Disulphide bond formation in the ER

Contact information

Professor Lloyd Ruddock Biocenter Oulu and Department of Biochemistry P.O.Box 3000 FIN-90014 University of Oulu FINLAND Tel: +358-8-553 1136 Fax: +358-8-553 1141 e-mail: lloyd.ruddock@oulu.fi

A major challenge for protein science over the coming decades will be to integrate the information obtained from in silico, in vitro and in vivo based studies to produce an understanding of how individual proteins function within the complex and highly interactive systems formed by cells and whole organisms. This ambitious goal will only be achieved by a multidisciplinary approach that utilises both established and new technologies and encompasses the full spectrum of molecular biosciences (from computational and structural biology through to cell and developmental biology).

In the system to be studied, the overall aim is to provide a complete molecular description of the processes by which native disulphide bond formation occurs within the endoplasmic reticulum (ER) of higher eukaryotes. The research will be focused on determining the function and mechanism of action of individual proteins in this biological system and on understanding the functional significance of cross-talk between interacting proteins. This will require detailed studies of the molecular activity and mechanism of action of individual protein components (what they do and how they do it) and of the expression characteristics of the proteins (where, when and in what form they are expressed). The use of a “quantitative” in vivo approach, combined with the results from the in vitro studies should allow for a detailed elucidation of which function(s) each gene product can provide and just as importantly which are relevant at physiological concentrations. This should allow for the rational manipulation of these systems to improve the efficiency of eukaryotic cells as factories for the production of high value secreted proteins.

Background

The lumen of the ER is a compartment specialised for protein folding; proteins destined for secretion, or for compartments accessed via the secretory pathway, enter the ER lumen unfolded, but only exit when correctly folded and assembled. Recent work has focused on the nature of the system responsible for quality control at this point, but there is still much to be learned about the cellular machinery involved in folding and assembly in this compartment.

Protein folding in this context is often associated with the formation of native disulphide bonds, and this is facilitated by the enzyme protein disulphide isomerase (PDI). PDI was the first catalyst of protein folding to be identified nearly 40 years ago, but many questions remain unanswered about its mechanism of action. The problem is difficult because of the complexity of the physiological substrates and the absence of a complete 3-dimensional structure for any PDI. When recent experimental data (proteolysis of native PDI and characterisation of recombinant fragments) is combined with bioinformatic approaches, the model that emerges is of PDI comprising four structural domains, a, b, b' and a' plus a linker region between b’ and a’ and a C-terminal acidic extension (Figure 1). The homologous a and a' domains of PDI, which contain the active site motif -WCGHC-, show significant sequence identity to thioredoxin, a small protein involved in many cytoplasmic redox functions. The single domain members of the thioredoxin superfamily (thioredoxins, glutaredoxins), all have the same characteristic fold, an alpha/beta fold, with the structure beta-alpha-beta-alpha-beta-alpha-beta-beta-alpha. The -WCXXC- active site motif, which is redox active and converts between the dithiol and disulphide forms, is found in an exposed turn linking beta2 to alpha2.

NMR studies on the recombinant a domain of human PDI have confirmed that it is a genuine structural domain with the thioredoxin fold and many of the characteristic features of thioredoxins (Figure 2). No high resolution structure of the a’ domain has yet been derived, although preliminary NMR data and secondary structure assignment confirm the similarity of the overall fold to that of a.

The b and b' domains show significant sequence similarity to each other but no obvious similarity to thioredoxin or to the a domain. Nevertheless, NMR analysis has revealed that the b region also forms a domain with the thioredoxin fold but in this domain the typical thioredoxin-like active site has been deleted and other residues associated with redox properties have been replaced. The sequence similarity of b’ to b implies that b’ also has the thioredoxin fold.

PDI is an excellent catalyst of disulphide bond formation and isomerisation. However, it is clear that in vivo there are multiple gene products that play a role in disulphide bond formation. For example in yeast the gene products of PDI1, ERO1, yFMO, EUG1, MPD1, MPD2 and EPS1 have all been implicated in the pathway for native disulphide bond formation (see figure 3). While none of the other yeast gene products can be currently considered to be a protein disulphide isomerase i.e. that the only gene encoding a “PDI” is PDI1, it is clear that in higher eukaryotes this is not the case. Recently a range of PDI homologues have been found in higher eukaryotes including Erp72, Erp57, P5, PDIp and PDIr, discoveries which add an additional complexity to the in vivo pathway (Figure 3).

