Hello Ruchi,

If you have not already done this, I would suggest making the following negative control experiment: carry out the whole process but without any antibodies coupled to your beads used for immunoprecipitation. This will indicate if your proteins 'Y' and 'Z' bind to the beads themselves, no matter what pull-down antibody you are using.

I have experienced myself, that my protein of interest also bound to the beads themselves, thus giving me false positive immunoblots all over. Also, you might want to try and use a different kind of beads or perhaps an isotype control from another manufacturer.

Best of luck.

Immnoblotting is quite sensitive, so it has a high probability of amplifying the presence of even minute amounts of your proteins non-specifically retained on immunoprecipitated material. You would have to adjust IP to get rid of them, for instance increasing blocking and washing stringency. In the case of blocking, I would strongly suggest to use skim milk powder as the blocking agent.

Dear Ruchi,

There are several things you can do to troubleshoot your IP. As suggested by Jeppe, your proteins may bind non-specifically to the beads you are using, and I agree that it is worth checking this. If this is the case, I would recommend the use of magnetic beads (e.g. Dynabeads™ Protein G from ThermoFisher). These also make wash steps significantly easier and more reproducible.

I would also suggest that you check if you are depleting your cell extract of the protein you are immunoprecipitating (protein X). It is possible that only a small fraction of your cellular protein X is bound to Y and Z, and by fully depleting the lysate you would increase the possibility of detecting this interaction.

It may be that the antibody you are using to IP protein X recognizes the same epitope on protein X that binds to Y and Z. This may mean that the antibody used for IP will compete away Y and Z. Do you have a tagged version of your protein X? If so, testing an IP against the tag would allow you to see if in this scenario you can see Y and Z co-precipitating.

I would also suggest that you try the reciprocal IPs - that is to say, IP Y and Z independently and probe for X,Z and X,Y respectively.

I hope this will be of some help,

Best of luck with your experiment,

Roland.

Hi Jeppe,

As you have suggested, I have already done the 'pre-clearing of lysates' wherein I incubated my lysate with the beads to get rid of proteins binding non specifically to the beads. Only after this step have i proceeded with immunoprecipitation. I could try changing beads or isotype control if possible.

Thank you for your suggestion

Are you sure the proteins detected are your target proteins and not Ab fragments i.e. light/heavy chain. If you carry out WB under reducing conditions you will see light chain and heavy chain. This is just a shot in the dark, but try carrying out WB under non reducing conditions. Also run Ab bound to beads only as control as well as lysate + beads (No Ab). I would also pre bind Ab to beads then block beads with BSA and wash. Then incubate with cleared lysate. This way all beads will be the same and more consistent.

Ian probably hit it on the head: you likely detect your IgG light or heavy chain (are your bands at ~25 or 50 kda?). If your IP Ab is derived from the same species as your primary WB Ab, it is even more likely to be case. Running the gel in non-reducing conditions will likely help diagnose, and if you're lucky, solve the problem; typically most (but rarely all) the signal will be shifted up to the IgG native size (~160) under non-denaturing conditions.

If you're unlucky (your band is predicted to run right at 25 or 50 and you can't run under non-reducing for whatever reason) other option is to switch your secondary that does not recognize the IgG, although that will require you to change your source of primary Ab to a different species (i.e. different from the IP Ab). Alternatively, there are some Abs on the market that only detect native Abs; consider using those.

Good luck!

Hi Ruchi,

you should definitely look into Ian's suggestion. Still, you must have been very unlucky to hit both Y and Z at the same MW as the heavy and light chain! So I imagine you possibly also have some non-specific binding of your Y and Z to the IgGs you use in the negative control. That would mean you should increase the stringency of your IP reaction. I see that you are using pretty standard conditions, so you have quite some margin of maneuver! To give you an idea:

1. Ionic strength. You are in standard 150 mM NaCl, you can increase stringency raising the salt concentration to 200, 250 or even 300 mM. Also in certain cases it helps using KCl, the intracellular salt.

2. Detergent. You are using TritonX-100, possibly 0.5 %... You can increase up to 2 %. But there are other detergents you can use in place or in combination with Triton, typically Na deoxycholate (DOC), up to 0.5 %, and, very stringent, 0.1 % SDS (you could essentially test your IP in the various formulations of RIPA buffer).

3. Dilution. While it is efficient to incubate the Ab in a small volume of protein extract (say 200 ul), it helps to clean the IP to increase the volume to 1 ml for the proteinA/G beads incubation (unless you follow Ian's other suggestion to first bind the Ab to the beads....).

4. Timing and temperature. You incubate the protein extract and the Ab O/N, but it might help to reduce this time and maybe increase a bit the temperature, for example 4 hrs at 15 °C... this alters the kinetics, besides the thermodynamics, of the binding equilibria, but it should affect more the non-specific ones compared to the specific ones.

Good luck.

Hey Ian, Thank you for your reply. The bands observed are at molecular weights higher than 100 kDa, so its not about antibody fragments. But I could look into other suggestions. @Ian R Wilkinson