Gastrin release after truncal vagotomy in fistula dogs: Hypersensitivity to bombesin but not bethanechol

Integrated gastrin response was measured by the serial changes in serum immunoreactive gastrin after various stimuli in three dogs with gastric fistula and highly selective fundic vagotomy, who were then subjected to truncal vagotomy. Truncal vagotomy eliminated the gastrin as well as the gastric acid response to vagal excitation by 2-deoxy-glucose, but did not significantly change the responses to bethanechol (20 or 120 μg/kg/hr by IV infusion). Acid output was the same with bombesin or its nonapeptide in the dogs with fundic vagotomy as it was after subsequent truncal vagotomy, but gastrin release was very much increased by truncal vagotomy. For a 3-hour infusion of bombesin integrated gastrin release was 65 and 143 ng/ml/min and for its nonapeptide 43 and 109 ng/ml/min in the dogs with fundic and truncal vagotomy respectively. The marked hypersensitivity of the gastrin response after truncal vagotomy to bombesin but not to a cholinergic agonist suggests that the antral denervation led to a post-denervation hyper-response to the putative transmitter, bombesin, and that the vagal release of antral gastrin may thus represent a peptidergic neurohormonal mechanism. Also, a long half-life of effect suggests that bombesin binds avidly to its receptors.     

Free amino acids in basal and vagally stimulated gastric secretion

The concentrations of the individual free amino acids were determined in one hour fraction of basal secretion and peak hydrogen ion secretion following stimulation with 2-deoxy-D-glucose (2-DG) (group I) or insulin (group II). Group I consisted of 9 patients with duodenal ulcer having hypersecretion of gastric acid as determined by histamine test; 7 patients with duodenal ulcer who underwent truncal vagotomy and had insulin test performed two weeks after the operation formed group II. The total concentration of free amino acids was similar in basal and in stimulated gastric juice in both groups. Also the concentrations of the individual amino acids did not change significantly after stimulation. There was, however, a significant increase following stimulation in the output of amino acids both in group I and in group II. This increase was parallel to that in the volume of gastric juice, which suggests that a definite amount of free amino acids is always present in the gastric juice, and that the secretion of these acids is not under vagal control.     

Effects of truncal,selective, and highly selective vagotomy on glucose tolerance and insulin secretion in patients with duodenal ulcer. Part I-Effect of vagotomy on response to oral glucose.

An oral glucose tolerance test was performed in patients who had undergone truncal vagotomy and pyloroplasty, bilateral selective vagotomy and pyloroplasty, or highly selective vagotomy without a drainage procedure at least six months earlier. The results were compared with those from patients with chronic duodenal ulcer before operation. In all three groups of patients after vagotomy more rapid rates of rise of blood glucose and higher peak concentrations were observed than in patients who were tested before operation. These differences were statistically significant only in patients who had undergone truncal or selective vagotomy with pyloroplasty and were probably due to more rapid rates of gastric emptying after these operations. Plasma insulin concentrations were lower after truncal vagotomy than after selective or highly selective vagotomy, the difference between truncal vagotomy and highly selective vagotomy being statistically significant. Truncal vagotomy resulted in a diminished insulin response to oral glucose, which could have been due to vagal denervation of the pancreas or, more probably, impaired release of small-bowel hormones which normally augment the pancreatic insulin response.     

Centrally applied atrial natriuretic factor diminishes bile secretion in the rat.

Little is known about the role of centrally applied peptides in the regulation of bile secretion. We previously reported that the intravenous injection of atrial natriuretic factor (ANF) reduces bile acid dependent flow without affecting portal venous pressure in the rat. In the present work, we studied the effects of centrally applied ANF on bile secretion and the possible pathways involved. Rats were cannulated in the brain lateral ventricle for the administration of 1, 10 and 100 ng/microl ANF. After 1 week, the common bile duct was cannulated and bile samples were collected every 15 min for 60 min after the administration of ANF. The excretion rate of various biliary components was assessed. Bile secretion experiments were also performed after bilateral truncal vagotomy or atropine administration to evaluate the participation of a vagal pathway. In addition, the role of the sympathetic system was addressed by combined administration of propranolol and phentolamine. Centrally applied ANF did not modify blood pressure but diminished bile flow and bile acid output. It also reduced sodium and potassium secretion but did not modify protein or phospholipid excretion. Neither bilateral truncal vagotomy nor atropine administration abolished ANF response. Furthermore, combined administration of adrenergic antagonists did not alter ANF inhibitory effect on bile flow. In conclusion, centrally applied ANF reduced bile acid dependent flow not through a vagal or adrenergic pathway in the rat, suggesting the involvement of a peptidergic pathway.     

