Serotonin type 1D receptors (5HTR) are differentially distributed in nerve fibres innervating craniofacial tissues.

We tested the hypothesis that the 5HT(1D)R, the primary antinociceptive target of triptans, is differentially distributed in tissues responsible for migraine pain. The density of 5HT(1D)R was quantified in tissues obtained from adult female rats with Western blot analysis. Receptor location was assessed with immunohistochemistry. The density of 5HT(1D)R was significantly greater in tissues known to produce migraine-like pain (i.e. circle of Willis and dura) than in structures in which triptans have no antinociceptive efficacy (i.e. temporalis muscle). 5HT(1D)R-like immunoreactivity was restricted to neuronal fibres, where it colocalized with calcitonin gene-related peptide and tyrosine hydroxylase immunoreactive fibres. These results are consistent with our hypothesis that the limited therapeutic profile of triptans could reflect its differential peripheral distribution and that the antinociceptive efficacy reflects inhibition of neuropeptide release from sensory afferents. An additional site of action at sympathetic efferents is also suggested.

Stimulating circle of Willis nerve fibers preserves the diffusion-perfusion mismatch in experimental stroke.

BACKGROUND AND PURPOSE: Stimulation of the nerves traversing the ethmoidal foramen (including postsynaptic, parasympathetic projections from the sphenopalatine ganglion [SPG], henceforth referred to as "SPG-stimulation") has been shown to elevate cerebral blood flow (CBF) and to be neuroprotective after permanent, middle cerebral artery occlusion (pMCAO). METHODS: Employing diffusion (DWI)- and perfusion (PWI) weighted MRI, the effect of SPG-stimulation (started at 60 minutes post-MCAO) on the spatiotemporal evolution of ischemia during and after pMCAO was investigated. In an additional experiment, regional CBF changes were investigated in the nonischemic brain. RESULTS: In the nonischemic brain, SPG stimulation significantly elevated CBF predominantly within areas supplied by the anterior cerebral artery (by 0.64 mL/g/min relative to baseline). In the ischemic brain, CBF only marginally increased within the penumbra and core (by up to 0.08 and 0.15 mL/g/min relative to prestimulation, respectively). However, the threshold-derived CBF lesion volume did not change significantly. Penumbral apparent diffusion coefficient (ADC)-values improved to almost baseline values and the threshold derived ADC/CBF-mismatch was preserved up to 180 minutes after MCAO. TTC-derived lesion volumes were significantly smaller in stimulated versus nonstimulated animals (120.4+/-74.1 mm(3) versus 239.3+/-68.5 mm(3), respectively). CONCLUSIONS: This study demonstrates that unilateral SPG-stimulation increases CBF bilaterally within the normal brain, acutely preserves the CBF/ADC mismatch largely independent of altering cerebral blood flow, and reduces infarct size in the rat permanent suture model.

Calbindin-D28k in cerebrovascular extrinsic innervation system of the rat.

Calbindin-D28k, one of the calcium-binding proteins, belongs to the EF hand family and is commonly found in neurons. It serves as a representative neuronal marker for neuroanatomical investigations. The authors' knowledge of its precise function, however, is yet very limited. In this study, we examined the existence of nerve fibers with calbindin-D28k immunoreactivity in the cerebral blood vessels and ganglia that innervate the cerebral blood vessels in the rat. Numerous nerve fibers with calbindin-D28k immunoreactivity were observed on the walls of the major extracerebral arteries forming the circle of Willis and its branches. Calbindin-D28k immunoreactivity was seen in many neurons of the trigeminal, dorsal root and jugular ganglia. A small number of neurons showed calbindin-D28k immunoreactivity in the otic and superior cervical ganglia. Calbindin-D28k immunoreactivity was not detected in the sphenopalatine or internal carotid ganglia. Pericellular basket-like formations of nerve terminals with calbindin-D28k immunoreactivity were observed in the sphenopalatine, otic, internal carotid and superior cervical ganglia. The present study demonstrated calbindin-D28k immunoreactivity in the cerebrovascular nerve fibers as well as in their origins--the cranial ganglia. These findings are significant in understanding the calcium-mediated mechanism of the neural control of the cerebral blood vessels.

Cholinergic, nitric oxidergic innervation in cerebral arteries of the cat.

