antiprogestin zk-98.299 and progesterone display differential binding characteristics in the human...

Biochimica et Biophysica Acta, 1175 (1992) 73-80 73 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4889/92/$05.00

BBAMCR 13287

Antiprogestin ZK-98.299 and progesterone display differential binding characteristics in the human myometrial cytosol

Anil D'souza a, Indira N. Hinduja b and Chander P. Puri a

a Institute for Research in Reproduction (ICMR) and b K~E.M. Hospital and Seth G.S. Medical College, Parel, Bombay (India)

(Received 21 July 1992)

Key words: Antiprogestin ZK-98.299; Progesterone; Receptor; (Human myometrium)

To investigate whether the synthetic progesterone antagonist ZK-98.299 binds to progesterone receptor or also has distinct binding sites, the binding characteristics of ZK-98.299 were compared with those of progesterone in the human myometrial cytosol. [3H]ZK-98.299 and [3H]progesterone showed specific binding in the myometrial cytosol and the binding of each radiolabelled ligand could be displaced with the respective ligand in a dose-response manner. However, while the binding of [3H]progesterone could be completely blocked with progesterone or ZK 98.299, the binding of [3H]ZK-98.299 could not be displaced more than 50%. The non-specific binding of [3H]ZK-98.299 was very high as compared to that of [3H]progesterone. Using [3H]progesterone, the relative binding affinity (RBA) of progesterone was more than that of ZK 98.299, whereas using [3H]ZK-98.299 the RBA of ZK 98.299 exceeded that of progesterone. Treatment of myometrial cytosol with increasing concentrations of -SH-modifying agents (iodoacetamide (IA) 0-10 mM or N-ethylmaleimide (NEM) 0-1000 nM) decreased the binding of progesterone by over 80%, whereas similar treatment did not have appreciable effect on the binding of [3 H]ZK-98.299. Although both preformed ligand-receptor complexes were relatively stable in the presence of IA and NEM, the [3H]progesterone-receptor complex was more sensitive as compared to the [3H]ZK-98.299-receptor complex. The addition of 20 mM molybdate in the cytosol had a protective effect against the -SH-modifying agents. [3H]ZK-98.299 and [3H]progesterone-receptor complexes also showed differential stability when incubated at elevated temperatures (25°C and 37°C), [3H]ZK-98.299-binding sites being more thermolabile as compared to [3H]progesterone binding sites. Prior occupation of the receptor by the two ligands gave the complexes the ability to resist an elevated temperature of 25°C. Moreover, molybdate stabilized both the iiganded and unoccupied receptors at 25°C. When the ligand-receptor complexes were applied onto a prefocused polyaerylamide gel, the progesterone and ZK-98.299-receptor complexes were resolved and focused at pH 7.2 and 8.4, respectively. The results of this study suggest that although progesterone and ZK-98.299 are mutually competitive for binding to progesterone receptor, ZK-98.299 also has distinct binding sites.

Introduction

ZK-98.299 [1], like R U 486 [2] and ZK-98.734 [3], is a progesterone antagonist. Trea tment with ZK-98.299 during the mid-luteal phase results in premature induction of menstruat ion [1], and t reatment during early pregnancy results in the expulsion of conceptus in bonnet monkeys [4]. The antiprogestational action of ZK-98.299 is mediated at the level of receptor binding [5,6]. Two 'models ' have been proposed to explain the antiprogestational action of antiprogestins [7,8]. The first 'model ' suggests that the binding of an agonist to the receptor triggers a conformational change which is associated with the release of hsp-90 subunit from the heteroligomer, 8S receptor, thereby unmasking the

Correspondence to: C.P. Puri, Institute for Research in Reproduc- tion (ICMR), Parel, Bombay-400012, India.

