- Email: [email protected]
Steroid-affinity purification of the rat liver glucocorticoid hormone receptor complex
J. steroid Biochem. Vol. 28, No.
6, pp. 76~717, 1987 Printed in Great Britain. All rights reserved
0022-4731/87$3.00+ 0.00 Copyright 0 1987Pergamon Journals Ltd
STEROID-AFFINITY PURIFICATION OF THE RAT LIVER GLUCOCORTICOID HORMONE RECEPTOR COMPLEX JANET P. HAPGOOD and CLAUS VON HOLT* The UCT-CSIR Research Centre for Molecular Biology, Department of Biochemistry, University of Cape Town, Republic of South Africa (Received 2 March 1987)
Summary-The molybdate-stabilized GHRC was isolated from rat liver cytosol with a 9000-fold purification and 46% yield. The major purification step was achieved using an affinity matrix consisting of an agarose support coupled to a dexamethasone ligand via an aliphatic spacer arm. Spacer arms containing disulfide bridges were found to be unsuitable due to their instability in cytosol. To reduce the non-specific binding properties of the affinity matrix, underivatized amino groups were acetylated, since the receptor was found to bind avidly to such groups thus evading elution by the ligand. Sodium molybdate present during biospecific elution from the gel stabilized the steroid-binding activity of the receptor. The use of denaturing and sulfhydryl modifying reagents (NaSCN, DMSO, Mersalyl) during elution led to partial or complete irreversible loss of steroid-binding activity of the unoccupied receptor. Efficient biospecific elution occurred at competing concentration of high affinity steroid in the presence of sodium molybdate. The ligand specific eluate was further purified by DEAE-Sephacel chromatography resulting in additional purification of 3.2-fold. The GHRC eluted from the DEAE-Sephacel column at a salt concentration characteristic of the untransformed GHRC. Molybdate was removed from the purified untransformed GHRC in the ligand eluate by DEAE-Sephacel chromatography in the absence of molybdate, for subsequent heat transformation.
INTRODUCTION
The cDNA of a 90-kDa glucocorticoid-binding protein has been identified and sequenced [l]. However, the isolation and characterization of the GHRC itself from cells is still controversial. It is necessary to purify the GHRC about 104-fold from cytosol to obtain a homogeneous preparation suitable for biochemical investigation. With increasing purification GHRC is known to become highly unstable due to its susceptibility to proteolysis [2, 31 and its tendency to lose steroid-binding activity [4-71. Despite this lability, several authors have reported the extensive purification of a steroid-binding receptor protein of approx 90 kDa. Such purifications have usually been accomplished using successive ion-exchange chromatography and DNA-cellulose chromatography, sometimes in combination with steroid-affinity *Correspondence: UCT-CSIR Research Centre for Molecular Biology, Department of Biochemistry, University of Cape Town, Private Bag, Rondebosch 7700, Republic of South Africa. Abbreviufions: Dexamethasone, 9a-fluoro- 16a-methyl-l l~, 17a,21-trihydroxy-pregna- 1,4-diene-3,20-dione; TA, triamcinolone acetonide, 9cc-fluoro-16a-methyl-ll/?,17a, 21-trihydroxypregna-1,4-diene-3,20-dione; GHRC, glucocorticoid hormone receptor complex; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrouhoresis: dam, disintegrations per min; DMF, dimethyl‘formamide; BSA, bovine ser;m albumin; TLC, thinlayer chromatography; NaSCN, sodium thiocyanate; DMSO, dimethylsulfoxide; Mersalyl, [3-[[2-(carboxymethoxy)-benzoyl]amino]-2-methoxypropyl]hydroxymercury monosodium salt; DTT, dithiothreitol.
[g-13]. The stabilizing effect of molybdate has been utilized to enrich the untransformed GHRC by steroid-affinity chromatography [6,7, 14, 151. 10% [6] to ~45% [7] pure molybdate-stabilised untransformed GHRC has been obtained with yields of 18% [14] to 46% [6]. Dexamethasone-containing matrices have resulted in purifications ranging from 900 to 8800-fold [7,9, 11, 13, 141. The great variation in the efficiency of similar matrices reflects the complexity of the problems inherent in the procedure. We report on the optimisation of conditions for the steroidaffinity purification of the transformed and untransformed GHRC. Mock steroid-affinity isolations can be performed by incubation of cytosol with affinity matrix in the presence of excess free dexamethasone to prevent receptor binding to the gel [15-181, followed by elution from the affinity matrix and DEAE-Sephacel chromatography as for experimental isolations. Such mock isolations aid in the identification of contaminants co-purifying with the receptor. We have previously shown that a 90-92 kDa non-steroidbinding protein as well as a Mg2+ dependent protein kinase copurify with the GHRC as contaminants [16], and have identified now other contaminants.
matrices
EXPERIMENTAL Materials
All glassware was acid cleaned and silylated. [ 1,2,4(n)-3 H]Triamcinolone acetonide (25 Ci/mmol) was from Amersham. All reagents used were ana769
110
JANETP. HAPGCODand CLAUS VON HOLT
lytical grade. Serum containing polyclonal antiglucocorticoid receptor antibodies raised in rabbits was a kind gift from Dr M. V. Govindan [ll]. The antigen was the 92-kDa transformed GHRC purified from rat liver cytosol by DNA-cellulose chromatography followed by preparative gel electrophoresis [ll]. The antibody is highly specific for a 92 kDa protein and does not cross-react with other cytoplasmic proteins (Fig. 5). Affi-Gel 102, an amino substituted agarose matrix, was from Bio-Rad. Phosphocellulose Pll was from Whatman and DEAESephacel from Pharmacia. Synthesis of dexamethasone derivatives
Synthesis of the 17P-carboxylic acid of dexamethasone by periodic acid oxidation was as described by Govindan and Manz [9]. The product migrated as a single spot on TLC (solvent system (1): R, = 0.06, solvent system (2): R, = 0.59). The N-hydroxybenzotriazole (HOBT) ester of the acid was prepared according to [7]. A single steroid ester spot was obtained on TLC system (1) Rr = 0.85 and system (2) Rf = 0.79. TLC was performed on silica gel plates with fluorescent indicator. Solvent system (1) = chloroform-methanol (9: l), v/v) and solvent system (2) = ethylacetate-ethanol-NH, (25% w/v H,O); (S:S:l, by vol). Preparation of affinity matrices Synthesis of Aji-Gel 102-Dexamethasone (Ajkity matrix 1) [l l] Affi-Gel 102 (40 ml, 14 pmol amino
groups/ml) was equilibrated in DMF and then suspended in 40ml of DMF containing 0.5 mmol of radioactive steroid ester. The specific activity of the ester was chosen such as to allow the monitoring of the washing procedure and the yield of bound steroid. The suspension was gently rolled at 20°C for 48 h. The affinity matrix was washed with 400 ml DMF, 5 1 absolute ethanol and 3 1 of water on a sintered glass funnel, after which it was stored at 4°C in the presence of 0.02% sodium azide. The presence of free steroid in the affinity matrix was monitored during the prolonged washing by the method of Sica et al. 1191 until the concentration of free steroid was 10.5 pmol/ml of affinity matrix. The concentration of bound ligand was determined by oxidation of a lyophilised measured volume of affinity matrix and counting the [3H]H,0 liberated. The affinity matrix prepared as above contained 2pmol steroid/ml packed gel. By decreasing the amount of ester during synthesis, matrices containing 0.1-2.0 pmol steroid/ ml gel were synthesized. Synthesis qf CL-Sepharose 4B-aminohexylsuccinylamidocystamine-dexamethasone (Aflnity matrix II)
CL-sepharose 4B (SOml) was activated with 15 g of CNBr and immediately derivatised with hexamethylenediamine (100 mmol) as described in [9,20]. The extent of derivatization was monitored by measuring the concentration of amino groups introduced
[21]. Aminohexyl-CL-Sepharose 48 (5 pmol aminohexyl groups/ml) was then succinylated with succinic anhydride [20] and coupled to cystaminium dichloride at 20°C as follows: The succinylated matrix (0.1 pmol amino groups/ml) was suspended in 25 ml of 40% DMF. Cystaminium hydrochloride (7 mmol) in 25 ml of 40% DMF was added and the pH brought to 4.5 with 1 M HCl. 1-ethyl-3-(3-diaminopropyl) carbodiimide (25 mmol) dissolved in 5 ml of water was then added over 10 min while shaking and maintaining the pH at 4.5. Thereafter, the suspension was shaken overnight, washed with 40% DMF and water and equilibrated in DMF. Amidocystamidosuccinamidohexyl-Cl-Sepharose 4B (5 pmol amino groups/ml) was coupled to the ester of dexamethasone-17/l-carboxylic acid as described for the synthesis of affinity matrix I. Affinity matrices containing 0.25 and 2.0 pmol steroid/ml of gel were synthesized. Synthesis of A#i-Gel 102-succinyl-amidocystamine dexamethasone (AfJinity matrix III)
Affi-Gel 102 was succinylated with 99% efficiency, coupled to cystaminium dichloride (to give 13.8 pmol amino groups/ml) and then reacted with steroid ester as described for affinity matrix II. Acetylation of gels
Acetylation was performed with distilled acetic anhydride in the presence of 50% saturated sodium acetate at 4C [22]. The gel was washed and suspended in an equal volume of this solution. 6 x 20 ~1 aliquots of acetic anhydride/ml of gel were added over a period of 1 h with gentle shaking. Thereafter the substituted gel was left to shake for 1 h and finally washed with sodium acetate and water. This procedure prevents the acetylation of hydroxyl groups and is selective for amino groups [22]. Regeneration of aflnity matrix 40 ml of used gel was washed with 11 of 1% SDS (6O”Q 2 1 of water (6O”Q then 1 1 of 1 M NaCl. This was followed by a wash with 11 of 50% ethanol and 3 1 of absolute ethanol to remove free steroid. If the concentration of free steroid was >O.S pmol/ml, the washing was continued until this level was attained. Finally the gel was equilibrated in buffer 1 by washing with 1 1 of water and 1 I of buffer 1. Washing was performed successively on a sintered glass funnel over a period of 3 h. Buffers
