| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (J.O., R.S.F., D.E.C.), Division of Metabolism, Endocrinology and Nutrition, University of Washington, Veterans Affairs Puget Sound Health Care System, Seattle, Washington 98108; and Department of Psychology (H.J.G., J.M.K.), University of Pennsylvania, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: David E. Cummings, M.D., Associate Professor of Medicine, University of Washington, Veterans Affairs Puget Sound Health Care System, 1660 South Columbian Way, S-111-Endo, Seattle, Washington 98108. E-mail: davidec{at}u.washington.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
70%) than did lipids (by 50%). Because jejunal nutrient infusions suppressed ghrelin levels as well as either gastric or duodenal infusions, we conclude that the inhibitory signals mediating postprandial ghrelin suppression are not derived discretely from either the stomach or duodenum. The relatively weak suppression of ghrelin by lipids compared with glucose or amino acids could represent one mechanism promoting high-fat dietary weight gain. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first goal of this study was to examine the contribution of nutrients within the duodenal lumen to ghrelin suppression. We hypothesized that the duodenum is necessary for the prandial ghrelin response based, in part, on observations that ghrelin regulation is disrupted after Roux-en-Y gastric bypass (RYGB) surgery (12, 25, 26). Ingested nutrients strongly regulate ghrelin, and RYGB excludes the majority of ghrelin-producing tissue (i.e. the stomach and duodenum) from contact with enteral nutrients. Because nutrients constrained within the stomach do not affect ghrelin levels (24), disordered ghrelin regulation after RYGB suggests a role for duodenal nutrients in ghrelin regulation. Moreover, nutrients within the duodenum regulate levels of other gut hormones involved in appetite control as well as endocrine and exocrine pancreatic output, gastric motility, and satiety (27, 28, 29, 30). If the nutrient-sensing mechanism mediating prandial ghrelin suppression is located discretely in the duodenum, then nutrients delivered into the stomach or duodenum should suppress ghrelin levels, whereas equivalent intestinal infusions delivered distal to the duodenum should not suppress ghrelin levels. To determine whether the duodenum is required for nutrient-related ghrelin suppression, we infused nutrients into the stomach, duodenum, or jejunum in conscious, unanesthetized rats fitted with chronic, indwelling intestinal catheters.
The second goal of this study was to compare the magnitude of ghrelin suppression after intestinal infusions of different macronutrient types. Absorption of food from the gut is accompanied by a characteristic pattern of humoral and neural responses for each class of macronutrient (28, 31). Therefore, differences in the dynamics of ghrelin suppression after isocaloric infusions of various macronutrients may shed light on the identity of signals mediating ghrelin suppression. Furthermore, such findings might have practical relevance for the design of diets intended to lower ghrelin levels. We examined the temporal profile of ghrelin suppression in rats infused with isocaloric solutions of glucose, lipids, or amino acids via indwelling intestinal catheters.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rat surgery
For surgery, animals were anesthetized with a mixture of 60 mg/kg ketamine and 7 mg/kg xylazine (Phoenix, St. Joseph, MO) administered ip.
Duodenal and jejunal feeding catheter placement and testing.
Rats received either a proximal duodenal catheter (entry point at 2 cm distal to the pylorus) or a proximal jejunal catheter (entry point at 25 cm distal to the pylorus, i.e. 15 cm distal to the duodenojejunal junction). Silastic tubing (outer diameter, 0.047 in; inner diameter, 0.025 in; Braintree Scientific, Braintree, MA) was fitted with a collar consisting of a 4-cm2 piece of Bard surgical mesh (Davol, Cranston, RI), which was used to tether the catheter to the intestine. After visualization of the stomach and small intestine, a puncture hole was made in the intestinal wall using an 18-gauge needle. The end of the SILASTIC brand silicon tubing (Dow Corning, Midland, MI) distal to the mesh was inserted approximately 2 cm into the intestinal lumen, pointing caudally. The mesh was tethered to the outer intestinal wall with a three-loop purse string of 6–0 silk suture (U.S. Surgical, Norwalk, CT). In pilot studies designed to verify that jejunal infusions did not flow in a retrograde direction into the duodenum, we infused barium sulfate into jejunal catheters, using the same flow rate and volume as were used for nutrient infusions in our experiments. Fluoroscopic visualization revealed only a minute amount of retrograde flow, extending less than 2 cm from the catheter tip, which was located approximately 17 cm distal to the duodenum. Similarly, visual inspection of slow infusions of methylene blue into the jejunal catheters revealed no detectable retrograde flow (data not shown).
