The Journal of Neuroscience dumped a collection of mini-reviews on us this week concerned with mRNA trafficking and dendritic protein synthesis. Most of them covered fairly familiar ground, but one struck me as containing some really novel and interesting perspectives. Schuman, Dynes, and Steward discussed the limitations of protein synthesis at synapses. For instance, there are, on average, 8 ribosomes per polyribosome and maybe 1 or 2 polyribosomes per synapse in the hippocampus. Polyribosomes are what they sound like, several ribosomes in a string presumably reading off of the same mRNA. This seems to indicate a rather stringent limit on the amount of new proteins that could be produced at a synapse, indicating that synapses better pick and choose wisely. Also, it brings some skepticism to recent discussions of activity-dependent protein synthesis regulation that rely on positive regulation of general translation initiation and elongation factors. What good does it do to have an initiation factor activated if you haven't got a free ribosomal subunit to stick it to?

Another idea that I had not even begun to consider was that the size of mRNAs hanging around synapses might be a factor. They note that dendritically localized mRNAs can be longer than the average dendritic spine when laid out straight and suggest that too many of these might lead to a 'traffic jam,' but mRNAs don't exist in a linearized form very often, so I think a more appropriate indicator might be volume. Let's see. The average spine head diameter in hippocampal CA1 pyramidal cells is 0.4 microns. If we pretend a spine head is a sphere or hemi-sphere we get volumes of either 0.03 or 0.015 microns^3. Schuman et al. provide a length for microtubule-associated protein 2 (MAP2) RNA of 3.17 microns. What is the diameter of a RNA molecule? Let's pretend it is smaller at least than a DNA helix which is about 2.5 nm and that we can approximate the volume as a cylinder: 1.55E-5 microns^3. So, conservatively, I can fit a little less than a thousand of the largest mRNA molecule in an average dendritic spine. Granted there are other proteins in there getting in the way, but I don't think the 'traffic jam' problem is going to end up being an issue.

A further limit discussed in the paper is that of homeostatic inhibition of synaptic modification. It gets only a cursory treatment in the review, but this is a fast-moving area that has only really begun to be explored. There are synaptic events called mini excitatory post-synaptic potentials (mEPSPs or minis) that result from neurotransmitter sort of leaking out of the pre-synaptic bouton even in the absence of action potentials. I hadn't really thought about what these do at all, but Schuman is doing some of the most interesting recent work in this area indicating that minis serve as a tonic suppressor of a natural drive to increase synaptic strength (synaptic scaling). The work is so interesting that I'm gonna devote another post to it instead of stashing it at the bottom of this one. In fact, I'm going to devote several posts to Schuman's recent work.

After I read about the work on mEPSPs, I searched "Schuman EM" in PubMed (XML feed) and had a WTF moment. Take a look at these titles, all within the past 5 months:

• MicroRNA: microRNAs reach out into dendrites.
• Single-trial learning of novel stimuli by individual neurons of the human hippocampus-amygdala complex.
• Miniature neurotransmission stabilizes synaptic function via tonic suppression of local dendritic protein synthesis.
• Selective identification of newly synthesized proteins in mammalian cells using bioorthogonal noncanonical amino acid tagging (BONCAT).
• Activity-dependent dynamics and sequestration of proteasomes in dendritic spines.

The breadth is unbelievable. In most of these papers she uses a different cutting edge technique to investigate each area, from inserting microwires to do single-neuron recording in humans to inventing new amino acids. While she hasn't yet published an empirical paper (rather than a review) on miRNAs and synaptic plasticity, I'm willing to bet a large sum of money that one is coming in the next 6 months or less, probably timed to coincide with the Society for Neuroscience meeting.

In my next five posts, I'll try to get into more detail about each of the five data papers listed there because they are all remarkably creative and apply unique approaches and new techniques to the problems at hand.