| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Psychology (I.G., R.Y.), The Hebrew University of Jerusalem, Jerusalem 91905, Israel; Department of Neurochemistry and Neurotoxicology (K.I.), Stockholm University, SE-106 91 Stockholm, Sweden; and Department of Neurology (J.W.), the Agnes Ginges Center for Human Neurogenetics, Hadassah University Hospital, Jerusalem 91120, Israel
Address all correspondence and requests for reprints to: Raz Yirmiya, Ph.D., Department of Psychology, The Hebrew University of Jerusalem, Mount Scopus, Jerusalem 91905, Israel. E-mail: razyirmiya{at}huji.ac.il.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Other factors that were implicated in the activation of the HPA axis are proinflammatory cytokines, particularly IL-1 (6, 7), which is produced by many types of cells, including immune cells in the periphery as well as glia and neurons within the brain (8). IL-1 is expressed throughout the components of the HPA axis (6, 7) and influences all of its levels: IL-1 induces the secretion of CRH from the hypothalamus (9, 10) and ACTH from the pituitary (11), via the IL-1 type I receptors in these regions (6, 7). Additionally, IL-1 stimulates GC secretion from the adrenal, although no known IL-1 receptors were found in this gland (6, 7). IL-1 may be also involved in the GC negative feedback mechanism by altering the levels of brain CG receptors in the hippocampus (12, 13).
In addition to its role in immunoregulation of inflammatory processes, IL-1 plays a major role in the modulation of neuroendocrine systems during illness (14). Immune challenges, such as local inflammation, viral infection, and endotoxin exposure, induce the production and secretion of brain IL-1, followed by HPA axis activation (6, 7). It was shown, by us and by others, that IL-1 blockade using IL-1 receptor antagonist (IL-1ra) during these conditions prevents the increase in GC (7, 15). Endogenous IL-1 activates the HPA axis not only during sickness, but also in various stressful situations. IL-1 gene expression, protein levels, and biological activity are increased after exposure to several types of stress, such as restraint (16, 17) and inescapable shock (18). Furthermore, impairments in IL-1 signaling, caused by either IL-1ra administration or genetically induced IL-1 deficiency, diminish the increase in GC secretion following stress (16, 19).
The interaction between IL-1 and the HPA axis is bidirectional. On the one hand, IL-1 activates the HPA axis, as detailed above; and on the other hand, GC suppress the production of IL-1, mainly by decreasing IL-1 mRNA levels, both by inhibiting its transcription and by destabilizing it (20). Accordingly, it has been shown that removal of endogenous GC augments IL-1 secretion, e.g. the increase in IL-1 protein levels, after stress, is more distinct in adrenalectomized rats (18, 21). Recently, we demonstrated an increase in IL-1 gene expression in rats after ADX. Furthermore, we found that IL-1 gene expression in astrocyte cultures can be detected only when cortisol is removed from the culture medium (22). Therefore, we hypothesized that, in addition to its putative role in the stress response, brain IL-1 may be involved in ADX-induced hypersecretion of ACTH.
In the present study, we further assessed the role of IL-1 in the regulation of adrenocortical activation after exposure to various stressors, using IL-1 receptor type I-deficient mice (IL-1rKO). Furthermore, we tested, for the first time, the role of IL-1 in HPA axis regulation after ADX, using IL-1rKO mice, as well as mice with transgenic overexpression of IL-1 receptor antagonist within the central nervous system (CNS) (IL-1raTG). To assess whether the results found using the genetically engineered mice can be ascribed to developmental influences of the IL-1 deficiency, we also tested ADX-induced ACTH hypersecretion in mice that were prenatally treated with IL-1ra.
| Materials and Methods |
|---|
|
|
|---|
Stressful stimuli
Twenty-four hours before the experiment, mice were divided into separate cages. At the experiment day, control nonstressed mice were taken from their new home-cages and killed by decapitation immediately, alternately with stressed mice. Mice subjected to acoustic stress were exposed to a ringing bell (109 decibels) for 4 min and were killed by decapitation 10 min after stress initiation. Mice subjected to restraint stress were placed in a perforated cylinder (3 cm in diameter; 10 cm in length) for 60 min and were killed immediately upon removal from the cylinder. To induce metabolic stress (cytoglucopenia), mice were ip injected with either 250 or 500 mg/kg 2-deoxyglucose (2DG, dissolved in 0.2 ml; Sigma, Rehovot, Israel) or saline (same volume) and were killed by decapitation 1 h after the injection. Administration of 2DG, which inhibits intracellular glucose use, is commonly used as a model of metabolic stress, influencing the HPA axis via central mechanisms (25).
