Interactions of mineralocorticoids and glucocorticoids in epithelial target tissues

Kidney International, Vol. 57 (2000), pp. 1370–1373

Interactions of mineralocorticoids and glucocorticoids in epithelial target tissues DAVID J. MORRIS, GRAHAM W. SOUNESS, ANDREW S. BREM, and MARIE-EDITH OBLIN Department of Pathology and Laboratory Medicine, Division of Pediatric Nephrology, The Miriam Hospital and Rhode Island Hospital, Brown University School of Medicine, Providence, Rhode Island, USA, and Inserm U478, Faculte´ de Me´decine, Hoˆpital Xavier Bichat, Paris, France

Interactions of mineralocorticoids and glucocorticoids in epithelial target tissues. Background. In Na⫹-transporting epithelial target tissues, such as mammalian kidney and the isolated toad bladder, glucocorticoids (GCs) do not normally elicit Na⫹ retention. In mammalian kidney, however, they do cause kaliuresis. The presence of 11␤-hydroxysteroid dehydrogenase isoform 2 (11␤-HSD2) in these target tissues inactivates the GCs, preventing them from accessing mineralocorticoid receptors (MRs) and stimulating Na⫹ transport. Results. The usually observed Na⫹ retention elicited by the mineralocorticoid aldosterone was blunted when the GC corticosterone was coadministered along with aldosterone. However, when corticosterone was administered along with a 11␤HSD2 inhibitor, a strong Na⫹ transport was elicited by an MRmediated mechanism. 11-Dehydrocorticosterone also blunted aldosterone-elicited Na⫹ transport in these target tissues. Conclusions. 11␤-HSD2 appears to play two important roles in the epithelial target tissues, kidney and toad bladder. The first is to protect GC access to MR, and the second involves the product of the enzyme to regulate the magnitude of aldosterone-induced Na⫹ retention.

In mineralocorticoid target tissues, such as the kidney and the parotid gland, mineralocorticoid receptors (MRs) appear to be protected from the physiological effects of circulating endogenous glucocorticoids (GCs) [1, 2]. The endogenous GCs corticosterone and cortisol and the mineralocorticoid aldosterone demonstrate equal binding affinities for MR in vitro [1, 3, 4]. Under normal circumstances, GCs do not bind to MR in vivo even though the circulating levels of cortisol (in humans) and corticosterone (rat) are approximately 500 times greater than that of aldosterone [5]. Cortisol and corticosterone do not induce mineralocorticoid-like actions (particularly a Key words: antinatriuresis, 11␤-HSD2, epithelial cells, kaliuresis, sodium transport, aldosterone.

 2000 by the International Society of Nephrology

Na⫹ retention) in large part because of the actions of the enzyme 11␤-hydroxysteroid dehydrogenase isoform 2 (11␤-HSD2). In the distal nephron, this enzyme functions as a unidirectional dehydrogenase [6] metabolizing endogenous GCs to their biologically inactive 11-dehydro derivatives (for example, cortisol to cortisone in humans and corticosterone to 11-dehydrocorticosterone in rats). The 11-dehydro products of cortisol and corticosterone are generally considered to be biologically inert and have little or no affinity for either steroid receptor [1]. Children with a syndrome of apparent mineralocorticoid excess (AME) provide the clinical evidence in support of the “guardian” role of a renal 11␤-HSD2. All of these children have homozygous mutations in the 11bHSD2 gene that result in diminished 11␤-HSD2 activity [7, 8]. Children with AME display increased Na⫹ retention, K⫹ wasting, marked hypertension, and severely impaired conversion of cortisol to cortisone [9]. Secretion of GCs and other adrenal steroid precursors is in general significantly greater than that of the mineralocorticoid steroids. Since these adrenal corticosteroids, under certain physiologic conditions, can also cause mineralocorticoid-like Na⫹ retention, many investigators have suggested the possibility of functional interactions between GCs and mineralocorticoids in the regulation of sodium and fluid homeostasis. The concept that 11␤-HSD2 is a “protective enzyme” preventing GCs from accessing MR and causing mineralocorticoid-like Na⫹ retention [1, 2] further supports the idea of functional interactions. The isoform 11␤-HSD1 has a wide tissue distribution and is also present in several target tissues of GCs and mineralocorticoids [10, 11]. Although 11␤-HSD1 may function as both a dehydrogenase and as an oxoreductase, in most tissues, this isoform functions largely as a reductase, activating 11-dehydro-derivatives back to their parent GC hormones. Thus, 11␤-HSD1 may also play a role in the interactions between GCs and mineralocorticoids in the regulation of electrolyte and fluid homeostasis.

