331
Pathways of Parasympathetic and Sensory
Cerebrovascular Nerves in Monkeys
Jan Erik Hardebo, MD; Mohammed Arbab, MD;
Norihiro Suzuki, MD; and Niels Aage Svendgaard, MD
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Using immunohistochemistry, we studied the origins and pathways of parasympathetic and
sensory nerve fibers to the pial arteries in four squirrel monkeys. Following its application to
the surface of the middle cerebral artery, the retrograde axonal tracer True Blue accumulated
in parasympathetic neurons of the sphenopalatine ganglion and the internal carotid ganglion.
The latter is strategically located where the internal carotid artery enters the cranium. Fibers
from the sphenopalatine ganglion reach the internal carotid artery in the cavernous sinus
region after running as rami orbitales. Before reaching the internal carotid artery, the fibers
bypass aberrant sphenopalatine ganglia, with the most distant, the cavernous ganglion, being
located in the cavernous sinus region. True Blue also accumulated in sensory neurons of the
ophthalmic and maxillary divisions of the trigeminal ganglion and in sensory neurons of the
internal carotid ganglion. Fibers from the ophthalmic division of the trigeminal ganglion reach
the internal carotid artery as a branch through the cavernous sinus, bypassing the cavernous
ganglion. Fibers from the maxillary division also bypass the cavernous ganglion after reaching
it via a recurrent branch of the orbitociliary nerve. Thus, the cavernous ganglion forms a
confluence zone for parasympathetic and sensory fibers in the region. In addition, parasympathetic and sensory fibers leave the confluence zone to follow the abducent and trochlear
nerves backward to the basilar artery and tentorium cerebelli, respectively. Clinical implications are discussed. (Stroke 1991^2:331-342)
B
lood vessels of the brain surface, and to some
extent those of its parenchymal branches,
are supplied with sympathetic, parasympathetic, and sensory nerve fibers, spreading like a
network in the adventitia.1 It has been long known
that all sympathetic fibers (except those in the caudal
basilar artery in some species) originate in the superior cervical ganglion.1 An origin for parasympathetic
fibers in the sphenopalatine and otic ganglia has been
demonstrated in rats and cats by the retrograde axonal
tracer technique, by nerve section, and by histochemical mapping of the putative transmitters acetylcholine
(cholinesterase, choline acetyltransferase [ChAT])
and vasoactive intestinal polypeptide (VIP).2-5 By this
latter technique (histochemical staining of the putative transmitters substance P and calcitonin geneFrom the Departments of Medical Cell Research (J.E.H., N.S.),
Neurology (J.E.H.), and Neurosurgery (M.A., N.A.S.), University
Hospital of Lund, Sweden.
Supported by grants from the Swedish Medical Research Council (14X-732/5680 and 04X-09045), Greta and Johan Kock's Foundation, Sweden, and the National Institutes of Health (IROINS
23722-01A1).
Address for reprints: J.E. Hardebo, Department of Medical Cell
Research, University of Lund, Biskopsgatan 5, S-223 62 Lund,
Sweden.
Received September 17, 1990; accepted November 6, 1990.
related peptide [CGRP]), an origin for supratentorial
cerebrovascular fibers in the trigeminal ganglion, primarily its ophthalmic division, has been demonstrated
in rats, cats, and monkeys.5-14 Parasympathetic and
sensory fibers in the basilar artery have their cell
bodies in the vagal and upper cervical dorsal root
ganglia and possibly in the sphenopalatine ganglion in
rats and cats,15-17 and in cynomolgus monkeys an
additional supply from the trigeminal ganglion has
been demonstrated.13 In rats, an additional source for
cerebrovascular parasympathetic and sensory nerves
has been identified, namely the internal carotid ganglion, located along the greater superficial petrosal
nerve pathway to the sphenopalatine ganglion, at its
junction with the sympathetic internal carotid nerve
on the surface of the internal carotid artery.4-512-18 The
presence of the internal carotid ganglion has also been
described in humans and monkeys.19-20 One further
possible source for innervation of the cerebrovascular
nerves in humans and monkeys is the miniganglia
described to be lateral to the internal carotid artery in
the matrix between vessels of the cavernous sinus.13-19-21 By using a combined retrograde axonal
tracer and immunohistochemical transmitter staining
technique, we studied the precise origins and probable
pathways of parasympathetic and pain fibers to vessels
on the brain surface in monkeys.