ADX
Mice were anesthetized with pentobarbital (60 mg/kg, ip), and the adrenal glands were bilaterally removed. Sham operations were similar in all aspects apart from the removal of the glands. After operation, mice were provided with 0.45% NaCl as their drinking solution instead of water. Eight days after operation, mice were killed by rapid decapitation, and trunk blood was collected. Blood was centrifuged for 20 min at 2000 rmp, and serum was collected for ACTH analysis. The complete removal of the adrenal glands was confirmed by measuring corticosterone (CS) levels, which were found to be below the detection limit of the kit in all the ADX-operated mice.
Determination of ACTH and CS levels
CS was determined by RIA, as previously described (26). The sensitivity limit of the assay was 0.5 µg/100 ml, and the intra- and interassay coefficients of variation were 6.4% and 7.2%, respectively. The percent of cross-reactivity of the antiserum with various steroids was: cortisol, 4.5; estradiol, less than 0.1; progesterone, 15.7; and testosterone, 7.9.
ACTH concentrations were determined by RIA kit (Nichols Institute Diagnostics, San Juan Capistrano, CA). The sensitivity limit of the assay was 10 pg/ml. There was no cross-reactivity of the antiserum with
-MSH, ß-MSH, ß-lipotrophic hormone, or ß-endorphin.
Prenatal IL-1ra exposure
Pregnancy was determined by the presence of vaginal plugs, and dams were sc implanted, on d 7 of gestation, with Alzet osmotic micropumps secreting human IL-1ra (20 mg/kg·d for 14 d; Amgen, Thousand Oaks, CA). Using ELISA, we found that exogenous human IL-1ra crossed both the placenta and the blood brain barrier and reached the fetal CNS (34.22 ± 7.8 and 10.74 ± 3.08 pg/mg brain tissue in IL-1ra and vehicle-treated fetuses, respectively) on d 15 of gestation. Two control groups were used in this experiment: The first group was implanted with vehicle-secreting micropumps, and the second group received no treatment. Within 24 h of birth, pups from all groups were transferred to untreated nursing dams. To minimize the influence of inherent familial effects, only one pup from each litter was assigned for each experimental group. ACTH levels after ADX were tested, as described above, at the age of 4 months.
Statistical analysis
The results were analyzed by a two-way ANOVA, followed by the Tukey post hoc analysis.
| Results |
|---|
|
|
|---|
|
Effects of impaired IL-1 signaling on ADX-induced ACTH hypersecretion
Eight days after ADX, the levels of ACTH were markedly increased in C57BL/6x129/Sv WT mice (P < 0.05), whereas the levels of ACTH in IL-1rKO mice remained unchanged (Fig. 2A
). This finding was reflected by a significant strain by treatment interaction (F1,22=4.78, P < 0.05). Similarly, 8 d after ADX ACTH, levels were significantly increased in C57BL/6xCBA WT mice (P < 0.01), whereas the levels of ACTH in IL-1raTG mice remained unchanged (Fig. 2B
). This finding was reflected by a significant strain by treatment interaction (F1,25 = 12.4, P < 0.005).
