Prefrontal cortex neuronal ensembles encoding fear drive fear expression during long-term memory retrieval

Dissociable dorsal medial prefrontal cortex ensembles are necessary for cocaine seeking and fear conditioning in mice

The dorsal medial prefrontal cortex (dmPFC) plays a dual role in modulating drug seeking and fear-related behaviors. Learned associations between cues and drug seeking are encoded by a specific ensemble of neurons. This study explored the stability of a dmPFC cocaine seeking ensemble over 2 weeks and its influence on persistent cocaine seeking and fear memory retrieval. In the first series of experiments, we trained TetTag c-fos-driven-EGFP mice in cocaine self-administration and tagged strongly activated neurons with EGFP during the initial day 7 cocaine seeking session. Subsequently, a follow-up seeking test was conducted 2 weeks later to examine ensemble reactivation between two seeking sessions via c-Fos immunostaining. In the second series of experiments, we co-injected viruses expressing TRE-cre and a cre-dependent inhibitory PSAM-GlyR into the dmPFC of male and female c-fos-tTA mice to enable "tagging" of cocaine seeking ensemble or cued fear ensemble neurons with inhibitory chemogenetic receptors. These c-fos-tTA mice have the c-fos promoter that drives expression of the tetracycline transactivator (tTA). The tTA can bind to the tetracycline response element (TRE) site on the viral construct, resulting in the expression of cre-recombinase, which enables the expression of cre-dependent inhibitory chemogenetic receptors and fluorescent reporters. Then we investigated ensemble contribution to subsequent cocaine seeking and fear recall during inhibition of the tagged ensemble by administering uPSEM792s (0.3 mg/kg), a selective ligand for PSAM-GlyR. In both sexes, there was a positive association between the persistence of cocaine seeking and the proportion of reactivated EGFP+ neurons within the dmPFC. More importantly, inhibition of the cocaine seeking ensemble suppressed cocaine seeking, but not recall of fear memory, while inhibition of the fear ensemble reduced conditioned freezing but not cocaine seeking. The results demonstrate that cocaine and fear recall ensembles in the dmPFC are stable, but largely exclusive from one another.

Reactivation of encoding ensembles in the prelimbic cortex supports temporal associations

Fear conditioning is encoded by strengthening synaptic connections between the neurons activated by a conditioned stimulus (CS) and those activated by an unconditioned stimulus (US), forming a memory engram, which is reactivated during memory retrieval. In temporal associations, activity within the prelimbic cortex (PL) plays a role in sustaining a short-term, transient memory of the CS, which is associated with the US after a temporal gap. However, it is unknown whether the PL has only a temporary role, transiently representing the CS, or is part of the neuronal ensembles that support the retrieval, i.e., whether PL neurons support both transient, short-term memories and stable, long-term memories. We investigated neuronal ensembles underlying temporal associations using fear conditioning with a 5-s interval between the CS and US (CFC-5s). Controls were trained in contextual fear conditioning (CFC), in which the CS-US overlaps. We used Robust Activity Marking (RAM) to selectively manipulate PL neurons activated by CFC-5s learning and Targeted Recombination in Active Populations (TRAP2) mice to label neurons activated by CFC-5s learning and reactivated by memory retrieval in the amygdala, medial prefrontal cortex, hippocampus, perirhinal cortices (PER) and subiculum. We also computed their co-reactivation to generate correlation-based networks. The optogenetic reactivation or silencing of PL encoding ensembles either promoted or impaired the retrieval of CFC-5s but not CFC. CFC-5s retrieval reactivated encoding ensembles in the PL, PER, and basolateral amygdala. The engram network of CFC-5s had higher amygdala and PER centralities and interconnectivity. The same PL neurons support learning and stable associative memories.

Sex differences in the rodent medial prefrontal cortex - What Do and Don't we know?

The prefrontal cortex, particularly its medial subregions (mPFC), mediates critical functions such as executive control, behavioral inhibition, and memory formation, with relevance for everyday functioning and psychopathology. Despite broad characterization of the mPFC in multiple model organisms, the extent to which mPFC structure and function vary according to an individual's sex is unclear - a knowledge gap that can be attributed to a historical bias for male subjects in neuroscience research. Recent efforts to consider sex as a biological variable in basic science highlight the great need to close this gap. Here we review the knowns and unknowns about how rodents categorized as male or female compare in mPFC neuroanatomy, pharmacology, as well as in aversive, appetitive, and goal- or habit-directed behaviors that recruit the mPFC. We propose that long-standing dogmatic concepts of mPFC structure and function may not remain supported when we move beyond male-only studies, and that empirical challenges to these dogmas are warranted. Additionally, we note some common pitfalls in this work. Most preclinical studies operationalize sex as a binary categorization, and while this approach has furthered the inclusion of non-male rodents it is not as such generalizable to what we know of sex as a multidimensional, dynamic variable. Exploration of sex variability may uncover both sex differences and sex similarities, but care must be taken in their interpretation. Including females in preclinical research needs to go beyond the investigation of sex differences, improving our knowledge of how this brain region and its subregions mediate behavior and health. This article is part of the Special Issue on "PFC circuit function in psychiatric disease and relevant models".

