WO1989009223A1

WO1989009223A1 - Dna binding proteins including androgen receptor

Info

Publication number: WO1989009223A1
Application number: PCT/US1989/001238
Authority: WO
Inventors: Shutsung Liao; Chawnshang Chang
Original assignee: Arch Development Corporation
Priority date: 1988-03-30
Filing date: 1989-03-24
Publication date: 1989-10-05
Also published as: EP0407462A1; EP0407462A4; JPH03504438A; CA1341305C

Abstract

Disclosed are DNA sequences encoding DNA binding polypeptides including androgen receptor (AR) and TR2 polypeptides. Illustratively, human and rat AR-cDNA have 79 kD and 98 kD polypeptide expression products which are immunoprecipitable by human auto-immune anti-androgen receptor antibodies and are capable of binding androgens specifically and with high affinity. Also disclosed are antibodies and immunological methods and materials for detection of androgen receptor and TR2 polypeptides and hybridization methods and materials for detection of AR and TR2-related nucleic acids.

Description

"DNA BINDING PROTEINS INCLUDING ANDROGEN RECEPTOR"

BACKGROUND OF THE INVENTION

The present invention relates generally to DNA binding regulatory proteins and more particularly to DNA sequences encoding androgen receptor protein and novel DNA binding proteins designated TR2, to the polypeptide products of recombinant expression of these DNA sequences_/ to peptides whose sequences are based on amino acid sequences deduced from these DNA sequences/ to antibodies specific for such proteins and peptides, and to procedures for detection and quantification of such proteins and nucleic acids related thereto.

There are five major classes of steroid hor¬ mones: progestins, glucocorticoids, mineralocorticoids, androgens_/ and estrogens. Receptor proteins/ each specific for a steroid hormone/ are distributed in a tissue specific fashion and in target cells, steroid hormones can form specific complexes with corresponding intracellular receptors. [Jensen, et al./ Proc. Nat'l. Acad. Sci. (USA), 5_9_:632 (1968); Gorski, et al., Ann. Rev. Phvsiol., 3_8:425-450 (1976); and LiaO/ et al._/ page 633 in Biochemistry of Hormones/ H. .J. Makin, ed. (Blackwell Sci. Publ. Oxford, 1984)]. The hormonal regulation of gene expression appears to involve inter¬ action of steroid receptor complexes with certain seg¬ ments of genomes and modulation of specific gene trans- cription. See, e.g., Ringold, Ann. Rev. Pharmacol. Toxicol♦ , 2_5:529 (1985); and Yamamoto, Ann. Rev. Genet., 19_:209 (1985). Many of the primary effects of hormones involve increased transcription of a subset of genes in specific cell types.

The successful cloning of e.g., cDNAs coding for various steroid receptors has allowed the structural and functional analysis of different steroid receptor domains involved in steroid and DNA binding. See, e.g., Hollenberg, et al.. Nature (London), 3_18:635 (1985); Miesfeld, et al.. Cell, 6:389 (1986); Danielsen, et al., EMBO J. , 5_:2513 (1986); Greene, et al.. Science, 231:1150 (1986); Green, et al.. Nature (London), 320:134 (1986); Krust, et al., EMBO J., 5_:891 (1986); Loosfelt, et al., Proc. Nat'l. Acad. Sci. (USA), 83_:9045 (1986); Conneely, et al. , Science, 233:767 (1987); Law, et al., Proc. Nat'l. Acad. Sci. (USA)/ 8_4:2877 (1987); Misrahi, et al., Biochem. Biophys. Res. Commun., 143:740 (1987); Arriza, et al.. Science, 237:268 (1987); Sap, et al., Nature (London), 3_24_:635 (1986); Weinberger, et al.. Nature (London), 318:641 (1986); Benbrook, et al.. Science, 238:788 (1987); and Evans, Science, 240:889. (1988).

Androgens, such as testosterone, are respons- ible for the development of male secondary sex charac¬ teristics and are synthesized primarily in testis. Cloning of a cDNA for androgen receptor (AR) has been difficult because, until recently, monospecific antibodies against AR have not been available for screening cDNA libraries. An abstract by Govindan, et al., J. Endocrinol. Invest., 10 (Suppl. 2 ) (1987), reported the isolation of cDNA clones encoding human androgen receptor isolated from a human testis λgt-11 cDNA library using synthetic oligonucleotides homologous to human glucocorticoid, estradiol, and progesterone receptors as probes. The expressed protein reportedly bound tritium-labelled DHT (dihydrotestosterone) with high affinity and specificity. However, no nucleotide or amino acid sequence analysis was provided for full length androgen receptors, nor was any description provided concerning isolation of the full length puta¬ tive androgen receptor clones.

Recently, Chang, C, et al.. Science, 240:324 (April 15, 1988), co-authored by the inventors herein, described cDNAs encoding androgen receptors obtained from human testis and rat ventral prostrate cDNA libraries. These cDNAs for human and rat androgen receptor were reported to be long enough to code for 94 kDa and 76 kDa receptors. The molecular weights were derived with the assistance of a software program known as: DNA Inspector II (Textco West Lebanon, New

Hampshire) open reading frame analysis. With a new DNA Inspector lie program, hAR (918 amino acids) has an estimated molecular weight 98,608 and rAR (902 amino acids) has a molecular weight of 98,133. Therefore, the reported "94 kDa" AR is now termed "98 kDa" AR; and the hAR or rAR polypeptides, from the second ATG/Met, reported as "76 kDa" are now termed "79 kDa". See also, Chang, C, ej al., Proc. Nat'l. Acad. Sci. (USA), 85:7211^' (October 5, 1988) also co-authored by the inventors herein.

In contrast, Lubahn, D., et al.. Science, 240:327 (1988), using libraries from human epididymis and cultured human foreskin fibroblasts obtained a human cDNA which was expressed in monkey kidney (COS) cells to yield a protein, present in the cytosol, capable of binding androgens. This cDNA, however, was only sufficient to code for a receptor having an estimated molecular weight of 41,000. Therefore, the cDNA obtained only coded for a portion of AR. Of interest to the present invention is Young, et al. , Endocrinol. , 123:601 (1988), wherein the production of anti-AR monoclonal antibodies was reported. Anti-AR autoantibodies were identified in the sera of prostate cancer patients, as described in Liao, S., et al., Proc. Nat'l. Acad. Sci. (USA), 82:8345 (1984) (one of the co-inventors herein), and were characterized with respect to their titer, affinity, and specificity. Subsequently, lymphocytes from the blood of those patients having high antibody titers were isolated, transformed with Epstein-Barr Virus (EBV), and cloned for anti-AR monoclonal antibody production. These monoclonal antibodies were found to interact with androgen receptors from rat prostate. An attempt to scale-up antibody production resulted in a decline of antibody secretion. It is not uncommon for transformed B-cells to be more unstable than hybridoma cells. Kozbor, et al. , Eur. J. Immunol./ 14, 23 (1984). Because of the instability associated with such cell lines, an alternate source of monoclonal antibodies is preferred.

There thus exists a need in the art for information concerning the primary structural conformation of androgen receptor protein and other DNA binding proteins such as might be provided by knowledge of human and other mammalian DNA sequences encoding the same. Availability of such DNA sequences would make possible the application of recombinant methods to the large scale production of the proteins in procaryotic and eukaryotic host cells, as well as DNA-DNA, DNA-RNA, and RNA-RNA, hybridization procedures for the detection, quantification and/or isolation of nucleic acids asso¬ ciated with the proteins. Possession of androgen receptor and related DNA-binding proteins and/or know- ledge of the amino acid sequences of the same would make possible, in turn, the development of monoclonal and polyclonal antibodies thereto (including antibodies to protein fragments or synthetic peptides modeled thereon) for the use in immunological methods for the detection and quantification of the proteins in fluid and tissue samples as well as for tissue specific delivery of sub¬ stances such as labels and therapeutic agents to cells expressing the proteins.

