Acetylcholinesterase-containing fibers and choline acetyltransferase activity in isolated cerebral microvessels from goats

The complex contribution of NOS interneurons in the physiology of cerebrovascular regulation

Following the discovery of the vasorelaxant properties of nitric oxide (NO) by Furchgott and Ignarro, the finding by Bredt and coll. of a constitutively expressed NO synthase in neurons (nNOS) led to the presumption that neuronal NO may control cerebrovascular functions. Consequently, numerous studies have sought to determine whether neuraly-derived NO is involved in the regulation of cerebral blood flow (CBF). Anatomically, axons, dendrites, or somata of NO neurons have been found to contact the basement membrane of blood vessels or perivascular astrocytes in all segments of the cortical microcirculation. Functionally, various experimental approaches support a role of neuronal NO in the maintenance of resting CBF as well as in the vascular response to neuronal activity. Since decades, it has been assumed that neuronal NO simply diffuses to the local blood vessels and produce vasodilation through a cGMP-PKG dependent mechanism. However, NO is not the sole mediator of vasodilation in the cerebral microcirculation and is known to interact with a myriad of signaling pathways also involved in vascular control. In addition, cerebrovascular regulation is the result of a complex orchestration between all components of the neurovascular unit (i.e., neuronal, glial, and vascular cells) also known to produce NO. In this review article, the role of NO interneuron in the regulation of cortical microcirculation will be discussed in the context of the neurovascular unit.

Cholinergic and vasoactive intestinal polypeptidergic innervation of the cerebral arteries

Acetylcholine and vasoactive intestinal polypeptide are not only two vasoactive agonists that predominantly induce a vasodilatation of the cerebral arteries, but also correspond to neurotransmitters that innervate the various anatomical segments of the cerebral vasculature. The distinct patterns of the cerebrovascular cholinergic and vasoactive intestinal polypeptidergic innervation, their neurochemistry, in vitro and in vivo pharmacology, as well as the putative pathophysiological implications of these neurotransmission systems are critically summarized on the basis of the most recently published literature.

Architectonic subdivisions of the motor thalamus of owl monkeys: Nissl, acetylcholinesterase, and cytochrome oxidase patterns

As the first part of an investigation of the motor thalamus and its cortical connections in the owl monkey, a New World anthropoid primate, we studied thalamic architecture by using stains for Nissl, acetylcholinesterase (AChE), and cytochrome oxidase (CO), in order to identify subdivisions of the ventrolateral thalamic region as well as other nuclei with motor connections. Material was obtained from brains cut in the frontal, horizontal, and parasagittal planes. Our results indicate that the ventrolateral thalamic region (VL) of owl monkeys is a heterogeneous structure composed of several architectonic subdivisions that resemble divisions that have been described in macaques and other Old World anthropoids. All of these subdivisions are more readily distinguished in AChE than in Nissl or CO preparations. The anterior part of VL, VLa (VLo of Olszewski), is characterized by clusters of medium-sized, darkly stained neurons. VLa is also distinguished by AChE-positive cells embedded in a matrix of neurites as well as by a characteristic dark, irregular net of blood vessels. The posterior part of VL is rather uniform cytoarchitectonically and contains large, darkly stained, and sparsely distributed neurons. However, we were able to distinguish three subdivisions of posterior VL that closely correspond to structures described by Olszewski in macaques: a principal segment, VLp (VPLo of Olszewski), a medial segment, VLx ("area X" of Olszewski), and a dorsal segment, VLd (VLc and VLps of Olszewski). In AChE, VLd is much darker than the other divisions. The distinction between VLp and VLx, which together make up the largest part of VL, is less marked, although VLp is somewhat darker and more irregular in appearance in AChE than is VLx.

Nicotinic cholinergic receptors associated with mammalian cerebral vessels

Current evidence suggests that the cerebral vasculature may be modulated by cholinergic nerves. We used ligand binding methods to examine the presence of nicotinic cholinergic receptors in brain vasculature. We found carbachol-displaceable [3H]acetylcholine (ACh) and [3H]nicotine (NIC) binding sites in preparations of intraparenchymal cerebral microvessels (CMV) and larger pial vessels from human and pig brains. Specific binding sites for [3H]ACh and [3H]NIC in cerebral microvessels were saturable and comparable in density to those in cerebral cortex. The Kds for the two ligands ranged 3-18 nM whereas the Bmaxs were 25-45 fmol/mg protein. In contrast, the binding of [3H]pirenzipine or [3H]quinuclidinyl benzilate, index for muscarinic receptors, was low (9-15% of cortex) in microvessels compared to the cerebral cortex. Our observations suggest the association of cholinergic nicotinic receptors with cerebral microvessels, which may be involved in the modulation of the cerebral circulation by cholinergic neurons.

