remain the same in normal tissue. Hybridization of a normalized multiple tissue expression aoay (Clontech) with an SRA-probe demonstrated that SRA expression is rather low in normal human tissues from mammary gland, ovary or uterus (Figures 9, 18, 19A and 19B), but is very high in the pituitary and adrenal glands (Figures 18, 21Aand 21B). So far, SRA-mediated transcriptional coactivation has been restricted to steroid receptors. In addition, in situ RNA hybridization techniques show that SRA and some steroid receptors are co-expressed in distinct regions and in particular cell types in the mouse brain. A cooelation between steroid- response and SRA expression was also demonstrated in Northern blots of human and mouse tissues. SRA is abundantly expressed in human skeletal muscle (Figures 19A and 19B ) but is almost absent in mouse muscle tissue, reflecting the fact that mouse muscle is not very sensitive to steroids.

All patents and publications mentioned in the specifications are indicative of the levels of those skilled in the art to which the invention pertains. All patents and publications are herein incorporated by reference to the same extent as if each individual publication was specifically and individually indicated to be incorporated by reference.

One skilled in the art readily appreciates that the patent invention is well adapted to carry out the objectives and obtain the ends and advantages mentioned as well as those inherent therein. SRA (Steroid Receptor RNA Activator), pharmaceutical compositions, treatments, methods, procedures and techniques described herein are presently representative of the prefeoed embodiments and are intended to be exemplary and are not intended as limitations of the scope Changes therein and other uses will occur to those skilled in the art which are encompassed within the spirit of the invention or defined by the scope of the pending claims.

Claims

CLAIMS:

1. An RNA transcript having the ability to selectively enhance transcriptional activation of steroid receptors without the requirement of translation of said RNA transcript into protein in eukaryotic cells.

2. The RNA transcript of claim 1 , wherein said transcript is sequence SRA1.

3. The RNA transcript of claim 1, wherein said transcript is sequence SRA2.

4. The RNA transcript of claim 1 , wherein said transcript is sequence SRA3.

5. The RNA transcript of claim 1, wherein said transcript complexes the core sequence.

6. An RNA having the ability to transactivate a DNA response element wherein said activation indicates RNA/protein binding or interaction.

7. A method for the detection of an RNA/protein interaction between an RNA sequence and a protein comprising the following steps:

(a) inserting a first DNA sequence coding for an R A transcript sequence into a first expression vector transcriptionally controlled by an inducer, wherein said inserted DNA sequence is inserted in series with vector DNA coding for a linker sequence and SRA and wherein expression of this first construct produces a first product containing a RNA sequence consisting of said RNA transcript sequence linked in series to said linker sequence and said SRA;

(b) inserting a second DNA sequence coding for a first protein into a second expression vector wherein said second DNA sequence is in frame with DNA coding for a GAL4 DNA binding domain and wherein expression and translation of this second construct produces a second product containing a fusion protein of first protein and GAL4 binding domain, said fusion protein capable of binding to GAL4- binding DNA element; (c) transfecting cells with the vector constructed in step (a), the vector constructed in step (b), an expression vector for the activation domain 2 of SRC- 1 and a reporter construct;

(d) inducing the vector constructed in step (a) with an appropriate inducing agent to promote transcription of the DNA coding for the RNA transcript;

(e) transcribing and translating said vector constructed in step (b), the expression vector and the reporter construct; and

(f) measuring the reporter product or activity.

8. The method of claim 7, wherein said DNA sequence coding for RNA transcript contains at least one in-frame stop codon.

9. The method of claim 8, wherein the at least one in-frame stop codon is in the 5' region of the sequence.

10. The method of claim 7, wherein the reporter construct codes for a protein conferring resistance to or improved survivability in an otherwise toxic environment to the host cell.

11. The method of claim 7, wherein an expression vector for the activation domain 2 SRC-1 is replaced with hPRΔDBD.

12. The method of claim 7 further comprising the step of contacting the transferred cells of step (c) with sufficient cycloheximide to substantially reduce or eliminate de novo protein synthesis,

13. A viral vector under the transcriptional control of an exogenous inducer in which a heterologous DNA sequence has been inserted in series with a linker sequence and an RNA coactivator sequence, wherein transfection and induction of transcription with said inducer produces an RNA transcript comprising the RNA sequences for the inserted DNA, the linker and the RNA coactivator in series.

14. The viral vector of claim 13 , wherein said RNA coactivator is SRA.

15. The viral vector of claim 13, wherein at least one stop codon is placed in- frame between the start of the transcription site and the 5' end of the heterologous DNA insert.

