- Email: [email protected]
11-Dehydrocorticosterone in the presence of carbenoxolone is a more potent sodium retainer than corticosterone
11-Dehydrocorticosterone in the presence of carbenoxolone is a more potent sodium retainer than corticosterone Graham W. Souness and David J. Morris Department of Pathology and Laboratory Medicine, The Miriam Hospital and the Division of Biology and Medicine, Brown University, Providence, Rhode Island, USA
In vivo, corticosterone and Il-dehydrocorticosterone are interconverted in the liver andpossibly kidney by II&hydroxysteroid dehydrogenase. In an effort to evaluate the relevance of this reversible reaction in relation to urinary sodium and potassium excretion, we investigated the effects of Il-dehydrocorticosterone in the presence and absence of carbenoxolone, a potent inhibitor of the oxidative component of I I@hydroxysteroid dehydrogenase, and compared them with the effects of similar doses of corticosterone in carbenoxolone-treated rats. All experiments were performed on adrenalectomized male rats. Here we describe that in carbenoxolone-treated rats II-dehydrocorticosterone and corticosterone display antinatriuretic activity, although under the conditions of this study II-dehydrocorticosterone is a more potent sodium retainer than its parent steroid corticosterone. In addition, the antinatriuretic effects of I1 -dehydrocorticosterone (like the antinatriuretic effects of corticosterone in carbenoxolonetreated rats) were blocked by the specific antimineralocorticoid RU28318. (Steroids S&24-28, 1993)
Keywords:
steroids;
1I-dehydrocorticosterone;
carbenoxolone;
Introduction Corticosterone, the main glucocorticoid in the rat, is metabolized in the liver and kidney to I l-dehydrocorticosterone by the enzyme 1l/3-hydroxysteroid dehydrogenase (1 I@OHSD).’ This enzymatic conversion is of particular significance given that (1) the circulating plasma level of corticosterone is approximately 500 times greater than that of aldosterone,2 and (2) in vitro, corticosterone and aldosterone have an equal affinity for renal mineralocorticoid receptors (MR), whereas 11-dehydrocorticosterone has been shown to have a very weak binding affinity for MR.3-6 The oxidative conversion of corticosterone to 1I-dehydrocorticosterone has been proposed as a mechanism that “protects” MR from corticosterone, and hence a mechanism by which the kidney can remain aldosterone selective in the regulation of electrolyte excretion.6,7 Indeed, we have previously shown that although certain doses of corticosterone in adrenalectomized rats had no ef-
Na+ retention; antimineralocorticoid
feet on Na+ excretion, in rats treated with carbenoxolone (CS), an inhibitor of 11/3-OHSD,8 the same doses of corticosterone caused significant antinatriuresis and kaliuresis;9 we were then able to block these effects with the specific MR antagonist RU28318.i0 1l/3-OHSD is made up of an oxidative component (which converts corticosterone to 1l-dehydrocorticosterone) and a reductive component (1 l-dehydrocorticosterone to corticosterone).8 In the kidney it has been reported that the main reaction is the conversion of corticosterone to 1l-dehydrocorticosterone, whereas in the liver both the oxidative and reductive components of the enzyme have been shown to be active.8,1’ The aims of the present study were to examine the effects of 1I-dehydrocorticosterone in C&treated animals in order to shed light on the relevance of the interconversion of corticosterone and 1l-dehydrocorticosterone in the liver and kidney.
Experimental Chemicals
Address reprint requests to Dr. G.W. Souness, Department of Pathology and Laboratory Medicine, The Miriam Hospital, 164 Summit Avenue, Providence, RI 02906, USA. Received February 16, 1592; accepted August 26, 1992.
24
Steroids,
1993, vol. 58, January
CS was obtained from Biorex Laboratories (London, England). Corticosterone was obtained from Sigma Chemicals (St. Louis, MO, USA) and 11-dehydrocorticosterone from Research Plus
0 1993 Butterworth-Heinemann
1 l-Dehydrocorticosterone Inc. (Bayonne, NJ, USA); the purity of each compound was established by high-performance liquid chromatography (HPLC) and nuclear magnetic resonance before use. Aldosterone was obtained from Andard Mount (London, England). RU28318 was kindly donated by Roussel Uclaf (Paris, France).
and Na+ retention:
Souness and Morris
1A. URINARY Na+/CREATININE (mmoles/g)
Rat bioassay Male Sprague-Dawley rats (Charles River Breeding Laboratories, Wilmington, MA, USA) were bilaterally adrenalectomized under ether anesthesia at 6-8 weeks of age. Thereafter, rats were allowed free access to 0.9% NaCl as drinking water at all times and to Purina Rat Chow (Ralston Purina, St. Louis, MO, USA) until 16 hours before experimentation. Rats (160-180 g body weight) were maintained in a temperature- and lightcontrolled room and were used 4-6 days postadrenalectomy. Experiments were performed as follows. Cortieosterone or 11.dehydrocorticosterone f CS. On the moming of the test, rats were injected subcutaneously (s.c.), with either CS (2.5 mg/rat) or vehicle. Thirty minutes later, at time 0, rats were then injected S.C. with 3 ml of 0.154 M NaCl and either corticosterone, 11-dehydrocorticosterone (each at doses of 1,5, 10, and 100 pglrat) or vehicle.
