Cat Cerebral Arteries Are Functionally Innervated by Serotoninergic Fibers From Central and Peripheral Origins

There is increasing evidence that cerebral blood vessels receive serotoninergic innervation, the main origin of which is the brain stem nuclei. Serotonin levels in cerebral blood vessels are increased after administration of tryptophan or a monoamine oxidase inhibitor and decreased after injecting the animals with p-chlorophenylalanine or destroying the dorsal and median raphe nuclei.¹²³⁴ Furthermore, tryptophan hydroxylase activity can be also estimated either in vivo or in vitro. The administration to the animals of an inhibitor of the aromatic amino acid decarboxylase brings about an increase in 5-hydroxytryptophan (5-HTP) content in this kind of vessel,⁵ which is reduced when the main serotoninergic pathway is destroyed with 5′-7′ dihydroxytryptamine.⁶ Cell-free extracts of rat cerebral arteries possess tryptophan hydroxylase activity that appears decreased after dorsal raphe nucleus destruction.⁷ This serotoninergic innervation seems functionally active. Lesions of the dorsal raphe nucleus enhance the contractile response to serotonin of isolated cat middle cerebral artery,⁸ and the electric stimulation of the dorsal raphe nucleus reduces cerebral blood flow in several brain areas.⁹

Most of the morphological evidence, however, shows that serotonin is bound to sympathetic nerve endings.¹⁰¹¹ The serotonin-like immunohistofluorescence found in cerebral blood vessels disappears after the animals are subjected to cervical gangliectomy¹²¹³ or treated with catecholamine uptake blockers before they are killed and when the blood vessels are perfused with saline solution.¹²¹³¹⁴ Thus, the serotonin revealed by histochemistry would be the result of its uptake by the sympathetic nerve terminals during the isolation procedure. Nevertheless, the morphological evidence is not unanimous on this issue. Serotonin-like immunoreactivity does not superimpose immunoreactivity to noradrenaline in rabbit cerebral arteries.¹⁵ Furthermore, application of horseradish peroxidase in cat middle cerebral artery results in horseradish peroxidase–labeled neurons in dorsal raphe nucleus.¹⁶ Attempts to show the presence of tryptophan hydroxylase have also yielded conflicting results. Although Cohen et al¹⁷ observed that the tryptophan hydroxylase–like immunoreactivity found in these vessels disappeared after cervical sympathectomy, they could not demonstrate the same immunoreactivity in the superior cervical ganglia.

The aim of the present work was to test whether cat cerebral arteries receive a functional serotoninergic innervation and to determine its origin. We assayed tryptophan hydroxylase activity in brain base arteries from untreated animals. Afterward, this was related to nerve structures by producing lesions of the dorsal and median raphe nuclei and performing cervical gangliectomy. In addition, the possible involvement of serotonin from a central origin in the contractile response to tyramine of cat isolated middle cerebral artery was studied after destruction of dorsal and median raphe nuclei and superior cervical ganglia removal. Since tyramine induces a contraction of cerebral arteries that is strikingly higher than that induced by noradrenaline¹⁸¹⁹²⁰ and serotoninergic nerve endings possess features in common with adrenergic terminals, one can assume that tyramine might release a substance different from noradrenaline in these vessels.

Materials and Methods

Mongrel cats weighing 2 to 2.5 kg were used in the present study. The animals were housed in the proper facilities, complying with the European Community directive 86/609/CEE and Spanish legislation (RD 223/1988) regarding the care of the animals used in experimentation and for other scientific purposes. The experiments reported here were approved by the Biosafety and Animal Care Unit Committee (Comisión de Bioseguridad y Gabinete Veterinario) of the Faculty of Medicine of the Autónoma University of Madrid.

The cats were injected with sodium pentobarbital (35 mg/kg IP) and killed by exsanguination, and their brains were quickly removed. The circle of Willis as well as some of its branches was dissected out, frozen on dry ice, and stored at −15°C for posterior assay of enzyme activity.