BRIEF SUMMARY OF THE INVENTION

The present invention provides novel purified and isolated DNA sequences encoding androgen receptor protein and a structurally related protein, designated TR2 protein, which also has DNA binding (and hence DNA replication or transcription regulatory) capacity. In presently preferred forms, novel DNA sequences comprise cDNA sequences encoding human and rat androgen receptor and human TR2 protein. Alternate DNA forms such as genomic DNA, and DNA prepared by partial or total chemical synthesis from nucleotides as well as DNA with deletions or mutations, is also within the contemplation of the invention.

Association of DNA sequences provided by the invention with homologous or heterologous species expression control DNA sequences, such as promoters, operators, regulators and the like, allows for iri vivo and ^n vitro transcription to form messenger RNA which, in turn, is susceptible to translation to provide androgen receptor and TR2 proteins, and related poly- and oligo-peptides in large quantities. In a presently preferred DNA expression system of the invention, AR and TR2 encoding DNA is operatively associated with a viral (T7) regulatory (promoter) DNA sequence allowing for in vitro transcription and translation in a cell free system to provide, e.g., a 79 kD and 98 kD human androgen receptor (hAR) protein, 79 kD and 98 kD rat androgen receptor (rAR) protein and smaller forms of these proteins, as well as TR2 protein, including 20 kD and 52 kD species.

Incorporation of DNA sequences into procaryotic and eucaryotic host cells by standard transformation and transfection processes, potentially involving suitable viral and circular DNA plasmid vectors, is also within the contemplation of the inven¬ tion and is expected to provide useful proteins in quan- tities heretofore unavailable from natural sources.

Systems provided by the invention included transformed E. coli DH5α cells, deposited January 25, 1989, with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 in accordance with the U.S. Patent and Trademark Office's requirements for microorganism deposits, and designated EC-hAR3600 under A.T.C.C. Accession No. 67879; EC-rAR 2830, A.T.C.C. No. 67878; EC-TR2-5, A.T.C.C. No. 67877; and EC TR2-7, A.T.C.C. No. 67876. Use of mammalian host cells is expected to provide for such post-translational modi¬ fications (e.g., truncation, glycosylation, and tyrosine, serine or threonine phosphorylation) as may be needed to confer optimal biological activity on recombinant expression products of the invention. Novel protein products of the invention include polypeptides having the primary structural con¬ formation (i.e., amino acid sequence) of AR and TR2 proteins as well as peptide fragments thereof and syn¬ thetic peptides assembled to be duplicative of amino acid sequences thereof. Proteins, protein fragments, and synthetic peptides of the invention are projected to have numerous uses including therapeutic, diagnostic and prognostic uses and will provide the basis for prepara¬ tion of monoclonal and polyclonal antibodies specifi- cally immunoreactive with AR and TR2 proteins.

Preferred protein fragments and synthetic peptides include those duplicating regions of AR and TR2 proteins which are not involved in DNA binding functions and the most preferred are those which share at least one anti¬ genic epitope with AR and TR2 proteins. Also provided by the present invention are polyclonal and monoclonal antibodies characterized by their ability to bind with high immunospecif city to AR and TR2 proteins and to their fragments and peptides, recognizing unique epitopes which are not common to other proteins especially DNA binding proteins.

Illustratively provided according to the present invention are monoclonal antibodies, designated AN1-6, AN1-7, AN1-15; and produced by hybridoma cell lines designated H-AN1-6, H-AN1-7, H-AN1-15; deposited January 25, 1989, under Accession Nos. HB 10,000;

HB 9,999; and HB 10,001, respectively, with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 in accordance with the U.S. Patent and Trademark Office's requirements for ^{'""""" ~} microorganism deposits. These antibodies are characterized by (a) capacity to bind androgen receptors from rat ventral prostate and synthetic peptides having sequences predicted from the structure of hAR-cDNA and rAR-cDNA; (b) specific immunological reactivity with, and capacity to reversibly immunobind to, naturally occurring and recombinant androgen receptors, in native and denatured conformations; and (c) specific immunological reactivity with, and capacity to reversibly immunobind to, proteinaceous materials including all or a substantially, immunologically significant, part of an amino acid sequence duplicative of that extant at residues 331 through 577 of hAR and corresponding amino acid sequences in rAR.

The monoclonal antibodies of the invention can be used for affinity purification of AR from human or rat prostate, and other sources such as AR-rich organs and cultured cells. Also provided by the present invention are novel procedures for the detection and/or quantification of normal, abnormal, or mutated forms of AR and TR2, as well as nucleic acids (e.g., DNA and mRNA) associated therewith. Illustratively, antibodies of the invention may be employed in known immunological procedures for quantitative detection of AR and TR2 proteins in fluid and tissue samples, of DNA sequences of the invention (particularly those having sequences encoding DNA binding proteins) that may be suitably labelled and employed for quantitative detection of mRNA encoding these proteins.

Among the multiple aspects of the present invention, therefore, is the provision of (a) novel AR and TR2-encoding DNA sequences set out in Figure 3, as well as (b) AR and TR2-encoding DNA sequences which hybridize thereto under hybridization conditions of the stringency equal to or greater than the conditions described herein and employed in the initial isolation of cDNAs of the invention, and (c) DNA sequences encoding the same allelic variant, or analog AR and TR2 polypeptides through use of, at least in part, degen¬ erate codons. Correspondingly provided are viral or circular "plasmid DNA vectors incorporating such DNA sequences and procaryotic and eucaryotic host cells transformed or transfected with such DNA sequences and vectors as well as novel methods for the recombinant production of AR and TR2 proteins through cultured growth of such hosts and isolation of these proteins from the hosts or their culture media.

Preferred polypeptide products of the inven¬ tion include the approximately 79 kD (starting from the second ATG/Met) and 98 kD (starting from the first ATG/Met) hAR polypeptides having the deduced amino acid sequence of 734 and 918 residues, respectively, as set out in Figure 3. Also preferred are the 79 kD and 98 kD - 9 -

rAR species polypeptides having the deduced sequence of 733 and 902 residues set out in Figure 3 and the 20 kD and 52 kD species human TR2 polypeptides having the same deduced amino acid sequence of 184 and 483 residues set out in Figure 4. The preferred 79 kD and 98 kD hAR and rAR polypeptides may be produced ii vitro and are characterized by a capacity to specifically bind androgens with high specificity and by their im unopre- cipitatability by human auto-immune anti-androgen receptor antibodies.

Other aspects and advantages of the present invention will be apparent upon consideration of the following detailed description thereof which includes numerous illustrative examples of the practice of the invention, reference being made to the drawing wherein:

Figure 1 illustrates the strategy employed in construction of a human androgen receptor cDNA vector;

Figure 2 illustrates the strategy employed in construction of rat androgen receptor cDNA vectors; Figure 3 provides a 3715 base pair nucleotide sequence for a human androgen receptor (hAR) DNA clone and the deduced sequence of 734 and 918 amino acid residues for hAR proteins; and in addition provides a 3218 base pair nucleotide sequence for a rat androgen receptor (rAR) DNA clone and the deduced sequences of 733 and 902 amino acids for two rAR species;

Figure 4 provides a 2029 base pair nucleotide sequence for a human TR2 DNA clone^and a deduced sequence of 483 amino acids for a "TR2-5" species and a deduced sequence of 184 amino acids for a "TR2-7" species; and

Figure 5 provides an amino acid sequence alignment of the cysteine-rich DNA binding domain of human androgen receptor, glucocorticoid receptor, ineralocorticoid receptor, progesterone receptor, estrogen receptor, TR2, rat AR, chick vitamin D receptor (c-VDR), and the v-erb A oncogene product of avian erythroblastosis virus.

Figures 6, 7, and 8 illustrate, respectively, the in-frame fusion of three different parts of the AR gene (the N-terminal, the DNA-binding domain and the androgen-binding domain) to the N-terminal half of the trpE gene using pATH expression vectors.

