Search

We sought to determine the source of the signal(s) that promotes expression of the catecholamine (CA) enzyme tyrosine hydroxylase (TH) in cultured neurons of embryonic rat cerebral cortex, a tissue which is not thought to contain CA cells in vivo. Cortical neurons were cultured with their non-neuronal constituents and 48 hr later immunostained for TH. Fibroblasts or glia had no effects, however, blood vessels increased the numbers of TH neurons nearly 4-fold. Coculture with either perinatal aorta, skeletal or cardiac muscle, clonal muscle cell lines 1440 (smooth) and L6 (skeletal), conditioned media from L6 cells, or a soluble extract of L6 cells increased the number of TH neurons up to 20-fold. The induction of TH by muscle extract was (1) dose dependent; (2) paralleled by a proportional increase in the steady-state levels of TH mRNA; (3) greatly reduced by the RNA synthesis inhibitor alpha-amanitin or the protein synthesis inhibitor cycloheximide; and (4) unassociated with change in the survival of neurons in culture. The response was not replicated by treatment with other established neurotrophic substances, including NGF, EGF, FGF, PDGF, neuroleukin, insulin, pyruvate, KCI, adenosine, or inosine. We conclude that muscle contains a potentially novel substance, muscle-derived differentiation factor (MDF) that promotes differentiation but not survival of neurons of cerebral cortex by de novo synthesis of TH mRNA and TH protein. Thus, neurons of the CNS, as in periphery, may undergo phenotypic interconversion in response to biologically derived molecules in their environment.

The electron microscopic localization of the adrenaline-synthesizing enzyme, phenylethanolamine N-methyltransferase (PNMT) was examined in the rostral ventrolateral medulla (RVL) of adult rats. The brains were fixed by perfusion with 3.75% acrolein and 2.0% paraformaldehyde in phosphate buffer. Coronal Vibratome sections through the RVL were immunocytochemically labeled using a rabbit polyclonal antiserum to PNMT and the peroxidase-antiperoxidase method. A semi-quantitative ultrastructure analysis revealed that the perikarya constituted 9% of the total immunoreactive profiles observed in the RVL. The labeled somata were large (18-24 microns) and were characterized by an indented nucleus and abundant cytoplasm with numerous mitochondria. An average of 136.8 +/- 11.6 mitochondria were present per 100 microns2 cytoplasm, which is 38% greater than the numbers found for PNMT-immunoreactive neurons in the nucleus of the solitary tract. Moreover, the labeled somata were often found in direct apposition to the basal lamina of small capillaries and neighboring astrocytic processes. The remaining labeled profiles were neuronal processes of which 72% were dendrites. Both the PNMT-labeled somata and dendrites received primarily symmetric contacts from unlabeled axon terminals. Only a few axons and terminals containing immunoreactivity for PNMT were observed. The axons were both unmyelinated and myelinated. The PNMT-immunoreactive terminals were characterized by a mixed population of vesicles and by the formation of synaptic junctions with both unlabeled dendrites and PNMT-labeled perikarya and dendrites. The ultrastructural morphology and proximity to blood vessels and glia suggest a high metabolic activity and possibly a chemosensory function of PNMT neurons in the RVL. The existence of myelinated and unmyelinated axons could imply that PNMT-containing neurons have different conduction velocities in efferent pathways to the spinal cord or other brain regions. Furthermore, the multiple types of synaptic interactions between labeled and unlabeled axons and dendrites support the concept that adrenergic neurons modulate and are modulated by neurons containing the same or other putative transmitters in the RVL.

The present study examined whether destruction of intrinsic neurons in the lateral hypothalamus of the rat would disrupt the acquisition of classically conditioned changes in arterial pressure. Ibotenic acid, a cellular toxin which spares axons of passage, was injected bilaterally in the hypothalamus either medial or lateral to the fornix. After 2 weeks the animals were subjected to classical fear conditioning trials involving the presentation of a tone in association with footshock. The next day changes in arterial pressure and emotional behavior elicited by the tone alone were measured. Destruction of intrinsic neurons in the lateral hypothalamus prevented the normal establishment of the arterial pressure conditioned response but did not affect the behavioral response. Unconditioned arterial pressure responses elicited by the tone and shock were not affected. Medial hypothalamic injections had no effect on any of the responses. The location of the lateral hypothalamic cell loss overlapped with the neurons projecting to the autonomic region of the spinal cord. Intrinsic neurons in the lateral hypothalamus therefore appear to be specifically involved in mediating learned cardiovascular adjustments.

Adrenaline-containing neurons of the rostral ventrolateral medulla (the C1 group) project selectively to autonomic spinal neurons in rats. Stimulation of these neurons electrically or chemically elevates arterial pressure, while neuronal blockade by microinjection of tetrodotoxin bilaterally drops arterial pressure to levels comparable to those produced by spinal cord transection. Adrenaline neurons of the ventral medulla appear necessary for maintaining normal levels of blood pressure, and thus may constitute a tonic vasomotor center.

We sought to determine in rat striatum whether the release of neurotransmitter amino acids aspartate (Asp), glutamate (Glu) and gamma-aminobutyric acid (GABA) were affected by local neurons. To do so, unilateral microinjections of ibotenic acid, and excitotoxin that destroys local neurons without affecting fibers of passage, were made into the striatum. Release of endogenous amino acids from lesioned and intact striatal slices were measured by HPLC one week later. The effectiveness and specificity of the lesion were confirmed by measuring the enzyme activity associated with extrinsic dopamine neurons (tyrosine hydroxylase; 111 +/- 14%), intrinsic GABA neurons (glutamic acid decarboxylase; 19 +/- 7%) and intrinsic acetylcholine neurons (choline acetyltranferase; 37 +/- 10%). Destruction of local striatal neurons markedly attenuated the release of GABA (41 +/- 12% of control) elicited by depolarization with K+ (35 mM), but did not significantly reduced the K+-evoked release of Asp (80 +/- 17%) and Glu (92 +/- 8%). However, spontaneous release of Asp and Glu was significantly greater than that observed in unlesioned tissue (159 +/- 18% and 209 +/- 27%, respectively), while the spontaneous release of GABA was not significantly reduced (75 +/- 43%). Although release of the neurotransmitter amino acids Asp, Glu and GABA were affected by the lesion, the release of the non-neurotransmitter amino acid tyrosine was unaffected. These data are consistent with the hypotheses that: 1) the predominant source of releasable stores of endogenous Asp and Glu in the striatum arises from extinsic neurons, and 2) that the spontaneous release of Asp and Glu from axon terminals in the striatum may be regulated, at least in part, by local inhibitory neurons.