332
Stroke Vol 22, No 3 March 1991
SPG
OCN
ophth.a.
dural penetrance of "
FIGURE 1. Schematic drawing
of right half of skull base of
squirrel monkey seen from
above. Dura of middle fossa has
been removed. n.TV runs rostrally in dural edge above n.V
and internal carotid artery
(dashed line) to join n. VI within
cavernous sinus; only its dural
entrance is shown. Inset: Relations between nerves and vessels
at coronal section A. SPG, sphenopalatine ganglion; RO, rami
orbitales; OCN, orbitociliary
nerve from n.V:2; VN, vidian
nerve; ICG, internal carotid ganglion; GSPN, greater superficial
petrosal nerve; ICA, internal carotid artery; CG, cavernous ganglion; opth. a., ophthalmic artery.
hypophyseal
fossa
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
SPN
OCN
+R0
Materials and Methods
We used four squirrel monkeys (Saimiri sciureus)
weighing 650-780 g. Anesthesia was initiated by the
inhalation of 4% isoflurane (Abbott Laboratories,
North Chicago, 111.) followed by 0.05 mg/kg i.m.
atropine. The monkeys were intubated, and anesthesia was maintained with 2% isoflurane in 70% N2O
and 30% O2. In two animals a left-sided frontotemporal craniotomy was performed under sterile conditions. The dura mater was incised, and the arachnoid
just distal to the trifurcation of the middle cerebral
artery was opened. A small piece of polyethylene film
was inserted between the brain cortex and the three
branches of the middle cerebral artery. A small piece
of Spongostan (Ferrosan, Copenhagen, Denmark)
preabsorbed with 25 y\ of the retrograde axonal
intracranial
entrance of ICA
tracer True Blue22 was then applied on the exposed
segments of the middle cerebral artery. The opened
arachnoid was made to adhere to the surface of the
piece of Spongostan, and another piece of polyethylene film was placed over the application site to
prevent contact with the dura of the water-soluble
dye. The dura was closed water-tight, the bone flaps
were replaced, and the skin was sutured. Pilot experiments in two monkeys revealed that the tracer does
not accumulate in cranial ganglia after its application
to an area of the brain surface devoid of larger
branches of the middle cerebral artery.
Six days after the application of True Blue, the two
animals were anesthetized with 14 mg/kg i.m. alfaxalone-alfadolone acetate followed by 10 mg/kg i.v.
thiopental. Their chests were opened, and the mon-
Hardebo et al
Cerebrovascular Nerves in Monkeys
333
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 2. Coexistence of vasoactive intestinal pofypeptide (VIP) and choline acetyltransferase (ChAT) (a and b) and of
substance P (SP) and calcUonin gene-related peptide (CGRP) (c and d) in perivascular nerve fibers in small pial arteries of squirrel
monkey. Fluorescence immunohistochemistry. Scale bars=50 \im.
keys were perfused through the left ventricle with
saline followed byfixationwith an ice-cold mixture of
2% paraformaldehyde and 15% saturated aqueous
picric acid solution in 0.1 M phosphate buffer (pH
7.2). The following tissues were dissected out: the
superior cervical, ciliary, sphenopalatine, otic, genie-
334
Stroke
Vol 22, No 3 March 1991
TABLE 1. Number of Cells in Various Ipsllateral Cranial Ganglia
That Accumulated True Blue After Application to Trifurcation of
One Middle Cerebral Artery in Squirrel Monkeys
Monkey
Ganglion
Superior cervical
Ciliary
Sphenopalatine
Aberrant sphenopalatine
Cavernous
Otic
Internal carotid
Geniculate
Trigeminal Ophthalmic division
Maxillary division
Glossopharyngeal
Vagal
1
2
52
0
11
0
0
0
16
0
41
10
0
0
47
0
8
0
0
0
11
0
33
6
0
0
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
ulate, trigeminal, superior and inferior glossopharyngeal, and vagal ganglia on both sides; the contents of
the cavernous sinus region on the left side; the left
internal carotid artery (at its entrance into the cranium through the carotid canal) with the internal
carotid ganglion and adjacent segments of the
greater superficial petrosal and vidian nerves (Figure
1); the branches of the left orbitociliary nerve with its
accompanying rami orbitales (Figure 1) running from
the maxillary nerve to the medial infraorbital fissure;
the nasociliary nerve and accompanying structures in
the ethmoidal foramen on both sides; the basal
frontal dura on both sides; the tracer application sites
with the underlying brain and the overlying dura
mater; and the remaining pial vessels. In the other
two animals, perfusion-fixed as above, the abducent
and trochlear nerves were dissected out separately
along their whole intracranial courses from the brain
stem to their exits through the supraorbital fissure.