Dissociable dorsal medial prefrontal cortex ensembles are necessary for cocaine seeking and fear conditioning in mice

The dmPFC plays a dual role in modulating drug seeking and fear-related behaviors. Learned associations between cues and drug seeking are encoded by a specific ensemble of neurons. This study explored the stability of a dmPFC cocaine seeking ensemble over two weeks and its influence on persistent cocaine seeking and fear memory retrieval. In the first series of experiments, we trained TetTag mice in cocaine self-administration and tagged strongly activated neurons with EGFP during the initial day 7 cocaine seeking session. Subsequently, a follow-up seeking test was conducted two weeks later to examine ensemble reactivation between two seeking sessions via c-Fos immunostaining. In the second series of experiments, we co-injected viruses expressing TRE-cre and a cre-dependent inhibitory PSAM-GlyR into the dmPFC of male and female c-fos -tTA mice to enable "tagging" of cocaine seeking ensemble or cued fear ensemble neurons with an inhibitory chemogenetic receptors. Then we investigated their contribution to subsequent cocaine seeking and fear recall during inhibition of the tagged ensemble by administering uPSEM792s (0.3 mg/kg), a selective ligand for PSAM-GlyR. In both sexes, there was a positive association between the persistence of cocaine seeking and the proportion of reactivated EGFP+ neurons within the dmPFC. More importantly, inhibition of the cocaine seeking ensemble suppressed cocaine seeking, but not recall of fear memory, while inhibition of the fear ensemble reduced conditioned freezing but not cocaine seeking. The results demonstrate that cocaine and fear recall ensembles in the dmPFC are stable, but largely exclusive from one another.

Exploring the memory: existing activity-dependent tools to tag and manipulate engram cells

The theory of engrams, proposed several years ago, is highly crucial to understanding the progress of memory. Although it significantly contributes to identifying new treatments for cognitive disorders, it is limited by a lack of technology. Several scientists have attempted to validate this theory but failed. With the increasing availability of activity-dependent tools, several researchers have found traces of engram cells. Activity-dependent tools are based on the mechanisms underlying neuronal activity and use a combination of emerging molecular biological and genetic technology. Scientists have used these tools to tag and manipulate engram neurons and identified numerous internal connections between engram neurons and memory. In this review, we provide the background, principles, and selected examples of applications of existing activity-dependent tools. Using a combination of traditional definitions and concepts of engram cells, we discuss the applications and limitations of these tools and propose certain developmental directions to further explore the functions of engram cells.

Approaches and considerations of studying neuronal ensembles: a brief review

First theorized by Hebb, neuronal ensembles have provided a framework for understanding how the mammalian brain operates, especially regarding learning and memory. Neuronal ensembles are discrete, sparsely distributed groups of neurons that become activated in response to a specific stimulus and are thought to provide an internal representation of the world. Beyond the study of region-wide or projection-wide activation, the study of ensembles offers increased specificity and resolution to identify and target specific memories or associations. Neuroscientists interested in the neurobiology of learning, memory, and motivated behavior have used electrophysiological-, calcium-, and protein-based proxies of neuronal activity in preclinical models to better understand the neurobiology of learned and motivated behaviors. Although these three approaches may be used to pursue the same general goal of studying neuronal ensembles, technical differences lead to inconsistencies in the output and interpretation of data. This mini-review highlights some of the methodologies used in electrophysiological-, calcium-, and protein-based studies of neuronal ensembles and discusses their strengths and weaknesses.