DNA-binding site on the 'activated' 4S receptor. In contrast, when the antihormone (RU 486) binds to the receptor, such a complex is stabilised and hsp 90 is not released, thereby preventing D N A binding. The second 'model ' proposes that binding of the ant ihormone to the receptor is capable of provoking transformation with dissociation of the hsp 90 from the 4S subunit. However, the latter does not undergo the necessary conformational change required for optimal DNA- binding and induction of gene transcription. In our earlier studies in the human myometrial cytosol [6] we have seen that ZK-98.299 has high binding capacity as compared to progesterone, thereby suggesting that ZK- 98.299 may have a greater number of binding sites than could be accounted for by the progesterone receptor alone. Moreover, our earlier studies [5,6] have demonstrated that in contrast to that of progesterone, binding of ZK-98.299 was not saturable, at least up to a concentration of 20 nM of the ligand, and had high non-

74

specific binding, thereby suggesting the involvement of hydrophobic interactions in ZK-98.299-receptor binding. In addition, studies on the mapping of progesterone receptor have revealed the existence of a large hydrophobic pocket in addition to the main steroid binding site [9]. This large pocket can accommodate C-11 and C-17 substituents of ZK-98.299 and related antiprogestins. These observations indicate that these antiprogestins in addition to binding to the progesterone receptor may also have distinct binding sites and that their antiprogestational action could be mediated by a mechanism involving these sites.

Therefore, studies were undertaken to investigate whether progesterone and ZK-98.299 show distinct binding sites and how they differ in their binding characteristics. A role for -SH groups has been impli- cated in the structural and functional aspects of steroid receptors [10,11]. It has been demonstrated that -SH groups are involved in steroid receptor interactions and that these may be different from those involved in DNA-binding. Moreover, it has also been suggested that molybdate, a steroid receptor stabilizer [12], may interact with the -SH groups of cysteine [13]. In order to understand whether these cysteine groups are also essential in the interaction of ZK-98.299 with the receptor, we sought to modulate the -SH groups in the receptor using -SH modifying agents and study the effects of these agents in the presence and absence of molybdate on the binding characteristics of ZK-98.299 and progesterone. Further, to understand the structural requirements for binding of the two ligands, the ZK-98.299 and progesterone-receptor complexes were subjected to iso-electric focusing and to elevated temperatures, the former to study any charge differences and the latter to study the effects of elevated temperature on the stability of the receptor before and after complex formation.

Materials and Methods

Steroids and reagents ZK-98.299 and [6,7-3H]ZK 98.299 (69 Ci/mM) were

received from Schering Ag, Berlin. [1,2,6,7-3H]proges - terone (96 Ci/mM) was purchased from Amersham International, UK. Ampholine (pH 3.5-10) used for iso-electric focusing was purchased from LKB, Sweden. All other biochemicals and reagents, including progesterone and cortisol were purchased from Sigma Chemi- cals, USA.

Human tissue Human uterus was obtained from women undergo-

ing vaginal hysterectomy for prolapse of uterus. The women were in the age group of 32-38 years and their menstrual cycles were of normal (25-31 days) duration. The surgery was performed during the mid-luteal

phase. The presence of freshly developed corpus lu- teum confirmed the secretory phase of the cycle. The uterine tissue was collected on ice and immediately brought to the cold room (0-4°C) where it was further processed. The myometrium was washed in saline and cut into small pieces. The tissue was again washed with saline, followed by buffer, blotted dry and weighed.

Buffers Essentially, two buffers were used: buffer TEG;

Tris-HCI (10 mM, pH 7.4) containing 1.0 mM EDTA and 10% glycerol and Buffer TEGM; buffer TEG containing 20 mM sodium molybdate.

Preparation of cytosol The tissue was divided into two equal portions, one

half was homogenized in buffer TEG and the other in buffer TEGM (1:5, w/v) using a Polytron homoge- nizer (Kinematica GmbH, Switzerland). These cytosolic preparations have been designated as cytosol (TEG) or cytosol (TEGM) under various experiments. The homogenate was centrifuged at 800 x g for 10 rain at 0-4°C using 65 Ti rotor on a Beckman Ultracentrifuge (Model L8-70M). The nuclear pellet was discarded and the supernatant centrifuged at 105 000 x g for 60 rain. The cytosol was stored at -70°C in several aliquots till further use. An aliquot was kept aside for protein estimation.

Stripping of endogenous steroids Cytosol was treated with Dextran-coated charcoal

(DCC, containing 1% charcoal and 0.1% Dextran in buffer TEG or TEGM) in a 1:1 ratio to remove endogenous steroids prior to its use in various experiments. The tubes were vortexed and allowed to stand for 30 min at 0-4°C with vortexing every 10 min. The samples were centrifuged at 6000 Xg for 10 min on Sorvall RC-5B. Supernatants from all the tubes were pooled and used for further studies.