With multiple members of the PDI family being present in mammalian tissues, even within the same cell, the question arises, as to why should there be so many members of the same family, all with the same putative function and all located in the ER? It is possible that there may be differences in tissue distribution, substrate specificity, induction specificity, reaction specificity and differential regulation of activity. Although there is now evidence that several of these may be occurring, the molecular characterisation required to fully define the functional differentiation is far from complete.

The resolution of these issues is perhaps best characterised by a comparison between PDI and PDIp, a pancreas specific homologue. Recently at the University of Kent we demonstrated that the two proteins show differences in tissue distribution, a putative difference in regulation and most importantly differences in substrate binding. This difference in substrate binding specificities implies that they most likely act on different subsets of folding proteins, an implication which has important consequences for the rational re-engineering of cells for the production of proteins of biotechnological or biomedical importance.

Top

Previous research

Before moving to Oulu in January 2001, Dr Ruddock was a lecturer at the University of Kent and before that Laboratory Research Manager for Prof Freedman, at the University of Kent.

In 1998 the group utilised a peptide binding assay, developed by Dr Klappa (University of Kent), which requires no prior purification of proteins expressed in E.coli, to determine for the first time that the b’ domain of PDI provided the principal peptide-binding site of the enzyme and that this site formed the core of the non-native protein binding site. This assay was subsequently used to study the peptide binding of PDIp, whose peptide binding is inhibited by certain oestrogens and xeno-oestrogens. The data obtained on the inhibition of PDIp peptide binding by oestrogens led to the hypothesis and subsequent demonstration that tyrosine and tryptophan residues are the recognition motifs for the binding of peptides to PDIp. This was the first paper detailing the specificity of substrate binding by a protein folding catalyst. The specificity of PDI has yet to be fully elucidated, but it is not the same as that of PDIp.

At the time this data was obtained no information was available on the structure of the substrate binding b’ domain of PDI. A homology model of this domain was constructed within the group and this was used to construct a model of the whole PDI and PDIp enzymes (unpublished data). This enabled a putative substrate binding site to be identified, the validity of which was tested by examining the possible binding of known oestrogenic inhibitors of peptide binding e.g. 17beta-estradiol inhibits peptide binding, 17alpha-estradiol does not. Based on this model a range of mutants have started to be constructed in PDI which would be expected to modulate peptide binding. To date there is a very good correlation between the predicted and the experimentally determined modulation of peptide binding, validating the model (unpublished data).

The structure of the a, b and a’ domains of PDI have been published, but the substrate binding b’ domain did not appear to be amenable to structural determination by NMR. A redefinition of the domain boundaries, combined with the use of a high affinity peptide ligand has generated a construct which is giving good, well-resolved NMR spectra, the structure of which is currently being solved in collaboration with Dr Williamson (University of Kent).

The yeast homologue of PDI (Pdi1p) is thought to be an essential gene product, which has complicated the study of the other proteins implicated in native disulphide bond formation in yeast. The group in Kent generated a PDI1 null strain, which is viable through a mutation in a second unrelated gene (unpublished observations). This second mutation has been characterised. The use of this strain in this project will allow, for the first time, a direct examination of the function of the other gene products in native disulphide bond formation.

In addition to its role in native disulphide bond formation, PDI is a subunit of two multisubunit complexes, one of which, prolyl-4-hydroxylase, is involved in collagen biogenesis. In collaboration with the group of Prof. Kivirikko (Biocenter Oulu) it has been determined that the b’ domain plays a critical role in the assembly process and that mutations that destabilise the a’ domain of PDI indirectly affect peptide binding and possibly are responsible for prolyl-4-hydroxylase assembly deficiencies.

Erp60 is involved in the formation of disulphide bonds in glycosylated proteins via its interaction with calnexin and calreticulin. Rationalisation of data in the literature combined with a peptide binding study on Erp60, led to a hypothesis regarding the interaction site between these proteins. This hypothesis is being tested in collaboration with Prof High (University of Manchester).

