Extended Poisson Gaussian-Process Latent Variable Model for Unsupervised Neural Decoding.

Dimension reduction on neural activity paves a way for unsupervised neural decoding by dissociating the measurement of internal neural pattern reactivation from the measurement of external variable tuning. With assumptions only on the smoothness of latent dynamics and of internal tuning curves, the Poisson gaussian-process latent variable model (P-GPLVM; Wu et al., 2017) is a powerful tool to discover the low-dimensional latent structure for high-dimensional spike trains. However, when given novel neural data, the original model lacks a method to infer their latent trajectories in the learned latent space, limiting its ability for estimating the neural reactivation. Here, we extend the P-GPLVM to enable the latent variable inference of new data constrained by previously learned smoothness and mapping information. We also describe a principled approach for the constrained latent variable inference for temporally compressed patterns of activity, such as those found in population burst events during hippocampal sharp-wave ripples, as well as metrics for assessing the validity of neural pattern reactivation and inferring the encoded experience. Applying these approaches to hippocampal ensemble recordings during active maze exploration, we replicate the result that P-GPLVM learns a latent space encoding the animal's position. We further demonstrate that this latent space can differentiate one maze context from another. By inferring the latent variables of new neural data during running, certain neural patterns are observed to reactivate, in accordance with the similarity of experiences encoded by its nearby neural trajectories in the training data manifold. Finally, reactivation of neural patterns can be estimated for neural activity during population burst events as well, allowing the identification for replay events of versatile behaviors and more general experiences. Thus, our extension of the P-GPLVM framework for unsupervised analysis of neural activity can be used to answer critical questions related to scientific discovery.

Sleep loss diminishes hippocampal reactivation and replay.

Memories benefit from sleep1, and the reactivation and replay of waking experiences during hippocampal sharp-wave ripples (SWRs) are considered to be crucial for this process2. However, little is known about how these patterns are impacted by sleep loss. Here we recorded CA1 neuronal activity over 12 h in rats across maze exploration, sleep and sleep deprivation, followed by recovery sleep. We found that SWRs showed sustained or higher rates during sleep deprivation but with lower power and higher frequency ripples. Pyramidal cells exhibited sustained firing during sleep deprivation and reduced firing during sleep, yet their firing rates were comparable during SWRs regardless of sleep state. Despite the robust firing and abundance of SWRs during sleep deprivation, we found that the reactivation and replay of neuronal firing patterns was diminished during these periods and, in some cases, completely abolished compared to ad libitum sleep. Reactivation partially rebounded after recovery sleep but failed to reach the levels found in natural sleep. These results delineate the adverse consequences of sleep loss on hippocampal function at the network level and reveal a dissociation between the many SWRs elicited during sleep deprivation and the few reactivations and replays that occur during these events.

Dynamics of spike transmission and suppression between principal cells and interneurons in the hippocampus and entorhinal cortex.

Synaptic excitation and inhibition are essential for neuronal communication. However, the variables that regulate synaptic excitation and inhibition in the intact brain remain largely unknown. Here, we examined how spike transmission and suppression between principal cells (PCs) and interneurons (INTs) are modulated by activity history, brain state, cell type, and somatic distance between presynaptic and postsynaptic neurons by applying cross-correlogram analyses to datasets recorded from the dorsal hippocampus and medial entorhinal cortex (MEC) of 11 male behaving and sleeping Long Evans rats. The strength, temporal delay, and brain-state dependency of the spike transmission and suppression depended on the subregions/layers. The spike transmission probability of PC-INT excitatory pairs that showed short-term depression versus short-term facilitation was higher in CA1 and lower in CA3. Likewise, the intersomatic distance affected the proportion of PC-INT excitatory pairs that showed short-term depression and facilitation in the opposite manner in CA1 compared with CA3. The time constant of depression was longer, while that of facilitation was shorter in MEC than in CA1 and CA3. During sharp-wave ripples, spike transmission showed a larger gain in the MEC than in CA1 and CA3. The intersomatic distance affected the spike transmission gain during sharp-wave ripples differently in CA1 versus CA3. A subgroup of MEC layer 3 (EC3) INTs preferentially received excitatory inputs from and inhibited MEC layer 2 (EC2) PCs. The EC2 PC-EC3 INT excitatory pairs, most of which showed short-term depression, exhibited higher spike transmission probabilities than the EC2 PC-EC2 INT and EC3 PC-EC3 INT excitatory pairs. EC2 putative stellate cells exhibited stronger spike transmission to and received weaker spike suppression from EC3 INTs than EC2 putative pyramidal cells. This study provides detailed comparisons of monosynaptic interaction dynamics in the hippocampal-entorhinal loop, which may help to elucidate circuit operations.

