Effect of heat treatment on the glucocorticoid-receptor complex dependence on steroid structure

196

Biochimica et Biophysica A cta,

761 (1983) 196-203 Elsevier

BBA 21624

EFFECT OF HEAT TREATMENT ON THE GLUCOCORTICOID-RECEPTOR COMPLEX DEPENDENCE ON STEROID STRUCTURE KLARA T()TH and PI~TERARANYI * Second Institute of Biochemistry, Semmelweis University Medical School, H-1444 Budapest (Hungary)

(ReceivedJune 6th, 1983)

Key words: Glucoeorticoid; Heat treatment," Steroid receptor

The Arrhenius plot of heat inactivation of the rat liver glucocorticoid receptor protein gave a straight line +

with A H + = 115 k J / m o l over the 0 - 4 4 0 C range. Molybdate ions were considerably protective but did not affect the linearity or slope of the Arrhenius plot. The effect of triamcinolone acetonide on heat stability of the receptor was similar to that of molybdate. On the other hand, glucocorticoid antagonists, although bound to the receptor, did not protect it from heat inactivation. Incubation of the complex of the glucocorticoid receptor with optimal glucocorticoids under activating conditions (elevated temperature or ionic strength) resulted in a considerable decrease in the dissociation rate. However, if the complex was incubated at 25°C in the presence of molybdate, its dissociation rate did not change. Heat treatment without molybdate of complexes of glucocorticoid antagonists did not decrease the dissociation rate. These findings indicate that the decrease in dissociation rate is probably related to nucleophilic transformation. An lift-hydroxyl group in the steroid structure seems to be an absolute requirement both for protection of the receptor against beat inactivation and for stabilization of the complex under conditions that promote nucleophilic transformation.

Introduction The heat stability of the glucocorticoid receptor protein has been extensively analyzed [1-3], mostly for practical reasons, since its extreme lability has hindered its study a great deal. Much useful information concerning the essential functional groups of the receptor as well as its interaction with other cytoplasmic constituents has been deduced from these experiments [4-7]. Experiments with the glucocorticoid receptor [3,5,8] and other receptor proteins [9,10] have demonstrated ligand protection against heat inactivation. (Heat inactivation, meaning the loss of steroid-binding capacity of the

* To whom reprint requests and correspondence should be sent. 0304-4165/83/$03.00 © 1983 ElsevierSciencePublishers B.V.

receptor protein, should not be confused with the term 'activation' or 'heat activation', which is defined below.) In these studies, however, only optimal agonists, which also displayed the highest possible affinity for the receptor, were used. Ligand specificity of the glucocorticoid receptor has also been studied in detail in a number of laboratories [11-13]. This property has been reported dependent upon temperature [14-15]. Nonetheless, almost all the information we have to date relates to low temperature (0°C). Glucocorticoid-receptor complexes undergo 'activation' (nucleophilic transformation) upon incubation at elevated temperatures or in media of high ionic strength. This operational term describes a transformation of the complex that renders it capable of binding to the nucleus. Activation is essential for glucocorticoid hormone action.

197

(For a review see Ref. 16.) It has been recently reported that activation of the dexamethasone-receptor complex resulted in a decreased dissociation rate [17]. Similar behavior of the estrogen-receptor and of the progesterone-receptor complex has also been described [18-20]. Data concerning the structural requirements for such a decrease in the rate of dissociation, under conditions that promote nucleophilic transformation (activation) of the complex, are lacking. Therefore, it seems to be of special interest to determine how incubation at moderately high temperatures (25°C) influences the stability of the complex of the glucocorticoid receptor with different steroids. This problem is, of course, also closely related to heat inactivation of the receptor in the presence of steroidal ligands. In order to address these questions, a reexamination of the heat inactivation of the unbound receptor was necessary, since this process is not yet fully understood. Anomalous heat inactivation kinetics have been observed [21] and it is not clear whether heat inactivation is a spontaneous or enzyme-catalyzed process. Heat inactivation as well as dissociation of the complex were monitored by the exchange assay developed in our laboratory [22,23]. This method offers the foil.owing advantages: (i) unlabeled steroids can be tested without the need for tritiated analogs, using a single tracer, and (ii) denaturation of and dissociation from the same steroid-binding protein is measured in all cases, regardless of other potential sites for binding the unlabeled steroid. Materials and Methods

