GABAA receptors expression pattern in rat brain following low pressure distension of the stomach.

It is known that gastric mechanoreceptor stimuli are widely integrated into neuronal circuits that involve visceral nuclei of hindbrain as well as several central brain areas. GABAergic neurons are widely represented in hindbrain nuclei controlling gastric motor functions, but limited information is available specifically about GABA(A)-responding neurons in brain visceral areas. The present investigation was designed to determine the central sensory neuronal pathways and their GABA(A)-alpha1 and -alpha3 receptor presenting neurons that respond to gastric mechanoreceptor stimulation within the entire rat brain. Low pressure gastric distension was used to deliver physiological mechanical stimuli in anesthetized rats, and different protocols of gastric distension were performed to mimic different stimulation patterns with and without sectioning vagal and/or splanchnic afferent nerves. Mapping of activated neurons was investigated using double colorimetric immunohistochemistry for GABA(A)-alpha1 or -alpha3 subunits and c-Fos. Following stomach distension, neurons expressing GABA(A) receptors with alpha1 or alpha3 subunits were detected. Low frequency gastric distension induced c-Fos expression in nucleus tractus solitarii (NTS) only, whereas in the high frequency gastric distension c-Fos positive nuclei were found in lateral reticular nucleus and in NTS in addition to some forebrain areas. In contrast, during the tonic-rapid gastric distension the neuronal activation was found in hindbrain, midbrain and forebrain areas. Moreover different protocols of gastric stimulation activated diverse patterns of neurons presenting GABA(A)-alpha1 or -alpha3 receptors within responding brain nuclei, which may indicate a probable functional significance of differential expression of GABA(A)-responding neurons. The same protocol of gastric distension performed in vagotomized rats has confirmed the primary role of the vagus in the response of activation of gastric brain areas, whereas neuronal input of splanchnic origins was shown to play an important role in modulating the mechanogastric response of brain areas.

Effect of distension of the gallbladder on plasma renin activity in anesthetized pigs.

BACKGROUND: Gallbladder pathology has been associated with cardiovascular disease. Recently, we showed that gallbladder distension in anesthetized pigs reflexly increased heart rate, arterial pressure, and coronary and renal vascular resistance through efferent sympathetic mechanisms. Renin release is affected by sympathetic output, and angiotensin liberation may result in vasoconstriction. This study was undertaken to determine whether gallbladder distension primarily causes a reflex change in plasma renin activity (PRA) and to assess its influence on observed pressor and coronary responses as well as on regional vascular resistance. METHODS AND RESULTS: In 34 alpha-chloralose-anesthetized pigs, balloons positioned within the gallbladder were distended for 30 minutes with volumes of Ringer's solution equal to those of withdrawn bile. In 19 pigs, gallbladder distension at constant heart rate, arterial pressure, and renal flow increased PRA in the absence of changes in urinary sodium excretion. This increase was abolished by cervical vagotomy, section of renal nerves, or blockade of beta-adrenergic receptors. In another 15 pigs, blockade of angiotensin II receptors significantly attenuated the pressor and coronary, mesenteric, and iliac vasoconstriction responses to gallbladder distension. CONCLUSIONS: The present study showed that innocuous gallbladder distension primarily caused a reflex increase in PRA. This increase, which involved afferent vagal pathways and efferent sympathetic mechanisms related to beta-adrenergic receptors, contributed significantly to the pressor and coronary, mesenteric, and iliac vasoconstriction responses to gallbladder distension.

Role of nitric oxide in the control of the heart rate within the nucleus ambiguus of rats.

The aim of this study was to determine whether NO plays a role in the control of heart rate (HR) within the nucleus ambiguus (NA). Experiments were performed in 29 male Wistar rats anaesthetized with urethane. Microinjections of the NO-donor sodium nitroprusside (SNP; 5 mmol) as well as of L-arginine (L-arg; 50 mmol) into functionally identified cardioinhibitory sites within the NA significantly decreased HR (-57.7 +/- 8.4 and -53.8 +/- 3.2 bpm, respectively), whereas the NO-synthase inhibitor N(omega)-nitro-L-arginine methyl ester (L-NAME) significantly increased HR (+40 +/- 2.7 bpm). Bilateral vagotomy and i.v. injection of atropine (0.5mg/kg) always abolished the HR decrease induced by SNP and L-arg, whereas propranolol did not affect the HR responses. These results demonstrated that NO mechanisms within the NA play a role in the parasympathetic control of the HR.