Evidence for intrinsic regulation of met-enkephalin-immunoreactivity in gastroenteropancreatic tissues of the rat

In a study whether gastrointestinal endogenous opioids can be modified by vagal denervation or by pharmacological application of an opiate, we examined met-enkephalin-immunoreactivity in gastrointestinal tissue in rats with and without truncal vagotomy and with and without subcutaneously implanted morphine pellets. The immunoreactivity of the tissue extracts gave dose-response lines in the radioimmunoassay for met-enkephalin which were near parallel to that for the standard. On Sephadex chromatography the met-enkephalin immunoreactivity eluted at a position similar to synthetic met-enkephalin. Tissue concentration of met-enkephalin immunoreactivity was not significantly different from the respective control after vagotomy and after morphine treatment. Total gastric met-enkephalin immunoreactive content increased significantly after vagotomy in line with gastric hypertrophy occurring after vagotomy without a drainage procedure. From these results it is concluded that met-enkephalin immunoreactivity in the rat gastrointestinal tract is regulated intrinsically, it is neither altered by vagal denervation nor by exogenous opiate administration.     

Treatment of severe side effects after vagotomy and gastroenterostomy by closure of gastroenterostomy without pyloroplasty.

We describe nine patients who had severe, persistent abdominal pain, vomiting, dumping, or diarrhoea several years after truncal vagotomy and gastroenterostomy had been performed for duodenal ulceration. Each patient was judged to have a bad clinical result (Visick grade 4). There was no evidence of recurrent ulceration in any of the patients, and in each the patency of the pyloric canal was confirmed radiologically or endoscopically. Each patient was treated by simply dismantling the gastroenterostomy without addition for a pyloroplasty. In one patient the surgeon suspected that a vagal trunk might have been left intact, and a revagotomy was performed by the "highly selective" technique. Postoperatively, none of the patients developed gastric retention. Symptomatic improvement occurred in eight patients, and four of them achieved perfect results (Visick grade 1). Side effects are common after vagotomy and gastroenterostomy, and are largely attributable to the presence of the gastroenterostomy stoma. Our results show that the symptoms may be alleviated by closing the gastroenterostomy, without precipitating gastric retention.     

Gastric emptying in response to IAPP and CCK in rats with subdiaphragmatic afferent vagotomy

In the subdiaphragmatic vagal deafferentation procedure (SDA), the afferent fibers of the vagus are surgically severed unilaterally where they enter the brain stem. The technique includes a subdiaphragmal truncal vagotomy performed on the contralateral side. This procedure has been used to study the control of food intake, but it has not been used previously to investigate the role of vagal afferent fibers in the control of gastric emptying (GE). The current experiment studied the effect of SDA on the inhibition of GE by islet amyloid polypeptide (IAPP or amylin) and cholecystokinin (CCK) in awake, unrestrained rats with gastric cannulas. The experimental group underwent subdiaphragmatic vagal deafferentation; the control group had sham operations. All rats received 20-min intravenous infusions of IAPP (1, 3, 9, 27, and 81 pmol/kg/min), CCK (3, 30 and 90 pmol/kg/min), and normal saline in random order. Gastric emptying of saline was measured by the phenol red method during the last 5 min of each infusion period. CCK dose-dependently inhibited gastric emptying in both the control and SDA animals. The inhibition of GE by CCK was significantly attenuated by SDA (p<0.01). IAPP also inhibited gastric emptying dose-dependently, but the difference between the SDA and control groups was not significant. The current experiment, which used a different methodology than previous studies, provides support for the hypothesis that the inhibition of gastric emptying by CCK, but not by IAPP, is mediated partly by afferent vagal fibers.     

