Unravelling the spatial and temporal appearance of the channel subunits is important to understand the molecular basis of GIRK signalling. In the adult, native GIRK channels are believed to be composed of either homomeric assemblies of GIRK2 (Jelacic et al., 2000; Koyrakh et al., 2005) or heteromeric complexes of GIRK1, GIRK2 and/or GIRK3 subunits (Chen et al., 1997; Karschin et al., 1996; Kobayashi et al., 1995). There is general agreement that GIRK2 is an integral subunit in most neuronal GIRK channels and GIRK1/GIRK2 heteromultimers are widely considered to be the prototypical GIRK channel in the brain (Cruz et al., 2004; Koyrakh et al., 2005; Lüscher et al., 1997; Marker et al., 2005, 2006; Torrecilla et al., 2002). The loss of GIRK1 staining that we have detected in a number of brain regions, as well as in different layers or subfields within a specific region, in GIRK2 knockout mice supports this contention, as has also been suggested previously (Koyrakh et al., 2005; Liao et al., 1996; Torrecilla et al., 2002). Nevertheless, the expression of GIRK2 does not preclude the formation of