DETAILED DESCRIPTION

The following examples illustrate practice of the invention. Example 1 relates to the isolation, preparation, and partial structural analysis of cDNA for human and rat androgen receptors. Example 2 relates to confirmation of the presence on the human X-chromosome of an AR-type cDNA sequence. Example 3 relates to the preparation of human and rat cDNAs containing AR-type cDNA from different clones and ligation into the pGEM-3Z plasmid. Example 4 relates to transcription and trans- lation of the AR-type cDNA plasmid DNA. Example 5 relates to steroid binding activity of the expression product of Example 4. Example 6 relates to the binding activity of the expression product of Example 4 to human auto-antibodies. Example 7 relates to the characteriza- tion of TR2-CDNA. Example 8 relates to the Ln vitro transcription and translation of TR2-CDNA. Example 9 relates to the binding activity of TR2-CDNA expression product. Example 10 relate to the androgen regulation of TR2 mRNA levels in the rat ventral prostate. Example 11 relates to recombinant expression systems of the invention. Example 12 relates to the production of fusion proteins and their use in producing polyclonal and monoclonal antibodies according to the invention. Example 13 relates to use of DNA probes of the inventions. Example 14 relates to development of transgenic animals by means of DNA sequences of the invention. These examples are for illustrative purposes only and are not intended in any way to limit the scope of the invention.

EXAMPLE 1

Preparation and Partial Structural Analysis of cDNA for Human and Rat Androgen Receptors

The isolation of cDNA for human androgen receptor (hAR) and rat androgen receptor (rAR) was accomplished using λGTll cDNA libraries. The human testis and prostate λGTll libraries were obtained from Clontech Co., Palo Alto, California and a rat ventral prostate λGTll library in E.coli Y1090 was constructed as described in Chang, et al., J. Biol. Chem., 262:11901 (1987). In general, clones were differentiated using oligonucleotide probes specific for various steroid receptors. The cDNA libraries were initially screened with a set of 41-bp oligonucleotide probes designed for homology to nucleotide sequences in the DNA-binding domain of glucocorticoid receptors (GR) , estrogen recep¬ tors (ER), progesterone receptors (PR), mineralocorti- coid receptors (MR), and the v-erb A oncogene product of avian erythroblastosis virus. The set of probes had the following sequence: TGTGGAAGCTGT/CAAAGTC/ATTCTTTAAAAGG/ AGCAA/GTGGAAGG.

The plaques were replicated on a nitrocellu¬ lose filter and screened with a 5'-end ³²P-labeled 41-bp oligonucleotide probes. The conditions of hybridization were 25% formamide, 5X Denhardt's solution (0.1% Ficoll 400, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin), 0.1% SDS, 5X SSC (IX SSC is 150 mM NaCl, 15 mM sodium citrate), 100 yg/ml denatured salmon sperm DNA, and 1 μg/ml poly(A) at 30^βC. Filters were washed with a solution containing 0.1% SDS, 0.05% sodium pyrophosphate and 0.4X SSC at 37°C. A less stringent hybridization condition (2X SSC at 37°C) was used for the first screen employing the 41 bp probes. The remaining clones were then probed again at more stringent conditions by reducing the con- centration of SSC, eventually to 0.4X SSC at 37°C, or by increasing the temperature, or by increasing the concen¬ tration of formamide. In some procedures, 5X SSC, 8% dextran sulfate, and 20% formamide, at 42°C was employed and the result was equivalent to that obtained with 0.6X SSC.

From approximately 3 X 10° human testis recom- binants and 6 X 10⁵ rat ventral prostate recombinants, 302 and 21 positive clones, respectively, were obtained. Based on the assumption that AR might have a cysteine-rich DNA binding domain highly homologous to the DNA-binding regions of other steroid receptors, positive clones from the first screenings were probed with 5'-end ³^P-labeled 24-bp oligonucleotides specific for the various steroid receptors for the possible presence of cDNA for AR through a process of elimina¬ tion. The GR-cDNA clones were eliminated by screening with two GR-specific 24-bp probes that had nucleotide sequences identical to nucleotide segments immediately f next to the 5'-end or the 3'-end of the DNA binding- region of hGR-cDNA , i.e., TGTAAGCTCTCCTCCATCCAGCTC and CAGCAGGCCACTACAGGAGTCTCA. 244 and 14 clones, respec_? tively, were eliminated as hGR- and rGR-cDNA clones.

Using similar procedures involving four 24-bp probes for the 5'-end of PR(CCGGATTCAGAAA/GCCAGT/-

CCAGAGC) and two 24-bp probes for the 3'-end of ER(GCA/- CGACCAGATGGTCAGTGCCTTG)/ no ER- or PR-cDNA clones were detected in the human testis library. In the rat pros¬ tate library, no ER-cDNA clones were detected but one positive clone was obtained with hPR-specific 24 bp probes. Following this process of eliminating clones putatively encoding other steroid receptors, the DNA inserts in the remaining clones were analyzed by restriction mapping and subcloned into M13 vectors for di-deoxy sequence analysis. See, Chang, et al., J. Biol. Chem., 262:2826 (1987). Nucleotide sequence analysis allowed four clones to be identified as hMR- cDNA clones.

Through this stepwise process of elimination, 54 human testis clones and 6 rat prostate clones were selected and were then categorized into two groups: 30 human testis clones had sequences overlapping to form a 2.1 kb cDNA; and 24 human testis and 6 rat prostate clones had sequences overlapping to form a cDNA of about 2.7 kb. The two groups of cDNA were designated, respec¬ tively, as "TR2-type" and "AR-type" cDNA.

EXAMPLE 2

Confirmation of the Presence on the Human X-Chromosome of an AR-type cDNA Sequence Rather than a TR2-type cDNA Sequence

The length between the putative polyadenylat- ion signa'l (AATAAA) and the 5'-end in the "TR-2 type" cDNA is only 2.0 kb, which is considerably shorter than that for the cDNA of other steroid receptors. There¬ fore, it was suspected that the "AR-type" cDNA, rather than the "TR2-type" cDNA, encoded androgen receptor. To obtain additional information, a human X-chromosome library prepared according to Kunkel, et al.. Nucleic

Acids Research, 11:7961 (1983) was probed with the TR2- type cDNA and AR-type cDNA of Example 1. With TR2-type cDNA fragments, no positive clones were detected, while 3 positive clones were obtained with a 1.9 kb fragment of AR-type cDNA from a human testis (clone AR 132), thereby confirming the presence of an AR-type cDNA sequence on the human X-chromosome. Because the X- chromosome has been implicated as the chromosome which contains an AR gene [Lyon, et al.. Nature (London), 2_27_:1217- (1970); Meyer, et al., Proc. Nat'l. Acad. Sci. (USA), 2i^:1469 (1975); and Amrhein, et al., Proc. Nat'l. Acad. Sci. (USA), 7_3:891 (1976)], this information suggested that "AR-type" cDNA, but probably not the "TR2-type" cDNA, contained the DNA sequence that could encode for androgen receptor. Two human clones containing DNA inserts that overlapped to form a 2.7 kb cDNA were designated AR 132 and AR 5. Two rat clones containing DNA inserts that overlapped to form a 2.8 kb cDNA were designated rAR 1 and rAR 4. After restriction enzyme digestion, the DNA segments from these AR-type clones were ligated, selected and amplified using pBR322 and pGEM-3Z vectors as described in Example 3 below.

EXAMPLE 3

A. Preparation of a Human cDNA

Containing AR-type cDNA from Two Different Clones and Ligation Into the Cloning Vector pGEM-3Z Plasmid

Figure 1 relates to the strategy employed in the construction of a full length hAR-cDNA clone. cDNA of clone AR 132 was digested with Eco RI to obtain a 1.9 kb fragment which was then digested with Kpn I to pro¬ vide a 1 kb Eco RI-Kpn I fragment. This 1 kb fragment was ligated to a 3 kb fragment obtained by digestion of clone AR 5 with Kpn I and Pvu I. The resulting 4 kb fragment was inserted into Eco RI and Pvu I-digested pBR322 vector and used to infect E. coli DH5α. The transformed clones were selected by tetracycline-resis- tance. The plasmid with the DNA insert was digested with Cla I and Nde I to obtain a 2.6 kb fragment. The fragment was blunt-ended with the Klenow fragment of E^ coli DNA polymerase I and ligated to the cloning vector pGEM-3Z' plasmid DNA (Promega Biotec, Madison WI.) which was previously blunt-ended by digestion with Sma I. E. coli DH5 α cells were transformed with the plasmid so formed (designated plasmid PhAR3600) and colonies containing the plasmid were selected by ampicillin resistance and amplified. E. coli DH5α cells, transformed with plasmid PhAR3600, were designated EC-hAR3600 and were deposited with the American Type Culture Collection, 12301 Parklawn Drive, Rockville,

Maryland 20852 on January 25, 1989 under Accession No. 67879.

