The adrenergic nervous control of fluid transport in the small intestine of normotensive and spontaneously hypertensive rats

Intestinal net fluid transport in normotensive Wistar Kyoto rats (WKR) and spontaneously hypertensive rats of the Okamoto strain (SHR) were studied during 'rest', during electrical stimulation of the regional sympathetic fibres as well as after acute denervation and alpha-adrenergic receptor blockade (phentolamine). During 'rest' no statistically significant difference in fluid transport rate could be demonstrated between WKR and SHR. Cutting the left splanchnic nerve, severing the periarterial nerves or giving phentolamine turned net fluid absorption to net fluid secretion in most SHR, whereas fluid absorption was little influenced in WKR by these procedures. Stimulating the left splanchnic nerve (2, 4, 8 Hz) markedly increased net fluid uptake or decreased net fluid secretion in SHR in a frequency-dependent manner. A small effect was seen in WKR at a stimulation rate of 4 Hz. The 'spontaneous' fluid secretion in denervated intestinal segments of SHR was accompanied by a net chloride secretion. Giving hexamethonium i.v. turned net fluid and chloride secretion into water and ion absorption, suggesting that the secretion was evoked by secretory nervous pathways in the enteric nervous system. It is concluded that the 'spontaneous' fluid and electrolyte secretion seen in denervated intestines of SHR is normally 'concealed' by an augmented rate of firing in the regional adrenergic nerve fibres controlling fluid and electrolyte transport. The possible importance of the 'spontaneous' intestinal secretion in SHR in the pathophysiology of arterial hypertension is tentatively discussed.     

Neural control of electrogenic transport in the rat jejunum; interaction between intramural and adrenergic mechanisms

The aim of the study was to test the hypothesis that extrinsic adrenergic neurons modulate electrogenic ion transport in the jejunum by inhibition of intramural secretory nerve activity. Isolated pieces of rat jejunum were mounted in the Ussing chamber. Noradrenaline (NA) was released from sympathetic nerve endings in the tissue by electrical stimulation of the mesenteric nerves or by the indirect sympathomimetic agent tyramine. Intramural neurons were activated by electrical field stimulation (EFS). Mesenteric nerve stimulation induced a frequency-related decrease in the transepithelial potential difference (PD) and short circuit current (SCC), effects which were abolished by the alpha-adrenergic antagonist phentolamine. A similar response was induced by the serosal application of tyramine and by exogenous NA. Tetrodotoxin (TTX), a nerve conduction inhibitor, had no significant effect on the response to NA, but attenuated the response to tyramine. Both NA and tyramine markedly attenuated the secretory response to EFS, but there was no significant effect of catecholamines on the secretory response to serosal theophylline. The results suggest at least two mechanisms behind the adrenergic modulation of electrogenic chloride secretion: (I) a ganglionic or presynaptic modulation of secretory nerve activity via innervated alpha-receptors, and (2) a modulation at the level of the enterocytes via non-innervated alpha-receptors. The latter effect does not seem to be mediated by the cAMP system.     

On the mode of action of the sympathetic fibres on intestinal fluid transport: Evidence for the existence of a glucose-stimulated secretory nervous pathway in the intestinal wall

The effect of sympathetic nerve stimulation or close i. a.infusion of noradrenaline on net fluid transport was investigated on anesthetized cats. In the presence of glucose in the solution perfusing the intestinal lumen the adrenergic mechanisms increased net fluid absorption in normal intestines. Substituting glucose with mannitol in the perfusate abolished this effect of adrenergic stimulation on the net fluid uptake. Furthermore, the effect of noradrenaline on net fluid transport in normal or choleraic intestines was abolished by tetrodotoxin (TTX), a nerve conductivity blocking agent. This suggests that the sympathetic influence is dependent on intraluminal glucose and that noradrenaline exerts its effect mainly via nerves. TTX significantly increased fluid uptake from normal intestines perfused with an isotonic electrolyte solution containing glucose while a considerably smaller effect was seen in intestinal segments perfused with a solution with mannitol. Based on these findings it is proposed that glucose in some way activates intramural nervous reflex(es) in the intestinal wall. According to this hypothesis the enhancement of fluid transport induced by adrenergic stimuli is explained by an inhibition of the glucose-activated nervous secretion     

