Help trouble shooting expression and folding in bacterial system

Dr. Ted noone at
Fri Sep 8 01:32:32 EST 1995

In article <42i80a$76a at>, Holers Lab
<holers-lab at> wrote:

> <BASE HREF=3D"file:///i:\joel\scr.htm">
> From: guthridj at (Joel Guthridge, Ph.D.)
>    We've tried:   Lower induction temp.
>                           Lower IPTG levels.
>    Possible Altern. (We need advise here)
>              -do expression in the presence of Glutatione, etc.   =
>              -? better periplasmic fractionation (how do others do this?)
>              -do expression in presence or absence of Ca2+  (FLAG req. Ca2+=
> )   =
>              -change epitope tag system ????
>              -change expression system (Pichia??) =
>              -reduce and refold protein in vitro  (methods ??)

   Your problem is a common one, if that makes you feel better. It is very
easy to get your typical disulfide bonded eukaryotic protein into
insoluble inclusion bodies by overexpressing it in E. coli. Note that
these inclusion bodies can be formed in the cytoplasm OR the periplasm.
You have probably checked to see if the signal peptide is cleaved in the
solubilized monomers from the BL21, so you can see if the processing/
transport is working. Typical periplasmic fractionation is by osmotic
shock, although this is awkward to scale up. Also note that the DH5alpha
can be grown to about a kilogram of wet cell mass per 10L fermentation
with the correct feed protocol, so low level expression is not the end.
The stable isotope label/minimal media will be tough, though. As for the
changes you mention:
   glutathione- E. coli is pretty particular about the redox potential it
maintains inside, this will most likely not work
   periplasmic fractionation-Check the lit. to see if your up to speed,
but this will not make the protein behave, simply make it more pure.
   Ca2+, epitope tag- your problem is the overexpression, and the
cysteines so the tag is probably not going to help, you could omit it to
see if it helps, but the purification will be harder. Invitrogen(Pichia)
markets a thioredoxin fusion system and NEB has the maltose binding
protein, these may help but probably not.
   Pichia- works great for some things but its only advantage over Sacc.
cerve is the huge cell mass, I'd try reg. yeast first, then there is the
odd glycosylation and stable isotope label problems... best to stick with
E. coli.
   Renature/reduce the protein- YES TRY THIS. Inclusion bodies can be a
shortcut to purification, they pellet in the centrifuge almost pure, and
there are currently several large articles discussing
renaturation/refolding, some using disulfide isomerase. You have only 4
disulfides= two possible bond combination; one right one wrong, worth a
shot. Check medline, Bio/technlogy etc. I've seen this work with over 5
S-S bridges.These papers should be easy to find. Methods in Enzymology?..

> Lower IPTG levels.

low level expression only occurs with the TAC promoter at IPTG levels
below 1µM. The promoter is leaky though, sometimes uninduced cells give
more active protein than induced cells...check it out. Good Luck.


Ted Michelini
Institute of Molecular Biology
University of Oregon
tedm at

More information about the Proteins mailing list