Direct Selection of Synthetic Binding Proteins That Recognize Post-Translationally Modified Proteins in Living Cells

Protein phosphorylation plays an important role in the regulation of protein function and many cellular processes. Aberrant phosphorylation has been shown to be a cause of cell death as well as maligination. The inhibition of aberrant phospho-modification of proteins is based on small molecules, which target the active site of kinases that are highly conserved; therefore, specificity is hard to achieve. As such, designing therapeutic antibodies which can specifically target phosphorylated targets holds much clinical promise. Currently, generation of phospho-specific antibodies relies primarily on hybridoma technology. Phospho-epitope mapping through mass spectrometry, selection of the phospho-epitope to be targeted, and synthesis of a short phosphopeptide to be injected are all performed before animal immunization. As an alternative to immunization, protein display technologies (e.g., phage, yeast, and ribosome display) have been employed. Though up-front phospho-amino acid identification is omitted in protein display technologies, purification of kinases is still required. To address the shortcomings of conventional antibodies, numerous groups have developed intracellular antibodies (intrabodies) that function inside cells; however, a major obstacle for intrabody development is that the reducing environment of the cytoplasm inhibits the formation of disulfide bridges, which are essential for proper folding of the antigen-binding site of an antibody. Hence, for most antibodies derived from hybridomas or isolated from synthetic libraries, the antigen binding capacity is diminished or even completely lost following expression as intrabodies in the cytosolic environment. To overcome the shortcomings of existing technologies and to rapidly identify intrabodies or other binding proteins that function intracellularly, we recently developed a novel technology termed FLI-TRAP (functional ligand-binding identification by Tat-based recognition of associating proteins). FLI-TRAP is based upon the hitchhiker mechanism of the bacterial twin-arginine translocation (Tat) system to efficiently co-localize non-covalent complexes of two correctly folded polypeptides to the periplasmic space of E. coli. Here, we extended this FLI-TRAP to mature the binding affinity of a binding molecule called DARPins targeting phosphorylated human ERK2 which is a potential target for cancer therapy. Following just a single round of survival-based enrichment using FLI-TRAP, this FLI-TRAP system has yielded a number of candidate binding proteins that specifically recognize the oncogenic phosphoprotein pERK2 but not their non-phosphorylated counterparts. Based on cellular assay, the candidate binding proteins possess superior traits simply by demanding bacterial growth on high concentrations of antibiotic. Further characterization of these mutants will be required to confirm maturation. Collectively, our results illustrate that FLI-TRAP has great potential to be an attractive alternative to classic animal-based technologies, and protein display technologies for the selection of phospho-specific binding molecules, all without the need for purification or immobilization of the binding target and its kinase.