Hi,

How are you doing your denaturation? By boiling in 1% SDS and diluting at least 10 times before the pull-down?

When I used to do this kind of experiment, I observed that the IP was less powerful, so I had to start with a higher quantity of total protein. For example for an overexpressed protein, at least 2-3mg of total protein in the extract (200uL before denaturation, to have 2mL for the IP after denaturation and 10 times dilution).

Also, maybe that your protein is not ubiquitinated K48 or K63, but maybe with K6, K29, linear... So using a K48/63 only is useless in this condition. I would suggest to set up first the condition for the pull-down using a HA-Ub wild-type and then use the K-only.

Hope this can help.

Best,

Anne

When I did my ubiquitination assays with overexpressed proteins, I would do the pull down using Ni-NTA beads under denaturing conditions but blot for the "tag" attached to the protein being expressed rather than blotting for the HA- tagged ubiquitin. 

If your protein is known to be ubiquitinated, you'd expect to see a smear of ubiquitinated protein running on the corresponding size of your protein. The amount/intensity of the smear depends on whether the overexpressed protein is mono- or poly-ubiquitinated. However, whenever I blotted against the HA-tagged ubiquitin (as you have done) I would see a very faint band or none at all at the site corresponding to the ubiquitin. I'm not certain why that is but one interpretation could be that most of the ubiquitin has been used up as they get attached to your overexpressed protein since it is known to be ubiquitinated. 

Not sure if this would answer your question but it's worth giving a try to just blot for your tagged protein instead of the ubiquitin itself.

Hi Gianni,

Have you tried to make dot plot and see how much Ubiq / HA have you expressed?

I suspect you have not optimized the transfectionen. What transfection method do you use? how much plasmid have you transfected with? all these parameter are very important to detect ubiq-proteins.

I would also recommend  treating the cells with the proteasome inhibitor MG115.

Read this article it might help you: one of my publications. 

The F-box protein Skp2 participates in c-Myc proteosomal degradation and acts as a cofactor for c-Myc-regulated transcription.

good Luck

"pull-down the target protein under denaturing conditions " could be the problem. Denaturing conditions affect pulldown.

Thanks to all for the answers. I did a pull-down experiment by using Ni-NTA matrix denaturing the sample with GuHCl 5M to avoid DAB activity. I tryed also NEM (20 mM) but results were worst than Guanidine....As Anne-Claire suggested, I first tested protein ubiquitination with wt-HA-Ub and a clear signal was detected in western with anti-HA antibody. In this case, I observed clear bands corresponding to mono- and poly-ubiquitination, and a smear at higher MW. For this kind of experiment I generally use one well from a 6-well plates (2x106 cells, more or less) transfected by ca-method or commercial reagents (superfect, genejuice...) and always treat samples with MG-132 (2.5 uM) for 12 hrs. Then I test samples by immunofluorescence for both target protein and Ub expression.....pull-down and western takes a lot of time, thus I prefere to be sure to start using a good sample!

Faud, for the transfection I use 500 ng of Ub expressing plasmids and 1.5 ug of target protein expressing plasmid. Transfection efficiency is really good as tested by IFA, and I didn't observed any significant difference when I used Ca-phosphate rather than SuperFect as transfection method.

Thanks a lot

Gianni

Superfect should work fine much better than Ca-Phosph. The amount of plasmid also sounds reasonable...we used to transfect with 0.5-2.0 µg (depends on cell type, plasmid and size of cell culture). All parameters you mentioned are sensible but only thing I've standpoint, the amount of cell extract. You wrote that you were using six well plate.. Take a 5 cm dish / point.

Good Luck