Retuning of hippocampal representations during sleep.

Hippocampal representations that underlie spatial memory undergo continuous refinement following formation1. Here, to track the spatial tuning of neurons dynamically during offline states, we used a new Bayesian learning approach based on the spike-triggered average decoded position in ensemble recordings from freely moving rats. Measuring these tunings, we found spatial representations within hippocampal sharp-wave ripples that were stable for hours during sleep and were strongly aligned with place fields initially observed during maze exploration. These representations were explained by a combination of factors that included preconfigured structure before maze exposure and representations that emerged during θ-oscillations and awake sharp-wave ripples while on the maze, revealing the contribution of these events in forming ensembles. Strikingly, the ripple representations during sleep predicted the future place fields of neurons during re-exposure to the maze, even when those fields deviated from previous place preferences. By contrast, we observed tunings with poor alignment to maze place fields during sleep and rest before maze exposure and in the later stages of sleep. In sum, the new decoding approach allowed us to infer and characterize the stability and retuning of place fields during offline periods, revealing the rapid emergence of representations following new exploration and the role of sleep in the representational dynamics of the hippocampus.

Extended Poisson Gaussian-Process Latent Variable Model for Unsupervised Neural Decoding.

Dimension reduction on neural activity paves a way for unsupervised neural decoding by dissociating the measurement of internal neural state repetition from the measurement of external variable tuning. With assumptions only on the smoothness of latent dynamics and of internal tuning curves, the Poisson Gaussian-process latent variable model (P-GPLVM) (Wu et al., 2017) is a powerful tool to discover the low-dimensional latent structure for high-dimensional spike trains. However, when given novel neural data, the original model lacks a method to infer their latent trajectories in the learned latent space, limiting its ability for estimating the internal state repetition. Here, we extend the P-GPLVM to enable the latent variable inference of new data constrained by previously learned smoothness and mapping information. We also describe a principled approach for the constrained latent variable inference for temporally-compressed patterns of activity, such as those found in population burst events (PBEs) during hippocampal sharp-wave ripples, as well as metrics for assessing whether the inferred new latent variables are congruent with a previously learned manifold in the latent space. Applying these approaches to hippocampal ensemble recordings during active maze exploration, we replicate the result that P-GPLVM learns a latent space encoding the animal's position. We further demonstrate that this latent space can differentiate one maze context from another. By inferring the latent variables of new neural data during running, certain internal neural states are observed to repeat, which is in accordance with the similarity of experiences encoded by its nearby neural trajectories in the training data manifold. Finally, repetition of internal neural states can be estimated for neural activity during PBEs as well, allowing the identification for replay events of versatile behaviors and more general experiences. Thus, our extension of the P-GPLVM framework for unsupervised analysis of neural activity can be used to answer critical questions related to scientific discovery.

Simultaneous electrophysiology and optogenetic perturbation of the same neurons in chronically implanted animals using µLED silicon probes.

