| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Endocrinology Unit, Department of Medical Sciences, University of Edinburgh, Western General Hospital, Edinburgh, Scotland, United Kingdom EH4 2XU
Address all correspondence and requests for reprints to: Dr. Ruth Andrew, Endocrinology Unit, Department of Medical Sciences, University of Edinburgh, Western General Hospital, Crewe Road, Edinburgh, Scotland, United Kingdom EH4 2XU. E-mail: ruth.andrew{at}ed.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have investigated the use of [9,11,12,12-2H4]cortisol (d4F) to distinguish the dehydrogenase and reductase activities. On metabolism by dehydrogenation, d4F loses 11
- deuterium, forming trideuterated cortisone (d3E) and is regenerated by reduction to trideuterated cortisol (d3F). Healthy men (n = 6) participated in a randomized, double blind, cross-over study comparing oral placebo and the 11ßHSD inhibitor, carbenoxolone (100 mg every 8 h for 7 d). d4F and its metabolites were measured in plasma and urine during a steady state infusion. Inhibition of 11ßHSDs by carbenoxolone was measured by increased steady state concentrations of d4F (41 ± 5.1 vs. 48 ± 7.7 nM; P < 0.05) and a fall in the rate of appearance of d3F (P < 0.05). 11ßHSD1 reductase activity could be measured specifically as conversion of d3E to d3F (28 ± 4.2 vs. 17 ± 3.1 nM; P < 0.05), whereas 11ßHSD2 could be measured by initial rates of appearance of d3E or from urinary ratios of d4F/(d3E + d3F) (0.73 ± 0.06 vs. 1.02 ± 0.03; P < 0.05).
This technique offers a significant advance in the methods available to measure turnover in 11ßHSDs and isozymes of 11ßHSDs in vivo in human studies, and this study confirms that carbenoxolone inhibits both isozymes of 11ßHSD.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
11ßHSD type 2 is a unidirectional dehydrogenase located close to MR and inactivates cortisol to cortisone, preventing illicit occupation of these receptors by cortisol. This is necessary because cortisol circulates in concentrations 1000-fold in excess of those of aldosterone, the intended ligand (1). Mutations in 11ßHSD type 2 are associated with cortisol-induced activation of MR and are manifest by sodium and water retention, hypokalemia, and severe hypertension (2). Impairment of the activity of 11ßHSD type 2 has been implicated also in essential hypertension (3, 4, 5). The 11ßHSD type 1 enzyme functions predominantly as a reductase in vivo, promoting cortisol production from cortisone near GR, e.g. in liver, skeletal muscle, and fat, and also adjacent to corticosteroid receptors in the brain. A few presumed cases of congenital deficiency of 11ßHSD type 1 have been reported (6, 7). These patients present with hirsutism attributable to overstimulation of steroidogenesis in the adrenal cortex to compensate for the reduced ability of peripheral tissues to generate active cortisol from cortisone. Changes in the activity of 11ßHSD type 1 have also been implicated in obesity (8, 9) and polycystic ovarian syndrome (10).
There is increasing interest in the control of local activation and inactivation of cortisol by the isozymes of 11ßHSD in patients with common disorders, e.g. hypertension and obesity. Most of the reports on the activity of these enzymes in humans are based on measurements of cortisol and cortisone in blood or the concentrations of their major metabolites in urine. These approaches provide limited information on the activities of the individual 11ßHSDs. The two isozymes work in opposition in vivo, and hence these measures merely reflect the balance of their activities. This is exemplified by the effects of inhibitors of 11ßHSDs. Glycyrrhetinic acid inhibits cortisol inactivation by 11ßHSD type 2 in vivo and increases the ratios of cortisol to cortisone metabolites (11, 12). However, carbenoxolone, which inhibits both 11ßHSD isozymes, does not alter the ratio of metabolites of cortisol and cortisone due to inhibition of both cortisol inactivation and reactivation (13). Nevertheless, carbenoxolone results in marked inhibition of turnover in the 11ßHSD pathways, with increased cortisol concentrations in kidney and decreased cortisol levels in the liver. Urinary metabolite ratios therefore can give false negative results. As the isozymes possess distinct physiological roles and are altered in different pathologies, determination of their separate activities is desirable.
