Engineered crispr-cas9 nucleases with altered pam specificity

Abstract

An engineered CRISPR-Cas9 nuclease having modified and improved PAM specificity, and its use in genome engineering, epigenome engineering and genome targeting are disclosed.

Description

Technology Field

[0001] Claim of priority

[0002] This application claims priority to U.S. provisional patent application serial number 61 / 127,634 filed March 3, 2015; 62 / 165,517 filed May 22, 2015; 62 / 239,737 filed October 9, 2015; and 62 / 258,402 filed November 20, 2015. The full contents of the above documents are incorporated herein by reference.

[0003] federally funded research or development

[0004] The present invention was made with government support under approval numbers DP1 GM105378, NIH R01 GM107427, and R01 GM088040 granted by the National Institutes of Health. The government holds specific rights to the present invention.

[0005] Technology field

[0006] At least in part, the present invention relates to an engineered clustered regularly interspaced short palindromic repeats (CRISPR) / CRISPR-associated protein 9 (Cas9) nuclease having modified and improved protospacer adjacent motif (PAM) specificity, and its use in genome engineering, epigenome engineering and genome targeting. Background Technology

[0007] CRISPR-Cas9 nucleases enable efficient, customized genome editing across a wide range of organisms and cell types (Sander & Joung, Nat Biotechnol 32, 347-355 (2014); Hsu et al., Cell 157, 1262-1278 (2014); Doudna & Charpentier, Science 346, 1258096 (2014); Barrangou & May, Expert Opin Biol Ther 15, 311-314 (2015)). Target site recognition by Cas9 is directed by two short RNAs known as crRNA and tracrRNA (Deltcheva et al., Nature 471, 602-607 (2011); Jinek et al., Science 337, 816-821 (2012)), which are chimeric single guide RNAs ( sgRNAIt can be fused to (Jinek et al., Science 337, 816-821 (2012); Jinek et al., Elife 2, e00471 (2013); Mali et al., Science 339, 823-826 (2013); Cong et al., Science 339, 819-823 (2013)). The 5' end of sgRNA (derived from crRNA) can form base pairs with target DNA sites, thereby allowing direct reprogramming of site-specific cleavage by the Cas9 / sgRNA complex (Jinek et al., Science 337, 816-821 (2012)). However, Cas9 must also recognize specific protospacer adjacent motifs (PAMs) located near DNA that form base pairs with sgRNA (Mojica et al., Microbiology 155, 733-740 (2009); Shah et al., RNA Biol 10, 891-899 (2013); Jinek et al., Science 337, 816-821 (2012); Sapranauskas et al., Nucleic Acids Res 39, 9275-9282 (2011); Horvath et al., J Bacteriol 190, 1401-1412 (2008)), and while this is a requirement for initiating sequence-specific recognition (Sternberg et al., Nature 507, 62-67 (2014)), the targeting of these nucleases for genome editing The scope can be limited. Widely used Streptococcus pyogenes Cas9 ( SpCas9) recognizes short NGG PAMs that occur once every 8 bp of random DNA sequences (Jinek et al., Science 337, 816-821 (2012); Jiang et al., Nat Biotechnol 31, 233-239 (2013)). In contrast, other Cas9 orthologues characterized to date can recognize longer PAMs (Horvath et al., J Bacteriol 190, 1401-1412 (2008); Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014); Esvelt et al., Nat Methods 10, 1116-1121 (2013); Ran et al., Nature 520, 186-191 (2015); Zhang et al., Mol Cell 50, 488-503 (2013)). For example, Staphylococcus aureus, one of several small Cas9 orthologues more suitable for viral transmission ( Staphylococcus aureus ) Cas9 ( SaCas9) (Horvath et al., J Bacteriol 190, 1401-1412 (2008); Ran et al., Nature 520, 186-191 (2015); Zhang et al., Mol Cell 50, 488-503 (2013)) recognizes the longer NNGRRT (SEQ ID NO:46) PAM, which is expected to occur once every 32 bp of random DNA. It is important to broaden the targeting range of Cas9 heterologous genes for various applications, including modification of small gene elements (e.g., transcription factor binding sites (Canver et al. Nature.;527(7577):192-7 (2015); Vierstra et al., Nat Methods. 12(10):927-30 (2015)) or the performance of allele-specific alterations by locating sequence differences within PAM (Courtney, DG et al. Gene Ther. 23(1):108-12 (2015)).

[0008] As described herein, the commonly used Streptococcus pyogenes ( Streptococcus pyogenes ) Cas9 ( SpCas9 In addition to ) Staphilococcus aureus Cas9 ( SaCas9) was engineered to recognize novel PAM sequences using structural information, bacterial selection-directed evolution, and combinatorial designs. These modified PAM-specific variants enable potent editing of endogenous gene regions in zebrafish and human cells that cannot be efficiently targeted by wild-type SpCas9 or SaCas9. Additionally, the inventors identified and characterized another SpCas9 variant exhibiting improved PAM specificity in human cells, which has reduced activity toward regions containing non-canonical NAG and NGA PAMs. Furthermore, the inventors [identified] two small Cas9 heterologous genes, Streptococcus thermophilus ( Streptococcus thermophilus ) Cas9 ( St1Cas9 ) and Staphilococcus aureus Cas9 ( SaCas9 It was discovered that ) functioned efficiently in the inventors' bacterial selection system and in human cells, which suggests that the inventors' manipulation strategy can be extended to Cas9 from other species. The inventors' discovery provides widely useful SpCas9 and SaCas9 variants, which are collectively referred to herein as “variants” or “the said variants”.

[0009] In a first aspect, the present invention relates to an isolated Streptococcus pyogenes Cas9 having a mutation at one or more of the following positions: G1104, S1109, L1111, D1135, S1136, G1218, N1317, R1335, T1337, comprising a sequence that is, for example, at least 80% identical to the amino acid sequence of sequence identification number: 1 ( SpCas9) provides a protein. In some embodiments, the variant SpCas9 protein comprises one or more of the following mutations: G1104K; S1109T; L1111H; D1135V; D1135E; D1135N; D1135Y; S1136N; G1218R; N1317K; R1335E; R1335Q and T1337R. In some embodiments, the variant SpCas9 protein comprises the following mutations: D1135; D1135V / R1335Q / T1337R (VQR variant); D1135E / R1335Q / T1337R (EQR variant); D1135V / G1218 / R1335Q / T1337R (VRQR variant); D1135N / G1218R / R1335Q / T1337R (NRQR variant); D1135Y / G1218R / R1335Q / T1337R (YRQR variant); G1104K / D1135V / G1218R / R1335Q / T1337R (KVRQR variant); S1109T / D1135V / G1218R / R1335Q / T1337R (TVRQR variant); L1111H / D1135V / G1218R / R1335Q / T1337R (HVRQR variant); D1135V / S1136N / G1218R / R1335Q / T1337R (VNRQR variant); Includes D1135V / G1218R / N1317K / R1335Q / T1337R (VRKQR variant); or D1135V / G1218R / R1335E / T1337R (VRER variant).

[0010] In some embodiments, the variant SpCas9 protein comprises one or more mutations that reduce nuclease activity, selected from the group consisting of mutations at D10, E762, D839, H983, or D986; and mutations at H840 or N863.

[0011] In some embodiments, the mutations are (i) D10A or D10N, and (ii) H840A, H840N or H840Y.

[0012] Additionally, the present invention relates to isolated Staphylococcus aureus Cas9 having a mutation at one or more of the following positions: E782, N968 and / or R1015, comprising a sequence that is at least 80% identical to the amino acid sequence of, for example, SEQ ID No. 2 ( SaCas9 ) protein is provided. Additionally, the present invention provides isolated Staphylococcus aureus Cas9 having a mutation at one, two or more of the following locations: E735, E782, K929, N968, A1021, K1044 and / or R1015 ( SaCas9 ) A protein is provided. In some embodiments, the variant SaCas9 protein comprises one or more of the following mutations: R1015Q, R1015H, E782K, N968K, E735K, K929R, A1021T, K1044N. In some embodiments, the variant SaCas9 protein comprises one or more mutations that reduce nuclease activity, selected from the group consisting of mutations in D10, D556, H557, and / or N580.

[0013] In some embodiments, the variant SaCas9 protein includes mutations in D10A, D556A, H557A, N580A, e.g., D10A / H557A and / or D10A / D556A / H557A / N580A.

[0014] The SpCas9 variants described herein may comprise the amino acid sequence of sequence identification number: 1 having mutations at one or more of the following positions: D1135, G1218, R1335, and T1337. In some embodiments, the SpCas9 variants may comprise one or more of the following mutations: D1135V; D1135E; G1218R; R1335E; R1335Q and T1337R. In some embodiments, the SpCas9 variants may comprise the following sets of mutations: D1135V / R1335Q / T1337R (VQR variant); D1135V / G1218R / R1335Q / T1337R (VRQR variant); D1135E / R1335Q / T1337R (EQR variant); Or it may include one of D1135V / G1218R / R1335E / T1337R (VRER variant).

[0015] The SaCas9 variants described herein may comprise the amino acid sequence of sequence identification number: 2 having mutations at one or more of the following positions: E735, E782, K929, N968, R1015, A1021, and / or K1044. In some embodiments, the SaCas9 variants may comprise one or more of the following mutations: R1015Q, R1015H, E782K, N968K, E735K, K929R, A1021T, K1044N. In some embodiments, the SaCas9 variants may comprise the following sets of mutations: E782K / N968K / R1015H (KKH variant); E782K / K929R / R1015H (KRH variant); Or it may include one of E782K / K929R / N968K / R1015H (KRKH variant).

[0016] Additionally, the present invention provides a fusion protein comprising the isolated variant SaCas9 or SpCas9 protein described herein fused to a heterofunctional domain with an optional intervening linker, wherein the linker does not interfere with the activity of the fusion protein. In some embodiments, the heterofunctional domain is a transcriptional activation domain. In some embodiments, the transcriptional activation domain is from VP64 or NF-κB p65. In some embodiments, the heterofunctional domain is a transcriptional silencer or a transcriptional inhibition domain. In some embodiments, the transcriptional inhibition domain is a Krüfel-associated box (KRAB) domain, an ERF inhibitor domain (ERD), or an mSin3A interaction domain (SID). In some embodiments, the transcriptional silencer is heterochromatin protein 1 (HP1), e.g., HP1α or HP1β. In some embodiments, the heterofunctional domain is an enzyme that modifies the methylation status of DNA. In some embodiments, the enzyme that modifies the methylation state of DNA is DNA methyltransferase (DNMT) or a TET protein. In some embodiments, the TET protein is TET1. In some embodiments, the heterofunctional domain is an enzyme that modifies histone subunits. In some embodiments, the enzyme that modifies histone subunits is histone acetyltransferase (HAT), histone deacetylase (HDAC), histone methyltransferase (HMT), or histone demethylase. In some embodiments, the heterofunctional domain is a biological tether. In some embodiments, the biological tether is MS2, Csy4, or lambda N protein. In some embodiments, the heterofunctional domain is FokI.

[0017] Additionally, the present invention provides a vector comprising an isolated nucleic acid encoding the variant SaCas9 or SpCas9 protein described herein, as well as an isolated nucleic acid optionally operably linked to one or more regulatory domains to express the variant SaCas9 or SpCas9 protein described herein. Additionally, the present invention provides a host cell, e.g., a mammalian host cell, comprising the nucleic acid described herein and optionally expressing the variant SaCas9 or SpCas9 protein described herein.

[0018] In addition, the present invention provides a method for modifying the genome of a cell by expressing in a cell the isolated variant SaCas9 or SpCas9 protein described herein and a guide RNA having a region complementary to a selected portion of the cell's genome.

[0019] In addition, the present invention provides a method for modifying the genome of a cell, for example, selectively, by expressing a variant protein and a guide RNA having a region complementary to a selected part of the cell's genome in the cell.

[0020] In addition, a method for selectively altering the genome of a cell is provided by contacting the cell with a protein variant described herein and a guide RNA having a region complementary to a selected portion of the cell's genome.

[0021] In some embodiments, the isolated protein or fusion protein comprises one or more of a nuclear localization sequence, a cell penetration peptide sequence, and / or an affinity tag.

[0022] In some embodiments of the method described herein, the cell is a stem cell, e.g., an embryonic stem cell, a mesenchymal stem cell, or an induced pluripotent stem cell; is in a living animal; or is in an embryo, e.g., a mammalian, insect, or fish (e.g., zebrafish) embryo or embryonic cell.

[0023] Additionally, the present invention provides a method for modifying a double-stranded DNA (dsDNA) molecule, for example, an in vitro method. The method comprises contacting the dsDNA molecule with one or more variant proteins described herein and a guide RNA having a region complementary to a selected portion of the dsDNA molecule.

[0024] Unless otherwise defined, all technical and scientific terms have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The methods and materials used in this invention are described herein, and other suitable methods and materials known in the relevant art may also be used. The materials, methods, and embodiments are merely illustrative and are not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In the event of any dispute, this specification, including definitions, shall prevail.

[0025] Other features and advantages of the present invention will be apparent from the following detailed description, drawings, and claims. Brief explanation of the drawing