Effect of vagotomy and atropine on plasma somatostatin response to a meal in conscious dogs

In 4 conscious dogs with gastric fistulas the somatostatin responses to a meal were measured and compared to the responses seen after i.v. infusion of atropine sulfate (20 and 50 micrograms.kg-1.h-1) or cimetidine (8 mg.kg-1.h-1). The experiments were repeated after truncal vagotomy. The somatostatin responses to bombesin (0.5 micrograms.kg-1.h-1) were also measured before and after vagotomy. Vagotomy decreased basal and postprandial somatostatin levels and reduced the somatostatin responses to feeding during the first 30-min period following the ingestion of the meal but not during subsequent periods. Bombesin-induced somatostatin release was increased after vagotomy. Atropine decreased the somatostatin responses to the meal before and after vagotomy. Cimetidine had no significant effect. These studies suggest that, in conscious dogs, somatostatin released into the circulation is partly under vagal control and that, as for gastrin release, vagal pathways for stimulation and inhibition are present. Our studies also suggest that cholinergic mechanisms are involved in the control of postprandial somatostatin release.     

The experimental study for efficacy and safety of pancreatic cryosurgery

Objective: This study was designed to basic information concerning the efficacy and safety of cryosurgery for pancreatic cancer. Fifteen healthy pigs were used to perform biochemical analysis and histological assessment. Methods: Following anesthesia and laparotomy, an argon–helium cryoprobe was inserted into the pancreas. The introduction of argon gas induced a rapid decrease in temperature to ?160 °C (Group I, 5 pigs) or ?110 °C (Group II, 5 pigs), respectively, resulting in ice-ball formation of 15–20 mm diameter after 5 min. Following freezing, helium gas was circulated in the probe tip to increase the temperature to 10–20 °C over 3 min to thaw. The freeze/thaw cycle was then repeated. Group III (3 pigs) had a cryoprobe inserted, but without freezing, and Group IV (2 pigs) included untreated or normal control animals. Levels of serum amylase (AMY), IL-6 and C-RP were measured prior to freezing and for 7 days following the procedure. All pigs were euthanized 7 days post-treatment and pancreases were examined histologically. Results: Neither hyperaemia, edema or hemorrhage were observed in the un-frozen parts of the pancreas. Histological assessment revealed a significant level of necrosis in the central and lateral regions of the tissue frozen within the ice-ball. All cellular ultrastructure was destroyed and only observable as a few of remaining nuclei with broken crests and degranulated mitochondria and rough endoplasmic reticulum. There was a significant increase of serum AMY levels for a brief period in both “deep frozen” and the “shallow frozen” groups. However, the AMY also increased in two pigs in the “normal control” group and one pig from the “inserted cryoprobe without freeze” control group. All experimental pigs appeared healthy until the sacrifice time. Conclusion: Cryosurgery is a safe and effective ablative procedure for pancreatic tissue resulting in minimal complications.     

Gastrin and the vagus interact in the trophic control of the rat oxyntic mucosa

Gastrin is a trophic hormone for the stomach, and permanent reduction of circulating gastrin by antrectomy leads to atrophy of the oxyntic mucosa, including a reduced density of histamine-storing endocrine cells (so-called ECL cells). Recently, it was proposed that also the vagal nerve has a trophic influence on the stomach. The two vagal trunks innervate the anterior and posterior side of the gastric wall, respectively. This arrangement makes it possible to denervate one side of the stomach selectively. The objective of the present study was to examine the consequences of combined antrectomy and vagotomy (unilateral or bilateral). Male Sprague-Dawley rats were subjected to unilateral or bilateral subdiaphragmatic truncal vagotomy with or without antrectomy. Control rats were sham-operated. The rats were killed 8 weeks after the operation. Bilateral vagotomy raised the basal serum gastrin concentration (fasting level). The thickness of the oxyntic mucosa and the density of ECL cells were not significantly different from age-matched vagally intact controls. Unilateral vagotomy induced no change in the basal serum gastrin concentration, nor did it affect the mucosa on the intact side. On the denervated side, however, there was reduced mucosal thickness and a greatly reduced ECL cell density. With a combination of antrectomy and vagal denervation the decrease in ECL cell density was exaggerated compared to the effect of antrectomy or unilateral vagotomy alone.(ABSTRACT TRUNCATED AT 250 WORDS)     

Bombesin microinjected into the dorsal vagal complex inhibits vagally stimulated gastric acid secretion in the rat