The tissues were immersed in fixative for 2 hours
and thoroughly rinsed for 24 hours in Tyrode's
solution containing 10% sucrose at 4°C. The pial
vessels were spread out on glass slides. The other
preparations were frozen and serially sectioned at
10-^.m thicknesses in a cryostat. The cavernous sinus
region was sectioned in the frontal plane, and the
proximal internal carotid artery was sectioned at
right angles to its length axis.
The sections and spread preparations were processed for imrnunohistochernical demonstration of
dopamine /3-hydroxylase, ChAT, VIP, substance P,
and CGRP with the indirect immunofluorescence
method.23 The dopamine /3-hydroxylase antiserum
was raised in rabbits (Lot No. 2012, Eugene Tech,
Allendale, N.J.) and used in a 1:320 dilution. Monoclonal ChAT antiserum was raised in rats (1E6, P.M.
Salvaterra,24 Duarte, Calif.) and used in a 1:20
dilution. The VIP antisera were raised in rabbits or
guinea pigs (Code No. 7854 or 8701, respectively;
Milab, Malmo, Sweden) and used in 1:320 or 1:1,280
dilutions. Rabbit substance P antiserum was supplied
by Milab (Code No. 8127) and used in a 1:160
dilution, and monoclonal rat substance P antiserum
was supplied by Seralab, Stockholm, Sweden (Code
No. NC1/34HL) and used in a 1:160 dilution. Rabbit
CGRP antiserum (Code No. 8425, Milab) and guinea
pig CGRP antiserum (BGP 470-1, Milab) were used
in 1:640 dilutions.
The specimens were exposed to the peptide antiserum for 24 hours at 4°C in a moist chamber. The
site of the antigen-antibody reaction was revealed by
the application of fluorescein isothiocyanate (FTTC)labeled pig anti-rabbit or goat anti-mouse immunoglobulin G (IgG) (Dakopatts, Copenhagen, Denmark
and Sigma Chemical Co., St. Louis, Mo., respectively) in a 1:20 or 1:80 dilution, respectively. For
double-staining we also used tetramethylrhodamine
isothiocyanate (TRITC)-labeled goat anti-rabbit IgG
(Milab) and FITC-labeled goat anti-guinea pig IgG
(Sigma) in 1:320 and 1:160 dilutions, respectively.
All antisera contained 0.3% Triton X-100 and 0.25%
bovine serum albumin (for the ChAT antiserum
Triton X-100 was excluded). Control sections were
exposed to antisera that had been preabsorbed with
excess antigen (10-100 pg synthetic or natural peptide per milliliter diluted antiserum) and displayed
no immunoreactivity.
The slides were examined in an epi-illumination
fluorescence microscope fitted with filter settings
appropriate for True Blue, FITC, or TRITC fluorescence. The number of True Blue-positive cells in the
ganglia were counted on serial sections, avoiding
recounting of the same cell on consecutive sections.
Every fifth section of the frontal dura was stained
with hematoxylin and eosin.
Results
In the first two monkeys, a network of fiber bundles
and individual nerve fibers positive for VIP, ChAT,
substance P, and CGRP was seen in even the smallest
pial vessels contained in the spread preparations
(diameters of approximately 50 /tm) (Figure 2).
ChAT displayed a weaker fluorescence than the
other peptides. All ChAT-positive fibers also contained (or ran in close association with fibers containing) VIP, whereas approximately 60% of the VIPpositive fibers were ChAT-positive. Almost all
substance P-positive fibers also contained (or ran in
close association with fibers containing) CGRP,
whereas about 80% of the CGRP-positive fibers were
substance P-positive. In the intracranial segment of
the internal carotid artery, fiber bundles of
VIP+ChAT- and CGRP+substance P o s i t i v e fibers were seen as were a few delicate varicose fibers
positive for VIP, CGRP, and substance P apparently
innervating the artery. The pial arterial wall at the
site of True Blue application was intensely labeled,
and a limited spread of the dye was seen to areas
immediately surrounding the artery (i.e., the pia
mater and underlying cortex). No dye was seen in the
overlying dura mater.
Hardebo et al
Cerebrovascular Nerves in Monkeys
335
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 3. Following its application to ipsilateral middle cerebral artery, True Blue (TB) accumulated in three neurons (arrows)
of sphenopalatine ganglion (SPG) (a) that were also positive for vasoactive intestinal polypeptide (VIP) (b). Most neurons in this
ganglion are VIP-positive. In rostral division of internal carotid ganglion (ICG), TB was found in two neurons (c) that were
VIP-positive (d). Fluorescence immunohistochemistry. Scale bars=50 urn.
