Harnessing Type I CRISPR–Cas Systems for Human Genome Engineering
Peter Cameron1, Mary Coons1, Sanne E. Klompe1, Alexandra M. Lied1, Stephen Smith1, Bastien Vidal1, Paul D. Donohoue1, Tomer Rotstein1, Bryan Kohrs1, David B. Nyer1, Rachel Kennedy1, Tim Künne2, John van der Oost2, Stan J. J. Brouns2,3, Euan M. Slorach1, Chris K. Fuller1, Scott Gradia1, Steven B. Kanner1, Andrew P. May1, and Samuel H. Sternberg1
1 Caribou Biosciences, Inc., Berkeley, CA, USA. 2 Laboratory of Microbiology, Department of Agrotechnology and Food Sciences, Wageningen University, Wageningen, The Netherlands. 3 Kavli Institute of Nanoscience, Department of Bionanoscience, Delft University of Technology, Delft, The Netherlands.
Type I CRISPR-Cas systems are the most abundant form of prokaryotic adaptive immunity found in nature, comprising ~50% of systems detected in sequenced bacterial and archaeal genomes. Target interference in Type I systems relies on a multi-subunit, RNA-guided complex called Cascade, which binds complementary double-stranded DNA and then recruits a trans-acting helicase-nuclease, Cas3, for target degradation. While many hallmark features of DNA targeting are functionally conserved between Cascade and the Cas9 effector in Type II systems, Type I systems have thus far been overlooked for eukaryotic genome engineering applications, due in large part to the relative difficulty of heterologous expression of the Cascade complex, and to the distinct way in which DNA is cleaved. Here we show that by using fusions between Cascade and the dimerization-dependent, non-specific FokI nuclease domain, we can achieve programmable, RNA-guided gene editing with Type I systems in human cells. First, we demonstrated that functional FokI-Cascade can be directly transfected as intact ribonucleoprotein (RNP) complexes or assembled in cellsvia delivery of individual plasmid-encoded components. Second, we combined all the CRISPR-associated (Cas)genes onto a single polycistronic vector, giving us a simplified “two-component” effector-guide expression system similar to the widely used Cas9-sgRNA platform. Third, we optimized various parameters of our system, including the FokI linker sequence, target DNA geometry, and Cascade homolog choice, to achieve editing efficiencies up to ~50%. Finally, we systematically explored both the PAM requirements and mismatch sensitivity during DNA targeting, revealing key differences between well-studied Cas9–RNA complexes and the FokI–Cascade nuclease platform. Taken together, our work highlights the vast potential to harness abundant, previously untapped Type I CRISPR–Cas systems for genome engineering applications in eukaryotic cells.
