11β-Hydroxysteroid Dehydrogenase Type 1 Regulation by Intracellular Glucose 6-Phosphate Provides Evidence for a Novel Link between Glucose Metabolism and Hypothalamo-Pituitary-Adrenal Axis Function*

Microsomal glucose-6-phosphatase-α (G6Pase-α) and glucose 6-phosphate transporter (G6PT) work together to increase blood glucose concentrations by performing the terminal step in both glycogenolysis and gluconeogenesis. Deficiency of the G6PT in liver gives rise to glycogen storage disease type 1b (GSD1b), whereas deficiency of G6Pase-α leads to GSD1a. G6Pase-α shares its substrate (glucose 6-phosphate; G6P) with hexose-6-phosphate-dehydrogenase (H6PDH), a microsomal enzyme that regenerates NADPH within the endoplasmic reticulum lumen, thereby conferring reductase activity upon 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1). 11β-HSD1 interconverts hormonally active C11β-hydroxy steroids (cortisol in humans and corticosterone in rodents) to inactive C11-oxo steroids (cortisone and 11-dehydrocorticosterone, respectively). In vivo reductase activity predominates, generating active glucocorticoid. We hypothesized that substrate (G6P) availability to H6PDH in patients with GSD1b and GSD1a will decrease or increase 11β-HSD1 reductase activity, respectively. We investigated 11β-HSD1 activity in GSD1b and GSD1a mice and in two patients with GSD1b and five patients diagnosed with GSD1a. We confirmed our hypothesis by assessing 11β-HSD1 in vivo and in vitro, revealing a significant decrease in reductase activity in GSD1b animals and patients, whereas GSD1a patients showed a marked increase in activity. The cellular trafficking of G6P therefore directly regulates 11β-HSD1 reductase activity and provides a novel link between glucose metabolism and function of the hypothalamo-pituitary-adrenal axis.

A vital function of the liver is to provide glucose during fasting. This occurs through two principal pathways, gluconeogenesis and glycogenolysis, in each case yielding glucose 6-phosphate (G6P), 5 which is then hydrolyzed by glucose-6-phosphatase (G6Pase), more recently described as G6Pase-␣ (1), to glucose (reviewed in Ref. 2). G6Pase-␣ is a transmembrane protein in the endoplasmic reticulum (ER) with the active enzyme site directed toward the ER lumen (2,3). G6P must be translocated from cytosol to ER via a ubiquitously expressed G6P transporter (G6PT) before hydrolysis can occur. Patients and rodent models lacking components of the G6Pase-␣, G6PT system emphasize the crucial role of this pathway in maintaining glucose homeostasis in the fasting state. Profound fasting hypoglycemia occurs in glycogen storage disease type 1; von Gierke's disease (type 1a) caused by G6Pase-␣ deficiency; and type 1b caused by a G6PT defect (2,3). Recombinant mice with global deletion of G6Pase-␣ (4) and G6PT (5) have similar phenotypes with profound hypoglycemia.
G6Pase-␣ shares its substrate (G6P) with another enzyme within the ER, hexose-6-phosphate-dehydrogenase (H6PDH), that catalyzes the first two steps of an ER-specific "pentose phosphate pathway," i.e. both G6P dehydrogenase and 6-phosphogluconolactonase reactions ( Fig. 1) (6,7). Our recent studies have indicated a pivotal link between H6PDH activity in the ER and the control of set point of 11␤-hydroxysteroid dehydrogenase type 1 (11␤-HSD1). 11␤-HSD1 is an ER-bound enzyme (8) catalyzing the interconversion of inactive glucocorticoids (cortisone in humans and 11-dehydrocorticosterone in rodents) and hormonally active glucocorticoids (cortisol and corticosterone) (9). The reaction direction, which 11␤-HSD1 catalyzes, is determined by the relative abundance of NADPϩ and NADPH (10). In its native purified state, 11␤-HSD1 acts as * This work was supported by Programme grant support (Ref. 066357) and Project grant support (Ref. 074088) from the Wellcome Trust (to E. A. W. and P. M. S.) and by National Institutes of Health support Grant 1Z01HD000912-27 (to J. Y. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Both authors contributed equally to the work. 2  a dehydrogenase-inactivating cortisol/corticosterone to cortisone/11-dehydrocorticosterone (11-DHC) (10,11). However, in the presence of reducing NADPH, generated through H6PDH activity (12), 11␤-HSD1 switches to reductase with the generation of active glucocorticoid in key tissues such as liver and adipose (13)(14)(15). In these tissues, this reductase activity has been shown to enhance hepatic glucose output (16) and adipogenesis (17), respectively. 11␤-HSD1 is thus an exciting candidate to explain features of obesity and the metabolic syndrome with selective inhibitors under active development that may prevent and/or reverse diabetes mellitus in obese subjects (18). Previously, we and others have shown that H6PDH is an essential requirement for 11␤-HSD1 reductase activity; here we have explored whether G6P availability to H6PDH could directly modulate 11␤-HSD1 activity, thereby representing a novel pathway linking the metabolism of glucose and glucocorticoids.

