(641g) Characterization of Protein-Metabolite Regulatory Interactions Via Complementary Parallel Approaches
Computational metabolic modeling plays an increasing role in both metabolic engineering and identification of therapeutic targets to combat disease. The construction of such models relies primarily on understanding of the network architecture conferred by large-scale characterization at the genomic, transcriptomic, proteomic, and increasingly, metabolomic levels. Regulatory interactions between proteins and metabolites, however, are only sparsely characterized by comparison, despite being a common phenomenon with great functional importance in cellular metabolism.
Available evidence suggests that current knowledge of these interactions encompasses only a fraction of their actual occurrence; therefore, a broad understanding of protein-metabolite regulatory interactions should confer substantial benefit in the development of predictive metabolic models. Efforts at characterization of this level of regulation face the challenge of capturing interactions that are often as weak as micromolar- to millimolar-level KDs. A systematic methodology for discovery and characterization of these protein-metabolite regulatory interactions is necessary for broad-scale understanding.
The model organism Saccharomyces cerevisiae provides an excellent subject of study for the development of this methodology, as it is an industrially important organism and has been the subject of many previous mathematical modeling efforts. Novel regulatory data for its metabolism will therefore be immediately useful for the improvement of such models. Here, we present our work in developing a multi-pronged strategy for the characterization of protein-metabolite regulatory interactions, which can be generalized to study this level of regulation in other organisms of interest. We have developed an in vitro binding assay as a medium-throughput screen for protein-metabolite binding interactions; cocktails of metabolites are incubated with known proteins, washed, and eluted prior to GC-MS analysis for enriched metabolites to identify putative regulatory pairs. These interactions may then be validated and further characterized using an in vitro reaction assay, also with GC-MS based analysis. We additionally present the in vitro reaction assay itself as a means of investigating putative regulatory interactions that are predicted by any other method. In parallel, small-molecule microarrays provide a high-throughput platform for broader-range protein-metabolite interaction discovery at the cost of sensitivity. Small-molecule microarrays have previously been applied to the discovery of strong-affinity binders for putative drug targets; we have worked toward adapting this strategy to qualitatively probe for weaker, endogenous interactions. Arrays of covalently-attached yeast metabolites are synthesized and probed with tagged, known proteins to identify protein-metabolite binding pairs, which may then be characterized via in vitro reaction assay.
The diversity of binding affinities across the scope of protein-metabolite regulatory pairs provides both challenge and opportunity; employing complementary discovery methods in parallel will allow throughput to be balanced with sensitivity. Data gathered via these complementary approaches will aid in the eventual construction of a systems-level model of S. cerevisiae metabolism.