| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
APPETITE, OBESITY AND METABOLISM
1Rheoscience, 2610 Rødovre, Denmark; 2Louisiana Scholars'College, Northwestern State University, Natchitoches; and 3Neurobiology of Nutrition Laboratory, Pennington Biomedical Research Center, Baton Rouge, Louisiana
Submitted 2 December 2002 ; accepted in final form 16 April 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
preproglucagon; fundus; corpus
-cells in
the pancreas, L cells in the gut, and neurons in the brain stem nucleus of the
solitary tract (NTS) (15).
Whereas posttranslational processing of proglucagon in the pancreas leads to
the formation of glucagon and the major proglucagon fragment, proteolytic
cleavage in the L cells of the gut and in the NTS yields the peptides
glicentin, oxyntomodulin, glucagon-like peptide (GLP)-1, and GLP-2
(2,
11,
18). GLP-1 is secreted from L
cells in response to a meal
(17) and plays an important
role as an incretin (12,
23). Additionally, peripheral
GLP-1 has been shown to delay gastric emptying
(30), and several groups have
demonstrated appetite-suppressive effects of peripherally administered GLP-1
or GLP-1 agonists in humans (7,
10,
47) and rats
(19). L cell-derived GLP-2 was
recently recognized to function as an intestinal epithelial growth factor
(5). In the central nervous system, GLP-1 and GLP-2 function as neurotransmitters in a discrete population of neurons located in the caudal part of the NTS. Immunohistochemical and neuroanatomical tract tracing methods have demonstrated dense GLP-1-containing projections to several hypothalamic areas involved in the regulation of food and water intake (14, 18, 22, 28), areas that also contain GLP-1 and GLP-2 receptors (22, 39). Consistent with these observations, intracerebroventricular administration of GLP-1 inhibits food and water intake in rats (38, 40, 44). Recently, GLP-2 (39) and oxyntomodulin (4) were shown to suppress food intake in rats when injected intracerebroventricularly. The precise function of GLP-1 in central appetite-regulating pathways has been intensely investigated over the last couple of years. Reports of GLP-1-induced taste aversion coupled with GLP-1-induced activation of central oxytocin-containing neurocircuits has led several groups to investigate the role of GLP-1 in aversive stimuli. Indeed, an increasing body of evidence points to a role of GLP-1 neurons in mediating anorexia associated with a variety of interoceptive stressors. Several chemical stimuli known to potently suppress food intake and produce conditioned taste aversion, such as intraperitoneal injections of LiCl or LPS as well as peripheral injections of cholecystokinin have been shown to induce c-Fos expression in brain stem GLP-1-immunoreactive (ir) neurons (28). Also, LiCl-induced anorexia and taste aversion can be significantly blunted by prior injection of GLP-1 receptor antagonists, further supporting a functional role of endogenous GLP-1-containing neural pathways as mediators of LiCl-induced aversive behavior (27, 35). Although these findings point to a role of GLP-1 neurons as important mediators of interoceptive stressors, the neurons are in an excellent position to respond also to more physiologically relevant gastrointestinal satiety signals in the course of a normal meal.
In the present experiment we sought to investigate the effect of a purely mechanical gastrointestinal stimulus (gastric balloon distension) on the activation of GLP-1/2 and catecholamine-synthesizing neurons in the caudal part of the NTS by the use of triple labeling immunohistochemistry for GLP-2, tyrosine hydroxylase (TH), and the immediate early gene c-Fos. The experiment was carried out in freely moving rats, and the rate and magnitude of distension were designed to mimic an average physiological meal in moderately food-deprived rats.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
Distension experiments and perfusion. In several sessions, animals were adapted in special acrylic test cages to rinsing the stomach with warm saline and inserting a small latex balloon. On the experimental day, after 15–17 h of overnight food deprivation, the empty balloon was inserted through the fistula and placed in the proximal stomach. After animals rested for 1 h, the balloon was inflated with 9 ml of warm saline at a rate of 0.9 ml/min, and inflation was maintained for an additional 50 min. In control animals, the balloon was inserted but not inflated, and the external tubing to the balloon was cut to avoid mucosal stimulation by the empty balloon. One animal was excluded from the experiment because of a leaky balloon. One hour after the start of distension, animals were killed with an overdose of pentobarbital sodium and perfused transcardially with 150 ml of heparinized saline (20 U/ml) followed by 400 ml of cold 4% phosphate-buffered paraformaldehyde (pH 7.4) containing 0.3% picric acid. After the perfusion, the exact location of the inflated balloon within the stomach was determined by estimating the percentage of the total volume of the balloon in the fundus, corpus, and antrum, with the limiting ridge delineating the border between fundus and corpus. On the basis of this evaluation, two groups were formed: one group (n = 6; 1 animal excluded due to a leaky balloon) with >70% of the total balloon volume in the fundus, and one group (n = 7) with >60% of the balloon volume in the corpus and antrum. Brains were removed and postfixed in the same fixative at 4°C overnight. A block containing the caudal brain stem was placed overnight in 4% paraformaldehyde containing 25% sucrose and frozen, and coronal sections were cut in a cryostat in five series at 30-µm intervals. The sections were stored at -20°C in cryoprotectant solution (50% PBS, 20% glycerol, 30% ethylene glycol) for later processing.