EXPERIMENTAL PROCEDURES
Preparation of Mouse Liver Microsomes-Mouse liver microsomes were prepared from recombinant male mice (n ϭ 3) with global deletion of G6Pase-␣, or separately, deletion of G6PT by differential centrifugation as described previously (19). The phenotype and background strain of these mice have been pre-viously reported (4,5). All animal experiments had the approval of the Institutional Animal Care and Use Committee and were performed according to procedures approved by that committee. Microsomal fractions were resuspended in a buffer containing 20 mM NaCl, 1 mM MgCl 2 , 100 mM KCl, 20 mM Mops, pH 7.2, and were snap-frozen under liquid nitrogen. Microsomal protein concentration was determined using the Bio-Rad protein assay with bovine serum albumin as a standard as per the manufacturer's instructions (Bio-Rad). The integrity of the microsomal membranes was assessed by using the mannose-6phosphatase assay (20), which showed a latency greater than 95% in all preparations.
11␤-HSD1 Enzyme Activity Assays-Microsomes (30 g) were preincubated at 37°C for 20 min in Mops buffer with 1 mM of the H6PDH substrate G6P. 11␤-HSD enzyme reactions were started by the addition of 200 nM 11-dehydrocorticosterone/ 500 nM corticosterone spiked with 20,000 cpm of tritiated 11-dehydrocorticosterone/corticosterone. All experiments were performed in triplicate. After incubation at 37°C for 30 min, steroids were extracted with dichloromethane, separated by thin layer chromatography using a mobile phase of ethanol and chloroform (8:92), and quantified using a Bioscan 2000 image analyzer (Lablogic, Sheffield, UK) (19). The percentage of substrate metabolized in each experiment was 10% or less, ensuring that initial rates of metabolism were being measured.
Immunoblotting-SDS-PAGE was performed by the method of Laemmli (21) with 10 g of mouse liver microsomal protein on 11% acrylamide minigels using a Bio-Rad Mini-PROTEAN II apparatus (Bio-Rad). Following electrophoresis, proteins were transferred to Immobilon-P membrane (Millipore Corp., Bedford, MA). Nonspecific protein binding was blocked by incubating membranes in 20% nonfat milk, 0.1% Tween 20 in phosphate-buffered saline at 25°C for 1 h. Membranes were then incubated with an in-house raised polyclonal antibody to human H6PDH at a dilution of 1:1000 for 16 h at 4°C. Following 3 ϫ 10-min washes in phosphate-buffered saline, 0.1% Tween 20, membranes were incubated with secondary antibody (goat anti-rabbit IgG peroxidase-conjugate) at a dilution of 1:25,000 for 1.5 h at room temperature. Bound peroxidase-conjugated IgG was visualized using ECL detection kit (Amersham Biosciences, Buckinghamshire, UK) by exposing membranes to x-ray film (Kodak, France). Following stripping, membranes were reprobed with a polyclonal antibody to human 11␤-HSD1 (1:1000) (22) in a similar method as above and incubated with goat anti-sheep IgG peroxidase secondary antibody.