[0026] This patent or application file includes at least one color drawing. A copy of this patent or patent application publication containing the color drawing(s) will be provided to the Patent Office upon request, provided the necessary fee is paid. Figs. 1A-J | Evolution and Characterization of SpCas9 Variants with Modified PAM Specificity . a Reasonable mutations in SpCas9 residues that induce base-specific contact with PAM bases are insufficient to alter PAM specificity in the enhanced green fluorescent protein (EGFP) disruption assay in U2OS human cells. The disruption frequency was quantified by flow cytometry, and this and subsequent panels ( c , g , h , and jThe average level of destruction observed in the background control group in ) is It is represented by a red dashed line, and the error bars represent sem, n = 3. b Schematic diagram of the 2-plasmid positive selection assay used to modify the PAM specificity of SpCas9. Splitting of the target site within the positive selection plasmid by the functional Cas9 / sgRNA complex is necessary for survival when bacteria are placed in the selective medium ( Fig. 12A-B Also refer to). c Combination assembly and testing of mutants obtained from positive selection for SpCas9 variants capable of cleaving target sites containing NGA PAM. SpCas9 variants were paired with sgRNAs targeting sites containing NGG or NGA PAM, and activity was evaluated using the EGFP disruption assay. Error bars represent sem, n = 3. d , schematic diagram of a negative selection assay in which apoptosis occurs via the cleavage of a selected plasmid when bacteria are placed in a selective medium. This system was adapted to profile the PAM specificity of Cas9 by generating a library of plasmids containing randomized sequences adjacent to the 3' end of the protospacer ( Fig. 13b Also refer to). e , scatter plot of PAM depletion values ​​(PPDV) after selection of wild-type SpCas9 and two randomized PAM libraries (each having different protospacers). PAMs were plotted by classifying them by their 2nd / 3rd / 4th positions. The red dashed line indicates the cutoff for statistically significant depletion (obtained from the dCas9 control experiment, Fig. 13c Note), the gray dashed line indicates 5-fold depletion (0.2 PPDV). f , PPDV scatter plot for VQR and EQR SpCas9 variants recognizing PAM distinct from that recognized by wild-type SpCas9. gEGFP disruption frequencies of wild-type, VQR, and EQR SpCas9 for sites containing , NGAN, and NGNG PAM. Error bars represent sem, n = 3. h Combination assembly and testing of mutations obtained from positive selection for SpCas9 variants capable of cleaving target sites containing NGCG PAM. NGGG or NGCG sgRNAs targeting PAM-containing sites were evaluated for Cas9 targeting using the EGFP disruption assay. Error bars represent sem, n = 3. i , PPDV scatter plot for VRER variants. j EGFP destruction frequency of wild-type and VRER SpCas9 for sites containing , NGCN and NGNG PAM. Error bars represent sem, n = 3. Fig. 2 | SpCas9 variants with evolved PAM specificity strongly modify endogenous sites in zebrafish embryos and human cells . a Quantification of mutagenicity in wild-type or VQR SpCas9-induced zebrafish embryos for endogenous gene regions possessing NGAG PAM. Mutation frequency was determined using the T7E1 test, and error bars represent sem, n = 5 to 9 individual embryos. b Mutation frequencies of VQR variants quantified by T7E1 assay at 16 target sites in four endogenous human genes with sgRNAs targeted to sites containing NGAG, NGAT, and NGAA PAM. Error bars represent sem, n = 3. c , mutation frequencies of wild-type SpCas9 against endogenous human gene target sites containing NGA PAM. For convenience of comparison, mutation frequencies for VQR variants using the same sgRNA are presented here again (panel b Same data as presented in ). Error bars represent sem, n = 3, and nd Is It is undetectable by T7E1. d Mutation frequencies of wild-type, VRER, and VQR SpCas9 at 9 target sites containing NGCG PAM in 3 endogenous human genes quantified by the T7E1 assay. sgRNA complementary lengths of 19 and 20 nt were used, and error bars represent sem, n = 3. e , indication of the number of sites with 20 nt spacers targetable by wild-type, VQR, and VRER SpCas9 in the human genome. f , panel b and d Number of off-target cleavage sites identified by GUIDE-seq for VQR and VRER SpCas9 variants using sgRNA from. Fig. 3 | The D1135E mutation improves PAM recognition and spacer specificity of SpCas9 . a , PPDV scatter plots of wild-type and D1135E SpCas9 for two randomized PAM libraries (left and right panels, respectively). PAMs were plotted classified by their 2nd / 3rd / 4th positions. The data presented for wild-type SpCas9 are Fig. 1d It is identical to the plot and is presented here again for convenience. The red dashed line indicates statistically significantly depleted PAM ( Fig. 13c Note), the gray dashed line indicates the 5-fold depletion cutoff (PPDV of 0.2). b , EGFP disruption activity of wild-type and D1135E SpCas9 on sites containing NGG, NAG, and NGA PAM in human cells. Disruption frequency was quantified by flow cytometry, and this panel and ( d In ), the average level of destruction observed by the background control was shown as a red dashed line, and error bars represent sem, n = 3, and the average multiple change in activity was presented. cMutagenicity detected by T7E1 for wild-type and D1135E SpCas9 in 6 endogenous sites in human cells. Error bars represent sem, n = 3, and show the mean multiple change in activity. d , titration of the amount of wild-type or D1135E SpCas9-coding plasmid transfected for EGFP disruption experiments in human cells. The amount of sgRNA plasmid used in all these experiments was fixed at 250 ng. Two sgRNAs targeting different EGFP sites were used, and error bars represent sem, n = 3. e Targeted deep sequencing of on- and off-target sites for three sgRNAs using wild-type and D1135E SpCas9. On-target sites are presented at the top, and off-target sites are listed below, highlighting mismatches with the on-target sites. In the off-target sites, the fold reduction in activity by D1135E compared to wild-type SpCas9 was highlighted in green, and control indel levels for each amplicon were recorded. f Summary of targeted deep-sequencing data plotting the fold reduction in activity at on- and off-target sites using D1135E compared to the indel frequency observed by wild-type SpCas9. g , change in specificity between wild-type and D1135E detected by GUIDE-seq in off-target sites, plotted by comparing the normalized fold-change in specificity using D1135E with the read counts in off-target sites using wild-type SpCas9 ( Fig. 18c See also). The estimated multiple-increase of specificity in regions without read counts for D1135E was not plotted ( Fig. 18c reference). Fig. 4 |Characterization of St1Cas9 and SaCas9 heterologous genes in bacteria and human cells. a , PPDV scatter plot for St1Cas9 using two randomized PAM libraries. PAMs were plotted and classified by their 3rd / 4th / 5th / 6th positions. St1Cas9 was programmed for both libraries using sgRNA complementary lengths of 20 and 21 nucleotides (left and right panels, respectively). The red dashed line indicates statistically significantly depleted PAM ( Fig. 13c Note), the gray dashed line indicates 5-fold depletion (PPDV of 0.2), α, PAM previously confirmed by a bioinformatics approach 27 ; β, previously identified PAM under strict experimental conditions 20 ; *, Novel PAM discovered in this study; γ, PAM previously identified under moderate experimental conditions 20 . b , PPDV scatter plot for SaCas9 using two randomized PAM libraries. PAMs were plotted by sorting them by their 3rd / 4th / 5th / 6th positions. SaCas9 was programmed for both libraries using sgRNA complementary lengths of 21 and 23 nucleotides (left and right panels, respectively). The identified PAMs for SaCas9 are presented, and PAMs 1–3 were consistently depleted across all combinations of spacers and spacer lengths used in these experiments. c , survival percentage of St1Cas9 and SaCas9 in bacterial-positive selection when tried with select plasmids possessing different target sites and PAMs shown on the x-axis. Panel for St1Cas9 and SaCas9 ( a ) and ( b Highly depleted PAM from ) was used for the target site in a positive selector plasmid. d , e, St1Cas9 (panel) for EGFP sites containing NNAGAA (sequence ID: 3) or NNGGGT (sequence ID: 4) / NNGAGT (sequence ID: 5) PAM, respectively d ) or SaCas9 (panel e EGFP disruption activity of ). Matched sgRNAs of different lengths for the same site were shown, disruption frequency was quantified by flow cytometry, and the average frequency of EGFP disruption obtained using a negative control is shown as a red dashed line, and error bars represent sem, n = 3. f , g , St1Cas9 (panel) quantified by T7E1 assay at regions in four endogenous human genes containing NNAGAA (Sequence ID: 3) or NNGGGT (Sequence ID: 4) / NNGAGT (Sequence ID: 5) / NNGAAT (Sequence ID: 6) PAM, respectively f ) and SaCas9 (panel g Mutation frequency of ). Error bars represent sem, n = 3, and nd is undetectable by T7E1. Figs. 5A-J . Sequences and Maps - Plasmids Used in This Study Fig. 6 | Alignment of Cas9 heterologous genes to predict PAM-interaction residues of SaCas9 The PAM-interaction domains of SpCas9, SaCas9, and 11 other Cas9 heterologs were aligned to identify PAM-containing residues in SaCas9 based on what is known for SpCas9. Top, top, S. pyogenes, amino acids 1229-1368 of sequence identification number:1, followed by sequence identification number:29-40. Fig. 7 | Substitutions in SaCas9 evaluated for activity against different PAMs on a bacterial screen . Fig. 6Based on alignment, single amino acid substitutions were tested in bacterial-positive selection to screen for the effects on activity against standard NNGAGT (Sequence ID: 5) and non-standard NNAAGT (Sequence ID: 41) and NNAGGT (Sequence ID: 42) PAMs. Bacterial colonies on the selected medium suggest that SaCas9 variants are active against sites containing the specified PAMs. Fig. 8A-B | Summary of amino acid substitutions enabling SaCas9 variants to target NNARRT (Sequence ID: 43) PAM . NNARRT (Sequence ID: 43) Amino acid sequence of the PAM-interaction domain of 52 selected mutant SaCas9 clones that enable survival in bacteria for the PAM-containing site; the presented sequence is a partial sequence of Sequence ID: 53-104 presented in Table 6. Fig. 9 | Human cell activity of wild-type and engineered SaCas9 variants In a human cell EGFP reporter assay, the activity of wild-type, KKQ, and KKH SaCas9 was evaluated for the site containing NNRRRT (sequence identification number: 45) PAM. Fig. 10. SaCas9 activity against non-standard PAM in bacteria, and how a specified mutation in R1015 affects activity against the same non-standard PAM. Fig. 11. The engineered variant can recognize the PAM of the NNNRRT form. Fig. 12A-B | Bacteria-based positive selection used to engineer modified PAM-specific variants of SpCas9 . a , Fig. 1b Extended schematic diagram of positive selection from (left panel), and demonstration that SpCas9 behaves as expected in positive selection (right panel). Spacer 1, sequence identification number: 105; Spacer 2, sequence identification number: 106. b Schematic diagram of a method for tuning positive selection to select SpCas9 variants with modified PAM recognition specificity. A library of SpCas9 clones with randomized PAM-interaction (PI) domains (residues 1097-1368) was tested with a selection plasmid containing modified PAM. By dividing the positive selection plasmid, selection-surviving SpCas9 variants were sequenced to determine the mutations enabling modified PAM specificity. Fig. 13A-D | Bacterial cell-site-depletion assay for profiling the overall PAM specificity of Cas9 nucleasea, an extended schematic diagram illustrating negative selection from Fig. 1d (left panel), and demonstrating that wild-type SpCas9 behaves as expected in the screening of sites containing functional (NGG) and non-functional (NGA) PAMs (right panel). b, a schematic diagram of a method using negative selection as a site-depletion assay to screen for functional PAMs by constructing a negative selection plasmid library containing six randomized base pairs instead of PAMs. The selection plasmid containing PAMs cleaved by the Cas9 / sgRNA of interest was depleted, while uncleaved (or poorly cleaved) PAMs were retained. The post-selection PAM depletion value (PPDV) was calculated by comparing the frequency of PAMs after selection to their pre-selection frequency in the starting library. Spacer 1, Sequence ID: 105; Spacer 2, Sequence ID: 106. c, d, a cutoff for statistically significant PPDV was established by plotting the PPDV of PAM against catalytically inactive SpCas9 (dCas9) (categorized and plotted by its 2nd / 3rd / 4th positions) for two randomized PAM libraries (c). For the two libraries, the titer of 3.36 standard deviations from the mean PPDV was calculated (red line in (d)), and it was established that any PPDV deviation of less than 0.85 was statistically significant compared to the dCas9 treatment (red dashed line in (c)). In (c), the gray dashed line indicates a 5-fold depletion in the test (PPDV of 0.2). Fig. 14 | Concordance between site-depletion assay and EGFP destructive activityData points represent the average EGFP destruction of two NGAN and NGNG PAM sites in VQR and EQR SpCas9 variants (Fig. 1g), plotted against the average PPDV observed in Libraries 1 and 2 for the corresponding PAMs (Fig. 1f). The red dashed line indicates statistically significant depletion of PAM (PPDV of 0.85, see Fig. 13c), and the gray dashed line indicates 5-fold depletion (PPDV of 0.2). Mean values ​​were plotted with 95% confidence intervals. Fig. 15 | VQR SpCas9 in endogenous zebrafish sites containing NGAG PAM Insertion or deletion mutations induced by variants For each target locus, the wild-type sequence is presented at the top with a yellow-highlighted protospacer (highlighted in green if present on the complementary strand), and the PAM is indicated by red underlined text. Deletions are indicated by red dashed lines highlighted in gray, and insertions are depicted in lowercase letters highlighted in blue. The net change in length caused by each indel mutation is presented to the right (+, insertion; -, deletion). Some alterations involved both insertions and deletions of the sequence; in these cases, the alterations are listed in parentheses. The retrieval multiple (if greater than 1) of each mutant allele is presented in parentheses. Fig. 16A-B | Endogenous genes targeted by wild-type and evolved variants of SpCas9 a, Sequences targeted by wild-type, VQR, and VRER SpCas9 are presented in blue, red, and green, respectively. The sequences of the sgRNA and primers used to amplify these loci for T7E1 are provided in Tables 1 and 2 below. b, Mean mutagenicity detected by T7E1 against wild-type SpCas9 at eight target sites harboring NGG PAM in four different endogenous human genes (corresponding to the annotations in the top panel). Error bars indicate sem, n = 3. Fig. 17A-B | Specificity profiles of VQR and VRER SpCas9 variants determined using GUIDE-seqIntended on-target sites are indicated by black squares, and mismatched locations within off-target sites are highlighted. a, The specificity of VQR variants was evaluated by targeting endogenous sites containing NGA PAM in human cells: EMX1 site 4 (Sequence ID: 142), FANCF site 1 (Sequence ID: 143), FANCF site 3 (Sequence ID: 144), FANCF site 4 (Sequence ID: 145), RUNX1 site 1 (Sequence ID: 146), RUNX1 site 3 (Sequence ID: 147), VEGFA site 1 (Sequence ID: 148), and ZSCAN2 (Sequence ID: 149). b, The specificity of the VRER variant was evaluated by targeting the NGCG PAM-containing endogenous sites in human cells: FANCF site 3 (sequence identification number: 150), FANCF site 4 (sequence identification number: 151), RUNX1 site 1 (sequence identification number: 152), VEGFA site 1 (sequence identification number: 153), and VEGFA site 2 (sequence identification number: 154). Fig. 18A-C | Difference in activity between D1135E and wild-type SpCas9 at off-target sites detected by GUIDE-seqa, Mean frequency of oligo-tag integration at on-target sites estimated by restriction fragment length polymorphism analysis. Error bars indicate sem, n = 4. b, Mean mutagenic frequency at on-target sites detected by T7E1. Error bars indicate sem, n = 4. c, GUIDE-seq read-count difference between wild-type SpCas9 and D1135E at three endogenous human cell sites (EMX1 site 3 (SEQ ID: 155); ZNF629 site (SEQ ID: 156), VEGFA site 3 (SEQ ID: 157)). On-target sites are presented at the top, and off-target sites are listed below with mismatches highlighted. In the table, the ratio of off-target activity to on-target activity was compared between the wild type and D1135E to calculate the normalized fold-change in specificity (increase in specificity is highlighted in green). For regions without detectable GUIDE-seq reads, a value of 1 was assigned to calculate the estimated change in specificity (indicated in orange). In Figure 3e, off-target regions analyzed by deep sequencing were numbered to the left of the EMX1 region 3 and VEGFA region 3 off-target regions. Figs. 19A-F | Additional PAM for St1Cas9 and SaCas9, and spacer length-based activity in human cells. a, Scatter plot of PPDV for St1Cas9 comparing 20 and 21 nucleotide sgRNA complementary lengths obtained using PAM libraries randomized for spacer 1 (top panel) or spacer 2 (bottom panel). PAMs were plotted by classifying them by their 3rd / 4th / 5th / 6th positions. The red dashed line indicates statistically significantly depleted PAMs (see Fig. 13c), and the gray dashed line indicates 5-fold depletion (PPDV of 0.2). b, Table of PAMs with PPDV less than 0.2 for St1Cas9 under each of the four tested conditions. The PAM numbers presented on the left are the same as in Fig. 4a. c, Scatter plot of PPDV for SaCas9 comparing 21 and 23 nucleotide sgRNA complementary lengths obtained using PAM libraries randomized for Spacer 1 (top panel) or Spacer 2 (bottom panel). PAMs were plotted sorted by their 3rd / 4th / 5th / 6th positions. The red and gray dashed lines are the same as in (a). d, Table of PAMs with PPDV < 0.2 for SaCas9 under each of the four tested conditions. PAM numbers are the same as in Fig. 4b. e, f, Human cellular activity of St1Cas9 and SaCas9 across various spacer lengths via EGFP disruption (Panel e, data from Fig. 4d, 4e) and endogenous gene mutagenesis detected by T7E1 (Panel f, data from Fig. 4f, 4g). The activity of all replicates was presented (n = 3 or 4); The bars represent the mean and 95% confidence interval; and indicate the number of regions per spacer length. Fig. 20A-B | Structural and functional roles of D1135, G1218, and T1337 in PAM recognition by SpCas9a, Structural representation of six residues associated with PAM recognition. The left panel illustrates the proximity of residues S1136 and D1135, which form water-mediated small groove contacts, to the third base position of PAM15. The right panel illustrates the proximity of residues R1335, G1218, E1219, and T1337, which form direct base-specific large groove contacts, to the third base position of PAM15. Angstrom distances are indicated by yellow dashed lines, non-target strand guanine bases dG2 and dG3 of PAM are presented in blue, other DNA bases in orange, and water molecules in red; images were generated from PDB:4UN3 using PyMOL. b, Mutation analysis of six residues of SpCas9 associated with PAM recognition. Clones containing one of three mutation types at each location were tested for EGFP disruption using two sgRNAs targeted to the NGG PAM-bearing site. For each location, the inventors generated alanine substitutions and two non-conservative mutations. S1136 and R1335 have been previously reported to mediate contact of PAM15 with the tertiary guanine, while D1135, G1218, E1219, and T1337 are reported in this study. EGFP disruption activity was quantified by flow cytometry, the background control is indicated by the red dashed line, and error bars represent sem, n = 3. Figs. 21A-F Selection and assembly of SaCas9 variants with modified PAM specificity. a ) Phylogenetic tree of Cas9 heterologous genes highlighting SpCas9 and SaCas9. b ) Activity of SaCas9 variants with a single amino acid substitution evaluated in a bacterial positive selection assay ( Fig. 31b See also). Error bars represent sem, n = 3, and NS = not survived. c) Human cell activity of wild-type and R1015H SaCas9. EGFP degrading activity was quantified by flow cytometry, error bars represent SEM, n = 3, and the mean level of background EGFP loss is indicated by the red dashed line (this panel and panel e (in the case of). d ) Total number of substitutions observed at each amino acid position when selecting SaCas9 variants with altered PAM specificity. Starter mutations in R1015 were not counted. e ) Human cell EGFP destructive activity of variants containing mutations observed when selected for altered PAM specificity. f ) Scatter plot of mean PAM depletion values ​​(PPDV) of wild-type SaCas9 for KKH variants (n = 2, Fig. 34c (See also). Using two libraries with different protospacers and 8 randomized base pairs instead of PAM, we determined whether PAM was targetable by each Cas9. Statistically significant depletion is indicated by a red dashed line (compared to the dCas9 control, Figs. 34a and 34b Note), 5-fold depletion is indicated by a gray dotted line. Fig. 22A-F. Activity of the SaCas9 KKH variant targeted to endogenous sites in human cells. a ) Mutagenicity of 55 different sites containing KKH SaCas9-induced NNNRRT PAM determined by T7E1 test. Error bars indicate sem, n = 3, and ND indicates undetectable by T7E1 test. ( b ) KKH variant preference for the third position of PAM. Panel a The average activity from the data in this panel and Panel b and c Presented in. c ) KKH variant preference for the 4th and 5th positions of PAM. d) Spacer length preference of the KKH SaCas9 variant. e Comparison of human cell EGFP disruption activity of wild-type and KKH SaCas9 targeted to various sites containing NNNRRT PAM. EGFP disruption was quantified by flow cytometry, error bars represent SEM, n = 3, and the mean level of background EGFP loss is indicated by the red dashed line. f ) Panel a Mutagenic frequency of wild-type SaCas9 at one site for each of the 16 possible NNNRRT sites (the site with the highest KKH activity was selected). Error bars indicate sem, n = 3, and ND indicates undetectable by T7E1 test. Figs. 23A-E Genome-wide specificity profiles of wild-type and KKH SaCas9. a ) and ( b ) Total number of off-targets at EMX site 1 (Sequence ID: 158) and VEGF site 8 (Sequence ID: 159) Panel a ) and mismatch observed at each off-target site ( Panel b Direct comparison of wild-type and KKH SaCas9 targeted to the NNGRRT (sequence identification number: 46) PAM-containing site, represented by ). Panel b and e For each region, GUIDE-seq read counts were indicated, on-target sequences were marked with black boxes, mismatched locations within off-target regions were highlighted, sequences were corrected for cell-type specific SNPs, and regions containing potential sgRNA or DNA bulge nucleotides were indicated with small red-bordered bases or dotted lines, respectively. c ) Venn diagram highlighting the difference in off-target site splitting by wild-type and KKH SaCas9 at VEGFA site 8. d ) and ( e ) Total number of off-targets ( Panel d ) and mismatch observed at each off-target site ( Panel e Specificity profile of a KKH variant targeted to EMX site 1 (Sequence Identification No.: 160), EMX site 4 (Sequence Identification No.: 161), EMX site 10 (Sequence Identification No.: 162), FANCF site 9 (Sequence Identification No.: 163), and FANCF site 16 (Sequence Identification No.: 164), which are regions having NNHRRT (Sequence Identification No.: 44) PAM, represented by ). Fig. 24: Activity of VQR-derivative clones in a bacterial 2-plasmid screen . Test of 24 different VQR derivative variants on bacterial sites containing NGAN PAM. Survival on selective plates compared to non-selective plates is an indicator of activity against the specified PAM. Fig. 25: Human cell EGFP disruption activity of SpCas9-VQR derivatives. The EGFP disruption activity of SpCas9 variants is a measure of activity toward sites containing a specified PAM. Fig. 26 : Human cell EGFP disruption activity of SpCas9-VQR and -VRQR variants. The EGFP disruption activity of SpCas9 variants is a measure of activity toward sites containing a specified PAM. Fig. 27 : Activity of SpCas9-VRQR derivative variants in a bacterial 2-plasmid screen. Testing of 12 different VQR derivative variants against bacterial sites containing NGAN PAM compared to VQR and VRQR variants. Survival on selective plates compared to non-selective plates is an indicator of activity against the specified PAM. Fig. 28 : Human cell EGFP disruption activity of SpCas9-VRQR variants. The EGFP disruption activity of SpCas9 variants is a measure of activity toward sites containing a specified PAM. Fig. 29 Cas9 heterologous gene ( Fig. 21a Protein domain alignment (from). The domain structure of SpCas9 is presented at the top (based on PDB:4UN3; Anders et al., 2014), highlighted the PAM contact residue of SpCas9 and presented a SaCas9 region mutated to select a modified PAM-specific variant. Fig. 30 Primary sequence alignment of Cas9 heterologous genes for the identification of PAM-interacting residues; sequence identification numbers: 165-176, respectively. SpCas9 residues previously identified as important for PAM contact (Anders et al ., 2014; Examples 1-2) are highlighted in blue, residues capable of modulating SaCas9 PAM specificity (identified in this study) are highlighted in orange, and positively charged residues adjacent to R1015 are highlighted in yellow. The structurally predicted PAM-interaction domain of SpCas9 is highlighted with a blue dashed line (based on PDB:4UN3; Anders et al ., 2014), the conservative estimate of the SaCas9 PAM-interaction domain used as a boundary for PCR mutagenesis is shown by the orange dotted line. Fig. 31A-B Schematic diagram of the bacterial positive selection test. a The selection plasmid can be modified to screen Cas9 variants capable of recognizing alternative PAM sequences. b Schematic diagram of the positive selection plasmid (left panel) and expected result (right panel) when screening functional or non-functional Cas9 / sgRNA pairs in positive selection. Fig. 32 Addition of K929R mutants as KNH and KKH variants. EGFP destructive activity was quantified by flow cytometry, error bars represent SEM, n = 3, and the average level of background EGFP loss is shown as a red dashed line. Fig. 33Schematic diagram of the bacterial site-depletion assay. Instead of wild-type or KKH SaCas9-refractory PAMs, a site-depletion plasmid with 8 randomized nucleotides was sequenced. Library 1 spacer sequence, sequence identification number: 105; Library 2 spacer sequence, sequence identification number: 106. Targetable PAMs were inferred by their depletion compared to the input library, calculated as the PAM depletion value (PPDV) after selection. Figs. 34A-E Site-depletion test results for wild-type and KKH SaCas9. ( a ) PPDV values ​​of the dCas9 control experiment for both libraries. The red dashed line indicates statistical significance (PPDV = 0.794, Panel b Note), the gray dashed line indicates 5-fold depletion, and PPDV was plotted for a window including the 3rd / 4th / 5th / 6th positions of PAM (this panel and Panel c (in the case of). b ) Statistically significant post-selection PAM depletion value (PPDV) Panel a It was determined from the dCas9 control group experiment. Statistical significance was determined by setting the titer to 3.36 times the standard deviation. c Comparison of PPDV for wild-type and KKH SaCas9 for each of the two libraries containing 8 randomized nucleotides instead of PAM. d ) and ( e ) PAM, and corresponding PPDV values ​​for all PAMs depleted more than 5-fold for wild-type and KKH SaCas9, respectively. Sequence motifs were presented for PAMs in two categories: 1) depletion more than 10-fold or 2) depletion between 5-fold and 10-fold. Figs. 35A-D Additional characteristics of endogenous sites targeted by KKH SaCas9. ( aActivity for each of the 55 endogenous site sgRNAs binned based on 16 possible NRR motifs of NNNRRT PAM. Fig. 2a The average activity from this panel and panel b and c Presented in. b ) and ( c ) Relationship between endogenous gene disruption activity and GC content of spacers and PAM, respectively. d Sequence logos for spacers and PAMs at target sites, binned based on ) activity. Sites have average mutation frequency ( Fig. 2a Based on (from), it was classified into low (0-10%, 17 sites), medium (10-30%, 17 sites), or high (>30%, 21 sites) activity. Fig. 36A-B On-target tag integration and mutagenicity for GUIDE-seq experiments. a ) Restricted Fragment Length Polymorphism (RFLP) analysis to determine the average GUIDE-seq tag integration frequency. Error bars represent sem, n = 3 (this panel and Panel b (in the case of). b ) Average mutagenicity was detected by the T7E1 test. Fig. 37A-B Terminated repeat: Anti-rep sgRNA outperforms full-length sgRNA, which is similar to previous results (Ran et al. , 2015). ( a ) Human cell EGFP disruption activity of wild-type SaCas9 against four sites containing NNGRRT (SEQ ID: 46) PAM. EGFP disruption activity was quantified by flow cytometry, error bars represent SEM, n = 3, and the mean level of background EGFP loss is indicated by the red dashed line (this panel and panel b (in the case of). b) Human cell EGFP disruption activity of KKH SaCas9 against 8 sites containing NNNRRT PAM. Specific details for implementing the invention

