Hi there,

In theory there shouldn't be any problem. This strategy is used as a routine in our lab when we have to coexpress chaperonins to ensure the proper folding of our POI. Plasmid encoding chaperonins allows their arabinose driven expression  and plasmid from pETserie allows IPTG driven expression of POI in BL21(DE3) strain. Both overexpressions occur independently and concomittently without any problem. You might be confronted with something specific with your proteins.

Hello Matteo

Do your two plasmids share the same origin of replication ? This is usually not recommended, as it leads to instability of one of the partners. If so, you may wish to exchange the origin of one of your plasmids with another one, like the origin of p15A,  which is found in pACYC184 and its derivatives. Of course, both plasmids ought to have also different selection markers.

pET and pBAD vector share the same pBR replication origin, As Pierre mentioned before it could lead to one of the partner unstable, usually the one with lower replication copy number (the pET vector in your case). Better to re-clone one of your target gene into a vector with different replication origin.

Please note that arabinose also induces lac-based promoters. 

http://www.ncbi.nlm.nih.gov/pubmed/16739941

Hello again and thanks for your answers: the arabinose-inducible vector, even if it's pBAD-based, is not a pBAD vector and thus has a p15A origin (when I said PBAD I was only referring to the promoter and not to the vector series). The pET vectors, as some of you already pointed out, have a pBR322 origin so no clash between the two is foreseeable based on their replication-related features. The fact that p15A is a low-copy origin of replication shouldn't cause problems either, as the NATB complex acts catalitically and thus an extremely high expression level shouldn't be necessary for the bacterium to acetylate all of the target protein expressed. We spotted a problem in the target protein expression vector (random mutagenesis, maybe) and we're trying to find out more about it (and hopefully fix it), but more insights are welcome!

@Syed A Ali: Unfortunately I'm no microbiology expert, so I can't really judge the solidity of their claims (I went across other reports saying that actually arabinose, via a catabolite-repressor mechanism, can reduce the leaky expression of the lac system), but the authors are using a completely different E.Coli strain (we're working with BL21(DE3)) and mention the fact that this ability to induce a lac/T7 operon could be specific of certain strains, due to deletions/mutations of some metabolism-related proteins.

Hello Matteo,

I did not come across this paper until I myself made similar observations. We had two plasmid vectors, one of p15A ori and another of pBR322 ori both were used to transform BL21(DE3). The p15A ori-based plasmid expressed TEV protease under the control of PBad promoter whereas pBR322-based vector expressed MBP-TEV-scFv under the control of T7 promoter. The idea was to express MBP-TEV-scFv by inducing cells with IPTG followed by TEV induction using Arabinose that will result in intracellular cleavage of scFv from MBP. We wanted TEV to express at later stage when MBP-TEV-scFV was already fully expressed. To do that, we induced the co-transformed cells with IPTG and removed the IPTG containing medium by centrifugation. Cells were then resuspended in fresh medium containing Arabinose. To our surprise, Arabinose not only induced TEV expression but also of MBP-TEV-scFv. Initially I did not believe those observations but when we got same results after repeated assays, we looked for some reference and came across the paper I pasted the link for. 

Is there a strain appropriate for this? It seems BL21(DE3) is not recommended for PBAD because it still contains the ara operon. Are there any (DE3) strains deficient in ara operon? I am not sure about BL21-AI...