co-imunoprecipitation

[D.K.]

GuilfoyleMR at cardiff.ac.uk (Mathew Guilfoyle) wrote:
>Hi, I'm making initial investigations as to whether
>coimmunoprecipitation would be a decent technique to try and detect
>receptor (mGluR) dimerization.  Can anyone direct we to a web page or
>written material on the various forms and protocols so I can a)
>evaluate whether it'll work and b) how much it is all likely to cost

This is really simple and one of the few cases where co-immunoprecipitation 
can deliver trustable results. Briefly: 

Constructs coding for two different tags, say Myc and HA. Co-transfect 
cells with two constructs. IP out one, stain Westerns for both, quantify:
Load 1, 1/2, 1/4, 1/8, 1/16, 1/32, 1/64 parts of total lysate. This is your 
calibration curve with 1 part being 100%. In the end of IP, resuspend 
beads in loading buffer so that final volume is equal to the volume of 
lysate taken for IP. This is your sample. Load several different volumes
of it to ensure you will get into sensible range of calibration. In parallel, 
run same amounts of control IP with same type of antibody but not directed
against either tag (processed in parallel with your sample, naturally). 
This is your negative control. From the two Westers you will be able to 
say that IP pulls down A% of total and co-IP B% of total. If A ~ B, then 
your protein most likely forms dimers. 

A guy in our lab did exactly this to show that published claim of 
dimerization is wrong. It was considered to be "negative result" and never 
published :-)

DK