- Email: [email protected]
Dimerization characteristics of the DNA- and steroid-binding domains of the androgen receptor
J. Steroid Biochem. Molec. Biol. Vol. 50, No. 5/6, pp. 225-233, 1994
Pergamon
0960-0760(94)00101-4
Copyright © 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0960-0760/94 $7.00 + 0.00
D i m e r i z a t i o n Characteristics of the D N A - and S t e r o i d - b i n d i n g D o m a i n s of the A n d r o g e n R e c e p t o r Takayuki
Nemoto,
1. Y u k o O h a r a - N e m o t o , and Minoru Ota I
2 Soichi
Shimazaki
1
Departments of 1Biochemistry and 2Microbiology, Iwate Medical University School of Dentistry, Morioka, Iwate 020, Japan
The DNA-binding domain (DBD) of the androgen, mineralocorticoid, and glucocorticoid receptors a n d t h e s t e r o i d - b i n d i n g d o m a i n ( S B D ) o f t h e a n d r o g e n r e c e p t o r ( A R ) w e r e e x p r e s s e d s e p a r a t e l y as f u s i o n p r o t e i n s w i t h g l u t a t h i o n e - S - t r a n s f e r a s e ( G S T ) in Escherichia coli. N a t i v e p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s a n d gel e x c l u s i o n H P L C d e m o n s t r a t e d t h a t t h e G S T - A R D B D f u s i o n p r o t e i n w a s p r e s e n t as a d i m e r . O n t h e o t h e r h a n d , t h e G S T - A R S B D f u s i o n p r o t e i n f o r m e d a h i g h - m o l e c u l a r w e i g h t o l i g o m e r , w h i c h s e e m e d to b e f o r m e d b y t w o s e p a r a t e i n t e r a c t i o n s , i.e. G S T - G S T a n d ARSBD-ARSBD between the fusion molecules. These findings strongly suggest that ARSBD has a p o t e n t a b i l i t y to f o r m a h o m o d i m e r a n d t h a t A R D B D d o e s n o t . G S T - A R D B D s p e c i f i c a l l y i n t e r a c t e d with the glucocorticoid response elements of the mouse mammary tumor virus long terminal repeat (GREMMTV). C l e a v a g e o f t h e f u s i o n p r o t e i n b y t h r o m b i n a b o l i s h e d t h e b i n d i n g , w h i l e t h e n o n s p e c i f i c DNA-cellulose binding ability was retained. Therefore, the dimeric configuration of GST-ARDBD, e v e n i f a c c o m p l i s h e d t h r o u g h t h e i n t e r a c t i o n w i t h t h e G S T m o i e t y , is n e e d e d f o r h i g h - a f f i n i t y b i n d i n g to t h e r e s p o n s e e l e m e n t . T h e b i n d i n g o f G S T - A R D B D to GRE~MTV w a s s t r o n g l y c o m p e t e d b y t h e g l u c o c o r t i c o i d r e s p o n s e e l e m e n t o f r a t t y r o s i n e a m i n o t r a n s f e r a s e gene, f o l l o w e d b y t h e a n d r o g e n r e s p o n s e e l e m e n t of the r a t p r o b a s i n gene. A p a l i n d r o m i c t h y r o i d r e s p o n s e e l e m e n t s h o w e d n o c o m p e t i t i o n . U n e x p e c t e d l y , n o a p p a r e n t d i f f e r e n t in t h e b i n d i n g a f f i n i t y to t h e s e r e s p o n s e elements was observed among the DBDs of androgen, mineralocorticoid and glucocorticoid receptors.
J. Steroid Biochem. Molec. Biol., Vol. 50, No. 5/6, pp. 225-233, 1994
INTRODUCTION
response element (GRE), is related to, but is distinct from, the H R E for the estrogen receptor (ER) [1]. M o u s e m a m m a r y t u m o r virus ( M M T V ) long terminal repeat contains several sequentially located G R E s , which also function as H R E s for androgen and progesterone receptors (AR and PR, respectively) [2]. At present, it is unclear whether or not specific response elements for androgen, mineralocorticoid and progesterone, distinct from G R E , are present. Claessens et al.[3] reported an androgen response element (ARE) in the first intron of the C3 gene of which expression is primarily regulated by androgens in vivo. However, the sequence of this half site is identical to the consensus sequence of G R E , i.e. T G T T C T ; and a transfection study revealed that the A R E of the C3 gene also functions as G R E or progesterone response element to a similar extent. Roche et al. [4] cloned several putative A R E by means of the in vitro D N A - b i n d i n g site
Steroid receptors exert their action through binding to a specific D N A sequence, called a hormone response element (HRE). T h e consensus sequence of the H R E for the glucocorticoid receptor (GR), i.e. glucocorticoid
*Correspondence to T. Nemoto. Abbreviations: AR, ER, GR, M R and PR, the androgen, estrogen,
glucocorticoid, mineralocorticoid and progesterone receptors, respectively; D B D and SBD, the respective D N A - and the steroidbinding domains of the steroid receptor; G S T , glutathioneS-transferase; M M T V , m o u s e m a m m a r y t u m o r virus; H R E , ARE, G R E and T R E , the hormone, androgen, glucocorticoid and thyroid response elements, respectively; AREpRoB , A R E 2 of the rat probasin gene; GRETAT, G R E of the rat tyrosine aminotransferase gene; GREMMTV, G R E of M M T V ; TREpAL, the palindromic T R E ; PCR, polymerase chain reaction; S D S - P A G E , polyacrylamide gel electrophoresis in the presence of s o d i u m dodecyl sulfate. Received 14 Feb. 1994; 25 Apr. 1994. 225
226
Takayuki
selection assay. The sequence is comprised of two 6-bp nucleotides with 3-bp spacers, S’-GGA/TACANNNTGTTCT-3’, which is similar to that for GREs. Very recently, Rennie et al. [5] reported two ARES of the rat probasin gene, which act together to mediate androgen-specific induction. The sequences with which the mineralocorticoid receptor (MR) specifically interacts have not been reported to date. Biochemical and crystallographic studies demonstrated that one DNA-binding domain of GR and ER interacts with a half site of HRE [6,7]. Moreover, GR, PR and ER form homodimers before binding to HRE [8-lo]. However, homodimerization is not common to all members of the steroid/thyroid receptor superfamily. At least, thyroid and retinoic acid receptors interact with cognate response elements primarily as a heterodimer with 94 retinoic acid receptor (also called RXR) [ll, 121. Vitamin D, receptor interacts with its response element either as a heterodimer with RXR or as a homodimer, depending on the structure of the response element [ 131. Recently, it has been shown that, like GR and PR, AR binds to ARE as a dimer in a ligand-dependent manner [14]. However, it has not been clear whether or not AR forms a homodimer before binding to DNA. In this study, the DBDs of AR, MR and GR, and the SBD of AR were expressed in E. coli as fusion proteins with glutathione-S-transferase (GST). Utilizing a dimerization/oligomerization characteristic of the GST-fusion proteins, we examined the dimerization ability of ARDBD and ARSBD and showed that ARSBD has the ability to form a dimeric configuration but that ARDBD does not. We also compared the DNA-binding activity of the three DBDs by the gel retardation assay. EXPERIMENTAL Materials
[3H]R1881 (80-86 Ci/mmol) was purchased from DuPont-New-England Nuclear (Boston, MA). [a-32P]dCTP (3000 Ci/mmol) was purchased from Amersham (Arlington Heights, IL). Restriction enzymes and DNA-modifying enzymes were obtained from Boehringer Mannheim, and Takara (Tokyo, Japan). An expression vector, pGEX-2T, and glutathioneSepharose 4B were from Pharmacia Biotech (Uppsala, Sweden). Thrombin (54 U/mg protein) and calf thymus double-stranded DNA-cellulose were from Sigma (St Louis, MO). All other reagents were of analytical grade. Methods Plasmid construction. The DNA fragments encoding DBDs of human AR, MR, and GR were amplified by the polymerase chain reaction (PCR). Sense primers for DBDs were designed to possess a BamHI site (italics), and the antisense primers were prepared with
Nemoto
et al.
an EcoRI site (underlined) as follows: ARDBD, 5’GTTGGA TCCATTGACTATTACTTTC-3’ and 5’-TCTGCAGAATTCGGCTCCCAGAGTCATC3’; MRDBD, 5’-ACTGGA TCCTCAAGACCTTC3’ and 5’-GACTGCAGAATTCCTAAATTCATTC-3’; GRDBD, 5’-TCCGGA TCCTCAACAGCAACAAC-3’ and 5’-TCTGCAGAATTCCAGGTTCATTCC-3’. Primers for the SBD of human AR were designed to possess a BamHI site (italics). The sequences of the sense and antisense primers were 5’-AAGCAGGA TCCACTCTGGGAGCCCG-3’ and 5’ - AGGGGATCCAATGCTTCACTGGGTG - 3’, respectively. cDNA clones encoding the full length of human AR, MR and GR, generously provided by Drs S. Liao (Chicago University), J. L. Arizza (University of California) and R. L. Evans (Salk Institute), respectively, were used as templates for the PCR. PCR was done for 25 cycles with denaturation for 2 min at 92°C annealing for 2 min at 50°C and polymerization for 2.5 min at 70°C. The amplified fragments encoding DBDs were cut with BamHI and EcoRI, and inserted into the BamHI/EcoRI site of a pGEX-2T vector. The constructed plasmids were designated pGEXARDBD for AR, pGEX-MRDBD for MR, and pGEX-GRDBD for GR. E. coli Y1090 was transformed with these plasmids separately. The amplified fragment encoding ARSBD [Fig. l(a)] was cut with BamHI and inserted into the BamHI site of pGEX-2T. E. coli Y1090 was transformed with the ligated plasmid, designated pGEX-ARSBD. A 139-bp fragment containing three GREs of the MMTV was amplified by PCR with synthetic primers: 5’-CCTTGCGGATCCCAGGGCT-3’ (BamHI site and 5’-GATTTGGATrepresented by italics) GAATTCCAAAAG-3’ (EcoRI site represented by italics). MMTV5031 (Japanese Cancer Research Resources) was used as a PCR template. The PCR product was cut with BamHI and EcoRI, and inserted into the corresponding site of pUC18, designated pUCMMTV. Induction
and purification
of recombinant
proteins.