Attempts have been made to establish measurements allowing distinction between the two isozymes. One of the first approaches used 11-tritiated cortisol (14), which is metabolized by 11ßHSD type 2, yielding unlabeled cortisone and tritiated water. There is some evidence that the type 1 isozyme may also catalyze this reaction. This method provides information about dehydrogenation only and is less desirable due to the ethical concerns about the administration of radioactivity to humans (5). The ratio of concentrations of unconjugated cortisol to those of cortisone in urine is also a useful index of the activity of renal 11ßHSD type 2, because the concentrations of these steroids in urine reflect the enzyme in the distal convoluted tubule (11, 15).
Unfortunately, a similar index has not been determined for the reductase activity of the type 1 isozyme. The best test to date has been the appearance of cortisol in the circulation after first pass metabolism of an oral dose of cortisone acetate (13, 16). However, differences in cortisol concentrations achieved in this manner may be due to alterations in the activity not only of 11ßHSD type 1, but also of all the other hepatic steroid-metabolizing enzymes, e.g. A ring reductases. Tissue-specific activity of 11ßHSDs has been measured in vivo by invasive means, e.g. arterio-venous sampling across the liver or fat beds (17), or, alternatively, in vitro by taking biopsies of adipose tissues (8).
In this paper we investigated whether turnover in the 11ßHSD pathway could be measured and, further, whether the dehydrogenase and reductase components of metabolism could be distinguished, using a stable isotope tracer of cortisol metabolism. The tracer, analogous to that prepared by Ulick et al. (18), contained four deuteriums in the steroid skeleton, one of which was in the 11 position. This tracer could be distinguished from endogenous cortisol by mass difference and was expected to be metabolized as shown in Fig. 1
. The 11-deuterium will be lost on metabolism by dehydrogenation, yielding a labeled cortisone molecule [trideuterated cortisone (d3E)] with a mass 3 greater than that of the endogenous cortisone. As an extension of Ulick’s proposal, we hypothesized that on reactivation by 11ßHSD type 1/reductase, d3E should be converted to a labeled cortisol species [trideuterated cortisol (d3F)], which can be distinguished from the [9,11,12,12-2H4]cortisol (d4F) infused. Thus, using this tracer it would be possible to distinguish the substrate for 11ßHSD type 2 from the product of 11ßHSD type 1 and from endogenous cortisol and cortisone. We now report the ability to measure the rates of appearances of these labeled steroids reflecting the selective activities of each isozyme of 11ßHSD.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cortisol, cortisone, NAD+, buffer reagents, and reagents for derivatization were obtained from Sigma (Poole, UK). d4F was obtained from Cambridge Isotope Laboratories (Cambridge, MA). Internal standards (epi-cortisol and epi-tetrahydrocortisol) were obtained from Steraloids (Newport, RI). Solvents were glass distilled and HPLC grade (Rathburn Chemicals, Walkerburn, UK). Lipidex 5000 was obtained from Canberra Packard (Pangbourne, UK). Sep-Pak C18 cartridges were obtained from Waters Corp. (Glasgow, UK). Tablets of placebo and carbenoxolone were obtained from Biorex Laboratories Ltd. (Middlesex, UK).
In vitro metabolism tracer by dehydrogenation
Human placental homogenates [minced tissue (0.3–0.6 g) in Krebs-Ringers buffer (5 ml)] were incubated with cortisol or d4F (5 µg) in the presence of NAD+ (200 nM) at 37 C. Aliquots (1 ml) were removed at 0, 30, 60, and 120 min and vortexed with ethyl acetate (10 vol). Dried steroid extracts were analyzed by HPLC with UV detection (Waters 441 absorbance detector, 254 nm), using a Luna C18 column (15 cm, 4.9 µm; Phenomenex, Macclesfield, UK) at 30 C with a mobile phase of water/acetonitrile/methanol (60:15:25; 0.7 ml/min). The percent conversion of substrate to product was calculated by the measurement of peak areas. Assays were performed in duplicate and also without tissue and without cofactor as controls.
