Small molecule-dependent genetic selection in stochastic nanodroplets as a means of detecting protein-ligand interactions on a large scale.

BACKGROUND: Understanding the cellular role of a protein often requires a means of altering its function, most commonly by mutating the gene encoding the protein. Alternatively, protein function can be altered directly using a small molecule that binds to the protein, but no general method exists for the systematic discovery of small molecule ligands. Split-pool synthesis provides a means of synthesizing vast numbers of small molecules. Synthetic chemists will soon be able to synthesize natural product-like substances by this method, so compatible screening methods that detect the activity of minute quantities of molecules among many inactive ones will be in demand.

RESULTS: We describe two advances towards achieving the above goals. First, a technique is described that uses a simple spray gun to create 5000-8000 droplets randomly, each having a volume of 50-200 nanoliters. The individual 'nanodroplets' contain a controlled number of cells and many also contain individual synthesis beads. As small molecules can be photochemically released from the beads in a time-dependent manner, the concentration of ligands that the cells are exposed to can be controlled. The spatial segregation of nanodroplets prevents the mixing of compounds from other beads so the effects of each molecule can be assayed individually. Second, a small molecule-dependent genetic selection involving engineered budding yeast cells was used to detect intracellular protein-ligand interactions in nanodroplets.

CONCLUSIONS: The technique described here should facilitate the discovery of new cell-permeable ligands, especially when combined with a positive selection assay that detects intracellular binding of small molecules to proteins. Using 'anchored combinatorial libraries', it may be possible to screen entire libraries of natural product-like molecules against the entire collection of proteins encoded within cDNA libraries in a single experiment.