A quantitative model for the dynamics of target recognition and off-target rejection by the CRISPR-Cas Cascade complex

Abstract

CRISPR-Cas effector complexes recognise nucleic acid targets by base pairing with their crRNA which enables easy re-programming of the target specificity in rapidly emerging genome engineering applications. However, undesired recognition of off-targets, that are only partially complementary to the crRNA, occurs frequently and represents a severe limitation of the technique. Off-targeting lacks comprehensive quantitative understanding and prediction. Here, we present a detailed analysis of the target recognition dynamics by the Cascade surveillance complex on a set of mismatched DNA targets using single-molecule supercoiling experiments. We demonstrate that the observed dynamics can be quantitatively modelled as a random walk over the length of the crRNA-DNA hybrid using a minimal set of parameters. The model accurately describes the recognition of targets with single and double mutations providing an important basis for quantitative off-target predictions. Importantly the model intrinsically accounts for observed bias regarding the position and the proximity between mutations and reveals that the seed length for the initiation of target recognition is controlled by DNA supercoiling rather than the Cascade structure.

Dynamic interplay between target search and recognition for the Cascade surveillance complex of type I-E CRISPR-Cas systems

“…Above, we determined the target recognition probability by modeling the target search efficiency. However, it can also be directly predicted using the target recognition model based on the discrete 1D energy landscape described in the introduction 16,17,27,29 (Fig. 1b; Supplementary Fig.…”

“…Incorporating an additional locking step into the random-walk model for target recognition and assuming a base pair stepping rate K bp (see Fig. 1a; Supplementary Methods 7) in the millisecond range 17 , we could obtain a prediction for the target recognition probability p recog as function of torque that included locking (Fig. 4e, black line).…”

“…Supplementary Methods 7) in the millisecond range 17 , we could obtain a prediction for the target recognition probability 𝑝 𝑟𝑒𝑐𝑜𝑔 as function of torque that included locking (Fig. 4e, black line).…”

“…The resulting structure is called an R-loop 14,15 . R-loop formation is achieved by unwinding the DNA from the PAM towards the PAM-distal end of the target in a reversible process driven by thermal fluctuations [16][17][18] . For St-Cascade, the R-loop is 32 base pairs (bp) long and the PAM corresponds to an AAN motif.…”

“…A rapid reversibility of R-loop formation, thus, circumvents stalling of the protein complex at partially matching sites to allow rapid scanning of all potential sites during target search. Reversible R-loop formation, and similarly other strand exchange reactions, have been previously modelled as a random walk of 1 bp steps within a simplified one-dimensional energy landscape, starting at the PAM and ending in the locked state 17,[27][28][29] . Within this model, mismatches introduce local energy barriers, while negative supercoiling (favoring DNA unwinding and thus R-loop formation) provides a global negative bias of the energy landscape (Fig.…”