Using immunoperoxidase labeling (IPL) and immunofluorescence labeling (IFL) methods, and each followed by NADPH diaphorase (NADPHd) histochemical staining in the same specimen, colocalization of choline acetyltransferase (ChAT) and NADPHd, indicative of nitric oxide synthase (NOS), in cerebral pial arteries and the sphenopalatine ganglia (SPG) of the cat was examined. In addition, retrograde axonal tracing using true blue was performed to determine if cerebral perivascular nerves containing ChAT and NADPHd originate in the SPG. Consistent results were obtained from IPL and IFL methods, indicating that the middle cerebral artery (MCA) and the circle of Willis received dense ChAT-immunoreactive (I) and NADPHd bundles and fine fibers. Almost all ChAT-I fibers and NADPHd fibers were found to be coincident in the arteries examined. A few fine fibers exhibited only NADPHd staining. In the SPG, approximately half of the ganglionic cells were both ChAT-I and NADPHd positive, while the remaining cells were positively only for NADPHd staining. One week after application of true blue on the middle cerebral arteries (MCA), the fluorescent true blue was found in the ganglionic cells of the SPG. Some of the true blue-positive cells contained both ChAT-immunoreactivity and NADPHd staining. These results provide morphological evidence indicating that all ChAT-I fibers in the MCA and the circle of Willis contain NOS, and that these fibers originate in the SPG, although not all NOS-I ganglionic cells in the SPG send fibers to pial vessels. These results also support the hypothesis that acetylcholine (ACh) and nitric oxide (NO) are synthesized and co-released in the same neurons in cerebral perivascular nerves. Based on the reported findings that NO mediates a major component of neurogenic vasodilation, and that ACh acts as a modulator, the present results demonstrate the presence of a cholinergic, nitric oxidergic innervation in cerebral arteries of the cat.

Patterns of peptide-containing perivascular nerves in the circle of Willis: their absence in intracranial arteriovenous malformations.

Using standard immunohistochemical techniques and an improved procedure for whole-mount vascular preparations, the authors describe the pattern and density of innervation of calcitonin gene-related peptide (CGRP)-like, neuropeptide Y (NPY)-like and vasoactive intestinal polypeptide (VIP)-like immunoreactivity in major arteries of postmortem adult human circles of Willis. Calcitonin gene-related peptide-, NPY-, and VIP-LI exhibited a variety of varicose and nonvaricose single axons, and small and large perivascular nerve bundles. Although the density of innervation within each vascular segment was highly variable, the pattern of innervation for each neuropeptide observed was consistent throughout the circle of Willis. With the use of human and rat circles of Willis as positive control preparations, the lack of CGRP-LI, NPY-LI, and VIP-LI in vessel segments taken from five cases of intracranial arteriovenous malformations (AVMs) is also reported. It is concluded that adult human circles of Willis exhibit CGRP-LI, NPY-LI, and VIP-LI perivascular nerves. In addition, intracranial AVMs do not possess these peptide-containing nerves that, in animals, normally mediate neurogenic control in the cerebrovasculature. It is hypothesized that this lack of innervation, and hence neurotrophic influence, may contribute to the development of AVMs.

NADPH-diaphorase-containing cerebrovascular nerve fibres and their possible origin in the pig.

NADPH-diaphorase histochemical technique was applied to demonstrate the catalytic activity of nitric oxide synthase (NOS) in the nerve fibers supplying some porcine cerebral blood vessels, as well as in ganglia thought to be their possible sites of origin. Five sexually mature Large With Polish race sows were used. The following blood vessels and their branches were studied: the basal cerebral artery, medial cerebral artery, arteries of the circle of Wills as well as arteries located in the arachnoidea. The activity of NADPH-d was visualised in whole-mount preparations from the above listed blood vessels. The presence of NADPH-d was additionally studied in the ganglia of trigeminal nerves, the sensory ganglia of vagus nerves, the pterygopalatine and cranial cervical ganglia. NADPH-d activity was found in nerve fibres supplying all the studied arteries. Larger blood vessels, the basal cerebral artery, medial cerebral artery and arteries of the circle of Wills possessed very dense NADPH-d-positive nerve plexuses while arachnoidal arteries were poorly innervated by only single nerves. The vascular nerve fibers formed bundles varying in thickness, from very thick bundles often interchanging nerve fibers to quite thin fascicles. Thick bundles were absent from the walls of medium sized vessels and small meningeal arteries where only smaller fascicles or single fibres occurred. NADPH-d-positive neurons and nerve fibres were found in all the ganglia investigated. However, pronounced differences in the number of the positive nerve structures were observed between the ganglia. In the pterygopalatine, trigeminal and sensory ganglia of the vagal nerve the vast majority of neurons were NADPH-d-positive. Numerous NADPH-d-positive nerve fibers occurred within the pterygopalatine and trigeminal ganglion while sensory ganglia of the vagal nerve comprised smaller number of fibres. Small numbers of the neurons and moderate numbers of the nerve fibres occurred in the cranial cervical ganglion. The intensity of NADPH-d reaction in the endothelium was constant independent of the size of the vessels studied.