1370

1371

Morris et al: Interactions of mineralocorticoids and GCs

REGULATION OF MINERALOCORTICOID Naⴙ RETENTION AND Kⴙ SECRETION BY GLUCOCORTICOIDS Physiological paradoxes in the action of mineralocorticoids and glucocorticoids Several years ago, our laboratory demonstrated that the administration of physiologic dosages of aldosterone to adrenally intact male rats elicited only a kaliuretic effect [12]. However, administration of the same dosage of aldosterone to adrenalectomized (ADX) male rats induced both antinatriuresis and kaliuresis, the effects commonly associated with mineralocorticoids [12]. This response pattern has also been reported in intact and ADX dogs [13]. These important observations are evidence that secretory products from the adrenal gland regulate the renal sodium retention in response to aldosterone. Interestingly, a kaliuresis can be observed following the administration of physiological dosages of either GC, cortisol, or corticosterone to ADX rats [14, 15]. This finding is paradoxical because the kaliuresis occurs following activation of MRs without accompanying antinatriuresis. The separability of K⫹ from Na⫹ transport could be explained by the presence of MRs in two or more distinct populations of renal epithelial cells, each responding differently to the same stimulus or by a postreceptor modification resulting in different responses occurring in the same cells. These experiments have not only provided additional evidence that renal 11␤-HSD2 does prevent GCs from causing excessive Na⫹ retention, but also supports the concept that separate mechanisms are responsible for each of the physiological effects of mineralocorticoids, Na⫹ retention, and K⫹ secretion. We have previously shown that corticosterone blunts the observed renal antinatriuretic response to aldosterone when both steroids are coadministered in physiologic doses to ADX rats [16]. Similar findings that endogenous and synthetic GCs can inhibit the actions of aldosterone on urinary Na⫹ excretion have also been reported in humans [17] and in dogs [18]. Subsequent experiments by Brem et al using the isolated toad bladder preparation also demonstrated that corticosterone attenuated Na⫹ transport elicited by aldosterone [19]. However, isolated toad bladders exposed to physiologic concentrations of corticosterone alone do not generate an increase in Na⫹ transport comparable to that caused by aldosterone, most likely because of the presence of the guardian enzyme 11␤-HSD2 [19]. When 11␤-HSD2 in the toad bladder is inhibited by pretreatment with carbenoxolone (CBX), corticosterone induces a rise in Na⫹ transport similar to that observed with aldosterone [20]. Thus, it would appear that corticosterone or one of its derivatives may be one of the adrenal products regulating the sodiumretaining effects of aldosterone. This presents us with an

Table 1. Effect of steroids on the urinary excretion of sodium and potassium in the ADX male rat Condition Control Aldo 0.1 ␮g/rat Aldo ⫹ 50 ␮g corticosterone Aldo ⫹ 100 ␮g corticosterone Aldo ⫹ 250 ␮g corticosterone Aldo ⫹ 50 ␮g 11-dehydro Aldo ⫹ 100 ␮g 11-dehydro Aldo ⫹ 250 ␮g 11-dehydro Corticosterone 50 ␮g Corticosterone 100 ␮g Corticosterone 250 ␮g 11-dehydro 50 ␮g 11-dehydro 100 ␮g 11-dehydro 250 ␮g

Na⫹/creatinine mmol/g

K⫹/creatinine mmol/g

530 ⫾ 58 202 ⫾ 18a 209 ⫾ 15a 341 ⫾ 18a,b 503 ⫾ 43b 217 ⫾ 21a 388 ⫾ 50a,b 473 ⫾ 30b 470 ⫾ 63 536 ⫾ 22 420 ⫾ 19 463 ⫾ 31 600 ⫾ 22 497 ⫾ 75

