Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Serotonin can modulate sympathetic neurotransmission in blood vessels (Humphrey, 1978; Marin et al., 1981; Cohen, 1985; Gaw et al., 1990; Molderings et al., 1990; Göthert et al., 1991; Barrús et al., 1993). For the most part, these effects of serotonin have resulted primarily in inhibition of norepinephrine release from sympathetic nerves in blood vessels. However, an indirect effect of high concentrations of serotonin to enhance norepinephrine release has been reported; an effect thought to be mediated by the uptake of serotonin into adrenergic nerves and the subsequent displacement of norepinephrine (Marin et al., 1981; Gaw 1990). With regard to the receptor mechanisms mediating the effects of serotonin on sympathetic neurotransmission, based on analogy to brain serotonergic neurons, both 5-hydroxytryptamine (5-HT)_1A and 5-HT_1B/1D receptors may be involved in regulating neurotransmitter release from noradrenergic neurons.

The rabbit saphenous vein is a tissue used most recently in the identification of several antimigraine compounds (Beer et al., 1994; Perez et al., 1995). Activation of the 5-HT_1B/1D receptor or a receptor highly similar to the 5-HT_1B/1D receptor is likely to be responsible for the direct contractile effects of serotonin agonists in the rabbit saphenous vein (Cohen et al., 1997). However, the rabbit saphenous vein is also sympathetically innervated (De Mey and Vanhoutte, 1978; Alabaster et al., 1985; Levitt and Hieble, 1986) and several of the antimigraine drugs in development have high 5-HT_1A and 5-HT_1B/1D receptor affinities. Thus, we designed experiments to explore the effects of 5-HT_1B/1D and 5-HT_1A receptor activation on the modulation of neurotransmission in the rabbit saphenous vein. We used 8-OH-DPAT [(±)-8-hydroxy-dipropylaminotetralin hydrobromide] and LY228729 [()-4-(dipropylamino)-1,3,4,5-tetrahydrobenz[CD]indole-6-carboxamide] (Foreman et al., 1993), two potent and selective 5-HT_1A receptor agonists; sumatriptan, an agonist with high affinity at 5-HT_1B/1D receptors; and two highly selective antagonists, GR127935 [N-[4-methoxy-3-(4-methyl-1-piperazinyl)phenyl]-2'-methyl-4'-(5-methyl-1,2,4-oxadiazol-3-yl)-1,1'-biphenyl-4-carboxamide], a 5-HT_1B/1D receptor antagonist (Skingle et al., 1996; Terrón, 1996), and WAY 100635 (N-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinyl-cyclohexanecarboxamide), a 5-HT_1A receptor antagonist (Khawaja et al. 1995). These studies document an important effect of 5-HT_1A but not 5-HT_1B/1D receptors to enhance neurotransmitter release in the rabbit saphenous vein.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of Vascular Tissue. Male New Zealand White rabbits (6-12 kg; Hazelton, Kalamazoo, MI) were sacrificed by a lethal dose of sodium pentobarbital (200 mg) injected into the ear vein. Tissues were dissected free of connective tissue, cannulated in situ with polyethylene tubing (PE50, outside diameter = 0.97 mm), and placed in Petri dishes containing Krebs' bicarbonate buffer (see below). The tips of two 30-gauge stainless steel hypodermic needles bent into an L-shape were slipped into the polyethylene tubing. Vessels (approximately 1.0 cm in length) were gently pushed from the cannula onto the needles. The needles were then separated so that the lower one was attached with thread to a stationary glass rod and the upper one was tied with thread to the transducer.

Contractile Responses. Cumulative agonist concentration-response curves or field-stimulated responses were generated in tissues either exposed to vehicle or antagonist for 1 h. No ring was used to generate more than two agonist concentration-response curves. All results are expressed as mean ± S.E. where n represents the number of rings examined. The data are expressed as a percentage of the response to 67 mM KCl administered initially in each tissue. The log EC₅₀ values were determined by least-squares linear regression analysis of the linear portion of the mean concentration-response curves and were reported as the negative logarithm of the agonist concentration required for half-maximal response.

