ArticleCpf1 Is a Single RNA-Guided Endonuclease of a Class 2 CRISPR-Cas System Graphical Abstract Authors Bernd Zetsche, Jonathan S. Gootenberg, Omar O. Abudayyeh, ..., Aviv Regev, Eugene V. Koonin, Feng Zhang Correspondence [email protected] In Brief Cpf1 is a RNA-guided DNA nuclease that provides immunity in bacteria and can be adapted for genome editing in mammalian cells. Highlights d CRISPR-Cpf1 is a class 2 CRISPR system d Cpf1 is a CRISPR-associated two-component RNA- programmable DNA nuclease d Targeted DNA is cleaved as a 5-nt staggered cut distal to a 50 T-rich PAM d Two Cpf1 orthologs exhibit robust nuclease activity in human cells Zetsche et al., 2015, Cell 163, 759–771 October 22, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.cell.2015.09.038 their genome editing applications. 2015). but they are all type II and employ homologous RNA-guided effector proteins and CRISPR RNAs (crRNAs) (Makarova et al.. Department of Agrotechnology and Food Sciences. crRNA and tracrRNA.10 Omar O. Abudayyeh. Diversity is observed at the level of demonstrate that Cpf1 mediates robust DNA interfer. 2011. MA 02139.doi.3.2. Sapranauskas includes three stages: (1) adaptation. Wageningen University. 2014.. MA 02139. USA 10Co-first author *Correspondence: [email protected]. RNA) (Horvath and Barrangou. MD 20894. 2008. (2) expression and processing of The microbial adaptive immune system CRISPR the precursor CRISPR (pre-cr) RNA resulting in the formation of mediates defense against foreign genetic elements mature crRNAs. the effector modules differ sub- over. Class 1 effectors utilize multi-protein tein—is guided by a crRNA to recognize and cleave target DNA complexes. In the latest classification.4.4 John van der Oost.4. 2008. Deltcheva et al. 2011.1016/j.2.7 and Feng Zhang1. when the effector mod- through two classes of RNA-guided nuclease ule—either another Cas protein complex or a single large pro- effectors. 6703 HB Wageningen. The defense activity of the CRISPR-Cas systems 2007. and (3) interference. 2012). the diverse CRISPR-Cas systems are divided into two classes ac- aminococcus and Lachnospiraceae. has Cell 163.3. tentatively assigned to type V. we report characterization are shared by all known CRISPR-Cas systems.4. have been identified and functionally charac- INTRODUCTION terized in detail. (or in some cases. Moreover. Sinkunas et al. 2010.4 Kira S. Harvard Medical School. 759–771. Bonn Medical School.09.3. Dreijenplein 10.1. MA 02115. Multiple class 1 CRISPR-Cas systems.1. 2014). 2010. whereas class 2 CRISPR systems this mechanism of interference broadens our under- employ a large single-component Cas protein in conjunction standing of CRISPR-Cas systems and advances with crRNAs to mediate interference (Makarova et al. Slaymaker.org/10. 2015). pu- proteins excises a segment of the target DNA (known as a tative class 2 CRISPR system. We volves additional Cas proteins.4 Sara E... More. Gasiunas et al. MA 02142. Several class 2 CRISPR-Cas spaced short palindromic repeats and CRISPR-associated pro. endonucleases of the Cas9 family as effectors (Barrangou et al.3. Sorek et al.2.2. 2012.1. Identifying form an effector complex.3.1. Institute of Pathology. Boston. each of which consists of a combination of Cas ized.. proceeding ence with features distinct from Cas9. revealing the complex architecture and dynamics of the effector complexes (Brouns et al. A second.9 Eugene V. processing of the pre-crRNA to mature crRNA guides. Netherlands 9Department of Biology.1. et al.. 2015 ª2015 Elsevier Inc. Charpentier et al. 2013.org http://dx. USA 7National Center for Biotechnology Information. ble-stranded break. a putative class 2 CRISPR effector. when a complex of Cas et al.2.3. 2009. Volz. Marraffini and Son- Almost all archaea and many bacteria achieve adaptive immunity theimer.5. 53127 Bonn. USA 8Laboratory of Microbiology. Cpf1 cleaves DNA via a staggered DNA dou. The adaptation stage gle-component effector proteins such as the well- is mediated by the complex of the Cas1 and Cas2 proteins.038 SUMMARY protospacer) and inserts it into the CRISPR array (where this sequence becomes a spacer).3.4 Ian M. USA 2McGovern Institute for Brain Research 3Department of Brain and Cognitive Sciences 4Department of Biological Engineering Massachusetts Institute of Technology. 2015).* 1Broad Institute of MIT and Harvard.7 Patrick Essletzbichler.cell.4 Julia Joung. we identified two candidate enzymes from Acid. 2015.1.6. 759 .. stantially among the CRISPR-Cas systems (Makarova et al.. Hale et al... 2011.. Garneau et al. Massachusetts Institute of Technology. USA 5Department of Developmental Pathology. Jackson through a diverse set of CRISPR-Cas (clustered regularly inter. Gootenberg. Here. and sometimes in- of Cpf1.1. Sigmund Freud Street 25.2. Makarova.2. Germany 6Department of Systems Biology.2. which include the type I and type III systems. and it that specifically cleaves double-stranded RNA hybrids of pre- utilizes a T-rich protospacer-adjacent motif. Jinek et al. Cpf1 is a single via either a Cas6-related ribonuclease or a housekeeping RNaseIII RNA-guided endonuclease lacking tracrRNA. with efficient cording to the configuration of their effector modules: class 1 CRISPR systems utilize several Cas proteins and the crRNA to genome-editing activity in human cells.1. October 22. 2013. whereas class 2 effectors rely on sin. 2011..Article Cpf1 Is a Single RNA-Guided Endonuclease of a Class 2 CRISPR-Cas System Bernd Zetsche. Mulepati et al. 2014). which characterized Cas9.8 Aviv Regev.10 Jonathan S. Out of 16 Cpf1-family proteins. Howard Hughes Medical Institute. Cambridge. Barrangou and Marraffini. Cambridge.3.. Koonin. National Institutes of Health.. National Library of Medicine. Bethesda. Cambridge. systems have also been identified and experimentally character- teins) systems.. Cpf1 introduces a staggered DNA double. The domain architectures of FnCas9 and FnCpf1 are compared. error bars show 95% Bayesian confidence interval. coli harboring pFnCpf1 provides robust interference against plasmids carrying 50 -TTN PAMs (n = 3.. these 2015). (B) Schematic illustrating the plasmid depletion assay for discovering the PAM position and identity. 2011. systems.A Figure 1.. Cpf1 differs from Cas9 in that it lacks a second. such as multiple strains of (tracrRNA) (Deltcheva et al. (C and D) Sequence logo for the FnCpf1 PAM as determined by the plasmid depletion assay. The Francisella novicida U112 Cpf1 CRISPR Locus Provides Immunity against Transformation of Plasmids Con- taining Protospacers Flanked by a 50 -TTN PAM (A) Organization of two CRISPR loci found in Francisella novicida U112 (NC_008601). 2015). Plasmids from surviving colonies were extracted and sequenced to determine depleted PAM B C D sequences. Chylinski et al. coli harboring either the heterologous FnCpf1 locus plasmid (pFnCpf1) or the empty vector control were transformed with a library of plasmids containing the matching pro- tospacer flanked by randomized 50 or 30 PAM se- quences and selected with antibiotic to deplete plasmids carrying successfully targeted PAM.. To explore the suitability of Cpf1 for genome-editing applications. effector of a CRISPR locus that is distinct from the Cas9-contain- ated CRISPR arrays are processed into mature crRNAs without ing type II CRISPR-Cas loci that are also present in the genomes the requirement of an additional trans-activating crRNA of some of the same bacteria. results establish Cpf1 as a class 2 CRISPR-Cas system that 1. However. 