Distribution and origins of cerebrovascular NADPH-diaphorase-containing nerve fibers in the rat.

Neuronal NADPH-diaphorase has been proved to be nitric oxide synthase itself. In this study, we investigated distribution and origins of NADPH-diaphorase-containing nerve fibers in the cerebral vessels in the rat. Adult male Sprague-Dawley rats were divided into 4 groups. Nasociliary nerves were transected bilaterally in group 1. In group 2, intracranial branches of the sphenopalatine ganglion were transected bilaterally. In group 3, both of these structures were transected. The remaining animals were served as control (group 4). Two weeks after the above procedures, they were perfused with paraformaldehyde and glutaraldehyde. The pial arteries and superior cervical, trigeminal, internal carotid, otic and sphenopalatine ganglia were dissected. All specimens were processed for NADPH-diaphorase histochemistry. Numerous NADPH-diaphorase-containing nerve fibers with varicosities forming plexuses were observed in the circle of Willis and its branches. Relatively thick nerve bundles were noted in the anterior half of the circle of Willis. They are most abundant in the internal ethmoidal artery. Approximately 5% of such fibers in anterior half of the circle of Willis disappeared in group 1, 90% in group 2, and no fibers were seen to remain in group 3. NADPH-diaphorase reaction was positive in the neurons of sphenopalatine, otic trigeminal and internal carotid ganglia. Among these ganglia, the reaction was prominent in sphenopalatine, otic and internal carotid ganglia. In summary: (1) NADPH-diaphorase-containing nerve fibers distribute to the circle of Willis and its branches.(ABSTRACT TRUNCATED AT 250 WORDS)

Distribution and pathway of the cerebrovascular nerve fibers from the otic ganglion in the rat: anterograde tracing study.

The distribution and pathway of cerebrovascular nerve fibers from the otic ganglion were studied by an anterograde tracing technique in the rat. Wheat germ agglutinin-horseradish peroxidase was injected as an anterograde axonal tracer into the otic ganglion on one side. Forty-eight hours later, the animals were killed and specimens were reacted with tetramethylbenzidine. Wheat germ agglutinin-horseradish peroxidase positive fine nerve fibers were observed in the circle of Willis and its branches, i.e., anterior cerebral artery, middle cerebral artery, internal ethmoidal artery and posterior cerebral artery, while no positive fiber could be detected in the vertebrobasilar artery. A positive reaction with tetramethylbenzidine was also observed in the lesser superficial petrosal nerve, the greater superficial petrosal nerve, the vidian nerve, the greater deep petrosal nerve, the internal carotid ganglion and the trigeminal ganglion. The sphenopalatine ganglion, however, failed to reveal any positive neurons or nerve fibers. It is concluded that the cerebrovascular nerve fibers originating from the otic ganglion run along the lesser superficial petrosal nerve to join the greater superficial petrosal nerve. They then reach the greater deep petrosal nerve and ascend along the internal carotid artery to distribute themselves to the cerebral blood vessels. This study demonstrated, for the first time, that the otic ganglion innervates the cerebral vessels and elucidated the pathway from the otic ganglion to the cerebral vessels directly by means of an anterograde axonal tracing technique.

Trigeminal sensory innervation on perforators of the circle of Willis in rabbits by wheat germ agglutinin-conjugated horseradish peroxidase anterograde tracing.