Jugular-vein catheter placement.
A 2-cm incision was made through the skin in the right jugular area, and the jugular vein was cannulated with SILASTIC brand silicon tubing (outer diameter, 0.037 in; inner diameter, 0.02 in; VWR Scientific, West Chester, PA). The other end of the tubing was tunneled sc to the head, onto which it was secured, along with the intestinal catheter, with an acrylic skull cap (Lang Dental, Wheeling, IL) (32).
Experimental protocol
Rats recovered from surgery for at least 2 wk before nutrient infusions began; by this time, body weights of all animals had returned to presurgical levels. Experiments were conducted in custom-built, Plexiglas test cages (30 x 25 x 35 cm) that facilitated intestinal infusions and repeated blood sampling without disturbance of the animals. Several days before the start of each experiment, rats were habituated overnight to the experimental cages. At 1700 h on the day before infusions, each animal was placed individually in an experimental cage and was food deprived for 18 h to establish high baseline ghrelin levels (33). One hour before infusions, each animal’s head cap was connected with respective lines (PE-100; VWR) for blood sampling and nutrient administration. Infusions were initiated at 1100 h (5 h into the light cycle).
Gastrointestinal infusions.
Infusion conditions were run in random order and spaced at least 1 wk apart from one another. The carbohydrate solution consisted of 25% glucose, which is known to suppress ghrelin levels when delivered into the stomach (24). Three milliliters were delivered into the small intestine over 10 min at a mean rate of 0.3 ml/min. This rate of delivery approximates that of normal gastric emptying in rats voluntarily ingesting 25% glucose (34). The lipid and amino acid infusions, also administered over 10 min, were made isocaloric to the glucose infusion. The lipid infusion consisted of Intralipid (Baxter Healthcare, Deerfield, IL), a broad-spectrum mixture of long-chain triglycerides and phospholipids, that was diluted with distilled water to obtain a volume of 3 ml. The amino acid solution was Prosol 20% Amino Acids, which is a water-soluble, broad-spectrum mixture of essential and nonessential amino acids (a generous gift from Baxter Healthcare). To render the amino acid infusion isocaloric to the glucose and lipid infusions, we used a slightly larger volume of this solution (3.75 ml). Two control conditions were also examined. First, 3 ml of 25% glucose was infused into the stomach by gavage, as a positive control in which nutrients entered the small intestine more physiologically than by direct infusion (i.e. mixed and diluted with gastric secretions, then passed though the pylorus). Second, we infused 3 ml of physiologic saline (0.9% NaCl) intestinally over 10 min to control for the volume-related effects of intestinal infusions on ghrelin levels.
Blood sampling and analysis.
Five minutes before the start of nutrient infusions, a baseline 250-µl blood sample was drawn from the jugular-vein catheter, and additional 250-µl samples were taken at 30, 60, 90, 120, 180, 240, and 300 min after the onset of infusions. Blood glucose levels were analyzed in a small drop from each sample using a portable glucose meter (Accu-Check; Roche, Indianapolis, IN). The remaining blood was transferred to 0.5-ml microcentrifuge tubes containing 10 µl of 7.5% EDTA and then placed immediately on ice. As soon as possible, blood samples were centrifuged, and the plasma was withdrawn and stored at –80 C. Plasma levels of total ghrelin were measured in duplicate samples of 25 µl each by RIA using a primary antibody against rat ghrelin and 131I-labeled ghrelin as the tracer (kit RK-031-31; Phoenix Pharmaceuticals, Belmont, CA) (11). This assay detects both acylated and des-acyl ghrelin. Although only acylated ghrelin is bioactive, levels of the total and acylated forms correlate closely with one another over a wide variety of physiological manipulations that affect ghrelin (35, 36, 37). Plasma insulin was measured in duplicate samples of 25 µl each using a commercial RIA (kit SRI-13K; Linco Research Inc., St. Charles, MO). The intra- and interassay coefficients of variance were 5.5 and 9.6%, respectively, for ghrelin, and 3.7 and 5.7%, respectively, for insulin.