We found that the catecholamine biosynthetic enzymes tyrosine hydroxylase (TH) (EC 1.14.16.2), dopamine beta-hydroxylase (EC 1.14.17.1), and phenylethanolamine N-methyltransferase (EC 2.1.1.28) share similar protein domains in their primary structures and that they share common gene coding sequences. In a recent report we also demonstrated that antiserums directed against choline acetyltransferase (EC 2.3.1.6), glutamic acid decarboxylase (EC 4.1.1.15), and TH cause specific complement-mediated lysis of cholinergic, gamma-aminobutyric acid-ergic, and dopaminergic subpopulations of synaptosomes, respectively. This interaction of specific antibodies to the specific subpopulation of synaptosomal membrane, e.g., recognition of antibody to TH to only the dopaminergic subpopulation of synaptosomal membrane protein, indicates that the neurotransmitter enzyme and membrane protein of its own synaptosomes may also share common protein domains. Therefore, we postulate that the specific neurotransmitter biosynthetic enzyme and a certain membrane protein of the nerve endings may share similar gene coding sequences, and that expression of these proteins may determine the phenotype of the neuron.

We have studied the effect of electrical stimulation of the parabrachial nucleus (PBN) and adjacent areas of dorsal pons on regional cerebral blood flow (rCBF) and glucose utilization (rCGU) in anesthetized (chloralose), paralyzed (tubocurarine) rats. rCBF and rCGU were measured in dissected tissue samples of 9 brain regions by the [14C]iodoantipyrine and [14C]2-deoxyglucose method, respectively. Electrical stimulation restricted to the medial parabrachial nucleus (PBNm, n = 5) elicited significant (P less than 0.05) reductions in rCBF in 7 out of 9 brain regions. Reductions were greatest in cerebral cortex (up to 35% in occipital cortex) and least in the white matter of the corpus callosum (23%). The effect on rCBF persisted after transection of the cervical sympathetic trunk (n = 5). In contrast, stimulation of the lateral portion of PBN (n = 5), periventricular gray (n = 5) and interestingly, the nucleus locus coeruleus (n = 5) failed to elicit similar changes in rCBF. PBNm stimulation also elicited decreases in rCGU (n = 4) in 5 out of 9 brain areas, most notably regions of cerebral cortex. The decreases in rCGU (delta rCGU) were linearly related to the decreases in rCBF (delta rCBF) according to the equation delta rCBF = 2.37 delta rCGU + 2.1 (r = 0.72; P less than 0.001). We conclude that excitation of neural pathways originating in, or passing through, PBNm elicits a widespread reduction in cerebral metabolism and secondarily in blood flow (secondary vasoconstriction). Since projections of the PBNm do not involve the entire cortex, it seems likely that the effect is mediated via inhibition of diffuse cortical projections through a subcortical site.

The greater activity of tyrosine hydroxylase (TH) in substantia nigra and corpus striata of adult BALB/cJ than CBA/J mice, is attributable to differences in the number of dopamine neurons in the ventral midbrain tegmentum. To determine if strain differences in TH activity develop postnatally we have measured the development of TH in the midbrain (SN) and in the corpus striatum (CS). In the midbrain neonatal TH activity was 20% of adult levels. Thereafter, TH activity increased rapidly achieving adult levels by 11 days. A 25% "overshoot' above adult values at 15 days was followed by a gradual decrease to adult activity at 4 weeks. In the CS neonatal activity was about 10% of adult levels and increased slowly to reach adult values at 4 weeks. Striatal choline acetyltransferase (CAT) activity in the neonate was only 3.7% of adult values and at 21 days had only reached 70% of adult activity. Neonatal glutamic acid decarboxylase (GAD) activity was relatively high in both brain regions and increased gradually to adult activity by 4 weeks. Strain differences in TH activity were not present at birth but first appeared at 9 days in SN and 11 days in CS. Once established, the differences were maintained. These results suggest that strain differences in TH are most probably a consequence of differences in postnatal neuron survival, although the possibility that some neurons lose their phenotypic expression of TH cannot be excluded.

The effects of somatostatin (SOM) and cholecystokinin octapeptide (CCK-8) on basal and potassium-induced release of acetylcholine (ACh) were investigated in slices of rat caudate nucleus (CN) and, for comparison, cerebral cortex (CX). Potassium (5-55 mM) produced a concentration-dependent increase in the release of [3H]ACh in the presence of extracellular Ca2+. SOM (1 microM), CCK-8 (1 microM) and the dopamine (DA) receptor agonist, apomorphine (APO, 30 microM) inhibited the K+-induced (35 mM) release of [3H]ACh by 26-32% from CN, but did not affect ACh release from CX. Other peptides (1 microM), such as Met-enkephalin, vasoactive intestinal peptide, thyrotropin-releasing hormone and substance P, had no effect on release of [3H]ACh in CN or CX. Sulpiride (SULP), a dopamine receptor antagonist, prevented the effects of APO and SOM, but not CCK-8, to inhibit [3H]ACh release. The results indicate that: (1) SOM and CCK-8 inhibit the release of [3H]ACh in CN, but not CX; and (2) the inhibitory effect of SOM, but not CCK-8, on [3H]ACh release is mediated by dopaminergic mechanisms.

We have postulated that the catecholamine-synthesizing enzymes tyrosine hydroxylase (TH), dopamine beta-hydroxylase (DBH), and phenylethanolamine N-methyltransferase (PNMT) are coded for by similar genes. To analyze the structural relationship of genes coding for these enzymes, we have cloned DNAs complementary (cDNA) to DBH and PNMT messenger RNAs (mRNAs). Using hybrid selection analysis to identify the cDNA clones positively, we discovered cross-hybridization between DBH cDNA clones and PNMT mRNA and between PNMT cDNA clones and DBH mRNA. Further analysis by RNA blot hybridization revealed that the DBH cDNA probe hybridized predominantly to a 5500 nucleotide mRNA and less strongly to a 1100 nucleotide species, and the PNMT cDNA probe hybridized strongly to the 1100 nucleotide mRNA and weakly to the 5500 nucleotide message. DNA blot hybridization analysis demonstrated that DBH and PNMT cDNA probes hybridized to several common restriction fragments of total cellular DNA. The evidence presented here suggests the existence of homologous gene-coding regions in DBH and PNMT cDNAs. These homologies may be the result of duplication of a common ancestral gene. DNA blot analysis suggests that these enzymes are coded for by single genes, which may be located in close proximity to each other in the DNA, and points to the existence of either a single gene or linked genes coding for all catecholamine enzymes.

