Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

CRISPRscan: designing highly efficient sgRNAs for CRISPR-Cas9 targeting in vivo

Abstract

CRISPR-Cas9 technology provides a powerful system for genome engineering. However, variable activity across different single guide RNAs (sgRNAs) remains a significant limitation. We analyzed the molecular features that influence sgRNA stability, activity and loading into Cas9 in vivo. We observed that guanine enrichment and adenine depletion increased sgRNA stability and activity, whereas differential sgRNA loading, nucleosome positioning and Cas9 off-target binding were not major determinants. We also identified sgRNAs truncated by one or two nucleotides and containing 5′ mismatches as efficient alternatives to canonical sgRNAs. On the basis of these results, we created a predictive sgRNA-scoring algorithm, CRISPRscan, that effectively captures the sequence features affecting the activity of CRISPR-Cas9 in vivo. Finally, we show that targeting Cas9 to the germ line using a Cas9-nanos 3′ UTR led to the generation of maternal-zygotic mutants, as well as increased viability and decreased somatic mutations. These results identify determinants that influence Cas9 activity and provide a framework for the design of highly efficient sgRNAs for genome targeting in vivo.

This is a preview of subscription content, access via your institution

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Measuring the activity of >1,000 sgRNAs.
Figure 2: Stable sgRNAs are more active, G rich and A depleted.
Figure 3: CRISPR-Cas9 activity is modulated by the sgRNA sequence.
Figure 4: Extending the CRISPR target repertoire with truncated, extended and 5′ mismatch–containing sgRNAs.
Figure 5: Targeting CRISPR-Cas9 activity to germ cells.

Similar content being viewed by others

Accession codes

Accessions

Gene Expression Omnibus

Sequence Read Archive

References

  1. Bogdanove, A.J. & Voytas, D.F. TAL effectors: customizable proteins for DNA targeting. Science 333, 1843–1846 (2011).

    Article  CAS  PubMed  Google Scholar 

  2. Cathomen, T. & Joung, J.K. Zinc-finger nucleases: the next generation emerges. Mol. Ther. 16, 1200–1207 (2008).

    Article  CAS  PubMed  Google Scholar 

  3. Hsu, P.D., Lander, E.S. & Zhang, F. Development and applications of CRISPR-Cas9 for genome engineering. Cell 157, 1262–1278 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819–823 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Jinek, M. et al. RNA-programmed genome editing in human cells. eLife 2, e00471 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  6. Mali, P. et al. RNA-guided human genome engineering via Cas9. Science 339, 823–826 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Bassett, A.R. & Liu, J.L. CRISPR/Cas9 and genome editing in Drosophila. J. Genet. Genomics 41, 7–19 (2014).

    Article  CAS  PubMed  Google Scholar 

  8. Friedland, A.E. et al. Heritable genome editing in C. elegans via a CRISPR-Cas9 system. Nat. Methods 10, 741–743 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Hwang, W.Y. et al. Efficient genome editing in zebrafish using a CRISPR-Cas system. Nat. Biotechnol. 31, 227–229 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. 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).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Doench, J.G. et al. Rational design of highly active sgRNAs for CRISPR-Cas9-mediated gene inactivation. Nat. Biotechnol. 32, 1262–1267 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. 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).

    Article  CAS  PubMed  Google Scholar 

  13. Gagnon, J.A. et al. Efficient mutagenesis by Cas9 protein-mediated oligonucleotide insertion and large-scale assessment of single-guide RNAs. PLoS One 9, e98186 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  14. Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816–821 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  15. Huppert, J.L. Four-stranded nucleic acids: structure, function and targeting of G-quadruplexes. Chem. Soc. Rev. 37, 1375–1384 (2008).

    Article  CAS  PubMed  Google Scholar 

  16. Beaudoin, J.D. & Perreault, J.P. Exploring mRNA 3′-UTR G-quadruplexes: evidence of roles in both alternative polyadenylation and mRNA shortening. Nucleic Acids Res. 41, 5898–5911 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Kuscu, C., Arslan, S., Singh, R., Thorpe, J. & Adli, M. Genome-wide analysis reveals characteristics of off-target sites bound by the Cas9 endonuclease. Nat. Biotechnol. 32, 677–683 (2014).

