:: Search forums :: Top Users :: Reagent
Position: Discussion forums > Bioscience biotechniques discussion forums > Proteomics and protein biochemistry forum
|
co-IP PROBLEM
Posted by: mb80 (IP Hidden, Junior member, 12)
Date: August 3, 2005 01:01PM
I'm having trouble performing coIPs--
Have been trying to determine whether protein A interacts with protein B. Previously it has been shown that protein A interacts with another protein, C. I cannot reproduce the A-C interaction, and hence am unable to determine whether my negative data for A-B is real or just a problem with the assay-- I need to get a positive control to work before I can accept a negative result. I also have not been able to reproduce a separate interaction, D-E. While it is possible the previously published interactions were not robust results and the authors themselves could not reproduce them reliably, it seems unlikely that these two independent binding interactions in the literature would fall into this category since the data themselves in the figures look good (?) I am transfecting 2ug of each plasmid into 293T cells; my transfections work, as I am able to detect the overexpressed tagged proteins by western blot in whole-cell lysates; when I IP for protein A, run out the IP on western and blot for protein A, I see a strong band; so I apparently can IP protein A successfully, but I just cant detect ANY binding partners (B or C). I use 300ug protein lysate per IP. I incubate lysate with both beads and primary antibody simultaneously at 4 degrees with rotation O/N. I'm wondering if I'm encountering a simple technical problem with my technique, or with my buffers; I used standard RIPA buffer (see below), while other groups used less harsh lysis buffers...and included magnesium and/or zinc in some cases. BUT, ONE study has actually already demonstrated the A-B interaction in cos7 cells, and they used standard RIPA buffer Thanks very much for any HELP (!!) Here are the previously published data and buffers: **MY LYSIS BUFFER** RIPA BUFFER 50mL 500 mL 150 mM NaCl 1.5 mL 5M NaCl 15 mL 5M NaCl 1% NP-40 0.5 mL NP-40 5.0 mL NP-40 0.5% deoxycholate (DOC) 0.25g DOC 2.5g DOC 0.1% SDS 0.5 mL 10% SDS 5.0 mL 10% SDS 50 mM Tris pH 7.5 1.25 mL 2M Tris-HCl pH7.5 12.5 mL 2M Tris-HCl ddH20 46.25 mL 462.5 mL water PROTEASE INHIBITORS TABLET NaF 0.5 mL/50mL RIPA NaOrthovanadate 0.5mL/50mL RIPA 1. protein A-protein B (PUBLISHED) cos7 cells 500ug lysate per IP add beads POST-Ab RIPA 140mM NaCl 1% NP-40 0.5% DOC 0.l% SDS 20mM sodium phosphate pH 7.5 4mM NaF, 10ug/mL aprotinin,10ug/mL leupeptin, 10ug/mL PMSF, 0.1mM NaOrtho, 10ug/mL glyercol phosphate 2. protein A-protein C (PUBLISHED) 293T cells transfect 6ug DNA of each plasmid lyse 20 minutes on ice with 1 million cells/100ul extraction buffer 150 mM NaCl 0.5% NP-40 50 mM Tris HCl pH 7.6 1 mM MgCl2 50 uM ZnSO4 1X EDTA–free protease inhibitor cocktail [Roche] 10 mM NaF, 1 mM Na3VO4 3. protein D-protein E (PUBLISHED) 293T cells lysis buffer 100 mM KCl 0.1% NP-40 20 mM Tris-HCl [pH 8.0] 5 mM MgCl2 10% glycerol supplemented with freshly prepared protease inhibitors.
Re: co-IP PROBLEM
Posted by: mitolab (IP Hidden, Senior member, 89)
Date: August 5, 2005 03:29PM
>> but I just cant detect ANY binding partners (B or C).
You mean can NOT detect B or C? Appropriate control is essential for interpretation IP results. You should also include whole lysate to demonstrate your antibody against B and C can actually detect the specific band. Lysis buffer is critical in terms of its possible interruption of complex stability. You can try different mild lysis buffer and see if it makes any different. You may also explore if the tagged protein interfere with the complex assembly by using vectors encoding protein tagged in another direction or a untagged protein. You can try to increase protein lysate to 1 mg for IP.
Re: co-IP PROBLEM
Posted by: EANON (IP Hidden, Junior member, 19)
Date: August 7, 2005 12:54PM
Have you ever successfully IP-ed two interacting proteins together or is this just a temporary anomaly? There very likely is a problem with your solutions/technique rather than the interaction itself. I recommend you remake all buffers. Try decreasing the detergent concentration in the lysis buffer to .5% vs 1%.
Since you can detect your transfected proteins in the lysate, there very likely is a problem with your handling of the co-IP itself rather than another stage. What buffer are you using to wash the beads with? Perhaps you are losing bound protein B or C during the wash step. You may want to describe your handling of the bead-antibody complexes after the overnight incubation. Also, unless your beads and antibody are conjugated, you should incubate with antibody first for one hour and then add beads for the overnight rotation, although I don't believe this is the cause of your problem.
