Selection and evaluation of new sites for splitting beta-lactamase to modulate auto-complementation enzyme activity

Split protein systems, including split fluorescent proteins and split enzymes, have predominantly been applied to monitor protein-protein interactions but have recently gained popularity as components in induced complementation diagnostic platforms. Many of these systems are limited by self-affinity which limits the dynamics or causes a false-positive high background signal. To overcome these challenges, it is necessary to either modify the interacting residues of established split fragments or by exploring new split-sites. Using the computational modeling tool SPELL to identify stable candidates, we developed five unique split sites for TEM-1 β-lactamase to compare structure and activity versus the original split site. We monitored structural changes through MMS IR spectroscopy as well as functional activity of each split enzyme. Expected structural changes of each fragment and changes in self-affinity driven complementation were observed. By utilizing specific binding proteins attached to the split enzyme, we monitored enzyme activity with forced complementation when bound to the target compared to auto-complementation when the enzymes were auto-complexed in solution. Of the five new split sites, one maintained high binding-mediated complementation activity with significant reduction in background auto-complementation. Changing the split site is a relatively simple approach to reduce background activity while maintaining on-target activity. The ability to tune split-systems has the potential to enable more sensitive and complex studies that previously were limited by the high background from self-affinity.