Immunohistochemisty. One series of sections from each rat was subjected to a triple immunohistochemical procedure to visualize c-Fos, GLP-2, and TH simultaneously. Because both the GLP-2 and TH antibodies were mouse monoclonal, it was necessary to visualize these antigens by the use of tyramide amplification technique as has been described by us and others previously (24, 36). Tyramide amplification allows the use of one of the primary antibodies in dilutions [monoclonal GLP-2 antibody (F1221) diluted 1:1,600] that are undetectable by a direct technique (fluorescent secondary antibodies). Therefore, once the tyramide signal amplification reaction has been terminated, the other primary mouse antibody can be used in a concentration suitable for direct visualization. This, combined with the fact that brain stem GLP-1 (which is costored with GLP-2)-immunopositive neurons are separate from TH-positive neurons (18), made the use of two primary mouse antibodies possible. All reactions were carried out on free-floating sections. First, sections were rinsed 3 x 10 min in phosphate-buffered saline containing 50 mM KCl (KPBS, pH 7.4). Next, sections were transferred to a 10 mM sodium citrate solution (pH 6.0) and heated in a water bath at 85°C for 90 min. Others (13) have described this method of unmasking antigens in formaldehyde and paraffin-embedded tissue in detail. Sections were allowed to cool for 20 min in the citrate buffer, then rinsed 2 x 5 min in KPBS followed by a 10-min incubation in 1 ml of 1% H2O2 in KPBS. After a 20-min incubation in 5% swine serum in KPBS containing 0.3% Triton X-100 (TX) and 1.0% BSA, the sections were incubated overnight at 4°C in rabbit anti-c-Fos [1:100,000; code 94012–4; (46)] and mouse anti-GLP-2 antibodies (1:1,600, code F1221, generously provided by Dr. Jes T. Clausen, NovoNordisk A/S) diluted in KPBS containing 0.3% TX and 1.0% BSA. The next day, the sections were washed for 3 x 10 min in KPBS with 0.1% TX (KPBS-T) before incubation for 60 min at room temperature in a biotinylated donkey anti-rabbit antibody (Jackson Immunoresearch Lab) diluted 1:200 in KPBS-T. After three rinses for 10 min in KPBS-T followed by 60 min in ABC-streptavidin-horseradish peroxidase (Vector Elite Kit; Vector Laboratories, Burlingame, CA) the sections were washed 3 x 10 min in KPBS-T. Sections were then reacted for 12 min in biotinylated tyramide [biotinylated according to the procedure described in Ref. 1 using tyramine-HCl (Sigma Aldrich) and sulfosuccinimidyl-6-(biotinimide) hexanoate (NHS-LC-biotin; Pierce, Rockford, IL)] diluted 1:100 in KPBS-T containing 0.005% H2O2. After the tyramide amplification, sections were rinsed 3 x 5 min in KPBS-T and incubated for 60 min in streptavidin-Alexa 546 (Molecular Probes, Eugene, OR) diluted 1:200 in KPBS-T. Three rinses for 5 min in KPBS-T preceded 10 min in KPBS-T + 1% H2O2, the latter to destroy remaining horseradish peroxidase activity. After 3 x 5 min in KPBS-T the sections were incubated for 60 min in a horseradish peroxidase coupled donkey anti-mouse antibody (to label the mouse GLP-2 antibody), rinsed 3 x 5 min in KPBS-T, and subsequently incubated for 12 min in FITC-labeled tyramine diluted 1:100 in KPBS-T + 0.005% H2O2 [labeled using tyramine-HCL (Sigma Aldrich) and HS-LCFITC (fluorescein, Pierce) in a procedure similar to the biotinylated tyramine described above]. After this detection reaction, sections were rinsed 3 x 5 min in KPBS-T and incubated overnight at 4°C with a mouse monoclonal antibody to tyrosine hydroxylase (Sigma Aldrich, batch 020K4836) diluted 1:4,000 in KPBS + 0.3% TX + 1.0% BSA. On the third day, sections were washed 3 x 10 min in KPBS-T, incubated for 60 min in a Cy5-coupled donkey anti-mouse antibody (Jackson Immunoresearch Laboratories) diluted 1:200 in KPBS-T, then rinsed 2 x 10 min in KPBS-T and finally 10 min in distilled water before being mounted in a 0.5% gelatine solution. Slides were dehydrated through a graded series of ethanol dilutions (30%, 60%, 80%, 96%, 99%, and xylene) before being placed under a coverslip in Pertex.