[0027] Although CRISPR-Cas9 nucleases are widely used for genome editing 1-4 The range of sequences capable of cleaving Cas9 is limited by the requirement for specific protospacer adjacent motifs (PAMs) at the target site. 5, 6 For example, SpCas9, the most powerful and widely used Cas9 to date, primarily recognizes NGG PAM. As a result, it can often be difficult to accurately target double-strand breaks (DSBs) required for various genome editing applications. Furthermore, incomplete PAM recognition by Cas9 can lead to the generation of unwanted off-target mutations. 7, 8 The ability to evolve Cas9 derivatives having modified or improved PAM specificity for the purpose would address these limitations, but to the knowledge of the inventors, no such Cas9 variant has ever been described.

[0028] A potential strategy to improve the targeting range of orthogonal Cas9 recognizing extended PAMs is to modify its PAM recognition specificity. As described herein, the PAM recognition specificity of SpCas9 can be modified by structure-directed design and direct evolution performed using a bacterial cell-based selection system, see Examples 1 and 2. Additionally, variants evolved to have relaxed or partially relaxed specificity for specific locations within the PAM are described herein, see Example 3. These variants extend the utility of Cas9 heterologous genes having specificity for longer PAM sequences.

[0029] Engineered Cas9 variant with modified PAM specificity

[0030] The engineered SpCas9 variants in this study significantly increase the sites accessible by wild-type SpCas9 and further improve the opportunities to utilize the CRISPR-Cas9 platform to perform efficient HDR, target NHEJ-mediated indels as small genetic elements, and develop the requirements for PAM to distinguish two different alleles within the same cell. The modified PAM-specific SpCas9 variants can efficiently disrupt endogenous genetic regions that are currently untargetable by SpCas9 in both zebrafish embryos and human cells, suggesting that they will function in various different cell types and organisms. Importantly, GUIDE-seq experiments indicate that the overall profiles of the VQR and VRER SpCas9 variants are similar to or better than those observed in wild-type SpCas9. Furthermore, the improved specificity D1135E variant identified and characterized by the inventors provides an excellent alternative to the widely used wild-type SpCas9. D1135E has activity similar to wild-type SpCas9 for regions containing standard NGG PAM, but reduces genome-wide splitting at off-target regions containing mismatched spacer sequences and standard or non-standard PAMs.

[0031] All SpCas9 and SaCas9 variants described herein can be rapidly introduced into existing widely used vectors, for example by simple site-directed mutagenesis, and because they require only a few mutations contained within the PAM-interaction domain, the variants are also applicable to SpCas9 platforms (e.g., truncated sgRNA (Tsai et al., Nat Biotechnol 33, 187-197 (2015); Fu et al., Nat Biotechnol 32, 279-284 (2014)), nikase mutations (Mali et al., Nat Biotechnol 31, 833-838 (2013); Ran et al., Cell 154, 1380-1389 (2013)), dimeric FokI-dCas9 fusions (Guilinger et al., Nat Biotechnol 32, 577-582 (2014); It should function with other improvements previously described for Tsai et al., Nat Biotechnol 32, 569-576 (2014)).

[0032] In addition to the mutation at R1335, which likely contacts the third PAM base position, the SpCas9 variants evolved in this study possess amino acid substitutions at D1135, G1218, and T1337, all of which are located near or adjacent to residues that make direct or indirect contact with the third PAM position in the SpCas9-PAM structure, but do not themselves mediate contact with the PAM base (Anders et al., Nature 513, 569-573 (2014)) Fig. 20a ( ). In accordance with this, the inventors confirmed that various mutations at these positions are not believed to affect SpCas9-mediated splitting of the NGG PAM-containing site ( Fig. 20bThese results, along with the characteristics of amino acid substitutions at G1218 and T1337 in VQR and VRER SpCas9 variants, suggest that changes at these two positions may be gain-of-function mutations. For example, the T1337R mutation is capable of forming a backbone or base-specific contact with or near the fourth position of PAM, particularly in the case of VRER variants. Although the mechanistic role of mutations at D1135 is less obvious, they may affect the activity of the adjacent S1136 residue, which is associated with forming a water-mediated contact with guanine through a small groove at the third position of PAM (Anders et al., Nature 513, 569-573 (2014)). The D1135E mutation can improve specificity by disrupting this network, possibly by reducing the overall interaction energy between the SpCas9 / gRNA complex and the target site, and the mechanism previously proposed by the inventors can reduce off-target effects by cleaving these unwanted sequences in an energetically less favorable manner (Fu et al., Nat Biotechnol 32, 279-284 (2014)).

[0033] These results establish the feasibility of manipulating Cas9 nucleases with clearly altered PAM specificity. As previously described, characterization of additional Cas9 heterologous genes (Esvelt et al., Nat Methods 10, 1116-1121 (2013); Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014)) or generation of domain-swapped Cas9 chimeras (Nishimasu et al., Cell. 156(5):935-49 (2014)) also provides potential methods for targeting different PAMs. The manipulation strategies described herein can also be performed using these heterologous genes or synthetic hybrid Cas9s to further diversify the range of targetable PAMs. St1Cas9 and SaCas9 create a particularly attractive framework for future manipulation efforts, as demonstrated by the inventors' demonstration of their smaller size and potent genome-editing activity compared to SpCas9 in the inventors' bacterial selection system and in human cells.

[0034] The inventors' results strongly suggest that R1015 of wild-type SaCas9 contacts a G at the third PAM position. While we do not wish to be constrained by theory, it is possible that R1015H substitutions eliminate this contact at the third position and mitigate specificity, but the loss of R1015 to a G contact may also reduce the energy associated with target site binding, which explains why the R1015H mutation alone is not sufficient for potent activity toward the NNNRRT site in human cells. Since both E782K and N968K substitutions add a positive charge, it is possible for them to interact non-specifically with the DNA phosphate backbone to energetically compensate for the loss of R1015 to a guanine contact.

[0035] The genetic approach described herein does not require structural information and should therefore be applicable to various different Cas9 heterologous genes. The only requirement for evolving Cas9 nucleases with expanded PAM specificity is that they function in bacterial-substance selection. Previous studies have demonstrated that PAM recognition can be modified by swapping the PAM-interaction domains of Cas9 heterologous genes highly associated with PAM recognition (Nishimasu et al., Cell (2014)), but it must be determined whether this strategy is generalizable or effective when using a wider variety of heterologous genes. In contrast, the evolutionary strategy described herein by the inventors can be used to manipulate PAM recognition specificity in addition to that encoded within naturally occurring Cas9 heterologous genes. By utilizing this overall strategy, the targeting range of numerous naturally occurring Cas9 heterologous genes can be expanded and their utility extended.

[0036] SpCas9 variants with altered specificity

[0037] Accordingly, the present invention provides spCas9 variants. The wild-type SpCas9 sequence is as follows:

[0038]

[0039]

[0040] The SpCas9 variants described herein may include mutations at one or more of the following positions: D1135, G1218, R1335, T1337 (or similar positions). In some embodiments, the SpCas9 variants include one or more of the following mutations: D1135V; D1135E; G1218R; R1335E; R1335Q; and T1337R. In some embodiments, the SpCas9 variants are at least 80% identical, e.g., at least 85%, 90%, or 95% identical to the amino acid sequence of SEQ ID NO:1, and differ, e.g., up to 5%, 10%, 15%, or 20% of the residues of SEQ ID NO:1 are replaced, e.g., by conservative mutations. In a preferred embodiment, the variant possesses the preferred activity of the parent, e.g., nuclease activity (except where the parent is nikase or dead Cas9) and / or the ability to interact with guide RNA and target DNA.

[0041] To determine the percentage identity of two nucleic acid sequences, the sequences are aligned for optimal comparison purposes (e.g., gaps may be introduced in one or both of the first and second amino acid or nucleic acid sequences for optimal alignment, and non-homologous sequences may be ignored for comparison purposes). The length of the reference sequence aligned for comparison purposes is at least 80% of the reference sequence length, and in some embodiments, at least 90% or 100%. Subsequently, nucleotides at corresponding amino acid positions or nucleotide positions are compared. If a position in the first sequence is occupied by the same nucleotide as the corresponding position in the second sequence, the molecules are identical at that position (as used herein, nucleic acid "identity" is equivalent to nucleic acid "homology"). The percentage identity between the two sequences is a function of the number of identical positions shared by the sequences, taking into account the number of gaps that need to be introduced for optimal alignment of the two sequences and the length of each gap. Percent identity between two polypeptide or nucleic acid sequences is determined in various ways within the art of the relevant technical field, for example, by publicly available computer software, such as Smith-Waterman alignment (Smith, TF and MS Waterman (1981) J Mol Biol 147:195-7); GeneMatcher Plus TMIt is determined using "BestFit" (Smith and Waterman, Advances in Applied Mathematics, 482-489 (1981)) introduced in the Atlas of Protein Sequence and Structure, Dayhof, MO, Ed, pp 353-358; BLAST programs (Basic Local Alignment Search Tool; Altschul, SF, W. Gish, et al. (1990) J Mol Biol 215: 403-10), BLAST-2, BLAST-P, BLAST-N, BLAST-X, WU-BLAST-2, ALIGN, ALIGN-2, CLUSTAL, or Megalign (DNASTAR) software. Additionally, a person skilled in the art can determine appropriate parameters for alignment measurements, such as any algorithm required to achieve maximum alignment over the length of the sequences to be compared. Generally, for proteins or nucleic acids, the length of the comparison may be any length including or less than the total length (e.g., 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, or 100%). For the purposes of the composition and method of the present invention, at least 80% of the total length of the sequence is aligned using a BLAST algorithm and default parameters.

[0042] For the purposes of the present invention, the comparison of sequences and the measurement of percentage identity between two sequences can be performed using a Blossum 62 scoring matrix with a gap penalty of 12, a gap extend penalty of 4, and a frameshift gap penalty of 5.

[0043] Conservative substitutions typically include substitutions within the following groups: glycine, alanine; valine, isoleucine, leucine; aspartic acid, glutamic acid, asparagine, glutamine; serine, threonine; lysine, arginine; and phenylalanine, tyrosine.

[0044] In some embodiments, the SpCas9 variant comprises one of the following sets of mutants: D1135V / R1335Q / T1337R (VQR variant); D1135V / G1218R / R1335Q / T1337R (VRQR variant); D1135E / R1335Q / T1337R (EQR variant); or D1135V / G1218R / R1335E / T1337R (VRER variant).

[0045] In some embodiments, the SpCas9 variant also has the following mutations that reduce or destroy the nuclease activity of Cas9: D10, E762, D839, H983 or D986, and one of H840 or N863, e.g., D10A / D10N and H840A / H840N / H840Y, which make the nuclease portion of the protein catalytically inactive, and the substitution at these positions may be alanine (as described in [Nishimasu al., Cell 156, 935-949 (2014)]), or other residues, e.g., glutamine, asparagine, tyrosine, serine, or aspartate, e.g., E762Q, H983N, H983Y, D986N, N863D, N863S, or N863H (see WO 2014 / 152432). In some embodiments, the variant includes a mutation in D10A or H840A (which produces a single-stranded nikase), or a mutation in D10A and H840A (which interferes with nuclease activity, and this mutant is known as dead Cas9 or dCas9).