In vivo metabolism of d4F
Healthy lean volunteers (six males, aged 22–37 yr), with normal thyroid, renal, and hepatic function and who had not received glucocorticoid treatment in the last 3 months, were recruited into a randomized, double blind, cross-over study in which they received oral placebo or carbenoxolone (100 mg every 8 h) for 3 d. Local ethical approval and written informed consent were obtained. After the third day of treatment, subjects attended at 0830 h, and infusions commenced at 0900 h. Subjects were given cortisol/d4F [20 atoms percent (AP)] as a bolus injection (3.6 mg total), followed by a constant infusion (20 AP, 1.74 mg/h; Fig. 2). Blood samples were withdrawn from the opposing arm at frequent intervals for 4 h. Subjects provided a urine sample before administration of the bolus dose. They were given water (250 ml) to drink and provided further urine samples at hourly intervals. Blood samples were maintained on ice, and plasma and urine were stored at -20 C pending analysis.
|
Plasma. Analysis of steroids in plasma was performed by gas chromatography-mass spectrometry. Plasma (1.5–2 ml) containing epi-cortisol (1 µg) as an internal standard was shaken with ethyl acetate (10 ml). The organic layer was evaporated to dryness under a stream of nitrogen at 60 C. The dried residue was reconstituted in methanol (40 µl), diluted in distilled water (to 2 ml), and passed through a Sep-Pak C18 column [conditioned with methanol (5 ml), then water (5 ml)]. The steroids were eluted in methanol (2 ml), the eluant was reduced to dryness, and the steroids were derivatized as reported previously to form methoxime-trimethylsilyl derivatives (11). Analysis was performed using a Voyager gas chromatograph mass spectrometer with a DB17 column (15 m, 0.25 mm id, 0.25 µm ft; J&W Scientific, Folsom, CA). The initial temperature was 200 C; this was increased by 5 C/min to 250 C and then by 2.5 C/min to 270 C, and finally by 15 C/min to 300 C and maintained for 2 min. Injection, source, and interface temperatures were 290, 200, and 280 C, respectively. Ionization was performed in electron impact mode at 70eV. The following ions were monitored to allow analysis of the methoxime-trimethylsilyl (MO-TMS) derivative of endogenous steroid and the tracer and its metabolites: cortisol, m/z 605; d4F, m/z 609; d3F, m/z 608; cortisone, m/z 531; and d3 cortisone, m/z 534. The ratios of the areas under the peaks of the analyte vs. those of the internal standard (epi-cortisol) were determined and compared with those of known quantities of standard compounds and internal standard over appropriate concentration ranges. The enrichment of steroid total steroid with tracer was determined as the APE (i.e. the amount of tracer divided by the sum of the amounts of tracer and tracee, expressed as a percentage), and this value was corrected for any naturally occurring m+3 and m+4 isotopic components.
Urine.
Urine was extracted as described previously (11) and analyzed as described above, except analysis of tetrahydrosteroids was performed on a DB5MS column (30 m, 0.25 mm id, 0.5 µm ft; J&W Scientific). The following ions were monitored to identify and quantify MO-TMS derivatives of urinary metabolites of endogenous cortisol and cortisone and the tracer: 5ß- and 5-tetrahydrocortisol (5ß- and 5THF), m/z 652; d4 5ß- and 5THF, m/z 656; d3 5ß- and 5THF, m/z 655; tetrahydrocortisone (THB), m/z 581; and d3THB, m/z 584. Ions identical to those detailed for plasma were used to quantify the derivatives of cortisol and cortisone.
Pharmacokinetic analysis and statistics
Absolute concentrations of tracer steroids were calculated from enrichment of each tracer and the concentrations of total endogenous steroid. Curve fitting and rate of appearance calculations were carried out using the Kinetica software package (Innaphase, Champs- sur-Marne, France). Curve fitting was accepted with a coefficients of variation less than 5%, and linear regression with r > 0.95. Data are the mean ± SEM. Mean data obtained at steady state were compared between groups using paired t tests. Accumulations of metabolites with time were compared by repeated measure ANOVA.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mass spectra of the MO-TMS derivatives of cortisol, d4F, and cortisone (Fig. 3) are shown for illustration; the spectra of cortisol and d4F differed only in that the major ions of the deuterated tracer detected were 4 mass units higher than those formed from the derivatized unlabeled steroid. The reproducibility of assays between batches showed coefficients of variation less than 10% for cortisol, cortisone, and their tetrahydrometabolites.
|
The presence of deuteriums on the steroid skeleton did not influence the rates of metabolism of substrate by 11ßHSD2 in vitro (Fig. 4), suggesting a negligible primary isotope effect.