The plasmid DNA was isolated and its structure analyzed by restriction enzyme mapping and sequencing. The 2.0 kb hAR fragment obtained by NruI-BamHI digestion of a 2.6 kb hAR in pGEM3Z was then ligated to another 1.6 kb ECORI-NruI fragment of hHR to obtain the full length 3715 bp hAR. The open reading frame is about 2.8 kb which is sufficient to code for a protein with^" more than 900 amino acids. Near the middle of the protein is a cysteine-rich region with a 72 amino acid sequence highly homologous to regions in other steroid receptors considered to be the DNA binding domain.

As set out in detail below and illustrated in Figure 2, formation of "full length" rat AR clones by slightly .varying procedures results in constructions providing RNA transcripts translatable to 79 kD and 98 kD protein products.

B. Preparation of a Rat

2.7 kb cDNA and Ligation Into the Cloning Vector PGEM-3Z Plasmid

The 2.4 kb Eco RI-Eco RI cDNA insert of clone rAR 1 was digested with Xmn I to obtain a 2.3 k b fragment. This 2.3 kb Xmn I-EcoR I fragment was ligated to a 400 bp fragment that was obtained by digestion of another cDNA clone insert (Eco RI-Eco RI insert of rAR 4) with Pst I. The ligated 2.7 kb fragment was inserted into Sma I and Pst I-digested pGEM-3Z vector and used to infect E. coli DH5o. The E_^ coli DH5α cells were transformed with the plasmid and colonies containing the plasmid were selected by ampicillin resistance and amplified. These cells were designated EC-rAR 2830 and were deposited with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 on January 25, 1989 under Accession No. 67878. As noted in Figure 2, this construction allowed for a transcription product translated beginning with the second of two in-frame methionine-specifying codons (designated ATG₂).

C. Preparation of a Rat 2.83 kb cDNA Ligation Into the Cloning Vector PGEM-3Z Plasmid

The 2.4 kb Eco RI-Eco RI cDNA insert of rAR 1 was digested with Hind III to obtain a 1.68 kb frag¬ ment. The 1.68 kb Eco RI-Hind III fragment was ligated to a 1.15 kb DNA fragment obtained by digestion of another cDNA clone insert (rAR 6) with Hind III and Pst I. The ligated 2.83 kb fragment was inserted into Eco RI and Pst I-digested pGEM 3Z vector and used to infect E. coli DH5α. E. coli (DH5α) cells were transformed with the plasmid and colonies containing the plasmid were selected by ampicillin resistance and amplified. As noted in Figure 2, this construction allowed for a transcription product translated beginning at the first of two in-frame methionine-specifying codons (designated ATG_χ) .

Figure 3 provides the nucleotide sequence of the DNA sequence of the longer "full length" rat and human AR clones and includes the deduced amino acid sequences. The first and second methionine-specifying codons are designated at amino acid positions 1 and 170 of rAR and positions 1 and 185 of hAR.

EXAMPLE 4

Transcription and Translation of the Human AR-type cDNA Plasmid in a Rabbit Reticulocyte Lysate System pGEM-3Z vector (20 μg) containing 2.6 kb hAR DNA segment, as described in Example 3, was linearized with restriction enzyme Bam HI, phenol/chloroform extracted, and precipitated with ethanol. The linearized plasmid was transcribed in a reaction mixture containing 40 mM Tris-HCl, pH 7.5, 6 mM MgCl₂, 2 mM spermidine, 10 mM NaCl, 10 mM DTT, 500 μM each of ATP, GTP, CTP, and UTP, 160 units ribonuclease inhibitor, 5 μg plasmid, 30 units T7 RNA polymerase (Promega Biotec, Madison, WI) and diethylpyrocarbonate (DEPC)- treated water to a final volume of 100 ul. T7 RNA polymerase was used in the transcription of the plasmid DNA, because a T7 promotor, rather than the SP6 promotor, was found ahead of the 5'-end of the ligated AR-cDNA.

The reaction was allowed to proceed for 2 hrs. at 40^βC. RQ1 DNase I (5 units) was added and the reaction continued for 15 mins. at 40°C. The reaction mixture was extracted with phenol/chloroform (1:1) and then with chloroform. RNA product was precipitated by the addition of 0.1 volume of 3 M Na-acetate and 2.5 volumes of ethanol, re-suspended in 0.5 M NaCl, and re- precipitated with 2.5 volumes of ethanol. RNA trans¬ cribed was isolated and then translated in a rabbit reticulocyte lysate system.

Translation of RNA was carried out in a micro- coccal nuclease-treated rabbit reticulocyte lysate

(Promega Biotec, Madison, WI) pre-mixed kit (100 μl) in the presence of 8 μg mRNA, 40 μCi of [³⁵S] methionine (800 Ci/mmol; Amersham Co. , Arlington Heights, IL) and 100 μM each of amino acid mixture without methionine. The reaction was allowed to proceed for 1 hour at 30°C. To quantitate the incorporation of radioactive methionine, 3 μl of the reaction mixture were added to 1 ml of 1 M NaOH containing 1.5% H2θ₂, 1 mM methionine, and 0.04% bovine serum albumin. The mixture was incu¬ bated for 15 mins. at 37°C to hydrolyze [³⁵S] methionine charged tRNA. The radioactive protein products were precipitated by the addition of 1 ml of 25% tricholoacetic acid and the radioactivity associated with the precipitates was determined.

By SDS-PAGE (8% acrylamide gel) analysis, performed as described in Saltzman, et al., J. Biol. Chem. , 262:432 (1987), it was found that a 79 kD protein comprised more than 85% of the translated products.

EXAMPLE 5

Binding Activity of the 79 kD hAR Protein to a Synthetic Androgen

To study the steroid binding activity of the protein coded for by the cloned cDNA, the reticulocyte lysate of Example 4, containing the newly synthesized protein was incubated with 17β[³H]-methyl-17s-hydroxy- estra-4,9,ll-trien-3-one ([³H] R1881), a potent syn¬ thetic androgen that binds AR with high affinity [Liao, et al., J. Biol. Chem., 248_:6154 (1973)].

Specifically, RNA transcribed from the cloned cDNA, as described in Example 4, was translated in a rabbit reticulocyte lysate system and aliquots of the lysate were then incubated with 5 nM [³H] R1881 (87 Ci/mmol) in the absence or presence of 25 nM, 50 nM, or 250 nM of non-radioactive steroid. The final incubation volume was 100 μl. The radioactive androgen binding was measured by the hydroxylapatite-filter method as des- - 19 -

cribed in Liao, S., et al., J. Steroid Biochem., 20:11 (1984). The result was expressed as a percentage of the radioactivity bound in the control tube (5000 dpm) with¬ out additional non-radioactive steroid and is listed in Table 1.

TABLE 1

Androgen-specific binding of hAR coded by cloned cDNA

Non-radioactive [³H] R1881-bound (% of control) steroid added 25 nM 50 nM 250 nM

R1881 13 10 1

5o-dihydrotestosterone 25 17 6

5s-dihydrotestosterone 89 89 81

178-Estradiol 91 91 86

Progesterone 100 91 92

Dexa ethasone 100 93 93

Hydrocortisone 96 90 90

Testosterone 38 28 Not tested

As shown in Table 1, the active natural andro¬ gen, 17s-hydroxy-5α-androstan-3-one(5α-dihydro-testos- terone) competed well with [³H] R1881 binding, but the inactive 58-isomer did not compete well with [³H] R1881 suggesting that it does not bind tightly to AR. The binding activity was steroid specific; dexamethasone, hydrocortisone, progesterone, and 17s-estradiol did not compete well with the radioactive androgen for binding to the 79 kD protein.

Similar steroid binding specificities have also been observed for rAR coded for by cloned cDNA. Chang, c, et al. , Proc. Nat'l. Acad. Sci. (USA), £5:7211-7215 (1988).

