Metaplasticity contributes to memory formation in the hippocampus

Abstract

Prior learning can modify the plasticity mechanisms that are used to encode new information. For example, NMDA receptor (NMDAR) activation is typically required for new spatial and contextual learning in the hippocampus. However, once animals have acquired this information, they can learn new tasks even if NMDARs are blocked. This finding suggests that behavioral training alters cellular plasticity mechanisms such that NMDARs are not required for subsequent learning. The mechanisms that mediate this change are currently unknown. To address this issue, we tested the idea that changes in intrinsic excitability (induced by learning) facilitate the encoding of new memories via metabotropic glutamate receptor (mGluR) activation. Consistent with this hypothesis, hippocampal neurons exhibited increases in intrinsic excitability after learning that lasted for several days. This increase was selective and only observed in neurons that were activated by the learning event. When animals were trained on a new task during this period, excitable neurons were reactivated and memory formation required the activation of mGluRs instead of NMDARs. These data suggest that increases in intrinsic excitability may serve as a metaplastic mechanism for memory formation.

Sex Differences in Context-Driven Reinstatement of Methamphetamine Seeking Is Associated with Distinct Neuroadaptations in the Dentate Gyrus<strong> </strong>

“…In females that experienced reinstatement, neuroadaptive changes in the GCNs indicate that the change in resting membrane potential closer to firing threshold in concert with increases in membrane resistance, function to enhance the propagation of depolarizing current and therefore intrinsic excitability. Enhanced intrinsic excitability of GCNs may contribute to synaptic plasticity such as long-term potentiation in the DG (Zhang et al, 2004;Sametsky et al, 2009;Williams et al, 2015;Crestani et al, 2018), which could result in an activity-dependent strengthening of synapses or network stability that is thought to underlie certain types of learning and memory, in our case preservation of extinguished responses. In males that reinstated, the opposing regulation of resting membrane potential and AHP to reduced excitability suggests that distinct neuroadaptations and perhaps compensatory to the other.…”

“…Regardless, this likely has important implications for the modulation of hippocampal circuitry underlying context-driven reinstatement (Milshtein-Parush et al, 2017;Crestani et al, 2018).…”

“…Increases and decreases in intrinsic excitability of hippocampal neurons may serve as a metaplastic mechanism for memory formation and retention (Milshtein-Parush et al, 2017;Crestani et al, 2018), suggesting that functionally adaptive modulation of excitability during hippocampal learning could promote network stability. Notably, CaMKII and GluNs are implicated in regulating intrinsic excitability of neurons in the hippocampus and elsewhere (Zhang et al, 2004;Nelson et al, 2005;Sametsky et al, 2009;van Welie & du Lac, 2011;Klug et al, 2012;Milshtein-Parush et al, 2017;Crestani et al, 2018), suggesting that altered expression of these proteins in the DG could correlate with changes in the neuronal excitability of GCNs. We examined the density of CaMKII and the ratio of GluN2A to GluN2B to show that higher CaMKII levels in the DG in males that reinstated were associated with reduced intrinsic excitability of GCNs, and enhanced GluN2A/2B ratio in the DG in females that did not reinstate were associated with increased excitability of GCNs.…”

Cognitive Neuroscience: Exciting Developments in Schematic Learning

“…Another new study provides some strong hints. Using the same two-learning, contextual fear conditioning procedures, Crestani and colleagues [14] find that contextual fear conditioning increases the intrinsic excitability of CA1 pyramidal neurons. These excitability increases were limited to the presumed encoding neuronal population (i.e., neurons tagged with an activity-dependent fluorescent protein during training, 'engram neurons'), and persisted for several days.…”

Synaptic and intrinsic plasticity coordinate spike output in cerebellar Purkinje cells

“…Emerging evidence has revealed that learning and memory is implemented not only by activitydependent synaptic plasticity but also by non-synaptic intrinsic plasticity in several neural circuits (Crestani et al, 2018;Lisman et al, 2018;Shim et al, 2018;Zhang and Linden, 2003). Of interest, intrinsic excitability contributes to integration of synaptic inputs and generation of net neuronal output (Hoffman et al, 1997;Lev-Ram et al, 2003, 1995Shim et al, 2017).…”