|
Cortisol and cortisone, d4F, d3F, and d3E were detected in plasma, and their major metabolites (unlabeled THF and THF, d4- and d3THF and THF, unlabeled THE and d3THE) were also measured in urine over the 4-h infusion period (Tables 1 and 2).
|
|
and Table 1
. Steady state concentrations of cortisol were achieved during the infusion, and mean data were calculated between 165 and 240 min. A sufficiently high dose of cortisol was administered such that endogenous production of steroid by the adrenal cortex was negligible; indeed, the enrichment of deuterated cortisol in plasma at steady state was, on the average, just in excess of the 20 APE infused. Cortisol pharmacokinetics were modeled with two compartments, with the assumption that the majority of first pass metabolism was irreversible and occurred only from the central pool. Clearance of d4F, was calculated from steady state plasma concentrations (css) and its rate of infusion (Table 1): rate of appearance = css x Cl. Similar calculations to calculate the apparent clearance of cortisol were performed and showed a more rapid rate of elimination of the tracer than the endogenous steroid in both groups (placebo, P = 0.009; carbenoxolone, P = 0.006). However, this difference actually reflects the additional source of appearance of F (i.e. synthesized from cortisone by 11ßHSD type 1). There is not a similar source contributing to the pool of d4F. Hence the rate of elimination of cortisol is slower than that of d4F, although the irreversible steps associated with clearance proceed probably at the same rate.
|
). Indexes of 11ßHSDs
A number of analyses were compared that infer changes in 11ßHSDs with carbenoxolone treatment.
Combined measurement of the isozymes of 11ßHSDs. The following indexes were devised to determine the turnover of d4F into regenerated d3F, i.e. the amount of tracer being metabolized by both isozymes of 11ßHSD.
(a) Enrichment of plasma cortisol pool with deuterated cortisol.
The enrichment of the circulating cortisol pool with tracer (d4F + d3F) increased with time (Fig. 6, A and B), reaching a steady state that was the same between groups (24.9 ± 1.5 vs. 23.2 ± 1.3 APE). The enrichment of the cortisone pool with d3E was the same as that of the cortisol pool with d4F at steady state, indicating that there was not an additional source of cortisone biosynthesis. Enrichment of the cortisone pool with d3E was not altered between groups. However, enrichment of the endogenous cortisol pool with d4F was increased by carbenoxolone (13.6 ± 1.3 vs. 16.3 ± 0.9 APE; P < 0.05; Fig. 7A). Under conditions in which the rate of administration of a tracer are fixed, changes in enrichment reflect alterations in pool size or, more likely, in the rate of appearance of the endogenous tracee. Hence increased enrichment of the plasma pool of F with d4F implies reduced regeneration of endogenous cortisol by 11ßHSD1, and this is demonstrated in that the enrichment of cortisol with d3F declined (10.2 ± 0.9 vs. 6.1 ± 0.7 APE; P < 0.001).
|
|
) (calculated between 180 and 240 min) were lower after carbenoxolone treatment. The rate of appearance of d3F (v11ßHSD1) could be calculated as follows: v11ßHSD1 = css d3F x ClF, where ClF is cortisol clearance. Absolute values for the clearance of d3F could not be determined, because a pure d3F was not available for administration. However, we infer from the lack of change in the clearance of cortisol (Table 1) that the clearance of d3F (Cl d3F) does not change with carbenoxolone. Hence, the observed difference in css d3F (P = 0.01) indicates a slower rate of appearance of d3F after treatment with carbenoxolone.
(c) Rate of appearance of d3F in plasma from rate of increase in concentrations.
The initial increase in the concentration of d3F in plasma with time (Fig. 5C) was slower with carbenoxolone treatment (by repeated measures ANOVA, P = 0.001).
(d) Urinary cortisol (Table 2).
Total cortisol or labeled cortisol (d4 + d3F) in urine did not change after treatment with carbenoxolone (P = 0.10 and P = 0.74, respectively; Table 2). The enrichment of cortisol with tracer increased with time (Fig. 6, C and D), and at steady state the enrichment of the cortisol pool with d4F increased with carbenoxolone treatment (15.8 ± 2.1 vs. 20.0 ± 1.6%; P = 0.01), whereas the enrichment of d3F decreased (10.8 ± 1.3 vs. 6.2± 0.6%; P = 0.006; Fig. 7B).
