Purification of Recombinant E. coli Proteins
Because E. coli lacks glycosylation machinery, expression of glycoproteins in E. coli systems results in the synthesis of nonglycosylated variants. Glycoproteins expressed in E. coli are often, but not always, insoluble. In vitro folding studies with glycosylated and nonglycosylated forms of proteins indicate that the carbohydrate can stabilize folding intermediates, and thus enhance folding, while not necessarily affecting the stability of the native state (Kern et al., 1993, and references cited therein). In eukaryotic cells, interference with protein glycosylation can lead to the formation of misfolded, aggregated, and degraded protein. This indicates that in vivo glycosylation (N-linked) may also prevent the aggregation of folding intermediates (reviewed by Helenius, 1994). Detailed NMR studies on glycoproteins have clearly shown that carbohydrates stabilize folded proteins and even prevent marginally stable proteins from unfolding (for a review, see Wyss and Wagner, 1996).
Despite potential pitfalls, many nonglycosylated protein variants have been successfully folded from E. coli inclusion bodies. Examples include cytokines of biomedical importance such as granulocyte/macrophage colony-stimulating factor (GM-CSF; Diederichs et al., 1991) and interleukin 5 (IL-5; Milburn et al., 1993). Inclusion body formation was avoided in some studies by using secretion vectors; examples include GM-CSF (Walter et al., 1992) and the extracellular domain of the human growth hormone receptor (deVos et al., 1992). The aforementioned proteins have been crystallized and their structures determined by X-ray crystallography, supporting the view that the structural integrity and conformation of the proteins were not affected by the lack of glycosylation and their respective preparative histories.
If a glycoprotein of interest is available from a eukaryotic recombinant expression system or if the natural protein is available, then before investing time with E. coli expression, it may be worthwhile to determine whether the protein can be denatured and refolded in vitro. Pilot experiments can be carried out on intact protein and on protein enzymatically deglycosylated with glycosidases and, if disulfides are present, with and without reduction. Of course, if the protein can be secreted to the periplasm, aggregation and the necessity for in vitro folding may be avoided.
The production of deglycosylated proteins in E. coli expression systems for in vitro biochemical and structural studies is obviously of great value; however, the proteins may not always be suitable for in vivo studies due to low biological activity. Compared to authentic proteins, nonglycosylated variants can have a reduced circulatory lifetime and can exhibit increased immunogenicity and protease sensitivity (Rasmussen, 1992).
Was this article helpful?