Improved specificity of TALE-based genome editing using an expanded RVD repertoire (original) (raw)

References

1. Moscou, M.J. & Bogdanove, A.J. A simple cipher governs DNA recognition by TAL effectors. Science 326, 1501 (2009).
   Article CAS Google Scholar
2. Boch, J. et al. Breaking the code of DNA binding specificity of TAL-type III effectors. Science 326, 1509–1512 (2009).
   Article CAS Google Scholar
3. Cermak, T. et al. Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res. 39, e82 (2011).
   Article CAS Google Scholar
4. Streubel, J., Blücher, C., Landgraf, A. & Boch, J. TAL effector RVD specificities and efficiencies. Nat. Biotechnol. 30, 593–595 (2012).
   Article CAS Google Scholar
5. Reyon, D. et al. FLASH assembly of TALENs for high-throughput genome editing. Nat. Biotechnol. 30, 460–465 (2012).
   Article CAS Google Scholar
6. Schmid-Burgk, J.L., Schmidt, T., Kaiser, V., Höning, K. & Hornung, V. A ligation-independent cloning technique for high-throughput assembly of transcription activator–like effector genes. Nat. Biotechnol. 31, 76–81 (2013).
   Article CAS Google Scholar
7. Kim, Y. et al. A library of TAL effector nucleases spanning the human genome. Nat. Biotechnol. 31, 251–258 (2013).
   Article CAS Google Scholar
8. Deng, D. et al. Structural basis for sequence-specific recognition of DNA by TAL effectors. Science 335, 720–723 (2012).
   Article CAS Google Scholar
9. Mak, A.N., Bradley, P., Cernadas, R.A., Bogdanove, A.J. & Stoddard, B.L. The crystal structure of TAL effector PthXo1 bound to its DNA target. Science 335, 716–719 (2012).
   Article CAS Google Scholar
10. Christian, M. et al. Targeting DNA double-strand breaks with TAL effector nucleases. Genetics 186, 757–761 (2010).
    Article CAS Google Scholar
11. Li, T. et al. TAL nucleases (TALNs): hybrid proteins composed of TAL effectors and FokI DNA-cleavage domain. Nucleic Acids Res. 39, 359–372 (2011).
    Article Google Scholar
12. Miller, J.C. et al. A TALE nuclease architecture for efficient genome editing. Nat. Biotechnol. 29, 143–148 (2011).
    Article CAS Google Scholar
13. Sun, N. & Zhao, H. Transcription activator-like effector nucleases (TALENs): a highly efficient and versatile tool for genome editing. Biotechnol. Bioeng. 110, 1811–1821 (2013).
    Article CAS Google Scholar
14. Hockemeyer, D. et al. Genetic engineering of human pluripotent cells using TALE nucleases. Nat. Biotechnol. 29, 731–734 (2011).
    Article CAS Google Scholar
15. Osborn, M.J. et al. TALEN-based gene correction for epidermolysis bullosa. Mol. Ther. 21, 1151–1159 (2013).
    Article CAS Google Scholar
16. Ding, Q. et al. A TALEN genome-editing system for generating human stem cell-based disease models. Cell Stem Cell 12, 238–251 (2013).
    Article CAS Google Scholar
17. Tsai, S.Q. et al. Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome editing. Nat. Biotechnol. 32, 569–576 (2014).
    Article CAS Google Scholar
18. Guilinger, J.P., Thompson, D.B. & Liu, D.R. Fusion of catalytically inactive Cas9 to FokI nuclease improves the specificity of genome modification. Nat. Biotechnol. 32, 577–582 (2014).
    Article CAS Google Scholar
19. Cong, L., Zhou, R., Kuo, Y.C., Cunniff, M. & Zhang, F. Comprehensive interrogation of natural TALE DNA-binding modules and transcriptional repressor domains. Nat. Commun. 3, 968 (2012).
    Article Google Scholar
20. Tesson, L. et al. Knockout rats generated by embryo microinjection of TALENs. Nat. Biotechnol. 29, 695–696 (2011).
    Article CAS Google Scholar
21. Mali, P. et al. CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat. Biotechnol. 31, 833–838 (2013).
    Article CAS Google Scholar