‘\
\ \
P(O.001 \T m
1
loo-! 1
10
100
CORTICOVERONE (B) (us/ret>
1B. URINARY KICREATININE
(mmoles/g)
11-Dehydrocorticoosterone + CS f RU28318. As in the preceding
section, animals were injected S.C. with CS (2.5 mglrat). Thirty minutes later (time 0) animals were injected with 3 ml of 0.154 M NaCl, 11-dehydrocorticosterone (5 pg/rat), and either RU283 18 (300 and 600 @g/rat) or vehicle. Corticosterone, 1 1-dehydrocorticosterone, and RU283 18 were dissolved in a mixture of 0.154 M NaCYethanol(90 : 10 vl v); CS was dissolved in 0.154 M saline. Each compound was injected in a volume of 0.2 ml. All animals were injected with 3.0 ml of 0.154 M NaCl at time 0 to ensure an adequate diuresis for subsequent urine collection. The 30-minute period between injections (in all experiments) provided a convenient window to allow the animals to empty their bladders; this was achieved with the aid of a whiff of ether and slight suprapubic pressure. Urine was collected for the time periods O-l, 1-3, and 3-4 hours postinjection of steroid (time 0) and analyzed for Na+ , K+ ,and creatinine content as previously described.‘* Mean (*SE) urinary Na+ /creatinine (Na+ICr) and K+lcreatinine (K+/Cr) ratios were computed for all experimental groups in each urine collection period. The results described here are those from the l-3 hour urine collection period only; the various compounds and combination of compounds used in these experiments did not significantly affect urinary electrolyte excretion in the O-l or 3-4 hour urine collection periods. The number of animals in each experimental group ranged from seven to 13 (Figures l-3). Statistical analyses [analysis of variance (ANOVA) and the Bonferroni test] were made between animals that received 1l-dehydrocorticosterone or corticosterone +- CS, and 1l-dehydrocorticosterone + CS + RU28318. In addition, the results of all experimental manipulations were compared with those of the controls.
-
1'0
i
100
CORTIC$;,Ta;NE (9)
Figure 1 Mineralocorticoid activity of 1, 5, 10, and 100 pglrat corticosterone with (W--- m) and without (O---O) CS pretreatment (2.5 mglrat). Corticosterone and CS were given S.C.to male ADXrats, and CS was administered 30 min before corticosterone. Also shown are the effects of CS (2.5 mg/rat) alone (A) and vehicle alone (A, control group). Urine collected l-3 hours postinjection of corticosterone or vehicle was analyzed for Nat, K+, and creatinine content. Antinatriuretic activity (A) is indicated by a decrease (from controls) in the mean urinary Na+/creatinine ratio and kaliuretic activity (6) by an increase in the mean urinary K+/creatinine ratio. Values shown are means 2 SE (n = 7-10 rats/group). Statistical comparisons were made (using ANOVA and the Bonferroni test) between groups of animels that received the same dosage of corticosterone 2 CS; statistically significant effects of corticosterone alone (versus controls) are described in the text.
Results Corticosterone
+ CS
Although we have previously reported the effects of corticosterone plus CS in the adrenalectomized rat,9 additional experiments, using lower dosages of corticosterone, were performed so that a comparison with the low doses of 11-dehydrocorticosterone plus CS could be better assessed. Corticosterone alone, at doses of 1,5, 10, and 100 pg/rat produced mean Na+/Cr ratios
of 583 * 85,489 f 46,461 + 46, and 536 + 22 mmol/ g, respectively (Figure 1A); these ratios were not signifkantly different from those of the control group (530 f 58 mmol/g). After pretreatment with CS (2.5 mg/rat) the same doses of corticosterone (1,5, 10, and 100 pg/rat) gave mean Na+/Cr ratios of 499 f 26,491 + 46, 439 -+ 22, and 162 + 39 mmol/g, respectively (Figure 1A). Thus, in C&treated rats, corticosterone displayed significant Na+-retaining activity only at a Steroids,
1993, vol. 58, January
25
Papers URINARY
2A.
Na+/CREATININE
Na~CREAnNlNE
(mmoles/g) 700
700 T 600 -500.-
T-y_&/+
1 fi
-
400 --
Q @NS -\ \
(mmoles/g)
T
1 1
\
300 --
\
\
‘\
P(O.01 P(O.001 -------_
200 --
100~
CTRL 318
1004 1
100
10 11 -DEHYDROCORTICOSTERONE (&rot)
28. URINARY
K+/CREATININE
(mmoles/g)
I
P(O.01
300 --
,i /’ 40’
200 --
cs
dhB
d;t
* P(O.01 VS CTRLand dhg
d;t
d;
318O 3’
Figure 3 Effects (in male ADX rats) of 11 -dehydrocorticosterone (dhS) alone, 11-dehydrocorticosterone after CS pretreatment, and 11-dehydrocorticosterone plus RU28318 (318) after CS pretreatment on urinary Na+ excretion. Also shown are the effects of CS alone, RU28318 alone, and vehicle alone (control group, CTRL). Compounds were administered (s.c.) alone or in combination at doses of 1 I-dehydrocorticosterone, 5 pg/rat; CS, 2.5 mg/ rat; and RU28318 200’ and 600b pg/rat. CS or vehicle was given 30 minutes before 11-dehydrocorticosterone or a combination of 1 l-dehydrocorticosterone and RU28318. Urine collected in the l-3 h postinjection period was analyzed for Na’, K+, and creatinine content. Antinatriuretic activity is expressed as a decrease (from controls) in the Na+/creatinine ratio. Values shown are means + SE (n = 7-13 rats/group); statistical comparisons were made using ANOVA and the Bonferroni test.
