, 1998; Dowd & Dunnett, 2005). The unilateral 6-OHDA lesion model has, since its introduction 40 years ago (Ungerstedt, 1968), remained the most

widely used PD model in rats. Its application in mice has proved more problematic, at least in part due to the smaller size of the mouse brain, which makes it more difficult to achieve reproducible stereotaxic placements of the toxin injections. Nevertheless, several investigators have explored the possibility of inducing stable behavioural deficits by injection of 6-OHDA into either the striatum (Von Voigtlander & Moore, 1973; Akerud et al., 2001; Lundblad et al., 2004; Alvarez-Fischer et al., 2008) or the MFB (Lundblad et al., 2004; Iancu et al., 2005). When successful, Roxadustat the MFB lesion is clearly very efficient but has a major disadvantage in that it is associated with a very high

mortality rate: Lundblad et al. (2004) reported a 14% success rate with a mortality rate of 82%, and our own experience (S. Grealish and A. Björklund, unpublished data), using the same lesion parameters as in the Lundblad et al. (2004) study, is in line with the results reported here. The experience obtained in studies using intrastriatal 6-OHDA delivery, on the other hand, is that C646 the ‘success rate’, i.e. the percentage of lesioned animals showing severe nigrostriatal neurodegeneration and behavioural deficits, is generally insufficient in this approach (Lundblad et al., 2004; S. Grealish and A. Björklund, unpublished Fenbendazole data). Intranigral injection of 6-OHDA has emerged as an interesting third alternative. This version, which was introduced by Parish et al. (2001), is attractive in that it makes it possible to induce more extensive DA neurodegeneration in the absence of the high mortality rate seen in MFB-lesioned mice.

Based on our own preliminary experiments, and results reported in studies from other laboratories (Moses et al., 2008; Parish et al., 2008), we feel that the intranigral 6-OHDA lesion is the one that holds greatest promise for long-term studies in mice. Regardless of the site of injection, however, 6-OHDA lesions in mice are highly variable, and in any single round of surgery only a sub-portion of the injected animals can be expected to be well lesioned. This has, so far, posed a serious limitation to the usefulness of this mouse PD model. Moreover, the behavioural tests commonly used in the 6-OHDA-lesioned rats have as yet not been properly validated for use in mice. The standard amphetamine- and apomorphine-induced rotation tests have been directly applied to 6-OHDA-lesioned mice although it is unclear whether they are equally informative in mice. Indeed, there is no consensus on what protocols to use for evaluation of behavioural impairments following the 6-OHDA lesion. The doses used in amphetamine-induced rotation tests in mice range from 2 mg/kg (Perez et al., 2005) to 10 mg/kg (Offen et al.