Urinary free cortisone and the assessment of 11β-hydroxysteroid dehydrogenase activity in man

Clinical Endocrinology (1996) 45, 605–611

Urinary free cortisone and the assessment of11¯-hydroxysteroid dehydrogenase activity in man

Mario Palermo*, Cedric H. L. Shackleton*,Franco Mantero y and Paul M. Stewart z*Children’s Hospital Oakland Research Institute,747 52nd Street, Oakland, CA 94609, USA, yCattedra diEndocrinologia, University of Ancona, Ancona, Italy andzDepartment of Medicine, Queen Elizabeth Hospital,Edgbaston, Birmingham, UK

(Received 14 February 1996; returned for revision 3 May 1996;finally revised 5 June 1996; accepted 29 August 1996)

Summary

OBJECTIVE Two isoforms of 11 ¯-hydroxysteroiddehydrogenase (11 ¯-HSD) catalyse the interconver-sion of cortisol to hormonally inactive cortisone;defects in the 11 ¯-HSD2 isoform result in hyperten-sion. The kidney, expressing high levels of 11 ¯-HSD2,is the principal source of cortisone in man. We havevalidated the measurement of urinary free cortisone(UFE) excretion in normals and in patients with dis-orders of the pituitary-adrenal axis in an attempt tomore accurately measure the activity of 11 ¯-HSD2in vivo .SUBJECTS Forty-one normal adults, 12 normal chil-dren <12 years of age, 15 patients with Cushing’ssyndrome, 12 with hypopituitarism on replacementhydrocortisone, 12 with the syndrome of apparentmineralocorticoid excess (AME) and 7 volunteersconsuming liquorice.MEASUREMENTS A complete 24-hour urine collec-tion was analysed by gas chromatography/massspectrometry for ‘A-ring’ reduced cortisol and corti-sone metabolites, i.e. tetrahydrocortisols (THF andallo-THF) and tetrahydrocortisone (THE). In addition,urinary free cortisol (UFF) and urinary free cortisonewere quantified using deuterium-labelled internalstandards.RESULTS In normal adults and children, UFE excre-tion exceeded that of UFF (UFF 30 .4� 2.4¹g/24h(mean �SE), UFE 54.6�4.1¹g/24h, adults) (for con-version to nmol/24h multiply E by 2 .78 and F by 2 .76respectively). Thus the normal UFF/UFE ratio was

0.54�0.05 in contrast to the (THF �allo-THF)/THEratio of 1 .21� 0.06. UFE excretion was normal inhypopituitary patients on replacement hydrocorti-sone. Although UFE was elevated in all forms ofCushing’s syndrome, the UFF/UFE ratio was grosslyelevated in patients with the ectopic ACTH syndrome(14.0� 6.7, n�6). UFE was below the lower limit of theassay (<1¹g/24h) in most patients with the so-calledtype 1 variant of AME and significantly reduced in 4patients described as having the type 2 variant of AME(10.5� 3.5¹g/h, P<0.05) and in 7 volunteers consum-ing liquorice (26 .8� 10.0¹g/24h, P<0.01). In ectopicACTH syndrome, AME, and liquorice ingestion theUFF/UFE ratio was more deranged than the (THF �

allo-THF)/THE ratio.CONCLUSION In normals the discrepant THF � allo-THF/THE and UFF/UFE ratio suggests that much moreof the UFE is derived from the kidney. Reduction inUFE excretion is seen following liquorice ingestionand in both variants of AME, though it is more pro-found in AME1. The high UFF/UFE ratio in the miner-alocorticoid excess state seen in the ectopic ACTHsyndrome is compatible with substrate-saturation ofrenal 11¯-HSD2. The measurement of UFE and theUFF/UFE ratio is a significant advance in the analysisof human 11 ¯-HSD activity in vivo ; in particular, theUFF/UFE ratio appears to be a more sensitive indexthan the (THF �allo-THF)/THE ratio of renal 11 ¯-HSD2activity.

