Efficient soluble expression of disulfide bonded proteins in the cytoplasm of Escherichia coli in fed-batch fermentations on chemically defined minimal media

Engineering Pathways in Central Carbon Metabolism Help to Increase Glycan Production and Improve N-Type Glycosylation of Recombinant Proteins in E. coli

Escherichia coli strains have been modified in a variety of ways to enhance the production of different recombinant proteins, targeting membrane protein expression, proteins with disulphide bonds, and more recently, proteins which require N-linked glycosylation. The addition of glycans to proteins remains a relatively inefficient process and here we aimed to combine genetic modifications within central carbon metabolic pathways in order to increase glycan precursor pools, prior to transfer onto polypeptide backbones. Using a lectin screen that detects cell surface representation of glycans, together with Western blot analyses using an O-antigen ligase mutant strain, the enhanced uptake and phosphorylation of sugars (ptsA) from the media combined with conservation of carbon through the glyoxylate shunt (icl) improved glycosylation efficiency of a bacterial protein AcrA by 69% and over 100% in an engineered human protein IFN-α2b. Unexpectedly, overexpression of a gene involved in the production of DXP from pyruvate (dxs), which was previously seen to have a positive impact on glycosylation, was detrimental to process efficiency and the possible reasons for this are discussed.

A Bacterial Expression Platform for Production of Therapeutic Proteins Containing Human-like O-Linked Glycans

We have developed an Escherichia coli strain for the in vivo production of O-glycosylated proteins. This was achieved using a dual plasmid approach: one encoding a therapeutic protein target, and a second encoding the enzymatic machinery required for O-glycosylation. The latter plasmid encodes human polypeptide N-acetylgalactosaminyl transferase as well as a β1,3-galactosyl transferase and UDP-Glc(NAc)-4-epimerase, both from Campylobacter jejuni, and a disulfide bond isomerase of bacterial or human origin. The effectiveness of this two-plasmid synthetic operon system has been tested on three proteins with therapeutic potential: the native and an engineered version of the naturally O-glycosylated human interferon α-2b, as well as human growth hormone with one engineered site of glycosylation. Having established proof of principle for the addition of the core-1 glycan onto proteins, we are now developing this system as a platform for producing and modifying human protein therapeutics with more complex O-glycan structures in E. coli.

The dimeric structure of wild-type human glycosyltransferase B4GalT1

Most glycosyltransferases, including B4GalT1 (EC 2.4.1.38), are known to assemble into enzyme homomers and functionally relevant heteromers in vivo. However, it remains unclear why and how these enzymes interact at the molecular/atomic level. Here, we solved the crystal structure of the wild-type human B4GalT1 homodimer. We also show that B4GalT1 exists in a dynamic equilibrium between monomer and dimer, since a purified monomer reappears as a mixture of both and as we obtained crystal forms of the monomer and dimer assemblies in the same crystallization conditions. These two crystal forms revealed the unliganded B4GalT1 in both the open and the closed conformation of the Trp loop and the lid regions, responsible for donor and acceptor substrate binding, respectively. The present structures also show the lid region in full in an open conformation, as well as a new conformation for the GlcNAc acceptor loop (residues 272–288). The physiological relevance of the homodimer in the crystal was validated by targeted mutagenesis studies coupled with FRET assays. These showed that changing key catalytic amino acids impaired homomer formation in vivo. The wild-type human B4GalT1 structure also explains why the variant proteins used for crystallization in earlier studies failed to reveal the homodimers described in this study.

Lactococcus lactis provides an efficient platform for production of disulfide-rich recombinant proteins from Plasmodium falciparum

BACKGROUND

The production of recombinant proteins with proper conformation, appropriate post-translational modifications in an easily scalable and cost-effective system is challenging. Lactococcus lactis has recently been identified as an efficient Gram positive cell factory for the production of recombinant protein. We and others have used this expression host for the production of selected malaria vaccine candidates. The safety of this production system has been confirmed in multiple clinical trials. Here we have explored L. lactis cell factories for the production of 31 representative Plasmodium falciparum antigens with varying sizes (ranging from 9 to 90 kDa) and varying degree of predicted structural complexities including eleven antigens with multiple predicted structural disulfide bonds, those which are considered difficult-to-produce proteins.

RESULTS

Of the 31 recombinant constructs attempted in the L. lactis expression system, the initial expression efficiency was 55% with 17 out of 31 recombinant gene constructs producing high levels of secreted recombinant protein. The majority of the constructs which failed to produce a recombinant protein were found to consist of multiple intra-molecular disulfide-bonds. We found that these disulfide-rich constructs could be produced in high yields when genetically fused to an intrinsically disorder protein domain (GLURP-R0). By exploiting the distinct biophysical and structural properties of the intrinsically disordered protein region we developed a simple heat-based strategy for fast purification of the disulfide-rich protein domains in yields ranging from 1 to 40 mg/l.

CONCLUSIONS

A novel procedure for the production and purification of disulfide-rich recombinant proteins in L. lactis is described.

Designing artificial metabolic pathways, construction of target enzymes, and analysis of their function

Artificial design of metabolic pathways is essential for the production of useful compounds using microbes. Based on this design, heterogeneous genes are introduced into the host, and then various analysis and evaluation methods are conducted to ensure that the target enzyme reactions are functionalized within the cell. In this chapter, we list successful examples of useful compounds produced by designing artificial metabolic pathways, and describe the methods involved in analyzing, evaluating, and optimizing the target enzyme reaction.