Diversity and Mechanisms of State-Dependent Regulation of Synaptic Plasticity.

Abstract

The Synaptic Homeostasis Hypothesis (SHY) posits that a fundamental sleep function lies in regulating synaptic strengths through synaptic weakening. However, the mechanistic details underlying this process, and how extensively it occurs in the mammalian brain, are largely unknown. Studies have focused on specific neocortical regions and excitatory synapses. And it is unclear how inhibitory neurons respond to sleep and sleep loss or whether synapses across all brain regions undergo sleep-dependent weakening. Nevertheless, SHY continues to significantly impact sleep research, with many designing their experiments and interpreting their data under the framework that sleep is a global and uniform process. My thesis work challenges this assumption, working from an overarching hypothesis that sleep dependent synaptic plasticity varies as function of brain region, cell type, and prior experience, and thus pushes the field to consider how sleep may engender divergent plasticity mechanisms across the brain. In my thesis, I interrogated cell type and region-specific changes in gene expression following sleep deprivation (SD). I used translating ribosome purification (TRAP) to isolate ribosome-associated mRNA from either Camk2a-expressing (excitatory) neurons or parvalbumin-expressing (PV+; inhibitory) interneurons of ad lib sleeping or sleep deprived (SD) mice. To look at region-specific changes in transcript abundance after sleep vs. sleep loss, mRNA was isolated from these cell populations in the neocortex and hippocampus. Using quantitative PCR (qPCR), we found significant cell type- and region-specific alterations in immediate early gene (IEG) and clock transcript abundance following SD. We found hippocampal populations to be substantially less responsive to SD, an effect heightened within parvalbumin cells. We used fluorescence in situ hybridization (FISH) on brain tissues from sleeping and sleep deprived mice to quantify changes in IEG expression in parvalbumin (identified by Pvalb mRNA expression) and non-parvalbumin (lacking Pvalb expression) cells. These results supported the TRAP findings and revealed layer-specific subregional differences in neocortical and hippocampal expression Overall, the data suggests that synaptic weakening across sleep (and strengthening across wake) is a variable phenomenon, dependent on both brain region and cell type. To test whether slow wave activity (SWA) plays a causal role in sleep dependent synaptic plasticity and visual memory consolidation, I optogenetically stimulated the visual cortices of mice expressing ChR2 in layer 6 corticothalamic neurons (Ntsr1::ChR2). Doing so allowed experimental mimicry of SWA in SD mice. I then collected the affected neocortical tissue to determine whether SWA mimicry results in sleep-like activity dependent gene expression. Our preliminary results are inconclusive but promising. While stimulating SD mice constitutively expressing ChR2 blocked SD-driven increases in Arc and Homer1a expression, we were unable to replicate these findings with AAV-transduced mice. Similarly, while we found that SWA mimicry improved performance of Ntsr1::ChR2 mice in a sleep dependent visual recognition task, we found no difference in performance between non-stimulated ad lib sleep and SD Ntsr1::ChR2 mice. Nevertheless, SWA mimicry appeared to improve visual recognition, suggesting that SWA mimicry during SD may partially rescue visual memory consolidation. My thesis work contributes to the sleep field in two significant ways: (1) it challenges SHY’s conception of a global and uniform sleep-associated plasticity mechanism and (2) it provides preliminary causal evidence for the role of SWA in sleep-induced gene expression and visual memory consolidation. In challenging this framework, it suggests that future studies should view sleep effects on the brain as inherently heterogenous, opening new avenues of research.