An inducible CRISPR-ON system for controllable gene activation in human pluripotent stem cells
- First Online:
- 2.4k Downloads
Human pluripotent stem cells (hPSCs) are an important system to study early human development, model human diseases, and develop cell replacement therapies. However, genetic manipulation of hPSCs is challenging and a method to simultaneously activate multiple genomic sites in a controllable manner is sorely needed. Here, we constructed a CRISPR-ON system to efficiently upregulate endogenous genes in hPSCs. A doxycycline (Dox) inducible dCas9-VP64-p65-Rta (dCas9-VPR) transcription activator and a reverse Tet transactivator (rtTA) expression cassette were knocked into the two alleles of the AAVS1 locus to generate an iVPR hESC line. We showed that the dCas9-VPR level could be precisely and reversibly controlled by the addition and withdrawal of Dox. Upon transfection of multiplexed gRNA plasmid targeting the NANOG promoter and Dox induction, we were able to control NANOG gene expression from its endogenous locus. Interestingly, an elevated NANOG level promoted naïve pluripotent gene expression, enhanced cell survival and clonogenicity, and enabled hESCs to integrate with the inner cell mass (ICM) of mouse blastocysts in vitro. Thus, iVPR cells provide a convenient platform for gene function studies as well as high-throughput screens in hPSCs.
KeywordsCRISPR transcription activation human pluripotent stem cells NANOG pluripotency
Human pluripotent stem cells (hPSCs), including human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs), are capable of self-renewal indefinitely and have the potential to differentiate into all cell types in the human body. Therefore this system offers a useful platform to study early human embryogenesis and a potential cell source for regenerative medicine. Moreover, functional cells derived from hESCs can be used to model human diseases in the context of drug toxicity tests and new drug development. These applications rely on methods to precisely control gene expression. However, because of difficulties in culture and transfection, targeted regulation of gene expression in hPSCs remains a technically challenging task. A method for efficient, rapid, and controllable gene activation is sorely needed.
Recently, the clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 system emerged as a powerful and versatile tool for genome editing (Wiedenheft et al., 2012). CRISPR was initially discovered as the adaptive immune system of bacteria and archaea (Wiedenheft et al., 2012). In response to viral and plasmid infection, bacteria and archaea could cut and degrade the foreign DNA recognized by a matching spacer RNA with the help of the Cas9 enzyme (Wiedenheft et al., 2012). CRISPR was rapidly transformed to a genome editing tool, and it has been shown to work in a wide range of systems, from plants to human cells, since the Cas9 nuclease can be directed easily to virtually anywhere in the genome using a short guide RNA and cutting the target DNA (Hsu et al., 2014). In pluripotent stem cells, the CRISPR system has been used to perform highly efficient gene knock-out and knock-in studies (Hsu et al., 2014). In addition to genome editing, a nuclease inactivated Cas9 (dCas9) was developed (Gilbert et al., 2014). By fusing dCas9 with transcription activators and repressors, such as VP64, and KRAB (Balboa et al., 2015; Gilbert et al., 2014; Mandegar et al., 2016; Genga et al., 2016), or with epigenetic modifiers, such as the catalytic domain of acetyltransferase p300 (Hilton et al., 2015) and Tet (ten eleven translocation) dioxygenase (Xu et al., 2016), one can use the CRISPR system to activate or inhibit gene expression or modify the histone and DNA methylation status at the desired locus.
