Integrating Proteomics Techniques with Transcriptome and Targeted Gene and Metabolite Profiling
Large-scale investigations of proteins using techniques such as shotgun proteomics provide an additional set of tools for enzyme and pathway discovery. When transcriptome datasets (e.g., large-scale EST or RNA-seq) are coupled with matching proteome datasets of the same tissues/treatments, the combined transcriptome-proteome dataset enables the identification of expressed proteins in the metabolically active tissue of interest. This approach has the potential to detect critical enzymes and pathways by finding correlations between candidate gene expression, observed enzyme activities, and target metabolite profiles.
Work on the biosynthesis of methylcinnamate in sweet basil (Kapteyn et al. 2007) offers a good example of this approach. In this case, four basil varieties were used for comparisons—the MC variety that accumulates high levels of methylcinnamate and the three varieties (i.e., EMX, SW, and SD) that produced little to none of this compound. Methylcinnamate biosynthesis had been hypothesized to involve either MT and/or an acyltransferase activities. Of the two possible routes, cell- free protein extracts from the leaves showed cinnamic acid carboxyl methyltransferase (CCMT) activities, thus supporting the route involving MT enzymes. Formation of methylcinnamate was readily detectable when extracts were incubated with cinnamoyl-CoA (or labeled cinnamate) and S-adenosyl methionine while labeled methylcinnamate remained undetected when incubated with cinnamoyl-CoA and labeled methanol. In this context, comparative transcriptomics of glandular trichomes indicated salicylic acid carboxyl methyltransferase (SAMT)-like proteins that were abundant in MC, but absent in the other three varieties, as candidates.
Further transcriptome analysis narrowed the focus to three highly expressed genes belonging to the SABATH family of carboxyl MTs. The corresponding proteins were confirmed to be expressed in the glandular trichomes using shotgun proteomics and their abundance correlated well with observed gene expression and CCMT activity. Each candidate gene was expressed in
E. coli and tested with 20 potential substrates including cinnamate and other structurally related molecules. The best substrates for methylation by all three tested recombinant proteins were cinnamate and p-coumarate and thus designated as cinnamate/p-coumarate carboxyl methyltransferases (ObCCMTl—3) (Kapteyn et al. 2007).