Mechanisms and applications of disulfide bond formation

Project Leader: Prof. Lloyd Ruddock, Ph.D.
Faculty of Biochemistry and Molecular Medicine,
University of Oulu

Background and Significance

In 1982 the FDA approved the use of recombinant insulin, triggering a healthcare revolution with the introduction of recombinant protein based therapeutics. Since then life expectancy has increased by 7 years in Finland and by 8 years worldwide. However, this has been accompanied by a massive expansion in healthcare costs bringing increased inequality, increasing moral dilemmas regarding the value of life quality vs cost of treatment and threatening to undermine the fabric of healthcare in the developed world. Cheaper more efficient means of producing high quality protein and peptide therapeutics must be found.

The limitation in both the affordability of protein based therapeutics and in the expansion of protein based therapeutics and analytics is heavily linked to the complexity of large-scale production of active proteins with appropriate homogenous post-translational modifications (PTM), in particular disulfide bonds. Disulfide bonds are naturally formed in the endoplasmic reticulum (ER), with the formation of the single native disulfide bonded state being essential to generate the biologically active conformation and disulfide bond formation often being the rate limiting step of protein folding in vivo and in vitro.

To lessen the costs and time of producing disulfide bond containing proteins we have developed CyDisCo technology. This technology for the efficient production of disulfide bond containing proteins in the cytoplasm of E.coli is simple to use and easily transfered from the laboratory to the market. CyDisCo is not only of potential use for industrial scale production of disulfide bonded proteins, but also as a facile system for the production of such proteins for academic use, for example for structure-function relationships.

Recent Progress

Development of CyDisCo
Based on more than two decades of studies by the PI on understanding the mechanisms for disulfide bond formation we have developed systems which allow efficient disulfide bond formation in the cytoplasm of E. coli. Unlike previously published systems, our systems do not require disruption of the reducing pathways naturally found in the cytoplasm, and they work in any media and in any E.coli strain tested to date.

Our system, known as CyDisCo (cytoplasmic disulfide bond formation in E.coli), has a variety of formats, but those most widely used share the common feature of co- or pre-expression of a sulfhydryl oxidase and a protein disulfide isomerase i.e. catalysts of the two steps of native disulfide bond formation. Patents in Europe and USA have been granted and the system has been commercialized.

Studies with the CyDisCo system shows that the system is very successful and high yields of active, correctly folded, eukaryotic proteins can be obtained. The patented variants of the system allow production of homogeneously folded human proteins with multiple disulfide bonds in E.coli grown in shake flasks with yields of up to 250 mg/litre culture. Recent results indicate that this can be extended into batch and fed-batch fermentation in defined minimal media even for proteins as complicated as full length human antibodies, with purified yields in excess of 1g/L of soluble folded protein being achieved for human growth hormone and interleukin 6 and up to 0.5g/L for human antibodies.

During the past two years we have gone back to redesign the fundamentals of the system and in parallel to introduce new factors into the system – again based on decades of fundamental studies on the mechanisms of protein folding. New variants that are more efficient in native disulfide bond formation as well as variants which aid other rate limiting steps in protein folding have been introduced. These new variants can increase the yield of folded protein obtained up to 4x.

Application of CyDisCo
The CyDisCo system was originally developed due to frustration with the inability to produce proteins involved in, or suspected to be involved in, protein folding and quality control in the ER in sufficient yields to undertake molecular characterization and structural studies.

This year one focus has been on increasing the breadth of proteins tested, both for academically interesting and industrially relevant proteins. During the past year three crystal structures have been solved of CyDisCo generated proteins. These include the fibrinogen domains of angiopoeitin-like proteins 3 and 4 (Angptl3 and Angptl4). Both proteins have been linked to coronary artery disease (CAD) with mutations reported to decrease CAD incidence by up to 34% and inhibition of function resulting in reduced plasma triglyceride and LDL levels and reduced artherosclerotic lesion. Angplt3 has been reported to be the next PSCK9. The crystal structures both reveal the structural effects of all reported mutations and open up the potential for rational design of low molecular weight inhibitors.

