Evidence for a Novel Peripheral Action of Leptin asa Metabolic Signal to the Adrenal GlandLeptin Inhibits Cortisol Release DirectlyStefan R. Bornstein, Katja Uhlmann, Andrea Haidan, Monika Ehrhart-Bornstein,and Werner A. Scherbaum

The crucial role of glucocorticoids in obesity andinsulin resistance and the actions of the OB proteinleptin on the hypothalamic-pituitary-adrenal (HPA)axis suggest that there is an important interaction ofleptin with the glucocorticoid system. Therefore, wedesigned a study to test the effect of leptin directly onadrenocortical steroidogenesis. Primary cultures ofbovine adrenocortical cells were incubated withincreasing concentrations (10-1,000 ng/ml) of recom-binant mouse leptin for 24 h, and the effects of leptinon basal and ACTH-stimulated cortisol secretion weredetermined. The accumulation of P450 17a mRNA fol-lowing incubation with ACTH (10 nmol/1) and leptin(10-1,000 ng/ml) was analyzed by Northern blot.Adrenocortical cells were characterized by immuno-histochemical staining for 17a-hydroxyprogesterone.Leptin (10-1,000 ng/ml) inhibited basal and ACTH-stimulated cortisol release. At a concentration thatoccurs in obese individuals in vivo (100 ng/ml), itreduced basal cortisol secretion to 52.7 ± 37% (mean ±SE). The rise in cortisol secretion following maximalACTH stimulation (10 nmol/1) was blunted to 55.2 ±27%. At more physiological concentrations of ACTH(0.1 nmol/1), the inhibition of cortisol release by coin-cubation with low doses of leptin (10 ng/ml) was evenmore pronounced, leading to a reduction to 32.8%(1,248 ± 134 vs. 410 ± 157 nmol/1). Addition of OB pro-tein (10-1,000 ng/ml) led to a dose-dependent reductionof ACTH-stimulated cytochrome P450 17a mRNA accu-mulation (from 80 to 45%), suggesting that leptin reg-ulates adrenal steroidogenesis at the transcriptionallevel. These data clearly demonstrate that leptininhibits cortisol production in adrenocortical cells andtherefore appears to be a metabolic signal that directlyacts on the adrenal gland. Diabetes 46:1235-1238,1997

From the Department of Internal Medicine III (S.R.B., K.U., A.H., M.E.-B.),University of Leipzig, Leipzig; and the Diabetes Research Institute at the Uni-versity of Diisseldorf (W.A.S.), Diisseldorf, Germany.

Address correspondence and reprint requests to Dr. Stefan R. Born-stein, Department of Internal Medicine III, University of Leipzig, Phillip-Rosenthal-Str. 27, 04103 Leipzig, Germany.

Received for publication 24 March 1997 and accepted in revised form 30April 1997.

CRH, corticotropin-releasing hormone; HPA, hypothalamic-pituitary-adrenal; NPY, neuropeptide Y.

Adrenal glucocorticoid production plays a funda-mental role in obesity and insulin resistance, andhypercortisolism is common in geneticallyobese fa/fa rats and db/db and ob/ob mice as well

as obese humans with insulin resistance (1-4). Adrenalectomyreduces or prevents the metabolic changes in obese animalmodels (5,6).

Leptin, a protein secreted by adipocytes (7), is an adipostatichormone that decreases food intake and body weight in obeseand lean animals (8). There is a clear reciprocal relationshipbetween leptin and corticosterone levels, and leptin appearsto be involved in the regulation of the diurnal pattern of cor-ticosterone secretion (9). In addition, administration of leptinreduces corticosterone levels in ob/ob mice (10).

It is well established that leptin acts at the level of thehypothalamus (11-13). However, recent studies suggest addi-tional important peripheral actions of leptin involving bothmodulation of insulin action on liver cells (14) and productionof steroid in the ovary (15). hi the light of these data, whenconsidering the crucial role of the adrenal in obesity andinsulin resistance, the question arises whether leptin candirectly influence adrenal glucocorticoid production. There-fore, we designed a study to test the direct effect of leptin onadrenocortical steroid production. A dose-dependent effectof leptin concentration on basal and ACTH-stimulated corti-sol release in isolated bovine adrenocortical cells was inves-tigated. A possible long-term effect of leptin on the accumu-lation of cytochrome P450 enzyme 17a-hydroxylase wasinvestigated by Northern blot analysis.