Animals Male Wistar rats (150-200 g body weight) were used. They were adrenalectomized and maintained on 0.9% saline for 2-4 days before the experiments. Chemicals [1,2-3H]Triamcinolone acetonide, 25 Ci/mmol, was purchased from Amersham International, U.K. Dexamethasone was a gift from Merck, Sharp and Dohme, Rahway, N J, U.S.A., llfl,20ct-dihydroxypregn-4-en-3-one and llfl,20fl-dihydroxy-pregn-4en-3-one were donated by the Steroid Reference

Collection, Hampstead, London (curator: Prof. D.N. Kirk). Other nonlabeled steroid were obtained from G. Richter Ltd., Budapest, Hungary. The purity of the steroid, chemical and radiochemical was verified by thin-layer chromatography. Charcoal (Norit A) and dithiothreitol were purchased from Serva, Heidelberg, F.R.G. Sephadex G-25 was from Pharmacia and cellulose CF-11 from Whatman Ltd. All other chemicals were obtained from Reanal, Budapest, Hungary. Preparation of the cytosol Livers were perfused in situ with physiological saline, then removed and weighed. All subsequent procedures were performed at 0°C, unless indicated otherwise. Livers were homogenized with 1.5 vol. of a 0.01 M Tris/1.5 mM EDTA/2 mM dithiothreitol (pH 7.4) buffer in a motor-driven Teflon-glass homogenizer. In some experiments homogenizing buffer also contained 10 mM sodium molybdate. The homogenate was spun at 100000 x g for 45 min in the cold and the supernatant was used as cytosol. DNA-cellulose was prepared by the method of Alberts and Herrick [24]. DNAcellulose binding was determined as published earlier [25]. Determination of denaturation rate constant Cytosol was incubated at various temperatures. At the indicated times, 50-#1 aliquots were removed to determine specific binding. To this end, aliquots were incubated at 0°C for 2.5 h with 50/~1 of 80 nM [3H]triamcinolone acetonide in the presence or absence of 5 ttM unlabeled triamcinolone acetonide. At the end of incubation period, excess free steroid was removed by the addition of 200/~1 of ice-cold 0.5% charcoal suspension containing 0.05% dextran. The samples were vigorously stirred and spun at 1000 Y
198

Determination of the receptor in the presence of steroids Cytosol containing 1 #M unlabeled steroid was incubated at 0°C for 2 h and at 25°C thereafter. 150 /~1 aliquots were withdrawn after various incubation times. They were put on ice and the free steroid was removed by addition of 30 #1 ice-cold 5% charcoal/0.5% dextran suspension. After centrifugation at 1000 × g for 5 min 100 #1 of the clear supematant was mixed with an equal volume of 80 nM [ 3H]triamcinolone acetonide to exchange unlabeled steroid, and to determine the amount of receptor still capable of binding glucocorticoid. Incubation was continued at 10°C for 5 h. In case of suboptimal agonists or antagonists, a full exchange was achieved under these conditions. For determination of nonspecific binding, incubation was at 10°C in the presence of 40 nM [3H]triamcinolone acetonide and 5 # M unlabeled triamcinolone acetonide (final concentration). At the end of the incubation, excess labeled steroid was removed by the addition of 200 btl of ice-cold 0.5% charcoal/0.05% dextran suspension. After centrifugation at 1000 × g for 5 min, 250 btl of clear supernatant was transferred into vials for scintillation counting. The total concentration of receptor was determined by incubation in the presence of a saturating concentration of [3H]triamcinolone acetonide for a sufficiently long time (3-4 h). This value was also corrected for nonspecific binding. Dissociation rate constants were determined by the indirect method of Ar/myi and Quiroga [22]. In brief, the glucocorticoid receptor was equilibrated with the unlabeled steroid under study. After the removal of the excess of unlabeled steroid by the addition of charcoal, [3H]triamcinolone acetonide was added and the dissociation of receptor-unlabeled steroid complex was detected by measuring the association of the receptor-[3H]triamcinolone acetonide complex. The temperature of dissociation was 10°C. Data were analyzed from semilogarithmic plot. When two phases were observed, data of the first phase were corrected for extrapolated values of the second phase. Miscellaneous Liquid scintillation counting was performed in a Beckman LS 9000 radiospectrofluorometer. Composition of the scintillation liquid was

t o l u e n e / P P O / P O P O P (100 ml : 4 g : 0.05 g). Counting efficiency was 55-58%. All data points shown in the figures are averages of two independent determinations. Experimental error was below 10% in all cases. Results