Buffer 1 was sodium phosphate (20mM), EDTA (1 mM), fl-mercaptoethanol (5 mM), sodium chloride (SOmM) and glycerol (lo%), pH 7.0. Buffer 2 was Tris (10 mM), EDTA (1 mM), fl-mercaptoethanol (5 mM) and glycerol (lo%), pH 7.4 (4°C). Buffer 3 was potassium phosphate (SOmM), EDTA (1 mM), /I-mercaptoethanol (5 mM) and glycerol (lo%), pH 7.0. Buffer 4 was Tris (10 mM), EDTA (1 mM), fl-mercaptoethanol (5 mM), glycerol (lo%), NaCl
111
Glucocorticoid receptor affinity purification
(50 mM) and BSA (0.2%), pH 7.5 (4°C). Buffer 5 was thiol-free buffer 1. For all buffers, plus molybdate refers to the addition of 10mM sodium molybdate, unless otherwise stated. Optimized isolation procedure
All procedures were performed at 4°C. Buffers contained 10mM molybdate unless otherwise indicated. Cytosol (200ml) was prepared from livers of 15 non-adrenalectomized male Wistar rats in buffer 1 [16]. The cytosol was then immediately passed at a flow rate of 20 ml/min over phosphocellulose (150 ml) packed in a silylated glass column (15 x 7 cm). Additional sodium molybdate (10 mM) was added to the clear yellow flow-through. The flow-through (200 ml) was incubated with acetylated affinity matrix I (40 ml) in a glass reagent bottle. The suspension was gently rolled on a Coulter mixer for 16 h. To determine the amount of receptor bound to the matrix, an aliquot of cytosol before and after incubation with affinity matrix was assayed for receptor content. The gel was then transferred to a sintered glass funnel and washed under low suction with 15 x 100 ml of buffer 1, 2 x 100 ml of buffer 1 plus 0.2 M NaCl then with 15 x 100 ml of buffer 2 or buffer 3. The washing of the affinity matrix was performed over a period of 90 min. The final wash did not absorb at 280 nm. The affinity matrix was then transferred to a 100 ml glass bottle and an equal volume of eluting buffer, 2 PM t3H]TA (5 pCi/nmol) in buffer 2 or 3, was added. The mixture was then gently rolled for 16 h and the eluate collected after centrifugation of the affinity matrix at 1,OOOgin a glass tube and a wash with one affinity matrix volume of elution buffer. The pooled eluates were then applied to a DEAE-Sephacel column (2 ml bed volume). Washing and elution were performed as described below. The GHRC eluting from the ionexchange column was stored in 50% glycerol in the presence of 1 PM [3H]TA at -20°C. DEAE-Sephacel
ion-exchange chromatography
The untransformed, molybdate-stabilized GHRC in the affinity matrix eluate was subjected to DEAESephacel chromatography in the absence or presence of sodium molybdate in buffer 3. DEAE-Sephacel was equilibrated by suspension and decanting twice with 20 vol of 0.5 M potassium phosphate pH 7.0. The ion-exchange gel was then packed into a glass column (0.5 x 5.0 cm) with a silylated glass wool plug and washed with 20 column volumes of buffer 3. 2 ml of packed DEAE-Sephacel was used for 40ml of affinity eluate at a flow rate of 0.3 ml/min. Affinity eluate in molybdate-containing buffer was applied to the column. Once all the affinity eluate had passed through the column, the column was washed with 10 ml of buffer in the absence or presence of molyb*The term matrix refers to the polymer substituted with linker arm and affinity matrix refers to the steroidsubstituted matrix. The term gel is used generally and refers to matrix or affinity matrix.
date as indicated in figures and tables. The GHRC was eluted with a 40 ml linear gradient of 50-500 mM potassium phosphate in buffer 3 with or without molybdate after washing with 10ml of this buffer. Occasionally buffer 2 instead of buffer 3 was used. In that case elution was performed with 40 ml of a linear O-500 mM NaCl gradient in buffer 2 with or without molybdate. Receptor content was determined by incubation of cytosol with 25 nM [3H]TA in the absence and presence of 5 PM TA for 16 h at 4°C followed by dextran/charcoal treatment [23] and liquid scintillation counting. SDS-PAGE, Western blotting and immunoassay were performed as described in [16]. In all cases,
control incubations with preimmune serum gave negative results. Siher staining was by the method of Ansorg [24]. Protein determination was by a modified Lowry method, with BSA as standard [25,26]. The sensitivity was 1 pg of protein and samples were analysed in triplicate. Radioactiuity determination
Determination of tritium activity in matrix bound steroid ligand was done after oxidation to [3H]Hr0 in a sample oxidizer (Packard Instruments) via liquid scintillation counting. Samples containing soluble radioactive compounds were dissolved in a suitable scintillation cocktail and likewise counted in a scintillation spectrometer. All counts were quench corrected for dpm. RESULTS
Afinity matrices *
TWO types of affinity matrix (Fig. 1) with different linker arms were synthesized. Affinity matrices II and III contain a cleavable disulfide bond in the linker arm, while affinity matrix I contains amide bonds in the hydrocarbon linker arm. Both types of linker arm have been used previously by others [6,7,9, 11, 14, 15, 27-301. Binding of receptor to a$nity matrix
Whereas the steroid ligand coupled to affinity matrix I was stable during incubation with cytosol for 16 h, affinity matrices II and III were unsuitable due to the lability of the disulfide bond in untreated cytosol. 50% of bound steroid was cleaved off affinity matrices II cr III (2 pmol steroid/ml packed gel) after incubation of 1 vol of affinity matrix with 5 vol of cytosol in buffer 5 for 3 h. Gel filtration or ammonium sulfate precipitation of cytosol prior to incubation with affinity matrix removed the reducing capability of cytosol, whereas phosphocellulose treatment did not (data not shown). The non-specific binding of receptor to the matrices could be almost entirely eliminated by acetylation of underivatized amino groups for matrix I (Table 1). However matrices II and III still exhibited high
JANET
P. HAPGOODand CLAUSVONHOLT Table 1. Effect of acetvlation of gels on recwtor binding
OCH,CONH(CH,),NH \
I I I I II II II II III III III III CL-Seoharose 4B
Affi-Gel 102-dexamethasone AFFINITY MAlRIX 1
0, /ImccH2);2~ ) 0
A
co22 OH
NHKH,),SS(CH,),NH \
?
/’
,
c=o -_OH
95 17
'-CH,
CL-Sepharose midocystaine-
Steroid (2 pmol/ml gel) _ _ + + _ + + _ + + -
Matrix
c-o
‘YIL-J 4B-minohexyl-succinyldexamethasone
AFFININ MATRIXII 0CH$ONH(CH2)zNH
Acetylation
44 3 86 46 84 63 95 76 64 46 68 41 2
+ + + + + + -
Matrices (100 PI), derivatized as shown in the table were gently shaken for 3 h at 4°C with cytosol (500~1) from adrenalectomized male rat livers in buffer 1. In the case of gels containing cystamine derivatives, cytosol was desalted prior to incubation and fl-mercaptoethanol was omitted from the buffers. The fraction of receptor bound to the gel was calculated after determination of the receptor content in cytosol before and after incubation.
Washing of afinity matrix after receptor binding
The affinity matrix was washed at 4°C batchwise on a silylated glass funnel under low suction with a total of 80 vol of molybdate-containing buffer (see Experimental) to remove the bulk of the contaminating proteins. The matrix was resuspended gently with a spatula between washes. Small amounts of the receptor leaked off the gel under more rigorous conditions.
c:R
I co2 2 NH(C%)2SS(CH2&NH
% bound receptor
100
.
.
1.
-.
\
Affi-Gel
102-succinyl-amidocystamine-
dexamethasone AFFINITY
WURIX III
Fig. 1. Affinity matrices for glucocorticoid receptor with non-cleavable and cleavable linker arms.
non-specific binding after acetylation, probably due to functional groups introduced during activation and coupling [7] and as the result of exchange reactions with cytosolic thiol proteins. The chemical reactions used in preparation of the affinity matrices did not affect the integrity as a ligand of the dexamethasone derivative (Fig. 2). The optimal ligand concentration on acetylated affinity matrix I was 2.0 pmol steroid/ml packed gel (Fig. 3). Optimal binding of receptor was at a ratio of 1 vol of acetylated affinity matrix I to 5 vol of cytosol for 16 h at 4°C (Fig. 4). Under these conditions the non-specific receptor binding was 5% and the specific binding was 70-80%.
,\,
I 1
0
ml
0
. 3
2
Ccompetitorl r3H triamcinolone
acetonidel
Fig. 2. Competition binding of steroid derivatives to receptor: [)H]Dexamethasone 17fi-CONH(CH,),SH was cleaved off affinity matrix II with b-mercaptoethanol to obtain a solution of this derivative in buffer 1. Increasing concentrations of dexamethasone 17/3-(2-mercaptoethylamide) dexamethasone and dexamethasone (A) (O), 17fl-carboxylic acid (0) were incubated with cytosol (100~1) from adrenalectomized rats in the presence of 10 nM 13H]TA. Incubations were for 16 h at 4°C in a final volume of 250 ~1 in buffer I plus molybdate. Thereafter, the amount of [‘HJTA bound to receptors was measured by the dextranicharcoal assay. 100% specific binding was taken as the difference in binding between [‘HITA and [‘HITA plus IOpM TA.