RESEARCH DESIGN AND METHODSReagents. Unless otherwise indicated, all reagents were purchased fromSigma (Munich, Germany). Recombinant murine leptin was obtained fromPeproTech (London, U.K.).Cell preparation, cell culture, and stimulation. Bovine adrenal glandswere obtained from freshly slaughtered 8- to 15-month-old steer and were iso-lated as previously reported (16). The cells were seeded on 24-well plates(200,000 cells per well) for cortisol measurement, on chamber slides (Nunc;50,000 cells per well) for immunohistochemistry, and on 6-well plates (106 cellsper well) for RNA isolation. The medium was replaced every 24-48 h. Exper-iments were performed after confluency reached at least 80%, after 3-4 days.The cells were washed and incubated for 24 h with serum-free medium con-taining ascorbic acid (10~7 mol/1), transferrin (0.001% wt/vol), and bacitracin(0.01% wt/vol). For measurement of cortisol, supernatants were removed andkept frozen at -20°C until assayed.Radioimmunoassay. Cortisol concentrations in the incubation media weremeasured by radioimmunoassay using a Biermann Cortisol-RIA kit (Bad

DIABETES, VOL. 46, JULY 1997 1235

LEPTIN AND CORTISOL RELEASE

Nauheim, Germany), which has a sensitivity of 5.5 nmol/1 and the followingcross-reactivities: cortisol 100%, prednisolone 76%, 11-deoxycortisol 11.4%,prednisone 2.3%, and other steroids <1%. It has intra- and interassay variationsof 5.1 and 6.4%, respectively.RNA isolation and Northern blot analysis. Total RNA was isolated fromadrenocortical cells using the RNAzol B-RNA-isolation kit from AGS (Heidel-berg, Germany). Electrophoresis and hybridization were performed as previ-ously described (17). Hybridization signals in the blots were normalized to theamount of 28S RNA on the nylon membrane and analyzed quantitatively by den-sitometric scanning.Immunohistochemistry. Immunostaining was performed by the avidin-biotin technique using the UniTect rabbit immunohistochemistry detection sys-tem (Dianova, Hamburg, Germany) as previously described (17). Cells wereincubated for 1 h at 37°C with a 1:10 dilution of the rabbit antiserum to 17a-hydroxyprogesterone (Sigma).Statistics. Treatments were administered in quadruplicate, and each exper-iment was repeated a minimum of three times. Data from the cell culture exper-iments are expressed as percentages of basal secretion (except in Fig. 2(7), cal-culated as ratio of stimulated to basal cortisol levels for each experiment.Results were expressed as means ± SE, and statistical significance was deter-mined by analysis of variance. P < 0.05 was considered significant.

RESULTS

Confluent monolayer cultures of bovine adrenocortical cellswere characterized by immunohistochemical staining with anantibody against 17a-hydroxyprogesterone. There was norelevant contamination with chromaffin cells, fibroblasts, orother non-steroid-producing cells (Fig. 1). Cultures wereincubated for 24 h with increasing doses of OB protein (lep-tin; 10-1,000 ng/ml). Basal rates of cortisol secretiondecreased in a dose-dependent manner to 78.5 ± 12.2% at 10ng/ml, 52.8 ± 37.2% at 100 ng/ml, and 25.3 ± 23.0% at 1,000ng/ml (Fig. 2A). Basal cortisol levels (42 ± 26.4 nmol/1) wereenhanced to 4,100% (1,738 ± 93.4 nmol/1) following a maximalstimulation with 10 nmol/1 ACTH. Leptin led to a dose-depen-dent (10-1,000 ng/ml) reduction of ACTH-stimulated cortisolrelease to 55.2 ± 27.1% (from 1,737 ± 208 to 958 ± 259 nmol/1)at 100 ng/ml leptin and to 47.6 ± 14.7% at 1,000 ng/ml (Fig. 2B).At more physiological concentrations of ACTH (0.1 nmol/1),the inhibition from coincubation with OB protein (10 ng/ml)was even more pronounced: ACTH-stimulated cortisolrelease was reduced to 32.8 ± 12.6% (from 1,248 ± 134 to 410±158 nmol/1) (Fig. 2(7). To see if leptin regulates steroidoge-nesis on the transcriptional level, the accumulation of P45017a mRNA was analyzed by Northern blot (Fig. 3). ACTH ina concentration of 10 nmol/1 led to a sixfold accumulation ofP450 17a mRNA. Addition of leptin (10-1,000 ng/ml) led to adose-dependent reduction (from 83 to 47%) of ACTH-stimu-lated P450 17a mRNA accumulation.

