| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Comments |
Department of Endocrinology (A.A.T., S.M.S.), Christie Hospital, Manchester M20 4BX; Department of Endocrinology (J.P.M.), St. Bartholomew’s and The Royal London School of Medicine and Dentistry, London EC1A 7BE; and Department of Clinical Biochemistry (N.F.T.), Kings College Hospital, London SE5 9RS, United Kingdom
Address correspondence and requests for reprints to: Dr. Andy Toogood, Department of Endocrinology, Christie Hospital, Manchester, M20 4BX, United Kingdom.
Abstract
11 ß-hydroxysteroid dehydrogenase type 1 (11ß-HSD1) functions as a net reductase converting cortisone to cortisol. GH inhibits 11ß-HSD1, resulting in a shift in cortisol metabolism favoring cortisone, an observation that may have significance in patients with ACTH deficiency who are unable to compensate for such changes. We have studied the effect of three doses of GH replacement (0.17, 0.33, and 0.5 mg each given for 12 weeks in ascending order) on cortisol metabolism in nine patients, aged 62–70 yr, with hypopituitarism who were receiving fixed doses of oral hydrocortisone.
Serum insulin-like growth factor I levels rose in a dose-dependent manner over the course of the study. Fat mass decreased significantly at 24 weeks (P = 0.02), a change that was maintained at 36 weeks. Fasting serum insulin levels did not change significantly over the course of the study.
The ratio of urine cortisol to cortisone metabolites (Fm/Em) fell significantly at 12 weeks (GH dose, 0.17 mg/day) from 1.32 (0.91–2.20) at baseline to 1.08 (0.89–2.11) (P < 0.05). Although it did not fall further as the dose of GH was increased, the reduction in the Fm/Em ratio persisted at 24 weeks (GH dose, 0.33 mg/day), 1.09 (0.8–2.11) (P < 0.05 vs. baseline), and 36 weeks (GH dose, 0.5 mg/day), 1.19 (0.82–2.31) (P < 0.05 vs. baseline). The Fm/Em ratio did not correlate with serum insulin-like growth factor I, fat mass, or fasting insulin levels at any time during the study.
This study confirms the inhibitory effect of GH on 11ß-HSD1 but has shown that the effect occurs maximally at very low GH doses and is not mediated indirectly by change in circulating insulin. Patients with partial or total ACTH deficiency, in whom cortisol replacement is suboptimal, may be at risk of the clinical manifestations of cortisol deficiency when they are commenced on GH therapy.
CORTISOL METABOLISM is substantially dependent on the interconversion of cortisol and cortisone under the influence of the isoforms of 11ß-hydroxysteroid dehydrogenase (11ß-HSD). Two isoforms of 11ß-HSD have been described. The Type 1 isoenzyme is a low-affinity NADP(H)-dependent enzyme which has been localized to liver, adipose tissue, gonads, and the central nervous system and functions as a net reductase converting cortisone to cortisol (1). The Type 2 isoenzyme is a high-affinity NAD-dependent dehydrogenase localized to kidney, placenta, colon, and the salivary glands that exhibits unidirectional dehydrogenase activity such that physiologically active cortisol is converted to cortisone (2).
Variations in glucocorticoid metabolism may be of physiological and pathological relevance for several reasons. First, an increase in cortisol metabolism may result in a reduction in circulating cortisol in patients with insufficient ACTH reserve for augmenting adrenal steroidogenesis. Conversely, there is evidence that essential hypertension may be associated with reduced 11ß-HSD activity (3, 4). Furthermore, a variation in cortisol-cortisone conversion within a particular organ or tissue may have specific local metabolic consequences. This is exemplified by the renal tubular action of 11ß-HSD-2, which protects the mineralocorticoid receptors from exposure to cortisol by catalyzing its conversion to inactive cortisone. In addition, cortisol-cortisone interconversion has been shown to be modulated by adipose tissue (5, 6) and in polycystic ovary syndrome (7, 8), raising the intriguing hypothesis that tissue-specific modulation of cortisol metabolism may have pathological consequences either by increasing local exposure to glucocorticoid action (5) or by accelerating cortisol clearance so that adrenal androgen production is increased under the influence of increased ACTH secretion (7, 8).
There has been accumulating evidence in recent years that GH may modulate cortisol metabolism and specifically inhibits 11ß-HSD1. The net effect of this is to shift net cortisol metabolism in favor of inactive cortisone (9, 10, 11, 12, 13). The effect of this phenomenon is an increase in cortisol clearance in active acromegaly (12, 13) and a reduction in cortisol clearance in GH deficiency (11), the latter reverting to normal with GH replacement therapy (9, 10). The mechanism of GH modulation of cortisol metabolism is likely to be insulin-like growth factor (IGF) I dependent (14), but because pathological GH excess, hypopituitarism, and GH therapy are all associated with perturbations of insulin sensitivity, and insulin exerts an inhibitory effect on 11ß-HSD1 activity (2), it is possible that changes in insulin secretion and action mediate the changes observed. We have investigated this further by determining the dose dependence of GH modulation of cortisol metabolism, with simultaneous measurement of serum insulin and body composition, in a cohort of elderly hypopituitary patients.