Glucocorticoid receptor affinity purification
173
Table 2. Summary of elution conditions Experiment (corresponds to lanes l-14
Eluant (in buffer 2 or buffer 5) plus
in Fig. 5)
I
Gel acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acptylated acetylated acetylated
2 3 4 5 6 I 8 9 10 II 12 13 14
Affi-Gel 102 affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix
I I I I I I I I I I I I I
Time of elution h
Cytosol
molybdate
adren adren adren adren adren adren adren adren adren adren non-adren t steroid adren + steroid non-adren adren
5 + Mersalyl 5 2 2tTA 2+DMSO+TA 2 t DMSO 2+TA 2 2 + NASCN 2 t NASCN + TA 5 t Mersalyl 5 t Mersalyl 5 t Mersalyl 5 + Mersalyl
1
I 5 5 5 5 16 16 5 5
I I
1 I
Receptor was eluted from acetylated affinity matrix I under the conditions summarized in the table (details in legend to Fig. 5). The eluates were then assayed for the presence of receptor and the results are shown in Fig. 5.
75
1 Ligond
I
I
I
0.5
1.0
1.5
concentration
(pm01
I 2.0
steroid/ml
/’
, t7
25
./* /
gel)
Fig. 3. Specific binding of receptor to acetylated affinity matrix I: 100~1 of acetylated affinity matrix I containing O&2.0 pmol steroid/ml of gel was incubated with 500 ~1 of cytosol for 3 h at 4°C as previously described (Table 1). After incubation, the percentage receptor binding to the affinity matrices was determined as described for Table 1.
__..---.(A)
_-.-
,./’
.z fr E 2 z
.- ..A-A-.-._._ __>_-_-..---.. .-(6)
r’ 75-J
._--
-._
i .,A’
- - ;-2”
.y
I1
Table 3. Steroid-binding activity of eluted receptor Steroid-binding activity (% of input)
Elution conditions
60 18
[‘HITA, buffer 2 plus molybdate, 5 h [‘HITA, buffer 2 minus molybdate, 5 h [‘HITA, buffer 2 plus molybdate, plus NASCN, 5 h [‘HITA, buffer 2 plus molybdate, plus DMSO, 5 h Mersalyl, buffer 5 plus, molybdate
35 41
Ih
(I)’ (2)’
1;
Phosphocellulose-treated cytosol was incubated with acetylated affinity matrix I followed by washing of the gel as described in Fig. 5. The gels were washed with buffer 2 or buffer 5 in the absence or presence of molybdate as indicated in Table 2. The concentrations of eluants were as indicated in Fig. 5. The eluates were harvested and assayed for [‘HITA-bound receptor by gel filtration on Sephadex G-75 in buffer 4 plus molybdate. The steroid-binding activity was calculated as a percentage of the maximum possible eluted steroid-binding activity for a known amount of receptor bound to the gel. In the case of Mersalyl elution, the eluates were reduced and Mersalyl was removed followed by incubation with [‘HITA by methods I or 2* before
determination of steroid-bound receptor. Method I involved gel filtration in buffer 4 containing DTT (25 mM) and molybdate (IOmM), followed by incubation of the void volume with [‘H]TA (I pM) for I6 h at 4°C. Method 2 involved exhaustive dialysis against buffer 2 containing DTT (25 mM) and molybdate (1OmM) followed by incubation with [‘HITA as for
method I. sa
28,6-M
[*.._*_;-;-;-;4
8
12
16
Time (hr) Fig. 4. Binding of receptor to matrix I and affinity matrix I; effects of acetylation, time of incubation and ratio of cytosol to gel: Matrix I (--A--), acetylated matrix I (--A-), affinity matrix I (--a--) and acetylated affinity matrix I (-•-) were incubated with 10 vol (A), 5 vol (B) or 2.5~01 (C) of cytosol from adrenalectomized rats for increasing times. The percentage of receptor bound to the gels was determined as described for Table 1. Affinity matrix I contained 2 pmol steroid/ml of gel.
JANET P. HAPGCKJD and CLAUS VON HOLT
774
(A) 205 $16 97
I 1
I 2
I 3
I 1
I
2
I 3
I 4
I 5
I 6
II 7
5
I 6
II 7
6
I 9
II 10
11
I 12
I 13
I 14
205 116 97
66
45 0 2
29
II 4
6
I 9
I 10
II 11
12
I 13
I 14
Fig. 5. Identification of receptor in eluates obtained by different methods of affinity purification of receptor: 5 vol of phosphocellulose-treated cytosol in buffer I plus molybdate were incubated with 1 vol of acetylated affinity matrix I or matrix I for 16 h at 4°C. The gels were then washed free of unbound protein. 1 ml aliquots of the gels were then packed into silylated columns (2 x 0.5 cm) and washed with an additional 4ml of buffer 2 plus molybdate or buffer 5 plus molybdate as the case may be. This was followed by incubation with 1ml of eluant for various times with occasional mixing. The elution conditions for incubations IL14 are summarized in Table 2. The effects of elution in the presence of TA (2 ,uM), NaSCN (50 mM), DMSO (10%) and Mersalyl (I 00 PM) were examined. In some cases the affinity matrix was incubated with cytosol in the presence of 4pM unlabelled dexamethasone. Cytosol was prepared either from the livers of adrenalectomized rats (adren) or non-adrenalectomized rats (non-adren) as shown in Table 2. The eluates from incubations l-14 were collected and pooled with an additional 1ml column wash. Analysis of the eluates was by Coomassie stained SDS-PAGE (IA) or anti-receptor antibody detection after SDS-PAGE and Western blotting (IB).
Glucocorticoid receptor affinity purification A)
1D) 116 97 6E
I 17
I
I ?S
I
I 21
I
I 23
I
I 19 Fraction
I
I 21
I
I 23
I
I 25
number
Fig. 6. Identification of purified GHRC: Experimental and control mock GHRC preparations were eluted with a 40 ml St%500 mM potassium phosphate gradient from DEAE-Sephacel in buffer 3 plus molybdate. Elution profile of [)H]TA for 35 ~1 of individual eluate fractions (1 ml) for experimental (a) and control (A) isolations (panel A). Coomassie (panel B) and silver (panels C, D) staining patterns after SDS-PAGE of individual fractions for experimental (panels B, C) and control (panel D) isolations. The arrow denotes the bands which react with anti-glucocorticoid receptor polyclonal antibodies after Western blotting, as shown previously [16].
Elution of receptor from affinity matrix Glucocorticoid receptor has been eluted from steroid-affinity matrices with varying degrees of success by competition with a high affinity steroid (l-4 PM, 4-16 h) [6,7, 14, 15,301, sometimes in combination with NaSCN [l 1, 131. We have investigated the effects of the chaotropic salt, NaSCN [31]; the hydrogen bond disrupting agent, DMSO; the sulfhydry1 modifying reagent, Mersalyl [32]; and the high affinity steroid, TA, on elution of bound receptor from acetylated affinity matrix I (Table 2, Fig. 5A, B). The integrity of the steroid-binding activity of the receptor eluted under different conditions is shown in
Table 3. NaSCN elution (lanes 9, 10) did not improve the yield of receptor above that eluted with [3H]TA alone (lane 4). It did however increase the amount of non-specifically eluted proteins and the receptor exhibited a decreased steroid-binding activity (Table 3).
Restoration of steroid-binding could not be achieved after removal of NaSCN by dialysis or gel filtration and recharging with [‘HITA (data not shown). Elution in the presence of 10% DMSO (lanes 5,6) increased the non-specific elution of contaminating proteins and did not improve the efficiency of elution of receptor in the presence of [‘HITA (lane 4). It did however decrease the yield of steroid-bound eluted receptor (Table 3). 1OOpM Mersalyl for 1 h was needed for quantitative elution of the receptor (lane 14). Only 10% was eluted on incubation with 10 pM Mersalyl for 1 h (data not shown). Attempts to restore steroid-binding activity by reduction of receptor eluted with 100 PM Mersalyl were unsuccessful (Table 3). Thus it appears that the minimum concentration of Mersalyl required to elute the receptor from the affinity matrix results in irreversible modification of the receptor. The degree of irreversibility for
176
JANETP. HAPCKX~D and CLAUSVONHOLT Table 4. Three step purification
Step Cytosol Phosphocellulose Affinity eluate DEAE eluate
Volume (ml)
Total protein (mg)
150 200 80 6
3333 2518 0.648 0.172
flow through
of untransformed Total receptor (%I 150 136 81 69
GHRC from
Total receptor (pmot) 1630 1478 880 750
‘The purification has been executed subsequently several times by one of the authors and independently twice with overall purification of 8800 and 8900.
