The steady state concentration of a protein that is ubiquitinated is usually very small compared with the steady state concentration of a protein that is not ubiquitinated. This is because ubiquitinated proteins are rapidly recognized by cellular degradation machinery (notably the 19S lid of the proteasome or ESCRT components that initiate lysosomal delivery). MG132 inhibits the proteasome (and to a lesser extent the lysosome), so targets of ubiquitin-proteasome dependent proteolysis will accumulate in MG132-treated cells. However, because the 19S lid's deubiquitinase activity is not affected by MG132, ubiquitin chains will still be removed from the target protein. NEM  inhbits deubiquitinase enzymes, but it also inhibits other enzymes with active site cysteines (e.g., E1 and E2, which are necessary for ubiquitination), such that no new protein ubiquitination can occur upon NEM treatment.

I have a few suggestions. First, it is possible that your protein is ubiquitinated so intensely that the flag epitope becomes masked. In this case, it would be helpful to blot with an antibody against ubiquitin (either K48 chains or monomeric ubiquitin). Secondly, you could try using a more specific deubiquitinase inhibitor, such as PR-619. Another approach would be mass spectrometry. When you carry out an immunoprecipitation, do you have enough protein to visualize by Coomassie blue staining? If so, I would recommend running this sample (the IP) on a gel, perform in-gel trypsin digestion on the entire sample lane, and see if you can find any lysine-linked Gly-Gly mass shifts (the MS signature of ubiquitin).

Good luck! 

I am thinking that you have confirmed the expression of your protein by western blots before performing IP. If you are getting very good over-expression and you think your protein is getting ubiquitinated the best way to check would be to use Urea-lysis buffer. You can just run the blots, and use FL antibody to see the ladder. I would not recommend using FL-HRP antibody to go overnight.

The steady state concentration of a protein that is ubiquitinated is usually very small compared with the steady state concentration of a protein that is not ubiquitinated. This is because ubiquitinated proteins are rapidly recognized by cellular degradation machinery (notably the 19S lid of the proteasome or ESCRT components that initiate lysosomal delivery). MG132 inhibits the proteasome (and to a lesser extent the lysosome), so targets of ubiquitin-proteasome dependent proteolysis will accumulate in MG132-treated cells. However, because the 19S lid's deubiquitinase activity is not affected by MG132, ubiquitin chains will still be removed from the target protein. NEM  inhbits deubiquitinase enzymes, but it also inhibits other enzymes with active site cysteines (e.g., E1 and E2, which are necessary for ubiquitination), such that no new protein ubiquitination can occur upon NEM treatment.

I have a few suggestions. First, it is possible that your protein is ubiquitinated so intensely that the flag epitope becomes masked. In this case, it would be helpful to blot with an antibody against ubiquitin (either K48 chains or monomeric ubiquitin). Secondly, you could try using a more specific deubiquitinase inhibitor, such as PR-619. Another approach would be mass spectrometry. When you carry out an immunoprecipitation, do you have enough protein to visualize by Coomassie blue staining? If so, I would recommend running this sample (the IP) on a gel, perform in-gel trypsin digestion on the entire sample lane, and see if you can find any lysine-linked Gly-Gly mass shifts (the MS signature of ubiquitin).

Good luck! 

Hi Damian,

Thanks for the feedback. Do you recommend that I do not add NEM to the lysis buffer since this might inhibit new protein ubiquitination? Or perhaps I should stick to the recommended dose of 2-10mM NEM?

I will try out the anti-ubiquitin staining and check if I could see the "ladder".

I ran my total cell lysate alongside my IP sample and is able to observe protein bands so I believe I have sufficient for IP. 

Will try out the other suggestions! Thanks so much for your quick reply and awesome suggestions :)

Hi Shyam,

Thanks for the idea of using Urea-Lysis buffer. I admit I have not much idea what is the main purpose until I google it. And it sounds like a good idea. I will definitely try it out for the next experiment! :)

The best thing is to use anti-ubiquitin antibodies. The ubiquitinated proteins may not  run as  ladders on WB but as very faint smear. 

