| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (R.W.G., C.D.M., G.J.M., M.W.S.), Harborview Medical Center, University of Washington, Seattle, Washington 98104; and Department of Medicine (J.O., R.S.F., D.E.C.), Veterans Affairs Puget Sound Healthcare System, Seattle, Washington 98108
Address all correspondence and requests for reprints to: Michael W. Schwartz, M.D., Harborview Medical Center, Division of Endocrinology, 325 Ninth Avenue, Box 359657, Seattle, Washington 98104. E-mail: mschwart{at}u.washington.edu.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Although the mechanisms whereby food consumption inhibits ghrelin secretion are poorly understood, absorption of ingested nutrients appears to play an essential role. This conclusion is based on the observations that intragastric gavage of a glucose solution rapidly lowers ghrelin levels, whereas an equal volume of water does not (3, 11, 12), and that the ghrelin-lowering effect of intragastric glucose is blocked if gastric emptying is prevented (12). Among several mechanisms forwarded to explain nutrient-induced inhibition of ghrelin secretion is the postprandial release of the pancreatic hormone insulin. This hypothesis is supported by the inverse temporal profiles of plasma ghrelin and insulin levels both before and after meals (9) and by evidence that sustained, pharmacological elevations of plasma insulin lower plasma ghrelin concentrations (13, 14, 15, 16, 17). Findings on this point are controversial (18, 19), however, and the physiologic role of insulin in the control of ghrelin release remains uncertain.
Peripheral or central ghrelin administration stimulates food intake and body weight gain in rodents (2, 3, 20) and humans (21) as potently as any known peptide (22). As is true of several other humoral regulators of energy balance (23, 24, 25), ghrelins feeding effects appear to be mediated via regulation of neurons in the hypothalamic arcuate nucleus. Specifically, ghrelin activates neurons that coexpress neuropeptide Y (NPY) and agouti gene-related protein (AgRP), peptides that potently stimulate feeding. This conclusion is based on evidence of ghrelin-induced activation of these NPY/AgRP neurons and on the finding that antagonism of neuronal signaling by either of these peptides attenuates ghrelins orexigenic action (26, 27, 28).
Insulin deficiency induced by ß-cell destruction after streptozotocin (STZ) administration provides a highly reproducible rodent model of uncontrolled, insulin-deficient diabetes mellitus and is characterized by weight loss, hyperglycemia, and markedly increased food intake (29). The hyperphagia of STZ-induced diabetes (STZ-DM) appears to arise, at least in part, from deficient hypothalamic signaling by insulin and leptin, which in turns lead to the activation of NPY/AgRP neurons (30, 31, 32). Furthermore, mice lacking NPY fail to increase food intake in response to STZ-DM, suggesting an obligatory role for increased NPY signaling in the effect of diabetes to increase food intake (32). In addition, STZ-DM markedly lowers plasma levels of both insulin and leptin and replacement of either hormone to physiological plasma concentrations attenuates the effects of STZ-DM on food intake and hypothalamic neuropeptide gene expression, as does intracerebroventricular (icv) infusion of insulin at a low dose (30, 31, 33). Whether the activation of NPY/AgRP neurons and hyperphagic feeding induced by STZ-DM is due solely to the loss of inhibitory input from leptin and insulin or involves increased ghrelin signaling in addition has yet to be determined.
In light of recent evidence showing that plasma ghrelin levels are elevated in rats with STZ-DM (34, 35), both increased ghrelin and reduced insulin and leptin signaling can be invoked to explain feeding and hypothalamic responses in this setting. Because food consumption potently inhibits ghrelin release, however, it seems paradoxical that hyperphagia due to uncontrolled diabetes should be characterized by elevated circulating levels of this peptide. Further limiting insight into the contribution of increased ghrelin signaling to diabetic hyperphagia is a lack of information regarding the temporal relationship between changes of plasma ghrelin levels, food intake, and other key metabolic parameters (e.g. plasma insulin, leptin and glucose levels, and body weight) after diabetes onset. A key question, for example, is whether ghrelin levels increase before the onset of diabetic hyperphagia (as expected if ghrelin participates causally in this phenomenon) and whether hyperphagia itself lowers plasma ghrelin in diabetic animals. To answer these questions, we determined the time course of alterations in circulating levels of ghrelin in relation to these parameters both before and after the onset of diabetic hyperphagia. We also wished to determine whether nutrient inhibition of ghrelin release requires a postprandial increase of plasma insulin levels and investigate whether ghrelin-induced feeding is enhanced in the setting of uncontrolled diabetes, as predicted in animals with reduced hypothalamic signaling by insulin and leptin.
| Materials and Methods |
|---|
|
|
|---|
Study protocols
Time course of the effect of STZ-DM on humoral parameters and food intake.
