Report from 2nd Programmable Nucleases Course April 2017
Lluis Montoliu
Research Scientist
Departament of Molecular and Cellular Biology, National Centre of Biotechnology (CNB-CSIC)
Dear colleagues and friends,
From the 3rd – 7th April we had the 2nd edition of the Nuclease hands-on Course at the CCP-BIOCEV, in Prague, nicely organized by Radislav Sedlacek, Inken Beck and colleagues. I think it was a great course, where participants (and instructors) could learn a lot and discuss about their own projects directly with people actively developing or using CRISPR tools. We enjoyed the Czech hospitality and the impressive Czech Centre for Phenogenomics, an excellent new and modern venue to host this type of hands-on courses and also where to successfully run mouse functional genomics projects, from generation to phenotyping.
I would like to highlight three aspects of this course, which might be relevant or hopefully be of some interest for some of you.
1.- Francis Stewart delivered a great talk sharing the latest results from his lab in using CRISPR strategies, combined with recombineering (a technique he devised and for which he was awarded the ISTT Prize at the TT2010 meeting in Berlin), for improved knockin projects. The bottom line and his take-home message was to return to use classical targeting constructs (that can be obtained through recombineering techniques) with up to 4 kb (in total, both arms) of homology, within a plasmid, delivered as intact circular DNA (hence closed). Combining the use of these targeting constructs, where recombineering could be most helpful for building them, with a guide RNA opening the locus where we expect the knockin to occur in mouse ES cells resulted in great efficiencies, in any case greater than using oligo-based strategies, apparently simpler and quicker, but not necessarily more efficient, as we all know. Whether the same high knockin efficiencies using plasmid templates carrying targeting constructs can be observed or not in mouse embryos is something currently being investigated. In any case, please keep one eye in current and forthcoming Francis papers. A couple of his recent contributions are:
RecET direct cloning and Redαβ recombineering of biosynthetic gene clusters, large operons or single genes for heterologous expression.
Wang H, Li Z, Jia R, Hou Y, Yin J, Bian X, Li A, Müller R, Stewart AF, Fu J, Zhang Y.
Nat Protoc. 2016 Jul;11(7):1175-90.
https://www.ncbi.nlm.nih.gov/pubmed/27254463
RAC-tagging: Recombineering And Cas9-assisted targeting for protein tagging and conditional analyses.
Baker O, Gupta A, Obst M, Zhang Y, Anastassiadis K, Fu J, Stewart AF.
Sci Rep. 2016 May 24;6:25529.
https://www.ncbi.nlm.nih.gov/pubmed/27216209
2.- Marie Christine Birling and Guillaume Pavlovic, from the Institute Clinique de la Souris (ICS-Phenomin-IGBMC-Univ Strasbourg), directed by Yann Herault, presented their amazing set of results using CRISPR strategies to obtain almost any type of gross chromosomal alterations, including deletions, duplications, inversions, alone or in any combination, both in rats and in mice, in their attempts to reproduce in rodents some of the features associated with Down syndrome. Some very big deletions were achieved in rats (up to 24.4 Mb), setting new world records in this field. Most of these results are included in their last publication:
Efficient and rapid generation of large genomic variants in rats and mice using CRISMERE.
Birling MC, Schaeffer L, André P, Lindner L, Maréchal D, Ayadi A, Sorg T, Pavlovic G, Hérault Y.
Sci Rep. 2017 Mar 7;7:43331. doi: 10.1038/srep43331.
https://www.ncbi.nlm.nih.gov/pubmed/28266534
This study nicely illustrates the power but also the allelic variability of results one can achieve using CRISPR strategies when attempting to trigger deletions, duplications and inversions. Most of the time the planned alleles will be found among the founder mice generated, and transmitted through germline to their progeny, but one has to invest time and carefully design genotyping strategies to adequately interpret the great variety of genome-edited alleles generated. This study also highlights that rats would appear better hosts for these type of challenging CRISPR-mediated genome alterations, and, in general for knockin attempts as well, considering the efficiencies reported, as compared to mice. Guillaume speculated that perhaps the fact that rats are mostly outbred whereas mice used are mostly inbred would play a role regarding what can be easily done or not done with them, at genome editing level. Something interesting to take into account and further investigate.