[0046] Additionally, a SaCas9 variant is provided herein. The wild-type sequence of SaCas9 is as follows:

[0047]

[0048]

[0049] The SaCas9 variants described herein contain mutations at one or more of the following positions: E782, N968, and / or R1015 (or at similar positions above them). In some embodiments, the variant contains one or more of the following mutations: R1015Q, R1015H, E782K, N968K, E735K, K929R, A1021T, K1044N. In some embodiments, the SaCas9 variant contains mutations E782K, K929R, N968K, and R1015X, where X is any amino acid other than R. In some embodiments, the SaCas9 variant is at least 80% identical, e.g., at least 85%, 90%, or 95% identical to the amino acid sequence of SEQ ID NO:2, and differs, e.g., up to 5%, 10%, 15%, or 20% of the residues of SEQ ID NO:2 are replaced, e.g., by conservative mutations. In a preferred embodiment, the variant possesses the preferred activity of the parent, e.g., nuclease activity (except where the parent is nikase or dead Cas9) and / or the ability to interact with guide RNA and target DNA.

[0050] In some embodiments, the SaCas9 variant also has one of the following mutations that reduce or destroy the nuclease activity of SaCas9: D10A, D556A, H557A, N580A, e.g., D10A / H557A and / or D10A / D556A / H557A / N580A, which makes the nuclease portion of the protein catalytically inactive, and the substitution at these positions may be alanine (as described in [Nishimasu al., Cell 156, 935-949 (2014)]), or other residues, e.g., glutamine, asparagine, tyrosine, serine, or aspartate. In some embodiments, variants include mutations in D10A, D556A, H557A, or N580A (which may produce a single-stranded nikase), or mutations in D10A / H557A and / or D10A / D556A / H557A / N580A (which may interfere with nuclease activity due to similarity to SpCas9, and are also referred to as dead Cas9 or dCas9).

[0051] Additionally, the present invention provides an isolated nucleic acid encoding a SpCas9 and / or SaCas9 variant, a vector comprising the isolated nucleic acid operably linked to one or more regulatory domains to express the variant protein, and a host cell comprising the nucleic acid and optionally expressing the variant protein, e.g., a mammalian host cell.

[0052] The genome of a cell can be modified using the variants described herein, and the method generally comprises expressing a variant protein in a cell together with a guide RNA having a region complementary to a selected portion of the cell's genome. Methods for selectively modifying the genome of a cell are known in the art, for example, US8,697,359; US2010 / 0076057; US2011 / 0189776; US2011 / 0223638; US2013 / 0130248; WO / 2008 / 108989; WO / 2010 / 054108; WO / 2012 / 164565; WO / 2013 / 098244; WO / 2013 / 176772; US20150050699; US20150045546; US20150031134; US20150024500; US20140377868; US20140357530; US20140349400; US20140335620; US20140335063; US20140315985; US20140310830; US20140310828; US20140309487; US20140304853; US20140298547; US20140295556; US20140294773; US20140287938; US20140273234; US20140273232; US20140273231; US20140273230; US20140271987; US20140256046; US20140248702; US20140242702; US20140242700; US20140242699; US20140242664; US20140234972; US20140227787; US20140212869; US20140201857; US20140199767; US20140189896; US20140186958; US20140186919; US20140186843; US20140179770; US20140179006; US20140170753; Makarova et al., "Evolution and classification of the CRISPR-Cas systems" 9(6) Nature Reviews Microbiology 467-477 (1-23) (Jun. 2011); Wiedenheft et al., "RNA-guided genetic silencing systems in bacteria and archaea" 482 Nature 331-338 (Feb. 16, 2012); Gasiunas et al., "Cas9-crRNA ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in bacteria" 109(39) Proceedings of the National Academy of Sciences USA E2579-E2586 (Sep. 4, 2012); Jinek et al., "A Programmable Dual-RNA-Guided DNA Endonuclease in Adaptive Bacterial Immunity" 337 Science 816-821 (Aug. 17, 2012); Carroll, "A CRISPR Approach to Gene Targeting" 20(9) Molecular Therapy 1658-1660 (Sep. 2012); 2012년 5월 25일에 출원된 미국 출원 번호 61 / 652,086; Al-Attar et al., Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs): The Hallmark of an Ingenious Antiviral Defense Mechanism in Prokaryotes, Biol Chem. (2011) vol. 392, Issue 4, pp. 277-289; Hale et al., Essential Features and Rational Design of CRISPR RNAs That Function With the Cas RAMP Module Complex to Cleave RNAs, Molecular Cell, (2012) vol. 45, Issue 3, 292-302.

[0053] Instead of the SpCas9 protein described in the above reference, the variant protein described herein may be used with a guide RNA that targets a sequence having a PAM sequence according to Table 4 below.

[0054]

[0055] Additionally, the variants described herein may be used as fusion proteins instead of wild-type Cas9 or other Cas9 mutations known in the art (e.g., the dCas9 or Cas9 nikase described above), for example, as fusion proteins with heterofunctional domains described in WO 2014 / 124284. For example, a variant preferably comprising one or more nuclease-reducing or death mutations may be fused to a transcriptional activation domain or other heterofunctional domain on the N or C terminus of Cas9, for example, a transcriptional inhibitor (e.g., KRAB, ERD, SID, etc., for example, amino acids 473-530 of the inhibitor domain (ERD) of the ETS2 inhibitor factor (ERF), amino acids 1-97 of the KRAB domain of KOX1, or amino acids 1-36 of the Mad mSIN3 interaction domain (SID); see reference [Beerli et al., PNAS USA 95:14628-14633 (1998)]) or a silencer, for example, heterochromatin protein 1 (also known as HP1, swi6), for example, HP1α or HP1β; Proteins or peptides capable of recruiting long non-coding RNA (lncRNA) fused to a fixed RNA binding sequence, as known in the relevant art, such as MS2 coat protein, endoribonuclease Csy4, or lambda N protein; enzymes that alter the methylation status of DNA (e.g., DNA methyltransferase (DNMT) or TET protein); or enzymes that alter histone subunits (e.g., histone acetyltransferase (HAT), histone deacetylase (HDAC), histone methyltransferase (e.g., for the methylation of lysine or arginine residues) or histone demethylase (e.g., for the demethylation of lysine or arginine residues)) may also be used.Numerous sequences for such domains are known in the relevant art, for example, domains that catalyze the hydroxylation of methylated cytosine in DNA. Exemplary proteins include the Ten-Eleven-Translocation (TET)1-3 family, which are enzymes that convert 5-methylcytosine (5-mC) to 5-hydroxymethylcytosine (5-hmC) in DNA.

[0056] The sequence for human TET1-3 is known in the relevant technical field and is presented in the table below:

[0057]

[0058] * Variant (1) represents a longer transcript and codes for a longer isoform (a). Variant (2) has different coding sequences in the 5' UTR and 3' UTR compared to variant 1. The resulting isoform (b) is shorter and has a C-terminus distinct from isoform a.

[0059] In some embodiments, the catalytic module may include the entire or part thereof of the full-length sequence of the catalytic domain, e.g., a cysteine-rich extension, and a 2OGFeDO domain encoded by seven highly conserved exons, e.g., the Tet1 catalytic domain containing amino acids 1580–2052, the Tet2 containing amino acids 1290–1905, and the Tet3 containing amino acids 966–1678. For example, for an alignment illustrating major catalytic residues in all three Tet proteins, refer to Fig. 1 of Iyer et al., Cell Cycle. 2009 Jun 1;8(11):1698–710. Epub 2009 Jun 27, and for the full-length sequence, refer to its supplementary material (e.g., see seq 2c); in some embodiments, said sequence includes amino acids 1418–2136 of Tet1 or the corresponding region in Tet2 / 3.

[0060] Other catalytic modules may be from proteins identified in Iyer et al., 2009.

[0061] In some embodiments, the heterofunctional domain is a biological tether and comprises all or part of the MS2 coat protein, endoribonuclease Csy4, or lambda N protein (e.g., a DNA binding domain therefrom). These proteins can be used to recruit RNA molecules containing a specific stem-loop structure to a site of specificity on the dCas9 gRNA targeting sequence. For example, using a dCas9 variant fused to the MS2 coat protein, endoribonuclease Csy4, or lambda N, long non-coding RNA (lncRNA), such as XIST or HOTAIR, linked to the Csy4, MS2, or lambda N binding sequence can be recruited (see, for example, Keryer-Bibens et al., Biol. Cell 100:125-138 (2008)). Alternatively, for example, as described in Keryer-Bibens et al., the Csy4, MS2, or lambda N protein binding sequence may be linked to another protein, and said protein may be targeted to a dCas9 variant binding site using the methods and compositions described herein. In some embodiments, Csy4 is catalytically inactive. In some embodiments, as described in WO 2014 / 204578, a Cas9 variant, preferably a dCas9 variant, is fused to FokI.

[0062] In some embodiments, the fusion protein comprises a linker between the dCas9 variant and the heterofunctional domain. The linker that may be used in these fusion proteins (or between fusion proteins in the case of linked structures) may comprise any sequence that does not interfere with the function of the fusion protein. In a preferred embodiment, the linker is short, for example, of 2 to 20 amino acids, and is typically flexible (i.e., includes amino acids with high degrees of freedom, such as glycine, alanine, and serine). In some embodiments, the linker comprises one or more units of GGGS (Sequence Identification No.: 188) or GGGGS (Sequence Identification No.: 189), for example, two, three, four or more repeats of the GGGS (Sequence Identification No.: 188) or GGGGS (Sequence Identification No.: 189) units. Other linker sequences may also be used.

[0063] Expression system

[0064] To use the Cas9 variant described herein, it may be desirable to express it from the nucleic acid encoding it. This can be done in various ways. For example, the nucleic acid encoding the Cas9 variant can be cloned into an intermediate vector to transform it into a prokaryotic or eukaryotic cell for replication and / or expression. To store or manipulate the nucleic acid encoding the Cas9 variant to produce the Cas9 variant, the intermediate vector is typically a prokaryotic vector, e.g., a plasmid, or a shuttle vector, or an insect vector. The nucleic acid encoding the Cas9 variant can also be cloned into an expression vector to administer it into plant cells, animal cells, preferably mammalian cells or human cells, fungal cells, bacterial cells, or protozoan cells.

[0065] To achieve expression, the sequence typically encoding a Cas9 variant is subcloned into an expression vector containing a promoter that directs transcription. Suitable bacterial and eukaryotic promoters are widely known in the art and are described, for example, in the literature [Sambrook et al., Molecular Cloning, A Laboratory Manual (3d ed. 2001); Kriegler, Gene Transfer and Expression: A Laboratory Manual (1990); and Current Protocols in Molecular Biology (Ausubel et al., eds., 2010)]. Bacterial expression systems for expressing engineered proteins are, for example, E. coli ( E. coli ), Bacillus( Bacillus ) species, and Salmonella( Salmonella It can be obtained from (Palva et al., 1983, Gene 22:229-235). Kits for such expression systems are commercially available. Eukaryotic expression systems for mammalian cells, yeast, and insect cells are widely known in the relevant art and are also commercially available.

[0066] The promoter used to direct nucleic acid expression depends on the specific application. For example, typically, a strong constructive promoter is used for the expression and proliferation of fusion proteins. In contrast, when administering Cas9 variants in vivo for gene regulation, a constructive or inductive promoter may be used depending on the specific use of the Cas9 variant. Additionally, a promoter preferred for the administration of Cas9 variants may be a weak promoter, such as HSV TK or a promoter with similar activity. The promoter may also include elements reactive to transcriptional activation, e.g., hypoxia-responsive elements, Gal4-responsive elements, lac inhibitor-responsive elements, and small molecule control systems, e.g., tetracycline-regulated systems and RU-486 systems (see, e.g., Gossen & Bujard, 1992, Proc. Natl. Acad. Sci. USA, 89:5547; Oligino et al., 1998, Gene Ther., 5:491-496; Wang et al., 1997, Gene Ther., 4:432-441; Neering et al., 1996, Blood, 88:1147-55; and Rendahl et al., 1998, Nat. Biotechnol., 16:757-761).

[0067] In addition to the promoter, the expression vector typically contains a transcription unit or expression cassette containing all additional elements necessary for nucleic acid expression in a host cell, which is a prokaryotic or eukaryotic cell. Thus, a typical expression cassette contains, for example, a promoter operably linked to a nucleic acid sequence encoding a Cas9 variant, and any signals necessary for, for example, efficient polyadenylation of the transcript, transcription termination, ribosome binding sites, or translation termination. Additional elements of the cassette may include, for example, enhancers and sliced ​​heterogeneous intron signals.

[0068] Specific expression vectors for delivering genetic information to cells are selected in relation to the intended use of Cas9 variants, e.g., expression in plants, animals, bacteria, fungi, protozoa, etc. Standard bacterial expression vectors include plasmids, e.g., pBR322-based plasmids, pSKF, pET23D, and commercially available tag-fusion expression systems, e.g., GST and LacZ.

[0069] Expression vectors containing regulatory elements from eukaryotic viruses are often used in eukaryotic expression vectors, e.g., SV40 vectors, papillomavirus vectors, and vectors derived from Epstein-Barr virus. Other exemplary eukaryotic vectors include pMSG, pAV009 / A+, pMTO10 / A+, pMAMneo-5, baculovirus pDSVE, and any other vectors that allow protein expression under the direction of SV40 early promoter, SV40 late promoter, metallothionein promoter, rat mammary tumor virus promoter, Rous sarcoma virus promoter, polyhedrin promoter, or promoters suggested to be effective for expression in eukaryotic cells.

[0070] A vector for expressing a Cas9 variant may include an RNA Pol III promoter, e.g., H1, U6, or 7SK promoter, to induce the expression of a guide RNA. These human promoters allow the expression of the Cas9 variant in mammalian cells after plasmid transfection.

[0071] Some expression systems have markers for the selection of stably transfected cell lines, such as thymidine kinase, hygromycin B phosphotransferase, and dihydrofolate reductase. High-yield expression systems, such as using baculovirus vectors in insect cells with gRNA-coding sequences under the direction of a polyhedrin promoter or other potent baculovirus promoter, are also suitable.

[0072] Elements typically included in the expression vector also include a replicon that functions in E. coli, a gene encoding antibiotic resistance to allow selection of bacteria harboring the recombinant plasmid, and a unique restriction site in a non-essential region of the plasmid to allow insertion of the recombinant sequence.

[0073] Bacterial, mammalian, yeast, or insect cell lines expressing large amounts of protein are generated using standard transfection methods and then purified using standard techniques (e.g., Colley et al., 1989, J. Biol. Chem., 264:17619-22; see Guide to Protein Purification, in Methods in Enzymology, vol. 182 (Deutscher, ed., 1990)). Transformation of eukaryotic and prokaryotic cells is performed according to standard techniques (e.g., Morrison, 1977, J. Bacteriol. 132:349-351; see Clark-Curtiss & Curtiss, Methods in Enzymology 101:347-362 (Wu et al., eds, 1983)).

[0074] Any known procedure for introducing foreign nucleotide sequences into host cells may be used. These procedures include calcium phosphate transfection, polybrene, protoplast fusion, electroporation, nucleofection, liposomes, microinjection, naked DNA, plasmid vectors, viral vectors, both episodic and incorporated forms, and the use of other widely known methods for introducing cloned genomic DNA, cDNA, synthetic DNA, or other foreign genetic material into host cells (e.g., see Sambrook et al. above). The specific genetic manipulation procedure used must be capable of successfully introducing at least one gene into a host cell capable of expressing a Cas9 variant.

[0075] The present invention includes a vector and a cell comprising said vector.

[0076] Examples

[0077] The present invention is further described in the following examples, and the following examples do not limit the scope of the invention as described in the claims.

[0078] method

[0079] The following materials and methods were used in Examples 1 and 2.

[0080] Plasmids and Oligonucleotides

[0081] The schematic map and DNA sequence used in this study are Figs. 5A-J and can be verified in Sequence Identification Number: 7-20. The sequences of the oligonucleotides used to generate the positive selection plasmid, negative selection plasmid, and site-depletion library are Table 1 It was described in. In this study, the sequences of all gRNA targets are Table 2 It was described in ). Point mutations in Cas9 were generated by PCR.

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088] Bacterial Cas9 / sgRNA expression plasmids were constructed using two T7 promoters to express Cas9 and sgRNA separately. These plasmids were the human codon-optimized variant of Cas9 for S. pyogenes (BPK764, SpCas9 sequence subcloned from JDS246). 17), S. thermophilus from CRISPR locus 1 Cas9 (MSP1673, St1Cas9 sequence modified from a previously published specification) 20 ), and S. aureus (BPK2101, SaCas9 sequence codons optimized from Uniprot J7RUA5). SpCas9 34, 35 and St1Cas9 20 In the case of, a previously prepared sgRNA sequence was used, whereas the SaCas9 sgRNA sequence was determined by searching the European Nucleotide Archive sequence HE980450 for the crRNA repeat using CRISPRfinder and identifying the tracrRNA using a bioinformatics approach similar to that previously described. 36 Annealed oligos to complete the spacer complementation region of sgRNA were ligated to BsaI-cut BPK764 and BPK2101, or BspMI-cut MSP1673 (5'-ATAG was attached to the spacer to generate the upper oligo, and 5'-AAAC was attached to the reverse complement of the spacer sequence to generate the lower oligo).

[0089] Mutated PAM-interaction (PI) domain libraries were generated by randomly mutating SpCas9 residues 1097–1368 with ~5.2 substitutions / kilobases using Mutazyme II (Agilent Technologies). The theoretical complexity of each PI domain library was 10 based on the number of obtained transformants. 7 It was presumed to be clonal excess. Annealed target site oligos were cut into XbaI / SphI or EcoRI / SphI p11-lacY-wtx1, respectively. 17 Positive and negative selection plasmids were generated by ligating.