Ventral hippocampus neurons encode meal-related memory

The ability to encode and retrieve meal-related information is critical to efficiently guide energy acquisition and consumption, yet the underlying neural processes remain elusive. Here we reveal that ventral hippocampus (HPCv) neuronal activity dynamically elevates during meal consumption and this response is highly predictive of subsequent performance in a foraging-related spatial memory task. Targeted recombination-mediated ablation of HPCv meal-responsive neurons impairs foraging-related spatial memory without influencing food motivation, anxiety-like behavior, or escape-mediated spatial memory. These HPCv meal-responsive neurons project to the lateral hypothalamic area (LHA) and single-nucleus RNA sequencing and in situ hybridization analyses indicate they are enriched in serotonin 2a receptors (5HT2aR). Either chemogenetic silencing of HPCv-to-LHA projections or intra-HPCv 5HT2aR antagonist yielded foraging-related spatial memory deficits, as well as alterations in caloric intake and the temporal sequence of spontaneous meal consumption. Collective results identify a population of HPCv neurons that dynamically respond to eating to encode meal-related memories.

Insights into the encoding of memories through the circuitry of fear

Associative learning induces physical changes to a network of cells, known as the memory engram. Fear is widely used as a model to understand the circuit motifs that underpin associative memories. Recent advances suggest that the distinct circuitry engaged by different conditioned stimuli (e.g. tone vs. context) can provide insights into what information is being encoded in the fear engram. Moreover, as the fear memory matures, the circuitry engaged indicates how information is remodelled after learning and hints at potential mechanisms for consolidation. Finally, we propose that the consolidation of fear memories involves plasticity of engram cells through coordinated activity between brain regions, and the inherent characteristics of the circuitry may mediate this process.

Astrocyte regulation of synaptic signaling in psychiatric disorders

Over the last 15 years, the field of neuroscience has evolved toward recognizing the critical role of astroglia in shaping neuronal synaptic activity and along with the pre- and postsynapse is now considered an equal partner in tripartite synaptic transmission and plasticity. The relative youth of this recognition and a corresponding deficit in reagents and technologies for quantifying and manipulating astroglia relative to neurons continues to hamper advances in understanding tripartite synaptic physiology. Nonetheless, substantial advances have been made and are reviewed herein. We review the role of astroglia in synaptic function and regulation of behavior with an eye on how tripartite synapses figure into brain pathologies underlying behavioral impairments in psychiatric disorders, both from the perspective of measures in postmortem human brains and more subtle influences on tripartite synaptic regulation of behavior in animal models of psychiatric symptoms. Our goal is to provide the reader a well-referenced state-of-the-art understanding of current knowledge and predict what we may discover with deeper investigation of tripartite synapses using reagents and technologies not yet available.

Sex differences in fear responses: Neural circuits

Women have increased vulnerability to PTSD and anxiety disorders compared to men. Understanding the neurobiological underpinnings of these disorders is critical for identifying risk factors and developing appropriate sex-specific interventions. Despite the clear clinical relevance of an examination of sex differences in fear responses, the vast majority of pre-clinical research on fear learning and memory formation has exclusively used male animals. This review highlights sex differences in context and cued fear conditioning, fear extinction and fear generalization with a focus on the neural circuits underlying these behaviors in rodents. There are mixed reports of behavioral sex differences in context and cued fear conditioning paradigms, which can depend upon the behavioral indices of fear. However, there is greater evidence of differential activation of the hippocampus, amygdalar nuclei and the prefrontal cortical regions in male and female rodents during context and cued fear conditioning. The bed nucleus of the stria terminalis (BNST), a sexually dimorphic structure, is of particular interest as it differentially contributes to fear responses in males and females. In addition, while the influence of the estrous cycle on different phases of fear conditioning is delineated, the clearest modulatory effect of estrogen is on fear extinction processes. Examining the variability in neural responses and behavior in both sexes should increase our understanding of how that variability contributes to the neurobiology of affective disorders. This article is part of the Special Issue on 'Fear, anxiety and PTSD'.

Technologies for large-scale mapping of functional neural circuits active during a user-defined time window

The mapping of neural circuits activated during behavior down to individual neurons is crucial for decoding how the brain processes information. Technologies allowing activity-dependent labeling of neurons during user-defined restricted time windows are rapidly developing. Precise marking of the time window with light, in addition to chemicals, is now possible. In these technologies, genetically encoded molecules integrate molecular events resulting from neuronal activity with light/drug-dependent events. The outputs are either changes in fluorescence or activation of gene expression. Molecular reporters allow labeling of activated neurons for visualization and cell-type identification. The transcriptional readout also allows further control of activated neuronal populations using optogenetic tools as reporters. Here we review the design of these technologies and discuss their demonstrated applications to reveal previously unknown connections in the mammalian brain. We also consider the strengths and weaknesses of the current approaches and provide a perspective for the future.