Data analysis
Data were analyzed using Unistat software (Unistat Ltd., London, UK) running under Excel (Microsoft, Redmond, WA). For most purposes, ghrelin levels were expressed as percentages of baseline values. Three parameters were calculated to describe the dynamics of ghrelin response to gastrointestinal infusions. Ghrelin nadir (GN) was the lowest ghrelin level (expressed as a percentage of baseline) reached within the 5-h test period after infusions. Time to ghrelin nadir (TGN) was the time required for ghrelin to reach its lowest postinfusion level. Decremental area under the curve (D-AUC) was calculated to obtain an overall index of the ghrelin response, reflecting both the depth and duration of suppression. The D-AUC was defined as 100% minus the area under the ghrelin curve, which was determined using the trapezoidal rule. Graphically, D-AUC represents the area enclosed by the preinfusion baseline level (100%) and the postinfusion ghrelin curve. Two-tailed paired t tests with a significance threshold of P = 0.05 were used to test differences between pre- and postinfusion ghrelin levels. The effect of duodenal vs. jejunal infusion site on ghrelin regulation was tested with two-sided unpaired t tests for each macronutrient, with a significance threshold of P = 0.05. Effects of macronutrient type on ghrelin suppression were tested by one-way ANOVA, followed by unpaired two-sided t tests. To account for multiple post hoc comparisons, we used the Bonferroni correction, setting the significance threshold at P = 0.05/3 = 0.0167. Results are expressed as mean ± SEM, unless otherwise stated.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). This finding demonstrates that the indwelling intestinal catheters did not impair normal nutrient-related ghrelin regulation. To assess whether duodenal exposure to nutrients is required for prandial ghrelin regulation, nutrients were infused either into the proximal duodenum or more distally into the jejunum. For each of the three macronutrient types, duodenal and jejunal infusions suppressed ghrelin levels equally well, with similar magnitude and timing of suppression. This can been seen in Fig. 1, as well as in Table 1, which quantifies the depth, speed, and overall magnitude of ghrelin suppression after duodenal or jejunal isocaloric infusions of glucose, amino acids, or lipids. For each macronutrient type, duodenal and jejunal infusions yielded equivalent results in terms of the postinfusion GN, TGN, and integrated D-AUC (P > 0.2 for all comparisons between duodenal and jejunal effects on GN, TGN, and D-AUC parameters; Fig. 1 and Table 1).
|
|
GN.
A one-way ANOVA revealed a significant effect of macronutrient type on GN (F(2,38) = 26.4, P < 0.001). GN was significantly lower after both glucose and amino acids than after lipids (P < 0.001 for both; Fig. 2). No differences were found between glucose and amino acid infusions.
|
). D-AUC after glucose infusions was larger than after either amino acids or lipids (P < 0.001). No significant difference in D-AUC was found between amino acids and lipids.
TGN.
A one-way ANOVA showed a significant effect of macronutrient type on TGN (F(2,38) = 16.9, P < 0.001), and results for TGN largely mirrored those for GN (Fig. 2). TGN was similar after glucose (58 ± 5 min) and amino acid infusions (64 ± 4 min), and each of these times was shorter than that after lipid infusions (133 ± 17 min, P < 0.001). As with GN and D-AUC, the pattern of results was identical when data for duodenal and jejunal infusions were considered separately, except that the difference in TGN for jejunal infusions of amino acids vs. lipids did not quite reach statistical significance (P = 0.06).
Blood glucose levels.
Compared with baseline values, blood glucose levels increased by 49.5 ± 4.2 mg/dl after glucose infusions (P < 0.001) and by 16.2 ± 3.0 mg/dl after amino acid infusions (P < 0.001). These levels returned to baseline by 120 min after the start of both infusions (Fig. 3A). Lipids did not affect blood glucose levels.
|
Insulin levels.
Plasma insulin levels increased from baseline by 374 ± 78 pmol/liter after glucose infusions (P = 0.003), by 162 ± 13 pmol/liter after amino acids (P < 0.001), and slightly after lipids (by 55 ± 9 pmol/liter, P = 0.007; Fig. 3B). As with blood glucose, insulin levels returned to baseline by 120 min after the start of all infusions. Maximum increases of insulin were larger after glucose and amino acid infusions than after lipids (P = 0.002 and P < 0.001, respectively). Maximum increases were larger after glucose than after amino acid infusions (P < 0.02).
Correlations among ghrelin, insulin, and blood glucose responses to gastrointestinal nutrient infusions.