Modulation of Vascular Serotonin Content. Male New Zealand White rabbits (2-4 kg; Hazelton, Kalamazoo, MI) were housed in hanging wire cages at 22°C with lights on from 0730 to 1930 h. Pargyline hydrochloride was dissolved in distilled water and injected at a dose of 30 mg/kg s.c., and rabbits were sacrificed 3 h after injection. dl-p-chlorophenylalanine methyl ester (PCPA) was dissolved in distilled water and injected at a dose of 300 mg/kg, s.c., and rabbits were sacrificed 72 h after injection. 6-Hydroxydopamine hydrobromide (OHDA) was dissolved in 0.01 N HCl and injected at a dose of 100 mg/kg s.c., and rabbits were sacrificed 24 h after injection. Anesthetized rabbits were sacrificed by cervical dislocation, and tissues were quickly dissected, rinsed in saline, frozen on dry ice, and stored at 70°C before analyses. The saphenous vein was sonicated in 1 ml of 0.1 N trichloroacetic acid, centrifuged at 12,000g for 2 min, and filtered to remove particulate matter. Tissue norepinephrine and serotonin concentrations were measured by HPLC with electrochemical detection using a modification of the method of Fuller and Perry (1989). Samples (25-30 µl) were injected onto a Waters 5C18-AR analytical column at a temperature of 40°C and a flow rate of 1.5 ml/min, a potential of 600 mV and sensitivity of 5 nA/V. Elution buffer was 0.1 M monochloroacetic acid, 0.1 mM EDTA, 500 mg/l 1-octanesulfonic acid sodium salt, 1% tetrahydrofuran, and 4.5% acetonitrile, pH 2.8. Peaks were measured and compared with standards added to tissue sonicates.

Chemicals. 5-HT, pargyline hydrochloride, PCPA and OHDA were purchased from Sigma Chemical Co. (St. Louis, MO). 8-OH-DPAT was purchased from Research Biochemicals Inc. (Wayland, MA). Sumatriptan, GR127935, WAY 100635, and LY228729 were provided by the Lilly Research Laboratories. Prazosin was a gift from Pfizer Laboratories.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Field-Stimulated Contractile Response in Rabbit Saphenous Vein. Field stimulation (1-16 Hz) induced a marked contractile response in the rabbit saphenous vein, an effect that increased with increasing voltage (20-40 V; Fig. 1). The contractile response to field stimulation, throughout the voltage range studied was abolished in the presence of tetrodotoxin (TTX; 100 ng/ml; Fig. 1), consistent with previous information that TTX could abolish the response to 3 Hz stimulation frequency in the saphenous vein (Levitt and Hieble, 1986). Thus, the contractile response to field stimulation in the rabbit saphenous vein resulted from neurotransmitter release even at the highest frequency and voltage used. Based on these observations, 30 V was selected as an appropriate voltage to use for the evaluation of mechanisms that would modulate (either increase or decrease) neuronal release mechanisms in this tissue.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Effect of increasing frequency of field stimulation at 20, 30, and 40 V to contract the rabbit saphenous vein in the presence and absence of TTX (100 ng/ml). Points are mean values and vertical bars represent S.E.M. for the number of tissues indicated in parentheses.

6

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Effect of prazosin (106 M), GR127935 (106 M), and WAY 100635 (106 M) on the contractile response to field stimulation in the rabbit saphenous vein. Points are mean values and vertical bars represent S.E.M. for the number of preparations indicated in parentheses.

Effect of GR127935 and WAY 100635 on Field-Stimulated Contraction. Because activation of 5-HT_1B and/or 5-HT_1D receptors has been evoked as a mechanism that can modulate release of neurotransmitters (El Mansari and Blier, 1996; Bühlen et al., 1996; Stanford and Lacy, 1996), we examined the effect of GR127935 (10⁶ M), a selective 5-HT_1B/1D receptor antagonist (Skingle et al., 1996) on the contractile response to field stimulation in the rabbit saphenous vein (Fig. 2). Unlike other preparations in which 5-HT_1B and/or 5-HT_1D receptor activation inhibited norepinephrine release from nerves (Göthert et al., 1986; Molderings et al., 1990), GR127935 did not alter the contractile response to field stimulation in the rabbit saphenous vein.