2013. which is inserted within the 760 Cell 163. Cpf1-associ. First. Jiang and Marraffini. we show that Cpf1-containing CRISPR-Cas loci of Immune Systems Francisella novicida U112 encode functional defense systems Cpf1 was first annotated as a CRISPR-associated gene in capable of mediating plasmid interference in bacterial cells TIGRFAM (http://www.org/cgi-bin/tigrfams/HmmReportPage. Cpf1-con. RESULTS we sought to explore the function of Cpf1-based putative CRISPR systems. cgi?acc=TIGR04330) and has been hypothesized to be the taining CRISPR systems have three features. 2015) (Figure 1A). however. Sec.cgi?acc=TIGR04330) Lachnospiraceae bacterium ND2006 that are capable of medi- (Schunder et al. Collectively. See also Figure S1. October 22. stranded break with a 4 or 5-nt 50 overhang. et al. Third. 2014. distinct properties that has the potential to substantially advance containing CRISPR loci indeed represent functional CRISPR our ability to manipulate eukaryotic genomes. Cpf1-crRNA complexes efficiently cleave target DNA pro. The putative type V CRISPR-Cas systems contain a large.. Makarova et al.. The Cpf1 protein ceeded by a short T-rich protospacer-adjacent motif (PAM).. whether Cpf1.org/cgi-bin/tigrfams/HmmReportPage. Francisella and Prevotella (Schunder et al. 2014. It remains unknown.jcvi. and we identified been recently identified in several bacterial genomes (http://www. Letter height at each position is measured by information content (C) or frequency (D). ating robust genome editing in human cells. HNH endonuclease domain.. systems. E (E) E. in contains a predicted RuvC-like endonuclease domain that is contrast to the G-rich PAM following the target DNA for Cas9 distantly related to the respective nuclease domain of Cas9. BV3L6 and jcvi. Makarova et al. error bars represent mean ± SEM). Unlike Cas9 systems. Competent E. 2014.. guided by the CRISPR spacers. Vestergaard ond. we charac- terized the RNA-guided DNA-targeting requirements for 16 Cpf1-family proteins from diverse bacteria. 2015 ª2015 Elsevier Inc. 2013).. two Cpf1 enzymes from Acidaminococcus sp. . Cpf1-Containing CRISPR Loci Are Active Bacterial Here. 2013. 759–771. Given the broad applications of Cas9 as a genome- engineering tool (Hsu et al. Vestergaard et al.300 amino acid protein called Cpf1 (CRISPR from Prevotella includes an effective single RNA-guided endonuclease with and Francisella 1). we did not observe any robustly expressed small tran- CRISPR systems. This crRNA from Francisella (Schunder et al. 2013) to ascertain the ac. but also that type V systems are functionally loci. October 22. we determined the PAM The finding that FnCpf1 can mediate DNA interference with sequence and location by identifying nucleotide motifs that are crRNA alone is highly surprising given that Cas9 recognizes preferentially depleted in cells heterologously expressing the crRNA through the duplex structure between crRNA and FnCpf1 locus.. 759–771. We found that all 50 -TTN PAMs were efficiently targeted Apart from these distinctions between the effector proteins. there are two requirements for DNA interference: Francisella cpf1 (FnCpf1) and the CRISPR array (pFnCpf1_min). apart from the targeted DNA interference. Beyond the identification of the of Cpf1 is predicted to adopt a mixed a/b structure and appears PAM. To ensure that crRNA is indeed sufficient location of the PAM is also observed in type I CRISPR systems. 2012. as well as stream of the 50 end of the displaced strand of the protospacer the 30 secondary structure of the tracrRNA (Hsu et al. Each sequences for similarity to the type V spacers that produced mature crRNA begins with 19 nt of the direct repeat followed several significant hits to prophage genes—in particular. 2015 ª2015 Elsevier Inc.. we found that the CRISPR array is pro- systems is further buttressed by the search of microbial genome cessed into short mature crRNAs of 42–44 nt in length. Zhang et al. where Cas9 employs PAM sequences guided DNA cleavage. library was transformed into E.. Given these observa. suggesting that the middle T is more critical that are more closely related to orthologs from types I and III for PAM recognition than the first T and that. tracrRNA (Jinek et al. By sequencing small RNAs extracted from a Francisella unique. By contrast. a plasmid with all of the cas genes plasmid depletion assay (Jiang et al. 2011. In addition.. After showing that cpf1-based CRISPR loci are able to mediate tems independently evolved through the association of different robust DNA interference. removed but retaining native promoters driving the expression tivity of Cpf1 and identify the requirement for a PAM sequence of FnCpf1 and the CRISPR array.. but not 50 -TCA.. Taken together. arrangement contrasts with that of type II CRISPR-Cas systems tions and the prevalence of Cpf1-family proteins in diverse in which the mature crRNA starts with 20–24 nt of spacer bacterial species. PAMs. 2013. E. prompted the classification of Cpf1-encoding CRISPR-Cas loci as the putative type V within class 2 (Makarova et al. To further characterize the PAM requirements. We found that the PAM for FnCpf1 is located up. a-helical recognition lobe of erologously expressed Cpf1 loci are capable of efficient interfer- Cas9 (Figure 1A). sequence followed by 22 nt of direct repeat (Deltcheva et al. especially the domain ar.. and (2) the target sequence robust processing of the CRISPR array into mature crRNA (Fig- complementary to the spacer (hereinafter protospacer) has to ure 2B). the results of the depletion assay clearly indicate that het- to be unrelated to the N-terminal. of Cas9 and Cpf1 are homologous to distinct groups of trans. Chylinski et al. Furthermore. we adapted a previously described well as pFnCpf1_DCas. we cloned the Francisella tracrRNAs... Typically. in agreement with than to those from type II CRISPR systems (Makarova et al. and cas4 genes) with different to determine the exact identity of the crRNA produced by these TnpB genes. the N-terminal portion 2009. It has been shown that the nuclease moieties ence with plasmid DNA. 2015). 2013). we constructed an expression Escherichia coli. Given the completely uncharacterized functionality of processing.RuvC-like domain of Cas9. indicating that FnCpf1 and its CRISPR array are the be flanked by the appropriate protospacer adjacent motif only elements required from the FnCpf1 locus to achieve crRNA (PAM). 50 -CTA.. The 50 Nishimasu et al. plasmid using synthetic promoters to drive the expression of Cas systems. We purified Cell 163. we analyzed poson-encoded TnpB proteins. Unexpectedly. coli expressing this plasmid still showed in the respective CRISPR array.. and has the sequence 50 -TTN (Figures 1C.. demonstrating that FnCpf1 and crRNA are suffi- (Figure 1B). Furthermore. was also efficiently the Cpf1-carrying loci encode Cas1.. novicida U112 Cpf1 (FnCpf1) locus (Figure 1A) into low-copy To confirm that no additional RNAs are required for crRNA plasmids (pFnCpf1) to allow heterologous reconstitution in maturation and DNA interference. we sought to test the hypothesis that Cpf1-en. coding CRISPR-Cas loci are biologically active and can mediate 2011. these differences from type II have ure S1D). those by 23–25 nt of the spacer sequence (Figure 2A). we performed small RNA sequencing adaptation modules (cas1. (1) the target sequence has to match one of the spacers present Small RNaseq of E. a protospacer matching the first spacer in the FnCpf1 CRISPR Cas9 requires both crRNA and tracrRNA to mediate targeted array with the 50 or 30 7 bp sequences randomized. coli strain carrying Cpf1 Is a Single crRNA-Guided Endonuclease the empty vector. ing cells with plasmids carrying protospacer 1 flanked by 50 -TTN ing the RuvC-like domain only (Makarova and Koonin. we investigated whether FnCpf1 supplied that are located on the 30 end of the protospacer (Mojica et al. for forming an active complex with FnCpf1 and mediating RNA- but not in type II systems. (Figure 1E).. 