22. Juillerat, A. et al. Comprehensive analysis of the specificity of transcription activator-like effector nucleases. Nucleic Acids Res. 42, 5390–5402 (2014).
    Article CAS Google Scholar
23. Guilinger, J.P. et al. Broad specificity profiling of TALENs results in engineered nucleases with improved DNA-cleavage specificity. Nat. Methods 11, 429–435 (2014).
    Article CAS Google Scholar
24. Fu, Y. et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat. Biotechnol. 31, 822–826 (2013).
    Article CAS Google Scholar
25. Hsu, P.D. et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat. Biotechnol. 31, 827–832 (2013).
    Article CAS Google Scholar
26. Pattanayak, V. et al. High-throughput profiling of off-target DNA cleavage reveals RNA-programmed Cas9 nuclease specificity. Nat. Biotechnol. 31, 839–843 (2013).
    Article CAS Google Scholar
27. Bogdanove, A.J. & Voytas, D.F. TAL effectors: customizable proteins for DNA targeting. Science 333, 1843–1846 (2011).
    Article CAS Google Scholar
28. Blackwell, T.K. & Weintraub, H. Differences and similarities in DNA-binding preferences of MyoD and E2A protein complexes revealed by binding site selection. Science 250, 1104–1110 (1990).
    Article CAS Google Scholar
29. Yang, J. et al. Complete decoding of TAL effectors for DNA recognition. Cell Res. 24, 628–631 (2014).
    Article CAS Google Scholar
30. Doyon, Y. et al. Transient cold shock enhances zinc-finger nuclease-mediated gene disruption. Nat. Methods 7, 459–460 (2010).
    Article CAS Google Scholar
31. Christian, M.L. et al. Targeting G with TAL effectors: a comparison of activities of TALENs constructed with NN and NK repeat variable di-residues. PLoS ONE 7, e45383 (2012).
    Article CAS Google Scholar
32. Yang, Y. & Gabriel, D.W. Intragenic recombination of a single plant pathogen gene provides a mechanism for the evolution of new host specificities. J. Bacteriol. 177, 4963–4968 (1995).
    Article CAS Google Scholar
33. Yang, B., Sugio, A. & White, F.F. Avoidance of host recognition by alterations in the repetitive and C-terminal regions of AvrXa7, a type III effector of Xanthomonas oryzae pv. oryzae. Mol. Plant Microbe Interact. 18, 142–149 (2005).
    Article CAS Google Scholar
34. Mussolino, C. et al. A novel TALE nuclease scaffold enables high genome editing activity in combination with low toxicity. Nucleic Acids Res. 39, 9283–9293 (2011).
    Article CAS Google Scholar
35. Doyon, Y. et al. Enhancing zinc-finger-nuclease activity with improved obligate heterodimeric architectures. Nat. Methods 8, 74–79 (2011).
    Article CAS Google Scholar
36. Miller, J.C. et al. Enhancing zinc-finger-nuclease activity with improved obligate heterodimeric architectures. Nat. Biotechnol. 25, 778–785 (2007).
    Article CAS Google Scholar
37. Lamb, B.M., Mercer, A.C. & Barbas, C.F. III. Directed evolution of the TALE N-terminal domain for recognition of all 5′ bases. Nucleic Acids Res. 41, 9779–9785 (2013).
    Article CAS Google Scholar
38. Doyle, E.L. et al. TAL effector specificity for base 0 of the DNA target is altered in a complex, effector- and assay-dependent manner by substitutions for the tryptophan in cryptic repeat-1. PLoS ONE 8, e82120 (2013).
    Article Google Scholar
39. Li, L. GADEM: a genetic algorithm guided formation of spaced dyads coupled with an EM algorithm for motif discovery. J. Comput. Biol. 16, 317–329 (2009).
    Article CAS Google Scholar
40. Perez, E.E. et al. Establishment of HIV-1 resistance in CD4+ T cells by genome editing using zinc-finger nucleases. Nat. Biotechnol. 26, 808–816 (2008).
    Article CAS Google Scholar

Download references

Acknowledgements

We thank S. Abrahamson for critically reviewing this manuscript. We thank M. Holmes and H. Zhang for helpful discussions.