Interaction between sympathetic nerve fibres and epithelial transport in the rat jejunal mucosa in vivo

The aim of this study was to elucidate the mechanisms behind the absorptive response to mesenteric nerve stimulation (MNS). Segments of rat jejunum were mounted with intact blood circulation in an 'in-vivo chamber', allowing parallel registration of net fluid transport rate (NFT) and electrical parameters of the tissue. All segments secreted fluid spontaneously, and the secretion rate was inversely correlated to the corresponding potential difference (PD) across the segment. The secretion was not inhibited by atropine. The nicotinic antagonist hexamethonium (Hx) inhibited the secretion and changed the relationship between PD and NFT. MNS reversed the secretion into an absorption (absorptive response) and decreased PD and short-circuit current (SCC) (electrogenic response). Total tissue resistance (Rt) was not significantly changed. The absorptive response increased at high control secretion rates and the electrogenic response increased at high control PD levels. The electrogenic response was inversely related to the absorptive response; however, the presence of Hx reversed this correlation. Guanethidine, a sympatholytic agent, eliminated all effects of MNS. Phentolamine eliminated the absorptive response but not the electrogenic response, and the remaining electrogenic response was not sensitive to propranolol (a beta-adrenergic antagonist). Hx induced similar but smaller effects than MNS, and after Hx treatment the MNS effects were reduced. We conclude: (1) Fluid secretion in our model involves a mechanism other than electrogenic Cl- secretion. (2) MNS activates sympathetic nerve fibres, which enhance fluid absorption and decrease PD and SCC. (3) The absorptive and electrogenic responses involve partly separate transport mechanisms. (4) Both responses are partly due to the inhibition of intramural secretory nervous pathways. (5) Some sympathetic nerve fibres interact directly with the epithelium.     

Afferent vagal control of fluid absorption in the feline jejunum

The aim of the study was to test experimentally whether vagal afferent pathways are involved in the reflex regulation of jejunal fluid absorption. Acute bilateral cervical vagotomy led to an increase in net jejunal fluid absorption rate, an effect which was abolished by previous division of the splanchnic nerves. Selective division of the right cardiac branch of the vagal nerve induced an increase in fluid absorption similar to that elicited by truncal cervical vagotomy. Afferent stimulation of the right cardiac nerve at frequencies within the physiological firing range for unmyelinated C-fibre afferents induced an inhibition of net fluid absorption. Based on these findings, we propose a reflex pathway containing a non-myelinated vagal afferent branch originating from cardiopulmonary receptor endings, and an efferent sympathetic branch reaching the jejunum via the splanchnic nerves. Such a pathway might be of physiological importance in extracellular volume control by regulating the rate of fluid transport across the intestinal mucosa.     

Sympathetic nerve activation decreases the release of vasoactive intestinal polypeptide from the feline intestine

The splanchnic nerves to the small intestine were stimulated in anaesthetized cats. Activation of the sympathetic nerves caused vasoconstriction, increased net fluid absorption and decreased release of vasoactive intestinal polypeptide (VIP) in the small intestine. In the colon, parasympathetic nerve stimulation elicited vasodilatation and increased release of VIP. Additional stimulation of the sympathetic lumbar colonic nerves decreased the colonic blood flow and inhibited the release of VIP. These effects of the stimulation of the lumbar colonic nerves were blocked by phentolamine. It is concluded that, in the feline intestine, sympathetic nerve stimulation presynaptically decreased the release of VIP via an alpha-adrenergic mechanism.     

α2-Adrenoceptors inhibit antidiuretic hormone-stimulated Na+ absorption across tight epithelia (Rana esculenta)

