(Co)Immunoprecipitation problem - (Apr/08/2008 )

Hi everybody,

I just wanted to describe my problem with IP here and see if someone could have an idea to help me to solve it. I'm trying to immunoprecipitate one protein from a cell lysate and test by western blot for different proteins eventually interacting with this proitein. The problems I have are these :

- At the last step (when the protein G sepharose beads are in the Laemmli 2X buffer), I cannot boil the solution of beads/antibody/proteins. If I do so, I obtain only two bands by western blot when I try to see the protein I precipitated with another antibody than the one used for IP (as a ctrl). These bands are approx 50 and 25 kDa, so Maybe these are the antibody light and heavy chains, maybe it's a degradation of the protein of interest. If I don't boil (I just let the mix at Room Temperature for 10 minutes then load) I get only one band corresponding to the correct protein (and not the other bands anymore !). Anyway the signal is not so strong (isn't IP supposed to concentrate the samples ?). I don't get neither any co-IP product whith this or any other protein (so at least the IP seems to be specific and precipitate only the protein of interest).

My conditions : Samples in RIPA buffer, frozen at -80°C then immunoprecipitated for 2 hours at 4°C (1 ug/µl samples, 2 µg antibody)

Thanks for your help !

Hello Raph,

I only have a very little experience with IP and I hope the experts can give you a complete answer to your problem.
What I would suggest is to try and chemically link your antibody to the beads so that you don't elute them in the final step. That way you should not expect the chains to come up on your gel. There is protocols to do that with dimethylpimelimidate for instance.
Are both of your antibodies against the protein of interest (the ne for IP and the one for detection) from the same species? If not, then you should not detect antibody chains on your blot. Can you detect that protein from lysates without IP? In that case, do you see the same problem with boiling your samples? Do you get one or two bands?

IP should concentrate you protein but if the IP isn't working well, then you can end up with actually less signal than you would detect from lysates. Incomplete elution can be a factor related to not boiling the beads. That could explain your low signal. The other thing could be that the incubation time with the lysate is not long enough. I usually do overnight incubation at 4C. But I'm not sure how much it matters in different cases of proteins.

Krisztina

Well, yes the bands you see are the heavy (50) and light (25) chains.
I would try using non-reducing elution buffer to elute your protein from antibodies and boil it. There are a couple of good kits out there for IP and Co-IP.
first I would try elution without reducing agent though.
I agree with what krisztina said about using different Ab for WB from different sp[ecies

The two antibodies used are from the same species indeed (mouse). The fact is that I don't understand that when I boil I obtain two bands (corresponding to the heavy and light chains) and when I do not boil I obtain only one band corresponding to my protein of interest. Boiling seems to make disappear my protein of interest (while making appear the bands corresponding to heavy and light chains !!).. I can detect the protein from lysates without IP with a very strong signal using both antibodies (the one used for IP and the one used for detection). I don't see any problem when boiling the samples. I get only only one strong band. When I precipitate, I still detect this band in the supernatant (the fraction which has not been precipitated).

Thanks for your answer,

How large is your protein of interest? Is it possible that the non-degraded antibody would run at the same size? If thats the case then your IP may not be working and you're just detecting antibody. Have you checked your lysate after the IP to see if your protein has been depleted?