The third crystal structure solved in the past year was microsomal triglyceride transfer protein (MTP). MTP is a heterodimer with a lipid binding a-domain and protein disulfide isomerase (PDI) as the β-subunit. MTP is involved in lipoprotein assembly, with mutations in MTP causing abetalipoproteinemia. Drugs targetting MTP are used in humans to treat familial hypercholesterolemia and in dogs to treat obesity. The crystal structure allows elucidation of the molecular basis for disease causing mutations, the potential design for inhibitors and give important insights into the mechanisms of action of MTP and in particular the role of PDI in the complex.

In parallel to these studies utilizing CyDisCo to determine protein structure, we have also continued work on the mechanisms of action of key enzymes in oxidative protein folding such as Ero1 and PDI. For example, it is known that PDI must be able to trigger conformational changes in bound non-native protein subtrates to allow access to buried thiols and disulfides. This is an area that has been poorly pursued due to the extreme difficulty in isolating/identifying intermediates and in studying the process. We have data from collaborative NMR studies that conformational exchange we had previously identified within PDI is linked to the ability to trigger change in the conformation of folding intermediate mimics (produced using CyDisCo). These studies feed into the development of the CyDisCo system.

Finally, we have used our knowledge of the mechanisms of oxidative folding in collaborative studies to elucidate the role of variants in a PDI-family members in Amyotrophic Lateral Sclerosis (ALS) and motor dysfunction.

Future Goals

The overall aim of the group is to provide a complete molecular description of the processes by which protein folding occurs within the ER and the application of this knowledge for the efficient production of disulfide bond containing proteins of scientific, medicinal or biotechnological importance.

With the final stages of development of CyDisCo in sight the primary foci of the group will switch towards:

  • The use of CyDisCo to obtain mechanistic understanding of the pathways and synergy of protein folding, quality control and ER-associated degradation.
  • The development of Gen2Co for efficient secretion of disulfide bonded proteins
  • The development of systems analogous to CyDisCo for the production of other PTMs in the cytoplasm of E.coli
  • Collaborative studies utilizing CyDisCo for mechanistic studies on disulfide containing proteins.

Publications 2017-

Gaciarz A., Khatri N.K., Velez-Suberbie M.L., Saaranen M.J., Uchida Y., Keshavarz-Moore E., Ruddock L.W. Efficient soluble expression of disulfide bonded proteins in the cytoplasm of Escherichia coli in fed-batch fermentations on chemically defined minimal media. Micro Cell Fact 15, 108, 2017.

Gaciarz A., Ruddock, L.W. Complementarity determining regions and frameworks contribute to the disulfide bond independent folding of intrinsically stable scFv. PLoS One 12, e0189964, 2017.

Sliz E., Taipale M., Welling M., Skarp S., Alaraudanjoki V., Ignatius J., Ruddock L., Nissi R., Männikkö M. TUFT1, a novel candidate gene for metatarsophalangeal osteoarthritis, plays a role in chondrogenesis on a calcium-related pathway. PLoS One 12, e0175474, 2017.

Patents 2017

EP2670844 Method for producing disulfide bond containing proteins in a prokaryotic cytoplasm (Lloyd Ruddock & Feras Hatahet)

Research Group Members

ww
Project Leader:
Lloyd Ruddock, Ph.D., Professor

Senior and Post-doctoral Investigators:
Ekaterina Biterova (Academy of Finland)
Mirva Saaranen, Ph.D. (Academy of Finland and University of Oulu)

Ph.D. Students:
Zhang Chi, M.Sc. (Academy of Finland and Biocenter Oulu)
Anna Gaciarz, M.Sc. (Biocenter Oulu)
Kati Korhonen, M.Sc. (ISB, Academy of Finland)
Antti Moilanen (Academy of Finland)
Lisette Van Tassel, M.Sc. (Biocenter Oulu)

Laboratory Technicians, 1 (Sigrid Jusélius Foundation and University of Oulu)

Main source of salary in brackets.

Foreign Scientists, 5

Last updated: 6.7.2018