After overnight fasting, some anesthetized animals (sodium pentobarbital, 35 mg/kg IP) received atropine (0.5 mg IP each), and their dorsal and median raphe nuclei were destroyed by thermocoagulation. The electrode was implanted according to the coordinates described by Reinoso-Suárez,²¹ and the animals were killed 15 days later. The sham-operated group were subjected to the same surgical procedure but did not receive any current. Segments of middle cerebral arteries were dissected out to study their response to tyramine; the rest of the vessels were kept frozen for tryptophan hydroxylase estimation, and the brains were stored for histological control of the lesions. Only the cerebral blood vessels from animals with the correct lesions were used.

Another group of animals that fasted overnight were anesthetized with sodium pentobarbital (35 mg/kg IP) and injected with atropine 0.5 mg IP each. Under aseptic conditions both superior cervical ganglia were excised, and the right one was stored at −15°C for posterior assay of tryptophan hydroxylase activity. The sham-operated animals were subjected to the same surgical procedure without ganglia removal. The animals were killed 15 days later. Segments of middle cerebral arteries were excised out for isometric tension recording, and the rest of the circle of Willis was kept at −15°C for posterior enzyme activity assay.

The assay of tryptophan hydroxylase was similar to that described by Meek and Neckers.²² The tissues were homogenized by sonication in 0.05 mol/L tris(hydroxymethyl)aminomethane (Tris) (pH 7.4) buffer containing 10⁻³ mol/L mercaptoethanol and 0.05% Triton X-100. We used a volume of 500 μL for the samples of cerebral arteries and a volume of 300 μL for the ganglia. The homogenates were centrifuged at 12 000 rpm for 10 minutes in a Beckman Microfuge. One hundred microliters of the supernatant was added to an Eppendorf microcentrifuge tube with 20 μL of a standard reaction mixture containing, in 0.05 mol/L Tris (pH 7.4) buffer, tryptophan (2 mmol/L), 6-methyltetrahydropterine (16 mmol/L), mercaptoethanol (1 mmol/L), and catalase (2.5 mg/mL). The tubes were incubated in a Suppelco Blok Heater at 37°C for 90 minutes. Some cell-free extracts from cerebral blood vessels were incubated under the same conditions for 180 minutes. The reaction was stopped by adding 20 μL of 11.64 mol/L HClO₄. Blank or 0-hour tubes were made by adding the perchloric acid to the supernatants and shaking them before adding the reaction mixture. The samples were centrifuged at 12 000 rpm for 5 minutes, and 5-HTP was assayed in 20 μL of the supernatant by high-performance liquid chromatography (HPLC) with fluorometric detection. Proteins were determined in the precipitates of the first homogenates by the method of Lowry et al,²³ with 5-HTP final concentrations referred to them. The HPLC system consisted of a pump (GILSON, model 305) with a manometric module (GILSON, model 805), giving a flow rate of 1200 mL/min, a Rheodyne injection valve with a 20-μL loop, and a reverse-phase column (μBondapak C₁₈, Waters) with guard column (μBondapak C₁₈/Corasil, Waters). 5-HTP was detected with an spectrofluorometric detector (F-2000 fluorescence spectrophotometer, Hitachi) at 286-nm excitation and 343-nm emission wavelengths. The mobile phase was prepared according to Lackovic et al²⁴ with 4% methanol.

In some instances, the standard reaction mixture was devoid of tryptophan or 6-methyltetrahydropterine or contained added 6-fluorotryptophan, p-chlorophenylalanine, or α-methyltyrosine (all 1 mmol/L).

Arterial segments 2 mm in length were set up for isometric tension recording according to Nielsen and Owman²⁵ as described elsewhere.¹⁸ Briefly, this consists of passing two fine stainless steel pins through the lumen of the segment; one of them is fixed to the wall of the isolated organ bath, and the other is connected to a strain gauge for isometric tension recording. The latter pin is parallel to the former and movable, allowing the application of resting tension at right angles to the long axis of the vascular cylinder. The recording system included force transducers (FT03 Grass) and a polygraph (Grass 7-E). A resting tension of 0.3 g was applied to the tissue and was readjusted every 15 minutes during a 60- to 90-minute equilibration period before any drug was added to the bath medium.