Hemodynamic effects of the intravenous administration of cyclovirobuxine D [correction of cyclorirobuxine D] in anesthetized pigs.

The present study was undertaken in anesthetized pigs to determine the primary effects of cyclovirobuxine D [corrected] given intravenously on hemodynamic variables. In eight pigs, the administration of 1.5 mg/kg of cyclovirobuxine D [corrected] caused a small increase in aortic blood pressure. When this response was prevented, a decrease in heart rate was obtained in each of the eight pigs. When this response was also prevented, an increase in the maximum rate of change of left ventricular systolic pressure (left ventricular dP/dtmax) was observed. In four pigs, the decrease in heart rate and the increase in left ventricular dP/dtmax were progressively augmented by graded increases in the dose of cyclovirobuxine D [corrected]. In six pigs, the responses of hemodynamic variables to cyclovirobuxine D [corrected] were not affected by blockade of cholinergic and adrenergic receptors. In a further six pigs, blockade of nitric oxide synthase with N omega-nitro-L-arginine methyl ester did not affect the decrease in heart rate caused by the drug, but abolished the increases in left ventricular dP/dtmax and aortic blood pressure. The present study showed that intravenous administration of cyclovirobuxine D [corrected] primarily caused a decrease in heart rate and an increase in left ventricular inotropic state, which secondarily determined an increase in aortic blood pressure, and suggested that the response of heart rate involved a direct effect of the drug on the heart, while the response of left ventricular contractility was related to mechanisms dependent on the release of nitric oxide.

The role of beta 2-adrenergic vascular receptors in the peripheral vasodilation caused by 17 beta-estradiol in anesthetized pigs.

It has been previously shown in anesthetized pigs that intravenous infusion of 2 microg/h of 17beta-estradiol primarily dilated renal, iliac and coronary circulations, while higher doses of the hormone were required to cause vasodilation also in the mesenteric vascular bed. In the same experimental model, a tonic beta2-adrenoceptor mediated vasodilation, which could be argued to attenuate the vasodilator effect of 17beta-estradiol, has been described. The present study was planned to investigate the role of beta2-adrenergic receptors in the hemodynamic responses of renal and mesenteric vascular beds to 17beta-estradiol. Changes in flow caused by intravenous infusion of 2 microg/h of the hormone at constant heart rate and aortic blood pressure in the left renal and superior mesenteric arteries were assessed using electromagnetic flowmeters. In six pigs, infusion of 17beta-estradiol caused an increase in renal blood flow, which averaged 12.1% of the control values, without affecting mesenteric blood flow. In the same pigs, after hemodynamic variables had returned to the baseline values, blockade of beta2-adrenergic receptors with butoxamine caused an increase in aortic blood pressure and an increase in renal and mesenteric resistance. The subsequent infusion of 17beta-estradiol elicited increases in renal and mesenteric blood flow which respectively averaged 19.6% and 12.8%. Therefore, the present study in anesthetized pigs have shown that the vasodilator responses of the renal and mesenteric circulations to 17beta-estradiol were attenuated and even masked by a tonic beta2-adrenoceptor mediated vasodilation. This indicates that some vasodilator effects elicited by normally used replacement doses of the hormone may not be apparent.

Coronary effects of cyclovirobuxine D in anesthetized pigs and in isolated porcine coronary arteries.

The present study was undertaken in anesthetized pigs and in isolated porcine coronary arteries to determine the primary coronary effects of cyclovirobuxine D. In six pigs, the intravenous administration of 1.5 mg/kg of cyclovirobuxine D whilst preventing changes in heart rate and aortic blood pressure caused increases in left ventricular dP/dtmax and coronary blood flow which respectively averaged 10% and 23.9%. These responses were progressively augmented by graded increases in the dose of the drug (four pigs) and were not affected by blockade of cholinergic and adrenergic receptors (five pigs). Intravenous blockade of nitric oxide synthase (L-NAME, five pigs) abolished both responses, while intracoronary injection of L-NAME (five pigs) abolished only the coronary vasodilatation. In ten isolated coronary segments, cyclovirobuxine D significantly reduced the degree of potassium chloride-induced contraction. This reduction was not affected by inhibition of cyclooxygenase with indomethacin (five segments) or potassium channels blockade with glibenclamide (five segments), but it was abolished by L-NAME (five segments) or removal of endothelium (five segments). The present study showed that cyclovirobuxine D caused a primary effect of coronary vasodilatation, which involved mechanisms related to the endothelial release of nitric oxide.

