Monday, December 25, 2006

Universal Granules   posted by amnestic @ 12/25/2006 01:47:00 PM

First order of business: RIP James Brown.

RNAs aren't just allowed to roam around the cell unescorted. They are most often found in association with RNA-binding proteins. One of the important stories from the past year was that assocation of RNAs with particular proteins involved in the microRNA interference pathway caused them to be localized to special processing centers in the cell soma called P-Bodies. Another relatively recent buzz follows the discovery that the protein lost in Fragile X mental retardation (FMRP) is an RNA-binding translation-repressor. Yet another area of broad interest has been the association of RNAs coding for synaptic plasticity-related proteins with protein particles attached to molecular motors that crawl up and down dendrites using the microtubule highway.

If you look at the figures from all of these discoveries you will find a lot of pictures of cells containing tiny bright dots. This is the visualization of an RiboNucleoProtein (RNP) granule. If your career didn't depend on it and you liked to dream, you might just go ahead and assume that those granules found in yeast, frog eggs, fly neurons, and mammal neurons were all pretty much the same things and continue to develop your mental model with this concept in mind. Fortunately, all you need is an international collaboration of 17 scientists to confirm your suspicions. Barbee et al just reported that, at least in fly neurons, we can find most of disparate translational repression machines in the very same granules. The summary table is below. Points of interest include AGO2, which is the catalytic center of RNA interference; DCP1, the hallmark of a P-Body; FMRP, the fragile X mental retardation protein; and Staufen, the marker for neuronal granules that can be found travelling up and down dendrites in just the right places to be useful for local protein synthesis.

There are so many pathways leading to RNA degradation these days, I've lost hope of keeping track. Perhaps I'll be able to make time to do a review in the next few months. I just ordered this lovely book. It should be noted that not all granules/particles/bodies were created equal. It is possible to have a granule with FMRP and no Staufen and vice versa. This leaves open the possibility of individual granule identities for specific translational control tasks. The relation of RNA interference to neuronal translational control is moving along surprisingly slowly, but we've had a couple good illustrations so far. One of the most interesting is was the discovery by Ashraf et al. (pdf) that olfactory learning in flies leads to proteasomal destruction of RNAi machinery at synapses and derepression of CaMKII translation. In other words, learning allows a very important synaptic constituent to be made, and the reason it isn't being made all the time is probably because its coding RNA is tied up in one of these repression granules.

I have visions of a gradient of RNP granule identities diffusing away from the focus of synaptic potentiation. Since FMRP seems especially good at repressing depression related proteins it should remain intact at the area of highest calcium influx, repressing depression proteins while repressors of LTP related proteins (i.e. maybe Cup or parts of the miRNA machinery) are destroyed by the proteasome. Further away the relationship flips over such that, within a dendritic neighborhood, an overall excitability is maintained by suppressing potentiation related proteins and allowing production of those that lead to depression. This is just an order that would please me. So far I have not found the science.