Hi Jaya, as usual in biology, it is the specific problem you are addressing that determines the protocol to use. RIPA buffer (which has two formulations, with 0.1-0.2% SDS or with 0.25% DOC, with decreasing stringency) is a very stringent buffer, meaning it is suited for strong and normally hydrophobic-based protein-protein interactions. Then come buffers with TritonX-100, from 0.5 to 2%, and finally the less stringent buffers which use NP-40. Same story for the ion concentration: physiological concentration in the cell is close to 150 mM K+, but in practice replacing K+ with Na+ has normally no or very little effect, while lowering or increasing the concentration alters the stringency of your environment, since protein interactions in higher ionic strength tend to fall apart. But do not forget that in cell compartments the ions concentrations can be different, so for your specific interaction you might have to find the best in vitro conditions. Last point is pH: again in the cytoplasm it is in average around 7.2, but different compartments can have different pH. Normally Abs bind better at pH 8-8.2, therefore if this does not affect your interaction using this slightly basic pH might help you get a more efficient and clean IP.

Washes are not necessarily carried on with the same stringency as the IP. You can increase it or decrease it, since the washing time is short compared to the IP incubation time and kinetics count in specific and unspecific interactions, the latter being normally faster.

My suggestion is to run in parallel your IP in a standard buffer such as Nicolas suggests and in a RIPA buffer (the one with DOC... in a first attempt I also would leave out SDS... then if you have something really dirty, you can try the SDS).  Do your washes in a standard isotonic  buffer with 1% Triton. Work always at 4 °C or on ice.

Run a standard SDS gel (non denaturing gels are used in special applications where you have to determine sizes of complexes or you are interested in size shifts due to some in vitro  manipulation), keeping in mind that you will have two prominent bands at 20-25 and ca. 55 kDa coming from your primary Ab (but often expanding over a larger range of kDa!), mostly when you will probe your WB with a primary of the same species as the Ab you used for the IP, and check the efficiency of your IP first (do you get the protein you are immunoprecipitating? and how much) comparing 1/10th-1/5th of the IP against 10-20 ug of your IP supernatant (from the proportions you can calculate how much of your protein you are IPing), and then check a co-IP partner to be sure that your conditions are not disrupting the interaction you are interested in. In parallel you should run a negative control without your Ab for IP, or even better, using the Ab on a knock-out or knock-down protein extract. Then you will be able to choose the IP conditions that gives you the best results in terms of IP and co-IP efficiency and cleanliness, or you will have some starting point to try new conditions.

Good luck

I agree about the detergents - I always used a buffer with NP-40 instead of SDS, as Nicolas suggested. I also limited my washes relative to a straight IP, since you can lose an interaction if it's diluted too far.

As for which gel to run, it really depends on how you're planning to detect an interaction. If you have an antibody for the interaction partner, and will detect with Western, you run standard SDS-PAGE. Similarly, if you are using enough protein that you anticipate visualizing by Coomassie staining, standard SDS-PAGE.  If you are trying to detect by Western with your primary protein, and are looking for a shift in migration, then you would need to run a native gel - but this might be very difficult if the antibody stays associated (it will shift the migration of the protein, and if it is the same species that you're using for your IP, the secondary antibody will recognize it, which will make it look like an interaction even if one isn't taking place).

If none of the above scenarios is realistic, you can attempt to cross-link your proteins, before IP, using something like DSP or DTSSP, and then immunoblot looking for a migration shift of your protein. Just remember if you do this to leave DTT or beta-ME out of your sample buffer, as it will reverse the crosslink.

2x RIPA for 1 million cells

I also agree in avoiding RIPA for Co-IP, due to the reasons they have mentioned before. However, I have to say I have used RIPA for IP (with 5 whasing steps max), and analyzed with standard SDS-PAGE, and there was no problem (with 3 different cell lines and proteins).

Hi Jaya, as usual in biology, it is the specific problem you are addressing that determines the protocol to use. RIPA buffer (which has two formulations, with 0.1-0.2% SDS or with 0.25% DOC, with decreasing stringency) is a very stringent buffer, meaning it is suited for strong and normally hydrophobic-based protein-protein interactions. Then come buffers with TritonX-100, from 0.5 to 2%, and finally the less stringent buffers which use NP-40. Same story for the ion concentration: physiological concentration in the cell is close to 150 mM K+, but in practice replacing K+ with Na+ has normally no or very little effect, while lowering or increasing the concentration alters the stringency of your environment, since protein interactions in higher ionic strength tend to fall apart. But do not forget that in cell compartments the ions concentrations can be different, so for your specific interaction you might have to find the best in vitro conditions. Last point is pH: again in the cytoplasm it is in average around 7.2, but different compartments can have different pH. Normally Abs bind better at pH 8-8.2, therefore if this does not affect your interaction using this slightly basic pH might help you get a more efficient and clean IP.

Washes are not necessarily carried on with the same stringency as the IP. You can increase it or decrease it, since the washing time is short compared to the IP incubation time and kinetics count in specific and unspecific interactions, the latter being normally faster.

My suggestion is to run in parallel your IP in a standard buffer such as Nicolas suggests and in a RIPA buffer (the one with DOC... in a first attempt I also would leave out SDS... then if you have something really dirty, you can try the SDS).  Do your washes in a standard isotonic  buffer with 1% Triton. Work always at 4 °C or on ice.

Run a standard SDS gel (non denaturing gels are used in special applications where you have to determine sizes of complexes or you are interested in size shifts due to some in vitro  manipulation), keeping in mind that you will have two prominent bands at 20-25 and ca. 55 kDa coming from your primary Ab (but often expanding over a larger range of kDa!), mostly when you will probe your WB with a primary of the same species as the Ab you used for the IP, and check the efficiency of your IP first (do you get the protein you are immunoprecipitating? and how much) comparing 1/10th-1/5th of the IP against 10-20 ug of your IP supernatant (from the proportions you can calculate how much of your protein you are IPing), and then check a co-IP partner to be sure that your conditions are not disrupting the interaction you are interested in. In parallel you should run a negative control without your Ab for IP, or even better, using the Ab on a knock-out or knock-down protein extract. Then you will be able to choose the IP conditions that gives you the best results in terms of IP and co-IP efficiency and cleanliness, or you will have some starting point to try new conditions.

Good luck

Thank you all for your suggestions.

Please read some earlier articles on Co-IP. Low non ionic detergent works great. They expose the interacting domain of the interactome and create accessibility. In native state , under physiology multiple proteins interact and increase accessibility. In vitro non ionic detergent like triton or np 40 but not SDS favors that.