Brief CommuniCationhttps://doi.org/10.1038/s41592-018-0262-1

Helmholtz-University Group ‘Cell Plasticity and Epigenetic Remodeling’, German Cancer Research Center (DKFZ) and Institute of Pathology, University Hospital, Heidelberg, Germany. *e-mail: d.tschaharganeh@dkfz.de

CRISPR–Cas9-based combinatorial perturbation approaches for orthogonal knockout and gene activation have been impeded by complex vector designs and co-delivery of multi-ple constructs. Here, we demonstrate that catalytically active CRISPR–Cas12a fused to a transcriptional-activator domain enables flexible switching between genome editing and tran-scriptional activation by altering guide length. By leveraging Cas12a-mediated CRISPR-RNA array processing, we illustrate that Cas12a-VPR enables simplified multiplexed knockout and transcriptional activation in vitro and in vivo.

Genes typically operate in networks, in which the effect of a gene’s alterations is determined by interacting genes1. Recently, CRISPR–Cas9-based genetic tools have enabled genetic interaction analyses by combinatorial knockouts, and contemporaneous gene activation and knockout or repression2–9. However, Cas9-based combinatorial perturbations require complex vector designs and simultaneous delivery of multiple plasmids, thus posing hurdles to multiplex in vivo applications.

To provide a simplified alternative for orthogonal gene control, we hypothesized that Cas12a, which enables multiplexing by pro-cessing its own CRISPR RNA (crRNA) arrays10,11, can be engineered to execute both gene editing and transcriptional activation, as has previously been done with Cas9 (ref. 8). Because loss of Cas12a activ-ity with truncated guides12–14 is not based on a complete absence of DNA binding15, we fused a transcriptional-activator complex (VPR) to catalytically active Cas12a from Acidaminococcus sp. BV3L6 (As) and Lachnospiraceae bacterium ND2006 (Lb) to investigate whether this system might enable orthogonal gene control dictated by guide length (Supplementary Fig. 1).

To initially test Cas12a-VPR for transcriptional activation, we used a tetracycline-response element (TRE)-driven RFP-fluorescence reporter system (Supplementary Fig. 2a,b) and inves-tigated the effect of guide length on AsCas12a-VPR-mediated RFP activation. Strikingly, in the absence of doxycycline (off Dox), RFP activation was achieved with 15- and 12-nt TRE-targeting guides, whereas 20- or 17-nt TRE guides did not induce RFP (Fig. 1a). For 20- or 17-nt TRE guides, we observed decreased RFP expression in on-Dox cells (Fig. 1a; 24 h doxycycline treatment), and PCR amplicons of the TRE promoter revealed multiple bands indicative of impaired promoter activity due to deletions (Supplementary Fig. 2c). A transcriptional-activation-dependent cell-growth reporter assay confirmed AsCas12a-VPR-mediated activation with 15-nt guides (Supplementary Fig. 2e). In cells expressing AsCas12a lack-ing VPR, 15- and 12-nt TRE guides did not induce reporter activa-tion, whereas 20- or 17-nt guides decreased TRE promoter activity in on-Dox conditions (Supplementary Fig. 2c–e).

Notably, RFP-reporter assays indicated that, for transcriptional activation, AsCas12a-VPR performed better than LbCas12a-VPR, and direct repeats within crRNAs were not interchangeable between these orthologs (Supplementary Fig. 3). RFP-reporter assays also showed that AsCas12a-VPR performed as well as other CRISPR-based transcriptional activators8,16 (Supplementary Fig. 4).

We next tested the AsCas12a-VPR-mediated transcriptional activation of endogenous genes in human HEK293T cells and observed that transient transfection of 15-nt guides targeting the promoter regions of NPY1R or TTN induced transcriptional activa-tion (Fig. 1b). AsCas12a-VPR-mediated transcriptional activation with 15-nt guides was also achieved for five additional endogenous genes (Supplementary Fig. 5a). Furthermore, a comparison of tran-scriptional activation of endogenous genes with AsCas12a-VPR, LbCas12a-VPR, and deadLbCas12a-VPR (Supplementary Fig. 5b) corroborated our findings in RFP-reporter cells.

We observed that, in contrast to the results of RFP-reporter experiments, promoter-targeting 20-nt guides also mediated tran-scriptional activation of endogenous genes, although to a lesser degree (Fig. 1b). We reasoned that this effect might be based on persistent binding of AsCas12a-VPR to target loci, because Cas12a leaves the 5′ protospacer-adjacent motif and large regions of the guide-matching DNA sequence unaltered12. Therefore, we tested whether 20-nt guides targeted to exons, the preferred loci to induce indels for knockout approaches, might induce transcriptional acti-vation. Using TTN, NPY1R, and MYOD1 as representative targets, we observed that 20-nt exon-targeting guides did not result in pro-nounced activation comparable to that of 15-nt or 20-nt promoter-targeting guides. However, for TTN, activation was still observed (approximately fivefold activation for 20-nt exon-targeting guides versus ~ 100-fold activation for 15-nt promoter-targeting guides; Supplementary Fig. 6). Corroborating previous reports12,14, we detected indels for 20-nt guides, whereas indel frequencies were dramatically decreased or absent for corresponding 15-nt guides (Fig. 1c). Together, these results demonstrate that AsCas12a-VPR enables transcriptional activation with 15-nt guides, whereas 20-nt guides produce indels.

