Temporal lobe epilepsy with distributed hippocampal seizure foci is often intractable and its secondary generalization might lead to sudden death. Early termination through spatially extensive hippocampal intervention is not feasible directly, because of the large size and irregular shape of the hippocampus. In contrast, the medial septum is a promising target to govern hippocampal oscillations through its divergent connections to both hippocampi. Combining this ‘proxy intervention’ concept and precisely timed stimulation, we report here that closed-loop medial septum electrical stimulation can quickly terminate intrahippocampal seizures and suppress secondary generalization in a rat kindling model. Precise stimulus timing governed by internal seizure rhythms was essential. Cell type-specific stimulation revealed that the precisely timed activation of medial septum GABAergic neurons underlaid the effects. Our concept of time-targeted proxy stimulation for intervening pathological oscillations can be extrapolated to other neurological and psychiatric disorders, and has potential for clinical translation.

In the past decades, the cellular and molecular mechanisms underlying memory consolidation, reconsolidation, and extinction have been well characterized. However, the neurobiological underpinnings of forgetting processes remain to be elucidated. Here we used behavioral, pharmacological and electrophysiological approaches to explore mechanisms controlling forgetting. We found that post-acquisition chronic inhibition of the N-methyl-D-aspartate receptor (NMDAR), L-type voltage-dependent Ca2+ channel (LVDCC), and protein phosphatase calcineurin (CaN), maintains long-term object location memory that otherwise would have been forgotten. We further show that NMDAR activation is necessary to induce forgetting of object recognition memory. Studying the role of NMDAR activation in the decay of the early phase of long-term potentiation (E-LTP) in the hippocampus, we found that ifenprodil infused 30 min after LTP induction in vivo blocks the decay of CA1-evoked postsynaptic plasticity, suggesting that GluN2B-containing NMDARs activation are critical to promote LTP decay. Taken together, these findings indicate that a well-regulated forgetting process, initiated by Ca2+ influx through LVDCCs and GluN2B-NMDARs followed by CaN activation, controls the maintenance of hippocampal LTP and long-term memories over time.

Memory fades over time, becoming more schematic or abstract. The loss of contextual detail in memory may reflect a time-dependent change in the brain structures supporting memory. It has been well established that contextual fear memory relies on the hippocampus for expression shortly after learning, but it becomes hippocampus-independent at a later time point, a process called systems consolidation. This time-dependent process correlates with the loss of memory precision. Here, we investigated whether training intensity predicts the gradual decay of hippocampal dependency to retrieve memory, and the quality of the contextual memory representation over time. We have found that training intensity modulates the progressive decay of hippocampal dependency and memory precision. Strong training intensity accelerates systems consolidation and memory generalization in a remarkable timeframe match. The mechanisms underpinning such process are triggered by glucocorticoid and noradrenaline released during training. These results suggest that the stress levels during emotional learning act as a switch, determining the fate of memory quality. Moderate stress will create a detailed memory, whereas a highly stressful training will develop a generic gist-like memory.

Systems consolidation is a time-dependent reorganization process involving neocortical and hippocampal networks underlying memory storage and retrieval. The involvement of the hippocampus during acquisition is well described; however we know much less about the concomitant contribution of cortical activity levels to the formation of stable remote memories. Here, after a reversible pharmacological inhibition of the anterior cingulate cortex (ACC) during the acquisition of a contextual fear conditioning, retrieval of both recent and remote memories were impaired, an effect that was reverted by a single memory reactivation session 48 h after training, through a destabilization-dependent mechanism interpreted as reconsolidation, that restored the normal course of systems consolidation in order to rescue a remote memory. Next we have shown that the integrity of both the anterior cingulate cortex and the thalamic nucleus reuniens (RE) were required for this reactivation-induced memory rescue. Because lidocaine infused into the RE inhibited LTP induction in the CA1-anterior cingulate cortex pathways, it seems that RE is a necessary component of the circuit underlying systems consolidation, mediating communication between dorsal hippocampus and cortical areas. To our notice, this is the first demonstration of the rescue of remote memories disrupted by ACC inhibition during acquisition, via a reconsolidation-driven mechanism. We have also shown the importance of RE to ensure the interconnection among brain areas that collectively seem to control the natural course of systems consolidation and allow the persistence of relevant emotional engrams. © 2017 Wiley Periodicals, Inc.

