Do vasomotor nerves significantly regulate cerebral blood flow?

Effect of sympathetic nerve stimulation and adrenoceptor blockade on pial arterial and venous calibre and on intracranial pressure in the cat

Pial arterial and venous calibre were continuously recorded through a closed cranial window preparation during cervical sympathetic nerve stimulation in 10 cats before and after alpha- and beta-adrenoceptor blockade. In addition, the intracranial pressure (ICP) was simultaneously recorded in 4 of the cats. Under resting conditions 33 arteries (mean diameter 130 micron) constricted by 11.7 +/- 0.8% and 80 venous portions (mean diameter 152 micron) constricted by 13.7 +/- 0.7% during sympathetic nerve stimulation. ICP decreased simultaneously by 16.5 +/- 6.2%. Administration of the alpha-adrenoceptor antagonist phenoxybenzamine i.v. (1.5 mg X kg-1) abolished the reduction of ICP and markedly reduced, but did not completely abolish, the constrictor response of arteries and veins. The beta-adrenoceptor antagonist propranolol (1.5 mg X kg-1) did not significantly alter the reduction of ICP or the response of pial veins and small arteries, whereas the response of arteries with a diameter greater than 150 micron was attenuated. It is concluded that the constriction of pial veins and arteries during sympathetic stimulation is mediated predominantly via alpha-adrenoceptors. The sympathetic nerves of cerebral blood vessels may have stronger influence on the cerebral capacitance than on resistance vessels under normotensive and normocapnic conditions.

Regional blood flow in canine brain during nicotine infusion: effect of autonomic blocking drugs

Radioactive microspheres (15 mu) were used to measure regional cerebral blood flow during intravenous infusion of nicotine (36 micrograms/kg/min) in anesthetized, open chest dogs. Experiments were conducted with uncontrolled mean aortic pressure and intact autonomic receptors (Series I; n = 9), and in four groups of dogs with mean aortic pressure held constant (Series II); 1) with intact autonomic receptors (n = 6), 2) after beta adrenergic blockade (n = 8), 3) after alpha and beta adrenergic blockade (n = 6), 4) after alpha and beta adrenergic and cholinergic blockade (n = 4). In Series I, nicotine raised mean aortic pressure (+ 72%) and increased flow in cerebral cortex (+ 67%), cerebellum (+ 38%), pons (+ 46%), medulla (+ 39%), and spinal cord (+ 48%). In all regions, but cortex, increases in vascular resistance limited nicotine-induced increases in flow. In Series II, nicotine changed flow only in cortex. Without blockade, nicotine increased cortical flow (+ 38%); but beta blockade abolished this increase in flow. After alpha and beta blockade nicotine again raised cortical flow (+ 29%), and additional cholinergic blockade had no effect on this response. It is concluded that nicotine causes predominant beta receptor mediated vasodilation in cerebral cortex, although it also activates alpha (vasoconstrictor) receptors and a non-adrenergic, non-cholinergic vasodilator mechanism in this region of brain.

Effects of agents that modulate potassium permeability on smooth muscle cells of the guinea-pig basilar artery