The interconversion of cortisol (F) to hormonally inactive cortisone(E) is catalysed by two distinct isoforms of 11�-hydroxysteroiddehydrogenase (11�-HSD). 11�-HSD1 is localized predomi-nantly to the glucocorticoid target tissues, liver, lung, gonad,cerebellum and pituitary and is a low affinity NADP(H)-dependent dehydrogenase/oxo-reductase (Tanninet al., 1991;Mooreet al., 1993; Whorwoodet al., 1995). In contrast, 11�-HSD2is a high affinity, NAD-dependent dehydrogenase foundprincipally in the placenta (Brownet al., 1993) and miner-alocorticoid target tissues, kidney and colon (Albistonet al.,1994; Stewartet al., 1994; Whorwoodet al., 1995). Thebiological importance of 11�-HSD relates to its role inprotecting both the glucocorticoid and mineralocorticoidreceptors (GR, MR) from cortisol excess.In vitro, the MRhas an equal affinity for cortisol and aldosterone (Arrizaet al.,

605# 1996 Blackwell Science Ltd

Correspondence: Paul M. Stewart, Department of Medicine, QueenElizabeth Hospital, Edgbaston, Birmingham B15 2TH, UK. Fax 44121 627 2384.

1987); 11�-HSD by inactivating F to E, enables aldosterone tobind to the MR thereby facilitating normal specificity for theMR (Edwardset al., 1988; Funderet al., 1988). Congenitaldeficiency of 11�-HSD or the syndrome of apparent miner-alocorticoid excess (AME) is known to be due to a defect in thehuman type 2 11�-HSD gene (Wilsonet al., 1995; Muneet al.,1995; Stewartet al., 1996), but defective 11�-HSD activity hasbeen postulated in forms of Cushing’s syndromes (Ulicket al.,1992b; Walkeret al., 1992; Stewartet al., 1995), polycysticovary syndrome (Rodinet al., 1994) and essential hypertension(Walker et al., 1993; Soroet al., 1995).

An obvious prerequisite for these studies is the ability toassess 11�-HSD activity in vivo. This has largely beenperformed by measuring the urinary ratio of ‘A’-ring reducedmetabolites of cortisol and cortisone (tetrahydrocortisols (THF,allo-THF) and tetrahydrocortisone (THE)) (Shackleton, 1993).However, the principal site of A-ring cortisol reduction is theliver, and the (THF�allo-THF)/THE ratio presumably reflectsthe set-point of F to E conversion as seen by the liver.This in turn will be dictated by the activity of type 1 and type 211�-HSD in any tissue expressing these isoforms. Because theprincipal site of 11�-HSD2 expression is the kidney(Hellman et al., 1971; Whitworth et al., 1989), we havedeveloped a gas chromatography/mass spectroscopy (GC/MS)method for the determination of urinary free E in an attempt tomeasure more accurately the activity of this novel isoform.

Patients and methods

The following subjects were studied.

Normal controls

Twenty adult females (age 44.7�1.9 years (mean�SE)),21 adult males (age 39.3�2.2 years) and 12 children (age 8.5�1.2years, all<12 years of age, 6 male) on no regular medicationcompleted a full 24-hour urine collection.

Cushing’s syndrome patients

Six patients with the ectopic ACTH syndrome (age 56.5�4.5years), 7 patients with pituitary-dependent Cushing’s syndrome(age 41.0�3.8 years) and 2 patients with cortisol secretingadrenal adenomas (age 38 and 62 years) were studied. Clinicaland biochemical tests on these patients have recently beenpublished (Stewartet al., 1995); in brief, the differentialdiagnosis of Cushing’s syndrome was made on the basis ofpotassium and ACTH measurements, high dose dexamethasonesuppression and corticotrophin releasing hormone stimulation

tests. Pituitary and adrenal disease were confirmed surgically inevery case.

Patients with hypopituitarism

Twelve patients (6 male) (age 42.5�2.9 years) with long-standing hypopituitarism were studied whilst taking 30 mg/day ofhydrocortisone replacement (20 mg a.m., 10 mg p.m.) and againfollowing a reduction in daily hydrocortisone dose to 15 mg/day(10 mg a.m., 5 mg p.m.). The aim of this study was to assess theeffect of hydrocortisone replacement on cardiovascular functionin patients with hypopituitarism and its results have been reported(Dunne et al., 1995). All patients were fully replaced, whereappropriate, with other endocrine drugs (thyroxine, sex steroidsand DDAVP), but no patient received GH or fludrocortisone,each of which may alter cortisol metabolism (Weaveret al.,1994; Oelkerset al., 1994). Patients had been taking the abovedoses of hydrocortisone for at least 3 months prior to the urinecollection.