The transformed bacteria were cultured at 37°C in 250ml LB medium with 50pg/ml ampicillin. When of the culture had reached 0.8-1.0, the temthe A,,, perature was lowered to 28°C and the bacteria were further cultured for 4-5 h in the presence of 0.2 mM B-isothiogalactopyranoside. The cells were harvested and stored at -80°C until used. The cell lysate was prepared and the recombinant proteins were purified according to the manufacturer’s protocol except that protease inhibitors (20 pgg/ml each of leupeptin, soybean trypsin inhibitor, antipain and pepstatin) were added as supplements to the lysate preparation buffer. In the case of GST-ARSBD, the sample eluted from glutathione-Sepharose contained a significant amount of a 27-kDa protein, presumably a proteolysed product of the fusion protein. This sample was further subjected to DEAE-5PW column
Dimerization of Androgen Receptor (0.75 × 5 c m , Tosoh, Japan) chromatography and eluted with a linear gradient of 0-0.6 M NaC1 in 20 m M Tris-HC1 (pH 7.3). Polyacrylamide gel electrophoresis. Bacterial lysates or purified proteins were subjected to electrophoresis at 4°C on 12.5 or 15% (w/v) polyacrylamide gels under denaturing conditions according to Laemmli [15], or on 7.5% (w/v) polyacrylamide gels under nondenaturing conditions [16].
Gel
exclusion
HPLC
on
TSK-G3OOOSWXL.
Purified proteins (0.3 ml) were subjected to gel chromatography at 4°C on a T S K - G 3 0 0 0 S W X L column (0.78 × 30 cm, Tosoh, Japan), and eluted with 20 m M Tris-HC1 (pH 7.3), 1 m M E D T A and 50 m M NaC1. Flow rate was 0.25 ml/min, and 1-min fractions were collected.
Molecular weight markers estimating molecular weights of native forms. For native polyacrylamide gel electrophoresis as well as gel exclusion H P L C analysis, ovalbumin (45-kDa), bovine serum albumin (monomer, 66-kDa; dimer, 132-kDa; trimer, 198-kDa) and catalase (240-kDa) were used as molecular weight markers. Additionally, cytochrome C (13-kDa) and sperm whale myoglobin (17-kDa) were used as markers for gel exclusion H P L C analysis. Gel retardation analysis. Interaction of the expressed proteins with specific D N A sequences was investigated by gel retardation analysis. T h e 139-bp fragment excised with EcoRI and XbaI from p U C M M T V , designated GREM~Tv, was labeled with [~-32p]dCTP and Klenow fragment, and used as a probe for gel retardation analysis. Purified proteins were subjected to gel retardation analysis as described previously [17]. After electrophoresis, the gels were dried and subjected to autoradiography. In some experiments, radioinert GRETAT, TREFA L and AREeRoB, synthesized with a Model 381A D N A synthesizer (Applied Biosystems, CA), were used as competitors. T h e sequences of GRETAx, TREpAL and AREpRoB were 5 ' - G G G T C T G C T G T A C A G GATGTTCTAGCTACGCCC-3' [18], 5 ' - G A T C CAAGATTCAGGTCATGACCTGAGGAGAGCC T A G - 3 ' [19] and 5 ' - C C G G G T A A A G T A C T C CAAGAACCTATTTGCCCGG-Y [5], respectively. RESULTS
Expression and purification of the recombinant proteins Our previous expression studies in E. coli revealed lower expression levels of larger molecules of steroid receptors [20, 21]. Accordingly, in this study, we did not express a full-length receptor, but rather expressed functional domains of AR, i.e. D B D and SBD, as fusion proteins with G S T [Fig. l(a)]. DBDs of M R and G R were also expressed [Fig. l(b)]. As shown in Fig. l(c) the amino acid sequence of A R D B D is less similar to that of M R D B D or G R D B D compared with the similarity between M R D B D and G R D B D .
227
T h e bacterially-expressed G S T - D B D fusion proteins, designated G S T - A R D B D , G S T - M R D B D and G S T - G R D B D , were analyzed by S D S - P A G E [Fig. 2(a)]. A predominant band at around Mr 35,000 was found in lysates of E. coli Y1090 [ p G E X - A R D B D , p G E X - M R D B D , and p G E X - G R D B D ] [Fig. 2(a), lanes 1-3], and at Mr 27,000 in the lysate of Y1090[pGEX-2T] (lane 4). These values are in good agreement with the molecular weights calculated from the predicted amino acid sequences of cloned receptor cDNAs [22-24]. T h e recombinant proteins were purified to near homogeneity by a single-step purification with glutathione-Sepharose affinity chromatography (lanes 5-8). In the lysate of Y1090[pGEX-ARSBD], specific [3H]R1881-binding was detected (20,455 + 477 dpm/ 0.1 ml lysate at 1 0 n M [3H]R1881), while no specific binding was found in the lysate of Y1090[pGEX-2T]. After purification by glutathione-Sepharose chromatography, Mr 60,000 and 27,000 bands were observed [Fig. 2(b), lane 2]. F r o m the calculation of the predicted molecular weight of G S T - A R S B D (60.9-kDa), the Mr 60,000 band was assigned to the fusion protein, and the band with M r 27,000 appeared to be G S T , presumably produced by proteolysis of the fusion protein. T h e Mr 60,000 band was subsequently purified to homogeneity by ion-exchange D E A E - H P L C (lane 3). T h e purified samples were used in the following experiments.
Analysis of the molecular configuration of recombinant proteins T o characterize the molecular forms of the expressed proteins, we subjected the purified proteins to native gel electrophoresis. G S T - A R D B D migrated at Mr 150,000 (Fig. 3, lane 2) on the native gel. This value was approximately four times as large as the value on S D S - P A G E (Mr 35,000), indicating a tetrameric structure of G S T - A R D B D . However, we supposed that G S T - A R D B D migrated slower than that expected from its real molecular weight, on native gel electrophoresis, because of the basic amino acid residues of D B D and its coordination with two zinc ions. In fact, the dimeric configuration of G S T - A R D B D was ascertained by gel exclusion H P L C analysis as described below (see Fig. 4). T o the contrary, most of G S T - A R S B D did not penetrate the gel, but remained concentrated at the top (Fig. 3, lane 1). This indicated that G S T - A R S B D formed an oligomeric structure with a large molecular mass. A faint band observed a t M r 140,000 may correspond to a dimer of G S T - A R S B D . T h e molecular form of the expressed proteins was further characterized by gel exclusion H P L C analysis [Fig. 4(a)]. T h e native Mr was calibrated in comparison with marker proteins [Fig. 4(b)]. It has been already shown that G S T exists as a homodimer [25]. In fact, Fig. 4 shows that G S T ( M r 27,000) eluted at the position of a dimer (Mr 55,000). G S T - A R D B D (M r
Takayuki Nemoto et al.
228
(a) AR
1
200
I
I
400 '
600"~R66 556'623
800 '
918 ,
I 548 627 ~i;i;i;;:G STi:::;iii;;]l~!ii.:..'!~
GST-ARDBD
fili~i!i i i!i i i i i~!ii i i ~
GST-ARSBD
(b) AR
1 ]i0 20 30 40 50 60 70 1 IDYYFPP~TCLICGDEASGCHYGALTCGSCKVFFKR/~%EGKQKYLCASRMDCTIDKFRRKMCPSCRLRKCYEKG~LGA
MR
80
1 ] i0 20 30 40 50 60 70 | 7 6 S S R - ~ I--V ........... W ............ V - - Q H N .... G .... I---I ...... A - - - Q - - L Q - - ~ - ~ 1
GR 1 I0] SSTATTG--~L--V-S
20 30 40 50 60 70 ]80 ......... V ............. V - - Q H N .... G .... I--- I ...... &--Y--- L Q - - ~ - E-
(c) AR AR
MR
GR
15 (22.1%)
MR
15
GR
15
15 (22.1%) 5
(7.4%)
5
Fig. 1. Schematic illustration of recombinant AR, GST-ARDBD and GST-ARSBD. (a) Human AR, GST-ARDBD and GST-ARSBD are schematically illustrated with putative DBD (amino acid residues 556-623) and SBD (amino acid residues 666-918). (b) The amino acid sequences of AR, MR and GR expressed a s GST-fusion proteins are illustrated. A box indicates the putative DBD. Arrows indicate the cysteine residues coordinated with zinc ions. (c) The number of substituted amino acids in DBD (68 amino acid residues) between the receptors and their percentages (parentheses) are represented.
35,000) also eluted at the position of a dimer (Mr 70,000). Therefore, as we postulated above, the migration of G S T - A R D B D on native gel electrophoresis was retarded due to the basic nature of DBD. T h e phenomenon was found even on S D S - P A G E of thrombin-cleaved G S T - A R D B D , i.e. A R D B D , as described below (see Fig. 6). Consistent with the results of native gel electrophoresis, G S T - A R S B D , forming a high-molecular weight oligomer, eluted at the void volume fraction of the gel chromatography (Fig. 4).
Interaction of the expressed proteins with specific D N A sequences Gel retardation analysis of G S T - A R D B D , G S T M R D B D and G S T - G R D B D was performed by use of a 139-bp fragment of GREMMTV which contains three GREs in its sequence. As shown in Fig. 5, all three fusion proteins bound to the probe, and gave similar shifted band patterns. T h e increase in input proteins resulted in heavier retardation of the D N A - D B D complexes, because several D B D s interacted with GREMMTV at higher protein concentrations. Essentially no binding was found with G S T . Unexpectedly, a higher protein concentration was needed for
G S T - G R D B D to obtain an intensity comparable to that of G S T - A R D B D and G S T - M R D B D . T h e junction of G S T - A R D B D can be cleaved by thrombin. Following thrombin cleavage, G S T A R D B D was separated into two species, i.e. G S T and A R D B D [Fig. 6(a)]. The Mr 13,000 of A R D B D obtained on S D S - P A G E (lane 2) was appreciably larger than that calculated from its c D N A seuqence (9.7kDa), because of its basic nature. T h e gel retardation analysis with the cleaved protein revealed a shifted band only at the highest protein concentration examined [Fig. 6(b), lane 10]. Accordingly, a greater than 64-fold molar excess of proteins was needed for A R D B D to obtain a p r o t e i n - D N A complex comparable to that of G S T - A R D B D [compare Fig. 6(b), lanes 2 and 10]. However, both A R D B D and G S T - A R D B D similarly bound to calf thymus DNA-cellulose [Fig. 6(c), lane 4]. These results strongly indicate that a dimeric configuration of A R D B D is required for highaffinity binding to HRE. In Fig. 6(c), although a small portion of G S T was also retained by the matrix (lane 4), this species may have been derived from a heterodimer of G S T and G S T - A R D B D bound to the matrix.
Dirnerization of Androgen Receptor (a)
229
(b)
97k
66k
45k
30k
20k
1
2
3
4
5
6
7
8
M
1
2
3
M
Fig. 2. S D S - P A G E of r e c o m b i n a n t p r o t e i n s . (a) The b a c t e r i a l lysates (lanes 1-4) a n d t h e a f f i n i t y - p u r i f i e d p r o t e i n s (lanes 5-8) of E. coli Y1090[pGEX-ARDBD], [ p G E X - M R D B D ] , [ p G E X - G R D B D ] , a n d [pGEX-2T] were a n a l y z e d on S D S - P A G E . Lanes 1 a n d 5, Y1090[pGEXARDBD]; lanes 2 a n d 6, Y1090[pGEX-MRDBD]; lanes 3 a n d 7, Y1090[pGEX-GRDBD]; lanes 4 a n d 8, Y1090[pGEX-2T]. (b) T h e lysate (lane 1), a f f i n i t y - p u r i f i e d (lane 2) a n d D E A E - H P L C p u r i f i e d (lane 3) s a m p l e s of Y1090[pGEX-ARSBD] were a n a l y z e d on S D S - P A G E . M, m o l e c u l a r weight m a r k e r s .