16. The viral vector of claim 13 , wherein said viral vector is selected from a group consisting of retroviral, adenoviral, and vaccinia viral vectors.

17. A method for the detection of RNA transcript that is capable of transactivating a gene or DNA construct, comprising the following steps:

(a) inserting a first DNA sequence coding for said RNA transcript into a first expression vector transcriptionally controlled by an inducer, wherein said inserted DNA sequence is in series with vector DNA coding for a linker sequence and SRA and wherein expression of the first construct produces an RNA sequence consisting of RNA transcript linked in series to a linker sequence and a SRA;

(b) inserting a second DNA into a second expression vector sequence coding for an SRC-1 mutant, wherein said mutant comprises the SRA-binding domain of SRC-1 but lacks the ability to transactivate a heterologous promoter, wherein said DNA sequence is in frame with DNA coding for the GAL4 DNA binding domain and a reporter and wherein expression and translation of this second construct produces a fusion protein comprised of the SRC-1 mutant and the GAL4 DNA binding domain and wherein said fusion protein is capable of binding to a GAL4-binding DNA element;

(c) transfecting cells with the vector constructed in step (a) and the vector constructed in step (b);

(d) transcribing and translating the vector constructed in step (b);

(e) inducing the vector constructed in step (a) with an appropriate inducing agent to promote transcription of the RNA transcript; and

(f) measuring the reporter message.

18. The method of claim 17, further comprising the step of contacting the transfected cells in step (c) with sufficient cycloheximide to substantially reduce or eliminate de novo protein synthesis.

19. A method for determining the presence of a specific RNA transcript comprising the steps of:

(a) incubating a cell extract independently with each of the following:

(1) no additional agents;

(2) RNase Tl at a concentration sufficient to substantially reduce or eliminate single- stranded RNA;

(3) RNase H at a concentration sufficient to substantially cleave RNA in RNADNA hybrids;

(4) multiple concentrations of deoxynucleotides antisense to said RNA transcript followed by RNase H at a concentration sufficient to substantially cleave RNA in RNADNA hybrids;

(5) deoxyoligonucleotides sense to such RNA transcript followed by RNase H at a concentration sufficient to substantially cleave RNA in RNADNA hybrids;

(b) isolating total RNA from each of said extracts in step (a);

(c) generating cDNA from each of said total RNA;

(d) amplifying selected cDNA products using RT-PCR with PCR primers specific to a region of said RNA transcript that contains the region which hybridizes to said antisense deoxynucleotides; and

(e) measuring the selected amplified PCR products.

20. The method of claim 19, wherein said amplified PCR products are detected by size fractionation.

21. The method of claim 20 wherein the RNA transcript is a SRA transcript.

22. The method of claim 21 , wherein said amplified PCR products are detected by size fractionation.

23. A kit for the detection of SRA transcripts comprising the following components:

1) RNase Tl;

2) RNase H;

3) antisense deoxyoligonucleotides to SRA;

4) sense deoxyoligonucleotides to SRA; and

5) sense and antisense primers to SRA for RT-PCR amplification.

24. The kit of claim 23, wherein said antisense deoxyoligonucleotides are derived for the sequence of SRA.

25. The kit of claim 24, wherein said antisense deoxyoligonucleotide to SRA is SEQ. ID. No. 4, said sense deoxyoligonucleotides to SRA is SEQ. ID. No. 5, said sense primer to SRA is SEQ. ID. No. 6, and antisense primer to SRA is SEQ. ID. No. 7.

26. A method for the evaluation of type I or "classical" nuclear receptor activity comprising the step of measuring SRA transcript.

27. The method of claim 26, wherein said evaluation comprises in situ hybridization with specific oligonucleotide probes to SRA.

28. The method of claim 26, wherein said evaluation comprises RT-PCR amplification of cDNA products with selective primers for SRA.

29. A method for modulating type I or "classical" nuclear receptor activation of target genes comprising the step of increasing or decreasing the availability of SRA, wherein increasing SRA results in increased nuclear receptor activity and decreasing SRA results in decreased nuclear receptor activity.

30. A method for the treatment of hormone-dependent tumors comprising the step of selective reduction of RNA coactivator availability.

31. The method of claim 30, wherein said RNA coactivators are selected from the group of sequences consisting of SRA-1, SRA-2 and SRA-3.

32. The method of claim 30, wherein said reduction is by introduction of deoxyoligonucleotides complementary to the RNA coactivator.

33. The method of claim 30, wherein said reduction is by the transfection of cells of said tumor with a vector encoding an RNA transcript which is complementary to the RNA coactivator.