2015). The Cpf1-Associated CRISPR Array Is Processed The features of the putative type V loci. The notion that Cpf1-carrying loci are bona fide CRISPR novicida U112 culture. which are associated with Cas9-based systems. scripts near the Francisella cpf1 locus that might correspond to To simplify experimentation. coli that heterologously ex- pressed the FnCpf1 locus or into a control E. 2014). the first one containing both plasmid interference activity by transforming cpf1-locus-express- RuvC and HNH nuclease domains and the second one contain. 2013). Cas2. Using this assay. only with crRNA can cleave target DNA in vitro. 2010). suggest not only that type II and type V sys. Independent of TracrRNA chitecture of Cpf1. also exhibited robust DNA and its respective location relative to the protospacer (50 or 30 ) interference. and Cas4 proteins targeted (Figure 1E). one of the primary functions of crRNAs. 1D and S1). 2014). Nishimasu et al. We constructed two libraries of plasmids carrying cient for mediating DNA targeting (Figure 2C). the sequence motifs depleted in the PAM discovery assay (Fig- 2015). 2013). the PAM might be more relaxed than 50 -TTN. in currently characterized CRISPR. coli expressing pFnCpf1_min as the FnCpf1 CRISPR locus. cas2. Garneau et al. 761 . Each plasmid DNA interference (Deltcheva et al. and FnCpf1(D1225A) (Figure 4A)—to test whether the conserved catalytic residues are essential for the nuclease activity of FnCpf1. Heterologous Expression of FnCpf1 and CRISPR Array in E. coli transformed with a plasmid-carrying synthetic promoter-driven FnCpf1 and CRISPR array shows crRNA pro- cessing independent of Cas genes and other sequence elements in the FnCpf1 locus. and S3A–S3D). Gasiunas et al. coli Is Suffi- cient to Mediate Plasmid DNA Interference and crRNA Maturation (A) Small RNA-seq of Francisella novicida U112 reveals transcription and processing of the FnCpf1 CRISPR array. mains converts SpCas9 into a DNA nickase (i. (Figure S2B).. Therefore.. 759–771.. and only consists of a single stem loop in the direct repeat 762 Cell 163. 3D.and crRNA-dependent manner (Figure 3B). Moreover. These results are in contrast to the mutagen- protospacer-1-containing plasmid used in the bacterial DNA esis results for Streptococcus pyogenes Cas9 (SpCas9). perhaps in a dimeric configuration. (C) E. which is different from the blunt cleavage product generated by Cas9 (Garneau et al. and D1255A significantly reduced nucleolytic FnCpf1 (Figure S2) and assayed its ability to cleave the same activity (Figure 4B). 2014). the guide RNA for FnCpf1 is notably simpler double-stranded oligo substrates with different PAM sequences.. These findings suggest that the RuvC-like DNA (Figure 3C). coli harboring different truncations of the FnCpf1 CRISPR locus shows that only FnCpf1 and the CRISPR array are required for plasmid DNA interference (n = 3. Sequence and Structural Requirements for the 2012. of 300 kD. (B) Small RNA-seq of E. The RuvC-like Domain of Cpf1 Mediates RNA-Guided DNA C Cleavage The RuvC-like domain of Cpf1 retains all of the catalytic residues of this family of endonucleases (Figures 4A and S4) and is thus predicted to be an active nuclease. where interference experiments (Figure 3A). ciency of FnCpf1 and crRNA for RNA-guided DNA cleavage. twice the molecular weight of a FnCpf1 monomer mediated cleavage results in a 5-nt 50 overhang (Figures 3A. Gasiunas et al. FnCpf1 was able to cleave both supercoiled and linear target 2012) (Figure 4B). we generated three mutants— FnCpf1(D917A). 2015 ª2015 Elsevier Inc. domain of FnCpf1 cleaves both strands of the target DNA.. 2012. one of the DNA strands) (Jinek et al. . We found that FnCpf1.. Jinek et al. The mature crRNA begins with a 19-nt partial direct repeat followed by 23–25 nt of spacer sequence. and S3A–S3D). B we also found that FnCpf1 requires the 50 - TTN PAM to be in a duplex form in order to cleave the target DNA (Figure 3E).. Interestingly. October 22. The staggered cleavage site of Cpf1 crRNA FnCpf1 is distant from the PAM: cleavage occurs after the 18th Compared with the guide RNA for Cas9. 2010. Using imasu et al. We found that FnCpf1 mutation of the RuvC (D10A) and HNH (N863A) nuclease do- along with an in-vitro-transcribed mature crRNA-targeting proto. 3D. These results clearly demonstrate the suffi. inactivation of spacer 1 was able to efficiently cleave the target plasmid in a each of the two nuclease domains abolished the cleavage of Mg2+. We found that the D917A and E1006A mutations completely inactivated the DNA cleavage activity of FnCpf1. which has elaborate base on the non-targeted (+) strand and after the 23rd base on RNA secondary structure features that interact with Cas9 (Nish- the targeted (–) strand (Figures 3A.A Figure 2.e. error bars show mean ± SEM). 2012). size-exclusion We also mapped the cleavage site of FnCpf1 using Sanger gel filtration of FnCpf1 shows that the protein is eluted at a size sequencing of the cleaved DNA ends. FnCpf1(E1006A). Cleavage sites are indicated by red arrows. We also found that the seed region of direct repeat. SsCpf1) were associated with direct repeat revealed that at least 16. These Cpf1-family of 18 nt of guide sequence to achieve efficient DNA cleavage proteins span a range of lengths between 1. However. only a small fraction of bac. Butyrivibrio proteoclasticus. proteins show strong conservation in the 19 nt at the 30 of the 2014. (D) Sanger-sequencing traces from FnCpf1- digested target show staggered overhangs. 2015 ª2015 Elsevier Inc. represented sequence to achieve detectable DNA cleavage and a minimum the entire Cpf1 diversity (Figures 6A and S5). Finally. Fu et al. but optimally more than 17 nt. denoted as N. Therefore. these results suggest that FnCpf1 recognizes the crRNA pected. (E) Dependency of cleavage on base-pairing at the 50 PAM.A Figure 3. See also Figures S2 and S3. the direct repeats from candidates 2 (Lb3Cpf1) and 6 (SsCpf1) were unable to support FnCpf1 cleavage activity Cpf1-Family Proteins from Diverse Bacteria Share (Figure 6D). Based on our previous experience in harnessing Cas9 for possibly due to the conservation of the 30 -most U.. Next. 763 . Of the 16 Cpf1-family proteins cho- Next. from which we chose 16 candidates that. Mutations in the stem direct repeat (Figure 6B). Truncation of the direct Smithella sp.. we applied the in vitro PAM identification assay (Fig- terial nucleases can function efficiently when heterologously ex. and 6. (B) FnCpf1 and crRNA alone mediated RNA- guided cleavage of target DNA in a crRNA. in order to assess the feasi- bility of harnessing Cpf1 as a genome- editing tool. Lachnospiraceae bacterium MC2017. structure completely abolished cleavage (Figure 5D). tures of the stem loop.. Collec. We were able to identify the PAM sequence for seven Cell 163. 1988). A BLAST search of the WGS database requirements of crRNA for mediating DNA cleavage with FnCpf1. The non-templated addition of an additional adenine. BpCpf1. even these direct repeat loop that preserved the RNA duplex did not affect the cleavage sequences preserved stem-loop structures that were identical activity. FnCpf1 can only recognize the PAM in E correctly Watson-Crick-paired DNA. 759–771. three (2.and Mg2+-dependent manner. we exploited the diversity of Cpf1-family proteins available in the pub- sequence (Figure 3A). RNA-guided DNA cleavage activity. we studied the effect of direct repeat mutations on the sen for analysis. repeat is much more diverse. at the NCBI revealed 46 non-redundant Cpf1-family proteins We first examined the length requirement for the guide (Figure S5A). 