Analysis of Urine from GSD1a and GSD1b Mice-Urine was collected on filter paper following bladder massage, and samples were pooled from three of each group: wild type (WT), GSD1b, and GSD1a. Three individual pooled samples were collected from WT and GSD1b, and a single pooled sample was analyzed from the GSD1a mice. The filter papers were cut up and eluted with water while vortexing and sonicating. Extracts were subjected to steroid analysis by gas chromatography/mass spectrometry as described previously (19,23,24). Multiple corticosterone metabolites were found in these analyses. Major metabolites were 6␤-hydroxy and 20-dihydro metabolites of corticosterone. Prominent saturated components were hydroxylated (6␣or 11␤-) derivatives of "tetrahydro" FIGURE 1. The microsomal G6Pase-␣, H6PDH, and 11␤-HSD1 systems and the hypothalamus-pituitary-adrenal axis. G6P entering the ER lumen via G6PT either is a substrate for G6Pase-␣ and undergoes hydrolysis to glucose (G) and inorganic phosphate (Pi) or is utilized by hexose-6-phosphate dehydrogenase (H6PDH). H6PDH converts G6P to 6-phosphogluconate (6PGL), and in doing so, it generates NADPH for 11␤-HSD, enabling it to convert inactive glucocorticoid, cortisone (11-dehydrocorticosterone in rodents), to active glucocorticoid cortisol (corticosterone in rodents). At an endocrine level, production of adrenocorticotropic hormone (ACTH) from the anterior pituitary stimulates the adrenal glands to produce cortisol. This forms a negative feedback (Ϫve feedback) loop to switch off cortisol production, thereby maintaining normal circulating levels.
(3␣-, 5␣-, or 3␣,5␤-) or "hexahydro" (additionally reduced at 20␣or 20␤-) corticosterone or 11-DHC. The basic structure of each compound and establishment of it as a corticosterone or 11-DHC metabolite was easily determined by mass spectral fragmentation, although stereochemistry of the 5-hydrogen and the 20-hydroxyl could not be determined in the absence of authentic compounds. Quantitation was achieved by measuring the total-ion-current response for each peak and relating this to a known amount of internal standard, assuming an equal mass spectrometry response. For the purpose of this study, excretions of individual 11-oxo and 11␤-hydroxy metabolites were separately summed, and the percentage of excretion of each group was calculated.
Patient Studies-Five patients with GSD1a were investigated (two males, three females; mean age Ϯ S.D., 28 Ϯ 1 years). All patients gave formal written consent for the study, which was approved by the local hospital ethics committee and carried out according to the recommendations of the declaration of Helsinki. All patients were diagnosed in early infancy by liver biopsy and G6Pase-␣ enzyme assay after presenting with severe hypoglycemia.
Urine steroid metabolite analysis was carried out on 24-h urine collections from five patients with GSD1a. In addition, we were able to obtain single spot urine samples from two children with GSD1b. All samples were analyzed by gas chromatography/mass spectrometry, as reported previously (23,25), measuring free and conjugated cortisol metabolites. Urinary steroid metabolite ratios and total 24-h cortisol metabolite production rates are presented in Table 1. GSD1a patients have been compared with 36 health controls (mean age Ϯ S.D., 33 Ϯ 8 years; mean body mass index Ϯ S.D., 28.4 Ϯ 5.6 kg/m2). Individual values are presented for the two GSD1b patients, alongside the age-adjusted reference ranges ( Table 1). The THFϩ5␣THF/THE, the cortols/cortolones, and the 11OH-androsteroneϩ11OH-etiocholanolone/11-oxoetiocholanolone ratios represent acknowledged markers of global 11␤-HSD1 activity, with a high ratio indicating increased 11␤-HSD reductase activity.
At 2300 h, GSD1a subjects were given 1 mg of dexamethasone orally to suppress endogenous cortisol production. All subjects attended the Clinical Research Facility at 0900 h the following morning, and after baseline 0900 h measurements of cortisol and adrenocorticotropic hormone, a further 0.5 mg of dexamethasone and cortisone acetate (25 mg) were given orally. Serum cortisol and cortisone concentrations were then measured at 30-min intervals for 240 min. Serum cortisol was assayed using a chemiluminescent immunoassay (Bayer Advia Centaur, Bayer Diagnostics, Newbury, UK) with interassay coefficients of variation of 10.2% at 76 nmol/liter, 7.7% at 528 nmol/liter, and 7.4% at 882 nmol/liter. Cortisone was assayed after extraction from serum followed by radioimmunoassay of the extract with 125 I-cortisone and Sac-Cel (IDS Ltd., Tyne and Weir, UK) second antibody separation. Results were compared with data from 34 age and body mass index-matched controls from our existing data base of normal controls.
Statistical Analysis-Statistical analysis of comparisons among groups was undertaken using the one-way analysis of variance with Tukey's post hoc testing (for normal distribution) or Mann-Whitney rank sum test (for non-normal distribution). Area under the curve (AUC) analysis was performed using the trapezoidal method. All analyses were performed using the SigmaStat 3.1 software package (Systat Software, Inc. Point Richmond, CA).

