Radioisotope-induced radioactivity in a protein molecule

[Igor 77]

Dear Forum Members,

I am a molecular biologist. One of my projects is focused on the identification of a protein that interacts with a known organic molecule. Namely, I try to chase a transmembrane protein that is known to transport one organic acid. If possible, I would like to get an idea on whether it is possible to induce measurable radioactivity in a protein molecule by subjecting it to a radio-labeled interacting substrate. My idea is to label the transport substrate with a radioisotope, then isolate the whole spectrum of proteins from a given biological membrane, and finally identify the protein that physically interacted with the radiolabeled substrate (namely, transported it) by tracing the induced radiation in that protein.

Specifically, my questions are as follows:

1. Would the strength of the induced signal be measurable?
2. If yes, which method would be the most sensitive?
3. Which isotope would be the most efficient in the induction (out of N, C, O, or H)?

Finally, do you see any alternative way of inducing a measurable change in a target protein?

Thank you in advance!

Igor.

 

[gleem]

No this is not practicable. AFAIK all radioisotopes used for labeling emit gamma/x-radiation or beta particles which will not induce radioactivity in other atoms. The reasons being that they are usually too low in energy and way too low in intensity. Another issue might be if the intensity and appropriate decay products where available the radiation level would probably damage the target material (radiolysis).

 

[mfb]

If your protein doesn't pick up atoms from the substrate it won't get radioactive. If it picks up some atoms you can work with them.

If the protein deposits some atoms you can do the reverse, mark the protein.

A sketch of the system would help.

 

[Igor 77]

Dear Gleem, dear Mfb,

Thank you for your informative replies.

Since in our system the transport protein does not chemically react with the transport substrate, we have no means to transfer the radiolabel from the substrate to the transporter. Currently I try to figure out if chemical cross-linking between the protein and the transported organic acid would do the trick. The problem in this approach is unspecific cross-linking of the radio-labeled organic acid to many other proteins on the membrane surface. There is hope that the preferential interaction of the acid with its dedicated transport protein would result in measurably higher radioactivity associated with that particular protein.

Kind regards,

Igor.

 

[mfb]

What does that interaction do then?

If your protein spends more time at the substrate (migrates slower across the surface?) this would also be something you could look for.

 

[Igor 77]

What does that interaction do then?

If your protein spends more time at the substrate (migrates slower across the surface?) this would also be something you could look for.

Dear Mfb,

Essentially, the substrate passes through a channel-like protein embedded into a lipid bi-layer (cell membrane). The substrate spends only very short time in the protein, but is physically very close to its surface. Also, more than one molecule passes through in a short period of time, which resembles a steady flow.

Thank you,

Igor.

 

[mfb]

What do you want to learn? Which protein let it flow through? Can you prepare substances with a different concentration of different proteins?

 

[snorkack]

A candidate would be hydrogen.
Marking C, N or O would usually be hard, because changing C, N or O requires breaking the molecular backbone.
What is slightly easier is marking hydrogens. There are several hydrogen exchange regions that do not break the molecular backbone and are not visible when all are H.
Note that it is not necessary to use a radioactive marker here (T). D is stable and can be traced by infrared and NMR spectroscopy - and IR or NMR observations of D show its location in molecule, not just its existence as radioactivity of T.
In case of organic acid, though...
Acid hydrogen is promptly exchanged in water. How about oxygen? How quick is the reaction
R-CO(17)O(17)H+H₂O(16)=R-CO(17)O(16)H+H₂O(17)?
O(17) should be conspicuous because unlike O(16) and O(18) it has a spin (although also a quadrupole moment).

 

[Igor 77]

Dear Mfb, dear Snorkack,

Thank you for your informative replies.

Unfortunately, for my application, it would be insufficient to visualize the interaction between homocitrate and its transport protein in the membrane. My goal is to tag the transport protein upon contact with its substrate, so that it can be recognized from thousands of other proteins after the purification step. Actually, isolation of this protein is needed for learning its polypeptide sequence. This way we can see which gene in our model organism codes for this important protein.

In short, we try to learn which protein out of very many (over 100,000) present in a given plant organism can transport homocitrate. So far, no protein with the ability to transport homocitrate has been described in nature. But we know that it exists and we know in which part of the cell it is located. We just need to learn its polypeptide sequence.

Thank you,

Igor.

 

[snorkack]

Found out that carboxylic acid oxygen exchange in water is in a decent timescale... of days.
And can be measurably catalyzed by enzymes.

Oxygen 17 natural concentration is 0.037 %.
Given surplus of water with 100 % oxygen 17, it should be easy to produce homocitrate which is 90+ oxygen 17.
Put it in ordinary water, say 1 % - well, a few days later you might have 50 % oxygen 17 homocitrate dissolved in water that is now 0,5 % oxygen 17. But if your experiment with membrane is done in hours?

You might identify, say water that is 0,05 % oxygen 17 (up from the 0,037 before you added homocitrate) and uninvolved enzymes which are 0,04 % oxygen 17 (exchanging oxygen from water into enzymes takes time in its turn) - and one enzyme, one position in molecule, which is 0,06 % oxygen 17.
Is that the correct approach?

 

[Igor 77]

Dear Snorkack,

Thank you for this suggestion. Since our unknown membrane protein does not exchange atoms/groups with the transported substrate (homocitrate), I consider cross-linking radio-labeled homocitrate to all proteins which it encounters while on the way to its "gate" trough the membrane. Maybe we can detect the cross-linked product and sequence it, to learn its molecular identity.

Just for the case, I attach a figure that explains the situation in the living system. It is from the following paper:

Hakoyama T, Niimi K, Watanabe H, Tabata R, Matsubara J, Sato S, Nakamura Y, Tabata S, Jichun L, Matsumoto T, Tatsumi K, Nomura M, et al (2009). Host plant genome overcomes the lack of a bacterial gene for symbiotic nitrogen fixation. Nature 462: 514-517

Igor.