Author information

Authors and Affiliations

1. Sangamo BioSciences Inc., Richmond, California, USA
   Jeffrey C Miller, Lei Zhang, Danny F Xia, John J Campo, Irina V Ankoudinova, Dmitry Y Guschin, Joshua E Babiarz, Xiangdong Meng, Sarah J Hinkley, Stephen C Lam, David E Paschon, Anna I Vincent, Gladys P Dulay, Kyle A Barlow, David A Shivak, Elo Leung, Jinwon D Kim, Rainier Amora, Fyodor D Urnov, Philip D Gregory & Edward J Rebar

Authors

1. Jeffrey C Miller
   You can also search for this author inPubMed Google Scholar
2. Lei Zhang
   You can also search for this author inPubMed Google Scholar
3. Danny F Xia
   You can also search for this author inPubMed Google Scholar
4. John J Campo
   You can also search for this author inPubMed Google Scholar
5. Irina V Ankoudinova
   You can also search for this author inPubMed Google Scholar
6. Dmitry Y Guschin
   You can also search for this author inPubMed Google Scholar
7. Joshua E Babiarz
   You can also search for this author inPubMed Google Scholar
8. Xiangdong Meng
   You can also search for this author inPubMed Google Scholar
9. Sarah J Hinkley
   You can also search for this author inPubMed Google Scholar
10. Stephen C Lam
    You can also search for this author inPubMed Google Scholar
11. David E Paschon
    You can also search for this author inPubMed Google Scholar
12. Anna I Vincent
    You can also search for this author inPubMed Google Scholar
13. Gladys P Dulay
    You can also search for this author inPubMed Google Scholar
14. Kyle A Barlow
    You can also search for this author inPubMed Google Scholar
15. David A Shivak
    You can also search for this author inPubMed Google Scholar
16. Elo Leung
    You can also search for this author inPubMed Google Scholar
17. Jinwon D Kim
    You can also search for this author inPubMed Google Scholar
18. Rainier Amora
    You can also search for this author inPubMed Google Scholar
19. Fyodor D Urnov
    You can also search for this author inPubMed Google Scholar
20. Philip D Gregory
    You can also search for this author inPubMed Google Scholar
21. Edward J Rebar
    You can also search for this author inPubMed Google Scholar

Contributions

J.C.M., L.Z. and E.J.R. designed experiments and supervised studies. J.J.C., J.E.B., D.F.X., I.V.A., K.A.B., J.D.K. and R.A. performed studies. J.C.M., D.A.S. and E.L. analyzed data. J.J.C., D.F.X., I.V.A., D.Y.G., X.M., S.J.H., S.C.L., D.E.P., A.I.V., G.P.D. and L.Z. developed new procedures and constructs. J.C.M., P.D.G., F.D.U. and E.J.R. wrote the manuscript.

Corresponding author

Correspondence toEdward J Rebar.

Ethics declarations

Competing interests

All authors are current or past employees of Sangamo BioSciences, a biopharmaceutical company focused on the research and development of engineered DNA-binding proteins for addressing unmet medical needs.

Integrated supplementary information

Supplementary Figure 1 Graphical representation SELEX data from large-scale SELEX studies.