Distribution patterns of sensory innervation from the trigeminal ganglion to the perforators of the circle of Willis in rabbits were investigated by wheat germ agglutinin-conjugated horseradish peroxidase (WGA-HRP) anterograde tracing. Twenty Japanese white rabbits were anesthetized by inhaling 1% halothane. Using a microsurgical technique, 4 microliters of 2% WGA-HRP in 1 M KCl solution, colored with brilliant blue, was micro-injected into the medial part of the left trigeminal ganglion in 14 animals with a pressure injection system. Another six served as controls to exclude the possibility of labeling non-trigeminal axons. Forty-eight hours later, the perforators in the cisternal and intracerebral segments along with their parent arteries were dissected from the brain according to Dacey's dissecting technique after transcardial perfusion, reacted with the 3,3',5,5'-tetramethyl benzidine method of Mesulam. The results revealed that sensory nerves on the perforators of the circle of Willis were less densely innervated than those on their parent arteries due to the difference in diameter. The posteromedial perforating arteries arising from the P1 segment of the posterior cerebral artery to the tegmentum, posteroventral thalamus and posterior hypothalamus were more prominently and consistently innervated than other perforators. The sensory fibers were seen on the cisternal segment of the perforating arteries. A parallel or twisted pattern was found in the perforators less than 100 microns in diameter, while a meshwork pattern was visualized in the proximal part of some bigger ones. Fine sensory fibers could be traced on the perforators as small as 40 microns in diameter.(ABSTRACT TRUNCATED AT 250 WORDS)

Possible origins and distribution of immunoreactive nitric oxide synthase-containing nerve fibers in cerebral arteries.

The distribution of perivascular nerve fibers expressing nitric oxide synthase (NOS)-immunoreactivity was examined in Sprague-Dawley and Long-Evans rats using affinity-purified rabbit antisera raised against NOS from rat cerebellum. NOS immunoreactivity was expressed within the endothelium and adventitial nerve fibers in both rat strains. Labeled axons were abundant and dense in the proximal anterior and middle cerebral arteries, but were less numerous in the caudal circle of Willis and in small pial arteries. The sphenopalatine ganglia were the major source of positive fibers in these vessels. Sectioning postganglionic parasympathetic fibers from both sphenopalatine ganglia reduced the density of NOS-immunoreactive (IR) nerve fibers by > 75% in the rostral circle of Willis. Moreover, NOS-IR was present in 70-80% of sphenopalatine ganglion cells. Twenty percent of these neurons also contained vasoactive intestinal polypeptide (VIP)-immunoreactivity. By contrast, the superior cervical ganglia did not contain NOS-IR cells. In the trigeminal ganglion, NO-IR neurons were found chiefly within the ophthalmic division; approximately 10-15% of neurons were positively labeled. Colocalization with calcitonin gene-related peptide (CGRP) was not observed. Sectioning the major trigeminal branch innervating the circle of Willis decreased positive fibers by < or = 25% in the ipsilateral vessels. In the nodose ganglion, 20-30% of neurons contained NOS-immunoreactivity, whereas less than 1% were in the C2 and C3 dorsal root ganglia. Three human circles of Willis obtained at autopsy showed sparse immunoreactive fibers, chiefly within vessels of the posterior circulation. Postmortem delay accounted for some of the reduced density. Our findings indicate that nerve fibers innervating cerebral arteries may serve as a nonendothelial source of the vasodilator nitric oxide (NO). The coexistence of NOS and VIP within sphenopalatine ganglion cells raises the possibility that two vasodilatory agents, one, a highly diffusable short-lived, low-molecular-weight molecule, and the other, a polar 28 amino acid-containing peptide, may serve as coneuromediators within the cerebral circulation.

Contributions from the upper cervical dorsal roots and trigeminal ganglia to the feline circle of Willis.

To further define the sensory projections to the circle of Willis, we measured concentrations of immunoreactive substance P in pial arteries of cats following either bilateral removal of the C1-3 dorsal root ganglia (six cats) or bilateral removal of the trigeminal ganglia (three cats). Removal of the dorsal root ganglia decreased concentrations of the tachykinin substance P in the vertebral artery and the basilar artery and its branches by 72% and 50-66%, respectively. Bilateral removal of the trigeminal ganglia decreased substance P concentrations in all forebrain vessels including the rostral basilar artery, although only concentrations in the anterior cerebral artery were significantly lower than those in unilaterally lesioned cats (p less than 0.01). Hence, the vertebrobasilar artery and its tributaries are invested by substance P-containing fibers originating from the upper cervical dorsal root ganglia, and the anterior cerebral artery is innervated by both trigeminal ganglia. If a similar anatomy exists in humans, our data provide an explanation for the occipital localization of headaches arising from the vertebrobasilar arteries and for bilateral headaches following stimulation of the anterior cerebral artery.

