Discovery of Diverse CRISPR-Cas Systems and Expansion of the Genome Engineering Toolbox

Abstract

CRISPR systems mediate adaptive immunity in bacteria and archaea through diverse effector mechanisms and have been repurposed for versatile applications in therapeutics and diagnostics thanks to their facile reprogramming with RNA guides. RNA-guided CRISPR-Cas targeting and interference are mediated by effectors that are either components of multisubunit complexes in class 1 systems or multidomain single-effector proteins in class 2. The compact class 2 CRISPR systems have been broadly adopted for multiple applications, especially genome editing, leading to a transformation of the molecular biology and biotechnology toolkit. The diversity of class 2 effector enzymes, initially limited to the Cas9 nuclease, was substantially expanded via computational genome and metagenome mining to include numerous variants of Cas12 and Cas13, providing substrates for the development of versatile, orthogonal molecular tools. Characterization of these diverse CRISPR effectors uncovered many new features, including distinct protospacer adjacent motifs (PAMs) that expand the targeting space, improved editing specificity, RNA rather than DNA targeting, smaller crRNAs, staggered and blunt end cuts, miniature enzymes, promiscuous RNA and DNA cleavage, etc. These unique properties enabled multiple applications, such as harnessing the promiscuous RNase activity of the type VI effector, Cas13, for supersensitive nucleic acid detection. class 1 CRISPR systems have been adopted for genome editing, as well, despite the challenge of expressing and delivering the multiprotein class 1 effectors. The rich diversity of CRISPR enzymes led to rapid maturation of the genome editing toolbox, with capabilities such as gene knockout, base editing, prime editing, gene insertion, DNA imaging, epigenetic modulation, transcriptional modulation, and RNA editing. Combined with rational design and engineering of the effector proteins and associated RNAs, the natural diversity of CRISPR and related bacterial RNA-guided systems provides a vast resource for expanding the repertoire of tools for molecular biology and biotechnology.

Figure 1

CRISPR diversity of class 1. This diagram illustrates the CRISPR-cas loci of each class 1 subtype and their distinct variants along with a dendrogram (left) showing the possible evolutionary relationships among the different subtypes. Labels on the right indicate the example organisms and gene ID ranges. Homologous genes across the different subtypes are color coded to indicate their evolutionary relationships. Gene names are derived from previous classifications of CRISPR systems.^, The small subunit involves csm2, cmr5, cse2, csa5, and several other homologous genes collectively denoted cas11. Dotted lines around cas1 and cas2 in subtype III-A and III-E systems indicate they are dispensable genes. Gene regions highlighted in cream and red represent the HD nuclease domain. The purple shading shows the effector module. Genes without functional characterization are colored gray. CRISPR arrays are shown on loci when detected near these subtype systems, although there are exceptions detected for each subtype. RT indicates reverse transcriptase. TPR indicates a tetratricopeptide repeat. Modern gene names are shown in bold with legacy names shown below. Adapted from ref (10).

Figure 2

CRISPR diversity of class 2. The diagram depicts representative CRISPR-cas loci for each class 2 subtype, including selected distinct variants. The dendrogram on the left illustrates the likely evolutionary relationships among the types and subtypes. The labeling on the right depicts the example host genome and gene ID ranges. Homologous genes are color-coded, and gene names are derived from previous classifications of CRISPR systems.^, The gray shading with different hues represents the two levels of classification: subtypes and variants. Dotted lines around adaptation genes cas1 and cas2 indicate they are not present in many systems. The genes encoding WYL domains and csx27 are also dispensable and shown by dashed lines. Additional genes encoding components of the interference module, such as transactivating CRISPR RNA (tracrRNA), are displayed. The effector protein domains are color-coded as follows: cream for RuvC-like nuclease, yellow for HNH nuclease, navy for nucleotide binding (HEPN) RNase of higher eukaryotes and prokaryotes, and blue for transmembrane domains. Modern gene names are shown in bold with legacy names shown below. Adapted from ref (10).

Figure 3

Pipeline for the computational discovery of class 2 CRISPR-Cas systems. A list of coordinates or protein sequences can be used as seeds for the search. Protein coding sequences within 10 kb of either side of the seeds are extracted and then annotated using known protein profiles. Both annotated proteins and unknown ORFs are then clustered and ranked for enriched association with seeds. Top candidates are selected, and the process can be repeated for expansion of the candidate families. Examples of this pipeline have been used in previous CRISPR system searches.^,

Figure 4

Mammalian DNA- and RNA-targeting applications of class 2 Cas programmed nucleases. Select class 2 families are represented in three ways: (1) genomic context within representative CRISPR loci, (2) structure, with target DNA and guide RNA colored, and (3) applications when used with nuclease, nicking, or binding activities. Protein Data Bank references are Cas9 (4OO8), Cas12a (5B43), Cas12b (5U30), and Cas13a (5XWP).

Figure 5

Emerging applications of CRISPR nucleases. (A) Miniature RNA-guided nucleases, such as CRISPR-Cas12f and the ancestral CRISPR proteins IscB, IsrB, and TpnB, can be used for genome editing. (B) Emerging programmable gene integration technologies that rely on CRISPR targeting include PASTE, which uses a Cas9-RT-Bxb1 integrase fusion to direct insertion of large payloads, and the CRISPR-associated transposons from type I-F and V-K systems. (C) New RNA-targeting technologies are safer with either no collateral activity, such as Cas7-11, or minimal collateral activity, such as with engineered or novel natural variants.

Figure 6

Nucleic acid detection using collateral RNase activity of Cas13. (A) CRISPR diagnostics, when paired with preamplification chemistries like RPA or LAMP, can allow for detection of single molecules of DNA or RNA by leveraging the Cas12 or Cas13 collateral activity for reporter-based detection. (B) Many readouts for CRISPR diagnostics have been developed, including by fluorescence, visual detection, colorimetric, lateral flow strips, and electrochemical modalities.