Transport of serotonin from the rat jejunal lumen into mesenteric veins in vivo.

The transport of labelled (hot) and non-labelled (cold) serotonin (5-HT) into the mesenteric venous circulation was studied after instillation of test solutions into an isolated jejunal loop of anaesthetized rats. After instillation of [3H]H2O and [14C]5-HT there was an almost parallel appearance of the isotopes in mesenteric venous blood. After instillation of 5-HT a marked early increase of the total amounts of cold 5-HT was observed in mesenteric veins compared with animals instilled with saline only. In a third type of experiment the label was detected in mesenteric venous whole blood after instillation of [3H]5-HT into the gut lumen. After hydrolysis of blood cells and protein precipitation the samples were fractionated and determined for 5-HT and metabolites. Only 5-HT was detected in these fractions. The label was present within 5-HT peaks in three out of eight animals. The experiments indicate rapid transport of 5-HT (or metabolites) across the rat jejunal mucosa. These substances may be bound to a binding protein in platelets since the isotope was detected in whole blood but more seldom in supernatants after hydrolysis and precipitation.

A possible mechanism for the release of serotonin from the gut caused by pentagastrin.

Pentagastrin (PG) is a potent agent causing release of serotonin (5-HT) from patients with carcinoid tumors. The physiological release of 5-HT from gut enterochromaffin cells is controlled by beta-adrenoceptors. Studies on carcinoid tumor cell suspensions, acute or in culture, have shown that catecholamines (CA), but not PG, release 5-HT, thus indicating an indirect mode of action by the peptide. In this study the mechanism for release of 5-HT from the gut induced by PG was investigated in animal models. The test protocol for patients was used in anesthetized cats. Portal blood samples were drawn after PG injection (0.6 microgram/kg iv), which resulted in significantly increased levels of 5-HT at 3 and 5 min postinjection. The PG-induced release was totally inhibited after blockade of beta-adrenoceptors (propranolol) or of slow calcium channels (verapamil) as well as after adrenalectomy. Blockade of beta-adrenoceptors or slow calcium channels decreased the basal levels of 5-HT, while adrenalectomy caused no change. In separate experiments CA were measured after PG injection in caval blood, drawn at the level of the adrenal veins. There was a significant increase in the levels of dopamine and epinephrine postinjection, while the levels of norepinephrine were stable. The changes of CA levels after PG injection could be prevented by adrenalectomy. These results further suggest an indirect action of PG in the release of 5-HT from the feline gut via activation of beta-adrenoceptors by epinephrine released from the adrenals.

Vagal release of serotonin into gut lumen and portal circulation via separate control mechanisms.

The mechanisms controlling vagally induced release of serotonin-like immunoreactivity (5-HTLI) into portal circulation and jejunal lumen were studied in individual cats. In control animals, electrical vagal nerve stimulation significantly enhanced both the endoluminal secretion rate of 5-HTLI and the release of 5-HTLI into the portal vein. The vagally induced release of 5-HTLI into portal circulation was blocked by pretreatment with propranolol or phenoxybenzamine, or by previous removal of the superior cervical ganglia, but was not blocked by atropine or hexamethonium. On the contrary, the luminal secretion of 5-HTLI after vagal stimulation was not blocked by adrenoceptor blocking agents or ganglionectomy, but instead was inhibited by cholinoceptor antagonists. Thus, in the same experimental animals it was shown that vagally induced release of 5-HTLI into portal circulation was mediated by adrenoceptor mechanisms, while endoluminal release of 5-HTLI was regulated via cholinoceptors. Based on indirect estimations, the apical release of 5-HT seems to be quantitatively small in comparison with the release into portal circulation.

Vagal release of serotonin into gut lumen and portal circulation via separate control mechanisms.

The mechanisms controlling vagally induced serotonin-like immunoreactivity (5-HTLI) release into portal circulation and jejunal lumen were studied in individual cats. In control animals, electrical vagal nerve stimulation significantly enhanced both the endoluminal secretion rate of 5-HTLI and the release of 5-HTLI to the portal vein. The vagally induced release of 5-HTLI to the portal circulation was blocked by pretreatment with propranolol or phenoxybenzamine, or by prior removal of the superior cervical ganglia, but was not blocked by atropine or hexamethonium. On the contrary, the luminal secretion of 5-HTLI after vagal stimulation was not blocked by adrenoceptor blocking agents or ganglionectomy, but instead was inhibited by cholinoceptor antagonists. Thus, in the same experimental animal it was shown that vagally induced release of 5-HTLI to the portal circulation was mediated by adrenoceptor mechanisms, while the luminal release of 5-HTLI was regulated via cholinoceptors. Based on indirect estimations, the apical release of 5-HT seems to be qualitatively small in comparison with the release into the portal circulation.

Adrenergic control of serotonin release from carcinoid tumor cells in vitro and in vivo.

