A brief introduction to CED
CED is short for Conformational Epitope Database.
Immune cells recognize "epitopes" (antigenic determinants) rather than entire antigens. Epitopes are special regions of an antigen that can bind to antigen-specific membrane receptors on lymphocytes or to secreted antibodies. Epitopes on macromolecules especially proteins can be classified into two categories. One is termed as linear epitope, which is a sequentially linear segment composed by continuous (contiguous) residues along the polymer chain. Another category is termed as conformational epitope, which is constituted by several sequentially discontinuous segments or noncontiguous residues that are brought together by the folding of the antigen to its native structure. Let's take a look at the pictures below.
The pictures above is about a linear epitope KPLEEVLNL, which is a continuous segment (64-72) from human interleukin 2. In the left picture, this linear epitope is shown by surface, and its paratope is shown by van der Waals dots spheres, the chains of the antibody are shown by backbone alpha carbon trace. In the middle picture, this linear epitope is also shown by surface, but Heavy and Light chains of antibody are in blue and green respectively. In right picture, this linear epitope is shown and colored by CPK, and the binding site of the antibody is shown by surface. All the pictures above are created by DeepView based on structure 1F90 from PDB. Now, let's look an example of conformational epitope.
The pictures above depict an conformational epitope of influenza virus neuraminidase. This epitope is constituted by 4 sequentially discontinuous segments ( PNDPT(328-332) + YPGN(341-344) + ISIAS(366-370) + NTDW(400-403) ) that are brought together by the folding of the antigen to its native structure. For more information about this conformational epitope, you can browse or search CE0184 by Conformational epitope ID. In the left picture, this epitope is shown and colored by surface, and its paratope is shown by van der Waals dots spheres, the chains of the antibody and antigen are both shown by backbone alpha carbon trace. In the middle picture, this epitope is colored by CPK. The 4 segments are circled by 4 ellipses and are pulled away from antibody binding site. In right picture, this epitope and its paratope are both shown by their van der Waals dots spheres. Good shape complementarity is clear. All the pictures above are created by DeepView based on structure 1NMB from PDB.
As the molecular basis of immune recognition and immune response, any types of epitope provide valuable information that is very useful for disease prevention, diagnosis, and treatment. The conformational epitope is no exception. For example, a few FDA approved drugs such as Herceptin (CE0096) and Erbitux (CE0199) which are quite effective for cancer, target at conformational epitopes. A good few drugs targeting at conformational epitopes have been on clinical trials too. In a word, conformational epitopes not only provide useful information for new vaccine design, new diagnostic reagent development and new drug research against all kinds of diseases such as virus infection and cancer, but also have implicit structure information, which makes it very attractive in theoretical biology research too.
The MySQL relational database management system has been used for CED to store, retrieve and manage the data. The web interface between the user and the CED database are coded by PHP with PEAR database abstraction layer support. All entries in this database are thoroughly sourced from articles published on the peer viewed journals. Initially, exhaustive queries are made to PubMed and ScienceDirect; more than 3000 references are loaded into an EndNote reference database. The references are further filtered manually to exclude articles that do not define a conformational epitope or the defined epitope is only at a very low resolution and completeness. Later, the well defined data about conformational epitope is checked and extracted manually into CED database from the primary experimental research reports.
CED is small in size at present. To identify a conformational epitope in detail experimentally is a hard, time-consuming, money and labor intensive work. It is also very difficult and hard to predict a conformational epitope accurately. According to reviews and research reports, as much as 90% of B cell epitopes on native proteins are conformational rather than linear and there are a lot of articles that reported new conformational epitopes. However, few are defined completely and precisely enough. Most conformational epitopes are defined and refined successively along with progress of experiment techniques. Take autoimmune epitopes as an example. Most auto-antibodies react with conformational epitopes and many of them has been successively defined and refined at the level from whole cellular organelles (immunofluorescence method), identified molecules (immunoblot, gene expression libraries) to epitope regions (truncated cDNAs, peptide scanning), but few are identified at the contact residues level. As the data quality is vital in bioinformatics research, the aim of CED is to provide high quality well defined conformational epitope information. Thus, a lot of conformational epitopes that are not defined clearly are discarded. These may be another reasons why CED is small in size. Along with the rapid progress of techniques, we can expect that more and more new conformational epitopes will be identified at a more accurate, more complete and more precise level. Thus CED will grow up too.