Encouraged by our in vitro results, we determined whether AsCas12a-VPR enables transcriptional activation in vivo. We therefore integrated AsCas12a-VPR and crRNAs into Sleeping Beauty–transposon vectors and delivered these plasmids along with Sleeping Beauty transposase via hydrodynamic tail-vein injection (HDTV)17 into the hepatocytes of mice bearing a TRE-GFP reporter in the Col1a1 locus (Fig. 1d). Strikingly, we observed GFP-positive hepatocytes in mice injected with a 15-nt TRE guide, but not with a corresponding 20-nt TRE guide, although BFP staining indicated

Multiplexed orthogonal genome editing and transcriptional activation by Cas12aMarco Breinig, Anabel Y. Schweitzer, Anna M. Herianto, Steffie Revia, Lisa Schaefer, Lena Wendler, Ana Cobos Galvez and Darjus F. Tschaharganeh   *

NATuRe MeTHoDS | www.nature.com/naturemethods

Brief CommuniCation NAtuRe MetHodS

crRNA expression for each treatment (Fig. 1d and Supplementary Fig. 7a). To test whether AsCas12a-VPR enables transcriptional activation of endogenous genes in vivo, we used HDTV of crRNAs with a 15-nt Rosa26 guide or 15-nt Krt19 guides to induce Krt19 expression in mouse hepatocytes. We detected foci of Krt19-expressing hepatocytes (arrows, Supplementary Fig. 7b) only in mice injected with 15-nt Krt19 guides, whereas Krt19-positive cholangiocytes (asterisks, Supplementary Fig. 7b) were observed in both conditions, in agreement with Krt19 being a marker for chol-angiocytes17. Additionally, we confirmed that AsCas12a-VPR gen-erates knockouts in vivo, because HDTV of an in vitro–validated 20-nt Trp53 guide in conjunction with enforced c-Myc expression led to tumor development as well as indel formation at the Trp53 locus and a loss of P53 expression in tumor cells (Supplementary Fig. 8a–i). Thus, AsCas12a-VPR enables transcriptional activation and knockout approaches in vivo.

Next, we investigated the multiplexing capabilities of the AsCas12a-VPR system. First, we used the cell-based TRE-RFP reporter and constructed crRNA arrays bearing a 15-nt TRE guide in alternating positions. RFP activation was observed for all arrays tested; however, activation was generally weaker when the 15-nt TRE guide was expressed from an array (Fig. 2a). Similar results were obtained with our cell-growth reporter assay (Supplementary Fig. 9a–c). To investigate the positional effects of 20-nt loss-of-function guides when embedded in arrays, we tested AsCas12a-VPR-mediated knockout and observed efficient Trp53 loss of function for all arrays tested (Supplementary Fig. 9d,e).

Thus, AsCas12a-VPR-mediated transcriptional activation or knockout can be achieved by array-based expression of 15-nt guides or 20-nt guides, respectively.

We then focused on multiplex gene activation in human HEK293T cells by embedding 15-nt guides for TTN and NPY1R activation in arrays, and we indeed observed simultaneous gene activation (Fig. 2b). In agreement with the results of our reporter assays, TTN activation was weaker when guides were expressed from arrays; however, this result was not the case for NPY1R acti-vation. Multiplexed activation was further shown for MYOD1 and MIAT (Supplementary Fig. 10). Thus, 15-nt guides embedded in crRNA arrays enable AsCas12a-VPR-mediated multiplexed tran-scriptional activation of endogenous genes.

Beyond using multiplexing to concurrently activate differ-ent genes, simultaneous targeting of multiple loci in the promoter region of a single gene can enhance the performance of CRISPR-based transcriptional activation16,18. This effect was observed for the AsCas12a-VPR system when three 15-nt guides were expressed from an array (Supplementary Fig. 11).

We next tested AsCas12a-VPR-mediated orthogonal gene con-trol in HEK293T cells by simultaneously activating TTN with 15-nt guides and inducing indels in the NPY1R promoter locus with 20-nt guides. Using the respective arrays, we observed TTN activation (Fig. 2c) alongside indels in the NPY1R locus, with frequencies comparable to that of single NPY1R-targeting crRNA (Fig. 2d). Orthogonal gene control was further found for additional endog-enous targets (Supplementary Fig. 12). Hence, AsCas12a-VPR

On DoxOff DoxGuidea

c d

b

DR

U6 EF1 BFP

IR

IR

EFs

IRAsCas12a-VPR

CMV SB

TRE (20 nt)

TRE (17 nt)

TRE (15 nt)

Rosa26 (20 nt)

Rosa26 (15 nt)

RFP-positive cells (%)

020 600 20 4040

GFPTRE

TRE (12 nt)

DAPI

20 nt TRE

15 nt TRE

GFPGuide

Guide

Length 20 nt 15 nt

TTN

Rel

ativ

e m

RN

A e

xpre

ssio

n

0

25

50

75

100

NPY1R

0

5

10

15

1 2

NPY1RTTNLoci:

