| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology Unit, Department of Medical Sciences, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland, United Kingdom
Address all correspondence and requests for reprints to: Prof. Brian R. Walker, Endocrinology Unit, Department of Medical Sciences, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, Scotland, United Kingdom. E-mail: b.walker{at}ed.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In brains from young adult male rats, in situ hybridization and Western blotting showed that obese rats had normal hippocampal GR mRNA and protein levels. In contrast, in obese rats, 11ßHSD1 mRNA levels were reduced in a subpopulation of hippocampal cells in the main neuronal layers (by 37–47%, P < 0.05), whereas 11ßHSD1 levels in sparse high-expressing cells did not differ. MR mRNA was decreased in all regions of the hippocampus (by 37–49%, P < 0.05 for CA1–2 and P < 0.01 for dentate gyrus) and in frontal cortex (by 16%, P < 0.05) in obese rats. In whole hippocampal homogenates, however, neither the protein concentration of MR by Western blot nor activity of 11ßHSD1 was measurably different between the phenotypes. To test the functional importance of lower central MR expression, groups of lean and obese rats were given spironolactone before restraint stress. In vehicle-treated animals, obese rats had higher plasma corticosterone levels than lean rats after stress (by ANOVA, P < 0.05). Spironolactone markedly increased the corticosterone response in both groups, but the incremental rise was smaller in the obese rats, so that spironolactone abolished the differences between groups.
We conclude that lower levels of MR, but not GR, contribute to the increased HPA activity in the obese Zucker rats and that this seems more influential during stress than in the basal state. This may be exacerbated by impaired local regeneration of corticosterone by 11ßHSD1. These abnormalities could contribute to the subtle changes in the HPA axis in rodent and human obesity.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HPA axis negative feedback is mediated in hippocampus, hypothalamus, and pituitary by corticosterone binding to low-affinity, high-capacity glucocorticoid receptors (GR) and to high-affinity, low-capacity mineralocorticoid receptors (MR). The basal activity is maintained by the proactive feedback of MR; whereas the response to acute stress, the reactive feedback, is mediated by GR in addition to MR (17). In addition, the availability of corticosterone to bind GR and MR in central feedback sites is influenced by local regeneration of corticosterone from 11-dehydrocorticosterone by 11ß-hydroxysteroid dehydrogenase type 1 (11ßHSD1) (18), which is expressed in hippocampus, hypothalamus, and anterior pituitary (19). Transgenic deletion of 11ßHSD1 results in mice with low intrahippocampal concentrations of corticosterone (20) and, hence, impaired negative feedback and elevated plasma levels of corticosterone and ACTH (21).
Previous studies of HPA axis feedback in obese Zucker rats have focused on GR. However, results are inconsistent. GR dissociation constant and binding maximum in the brain of obese rats have been reported to be unchanged or increased (22, 23, 24). Treatment with the GR antagonist RU486 has been reported to increase (25), decrease (6), or not affect (26, 27) plasma corticosterone levels in obese animals.
The influence of MR and 11ßHSD1 on HPA feedback has not been evaluated previously in animal models of obesity. Recent studies in obese men suggest resistance to negative feedback that is specific to cortisol rather than dexamethasone and affects the HPA during the diurnal nadir rather than the diurnal peak, suggestive of MR (rather than GR) dysfunction (28). Regarding 11ßHSD1, peripheral tissue-specific abnormalities have been documented in obese Zucker rats, with decreased 11ßHSD1 enzyme activity and mRNA levels in liver and increased activity in visceral adipose tissue (7). Similar changes have been found in liver and adipose of overweight men and women (29, 30, 31, 32), but central nervous system (CNS) 11ßHSD1 has not been evaluated in obesity.
In the present study, we evaluated mRNA for MR, GR, and 11ßHSD1 in brain regions important for the regulation of the HPA axis and found a decrease of hippocampal mRNA encoding both MR and 11ßHSD1. The significance of the diminished MR expression was examined by measuring the corticosterone response to stress with and without the MR antagonist spironolactone.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For dynamic tests, a further 11 lean (196 ± 4 g) and 12 obese (262 ± 5 g) male Zucker rats (Harlan Orlac) were used. The housing conditions were similar to those described above. For 2 wk before the experiment, the animals were handled and injected each day with 0.2 ml 0.9% saline sc. On the experimental day, groups of 6 animals of each phenotype were given spironolactone (Sigma, Gillingham, UK; 100 mg/kg sc) or vehicle (propyleneglycole; Sigma; 1 ml/kg sc) at 0800 h. One hour later, the animals were subjected to restraint stress in Plexiglas tubes for 30 min. Blood samples for corticosterone measurements were taken, by tail clip, immediately before the onset of stress and at 30, 60, and 120 min after the onset of stress.