Subjects and Methods
Patients
We studied nine patients (seven males), between 62 and 70 yr of age, who had organic disease of the hypothalamic-pituitary axis: seven had been treated for a nonfunctioning pituitary adenoma, one was treated for a prolactinoma, and one was treated for a FSH-secreting adenoma. All the patients had ACTH deficiency that had been previously determined by an insulin tolerance test or a glucagon stimulation test and had been receiving stable hydrocortisone replacement therapy for 6 months before entering the study. The dose remained unchanged throughout the study period; seven patients were receiving 30 mg/day hydrocortisone, and the other two patients were taking 25 and 35 mg/day. All patients had severe GH deficiency documented by an arginine stimulation test. All patients had gonadotropin deficiency; six of the men were receiving testosterone therapy, and neither of the women were taking oestrogen replacement. Six subjects had TSH deficiency and were taking T4.
Study design
The protocol for this study was approved by the South Manchester Area Health Authority Ethics Committee. This was an open study examining the effects of three doses of GH (0.17, 0.33, and 0.5 mg/day) (Genotropin; Pharmacia & Upjohn, Inc., Stockholm, Sweden) in a group of elderly patients with organic GH deficiency. Each dose of GH was taken for 12 weeks, commencing on 0.17 mg/day and finishing on 0.5 mg/day, administered sc at 2200 h. GH was discontinued after 36 weeks of therapy. To determine the effects of withdrawal, the patients were reassessed again 12 weeks after cessation of GH therapy.
Patients were assessed at baseline, 12, 24, 36, and 48 weeks. A 24-h urine collection was carried out on the day before the visit. Blood was drawn after an overnight fast for estimation of serum IGF-I, insulin, and IGF binding protein-3. The subjects underwent a total body dual-energy x-ray absorptiometry (DXA) scan to determine body composition at each visit. The data from this study relating to dose-dependent IGF-I generation and body composition have been published previously (15).
Assays
Urinary steroid profiles were measured by gas chromatography, as described previously (16). The intra- and interassay coefficients of variation (CV) were between 7.1 and 21.1% and 11.2 and 21.9%, respectively.
Total cortisol metabolites were determined from the sum of
tetrahydrocortisone (THE), tetrahydrocortisol (THF), 5
-THF,
-cortolone, ß-cortolone, -cortol, and ß-cortol. 11-Hydroxy
cortisol metabolites (Fm) were derived from the sum of THF, 5-THF,
(ß-cortolone + ß-cortol) x 0.5, and -cortol. 11-Oxo
cortisol metabolites (Em) were determined from the sum of THE,
-cortolone, and (ß-cortolone + ß-cortol) x 0.5. 20-Hydroxy
cortisol metabolites were determined from the sum of -cortolone,
ß-cortolone, -cortol, and ß-cortol and 20-oxo cortisol
metabolites from THE, THF, and 5-THF. The Fm/Em ratio was calculated
as an index of total net 11ß-HSD activity.
Serum IGF-I was measured, following acid/alcohol extraction, by an in-house RIA. The reference preparation used was NIBSC 87/518. The intra-assay CV for mean IGF-I concentrations of 46, 246, and 706 µg/L were 11.3%, 6.5%, and 4.7%, respectively. The sensitivity of this assay was 14 µg/L.
Serum insulin was measured using an immunoradiometric assay (DSL, Houston, TX). The sensitivity of this assay was 0.05 µg/L. The intra-assay CV for mean serum concentrations of 0.19, 0.70, and 2.18 µg/L were 8.3%, 4.5%, and 6.4%, respectively.
Body composition
Body composition of each patient was determined by DXA using a QDR4500A whole body scanner (Hologic Inc., MA). Patients were scanned supine, wearing a cotton gown. Truncal fat mass was obtained using the anterior superior iliac spine and the lateral border of the clavicle as reproducible anatomical landmarks. The CV was 1.75% for fat mass.
Statistics
Results are expressed as median (range). Comparisons between individual time points were made using the Wilcoxon signed rank test. Relationships between variables were determined using Spearman’s correlation. P < 0.05 was considered statistically significant.
Results
The serum IGF-I level rose significantly over the 9 months of GH
therapy in a dose-dependent manner (Table 1
).
|
Total urine cortisol metabolites (Fm) and metabolites of cortisone
(Em) did not change significantly over the course of the study.