    Article  CAS  PubMed  Google Scholar 

  18. Farboud, B. & Meyer, B.J. Dramatic enhancement of genome editing by CRISPR/Cas9 through improved guide RNA design. Genetics 199, 959–971 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. White, R.M. et al. Transparent adult zebrafish as a tool for in vivo transplantation analysis. Cell Stem Cell 2, 183–189 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Fu, Y., Sander, J.D., Reyon, D., Cascio, V.M. & Joung, J.K. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nat. Biotechnol. 32, 279–284 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Ren, X. et al. Enhanced specificity and efficiency of the CRISPR/Cas9 system with optimized sgRNA parameters in Drosophila. Cell Rep. 9, 1151–1162 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Köprunner, M., Thisse, C., Thisse, B. & Raz, E. A zebrafish nanos-related gene is essential for the development of primordial germ cells. Genes Dev. 15, 2877–2885 (2001).

    PubMed  PubMed Central  Google Scholar 

  23. Mishima, Y. et al. Differential regulation of germline mRNAs in soma and germ cells by zebrafish miR-430. Curr. Biol. 16, 2135–2142 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Chen, J.N. et al. Mutations affecting the cardiovascular system and other internal organs in zebrafish. Development 123, 293–302 (1996).

    CAS  PubMed  Google Scholar 

  25. Halpern, M.E., Ho, R.K., Walker, C. & Kimmel, C.B. Induction of muscle pioneers and floor plate is distinguished by the zebrafish no tail mutation. Cell 75, 99–111 (1993).

    Article  CAS  PubMed  Google Scholar 

  26. Schulte-Merker, S. et al. Expression of zebrafish goosecoid and no tail gene products in wild-type and mutant no tail embryos. Development 120, 843–852 (1994).

    CAS  PubMed  Google Scholar 

  27. Wienholds, E., Koudijs, M.J., van Eeden, F.J., Cuppen, E. & Plasterk, R.H. The microRNA-producing enzyme Dicer1 is essential for zebrafish development. Nat. Genet. 35, 217–218 (2003).

    Article  CAS  PubMed  Google Scholar 

  28. Giraldez, A.J. et al. MicroRNAs regulate brain morphogenesis in zebrafish. Science 308, 833–838 (2005).

    Article  CAS  PubMed  Google Scholar 

  29. Duquette, M.L., Handa, P., Vincent, J.A., Taylor, A.F. & Maizels, N. Intracellular transcription of G-rich DNAs induces formation of G-loops, novel structures containing G4 DNA. Genes Dev. 18, 1618–1629 (2004).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Montague, T.G., Cruz, J.M., Gagnon, J.A., Church, G.M. & Valen, E. CHOPCHOP: a CRISPR/Cas9 and TALEN web tool for genome editing. Nucleic Acids Res. 42, W401–W407 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Varshney, G.K. et al. High-throughput gene targeting and phenotyping in zebrafish using CRISPR/Cas9. Genome Res. 25, 1030–1042 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Shah, A.N., Davey, C.F., Whitebirch, A.C., Miller, A.C. & Moens, C.B. Rapid reverse genetic screening using CRISPR in zebrafish. Nat. Methods 12, 535–540 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Ciruna, B. et al. Production of maternal-zygotic mutant zebrafish by germ-line replacement. Proc. Natl. Acad. Sci. USA 99, 14919–14924 (2002).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Cunningham, F. et al. Ensembl 2015. Nucleic Acids Res. 43, D662–D669 (2015).

    Article  CAS  PubMed  Google Scholar 

  35. 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).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Meeker, N.D., Hutchinson, S.A., Ho, L. & Trede, N.S. Method for isolation of PCR-ready genomic DNA from zebrafish tissues. Biotechniques 43 610, 612, 614 (2007).

  37. Cifuentes, D. et al. A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity. Science 328, 1694–1698 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Beaudoin, J.D., Jodoin, R. & Perreault, J.P. In-line probing of RNA G-quadruplexes. Methods 64, 79–87 (2013).