a particular preparation at a given concentration of Mersalyl will depend on the concentration of contaminating thiol proteins. NaSCN, DMSO and Mersalyl all lead to irreversible partial or complete destruction of the steroidbinding activity of the unoccupied glucocorticoid receptor. However incubation of affinity matrix with 2pM [3H]TA (a synthetic analogue of the natural ligand) for 16 h at 4°C in the presence of molybdate results in efficient biospecific elution of the GHRC (lane 7). This method of elution was thus routinely used in subsequent purification procedures. Similar amounts of receptor are isolated from the cytosol of non-adrenalectomized or adrenalectomized rats (Fig. 5, lanes 13, 14). Although the cytosol from non-adrenalectomized rats was found to contain 25% less receptor than that from adrenalectomized rats, the percentage of specific binding of receptor to the affinity matrix did not differ after 16 h incubation. We ascribe this to insignificant competion for receptor binding of endogenous steroid in cytosol with the high concentration of high affinity ligand on the affinity matrix (2 mM). DEAE-Sephacel
chromatography
GHRC eluted from the acetylated affinity matrix I under optimized conditions elutes as a symmetrical peak from DEAE-Sephacel at 0.24-0.26 M potassium phosphate in buffer 3 in the absence or presence of 10 mM molybdate, as also found by others [15]. 13H]TA activity and 9&92-kDa protein are coeluted (Fig. 6). The yield of steroid-bound receptor was 60% in the absence and 85% in the presence of molybdate, reflecting the stabilization of steroid-binding by molybdate [33]. GHRC eluted in buffer 2 with or without molybdate as a sharp peak at 0.140.16 M NaCl. Occasionally a small peak (510%) of radioactivity was eluted at (4&50mM NaCI) together with a 40-kDa polypeptide (data not shown). The presence of contaminating fractions in the GHRC preparation was analysed by SDS-PAGE and silver staining for experimental and control mock isolations (Fig. 6). Only the 90-92 kDa and a 20-22 kDa fraction became significantly reduced in mock isolations, showing that the other detectable bands are contaminants. The 20-22-kDa band behaved similarly to the 24-kDa band observed by others [15], eluting at a higher ionic strength than GHRC from DEAE-Sephacel. There is controversy
I5 rat livers
Specific activity (pm01 receptor per mg total protein)
% Yield of receptor
0.489 0.587 1.36 x IO’ 4.36 x IO’ with overall purification
100 90 54 46
Fold purification
I 1.2 2777 8916’
factors between 85Ot&9000
as to the nature of this 2t324-kDa fraction and whether it is associated with the GHRC. The presence of RNA in that fraction has been suggested by some [34,35], while others have been unable to detect this fraction by silver staining [36]. We could not detect a 72-kDa fraction specifically associated with GHRC [12], neither did we observe the 4&41-kDa fraction found by Grandics et al. [I 51. DISCUSSION Based on the preceding investigations, we routinely isolate the untransformed GHRC as described in Experimental to obtain a 9,000-fold purified GHRC preparation with 46% yield (see Table 4). Idziorek et al. [7] have recently published a very similar isolation procedure using high performance size exclusion chromatography as a final step to obtain >45% pure GHRC. Presuming one steroid-binding subunit per 92-kDa polypeptide, the GHRC eluted off DEAE-Sephacel in our experiments in the presence of molybdate is 3040% pure. However the Coomassie dye staining of the gel suggests an almost homogeneous receptor preparation (Fig. 6). This discrepancy was observed by others [7] and is most likely explained by the presence of a 90-92-kDa non-steroid-binding protein contaminant [ 161, which is also evident from Fig. 6C, D. We have found that the amount of 9&92-kDa contaminant varies for different isolations. The sedimentation and DNA-binding properties of the highly purified transformed and untransformed GHRC will be reported elsewhere [37]. The affinity purification of molybdate-stabilized untransformed GHRC followed by heat transformation of the DEAE-Sephacel eluate in the absence of molybdate and in the presence of BSA was found to reproducibly result in highly purified, undegraded, transformed GHRC with a high yield. Acknowledgements-We thank Dr M. V. Govindan (Laboratoire de Genttique Moieculair des Eucaryotes, Strasbourg, France) for his kind gift of polyclonal anti-glucocorticoid receptor antibodies. This work was supported by a grant from the Council for Scientific and Industrial Research (S.A.) and the University of Cape Town Research Committee to C.V.H. REFERENCES I. Hollenberg
S. M., Weinberger Oro A., Lebo R., Thompson
C., Ong E. S., Cerelli G., E. B., Rosenfield M. G.
Glucocorticoid
receptor affinity purification
and Evans R. M.: Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318 (1985) 6355641. 2. Wrange t). and Gustafsson J. A.: Separation of the hormone- and DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. J. biol.
19.
20.
Chem. 253 (1978) 856865.
3. Carlstedt-Duke J., Wrange t), Dahlberg E., Gustafsson J. A. and Hogberg B.: Transformation of the glucocorticoid receptor in rat liver cytosol by lyosomal enzymes. J. biol. Chem. 254 (1979) 1537-1539. 4. Govindan M. V. and Sekeris C. E.: Purification of two dexamethasone-binding proteins from rat-liver cytosol. Eur. J. Biochem. 89 (1978) 955104. 5. Manz B., Heubner A., Kijhler I., Grill H. J. and Pollow K.: Synthesis of biotin-labelled dexamethasone derivatives. Novel hormone affinity probes. Eur. J. Biochem. 131 (1983) 333-338.
6. Weisz A., Baxter J. D. and Lan N. C.: Interaction of sodium molybdate with highly purified glucocorticoid receptor. J. steroid Biochem. 20 (1984) 289-293. 7. Idziorek T., Formstecher P., Danze P. M., Sablonniere B., Lustenberger P., Richard C., Dumur V. and Dautrevaux M.: Characterization of the purified molybdatestabilized glucocorticoid receptor from rat liver. An in vitro transformable complex. Eur. J. Biochem. 153 (1985) 65-74.
8. Wrange G., Carlstedt-Duke J. and Gustafsson J. A.: Purification of the glucocorticoid receptor from rat liver cytosol. J. biol. Chem. 254 (1979) 92849290. 9. Govindan M. V. and Manz B.: Three-step purification of glucocorticoid receptor from rat liver. Eur. J. Biothem. 108 (1980) 47-53. 10. Westphal H. M. and Beato M.: The activated glucocorticoid receptor of rat liver. Eur. J. Biochem. 106 (1980) 395403. 11. Govindan M. V. and Gronemeyer H.: Characterization of the rat liver glucocorticoid receptor purified by DNA-cellulose and ligand affinity chromatography. J. biol. Chem. 259 (1984) 12915-12924. 12. Wrange d., Okret S., Radojcic M., Carlstedt-Duke J. and Gustafsson J. A.: Characterization of the purified activated glucocorticoid receptor from rat liver cytosol. J. biol. Chem. 259 (1984) 45344541. 13. Kurl R. N. and Jacob S. T.: Phosphorylation of purified glucocorticoid receptor from rat liver by an endogeneous protein kinase. Biochem. biophys. Res. Commun. 119 (1984) 700-705.
14. Lustenberger P., Formstecher P. and Dautrevaux M.: Purification of rat liver glucocorticoid receptor by affinity chromatography. Design of a suitable adsorbent. J. steroid Biochem.
14 (198i) 697-703.
15. Grandics P.. Miller A.. Schmidt T. J.. Mittman D. and Litwack G.:’ Purification of the unactivated glucocorticoid receptor and its subsequent in uitro activation. J. biol. Chem. 259 (1984) 3173-3180. 16. Hapgood J. P., Sabbatini G. P. and von Holt C.: Rat liver glucocorticoid receptor isolated by affinity chromatography is not a Mg*+- or Ca2+-dependent protein kinase. Biochemistry 25 (1986) 7529-7534. 17. Lan N. C., Karin M., Nguyen T., Weisz A., Birnbaum M. J., Eberhardt N. L. and Baxter J. D.: Mechanisms of glucocorticoid hormone action. J. steroid Biochem. 20 (1984) 77-88.
18. Webb M. L., Miller-Diener A. S. and Litwack G.: Purification, characterization and activation of the
21.
22. 23.
777
glucocorticoid-receptor complex from rat kidney cortex. Biochemistry 24 (1985) 1946-1952. Sica V., Parikh I., Nola E., Puca G. A. and Cuatrecasas P.: Affinity chromatography and the purification of estrogen receptors. J. biol. Chem. 248 (1973) 6543-6558. Cuatrecasas P.: Protein purification by affinity chromatography. J. biol. Chem. 245 (1970) 3059-3065. Lustenberger P., Formstecher P. and Dautrevaux M.: Quantitative determination of alkvlamino side-chains coupled to agarose beads. J. Chromat. 193 (1980) 451457. Riordan J. F. and Vallee B. L.: Acetylation. Merh. Enzym. 11 (1967) 565-570. Rousseau G. G., Baxter J. D. and Tomkins G. M.: Glucocorticoid receptors: relations between steroid binding and biological effects. J. molec. Biol. 67 (1972)
99-105. 24. Ansorg W.: In Proceedings
Electrophoresis (Edited by D. Stathakos). Walter de Gruyter, Berlin (1982) pp. 235-245. 25. Bensadoun A. and Weinstein D.: Assay of proteins in the presence of interfering materials. Analyt. Biochem. 70 (1976) 241-250. 26. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall
R.: Protein measurement with Folin nhenol reaeent. J. biol. Chem. 193 (1951) 265-275.
27. Govindan M. V. and Sekeris C. E.: Partial purification of rat liver glucocorticoid binding proteins by affinity chromatography. Steroids 28 (1976) 499-507. 28. Govindan M. V. and Sekeris C. E.: Purification of two dexamethasone-binding proteins from rat liver cytosol. Eur. J. Biochem. 89 (1978) 95-104. 29. Tsawdaroglou N. G., Govindan M. V., Schmidt W. and
Sekeris C. E.: Dexamethasone binding proteins in cytosol and nucleus of rat thvmocvtes. Eur. J. Biochem. 114 (1981) 305-313. _ _ 30. Failla D.. Tomkins G. M. and Santi D. V.: Partial purification of a glucocorticoid receptor. Proc. natn. Acad. Sci. U.S.A. 72 (1975) 3849-3852. 31. Kalimi M. and Hubbard J.: Interaction
of sodium thiocyanate with rat hepatic glucocorticoid-receptor complexes. Biochim. biophys. Acta 719 (1982) 488-494. 32. Coty W. A.: Reversible dissociation of steroid hormone receptor complexes by mercurial reagents. J. biol. Chem. 255 (1980) 8035-8037.
33. Leach K. L., Dahmer M. K., Hammond N. D., Sando J. J. and Pratt W. B.: Molybdate inhibition of glucocorticoid receptor inactivation and transformation. J. biol. Chem. 254 (1979) 1188411890. 34. Grandics P., Miller A., Schmidt T. J. and Litwack G.: Phosphorylation in uiuo of rat hepatic glucocorticoid receptor. Biochem. biophys. Res. Commun. 120 (1984) 59-65.
35. Housley P. R. and Pratt W. B.: Small 32P-labeled components of L-cells that coelute with the molybdatestabilized glucocorticoid receptor during purification. Fedn. Proc. Fedn. Am. sot. Exp. Biol. 43 (1984) 1572.
36. KovaEiE-Milivojevic B. and Vedeckis W. V.: Absence of detectable ribonucleic acid in the purified untransformed mouse glucocorticoid receptor. Biochemistry 25 (1986) 82668273.
37. Hapgood J. P. and von Holt C.: Transformation in vitro and covalent modification with biotin of steroidaffinity-purified rat-liver glucocorticoid-hormonereceptor complex. Eur. J. Biochem. 166 (1987) 415420.
