Purification yields protein complex between recombinant and native p97

Dima Klenchin klenchin at REMOVE_TO_REPLY.facstaff.wisc.edu
Thu Sep 7 19:02:47 EST 2000

vbutel at ems.salk.edu (Virginia Butel) writes:
>Hi everybody,
>I have expressed a recombinant mouse protein, p97, in High Five insect
>cells using Gibco's baculovirus protocol. This is a fusion protein with
>Maltose binding protein at the N-terminal end. After purifying on amylose
>resin, I get very pure protein; only two bands show up by Coomassie
>staining. One is my recombinant fusion protein, MBP-p97, the other is the
>native p97 from the insect cells. Since p97 forms normally forms
>homohexamers, I assume the insect p97 is similar enough to complex with
>the overexpressed recombinant p97. 
>Has anyone come across this problem? 

I doubt anyone did. Everyone expresses p97 in E.coli. It is chaperone-like
protein which is perfectly soluble at 37C in E.coli and is expressed to huge
levels. Same for it's homologue, NSF. I would never go for a trouble and
expense of baculovirus when needing VCP/p97 (BV in my book is reserved
for proteins that are never soluble or never active in E.coli). Since p97
is present in almost any living cell, you problem was bound to happen.

>If so, I need ideas to get rid of the
>native protein. Amylose resin is not compatible with denaturing protocols
>and I would prefer not to have to try to renature to restore ATPase

OK, there are only few things to try. I am 99% positive that p97, like
NSF, can be dissociated into monomers while not denaturing and not
losing ATPase activity. All you need to do is to find such conditions
and repurify. 2.0 M NaCl and/or 50% ethylene glycol would be things
to try. Problem is your have MBP fusion and I don't know how resistant
its amylose binding to lowering ionic/hydrophobic interactions. His 
tag would work perfectly because IMAC does not rely on any of them. 
You might even get licky with very moderate GuHCl treatment, 0.5 M
or something.

Hope this helps, 

- Dima

P.S. I just got a p97 expression construct from Graham Warren at
Yale. Many other labs have same or similar. Maybe it is easier to ask. 

More information about the Methods mailing list