loo-
I I -Dehydrocorticosterone
01 1
10
loo
11 -DEHYDROCORTlCOSTERONE (&rot) Figure 2 Mineralocorticoid activity of 1, 5, 10, and 100 pg/rat II-dehydrocorticosterone with (m---m) and without (O---O) CS pretreatment (2.5 mg/rat). Also shown is the MC activity of CS (2.5 mg/rat) alone (A) and of vehicle alone (AI (see Figure 1 and methods for details). Values shown are means -C SE (n = 7-10 rats/group). Statistical analyses were made between groups of animals that received the same dosage of 1 l-dehydrocorticosterone f CS.
dose of 100 pg/rat, the highest dose used in the study. CS (2.5 mg/rat) alone gave a mean Na+/Cr ratio of 519 of: 24 mmol/g, a value not significantly different from controls (Figure 1A). Corticosterone at doses of 1, 5, and 10 pg/rat did not significantly affect urinary K+ excretion, giving mean K+/Cr ratios of 146 f 12, 134 ? 4, and 163 + 14 mmol/g, respectively, compared with 139 t 14 mmol/ g for the control group (Figure 1B). The urinary K+ excretory response to these doses of corticosterone was not affected by CS pretreatment. At a dose of 100 ,uglrat, corticosterone produced a significant kaliuresis, giving a mean K+/Cr ratio of 240 2 10 mmol/g (Figure 1B). This ratio was further increased to 339 + 33 mmol/g after pretreatment with 2.5 mg/rat CS. CS alone did not significantly affect urinary K+ excretion, giving a K+/Cr ratio of 161 + 16 mmol/g (Figure 1B). 26
Steroids,
1993,
vol.
58, January
f CS
1I-Dehydrocorticosterone alone did not affect urinary Nat excretion; doses of 1, 5, 10, and 100 pg/rat gave mean Na+/creatinine ratios of 531 + 42, 557 2 42, 552 + 57, and 600 + 50 mmol/g, respectively (Figure 2A). After CS (2.5 mg/rat) pretreatment, the same doses of 11-dehydrocorticosterone resulted in mean Na+/creatinine ratios of 449 ? 23,271 _+ 19,224 -+ 18, and 214 t 33 mmol/g (Figure 2A). Thus, in the presence of CS, 11-dehydrocorticosterone displayed significant Na+-retaining activity at doses as low as 5 pug/rat, indicating that under these conditions, 1l-dehydrocorticosterone is a more potent Na+ retainer than its parent steroid, corticosterone. 1I-Dehydrocorticosterone alone did not affect urinary K+ excretion (Figure 2B); however, in CS-treated rats, 100 pg/rat elicited a significant kaliuresis, increasing the mean K+/Cr ratio to 295 + 16 mmol/g.
II-Dehydrocorticosterone
+ CS -+ RU28318
In order to determine the effects of the specific antimineralocorticoid RU283 18 on the actions of 11-dehydrocorticosterone in CS-treated rats, we chose the lowest dosage of 11-dehydrocorticosterone (5 pg/rat) that gave “mineralocorticoid-like” effects. In a further set of experiments, 5 pug/rat 1I-dehydrocorticosterone gave a mean Na+/Cr ratio of 225 + 9 mmol/g in CStreated animals, whereas the co-administration of this dosage of ll-dehydrocorticosterone and 300 and 600
1 I-Dehydrocorticosterone
pglrat RU28318 gave mean Na+/Cr ratios of 473 + 18 and 567 + 53 mmol/g, respectively (Figure 3). Thus, it is clear that RU28318 can block the Na+ retaining actions of 11-dehydrocorticosterone in (X-treated rats. This dosage of 1l-dehydrocorticosterone (5 pg/rat) did not affect urinary K+ excretion in either CS-treated or nontreated rats. In whole bioassays similar to the one used here, a change in total urinary creatinine excretion is commonly used as an index of a change in glomerular filtration rate. Although there was some variation between experimental groups, the different compounds and combination of compounds used in this study did not significantly affect urinary creatinine excretion compared with that of the control group (data not shown).
Discussion In previous studies we have shown that corticosterone at a dosage of 100 pg/rat displays kaliuretic but not antinatriuretic activity in adrenalectomized rats; however, in CS-treated rats this same dose of corticosterone causes a significant antinatriuresis and the kaliuretic actions of corticosterone are enhancedm9 Further, we demonstrated that these effects of corticosterone in CS-treated animals could be blocked by RU28318.‘O These results suggest that CS has inhibited the oxidative component of 1l@OHSD blocking the conversion of corticosterone to 1l-dehydrocorticosterone; this allows corticosterone to bind to renal MR, which in turn leads to the subsequent changes in urinary Na+ and K+ excretion. 1l-Dehydrocorticosterone has been reported to have l/300 the affinity of corticosterone or aldosterone for MR, and is thus believed to have little or no mineralocorticoid activity.6 However, the present study shows that 1l-dehydrocorticosterone given to CS-treated rats results in a significant antinatriuresis, and that the dosages of 1 1-dehydrocorticosterone used to achieve these effects are much lower than the doses of corticosterone used in CS-treated rats to produce a similar antinatriuresis (5 pg/rat 1I-dehydrocorticosterone compared with 100 pg/rat corticosterone). In addition, the Na+-retaining effects of 1 1-dehydrocorticosterone in CS-treated rats were blocked by RU28318, suggesting that, like the Na+-retaining effects conferred on corticosterone in CS-treated rats, they are mediated via MR. Given that the binding of 1l-dehydrocorticosterone to MR is relatively weak, it is likely that a metabolite of 1I-dehydrocorticosterone is responsible for the antinatriuretic effects of exogenously administered 1ldehydrocorticosterone. The most probable candidate for this metabolite is corticosterone itself, and preliminary studies in this laboratory using HPLC have indicated that there is a substance that co-chromatographs with corticosterone in kidney extracts from adrenalectomized rats injected S.C. with tritiated 1l-dehydrocorticosterone; however, we do not know as yet if the source of this material is hepatic, renal, or both. If corticosterone is indeed responsible for the antina-
