The CRISPR-Cas9 system is a powerful tool to edit eukaryotic genomes that has recently been adapted for functional screens. Several of its applications—including the disruption of genes using Cas9-nickase and the generation of large deletions—require co-expression of two distinct guide RNAs (gRNAs). However, the lack of experimental approaches to generate pools of paired gRNA vectors prevents these applications from being scalable. Here we report a simple, inexpensive, one-step method that allows for the rapid and efficient cloning of gRNA pairs into expression vectors. We show that this method can be used to generate pooled libraries and is therefore suitable for in vivo and in vitro functional screens.
The CRISPR-Cas9 system for genome editing is a powerful tool for functional screens in vitro and in vivo1,2. Co-expression of the bacterial Cas9 endonuclease and a short guide RNA molecule (gRNA) is sufficient to generate double-stranded DNA breaks in eukaryotic cells. These cuts occur at sites that have a short (∼20 nucleotides) homology to the 5’ end of the gRNA and are followed by an NGG sequence—known as the protospacer-adjacent motif1,3.
In eukaryotes, double-stranded DNA breaks are primarily repaired through the error-prone non-homologous end-joining pathway4, often leading to the generation of small indels at the target site. Thus, the CRISPR-Cas9 system provides a simple way of disrupting the frame of protein-coding genes to produce loss of function alleles2,5,6,7,8,9,10. Gene disruption using the CRIPSR-Cas9 system has proven highly efficient and has been used for biallelic targeting of multiple genes simultaneously6,11,12,13. Because target recognition by Cas9 requires only a short homology between the gRNA and the genomic locus, the CRISPR-Cas9 system is suitable for large scale loss of function screens. Libraries of vectors expressing gRNAs designed to target genes of interest can be easily and inexpensively generated in one cloning step starting from on-chip synthesized oligonucleotides pools8,9,10, a strategy that has been successfully applied to the generation of RNAi libraries14.
However, a growing number of applications of the CRISPR-Cas9 system require the simultaneous expression of two guide RNAs transcribed from independent promoters. For instance, a strategy that significantly reduces off-target mutations while retaining on-target cutting efficiency relies on the expression of a mutant Cas9 having one of the two nuclease domains disrupted (Cas9n; nickase) together with two gRNAs targeting off-set sites on opposite DNA strands15,16.
Because the sequences encoding the gRNA pairs need to be cloned downstream of independent promoters in the same plasmid, available strategies to rapidly generate pooled libraries14 cannot be used. This severely limits the scalability of this approach for functional screens. To overcome this limitation, we have developed a simple one-step method that allows the rapid and efficient cloning of specific gRNA-pairs into virtually any CRISPR-expression vector starting from pools of short oligonucleotides. This methods uses an intermediate circularization-linearization step that ensures that the two gRNAs are cloned downstream of independent U6 promoters in the final plasmid. Importantly, because the sequence of the two gRNAs paired in the dual-expression construct is determined at the oligonucleotide design step, correct pairing of gRNAs for the same genomic locus in the resulting plasmid is ensured by default in our experimental scheme.
We also show that lentiviral vectors expressing gRNA pairs from two identical U6 promoters are prone to recombination and consequent loss of the proximal gRNA. We overcome this problem by generating a novel lentiviral vector containing a human U6 promoter and a modified murine U6 promoter in which key regulatory sequences are replaced by the human equivalents.
By providing a simple, fast and inexpensive way to generate pooled library of paired gRNAs expressing vectors this method greatly expands the potential applications of the CRISPR technology for functional genomic studies.
A single vector for co-delivery of Cas9 and paired gRNAs
To determine whether a pair of gRNAs can be effectively expressed from a single vector, we tested the editing efficiency of a Cas9 expression plasmid containing two tandem U6 promoters, each followed by unique cloning sites and a gRNA-scaffold sequence18.
