TOPLACIR · A two-photon survey of the plasticity of the neocortical microcircuit: searching for plasticity hotspots

A two-photon survey of the plasticity of the neocortical microcircuit: searching for plasticity hotspots.

In many structures on the central nervous system, inhibitory GABAergic circuits are determinant in the control of the excitability and activity of principal cells. Many pathologies such as epilepsy, autism, schizophrenia or dystonia are associated to dysfunctions in inhibitory circuits, highlighting their central role in the modulation of neuronal network dynamics. Nevertheless, despite their central role, the mechanisms of regulation and their role in shaping neural activity is still not fully understood. The great diversity that interneurons exhibit in electrophysiological properties, anatomical characteristics and sub-cellular compartment targeting has made difficult to study them in a systematic manner. The variety of interneurons might also reflect a division of labour for the function of inhibition in neuronal networks. Therefore, it is critical to understand how interneurons integrate into neuronal circuits and determine how they could influence computation. The different phases of the present project allowed the fellow to study the functional organization and role of GABAergic networks in two different brain areas, the cerebral cortex and the striatum. 1- The mammalian neocortex is the largest part of the brain and is responsible for numerous functions including complex cognitive functions. Crucial to understanding how the cortex processes information is a complete description of the neurons composing its microcircuits, including the structural and functional dynamic characteristics of the connections between them. The cortical microcircuit is composed of a majority of excitatory pyramidal cells, the output neurons, and a large variety of inhibitory GABAergic interneurons. Most studies have focused on the excitatory pyramidal cells but over the last decades there has been a growing body of evidence supporting the important role of inhibitory cells in cortical functions. We focused our study on inhibitory connections to pyramidal cells and explored inhibitory subnetworks formed by different subtypes of interneurons (parvalbumin- or somatostatin-positive interneurons). To study neuronal connectivity at the network scale, we used a technique allowing us to quickly probe many neuronal connections with single-cell resolution, the two-photon glutamate uncaging (Nikolenko et al., 2011, Cold Spring Harbor Protocols), with a new caged compound that we developed, the RuBi-Glutamate (Fino et al., 2009, Frontiers in Neural Circuits). We investigated the basic structure of inhibitory cortical microcircuits and first focused on the inhibitory connectivity between somatostatin-positive interneurons and pyramidal cells in neocortex. We observed a very dense inhibitory connectivity at both young and mature developmental ages, regardless of whether pyramidal cells were part of the same functional circuits or not (Fino and Yuste, 2011, Neuron). We obtained similar results when considering another subtype of interneurons, the parvalbumin-positive interneurons (Packer and Yuste, 2011, Journal of Neuroscience). In addition, we explored the anatomical mechanisms explaining this dense connectivity (Packer et al., 2012, Cerebral Cortex). We conclude that local inhibitory connectivity is promiscuous and does not form specific subnetworks, highlighting the potential very important role of inhibitory interneurons in cortical circuit computation (for review see Fino et al., The Neuroscientist, 2013). In a second step, we explored the role of distinct subclasses of interneuron in cortical circuit activity. Using an optogenetic approach, we have been able to either activate or inhibit specifically the whole subpopulation of either parvalbumin-positive interneurons or somatostatin-positive interneurons while recording electrophysiologically the inputs to pyramidal cells and then determine the role of distinct subpopulations of interneurons in information processing. 2- Basal ganglia are involved in adaptive control of behavior and procedural learning. Striatum, the primary input nucleus of basal ganglia, acts as a coincidence detector of cortical afferents and is the major site of procedural memory formation. It is now well established that learning and memory rely mainly on long-term changes of synaptic efficacy, named long-term plasticity. Therefore, the occurrence of different forms of plasticity at corticostriatal synapses is determinant in the occurrence of procedural learning. We previously studied extensively the different forms of long-term plasticity occurring at corticostriatal synapses depending on the frequency of activation of the cortex (Fino et al., 2005, Journal of Neuroscience) or on the relative concomitant activation of the cortex and the striatum, plasticity named spike-timing dependent plasticity or STDP (for review see Fino and Venance, 2010, Frontiers in Synaptic Neuroscience). The striatum is a heterogeneous structure, mainly composed of the output neurons, medium-spiny neurons (MSNs), but also different types of interneurons, most of them being GABAergic interneurons. Interestingly we have observed that all the striatal subtypes receives cortical information and are able to develop different forms of synaptic plasticity after different patterns of corticostriatal activity (for review see Fino and Venance, 2011, Neuropharmacology). The most potent GABAergic circuits within the striatum are formed by the different subtypes of interneurons. Striatal interneurons had been evidenced to control neuronal excitability, and consequently the spike timing, in MSNs. Nevertheless, the weight of the interneurons in the cortical information processing within the striatum had not been explored. We first asked whether and how GABAergic microcircuits could regulate corticostriatal plasticity. Using pharmacological approach to block all the GABAergic transmission, we showed that GABAergic circuits exert a potent control on the orientation of corticostriatal STDP in MSNs (Fino, Paille et al., 2010, Journal of Physiology). Moreover, with a combination of modeling and experimental approaches, we deciphered the mechanisms behind this strong effect of inhibitory circuits (Paille, Fino et al., 2013, Journal of Neuroscience). The previous results highlighted the central role of GABAergic interneurons in the integration of cortical inputs in the striatum. To go further, we addressed the question of the specific roles of the different subpopulations of GABAergic interneurons in the control of corticostriatal information processing. Striatum comprises three main subpopulations of GABAergic interneurons, parvalbumin-positive, somatostatin-/NO-synthase-positive and calretinin-positive interneurons that are likely to perform different computational function in the striatum. This question is central but very little information was available. To study the role of the different subpopulations of GABAergic interneurons, we used an optogenetic approach to optically either activate or silence specific GABAergic interneuronal populations. We coupled activation of corticostriatal inputs at various frequencies (in MSNs) with either activation or inhibition of parvalbumin- or somatostatin-positive GABAergic networks to assess the weight of each interneuron in the control of cortical input integration. The first results showed that one subpopulation (parvalbumin-positive interneurons), but not the other, strongly control the integration of cortical inputs in striatal ouptput neurons, MSNs. Altogether, in this project, we were able to characterize the functional organization and the role of inhibitory circuits in neocortex and striatum. These findings establish a central role for GABAergic microcircuits in shaping neuronal activity and highlight the importance of considering the heterogeneity of neuronal networks to understand how they can compute information. This will allow us to understand the dysfunction of the inhibitory networks in pathological states and try to find therapeutic approaches.

Data: CORDIS, © European Union