RESULTS
In Vitro Analysis of 11␤-HSD1 Activity and Protein Expression in GSD1b and GSD1a Mice-Liver microsomal preparations have been extensively used to assess the kinetics of all the enzyme systems analyzed in this study and are considered a representative model system of the ER (3,13,27,28). 11␤-HSD1 enzyme activity was assessed by examining the 11-reductase (11-DHC to corticosterone) and dehydrogenase (corticosterone to 11-DHC) activity in mouse liver microsomes from WT, GSD1b, and GSD1a animals.
In hepatic microsomes from the WT animals, reductase activity predominated ( Fig. 2A) and was significantly higher than dehydrogenase activity (p Ͻ 0.001; Fig. 2A). In the microsomes of GSD1b animals, where there is no transporter protein present for G6P, reductase activity was significantly lower than that from WT animals (p Ͻ 0.001; Fig. 2A), and dehydrogenase activity was at a similar level to WT controls. Expression of H6PDH and 11␤-HSD1 protein was similar in WT and GSD1b animals (Fig. 2B), indicating that changes in the levels of these proteins could not account for the differences in activity.
In the GSD1a mice, both reductase and dehydrogenase activities were reduced when compared with WT animals. This unexpected finding was explained by a reduction in expression of 11␤-HSD1 protein in these animals, despite no change in expression of H6PDH (Fig. 2B). 11␤-HSD1 reductase and dehydrogenase activity and protein expression from liver microsomes of WT, GSD1b, and GSD1a mice. A, in the WT microsomes, reductase activity (black bars) was significantly higher than dehydrogenase activity (white bars). Reductase activity from GSD1b microsomes was significantly lower when compared with WT and at a similar level to dehydrogenase activity. Microsomes prepared from livers of GSD1a had much lower levels of reductase and dehydrogenase activity when compared with WT, but within these animals, reductase activity still predominates. Values indicate mean activity Ϯ S.E.; n ϭ 3 for each group. ***, p Ͻ 0.001. B, SDS-PAGE Western blot analysis revealed similar expression levels of H6PDH protein across all animals; however, levels of 11␤-HSD1 protein were lower in the GSD1a.

11␤-HSD1 Reductase Activity Is Regulated by G6P Availability
In Vivo Analysis of Corticosterone Metabolism in GSD1b and GSD1a Mice-In vivo assessment of 11␤-HSD1 activity in mice was carried out using gas chromatography/mass spectrometry analysis of pooled urine collections (n ϭ 3) from each group. WT mice were found to excrete almost exclusively 11␤-hydroxy metabolites (92.3 Ϯ 3.4%; mean Ϯ S.E.) with only minor amounts of 11-oxo metabolites (Fig. 3A). The converse was true for the GSD1b mice, where the dominant steroids excreted in the urine were 11-oxo metabolites (92.3 Ϯ 3.4% WT versus 60.2 Ϯ 4.6%; GSD1b p Ͻ 0.001; Fig. 3A). This pattern of metabolism was mirrored in vitro, where the ratio of reductase to dehydrogenase activity obtained from liver microsomal preparations of GSD1b mice was significantly lower than WT (5.21 Ϯ 0.64 WT versus 1.13 Ϯ 0.28 GSD1b p Ͻ 0.01; Fig. 3B). Results from a single pooled urine sample from GSD1a mice indicated a similar level of 11␤-hydroxy metabolites to WT (87.5%; Fig.  3A). This was consistent with the relative ratio of reductase: dehydrogenase activity seen in vitro, being comparable with that of WT animals (5.21 Ϯ 0.64 WT versus 5.7 Ϯ 0.7 GSD1a; Fig. 3B).