The cytosol samples were incubated with 1 /~M cortisol for 30 min at 0-4°C in order to saturate all the corticosteroid binding globulins prior to its use for various experiments [14].

Formation of ligand-receptor complex Cytosol was incubated for 3 h at 0-4°C with 10 nM

[3H]progesterone or [3H]ZK-98.299. Non-specific binding (NSB) was determined by simultaneously incubat- ing cytosol with a 100-fold molar excess of radioinert ligand.

Relative binding affinity Cytosol (TEG) samples were incubated with 10 nM

[3H]ZK-98.299 and increasing concentrations of ZK- 98.299 or progesterone (0-1000 nM) at 0-4°C for 3 h. Similarly, samples were also incubated with 10 nM

[3H]progesterone. The bound fraction was separated and counted for tritium content.

Effect of-SH-modifying agents Cytosol (TEG) was incubated with increasing con-

centrations of IA (0-10 mM) or NEM (0-1000 nM) for 30 min at 0-4°C and the mixture incubated subsequently for the formation of the ligand-receptor complex. In an another experiment, cytosol (TEGM) was incubated with increasing concentrations of IA (0-10 mM) for 30 min at 0-4°C and the mixture incubated subsequently for ligand-receptor complex formation. The cytosol (TEG) samples were also used to form the ligand-receptor complex (liganded receptor) prior to incubation with increasing concentrations of IA (0-10 mM) and NEM (0-1000 nM) for 30 min at 0-4°C.

Effect of temperature on the unoccupied and liganded receptor

Cytosol (TEG) and (TEGM) samples were incubated with [3H] ligands for 3 h at 0-4°C. These ligand- receptor complexes were then further incubated for 10, 20, 30 and 60 min at 25°C or 37°C. In another experiment cytosol (TEG and TEGM) samples were first kept at 25°C or 37°C for 10, 20, 30 and 60 min before subjecting them to ligand-receptor complex formation at 0-4°C. Simultaneous incubations were also carried out at 0-4°C which served as controls.

Iso-electric focusing The method of Vollmer et al. [15] was essentially

followed, except for the modification that 1.5% of ampholine (pH 3.5-10) was used. The gel was prefocused for 30 min at 1200 V, 15 mA and 15 W, corre- sponding to 500 V h. Samples (25/xl of bound ligand-

75

receptor complexes) were loaded onto the gel at three pH positions (acidic, neutral and basic) using paper applicators. Ferritin which focuses as a yellow band around pH 4.5 was used as a visual marker. Focusing was then continued on the preset power conditions up to 3000 V h. After focusing was complete, the gel was cut into 5-mm pieces, put into scintillation fluid vials, vortexed thoroughly and counted for tritium content. Part of the gel was checked immediately for pH meas- urements using a Orion surface electrode.

Estimation of protein Protein was estimated by the method of Lowry et

al., using bovine serum albumin as the standard [16].

Data analysis Specific binding was calculated by subtracting the

binding of the radiolabelled ligand observed in the presence of 1 /~M of radioinert ligand (NSB) from the total binding observed in the absence of radioinert ligand. In the elevated temperature experiments, the binding at 0-4°C at various time points was taken as 100% binding. In the iso-electric focusing experiments the radioactivity in the peak fraction was calculated as a percentage of the radioactivity obtained in all the fractions of the gel.

Results

Relative binding affinity [3H]progesterone showed binding in the myometrial

cytosol which could be displaced by both progesterone as well as ZK-98.299 in a dose-dependent manner. However, progesterone had higher RBA as compared to ZK-98.299. Moreover, the binding of [3H]proges-

® [5 HI PROGESTERONE

lOO-

so-

60- o

-

4O..

0 , | ! - w . •

t-9 7-e 31"Z 125 600

0 P R O G E S T E R O N E • Z K 9 8 . s ' 9 9

[SH] ZK " " S ,

,- ¢ , , ~ , t-S 8 3,1.2 tZS 5 0 2000

l l l l L I | A N I )

Fig. 1. Relative binding affinity of ligands for progesterone receptor. Myometrial cytosol (TEG) samples were incubated with 10 nM [3H]progesterone (A) or [3H]ZK-98.299 (B) and increasing concentrations of progesterone or ZK-98.299 (0-1000 nM) at 0-4°C for 3 h. The

concentration of ligand which displaced B / B 0 to 50% was used to calculate RBA.

