Overexpression of a Partial Human Androgen Receptor in £. coli: Characterization of Steroid Binding, DNA Binding, and Immunological Properties

Charles Y.-F. Young, Shudong Qiu, James L. Prescott, and Donald J. Tindall Departments of Urology (C.Y.-F.Y., S.Q., J.L.P., D.J.T.) and Biochemistry and Molecular Biology (D.J.T.) Mayo Clinic/Foundation Rochester, Minnesota 55905

INTRODUCTION

The recent cloning of human androgen receptor (AR) cDNAs in this and other laboratories has provided valuable probes for investigating the structure and function of the AR at the molecular level. We now report the overexpression of a region of the human AR containing both the DNA- and hormone-binding domains in E. coli, which provides a means to produce large amounts of AR for analysis and use in functional studies. Under isopropyl-£-D-thiogalactopyranoside induction, a tripartite protein, consisting of 0-galactosidase, a collagenase recognition site, and AR polypeptide, was produced in E. coli JM109 using pSS20*a as a vector. About 1 mg of the fused AR could be recovered per liter bacterial culture. The induced protein could readily be detected in a sodium dodecyl sulfate-polyacrylamide gel by Coomassie blue staining. Its identity was confirmed by Western blot analysis using antibodies to both /?galactosidase and the AR. Scatchard analysis of the androgen-binding activity of the hybrid AR revealed high affinity binding to the synthetic androgen, Mibolerone (Kd, -1.2 niui). Competition studies demonstrated the fusion protein's specificity for androgens. The hybrid receptor formed immune complexes with human anti-AR serum that sedimented at about 19S in 10-50% linear sucrose gradients containing 0.4 M KCI. Gel band shift assays revealed that the hybrid receptor protein forms specific complexes with a synthetic steroid response element derived from the mouse mammary tumor virus long terminal repeat region. These results demonstrate that the recombinant AR expressed in E. coli possesses many of the functional properties characteristic of DNA- and steroid-binding domains of the native AR. (Molecular Endocrinology 4: 1841-1849, 1990)

Androgenic hormones regulate the development and growth of many normal and tumor cells. It is clear that the androgen receptor (AR) plays a key role in the mechanism of androgen action. Activation of the AR by binding of the male hormone (either testosterone or dihydrotestosterone) results in the differential regulation of both mRNA and protein synthesis in target cells ( 1 5). AR cDNA has been cloned in several laboratories, including our own (6-11). Sequence comparison with the other steroid hormone receptor genes has revealed that the AR gene contains two highly conserved regions. Functional analysis of the estrogen (12), progesterone (13), and glucocorticoid receptor (14) cDNAs has demonstrated that these two discrete regions correspond to DNA binding and steroid binding, respectively. The low abundance of the AR within tissues (-0.001% of total cellular protein) and its relative instability have complicated studies of the receptor's physical and functional properties. Moreover, these same factors have contributed significant difficulties in achieving purification of the receptor in large quantities. Although the AR cDNA has been expressed in AR-deficient eukaryotic cells (7, 8) or in in vitro translation systems (6, 10), the quantity of the expressed protein still remains at a level similar to that found in androgen target tissues. Expression of the AR in a suitable heterologous system could circumvent previous difficulties of obtaining sufficient quantities of the receptor for analyzing the molecular properties of the protein. The advantage of using E. coli as an expression host is that foreign genes can be easily manipulated into a suitable vector for producing large quantities of protein. Recently, a number of mammalian proteins have been expressed in E. coli in their biologically active forms (15-17). Moreover, the rat glucocorticoid receptor (18) and the chicken progesterone receptor (19, 20) cDNAs were success-

0888-8809/90/1841-1849S02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society

1841

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 04:41 For personal use only. No other uses without permission. . All rights reserved.