Because of its potential applications in regenerative medicine, random insertion of foreign DNA into the genome of hPSCs should be avoided, since this may cause harmful mutations. The Adeno-Associated Virus Integration Site 1 (AAVS1) locus resides in the first intron of the PPP1R12C gene and has been used as a safe harbor for transgene integration (Smith et al., 2008; Hockemeyer et al., 2009; Lombardo et al., 2011; Qian et al., 2014; Zhu et al., 2014; Genga et al., 2016). Here we generated an iVPR hESC line by knocking-in the inducible dCas9-VPR system into the two alleles of the AAVS1 locus. Detailed characterization of the iVPR hESC demonstrated that dCas9-VPR protein could be induced by Dox within 12 h and disappear after Dox withdrawal. An inducible NANOG overexpression line (iNANOG) was established based on the iVPR system. We found a significant increase in NANOG protein after Dox induction. INANOG cells upregulated naïve pluripotency genes and were able to grow for a significant length of time in a naïve state medium containing ERK and GSK3 inhibitors and human LIF. The iVPR system can be a valuable system to control gene expression from endogenous loci and serve as platform for genome wide screens to identify new genes that can regulate stem cell self-renewal and differentiation.
DCas9-VPR mediated robust ectopic and endogenous gene activation in human cell lines
We next tested the dCas9-VPR function in hESCs. DCas9-VPR, gTetO, and TRE-BFP plasmids were co-transfected into H9 hESCs. In another group, rtTA and TRE-BFP plasmids were co-transfected. FACS analysis showed that nearly 17% of cells in the dCas9-VPR group turned on BFP, while 24.7% of cells in the rtTA group were BFP positive after Dox induction, and only 0.6% of cells exhibited BFP fluorescence without Dox (Fig. 1E). Interestingly, the dCas9-VPR group showed the strongest mean fluorescence intensity (Fig. 1F). This is consistent with our result based on 293FT cells and proves that dCas9-VPR is a robust transcription activator, even compared with rtTA. We also tested the activation effect of dCas9-VPR in mouse embryonic stem cells (mESCs) and mouse embryonic fibroblasts (MEFs) and obtained similar results (Fig. S1A and S1B).
Next, we tested whether the dCas9-VPR system could simultaneously activate multiple genes in human cells, we designed 2 different gRNAs per gene promoter for HOXA10, SNAIL1, MESP1, GATA5 and HOXA9. First we tested the activation efficiency of these gRNAs towards their target genes when transfected separately in 293FT cells (Fig. 2D). Q-PCR analysis showed all of the five pairs of gRNAs can activate their target gene upon co-transfection with dCas9-VPR (Fig. 2D). We next pooled gRNA pairs of two genes (2× gRNAs: MESP1, GATA5), three genes (3× gRNAs: HOXA10, SNAIL1, HOXA9) or five genes (5× gRNAs: HOXA10, SNAIL1, MESP1, GATA5 and HOXA9) to test the co-activation efficiency. Upon co-transfection with dCas9-VPR, different combination of gRNAs upregulated their target genes together (Fig. 2E), indicating that dCas9-VPR system could be a useful tool for multiplexed endogenous gene activation.
Generation of an inducible idCas9-VPR hESC knock-in line
Upregulation of NANOG by dCas9-VPR promoted naïve state of pluripotency
Upregulation of NANOG enabled hESCs to integrate with mouse ICM in vitro
In this study, we generated an inducible CRISPR-ON hESC line by targeting the AAVS1 locus. Based on both our results and those of Chavez et al. (2015), dCas9-VPR appeared to be a stronger activator than VP64 to induce gene expression from both ectopic and endogenous promoters. It even led to a higher level of reporter gene activation compared with Tet-ON rtTA, where VP64 was fused with Tet protein directly bound to the TRE elements. This is likely due to the combined effects of VP64, NF-κB transcactivating subunit p65, and the viral transcription factor Rta, which together can recruit a multitude of endogenous factors to achieve dramatically enhanced transcriptional activation. Other dCas9 based transcription activators have been generated. For example, Balboa et al. found increased activation ability with more VP16 fusing together. Using the longest version of dCas9-VP192 combined with inducible systems, they sucessfully facilitated human cell reprogramming and differentiation (Balboa et al., 2015). Konermann et al. engineered a structure-guided CRISPR synergistic activation mediator system (SAM), where they engineered gRNA2.0 by replacing the tetraloop and stem loop 2 of the original gRNA with a minimal hairpin aptamer that specifically binds to MS2 bacteriophage coat proteins (Konermann et al., 2014). By co-expression of dCas9-VP64, gRNA2.0, and MS2 fused with p65 and the activation domain of the human heat-shock factor 1 (HSF1), highly effective gene activation can be achieved (Konermann et al., 2014). Tanenbaum et al. constructed a SunTag system: dCas9 was joined with 10 copies of GCN4 peptide (SunTag), while VP64 was fused with scFv-GCN4 (the single-chain variable fragment (scFv) antibody of GCN4) (Tanenbaum et al., 2014). When co-expressed in the cell, SunTag was bound by scFv-GCN4, and multiple copies of VP64 resulted activation of the target gene (Tanenbaum et al., 2014). Compared with the systems discussed above, which required introducing tandem repeat large cassette or the co-expression of two components in addition to the gRNA, dCas9-VPR is a simple and effective option.