DISCUSSIONThere is a well-established relationship between the nutri-tional status of mammals and the activity of the hypothala-mic-pituitary-adrenal (HPA) axis (4,18). Leptin, a newly dis-covered adipostatic hormone (7), inhibits central appetitestimulation and regulates the HPA axis at the level of thehypothalamus (19). In our study, we provide the first evidencethat leptin can also act on the HPA axis peripherally at thelevel of the adrenal gland. Leptin at concentrations thatoccur in vivo led to a dose-dependent reduction of basalcortisol release from adrenocortical cells in primary cul-ture. In addition, leptin blunted the ACTH-induced rise in cor-tisol release.

Elevated plasma cortisol levels are found in patients andanimal models with obesity, glucose intolerance, and hyper-

FIG. 1. Confluent primary cultures of bovine adrenocortical cellswere characterized by immunohistochemical staining with an anti-body against 17a-hydroxyprogesterone. There was no relevant cont-amination with chromaffin cells, fibroblasts, or other non-steroid-producing cells. Bar = 20 um.

plasia of islets in the pancreas (1-3,20). Patients with Cush-ing's syndrome gain weight and have impaired glucose tol-erance or develop secondary diabetes with insulin resistance(21). Likewise, exogenous administration of glucocorticoidsresults in the metabolic characteristics of obesity, includinghyperinsulinemia, p-cell hyperplasia, and insulin resistance(22,23). In addition, stroma cells in human adipose tissuedevelop the characteristic features of adipocytes after stim-ulation with glucocorticoids (24). Adrenalectomy can partiallynormalize or prevent the weight gain and metabolic changesin obesity (8,9,11,12), and treatment of adrenalectomizedanimals with glucocorticoids results in reappearance of theobesity syndrome (5,6,25). These facts highlight the crucialrole of adrenal steroids in the development of obesity andinsulin resistance, and it becomes clear that any factorsinvolved in weight regulation will also influence adrenalsteroid production. The inhibiting effects of leptin on neu-ropeptide Y (NPY) gene expression and secretion led to theproposal that NPY, a potent stimulator of feeding and of theHPA axis, was the major mediator of the actions of OB pro-tein (10,26). However, mice totally deficient in NPY geneexpression have normal body weight and respond well to lep-tin, suggesting that other important control mechanismsshould exist (27). Several arguments suggest that an inhibi-tion of the HPA axis by leptin directly at the level of theadrenal gland, in addition to a central action, may be of phys-iological relevance.

1236 DIABETES, VOL. 46, JULY 1997

S.R. BORNSTEIN AND ASSOCIATES

120

basal 10 100 1000

leptin (ng/ml)

.2 g2i,4! 120

0 10 100 100010 nM ACTH + leptin (ng/ml)

Hi

c/>

nm

ol/L

(:

cort

iso

l

2000

1500

1000

500

0

- - O - ACTH (control)

I: D " ACTH + 10 ng/ml leptin *

[ / ^

T I I I I

0 0.01 0.1 1.0 10

ACTH (nM)

FIG. 2. Confluent bovine adrenocortical cells were incubated withincreasing doses of leptin (10-1,000 ng/ml) for 24 h. A: the basal rate(42 ± 26 nmol/1 = 100%) of cortisol secretion decreased in a dose-dependent manner. B: leptin caused a dose-dependent (10-1,000ng/ml) reduction of cortisol secretion after coincubation with 10nmol/1 ACTH. C: incubation with increasing doses (0.01-10 nmol/1) ofACTH caused a dose-dependent increase in cortisol production, whichwas blunted by 10 ng/ml of leptin. Concentration of cortisol in thesupernatants was determined by radioimmunoassay in quadruplicatesamples from three to five different cell preparations. Results areexpressed as means ± SE. *P 0.05 was considered significant.