Heat inactivation The heat stability of the glucocorticoid receptor protein was investigated over a 0-44°C range. Fig. 1 shows the time-course of heat inactivation at selected temperatures. Denaturation of the receptor followed first-order kinetics at the higher temperatures (Fig. la), whereas the kinetics were more

lOC

x~ 5G

20

of

0

10

20

30

z.O

50 t (ram)

Io

~

I00

~°

u_

20 \ '60

"120 '180 '2L.0 '300 '360 t (rain)

Fig. 1. Heat inactivation of the receptor protein. Rat liver cytosol was incubated at different temperatures immediately after preparation (filled symbols in (a) and all symbols in (b) (or after preincubation at 0°C with [3H]triamcinolone acetonide (open symbols in (a)). Remaining specific binding was determined after various incubation times and was expressed as percentage of initial specific binding. Incubation temperatures: (a) 37°C (e e, © ©) and 44°C (A A, A zx); (b) 17°C ( × - x ) and 24°C

(o

o).

199 complex at low or intermediate temperatures (Fig. lb). Between 0 - 2 7 ° C , an initial rise or plateau of binding capacity was observed, in accordance with our earlier findings [21]. The simplest explanation of this behavior is that a fraction of the receptor molecules was initially b o u n d by endogenous steroids, p r o b a b l y by testosterone or its metabolites, because the animals had not been castrated. Testosterone can be easily exchanged by triamcinolone acetonide through a process with a 40 min half-time at 0 ° C [13]. Thus, the a m o u n t of triamcinolone acetonide complex increased if the cytosol was preincubated. However, at higher temperatures, dissociation of the masked receptors m a y require only a few minutes or seconds, and accordingly this process was not reflected in the denaturation kinetics. F o r construction of an Arrhenius plot, straight lines were fitted to the experimental points by m e t h o d of least squares. In cases where the timecourse showed a maximum, only the points bey o n d that m a x i m u m were used for fitting. The Arrhenius plot gave a straight line over a large temperature range ( 0 - 4 4 ° C , Fig. 2). It revealed no sign of conformational change in the absence of steroid, and showed that heat inactivation was not

associated with changes in the heat capacity of the protein. Activation enthalpy, AHS, was only 115 kJ/mol. We next analyzed how complexing with steroidal ligands influences the heat stability of the receptor. Since we decided to include suboptimal glucocorticoids and antiglucocorticoids rather than the most studied optimal glucocorticoids [26], the use of an exchange assay for the remaining triamcinolone acetonide binding seemed most useful. T o this end, the preformed complex of the glucocorticoid receptor with unlabeled steroids was incubated at 25°C. The complex was then allowed to dissociate ~ind the remaining [3H]triamcinolone acetonide binding capacity was determined as described in Materials and Methods. U n d e r the conditions employed all the b o u n d steroids were fully exchanged (see Fig. 4 and Table II). N o protective effect whatsover by these steroids was observed (Table I). In fact, denaturation was a little more p r o n o u n c e d in the presence of cortexolone and progesterone than in their absence. The reason of this finding is u n k n o w n at present. F o r comparison with an optimal glucocorticoid, triamcinolone acetonide was used. The time-course of denaturation of the glucocorticoid receptor protein in the

,'7

TABLE I

e-

HEAT INACTIVATION OF THE GLUCOCORTICOID RECEPTOR IN THE PRESENCE OF STEROIDS Rat-liver cytosol was equilibrated with unlabeled steroid for 2 h at 0°C or incubated under similar conditions without additives. Incubation was then continued at 25°C for 45 or 90 min. At that time excess steroid, where present, was removed by charcoal treatment and unlabeled steroid was exchanged for [3H]triamcinolone acetonide during a 5 h incubation at 10°C. Protein bound radioactivity was determined, corrected for nonspecific binding and expressed as percentage of zero-time binding.

1.0

lo-1

\\

10-2

lo-3

*\ ×

\

I(~4 .