336
Stroke
Vol 22, No 3 March 1991
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 4. Following its application to ipsilateral middle cerebral artery, True Blue (TB) was found in neuron (a) of ophthalmic
division of trigeminal ganglion (TG) that contained both substance P (SP) (b) and calcitonin gene-related peptide (CGRP) (c).
In rostral division of internal carotid ganglion (ICG), TB was found in neurons (d) that were positive for both vasoactive intestinal
peptide (VIP) (e) and choline acetyltransferase (ChAT) (f). Scale bars=50 urn.
Hardebo et al
Cerebrovascular Nerves in Monkeys
337
FIGURE 5. Most neurons in cavernous ganglion were positive
for vasoactive intestinal polypeptide (VIP) (a and b). Some
were also positive for choline acetyltransferase (ChAT) (arrows) (b and c). Scale bars=50 \an.
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
True Blue accumulated in neurons of the superior
cervical, sphenopalatine, internal carotid, and
trigeminal ganglia of both monkeys (Table 1) but not
in the ciliary, cavernous, otic, geniculate, glossopharyngeal, or vagal ganglia or the aberrant sphenopalatine ganglia along the rami orbitales. Within the
338
Stroke
Vol 22, No 3 March 1991
r\
ICA
n.1
n.Y2
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 6. Schematic figure of principal pathways connecting parasympathetic sphenopalatine ganglion (SPG) and cavernous ganglion (CG) with internal carotid artery (ICA), n. VI,
and (possibly) n.IV, and pathways connecting sensory trigeminal ganglion's ophthalmic (n.V:l) and maxillary (n.V:2)
divisions with ICA, n. VI, and (possibly) n.IV. Fibers between
SPG and CG are named rami orbitales (RO), and fibers
between n.V:2 and CG are named orbitociliary nerve (OCN).
Although not found in our present experiment, connections
between ICA, n. V: 1, SPG, and n. VI may be severalfold in
individual monkeys, and CG may be scattered in multiple
miniganglia along these connections'3 so that actual nerve
plexus is formed within matrix between venous vessels of
cavernous sinus. Aberrant sphenopalatine ganglia along RO
are not indicated.
trigeminal ganglion True Blue accumulated in both
the ophthalmic and maxillary divisions. Almost all
True Blue-positive cells in the superior cervical
ganglion were positive for dopamine £-hydroxylase,
and in the sphenopalatine ganglion the same was true
for VIP. In this ganglion approximately 60% of all
neurons were positive for VIP (Figure 3b). About
30% of the True Blue+VLP-positive cells were also
positive for ChAT. In the trigeminal ganglion approximately 60% and 70% of the True Blue-positive cells
were also positive for substance and CGRP, respectively (Figures 4b and 4c). The internal carotid
ganglion (Figure 1) consisted of a larger group (about
30) of VIP-positive cells (Figure 3d), several of which
also contained ChAT (Figures 4e and 4f), located
rostral to a group of about 10 cells that were CGRPand substance P-positive. Double-staining with VIP
and CGRP revealed that the larger group of cells did
not contain CGRP and that the smaller group of cells
did not contain VIP. All True Blue-positive cells in
the internal carotid ganglion were either
VIP+ChAT-positive (Figures 3c and 3d; Figures 4d,
4e, and 4f) or CGRP+substance P-positive.
Among the ganglia that did not accumulate the
tracer, a large percentage of VIP-positive cells were
found in the cavernous, otic, glossopharyngeal, and
vagal ganglia (Figures 5 a and 5b) and the aberrant
sphenopalatine ganglia. Several VIP-positive neu-
rons were also ChAT-positive (as demonstrated for
the cavernous ganglion in Figures 5b and 5c). Only
one cavernous ganglion was found within the cavernous sinus in each monkey (see References 13,19, and
21). The location and probable connections of the
cavernous ganglion are indicated in Figures 1 and 6.
None of the approximately 30 cells in this ganglion
were CGRP- or substance P-positive (every second
section was double-stained for either VIP+ChAT or
CGRP+substance P).
Aberrant VIP+ChAT-containing ganglia (negative for CGRP and substance P) and VIP+ChATpositive fibers were found along the whole course of
the rami orbitales. The rami orbitales appeared to
mingle with fibers of the orbitociliary nerve branch
on their course from the sphenopalatine ganglion
toward the medial infraorbital fissure. No neurons
were found directly on the surface of the internal
carotid artery. Several CGRP + substance P-positive
cells were found in the upper glossopharyngeal and
vagal ganglia, and a few such cells were observed in
the geniculate ganglion.