We have sought to determine if adrenaline neurons of the C1 group are responsible for cardiovascular functions heretofore attributed to neurons in the rostral ventrolateral medulla. C1 neurons were identified in rat with antibodies to the adrenaline synthesizing enzyme, PNMT. These project to spinal cord wherein they selectively innervate the sympathetic columns. C1 neurons are also innervated by projections, mostly unilateral, from the nucleus tractus solitarii (NTS). Stimulation of the C1 area electrically, by local injection of the excitatory amino acid, L-glutamate, or with the GABA antagonist bicuculline, elevates arterial pressure (AP). Bilateral electrolytic lesions, microinjection of GABA, or administration of tetrodotoxin, in contrast, collapses AP to levels comparable to that of spinal cord transection. After lesions of one NTS, a lesion of the contralateral C1 area abolishes all reflex activity elicited by electrical or natural stimulation of baroreceptors on the side of C1 lesion without modifying resting AP. Lesions of axon bundles of PNMT neurons in the medulla also abolish baroreflexes after unilateral NTS lesions. C1 neurons appear to be the neurons mediating cardiovascular effects of application of drugs or cold to rostral portions of the ventrolateral medulla. We conclude adrenaline neurons of the C1 area represent the purportedly tonic vasomotor neurons of the rostral ventrolateral medulla and mediate the vasodepressor limb of reflexes arising from arterial baroreceptors and other cardiopulmonary afferents. Whether the tonic vasomotor response to stimulation of C1 neurons is dependent upon the release of adrenaline is not yet certain.

Binding sites labeled by [3H]p-aminoclonidine ([3H]PAC) were characterized in bovine brain membranes prepared from the ventrolateral medulla, the probable site of the antihypertensive action of clonidine and analogs. Comparison was made with [3H]PAC binding to membranes prepared from frontal cortex, which has been studied extensively. Saturation binding isotherms for [3H]PAC were similar in the two regions, although Bmax values were approximately two-fold lower in ventrolateral medulla relative to frontal cortex. Norepinephrine and other phenylethylamines displaced [3H]PAC from a maximum of 70% of the total sites in the ventrolateral medulla. The remaining 30% were norepinephrine-insensitive, non-adrenoceptor sites which displayed high affinity for imidazole compounds. Ligand selectivity differed markedly between ventrolateral medulla and frontal cortex, since some imidazole compounds which potently inhibited [3H]PAC binding in the ventrolateral medulla had no effect in frontal cortex. Imidazole binding sites may mediate, in part, the hypotensive action of clonidine and other imidazole compounds in the ventrolateral medulla. These sites may also participate in the functions of a putative endogenous clonidine-like substance.

The hemodynamic changes associated with hypotension elicited by electrical stimulation of the spinal trigeminal complex (trigeminal depressor response, TDR) or the aortic nerve (aortic depressor reflex, ADR) were compared in rabbits anesthetized with urethan. The hypotension associated with each response was accompanied by bradycardia, a marked fall in total peripheral resistance, a small decrease in cardiac output, and a nonuniform decrease in regional vascular resistances, with the order of magnitude of the decrease being femoral greater than mesenteric greater than renal arterial resistance. In individual experiments the percent decrease in heart rate, total peripheral resistance, or regional resistances was plotted against the percent fall in arterial pressure to obtain a pair of regression lines during the TDR and ADR. There was no significant difference in the slope or y-intercept of the regression line between the TDR and ADR for all of the hemodynamic variables examined. In both responses, however, the slope of the femoral resistance/arterial pressure relationship was significantly greater than that of the renal resistance/arterial pressure relationship. We conclude that the TDR is characterized by a pattern of hemodynamic changes similar to that of the ADR.

Electrical stimulation within discrete sites of the spinal trigeminal complex in anesthetized or decerebrated rabbits results in arterial hypotension, often over 50 mm Hg, bradycardia of up to 60 beats/min, apnea, and gastric hypermotility, collectively termed the trigeminal depressor response (TDR). The threshold for the TDR is less than or equal to 10 muA and is graded up to 3-6 times threshold. It can only be elicited by trains of stimuli of low frequency (0.5-20 Hz); at 50 Hz the response disappears or becomes pressor. The bradycardia is only abolished by bilateral vagotomy combined with beta-adrenergic blockade, and thus results from combined excitation of cardio-vagal and inhibition of cardiac sympathetic nerves. The hypotension is unassociated with changes in cardiac output, does not change after blockade of the bradycardia, but disappears after alpha-adrenergic blockade and hence is entirely attributable to inhibition of ongoing sympathetic vasoconstrictor nerve activity. Below threshold stimulation the TDR can only be elicited from the root entry zone of the Vth nerve, from dorsal portions of the spinal tract of the Vth nerve, and to portions of the nucleus of the spinal tract, notably the nucleus caudalis. A TDR of reduced magnitude can also be elicited by low frequency stimulation of numerous branches of the Vth nerve arising from all three divisions and including the supra- and infra-orbital, the inferior alveolar, and lingual nerves. Bilateral electrolytic lesions of the nucleus tractus solitarii at the obex, with complete abolition of baroreceptor reflexes from carotid sinus and aortic depressor nerves, fail to alter the TDR elicited from the brain or from branches of the Vth nerve, or the vasodepressor responses elicited by electrical stimulation of the central ends of the IXth and Xth cranial nerves transescted distal to the branches of baro-receptor nerves. In contrast, caudal lesions of the trigeminal complex abolish the TDR elicited from brain and Vth nerve and substantially reduces the vasodepressor responses from the IXth and Xth nerves, without altering baroreceptor reflexes. We conclude that the TDR represents a heretofore recognized vasodepressor response dependent upon the spinal trigeminal complex which is at least in part anatomically distinct from pathways subserving arterial baroreceptor and somatic vasodepressor reflexes. The TDR can be reflexly elicited from widely distributed but yet unidentified receptors innervated by branches of the Vth and of the IXth and Xth cranial nerves other than those innervating arterial baroreceptors. It is of unknown function, but may be related to pain mechanisms.