1

% in

del

20

10

5

1

4.82%

7.55%

0.64%

20.04%22.29%

0% 0% 0%0

IR

1 2

20 nt 15 ntLength

Guide 1

1

Fig. 1 | Transcriptional activation with catalytically active AsCas12a-VPR and 15-nt guides. a, Flow cytometry–based RFP quantification in mouse reporter cell lines stably expressing AsCas12a-VPR transduced with crRNAs expressing different guide lengths targeting the TRE promoter in off-Dox or on-Dox conditions. Data are shown as mean ±  s.d.; n =  3 independent experiments. b, Relative mRNA expression after cotransfection of HEK293T cells with AsCas12a-VPR and crRNAs. Two separate crRNAs (20-nt and corresponding 15-nt guides) were targeted to the promoters of TTN or NPY1R. Data are shown as mean ±  s.d.; n =  3 independent experiments. c, Indel formation after cotransfection of HEdK293T cells with AsCas12a-VPR and crRNAs. Two separate crRNAs (20-nt and corresponding 15-nt guides) were targeted to the promoters of TTN or NPY1R. n =  2 independent experiments. d, HDTV of AsCas12a-VPR, crRNA-expressing vectors, and SB into TRE-GFP-reporter knock-in mice. crRNAs with a 20-nt TRE guide and the corresponding 15-nt TRE guide were tested. At 2 weeks after HDTV, the expression of GFP in livers was investigated through immunofluorescence. DAPI staining visualizes nuclei. n =  3 mice per group (representative images shown). IR, inverted repeat; DR, direct repeat; SB, CMV-SB13 transposase; EF, elongation factor-1 promoter. Scale bars, 100 µ m.

NATuRe MeTHoDS | www.nature.com/naturemethods

Brief CommuniCationNAtuRe MetHodS

and crRNA arrays enable multiplexed orthogonal gene control in mammalian cells.

Next, we used dual fluorescence-reporter cells stably express-ing BFP alongside TRE-driven RFP and generated arrays to target AsCas12a-VPR for RFP activation and BFP knockout. Flow cytom-etry analyses revealed that simultaneous activation and knockout were achieved in a substantial fraction of cells (Supplementary Fig. 13), thus indicating that orthogonal gene control can be accomplished within the same cell.

Furthermore, we investigated whether transcriptional activa-tion and orthogonal gene control of endogenous genes could be established by using lentiviral delivery of crRNAs. Although these experiments largely recapitulated the findings obtained for transient transfection of crRNAs, we observed that activation strength was generally weaker with lentiviral-mediated expression of crRNAs

(Supplementary Fig. 14). These results indicate that different deliv-ery routes for crRNA expression can impair the efficacy of orthogo-nal gene control with AsCas12a-VPR.

Finally, we investigated whether AsCas12a-VPR–mediated orthogonal gene control is possible in vivo by simultaneously target-ing the TRE promoter for activation and Trp53 for indel formation in TRE-GFP-reporter mice. Strikingly, we observed GFP-positive hepatocytes together with indels in the desired Trp53 locus, by using an array in which a 20-nt Trp53 guide was combined with a 15-nt TRE guide (Fig. 2e and Supplementary Fig. 15). Together, these findings demonstrate in vivo orthogonal gene control with a single CRISPR effector and array-based guide multiplexing.

The ability of AsCas12a-VPR to activate genes while main-taining DNA cleavage activity, together with the simplicity of Cas12a-mediated multiplexing, is particularly appealing for in vivo

Arraya

d e

b c

15 nt TRE guide

20 nt guide

15 nt guide

DR

RFP-positive cells (%)

10 200

NPY1RTTN

15 1515

15 1515

TTN

NPY1R

Rel

ativ

e m

RN

A e

xpre

ssio

n

Rel

ativ

e m

RN

A e

xpre

ssio

n

5

12.5

25

50

100

0

NPY1R locus

15

15

20

15

2020

20

5

1

% in

del

10

15 15

20

15

20 20

25

50

100

0

75

IR

EFs

IR

AsCas12a-VPRCMV SB

GFPTRE

DRIR

U6 EF1 BFP

IR

Length (nt); position: 1,2

0

TTN

NPY1R

Length (nt); position: 1,2

TTN

NPY1R

Length (nt); position: 1,2

NPY1RTTN

DAPI BFP GFP Merge

0.440.29 207/69,906

307/69,906

PAM 20-nt target sequence % reads Read countTrp53 locus

crRNA array: 15-nt TRE guide + 20-nt Trp53 guide

1

Fig. 2 | Multiplexed transcriptional activation and orthogonal gene control by catalytically active AsCas12a-VPR and crRNA arrays. a, Flow cytometry–based RFP quantification in mouse reporter cell lines stably expressing AsCas12a-VPR transduced with crRNA arrays. A 15-nt TRE guide was incorporated into varying arrays as indicated. Data are shown as mean ±  s.d.; n =  3 independent experiments. b,c, Relative mRNA expression after cotransfection of HEK293T cells with AsCas12a-VPR and crRNA arrays. b, 15-nt guides targeting the TTN and NPY1R promoter were incorporated into arrays as indicated. Data are shown as mean ±  s.d.; n =  3 independent experiments. c, Multiplex expression of a 20-nt guide targeting NPY1R and a 15-nt guide targeting the TTN promoter, with arrays as indicated, versus the respective single crRNAs. Data are shown as mean ±  s.d.; n =  3 independent experiments. d, Indel formation after cotransfection of HEK293T cells with AsCas12a-VPR and crRNA arrays. Multiplex expression of a 20-nt guide targeting NPY1R and a 15-nt guide targeting the TTN promoter, with a single array versus the respective single crRNAs. n =  2 independent experiments. e, HDTV of AsCas12a-VPR, crRNA-array-expressing vectors, and SB into mice stably expressing a TRE-GFP reporter. Arrays for multiplex expression of a 20-nt Trp53 guide combined with a 15-nt TRE guide. GFP expression in livers taken 8 weeks after HDTV was visualized by immunofluorescence. BFP staining indicates crRNA expression. DAPI staining visualizes nuclei. PAM, protospacer-adjacent motif. n =  5 mice (representative images shown). Scale bars, 80 µ m. Indel occurrence at the Trp53 locus. n =  2 independent liver samples (representative results shown).