Molecular characterisation of the mechanism of action of these enzymes will require not only a multidisciplinary approach that utilises both established and new technologies and encompasses the full spectrum of molecular biosciences, but also innovative application of these approaches. Currently new insights are being obtained into the peptide binding site (including specificities and affinities and a conformational change switching mechanism), through a combination of biophysical techniques (unpublished data).

Group members

Alanen Heli, lab.technician
Karala Anna, B.Sc.
Lappi Anna-Kaisa, M.Sc.
Peltoniemi Mirva, M.Sc.
Ruddock Lloyd Ph.D.
Salo Kirsi, lab.technician

Collaborators

• Prof Stephen High, University of Manchester: The interaction between Erp60 and Calnexin/Calreticulin
  
• Prof Kari Kivirikko, Biocenter Oulu: The role of PDI as the beta-subunit of prolyl-4-hydroxylase
  
• Dr Peter Klappa, University of Kent: Interactions between PDI family members and small peptide ligands
  
• Dr Margherita Ruoppolo, University of Salerno: In vitro enzymatic studies on the role of the substrate binding site of PDI
  
• Prof Rikkert Wierenga, Biocenter Oulu: Crystallisation trials of PDI family members
  
• Dr Richard Williamson, University fo Kent: The structure of the b’ domain of PDI complexed to a peptide ligand

Recent publications in this field

1. Klappa, P., Ruddock, L.W., Darby, N.J. and Freedman, R.B. (1998) The b’ domain provides the principal peptide-binding site of protein disulfide isomerase but all domains contribute to binding of misfolded proteins. EMBO.J., 17, 927-935.
   
2. Klappa, P., Stromer, T., Zimmermann, R., Ruddock, L.W. and Freedman, R.B. (1998) A pancreas-specific glycosylated protein disulphide-isomerase binds to misfolded proteins and peptides with an interaction inhibited by oestrogens. Eur.J.Biochem., 254, 63-69.
   
3. Freedman, R.B., Dunn, A.D. and Ruddock, L.W. (1998) A missing redox link in the endoplasmic reticulum. Current Biology, 8, R468-R470.
   
4. Ruddock, L.W. and Klappa, P. (1999) Oxidative stress: Protein folding with a novel redox switch. Current Biology, 9, R400-R402.
   
5. Ruddock, L.W., Freedman, R.B. and Klappa, P (2000). Specificity in substrate binding by folding catalysts: Tyrosine and tryptophan residues are the recognition motifs for the binding of peptides to the pancreas-specific protein disulphide isomerase PDIp. Protein Science, 9, 758-764.
   
6. Klappa, P., Karpinnen, P., Pirneskoski, A., Ruddock, L.W., Kivirikko, K.I. and Freedman, R.B. (2000) Mutations which destabilise the a’ domain of human protein disulphide isomerase indirectly affect peptide binding. J.Biol.Chem., 275, 13213-13218.
   
7. Pirneskoski, A., Ruddock, L.W., Klappa, P., Freedman, R.B., Kivirikko, K.I. and Koivunen, P (2001). Domains b’ and a’ of protein disulfide isomerase fulfill a minimum requirement for prolyl 4-hydroxylase subunit function. J.Biol.Chem, 276, 11287-11293.
   
8. Klappa, P., Freedman, R.B., Langenbuch, M., Lan, M.S., Robinson, G.K. and Ruddock, L.W (2001) The pancreas-specific protein disulphide-isomerase PDIp interacts with hydroxyaryl group in ligands. Biochem. J 354, 553-9.
   
9. Freedman, R.B., Klappa, P. and Ruddock, L.W. Model peptide substrates and ligands in the analysis of the action of mammalian protein disulfide isomerase. Methods Enzymol. Manuscript in press.
   
10. Webb, H.M., Ruddock, L.W., Marchant, R., Jonas, K. and Klappa, P. Interaction of the periplasmic peptidyl-prolyl cis-trans isomerase SurA with model peptides: The N-terminal region of SurA is essential and sufficient for peptide binding. J.Biol.Chem. Manuscript in press.