We sought to establish whether neurons of the C1 area of the rostral ventrolateral medulla (RVL) mediate changes in arterial pressure and heart rate evoked by topical application of drugs to the ventral medullary surface of the rat. Animals were anesthetized, paralyzed, and ventilated. The ventral surface was mapped with L-glutamate, and a restricted zone was identified from which L-glutamate, as well as kainic acid, bicuculline, strychnine, carbachol, or physostigmine, increased arterial pressure and heart rate. The hypertensive effects of carbachol and physostigmine were blocked by previous local application of atropine but not hexamethonium. Application of gamma-aminobutyric acid (GABA) or glycine to this area produced hypotension and bradycardia. Located caudal to the trapezoid bodies and lateral to the pyramids, this area corresponded to points with lowest threshold for pressor responses evoked by electrical stimulation and overlapped the distribution of epinephrine-synthesizing cells of the RVL. Processes arising from these neurons were identified reaching and contacting the ventral surface. Unilateral lesions involving the C1 area or phenylethanolamine-N-methyltransferase-labeled descending axons derived from this area imparied by greater than 70% the response to ipsilateral application of L-glutamate, GABA, or glycine to the ventral surface. We suggest that neurons within the C1 area of RVL adjacent to or including epinephrine cells may mediate cardiovascular changes elicited from a restricted chemosensitive zone of the ventral medullary surface of the rat.

Neurons located within 100 micron of the ventral surface of the hindbrain were labeled following injections of horseradish peroxidase into the thoracic and lumbar spinal cord in the rat. Labeled neurons were distributed into two groups, one ventrolateral to the inferior olive and ventral to the facial nucleus, and one ventral to the raphe pallidus. The location of the ventrolateral subpial group is very similar to that of the brain stem chemosensitive zones, and suggests that the neurons in this group may be important in cardiovascular regulation.

The activity of the epinephrine biosynthetic enzyme phenylethanolamine N-methyltransferase (PNMT, EC 2.1.1.28) is 3- to 8-fold greater in rats of the Fischer 344 and Buffalo strains. The biochemical basis for the strain differences has been analyzed at the level of PNMT protein and messenger RNA production. Fischer rat adrenals possess approximately 5-fold more PNMT protein than those of the Buffalo rat as established by Western blotting and immunoprecipitation of adrenal gland homogenates. Poly(A)+ RNAs purified from adrenal glands of each strain were translated in a reticulocyte lysate system, immunoprecipitated with antibody to PNMT and fractionated by SDS-PAGE. A 35S-labelled protein of Mr = 34,000 was immunoprecipitated from adrenals of Fischer and Buffalo rats, indicating that the molecular weights of PNMT do not differ in these strains prior to post-translational processing. Hybridization of a 740 base pair (bp) cDNA for PNMT indicated that the mRNAs for PNMT are the same size in the adrenals of both strains. However, the adrenals of Fischer rats contain 2- to 4-fold more PNMT mRNA than Buffalo rats, as established by quantitative dot blot hybridization and Northern blot analysis. The medulla oblongata, the site of cell bodies of central adrenergic neurons, also contains approximately 2-fold more PNMT mRNA in Fischer rats. The strain specificity in the production of PNMT reflects differences in the expression of the gene for PNMT. Thus, an inherited capacity for PNMT expression may in fact provide the intrinsic determinants responsible for neurotransmitter production. These data provide a direct link between regulation of catecholamine enzyme biosynthesis at the genomic level and the availability of specific catecholamines for neurotransmitter and hormonal functions.

The purpose of the present study was to determine whether lesions of areas projected to by the central amygdaloid nucleus (ACE) would disrupt the classical conditioning of autonomic and/or behavioral emotional responses. The areas studied included 3 projection targets of the ACE: the lateral hypothalamic area (LH), midbrain central gray (CG) region, and bed nucleus of the stria terminalis (BNST). Lesions were made either electrolytically or by microinjection of ibotenic acid, which destroys local neurons without interrupting fibers of passage. Two weeks later, the animals were classically conditioned by pairing an acoustic stimulus with footshock. The next day, conditioned changes in autonomic activity (increases in arterial pressure) and emotional behavior ("freezing," or the arrest of somatomotor activity) evoked by the acoustic conditioned stimulus (CS) were measured during extinction trials. Electrolytic and ibotenic acid lesions of the LH interfered with the conditioned arterial pressure response, but did not affect conditioned freezing. Electrolytic lesions of the rostral CG disrupted conditioned freezing but not conditioned changes in arterial pressure. Ibotenic acid injected into the rostral CG reduced neither the arterial pressure nor the freezing response. Injection of ibotenic acid in the caudal CG, like electrolytic lesions of the rostral CG, disrupted the freezing, but not the arterial pressure response. Injection of ibotenic acid into the BNST had no effect on either response. These data demonstrate that neurons in the LH are involved in the autonomic, but not the behavioral, conditioned response pathway, whereas neurons in the caudal CG are involved in the behavioral, but not the autonomic, pathway. Different efferent projections of the central amygdala thus appear to mediate the behavioral and autonomic concomitants of conditioned fear.

In embryonic mice, the catecholamine biosynthetic enzyme tyrosine hydroxylase [L-tyrosine, tetrahydropteridine:oxygen oxidoreductase (3-hydroxylating), EC 1.14.16.2] can be visualized immunocytochemically in a population of cells in epithelial cords of the developing pancreas. These embryonic catecholamine cells, first seen by day 11, are large and vacuolated and have a folded nuclear membrane. One day later, at day 12, glucagon is first detected immunocytochemically in pancreatic cells similar in location and morphology to the embryonic catecholamine cells. By use of a method for detecting both antigens in the same cell, both the hydroxylase and glucagon can be visualized between day 12 and day 14 in 10-40% of stained cells. From day 14, the number of cells stained for hydroxylase decreases; they cannot be detected after day 18. In contrast, the cells containing glucagon increase during development and persist throughout life. Endocrine cells of the embryonic pancreas also contain dopa decarboxylase but not dopamine-beta-hydroxylase or phenylethanolamine-N-methyl transferase. In adult mice, small cells containing tyrosine hydroxylase but differing in location and morphology from the embryonic catecholaminergic cells are seen in pancreatic islets. The adult catecholaminergic cells never store glucagon. We suggest that adult glucagon (A)-containing cells arise from transformation in situ of cells that transiently express a catecholaminergic (probably dopaminergic) phenotype. These results suggest that one class of peptidergic cells may arise from transformation of an aminergic precursor.