To determine the temporal relationships among changes of plasma ghrelin concentrations, food intake, and levels of glucose, leptin, and insulin after STZ administration, rats were outfitted with an indwelling intrajugular (IJ) catheter, as previously described (33), to allow the collection of daily blood samples. Postoperatively, animals were treated prophylactically with a broad-spectrum antibiotic (ceftriaxone, 100 mg/kg ip; Roche Laboratories, Nutley, NJ) and allowed to recover to presurgical levels of body weight and daily food intake before study. Rats were subsequently divided into two groups (n = 8/group) of equivalent body weight and were either made diabetic with a single iv injection of STZ [65 mg/kg dissolved in ice-cold 0.1 M sodium citrate (pH 4.5)] or received iv vehicle alone [0.1 M sodium citrate (pH 4.5)] and remained nondiabetic. Food intake and body weight were measured daily, and blood samples (
150 µl) were obtained during mid-light cycle (12001300 h) of each day for the determination of immunoreactive insulin, leptin, and ghrelin levels. Blood samples were placed into heparinized tubes, separated into plasma, and stored at 80 C until assay.
Effect of hyperphagic feeding on plasma ghrelin levels.
To determine whether the diabetic hyperphagia has an inhibitory effect on plasma ghrelin levels, rats were prepared with IJ catheters and received iv injections of either STZ (n = 23) or vehicle (sodium citrate) (n = 14), as described above. Once stable hyperphagia was established among STZ-treated rats (d 13 after STZ or vehicle injections), diabetic animals were subdivided into two groups. One was provided ad libitum access to food (n = 14), whereas the other was pair fed to the intake of nondiabetic controls (n = 9). Pair-feeding was accomplished by measuring food intake in nondiabetic rats at 2-h intervals from 0800 to 2000 h on the day before study and providing the pair-fed group with only that amount of food consumed by the control animals over the same time interval on the previous day. Food intake was measured at 2- to 4-h intervals in each of the three groups over this 12-h period and blood samples (
150 µl) were collected every 24 h for determination of plasma ghrelin levels.
Nutrient suppression of ghrelin levels in control and STZ-DM rats.
To determine the effect of STZ-DM on acute, nutrient-induced inhibition of ghrelin levels, rats were outfitted with IJ catheters as above and treated with either iv vehicle (sodium citrate; n = 4) or STZ (n = 6). Nineteen days later (when stable diabetic hyperphagia was well established in the diabetic group), the effect of intragastric gavage of 5 ml of a nutrient-rich solution (1 kcal/ml; Ensure, Abbot Laboratories, Abbott Park, IL) on plasma ghrelin and insulin levels was determined. Nondiabetic control animals were fasted for 16 h (18001000 h) before study, whereas STZ-treated diabetic rats were fasted for only 6 h (04001000 h) to match baseline plasma ghrelin levels between groups. Blood samples (150 µl) were collected before and at 30-min intervals for 2 h after gavage and were subsequently processed for measurement of plasma ghrelin and insulin levels.
Feeding response to a subthreshold dose of icv ghrelin.
To determine the effect of STZ-DM on sensitivity to ghrelin-induced feeding, we first identified a dose of icv ghrelin that is a subthreshold for feeding effects in nondiabetic rats and subsequently measured the feeding response to this dose (relative to icv vehicle) after the onset of STZ-induced diabetes. One week after cannulation of the third cerebral ventricle (36), cannula placement was assessed by measuring the drinking response to icv injection of angiotensin II (10 µg; American Peptide, Sunnyvale, CA) diluted saline (injection volume: 1 µl). Animals consuming less than 8 ml water over the 30-min period after injection were excluded as cannula failures (<5% of all animals). Two weeks after cannulation, either saline or ghrelin at one of four doses (2.5, 5, 50, or 500 pmol) was injected icv in a volume of 2 µl. Injections were given at 0800 h (early in the light cycle), and food intake was measured over the ensuing 2-h interval. Treatment order was randomized and a washout period of at least 24 h was given between injections. Three days after completing this dose-response study, all animals were made diabetic by iv injection of STZ. On d 12 and 13 after STZ, 2-h food intake was measured after icv injection of either vehicle or the lowest dose of ghrelin (2.5 pmol, which proved to be subthreshold for feeding effects when given before STZ injection). Data are presented as the difference in food intake after icv vehicle or ghrelin injection within each animal before and after induction of uncontrolled diabetes.
