C2c2 (green) or dCas9 with a PAM-providing DNA oligo (orange) and their respective guide RNAs (gray) bring effector proteins (yellow) to RNA (blue).

Adapting the CRISPR bacterial immune system to eukaryotic genomes has enabled an unprecedented ability to modify DNA from disabling to swapping or tagging genes, to introducing specific point mutations, to enhancing or repressing select genes' activities and carrying out genome-wide functional screens. In the CRISPR system, a short guide RNA (gRNA) molecule steers an endonuclease to a target sequence complementary to the gRNA and located next to a protospacer-adjacent motif (PAM).

While CRISPR has mostly been applied to DNA, recent advances have expanded its range to RNA editing. RNA interference has been a mainstay to knockdown gene expression, but to target particular transcripts for imaging or subcellular localization has required cumbersome design of RNA aptamers or proteins such as Pumilio for each target RNA.

To target RNA in a way that can easily be programmed and scaled up, two independent groups took different routes. Gene Yeo, in collaboration with Jennifer Doudna, expanded on previous findings from the Doudna lab showing that Cas9 can be targeted to RNA if the PAM is provided by a separate DNA oligonucleotide that binds the target RNA. The researchers recently demonstrated that this approach can target specific RNAs; a fusion between a catalytically inactive Cas9 and GFP allowed them to track the subcellular localization of RNA in live mammalian cells (Cell 165, 488–496, 2016). The teams of Feng Zhang and Eugene Koonin took a different approach. Instead of Cas9, they used the C2c2 nuclease, which has an RNase domain but no known DNase domains. They targeted C2c2 by a single 28-nucleotide gRNA and saw single-strand RNA cleavage in the bacterial target transcripts (Science 353, aaf5573, 2016). Simple base substitutions converted C2c2 into a catalytically inactive RNA binding protein that can now be coupled to different effector proteins.

Both approaches open the possibility of influencing the many post-transcriptional processing steps of RNA. C2c2 is a particularly good candidate, since recent work on its catalytic activity showed that it has two independent RNase activities, one to process its own RNA guides and another to cleave the target RNA (Nature 538, 270–273, 2016), which will allow multiplexed applications.

With the appropriate effector fused to the nuclease, one can envision a myriad of uses, such as regulating splicing, directing RNAs to particular subcellular localizations, attaching or removing chemical modifications, and affecting degradation, to name only a few. RNA-targeted CRISPR will give researchers access to a regulatory layer of which we have so far only scratched the surface.