| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Neurobiology and Physiology, College of Life Sciences and Institute of Neuroscience, School of Medicine, Zhejiang University, Hangzhou 310058, China
Address all correspondence and requests for reprints to: Dr. Xue-Qun Chen or Professor Ji-Zeng Du, Division of Neurobiology and Physiology and Institute of Neuroscience, School of Medicine, Zhejiang University, Hangzhou 310058, China. E-mail: chewyg{at}zjuem.zju.edu.cn or dujz{at}cls.zju.edu.cn.
| Abstract |
|---|
|
|
|---|
10.8% O2) in a hypobaric chamber for 1, 2, 5, 10, and 15 d. Potential involvement of CRFR1 was assessed through five daily sc injections of a CRFR1 antagonist (CP-154,526) prior to hypoxia. Results showed that the levels of body weight, food intake, blood glucose, and plasma insulin were significantly reduced; the content of hepatic glycogen initially and transiently declined, whereas the early plasma glucagon and leptin remarkably increased; plasma corticosterone was markedly increased throughout the hypoxic exposure of 115 d. Compared with hypoxia alone, CRFR1 antagonist pretreatment in the hypoxic groups prevented the rise in corticosterone, whereas the levels of body weight and food intake were unchanged. At the same time, the reduction in blood glucose was greater and the pancreatic glucose was increased, plasma insulin reverted toward control, and plasma glucagon decreased. In summary, prolonged hypoxia reduced body weight, food intake, blood glucose, and plasma insulin but transiently enhanced plasma glucagon and leptin. In conclusion, CRFR1 is potentially involved in the plasma insulin reduction and transient glucagon increase in hypoxic rats. | Introduction |
|---|
|
|
|---|
We previously found that hypoxia stimulates corticotropin-releasing factor (CRF) secretion and its mRNA expression in the hypothalamic paraventricular nucleus (PVN) and enhances plasma ACTH and corticosterone (9). Recently, we demonstrated in the rat that these actions of hypoxia were associated with activation of the CRF receptor type 1 (CRFR1) and CRFR1 mRNA in the PVN (10) and in the pituitary (11). We also found that in lowland mice, hypoxia significantly increases hypoxia-inducible factor-1
and lactate dehydrogenase mRNA expression, whereas it reduces isocitric acid dehydrogenase mRNA expression in the rat liver, indicating that hypoxia interferes with both anaerobic and aerobic metabolism. These changes were different from those found in native high-altitude mammals. These findings raise the question of whether lowland animals exposed to continuous hypoxia in time become more like native high-altitude animals with respect to the regulation of metabolism. In this study, we first confirmed the effects of continuous hypoxia on body weight, food intake, circulating blood glucose, insulin, glucagon, corticosterone, leptin, and glycogen in the livers of lowland rats. We then examined the role of CRFR1 in the changes in those hormones, in particular in circulating insulin and glucagon during 5 d of continuous hypoxia.
| Materials and Methods |
|---|
|
|
|---|
Hypoxic stress
Altitude hypoxia of 5 km (54.02 kPa, equivalent to
10.8% O2 at sea level) was simulated in a ventilated and controllable hypobaric chamber (1 m3 in volume). Experimental animals were placed in this chamber for either 1, 2, 5, 10, or 15 d with the same light/dark cycle, nutrition, and temperature as described above. After hypoxic exposure, all rats were killed immediately by decapitation at 11001130 h to minimize circadian rhythm effects. Control rats were placed in the same chamber set at sea level (100.08 kPa, equivalent to
20.9% O2) for the same times as the hypoxic groups.
Further groups of rats were housed in the chambers and exposed to hypoxia (slight hypoxia) of 2 km altitude (79.97 kPa, equivalent to
16.0% O2 at sea level) for similar periods for comparison with animals in 5 km altitude hypoxia.