(a) SELEX-derived base frequency matrices for 10 TALEs characterized in large scale SELEX studies. At each matrix position, the frequency of the intended target base is projected above the x-axis, whereas the remaining base frequencies are plotted below the x-axis. Matrix positions are arranged in 5' to 3' order (left to right) with corresponding RVDs listed above. The 3' matrix position corresponds to the C-terminal half-repeat. As shown, base preferences for each RVD can vary over a wide quantitative range. Numerical identifiers are shown adjacent to each plot. For base frequency values see Supplementary Table 1. (b) Left: Average base preference for three of the canonical RVDs when deployed within the C-terminal half-repeat of engineered TALEs. NI and HD exhibit substantially reduced average preference for their target bases in this context as compared with non-terminal repeats (compare with Fig. 1d; ***: p < 10-6; **: p < 10-4). Note that NN was not tested in the C-terminal repeat in these studies. Right: Average base preference of NG observed in successively longer arrays of this RVD. Thymine is highly specified when neither flanking RVD is NG (left bar; 131 examples in this study), or when NG is deployed in two adjacent repeats (second panel from left; 27 examples of an NG-NG pair). However average thymine preference is substantially reduced for NG RVDs that are centrally located within longer runs of this RVD (three rightmost panels; 7 examples of (NG)3 and 2 examples each of (NG)4 and (NG)5; *:p< 10-4 for indicated context compared with the left-most bar). Plotted values are derived from the large scale SELEX study of 76 synthetic TALEs summarized in Supplementary Table 1, and were calculated using the base preferences of every non-terminal repeat from that study. All p values are Mann-Whitney test corrected for false discovery rate.

Supplementary Figure 2 Average base preference for the canonical RVDs when deployed within the N-terminal, C-terminal or nonterminal repeats of a TALE-DNA interface.

Average base preference for the canonical RVDs when deployed within the N-terminal repeat (left panel), C-terminal half repeat (right panel) or other repeat positions (center panel) of a TALE-DNA interface. Values are derived from the large scale SELEX study of 76 synthetic TALEs summarized in Supplementary Table 1, and are plotted in Fig. 1d and Supplementary Fig. 1b. Yellow highlights RVD:base correspondences from the natural code. "N" indicates the number of RVD instances underlying each preference calculation. Note that NN was not tested in the C-terminal half repeat in these studies.

Supplementary Figure 3 Average base preferences for every possible two-repeat unit bearing canonical RVDs.

Values are derived from the large scale SELEX study of 76 synthetic TALEs summarized in Supplementary Table 1, and were calculated using the base preferences of every non-terminal repeat from that study. Yellow highlights RVD:base correspondences from the natural code. "N" indicates the number of instances of the indicated two-repeat unit underlying each preference calculation. A superscript highlights a target base preference that exhibits context dependent variation that is both highly significant (p < 0.001) and substantial (defined as a change in the aggregate frequency of non-targeted bases of > 2-fold or > 0.1). For example, the superscript "B" highlights the observation that the NN RVD exhibits a higher enrichment of guanine when it is flanked by NI than by HD (0.85 vs 0.60; compare panel "NN-NI" with "NN-HD"). F, K, L: p < 0.001; all others: p < 0.10-5. P-values are Mann Whitney test with FDR correction.

Supplementary Figure 4 Gene-modification activities of a PITX3-targeted TALEN1 and five variants that employ noncanonical RVDs for recognition of key bases.

(a) Surveyor assay. Each TALEN was delivered to K562 cells along with an invariant TALEN partner1 via transient transfection of plasmid expression constructs. Transfections were performed in triplicate. After incubation for 3 days at 37C, genomic DNA was isolated, PCR-amplified, and quantified for % indels using the Surveyor assay2, 3. The resultant gel image is shown, with % indels provided below each lane. Each TALE is indicated at the top of its triplicate set of lanes, and is identified by the RVD composition of its first four repeats, with red highlighting noncanonical RVDs. Remaining repeats are as shown in Fig. 3. The rightmost TALE, with all black RVDs, is the original parent. GFP indicates modification levels observed in cells transfected with a GFP expression construct. (b) Table of results from panel a, with calculations of mean and standard deviation (SD).

References

1. Hockemeyer, D. et al. Genetic engineering of human pluripotent cells using TALE nucleases. Nat. Biotechnol. 29, 731-734 (2011).

2. Miller, J.C. et al. An improved zinc-finger nuclease architecture for highly specific genome editing. Nat. Biotechnol. 25, 778-785 (2007).