6, pp. 76~717, 1987 Printed in Great Britain. All rights reserved
0022-4731/87$3.00+ 0.00 Copyright 0 1987Pergamon Journals Ltd
STEROID-AFFINITY PURIFICATION OF THE RAT LIVER GLUCOCORTICOID HORMONE RECEPTOR COMPLEX JANET P. HAPGOOD and CLAUS VON HOLT* The UCT-CSIR Research Centre for Molecular Biology, Department of Biochemistry, University of Cape Town, Republic of South Africa (Received 2 March 1987)
Summary-The molybdate-stabilized GHRC was isolated from rat liver cytosol with a 9000-fold purification and 46% yield. The major purification step was achieved using an affinity matrix consisting of an agarose support coupled to a dexamethasone ligand via an aliphatic spacer arm. Spacer arms containing disulfide bridges were found to be unsuitable due to their instability in cytosol. To reduce the non-specific binding properties of the affinity matrix, underivatized amino groups were acetylated, since the receptor was found to bind avidly to such groups thus evading elution by the ligand. Sodium molybdate present during biospecific elution from the gel stabilized the steroid-binding activity of the receptor. The use of denaturing and sulfhydryl modifying reagents (NaSCN, DMSO, Mersalyl) during elution led to partial or complete irreversible loss of steroid-binding activity of the unoccupied receptor. Efficient biospecific elution occurred at competing concentration of high affinity steroid in the presence of sodium molybdate. The ligand specific eluate was further purified by DEAE-Sephacel chromatography resulting in additional purification of 3.2-fold. The GHRC eluted from the DEAE-Sephacel column at a salt concentration characteristic of the untransformed GHRC. Molybdate was removed from the purified untransformed GHRC in the ligand eluate by DEAE-Sephacel chromatography in the absence of molybdate, for subsequent heat transformation.
INTRODUCTION
The cDNA of a 90-kDa glucocorticoid-binding protein has been identified and sequenced [l]. However, the isolation and characterization of the GHRC itself from cells is still controversial. It is necessary to purify the GHRC about 104-fold from cytosol to obtain a homogeneous preparation suitable for biochemical investigation. With increasing purification GHRC is known to become highly unstable due to its susceptibility to proteolysis [2, 31 and its tendency to lose steroid-binding activity [4-71. Despite this lability, several authors have reported the extensive purification of a steroid-binding receptor protein of approx 90 kDa. Such purifications have usually been accomplished using successive ion-exchange chromatography and DNA-cellulose chromatography, sometimes in combination with steroid-affinity *Correspondence: UCT-CSIR Research Centre for Molecular Biology, Department of Biochemistry, University of Cape Town, Private Bag, Rondebosch 7700, Republic of South Africa. Abbreviufions: Dexamethasone, 9a-fluoro- 16a-methyl-l l~, 17a,21-trihydroxy-pregna- 1,4-diene-3,20-dione; TA, triamcinolone acetonide, 9cc-fluoro-16a-methyl-ll/?,17a, 21-trihydroxypregna-1,4-diene-3,20-dione; GHRC, glucocorticoid hormone receptor complex; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrouhoresis: dam, disintegrations per min; DMF, dimethyl‘formamide; BSA, bovine ser;m albumin; TLC, thinlayer chromatography; NaSCN, sodium thiocyanate; DMSO, dimethylsulfoxide; Mersalyl, [3-[[2-(carboxymethoxy)-benzoyl]amino]-2-methoxypropyl]hydroxymercury monosodium salt; DTT, dithiothreitol.
[g-13]. The stabilizing effect of molybdate has been utilized to enrich the untransformed GHRC by steroid-affinity chromatography [6,7, 14, 151. 10% [6] to ~45% [7] pure molybdate-stabilised untransformed GHRC has been obtained with yields of 18% [14] to 46% [6]. Dexamethasone-containing matrices have resulted in purifications ranging from 900 to 8800-fold [7,9, 11, 13, 141. The great variation in the efficiency of similar matrices reflects the complexity of the problems inherent in the procedure. We report on the optimisation of conditions for the steroidaffinity purification of the transformed and untransformed GHRC. Mock steroid-affinity isolations can be performed by incubation of cytosol with affinity matrix in the presence of excess free dexamethasone to prevent receptor binding to the gel [15-181, followed by elution from the affinity matrix and DEAE-Sephacel chromatography as for experimental isolations. Such mock isolations aid in the identification of contaminants co-purifying with the receptor. We have previously shown that a 90-92 kDa non-steroidbinding protein as well as a Mg2+ dependent protein kinase copurify with the GHRC as contaminants [16], and have identified now other contaminants.
matrices
EXPERIMENTAL Materials
All glassware was acid cleaned and silylated. [ 1,2,4(n)-3 H]Triamcinolone acetonide (25 Ci/mmol) was from Amersham. All reagents used were ana769
110
JANETP. HAPGCODand CLAUS VON HOLT
lytical grade. Serum containing polyclonal antiglucocorticoid receptor antibodies raised in rabbits was a kind gift from Dr M. V. Govindan [ll]. The antigen was the 92-kDa transformed GHRC purified from rat liver cytosol by DNA-cellulose chromatography followed by preparative gel electrophoresis [ll]. The antibody is highly specific for a 92 kDa protein and does not cross-react with other cytoplasmic proteins (Fig. 5). Affi-Gel 102, an amino substituted agarose matrix, was from Bio-Rad. Phosphocellulose Pll was from Whatman and DEAESephacel from Pharmacia. Synthesis of dexamethasone derivatives
Synthesis of the 17P-carboxylic acid of dexamethasone by periodic acid oxidation was as described by Govindan and Manz [9]. The product migrated as a single spot on TLC (solvent system (1): R, = 0.06, solvent system (2): R, = 0.59). The N-hydroxybenzotriazole (HOBT) ester of the acid was prepared according to [7]. A single steroid ester spot was obtained on TLC system (1) Rr = 0.85 and system (2) Rf = 0.79. TLC was performed on silica gel plates with fluorescent indicator. Solvent system (1) = chloroform-methanol (9: l), v/v) and solvent system (2) = ethylacetate-ethanol-NH, (25% w/v H,O); (S:S:l, by vol). Preparation of affinity matrices Synthesis of Aji-Gel 102-Dexamethasone (Ajkity matrix 1) [l l] Affi-Gel 102 (40 ml, 14 pmol amino
groups/ml) was equilibrated in DMF and then suspended in 40ml of DMF containing 0.5 mmol of radioactive steroid ester. The specific activity of the ester was chosen such as to allow the monitoring of the washing procedure and the yield of bound steroid. The suspension was gently rolled at 20°C for 48 h. The affinity matrix was washed with 400 ml DMF, 5 1 absolute ethanol and 3 1 of water on a sintered glass funnel, after which it was stored at 4°C in the presence of 0.02% sodium azide. The presence of free steroid in the affinity matrix was monitored during the prolonged washing by the method of Sica et al. 1191 until the concentration of free steroid was 10.5 pmol/ml of affinity matrix. The concentration of bound ligand was determined by oxidation of a lyophilised measured volume of affinity matrix and counting the [3H]H,0 liberated. The affinity matrix prepared as above contained 2pmol steroid/ml packed gel. By decreasing the amount of ester during synthesis, matrices containing 0.1-2.0 pmol steroid/ ml gel were synthesized. Synthesis qf CL-Sepharose 4B-aminohexylsuccinylamidocystamine-dexamethasone (Aflnity matrix II)
CL-sepharose 4B (SOml) was activated with 15 g of CNBr and immediately derivatised with hexamethylenediamine (100 mmol) as described in [9,20]. The extent of derivatization was monitored by measuring the concentration of amino groups introduced
[21]. Aminohexyl-CL-Sepharose 48 (5 pmol aminohexyl groups/ml) was then succinylated with succinic anhydride [20] and coupled to cystaminium dichloride at 20°C as follows: The succinylated matrix (0.1 pmol amino groups/ml) was suspended in 25 ml of 40% DMF. Cystaminium hydrochloride (7 mmol) in 25 ml of 40% DMF was added and the pH brought to 4.5 with 1 M HCl. 1-ethyl-3-(3-diaminopropyl) carbodiimide (25 mmol) dissolved in 5 ml of water was then added over 10 min while shaking and maintaining the pH at 4.5. Thereafter, the suspension was shaken overnight, washed with 40% DMF and water and equilibrated in DMF. Amidocystamidosuccinamidohexyl-Cl-Sepharose 4B (5 pmol amino groups/ml) was coupled to the ester of dexamethasone-17/l-carboxylic acid as described for the synthesis of affinity matrix I. Affinity matrices containing 0.25 and 2.0 pmol steroid/ml of gel were synthesized. Synthesis of A#i-Gel 102-succinyl-amidocystamine dexamethasone (AfJinity matrix III)
Affi-Gel 102 was succinylated with 99% efficiency, coupled to cystaminium dichloride (to give 13.8 pmol amino groups/ml) and then reacted with steroid ester as described for affinity matrix II. Acetylation of gels
Acetylation was performed with distilled acetic anhydride in the presence of 50% saturated sodium acetate at 4C [22]. The gel was washed and suspended in an equal volume of this solution. 6 x 20 ~1 aliquots of acetic anhydride/ml of gel were added over a period of 1 h with gentle shaking. Thereafter the substituted gel was left to shake for 1 h and finally washed with sodium acetate and water. This procedure prevents the acetylation of hydroxyl groups and is selective for amino groups [22]. Regeneration of aflnity matrix 40 ml of used gel was washed with 11 of 1% SDS (6O”Q 2 1 of water (6O”Q then 1 1 of 1 M NaCl. This was followed by a wash with 11 of 50% ethanol and 3 1 of absolute ethanol to remove free steroid. If the concentration of free steroid was >O.S pmol/ml, the washing was continued until this level was attained. Finally the gel was equilibrated in buffer 1 by washing with 1 1 of water and 1 I of buffer 1. Washing was performed successively on a sintered glass funnel over a period of 3 h. Buffers