CRFR1 antagonist treatments
CP-154,526 (butyl-[2, 5-dimethyl-7-(2, 4, 6-trimethylphenyl)-7H-pyrrolo [2, 3-d] pyrimidin-4-yl]-ethylamine), a selective antagonist of CRFR1, was kindly donated by Pfizer Inc. (Groton, CT). The CP-154,526 was suspended in 0.5% methylcellulose (Sigma, St. Louis, MO). The rats received five daily sc injections with the antagonist (30 mg/kg·d, 10 mg/ml) or vehicle 11001130 h in the morning, and were then placed in the hypobaric chamber, and exposed to hypoxia (at 5 km altitude) until next injection. A total of 24 male rats were randomized into four groups: 1) sea level control + vehicle group (the chamber set at sea level and rats were given injections with the same volume of vehicle as antagonist group); 2) hypoxic exposure + vehicle group (5 d of hypoxia at 5 km altitude with an injection of the same volume of vehicle as the antagonist-treated group); 3) antagonist-treated control group (injections of CP-154,526, 30 mg/kg·d, 10 mg/ml only and the chamber set at sea level); and 4) antagonist + hypoxia group (5 d of hypoxic exposures at 5 km altitude with sc injections of CP-154,526.
Measurement of food intake and body weight
Food intake was measured by weighing the reduction of food mass per rat after consumption daily (12) and body weight (BW) was determined by directly weighing each rat each day at 1100 h.
Plasma, liver, and pancreatic tissue preparation
At the end of all exposures, rats were killed by decapitation between 1100 and 1130 h, and trunk blood was collected into 15-ml tubes containing sodium heparin. The tube was placed in ice water till separation of plasma by centrifugation at 4 C. Plasma was stored at 80 C before analysis (13). The liver and pancreas were immediately removed, frozen in liquid nitrogen, and stored at 80 C until analysis.
Pancreatic homogenate preparation and protein concentration assay
The pancreatic tissue was divided into two parts, one for measurement of glucagon, and the other for insulin. For glucagon assay, the pancreatic tissue was quickly weighed, placed in an Eppendorf tube (Eppendorf AG, Hamburg, Germany) with saline (covered) and boiled for 3 min (to degrade enzymes), and then homogenized (electronic homogenizer) in 0.25 ml 1.0 mol/liter frozen acetic acid (HAc) and a cocktail of protease inhibitors (aprotinin, leupeptin) for 1 min at 4 C according to the manufacturers instructions (14). For insulin assay, the pancreas sample from the 80 C freezer was quickly weighed, removed to 0.25 ml 1.0 mol/liter HAc with aprotinin, and homogenized (electronic homogenizer) for 1 min at 4 C. After keeping at 4 C for 2 h, the homogenate was neutralized with 0.25 ml 1.0 mol/liter sodium hydroxide (NaOH) and centrifuged at 3000 rpm for 30 min at 4 C, and then the supernatant (100 µl) was collected for protein assay (Bio-Rad, Hercules, CA) and insulin RIA analysis (13).
Measurement of glycogen and glucose
The livers, after being washed twice with saline, were incubated with 1 ml of 30% KOH for 30 min at room temperature, and then heated for 30 min at 100 C, mixed with 2 ml ethanol, kept at 4 C overnight and precipitated by centrifugation at 2000 x g for 10 min at 4 C. The pellet was dissolved in distilled H2O and the hepatic glycogen content was determined by the anthrone-sulfuric acid method (15). The blood and pancreatic glucose were assayed by an enzymatic colorimetric method (16) using commercial kits (Institute of Biological Production, Shanghai, China).
Insulin, glucagon, leptin, and corticosterone assay
The levels of plasma and pancreatic insulin were estimated with a rat RIA kit (Institute of Biological Production, Shanghai, China) (17). The sensitivity of the assay was about 2 µIU/ml and the inter- and intraassay coefficients of variation were 9.5 and 6.5%, respectively. The levels of plasma and pancreatic glucagon were determined using a rat RIA kit (Institute of Biological Production) (18), and its sensitivity was about 10 pg/ml. The levels of plasma leptin were determined using a rat RIA kit (Institute of Biological Production) (13), and its sensitivity was about 0.1 ng/ml. Plasma levels of corticosterone were estimated with a rat RIA kit (China Institute of Atomic Energy, Beijing, China); its sensitivity was about 0.40 ng/ml, and the inter- and intraassay coefficients of variation were 6.5 and 4.5%, respectively (11).