In the other two monkeys, fibers positive for both
VIP and ChAT or substance P in the abducent nerve
and for substance P in the trochlear nerve were seen
near the margin of these nerves from about halfway
along their intracavernous courses and caudally
toward their origins from the brain stem. In the
caudal intracavernous part of the abducent nerve a
few CGRP+substance P-positive neurons were seen,
and in one monkey a single VIP+ChAT-positive
neuron was found.
No VIP- or ChAT-positive fibers were seen in the
nasociliary nerve or other structures passing through
the ethmoidal foramen, except in the adventitia of
blood vessels. No VIP-, ChAT-, CGRP-, or substance
P-positive fibers bundles could be found in the basal
frontal dura from the level of the ethmoidal foramen
and caudally. Also, no fiber bundles could be observed in hematoxylin and eosin-stained sections of
the basal frontal dura.
Discussion
We confirm in monkeys that perivascular pial
sympathetic nerves originate in the superior cervical
ganglion and that there is no contribution of nerve
branches from the contralateral ganglion at the level
of the trifurcation of the middle cerebral artery.
One major finding is that parasympathetic fibers to
this part of the pial vasculature originate in the
ipsilateral sphenopalatine and internal carotid ganglia. According to the number of True blue-labeled
cells, these two ganglia contribute to innervation to
similar extents. The probable pathways are summarized in Figure 7. The pathway from the sphenopalatine ganglion to the pial vessels is not identical to that
found in rats, where the fibers take a short course on
the medial orbital wall up to the ethmoidal foramen
to enter the skull and follow the internal ethmoidal
artery back to the anterior part of the circle of
Willis.4-5 No such fiber connection was found in
Hardebo et ai Cerebrovascular Nerves in Monkeys
339
MCA
ACA
caudal
cranial
nerves? n.lZII
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 7. Schematic overview of origins and pathways ofparasympathetic fibers to right intracranial internal carotid artery and
pial arteries in monkeys. ACA, anterior cerebral artery; BA, basilar artery; CG, cavernous ganglion; DPN, deep petrosal nerve; EF,
ethmoidal foramen; GSPN, greater superficial petrosal nerve; ICA, internal carotid artery; ICG, internal carotid ganglion; ICN,
internal carotid nerve; LSPN, lesser superficial petrosal nerve; MCA, middle cerebral artery; NCN, nasociliary nerve; OCN,
orbitociliary nerve; OTG, otic ganglion; PCA, posterior cerebral artery; RO, rami orbitales; SCG, superior cervical ganglion; VN,
vidian nerve.
monkeys. Instead, a highly likely course for the fibers
from the sphenopalatine ganglion is via rami orbitales,25 which in monkeys join a recurrent branch of
the orbitociliary nerve from the maxillary nerve to
reach the cavernous sinus region (via the medial
infraorbital fissure)13 (Figures 1 and 6). Direct documentation of such a pathway may be obtained from
studies in which True Blue application is combined
with nerve section. In the cavernous sinus region a
plexus-like nerve formation has been demonstrated
in cynomolgus monkeys,13 with fiber branches to/
from the adjacent internal carotid artery and ophthalmic, trochlear, and abducent nerves11-26 (Figure
6). Although we could not satisfactorily demonstrate
branches from the plexus to the internal carotid
artery, evidence from nerve sections in cynomolgus
monkeys have shown that at least some of the
branches are parapathetic in nature. 13
The internal carotid ganglion is located on the wall
of the internal carotid artery, where the greater
superficial petrosal nerve is joined by the internal
carotid nerve to form the vidian nerve. The internal
carotid ganglion is also found in rats and hu-
mans.4'12-18'19 It appears to be strategically located on
the lateral wall of the internal carotid artery just at its
entrance through the skull base (Figure 1). In all
three species the internal carotid ganglion consists of
a distal (rostral) parasyrnpathetic subdivision and a
proximal sensory subdivision, as characterized by
their neuronal contents of VIP+ChAT and CGRP+
substance P, respectively. The fibers travel a very
short course in the greater deep petrosal nerve to the
arterial adventitia. The fibers seem not only to course
as bundles along the intracranial internal carotid
artery to reach its pial ramifications, but also to
innervate the intracranial segment of the artery with
some delicate fibers. In line with this, substantial
amounts of VIP are found by radioimmunoassay in
the intraaranial segment of the human internal carotid artery.19
No True Blue-positive cells were found in the
VIP+ChAT-positive miniganglia along the rami orbitales and orbitociliary nerve directed to the cavernous sinus nor in the VIP+ChAT-positive ganglion in
the trabeculae of the cavernous sinus. This indicates
that these ganglia do not innervate structures as far
340
Stroke
Vol 22, No 3 March 1991
MCA
ACA
<5>
n.TZE
caudal
cranial
nerves? n . H
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
FIGURE 8. Schematic overview of origins and pathways of sensoryfibersto right intracranial internal carotid artery and pial
arteries in monkeys. ACA, anterior cerebral artery; BA, basilar artery; DPN, deep petrosal nerve; EF, ethmoidal foramen; GSPN,
greater superficial petrosal nerve; ICA, internal carotid artery; ICG, internal carotid ganglion; ICN, internal carotid nerve; MCA,
middle cerebral artery; NCN, nasocUiary nerve; OCN, orbitociUary nerve; PCA, posterior cerebral artery; RO, rami orbitales; SCG,
superior cervical ganglion; SPG, sphenopalatine ganglion; TG, trigeminal ganglion; VN, vidian nerve.
