WO1990004399A1

WO1990004399A1 - Administration of corticosteroids

Info

Publication number: WO1990004399A1
Application number: PCT/GB1989/001266
Authority: WO
Inventors: Christopher Richard Watkin Edwards
Original assignee: The University Court Of The University Of Edinburgh
Priority date: 1988-10-24
Filing date: 1989-10-23
Publication date: 1990-05-03
Also published as: GB8824835D0

Abstract

The effect of an 11-betahydroxy corticosteroid at the site of its desired action on a patient can be potentiated by the use of an inhibitor for the patient's 11-betahydroxy steroid dehydrogenase. The active components of liquorice, principally glycyrrhizic acid and glycyrrhetinic acid and their derivatives, have this potentiating effect if they are administered in association with the 11-betahydroxy corticosteroid, e.g. hydrocortisone, for example as a pharmaceutical composition.

Description

ADMINISTRATION OF CORTICOSTEROIDS

Field of the invention

Tne present invention relates to the administration of corticosteroids to human ano animal patients. Background of the invention

Corticosteroids such as cortisol are administered to patients for physiological replacement therapy ano for their anti-inflammatory efrect. The active form of these steroids is the 11-betahyoroxy from. However, there is a tendency for the administered 11-betanyoroxy steroid to be deactivated after administration by conversion in tissues such as the liver and kidney to the corresponding 11-keto steroid, e.g. cortisol is converted to cortisone. In order to decrease or eliminate such conversion the 11-oetanydroxy steroids administered for anti-inflammatory effect have been modified, primarily by addition of a fluorine substituent at tne 9-bpsition of the steroid nucleus, however, the enhanced potency of these 3-alpnafluorinated corticosteroids. whilst being an advantage for therapy, is associated with more marked suppression of the hyootnalamic-pituitary-adrenal axis. This can be a serious problem in patients given topical steroid therapy. It would thus be an advantage if it was possible to potentiate the effect of a corticosteroid at the site of the desired action without having to have a steroid structure that was designed to delay its systemic metaoolism.

It is known that conversion of cortisoi to cortisone in humans is catalysed by 11-betanydroxysterold denydrogenase, a microsomal enzyme complex consisting of 11- betaoehydrogenase converting cortisol to cortisone and 11- oxoreouctase converting cortisone to cortisol (Endocrinology 1965 : 116 : 552-560). A similarly named enzyme in the rat catalyses conversion of corticosterone into 11-dehydro-cortιcosterone.

It has been disclosed in The Lancet. 10 October 1987, 821-824. that administration of liouorice. tne active components of whicn are glycyrrhizic acio and its hydrolytic product, glycyrrhetinic acid, to normal adult human subjects in amounts of 200 g/day causes an increase in urinary free cortisoi and has other effects characteristio of 11-betahydroxysteroιd dehydrogenase deficiency indicating inhibition of 11-betahydroxysteroιd dehydrogenase by administration of liquorice.

Summary of the invention

The present invention seeks to utilise that discovery in the administration of corticosteroids. We have found that the effect of a corticosteroid at the site of the desired action can be potentiated by localiy inhibiting the conversion of the 11-betahydroxysteroιd into its inactive form.

According to the invention an inhibitor for 11-beta- hydroxysteroid dehydrogenase is administered to a patient in association with administration of an 11-betahydroxysteroid capable of being metaboiised by the 11-betanydroxysteroid dehydrogenase.

The administration may take place shortly before (e.g. 60 minutes before) or shortly after administration of the 11-betahydroxy steroid or simultaneously with it.

Thus, the invention includes a pharmaceutical comoosition for administration to a oatient, comprising an 11-betahydroxy corticosteroid and an inhibitor for the patient's 11-betahydroxysteroid dehydrogenase. Compounds having aldosterone-like actions, for examble carbenoxolone (3ß-(3-carboxypropιonyloxy)-11-oxo-olean-12- en-30-olc acid) and its bhysiologically acceptable salts sucn as the sodium salt, may be used to potentiate the effect of the corticosteroid. Preferably, however tne material used is a liquorice product, especially giycyrrhizic acid or glycyrrhetimc acid or a physiologically acceptable derivative thereof such as an alkali metal or alkaline earth metal salt. Glycyrrhizic acid and its salts are preferred in compositions intended for internal administration owing to higher water-solubility but they are converted in vivo to glycyrrhetinlc acid or its salts which are more effective in their inhibiting action.