1 Effects of various chemical agents, with known actions on K-permeability, on the membrane potential, membrane resistance and spike activity of smooth muscle cells of the guinea-pig basilar artery were investigated using the microelectrode method. The K-permeability suppressing agents procaine, tetraethylammonium (TEA), 4-aminopyridine (4-AP), and the K-permeability increasing agents acetylcholine (ACh), caffeine and 2-nicotinamidoethyl nitrate (2-NN) were used.2 The mean resting membrane potential of smooth muscle cells was -50.6 mV and was electrically quiescent. The maximum slope of the membrane depolarization produced by a ten fold increase in [K](o) plotted on a log scale was 42 mV. The contribution of the Na-K pump mechanism to the membrane potential was apparent, i.e. the maximum hyperpolarization induced by activation of the Na-K pump was -71 mV. This hyperpolarization ceased following application of ouabain. Reduction in the [Na](o) slightly hyperpolarized the membrane.3 TEA (1-10 mM) and procaine (0.3-10 mM) depolarized the membrane dose-dependently, and increased the membrane resistance. TEA markedly inhibited and procaine slightly inhibited rectification of the membrane.4 Caffeine had dual actions on the membrane potential, i.e. a low concentration (below 1 mM) hyperpolarized the membrane and reduced the membrane resistance, while increased concentrations (above 1 mM) transiently hyperpolarized, and then depolarized the membrane with an increase in the membrane resistance.5 ACh hyperpolarized the membrane dose-dependently (10(-7)-10(-5)M), and reduced the membrane resistance. The hyperpolarizing action of ACh did not persist and even in the presence of ACh, the membrane was repolarized to near the control level.6 2-NN (up to 10(-4)M) and 4-AP (up to 1 mM) did not modify the membrane potential or the membrane resistance.7 TEA (10 mM) generated a spike either spontaneously or by application of electrical stimulation and the spike was preceded or followed by slow oscillatory potential changes. These potential changes ceased with application of diltiazem (10(-5)M).8 Low concentrations of procaine (below 1 mM) accelerated but high concentrations (5-10 mM) inhibited the spike generation in the presence of 10 mM TEA. Low concentrations of caffeine (below 1 mM) inhibited the spike generation and higher concentrations (above 1 mM) of caffeine accelerated the spike generation in the presence of 10 mM TEA. ACh (10(-6)-10(-4)M) consistently inhibited the spike generation, with 10 mM TEA pretreatment.9 4-AP (1 mM) did not modify the membrane potential yet accelerated the spike generation, in the presence of 10 mM TEA. 2-NN (10(-4)M) had no effect on the spike evoked in the presence of 10 mM TEA.10 The results show that the low membrane potential in smooth muscle cells of the guinea-pig basilar artery is mainly due to the low permeability of the membrane to K ion, presumably due to the lack of a K channel sensitive to 2-NN and 4-AP. Similarities and differences between this vascular tissue and other regions are also discussed.

Neurogenic modification of the vulnerability of the blood-brain barrier during acute hypertension in conscious rats

To study the possible influence of sympathetic adrenergic tone on the blood-brain barrier function during acute hypertension in conscious unrestrained rats with indwelling catheters in the aorta and a jugular vein the blood pressure was increase by noradrenaline, 6-hydroxydopamine (6-OHDA) or baclofen. One or 60 min later the rats were sacrificed and the extravasation of 125I labelled albumin determined in the brain. After i.v. injection of noradrenaline the baroreceptor reflex will decrease the sympathetic tone whereas the blood pressure increased induced by the other two drugs is accompanied by an increased sympathetic activity. One minute after a corresponding rise in blood pressure the albumin content in the brain was considerably lower in rats given 6-OHDA than in those given noradrenaline. 60 min after the injection of 6-OHDA or baclofen the extravasation in the brain did not differ despite a considerably more rapid increase in pressure after 6-OHDA. Pretreatment with clonidine increased the blood-brain barrier dysfunction in rats given 6-OHDA but not in those given baclofen, probably because the slower rise in pressure facilitates myogenic autoregulation. It is concluded that neurogenic influences on vessel tone can modify the response of the blood-brain barrier during acute hypertension in conscious rats.

Video camera method for simultaneous measurement of blood flow velocity and pial vessel diameter

A new method for the simultaneous measurement of blood flow velocity and pial vessel diameter is described. The system consists basically of a high-sensitivity vidicon camera, camera control, width analyzer, video densitometer, TV monitor, desktop computer, and multi-pen recorder. The pial vessels are visualized through a cranial window at 25-200x magnification on the TV monitor. The diameter of three target vessels can be recorded simultaneously on the recorder by adjustment of controllable video signal gates using the width analyzer. At the same time, the optical densities of two targets at points upstream and downstream of the pial vessel are measured continuously with video densitometers, and their outputs are recorded on the polygraph and analyzed by the computer. The time difference in the two peaks of time--concentration curves, produced very 2-3 s at the highest frequency by the injection of a small amount of saline through the lingual artery, is measured on-line using the computer. The flow velocity in the vessel is calculated from the time difference and the distance between the two targets. The system was shown to be stable, reliable, and rapid in response. This method may provide a useful tool for research in the field of blood circulation in the brain or any other organ.