76

terone ( B / B 0) could be displaced by over 80% with radioinert progesterone or ZK-98.299 (Fig. 1A).

Similarly, [3H]ZK-98.299 also showed binding which could be displaced by both ZK-98.299 and progesterone. However, the binding could not be displaced by more than 50% even when higher concentrations of the ligands were used. Moreover, RBA of ZK-98.299 was more than that of progesterone (Fig. 1B).

Effect of-SH-modifying agents Fig. 2 illustrates the effect of IA on the unoccupied

and liganded receptor in the human uterine cytosol. The incubation of cytosol with 0-10 mM IA, prior to the formation of the ligand-receptor complexes (unoccupied receptor), reduced the binding of [3H]progesterone to the receptor by 80%. On the other hand, similar t reatment with IA reduced the binding of [3H]ZK-98.299 by only 23%.

When the preformed ligand-receptor complexes (occupied receptor) were treated with IA, no significant effect on the binding of either [3H]ZK-98.299 and [3H]progesterone was observed.

The data in Fig. 3 illustrates the effect of IA (0-10 mM) on the binding of [3H]progesterone to the receptor in the presence and absence of molybdate. The incubation of cytosol with 10 mM IA, prior to the

• I TREATMENT AFTER COMPLEX FORMATION

O r- ] TREATMENT BEFORE COMPLEX FORMATION

O • 1"3H'1 PROGESTERONE

n. , [S . ]ZK 9 s . = .

t00

t9

Z

o u. 5©

bJ O-

O ! ~ • r i = i 0 0 " 6 2 2"5 l(J'O

I O D O A C E ' I r A M I D E mM

Fig. 2. Effect of IA on the unoccupied and liganded receptor in human uterine myometrial cytosol (TEG). In the first experiment cytosol samples were incubated with varying concentrations of IA (0-10 mM) at 0-4°C for 30 min and then subjected to complex formation with 10 nM [3H]progesterone or [3H]ZK-98.299 for 3 h at 0-4°C. In the other experiment, the cytosolic ligand-receptor complexes were formed prior to the treatment with IA. NSB was determined by running parallel assays with 1 ~M of respective radioinert

ligand.

4OO-

(9 Z

O 50

hJ a. (/)

• CYTOSOL WITH MOLYBDATE

0 CYTOSOL WITHOUT MOLYBDATE

0 , # , , o o . 6 z z ~ 4 ~ . o

. I O D O A C E T A M I D E I M

Fig. 3. Effect of IA on the binding of [3H]progesterone to human myometrial cytosol in the presence or absence of molybdate. Cytosol samples, prepared in Tris-HCI buffer with or without molybdate, were incubated with increasing concentrations of IA (0-10 mM) at 0-4°C for 30 min following which 10 nM [3H]progesterone was added and incubation continued for another 3 h. NSB was determined by running parallel assay with 100-fold molar excess of radio-

inert progesterone.

formation of progesterone-receptor complexes, reduced the binding of [3H]progesterone by about 80%. The presence of molybdate in the cytosol, on the other hand, diminished the inhibitory effects of IA on progesterone receptor binding.

Treatment of unoccupied or liganded receptor with NEM had similar effect on the receptor binding as was observed with IA (Fig. 4). However, NEM abolished the binding of unoccupied progesterone receptor at a much lower concentration (0-1000 nM) as compared to IA.

Iso-electric focusing of the receptor-ligand complex Fig. 5 illustrates that when the ligand-bound recep-

tor complexes were applied at a basic pH (pH 8.9), progesterone-bound receptor focused as a thin band at a pH of 7.2 and ZK-98.299-receptor complex focused at a pH of 8.4. Similar results were obtained when the samples were applied at a neutral pH (pH 7.6). How- ever, when the samples were applied at an acidic pH (pH 5.0), no movement was observed in the case o f progesterone-bound receptor complex, though the antiprogestin-bound receptor complex still focused at a pH of 8.4 (result not shown).

