Manipulating Hippocampal Place Cell Activity by Single-Cell Stimulation in Freely Moving Mice

Abstract

Learning critically depends on the ability to rapidly form and store non-overlapping representations of the external world. In line with their postulated role in episodic memory, hippocampal place cells can undergo a rapid reorganization of their firing fields upon contextual manipulations. To explore the mechanisms underlying such global remapping, we juxtacellularly stimulated 42 hippocampal neurons in freely moving mice during spatial exploration. We found that evoking spike trains in silent neurons was sufficient for creating place fields, while in place cells, juxtacellular stimulation induced a rapid remapping of their place fields to the stimulus location. The occurrence of complex spikes was most predictive of place field plasticity. Our data thus indicate that plasticity-inducing stimuli are able to rapidly bias place cell activity, simultaneously suppressing existing place fields. We propose that such competitive place field dynamics could support the orthogonalization of the hippocampal map during global remapping.

Transient effect of mossy fiber stimulation on spatial firing of CA3 neurons

“…Previous studies in CA1 have shown that a single episode of intracellular stimulation, that is strong enough to evoke complex spikes associated with dendritic plateau potentials, can create a new place field at the stimulation location in head‐fixed mice (Bittner et al, ; Bittner, Milstein, Grienberger, Romani, & Magee, ). A recent study also has shown that juxtacellular stimulation, that is strong enough to induce complex spikes in CA3 or CA1 neurons, can create a new place field at the stimulation location in freely‐moving mice (Diamantaki et al, ). The stimulation effects on CA3 versus CA1 neurons were not compared quantitatively in this study, hence it is unclear whether and how spatial firing stability differs between CA3 and CA1 neurons.…”

“…Previous studies have shown more stable spatial firing in CA3 than CA1 over time (Mankin et al, ; Manns, Howard, & Eichenbaum, ) and in response to ventral tegmental area stimulation (Martig & Mizumori, ) and reward location changes (Dupret, O'Neill, Pleydell‐Bouverie, & Csicsvari, ). Nevertheless, the study by Diamantaki et al () suggests that it is possible to induce long‐lasting changes in CA3 spatial firing by artificially activating CA3 pyramidal neurons. Our mossy fiber stimulations may have failed to create permanent place fields because they came short of generating sufficiently strong complex spikes, even at 50 Hz, due to accordingly increased feedforward inhibition.…”

Bidirectional synaptic plasticity rapidly modifies hippocampal representations

“…9 When both initial ramp amplitude and relative input timing are considered, it is apparent 10 that the preferred conditions for large synaptic depression are that spatial inputs 1) have 11 already been strengthened by previous plasticity, resulting in elevated postsynaptic 12 depolarization at the time of presynaptic spikes, and 2) are activated within a time window 13 ~2 -4 seconds away from a plateau ( Fig. 1J, trace color indicates initial ramp amplitude; 14 Fig 1K; two-dimensional interpolation from data, trace color indicates change in ramp 15 amplitude, see Materials and Methods). In summary, BTSP can either strengthen or 16 weaken synapses in a small number of trials, providing a bidirectional learning 17 mechanism capable of both rapid memory storage and erasure.…”

“…To account for the long time course of BTSP, both models (1,3,15,16,18). 13 Biologically, these traces could correspond to the enzymatic activity of calcium- 14 dependent kinases and phosphatases, and post-translational modification and synaptic 15 localization of proteins that regulate synaptic function (18)(19)(20)(21)(22)(23). While in the voltage- 16 dependent model, the amplitudes of these eligibility signals were modulated by the 17 value of postsynaptic depolarization at the time of presynaptic activation ( Fig.…”

“…However, the two models made 11 qualitatively different predictions about the causal role of the activation state of the 12 postsynaptic neuron in controlling the magnitude and direction of plasticity. While in the 13 voltage-dependent model, correlation between presynaptic activity and postsynaptic 14 depolarization influences the sign of plasticity, in the alternative model, the sign of 15 plasticity is independent of postsynaptic voltage, and is modulated instead by current 16 synaptic weight such that weak synapses tend to potentiate and strong synapses tend to 17 depress (3,(24)(25)(26)(27)(28)(29). 18 We next sought to distinguish between the two model classes with an in vivo 19 perturbation experiment where neuronal Vm was experimentally depolarized by 20 intracellular current injection during plasticity induction trials.…”