GST fusion prot.- HELP

*. * Troianovsk_s at MSDISK.WUSTL.EDU
Sat Nov 19 17:30:22 EST 1994

In article:
>Date:    7-NOV-1994 14:09
>From: aduarte at
>Description: GST fusion prot.- HELP

Antonio Duarte (aduarte at wrote:
>I am currently trying to express GST fusions proteins  (Approx. 55 KDa) 
>using the pGEX expression vector. These proteins are purified by affinitty 
>chromatography with a glutathione collumn and are used in DNAse I footprinting 
>The level of expression is good and I consistently collect high levels of
>protein but there is considerable degradation, with the full-size protein
>representing only about 50% of the purified extract with various other smaller
>products. This causes me problems both when trying to quantify the protein and
>in terms of validating the DNAse footprinting data, because of the
>possibility that the degradation products bind to the DNA target sequences with
>modified specificity.
>I suspect the degradation occurs while still growing the cells because I use
>protease inhibitors (aprotinin, leupeptin and pepstatin, 3ug/ml each) and
>apply a brief sonication treatment while keeping the cells on ice all the time
>and working in the cold room.
>Can you suggest me potential good strains for protein expression or any tips
>that may help to reduce the protein degradation (I'll enclose a summary of my
>Thank you very much.

Hi, there.

                  Warning. Codon Usage.
   Some proteins are pooly expressed in E.Coli if they contain rarely used
codons, which often caused premature termination of translation (Grosjean &
Fiers, 1992). This can result in a variety of turncated protein products,
particulary from large recombinant proteins.
   A short proteins can arise due to initiation of translation 3' to the
correct start site. Such internal starts arise when there is a ribosome
binding consensus sequence (Shine-Dalgarno sequence) 5' to an internal ATG
or GUG codon. Internal starts can be avoided by mutation of any potential
Shine-Dalgarno sequence in the coding region followed by a start codon
without affecting the protein sequence.

   In your case I'd suspect that the protein of interest fused to GST is
truly mammalian protein whose sequence caused premature termination of the
translation. So that the majoriry of your purified product is GST itself and
couple species of higher m/w. while there is present product of interest at
appr. 30-50% of total.
We also sruggeled with this in some cases. I saw the best results when such
a constructs without modification of a coding sequence, are expressed in DH5
alpha. LB medium + 2% glucose. But in any way, this chahged
product/contaminant ratio to 70/30%. Shure, there is nothing to do with
proteolysis of the product. I used evrything, nothing helped.

Hope this helps.

Child says: "If All Fails Carefully Read the Manual"

More information about the Methods mailing list