Apparent mineraolocorticoid excess

Eight patients with the so-called ‘type 1’ variant of AME (age7.0�2.1 years) and 4 patients with the ‘type 2’ variant (age18.0�5.1 years) were studied whilst taking no medication. Theclinical details of these patients have been previously reported(Shackleton & Stewart, 1990; Ulicket al., 1990; Manteroet al.,1994; Milford et al., 1995). The distinction between the type 1and type 2 variants is at present arbitrary but has been made byother authors on the basis of the THF�allo-THF/THE ratio,which is grossly elevated in the type 1 variant, but only mildlyelevated in the type 2 variant. The molecular basis for AMEtype 1 has been explained as a defect in the 11�-HSD2 genebut, as yet, the 11�-HSD2 gene has not been sequenced inpatients with the type 2 variant.

Liquorice ingestion

Seven normal males (age 33.3�2.8 years) consumed the activecomponent of liquorice, glycyrrhetinic acid, 240 mg/day for 45days. A full 24-hour urinary steroid profile was determined afterthis 45-day period.

Every subject studied had normal renal function, as judgedby a normal plasma creatinine.

Urinary steroid analysis

A full urinary steroid metabolite profile, including THF, allo-THF and THE was determined by gas chromatography/mass

606 M. Palermo et al.

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 45, 605–611

spectrometry as previously reported (Shackleton, 1993).Urinary free steroids were measured by a method suitable for Fand its 3-oxo-4-ene metabolites including E. For F and E we usedstable isotope-labelled internal standards ([9,11,12,12]2H4-cortisol and [9,12,12]2H3-cortisone respectively), kindlydonated by Dr Walter Hubbard, Johns Hopkins University,Baltimore. These standards were calibrated against accuratelyweighed and solubilized non-labelled standards by HPLCanalysis. Following the addition of the deuterated internalstandard cocktail to 5 ml of urine (0.12�g d3-cortisone, 0.18�gd4-cortisol), steroid extraction was carried out by Sep-pakC18 cartridges. To this extract 200 ng of stigmasterol andcholesteryl-butyrate were added as external standard andmethyloxime-trimethylsilyl ether were prepared according toestablished procedures. Following derivatization, the excessreagent was removed by Lipidex chromatography; the sampleswere automatically introduced in a Hewlett-Packard 5970mass spectrometer with a 15-m DB1 capillary column.Quantification was achieved by monitoring selected ions forthe analytes (fragment 531 for E and 605 for F) and internalstandards (m=Z 534 and 609 respectively). Relative peak areaswere determined and reported as�g/24 hours of the individualcompound. Validation of the method was performed bycomparing the areas under the curve of increasing amountsof the analyte to the ratio between the areas of differentamounts of the analyte and fixed concentration of theinternal standard (r � 0:998 for F and 0.999 for E).Coefficients of intra and inter-assay variability were<10%for both F and E.

Statistical comparisons between groups were made using theMann–Whitney U-test and comparisons between the samegroup using Student’s pairedt-test.

Results

In normal adults the urinary excretion of A-ring reducedcortisol metabolites was higher than that of the A-ring reducedcortisone metabolites (THF�allo-THF/THE 1.21�0.06,n� 41), yet the converse was true for UFF and UFE, with,on average, twice as much UFE excreted compared to UFF(UFF/UFE ratio 0.54�0.05) (Table 1). The same pattern wastrue for normal children (Table 2).

In 14 of the 41 adult subjects, cortols and cortoloneswere also measured; if the cortisol/cortisone metabolites areexpressed as (THF�allo-THF� �- and�-cortols)/(THE��- and�-cortolones) the ratio was 0.72�0.16 (0.65� 0.13 in females,0.80�0.19 in males). In adult normals, both the (THF� allo-THF)/THE and the UFF/UFE ratios were higher in males than infemales, but this was not statistically significant.