(e) Urinary cortisol metabolites (Table 2).
The traditional index of overall activity of 11ßHSDs has been the ratio of endogenous THFs to THE, and in this study this was reduced after treatment with carbenoxolone (P = 0.009). Among analyses of deuterated metabolites the most sensitive index of the activities of these enzymes was the ratio of (d4THF + d4THF)/(d3THF + d3THF), which increased substantially with carbenoxolone treatment (P = 0.0001). This was a consequence of lower concentrations of d3- labeled steroids, rather than a difference in d4-labeled steroids.
11ßHSD type 2/11ß-dehydrogenase activity. Having shown the impact of carbenoxolone on the combined enzyme activities, i.e. generation of d3F from d4F, we explored indexes of selective enzyme activities.
(a) Rate of appearance of d3E in plasma from steady state concentrations. The change in concentration of d3E in plasma with time (dc(d3E)/dt) is dependent on the velocity of the reaction catalyzed by 11ßHSD type 2 (v11ßHSD2) and the rate of irreversible clearance (Cld3E) from the central pool. At steady state, dc(d3E)/dt = 0, therefore v11ßHSD2 = c d3E x Cld3E. The concentration of d3E in plasma tended to be lower in subjects receiving carbenoxolone (P = 0.07). However, it was not possible to calculate the rate of appearance, i.e. the velocity of dehydrogenation, from steady state concentrations of d3E, because in the absence of pure d3E for bolus injection we were unable to determine clearance of this steroid.
(b) Rate of appearance of d3E in plasma from initial rates.
The initial rate of formation of d3E by 11ßHSD type 2 was calculated by curve-fitting to a model with two compartments. During the initial phase (0–30 min) it was assumed that d3E was being generated from d4F alone and not regenerated from d3F. Hence, the curve describing the rate of appearance of d3E in plasma by dehydrogenation of d4F by 11ßHSD type 2 with time fitted the following equation: cd3E = Re-
t + Be-ßt + Ae-t, where at any given time Re-t is a function of the amount of d3E being added into plasma as a consequence of 11ßHSD type 2, Ae-t is a function of the amount of d3E being lost from plasma as a consequence of irreversible elimination, and Be-ßt is a function of the amount of d3E being lost from plasma as a consequence of distribution. , ß, and are hybrid rate constants related to appearance, elimination, and distribution.
The initial rates of appearance calculated in this manner from mean data from all six subjects were 0.76 nM/min for placebo vs. 0.50 nM/min for carbenoxolone. These represent the velocity of reaction of 11ßHSD type 2, because at this time elimination and distribution were negligible. However, it was not possible to calculate initial rates of appearance for each subject, because curve fitting was insufficiently precise with the available data points. To correct for the concentration of substrate, these rates were divided through by the sum of the concentrations of F, d4F, and d3F, yielding results as 2.1 vs. 1.4 pM/min.
(c) Urinary cortisol and its metabolites.
The ratio of urinary cortisol to cortisone for both endogenous and labeled steroids increased after treatment with carbenoxolone (Table 2). The ratio of (d4 + d3F)/d3E may reflect the renal activity of 11ßHSD type 2, whereas d4F/(d3E + d3F) may reflect the activity of whole body 11ßHSD type 2.
(d) Urinary metabolite ratios.
Although the ratio of (d4THF + d4THF)/(d3THF + d3THF + d3THE), i.e. metabolites of intact tracer vs. those of the component of the tracer that had been metabolized by 11ßHSD type 2, increased after treatment with carbenoxolone, this did not reach statistical significance (P = 0.21; Table 2).
11ßHSD type 1/11ß reductase activity. Similar approaches were used to assess selective indexes of 11ßHSD type 1/reductase activity.
(a) Rate of appearance of d3F in plasma from steady state concentrations.