and Na+ retention:
Souness and Morris
triuretic actions conferred on 1l-dehydrocorticosterone by CS, it would indicate that under the conditions described here CS does not inhibit the reductive component of 1I@-OHSD. Thus, exogenous 1l-dehydrocorticosterone could be converted back to corticosterone, most probably in the liver and/or kidney, after which corticosterone produced in the liver would need to be transported to the kidney, where together with locally produced corticosterone it would then bind to MR. The conversion of corticosterone to 1l-dehydrocorticosterone by llfl-OHSD is reversible, and although studies have indicated that although the conversion of 1I-dehydrocorticosterone to corticosterone does not readily occur in kidney microsomal preparations,8*1’this is not true of the liver, where the reduction of ll-dehydrocorticosterone has been shown to be a major metabolic pathway.* However, it should be noted that Stewart et al. have shown that CS appears to inhibit both the oxidative and reductive components of lip-OHSD in humans.13 The difference in potency of exogenously administered corticosterone and 1 1-dehydrocorticosterone could be explained by the relative plasma binding of each compound to corticosteroid-binding globulin (CBG). Corticosterone binds extensively to CBG, whereas the binding of 1l-dehydrocorticosterone to CBG is minimal;6 hence, the amount of 1l-dehydrocorticosterone that reaches the kidney from a S.C. dose of 5 pg/rat could actually be similar or even greater than the amount of free corticosterone that reaches the kidney from a S.C. dose of 100 pg/rat. However, this theory relies on the ability of the kidney to convert the 1ldehydrocorticosterone (that reached it) back to corticosterone in order for corticosterone to bind to MR. Given that the conversion of 1l-dehydrocorticosterone to corticosterone in the kidneys,” and in isolated cultured cortical collecting duct cellsI does not occur to any great extent, an alternative explanation for our findings is that under these conditions the renal bioavailability of hepatically produced corticosterone is greater than that of exogenously administered corticosterone, or that other processes or mechanisms are involved. Further experiments are now necessary to explore in detail the interconversions of corticosterone and 1ldehydrocorticosterone under conditions where each of these compounds displays MC-like activity.
Acknowledgments This work was supported by National Institutes of Health Grant DK-21404 and The Miriam Hospital Research Foundation.
References 1. 2. 3.
Bush IE, Hunter SA, Meigs RA (1968). Metabolism of lloxygenated steroids. Biochem J 107:239-257. Holbrook MM, Dale SL, Melby JC (N80). Peripheral plasma steroid concentrations in rats sacrificed by anoxia. J Steroid Biochem W:135.5-1358. Krozowski ZS, Funder JW (1983). Renal mineralocorticoid
Steroids,
1993, vol. 59, January
27
Papers receptors and hippocampal corticosterone binding species have identical intrinsic steroid specificity. Prot Nutl Acud Sci
9.
Souness GW, Morris DJ (1989). The antinatriuretic and kaliuretic effects of corticosterone and cortisol following pretreatment with carbenoxolone sodium (a liquorice derivative) in the adrenalectom~zed rat. ~ndocri~oio~v X&&1588-
10.
Souness GW, Morris DJ (1991). The “mineralocorticoid-like” actions conferred on corticosterone by carbenoxoione are inhibited by the mineralocorticoid receptor (type I) antagonist RU28318. Endocrinology X29:2451-2&6. Hierholzer K, Siebe H. Fromm M (1990). Inhibition of I IBhydro~ystero~d dehydro~enase and its effect on epithelial sodium transport. Kidney int 38:673-678. _ Morris DJ, Kenyon CJ, Latif SA, McDermott M, Goodfriend T (1983). The possible significance of aldosterone metabolites. Nypertension (suppl 1) 5535~I40. Stewart PM. Wallace AM. Atherden SM. Shearine CH. Edwards CRH’ (1990). The ‘mineralocorticoid actio& of’ carbenoxolone: contrasting effects of carbenoxolone and liquorice on Ilk-hydroxysteroid dehydrogenase in man. Clin Sci 78:49-54. Naray-Fejes-Toth A, Watlington CO, Fejes-Toth G (1991). 1I&Hydroxysteroid dehydrogenase activity in the renal target cells of aldosterone. Endocrinology 129: 17-21.
USA ~6056-~. 4‘
5.
6.
7.
8.
28
Sheppard KE, Funder JW (1987). Equivalent affinity of aldostercme and corticosterone for type I receptors in kidney and hippocampus: direct binding studies. I Steroid Biochem 28:737-742. Arriza JL, Wienberger C, Cerelli G, Glaser TM, Handelin BL, Houseman DE, Evans RM (t 987). Cloning of human minemlocorticoid receptor complementary DNA: structural and functional kinship with the glucoco~icoid receptor. Science 237:X&-275. Funder JW, Pearce PT, Smith R, Smith AI (1988). Mineralocorticoid action: target tissue specificity is enzyme not receptor mediated. Scien& 242:583?85. Edwards CRW. Stewart PM. Burt D. Brett L. McIntvre MA. Sutanto WS, DeKloet ER, Monder 6 (1988).‘Localization of 1 l/3-hydroxysteroid dehydrogenase; tissue specific protector of the mineraloco~coid receptor. Lancet 2~986-989. Monder C, Stewart PM, Lakshmi V, Valentino R, Burt D, Edwards CRW (1989). Liquorice inhibits corticosteroid 1I@dehydrogenase of rat kidney and liver: in vivo and in vitro studies, Endocrinology l25:1046-1053.
