What do you mean? What have you done so far? Or what do you plan to do?

Do you want to discuss experimental controls or physiological relevance or functional significance, for example.

You need to give us more information to start a discussion.

Dear Satyajeet,

if I got your questions, I would like to try to suggest you some ways to validate your IPs. In my opinion, you can use different positive and negative controls in order to be sure that your experiment was made in the right way.

1. As positive control you can use a total lysate of your cells in order to understand if the band in the IP lanes appears at the right molecular weight.

2. As negative control you can use different strategies. The most used is to load a lysate that was treated with non specific IgG  (and Protein A/G) instead of the antibody. Teorically, your IgG (mouse/rabbit/goat depend by your antibodies) should not bind any proteins in this sample then, when you  perform western blott analysis, you should not see any band at the level of the protein you are interested in. Another good control is to load only your lysis buffer and the antibody you used for the IP. This sample will tell you if there are any cross-reactions between the two antibodies used in the experiment. Finally, you can load a sample treated with just Protein A/G to see if these can interfere with the IP.

3. You can use other proteins already known to bind the molecules you are interested in. You pull down one of them and see if your enzymes are linked with these proteins too.

4. You can choose to divide your proteins in pieces to understand which part is important to form the complex (it is quite long).

5. You can overexpress your proteins tagged with GFP/GST etc and pull down this tags.

I hope these advices can be helpful for you  and I do hope I was enough clear. 

Good luck,

Alle

Hi Satyajeet, as Amanda suggested, you need to give more information. Are you trying to verify interactions of your co-immunoprecipitated proteins by Western blots, native PAGE, dot blots, gel filtration?

Plus, it would help if you have a negative control for your co-IP.

You need positive controls and negative controls.

Negative controls:

1. Do a IP in parallel with preimmune serum (of the same animal)

2. Do an IP in parallel with unrelated serum, or unrelated monoclonal antibody (if mouse)

3. Do an IP from cells that do not express that protein (KO par example)

Positive controls:

1. Check that you protein is present in its own IP, if possible with an antibody of a different species (doing a rabbit blot on a rabbit IP can give a lot of background

2. Check that the protein is present in the lysate (and absent in a KO lysate)

3. If an antibody that is known to work for IP with that protein, use that antibody as a positive control

I will suggest you to keep IgG as a negative control, and also run actin during western blot. If your IP is successful, you will see bands for Actin and no interaction with IgG normal. Keep you lysate  prior to pre-clear step(if using kit) as a input which will show you whether your protein is expressed or not. If you are raising Ab against a tag like flag, then you can use both IgG and flag as a control.

All the above suggestions are great. Again, as has already been stated, more information would be helpful to specifically address your question. Having said that, here are a couple of things you can do. First, reverse the Co-IP. In other words, if you are trying to establish that protein X binds to protein Y, bind X to the beads, and blot for Y. If that is successful, do just the reverse; bind Y and blot for X. Another thing that can be helpful is to increase the stringency of your washes to see if you can disrupt the association.

Hi Satyajeet,

The question you are asking has to do with the specificity of binding. Hence, this issue pertains of 'off-rate or 'on-rate constants.' One of the above suggestions talks about having your purified protein. Of course, one major question you need to address is the potential existence of a shared epitope for your IgG with the protein of interest and the protein which is co-precipitating.

There is your first experiment, determining the specificity of your IgG. One of your colleagues made an excellent suggestion of using albumin to block nonspecific binding sites...always helpful.

A question for you, is your IgG monoclonal or polyclonal? Is it a blocking or non-blocking relative to the function of your protein of interest? This could be probed in a functional context once you've determined the next question.

I'm assuming you have a monovalent IgG? Here is my suggestion to you. My first step in my experiment, assuming a monoclonal Ab, would be to prepare an Fab fragment of your antibody and bind that to beads. Once you have removed any non-specific binding of your bead-Fab complex, run a purified preparation of your main protein over the column. Based on the binding constant of your protein...which should be very high, you should be able to determine the total binding value for your column. Depending on your analytical sensitivity, you don't need a very big column to do this. I'm sure you know what options would be available. If not, we can discuss that.

Since you know total binding, you do two things: First block your antibody with an epitope to your protein and demonstrate no binding...all of your protein comes out in the primary eluate.

Clear your gel, add you protein then add a series of titrations of your co-preciptated protein. For each, elute the column and run your gel to determine % binding at each concentration of your co-precipitated protein. Of course, one obvious experiment is to run you co-precipitated protein over your column and demonstrate that it doesn't bind without it's immuno-precipitating partner.

Based on the concentration series and knowing that your column is run under equilibrium conditions, you can determine the 1/2 binding constant for your immuno-preciptant complex. 

I'm wondering how you might do the negative control since you don't know the epitope for your second protein and it's bindnig. However, at this point, it would seem to me that you have convincingly demonstrated that your immunoprecipitation is, in fact, specific and titratable. This demonstrates, beyond any shadow of doubt that the interaction is real.

The next experiment you can do is check your system for off-rates, if you will. You would saturate your preparation with both proteins in the immunoprecip complex. This would be done at a concentration such that a significant dilution would allow you to see your unbound co-precipitant. Let's assume, for the moment, that your co-precipitating macromolecule is radiolabelled or fluorescent. After a large dilution, you run your complex over your column, you can use the "spin column" approach to do this so you capture all of the void volume. The void volume is assayed at to, t1, t2, etc to determine how much of your co-precipitant remains unbound (this is better than a bead spin approach). Having gone through this process, you can determine bmax, t1./2 and bmin. 

You've now determined everything you'd need to know about this immuno-coprecipitant except the 'why?' part. The example would be the binding of factor X to K+Channels. Factor X binding to these ion channels is directly correlated to down-syndrome, if I'm not mistaken. What factor X is doing in terms of the ion channel function is still a question under study I believe.

Well, Satyajeet, this is the best I can provide and hope it has some bearing on your experiment and interpretation. Of all of the suggestions you've received, one of the important ones has to do with finding that "negative" control so you know whether your binding is a 'real' event or just some effect of an uncontrolled variable such as a polyvalent Ab or common epitopes shared by both proteins in your complex. That's the important piece of data you need to procure.

I pray your experiments provide both productive and useful results.

Sincerely,

David W. Chester, Ph.D.

BiFC is a good method for confirmation of protein-protein interactions.

All the above suggestions are great.

HI all 

Suggestion provided all are great,

It will be helpful