Using the hydroxylapatite filter assay method, it was observed that approximately one molecule of the ³⁵S-labelled 79 kD protein obtained from the lysate bound about one molecule of the tritiated androgen at a saturating concentration of ligand. By Scatchard plot analysis, the apparent dissociation constant was 0.31 nM, which is similar to the binding constant (0.65 nM) reported previously for AR of rat ventral prostate as reported in Schilling, et al.. The Prostate, 5_:581 (1984).

EXAMPLE 6

Binding Activity of the 79 kD Protein to Human Auto-antibodies

It has previously been reported [Liao, et al., Proc. Nat'l. Acad. Sci. (USA), 8J2:8345 (1985)] that some older men with prostate cancers have high titers of auto-immune antibodies to AR in their serum samples. The ability of human auto-antibodies to recognize the 79 kD protein made by the reticulocyte lysate system was therefore studied. The receptor protein made in the lysate system of Example 4 was incubated with [³H] R1881 to allow the formation of radioactive androgen-androgen receptor (A-AR) complexes and was then mixed with serum containing auto-antibodies. Reticulocyte lysate containing translated AR was incubated with [³H] R1881, as described in Example 4, and then incubated again in either the presence of or absence of 5 μl of human male serum containing anti¬ bodies to AR (anti-AR serum) for 4 hrs. at 4°C. Rabbit serum containing anti-human immuglobulins (Anti-IgG) was then added as the second antibody. After 18 hrs. of incubation at 4°C, the mixture was centrifuged and the radioactivity associated with the precipitate was esti¬ mated. Human female serum, not containing anti-AR anti- body, was also used for comparison. The results shown in Table 2 below, indicate a quantitative immunoprecipitation of the radioactive A-AR complexes in the presence of both the high titer human serum and a rabbit anti-human immunoglobulin IgG. By SDS-PAGE, it was also observed that the immunoprecipi- tated protein was the 79 kD protein.

TABLE 2

Anti-human immunoglobulin- dependent precipitation of hAR made by the translation of RNA transcribed from cloned cDNA

Sample incubated Anti-serum Immunoprecipitable with [³H]R1881 addition radioactivity(dpm)

AR coded by cDNA^a None 32

+Anti-AR serum + Anti-IgG 8212 +Female serum + Anti-IgG 430 +Anti-IgG 8 Heated AR° +Anti-AR serum + Anti-IgG 42 BMW-lysate⁽ +Anti-AR serum + Anti-IgG 204

^a 8500 dpm of the radioactive AR complexes made were used. ^D Reticulocyte lysate containing AR was heated at 50°C for 20 mins. to inactivate receptor and release the radioactive androgen bound before the addition of antiserum. ^c Brome Mosaic Virus RNA was used in the reti¬ culocyte lysate translation system instead of RNA trans¬ cribed from cloned cDNA. EXAMPLE 7

Characterization of TR2-CDNA

Of the more than 40 TR2-type human cDNA clones obtained, including the 30 described in Example 1, the clone designated TR2-5 was found to be 2029 base pairs

_* in length as indicated in Figure .4. The open reading frame between the first ATG and terminator TAA can encode 483 amino acids with a calculated molecular weight of 52 kD. The putative DNA binding region is underscored. The putative initiator ATG matched closely with Kozak's concensus sequence for active start codons. [See, Kozak, M. , Nature, 308:241 (1984).] Two triplets upstream of this ATG codon is an in-frame terminator (TAA) further supporting initiator function for the ATG. Eleven out of the 30 TR2-type clones of Example 1, as represented by the clone designated TR2-7, contain an internal 429 bp insertion between nucleotide sequence 669 and 670 (designated by an asterisk in Figure 4). This internal insertion introduces a termination codon TAG (underscored in the insert sequence footnote) which reduces the open reading frame to 184 amino acids with a calculated molecular weight of 20 kD. It is likely that the insertion in these 11 TR2 clones (or deletion in the 19 other TR2 clones) represents either the existence of two types of mRNA in the human testis or an artifact of cDNA construction. In the 3'-nontranslated region, a eukarotic polyadenylation signal AATAAA is present between the nucleotide sequence 2000 and 2007 of the TR2-5 clone.

Other variants of TR-2 with open reading frames at the putative ligand-binding domains have been obtained. Some of these may code for receptors for new hormones or cellular effectors. It is anticipated that the knowledge of TR2-CDNA sequences will be utilized in isolation and structural analysis of other cellular receptors, their genes, and ligands (endogenous or therapeutic agents) that can regulate cellular growth and functions in both normal and diseased organs.

Figure 5 depicts an amino acid sequence align- ment of the cysteine-rich DNA binding domain of human androgen receptor, glucocorticoid receptor, mineralocor- ticoid receptor, progesterone receptor, estrogen recep¬ tor, human TR2 protein, rat AR, chick vitamin D receptor (c-VDR), and the v-erb A oncogene product of avian erythroblastosis virus. The numbers in the left margin represent the positions of amino acid residues in the individual receptors. Common residues are boxed with solid lines. The residues in dotted boxed represent those not in common with those in the solid boxes. V-erb A has two more amino acids at the starred posi¬ tion.

In this region, the human and rat cDNAs for AR have identical amino acid sequences, although for some amino acids different codons are employed. Also in this region, the homology between human AR or rat AR and other receptors is as follows: glucocorticoid receptor (GR), 76.4%; mineralo-corticoid receptors (MR), 76.4%; progesterone receptors (PR), 79.2%; estrogen receptors (ER), 55.6%; TR2, 45.8%; chick vitamin D receptor (c- VDR), 40,3%; and the v-erb A oncogene product of avian erythroblastosis virus, 40.3%. In the putative region for steroid binding, which has about 200 amino acids near the -COOH terminal of steroid receptors, the homology between human AR or rat AR and hGR, hMR, or hPR is about 45-55%, whereas the homology between human AR and rat AR and hER is less than 20%. Thus, human and rat AR appear to be more closely related to GR, MR, and PR, than to v-erb A or to receptors for estrogen, vitamin D, and thyroid hormones. The DNA binding domain of TR2 (amino acids 111 to 183) has a high homology with the steroid receptor super-family as follows: retinoic acid receptor (RAR), [Giguere, et al.. Nature, 33£:624 (1987)], 65%; thyroid receptor (T₃R) [Sap, et al.. Nature, 324:635 (1987)], 59%; mineralocorticoid receptor (MR), [Arriza, et al., Science, 235:268 (1987)], 54%; vitamin D₃ receptor

(VD₃R) [McDonnell, et al. , Science, £35:1214 (1987)], 53%; hERRl and hEER2, [Giguere, V., et al.. Nature, 331:91 (1988)], 51% estrogen receptor (ER), [Hollenberg, et al.. Nature, 318:635 (1985)], 51%; glucocorticoid receptor (GR) [Hollenberg, et al.. Nature, 318:635

(1985)], 50%; androgen receptor (AR), 50%; progesterone receptor (PR), 49%; [Loosfelt, et al., Proc. Nat'l. Acad. Sci., (USA), 8_3_:9045 (1986)]. As noted in Figure 5, the positions of 20 amino acids (9 Cys, 3 Arg, 2 Gly, 2 Phe, 1 Lys, 1 Met, 1 Asp, 1 His) in the putative DNA binding domain are identical among all isolated thyroid steroid receptor genes. It has been proposed that this highly conserved region may be involved in the formation of a DNA binding finger. See, Weinberger, et al., Nature, 3_18:670 (1985). Like the other steroid receptors, TR2 does not have the two extra amino acids (Lys-Asn) found only in the thyroid receptors' DNA binding domain. See, Sap, et al.. Nature, 324:635 (1987).

EXAMPLE 8

In Vitro Transcription and Translation of TR2 cDNA

The Eco RI-Eco RI DNA inserts from clones TR2-

5 and TR2-7 were isolated and ligated to an EcoRI digested pGEM-3Z vector for i i vitro transcription essentialy as described in Example 3. E_* coli DH5α cells, transformed with these plasmids were designated EC TR2-5 and EC TR2-7 and were deposited with the

American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 on January 25, 1989 under Accession Nos. 67877 and 67876.

