Re: Can bacteria express proteins with disufide bonds?

Hi Jimmy,

Thanks for the great questions. You asked two slightly different questions, which turn out to have vastly different answers. So, the answer to your first question, "can bacteria express proteins with disufide bonds?" is yes, but the answer to your second question, "is the cytoplasmic environment of bacteria suitable to express proteins with disulfide bonds?" is not usually, no!

Lets take a closer look. Disulfide bond formation is basically a redox reaction in which the sulfhydryl sidegroups (-SH) of two cysteine residues are oxidized, forming cystine (-S-S-) disulfides. In general, intra-cellular proteins usually lack cystines, while extra-cellular proteins often contain many. This observation is true for both prokaryotes (bacteria and to a lesser extent, archaea) as well as eukaryotes, and it was originally attributed to the reducing nature of the cytoplasm of both types of organisms. Even in eukaryotes, cystines are often formed in the oxidizing environment of the endoplasmic reticulum (ER), which is separate from the cytoplasm.

Until the early 1990s, it was assumed that disulfide bond formation was a largely spontaneous process that took place in an oxidizing environment. However, it is now clear that there is a rather complex electron transport pathway that is responsible for the proper formation of disulfides in both prokaryotes and eukaryotes. Whereas eukaryotes can create oxidizing environments in their intra-cellular organelles, prokaryotes catalyze the formation of disulfides in the periplasmic space between their plasma membranes and their cell walls. As with the eukaryotic ER, the periplasmic space is an oxidizing environment, and these electron transfers of disulfide bond formation are carried out by a series of enzymes known as DsbA, DsbB, DsbC, and DsbD. After a decade of study of these enzymes, it turns out that similar molecules are used in eukaryotes as well.

So now you ask, why can’t insulin be produced in a properly crosslinked state by Escherichia coli? Well, insulin is normally produced as a single proinsulin peptide, which has to be proteolytically cleaved after it has been synthesized in order to be properly aligned for crosslinking. E. coli don't have the proper proteaase for this cleavage. So, instead of trying to give them a protease that could conceivably chop up the rest of the bacterial proteins, insulin is expressed in E. coli as two separate chains, which are then purified and crosslinked synthetically.

A lot of the work on the electron transport pathway for the synthesis of disulfides has been carried out by Dr. Johnathan Beckwith. If you want more information on how the pathway works, as well as some other examples of prokaryotic disulfide synthesis pathways, take a look at the following papers. The first two are early papers describing the Dsb system, and the third is an excellent review of bacterial disulfide synthesis pathways. This review also discusses recent findings of disulfide bonded proteins in the cytoplasm of archaea, which is why the answer to your second question was not a flat out no (even though, technically, bacteria and archaea are very different).

Bardwell JC, McGovern K, Beckwith J. (1991) Identification of a protein required for disulfide bond formation in vivo. Cell.;67(3):581-9.

Bardwell JC, Lee JO, Jander G, Martin N, Belin D, Beckwith J. (1993) A pathway for disulfide bond formation in vivo. Proceedings of the National Academy of Sciences of the U S A. 90(3):1038-42.

Kadokura H, Katzen F, Beckwith J. (2003) Protein disulfide bond formation in prokaryotes. Annual Review of Biochemistry.72:111-35. (Epub 2003 Jan 09)