Network Heterogeneities, Intrinsic Plasticity and Decorrelation in the Dentate Gyrus

Abstract

A central premise for research in neuroscience is to mechanistically decipher the physiological roles of the brain in effectuating learning and memory. The tripartite framework that is at the core of learning and memory entails encoding of inputs received through the different sensory modalities, selection and storage of salient information through physical and chemical changes, and retrieval of such stored information. Behavior is then postulated to emerge as a consequence of synergistic interactions and coordinated activity between these interdependent processes, executed by the involvement of multiple brain regions in conjunction with sensory and motor systems. In this thesis, we chose the dentate gyrus (DG), a hippocampal sub region that is endowed with several unique features (including divergent and sparse afferent connectivity, and their ability to sustain adult neurogenesis), to explore cellular and network mechanisms involved in encoding and storage. Physiologically, from the encoding perspective, the DG has been implicated in response decorrelation and pattern separation, and has been demonstrated to encompass engram cells that aid in memory storage and retrieval processes. The principal question addressed in this thesis is on how sparse and divergent connectivity in the DG, in conjunction with disparate network heterogeneities and various plasticity mechanisms, form the basis for its encoding function of response decorrelation and its storage function of engram formation? First, from the encoding perspective, we assess the mechanistic basis for response decorrelation and pattern separation in the DG from the perspective of the different forms of heterogeneities that express in the DG. Although there are several lines of evidence for a role of adult neurogenesis and sparse connectivity in mediating decorrelation, the role of various biological heterogeneities in executing response decorrelation or pattern separation have not been systematically assessed. To address this lacuna, we first characterized one form of heterogeneity, in intrinsic neuronal properties, employing whole-cell patch-clamp recordings from granule cells in different sectors of the DG. We, then employ a multi-scale conductance-based network model to systematically assess the implications for four different forms of heterogeneities, expressed independently or synergistically, in the emergence of response decorrelation and pattern separation. Although different forms of heterogeneities were independently and synergistically capable of mediating channel decorrelation, we found the dominance of afferent heterogeneities (mediated by the unique sparse and divergent afferent connectivity) over all local heterogeneties in effectuating channel decorrelation. With reference to memory storage, we explore putative plasticity mechanisms that could underlie the emergence of engram cells. Although there is evidence for the manifestation of engram cells in the DG, and despite widespread acknowledgement for a role of plasticity in intrinsic neuronal properties in the emergence of engram cells, intrinsic plasticity mechanisms in the DG have not been systematically assessed from a mechanistic perspective. Here, we demonstrate that intrinsic properties of DG granule cells change in response to behaviorally relevant theta-patterned activity patterns. We show that this calcium-dependent form of intrinsic plasticity is consequent to conjunctive changes in two different ion channel subtypes: HCN ion mediating sub-threshold changes and persistent sodium ion channels regulating supra-threshold changes. Finally, in synthesizing our two main analyses motifs, on the role of heterogeneities in decorrelation and on modulation of intrinsic properties induced by ion plasticity, we analyzed the cascading implications for dramatic changes to ion channel properties on neuronal intrinsic properties and on response decorrelation. In doing this, we report a critical role of neurogenesis-induced heterogeneities in mediating resilience of network decorrelation under extreme conditions involving ion channel knockouts.