139 ⫾ 14 291 ⫾ 24a 288 ⫾ 27a 211 ⫾ 24 444 ⫾ 20a,b 207 ⫾ 18a,b 256 ⫾ 15a 333 ⫾ 37a 226 ⫾ 20a 240 ⫾ 16a 345 ⫾ 75a 109 ⫾ 10 143 ⫾ 20 441 ⫾ 20a

Mean ⫾ SE; N ⫽ 6–8 animals per group. a P ⬍ 0.01 vs. control b P ⬍ 0.01 vs. aldosterone alone

additional paradox in that inhibition of 11␤-HSD2 in the kidney following treatment of ADX rats with either CBX or other 11␤-HSD2 inhibitors allows GCs to access MR with resulting antinatriuresis and enhanced kaliuresis [15]. One might have expected that access of corticosterone to MRs enhances the sodium-retaining actions of aldosterone rather than inhibiting them. ROLE OF 11-DEHYDRO-DERIVATIVES The above findings prompted us to look at the physiological role of the product of the enzyme 11␤-HSD2, 11dehydrocorticosterone, in both the mammalian kidney and the isolated toad bladder. 11-Dehydrocorticosterone alone at physiological dosages is biologically inactive in ADX rats, but when administered at higher dosages, it does elicit a kaliuresis [21]. When 11-dehydrocorticosterone is coadministered with aldosterone to an ADX rat, the renal antinatriuretic component of the physiological response to aldosterone was blunted in a dose-dependent fashion similar to that observed with corticosterone and aldosterone (Table 1). When given alone, neither corticosterone nor 11-dehydrocorticosterone at a dose of 250 ␮g per rat had significant effects on urinary Na⫹ excretion [21]. Corticosterone alone possessed intrinsic kaliuretic activity at doses of 50, 100, and 250 ␮g per rat compared with the control group. 11-Dehydrocorticosterone, however, only exhibited intrinsic kaliuretic activity at the highest dosage used (250 ␮g per rat). Aldosterone (0.1 ␮g per rat) alone caused a typical kaliuresis, while the effects of a combination of corticosterone and aldosterone on urinary K⫹ excretion appeared to be additive at the highest dosage of corticosterone only. Similar effects were seen with the administration of aldosterone plus

1372

Morris et al: Interactions of mineralocorticoids and GCs

Fig. 1. Effect of 11-dehydrocorticosterone (11␤-HSD2; 10 ␮mol/L) in the serosal bath on the short circuit current (SCC) generated by aldosterone (10 nmol/L). The 11-dehydrocorticosterone was added to the serosal bath at –60 minutes, and aldosterone was added at 0 time. Data are expressed as mean ⫾ SE (N ⫽ 6 pairs of toad bladders). Symbols are: (䊏) aldosterone ⫹ 11␤-HSD2; (䉫) aldosterone; *P ⬍ 0.001; **P ⬍ 0.01. (This figure is reproduced from Am J Physiol 261: 873–879, 1999 with permission from the American Physiological Society.)

11-dehydrocorticosterone. Thus, both the end product of 11␤-HSD2, 11-dehydrocorticosterone, and its substrate, corticosterone, can modify the magnitude of the Na⫹retaining activity of aldosterone in the kidney, which is consistent with the view that the adrenal GCs modulate mineralocorticoid action. Similar experiments were then performed using the isolated toad bladder preparation. 11-Dehydrocorticosterone did not cause any effect on Na⫹ transport, as demonstrated by the short-circuit current in this model. However, when bladders were pretreated with 11-dehydrocorticosterone and then stimulated with physiological concentrations of aldosterone [19], the short circuit current elicited by aldosterone was significantly suppressed (Fig. 1). Interestingly, when bladders were exposed to the enzyme end product 11-dehydrocorticosterone and then stimulated with corticosterone [22], the short-circuit current was enhanced relative to that seen with corticosterone alone. This GC-generated enhanced Na⫹ transport most likely occurs from end-product inhibition of 11␤-HSD2 in this target tissue. Thus, although previously thought to be inactive, 11-dehydrocorticosterone and possibly other 11-dehydro-derivatives may play a functional role in regulating the magnitude of aldosteroneinduced renal Na⫹ retention and the extent to which GCs induce Na⫹ retention in the kidney. In a preliminary set of experiments conducted by M.E. Oblin, either human mineralocorticoid or GC receptors were transfected into a COS-7 cell line along with a luciferase reporter system. When cells were transfected with MRs and stimulated with aldosterone, a significant