Pain mechanisms underlying vascular headaches. Progress Report 1989.

Vascular headaches are among the most prevalent yet poorly understood problems in clinical neurology. Headaches may develop in association with hypertension, seizures, stroke or without a recognizable pathophysiology such as during migraine and cluster headaches. Cephalic blood vessels (pial and dural vessels) are implicated as the most important source for all headaches and are innervated by sensory fibers which arise from ganglia innervating the forehead, scalp and neck. Sensory fibers contain vasoactive neuropeptides which become released from peripheral (perivascular) and central terminations to mediate vasodilation and pain, respectively. The presence of vascular headache implies activation of this final common pain pathway which we have termed the trigeminovascular system. The presence of vascular headache implies activation of this final common pain pathway which we have termed the trigeminovascular system. The existence of such a system a) clarifies certain pain patterns which develop following stimulation of cephalic blood vessels, b) suggests a mechanism to explain the referral of pain to the forehead, c) provides a mechanism to explain the action of certain antimigraine drugs, d) suggests a local mechanism which enhances blood flow under certain pathological conditions. Hence, this review will update existing knowledge about the trigeminovascular system and its role in headache pathophysiology.

Distribution and termination of trigeminal nerves to the cerebral arteries in monkeys.

Wallerian degeneration was used to study the contributions of the first and second divisions of the trigeminal nerve to cerebral arterial innervation in the cynomolgus monkey. Animals were killed by intracardiac perfusion of fixative three to seven days after left ophthalmic or maxillary neurotomy or a combination of both, using three animals for each procedure. Cerebral arteries were dissected, removed and prepared for light and electron microscopy. The anterior vessels of the circle of Willis received nerve fibres, distributed via the internal carotid artery, from the ipsilateral ophthalmic branch of the trigeminal nerve and a small maxillary contribution was also observed in some animals. The posterior vessels were supplied from the same trigeminal source but by a different route, the nerves moving onto the basilar artery bilaterally or unilaterally via the recurrent nerve of the cavernous plexus using the abducent nerve for access. From the basilar, fibres distributed to both posterior cerebral arteries. Augmentation of the vascular nerve supply apparently from branches of the vagus and/or hypoglossal nerves was noted but otherwise unexamined. Trigeminal terminals were found on all vessels of the circle of Willis and their distal branches throughout the thickness of the adventitia, often lying close to the media but never contacting smooth muscle cells. These observations are consistent with results from other studies employing neurohistochemical and tracer techniques in subprimates. Comparison of operated and control material failed to reveal any distinctive features of terminals attributable to a trigeminal source.

Co-existence of immunoreactivity to neuropeptide Y and vasoactive intestinal peptide in non-noradrenergic axons innervating guinea pig cerebral arteries after sympathectomy.

We have used double-labelling immunofluorescence to examine the coexistence of immunoreactivity (IR) to neuropeptide Y (NPY), tyrosine hydroxylase (TH) and vasoactive intestinal peptide (VIP) in autonomic neurons innervating guinea pig cerebral arteries. In the rostral circle of Willis of control animals. NPY-IR was detected in 86% of axons with TH-IR (noradrenergic) and 18% of VIP-IR (non-noradrenergic) axons. No axons contained both VIP-IR and TH-IR. Ten to 12 days after bilateral removal of the superior cervical ganglia all TH-IR axons had disappeared. The density of VIP-IR axons was unchanged but now 70% of VIP-IR axons contained NPY-IR. These results show that NPY is not exclusively associated with noradrenergic axons in the cerebral vasculature. Furthermore, NPY levels in non-noradrenergic axons increased following sympathetic denervation.

Concentrations of putative neurovascular transmitters in major cerebral arteries and small pial vessels of various species.

The levels of noradrenaline, neuropeptide Y, 5-hydroxytryptamine, and substance P were measured and compared between the large arteries of the circle of Willis and the small cerebral vessels of the pia mater in the rat, rabbit, cat, and monkey. In all species, noradrenaline and neuropeptide Y concentrations were greater in the larger arteries than in small pial vessels. Noradrenaline concentrations were dramatically reduced following cervical sympathectomy, with the extent of diminution differing greatly in the various species; the effects of cervical ganglionectomy on neuropeptide Y concentrations were less pronounced. 5-Hydroxytryptamine concentrations in rats, cats, and rabbits were significantly greater in the small pial vessels, although measurable concentrations existed in the circle of Willis. In cats and monkeys, substance P was found in major arteries, but was not detectable at the level of the small pial vessels. The differences in the regional distribution of the various neurotransmitter candidates in the cerebrovascular bed may reflect their physiological significance.