[0090] Two randomized PAM libraries (each having a different protospacer sequence) were constructed using Klenow(-exo) to fill the downstrand of an oligo containing six randomized nucleotides immediately adjacent to the 3' end of the protospacer ( Table 1 Note). The double-stranded product was cleaved by EcoRI, leaving the EcoRI / SphI ends for ligation to cut p11-lacY-wtx1. The theoretical complexity of each randomized PAM library is 10 based on the number of obtained transformants. 6 It was estimated to be in excess.

[0091] SpCas9 and SpCas9 variants were expressed in human cells from a vector derived from JDS246. 16 For St1Cas9 and SaCas9, Cas9 ORFs from MSP1673 and BPK2101 were subcloned into CAG promoter vectors to generate MSP1594 and BPK2139, respectively. Plasmids for U6 expression of sgRNA (where the desired spacer oligo can be cloned) were generated for SpCas9 sgRNA (BPK1520), St1Cas9 sgRNA (BPK2301), and SaCas9 gRNA (VVT1) using the sgRNA sequences described above. Annealed oligos to complete the spacer complementation region of the sgRNAs were ligated to the BsmBI overhangs of these vectors (5'-CACC was attached to the spacer to generate the upper oligo, and 5'-AAAC was attached to the reverse complement of the spacer sequence to generate the lower oligo).

[0092] Bacterial-based positive selection test for the evolution of SpCas9 variants

[0093] Qualified E. coli BW25141(λDE3) containing a positive selection plasmid (having an embedded target site) 23... was transformed with a Cas9 / sgRNA-coding plasmid. After recovery in SOB medium for 60 minutes, the transformants were plated on LB plates containing either chloramphenicol (non-selective) or chloramphenicol + 10 mM arabinose (selective). The splitting of the positive-selective plasmid was estimated by calculating the viability frequency: (colonies on the selective plate / colonies on the non-selective plate) Fig. 12 Also refer to).

[0094] To select SpCas9 variants capable of cleaving novel PAMs, a library of PI-domain mutated Cas9 / sgRNA plasmids was electroporated into E. coli BW25141 (λDE3) cells containing a positive selection plasmid encoding the target site + the PAM of interest. Typically, approximately 50,000 clones were screened to yield 50 to 100 survivors. The PI domains of the surviving clones were subcloned into new backbone plasmids and retested for positive selection. Clones with a survival rate of over 10% in this secondary screen for activity were sequenced. Mutations observed in the sequenced clones were analyzed based on their frequency, substitution type, and proximity to PAM bases in the SpCas9 / sgRNA crystal structure (PDB:4UN3) in the surviving clones 14 , and (in some cases) were selected for further evaluation based on activity in human cell-based EGFP destruction assays.

[0095] Bacterial-Repository-Depletion Scheme for Profiling Cas9 PAM Specificity

[0096] Qualified E. coli BW25141 (λDE3) containing a Cas9 / sgRNA expression plasmid were transformed with a negative-selection plasmid containing a cleavable or non-cleavable target site. After recovery in SOB medium for 60 minutes, the transformants were plated on LB plates containing chloramphenicol + carbenicillin. Cleavage of the negative-selection plasmid was estimated by calculating the colony-forming units per μg of transformed DNA ( Fig. 13 Also refer to).

[0097] The PAM specificity profile of Cas9 nucleases was determined by electroporating each randomized PAM library into E. coli BW25141 (λDE3) cells already possessing the appropriate Cas9 / sgRNA plasmid, by adjusting negative selection. 80,000–100,000 colonies were plated at low density on LB + chloramphenicol + carbenicillin plates. Surviving colonies containing negative-selected plasmids refractory to Cas9-induced cleavage were harvested, and the plasmid DNA was isolated by maxi-prep (Qiagen). The generated plasmid library was amplified by PCR using Phusion Hot-start Flex DNA Polymerase (New England BioLabs), followed by the Agencourt Ampure XP cleanup step (Beckman Coulter Genomics). In each site-depletion experiment, a Dual-indexed Tru-Seq Illumina deep-sequencing library was prepared from ~500 ng of clean PCR product using the KAPA HTP Library Preparation Kit (KAPA BioSystems). The Dana-Farber Cancer Institute Molecular Biology Core performed 150-bp paired-terminal sequencing on an Illumina MiSeq Sequencer.

[0098] The raw FASTQ files generated for each MiSeq operation were analyzed using a Python program to determine the relative PAM depletion. Program ( methodNote) The procedure was operated as follows: First, a file dialog was presented to the user, enabling the selection of all FASTQ read files for a given experiment. For these files, each FASTQ entry was scanned against fixed spacer regions on both strands. If a spacer region was found, six variable nucleotides flanking the spacer region were captured and countered. From the set of detected variable regions, the counts and frequencies of each window of 2–6 nt length at each possible location were tabulated. Site-depletion data for both randomized PAM libraries were analyzed by calculating the post-selection PAM depletion value (PPDV: the post-selection frequency of PAM in the selected population divided by the pre-selection library frequency of PAM). For each experiment, PPDV analysis was performed across all possible 2–6 length windows in a 6 bp randomized region. The windows used by the inventors to visualize PAM preference were a 3 nt window representing the 2nd, 3rd, and 4th PAM positions in wild-type and variant SpCas9 experiments, and a 4 nt window representing the 3rd, 4th, 5th, and 6th PAM positions in St1Cas9 and SaCas9.

[0099] Two significant titers for PPDV were determined based on the following: 1) Statistical significance titer based on the dCas9 distribution for the ratio of library log read counts prior to selection ( Figs. 13c & 13d1) and 2) biological activity titers based on the experimental relationship between depletion value and activity in human cells. Statistical titers were set at 3.36 standard deviations from the mean PPDV for dCas9 (equivalent to a relative PPDV of 0.85), which corresponded to a two-sided normal p-value of 0.05 (i.e., p=0.05 / 64) after adjustment for multiple comparisons. Since depletion levels serve as reasonable predictors of activity in human cells, biological activity titers were set at 5-fold depletion (equivalent to a PPDV of 0.2) ( Fig. 14 Also refer to). Fig. 14 The 95% confidence interval was calculated by dividing the standard deviation of the mean by the square root of the sample size multiplied by 1.96.

[0100] Human cell culture and transfection

[0101] U2OS.EGFP cells possessing a single integrated copy of the constitutively expressed EGFP-PEST reporter gene 15 Cells were cultured at 37°C under 5% CO2 in Advanced DMEM medium (Life Technologies) supplemented with 10% FBS, 2 mM GlutaMax (Life Technologies), penicillin / streptomycin, and 400 μg / ml G418. Cells were co-transfected with 750 ng of Cas9 plasmid and 250 ng of sgRNA plasmid (unless otherwise noted) using the Lonza 4D-NucleoFactor DN-100 program according to the manufacturer's protocol. Cas9 plasmids transfected with an empty U6 promoter plasmid were used as negative controls in all human cell experiments. In endogenous gene experiments, target sites were selected within a 200 bp NGG region cleavable by wild-type SpCas9 ( Fig. 16a and Table 2 reference).

[0102] Zebrafish care and injections

[0103] The care and use of zebrafish were approved by the Massachusetts General Hospital Subcommittee on Research Animal Care. As previously described, Cas9 mRNA was transcribed by PmeI-digested JDS246 (wild-type SpCas9) or MSP469 (VQR variant) using the mMESSAGE mMACHINE T7 ULTRA kit (Life Technologies). 21 In this study, all sgRNAs were prepared according to the cloning-independent sgRNA generation method. 24 sgRNA was transcribed using the MEGAscript SP6 transcription kit (Life Technologies), purified using RNA Clean & Concentrator-5 (Zymo Research), and eluted with RNase-free water.

[0104] sgRNA- and Cas9-coding mRNAs were co-injected into 1-cell stage zebrafish embryos. Each embryo was injected with ~2-4.5 nL of solution containing 30 ng / μL gRNA and 300 ng / μL Cas9 mRNA. The next day, the injected embryos were examined stereoscopically for normal morphological development, and genomic DNA was extracted from 5 to 9 embryos.

[0105] Human cell EGFP destruction test

[0106] The EGFP destruction experiment was performed as previously described. 16 Transfected cells were analyzed for EGFP expression approximately 52 hours after transfection using a Fortessa flow cytometer (BD Biosciences). Background EGFP loss was approximately 2.5% in all experiments (indicated by the red dashed line in the graph).

[0107] T7E1 assay, targeted deep sequencing, and GUIDE-seq for quantifying nuclease-induced mutation rates

[0108] human cells 15 and zebrafish 21 T7E1 assays were performed as previously described. For U2OS.EGFP human cells, genomic DNA was extracted from transfected cells approximately 72 hours after transfection using the Agencourt DNAdvance Genomic DNA Isolation Kit (Beckman Coulter Genomics). Target loci from zebrafish or human cell genomic DNA Table 1 Amplification was performed using the primers listed in [document]. Approximately 200 ng of purified PCR product was denatured, annealed, and digested with T7E1 (New England Biolabs). Mutagenicity was human cells. 15 and zebrafish 21 As previously described, it was quantified using a Qiaxcel capillary electrophoresis device (Qiagen).

[0109] For targeted deep sequencing, previously characterized on- and off-target sites (Tsai et al., Nat Biotechnol 33, 187-197 (2015); Fu et al., Nat Biotechnol 31, 822-826 (2013); Fu et al., Nat Biotechnol 32, 279-284 (2014)) Table 1Amplification was performed using a fusion hot-start plex with the primers listed in [reference]. Genomic loci were amplified for control conditions (empty sgRNA), wild-type, and D1135E SpCas9. Before pooling ~500 ng of DNA from each condition for library preparation, the ArgenCort Ampure XP cleanup step (Beckman Coulter Genomics) was performed. A dual-indexed Tru-Seq Illumina deep-sequencing library was generated using the KAPA HTP Library Preparation Kit (KAPA Biosystems). 150-bp paired-terminal sequencing was performed on an Illumina MiSeq sequencer using the Dana-Farber Cancer Institute Molecular Biology Core. Mutation analysis of the targeted deep-sequencing data was performed as previously described (Tsai et al., Nat Biotechnol 32, 569-576 (2014)). Briefly, Illumina MiSeq paired-terminal read data were mapped to the human genome-referenced GRChr37 using bwa (Li et al., Bioinformatics 25, 1754-1760 (2009)). High-quality reads (quality score >= 30) were evaluated for indel mutations overlapping with target or off-target sites. If a 1-bp indel mutation did not occur within the expected 1-bp cutoff point, the mutation was excluded from the analysis. Changes in activity at on- and off-target sites comparing D1135E and wild-type SpCas9 were calculated by comparing indel frequencies from both conditions (in the case of mutation rates exceeding background control amplicon indel levels).

[0110] GUIDE-seq experiments were performed as previously described (Tsai et al., Nat Biotechnol 33, 187-197 (2015)). Briefly, phosphorylated phosphorothioate-modified double-stranded oligodeoxynucleotides (dsODNs) were transfected into U2OS cells along with Cas9 and sgRNA expression plasmids using Cas9 nuclease as described above. To identify genomic intervals containing DSB activity, dsODN-specific amplification, high-yield sequencing, and mapping were performed. For wild-type versus D1135E experiments, off-target read counts were normalized to on-target read counts to correct for differences in sequencing depth between samples. Subsequently, the fold-variance of activity at off-target sites was calculated by comparing the normalization ratios of wild-type and D1135E SpCas9. To determine whether wild-type and D1135E samples for GUIDE-seq had similar oligo-tag integration rates at the intended target site, Table 1 Restriction Fragment Length Polymorphism (RFLP) assays were performed by amplifying intended target loci from 100 ng of genomic DNA (isolated as described above) using a fusion hot-start plex with the primers listed above. Before cleanup using the ArgenCort AmPure XP kit, approximately 150 ng of PCR product was digested with 20 U of NdeI (New England BioLabs) for 3 hours at 37°C. The RFLP results were quantified as approximate oligo-tag integration rates using a Qiagen capillary electrophoresis instrument. A T7E1 assay was performed for a similar purpose as described above.

[0111] Software - For analyzing PAM-depleted MiSeq data

[0112] Run from the command prompt (in the directory containing the file) using the command "python PAM_depletion.py".

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] Example 1

[0119] One potential approach to addressing the limitation of the targeting range is to engineer Cas9 variants with novel PAM specificity. Previous attempts to modify PAM specificity utilized structural information regarding base-specific SpCas9-PAM interactions by mutating arginine residues (R1333 and R1335, respectively) contacting guanine nucleotides at the second and third PAM positions (Anders et al., Nature 513, 569-573 (2014)). Substitution of both arginines with glutamine (whose side chains could be expected to interact with adenine) failed to generate SpCas9 variants capable of cleaving targets containing the expected NAA PAM in vitro (Anders et al., Nature 513, 569-573 (2014)). The inventors confirmed that the R1333Q / R1335Q SpCas9 variants failed to efficiently cleave target sites containing NAA PAM using a human cell-based U2OS EGFP reporter gene disruption assay in which nuclease-induced indels cause loss of fluorescence (Reyon et al., Nat Biotechnol 30, 460-465 (2012); Fu et al., Nat Biotechnol 31, 822-826 (2013)). Fig. 1aAdditionally, the inventors confirmed that a single R1333Q and R1335Q SpCas9 variant each failed to efficiently segment target sites having their expected NAG and NGA sites, respectively ( Fig. 1a Accordingly, the inventors determined that the re-engineered PAM specificity might require additional mutations at positions other than R1333 and R1335. For example, available structural information suggests that K1107 and S1136 make direct or indirect small groove contacts with the second and third bases of PAM, respectively (Anders et al., Nature 513, 569-573 (2014)). Therefore, it is reasonable to assume that additional changes at or near these positions might be required to alter PAM specificity.

[0120] To identify additional locations that may be critical for modifying PAM specificity, the inventors [researched] a bacterial selection system previously used to study the characteristics of homing endonucleases (hereinafter Positive selection (referred to as) was used (Chen & Zhao, Nucleic Acids Res 33, e154 (2005); Doyon et al., J Am Chem Soc 128, 2477-2484 (2006)). In the inventors' modification of this system, Cas9-mediated cleavage of a positively selected plasmid encoding an inducible toxicity gene enables cell survival due to the subsequent degradation and loss of the linearized plasmid ( Fig. 1b and Fig. 12a After establishing that SpCas9 can function in a positive selection system, the inventors tested the ability of both the wild-type and R1335Q variants to split a selection plasmid containing a target site with NGA PAM, and failed to observe survival as expected ( Fig. 12a( ). To screen for function-gain mutations, the inventors generated a library of wild-type and R1335Q SpCas9 containing randomly mutated PAM-interaction domains (amino acid positions 1097-1368) with an average mutation rate of 5.2 mutations / kilobases ( Fig. 12b and method These libraries were introduced into bacteria carrying a positive-selection plasmid containing a target site with NGA PAM and plated on selective media. From the sequences of surviving clones from the R1335Q-described library, it was revealed that the most frequent substitutions other than the existing R1335Q mutations were D1135V / Y / N / E and T1337R ( Table 3 ). The inventors obtained fewer survivors with wild-type SpCas9-based library selections, but the sequences of these clones also included D1135V / Y / N and R1335Q mutations. Next, the inventors assembled and tested SpCas9s possessing all possible single, double, and triple combinations of D1135V / Y / N / E, R1335Q, and T1337R mutations using a human cell-based EGFP disruption assay. This analysis suggested that SpCas9 variants with substitutions at all three positions exhibited peak activity against NGA PAM but trough activity against NGG PAM ( Fig. 1c ). For further characterization, the inventors [referenced] two SpCas9 variants, D1135V / R1335Q / T1337R and D1135E / R1335Q / T1337R (hereinafter each VQR and EQR We selected (referred to as SpCas9 variants) because they had the greatest difference between NGA and NGG PAM. Fig. 1c ).

[0121] To evaluate the overall PAM specificity profile for the inventors' novel SpCas9 variants, the inventors used a bacteria-based negative selection system ( Fig. 1d and Fig. 13a ). Previous studies have used a similar type of selection system to confirm the cleavage site preference of Cas9 nucleases (Jiang et al., Nat Biotechnol 31, 233-239 (2013); Esvelt et al., Nat Methods 10, 1116-1121 (2013)). In the inventors' version of this assay (the inventors Site-depletion test (referred to as) , A library of plasmids containing randomized 6 bp sequences located adjacent to the protospacer was tested for cleavage by the Cas9 / sgRNA complex in E. coli ( Fig. 13b Plasmids containing protospacer-adjacent sequences resistant to cleavage by the Cas9 / sgRNA complex enabled cell survival due to the presence of antibiotic resistance genes, whereas plasmids containing cleavable sequences were degraded and depleted from the library ( Fig. 13b ). High-yield sequencing of ~100,000 untargetable sequences is the PAM depletion value after selection for any given PAM ( PPDV ) enabled the calculation of. The PPDV of a PAM (or group of PAMs) is defined by dividing the frequency of the corresponding PAM in the post-selection group by its frequency in the pre-selection library. This quantitative value provides an estimate of the Cas9 activity for the corresponding PAM. For two randomized PAM libraries (each having different protospacers), catalytically inactive Cas9 ( dCas9 The profile obtained by ) makes it possible to define what indicates a statistically significant change in PPDV for any given PAM or group of PAMs ( Fig. 13c). Subsequently, the inventors validated their site-depletion test by demonstrating that PPDV against wild-type SpCas9 obtained using two randomized PAM libraries reproduced the previously described profile of the targetable PAM (Jiang et al., Nat Biotechnol 31, 233-239 (2013)) Fig. 1e ).