Transcribed RNA was then translated in a rabbit reticulocyte lysate system. By SDS- polyacrylamide gel electrophoresis (PAGE), it was found that the major translated product of TR2-7, which has an internal 429 bp, insertion, was a 20 kD protein. The major translated product of TR2-5 was a 52 kD protein.

To further characterize these translated pro- teins, the translation lysate was passed over a DNA cellulose column. The bound product was then eluted, concentrated and applied to SDS-PAGE. The results indi¬ cated that the translated proteins were indeed DNA- binding proteins.

EXAMPLE 9

Binding Activity of TR2-5 cDNA Expression Product

To study the steroid binding activity of the translation products of the TR2-5 clone, the products were incubated with all major classes of steroids, including androgens, progesterone, glucocorticoid and estrogen but no significant binding with the above steroids was observed. This does not necessarily rule out a steroid binding function for the protein.

Possibly the TR2-5 expression product steroid binding activity may involve some post-translation modifications missing in the rabbit reticulocyte lysate system.

Alternatively, the TR2-5 translated protein may be steroid independent or may bind to an unidentified ligand present in the human testis or rat ventral prostate.

The size of TR2 mRNA was determined by Northern blot analysis with TR2-5 cDNA insert as a probe. One 2.5 kb band was detected which should include enough sequence information to code for a 52 kD protein. The TR2 mRNA tissue distribution was also analyzed by dot hybridization. The hybridization was visualized by densitometric scanning of the autoradio- graphs, individual dots were cut and radioactivity measured by liquid scintillation counting [See, Chang, et al., J. Biol. Chem., 26_2:2826 (1987)]. The results showed that TR2 mRNA was most abundant in the rat ventral prostate with the relative amounts in other tissues being: prostate.100%, seminal vesicle 92%; testis, 42%; submaxillary gland, 18%; liver, 13%; kidney, <1%; and uterus, <1%.

EXAMPLE 10

Analysis of Androgen Regulation of AR and TR2 mRNA Levels in the Rat Ventral Prostate

Because rat ventral prostate is an androgen- sensitive organ and contains the greatest amount of AR and TR2 mRNA, the effect of androgen depletion and replacement on the mRNA levels was studied by RNA dot hybridization and Northern blot analysis. Total RNA was extracted from the ventral prostate of normal rats, rats castrated and rats previously castrated and treated with

5α-dihydrotestosterone (17s-hydroxy-5α-androstand-3- one) . AR mRNA levels per unit of DNA increased 200 to

#

300% of the level for normal rats within 2 days after castration. Administration of 5α-dihydrotestosterone (5 mg/rat/day) into castrated rats reduced the AR mRNA level to that of normal rats. TR2 mRNA levels, per unit of DNA, were increased to 170% of the normal rat within 2 days after castration. Injection of 5α-dihydrostes- tosterone (5 mg/rat/day) into castrated rats reduced the TR2 mRNA to the levels of normal rats. Interestingly, the total prostate RNA levels, at the same period of time, were decreased to 40% of the normal level. The effects of androgen on the levels of prostatic TR2 mRNA were further confirmed by flutamide injection experi¬ ments. Flutamide, an anti-androgen which antagonizes the effects of 5α-dihydrotestosterone on the ventral prostate weights in castrated rats [Neri, et al..

Invest. Urol., ljO:123 (1972)], was injected into normal rats for from 2 to 6 days. TR2 mRNA levels were then measured by dot hybridization as described above. The results show that flutamide injection, like castration, increased TR2 mRNA levels. The change in the AR or TR2 protein levels could be due to a change in mRNA stab¬ ility and utilization or a change in the regulation of gene transcription. The activation or inactivation by androgen of specific genes to different degrees in the same organ may suggest that androgen is involved in the structuring of the pattern of gene expression in the target cell. Also, if androgen-mediated gene repression mechanisms are related to growth of the prostate, then a further study of the mechanism and structure of genes, repressed AR and TR2 mRNA may provide a better under¬ standing of androgen action in the normal and abnormal prostate and other hormone sensitive organs.

Also, defects in the structures of AR and androgen sensitive genes and/or losses of the control of the production and function of these gene products can be the causes of the abnormal growth of androgen sensi¬ tive or insensitive tumors like prostate cancers. These lines of research may, therefore, be helpful in design¬ ing new diagnostic methods and treatments for patients. EXAMPLE 11

Expression of Cloned AR-Genes and Androgen Sensitive Genes in Eukaryotic and Prokaryotic Cells

The ability of cloned genes to function when introduced into mammalian, yeast, and bacterial cells has proved to be very valuable in understanding the function and regulatory mechanism of genes. Recombinant techniques can provide, in large quantities, gene expression products (proteins) which are not readily obtainable from natural sources. While bacterial systems are very useful in large scale production of those proteins which do not require substantial post- translational modification for optimal biological activity, eukaryotic systems are particularly advantageous because of their ability to correctly modify the expressed proteins to their functional forms.

Using well known techniques, AR-cDNA and TR2- cDNA may readily be used for large scale production of gene products. For this purpose, the most efficient transcription units can be constructed using viral, as well as non-viral, vectors with regulatory signals that can function in a variety of host cells. SV40, pSV2, adenoviruses, and bovine papilloma virus DNA have been used successfully for introduction of many eukaryotic genes into eukaryotic cells and permit their expression in a controlled genetic environment. These and similar systems are expected to be appropriate for the expression of AR- and TR2-genes. To assist gene transfer,* the two most widely used methods, the "calcium phosphate precipitation" and the "DEAE-dextran technique" can be used. Genes can be introduced into cells either transiently, where they continue to express for up to 3 days, or, more permanently to form stably transformed cell-lines. The expressed proteins can be detected by androgen binding or antibody assays. The expression of cloned AR-genes was achieved as follows in a eukaryotic system. NIH 3T3 cells, contact-inhibited cells established from NIH Swiss mouse embryo, were co-transfected with hAR cDNA inserted into pBPVMTH vectors as described by Gorman, "DNA Cloning", 2:143-190 D. M. Glover, ed.; (Oxford, Washington, D.C. 1985). Transfected cells were cloned and grown in multiple-well cell culture plates. About 100 individual cell lines were isolated. Of these, 6 demonstrated [³H] R1881-binding activity at least 4-fold the activity of cells transfected with pSV2 vector alone, i.e., without the hAR cDNA sequence.

To express AR cDNA in prokaryotic systems, hAR and rAR cDNAs were inserted into a number of expression vectors including pUR, λGTll, pKK223-3, pKK233-2, pLEX, pATHl, pATH2, pATHIO, and pATHll. Vectors with AR cDNA inserts were used to infect E. coli strains (JM109, DH5o, Y1089, JM105, and RR1). According to polyacrylamide gel electrophoresis analysis, the infected bacteria can synthesize AR fragments coded for by the AR cDNA inserts. Some of these AR polypeptides are degraded in culture. Amino terminal, DNA-binding, and androgen binding domains were used, as described in Example 12, to construct fusion proteins representing these domains.

EXAMPLE 12

Production of Polyclonal and Monoclonal Antibodies to AR

The isolation of AR in significant amounts from androgen sensitive organs has been exceedingly difficult. Therefore, the high-level expression of hAR or rAR cDNAs, as shown in Example 11, is expected to be an ideal way for the large scale production of AR. In addition, oligopeptides, with sequences identical to the deduced amino acid sequences of portions of AR molecules, can be chemically synthesized inexpensively in large quantities. Both AR produced by expression vectors in eukaryotic or prokaryotic cells and AR oligopeptides chemically synthesized were used as antigens for the production of monoclonal antibodies as described in greater detail below.

Generally, several chemically synthesized oligopeptides, representing sequences unique to AR, (i.e., PYGDMRLETARDHVLP; CPYGDMRLETARDHVLP; and

SIRRNLVYSCRGSKDCIINK) were bound to BSA or KLH carrier proteins and were used to immunize mice. Spleen cells from these mice were fused to myeloma cells to produce hybrid antibody producing cells. Analysis by ELISA (enzyme-linked immunoassay) of the supernatants of 4 hybrid cultures appeared to indicate the presence of immunoglobulin that interacts with AR of rat ventral prostate. It is anticipated that these cells which produce monoclonal antibodies can be injected intraperitoneally into BALB/c mice previously treated with pristane. Ascites fluids can then be harvested and antibodies precipitated with ammonium sulfate.