Further evidence for a muscarinic component to the neural vasodilator innervation of cerebral and cranial extracerebral arteries of the cat

Transmural electrical stimulation of segments of lingual and cerebral (basilar, middle and posterior cerebral) and also other cranial arteries of the cat results after a long latency in a dilator response. The response may be resolved into two components--an initial transient atropine-sensitive component and a slower more ponderous one that is atropine-resistant. The variability in pattern of dilation responses from segments of different vessels or even those from the same segment of different cats is considerable. Some responses are entirely atropine-sensitive and others atropine-resistant; however the vast majority show a dilation that can be considered to be made up of both components. The latencies of the atropine-sensitive and atropine-resistant components are not different. The effect of atropine on the lingual but not the cerebral arteries is frequency dependent, being proportionately greater at low than at high frequencies. In both vessels, the effect of atropine is independent of train length at 1 Hz. Physostigmine potentiates significantly the dilation of the lingual artery but not that of the cerebral arteries. The potentiation is reversed by atropine. The endogenous acetylcholine level was measured in a series of vessels. It can be correlated with the activity of choline acetyltransferase and the presence of neurogenic dilation. It is proposed that there are two transmitters released in parallel from nerve(s) in the walls of cerebral, lingual, and possibly, other cranial arteries to cause vasodilation. It seems that one of these is acetylcholine.

Pial artery diameter and blood flow velocity during sympathetic stimulation in cats

The effects of sympathetic stimulation on the cerebral circulation in cats are in dispute. One unexplained observation is that sympathetic nerve stimulation constricts pial arteries but does not decrease cerebral blood flow (CBF). To reconcile these findings, we studied effects of sympathetic nerves on cerebral vessels using a new method that permits virtually continuous measurement of pial artery diameter and blood flow velocity, and calculation of change in CBF. Change in CBF was calculated as the product of cross-sectional area (CSA) and blood flow velocity in a large pial artery. Pial artery diameter was measured with the cranial window method and CSA was calculated (pi r2). Blood flow velocity was measured with a pulsed Doppler velocity meter. In 11 cats, CBF was measured duringg the control period and during electrical stimulation of sympathetic nerves. During stimulation, arterial diameter decreased 6 +/- 1%, but CBF did not change because velocity increased 19 +/- 5%. We also measured CBF with microspheres during control and after 60 s of sympathetic stimulation. Cerebral blood flow was 28 +/- 3 ml/min/100 g during control and 26 +/- 3 ml/min/100 g after 60 s of sympathetic stimulation (n = 5). It is likely that the increase in velocity durin sympathetic stimulation was due to compensatory dilatation of downstream arteries.

Comparison of the effects of potassium and pH on the calibre of cerebral veins and arteries

The vasomotor response of individual pial veins and arteries on the convexity of the cerebral cortex to perivascular microinjection of mock cerebrospinal fluid (CSF) containing various concentrations of potassium (K+) and of various pH (achieved by altering the bicarbonate, HCO-3 concentration) have been examined in cats anaesthetised with alpha-chloralose. Microapplication of CSF containing 0 mM HCO-3 (pH 4.80) effected significant increases in calibre of pial veins and arteries of 9.3 +/- 2.4% and 38.2 +/- 4% respectively (mean calibre change +/- SE), whereas CSF containing 22 mM HCO-3 (pH 7.45) which constricted pial arteries significantly (-18.5 +/- 2.9%) minimally altered venous calibre (-4.3 +/- 2.2%). Microapplication of CSF containing 0 mM potassium resulted in a significant reduction in pial arterial calibre (-11.4 +/- 2.8%) but failed to alter pial venous calibre (-0.3 +/- 0.6%). Perivascular microapplication of CSF containing moderately elevated potassium concentrations (10 mM) which effected marked, significant increases in pial arterial calibre (49.3 +/- 3.9%) did not significantly alter the calibre of the pial veins (mean response -1.6 +/- 2.4%). The perivascular administration of CSF containing a high concentration of potassium (40 mM) resulted in the significant constriction of both pial veins (-13.5 +/- 0.9%) and pial arteries (-47.2 +/-6.3%). The magnitude of the response was significantly smaller in the pial veins. The relative insensitivity to K+ and pH of the pial veins as compared to pial arteries suggests that alteration in the chemical composition of of the perivascular fluid are of lesser importance in the control of cerebrovascular capacitance than for the regulation of cerebrovascular resistance.