Image acquisition and estimation of cell counts. A Zeiss confocal
laser-scanning microscope (LSM 510) equipped with two HeNe lasers and one
argon laser was used to visualize the three fluorophores simultaneously.
Images used for the semiquantitative analysis were acquired using a x20
objective and a pinhole size that yielded optical sections 
6 µm thick.
One optical section was acquired per section where all three antigens were
optimally stained. Images were acquired from all sections containing the NTS
caudal to the rostral tip of the area postrema (10–12 sections per
animal, each section spaced 150 µm). In the commissural part of the NTS,
one image covered the entire NTS. At more rostral levels, two images, covering
each side of the NTS, were acquired. Cells [c-Fos-ir nuclei, GLP-2-ir cells,
TH-ir cells] were counted automatically using Image Pro Plus 4.6. The program
was set to count the number of c-Fos-, GLP-2-, and TH-positive cells based on
size and color/intensity. The number of GLP-2- and TH-positive cells
colocalizing c-Fos was scored manually in the same session by an observer
blinded to the treatment. To look for possible regional differences, the
analyzed NTS sections were divided into rostral sections (sections of the NTS
underlying the area postrema), midsections (caudal to the area postrema,
rostral to the pyramidal decussation), and caudal sections (at the level of
the pyramidal decussation). The fraction (percentage) of GLP-2-ir cells
containing c-Fos for the NTS as a whole was calculated as (total number of
GLP-2-positive cells with c-Fos)/(total number of GLP-2-positive
cells)x100. The fraction of GLP-2/TH-ir cells containing c-Fos for the
three rostrocaudal levels of the NTS was calculated in a similar fashion. TH-,
GLP-2-, and c-Fos-ir cell counts are expressed as number of cells per section
rather than total number of cells to adjust for differences in numbers of
sections (varied from 10–12 sections per animal).
Production of photomicrographs. Tricolor images acquired using the LSM 510 Zeiss laser confocal microscope were enhanced for brightness and contrast in Adobe Photoshop 7.0 and combined into plates using Adobe Illustrator 10.0 (Fig. 3). Adobe Illustrator plates (plates 70 to 73) from the atlas by Swanson (37) were used to generate the map of labeled cells in Fig. 4.
|
|
Statistics. Statistical evaluation of the cell counts and percentage colocalization used one-way ANOVA followed by Fisher's post hoc analysis. Results are expressed as means ± SE from groups of four to six rats. P values <0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
The aim of our study was to elucidate the specific role of the gastric phase of meal ingestion on activity of medullary GLP-1/2 neurons. Whereas normal ingestion of a meal activates many additional neural and humoral sensory systems, including signals from the oral cavity, the small intestine, and postabsorptive sites, balloon distension is a method to selectively activate gastric mechanoreceptors. Notably, to reflect the gastric distension associated with the ingestion of a large meal, the balloon was inflated to mimic the normal fill pattern of rats given a liquid diet after an overnight fast (9). Several studies have examined the pattern of c-Fos expression in the NTS after gastric distension, and it has been shown that c-Fos expression increases with both volume and the rate of distension (pressure) (6, 43, 45). This is in line with electrophysiological recordings from the afferent vagus nerve, showing a dose-dependent relationship between firing rate and gastric volume (31, 32). Additionally, it has been demonstrated that the load-sensitive gastric vagal afferents encode volume rather than nutrients (20), in line with conclusions from behavioral experiments examining appetite-suppressive feedback signals arising from the stomach (26).