Light and electron microscopic immunocytochemical analysis of the neurovascular relationships of choline acetyltransferase and vasoactive intestinal polypeptide nerve terminals in the rat cerebral cortex

Acetylcholine or vasoactive intestinal peptide (VIP) nerve terminals closely related to intracortical blood vessels have previously been reported. Recent physiological evidence indicates that these central neuronal systems are involved in the fine control of local cerebral blood flow. In the present study, the intimate associations between choline acetyltransferase (ChAT) and VIP axon terminals and intracortical microvessels were characterized by light (LM) and electron microscopic (EM) immunocytochemistry. In semithin sections, LM analysis of the distribution of ChAT- and VIP-immunostained puncta juxtaposed to small intraparenchymal blood vessels demonstrated that neither type of terminal was enriched or impoverished around microvessels within the cerebral cortex. At the EM level, most ChAT- or VIP-immunolabelled elements located within a 3 microns perimeter around vessel walls were axon terminals. These perivascular terminals were associated primarily with capillaries but also, to a lesser extent, with microarterioles. Even though ChAT and VIP terminals were frequently found in the immediate vicinity (< or = 0.25 microns) of microvessels, they almost never contacted the outer basal lamina, usually abutting onto perivascular astroglial leaflets. There were no membrane specializations at the site of contact between ChAT or VIP terminals and perivascular astroglia. In all cortical areas examined, the average size of VIP-immunolabelled varicosities (0.56 +/- 0.04 microns 2) was significantly larger than that of their ChAT counterparts (0.32 +/- 0.02 microns 2; P < 0.001). Perivascular VIP terminals were more frequently engaged in synaptic contact than those immunostained for ChAT, which rarely exhibited a synaptic junction even in serial thin sections. Neither VIP nor ChAT immunostaining was ever observed in endothelial cells. These results suggest that both acetylcholine and VIP exert their effects on intracortical microvessels through indirect, paracrine mechanisms. The marked difference in synaptic incidence and average size between both types of perivascular terminals indicates that these two vasoactive agents are primarily located in distinct neuronal populations. Further, our results show that the astrocytic glia is the major direct target for both ChAT and VIP perivascular terminals and suggest that neuronal/glial/vascular interactions are a key element in the neurogenic control of the intracortical microcirculation.

Ouabain-sensitive choline transport system in capillaries isolated from bovine brain

In physiological conditions, there is a net transport of choline from brain to blood, despite the fact that the choline concentration is higher in plasma than in CSF. Because of the blood-brain barrier characteristics, such passage against the concentration gradient takes place necessarily through endothelial cells. To get a better understanding of this phenomenon, [3H]choline uptake properties have been analyzed in capillaries isolated from bovine brain. [3H]Choline uptake was linear with time for up to 1 h. Nonlinear regression analysis of the uptake rates at different substrate concentrations gave the best fit to a system of two components, one of which was saturable (Km = 17.8 +/- 4.8 microM; Vmax = 11.3 +/- 3.4 pmol/min/mg of protein) and the other of which was nonsaturable at concentrations up to 200 microM. The [3H]choline transport was significantly reduced in the absence of sodium and after incubation with 10(-4) M ouabain for 30 min. Ouabain also inhibited choline uptake in purified cerebral endothelial cells, but not in the endothelium isolated from bovine aorta. Accordingly, cerebral endothelial cells were able to concentrate [3H]choline, with this effect being abolished by ouabain, whereas in aortic endothelial cells the [3H]choline intracellular concentration was never higher than that of the incubation medium. These results suggest that the blood-brain barrier endothelium is specifically provided with an energy-dependent choline transport system, which may explain the choline efflux from the brain and the maintenance of a low choline concentration in the cerebral extracellular space.

Effects of locus ceruleus lesions on the pericapillary nerve terminals in the feline brain

The effects of bilateral locus ceruleus (LC) lesions on the pericapillary nerve terminals were investigated in the feline brain parenchyma using electron microscopy. LC lesions were induced stereotaxically and the animals were sacrificed after intravenous administration of 5-hydroxydopamine (5-OHDA). The diameter and number of dense-cored vesicles (DCVs) and clear vesicles (CVs) in the pericapillary nerve terminals were measured. The number of DCVs in the nerve terminal was significantly decreased by bilateral LC lesions. The diameters of the DCVs and CVs decreased significantly as compared with those in the non-operated control group. These data suggest that the LC is closely related to the pericapillary nerve terminals in the brain parenchyma and that not only nerve terminals with DCVs but also those with CVs are affected by LC lesions.