However, the Fm/Em ratio had fallen significantly at 12 weeks (GH dose,
0.17 mg/day) from 1.32 (0.91–2.20) at baseline to 1.08 (0.89–2.11)
(P < 0.05). Although it did not fall further as the
dose of GH was increased, the reduction in the Fm/Em ratio persisted at
24 weeks (GH dose, 0.33 mg/day), 1.09 (0.8–2.11)
(P < 0.05 vs. baseline) and 36 weeks (GH
dose 0.5 mg/day), 1.19 (0.82–2.31) (P < 0.05
vs. baseline) (Table 1
).
There was an increase in 20-OH/20-oxo steroids at week 24 of the study,
however, this was not apparent at week 12 or week 36. These changes did
not correlate with the changes in the Fm/Em ratio. 5-THF/THF did not
change significantly over the course of the study (Table 1).
Changes in body composition
Fat mass did not fall significantly during the first 12 weeks of
the study (P = 0.07) but was significantly reduced
compared with baseline at 24 weeks of GH therapy (P =
0.02 vs. baseline). This reduction was maintained at 36
weeks (P = 0.02 vs. baseline,
P = 0.11 vs. 24 weeks) (Table 1). Trunk fat
mass decreased significantly at week 24 (P = 0.02
vs. baseline), a change that was maintained at week 36
(P = 0.02 vs. baseline; P =
0.17 vs. week 24) (Table 1).
Fasting serum insulin levels
The fasting serum insulin levels at baseline, 12, 24, and 36 weeks were 10.86 µg/L (1.71–30.36), 14.15 µg/L (1.31–24.79), 16.43 µg/L (5.5–44.73), and 24.26 µg/L (5.06–47.15), respectively. The changes were not significant over the course of the study although there was a trend toward an increase at 36 weeks (P = 0.09 vs. baseline).
Relationships between steroid metabolites, body composition, and fasting insulin levels
The Fm/Em ratio did not correlate with serum IGF-I, fasting serum insulin, fat mass, or trunk fat mass at baseline, 12, 24, or 36 weeks.
Withdrawal of GH
Seven patients returned urine collections after 3 months off GH
therapy. In those patients Fm/Em rose significantly from 0.99
(0.82–1.35) to 1.11 (0.92–1.60) (P < 0.05), total
fat mass rose from 20.69 kg (15.87–29.82) to 22.08 kg (17.03–31.42)
(P < 0.02), and trunk fat rose from 11.46 kg
(9.60–16.15) to 12.51 kg (10.51–16.97) (P < 0.02).
Fasting insulin levels fell from 24.3 µg/L (5.1–47.2) to 16.5 µg/L
(3.9–27.5) (P < 0.05), and serum IGF-I fell from 277
µg/L (140–502) to 108 µg/L (91–182) (P < 0.02).
The urinary ratios of 5-THF/THF and 20-OH/20-oxo did not change
significantly when GH therapy was withdrawn.
Discussion
The results of this study are consistent with our previous findings of an inhibitory effect of GH on net cortisone to cortisol conversion (9, 10). The determination of the Fm/Em ratio provides an index of overall 11ß-HSD activity, and a decrease in the ratio could be explained by an inhibitory effect on 11ß-HSD1, enhanced activity of 11ß-HSD2, or a combination of these effects. We have previously demonstrated that this effect of GH is not associated with any change in urine free cortisone or the ratio of urine free cortisol to free cortisone (as an index of 11ß-HSD2 activity; Ref. 17), and we have, therefore, inferred that the change in the Fm/Em ratio is substantially due to a direct or indirect inhibitory effect of GH on 11ß-HSD1 activity (10). This conclusion is supported by observations in treated and untreated acromegaly (12), and it is reasonable to assume that similar mechanisms are operating in the effect of GH in this elderly patient cohort.
In the present study, the change in cortisol metabolism was evident at very low GH doses, indicating that 11ß-HSD1 activity is particularly sensitive to changes in GH status. There were no significant changes observed in serum T4 or triiodothyronine, which might be implicated in changes in steroid metabolism. Previous observations in GH-deficient hypopituitary patients have shown an inverse relationship between indices of adiposity and the Fm/Em ratio, indicating increased net cortisol to cortisone conversion under the influence of body fat (11). These findings were in contrast to in vitro observations of increased 11ß-HSD1 activity in adipocytes from omental fat depots, the effect being enhanced further by the addition of glucocorticoid (5). However, recent in vivo data in subjects with normal GH status has confirmed the inverse relationship between body mass index, central adiposity, and the Fm/Em ratio (18). Importantly, we observed changes in cortisol metabolism before any reduction in total fat mass, determined by the sensitive technique of DXA or of fasting serum insulin. These latter findings suggest that this effect of GH is not mediated by insulin or by a change in body fat. The change in the Fm/Em ratio observed after withdrawal of GH therapy for 12 weeks was predictably associated with an increment in body fat and a decrement in serum insulin. These findings are not surprising given that patients were on the highest doses of GH before discontinuation and do not imply a causal relationship. Although the activity of 11ß-HSD1 exhibits sexual dimorphism (11, 19), with increased activity in males, the longitudinal within subject analysis that we used was not biased by this phenomenon.