Steroid

O: 3.0

'3.2

'3.4 '3.6 10-3x 1_.. (K-1) T

Fig. 2. Arrhenius plots of heat inactivation. Arrhenius plots were constructed from the slopes of the linear parts of the kinetic curves presented in Fig. 1, and from similar time-courses. Temperature ranges were as follows: 25-40°C for heat inactivation in the presence of 10 mM/molybdate (e e) or of triamcinolone acetonide (zx A) and 0-r44°C in the absence of additives ( x - × ).

None Cortexolone Progesterone Testosterone Deoxycorticosterone

Remaining activity at 25°C (%) after 45 min

90 min

62.8 36.8 58.6 71.1

55.9 32.9 41.2 48.6

74.3

51.8

200 continuous presence of [3H]triamcinolone acetonide is shown in Fig. la. In agreement with earlier observations [3,5,8], a p r o n o u n c e d stabilizing effect of the glucocorticoid agonist was observed, in this case, the temperature range was restricted to 2 5 - 4 4 ° C . Interestingly, the protective effect of 10 m M molybdate was similar to that of triamcinolone acetonide (Fig. 2), supporting the novel suggestion that molybdate m a y act directly on the receptor protein [6].

x

E

20 10

\, \,

Dissociation after heat treatment

\

Heat treatment of a preformed hydrocortisonereceptor complex resulted in a slowing d o w n of the dissociation rate along with a striking change in the time-course (Fig. 3). If dissociation were measured without any heat treatment, it followed first-order kinetics. However, the kinetics became biphasic if the complex had been incubated at 25°C before its dissociation was monitored. This change became more and more p r o n o u n c e d with increasing duration of heat treatment. The second phase was characterized by a very slow dissociation rate even at 10°C.

m x

E

01 0

1

2

3

t,

*\ 5

6

7

t (hours}

Fig. 3. Dissociation of the hydrocortisone-receptor complex after heat treatment. Rat-liver cytosol glucocorticoid receptor was complexed with unlabeled hydrocortisone at 0°C. The complex was then incubated at 25°C for 10 min (×--×), 30 min ( + - - + ) , 60 min (O ©) or 90 min (e I). Time-course of dissociation was then measured by exchange for [3H]triamcinolone acetonide at 10°C. Total receptor-[aH]triamcinolone acetonide-receptor complex (clam) is plotted semilogarithmically against time.

\ o

o.~

~\'~\ \,

o~

0

~-

'2

'3

7. t (hours)

Fig. 4. Dissociation of receptor-ligand complex after heat treatment. The glucocorticoid receptor was complexed with corticosterone (O O, • $) or ll-deoxycorticosterone ('7 ~, • •). It was further incubated at 0°C (open symbols) or 25°C (filled symbols) for 60 min and the dissociation kinetics were followed at 10°C thereafter.

The corticosterone-receptor complex behaved similarly and so did complexes of a few other, but not all, steroids (Fig. 4). The dissociation of progesterone, deoxycorticosterone, testosterone, e p i p r e d i n i s o l o n e or cortexolone from the glucocorticoid receptor did not change much, whether or not they were heat treated. F o r quantitative measurement of the changes brought about by incubation of the complex at 25°C, rate constants were calculated by fitting straight lines to the time points. These are compiled in Table II. The ratio of the rate constants before and after heat treatment of the complexes was larger than those for optimal glucocorticoids and for llfl,20-dihydroxypregn-4-en-3ones. This ratio, however, was close to 1 for steroids lacking the l i f t - h y d r o x y l substituent (all suboptimal agonists or antagonists).

Relation of nucleophilic transformation Incubation of optimal glucocorticoid-receptor complexes at 25°C for times that considerably reduced their dissociation rate is k n o w n to result

201 TABLE II D I S S O C I A T I O N R A T E CONSTANTS OF S T E R O I D - G L U C O C O R T I C O I D R E C E P T O R C O MP LEX ES Steroid

Testosterone a Progesterone a Cortexolone a 11-Deoxycorticosterone a Epiprednisolone b Corticosterone c Hydrocortisone c 1 lf120 a-Dihydroxypregn-4-en-3-one c.r 1 lfl,20fl-Dihydroxypregn-4-en-3-one ~,t

k_ 2 d (ooc)

k_ 2 (25oc) o

k_ 2 (OoC)/k_2 (25oc)