First, we found that leptin concentrations that occur in vivo(28) decrease adrenocortical cortisol release. The adrenalgland itself is embedded in adipose tissue, and local con-centrations of leptin may even be higher.

Second, our finding that leptin inhibits accumulation ofP450 17a mRNA provides evidence for a long-term effect on

o

XHO

A

c

a

E

coT"

Xh-

o

B

c

ao

/ml

I

G)coo"*"

X1 -O

C

c

a~~E

cooo

XHO

D

- 28S

- 18S

FIG. 3. Comparison of the effects of leptin (10-1,000 ng/ml) andACTH (10 nmol/1) on the accumulation of mRNA encoding P450 17«enzyme. Total mRNA was isolated from adrenocortical cells after 24-h incubation with ACTH (A) and coincubation with ACTH and leptin:10 ng/ml (5), 100 ng/ml (C), and 1,000 ng/ml (D). Blots were normal-ized to 28S and analyzed by densitometric scanning. Leptin reducedACTH-stimulated accumulation of P450 17a mRNA from 80 to 45%.

adrenal steroidogenesis. Although the rate-limiting step insteroid hormone biosynthesis is the side-chain cleavage ofcholesterol (29), 17a-hydroxylase is a major regulatoryenzyme; P450 17a mRNA expression is regulated morerapidly and enzyme activity is increased more by ACTH thanby other steroidogenic enzymes (30). This may suggest thatleptin regulates the HPA axis acutely at the level of the hypo-thalamus and that it chronically depresses steroidogenesisand the molecular expression of steroid enzymes at the levelof the adrenal.

Third, diurnal variations in adrenal steroidogenesis donot seem to be directly related to plasma ACTH concentra-tions (31,32), whereas in rats a strict reciprocal diurnal rela-tionship exists between leptin and corticosterone levels (9).It will be interesting to see if leptin is responsible for this dis-crepancy, which possibly involves other identified pituitaryor extrapituitary mechanisms that can influence adrenalsteroidogenesis (33-35).

A peripheral inhibition of glucocorticoids by leptin at thelevel of the adrenal gland is in accordance with the role of cor-

DIABETES, VOL. 46, JULY 1997 1237

LEPTIN AND CORTISOL RELEASE

ticotropin-releasing hormone (CRH) in weight regulation.On the one hand, CRH inhibits feeding in rodents (8,36) andreduces weight and insulin levels. Since leptin increases CRHmRNA expression in the paraventricular nucleus of the hypo-thalamus, the activity of OB protein appears partly to bemediated through these increased CRH levels (8). On theother hand, CRH is the major stimulator of the HPA axis andleads to the release of glucocorticoids, which have the oppo-site effects, resulting in weight gain and hyperinsulinemia.Therefore, a central activation of CRH and a peripheral glu-cocorticostatic action of leptin would be complementary toeach other. In this context, it is of interest that adrenalectomyprevents the obesity syndrome produced by chronic centralNPY infusion in normal rats and that the elevated CRHexpression in these animals does not influence food intake,body weight gain, or insulinemia (37). As predicted, it nowbecomes clear that a fundamental physiological function inthe maintenance of energy balance appears to be regulatedby a multidimensional system with overlapping control path-ways (26). The fact that leptin is a direct inhibitor of adrenalcortisol production is important for our understanding ofboth adrenal stress regulation and the role of the adrenal inmetabolic diseases.

ACKNOWLEDGMENTS

This work was supported by the Deutsche Forschungsge-meinschaft (Heisenberg grant to S.R.B.) and the Interdisci-plinary Center for Clinical Research at the University ofLeipzig (IDZL Projects Bl to M.E.-B. and S.R.B. and B3 toW.A.S.).