The bath medium consisted of 6 mL of Krebs-Henseleit solution kept at 37°C and continuously bubbled with a 5% CO₂/95% O₂ mixture, which gave a final pH of 7.2 to 7.4. The composition of the Krebs-Henseleit solution was as follows (in mmol/L): NaCl 115, KCl 4.6, CaCl₂ 2.5, KH₂PO₄ 1.2, MgSO₄ · 7 H₂O 1.2, NaCOH₃ 25, and glucose 11.1.

The drugs were dissolved in physiological saline solution containing 0.01% (wt/vol) ascorbic acid. Dose-response curves to tyramine and noradrenaline were performed in a cumulative manner. The changes in tension are expressed in milligrams. To study the effect of ketanserine (10⁻⁶ mol/L) on the vascular response to tyramine, the serotoninergic antagonist was added to the bath medium 10 minutes before determining the dose-response curve and remained until its completion. At the end of the experiments the arterial segments were challenged with 80 mmol/L KCl to test the vascular muscle responsiveness.

The results were statistically analyzed by means of Student’s t test for unpaired data.

l-Tryptophan, 5-hydroxy-l-tryptophan, 6-fluoro-dl-tryptophan, dl-6-methyl-5,6,7,8,-tetrahydropterine, mercaptoethanol, tyramine, norepinephrine bitartrate, p-chlorophenylalanine, α-methyltyrosine, and catalase from bovine liver were purchased from Sigma; Tris was purchased from Serva. Ketanserine tartrate was kindly supplied by Janssen Pharmaceutica. The rest of the reagents were of high purity suitable for HPLC.

Results

When cell-free extracts of untreated cat cerebral arteries remained in contact with the standard reaction mixture, the production of 5-HTP increased with time (Fig 1). No 5-HTP was detected when the reaction did not take place or the substrate or the cofactor of the reaction mixture was absent (Fig 1 and Table 1, respectively). The addition of 6-fluorotryptophan to the reaction mixture brought about a significant decrease in the production of 5-HTP, whereas this was unaffected in the presence of 1 mmol/L p-chlorophenylalanine or 1 mmol/L α-methyltyrosine (Table 1).

Two weeks after median raphe nucleus destruction, tryptophan hydroxylase activity in cat brain base arteries was significantly reduced (Fig 2). A similar result was obtained if the area affected by the lesion was dorsal raphe nucleus or the animals had been subjected to sympathectomy (Fig 2).

Tryptophan hydroxylase activity measured in superior cervical ganglia was significantly reduced in the presence of 6-fluorotryptophan but remained unchanged when α-methyltyrosine was present in the incubation medium (Table 2).

Removal of both superior cervical ganglia evoked a significant reduction in the contractile response to tyramine of isolated cat middle cerebral arteries at the first three doses (Fig 3A). This response was further decreased in the presence of 10⁻⁶ mol/L ketanserine (Fig 3B).

Destruction of the dorsal raphe nucleus induced an enhancement in the contractile response to tyramine of isolated arterial segments at all concentrations used (Fig 4). When response curves were determined in cerebral blood vessels from cats whose median raphe nucleus had been destroyed 2 weeks before, only the effects of the three last concentrations of tyramine were significantly higher than those of the corresponding controls (Fig 5). The contractile response to noradrenaline (10⁻⁹ to 10⁻⁵ mol/L) was not affected by these treatments (results not shown).

None of the treatments altered the contractile response to 80 mmol/L KCl (data not shown).

Discussion

The present results support the existence of serotoninergic innervation impinging on cat brain base arteries from peripheral and central origins.

Tryptophan hydroxylase was assayed in these blood vessels by means of a biochemical method. When the standard reaction mixture was in contact with cell-free extracts of cat cerebral arteries, we observed a time-dependent 5-HTP production. This 5-HTP was synthesized from tryptophan by the action of tryptophan hydroxylase since it was absent after removal of the amino acid or the cofactor from the medium or diminished in the presence of an enzyme inhibitor such as 6-fluorotryptophan. The possibility of its synthesis by the action of tyrosine hydroxylase is very unlikely because α-methyltyrosine did not affect 5-HTP formation. The present results agree with experiments demonstrating tryptophan hydroxylase activity in cerebral blood vessels in vivo⁵⁶ as well as in vitro.⁷