Peptide-containing nerve fibers in human cerebral arteries: immunocytochemistry, radioimmunoassay, and in vitro pharmacology.

Nerve fibers containing neuropeptide Y, vasoactive intestinal peptide (VIP), substance P (SP), and calcitonin gene-related peptide (CGRP) were seen in the adventitia or at the adventitia-media border of human cerebral arteries obtained during neurosurgical procedures. Radioimmunoassay of human cerebral arteries, removed at autopsy, revealed that the levels of the four peptides did not differ among the major cerebral arteries. There was, however, a gradual decline in peptide concentrations with increasing age of the patients, as measured in the proximal part of the middle cerebral artery. Pharmacological experiments on fresh segments of cerebral (pial) arteries in vitro revealed that neuropeptide Y caused vasoconstriction per se but did not potentiate the contractile response of noradrenaline. VIP, peptide histidine methionine-27 (PHM-27), SP, neurokinin A (NKA), and human CGRP potently relaxed vessels precontracted by prostaglandin F2 alpha, the relative potency being human CGRP greater than SP greater than VIP greater than NKA greater than PHM-27. The amount of relaxation varied between 55% (SP) and 96% (human CGRP) of the prostaglandin F2 alpha-induced contraction. The peptide effects were not antagonized by propranolol, atropine, or cimetidine, suggesting an action that does not involve adrenergic, cholinergic, or histaminergic receptors.

Substance P-like immunoreactivity in rat forebrain leptomeninges and cerebral vessels originates from the trigeminal but not sympathetic ganglia.

Substance P-like immunoreactivity (SPLI) is present in the pia mater, arachnoid and pial vessels (leptomeninges) of the rat. Unilateral electrolytic lesions of the trigeminal ganglion decreased levels of SPLI in leptomeninges on the lesioned side. Levels did not change following unilateral or bilateral superior cervical ganglionectomies or after i.c.v. injections of 6-hydroxydopamine, sufficient to deplete norepinephrine in pial arteries by more than 90%. SP levels did not decrease in leptomeninges or in blood vessels within the circle of Willis following treatment of adult or neonatal rats with capsaicin despite the fact that levels were reduced in the cornea and dorsal spinal cord in these same animals. These results suggest that not all substance P-containing sensory fibers are susceptible to the peptide-depleting effects of capsaicin.

Origin and distribution of cerebral vascular innervation from superior cervical, trigeminal and spinal ganglia investigated with retrograde and anterograde WGA-HRP tracing in the rat.

Peripheral sources of cerebral vascular innervation have been investigated with retrograde and anterograde neuronal tracing of wheat germ agglutinin conjugated with horseradish peroxidase (WGA-HRP) in the rat. For retrograde identification of sources of innervation, WGA-HRP was applied to the exposed basilar artery through a fine slit in the overlying meninges, and sections of brain and peripheral ganglia were reacted with tetramethylbenzidine for detection of the tracer. A high density of tetramethylbenzidine reaction product was observed around the basilar artery and in the surrounding pial tissue, but the application sites were not completely selective since some tracer always had spread into the ventral brain stem. Retrogradely labelled cell bodies were identified in the superior cervical, stellate, first and second spinal, and trigeminal ganglia, i.e. these ganglia may represent origins of basilar artery innervation. In a second series of experiments, microinjections of WGA-HRP were placed into the indicated ganglia to obtain anterograde labelling of nerve fibres on whole-mounts of the cerebral vessels. Injections into trigeminal ganglia labelled nerve fibres on the ipsilateral half of the circle of Willis, as well as the contralateral anterior cerebral artery and the rostral part of the basilar artery. The first and second spinal ganglia projected to the vertebrobasilar arteries, while the ipsilateral part of the internal carotid (outside the circle of Willis) received fibres from the second spinal ganglion. Nerve fibres originating in trigeminal and spinal ganglia were organised in bundles, and between these a sparse plexus of thin single fibres appeared. Injection of WGA-HRP into superior cervical ganglion labelled a plexus of nerve fibres on the ipsilateral circle of Willis and the (rostral) basilar artery. These experiments demonstrated the origin and distribution of sympathetic and sensory innervation to major cerebral arteries in the rat.