3.- Bernd Zetsche (BROAD-MIT), from Feng Zhang’s lab, was also participating in this course and delivered a great talk on their newest data, as well as data from other labs, on the Cpf1 nuclease, the novel CRISPR-Clas II nuclease, alternative to Cas9, they discovered and first reported in 2015 (http://www.ncbi.nlm.nih.gov/pubmed/26422227). As you know, the Cpf1 has unique and interesting properties, different from Cas9. Cpf1 uses a PAM located at 5′, it appears to cut leaving protruding ends and only requires one short RNA guide (not two, as Cas9). Besides acting as RNA driven DNA endonuclease has also been found to act as RNAse, being able to process a multiple array of crRNA guides from a single RNA into the respective short RNA guides later used to target DNA sequences at genomic loci. Cpf1 seems to be used a lot in the Plant world. Our colleagues from plants appear to have discovered that Cpf1 is more attractive than Cas9 in producing genome-edited modifications, perhaps due to the additional simplicity of Cpf1 (one RNA, crRNA), as compared to Cas9 (two RNAs, crRNA and tracrRNA). The combined DNA and RNA nuclease activities of Cpf1 were recently reported by Bernd and colleagues in this publication:
Multiplex gene editing by CRISPR-Cpf1 using a single crRNA array. Zetsche B, Heidenreich M, Mohanraju P, Fedorova I, Kneppers J, DeGennaro EM, Winblad N, Choudhury SR, Abudayyeh OO, Gootenberg JS, Wu WY, Scott DA, Severinov K, van der Oost J, Zhang F. Nat Biotechnol. 2017 Jan;35(1):31-34.
https://www.ncbi.nlm.nih.gov/pubmed/27918548
Bernd also shared the latest news concerning the PAM specificity of Cpf1. This nuclease was originally reported to use a PAM at 5′ consisting of a recognition sequence of TTN or TTTN, but this is no longer correct. Updated experiments and results have led to the conclusion that the correct PAM for Cpf1 nucleases at 5′ is “TTTV”, where V can be A, G or C, but not T. The matching sequence (the length of the RNA guide) was originally announced as 23, but the reality is that only the first 18 positions contribute to drive the specificity, the homology, whereas the last four positions, from 20 to 23, can be in fact any nucleotide “NNNN”, therefore these potential off-targets are in reality, and must be considered as, on-targets! Also, Bernd announced a new set of Cpf1 mutants derived in Feng Zhang’s lab, with altered PAM specificities. In particular two of these mutants are the so-called Cpf1 RVR, whose PAM is “TVTV” and the so-called Cpf1 RR, whose PAM is “TYCV”. These two mutants were reported in a manuscript currently under review but already posted at the BioRchiv pre-print server. Bernd announced that these two Cpf1 mutants will be soon available through AddGene. The deposited Cpf1 wild-type plasmid has one NLS but recent data, particularly in cells, suggest that an additional NLS must be added to ensure efficient translocation into the nucleus, something which might be not that relevant when injecting these reagents into mouse embryos, particularly if injected into pronucleus. In this regard, it was noted that IDT is the first company commercializing Cpf1 protein (https://eu.idtdna.com/pages/products/genome-editing/crispr-genome-editing/crispr-cpf1-genome-editing). Disclosure: I have no interests with IDT products
Genome-wide analysis reveals specificities of Cpf1 endonucleases in human cells. Kim D, Kim J, Hur JK, Been KW, Yoon SH, Kim JS. Nat Biotechnol. 2016 Aug;34(8):863-868.
http://www.ncbi.nlm.nih.gov/pubmed/27272384
Genome-wide specificities of CRISPR-Cas Cpf1 nucleases in human cells. Kleinstiver BP, Tsai SQ, Prew MS, Nguyen NT, Welch MM, Lopez JM, McCaw ZR, Aryee MJ, Joung JK. Nat Biotechnol. 2016 Aug;34(8):869-874.
http://www.ncbi.nlm.nih.gov/pubmed/27347757
Engineered Cpf1 Enzymes with Altered PAM Specificities. Linyi Gao, David B.T. Cox, Winston X Yan, John Manteiga, Martin Schneider, Takashi Yamano, Hiroshi Nishimasu, Osamu Nureki, Feng Zhang. BioRxiv, Dec 4, 2016. doi: https://doi.org/10.1101/091611
http://biorxiv.org/content/early/2016/12/04/091611
In vivo high-throughput profiling of CRISPR-Cpf1 activity. Kim HK, Song M, Lee J, Menon AV, Jung S, Kang YM, Choi JW, Woo E, Koh HC, Nam JW, Kim H. Nat Methods. 2017 Feb;14(2):153-159.
https://www.ncbi.nlm.nih.gov/pubmed/27992409
The fact that new PAM sequences are required for devising correct RNA guides to be used in combination with Cpf1 prompted me to contact our bioinformatician at CNB-CSIC, Juan Carlos Oliveros, with whom we generated and launched Breaking-Cas (https://www.ncbi.nlm.nih.gov/pubmed/27166368). He efficiently responded to my request and already implemented the changes and now, you can seek for Cpf1 cutting sites at your favourite genomic sequence using the updated version of Breaking-Cas (click on nuclease presetting and select either wild-type or mutant Cpf1). I believe Breaking-Cas is currently the only RNA guide designing web-based program where Cpf1 guides can be designed with the correct PAM sequences.
http://bioinfogp.cnb.csic.es/tools/breakingcas/
All these CRISPR-related updates and improvements have been annotated in the CRISPR web page we maintain at CNB-CSIC
http://wwwuser.cnb.csic.es/~montoliu/CRISPR/
with best regards
Lluis