We designed two guide RNAs targeting two sites 760 bp apart on the murine Trp53 locus and sequentially cloned them downstream of the two U6 promoters (Fig. 1a). Northern blot analysis showed that this plasmid configuration leads to expression of the two guides at approximately equimolar ratios (Fig. 1b). Importantly, gRNAs expressed from a single vector were able to generate on-target indels at approximately the same frequency as gRNAs co-expressed from two different plasmids (Fig. 1c) and could generate the desired genomic deletion (Fig. 1d and Supplementary Fig. 1a).
These data, together with previous work from our lab showing that tandem gRNA expression from a single vector can be used to engineer chromosomal rearrangements in vitro and in vivo18, demonstrate that a single plasmid expressing Cas9 and two guide RNAs can be used for efficient genome editing.
A one-step cloning method for paired-gRNA vectors
For the experiments described above, the two gRNA sequences were cloned sequentially in the recipient vector, a strategy that is incompatible with the generation of medium or large-scale pooled libraries. We therefore devised a method to simultaneously clone two guide RNAs starting from a short-(110 nt) DNA oligonucleotide (Fig. 2). The oligonucleotide contains the sequences corresponding to the two gRNAs separated by a short spacer harbouring two BbsI sites. At the 5’ and 3’ ends of the oligo are short sequences with homology to the U6 promoter and the gRNA scaffold, respectively. Amplification of the oligo by PCR using primers that bind to these regions generates a 148-bp dsDNA molecule that contains 40 bp homologies to the 3’ end of the U6 promoter and to the 5’ end of the gRNA scaffold (Fig. 2).
In addition, we generated a plasmid (pDonor) whose digestion with BbsI yields a 415 bp fragment consisting of a gRNA scaffold and a U6 promoter (Fig. 2, Supplementary Fig. 2). The presence of overlapping sequences at both ends of the Donor fragment and of the PCR product allows for their assembly into an intermediate circular molecule using the Gibson reaction20. Digestion of this circular intermediate with BbsI produces a linear fragment with two distinct 5’ overhangs and allows for directional cloning into a variety of CRISPR-expression vectors9,13,21,22. This generates a final plasmid in which the two gRNAs present in the oligonucleotide are cloned downstream of separate U6 promoters (Figs 1a, 2).
Pooled cloning of gRNA pairs
To test whether this strategy was suitable for cloning pools of gRNA pairs, we designed oligos corresponding to nine different gRNA pairs (A-I; see Table 1) and pooled them at equimolar concentrations (Fig. 3a). We amplified the pool in a single-PCR reaction and ligated the resulting product to the U6:gRNA-scaffold fragment using the Gibson method20 (Fig. 3a). Linearization of the Gibson product with BbsI resulted in the expected 461 bp band (Fig. 3b), which was gel purified and cloned into a linearized vector9 containing a U6 promoter and a gRNA scaffold (Fig. 2). Digestion of DNA from individual bacterial colonies released the expected 1.6-kb band in 10/10 clones (Fig. 3c). To determine the frequency at which individual oligos were correctly cloned, we sequenced plasmid DNA from 90 bacterial colonies. Although 10% of clones contained chimeric inserts—probably generated at the Gibson step (data not shown)—each of the remaining clones contained one of the nine gRNA pairs correctly assembled (Fig. 3d). Importantly, each pair was represented at an approximately equal frequency (min=5, max=13 and average=8.8; Fig. 3e). These results show that the method described here can be used for pooled cloning of multiple gRNA pairs.
Paired gRNA lentiviral vectors for stable transduction of cells
CRISPR-Cas9-based screens have mainly relied on the stable transduction of mammalian cells with pools of lentiviral vectors8,9,10. Since lentiviral vectors harbouring direct repeats are unstable, the presence of two identical human U6 (hU6) promoters in the same vector might lead to viral recombination, loss of gRNA expression and consequently lower genome editing efficiency.