T h e D N A - b i n d i n g specificity of G S T - D B D fusion proteins of AR, M R and G R was examined. As a potent response element for androgen, i.e. ARE, Rennie et al. [5] recently reported two cis-acting D N A elements involved in androgen regulation of the rat probasin
gene. Here, we analyzed one of the sequences (designated AREpRoB in this report, and ARE2 in Ref.[5]), because the binding affinity of AR to this sequence is higher than that to the other. T h e binding affinities of GRETA T and TREpA c were also compared.
~~
240k 198k 132k
240k-198k--
66k 45k 132k--,
66k
.
45k --.
I
I M
I 1
2
I M
Fig. 3. Native gel e l e c t r o p h o r e s i s of p u r i f i e d p r o t e i n s . G S T - A R S B D (lane 1) a n d G S T - A R D B D (lane 2) were s u b j e c t e d to n a t i v e gel e l e c t r o p h o r e s i s . M o l e c u l a r weight m a r k e r s (M) of n a t i v e gel e l e c t r o p h o r e s i s were r u n in parallel.
230
Takayuki Nemoto et al.
(a)
(b) ~..Q--45k
G S T - A R D B D ---~
w
qllI~
GST-ARSBD
/
~-.~--30k
400
29 31 33 35 37 39 41 43 45 47 49 M
1
2
~o~ 200 ---
"~
100 80
3 o
40
~45k M
21 23 25 27 29 31 33 35 37 39
GST-AF:IDBD
/
-',~--66k
GST-ARSBD~
60
¢l 0
-20 _
:~
10 20
GST
GST~
b7
[
I
I
30
40
50
I° 8 60
Fraction number 21
23
25
27
29
31
Fraction
33
35
37
39
M
number
Fig. 4. Gel exclusion HPLC analysis on TSK-G3000SWXL. (a) GST-ARDBD, GST-ARSBD, and GST were analyzed on a TSK-G3000SWXL column equilibrated with 20 mM Tris-HC1 (pH 7.3), I mM EDTA and 50 mM NaCI. Aliquots (20/Jl) of the eluted samples were analyzed on SDS-PAGE. (b) The relative molecular weights of eluted samples in "a" are plotted. Markers are as follows: 1, bovine serum albumin tetramer (264-kDa); 2, catalase (240-kDa); 3, albumin trimer (198-kDa); 4, albumin dimer (132-kDa); 5, albumin monomer (66-kDa); 6, ovalbumin (45-kDA); 7, sperm whale myoglobin (17-kDa); 8, swine cytochrome C (13-kDa). T h e b i n d i n g of G S T - A R D B D to GREMMTV was m o s t strongly c o m p e t e d b y GRETA T a n d weakly b y AREpRoB (Fig. 7). TREpAL showed n o c o m p e t i t i o n w i t h i n the c o n c e n t r a t i o n s used in this study. C o m pletely identical results were obtained with G S T - M R D B D . I n the case of G S T - G R D B D , G R E TAT was the strongest c o m p e t i t o r as in the other cases, b u t the difference b e t w e e n AREpRoB a n d TREpA L was a m b i g u o u s because of a weak b i n d i n g of G S T G R D B D to the probe. I n c o n c l u s i o n , AREpRoB had a weaker affinity t h a n GRETA T for the b i n d i n g of GST-ARDBD, GST-MRDBD, or G S T - G R D B D .
4) a n d the oligomeric one of G S T - A R S B D (Fig. 4), G S T - A R S B D s h o u l d possess a n a d d i t i o n a l i n t e r a c t i n g GST
AR
MR
GR
DISCUSSION I n this study, we expressed D B D s of three receptors a n d A R S B D as G S T fusion proteins. T h e G S T - f u s i o n expression system has b e e n already e m p l o y e d for n u m e r o u s p r o t e i n s i n c l u d i n g A R [26]. I n this study, however, we note a novel characteristic of o u r G S T f u s i o n proteins: T h e G S T p o r t i o n of the f u s i o n p r o t e i n possesses the p r o p e r t y of d i m e r f o r m a t i o n ; a n d therefore, w h e n a fused p o r t i o n t e n d s to b i n d each other, the G S T - f u s i o n p r o t e i n results in f o r m i n g a h i g h - m o l e c u lar weight oligomer. U t i l i z i n g this u n i q u e characteristic, we analyzed the d i m e r i z a t i o n abilities of A R D B D and ARSBD. Both native gel electrophoresis a n d gel exclusion c h r o m a t o g r a p h y clearly d e m o n s t r a t e d that G S T A R S B D existed as an oligomeric form. T a k i n g into a c c o u n t the d i m e r i c form of G S T (Ref. [25] a n d Fig.
1 2 3 4 5
6
7
8
9 1011 121314
1516
F
Fig. 5. Gel retardation analysis of GST-ARDBD, GSTMRDBD and GST-GRDBD. GST, GST-ARDBD, GSTMRDBD and GST-GRDBD were incubated with the 3zp-labeled GREMMTVprobe (5000 cpm). After 20 min, DNA-protein complexes were separated on a 5% (w/v) polyacrylamide gel. Lanes 1, 5, 9 and 13, 0.5 #g proteins; lanes 2, 6, 10 and 14, 1 ~g; lanes 3, 7, 11 and 15, 2#g; lanes 4, 8, 12 and 16, 4/1g. F, free probe.
Dimerization of Androgen Receptor
(a)
231
(c)
(b)
9Y ~)7k
97k 66k 45k GST-ARDBD GST
66k 45k
30k
GST-ARDBD 30k
G ST 20k
20k
ARDBD
4k 1
2
M
14k
AR D BD F 12345
F678910
1
2
3
4
Fig. 6. Effect of thrombin cleavage on the DNA-binding of GST-ARDBD. (a) GST-ARDBD (210 gg) was incubated with or without 1.6 U thrombin at 30°C for 3 h. Aliquots of the sample uncleaved (lane 1) or cleaved (lane 2) with thrombin were analyzed on a 15% (w/v) gel of SDS-PAGE. (b) Uncleaved (GST-ARDBD) or cleaved (ARDBD and GST) sample obtained as in "a" was incubated with 32P-labeled GREM~TVand separated on a 5% polyacrylamide gel. Lanes 1 and 6, 0.1/~g; lanes 2 and 7, 0.4gg; lanes 3 and 8, 1.6 #g; lanes 4 and 9, 6.4 #g; lanes 5 and 10, 25.6 ~g proteins. (c) Thrombin-treated GST-ARDBD was applied to a DNA-cellulose column (1 x 5 cm) equilibrated with 20 m M Tris-HC1 (pH 7.3) containing 1 mM EDTA, 1 m M dithiothreitol, 10% (v/v) glycerol, and 50 mM NaCl. The column was washed with the same buffer, and then eluted with the same buffer with 0.3 M NaCI as a final concentration. Samples were analyzed by SDS-PAGE. Lane 1, untreated GST-ARDBD; lane 2, thrombin-treated GST-ARDBD; lane 3, DNA-cellulose nonadsorbed fraction; lane 4, 0.3 M NaCl-eluted fraction. Data of "a and b" and "c" are from separate experiments.
site o t h e r t h a n the G S T m o i e t y ; and, h e n c e , G S T A R S B D is c a p a b l e o f i n t e r a c t i n g w i t h t w o o t h e r m o l ecules. F i n a l l y , G S T - A R S B D f o r m s an o l i g o m e r as i l l u s t r a t e d in Fig. 8. T h u s , the o l i g o m e r i c s t r u c t u r e o f GST-ARSBD i m p l i e s that A R S B D ( m o r e precisely, a m i n o acid r e s i d u e s 6 2 4 - 9 1 8 ) h a v e t h e ability to f o r m a h o m o d i m e r . S i n c e no l i g a n d was a d d e d u n d e r these e x p e r i m e n t a l c o n d i t i o n s , l i g a n d - b i n d i n g is not req u i r e d for the d i m e r i z a t i o n o f A R S B D . P r e s u m a b l y , the l i g a n d - b i n d i n g is p r i m a r i l y n e e d e d for the l i b e r a ti on o f an as s o ci at i n g c o m p o n e n t o f the n o n D N A b i n d i n g f o r m o f h e t e r o m e r i c A R [21].
GRETA T
TREpAL AREpRoB
AR
W e p r e v i o u s l y r e p o r t e d that b act er i al l y e x p r e s s e d t r u n c a t e d A R s r e t a i n i n g D B D an d S B D ( A R 4 3 8 ) or S B D ( A R 6 1 2 ) s e d i m e n t e d at 4 - 5 S an d 3.5--4.5S, resp ect i v el y , u p o n g l y c e r o l g r a d i e n t c e n t r i f u g a t i o n [21]. T h o s e species m a y r e p r e s e n t d i m e r i c f o r m s . M o r e o v e r , the s t u d y s h o w e d that A R S B D itself does not f o r m an o l i g o m e r i c s t r u c t u r e such as G S T - A R S B D . T o directly d e m o n s t r a t e a d i m e r i c f o r m o f A R S B D , we a t t e m p t e d to separate G S T an d A R S B D by t h r o m b i n cleavage. H o w e v e r , G S T - A R S B D was n e v e r c l e a v e d for s o m e u n k n o w n reason. W h i l e t r y p s i n efficiently d e g r a d e d G S T - A R S B D , a s t a b l y - e x i s t i n g species c o r -
GRETAT TREpAL AREpRoB MR
12345
6 7 8 10 1113 15 9 12 14
GRETAT TREpAL AREpRoB GR
F12345
678
........................... 678 10 11 13 15 9 12 14 Fig. 7. DNA-binding specificity of GST-ARDBD, GST-MRDBD and GST-GRDBD. GST-ARDBD (0.75 ~g), GST-MRDBD (0.75 gg), and GST-GRDBD (4 gg) were incubated with 32P-labeled GRE-MMTV in the absence or presence of competitors. DNA-protein complexes were separated on a polyacrylamide gel. Lanes 1, 6 and 11, no competitor; lanes 2, 7 and 12, 0.04 pg; lanes 3, 8 and 13, 0.2/~g; lanes 4, 9 and 14, 1/~g; lanes 5, 10 and 15, 5 ~g competitors. F, free probe. 9
10 11 13 15 12 14
12 3 4 5
232
Takayuki Nemoto et al.