The especially strong suppression of ghrelin after intestinal infusions of either glucose or amino acids, both of which stimulate insulin secretion, suggests a possible dominant role for insulin and/or glucose in the postprandial ghrelin response. If this were true, the depth and total magnitude of nutrient-related ghrelin suppression should correlate with the height and total magnitude of increases in insulin and/or glucose. Thus, we analyzed these correlations, as shown in Table 2. For all macronutrient infusions analyzed together, GN correlated inversely with the maximum rise in both insulin and glucose levels (Spearman rank correlations, P < 0.001 for both; Table 2). Similarly, the D-AUC of ghrelin correlated significantly with the AUCs of insulin and glucose (P < 0.001 for both; Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Presuming that ghrelin is regulated similarly in rats and humans, as has generally been found in studies to date (14, 15, 38), one implication of these findings is that the perturbation of ghrelin regulation that typically follows RYGB surgery (12, 25, 26) is unlikely to result from lack of nutrient flux through the stomach and duodenum. We had initially hypothesized that prandial ghrelin suppression results from direct contact between ingested nutrients within the gastrointestinal lumen and ghrelin-producing cells, most of which are in the stomach and duodenum (3). Accordingly, we had hypothesized that the condition of a permanently empty stomach and duodenum after RYGB would cause unrelenting ghrelin stimulation, leading ultimately to paradoxical ghrelin suppression via override inhibition (12). This phenomenon would be analogous to the paradoxical suppression of gonadotropins or GH by continuous infusions of GnRH or GHRH, respectively (39, 40). In view of our current findings, however, showing that nutrient-related ghrelin regulation does not require nutrients within the lumen of the stomach or duodenum, it now seems unlikely that this override inhibition hypothesis is valid.
Our second principal finding was that isocaloric intestinal infusions of either glucose or amino acids suppressed ghrelin levels more rapidly and effectively than did lipid infusions. These observations are consistent with and extend those of a recent report showing that intragastric infusions of glucose, peptone, casein, or corn oil all suppressed plasma ghrelin in anesthetized rats (41). In that work, however, infusions across the different macronutrient classes were not isocaloric, and the ghrelin responses were only assessed at a single postinfusion time point. Thus, quantitative comparisons of the ghrelin responses to various macronutrients are not possible from the prior study. Our finding that ghrelin was suppressed less effectively by lipids than by glucose or amino acids could have clinical implications. Theoretically, weak suppression of an orexigenic hormone by ingested lipids could be one of the mechanisms underlying high-fat diet-induced weight gain (42).
For all infusions combined, the depth of ghrelin suppression correlated significantly with the magnitude of increase in both glucose and insulin levels; similarly, the D-AUC for ghrelin correlated with the AUCs for glucose and insulin. Although we did not directly test whether increases in blood glucose and/or insulin cause ghrelin suppression, studies of rats made insulin deficient with the ß-cell toxin, streptozotocin, suggest that insulin participates in postprandial ghrelin suppression in this species but that other factors are also involved (43). Two observations that support this assertion can be made from our current data. First, lipid infusions suppressed ghrelin levels by approximately 50% (Table 1), without any increase in glucose levels and with only a marginal increase in insulin (Fig. 3). Second, regardless of the macronutrient type infused, ghrelin levels remained suppressed long after blood glucose and insulin levels had returned to baseline (Figs. 2 and 3). Thus, it is unlikely that nutrient-related ghrelin suppression is driven solely by circulating glucose and insulin, although these factors may contribute to the response, as suggested by several studies (43, 44).
Two additional factors, rate of nutrient absorption and increases in osmolarity within the intestinal lumen, may partly explain the dynamics of ghrelin responses in our experiments. Glucose and amino acids, which are quickly absorbed from the gut, suppressed ghrelin rapidly and deeply. In contrast, lipids, which require intestinal digestion before absorption (45), suppressed ghrelin more gradually and to a lesser extent. This difference could imply that nutrient-related ghrelin suppression results from signals generated during or after nutrient absorption. Moreover, although our infusions of various macronutrients were isocaloric with one another, they varied significantly in osmolarity, and the strength of ghrelin suppression followed a pattern consistent with a possible contribution from changes in intestinal osmolarity. Specifically, infusions of glucose and amino acids (with osmolarities of 1450 and 1800 mOsm/liter, respectively) suppressed ghrelin more strongly than did the comparatively less osmotic lipid infusions (
450 mOsm/liter).
Other potential pathways governing meal-related ghrelin suppression could include signaling from factors secreted by intestinal enteroendocrine cells (e.g. serotonin, glucose-dependent insulinotropic peptide, cholecystokinin, glucagon-like peptide-1, etc.). These cells respond to all macronutrients in vitro and in vivo (28). Prandial ghrelin suppression could also involve neural pathways, specifically the myenteric plexus but probably not the vagus nerve. Although vagal afferent pathways are involved in other nutrient-related signaling from the gut (28, 29), the vagus is not required for prandial ghrelin regulation, as judged from experiments in vagotomized rats (33).
