Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Several antimigraine agents have recently been identified with an improved oral pharmacokinetic profile relative to the prototypic 5-hydroxytryptamine (or serotonin) 1B/1D (5-HT_1B/1D) receptor agonist sumatriptan (Johnson et al., 1998). However, most of these recent additions to the clinical therapy of migraine have been developed based on vasoconstrictor potency and the ability to produce cerebral vascular contractile effects (Goldstein, 1996). The efficacy of these agents in the treatment of migraine has classically been attributed to the ability of these agents to activate 5-HT_1B/1D receptors (Beer et al., 1994; Perez et al., 1995), although, recently, Phebus et al. (1997) have advanced the theory that 5-HT_1F receptor activation affords migraine pain relief. To this end, LY344864 has been identified as the initial member of a new class of compounds known as selective 5-HT_1F receptor agonists (SSOFRAs) that are highly effective as inhibitors of trigeminal nerve-mediated dural extravasation (Phebus et al., 1997).

This study was designed to compare the contractile effects of LY344864 in the rabbit saphenous vein, a tissue with a contractile response to serotonergic agonists that resembles the response observed in human cerebral and coronary arteries (Cohen et al., 1997), with sumatriptan and other more recently developed triptan derivatives. A secondary objective was to shed light on the identity of the vascular 5-HT receptor that mediates the contractile response to these agents. Although contraction to serotonergic agonists in the rabbit saphenous vein resembles the contractile response occurring in human cerebral and coronary arteries (Cohen et al., 1997), some (Kaumann et al., 1994; Verheggen et al., 1996) have postulated that the receptor mediating this response is the 5-HT_1B receptor. Others (Saxena et al., 1997) have suggested that 5-HT_1D or 5-HT_1F receptors may also be involved in this response. In this regard, 5-HT_1F, 5-HT_1B, and, to a lesser extent, 5-HT_1D receptor mRNA have been identified in human cerebral arteries (Bouchelet et al., 1996a). The 5-HT_1B receptor mRNA has also been found in peripheral blood vessels (Ullmer et al., 1995; Wurch et al., 1997) and in human coronary arteries (Bouchelet et al., 1996b). As observed with human vasculature, the 5-HT_1B, 5-HT_1D, and 5-HT_1F receptor mRNAs have been detected in the rabbit saphenous vein (Bard et al., 1996a,b; Wurch et al., 1996). Thus, examination of the SSOFRA LY344864 in the rabbit saphenous vein would assist in understanding the receptor(s) mediating this vasoconstrictor response and would provide data on the role of 5-HT_1F receptors in vascular contractility.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of Vascular Tissue. Male New Zealand White rabbits (6.6-13.3 kg; Hazelton, Kalamazoo, MI) were sacrificed by a lethal dose of sodium pentobarbital (200.0 mg) injected into the ear vein. The saphenous vein was dissected free of connective tissue, cannulated in situ with polyethylene tubing (PE 50), placed in Petri dishes containing Krebs-bicarbonate buffer (see below), and ring preparations were obtained.

Experimental Protocol. Cumulative agonist concentration-response curves were generated, and no tissue was used to generate more than one agonist concentration-response curve. In some experiments, tissues were precontracted [7.62 + 2.76% (n = 12) of a maximal KCl response] with prostaglandin F₂ (PGF₂) (3 × 10⁷ M) before initiating a response to sumatriptan or LY344864. All results are expressed as means ± S.E., where n represents the number of tissues examined. The data are expressed as a percentage of the response to 67 mM KCl administered initially in each tissue. The log EC₅₀ values (pD₂) were determined by least-squares linear regression analysis of the linear portion of the concentration-response curves, taking maximal response as 100%.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

As previously reported (Cohen et al., 1997), 5-HT was a potent contractile agonist in the rabbit saphenous vein. Like 5-HT, the four 5-HT_1B/1D receptor agonists examined also contracted the rabbit saphenous vein. Of these agents, sumatriptan was roughly equipotent to rizatriptan, whereas naratriptan produced the lowest maximal response, and zolmitriptan was approximately 3- to 10-fold more potent than sumatriptan in contracting the rabbit saphenous vein (Fig. 1). In contrast, the 5-HT_1F-selective agonist LY344864 did not contract the rabbit saphenous vein in concentrations up to 10⁴ M. 

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Contractile concentration-response curves for serotonergic agonists in the rabbit saphenous vein. Points are mean values, and vertical bars represent the S.E.M. for the number of tissues indicated in parentheses.