GST-ARDBD
GST-ARSBD GST
"--TD
ARDBD
Fig. 8. Model of dimerization and oligomerization of G S T fusion proteins. The model of dimerization of G S T - A R D B D and oligomerization of GST-ARSBD are schematically represented.
r e s p o n d i n g to the Mr o f A R S B D was n o t o b t a i n e d (data not shown). O n the o t h e r h an d , the d i m e r i c s t r u c t u r e o f G S T A R D B D i m p l i es that A R D B D ( m o r e precisely, a m i n o acid residues 548-627) has no i n t e r a c t i n g site. In o t h e r w o r d s , A R D B D itself does not f o r m a h o m o d i m e r . It has already b e e n a c c e p t e d that t w o m o l e c u l e s o f D B D o f G R and E R b i n d to a p a l i n d r o m i c s t r u c t u r e o f H R E . M o r e o v e r , G R [8], E R [10] and P R [9] f o r m h o m o d ime rs after l i b e r a t i n g b i n d i n g c o m p o n e n t s , such as 9 0 - k D a heat shock p r o t e i n , b e f o r e b i n d i n g w i t h H R E . N M R an d c r y s t a l l o g r a p h i c studies on G R a n d E R clearly s h o w e d that the D B D s i n t e r a c t w i t h each o t h e r p r i m a r i l y w i t h i n the s e c o n d zinc finger [6, 7]. H o w e v e r , this i n t e r a c t i o n ap p e a r s to be insufficient to f o r m a d i m e r i c c o m p l e x w h e n the t w o are not associated w i t h H R E [27, 28]. T h i s s t u d y also d e m o n s t r a t e d that A R D B D alone does not h o m o d i m e r i z e . M o r e o v e r , o u r data s t r o n g l y suggest that the native A R m o l e c u l e h o m o d i m e r i z e s t h r o u g h the i n t e r a c t i o n o f S B D . H e n c e , one o f the f u n c t i o n s o f S B D m a y be the stabilization o f the d i m e r i c s t r u c t u r e o f D B D . T h e dimeric configuration of G S T - A R D B D m a y be app r e c i a b l y d i f f eren t f r o m that o f an intact A R . N e v e r t h e less, it s eem s i m p o r t a n t to note that, at least w i t h r e spe ct to e n h a n c e m e n t o f D N A - b i n d i n g activity o f A R D B D , G S T is able to s u b s t i t u t e the f u n c t i o n o f ARSBD. In this study, we did not analyze the N - t e r m i n a l A / B d o m a i n o f A R D B D , w h i c h m i g h t be i n v o l v e d in the d i m e r i z a t i o n o f A R . I n fact, W o n g et al. [14] r e p o r t e d that A R d i m e r i z a t i o n r e q u i r e s a n d r o g e n - b i n d i n g only w h e n the N - t e r m i n a l A / B d o m a i n o f A R is p r e s e n t , and s u g g e s t e d that the N - t e r m i n a l A / B d o m a i n acts to i n h i b i t A R d i m e r i z a t i o n a n d D N A - b i n d i n g . O n the o t h e r h a n d , this d o m a i n is n e e d e d for the full activity o f t r a n s c r i p t i o n a c t i v a t i o n o f G R , and is a s s u m e d to
i n t e r a c t w i t h r e c e p t o r - s p e c i f i c t r a n s c r i p t i o n factors [29]. F u r t h e r studies are n e e d e d to e l u c i d a t e the f u n c tion o f the N - t e r m i n a l A / B d o m a i n o f steroid receptors. T h e D N A - b i n d i n g specificity o f the t h r e e D B D s f u sed to G S T d i d not reveal any difference w i t h t h r e e r e s p o n s e e l e m e n t s ; all t h r e e D B D s possessed a h i g h e r b i n d i n g affinity for GRETA-r. A l t h o u g h the a m i n o acid s e q u e n c e o f A R D B D is 2 2 . 1 % different f r o m t hos e s e q u e n c e s o f M R and G R [Fig. l ( b and c)], the difference seems n o t to affect the D N A - b i n d i n g p r o p erties o f A R , at least n o t its in v i t r o D N A - b i n d i n g characteristics. T h e r e f o r e , the D N A s e q u e n c e o f H R E an d a m i n o acid s e q u e n c e s o f D B D do n o t a p p e a r the p r i m a r y d e t e r m i n a n t for the specificity o f the actions o f a n d r o g e n , m i n e r a l o c o r t i c o i d and g l u c o c o r t i c o i d . Acknowledgements--We thank Dr S. Liao (Chicago University) for
the rat androgen receptor cDNA plasmid, Dr R. Evans (Salk Institute) for human GR cDNA plasmid, and Dr J. L. Ariza (University of California) for human MR cDNA plasmid, pMR3750. This work was supported by grants-in-aid for scientific research for the Ministry of Education, Science and Culture of Japan (to T.N., Y.O.-N. and M.O.) and from the Asahi Glass Foundation (to T.N.).
REFERENCES
1. Klock G., Str/ihle U. and Schfitz G.: Oestrogen and glucocorticold response elements are closely related but distinct. Nature 329 (1987) 734-736. 2. Cato A. C. B., Henderson D. and Ponta H.: The hormone response element of the mouse mammary tumour virus DNA mediates the progestin and androgen induction of transcription in the proviral long terminal repeat region. E M B O J 6 (1987) 363-368. 3. Claessens F., Celis B., Peeters W., Heyns G., Verhoeven G. and Rombauts W.: Functional characterization of an androgen response element in the first intron of the C3(1) gene of prostatic binding protein. Biochem. Biophys. Res. Commun. 164 (1989) 833-840. 4. Roche P. J., Hoare S. A. and Parker M. G.: A consensus DNA-binding site for the androgen receptor. Molec. Endocr. 12 (1992) 2229-2235. 5. Rennie P. S., Bruchovsky N., Leco K. J., Sheppard P. C., McQueen S. A., Cheng H., Snoek R., Hamel A., Bock M. E., MacDonald B. S., Nickel B. E., Chang C., Liao S., Cattini P. A. and Matusik R. J.: Characterization of two cis-acting DNA elements involved in the androgen regulation of the probasin gene. Molec. Endocr. 7 (1993) 23-36. 6. H~ird T., Kellenbach E., Boelens R., Maler B. A., Dahlman K., Freedman L. P., Carlstedt-Duke J., Yamamoto K. R., Gustafsson J.-/k. and Kaptein R.: Solution structure of the glucocorticoid receptor. Nature 249 (1990) 157-160. 7. Luisi B. F., Xu W. X., Otwinowski Z., Freedman L. P., Yamamoto K. R. and Siegler P. B.: Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352 (1991) 497-505. 8. Cairns W., Cairns C., Pongratz I., Poellinger L. and Okret S.: Assembly of a glucocorticoid receptor complex prior to DNA binding enhances its specific interaction with a glucocorticoid response element. J. Biol. Chem. 266 (1991) 11221-11226. 9. Rodriguez R., Weigel N+ L., O'Malley B. W. and Schrader W. T.: Dimerization of the chicken progesterone receptor in vitro can occur in the absence of hormone and DNA. Molec. Endocr. 4 (1990) 1782-1790. 10. Fawell S. E., Lees J. A., White R. and Parker M. G.: Characterization and colocalization of steroid binding and dimerization activities in the mouse estrogen receptor. Cell 60 (1990) 953-962. 11. Zhang X.-K., Hoffmann B., Tran P. B. V., Graupner G. and Pfahl M.: Retinoid X receptor is an auxiliary protein for thyroid hormone and retinoic acid receptors. Nature 355 (1992) 441-446.
D i m e r i z a t i o n of A n d r o g e n R e c e p t o r 12. Kliewer S. A., Umesono K., Mangelsdorf D. J. and Evans R. M.: Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D~ signalling. Nature 355 (1992) 446-449. 13. Carlberg C., Bendik I., Wyss A., Meier E., Struzenbecker L. J., Grippo J. F. and Hunziker W.: Two signalling pathways for vitamin D. Nature 361 (1993) 657-660. 14. Wong C.-I., Zhou Z.-X., Sar M. and Wilson E. M.: Steroid requirement for androgen receptor dimerization and DNA binding. J. Biol. Chem. 268 (1993) 19004-19012. 15. Laemmli U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680~85. 16. Davis B. J.: Disc electrophoresis-II. Method and application to human serum proteins. Ann. N . Y . Acad. Sci. 121 (1964) 404-436. 17. Nemoto T., Ohara-Nemoto Y., Sato N. and Ota M.: Dual roles of 90-kDa heat shock protein in the function of the mineralocorticoid receptor. J. Biochem. 113 (1993) 769-776. 18. Strfihle U., Klock U. and Sch/.itz G.: A DNA sequence of 15 base pairs is sufficient to mediate both glucocorticoid and progesterone induction of gene expression. Proc. Natn. Acad. Sci. U . S . A . 84 (1987) 7871-7875. 19. Glass C. K., Holloway J. M., Devary O. V. and Rosenfeld M. G.: The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid and estrogen response elements. Cell 54 (1988) 313-323. 20. Nemoto T., Ohara-Nemoto Y., Denis M. and Gustafsson J.-,~.: The transformed glucocorticoid receptor has a lower steroidbinding affinity than the nontransformed receptor. Biochemistry 29 (1990) 1880-1886.
233
21. Nemoto T., Ohara-Nemoto Y. and Ota M.: Association of the 90-kDa heat shock protein does not affect the ligand-binding ability of androgen receptor. J. Steroid Biochem. Molec. Biol. 42 (1992) 803-812. 22. Chang C., Kokontis J. and Liao S.: Structural analysis of complementary DNA and amino acid sequences of human and rat androgen receptors. Proc. Natn. Acad. Sci. U . S . A . 85 (1988) 7211-7215. 23. Arriza J. L., Weinberger C., Cerelli G,, Glaser T. M., Handelin B. L., Housman D. E. and Evans R. M.: Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237 (1987) 267-275. 24. Hollenberg S. M., Weinberger C., Ong E. S., Cerelli G., Oro A., Lebo R., Thompson E. M., Rosenfeld M. G. and Evans R. M.: Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318 (1985) 635-641. 25. Habig W. H., Pabst M. J. and Jakoby W. B.: Glutathione S-transferases. J. Biol. Chem. 249 (1974) 7130-7139. 26. Roehrborn C. G., Zoppi S., Gruber J. A., Wilson C. M. and McPhaul M. J.: Expression and characterization of full-length and partial human androgen receptor fusion proteins. Implications for the production and applications of soluble steroid receptors in Escherichia coli. Molec. Cell. Endocr. 84 (1992) 1-14. 27. Schwabe J. W. R., Neuhaus D. and Rhodes D.: Solution structure of the DNA-binding domain of the oestrogen receptor. Nature 348 458-461. 28. Freedman L. P., Yamamoto R. K., Luisi B. F. and Sigler P. B.: More fingers in hand. Cell 54 (1988) 445. 29. Miesfeld R., Godowski P. J., Maler B. A. and Yamamoto K. R.: Glucocorticoid receptor mutants that define a small region sufficient for enhancer activation. Science 236 (1987) 423-427.