Expression of Androgen Receptor Fusion Protein in E. coli

Three different parts of the AR gene (encompassing the N-terminal domain, the DNA-binding domain and the androgen-binding domain) were fused, in frame, to the N-terminal half of the trpE gene (trpE promoter-the first 969 bp of trpE coding region-multiple cloning region of pUC12) by using the pATH expression vectors as shown in^' Figures 6, 7, and 8, respectively. Dieckmann, et al. , J. Biol. Chem., 260:1513 (1985).

These constructions resulted in the fusion of approximately 25 kDa of AR, including a portion of the N-terminal domain; 29 kDa of AR, including a major portion of the DNA-binding domain; and 12 kDa of AR, including a portion of the androgen-binding domain; to the 33 kDa trpE protein. Because the trpE protein is insoluble, partially purified induced fusion proteins were obtained simply by lysing the E. coli and precipitating the insoluble fusion proteins. After electrophoresis on SDS-polyacrylamide gels, the induced fusion proteins, i.e., those proteins not present in the control pATH vector (no AR gene insert), were sliced from the gels and then used for immunization.

Fusion proteins, other than the three specifically exemplified, can also be constructed using these means.

Production and Purification of Anti-AR Antibodies

Rabbits, rats, and mice were immunized with either SDS-polyacrylamide gel slices containing denatured fusion proteins or electro-eluted, SDS-free, fusion protein, as well as fusion proteins obtained by other protein purification methods. The presence of antibodies to the fusion proteins in the antisera was assayed by ELISA. Positive serum having a higher titer was further assayed by the double antibody precipitation method using rat ventral prostate cytosol [³H]AR as antigen. The results showed that 1 μl of crude serum precipitated 10 to 20 fmole [³H]AR. Anti-AR crude serum was then affinity-purified by differential suspension of immune serum containing TrpE protein(s) (both those TrpE proteins having and those TrpE proteins not having inserted AR sequences) expressed by pATH vectors. The bound antibodies can be removed from the suspension because TrpE protein is insoluble. Antibodies specific against only the trpE protein were removed; antibodies specific for AR were isolated and again confirmed by both ELISA and double antibody precipitation. Production of Monoclonal Anti-Androgen Receptor Antibodies

The immunized rats were judged ready to be sacrificed for a fusion when their serum tested positive anti-AR antibodies by ELISA. Spleens were removed and grinded to release the cells into DMEM (Dulbeco's

Modified Engle's Medium) medium. Through a series of centrifugations using DMEM + DMEM with Ficoll Hypaque, the spleen cells were isolated. The SP2/0 myeloma cells were grown, split and diluted in 50 ml of DMEM with 20%

FCS, 1% MOPS, and IX L-Gln for two days before ready for the fusion. SP2/0 cells (5 x 10⁶) and 5 x 10⁷ spleen cells were used in the fusion. After incubating overnight, the fused cells were collected, suspended in

DMEM with IX H-T, IX Methotrexate, 20% FCS, and IX PBS and distributed in 96-well plates. Plates were supplemented after 6 days with DMEM and 20% FCS.

Hybridomas were identified and assayed, using the ELISA assay of Engrail, et al_^. , Bio. Chem. et Biophys. ACTA,

251:427-439 (1971). In this assay, plates were coated with either the AR fusion proteins or the TrpE protein as antigen and read on an ELISA reader.

Only those hybridomas that caused a positive reaction with the AR fusion protein were "limit diluted" to a concentration of 10 cells/ml and were then distributed among half of a 96-well plate. The remaining cells from the original well were transferred to a 24-well plate. Each of these plates had a thymocyte feeder layer. The tymocyte feeder layer was made up of thymus cells isolated from an un-injected rat, purified through centrifugation, irradiated with 1200 to 1400 RADS, and diluted to 1 x 10⁷ cells/ml of DMEM with 20% FCS. Positives from these thymocyte 96-well plates were again tested by ELISA. Only those which again tested positive with the AR fusion protein were grown up for monoclonal antibody purification. Three of the wells produced monoclonal antibody against AR. Both ELISA and double antibody assays were positive. The monoclonal antibodies were designated AN1-6, AN1-7, and AN1-15 and the three cell lines were designated HAN1-6, HAN1-7, and HAN1-15; Accession Nos. 10,000; 9,999; and 10,001; respectively, deposited on January 25, 1989 with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852.

Specificity of Anti-AR Antibodies

Sucrose gradient centrifugation was used to characterize the specificity of the three monoclonal anti-AR antibodies and their ability to react with non- denatured [³H]AR.

Cytosol was prepared from the ventral prostates of castrated rates as follows. Rats were castrated by the scrotal route while under anesthesia. They were killed 18 hrs. laters by cervical dislocation and their ventral prostates were removed, minced with scissors, washed in Buffer A (50 mM sodium phosphate, pH 7.5, 1 mM EDTA, 2 mM DTT, 10 mM sodium molybdate, 10% (v/v) glycerol and 10j mM sodium floride) and homogenized in 2x the tissue volume of Buffer A + 0.1 mM bacitracin, 1 mM PMSF, and aprotinin (lTIU/ml). The homogenate was centrifuged at 5,000 * g for 10 mins., adjusted to 10 nM ³H-androgen, spun at 225,000 x g for 45 mins. and treated with dextran-coated charcoal. One hundred μl of the cytosol solution, containing 3H-A-AR complexes, was incubated for 6 hrs. with 100 μl of the purified anti- androgen receptor monoclonal antibody, AN1-6, (20x as concentrated as the tissue culture media). Sucrose gradient centrifugation was performed by centrifugation at 257,000 x g for 16 hrs. at 4°C on a 3.8 ml, linear 5-20% (w/v) sucrose gradient containing 20 mM Tris-HCl, pH 7.5, 1 mM EDTA, 1 mM DTT, 10% (v/v) glycerol, and 0.4 M KC1. Gradients were fractionated and numbered from the bottom and 0.2 ml per fraction collected. The results obtained indicated that all three of the monoclonal antibodies, AN1-6, AN1-7, and AN1-15, recognized and effectively bound the radioactively labeled androgen receptor ([³H] AR) .

The [ H]AR and other steroid receptor complexes had a sedimendation coefficient of about 4-5S in the sucrose gradient media containing 0.4M KC1. Anti-AR antibodies do not alter the sedimentation coefficient of 4-5S for [³H]glucocorticoid receptors complexes of rat liver, estrogen receptor complexes of MCF-7 cells, and progesterone receptor complexes of T47D cells, but do shift the sedimentation coefficient of [³H]A-AR complexes of rat ventral prostate from 4S to 9- 12S or to heavier units. By SDS-polyacrylamide gel electrophoresis analysis it was also found that all major ^n vitro transcription/translation products of human and rat AR cDNAs were immunoprecipitatable by the anti-AR antibodies.

EXAMPLE 13

Use of AR cDNA and TR2 cDNA as Probes in the Study of Abnormality in Human and Animal Organs and Cancer Cells

Patients with metastatic prostatic cancer initially often respond favorably to androgen withdrawal therapy (castration or antiandrogen treatments). Most patients, however, eventually relapse to an androgen- state for which no chemotherapy, which would significantly increase the survival rate, is available. Regardless of the origin of androgen

-independent or -insensitive cancer cells, it is important to understand whether the androgen insensitivity or abnormality in the diseased cells are due to qualitative or quantitative changes in (a) the AR or TR2 genes, (b) regulation of their transcription, or translation, or (c) other cellular factors. AR cDNA, TR2 cDNA, or their partial segments can be used as specific probes in these studies.