Whereas the effect of GH on cortisol metabolism demonstrated in this study was not dependent on change in body fat, it is nonetheless possible that a decrease in adipocyte cortisol exposure may partly explain the lipolytic effects of GH. We would hypothesize that the net consequences for overall cortisol metabolism in this situation depend on the summation of the effect of GH/IGF-I increasing cortisol clearance, on the one hand, and the lipolytic effect of GH, on the other hand. Decreased tissue exposure to glucocorticoid may result in further reduction in body fat, and the inverse relationship between fat mass and net cortisol-cortisone conversion may have the effect of partially off-setting the GH-induced increase in cortisol metabolism.
Systemic glucocorticoid availability will not be affected by these phenomena in patients with intact ACTH reserve because of feedback considerations. However, the sensitivity of 11ß-HSD1 to the low doses of GH used in the present study indicate the potential for precipitating clinically significant manifestations of cortisol deficiency in patients with reduced ACTH reserve who are on suboptimal glucocorticoid replacement therapy.
Acknowledgments
We are grateful to Dr. A. A. Palizban for excellent technical assistance with the urinary steroid metabolite analysis.
Footnotes
1 Supported by Pharmacia and Upjohn. 
Received August 11, 1999.
Revised December 8, 1999.
Accepted December 20, 1999.
References
-reductase activity in polycystic ovary syndrome. Lancet. 335:431–433.[CrossRef][Medline]
cortisol conversion in subjects with central adiposity. J Clin Endocrinol Metab. 84:1022–1027.This article has been cited by other articles:
 |
|
 |
|
  H. A Sigurjonsdottir, R. Andrew, R. H Stimson, G. Johannsson, and B. R Walker Lack of regulation of 11{beta}-hydroxysteroid dehydrogenase type 1 during short-term manipulation of GH in patients with hypopituitarism Eur. J. Endocrinol., September 1, 2009; 161(3): 375 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Gibney, M.-L. Healy, and P. H. Sonksen The Growth Hormone/Insulin-Like Growth Factor-I Axis in Exercise and Sport Endocr. Rev., October 1, 2007; 28(6): 603 - 624. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A Sigurjonsdottir, J. Koranyi, M. Axelson, B.-A. Bengtsson, and G. Johannsson GH effect on enzyme activity of 11{beta}HSD in abdominal obesity is dependent on treatment duration Eur. J. Endocrinol., January 1, 2006; 154(1): 69 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Tomlinson, E. A. Walker, I. J. Bujalska, N. Draper, G. G. Lavery, M. S. Cooper, M. Hewison, and P. M. Stewart 11{beta}-Hydroxysteroid Dehydrogenase Type 1: A Tissue-Specific Regulator of Glucocorticoid Response Endocr. Rev., October 1, 2004; 25(5): 831 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Tomlinson, N. Crabtree, P. M. S. Clark, G. Holder, A. A. Toogood, C. H. L. Shackleton, and P. M. Stewart Low-Dose Growth Hormone Inhibits 11{beta}-Hydroxysteroid Dehydrogenase Type 1 but Has No Effect upon Fat Mass in Patients with Simple Obesity J. Clin. Endocrinol. Metab., May 1, 2003; 88(5): 2113 - 2118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Friedberg, E. Zoumakis, N. Hiroi, T. Bader, G. P. Chrousos, and Z.'e. Hochberg Modulation of 11{beta}-Hydroxysteroid Dehydrogenase Type 1 in Mature Human Subcutaneous Adipocytes by Hypothalamic Messengers J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 385 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Charmandari, C G D Brook, and P C Hindmarsh Why is management of patients with classical congenital adrenal hyperplasia more difficult at puberty? Arch. Dis. Child., April 1, 2002; 86(4): 266 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Trainer, W. M. Drake, L. A. Perry, N. F. Taylor, G. M. Besser, and J. P. Monson Modulation of Cortisol Metabolism by the Growth Hormone Receptor Antagonist Pegvisomant in Patients with Acromegaly J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 2989 - 2992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Charmandari, P. C. Hindmarsh, A. Johnston, and C. G. D. Brook Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency: Alterations in Cortisol Pharmacokinetics at Puberty J. Clin. Endocrinol. Metab., June 1, 2001; 86(6): 2701 - 2708. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