From the homepage of CED, you have found this brief introduction. And you can browse epitope records page by page, entry by entry from there too. When browsing CED database, the entries of CED will appear firstly in a brief paged table. Only the conformational epitope ID and the constitution and location field will be shown in the brief table, which is ordered by corresponding IDs. Simply a click on each ID in the table will open a new window that displays the information of each epitope in detail. You can also search the CED database for conformational epitopes of your interest. A help page for searching can be found on the search interface.
You can also view the conformational epitopes in CED interactively in context of antigen-antibody complex or antigen structure or known theoretical model, if they have corresponding PDB structures in PDB database. The visualization function of CED is powered by Jmol. This visualization tool is quite useful as it can help you to judge if an entry defined by other experimental method rather than X-ray diffraction reasonable. When you browse or search CED, the one has PDB structure will have a red view icon in both summary table and entry table. Click the view icon will initialize the loading of Jmol Java applet. After that, you can operate (turn on or turn off the epitope segments, antigen chains, antibody chains, zoom in, zoom out, move, spin, rotate the structure, or even measure distance, angle, dihedral, etc.) the epitope and its corresponding PDB structure easily. A help page for viewing epitope can be found here .
The information about epitope is valuable. So more and more databases focusing on different epitope categories become available. The existing epitope databases includes MHCPEP, SYFPEITHI, FIMM, AntiJen, SDAP, MHCBN, Bcipep and so on. These databases mainly collect MHC ligands and motifs, T cell epitopes and linear B cell epitopes, as large number of these linear epitopes have been identified in the past 20 years. There are several conformational epitopes stored in HIV Molecular Immunology Database and HCV Molecular Immunology Database; but they are limited to a single virus system. And we also noticed that there is a big project name by "The Immune Epitope Database and Analysis Resource", which will also include conformational B cell epitopes in its scope. A recent released database named by "Epitome" is very similar to CED except that it collect B cell epitope data only from PDB and we build CED from literature. Two different ways, similar result. In addition, Epitome also has information of paratope and CED also has conformational B cell epitopes identified by scanning mutagenesis, overlapping peptides, phage display other than X-ray diffraction and NMR. Anyway, we believe that a special database have a special value different from those complex database and CED may complement other existing special epitope database.
First, like all other databases, random errors will have occurred during the data accumulation, though the data was checked repeatedly. We hope you guys send your feedback to help us maintain and revise the CED database in future. Second, we will scan newly published peer view articles for well defined or refined conformational epitopes routinely. New epitopes will be added; refined epitopes will be updated. Thus, the quantity and quality of CED entry will both be further improved. Last, we have noticed that a formal ontology of epitope have been developed and suggested recently. To represent and communicate epitope information systematically and effectively, we will make the next main version of CED database complete compatible to IEDB's ontology. Other web tools are also under consideration.
Aha, it is. Great minds think alike. It makes a story. The very original version of this logo is on p31 of the famous book "Clonal Selection Theory of Acquired Immunity", which was published in 1959. The author of the book, Sir Frank MacFarlane Burnet, who won the Nobel Prize in Physiology or Medicine 1960 for discovery of acquired immunological tolerance, depicted a representative antigenic molecule showing the entire surface occupied by potential antigenic determinants with only a few active. Professor Mackay of Monash University adapted the diagram of Burnet slightly and did it in color in a review about autoepitopes published in 2004 on Autoimmunity Reviews. We like the figure very much and adapted it slightly to be the logo of CED. We thank Professor Mackay and Autoimmunity Reviews for permitting us to adapt the figure to the logo of CED.
Professor Tonegawa (left photo), who once was a student of University of Kyoto, perfectly answered the question why limited genes can generate unlimited antibodies binding to unlimited epitopes and won the Nobel Prize in Physiology or Medicine 1987 for discovery of the genetic principle for generation of antibody diversity. Nearly 50 years ago, Sir Frank MacFarlane Burnet (right photo) depicted a diagram that the entire antigen might be full of potential epitopes. However, in real case, only some part of the antigen really are epitopes in a given condition. Though we have known something on how a "epitope" can be a epitope, we still cannot completely answer the question why some parts of antigen are epitopes and others are not. It is beyond self and nonself theory. Exploring for a panorama of rules to epitope and to predict a epitope is still on the way. Databases of epitopes as we and many others have done might be helpful to the long march.
HLAB | Center of Bioinformatics | UESTC | Chengdu, 610054, China  [Feedback]