The 11-betahydroxy corticosteroids administered may be any of those previously proposed for administration, for example hydrocortisone (cortisol). Because of the inmbition of the 11-betanydroxysteroid dehydrogenase it is not necessary to use the stronger corticosteroids naving adverse side effects such as those bearing a fluorine substituent at tne 9-posιtion and they are preferably avoided. The corticosteroid and inhibitor may be administered in any convenient way, externally (topically) or internally by injection (e.g. into a joint), or by mouth le.g. by inhalation). The steroid may be administered in conventional amounts, for example topical application of an ointment, cream or the like containing about 0.5% by weight hydrocortisone, and the inhibitor may be included in any amount sufficient to achieve significant inhibition of the 11-betahydroxysteroid dehydrogenase, for example a concentration of about 2 % by weight of glycyrrhetinic acid or a derivative in such an ointment or cream. We have found that whereas an ointment or cream containing hydrocortisone normally has little effect on the skin owing to the high concentration of 11-betanydroxysteroid dehydrogenase there, the inclusion of glycyrrhetimc acid in the olntment or cream causes the ointment or cream to nave a significant effect even at a hydrocortisone dilution of 1 in 100,000. The Tormuiations administered may contain conventional pharmaceutical excipients or diluents and may be suitably packaged. If the two ingredients are not administered in a single composition a pack may comprise separate but associated compositions each containing one of the two active ingredients and designed to be administered in association with the other.

In vitro studies have shown that the mineralocorticoid receptor is non-specific and does not distinguish between aidosterone and cortisoi. In vivo certain tissues with this receptor are aidosterone selective (e.g. kidney and parotid) whereas others with the same receptor are not (e.g. hippocampus, heart). Studies in rats using immunohistochemi stry and in vitro incubation have shown that the enzyme 11ß-hyoroxysteroid dehydrogenase (11ß-OHSD) is present in the aldosterone-selective tissues in much higher concentrations than in those whicn are non-selective. The localisation in the selective tissues is sucn that it is ideally situated to act as either a paracnne or possibly autocrine mechanism protecting the receptor from exposure to corticosterone (and therefore presumably cortisoi in man) and hence determining its apparent specificity. Autoradiography experiments confirmed that inhibition of the enzyme results in loss of protection. Under these conditions ³H-corticosterone was bound to similar sites as was ³H-aidosterone. These flndings appear to explain why patients witn congenital deficiency of 11ß-OHSD or those in whom the enzyme has been inhibited by liquorice develop sodium retention, hypokaiemia and hypertension.

