Engineering the Protease Substrate Specificity of Human Kallikrein 7

     The ability to redesign the specificity of a protease to cleave a target substrate and concurrently minimize its recognition of undesirable substrates would be a powerful tool.  Most proteases are thought to have ~10-100 physiological substrates, although the mechanisms by which proteases selectivity recognize their substrates in vivo are not well understood.  The focus of my research is to create a method for engineering target-specific proteolytic enzymes.  As a model system, I will apply these methods to engineer the human protease kallikrein 7 (hK7) to selectively cleave the hydrophobic region of amyloid beta (Aβ) peptide.  The hydrophobic core of Aβ was chosen as the target substrate since its aggregation has been linked to Alzheimer’s pathology, therefore a protease that selectively decreases Aβ could serve as a potential therapeutic.

     We have designed a microbial cell based assay for screening protease libraries against a target substrate sequence and counterselecting for non-cleavage of a non-target sequence using fluorescence activated cell sorting.  Initial attempts to produce hK7 in the periplasm of E. coli were unsuccessful due to the extreme inefficiency of bacteria to form the six native disulfide bonds in hK7.  Therefore, we have designed and constructed a system in Saccharomyces cerevisiae for the expression of active hK7.  Initial hK7 expression levels were determined using an hK7-yellow fluorescent protein (YPet) fusion by monitoring yellow fluorescence inside the cell using flow cytometry.  Expression of correctly folded, active hK7 was detected in the cell lysate using an hK7 substrate FRET probe and confirmed through Western blot.