Effect of temperature on the unoccupied and liganded receptor

The effects of 25°C and 37°C on the binding of [3H]progesterone and [3H]ZK-98.299 to the human

77

• • TRtrATMtrNT A F T E R COMPLEX FORMATION

o = r.tArmr . s ~ CO.~LEX , o , . , T , o .

o • [~] pRoGtsrt.o.E t~ • [3.] zK , . , 9

'tOO

z

z GI

I I

0 7 '8 3t-2 125 5 0 0

N - E T H Y L M A L E I M I D E a M

Fig. 4. Effect of NEM on unoccupied and liganded receptor in the human uterine myometrial cytosol (TEG). Aliquots of cytosol were incubated with increasing concentrations of NEM (0-1000 nM) for 30 rain at 0-4°C either prior or subsequent to complex formation with the ligands (10 nM [3H]progesterone or [3H]ZK-98.299) for 3 h at 0°C. Parallel incubations with radiolabelled ligands and 1 ~M of respective radioinert ligand were carried out to determine the NSB.

myometrial cytosol, in the absence and presence of molybdate, are shown in Fig. 6 (unoccupied receptor) and Fig. 7 (occupied receptor). The results have been

1 0 0 -

50

I " Z w U 0 e, ;,I ~00. a

0

4

PROGESTERONE

ZK 98"299 APPLICATION SITE

• t 0

' 8

. 6

. 4

- 2

0

"0 ,.p

-,tO

- B

- 6

. 4

- 2

r

L0

5(3

O-

F R A C T I O N S

Fig. 5. IEF of the ligand-receptor complexes. Cytosol samples were incubated with 10 nM [3H]progesterone or [3H]ZK-98.299 at 0-4°C for 3 h. The bound ligand-receptor complexes were separated by using DCC and loaded (25/zl) onto the prefocused polyacrylamide gel at three pH positions, viz. pH 8.9, 7.6, and 5.0. The electrophore- sis was continued until the current dropped to 2 mA. The gel was cut into 5-mm pieces and counted for radioactivity. A portion of the gel

was used to assess the pH gradient using surface electrode.

100-

.

80-

Z s o

Z

W O.

~ 20,

® WITHOUT MOLYBDATE

\ x

I1~ 0 \ '~0-- . . . . . . . . .

" 0 " I ' - - Q _

i i i / i i i

o 3o 60

0 PROGESTERONE • ZK 9 8 , 2 9 9

25"C . . - - 37eC

® WITH MOLYBOATE

- - 0

I i I I i I I

0 30 60

M I N U T E S

Fig. 6. Effect of temperature on the unoccupied [3H]progesterone or [3H]ZK-98.299 binding sites in human myometrial cytosol in the absence (A) and presence (B) of molybdate: Cytosol samples were incubated at either 25°C or 37°C for various lengths of time. Samples were then transferred at 0-4°C, 10 nM of [3H]progesterone or [3H]ZK-98.299 added and incubation continued further for 3 h. Parallel assays were also

done in the presence of 100-fold molar excess of radioinert ligand to account for NSB.

78

~40-

120-

tOO

Z• 3 so Z I

(J

~'- 6o, u W O.

40,

:t

®

WITHOUT MOLYBDATE

i \

~ O. " . .o

,, - o . .

I

, ¥ - - . - - - - ~ - - . - - - ; - - ~ , , O 30 60 0

M I N U T E S

O P R O O E I T E R O N E • ZK 9 8 " 2 9 9

® - - . * c w . v . . O L Y . O , T E - - - S . * C

O

', \, \

I \

\ . . . . .

"o..

30 60

Fig. 7. Effect of temperature on the [3H]progesterone and [3H]ZK-98.299 liganded receptor complexes in human uterine myometrial cytosol, in the absence (A) and presence (B) of molybdate. Cytosol samples were incubated with 10 nM of [3H]progesterone or [3H]ZK-98.299 at 0-4°C for 3 h. Samples were then left at either 25°C or 37°C for various periods of time. Parallel assays were also carried out in the presence of 100-fold

molar excess of radioinert ligands to account for NSB.

expressed with respect to the specific binding (expressed as 100%) observed at 0-4°C in parallel assays.