Pergamon
0960-0760(94)00101-4
Copyright © 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0960-0760/94 $7.00 + 0.00
D i m e r i z a t i o n Characteristics of the D N A - and S t e r o i d - b i n d i n g D o m a i n s of the A n d r o g e n R e c e p t o r Takayuki
Nemoto,
1. Y u k o O h a r a - N e m o t o , and Minoru Ota I
2 Soichi
Shimazaki
1
Departments of 1Biochemistry and 2Microbiology, Iwate Medical University School of Dentistry, Morioka, Iwate 020, Japan
The DNA-binding domain (DBD) of the androgen, mineralocorticoid, and glucocorticoid receptors a n d t h e s t e r o i d - b i n d i n g d o m a i n ( S B D ) o f t h e a n d r o g e n r e c e p t o r ( A R ) w e r e e x p r e s s e d s e p a r a t e l y as f u s i o n p r o t e i n s w i t h g l u t a t h i o n e - S - t r a n s f e r a s e ( G S T ) in Escherichia coli. N a t i v e p o l y a c r y l a m i d e gel e l e c t r o p h o r e s i s a n d gel e x c l u s i o n H P L C d e m o n s t r a t e d t h a t t h e G S T - A R D B D f u s i o n p r o t e i n w a s p r e s e n t as a d i m e r . O n t h e o t h e r h a n d , t h e G S T - A R S B D f u s i o n p r o t e i n f o r m e d a h i g h - m o l e c u l a r w e i g h t o l i g o m e r , w h i c h s e e m e d to b e f o r m e d b y t w o s e p a r a t e i n t e r a c t i o n s , i.e. G S T - G S T a n d ARSBD-ARSBD between the fusion molecules. These findings strongly suggest that ARSBD has a p o t e n t a b i l i t y to f o r m a h o m o d i m e r a n d t h a t A R D B D d o e s n o t . G S T - A R D B D s p e c i f i c a l l y i n t e r a c t e d with the glucocorticoid response elements of the mouse mammary tumor virus long terminal repeat (GREMMTV). C l e a v a g e o f t h e f u s i o n p r o t e i n b y t h r o m b i n a b o l i s h e d t h e b i n d i n g , w h i l e t h e n o n s p e c i f i c DNA-cellulose binding ability was retained. Therefore, the dimeric configuration of GST-ARDBD, e v e n i f a c c o m p l i s h e d t h r o u g h t h e i n t e r a c t i o n w i t h t h e G S T m o i e t y , is n e e d e d f o r h i g h - a f f i n i t y b i n d i n g to t h e r e s p o n s e e l e m e n t . T h e b i n d i n g o f G S T - A R D B D to GRE~MTV w a s s t r o n g l y c o m p e t e d b y t h e g l u c o c o r t i c o i d r e s p o n s e e l e m e n t o f r a t t y r o s i n e a m i n o t r a n s f e r a s e gene, f o l l o w e d b y t h e a n d r o g e n r e s p o n s e e l e m e n t of the r a t p r o b a s i n gene. A p a l i n d r o m i c t h y r o i d r e s p o n s e e l e m e n t s h o w e d n o c o m p e t i t i o n . U n e x p e c t e d l y , n o a p p a r e n t d i f f e r e n t in t h e b i n d i n g a f f i n i t y to t h e s e r e s p o n s e elements was observed among the DBDs of androgen, mineralocorticoid and glucocorticoid receptors.
J. Steroid Biochem. Molec. Biol., Vol. 50, No. 5/6, pp. 225-233, 1994
INTRODUCTION
response element (GRE), is related to, but is distinct from, the H R E for the estrogen receptor (ER) [1]. M o u s e m a m m a r y t u m o r virus ( M M T V ) long terminal repeat contains several sequentially located G R E s , which also function as H R E s for androgen and progesterone receptors (AR and PR, respectively) [2]. At present, it is unclear whether or not specific response elements for androgen, mineralocorticoid and progesterone, distinct from G R E , are present. Claessens et al.[3] reported an androgen response element (ARE) in the first intron of the C3 gene of which expression is primarily regulated by androgens in vivo. However, the sequence of this half site is identical to the consensus sequence of G R E , i.e. T G T T C T ; and a transfection study revealed that the A R E of the C3 gene also functions as G R E or progesterone response element to a similar extent. Roche et al. [4] cloned several putative A R E by means of the in vitro D N A - b i n d i n g site
Steroid receptors exert their action through binding to a specific D N A sequence, called a hormone response element (HRE). T h e consensus sequence of the H R E for the glucocorticoid receptor (GR), i.e. glucocorticoid
*Correspondence to T. Nemoto. Abbreviations: AR, ER, GR, M R and PR, the androgen, estrogen,
glucocorticoid, mineralocorticoid and progesterone receptors, respectively; D B D and SBD, the respective D N A - and the steroidbinding domains of the steroid receptor; G S T , glutathioneS-transferase; M M T V , m o u s e m a m m a r y t u m o r virus; H R E , ARE, G R E and T R E , the hormone, androgen, glucocorticoid and thyroid response elements, respectively; AREpRoB , A R E 2 of the rat probasin gene; GRETAT, G R E of the rat tyrosine aminotransferase gene; GREMMTV, G R E of M M T V ; TREpAL, the palindromic T R E ; PCR, polymerase chain reaction; S D S - P A G E , polyacrylamide gel electrophoresis in the presence of s o d i u m dodecyl sulfate. Received 14 Feb. 1994; 25 Apr. 1994. 225
226
Takayuki
selection assay. The sequence is comprised of two 6-bp nucleotides with 3-bp spacers, S’-GGA/TACANNNTGTTCT-3’, which is similar to that for GREs. Very recently, Rennie et al. [5] reported two ARES of the rat probasin gene, which act together to mediate androgen-specific induction. The sequences with which the mineralocorticoid receptor (MR) specifically interacts have not been reported to date. Biochemical and crystallographic studies demonstrated that one DNA-binding domain of GR and ER interacts with a half site of HRE [6,7]. Moreover, GR, PR and ER form homodimers before binding to HRE [8-lo]. However, homodimerization is not common to all members of the steroid/thyroid receptor superfamily. At least, thyroid and retinoic acid receptors interact with cognate response elements primarily as a heterodimer with 94 retinoic acid receptor (also called RXR) [ll, 121. Vitamin D, receptor interacts with its response element either as a heterodimer with RXR or as a homodimer, depending on the structure of the response element [ 131. Recently, it has been shown that, like GR and PR, AR binds to ARE as a dimer in a ligand-dependent manner [14]. However, it has not been clear whether or not AR forms a homodimer before binding to DNA. In this study, the DBDs of AR, MR and GR, and the SBD of AR were expressed in E. coli as fusion proteins with glutathione-S-transferase (GST). Utilizing a dimerization/oligomerization characteristic of the GST-fusion proteins, we examined the dimerization ability of ARDBD and ARSBD and showed that ARSBD has the ability to form a dimeric configuration but that ARDBD does not. We also compared the DNA-binding activity of the three DBDs by the gel retardation assay. EXPERIMENTAL Materials
[3H]R1881 (80-86 Ci/mmol) was purchased from DuPont-New-England Nuclear (Boston, MA). [a-32P]dCTP (3000 Ci/mmol) was purchased from Amersham (Arlington Heights, IL). Restriction enzymes and DNA-modifying enzymes were obtained from Boehringer Mannheim, and Takara (Tokyo, Japan). An expression vector, pGEX-2T, and glutathioneSepharose 4B were from Pharmacia Biotech (Uppsala, Sweden). Thrombin (54 U/mg protein) and calf thymus double-stranded DNA-cellulose were from Sigma (St Louis, MO). All other reagents were of analytical grade. Methods Plasmid construction. The DNA fragments encoding DBDs of human AR, MR, and GR were amplified by the polymerase chain reaction (PCR). Sense primers for DBDs were designed to possess a BamHI site (italics), and the antisense primers were prepared with
Nemoto
et al.
an EcoRI site (underlined) as follows: ARDBD, 5’GTTGGA TCCATTGACTATTACTTTC-3’ and 5’-TCTGCAGAATTCGGCTCCCAGAGTCATC3’; MRDBD, 5’-ACTGGA TCCTCAAGACCTTC3’ and 5’-GACTGCAGAATTCCTAAATTCATTC-3’; GRDBD, 5’-TCCGGA TCCTCAACAGCAACAAC-3’ and 5’-TCTGCAGAATTCCAGGTTCATTCC-3’. Primers for the SBD of human AR were designed to possess a BamHI site (italics). The sequences of the sense and antisense primers were 5’-AAGCAGGA TCCACTCTGGGAGCCCG-3’ and 5’ - AGGGGATCCAATGCTTCACTGGGTG - 3’, respectively. cDNA clones encoding the full length of human AR, MR and GR, generously provided by Drs S. Liao (Chicago University), J. L. Arizza (University of California) and R. L. Evans (Salk Institute), respectively, were used as templates for the PCR. PCR was done for 25 cycles with denaturation for 2 min at 92°C annealing for 2 min at 50°C and polymerization for 2.5 min at 70°C. The amplified fragments encoding DBDs were cut with BamHI and EcoRI, and inserted into the BamHI/EcoRI site of a pGEX-2T vector. The constructed plasmids were designated pGEXARDBD for AR, pGEX-MRDBD for MR, and pGEX-GRDBD for GR. E. coli Y1090 was transformed with these plasmids separately. The amplified fragment encoding ARSBD [Fig. l(a)] was cut with BamHI and inserted into the BamHI site of pGEX-2T. E. coli Y1090 was transformed with the ligated plasmid, designated pGEX-ARSBD. A 139-bp fragment containing three GREs of the MMTV was amplified by PCR with synthetic primers: 5’-CCTTGCGGATCCCAGGGCT-3’ (BamHI site and 5’-GATTTGGATrepresented by italics) GAATTCCAAAAG-3’ (EcoRI site represented by italics). MMTV5031 (Japanese Cancer Research Resources) was used as a PCR template. The PCR product was cut with BamHI and EcoRI, and inserted into the corresponding site of pUC18, designated pUCMMTV. Induction
and purification
of recombinant
proteins.
The transformed bacteria were cultured at 37°C in 250ml LB medium with 50pg/ml ampicillin. When of the culture had reached 0.8-1.0, the temthe A,,, perature was lowered to 28°C and the bacteria were further cultured for 4-5 h in the presence of 0.2 mM B-isothiogalactopyranoside. The cells were harvested and stored at -80°C until used. The cell lysate was prepared and the recombinant proteins were purified according to the manufacturer’s protocol except that protease inhibitors (20 pgg/ml each of leupeptin, soybean trypsin inhibitor, antipain and pepstatin) were added as supplements to the lysate preparation buffer. In the case of GST-ARSBD, the sample eluted from glutathione-Sepharose contained a significant amount of a 27-kDa protein, presumably a proteolysed product of the fusion protein. This sample was further subjected to DEAE-5PW column
Dimerization of Androgen Receptor (0.75 × 5 c m , Tosoh, Japan) chromatography and eluted with a linear gradient of 0-0.6 M NaC1 in 20 m M Tris-HC1 (pH 7.3). Polyacrylamide gel electrophoresis. Bacterial lysates or purified proteins were subjected to electrophoresis at 4°C on 12.5 or 15% (w/v) polyacrylamide gels under denaturing conditions according to Laemmli [15], or on 7.5% (w/v) polyacrylamide gels under nondenaturing conditions [16].
