Thursday, October 26, 2006

LTP protein story breaks down   posted by amnestic @ 10/26/2006 07:50:00 AM

The new Neuron has a bunch of good stuff in it. I don't understand how they expect me to do lab work if they keep giving me all this good junk to read. I think the PI for this report was Tobias Bonhoeffer. It's a short paper with just a few experiments but it has wide implications. The standard story is that activity related to memory storage should cause protein synthesis and those proteins are used to build the changes into synapses to make a lasting change in cellular response to input. Some papers recently have shown that protein degradation may be as important for long-term potentiation (the cellular model for memory) as protein synthesis. Besides the Bingol and Schuman paper linked above there is the example in drosophila of an activity-dependent degradation of the protein, armitage, which released CaMKII RNA from translational repression (Ashraf et al, 2006).

In the Bonhoeffer paper, they did the normal thing experiment where you block the longest lasting form of LTP (late-LTP) with a protein synthesis inhibitor. But then they also tried the same experiment with a protein degradation inhibitor in the solution as well. BooyaShocka! The late-LTP was back to normal levels. Activity must be degrading proteins that could've been used for synaptic plasticity. This is really odd. I don't really know how to interpret it yet. A cluster of synapses could be sharing proteins and so very local degradation could increase specificity of a modification, just like you can add inhibition to a neural network to increase your signal to noise ratio, but when protein synthesis and degradation are both inhibited perhaps the synapses that need to be modified just borrow from their neighbors instead of making new proteins. The proteins that are degraded must be the same ones that are being synthesized as if the overall metabolic rate is increased rather than any specific synthesis. One wonders if there is any sort of local cellular metabolic stress state due to use of amino acids and ATP in these processes.

If this was the case then you could do an interesting "synaptic competition" experiment where you establish LTP in one input and then establish it in a second input under the influence of synthesis and degradation blockers, perhaps the second input would pull some of the material away from the first. This will take some time to digest. The need for protein synthesis was just on the verge of dogma in the field of synaptic plasticity. At the very least, we can expect a much more detailed study of the proteasome (the protein degradation machinery) at synapses in coming months/years.