Statistical analysis
The data were presented as mean ± SD. Students t tests, one- and two-way ANOVA (factor 1 = control, hypoxia; factor 2 = control, CP-154,526) were performed as appropriate. Statistical significance was fixed at P < 0.05.
| Results |
|---|
|
|
|---|
|
|
|
|
3) Compared with the group with continuous 5 km hypoxia alone, the hypoxia + CP-154,526 treatment for 5 d reduced blood glucose (10%, Fig. 3C
) but reversed the decreases in pancreatic glucose (up 90%, Fig. 3D
) and plasma insulin (Fig. 3E
). It did not do so for pancreatic insulin (Fig. 3F
), CP-154,526 markedly decreased the levels of plasma glucagon (24%, Fig. 3G
) but not those of pancreatic glucagon (Fig. 3H
), and significantly reversed the increase in plasma corticosterone (Fig. 3I
).
4) Plasma leptin levels were significantly decreased by continuous hypoxia (5 km for 5 d), compared with the sea level control + vehicle. Treatment with CP-154,526 alone dramatically enhanced plasma leptin levels. Continuous hypoxia (5 km) + the CP-154,526 reversed the hypoxia-induced decrease in plasma leptin (Fig. 3J
).
The above data indicate that CRFR1 mediates the inhibition of insulin release and the increase of glucagon and corticosterone release.
| Discussion |
|---|
|
|
|---|
The hypothalamo-pituitary-adrenal (HPA) axis is well known to be involved in the modulation of feeding. BW is normally regulated by food intake, metabolic consumption, and hormonal regulation. Dallman and colleagues (19) indicate that feeding is modified by glucocorticoids and possibly by insulin. Activity in the HPA axis is normal related to food intake in rodents and humans. HPA activity tracks feeding, metabolism, and energy disposition under normal conditions. In normal male rodents, adrenalectomy reduces daily food intake (by 1020%), fat stores, and the rate of ponderal weight gain (20, 21, 22) in a glucocorticoid-reversible manner (22, 23). Physiologically appropriate concentrations of naturally secreted glucocorticoids (corticosterone in rodents, cortisol in humans) have major stimulatory effects on caloric intake and, in the presence of insulin. Under acute stress, the rapidly increased HPA activity interacts with elevated epinephrine, glucagon, and sympathetic neural activity to elevate blood glucose concentration, ensuring adequate substrates for brain and muscle that may be necessary for survival. This increase in glucocorticoids inhibits further activity in the HPA axis through a negative feedback mechanism (23). Under chronic stress, however, food intake is usually decreased in rodents (24). In the presence of low energy stores and low insulin, daily mean glucocorticoid concentrations are invariably elevated in rodents and humans. Starvation induces a rapid and persistent increase of ACTH and, through corticosterone secretion, reduces the negative feedback efficiency of corticosterone on the HPA axis, and reduces insulin and leptin secretion (24). Besides, in a model of anorexia nervosa where rats are allowed restricted food with free access to running wheels, energy stores and insulin are reduced and HPA axis activity is increased (25). In contrast under acute and chronic hypoxic stress, HPA activity is enhanced over sea-level controls, such that CRF protein/CRF mRNA in the rat PVN, plasma ACTH, and corticosterone are significantly increased in a "sinusoidal pattern" within 15 d of hypoxia (9, 26, 27), but body weight is markedly decreased, in particular in the first 12 d of hypoxia (28). In this study of male rats, body weight, food intake, blood glucose and plasma insulin all decreased, but the activity of the HPA axis was elevated, and these effects on body weight and food intake were similar to those discussed above. These new observations, including blood glucose, plasma insulin and glucagon, provide new evidence to support the close relationships between activity in the HPA axis and food intake. HPA axis activation and feeding reduction (Fig. 1B