Detailed disclosure

In vitro experiments with either cytosolic preparations of the mineralocorticoid (type 1) receptor or where the cloned receptor has peen expressed in transfected cells have shown that its affιnity is similar for aidosterone. cortisol, corticosterone ano deoxycorticosterone. Tnese contrast witn in vivo studies in which these type 1 receptors in the Kidney, parotid and colon are aloosteroneselective, whereas those in the nippocampus do not distinguisn between aidosterone ano corticosterone. Tnese results led to a suggestion that there must be a factor other than the receptor responsiole for determining the aidosterone tissue specificity, perhaps extravascular corticosteroid binding giobulin (CBG) wnich prererentially bound cortisol or corticosterone ( the major glucocorticold in the rat). However, recent studies in the 10 day old rat which has very low levels of CBG snowed that the in vivo speci ficity of aidosterone was maintained despite the much higner levels of corticosterone, 11ß-OHSD is the microsomal enzyme complex responsible for the interconversion of cortisol and cortisone. Recent worκ nas snown that it consists of two separate enzymes, one converting cortisol to cortisone (11ß-denyorogenase) and the other cortisone to cortisol ( 11-oxoreductase ) . Congenital deficiency of the dehydrogenase enzyme is associated witn severe hypertension, hyooκalemιa and suppression of plasma aidosterone and plasma renin activity. Tms nas been called the syndrome of apparent mineralocorticoid excess. Our studies in the fιrst aduit to be found to have the syndrome suggested that cortisol was acting as a mineralocorticoid. Tnese were in Keeping with other results. we put forward the hypothesis that the normal kidney used 11ß-OHSD to convert cortisol to the inactive steroid cortisone and was tnus protected from this effect. If this was the case tnen inniDition of 11ß-OHSD would result in a fallure of this protective mechanism and thus allow access of cortisol to the non-specιτιc renal mineralocorticoid receptors, resulting in sodium retention, We then found that the active component of liquorice (glycyrrhetinic acid) was a potent inmoitor of 11ß-OHSD and proposed that this was the explanation for the sodium- retaining and potassium-losing actions of liquorice. This resolved tne problem as to why liquorice did not have these effects in patients with severe adrenocortical insufficiency or following bilateral adrenalectomy. 11-ß-OHSD has been shown to be present in the liver. kidney, gonads, placenta, iung and intestinal mucosa. The recent purification of 11ß-OHSD has allowed the production of a specific antiserum and hence precise tissue localisation. We have sought to make a further examination of the tissue distribution of 11ß-OHSD to determine whetner the enzyme is appropriately situated to act as either a paracnne or autocnne protector of the mineralocorticoid receptor in aidosterone-selective organs. (In the usual paracrine system a hormone produced by one cell type acts on adjacent cells as compared to an autocnne mechanism where the hormone is produced by and acts on the cell of origin. Here the term paracnne is used to denote the metabolism of the steroid by 11ß-OHSD in cells which do not contain the mineralocorticoid receptor but wnich can influence the hormonal environment of other eel is with the receptor. This contrasts with an autocnne system in which the enzyme and the receptor are in the same cells.) In addition we have looked at tissues which are not aidosterone-selective but which have mineralocorticold receptors to determine whether this non-selectivity could be explained by an absence of 11ß-OHSD. Finally we have examined the effect of inhibiting 11ß-OHSD on the specific binding of corticosterone by the kidney to determine whether this would result in corticosterone binding in the same sites as aidosterone.

Methodology

A i) Preparation of tissue for enzyme activity

Renal cortex, parotid, heart ano hippocampus was obtained from 10 day old male Sprague-Dawley rats and 0.5g wet weignt tissue nomogemsed in 10ml k rebs Ringer buffer with a Dounce tissue grinder for 30 stroκes. using a fied mg protein/g wet weignt tissue (Bio-Rad protein assay kit) 400μl of a diluted homogenate preoaration was incubated at 37°C for 60 mins with 600μl Krebs Ringer pufrer (+ 0.2% giucose, 0.2% oovine serum aloumin) and 1.2 x 10^-8M ³H-corticosterone (sp. activity 64 Cl/mmol, Amersnam International). Following centrifugation, steroids were extracted from the supernatant using etnyl acetate and ³H- corticosterone separated from ³H-11-denydrocorticosterone using thin-layer cnromatography. Tne percentage conversion of corticosterone to 11-dehydrocortιcosterone by 11ß-OHSD was then calculated. i i ) Isolation of rat Kidney cortical tupuies

enriched in proximal and distal segments

using a Ficoll gradient

2g wet weight of renal cortical tissue was taken from 3 month old male Sprague-Dawley rats and tupuies prepared using a mecnamcally dispersed enzyme solution 10.05% (w/v) collagenase, 0.1% (w/v) hyaluronidase). Following filtration ( to remove giomerull) and successive washing procedures the tubules were subjected to unit gravity sedimentation througn a Ficoll gradient. From this both a distal enriched fraction (narrow, trahsparent tubules) and a proximal enriched fraction (yellowlsh, broad tubules) were established with purities of 85% and 70% respectively. Tubular integrity was evaluated by the dye exclusion method. A tubular count was maoe using a naemocytometer and 400μl of each tubular preparation (containing approximately 10⁵ tubules) incubated with 600μl Krebs Ringer buffer (contaihing 0.2% giucose. 0.2% BSA ) and 1.2 x 10^-8M