2013. protein. FnCpf1 Is Guided by crRNA to Cleave DNA In Vitro (A) Schematic of the FnCpf1 crRNA-DNA-targeting complex. The direct repeat from candidate 3 (BpCpf1) sup- Common crRNA Structures and PAMs ported only a low level of FnCpf1 nuclease activity (Figure 6D). the portion of the repeat that is included in the the FnCpf1 guide RNA is approximately within the first 5 nt on processed crRNA (Figure 6B). is an artifact of the polymerase used in sequencing (Clark. SC_K08D17. October 22. we activity.. The 50 sequence of the direct the 50 end of the spacer sequence (Figures 5B and S3E). B C D (C) FnCpf1 cleaves both linear and supercoiled DNA.200 and 1. 3. demonstrated for SpCas9. Reverse primer read represented as reverse complement to aid visual- ization. The direct repeat portion Lb3Cpf1. See also Figure S3. Given the strong structural conservation of the direct repeats base substitutions in the loop region did not affect nuclease that are associated with many of the Cpf1-family proteins. As ex- tively. Ran et al. genome editing in mammalian cells. whereas the uracil base immediately proceeding the first tested whether the orthologous direct repeat sequences spacer sequence could not be substituted (Figure 5E).500 in vitro (Figure 5A). the direct repeats that contained conserved stem through a combination of sequence-specific and structural fea. in which a minimum of 16–17 nt of The direct repeat sequences for each of these Cpf1-family spacer sequence is required for DNA cleavage (Cencic et al. 2015). based sequence and found that FnCpf1 requires at least 16 nt of guide on our phylogenetic reconstruction (Figure S5A). of mature crRNA is 19 nt long (Figure 2A). By contrast. We explored the sequence and structural lic sequences databases. ure S6A) to determine the PAM sequence for each Cpf1-family pressed in mammalian cells (Cong et al. 2014). These requirements are similar to those amino acids. are able to support FnCpf1 nuclease activity in vitro. of repeat sequences that are notably divergent from the FnCpf1 the direct repeat is required for cleavage. whereas mutations that disrupted the stem loop duplex or nearly identical to the FnCpf1 direct repeat (Figure 6C). sequences were able to function interchangeably with FnCpf1. the only other experimentally charac- terized class 2 effector—in terms of structure and function and Figure 4. only two out of the eight Cpf1-family proteins (7. 606 nt denaturing TBE-Urea 424 nt DISCUSSION In this work. AsCpf1 and LbCpf1 mediated comparable levels of indel forma- tion (Figure 7E). which requires screen confirmed the PAM for FnCpf1 as 50 -TTN. processing of the pre-crRNA is catalyzed each Cpf1-family protein. is a single RNA-guided endonuclease. Cpf1 substantially differs from Cas9—to date. we used in vitro cleavage followed by Sanger sequencing of the cleaved DNA ends and found that 182 bp 7. Unlike Cas9.. with the RuvC-like domains of each of the two subunits cleaving Urea PAGE gel showing that mutation of the RuvC catalytic residues of FnCpf1 one DNA strand. and S6C). in FnCpf1. Structural RuvC (D10A) catalytic residue of SpCas9 results in nicking of the target site. we selected a guide RNA target site by the bacterial RNase III. We first found that each of formed by the tracrRNA and the complementary portion of the the Cpf1-family proteins along with its respective crRNA de. we cannot rule out that FnCpf1 con- (D917A and E1006A) prevents DNA-nicking activity. in contrast to the two nuclease domains present in catalytic residues were identified based on sequence homology to Thermus thermophilus RuvC (PDB: 4EP5). without of Ts constituting each PAM (Figures 6E. The PAM sequences for the Cpf1-family processes crRNA arrays independent of tracrRNA. 2015). which recognizes the long duplex within the DNMT1 gene (Figure 7B). When compared to Cas9. LbCpf1 also generated staggered cleavage sites (Figures S6E and S6F. inactiva- (B) Native TBE PAGE gel showing that mutation of the RuvC catalytic residues tion of RuvC-like domain abolishes cleavage of both DNA of FnCpf1 (D917A and E1006A) and mutation of the RuvC (D10A) catalytic strands. The domain. RuvC I RuvC II III when tested in human embryonic kidney 293FT (HEK293FT) D917 E1006 D1255 cells. Additionally. whereas the 606 bp remaining Cpf1 proteins showed either no detectable activity or native TBE PAGE only sporadic activity (Figures 7E and S7) despite robust expres- 424 bp sion (Figure S6D). S6B. Specifically. Conceivably. Catalytic Residues in the C-Terminal RuvC Domain of might provide important advantages for genome-editing appli- FnCpf1 Are Required for DNA Cleavage cations. In clear targeting in human cells (Figure 7A). Cpf1 in vitro reconstitution. See also Figure S4. processing mechanism of the type V CRISPR-Cas systems. The results presented here show that. 759–771. The remaining tracrRNA to process crRNA arrays and both crRNA and eight tested Cpf1 proteins did not show efficient cleavage during tracrRNA to mediate interference (Deltcheva et al. S6B. However. The only Cpf1 candidate that expressed poorly was PdCpf1 (Figure S6D). AsCpf1 and 13. However. signed to target DNMT1 was able to cleave a PCR amplicon region like domain of the DNMT1 genomic region in vitro (Figure 7C). FnCpf1 forms a homodimer (Figure S2B). In addition. n We further tested each Cpf1-family protein with additional ei A A 7A ot 55 06 A pr 10 91 12 T T 0 genomic targets and found that AsCpf1 and LbCpf1 consistently E1 W W no D D D mediated robust genome editing in HEK293FT cells. AsCpf1 and 13. Denaturing TBE. 182 nt Cpf1. 2011). and show that its effector protein. This feature could simplify the design and delivery of genome-editing tools. October 22. characterization of Cpf1-RNA-DNA complexes will allow testing of these hypotheses and elucidation of the cleavage mechanism. and Cpf1- proteins were predominantly T rich.. only varying in the number crRNA complexes alone cleave target DNA molecules. We cantly easier and cheaper to synthesize. Perhaps the most notable feature of Cpf1 is that it is a single new Cpf1-family proteins (Figures 6E. A FnCpf1 helical zinc finger. whereas mutation of the tains a second yet-to-be-identified nuclease domain. the shorter (42 nt) crRNA employed by Cpf1 Human Cells has practical advantages over the long (100 nt) guide RNA in We tested each Cpf1-family protein for which we were able to Cas9-based systems because shorter RNA oligos are signifi- identify a PAM for nuclease activity in mammalian cells. 2015 ª2015 Elsevier Inc. the requirement for any additional RNA species. To test the activity of the case of type II. classified as type V. Cpf1 Can Be Harnessed to Facilitate Genome Editing in For example. Such long duplexes 764 Cell 163. we characterize Cpf1-containing class 2 CRISPR systems. 2011). residue of SpCas9 prevents double-stranded DNA cleavage. these find- codon optimized each of these genes and attached a C-terminal ings raise more fundamental questions regarding the guide nuclear localization signal (NLS) for optimal expression and nu.. direct repeat (Deltcheva et al. Cas9. and the crRNA-guided endonuclease. Cpf1 contains a single identified nuclease (A) Domain structure of FnCpf1 with RuvC catalytic residues highlighted. LbCpf1) exhibited detectable levels of nuclease-induced catalytic residues indels (Figures 7C and 7D). and S6C). This result is consistent with previous B experiments with Cas9 in which only a small number of Cas9 FnCpf1 SpCas9 orthologs were successfully harnessed for genome editing in mammalian cells (Ran et al. . respectively). in which genome editing via homology-directed repair to the blunt ends generated by Cas9 (Garneau et al. (B) Effect of crRNA-target DNA mismatch on FnCpf1 cleavage activity. 