) may reduce the body weight gain through decreased pituitary GH release and GH mRNA expression (28). This reduction in GH and GH mRNA expression may be due to an activation of the CRFR1 in the PVN by either sustained or intermittent hypoxia at 5 km altitude (29), which further mediates somatostatin (SS) secretion because the hypoxia stimulates SS secretion and mRNA expression in the PVN and periventricular nuclei of the rat hypothalamus (30). Here, we emphasize that the hypoxia-induced reduction in body weight gain and food intake were dose dependent, such that the reduction under hypoxia of 5 km was much stronger than that of 2 km (Fig. 1B
and Table 1
). Several studies in rats show that hypoxia acutely influences body weight, feeding, activity of metabolic enzymes (31, 32, 33), and carbohydrate supplement effects on feeding behavior (34). Anorexia behavior is directly responsible for reduced food intake (32, 33). We also found that an injection of CoCl2 (mimicking hypoxia) and normobaric hypoxia in mice cause an increase in anaerobic glycolysis (lactate dehydrogenase mRNA) and a decrease in activity of the aerobic Krebs cycle (isocitric acid dehydrogenase mRNA), which may lead to an increase in consumption of both energy and body mass, which may contribute to the loss of body weight (35), Taken together, hypoxia-induced reduction in body weight may result partly from hypoxia-inhibited GH release/GH mRNA expression (through CRFR1 and SS) hypoxia-reduced feeding, and hypoxia-enhanced energy mass consumption.
It is well known that leptin is associated with inhibition of food intake (36). In the present study, continuous hypoxia of 5 km markedly increased plasma leptin levels on d 1 and 2 (Fig. 2E
). This may contribute early in hypoxic exposure to the dramatic reduction in food intake (Fig. 1B
). With adaptation to hypoxia, the leptin levels declined on d 5 were restored after 10 d, when the food intake was also maintained at a low rate. Recently, Huang et al. (37) reported that leptin treatment significantly decreased food intake, body weight, glucose and insulin plasma content, and blunted the treadmill running-induced elevation in plasma levels of corticosterone. This supports our findings at hypoxic d 1 and 2 (early phase of hypoxia) that is hypoxia-induced increase in plasma leptin would contribute to the reduction in food intake and BW. In contrast, leptin treatment strongly increased the expression of the CRFR2 in the ventromedial hypothalamic nucleus, suggesting that subchronic elevation of central levels of leptin blunts stress-induced activation of the HPA axis through the inhibition of activation of the CRFergic PVN neurons, and potentially enhances the anorectic CRF effects via the stimulation of expression of CRFR2 in the ventromedial hypothalamic nucleus. This is much similar to hypoxia-induced effects.
Urocortin III, a newly discovered member of the CRF family, is reported to bind specifically with CRFR2, a receptor that plays a potential role in the inhibition of food intake (38). Under continuous or intermittent hypoxia, the CRF and the CRF mRNA expression in the rat PVN markedly increase for 5 d through hypoxia-activated CRFR1 (10). Both CRFR1 and CRFR2 mRNA expression in the rat pituitary are significantly enhanced (11), and Urocortin III mRNA levels are enhanced in the amygdala and the arcuate nucleus (Chen, X.-Q., and C.-Y. Niu, unpublished data). In this study, blood glucose, plasma insulin, and plasma glucagon levels were linked to CRFR1 activity (Fig. 3
). Therefore, both CRFR1 and CRFR2 may contribute to the lowered food intake over 5 d of 5 km continuous hypoxia in rats. These data suggest that not only glucocorticoids but CRF family members in the brain, in particular in the hypothalamus, are correlated with feeding and glucose metabolism.
