Extending Metabolic Networks by in vitro Metabolomics

Most of the current metabolic networks in microbes end up somewhere around 1000 enzymes with highly-specific reactions. While there are still some gaps, the impression manifests that we are nearing completion. By using our high-throughput metabolomics approach on proteome-wide collections of purified proteins, we are asking whether how much of metabolism is currently not covered in out models. Specifically we follow two questions: i) how many yet unknown enzymatic reactions hide in the large fraction of still not annotated proteins and ii) how specific are enzymes really. For each of these questions we purified >1000 proteins from the E. coli ASKA collection. In the first case, we purified 1275 proteins of unknown function (ie y-genes) and determined their in vitro ability to catalyse metabolic reactions. In the second case, we investigated promiscuity of >1000 purified E. coli enzymes whose functions are known. We find a surprisingly large number of promiscuous enzymes and assess their evolutionary potential by FBA. During this talk I will present the results from both of these so far unpublished studies.