In the present study we examined the effect of alpha-adrenergic regulation of active transepithelial Na+ absorption across the isolated frog skin epithelium. alpha-Adrenergic stimulation was achieved by addition of the adrenergic agonist noradrenaline in the presence of the beta-adrenergic blocker propranolol. alpha-Adrenergic stimulation inhibited basal as well as antidiuretic hormone (ADH)-stimulated Na+ transport. The ADH-induced increase in Na+ transport was accompanied by a membrane depolarisation due to an increase in the apical Na+ permeability. The subsequent application of noradrenaline inhibited the Na+ transport and repolarised the membrane potential, suggesting that alpha-adrenergic stimulation had reduced the apical Na+ permeability. The inhibition was abolished by the alpha2-adrenergic antagonist yohimbine whereas it was insensitive to the alpha1-adrenergic antagonist prazosin. alpha-Adrenergic stimulation had no effect on the cytosolic free [Ca2+] ([Ca2+]i). Incubation of the epithelium in the presence of ADH increased the cellular adenosine 3',5'-cyclic monophosphate (cAMP) content, an increase which was abolished by alpha-adrenergic activation. The effect of alpha-adrenergic stimulation on cAMP production was abolished by the alpha2-adrenergic antagonist yohimbine. We conclude that the noradrenaline-induced inhibition of the ADH-stimulated Na+ absorption and cAMP content is mediated by activation of alpha2-adrenoceptors. The data further indicate that the principal cells of the epithelium do not express alpha1-adrenoceptors. The noradrenaline-induced inhibition of the ADH-stimulated Na+ transport was concentration dependent, with 0.24+/-0.03 microM eliciting a half-maximal response. This alpha2-adrenergic-mediated down-regulation of Na+ absorption is achieved at a concentration of noradrenaline which begins to activate the NaCl secretion via the skin glands. The alpha2-adrenoceptors therefore appear to have considerable physiological importance.     

Adrenergic sensitivity of uninjured C-fiber nociceptors in neuropathic rats

We investigated the adrenergic sensitivity of afferent fibers in the L4 dorsal roots of rats with a unilateral ligation of the L5-L6 spinal nerves. About 12% of nociceptive fibers on the affected side were excited by sympathetic stimulation or by intra-arterial injection of norepinephrine which did not affect A beta-fiber activity. Sympathetic excitation of nociceptive fibers was suppressed by alpha 1-antagonist prazosin, while it was unaffected by alpha 2-antagonist yohimbine. Most of these fibers were excited by intra-arterial injection of alpha 1-agonist phenylephrine, without being affected by an injection of alpha 2-agonist clonidine. Sympathetic excitation was blocked by lidocaine applied near the receptive fields of recorded fibers. The results suggested that some nociceptors remaining intact after partial nerve injury become sensitive to sympathetic activity by the mediation of alpha 1-adrenoceptors in the peripheral endings.     

Sympathetic control of Na, K transport in perfused submaxillary main duct of rat.

The effect of stimulating the sympathetic innervation to rat submaxillary gland on ductal transport of Na, K, and water and on transepithelial PD was tested in the main excretory duct during perfusion through its lumen. Stimulation of the sympathetic nerve, supramaximally, caused a decrease of 30-40% in net flux of Na from, and of K to, the lumen in ducts perfused with medium containing Na and K in isotonic concentrations. Net flux of water was unaffected. Transductal PD decreased by about 30% during supramaximal stimulation. Changes in PD and net cation fluxes were reversible. These effects of supramaximal stimulation of the sympathetics on ductal transport resemble those reported to occur after large doses of isoproterenol and suggest an adrenergic secretomotor innervation to the ducts. However, changes in PD evoked by supramaximal stimulation of the sympathetic nerve could not be suppressed with propranolol, but were with phenoxybenzamine, indicating that alpha-adrenergic receptors are primarily involved in mediating at least the electrical responses of duct cells to sympathetic nerve stimulation.     

Studies on the integrated extrinsic nervous control of rectal motility in the cat

The effect of sympathetic nervous activity on rectal motility induced by pelvic nerve stimulation (PNS) was studied in anaesthetized cats. Division of the sympathetic lumbar colonic and hypogastric nerves or alpha-adrenoreceptor blockade, both of which reduced rectal tone, also reversed a predominantly relaxatory pelvic nerve response into a pure contraction. Contractions to pelvic nerve stimulation were reduced by simultaneous lumbar colonic nerve stimulation. This lumbar colonic nerve-induced inhibition was augmented by alpha-adrenoceptor blockade and abolished by beta-blockade. Close intra-arterial injection of a beta-adrenergic agonist reduced contractions to PNS, while an alpha-adrenergic agonist had no effect. Stimulation of the hypogastric nerves enhanced rectal contractions to simultaneous PNS. The apparent similarity with the arrangement of extrinsic nervous control of the internal anal sphincter suggests that the rectum is functionally involved in continence mechanisms.     