increase in luciferase activity was observed as expected. However, if these same cells were incubated with both 11-dehydrocorticosterone and aldosterone, the luciferase activity was almost completely blocked. The 11-dehydrocorticosterone alone produced little or no luciferase activity. When COS-7 cells were transfected with human GC receptors and incubated with dexamethasone, a large degree of luciferase activity could again be seen. However, 11-dehydrocorticosterone had no effect on the luciferase response when cells were incubated with both dexamethasone and 11-dehydrocorticosterone. Thus, the otherwise biologically inert metabolic product of 11␤HSD2, 11-dehydrocorticosterone, appears able to directly affect the aldosterone action at the receptor level. CONCLUSION Based on our findings, it appears that 11␤-HSD2 serves two purposes. In addition to the “guardian” role of 11␤-HSD2, in which the enzyme limits the access of glucocorticoids to MRs, the 11-dehydro-product of 11␤HSD2 serves to regulate the magnitude of aldosteroneinduced Na⫹ retention. The fact that at this moment some of these findings appear paradoxical should spur on our efforts in this area. It is uncertain at this time whether 11-dehydro-GCs, which do not bind well to MR [1], act as competitive antagonists. However, earlier studies with toad bladder had indicated that 11-dehydrocorticosterone lowered the binding of aldosterone in the nuclei of cells in this target tissue [23]. Heterodimers of MR and glucocorticoid receptors have been suggested to play a role(s) in the magnitude of target tissue responses [24]. Similarly, our findings that 11␤-HSD1 inhibitors, such as the bile acid chenodeoxycholic acid, can also cause the GC corticosterone to elicit Na⫹ retention [25] present a third paradox. These results may not only suggest a role for 11␤-HSD1 in Na⫹ transport, but also for interactions in these processes by the newly discovered bile acid receptors (BARs), which are also present in kidney [26]. ACKNOWLEDGMENTS This work was supported by National Institutes of Health Grant DK-21404 and The Miriam Hospital Research Foundation. The authors would like to thank Ms. Patricia A. Blade for secretarial skills. Reprint requests to Dr. David J. Morris, The Miriam Hospital, Department of Pathology and Laboratory Medicine, 164 Summit Avenue, Providence, Rhode Island 02906, USA. E-mail: [email protected]

REFERENCES 1. Funder JW, Pearce PT, Smith R, Smith AI: Mineralocorticoid action: Target tissue specificity is enzyme, not receptor, mediated. Science 242:583–585, 1988 2. Edwards CR, Stewart PM, Burt D, Brett L, McIntyre MA, Sutanto WS, de Kloet ER, Monder C: Localisation of 11 beta-

Morris et al: Interactions of mineralocorticoids and GCs

3.

4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

hydroxysteroid dehydrogenase: Tissue specific protector of the mineralocorticoid receptor. Lancet 2:986–989, 1988 Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, Evans RM: Cloning of human mineralocorticoid receptor complementary DNA: Structural and functional kinship with the glucocorticoid receptor. Science 237:268–275, 1987 Krozowski ZS, Funder JW: Renal mineralocorticoid receptors and hippocampal corticosterone-binding species have identical intrinsic steroid specificity. Proc Natl Acad Sci USA 80:6056–6060, 1983 Holbrook MM, Dale SL, Melby JC: Peripheral plasma steroid concentrations in rats sacrificed by anoxia. J Steroid Biochem 13:1355–1358, 1980 Naray-Fejes-Toth A, Fejes-Toth G: Subcellular localization of the type 2, 11beta-hydroxysteroid dehydrogenase: A green fluorescent protein study. J Biol Chem 271:15436–15442, 1996 Wilson RC, Krozowski ZS, Li K, Obeyesekere VR, RazzaghyAzar M, Harbison MD, Wei JQ, Shackleton CH, Funder JW, New MI: A mutation in the HSD11␤2 gene in a family with apparent mineralocorticoid excess. J Clin Endocrinol Metab 80:2263– 2266, 1995 Mune T, Rogerson FM, Nikkila H, Agarwal AK, White P: Human hypertension caused by mutations in the kidney isozyme of 11 beta-hydroxysteroid dehydrogenase. Nat Genet 10:394–399, 1995 Ulick S, Levine LS, Gunczler P, Zanconato G, Ramirez LC, Rauh W, Rosler A, Bradlow HL, New MI: A syndrome of apparent mineralocorticoid excess associated with defects in the peripheral metabolism of cortisol. J Clin Endocrinol Metab 49:757– 764, 1979 Bush IE, Hunter SA, Meigs RA: Metabolism of 11-oxygenated steroids: Metabolism in vitro by preparations of liver. Biochem J 107:239–258, 1968 Monder C, White PC: 11 beta-hydroxysteroid dehydrogenase. Vitam Horm 47:187–271, 1993 Morris DJ, Berek JS, Davis RP: The physiological response to aldosterone in adrenalectomized and intact rats and its sex dependence. Endocrinology 92:989–993, 1973 Barger AC, Berlin RD, Tulenko JK: Infusion of aldosterone, 9-fluorohydrocortisone and antidiuretic hormone into the renal artery of normal and adrenalectomized, unanesthetized dogs: Ef-

