Skip Navigation


NAR Molecular Biology Database Collection entry number 893
Holliday, Gemma1; Andreini, Claudia2,3; Fischer, Julia1; Rahman, Syed1; Almonacid, Daniel4; Williams, Sophie1 and Pearson, William5
1EMBL-EBI, Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SD, UK, 1Magnetic Resonance Center (CERM), University of Florence, Via L. Sacconi 6, 1Department of Chemistry, University of Florence, Via della Lastruccia 3, 50019 Sesto Fiorentino, Italy, 1Department of Bioengineering and Therapeutic Sciences, University of California San Francisco, San Francisco, CA 94158 and 1Department of Biochemistry and Molecular Genetics, University of Virginia, Charlottesville, VA 22908, USA

Database Description

Enzymes are nature's catalysts, essential for the chemistry of life. Industrially relevant applications range from food science to laundry. We have now released MACiE, an electronic database of well-characterised enzymatic reactions. MACiE is the result of more than three years' collaboration between the Mitchell and Murray-Rust groups in the Unilever Centre and Prof. Janet Thornton at the European Bioinformatics Institute. The database contains the reaction mechanisms for 100 individual enzymes; this includes the overall reactions and the multiple steps that constitute them. We hold the data in CMLReact, a novel XML application. We are developing software for classifying enzymes in a manner complementary to the existing EC (Enzyme Commission) scheme, which has been the state-of-the-art since the 1960's.

MACiE (Mechanism, Annotation and Classification in Enzymes) is available at Version 1 of MACiE consists of a non-homologous dataset of enzymes with known structure. 100 overall reaction mechanisms comprising 358 reaction steps are included. The MACiE website contains images of every reaction step, showing each starting material, intermediate and product. Annotations describe the part each amino acid plays in each reaction step, as well as listing the mechanistic attributes. Most of the reactions can also be viewed as animations. It is even possible interactively to browse MACiE's chemical space as a HyperGraph projection using a Java applet.

Our preliminary analysis has enabled us to draw some conclusions about the roles played by the different amino acid residues in enzyme catalysis. Amino acids with alkyl or aryl side-chains have been shown to act mainly as main-chain spectators. In contrast, amino acids with side-chain chemical functionality rarely act through their main-chains. A study of the mechanisms of the 358 distinct reaction steps in MACiE reveals that amino acids accomplish nearly all their chemistry through simple acid/base catalysis - 236 of these steps are proton transfers. Nonetheless, a significant variety of mechanisms are used, including some, such as the Claisen condensation, Claisen rearrangement, Amadori rearrangement and aldol condensation, familiar from in vitro organic chemistry, and 26 steps involve radical chemistry.

Following a major revision of the annotation used in MACiE we have now a consistent way to describe the chemistry performed by enzymes in each step of the reactions they catalyse. This allows us to address classification in enzymes in a way different to that established by the EC. Whereas the latter classification considers only the overall reaction catalysed, ours is based on the mechanistic approaches undertaken by the different enzymes to perform their functions. Enzymes that catalyse completely different overall reactions can share a very similar catalytic site, and therefore the mechanism used to transform substrates into products can be similar. On the other hand, diverse enzymes can catalyse the same overall reaction by quite different sequences of steps, as we find with four distinct classes of beta lactamases. With the detailed knowledge about reaction mechanisms we have nowadays, this new approach to classifying enzymes seems a natural evolution of the traditional classification methods. Finally, amino acids that bind the substrate but play no direct catalytic role will be considered in future versions of MACiE and some interesting results are expected in this area.


1. Holliday GL, Bartlett GJ, Almonacid DE, O'Boyle NM, Murray-Rust P, Thornton JM, Mitchell JB. (2005) MACiE: a database of enzyme reaction mechanisms (2005). Bioinformatics 21:4315-4316

Go to the abstract in the NAR 2012 Database Issue.
Oxford University Press is not responsible for the content of external internet sites