Glial cells in coculture can increase the acetylcholinesterase activity in human brain endothelial cells

The elements of the cholinergic system (acetylcholinesterase and choline acetyltransferase) and butyrylcholinesterase were studied in human cortical capillary samples, brain-derived endothelial cell cultures and glial cell cultures. It was shown that the elements of the cholinergic system are present in the microvessels, but the choline acetyltransferase activity may be due to contamination with cholinergic nerve terminals since no choline acetyltransferase could be demonstrated in endothelial cell cultures. The present results revealed that the activity of acetylcholinesterase is reduced in the cortical endothelial cell cultures after longer culture times, while butyrylcholinesterase activity is not altered. In a system where endothelial cells were cocultured with embryonic human brain astroglial cells for 12 days in vitro, the acetylcholinesterase activity was increased 2-fold. These results support a glial influence on the enzyme activity of the cerebral endothelium.

Periendothelial acetylcholine synthesis and release in bovine cerebral cortex capillaries

Choline acetyltransferase (ChAT) activity is present in isolated cerebral capillaries, where it has been considered to be a marker for perivascular cholinergic nerve terminals. However, ChAT-like immunoreactivity has been visualized in endothelial cells. This finding raised the possibility that at least part of the biochemically detected ChAT has a nonneuronal origin. To evaluate the relative contribution of endothelial cells and nerve fibers to the total acetylcholine (ACh)-synthesizing capacity of cerebral capillaries, ChAT activity and ACh release were measured in capillaries and in purified endothelial cells isolated from bovine cerebral cortex. Isolated capillaries showed ChAT activity, which was inhibited by 2-benzoylethyl trimethylammonium to the same extent as cerebral ChAT. When preincubated with [3H]choline, these capillaries presented a calcium-dependent enhancement in tritium release upon electrical field stimulation. Purified endothelial cells had minor ChAT activity and lacked the ability to release tritium in response to electrical stimulation, although the endothelial markers alkaline phosphatase, gamma-glutamyltranspeptidase, and 1,1'-dioctadecyl-1,3,3',3'-tetramethyl-iodocarbocyanide perchlorate-labeled acetylated low-density lipoprotein uptake were fully preserved. These data indicate that, within isolated cerebral capillaries, ACh is synthesized and released by a periendothelial structure. The fact that ACh release is provoked by electrical stimulation and by a calcium-dependent mechanism strongly suggests that cerebrovascular ACh has a neuronal origin.

Choline acetyltransferase activity associated with cerebral cortical microvessels does not originate in basal forebrain neurons

Cerebral cortical microvessels are innervated by cholinergic fibers that are probably involved in the regulation of local cerebral blood flow and blood-brain barrier permeability. The possibility exists that the cholinergic terminals associated with the cortical microvasculature belong to neurons from the nucleus basalis magnocellularis (NBM), where 70% of the cortical cholinergic projections originate. To test this hypothesis, ibotenic acid (25 nmol) was injected unilaterally in the NBM in rats, and 14 days later, choline acetyltransferase (ChAT) activity was measured in the frontoparietal cortex and in a blood vessel fraction isolated from this region. Lesions of the NBM resulted in a 50% decrease of cortical ChAT as compared with control or sham-operated hemispheres; however, no changes were observed in the ChAT activity associated with cortical microvessels. These results indicate that, in rat cerebral cortex, the perivascular cholinergic terminals do not originate in the basal forebrain.

Endothelial cells from human fetal brain microvessels may be cholinoceptive, but do not synthesize acetylcholine

Brain homogenate, cerebral microvessels, and endothelial cells (ECs) were prepared from 15-18-week-old human fetuses and analyzed biochemically for the presence of elements of the cholinergic system [acetylcholinesterase (AChE), choline acetyltransferase (ChAT), and butyrylcholinesterase]. The ECs were cultured, and their purity was checked by light microscopic immunohistochemistry with the application of anti-human factor VIII and glial fibrillary acidic protein. The highest activity of ChAT was found in the brain homogenate and the lowest in the microvessel fraction. No ChAT activity could be detected in the cultured ECs, despite the presence of high AChE activity. It is suggested that human brain ECs may be under the control of acetylcholine released from cholinergic nerve terminals but that the cells do not produce the transmitter itself. In coculture experiments, when ECs were plated on the upper surface of a polycarbonate filter and glial cells were seeded on the lower surface, the electric resistance was measured. During the culture period, the resistance first increased up to 5 days in vitro (297 +/- 17 ohm.cm2) but later gradually declined. These results demonstrate that human ECs cocultured with glial cells provide a useful model for study of the function of the blood-brain barrier in vitro.