Steroids, 1993, vol. 58, January
11. 12.
13.
14.
In vivo, corticosterone and Il-dehydrocorticosterone are interconverted in the liver andpossibly kidney by II&hydroxysteroid dehydrogenase. In an effort to evaluate the relevance of this reversible reaction in relation to urinary sodium and potassium excretion, we investigated the effects of Il-dehydrocorticosterone in the presence and absence of carbenoxolone, a potent inhibitor of the oxidative component of I I@hydroxysteroid dehydrogenase, and compared them with the effects of similar doses of corticosterone in carbenoxolone-treated rats. All experiments were performed on adrenalectomized male rats. Here we describe that in carbenoxolone-treated rats II-dehydrocorticosterone and corticosterone display antinatriuretic activity, although under the conditions of this study II-dehydrocorticosterone is a more potent sodium retainer than its parent steroid corticosterone. In addition, the antinatriuretic effects of I1 -dehydrocorticosterone (like the antinatriuretic effects of corticosterone in carbenoxolonetreated rats) were blocked by the specific antimineralocorticoid RU28318. (Steroids S&24-28, 1993)
Keywords:
steroids;
1I-dehydrocorticosterone;
carbenoxolone;
Introduction Corticosterone, the main glucocorticoid in the rat, is metabolized in the liver and kidney to I l-dehydrocorticosterone by the enzyme 1l/3-hydroxysteroid dehydrogenase (1 I@OHSD).’ This enzymatic conversion is of particular significance given that (1) the circulating plasma level of corticosterone is approximately 500 times greater than that of aldosterone,2 and (2) in vitro, corticosterone and aldosterone have an equal affinity for renal mineralocorticoid receptors (MR), whereas 11-dehydrocorticosterone has been shown to have a very weak binding affinity for MR.3-6 The oxidative conversion of corticosterone to 1I-dehydrocorticosterone has been proposed as a mechanism that “protects” MR from corticosterone, and hence a mechanism by which the kidney can remain aldosterone selective in the regulation of electrolyte excretion.6,7 Indeed, we have previously shown that although certain doses of corticosterone in adrenalectomized rats had no ef-
Na+ retention; antimineralocorticoid
feet on Na+ excretion, in rats treated with carbenoxolone (CS), an inhibitor of 11/3-OHSD,8 the same doses of corticosterone caused significant antinatriuresis and kaliuresis;9 we were then able to block these effects with the specific MR antagonist RU28318.i0 1l/3-OHSD is made up of an oxidative component (which converts corticosterone to 1l-dehydrocorticosterone) and a reductive component (1 l-dehydrocorticosterone to corticosterone).8 In the kidney it has been reported that the main reaction is the conversion of corticosterone to 1l-dehydrocorticosterone, whereas in the liver both the oxidative and reductive components of the enzyme have been shown to be active.8,1’ The aims of the present study were to examine the effects of 1I-dehydrocorticosterone in C&treated animals in order to shed light on the relevance of the interconversion of corticosterone and 1l-dehydrocorticosterone in the liver and kidney.
Experimental Chemicals
Address reprint requests to Dr. G.W. Souness, Department of Pathology and Laboratory Medicine, The Miriam Hospital, 164 Summit Avenue, Providence, RI 02906, USA. Received February 16, 1592; accepted August 26, 1992.
24
Steroids,
1993, vol. 58, January
CS was obtained from Biorex Laboratories (London, England). Corticosterone was obtained from Sigma Chemicals (St. Louis, MO, USA) and 11-dehydrocorticosterone from Research Plus
0 1993 Butterworth-Heinemann
1 l-Dehydrocorticosterone Inc. (Bayonne, NJ, USA); the purity of each compound was established by high-performance liquid chromatography (HPLC) and nuclear magnetic resonance before use. Aldosterone was obtained from Andard Mount (London, England). RU28318 was kindly donated by Roussel Uclaf (Paris, France).
and Na+ retention:
Souness and Morris
1A. URINARY Na+/CREATININE (mmoles/g)
Rat bioassay Male Sprague-Dawley rats (Charles River Breeding Laboratories, Wilmington, MA, USA) were bilaterally adrenalectomized under ether anesthesia at 6-8 weeks of age. Thereafter, rats were allowed free access to 0.9% NaCl as drinking water at all times and to Purina Rat Chow (Ralston Purina, St. Louis, MO, USA) until 16 hours before experimentation. Rats (160-180 g body weight) were maintained in a temperature- and lightcontrolled room and were used 4-6 days postadrenalectomy. Experiments were performed as follows. Cortieosterone or 11.dehydrocorticosterone f CS. On the moming of the test, rats were injected subcutaneously (s.c.), with either CS (2.5 mg/rat) or vehicle. Thirty minutes later, at time 0, rats were then injected S.C. with 3 ml of 0.154 M NaCl and either corticosterone, 11-dehydrocorticosterone (each at doses of 1,5, 10, and 100 pglrat) or vehicle.
