Overexpression of Eukaryotic proteins in E. coli

mcp at canctr.mc.duke.edu mcp at canctr.mc.duke.edu
Mon Nov 2 11:38:10 EST 1992

Dear Colleagues,

	I am currently trying to establish a system for overexpressing my favorite gene in E. coli.  I have a plasmid which contains MFG as a fusion with
the maltose binding protein (MBP - using the pMal-p2 vector from NEBL) which
is under the control of the Lac promoter.  I get expression at first
after transforming the plasmid into the host, but as I subculture the
little bugs, they decide to quit making my protein.  Now, I am
not streaking these guys on media with inducer, so they shouldn't
be unhappy about a foreign protein.  Also, upon induction, the cells
continue to grow even if they are making MFP, so the product isn't
lethal.  I have tried to express MFG in five different E. coli
lines: HB101, DH5a, BL21 (Novagen), HMS174 (Novagen) and Nova Blue.
No line was better at producing MFP.  Also, compared to the vector
with no insert, ie. just the MBP, the expression of my favorite
fusion is pitiful. I have tried a billion experiments with various
media and various growth conditions - temperature, amount of
inducer added, etc. and nothing seems to help. Plus, while I am
doing these experiments, the colonies that are on the plates in
the 'fridge are conspiring against me to refuse to express the
protein.  Does anyone have an idea as to:

1. Why should the amount of expression decline in cells grown
without induction. (The vector has a copy of the lacI gene on
it as well, so that leaky expression shouldn't be a problem -
and I can't see any expression in uninduced cells on a PAGE)

2.  Why can't I get this protein to be overexpressed at a 
reasonable level, and what can I do to remedy this?  I have
heard that Stratagene has an E. coli line "TOPP" that costs
a million and a half (:-%) dollars, but is supposed to
allow expression of tempermental proteins - has anyone
used this ? Is it worth $130 for a stab of E. coli?

3.  If you can't help, can you find me a new job in the future?


Laura J. Weislo
Weisl001 at dukemc.bitnet

More information about the Methods mailing list