10,000th structure curated by the Protein Data Bank in Europe
This week, the Protein Data Bank in Europe (PDBe) annotated its 10 000th structure for inclusion in the Protein Data Bank (PDB), a freely available resource for life science researchers. Gerard Kleywegt, Head of PDBe, tells us about the project and what’s special about this milestone structure.
PDBe is part of the European Bioinformatics Institute, and we work with our colleagues in the Worldwide Protein Data Bank (wwPDB) to provide a unified deposition, processing and distribution centre for experimentally determined biomacromolecular structures. Collectively, the wwPDB has already curated over 72 000 structures and the archive is growing rapidly. Our main goal is to provide a fully integrated resource for the biomedical and agri-science communities, and to make structural biology more accessible to all life science researchers.
The 10 000th entry, 2yf6, is of a Major Histocompatibility Complex (MHC)-peptide complex and was determined in the lab of Professor Susan Lea at the University of Oxford. This structure is helping to clarify how the immune system recognises an infected cell. The structure is of a chicken MHC from the B21 haplotype, which confers resistance to Marek’s disease, a contagious disease that can wipe out entire flocks. MHC proteins are present in all cells. Their job is to continually sample the internal protein environment and display short fragments of these proteins – peptides – on the cell’s surface. Passing T-cells determine whether these peptides are ‘self’ or foreign, and hopefully destroy the right ones.
One interesting question is how these complexes manage to display so many peptides, since this would require many different types of MHC proteins. Mammals express a wide variety of highly specific MHCs; chickens, on the other hand, express far fewer but each of these is capable of displaying one of several peptides on the cell surface. So while chicken and human MHCs may be structurally and functionally quite similar, the way they achieve their aim is subtly different.
Examining 2yf6 and similar structures in fine detail is helping to illuminate how chicken MHCs recognise and bind to peptides. 2yf6, a structure of a chicken MHC in combination with a chicken self-peptide, was determined by a senior postdoc, Pietro Roversi, and an undergraduate student, Paul Chappell, in Professor Lea’s lab using X-ray crystallography. The protein was generated in a collaboration with Professor Jim Kaufman’s group at the University of Cambridge.
The Cambridge group used high-throughput peptide generation to see which peptides bind to these MHCs; this made it possible to target key peptide complexes for structure determination. Understanding these structures will enable the scientists to define the structural rules that determine the peptides’ specificity. Professor Lea’s lab has deposited structures of three closely related chicken MHC-peptide complexes at PDBe and a few more are on their way.
All structures in the PDB archive are carefully curated before they are released. They are checked and annotated with information about, for example, small chemical entities present in the structure, and cross-linked with other biological data, such as protein sequences and functional information held in other databases. The process is rigorous but swift – the PDBe annotation team can usually check and return a deposition within one working day.
The PDBe website offers a suite of tools that help users access and analyse structures. For example, you can compare a protein structure to the entire PDB to find similar structures – which may yield clues about unexpected evolutionary relationships – or carry out sophisticated analyses of 3D data in seconds. We are currently working with our wwPDB partners to develop a software system that will let us handle the growing number of ever-larger and more complex structures being deposited in the archive every day. We are also working to integrate PDBe with chemical resources (e.g. ChEBI and ChEMBL) and biological resources (e.g. Reactome, a biological pathway resource).
We see 300 million structure downloads a year from wwPDB sites by users who wish to explore and understand remote evolutionary relationships between proteins, clarify biological mechanisms and discover new therapies for biotechnology and medicine. If you tend to shy away from structural biology, don’t – have a look at PDBe and let us know how we can make it easier for you to access and use this incredibly rich and diverse resource in your research.
Gerard Kleywegt, Head of PDBe, Senior Team Leader, EMBL-EBI
To learn more about 2yf6 and other interesting structures, visit http://pdbe.org/quips
Acknowledgment: The structure was determined by Chappell, P., Roversi, P., and Lea, S.M. (Sir William Dunn School of Pathology, University of Oxford) in collaboration with Harrison, M.C., Mears, L.E., and Kaufman, J.F. (Department of Pathology,University of Cambridge). X-ray data were collected at the Diamond Light Source, the UK’s national synchrotron facility in Harwell.