‘\
\ \
P(O.001 \T m
1
loo-! 1
10
100
CORTICOVERONE (B) (us/ret>
1B. URINARY KICREATININE
(mmoles/g)
11-Dehydrocorticoosterone + CS f RU28318. As in the preceding
section, animals were injected S.C. with CS (2.5 mglrat). Thirty minutes later (time 0) animals were injected with 3 ml of 0.154 M NaCl, 11-dehydrocorticosterone (5 pg/rat), and either RU283 18 (300 and 600 @g/rat) or vehicle. Corticosterone, 1 1-dehydrocorticosterone, and RU283 18 were dissolved in a mixture of 0.154 M NaCYethanol(90 : 10 vl v); CS was dissolved in 0.154 M saline. Each compound was injected in a volume of 0.2 ml. All animals were injected with 3.0 ml of 0.154 M NaCl at time 0 to ensure an adequate diuresis for subsequent urine collection. The 30-minute period between injections (in all experiments) provided a convenient window to allow the animals to empty their bladders; this was achieved with the aid of a whiff of ether and slight suprapubic pressure. Urine was collected for the time periods O-l, 1-3, and 3-4 hours postinjection of steroid (time 0) and analyzed for Na+ , K+ ,and creatinine content as previously described.‘* Mean (*SE) urinary Na+ /creatinine (Na+ICr) and K+lcreatinine (K+/Cr) ratios were computed for all experimental groups in each urine collection period. The results described here are those from the l-3 hour urine collection period only; the various compounds and combination of compounds used in these experiments did not significantly affect urinary electrolyte excretion in the O-l or 3-4 hour urine collection periods. The number of animals in each experimental group ranged from seven to 13 (Figures l-3). Statistical analyses [analysis of variance (ANOVA) and the Bonferroni test] were made between animals that received 1l-dehydrocorticosterone or corticosterone +- CS, and 1l-dehydrocorticosterone + CS + RU28318. In addition, the results of all experimental manipulations were compared with those of the controls.
-
1'0
i
100
CORTIC$;,Ta;NE (9)
Figure 1 Mineralocorticoid activity of 1, 5, 10, and 100 pglrat corticosterone with (W--- m) and without (O---O) CS pretreatment (2.5 mglrat). Corticosterone and CS were given S.C.to male ADXrats, and CS was administered 30 min before corticosterone. Also shown are the effects of CS (2.5 mg/rat) alone (A) and vehicle alone (A, control group). Urine collected l-3 hours postinjection of corticosterone or vehicle was analyzed for Nat, K+, and creatinine content. Antinatriuretic activity (A) is indicated by a decrease (from controls) in the mean urinary Na+/creatinine ratio and kaliuretic activity (6) by an increase in the mean urinary K+/creatinine ratio. Values shown are means 2 SE (n = 7-10 rats/group). Statistical comparisons were made (using ANOVA and the Bonferroni test) between groups of animels that received the same dosage of corticosterone 2 CS; statistically significant effects of corticosterone alone (versus controls) are described in the text.
Results Corticosterone
+ CS
Although we have previously reported the effects of corticosterone plus CS in the adrenalectomized rat,9 additional experiments, using lower dosages of corticosterone, were performed so that a comparison with the low doses of 11-dehydrocorticosterone plus CS could be better assessed. Corticosterone alone, at doses of 1,5, 10, and 100 pg/rat produced mean Na+/Cr ratios
of 583 * 85,489 f 46,461 + 46, and 536 + 22 mmol/ g, respectively (Figure 1A); these ratios were not signifkantly different from those of the control group (530 f 58 mmol/g). After pretreatment with CS (2.5 mg/rat) the same doses of corticosterone (1,5, 10, and 100 pg/rat) gave mean Na+/Cr ratios of 499 f 26,491 + 46, 439 -+ 22, and 162 + 39 mmol/g, respectively (Figure 1A). Thus, in C&treated rats, corticosterone displayed significant Na+-retaining activity only at a Steroids,
1993, vol. 58, January
25
Papers URINARY
2A.
Na+/CREATININE
Na~CREAnNlNE
(mmoles/g) 700
700 T 600 -500.-
T-y_&/+
1 fi
-
400 --
Q @NS -\ \
(mmoles/g)
T
1 1
\
300 --
\
\
‘\
P(O.01 P(O.001 -------_
200 --
100~
CTRL 318
1004 1
100
10 11 -DEHYDROCORTICOSTERONE (&rot)
28. URINARY
K+/CREATININE
(mmoles/g)
I
P(O.01
300 --
,i /’ 40’
200 --
cs
dhB
d;t
* P(O.01 VS CTRLand dhg
d;t
d;
318O 3’
Figure 3 Effects (in male ADX rats) of 11 -dehydrocorticosterone (dhS) alone, 11-dehydrocorticosterone after CS pretreatment, and 11-dehydrocorticosterone plus RU28318 (318) after CS pretreatment on urinary Na+ excretion. Also shown are the effects of CS alone, RU28318 alone, and vehicle alone (control group, CTRL). Compounds were administered (s.c.) alone or in combination at doses of 1 I-dehydrocorticosterone, 5 pg/rat; CS, 2.5 mg/ rat; and RU28318 200’ and 600b pg/rat. CS or vehicle was given 30 minutes before 11-dehydrocorticosterone or a combination of 1 l-dehydrocorticosterone and RU28318. Urine collected in the l-3 h postinjection period was analyzed for Na’, K+, and creatinine content. Antinatriuretic activity is expressed as a decrease (from controls) in the Na+/creatinine ratio. Values shown are means + SE (n = 7-13 rats/group); statistical comparisons were made using ANOVA and the Bonferroni test.
loo-
I I -Dehydrocorticosterone
01 1
10
loo
11 -DEHYDROCORTlCOSTERONE (&rot) Figure 2 Mineralocorticoid activity of 1, 5, 10, and 100 pg/rat II-dehydrocorticosterone with (m---m) and without (O---O) CS pretreatment (2.5 mg/rat). Also shown is the MC activity of CS (2.5 mg/rat) alone (A) and of vehicle alone (AI (see Figure 1 and methods for details). Values shown are means -C SE (n = 7-10 rats/group). Statistical analyses were made between groups of animals that received the same dosage of 1 l-dehydrocorticosterone f CS.
