Tetrode recordings in the cerebellar cortex HongYing Gao a,b,c,d,e , Camille de Solages a,b,c,d , Clément Lena a,b,c,d,⇑ a Institut de Biologie de l’Ecole Normale Supérieure, IBENS, Paris F-75005, France b CNRS, UMR 8197, Paris F-75005, France c Inserm, U1024, Paris F-75005, France d Ecole Normale Supérieure, Paris F-75005, France e Institutes for Advanced Interdisciplinary Research, East China Normal University, Shanghai 200062, China article info Article history: Available online 28 October 2011 Keywords: Cerebellum Tetrode In vivo recording Purkinje cells Synchrony abstract Multi-unit recordings with tetrodes have been used in brain studies for many years, but surprisingly, scarcely in the cerebellum. The cerebellum is subdivided in multiple small functional zones. Understand- ing the proper features of the cerebellar computations requires a characterization of neuronal activity within each area. By allowing simultaneous recordings of neighboring cells, tetrodes provide a helpful technique to study the dynamics of the cerebellar local networks. Here, we discuss experimental config- urations to optimize such recordings and demonstrate their use in the different layers of the cerebellar cortex. We show that tetrodes can also be used to perform simultaneous recordings from neighboring units in freely moving rats using a custom-made drive, thus permitting studies of cerebellar network dynamics in a large variety of behavioral conditions. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introduction The cerebellar cortex consists of three layers and is divided into many functional zones. Most incoming information reaches the cortex via mossy fibers terminating in the granule layer, and is then relayed to the output layer formed by the Purkinje cells. Gran- ule cell axons, referred to as parallel fibers, ascend to the molecular layer and synapse onto Purkinje cell dendrites. This relatively sim- ple organization is repeated throughout the cerebellar cortex, con- cealing the topography of the functional organization of the cerebellum. Indeed, the cerebellar cortex is divided into multiple functional zones (Apps and Hawkes, 2009). These zones may be evidenced by molecular markers, the zebrins (Brochu et al., 1990), whose expression is organized in roughly parasagittal bands perpendicular to the parallel fibers and results from developmental processes (Larouche and Hawkes, 2006). Since parallel fibers prop- agate the information across the boundaries of the zones (e.g. Gao et al., 2006), the zebrin zonation might not reflect functional divi- sions. However, the bands receive distinct inputs from mossy fibers and specific input from climbing fibers to the Purkinje cells that, in turn, project to specific areas of the deep cerebellar nuclei (Buisseret-Delmas and Angaut, 1993; Sugihara and Shinoda, 2004; Voogd and Bigaré, 1980; Voogd and Glickstein, 1998; Voogd and Ruigrok, 2004). Studies of sensory receptive fields suggest an even finer functional organization of the inputs and outputs of the cerebellar cortex (Bower and Woolston, 1983; Cohen and Yar- om, 1998; review in Garwicz et al., 1998). The functional impor- tance of the parasagittal zonation of the cerebellar cortex is also reflected by the temporal organization of inputs from the inferior olive. These inputs take the form of unitary climbing fibers contact- ing the Purkinje cells and their activation produces an extremely powerful excitation that drives complex spikes in these cells. Mul- tiple single-electrode (Lang et al., 1999; Lou and Bloedel, 1992; Sasaki et al., 1989) and optical (Ozden et al., 2008; Schultz et al., 2009) recordings have shown a strong synchrony of activation of climbing fibers in narrow parasagittal bands which correspond to zebrin bands (Sugihara et al., 2007). This circumscribed functional topography of the cerebellar extrinsic connectivity is comple- mented by the stronger efficiency of local granule cells to excite neighboring Purkinje cells (Isope and Barbour, 2002) and by the structure of the intrinsic inhibition, characterized by a relatively limited spatial extension of the territory innervated by most inhibitory interneurons (e.g. Dizon and Khodakhah, 2011) and a compartmentalization of local inhibition by the zebrin zones (Gao et al., 2006; Ozden et al., 2009; Sillitoe et al., 2008). Specific cerebellar computations are thus likely to be operated in small functional zones within the network, and are thus best analyzed by studying the dynamics of neuronal activity in local cerebellar network. In vivo recordings of neighboring cerebellar cells with pairs of electrodes are difficult experiments in 0928-4257/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jphysparis.2011.10.005 ⇑ Corresponding author at: Institut de Biologie de l’Ecole Normale Supérieure, IBENS, 46 rue d’Ulm, Paris F-75005, France. E-mail address: lena@biologie.ens.fr (C. Lena). Journal of Physiology - Paris 106 (2012) 128–136 Contents lists available at SciVerse ScienceDirect Journal of Physiology - Paris journal homepage: www.elsevier.com/locate/jphysparis