Hi Kamil,

When I look at my film, I do see a little smear from the size of protein of interest upwards. I am expecting to see a larger smear or even bands of proteins. I am curious what might account for the difference in the size of the smear.

I will definitely try out anti-ubiquitin staining for my next experiment.

  I think that depends on the size or molecular weight of your target protein. If the protein is large, it is difficult to see a ladder or higher bands.  You may need to run a lower percentage gel to get better resolution. Remember there is only a small fraction of your protein that is ubiquitinated. Protein molecules are not ubiquitinated to the same extent. Some molecules have one ubiquitin, some two..3...5....hundreds....etc.. That's why you are likely to see a smear and not ladder. (google inositol,1,4,5,trisphosphate receptor and ubiquitination). This protein is ubiquitninated when cells are stimulated but it does not run as ladder. The best we could see is smear. If your target protein is small (e.g 25kDa) you may see a ladder, but not always.

Hi Sy Gun,

As Damian and Kamil said the levels of  the ubiquitinatinated protein-X would be very less (may be 1%) in comparison to the overexpressed protein-X; is completely true.

I understand your concern and I have some suggestions:

1- Check what is the half life of your overexpressed protein-X. That always affect the overall ubiquitination levels in the lyaste at a given time upon MG132 treatment. If it is more than 12h then it is difficult to detect. In my experiments I could detect ubiquitinated ladder of IkBa protein (6h half life). [You may see my paper -- I Sahu. et al 2014, FEBSJ]

2-First overexpress your protein for 24h then block protein synthesis by cycloheximide and perform your all other steps (including MG132 for 3h and then IP). Sometimes CHX with MG132 treatments works; to detect ubiquitinated form of your protein. Here you have to play around with the time points (or use half life time point) of prior treatment of CHX followed by 3h MG132/velcade treatment.

3- Have you seen any ubiquitination ladder of your endogenous protein? And you can try IP with anti-X antibody after overexpressing. As suggested by others you can also use anti Ub antibody after IP to detect.

4- As already suggested you can try in solution MS/MS or cut the whole lane/well gel and send it for ub-Protein-X unique peptide. NOTE: Cut out the overexpressed Protein-X, Heavy and light chain bands from the gel; Since they may interfere in MS/MS sensitivity.

Using NEM/IAN in lysis buffer should not be a concern for its inhibitory action against E1/E2/E3..!!! Since there is no more new ubiquitination occur after cell lysis (as you said), however they will inhibit DUBs.

ONE TRICK: (Technical)

You must have observed; "While probing with antibody,  if you have one overexpressed protein in one well (lane) it scavenges all the antibodies, although we saturate the membrane with enough antibody". So what you can do-- Cut the membrane just above the main band of your overexpressed protein-X and probe the two pieces of membrane separately. Then develop them together (for WB). This works...!!! You may see those 1% ubiquitinated Protein-X. (Try with all the antibodies).

I hope this may help you.

Good Luck.

Hi all,

I managed to work out the problem. Apparently, the cell line seems to be the major problem. I was using the general HEK293T cells transfected with my protein of interest (POI) and after that I switched to a cell line that is more specific to my in vivo model. 

Another thing I observed is that IP samples should be run on SDS-PAGE on the same day. Otherwise, the protein will lose its ubiquitin. When I compared samples run on the same day vs the following day, I get clear ubiquitinated bands for the same day gel and a smear for samples run on the following day.

I have done a Cycloheximide chase assay and determined my POI to have a half-life of ~20h. I understand that if the half life is more than 12h, it might be hard to detect. I hope that the fact that I detect the ubiquitinated protein suggest that my proteins are easily ubiquitinated for degradation. 

Very thankful to all for your suggestions. I am replying after so long because I hope that anyone with the same problem as me will find solutions in this Q&A. And lastly, I am really grateful that our research community is so willing to share and help amateurs like me. May the goddess of successful experiment shine on all of you!

I’m jumping on the thread to ask a similar question- Would i be able to perform the same experiment however instead of doing IP i would do CO-IP and look for ubiquitination of interacting protein.. Would this work?