NATuRe MeTHoDS | www.nature.com/naturemethods

Brief CommuniCation NAtuRe MetHodS

applications in which simultaneous delivery of multiple plasmids poses a roadblock to systematically modeling complex genotypes. However, we observed limitations of the system, such as lower acti-vation strength when guides are expressed from arrays or delivered lentivirally. Thus, the particular characteristics that define highly functional crRNA arrays for multiplexed gene control remain to be further elucidated. As shown previously11, the performance of crRNA arrays might be enhanced when expressed from a RNA polymerase II promoter, and improved Cas12a guide design19 is likely to contribute to the optimization of our system.

Nonetheless, we anticipate that the AsCas12a-VPR system may open up new avenues to deciphering the higher-order genetic interactions that underlie phenotypic traits, in both healthy and diseased states.

online contentAny methods, additional references, Nature Research reporting summaries, source data, statements of data availability and asso-ciated accession codes are available at https://doi.org/10.1038/s41592-018-0262-1.

Received: 5 June 2018; Accepted: 5 November 2018; Published: xx xx xxxx

References 1. Costanzo, M., Baryshnikova, A., Myers, C. L., Andrews, B. & Boone, C.

Curr. Opin. Biotechnol. 22, 66–74 (2011). 2. Zalatan, J. G. et al. Cell 160, 339–350 (2015). 3. Gao, Y. et al. Nat. Methods 13, 1043–1049 (2016). 4. Du, D. et al. Nat. Methods 14, 577–580 (2017). 5. Han, K. et al. Nat. Biotechnol. 35, 463–474 (2017). 6. Najm, F. J. et al. Nat. Biotechnol. 36, 179–189 (2018). 7. Boettcher, M. et al. Nat. Biotechnol. 36, 170–178 (2018). 8. Kiani, S. et al. Nat. Methods 12, 1051–1054 (2015). 9. Dahlman, J. E. et al. Nat. Biotechnol. 33, 1159–1161 (2015). 10. Zetsche, B. et al. Nat. Biotechnol. 35, 31–34 (2017). 11. Zhong, G., Wang, H., Li, Y., Tran, M. H. & Farzan, M. Nat. Chem. Biol. 13,

839–841 (2017).

12. Zetsche, B. et al. Cell 163, 759–771 (2015). 13. Kleinstiver, B. P. et al. Nat. Biotechnol. 34, 869–874 (2016). 14. Kim, H. K. et al. Nat. Methods 14, 153–159 (2017). 15. Singh, D. et al. Proc. Natl. Acad. Sci. USA 115, 5444–5449 (2018). 16. Tak, Y. E. et al. Nat. Methods 14, 1163–1166 (2017). 17. Tschaharganeh, D. F. et al. Cell 158, 579–592 (2014). 18. Perez-Pinera, P. et al. Nat. Methods 10, 973–976 (2013). 19. Kim, H. K. et al. Nat. Biotechnol. 36, 239–241 (2018).

AcknowledgementsWe thank all members of the laboratory of D.F.T. for insightful comments on this work. We thank the DKFZ Genomics Core Facility, the DKFZ Central Animal Laboratory, and the CMCP of the Institute of Pathology Heidelberg for excellent technical support. This work was supported by the Helmholtz Association (VH-NG-1114), the Helmholtz Association ERC Recognition Award, and the German Research Foundation (DFG) project B05, SFB/TR 209 ‘Liver Cancer’ (all awarded to D.F.T.). We thank F. J. Sánchez-Rivera (Cancer Biology and Genetics, Memorial Sloan Kettering Cancer Center), G. Church (Department of Genetics, Harvard Medical School), E. Welker (Institute of Biochemistry, Biological Research Centre, Hungarian Academy of Sciences), F. Zhang (Department of Biological Engineering, Massachusetts Institute of Technology), and D. Trono (School of Life Sciences, Ecole Polytechnique Federale de Lausanne (EPFL)) for providing reagents.

Author contributionsM.B. performed experiments, analyzed data, designed the study, and wrote the manuscript. S.R., A.Y.S., A.M.H., A.C.G., L.S., and L.W. performed experiments. D.F.T. performed experiments, analyzed data, designed the study, and wrote the manuscript.

Competing interestsThe authors declare no competing interests.

Additional informationSupplementary information is available for this paper at https://doi.org/10.1038/s41592-018-0262-1.

Reprints and permissions information is available at www.nature.com/reprints.

