This model has been used to provide a mechanistic explanation of the symptoms associated with several basal ganglia disorders, most notably Parkinson disease (PD), (Albin et al., 1989). In PD, loss of dopamine Alectinib input mainly from substantia nigra pars compacta, would have opposing effects on direct and indirect pathway neurons, which express mostly D1 versus D2-type dopamine receptors, respectively (Gerfen et al., 1990). This would result in overactivation of the indirect pathway (and the consequent inhibition of GPe) and less activation of the direct pathway and to lack of movement (Kravitz et al., 2010). Other studies show that PD is accompanied by the emergence of abnormal oscillations in basal ganglia,

most notably prominent beta oscillations in STN and GPe (Mallet et al., 2008 and Nini et al., 1995), which are thought to constitute a pacemaker circuit (Plenz and Kital, 1999). The GPe, central to basal ganglia function, has been traditionally portrayed as a structure organized in different domains of homogeneous cell populations of projection neurons, all projecting to the STN with some collaterals reaching other structures. In this issue of Neuron, Mallet and colleagues ( Mallet et al., 2012) demonstrate

that the organization of the GPe is more complex than previously thought, and that it is composed of at least two populations of GABAergic projection neurons. selleck kinase inhibitor The authors had previously shown that in a PD rat model, two different types of GPe neurons could be identified based on their entrainment to different phases of

cortical slow wave oscillations ( Mallet et al., 2008): some fired preferentially during the surface-negative component of the cortical oscillation (inactive, hence named GP-TI); others during the surface-positive phase of the cortical oscillation (active phase, GP-TA). In this study, Mallet et al. (2012) used juxtacellular labeling of in vivo recorded cells to establish that these two types of neurons, identified based on their firing dynamics, constitute indeed different cell types within GPe, also with quite distinct molecular profiles, neuronal structures, and projection patterns. The authors observed that all GP-TA neurons expressed the neuropeptide precursor preproenkephalin (PPE), while none of the GP-TI neurons did. Other markers, like parvalbumin, were more expressed in GP-TI neurons, but were also found in GP-TA neurons. Therefore, PPE could be used as a specific marker for GP-TA neurons. Using this marker, the authors showed that GP-TA and GP-TI neuronal populations are spatially intermingled in GPe, and that they are both GABAergic neurons. Next, they characterized the structure and projection specificity of individual neurons from both populations. They observed that while GP-TI neurons have the projection profile expected for GPe neurons—descending projections to downstream BG nuclei such as STN, which sometimes sent collaterals to striatum—GP-TA neurons had an unanticipated projection pattern.