A few suggestions:

1. Cysteine residues need to bee ionized in order to react (That's why they dont react at low pH). A major requirement for reactivity and ionizability is for the residue to be relatively exposed. So you shouldn't expect to be able to label all residues in a folded protein. Some of them simply do not label...

2. When the protein is solubilized in detergent, you may loose conformational dynamics of the protein and the Cys may be less accesible for that reason. Thus, in my hands, Cys labeling is usualy better in the membrane - first label, then solubilize. Dont forget to check if the labeling agent is membrane-permeable (I thinkl that MTS reagents are generally not). If not, a sonication step after reagent addition might be necessary (you can see my 2009 paper).

3. Try increasing reagent concentration much more, Chemical reaction rates are proportional to the reactant and, also, the reagent is unstable. I would try to work with ~ 2 mM reagent and see what happens.

Good luck!

I do MTSL labeling of my membrane protein at 1:10 molar ratio in PBS, 0.2% SDS at room temperature with shaking for ~12-16 hrs. The mixing is important. With membrane proteins sometimes room temp is enough to make the protein unstable which could be why you are getting some precipitation. Another method I have tried is 16 hrs at 4 degrees (with mixing) and then 4 hrs room temp. As mentioned above, you are not going to get full labeling if your protein is folded. I had trouble with labeling until I got my protein unfolded by increasing the SDS concentration (sometimes gentle heating helps too).

You will need to deal with the precipitation problem I think before you get much improvement. Does your protein precipitate if you leave in cold room overnight without the label present? You may need to increase your detergent concentration some or add lipids as already suggested though I have never done any MTSL labeling in lipids. What concentration is your MTSL stock solution? If you add too much volume the ethanol from the stock solution could also make your protein less stable.

Membrane proteins are often difficult to work with so just keep trying.

Sorry for my late response. I did not check this site in a while. Thank you Amanda and Nir for the informative answers. I have tried both room temperature and 4 degree labeling for several cysteine mutants of my protein . In either case almost 80% of the protein gradually precipitates and the protein already express very low (1mg/L). I have tried 1:10 molar ratio to 1:50 molar ratio. My protein is in 25 mM HEPES, 500 mM NaCl and 0.03 % DDM, pH 7.5. I tried the labeling after first step of My Ni-NTA (after desalting ) purification, as well as after two step purification (final step Size exclusion). I tried adding and removing DTT while purification too, but that did not improve anything. The mutants are single cysteine though my ultimate goal is to make double cysteine mutants for distance measurements. For this I must get efficiency at least greater than 80%. Right now i am still fighting with precipitation and efficiency issue. All the cysteine mutants are quite exposed based on the structure.

If the Cys is on exposed site the original amino acid was probably hydrophilic, correct? Since MTSL is hydrophobic (at least after labeling) you would now have a hydrophobic side-chain at a formerly hydrophillic site. This is likely the source of the aggregation you see. You could try a more hydrophillic spin label like TEMPONE

I have several exposed residues which are actually hydrophobic and still the protein precipitates. Though i would still like to try TEMPONE on hydrophobic mutants as it seems much cheaper than MTSSL.

What are you dissolving the MTSL in? Some solvents such as DMSO can have issues with membrane proteins such as the aggregation problems you've observed.

I am indeed using DMSO so far..do you think changing solvent can help ?