REFERENCES1. Coleman DL: Classical diabetes models: past lessons and potential new ther-

apies. In Frontiers in Diabetes Research: Lessons from Animal Diabetes II.Shafir E, Renoid AE, Eds. London, Libbey, 1988, p. 253-256

2. Terrettaz J, AssimacopoulosnJeannet F, Jeanrenaud B: Severe hepatic andperipheral insulin resistance as evidenced by euglycemic clamps in geneti-cally obese fa/fa rats. Endocrinology 118:674-678,1986

3. Bray GA, York DA: Hypothalamic and genetic obesity in experimental animals:an autonomic and endocrine hypothesis. Physiol Rev 59:719-809, 1979

4. Bjorntorp P: Endocrine abnormalities of obesity. Metabolism 44:21-23,19955. Ohshima K, Shargill NS, Chan TM, Bray GA: Adrenalectomy reverses insulin

resistance in muscle from obese (ob/ob) mice. Am J Physiol 246:E193-E197,1984

6. Freedman MR, Horwitz BA, Stern JS: Effect of adrenalectomy and gluco-corticoid replacement on development of obesity. Am J Physiol250:R595-R607,1986

7. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM: Positionalcloning of the mouse obese gene and its human homologue. Nature372:425-432, 1994

8. Campfield LA, Smith FJ, Burn P: The OB protein (leptin) pathway: a linkbetween adipose tissue mass and central neural networks. Horm Metab Res28:619-632, 1996

9. Ahima RS, Prabakaran D, Mantzoros C, Qu D, Lowell B, Maratos-Flier E, FlierJS: Role of leptin in the neuroendocrine response to fasting. Nature382:250-252, 1996

10. Stephens TW, Basinski M, Bristow PK, Bue-Valleskey JM, Burgett SG, CraftL, Hale J, Hoffmann J, Hsiung HM, Kriauciunas A, MacKellar W, Rosteck PRJ,Schoner B, Smith D, Tinsley FC, Zhan X-Y, Heiman M: The role of neu-ropeptide Y in the antiobesity action of the obese gene product. Nature377:530-532, 1995

11. Tartaglia LA, Dembski M, Weng X, Deng N, Culpepper J, Devos R, RichardsGJ, Campfield LA, Clark FT, Deeds J, Muir C, Sanker S, Moriaty A, Mure KJ,Smutko JS, Mays GG, Woolf EA, Monroe CA, Tepper RI: Identification andexpression cloning of a leptin receptor, OB-R. Cell 83:1263-1271, 1995

12. Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG, Lee JI, Fried-man JM: Abnormal splicing of the leptin receptor in diabetic mice. Nature

379:632-635, 199613. Vaisse C, Halaas JL, Horvath CM, Darnell JE Jr, Stoffel M, Friedman JM: Lep-

tin activation of stat3 in the hypothalamus of wild-type and ob/ob mice butnot db/db mice. Nature Genet 14:95-97, 1996

14. Cohen B, Novick D, Rubinstein M: Modulation of insulin activities by leptin.Science 274:1185-1188, 1996

15. Zachow RJ, Magoffin DA: Direct intraovarian effects of leptin: impairment ofthe synergistic action of insulin-like growth factor-I on follicle-stimulating hor-mone-dependent estradiol-17-beta production by rat ovarian granulosa cells.Endocrinology 138:847-850,1997

16. Guse-Behling H, Ehrhart-Bornstein M, Bornstein SR, Waterman MR,Scherbaum WA, Adler G: Regulation of adrenal steroidogenesis by adrena-line: expression of cytochrome P450 genes. JEndocrinol 135:229-237,1992

17. Glasow A, Breidert M, Haidan A, Anderegg U, Kelly PA, Bornstein SR: Func-tional aspects of the effect of prolactin (PRL) on adrenal steroidogenesis anddistribution of the PRL receptor in the human adrenal gland. J ClinEndocrinol Metab 81:3103-3111,1996

18. Akana SF, Strack AM, Hanson ES, Dallman MF: Regulation of activity in thehypothalamo-pituitary-adrenal axis is integral to a larger hypothalamic sys-tem that determines caloric flow. Endocrinology 135:1125-1134,1994

19. Schwartz MW, Seeley RJ, Campfield LA, Burn P, Baskin DG: Identification oftargets of leptin action in rat hypothalamus. J Clin Invest 98:1101-1106,1996