The effect of vagal nerve stimulation on net fluid transport in the small intestine of the cat

The effect of electrical vagal nerve stimulation on intestinal net fluid transport rate was studied in the small intestine of the cat. The splanchnic nerves were severed in all experiments. Absorption was quantified with a new gravimetric technique which made it possible to study fluid transport also during intestinal motility. The stimulation characteristics were varied to activate selectively low threshold fibres or low and high threshold fibres. The observations did not reveal any affects of low threshold stimulation on intestinal fluid transport whereas an inhibition was seen when also the high threshold fibres were stimulated. This inhibitory vagal mechanism could also be elicited after the administration of atropine. Atropine in itself increased “resting” net fluid absorption. The results speak against a role for vagal cholinergic mechanisms in the control of net fluid absorption. There seem, however, to be tonically active intramural cholinergic pathways and noncho-linergic inhibitory vagal neurons of unknown physiological significance.     

Effect of sympathetic nerves on cerebral vessels during seizures.

The purpose of this study was to determine the effect of activation of sympathetic pathways during seizures on cerebral blood flow and integrity of the blood-brain barrier. We measured cerebral blood flow with microspheres and disruption of the blood-brain barrier with labeled albumin in cats. One cerebral hemisphere was denervated by cutting the superior cervical sympathetic trunk on one side. During bicuculline-induced seizures, superior cervical sympathetic nerve activity increased about threefold. Blood flow to the innervated hemibrain was significantly lower than flow to denervated hemibrain. However, in relation to the total increase in flow, this effect of nerves was minor. Blood-brain barrier permeability increased about sixfold during seizures, but there was no difference between the innervated and denervated sides of the brain. We conclude that sympathetic nerves attenuate the increase in cerebral blood flow during seizures, despite the increase in metabolism, but this effect is small. Activation of sympathetic nerves does not reduce disruption of the blood-brain barrier during seizures.     

Baroreflex control of jejunal blood flow and fluid transport in cats; effects of yohimbine,an α.-adrenergic antagonist

The aim of the study was to test the hypothesis that baroreceptor unloading increases jejunal fluid absorption rate via an α₂-adrenergic effect on electrogenic active transport. In 13 chloralose-anaesthetized cats, the carotid sinus baroreceptors were isolated and perfused with arterial blood, and we studied the effects of a graded decrease in carotid sinus pressure on intestinal vascular resistance, net fluid absorption rate and the potential difference between the intestinal lumen and the peritoneal cavity (PD). Experiments were performed in seven control animals and in six animals pretreated with yohimbine, an α₂-adrenergic antagonist, at a dose of 0.1 mg kg^-I i.v. Yohimbine per se had no significant effects on systemic arterial pressure, intestinal vascular resistance, net fluid absorption rate or PD. In the control animals, baroreceptor unloading induced an increase in systemic arterial pressure, intestinal vascular resistance and net fluid absorption rate, and a decrease in the PD. Yohimbine pretreatment did not significantly affect the systemic blood pressure response to baroreceptor unloading, but abolished the effect on intestinal vascular resistance and PD. After yohimbine treatment, decreases in carotid sinus pressure still enhanced net fluid absorption rate, but this response was observed in a higher range of carotid sinus pressures than in control animals. We conclude that (1) a major component of the increase in jejunal absorption rate during baroreceptor unloading is due to a non-electrogenic mechanism, which may be either active or passive; (2) this component of the response is not blocked by yohimbine at a dose sufficient for an effect on presynaptic α₂-receptors; (3) the absorptive response to baroreceptor unloading is not a phenomenon secondary to the concomitant jejunal vasoconstriction.     

Effect of Sympathetic Nerve Stimulation on Net Transvascular Movement of Fluid in Canine Adipose Tissue

Net transvascular movement of fluid has been studied in the isolated, autoperfused subcutaneous adipose tissue of the dog, during and after sympathetic nerve stimulation (1–15 Hz) and during infusion of 50% glucose i.a. Net fluid movement was calculated as the difference between change in tissue volume and change in blood volume. Tissue volume was measured by plethysmography and blood volume by external monitoring of circulating ¹³¹I-albumin. No net fluid movement of statistical significance was found during or after nerve stimulation except during the first minute of stimulation at 15 Hz when a small net absorption (p<0.05) was obtained. In contrast, infusion of glucose at 25–75 mOsm/kg H₂O produced a dose-dependent net absorption lasting several minutes, amounting maximally to 0.30 ml × min^-1× 100 g^-1. The absence of prolonged net absorption in subcutaneous adipose tissue during nerve stimulation as well as the absence of net filtration after stimulation may be explained by an essentially unaltered mean hydrostatic capillary pressure. The results indicate that adipose tissue does not contribute to the fluid homeostasis of the body via sympathetic resetting of the pre-postcapillary resistance ratio. Thus, mobilisation of fluid from the nterstitial space in adipose tissue into the blood does not seem to occur by nerve activity.     