³H-cortιsol. Steroids were extracted ana separated as above and 11ß-OHSD activity caicuiated in proximal and distal tubules as % conversion of cortisol to cortisone. B Immunohistochemical localisation of enzyme activity with a specific antibody 11ß-dehydrogenase 11ß-dehydrogenase was purified to apparent homogeneity from a rat hepatic microsomai preparation. 800-foid purification was achieved using agarose-NADP affinity chromatography. The purified enzyme was a giycoprotein (mw 34000) and had no reductase activity. Antibodies to this homogeneous 11ß-dehydrogenase were raised in female New Zealand white rabbits. Pre-immune and immune sera were stored as 1ml aliquots at -20°C. Three month old maie Sprague-Dawley rats (Charies River, Kent) were sacrificed and the kidneys, parotid, heart and hippocampus removed, sliced longitudinally and placed in Bourns fixative for 24 hrs. They were then processed tnrough to paraffin block and sections cut at 4μm. Immunostai mng was accomplished using the Avidin-Biotin-Peroxiαase method. Reagents used were the vector "Elite" (Vector Laboratories Inc, Burlinghame, California, USA).

C Effect of inhibition of 11ß-OHSD on the autoradiographic localisation of H-corticosterone in the

kidndy

H-aIdosterone (100μCl/100g body weight) or ³H-corticosterone (100 μCl/100g body weight; was given to aouit Wistar rats via a jugular vein cannula 1 hour before sacrifice. In a separate experiment 5mg glycyrrhizic acid was given subcutaneously 60 minutes before the ³H-corticosterone to inhibit 11ß-OHSD. In a further experiment

1mg unlabelled corticosterone was given subcutaneously 30 minutes prior to the ³H-corticosterone to determine the non-specific binding. After sacrifice kicneys were removed, frozen, cryostat sectioned (25μ thickness) and then exposed to ³H-ultrofilm for two weeks.

Results The activity of 11ß-OHSD in the homogenates or renal cortex, parotid, hippocampus and neart as determmed by % conversion of ³H-corticosterone to ³H-11-dehydrocorticosterone was compared. The hignest level of activity was present in the Kidney (35 ± 2%) with lower levels in the parotid (10 ± 2%). Little or no enzyme was present in the hippocampus ( 1 ± 1%) or neart.

Density gradient separation snowed that both the proximal and the aistal tubular preparations were capable of converting cortisol to cortisone, % conversion of ³H- cortisol to ³H-cortιsone peing 33% and 23% respectively arter 60 minutes and 18% and 15% respectively after ι5 minutes. In the three experiments performed enzyme activity was higher in the distal than in the proximal tuoule.

Immunohistocnemistry confirmed the high level of enzyme present within the Kidney. In contrast to the density gradient separation the enzyme was mainiy in the proximal tupuie ano not in tne distal nephron. However the enzyme appeared also to be localised either in or immediately adjacent to the vasa recta alongside the papillary collecting tubules.

In the parotid 11ß-OHSD was present in both the intercalated and striated ducts but was not round in the acini. No localised enzyme was present in the heart or hippocampus. Autoradiography with ³H-aldosterone showed the expected binding in the cortex-outer medulla and papilla-inner medulla. In contrast the uptake of ³ H- corticosterone in these sites was very low and little different from the non-specific binding. However, after inhibition of 11ß-OHSD the pattern of ³H-corticosterone binding was marKediy changed and was now similar to that with ³H-aldosterone. The above results show that two major aidosteroneseiective tissues (kidney and parotid) have much higher ievels of 11ß-OHSD than those organs with the same mineralocorticoid receptor but which are not aidosterone specific (heart and hippocampus). The position of enzyme in the kidney as assessed by immunomstochemistry (proximal tubule and vasa recta) suggests that the enzyme is situated in a position which would allow it to act as a paracnne protector of the type 1 receptor in the cortical and papillary collecting tubule. The localisation of this receptor in the rat has been demonstrated using tritiated aidosterone binding. This was found in both the renal cortex-outer medulla and papilla-inner medulla. These results were in keeping with those obtained by others who measured aidosterone binding along the nephron and showed high levels of nuclear labelling in the cortical collecting tubule. The density gradient studies confirmed the presence of enzyme activity in the proximal tubule but also suggested that there was conversion of cortisol to cortisone by the distal nephron. One possible explanation for these apparently discrepant results is that tnere may be a different 11ß-dehydrogenase in the distal tubule.