The vasodilation elicited in cerebral cortex by stimulation of the cerebellar fastigial nucleus (FN) is mediated by input pathways coming from the basal forebrain. We studied whether these pathways mediate the cortical vasodilation via a direct action on local blood vessels or via interposed local neurons. Neurons were destroyed in the primary sensory cortex by local microinjection of the excitotoxin ibotenic acid (IBO) (10 micrograms/l microliter). Five days later rats were anesthetized (alpha-chloralose), paralyzed, and ventilated. Arterial pressure and blood gases were controlled, and FN was stimulated electrically. Local cerebral blood flow (LCBF) was measured using the [14C]iodoantipyrine technique with autoradiography. Five days after IBO, neurons were destroyed in a restricted cortical area, and afferent fibers and terminals were preserved. The selectivity of the neuronal loss was established by histological and biochemical criteria and by transport of horseradish peroxidase from or into the lesion. Within the lesion, resting LCBF (n = 7) was unaffected, but the increase in LCBF evoked from the FN was abolished (P greater than 0.05); n = 6). In contrast the vasodilation elicited by hypercapnia (arterial CO2 partial pressure = 62.7 +/- 3; n = 5) was preserved. In the rest of the brain the vasodilation elicited from FN was largely unaffected. We conclude that the vasodilation evoked from FN in cerebral cortex depends on the integrity of a restricted population of local neurons that interact with the local microvasculature.

We sought to determine by using kainic acid (KA), an agent stimulating neuronal cell bodies but not fibers, whether the elevations in arterial pressure (AP) and heart rate (HR) elicited by electrical stimulation of the cerebellar fastigial nucleus (FN) are due to excitation of local neurons in FN or fibers passing through the area. Rats were anesthetized (chloralose 40 mg/kg), paralyzed (tubocurarine) and artificially ventilated. Electrical stimulation of FN (50 Hz, 10-100 microA) increased AP and HR. In contrast restricted microinjections of KA (0.05-5 nmol/100 nl) in FN elicited a sustained, dose-dependent decrease in AP and HR. The response is not due to diffusion of KA to adjacent sites. Excitation by KA of neurons intrinsic to FN elicits a heretofore unrecognized vasodepressor response, the fastigial depressor response.

In an investigation of the mechanism by which brain lesions result in delayed degeneration of neurons remote from the site of injury, neurons within the caudate nucleus of rats were destroyed by local injection of the excitotoxin ibotenic acid. Treatment resulted in the rapid degeneration of the striatonigral pathway including projections containing the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) and delayed transneuronal death of neurons in the substantia nigra pars reticulata. The distribution of nigral cell loss corresponded to the loss of GABAergic terminals. Neuronal death was prevented by long-term intraventricular infusion of the GABA agonist muscimol. Delayed transneuronal degeneration may be produced by neuronal disinhibition consequent to loss of inhibitory inputs. Replacement of inhibitory transmitters by suitable drugs may prevent some forms of delayed neuronal death.

The monoamine-synthesizing enzymes tyrosine hydroxylase (TH), dopamine-beta-hydroxylase (DBH) and tryptophan hydroxylase (TrH) were immunocytochemical localized in dopaminergic, noradrenergic and serotonergic neurons of rat brain by light and electron microscopy. In dopaminergic and serotonergic neurons, the respective synthesizing enzymes. TH and TrH, were distributed throughout the cytoplasm of the neuronal perikarya, dendrites, axons and terminals. The most selective accumulation of reaction product for the specific enzyme was associated: (a) in perikarya with endoplasmic reticulum, Golgi apparatus and microtubules, (b) in processes with microtubules, and (c) in terminals with dense granules or clear vesicles. The labeled terminals were characterized by their content of labeled organelles and the absence of synaptic junctions. In noradrenergic neurons, both TH and DBH were localized in the perikarya, similar to TH in dopamine neurons. TH and DBH differed in their localization within proximal axons and dendrites in that TH was associated with microtubules but DBH was not. These results provide ultrastructural evidence to suggest that monoamines may be: (a) synthesized by enzymes which are associated with different organelles depending on the portion of the neuron and the type of enzyme; (b) synthesized in both axons and dendrites and (c) released from terminals without postsynaptic membrane specializations.

The first part of this chapter demonstrates that the C1 adrenergic neurons have high mitochondrial content and a close proximity to capillaries and glia suggestive of a high metabolic activity and a possible chemosensory function. Adrenergic terminals arising primarily from these neurons (1) can influence sympathetic nerve discharge through direct contacts on sympathetic preganglionic neurons in the IML of the spinal cord; and (2) are one of the more prevalent synaptic inputs to the principally noradrenergic neurons in the locus coeruleus. In both the IML and locus coeruleus, adrenergic terminals may be either excitatory (asymmetric synapses) or inhibitory (symmetric synapses) depending on their distribution on the post-synaptic target. The second part of this chapter shows that C1 adrenergic neurons in the RVL are modulated by synaptic associations with a variety of transmitter systems (see schematic Fig. 8). Specifically, C1 adrenergic neurons receive (1) major inhibitory input (symmetric synapses) from GABA-ergic and opioid terminals as well as from unidentified (unlabelled) transmitter-containing terminals; (2) major excitatory input (asymmetric synapses) from terminals containing substance P as well as other unidentified terminals and (3) minor inputs from cholinergic, adrenergic and noradrenergic pathways. Moreover, cholinergic terminals in the RVL form symmetric synapses mainly on unidentified transmitter-containing neurons rather than the C1 neurons suggesting that the reported cardiovascular effects of cholinergic agents in the RVL are most likely mediated via inhibitory interneurons. Within the RVL, adrenergic and noradrenergic terminals innervate cholinergic and opioid neurons. Thus, these results not only provide direct evidence that a number of transmitters modulate the activity of C1 adrenergic neurons, but also suggest new directions for studies of functional interactions involving catecholaminergic regulation of other transmitter-containing neurons within the RVL.