Blood glucose and plasma hormone assays.
Blood glucose was determined using a handheld glucometer (Accu-Check, Roche Diagnostics, Indianapolis, IN). Immunoreactive insulin and leptin levels were determined by ELISA (Crystal Chem, Inc., Chicago, IL), and ghrelin levels were determined with a commercially available RIA (Phoenix Pharmaceuticals, Belmont, CA) as previously described (9).
Statistical analyses
Data are presented as mean ± SEM. For multiple group comparisons, statistical significance was determined by one-way ANOVA using a Bonferronis post hoc test to assess differences between groups. For two-group comparisons, an unpaired, two-tailed Students t test was employed. Within-animal comparisons (e.g. effect of STZ-DM on ghrelin-induced food intake) were performed using a paired, two-tailed t test.
| Results |
|---|
|
|
|---|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Our finding of increased plasma ghrelin concentrations within 24 h of STZ administration, a time point well before the onset of hyperphagic feeding, is compatible with the hypothesis that increased ghrelin signaling contributes to the stimulatory effect of uncontrolled diabetes on food intake. As food intake increased, however, plasma ghrelin levels declined to values that, although still increased over nondiabetic controls, were well below those observed before hyperphagia onset. This finding suggests that even in the setting of severe, insulin-deficient diabetes, ghrelin secretion is sensitive to inhibition by food consumption.
Increased ghrelin concentrations in animals exhibiting robust hyperphagia seems paradoxical in light of the well-documented effect of food consumption to acutely and potently reduce plasma ghrelin levels in normal animals and humans (3, 9, 10, 11). To explain this paradox, we hypothesized that insulin deficiency and/or other consequences of uncontrolled diabetes stimulate ghrelin secretion and raise plasma concentrations but that as food intake increases, nutrient-induced inhibition of ghrelin secretion partially offsets the stimulatory effect of uncontrolled diabetes and lowers plasma levels of this hormone.
To test this hypothesis directly, we performed two experiments. First, we pair fed STZ-DM rats to the intake of nondiabetic controls to prevent hyperphagic feeding and demonstrated that as a consequence, ghrelin levels increased to values higher than those seen in diabetic animals that were allowed to overeat. In addition, we determined whether ghrelin levels are acutely reduced after an intragastric nutrient challenge in STZ-DM rats, as is documented in nondiabetic controls. Our finding that ghrelin levels were reduced comparably after an intragastric nutrient challenge in STZ-DM rats and controls demonstrates that nutrient suppression of ghrelin release is intact in diabetic rats. Together, these observations provide clear evidence that hyperphagic feeding offsets the stimulatory effect of STZ-diabetes on ghrelin secretion and that the interaction between these two opposing effects results in steady-state values of plasma ghrelin that are higher than in nondiabetic animals but are constrained by hyperphagic feeding.
The hypothesis that postprandial hyperinsulinemia contributes to meal-induced suppression of circulating ghrelin levels is supported by the inverse temporal relationship between plasma ghrelin and insulin concentrations, both before and after food consumption (9). The question of whether insulin plays a physiological role to inhibit ghrelin secretion remains unanswered, however, and previous studies have yielded mixed results. For example, experiments using supraphysiological insulin doses or a sustained hyperinsulinemic clamp have shown insulin-dependent suppression of circulating ghrelin levels (13, 14, 15, 16, 17, 19). However, stimulation of endogenous insulin release or administration of insulin at physiological doses did not reliably lower ghrelin levels (18), and the ability of specific nutrients to reduce ghrelin levels appears to involve mechanisms independent of insulin release (39).
Our observation that meal-induced suppression of plasma ghrelin is intact in STZ-DM rats, and that this response occurs despite the absence of any postprandial increase of plasma insulin levels, provides unequivocal evidence that meal- related increases of insulin are not required for acute nutrient-induced lowering of circulating ghrelin. Meal-induced suppression of plasma ghrelin levels was relatively short-lived in STZ-DM rats, however, suggesting that inhibitory effects of postprandial insulin release are a key determinant of the duration of meal-related inhibition of ghrelin release and thus the time course over which plasma levels of ghrelin are restored to preprandial values. This interpretation is compatible with findings from studies in humans with type 1 diabetes suggesting that insulin is a determinant of the duration of meal-induced ghrelin suppression, although controversy exists on this point (40, 41).