|
|
| Discussion |
|---|
|
|
|---|
and IL-1ß (23). IL-1raTG mice have astrocyte-directed overexpression of the human IL-1ra gene under the control of the murine glial fibrillary acidic protein promoter and are insensitive to the administration of exogenous IL-1 (24). The deficits in ADX-induced HPA axis activation that accompany the impairments in IL-1 signaling cannot be attributed to any peripheral influences, because IL-1raTG mice overexpress IL-1ra only within the CNS (24). Our results regarding the role of IL-1 in stress response are in accordance with previous studies showing that IL-1 levels are increased after exposure to various stressors, such as restraint (16, 17) and inescapable shock (18), and that IL-1 receptor type I gene expression in the pituitary is also enhanced after stress or CRH injection (27, 28). IL-1 by itself can activate the HPA axis, and either pharmacological or genetic disruption of IL-1 signaling, using IL-1ra administration or IL-1 knockout mice, diminishes stress-induced increase in CS secretion (6, 7, 19). In the present study, a diminished increase in CS levels was found in IL-1rKO mice only in response to relatively mild stressors, such as bell ringing and cytoglucopenia induced by 250 mg/kg 2DG injection, suggesting that IL-1 signaling via its type I receptor is important for adrenocortical activation after moderate stress. However, in response to more severe stressors, such as 60 min of restraint or 500 mg/kg 2DG injection, no difference was found between IL-1rKO and WT controls. This finding is consistent with the findings of a previous study reporting that, 2 h after turpentine injection, IL-1-deficient mice demonstrate normal HPA axis activation; whereas at 8 h after injection, their response was abolished (19). In contrast with our findings, treatment with IL-1ra before (but not during) exposure to restraint stress significantly inhibited ACTH secretion in rats (16). Differences between species and methodologies may account for this inconsistency. For example, in the pharmacological study (16), young rats were subjected to stress in relatively large restrainers; whereas in the present study, adult mice were entered into much tighter restraining tubes. Thus, it is possible that the stress induced in the pharmacological study was milder than the one induced in the present study and, therefore, its effects were attenuated by blockade of IL-1 receptors.
It is possible that the response to robust stressors may also involve other mediators, which may compensate for the absence of IL-1 signaling. One of these mediators may be the cytokine TNF-
, which often works in synergism with IL-1 (8, 29). Indeed, in a preliminary study, we found that TNF p55 receptor-deficient mice demonstrated normal CS secretion after exposure to auditory stress but decreased CS secretion after 60 min of restraint stress. It should be noted that the finding that mild stress-induced CS secretion in IL-1rKO mice is blunted during a single time point (which coincides with maximal activation in normal animals) (16, 25) cannot exclude the possibility that the strain differences are attributable to changes in the kinetics, rather than the amplitude, of CS secretion from the adrenal. Thus, the possibility that the CS response to the auditory stress and the low dose of 2DG are slower to develop in the IL-1rKO mice should be further explored.
Feedback regulation of the HPA axis is considered to involve inhibitory actions of GC via GC receptors. Several studies (e.g. Ref. 3) suggest that the primary target for GC feedback inhibition is not only the hypothalamic CRH neurons but also other hypothalamic and extrahypothalamic (e.g. hippocampus, pituitary) systems that regulate CRH expression and release (3, 4). The action of this system is exemplified by the marked increase in ACTH levels after ADX (5). In addition to ACTH hypersecretion, ADX is also associated with enhanced expression of IL-1 gene in the hypothalamus, pituitary, and brain stem, as well as increased production of IL-1 protein in the pituitary (22, 30). These findings suggest that GC may have a tonic inhibitory effect on brain IL-1. In the present study, after removal of this inhibitory effect by ADX, ACTH levels were markedly elevated in WT controls but did not increase in IL-1rKO or IL-1raTG mice, suggesting that brain IL-1 may have an important role in ADX-induced ACTH hypersecretion. Together, these findings suggest that the removal of the direct inhibitory signal exerted by GC is not sufficient for ADX-associated ACTH hypersecretion. Rather, this hypersecretion is critically dependent on an excitatory signal, provided by the elevated levels of IL-1, which is a potent HPA axis activator (6, 7).