In states of cortisol excess, such as Cushing’s syndrome,the UFF/UFE ratio was significantly elevated compared tocontrols, despite an often impressive increase in the excretionof UFE. Thus in perhaps the most florid example ofhypercortisolism, the ectopic ACTH syndrome, UFE excre-tion rose tenfold above normal (Table 1, Fig. 1). UFF/UFEwas also slightly elevated in hypopituitary patients takinghydrocortisone 30 mg/day (but not on 15 mg/day), and thiswas due to an increase in UFF excretion (88.9�21.2�g/24hours on 30 mg/dayvs 40.1�6.4�g/day on 15 mg/day,P < 0:05) rather than to any decrease in UFE excretion.Conversely, in volunteers consuming glycyrrhetinic acid (theactive component of liquorice) 240 mg/day for 45 days, UFEwas significantly reduced (Table 1).

We have also studied 12 patients with the so-called ‘type 1and type 2 variants’ of AME. When compared to an age-matched

Urinary free cortisone and 11�-HSD 607


Table 1 Twenty-four-hour urinary excretion of free cortisol (UFF) and cortisone (UFE) depicted as mean�SE (�g/24 hours)

Patient group UFF UFE UFF/UFE (THF�allo-THF)/THE

Normal male (n� 21) 36.7� 3.3 60.9� 5.7 0.60� 0.07 1.30� 0.07Normal female (n� 20) 23.9� 3.0 48.0� 5.9 0.47� 0.05 1.15� 0.11Total normals (n� 41) 30.4� 2.4 54.6� 4.1 0.54� 0.05 1.21� 0.06Ectopic ACTH (n� 6) 6612�2075z 584�211z 14.0�6.7z 3.95�0.69zPituitary Cushing’s (n� 7) 293�116y 175� 58 1.86�0.42z 1.74� 0.24*Adrenal Cushing’s (n� 2) 264 299 0.75 1.7Hypopituitary 30 mg/day (n� 12) 88.9�21.2y 73.4� 10.6 1.22� 0.27* 2.30�0.12zHypopituitary 15 mg/day (n� 12) 40.1� 6.4 67.8� 37.4 0.65� 0.10 2.06�0.15zLiquorice 240 mg/day for 45 days (n� 7) 49.4� 12.2 26.8�10.0y 1.80�0.31y 2.31�0.27y

(For conversion to nmol/24 hours multiply UFF by 2.76 and UFE by 2.78). Values are shown for normal adult controls, patients with Cushing’ssyndrome, hypopituitary patients taking 30 and 15 mg/day hydrocortisone and volunteers taking liquorice 240 mg/day for 45 days. The UFF/UFE and(THF�allo-THF)/THE ratios are also shown.* P < 0:05, yP < 0:01, zP < 0:001vs total normals.

control group, the most striking defect in the type 1 variant wasthat UFE was below the detection limit for the method in allcases; UFF was also elevated, contributing to a grosslyelevated, but technically unrecordable, UFF/UFE ratio (Table2, Fig. 1). For the type 2 variant, the results were less striking.UFE was reduced when compared to both the child controls(P < 0:05) or adult control group (P < 0:01), but not to thesame extent as that seen in the type 1 variant; in every case UFEwas detectable.

Comparing the results of the UFF/UFE and the (THF�allo-THF)/THE ratios, a reasonable correlation was observed,though it was clear that any deviation from normal in the(THF�allo-THF)/THE ratio resulted in a much more markedchange in UFF/UFE ratio. Thus, expressed as a fold increaseabove normal control values, it was evident from patients withAME, liquorice ingestion and ectopic ACTH syndrome, that

the UFF/UFE ratio appeared to be a more sensitive index of11�-HSD2 activity than the (THF�allo-THF)/THE ratio(Fig. 1).