759–771. are not present in the pre-crRNA of type V systems.. (D) FnCpf1 cleavage activity depends on secondary structure in the stem of the direct repeat RNA structure. 2015 ª2015 Elsevier Inc. crRNA Requirements for FnCpf1 Nuclease Activity In Vitro (A) Effect of spacer length on FnCpf1 cleavage activity. See also Figure S4. 2012). Specifically. 2013). Cpf1 could provide an effective way to precisely introduce product could be particularly advantageous for facilitating non. 2012. mammalian genome (Maresca et al. A B C D E Figure 5. Cell 163. et al. (E) FnCpf1 cleavage activity is unaffected by loop mutations but is sensitive to mutation in the 30 -most base of the direct repeat... See also Figure S3E. Further ex. (C) Effect of direct repeat length on FnCpf1 cleavage activity. genome in the proper orientation. October 22. in contrast cells. Being able to pro- periments aimed at elucidating the processing mechanism of gram the exact sequence of a sticky end would allow re- type V systems will shed light on the functional diversity of searchers to design the DNA insert so that it integrates into the different CRISPR-Cas systems. Jinek (HDR) mechanisms is especially challenging (Chan et al. Gasiunas et al. DNA into the genome via non-HDR mechanisms. 2010. This structure of the cleavage 2011). 765 . in non-dividing Cpf1 generates a staggered cut with a 50 overhang.. making it homologous end joining (NHEJ)-based gene insertion into the unlikely that RNase III is responsible for processing.. October 22. 2015 ª2015 Elsevier Inc. A B C D E (legend on next page) 766 Cell 163. . 759–771. Analysis of Cpf1-Family Protein Diversity and Function (A) Phylogenetic tree of 16 Cpf1 orthologs selected for functional analysis.. In the case of Cpf1. it appears England Biolabs). coli cells carrying genome editing and other areas of biotechnology as well as shed the FnCpf1 locus or pACYC184 control. Plasmid DNA was harvested using a Maxi-prep kit characteristics that await exploration and could further enhance (QIAGEN). The random- opportunities for understanding the origin and evolution of pro. Cpf1-induced indels will be located far Prep Set for Illumina (New England Biolabs) and size selected using the Pippin from the target site.01 psuedocount adjustment. 767 . The target PAM region EXPERIMENTAL PROCEDURES was amplified and sequenced using a MiSeq (Illumina) with a single-end 150 cycle kit. Competent Stbl3 E. so the T. such as scaffold/matrix Durbin. any indel resulting from For heterologous E. There is little doubt that. such as Plasmodium falciparum (Gardner et al. Cells harboring to pACYC184 control. enrichment was measured as the log ratio compared pACYC-184 using Gibson cloning (New England Biolabs). For a given PAM. libraries were prepared from rRNA-depleted RNA using a derivative of the pre- effectively eliminating the possibility of inserting new DNA at viously described CRISPR RNA-sequencing method (Heidrich et al. a Ligase) High Concentration (New England Biolabs). (E) PAM sequences for eight Cpf1-family proteins identified using in vitro cleavage of a plasmid library containing randomized PAMs flanking the protospacer.5 (Biomatters). novicida (generous gift from David Weiss) or E.. 2015 ª2015 Elsevier Inc. nucleotides (IDT) consisting of eight or seven randomized nucleotides either The natural diversity of CRISPR systems provides a wealth of upstream or downstream. Non-conserved bases are colored red.io/picard). 2002) and aligned to the appropriate RefSeq reference genome using BWA (Li and or areas of interest with AT enrichment. Predictions for FnCpf1 along with three diverged type V loci are shown. Bacterial RNA Sequencing PAM Validation RNA was isolated from stationary-phase bacteria by first resuspending Sequences corresponding to both PAMs and non-PAMs were cloned into di- F. PAMs above a 3. (Hsu et al. ligated with 50 RNA adapters using T4 RNA Ligase 1 (ssRNA possible that. 2009). 2011) prediction of the direct repeat sequence in the mature crRNA. Figure 6. Therefore. (C) RNAfold (Lorenz et al. The stem duplex is highlighted in gray. After 16 hr of growth. Colonies were counted after 18 hr. (B) Alignment of direct repeats from the 16 Cpf1-family proteins. coli in TRIzol and then ho.5 plasmids were made competent using the Z-competent kit (Zymo). Cell 163.. and normalized to total reads for each using Herculase II polymerase (Agilent Technologies) and cloned into sample. The dsDNA product was assembled into a linearized beyond the already classified and characterized diversity of the pUC19 using Gibson cloning (New England Biolabs). See also Figures S5 and S6. 2014. with a 0. cDNA was PCR amplified with barcoded primers using Herculase II and other class 2 effectors is expected to bring solutions for polymerase (Agilent Technologies).github. respectively. rounds of Cpf1 cleavage. Paired-end alignments were used to extract entire transcript se- attachment regions. RNA libraries DNA at the distal end of the protospacer. RNA-sequencing the dominant NHEJ repair pathway will disrupt the target site.and T/C-dependent PAMs of Cpf1-family proteins expand the targeting range of In Vivo FnCpf1 PAM Screen Randomized PAM plasmid libraries were constructed using synthesized oligo- RNA-guided genome editing nucleases. cells were plated further light on the evolution of these defense systems. In brief. Sequences that are removed post crRNA maturation are colored gray. coli Poly(A) Polymerase (New that site in that particular cell. Another potentially useful feature of Cpf1 that might aid the DNase treated with TURBO DNase (Life Technologies). 2015). 2004). counted.. gested pUC19 and ligated with T4 ligase (Enzymatics). far away from the were prepared from rRNA-depleted RNA using NEBNext Small RNA Library seed region. with AffinityScript Multiple Temperature Reverse Transcriptase (Agilent Tech- Future exploration of these and other strategies using Cpf1 nologies). transcripts were poly-A tailed with E. on ampicillin. ized ssDNA oligos (Table S1) were made double stranded by annealing to a karyotic adaptive immunity. With Cas9. if the first round of targeting results in an indel. and 30 dephosphory- introduction of new DNA sequences is that Cpf1 cleaves target lated with T4 Polynucleotide Kinase (New England Biolabs). We transformed 30 ng of the pooled library into E. there are additional systems with distinctive collected and pooled. Generation of Heterologous Plasmids To generate the FnCpf1 locus for heterologous expression. genomic DNA from Computational PAM Discovery Pipeline Francisella novicida (generous gift from Wayne Conlan) was PCR amplified PAM regions were extracted. as well as for harnessing potentially short primer and using the large Klenow fragment (New England Biolabs) for transformative biotechnological tools. coli expression of the FnCpf1 locus. coli (Invitrogen) were transformed with the cloned products. Total RNA was purified formed with 20 ng of PAM plasmid and plated on LB agar plates supplemented from homogenized samples with the Direct-Zol RNA miniprep protocol (Zymo). Sequences enrichment threshold were collected and used to generate sequence logos of all bacterial expression plasmids can be found in Table S1. (Crooks et al. Conserved sequences are shown in dark gray. (D) Type V crRNAs from different bacteria with similar direct repeat sequences are able to function with FnCpf1 to mediate target DNA cleavage. all characterized mammalian quences using Picard tools (http://broadinstitute. and zinc finger are highlighted. and reverse transcribed subsequent round of targeting could yet be repaired via HDR. with ampicillin and chloramphenicol. After transformation.1. of the FnCpf1 spacer 1. Reads in genome editing in organisms with particularly AT-rich ge- from each sample were identified on the basis of their associated barcode nomes. The RuvC domain. 