The free (non-protein bound) fraction of cortisol in plasma is filtered by the kidney; 80-90% of this is reabsorbed passively by the tubule and only about 0.5% is excreted unchanged in the urine. If this reabsorption is by the proximal tubule then this would allow metabolism of cortisoi to cortisone. In addition there would be a need for cortisol which is not filtered to be metabolised. This could be by diffusion from the pentubular capil lanes i nto the proxi mal tubu l ar ce l l . It has been suggested that 11ß- OHSD was involved in the cellular capture mechanism for cortisol and that, at least in the kidney, oxidation to cortisone was apparently a prerequisite for celiular release of the steroid. By way of contrast aidosterone which is not signifleantly metabolised by 11ß-OHSD would be reabsorbed and pass into the peritubuiar piexus without inactivation. Aldosterone and the inactive glucocorticold metabollte would then pass down the peritupular plexus to reach the collecting tubule where aidosterone diffuses across the pasolateral cell membrane to gain access to the cytoplasmic or possibly intranuclear receptor.

Tne blood supply to the renal medulla suggests that there would probably be a need for a different localisation of 11ß-OHSD to prevent access of cortisol to the papillary collecting tubule. Some vasa recta arise directly from interlobular arteries and do not have an initial circulation througn a glomerular tuft. In this case the enzyme would need to be in close relation to the descending vasa recta. Our lmmunohistocnemistry results would be in keeping witn this. A crucial test of our paracnne hypothesis is the demonstration that inhibition of 11ß-OHSD resurts in loss of the selectivity of binoing of aidosterone in tissues such as the kidney. The autoradiography results strongly suggest that this is the case. when the enzyme is intact corticosterone (the steroid equivalent to cortisol in man) is not taken up by the kidney. However, when 11ß-OHSD is inhibited by glycyrrhizic acid (one of the components of liquorice which is hydrolysed in vivo to the major active component glycyrrhetihic acid) the aldosterone selectivity is lost and corticosterone now binds in a distribution which is similar to that of aldosterone. It remains to be determined whether all the corticosterone uptake is by the same receptors as tnose binding aidosterone. Using isolated tubules it has been showh that corticosterone binding sites were concentrated in the cortical collecting tubule. Part but not all of this steroid could be displaced by unlabelled aldosterone. In these studies corticosterone had direct access to the nephron from the incuoation medium and was thus able to bypass the protective moat provided by 11ß-OHSD. In the parotid the enzyme was found in both the intercalated and striated ducts but not in the acini. Previous studies using micropuncture have shown that sodium reabsorption and potassium secretion occurs in the striated but not in the intercalated ducts in rat suomaxillary glands. This is in keeping with the salivary gland localisation of (Na⁺ + K⁺ )-ATPase in the cat as assessed using ³H-ouabain where the ceils of the striated duct were heavily labelled. These results would suggest that the type 1 receptor is likely to be in the striated duct.

It is not clear how mineraiocorticoids gain access to the striated duct. The blood supply to the parotid is very different to that in the kidney. The arteries accompany the ducts within the lobules and break up into capillary and precapillary plexuses around the striated ducts. Arteriolar arcades then arise which continue to supply the acini. Blood flow studies have suggested that the flow was mainly countercurrent to that of saliva. If aidosterone then entered via the basal cell membrane of the striated duct then 11ß-OHSD would need to be present within the cells of the duct. This appears to be the case and would be in keeping with an autocrine system. If however aidosterone enters via the luminal membrane then this would allow the possibility of upstream metabolism by the enzyme in the intercalated duct.

The very low levels of the enzyme in the nippocampus and the heart associated with the lack of any specific tissue localisation suggest that these tissues contain no 11ß-OHSD mechanism for the inactivation of cortisoi or corticosterone. This would allow direct access of these steroids to the receptor and would thus be in keeping with previous work showing that there are type 1 receptors in these tissues but they are not al dosterone-seiective. In fact, behavioural and biochemicai studies in the rat have revealed a number of responses that are unαer stringent control of corticosterone acting via the limbic type i receptor. Aldosterone appeared a competitive antagonist in these studies.