The exact mechanism by which endogenous IL-1 is involved in the regulation of the HPA axis is not clear. Among the possible mediators of its actions are NE and 5-HT, which are known to play an important role in regulating HPA axis responses to various stressful stimuli (1, 31, 32, 33). Furthermore, it has been shown that IL-1 has a facilitatory role on noradrenergic and serotonergic systems; for example, either peripheral or intrahypothalamic IL-1 administration stimulates NE secretion in the hypothalamus in vivo (16, 34, 35), and intrahypothalamic or intrahippocampal IL-1 administration induces the secretion of 5-HT in these regions (35, 36). IL-1ra can block the increase in NE and 5-HT after stress, as well as the increase in ACTH plasma levels (16). Additionally, neurotoxic lesions that produce depletion of hypothalamic NE inhibit the increase in plasma ACTH and CS levels in response to IL-1 (37, 38). NE may be also involved in the effects of IL-1 on ACTH hypersecretion after ADX, because it was shown that hypothalamic NE depletion by 6-OHDA injection does not affect basal ACTH levels but reduces the increase in ACTH hypersecretion after ADX (39). Together, these findings suggest that the influence of IL-1 signaling on HPA axis regulation after stress or ADX may be, at least partly, mediated by NE and/or 5-HT systems.
The impairment in ADX-induced ACTH hypersecretion in IL-1rKO and IL-1raTG mice may result from the lack of IL-1 signaling, either acutely or during brain development. To assess the latter possibility, we tested the effect of IL-1 blockade during the last 2 wk of gestation, mimicking the effect of excess IL-1ra during the same developmental period in IL-1raTG mice. IL-1 influences brain development both directly by inducing neuronal differentiation (e.g. Ref. 40), as well as indirectly by the enhancement of trophic factors, e.g. NGF and CNTF (41, 42). Similar to the results found in the two IL-1 signaling-deficient mice strains, IL-1ra prenatally treated mice exhibited no ACTH hypersecretion after ADX. This result suggests that during fetal development, IL-1 may have a role in the maturition of brain systems that regulate the HPA axis. It should be noted that this developmental effect does not rule out the possibility that IL-1 is also involved in activation of the HPA axis after various stimuli, as suggested by previous studies using IL-1ra in adult animals (7, 15, 16). Further research will be needed to fully understand the developmental and acute influences of IL-1, or their combination, on the HPA axis.
In conclusion, the results of the present study demonstrate that, in addition to its role in mediating illness-associated neuroendocrine response, IL-1 has an important role in the activation of CS secretion after exposure to different modalities of mild, but not severe, stress. Furthermore, our study demonstrates a novel modulatory role for IL-1 in ADX-induced ACTH hypersecretion. Thus, alterations in IL-1 levels during noninflammatory conditions play a role in HPA axis activation and its regulation by GC-associated feedback control mechanisms.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: ADX, Adrenalectomy; CNS, central nervous system; CS, corticosterone; 2DG, 2-deoxyglucose; GC, glucocorticoids; HPA, hypothalamo-pituitary-adrenal; 5-HT, serotonine; IL-1rKO, knockout of the IL-1 receptor type I; IL-1raTG, transgenic overexpression of IL-1 receptor antagonist; NE, norepinephrine; PVN, paraventricular nucleus; WT, wild-type.
Received March 18, 2003.
Accepted for publication June 30, 2003.
| References |
|---|
|
|
|---|
, IL-1ß, IL-1
/ß, and IL-1 receptor antagonist shows that IL-1ß is crucial in turpentine-induced fever development and glucocorticoid secretion. J Exp Med 187:14631475
synergistically stimulate nerve growth factor (NGF) release from cultured rat astrodytes. Neurosci Lett 117:335340[CrossRef][Medline]
This article has been cited by other articles:
![]() |
Y. Shavit, G. Fish, G. Wolf, E. Mayburd, Y. Meerson, R. Yirmiya, and B. Beilin The Effects of Perioperative Pain Management Techniques on Food Consumption and Body Weight After Laparotomy in Rats Anesth. Analg., October 1, 2005; 101(4): 1112 - 1116. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Bajayo, I. Goshen, S. Feldman, V. Csernus, K. Iverfeldt, E. Shohami, R. Yirmiya, and I. Bab Central IL-1 receptor signaling regulates bone growth and mass PNAS, September 6, 2005; 102(36): 12956 - 12961. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Karpac, D. Ostwald, S. Bui, P. Hunnewell, M. Shankar, and U. Hochgeschwender Development, Maintenance, and Function of the Adrenal Gland in Early Postnatal Proopiomelanocortin-Null Mutant Mice Endocrinology, June 1, 2005; 146(6): 2555 - 2562. [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 |