Consistent with this hypothesis, we observed reduced editing efficiency at the 3’ cut site (targeted by the proximal gRNA; gRNA2) in cells transduced with a lentivirus expressing the two gRNAs from identical hU6 promoters (Fig. 4b; lanes 9 and 10). Genomic PCR analysis confirmed that this was due to loss of the proximal hU6-gRNA2 sequence in the proviral genome (Fig. 4c and Supplementary Fig. 1b).
We reasoned that reducing the sequence identity between the two promoters should prevent recombination. We therefore generated two additional pDonor vectors carrying the murine U6 (mU6) promoter, or a synthetic mouse U6 (sU6) promoter harbouring regulatory sequence elements from hU6 (Supplementary Fig. 2). We then used these pDonor vectors to clone the gRNA pair into a recipient lentivirus containing the hU6 promoter, thus producing two new lentiviral constructs (Fig. 4a).
In contrast to cells infected with lentiviruses carrying two hU6 promoters, cells infected with lentiviruses expressing gRNAs from two different promoters displayed largely intact proviruses (Fig. 4c). Accordingly editing efficiency at the two cut sites was comparable to what observed when each gRNA was individually expressed (Fig. 4b), as were the expression levels of the gRNAs (Fig. 4d). Finally, the desired genomic deletion was readily detectable in these cells (Fig. 4d).
These results demonstrate that the use of two different promoters prevents lentiviral recombination and allows simultaneous editing at two sites. The series of pDonor plasmids we have generated can therefore be used to rapidly build paired gRNA libraries in a variety of currently available vectors.
We describe here a novel method that allows the generation of pooled libraries of vectors containing paired gRNAs. The small size of the oligos that contain the gRNA sequences makes them compatible with ‘on-chip’ oligonucleotide synthesis23, enabling the generation of large paired gRNA pooled libraries in a fast and cost-effective manner that is analogous to what has been previously described for shRNA14 and single gRNA screens8,9,10.
In addition to providing a simple method to generate Cas9 nickase-based12,13 pooled libraries, the method described here extends the use of CRISPR-Cas9 screenings to a wider range of biological questions. It allows for instance the generation of deletion libraries against long-noncoding RNA genes, whose functions are unlikely to be compromised by the short insertions/deletions introduced by conventional single-guide CRISPR libraries. It also allows the generation of pooled libraries to delete chromosomal regions that are recurrently lost in human cancers, as well as the systematic functional characterization of DNA-regulatory elements.
pDonor plasmids were generated by cloning the Donor fragments into either an EcoRV-digested pBluescript KS+ vector (pDonor_hU6) or into the Topo Blunt II plasmid (Invitrogene). The sU6 promoter was generated by replacing the regulatory elements of mU6 for those of hU6 (that is, the Octamer motif, the Proximal sequence element and the TATA-box). See Supplementary Fig. 2a for a schematic representation of these plasmids. The sequences contained in each of the pDonors are provided in Supplementary Fig. 2, along with the sequence of the DNA oligos used for cloning gRNA pairs under the various promoters.
Paired gRNA cloning
For the pooled cloning oligonucleotides were mixed at equimolar concentrations and amplified with phusion polymerase (New England Biolabs) using primers that add homology regions to the 3’ region of the hU6 promoter (primer F1) and to the 5’ region of the gRNA scaffold (primer R1). The gel-purified 148-bp amplicon was ligated to the 415-bp Donor fragment—generated by BbsI digestion of the pDonor plasmid—in a 3:1 molar ratio, using the Gibson Assembly Master Mix (New England Biolabs; 1 h at 50 °C). The Gibson reaction was treated with Plasmid Safe exonuclease (Epicenter) for 1 h at 37 °C to remove unligated fragments, column purified (QIAquick PCR purification kit; Qiagen), and digested with BbsI at 37 °C for 3 h. The linearized 461 bp fragment was gel purified and cloned into BsmBI-digested lentiCRISPR vector4 (Addgene plasmid 49535). For 10 bacterial clones, correct assembly was confirmed by digestion of plasmid DNA with NotI and EcoRI enzymes. Sequencing of vectors was done using the F2/R2 primer set. Cloning of the paired lentivirus vectors carrying distinct pol III promoters was done as described above, but we used modified oligos and primers (Supplementary Fig. 2b and Supplementary Table 1). A list of all primer sequences and primer pairs (including corresponding amplicon sizes) used throughout this study is provided in Supplementary Tables 1 and 2, respectively. A detailed step-by-step protocol for the generation of paired gRNA libraries is provided as a Supplementary Method.