In our study, we chose to insert the iVPR system into the AAVS1 locus, since it has been used as a ‘safe habor’ for transgene insertion in human stem cell systems (Dekelver et al., 2010). For example, Genga et al. constructed a GFP labled H1 hESC line by knocking-in a CAG-GFP into the AAVS1 locus. Besides, an inducible dCas9-KRAB gene inhibition system was also introduced into the GFP-H1 cells. By infecting sgRNA targeting the exdogenous CAG promoter, they successfully realized CRISPR based inhibition of exdogenous gene in hESCs (Genga et al., 2016). González et al. inserted the Dox inducible Cas9 system into the 2 alleles of the AAVS1 locus of HUES8 hESCs (González et al., 2014; Zhu et al., 2014; Zhu et al., 2015). The resulting iCRISPR hESC line enabled selection-free gene knock-out and the generation of lineage-specific knock-in reporters. This demonstrated that when Cas9 was expressed in a controllable manner from a suitable locus, the resulting cell line can be a powerful platform for genome editing in normally hard to transfect human stem cells (Zhu et al., 2015). Similarly, using the iVPR line, we found that the efficiency to generate an iNANOG line was much improved. Upon Dox addition and withdrawal, NANOG transcripts and proteins could be up- and down-regulated in a highly repeatable manner, which greatly facilitated downstream experiments. Recently, Ordovás et al. reported AAVS1-locus mediated transgene inhibition in hESCs, and that inhibition may due to different cassettes inserted into the locus (Ordovás et al., 2015). We tested the iVPR expression in both undifferentiated hESCs and after induction of mesoderm differentiation. The level of dCas9-VPR transcripts was even higher upon Dox treatment after mesoderm induction (Fig. 4E). The iVPR and iNANOG cells have been maintained for more than 6 months, and we did not observe any reduction in the level of dCas9-VPR or NANOG induced by Dox. Thus, results of us and other groups suggested that, in most cases, AAVS1 locus integration is a reliable approach to generate transgenic hPSCs.