3. Guschin, D.Y. et al. A rapid and general assay for monitoring endogenous gene modification. Methods Mol. Biol. 649, 247-256 (2010).

Supplementary Figure 5 RVD usage and gene-modification activity of TALEN variants that employ progressively higher fractions of alternative RVDs for DNA recognition.

(a) Variants of TALEN "L" were generated using the indicated RVD sets (columns "A", "C", "G" and "T"; red highlights noncanonical RVDs). Plasmids expressing each protein were introduced into K562 cells with the partner TALEN "R" (the "R557" protein from reference 1). Cells were harvested after three days of culture at 37C or with an intervening 30C cold shock2, and modification levels were determined using the Surveyor assay3, 4. Righthand entries provide TALEN design information; red highlights noncanonical RVDs. (b) As in a, except a higher fraction of RVDs were substituted and cellular studies were replicated either 6 times (TALEN L) or 3 times (remaining variants) without a cold shock arm. Modification levels were quantified via high throughput sequencing with averages and standard deviations shown. (c) As in b except TALENs use exclusively RVDs that differ from the natural code. Note that repeats targeting guanine with a new RVD bear one additional residue substitution (N11S) relative to the parent "NN" repeat.

References

1. Miller, J.C. et al. A TALE nuclease architecture for efficient genome editing. Nat. Biotechnol. 29, 143-148 (2011).

2. Doyon, Y. et al. Transient cold shock enhances zinc-finger nuclease-mediated gene disruption. Nat. Methods 7, 459-460 (2010).

3. Miller, J.C. et al. An improved zinc-finger nuclease architecture for highly specific genome editing. Nat. Biotechnol. 25, 778-785 (2007).

4. Guschin, D.Y. et al. A rapid and general assay for monitoring endogenous gene modification. Methods Mol. Biol. 649, 247-256 (2010).

Supplementary Figure 6 Overview of the process used for generating TALENs that recognize DNA using an expanded set of RVDs.

The generation of L* is provided as an example, while a very similar procedure was used for assembling R*. Activities proceeded in two stages. (a) In the first stage, libraries of TALE tetramers (units comprising four repeats) were created via combinatorial gene assembly, each targeting a distinct quartet base sequence from the TALEN L binding site (shown at top with quartets boxed). RVD mixtures encoded during gene synthesis are listed above each tetramer library repeat. In constructing the libraries, RVDs were chosen that had performed well in the ELISA studies, except that the four canonical RVDs were deliberately excluded in order to maximize the compositional difference between the new TALENs and the original L/R pair. A panel of constructs was then screened via ELISA to identify binding-validated tetramers (three shown for each quartet). These screens identified dozens of binding-competent tetramers for the targeted quartets (Supplementary Table 3). Note that tetramer libraries were assembled in the context of longer host TALE proteins in order to provide sufficient affinity for ELISA screens (Supplementary Fig. 7). (b) In the second stage of assembly, DNA segments encoding binding-validated tetramers were randomly linked with each other and with a degenerate 17th repeat (encoding HG or KG) via combinatorial gene assembly. The resulting 17-repeat library was cloned into a TALEN vector, and 72 randomly chosen constructs, along with 16 discretely assembled TALENs, were screened for gene modification activity when introduced into K562 cells with the partner TALEN R. These studies yielded TALEN L* as the most active variant (at bottom, boxes highlight component tetramers). For screening data see Supplementary Tables 4–5. (c) A similar two-stage process was used to generate TALEN R*. For details see Supplementary Fig. 7 and Supplementary Tables 3 and 6.

Supplementary Figure 7 Tetramer libraries used for making RVD-diversified variants of TALEN L and TALEN R.

