- Email: [email protected]
Androgen receptor binding and antiandrogenic activity of some 4,5-secoandrostanes and ring B cyclopropanoandrostanes
0022-473 1/82/09033 I -~4~3.~/~ Copyright Q 1982 Pergamon Press Ltd
J. srwuid BiO~~~f~l.Vol. 17, pp. 331 to 334, 1982 Printed in Great Britain. All rights reserved
ANDROGEN RECEPTOR BINDING AND ANTIA~~ROGEN~C ACTIVITY OF SOME 4,~-SEC~AN~R~STA~ES AND RING B ~YCLOPR~PANOAN~ROSTANES L. ST~RKA, R.
HAMPL,
A. KASAL
and L.
KOWWT
Research Institute of Endocrinology, 11594 Praha, Institute of Organic Chemistry and Biochemistry of CSAV, 165 10 Praha 6, Czechoslovakia (Receiued 30 July 1981) SUMMARY Eight androstane derivatives with modified ring A or B (4,5-secoandrostanes and ring I3 cyclopropanoandrosta~~~ were assayed in uiuu on mice for their ~t~aadro8en~c activity and the effect was compared with that of cyproterone acetate. The inhibition of dihydrotestosterone binding to rat prostate cytosol and to human plasmatic sex hormone binding protein was correlated with the in uiuo effect. The antiandrogenic activity of ~,7a-cyclopropano-S~-androstane-3~,17~-dioi was nearly as high as that of cyproterone acetate. The opening of ring A of androgens, such as testosterone or 17a-methyltestosteron~ moderately reduced the binding and changed the biological activity to weakly antiandro~eni~.
INTRODUCTION
search for hormonal and antihormonal agents has led to investigation of steroids with modified ring A or B among which several potent antiandrogens have been found, such as cyproterone acetate, 6c(bromo-l7~-hydroxy-l~~-methyl-4-oxa-5~-andr~stan~ 3-one (BOMT), A-norprogesterone and i?cc-methyl-Bnortestosterone [ 11. In connection with other programmes several Aring secosteroids and 5&6@- or 6~,7~-cy~lopropano” androstane derivatives were obtained. It seemed reasonable to screen these compounds for their hormonal or antihormonal potential both by bioassay and by the assessment of their binding properties. Binding studies are frequently used as a useful tool for screening the potential biological activity of steroid hormones and antihormones [I-S]. The
EXPERIMENTAL
All compounds used in this study were crystalline substances of a purity higher than 997;. The methods of preparation and the physiochemical properties of the steroids mentioned here were described recently [t&-9] and are summarized briefly in Table I. [1,24,5,6,7-3H]-5a-dihydrotestosterone, SA 4.81 TBq (130 Ci)/mmol, Radiochemical Centre, Amersham, was purified by paper chromatography in the system cy~lohexane-toluene-methanol-water (9: f : 8 : 2, by vol.).
Prostates from Wistar male rats (15@2OOg body weight), castrated 24 h before sacrifice, were removed, minced and homogenized in Potter-Elvehjem homogenizer in three volumes of Tris-WC1 buffer, pH 7.4 (20 mmol - 1_’ 1, containing EDTA and dithiuthreito~ (1.5 mmof+l- ’ each). The homogenate was centrifuged at 105,~~ for 1 h to obtain the cytosol fraction. Cytosol (2ml) was pre-incubated with [3H]-5adihydrot~tosterone (2.0 x 10~9mol~I-L) at O’C for 30 min to ensure equal concentration of the labei in all parallelly processed samples. Aliquot portions (150$) of cytosol iabelled with [3Hf-dihydrotestosterone were added to the dry residue of tested steroid, mixed and incubated at O’C under shaking for 2 h. The following concentrations were used: for dihydrotestosterone 2.5, 5.0, 10.0 and 20.0 x 10-9mol~l-’ and 2.0 x 10-6mol~l-‘, for synthetic competitors 2.5, 5.0, 10.0 and 20.0 x 10-‘mot.t-‘, respectively. Separation of free and bound figand and the determination of the binding characteristics was carried out by a polyacrylamide gel eiectrophores~s technique, similar to the agar gel electrophoresis method of Krieg et at.[ to], as described in detail elsewhere [El]. Slightly modified experimental conditions were used as follows: 30~1 portions of the incubation mixture were layered on the gel, the preset current was 180 mA and the time of electrophoresis 16 h. The apparent association constant for dihydrotestosterone was 4.82 x lo8 I.mol‘ t, the concentration of dihydrotestosterone binding sites 82 fmol per mg of protein (determined by Lowry’s method [12]). ~i~d~~~ rn sex ~~)rrn~~n~ binning ~~#~uli~
Address for correspondence: Endocrinology akia.