dose of 100 pg/rat, the highest dose used in the study. CS (2.5 mg/rat) alone gave a mean Na+/Cr ratio of 519 of: 24 mmol/g, a value not significantly different from controls (Figure 1A). Corticosterone at doses of 1, 5, and 10 pg/rat did not significantly affect urinary K+ excretion, giving mean K+/Cr ratios of 146 f 12, 134 ? 4, and 163 + 14 mmol/g, respectively, compared with 139 t 14 mmol/ g for the control group (Figure 1B). The urinary K+ excretory response to these doses of corticosterone was not affected by CS pretreatment. At a dose of 100 ,uglrat, corticosterone produced a significant kaliuresis, giving a mean K+/Cr ratio of 240 2 10 mmol/g (Figure 1B). This ratio was further increased to 339 + 33 mmol/g after pretreatment with 2.5 mg/rat CS. CS alone did not significantly affect urinary K+ excretion, giving a K+/Cr ratio of 161 + 16 mmol/g (Figure 1B). 26
Steroids,
1993,
vol.
58, January
f CS
1I-Dehydrocorticosterone alone did not affect urinary Nat excretion; doses of 1, 5, 10, and 100 pg/rat gave mean Na+/creatinine ratios of 531 + 42, 557 2 42, 552 + 57, and 600 + 50 mmol/g, respectively (Figure 2A). After CS (2.5 mg/rat) pretreatment, the same doses of 11-dehydrocorticosterone resulted in mean Na+/creatinine ratios of 449 ? 23,271 _+ 19,224 -+ 18, and 214 t 33 mmol/g (Figure 2A). Thus, in the presence of CS, 11-dehydrocorticosterone displayed significant Na+-retaining activity at doses as low as 5 pug/rat, indicating that under these conditions, 1l-dehydrocorticosterone is a more potent Na+ retainer than its parent steroid, corticosterone. 1I-Dehydrocorticosterone alone did not affect urinary K+ excretion (Figure 2B); however, in CS-treated rats, 100 pg/rat elicited a significant kaliuresis, increasing the mean K+/Cr ratio to 295 + 16 mmol/g.
II-Dehydrocorticosterone
+ CS -+ RU28318
In order to determine the effects of the specific antimineralocorticoid RU283 18 on the actions of 11-dehydrocorticosterone in CS-treated rats, we chose the lowest dosage of 11-dehydrocorticosterone (5 pg/rat) that gave “mineralocorticoid-like” effects. In a further set of experiments, 5 pug/rat 1I-dehydrocorticosterone gave a mean Na+/Cr ratio of 225 + 9 mmol/g in CStreated animals, whereas the co-administration of this dosage of ll-dehydrocorticosterone and 300 and 600
1 I-Dehydrocorticosterone
pglrat RU28318 gave mean Na+/Cr ratios of 473 + 18 and 567 + 53 mmol/g, respectively (Figure 3). Thus, it is clear that RU28318 can block the Na+ retaining actions of 11-dehydrocorticosterone in (X-treated rats. This dosage of 1l-dehydrocorticosterone (5 pg/rat) did not affect urinary K+ excretion in either CS-treated or nontreated rats. In whole bioassays similar to the one used here, a change in total urinary creatinine excretion is commonly used as an index of a change in glomerular filtration rate. Although there was some variation between experimental groups, the different compounds and combination of compounds used in this study did not significantly affect urinary creatinine excretion compared with that of the control group (data not shown).
Discussion In previous studies we have shown that corticosterone at a dosage of 100 pg/rat displays kaliuretic but not antinatriuretic activity in adrenalectomized rats; however, in CS-treated rats this same dose of corticosterone causes a significant antinatriuresis and the kaliuretic actions of corticosterone are enhancedm9 Further, we demonstrated that these effects of corticosterone in CS-treated animals could be blocked by RU28318.‘O These results suggest that CS has inhibited the oxidative component of 1l@OHSD blocking the conversion of corticosterone to 1l-dehydrocorticosterone; this allows corticosterone to bind to renal MR, which in turn leads to the subsequent changes in urinary Na+ and K+ excretion. 1l-Dehydrocorticosterone has been reported to have l/300 the affinity of corticosterone or aldosterone for MR, and is thus believed to have little or no mineralocorticoid activity.6 However, the present study shows that 1l-dehydrocorticosterone given to CS-treated rats results in a significant antinatriuresis, and that the dosages of 1 1-dehydrocorticosterone used to achieve these effects are much lower than the doses of corticosterone used in CS-treated rats to produce a similar antinatriuresis (5 pg/rat 1I-dehydrocorticosterone compared with 100 pg/rat corticosterone). In addition, the Na+-retaining effects of 1 1-dehydrocorticosterone in CS-treated rats were blocked by RU28318, suggesting that, like the Na+-retaining effects conferred on corticosterone in CS-treated rats, they are mediated via MR. Given that the binding of 1l-dehydrocorticosterone to MR is relatively weak, it is likely that a metabolite of 1I-dehydrocorticosterone is responsible for the antinatriuretic effects of exogenously administered 1ldehydrocorticosterone. The most probable candidate for this metabolite is corticosterone itself, and preliminary studies in this laboratory using HPLC have indicated that there is a substance that co-chromatographs with corticosterone in kidney extracts from adrenalectomized rats injected S.C. with tritiated 1l-dehydrocorticosterone; however, we do not know as yet if the source of this material is hepatic, renal, or both. If corticosterone is indeed responsible for the antina-