Vol4No. 12

MOL ENDO-1990 1842

fully expressed in E. coli and shown to exhibit specific binding to their cognate steroids and/or steroid response elements. Thus, these bacterially produced receptor derivatives exhibited biological activities similar to their in vivo counterparts. In this paper we report the overexpression of a truncated human AR as a fusion protein in E. coli. The bacterially expressed human (h) AR is antigenically and functionally similar to the endogenous receptor protein. This fusion protein should facilitate the elucidation of the structure and function of the DNA- and steroidbinding domains of the AR and the further characterization of other androgen-regulated genes.

RESULTS

We have isolated four cDNA clones and one genomic DNA clone, containing human AR DNA sequences, from a testis cDNA and human X-chromosome library, respectively. These DNAs have allowed us to construct a cDNA encoding a full-length hAR protein with 926 amino acid residues. As summarized in Fig. 1, various lengths of the hAR polypeptide, deduced from cDNAs that were isolated in different laboratories, have been reported (6-8). The discrepancy is apparently attributable to the different lengths of two homopolyamino acid regions in the N-terminal portion of the hAR protein. Thus, it is possible that the hAR protein in different individuals varies to a small extent in both size and amino acid content. Elucidation of the structure and function of the AR protein has always been hampered by its scarcity and lability. Overexpression of AR in a heterologous system, such as E. coli, would provide a good source for obtaining large quantities of AR. To accomplish this, the hAR cDNA from hAR52 was subcloned into pSP72. An Asp-718-C/al fragment, which codes for 415 amino acid residues and contains both the putative DNA-binding and the steroid-binding domains, was generated from this vector, blunt-ended with the Klenow fragment of DNA polymerase, and fused to the C-terminus of the E. coli /3-galactosidase gene in the plasmid pSS20*a

NM2 Polyglulamine

Polyglycine

COOH Androgen

No. a.a.

No. a.a.

Total a.a.

17

27

918'

21

24

919 b

20

23

917C

25

27

926 d

Fig. 1. Polymorphism of the hAR Gene The deduced amino acid sequences reported from different laboratories were compared. Only the two regions of homopolyamino acids indicated were variable in these reports. The DNA-binding domain (DNA) and the androgen-binding domain (androgen) are indicated, a.a., Amino acid; a, Chang et al. (6); b, Lubahn et al. (7); c, Tilley et al. (8); d, Young ef al. (10).

UGA

AUG

Wild type hAR DNA

Androgen

926 Truncated hAR '926

Fig. 2. A Prokaryotic Vector Construct, pSS20*a/AR1, for AR Expression in E. coli The truncated human AR from amino acids (aa) 511-926 was fused in frame to the 3' end of the Lac Z gene that contains the collagenase recognition site (CRS) at its C-terminus. Note that the amino acid 511 of the AR was changed from tryptophan to serine due to this construction process. The expression of the /S-galactosidase-collagenase recognition site-AR fusion gene is induced by IPTG. P, Promoter; O, operator.

(Fig. 2). As shown in Fig. 3A, the fusion protein, induced by isopropyl-/3-D-thiogalactopyranoside (IPTG), was readily detected by Coomassie blue staining on a sodium dodecyl sulfate (SDS)-polyacrylamide gel as a 150-kDa protein. Noninduced bacteria or bacteria transformed with the plasmid pSS20*a produced very little or none of the fusion protein. The fused hAR could be solubilized by sonication in low salt Tris buffer, pH 7.4, and detected as a 150-kDa protein in the gel (Fig. 3B). After 40% saturated ammonium sulfate precipitation, about 1 mg of the fusion protein could be recovered per liter bacterial culture. This represented approximately 2-5% of the solubilized bacterial proteins, as estimated by both densitometric scanning of Coomassie blue protein profile in SDS-polyacrylamide gel electrophoresis and protein assays of the solubilized bacterial proteins (data not shown). Attempts to express hAR in E. coli with a longer coding region (720 amino acid residues) of the hAR cDNA failed, where neither the fusion protein nor specific hormone-binding activity was detected. The identity of the inducible 150-kDa protein was verified by anti-AR and anti-/3-galactosidase antibodies. The anti-AR antibodies were produced in rabbits that had been injected with a synthetic AR peptide conjugated to keyhole limpet hemocyanin. The immunoglobulin G (IgGO fraction of the antiserum was demonstrated to have high specificity for AR by sucrose gradient, double immunoprecipitation, and Western blot analyses (Young, C. Y. F., J. L. Prescott, and D. J. Tindall, unpublished data). As shown in Fig. 3, C and D, the 150kDa protein in IPTG-induced E. coli harboring pSS20*a/ AR1 was recognized by both the anti-AR antibody and an anti-/3-gal monoclonal antibody, respectively. The interaction of the 150-kDa protein and the anti-AR

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 04:41 For personal use only. No other uses without permission. . All rights reserved.