Mechanisms of the renal vasodilation caused by insulin in anesthetized pigs.

The present study was planned to determine the mechanisms involved in the renal vasodilation caused by insulin. Changes in flow caused by the intravenous infusion of 0.004 IU/kg/min of insulin at constant heart rate, aortic blood pressure, left ventricular contractility and blood levels of glucose and potassium in the left renal artery were assessed using an electromagnetic flowmeter. In ten pigs, infusion of insulin caused an increase in renal blood flow which averaged 12.8% of the control values. After hemodynamic variables had returned to control values, insulin infusion was repeated in five pigs following blockade of alpha-adrenergic receptors with injection of phentolamine into the renal artery and in the other five pigs following blockade of nitric oxide formation with injection in the same artery of Nomega-nitro-L-arginine methyl ester (L-NAME). After blockade of alpha-adrenergic receptors, insulin infusion caused an increase in renal blood flow which averaged 18.1% of the control values, being significantly enhanced with respect to the increase previously obtained in the same pigs. On the contrary, after blockade of nitric oxide formation insulin infusion caused a decrease in renal blood flow which averaged 6.5% of the control values. These responses were respectively abolished by the subsequent injection into the renal artery of L-NAME and phentolamine. The present study showed that the renal vasodilation caused by insulin in the anesthetized pig was the result of two opposite effects which involved a predominant vasodilation mediated by the release of nitric oxide from the endothelium and a sympathetic vasoconstrictor mechanism mediated by alpha-adrenergic receptors.

The role of activation of the renin-angiotensin system on the reflex regional vasoconstriction caused by distension of the uterus in anaesthetized pigs.

Distension of the uterus in anaesthetized pigs has been shown to cause a reflex regional vasoconstriction and an increase in plasma renin activity (PRA) through efferent sympathetic mechanisms which respectively involved alpha- and beta-adrenergic receptors. The present study was undertaken to determine the possible contribution of the activation of the renin-angiotensin system (RAS) to the observed regional vasoconstrictive responses to uterus distension. In pigs anaesthetized with alpha-chloralose, blood flow in the left circumflex or anterior descending coronary, superior mesenteric, left renal and left external iliac arteries was assessed using electromagnetic flowmeters. Distension of the uterus for periods of 30 min was performed by injecting 20 ml of warm Ringer solution into balloons positioned within the viscus before and after blockade of angiotensin II receptors with losartan. Changes in heart rate and renal blood flows were respectively prevented by atrial pacing and injection of phentolamine into the renal arteries. Changes in baroreceptors activity and in regional perfusion pressure were minimized by section of cervical vagus nerves and denervation of carotid sinuses and by an aortic constriction. PRA was assessed during the last minute of distension by radioimmunoassay of angiotensin 1. Before blockade of angiotensin II receptors, in six pigs, distension of the uterus decreased coronary blood flow by 19%, and in other six pigs, decreased mesenteric and iliac blood flows by 13.1% and 29.4% in the absence of changes in arterial perfusion pressure. After losartan, these decreases were significantly reduced to 11.7%, 8.2% and 18%. These results showed that the activation of the RAS significantly contributed to the alpha-adrenergic receptor-mediated regional vasoconstrictive responses reflexly elicited by distension of the uterus.

Calcium handling in porcine coronary endothelial cells by gastrin-17.