NANOG is a master transcription factor for pluripotency in both human and mouse ESCs (Mitsui et al., 2003; Boyer et al., 2005; Chambers et al., 2007). During somatic cell reprogramming to pluripotent stem cells, ectopic expression of NANOG helped to speed up reprogramming and restrict partially reprogrammed cells to the ground state (Hanna et al., 2009; Silva et al., 2009). Different from mESCs, conventional cultured hESCs are in a primed state, similar to the epiblast stem cells in mice (Brons et al., 2007; Tesar et al., 2007). Recently, multiple groups reported methods to obtain naïve state hPSCs that resemble ground-state mESCs (Gafni et al., 2013; Duggal et al., 2015; Takashima et al., 2014; Theunissen et al., 2014). Takashima et al. showed that ectopic expression of NANOG and KLF2 could reset the self-renewal requirements of hPSCs so that they can be grown in a medium containing ERK1/2 inhibitor PD0325901 and GSK3 inhibitor CHIR99021, and adopt a domed-shaped morphology similar to that of mESCs (Takashima et al., 2014). Here we increased the expression of endogenous NANOG by targeting a strong transcription activator, dCas9-VPR, to its promoter. As expected, we observed upregulation of naïve state genes such as GDF3, PRDM14, and LEFTYB and downregulation of early differentiation gene AFP. Interestingly, these iNANOG cells showed a significantly improved survival ability and clonogenicity when cultured in the primed state, and they could grow in 2i plus LIF conditions for more than nine passages. The improved survival and self-renewal of iNANOG cells was not due to the effect of Dox treatment as described by Chang et al. (2014), because Dox treated iNANOG cells showed significantly higher clonogenicity over Dox treated iVPR cells (Fig. 5H and 5I). The enhanced survival ability seemed to have a significant influence on whether hPSCs can integrate with the ICM of mouse blastocysts during in vitro culture. We found that even when iNANOG cells were in the primed state, after injection into mouse blastocysts, more cells remained inside the blastocysts and some of the cells were able to integrate with mouse ICM cells (Fig. 6F and 6G). Culturing iNANOG cells in 2iL/FK naïve state medium (Duggal et al., 2015) further improved the ICM integration rate (Fig. 6F and 6G). INANOG cells displayed highly dynamic interactions with mouse ICM cells, as observed in time-lapse movies (Supplementary movie S1). They migrated with mouse ICM cells as blastocysts hatched from zona pellucida. However, despite enhanced survival ability of iNANOG cells, many injected cells died over time. After 24 h, more than 30% of injected blastocysts lost all iNANOG cells and more than 60% of blastocysts lost the injected hESCs if NANOG was not overexpressed (Fig. 6F). This was partially caused by poor survival of hESCs in the IVC-1 and -2 media designed to culture peri-implantation mouse and human embryos (Bedzhov and Zernicka-Goetz, 2014; Deglincerti et al., 2016; Shahbazi et al., 2016) (Fig. S4D). Thus, to achieve better naïve hPSC and mouse ICM integration, a culture medium suitable for both mouse blastocysts and hPSCs may be needed. The effect of NANOG overexpression on cell survival and self-renewal is also in accordance with the observation that chromosome 12, where the NANOG gene is located, is the most frequently gained chromosome in culture adapted hPSCs (Baker et al., 2007) and during hiPSC generation (Taapken et al., 2011). Moreover, NANOG was reported to be upregulated by a number of factors such as STAT3, Hedgehog signaling, hypoxia, etc., in human cancers, and repression or ablation of NANOG inhibited tumor initiation (Gong et al., 2015). Thus, iNANOG hESCs, where the endogenous NANOG can be activated by dCas9-VPR in a controllable manner, may also be a good system to study the process of hPSC adaptation and cancerous transformation.
In summary, the iVPR hESC line generated and characterized in this study offered a convenient, stable, and highly controllable platform for gene activation studies. It can also be used to investigate the function of regulatory elements in the genome such as super enhancers as well as for genome wide screens using established human gRNA libraries.
MATERIALS AND METHODS
H9 hESCs (WiCell Institute) were maintained on inactivated mouse embryonic fibroblast (MEF) cells in standard hESC medium consisting of KO-DMEM (Invitrogen) supplemented with 1× Nonessential Amino Acids (NEAA) (Invitrogen), 0.1 mmol/L 2-mercaptoethanol (Sigma-Aldrich), 1 mmol/L GlutaMAX (Invitrogen), 20% Knock-out serum-replacement (KOSR) (Invitrogen) and 4 ng/mL bFGF (Peprotech). Cells were cultured at 37°C in a humidified atmosphere with 5% CO2 in air. They were passaged with 1 mg/mL collagenase IV (Invitrogen) and seeded onto MEFs. For feeder-free culture, hESCs were grown for more than three passages on Matrigel (growth factor reduced, BD Biosciences) in the absence of feeders in E8 medium (Invitrogen).
