Robbins and Hartwig have turned catalyst discovery in organic chemistry into a high throughput process. This process uses 96 well plates, into which 17 diverse reactants were added. Theoretically, a much higher number could have been used. Into each well, a different metal/ligand combination are added. After incubation, mass spectrometry was used to characterize the mixtures. To reduce the number of samples that need to be analyzed by MS, the contents of each row were added (8 samples) and the contents of each column were added (12 samples). If the same new ion was seen in the column 1 sample and the row A sample, it was evident that the reaction took place in A1. This effectively speeds up screening ~5 fold (96 samples to 20). In some cases further deconvolution was necessary.
They tested this method to "discover" known reactions, and in each case the catalyst was identified. They also discovered some new reactions, including a copper-catalysed alkyne hydroamination and two nickel-catalysed hydroarylation reactions. I think this is an exciting new approach to catalyst discovery.
I realize this isn't really a chemical biology or biomolecular engineering paper, but I think it's still a good fit for this blog.
ReplyDeleteFirst of all, bioorthogonal chemical reactions play an important role in chemical biology, especially for metabolite and protein labeling. This screening technique could be used to search for new bioorthogonal reactions. One would simply need to choose substrates with functional groups that are relatively biologically inert, use metals with low cellular toxicity, and do the reactions under biological conditions.
Secondly, this is inherently a combinatorial approach to catalyst discover, and combinatorial chemistry can be thought of as applied molecular evolution. With this in mind, the relationship to biomolecular engineering becomes apparent.