Medullary sites of action for bombesin-induced inhibition of gastric acid secretion were investigated in urethane-anesthetized rats with gastric fistula. Unilateral microinjection of bombesin or vehicle into the dorsal vagal complex was performed using a glass micropipet and pressure ejection of 100 nl volume; gastric acid output was measured every 10 min by flushing the stomach. Microinjection of vehicle into the dorsal vagal complex did not alter gastric acid secretion (1.9 +/- mumol/10) from preinjection levels (2.9 +/- 0.8 mumol/10 min). Microinjection of the stable thyrotropin-releasing hormone (TRH) analog, RX 77368, at a 77 pmol dose into the dorsal vagal complex stimulated gastric acid secretion for 100 min with a peak response at 40 min (24.1 +/- 3.2 mumol/10 min). Concomitant microinjection of RX 77368 (77 pmol) with bombesin (0.6-6.2 pmol) into the dorsal vagal complex dose dependently inhibited by 35-86% the gastric acid response to the TRH analog. Bombesin (6.2 pmol) microinjected into the dorsal vagal complex inhibited by 17% pentagastrin infusion-induced stimulation of gastric acid secretion (13.2 +/- 0.8 mumol/10 min) whereas intracisternal injection induced a 69% inhibition of the pentagastrin response. These results demonstrate that the dorsal motor complex is a sensitive site of action for bombesin-induced inhibition of vagally stimulated gastric secretion. However, other medullary sites must be involved in mediating the inhibitory effect of intracisternal bombesin on pentagastrin-stimulated gastric acid secretion.     

Vagotomy without Diarrhoea

The incidence of diarrhoea after three types of vagotomy was assessed “blind” at a gastric follow-up clinic one year after operation. Diarrhoea was recorded in 24% of patients after truncal vagotomy and pyloroplasty, in 18% after selective vagotomy and pyloroplasty, but in only 2% of patients after highly selective vagotomy without a drainage procedure. The incidence of diarrhoea was significantly less (P < 0·01) after highly selective vagotomy than after either of the other procedures.Hypertonic glucose solution given by mouth to 15 representative patients from each group and to 15 patients before operation provoked the onset of diarrhoea in 67% of the patients who had undergone truncal vagotomy and pyloroplasty, in 60% of those who had undergone selective vagotomy and pyloroplasty, in 13% of those who had undergone highly selective vagotomy without a drainage procedure, and in none of the preoperative patients. Again the difference between the “highly selective” group and the other two groups of vagotomized patients was statistically significant.It is suggested that postvagotomy diarrhoea is attributable both to unregulated gastric emptying after truncal or selective vagotomy with a drainage procedure and to the extragastric denervation produced by truncal vagotomy. “Postvagotomy” diarrhoea can be virtually eliminated by using highly selective vagotomy without a drainage procedure.     

Incidence of Dumping after Truncal and Selective Vagotomy with Pyloroplasty and Highly Selective Vagotomy without Drainage Procedure

The incidence of dumping after truncal or selective vagotomy with pyloroplasty and highly selective vagotomy without a drainage procedure was assessed both clinically and experimentally. At a gastric follow-up clinic dumping was found to be significantly less frequent in patients who had undergone highly selective vagotomy without a drainage procedure than in patients who had undergone truncal or selective vagotomy with pyloroplasty (P < 0·05 or < 0·001, respectively). Hypertonic glucose given by mouth provoked the onset of dumping in 20% of patients with duodenal ulcer before operation, in 73% after truncal vagotomy and pyloroplasty, in 80% after selective vagotomy and pyloroplasty, and in 47% after highly selective vagotomy. The test meal also produced significantly greater decreases in blood pressure and increases in pulse rate in patients who had undergone vagotomy with pyloroplasty than in patients who had undergone highly selective vagotomy.     

Inhibition of gastric acid secretion by intracerebral injection of calcitonin gene related peptide in rats

Calcitonin gene related peptide (CGRP), a 37 amino acid peptide recently characterized from the rat brain, injected into the cisterna magna or the lateral hypothalamus, inhibited gastric acid secretion and increased serum gastrin levels in conscious pylorus-ligated rats. CGRP suppressed pentagastrin- and histamine-induced stimulation of gastric secretion in urethane-anesthetized gastric fistula rats. Peptide action was dose-dependent, not modified by adrenalectomy and peripheral sympathectomy induced by guanethidine pretreatment and reversed by vagotomy. These results demonstrate that intracisternal or intralateral hypothalamic injection of CGRP inhibits stimulated gastric acid secretion. The mechanism of action is vagal dependent and unrelated to modification of gastrin secretion or activation of sympathetic outflow.     