and Na+ retention:
Souness and Morris
triuretic actions conferred on 1l-dehydrocorticosterone by CS, it would indicate that under the conditions described here CS does not inhibit the reductive component of 1I@-OHSD. Thus, exogenous 1l-dehydrocorticosterone could be converted back to corticosterone, most probably in the liver and/or kidney, after which corticosterone produced in the liver would need to be transported to the kidney, where together with locally produced corticosterone it would then bind to MR. The conversion of corticosterone to 1l-dehydrocorticosterone by llfl-OHSD is reversible, and although studies have indicated that although the conversion of 1I-dehydrocorticosterone to corticosterone does not readily occur in kidney microsomal preparations,8*1’this is not true of the liver, where the reduction of ll-dehydrocorticosterone has been shown to be a major metabolic pathway.* However, it should be noted that Stewart et al. have shown that CS appears to inhibit both the oxidative and reductive components of lip-OHSD in humans.13 The difference in potency of exogenously administered corticosterone and 1 1-dehydrocorticosterone could be explained by the relative plasma binding of each compound to corticosteroid-binding globulin (CBG). Corticosterone binds extensively to CBG, whereas the binding of 1l-dehydrocorticosterone to CBG is minimal;6 hence, the amount of 1l-dehydrocorticosterone that reaches the kidney from a S.C. dose of 5 pg/rat could actually be similar or even greater than the amount of free corticosterone that reaches the kidney from a S.C. dose of 100 pg/rat. However, this theory relies on the ability of the kidney to convert the 1ldehydrocorticosterone (that reached it) back to corticosterone in order for corticosterone to bind to MR. Given that the conversion of 1l-dehydrocorticosterone to corticosterone in the kidneys,” and in isolated cultured cortical collecting duct cellsI does not occur to any great extent, an alternative explanation for our findings is that under these conditions the renal bioavailability of hepatically produced corticosterone is greater than that of exogenously administered corticosterone, or that other processes or mechanisms are involved. Further experiments are now necessary to explore in detail the interconversions of corticosterone and 1ldehydrocorticosterone under conditions where each of these compounds displays MC-like activity.
Acknowledgments This work was supported by National Institutes of Health Grant DK-21404 and The Miriam Hospital Research Foundation.
References 1. 2. 3.
Bush IE, Hunter SA, Meigs RA (1968). Metabolism of lloxygenated steroids. Biochem J 107:239-257. Holbrook MM, Dale SL, Melby JC (N80). Peripheral plasma steroid concentrations in rats sacrificed by anoxia. J Steroid Biochem W:135.5-1358. Krozowski ZS, Funder JW (1983). Renal mineralocorticoid
Steroids,
1993, vol. 59, January
27
Papers receptors and hippocampal corticosterone binding species have identical intrinsic steroid specificity. Prot Nutl Acud Sci
9.
Souness GW, Morris DJ (1989). The antinatriuretic and kaliuretic effects of corticosterone and cortisol following pretreatment with carbenoxolone sodium (a liquorice derivative) in the adrenalectom~zed rat. ~ndocri~oio~v X&&1588-
10.
Souness GW, Morris DJ (1991). The “mineralocorticoid-like” actions conferred on corticosterone by carbenoxoione are inhibited by the mineralocorticoid receptor (type I) antagonist RU28318. Endocrinology X29:2451-2&6. Hierholzer K, Siebe H. Fromm M (1990). Inhibition of I IBhydro~ystero~d dehydro~enase and its effect on epithelial sodium transport. Kidney int 38:673-678. _ Morris DJ, Kenyon CJ, Latif SA, McDermott M, Goodfriend T (1983). The possible significance of aldosterone metabolites. Nypertension (suppl 1) 5535~I40. Stewart PM. Wallace AM. Atherden SM. Shearine CH. Edwards CRH’ (1990). The ‘mineralocorticoid actio& of’ carbenoxolone: contrasting effects of carbenoxolone and liquorice on Ilk-hydroxysteroid dehydrogenase in man. Clin Sci 78:49-54. Naray-Fejes-Toth A, Watlington CO, Fejes-Toth G (1991). 1I&Hydroxysteroid dehydrogenase activity in the renal target cells of aldosterone. Endocrinology 129: 17-21.
USA ~6056-~. 4‘
5.
6.
7.
8.
28
Sheppard KE, Funder JW (1987). Equivalent affinity of aldostercme and corticosterone for type I receptors in kidney and hippocampus: direct binding studies. I Steroid Biochem 28:737-742. Arriza JL, Wienberger C, Cerelli G, Glaser TM, Handelin BL, Houseman DE, Evans RM (t 987). Cloning of human minemlocorticoid receptor complementary DNA: structural and functional kinship with the glucoco~icoid receptor. Science 237:X&-275. Funder JW, Pearce PT, Smith R, Smith AI (1988). Mineralocorticoid action: target tissue specificity is enzyme not receptor mediated. Scien& 242:583?85. Edwards CRW. Stewart PM. Burt D. Brett L. McIntvre MA. Sutanto WS, DeKloet ER, Monder 6 (1988).‘Localization of 1 l/3-hydroxysteroid dehydrogenase; tissue specific protector of the mineraloco~coid receptor. Lancet 2~986-989. Monder C, Stewart PM, Lakshmi V, Valentino R, Burt D, Edwards CRW (1989). Liquorice inhibits corticosteroid 1I@dehydrogenase of rat kidney and liver: in vivo and in vitro studies, Endocrinology l25:1046-1053.
Steroids, 1993, vol. 58, January
11. 12.
13.
14.