Very nicely described.

Any idea how they can induce LTP in a living animal? I've read stuff about that, but it seems like it would be hard to do if they don't have a slice and can't see what they're doing.

lol, two of my favorite things, chamillionaire and the calcium gradient for potentiation.
I had to model that in software for a neural-net project in school.

chamillionaire is so hot right now, i love Southern Uprising the best. a perfect example of oldskool "fight the power" hiphop. ;)

haha.. matoko - if (or maybe when) my girlfriend finally figures out how big a dork i am and leaves me, me and you is getting married.. i got trapped inside "sound of revenge" in the middle of winter.. h-town thing seems to have lulled a little bit, 'specially cos H.A.W.K. got shot, but I think with Pimp C out of prison, we're just in the quiet before the storm.. enough to make a guy feel proud to be from texas..

mortimer -
i know that they do in vivo LTP, but i'm not entirely certain how they pull it off. i assume they use stereotaxic coordinates and then record as tehy move the electrodes in until they get a proper electrophys signal that indicates the cell-type they're after.. it can last for weeks and weeks while in vitro LTP can only last as long as they can keep the slice alive in a dish..

Coffee Mug: ”One further thing I’d like them to investigate if they got the chance is the synapse-specificity of the LTP 3 that is generated in the presence of an NMDA receptor antagonist.”

Yes. And are proteins produced in the soma that are then transported to the dendritic spines or is mRNA produced and then transported to the spines? How does the process differ for neurons with extended cell bodies? (Some neurons are LONG.)


What about the role of signal backpropagation along the dendrites as a cell fires a signal along its axon?


What role do the glia cells play in memory?

http://www.sfn.org/index.cfm? pag...ings_astrocytes

i assume they use stereotaxic coordinates and then record as tehy move the electrodes in until they get a proper electrophys signal that indicates the cell-type they're after..

It would be interesting to see if they could do something like that with tone conditioning in the hippocampus - stimulate various neurons during different phases of the tone conditioning to try to isolate the synapses responsible for the change in neural response associated with the conditioning. Then maybe see if both artificially induced LTP and the changes associated with the tone conditioning can be blocked by the same chemical. I'm not sure if that would count as demonstrating that LTP is really the mechanism behind learning during tone conditioning - one problem perhaps being that they can remove the hippocampus and still get the same behavioral response.

Fly -
I think they demonstrated in a previous paper that LTP 3 is dependent on transcription. This would imply that mRNA has to be transported to the spines. I think that's rather plausible, but I had been thinking of the NMDA receptor as being crucial in producing the synapse-specific "tag".

Glia probably play a role in memory as they seem to do in everything, but even though I work right next to a lab that specializes in glia and plasticity I am far from understanding the specifics. Seems like there might be some back-and-forth between BDNF and cytokines that could determine an overall level of synaptic change. I dunno. I'll consider glia after I get neurons figured out.

Mortimer -
You can take a shortcut and show that several of the same chemicals (for instance NMDA-antagonists) that block LTP in the amygdala also block fear conditioning. The hippocampus seems responsive to tone-fear, but it isn't required as you mentioned. I think the best way to think of LTP is a large collection of mechanisms by which synaptic connectivity can be altered, some smaller portion of which will actually constitute the changes underlying memory. Use LTP to generate hypotheses, but test them in a live animal learning something. Also, none of this will really be satisfactory as an explanation because memory is a network property, so you've got to understand at least plasticity and connectivity across many synapses.

Coffee Mug: “I'll consider glia after I get neurons figured out.”

I look forward to reading your interpretation.

"after I get neurons figured out"

You mean that they're already figured out & you just want to increase your knowledge of them, or are you actually a scientist about to make a landmark discovery/ies?

Anyway, i hope that you can tell us what neurons do that is *subjective* as well as objective & readily observed in the lab.

Observing how neurons function is only part of the story, imho.
I want you to translate the *objectivity* you experience in the lab when observing neurons from the outside to the *subjectivity* you experience when you *think* about, & with, them.

We can both view the same thing, but we're viewing from different brains. Very *subjective* there, i think.

Upon reading about the molecular basis for LTP2, two questions come to mind:

1) Why is the NMDA receptor required for LTP2, if the calcium entry occurs through the IP3R, and the IP3 formation is triggered by mGluR activation?

2) How does calcium activate translation of specific pre-transcribed mRNAs? Is the mRNA bound in some kind of ribonucleoprotein complex that is dissociated somehow when [Ca2+] increases, for example by binding of Ca-calmodulin?

arosko -

the authors asked the same question.. and i quote: "How might NMDARs gate IP3R-mediated calcium release?"
If you look close at that pretty figure they gave us you can see that they put a little NMDAR at the base of the spine. they're suggesting that glutamate can spill over and activate extra-synaptic NMDARs.. the calcium influx from there is supposed to work in concert with mGluRs somehow to generate dendritic calcium.. they cite a series of experiments by Nakamura and colleagues..

with regard to protein synthesis.. it is known that there are a bunch of mRNAs hanging out near synapses in RNP complexes as you described. the signal you're looking for could be PKC which is activated by calcium and diacylglycerol (the other product of mGluRs).. the authors also suggest ERK/MAPK phosphorylation of translation factors.. i should really get the pdf up.. in fact i'll send it to you.. i think i have your email somewhere.. their explanation is very clear..