Incubation of the cytosol in the absence of molybdate, at 25°C and 37°C prior to complex formation with radiolabelled ligands, resulted in decrease in the binding of both [3H]progesterone and [3H]ZK-98.299. This decrease was higher at 37°C as compared to 25°C. However, [3H]ZK-98.299-binding sites were more thermolabile than [3H]progesterone-binding sites (Fig. 6A). The presence of molybdate in the c3rtosol had stabilizing effect on the binding of both the unoccupied receptors at 25°C, whereas the binding at 37°C was still impaired (Fig. 6B).

On the other hand, when the binding sites were occupied with [3H]progesterone or [3H]ZK-98.299 at 0°C, a further incubation at 25°C did not impair the binding of both the ligands either in the absence (Fig. 7A) or presence (Fig. 7B) of molybdate. However, when the occupied receptor was incubated at 37°C the binding of both the ligand-receptor complexes was impaired (Fig. 7).

Discussion

This study suggests that antiprogestin ZK-98.299 and progesterone are mutually competitive for binding to progesterone receptor in the human myometrial cytosol; however, ZK-98.299 appears to also have distinct binding sites. Using [3H]progesterone, the RBA

of progesterone is more than that of ZK-98.299. On the other hand, when [3H]ZK-98.299 is used as the ligand, the RBA of ZK-98.299 exceeds that of progesterone. The binding sites for progesterone are more susceptible to -SH-modifying agents than ZK-98.299. [3H]ZK-98.299 and [3H]progesterone-receptor complexes also show differential stability at elevated temperatures, [3H]ZK-98.299-binding sites being more thermolabile as compared to [3H]progesterone-binding sites. Moreover, the two ligand-receptor complexes focus at different iso-electric points when applied on prefocused polyacrylamide gel. These observations, therefore, suggest that ZK-98.299 apart from binding to the cytosolic progesterone receptor, has its own separate and distinct binding sites and that the an- tagonistic properties of ZK-98.299 may be manifested by a mechanism involving these sites.

The ability of both progesterone and ZK-98.299 to completely inhibit the binding of [3H]progesterone in the myometrial cytosol and the parallel nature of the inhibition curves suggest that ZK-98.299 binds to the progesterone receptor. [3H]ZK-98.299 also shows specific binding which can only be displaced partially with the two ligands. Moreover, using [3H]ZK-98.299, the RBA of ZK-98.299 for the receptor is higher relative to that of progesterone. The change in RBA of ZK-98.299 in relation to that of progesterone, depending upon which radiolabelled ligand is used, suggests that ZK- 98.299 binds to two separate binding components. Had

progesterone and ZK-98.299 been binding to only one class of binding sites, the RBA of one of the two ligands would have been more than that of the other, irrespective of the radiolabelled ligand used. This is supported by the observation that the RBA of the synthetic progestin R5020 for progesterone receptor is more than that of progesterone irrespective of whether [3H]progesterone or [3H]R5020 is used as the ligand [17]. Moreover, our earlier studies have shown that the human myometrial cytosol contains significantly more ZK-98.299 binding sites than could be accounted for by progesterone receptor alone [6]. However, these binding sites could not be resolved by using either double labelling of the cytosol or by blocking the receptor with one of the two ligands. Similarly, the antiestrogen tamoxifen which inhibits the binding of tritiated estra- diol to cytosol from estrogen target tissues also has binding sites distinct from the estrogen receptor [18,19]. These binding sites are also difficult to resolve in vitro because of their relative concentrations and affinities unless the target organ contains very high titers of the receptor, such as in human mammary carcinoma [20].

The structural requirements for binding to the ZK- 98.299-binding sites appear to be different from those involved in the binding of progesterone to progesterone receptor. Progesterone binding requires -SH groups, unlike the case of ZK-98.299 binding. Addition of IA and NEM to the cytosol inhibits the binding of progesterone to progesterone receptor whereas binding of ZK-98.299 is not affected significantly, NEM was more effective as compared to IA because it shows high selectivity for reaction with more exposed protein -SH groups. This study also confirms an earlier report on the presence of free -SH groups in the progesterone receptor and their involvement in ligand binding [21]. Our observation that -SH modulating agents do not affect preformed progesterone-receptor complexes and that molybdate is able to prevent the inhibitory effects of IA on progesterone binding to the receptor suggest that the occupied ligand sterically shields the -SH groups against these agents and that these groups may also be interacting with molybdate. This finding therefore suggests that there is a difference in the amino-acid structure at the steroid-binding site for the two ligands, at least as far as the composition of the cysteine residues is concerned. It is quite possible that in our studies we were measuring ZK-98.299 binding to glucocorticoid receptors since ZK-98.299 possesses weak antiglucocorticoid activity [22], however, -SH-modifying agents render the glucocorticoid receptor, like that of progesterone receptor [23], to a state that is unable to bind glucocorticoids [24]. Moreover, in our studies cytosol preparations were mixed with 1 tzM cortisol prior to addition of labelled steroids which saturates all the glucocorticoid binding sites. ZK-98.299, like RU 486, competes with [3H]testosterone for binding to andro-