6

Effect of GR127935 and WAY 100635 on Contraction to Sumatriptan. Because receptor densities and sensitivity to agonists and antagonists may vary among tissues, we needed to establish the selectivity of the antagonists in this preparation. To do this, we first examined the effect of WAY 100635 and GR127935 on the contractile response to sumatriptan, a prototypic 5-HT_1B/1D receptor agonist (Fig. 3). WAY 100635, consistent with its selectivity as a 5-HT_1A receptor antagonist, did not inhibit the contractile response to sumatriptan in the rabbit saphenous vein. In contrast, as previously reported for canine arteries (Terrón, 1996), GR127935 (10⁸ M) potently inhibited the contractile response to sumatriptan in the rabbit saphenous vein, supporting the selectivity of GR127935 as a 5-HT_1B/1D receptor antagonist in the rabbit saphenous vein. Thus, the inability of GR127935 at 10⁶ M to modulate neurotransmission in the rabbit saphenous vein occurred in spite of its high antagonist affinity at 5-HT_1B/1D-like receptors in the rabbit saphenous vein.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Effect of prazosin (106 M), GR127935 (108 M), and WAY 100635 (106 M) on the contractile response to sumatriptan in the rabbit saphenous vein. Points are mean values and vertical bars represent the S.E.M. for number of tissues indicated in parentheses.

Effect of GR127935 and WAY 100635 on Contraction to Norepinephrine. Tissues were contracted with norepinephrine to examine the ability of these antagonists to block  receptors in the rabbit saphenous veins. Because the contraction to field stimulation resulted from activation of  receptors, the  receptor antagonist activity of WAY 100635, if it existed, could account for the inhibition of field stimulation that was observed. Norepinephrine (10⁸ M-10⁴ M) produced a marked contractile response in the rabbit saphenous vein, an effect that was dextrally shifted by prazosin (10⁶ M) but unaffected by GR127935 (10⁶ M) or by WAY 100635 (10⁶ M; Fig. 4). Therefore, neither GR127935 nor WAY 100635, in the concentrations used, blocked  receptors in the rabbit saphenous vein.

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Effect of prazosin (106 M), GR127935 (106 M), and WAY 100635 (106 M) on the contractile response to norepinephrine in the rabbit saphenous vein. Points are mean values and vertical bars represent S.E.M. for the number of tissues indicated in parentheses.

Effect of the 5-HT_1A Receptor Agonists, 8-OH-DPAT and LY228729 on Field-Stimulated Contraction. To examine further the ability of 5-HT_1A receptors to modulate neuronal release of norepinephrine in the rabbit saphenous vein, we examined the effect of two known but structurally dissimilar 5-HT_1A receptor agonists, 8-OH-DPAT and LY228729 (Foreman et al., 1993), on the response to field stimulation in this tissue. Both 8-OH-DPAT (10⁶ M) and LY228729 (10⁶ M) increased the response to field stimulation in the rabbit saphenous vein (Fig. 5), consistent with the contention that 5-HT_1A receptor activation can enhance norepinephrine release from nerves in the rabbit saphenous vein. Field-stimulation responses in time-dependent preparations were identical (data not shown) for the first and second responses.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Effect of 8-OH-DPAT (106 M) and LY228729 (106 M) on the contractile response to field stimulation in the rabbit saphenous vein. Points are mean values and vertical bars represent S.E.M. for the number of tissues indicated in parentheses.

Effect of GR 127935 and WAY 100635 on Contraction to 8-OH-DPAT and LY228729. Because 8-OH-DPAT is also known to contract the human saphenous vein in the absence of field stimulation (Bax et al., 1992; Glusa and Miller-Schweinitzer 1993), we examined the effect of the 5-HT_1A receptor agonists 8-OH-DPAT and LY228729 to contract the rabbit saphenous vein in the absence of field stimulation. 8-OH-DPAT (Fig. 6) and LY228729 (Fig. 7) were relatively potent contractile agonists (8-OH-DPAT log ED₅₀ = 5.83 ± 0.05 (n = 7), LY228729 log ED₅₀ = 5.66 ± 0.80 (n = 27) in the rabbit saphenous vein, producing a similar maximum contraction.