DCas9-VPR was constructed by fusing the nuclease deficient Cas9 (dCas9) with transcription activator VP64, p65, and Rta in tandem as described by Chavez et al. (Maeder et al., 2013). For constitutive expression, dCas9-VPR was placed behind a CAG promoter in a PiggyBac vector also containing a PGK promoter driving a hygromycin resistance gene. For inducible expression from the AAVS1 locus, dCas9-VPR was placed behind a TRE promoter in the AAVS1 homologous recombineering donor plasmid, as shown in Fig. S2A. DCas9-VP64 was constructed by fusing dCas9 with VP64. Tet-On system was obtained from Clontech (http://www.clontech.com). PiggyBac plasmids were generous gift from the Sanger institute, Cambridge, UK (http://www.sanger.ac.uk). The multiple NANOG gRNA expression plasmid was constructed by SynGene (http://syngen.tech) as depicted in Fig. 2A.
Naïve state culture condition for hPSCs
For naïve state conversion, cells cultured in standard hESC medium on MEFs were dissociated to single cells using 0.05% trypsin/EDTA solution (Invitrogen), replated on MEFs, and cultured overnight in standard hESC medium supplement with 10 μmol/L Rho Kinase (ROCK)-inhibitor Y-27632 (Calbiochem). The next day, the standard medium was changed to the 2iL or 2iL/FK (for injection) medium, which consisted of KO-DMEM (Invitrogen), 20% KOSR, 1×NEAA, 0.1 mmol/L 2-mercaptoethanol, 1 mmol/L GlutaMAX, 12 ng/mL bFGF, 10 ng/mL human recombinant LIF (Peprotech), 1 μmol/L ERK1/2 inhibitor PD0325901 (Peprotech), 3 μmol/L GSK3 inhibitor CHIR99021 (Peprotech), 10 μmol/L Forskolin (Peprotech), and 50 μg/mL ascorbic acid (Sigma). HESCs changed to a dome-shaped morphology within 4–6 days after culturing in the 2iL or 2iL/FK medium and were passaged every 4 days as single cells using 0.05% trypsin/EDTA.
Cardiac mesoderm differentiation from hESCs
For cardiac mesoderm differentiation, hESCs maintained on Matrigel (growth factor reduced, BD Biosciences) in E8 were dissociated into single cells with Accutase (Invitrogen), then seeded onto Matrigel-coated tissue culture dishes at a density of 5 × 104 cells/cm2 and cultured in E8 for 3 days. Then the medium was switched to the RPMI1640 medium supplemented with Albumin, Ascorbic acid, transferrin, selenite, 5 ng/mL BMP4 (R&D Systems), and CHIR99021 to induce cardiac mesoderm formation.
Total RNA was extracted with TRIZOL (Invitrogen). 1 μg RNA of each sample was used for reverse transcription with Superscript III (Invitrogen). Q-PCR reactions were performed using GoTaq qPCR Master Mix (Promega) in a CFX96 Real-Time System (Bio-Rad). The relative expression level of each gene was normalized against the Ct (Critical Threshold) value of the house-keeping gene GAPDH using the Bio-Rad CFX Manager program. Primer sequences are listed in table S2.
Antibodies, immunostaining, Western blot, and FACS analysis
For immunostaining, cells were fixed in 4% paraformaldehyde (PFA) in PBS, permeabilized in 0.5% Triton X-100 (Sigma), blocked in 5% normal goat serum (Origene) and incubated with primary antibodies against NANOG (1:200), SSEA3 (1:200) in 4°C overnight and detected by DyLight 488- or 549-conjugated secondary antibodies (Thermo). Nuclei were stained with DAPI (Sigma). A Nikon Ti-U fluorescence microscope was used for image acquisition. For western blot, cells were lysed in a RIPA buffer (Applygen, http://applygen.com.cn) with Protease Inhibitor Cocktail (Roche). Total proteins were separated on a 12% SDS/PAGE gel, transferred to nitrocellulose membrane (Whatman). The membrane was blocked with 5% non-fat dry milk in TBST and then incubated with primary antibodies against Cas9 (Genetex, 1:1000), GAPDH (CWBio, 1:1000), OCT4 (Santa Cruz, 1:1000) and NANOG (Cell Signaling Technology, 1:1000). After washing, the membrane was incubated with anti-mouse or anti-rabbit peroxidase-conjugated secondary antibodies (ZSGB-Bio http://www.zsbio.com/). Bands recognized by antibodies were revealed by ECL reagent (Pierce). For FACS analysis, cells were first dissociated with 0.05% Trypsin in 0.2% EDTA and PBS. FACS was performed on a Fortessa flow cytometer (Becton Dickinson).