2015). 759–771. and these genome-editing proteins require the presence of at least one G sequences were analyzed using Geneious 8. Competent E.. The T-rich PAMs of the Cpf1-family also allow for applications RNA-Sequencing Analysis The prepared cDNA libraries were sequenced on a MiSeq (Illumina). To date. second-strand synthesis. Jiang et al. >4E6 cells were harvested and plasmid DNA was extracted using a Maxi-prep kit (QIAGEN).. October 22. which is thus preserved for subsequent Prep (Sage Science). coli with mogenizing the bacteria with zirconia/silica beads (BioSpec Products) in a either the FnCpf1 locus plasmid or pACYC184 control plasmid were trans- BeadBeater (BioSpec Products) for three 1-min cycles. and >107 cells were CRISPR-Cas types. helical region. rRNA was removed with the bacterial Ribo-Zero rRNA removal kit (Illumina). some of the biggest challenges facing genome editing. Fractions containing protein of the expected size were pooled. and the sample was dialyzed overnight into TEV buffer 768 Cell 163. The culture was induced for 14–18 hr before harvesting cells and freezing were synthesized from IDT and annealed to a short T7 priming sequence. The corresponding crRNA was expressed using a PCR fragment containing a U6 promoter fused to the crRNA sequence. Cpf1 Mediates Robust Genome Editing in Human Cell Lines (A) Eight Cpf1-family proteins were individually expressed in HEK293FT cells using CMV-driven expression vectors. 20 mM imidazole) supplemented with protease inhibitors (Roche cOmplete. then the temperature All crRNAs and sgRNAs used in biochemical reactions were synthesized using was decreased to 21 C. (D) Cpf1 and SpCas9 target sequences in the human DNMT1 and EMX1 loci. (C) Comparison of in vitro and in vivo cleavage activity. See also Figure S7. fuged at 10. with 10 ml overnight culture Rosetta (DE3) pLyseS (EMD Millipore) cells con.22 micron filters (Millipore. Target sites correspond to sequences shown in Figure 7D. 2015 ª2015 Elsevier Inc. A B C D E Figure 7. Once homog- Purification of Cpf1 Protein enized.2 OD600. (E) Comparison of Cpf1 and SpCas9 genome-editing efficiency. transcription was performed for 4 hr. The DNMT1 target region was PCR amplified. AsCpf1 and 13. Two through 0. Sequencing reads show representative indels. and then eluted with a gradient of imidazole. and the genomic fragment was used to test Cpf1-mediated cleavage. TEV protease taining the Cpf1 expression construct. 5 mM MgCl2. but only candidates 7. All eight Cpf1-family proteins showed DNA cleavage in vitro (top). Growth was continued until OD600 reached 0. Lb3Cpf1 facilitated robust indel formation in human cells. ssDNA oligos (Table S2) a final concentration of 500 mM IPTG was added to induce MBP-Cpf1 expres- corresponding to the reverse complement of the target RNA sequence sion. a pET based vector generously provided by Doug Daniels). Transfected cells were analyzed using either Surveyor nuclease assay or targeted deep sequencing. .000 3 g for 1 hr to clear the lysate. EDTA-free) and lysozyme (Sigma). Synthesis of crRNAs and sgRNAs grown at 37 C until the cell density reached 0. October 22. 2M NaCl.6 when the HiScribe T7 High Yield RNA Synthesis Kit (NEB). (B) Schematic showing the sequence of DNMT1-targeting crRNA 3. Growth media plus inoculant was (Sigma) was added. The lysate was filtered TEV-Cpf1. 7]. and then RNA was purified using the Cell paste was resuspended in 200 ml of Lysis Buffer (50 mM HEPES [pH MEGAclear Transcription Clean-Up Kit (Ambion). washed. Stericup) and applied to a nickel col- liters of Terrific Broth growth media with 100 mg/ml ampicillin were inoculated umn (HisTrap FF. cells were lysed by sonication (Branson Sonifier 450) and then centri- FnCpf1 protein was cloned into a bacterial expression vector (6-His-MBP. 759–771. 5 ml). T7 at 80 C until purification. Membranes were incubated 2015). BLOT-QuickBlocker Reagent (Millipore) for 1 hr.000 ng of Cpf1 expression plasmids were TBE polyacrylamide gels (Life Technologies). assisted with RNA sequencing and analysis.3 Read Mapper...S. K.Z. After re-annealing.S. 769 . 95 C to 85 C ramping at 2 C/s. A.A. AUTHOR CONTRIBUTIONS SURVEYOR Nuclease Assay for Genome Modification F.. E. SYBR Gold DNA stain (Life Technologies) for 10 min and imaged with a Gel ofectamine 2000 reagent (Life Technologies). and F..E. Sequencing results were entered into the PAM discovery 2015) as the query.09.Z. 5mM MgCl2. Indel percentage was determined by the formula. Supplemental Information includes seven figures and three tables and can be luminescent Substrate (Life Technology) and imaged using the ChemiDoc found with this article online at http://dx.G. B. J. and F.O. O.R. Proteins were visualized with West Pico Chemi..Z.d. 759–771. 50 mM HEPES [pH 7]. Supernatant was frozen for subsequent use in in vitro cleavage assays. and products were purified using QiaQuick Spin Column (QIAGEN) trated to 200 ml and either used directly for biochemical assays or frozen at following the manufacturer’s protocol. B. 2010). 2.v.. and the sample was concen. The genomic-region-flanking DNMT1 targets were Cleavage in vitro was performed either with purified protein (25 nM) or amplified using a two-round PCR region to add Illumina P5 adapters as well mammalian lysate with protein at 37 C in cleavage buffer (NEBuffer 3....G.01 and low-complexity filtering and E-gels (Life Technologies).M. were pooled and quantified by Qubit 2. J. cleaned-up reactions were run on TBE 6% polyacrylamide or TBE-Urea 6% polyacrylamide gels (Life Technologies). 48 hr later. Results of all searches were combined (Table S3). Multiple sequence alignments were constructed using Cells were lysed in 13RIPA buffer (Cell Signaling Technology) supplemented MUSCLE (Edgar. lated from HEK293 cells.(500 mM NaCl. J.O. and J. Bands corresponding to un-cleaved target were gel composition-based statistics turned off. and mids and 100 ng of the U6::crRNA expression cassettes were transfected F. F.1% Triton X-100.V. 2015 ª2015 Elsevier Inc.V.0. 1997) was used to identify Cpf1 homologs In Vitro Cpf1-Family Protein PAM Screen in the NCBI NR database using several known Cpf1 sequences as queries with In vitro cleavage reactions with Cpf1-family proteins were run on 2% agarose the Cpf1 with the E-value cut-off of 0. Computational Analysis of Cpf1 loci PSI-BLAST program (Altschul et al. I. After dialysis. Fractions from gel filtration were genomic region flanking the CRISPR target site for each gene was PCR ampli- analyzed by SDS-PAGE.S. 300 cycle kit (Illumina). Samples cloned into pUC19 or PCR amplicons of gene regions from genomic DNA iso.S. a re-annealing process to enable heteroduplex formation: 95 C for 10 min. cells were harvested Doc gel imaging system (Bio-rad). or HRP-conjugated GAPDH [Cell Signaling Technology 14C10]) for 1 hr in TBS-T with 1% BLOT-QuickBlocker.. J. CRISPR repeats were identified using PILER-CR (Edgar. 2004) with manual correction based on pairwise alignments with protease inhibitor cocktail (Roche). 1X cOmplete Protease Inhibitor Cocktail Tablets [Roche]). Quantification was based on relative band by washing once with DPBS (Life Technologies) and scraping in lysis buffer intensities. 