For the analysis of AR or TR2 genes, high molecular weight genomic DNA isolated from target organs, tumors, and cultured cells can be used in identifying and characterizing AR genes. Different restriction endonucleases can be used to cleave DNA. The fragments can be analyzed by Southern analysis (agarose. electrophoresis, transfer to nitrocellulose and hybridization with AR cDNA probes). After identification, selected fragments can be cloned and sequenced. It is also possible to use appropriate oligonucleotide fragments of AR or TR2 cDNA as primers to amplify genomic DNA isolated from normal and abnormal organs or cells by specific DNA polymerases. The amplified genomic DNA can then be analyzed to identify sequence abnormality using the polymerase chain reaction (PCR) assay. Saiki, et al. , Science, 230, 1350 (1985). See also, Mullis, K.B., U.S. Patent No. 4,683,202; July 28, 1987; and Mullis, K.B., U.S. Patent No. 4,683,195; July 28, 1987. For the analysis of mRNA for ARs or related proteins, dot hybridization and Northern hybridization analysis could be used to characterize mRNA and AR or receptor-like molecules quantitatively and qualitatively. From these studies valuable information about the number of different forms of AR genes and their expression in androgen insensitive and sensitive tumor cells can be obtained.

DNAs and RNAs obtained from androgen sensitive and insensitive tumors and from cell lines from rats and humans with testicular feminization syndromes have been analyzed by the above methods. Preliminary studies indicated that abnormality in androgen responses may be due to sequence deletion/mutation in genes for ARs.

EXAMPLE 14

Development of Transgenic Animals

Transgenic techniques have been employed for expression of exogenous DNA. It may therefore be poss¬ ible to confer androgen sensitivity to animals with androgen receptor defects. For example, androgen insen¬ sitive animals, such as testicular feminized mice or rats, are known to have defective AR genes or defective AR itself. If DNA containing a normal AR gene is injected into fertilized mouse embryos, the transgenic mice may carry and express the gene and produce a functional AR necessary for androgen responses. For micro-injection, it is necessary to use AR genes containing DNA that can be expressed in the insensitive animals. A number of genomic receptor clones from human

X-chromosome libraries and rat genomic DNA libraries have been obtained and analyzed for their structures. Clones containing AR sequences will be characterized by endonuclease mapping, by Southern hybridization and by Sl-nuclease mapping. The 5' and 3' untranslated regions thus identified will aid in determining the minimal size of the DNA that would be required for tissue specific expression of the AR coding region.

Partial sequence analysis of the 5' and 3 ' regions would locate the minimal region that represents the promoter and the polyadenylation region. Approxi¬ mately 2 to 5 kb of upstream un-translated region and 0.5 to 1 kb of sequences downstream from the poly(A) site may be fused to the cDNA clone (minimal-gene) and injected into embryos of mice. Transgenic mice would be identified by analysis of their tail DNA using mini-gene specific probe(s). Normally only some of the transgenic mouse lines can express their transgenes. Transgenes may be inactive because of the presence of inhibitory sequences, integration of the exogenous gene into a transcriptionally inactive chromosomal location, or the juxtaposition of the transgene and an endogenous enhancer. In addition, androgen insensitivity may be due to various other factors and not due to abnormality in the AR gene or its expression. The foregoing illustrative examples relate to the isolation of human and rat cDNAs encoding DNA bind¬ ing proteins including androgen receptor and TR-2 and more particularly describe the transcription of the corresponding cDNAs and translation of the corresponding mRNAs in cell-free systems. While the present invention has been described in terms of specific methods and compositions, it is understood that variations and modi¬ fications will occur to those skilled in the art upon consideration of the present invention. Accordingly it is intended in the appended claims to cover all such equivalent variations which come within the scope of the invention as claimed.

Claims

WHAT IS CLAIMED IS

1. A purified and isolated DNA sequence encoding androgen receptor polypeptide.

2. The DNA sequence according to claim 1 encoding human androgen receptor polypeptide.

3. The DNA sequence according to claim 1 encoding rat androgen receptor polypeptide.

4. A purified and isolated DNA sequence encoding TR2 polypeptide.

5. The DNA sequence according to claim 1 or 4 which is a cDNA sequence.

6. The DNA sequence according to claim 1 or 4 which is a genomic DNA sequence.

7. The DNA sequence according to claim 1 or 4 which is a partially synthetic DNA sequence.

8. The DNA sequence according to claim 1 and as set forth in Figure 3.

9. The DNA sequence according to claim 4 and as set forth in Figure 4.

10. A procaryotic or eucaryotic host cell transformed or transfected with a DNA sequence according to claim 1 or 4.

11. The procaryotic transformed host cell according to claim 10 which is E. coli DH5α cells designated as, and corresponding to A.T.C.C. deposit Nos.: EC-hAR 3600, A.T.C.C. No. 67879; EC-rAR 2830, A.T.C.C. No. 67878; EC TR2-5, A.T.C.C. 67877; and EC TR2-7, A.T.C.C. No. 67876.

12. A viral or circular DNA plasmid compris¬ ing a DNA sequence according to claim 1 or 4.

13. A viral or circular DNA plasmid according to claim 11 further comprising an expression control DNA sequence operatively associated with said androgen receptor or TR2 encoding DNA.

14. A method for the production of androgen receptor polypeptide comprising: growing, in culture, a host cell transformed or transfected with a DNA sequence according to claim 1; and isolating from said host cell or culture the polypeptide product of the expression of said DNA sequence.

15. A method for the production of androgen receptor polypeptide comprising: disposing a DNA sequence according to claim 1 in a cell free transcription and translation system; and isolating from said system the polypeptide product of the expression of said DNA sequence.

16. A method for the production of TR2 poly¬ peptide comprising: growing, in culture, a host cell transformed or transfected with a DNA sequence according to claim 4; and isolating from said host cell or culture the polypeptide product of the expression of said DNA sequence.

17. A method for the production of TR2 poly¬ peptide comprising: disposing a DNA sequence according to claim 4 in a cell free transcription and translation system; and isolating from said system the polypeptide product of the expression of said DNA sequence.

18. The polypeptide product of the ij vitro or in vivo expression of a DNA sequence according to claim 1.

19. An amino acid sequence as set out in Figure 3.

20. The polypeptide product of claim 18 characterized by a molecular weights of 98 kD and 79 kD by SDS-PAGE and the ability to bind an androgen.

21. The polypeptide product of the iι vitro or i_jn vivo expression of a DNA sequence according to claim 4.

22. TR2 polypeptides.

23. A synthetic peptide duplicative of a sequence of amino acids present in AR or TR2 proteins in a region of the proteins not involved with DNA binding functions and sharing at least one antigenic epitope with AR or TR2 proteins.

24. An antibody specifically immunoreactive with at least one epitope of androgen receptor polypep¬ tide or TR2 polypeptide other than an epitope within the DNA binding functional region thereof.

25. The monoclonal antibody according to claim 24.

26. The monoclonal antibody according to claim 24 and produced by hybridoma cell line Nos. HB

10,000; HB 9,999; and HB 10,001.

27. The polyclonal antibody according to claim 24.

28. A method for quantitative detection of androgen receptor based on the immunological reaction of androgen receptor with an antibody according to claim 24.

29. A method for quantitative detection of TR2 receptor based on the immunological reaction of TR2 receptor with an antibody according to claim 24.

30. A method for the quantitative detection of androgen receptor encoding DNA or RNA based on hybridization of said nucleic acids with a DNA sequence according to claim 1.

31. A method for the quantitative detection of TR2 receptor encoding DNA or RNA based on hybridization of said nucleic acids with a DNA sequence according to claim 4.

32. A method for the quantitative and qualitative detection of AR or TR2 specific gene sequence or sequences present in a sample comprising the steps of: a) treating said sample with one oligonucleotide primer for each strand for said specific sequence, under hybridizing conditions such that for each strand of each sequence to which an oligonucleotide primer is hybridized an extension product of each primer is synthesized which is complementary to each nucleic acid strand, wherein said primer or primers are selected so as to be sufficiently complementary to each strand of each specific sequence to hybridize therewith such that the extension product synthesized from one primer, when it is separated from its complement, can serve as a template for synthesis of the extension product of the other primer; b) treating the sample under denaturing conditions to separate the primer extension products from their templates if the sequence or sequences to be detected are present; c) treating the sample with oligonucleotide primers such that a primer extension product is synthesized using each of the single strands produced in step (b) as a template, resulting in amplification of the specific nucleic acid sequence or sequences if present; d) adding to the product of step (c) a labeled oligonucleotide ^"probe for each sequence being detected capable of hybridizing to said sequence or a mutation thereof; and e) determining whether said hybridization has occurred.