out in the pial arterial tree as the smaller branches.
The possibility remains that the internal carotid
artery itself, the circle of Willis, and its proximal
branches may be reached by fibers from these ganglia. Other nearby target structures are, however,
equally likely as candidates: 1) the walls of venous
vessels of the cavernous sinus, 2) the capsule of the
glandular tissue of the hypophysis, 3) the basilar
artery and its branches, reached by fibers running on
the surface of the abducent nerve backward,13 as
supported by our findings of VIP+ChAT-positive
fibers on the nerve surface, 4) the tentorium cerebelli, reached by a fiber bundle forming the nervus
tentorii, which follows the trochlear nerve backward,26 and 5) fibers from the plexus joining the
adjacent cranial nerves, which may follow them rostrally to orbital structures.
In rats and cats the otic ganglion has been shown
to contribute parasympathetic fibers to the cerebrovascular innervation.34 We observed no such contribution in our two monkeys. The same was valid for
the ciliary, geniculate, glossopharyngeal, and vagal
ganglia. However, the limited number of animals
studied and the single application site for the tracer
along the vascular tree implies that negative findings
should be interpreted with caution.
The source for sensory innervation of the middle
cerebral artery with CGRP+substance P-positive
fibers appears to be threefold, as summarized in
Figure 8. As judged from the number of True Bluepositive cells, the major contribution comes from the
ophthalmic division of the ipsilateral trigeminal ganglion. The pathway for these fibers has been demonstrated in cynomolgus monkeys by the aid of electron
microscopy and nerve sections.13-27 The fibers bridge
to the cavernous plexus from the ophthalmic nerve
shortly rostral to the ganglion, and an additional
bridge exists between the plexus and the rostral
segment of the internal carotid artery just before the
artery bends upward in the sinus (Figure 6). A few
early reports in humans have identified fine branches
connecting the internal carotid artery with the medial
surface of the ophthalmic nerve shortly rostral to the
ganglion, but it was not clarified whether such fibers
were of sympathetic or sensory origin.28-30 Recently,
a pathway between the ophthalmic nerve and the
plexus has been demonstrated in humans, as well as
a probable extension of the pathway to the intracavernous internal carotid artery.19 Substantial levels of
Hardebo et al Cerebrovascular Nerves in Monkeys
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
CGRP and substance P are found in this segment of
the artery in humans.19
No CGRP- or substance P-positive fiber bundles
were observed in the basal frontal dura. This and
other anatomic data given above speak against a
pathway via the nasociliary nerve (from the ophthalmic division of the trigeminal ganglion) through
the ethmoidal foramen and basal frontal dura to the
circle of Willis, as is found in rats.12
Fibers from the ipsilateral maxillary division of the
trigeminal ganglion apparently travel in the orbitociliary nerve branch to run through the cavernous nerve
plexus and bridge to the internal carotid artery
together with fibers from the ophthalmic division.13
Apparently, some substance P-positive fibers from
the ophthalmic and maxillary divisions join the recurrent nerve of the plexus to run along the abducent
nerve to innervate the basilar artery and its branches,
as also shown by denervation experiments in cynomolgus monkeys.13 The unexpected finding of a few
CGRP+substance P-positive neurons in the abducent nerve may represent aberrant trigeminal neurons along this pathway. Further, in monkeys some
substance P-positive fibers from the ophthalmic division leave the plexus to join the tentorial nerve,
which follows the trochlear nerve to the rostral origin
of the tentorium cerebelli and innervates the tentorium, occipital dura, and falx and their sinuses.26
A third source of sensory fibers to the middle cerebral artery was found in the few CGRP+substance
P-positive cells in a subdivision of the ipsilateral internal carotid ganglion. The fibers reach the internal
carotid artery via the greater deep petrosal nerve and
run in bundles along the artery to reach the pial vessels
or form a terminal network on the artery.