In Vivo Analysis of Cortisol Metabolism in GSD1a and GSD1b Patients, Urinary Cortisol Metabolite Excretion-In
GSD1a patients, total cortisol production rate, as determined from the summation of the metabolite excretion in a 24-h urine collection (29), was significantly decreased (Table 1). In addition, the THFϩ5␣THF/THE, the cortols/cortolones, and the 11OH-androsteroneϩ11OH-etiocholanolone/11-oxo-etiocholanolone ratios, all of which reflect global 11␤-HSD1 activity, were significantly increased, in keeping with an increased cortisone to cortisol conversion. Absolute levels of THF and THE were significantly lower than in control subjects; however, there was little change in the relative activity of either 5␣-reductase or 5␤-reductase as measured by the THF/5␣THF and etiocholanolone/androsterone ratios (Fig. 4), although it is possible that there is a reduction in both 5␣-reductase and 5␤-reductase activity. Small numbers of cases of GSD1b patients precluded statistical analysis, but in contrast to patients with GSD1a, in two cases of GSD1b, we observed a relative decrease in 5␣ and/or increase in 5␤-reductase activity (increased THF/ 5␣THF ratio). However, the etiocholanolone/androsterone ratio was within the reference range. Importantly, 11␤-HSD1 activity, as measured by the THFϩ5␣ THF/THE and the cortols/cortolones ratios, was decreased (Table 1). These were spot urines from overnight collections, and we were unable therefore to assess 24-h cortisol metabolite secretion rates.

DISCUSSION
This work presents compelling evidence for the critical importance of G6P concentrations in the ER lumen in determining the "set point" of 11␤-HSD1 activity. It further endorses recent findings of a close functional cooperativity between 11␤-HSD1 and H6PDH (13,14,31) and the existence of a separate pool of NADP/H within the ER (15, 32, 33).
The studies on humans lacking G6Pase-␣ (GSD1a) show for the first time an impressive increase in 11␤-HSD1 reductase activity in vivo, resulting in increased local generation of active glucocorticoid. In these patients, G6P availability to H6PDH is increased, resulting in enhanced NADPH generation for 11␤-HSD1. Activity was inferred in humans through an increase in the urinary THFϩ5␣THF/THE, cortols/cortolones, and 11OH-androst-eroneϩ11OH-etiocholanolone/11oxo-etiocholanolone ratios and increased generation of circulating cortisol following an oral dose of cortisone acetate, with unchanged markers of 5␣-reductase and 5␤-reductase activity. We and others have extensively utilized the cortisol generation test post-cortisone as a marker of liver 11␤-HSD1 activity; a reduced generation of cortisol has been observed in patients with obesity (30) and in subjects with apparent cortisone reductase deficiency (34). However, this is the first time an increase in cortisol concentrations following cortisone has been observed. The normal ratio of urinary C11␤-hydroxy:C11-oxo metabolites of glucocorticoids in humans is ϳ1:1, but in mice, this ratio is ϳ20:1, reflecting a more efficient reductase enzyme (19). As a consequence, a further increase in the percentage of urinary 11␤hydroxy metabolites was not observed in mice lacking G6Pase-␣. However, in hepatic microsomes from mice lacking G6Pase-␣, absolute levels of both reductase and dehydrogenase activities were lower than WT controls. Since both reductase and dehydrogenase activities were reduced by the same extent, the ratio of these activities remained unchanged. This was explained by an overall reduction in expression of 11␤-HSD1 protein in the GSD1a liver microsomes. The down-regulation of 11␤-HSD1 in the liver of these animals may represent a negative feedback mechanism whereby an attempt is made to increase G6P hydrolysis by limiting the requirement for G6P by H6PDH, although this hypothesis requires further investigation.
Conversely, when delivery of G6P to the ER is compromised, as seen in mice and humans lacking G6PT (GSD1b), a significant impairment of 11␤-HSD1 reductase activity was observed. It was not possible to undertake cortisol generation profiles following oral cortisone in two patients with GSD1b, but the urinary THFϩ5␣THF/THE ratio was reduced by 71% when compared with controls. Similar changes were seen in mice, where the percentage of 11-oxo metabolites fell from 92.3 to 39.8%. Our in vitro data indicated that this was a direct result of impaired reductase activity. These data compare favorably with those obtained from our H6PDH knockout mouse. In these animals, we also found that the set point of 11␤-HSD1 activity switched from reductase to dehydrogenase with greater dehydrogenase activity evident from liver microsomal preparations Markers of 11␤-HSD1 activity (cortols/cortolones, THFϩ5␣THF/THE, 11OH-andϩ11OH-etio/11-oxo-etio; mean ratio Ϯ S.E.) were significantly increased in the GSD1a patients, whereas indicators of 5␣ and 5␤-reductase activity were unchanged. and, androsterone; etio, etiocholanolone. The data represent the mean Ϯ S.E. of three individuals with GSD1a (two premenopausal females and one male) and compared with data from 34 age and body mass index-matched controls. A, the GSD1a group showed a significant increase in serum cortisol concentrations (mean cortisol 248 Ϯ 3 versus 75 Ϯ 4 mol/liter⅐min, p Ͻ 0.001), indicating enhanced production of cortisol and therefore increased 11␤-HSD1 reductase activity. B, there were no significant changes in serum cortisone concentrations between the controls and the GSD1a group (mean cortisone 15 Ϯ 3 versus 13 Ϯ 5 mol/ liter⅐min, p ϭ 0.4).

