It is important to clarify whether the decrease

It is important to clarify whether the decrease in USV after administration was secondary to the central depressant actions of the test compounds. Several reports have addressed the possibility that central depressant actions such as motor incoordination or changes in body temperature might affect USV (Olivier et al., 1998a, Olivier et al., 1998b). However, while diazepam is known to induce muscle relaxation at higher doses, it has only been shown to suppress USV at lower doses (Olivier et al., 1998a). And although diazepam reduces rectal temperature at effective doses, overall the hypothermic actions of the benzodiazepines are not thought to contribute to their anxiolytic actions, at least in the clinical setting (Olivier et al., 1998a). Furthermore, in the present study, diazepam did not decrease rectal temperature at the effective doses used, which provides support for the idea that diazepam decreases USV via its anxiolytic actions. Our results also show that escitalopram did not affect rectal temperature at the effective doses used, which is consistent with the results of previous studies showing that selective serotonin reuptake inhibitors have no effect on motor coordination, body temperature, or the righting PF 04418948 (Hodgson et al., 2008, Olivier et al., 1998b). We also showed that SSR504734 did not affect rectal temperature at the effective doses used, which supports previous work by Depoortère et al. (2005) that did not report any abnormal behavior in rats and mice caused by SSR504734, and that administration of ALX5407 at effective doses also did not cause a decrease in rectal temperature. Several studies have shown that high-dose ALX5407 induces motor incoordination. For example, Harsing et al. (2006) have reported that intraperitoneal administration of (±)-NFPS at 30mg/kg produces motor dysfunction in mice, and Perry et al. (2008) have reported that orally administered (±)-NFPS at 30mg/kg, but not at 10mg/kg, produces motor dysfunction, impaired gait in rats. Kopec et al. (2010) have shown that intraperitoneal administration of ALX5407 at 6mg/kg, but not at 3mg/kg, exhibited hyperlocomotion. Here, we observed the decrease in USV at the lower dose of 1mg/kg without affecting rectal temperature and we did not observe any abnormal behaviors in rat pups at the same dose. Therefore, we consider the ALX5407-induced decrease in USV to not be secondary to the central depressant actions of the compound. Together, the above results suggest that GlyT1 inhibitors have an anxiolytic action. Next, we examined whether GlyA or GlyB is the main contributor to the anxiolytic action of GlyT1 inhibitors. Inhibitory GlyA and excitatory GlyB mediate opposing actions on neuronal excitability. GlyT1 inhibitors efficiently activate NMDA receptors because GlyT1 is distributed closely to the NMDA receptor (Smith et al., 1992). However, a recent autoradiograph study (Herdon et al., 2010) confirmed the results of a previous study that showed that GlyT1 is enriched in the hindbrain (Jursky et al., 1994), where the distribution is consistent with that of GlyA (Zarbin et al., 1981). Therefore, it might be possible that GlyT1 inhibitors contribute to the activation of GlyA. Consistent with this hypothesis, our results indicate that the anxiolytic action of GlyT1 inhibitors is reversed by the GlyA antagonist strychnine, but not by the GlyB antagonist L-687,414, and therefore that the anxiolytic action of GlyT1 inhibitors is mediated by GlyA. Strychnine administration (0.4mg/kg) slightly increased USV; however, we can exclude the possibility that this apparent reversal of the action of SSR504734 and ALX5407 by strychnine was actually due simply to an independent increase in USV, because strychnine also showed significant reversal when administered at 0.2mg/kg, a dose at which USV remained unchanged when strychnine was administered alone. Administration of L-687,414 alone at 30mg/kg decreased USV, which agrees with the results of previous studies in which GlyB antagonists were reported to have anxiolytic actions (Trullas et al., 1989, Winslow et al., 1990). These data support the notion that activation of GlyB does not induce anxiolytic actions. Although L-687,414 has been reported to be a partial agonist of GlyB (Priestley et al., 1998), it acts as a substantial NMDA receptor antagonist in vivo because of its weak intrinsic agonist activity. Tricklebank et al. (1994) have shown that L-687,414 has dose-dependent anticonvulsant effects in a variety of animal models, and that the anticonvulsant effects of L-687,414 are completely reversed by administration of the GlyB agonist d-serine. In contrast to GlyT1 inhibitors, neither the diazepam- nor escitalopram-induced reductions in USV were reversed by strychnine. These results suggest that GlyA is not associated with GABAA receptor- or serotonin transporter-mediated anxiolytic actions.