Evidence for sympathetic regulation of the hydraulic conductance of rat jejunal mucosa in vivo

The aim of the study was to test the hypothesis that sympathetic nerve stimulation inhibits jejunal fluid secretion by decreasing the hydraulic conductance (L_p)of the mucosa. Jejunal segments of anaesthetized rats were mounted in a specially constructed chamber which makes it possible to measure net fluid transport together with transmural potential difference or short circuit current in vivo. L_p was estimated by measuring the effects on net fluid transport rate elicited by stepwise changes in luminal pressure. L_p was measured in four groups of animals: (1) under control conditions; (2) after hexamethonium administration, to block intramural secretory neurons; (3) during efferent stimulation of the mesenteric nerves, to activate sympathetic neurons, and (4) during mesenteric nerve stimulation (MNS) after pre-treatment with phentolamine, an a-adrenergic antagonist, to block the sympathetic effector mechanism. L_p measured at positive luminal pressures in control animals was approximately 4 μl min^-1 cmH₂O^-1 100 cm^-2 serosal surface area. When measured at negative luminal pressures, this value increased approximately threefold. Hexamethonium did not significantly influence L_p. At positive luminal pressures, MNS decreased L_p to a value not significantly different from zero. This effect was abolished by phentolamine. At negative lumina] pressures, MNS decreased L_p by approximately 50% and this effect was partly antagonized by phentolamine. The results suggest that a decrease in mucosal hydraulic conductance may contribute to the antisecretory effect of sympathetic nerve stimulation in vivo.     

Blood flow changes in the submaxillary gland of the rat on parasympathetic and sympathetic nerve stimulation.

Blood flow and secretion from the submaxillary gland of the rat was recorded. Stimulation of the chordalingual nerve evoked vasodilatation which was little, it at all, decreased after administration of atropine. The secretion evoked on parasympathetic nerve stimulation was greatly reduced but not totally abolished by parasympatholytic agents. Stimulation of the cervical sympathetic trunk evoked secretion, which was mediated both via alpha and beta-adrenoceptors and vasoconstriction which was abolished by dihydroergotamine, and thus seemed to be mediated via alpha-adrenergic receptors. After alpha-adrenoceptor blockade vasodilatation was seen on sympathetic nerve stimulation. This was abolished by the beta-adrenoceptor blocking agent propranolol.     

On the mechanisms of the basal alkaline secretion in the rat ileum in vivo.

Basal alkaline secretion of the denervated rat ileum was monitored by a pH-stat method. Changes of transepithelial electrical potential difference (PD) were also continuously registered. In other experiments net fluid transport was measured with a gravimetric method. The importance of the enteric nervous system for the recorded variables was investigated by giving i.v. hexamethonium, neuropeptide Y (NPY) or methionine-enkephalin or by stimulating electrically the mesenteric nerves surrounding the superior mesenteric artery. Alkaline secretion was inhibited by about 20% by mesenteric nerve stimulation or by neuropeptide Y (NPY) or met-enkephalin i.v. A somewhat greater inhibition (approximately 30%) of transepithelial electrical potential difference was elicited by the mesenteric nerve stimulation and NPY whereas met-enkephalin did not cause any transepithelial electrical potential difference change. Net fluid absorption was markedly diminished (by approximately 65-75%) by met-enkephalin but not by NPY. The cellular mechanisms underlying alkaline secretion were investigated by means of amiloride, SITS and acetazolamide. The basal alkaline secretion and transepithelial electrical potential difference were not influenced by 10(-3)M or 10(-4) SITS. In contrast 10(-3) M amiloride caused a significant increase of alkaline secretion but not of transepithelial electrical potential difference. A 35% reduction in the alkaline secretion but not transepithelial electrical potential difference was observed after acetazolamide had been given intravenously. A similar decrease was observed after giving hexamethonium. We conclude: (1) Enteric nerves are of comparatively small importance in controlling the ileal alkaline secretion recorded during basal conditions; (2) About 35% of the basal ileal alkaline secretion is carbonic anhydrase dependent. This mechanism is not influenced by nicotinic receptor blockade; (3) Under the present experimental conditions there may be an alkaline secretion which is concealed by a simultaneously operating Na+/H+ exchanger and; (4) No consistent quantitative correlation exists between alkaline secretion, transepithelial electrical potential difference and net field transport in the denervated rat ileum.     