34. The method of claim 30, wherein said reduction comprises the contacting of said tumor cells with a chemical agent which inhibits the transcription of RNA coactivators.

35. The method of claim 30, wherein said reduction comprises the contacting of said tumor cells with a chemical agent which promotes the degradation of RNA coactivators.

36. The method of claim 30, wherein said reduced availability results from the introduction of an SRA dominant negative.

37. A method for the creation of a non-human animal wherein SRA is overexpressed for the purpose of increasing responsiveness of a hormone receptor to a hormone.

38. A genetically altered non-human animal in which SRA is overexpressed for the purpose of increasing responsiveness of a hormone receptor to a hormone.

39. The animal of claim 38, wherein said hormone receptor is the growth hormone receptor.

40. A method for the detection of hormone-dependent tumors comprising the measurement of RNA coactivator in a tissue sample, wherein an elevated level of SRA is predictive of a hormone-dependent tumor.

41. The method of claim 40 wherein said RNA coactivator is SRA.

42. The method of claim 40 wherein said tumors are selected from a group consisting of breast tumors, endometrial tumors, and prostate tumors.

43. A method for the prenatal or postnatal detection of cortisol resistance comprising the step of measuring RNA coactivator in a tissue sample, wherein decreased RNA coactivator expression is predictive of cortisol resistance.

44. A kit for the prenatal or postnatal detection of cortisol resistance comprising the following elements: a) RT-PCR primers for SRA of SEQ. ID. No. 6 and SEQ. ID. No. 7; b) normal tissue sample standard; and c) confirmed cortisol resistance tissue sample standard.

45. The kit of claim 44, herein said tissue sample standards comprise immortalized cells of human origin.

46. A method for the evaluation of an agent for the ability to modulate RNA coactivator levels, comprising contacting a cell with said agent and subsequently measuring RNA coactivator transcript levels.

47. The method of claim 46, wherein said RNA coactivator is SRA.

48. The method of claim 46, wherein said cell is selected from a group consisting of fibroblasts, epithelial cells, blood cells, HeLa cells, COS-7 cells, T-47D cells, and MCF-7 cells.

49. The method of claim 46, wherein RNA coactivator transcript levels are measured via activation of a reporter construct.

50. A method for the evaluation of an agent for the ability to affect the half-life of SRA transcript, comprising contacting said transcript cell with said agent and subsequently measuring SRA.

51. The method of 50, wherein said measurement of SRA comprises RT-PCR with primers for SRA.

52. The method of 51, wherein said primers are of the sequence SEQ. ID. No. 6 and SEQ. ID. No. 7.

53. A method for the selective isolation of specific RNA transcripts expressed in an expression vector comprising the steps of:

(a) inserting the cDNA coding for the desired RNA transcript into a vector in any reading frame with an SRA insert such that expression of said vector produces an RNA transcript comprising said desired RNA transcript and SRA in series,

(b) transfecting cells with said vector;

(c) harvesting total RNA from said cells;

(d) incubating said total RNA with SRC-1 protein; and

(e) contacting said total RNA/SRC-1 protein mixture with conjugated antibodies to SRC-1 such that the RNA/SRC-1 /antibody complex is bound to a substance which facilitates isolation of said complex.

54. The method of claim 53, wherein said RNA of interest and SRA can be cleaved by RNase H by incorporation of a specific nucleotide sequence into said vector such that expression of said vector produces an RNA transcript comprising said desired RNA transcript, said specific nuleotide sequence, and SRA in series.

55. The method of claim 53 , wherein said vector contains at least one stop codon in-frame with the inserted RNA between said inserted RNA and SRA.

56. A kit for the expression and isolation of a specific RNA transcript comprising the following components,

(a) a vector comprising a restriction site for the insertion of a selected cDNA such that expression of said vector produces a single RNA transcript comprising said desired RNA transcript and SRA in series;

(b) an SRC family protein; and

(c) conjugated antibody to SRC-1.

57. The kit of claim 56, wherein said SRC family protein is selected from the group consisting of SRC-1, SRC-2, and SRC-3.

58. The kit of claim 56, wherein said SRC family protein is conjugated to a substance for the purpose of facilitating isolation.

59. The kit of claim 56, wherein said SRC family protein interacts with an antibody for the purpose of facilitating isolation.

60. A method of drug design whereby the SRA/SRC-1 interaction is used as a model to evaluate agents for the ability to modulate hormone receptor activity.