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Effect of prazosin (106 M), GR127935 (106 M), and WAY 100635 (106 M) on the contractile response to 8-OH-DPAT in the rabbit saphenous vein. Points are mean values and vertical bars represent S.E.M. for the number of tissues indicated in parentheses.

View larger version (28K): [in this window] [in a new window]   Fig. 7.   Effect of prazosin (106 M), GR127935 (106 M), and WAY 100635 (106 M) on the contractile response to LY228729 in the rabbit saphenous vein. Points are mean values and vertical bars represent S.E.M. for the number of tissues indicated in parentheses.

Norepinephrine and Serotonin Concentrations in the Rabbit Saphenous Vein. Presynaptic 5-HT_1A receptors appear to be tonically activated in the rabbit saphenous vein, because WAY 100635 inhibited the field-stimulated contraction (Fig. 2) in the absence of agonists. We further asked whether the saphenous vein could provide a source of serotonin. Low, but measurable concentrations of serotonin were detected in the rabbit saphenous vein (Table 1). Concentrations of serotonin in the rabbit saphenous vein were approximately 15% of the norepinephrine concentration found in this tissue.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Serotonin exerts diverse effects on neurotransmitter release. In blood vessels, activation of presynaptic serotonergic receptors has been widely documented to inhibit norepinephrine release from sympathetic nerves (Göthert et al., 1990). In the human saphenous vein (Molderings et al., 1990; Göthert et al., 1991) and bovine arteries (Barrús et al., 1993), inhibition of norepinephrine release has been attributed to 5-HT_1D receptor activation, whereas in the rat vena cava, 5-HT_1B receptors have been proposed to mediate inhibition of norepinephrine release (Göthert et al., 1991). In pig coronary arteries (Göthert et al., 1991), canine coronary arteries (Cohen, 1985), and canine saphenous veins (McGrath and Shepherd, 1978; Müller-Schweinitzer, 1981), serotonin-induced inhibiton of norepinephrine release was reported, although the receptor was not delineated.

In contrast to these studies, an effect of serotonin to enhance norepinephrine release from presynaptic nerves has been reported in canine saphenous vein (Humprey, 1978), goat cerebral arteries (Marin et al. 1981), and sheep middle cerebral arteries (Gaw et al., 1990). The enhanced release of norepinephrine from nerves in these blood vessels was attributed to the displacement of norepinephrine from neuronal stores by serotonin, because serotonin has been demonstrated to be taken up by adrenergic nerves in certain blood vessels and subsequently released by depolarization and tyramine (Verbeuren et al., 1983). Thus, presynaptic effects attributed to serotonergic receptor activation in blood vessels have included both inhibition of norepinephrine release and increased norepinephrine neuronal release, the latter attributed to norepinephrine displacement by accumulated serotonin.

The rabbit saphenous vein, used to identify potent contractile agonists at the 5-HT_1B/1D-like receptor (Beer et al., 1994; Perez et al., 1995) is adrenergically innervated (De Mey and Vanhoutte, 1978; Levitt and Hieble, 1986), yet little is known about the ability of serotonin to modulate innervation in this tissue. Presynaptic modulation of norepinephrine release (Levitt and Hieble, 1986) has been demonstrated for ₂ receptor agonists, suggesting that presynaptic modulation by serotonergic agonists was likely to occur. However, in the present study, the 5-HT_1B/1D receptor antagonist GR127935 did not either increase or decrease contraction to field stimulation, suggesting that under the in vitro conditions of these experiments, 5-HT_1B/1D receptors were not tonically activated. This is in contrast to the profound inhibitory role of 5-HT_1D receptor activation on neurotransmitter release in other blood vessels (Molderings et al., 1990; Göthert et al., 1991; Barrús et al., 1993). These studies underscore the importance of presynaptic serotonergic receptor heterogeneity in blood vessels, a concept previously noted for postsynaptic serotonergic mechanisms (Glusa and Müller-Schweinitzer, 1993).