To calculate the velocity of 11ßHSD type 1, the individual steady state concentrations of d3F (above) should be corrected for the clearance of d3F; thus, v11ßHSD1 = cssd3F x ClF. As explained above this was not possible, and concentrations were used as surrogates for velocity in the face of no change in cortisol clearance. The velocities of these reactions are dependent on the concentration of substrate (cssd3E) and also endogenous steroid, cssE. The concentrations of d3E at steady state (an index of rates) with carbenoxolone were normalized to the amount of substrate, i.e. E + d3E (cssEd3E) in plasma in the same subject after placebo treatment. The apparent rates thus calculated demonstrated a significant difference (corrected cssd3F 27.7 ± 4.2 vs. 21.5 ± 3.3 nM; placebo vs. carbenoxolone (P = 0.03), indicating that rates of production of d3F by 11ßHSD1 were significantly lower (by 
22%) with carbenoxolone.
(b) Rate of appearance of d3F in plasma from initial rates. Curve fitting as described for d3E was performed using Kinetica software, modeling with two compartments, and the initial rate was calculated by extrapolation to time zero. The rate of appearance of d3F calculated from the mean data for all six subjects was 1.51 nM/min (placebo) vs. 0.68 nM/min (carbenoxolone).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The "proof of principle" experiment identified several strengths and weaknesses of the method. The pharmacokinetic calculations require knowledge of the rate of appearance of total cortisol in steady state. Because of the diurnal variation and response to stress of endogenous cortisol secretion and because of the nonlinear relationship between cortisol secretion and free plasma cortisol concentrations according to availability of corticosteroid-binding globulin-binding sites, we aimed to infuse exogenous unlabeled cortisol to suppress the hypothalamic-pituitary-adrenal axis and occupy a fixed amount of corticosteroid-binding globulin (without saturation). This was successful in these healthy subjects, as no fluctuation of the dilution of tracer d4F was observed by endogenous cortisol secretion in vivo. Care will be required to ensure the same steady state conditions in other circumstances where volume of distribution and total metabolic clearance of cortisol may be altered, e.g. in obesity.
We examined several indexes of turnover of d4F through the 11ßHSD pathway. The advantage of studying kinetics of a tracer with a label in the 11 position, rather than endogenous cortisol or tracer with no 11 label (20), is apparent even without observing the effect of carbenoxolone. The calculated rate of elimination of d4F was almost twice as fast as that of endogenous cortisol because, unlike reversible interconversion of cortisol and cortisone, there is no reactivation of d4F from d3E. This subtlety has been observed previously only with a radioactive tracer, [11-3H]cortisol (14). Administration of carbenoxolone had no measurable effect on cortisol concentrations during the infusion, but, by inhibiting turnover, markedly reduced d3F (measured as lower enrichment, steady state concentrations, or rate of appearance) and increased d4F (measured as higher enrichment and steady state concentrations, or reduced rate of elimination).
Moreover, our data suggest that the technique may be employed in some circumstances to measure turnover of 11ßHSDs without detailed plasma sampling. This would be especially useful in studies involving large numbers and in children. The enrichment of the urinary cortisol metabolite pool with d3 and d4 metabolites followed a similar pattern as the plasma pools. However, we have not confirmed that the urinary metabolite levels reached steady state. Reliance on the urinary indexes in nonsteady state assumes that A ring reduction of cortisol and cortisone and subsequent conversion of these metabolites to cortols and cortolones is unaltered, and this may not be the case. Further extended studies will be required to establish the ideal timing of urinary samples to reflect steady state.
These experiments also illustrate that it is possible to discriminate which isozyme of 11ßHSD is responsible for changes in turnover, but this is more difficult. To measure 11ßHSD type 2/dehydrogenase activity, it would be ideal to calculate rates of appearance of d3E at steady state. However, this calculation requires an estimate of the clearance of E and d3E, which cannot be measured directly without separate administration of d3E (which is not currently available). It is known that the half-life of E is approximately a third of that of F, and hence, extrapolation from F is not appropriate. The steady state concentrations of d3E alone may reflect either production or clearance. Instead, we attempted to extrapolate the initial rate of appearance of d3E to time zero, when distribution and elimination are negligible and d3E is not being formed from d3F as well as from d4F. From the mean data obtained from all six subjects, we showed a reduction in the initial rate of formation of d3E, but were not able to calculate these values for all individuals due to the lack of sufficiently precise data for complex curve fitting. In future studies increased frequency of sampling in the early phase of the study should overcome this.