    Article  CAS  PubMed  Google Scholar 

  39. Beaudoin, J.D. & Perreault, J.P. 5′-UTR G-quadruplex structures acting as translational repressors. Nucleic Acids Res. 38, 7022–7036 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. del Viso, F., Bhattacharya, D., Kong, Y., Gilchrist, M.J. & Khokha, M.K. Exon capture and bulk segregant analysis: rapid discovery of causative mutations using high-throughput sequencing. BMC Genomics 13, 649 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  41. Nieuwkoop, P.D. & Faber, J. Normal Table of Xenopus laevis (Daudin): A Systematical and Chronological Survey of the Development from the Fertilized Egg till the End of Metamorphosis (Garland, 1994).

  42. Khokha, M.K. et al. Techniques and probes for the study of Xenopus tropicalis development. Dev. Dyn. 225, 499–510 (2002).

    Article  CAS  PubMed  Google Scholar 

  43. Langmead, B. & Salzberg, S.L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Wu, T.D. & Watanabe, C.K. GMAP: a genomic mapping and alignment program for mRNA and EST sequences. Bioinformatics 21, 1859–1875 (2005).

    Article  CAS  PubMed  Google Scholar 

  45. Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).

    PubMed  PubMed Central  Google Scholar 

  46. Lorenz, R. et al. ViennaRNA Package 2.0. Algorithms Mol. Biol. 6, 26 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  47. Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning (Springer, 2001).

  48. Meinshausen, N. & Bühlmann, P. Stability selection. J. R. Stat. Soc. Series B Stat. Methodol. 72, 417–473 (2010).

    Article  Google Scholar 

  49. Zhang, Y. et al. Canonical nucleosome organization at promoters forms during genome activation. Genome Res. 24, 260–266 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank E. Fleming and H. Codore for technical help; D. Cifuentes, A. Bazzini and M. Lee for discussions; all the members of the Giraldez laboratory for intellectual and technical support; and S. Lau, M. Lee and M. Fernandez-Fuertes for cloning of nanos 3′ UTR, help with MNase analysis and help with pictures of the adult fish, respectively. We thank C. Takacs, M. Lee and K. Divito for manuscript editing. Supported by the Swiss National Science Foundation (grant P2GEP3_148600 to C.E.V.), Programa de Movilidad en Áreas de Investigación priorizadas por la Consejería de Igualdad, Salud y Políticas Sociales de la Junta de Andalucía (M.A.M.-M.), the Fonds de Recherche du Québec (grant 29818 to J.-D.B.) and the US National Institutes of Health (grants R21 HD073768, R01 GM103789, R01 GM102251, R01 GM101108 and GM081602 to A.J.G. and grant R01 HD081379 to E.K.M. and M.K.K.). M.K.K. is supported by the Edward Mallinckrodt Jr. Foundation.

Author information

Authors and Affiliations

Authors

Contributions

M.A.M.-M., C.E.V. and A.J.G. designed the project, performed experiments and data analysis and wrote the manuscript. J.-D.B. created the sgRNA libraries, performed G-quadruplex experiments and helped write part of the manuscript. J.P.F. performed F0 phenotype analysis and Xenopus phenotype analysis with M.A.M.-M. E.K.M. carried out Xenopus injections and phenotype analysis. M.K.K. provided reagents and materials.

Corresponding author

Correspondence to Antonio J Giraldez.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Measuring the activity of >1,000 sgRNAs.

(a) Diagram illustrating the 1,280 sgRNAs targeting the 128 loci (gray lines) with the genomic orientation. (b) Schematic diagram illustrating the experimental setup to analyze CRISPR-Cas9–mediated mutations in vivo. 1,280 sgRNAs in pools of 80 along with cas9 mRNA were injected into one-cell-stage zebrafish embryos collected after 9 h of development. In parallel, sgRNA levels at 0 and 1.25 hpf were measured to analyze sgRNA stability and to normalize sgRNA activity. (c) Schematic diagram illustrating the bioinformatics pipeline developed to characterize CRISPR-Cas9–induced mutations in vivo. egr2b and smarce1 are two representative genes among the 128 targeted loci. (d) Number of WT reads (used for normalization), aligned reads to the 128 targeted loci with potential indels and unaligned/filtered reads (PCR oligo-dimers and mispriming, etc.). (e) PCR approach to obtain a 117-bp product template for sgRNA in vitro transcription (details in Online Methods). An oligonucleotide containing the T7 promoter (green), the 20 nt of the specific sgRNA DNA binding sequence (red) and a constant 15-nt tail for annealing is used in combination with an 80-nt reverse oligo (tail primer) to add the sgRNA constant 3ʹ end (in blue). (f) Histogram of insertion lengths induced by single sgRNAs (median of 4 nt). (g) Scheme illustrating an sgRNA (binding sequence in red, tail in blue) binding to the genomic target site (black) and the PAM sequence NGG (orange). Purple triangles indicate predicted cleavage sites. (Based on Jao et al.35.)