Micro-light-emitting-diode (µLED) silicon probes feature independently controllable miniature light-emitting-diodes (LEDs) embedded at several positions in each shank of a multi-shank probe, enabling temporally and spatially precise optogenetic neural circuit interrogation. Here, we present a protocol for performing causal and reproducible neural circuit manipulations in chronically implanted, freely moving animals. We describe steps for introducing optogenetic constructs, preparing and implanting a µLED probe, performing simultaneous in vivo electrophysiology with focal optogenetic perturbation, and recovering a probe following termination of an experiment. For complete details on the use and execution of this protocol, please refer to Watkins de Jong et al. (2023).1.

Optogenetics reveals paradoxical network stabilizations in hippocampal CA1 and CA3.

Recurrent connectivity between excitatory neurons and the strength of feedback from inhibitory neurons are critical determinants of the dynamics and computational properties of neuronal circuits. Toward a better understanding of these circuit properties in regions CA1 and CA3 of the hippocampus, we performed optogenetic manipulations combined with large-scale unit recordings in rats under anesthesia and in quiet waking, using photoinhibition and photoexcitation with different light-sensitive opsins. In both regions, we saw striking paradoxical responses: subsets of cells increased firing during photoinhibition, while other cells decreased firing during photoexcitation. These paradoxical responses were more prominent in CA3 than in CA1, but, notably, CA1 interneurons showed increased firing in response to photoinhibition of CA3. These observations were recapitulated in simulations where we modeled both CA1 and CA3 as inhibition-stabilized networks in which strong recurrent excitation is balanced by feedback inhibition. To directly test the inhibition-stabilized model, we performed large-scale photoinhibition directed at (GAD-Cre) inhibitory cells and found that interneurons in both regions increased firing when photoinhibited, as predicted. Our results highlight the often-paradoxical circuit dynamics that are evidenced during optogenetic manipulations and indicate that, contrary to long-standing dogma, both CA1 and CA3 hippocampal regions display strongly recurrent excitation, which is stabilized through inhibition.

Simultaneous Electrophysiology and Optogenetic Perturbation of the Same Neurons in Chronically Implanted Animals using µLED Silicon Probes.

Optogenetics are a powerful tool for testing how a neural circuit influences neural activity, cognition, and behavior. Accordingly, the number of studies employing optogenetic perturbation has grown exponentially over the last decade. However, recent studies have highlighted that the impact of optogenetic stimulation/silencing can vary depending on the construct used, the local microcircuit connectivity, extent/power of illumination, and neuron types perturbed. Despite these caveats, the majority of studies employ optogenetics without simultaneously recording neural activity in the circuit that is being perturbed. This dearth of simultaneously recorded neural data is due in part to technical difficulties in combining optogenetics and extracellular electrophysiology. The recent introduction of µLED silicon probes, which feature independently controllable miniature LEDs embedded at several levels of each of multiple shanks of silicon probes, provides a tractable method for temporally and spatially precise interrogation of neural circuits. Here, we provide a protocol addressing how to perform chronic recordings using µLED probes. This protocol provides a schematic for performing causal and reproducible interrogations of neural circuits and addresses all phases of the recording process: introduction of optogenetic construct, implantation of the µLED probe, performing simultaneous optogenetics and electrophysiology in vivo , and post-processing of recorded data. SUMMARY: This method allows a researcher to simultaneously perturb neural activity and record electrophysiological signal from the same neurons with high spatial specificity using silicon probes with integrated µLEDs. We outline a procedure detailing all stages of the process for performing reliable µLED experiments in chronically implanted rodents.

Erasable Hippocampal Neural Signatures Predict Memory Discrimination.