The upper cervical dorsal root ganglia were not
studied since it is highly unlikely that they contribute to
the sensory innervation of supratentorial vessels.912-17
As recently demonstrated, activation of the cerebrovascular parasympathetic and sensory nerves
leads to a marked increase in cortical blood flow.31-35
Thus, upon activation these systems may ensure a
sufficient blood supply to the brain when it is threatened or when the demand is raised as during ischemia, seizures, and severe hypertension.31-33 If the
perivascular parasympathetic nerves terminate close
to the pain fiber endings, released VIP may even
enhance the action of the pain fiber transmitter(s).36
Our demonstration of a sensory innervation of the
intracranial segment of the internal carotid artery
may contribute to the understanding of the pathophysiology of syndromes with retro-orbital pain (cluster headache, ophthalmoplegic migraine, and Tolosa
Hunt's syndrome).18^"
Acknowledgments
The skillful technical assistance of Joanna Nilsson
and excellent word-processing by Maud Martensson
are highly appreciated.
341
References
1. Owman C, Hanko J, Hardebo JE, Kahrstiom J: Neuropeptides and classical autonomic transmitters in the cardiovascular system: Existence, coexistence, action, interaction, in
Owman C, Hardebo JE (eds): Neural Regulation of Brain
Circulation. Amsterdam, Elsevier Science Publishing Co, Inc,
1986, pp 299-331
2. Hara H, Weir B: Pathway of acetylcholinesterase containing
nerves to the major cerebral arteries in rats. J Comp Neurol
1986;250:245-252
3. Walters BB, Gillespie SA, Moskowitz MA: Cerebrovascular
projections from the sphenopalatine and otic ganglia to the
middle cerebral artery of the cat. Stroke 1986;17:488-494
4. Suzuki N, Hardebo JE, Owman C: Origins and pathways of
cerebrovascular vasoactive intestinal polypeptide-positive
nerves in rat. / Certb Blood Flow Metab 1988;8:697-712
5. Suzuki N, Hardebo JE, Owman C: Origins and pathways of
choline acetyltransferase parasympathetic nerve fibers to cerebral vessels in rat. / Cereb Blood Flow Metab 1990;10:399-408
6. Edvinsson L, Hara H, Uddman R: Retrograde tracing of nerve
fibers to the rat middle cerebral artery with True Blue:
Colocalization with different peptides. / Cereb Blood Flow
Metab 1989;9:212-218
7. Liu-Chen LY, Han DH, Moskowitz MA: Pia arachnoid contains substance P originating from trigeminal neurons. Neurosdence 1983;9:212-218
8. Liu-Chen LY, Mayberg MR, Moskowitz MA: Immunohistochemical evidence for a substance P-containing trigeminovascular pathway to pial arteries in cat. Brain Res 1983;268:
162-166
9. Mayberg MR, Zervas NT, Moskowitz MA: Trigeminal projections to supratentorial pial and dural blood vessels in cat,
demonstrated by horseradish peroxidase histochemistry. /
Comp Neurol 1984;223:46-56
10. O'Connor TP, van der Kooy D: Pattern of intracranial and
extracranial projections of trigeminal ganglion cells. / Neurosci
1986;6:2200-2207
11. O'Connor TP, van der Kooy D: Enrichment of a vasoactive
neuropeptide (calcitonin gene related peptide) in the trigeminal sensory projection to the intracranial arteries. / Neurosci
1988;8:2468-2476
12. Suzuki N, Hardebo JE, Owman C: Origins and pathways of
cerebrovascular nerves storing substance P and calcitonin
gene-related peptide in rat Neuroscience 1989^1:427-438
13. Ruskell GL, Simons T: Trigeminal nerve pathways to the
cerebral arteries in monkeys. J Anat 1987;155:23-37
14. Liu-Chen LY, Gillespie S, Norregaard TV, Moskowitz MA:
Colocalization of retrogradely transported wheat germ agglutinin and the putative neurotransmitter substance P within
trigeminal ganglion cells projecting to cat middle cerebral
artery. J Comp Neurol 1984;225:187-192
15. Keller JT, Beduk A, Saunders MC: Origin of fibers innervating
the basilar artery of the cat. Neurosci Lett 1985^8:263-268
16. Arbab MAR, Wiklund L, Svendgaard NA: 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. Neuroscience 1986;
19:695-708
17. Saito K, Moskowitz MA: Contributions from the upper cervical dorsal roots and trigeminal ganglia to the feline circle of
Willis. Stroke 1989;20:524-526
18. Hardebo JE, Suzuki N, Owman C: Origins of substance P - and
calcitonin gene-related peptide-containing nerves in the internal carotid artery of rat. Neurosci Lett 1989;101J9-45
19. Suzuki N, Hardebo JE: The anatomical basis for a parasympathetic and sensory innervation of the intracranial segment of
the internal carotid artery in man. Possible implication for
vascular headache. J Neurol Sci (in press)
20. Chorobski J, Penfield W; Cerebral vasodilator nerves and their
pathway from the medulla oblongata. Arch Neurol Psychiatry
1932;28:1257-1289
21. Carvalho VC: Nerve cells in the human cavernous sinus. Anat
Anz 1985; 159:29-32
342
Stroke Vol 22, No 3 March 1991
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
22. Hokfelt T, Skagerberg G, Skirboll L, Bjorklund A: Combination of retrograde tracing and neurotransmitter histochemistry, in Bjorklund A, Hokfelt T (eds): Handbook of Chemical
Neuroanatomy. Vol I. Methods in Chemical Neuroanatomy.