The lack of effect of p-chlorophenylalanine on tryptophan hydroxylase activity indicates that this drug seems to be effective in inhibiting serotonin formation when administered in vivo, but it is a poor inhibitor of the enzyme in vitro.²⁶ This result might be also due to a higher resistence to inhibition of the enzyme present in cat cerebral arteries since p-chlorophenylalanine at the same concentration reduces tryptophan hydroxylase activity in rat cerebral arteries, whereas 6-fluorotryptophan practically abolishes it.⁷

On the other hand, destruction of dorsal and median raphe nuclei, as well as cervical sympathectomy, brings about a significant decrease in the enzymatic activity, indicating that the enzyme is bound, at least in part, to nerve fibers whose cell bodies are located in those brain stem nuclei and in the superior cervical ganglia. Previous data showing that the serotonin content of cat cerebral arteries appeared reduced when the animals were subjected to similar procedures² suggest the existence of serotoninergic innervation in cat cerebral arteries from both peripheral and central origins, although they were challenged later by morphological findings. The present work also confirms biochemical results indicating that cerebral arteries receive serotoninergic innervation originating from dorsal and median raphe nuclei.¹³⁴ This would explain the decrease in cerebral blood flow obtained in several brain areas when the dorsal raphe nucleus is stimulated.⁹

Morphological evidence suggests that cerebral blood vessels do not possess a true serotoninergic innervation and that the serotonin found in them is a mere artifact because of its uptake by the sympathetic nerve endings.^11121314 Furthermore, attempts to reveal tryptophan hydroxylase immunohistochemically failed to give conclusive results.¹⁷ Indeed, they could show the presence of the enzyme associated with sympathetic innervation of these vessels but were unable to find it in the superior cervical ganglia. According to the present data, enzyme activity is reduced after cervical sympathectomy and can be assayed in superior cervical ganglia, which suggests the existence of sympathetic fibers containing serotonin as neurotransmitter. They also point out that the discrepancies between morphological evidence and biochemical results might be due to some methodological reason, which needs further investigation.

This serotoninergic sympathetic innervation seems to be functionally active since cervical gangliectomy induces a great enhancement in the contractile response to serotonin of isolated cat middle cerebral artery segments.²⁷ Whether this serotonin coexists with noradrenaline in the same fibers or is located in independent ones cannot be determined from the present results. If serotonin is found in sympathetic nerve terminals of other animal species, including humans, this will reinforce the role of these fibers in regulating cerebral blood flow because this amine is a more potent constrictor agent of cerebral arteries than noradrenaline.¹⁸

The results obtained with tyramine also confirm the existence of a functional serotoninergic innervation, at least from a central origin.

The contractile response to the amine after performing cervical gangliectomy to eliminate noradrenaline from vessels was further reduced in the presence of ketanserine, suggesting the participation of serotonin in the effect of tyramine. The release of serotonin by tyramine would explain why contractions elicited by medium-high concentrations of tyramine reached the same level in arterial segments from sham-operated and sympathectomized animals since, as indicated, superior cervical ganglia removal brings about supersensitivity to serotonin in these vessels that might compensate for the loss of the adrenergic component. This effect of cervical sympathectomy on the response of cat cerebral arteries to tyramine has already been described.²⁰

Furthermore, destruction of brain stem nuclei altered the tyramine-induced contraction in a manner similar to its effect on serotonin response.⁸ Thus, the production of lesions of the dorsal raphe nucleus resulted in supersensitivity to serotonin in isolated cat middle cerebral artery and increased the contractile response to tyramine at all doses used. This enhanced tyramine response would be the result of serotonin release from other sources, evoking an augmented response because of the development of supersensitivity. Destruction of the median raphe nucleus, however, only slightly increased the responses to both serotonin and tyramine. This would suggest that cat middle cerebral artery is preferentially innervated by the dorsal raphe nucleus. An increased response of tyramine due to subarachnoid hemorrhage production during the surgical procedure must be ruled out because there was no supersensitivity to noradrenaline, as would occur after experimental subarachnoid hemorrhage.²⁸

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

• Download figure
• Download PowerPoint

Footnotes