2006) to identify remote sequence similarity using a subset of representative Cpf1 Western Blot Analysis sequences queries..2015.. Deep Sequencing to Characterize Cpf1 Indel Patterns in 293FT Cells HEK293FT cells were transfected and harvested as described for assessing In Vitro Cleavage Assay activity of Cpf1 cleavage. 85 C to 25 C at 0. and 25 C hold Generation of Cpf1 Protein Lysate for 1 min.. Phylogenetic analysis was run on BOLT 4%–12% Bis-Tris gradient gels (Invitrogen) and transferred performed using the FastTree program with the WAG evolutionary model to PVDF membranes (Millipore). Non-specific antigen binding was blocked and the discrete gamma model with 20 rate categories (Price et al. and J.0 Fluorometer (Life Technologies).S. Gels were stained with transfected into 6-well plates of HEK293FT cells at 90% confluency using Lip. 200–500 ng total of the purified PCR 80 C for storage..A.Z. Target DNA involved either protospacers umn (QIAGEN) as per the manufacturer’s recommended protocol. performed the computational sequence analysis. into 24-well plates of HEK293FT cells at 75%–90% confluency using Lipofect.A.01 and low-complexity filtering turned off was used to gion was amplified and sequenced using a MiSeq (Illumina) with a single-end search the NCBI WGS database using the Cpf1 profile (Makarova et al. The TBLASTN program with the extracted using QIAquick Gel Extraction Kit (QIAGEN).M. 150 mM NaCl and 0.cell. The cleavage reaction used 500 ng of synthesized crRNA or were run on 2% E-gel (Invitrogen) and gel extracted using QiaQuick Spin Col- sgRNA and 200 ng of target DNA. 100 3 (1 (20 mM HEPES [pH 7. The pipeline.O. Membranes were washed for three 10 min washes and anti-HA-tag membranes were further incubated with SUPPLEMENTAL INFORMATION anti-rabbit antibody (Cell Signaling Technology 7074) for 1 hr followed by six 10 min washes in TBS-T.05% Tween-20) with 5% tein secondary structure was predicted using Jpred 4 (Drozdetskiy et al..Z.1016/j. fractions containing Cpf1 were pooled and concen.5].Z. B. S. DNA extraction.0) to calculate the approximate size matics) and ultrapure water to a final volume of 10 ml and were subjected to of FnCpf1.Z.S. 5 mM MgCl. Cells were incubated at 37 C for 72 hr post-transfection before genomic TEV cleavage was confirmed by SDS-PAGE. 150 cycle kit.doi.038. per- sequence were generated using Herculase II (Agilent Technologies) and formed the experiments and analyzed the data. centrifuged. PCR products DTT) for 20 min. 2007) and with primary antibodies (anti-HA-tag [Cell Signaling Technology C29F4] CRISPRfinder (Grissa et al. Pro- with TBS-T (50 mM Tris. conceived this study. wrote the manuscript. B. HHpred program was used with default parameters (Söding et al. rification.. where a is the integrated intensity of the undigested 0. of the Geneious 6.J...K. Equal volumes of cell lysate were obtained using PSI-BLAST and HHpred programs. 400 ng of Cpf1 expression plas.V.K.. Genomic DNA was extracted using the QuickExtract DNA trated to 500 ml prior to loading on a gel filtration column (HiLoad 16/600 Extraction Solution (Epicenter) following the manufacturer’s protocol. O. 2 mM DTT). which was read amine 2000 (Life Technologies). Cell 163. MP Imaging System (Bio-Rad) and processed with ImageLab software (Bio-Rad). fied. 2007). and approved by all authors.S. 1 mM DTT. PCR product.. HEPES (pH 7..Z. P. O.G.G. PCR amplicons comprised of a U6 promoter driving expression of the crRNA designed the experiments.M. conducted FnCpf1 pu- appropriate U6 reverse primers (Table S2). October 22. E. and the target PAM re- E-value cut-off of 0..Z. Gel filtration standards were run on the same column products were mixed with 1 ml 10 3 Taq DNA Polymerase PCR buffer (Enzy- equilibrated in 2M NaCl. 5% glycerol.O. sqrt(1 (b + c)/(a + b + c))). Reactions were cleaned up using PCR purification The prepared cDNA libraries were sequenced on a MiSeq with a single-end columns (QIAGEN) and were run on 2% agarose E-gels (Life Technologies). and b and c are the integrated intensities of each cleavage Lysate was sonicated for 10 min in a Biorupter sonicator (Diagenode) and then product. Indels were mapped using a Python implementation For native and denaturing gels to analyze cleavage by nuclease mutants. The Superdex 200) via FPLC (AKTA Pure).E. 5 mM as unique sample-specific barcodes to the target amplicons....1 expression manufacturer’s recommended protocol and analyzed on 4%–20% Novex plasmid by Genscript (Table S1). products were treated with SURVEYOR nuclease Cpf1 proteins codon optimized for human expression were synthesized with and SURVEYOR enhancer S (Integrated DNA Technologies) following the a C-terminal nuclear localization tag and cloned into the pcDNA3.25 C/s.org/10. 100 mM KCl. K. F. Yakunin. Horvath. T. eds. the authors plan to make the reagents widely available to the academic com. S. J. Li.R. Procter. F. and Pourcel. Barrangou.. S. Boyaval. J. J. Flamm. Villion. Mol. C. Ma.. Dugar.. Rev..L. bacteria and archaea.M.. Moineau. Charpentier. CRISPRFinder: a web tool to identify clustered regularly interspaced short palindromic repeats.... R. van Erp. secondary structure prediction server. Zhao.. Salzberg. October 22. J.. WebLogo: a Mol. K. (2010)..J. Barrangou. Structural biology. J. O. S. E. Boyaval. Romero. Brouns. (2012). 18. Nature 419. Bar- Edgar... Biol. A.A.. W. R. e109213. 67–71.F.. R. Burrows-Wheeler transform.R.. 945–956. R. 233–239.A.. Makarova. and Sergei Shmakov and Gardner. (2004). et al. Computational Science Graduate Fellowship. 759–771.J. Zhou.. 47–75. Golden. RNA-guided RNA cleavage by a CRISPR RNA-Cas protein complex.. V. Microbiol. B.A. Clark. RNA- cleic Acids Res. M.. Konermann. Grissa. E. (2007). Science 315. and Lipman.. D. Macrae for critical reading of the manuscript. S. Jiang. G.. Revised: September 15.. Development and applications tance against viruses in prokaryotes..S. Magadán.. BIOLOGY.J... Nature 468.. Horvath. Investigating CRISPR RNA Biogenesis and Function Using RNA-seq. M. Gapped BLAST and PSI-BLAST: a new generation Horvath. Chylinski.. P. 25. J. Richards. and Brenner. Westra. Nat. Fine. L. Vergnaud.. 602–607. 39.H.J.. Tom Harriman.S. Shoaibi. R... E. S. Fremaux.. 827–832. R. H.. Science 348. Hauswirth. 1477–1481. BMC Bioinformatics 8. Li. A. J. K. (1988). Barrangou.. (2008). the immune system of of protein database search programs. 819–823. Sharma. Reyon. S. would like to thank Wayne Conlan and David S... L.H. Biotechnol. Allen.. http://dx.J.Z.A. Chan. D. evolutionary classification of CRISPR-Cas systems and cas genes.... The and New York Stem Cell Foundations. We Fu. 14. 9677–9686.A.. Lander. M.. Biogen- esis pathways of RNA guides in archaeal and bacterial CRISPR-Cas adaptive Jiang. and Barton.. of CRISPR-Cas9 for genome engineering. Nucleic Received: August 28. (2011). Hauer. E. Bioinformatics 25. PILER-CR: fast and accurate identification of CRISPR re- peats.. upgrade to adaptive immunity. scientific advisor for Editas Medicine and Horizon Discovery. Madden. and Koonin. 279–284.. Pirzada.. K. P. Wolf. FEMS Microbiol. 39. M. and Wilson.A.. J. Doug Daniels for kindly Nat. F. (2015). A programmable dual-RNA-guided DNA endonuclease in adaptive lyzed by procaryotic and eucaryotic DNA polymerases. and Pelletier.A. (2015).... We would like to thank R.. D.. Sequence of Plas- by the intramural program of the US Department of Health and Human modium falciparum chromosomes 2... F... Costa. The tracrRNA and Cas9 families of type II CRISPR-Cas immunity systems. M. 770 Cell 163. and CRISPR/Cas bacterial immune system cleaves bacteriophage and plasmid Bob Metcalfe. Vogel.. 