The importance of vagal mechanoreceptors in short-term satiety was recently examined in neurotrophin-4 (NT-4)-deficient mice (8). NT-4 knockout mice display a selective loss of one class of vagal mechanoreceptor, the intraganglionic laminar ending (IGLE), whereas the other class of mechanoreceptor, the intramuscular array, remains intact (8). IGLE-deficient mice have longer and larger meals, suggesting a role for IGLEs in short-term satiety, probably by conveying distension feedback to the NTS (8).
Interestingly, the GLP-1 receptor-null mutant mouse also displays mild disturbances in short-term satiety (34). Initial characterization of the GLP-1 receptor knockout mouse questioned the role of GLP-1 in appetite regulation due to the lack of an obese pheno-type (33). However, closer examination of the data presented by Scrocchi et al. (34), shows that indeed the GLP-1 receptor knockout mice display a disturbed satiety sequence by terminating their initial feeding period of the dark phase significantly later than that of wild-type controls, resulting in an increase in food intake in the first hours of the dark phase (34). Coupled with our observation of GLP-1/2 neuron activation by gastric distension, it is tempting to speculate that these neurons play a role in mechanoreceptor-mediated short-term satiation. It would be highly interesting to examine gastric distension-induced activation of GLP-1/2 neurons in NT-4-deficient mice.
The induction of c-Fos in GLP-containing neurons by gastric distension demonstrated in the present experiments contrasts with recent data from Rinaman (28), who examined c-Fos expression in GLP-1-ir neurons after the consumption of a large meal. Whereas ingestion of a large liquid meal was unable to induce c-Fos expression in NTS preproglucagon neurons, Rinaman showed that up to 70% of these cells contained c-Fos after the administration of interoceptive stressors (LPS, LiCl, CCK in high doses) (28). Although it is possible that the differences could have been influenced by the immunohistochemical methods used, the completely different nature of the gastric stimulation used in the two studies clearly makes direct comparison difficult. These differences are emphasized by data from a recent study by Emond et al. (6). Using c-Fos immunocytochemistry, Emond et al. (6) showed that a large degree of signal convergence and integration of different meal-related stimuli occur at the level of the NTS. Notably, the same gastric load resulted in a greater induction of c-Fos in the NTS when present alone than when present together with other meal-related stimuli (6). This clear example of integration at the level of the NTS could explain why selective stimulation of mechanoreceptors (present study) and ingestion of a large meal (28) lead to differential activation of GLP-1/2-containing neurons.
Our finding of distension-induced activation of GLP-1/2-containing neurons raises a number of questions regarding the functional role of these neurons in the regulation of food intake. To date, preproglucagon-derived peptides, GLP-1, GLP-2, and oxyntomodulin have all been shown to inhibit food intake when administered intracerebroventricularly to rats (4, 38, 39, 41, 44). Although the specificity of GLP-1-induced anorexia has been questioned by reports showing taste aversion after centrally administered GLP-1 (16, 41, 42), specific (that is nonaversive) appetite suppression by GLP-1 can be elicited by direct injections into the hypothalamic paraventricular nucleus (21) or into the fourth ventricle (16). Furthermore, in monosodium glutamate-treated rats, central GLP-1-induced appetite suppression is eliminated, whereas GLP-1-induced taste aversion is intact (40), suggesting that indeed GLP-1 acts specifically on central appetite-regulating neurocircuits. A firm body of evidence now supports a role of NTS preproglucagon neurons as mediators of the anorexia elicited by visceral malaise. Rinaman (28) has shown that three potent interoceptive stressors (CCK, LiCl, and LPS) all induce c-Fos expression in a large proportion of GLP-1-ir neurons in the NTS, and LiCl-induced suppression of food intake can be partly blocked by prior central administration of GLP-1 receptor antagonists (27, 35), pointing to a functional role of central GLP-1 neurons in LiCl-induced anorexia. In line with these observations, it was recently found that GLP-1 neurons are activated by central oxytocin administration and that GLP-1 receptor blockade diminishes anorectic responses to oxytocin (29). As central blockade of oxytocinergic neural systems have also been implicated in LiCl anorexia (25), these data suggest that GLP-1 neurons are activated by nauseating stimuli tapping into central oxytocinergic neural pathways (29). In light of the literature supporting a role of GLP-1 in mediating the anorexia associated with gastric malaise, it would be interesting to investigate the ability of gastric balloon distension to produce a conditioned taste aversion.