Preganglionic parasympathetic innervation in normal and partially denervated rat stomach

Retrograde transport of horseradish peroxidase (HRP) was used to study the organization of the preganglionic parasympathetic innervation of the normal and vagotomized rat stomach. In normal rats, following application of HRP to either the gastric fundus or corpus, HRP accumulated bilaterally in nerve cells in the dorsal vagal motor nuclei, but predominantly in the left nucleus. The biggest populations of cells innervated the lesser curvature of the stomach. When HRP was applied at different periods of time after truncal or selective proximal vagotomy, there was no labeling of nerve cells in the dorsal vagal motor nuclei. Regeneration of efferent vagal nerve fibers and the recovery of gastric activity is discussed.A. A. Bogomolets Institute of Physiology, Academy of Sciences of the Ukrainian SSR, Kiev. Institute of Physiology, T. G. Shevchenko Kiev State University. Translated from Neirofiziologiya, Vol. 23, No. 2, pp. 231–238, March–April, 1991.     

Effect of truncal vagotomy on pancreatic polypeptide response after intravenous glucose administration

Intravenous glucose infusion was performed in six dogs with and without truncal vagotomy, and plasma pancreatic polypeptide (PP) responses were compared before and after truncal vagotomy. Following truncal vagotomy, basal PP levels decreased significantly from 286 ± 64 pg/ml (mean ± S.E.) to 94 ± 14 pg/ml (P < 0.05). Basal plasma insulin and blood glucose levels also tended to be lower, but not significantly. During the influsion of glucose, blood glucose concentrations rose rapidly in both groups and after 15 min reached peak values which were not significantly different from each other. In the vagotomized group the plasma insulin response to intravenous glucose infusion was significantly lower than in the control group. Following intravenous glucose loading, plasma PP concentrations decreased rapidly in both groups, but the PP level in the vagotomized group was suppressed only to 77 ± 4% of the basal level whereas in the control group it decreased to 45 ± 8%, significantly lower than in the vagotomized group (P < 0.01).These results suggest that basal PP is regulated by vagal tonus and that vagus controls, at least in part, suppression by intravenous glucose administration.     

Motilin and the vagus in dogs

The aim of this work was to determine the influence of the vagus on the circulating levels of immunoreactive (IR) motilin. Five mongrel dogs were equipped with chronically implanted electrodes in the small intestine to record the myoelectrical activity. The release of IR motilin during fasting, after a meal, and during an infusion of insulin was studied before and after truncal vagotomy at the diaphragmatic level. When tested at least two weeks after the operation, the motility pattern of the small intestine and the secretion of IR motilin remained unaltered by vagal section. Cyclic increases in IR motilin associated with phase III's of the interdigestive myoelectric complexes were still observed after vagotomy (maximum levels of IR motilin: 250 +/- 37 versus 239 +/- 19 fmol X mL-1, not significant), and they were still abolished by feeding or by insulin. However, an inhibitory influence can probably be mediated by the vagus since, in normal animals, vagal stimulation by a "modified sham feeding" (tease feeding or presentation of food) at the beginning of a period of phase III activity promptly interrupted this part of the complex and decreased significantly the release of IR motilin by about 20%. The release of motilin is not chronically altered by distal vagotomy in dogs.     

Gastric relaxation induced by hyperglycemia is mediated by vagal afferent pathways in the rat