Supplementary Figure 2 Stable sgRNAs are G rich, A depleted and more active.

(a) Diagram illustrating the experiment to analyze the Cas9 loading activity in vivo. One-cell-stage embryos were injected with 1,280 sgRNAs in pools of 80 along with FLAG-cas9 mRNA (Online Methods). sgRNA levels were measured at 0 and 1.25 hpf, and immunoprecipitation of FLAG-Cas9 was performed at 6 and 9 hpf to analyze levels of loaded sgRNAs. (b) Analysis of the immunoprecipitation of FLAG-Cas9 at 6 hpf (bottom). 1/50 of the total input (IN), the immunoprecipitation (IP) and the supernatant after the immunoprecipitation (SN) were analyzed by western blotting using FLAG and γ-tubulin (as a loading control) antibodies. (c) Biplot of sgRNA levels (log2 RPM) comparing 0 and 1.25 hpf (experiment shown in a), colored to indicate the frequencies of the four nucleotides in each sgRNA. Corresponding Spearman correlations between nucleotide frequencies and sgRNA stability (ratio of 1.25 hpf to 0 hpf levels) are shown (right), with P values indicated. Bar plot representing the nucleotide composition of the 20% most stable sgRNAs compared to all others (bottom). Bars show log-odds scores of nucleotide frequencies for each position in the sgRNA (1 to 20) (G-test: *<0.05, **<0.01). (d) Biplot illustrating stable and unstable groups of sgRNAs, defined by >2-fold enrichment or depletion between 0 and 1.25 hpf (log2 RPM). sgRNAs with low read counts (bottom 10%) were excluded (gray lines). (e) Box-and-whisker plots (box spans first to last quartiles with 1.5× the interquartile range distance for whiskers) showing sgRNA activity (left) and the input levels (right) in the stable and unstable sgRNAs. Mann-Whitney U-test (****P < 0.0001). (f) Bar plot representing the nucleotide composition of the 20% most loaded sgRNAs normalized to the 1.25-hpf input compared to all others (see c). (g) Biplot of sgRNA levels (log2 RPM) comparing 1.25 hpf and loaded into Cas9 at 6 hpf, colored to indicate the frequencies of C and T in each sgRNA. Corresponding Spearman correlations between nucleotide frequencies and sgRNA stability (ratio of loaded at 6 h to 1.25 hpf levels) are shown (right), with P values indicated. (h) Bar plot representing the nucleotide composition of the 20% most loaded sgRNAs normalized to the 0-hpf input compared to all others (see c). (i) Biplot of sgRNA levels comparing 0 hpf and loaded into Cas9 at 6 hpf (see g). (j) Biplot illustrating most and least loaded into Cas9 groups of sgRNAs (see d). (k) Box-and-whisker plots showing sgRNA activity (left) and the input levels (right) in the most and least loaded into Cas9 sgRNAs. Mann-Whitney U-test (****P < 0.0001).

Supplementary Figure 3 sgRNA stability depends on folding energy and G-quadruplex formation.