[0122] Using a site-depletion test, the inventors obtained PAM specificity profiles for VQR and EQR SpCas9 variants using two randomized PAM libraries. The VQR variant strongly depleted sites containing NGAN and NGCG PAMs, and weakly depleted sites containing NGGG, NGTG, and NAAG PAMs ( Fig. 1f In contrast, the EQR variant strongly depleted NGAG PAM, and more weakly depleted NGAT, NGAA, and NGCG PAM ( Fig. 1f ), this demonstrates a potentially more limited targeting range for VQR variants. To test whether the PAMs identified by the site-depletion assay can also be recognized in human cells, the inventors evaluated the cleavage by VQR and EQR SpCas9 variants to target sites using the EGFP depletion assay. VQR variants strongly cleaved the EGFP regions containing NGAN PAM (with relative efficiency in the order NGAG > NGAT = NGAA > NGAC), and also cleaved the regions containing NGCG, NGGG, and NGTG PAMs with overall lower efficiency ( Figure 1g The EQR variants also replicated their preference for NGAG and NGNG PAMs compared to other NGAN PAMs in human cells, with lower activity compared to the VQR variants ( Figure 1gIn summary, these results in human cells strongly reflect what was observed in the bacterial site-depletion test and ( Fig. 14 ), suggesting that a PPDV of 0.2 (indicating 5-fold depletion) in bacterial assays provides a reasonably expected titer for activity in human cells ( Fig. 14 ).

[0123] Next, the inventors sought to expand the generalizability of their engineering strategy by attempting to identify SpCas9 variants capable of recognizing NGC PAM. First, the inventors devised Cas9 mutants possessing amino acid substitutions (D, E, S, or T) of R1335 expected to interact with cytosine, and confirmed using a positive selection system that they were inactive toward the NGC PAM site. Subsequently, the inventors randomly mutated the PAM-interaction domain of each of these single-substituted SpCas9 variants, but still failed to obtain viable colonies in positive selection. Because the T1337R mutation increased the activity of the inventors' VQR and EQR SpCas9 variants ( Fig. 1c ), the inventors combined this mutation with R1335 substitutions of A, D, E, S, T, or V and randomly mutated their PAM-interaction domains again. The selection of two of these six mutated libraries (containing the existing R1335E / T1337R and R1335T / T1337R substitutions) generated viable colonies containing various additional mutations ( Table 3Characterization of various selected clones using both bacterial and human cell-based assays suggested that substitutions at four specific positions (D1135V, G1218R, R1335E, and T1337R) are considered important for the cleavage of NGC PAM. The assembly and testing of all potential single, double, triple, and quadruple combinations of these mutants using EGFP disruption assays revealed that the quadruple VRER It was established that the variant exhibits maximum activity against NGCG PAM and minimum activity against NGGG PAM ( Fig. 1h Analysis of VRER variants using site-depletion tests was revealed to be highly specific for NGCG PAM ( Fig. 1i Consistent with these results, EGFP disruption assays performed on human cells with VRER variants revealed efficient cleavage of the site containing NGCG PAM, significantly reduced and inconsistent cleavage of the sites containing NGCA, NGCC, and NGCT PAM, and intrinsic inactivity for the sites containing NGAG, NGTG, and NGGG PAM ( Figure 1j ).

[0124] To directly demonstrate that the inventors' VQR and VRER SpCas9 variants can enable the targeting of regions currently unmodifiable by wild-type SpCas9, the inventors tested their activity against endogenous genes in zebrafish embryos and human cells. In single-cell zebrafish embryos, the inventors found that the VQR variant can efficiently modify endogenous gene regions containing NGAG PAM with an average mutagenicity of 20 to 43% ( Fig. 2a ), that the indel originated from the expected split site ( Fig. 15) confirmed. In human cells, the inventors confirmed that the VQR variant strongly modified 16 sites across four different endogenous genes possessing NGAG, NGAT, and NGAA PAM (range of 6 to 53%, average 33%; Fig. 2b and Fig. 16a Importantly, although the inventors were unable to efficiently modify most of the same sites containing NGAG and NGAT PAM in zebrafish and human cells with wild-type SpCas9 ( Fig. 2a and 2c ), the area possessing NGG PAM was modified almost efficiently ( Fig. 16b ) demonstrated. Similarly, when VRER variant activity was tested at nine sites containing NGCG PAM across three endogenous human genes, the inventors demonstrated a potent average disruption frequency (ranging from 5 to 36%, average 21%; Fig. 2d ) was also observed. Consistent with the inventors' site-depletion data ( Fig. 1e & 1f ), VQR variants modified the NGCG PAM site with efficiency similar to that observed in VRER variants, whereas wild-type SpCas9 did not. Fig. 2d Computer analysis of the reference human genome sequence suggests that the addition of the inventors' VQR and VRER SpCas9 variants has doubled the range of potential target sites compared to what was previously possible with only wild-type SpCas9 ( Fig. 2e Taken together, these results demonstrate that the inventors' engineered SpCas9 variants expand the targeting range of SpCas9 by enabling modification of previously inaccessible endogenous sites in zebrafish embryos and human cells.

[0125] To determine the genome-wide specificity of the inventors' VQR and VRER SpCas9 nucleases, the inventors use a recently described GUIDE-seq (Genome-wide Unbiased Identification of Double-stranded breaks Enabled by sequencing) method to profile off-target splitting events of these SpCas9 variants in human cells. 10 The inventors profiled the genome-wide activity of VQR and VRER SpCas9 variants using a total of 13 different sgRNAs (each Fig. 2b and 2d (8 for VQR and 5 for VRER), thereby demonstrating that the inventors can induce high efficiency at their intended on-target sites. These GUIDE-seq experiments yielded numerous important observations: the number of off-target DSBs induced by the inventors' SpCas9 variants in human cells was comparable to that previously observed with wild-type SpCas9 (or perhaps even better in the case of VRER variants) Fig. 2f The inventors noted that the high genome-wide specificity observed by VRER could be generated by both its limited specificity for NGCG PAM and possibly the relative depletion of sites containing NGCG PAM in the human genome ( Fig. 2e ) 21 Additionally, the observed off-target site generally possesses the PAM sequence expected by the inventors' site-depletion experiments, including some resistance to PAM with the 3' shifted by one base ( Fig. 1f and 1i PAM from Fig. 17 (Compared to that from). Finally, the locations and number of mismatches identified in off-target sites of the inventors' VQR and VRER SpCas9 variants ( Fig. 17The inventors found that the distribution of sgRNAs targeted with non-repetitive sequences was similar to that previously observed with wild-type SpCas9. 10 .

[0126] Previous studies have suggested that incomplete PAM recognition by SpCas9 can lead to the recognition of unwanted sites containing non-standard NAG, NGA, and other PAMs in human cells (Hsu et al., Nat Biotechnol 31, 827-832 (2013); Tsai et al., Nat Biotechnol 33, 187-197 (2015); Jiang et al., Nat Biotechnol 31, 233-239 (2013); Mali et al., Nat Biotechnol 31, 833-838 (2013); Zhang et al., Sci Rep 4, 5405 (2014)). Therefore, the inventors were interested in investigating whether mutations in or near residues mediating PAM interactions improve SpCas9 PAM specificity. While manipulating VQR variants, the inventors noted that the D1135E SpCas9 mutant is believed to distinguish between standard NGG PAM and non-standard NGA PAM better than wild-type SpCas9 ( Fig. 1c ( ). Under these observations, the inventors thoroughly evaluated the PAM recognition profile of the D1135E variant using the inventors' site-depletion test. From this experiment, it was revealed that the depletion of non-standard NAG, NGA, and NNGG PAM by D1135E SpCas9 was reduced compared to wild-type SpCas9 ( Fig. 3aInterestingly, this effect was more pronounced for one of the two protospacers used by the inventors, which suggests that the effect of the D1135E substitution on non-standard PAM recognition can be changed to some extent in a protospacer-dependent manner. Importantly, the inventors did not observe the emergence of any new non-standard PAM specificity.

[0127] Next, the inventors tested whether the improved PAM specificity of D1135E SpCas9 could also be observed in human cells. In a direct comparison of wild-type and D1135E SpCas9 for eight target sites with non-standard NAG or NGA PAMs, the inventors observed that these sites were consistently cleaved less efficiently by D1135E than by wild-type SpCas9 in the EGFP disruption assay ( Fig. 3b , with an average multiple decrease of 1.94 in activity). Importantly, both wild-type and D1135E SpCas9 demonstrated comparable activity for four EGFP reporter gene regions and six endogenous human gene regions with standard NGG PAM (each Figures 3b and 3c ), this demonstrates that the D1135E variant does not have a noticeable effect on the splitting of on-target sites containing NGG PAM (an average multiple decrease of 1.04 in activity across all 10 sites). In titration experiments in which the inventors reduced the concentration of Cas9-coding plasmids transfected in human cells, it was revealed that there was no substantial difference in the activity of wild-type and D1135E SpCas9 when targeting the same sites ( Fig. 3d This suggests that the increased specificity observed in the D1135E variant is not simply a result of protein destabilization.

[0128] To more directly evaluate whether the introduction of D1135E could reduce the off-target splitting effect of SpCas9, the inventors used deep sequencing to compare the mutation rates induced by wild-type and D1135E SpCas9 for 25 previously known off-target sites of three different sgRNAs (Hsu et al., Nat Biotechnol 31, 827-832 (2013); Tsai et al., Nat Biotechnol 33, 187-197 (2015); Fu et al., Nat Biotechnol 31, 822-826 (2013)). These 25 sites included off-target sites with various mismatches in spacer sequences and both standard NGG and non-standard PAM ( Fig. 3e The results of these deep-sequencing experiments revealed that the D1135E variant exhibited reduced mutation frequencies in 19 of the 22 off-target sites with activity exceeding the background indel ratio compared to the mutation frequencies observed in 3 on-target sites ( Fig. 3e & 3fInterestingly, these reduced off-target mutation frequencies were observed in several sites with standard PAMs, suggesting that the increase in specificity caused by D1135E is not limited only to sites with non-standard PAMs. To evaluate the improvement in specificity associated with D1135E on a genome-wide scale, the inventors performed GUIDE-seq experiments using wild-type and D1135E SpCas9 with three different sgRNAs (two of which were previously known to have off-target sites with standard and non-standard PAMs (Hsu et al., Nat Biotechnol 31, 827-832 (2013); Tsai et al., Nat Biotechnol 33, 187-197 (2015); Fu et al., Nat Biotechnol 31, 822-826 (2013)). The inventors observed an overall improvement in genome-wide specificity when comparing the D1135E SpCas9 variants with wild-type SpCas9 ( Figure 3g For all three sgRNAs tested by the inventors, these improvements in specificity were observed at off-target sites containing mismatched spacers with standard or non-standard PAMs ( Fig. 18 Importantly, these GUIDE-seq experiments demonstrated that the introduction of the D1135E mutation did not increase the number of off-target effects induced by SpCas9. Taken together, these results suggest that the D1135E substitution can increase the overall specificity of SpCas9.

[0129] Although all experiments described above were performed using SpCas9, there are several Cas9 heterologs from other bacteria that could be attractive candidates for characterizing and manipulating Cas9 with novel PAM specificity (Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014); Ran et al., Nature 520, 186-191 (2015)). To explore the feasibility of this, the inventors [investigated] two similar-sized heterologs, Streptococcus thermophilus Cas9 from a CRISPR1 locus ( St1Cas9 ) (Deveau et al., J Bacteriol 190, 1390-1400 (2008); Horvath et al., J Bacteriol 190, 1401-1412 (2008)) and Staphylococcus aureus ( SaCas9 ) (Hsu et al., Cell 157, 1262-1278 (2014); Ran et al., Nature 520, 186-191 (2015)) determined whether it could also function in the inventors' bacterial selection test. While the PAM of St1Cas9 was previously characterized as NNAGAA (sequence identification number: 3) (Esvelt et al., Nat Methods 10, 1116-1121 (2013); Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014); Deveau et al., J Bacteriol 190, 1390-1400 (2008); Horvath et al., J Bacteriol 190, 1401-1412 (2008)), the inventors' attempt to derive the bioinformatics of the SaCas9 PAM using the previously described approach (Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014)) failed to obtain a consensus sequence ( Data not presented). Accordingly, the inventors utilized their site-depletion assay to determine PAMs for SaCas9 and St1Cas9 as a positive control. These experiments were performed using two different protospacers and sgRNAs with two different complementary lengths for each protospacer, resulting in four choices for Cas9. For St1Cas9, the inventors identified two novel PAMs in addition to the six previously described PAMs (Esvelt et al., Nat Methods 10, 1116-1121 (2013); Fonfara et al., Nucleic Acids Res 42, 2577-2590 (2014); Horvath et al., J Bacteriol 190, 1401-1412 (2008)) Fig. 4a and Fig. 19c and 19d , consistent with the recent definition of SaCas9 PAM specificity (Ran et al., Nature 520, 186-191 (2015))). In the case of SaCas9, PPDV variability was present across the four selections due to a restricted PAM preference observed primarily by a single protospacer. Consequently, only three PAMs were depleted beyond fivefold in all four experiments: NNGGGT (Sequence ID: 4), NNGAAT (Sequence ID: 6), and NNGAGT (Sequence ID: 5) ( Fig. 4b However, the inventors identified several more targetable PAMs using the second protospacer library, which suggests that SaCas9 can recognize numerous additional PAMs ( Fig. 18c and 1 8dUsing the PAM identified in the inventors' site-depletion experiments (NNAGAA for St1Cas9 (sequence identification number: 3) and NNGAGT for SaCas9 (sequence identification number: 5)), the inventors confirmed that both St1Cas9 and SaCas9 can function efficiently in a bacteria-positive selection system ( Fig. 4c This suggests that their PAM specificity can be modified by mutagenicity and selection.

[0130] Because not all Cas9 heterologs function efficiently outside of their natural state (Esvelt et al., Nat Methods 10, 1116-1121 (2013)), the inventors tested whether St1Cas9 and SaCas9 could potently cleave target sites in human cells. It has been previously suggested that St1Cas9 functions as a nuclease in human cells, but only in certain regions (Esvelt et al., 2013; Cong et al., Science 339, 819-823 (2013)). The inventors evaluated St1Cas9 activity toward the region possessing NNAGAA (SEQ ID: 3) PAM using sgRNA with a variable-length complementary region, and high activity was confirmed at three out of five target sites ( Fig. 4d In the case of SaCas9, the inventors observed efficient activity at eight sites possessing NNGGGT (Sequence Identification No.: 4) or NNGAGT (Sequence Identification No.: 5) PAM ( Fig. 4e For both St1Cas9 and SaCas9, no apparent correlation was observed between the activity and length of spacer complementarity ( Fig. 19e). Next, the inventors determined whether St1Cas9 and SaCas9 could efficiently modify endogenous loci in human cells. In the case of St1Cas9, as determined by the T7E1 test, 7 out of 11 sites across 4 genes were efficiently destroyed (1 to 25%, average 13%; Fig. 4f ), SaCas9 showed somewhat more potent activity at 16 sites tested across 4 genes (1% to 37%, average 19%; Figure 4g Once again, for St1Cas9 and SaCas9, no apparent trend was observed when considering sgRNA spacer length ( Fig. 19f Taken together, the inventors' results suggest that St1Cas9 and SaCas9 function potently in both the inventors' bacterial-substrate selection and human cells, making them attractive candidates for fabricating additional SpCas9 variants with novel PAM specificity.

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138] Example 2. Manipulation of PAM specificity of Staphylococcus aureus Cas9

[0139] The inventors of the present invention are Streptococcus pyogenes Since it was known that residues (R1333 and R1335) of Cas9 (SpCas9) are important for PAM recognition, the inventors generated an alignment of Cas9 heterologous genes to find homologous residues in the PAM-interaction domain (PI domain) of Staphylococcus aureus Cas9 (SaCas9). Fig. 6 Note). The inventors have previously suggested that the PAM of SaCas9 is NNGRRT (sequence identification number: 46) (where N is any nucleotide and R is A or G). Based on the inventors' data, the preference for G at the third position of the PAM was considered to be the strictest requirement, and the inventors hypothesized that a positively charged residue, such as lysine (K) or arginine (R), could mediate the interaction. Fig. 6 As shown in [the text], there are numerous candidate residues of SaCas9, such as K1101, R1012, R1015, K1018, and K1023, in the homology region for R1333 and R1335 of SpCas9.

[0140] Whether the mutant clone can still split the site containing the standard NNGRRT PAM (Sequence ID: 46), or possibly the NN that was previously untargetable A To determine whether the PAM (sequence identification number: 43) of RRT can be cleaved, the inventors produced alanine (A) and glutamine (Q) substitutions at these five positions ( Fig. 7 The inventors utilized the inventors' bacterial assay (as described in a previous patent application), wherein the activity of Cas9 can be visualized by the survival of bacterial colonies when plated under selective conditions. The relative activity of Cas9 could be quantified by calculating the ratio of bacterial colonies grown on a selective medium compared to a non-selective medium. Fig. 7In this, the inventors state that only R1015A and R1015Q mutations affect the ability of SaCas9 to recognize standard NNGAGT (Sequence Identification No.: 5) PAM, whereas no mutations NN A It was shown that RRT (Sequence Identifier: 43) did not enable the targeting of PAM (NNAAGT (Sequence Identifier: 41) or NNAGGT (Sequence Identifier: 42)). These results suggest that R1015 plays a specific role in PAM recognition by SaCas9.

[0141] Subsequently, the inventors selected a randomly mutated wild-type SaCas9 or R1015Q variant, and NN A AGT (Sequence Identification Number: 41) or NN A A modified PAM-specific clone for the GGT (Sequence Identification No.: 42) PAM-containing region was selected (as previously described for SpCas9). The inventors identified, re-screened, and sequenced numerous mutant clones capable of targeting these PAMs, and their amino acid sequences Fig. 8 (and Table 6 These sequences were presented in ). To summarize these sequences, while numerous changes were considered to be very important in altering SaCas9 specificity (R1015Q, R1015H, E782K), many other mutations may also contribute (N968K, E735K, K929R, A1021T, K1044N).