Role of nucleus basalis in cholinergic control of cortical blood flow

The present investigation was designed to determine the effect of lesions localized to the nucleus basalis/substantia innominata (NB) on resting and cholinergically activated regional cerebral cortical blood flow (rCBF). Ibotenic acid (10 micrograms) was infused locally at 1 mm caudal to bregma, 3 mm lateral to the midline, and 8 mm below the cortical surface. Effectiveness of lesions was demonstrated by histological verification of lesion sites and determination of choline acetyltransferase activity in cerebral cortex homogenates. rCBF was measured with the autoradiographic iodo-14C-antipyrine technique. Resting rCBF was similar in the hemisphere that received the NB lesion and in the contralateral (intact) side in all regions examined. Physostigmine intravenous infusion (3.3 micrograms.kg-1.min-1) enhanced rCBF in frontal, parietal, occipital, and temporal cortex. The increase was symmetrical, however, indicating inability of NB lesion to affect this phenomenon. It is concluded that the cortical cholinergic afferents originating in the NB are not involved in the control of rCBF.

Activation of endothelial L-arginine pathway in resistance arteries. Effect of age and hypertension

In conduit arteries, nitric oxide is formed from L-arginine in the endothelium and released after stimulation with acetylcholine. The contribution of the L-arginine pathway and the effects of age and hypertension on endothelium-dependent vascular regulation were studied, using a video dimension analyzer, in pressurized and perfused mesenteric resistance arteries of 8- and 16-20-week-old Wistar-Kyoto and spontaneously hypertensive rats. Norepinephrine and phenylephrine caused contractions, which were similarly augmented after removal of the endothelium. NG-Monomethyl-L-arginine, an inhibitor of nitric oxide formation, augmented the contraction, but less than endothelial removal. Acetylcholine caused endothelium-dependent relaxations that were much more pronounced with intraluminal than with extraluminal application. NG-Monomethyl-L-arginine, methylene blue, and hemoglobin only partially inhibited the response. With aging, the endothelium-dependent inhibition of the response to norepinephrine decreased in Wistar-Kyoto rats; in spontaneously hypertensive rats this inhibition was smaller as compared with age-matched Wistar-Kyoto rats. In Wistar-Kyoto rats, the difference between intraluminal and extraluminal activation became more pronounced in adult rats. In the adult but not the young spontaneously hypertensive rats, the response to intraluminal but not extraluminal acetylcholine was reduced as compared with Wistar-Kyoto rats. Thus, in mesenteric resistance arteries of the rat, nitric oxide is released from L-arginine under basal conditions and after stimulation with acetylcholine but only in part accounts for endothelium-dependent responses. With aging and hypertension, the inhibitory effects of the endothelium against norepinephrine-induced contractions decrease. In hypertension, the intraluminal but not extraluminal activation of the release of endothelium-derived relaxing factors is impaired.

Neuronal and microvascular alterations induced by the cholinergic toxin AF64A in the rat retina

The choline analogue ethylcholine mustard aziridinium ion (AF64A) produces both neuronal and non-neuronal alterations in the rat retina. The possible involvement of the retinal capillaries in the origin of the apparently non-specific lesions has been investigated. Two hours after a single intraocular injection of 5 nmol AF64A, ultrastructural alterations were observed in neurons of the inner nuclear layer and the ganglion cell layer, where cholinergic cells are located. One week later, the number of cholinergic neurons, identified by choline acetyltransferase immunohistochemistry, was decreased to 65% of control, the neurons located in the inner nuclear layer being more sensitive than those in the ganglion cell layer. The same dose of AF64A also induced ultrastructural changes in retinal capillaries, which showed a significant increase in the number of pinocytotic vesicles and microvilli in the endothelial cells, 2-5 h after the toxin administration. One day later, arterioles and capillaries presented contracted profiles and the lumen was occasionally lost. The sensitivity of endothelial cells to the toxic effects of AF64A may be explained by the presence in the cerebral endothelium of a choline transport mechanism with an affinity close to that of cerebral synaptosomes. In vitro, both neuronal and endothelial choline uptake systems were equally sensitive to the toxin inhibitory effect. The early and severe vascular alterations induced in the retinal microvessels by AF64A may produce changes in blood perfusion and capillary permeability that could account for the apparently non-specific histological damage.

