Immunoprecip gurus?

Ian A. York iayork at panix.com
Fri Nov 19 18:02:29 EST 1999


In article <811qdj$fbe$1 at news.doit.wisc.edu>,
Dima Klenchin <klenchin at facstaff.REMOVE_TO_REPLY.wisc.edu> wrote:
>:Ok, let's say protein A binds to protein B which binds to protein C.  If
>:I pull down with an antibody to protein C (from cell lysates), should I
>:expect to be able to pull down protein A with it and probe for protein A
>:on a Western, or is it too much to hope for that everthing will stay
>
>My experience with IP is that if one tries really hard, it is possible
>to co-IP anything with anything. Particularly when a criterion for 
>co-IP is a sensitive western. And that is the major problem with
>this overused undervalidated technique.

This is absolutely true, but there's nothing intrinsically wrong with the
technique.  It's just that you ABSOLUTELY MUST do the appropriate 
controls.  A student working for me is doing this right now, and I
insisted on seeing multiple different antibodies that do and do not bind
the protein; the same antibodies in cells that don't express the protein;
and much more.  We started out seeing four proteins that apparently
specifically co-precipitated, and only by doing the full series of
controls were we able to determine that three of the four are
non-specific.  

Other people have offered good sugegstions, like playing with different
detergents (in order of increasing stringency, I use digitonin, CHAPS,
NP40, NP40/DOC, NP40/DOC/SDS).  You can also play with salt concentration
(low salt is often less stringent, but not always; it depends on the
nature of the interaction).  But don't be TOO gentle, or you'll almost
certainly get non-specific interactions.

Ian
-- 
    Ian York   (iayork at panix.com)  <http://www.panix.com/~iayork/>
    "-but as he was a York, I am rather inclined to suppose him a
     very respectable Man." -Jane Austen, The History of England




More information about the Methods mailing list