(a) Biplot comparing sgRNA level at 1.25 hpf (log2 RPM) with their activity. Spearman correlation is indicated. (b) Biplots comparing sgRNA stability with their ensemble free energy (EFE) in kcal/mol. Each sgRNA is colored to indicate the frequencies of the four nucleotides. (c) Linear regressions of guanine content below (green) and above seven guanines (yellow) in the variable part (binding sequence) of the sgRNA (which is the minimum number required to form a G-quadruplex). The slopes are shown in the inset. The stability increases more rapidly after seven guanines. sgRNA stability is the ratio of sgRNA levels (log2 RPM) between 0 and 1.25 hpf inputs from Supplementary Figure 1b. (d) Bar graphs showing the intensity band ratio (KCl/LiCl) quantified using the SAFA software for each nucleotide found in the variable part (green) and 5ʹ of the constant part (purple) of each sgRNA. Red bars indicate positions for which the ratio is above 2. For nucleotide A18 of stable sgRNA #1 and C16 of stable sgRNA #2, the intensity ratio was over 8; therefore their corresponding value has been added on top of the bar. Error bars correspond to quantification of two technical replicates. (e) Autoradiogram of a 10% denaturing PAGE of the in-line probing of two stable and two unstable sgRNAs in the presence of either 100 mM LiCl (Li) or 100 mM KCl (K). The L and T1 lanes indicate, respectively, the alkaline hydrolysis and ribonuclease T1 mapping for each sgRNA. The constant (purple) and variable (green) parts of the sgRNA are indicated (left). A purple asterisk indicates the guanine 21, the first nucleotide of the constant part, for each sgRNA. Red lines or dots reveal nucleotides for which a band intensity ratio (KCl/LiCl) higher than 2 was observed after quantification. (f) Analysis of the immunoprecipitation of Flag-Cas9 at 9 hpf. 1/50 of the total input (IN), the immunoprecipitation (IP) and the supernatant after the immunoprecipitation (SN) were analyzed by western blotting using Flag and γ-tubulin antibodies. Negative control with uninjected embryos is shown (bottom). (g) Biplot of sgRNA levels pulled down at 6 and 9 hpf (log2 RPM). Pearson correlation is shown.

Supplementary Figure 4 CRISPRscan independent validation.

(a) Phenotypic evaluation of 11 sgRNA sites targeting albino in zebrafish. Top: Diagram showing 11 sgRNA sites targeting albino exons 1 and 2 used in an independent validation of the prediction model. Middle: Phenotypes obtained after the injection of the sgRNAs, showing different levels of mosaicism compared to the wild type (WT). Lateral views and insets of the eyes of 48-hpf embryos are shown. Picture of an albino loss-of-function mutant (-/-) described by White et al.19 (right). Bottom: Stacked bar plots showing the percentage of albino-like (white), mosaic (gray) and phenotypically WT (black) embryos 48 hpf after injection. Predicted CRISPRscan and Doench et al. scores, ranks and number of embryos evaluated (n) are shown for each sgRNA. 100 pg of cas9 mRNA and 10 pg of each sgRNA were injected. (b) Phenotypic evaluation of seven sgRNA sites targeting golden exons 1, 2, 3 and 6 in zebrafish (see a). Injections were performed in golden heterozygous (+/-) embryos. Picture of a golden loss-of-function mutant (-/-) described by White et al.19 (right). (c) Diagram showing seven sgRNA sites targeting ntla exons 1 and 2 in zebrafish (see a). Levels of mosaicism compared to the wild type (WT) were evaluated at 32 hpf. Class I: Short tail (least extreme). Class II: Absence of notochord and short tail (medium level). Class III: Absence of notochord and extremely short tail (most extreme). (d) Diagram showing ten sgRNA sites targeting albino exons 1 and 3 in frog (see a). Phenotypes were evaluated by comparison to WT at stage 46 (head dorsal views and insets of the eyes of the embryos are shown). 500 pg of cas9 mRNA and 400 pg of each sgRNA were injected in one-cell-stage embryos.

Supplementary Figure 5 Chromatin accessibility and off-target seed binding have no effect on CRISPR-Cas9 activity.