[0142] After identifying the locations and mutations essential for altering the PAM specificity of SaCas9 to NNARRT (SEQ ID: 43), the inventors created single, double, and triple mutant combinations and evaluated the contribution of the most abundant mutation to the change in specificity ( Table 5When these mutations were tested against various PAMs in the inventors' positive selection (previously described), the inventors observed that while numerous mutations allowed activity against both standard NNGAGT (SEQ ID: 5) and non-standard NNAAGT (SEQ ID: 41) or NNAGGT (SEQ ID: 42) PAMs, the wild-type SaCas9 enzyme had very low activity against non-standard PAMs. Specifically, triple mutations were thought to enable relaxed specificity at the third position of PAM (KKQ, KKH, GKQ, GKH—named based on mutations at positions E782 / N968 / R1015), inducing the consensus PAM motif of NNRRRT (SEQ ID: 45) against standard NNGRRT (SEQ ID: 46). This relaxation of the PAM requirement theoretically doubles the targeting range of SpCas9. Subsequently, variants will be named based on their identities for positions 782, 968, and 1015. For example, E782K / N968K / R1015H will be named the SaCas9 KKH variant.

[0143]

[0144] Next, the inventors evaluated two of the triple mutants in a human cell EGFP disruption assay (previously described) to determine whether the engineered variant could target non-standard PAM in human cell conditions ( Fig. 9 Variants capable of targeting regions within the EGFP gene containing non-standard PAMs will disrupt the EGFP coding frame, thereby inducing a loss of signal. As a result, both KKQ and KKH mutants were found to possess activity similar to wild-type SaCas9 against standard NNGRRT (sequence identification number: 46) PAMs, whereas they had much higher activity against NNARRT (sequence identification number: 43) PAMs.

[0145] Overall, the inventors have identified mutations in SaCas9 (KKQ or KKH variants) that are believed to mitigate the wild-type enzyme's preference from G to R (A or G) at the third position of PAM. This effectively mitigates the targeting of SaCas9 from the NNGRRT (SEQ ID: 46) PAM restriction to the NNRRRT (SEQ ID: 45) PAM restriction.

[0146] Since the inventors successfully induced a variant capable of targeting NNARRT (SEQ ID: 43) PAM in human cells, the inventors next questioned whether they could engineer variants specific to NNCRRT (SEQ ID: 47) or NNTRRT (SEQ ID: 48). To this end, the inventors first mutated R1015 to E (specific to C at the third position of PAM) and to L or M (specific to T at the third position of PAM), and tested them against their expected PAMs in the inventors' bacterial positive selection assay (previously described). Fig. 10 The inventors observed that wild-type SaCas9 can inefficiently cleave the site containing NNCAGT (SEQ ID: 511) PAM, that the R1015E variant has slightly more activity for the same site, and that the wild-type or any other designated mutant does not transfer activity for other PAMs ( Fig. 10 As confirmed by the inventors in R1015Q, this suggested that other mutations would be required to fabricate SaCas9 variants capable of targeting NNCRRT (sequence identification number: 47) and NNTRRT (sequence identification number: 48) PAMs.

[0147] For the SaCas9 evolved variant of NNARRT (sequence identification number: 43) PAM, E782K and N968K mutations along with R1015(H / Q) were required and essential. To test whether these mutations increase the activity of the R1015(E / L / M) variant of their expected PAM, the inventors generated KKE, KKL, and KKM variants. Fig. 11 As presented, KKE, KKL, and KKM all had strong activity against their expected PAM.

[0148] In addition, the inventors questioned whether the KKQ, KKH, KKE, KKL, or KKM variants had relaxed specificity for any nucleotide at the third position of PAM, and accordingly, the inventors investigated numerous sites in the inventors' bacterial positive selection test containing NNNRRT PAM. Fig. 11 As presented in [document], with a few exceptions, almost all of these variants can cleave all tested sites containing NNNRRT PAM. This indicated that they possessed relaxed specificity at the third position of PAM because they could efficiently target the NNNRRT site. This contrasts with the wild-type protein (ENR), which can efficiently target only the NNGAGT (SEQ ID: 5) site while exhibiting very low activity for some NNNRRT sites. In summary, KKH (and Fig. 11 Other similar derivatives presented in [the text]) variants can target sites containing NNNRRT PAM in bacteria, which efficiently quadruples the targeting range of SaCas9.

[0149] Therefore, the KKH variant (and some other variants shown in Fig. 6) can target NNNRRT PAM in bacteria, which efficiently quadruples the targeting range of SaCas9.

[0150]

[0151]

[0152]

[0153]

[0154]

[0155]

[0156]

[0157]

[0158] Method for Example 3

[0159] The following materials and methods were used in Example 3.

[0160] Plasmids and Oligonucleotides

[0161] Oligonucleotides Table 11 Listed in, and sgRNA target sites Table 12 Listed in, and the plasmids used in this study Table 10 Listed in.

[0162] Using bacterial Cas9 / sgRNA expression plasmids, both human codon-optimized versions of SaCas9 and sgRNA were expressed under separate T7 promoters. The bacterial expression plasmid used in this selection was derived from BPK2101 (see Examples 1-2), whereas the one used for the site-depletion assay was modified to express sgRNA with a shortened repeat:anti-repeat sequence (see below). In these bacterial expression plasmids, all sgRNAs contained two guanines at the 5' end of the spacer sequence for proper expression from the T7 promoter.

[0163] To generate a library of SaCas9 variants, amino acids M657–G1053 of SaCas9 were randomly mutated using Mutazim II (Agilent Technologies) at a frequency of ~5.5 mutations / kilobase. Both wild-type and R1015Q SaCas9 were used as starting templates for mutagenesis, 6x10 6 Two libraries were created with the estimated complexity of excess clones.

[0164] Positive selection plasmids were assembled by ligating an oligonucleotide duplex encoding the target site with XbaI / SphI-digested p11-lacY-wtx1 (Chen, Z. & Zhao, H. A highly sensitive selection method for directed evolution of homing endonucleases. Nucleic Acids Res 33, e154 (2005)). For site-depletion experiments, two separate libraries containing different spacer sequences were generated. For each library, an oligonucleotide containing eight randomized nucleotides adjacent to the spacer sequence (instead of PAM) was complexed with a downstrand primer and filled using Klenow(-exo) ( Table 11 Note). The generated product was digested with EcoRI and ligated to EcoRI / SphI-digested p11-lacY-wtx1. The estimated complexity of the two site-depletion libraries is 4x10 6 It was an excess clone.

[0165] For human cell experiments, human codon-optimized wild-type and variant SaCas9 were expressed from a plasmid containing a CAG promoter ( Table 12). sgRNA expression plasmids (containing the U6 promoter) were generated by ligating an oligonucleotide duplex encoding the spacer sequence into BsmBI-digested VVT1 (see Examples 1-2, or BPK2660 (containing full-length 120 nt crRNA:tracrRNA sgRNA or 84 nt shortened repeat:anti-repeat versions, respectively). For human expression, all sgRNAs used in this study included one guanine at the 5' end of the spacer to ensure proper expression from the U6 promoter, and shortened sgRNAs similar to those previously described were also used ( Fig. 37A-B ) (Ran, FA et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186-191 (2015)).

[0166] Bioinformatics analysis of Cas9 heterologous gene sequences

[0167] Similar to what was done in previous studies (Fonfara, I. et al. Phylogeny of Cas9 determines functional exchangeability of dual-RNA and Cas9 among orthologous type II CRISPR-Cas systems. Nucleic Acids Res 42, 2577-2590 (2014); Ran, FA et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186-191 (2015); Anders, C., Niewoehner, O., Duerst, A. & Jinek, M. Structural basis of PAM-dependent target DNA recognition by the Cas9 endonuclease. Nature 513, 569-573 (2014)), Cas9 heterologs similar to both SpCas9 and SaCas9 were analyzed using ClustalW2 (ebi.ac.uk / Tools / msa / clustalw2 / ). Aligned. The generated phylogenetic tree and protein alignment were visualized using Genious version 8.1.6 and ESPript (espript.ibcp.fr / ESPript / ESPript / ).

[0168] Bacteria-Instrumented Positive Selection Test

[0169] Bacterial positive selection assays were performed as previously described (see Examples 1-2). Briefly, the Cas9 / sgRNA plasmid was transformed into E. coli BW25141(λDE3) (Kleinstiver et al., Nucleic Acids Res 38, 2411-2427 (2010)) containing the positive selection plasmid. Transformers were plated under both non-selective (chloramphenicol) and selective (chloramphenicol + 10 mM arabinose) conditions. Cas9 splitting of the selective plasmid was estimated by calculating the percentage survival: (number of colonies on the selective plate / number of colonies on the non-selective plate) x 100. To select SaCas9 variants capable of recognizing alternative PAMs, wild-type and R1015Q libraries with mutated PI domains were transformed into qualified E. coli BW25141(λDE3) containing positive-selection plasmids with NNAAGT (SEQ ID: 41), NNAGGT (SEQ ID: 42), NNCAGT (SEQ ID: 511), NNCGGT (SEQ ID: 512), NNTAGT (SEQ ID: 513), or NNTGGT (SEQ ID: 514) PAMs. Approximately 1 x 10⁶ 5 Clones were screened by plating under selective conditions, and viable colonies containing SaCas9 variants presumed to split the selected plasmid were isolated by miniprep (MGH DNA Core). All variants were individually re-screened in a positive selection assay, and those with a viability greater than ~20% were sequenced to determine the mutations required for the recognition of alternative PAMs.

[0170] Bacteria-based site-depletion test

[0171] Site-depletion experiments were performed as previously described (see Examples 1-2). Briefly, randomized PAM libraries were electroporated onto eligible E. coli BW25141 (λDE3) containing wild-type, catalytically inactive (D10A / H557A), or KKH variant SaCas9 / sgRNA plasmids. 1x10 5 Excess colonies were plated on chloramphenicol / carbenicillin plates, and a selected plasmid containing a PAM resistant to Cas9 targeting contained within the surviving colonies was isolated by maxiprep (Qiagen). The spacer sequence and the region of the plasmid containing the PAM Table 11 Amplification was performed by PCR using the primers listed in [the document]. Before running Illumina MiSeq high-yield sequencing on the Dana-Farber Cancer Institute Molecular Biology Core, a dual-indexed Tru-Seq Illumina sequencing library was generated using ~500 ng of purified PCR product from each site-depletion condition using the KAPA HTP Library Preparation Kit (KAPA Biosystems). Data from the site-depletion experiments were analyzed as previously described, except for the analysis of 8 randomized nucleotides by modifying the script (see Examples 1-2). The ability of Cas9 to recognize PAM was determined by calculating the PAM depletion value (PPDV) after the selection of any given PAM: the ratio of the frequency of the corresponding PAM after selection to the library frequency before selection. Using a control experiment with catalytically inactive SaCas9, it was established that a PPDV of 0.794 indicates statistically significant depletion for the input library.

[0172] Human cell culture and transfection

[0173] U2OS cells were obtained from the inventors' collaborator T. Cathomen (Freiburg), and U2OS.EGFP cells possessing a single integrated copy of the EGFP-PEST reporter gene (Reyon, D. et al. FLASH assembly of TALENs for high-throughput genome editing. Nat Biotechnol 30, 460-465 (2012)) were cultured in Advanced DMEM medium (Life Technologies) containing 10% FBS, penicillin / streptomycin, and 2 mM GlutaMAX (Life Technologies) at 37°C under 5% CO2. Cell line identity was confirmed by STR profiling (ATCC) and deep sequencing, and cells were tested for Mycoplasma contamination every two weeks. 400 μg / mL G418 was additionally supplemented to the U2OS.EGFP culture medium. Cells were co-transfected with 750 ng Cas9 plasmid and 250 ng sgRNA plasmid using Lonza 4D-NucleoFactor's DN-100 program according to the manufacturer's instructions.

[0174] Human cell EGFP destruction test

[0175] EGFP degradation experiments were performed as previously described (Fu, Y. et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol 31, 822-826 (2013); Reyon, D. et al. FLASH assembly of TALENs for high-throughput genome editing. Nat Biotechnol 30, 460-465 (2012)). After approximately 52 hours of transfection, EGFP fluorescence was measured in transfected U2OS.EGFP cells using a Fortessa flow cytometer (Vidi Biosciences). A background EGFP loss of approximately 2.5% was established for all experiments using negative control transfections of Cas9 and empty U6 promoter plasmids (indicated by the red dashed line in the figure).

[0176] T7E1 test

[0177] The T7E1 assay was performed as previously described (Reyon, D. et al. FLASH assembly of TALENs for high-throughput genome editing. Nat Biotechnol 30, 460-465 (2012)) to quantify Cas9-induced mutagenicity at endogenous loci in human cells. After transfection for approximately 72 hours, genomic DNA was isolated using the Argencort DN-Advance Genomic DNA Isolation Kit (Beckman Coulter Genomics). Target loci were amplified by PCR from ~100 ng of genomic DNA using the primers listed in Table 11. Following the Argencort Ampure XP cleanup step (Beckman Coulter Genomics), ~200 ng of the purified PCR product was denatured and hybridized before digestion by T7E1 (New England Biolabs). After the second cleanup step, the mutagenic frequency was quantified using a Quaxel capillary electrophoresis device (Quiagen).

[0178] GUIDE-seq experiment

[0179] GUIDE-seq experiments were performed and analyzed as previously described (Tsai, SQ et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 33, 187-197 (2015)). Briefly, U2OS cells were transfected with Cas9 and sgRNA plasmids as described above, as well as 100 pmol of phosphorylated phosphorothioate-modified double-stranded oligodeoxynucleotide (dsODN) containing an embedded NdeI site. Restriction Fragment Length Polymorphism (RFLP) analysis was performed to determine the frequency of dsODN-tag integration (see Examples 1-2; Tsai, SQ et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 33, 187-197 (2015)), and the frequency of on-target Cas9 mutagenicity was quantified by performing T7E1 assay. dsODN tag-specific amplification and library preparation (Tsai, SQ et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 33, 187-197 (2015)) were performed prior to high-yield sequencing using an Illumina MiSeq sequencer. When mapping potential off-target sites, the alignment cutoff for the on-target spacer sequence was set to 8 mismatches for 21-nucleotide spacers, 9 mismatches for 22-nucleotide spacers, and 10 mismatches for 23-nucleotide spacers. Off-target sites with potential DNA- or RNA-bulges (Lin, Y. et al.CRISPR / Cas9 systems have off-target activity with insertions or deletions between target DNA and guide RNA sequences. Nucleic Acids Res 42, 7473-7485 (2014)) was confirmed by manual alignment.

[0180] Table 10 - Plasmid used in Example 3

[0181]

[0182] Table 11 - Oligonucleotides used in Example 3

[0183]

[0184] Table 12 - sgRNA target sites for Example 3

[0185]

[0186]

[0187]

[0188] * Fig. 1c and 1e, Fig. 32 Used in

[0189] ** Fig. 3 , Fig. 36A-B at Used for GUIDE-seq experiments

[0190] Example 3. Manipulation of PAM specificity of Staphylococcus aureus Cas9

[0191] Site-specific DNA cleavage by CRISPR-Cas9 nucleases is primarily directed by RNA-DNA interactions, but protospacer adjacent motifs ( PAM Cas9-mediated recognition of ) is also required. Commonly used Streptococcus pyogenes While Cas9 is specific to only two nucleotides within its NGG PAM, other Cas9 heterologous genes possessing the desired characteristics recognize longer PAMs. Although there is a potential advantage in terms of specificity, extended PAM sequences can limit the targeting range of Cas9 heterologous genes for genome editing applications. A possible strategy to broaden the range of sequences targetable by these Cas9 heterologous genes may be to evolve variants with relaxed specificity for specific locations within the PAM. Here, the inventors [discuss] Staphylococcus aureus Cas9, a small heterologous gene useful for applications requiring viral delivery ( SaCas9 Molecular evolution was utilized to modify the NNGRRT (Sequence Identification No.: 46) PAM specificity of ). One variant identified by the inventors, designated as KKH SaCas9, exhibited potent genome-editing activity at endogenous human target sites containing NNNRRT PAM. Importantly, using the GUIDE-seq method, the inventors demonstrated that both the wild-type and KKH SaCas9 induce a comparable number of off-target effects in human cells. KKH SaCas9 increased the targeting range of SaCas9 by nearly two to four times, enabling the targeting of sequences that cannot be altered by the wild-type nuclease. More generally, these results demonstrate the feasibility of relaxing PAM specificity to broaden the targeting range of Cas9 heterologous genes. The inventors' molecular evolution strategy does not require structural information or prior knowledge regarding specific residues in contact with PAM and should therefore be applicable to a wide range of Cas9 heterologous genes.

[0192] result

[0193] The inventors devised an unbiased genetic approach to engineer Cas9 variants with relaxed PAM recognition specificity that do not require structural information. The inventors tested this strategy using SaCas9, for which structural data was not available at the time the inventors began these studies. In the initial stages, the inventors sought to carefully estimate the PAM-interaction domain for SaCas9 by sequence comparison with structurally well-characterized SpCas9 (Jiang et al., Science 348, 1477-1481 (2015); Anders et al., Nature 513, 569-573 (2014); Jinek et al., Science (2014); Nishimasu et al., Cell (2014)). Although SpCas9 and SaCas9 were substantially different at the primary sequence level ( FIG. 21a, FIG. 29 Alignment of both using 10 additional heterologous genes made it possible to carefully define the expected PAM-interaction domain for SaCas9 ( Method for Example 3 ; Fig. 29 and 30 reference).