Effects of hypoxia on hepatic glycogen and blood glucose, plasma insulin and glucagon
In the present study, an early reduction in hepatic glycogen (at 12 d) and a delayed reduction in blood glucose (at 5, 10, and 15 d) were induced by the hypoxia of 5 km (Fig. 2
, A and B) but not by the hypoxia of 2 km (Table 1
), clearly showing dose-dependent hypoxic effects. The changes in blood glucose are not too dissimilar to those reported in several studies which have used various periods of hypoxia exposure (36, 39, 40, 41). Thus, hypoxia in lowland rats is partly compensated by changes in glucose metabolism. These effects contrast with the limited effects of hypoxia on hepatic glycogen that we found in a previous study in a high altitude native mammal, the Tibetan pika (Ochotonia curzoniae) (42). Possible mechanisms involved in lowland rats response to hypoxia were partly revealed by the present study, and confirm some of the work by others (43). When animals are acutely subjected to hypoxia, food intake declines and plasma glucagon is enhanced in the early phase (an acute adjustment phase) to transform hepatic glycogen to provide sufficient blood glucose and energy (through an anaerobic glycolysis) to ensure adequate substrates for brain and muscle that may be essential for survival. In our study, this effect generally occurred before 5 d of hypoxic exposure. After about 5 d of hypoxia (in the subacute phase, when adaptation to hypoxia began), food intake was sustained at a low rate (possibly due to anorexia nervosa by activating CRFR2 and CRFR1), whereas plasma glucagon (Fig. 2D
) and hepatic glycogen (Fig. 2A
) were restored. But the blood glucose (Fig. 2B
) and plasma insulin (Fig. 2C
) were reduced, which may have resulted from sustained HPA axis activation and high levels of plasma corticosterone (Fig. 2F
). Therefore, transiently reducing hepatic glucagon levels to maintain blood glucose homeostasis is an acute response to hypoxia, maintaining glucose at nearly normoxic levels. Because glycogen reserves are physiologically regulated by glycogenolysis and glycogen synthesis through a dephosphorylation and phosphorylation, glucagons activates the corresponding protein kinase by phosphorylation (44). In this aspect, the detailed mechanism remains unknown.
Circulatory glucagon and insulin are modulated by hypoxia. Some studies show that acute hypoxia or intermittent hypoxia increases plasma glucose and insulin levels in the early phase, indicating that hypoxia induces glucose intolerance in humans (7) and increases insulin resistance in genetically obese mice (6). We found in this study that continuous hypoxia did not changed circulating glucose before d 5 and insulin levels before d 2 of exposure but caused a sustained reduction of blood glucose and insulin levels without high insulin resistance thereafter (Fig. 2
, B and C). Meanwhile, hepatic glycogen was decreased on d 1 and 2 of exposure (Fig. 2A
). Miki and Seino (45) recently reported that K+-ATP channels act as metabolic sensors of acute metabolic changes When a high concentration of extracellular glucose is transported into pancreatic ß-cells through the glucose transporter, intracellular glucose generates ATP through the tricarboxylic acid cycle in the mitochondrial, the rise of ATP closes K+-ATP-sensitive channels, which allows Ca2+ influx that triggers insulin secretion. However, hypoxia not only reduced plasma glucose during 515 d of exposure (Fig. 2B
) but also reduced insulin during d 215 of exposure (Fig. 2C
). In this case, reduced plasma glucose may result from lowered food intake, whereas reduced plasma insulin may result from inefficient ATP production (through suppression of the tricarboxylic acid pathway and triggered anaerobic glycolytic pathway), which opens K+-ATP sensitive channels, limiting Ca2+ influx and arresting insulin release from pancreatic ß-cells, resulting in decreased plasma insulin levels. In addition, continuous hypoxia of 5 km for 2 d transiently decreased hepatic glycogen (Fig. 2A
) and increased plasma glucagon levels (Fig. 2D
). Similarly increased circulating glucagon has also been found in the term human infant and rats exposed to severe hypoxia (46, 47). This may be due to a reduced blood glucose level. Because hypoglycemic stress contributes to changes of glucagon and insulin secretion through the autonomic nervous system (48), it is possible that the low blood glucose caused by hypoxia triggered hypothalamic glucose-responsive neurons through K+-ATP sensitive channels to activate the sympathetic system, resulting in stimulation of pancreatic
-cells to secrete glucagon to maintain glucose homeostasis (activating glycogenolysis and enhancing circulating glucose). However, this possible mechanism may not entirely explain the long-lasting low circulatory glucose levels because the hypoxia-activated high plasma glucagon level was observed at 2 d exposure but was not sustained thereafter.