Sex-Specific Neural Networks of Cued Threat Conditioning: A Pilot Study

Cued threat conditioning is the most common preclinical model for emotional memory, which is dysregulated in anxiety disorders and post-traumatic stress disorder. Though women are twice as likely as men to develop these disorders, current knowledge of threat conditioning networks was established by studies that excluded female subjects. For unbiased investigation of sex differences in these networks, we quantified the neural activity marker c-fos across 112 brain regions in adult male and female mice after cued threat conditioning compared to naïve controls. We found that trained females engaged prelimbic cortex, lateral amygdala, cortical amygdala, dorsal peduncular cortex, and subparafasicular nucleus more than, and subparaventricular zone less than, trained males. To explore how these sex differences in regional activity impact the global network, we generated interregional cross-correlations of c-fos expression to identify regions that were co-active during conditioning and performed hub analyses to identify regional control centers within each neural network. These exploratory graph theory-derived analyses revealed sex differences in the functional coordination of the threat conditioning network as well as distinct hub regions between trained males and females. Hub identification across multiple networks constructed by sequentially pruning the least reliable connections revealed globus pallidus and ventral lateral septum as the most robust hubs for trained males and females, respectively. While low sample size and lack of non-associative controls are major limitations, these findings provide preliminary evidence of sex differences in the individual circuit components and broader global networks of threat conditioning that may confer female vulnerability to fear-based psychiatric disease.

Spatiotemporal reorganization of corticostriatal networks encodes motor skill learning

Motor skill learning requires the activity of the dorsal striatum, with a differential global implication of the dorsomedial and dorsolateral territories. We investigate here whether and how specific striatal neurons encode the acquisition and consolidation of a motor skill. Using ex vivo two-photon calcium imaging after rotarod training, we report that highly active (HA) striatal populations arise from distinct spatiotemporal reorganization in the dorsomedial (DMS) and dorsolateral (DLS) striatum networks and are correlated with learning performance. The DMS overall activity decreases in early training, with few and sparsely distributed HA cells, while the DLS shows a progressive and long-lasting formation of HA cell clusters. These reorganizations result from reinforcement of synaptic connections to the DMS and anatomical rearrangements to the DLS. Targeted silencing of DMS or DLS HA cells with the cFos-TRAP strategy strongly impairs individual performance. Our data reveal that discrete domains of striatal populations encode acquisition and long-lasting retention of a motor skill.

Comparison of prefrontal cortex sucrose seeking ensembles engaged in multiple seeking sessions: Context is key

Encoding of memories, including those associated with prior drug or reward, is thought to take place within distinct populations of neurons, termed ensembles. Neuronal ensembles for drug- and reward-seeking have been identified in regions of the medial prefrontal cortex, but much of our understanding of these ensembles is based on experiments that take place in a single reward-associated environment and measure ensemble encoding over short durations of time. In contrast, reward seeking behavior is evident across different reward-associated environments and persists over time. Using TetTag mice and Fos immunohistochemistry, we examined the relationship between persistent sucrose-seeking and ensemble encoding in mice that undergo seeking sessions in the same or different sucrose self-administration contexts 2 weeks apart. We found that prelimbic (PrL) and anterior cingulate cortex ensembles tagged in the first seeking session were highly sensitive to the context in which a second seeking session took place: reactivation of these ensembles was reduced in the same context but elevated in a distinct sucrose self-administration context. Correlational analyses revealed that ensemble reactivation in the PrL was proportional to the persistence of sucrose seeking behavior across sessions in differing ways in female mice. In the same context, reactivation was proportional to the persistence of non-reinforced operant responses, whereas in a distinct context, reactivation was proportional to the persistence of non-reinforced head entries into the sucrose receptacle. This study underlines the importance of the medial prefrontal cortex importance in maintaining a reward-seeking ensemble over time and identifies context-dependent changes in behavioral correlates of ensemble reactivation.

Glutamatergic Neurotransmission Controls the Functional Lateralization of the mPFC in the Modulation of Anxiety Induced by Social Defeat Stress in Male Mice

