Not sure if this has already been mentioned, but if your fusion protein is toxic to E coli, you may be selecting for mutations during growth. If this is the case, you may have noticed low transformation efficiencies into the expression strain etc.

Mark J van Raaij
Dpto de Estructura de Macromoleculas
Centro Nacional de Biotecnologia - CSIC
c/Darwin 3
E-28049 Madrid, Spain
tel. (+34) 91 585 4616
http://wwwuser.cnb.csic.es/~mjvanraaij






> On 15 Feb 2016, at 18:40, Elias Fernandez <[log in to unmask]> wrote:
> 
> Dear colleagues,
> Thanks for the many approaches to address this problem. I neglected to mention that 
> 1.       We have an N-terminal His6 tag to the MBP-fusion and first purify the complex with Ni-NTA. Thus, it’s likely that either the fusion protein is getting cleaved or it’s a transcription/translation problem.
> 2.       We get approx. 50:50 MBP and fusion but the total MBP + MBP-fusion are very low.
> 3.       We can partially separate the two by size exclusion. 
> 4.       We’ve not had much luck with other tags, such as SUMO and GST.
> Best regards,
> Elias
> From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Julia Griese
> Sent: Monday, February 15, 2016 10:29 AM
> To: [log in to unmask]
> Subject: Re: [ccp4bb] MBP fusion
>  
> Dear Elias,
> 
> While, as suggested by Mark and others, there can be many reasons for your problems, such as your fusion protein being largely insoluble, I would second Michael's suggestion to put a His tag in front of the MBP. Ni-NTA has much much higher capacity than amylose (and is also much cheaper), so you can increase your yield hugely by using the MBP merely as a solubility enhancer, combined with a His tag as the actual affinity tag. It has the additional advantage of getting rid of host MBP of course. 
> I don't know which plasmid you're using, or if you secrete your fusion protein or not, but an easy way to put a His tag in front of the MBP that I used once was to clone MFG (my favourite gene) into pMAL-c2x (i.e. the cytosolic version), and then cut out the whole MBP-MFG construct from the plasmid with NdeI (cuts just 5' of malE) and whatever I had used at the 3' end of MFG, and put that into pET28 - thus generating a His6-(thrombin site)-MBP-(factor Xa site)-MFG fusion. I had had problems with low yield with the original pMAL-c2x construct, but this His6-MBP construct solved all my problems. 
> 
> Cheers,
> 
> Julia
> 
> On 15/02/16 15:39, R. Michael Garavito wrote:
>> Elias, 
>>  
>> Your experience is not unusual, in fact it is completely expected unless you work with a MalE null mutant, which I don’t think you can get versions of the usual BL21 expression hosts.  We always want to see native MBP as it means the amylose column is working (see note about amylase destruction of the  amylose resin) .  However, I can see that you don’t want to use 20-30 mL of expensive resin just to ensure that your MBP fusion protein is not out competed by the native MBP for the amylose resin.  If you have a real gene jockey at hand, you might engineer a N-terminal fusion to the MBP so that you create a His6-MBP fusion protein.  Keep the normal start/signal peptidase cleavage site, but add a 6- or 10-His tag in front.  Run a single step Ni-column to remove background protein and the native MBP, the load the elution fraction on an amylose column.
>>  
>> If you are not secreting your MBP fusion protein, another option is to pretreat your cells.  Native MBP is in the periplasm (at concentrations up to 1 mM).  There are protocols to pop open the periplasm (using high [Ca2+] or [EDTA]) then centrifuging your cells; the native MBP remains in the supernatant.  People have used this method to replace native MBP with labeled MBP to study periplasmic function (see work in the 1980s by Nikaido’s group).  Most of the time the cells are not broken, but I have never tested this method on a large scale (4-20 wet cell wt.).
>>  
>> I also noted that you are not using the typically recommended host from NEB with the REP3 plasmid.  While it may not be an issue, the NEB host has reduced amylase production which can chew up your amylose resin faster.
>>  
>> Good luck,
>>  
>> Michael
>>  
>> ****************************************************************
>> R. Michael Garavito, Ph.D.
>> Professor of Biochemistry & Molecular Biology
>> 603 Wilson Rd., Rm. 513   
>> Michigan State University      
>> East Lansing, MI 48824-1319
>> Office:  (517) 355-9724     Lab:  (517) 353-9125
>> FAX:  (517) 353-9334        Email:  [log in to unmask]
>> ****************************************************************
>> 
>> 
>>  
>>> On Feb 14, 2016, at 6:17 PM, Fernandez, Elias J <[log in to unmask]> wrote:
>>>  
>>> 
>>> Dear Colleagues,
>>> We are attempting to express our target protein fused to MBP in BL21(DE3) and BL21(RIPL). Unfortunately, we get a mixture of MBP-target protein and MBP only, which severely depletes our overall yield of MBP-target protein. We've tried enriched media, such as 2xYT & TB and also optimal high-density induction in minimal media at temperatures ranging from 20-37deg and IPTG from 0.05-0.5microM.
>>> Any suggestions to address this competitively high yield of MBP-only protein?
>>> Cheers,
>>> Elias
>>  
> 
> 
> -- 
> Dr. Julia Griese
> Postdoctoral Researcher
> Stockholm Center for Biomembrane Research
> Department of Biochemistry and Biophysics
> Stockholm University
> 106 91 Stockholm
> Sweden
>  
> phone: +46-(0)8-163 246
> email: [log in to unmask]