Plant protein expression/Ecoli codon preference

CSMITH at CRCVMS.UNL.EDU CSMITH at CRCVMS.UNL.EDU
Fri Aug 6 23:39:00 EST 1993


Hello fellow biochemists/molecular biologists,

     YOUR HELP IS DEARLY NEEDED ... READ ON ... P-L-E-A-S-E 

     I am trying to express the maize gene (3 kb cDNA) for
pyruvate,orthophosphate dikinase (a mouthfull, so lets use PPDK)
in the expression vectors pTrc99a [Pharmacia] and pET11a
[Novagen], but am having very little luck. I'd like to get some
insights or comments from fellow "expressionists" (not the
painters !!!) on why things aren't happening for me. I'm
presenting a tad-bit of detail of what I've done and what I
suspect to give you the "big picture".
     
     I know the full-length PPDK cDNA is in my expression vectors
based on restriction mapping and PCR analyses. I have not (yet!)
gone back to sequence these putative expression clones to make
sure my inserts are in-frame. My cloning strategy should have put
them in-frame, but theory, strategy and REALITY seem sometimes to
be completely different things. Based on prior research successes
(not in this lab !!) with both pTrc99a and pET11a, I find it hard
to beleive my problem resides with the promoter or induction.
I've induced with 0.4-5 mM IPTG. I have gotten expression from
one clone of what appears to be a truncated PPDK protein (by
Western blotting of induced whole cell pellet).

     I have since turned my attention to further analysis of the
PPDK protein and its sequence. When comparing codon preference in
E. coli and the actual codon usage in the PPDK message a very
interesting and/or potential problem becomes apparent.

     E. coli codon preference indicates that the arginine codon
AGG comprises only 3% of all 6 possible arginine codons. Maize
PPDK message contains 57 arginine codons, and 26 (or 46%) are
coded for by the AGG codon. Thus there is a 15+fold difference
between E.coli codon preference and PPDK message composition.
Because of this difference, I am questioning whether the E. coli
translational machinery (namely the number of available AGG-tRNA-
arginine) can provide enough AGG-tRNA-arginine to meet the PPDK's
translational demand for AGG-tRNA-arginine ?!!!  
     
     If the E. coli translational machinery cannot provide enough
of the correct tRNA I would expect synthesis of truncated protein
... which would provide one possible explanation for my truncated
protein results. BUT, is this a REAL possibility or am I not
giving "mother nature" more credit than is do ?

     I should also note that the AGG arginine codon is not the
only potential problem codon, there are eight others. If the
codon preference/usage differences were restricted to one or two
codons, I would be hesitant to suspect their detrimental effect
on the translational process. BUT, the fact that there are
several potential problem codons, and clustered within the first
100-200 bases of the PPDK message has got me very concerned about
the REAL feasibility of getting plant PPDK expressed in E. coli.

     A bacterial (Bacteroides symbiosus) PPDK, as well as two
other plant proteins (phosphoenolpyruvate carboxylase, NADP-MDH)
have been successfully cloned/expressed in an E. coli host, but
neither of these have the same degree of potential codon
problems. Is this relevant ? It seems to suggest to me that I
have a better chance of being struck by lightning than getting
maize PPDK protein expressed in E. coli. 

     I don't have the time or resources to try other non-
bacterial expression systems. What I need is some comfort that
I'm not missing something, seeing things, or going insane !!!

     Any ideas, suggestions, insights would be GREATLY
APPRECIATED. I do have to answer to a higher "authority" as to
why I'm not being >>PRODUCTIVE<< (isolating a full-length PPDK
cDNA clone, cloning into an expression vector, AND getting
EXPRESSION, all in 12 months !!!). 

                              Thanks in Advance,

                                   Chris


Christopher Smith
Department of Biochemistry
University of Nebraska-Lincoln
Lincoln, Nebraska 68583-0718
(402) 472-6806/2932
e-mail:csmith at crcvms.unl.edu



More information about the Plantbio mailing list