The effect of vasodilatation and sympathetic nerve activation on net water absorption in the cat's small intestine

The rate of net water uptake from the feline small intestine has been investigated during control conditions, during graded infusions of the vasodilator drug isopropylnoradrenaline, and during electrical stimulation of the regional sympathetic nerve fibres to the gut. Net water absorption rate was largely unaffected by intestinal vasodilatation. The fraction of the absorbate transported via the lymphatics remained also constant at 20–40% of the total absorption regardless of blood flow rate. Stimulating the sympathetic nerve fibres to the small intestine increased, however, net water absorption rate. The increase was particularly pronounced when blood pressure was kept constant during the period of stimulation. The absorption rate was on an average almost doubled at a stimulation frequency of 8 Hz during constant pressure conditions. The mechanism(s) explaining this nervous control of water absorption are tentatively discussed.     

The effect of vasodilatation and sympathetic nerve activation on net water absorption in the cat's small intestine.

The rate of net water uptake from the feline small intestine has been investigated during control conditions, during graded infusions of the vasodilator drug isopropylnoradrenaline, and during electrical stimulation of the regional sympathetic nerve fibres to the gut. Net water absorption rate was largely unaffected by intestinal vasodilatation. The fraction of the absorbate transported via the lymphatics remained also constrant at 20-40% of the total absorption regardless of blood flow rate. Stimulating the sympathetic nerve fibres to the small intestine increased, however, net water absorption rate. The increase was particularly pronounced when blood pressure was kept constant during the period of stimulation. The absorption rate was on an average almost doubled at a stimulation frequency of 8 Hz during constant pressure conditions. The mechanism(s) explaining this nervous control of water absorption are tentatively discussed.     

On the mechanisms of the basal alkaline secretion in the rat ileum in vivo

Basal alkaline secretion of the denervated rat ileum was monitored by a pH-stat method. Changes of transepithelial electrical potential difference (PD) were also continuously registered. In other experiments net fluid transport was measured with a gravimetric method. The importance of the enteric nervous system for the recorded variables was investigated by giving i.v. hexamethonium, neuropeptide Y (NPY) or methionine-enkephalin or by stimulating electrically the mesenteric nerves surrounding the superior mesenteric artery. Alkaline secretion was inhibited by about 20% by mesenteric nerve stimulation or by neuropeptide Y (NPY) or met-enkephalin i.v. A somewhat greater inhibition (～ 30%) of transepithelial electrical potential difference was elicited by the mesenteric nerve stimulation and NPY whereas met-enkephalin did not cause any transepithelial electrical potential difference change. Net fluid absorption was markedly diminished (by ～ 65–75%) by met-enkephalin but not by NPY. The cellular mechanisms underlying alkaline secretion were investigated by means of amiloride, SITS and acetazolamide. The basal alkaline secretion and transepithelial electrical potential difference were not influenced by 10^-3 M or 10^-4 SITS. In contrast 10^-3 m amiloride caused a significant increase of alkaline secretion but not of transepithelial electrical potential difference. A 35% reduction in the alkaline secretion but not transepithelial electrical potential difference was observed after acetazolamide had been given intravenously. A similar decrease was observed after giving hexamethonium. We conclude: (1) Enteric nerves are of comparatively small importance in controlling the ileal alkaline secretion recorded during basal conditions; (2) About 35% of the basal ileal alkaline secretion is carbonic anhydrase dependent. This mechanism is not influenced by nicotinic receptor blockade; (3) Under the present experimental conditions there may be an alkaline secretion which is concealed by a simultaneously operating Na⁺/H⁺ exchanger and; (4) No consistent quantitative correlation exists between alkaline secretion, transepithelial electrical potential difference and net fluid transport in the denervated rat ileum.