Gel
exclusion
HPLC
on
TSK-G3OOOSWXL.
Purified proteins (0.3 ml) were subjected to gel chromatography at 4°C on a T S K - G 3 0 0 0 S W X L column (0.78 × 30 cm, Tosoh, Japan), and eluted with 20 m M Tris-HC1 (pH 7.3), 1 m M E D T A and 50 m M NaC1. Flow rate was 0.25 ml/min, and 1-min fractions were collected.
Molecular weight markers estimating molecular weights of native forms. For native polyacrylamide gel electrophoresis as well as gel exclusion H P L C analysis, ovalbumin (45-kDa), bovine serum albumin (monomer, 66-kDa; dimer, 132-kDa; trimer, 198-kDa) and catalase (240-kDa) were used as molecular weight markers. Additionally, cytochrome C (13-kDa) and sperm whale myoglobin (17-kDa) were used as markers for gel exclusion H P L C analysis. Gel retardation analysis. Interaction of the expressed proteins with specific D N A sequences was investigated by gel retardation analysis. T h e 139-bp fragment excised with EcoRI and XbaI from p U C M M T V , designated GREM~Tv, was labeled with [~-32p]dCTP and Klenow fragment, and used as a probe for gel retardation analysis. Purified proteins were subjected to gel retardation analysis as described previously [17]. After electrophoresis, the gels were dried and subjected to autoradiography. In some experiments, radioinert GRETAT, TREFA L and AREeRoB, synthesized with a Model 381A D N A synthesizer (Applied Biosystems, CA), were used as competitors. T h e sequences of GRETAx, TREpAL and AREpRoB were 5 ' - G G G T C T G C T G T A C A G GATGTTCTAGCTACGCCC-3' [18], 5 ' - G A T C CAAGATTCAGGTCATGACCTGAGGAGAGCC T A G - 3 ' [19] and 5 ' - C C G G G T A A A G T A C T C CAAGAACCTATTTGCCCGG-Y [5], respectively. RESULTS
Expression and purification of the recombinant proteins Our previous expression studies in E. coli revealed lower expression levels of larger molecules of steroid receptors [20, 21]. Accordingly, in this study, we did not express a full-length receptor, but rather expressed functional domains of AR, i.e. D B D and SBD, as fusion proteins with G S T [Fig. l(a)]. DBDs of M R and G R were also expressed [Fig. l(b)]. As shown in Fig. l(c) the amino acid sequence of A R D B D is less similar to that of M R D B D or G R D B D compared with the similarity between M R D B D and G R D B D .
227
T h e bacterially-expressed G S T - D B D fusion proteins, designated G S T - A R D B D , G S T - M R D B D and G S T - G R D B D , were analyzed by S D S - P A G E [Fig. 2(a)]. A predominant band at around Mr 35,000 was found in lysates of E. coli Y1090 [ p G E X - A R D B D , p G E X - M R D B D , and p G E X - G R D B D ] [Fig. 2(a), lanes 1-3], and at Mr 27,000 in the lysate of Y1090[pGEX-2T] (lane 4). These values are in good agreement with the molecular weights calculated from the predicted amino acid sequences of cloned receptor cDNAs [22-24]. T h e recombinant proteins were purified to near homogeneity by a single-step purification with glutathione-Sepharose affinity chromatography (lanes 5-8). In the lysate of Y1090[pGEX-ARSBD], specific [3H]R1881-binding was detected (20,455 + 477 dpm/ 0.1 ml lysate at 1 0 n M [3H]R1881), while no specific binding was found in the lysate of Y1090[pGEX-2T]. After purification by glutathione-Sepharose chromatography, Mr 60,000 and 27,000 bands were observed [Fig. 2(b), lane 2]. F r o m the calculation of the predicted molecular weight of G S T - A R S B D (60.9-kDa), the Mr 60,000 band was assigned to the fusion protein, and the band with M r 27,000 appeared to be G S T , presumably produced by proteolysis of the fusion protein. T h e Mr 60,000 band was subsequently purified to homogeneity by ion-exchange D E A E - H P L C (lane 3). T h e purified samples were used in the following experiments.
Analysis of the molecular configuration of recombinant proteins T o characterize the molecular forms of the expressed proteins, we subjected the purified proteins to native gel electrophoresis. G S T - A R D B D migrated at Mr 150,000 (Fig. 3, lane 2) on the native gel. This value was approximately four times as large as the value on S D S - P A G E (Mr 35,000), indicating a tetrameric structure of G S T - A R D B D . However, we supposed that G S T - A R D B D migrated slower than that expected from its real molecular weight, on native gel electrophoresis, because of the basic amino acid residues of D B D and its coordination with two zinc ions. In fact, the dimeric configuration of G S T - A R D B D was ascertained by gel exclusion H P L C analysis as described below (see Fig. 4). T o the contrary, most of G S T - A R S B D did not penetrate the gel, but remained concentrated at the top (Fig. 3, lane 1). This indicated that G S T - A R S B D formed an oligomeric structure with a large molecular mass. A faint band observed a t M r 140,000 may correspond to a dimer of G S T - A R S B D . T h e molecular form of the expressed proteins was further characterized by gel exclusion H P L C analysis [Fig. 4(a)]. T h e native Mr was calibrated in comparison with marker proteins [Fig. 4(b)]. It has been already shown that G S T exists as a homodimer [25]. In fact, Fig. 4 shows that G S T ( M r 27,000) eluted at the position of a dimer (Mr 55,000). G S T - A R D B D (M r
Takayuki Nemoto et al.
228
(a) AR
1
200
I
I
400 '
600"~R66 556'623
800 '
918 ,
I 548 627 ~i;i;i;;:G STi:::;iii;;]l~!ii.:..'!~
GST-ARDBD
fili~i!i i i!i i i i i~!ii i i ~
GST-ARSBD
(b) AR
1 ]i0 20 30 40 50 60 70 1 IDYYFPP~TCLICGDEASGCHYGALTCGSCKVFFKR/~%EGKQKYLCASRMDCTIDKFRRKMCPSCRLRKCYEKG~LGA
MR
80
1 ] i0 20 30 40 50 60 70 | 7 6 S S R - ~ I--V ........... W ............ V - - Q H N .... G .... I---I ...... A - - - Q - - L Q - - ~ - ~ 1
GR 1 I0] SSTATTG--~L--V-S
20 30 40 50 60 70 ]80 ......... V ............. V - - Q H N .... G .... I--- I ...... &--Y--- L Q - - ~ - E-
(c) AR AR
MR
GR
15 (22.1%)
MR
15
GR
15
15 (22.1%) 5
(7.4%)
5
Fig. 1. Schematic illustration of recombinant AR, GST-ARDBD and GST-ARSBD. (a) Human AR, GST-ARDBD and GST-ARSBD are schematically illustrated with putative DBD (amino acid residues 556-623) and SBD (amino acid residues 666-918). (b) The amino acid sequences of AR, MR and GR expressed a s GST-fusion proteins are illustrated. A box indicates the putative DBD. Arrows indicate the cysteine residues coordinated with zinc ions. (c) The number of substituted amino acids in DBD (68 amino acid residues) between the receptors and their percentages (parentheses) are represented.
35,000) also eluted at the position of a dimer (Mr 70,000). Therefore, as we postulated above, the migration of G S T - A R D B D on native gel electrophoresis was retarded due to the basic nature of DBD. T h e phenomenon was found even on S D S - P A G E of thrombin-cleaved G S T - A R D B D , i.e. A R D B D , as described below (see Fig. 6). Consistent with the results of native gel electrophoresis, G S T - A R S B D , forming a high-molecular weight oligomer, eluted at the void volume fraction of the gel chromatography (Fig. 4).
Interaction of the expressed proteins with specific D N A sequences Gel retardation analysis of G S T - A R D B D , G S T M R D B D and G S T - G R D B D was performed by use of a 139-bp fragment of GREMMTV which contains three GREs in its sequence. As shown in Fig. 5, all three fusion proteins bound to the probe, and gave similar shifted band patterns. T h e increase in input proteins resulted in heavier retardation of the D N A - D B D complexes, because several D B D s interacted with GREMMTV at higher protein concentrations. Essentially no binding was found with G S T . Unexpectedly, a higher protein concentration was needed for
G S T - G R D B D to obtain an intensity comparable to that of G S T - A R D B D and G S T - M R D B D . T h e junction of G S T - A R D B D can be cleaved by thrombin. Following thrombin cleavage, G S T A R D B D was separated into two species, i.e. G S T and A R D B D [Fig. 6(a)]. The Mr 13,000 of A R D B D obtained on S D S - P A G E (lane 2) was appreciably larger than that calculated from its c D N A seuqence (9.7kDa), because of its basic nature. T h e gel retardation analysis with the cleaved protein revealed a shifted band only at the highest protein concentration examined [Fig. 6(b), lane 10]. Accordingly, a greater than 64-fold molar excess of proteins was needed for A R D B D to obtain a p r o t e i n - D N A complex comparable to that of G S T - A R D B D [compare Fig. 6(b), lanes 2 and 10]. However, both A R D B D and G S T - A R D B D similarly bound to calf thymus DNA-cellulose [Fig. 6(c), lane 4]. These results strongly indicate that a dimeric configuration of A R D B D is required for highaffinity binding to HRE. In Fig. 6(c), although a small portion of G S T was also retained by the matrix (lane 4), this species may have been derived from a heterodimer of G S T and G S T - A R D B D bound to the matrix.
Dirnerization of Androgen Receptor (a)
229
(b)
97k
66k
45k
30k
20k
1
2
3
4
5
6
7
8
M
1
2
3
M
Fig. 2. S D S - P A G E of r e c o m b i n a n t p r o t e i n s . (a) The b a c t e r i a l lysates (lanes 1-4) a n d t h e a f f i n i t y - p u r i f i e d p r o t e i n s (lanes 5-8) of E. coli Y1090[pGEX-ARDBD], [ p G E X - M R D B D ] , [ p G E X - G R D B D ] , a n d [pGEX-2T] were a n a l y z e d on S D S - P A G E . Lanes 1 a n d 5, Y1090[pGEXARDBD]; lanes 2 a n d 6, Y1090[pGEX-MRDBD]; lanes 3 a n d 7, Y1090[pGEX-GRDBD]; lanes 4 a n d 8, Y1090[pGEX-2T]. (b) T h e lysate (lane 1), a f f i n i t y - p u r i f i e d (lane 2) a n d D E A E - H P L C p u r i f i e d (lane 3) s a m p l e s of Y1090[pGEX-ARSBD] were a n a l y z e d on S D S - P A G E . M, m o l e c u l a r weight m a r k e r s .
