[ccp4bb]: Summary of replies: soluble aggregation of proteins

[Janet Moloney <janet@strubi.ox.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***


Dear all,

A while ago, I posted an email enquiring about any remedies people could
advise about the soluble aggregation of proteins. The proteins in this
case are being expressed in E. coli in the pMALp2 vector as
maltose-binding protein fusions. This is a periplasmic expression and we
found that our three proteins in the 22-38 KDa range all aggregated in
solution. The proteins are linked by the 3C recognition site and cleaved
by Rhinovirus 3C protease after several hours at room temperature.

Here is a summary of the replies that I got:

(1) Francisco Enquita wrote to suggest tactics to deal with the
aggregated protein such as increasing the strength of the resuspension
buffer from 200 mM NaCl to 300 mM. Also, 5-15% glycerol can be added to
the resuspension buffer. 
In the case of intermolecular disulfide bridge formation, DTT or
beta-mercaptoethanol can be included in the buffer to break these.
To deal with intermolecular disulfide bridges because of the high
expression levels and different codon usage than E. coli, he suggested
including a reducing agent in the column buffer, like DTT or
beta-mercaptoethanol, in the 1-5 mM concentration range. In the case of
membrane associated or bound proteins, a detergent must be added to the
buffer such as Nonidet PD-40 or Triton X-100 at a low concentration like
0.1-0.2 %. 
 
(2) Steve Soisson said that lysis could be carried out under
non-denaturing conditions of guanidine or urea, followed by a later
dialysis step to remove them. Cytoplasmic expression using the pMALc2
vector could be tried, since oxisdation should be greatly reduced inside
the cell. He also suggested trying a different expression system
altogether like His or GST tagged.

(3) Peter Adams recommended the review on general solubility questions
by C.H. Schein (Schein CH, Biotechnology vol 8 april 1990 pages
308-315.)

(4) Chris Garcia's email on the pMALp2 expression system dealt with the
reason why aggregation is observed and suggested an interesting strategy
in denaturing the entire fusion and refolding it:

pMAL-p2 is a "forced" secretion system with the logic that if you 
fuse your normally-secreted protein to it, the pMAL ( which contains 
a bacterial leader) will pull it out into the periplasm using the 
bacterial secretion machinery to correctly fold and form disulfide 
bonds.  Unfortunately, while a clever idea, this rarely works because 
the T7-lac based promoter in these vectors is way too fast for a 
heterologous protein to get properly processed through the secretion 
machinery.  Usually what happens is your protein will get backed up 
inside the bug, and not get secreted.  This is obvious when you have 
to sonicate, as opposed to osmotically-shock the bugs to release it. 
This should be a first warning sign that your protein, even though it 
appears to be soluble, is not folded properly and is aggregated. 
Sometimes, your protein will actually get secreted and is released by 
shocking, and is still not folded properly, as evidenced by gel 
filtration.   In this case there are a couple strategies which I have 
used:   1-  carry out the expression with lower IPTG concentrations 
at reduced temperature (28degC) to slow down the 
transcription/translation, and 2- denature the entire pMAL fusion and 
refold that, which often works much better than trying to refold your 
protein by itself.   So what I am saying is that the aggregation 
problems you are seeing is less a matter of the buffer 
conditions,etc., but more of an endemic problem of misfolding.  One 
can mask a lot of misfolded proteins by using an array of chaotropes 
to fool yourself into thinking you have the right thing. Chris added
that his success rate for secreting proteins with pMal was higher with
small domains (less than 10kDa). 

(5) Chris Dealwis suggested trying to optimize the correct temperature
for induction and failing that using detergents.

(6) Filip Van Petegem also suggested working at lower temperatures. In
addition to this, centrifuging the sample for more than an hour in order
to remove any protein that has already aggregated.  This aggregated
product can catalyze the formation of more aggregates.  But when these
'seeds' are removed from the solution, it can be safer to concentrate
afterwards. The other idea here was to change the pH of the solution
(far from the pI). 

Thanks very much to all that took an interest in the problem.

Best Wishes,
Janet.

-- 
Janet Moloney, Division of Structural Biology,
Wellcome Trust Centre for Human Genetics,
University of Oxford, Roosevelt Drive, 
Oxford OX3 7BN, U.K.

Tel: 0044(0)1865 287551	   Fax: 0044(0)1865 287547