Preexisting hippocampal network dynamics constrain optogenetically induced place fields

Abstract

Neuronal circuits face a fundamental tension between maintaining existing structure and changing to accommodate new information. Memory models often emphasize the need to encode novel patterns of neural activity imposed by "bottom-up" sensory drive. In such models, learning is achieved through synaptic alterations, a process which potentially interferes with previously stored knowledge 1-3 . Alternatively, neuronal circuits generate and maintain a preconfigured stable dynamic, sometimes referred to as an attractor, manifold, or schema 4-7 , with a large reservoir of patterns available for matching with novel experiences 8-13 . Here, we show that incorporation of arbitrary signals is constrained by preexisting circuit dynamics. We optogenetically stimulated small groups of hippocampal neurons as mice traversed a chosen segment of a linear track, mimicking the emergence of place fields 1,14,15 , while simultaneously recording the activity of stimulated and nonstimulated neighboring cells. Stimulation of principal neurons in CA1, but less so CA3 or the dentate gyrus, induced persistent place field remapping. Novel place fields emerged in both stimulated and non-stimulated neurons, which could be predicted from sporadic firing in the new place field location and the temporal relationship to peer neurons prior to the optogenetic perturbation. Circuit modification was reflected by altered spike transmission between connected pyramidal cell -inhibitory interneuron pairs, which persisted during post-experience sleep. We hypothesize that optogenetic perturbation unmasked subthreshold, pre-existing place fields 16,17 . Plasticity in recurrent/lateral inhibition may drive learning through rapid exploration of existing states.The ability for hippocampal circuits to imprint a random, novel pattern was tested in transgenic mice in which channelrhodopsin2 (ChR2) was expressed in excitatory neurons (N = 6 CaMKIIα-Cre:: Ai32 mice). Stimulation was achieved through μLED illumination 18 , which was delivered as mice ran on a linear track (1.2 m) for water reward ( Figure 1A). After ten baseline trials, stimulation (1s half sine wave) was given for one to ten trials (see Supplemental Table 1) at a fixed position and running direction that changed daily. Optogenetic stimulation induced highly focal drive in CA1 neurons (N = 715; rate change on stimulated shank, Wilcoxon signed-rank test p = 4.1 -183 , median number stimulated = 12 neurons , range 1-50 neurons), as pyramidal cells on the neighboring shanks (≥ 250μm away; N = 420) showed no increase in firing even at the highest stimulation intensity ( Figure S1, rate change on non-stim shank, Wilcoxon signed-rank test p = 0.40, median number stimulated = 0 neurons, range 0-5 neurons). On the track, pyramidal cells on the non-stimulated shanks were moderately suppressed in the stimulation zone, as compared to baseline, non-stimulated trials (mean rate change -0.10 ± 0.07 Hz, sign-test, p =7.34 -5 ). In CA3 and the dentate gyrus (DG), the activity of pyramidal neurons on neighboring sha...

The brain in motion: How ensemble fluidity drives memory-updating and flexibility

“…Synapses are constantly being remodeled, which could be one underlying factor for drift in ensemble codes ( Buzsáki, 2010 ; Rumpel and Triesch, 2016 ; Ziv and Brenner, 2018 ). While neural activity is a well-known mediator of synaptic plasticity ( Bi and Poo, 1998 ; Bliss and Collingridge, 1993 ), the reverse relationship is also possible—synaptic structure and weights could influence neural activity patterns ( Buzsáki, 2010 ; Fauth et al, 2015 ; Felipe et al, 2020 ; McKenzie et al, 2019 ; Turrigiano and Nelson, 2004 ). Along those lines, when action potentials are blocked in cell cultures, synapses continue to turn over across days, demonstrating that spontaneous synaptic remodeling occurs even in the absence of neuronal firing, and that these intrinsic remodeling events could impact the participation of individual neurons within ensembles ( Minerbi et al, 2009 ; Yasumatsu et al, 2008 ).…”

“…While excitability could determine the identity of the neurons to be recruited into an ensemble during memory-updating, pre-existing firing patterns (i.e., functional connectivity) at the time of encoding could mediate the temporal structure of the new neural motif destined for eventual potentiation ( Dragoi and Tonegawa, 2014 ; McKenzie et al, 2019 ). The temporal structure of spiking patterns could then be interpreted by downstream reader regions ( Buzsáki and Tingley, 2018 ).…”

Targeted Activation of Hippocampal Place Cells Drives Memory-Guided Spatial Behavior

“…Importantly, our all-optical approach allows us to go beyond these results by demonstrating that remapping effects are not limited to the targeted neurons: indeed, we find widespread and persistent network-level remapping, with changes not being restricted to neurons that responded to stimulation or the subset of place cells that were targeted on that day. The mechanisms underlying these effects may include plasticity of excitatory synapses onto the stimulated cells (Bittner et al, 2015(Bittner et al, , 2017 as well as changes in the synaptic weight matrix underlying lateral inhibition in the CA1 network (English et al, 2017;McKenzie et al, 2019), supported by our observation of the inhibitory influence of stimulation on the endogenous place cell population. This in turn complements previous work showing that interneuron activity and pyramidal-interneuron functional connectivity are altered during spatial learning (Dupret et al, 2013) and that interneurons play a key role in regulating place cell firing (Royer et al, 2012) and place field formation (Sheffield et al, 2017;Turi et al, 2019).…”

“…Our results are consistent with a model in which stimulation creates a bias in representation toward the targeted location, and the downstream effects of this are facilitated by inhibition of the endogenous spatial code. This could be mediated by the recruitment of recurrent local inhibitory neurons ( Freund and Buzsáki, 1996 ; Pouille and Scanziani, 2004 ; Grienberger et al., 2017 ; McKenzie et al., 2019 ), which cause disynaptic inhibition of the endogenous place cell population. The observed increase in suppression following place cell stimulation, but not Non-PC stimulation, suggests that place cells may preferentially interact with members of the same map, communicating through local interneurons to guide network dynamics.…”

Behavioral Time Scale Plasticity of Place Fields: Mathematical Analysis

“…The underlying structure of such plasticity is significantly different than the commonly studied Hebbian forms of plasticity that assume near temporal coincidence of pre and postsynaptic activity. Experimental results in CA1 using both in vivo and slice preparations strongly suggest that presynaptic activity generates synaptic eligibility traces for both LTP and LTD [17,19]. Prior modeling work [18] shows, using simulations, that these two opposing traces, combined with a weight dependence can account for the place field plasticity observed in vivo.…”