Feline cerebral veins and arteries: comparison of autonomic innervation and vasomotor responses

1. The innervation of feline cerebral (pial) vessels by nerve fibres containing noradrenaline, substance P or vasoactive intestinal polypeptide (VIP) has been examined using the Falck-Hillarp histo-fluorescence method and immunohistochemical techniques. Cerebral veins were shown to be innervated by nerve fibres containing noradrenaline, substance P or VIP. Nerve fibres containing noradrenaline were the most numerous, while fibres containing substance P were observed least frequently in both types of vessel. For each putative neurotransmitter, the density of the innervation of the cerebral veins was less than that of cerebral arteries.2. The vasomotor responses of individual pial arteries and veins on the convexity of the cerebral cortex to perivascular micro-injection of noradrenaline, substance P and VIP were examined in twenty-five cats anaesthetized with alpha-chloralose.3. The perivascular micro-application of noradrenaline resulted in pronounced dose-dependent reductions in the diameter of pial veins (maximum calibre reduction: 32+/-3% noradrenaline 10(-5) M) and arteries (22+/-3% noradrenaline 10(-5) M). Pial veins were more sensitive to noradrenaline than were pial arteries tested under similar conditions. The reductions in the diameter of cerebral veins and arteries resulting from the administration of noradrenaline could be attenuated by the concomitant micro-application of phentolamine (10(-6) M).4. The perivascular micro-application of substance P effected significant dose-dependent increases in the calibre of pial veins (maximum calibre increase: 16+/-4% substance P 10(-7) M) which were of a similar magnitude to those observed in pial arteries in response to this peptide (21+/-4% substance P 10(-6) M).5. The perivascular micro-application of VIP resulted in small increases in the calibre of pial veins (maximum calibre increase: 9+/-2% VIP 10(-8) M) which were proportionately smaller than those observed in pial arteries in response to this peptide (23+/-5% VIP 10(-7) M).

Electrical properties of smooth muscle cell membrane and neuromuscular transmission in the guinea-pig basilar artery

1 The membrane properties of smooth muscle cells and neuromuscular transmission in the guinea-pig basilar artery were investigated by use of microelectrodes.2 The membrane potential was -47.0 mV and the muscle tissue possessed cable-like properties as determined by the current-voltage relationships. The mean value of the spacè constant was 0.78 mm.3 An outward current produced a graded response and, in some cases, spike generation. This membrane response was enhanced in the presence of tetraethylammonium (TEA, 5 mM), and an increased concentration of TEA (10 mM) generated spontaneous spikes in most of the cells. Action potentials induced by TEA were abolished in the presence of MnCl(2) (5 mM) but not by isoprenaline (4 x 10(-6) M).4 Acetylcholine (ACh), over 10(-7) M, hyperpolarized the membrane and decreased the membrane resistance. This hyperpolarization increased in the presence of low [K](o) (below 5.9 mM), but decreased in [K](o) concentrations over 17.8 mM. Pretreatment with atropine (10(-6) M) suppressed the ACh-induced hyperpolarization. Therefore, this action of ACh is due to an increase in the K-conductance of the membrane produced by activation of the muscarinic receptors.5 Noradrenaline in concentrations up to 10(-4) M did not modify the membrane potential and resistance, while 10(-5) M, histamine, 5-hydroxytryptamine and adenosine triphosphate (ATP) depolarized the membrane. The depolarization induced by histamine or ATP was suppressed by reducing [Na](o). The histamine-induced depolarization was accompanied by an increase and the ATP-induced one by a decrease in the membrane resistance. The action of histamine was suppressed by treatment with H(1)- but not H(2)-receptor blocking agents (dephenhydramine and cimetidine, respectively).6 Perivascular nerve stimulation (0.2 ms pulse duration) evoked excitatory junction potentials (e.j.ps). An increase in the number and frequency of stimuli enhanced the e.j.p. amplitude. In the presence of 1 mM TEA, a spike was evoked on the e.j.ps. A very high concentration of phentolamine (3.6 x 10(-4) M) or the usual concentration of tetrodotoxin (10(-7) M) abolished the generation of e.j.ps. Spontaneously generated miniature e.j.ps were never recorded from the resting membrane.7 The results are discussed in relation to regional specificities of smooth muscle cells of cerebral arteries in the guinea-pig.

