Efficiency and Outcomes of Sequence-Specific Self-Targeting in Streptococcus Thermophilus Utilizing Native CRISPR-Cas Type I-E and Type II-a Systems

CRISPR-Cas systems have gained recognition due the specificity of their RNA-mediated, nuclease-dependent targeting, to provide adaptive immunity against phages. This machinery has been repurposed in eukaryotes which has revolutionized the means and potential for genome editing. These systems hinge on Cas effector nucleases to target and generate sequence-specific DNA damage via a double-stranded DNA break (Cas9) or substantial degradation of ssDNA (Cascade-Cas3). Due to notoriously limited DNA repair pathways in bacteria, CRISPR-Cas systems can be repurposed for lethal sequence-specific targeting, as an innovative form of precision antimicrobials. Our objective was to compare the relative killing efficiencies of endogenous Type I and Type II CRISPR-Cas systems in the model organism Streptococcus thermophilus DGCC7710. The hypothesis was that endogenous targeting via Cas3-mediated exonucleolytic DNA damage would elicit more deleterious effects than Cas9-directed endonucleolytic processing. Targeting of lacZ by delivering plasmids carrying CRISPR arrays with targeting spacers repurposing either the native Type I-E or Type II-A systems. We observed efficient killing in both cases, in a dose-dependent manner when delivering 0.4-400ng of plasmid DNA. We next characterized survivors from both targets via PCR screening and sequencing to determine the genetic basis enabling survival. Results showed that Type I-E survival was primarily the result of low-frequency targeting-defective plasmids that lacked the targeting spacer, likely generated via loss through homologous recombination between CRISPR repeats. In contrast, Type II-A survivors had mutations in the chromosomal target. Surprisingly, we observed the excision of a 35 kB genomic segment, putatively generated via homologous recombination between flanking UDP-glucose-4-epimerase sequences (~250 bp identity) to circumvent self-targeting. Future efforts will determine whether Type I-based targeting generates more extensive damage that could yield more potent antimicrobials.