61. A method of designing a therapeutic agent wherein said therapeutic agent alters activation of a promoter, said method comprising: selectively mutating an RNA transcript; contacting said RNA transcript with a mixture comprising a chimeric protein consisting of a coactivator protein fused with the GAL4 DNA binding domain and a reporter construct; and measuring the product of said reporter construct.

62. A cell line which lacks endogenous production of wildtype SRA.

63. An animal which has been genetically engineered such that said animal lacks the ability to produce wildtype SRA.

64. A method of designing a therapeutic agent wherein said therapeutic agent alters coactivation of a steroid hormone receptor, said method comprising: selectively mutating wildtype SRA; introducing a reporter gene construct responsive to said steroid hormone receptor into a cell lacking the ability to produce wildtype SRA; contacting one fraction containing said cell with said SRA mutant transcript; contacting a second fraction containing said cell with SRA wildtype transcript; contacting said cell fractions with a specific steroid receptor ligand; and measuring the production of product from said reporter construct, wherein production of said reporter construct product over basal values indicates coactivation and where said coactivation by SRA mutant can be compared to that of SRA wildtype.

65. The method of claim 64 wherein the mutation is selected from the group consisting of mutation in the core sequence of SRA, mutation in the 5' region upstream of the core sequence of SRA and a mutation in the 3' region downstream of the core sequence of SRA.

66. The method of claim 64, wherein said steroid hormone receptor is selected from the group consisting of androgen receptor (AR), estrogen receptor (ER), glucocorticoid receptor (GR), mineralocorticoid receptor (MR), and progestin receptor (PR).

67. A method of designing a therapeutic agent wherein said therapeutic agent reverses interference by transcriptional activators with common coregulators, said method comprising: selectively mutating an RNA transcript; introducing said RNA transcript into a cell in which a first steroid receptor, a second steroid receptor and a responsive reporter gene construct have been transfected; contacting said transfected cell with receptor specific ligands; and, measuring the product of said reporter construct.

68. A method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to interact with a protein target, said method comprising: selectively mutating and transcribing said wildtype SRA; contacting said protein target with labeled wildtype SRA; washing away unbound labeled wildtype SRA; contacting one fraction containing said SRA-bound protein target with unlabeled said SRA mutant transcript; contacting a second fraction containing said SRA-bound protein target with unlabeled wildtype SRA; and measuring displaced labeled wildtype SRA in each fraction, wherein a greater displacement of labeled SRA wildtype by said SRA mutant compared with unlabeled SRA wildtype indicates a greater binding affinity to said protein target.

69. A method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to interact with a protein target, said method comprising: selectively mutating wildtype SRA (SRA mutant) and generating labeled transcript; contacting one fraction containing a target protein with said SRA mutant; contacting a second fraction containing said target protein with labeled SRA wildtype; isolating said RNA/target protein complexes; and measuring R A transcript label in each of said complexes, wherein an increase in retained label in mutant SRA complex compared to wildtype SRA complex indicates a greater binding affinity to said protein target.

70. A method of designing a therapeutic agent wherein wildtype SRA is mutated and subsequently evaluated for the ability to coactivate individual steroid hormone receptors, said method comprising: selectively mutating wildtype SRA (mutant SRA); introducing a responsive reporter construct for steroid a hormone receptor into a group of cells; contacting one fraction containing said cells with said mutant SRA; contacting a second fraction containing said cells with wildtype SRA; contacting said cells with a ligand specific for said steroid hormone receptor; and measuring the production of product from said reporter construct, wherein production of said reporter construct product over basal values indicates coactivation and where said coactivation by SRA mutant can be compared to that of SRA wildtype.

71. A method for the expression of a protein wherein the RNA transcript acts as a coactivator, said method comprising the steps of: inserting a gene of interest into a vector, wherein expression of said vector results in an RNA transcript comprising SRA, an internal ribosomal entry site, and the mRNA for said protein in series; and inducing expression of said vector.

72. The method of claim 40 wherein said tumor is an ovarian tumor.

73. A method for the diagnosis of hormone-dependent tumors comprising the measurement of RNA coactivator in a tissue sample, wherein an elevated level of SRA indicates a hormone-dependent tumor.

74. The method of claim 73 wherein said RNA coactivator is SRA.

75. The method of claim 73 wherein said tumors are selected from a group consisting of breast tumors, endometrial tumors, ovarian tumors, and prostate tumors.

76. The method of claim 30, wherein said hormone-dependent tumors are resistant to hormone or anti-hormone therapies.

77. A method for the detection of expression patterns of SRA levels wherein said patterns are predictive of expression patterns of steroid receptors.