Hyperglycemia has a profound effect on gastric motility. However, little is known about the site and mechanism that sense alteration in blood glucose level. The identification of glucose-sensing neurons in the nodose ganglia led us to hypothesize that hyperglycemia acts through vagal afferent pathways to inhibit gastric motility. With the use of a glucose-clamp rat model, we showed that glucose decreased intragastric pressure in a dose-dependent manner. In contrast to intravenous infusion of glucose, intracisternal injection of glucose at 250 and 500 mg/dl had little effect on intragastric pressure. Pretreatment with hexamethonium, as well as truncal vagotomy, abolished the gastric motor responses to hyperglycemia (250 mg/dl), and perivagal and gastroduodenal applications of capsaicin significantly reduced the gastric responses to hyperglycemia. In contrast, hyperglycemia had no effect on the gastric contraction induced by electrical field stimulation or carbachol (10(-5) M). To rule out involvement of serotonergic pathways, we showed that neither granisetron (5-HT(3) antagonist, 0.5 g/kg) nor pharmacological depletion of 5-HT using p-chlorophenylalanine (5-HT synthesis inhibitor) affected gastric relaxation induced by hyperglycemia. Lastly, N(G)-nitro-L-arginine methyl ester (L-NAME) and a VIP antagonist each partially reduced gastric relaxation induced by hyperglycemia and, in combination, completely abolished gastric responses. In conclusion, hyperglycemia inhibits gastric motility through a capsaicin-sensitive vagal afferent pathway originating from the gastroduodenal mucosa. Hyperglycemia stimulates vagal afferents, which, in turn, activate vagal efferent cholinergic pathways synapsing with intragastric nitric oxide- and VIP-containing neurons to mediate gastric relaxation.     

Vagal modulation of intestinal afferent sensitivity to systemic LPS in the rat

The central nervous system modulates inflammation in the gastrointestinal tract via efferent vagal pathways. We hypothesized that these vagal efferents receive synaptic input from vagal afferents, representing an autonomic feedback mechanism. The consequence of this vagovagal reflex for afferent signal generation in response to LPS was examined in the present study. Different modifications of the vagal innervation or sham procedures were performed in anesthetized rats. Extracellular mesenteric afferent nerve discharge and systemic blood pressure were recorded in vivo before and after systemic administration of LPS (6 mg/kg iv). Mesenteric afferent nerve discharge increased dramatically following LPS, which was unchanged when vagal efferent traffic was eliminated by acute vagotomy. In chronically vagotomized animals, to eliminate both vagal afferent and efferent traffic, the increase in afferent firing 3.5 min after LPS was reduced to 3.2 +/- 2.5 impulses/s above baseline compared with 42.2 +/- 2.0 impulses/s in controls (P < 0.001). A similar effect was observed following perivagal capsaicin, which was used to eliminate vagal afferent traffic only. LPS also caused a transient hypotension (<10 min), a partial recovery, and then persistent hypertension that was exacerbated by all three procedures. Mechanosensitivity was increased 15 min following LPS but had recovered at 30 min in all subgroups except for the chronic vagotomy group. In conclusion, discharge in capsaicin-sensitive mesenteric vagal afferents is augmented following systemic LPS. This activity, through a vagovagal pathway, helps to attenuate the effects of septic shock. The persistent hypersensitivity to mechanical stimulation after chronic vagal denervation suggests that the vagus exerts a regulatory influence on spinal afferent sensitization following LPS.     

Acetylcholine content in superior cervical ganglion following reinnervation of vagal afferent fibers in cat

The superior cervical ganglion (SCG) was reinnervated by vagal afferent fibers by cross anastomosis between the cranial end of nodose ganglion and the caudal end of SCG in cats. Formation of functional synapses was evidenced by unilateral mydriasis and contraction of the nictitating membrane in response to inflation of the stomach with a balloon or to electrical stimulation of the afferent vagus. The acetylcholine (ACh) content in the cross-anastomosed SCG (reinnervated by vagal afferent fibers) was measured. In anastomosed SCG, the ACh content was about half of normal SCG, but significantly higher than chronically decentralized SCG. Also the ACh content in nodose ganglion (NDG) was investigated in situations in which there was anastomosis, chronic supra, infra, or supra-/infranodose vagotomy. The ACh content of anastomosed NDG was near that of supranosdose vagotomized ganglion. The ACh content of supra-/infranodose vagotomized NDG, which can be considered the NDG itself, was as much as that of normal intact NDG. It was found that the ACh content of infranodose vagotomized NDG was increased, possibly the result of vagal efferent axonal flow or transport. The ACh content of vagal trunk with or without infranodose vagotomy was also measured. The ACh content of vagal trunk with infranodose vagotomy was smaller than that of the normal trunk, but there was still a considerable quantity of ACh. There was no significant change in wet weight of the SCG and NDG before or after the operations. From these results we have concluded that the transmission of the cross-anastomosed SCG (reinnervated with vagal afferent nerve) was cholinergic; and that the vagal afferent nerve have afferent cell bodies not only in NDG but also in peripheral vagal trunks (infranodose portion). These results strongly suggest that vagal afferent fibers are in part cholinergic.