| Popis: |
The natural product willardiine is an AMPA receptor agonist. We have examined the structural changes required to convert willardiine into an antagonist at AMPA and kainate receptors. Structure–activity analysis has been carried out to discover the structural features required to increase the potency and/or selectivity of the antagonists at AMPA or kainate receptors. Reduction of the fast component of the dorsal root-evoked ventral root potential (fDR-VRP) has been used to investigate AMPA receptor antagonist activity. To examine antagonist activity at kainate receptors, the ability of compounds to depress kainate-induced depolarisations of dorsal root fibres was assessed. Blocking ionisation of the uracil ring by adding a methyl group to the N3 position was not sufficient to convert willardiine into an antagonist. However, willardiine derivatives with a side-chain bearing a carboxylic acid group at the N3-position of the uracil ring could antagonise AMPA and kainate receptors. S stereochemistry was optimal for antagonism. When compounds with differing interacidic group chain lengths were compared, a group chain length of two methylene groups was preferable for AMPA receptor antagonism in the series of compounds bearing a carboxyalkyl side chain (UBP275, UBP277 and UBP279 reduced the fDR-VRP with IC50 values of 287±41, 23.8±3.9 and 136±17 μM, respectively). For kainate receptor antagonism, two or three methylene groups were almost equally acceptable (UBP277 and UBP279 reduced dorsal root kainate responses with apparent KD values of 73.1±4.5 and 60.5±4.1 μM, respectively). Adding an iodo group to the 5-position of UBP277 and UBP282 enhanced activity at kainate receptors (UBP291 and UBP301 antagonised kainate responses on the dorsal root with apparent KD values of 9.83±1.62 and 5.94±0.63 μM, respectively). The most useful antagonist identified in this study was UBP301, which was a potent and ∼30-fold selective kainate receptor antagonist. UBP282 may also be of use in isolating a non-GluR5-mediated kainate response. Keywords: Neonatal rat spinal cord, willardiine, 3-CBW (UBP282), UBP301, kainate, AMPA, antagonist Introduction Ionotropic glutamate receptors in the mammalian central nervous system have been divided into three main types–NMDA, AMPA and kainate receptors–depending on their pharmacology (for comprehensive reviews see Jane et al., 2000; Jane, 2002). AMPA receptors are made up from a combination of GluR1–4 subunits, while kainate receptors consist of a combination of GluR5–7, KA1 and KA2 subunits (Bleakman & Lodge, 1998). Although selective AMPA receptor antagonists are available, few discriminate between individual subunits and selective kainate receptor antagonists remain scarce (for reviews see Chittajallu et al., 1999; Jane et al., 2000). Recently, a number of decahydroisoquinoline analogues have been shown to be GluR5-selective antagonists. These compounds have been used to show that GluR5-selective antagonists may have utility in the treatment of neuropathic pain (Simmons et al., 1998), cerebral ischaemia (O'Neill et al., 1998) and epilepsy (Smolders et al., 2002). The natural product willardiine acts as an agonist at AMPA receptors and a range of willardiine analogues have been synthesised with selectivity for either AMPA or GluR5-containing kainate receptors depending on the nature of the 5-substituent on the uracil ring (Evans et al., 1980; Patneau et al., 1992; Wong et al., 1994; Jane et al., 1997). For example, (S)-5-iodowillardiine is a highly selective GluR5 agonist whereas (S)-5-fluorowillardiine is an AMPA receptor agonist, which binds with higher affinity to GluR1 or GluR2 compared to GluR3 or GluR4 (Jane et al., 1997; Thomas et al., 1998; Varney et al., 1998). A number of studies have demonstrated the conversion of glutamate receptor agonists into antagonists by extending the chain length between the α-carboxylic acid and the terminal acidic group (Davies et al., 1982; Krogsgaard-Larsen et al., 1991; Madsen et al., 1996; Jane et al., 2000). We recently demonstrated that the agonist willardiine could be converted into an antagonist of AMPA and kainate receptors by increasing the interacidic group chain length when it was shown that 3-CBW ((S)-3-(4-carboxybenzyl)willardiine, UBP282) is an antagonist of AMPA receptors on motor neurones and kainate receptors on dorsal root C-fibres (More et al., 2002a, b). The aim of the current study was to investigate the structural changes required to convert willardiine into an antagonist and to find ways of increasing the potency of antagonism and the selectivity towards either AMPA or kainate receptors. To achieve this, several derivatives of willardiine have been assessed for their antagonist activity at AMPA and kainate receptors. In order to investigate whether adding a substituent to the N3-position is sufficient to convert willardiine into an antagonist, we synthesized the 3-methyl analogue ((S)-3-methylwillardiine, UBP294). Other structural changes made to the willardiine structure included changing the stereochemistry at the stereogenic centre, changing the interacidic group chain length and adding substituents to the 5-position of the uracil ring. Thus, the following compounds were investigated for their AMPA and/or kainate receptor antagonist activity (see Figure 1 for structures): (S)-3-carboxymethylwillardiine (UBP275), (R)-3-carboxymethylwillardiine (UBP276), (S)-3-(2-carboxyethyl)willardiine (UBP277), (R)-3-(2-carboxyethyl)willardiine (UBP278), (S)-3-(3-carboxypropyl)willardiine (UBP279), (R)-3-(3-carboxypropyl)willardiine (UBP280), (S)-3-methylwillardiine (UBP294), (S)-1-carboxymethyl-5-methylwillardiine (UBP293), (S)-1-carboxymethylwillardiine (UBP281), (S)-3-(2-carboxyethyl)-5-nitrowillardiine (UBP290), (S)-3-(2-carboxyethyl)-5-iodowillardiine (UBP291) and (S)-3-(4-carboxybenzyl)-5-iodowillardiine (UBP301). Figure 1 Structure of willardiine and a number of new derivatives. To compare the antagonists for activity at AMPA receptors, their ability to depress the fast component of the dorsal root-evoked ventral root potential (fDR-VRP) in the neonatal rat hemisected spinal cord preparation was measured. As reported previously, the fDR-VRP has been shown to be evoked by the stimulation of AMPA receptors and can therefore be used as a convenient method to compare AMPA receptor antagonists using native receptors (More et al., 2002b). To investigate the ability of the novel compounds to antagonise kainate receptors, their ability to depress kainate-induced depolarisations of neonatal rat dorsal roots was assessed. Previous studies have shown that this preparation contains predominantly kainate receptors of the GluR5 subtype (Bettler et al., 1990; Partin et al., 1993), although possibly combined with KA1 or KA2 (Fletcher & Lodge, 1996), making it useful for the examination of selective kainate receptor antagonists using a native receptor population (Agrawal & Evans, 1986; Thomas et al., 1998; More et al., 2002a, b). As there are a limited number of selective antagonists available for AMPA and kainate receptors, it is important to discover new families of compounds that have the potential to be used as antagonists. Small changes in substituents can have drastic selectivity effects in willardiine derivatives acting as agonists, as highlighted by (S)-5-iodowillardiine and (S)-5-fluorowillardiine. It is therefore possible that careful structure–activity analysis of willardiine derivatives may lead to novel selective and potent antagonists for either AMPA or kainate receptors. Preliminary reports of this work have been published (More et al., 2001; More et al., 2002a). |
