Exprssing these kind of protein is sometime difficult, but you can try it with keeping these points in considetion-

start with 10/20 micro molar conc. of IPTG, grow till Abs (600 nm) 0.8, keep in ice for 30 min before induction, after induction droe for 2/3 h, check on gel, do silver staining if required.

I have used pMALC2X for xpression of MBP fused proteins.

It sounds like you may have a problem with proteolysis of the fusion protein.  The linker between MBP and the target polypeptide is indeed rather sensitive to proteases. Do you see production of MBP but not of the full-length fusion ? In some cases, expressing the protein at very low tempeerature (14 °C) may help. If not, I would suggest using another tag, e.g. His6, either at the N- or C-terminus.

There are a lot of points that you need to consider when you are trying to express something in E.coli...I think that it is necessary to think about them before trying to change expression conditions:

1) what is the predicted MW of the protein? Expression of the proteins with high MW (more than 70 kda)may be challenging...(as well as the expression of small proteins)

2) have you tried to purify your protein? sometimes the protein is undetectable on the coomassie-stained gel (when you are trying to compare induced and uninduced controls) but is purified pretty well...

3) western is more sensitive than coomassi-staining so you can try western in order to confirm presence of yours protein in E.coli lysate...

PS

try different vectors and different fusions...for ex. thioredoxin usually works very well as fusion protein...it is included as the fusion protein into the pET32a vector backbone...

Have you thought about what the endonuclease would do to your cells if you actually would get soluble and active full-length protein? It's not unlikely that your protein is lethal to your cells thus you might only get those cells which managed to survive, e.g. by losing the plasmid when amicillin conc. gets low enough to survive, deleting (parts of) your insert gene, proteolysis of the part fused to MBP, mutation of one codon downstream of MBP to stop codon, etc.

You could try to apply strategies that are useful for toxic proteins. Here are some suggestions that could help:

http://www.embl.de/pepcore/pepcore_services/protein_expression/ecoli/decreasing_protein_toxicity/index.html

I'm guessing by your question that you are actually getting truncated protein expressed?

there are a few points to consider

1. your protein is toxic to the cells: do your cells grow as expected. Use tighter expression control

2. codon bias is a problem: the expression is stalled because bacteria lack sufficient quantities of a particular tRNA. Use a strain of bacteria that can cope with this: BL21 RIPL etc. 

3. Your protein has repeat sequences: bacteria do not like this and will chop your protein up. SURE cells can cope with this

4. Your protein has multiple disulphide bonds, protein is unable to fold correctly and therefore is degraded or in inclusion bodies: the reducing environment in bacteria cytosol make it difficult to express some of these proteins: do you get insoluble protein of correct size? there are strains of bacteria available for this.

I would as a first do a rigorous time course of induction with controls: at 37C/30C if possible

Set up small scale cultures grow to OD ~0.4 and induce with IPTG. Take samples at 30minutes and on every hour after. leave o/n as well. Check cells for both soluble and insoluble expression. Use the same approach with different media: 2xYT, LB, TB (buffered media) etc.  Some proteins are expressed quickly but get degraded if left too long as we have experienced.

good luck

Fusion of the target protein to MBP does not always result in soluble protein. Have you checked the insoluble fraction?  you could try other solubility tags like SUMO.

@Rakesh Kumar

Thanks for the reply. 

@Rakesh Kumar : Few years back I tried expressing my protein in a similar manner which you have indicated but did not work with the fusion one though. I will keep your pointers in mind.

@ Pierre Béguin: I guess so. Proteolysis may be the problem. I am trying out various other lower temps. to see whether that makes a difference or not. I have tried his tag but the protein was unable to purify. I appreciate you for the pointers.

@Zeeshan Chaudhry: That will be beyond the scope for the moment, however I thank you for the reply.

@Anna Leopold: Thanks for the reply.I have been doing small scale protein expression to optimize the expression of this fused protein. the prdicetd molecular weight of the fusion protein is 74 Kda (Maltose binding protein is 42 KDa and the recombinant protein is 32 KDa). Ya I agree with you but one time I expressed efficiently 110 KDa yeast peroxisomal protein in E.coli. Well, I tried to purify using his tag but something always goes wrong with that purification.I kinda switched to the pMAL  expression system. Western does give the impression that the full length protein is formed at some point and there are few intermediate bands when detected with anti-MBP antibody. pET32 may be a good choice, I can think about that later.

 @Hüseyin Besir: Thank you. I tried in vivo endonuclease assay where I co transformed both the expression construct and the substrate plasmid and did in vivo cutting assay. I found the cells were quite viable, so this is not a toxic protein. the sequence of the fusion protein is in frame and we checked it twice and did not observe any mutation whatsoever.

@Ian R Wilkinson: Thanks for the reply. Yes we got me right. The protein is not toxic, the cells are happy even after induction. Its under tight promoter ptac so that's not the problem. Currently, I am transforming my expression construct (in pMAL vector) with pRARE plasmid and hope to see some more full length expression. Time scale experiments and changing media had been done with no such great difference in the expression of the full length fusion protein.

I did not think about the repeat sequences, but we checked the in silico model of the protein, there are no such numerous repeat sequences. pMALp5x can express my fused protein in the periplasm if the protein expression is way too much affected with reducing environment or disulfide bond formation. I have that in mind.

@Christopher Wilson: Thanks for the reply. The insoluble fraction and the soluble fraction kind of look the same. SUMO approach ..lets see you far I can stretch this pMAL stuff.