Using Cpf1 for CRISPR

What is Cpf1?

Cpf1 is an RNA-guided nuclease, similar to Cas9. It recognizes a T-rich PAM, TTTN, but on the 5' side of the guide. This makes it distinct from Cas9, which uses an NGG PAM on the 3' side. The cut Cpf1 makes is staggered. In AsCpf1 and LbCpf1, it occurs 19 bp after the PAM on the targeted (+) strand and 23 bp on the other strand, as shown here:

20 nt guides can be used for now, and updates will be made here on optimal guide length. Robust guide scores for Cpf1 are still in development, but simple scoring based on the number of off-target sites is available. Cpf1 requires only a crRNA for activity and does not need a tracrRNA to also be present.

Two Cp1-family proteins, AsCpf1 (from Acidaminococcus) and LbCpf1 (from Lachnospiraceae), have been shown to perform efficient genome editing in human cells.

Why use Cpf1 over Cas9?

There's some features of Cpf1 that make it advantageous for certain use cases:

• Editing in AT-rich genomes/regions, or SNP-specific editing
  All characterized genome editing nucleases before Cpf1 required at least one G in their PAMs. Cpf1's PAM is TTTN, so it broadens the range of genome editing experiments. This is particularly useful in AT-rich genomes (e.g. Plasmodium falcipracum) or regions (e.g. scaffold/matrix attachment regions). It's also useful if a particular SNP is present that results in an AT-rich site.
• Applications (e.g. chemically modified guides) where shorter RNA species are useful
  Cpf1 only requires a crRNA and doesn't need tracrRNA - it processes crRNA arrays without needing tracrRNA and Cpf1-crRNA cuts target DNA without needing any other RNA types. Shorter oligos are easier to deliver and cost less to synthesize. This is particularly useful when chemically modifying guides, which was recently shown to enhance efficiency in human primary cells.

There are also a few potential benefits of using Cpf1 that are still being explored:

• Using the staggered cut to perform knock-in without HDR, such as in non-dividing cells
  Cpf1 generates a staggered cut with a 5' overhang, in contrast to the blunt ends generated by Cas9. This allows for editing through non-homologous end joining (NHEJ). Being able to program the exact sequence of a sticky end would allow researchers to design the DNA insert so that it integrates in the proper orientation. This is an alternative to HDR, which can be challenging in applications such as non-dividing cells.
• Increased efficiency of HDR-based knock-ins from repeated cutting after indel formation
  Cpf1 cleaves target DNA far away from the PAM at the end of the protospacer, unlike Cas9. The indels Cpf1 causes will therefore be located far from the target site. Cpf1 can continue cutting at the target site as a result, which may increase the chance that new DNA can be inserted at that site.

Designing gRNAs for Cpf1

Benchling supports finding guides for Cpf1, and is completely free to use. Robust guide scores for Cpf1 are still in development, but simple scoring based on the number of off-target sites is available.

Sign up and watch the video below to learn how to design guides. "TTTN (on the 5' side)" is now an option for the PAM setting.

Cloning into Cpf1 Plasmids

Two Cpf1-family proteins (AsCpf1 and LbCpf1) were shown to cause indels in human cells. pY010 (which expresses humanized AsCpf1) and pY016 (which expresses humanized LbCpf1) are now available on Addgene. Other plasmids from the paper can be found here.

Plasmids for cloning in custom sgRNA for expression are planned for release. Until they are available, U6 PCR products can be used, as described here. You can use this template spreadsheet to design U6 PCR primers for now. (You'll also need this macro.)

FAQ

Coming soon. If you have questions, please ask them on the CRISPR Google Group.