T h e D N A - b i n d i n g specificity of G S T - D B D fusion proteins of AR, M R and G R was examined. As a potent response element for androgen, i.e. ARE, Rennie et al. [5] recently reported two cis-acting D N A elements involved in androgen regulation of the rat probasin
gene. Here, we analyzed one of the sequences (designated AREpRoB in this report, and ARE2 in Ref.[5]), because the binding affinity of AR to this sequence is higher than that to the other. T h e binding affinities of GRETA T and TREpA c were also compared.
~~
240k 198k 132k
240k-198k--
66k 45k 132k--,
66k
.
45k --.
I
I M
I 1
2
I M
Fig. 3. Native gel e l e c t r o p h o r e s i s of p u r i f i e d p r o t e i n s . G S T - A R S B D (lane 1) a n d G S T - A R D B D (lane 2) were s u b j e c t e d to n a t i v e gel e l e c t r o p h o r e s i s . M o l e c u l a r weight m a r k e r s (M) of n a t i v e gel e l e c t r o p h o r e s i s were r u n in parallel.
230
Takayuki Nemoto et al.
(a)
(b) ~..Q--45k
G S T - A R D B D ---~
w
qllI~
GST-ARSBD
/
~-.~--30k
400
29 31 33 35 37 39 41 43 45 47 49 M
1
2
~o~ 200 ---
"~
100 80
3 o
40
~45k M
21 23 25 27 29 31 33 35 37 39
GST-AF:IDBD
/
-',~--66k
GST-ARSBD~
60
¢l 0
-20 _
:~
10 20
GST
GST~
b7
[
I
I
30
40
50
I° 8 60
Fraction number 21
23
25
27
29
31
Fraction
33
35
37
39
M
number
Fig. 4. Gel exclusion HPLC analysis on TSK-G3000SWXL. (a) GST-ARDBD, GST-ARSBD, and GST were analyzed on a TSK-G3000SWXL column equilibrated with 20 mM Tris-HC1 (pH 7.3), I mM EDTA and 50 mM NaCI. Aliquots (20/Jl) of the eluted samples were analyzed on SDS-PAGE. (b) The relative molecular weights of eluted samples in "a" are plotted. Markers are as follows: 1, bovine serum albumin tetramer (264-kDa); 2, catalase (240-kDa); 3, albumin trimer (198-kDa); 4, albumin dimer (132-kDa); 5, albumin monomer (66-kDa); 6, ovalbumin (45-kDA); 7, sperm whale myoglobin (17-kDa); 8, swine cytochrome C (13-kDa). T h e b i n d i n g of G S T - A R D B D to GREMMTV was m o s t strongly c o m p e t e d b y GRETA T a n d weakly b y AREpRoB (Fig. 7). TREpAL showed n o c o m p e t i t i o n w i t h i n the c o n c e n t r a t i o n s used in this study. C o m pletely identical results were obtained with G S T - M R D B D . I n the case of G S T - G R D B D , G R E TAT was the strongest c o m p e t i t o r as in the other cases, b u t the difference b e t w e e n AREpRoB a n d TREpA L was a m b i g u o u s because of a weak b i n d i n g of G S T G R D B D to the probe. I n c o n c l u s i o n , AREpRoB had a weaker affinity t h a n GRETA T for the b i n d i n g of GST-ARDBD, GST-MRDBD, or G S T - G R D B D .
4) a n d the oligomeric one of G S T - A R S B D (Fig. 4), G S T - A R S B D s h o u l d possess a n a d d i t i o n a l i n t e r a c t i n g GST
AR
MR
GR
DISCUSSION I n this study, we expressed D B D s of three receptors a n d A R S B D as G S T fusion proteins. T h e G S T - f u s i o n expression system has b e e n already e m p l o y e d for n u m e r o u s p r o t e i n s i n c l u d i n g A R [26]. I n this study, however, we note a novel characteristic of o u r G S T f u s i o n proteins: T h e G S T p o r t i o n of the f u s i o n p r o t e i n possesses the p r o p e r t y of d i m e r f o r m a t i o n ; a n d therefore, w h e n a fused p o r t i o n t e n d s to b i n d each other, the G S T - f u s i o n p r o t e i n results in f o r m i n g a h i g h - m o l e c u lar weight oligomer. U t i l i z i n g this u n i q u e characteristic, we analyzed the d i m e r i z a t i o n abilities of A R D B D and ARSBD. Both native gel electrophoresis a n d gel exclusion c h r o m a t o g r a p h y clearly d e m o n s t r a t e d that G S T A R S B D existed as an oligomeric form. T a k i n g into a c c o u n t the d i m e r i c form of G S T (Ref. [25] a n d Fig.
1 2 3 4 5
6
7
8
9 1011 121314
1516
F
Fig. 5. Gel retardation analysis of GST-ARDBD, GSTMRDBD and GST-GRDBD. GST, GST-ARDBD, GSTMRDBD and GST-GRDBD were incubated with the 3zp-labeled GREMMTVprobe (5000 cpm). After 20 min, DNA-protein complexes were separated on a 5% (w/v) polyacrylamide gel. Lanes 1, 5, 9 and 13, 0.5 #g proteins; lanes 2, 6, 10 and 14, 1 ~g; lanes 3, 7, 11 and 15, 2#g; lanes 4, 8, 12 and 16, 4/1g. F, free probe.
Dimerization of Androgen Receptor
(a)
231
(c)
(b)
9Y ~)7k
97k 66k 45k GST-ARDBD GST
66k 45k
30k
GST-ARDBD 30k
G ST 20k
20k
ARDBD
4k 1
2
M
14k
AR D BD F 12345
F678910
1
2
3
4
Fig. 6. Effect of thrombin cleavage on the DNA-binding of GST-ARDBD. (a) GST-ARDBD (210 gg) was incubated with or without 1.6 U thrombin at 30°C for 3 h. Aliquots of the sample uncleaved (lane 1) or cleaved (lane 2) with thrombin were analyzed on a 15% (w/v) gel of SDS-PAGE. (b) Uncleaved (GST-ARDBD) or cleaved (ARDBD and GST) sample obtained as in "a" was incubated with 32P-labeled GREM~TVand separated on a 5% polyacrylamide gel. Lanes 1 and 6, 0.1/~g; lanes 2 and 7, 0.4gg; lanes 3 and 8, 1.6 #g; lanes 4 and 9, 6.4 #g; lanes 5 and 10, 25.6 ~g proteins. (c) Thrombin-treated GST-ARDBD was applied to a DNA-cellulose column (1 x 5 cm) equilibrated with 20 m M Tris-HC1 (pH 7.3) containing 1 mM EDTA, 1 m M dithiothreitol, 10% (v/v) glycerol, and 50 mM NaCl. The column was washed with the same buffer, and then eluted with the same buffer with 0.3 M NaCI as a final concentration. Samples were analyzed by SDS-PAGE. Lane 1, untreated GST-ARDBD; lane 2, thrombin-treated GST-ARDBD; lane 3, DNA-cellulose nonadsorbed fraction; lane 4, 0.3 M NaCl-eluted fraction. Data of "a and b" and "c" are from separate experiments.
site o t h e r t h a n the G S T m o i e t y ; and, h e n c e , G S T A R S B D is c a p a b l e o f i n t e r a c t i n g w i t h t w o o t h e r m o l ecules. F i n a l l y , G S T - A R S B D f o r m s an o l i g o m e r as i l l u s t r a t e d in Fig. 8. T h u s , the o l i g o m e r i c s t r u c t u r e o f GST-ARSBD i m p l i e s that A R S B D ( m o r e precisely, a m i n o acid r e s i d u e s 6 2 4 - 9 1 8 ) h a v e t h e ability to f o r m a h o m o d i m e r . S i n c e no l i g a n d was a d d e d u n d e r these e x p e r i m e n t a l c o n d i t i o n s , l i g a n d - b i n d i n g is not req u i r e d for the d i m e r i z a t i o n o f A R S B D . P r e s u m a b l y , the l i g a n d - b i n d i n g is p r i m a r i l y n e e d e d for the l i b e r a ti on o f an as s o ci at i n g c o m p o n e n t o f the n o n D N A b i n d i n g f o r m o f h e t e r o m e r i c A R [21].
GRETA T
TREpAL AREpRoB
AR
W e p r e v i o u s l y r e p o r t e d that b act er i al l y e x p r e s s e d t r u n c a t e d A R s r e t a i n i n g D B D an d S B D ( A R 4 3 8 ) or S B D ( A R 6 1 2 ) s e d i m e n t e d at 4 - 5 S an d 3.5--4.5S, resp ect i v el y , u p o n g l y c e r o l g r a d i e n t c e n t r i f u g a t i o n [21]. T h o s e species m a y r e p r e s e n t d i m e r i c f o r m s . M o r e o v e r , the s t u d y s h o w e d that A R S B D itself does not f o r m an o l i g o m e r i c s t r u c t u r e such as G S T - A R S B D . T o directly d e m o n s t r a t e a d i m e r i c f o r m o f A R S B D , we a t t e m p t e d to separate G S T an d A R S B D by t h r o m b i n cleavage. H o w e v e r , G S T - A R S B D was n e v e r c l e a v e d for s o m e u n k n o w n reason. W h i l e t r y p s i n efficiently d e g r a d e d G S T - A R S B D , a s t a b l y - e x i s t i n g species c o r -
GRETAT TREpAL AREpRoB MR
12345
6 7 8 10 1113 15 9 12 14
GRETAT TREpAL AREpRoB GR
F12345
678
........................... 678 10 11 13 15 9 12 14 Fig. 7. DNA-binding specificity of GST-ARDBD, GST-MRDBD and GST-GRDBD. GST-ARDBD (0.75 ~g), GST-MRDBD (0.75 gg), and GST-GRDBD (4 gg) were incubated with 32P-labeled GRE-MMTV in the absence or presence of competitors. DNA-protein complexes were separated on a polyacrylamide gel. Lanes 1, 6 and 11, no competitor; lanes 2, 7 and 12, 0.04 pg; lanes 3, 8 and 13, 0.2/~g; lanes 4, 9 and 14, 1/~g; lanes 5, 10 and 15, 5 ~g competitors. F, free probe. 9
10 11 13 15 12 14
12 3 4 5
232
Takayuki Nemoto et al.
