Contractile 5-HT1B receptors in human cerebral arteries: pharmacological characterization and localization with immunocytochemistry

The cerebrovascular receptor(s) that mediates 5-hydroxytryptamine (5-HT)-induced vasoconstriction in human cerebral arteries (HCA)has proven difficult to characterize, yet these are essential in migraine. We have examined 5-HT receptor subtype distribution in cerebral blood vessels by immunocytochemistry with antibodies selective for human 5-HT1B and human 5-HT1D receptors and also studied the contractile effects of a range of 5-HT receptor agonists and antagonists in HCA. Immunocytochemistry of cerebral arteries showed dense 5-HT1B receptor immunoreactivity (but no 5-HT1D receptor immunoreactivity) within the smooth muscle wall of the HCA. The endothelial cell layer was well preserved and weak 5-HT1B receptor immunoreactivity was present. Pharmacological experiments on HCA with intact endothelium showed that 5-carboxamidotryptamine was significantly more potent than α-methyl-5-HT, 2-methyl-5-HT and 5-HT in causing vasoconstriction. The 5-HT1B/1D receptor agonists naratriptan, sumatriptan, zolmitriptan and 181C91 (N-desmethyl zolmitriptan), all induced equally strong contractions and with similar potency as 5-HT. The maximum contractile response was significantly less for avitriptan and dihydroergotamine. There was a significant correlation between vasoconstrictor potency and 5-HT1B- and 5-HT1D-receptor affinity, but not with 5-HT1A-, 5-ht1F or 5-HT2- receptor affinity. The 5-HT1B/1D-receptor antagonist GR 55562 (10−7–10−6 M) inhibited the contractile responses to sumatriptan and 5-CT in a competitive manner with a pKB value for GR 55562 of 7.4. Furthermore, ketanserin (10−7 M), prazosin (10−7 M), and sulpiride (10−7 M) were devoid of significant antagonistic activity of 5-HT-induced contraction in the HCA. The results are compatible with the hypothesis that the 5-HT1B receptors play a major role in 5-HT-induced vasoconstriction in HCA. Keywords: 5-HT receptors, immunocytochemistry, in vitro pharmacology, endothelium, smooth muscle Introduction Over several decades circumstantial evidence has been collected to implicate the involvement of serotonin in migraine headache. Serotonin is a powerful vasomotor agent of the intracranial vasculature (Humphrey & Feniuk, 1991; Friberg et al., 1991). The characterization of the 5-HT receptors involved in mediating the vasomotor effects of 5-HT in these vessels has proven difficult since up to 17 subtypes of 5-HT receptors exist at present, which can be divided into seven major classes, and selective agonists and antagonists are available for some, but not all, of these receptors (Hoyer et al., 1994). The recently developed antimigraine drugs which act as acute abortive antimigraine therapies are agonists at 5-HT1B/1D-receptors (Weinshank et al., 1992; Ferrari & Saxena, 1995). These drugs cause vasoconstriction in human blood vessels both in vitro and in vivo (Hamel et al., 1993; Kaumann et al., 1994; van Es et al., 1995) and haemodynamic studies have shown a selective action in causing vasoconstriction in the cranial vascular bed compared to peripheral arteries (Perren et al., 1989). At the time of the discovery of sumatriptan, the prototypic drug in this class, only one subtype namely the 5-HT1B (or 5-HT1Dβ) was described in man and drug development was targeted against this receptor. Subsequently, molecular cloning revealed the existence of two subtypes, namely 5-HT1B- and 5-HT1D-receptors (Hamblin & Metcalf, 1991; Hartig et al., 1992; Jin et al., 1992; Peroutka & McCarthy, 1989), and since in man these receptors are closely related, sumatriptan-like drugs have similar affinity at both subtypes (Weinshank et al., 1992). Thus, in human blood vessels vasoconstrictor potency can correlate with affinity at both 5-HT1B and 5-HT1D receptors (Hamel & Bouchard, 1991; Hamel et al., 1993; Jansen et al., 1992; 1993; Kaumann et al., 1994). Interestingly, studies with RT–PCR or Northern blot hybridization delineating the distribution of mRNA coding for 5-HT receptors in human cerebral vasculature, have shown expression of 5-HT1B receptor mRNA but not 5-HT1D receptor mRNA (Hamel et al., 1993). More recently immunocytochemical studies using selective antibodies have shown expression of 5-HT1B and not 5-HT1D receptor protein in human middle meningeal (dural) artery (Longmore et al., 1997). This contrasts with human trigeminal ganglia containing the cell bodies of sensory neurones innervating cranial vessels which express mRNA or receptor protein coding for both receptor subtypes, indicative of a prejunctional localization (Bouchelet et al., 1996; Hamel et al., 1993; Rebeck et al., 1994; Longmore et al., 1997). However, the expression of 5-HT1B receptor protein or its mRNA in blood vessels does not allow the extrapolation that this receptor mediates the vasoconstrictor action of 5-HT1B/1D receptor agonists. For example, mRNA coding for 5-ht1F receptors is also present in blood vessels yet selective 5-ht1F receptor agonists are devoid of vasoconstrictor properties (Johnson et al., 1997). The aim of the present studies was to further characterize the contractile 5-HT receptors in human cerebral arteries since previous studies only have examined the dura mater middle meningeal artery (Longmore et al., 1998; Razzaque et al., 1999) or the extracranial temporal artery (Verheggen et al., 1996; 1998) which according to recent positron emission tomography studies may have a less likely role in the primary part of migraine (Weiller et al., 1995) or cluster headache attacks (May et al., 1998). We have used two strategies; (i) receptor mapping studies using immunocytochemistry with selective antibodies towards 5-HT1B and 5-HT1D receptors; (ii) a pharmacological approach to correlate the vasoconstrictor potency of a series of 5-HT receptor agonists with their affinity at 5-HT receptors determined in previously reported binding studies and also to examine the effects of selective antagonists on vasoconstrictor responses in human cerebral arteries (HCA).