Table 1 compares the radioligand binding affinities of these serotonergic agonists at human 5-HT_1B, 5-HT_1D, and 5-HT_1F receptors. Binding affinities of these and other agonists at the human 5-HT_1B and 5-HT_1D receptors significantly correlated with the affinities at the rabbit receptors (Bard et al., 1996a). Interestingly, zolmitriptan and naratriptan both possessed similar and relatively high affinities at 5-HT_1B and 5-HT_1D receptors, but naratriptan had about 8-fold higher affinity at 5-HT_1F receptors. The contractile efficacy in the rabbit saphenous vein was markedly dissimilar for these agonists, with zolmitriptan possessing the highest potency and efficacy as a contractile agonist in the rabbit saphenous vein and naratriptan presenting the lowest agonist efficacy of the triptan agonists examined.

Vascular contractions to sumatriptan have been reported to be markedly potentiated and augmented in the presence of low concentrations of other vasoconstrictors such as angiotensin (Smith et al., 1996), thromboxane (VanDenBrink et al., 1996), and other agonists (Yildiz and Tuncer, 1995). Potentiated vascular contractions to sumatriptan by angiotensin have also been reported in rabbits in vivo (Choppin and O'Connor, 1996). Based on these observations and the similarity between contractile responses in the rabbit saphenous vein and contractions in human coronary and cerebral arteries (Cohen et al., 1997), we examined whether precontraction would unmask a contractile response to LY344864. In fact, precontraction of the rabbit saphenous vein with low concentrations of PGF₂ resulted in an enhanced contractile response to sumatriptan relative to sumatriptan-induced contraction in rabbit saphenous vein lacking the initial tone induced by PGF₂ (Fig. 2). Although the contractile response to sumatriptan was markedly augmented, tissues precontracted with PGF₂ remained nonresponsive to LY344864 in concentrations up to 10⁵ M. Thus, induction of tone did not enhance or unmask a contractile response to LY344864 until concentrations exceeded 10⁵ M, well in excess of 5-HT_1F receptor-activating concentrations.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Contractile concentration-response curves for sumatriptan (circles) and LY344864 (squares) in the presence of PGF2 (3 × 107 M) (open symbols) and in the absence of precontractile tone (closed symbols). Points are mean values, and vertical bars represent the S.E.M. for the number of tissues indicated in parentheses. All tissues contracted to KCl (67 mM) with a force ranging between 4.1 and 5.1g, with no significant differences in contraction to KCl among the four groups of tissues.

Relationship between Serotonergic Agonist Contractility in Rabbit Saphenous Vein and Affinity at Human 5-HT_1B, 5-HT_1D, and 5-HT_1F Receptors. A significant correlation existed between the contractile potency in the rabbit saphenous vein and the affinities at 5-HT_1B and 5-HT_1D receptors for these serotonergic ligands (Fig. 3). However, for both the 5-HT_1B and 5-HT_1D receptor correlations, potencies as contractile agonists in the rabbit saphenous vein (measured by EC₅₀ values) were not quantitatively similar to the affinities at either 5-HT_1B or 5-HT_1D receptors.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Correlation of contractile potency in the rabbit saphenous vein (pD2) for serotonergic agonists with their binding affinities at human 5-HT1B (left), 5-HT1D (middle), and 5-HT1F (right) receptors. Correlation coefficients are indicated.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Contractile responses in the rabbit saphenous vein to serotonergic agonists have been proposed to mimic responses observed in human cerebral and coronary arteries (Cohen et al., 1997). The receptor mechanisms mediating the contractile responses to these serotonergic agonists have been of intense interest. The 5-HT_1B, 5-HT_1D, and 5-HT_1F receptor mRNAs have been detected in rabbit saphenous vein (Bard et al., 1996a,b; Wurch et al., 1996) and in human cerebral vessels (Bouchelet et al., 1996a). Contraction to serotonergic agonists in the saphenous vein has been proposed to be mediated by activation of 5-HT_1B/1D receptors based primarily on the use of nonselective pharmacological tools (Branchek et al.,1995; Razzaque et al., 1995). Furthermore, the presence of 5-HT_1B receptor mRNA in vascular smooth muscle of pial arteries and the absence or minimal presence of 5-HT_1D receptor mRNA has led to the suggestion that the activation of 5-HT_1B receptors may be responsible for the vasoconstriction of some of these antimigraine serotonergic receptor agonists (Hamel et al., 1993; Bouchelet et al., 1996a). In fact, the 5-HT_1B receptor gene has been identified and cloned from the rabbit saphenous vein (Wurch et al., 1996). In addition, although the 5-HT_1F receptor mRNA has also been identified in human vasculature, to date, no functional effects of 5-HT_1F receptor activation have been documented in blood vessels.