(a) Biplot of RPM from MNase assay (log2) with normalized sgRNA activity (rank). Pearson correlation is shown. To determine target site accessibility, MNase-Seq maps in zebrafish embryos49 were analyzed. However, no significant correlation between average MNase read coverage and normalized sgRNA activity was observed (r = –0.046, P = 0.15). Thus, no chromatin accessibility effect on sgRNA activity was detected. (b) Biplot of number of potential Cas9 binding sites (log2) per sgRNA with normalized activity (rank). Potential binding sites contained the 5 nt or 7 nt closest to the PAM sequence and NGG of each analyzed sgRNA. Pearson correlation is shown. Cas9 can bind off-target sites in genomic regions with partial complementarity to the sgRNA17,50. This depends on pairing of 5–10 nt within the 3' end of the sgRNA (positions 11–20), termed the ‘seed’. Although this form of binding does not result in significant cleavage activity50, whether it could reduce sgRNA activity by competing with on-target binding was tested. To this end, potential off-target binding sites in the zebrafish genome for the sgRNAs in our study were identified, consisting of 5- or 7-nt perfect-match seeds followed by a PAM sequence. As a control, the 5' ends of the sgRNA distal to the PAM were used. The number of potential off-target binding sites was not significantly correlated with sgRNA activity, either across the whole genome or only within genes. Together, these results indicate that nucleosome positioning and off-target seed binding have no significant effect on CRISPR-Cas9 activity.

Supplementary Reference

50. Wu, X. et al. Genome-wide binding of the CRISPR endonuclease Cas9 in mammalian cells. Nat. Biotechnol. 32, 670–676 (2014).

Supplementary Figure 6 Alternative sgRNA formulations have variable activity.

(a) Number of WT reads (used for normalization), aligned reads to the 64 targeted loci with potential indels and unaligned/filtered reads (PCR oligo-dimers and mispriming, etc.). (b) Diagram describing the 11 classes of alternative sgRNA targets analyzed in this study. PAM sequence and the alternative features are highlighted in red and in green, respectively. (c) Diagram illustrating the 640 sgRNAs targeting the 64 loci (gray lines). The 11 alternative sgRNAs are colored by type and detailed in b. CRISPRscan was adapted to predict the activity of the most efficient alternative sgRNAs (gG18; Gg18; GG17; GG16). (d) Biplot of sgRNA levels (log2 RPM) comparing 0 and 1.25 hpf, colored to indicate the frequencies of C and T in each sgRNA. Corresponding Spearman correlations between nucleotide frequencies and sgRNA stability (ratio of 1.25 hpf to 0 hpf levels) are shown (right), with P values indicated. (e) Diagram showing 5ʹ mismatch-containing sgRNA binding sites targeting golden exon 1, 3 and 5 analyzed in this study. (f) Percentage of golden like, mosaic and WT embryos 48 h after being injected with the sgRNAs (from e). Gg18 sgRNAs are compared with their canonical versions (GA18). Injections were performed as described in Supplementary Figure 4b. Phenotype analysis was based on Supplementary Figure 4b. χ2 test (***P < 0.001). (g) Diagram showing truncated and 5ʹ mismatch–containing sgRNA binding sites targeting albino exons 1 and 2 analyzed in this study. (h) Percentage of albino-like, mosaic and WT embryos 48 h after being injected with the sgRNAs (from g). Shorter (GG17, GG16) and Gg18 sgRNAs are compared with their canonical versions GG18 (left) and GA18 (right), respectively. Phenotype analysis was based on Figure 3c. χ2 test (*P < 0.05). Injections were performed as described in Supplementary Figure 4a. (i) Diagram showing two options evaluated to apply the canonical scoring model to shorter GG17 sgRNAs: (i) index position (red asterisk) is omitted from the encoding of shorter sgRNAs and thus not scored (left); or (ii) index position is assigned best-approximation base identities (red letter), defined by the next nucleotide downstream (right) (Online Methods). The retained highest performing option consisting of duplicating position 3 of the sgRNA is indicated. (j) Performance of two options evaluated to apply the canonical scoring model to shorter sgRNAs. Distributions (kernel density estimations) of correlations between experimental rank and CRISPRscan predicted score for GG16 and GG17 alternatives sgRNAs. Despite the smaller number of sgRNAs analyzed, a significant correlation with the experimental activity was found for GG17 (r = 0.40, P = 0.007) but not with GG16.

Supplementary Figure 7 sgRNA/Cas9 activity can result in a lethal phenotype for many essential genes.