Fear memory overgeneralization contributes to the genesis and persistence of anxiety disorders and is a central hallmark in the pathophysiology of post-traumatic stress disorder (PTSD). Recent findings suggest that fear generalization is closely related to hippocampal dependency during retrieval. The selective serotonin reuptake inhibitor (SSRI) fluoxetine has been used as a first-line treatment for PTSD; however, how it exerts its therapeutic effect remains a matter of debate. Here, using contextual fear conditioning in rats, we show that chronic fluoxetine treatment prevents fear generalization and enhances subsequent extinction. Moreover, fluoxetine treatment after extinction prevents spontaneous recovery. The mechanism through which fluoxetine affects generalization and extinction seems to be through the postponement of systems consolidation, thereby maintaining hippocampal involvement during retrieval. Such an effect relies on a remodeling of dendritic spines in the hippocampus, as well as the number of mature, mushroom-type spines promoted by fluoxetine treatment. In order to further investigate whether fear generalization is a potential predictor of extinction effectiveness, we categorized a large naive population according to their generalization rate. We found that discriminator rats showed a better extinction profile compared to generalizers, suggesting that the generalization rate predicts extinction effectiveness. Hence, we propose that the therapeutic strategy of choice should take into account the extension of memory generalization, in which therapies based on extinction could induce a better outcome in patients who present less fear overgeneralization. These results open new avenues for the development of interventions that prevent fear generalization by maintaining memory dependency of the hippocampus.

Long-lasting changes in dendritic spines provide a physical correlate for memory formation and persistence. LIM kinase (LIMK) plays a critical role in orchestrating dendritic actin dynamics during memory processing, since it is the convergent downstream target of both the Rac1/PAK and RhoA/ROCK pathways that in turn induce cofilin phosphorylation and prevent depolymerization of actin filaments. Here, using a potent LIMK inhibitor (BMS-5), we investigated the role of LIMK activity in the dorsal hippocampus during contextual fear memory in rats. We first found that post-training administration of BMS-5 impaired memory consolidation in a dose-dependent manner. Inhibiting LIMK before training also disrupted memory acquisition. We then demonstrated that hippocampal LIMK activity seems to be critical for memory retrieval and reconsolidation, since both processes were impaired by BMS-5 treatment. Contextual fear memory extinction, however, was not sensitive to the same treatment. In conclusion, our findings demonstrate that hippocampal LIMK activity plays an important role in memory acquisition, consolidation, retrieval, and reconsolidation during contextual fear conditioning.

Systems consolidation has been described as a time-dependent reorganization process involving the neocortical and hippocampal networks underlying memory storage and retrieval. Previous studies of our lab were able to demonstrate that systems consolidation is a dynamic process, rather than a merely passive, time-dependent phenomenon. Here, we studied the influence of sequential learning in contextual fear conditioning (CFC) with different training intensities in the time-course of hippocampal dependency and contextual specificity. We found that sequential learning with high-intensity shocks during CFC induces generalization of the first learning (context A) and maintains contextual specificity of the second learning (context B) 15 days after acquisition. Moreover, subsequent experiences reorganize brain structures involved in retrieval, accelerating the involvement of cortical structures and diminishing the hippocampal participation. Exposure to original context before novelty seems to only induce context specificity in hippocampal-dependent memories. We propose that systems consolidation could be considered a potential biological mechanism for reducing possible interferences between similar memory traces. © 2017 Wiley Periodicals, Inc.

Although the etiology of major depressive disorder remains poorly understood, impairment of gamma oscillations recently emerged as a potential biomarker for major depression. The olfactory bulb (OB) is a major source of brain wide gamma oscillations and bulbectomy is an animal model for depression. Here we demonstrate that chemogenetic suppression of OB neuronal activity or temporally suppressing the OB to pyriform cortex synaptic transmission decreased gamma oscillation power in multiple brain areas associated with depression-like behaviors. To assess the hypothesized link between depression and diffuse depression of gamma oscillations, we employed gamma phase-dependent closed loop neuromodulation of cortical areas, paced by the native OB output. This procedure alleviated depressive-like behaviors in animals and suggests that restoring gamma oscillations may improve depression in humans.