Buffer 1 was sodium phosphate (20mM), EDTA (1 mM), fl-mercaptoethanol (5 mM), sodium chloride (SOmM) and glycerol (lo%), pH 7.0. Buffer 2 was Tris (10 mM), EDTA (1 mM), fl-mercaptoethanol (5 mM) and glycerol (lo%), pH 7.4 (4°C). Buffer 3 was potassium phosphate (SOmM), EDTA (1 mM), /I-mercaptoethanol (5 mM) and glycerol (lo%), pH 7.0. Buffer 4 was Tris (10 mM), EDTA (1 mM), fl-mercaptoethanol (5 mM), glycerol (lo%), NaCl
111
Glucocorticoid receptor affinity purification
(50 mM) and BSA (0.2%), pH 7.5 (4°C). Buffer 5 was thiol-free buffer 1. For all buffers, plus molybdate refers to the addition of 10mM sodium molybdate, unless otherwise stated. Optimized isolation procedure
All procedures were performed at 4°C. Buffers contained 10mM molybdate unless otherwise indicated. Cytosol (200ml) was prepared from livers of 15 non-adrenalectomized male Wistar rats in buffer 1 [16]. The cytosol was then immediately passed at a flow rate of 20 ml/min over phosphocellulose (150 ml) packed in a silylated glass column (15 x 7 cm). Additional sodium molybdate (10 mM) was added to the clear yellow flow-through. The flow-through (200 ml) was incubated with acetylated affinity matrix I (40 ml) in a glass reagent bottle. The suspension was gently rolled on a Coulter mixer for 16 h. To determine the amount of receptor bound to the matrix, an aliquot of cytosol before and after incubation with affinity matrix was assayed for receptor content. The gel was then transferred to a sintered glass funnel and washed under low suction with 15 x 100 ml of buffer 1, 2 x 100 ml of buffer 1 plus 0.2 M NaCl then with 15 x 100 ml of buffer 2 or buffer 3. The washing of the affinity matrix was performed over a period of 90 min. The final wash did not absorb at 280 nm. The affinity matrix was then transferred to a 100 ml glass bottle and an equal volume of eluting buffer, 2 PM t3H]TA (5 pCi/nmol) in buffer 2 or 3, was added. The mixture was then gently rolled for 16 h and the eluate collected after centrifugation of the affinity matrix at 1,OOOgin a glass tube and a wash with one affinity matrix volume of elution buffer. The pooled eluates were then applied to a DEAE-Sephacel column (2 ml bed volume). Washing and elution were performed as described below. The GHRC eluting from the ionexchange column was stored in 50% glycerol in the presence of 1 PM [3H]TA at -20°C. DEAE-Sephacel
ion-exchange chromatography
The untransformed, molybdate-stabilized GHRC in the affinity matrix eluate was subjected to DEAESephacel chromatography in the absence or presence of sodium molybdate in buffer 3. DEAE-Sephacel was equilibrated by suspension and decanting twice with 20 vol of 0.5 M potassium phosphate pH 7.0. The ion-exchange gel was then packed into a glass column (0.5 x 5.0 cm) with a silylated glass wool plug and washed with 20 column volumes of buffer 3. 2 ml of packed DEAE-Sephacel was used for 40ml of affinity eluate at a flow rate of 0.3 ml/min. Affinity eluate in molybdate-containing buffer was applied to the column. Once all the affinity eluate had passed through the column, the column was washed with 10 ml of buffer in the absence or presence of molyb*The term matrix refers to the polymer substituted with linker arm and affinity matrix refers to the steroidsubstituted matrix. The term gel is used generally and refers to matrix or affinity matrix.
date as indicated in figures and tables. The GHRC was eluted with a 40 ml linear gradient of 50-500 mM potassium phosphate in buffer 3 with or without molybdate after washing with 10ml of this buffer. Occasionally buffer 2 instead of buffer 3 was used. In that case elution was performed with 40 ml of a linear O-500 mM NaCl gradient in buffer 2 with or without molybdate. Receptor content was determined by incubation of cytosol with 25 nM [3H]TA in the absence and presence of 5 PM TA for 16 h at 4°C followed by dextran/charcoal treatment [23] and liquid scintillation counting. SDS-PAGE, Western blotting and immunoassay were performed as described in [16]. In all cases,
control incubations with preimmune serum gave negative results. Siher staining was by the method of Ansorg [24]. Protein determination was by a modified Lowry method, with BSA as standard [25,26]. The sensitivity was 1 pg of protein and samples were analysed in triplicate. Radioactiuity determination
Determination of tritium activity in matrix bound steroid ligand was done after oxidation to [3H]Hr0 in a sample oxidizer (Packard Instruments) via liquid scintillation counting. Samples containing soluble radioactive compounds were dissolved in a suitable scintillation cocktail and likewise counted in a scintillation spectrometer. All counts were quench corrected for dpm. RESULTS
Afinity matrices *
TWO types of affinity matrix (Fig. 1) with different linker arms were synthesized. Affinity matrices II and III contain a cleavable disulfide bond in the linker arm, while affinity matrix I contains amide bonds in the hydrocarbon linker arm. Both types of linker arm have been used previously by others [6,7,9, 11, 14, 15, 27-301. Binding of receptor to a$nity matrix
Whereas the steroid ligand coupled to affinity matrix I was stable during incubation with cytosol for 16 h, affinity matrices II and III were unsuitable due to the lability of the disulfide bond in untreated cytosol. 50% of bound steroid was cleaved off affinity matrices II cr III (2 pmol steroid/ml packed gel) after incubation of 1 vol of affinity matrix with 5 vol of cytosol in buffer 5 for 3 h. Gel filtration or ammonium sulfate precipitation of cytosol prior to incubation with affinity matrix removed the reducing capability of cytosol, whereas phosphocellulose treatment did not (data not shown). The non-specific binding of receptor to the matrices could be almost entirely eliminated by acetylation of underivatized amino groups for matrix I (Table 1). However matrices II and III still exhibited high
JANET
P. HAPGOODand CLAUSVONHOLT Table 1. Effect of acetvlation of gels on recwtor binding
OCH,CONH(CH,),NH \
I I I I II II II II III III III III CL-Seoharose 4B
Affi-Gel 102-dexamethasone AFFINITY MAlRIX 1
0, /ImccH2);2~ ) 0
A
co22 OH
NHKH,),SS(CH,),NH \
?
/’
,
c=o -_OH
95 17
'-CH,
CL-Sepharose midocystaine-
Steroid (2 pmol/ml gel) _ _ + + _ + + _ + + -
Matrix
c-o
‘YIL-J 4B-minohexyl-succinyldexamethasone
AFFININ MATRIXII 0CH$ONH(CH2)zNH
Acetylation
44 3 86 46 84 63 95 76 64 46 68 41 2
+ + + + + + -
Matrices (100 PI), derivatized as shown in the table were gently shaken for 3 h at 4°C with cytosol (500~1) from adrenalectomized male rat livers in buffer 1. In the case of gels containing cystamine derivatives, cytosol was desalted prior to incubation and fl-mercaptoethanol was omitted from the buffers. The fraction of receptor bound to the gel was calculated after determination of the receptor content in cytosol before and after incubation.
Washing of afinity matrix after receptor binding
The affinity matrix was washed at 4°C batchwise on a silylated glass funnel under low suction with a total of 80 vol of molybdate-containing buffer (see Experimental) to remove the bulk of the contaminating proteins. The matrix was resuspended gently with a spatula between washes. Small amounts of the receptor leaked off the gel under more rigorous conditions.
c:R
I co2 2 NH(C%)2SS(CH2&NH
% bound receptor
100
.
.
1.
-.
\
Affi-Gel
102-succinyl-amidocystamine-
dexamethasone AFFINITY
WURIX III
Fig. 1. Affinity matrices for glucocorticoid receptor with non-cleavable and cleavable linker arms.
non-specific binding after acetylation, probably due to functional groups introduced during activation and coupling [7] and as the result of exchange reactions with cytosolic thiol proteins. The chemical reactions used in preparation of the affinity matrices did not affect the integrity as a ligand of the dexamethasone derivative (Fig. 2). The optimal ligand concentration on acetylated affinity matrix I was 2.0 pmol steroid/ml packed gel (Fig. 3). Optimal binding of receptor was at a ratio of 1 vol of acetylated affinity matrix I to 5 vol of cytosol for 16 h at 4°C (Fig. 4). Under these conditions the non-specific receptor binding was 5% and the specific binding was 70-80%.
,\,
I 1
0
ml
0
. 3
2
Ccompetitorl r3H triamcinolone
acetonidel
Fig. 2. Competition binding of steroid derivatives to receptor: [)H]Dexamethasone 17fi-CONH(CH,),SH was cleaved off affinity matrix II with b-mercaptoethanol to obtain a solution of this derivative in buffer 1. Increasing concentrations of dexamethasone 17/3-(2-mercaptoethylamide) dexamethasone and dexamethasone (A) (O), 17fl-carboxylic acid (0) were incubated with cytosol (100~1) from adrenalectomized rats in the presence of 10 nM 13H]TA. Incubations were for 16 h at 4°C in a final volume of 250 ~1 in buffer I plus molybdate. Thereafter, the amount of [‘HJTA bound to receptors was measured by the dextranicharcoal assay. 100% specific binding was taken as the difference in binding between [‘HITA and [‘HITA plus IOpM TA.
