br Discussion The pattern of activity exhibited
Discussion The pattern of activity exhibited by SSR 504734 and Lu AA21279 in the MEST test was very similar. Both compounds were inactive at the lowest dose (3.0mg/kg), but exhibited robust increases (∼150%) at 10mg/kg followed by a maximal detectable response at 30mg/kg. Based on the magnitude of the responses at 10mg/kg (∼150%), it is very likely that the minimum effective dose (MED) for these compounds in the MEST test lies somewhere between 3.0 and 10mg/kg. The pattern of activity of NFPS and Org 25935 in the MEST test was overall similar to those of SSR 504734 and Lu AA21279 with the exception of lower MEDs (∼3.2mg/kg and 1.0mg/kg, respectively). Two novel GlyT1 inhibitors, SB-710622 and GSK931145, exhibited robust anticonvulsant activity in the MEST test characterized with a higher potency and steeper dose–response curves compared to those of other compounds tested in this study. For example, a modest (41%) increase in seizure threshold in response to 3.2mg/kg SB-710622 was followed by the maximal detectable response at 10mg/kg. Furthermore, a detailed analysis of the lower dose range (0.1–3.0mg/kg) of GSK931145 revealed that the MED for this compound in the MEST test could be as low as 0.1mg/kg. For Org 25935, there was a clear relationship among doses administered, blood and NS 1619 concentrations of the drug and magnitude of the anticonvulsant effect at estimated tmax. For example, 4h post-dosing Org 25935 at increasing doses from 0.1 to 1.0mg/kg, concentrations of the compound increased from 21ng/mL blood and 11ng/g brain to 256ng/mL blood and 117ng/g brain. These increases were paralleled by changes in seizure threshold from vehicle-like level to 305% increase. We can assume that anticonvulsant activity of GlyT1 inhibitors in the MEST test is due to increased levels of extracellular glycine and increased action of glycine at its binding site. There is ample evidence that administration of the GlyT1 inhibitors investigated in this study results in significant increases in the levels of extracellular glycine. For example, in the rat 6mg/kg Org 26935 increases striatal extracellular glycine levels by ∼50–80% (Ge et al., 2001). Also, increases in extracellular glycine levels in prefrontal cortex have been reported in animals treated with NFPS (Atkinson et al., 2001). Similar effects on extracellular glycine have been observed in response to SB-710622 and GSK931145 treatment in the rat (unpublished data). The question that remains to be answered is whether the anticonvulsant effect of GlyT1 inhibitors described in this study is mediated by strychnine-sensitive or strychnine-insensitive binding sites in the CNS. In the latter case there is evidence that, depending on the model of epilepsy used, strychnine-insensitive glycine binding sites can mediate both pro- and anticonvulsant activity. For example, in mice NMDA-induced convulsions can be potentiated by d-serine (Chiamulera et al., 1990, Koek and Colpaert, 1990, Singh et al., 1991) and antagonized by a strychnine-insensitive glycine site antagonist SM-31900 (Ohtani et al., 2002). On the other hand electroshock-induced tonic–clonic seizures are antagonized by d-serine (as seen in the present study) or by d-cycloserine, a partial agonist at the strychnine-insensitive site (Peterson, 1992, Peterson and Schwade, 1993, Löscher et al., 1994, Wlaz et al., 1994). One hypothesis which could be put forward is that pro- and anticonvulsant effects of d-serine might be mediated by distinct populations of strychnine-insensitive binding sites (Danysz et al., 1990). The fact that in our study the anticonvulsant effect of d-serine lacked dose-dependency and disappeared at the highest dose (1000mg/kg) could be used as evidence supporting this hypothesis. Alternatively, some non-specific effect of d-serine may have led to the lack of activity at the highest dose. It is also possible that different populations of NMDA receptors might mediate proconvulsant and anticonvulsant effects. For example, it has been reported that synaptic and extrasynaptic NMDA receptors can have opposite functional effects (Hardingham et al., 2002). Another possible mechanism for the anticonvulsant effect of GlyT1 inhibition could be via glycine priming of NMDA receptor internalisation (Nong et al., 2003). However, this effect would only be expected at high levels of GlyT1 inhibition (Martina et al., 2004), whereas in our study anticonvulsant effects were observed across the full range of compound doses.