Discussion

A prerequisite for the investigation of 11�-HSD activity in thepathogenesis of human diseases such as hypertension, is theability to accurately measure enzyme activityin vivo. Untilnow this has been performed through estimation of plasmacortisol half-life using radioisotope studies (Zumoffet al.,1983; Stewart et al., 1988; Walker et al., 1993), themeasurement of plasma cortisol and cortisone (Whitworthet al., 1989; MacKenzieet al., 1990; Walkeret al., 1992) oron the measurement of urinary A-ring reduced cortisol andcortisone metabolites (Shackleton, 1993). All of these haveinherent problems; radioisotopic studies involve the adminis-tration of radioactivity, and because of a marked isotope effectgive unreliable results in patients with hypercortisolaemia.Plasma F/E and urinary (THF�allo-THF)/THE ratios arevaluable but reflect the global activity of 11�-HSD over thesampling time. It is now established that this, in turn, reflectsthe coordinated action of two distinct isoforms of 11�-HSD,11�-HSD1 acting predominantly as an oxo-reductase (i.e.conversion of E to F) and 11�-HSD2 as a high affinitydehydrogenase (F to E). 11�-HSD2 is the predominant, if notthe exclusive, isoform in human kidney (Whorwoodet al.,1995); it is this enzyme which confers specificity upon theMR, and a defect in its activity explains the syndrome ofapparent mineralocorticoid excess. Furthermore, plasma corti-sone falls in patients with renal failure and is reduced tovalues 1/10th of normal in patients who have undergonebilateral nephrectomy (Whitworthet al., 1989), confirmingearlier studies that the kidney is the principal source ofcortisone in man (Hellmanet al., 1971). Our hypothesis,therefore, was that the measurement of urinary free cortisone



Table 2 Twenty-four-hour urinary free cortisol (UFF) and cortisone (UFE) in normal adults and children<12 years of age compared to patients withtype 1 and type 2 variants of AME

Patient group UFF UFE UFF/UFE (THF�allo-THF)/THE

Normal adults (n� 41) 30.4� 2.4 54.6�4.1 0.54� 0.05 1.21� 0.06Normal children (n� 12) 9.1� 1.3 26.3�3.9 0.43� 0.08 0.79� 0.09Type 1 AME (n� 8) 23.3�18.1y <1:0 (0–5)z >23z 30.0�20.7zType 2 AME (n� 4) 31.0�5.2y 10.5�3.5* 3.5�1.0y 2.9� 0.44y

Results are given as mean�SE in �g/24 hours. (For conversion to nmol/24 hours multiply UFF by 2.76 and UFE by 2.78.) The UFF/UFE andTHF�allo-THF/THE ratios are also shown.* P < 0:05, yP < 0:01, zP < 0:001vsnormal children.

60

0Liquorice

Fo

ld in

cre

ase v

s c

on

tro

l

ectopicACTH

pituitaryCushings

hypopit15

35

55

50

45

40

30

25

20

15

hypopit30

AME2AME1

10

5

> 54

Fig. 1 A comparison of the&, UFF/UFE andd, (THF�allo-THF)/THE ratios expressed as fold increase over the normal control value.Note that for the patients with AME, controls were normal children,otherwise adults were used.

or the UFF/UFE ratio would be a more accurate measure of11�-HSD2 than would be the (THF�allo-THF)/THE ratio,which reflects 11�-HSD1 and 11�-HSD2 activities.

Our results would suggest that this is the case. Whilst the(THF�allo-THF)/THE ratio was invariably greater than 1,urinary free cortisone excretion usually exceeded UFF excre-tion with the result that, in normal subjects, the UFF/UFE ratiowas approximately 0.5. This result suggests that much of theUFE is indeed derived from the kidney. We have then evaluatedUFE in a series of clinical disorders. UFE was undetectable inpatients with established and characterized mutations in the11�-HSD2 gene (Wilsonet al., 1995; Muneet al., 1995;Stewartet al., 1996), the so-called ‘type 1’ variant of AME, andsignificantly reduced in patients with the ‘type 2’ variant. In theabsence of any sequence data on the 11�-HSD2 gene in the type2 variant, this division is arbitrary. In both of these variants,cortisol has been shown to be the offending mineralocorticoid,but because the type 2 variant has been associated with a lower(THF�allo-THF)/THE ratio than that seen in the type 1 variant(Ulick et al., 1990; Manteroet al., 1994), it has been suggestedthat the principal steroid abnormality is not a defect in 11�-HSD, but a generalized reduction in cortisol A-ring metabolism(Ulick et al., 1992a). In this study, however, the UFF/UFE ratiowas certainly elevated in patients with type 2 AME (3.5�1.0)suggesting defective 11�-HSD2 activity, though this was not asmarked as the type 1 variant (>23). One can speculate that type2 AME may be caused by mutations in the 11�-HSD2 gene,which result in an enzyme protein with reduced activity, ratherthan very low or absent activity which appears to be the case fortype 1 AME (Muneet al., 1995); sequencing of the 11�-HSD2gene in the type 2 AME patients described herein is under wayand should clarify this situation.