Glucocorticoid receptor affinity purification
173
Table 2. Summary of elution conditions Experiment (corresponds to lanes l-14
Eluant (in buffer 2 or buffer 5) plus
in Fig. 5)
I
Gel acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acetylated acptylated acetylated acetylated
2 3 4 5 6 I 8 9 10 II 12 13 14
Affi-Gel 102 affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix affinity matrix
I I I I I I I I I I I I I
Time of elution h
Cytosol
molybdate
adren adren adren adren adren adren adren adren adren adren non-adren t steroid adren + steroid non-adren adren
5 + Mersalyl 5 2 2tTA 2+DMSO+TA 2 t DMSO 2+TA 2 2 + NASCN 2 t NASCN + TA 5 t Mersalyl 5 t Mersalyl 5 t Mersalyl 5 + Mersalyl
1
I 5 5 5 5 16 16 5 5
I I
1 I
Receptor was eluted from acetylated affinity matrix I under the conditions summarized in the table (details in legend to Fig. 5). The eluates were then assayed for the presence of receptor and the results are shown in Fig. 5.
75
1 Ligond
I
I
I
0.5
1.0
1.5
concentration
(pm01
I 2.0
steroid/ml
/’
, t7
25
./* /
gel)
Fig. 3. Specific binding of receptor to acetylated affinity matrix I: 100~1 of acetylated affinity matrix I containing O&2.0 pmol steroid/ml of gel was incubated with 500 ~1 of cytosol for 3 h at 4°C as previously described (Table 1). After incubation, the percentage receptor binding to the affinity matrices was determined as described for Table 1.
__..---.(A)
_-.-
,./’
.z fr E 2 z
.- ..A-A-.-._._ __>_-_-..---.. .-(6)
r’ 75-J
._--
-._
i .,A’
- - ;-2”
.y
I1
Table 3. Steroid-binding activity of eluted receptor Steroid-binding activity (% of input)
Elution conditions
60 18
[‘HITA, buffer 2 plus molybdate, 5 h [‘HITA, buffer 2 minus molybdate, 5 h [‘HITA, buffer 2 plus molybdate, plus NASCN, 5 h [‘HITA, buffer 2 plus molybdate, plus DMSO, 5 h Mersalyl, buffer 5 plus, molybdate
35 41
Ih
(I)’ (2)’
1;
Phosphocellulose-treated cytosol was incubated with acetylated affinity matrix I followed by washing of the gel as described in Fig. 5. The gels were washed with buffer 2 or buffer 5 in the absence or presence of molybdate as indicated in Table 2. The concentrations of eluants were as indicated in Fig. 5. The eluates were harvested and assayed for [‘HITA-bound receptor by gel filtration on Sephadex G-75 in buffer 4 plus molybdate. The steroid-binding activity was calculated as a percentage of the maximum possible eluted steroid-binding activity for a known amount of receptor bound to the gel. In the case of Mersalyl elution, the eluates were reduced and Mersalyl was removed followed by incubation with [‘HITA by methods I or 2* before
determination of steroid-bound receptor. Method I involved gel filtration in buffer 4 containing DTT (25 mM) and molybdate (IOmM), followed by incubation of the void volume with [‘H]TA (I pM) for I6 h at 4°C. Method 2 involved exhaustive dialysis against buffer 2 containing DTT (25 mM) and molybdate (1OmM) followed by incubation with [‘HITA as for
method I. sa
28,6-M
[*.._*_;-;-;-;4
8
12
16
Time (hr) Fig. 4. Binding of receptor to matrix I and affinity matrix I; effects of acetylation, time of incubation and ratio of cytosol to gel: Matrix I (--A--), acetylated matrix I (--A-), affinity matrix I (--a--) and acetylated affinity matrix I (-•-) were incubated with 10 vol (A), 5 vol (B) or 2.5~01 (C) of cytosol from adrenalectomized rats for increasing times. The percentage of receptor bound to the gels was determined as described for Table 1. Affinity matrix I contained 2 pmol steroid/ml of gel.
JANET P. HAPGCKJD and CLAUS VON HOLT
774
(A) 205 $16 97
I 1
I 2
I 3
I 1
I
2
I 3
I 4
I 5
I 6
II 7
5
I 6
II 7
6
I 9
II 10
11
I 12
I 13
I 14
205 116 97
66
45 0 2
29
II 4
6
I 9
I 10
II 11
12
I 13
I 14
Fig. 5. Identification of receptor in eluates obtained by different methods of affinity purification of receptor: 5 vol of phosphocellulose-treated cytosol in buffer I plus molybdate were incubated with 1 vol of acetylated affinity matrix I or matrix I for 16 h at 4°C. The gels were then washed free of unbound protein. 1 ml aliquots of the gels were then packed into silylated columns (2 x 0.5 cm) and washed with an additional 4ml of buffer 2 plus molybdate or buffer 5 plus molybdate as the case may be. This was followed by incubation with 1ml of eluant for various times with occasional mixing. The elution conditions for incubations IL14 are summarized in Table 2. The effects of elution in the presence of TA (2 ,uM), NaSCN (50 mM), DMSO (10%) and Mersalyl (I 00 PM) were examined. In some cases the affinity matrix was incubated with cytosol in the presence of 4pM unlabelled dexamethasone. Cytosol was prepared either from the livers of adrenalectomized rats (adren) or non-adrenalectomized rats (non-adren) as shown in Table 2. The eluates from incubations l-14 were collected and pooled with an additional 1ml column wash. Analysis of the eluates was by Coomassie stained SDS-PAGE (IA) or anti-receptor antibody detection after SDS-PAGE and Western blotting (IB).
Glucocorticoid receptor affinity purification A)
1D) 116 97 6E
I 17
I
I ?S
I
I 21
I
I 23
I
I 19 Fraction
I
I 21
I
I 23
I
I 25
number
Fig. 6. Identification of purified GHRC: Experimental and control mock GHRC preparations were eluted with a 40 ml St%500 mM potassium phosphate gradient from DEAE-Sephacel in buffer 3 plus molybdate. Elution profile of [)H]TA for 35 ~1 of individual eluate fractions (1 ml) for experimental (a) and control (A) isolations (panel A). Coomassie (panel B) and silver (panels C, D) staining patterns after SDS-PAGE of individual fractions for experimental (panels B, C) and control (panel D) isolations. The arrow denotes the bands which react with anti-glucocorticoid receptor polyclonal antibodies after Western blotting, as shown previously [16].
Elution of receptor from affinity matrix Glucocorticoid receptor has been eluted from steroid-affinity matrices with varying degrees of success by competition with a high affinity steroid (l-4 PM, 4-16 h) [6,7, 14, 15,301, sometimes in combination with NaSCN [l 1, 131. We have investigated the effects of the chaotropic salt, NaSCN [31]; the hydrogen bond disrupting agent, DMSO; the sulfhydry1 modifying reagent, Mersalyl [32]; and the high affinity steroid, TA, on elution of bound receptor from acetylated affinity matrix I (Table 2, Fig. 5A, B). The integrity of the steroid-binding activity of the receptor eluted under different conditions is shown in
Table 3. NaSCN elution (lanes 9, 10) did not improve the yield of receptor above that eluted with [3H]TA alone (lane 4). It did however increase the amount of non-specifically eluted proteins and the receptor exhibited a decreased steroid-binding activity (Table 3).
Restoration of steroid-binding could not be achieved after removal of NaSCN by dialysis or gel filtration and recharging with [‘HITA (data not shown). Elution in the presence of 10% DMSO (lanes 5,6) increased the non-specific elution of contaminating proteins and did not improve the efficiency of elution of receptor in the presence of [‘HITA (lane 4). It did however decrease the yield of steroid-bound eluted receptor (Table 3). 1OOpM Mersalyl for 1 h was needed for quantitative elution of the receptor (lane 14). Only 10% was eluted on incubation with 10 pM Mersalyl for 1 h (data not shown). Attempts to restore steroid-binding activity by reduction of receptor eluted with 100 PM Mersalyl were unsuccessful (Table 3). Thus it appears that the minimum concentration of Mersalyl required to elute the receptor from the affinity matrix results in irreversible modification of the receptor. The degree of irreversibility for
176
JANETP. HAPCKX~D and CLAUSVONHOLT Table 4. Three step purification
Step Cytosol Phosphocellulose Affinity eluate DEAE eluate
Volume (ml)
Total protein (mg)
150 200 80 6
3333 2518 0.648 0.172
flow through
of untransformed Total receptor (%I 150 136 81 69
GHRC from
Total receptor (pmot) 1630 1478 880 750
‘The purification has been executed subsequently several times by one of the authors and independently twice with overall purification of 8800 and 8900.