Re: co-IP PROBLEM
Posted by: mb80 (IP Hidden, Junior member, 12)
Date: August 8, 2005 02:52PM
Thanks very much for the responses--
To answer your question- No, I have never done coIPs before. So yeah, seems likely I'm doing SOMETHING that is disrupting the interactions between protein A and proteins B,C. (technique or buffers) After the overnight incubation with primary antibody and beads, I spin down the beads at 1250rpm in the cold. I aspirate the supernatant, and wash 4 times with lysis buffer (previously, this was RIPA buffer: 1% NP-40 and 0.1% SDS). For each wash step, I typically just added 500-1000ul of lysis buffer, inverted the tubes a few times, and spun down the beads, before repeating. Occasionally I would incubate in the cold room on a rocker for 5 minutes during the first wash. After the four washes, I add SDS loading buffer, boil, and transfer the supernatant to clean tubes. Does this sound like it could be a problem? What wash buffer do you recommend? Since my original post, I also tried a new lysis buffer with 0.1% NP-40 (very mild)... 100 mM KCl 0.1% NP-40 20 mM Tris-HCl [pH 8.0] 5 mM MgCl2 10% glycerol Unfortunately, I still am not able to reproduce the interaction between protein A-protein B, or a separate interaction, protein D-protein E. My ability to interpret my data for D-E is complicated by the fact that protein E runs almost at the same mobility as heavy chain... My antibody for protein D works better for IPs, so I use that to IP D and blot for protein E...it's difficult to know if the band I see is really protein E in the anti-D IP or whether it's simply heavy chain. Anyway, this is just an annoying collateral problem... I've been banging away at this problem for weeks, and am getting desperate to get SOMETHING to work.. I guess I should just go find a really robust known protein-protein interaction and focus on getting that to work in 293T cells. but of course you need reagents, expression vectors+antibodies.. Thanks for any help!
Re: co-IP PROBLEM
Posted by: EANON (IP Hidden, Junior member, 19)
Date: August 9, 2005 10:32AM
Your protocol seems OK with me. After boiling and centrifuging, why do you transfer the supernatant to clean tubes instead of loading directly?
To answer your question, I don't use RIPA buffer myself, I just use PBS with protease inhibitor cocktail and some detergent (.5-1% triton) for lysis. The wash buffer I use is the same as lysis - without protease inhibitors. Perhaps your antibody concentration against your protein's binding partners are not high enough. It is also possible that the RIPA buffer is disrupting your protein interaction (this is the trade-off for its low background). In any case, don't be discouraged, it is a foregone conclusion that you will have your co-IPs working.
Re: co-IP PROBLEM
Posted by: Jurek (IP Hidden, Unregistered user, )
Date: August 10, 2005 03:35PM
You probably should not use SDS in lysis buffer, you will denaturate your proteins and they will not interact. Try also to cut out Na-deoxycholate or maybe change NP-40 to Triton X100.
Re: co-IP PROBLEM
Posted by: mb80 (IP Hidden, Junior member, 12)
Date: August 16, 2005 04:24PM
Thanks again for the input--
I finally got some coIPs to work--ended up using different interactions than what I was using originally for a "positive control"... also changed the lysis buffer to 0.1% NP-40 and no SDS, and washed the beads post-IP only 2X instead of 4X, but i'm not sure if those changes were really what caused it to work, or whether i just picked more robust interactions to test as positive controls...in any event, now i can be more confident in my negative result for the primary interaction under investigation, although of course the absence of evidence is not evidence of absence
Re: co-IP PROBLEM
Posted by: Chenoeh (IP Hidden, New member, 5)
Date: August 18, 2005 10:40AM
I did several co-IPS a couple of years ago that took me months to get working. I was transfecting tagged proteins into HeLa cells. I eventually realised that while I could easily see the individual proteins, I could only see the interaction when the HeLa cells were transfected at a relatively low confluency, about 50-60%, while the transfection method suggested about 80-90%. I also found that using the gentler (and cheaper!) calcium chloride transfection method seemed to work better then superfect.
I've no idea why this might be, but it might be worth your while trying transfections at different cell confluencies. If you're still having problems with the band you're looking at overlapping the heavy chain from the antibody used to IP, you could try using protein A/G bound to HRP rather then the secondary antibody. The protein should bind to the three dimension structure of the Fc region of the primary antibody, and not the primary structure of the denatured antibody on the membrane, so you should see only one band. I had mixed results with this, but it might be worth a try. Edited 1 times. Last edit at 08/18/05 10:46AM by Chenoeh.
Re: co-IP PROBLEM
Posted by: Jialing Huang (IP Hidden, Unregistered user, )
Date: November 13, 2005 09:12AM
I am doing CO-IP with negative control problem-normal pre-immunization serum precipitate protein of interest! I repeated it several times with precipitating antibodies from different species and corresponding normal serum, but I saw the bands in the negative control lane.
I can not figure out what is the problem because I got confortable results while I doing CO-IP with the same negative control serum half year ago. Any help is highly appreciated. Jialing Huang
|
We are moving ... Please post to our new community forums