In situ hybridization for MR, GR, and 11ßHSD1
Ten-micrometer sections were cut through frontal cortex, pituitary, paraventricular nucleus, and hippocampus. The sections were thawed-mounted on gelatin/poly-L-lysine-coated slides and kept at -80 C. In situ hybridization, using [35S]-uridine 5'-triphosphate-labeled cRNA antisense probes for GR, MR, and 11ßHSD1, was performed as previously described (33). Templates for transcribing riboprobes were generated by linearizing plasmids containing fragments of rat cDNA for GR, MR, and 11ßHSD1, as described earlier (19, 34), with the adaption that restriction enzyme Sty1 was used to linearize the vectors containing 11ßHSD cDNA insert. Before hybridization for 11ßHSD1, the tissues were prehybridized in buffer containing 50% formamide, 0.6 M NaCl, 0.01 M Tris (pH 7.5), 1x Denhardt’s solution, and 0.125 µg/ml yeast transfer RNA. Two hundred microliters of prehybridization solution was applied to the sections/slide and then incubated in humid boxes at 50 C for 2 h. Two hundred microliters of radiolabeled probe (10 x 106 dpm/ml in hybridization buffer) was added to each slide. After hybridization and high stringency washes, slides were dehydrated, dipped in NTB-2 emulsion (Eastman Kodak Co., Rochester, NY), and exposed for 3 wk (MR, GR) or 4 wk (11ß-HSD1) at 4 C and thereafter developed and counterstained with 1% pyronin Y. The mRNA expression was quantified by counting silver grains over individual neurones under bright field conditions using a computerized image analysis system (Carl Zeiss, Welwyn Garden City, UK). Background counted over areas of white matter was subtracted. In frontal cortex, layer 2 was analyzed.
11ßHSD1 enzyme activity in the hippocampus
Hippocampus was homogenized in C-buffer (8% glycerol, 150 mM NaCl, 1 mM EDTA, 50 mM Tris), whereafter 0.25 mg of protein was incubated with 200 nM nicotinamide adenine dinucleotide phosphate and 100 nM 1,2,6,7,3H-corticosterone (specific activity, 84 Ci/mmol; Amersham International, Ayelsbury, UK) in a total vol of 250 µl C-buffer for 30 min at 37 C. The steroids were extracted with ethyl acetate and separated by HPLC with on-line ß-counting. The 11ßHSD1 activity was calculated as percentage conversion from 3H-corticosterone to 3H-11-dehydrocorticosterone. Enzyme activity was measured in the dehydrogenase (rather than reductase) reaction because this is the preferred direction when it is liberated from its intracellular environment (18); conditions chosen ensured a linear increase in dehydrogenase activity with increasing enzyme protein.
Western blot of GR and MR levels in the hippocampus
Hippocampi were homogenized in ice-cold lysis buffer containing 20 mM HEPES, 10 mM potassium chloride, 20 mM sodium molybdate, 1 mM EDTA, 1 mM EGTA, 25 µM sodium vanadate, 0.2% Nonidet P-40, 10% glycerol, 1 mM leupeptin, 1 µg/ml aprotinin, and 1 µg/ml pepstatin. Homogenates were microcentrifuged for 5 min at 4 C, and protein concentrations in the supernatants were determined colorimetrically. Supernatants were mixed with Laemmeli’s sample buffer and boiled for 5 min. Samples (30 µg protein) were loaded and separated on 8% sodium dodecyl sulfate polyacrylamide gels and then electrophoretically transferred to Hybond-ECL nitrocellulose membranes (Amersham Pharmacia Biotech, Chalfont St. Giles, UK). Nonspecific binding was minimized by blocking in Tris-buffered saline (pH 8.0), 0.1% Tween 20, and 5% (wt/vol) nonfat milk before GR protein was detected using a polyclonal anti-GR antibody at 1:500 dilution (Affinity BioReagents, Inc., Golden, CO). Immunopositive bands were visualized by chemiluminescence detection (ECL; Amersham Pharmacia Biotech). Bound antibodies were stripped from membranes by incubation in 0.2 M glycine (pH 2.0), 0.1% sodium dodecyl sulfate, and 0.1% Tween 20 for 1 h at room temperature. Membranes were reblocked in 5% milk solution before tubulin protein (used as the loading control) was detected using a monoclonal anti-
-tubulin antibody (Sigma) at 1:5000 dilution, followed by chemiluminescence detection as above. Membranes were stripped again before MR protein detection using a polyclonal anti-MR antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at 1:400 dilution. MR immunopositivity was detected using the Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA). Images of blots were digitized, and ODs of GR, MR, and tubulin bands were determined using Aida 2.0 Auto Image Data Analyser software (Raytest Scientific Ltd., Sheffield, UK). Levels of GR and MR were expressed as a ratio of tubulin.