11␤-HSD1 Reductase Activity Is Regulated by G6P Availability
and a higher percentage of 11-oxo metabolites present in the urine (19).
Glycogen storage disease type 1 is a complex liver disorder, and both GSD1a and GSD1b patients manifest the symptoms of failed G6P hydrolysis, characterized by growth retardation, hypoglycemia, hepatomegaly, nephromegaly, hyperlipidemia, hyperuricemia, and lactic academia (2). However, there are distinct phenotypic differences between GSD type 1a and 1b, which are not obviously related to G6P metabolism in gluconeogenic tissues, and local alterations in 11␤-HSD1 activity may explain some of these differences. The pathogenesis of osteoporosis in GSD1a, which is not present in GSD1b, has been studied in some detail and demonstrates many similarities with glucocorticoid-induced bone loss (35,36). Additionally, one unique feature of GSD1b that is not seen in GSD1a is neutropenia (37)(38)(39)(40). G6PT is widely distributed, but G6Pase-␣ is not present in neutrophils (2), suggesting that the neutropenia is in some way dependent upon loss of substrate for H6PDH. It remains to be seen whether this is secondary to a reduction in glucocorticoid generation in bone marrow precursors.
However, perhaps the most important clinical implications of these findings relates to the putative role of 11␤-HSD1 in the pathogenesis and treatment of patients with obesity-metabolic syndrome. Inhibition of 11␤-HSD1 lowers glucocorticoid levels in liver and adipose, thereby reducing hepatic gluconeogenesis and hepatic glucose output and adipogenesis (17,41). In each case, the assumption is that 11␤-HSD1 reductase activity is inhibited; our data highlight the key role of G6P delivery and H6PDH activity within the ER to direct this activity.
These studies also offer an exciting link between cellular glucose disposal, local glucocorticoid metabolism, and the function of the hypothalamo-pituitary-adrenal axis. During hyperglycemia, cytosolic G6P concentrations increase (42). In addition, under these conditions, free fatty acids limit the availability of G6P (42). As a consequence, this would increase G6P availability to the ER lumen, providing substrate for H6PDH, which would, in turn, drive 11␤-HSD1 reductase activity and increase local glucocorticoid regeneration. In support of this hypothesis, within 3 h of a mixed meal, both 11␤-HSD1 reductase activity (43) and total cortisol production rates increase (43)(44)(45)(46). Furthermore, at a cellular level, the concentration of glucose within cell culture medium has a profound effect on the directionality of 11␤-HSD1, omission of glucose leading to decreased 11␤-HSD1 reductase activity (47). This hypothesis may also offer an explanation for observations made in other clinical studies. We have previously observed a failure to downregulate 11␤-HSD1 activity (as measured by the urinary THFϩ5␣ THF/THE ratio) in obese patients with type 2 diabetes (48). In these patients, hyperglycemia, resulting in an increase in cytosolic and endolumenal G6P concentrations, would drive 11␤-HSD1 reductase activity and explain the relative increase in the urinary THFϩ5␣ THF/THE ratio seen. In addition, it has been reported that selective 11␤-HSD1 inhibitors are more efficacious in rodent models of hyperglycemia. This could be explained by the hyperglycemia in these animals increasing reductase activity in key target tissues, thereby making them more responsive to selective 11␤-HSD1 inhibition (26). Further studies are now required to determine the impact of short term fasting and feeding upon glucose trafficking, cortisol metabolism, and specifically, the impact upon 11␤-HSD1 activity.
In summary, human and mouse models of GSD elicit dramatic changes in 11␤-HSD1 activity with induction observed in GSD1a and loss of reductase activity in GSD1b. These studies highlight the importance of a novel metabolic pathway involving G6P metabolism via H6PDH and the regulation of redox potential within the ER, linking cellular glucose metabolism to the function of the hypothalamo-pituitary-adrenal axis.