Native or promiscuous? Analyzing putative dimethylsulfoniopropionate lyases using a substrate proofing approach

Enzyme promiscuity is widely spread. Foremost, within superfamilies, the native function of one enzyme is typically observed as promiscuous activity in related enzymes. The native function usually exhibits high catalytic efficiency while promiscuous activities are weak, but this is not always the case. Thus, for certain enzymes it remains questionable whether their currently known activity is native or promiscuous. Dimethylsulfoniopropionate (DMSP) is an abundant marine metabolite cleaved via beta-elimination to release dimethylsulfide (DMS). Eight different gene families have been identified as putative DMSP lyases, 5 of them belonging to the same superfamily (cupin-DLL; see the accompanying paper). Some of these enzymes exhibit very low activity, but this can be due to suboptimal folding or reaction conditions. We developed a substrate profiling approach with the aim of distinguishing native DMSP lyases from enzymes that promiscuously act as DMSP lyases. In a native DMSP lyase, relatively small changes in the structure of DMSP should induce significant activity drops. We thus profiled substrate selectivity by systematically modifying DMSP while retaining reactivity. Three enzymes that exhibit the highest activity with DMSP also exhibited high sensitivity to perturbation of DMSP structure (Alma, DddY, and DddL). The two enzymes with the weakest DMSP lyase activity also showed the highest crossreactivity (DddQ, DddP). Combined with other indications, it appears that the DMSP lyase activity of DddQ and DddP is promiscuous although their native function remains unknown. Systematic substrate profiling could help identify and assign potential DMSP lyases, and possibly applied to other enzymes.