GST-ARDBD
GST-ARSBD GST
"--TD
ARDBD
Fig. 8. Model of dimerization and oligomerization of G S T fusion proteins. The model of dimerization of G S T - A R D B D and oligomerization of GST-ARSBD are schematically represented.
r e s p o n d i n g to the Mr o f A R S B D was n o t o b t a i n e d (data not shown). O n the o t h e r h an d , the d i m e r i c s t r u c t u r e o f G S T A R D B D i m p l i es that A R D B D ( m o r e precisely, a m i n o acid residues 548-627) has no i n t e r a c t i n g site. In o t h e r w o r d s , A R D B D itself does not f o r m a h o m o d i m e r . It has already b e e n a c c e p t e d that t w o m o l e c u l e s o f D B D o f G R and E R b i n d to a p a l i n d r o m i c s t r u c t u r e o f H R E . M o r e o v e r , G R [8], E R [10] and P R [9] f o r m h o m o d ime rs after l i b e r a t i n g b i n d i n g c o m p o n e n t s , such as 9 0 - k D a heat shock p r o t e i n , b e f o r e b i n d i n g w i t h H R E . N M R an d c r y s t a l l o g r a p h i c studies on G R a n d E R clearly s h o w e d that the D B D s i n t e r a c t w i t h each o t h e r p r i m a r i l y w i t h i n the s e c o n d zinc finger [6, 7]. H o w e v e r , this i n t e r a c t i o n ap p e a r s to be insufficient to f o r m a d i m e r i c c o m p l e x w h e n the t w o are not associated w i t h H R E [27, 28]. T h i s s t u d y also d e m o n s t r a t e d that A R D B D alone does not h o m o d i m e r i z e . M o r e o v e r , o u r data s t r o n g l y suggest that the native A R m o l e c u l e h o m o d i m e r i z e s t h r o u g h the i n t e r a c t i o n o f S B D . H e n c e , one o f the f u n c t i o n s o f S B D m a y be the stabilization o f the d i m e r i c s t r u c t u r e o f D B D . T h e dimeric configuration of G S T - A R D B D m a y be app r e c i a b l y d i f f eren t f r o m that o f an intact A R . N e v e r t h e less, it s eem s i m p o r t a n t to note that, at least w i t h r e spe ct to e n h a n c e m e n t o f D N A - b i n d i n g activity o f A R D B D , G S T is able to s u b s t i t u t e the f u n c t i o n o f ARSBD. In this study, we did not analyze the N - t e r m i n a l A / B d o m a i n o f A R D B D , w h i c h m i g h t be i n v o l v e d in the d i m e r i z a t i o n o f A R . I n fact, W o n g et al. [14] r e p o r t e d that A R d i m e r i z a t i o n r e q u i r e s a n d r o g e n - b i n d i n g only w h e n the N - t e r m i n a l A / B d o m a i n o f A R is p r e s e n t , and s u g g e s t e d that the N - t e r m i n a l A / B d o m a i n acts to i n h i b i t A R d i m e r i z a t i o n a n d D N A - b i n d i n g . O n the o t h e r h a n d , this d o m a i n is n e e d e d for the full activity o f t r a n s c r i p t i o n a c t i v a t i o n o f G R , and is a s s u m e d to
i n t e r a c t w i t h r e c e p t o r - s p e c i f i c t r a n s c r i p t i o n factors [29]. F u r t h e r studies are n e e d e d to e l u c i d a t e the f u n c tion o f the N - t e r m i n a l A / B d o m a i n o f steroid receptors. T h e D N A - b i n d i n g specificity o f the t h r e e D B D s f u sed to G S T d i d not reveal any difference w i t h t h r e e r e s p o n s e e l e m e n t s ; all t h r e e D B D s possessed a h i g h e r b i n d i n g affinity for GRETA-r. A l t h o u g h the a m i n o acid s e q u e n c e o f A R D B D is 2 2 . 1 % different f r o m t hos e s e q u e n c e s o f M R and G R [Fig. l ( b and c)], the difference seems n o t to affect the D N A - b i n d i n g p r o p erties o f A R , at least n o t its in v i t r o D N A - b i n d i n g characteristics. T h e r e f o r e , the D N A s e q u e n c e o f H R E an d a m i n o acid s e q u e n c e s o f D B D do n o t a p p e a r the p r i m a r y d e t e r m i n a n t for the specificity o f the actions o f a n d r o g e n , m i n e r a l o c o r t i c o i d and g l u c o c o r t i c o i d . Acknowledgements--We thank Dr S. Liao (Chicago University) for
the rat androgen receptor cDNA plasmid, Dr R. Evans (Salk Institute) for human GR cDNA plasmid, and Dr J. L. Ariza (University of California) for human MR cDNA plasmid, pMR3750. This work was supported by grants-in-aid for scientific research for the Ministry of Education, Science and Culture of Japan (to T.N., Y.O.-N. and M.O.) and from the Asahi Glass Foundation (to T.N.).
REFERENCES
1. Klock G., Str/ihle U. and Schfitz G.: Oestrogen and glucocorticold response elements are closely related but distinct. Nature 329 (1987) 734-736. 2. Cato A. C. B., Henderson D. and Ponta H.: The hormone response element of the mouse mammary tumour virus DNA mediates the progestin and androgen induction of transcription in the proviral long terminal repeat region. E M B O J 6 (1987) 363-368. 3. Claessens F., Celis B., Peeters W., Heyns G., Verhoeven G. and Rombauts W.: Functional characterization of an androgen response element in the first intron of the C3(1) gene of prostatic binding protein. Biochem. Biophys. Res. Commun. 164 (1989) 833-840. 4. Roche P. J., Hoare S. A. and Parker M. G.: A consensus DNA-binding site for the androgen receptor. Molec. Endocr. 12 (1992) 2229-2235. 5. Rennie P. S., Bruchovsky N., Leco K. J., Sheppard P. C., McQueen S. A., Cheng H., Snoek R., Hamel A., Bock M. E., MacDonald B. S., Nickel B. E., Chang C., Liao S., Cattini P. A. and Matusik R. J.: Characterization of two cis-acting DNA elements involved in the androgen regulation of the probasin gene. Molec. Endocr. 7 (1993) 23-36. 6. H~ird T., Kellenbach E., Boelens R., Maler B. A., Dahlman K., Freedman L. P., Carlstedt-Duke J., Yamamoto K. R., Gustafsson J.-/k. and Kaptein R.: Solution structure of the glucocorticoid receptor. Nature 249 (1990) 157-160. 7. Luisi B. F., Xu W. X., Otwinowski Z., Freedman L. P., Yamamoto K. R. and Siegler P. B.: Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352 (1991) 497-505. 8. Cairns W., Cairns C., Pongratz I., Poellinger L. and Okret S.: Assembly of a glucocorticoid receptor complex prior to DNA binding enhances its specific interaction with a glucocorticoid response element. J. Biol. Chem. 266 (1991) 11221-11226. 9. Rodriguez R., Weigel N+ L., O'Malley B. W. and Schrader W. T.: Dimerization of the chicken progesterone receptor in vitro can occur in the absence of hormone and DNA. Molec. Endocr. 4 (1990) 1782-1790. 10. Fawell S. E., Lees J. A., White R. and Parker M. G.: Characterization and colocalization of steroid binding and dimerization activities in the mouse estrogen receptor. Cell 60 (1990) 953-962. 11. Zhang X.-K., Hoffmann B., Tran P. B. V., Graupner G. and Pfahl M.: Retinoid X receptor is an auxiliary protein for thyroid hormone and retinoic acid receptors. Nature 355 (1992) 441-446.
D i m e r i z a t i o n of A n d r o g e n R e c e p t o r 12. Kliewer S. A., Umesono K., Mangelsdorf D. J. and Evans R. M.: Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D~ signalling. Nature 355 (1992) 446-449. 13. Carlberg C., Bendik I., Wyss A., Meier E., Struzenbecker L. J., Grippo J. F. and Hunziker W.: Two signalling pathways for vitamin D. Nature 361 (1993) 657-660. 14. Wong C.-I., Zhou Z.-X., Sar M. and Wilson E. M.: Steroid requirement for androgen receptor dimerization and DNA binding. J. Biol. Chem. 268 (1993) 19004-19012. 15. Laemmli U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680~85. 16. Davis B. J.: Disc electrophoresis-II. Method and application to human serum proteins. Ann. N . Y . Acad. Sci. 121 (1964) 404-436. 17. Nemoto T., Ohara-Nemoto Y., Sato N. and Ota M.: Dual roles of 90-kDa heat shock protein in the function of the mineralocorticoid receptor. J. Biochem. 113 (1993) 769-776. 18. Strfihle U., Klock U. and Sch/.itz G.: A DNA sequence of 15 base pairs is sufficient to mediate both glucocorticoid and progesterone induction of gene expression. Proc. Natn. Acad. Sci. U . S . A . 84 (1987) 7871-7875. 19. Glass C. K., Holloway J. M., Devary O. V. and Rosenfeld M. G.: The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid and estrogen response elements. Cell 54 (1988) 313-323. 20. Nemoto T., Ohara-Nemoto Y., Denis M. and Gustafsson J.-,~.: The transformed glucocorticoid receptor has a lower steroidbinding affinity than the nontransformed receptor. Biochemistry 29 (1990) 1880-1886.
233
21. Nemoto T., Ohara-Nemoto Y. and Ota M.: Association of the 90-kDa heat shock protein does not affect the ligand-binding ability of androgen receptor. J. Steroid Biochem. Molec. Biol. 42 (1992) 803-812. 22. Chang C., Kokontis J. and Liao S.: Structural analysis of complementary DNA and amino acid sequences of human and rat androgen receptors. Proc. Natn. Acad. Sci. U . S . A . 85 (1988) 7211-7215. 23. Arriza J. L., Weinberger C., Cerelli G,, Glaser T. M., Handelin B. L., Housman D. E. and Evans R. M.: Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237 (1987) 267-275. 24. Hollenberg S. M., Weinberger C., Ong E. S., Cerelli G., Oro A., Lebo R., Thompson E. M., Rosenfeld M. G. and Evans R. M.: Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318 (1985) 635-641. 25. Habig W. H., Pabst M. J. and Jakoby W. B.: Glutathione S-transferases. J. Biol. Chem. 249 (1974) 7130-7139. 26. Roehrborn C. G., Zoppi S., Gruber J. A., Wilson C. M. and McPhaul M. J.: Expression and characterization of full-length and partial human androgen receptor fusion proteins. Implications for the production and applications of soluble steroid receptors in Escherichia coli. Molec. Cell. Endocr. 84 (1992) 1-14. 27. Schwabe J. W. R., Neuhaus D. and Rhodes D.: Solution structure of the DNA-binding domain of the oestrogen receptor. Nature 348 458-461. 28. Freedman L. P., Yamamoto R. K., Luisi B. F. and Sigler P. B.: More fingers in hand. Cell 54 (1988) 445. 29. Miesfeld R., Godowski P. J., Maler B. A. and Yamamoto K. R.: Glucocorticoid receptor mutants that define a small region sufficient for enhancer activation. Science 236 (1987) 423-427.