Another mechanism implicated in the nutritional regulation of ghrelin release involves modulation of vagal nerve activity. For example, although vagotomy affects neither baseline ghrelin levels nor the ability of a nutrient load to decrease ghrelin levels in rodents, both atropine treatment and vagotomy block the rise of ghrelin in response to a prolonged fast (42). These data suggest that whereas acute, meal-induced suppression of ghrelin levels does not involve the vagus nerve, the stimulatory effect of prolonged fasting does. Whether increased vagal efferent activity contributes to the effect of uncontrolled diabetes to increase ghrelin release is an interesting possibility for future study.
Diabetic hyperphagia is hypothesized to arise, at least in part, from reduced hypothalamic signaling by the adiposity-related hormones, insulin and leptin. Specifically, marked reductions in circulating levels of insulin and leptin are hypothesized to activate orexigenic NPY/AgRP neurons (43, 44, 45, 46) and inhibit adjacent melanocortin neurons (30, 47, 48) in the hypothalamic arcuate nucleus, a combination of responses that potently stimulates food intake. This model is supported by the observations that replacement of peripheral insulin or leptin in rats with STZ-DM corrects the hyperphagia and partially or completely normalizes the associated changes in hypothalamic neuropeptide gene expression (30, 31, 33, 46, 47, 49), and a similar effect occurs after icv infusion of insulin in these animals (31). Because NPY/AgRP neurons are also targets for the orexigenic action of ghrelin (22, 26, 28, 50), however, and because ghrelin levels are elevated in uncontrolled diabetes, activation of these orexigenic neurons in STZ-DM and attendant diabetic hyperphagia may arise, at least in part, from increased ghrelin signaling. Furthermore, it is possible that in the absence of the opposing effects of leptin and insulin, animals with uncontrolled diabetes are more sensitive to the feeding effects of ghrelin.
To investigate these possibilities further, we determined whether diabetic animals respond to a dose of icv ghrelin that is subthreshold for feeding effects in the nondiabetic state. Our finding that food intake was increased in response to a subthreshold dose of ghrelin after the induction of STZ-DM suggests an increased behavioral sensitivity to ghrelin, as might be expected under conditions of minimal opposing input from insulin and leptin. In light of this increased ghrelin sensitivity, elevated plasma ghrelin levels may contribute to diabetic hyperphagia even after the hyperphagia itself lessens the magnitude of ghrelin elevation. Further studies are therefore warranted to determine whether heightened sensitivity to the feeding effects of ghrelin detected in STZ-DM animals extends to circulating ghrelin as well and whether such an effect contributes to diabetic hyperphagia. Based on published evidence suggesting that ghrelin is synthesized in the hypothalamus (51), it will also be of interest to determine whether increased local ghrelin production and release is a component of the hypothalamic response to uncontrolled diabetes.
A recent report that daily food intake is not affected by targeted deletion of the ghrelin gene, and that ghrelin-deficient mice are not protected against diet-induced obesity, suggests that ghrelin may not be a critical determinant of overall energy homeostasis (52). Because compensatory mechanisms involving overlapping regulatory systems can be engaged when a key signaling molecule is deleted, however, negative findings from knockout models must be interpreted cautiously. An example relevant to energy homeostasis is that of NPY-null mice. Although these animals consume normal amounts of food on a daily basis and maintain normal body weight (53), hyperphagia in response to both STZ-induced diabetes (32) and insulin-induced hypoglycemia (54) is markedly attenuated or completely absent in these animals. Future studies that employ ghrelin-null mice and specific, high-affinity ghrelin antagonists are needed to clarify whether ghrelin plays an important physiologic role in the control of food intake and contributes to the pathogenesis of diabetic hyperphagia.
In conclusion, we report that increased circulating ghrelin levels in rats with uncontrolled diabetes occurs before the development of diabetic hyperphagia, compatible with a causal role for increased ghrelin signaling in the pathogenesis of this disorder. This hypothesis is further strengthened by evidence that sensitivity to the orexigenic effect of ghrelin is increased in rats with STZ-DM. Even in the diabetic state, however, food consumption acutely and potently suppresses plasma ghrelin levels, and our data show that the effect of STZ-DM to increase plasma ghrelin levels is partially offset by hyperphagic feeding. Collectively, these data support a model wherein increased ghrelin signaling acts in combination with decreased circulating insulin and leptin to activate hypothalamic neuropeptide responses that underlie increased food intake in uncontrolled diabetes.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: AgRP, Agouti gene-related protein; icv, intracerebroventricular; IJ, intrajugular; NPY, neuropeptide Y; STZ, streptozotocin; STZ-AL, STZ-DM with ad libitum access to food; STZ-DM, STZ-induced diabetes; STZ-PF, pair-fed STZ-DM.