Cells were cultured at 37 °C (5% CO2) in DME-HG supplemented with 10% FCS, L-glutamine (2 mM), penicillin (100 U ml−1) and streptomycin (100 μg ml−1). One day after seeding (1 × 106 cells per well; 6-well plate), cells were transfected with 4 μg of plasmid DNA using Lipofectamine 2000 (Invitrogen) according to manufacturer’s instructions. NIH3T3 cells (ATCC; #CRL-1658) were collected either 48 h or 5 days after transfection to detect the expression of the gRNAs or the generation of genomic editing, respectively. For cell infection, 293T cells (ATCC; # CRL-3216) were transfected with lentiviral constructs together with ecotropic packaging plasmids using the protocol described above. Media containing viruses were collected 48 h after transfection and used to infect NIH3T3 cells. Infected cells were selected with puromycin (2 μg ml−1) for 3 days and then collected for further analysis. Analysis of proviral integrity was done by PCR using primers F2 and R2 (Supplementary Fig. 3b). In cells displaying detectable levels of proviral recombination the amplicon corresponding to the recombined viral genome was cloned into Topo Blunt II (Invitrogene) and six clones of the resulting bacterial clones sequenced.
Northern blot analysis and detection of genomic editing
To detect gRNA expression, transduced cells were collected in TRIZOL (invitrogen) and total RNA was isolated according to manufacturers’ protocols. For each sample, 10 μg of RNA were resolved in a 15% Urea-PAGE gel and blotted onto a Hybond-N+ nylon membrane (GE Healthcare). Membranes were ultraviolet-cross-linked and hybridized overnight with 32P-labelled probes against the 5’ region of each gRNA (gRNA1 probe, TTGGACGCCCTCGCAGTGGC; gRNA2 probe, CCTGTTCGGCACACCTGCTG) and against mU6 as a loading control (GCAGGGGCCATGCTAATCTTCTCTGTATCG).
For detection of genomic deletions and indels, cells were collected in lysis buffer (100 mM Tris-HCl pH8.5, 200 mM NaCl, 5 mM EDTA, 0.2% SDS and 100 ng ml−1 proteinase K) and incubated at 55 °C for 4 h. Genomic DNA was extracted with phenol-chloroform followed by ethanol precipitation and amplified by PCR with Phusion polymerase (New England Biolabs). Detection of the genomic deletion was done using primers that flank the gRNA target sites (F3/R3), which leads to the amplification of a ∼1-kb band in cells carrying a wild-type locus, and a 340 bp band in cells carrying the deletion (Supplementary Fig. 3a). Sequencing of the genomic deletion was done after cloning the corresponding amplicon into Topo Blunt II vector (Invitrogene). For the detection of indel formation DNA was amplified with primers that flank the cut sites (5’cut with primers F3/R4; 3’ cut with primers F4/R3) followed by generation of DNA heteroduplexes and DNA digestion with the mismatch-sensitive SURVEYOR nuclease (Transgenomic) according to manufacturer’s instructions. Digestion fragments were resolved on a 2.5% agarose gel (Supplementary Fig. 3c,d).
How to cite this article: Vidigal, J. A. & Ventura, A. Rapid and efficient one-step generation of paired gRNA CRISPR-Cas9 libraries. Nat. Commun. 6:8083 doi: 10.1038/ncomms9083 (2015).
Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816–821 (2012) .
Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826 (2013) .
Sternberg, S. H., Redding, S., Jinek, M., Greene, E. C. & Doudna, J. A. DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507, 62–67 (2014) .
