CRISPR-Cas systems confer bacteria adaptive immunity against invading pathogens through sequence-specific recognition and cleavage of foreign nucleic acids. Reprogrammed CRISPR-Cas effectors offer a great opportunity to silence tumour-associated genes at the DNA or RNA level. CRISPR-Cas13 was recently reported to targets single-stranded RNAs with high efficacy and specificity. However, the molecular bases that govern Cas13 targeting remain largely unknown, which limits future therapeutic use of this promising tool to silence tumour RNAs.
To investigate Cas13b target recognition and silencing, we developed innovative library screens in mammalian cells where the spacer sequence of CRISPR RNA (crRNA) is designed to target several reporter transcripts with single-nucleotide increments. We built in-house bioinformatic analysis pipelines to comprehensively investigate hidden parameters that govern the silencing efficiency in datasets obtained from 200 crRNAs. The single-nucleotide resolution data revealed that Cas13b is not constrained by any protospacer-flanking sites that are commonly associated with other CRISPR systems, and therefore highlights its high design flexibility. On the other hand, we revealed a GG motif at the 5’ end of the spacer that greatly enhance Cas13b silencing efficiency. To further validate the importance of these parameters, we used our prediction model for de novo design of highly potent crRNAs targeting several other mRNAs, and demonstrated high accuracy of our prediction. Together, this single-nucleotide screen and bioinformatic analyses revealed how to reprogram Cas13b for efficient gene silencing.
Furthermore, we explored Cas13b target specificity through investigating spacer-target interaction at the single-nucleotide level. Our comprehensive mutagenesis analysis showed that Cas13b is highly specific since mismatches greater than three nucleotides destabilized spacer-target interaction and largely impaired Cas13b silencing in mammalian cells. Finally, we leveraged these rules to reprogram Cas13b to silence gene fusion tumour drivers and SARS-CoV-2 transcripts with high efficiency and specificity (Hu et al, in preparation; Fareh et al, Nature Commun, 2021).