Hi,

Beth from Practichem here, and I have had a lot of experience with protein expression and purification.  The previous responses are very helpful, and I wanted to add to them.

The first thing I recommend is double-checking your current clone sequence.  (If you haven't sequenced your clone completely, now is a good time to make sure you haven't introduced any mutations.)  To check for rare codon usage, a helpful online tool is www.genscript.com/tools/rare-codon-analysis or http://nihserver.mbi.ucla.edu/RACC/.  Next, you should check for predicted proteolytic sites.  While a poly-Gly-Ser linker is common, you may have introduced new proteolytic sites since the linker is flanked by native protein sequences to either side.  A few online tools for proteolysis analysis are http://web.expasy.org/peptide_cutter and http://cutdb.burnham.org.  If you've introduced rare codons or new proteolytic sites, I would recommend using site-directed mutagenesis to optimize codon usage and alter the linker sequence to reduce proteolysis.

You may want to consider re-engineering your linker.  I've included the PDFs of a few very useful reviews about linkers and linker design.  I would clone a series of constructs with varying linker lengths and sequences to find the one that works best for expression, purification, and your protein-protein interactions studies.  I've had colleagues who tried a half-dozen linkers before they found "the one".  You may also have to alter the flanking sequences from the native proteins to improve linker stability.  You can look for structural information on your proteins of interest in the PDB or use online tools to predict secondary structure, and I have included an article on structural features that affect proteolysis.  

I hope you have found these resources useful.  Please feel free to send me a message if you have any additional questions, or if you found a particular suggestion useful.

Good Luck!

Beth

Are you certain it is degradation and not premature translational stop? Is the protein full-length before purification?

If it isn't then you may get premature translational stop. For example GGA is a rare glycine codon in E. coli and if you have a few of these in tandem they might cause the translation to terminate prematurely.

We use (GGGGS)n codons a lot and have never had any problems with them except that you gave to optimise the codons for the organism in which you express them.

If it is protease degradation, try using a protease inhibitor cocktail.

Good luck! 

Possibly there is too much flexibility in the connecting region and this leads to instability. What I would do is to analyze the two proteins individually by sequence alignments with related proteins or (if available) have a look at the structure to see if the termini are important or structurally well defined. If not, it is advisable to fuse the two proteins with a shorter linker or without a linker or even to create fusions that even cut back the original sequences. Sometimes shorter is better. See also PMID 9862457.

Hi,

Beth from Practichem here, and I have had a lot of experience with protein expression and purification.  The previous responses are very helpful, and I wanted to add to them.

The first thing I recommend is double-checking your current clone sequence.  (If you haven't sequenced your clone completely, now is a good time to make sure you haven't introduced any mutations.)  To check for rare codon usage, a helpful online tool is www.genscript.com/tools/rare-codon-analysis or http://nihserver.mbi.ucla.edu/RACC/.  Next, you should check for predicted proteolytic sites.  While a poly-Gly-Ser linker is common, you may have introduced new proteolytic sites since the linker is flanked by native protein sequences to either side.  A few online tools for proteolysis analysis are http://web.expasy.org/peptide_cutter and http://cutdb.burnham.org.  If you've introduced rare codons or new proteolytic sites, I would recommend using site-directed mutagenesis to optimize codon usage and alter the linker sequence to reduce proteolysis.

You may want to consider re-engineering your linker.  I've included the PDFs of a few very useful reviews about linkers and linker design.  I would clone a series of constructs with varying linker lengths and sequences to find the one that works best for expression, purification, and your protein-protein interactions studies.  I've had colleagues who tried a half-dozen linkers before they found "the one".  You may also have to alter the flanking sequences from the native proteins to improve linker stability.  You can look for structural information on your proteins of interest in the PDB or use online tools to predict secondary structure, and I have included an article on structural features that affect proteolysis.  

I hope you have found these resources useful.  Please feel free to send me a message if you have any additional questions, or if you found a particular suggestion useful.

Good Luck!

Beth