Email updates

Keep up to date with the latest news and content from BMC Evolutionary Biology and BioMed Central.

Open Access Highly Accessed Open Badges Research article

Evolutionary constraints permeate large metabolic networks

Andreas Wagner

Author Affiliations

University of Zurich, Dept. of Biochemistry, Bldg. Y27, Winterthurerstrasse 190 CH-8057 Zurich, Switzerland

Department of Biology, University of New Mexico, Albuquerque, New Mexico, USA

The Santa Fe Institute, Santa Fe New Mexico, USA

The Swiss Institute of Bioinformatics, Quartier Sorge - Batiment Genopode, 1015 Lausanne, Switzerland

BMC Evolutionary Biology 2009, 9:231  doi:10.1186/1471-2148-9-231

Published: 11 September 2009



Metabolic networks show great evolutionary plasticity, because they can differ substantially even among closely related prokaryotes. Any one metabolic network can also effectively compensate for the blockage of individual reactions by rerouting metabolic flux through other pathways. These observations, together with the continual discovery of new microbial metabolic pathways and enzymes, raise the possibility that metabolic networks are only weakly constrained in changing their complement of enzymatic reactions.


To ask whether this is the case, I characterized pairwise and higher-order associations in the co-occurrence of genes encoding metabolic enzymes in more than 200 completely sequenced representatives of prokaryotic genera. The majority of reactions show constrained evolution. Specifically, genes encoding most reactions tend to co-occur with genes encoding other reaction(s). Constrained reaction pairs occur in small sets whose number is substantially greater than expected by chance alone. Most such sets are associated with single biochemical pathways. The respective genes are not always tightly linked, which renders horizontal co-transfer of constrained reaction sets an unlikely sole cause for these patterns of association.


Even a limited number of available genomes suffices to show that metabolic network evolution is highly constrained by reaction combinations that are favored by natural selection. With increasing numbers of completely sequenced genomes, an evolutionary constraint-based approach may enable a detailed characterization of co-evolving metabolic modules.