In conclusion, we showed that moderate gastric distension can activate hindbrain GLP-1/2-containing neurons, consistent with a role of central preproglucagon-derived peptides as inhibitors food intake. The tendency for corpus distension to activate more GLP-2-positive neurons than fundus distension further suggests that, even among gastric distension sensors, the capacity to signal to the brain stem can be different.
|
DISCLOSURES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONDISCLOSURESREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Babic, R. L. Townsend, L. M. Patterson, G. M. Sutton, H. Zheng, and H.-R. Berthoud Phenotype of neurons in the nucleus of the solitary tract that express CCK-induced activation of the ERK signaling pathway Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2009; 296(4): R845 - R854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J Larsen Mechanisms behind GLP-1 induced weight loss The British Journal of Diabetes & Vascular Disease, November 1, 2008; 8(2_suppl): S34 - S41. [Abstract] [PDF]
|
|
|
|
 |
|
 D. A. Sandoval, D. Bagnol, S. C. Woods, D. A. D'Alessio, and R. J. Seeley Arcuate Glucagon-Like Peptide 1 Receptors Regulate Glucose Homeostasis but Not Food Intake Diabetes, August 1, 2008; 57(8): 2046 - 2054. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Faipoux, D. Tome, S. Gougis, N. Darcel, and G. Fromentin Proteins Activate Satiety-Related Neuronal Pathways in the Brainstem and Hypothalamus of Rats J. Nutr., June 1, 2008; 138(6): 1172 - 1178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Holst The Physiology of Glucagon-like Peptide 1 Physiol Rev, October 1, 2007; 87(4): 1409 - 1439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wan, F. H. Coleman, and R. A. Travagli Glucagon-like peptide-1 excites pancreas-projecting preganglionic vagal motoneurons Am J Physiol Gastrointest Liver Physiol, June 1, 2007; 292(6): G1474 - G1482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zheng, L. M. Patterson, C. B. Phifer, and H.-R. Berthoud Brain stem melanocortinergic modulation of meal size and identification of hypothalamic POMC projections Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2005; 289(1): R247 - R258. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. A. Cupples Physiological regulation of food intake Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2005; 288(6): R1438 - R1443. [Full Text] [PDF]
|
|
|
|
|
|
P. K. Chelikani, A. C. Haver, and R. D. Reidelberger Intravenous infusion of glucagon-like peptide-1 potently inhibits food intake, sham feeding, and gastric emptying in rats Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2005; 288(6): R1695 - R1706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Badman and J. S. Flier The Gut and Energy Balance: Visceral Allies in the Obesity Wars Science, March 25, 2005; 307(5717): 1909 - 1914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Rinaman Hindbrain contributions to anorexia Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2004; 287(5): R1035 - R1036. [Full Text] [PDF]
|
|
|
|
|
|
L. L. Baggio, Q. Huang, T. J. Brown, and D. J. Drucker A Recombinant Human Glucagon-Like Peptide (GLP)-1-Albumin Protein (Albugon) Mimics Peptidergic Activation of GLP-1 Receptor-Dependent Pathways Coupled With Satiety, Gastrointestinal Motility, and Glucose Homeostasis Diabetes, September 1, 2004; 53(9): 2492 - 2500. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Mazda, H. Yamamoto, M. Fujimura, and M. Fujimiya Gastric distension-induced release of 5-HT stimulates c-fos expression in specific brain nuclei via 5-HT3 receptors in conscious rats Am J Physiol Gastrointest Liver Physiol, July 1, 2004; 287(1): G228 - G235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. A. Cupples Endo-neuro-endocrine incretin pathways Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2004; 286(2): R250 - R250. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
)-, TH (
)-, and GLP-2-positive (
) neurons are
plotted in single 30-µm tissue sections through the caudal NTS. GLP-2-ir
and TH-ir neurons containing c-Fos are shown as
and
, respectively.
Distribution of cells in 1 control animal and in 1 corpus-distended rat
(PG44b) is shown. Drawings are arranged from rostral (top) to caudal
(bottom) and are from the atlas by Swanson (plates 70 to 73)
(
