ARTICLE DOI: 10.1038/s41467-018-05067-x OPEN Human pluripotent reprogramming with CRISPR activators Jere Weltner1, Diego Balboa 1, Shintaro Katayama2, Maxim Bespalov1, Kaarel Krjutškov2,3, Eeva-Mari Jouhilahti1, Ras Trokovic1, Juha Kere 1,2,4,5 & Timo Otonkoski 1,6 CRISPR-Cas9-based gene activation (CRISPRa) is an attractive tool for cellular reprogram- ming applications due to its high multiplexing capacity and direct targeting of endogenous 1234567890():,; loci. Here we present the reprogramming of primary human skin fibroblasts into induced pluripotent stem cells (iPSCs) using CRISPRa, targeting endogenous OCT4, SOX2, KLF4, MYC, and LIN28A promoters. The low basal reprogramming efficiency can be improved by an order of magnitude by additionally targeting a conserved Alu-motif enriched near genes involved in embryo genome activation (EEA-motif). This effect is mediated in part by more efficient activation of NANOG and REX1. These data demonstrate that human somatic cells can be reprogrammed into iPSCs using only CRISPRa. Furthermore, the results unravel the invol- vement of EEA-motif-associated mechanisms in cellular reprogramming. 1 Research Programs Unit, Molecular Neurology and Biomedicum Stem Cell Centre, Faculty of Medicine, University of Helsinki, Helsinki 00014, Finland. 2 Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge 141 83, Sweden. 3 Competence Centre on Health Technologies, Tartu 50410, Estonia. 4 School of Basic and Medical Biosciences, Guy’s Hospital, King’s College London, London SE1 9RT, UK. 5 Folkhälsan Institute of Genetics, Helsinki 00290, Finland. 6 Children’s Hospital, Helsinki University Central Hospital, University of Helsinki, Helsinki 00290, Finland. These authors contributed equally: Juha Kere, Timo Otonkoski. Correspondence and requests for materials should be addressed to J.W. (email: jere.weltner@helsinki.fi) or to J.K. (email: [email protected]) or to T.O. (email: timo.otonkoski@helsinki.fi) NATURE COMMUNICATIONS | (2018) 9:2643 | DOI: 10.1038/s41467-018-05067-x | www.nature.com/naturecommunications 1 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05067-x RISPRa system relies on sequence specific recruitment of a reprogramming human cells, including primary adult human catalytically inactivated version of Cas9 protein (dCas9) to skin fibroblasts, into induced pluripotent stem cells by CRISPRa. C fi genomic sequences de ned by short guide RNA (gRNA) This opens up important possibilities for the development of molecules1–3. The fact that dCas9 effectors can be used to control more extensive CRISPRa reprogramming approaches for human transcription of targeted endogenous loci makes it useful for cells. Efficiency of the method depends on the targeting of the mediating cellular reprogramming, which requires silencing and EEA-motif, which results in improved activation of a subset of activation of endogenous gene sets for proper cell type conver- endogenous genes that work as reprogramming factors, including sion. CRISPRa may therefore be beneficial in overcoming NANOG and ZFP42 (REX1). These results also exemplify the reprogramming barriers that limit reprogramming efficiency and potential in targeting cell type enriched regulatory elements for contribute to the emergence of partially reprogrammed stable cell controlling cell fate. populations, often associated with inadequate endogenous gene activation or silencing4–6. Previously, dCas9 effectors have been Results used to mediate differentiation, transdifferentiation, and repro- gramming of various mouse and human cell types, but complete CRISPRa-mediated reprogramming of NSCs and EEA target- ing. We began human cell reprogramming with CRISPRa using a pluripotent reprogramming of human cells using only CRISPRa fi has not yet been reported7–15. simpli ed reprogramming scheme. CRISPRa-mediated POU5F1 In addition to gene activation, dCas9 effector mediated DNA (OCT4) activation has been used to replace transgenic OCT4 in human fibroblast reprogramming, while the transgenic expres- targeting can be used to decipher the functions of genomic reg- fi ulatory elements16–18. Combining reprogramming factor pro- sion of only OCT4 has been shown to be suf cient for the reprogramming of neuroepithelial stem cells (NSCs) into moter targeting gRNAs with targeting of other regulatory 12,22 elements has high potential in mediating comprehensive resetting iPSCs . We therefore combined CRISPRa-mediated OCT4 of gene regulatory networks. A conserved Alu-motif was recently activation and NSC reprogramming as an initial model using fi trimethoprim (TMP) stabilized SpdCas9VP192 fused with P65- reported to be enriched in the promoter areas of the rst genes 23 expressed during human embryo genome activation (EGA)19. HSF1 activator domain (DDdCas9VPH) under doxycycline This sequence is thus likely to be involved in the control of early (DOX) inducible promoter (Fig. 1a, b). Expression of embryonic transcriptional networks. As human embryos can DDdCas9VPH and OCT4 targeting guides in iPSC-derived NSCs 20 resulted in the