In porcine coronary artery endothelial cells (PCAEC), gastrin-17 has recently been found to increase nitric oxide (NO) production by the endothelial NO synthase (eNOS) isoform through cholecystokinin 1/2 (CCK1/2) receptors and the involvement of protein kinase A (PKA), PKC and the ß2-adrenoreceptor-related pathway. As eNOS is the Ca(2)(+)-dependent isoform of the enzyme, we aimed to examine the effects of gastrin-17 on Ca(2)(+) movements. Thus, experiments were performed in Fura-2-acetoxymethyl-ester-loaded PCAEC, where changes of cytosolic Ca(2)(+) ([Ca(2)(+)]c) caused by gastrin-17 were analysed and compared with those of CCK receptors and ß2-adrenoreceptors agonists/antagonists. In addition, some experiments were performed by stimulating cells with gastrin-17 in the presence or absence of cAMP/PKA activator/inhibitor and of phospholipase C (PLC) and Ca(2)(+)-calmodulin dependent protein kinase II (CaMKII) blockers. The results have shown that gastrin-17 can promote a transient increase in [Ca(2)(+)]c mainly originating from an intracellular pool sensitive to thapsigargin and from the extracellular space. In addition, the response of cells to gastrin-17 was increased by the adenylyl cyclase activator and the ß2-adrenoreceptor agonists and affected mainly by the CCK2 receptor agonists/antagonists. Moreover, the effects of gastrin-17 were prevented by ß2-adrenoreceptors and CaMKII blockers and the adenylyl cyclase/PKA and PLC inhibitors. Finally, in PCAEC cultured in Na(+)-free medium or loaded with the plasma membrane Ca(2)(+) pump inhibitor, the gastrin-17-evoked Ca(2)(+) transient was long lasting. In conclusion, this study shows that gastrin-17 affected intracellular Ca(2)(+) homeostasis in PCAEC by both promoting a discharge of an intracellular pool and by interfering with the operation of store-dependent channels through mainly CCK2 receptors and PKA/PLC- and CaMKII-related signalling downstream of ß2-adrenoreceptor stimulation.

Levosimendan induces NO production through p38 MAPK, ERK and Akt in porcine coronary endothelial cells: role for mitochondrial K(ATP) channel.

BACKGROUND AND PURPOSE: Levosimendan acts as a vasodilator through the opening of ATP-sensitive K(+) channels (K(ATP)) channels. Moreover, the coronary vasodilatation caused by levosimendan in anaesthetized pigs has recently been found to be abolished by the nitric oxide synthase (NOS) inhibitor N(omega)-nitro-L-arginine methyl ester, indicating that nitric oxide (NO) has a role in the vascular effects of levosimendan. However, the intracellular pathway leading to NO production caused by levosimendan has not yet been investigated. Thus, the purpose of the present study was to examine the effects of levosimendan on NO production and to evaluate the intracellular signalling pathway involved. EXPERIMENTAL APPROACH: In porcine coronary endothelial cells (CEC), the release of NO in response to levosimendan was examined in the presence and absence of N(omega)-nitro-L-arginine methyl ester, an adenylyl cyclase inhibitor, K(ATP) channel agonists and antagonists, and inhibitors of intracellular protein kinases. In addition, the role of Akt, ERK, p38 and eNOS was investigated through Western blot analysis. KEY RESULTS: Levosimendan caused a concentration-dependent and K(+)-related increase of NO production. This effect was amplified by the mitochondrial K(ATP) channel agonist, but not by the selective plasma membrane K(ATP) channel agonist. The response of CEC to levosimendan was prevented by the K(ATP) channel blockers, the adenylyl cyclase inhibitor and the Akt, ERK, p38 inhibitors. Western blot analysis showed that phosphorylation of the above kinases lead to eNOS activation. CONCLUSIONS AND IMPLICATIONS: In CEC levosimendan induced eNOS-dependent NO production through Akt, ERK and p38. This intracellular pathway is associated with the opening of mitochondrial K(ATP) channels and involves cAMP.

The role of nitric oxide in the peripheral vasoconstriction caused by human placental lactogen in anaesthetized pigs.

Regional intra-arterial infusion of human placental lactogen in anaesthetized pigs has been shown to cause coronary, renal and iliac vasoconstriction by antagonizing the vasodilatory effects of beta2-adrenergic receptors. Since nitric oxide is known to modulate or mediate beta2-adrenergic effects, the present study was planned in the same experimental model to determine the role of nitric oxide in the above vascular responses to human placental lactogen. In eight pigs anaesthetized with sodium pentobarbitone, changes in anterior descending coronary, left renal and left internal iliac blood flow caused by intra-arterial infusion of human placental lactogen at constant heart rate and arterial blood pressure were assessed using electromagnetic flowmeters. Intra-arterial infusion of the human placental lactogen caused decreases in coronary, renal and iliac blood flow which, respectively, averaged 16.7, 8.1 and 12.2% of the baseline values. The role of nitric oxide in this response was studied in the same pigs by repeating the experiments, after measured blood flows had returned to baseline values, following intra-arterial administration of N(omega)-nitro-L-arginine methyl ester (L-NAME). The subsequent intra-arterial infusion of human placental lactogen did not cause any significant changes in measured blood flows, even when performed after reversing the increase in arterial blood pressure and coronary, renal and iliac resistance caused by L-NAME with continuous intravenous infusion of papaverine. These results indicate that the coronary, renal and iliac vasoconstriction caused by human placental lactogen, known to involve antagonism of beta2-adrenergic vasodilatory effects, was mediated by inhibition of nitric oxide release.

