Design and Application of Conditionally Activated Split Inteins

Abstract

The discovery and characterization of inteins has led to the development of powerful applications capable of modulating protein structure and function through protein splicing. However, while inteins have found great utility in-vitro with protein semi-synthesis applications, their use as tools to modify proteins in living systems has been limited by the-auto catalytic nature of the splicing reaction, which makes temporal control of their activity difficult. To address this issue, a variety of conditional protein splicing (CPS) methods have been developed to try and regulate the splicing reaction so that intein activity can be activated only upon addition of a specified trigger. However, these CPS methods have largely relied on inteins that suffer from a variety of unfavorable characteristics that constrain their broader utility, including slow splicing kinetics, poor stability, and stringent extein dependency. In contrast, the recently characterized fast splicing split inteins such as the DnaE intein Npu, are less susceptible to these limitations. Therefore, CPS methods based on naturally split fast splicing inteins should help to make in-cell intein based tools more broadly accessible. This thesis attempts to address this issue by utilizing a basic mechanistic understanding of how the split intein Npu assembles and splices to develop intein zymogens, the first CPS method that effectively controls the association and splicing of multiple fast splicing split inteins while remaining amenable to a diverse set of triggers. Utilizing our understanding of how the caged inteins function, we then developed a general method that can be easily and effectively applied to increase expression and functionality of split N-inteins and split protein-intein fusions, which are otherwise prone to aggregation. Lastly, we further engineer the intein zymogens so that close proximity of the caged intein fragments can be used to overcome splicing inhibition, paving the road for applications that have both temporal and spatial control over protein function. In combination, these methods should make it possible to develop previously impractical or intractable intein based applications and generally expand their utility in cells.