a particular preparation at a given concentration of Mersalyl will depend on the concentration of contaminating thiol proteins. NaSCN, DMSO and Mersalyl all lead to irreversible partial or complete destruction of the steroidbinding activity of the unoccupied glucocorticoid receptor. However incubation of affinity matrix with 2pM [3H]TA (a synthetic analogue of the natural ligand) for 16 h at 4°C in the presence of molybdate results in efficient biospecific elution of the GHRC (lane 7). This method of elution was thus routinely used in subsequent purification procedures. Similar amounts of receptor are isolated from the cytosol of non-adrenalectomized or adrenalectomized rats (Fig. 5, lanes 13, 14). Although the cytosol from non-adrenalectomized rats was found to contain 25% less receptor than that from adrenalectomized rats, the percentage of specific binding of receptor to the affinity matrix did not differ after 16 h incubation. We ascribe this to insignificant competion for receptor binding of endogenous steroid in cytosol with the high concentration of high affinity ligand on the affinity matrix (2 mM). DEAE-Sephacel
chromatography
GHRC eluted from the acetylated affinity matrix I under optimized conditions elutes as a symmetrical peak from DEAE-Sephacel at 0.24-0.26 M potassium phosphate in buffer 3 in the absence or presence of 10 mM molybdate, as also found by others [15]. 13H]TA activity and 9&92-kDa protein are coeluted (Fig. 6). The yield of steroid-bound receptor was 60% in the absence and 85% in the presence of molybdate, reflecting the stabilization of steroid-binding by molybdate [33]. GHRC eluted in buffer 2 with or without molybdate as a sharp peak at 0.140.16 M NaCl. Occasionally a small peak (510%) of radioactivity was eluted at (4&50mM NaCI) together with a 40-kDa polypeptide (data not shown). The presence of contaminating fractions in the GHRC preparation was analysed by SDS-PAGE and silver staining for experimental and control mock isolations (Fig. 6). Only the 90-92 kDa and a 20-22 kDa fraction became significantly reduced in mock isolations, showing that the other detectable bands are contaminants. The 20-22-kDa band behaved similarly to the 24-kDa band observed by others [15], eluting at a higher ionic strength than GHRC from DEAE-Sephacel. There is controversy
I5 rat livers
Specific activity (pm01 receptor per mg total protein)
% Yield of receptor
0.489 0.587 1.36 x IO’ 4.36 x IO’ with overall purification
100 90 54 46
Fold purification
I 1.2 2777 8916’
factors between 85Ot&9000
as to the nature of this 2t324-kDa fraction and whether it is associated with the GHRC. The presence of RNA in that fraction has been suggested by some [34,35], while others have been unable to detect this fraction by silver staining [36]. We could not detect a 72-kDa fraction specifically associated with GHRC [12], neither did we observe the 4&41-kDa fraction found by Grandics et al. [I 51. DISCUSSION Based on the preceding investigations, we routinely isolate the untransformed GHRC as described in Experimental to obtain a 9,000-fold purified GHRC preparation with 46% yield (see Table 4). Idziorek et al. [7] have recently published a very similar isolation procedure using high performance size exclusion chromatography as a final step to obtain >45% pure GHRC. Presuming one steroid-binding subunit per 92-kDa polypeptide, the GHRC eluted off DEAE-Sephacel in our experiments in the presence of molybdate is 3040% pure. However the Coomassie dye staining of the gel suggests an almost homogeneous receptor preparation (Fig. 6). This discrepancy was observed by others [7] and is most likely explained by the presence of a 90-92-kDa non-steroid-binding protein contaminant [ 161, which is also evident from Fig. 6C, D. We have found that the amount of 9&92-kDa contaminant varies for different isolations. The sedimentation and DNA-binding properties of the highly purified transformed and untransformed GHRC will be reported elsewhere [37]. The affinity purification of molybdate-stabilized untransformed GHRC followed by heat transformation of the DEAE-Sephacel eluate in the absence of molybdate and in the presence of BSA was found to reproducibly result in highly purified, undegraded, transformed GHRC with a high yield. Acknowledgements-We thank Dr M. V. Govindan (Laboratoire de Genttique Moieculair des Eucaryotes, Strasbourg, France) for his kind gift of polyclonal anti-glucocorticoid receptor antibodies. This work was supported by a grant from the Council for Scientific and Industrial Research (S.A.) and the University of Cape Town Research Committee to C.V.H. REFERENCES I. Hollenberg
S. M., Weinberger Oro A., Lebo R., Thompson
C., Ong E. S., Cerelli G., E. B., Rosenfield M. G.
Glucocorticoid
receptor affinity purification
and Evans R. M.: Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318 (1985) 6355641. 2. Wrange t). and Gustafsson J. A.: Separation of the hormone- and DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. J. biol.
19.
20.
Chem. 253 (1978) 856865.
3. Carlstedt-Duke J., Wrange t), Dahlberg E., Gustafsson J. A. and Hogberg B.: Transformation of the glucocorticoid receptor in rat liver cytosol by lyosomal enzymes. J. biol. Chem. 254 (1979) 1537-1539. 4. Govindan M. V. and Sekeris C. E.: Purification of two dexamethasone-binding proteins from rat-liver cytosol. Eur. J. Biochem. 89 (1978) 955104. 5. Manz B., Heubner A., Kijhler I., Grill H. J. and Pollow K.: Synthesis of biotin-labelled dexamethasone derivatives. Novel hormone affinity probes. Eur. J. Biochem. 131 (1983) 333-338.
6. Weisz A., Baxter J. D. and Lan N. C.: Interaction of sodium molybdate with highly purified glucocorticoid receptor. J. steroid Biochem. 20 (1984) 289-293. 7. Idziorek T., Formstecher P., Danze P. M., Sablonniere B., Lustenberger P., Richard C., Dumur V. and Dautrevaux M.: Characterization of the purified molybdatestabilized glucocorticoid receptor from rat liver. An in vitro transformable complex. Eur. J. Biochem. 153 (1985) 65-74.
8. Wrange G., Carlstedt-Duke J. and Gustafsson J. A.: Purification of the glucocorticoid receptor from rat liver cytosol. J. biol. Chem. 254 (1979) 92849290. 9. Govindan M. V. and Manz B.: Three-step purification of glucocorticoid receptor from rat liver. Eur. J. Biothem. 108 (1980) 47-53. 10. Westphal H. M. and Beato M.: The activated glucocorticoid receptor of rat liver. Eur. J. Biochem. 106 (1980) 395403. 11. Govindan M. V. and Gronemeyer H.: Characterization of the rat liver glucocorticoid receptor purified by DNA-cellulose and ligand affinity chromatography. J. biol. Chem. 259 (1984) 12915-12924. 12. Wrange d., Okret S., Radojcic M., Carlstedt-Duke J. and Gustafsson J. A.: Characterization of the purified activated glucocorticoid receptor from rat liver cytosol. J. biol. Chem. 259 (1984) 45344541. 13. Kurl R. N. and Jacob S. T.: Phosphorylation of purified glucocorticoid receptor from rat liver by an endogeneous protein kinase. Biochem. biophys. Res. Commun. 119 (1984) 700-705.
14. Lustenberger P., Formstecher P. and Dautrevaux M.: Purification of rat liver glucocorticoid receptor by affinity chromatography. Design of a suitable adsorbent. J. steroid Biochem.
14 (198i) 697-703.
15. Grandics P.. Miller A.. Schmidt T. J.. Mittman D. and Litwack G.:’ Purification of the unactivated glucocorticoid receptor and its subsequent in uitro activation. J. biol. Chem. 259 (1984) 3173-3180. 16. Hapgood J. P., Sabbatini G. P. and von Holt C.: Rat liver glucocorticoid receptor isolated by affinity chromatography is not a Mg*+- or Ca2+-dependent protein kinase. Biochemistry 25 (1986) 7529-7534. 17. Lan N. C., Karin M., Nguyen T., Weisz A., Birnbaum M. J., Eberhardt N. L. and Baxter J. D.: Mechanisms of glucocorticoid hormone action. J. steroid Biochem. 20 (1984) 77-88.
18. Webb M. L., Miller-Diener A. S. and Litwack G.: Purification, characterization and activation of the
21.
22. 23.
777
glucocorticoid-receptor complex from rat kidney cortex. Biochemistry 24 (1985) 1946-1952. Sica V., Parikh I., Nola E., Puca G. A. and Cuatrecasas P.: Affinity chromatography and the purification of estrogen receptors. J. biol. Chem. 248 (1973) 6543-6558. Cuatrecasas P.: Protein purification by affinity chromatography. J. biol. Chem. 245 (1970) 3059-3065. Lustenberger P., Formstecher P. and Dautrevaux M.: Quantitative determination of alkvlamino side-chains coupled to agarose beads. J. Chromat. 193 (1980) 451457. Riordan J. F. and Vallee B. L.: Acetylation. Merh. Enzym. 11 (1967) 565-570. Rousseau G. G., Baxter J. D. and Tomkins G. M.: Glucocorticoid receptors: relations between steroid binding and biological effects. J. molec. Biol. 67 (1972)
99-105. 24. Ansorg W.: In Proceedings
Electrophoresis (Edited by D. Stathakos). Walter de Gruyter, Berlin (1982) pp. 235-245. 25. Bensadoun A. and Weinstein D.: Assay of proteins in the presence of interfering materials. Analyt. Biochem. 70 (1976) 241-250. 26. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall
R.: Protein measurement with Folin nhenol reaeent. J. biol. Chem. 193 (1951) 265-275.
27. Govindan M. V. and Sekeris C. E.: Partial purification of rat liver glucocorticoid binding proteins by affinity chromatography. Steroids 28 (1976) 499-507. 28. Govindan M. V. and Sekeris C. E.: Purification of two dexamethasone-binding proteins from rat liver cytosol. Eur. J. Biochem. 89 (1978) 95-104. 29. Tsawdaroglou N. G., Govindan M. V., Schmidt W. and
Sekeris C. E.: Dexamethasone binding proteins in cytosol and nucleus of rat thvmocvtes. Eur. J. Biochem. 114 (1981) 305-313. _ _ 30. Failla D.. Tomkins G. M. and Santi D. V.: Partial purification of a glucocorticoid receptor. Proc. natn. Acad. Sci. U.S.A. 72 (1975) 3849-3852. 31. Kalimi M. and Hubbard J.: Interaction
of sodium thiocyanate with rat hepatic glucocorticoid-receptor complexes. Biochim. biophys. Acta 719 (1982) 488-494. 32. Coty W. A.: Reversible dissociation of steroid hormone receptor complexes by mercurial reagents. J. biol. Chem. 255 (1980) 8035-8037.
33. Leach K. L., Dahmer M. K., Hammond N. D., Sando J. J. and Pratt W. B.: Molybdate inhibition of glucocorticoid receptor inactivation and transformation. J. biol. Chem. 254 (1979) 1188411890. 34. Grandics P., Miller A., Schmidt T. J. and Litwack G.: Phosphorylation in uiuo of rat hepatic glucocorticoid receptor. Biochem. biophys. Res. Commun. 120 (1984) 59-65.
35. Housley P. R. and Pratt W. B.: Small 32P-labeled components of L-cells that coelute with the molybdatestabilized glucocorticoid receptor during purification. Fedn. Proc. Fedn. Am. sot. Exp. Biol. 43 (1984) 1572.
36. KovaEiE-Milivojevic B. and Vedeckis W. V.: Absence of detectable ribonucleic acid in the purified untransformed mouse glucocorticoid receptor. Biochemistry 25 (1986) 82668273.
37. Hapgood J. P. and von Holt C.: Transformation in vitro and covalent modification with biotin of steroidaffinity-purified rat-liver glucocorticoid-hormonereceptor complex. Eur. J. Biochem. 166 (1987) 415420.