[Bronwen Dekker]

I recently edited a protocol for looking at in-cell nmr of proteins in E.coli and Xenopus eggs. Part of the protocol involved getting an nmr spectrum of the stable-isotope-labeled protein in egg extracts.

It occured to me that it would be interesting to ask whether a protein's structure was different in plasma (in comparison to buffer, water, solid matrices...). I did a PubMed and scopus search, but was not able to find a paper where this had obviously been done - perhaps I am not using the correct words.

Is this an experiment that is done?

If yes, do you have a nice reference / or know someone good that I could ask to write a method for it?

If not, why would it not work?

[Yggdrasil]

I think I've seen some work which has been trying to do in vivo NMR, but I don't remember any references.  Crystallography, obviously, would not be a good way to approach this problem, but you could do cryoEM or single particle reconstruction to get low resolution images of protein structure in vivo.  Other people have also done some work to examine the kinetics of protein-protein interactions in crowded in vitro environments to simulate the crowded environments found in the cell (I just saw a seminar by Pat Collier at Caltech on this topic, so you may want to look up his research although he doesn't do structure).

[Palmbeach]

Hi! Maybe this article can give you some ideas? "NMR spectroscopy of Group 13 metal ions: biologically relevant aspects" It is on pubmed, I just searched for "IN Vivo NMR". I think your question was very interesting since I never thought of NMR in In Vivo terms. But how could you differ all the proteins in a crowded cellular environment for each other, should not the 13C signals in a NOESY spectra somehow overlap each other if the proteins are really close to each other and so on? Would be nice to hear what you come up with!

Best regards Palmbeach

[Yggdrasil]

I think ideally, you would do multidimensional NMR with ²H, ¹⁵N, and ¹³C labeled proteins, so that the probability of seeing coupling between unlabeled neighboring protiens would be very low.  The one huge challenge with in vivo NMR would be issues with concentration.  If you were to inject your labeled protein at physiological copy number, I doubt you would get a strong enough signal to do any analysis.  IIRC, typical protein concentration for NMR are around 0.25-0.5mM, which is pretty high.

[Bronwen Dekker]

While I agree with Yggdrasil that it would be virtually impossible to study proteins at physiological copy number using nmr at this stage, I think that a lot of interesting information could be found by looking at these isotopically labeled proteins at a much-higher-than-physiological concentrations.

People who have done such work (e.g. the groups of Volker Doetsch and Alexander Shekhtman) have looked at the protein-in-question by overexpressing it in E.coli (feeding the bugs with isotopically labeled amino acids) and then either performing nmr of a bacterial-suspension or purifying the protein and injecting it into Xenopus oocytes (and/or egg extracts).

I originally posted the question, because I found it odd that these researchers had done such complicated experiments, but that I could not find a citation for just bunging the purified, labeled protein into plasma. 

I must look into the Group 13 metal ions! My radar had not picked up on those at all - thank you for the tip!

[pselenko]

high-resolution NMR of labeled compounds/proteins in cellular extracts or any other biofluids will prove a powerful tool to evaluate many more parameters than just protein conformation. And in many of these applications protein concentrations will be secondary. Why have people not done these types of measurements before? Well, I guess that's just the nature of science.

[Bronwen Dekker]

On the off-chance that anyone is still interested in chatting about this, there are a few replies to the same question posted in the NMR forum on the Nature Network.

http://network.nature.com/forums/nmr/471

[Yggdrasil]

Holy crap, they actually got this to work!

Sakakibara et al.  (2009)  Protein structure determination in living cells by in-cell NMR spectroscopy.  Nature 458, 102-105.  doi:10.1038/nature07814