Blanco · PLoS computational biology 2015 · Animal experimental study and computational modeling · n=?

Synaptic Homeostasis and Restructuring across the Sleep-Wake Cycle.

Cited 65 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Mechanism-based animal research and computational neural network modeling

PubMed 26020963 · doi:10.1371/journal.pcbi.1004241 · record verified 2026-08-29

What was done

Researchers measured hippocampal phosphorylated Ca2+/calmodulin-dependent protein kinase II (pCaMKIIα) using immunohistochemistry in rats sacrificed immediately after specific sleep-wake states (waking, slow-wave sleep [SWS], and rapid-eye-movement [REM] sleep) with or without prior novel object exposure. They subsequently developed computational network models (a fully connected excitatory network driven by recorded rat hippocampal spike trains and a detailed hippocampal-cortical network) to simulate synaptic homeostasis (downscaling/depression) and synaptic embossing (long-term potentiation, LTP) across sleep stages.

What was found

The abstract reports no numerical values. Control rats exhibited stable pCaMKIIα levels across the sleep-wake cycle. In contrast, rats exposed to novel objects showed a decrease in pCaMKIIα during subsequent SWS followed by a rebound during REM sleep, with REM pCaMKIIα levels correlating proportionally with cortical spindles near SWS-to-REM transitions. In computational simulations, sleep without LTP rescaled synaptic weights toward an intermediate range, whereas LTP triggered near SWS/REM transitions rearranged synaptic weight rankings. Synaptic homeostasis facilitated this controlled restructuring.

Why it matters

This study provides a mechanistic framework linking SWS-related synaptic downscaling and REM-related LTP, suggesting that SWS and REM act synergistically to stabilize overall synaptic weight while selectively restructuring memory traces.

Limits

The abstract does not state the sample size (n) or specific quantitative effect sizes. Findings rely on rodent biology and idealized computational models, which may not capture the full complexity of human sleep-dependent memory consolidation.

Cited by