Human 11ß-OHSD has yet to De purified and nence no antisera nave oeen proouceα. It woulα seem nkeiy, oased on measurement of enzyme activity, that the tissue localisation would be similar to that we have found in the rat. The human kidney is known to be an important site for the conversion of cortisol to cortisone. In patients with renal disease one mignt anticipate that this mecnahism would be impaired. we have recently shown that plasma cortisone levels are reαuced in such patients and there is a hignly significant negative correlation between piasma cortisone and creatimne. It remains to be determined what role this mignt play in sodium retention in renal failure. It is also possible that the lacκ of renal 11ß-OHSD mignt be important in delaying the deveiopment of hyperkalemia in such patients. Tnese effects would require the loss of enzyme activity to be dissociated from that of the mineralocorticoid receptors in the collecting tuoule.

Tnese studies all indicate the importance of this steroid shuttle. Congenital or acquired deficiency of the enzyme converting cortisol to cortisone results in cortisol functioning as a potent mineralocorticoid. We would suggest that without this paracnne mechanism we would be 'pillars of salt'.

Having demonstrated the role that 11ß-OHSD plays in protecting the mineralocorticoid receptor we nave now begun to examine the possibility that 11ß-OHSD may be important in regulating access of cortisoi to the glucocorticoid receptor. To do this we have employed the skin vasoconstrictor assay, used to determine the potency of topical corticosteroids. The assay is based on the degree of skin pallor produced by application of corticosteroid to the forearm. Such pallor can be assessed on a visual analogue scale and has been validated as an accurate measure of glucocorticoid potency and dose-response.

We have examined the effect of 11ß-OHSD inhibition by topical glycyrrhetinic acid on hydrocortisone skin vasoconstrictor potency. Twenty five tests were performed on 23 volunteers who had had no previous exposure to exogenous corticosteroids.

Test solutions comprised:-

(i) hydrocortisone acetate alone

(ii) glycyrretinic acid (GE) alone

(iii) hydrocortisone acetate plus GE (singie solution)

All solutions were freshly prepared in 95% ethanol in the following concentrations:- hydrocortisone: 1 , 3, 10, 30 and 100 mg/ml

GE : 20 mg/ml (for all tests). lest substances (10 μl) were applied to 7 mm² sites, demarcated with silicone grease, on the flexor aspect of the forearm. After drying, the solutions were occiuoed with polyester film for 16 h. The film was then removed. The degree of blanching of each test area was assessed 1,

2, 3 and 6 h after removal of the film by 2 observers using a linear analogue scale. Scoring was:-

0 no blanching

1 miId blanching

2 definite blanching

3 intense blanching

All solutions were applied in random sequence and scoring was double blind.

Time-effect curves for each test substance were plotted and tne area under the curves calculated. GE alone (solution ii) had no effect. The results for solutions (i) and (iii) were as follows:-

Hyorocortisone Area unoer curve of skin vasoconstriction oose mg Solution (i) Solution (iii)

Tne differences at hyorocortisone doses of 10 - 100mg are mgnly significant (p < 0.01).

Tnese results demonstrate unequivocal potentiation of glucocorticoid action by GE. This has important consequences for topical and other targetteo glucocorticoid theraby. Using immunohistocnemistry we have confirmed that 11ß-OHSD is present in the skin with high concentrations in the epidermis and in a number of dermal structures. Studies on rats, mice and numan skin nave shown 11ß-OHSD bioactivity. very recently we have found that cerebellum, one of the few tissues that lacks mineralocorticoid receptor out has a high concentration of glucocorticoid receptor, expresses 11ß-OHSD mRNA ana shows both enzyme Dioactivity and immunoreactivιty. Other non-mineralocorticoid target tissues showing 11ß-OHSD activity incluαe the testis, liver and lung. In the testis we have shown that the enzyme is in interstitial cells. Of great interest is the ooservation that, in the rat, the enzyme is not present in testis at birth put appears at about 20 days, coinciding witn the onset of puoerty. In vitro, corticosteroids block the production of testosterone by interstitial cells. Thus novel 11ß-OHSD activity at puberty may enhance testicuiar testosterone production by locally inactivating glucocorticoid. This could be a crucial step in regulation of the onset of puberty.

