Developing methods to understand and engineer protease cleavage specificity

Abstract

Proteases are ubiquitous enzymes that comprise nearly 2% of all human genes. These robust enzymes are attractive potential therapeutics due to their catalytic turnover and capability for exquisite specificity. While most existing drugs require a stoichiometric ratio to function, therapeutic proteases could clear their targets much more efficiently. Unfortunately, existing technologies are inadequate for understanding and engineering therapeutic proteolytic specificities. My thesis work has focused on building the groundwork to enable these technologies to thrive. For the goal of engineering a new protease, it is currently necessary to identify prototype proteases for engineering efforts that have specificities similar to the desired target substrate. Current technologies are unable to characterize proteases adequately for this goal. Accordingly, I invested in developing a method for the accurate characterization of protease cleavage specificity. Our unique combination of mRNA display technology, Next-Generation Sequencing, and mass spectrometry enables the sampling of all possible permutations of octamer substrates and the identification of millions of cleavage sites. The throughput of our approach is orders of magnitude greater than the current state-of-the-art methods. The resulting high-resolution specificity maps can be applied to identify promising protease prototypes, predict human cross-reactivity, or lead to a better understanding of this critical component of natural physiology. In the work presented here, I applied my new specificity-screening method to assess the specificities of the proteases factor Xa, ADAM17, and streptopain. The resulting cleavage preference maps confirmed known specificities, and revealed new insight into the broad preferences of both narrow- and broad-specificity proteases. In particular, disfavored amino acids were illuminated better than ever before. The next focus of my work was to engineer multiple-subsite novel protease specificity. I chose streptopain as the prototype for my efforts to neutralize the superantigen exotoxin SpeA. I identified a target loop of SpeA wherein cleavage would result in inactive fragments. Further, I confirmed that streptopain can be successfully presented as an mRNA displayed fusion. In summary, my thesis work established crucial methodologies for applying mRNA display technology to enable the understanding and ultimately engineering the specificity of therapeutic proteases.