79

gen receptors in the human myometrial cytosol, however, the binding of [3H]testosterone (on nM basis) is less than 10% that of [3H]progesterone (unpublished observations). Moreover, the relative binding affinity of ZK-98.299 for binding to testosterone receptor is only 22% as compared to that of testosterone. Thus it is clear that ZK-98.299 is not binding to glucocorticoid receptors and has very low affinity for androgen receptors in our experiments. That ZK-98.299 has distinct binding sites is also reflected by the observation that the binding sites for ZK-98.299 are more thermolabile as compared to those of progesterone. This difference in the dissociation pattern observed for ZK-98.299-receptor complex in the presence and absence of molybdate at 25°C has also been observed in the case of RU-486-receptor complex and has been attributed to the conversion of the complex to a less stable form at elevated temperature. It has been further suggested that molybdate has a stabilizing effect on this complex under identical conditions [25].

The differential behaviour of the cytosolic receptor when complexed with either progesterone or ZK-98.299 is further confirmed by iso-electric focusing wherein the ZK-98.299-receptor and progesterone-receptor complexes focus at a different pI. This observation demonstrates the differences in the net charge of the two types of binding sites, suggesting a differential amino acid composition at least at the sites where ZK-98.299 and progesterone bind. Moreover, progesterone-bound receptor complex does not move when applied at an acidic pH, whereas antiprogestin-receptor complex still focuses at a pH of 8.4. Similar observations have also been made in the case of human breast cancer cytosol wherein progesterone-receptor complex becomes dissociated or destroyed in the acidic pH near the anode [26]. The iso-electric point of the progesterone-receptor complex in our study is quite in agreement with observations made in human breast cancer cytosol [27].

Although the binding of progesterone can be displaced both by progesterone and ZK-98.299, the binding characteristics of [3H]ZK-98.299 appear to be distinct from those of [3H]progesterone. However, at pre- sent it is difficult to resolve whether progesterone and ZK-98.299 bind to different receptor proteins or the ligands bind to two separate sites on the same receptor. Experiments carried out by blocking the receptor sites with excess radioinert progesterone did not result in any increased binding with either radiolabelled lig- and, thereby suggesting that at any given time only one site is available for interaction with a ligand. The binding of one ligand results in inhibition of subsequent binding by the second steroid. Recent studies on the mapping of progesterone receptors have revealed the existence of a large hydrophobic pocket in addition to the main steroid-binding sites. This large pocket can

80

accommodate C-11 and C-17 substituents of ZK-98.299 and related antiprogestins [9]. Thus it is also quite possible that once the two ligands bind to the receptor sites they undergo differential alterations due to different configurational requirements. Studies on the rat liver glucocorticoid receptor [28] have shown that the kinetics of steroid binding are dependent upon whether the receptor site is occupied by the agonist triamci- nolone acetonide or its antagonist RU 486. Taken together, these results suggest that distinct binding sites for antiprogestin ZK-98.299 may be of physio- logical relevance, however, the role of these sites in the mechanism of antiprogestin action warrants further investigation. Further studies to resolve whether or not these sites (for agonist and antagonist) are on the same molecule or on distinct molecules are ongoing in our laboratory.

Acknowledgements

The authors are grateful to the Council for Scien- tific and Industrial Research for providing Senior Re- search Fellowship to one of the authors (A.D.), and to Dr. W. Elger, Schering Ag, Berlin, for providing antiprogestin ZK-98.299. The secretarial assistance of Miss G. Basantani is appreciated.

References

1 Puri, C.P., Elger, W.A.G. and Pongubala, J.M.R. (1987) Contra- ception 35, 409-421.

2 Philibert, D. (1985) in Adrenal Steroid Antagonism (Agarwal, M.K., ed.), pp.77-101, Walter de Gruyter, Berlin.

3 Puri, C.P., Katkam, R.R., D'souza, A., Elger, W.A.G. and Patil, R.K. (1990) Biol. Reprod. 43, 437-444.

4 Purl, C.P., Patil, R.K., Elger, W.A.G. and Pongubala, J.M.R. (1990) Contraception 41, 197-205.

5 Pongubala, J.M.R., Elger, W.A.G. and Puri, C.P.(1987) J. Recep. Res. 7, 903-920.

6 Puri, C.P., Hinduja, I.N., D'souza, A., Elger, W.A.G. and Pongubala, J.M.R. (1989) Biochim. Biophys. Acta. 1011, 176-182.

7 Baulieu, E.E., Ulmann, A., and Philibert, D. (1987) in Fertility Regulation Today and Tommorrow (Diczfaulsy, E. and Bygde- man, M., eds.), pp. 55-73, Raven Press, New York.

8 Groyer, A., Schweizer-Groyer, G., Cadepond, F., Mariller, M. and Baulieu, E.E. (1987) Nature 328, 624-626.

9 Teutsch, G. (1985) in The Antiprogestin Steroid RU 486 and Human Fertility Control (Baulieu, E.E. and Segal, S.J., eds.), pp. 27-47, Plenum Press, New York.

10 Bodwell, J.E., Holbrook, N.J. and Munck, A. (1984) Biochemistry 23, 4237-4242.

11 Smith, L.I., Bodwell, J.E., Mendel, D.B., Gardelli, T., North, W.G. and Munck, A. (1988) Biochemistry. 27, 3747-3753.

12 Ogle, T.F. (1983) J. Biol. Chem. 258, 4982-4988. 13 Housley, P.R., Dahmer, M.K. and Pratt, W.B. (1982) J. Biol.

Chem. 257, 8615-8618. 14 Milgrom, E., Atger, M. and Baulieu, E.E. (1971) in Advances in

Biosciences, Vol. 7, pp. 235, Pergamon Press, Vieweg. 15 Vollmer, G., Helmchen, B. and Knuppen, R. (1989) J. Clin.

Chem. Clin. Biochem. 27, 953-959. 16 Lowry, O.H., Rosebrough, N.J., Farr, N.J. and Randall, R.J.

(1951) J. Biol. Chem. 193, 265-272. 17 Waiters, M.R. and Clark, J.H. (1977) J. Steroid Biochem. 8,

1137-1144. 18 Ruh, M.F., Brzyski, R.G., Strange, L. and Ruh, T.S. (1983)

Endocrinology 112, 2203-2205. 19 Faye, J.C., Fargin, A., Valette, A. and Bayard, F. (1987) Horm.

Res. 28, 202-211. 20 Sutherland, R.L., Murphy, L.C., Foo, M.S., Green, M.D., Why-

bourne, A.M. and Krozowski, I.S. Nature (1980) 288, 273-275. 21 Kalimi, M. and Banerji, A. (1981) J. Steroid Biochem. 14, 593-597. 22 Neef, G., Beier, S., Elger, W.A.G., Henderson, D. and Wiechert,

R. (1984) Steroids 44, 349-402. 23 Moudgil, V.K., Anter, M.J. and Hurd, C. (1989) J. Biol. Chem.

264, 2203-2211. 24 Bodwell, J.E., Holbrook, N.J. and Munck, A. (1984) Biochemistry

23, 1392-1398. 25 Bell, A.P. and Weatherhill, P.J.(1986) J. Steroid Biochem. 25,

473-481. 26 Wrange, O., Humla, S., Ramberg, I., Gustafsson, S.A., Skoog., L.,

Nordenskjold, B. and Gustafsson, J.A. (1981) J. Steroid Biochem. 14, 141-148.

27 Wrange, O. (1983) Breast Cancer Res. Treatmt. 3, 97-102. 28 Agarwal, M.K., Lombardo, G., Eliezer, N. and Moudgil, V.K.

(1985). Biochem. Biophys. Res. Commun. 133, 745-752.

antiprogestin zk-98.299 and progesterone display differential binding characteristics in the human...

Documents