20. Thorn GW, Reinhold AE, Cahill GF: The adrenal and diabetes. Some inter-actions and intercorrelations. Diabetes 8:337-351,1959

21. Nosadini R, Del Prato S, Tiengo A, Valerio A, Muggeo M, Opocher G, ManteroF, Duner E, Marescotti C, Mollo F, Belloni F: Insulin resistance in Cushing'ssyndrome. J Clin Endocrinol Metab 57:529-536,1983

22. Guillaume-Gentil C, Assimacopoulos-Jeannet F, Jeanrenaud B: Involvementof non-esterifled fatty acid oxidation in glucocorticoid-induced peripheralinsulin resistance in vivo in rats. Diabetologia 36:899-906, 1993

23. McMahon M, Gerich J, Rizza R: Effects of glucocorticoids on carbohydratemetabolism. Diabetes Metab Rev 4:17-30,1988

24. Hauner H, Entenmann G, Wabisch M, Gaillard D, Ailhaug G, Negrel R, Pfeif-fer EF: Promoting effects of glucocorticoids on the differentiation of humanadipocyte precursor cells cultured in a chemically defined medium. J ClinInvest 84:1663-1670,1989

25. Debons AF, Zurek LD, Tse CS, Abrahamsen S: Central nervous system con-trol of hyperphagia in hypothalamic obesity: dependence on adrenal gluco-corticoids. Endocrinology 118:1678-1681,1986

26. Caro JF, Sinha MK, Kolaczynski JW, Zhang PL, Considine RV: Leptin: the taleof an obesity gene. Diabetes 45:1455-1462, 1996

27. Erickson JC, Clegg KE, Palmiter RD: Sensitivity to leptin and susceptibilityto seizures of mice lacking neuropeptide Y. Nature 381:415-421, 1996

28. Rohner-Jeanrenaud F, Jeanrenaud B: Obesity, leptin, and the brain. NEnglJMed 334:324-325, 1996

29. DuBois RN, Simpson ER, Kramer RE, Waterman MR: Induction of synthesisof cholesterol side chain cleavage cytochrome P-450 by adrenocorticotropinin cultured bovine adrenocortical cells. JBiol Chem 256:7000-7005, 1981

30. Zuber MX, John ME, Okamura T, Simpson ER, Waterman MR: Bovineadrenocortical cytochrome P-45017a. J Biol Chem 261:2475-2482, 1986

31. Krieger DT: Plasma ACTH and corticosteroids. In Endocrinology. Vol. 2.Groot LJ, Cahill GF, Martini L, Nelson DH, Odell WD, Potts JT, SteinbergerE, Winegard AP, Eds. New York, Grune and Stratton, 1979, p. 1139-1156

32. Fehm HL, Klein E, Holl R, Voigt KH: Evidence for extrapituitary mecha-nisms mediating the morning peak of plasma cortisol in man. J ClinEndocrinol Metab 58:410-414,1984

33. Bornstein SR, Ehrhart-Bornstein M, Scherbaum WA, Pfeiffer EF, Hoist JJ:Effects of splanchnic nerve stimulation on the adrenal cortex may be mediatedby chromaffin cells in a paracrine manner. Endocrinology 127:900-906,1990

34. Bornstein SR, Gonzalez-Hernandez JA, Ehrhart-Bornstein M, Adler G,Scherbaum WA: Intimate contact of chromaffin and cortical cells within thehuman adrenal gland forms the cellular basis for important intraadrenalinteractions. J Clin Endocrinol Metab 78:225-232, 1994

35. Ehrhart-Bornstein M, Bornstein SR, Scherbaum WA: Sympathoadrenal sys-tem and immune system in the regulation of adrenocortical function. Eur JEndocrinol 135:19-26, 1996

36. Rohner-Jeanrenaud F: A neuroendocrine reappraisal of the dual-centrehypothesis: its implications for obesity and insulin resistance. Int J Obes RelatMetab Disord 19:517-534, 1995

37. Sainsbury A, Cusin I, Rohner-Jeanrenaud F, Jeanrenaud B: Adrenalectomy pre-vents the obesity syndrome produced by chronic central neuropeptide Y infu-sion in normal rats. Diabetes 46:209-214, 1997

1238 DIABETES, VOL. 46, JULY 1997