Received May 12, 2004.
Accepted for publication July 2, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Ariga, K. Imai, C. Chen, C. Mantyh, T. N. Pappas, and T. Takahashi Does ghrelin explain accelerated gastric emptying in the early stages of diabetes mellitus? Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2008; 294(6): R1807 - R1812. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. E. Foster-Schubert, J. Overduin, C. E. Prudom, J. Liu, H. S. Callahan, B. D. Gaylinn, M. O. Thorner, and D. E. Cummings Acyl and Total Ghrelin Are Suppressed Strongly by Ingested Proteins, Weakly by Lipids, and Biphasically by Carbohydrates J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1971 - 1979. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Lechuga-Sancho, A. I. Arroba, L. M. Frago, C. Garcia-Caceres, A. D.-R. de Celix, J. Argente, and J. A. Chowen Reduction in the Number of Astrocytes and Their Projections Is Associated with Increased Synaptic Protein Density in the Hypothalamus of Poorly Controlled Diabetic Rats Endocrinology, November 1, 2006; 147(11): 5314 - 5324. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Arafat, F. H. Perschel, B. Otto, M. O. Weickert, H. Rochlitz, C. Schofl, J. Spranger, M. Mohlig, and A. F. H. Pfeiffer Glucagon Suppression of Ghrelin Secretion Is Exerted at Hypothalamus-Pituitary Level J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3528 - 3533. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. W. Gelling Diabetic Hyperphagia--Ghrelin in the Driver's Seat. Endocrinology, June 1, 2006; 147(6): 2631 - 2633. [Full Text] [PDF] |
||||
![]() |
J. Dong, T. L. Peeters, B. De Smet, D. Moechars, C. Delporte, P. Vanden Berghe, B. Coulie, M. Tang, and I. Depoortere Role of Endogenous Ghrelin in the Hyperphagia of Mice with Streptozotocin-Induced Diabetes Endocrinology, June 1, 2006; 147(6): 2634 - 2642. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. O. Mundinger, D. E. Cummings, and G. J. Taborsky Jr Direct Stimulation of Ghrelin Secretion by Sympathetic Nerves Endocrinology, June 1, 2006; 147(6): 2893 - 2901. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. A. Tritos and E. G. Kokkotou The Physiology and Potential Clinical Applications of Ghrelin, a Novel Peptide Hormone Mayo Clin. Proc., May 1, 2006; 81(5): 653 - 660. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. C. Griffen, K. Oostema, K. L. Stanhope, J. Graham, D. M. Styne, N. Glaser, D. E. Cummings, M. H. Connors, and P. J. Havel Administration of Lispro Insulin with Meals Improves Glycemic Control, Increases Circulating Leptin, and Suppresses Ghrelin, Compared with Regular/NPH Insulin in Female Patients with Type 1 Diabetes J. Clin. Endocrinol. Metab., February 1, 2006; 91(2): 485 - 491. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Tsubone, T. Masaki, I. Katsuragi, K. Tanaka, T. Kakuma, and H. Yoshimatsu Leptin downregulates ghrelin levels in streptozotocin-induced diabetic mice Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2005; 289(6): R1703 - R1706. [Abstract] [Full Text] [PDF] |
||||
![]() |
M A Arafat, B Otto, H Rochlitz, M Tschop, V Bahr, M Mohlig, S Diederich, J Spranger, and A F H Pfeiffer Glucagon inhibits ghrelin secretion in humans Eur. J. Endocrinol., September 1, 2005; 153(3): 397 - 402. [Abstract] [Full Text] [PDF] |
||||
![]() |
N Govoni, R De Iasio, C Cocco, A Parmeggiani, G Galeati, U Pagotto, C Brancia, M Spinaci, C Tamanini, R Pasquali, et al. Gastric immunolocalization and plasma profiles of acyl-ghrelin in fasted and fasted-refed prepuberal gilts J. Endocrinol., September 1, 2005; 186(3): 505 - 513. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Overduin, R. S. Frayo, H. J. Grill, J. M. Kaplan, and D. E. Cummings Role of the Duodenum and Macronutrient Type in Ghrelin Regulation Endocrinology, February 1, 2005; 146(2): 845 - 850. [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 |