2015 Acids Res. Small CRISPR RNAs guide antiviral defense in prokaryotes.H.R. K. K. D. V. Fonfara. Cheng. R.. Y... P. K. and P.I. MUSCLE: multiple sequence alignment with high accuracy rangou. Y. H.K.. Haft. L. M. E.J.M.. Nu. sequence logo generator.M are supported A.. Microbiol.J. Stadler.N. Jinek. tion by trans-encoded small RNA and host factor RNase III.H. 234–244. Makarova. Moineau. Services (to the National Library of Medicine). Science 339..A.. (2014).O. Graveley. Protospacer adjacent motif (PAM)-distal sequences Jiang.. Acad. CRISPR/Cas. Chylinski. P. R. is supported by a D. P.. 31. Crystal structure of the CRISPR RNA-guided sur- 960–964. Makarova. S. Terwilliger. (2004). R.org/10. JPred4: a protein 9.K. E2579–E2586. P.... Makarova. Lundgren. W. S. (2007). J.. P. Marraffini. (2014). S. Sander. Novel non-templated nucleotide addition reactions cata.F. X.. (2014). (2013).. Cole.. and Zhang.E.A. D.S. Schmeing. Cencic.. Ethier. CRISPR-Cas systems: Prokaryotes Hsu. M. Yuri Wolf for help with sequence analysis. (2009).. J.M. I.. Cox. (2015). Drozdetskiy. R. K. A. W52–W57.M.M.M. G.. G.. veillance complex from Escherichia coli. M... Bio- technol. and Marraffini. T. Dickman. C. (2014).. Chandonia. and Wiedenheft. Le Rhun. 32.. Rev.. Published online July 22. and Marraffini. Cascio. CRISPR provides acquired resis- Hsu. P. 3389–3402.. (2015). F.E. nipulations from Bacterial Immunity Systems. CRISPR RNA matura. Haft.. DNA targeting specificity of Barrangou..... D..E.P.B. M. Zhang. Fremaux. Nature 471. C.A. W389–W394. C. Charpentier. Simons. S. ribonucleoprotein complex mediates specific DNA cleavage for adaptive im- site (www. Multiplex genome engineering Lorenz. 10. C.E. Romero. Carlton. and Hofacker. STRUCTURAL immunity. Algorithms Crooks. Brouns. the NIMH (5DP1-MH100706). Slijkhuis. and K. W.J.J.W. Eckert.. Natl.. M. Z. 11 and 14. Habib. D. 2015.A. and Marraffini.. (2015). Lundgren.. Methods Mol. and Zhang.V.. E. S. Wells.D. Schäffer. Wu. P. L. Sci. 1792–1797. (2011).. Genome Res. Deveau. Gasiunas. Malina. Charpentier. Dostie. Science 327.A. W.J.. Westra. 531–534.. S. pp. I. C..D.. Charpentier..genome-engineering. E. et al. H. P. E.P. (2014)... 16.. (2015). (2007). E. F. B.. Agarwala. Snijders. J. Cox. Weiss for generously providing Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. 32... E. Shallom. R. Koonin. Nat. F... J. and DNA.. (Springer New York). F. D.C. R. Höner Zu Siederdissen... K. 1188–1190. (2011).H. R. (2002).. G. Nat.0.R. Wolf. Nucleic Acids Res. Vogel.. R. S. (1997).. J.Z. G. Zhang. Hsu. E. (2013).. USA 109.S. 10. Wensel.J.. A. Ciecko. Read. Efficient mutagenesis of the rhodopsin gene in rod photoreceptor neurons in mice. E. J. . 816–821. Zhang. Ran. and Charpentier. 2015 R. and van der Oost. Y.V.. F. Saunders.. T.. Shalem. Evolution and classification of the CRISPR-Cas systems. J..S. Doudna. Gonzales. Horvath. R. et al.. Weinstein. M. Jore.G. C. N.. guided editing of bacterial genomes using CRISPR-Cas systems. 31. Rev. (2013). (2015)... M..D. Ran. and Barrangou. J.. Nucleic Acids Res. A patent application has been filed related to this work. 167–170. J. A.... F. Lynn. 43. Fast and accurate short read alignment with Cong. N.G. Altschul.. Chylinski.doi.. F. Carter. providing us with the bacterial expression vector. Bernhart. Cell 139.. 1262–1278.C. R. C.. (2009).S.org).J.F. (2010). Cell 157. A Cas9-guide RNA complex preorganized for target DNA recogni- tion. Chao. van der Oost. Miller. S.M.. 1–21. A. Dupuis. Rizzo. (2013).. Tafer. 1473–1479. S.. and Charpentier.. J.S. F. Paul G. 428–441. and Horvath.. J. 1754–1760. CRISPR-Cas: New Tools for Genetic Ma- engage CRISPR Cas9 DNA target cleavage.. and Durbin. Robert. Nucleic Acids Res. REFERENCES Heidrich. (2012). H. and Siksnys... E...V.A. H. Duff. and Moineau. 1709–1712. E. T. L. and Terns.S.. 26. S. Brouns.M. and Charpentier.. K. Barrangou.. Gressel. Cas9-crRNA munity through Addgene and to provide software tools via the Zhang lab web. et al.. PLoS ONE 9. S. 5955–5966. Cell 54. E. A. 726–737. 1311.J. Biol. Terns. Nat. J. David R. Barretto. 6. J. RNA Biol. Published: September 25. A.. RNA-guided Cas9 nucleases. Mojica. and Doudna. Science 337. Richter..O. novicida genomic DNA and cell pellets. Biotechnol. J.. X. Science 345. Lin. 467–477. P. M. Wu. Fineran. Miura. Vallee. Weaver. bacterial immunity..L. respectively.A. 35.J. Rev.C. M. 2015 Accepted: September 17. L. and Sharma.E. ViennaRNA Package 2. 2015 Hale. S. 2015 ª2015 Elsevier Inc. is supported by Garneau. Olson. van der Oost.R. (2011). Annotation and classification of Deltcheva. Annu. D.1146/annurev-micro-091014-104441. D. Updated and high throughput. V.S.J. is a founder of Editas Medicine and a munity in bacteria.. Jackson.C.. Nucleic Acids Res. In CRISPR. W. Brouns.M. P. I..I.. Alkhnbashi.L. C.F. Nene.D. CRISPR-Cas systems. Jiang. Y. B. P. Y... Hon.. the Poitras Center. F. Science 321. Z. Scott. Proc. G. S. V. O. and White. and Joung.ACKNOWLEDGMENTS Edgar. using CRISPR/Cas systems. L. Bikard. M.. Konermann.. W374–W378. 488–503. K.. 385–394. E. Gunderson. O.. T.. Siksnys. Horvath... Int. K. editing using Staphylococcus aureus Cas9. V. 935–949.P.. J.A.. Rydzewski. Grunow.. Dı́ez-Villaseñor. e9490. Nature 520. and Bailey. P. 51–60.J. W.S. and Nureki. ated CRISPR immunity in Streptococcus thermophilus. S. and Lupas.M. Genome Res.. 34. Biegert. A.. Rev. B.. (2014).D.. Schoen.X.doi. A. G. A.. Garcı́a-Martı́nez.W.. J. H. R. (2010). Dickman. Microbiology 155. Makarova. Yan.1038/ Sapranauskas. 23. 733–740. and Siksnys. horizontal gene transfer in staphylococci by targeting DNA. Published online September 29.Microbiol.. RNA Biol.J. Gasiunas. Fremaux. CRISPR-mediated adaptive immune systems in bacteria and archaea... V. (2011). Seifert. Vogel. G. S..J. Horvath. recombination (ObLiGaRe): custom-designed nuclease-mediated targeted Schunder. Processing-independent sche. Crystal Acids Res.J. In vitro reconstitution of Cascade-medi- Mojica. 759–771.S. F. 237–266. 9275–9282. G. F.G. M. K. C. Héroux.. and Sontheimer. Science 345. Garrett. J. Ampattu. structure of a CRISPR RNA-guided surveillance complex bound to a ssDNA target. S.. P. R. (2008).. and Shah. Sorek. B. S.. O. Barrangou. Scott. Cell 156. P.. L. (2006). C. Shehata. (2013).A. Remmert.... C.. R. B. M. Wu. Y. Y. 32. F. N. E.. Biochem. 2015.S. and Arkin. In vivo genome CRISPR RNAs limit natural transformation in Neisseria meningitidis. 11. 303. E. A. Med.. maximum-likelihood trees for large alignments.. Ran.. Obligate ligation-gated vides immunity in Escherichia coli. The Streptococcus thermophilus CRISPR/Cas system pro- Maresca. S... C. M. Zet. Nucleic Mulepati. exhaustive transitive profile search using HMM-HMM comparison. EMBO J.J. Lin. L.. 1843–1845. October 22. Cong. P. (2014). Crystal structure of Cas9 in complex with guide RNA and target DNA. N. and nrmicro3534.. 1479–1484. et al.A. FastTree 2–approximately mune systems of Archaea. Cell 163. Dohmae. (2013). D. Heidrich.J. (2013). 39.P. (2013). 82. tion for a functional CRISPR/Cas system in Francisella tularensis. A... S. H. Dehal.. http://dx. R.A.N. Zhang. Lawrence. CRISPR adaptive im- Price. Nucleic Acids Res. Short motif sequences determine the targets of the prokaryotic Söding. 186–191.. R. F. First indica- integration through nonhomologous end joining. HHsenser: CRISPR defence system.. (2013). (2014). Marraffini. Structural biology... Mol. Gasiunas. 156–167.A.. M. Cell 50.. CRISPR interference limits Microbiol...A.S. N. Waghmare. Nishimasu.. J.I. X. Hsu. 539–546. Gootenberg. PLoS ONE 5. and Almendros. (2015). (2009). Zhang. R. Ishitani. Ran..N... Vestergaard.. C. and Yang. Science 322. 2015 ª2015 Elsevier Inc. C. and Wiedenheft. R.. and Sontheimer. Sinkunas. Shalem.. Kriz. Barrangou. J. Annu.org/10. and Heuner... Guo.. V. 771 .