Again, a less invasive alternative assessment of 11ßHSD type 2 activity may be inferred from the urine. When analyzing endogenous steroids, the best index of 11ßHSD type 2 is the ratio of urinary cortisol to cortisone, thought to reflect specifically intrarenal equilibrium (11, 15). We have been able to refine this by calculating the ratio of d4F (original tracer) to d3E + d3F (all tracer that has been metabolized by dehydrogenation). Although a similar index can be calculated for the ratio of d4 tetrahydrometabolites of cortisol to the sum of the d3 tetrahydrometabolites of cortisol and cortisone, the increase we observed in this ratio was not statistically significant. Measurement of further reduction through to cortols and cortolones, which also contribute to the ideal metabolite ratio, might improve this index.
Finally, a specific index of reactivation of cortisol by 11ßHSD type 1 was calculated. This can be inferred from the change in enrichment of the cortisol pool with d4F. Any changes in this measure during steady state infusion must reflect changes in endogenous biosynthesis of cortisol. As the hypothalamic-pituitary-adrenal axis was suppressed in these subjects, these changes are not due to adrenal production, but, instead, must be due to 11ßHSD type 1. An alternative approach is by measuring the rate of appearance of d3F. An initial rate of appearance of d3F could not be calculated reliably, again as a consequence of insufficient sampling points during this phase. However, unlike d3E, we could apply steady state kinetics to the data for d3F. The steady state concentrations of d3F were significantly lower with carbenoxolone, which could reflect differences in appearance or clearance. It was not possible to measure the clearance of d3F directly, because this steroid is not available for in vivo administration. However, d3F clearance could be extrapolated from clearance of endogenous cortisol, because, like endogenous cortisol, d3F is reversibly interconverted with d3E. A further adjustment was made to account for the differences in concentration of substrate, d3E. This index confirmed that carbenoxolone inhibits 11ßHSD type 1 in vivo (21, 22).
It has been argued that the use of radioactively labeled tracers (e.g 3H or 14C) is ethically unacceptable in an era when stable isotope tracers can be used for many pathways of metabolism. In this report we have confirmed that deuterated cortisol tracer can usefully be employed to measure, for the first time, the turnover in the 11ßHSD pathway and to infer activities of the individual isozymes of 11ßHSD. Crucially, this requires labeling in the 11 position so that substrate of 11ßHSD type 2 and product of 11ßHSD type 1 can be distinguished. We anticipate that this technique will be used extensively to elucidate the alterations in 11ßHSD activities in common disease, the functional importance of genetic polymorphisms of each isozyme, and the effects of pharmacological manipulation of the 11ßHSD system. It may be possible to extend the approach, in combination with tissue-specific sampling, to elucidate tissue-specific changes in 11ßHSD activity obesity (8). Indeed, the advantages of the approach may make it the gold standard against which to test less sensitive and specific measurement of endogenous steroids.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 R.A. and K.S. contributed equally to the work presented in this paper and are represented in alphabetical order. 
Abbreviations: AP, Atoms percent; d3E, trideuterated cortisone; d3F, trideuterated cortisol; d4F, [9,11,12,12-2H4]cortisol; 11ßHSD, 11ß- hydroxysteroid dehydrogenase; MO-TMS, methoxime-trimethylsilyl; THE, tetrahydrocortisone; THF, tetrahydrocortisol.
Received May 21, 2001.