The ingestion of glycyrrhetinic acid, the active ingredient ofliquorice, also significantly reduced UFE, though not to the lowlevels seen in AME. Glycyrrhetinic acid is known to inhibitboth 11�-HSD1 and 11�-HSD2 (Stewartet al., 1987; 1994;Monder et al., 1989; Whorwoodet al., 1993; Albistonet al.,1994), and this may explain why the rise in the (THF� allo-THF)/THE ratio (1.9-fold) was not as impressive as the rise inthe UFF/UFE ratio (3.3-fold) (Fig. 1). Furthermore, the dose ofglycyrrhetinic acid given in this study was significantly lessthan in earlier studies.

The capacity for the conversion of F to E by 11�-HSD2seems considerable. Thus in states of extreme hypercortiso-lism as seen in the ectopic ACTH syndrome, UFE waselevated some tenfold over control values. Despite thisincreased conversion of F to E, the UFF/UFE ratiowas elevated to values seen in patients with AME, inkeeping with the hypothesis that substrate saturation of renal11�-HSD2 in the ectopic ACTH syndrome may lead to amineralocorticoid excess state in an analogous fashion to

AME, with cortisol gaining inappropriate access to the MR(Ulick et al., 1992b; Stewartet al., 1995). The interpretationof more minor changes in this ratio in patients with normal orelevated UFE excretion remains unclear. Thus a UFF/UFEratio of 1.86 in our patients with Cushing’s disease is similarto normal volunteers following liquorice ingestion, yet, incontrast to the ectopic ACTH group all these patientswere normokalaemic (Stewartet al., 1995). Similarly, thesignificance of the slight increase in the UFF/UFE ratio inhypopituitary patients receiving 30 mg rather than 15 mg/dayof hydrocortisone remains speculative. We have argued that30 mg/day of hydrocortisone for such patients is excessive(Dunne et al., 1995), but whether this results in any long-term harm remains unknown.

In summary, we have described a method for the measure-ment of urinary free cortisone and have reported values in bothnormal volunteers and patients with a variety of clinicalconditions involving the pituitary-adrenal axis. Inhibition of11�-HSD2 activity as assessed by this technique is common toboth variants of AME and is found in patients consumingglycyrrhetinic acid. Substrate excess as seen in the ectopicACTH syndrome overwhelms the immense capacity forrenal 11�-HSD2 to convert F to E, in keeping with thehypothesis that cortisol is the offending mineralocorticoid inthis condition. The significance of more subtle alterations in theUFF/UFE ratio remains unknown, but we suggest thatthe measurement of UFE is a significant advance in theassessment of 11�-HSD activity in man, particularly the novelrenal 11�-HSD2 isoform.

Acknowledgements

We wish to thank Jayne Franklyn, David Heath andMichael Sheppard for allowing us to study their patients,Brian Walker for additional clinical material and the MedicalResearch Council for support. PMS is an MRC Senior ClinicalFellow.

References

Albiston, A.L., Obeyesekere, V.R., Smith, R.E. & Krozowski, Z.S.(1994) Cloning and tissue distribution of the human 11�-hydroxysteroiddehydrogenase type 2 enzyme.Molecular and CellularEndocrinology, 105,R11–R17.

Arriza, J.L., Weinberger, C., Cerelli, G., et al. (1987) Cloning of humanmineralocorticoid receptor complementary DNA: structural andfunctional kinship with the glucocorticoid receptor.Science, 237,268–275.

Brown, R.W., Chapman, K.E., Edwards, C.R.W. & Seckl, J.R. (1993)Human placental 11�-hydroxysteroid dehydrogenase: Evidence forand partial purification of a distinct NAD-dependent isoform.Endocrinology, 132,2614–2621.



Dunne, F.P., Elliot, P., Gammage, M.D., Stallard, T., Ryan, T.,Sheppard, M.C. & Stewart, P.M. (1995) Cardiovascular functionand hydrocortisone replacement regimes in patients withhypopituitarism.Clinical Endocrinology, 43, 623–629.

Edwards, C.R.W., Stewart, P.M., Burt, D., Brett, L., McIntyre, M.,Sutano, W.S., de Kloet, E.R. & Monder, C. (1988) Localisation of11�-hydroxysteroid dehydrogenase—tissue specific protector of themineralocorticoid receptor.Lancet, ii, 836–841.

