Novel Type V-A CRISPR Effectors Are Active Nucleases with Expanded Targeting Capabilities

Abstract

Cas12a enzymes are quickly being adopted for use in a variety of genome-editing applications. These programmable nucleases are part of adaptive microbial immune systems, the natural diversity of which has been largely unexplored. Here, we identified novel families of Type V-A CRISPR nucleases through a large-scale analysis of metagenomes collected from a variety of complex environments, and developed representatives of these systems into gene-editing platforms. The nucleases display extensive protein variation and can be programmed by a single-guide RNA with specific motifs. The majority of these enzymes are part of systems recovered from uncultivated organisms, some of which also encode a divergent Type V effector. Biochemical analysis uncovered unexpected protospacer adjacent motif diversity, indicating that these systems will facilitate a variety of genome-engineering applications. The simplicity of guide sequences and activity in human cell lines suggest utility in gene and cell therapies.

FIG. 1.

Diversity of CRISPR Type V-A effectors. (A) Per family distribution of taxonomic classification of contigs encoding the novel Type V-A and V-L effectors. (B) Phylogenetic tree inferred from an alignment of 119 novel Type V-A and V-L (colored branches) and 89 reference (gray branches) sequences. Cas12a-M families are denoted by numbers in parentheses. Protospacer adjacent motif (PAM) requirements for active nucleases described here are outlined within boxes colored by family. Non-Type V-A reference sequences were used to root the tree. *Cas12a-M61 family requires a crRNA with an alternative repeat motif.

FIG. 2.

Type V-A effectors are active nucleases. (A) PAM SeqLogo determined for six Type V-A nucleases. (B) Boxplot of plasmid transfection activity assays inferred from frequency of indel edits for active nucleases. The boundaries of the boxplots indicate first and third quartile values. The mean is indicated with an “x” and the median is represented by the midline within each box. (C) Plasmid transfection editing frequencies at four target sites for Cas12a-M29-1 versus AsCpf1. Side-by-side comparisons represent one experiment. (D) Plasmid versus ribonucleoprotein (RNP) editing activity for nuclease Cas12a-M29-1 at 14 target loci. (E) Editing profile of nuclease Cas12a-M29-1 from RNP transfection assays. Editing frequency experiments in (B), (D), and (E) were done in duplicate. The bar plots in (C) and (D) show mean editing frequency with one standard deviation error bars (D).