(rain - 1)

(min- ])

2.89.10- 2 1.53-10 - 2 1.87-10 - 2 1.66.10 - 2 3.52-10-2 1.02-10- 2 9.30.10- 3

4.07.10- 2 1.81-10- 2 1.90.10 - 2 1.59.10 - 2 3.91.10-2 1.70-10- 3 3.53.10-3

0.71 0.84 0.99 1.05 0.90 6.01 2.61

2.3 .10 -2

1.1 .10 -2

2.10

3.9 • 1 0 - 2

1.4 - 10- 2

2.80

a b c d e

No 11-hydroxyl substituent. l l a - H y d r o x y l substituent. l i f t - H y d r o x y l substituent. Dissociation rate constant was determined at 10°C without preincubation at 25°C. Dissociation rate constant was determined at 10°C after 60 min preincubation at 25°C, from the second, linear phase of the kinetics. f Dissociation rate constant was determined at 0°C, heat treatment was 30 min at 25°C.

in nucleophilic transformation as well. In order to decide whether the observed decrease in dissociation rate was due to transformation of the complex or to some otherwise unrecognized effect of

\

cp

2 1

~J

\.

\v 0.5

\. \

o

\

Olo

\

\

'1 '1

'2

'3

'2

'3

't.

'5 t (hours)

'& t (hours}

Fig. 5. Dissociation of the hydrocortisone-receptor complex in the presence of molybdate ions. Cytosol glucocorticoid receptor was complexed with unlabeled hydrocortisone at 0°C. The solution was then made 10 mM with respect to sodium molybdate and the incubation was continued. After 90 min of incubation, dissociation rate was m e a s u r e d a s described in the legend to Fig. 3. Incubation temperatures: 90 min, 0°C (O); 60 min, O°C; 30 min, 25°C (v); 90 rain, 25°C ( x ) .

Fig. 6. Dissociation of the hydrocortisone-receptor complex after KC1 treatment. Cytosol glucocorticoid receptor was complexed with unlabeled hydrocortisone at 0°C. The solution was then made to 0.3 M with respect to KCI. The complexed receptor was separated from KCI and free hydrocortisone by filtration through a Sephadex G-25 column immediately after the addition of KCI ( O ©) or after 2 h of incubation at 0° C in the presence of KCI (@ a). Dissociation rate was measured as described in the legend to Fig. 3.

202

heat treatment, the following experiments were performed. Hydrocortisone-receptor complex was incubated at 25°C in the presence of molybdate ions that prevent activation. A heat treatment of 30 or even 90 min had no significant effect on the time-course of dissociation, in contrast to what was seen in the absence of molybdate (Fig. 5). On the other hand, if transformation of the complex was induced by increased ionic strength, dissociation of the performed hydrocortisone-receptor complex was slowed down in a manner similar to that observed in the case of heat-induced transformation (Fig. 6). Discussion

Heat inactivation of the rat liver glucocorticoid receptor followed first-order kinetics (after a lag phase at lower temperatures) with a linear Arrhenius plot over the 0 - 4 4 ° C range. Linearity of the plot revealed no interconversion of receptor forms with different heat stabilities. Moreover, a curvilinear Arrhenius plot would be expected if denaturation were due to unfolding of the receptor protein [27,28]. However, this was not the case. The conformation change that resulted in the loss of steroid-binding capacity should thus be limited to a relatively small poretion of the receptor protein. The side-chains in question should be mostly in contact with the solvent in both their 'native' and 'denatured' state. This mechanism is compatible also with the AH* value of the denaturation process, 115 k J / m o l , which is relatively small for thermal denaturation of a protein [29]. An alternative explanation is that what is commonly called heat inactivation is, in fact, an enzyme-catalyzed reaction, for which a AH* of this magnitude is not impossible [30]. Phosphatases [31,32] or proteases [33] present in the cytosol are candidates for this protein. Complexing of the receptor with antiglucocorticoids did not protect it from heat inactivation. On the other hand, the optimal glucocorticoid, triamcinolone acetonide stabilized the receptor considerably, without influencing the linearity or the slope of the Arrhenius plot. Heat inactivation of the receptor protein probably follows similar mechanisms in the uncomplexed and complexed states. Incubation of the complex of the gluco-