Correspondence and requests for materials should be addressed to D.F.T.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© The Author(s), under exclusive licence to Springer Nature Limited 2018

NATuRe MeTHoDS | www.nature.com/naturemethods

Brief CommuniCationNAtuRe MetHodS

MethodsVector construction. pLenti-EFs-hCas9-VPR-PGK-puro (pMB1) was constructed by PCR amplification of Cas9-VPR from Addgene plasmid 68497 and cloning into pLenti-PGK-puro digested with NheI and AvrII with NEBuilder HiFi DNA Assembly (NEB) according to the manufacturer’s protocol. pLenti-EFs-dCas9-VPR-PGK-puro was a kind gift from F. J. Sánchez-Rivera. pT3-EFs-hCAS9-VPR (pMB19) was constructed by PCR amplification of hCas9-VPR from pMB1 and cloning into pT3-EFICS digested with NheI and NsiI with NEBuilder HiFi DNA Assembly. pLenti-EFs-AsCas12a-VPR-2A-puro (pMB5) and pLenti-EFs-LbCas12a-VPR-2A-puro (pMB26) were constructed through NEBuilder HiFi-DNA Assembly with a synthesized VPR sequence (gBlock, IDT) into Addgene plasmid 84739 or Addgene plasmid 84740 digested with BamHI, according to the manufacturer’s protocol. pLenti-EFs-AsCas12a-VPR-PGK-puro (pMB22) and pLenti-EFs-LbCas12a-VPR-PGK-puro (pMB34) were constructed via PCR amplification of AsCas12a-VPR or LbCas12a-VPR from pMB5 or pMB26, respectively, and cloning into pLenti-PGK-puro digested with NheI and AvrII with NEBuilder HiFi DNA Assembly. pT3-EFs-AsCas12a-VPR (pMB21) and pT3-EFs-LbCas12a-VPR (pMB33) were constructed by PCR amplification of EFs-AsCas12a-VPR from pMB22 and EFs-LbCas12a-VPR from pMB34 and cloning into pT3-EFICS digested with NheI and NsiI with NEBuilder HiFi DNA Assembly. pLKO-antisenseU6-AsDR-BsmBI (pMB4) was constructed through NEBuilder HiFi DNA Assembly with a synthesized antisenseU6-AsDR-BsmBI sequence (gBlock IDT) into plko-EF1s-GFP-2A-Blas digested with KflI and XhoI. pLKO-antisenseU6-LbDR-BsmBI (pMB30) was constructed likewise. pT3-antisenseU6-AsDR-BbsI (pMB13) and pT3-antisenseU6-LbDR-BbsI (pMB32) were constructed likewise with synthesized DNA fragments cloned into pT3-EF1a-BFP-mirE digested with NsiI and PacI. pLenti-deadLbCas12a-VPR-PGK-puro (pMB36) was constructed through AarI digestion of Addgene plasmid 80441 to obtain the LbCas12a fragment bearing the inactivating D832A mutation and ligation of this fragment into AarI-digested pMB34. pT3-deadLbCas12a-VPR (pMB37) was likewise constructed with AarI-digested pMB33. hCas9-VPR (Addgene plasmid 68497) was a gift from G. Church; WN10151 (Addgene plasmid 80441) was a gift from E. Welker; and pY108 (lenti-AsCas12a, Addgene plasmid 84739) and pY109 (lenti-LbCas12a, Addgene plasmid 84740) were gifts from F. Zhang. Relevant plasmids used in this study will be made available via Addgene.

Cas12a guides were designed with chopchop20 with TTTV as the PAM14. Unless otherwise indicated, a guide length of 20-nt was used for indel formation, and a length of 15-nt was used for transcriptional activation. Guides were cloned by oligonucleotide annealing and standard ligation. Briefly, top and bottom oligonucleotides were annealed and phosphorylated. Thereafter, they were ligated into either pMB4 or pMB13 digested with BsmBI or either pMB30 or pMB34 digested with BbsI. Sanger sequencing (GATC Biotech) was used to validate plasmids. Sequences for all guides and oligonucleotides are listed in Supplementary Tables 1 and 2.

Cell culture. HEK293T cells were obtained from the ATCC, and grown at 37 °C, under 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM, Sigma) with 10% FBS and 1% penicillin and streptomycin. HEK293T cells were further modified to stably express AsCas12a-VPR with lentiviral transduction and selection with puromycin. HEK293T cells were split twice per week at a ratio of 1:10 to 1:20.

Mouse liver cancer cell lines were derived from liver tumors established after HDTV. Briefly, genotypes comprised NRAS overexpression and simultaneous p53 knockdown under on-Dox conditions (NP5 cells) or Myc overexpression and simultaneous p53 knockdown under on-Dox conditions (MP2 cells).

NP5 and MP2 cells were cultivated in DMEM, with 10% FBS and 1% penicillin and streptomycin, and doxycycline (VWR; 1 µ g/ml). They were split twice per week at a ratio of 1:20 with collagen-precoated (PureCol, Cell Systems; 0.05 mg/ml) 10 cm dishes (Greiner).

NP5 cells were further modified to stably express constitutive BFP and shp53 (NP5-BFP + constitutive shp53). NP5 cells were further modified to stably express AsCas12a-VPR, LbCas12a-VPR, or deadLbCas12a-VPR with lentiviral transduction and selection with puromycin. NP5 (BFP + constitutive shp53) cells were further modified to stably express AsCas12a-VPR, LbCas12a-VPR, or deadLbCas12a-VPR with lentiviral transduction and selection with puromycin. MP2 cells were further modified to stably express dCas-VPR or Cas9-VPR with lentiviral transduction and selection with puromycin. Cell lines were annually tested for mycoplasma contamination (GATC Biotech).