Neurons containing the enzyme aromatic L-amino acid decarboxylase were immunocytochemically localized in the brain stem of the rat. The enzyme occurred as expected in previously well characterized monoaminergic cell groups, and in addition in some nuclei with unknown neurotransmitters. Major aggregates of neurons that were immunoreactive for aromatic L-amino acid decarboxylase but contained neither tyrosine hydroxylase nor serotonin, were found in the pretectal nuclei, the lateral parabrachial nucleus, and the dorsolateral subdivision of the nucleus tractus solitarius. Aromatic L-amino acid decarboxylase was also present in serotonin neurons and the majority of catecholamine cell groups. Dopamine, noradrenaline, and adrenaline cells exhibited characteristic staining intensities to anti-aromatic L-amino acid decarboxylase reflective of relative enzyme levels in the different groups. Some cells in the dorsal motor nucleus of the vagus that were previously classified as dopaminergic lacked immunoreactivity to aromatic L-amino acid decarboxylase.

In urethane-anesthetized rats the i.v. administration of the potent short-acting opioid, fentanyl, elicited inhibition of rhythmic spontaneous reflex increases in vesical pressure (VP) evoked by urinary bladder distension. The duration of fentanyl-induced inhibition was dose-related between 0.8 and 8.0 micrograms/kg i.v. Pretreatment with the opiate receptor antagonist, naloxone HCl (250 micrograms/kg i.v.), antagonized completely fentanyl's effect on VP rhythm. Results similar to those obtained with fentanyl were observed after the bilateral stimulation of opiate receptors in the dorsolateral tegmental-subceruleus region of the pons (DLT-SC region) with D-Ala2-Met5-enkephalinamide (DAMA) microinjections. These inhibitory effects were dose-related and significant between 1.0 and 100.0 ng/site. DAMA in the DLT-SC did not alter basal cardiorespiratory parameters. The systemic and local administration of naloxone reversed rapidly the VP effects elicited by DAMA in the DLT-SC region. Naloxone alone, administered either i.v. or locally in the DLT-SC region, increased the frequency of spontaneous reflex increases in VP. Moreover, pretreatment of the DLT-SC region with naloxone microinjections significantly (55% reduction) decreased the duration of fentanyl (8.0 micrograms/kg i.v.)-induced inhibition of VP rhythm. Neurons in DLT-SC region whose discharge increased just before and throughout spontaneous increases in VP (vesical contraction-related units), were inhibited rapidly and completely by the systemic administration of fentanyl. However, fentanyl elicited a prolonged decremental discharge in neurons of the DLT-SC region with discharge patterns related to vesical relaxation (vesical relaxation-related units).(ABSTRACT TRUNCATED AT 250 WORDS)

Changes in the fractional distribution of cardiac output (FF), organ blood flow, and regional vascular resistance were measured by the isotope dilution technique of Sapirstein using 86Rb as indicator in unanesthetized rats during acute arterial hypertension produced by bilateral lesions of the nucleus tractus solitarii (NTS). After NTS lesions, the FF was significantly reduced in skin, muscle, and colon, increased in ventricular myocardium, spleen, and adrenal glands, and was unchanged elsewhere. Because of a marked reduction in cardiac output (CO) during hypertension, the absolute organ blood flow (FF X CO) was reduced in lesioned rats to 30-40% of control in skin, muscle, and colon and between 60% and 75% of control in most of the remainder of the gastrointestinal tract and renal cortex; it was unchanged in myocardium and endocrine glands. Resistance was substantially increased (4- to 6-fold) in skin, muscle and colon but was only moderately increased (1.5- to 2.5-fold) in the remaining organs. The results indicate that, while NTS lesions will increase resistance in most vascular beds, the response is unequally distributed, influencing skin, muscle, and colon disproportionately to other tissues. Because of an interaction between a reduction in CO and little autoregulation, blood flow is reduced primarily in skin, muscle, and colon. The pattern of redistribution of CO was consistent with the interpretation that NTS hypertension results from interrupting baroreceptor reflexes centrally. The pattern of redistribution of blood flow in rats with NTS lesions differs from that produced by deoxycorticosterone acetate-salt and renal ischemia.

Bilateral electrolytic lesions of the nucleus tractus solitarii (NTS) were made at the level of the obex in seven cats. Within 1 hour the mean arterial pressure (MAP) rose to a maximum of 144 mm Hg (141% of control), and by 7 hours heart rate reached a peak of 236 beats/min (148% of control). The baroreceptor reflexes were abolished. After 24 hours the arterial pressure became extremely labile, with variations of 80-100 mm Hg observed. The lability occurred spontaneously and during behaviors that were self-initiated or elicited by environmental stimuli. The MAP in the lesion group was 144 mm Hg (180% of control) during the day, and 96 mm Hg (120% of control) at night. The lability, measured by the standard deviation, during the day in the lesion group was 4 times greater than in the control group and at night there were no differences. The heart rate of the lesion group was always higher than that of the control group but the lability of both groups was the same. We conclude that lesions of the NTS produced labile hypertension, probably by disinhibition of sympathetic activity through central interruption of the baroreceptor reflexes. The higher, more labile arterial pressures during the day may be caused by uninhibited increases in sympathetic activity elicited by environmental stimuli that are present during the day and absent at night. The daily variation of pressure may also be caused by somatomotor activity or by a daily rhythm of sympathetic activity which is unmasked by the lesions.