corticoid receptor with optimal glucocorticoids at 25°C (conditions that promote nucleophilic transformation) resulted in a considerable stabilization of t h e complex, i.e., dissociation of the steroid from the receptor became slower. The decrease in the dissociation rate constant and nucleophilic transformation ('activation') of the complex seem to be interdependent. Namely, if nucleophilic transformation was promoted at 0°C, by increased ionic strength, stabilization of the complex also took place. On the other other hand, if nucleophilic transformation at 25°C was blocked by molybdate ions, rate of dissociation did not change upon incubation. Moreover, complexes of antiglucocorticoids with the receptor, which cannot undergo nucleophilic transformation upon heat treatment [34,35] did not become more stable. By use of to the recently developed exchange method for determination of dissociation rate constants [22,23], steroids whose tritiated analogues were not avaible could also be included in our studies. Dependence of steroid structure of both heat stability and dissociation kinetics of the ligand-receptor complex could be analyzed this way. An liB-hydroxyl group seems to be an absolute requirement for a steroid to show a decreased dissociation rate upon heat treatment. It cannot be substituted by an lla-hydroxyl function. In agreement with this, complexes of steroids that display low affinity for the glucocorticoid receptor at 0°C but do have an l i f t - h y d r o x y l (llB-20-dihydroxysteroids) also show a considerable stabilization after incubation at 25°C. On the basis of the above data, the following model of steroid binding and transformation process can be put forth (Fig. 7). The receptor (R)

= ___

~

7

o=~:o

RS

heat

o=~]=o

R S

~

_

RS-OH

~

R'*' S - OH

S-OH

Fig. 7. Conformational change of the glucocorticoid receptor. Symbols: R: receptor, S-OH: llfl-hydroxysteroid-3,20-dione, S: steroid without an llfl-hydroxy function, R. S or R. S-OH: initial complexes, R * . S - O H : transformed complex. Hatched area: DNA-binding site of the R*. S-OH.

203 a n d the steroid (S or S-OH) form a complex ( R . S or R - S-OH) o n their encounter. If hydrocortisone molecule is taken as a reference, b o t h the 3- a n d 20-oxo groups b u t n o n e of the hydroxyls are required for a specific interaction at this stage [13,36]. A c o n f o r m a t i o n a l change is required, then, for t r a n s f o r m a t i o n of the complex. This is initiated b y a specific i n t e r a c t i o n of the l i f t - h y d r o x y l group with some u n i d e n t i f i e d side-chain of the receptor. R . S is therefore u n a b l e to transform. The conform a t i o n a l change requires i n c u b a t i o n at elevated t e m p e r a t u r e a n d / o r at increased ionic strength a n d is blocked b y m o l y b d a t e . The t r a n s f o r m e d complex ( R * . S-OH) is able to b i n d to nuclei or D N A . Dissociation of the steroid from R* • S-OH is c o n s i d e r a b l y slower t h a n from R - S-OH. T r a n s f o r m a t i o n of the complex slows heat i n a c t i v a t i o n (not shown in the figure). Therefore, the glucocortiocid receptor p r o t e i n is more stable in the presence of l lfl-hydroxysteroids t h a n in the presence of steroids lacking that group or in the absence of ligands.

Acknowledgements This work was supported b y the C e n t r a l Research F u n d of the H u n g a r i a n A c a d e m y of Sciences. The authors are i n d e b t e d to Professor I, Horv/tth for his c o n t i n u o u s interest a n d criticism. T h a n k s are due to Ms. Olga Losonczi a n d Ms. J u s z t i n a N a g y for the expert technical assistance.