Human AR Expressed in E. coli

1843

B

150 KDa

0

1 12

3 4

12

12

Fig. 3. SDS-Polyacrylamide Gel Analysis of the hAR Hybrid Protein Expression in E. coli A, Bacterial pellets (3.5-7.0 x 109 cells) washed with Tris buffer were resuspended in Laemli's sample buffer, heated at 25 C until completely dissolved, and dialyzed against Tris-SDS buffer, pH 6.8. The dialyzed bacterial lysate was heated in the presence of 2-mercaptoethanol at 96 C for 4 min and electrophoresed in a 10% SDS-polyacrylamide gel at 50 mamp. The gel was stained with Coomassie blue for 1 h and destained in a mixture of 5% acetic acid and 20% methanol for 18 h with several changes. Lane 0, SDS-polyacrylamide gel electrophoresis mol wt standards. Lanes 1 and 2, pSS20*a/AR1 harboring E. coli grown in the absence or presence of IPTG, respectively. Lanes 3 and 4, Same conditions as lanes 1 and 2, respectively, except that E. coli contained pSS20*a alone, without the AR insert. B, Sonicated E. coli lysate was prepared and precipitated with 40% saturated ammonium sulfate as described in Materials and Methods. The ammonium sulfate precipitate was dialyzed against Laemlli sample buffer, heated at 96 C, and subjected to electrophoresis as described above. The gel was stained with Coomassie blue. Lanes 1 and 2, pSS20*a/AR1 harboring E. coli grown in the presence or absence of IPTG, respectively. C and D, Protein from whole cells was prepared and electrophoresed as described in A. After transfer onto nitrocellulose, the proteins were probed with an IgG fraction of the rabbit anti-AR peptide antiserum (1:2000 dilution; C) or mouse monoclonal anti-/3-galactosidase antibody (D) and visualized by alkaline phosphatase conjugated with a secondary antibody against rabbit or mouse IgG, respectively. C, pSS20*a/AR1 harboring E. coli in the absence (lane 1) or presence (lane 2) of IPTG induction, and pSS20*a harboring E. coli in the absence (lane 3) or presence (lane 4) of IPTG induction. D, pSS20*a/AR1 harboring E. coli in the absence (lane 1) or presence (lane 2) of IPTG induction. E, An ammonium sulfate precipitate prepared from pSS20*a/AR1 harboring E. coli lysate was electrophoresized, transferred onto nitrocellulose, and probed with an IgG fraction of the rabbit anti-AR peptide antiserum in the absence (lane 1) or presence (lane 2) of the AR peptide (0.3 /*g/ml). The recognition of protein on the nitrocellulose by the antibodies was visualized by alkaline phosphatase-conjugated secondary antibody against rabbit IgG.

antibody on the Western blot was specific, since it could be abolished by the AR-specific peptide (Fig. 3E). Although two other bacterial proteins were recognized by this IgG fraction, the interaction of these bands could not be competed with the AR peptide; therefore, these bands represent nonspecific interactions. It was concluded that the 150-kDa protein contains both the AR and /3-galactosidase moieties. To determine whether the fusion protein could bind its cognate ligand with an affinity similar to that of the endogenous hAR, saturation analyses were performed using soluble protein or whole bacterial cells with the synthetic androgen Mibolerone. Scatchard analysis (Fig. 4) revealed nonlinear binding activity of the soluble fusion protein for Mibolerone that was resolved into two binding components (21). One component had a Ko of 1.2 nM, whereas the other had a K