Bilateral anodal lesions performed with stainless steel electrodes placed either in the anterior medial (AMH) or lateral (ALH) hypothalamus, or in the ventromedial nucleus (VMH), induced in unrestrained rats the rapid development of arterial hypertension, tachycardia and death. Similarly placed cathodal lesions performed with platinum electrodes failed to elicit the cardiovascular syndrome. The electrical stimulation of the AMH, ALH or VMH caused an increased in the arterial blood pressure in anesthetized rats. This pressor response was characteristically biphasic and consisted of a sharp increase in arterial pressure at the onset of the stimulation, followed by a second elevation at the end of the stimulation. The hypertension evoked either by lesions or by stimulations of the hypothalamus, appeared to depend largely on a neurally mediated release of adrenal medullary catecholamines, and to some extent on the activation of the sympathetic vasoconstrictor fibers. Bilateral adrenalectomy, or adrenal demedullation, prevented the hypertension evoked by lesions, and selectively blocked the important secondary phase of the pressor response elicited by stimulation, but did not affect the primary phase. The latter was specifically eliminated by the destruction of the sympathetic vasomotor axons with 6-hydroxydopamine (6-OHDA). On the other hand, the tachycardia evoked by lesions or stimulations of the medial hypothalamus, resulted from an increase in sympathetic neural discharges to the heart, and it was abolished either by beta-receptor blockade with sotalol or by chemical sympathectomy with 6-OHDA. In contrast, the tachycardia occurring after lesions of the lateral hypothalamus was entirely due to circulating adrenal medullary catecholamines and it was eliminated by adrenalectomy. It is concluded that acute hypertension and tachycardia produced by anodal lesions performed with stainless steel electrodes results from the excitation of the hypothalamus, possibly due to the irritative action of the metallic ions deposited at the lesion sites. The observations of cardiovascular responses entirely due to adrenomedullary secretions suggests that the control of the adrenal medulla is at least partially distinct from that of the sympathetic vasoconstrictor and cardiac fibers, at the rostral hypothalamic level.

An endogenous substance in brain, clonidine-displacing substance, binds to the same receptor populations as clonidine and is biologically active. Since receptor binding sites can be modeled by using specific antiligand antibodies, we tested the hypothesis that polyclonal antibodies raised in rat and rabbit against the clonidine analog p-aminoclonidine coupled to hemocyanin would recognize compounds structurally related to clonidine, including clonidine-displacing substance. Binding to anti-p-aminoclonidine antibodies was examined by using a competitive radioimmunoassay with tritiated p-aminoclonidine as the radioligand. Central vasodepressor agents that, like clonidine, are known to bind with high affinity to both imidazole sites and alpha 2-adrenergic receptors in brain inhibited radioligand binding to anti-p-aminoclonidine antibodies. All of these agents contain imidazol(in)e and phenyl ring moieties as part of their chemical structures (e.g., oxymetazoline); a number of other compounds without one or both of these rings failed to cross-react with the antisera. Clonidine-displacing substance, partially purified from bovine brain, also inhibited specific radioligand binding to anti-p-aminoclonidine antibodies. The inhibition was dose dependent and high affinity (IC50, 4 Units). The endogenous substance had no effect on the apparent affinity of the antibodies for the radioligand, but blocked a specific number of binding sites. Immunoprecipitation experiments showed that authentic clonidine-displacing substance, that which displaces tritiated p-aminoclonidine binding to membrane receptors, is recognized by anti-p-aminoclonidine antibodies. We conclude that a unique subset of structural determinants required for ligand interaction with both imidazole and alpha 2-adrenergic receptors is critical for binding to anti-p-aminoclonidine antibodies, and that since clonidine-displacing substance is recognized by highly clonidine-specific antisera, it may also contain these determinants within its structure, namely the imidazol(in)e and phenyl ring systems.

Recent studies have demonstrated that the neurons of the lower brainstem that are responsible for maintaining normal levels of arterial pressure reside in a specific area of the rostral ventrolateral medulla. In rat, the critical zone corresponds to a small region containing a subpopulation of the adrenergic C1 group, defined immunocytochemically by the presence of the epinephrine-synthesizing enzyme phenylethanolamine N-methyltransferase. Neurons of this region (the C1 area), possibly including the adrenergic neurons, directly innervate preganglionic neurons in the spinal cord, and are tonically active and sympathoexcitatory. The excitatory transmitter released into the spinal cord is unknown. The discharge of C1 area neurons is locked to the cardiac cycle and, in turn, leads to firing of sympathetic preganglionic neurons. The C1 area neurons are inhibited by baroceptor input and mediate the vascular component of baroceptor reflexes. They also mediate somato-sympathetic pressor responses from skin and muscle and participate in reflex responses to hypoxia. The neurons are directly innervated by local neurons containing gamma-aminobutyric acid, acetylcholine, enkephalin, and substance P, all of which modulate arterial pressure. The C1 area is the site of the hypotensive actions of clonidine. Clonidine appears to act on imidazole receptors in the C1 area to lower arterial pressure. The natural ligand for these receptors may be a newly defined substance in brain, clonidine-displacing substance. Neurons of the C1 area appear to be the critical neuronal group governing the normal resting and reflex control of arterial pressure. They may play a critical role in the maintenance of elevated arterial pressure in hypertension and as a site of action of antihypertensive drugs.

Electrical stimulation of the rostral fastigial nucleus (FN) in anesthetized, paralyzed, and artificially ventilated rats with a 10-s stimulus train elicited a stimulus-locked elevation of arterial pressure (AP) and heart rate (HR) (the fastigial pressor response, FPR) and elevated plasma catecholamines (CA) within 20 s from the onset of stimulus. Norepinephrine (NE) increased from 139 +/- 24 to 280 +/- 43 pg/ml (P less than 0.05, n = 8) and epinephrine (E) from 70 +/- 26 to 360 +/- 107 pg/ml (P less than 0.02, n = 8). Acute adrenalectomy increased basal plasma NE (362 +/- 108 pg/ml, P less than 0.05, n = 6) and reduced E (9 +/- 4 pg/ml, P less than 0.02, n = 6). The magnitude and duration of the FPR and the relative increase of NE were unchanged; however, the elevation of E was abolished. Chemosympathectomy, produced by 6-hydroxydopamine hydrobromide (100 mg/kg iv, 24 h before the experiment), lowered resting AP (from 122 +/- 2 to 77 +/- 1 mmHg, P less than 0.001) and NE (16 +/- 5 pg/ml, P less than 0.01), but not E. After chemosympathectomy, FN stimulation induced a pressor response of greater magnitude and longer latency and duration than in controls, increased NE 3.5-fold (from 16 +/- 5 to 56 +/- 14 pg/ml, P less than 0.05, n = 5) and E 9-fold (from 38 +/- 11 to 336 +/- 88, P less than 0.05, n = 5). The increases in CA were abolished by adrenalectomy. Chemosympathectomy shifted the pressor-dose-response curves of NE and E to the left; thus, the enhanced pressor response to FN stimulation after chemosympathectomy was possibly a consequence of supersensitivity to circulatory CA. Stimulation of cerebellar FN increased plasma CA, as a consequence of coexcitation of both neural and adrenomedullary components of the autonomic nervous system. However, in rats with intact sympathetic nerves the release of adrenomedullary CA did not contribute to the elevation in AP.