Memories involving the hippocampus can take several days to consolidate, challenging efforts to uncover the neuronal signatures underlying this process. Using calcium imaging in freely moving mice, we tracked the hippocampal dynamics underlying memory formation across a ten-day contextual fear conditioning (CFC) task. We found that cell turnover between the conditioning chamber and a neutral arena even prior to learning predicted the accuracy of subsequent memory recall the next day. Following learning, context-specific place field remapping correlated with memory performance. To causally test whether these hippocampal dynamics support memory consolidation, we induced amnesia in a group of mice by pharmacologically blocking protein synthesis immediately following learning. We found that halting protein synthesis following learning paradoxically accelerated cell turnover and also arrested learning-related remapping, paralleling the absence of remapping observed in untreated mice that exhibited poor memory expression. Finally, coordinated neural activity that emerged following learning was dependent on intact protein synthesis and predicted memory-related freezing behavior. We conclude that context-specific place field remapping and the development of coordinated ensemble activity require protein synthesis and underlie contextual fear memory consolidation.

Sleep loss diminishes hippocampal reactivation and replay.

Memories benefit from sleep, and sleep loss immediately following learning has a negative impact on subsequent memory storage. Several prominent hypotheses ascribe a central role to hippocampal sharp-wave ripples (SWRs), and the concurrent reactivation and replay of neuronal patterns from waking experience, in the offline memory consolidation process that occurs during sleep. However, little is known about how SWRs, reactivation, and replay are affected when animals are subjected to sleep deprivation. We performed long duration (~12 h), high-density silicon probe recordings from rat hippocampal CA1 neurons, in animals that were either sleeping or sleep deprived following exposure to a novel maze environment. We found that SWRs showed a sustained rate of activity during sleep deprivation, similar to or higher than in natural sleep, but with decreased amplitudes for the sharp-waves combined with higher frequencies for the ripples. Furthermore, while hippocampal pyramidal cells showed a log-normal distribution of firing rates during sleep, these distributions were negatively skewed with a higher mean firing rate in both pyramidal cells and interneurons during sleep deprivation. During SWRs, however, firing rates were remarkably similar between both groups. Despite the abundant quantity of SWRs and the robust firing activity during these events in both groups, we found that reactivation of neurons was either completely abolished or significantly diminished during sleep deprivation compared to sleep. Interestingly, reactivation partially rebounded upon recovery sleep, but failed to reach the levels characteristic of natural sleep. Similarly, the number of replays were significantly lower during sleep deprivation and recovery sleep compared to natural sleep. These results provide a network-level account for the negative impact of sleep loss on hippocampal function and demonstrate that sleep loss impacts memory storage by causing a dissociation between the amount of SWRs and the replays and reactivations that take place during these events.

Chemogenetic Enhancement of cAMP Signaling Renders Hippocampal Synaptic Plasticity Resilient to the Impact of Acute Sleep Deprivation.

Sleep facilitates memory storage and even brief periods of sleep loss lead to impairments in memory, particularly memories that are hippocampus dependent. In previous studies, we have shown that the deficit in memory seen after sleep loss is accompanied by deficits in synaptic plasticity. Our previous work has also found that sleep deprivation (SD) is associated with reduced levels of cyclic adenosine monophosphate (cAMP) in the hippocampus and that the reduction of cAMP mediates the diminished memory observed in sleep-deprived animals. Based on these findings, we hypothesized that cAMP acts as a mediator for not only the cognitive deficits caused by sleep deprivation, but also the observed deficits in synaptic plasticity. In this study, we expressed the heterologous Drosophila melanogaster Gαs-protein-coupled octopamine receptor (DmOctß1R) in mouse hippocampal neurons. This receptor is selectively activated by the systemically injected ligand (octopamine), thus allowing us to increase cAMP levels in hippocampal neurons during a 5-h sleep deprivation period. Our results show that chemogenetic enhancement of cAMP during the period of sleep deprivation prevents deficits in a persistent form of long-term potentiation (LTP) that is induced at the Schaffer collateral synapses in the hippocampal CA1 region. We also found that elevating cAMP levels in either the first or second half of sleep deprivation successfully prevented LTP deficits. These findings reveal that cAMP-dependent signaling pathways are key mediators of sleep deprivation at the synaptic level. Targeting these pathways could be useful in designing strategies to prevent the impact of sleep loss.

Hippocampal sharp-wave ripples in cognitive map maintenance versus episodic simulation.

Hippocampal sharp-wave ripples (SWRs) have been proposed to support memory-based decision-making. A new study by Gillespie et al. (in this issue of Neuron) provides important new insights on how past experiences and future choices are reflected in neuronal activity during SWRs.

Neuronal sequences during theta rely on behavior-dependent spatial maps.

Navigation through space involves learning and representing relationships between past, current, and future locations. In mammals, this might rely on the hippocampal theta phase code, where in each cycle of the theta oscillation, spatial representations provided by neuronal sequences start behind the animal's true location and then sweep forward. However, the exact relationship between theta phase, represented position and true location remains unclear and even paradoxical. Here, we formalize previous notions of 'spatial' or 'temporal' theta sweeps that have appeared in the literature. We analyze single-cell and population variables in unit recordings from rat CA1 place cells and compare them to model simulations based on each of these schemes. We show that neither spatial nor temporal sweeps quantitatively accounts for how all relevant variables change with running speed. To reconcile these schemes with our observations, we introduce 'behavior-dependent' sweeps, in which theta sweep length and place field properties, such as size and phase precession, vary across the environment depending on the running speed characteristic of each location. These behavior-dependent spatial maps provide a structured heterogeneity that is essential for understanding the hippocampal code.

3D-printed Recoverable Microdrive and Base Plate System for Rodent Electrophysiology.

Extracellular recordings in freely moving animals allow the monitoring of brain activity from populations of neurons at single-spike temporal resolution. While state-of-the-art electrophysiological recording devices have been developed in recent years (e.g., µLED and Neuropixels silicon probes), implantation methods for silicon probes in rats and mice have not advanced substantially for a decade. The surgery is complex, takes time to master, and involves handling expensive devices and valuable animal subjects. In addition, chronic silicon neural probes are practically single implant devices due to the current low success rate of probe recovery. To successfully recover silicon probes, improve upon the quality of electrophysiological recording, and make silicon probe recordings more accessible, we have designed a miniature, low cost, and recoverable microdrive system. The addition of a novel 3D-printed skull baseplate makes the surgery less invasive, faster, and simpler for both rats and mice. We provide detailed procedural instructions and print designs, allowing researchers to adapt and flexibly customize our designs to their experimental usage.

Progress and issues in second-order analysis of hippocampal replay.

Patterns of neural activity that occur spontaneously during sharp-wave ripple (SWR) events in the hippocampus are thought to play an important role in memory formation, consolidation and retrieval. Typical studies examining the content of SWRs seek to determine whether the identity and/or temporal order of cell firing is different from chance. Such 'first-order' analyses are focused on a single time point and template (map), and have been used to show, for instance, the existence of preplay. The major methodological challenge in first-order analyses is the construction and interpretation of different chance distributions. By contrast, 'second-order' analyses involve a comparison of SWR content between different time points, and/or between different templates. Typical second-order questions include tests of experience-dependence (replay) that compare SWR content before and after experience, and comparisons or replay between different arms of a maze. Such questions entail additional methodological challenges that can lead to biases in results and associated interpretations. We provide an inventory of analysis challenges for second-order questions about SWR content, and suggest ways of preventing, identifying and addressing possible analysis biases. Given evolving interest in understanding SWR content in more complex experimental scenarios and across different time scales, we expect these issues to become increasingly pervasive. This article is part of the Theo Murphy meeting issue 'Memory reactivation: replaying events past, present and future'.

Author Correction: Neuronal firing rates diverge during REM and homogenize during non-REM.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Neuronal firing rates diverge during REM and homogenize during non-REM.

Neurons fire at highly variable intrinsic rates and recent evidence suggests that low- and high-firing rate neurons display different plasticity and dynamics. Furthermore, recent publications imply possibly differing rate-dependent effects in hippocampus versus neocortex, but those analyses were carried out separately and with potentially important differences. To more effectively synthesize these questions, we analyzed the firing rate dynamics of populations of neurons in both hippocampal CA1 and frontal cortex under one framework that avoids the pitfalls of previous analyses and accounts for regression to the mean (RTM). We observed several consistent effects across these regions. While rapid eye movement (REM) sleep was marked by decreased hippocampal firing and increased neocortical firing, in both regions firing rate distributions widened during REM due to differential changes in high- versus low-firing rate cells in parallel with increased interneuron activity. In contrast, upon non-REM (NREM) sleep, firing rate distributions narrowed while interneuron firing decreased. Interestingly, hippocampal interneuron activity closely followed the patterns observed in neocortical principal cells rather than the hippocampal principal cells, suggestive of long-range interactions. Following these undulations in variance, the net effect of sleep was a decrease in firing rates. These decreases were greater in lower-firing hippocampal neurons but also higher-firing frontal cortical neurons, suggestive of greater plasticity in these cell groups. Our results across two different regions, and with statistical corrections, indicate that the hippocampus and neocortex show a mixture of differences and similarities as they cycle between sleep states with a unifying characteristic of homogenization of firing during NREM and diversification during REM.

Hippocampal Reactivation Extends for Several Hours Following Novel Experience.

New memories are believed to be consolidated over several hours of post-task sleep. The reactivation or "replay" of hippocampal cell assemblies has been proposed to provide a key mechanism for this process. However, previous studies have indicated that such replay is restricted to the first 10-30 min of post-task sleep, suggesting that it has a limited role in memory consolidation. We performed long-duration recordings in sleeping and behaving male rats and applied methods for evaluating the reactivation of neurons in pairs as well as in larger ensembles while controlling for the continued activation of ensembles already present during pre-task sleep ("preplay"). We found that cell assemblies reactivate for up to 10 h, with a half-maximum timescale of ∼6 h, in sleep following novel experience, even when corrected for preplay. We further confirmed similarly prolonged reactivation in post-task sleep of rats in other datasets that used behavior in novel environments. In contrast, we saw limited reactivation in sleep following behavior in familiar environments. Overall, our findings reconcile the duration of replay with the timescale attributed to cellular memory consolidation and provide strong support for an integral role of replay in memory.SIGNIFICANCE STATEMENT Neurons that are active during an experience reactivate again afterward during rest and sleep. This replay of ensembles of neurons has been proposed to help strengthen memories, but it has also been reported that replay occurs only in the first 10-30 min of sleep, suggesting a circumscribed role. We performed long-duration recordings in the hippocampus of rats and found that replay persists for several hours in sleep following novel experience, far beyond the limits found in previous reports based on shorter recordings. These findings reconcile the duration of replay with the hours-long timescales attributed to memory consolidation.

Storage fidelity for sequence memory in the hippocampal circuit.

Episodic memories have been suggested to be represented by neuronal sequences, which are stored and retrieved from the hippocampal circuit. A special difficulty is that realistic neuronal sequences are strongly correlated with each other since computational memory models generally perform poorly when correlated patterns are stored. Here, we study in a computational model under which conditions the hippocampal circuit can perform this function robustly. During memory encoding, CA3 sequences in our model are driven by intrinsic dynamics, entorhinal inputs, or a combination of both. These CA3 sequences are hetero-associated with the input sequences, so that the network can retrieve entire sequences based on a single cue pattern. We find that overall memory performance depends on two factors: the robustness of sequence retrieval from CA3 and the circuit's ability to perform pattern completion through the feedforward connectivity, including CA3, CA1 and EC. The two factors, in turn, depend on the relative contribution of the external inputs and recurrent drive on CA3 activity. In conclusion, memory performance in our network model critically depends on the network architecture and dynamics in CA3.