Lentivirus production. 1.5 million HEK293T cells were seeded in 2 ml medium into each well of a six-well plate. The next day, the cells were transfected with a plasmid mix of 0.375 µ g pMD.2 G, 1.2 µ g psPAX2 (Addgene plasmids 12259 and 12260, both gifts from D. Trono), and 1.5 µ g pMB4 bearing the respective guides in 150 µ l DMEM with 9 µ l polyethylenimine (PEI, 1 µ g/µ l). After vortexing for 5 s and a 30 min incubation, the plasmid mix was added dropwise to cells. 24 h later, the medium was exchanged, and lentiviral supernatant was harvested 48 h after transfection. The supernatant was sterile-filtered with 0.45-µ m Cellulose Acetate Membrane filters (VWR) and stored at –80 °C until use.

Lentiviral transduction. Target cells were seeded in 2 ml medium at a concentration of 10,000 cells per well of a six-well plate. The next day, the medium was exchanged with fresh medium containing polybrene (4 µ g/ml), and viral supernatant was added. Cells stably expressing AsCpf-1VPR, LbCas12a-VPR, Cas9-VPR, dCas9-VPR, or deadLbCas12a-VPR were generated by lentiviral transduction with an MOI of ~ 0.5, as calculated by cell growth after antibiotic selection and serial dilutions of lentiviral supernatant. 3 d after transduction, cells were further selected with puromycin (4 µ g/ml). Unless otherwise indicated, for crRNA expression with pMB4 or pMB30, cells were transduced with lentiviral supernatant at an MOI of ~ 0.3, as determined by flow cytometry analysis of GFP expression. 3 d after transduction, cells were selected with blasticidin (10 µ g/ml) for 3 d. Thereafter, cells were grown under off-Dox or on-Dox conditions as indicated. If not indicated otherwise, cells were used for flow cytometry analysis and lysed for RNA, genomic DNA, or protein extraction 10 d after transduction.

Transient transfection. 1.5 million HEK293T cells were seeded in 2 ml medium into each well of a six-well plate. The next day, cells were transfected with PEI. Briefly, 1.5 µ g plasmid pMB21 and 1.5 µ g plasmid pMB13 or pMB4 bearing the respective guides were mixed in 150 µ l DMEM. Then 9 µ l PEI was added, and the mix was vortexed for 5 s. After an incubation of 30 min, the plasmid mix was added dropwise to cells. After transfection, cells were grown for 96 h and lysed for either RNA or genomic DNA extraction.

Indel-frequency detection via T7E1 assays. Briefly, genomic DNA was extracted with a Puregene Core Kit A (Qiagen), and target loci were PCR amplified with Q5 Hot Start DNA Polymerase (New England BioLabs) with 100–200 ng of genomic DNA as a template. The primers used for PCR amplification are listed in Supplementary Table 3. After purification (PCR Purification Kit (Qiagen)), 250–400 ng of PCR product was denatured, reannealed, and digested with T7E1 (New England BioLabs), according to the manufacturer’s recommended protocol, and analyzed on 2–3% agarose gels or 10% PAGE gels. Gels were stained with EtBr to visualize DNA.

Indel-frequency detection. PCR was performed on genomic DNA to amplify the target loci, as described for the T7EI assay. The primers used for PCR amplification are listed in Supplementary Table 4. PCR reactions were purified (PCR Purification Kit (Qiagen)) and controlled for specificity (i.e., detection of a single band) with 2–3% agarose gels. NGS and indel detection were performed by the MGH CCIB DNA Core Facility. Data were inspected manually, and reads with single base alterations were not considered for analysis of indel frequency.

Quantitative reverse-transcription PCR (qPCR). Total RNA was extracted with an RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. 1 µ g of purified total RNA was converted to cDNA with TaqMan Reverse Transcription Reagents (Thermo). cDNA was diluted 1:20, and 1–5 µ l of cDNA was used for qPCR. The qPCR reaction samples were prepared with cDNA, PowerUp SYBR Green Master Mix (Thermo Fisher), and primers detecting each target transcript. Primers are listed in Supplementary Table 5. Each qPCR was performed in triplicate with a StepOnePlus system (Applied Biosystems). Samples that were transfected with pMB4 empty vector were used as negative controls, and the fold activation over negative controls was normalized to the expression of RPL41 (for human cells) or Gapdh (for mouse cells). Transcript levels were calculated with the deltaCt method. Because Ct values fluctuate for very low expressed transcripts, Ct values greater than 40 were considered as 40.

Flow cytometry. GFP and RFP were measured with a Guava easyCyte benchtop flow cytometer (Merck). Cells were washed with PBS, trypsinized, and resuspended in DMEM to a total volume of 1 ml. 180 µ l cell suspension was transferred to 96-well round-bottom plates (Falcon). GFP, RFP, and BFP were measured with an LSRFORTESSA system (BD Bioscience) kindly provided by the Core Facility Flow Cytometry of the German Cancer Research Center. The cells were trypsinized, and 500 µ l cell suspension was passed through a cell strainer into 5-ml round-bottom tubes (Falcon). Flow cytometry analysis was performed with either Guava InCyte or FlowJo (LLC).

