Fig. 1: Split intein system for in vivo detection of protein–protein interactions.

a A schematic is shown for the detection of a binding event between a RiPP and target. A split intein from Nostoc punctiforme PCC73102 (Npu), is used to splice together a σ factor for downstream transcription. b Inducible interaction-mediated splicing. The median fluorescence is shown as a function of the expression of the two halves of the sensor proteins. The induction of PMI- and Mdm2*-driven association (left) or split intein alone (right) are shown. The plots are a representative replicate selected from three performed on different days. c The specificity is calculated using the data in part b: (Fluorescence^Mdm2*+PMI/Fluorescence^{no bait+no peptide}). The white dot marks the highest fold-change in expression. d The fluorescence measured from the circuit containing a binding pair (Mdm2*:PMI) and non-binding pairs are compared. The 3OC6-AHL concentration for all inductions was 1 µM. Three replicates performed on different days are shown. e A schematic of the RiPP-containing half of the split intein system is shown, with the modified core residues in blue. f The structure of the wild-type PapA modified peptide⁵⁶ is shown with a box around the region used to design the peptide library and the resulting scaffold sequence. g Library sequence weblogo for the 14 modified library variants observed (Methods).