Sumatriptan and related compounds possess high affinity at 5-HT_1B and 5-HT_1D receptors, and the development of these agents as antimigraine therapies was based on the hypothesis that cerebral vasoconstriction would redistribute cerebral blood flow to alleviate migraine pain. More recently, however, evidence has been accumulating to suggest that alterations in primary efferent fibers, inhibition of neuropeptide release, and the block of neurogenic inflammation may better reflect the efficacy of antimigraine agents (Moskowitz and Waeber, 1996). In this regard, inhibition of dural extravasation produced by activation of the trigeminal nerves has been used to identify new classes of antimigraine agents.

Most recently, attention has been directed toward activation of 5-HT_1F rather than 5-HT_1B or 5-HT_1D receptors in the alleviation of migraine pain. In fact, both the 5-HT_1D and the 5-HT_1F receptor mRNAs have been identified in trigeminal ganglia (Bouchelet et al., 1996a). In addition, a selective 5-HT_1F agonist LY344864 with low affinity at 5-HT_1B or 5-HT_1D receptors was effective in inhibiting neurogenic dural extravasation in animal models of migraine (Phebus et al., 1997), raising the possibility that a new class of antimigraine agents can be developed with minimal affinity at 5-HT_1B or 5-HT_1D receptors. If the serotonergic receptors responsible for vasoconstriction are indeed either identical or similar to the 5-HT_1B or 5-HT_1D receptors, then agents with low affinity for these receptors should have minimal propensity to produce vasoconstrictor effects. In this article, we document the inability of LY344864 to contract the rabbit saphenous vein in concentrations that markedly activate 5-HT_1F receptors (Phebus et al., 1997). LY344864 possesses high affinity at human (log K_i = 8.2) and rabbit (log K_i = 7.8) 5-HT_1F receptors. Moreover, the saphenous vein markedly contracted to the 5-HT_1B/1D receptor agonists such as sumatriptan, zolmitriptan, rizatriptan, and naratriptan. These data support the inability of a selective 5-HT_1F receptor agonist to produce vasoconstriction, in contrast to other serotonergic compounds either under development or marketed for migraine therapy.

Because responses to sumatriptan can be augmented in the presence of modest vascular tone induced by other vasoconstrictor agonists (Yildiz and Tuncer, 1995; Bouchelet et al., 1996a; Smith et al., 1996; VanDenBrink et al., 1996) and because mRNA for 5-HT_1F receptors is present in many vascular beds including the rabbit saphenous vein (Bard et al., 1996b), we wanted to ensure that the augmented vascular contractile response observed with sumatriptan was unrelated to activation of 5-HT_1F receptors. For this reason, we examined the effect of LY344864 to cause vasoconstriction relative to sumatriptan in the presence of modest tone induced by PGF₂. As reported in other vascular beds, contraction to sumatriptan was markedly augmented with regard to both potency and maximal response when tissues were exposed to a previous modest tone. However, no augmentation to LY344864 occurred until concentrations of LY344864 exceeded 10⁵ M, concentrations well in excess of those required to activate 5-HT_1F receptors. These data suggest that 5-HT_1F receptors are unlikely to participate in vascular contractile responses either from baseline tone or when tone is augmented by the previous in vitro addition of a vasoconstrictor.

The inability to detect contractile effects of SSOFRAs may suggest that the presence of 5-HT_1F receptor mRNA detected in peripheral blood vessels (Bard et al., 1996b) and in some cerebral arteries (Bouchelet et al., 1996a) but not in temporal arteries (Verheggen et al., 1998) has not translated into receptor protein density sufficient to induce a contractile response. Alternatively, the receptor may reside on noncontractile elements in the tissue and may be responsible for effects other than contraction. To date, neither selective 5-HT_1F receptor antibodies nor high specific activity 5-HT_1F receptor radioligands are available to detect the presence and/or localization of 5-HT_1F receptor protein in vascular tissues. Nevertheless, these data with a SSOFRA provide evidence that activation of 5-HT_1F receptors does not participate in vascular contractile responses.

In summary, LY344864 is a high-affinity SSOFRA, effective after oral administration in reducing dural extravasation produced by trigeminal nerve stimulation (Phebus et al., 1997). Yet, LY344864 did not produce vasoconstriction in concentrations known to activate 5-HT_1F receptors, either from baseline or in the presence of augmented tone. These data suggest that 5-HT_1F receptor activation is not responsible for vascular contraction or for the enhanced contraction observed to sumatriptan after the induction of modest tone with other vascular contractile agents.