14.

15.

16.

17.

18. 19. 20. 21. 22. 23. 24. 25. 26.

1373

fect on electrolyte and water excretion. Endocrinology 62:804–809, 1958 Souness GW, Morris DJ: The antinatriuretic and kaliuretic effects of the glucocorticoids corticosterone and cortisol following pretreatment with carbenoxolone sodium (a liquorice derivative) in the adrenalectomized rat. Endocrinology 124:1588–1590, 1989 Souness GW, Morris DJ: The “mineralocorticoid-like” actions conferred on corticosterone by carbenoxolone are inhibited by the mineralocorticoid receptor (type I) antagonist RU28318. Endocrinology 129:2451–2456, 1991 Kenyon CJ, Saccocio NA, Morris DJ: The “mineralocorticoidlike” actions conferred on corticosterone by carbenoxolone are inhibited by the mineralocorticoid receptor (type I) antagonist RU 28318. Endocrinology 129:2451–2456, 1991 Uete T, Venning EH: The effect of cortisone, hydrocortisone, and 9␣-fluoro-16␣-hydroxy-hydrocortisone on the action of deoxycorticosterone and aldosterone with respect to electrolyte excretion. Endocrinology 67:52–64, 1960 Liddle GW: Effects of anti-inflammatory steroids on electrolyte metabolism. Ann NY Acad Sci 82:855–867, 1959 Brem AS, Metheson KL, Barnes JL, Morris DJ: 11-Dehydrocorticosterone, a glucocorticoid metabolite, inhibits aldosterone action in toad bladder. Am J Physiol 261(5 Pt 2):F873–F879, 1991 Brem AS, Metheson KL, Barnes JL, Morris DJ: Effect of carbenoxolone on glucocorticoid metabolism and Na transport in toad bladder. Am J Physiol 257(4 Pt 2):F700–F704, 1989 Souness GW, Myles K, Morris DJ: Other physiological considerations of protective mechanisms of mineralocorticoid action. Steroids 59:142–147, 1994 Brem AS, Matheson KL, Syed L, Morris DJ: Activity of 11 betahydroxysteroid dehydrogenase in toad bladder: Effects of 11-dehydrocorticosterone. Am J Physiol 264(5 Pt 2):F854–F858, 1993 Alberti KGMM, Sharp WG: Identification of four types of steroid by their interaction with mineralocorticoid receptors in the toad bladder. J Endocrinol 48:563–574, 1970 Pearce D, Yamamoto KR: Mineralocorticoids, glucocorticoids, receptors and response elements. Science 259:1132–1133, 1993 Latif S, Hartmann LR, Souness GW, Morris DJ: Possible endogenous regulators of steroid inactivating enzymes and glucocorticoid-induced Na⫹ retention. Steroids 59:352–356, 1994 Parks DJ, Blanchard SG, Bladsoe RK, Chandra G, Consler TG, Kliewer SA, Stimmel JB, Willson TM, Zavacki AM, Moore DD, Lehmann JM: Bile acids: Natural ligands for an orphan nuclear receptor. Science 284:1365–1368, 1999