Funder, J.W., Pearce, P.T., Smith, R. & Smith, A.I. (1988)Mineralocorticoid action: target tissue specificity is enzyme, notreceptor, mediated.Science, 242,583–585.

Hellman, L., Nakada, F., Zumoff, B., Fukushima, D., Bradlow, H.L. &Gallagher, T.F. (1971) Renal capture and oxidation of cortisol inman.Journal of Clinical Endocrinology, 33, 52–62.

MacKenzie, M.A., Hoefnagels, W.H.L., Jansen, R.W.M.M.,Bernard, T.J. & Kloppenborg, P.W.C. (1990) The influence ofglycyrrhetinic acid on plasma cortisol and cortisone in healthy youngvolunteers.Journal of Clinical Endocrinology and Metabolism, 70,1637–1643.

Mantero, F., Tedde, R., Opocher, G., Fulgheri, P.D., Arnaldi, G. &Ulick, S. (1994) Apparent mineralocorticoid excess type II.Steroids,59, 80–83.

Milford, D.V., Shackleton, C.H.L. & Stewart, P.M. (1995) Miner-alocorticoid hypertension and congenital deficiency of 11�-hydro-xysteroid dehydrogenase in a family with the syndrome of ‘apparent’mineralocorticoid excess.Clinical Endocrinology, 43, 242–246.

Monder, C., Stewart, P.M., Lakshmi, V., Valentino, R., Burt, D. &Edwards, C.R.W. (1989) Liquorice inhibits corticosteroid 11�-dehydrogenase of rat kidney and liver: in vivo and in vitro studies.Endocrinology, 125,1046–1053.

Moore, C.C.D., Mellon, S.H., Murai, J., Siiteri, P.K. &Miller, W.L. (1993) Structure and function of the hepatic form of11�-hydroxysteroid dehydrogenase in the squirrel monkey, ananimal model of glucocorticoid resistance.Endocrinology, 133,368–375.

Mune, T., Rogerson, F.M., Nikkila, H., Agarwal, A.K. & White, P.C.(1995) Human hypertension caused by mutations in the kidneyisozyme of 11�-hydroxysteroid dehydrogenase.Nature Genetics, 10,394–399.

Oelkers, W., Buchen, S., Diederich, S., Krain, J., Muhme, S.& Schoneshofer, M. (1994) Impaired renal 11�-oxidation of9�-fluorocortisol: an explanation for its mineralocorticoidpotency.Journal of Clinical Endocrinology and Metabolism, 78,928–932.

Rodin, A., Thakkar, H., Taylor, N. & Clayton, R.N. (1994)Hyperandrogenism in polycystic ovary syndrome: evidence ofdysregulation of 11�-hydroxysteroid dehydrogenase.New EnglandJournal of Medicine, 330,460–465.

Shackleton, C.H.L. (1993) Mass spectrometry in the diagnosis ofsteroid-related disorders and hypertension research.Journal ofSteroid Biochemistry and Molecular Biology, 45, 127–140.

Shackleton, C.H.L. & Stewart, P.M. (1990) The hypertension ofapparent mineralocorticoid excess (AME) syndrome. In: Biglieri EG,Melby JC (eds). Endocrine Hypertension, pp155–173. New York,Raven Press.

Soro, A., Ingram, M.C., Tonolo, G., Glorioso, N. & Fraser, R.(1995) Evidence of reduced 11�-hydroxysteroid dehydrogenase and5�-reductase activity in patients with untreated essentialhypertension.Hypertension, 25, 67–72.

Stewart, P.M., Wallace, A.M., Valentino, R., Burt, D.,Shackleton, C.H.L. & Edwards, C.R.W. (1987) Mineralocorticoid

activity of liquorice: 11�-hydroxysteroid dehydrogenase comes ofage.Lancet, ii , 821–824.

Stewart, P.M., Corrie, J.E.T., Shackleton, C.H.L. & Edwards, C.R.W.(1988) The syndrome of apparent mineralocorticoid excess: a defectin the cortisol—cortisone shuttle.Journal of Clinical Investigation,82, 340–349.