Corticosterone analyses
Plasma corticosterone levels were measured by RIA (35) modified for microtiter plate scintillation proximity assay. The intra- and interassay coefficients of variation were less than 10%.
Data analyses
Data are presented as mean ± SEM. Data from the in situ hybridization, Western blot, and 11ßHSD1 enzyme activity assay were analyzed with unpaired Student’s t tests; and data from the stress experiment were analyzed with Kruskal-Wallis nonparametric ANOVA followed by post hoc Mann-Whitney U tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, 2, and 3).
|
|
|
). The remainder of typical pyramidal cells in the cornu ammonis and granule cells of the dentate gyrus had relatively low expression of 11ßHSD1 mRNA. In obese rats, 11ßHSD1 mRNA was significantly decreased in the low-expressing cells in all hippocampal regions (Fig. 1B
). 11ßHSD1 mRNA levels in the high-expressing cells were not statistically different between lean and obese rats (Fig. 1C). 11ßHSD1 activity, measured in whole hippocampal homogenate, did not differ between the lean and obese groups (conversion of substrate to product, 34.4 ± 2.05% lean vs. 34.3 ± 1.75% obese; P = 0.95).
|
) and layer 2 of the frontal cortex (obese, 7.51 ± 0.43; lean, 8.91 ± 0.38; P < 0.05). However, in whole hippocampal homogenate, the protein concentration of MR did not significantly differ between the groups, being only marginally lower in the obese animals (Figs. 2 and 3).
Because of the possibility that the Western blotting procedure had poorer sensitivity than the in situ hybridization approach, the functional effect of lower MR in hippocampus was tested in vivo using the MR antagonist spironolactone. In lean animals, spironolactone had no effect on basal plasma corticosterone levels but markedly enhanced the response to stress (Fig. 5). In obese vehicle-treated animals, basal plasma corticosterone was not different, but the response to stress was enhanced, compared with lean animals. Spironolactone had a smaller effect on the increase of stress-induced hypercorticosteronemia in the obese than in the lean animals. As a result, in the presence of spironolactone, the difference between lean and obese groups was abolished.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Down-regulation of MR expression in negative feedback sites could explain some of the changes in the HPA axis previously described in obesity. In obese Zucker rats, plasma glucocorticoid levels are characteristically elevated during the diurnal nadir (i.e. in so-called basal conditions) rather than during the diurnal peak, so that animals lack the normal circadian rhythm of corticosterone (14, 37, 38, 39). Similar flattening of the diurnal rhythm has been observed in human obesity (40, 41). In the current study, we did not find any difference in the basal plasma corticosterone levels between the phenotypes. We did, however, confirm previous findings that the response to stress is enhanced in obese animals (5, 9, 14). It has been shown that, in addition to the role of GR, MR activation also contributes to suppressing the response to acute stress; administration of an MR antagonist elevates both the basal and the peak corticosterone levels (42, 43, 44) and gives rise to a delayed shut-off of the stress response (43). Similar effects have been observed after administration of the MR antagonists spironolactone or canrenoate in humans (45, 46, 47, 48). Indeed, the abolition, by spironolactone, of differences in stress responses in lean and obese animals strongly supports the inference that impaired MR function has an important effect on dysregulation of the HPA axis in obese Zucker rats. This effect is most obvious for the so-called reactive feedback, which regulates the stress response. However, the difference in MR does not seem to impact on basal corticosterone levels, dictated by so-called proactive negative feedback (17). Earlier reported increased basal concentrations of corticosterone in obese Zucker rats (5, 7, 8, 10, 11, 12, 13) may indicate that previous reports have, in fact, documented an increased response to the stress of the sampling procedure in obese animals.