Immunoblotting. Total protein was extracted with Cell Lysis Buffer (NEB) supplemented with protease inhibitor (Sigma). Cell pellets were lysed in Laemmli buffer (100 mM Tris-HCl, pH 6.8, 5% glycerol, 2% SDS, and 5% 2-mercaptoethanol). Protein concentration was determined with the Bradford method (Bio-Rad). Equal amounts of protein were separated with SDS–PAGE, transferred to PVDF membranes and immunoblotted. Clarity Western ECL Substrate Solution (Bio-Rad) was used for detection, and images were taken with AlphaView software (ProteinSimple). The following antibodies were used: anti-p53 (Cell Signaling, CST 2524, 1:1,000), HRP-conjugated anti-actin (Sigma, A3854, 1:10,000), anti-tRFP (Evrogen, AB233, 1:1,000), and HRP-conjugated goat anti-mouse (Jackson Immuno Research, 115-035-146, 1:5,000) or goat anti-rabbit (Jackson Immuno Research, 111-035-144, 1:5,000).

NATuRe MeTHoDS | www.nature.com/naturemethods

Brief CommuniCation NAtuRe MetHodS

Animal experiments. 8- to 10-week-old female C57BL/6 mice were purchased from Envigo (formerly Harlan). TRE-GFP knock-in mice were generated by electroporating 200,000 D34 ES cells with 5 µ g of the CTGM targeting vector and 2.5 µ g pCaggs-Flpe with a Lonza Primary Cell P3 Kit according to the manufacturer’s instructions. Positive clones were selected with hygromycin (150 µ g/ml). ES cells were injected into blastocysts to generate chimeric mice. The chimeras were crossed to C57/BL6 mice to obtain germline transmission of the knock-in. Mice were genotyped with primers ColA1-For (AATCATCCCAGGTGCACAGCATTGCGG) and ColA1-Rev (CTTTGAGGGCTCATGAACCTCCCAGG).

For HDTV, a sterile 0.9% NaCl solution/plasmid mix was prepared containing 20 µ g pMB21 DNA, 20 µ g DNA of pMB13 bearing the respective guides, and 5 µ g pT3-myc DNA when applicable, together with CMV-SB13 transposase (1:5 ratio). Mice were randomly assigned to experimental groups and injected with the 0.9% NaCl solution/plasmid mix into the lateral tail vein, with a total volume corresponding to 10% of body weight, in 5–7 s. If applicable, tumor development was monitored by palpation. Mice were euthanized in compliance with all relevant ethical regulations determined in the animal permit, inner organs were visually inspected, and livers were harvested and photographed. Liver and tumor samples were collected to obtain genomic DNA, and livers were incubated in 4% PFA for at least 24 h for further use. All animal experiments were approved by the regional board, Karlsruhe, Germany.

Immunofluorescence. Livers fixed in 4% paraformaldehyde were embedded in paraffin and sectioned. Slides containing 3-μ m-thick liver sections were deparaffinized and rehydrated with a descending alcohol series. For antigen retrieval, slides were cooked in sodium citrate buffer, pH 6.0, in a pressure cooker for 8 min. Slides were blocked with PBS containing 5% BSA and 0.05% Triton X-100 for 1 h before incubation with primary antibodies overnight at 4 °C. Slides were washed three times with PBS and incubated with secondary antibody for 1 h at room temperature in a dark humid chamber.

Slides were mounted in DAPI. The primary antibodies used were anti-tRFP (Evrogen, AB233, 1:500) and anti-GFP (Abcam, ab13970, 1:500). The secondary

antibodies used were Alexa Fluor 488 goat anti-chicken (Invitrogen, A-11039, 1:500) and Alexa Fluor 594 donkey anti-rabbit (Invitrogen, A-21207, 1:500).

Immunohistochemistry. Slides containing 3 μ m thick liver sections were deparaffinized and rehydrated with a descending alcohol series. For antigen retrieval, slides were treated with Dako Target Retrieval Solution (Agilent, S2367) for 30 min. Slides were washed with TBS for 10 min before incubation with the primary antibody (anti-rabbit CK19 (KRT19), Abcam, ab-133496; 1:100) overnight. Slides were washed twice with TBS, and staining was visualized with DCS SuperVision Red 2 AP + PermaRed according to the manufacturer’s instructions.

Derivation of primary liver tumor cell lines. Liver tumors were resected with sterile instruments, and 10–50 mg of tumor tissue was minced, incubated in a mix of 4 mg/ml collagenase IV and dispase in sterile, serum-free DMEM with gentle shaking at 37 °C for 30 min, washed in complete DMEM (10% FBS, 1× penicillin–streptomycin), and plated on collagen-coated plates. Primary cultures were passaged until they were visibly free of fibroblasts. Established cell lines were tested for mycoplasma contamination (GATC Biotech).

Statistical analysis. Unless otherwise stated, each replicate represents a biologically independent experiment, i.e., an independent cell transfection, independently transduced cell line, or independent animal. No statistical tests to determine significance were performed.

Reporting Summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availabilityThe data that support the findings of this study are available from the corresponding author upon reasonable request.

References 20. Labun, K., Montague, T. G., Gagnon, J. A., Thyme, S. B. & Valen, E.

Nucleic Acids Res. 44, W272–W276 (2016).

NATuRe MeTHoDS | www.nature.com/naturemethods

1

nature research | reporting summ

aryApril 2018

Corresponding author(s): Darjus Tschaharganeh

Reporting SummaryNature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see Authors & Referees and the Editorial Policy Checklist.

Statistical parametersWhen statistical analyses are reported, confirm that the following items are present in the relevant location (e.g. figure legend, table legend, main text, or Methods section).

n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement

An indication of whether measurements were taken from distinct samples or whether the same sample was measured repeatedly

The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section.