In the iung there is evidence for both 11ß-oehydrogenase and 11-oxoreductase activity. It has been suggested that increased cortisone to cortisoi conversion (oxoreductase) is important in the maturation of the lung, but when using human foetai lung explants as opposed to monolayer cultures only dehydrogenase activity has been found. This suggests that inhibition of 11ß-OHSD could enhance lung maturation by locally increasing giucocorticoid concentrations. Of greater clinical application is the possibility that 11ß-OHSD inhibition might potentiate endogenous or inhaled glucocorticoid therapy in the treatment of asthma, in parallel with the effect of GE on topical glucocorticoid activity. This also raises the possibility that some corticosteroid-resistant cases of disorders that are usually glucocorticoid-sensitive could be, at least in part, due to excessive tissue dehydrogenase activity.

Earlier studies on 11ß-OHSD demonstrated that the liver was an important site both for the conversion of cortisol to cortisone and also for the reverse oxoreductase reaction. The function of this dual activity within the same organ is unknown. However, liquorice extract is extensively used by general practitioners in Japan in the treatment of hepatitis. Apparently, abnormally elevated plasma concentrations of liver enzymes fall on therapy with liquorice but rise again when this is stopped. The effect of liquorice could be due to inhibition of liver dehydrogenase leading to enhanced hepatic cortisoi concentrations, providing local suppression of inflammation.

Potentiation of systemic glucocorticoid action might be expected to also increase deleterious side-effects such as hypothalamo-pituitary-adrenal axis suppression. However, in rat pituitary there is little or no 11ß-OHSD activity and enzyme immunoreactivity is low or absent. Furthermore, our patient with 11ß-OHSD deficiency had normal circulating levels of ACTH . This suggests tnat enzyme inmbitor-meoiated glucocorticoid potentiation may not lead to an eoual increase of inhibition of the hypothalamo-pituitary-aorenal axis. Thus organ-specific manipulation of glucocorticoid potency might be achieved by topical or systemic 11ß-OHSD inhibitinn.

Claims

1. The use of an inhibitor for 11-betahydroxysteroid dehydrogenase for the preparation of a medicament for potentiating the effect of an 11-betahydroxy corticosteroid at the site of its desired action on a patient.

2. The use of glycyrrhizic acid or glycyrrhetihic acid or a physiologically acceptable derivative thereof for the preparation of a medicament for potentiating the effect of hydrocortisone at the site of its desired action on a human patient.

3. A method of treatment of a patient by administration of an 11-betahydroxy corticosteroid, in which method the local conversion of the 11-betahydroxy corticosterold into an inactive form catalysed by the 11-betanydroxysteroid dehydrogenase occurring naturally in the patient is inhibited by the administration of an inmbitor for the 11-betahydroxysteroid dehydrogenase to the patient in association with the 11-betahydroxy corticosteroid treatment.

4. A method of treatment of a human patient by administration of hydrocortisone in which method a liquorice product selected from the group consisting of glycyrrhizic acid, glycyrrhetimc acid and derivatives thereof is administered to the patient in association with the hydrocortisone treatment for the local inhibition of 11-beta-hydroxysteroid dehydrogenase.

5. A method as claimed in claim 3 in which the inhibitor is administered between 60 minutes before and shortly after administration of the 11-betanydroxy corticosteroid.

6. A method as claimed in ciaim 4 in which the nyorocortisone and the liquorice product are admi n i ste red essentially simultaneously.

7. A pnarmaceutical composition for administration to a patient. comprising an innibitor for the patient's endogenous 11-Detanyoroxysteroid denydrogenase and an 11- betanyoroxy corticosteroid cabable of being metabolised by tne 11-Detanydroxysteroid denydrogenase.

8. A pnarmaceutical composition for administration to a numan patient comprising hydrocortisone and a Iiquorice product selected from carbenoxolone, glycyrrhizic acid giycyrrnetinic acid and derivatives thereof.

9. A pnarmaceutical composition as claimed in claim 8 in the form of an ointment or cream for topical application.

10. A pnarmaceutical composition as ciaimed in claim 8 in a form suitable for administration internally by injection or by moutn and containing glycyrrhizic acid or carbenoxolone.