Human placental lactogen decreases regional blood flow in anesthetized pigs.

In 22 pigs anesthetized with sodium pentobarbitone, changes in blood flow caused by infusion of human placental lactogen into the left renal, external iliac, and anterior descending coronary arteries were assessed using electromagnetic flowmeters. In 17 pigs, infusion of human placental lactogen whilst keeping the heart rate and arterial pressure constant decreased coronary, renal and iliac flow. In 5 additional pigs, increasing the dose of human placental lactogen produced a dose-related decrease in regional blood flow. The mechanisms of the above response were studied in 15 of the 17 pigs by repeating the experiment of infusion. The human placental lactogen-induced decrease in regional blood flow was not affected by blockade of cholinergic receptors (5 pigs) or of alpha-adrenergic receptors (5 pigs), but it was abolished by blockade of beta2-adrenergic receptors (5 pigs). The present study showed that intra-arterial infusion of human placental lactogen primarily decreased coronary, renal and iliac blood flow. The mechanism of this response was shown to be due to the inhibition of a vasodilatory beta2-adrenergic receptor-mediated effect.

Hemodynamic effect of intracoronary administration of levosimendan in the anesthetized pig.

In this study the hemodynamic effects of intracoronary injection of levosimendan in anesthetized pigs and the mechanisms involved were examined. In 12 anesthetized pigs instrumented for measurement of heart rate (HR), aortic blood pressure (ABP), central venous pressure (CVP), left ventricular end-diastolic blood pressure, left ventricular contractility and relaxation, and mean coronary blood flow (CBF), levosimendan has been injected into the left anterior descending coronary artery at doses corresponding to the ones commonly used in clinics as bolus administration but adapted to the measured CBF. In a further 9 pigs levosimendan has been administered after the blockade of alpha and beta adrenoceptors, muscarinic receptors, and coronary nitric oxide synthase (NOS) to investigate the action mechanism of the drug. The intracoronary bolus administration of doses of levosimendan corresponding to 12 and 24 microg/kg in 10 minutes exerted, respectively, CBF increases of 26.3% and 41.3% of the control values in the absence of changes in the other hemodynamic variables. The blockade of the autonomic nervous system did not prevent the coronary vasodilation, which was, however, abolished by the NOS inhibition. The intracoronary administration of levosimendan exerts positive effects on myocardial blood supply without changes in ABP, HR, CVP, or in myocardial kinetics. The coronary effects of levosimendan are related to NO production.

The effect of distension of the stomach on plasma renin activity in the anesthetized pig.

It has recently been shown that distension of the stomach in anesthetized pigs causes reflex hemodynamic responses through efferent sympathetic mechanisms. The present study was undertaken to investigate whether these mechanisms include activation of the renin-angiotensin system. In twelve anesthetized pigs, intragastric balloons were distended for periods of 30 minutes by 0.81 of warm Ringer solution (mean gastric transmural pressure of about 12 mmHg). Changes in arterial blood pressure and heart rate were respectively prevented by a pressurized reservoir connected to the left femoral artery and by atrial pacing. Plasma renin activity was measured during the last minute of distension by radioimmunoassay of angiotensin I. In each of the twelve pigs distension of the stomach caused an increase in plasma renin activity. In five pigs, this response was graded with step increments of the distension. The increase in plasma renin activity to gastric distension was abolished by bilateral subdiaphragmatic vagotomy (six pigs) and by bilateral section of the renal nerves (six pigs). The present study showed that innocuous distension of the stomach in the anesthetized pig reflexly increased plasma renin activity. The afferent limb of the reflex was in the vagal nerves and the efferent limb involved renal nerves.

Chronic hepatitis C is mild in menstruating women.