Microangioarchitecture of rat parietal cortex with special reference to vascular "sphincters". Scanning electron microscopic and dark field microscopic study

Microangioarchitecture of the rat parietal cortex was studied by means of scanning electron microscopy and dark field microscopy. The richest supply of blood vessels in the parietal cortex was found in layer III + IV and layer V, where 2 isolated plexuses of microvessels were prominent. The appearance of the plexuses was quite different between motor and sensory areas. In the motor area the capillary plexuses were narrow and compact, while in sensory area the plexuses were wide and diffuse. Characteristic ring formations, called ring-shaped-compressions in the present study, were frequently observed at branching sites of arterioles. The ring-shaped-compression probably corresponds to the precapillary sphincter. A similar structure was also seen in capillaries and venules and, therefore, it is likely that not only arterioles, but also capillaries and even venules, can actively change diameter to control cerebral blood flow.

Sympathetic nerves protect against stroke in stroke-prone hypertensive rats. A preliminary report

Studies were performed to determine whether sympathetic nerves protect against stroke in hypertensive rats. The superior cervical ganglion was removed on one side in 28 stroke-prone spontaneously hypertensive rats (SHRSP) when the rats were 4 weeks old. The rats were fed Japanese rat chow and 1% saline drinking water. When the rats were 19 weeks old, systolic pressure was 206 +/- 4 mm Hg (mean +/- SE). All rats died between 19 and 23 weeks of age. Microscopic and histological examination demonstrated cerebral hemorrhage in seven rats. All the hemorrhages occurred in the denervated hemispheres. Ischemic cerebral infarctions were found in 13 rats; in 10 rats, the infarcts were only in the denervated hemisphere. Pathological changes of cerebral arteries (hyalinosis, fibrinoid changes, and thrombus formation) were observed primarily in denervated hemispheres. Wall-to-lumen ratio was less in arteries of the denervated hemisphere than in arteries of the innervated hemisphere. These preliminary observations suggest that denervation of cerebral vessels increases susceptibility to stroke and inhibits development of cerebral vascular hypertrophy in SHRSP.

Norepinephrine depletion alters cerebral oxidative metabolism in the "active" state

Unilateral chemical lesion of the nucleus locus coeruleus in rats produced unilateral depletion of ipsilateral cortical norepinephrine. Norepinephrine depletion was not associated with changes in "resting" metabolic balance of the cerebral cortex, as determined by in situ reflection spectrophotometry of the redox state of cytochrome oxidase. Norepinephrine depletion, however, caused slowing of the transient metabolic response to sudden increases in energy demand produced by direct cortical electrical stimulation. The effect was apparent in the redox state of both cytochrome oxidase and NAD, the latter being measured, also in situ, by reflection microfluorometry. These results demonstrate that norepinephrine has a role in modulating the response to increased metabolic demand in the cerebral cortex. Norepinephrine may mediate its effect by potentiating Na+, K+-ATPase or through its effects on vascular reactivity, or both.

