Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

Immunoprecipitation Troubleshooting


  • Please log in to reply
7 replies to this topic

#1 Priya914

Priya914

    member

  • Active Members
  • Pip
  • 18 posts
-1
Neutral

Posted 02 November 2010 - 12:06 PM

Hi all,
I need some help.We assume there is an interaction between our two proteins of interest A and B. When I pull down Protein A and do IP, protein B is also pulled down and it is an absolutely clear result with no background in control IgG lane.
However, if I do the other way round i.e, pulldown Protein B and blot for A, I dont see anything. Does anyone have any clue what may be going on ?
My boss wants to see an interaction in both the ways to be really convinced of the result.
I really appreciate your help and time.
Thanks

#2 vered

vered

    member

  • Active Members
  • Pip
  • 8 posts
0
Neutral

Posted 02 November 2010 - 12:28 PM

hello
are your proteins tagged or do you use an antibody directly against your protein? that might be one problem that your antibody for detection is not good enough,
do you see both protein in the whole cell lysate (before the IP), if not than you might have an expression problem of this protein A. in other words, when you IP with protein B which is highly expressed you pick up on protein A, but if protein A is poorly expressed it might not be enough to IP protein B.
- do you use for the IP with protein A a direct antibody? it just might be that the antibody you are using is masking the binding site for protein B (this might be especially if your antibody is monoclonal)

hope it helps

V
[
quote name='Priya914' timestamp='1288728394' post='91198']
Hi all,
I need some help.We assume there is an interaction between our two proteins of interest A and B. When I pull down Protein A and do IP, protein B is also pulled down and it is an absolutely clear result with no background in control IgG lane.
However, if I do the other way round i.e, pulldown Protein B and blot for A, I dont see anything. Does anyone have any clue what may be going on ?
My boss wants to see an interaction in both the ways to be really convinced of the result.
I really appreciate your help and time.
Thanks
[/quote]

#3 Priya914

Priya914

    member

  • Active Members
  • Pip
  • 18 posts
-1
Neutral

Posted 02 November 2010 - 12:57 PM

Hi
Thanks so much for your reply. Both proteins A and B are tagged. A is flag tagged and B is HA tagged. In my lysates both proteins are expressed so the transfection has worked out fine.
Now when I pull down with Flag( mouse monoclonal) for Protein A, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and I see an interaction B) .

But when I pull down with HA (mouse monoclonal) for Protein B, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and I dont see an interaction :o .So I can successfully pulldown Protein B but no Protein A in the blot.

Then I tried pulling down with direct antibody (mouse monoclonal) for Protein B, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and same result. I dont see an interaction :( . So I can successfully pulldown Protein B but no Protein A in the blot.

Should I try pulling down Protein B with a polyclonal antibody ?
Do you think I may be getting a false positive result when I am pulling down with the Flag antibody ?

Thanks so much for your help and time :)

#4 BioBM

BioBM

    member

  • Active Members
  • Pip
  • 15 posts
3
Neutral

Posted 02 November 2010 - 01:56 PM

Priya,

If you expect a pulldown and don't get one, I always recommend pulling down using a polyclonal antibody (when available, of course) and unmodified proteins. A false positive is always a possibility as well. Tagging the proteins can change their conformation.

One of the things that always bothers the heck out of me regarding IP is that there's no single ideal way to perform one - any of them can potentially introduce problems.

Cheers,
-Carlton
Posted Image
BioBM™ Consulting - www.biobm.com - twitter: @BioBM

#5 rkay447

rkay447

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 177 posts
20
Excellent

Posted 02 November 2010 - 01:57 PM

There are times when you just can't get the reciprocal IP to work. For various reasons, you can only detect the interaction in one direction. Perhaps the IP with the HA antibody disrupts the interaction and it's worth trying with the polyclonal. Best HA antibody I've found for IP is the ab9110 from Abcam. Be sure you are doing the proper controls to ensure specificity. Are you doing the anti-flag IP from lysates only expressing your HA-tagged protein (in other words, there is no flag-tagged protein in the lysate) to make sure the flag antibody isn't non-specifically IPing the HA-tagged protein? This is your most important control and is essential. If you don't see the HA-tagged protein in the Flag IP when no flag-tagged protein is present, then your IP is specific and the interaction is very likely true. If your advisor is absolutely dead-set on you doing the reciprocal IP, you may need to move the HA tag to the other terminus or swap the tags on the two proteins so this time you have flag-tagged protein B and HA-tagged protein A.

#6 Priya914

Priya914

    member

  • Active Members
  • Pip
  • 18 posts
-1
Neutral

Posted 04 November 2010 - 11:53 AM

@ Carlton
Thanks so much. I will try pulling down using Polyclonal Ab to see if it works :)

@ Veteran
Thanks a lot. Thank you for letting me know about the control by transfecting the cells only with HA tagged protein and doing an IP. I havent yet done it but I completely agree that if I dont get an interaction there, it should be very specific. I am really glad you brought it up. I guess my mentor should be happy with that and hopefully wont ask me to show the interaction both the ways. Thanks again :)

#7 BioMiha

BioMiha

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 235 posts
9
Neutral

Posted 06 November 2010 - 12:41 PM

You could also use a label transfer agent or cross linker to see if the interaction is the same both ways.
http://www.piercenet...0D-78A121076770

Edited by BioMiha, 06 November 2010 - 12:42 PM.


#8 Jason S LEUNG

Jason S LEUNG

    member

  • Active Members
  • Pip
  • 6 posts
0
Neutral

Posted 29 November 2010 - 07:45 PM

Hi
Thanks so much for your reply. Both proteins A and B are tagged. A is flag tagged and B is HA tagged. In my lysates both proteins are expressed so the transfection has worked out fine.
Now when I pull down with Flag( mouse monoclonal) for Protein A, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and I see an interaction B) .

But when I pull down with HA (mouse monoclonal) for Protein B, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and I dont see an interaction :o .So I can successfully pulldown Protein B but no Protein A in the blot.

Then I tried pulling down with direct antibody (mouse monoclonal) for Protein B, then I used rabbit antibodies to blot for Protein A and B (DIRECT ANTIBODIES NO TAG) and same result. I dont see an interaction :( . So I can successfully pulldown Protein B but no Protein A in the blot.

Should I try pulling down Protein B with a polyclonal antibody ?
Do you think I may be getting a false positive result when I am pulling down with the Flag antibody ?

Thanks so much for your help and time :)


As some folks had said, it's very important to do the protein A or protein B alone with your flag-antibody. I have a bad experience: when I use invirtogen flag(M2) agarose, it can pull down any HA-tag protein(single protein alone control still huv HA-bands!! :angry: I huv changed different HA-protein/using new batches of agarose and the result is reproducible) non-specificially in any 150mM NaCl buffer. I'm still working on fixing this problem.

Some old threads here said you can increase the amount of salt in the (washing?) buffer to get rid of those non-specific bands :lol:

Edited by Jason S LEUNG, 29 November 2010 - 07:48 PM.





Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.