BACKGROUND AND AIMS: Women with chronic hepatitis C may have a slower rate of disease progression than men. We have previously demonstrated a relationship between hepatic iron concentration and liver fibrosis in patients with chronic hepatitis C. Our aim was to compare hepatic histologic findings, iron status and other factors putatively capable of determining the severity of chronic hepatitis between menstruating women and men of comparable age. METHODS: We studied 21 consecutive hepatitis C virus (HCV)-RNA positive menstruating women and 24 consecutive HCV-RNA positive men of comparable age, who underwent liver biopsy for chronic hepatitis C. Alcohol intake was recorded and blood tests, HCV genotyping, serum iron, unsaturated iron binding capacity, serum ferritin, hepatic iron concentration, and liver histology were evaluated. RESULTS: Menstruating women showed lower grading (2.7 +/- 1.5 vs 3.6 +/- 2, P = 0.09) and significantly lower staging (1.38 +/- 1.11 vs 2.42 +/- 1.64, P = 0.037) scores than men of comparable age. Among the factors putatively capable of determining the severity of chronic hepatitis, only the hepatic iron concentration correlated with the hepatic histologic staging in a multivariate analysis. Iron-depleted women (transferrin saturation < 20% and/or serum ferritin < 9 micrograms/L) showed significant lower hepatic histologic grading (1.75 +/- 0.7 vs 3.23 +/- 1.55, P = 0.027) and staging (0.75 +/- 1.03 vs 1.77 +/- 1.01, P = 0.026) scores than women with normal iron status. CONCLUSIONS: Menstruating women with chronic hepatitis C may have a milder disease compared to men of comparable age, possibly because of menstrual blood loss and lower hepatic iron concentration. Women with chronic hepatitis C and iron deficiency have a milder disease compared to women with normal iron status, suggesting that iron deficiency results in a slower rate of disease progression.

Effect of progesterone on peripheral blood flow in prepubertal female anesthetized pigs.

This study was undertaken to determine the effects of progesterone on the peripheral circulation. In prepubertal female pigs anesthetized with sodium pentobarbitone, changes in the superior mesenteric, left renal and left external iliac flow caused by intravenous infusion of progesterone were assessed using electromagnetic flow meters. Changes in heart rate and arterial blood pressure were prevented by atrial pacing and by connecting the arterial system to a pressurized reservoir containing Ringer solution. In 20 pigs, infusion of 1 mg/kg of progesterone increased mesenteric, renal and iliac flow. In a further 4 pigs, the vasodilatory effects of the hormone were enhanced by graded increases in the dose between 1, 2 and 3 mg/kg. The mechanisms of these responses were studied in the 20 pigs by repeating the experiment after hemodynamic variables had returned to the control values before infusion. In 5 pigs, blockade of adrenergic receptors with propranolol and phentolamine did not affect the responses elicited by progesterone. The increases in mesenteric, renal and iliac flow to progesterone were prevented, respectively, by the injection of N(omega)-nitro-L-arginine methyl ester into the mesenteric (5 pigs), the renal (5 pigs) or the iliac artery (5 pigs). The present study shows that intravenous infusion of progesterone dilated mesenteric, renal and iliac circulations. The mechanism of this response involved the release of nitric oxide.

The effects of hypertonic saline solution on coronary blood flow in anaesthetized pigs.

1. The effects of intracoronary bolus infusion of hypertonic saline solution on left circumflex coronary blood flow were examined in sixteen anaesthetized and artificially ventilated pigs whilst preventing changes in heart rate and arterial blood pressure. 2. In fourteen pigs, bolus infusion of 7.5% hypertonic saline solution (2 ml within 30 s) caused a steady-state increase in coronary blood flow without significantly affecting right atrial or left ventricular pressure and its rate of rise (dP/dtmax). Infusing normal saline solution (0.9%) at the same rate and volume in seven pigs did not have this effect. 3. In five pigs, the magnitude and the duration of the response of increase in coronary blood flow were increased in a graded manner by graded increases in the concentration of the hypertonic saline solution between 2.5, 5 and 7.5%. 4. In nine pigs, the response of increase in coronary blood flow to the bolus infusion of hypertonic saline solution was not affected by the blocking agents atropine, propranolol and phentolamine, but it was completely abolished in the same nine pigs by the subsequent intracoronary administration of N omega-nitro-L-arginine methyl ester (L-NAME) which blocks the synthesis of endothelium-derived relaxing factor (EDRF) and in seven pigs by solely giving L-NAME. 5. These results showed that the intracoronary bolus infusion of hypertonic saline solution in anaesthetized pigs caused a coronary vasodilatation which involved mechanisms dependent on the release of EDRF.