Stewart, P.M., Krozowski, Z.S., Gupta, A., Milford, D.V., Howie, A.J.,Sheppard, M.C. & Whorwood, C.B. (1996) Hypertension in thesyndrome of apparent mineralocorticoid excess due to mutation ofthe 11�-hydroxysteroid dehydrogenase type 2 gene.Lancet, 347,88–91.

Stewart, P.M., Murry, B.A. & Mason, J.I. (1994) Human kidney 11�-hydroxysteroid dehydrogenase is a high affinity nicotinamideadenine dinucleotide-dependent enzyme and differs from thecloned type 1 isoform.Journal of Clinical Endocrinology andMetabolism, 79, 480–484.

Stewart, P.M., Walker, B.R., Holder, G., O’Halloran, D. &Shackleton, C.H.L. (1995) 11�-hydroxysteroid dehydrogenaseactivity in Cushing’s syndrome: explaining the mineralocorticoidexcess state of the ectopic ACTH syndrome.Journal of ClinicalEndocrinology and Metabolism, 80, 3617–3620.

Tannin, G.M., Agarwal, A.K., Monder, C., New, M.I. & White, P.C.(1991) The human gene for 11�-hydroxysteroid dehydrogenase.Journal of Biological Chemistry, 266,16653–16658.

Ulick, S., Tedde, R. & Mantero, F. (1990) Pathogenesis of the type 2variant of the syndrome of apparent mineralocorticoidexcess.Journal of Clinical Endocrinology and Metabolism, 70,200–206.

Ulick, S., Tedde, R. & Wang, J.Z. (1992a) Defective ring A reduction ofcortisol as the major metabolic error in the syndrome of apparentmineralocorticoid excess.Journal of Clinical Endocrinology andMetabolism, 74, 593–599.

Ulick, S., Wang, J.Z., Blumenfeld, J.D. & Pickering, T.G. (1992b)Cortisol inactivation overload: a mechanism of mineralocorticoidhypertension in the ectopic adrenocorticotropin syndrome.Journalof Clinical Endocrinology and Metabolism, 74, 963–967.

Walker, B.R., Campbell, J.C., Fraser, R., Stewart, P.M. & Edwards,C.R.W. (1992) Mineralocorticoid excess and inhibition of 11�-hydroxysteroid dehydrogenase in patients with ectopic ACTHsyndrome.Clinical Endocrinology, 37, 483–492.

Walker, B.R., Stewart, P.M., Shackleton, C.H.L., Padfield, P.L. &Edwards, C.R.W. (1993) Deficient inactivation of cortisol by 11�-hydroxysteroid dehydrogenase in essential hypertension.ClinicalEndocrinology, 39, 221–227.

Weaver, J.U., Thaventhiran, L., Noonan, K., Burrin, J.M., Taylor, N.F.,Norman, M.R. & Monson, J.P. (1994) The effect of growth hormonereplacement therapy on cortisol metabolism and glucocorticoidsensitivity in hypopituitary adults.Clinical Endocrinology, 41,639–648.

Whitworth, J.A., Stewart, P.M., Burt, D., Atherden, S.M. &Edwards, C.R.W. (1989) The kidney is the major site of cortisoneproduction in man.Clinical Endocrinology, 31, 355–361.

Whorwood, C.B., Sheppard, M.C. & Stewart, P.M. (1993) Licoriceinhibits 11�-hydroxysteroid dehydrogenase gene expression andpotentiates glucocorticoid hormone action.Endocrinology, 132,2287–2292.

Whorwood, C.B., Mason, J.I., Ricketts, M.L., Howie, A.J. &Stewart, P.M. (1995) Detection of human 11�-hydroxysteroiddehydrogenase isoforms using RT-PCR and localization of the type2 isoform to renal collecting ducts.Molecular and CellularEndocrinology, 110,R7–R12.



Wilson, R.C., Krozowski, Z.S., Li, K., et al. (1995) A mutation inthe HSD11B2 gene in a family with apparent mineralocorticoidexcess.Journal of Clinical Endocrinology and Metabolism, 80,2263–2266.

Zumoff, B., Bradlow, H.L., Levin, J. & Fukushima, D.K. (1983)Influence of thyroid function on the in vivo cortisol-cortisoneequilibrium in man. Journal of Steroid Biochemistry, 18, 437–440.



Documents

Urinary free cortisone and the assessment of 11β-hydroxysteroid dehydrogenase activity in man