Perivascular substance P: occurrence and distribution in mammalian pial vessels

Nerve fibres containing immunoreactive substance P (SP) were demonstrated in the wall of cerebral blood vessels of several mammalian species. Pial arteries of cat and guinea-pig were richly supplied with SP nerve fibres, while those of rat, rabbit, pig, and man had a moderate number. SP fibres were more numerous in pial vessels belonging to the rostral parts of the circle of Willis as compared to more caudally located blood vessels. In cat and guinea-pig, blood vessels in the choroid plexus were surrounded by few SP nerve fibres; also spinal cord blood vessels of cat contained few such fibres.

Adrenergic and cholinergic receptors of cerebral microvessels

The presence of alpha- and beta-adrenergic and muscarinic cholinergic receptors in cerebral microvessels of the rat and pig was assessed by ligand binding techniques. The results demonstrate the presence of specific binding to alpha 2- and beta-adrenergic receptors but no appreciable specific binding to alpha 1-adrenergic or muscarinic cholinergic receptors. beta-Adrenergic receptors of pig cerebral microvessels are similar to those of the brain and other organs in their binding characteristics to the tritiated ligand and in their stereospecificity of binding to the biologically active isomers of beta-adrenergic agonists. Further evidence derived from the differential potency of binding displacement by the various beta-adrenergic agonists and selective beta 1- and beta 2-adrenergic antagonists indicates that beta-adrenergic receptors of pig cerebral microvessels are mostly of the beta 2-subtype.

Cerebrovascular sympathetic denervation and blood-brain barrier function in conscious rats

Electrical stimulation of the sympathetic nerves to the cerebrovascular bed enables the resistance vessels to better withstand a high blood pressure in terms of blood-brain barrier integrity. Sympathetic denervation could hence be expected to lead to a decrease in cerebrovascular tone and increased vulnerability of the blood-brain barrier. In the present study acute hypertension was induced in conscious unrestrained rats by administration of angiotensin or bicuculline. The albumin leakage into the brain, as studied by Evans blue-albumin and 125I labelled human serum albumin, was not enhanced in acutely or chronically sympathectomized rats compared to controls.

Validity of cerebral blood flow measurements obtained with quantitative tracer techniques

A great number of results for the cerebral blood flow obtained in the animal with quantitative tracer techniques have been collected from the literature. They are exposed in order to compare both normal flow values in different laboratory species, and the characteristics, accuracy and sensitivity of each technique. A dramatic overall dispersion of flow values is observed, allowing neither the flow level particular to each species to be estimated, nor the average value provided by a given technique to be found. The physiological and technological causes of such a dispersion are discussed. Several techniques seem to have limitations which even alter the interpretation of their results, and especially the origin of the local or regional blood flow results. Other techniques may be criticized from the quantitative standpoint, but give more reliable results.

Relationship of cerebral blood flow to cardiac output, mean arterial pressure, blood volume, and alpha and beta blockade in cats

The relationship among cerebral blood flow (CBF), blood volume, cardiac output (CO), and mean arterial blood pressure (MABP) at varying levels of arterial CO2 tensions (PaCO2) were studied in 70 normal cats. The CBF was measured from the clearance curve of xenon-133 and CO with a thermal dilution catheter placed in the pulmonary artery. The CBF, CO, and MABP values varied appropriately with changes in PaCO2, confirming the reliability of the preparations and the presence of normal autoregulatory responses. Moderate hypovolemia that did not change MABP did, nevertheless, significantly decrease CO and CBF. In an effort to determine if this decrease in CO and CBF were coupled responses, the effects of beta stimulation, hypervolemia, and alpha and beta blockade were investigated. Propranolol, in a dosage insufficient to change MABP, decreased both CO and CBF. This agent abolished the CO response to elevations in PaCO2 but not the CBF response, making it unlikely that this CBF reduction resulted from impaired cerebral autoregulation. Isoproterenol, which, in contrast to propranolol, does not cross the normal blood-brain barrier, alone or in combination with phenoxybenzamine, produced a 38% and 72% increase in CO, respectively, without a change in CBF. Alpha blockade (no major change in CO) and beta blockage (major decrease in CO) did not significantly effect cerebral autoregulation to changes in MABP from angiotensin. The ability of the brain to resist increases in MABP and CO and maintain normal CBF is explained by normal cerebral autoregulation. However, its vulnerability to modest decreases in blood volume, which cannot be attributed to variations in perfusion pressure, is unexplained but obviously has important therapeutic implications. This may be related to reduction in CO, changes in autonomic activity, or a decrease in the size of the perfused capillary bed.