A description of all covariates tested

A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons

A full description of the statistics including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)

For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable.

For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings

For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes

Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated

Clearly defined error bars State explicitly what error bars represent (e.g. SD, SE, CI)

Our web collection on statistics for biologists may be useful.

Software and codePolicy information about availability of computer code

Data collection AlphaView software AlphaEaseFC Version 4.1 was used to image gels. StepOne software Version 2.3 was used for qPCR.

Data analysis R Studio Version 0.98.1103 was used to calculate mean and s.d. and plot data. Flow cytometry analysis was performed using either GuavaSoft InCyte Version 3.2 or FlowJo Version 10.3.

For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors/reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

2

nature research | reporting summ

aryApril 2018

DataPolicy information about availability of data

All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A list of figures that have associated raw data - A description of any restrictions on data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request. Plasmids are available upon request and we will deposit relevant plasmids to Addgene.

Field-specific reportingPlease select the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences

For a reference copy of the document with all sections, see nature.com/authors/policies/ReportingSummary-flat.pdf

Life sciences study designAll studies must disclose on these points even when the disclosure is negative.

Sample size No sample-size calculation was performed. In vitro experiments were performed in at least biological duplicate n=2 unless otherwise noted. In previous studies using related experiments we determined this sample size to be sufficient to ensure reproducibility. For in vivo studies, where variance is increased, we used 3-5 mice per experimental group. In previous studies using related experiments we determined this sample size to be sufficient to ensure reproducibility.

Data exclusions No data was excluded from analyses.

Replication Experiments were performed on at least two independent occasions. For in vitro studies, transduction- based experiment that make up the majority of this manuscript, this means at least two independent transductions to verify reproducibility. Methods and materials used in our experiments are described in the manuscript to facilitate reliable reproduction of the experimental findings.

Randomization For animal experiments allocation into experimental groups was random. For in vitro studies, different cell passages were used for each biological replicate.

Blinding Blinding was not relevant to the studies as samples with particular genetic constituents were needed for the experiments.

Reporting for specific materials, systems and methods

Materials & experimental systemsn/a Involved in the study

Unique biological materials

Antibodies

Eukaryotic cell lines

Palaeontology

Animals and other organisms

Human research participants

Methodsn/a Involved in the study

ChIP-seq

Flow cytometry

MRI-based neuroimaging

Unique biological materialsPolicy information about availability of materials

Obtaining unique materials Murine liver cancer cell lines derived from established tumors are available upon request.

3

nature research | reporting summ

aryApril 2018

AntibodiesAntibodies used tRFP-Ab (Evrogen, AB233, 1:500), GFP-Ab (Abcam, ab13970, 1:500), p53-Ab (Cell Signaling, CST 2524, 1:1000), HRP-conjugated

Actin-Ab (actin-HRP, Sigma, 1:10000), Alexa Fluor 488 goat α-chicken-Ab (Invitrogen,A-11039, 1:500) , Alexa Fluor 594-donkey α-rabbit-Ab (Invitrogen, A-21207, 1:200) , CK19 (KRT19; Abcam ab-133496; 1:100)

Validation CK19 specificity was evaluated by positivity of cholangiocytes in murine livers. p53 specificitiy was evaluated by using p53-shRNA cells as well as testing p53 wt and p53 KO cells side-by-side. tRFP specificity was evalauted using tRFP/tBFP negative samples-no staining was observed. GFP specificity was evalauted using GFP negative samples-no staining was observed. Specificity of the Alexa488 and Alexa594 secondary antibodies was confirmed by performing a 'secondary only' staining control - no staining was observed.

Eukaryotic cell linesPolicy information about cell lines

Cell line source(s) HEK293T from AATC; Murine liver cancer cell lines derived from established tumors.

Authentication HEK293T cells were authenticated by the supplier. Murine liver cancer cell lines derived from established tumors were tested for marker expression, e.g p53 KO.

Mycoplasma contamination Cell lines were tested for mycoplasma contamination (GATC Biotech) and were found to be negytive.

Commonly misidentified lines(See ICLAC register)

No commonly misidentified cell lines were used.

Animals and other organismsPolicy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research

Laboratory animals For hydrodynamic liver transfections, 8-10 week old female C57Bl/6 animals were purchased from Envigo and used in this study. Animals were sacrificed 2-10 weeks following injection as indicated in respective Figures.

Wild animals No wild animals were used in this study.

Field-collected samples No field-collected samples were used in this study.

Flow CytometryPlots

Confirm that:

The axis labels state the marker and fluorochrome used (e.g. CD4-FITC).

The axis scales are clearly visible. Include numbers along axes only for bottom left plot of group (a 'group' is an analysis of identical markers).

All plots are contour plots with outliers or pseudocolor plots.

A numerical value for number of cells or percentage (with statistics) is provided.

Methodology

Sample preparation Cells were washed with PBS, trypsinized, resuspended in DMEM and passed through a cell-strainer.

Instrument Guava® easyCyte benchtop flow cytometer (Merck). LSRFORTESSA system (BD Bioscience).

Software Analysis was performed using either GuavaSoft InCyte Version 3.2 or FlowJo Version 10.3.

Cell population abundance n/A

Gating strategy Gating was performed based on positive and negative controls (see Supplementary Fig. 14b)

Tick this box to confirm that a figure exemplifying the gating strategy is provided in the Supplementary Information.