PROTEIN DOMAIN TRAPPING (PDT)
Structural biology is a critical part of the modern process of drug design and obtaining high expression of soluble, well behaved protein is essential. Standard bioinformatics approaches to design constructs for E. coli expression may not always be appropriate and often constructs selected based on sequence homology to a known structure or gene conservation are not successful. To overcome this limitation, ZoBio has implemented Protein Domain Trapping (PDT, Cabantous and Waldo 2006) in its workflow, which enables selection of highly expressing soluble proteins from a randomly generated library of gene fragments.
The Protein Domain Trapping procedure relies on the self-complementing split GFP assay (see figure below). Following several steps to select DNA fragments that are in frame, have the correct orientation and fall within a certain size range, the DNA fragments are fused to a sequence encoding a 15 amino-acid GFP fragment, GFP 11. When the GFP 11 fusion protein is co-expressed with the non-fluorescent GFP 1-10 detector protein, they spontaneously associate to form functional fluorescent GFP, as summarized schematically in the figure. Reconstitution of fluorescence only occurs when the GFP 11 fusion protein is well expressed and remains soluble. By screening for fluorescent clones from a pool of transformants, the soluble regions in the protein of interest can be selected and identified.
Schematic of reconstitution of GFP fluorescence, by self-association
of the GFP 1-10 detector and the GFP 11 tagged fusion protein.
Since E. coli is used as the host, the results are immediately translatable to NMR structural studies. Naturally the proteins are also amenable to X-ray crystallography, allowing more shots on the structure goal, thereby rendering drug discovery projects more robust.