Mao, Z., Bozzella, M., Seluanov, A. & Gorbunova, V. DNA repair by nonhomologous end joining and homologous recombination during cell cycle in human cells. Cell Cycle 7, 2902–2906 (2008) .
Sanchez-Rivera, F. J. et al. Rapid modelling of cooperating genetic events in cancer through somatic genome editing. Nature 516, 428–431 (2014) .
Wang, H. et al. One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell 153, 910–918 (2013) .
Swiech, L. et al. In vivo interrogation of gene function in the mammalian brain using CRISPR-Cas9. Nat. Biotechnol. 33, 102–106 (2014) .
Zhou, Y. et al. High-throughput screening of a CRISPR/Cas9 library for functional genomics in human cells. Nature 509, 487–491 (2014) .
Shalem, O. et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343, 84–87 (2014) .
Wang, T., Wei, J. J., Sabatini, D. M. & Lander, E. S. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343, 80–84 (2014) .
Platt, R. J. et al. CRISPR-Cas9 Knockin Mice for Genome Editing and Cancer Modeling. Cell 159, 440–455 (2014) .
Jao, L. E., Wente, S. R. & Chen, W. Efficient multiplex biallelic zebrafish genome editing using a CRISPR nuclease system. Proc. Natl Acad. Sci. USA 110, 13904–13909 (2013) .
Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819–823 (2013) .
Zuber, J. et al. Toolkit for evaluating genes required for proliferation and survival using tetracycline-regulated RNAi. Nat. Biotechnol. 29, 79–83 (2011) .
Ran, F. A. et al. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154, 1380–1389 (2013) .
Shen, B. et al. Efficient genome modification by CRISPR-Cas9 nickase with minimal off-target effects. Nat. Methods. 11, 399–402 (2014) .
Yang, H. et al. One-step generation of mice carrying reporter and conditional alleles by CRISPR/Cas-mediated genome engineering. Cell 154, 1370–1379 (2013) .
Maddalo, D. et al. In vivo engineering of oncogenic chromosomal rearrangements with the CRISPR/Cas9 system. Nature 516, 423–427 (2014) .
Han, J. et al. Efficient in vivo deletion of a large imprinted lncRNA by CRISPR/Cas9. RNA. Biol. 11, 829–835 (2014) .
Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods. 6, 343–345 (2009) .
Ran, F. A. et al. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 8, 2281–2308 (2013) .
Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods. 11, 783–784 (2014) .
Cleary, M. A. et al. Production of complex nucleic acid libraries using highly parallel in situ oligonucleotide synthesis. Nat. Methods. 1, 241–248 (2004) .
We are grateful to the members of the Ventura lab for helpful discussions, in particular Ciro Bonetti and Danilo Maddalo for insightful comments on the cloning strategy, Paul Ogrodowski for providing technical support and Jennifer Hollenstein for editing the manuscript. We would also like to acknowledge an anonymous reviewer of the manuscript for suggesting the possibility of recombination among identical U6 promoters. This work was funded by a grant from the Geoffrey Beene Cancer Research Foundation and by an NIH Core Grant (P30 CA008748).
The authors declare no competing financial interests.
About this article
Cite this article
Vidigal, J., Ventura, A. Rapid and efficient one-step generation of paired gRNA CRISPR-Cas9 libraries. Nat Commun 6, 8083 (2015). https://doi.org/10.1038/ncomms9083
Unbiased and Tailored CRISPR/Cas gRNA Libraries by Synthesizing Covalently-closed-circular (3Cs) DNA
Genetic interaction mapping and exon-resolution functional genomics with a hybrid Cas9–Cas12a platform
Nature Biotechnology (2020)
Nucleic Acids Research (2020)
ACS Synthetic Biology (2020)
Filoviruses Use the HOPS Complex and UVRAG To Traffic to Niemann-Pick C1 Compartments during Viral Entry
Journal of Virology (2020)