The rodent medial prefrontal cortex (mPFC) is anatomically divided into cingulate (Cg1), prelimbic (PrL), and infralimbic (IL) subareas. The left and right mPFC (L and RmPFC) process emotional responses induced by stress-related stimuli, and LmPFC and RmPFC inhibition elicit anxiogenesis and anxiolysis, respectively. Here we sought to investigate (i) the mPFC functional laterality on social avoidance/anxiogenic-like behaviors in male mice subjected to chronic social defeat stress (SDS), (ii) the effects of left prelimbic (PrL) inhibition (with local injection of CoCl₂) on the RmPFC glutamatergic neuronal activation pattern (immunofluorescence assay), and (iii) the effects of the dorsal right mPFC (Cg1 + PrL) NMDA receptor blockade (with local injection of AP7) on the anxiety induced by left dorsal mPFC inhibition in mice exposed to the elevated plus maze (EPM). Results showed that chronic SDS induced anxiogenic-like behaviors followed by the rise of ΔFosB labeling and by ΔFosB + CaMKII double-labeling bilaterally in the Cg1 and IL subareas of the mPFC. Chronic SDS also increased ΔFosB and by ΔFosB + CaMKII labeling only on the right PrL. Also, the left PrL inhibition increased cFos + CaMKII labeling in the contralateral PrL and IL. Moreover, anxiogenesis induced by the left PrL inhibition was blocked by NMDA receptor antagonist AP7 injected into the right PrL. These findings suggest the lateralized control of the glutamatergic neurotransmission in the modulation of emotional-like responses in mice subjected to chronic SDS.

A neural substrate of compulsive alcohol use

Alcohol intake remains controlled in a majority of users but becomes "compulsive," i.e., continues despite adverse consequences, in a minority who develop alcohol addiction. Here, using a footshock-punished alcohol self-administration procedure, we screened a large population of outbred rats to identify those showing compulsivity operationalized as punishment-resistant self-administration. Using unsupervised clustering, we found that this behavior emerged as a stable trait in a subpopulation of rats and was associated with activity of a brain network that included central nucleus of the amygdala (CeA). Activity of PKCδ+ inhibitory neurons in the lateral subdivision of CeA (CeL) accounted for ~75% of variance in punishment-resistant alcohol taking. Activity-dependent tagging, followed by chemogenetic inhibition of neurons activated during punishment-resistant self-administration, suppressed alcohol taking, as did a virally mediated shRNA knockdown of PKCδ in CeA. These findings identify a previously unknown mechanism for a core element of alcohol addiction and point to a novel candidate therapeutic target.

Differential roles of prelimbic and anterior cingulate cortical region in the modulation of histaminergic and non-histaminergic itch

Itch is an unpleasant sensation that evokes a desire to scratch. Itch processing in the peripheral and spinal cord has been studied extensively, but the mechanism of itch in the central nervous system is still unclear. Anterior cingulate cortex (ACC) and prelimbic cortex (Prl), two subregions of the prefrontal cortex closely related to emotion and motivation, have been reported to be activated during itching in a series of functional imaging studies. However, the exact role of Prl and the differences between ACC and Prl in itch modulation remains unknown. To directly test the differential roles of ACC and Prl in itch processing, we chemogeneticlly inhibited the caudal ACC and Prl, respectively. We found that inhibition of caudal ACC reduced histaminergic but not non-histaminergic itch-induced scratching behaviors. In contrast, inhibition of Prl reduced both histaminergic and non-histaminergic itch-induced scratching behaviors. Our study provided direct evidence of Prl involvement in itch modulation and revealed the differential roles of caudal ACC and Prl in regulating histaminergic and non-histaminergic itch.

Odor modulates the temporal dynamics of fear memory consolidation

Systems consolidation (SC) theory proposes that recent, contextually rich memories are stored in the hippocampus (HPC). As these memories become remote, they are believed to rely more heavily on cortical structures within the prefrontal cortex (PFC), where they lose much of their contextual detail and become schematized. Odor is a particularly evocative cue for intense remote memory recall and despite these memories being remote, they are highly contextual. In instances such as posttraumatic stress disorder (PTSD), intense remote memory recall can occur years after trauma, which seemingly contradicts SC. We hypothesized that odor may shift the organization of salient or fearful memories such that when paired with an odor at the time of encoding, they are delayed in the de-contextualization process that occurs across time, and retrieval may still rely on the HPC, where memories are imbued with contextually rich information, even at remote time points. We investigated this by tagging odor- and non-odor-associated fear memories in male c57BL/6 mice and assessed recall and c-Fos expression in the dorsal CA1 (dCA1) and prelimbic cortex (PL) 1 or 21 d later. In support of SC, our data showed that recent memories were more dCA1-dependent whereas remote memories were more PL-dependent. However, we also found that odor influenced this temporal dynamic, biasing the memory system from the PL to the dCA1 when odor cues were present. Behaviorally, inhibiting the dCA1 with activity-dependent DREADDs had no effect on recall at 1 d and unexpectedly caused an increase in freezing at 21 d. Together, these findings demonstrate that odor can shift the organization of fear memories at the systems level.