Serotonin (5-HT)-producing human carcinoid tumors of midgut origin were transplanted to the anterior eye chamber of cyclosporine-treated rats. The release of 5-HT from in oculo transplants was studied after stimulation with adrenoceptor agonists applied locally to the eye. Chamber fluid was collected by micropuncture of the eye. 5-HT levels were determined by liquid chromatography with electrochemical detection. The release of 5-HT in cell suspensions of the same tumors was similarly studied after incubation with adrenoceptor agonists. In both experimental models activation of adrenoceptors caused release of 5-HT from carcinoid tumor cells. Individual variations in the type of response to adrenoceptor stimulation could be demonstrated for the different tumors. In two tumors there was good agreement between in vitro and in vivo findings with release of 5-HT at selective activation of beta-adrenoceptors with isoprenaline. In a third tumor there was a release of 5-HT at activation of alpha-adrenoceptors with norepinephrine in vitro. However, 5-HT release from this tumor in vivo was demonstrated at activation of beta-adrenoceptors. This finding may reflect different adrenoceptor populations on the tumor cell surface, each population activated to various degrees in the in vivo and in vitro situation.

Studies on the mucosal hyperaemia of the feline small intestine observed at endoluminal perfusion with substance P.

Substance P-like immunoreactivity (SPLI) was found in the luminal contents of the feline small intestine. A physiological role for endoluminal substance P (SP) was explored in in vivo experiments, where feline jejunal segments were endoluminally perfused with saline or a low concentration of SP. A jejunal segment was initially perfused with saline, followed by SP in saline, and finally perfused with saline alone. The regional blood flow to the small intestine was determined by the microsphere technique. Endoluminal perfusion with SP caused a selective mucosal/submucosal hyperaemia of the experimental jejunum, which was normalized upon subsequent saline perfusion. Since neither acute vagotomy, local neural blockade (TTX, lidocaine), adrenergic blockade (phenoxybenzamine, propranolol), cholinergic blockade (hexamethonium, atropine), histamine - nor prostaglandin blockade - could antagonize the hyperaemic effects of luminally administered SP, it is suggested that this effect is directly paracrine in nature. Substance P may be released from SP nerves on physiological stimulation, that is, augmented vagal tone by a meal, and act at the effector cell, that is, vascular smooth muscle.

Verapamil and diarrhoea in the carcinoid syndrome--clinical and experimental observations on serotonin release.

A patient with the midgut carcinoid syndrome with severe diarrhoea and proven hypersecretion of serotonin (5-HT) was treated with low doses of verapamil perorally. During treatment the patient was completely relieved of diarrhoea but discrete facial flushing persisted during treatment. When treatment was cessated, diarrhoeas recurred. This patient underwent pentagastrin (PG) provocation repeatedly; during untreated conditions injection of PG released 5-HT, detectable in peripheral venous blood. Such release was abolished during verapamil treatment, but recurred after withdrawal of the drug. Surgical biopsies from this tumour were studied in two experimental models: cell suspensions and heterotransplants grown in the anterior eye-chamber of immunosuppressed rats. Release of 5-HT from the cell suspensions was elicited in a dose-dependent manner after stimulation with isoprenaline (IP) suggesting activation of beta-adrenoceptors on the tumour cells. Such release was reduced after pretreatment with verapamil indicating a calcium dependent mechanism. Intraocular tumour transplants also responded with release of 5-HT into the chamber fluid after conjunctival application of IP. However, pretreatment of the rats with verapamil significantly reduced the IP-stimulated release of 5-HT.

Regional and selective changes in blood flow of the feline small intestine induced by endoluminal serotonin.

A physiological role for endoluminal serotonin (5-HT) was explored in in vivo experiments, where feline jejunal segments were endoluminally perfused with saline or a low concentration of 5-HT. The upper jejunal segment was initially perfused with saline, followed by 5-HT in saline, and finally perfused with saline alone. The lower jejunal segment served as a control with constant perfusion with saline. The regional blood flow to the small intestine was determined by the microsphere technique. Endoluminal perfusion with 5-HT caused a selective muscular hyperaemia of the experimental gut segment, which was normalised upon subsequent saline perfusion. The blood flow to the saline perfused control segment was unchanged. This was also the case for control tissues; that is, adjacent small intestinal regions and kidneys. The 5-HT-induced muscular hyperaemia was confined to the experimental gut segment and seems to be mediated by a local cholinergic neural mechanism, activated from the mucosa, since the hyperaemic response was prevented by muscarinic blockade or local anaesthesia applied to the luminal surface of the experimental segment. The vascular response does not seem to involve 5-HT2 receptors, since selective blockade of such receptors did not prevent 5-HT-induced hyperaemia.

The influence of cholinergic and adrenergic blockade on the portal release of substance P in the cat.