(a) Wild-type embryos were injected with a combination of three sgRNAs (20 pg each) targeting sox32, s1pr2 or cdh1 and 100 pg of cas9 mRNA. Pictures were taken at 48 (ventral view) or 10 hpf. Arrows are indicating two functional hearts (sox32 and s1pr2), and the arrowhead is pointing out the lack of complete epiboly (cdh1). (b) Stacked bar graph showing the percentage of coherent F0 phenotype (Mutant phenotype) after injection with sox32, s1pr2, cdh1, ntla, tbx6 or ndr1/2 sgRNAs. (c) Table showing the references where the analyzed loss-of-function phenotypes are described.

Supplementary References

51. van Eeden, F.J. et al. Mutations affecting somite formation and patterning in the zebrafish, Danio rerio. Development 123, 153–164 (1996).

52. Schier, A.F. et al. Mutations affecting the development of the embryonic zebrafish brain. Development 123, 165–178 (1996).

53. Pei, W., Williams, P.H., Clark, M.D., Stemple, D.L. & Feldman, B. Environmental and genetic modifiers of squint penetrance during zebrafish embryogenesis. Dev. Biol. 308, 368–378 (2007).

54. Kane, D.A. et al. The zebrafish epiboly mutants. Development 123, 47–55 (1996).

Supplementary Figure 8 Bypassing lethal or deleterious sgRNA/Cas9-induced F0 phenotypes with Cas9-nanos 3ʹ UTR.

(a) Group picture of 48-hpf-old embryos injected with a combination of three sgRNAs (20 pg each) targeting ntla and 100 pg of cas9-nanos or cas9–β-globin mRNA. (b) Scheme illustrating the F0 in-cross between Cas9-nanos–injected fishes. Pictures of 32-hpf-old embryos showing the expected phenotype for a loss-of-function mutant of ntla (bottom).

Supplementary Figure 9 Comparison of CRISPRscan performance with that of published models.

(a) Performance of linear regression–based prediction model (CRISPRscan). sgRNAs were divided into quintiles on the basis of CRISPRscan scores (horizontal axis), and then each quintile was evaluated based on its experimentally determined activity (colors indicate five activity levels). The top-scoring quintile of sgRNAs (far right bar) contains 54% of the most active sgRNAs by experimental evaluation (dark green) and only 2% of the least active (dark red). (b) Scatter plot showing the correlation between CRISPRscan scores and experimentally measured activities (rank-percentile) based on all phenotypes used to independently validate CRISPRscan (Fig. 3c,d and Supplementary Fig. 4). Spearman correlation and P value are indicated. (c) Same analysis as in a with Doench et al. scores11. (d) Same analysis as in b with Doench et al. scores11. (e) Performance of Montague et al.30 categorization of sgRNAs in three classes: (i) extreme GC content (40% to 80%), (ii) presence of G at position 20, and (iii) the rest of the sgRNAs divided into quintiles on the basis of their experimentally determined activities (horizontal axis), with the proportion of the three classes in each quintile determined (colors indicate three classes). (f) Performance of CRISPRscan and Doench et al.11 models evaluated on F1 germline transmission rates measured by Varshney et al.31. Top and bottom 25% transmission rates were scored with both models. Box plot of each quartile is shown. Mann-Whitney U-test (**P < 0.01).

Supplementary information

Supplementary Text and Figures

Supplementary Figure 1–9 (PDF 2897 kb)

Supplementary Table 1

Oligo sequences (XLSX 76 kb)

Supplementary Table 2

Formation of G-quadruplex in sgRNAs (XLSX 108 kb)

Supplementary Table 3

Parameters of CRISPRscan model (XLSX 6 kb)

Supplementary Data Set 1

Schematic representation of deletions and insertions found on 128 targeted loci by canonical sgRNAs (PDF 6938 kb)

Supplementary Data Set 2

Schematic representation of deletions and insertions found on 64 targeted loci by alternative sgRNAs (PDF 3482 kb)

Source data

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Moreno-Mateos, M., Vejnar, C., Beaudoin, JD. et al. CRISPRscan: designing highly efficient sgRNAs for CRISPR-Cas9 targeting in vivo. Nat Methods 12, 982–988 (2015). https://doi.org/10.1038/nmeth.3543

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nmeth.3543

This article is cited by

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing