Neuronal circuitry and molecular mechanisms regulating memory engrams in health and Alzheimer’s disease

Memories are the basis of our existence and shape who we are. Understanding how and where memories are stored has been a central focus of neuroscience research for more than a century. Memories are stored in the brain by sparse populations of neurons, or ensembles, of engram cells. These neurons undergo structural, physical and molecular changes to encode, store and retrieve memories. A key player in this process is the transcription factor cAMP response element binding protein (CREB), which initiates the transcription of immediate early genes (IEG). However, less is known about the transcriptional mechanisms that spatially and temporally limit a learning-triggered gene response to enhance memory specificity. In the first part of my thesis, I explored the role of transcriptional repression in memory consolidation using healthy mice. Another key question is how memory engrams contribute to pathological memory loss. In the second part of my thesis, I explored whether engram dysfunction contributes to progressive memory impairments in an Alzheimer’s disease (AD) mouse model. AD patients initially show temporally graded retrograde amnesia, which progresses into more severe retrograde amnesia. While AD mouse models provided insight into impaired formation and retrieval of new memories, mechanisms underlying the loss of (far) remote memories remain unclear. This thesis advances our understanding of how engrams support memory in health and disease. In Chapter 2, I identified a previously unknown role for the transcriptional repressor NFIL3 in memory consolidation. While NFIL3 has been implicated in the regulation of CREB-targeted regeneration-associated genes, my work showed that it is also expressed in hippocampal cells following learning. Spatial memory was enhanced in Nfil3 KO mice, while contextual generalization and avoidance learning were impaired, suggesting that NFIL3 is important for adapting behaviour to new experiences. Furthermore, contextual memory retrieval was accompanied by elevated IEG expression in Nfil3 KO mice, highlighting a role for NFIL3 in transcriptional regulation after a behaviourally relevant stimulus. Importantly, Nfil3 was upregulated in Fos- and Arc-expressing neurons in the hippocampal dentate gyrus (DG) after learning, and engram-specific disruption of NFIL3 function suggests that NFIL3 acts within these cells to regulate memory strength. In Chapter 3, I investigated mechanisms underlying remote (1-month-old) memory decline in a widely used AD mouse model, the APP/PS1 mouse, focusing on cortical engrams and inhibitory interneurons. Behavioural experiments revealed an age-dependent deficit in remote memory, emerging around 20 weeks of age, coinciding with hyperexcitability of parvalbumin (PV) interneurons in the medial prefrontal cortex (mPFC). Reactivation of mPFC engram cells remained intact, suggesting that the memory deficit was not caused by a loss of memory-representing engram neurons. Further analysis of these cells showed increased perisomatic PV labelling and enhanced inhibitory input onto these neurons after retrieval, indicating that remote memory deficits may reflect dysregulated inhibitory control over cortical engrams. In Chapter 4, I examined how memory age affects memory in APP/PS1 mice by assessing far-remote (2-month-old) memories. These were impaired at 16 weeks of age, indicating that this deficit precedes remote memory decline. Notably, 20-week-old APP/PS1 mice showed increased reactivation of mPFC engram cells alongside reduced (re)activation of PV interneurons during far-remote memory retrieval, whereas these alterations were absent in 16-week-old mice, suggesting that additional mechanisms contribute to early impairment. I observed that engram reactivation and inhibitory input onto engram cells changed over time, independent of pathology, indicating that memory aging alters how mPFC engram cells are inhibited and reactivated. Together, these data underscore the need to investigate the mechanisms that support far-remote memories and their disruption in AD. Overall, my thesis provides new insight into the mechanisms of engrams across health and disease and identify engram-level mechanisms relevant to AD.