Tetramer libraries used for making RVD-diversified variants of TALEN L (top panel) and TALEN R (bottom panel). The full-length TALEN target is shown at the top of each panel, with boxes highlighting component quartets. Beneath, tetramer libraries are sketched, each paired with its quartet target (boxed) and bearing the indicated base-specific RVD mixtures within its repeats (red letters over red TALE repeat segments). Each library is embedded within a larger host TALE bearing five additional, invariant repeats (black letters over grey repeat segments). The libraries targeting the 5' quartets ("TCAT" or "CTTC" - at left) were assembled into repeats 1-4 of the host TALE, in order to match screening context to the ultimate location of this tetramer within the final full-length TALEN. The remaining libraries (at right) were placed at repeats 5-8 to better reflect their ultimate environment within the new TALENs.

Supplementary Figure 8 Candidate loci for off-target cleavage by the L/R and L*/R* TALENs.

The human genome (hg19) was scanned for potential heterodimer (LR and RL) and homodimer (LL and RR) cleavage targets in which the TALE binding sites were separated by 10 - 24 bp. This yielded 23 loci (OT1 - OT23) bearing eight or fewer mismatches between their component TALE binding sites and the intended L and R targets. Genome coordinates are provided for each candidate cleavage site, along with the identity of any surrounding gene and the configuration and separation distance of the component TALEN binding sites ("type" and "spacer" columns). Columns at right list TALEN monomer sites for each candidate off-target locus, with lower case letters highlighting mismatches relative to the identically shaded intended targets provided at top (CCR5 locus). The aggregate number of mismatches is given in the "mismatch" column.

^ OT3 targets a pseudogene homologous to PLEKHM1. OT23 targets a HERV17-int LTR

Supplementary Figure 9 Modification levels at CCR5 and 23 off-target loci as gauged by deep sequencing.

Constructs encoding the indicated TALEN dimers (L/R, L/R*, L*/R, L*/R*) or GFP were introduced into K562 cells via 12 replicate transfections and expressed for three days using a cold shock protocol1. Genome DNA was then isolated, pooled, amplified using locus specific primers (Supplementary Tables 7–8) and submitted for deep sequencing using an Illumina MiSeq. Reads were quality-filtered for identity to the queried locus within two 25-bp sequence tags at each amplicon end, yielding the indicated read totals ("total" columns). Note that for OT14 and OT10, homology with other genome sequences required the application of a more stringent identity filter (46- and 97-bp tags) in order to screen out mispriming events. Reads were then evaluated for indels via application of a length filter (observed length ≠ expected wild-type length) and a location filter (indel must occur within 10 bp of the midpoint between TALEN sites).

Indel frequencies were then adjusted by comparison of TALEN and GFP-treated samples, as follows: first, indels that occurred with similar frequencies (<10-fold difference) in the GFP sample and at least one TALEN sample were considered to be background events and were removed. Next, indels that occurred in both GFP and TALEN samples, but at substantially different frequencies (>10 fold difference) were normalized by subtracting the lower frequency from all samples, yielding the indicated total reads, indel reads and modification frequencies ("total", "indels" and "% indels" columns). Indels emerging from this step are provided in Supplementary Table 9 for the L/R, L*/R* and GFP samples.

A background level of modification was then defined based on the highest signal for any locus after GFP treatment, which was 6 indels comprising 2 unique sequences, for OT9. By these criteria, 13 loci in the L/R-treated cells exhibit modification levels that are above background (OT1-13) of which at least three (OT3, 4 and 11) drop below background after treatment with L*/R*.

Loci that failed to exceed background modification levels (minimum of 7 indels comprising 3 unique sequences) for any treatment were deemphasized via placement at bottom and use of a lighter shade of gray for these entries.

Note that acquisition of an indel will frequently render OT3 and OT4 indistinguishable, since these loci differ by just a single SNP. Ambiguous reads were therefore distributed evenly between these two loci.

References:

1. Doyon, Y. et al. Transient cold shock enhances zinc-finger nuclease-mediated gene disruption. Nat. Methods 7, 459-460 (2010).

Supplementary Figure 11 Study of TALEN activity as a function of the length of the C-terminal region.