In contrast to the apparent lack of tonic 5-HT_1B/1D receptor presynaptic modulation of norepinephrine release, activation of presynaptic 5-HT_1A receptors increased norepinephrine release from nerves in the rabbit saphenous vein. The evidence to support this conclusion is based on the following observations: 1) The effect of two structurally distinct 5-HT_1A receptor agonists to increase contraction induced by field stimulation. Both 8-OH-DPAT and LY228729 have similar affinity and efficacy at 5-HT_1A receptors (Foreman et al., 1993) and produced a similar increase in field-stimulated contraction of the rabbit saphenous vein. 2) Contraction to both 5-HT_1A receptor agonists in the absence of field stimulation was blocked by the -adrenergic antagonist prazosin, and partially blocked by the 5-HT_1A receptor antagonist WAY 100635. These observations are consistent with the contention that these 5-HT_1A receptor agonists enhanced norepinephrine release by a presynaptic mechanism involving 5-HT_1A receptor activation in the rabbit saphenous vein. 3) Most compelling is the evidence that the 5-HT_1A receptor antagonist WAY 100635 inhibited the contractile response to field stimulation in the absence of exogenously added 5-HT_1A receptor agonist. Thus, by a totally independent measure, the response to field stimulation was modulated by tonic or endogenous 5-HT_1A receptor activation, providing compelling evidence for the presence of this unique mechanism in this tissue.

Interestingly, the direct contractile effect of both 5-HT_1A receptor agonists was antagonized by prazosin and to a more modest extent by the 5-HT_1A receptor antagonist WAY 100635. The marked inhibition by prazosin may suggest that in the high concentrations used, 5-HT_1A receptor agonists may be directly activating postsynaptic  receptors or may be inhibiting neuronal uptake mechanisms. Most surprising was the observation that GR127935 (10⁶ M) dramatically inhibited the baseline contraction induced by both 8-OH-DPAT and LY228729. GR127935 was used in a concentration that selectively inhibited the 5-HT_1B/1D-like contractile response to sumatriptan without altering either the field-stimulated response or the contractile response to norepinephrine. These data suggest that although 8-OH-DPAT and LY228729 exerted indirect effects via presynaptic 5-HT_1A receptor activation to enhance norepinephrine release, those agents also appear capable of activating 5-HT_1B/1D-like receptors in this preparation. This possibility is supported by the documented weak affinity of 8-OH-DPAT and LY228729 at 5-HT_1D receptors (Foreman et al., 1993; Pauwels and Colpaert, 1996). Thus, in concentrations of 10⁶ M and higher, 8-OH-DPAT and LY228729 can exert a direct effect on postsynaptic 5-HT_1B/1D-like receptors to produce a contractile response.

The observation that the 5-HT_1A receptor antagonist WAY 100635 markedly inhibited the contractile response produced by field stimulation in the absence of exogenously added 5-HT_1A receptor agonist suggests that 5-HT_1A receptors were activated in the presence of field stimulation. These results may be best explained by postulating tonic activation of the presynaptic 5-HT_1A receptor by field stimulation, an effect that increased endogenous norepinephrine release. This observation requires a source of endogenous serotonin able to activate these presynaptic 5-HT_1A receptors. To support this possibility, we document measurable concentrations of serotonin in the rabbit saphenous vein and further show that the serotonin concentration could be modulated by tryptophan hydroxylase or monoamine oxidase inhibition. This serotonin did not result from serotonin uptake into adrenergic nerves, because 6-OHDA pretreatment, while abolishing norepinephrine levels, did not alter serotonin concentration in the saphenous vein. These data suggest that either neuronal or possibly mast cell serotonin can be released with field stimulation to activate presynaptic 5-HT_1A receptors, which enhanced norepinephrine release.

Thus, although presynaptic 5-HT_1A receptor activation is usually inhibitory, the observation that 5-HT_1A receptor activation can increase norepinephrine release from sympathetic nerves is novel as it applies to blood vessels. Furthermore, these results are consistent with other reports suggesting a role for 5-HT_1A receptor activation to enhance hippocampal acetylcholine release (Izumi et al., 1994), norepinephrine release from sympathetic preganglionic neurons (Lewis and Coote, 1990), and norepinephrine release as measured by microdialysis in the ventral tegmental area of rats (Chen and Reita, 1995). Thus, these previous but limited proposals that 5-HT_1A receptor activation can increase neurotransmitter release may not be restricted to brain, but may also apply to certain vascular beds.