5 Months of Immunoprecipitation Woes - Disassembly of Protein-G Sepharose

Melissa Greeve gliftor at hotmail.com
Tue Sep 4 19:53:31 EST 2001

Hi Ian, thanks for your reply...

> >Has anyone else encountered protein g sepharose dissociating
> >throughout the IP process??? As you can imagine, this results in a
> >huge band at 20kD, representing protein G, that binds and fluoresces
>This should not happen.  I've never seen it, in hundreds (thousands?
>haven't kept track) of IPs.  I've also never used Sigma protein A/G,
and I
>suspect that's related.  Try a different source.  It may be more 
>expensive, but it can't possibly be mroe expensive than 5 months of

When I initially started the IPs, we had some protein-A agarose from
Biorad knocking around, and we used it just to start off the protocol.
But sure enough, there were big bands (this time smeared) at 30kD, I
think, this time. Since this batch wasn't very new, we looked around
at what other people were using, and decided to order some protein G
sepharose (since this is meant to bind to a wider selection of
antibodies more strongly) from Sigma (where other people had bought
stocks from). This bottle came to us semi-evaporated (d'oh!), but the
technicians assured us it was ok if we redissolve the beads in some
more ethanol. That I did, but still had the problems. So eventually
they replaced it for us, and with a newer batch number, which I am
using now - with the same problem.

> For some reason, I've always found protein G less reliable (more 
> background, worse binding) than protein A.  

FYI, so far the blots are quite clean - little background.

> If you have lots of the primary antibody, consider coupling it directly to 
> sepharose.  That way you'll get rid of not only the protein G band, but 
> the antibody bands as well.  

This has been suggested to me by another reply to my email as well...
Might be worth looking in to.

> Those, basically, suck.  

This I know!!

> They're much less 
> reliable than straight IPs or straight westerns: you have all the errors 
> of each procedure, and they amplify each other.  

I think that is the biggest problem. There are likely to be many
factors involved. The westerns are running perfectly, though, so I
guess that's something.

On the side, I did an IP yesterday and finally found some protein
bands, but these turned out to be fainter than those in the basically
more dilute whole cell lysate!! The saga continues, and this time it
involves protein binding.

Thanks again,

More information about the Proteins mailing list