Accepted September 28, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Jang, V. R Obeyesekere, F. P Alford, and W. J Inder Skeletal muscle 11{beta} hydroxysteroid dehydrogenase type 1 activity is upregulated following elective abdominal surgery Eur. J. Endocrinol., February 1, 2009; 160(2): 249 - 255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Basu, A. Basu, M. Grudzien, P. Jung, P. Jacobson, M. Johnson, R. Singh, M. Sarr, and R. A. Rizza Liver Is the Site of Splanchnic Cortisol Production in Obese Nondiabetic Humans Diabetes, January 1, 2009; 58(1): 39 - 45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Stimson, J. Andersson, R. Andrew, D. N. Redhead, F. Karpe, P. C. Hayes, T. Olsson, and B. R. Walker Cortisol Release From Adipose Tissue by 11{beta}-Hydroxysteroid Dehydrogenase Type 1 in Humans Diabetes, January 1, 2009; 58(1): 46 - 53. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Stimson, A. M. Johnstone, N. Z. M. Homer, D. J. Wake, N. M. Morton, R. Andrew, G. E. Lobley, and B. R. Walker Dietary Macronutrient Content Alters Cortisol Metabolism Independently of Body Weight Changes in Obese Men J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4480 - 4484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Michailidou, A. P Coll, C. J Kenyon, N. M Morton, S. O'Rahilly, J. R Seckl, and K. E Chapman Peripheral mechanisms contributing to the glucocorticoid hypersensitivity in proopiomelanocortin null mice treated with corticosterone J. Endocrinol., July 1, 2007; 194(1): 161 - 170. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Wake, R. H. Stimson, G. D. Tan, N. Z. M. Homer, R. Andrew, F. Karpe, and B. R. Walker Effects of Peroxisome Proliferator-Activated Receptor-{alpha} and -{gamma} Agonists on 11{beta}-Hydroxysteroid Dehydrogenase Type 1 in Subcutaneous Adipose Tissue in Men J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1848 - 1856. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Tomlinson, M. Sherlock, B. Hughes, S. V. Hughes, F. Kilvington, W. Bartlett, R. Courtney, P. Rejto, W. Carley, and P. M. Stewart Inhibition of 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Activity in Vivo Limits Glucocorticoid Exposure to Human Adipose Tissue and Decreases Lipolysis J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 857 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Wake, N. Z. M. Homer, R. Andrew, and B. R. Walker Acute In Vivo Regulation of 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Activity by Insulin and Intralipid Infusions in Humans J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4682 - 4688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Basu, D. S. Edgerton, R. J. Singh, A. Cherrington, and R. A. Rizza Splanchnic Cortisol Production in Dogs Occurs Primarily in the Liver: Evidence for Substantial Hepatic Specific 11{beta} Hydroxysteroid Dehydrogenase Type 1 Activity Diabetes, November 1, 2006; 55(11): 3013 - 3019. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Garbrecht, T. J. Schmidt, Z. S. Krozowski, and J. M. Snyder 11beta-Hydroxysteroid dehydrogenase type 2 and the regulation of surfactant protein A by dexamethasone metabolites Am J Physiol Endocrinol Metab, April 1, 2006; 290(4): E653 - E660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Basu, R. Singh, A. Basu, C.M. Johnson, and R. A. Rizza Effect of Nutrient Ingestion on Total-Body and Splanchnic Cortisol Production in Humans Diabetes, March 1, 2006; 55(3): 667 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Basu, R. J. Singh, A. Basu, E. G. Chittilapilly, M. C. Johnson, G. Toffolo, C. Cobelli, and R. A. Rizza Obesity and Type 2 Diabetes Do Not Alter Splanchnic Cortisol Production in Humans J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 3919 - 3926. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Andrew, J. Westerbacka, J. Wahren, H. Yki-Jarvinen, and B. R. Walker The Contribution of Visceral Adipose Tissue to Splanchnic Cortisol Production in Healthy Humans Diabetes, May 1, 2005; 54(5): 1364 - 1370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Sandeep, R. Andrew, N. Z.M. Homer, R. C. Andrews, K. Smith, and B. R. Walker Increased In Vivo Regeneration of Cortisol in Adipose Tissue in Human Obesity and Effects of the 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Inhibitor Carbenoxolone Diabetes, March 1, 2005; 54(3): 872 - 879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Basu, R. J. Singh, A. Basu, E. G. Chittilapilly, C. M. Johnson, G. Toffolo, C. Cobelli, and R. A. Rizza Splanchnic Cortisol Production Occurs in Humans: Evidence for Conversion of Cortisone to Cortisol Via the 11-{beta} Hydroxysteroid Dehydrogenase (11{beta}-HSD) Type 1 Pathway Diabetes, August 1, 2004; 53(8): 2051 - 2059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Andrews, O. Rooyackers, and B. R. Walker Effects of the 11{beta}-Hydroxysteroid Dehydrogenase Inhibitor Carbenoxolone on Insulin Sensitivity in Men with Type 2 Diabetes J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 285 - 291. [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 |
, Endogenous steroids; 
, tracer steroids.
) and d3F (
) and cortisone with d3E (
) in plasma from subjects receiving placebo (A), plasma with subjects receiving carbenoxolone (B), urine from subjects receiving placebo (C), and urine from subjects receiving carbenoxolone (D).