References 1 Schmid, W., Grote, H. and Sekeris, c.E. (1976) Mol. Cell. Endocrinol. 5, 223-241 2 Schaumburg, B.P. (1972) Biochim. Biophys. Acta 261, 219-235 3 Bell, P.A. and Munck, A. (1973) Biochem. J. 136, 97-107 4 Rees, A.M. and Bell, P.A. (1975) Biochim. Biophys. Acta 411, 121-132 5 Nielsen, C.J., Sando, J.J., Vogel, W.M. and Pratt, W.B. (1977) J. Biol. Chem. 252, 7568-7578 6 Housley, P.R., Dahmer, M.K. and Pratt, W.B. (1982) J. Biol. Chem. 257, 8615-8618 7 McBlain, W.A. and Shyamala, G. (1980) J. Biol. Chem. 257, 8615-8618 8 Rousseau, G.G., Baxter, J.D. and Tomkins, G.M. (1972) J. Mol. Biol. 67, 99-115 9 Wilson, E.M. and French, F.S. (1976) J. Biol. Chem. 251, 5620-5629

10 Zak~, T. and Tbth, M. (1977) Steroids 30, 751-764 11 Ojasoo, T. and Raynaud, J.P. (1978) Cancer Res. 38, 4186-4198 12 Rousseau, G.G. and Schmit, J.P. (1979) J. Steroid Biochem. 8, 911-919 13 Arhnyi, P. (1982) J. Steroid Biochem. 17, 147-141 14 Jones, T.L., Sloman, J.C. and Bell, P.A. (1979) Mol. Cell Endocrinol. 13, 83-92 15 McDonald, R.G. and Cidlowsky, J.A. (1979) J. Steroid Biochem. 10, 21-29 16 Schmidt, T.J., Barnett, C.A. and Litwack, G. (1982) J. Cell. Biochem. 20, 15-27 17 McBlain, W.A., Tort, D.O. and Shyamala, G. (1981) Biochemistry 20, 6790-6798 18 Weichman, B.M. and Notides, A.C. (1977) J. Biol. Chem. 252, 8856-88,62 19 Weichman, B.M. and Notides, A.C. (1979) Biochemistry18, 220-225 20 Chang-Ren, Y., Seeley, D.H., Mester, J., Wolfson, A. and Baulieu, E-E. (1983) Biochim. Biophys. Acta 755, 428-433 21 Arfinyi, P. and N~tray, A. (1980) Acta Biochim. Biophys. Acad. Sci. Hung. 15, 185-198 22 Affmyi, P. and Quiroga, v. (1980) J. Steroid. Biochem. 13, 1167-172 23 Arhnyi, P. (1980) in Progress in Cancer Research and Therapy (Iacobelli, S., King, R.J.B., Lindner, H.R. and Lippman, M.E., eds.), Vol. 14, p~. 217-226, Raven Press, New York 24 Alberts, B. and Herrick, G. (1971) Methods Enzymol. 21, 198-217 25 Arhyi, P. and Nhrat, A. (1980) J. Steroid Biochem. 12, 267-272 26 Samuels, H.H. and Tomkins, G.M. (1970) J. Mol. Biol. 52, 57-74 27 Privalov, P.L. (1982) in Advances in Protein Chemistry (Anfinsen, C.B., Edsall, J.T. and Richards, F.M., eds.), Vol. 35, pp. 1-104, Academic Press, New New York 28 Tanford, Ch. (1970) in Advances in Protein Chemistry (Anfinsen, C.B., Edsall, J.I. and Richards, F.M., eds.), Vol. 24, pp. 2-95, Academic Press, New York 29 Joly, M. (1965) in A Physico-chemical Approach to the Denaturation" of Proteins. Molecular Biology, Vol. 6, pp. 190-216, Academic Press, New York 30 Dixon, M. and Webb, M.C. (1964) Enzymes, 3rd Edn., pp. 169-181, Academic Press, New York 31 Nielsen, C.J., Sando, J.J. and Pratt, W.B. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 1398-1402 32 Barnett, C.A., Schmidt, T.J. and Litwack, G. (1980) Biochemistry 19, 5446-5455 33 Sherman, M.R., Picketing, L.A., Rollwagen, F.M. and Miller, L.K. (1978) Fed. Proc. 37, 167-173 34 Rousseau, G.G., Higgins, S.J., Baxter, J.O., Gelfand, D. and Tomkins, G.M. (1975) J. Biol. Chem. 250, 6015-6021 35 Wira, C.R. and Munck, A. (1974) J. Biol. Chem. 249, 5328-5336 36 Raynaud, J.P., Ojasoo, T., Bonton, M.M. and Philibert, D. (1979) in Drug Design (Ariens, E.J., ed.), Vol. VIII, pp. 169-214, Academic Press, New York