emergence of pluripotent cells in a DOX and TMP reprogram somatic cell nuclei , we hypothesized that targeting – this EGA-enriched Alu-motif (EEA-motif) could enhance dependent manner (Fig. 1c e). These cells could be expanded into reprogramming of somatic cells to pluripotency. stable dCas9 independent cell lines (Fig. 1c and Supplementary Fig. 1). This demonstrated that CRISPRa mediated activation of Development of reprogramming approaches for faithful reca- fi pitulation of cellular phenotypes is an important task, considering endogenous OCT4 alone was suf cient to reprogram NSCs to the increasing pace with which reprogrammed cells are moving iPSCs. toward clinical trials21. Here we describe a method for To determine if EEA-motif targeting could improve CRISPRa reprogramming of NSCs, we designed a set of five 14 nt gRNAs a c OCT4 SOX2 TRA-1-60 NANOG TRA-1-81 OCT4 dCas9VPH dCas9 VP192 P65 HSF1 b dCas9VPH + VIM TUBB3 α-SMA AFP SOX17 iPSC NSC OCT4 activation iPSC d EGA-enriched Alu-motif targeting gRNAs e CRISPRa NSC reprogramming with dCas9VPH 4000 3500 EEA-g10 3000 EEA-g7 2 2500 1 Bits 2000 0 1 3 5 6 7 8 9 2 4 1500 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 EEA-g3 1000 EEA-g1 500 EEA-g2 Colonies per 100,000 cells 0 no DOX no OCT4 gRNA OCT4 gRNA + EEA gRNA TMP EEA gRNA + DOX + TMP Fig. 1 CRISPRa-mediated reprogramming of NSCs and EEA-motif targeting. a Schematic representation of dCas9VPH structure. b Schematic representation of NSC reprogramming into iPSCs with dCas9VPH mediated OCT4 activation. c Immunocytochemical detection of pluripotency markers in NCS-derived iPSCs (top row) and tri-lineage differentiation in plated embryoid bodies (bottom row). Nuclei stained blue. Scale bar = 200 µm. d Targeting of EGA enriched Alu-motif with SpdCas9 gRNAs. e Quantification of iPSC-like alkaline phosphatase positive colonies induced from NSCs. n = 6 independent inductions (P = 0.053, OCT4 targeting with EEA-gRNAs vs. without EEA-gRNAs). Data presented as mean ± s.e.m., two-tailed Student’s t-test 2 NATURE COMMUNICATIONS | (2018) 9:2643 | DOI: 10.1038/s41467-018-05067-x | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05067-x ARTICLE a OCT4 4 53 2 1 SOX2 1 23 45 KLF4 5 4 321 MYC 54 3 21 LIN28A 5 4 32 1 NANOG 1 23 45 GENE –600 bp –400 bp –200 bp TSS b d dCas9VPH HEK293 gene activation with dCas9VPH + OMKSL gRNA plasmid 100 *** * 10 OCT4 4 53 2 1 Pool 1–5 *** 1 *** 0.1 ** SOX2 1 23 4 5 Pool 1–5 0.01 0.001 KLF4 5 43 2 1 Pool 1–5 mRNA expression relative to hESC 0.0001 TdT ctrl TdT ctrl TdT ctrl TdT ctrl TdT ctrl MYC 5 43 2 1 Pool 1–5 OMKSL OMKSL OMKSL OMKSL OMKSL OCT4 SOX2 KLF4 LIN28A MYC Fibroblast gene activation with dCas9VPH + OMKSL gRNA plasmid LIN28A 5 43 2 1 Pool 1–5 100 10 1 NANOG 1 23 4 5 Pool 1–5 ** ** 0.1 c MYC.1 OCT4.4 SOX2.3 LIN28.4 0.01 OCT4.1 KLF4.3 OCT4.5 LIN28.1 LIN28.3 SOX2.4 0.001 mRNA expression relative to hESC 0.0001 TdT ctrl TdT ctrl TdT ctrl TdT ctrl TdT ctrl OMKSL gRNA plasmid OMKSL OMKSL OMKSL OMKSL OMKSL OCT4 SOX2 KLF4 LIN28A MYC Fig. 2 Optimization of dCas9 activator and gRNA targeting in HEK293 for reprogramming factor activation. a Locations of promoter targeting gRNAs for reprogramming factors (OCT4, SOX2, KLF4, C-MYC, LIN28A, and NANOG) in relation to transcription start site. b Immunocytochemical staining of reprogramming factors after single gRNA activation and pooled mixture of five guides in HEK293 with dCas9VPH. Pictures are in similar order to guides in Fig. 2a. Best performing guides used for plasmid cloning are marked with dotted lines. Scale bar = 400 µm. c Schematic representation of concatenated reprogramming factor gRNA plasmid construction. d Reprogramming factor activation by qRT-PCR, in HEK293 cells 3 days after transfection and HFFs 4 days after electroporation, using transiently expressed dCas9VPH effector. n = 3, data are from three independently treated samples. Data presented as mean ± s.e.m., two tailed Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001 targeting the 36 bp EEA consensus sequence (Fig. 1d). Addition rest of the factors. This suggested that additional guides for KLF4, of the EEA-motif gRNAs in the reprogramming mixture LIN28A, and MYC targeting would be required for efficient demonstrated a trend in increasing the number of alkaline activation of these genes. phosphatase (AP) positive colonies (P = 0.053, Student’s t-test) (Fig. 1e). This suggested that EEA-motif targeting could be useful Fibroblast reprogramming with CRISPRa. Electroporation of for improving CRISPRa reprogramming efficiency. primary skin fibroblasts with episomally replicating dCas9VPH plasmid, containing TP53 targeting shRNA, EEA-motif targeting Pluripotency factor activation with CRISPRa. To devise a gRNA plasmid, reprogramming factor targeting gRNA plasmid reprogramming system for fibroblasts based solely on CRISPRa, (OMKSL), and an additional KLF4 and MYC targeting gRNA we optimized the promoter targeting of single gRNAs to the plasmid (KM) resulted in the emergence of iPSC-like colonies canonical reprogramming factors OCT4, MYC, KLF4, SOX2, (Fig.