Squirrels lose half of their synapses when they go into hibernation. The synapses fall apart in relation to temperature changes. The proteins aren't destroyed. They are just detached, possibly as protein complexes floating in the cytoplasm. They get more synapses when they wake back up, but it's not clear that they are the same ones. In fact, they are impaired in recognizing squirrels they've met before after hibernation. Can you imagine? Half of synapses in thalamus, hippocampus, and cortex? If Synaptic Self is at all appropriate book title, do squirrels lose their squirreldentity every winter? Maybe some place holder protein is left behind so the synapses can grow back how they're supposed to. It's not easy to tell that you got the same synapses before and after hibernation. Guess you could induce in vivo LTP and see if it can be maintained over that span. Or use one of Svoboda's fancy window-on-the-brain techniques.