We studied the effects of acute bilateral electrolytic lesions of the nucleus tractus solitarii (NTS) on regional cerebral blood flow (rCBF) and its autoregulation in rats anesthetized (alpha-chloralose, 40 mg/kg), paralyzed (tubocurarine), and artificially ventilated. rCBF or regional cerebral glucose utilization (rCGU) was measured 30 min after NTS lesions, by the 14C-iodoantipyrine technique or 2-deoxyglucose method, respectively. Cerebrovascular autoregulation was assessed in groups of 4-5 rats at three levels of arterial pressure (AP): 90, 125, and 140 mmHg. AP was lowered by hemorrhage or elevated by intravenous infusion of phenylephrine. NTS lesions did not alter rCBF at 125 mmHg (P greater than 0.05) but resulted in loss of autoregulation (P less than 0.05, analysis of variance). In contrast, lesions of the cuneate nucleus or transection of the baroreceptor afferents did not alter autoregulation. NTS lesions did not affect the reactivity of the cerebrovascular bed to hypercarbia (PaCO2 57.4 +/- 1; n = 5) or hypocarbia (PaCO2 24.4 +/- 1; n = 5) nor the rCGU in any brain regions (P greater than 0.05; n = 5). We conclude that lesions of the NTS impair cerebrovascular autoregulation. The effect is not due to changes in metabolism, nonspecific effects of the lesions, vasoparalysis, or interruption of the baroreceptor reflex arch. Neural pathways originating in or passing through the NTS can regulate the cerebrovascular autoregulation of the entire brain.

A substance has been isolated from brain which potently inhibits the binding of clonidine to brain membranes (clonidine displacing substance, CDS). We sought to determine if CDS is biologically active on smooth muscle. CDS had no effect on vascular smooth muscle. In contrast, CDS potently contracted rat gastric fundus strips in a dose dependent manner. The contractile effect of CDS was not blocked by antagonists selective for biologically active substances known to contract the fundus strip. These results demonstrate that CDS has a unique and potent ability to selectively contract smooth muscle.

Clonidine acts within the rostral ventrolateral medulla to lower arterial pressure. The receptor mechanism for this action is unknown. In the rostral ventrolateral medulla clonidine binds not only to the alpha 2-adrenergic receptor but also to a novel class of sites which are specific for imidazolines and imidazoles and are distinct from adrenergic or histaminergic receptors. In order to distinguish whether a putative imidazole receptor or the alpha 2-receptor mediates the clonidine hypotensive response, a series of clonidine analogues were tested (1) in radioligand binding assays of their affinity at imidazole and alpha 2-adrenergic receptors, and (2) by microinjection into the rostral ventrolateral medulla of anaesthetized rats to measure their capacity to lower arterial pressure. The hypotensive response elicited by the test agents was strongly correlated with affinity at imidazole sites (r = 0.92) but not with alpha 2-adrenergic affinity (r = 0.36). An imidazole receptor in the rostral ventrolateral medulla may participate in the hypotensive action of clonidine.

Met-enkephalin (1 microM) increased the spontaneous release of endogenous glutamate (+155%), taurine (+80%) and glycine (+50%) from rat striatal slices, but was without effect in the cerebral cortex. This effect was calcium-dependent and significantly reduced in the presence of naloxone (1 microM). Naloxone alone had no effect on release of any substance. Release of aspartate, gamma-aminobutyric acid and [3H]acetylcholine was not significantly affected. Met-enkephalin did not affect potassium (35 mM)-evoked neurotransmitter release, nor did it affect the uptake of D-[3H]aspartate, [3H]taurine or [14C]glycine. The data indicate that opioid receptor activation selectively increases the spontaneous calcium-dependent release of putative amino acid neurotransmitters in the rat striatum.

Changes in cardiodynamics and regional blood flow were examined in chronically prepared paralyzed cats during seizures induced electrically by transcerebral or direct cortical stimulation or by administration of flurothyl ether (Indoklon) or pentylenetetrazol (Metrazol). Transcerebral and chemical stimuli produced the greatest vascular responses. During seizures there was an abrupt elevation of arterial pressure unassociated with consistent changes in heart rate. Vascular resistance was increased in femoral, renal and mesenteric arteries with variable reductions in blood flow. Resistance was decreased and flow passively increased in the common carotid artery reflecting the loss of cerebral autoregulation. Cardiac output was unchanged. With seizures associated with large elevations of arterial pressure, the central venous and left ventricular end-diastolic pressures were markedly increased indicating incipient congestive failure. The pressor response was blocked by alpha-adrenergic blockade with phentolamine. Increased regional vascular resistance was abolished by regional sympathectomy. While either adrenalectomy or treatment with 6-hydroxydopamine alone failed to abolish the pressor response, combined, they did. Such treatment unmasked an atropine-sensitive bradycardia. The pressor response with seizures is a consequence of increased vascular resistance in viscera and muscles due to widespread activation of sympathetic neurons and release of adrenomedullary catecholamines. Co-activation of cardiovagal and cardiosympathetic neurons may underlie some associated arrhythmias. Cardiovascular events may severe, by redistribution of the cardiac output, to assure increased availability of oxygen and nutrients to brain to meet the metabolic demands of convulsions.

The human fetal sympathetic ganglia were studied using the indirect peroxidase-antiperoxidase PAP method for immunocytochemical demonstration of three catecholamine-synthesizing enzymes, tyrosine hydroxylase (TH), dopamine-beta-hydroxylase (DBH) and phenylethanolamine-N-methyltransferase (PNMT) as well as the neuropeptides leucine (Leu5)-enkephalin and substance P. The neuroblasts of the ganglia showed intense peroxidase immunoreactivity for TH, moderate reaction to DBH, and no reaction to PNMT. The small intensely fluorescent (SIF) cells situated along the blood vessels also showed positive labelling for only two enzymes, TH and BDH. The immunocytochemical localization of these enzymes suggests that both neuroblasts and SIF cells synthesize noradrenalin. Neither the neuroblasts nor SIF cells showed a reaction to substance P, and only the SIF cells contained enkephalin-like immunoreactivity. The role of enkephalin in the noradrenalin-containing SIF cells is unknown, but may be related to neuromodulation of ganglionic transmission.