Wild type E. coli strains are able to use L-arabinose as carbon source, which is the reason why you might use an ara deficient strain. E. coli LMG194 proposed by Dr. Duez is one option. If you are interested in K12 phenotypes, you could use the E. coli BW25113 strain or the mutants constructed from this strain (which is also ara deficient). The KO collection from Keio University has KO mutants for almost all E. coli genes. 

http://msb.embopress.org/content/2/1/2006.0008.abstract

Moreover, if you want to be sure that the protein is expressed, you could perform O/N induction at 20-25ºC with your plasmid in a good producer strain such as E. coli BL21. This strain is deficient in proteases and protein yields are usually much higher than in K12.

Two hours of induction/expression is usually too short to see a clear overexpression of a protein through SDS-PAGE. Try to make a time course experiment to determine optimal induction time for your system.

Good luck.

Which gene is knocked out, and why you trying to over-express with arabinose as opposed to other expression systems? Did you induce expression during exponential growth? Or stationary phase? Are you using LB? 

Is the protein you are trying to over-express a difficult, regulated, or chaperoned protein? 

Many laboratory strains of E. coli cannot metabolize arabinose effectively. This seems beneficial, since the method doesn't require multiple inputs of arabinose. If your strain has both WT araA and WT araB genes, then your cells might be transforming arabinose into other chemicals. However, if you have an araD mutant, then adding arabinose to the cells will result in a build up of a toxic intermediate (Ribulose-5P) . Your strain appears to perhaps metabolize arabinose, since 'F- λ- ilvG- rfb-50 rph-1' is its genotype ("E. coli genetics OpenWetWear") Are you adding glucose?

Dear Mr. Nicolas,

First of all, thank you very much for your response. 

My plan is trying to compare the difference between the non-induced growth curve and induced growth curve of my strain. 

I transfered vector PBAD + Insert into MG1655, this vector has ara promoter and arabinose is its inducer. 

I am using LB media as it nutritive source. I did not add glucose into the media. Why do ask about adding glucose? Could you explain to me more?

Thank you very much for your precious time

Yours sincerely,

Le Phuong

Dr Mr

pBAD expression vectors are interesting for producing toxic proteins or for getting a inducer dose-response expression but you must work in RM medium (see the Invitrogen pBAD manual for composition) and with E. coli Top10 (if the protein is not toxic) or with E. coli LMG194 if the recombinannt protein is toxic for the host.

It is a well-regulated system but much less powerful than systems based on pET vectors.

Good luck

Dr Colette Duez

Dear Dr. Colette Duez,

I would like to express my thankfulness to your advice. 

In my case, I have to overexpress my protein in a nearly native E.coli with lowest genetic engineered strain to observe the real effect. That is the reason explaining why I have to overexpress in E.coli MG1665.

For knock out experiment, I used (PBAD vector inserted into MG1655) for overexpressing homologous recombinase at 30 degree. I also tried this condition with my protein but failed also.

Wild type E. coli strains are able to use L-arabinose as carbon source, which is the reason why you might use an ara deficient strain. E. coli LMG194 proposed by Dr. Duez is one option. If you are interested in K12 phenotypes, you could use the E. coli BW25113 strain or the mutants constructed from this strain (which is also ara deficient). The KO collection from Keio University has KO mutants for almost all E. coli genes. 

http://msb.embopress.org/content/2/1/2006.0008.abstract

Moreover, if you want to be sure that the protein is expressed, you could perform O/N induction at 20-25ºC with your plasmid in a good producer strain such as E. coli BL21. This strain is deficient in proteases and protein yields are usually much higher than in K12.

Two hours of induction/expression is usually too short to see a clear overexpression of a protein through SDS-PAGE. Try to make a time course experiment to determine optimal induction time for your system.

Good luck.

Dear Vicente,

thank you for your help here. Can you maybe put in the direction of any paper where they do a comparison on the expression levels of E. coli BL21 and K12?

Thank you

Jacopo