The effect of efferent electrical stimulation of the divided thoracic vagus nerves on the release of substance P-like immunoreactivity (SPLI) into the portal vein was studied in anesthetized cats. Under basal conditions portal SPLI levels were fairly constant. During the first 5 min of vagal stimulation there was a significant increase of the portal plasma levels of SPLI. Since we have demonstrated that infracardiac vagal stimulation causes only minor changes in small intestinal bloodflow, a true vagal neurogenic release of SPLI seems likely. We have shown that SPLI is exclusively localized within feline gut neurons (not the enterochromaffin (EC) cells). Thus, SPLI released into the portal circulation probably represents overflow of transmitter into the circulation. Neither blockade of cholino- or adrenoceptors nor ganglionic blockade influenced the enhanced release of SPLI induced by vagal stimulation, suggesting activation of SP neurons via nonclassical receptors.

Adrenergic control of serotonin release from a midgut carcinoid tumour.

A case of recurrent midgut carcinoid tumour with disseminated spread is described, in which the clinical diagnosis was supported by measurements of elevated basal serotonin (5-HT) levels in peripheral blood and increased 5-HT responses to pentagastrin provocation, despite normal urinary levels of 5-hydroxyindoleacetic acid. Preoperative diagnosis was obtained by concomitant determinations of 5-HT in mesenteric and hepatic veins. The carcinoid tumour was studied immunocytochemically using antisera to tyrosine hydroxylase and 5-HT. Neuroendocrine complexes between adrenergic nerve terminals and 5-HT-containing tumour cells could be demonstrated. 5-HT release from tumour cells in suspension was studied in vitro after incubation with adrenoceptor agonists or pentagastrin. Tissue pieces from the tumour were also transplanted into the anterior eye chamber of Sprague-Dawley rats, some of which were subjected to immunosuppression (Cyclosporin A 20 mg/kg s.c.). After 10 days in oculo the tumour transplants (with preserved immunocytochemical characteristics) were stimulated with adrenoceptor agonists. Tumour cells in suspension as well as tumour transplants released 5-HT upon adrenoceptor stimulation but no release was induced by pentagastrin. The pentagastrin test is suggested to cause release of 5-HT in carcinoid tumour patients via release of endogenous catecholamines, in turn activating adrenoreceptors on carcinoid tumour cells.

On the mode of action of the pentagastrin test in the carcinoid syndrome.

Provocation with pentagastrin (PG) (0.6 micrograms/kg intravenously) causes a release of serotonin (5-hydroxytryptamine (5-HT] in excess of the metabolizing capacity associated with carcinoid symptoms and a moderate fall in systemic arterial blood pressure in patients with midgut carcinoids and hepatic metastases. In this study PG also caused a release of 5-HT into portal circulation of anesthetized cats with stable peripheral levels of 5-HT in whole blood, indicating an effective hepatic metabolization. A similar response was obtained in the same animal when the PG test was repeated after 3 h. PG provocation was also performed in animals before and after adrenalectomy. Adrenalectomy seems to prevent the PG-induced release of 5-HT into portal circulation, indicating involvement of an adrenal mechanism. PG does not induce 5-HT release from cell suspensions of midgut carcinoid tumors, but such release was induced by incubation with adrenoceptor agonists. These findings indicate the presence of adrenoceptors on carcinoid tumor cells. The mode of action of the PG test may therefore be activation of such adrenoceptors by catecholamines, released from the adrenal medulla at the fall of arterial blood pressure at PG provocation.

The effects of vagal nerve stimulation on endoluminal release of serotonin and substance P into the feline small intestine.

Endoluminal release of serotonin (5-HT), substance P (SP), and motilin was quantitated after thoracic vagal nerve stimulation in the cat. In duodenum and jejunum, simultaneous release of these compounds was observed. In contrast, vagal stimulation did not augment the rate of luminal secretion of either 5-HT or SP in the distal ileum. Immunohistochemical studies demonstrated 5-HT in both enterochromaffin (EC) cells and nerves throughout the small bowel. However, we were unable to visualize any SP-containing EC cells in the cat, which suggests that the source of luminal SP in this species must be intramural nerves.

Endoluminal substance P as a cause of mucosal hyperemia in the feline gut by a nonneural mechanism.

Exogenous substance P (SP) was perfused into the lumen of the proximal jejunum in 16 fasted cats at a rate of 6.04 +/- 1.96 ng/min. Regional blood flow was measured by radioactive microspheres after successive 15-minute perfusions with saline (basal), neural blockers, SP, and saline. Endoluminal SP caused mucosal hyperemia in the perfused jejunal segment that was not blocked by preperfusion of the segment with lidocaine (2% or 4%) or tetrodotoxin (3 X 10(-7) M). In addition, there was no increase in blood flow to any other organ tested, gastrointestinal or systemic. Local luminal SP levels rose but there was no significant rise in portal SP levels, indicating that the peptide was not absorbed. The data suggest that SP acts locally by a nonneural mechanism to influence local gastrointestinal blood flow.