Amsterdam, Elsevier Science Publishing Co, Inc, 1983,
pp 228-285
23. Coons AH, Leduc EG, Connolly JM: Studies on antibody
production. I. A method for the histochemical demonstration
of specific antibody and its application to the study of the
hyperimmune rabbit. / Exp Med 1955;102:49-60
24. Crawford GD, Correa L, Salvaterra PM: Interaction of monoclonal antibodies with mammalian cholineacetyltransferase.
Proc NatlAcad Sci USA 1982;79:7031-7035
25. Ruskell GL: The orbital branches of the pterygopalatine
ganglion and their relationship with internal carotid nerve
branches in primates. J Anat 1970;106:323-339
26. Ruskell GL: The tentorial nerve in monkeys is a branch of the
cavernous plexus. J Anat 1988;157:67-77
27. Simons T, Ruskell GL: Distribution and termination of
trigeminal nerves to the cerebral arteries in monkeys. J Anat
1988;159:57-71
28. Hovelaque I: Anatomie des nerfs craniens et du systeme grand
sympathique. Paris, Doin, 1927, p 82
29. McNaughton FL: The innervation of the intracranial blood
vessels and dural sinuses. Assoc Res Nerv Ment Dis 1938;18:
178-200
30. Northfield DWC: Some observations on headache. Brain
1938;61:133-162
31. Moskowitz MA, Wei EP, Saito K, Kontos HA: Trigeminalectomy modifies pial arteriolar responses to hypertension and
norepinephrine. Am J Physiol 1988;255:H1-H6
32. Sakas DE, Moskowitz MA, Wei EP, Kontos HA, Kano M,
Ogilvy CS: Trigeminovascular fibers increase blood flow in
cortical gray matter by axon reflex-like mechanisms during
acute severe hypertension or seizures. Proc NatlAcad Sci USA
1989;86:1401-1405
33. Moskowitz MA, Sakas DE, Wei EP, Kano M, Buzzi MG,
Ogilvy C, Kontos HA: Postocclusive cerebral hyperemia is
markedly attenuated by chronic trigeminal ganglionectomy.
Am J Physiol 1989;257:H1736-H1739
34. Suzuki N, Hardebo JE, Kahrstrom J, Owman C: Effect on
cortical blood flow of electrical stimulation of trigeminal
cerebrovascular nerve fibres in the rat. Ada Physiol Scand
1990;138:307-315
35. Suzuki N, Hardebo JE, Kahrstrom J, Owman C: Selective
electrical stimulation of postganglionic cerebrovascular parasympathetic nerve fibers originating from the sphenopalatine
ganglion enhances cortical blood flow in the rat. J Cereb Blood
Flow Metab 1990;10:383-391
36. Khalil Z, Andrews PV, Helme RD: VIP modulates substance
P-induced plasma extravasation in vivo. Eur J Pharmacol
1988;151:281-287
37. Moskowitz MA: Cluster headache: Evidence for a pathophysiologic focus in the superior pericarotid cavernous sinus
plexus. Headache 1988;28:584-586
KEY WORDS
monkeys
neurons, afferent • ganglion, parasympathetic
Pathways of parasympathetic and sensory cerebrovascular nerves in monkeys.
J E Hardebo, M Arbab, N Suzuki and N A Svendgaard
Stroke. 1991;22:331-342
doi: 10.1161/01.STR.22.3.331
Downloaded from http://stroke.ahajournals.org/ by guest on June 18, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1991 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/22/3/331
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/