Intracellular brain pH and the pathway of a fat soluble pH indicator across the blood-brain barrier

Umbelliferone, a pH sensitive fluorescent indicator, can be used to determine intracellular pH measurements and analyze the pathway of a fat soluble substance across the blood--brain barrier (BBB). The 'equivalent intracellular pH' determined by this technique corresponds closely with the calculated intracellular pH derived from mathematical abstractions. The pH of the indicator's environment along its pathway into and out of brain tissue can be determined from a ratio analysis of various points along the calibrated fluorescent tissue clearance curves from 340 and 370 nm excitation. This analysis indicates that immediately upon leaving the intravascular space the indicator enters an environment that is too acid to represent the extracellular space. This suggests that fat soluble substances follow an intracellular pathway across the BBB (capillary endothelium to glial cell to neuron).

Transient responses of rabbit cerebral blood vessels to norepinephrine: correlation with intrinsic myogenic tone

Transient contractile responses to norepinephrine (NE) of vascular segments from the rabbit vertebral, internal carotid, and basilar arteries rise to a peak within several seconds, and in the presence of the agonist, reverse rapidly, relaxing with a half-time of 15 +/- 3.2 seconds. In the basilar artery, peak contraction is approximately 25% of the maximum response mediated via the "alpha-like" adrenoreceptors and is elicited by NE 10(-7) M. Steady state contractions are seen with higher concentrations. Transient contractile responses are absent in segments from the brachiocephalic and external carotid arteries, and their incidence increases the more rostral along the length of the vertebral and internal carotid artery the origin of the segment studied. They were seen in all preparations of the intracranial vertebral and basilar arteries. There is a good correlation between the occurrence in any particular vascular segment of the transient contractile response and intrinsic tone as assessed by relaxation to papaverine (10(-6) M). The response was blocked by alpha-adrenergic receptor blocking agents and was not elicited by d-NE nor tetrahydrazoline or oxymetazoline. This response may be analogous to the first phase of the biphasic contraction found in many other blood vessels. Since in cerebral vessels the agonist concentration to elicit the first phase is several orders of magnitude lower than the second, it can appear in the absence of the latter.

The cholinergic pathway to cerebral blood vessels. II. Physiological studies

The effect of stimulating the greater superficial petrosal nerve (g.s.p.n.) upon retroglenoid venous blood flow has been tested in anaesthetized, paralysed and artificially ventilated rats. In 11 out of 15 tests, blood flow increased by an average of 25% with a time to peak response of 28 s. This response was abolished with the injection of atropine 0.1 mg kg-1 injected intra-arterially. With both petrosal nerves intact, the administration of 6-7% CO2 in air or 15% O2 in N2 caused average increases in blood flow of 105% and 45% respectively. These responses were not affected by bilateral section of the g.s.p.n. Similar experiments were carried out in 5 anaesthetized, spontaneously breathing rabbits in which, in addition to PaCO2 and PaO2, PO2, PCO2 and blood flow in the caudate nucleus were measured continuously using chronically implanted mass spectrometer catheters and heated thermistors. Caudate nucleus blood flow increased in response to hypoxia and hypercapnia and this response was not significantly affected by section of one or both g.s.p.n., sinus or vagus nerves. With section of sinus and vagus nerves, blood flow changed passively with arterial pressure.