Mouse blastocyst injection and in vitro culture
The animal facility of Tsinghua University has been accredited by the AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care International) and all animal protocols used in this study were approved by the IACUC (Institutional Animal Care and Use Committee) of Laboratory Animal Research Center of Tsinghua University. Mouse morula were collected from ICR females 2.5 days post-coitus and cultured in KSOM medium (95 mmol/L NaCl, 2.5 mmol/L KCl, 0.35 mmol/L KH2PO4, 0.2 mmol/L MgSO4·7H2O, 0.2 mmol/L glucose, 10 mmol/L sodium lactate, 25 mmol/L NaHCO3, 0.2 mmol/L sodium pyruvate, 1.71 mmol/L CaCl·2H2O, 0.01 mmol/L EDTA, 1 mmol/L L-glutamine, 0.1 mmol/L EAA, 0.1 mmol/L NEAA, 4 mg/mL BSA) at 37°C, 5% CO2 for 24 h to get blastocysts (Hogan et al., 1986). HESCs were briefly treated with Accutase for single cell and injected (~10–15 cells for each embryo) on a Nikon microscope fitted with piezo-driven Eppendorf NK2 micromanipulator, CellTram air and CellTram Vario. After injection, embryos containing hESCs were cultured in medium supplemented with naïve culture medium:KOSM (1:1) (Chen et al., 2015) in 37°C, 5% CO2 incubator. After injected embryos reformed blastocoel, the chimera embryos were live cell imaged using Leica microscope fitted with a live cell imaging system and fixed after 24–36 h post-injection for staining and confocal imaging. For embryo immunostaining, zona pellucida-free injected embryos were fixed with 3.5% paraformaldehyde, permeabilized in 0.5% Triton X-100 (Sigma) and blocked with 5% BSA and then incubated with primary antibodies against CDX2 (BioGenex), β-Catenin (1:50, Abcam) and detected by DyLight 549- or 633- conjugated secondary antibodies(Thermo). Nuclei were stained with DAPI (Sigma). A Nikon-A1 fluorescence microscope was used for image acquisition.
Data are presented as mean ± standard error of the mean (SEM). Statistical significance was determined by Student’s t-test (two-tail) for two groups or one-way Analysis of Variance (ANOVA) for multiple groups using Graphpad software. P < 0.05 was considered significant.
This work was supported by the National Basic Research Program (973 Program) (No. 2012CB966701), the National Natural Science Foundation of China (Grant No. 31171381 to J.N.), and core facilities of the Tsinghua-Peking University Center for Life Sciences. TNLIST Interdisciplinary research foundation grant 042003171 (to Z.X. and N.J.). We thank Dr. Danwei Huangfu for the AAVS1 homologous recombineering donor plasmids and Dr. Xiaohua Shen lab for assistance in Southern blot.
J.G.: concept and design, collection and/or assembly of data, data analysis and interpretation, manuscript writing; D.M., R.H., M.Y., J.M.: collection and/or assembly of data; K.K. provided essential reagents, technical and scientific advice to the experiments and manuscript; Z.X.: concept and design; J.N.: concept and design, manuscript writing, and final approval of the manuscript.
COMPLIANCE WITH ETHICAL GUIDELINES
Jianying Guo, Dacheng Ma, Rujin Huang, Jia Ming, Min Ye, Kehkooi Kee, Zhen Xie, and Jie Na declare no conflict of interests.
This article does not contain any studies with human subjects performed by the any of the authors. All institutional and national guidelines for the care and use of laboratory animals were followed.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.