Effects of CRFR1
The CRF receptor (CRFR) is a member of the secretin family of G protein-coupled receptors. Wide expression of CRFRs in the central nervous system and periphery ensures that their cognate agonists, the family of CRF-like peptides, are capable of exerting a wide spectrum of actions that underpin their critical role in integrating the stress response and coordinating the activity of fundamental physiological functions, particularly in the regulation of the energy balance and homeostasis. Two types of mammal CRFR exist, CRFR1 and CRFR2, each with remarkably distinct physiological properties.
For the first time, we present evidence that CRFR1 is involved in the modulation of hypoxia-induced reduction in circulating insulin and enhancement of glucagon (Fig. 3
). This modulation depends on CRFR1 activation. We assume this modulation occurs via an increase in circulating corticosterone and activation of CRFR1 in the pancreatic ß-cells. Pretreatment with a CRFR1 antagonist (CP-154,526) prevented the increase in circulating corticosterone (Fig. 3I
), blocked the reduction in plasma insulin (Fig. 3E
), and reversed the enhancement of plasma glucagon (Fig. 3G
). Evidence shows that CRFR1 exists in rat pancreatic ß-cells and CRF at a low concentration (2 nmol/liter), potentiates Ca2+ influx via the L-type Ca2+ channel by activating of the cAMP/protein kinase a signaling pathway through its own receptor (CRFR1). But CRF at a high concentration (20 nmol/liter) also inhibits Ca2+ influx through an unknown signaling pathway (49). The rise of Ca2+ triggers insulin secretion (45). We previously reported that hypoxia in rat causes an increase in CRF in the circulation (50) and CRF/CRF mRNA increases in the PVN (10), which can be blocked by subcutaneous pretreatment with CP-154,526, a CRF antagonist (10). These data suggest that hypoxia activates the HPA axis through CRFR1 and glucocorticoids may be involved in the early stimulation of glucagon secretion by the pancreatic
-cells and persistent inhibition of insulin secretion by the pancreatic ß-cells (Fig. 2C
). There is also evidence showing that circulating hormones antagonistic to insulin, such as adrenaline, are involved in reduced glucose-stimulated insulin secretion (51). Hypoxia causes norepinephrine increase in the circulation (52) and in the PVN and Amygdala of rats (27). These findings may support the idea that the sympathetic-adrenal systems may also be involved in the hypoxia-induced reduction in circulating insulin.
Overall, the evidence shows a key involvement of CRFR1 in the plasma and tissue responses to sustained hypoxia and suggests that activation of the HPA system is key for the changes in circulating hormones. As part of this activation, CRF increases corticotropin leading to an increase in circulating corticosterone. The latter glucocorticoid improves pancreatic
-cell secretion of glucagon and suppresses pancreatic ß-cell secretion of insulin. This is further enhanced by increased plasma adrenalin known to be increased by hypoxia, although we did not measure it in the present study.
In conclusion, continuous hypoxia of 5 km causes a loss of body weight, arrests food intake, causes a transient, early increase in circulating glucagon and leptin, and reduces hepatic glycogen. Furthermore, hypoxia induces a long-lasting low circulating insulin and blood glucose levels that are linked to an increase in circulating corticosterone via hypoxia-activated CRFR1, and activation of the sympathetic nervous system.
| Acknowledgments |
|---|
| Footnotes |
|---|
Xue-Qun Chen, Jing Dong, Chen-Ying Niu, Jun-Ming Fan, and Ji-Zeng Du have nothing to declare.
First Published Online March 22, 2007
Abbreviations: BW, Body weight; CP-154,526, CRFR1 antagonist; CRF, corticotropin-releasing factor; CRFR, CRF receptor; CRFR1, CRFR type 1; HPA, hypothalamus-pituitary-adrenal; PVN, hypothalamic paraventricular nucleus; SS, somatostatin.
Received September 8, 2006.
Accepted for publication March 15, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Ream, A. M. Ray, R. Chandra, and D. M. Chikaraishi Early fetal hypoxia leads to growth restriction and myocardial thinning Am J Physiol Regulatory Integrative Comp Physiol, August 1, 2008; 295(2): R583 - R595. [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 |