Furthermore, a detailed analysis of the striatal innervation of reconstructed GP-TA neurons revealed that each axon could split into several axonal collaterals, and form thousands of axonal boutons in the striatum, constituting the largest extrinsic GABAergic input to the striatum. Additional observations revealed that GP-TA neurons target all main populations of neurons in the striatum, i.e., projection neurons and the major classes of

interneurons. Lastly, the authors also show that axon collaterals from GP-TI and GP-TA neurons can form local connections with both GP-TI and GP-TA neurons, i.e., these two populations Dabrafenib manufacturer can communicate directly within and between each other. These observations indicate yet another potential Selleckchem SCH-900776 degree of regulation in GPe networks. This new population of striatal projecting pallidal neurons (arkypallidal) adds to an increasingly complex picture of basal ganglia connectivity that challenges the basic feedforward view of corticobasal ganglia loops. In particular, it presents a new way of looking at GPe, not simply as a relay area that forwards information from the striatum to downstream structures like the STN, but as a region with different circuits that can differentially target specific points of a larger network. Because of the large projection of the GP-TA neurons to the striatum, this population can potentially have a major impact on the dynamics

of this structure. For example, it may shape models about the balance between direct and indirect pathways, and how these pathways dynamically influence Cell press each other. This will depend largely on a more extensive characterization of the projections of GP-TA neurons. Are they synapsing preferentially into direct or indirect neurons? If they target neurons from the indirect

pathway, is this a complete feedback loop where they are projecting back to same neurons from which they receive input? These and other questions can entirely change the predictions of what these neurons do in basal ganglia circuits, with different potential combinations resulting in the emergence of rather different network dynamics. Also, since ventral, medial and lateral networks in corticobasal ganglia loops have been implicated in different aspects of behavior (Balleine et al., 2007), it would be interesting to investigate if the projections of arkypallidal neurons are topographically structured or not, as this could constitute yet another level of organization. Another pending question relates to input to both populations of GPe neurons, that is, which cells project to GP-TI neurons and which project to GP-TA neurons. One can consider situations where both cell types receive projections from the same neurons, or a sort of functional organization of the inputs, which could be at least partially responsible for the diverse firing pattern of the two populations.