(a) sketch of a TALEN with the C-terminal region explored in this study provided as amino acid sequence. Blue letters highlight points of truncation and Fok attachment that were examined for TALEN activity. Associated numbers indicate the length of C-terminal region retained in each construct. Negative numbers indicate truncations that eliminated the entire C-terminal region as well as the indicated residues of the terminal half repeat. Red letters highlight positively charged residues that were recently implicated in nonspecific binding1. The N terminus of the TALEN figure is denoted by the large "N". (b) Diagram of the human CCR5 Δ32 locus with the binding sites for the four TALENs used in this study indicated. Binding sites include the invariant 5' thymine. Note that TALENs L2 and R2 were previously described2 and were referred to as L543 and R551 in that study. Truncation variants sketched in panel a were generated for each of these four TALENs. (c) Plot of gene modification activity as a function of C-terminal region length for the three indicated TALEN pairs. Human K562 cells were transfected with 400 ng of plasmid encoding each construct and incubated at 37° C. Genomic DNA was harvested 3 days post transfection and the percent indels was measured with the surveyor nuclease assay. Maximal activity was observed with the L+17/R2+17 TALEN pair (indicated by "+17" / arrow). Activity levels seen with the standard TALEN architecture are also indicated ("+63" / arrow).

References

1. Guilinger, J.P. et al. Broad specificity profiling of TALENs results in engineered nucleases with improved DNA-cleavage specificity. Nat. Methods 11, 429-435 (2014).

2. Miller, J.C. et al. A TALE nuclease architecture for efficient genome editing. Nat. Biotechnol. 29, 143-148 (2011).

Supplementary Figure 12 Activity and specificity of the L and L* TALEN designs in combination with the +17 architecture.

(a) Sketch of the TALEN pairs examined in this study (L+17/R2+17 and L*+17/R2+17) bound to their intended target in the CCR5 locus. On-target activity for L+17/R2+17, L*+17/R2+17, or eGFP is plotted at right. Human K562 cells were treated with 400 ng of each plasmid encoding the indicated constructs and subjected to a cold shock protocol1. Genomic DNA was harvested 3 days post transfection and genomic DNA from 12 separate replicate tranfections was pooled prior to measuring indels by Illumina MiSeq sequencing. Error bars indicate standard deviation of two technical replicates of the same genomic DNA pools. (b) Diagram showing homodimers of L+17 or L*+17 bound to the off-target site "+17_OT1". Activity of L+17 /R2+17, L*+17 /R2+17, or eGFP at this off-target site is plotted at right.

References:

1. Doyon, Y. et al. Transient cold shock enhances zinc-finger nuclease-mediated gene disruption. Nat. Methods 7, 459-460 (2010).

Supplementary Figure 13 Activity and specificity of the L, R, L* and R* TALEN designs in combination with the Q3 mutations.

(a) On- and off-target activity of the TALEN dimers L/R and L*/R*. The analysis encompassed 20 off-target loci that had been previously validated as detectably modified by L/R, in either this work (OT1-9, OT11-13; Supplementary Fig. 9) or a prior study (offC-5, -15, -16, -36, -38, -49, -69, and -76 from reference 1). Experimental conditions, data processing, and table entries are as described in Supplementary Fig. 9. (b) On- and off-target activity of the L/R and L*/R* TALEN dimers further modified to include the "Q3" mutations1 (modification denoted by "_Q3" appendage). As shown in the rightmost columns (L*_Q3/R*_Q3), use of the novel RVDs in combination with the Q3 mutations yields the lowest aggregate levels of off-target cleavage.

References

1. Guilinger, J.P. et al. Broad specificity profiling of TALENs results in engineered nucleases with improved DNA-cleavage specificity. Nat. Methods 11, 429-435 (2014).

Supplementary information

Rights and permissions

About this article

Cite this article

Miller, J., Zhang, L., Xia, D. et al. Improved specificity of TALE-based genome editing using an expanded RVD repertoire.Nat Methods 12, 465–471 (2015). https://doi.org/10.1038/nmeth.3330

Download citation

• Received: 08 May 2014
• Accepted: 30 January 2015
• Published: 23 March 2015
• Issue Date: May 2015
• DOI: https://doi.org/10.1038/nmeth.3330