L. StBrka, Res. Inst. for N&rodni 8, 116 94 Praha 1, Czechoslov-
Full term human placental serum precipitated by ammonium sulphate, as described in detail pre331
332
L.
STARKA et (11.
Table 1. Characteristic physicochemical data on steroids used for screening Compound No.
m.p. (‘C)
Optical rotation a;0 (CCIJ
159-161
fll”
3620, 1053 (CHCl,)
C61
-10
3625, 1721, 1714 (ink), 1355
C61
Name
IX. (in Ccl,) (cm-‘)
Reference
1
4-nor-3,5-secoandrostane-3,17p-diol
II
178-hydroxy-17a-methyl-4,5-secoandrostan-3-one
III
17/Ghydroxy-4,S-secoandrostan-3-one
58-60
+15’
3615, 1710, 1359, 1411
C61
IV
17/I-hydroxy-5/?,6/%cyclopropano4,5-secoandrostan-3-one
93-94
-29’
1720, 1715 (infl.), 1357, 3625, 1057, 3075, 3010
c71
V
17b-hydroxy-5/I,6/Lcyclopropano5@-androstan-3-one
124-126
+ 77’
3060, 1741, 1721
VI
VI
6a,7a-cyclopropano-5a-androstane3, I ‘I-dione
172-174
+ 39”
3070, 1742, 1719
c91
VII
17/%hydroxy-6a,7a-cyclopropano5a-androstan-3-one
215-218
-25”
3630, 1071, 1049, 1023
c91
VIII
6a,7a-cyclopropano-5a-androstane3/?,17/?-diol
209-2 11
+ 27”
3630, 3070, 1719, 1712, 1071, 1049, 1023
c91
viously [13], was used to obtain the protein fraction rich in sex hormone binding globulin (SHBG). The relative binding of steroids to SHBG was determined in a system containing constant amounts of SHBG (binding capacity 2.3 pmol), 0.055 pmol [3H]-5a-dihydrotestosterone and increasing concentrations of the competitors (0.1414.0pmol) in total volume of 0.5 ml of 20 mmol.ll ’ Tris-HCI buffer, pH 7.4, with 1.O mmol I- ’ CaCI,. Following incubation at 25°C for 2 h free steroids were separated by charcoal-dextran treatment and the radioactivity was measured in an aliquot portion of supernatant. The amounts of competitors required for 507,; displacement of the radioligand were obtained from the loglogit plot as described previously [ 131. Bioussuy for the untiandroqenic activity The method used for screening the antiandrogenic activity in vivo [4] was simultaneous administration of testosterone propionate and of the assayed compound to castrated male mice. Male strain H mice (Velaz, Praha), weighing approx. 35 g, were fed standard Velaz laboratory diet containing 237:, protein with water ud lib. The animals were kept in a room with indirect light and a controlled temperature of 24 k 2°C. The mice were castrated 21 days before starting the administration of steroids and kept in groups of seven animals each. Castrated mice were given the vehicle (0.2 ml olive oil) only or testosterone propionate or testosterone propionate in combination with the tested compound S.C. (in 0.2 ml) for 3 weeks on alternate days. Each steroid treated mouse received a total of 1.7 mg, i.e. 5.0 gmol testosterone propionate and those receiving the combination an additional 50pmol of the tested
compound. At the end of the experiment the animals were sacrificed by ether narcosis, weighed and the seminal vesicles, kidneys, adrenals and spleen were removed and weighed on a torsion balance. The organ weights were expressed in relative values (mg/lOO g body weight) and the results were evaluated statistically by means of Student’s test. Those compounds which showed a decrease of the seminal vesicle weight significant from the group of mice treated with testosterone propionate alone, were considered to be antiandrogenic. RESULTS
Out of the eight steroids tested seven inhibited the specific binding of the radioactive ligand to the intracellular prostatic androgen receptor. Four of them (compounds II, V, VII and VIII) were also bound to SHBG with relatively high affinity. The data on binding properties of these compounds are listed in Table 2. Very low displacement of dihydrotestosterone from binding to both receptor protein and SHBG was found in experiments with 6a,7a-cyclopropano-5aandrostane-3,17-dione (VI), the only compound lacking the 17fi-hydroxy group. The steroids tested were assayed for antiandrogenic activity by seminal vesicle weight. Body weight increase, increments and organ weights of the mice are given in Table 3. The changes in adrenal, spleen and kidney weights were measured as well, however, no conclusions are derived from them. The reduction of adrenal weight under treatment with cyclopropanoderivatives IV, V and VI is remarkable. However, this action can hardly be ascribed to a corticosteroid-like effect with respect to an increase in spleen weight.
Antiandrogenic Table 2. Competition
of various
synthetic
androstane
derivatives
333
steroids with androgens for specific binding of their biological activity in ciuo
proteins
in uirro and the evaluation
Binding to rat prostate, cytosol receptor Compound No.
Ki (nmol.l-
I II III IV
V VI VII VIII
‘)
y0 Inhibition*
5.55 5.08 3.81 3.00 5.88 11.30 6.27 8.71
33.9 36.1 48.1 61.1 31.2 16.2 29.2 21.0
1.83
100.0
Dihydrotestosterone
Binding to SHBG “/:, Inhibition*
Seminal vesicles weight “/:, Decrease?
2.64 23.10 0.99 8.62 15.60 0.10 78.30 82.20
to dihydrotestosterone. to the group of testosterone
14.9 25.3 19.7 24.1 21.5 11.6 -0.7 29.3
No Antiandrogenic Antiandrogenic Antiandrogenic Antiandrogenic No No Antiandrogenic
34.0
Antiandrogenic
propionate
treated
Biological testing showed that 4-nor-3,5-secoandrostan-3,17/I-diol (I) and cyclopropanoandrostanes VI and VII were devoid of any antiandrogenic activity. Compound V and its secoderivative IV as well as the secoderivative (III) of testosterone were weak antiandrogens. A higher antiandrogenic activity was exhibited by the 4,5-secoderivative (II) of 17~methyltestosterone and by 6a,7a-cyclopropano-Su-androstane-3/$17B-diol (VIII), a compound derived from compound VII by 3/&reduction, which has no antiandrogenic activity itself.
DISCUSSION
Seven of the eight steroids tested showed such an affinity for the androgen receptor that the majority of
animals
(see Table 3).
binding sites were occupied at steroid concentrations in the nanomolar range. All secosteroids exhibited a relatively high affinity to rat prostate cytosol receptors for androgens. The binding of 17/?-hydroxy-Sfi,6@yclopropano-4,5-secoandrostan-3-one (IV) was even higher than that of the parent compound with intact ring A (compound V) probably because the former compound better fulfils the steric requirements for planarity. However, the biological activity was shifted by opening the ring A from androgenic to distinctly antiandrogenic in 4,5+ecoderivatives of testosterone and 17c+methyltestosterone. The biological activity was fully lost in the A-nor-3,5-secosteroid I. Oxidation of the 17fi-hydroxy group to the 17-0~0 group reduced the receptor binding of 6a,7a-cyclopropano-5a-androstane-3,17-dione (VI) and practically
Table 3. Body weight and relative organ weights of castrated mice treated with testosterone propionate 5.0 pmol, during 3 weeks) and with tested steroids (50 pmol during 3 weeks)
Oil TP TP TP TP TP TP TP TP TP TP
+ + + + + + + + +
CPAt I II III IV V VI VII VIII
+2.16 +1.48 -0.38 -0.42 - 1.67 -4.47 + 1.67 +2.46 +4.68 - 1.14 - 1.38
(TP, 1.7 mg, i.e.
(g)
Seminal vesicles (mg/lOO g b.wt)
Adrenals (mg/lOO g b.wt)
Spleen (mg/lOO g b.wt)
Kidney (g/l00 g b.wt)
f 2.40 &- 1.46 + 2.06** k 2.97 + 2.70** + 2.51** k 0.86 + 1.63 + 1.92** k 2.18** + 2.02**
151.3 761.9 507.7 654.5 573.6 617.5 583.6 603.7 680.1 774.8 543.7
35.79 28.46 30.47 38.67 32.37 28.51 20.19 16.85 17.81 34.12 32.28
381.5 385.1 267.5 356.8 336.3 292.5 425.2 525.1 474.7 438.2 386.4
1.27 1.78 1.69 1.65 1.72 1.61 1.57 1.68 1.71 1.72 1.73
Body weight gain Compound
activity
100.0
Cyproterone acetate * Relative t Relative
Biological
* + f + + + f + i + +
53.9** 120.2 149.8** 81.7 157.8** 99.7* 211.3* 206.3* 256.9 134.8 75.2**
k + f + f + + f + f +
5.67** 7.58 5.31 10.24* 8.75 5.48 5.18* 2.65** 2.75** 8.07 8.42
* Difference relative to TP significant at the level P < 0.05 **P < 0.01. t CPA = cyproterone acetate. Mean values If: SD are given (n = 7, for TP, CPA + TP and oil group n = 14).
k + f f + + + + + + +
58.4 108.3 112.0* 181.9 71.6 72.3 70.2 115.7* 226.8 94.8 91.3
k f f f $+ + + + + +
0.06** 0.18 0.17 0.20 0.10 0.20 0.19* 0.25 0.27 0.16 0.15
L. STARKA et cd.
334
fully abolished its binding to SHBG. However, even the presence of 17/%hydroxy group in 4,5-secosteroids I, 111 and IV was not sufficient to achieve a distinct binding to SWBG. Binding of the antiandrogenic 3,17-diol VIII confirms that a 3-0~0 group is not essential for high affinity interaction with the androgen receptor, as has been demonstrated by other authors [5, 14, IS]. The binding properties are only in approximate agreement with the biological activity found in ai~o which, however, couid be influenced by the metabolism of the compounds. Among steroids tested here there were no compounds with surprisingly high antiandrogenic activity. However, in any event the data presented are of interest for the understanding of structure-binding relationship of synthetic steroids.
REFERENCES
Strirka L. and Hampl R.: Antihormonal activities of steroids with modified rings A or B. in Antihormones (Edited by M. K. Agarwal). W. de Gruyter. Berlin (1982). Raynaud J. P. A.: A strategy for the design of potent hormones, Med. Chem. 5 (1977) 451-4.56. Starka L., Sulcova J., Broulik P. D., Joska J., FajkoS J. and DoskoEil M.: Screening for antiandrogenic activity of some 4,5-cyclo-A-homo-B-norandrostane derivatives. J. .steroid Biochem. 8 (1977) 939-941. Starka L., Hampl R., BiEikova M., Cerny V., FajkoS J., Kasal A., KoEovsky P., Kohout L. and V&ova H.: Steroids with modified ring A or B: screening for potential antiandrogenic and synandrogenic activity. J. steroid ~i~)~~~~z.13 (1980)455460.
5. Rousseau G. C., Quivy J. 1. and Crabbe P.: Interaction of A-nor, A-l 9-dinor, and A-homo-Sa-androstane derivatives with the androgen receptor and the epididyma1 androgen binding protein. Steroids 37 (19Sl) 383-392. 6. Kasal A.: Anaiogues of androgens with the open ring A. Coil. czecft. Chem. Commun. 45 (1980) 2541-2549. 7. Kasai A. and Kohout L.: 4,56ecoanalogues of dihydrotestosterone with a cyclopropane ring in 5,6-position. Cull. czech. Gem. Commlm. in press. g. Kohout L. and FajkoS J.: The epimeric 5,7-cycle-Bhomoandrostane derivatives. Co/I. cc&. Gem. Commun. 38 (1973) 1415-1427. 9. Kohout L. and FajkoS J.: Anaiogues of androgens with a cyclopropane ring in a 6a,?ol-position. Co/l. Czech. Chem. C~~~rnu~l. 45 (1980) 1974.-1981. 10, Krieg M., Steins P., Szalay R. and Voigt K. D.: Characterization of specific androgen receptor in rat prostate cytosol by agar gel eiectrophoresis. in aivo and in vitro studies. J. steroid Biochem. 5 (1974) 87-92. 11. Hampl R., Horakova J., BiCikova M., Dvoiak P. and Starka L.: Iododerivatives of testosterone as potential biological markers. J. steroid &o&em. 13 (1980) 1035TrO38 12. Lowry 0. H., Rosebrough N. J.. Farr A. F. and Randall R. J.: Protein measurement with the Folin phenol reagent. J. hic)i. Chem. 193 (1951) 265-271. t3. Bicikova M., Hampl R. and Starka L.: Binding of synthetic anabolic steroids to testosterone-estradiol binding globulin and to rat prostate cytosol. Endocr. exper. 1I (l977) 85-90. 14 Cunningham G. R., Tindall D. J. and Means A. R.: Difference in steroid specificity for rat androgen binding protein and the cytoplasmic protein. Stewids 33 (1979) 261-276. 15 Kirchhoff J., Soffie M. and Rousseau Ci. G.: Differences in the steroid-binding site specificities of rat prostate androgen receptor and epididymal androgenbinding protein (ABP). J. steroid B&hem. 10 (1979) 487-497.
J. srwuid BiO~~~f~l.Vol. 17, pp. 331 to 334, 1982 Printed in Great Britain. All rights reserved
ANDROGEN RECEPTOR BINDING AND ANTIA~~ROGEN~C ACTIVITY OF SOME 4,~-SEC~AN~R~STA~ES AND RING B ~YCLOPR~PANOAN~ROSTANES L. ST~RKA, R.
HAMPL,
A. KASAL
and L.
KOWWT
Research Institute of Endocrinology, 11594 Praha, Institute of Organic Chemistry and Biochemistry of CSAV, 165 10 Praha 6, Czechoslovakia (Receiued 30 July 1981) SUMMARY Eight androstane derivatives with modified ring A or B (4,5-secoandrostanes and ring I3 cyclopropanoandrosta~~~ were assayed in uiuu on mice for their ~t~aadro8en~c activity and the effect was compared with that of cyproterone acetate. The inhibition of dihydrotestosterone binding to rat prostate cytosol and to human plasmatic sex hormone binding protein was correlated with the in uiuo effect. The antiandrogenic activity of ~,7a-cyclopropano-S~-androstane-3~,17~-dioi was nearly as high as that of cyproterone acetate. The opening of ring A of androgens, such as testosterone or 17a-methyltestosteron~ moderately reduced the binding and changed the biological activity to weakly antiandro~eni~.
INTRODUCTION
search for hormonal and antihormonal agents has led to investigation of steroids with modified ring A or B among which several potent antiandrogens have been found, such as cyproterone acetate, 6c(bromo-l7~-hydroxy-l~~-methyl-4-oxa-5~-andr~stan~ 3-one (BOMT), A-norprogesterone and i?cc-methyl-Bnortestosterone [ 11. In connection with other programmes several Aring secosteroids and 5&6@- or 6~,7~-cy~lopropano” androstane derivatives were obtained. It seemed reasonable to screen these compounds for their hormonal or antihormonal potential both by bioassay and by the assessment of their binding properties. Binding studies are frequently used as a useful tool for screening the potential biological activity of steroid hormones and antihormones [I-S]. The
EXPERIMENTAL
All compounds used in this study were crystalline substances of a purity higher than 997;. The methods of preparation and the physiochemical properties of the steroids mentioned here were described recently [t&-9] and are summarized briefly in Table I. [1,24,5,6,7-3H]-5a-dihydrotestosterone, SA 4.81 TBq (130 Ci)/mmol, Radiochemical Centre, Amersham, was purified by paper chromatography in the system cy~lohexane-toluene-methanol-water (9: f : 8 : 2, by vol.).
Prostates from Wistar male rats (15@2OOg body weight), castrated 24 h before sacrifice, were removed, minced and homogenized in Potter-Elvehjem homogenizer in three volumes of Tris-WC1 buffer, pH 7.4 (20 mmol - 1_’ 1, containing EDTA and dithiuthreito~ (1.5 mmof+l- ’ each). The homogenate was centrifuged at 105,~~ for 1 h to obtain the cytosol fraction. Cytosol (2ml) was pre-incubated with [3H]-5adihydrot~tosterone (2.0 x 10~9mol~I-L) at O’C for 30 min to ensure equal concentration of the labei in all parallelly processed samples. Aliquot portions (150$) of cytosol iabelled with [3Hf-dihydrotestosterone were added to the dry residue of tested steroid, mixed and incubated at O’C under shaking for 2 h. The following concentrations were used: for dihydrotestosterone 2.5, 5.0, 10.0 and 20.0 x 10-9mol~l-’ and 2.0 x 10-6mol~l-‘, for synthetic competitors 2.5, 5.0, 10.0 and 20.0 x 10-‘mot.t-‘, respectively. Separation of free and bound figand and the determination of the binding characteristics was carried out by a polyacrylamide gel eiectrophores~s technique, similar to the agar gel electrophoresis method of Krieg et at.[ to], as described in detail elsewhere [El]. Slightly modified experimental conditions were used as follows: 30~1 portions of the incubation mixture were layered on the gel, the preset current was 180 mA and the time of electrophoresis 16 h. The apparent association constant for dihydrotestosterone was 4.82 x lo8 I.mol‘ t, the concentration of dihydrotestosterone binding sites 82 fmol per mg of protein (determined by Lowry’s method [12]). ~i~d~~~ rn sex ~~)rrn~~n~ binning ~~#~uli~
Address for correspondence: Endocrinology akia.
L. StBrka, Res. Inst. for N&rodni 8, 116 94 Praha 1, Czechoslov-
Full term human placental serum precipitated by ammonium sulphate, as described in detail pre331
332
L.
STARKA et (11.
Table 1. Characteristic physicochemical data on steroids used for screening Compound No.
m.p. (‘C)
Optical rotation a;0 (CCIJ
159-161
fll”
3620, 1053 (CHCl,)
C61
-10
3625, 1721, 1714 (ink), 1355
C61
Name
IX. (in Ccl,) (cm-‘)
Reference
1
4-nor-3,5-secoandrostane-3,17p-diol
II
178-hydroxy-17a-methyl-4,5-secoandrostan-3-one
III
17/Ghydroxy-4,S-secoandrostan-3-one
58-60
+15’
3615, 1710, 1359, 1411
C61
IV
17/I-hydroxy-5/?,6/%cyclopropano4,5-secoandrostan-3-one
93-94
-29’
1720, 1715 (infl.), 1357, 3625, 1057, 3075, 3010
c71
V
17b-hydroxy-5/I,6/Lcyclopropano5@-androstan-3-one
124-126
+ 77’
3060, 1741, 1721
VI
VI
6a,7a-cyclopropano-5a-androstane3, I ‘I-dione
172-174
+ 39”
3070, 1742, 1719
c91
VII
17/%hydroxy-6a,7a-cyclopropano5a-androstan-3-one
215-218
-25”
3630, 1071, 1049, 1023
c91
VIII
6a,7a-cyclopropano-5a-androstane3/?,17/?-diol
209-2 11
+ 27”
3630, 3070, 1719, 1712, 1071, 1049, 1023
c91
viously [13], was used to obtain the protein fraction rich in sex hormone binding globulin (SHBG). The relative binding of steroids to SHBG was determined in a system containing constant amounts of SHBG (binding capacity 2.3 pmol), 0.055 pmol [3H]-5a-dihydrotestosterone and increasing concentrations of the competitors (0.1414.0pmol) in total volume of 0.5 ml of 20 mmol.ll ’ Tris-HCI buffer, pH 7.4, with 1.O mmol I- ’ CaCI,. Following incubation at 25°C for 2 h free steroids were separated by charcoal-dextran treatment and the radioactivity was measured in an aliquot portion of supernatant. The amounts of competitors required for 507,; displacement of the radioligand were obtained from the loglogit plot as described previously [ 131. Bioussuy for the untiandroqenic activity The method used for screening the antiandrogenic activity in vivo [4] was simultaneous administration of testosterone propionate and of the assayed compound to castrated male mice. Male strain H mice (Velaz, Praha), weighing approx. 35 g, were fed standard Velaz laboratory diet containing 237:, protein with water ud lib. The animals were kept in a room with indirect light and a controlled temperature of 24 k 2°C. The mice were castrated 21 days before starting the administration of steroids and kept in groups of seven animals each. Castrated mice were given the vehicle (0.2 ml olive oil) only or testosterone propionate or testosterone propionate in combination with the tested compound S.C. (in 0.2 ml) for 3 weeks on alternate days. Each steroid treated mouse received a total of 1.7 mg, i.e. 5.0 gmol testosterone propionate and those receiving the combination an additional 50pmol of the tested
compound. At the end of the experiment the animals were sacrificed by ether narcosis, weighed and the seminal vesicles, kidneys, adrenals and spleen were removed and weighed on a torsion balance. The organ weights were expressed in relative values (mg/lOO g body weight) and the results were evaluated statistically by means of Student’s test. Those compounds which showed a decrease of the seminal vesicle weight significant from the group of mice treated with testosterone propionate alone, were considered to be antiandrogenic. RESULTS
Out of the eight steroids tested seven inhibited the specific binding of the radioactive ligand to the intracellular prostatic androgen receptor. Four of them (compounds II, V, VII and VIII) were also bound to SHBG with relatively high affinity. The data on binding properties of these compounds are listed in Table 2. Very low displacement of dihydrotestosterone from binding to both receptor protein and SHBG was found in experiments with 6a,7a-cyclopropano-5aandrostane-3,17-dione (VI), the only compound lacking the 17fi-hydroxy group. The steroids tested were assayed for antiandrogenic activity by seminal vesicle weight. Body weight increase, increments and organ weights of the mice are given in Table 3. The changes in adrenal, spleen and kidney weights were measured as well, however, no conclusions are derived from them. The reduction of adrenal weight under treatment with cyclopropanoderivatives IV, V and VI is remarkable. However, this action can hardly be ascribed to a corticosteroid-like effect with respect to an increase in spleen weight.
Antiandrogenic Table 2. Competition
of various
synthetic
androstane
derivatives
333
steroids with androgens for specific binding of their biological activity in ciuo
proteins
in uirro and the evaluation
Binding to rat prostate, cytosol receptor Compound No.
Ki (nmol.l-
I II III IV
V VI VII VIII
‘)
y0 Inhibition*
5.55 5.08 3.81 3.00 5.88 11.30 6.27 8.71
33.9 36.1 48.1 61.1 31.2 16.2 29.2 21.0
1.83
100.0
Dihydrotestosterone
Binding to SHBG “/:, Inhibition*
Seminal vesicles weight “/:, Decrease?
2.64 23.10 0.99 8.62 15.60 0.10 78.30 82.20
to dihydrotestosterone. to the group of testosterone
14.9 25.3 19.7 24.1 21.5 11.6 -0.7 29.3
No Antiandrogenic Antiandrogenic Antiandrogenic Antiandrogenic No No Antiandrogenic
34.0
Antiandrogenic
propionate
treated
Biological testing showed that 4-nor-3,5-secoandrostan-3,17/I-diol (I) and cyclopropanoandrostanes VI and VII were devoid of any antiandrogenic activity. Compound V and its secoderivative IV as well as the secoderivative (III) of testosterone were weak antiandrogens. A higher antiandrogenic activity was exhibited by the 4,5-secoderivative (II) of 17~methyltestosterone and by 6a,7a-cyclopropano-Su-androstane-3/$17B-diol (VIII), a compound derived from compound VII by 3/&reduction, which has no antiandrogenic activity itself.
DISCUSSION
Seven of the eight steroids tested showed such an affinity for the androgen receptor that the majority of
animals
(see Table 3).
binding sites were occupied at steroid concentrations in the nanomolar range. All secosteroids exhibited a relatively high affinity to rat prostate cytosol receptors for androgens. The binding of 17/?-hydroxy-Sfi,6@yclopropano-4,5-secoandrostan-3-one (IV) was even higher than that of the parent compound with intact ring A (compound V) probably because the former compound better fulfils the steric requirements for planarity. However, the biological activity was shifted by opening the ring A from androgenic to distinctly antiandrogenic in 4,5+ecoderivatives of testosterone and 17c+methyltestosterone. The biological activity was fully lost in the A-nor-3,5-secosteroid I. Oxidation of the 17fi-hydroxy group to the 17-0~0 group reduced the receptor binding of 6a,7a-cyclopropano-5a-androstane-3,17-dione (VI) and practically
Table 3. Body weight and relative organ weights of castrated mice treated with testosterone propionate 5.0 pmol, during 3 weeks) and with tested steroids (50 pmol during 3 weeks)
Oil TP TP TP TP TP TP TP TP TP TP
+ + + + + + + + +
CPAt I II III IV V VI VII VIII
+2.16 +1.48 -0.38 -0.42 - 1.67 -4.47 + 1.67 +2.46 +4.68 - 1.14 - 1.38
(TP, 1.7 mg, i.e.
(g)
Seminal vesicles (mg/lOO g b.wt)
Adrenals (mg/lOO g b.wt)
Spleen (mg/lOO g b.wt)
Kidney (g/l00 g b.wt)
f 2.40 &- 1.46 + 2.06** k 2.97 + 2.70** + 2.51** k 0.86 + 1.63 + 1.92** k 2.18** + 2.02**
151.3 761.9 507.7 654.5 573.6 617.5 583.6 603.7 680.1 774.8 543.7
35.79 28.46 30.47 38.67 32.37 28.51 20.19 16.85 17.81 34.12 32.28
381.5 385.1 267.5 356.8 336.3 292.5 425.2 525.1 474.7 438.2 386.4
1.27 1.78 1.69 1.65 1.72 1.61 1.57 1.68 1.71 1.72 1.73
Body weight gain Compound
activity
100.0
Cyproterone acetate * Relative t Relative
Biological
* + f + + + f + i + +
53.9** 120.2 149.8** 81.7 157.8** 99.7* 211.3* 206.3* 256.9 134.8 75.2**
k + f + f + + f + f +
5.67** 7.58 5.31 10.24* 8.75 5.48 5.18* 2.65** 2.75** 8.07 8.42
* Difference relative to TP significant at the level P < 0.05 **P < 0.01. t CPA = cyproterone acetate. Mean values If: SD are given (n = 7, for TP, CPA + TP and oil group n = 14).
k + f f + + + + + + +
58.4 108.3 112.0* 181.9 71.6 72.3 70.2 115.7* 226.8 94.8 91.3
k f f f $+ + + + + +
0.06** 0.18 0.17 0.20 0.10 0.20 0.19* 0.25 0.27 0.16 0.15
L. STARKA et cd.
334
fully abolished its binding to SHBG. However, even the presence of 17/%hydroxy group in 4,5-secosteroids I, 111 and IV was not sufficient to achieve a distinct binding to SWBG. Binding of the antiandrogenic 3,17-diol VIII confirms that a 3-0~0 group is not essential for high affinity interaction with the androgen receptor, as has been demonstrated by other authors [5, 14, IS]. The binding properties are only in approximate agreement with the biological activity found in ai~o which, however, couid be influenced by the metabolism of the compounds. Among steroids tested here there were no compounds with surprisingly high antiandrogenic activity. However, in any event the data presented are of interest for the understanding of structure-binding relationship of synthetic steroids.
REFERENCES
Strirka L. and Hampl R.: Antihormonal activities of steroids with modified rings A or B. in Antihormones (Edited by M. K. Agarwal). W. de Gruyter. Berlin (1982). Raynaud J. P. A.: A strategy for the design of potent hormones, Med. Chem. 5 (1977) 451-4.56. Starka L., Sulcova J., Broulik P. D., Joska J., FajkoS J. and DoskoEil M.: Screening for antiandrogenic activity of some 4,5-cyclo-A-homo-B-norandrostane derivatives. J. .steroid Biochem. 8 (1977) 939-941. Starka L., Hampl R., BiEikova M., Cerny V., FajkoS J., Kasal A., KoEovsky P., Kohout L. and V&ova H.: Steroids with modified ring A or B: screening for potential antiandrogenic and synandrogenic activity. J. steroid ~i~)~~~~z.13 (1980)455460.
5. Rousseau G. C., Quivy J. 1. and Crabbe P.: Interaction of A-nor, A-l 9-dinor, and A-homo-Sa-androstane derivatives with the androgen receptor and the epididyma1 androgen binding protein. Steroids 37 (19Sl) 383-392. 6. Kasal A.: Anaiogues of androgens with the open ring A. Coil. czecft. Chem. Commun. 45 (1980) 2541-2549. 7. Kasai A. and Kohout L.: 4,56ecoanalogues of dihydrotestosterone with a cyclopropane ring in 5,6-position. Cull. czech. Gem. Commlm. in press. g. Kohout L. and FajkoS J.: The epimeric 5,7-cycle-Bhomoandrostane derivatives. Co/I. cc&. Gem. Commun. 38 (1973) 1415-1427. 9. Kohout L. and FajkoS J.: Anaiogues of androgens with a cyclopropane ring in a 6a,?ol-position. Co/l. Czech. Chem. C~~~rnu~l. 45 (1980) 1974.-1981. 10, Krieg M., Steins P., Szalay R. and Voigt K. D.: Characterization of specific androgen receptor in rat prostate cytosol by agar gel eiectrophoresis. in aivo and in vitro studies. J. steroid Biochem. 5 (1974) 87-92. 11. Hampl R., Horakova J., BiCikova M., Dvoiak P. and Starka L.: Iododerivatives of testosterone as potential biological markers. J. steroid &o&em. 13 (1980) 1035TrO38 12. Lowry 0. H., Rosebrough N. J.. Farr A. F. and Randall R. J.: Protein measurement with the Folin phenol reagent. J. hic)i. Chem. 193 (1951) 265-271. t3. Bicikova M., Hampl R. and Starka L.: Binding of synthetic anabolic steroids to testosterone-estradiol binding globulin and to rat prostate cytosol. Endocr. exper. 1I (l977) 85-90. 14 Cunningham G. R., Tindall D. J. and Means A. R.: Difference in steroid specificity for rat androgen binding protein and the cytoplasmic protein. Stewids 33 (1979) 261-276. 15 Kirchhoff J., Soffie M. and Rousseau Ci. G.: Differences in the steroid-binding site specificities of rat prostate androgen receptor and epididymal androgenbinding protein (ABP). J. steroid B&hem. 10 (1979) 487-497.