The current results also illustrate some of the complexities of dissecting central control of the HPA axis, which is dependent on specific cells with restricted localization. The hippocampus is one of the major sites where MR influences HPA function. Although MR mRNA was decreased in pyramidal cells in hippocampal subregions, MR protein in whole hippocampal homogenates was not different, as judged by Western blotting. This probably reflects dilution of the sample with surrounding tissue during dissection, and/or the lack of sensitivity of Western blots to detect differences of small magnitude. Of course, it could also be that the decreased mRNA expression does not give rise to a similar reduction in MR protein. The in vivo experiment with spironolactone shows, crucially, that the incremental effect of MR antagonism is smaller in obesity, consistent with lower levels of receptor in key cells responsible for negative feedback. It is possible that, if the change in MR protein level is of small magnitude, it does not have any relevance until the animals are exposed to stress. Another explanation could be that the most important changes in MR occur outside the hippocampus. We also observed down-regulation of MR in frontal cortex in obese animals, a region implicated in HPA axis control (49). However, MR levels in extrahippocampal forebrain sites are low, especially so in the hypothalamus, so these changes may not be as important as those in hippocampus.
The mechanism of down-regulation of MR in obese animals has not been investigated here. One hypothesis for the etiology of obesity is that it results from the neuroendocrine response to chronic stress (50). MR is decreased in hippocampus by exposing rats to varying stressors for several days (51, 52, 53), long-term treatment with corticosterone (54), or exposure of tree shrews to psychosocial stress (55). However, these manipulations also reduce GR expression (51, 52, 53, 54, 55), which we find is not a feature in obese rats. Parallel changes in MR expression and HPA function have been described in human depression (52, 56). Interestingly, selective serotonin re-uptake inhibitors, which potently up-regulate MR (57, 58, 59, 60) and normalize the HPA feedback in depression (61), also diminish weight gain in adult ob/ob mice and obese Zucker rats (62, 63) and decrease leptin levels in obese Zucker rats (64). Moreover, treatment with selective serotonin re-uptake inhibitors in a group of obese men normalized the morning plasma cortisol levels (65). It will be intriguing to dissect the regulation of hippocampal MR expression by factors that are known to be disrupted in obesity and to influence the HPA.
The influence of 11ßHSD1 on central regulation of the HPA axis is less well characterized than that of MR. Pharmacological experiments with 11ßHSD1 inhibitors, such as liquorice or carbenoxolone, are difficult to interpret because these agents have potent effects on peripheral glucocorticoid clearance rate; hence, changes in the HPA axis may occur without any alteration in feedback control, simply to maintain normal plasma glucocorticoid levels. 11ßHSD1 knockout mice also have deranged peripheral corticosterone clearance, but a recent study has shown that they have several HPA axis abnormalities that mimic, in part, those in obesity, including higher basal (but normal peak) plasma corticosterone levels, and enhanced corticosterone responses to restraint stress, in part attributed to impaired negative feedback to endogenous glucocorticoids (21). In the current study, the down-regulation of 11ßHSD1 mRNA was restricted to the major hippocampal neuronal layers (pyramidal and granule cells) but not found in a minority of high-expressing cells. The latter are located in regions of sparse interneurons and glia but remain unidentified. Probably as a result of dilution by the higher-expressing cells, we were not able to confirm differences in protein levels in homogenates of the whole hippocampus. Systemic pharmacological manipulation of 11ßHSD1 is unlikely to resolve the importance of the hippocampal enzyme because of confounding effects on peripheral corticosterone clearance. Future experiments with intracerebroventricular administration of 11ßHSD1 inhibitors in Zucker rats may establish the importance of local regeneration of glucocorticoids to negative feedback control of the HPA axis.