In summary, we find that nutrient-related ghrelin suppression does not require the presence of nutrients in either the stomach or duodenum, which are the principal sites of ghrelin production. Thus, prandial ghrelin regulation is probably mediated by intestinal signals located downstream of the ligament of Treitz (or at least distributed throughout the intestine) and/or by postabsorptive mechanisms. Among the latter, circulating glucose and insulin probably contribute to nutrient-related ghrelin suppression. However, they are unlikely to explain the entire response because ghrelin remained suppressed in our study long after normalization of glucose and insulin levels and because lipids suppressed ghrelin in the absence of substantial increases in glucose or insulin levels. Our findings pertain only to the regulation of circulating ghrelin, which is produced primarily by the stomach and proximal small intestine. The data do not address regulation of the very small amount of ghrelin that has been reported to be produced in the hypothalamus (46). The weaker suppression of peripheral ghrelin, an orexigenic hormone, by lipids than by glucose or amino acids could constitute one of many reasons why high-fat diets promote weight gain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online November 4, 2004
Abbreviations: D-AUC, Decremental area under the curve; GN, ghrelin nadir; RYGB, Roux-en-Y gastric bypass; TGN, time to ghrelin nadir.
Received May 12, 2004.
Accepted for publication October 13, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. A Parnell and R. A Reimer Weight loss during oligofructose supplementation is associated with decreased ghrelin and increased peptide YY in overweight and obese adults Am. J. Clinical Nutrition, June 1, 2009; 89(6): 1751 - 1759. [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]
|
|
|
|
 |
|
 B. J. Bradford, K. J. Harvatine, and M. S. Allen Dietary Unsaturated Fatty Acids Increase Plasma Glucagon-Like Peptide-1 and Cholecystokinin and May Decrease Premeal Ghrelin in Lactating Dairy Cows J Dairy Sci, April 1, 2008; 91(4): 1443 - 1450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Frecka and R. D. Mattes Possible entrainment of ghrelin to habitual meal patterns in humans Am J Physiol Gastrointest Liver Physiol, March 1, 2008; 294(3): G699 - G707. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Sun, N. F. Butte, J. M. Garcia, and R. G. Smith Characterization of Adult Ghrelin and Ghrelin Receptor Knockout Mice under Positive and Negative Energy Balance Endocrinology, February 1, 2008; 149(2): 843 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J Little, M. Horowitz, and C. Feinle-Bisset Modulation by high-fat diets of gastrointestinal function and hormones associated with the regulation of energy intake: implications for the pathophysiology of obesity Am. J. Clinical Nutrition, September 1, 2007; 86(3): 531 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rauh, M. Groschl, and W. Rascher Simultaneous Quantification of Ghrelin and Desacyl-Ghrelin by Liquid Chromatography-Tandem Mass Spectrometry in Plasma, Serum, and Cell Supernatants Clin. Chem., May 1, 2007; 53(5): 902 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Roche, A. J. Sheahan, L. M. Chagas, and D. P. Berry Concentrate Supplementation Reduces Postprandial Plasma Ghrelin in Grazing Dairy Cows: A Possible Neuroendocrine Basis for Reduced Pasture Intake in Supplemented Cows J Dairy Sci, March 1, 2007; 90(3): 1354 - 1363. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Little, S. Doran, J. H. Meyer, A. J. P. M. Smout, D. G. O'Donovan, K.-L. Wu, K. L. Jones, J. Wishart, C. K. Rayner, M. Horowitz, et al. The release of GLP-1 and ghrelin, but not GIP and CCK, by glucose is dependent upon the length of small intestine exposed Am J Physiol Endocrinol Metab, September 1, 2006; 291(3): E647 - E655. [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]
|
|
|
|
|
|
M. Romon, S. Gomila, P. Hincker, B. Soudan, and J. Dallongeville Influence of Weight Loss on Plasma Ghrelin Responses to High-Fat and High-Carbohydrate Test Meals in Obese Women J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 1034 - 1041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Williams and D. E. Cummings Regulation of Ghrelin in Physiologic and Pathophysiologic States J. Nutr., May 1, 2005; 135(5): 1320 - 1325. [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 |
