Winter has come for most places in North America, and for many creatures that means settling in somewhere and hibernating until the weather warms up. During hibernation, a number of physiological changes occur, such as decreased metabolic rates and lowered core temperatures, in order to conserve energy. Interestingly, the brains of hibernators also undergo morphological changes; specifically, scientists have shown that there is a loss of synaptic protein clustering in hibernating animals, and that upon rewarming to normal body temperatures, these synapses can be rapidly reformed. This process of synapse dismantling and reformation has been proposed as a model of adult synaptic plasticity.
Synapse loss is a hallmark of neurodegenerative diseases, so a group of UK scientists decided to investigate the mechanisms underlying synapse dismantling and reassembly in hibernating animals could give insight into the maintenance and subsequent loss of synapses in models of neurodegenerative disease.
First, Peretti and colleagues showed that laboratory mice demonstrated synaptic dismantling in the hippocampus with artificial cooling to core temperatures of 16-18C and subsequent reassembly with rewarming, similar to that of other small hibernators. Mechanistically, Peretti and colleagues linked an RNA binding protein, RBM3 (RNA-binding motif protein 3) to synaptic reassembly. RBM3 has previously been shown to be upregulated in hypothermic conditions and have neuroprotective effects. It plays a role in promoting global protein synthesis, and is expressed in both neurons and glia (Chip et. al., 2011).
Here, Peretti and colleagues showed that RBM3 is upregulated following artificial cooling, and this upregulation persisted for up to 6 weeks in wild-type mice. What is interesting is that RBM3 seems to be dysregulated in two mouse models of neurodegenerative disease: the 5xFAD model of Alzheimer’s disease, and prion disease (tg37+/- mice infected with Rocky Mountain Laboratory prions). Both models have a delayed onset of synaptic loss and associated behavioral and learning deficits under normal conditions (around 4 months in the 5xFAD model, and 7 weeks post infection in the prion model). In both models, prior to the onset of the disease symptoms animals that were cooled and rewarmed showed synaptic structural plasticity and upregulated RBM3 levels similar to that of wild-type mice. However, animals that were in the disease-stage showed a failure to reassemble synapses upon rewarming, as well as a failure to upregulate RBM3 after cooling. However, efforts to boost RBM3 levels, either through early therapeutic cooling to boost endogenous protein levels or through viral overexpression, showed a rescue of these structural deficits, and consequent behavioral changes. Additionally, viral knockdown of RBM3 accelerated disease progression. In all, it appears that RBM3 is important in synapse structure, either in the formation of new synapses or perhaps in the maintenance of existing ones.
In humans, therapeutic cooling is already used in certain clinical settings, such as cardiac arrest, stroke, traumatic brain/spinal injury, and neonatal encephalopathy, with varying amounts of evidence as to its efficacy. A 2009 study in human Alzheimer’s disease patients has shown that RBM3 mRNA is significantly downregulated compared to age-matched controls. It could be interesting to see whether RBM3 is a viable drug target for human treatments, given the effects of RBM3 overexpression in the rodent models. Alternatively, depending on when the downregulation of RBM3 occurs relative to disease progression, it could be a predictor or a diagnostic marker for the onset of Alzheimer’s disease.