The hippocampus is not the only site where 11ßHSD1 is altered in obese Zucker rats. Enzyme activity and mRNA are reduced in liver but increased in adipose tissue, especially in omental adipose (7). A similar magnitude of increased adipose 11ßHSD1 in mice, induced by transgenic overexpression of the rat gene under the adipose specific AP2 promoter/enhancer, results in marked central obesity and many features of the metabolic syndrome (66). In contrast, 11ßHSD1 knockout protects the liver from adverse effects of glucocorticoids on carbohydrate and lipid metabolism (67, 68). Similar down-regulation of liver 11ßHSD1 and up-regulation of adipose 11ßHSD1 has been observed in human obesity (29, 30, 31, 32). These observations have stimulated interest in 11ßHSD1 as a mediator and therapeutic target for peripheral metabolic complications of obesity. The mechanism(s) for tissue-specific alterations of 11ßHSD1 in obesity remain speculative (7, 69). If reduced CNS 11ßHSD1 is proven to be an important determinant of enhanced HPA axis activity in obesity, then it may be advantageous for 11ßHSD1 inhibitors to be excluded from CNS if they are to have peripheral metabolic benefits without inducing compensatory activation of the HPA axis.
In summary, these results support an emerging concept that disruption to feedback regulation of the HPA axis in obesity is explained by mediators other than GR. Both MR and 11ßHSD1 mediate feedback signals that respond to endogenous glucocorticoids (corticosterone in rat and cortisol in man, both of which bind to MR as well as GR and are substrates for 11ßHSDs) but less well to exogenous synthetic glucocorticoids [e.g. dexamethasone, which is a GR (not MR) agonist and is less readily metabolized by 11ßHSDs]. This suggests that in obesity and in parallel conditions, including depression, more discriminant tests of feedback control of the HPA axis are required to dissect these abnormalities in patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: CNS, Central nervous system; GR, glucocorticoid receptors; HPA, hypothalamic-pituitary-adrenal; HSD, hydroxysteroid dehydrogenase; MR, mineralocorticoid receptors.
Received October 1, 2002.
Accepted for publication February 14, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. Mattsson, R. M. Reynolds, K. Simonyte, T. Olsson, and B. R. Walker Combined Receptor Antagonist Stimulation of the Hypothalamic-Pituitary-Adrenal Axis Test Identifies Impaired Negative Feedback Sensitivity to Cortisol in Obese Men J. Clin. Endocrinol. Metab., April 1, 2009; 94(4): 1347 - 1352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. E. Bates, A. S. Sirek, M. A. Kiraly, J. T. Y. Yue, D. Goche Montes, S. G. Matthews, and M. Vranic Adaptation to Mild, Intermittent Stress Delays Development of Hyperglycemia in the Zucker Diabetic Fatty Rat Independent of Food Intake: Role of Habituation of the Hypothalamic-Pituitary-Adrenal Axis Endocrinology, June 1, 2008; 149(6): 2990 - 3001. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Garcia-Bueno, J. L. M. Madrigal, B. G. Perez-Nievas, and J. C. Leza Stress Mediators Regulate Brain Prostaglandin Synthesis and Peroxisome Proliferator-Activated Receptor-{gamma} Activation after Stress in Rats Endocrinology, April 1, 2008; 149(4): 1969 - 1978. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Buren, S.-A. Bergstrom, E. Loh, I. Soderstrom, T. Olsson, and C. Mattsson Hippocampal 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Messenger Ribonucleic Acid Expression Has a Diurnal Variability that Is Lost in the Obese Zucker Rat Endocrinology, June 1, 2007; 148(6): 2716 - 2722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. W. W. Pace and R. L. Spencer Disruption of mineralocorticoid receptor function increases corticosterone responding to a mild, but not moderate, psychological stressor Am J Physiol Endocrinol Metab, June 1, 2005; 288(6): E1082 - E1088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Drake, D. E. W. Livingstone, R. Andrew, J. R. Seckl, N. M. Morton, and B. R. Walker Reduced Adipose Glucocorticoid Reactivation and Increased Hepatic Glucocorticoid Clearance as an Early Adaptation to High-Fat Feeding in Wistar Rats Endocrinology, February 1, 2005; 146(2): 913 - 919. [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. R. Seckl, N. M. Morton, K. E. Chapman, and B. R. Walker Glucocorticoids and 11beta-Hydroxysteroid Dehydrogenase in Adipose Tissue Recent Prog. Horm. Res., January 1, 2004; 59(1): 359 - 393. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
116 kDa), corresponding to the nondegradable MR protein, was quantified.