Origins and pathways of choline acetyltransferase-positive parasympathetic nerve fibers to cerebral vessels in rat

The presence of cholinergic nerve fibers in the brain vasculature has been a matter of controversy, partly due to the lack of a reliable histochemical marker. Accordingly, no distinct information about the origin and pathways for such fibers has been available. In the present study on the rat pial vasculature, utilizing a choline acetyltransferase (ChAT) antibody, which is able to demonstrate this enzyme in peripheral nervous tissue, evidence was obtained for an innervation by cholinergic fibers of large pial arteries. Vasoactive intestinal polypeptide (VIP) was present in or in close association with these fibers. By the aid of the retrograde axonal tracer True Blue (TB) applied to the middle cerebral arterial wall, such fibers were shown to originate in a subgroup of ChAT-positive cells in the sphenopalatine, otic, and internal carotid ganglia, which, in addition, contained VIP. The ChAT-positive pial nerve fibers were few in relation to the VIP-immunoreactive fibers, as was also illustrated by the few TB-positive cells in the ganglia that were ChAT positive as compared with the number of cells that were VIP positive. Only a small population of ChAT-containing neurons in these ganglia appeared to project to the pial vessels. The pathway from the sphenopalatine ganglion is via a membranous structure on the medial orbital wall, through the ethmoidal foramen, and along the internal ethmoidal artery to reach the circle of Willis. The fibers from the internal carotid and otic ganglia probably bridge to the internal carotid artery in the carotid canal, those from the otic ganglion after an initial course in the lesser superficial petrosal nerve.(ABSTRACT TRUNCATED AT 250 WORDS)

Muscarinic M1 receptors stimulate phosphoinositide hydrolysis in bovine cerebral arteries

The muscarinic agonist oxotremorine-M produced a concentration-dependent increase in phosphoinositide hydrolysis in bovine pial arteries. The maximal effect was 5.9 +/- 0.89 fold over basal levels, and the EC50 for oxotremorine-M was 8.9 x 10(-6) M. The phosphoinositide response in arteries with the luminal endothelium removed was similar to the response in intact arteries. The specific muscarinic antagonists pirenzepine, 4-DAMP and methoctramine produced parallel shifts of the concentration-response curve to oxotremorine-M, with the following order of potency (pKB): 4-DAMP (8.59 +/- 0.10) greater than pirenzepine (8.12 +/- 0.11) greater than methoctramine (6.77 +/- 0.20). These results indicate that muscarinic stimulation activates phosphoinositide hydrolysis in cerebral arteries, and that the muscarinic receptors mediating this increase are similar to the M1 subtype.

Vascular cholinesterases and choline uptake in isolated rat forebrain microvessels: a possible link

The two parameters of the active [methyl-3H]choline uptake into isolated rat forebrain microvessels, Km and Vmax, were determined for 1-, 3-, 10-, and 24-month-old Charles River male rats and compared with the activities of the enzymes choline acetyltransferase (ChAT), acetylcholinesterase (AChE), and butyrylcholinesterase (BuChE) in these microvessels over the same time course. The value of Km remained constant over the entire period, but that of Vmax increased from 8.5 +/- 1.0 to 80.6 +/- 16.4 nmol g-1 (mean +/- SEM) over the first 3 months of life. Over the same period, the increase in ChAT activity, from an initial value of 7.1 +/- 1.6 to 10.2 +/- 0.3 nmol g-1 min-1, was not proportional to that of choline uptake. Levels of BuChE activity (0.9-1.3 mumol g-1 min-1) were almost unchanged throughout the entire 24-month period, but those of AChE showed a steady and significant increase from 1 to 24 months, remaining relatively high at senescence (4.7 mumol g-1 min-1), when choline uptake had decreased to one-third of its optimal value. Selective inhibition of AChE with 1,5-bis(4-allyldimethylammonium-phenyl)pentan-3-one dibromide (0.5 microM) in unruptured capillaries from 3-month-old rats resulted in a decrease in Vmax of choline uptake from approximately 81 to 59 nmol g-1 min-1 or with 9-amino-1,2,3,4-tetrahydroacridine (10 microM) in capillaries from 2-month-old rats from approximately 30 to 15 nmol g-1 min-1. Selective inhibition of BuChE with tetraisopropyl pyrophosphoramide (100 microM) resulted in an increase in Vmax from approximately 81 to 96 nmol g-1 min-1. It is possible that the two vascular enzyme systems are coupled to a hypothetical endothelial choline transporter, but with an action opposite to each other.