[0194] Because guanine at the third position of SaCas9 PAM is the base with the strictest specificity (Ran et al., Nature 520, 186-191 (2015)), the inventors randomly mutated the expected PI domain and, using the inventors' previously described bacterial cell-reference method (see Examples 1-2), obtained three other possible nucleotides at the third PAM position (i.e., NN[A / C / T]RRT PAM (NNHRRT (Sequence Identification No.: 44)); Fig. 31aAttempts were made to select mutants capable of cleaving the sites containing each of ). While one, but not all, of the surviving mutants from the selection for the sites containing NNARRT (SEQ ID: 43) and NNCRRT (SEQ ID: 47) PAM possessed the R1015H mutation, the inventors did not obtain any mutants from the selection for NNTRRT (SEQ ID: 48) PAM. These results strongly suggest that R1015 may participate in the recognition of guanine at the third position of SpCas9 PAM. Indeed, in the inventors' alignment, SpCas9 R1335 (a residue previously associated with the recognition of the third base position of PAM) Fig. 30 It was confirmed that R1015 of SaCas9 is located in the vicinity (see Examples 1-2; Anders, C., Niewoehner, O., Duerst, A. & Jinek, M. Structural basis of PAM-dependent target DNA recognition by the Cas9 endonuclease. Nature 513, 569-573 (2014)). Consistent with this, when the inventors tested in the inventors' bacterial selection system ( Fig. 31b It was confirmed that mutations in R1015 to alanine or glutamine substantially reduced SaCas9 activity at the target site containing NNGRRT (SEQ ID: 46) PAM ( Fig. 21b ). Alanine or glutamine substitution of other positively charged residues near R1015 did not have a potent effect on SaCas9 activity ( Fig. 21b, do 30 ).

[0195] The inventors' bacterial-based selection results also suggest that the R1015H mutation can at least partially attenuate the specificity of SaCas9 at the third position of PAM. However, when the inventors tested for a site having any nucleotide at the third position of NNNRRT PAM ( Fig. 21c It was confirmed that the R1015H single mutant exhibited lower-than-optimal activity in the inventors' previously described human cell-based EGFP disruption assay (Fu et al., Nat Biotechnol 31, 822-826 (2013); Reyon et al., Nat Biotechnol 30, 460-465 (2012)). Since this suggests that additional mutations may be required to increase or optimize the activity of the R1015H mutant in human cells, the inventors randomly mutated the region containing the expected PI domain of SaCas9, which also possesses the R1015Q mutant. Subsequently, the inventors selected variants from this library capable of cleaving target sites with each of the three different NNHRRT (SEQ ID: 44) PAMs using the inventors' bacterial selection system. The inventors used the R1015Q mutant because, unlike R1015H, it does not exhibit activity in bacteria ( Fig. 21b Although no surviving clones were observed upon selection for NNTRRT (sequence identification number: 48) PAM, selection with the R1015Q variant for NNARRT (sequence identification number: 43) or NNCRRT (sequence identification number: 47) produced mutations in E782, K929, and N968, and surprisingly mutated the Q of 1015 to H.

[0196] In combination with the selection results from wild-type SaCas9, the most frequent missense mutations identified across all selections were E782K, K929R, N968K, and R1015H ( Fig. 21d), this suggests that the combination of these mutations may allow for the efficient cleavage of a region containing A or C at the third position of the SaCas9 PAM. Accordingly, the inventors tested SaCas9 variants containing different combinations of these mutations using a human cell-based EGFP disruption assay with sgRNAs targeted to regions containing each of the four bases at the third position of the PAM (i.e., on the NNNRRT PAM). FIG. 21e, FIG. 32 The inventors found that variants having the triple mutant combinations E782K / N968K / R1015H and E782K / K929R / R1015H are highly active at the site containing NNNRRT PAM, whereas ( FIG. 21e, FIG. 32 It was confirmed that the quadruple mutant variant containing all four mutations (E782K / K929R / N968K / R1015H) has overall lower activity for these sites ( Fig. 32 ). The inventors selected E782K / N968K / R1015H (hereinafter referred to as the KKH variant) for further characterization and demonstrated, using the inventors' human cell-based EGFP disruption assay, that all three substitutions including the KKH variant are required for activity ( Fig. 21e ).

[0197] To more comprehensively define the PAM specificity of KKH and wild-type SaCas9, the inventors used the previously described bacterial cell-substance-depletion assay of the inventors (see Examples 1-2). Fig. 33 This method generated a Cas9 PAM specificity profile by confirming the relative splitting (and consequently depletion in bacterial cells) of DNA plasmids containing randomized PAM sequences, and the PAM depletion value after selection ( PPDVIt was quantified as ). The inventors performed site-depletion experiments on both wild-type and KKH SaCas9 using libraries having two different spacer sequences, each with 8 randomized bases, instead of PAM ( Fig. 33 ). Control experiments using catalytically inactive SaCas9 suggested that arbitrary PAM sequences were less depleted ( Fig. 34a This made it possible to establish a titer for statistically significant depletion with a PPDV of 0.794 ( Fig. 34b ). Previous experiments have shown that PAMs with a PPDV <0.2 in the inventors' bacterial site-depletion assay can be efficiently cleaved in the inventors' human cell-based EGFP destruction assay (see Examples 1-2). In the case of wild-type SaCas9, the most depleted PAM (based on the average PPDV obtained from the two libraries) was, as expected, four NNGRRT (SEQ ID: 46) PAMs ( Fig. 21f and Fig. 34c Interestingly, other PAMs with an average PPDV <0.1 included those having the NNGRRN (sequence identification number: 49) form ( Fig. 34d), this suggests that for some spacer sequences, the final position of PAM could not be fully identified as T in the inventors' bacterial-description assay (although previous reports demonstrated that thymine at the 6th position of PAM was most desirable by in vitro PAM depletion assays, ChIP-seq, and targeting of endogenous human sites (Ran, FA et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186-191 (2015))). In contrast, for the KKH variant, PAMs with an average PPDV <0.2 contained not only NNGRRT (SEQ ID: 46) PAM but also all four NNARRT (SEQ ID: 43), all four NNCRRT (SEQ ID: 47), and three of the four NNTRRT (SEQ ID: 48) PAMs ( FIG. 21f, FIG. 34c and 34e These results suggest that KKH SaCas9 is believed to have a wider PAM targeting range compared to its wild counterpart.

[0198] To evaluate the potent nature of the KKH SaCas9 variant in human cells, the inventors tested its activity against 55 different endogenous gene target sites containing various NNNRRT PAMs ( Fig. 22a ). The KKH variants demonstrated efficient activity across all sites with an average mutagenicity of 24.7%, and 80% of the sites (44 out of 55) demonstrated destruction exceeding 5%. Analysis of KKH SaCas9 activity across all 55 sites was based on the preference order for the third position of PAM (NN[G>A=C>T]RRT; Fig. 22b ) as well as the preference order for the 4th / 5th positions of PAM (NNN[AG>GG>GA>AA]T; Fig. 22c...was indicated. In accordance with this, the inventors observed differences in 16 possible combinations of the 3rd / 4th / 5th positions of the NNNRRT PAM ( Fig. 35a ). KKH SaCas9 has a spacer length in the range of 21–23 nucleotides ( Fig. 22d ), spacer sequence having variable GC content ( Fig. 35b PAM having variable GC content ( Fig. 35c It functioned efficiently for ). Sequence logos derived from sites split with low, medium, and high efficiency (mean mutagenic frequencies of 0–10%, 10–30%, and >30%, respectively) were the 4th of NNNRRT PAM. In addition to the 5th position, it showed little sequence preference across the entire target site and likely showed a slight preference for guanine at the 2nd PAM position on the highly efficient segmented site ( Fig. 35d ).

[0199] To demonstrate that KKH variants enable modifications of PAM that cannot be targeted by wild-type SaCas9, the inventors performed a direct comparison of these nucleases in human cells for sites harboring various NNNRRT PAMs. The inventors' EGFP disruption assay and evaluation of 16 sites using 16 endogenous human gene targets (each FIGS. 22e and 22fIt was suggested that KKH SaCas9 strongly modified target sites possessing NNNRRT PAM, whereas wild-type SaCas9 efficiently targeted only sites possessing NNGRRT (SEQ ID: 46) PAM. For all 24 sites possessing NNHRRT (SEQ ID: 44) PAM, the KKH variant induced mutagenicity at a substantially higher rate compared to wild-type SaCas9, while for the 8 sites possessing NNGRRT (SEQ ID: 46) PAM, KKH SaCas9 induced mutagenicity at a level comparable to or slightly lower than that of the wild type. Fig. 22e and 22f Taken together, these results demonstrate that the KKH variant can cleave the region containing NNNRRT PAM, thereby enabling the targeting of the region containing NNHRRT (sequence identification number: 44) PAM, which cannot be efficiently modified by wild-type SaCas9 in human cells.

[0200] To evaluate the effect of KKH mutations on the genome-wide specificity of SaCas9, the inventors used the GUIDE-seq (Genome-wide Unbiased Identification of DSBs Enabled by sequencing) method (Tsai, SQ et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 33, 187-197 (2015)) to directly compare the off-target profiles of wild-type and KKH SaCas9 with the same sgRNA. When tested using sgRNAs targeted to six endogenous human gene regions containing NNGRRT (SEQ ID: 46) PAM, the inventors observed that wild-type and KKH SaCas9 induced nearly identical GUIDE-seq tag integration rates and on-target cleavage frequencies for all six regions (each Fig. 36a and 36b In addition, wild-type and KKH SaCas9 induced a similar number of off-target site mutations by each of the six sgRNAs ( Fig. 23a and 23b In the case of the KKH variant, the off-target site was generally attached to the NNNRRT PAM motif, while in the case of wild-type SaCas9, the off-target site was attached to the NNGRR[T>G] motif ( Fig. 22b ). For one of the sgRNAs that induced the largest number of off-target sites among the six tested sgRNAs, the inventors observed a similar number of off-target sites by the wild type and KKH SaCas9. However, as could be expected from their given different PAM specificities, the off-target sites overlapped only partially between the wild type and KKH SaCas9 ( Fig. 23b and 23cWhile the inventors support the use of the KKH variant to target sites containing NNGRRT (sequence identification number: 46) PAM (since wild-type SaCas9 may present higher on-target activity for these sites compared to KKH), these results suggest that only KKH SaCas9 divides off-target sites containing the expected PAM and induces a number of off-target sites comparable to those observed in wild-type SaCas9 overall.

[0201] To further test the genome-wide specificity of KKH SaCas9, the inventors tested five additional sgRNAs targeted to a region containing NNHRRT (Sequence Identification No.: 44) PAM ( Fig. 23d and 23e The number of off-target sites detected by GUIDE-seq was generally low (comparable to the number observed in wild-type SpCas9 and SpCas9 variants in previously published experiments (see Examples 1-2 (Tsai, SQ et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nat Biotechnol 33, 187-197 (2015)), indicated potential DNA- and RNA-bulge off-targets (Lin, Y. et al. CRISPR / Cas9 systems have off-target activity with insertions or deletions between target DNA and guide RNA sequences. Nucleic Acids Res 42, 7473-7485 (2014)), and contained expected PAM sequences). Taken together, the inventors' experiments showed similar specificity across the genomes of wild-type and KKH SaCas9, and overall As determined by GUIDE-seq, a low number of off-target mutations were presented in human cells.

[0202] Although wild-type SaCas9 is the optimal choice for targeting NNGRRT (SEQ ID: 46) PAM, the KKH SaCas9 variant described herein by the inventors can potently target regions containing NNARRT (SEQ ID: 43) and NNCRRT (SEQ ID: 47) PAM and has a reasonable success rate for regions containing NNTRRT (SEQ ID: 48) PAM. Accordingly, the inventors have cautiously estimated that the KKH variant increases the targeting range of SaCas9 in random DNA sequences by nearly two to four times, thereby improving the possibility of more extensive use of SaCas9 in various different applications requiring highly precise targeting. Using GUIDE-seq, the inventors demonstrated that KKH SaCas9 induces a similar number of off-target mutations as wild-type SaCas9 when targeted to the same regions containing NNGRRT (SEQ ID: 46) PAM. Furthermore, when targeted to a region possessing NNHRRT (sequence identification number: 44) PAM, KKH SaCas9 induced only a small number of off-target mutations. Even if KKH SaCas9 recognizes a modified PAM sequence compared to wild-type SaCas9, the inventors' findings are not surprising at all if the total combined length of the protospacer and PAM is sufficiently long (24 to 26 bp) so that the KKH variant is reasonably orthogonal to the human genome. Additionally, it is possible that modifications in PAM recognition can improve specificity by altering the Cas9 / sgRNA interaction energy with its target site (the mechanism previously proposed for improved specificity of truncated sgRNA (Fu, Y., Sander, JD, Reyon, D., Cascio, VM & Joung, JK Improving CRISPR-Cas nuclease specificity using truncated guide RNAs.Nat Biotechnol 32, 279-284 (2014)) or similar to the D1135E SpCas9 mutant (see Examples 1-2).

[0203] Example 4. Improvement of the activity of the SpCas9-VQR variant

[0204] Since SpCas9-VQR variants have an affinity for NGAN PAM: NGAG > NGAA = NGAT > NGAC, the inventors sought to select derivative variants having improved activity for NGAH PAM (where H = A, C, or T). Selection using the R1335Q library (containing randomly mutated PI domains) for cells containing target sites with NGAA, NGAT, or NGAC PAM made it possible to sequence additional clones containing mutations that convey altered PAM specificity. The sequences of these clones revealed additional mutations that could be important in altering PAM specificity for NGAA, NGAT, or NGAC PAM.

[0205] Based on the results of these selections, VQR variants and 24 other derivative variants were tested for the NGAG, NGAA, NGAT, and NGAC PAM sites in bacteria. These numerous derivative variants survived more than VQR variants for the NGAH PAM site, and most of them contained the G1218R mutation ( Table 7 and Fig. 24 ).

[0206]

[0207] Since results from bacterial screens demonstrated that some of these additional mutations improved activity toward the NGAH PAM site, the inventors tested some of the best candidates in human cells using an EGFP disruption assay. The inventors observed that numerous of these variants, including VRQR, NRQR, and YRQR variants, outperformed the VQR variant in targeting the NGAH site (Table 8 and Fig. 25). The main difference between these clones and the VQR variant is that they contain the G1218R mutation.

[0208]

[0209] Since the VRQR variant was considered to be the most potent among those tested, the inventors compared its activity to that of VQR at nine different endogenous sites in human cells (two sites for each of NGAA, NGAC, NGAT, and NGAG PAM, and one site for NGCG PAM). From this data, it was revealed that the VRQR variant outperformed the VQR variant at all sites tested in human cells (Fig. 26).

[0210] After demonstrating that VRQR variants possessed improved activity compared to VQR variants, the inventors sought to determine whether adding additional substitutions could further improve activity. Since the inventors observed additional mutations located very close to the PAM interaction pocket of SpCas9 from selection, they added a subset of these mutations to VQR and VRQR variants and screened them for regions containing NGAG, NGAA, NGAT, and NGAC PAM in bacteria (Table 9 and Fig. 27). Numerous derivative variants appeared to have higher activity for the NGAT and NGAC PAM regions, and thus the inventors tested these variants in human cells. The inventors tested additional variants containing mutations added to the VQR or VRQR background in a human cell EGFP disruption assay. These experiments again showed that VRQR has more potent activity against NGAH PAM compared to VQR variants, and that additional mutations in the VRQR backbone are beneficial.

[0211]

[0212] Taken together, these results suggest that including additional mutations in SpCas9-VQR variants can improve activity at sites containing NGAN PAM, specifically NGAH PAM.

[0213] References

[0214]

[0215]

[0216] Other embodiments

[0217] Although the present invention has been described in detail, it should be understood that such description is intended to illustrate the invention as defined by the scope of the appended claims and is not intended to limit the scope thereof. Other aspects, advantages, and variations are within the scope of the following claims.

Claims

Claim 1 A variant Staphylococcus aureus comprising an amino acid sequence having at least 95% sequence identity with respect to the amino acid sequence of Sequence Identification No. 2 ( Staphylococcus aureus ) Isolated nucleic acid encoding a protein including the Cas9 (SaCas9) protein, wherein in the variant SaCas9 protein, the amino acid corresponding to E782 of SEQ ID NO:2 is mutated to K and / or the amino acid corresponding to N968 of SEQ ID NO:2 is mutated to K and / or the amino acid corresponding to R1015 of SEQ ID NO:2 is mutated to Q or H, the variant SaCas9 protein binds to guide RNA and target DNA, wherein any additional difference between the amino acid sequence of the variant SaCas9 protein and the amino acid sequence of SEQ ID NO:2 arises from a conservative substitution of the amino acid sequence of SEQ ID NO:2 or a substitution with alanine. Claim 2 A vector comprising the isolated nucleic acid of claim 1. Claim 3 A vector in which the isolated nucleic acid is operably linked to one or more regulatory domains to express a SaCas9 protein. Claim 4 Isolated host cell containing the nucleic acid of claim 1. Claim 5 The isolated nucleic acid of claim 1 further comprises a heterofunctional domain fused to a variant SaCas9 protein with an optional intervening linker, wherein the linker does not interfere with the activity of the fused protein. Claim 6 In paragraph 5, the isolated nucleic acid in which the heterogeneous functional domain is a transcriptional activation domain. Claim 7 In claim 6, the isolated nucleic acid in which the transcriptional activation domain is from VP64 or NF-κB p65. Claim 8 In paragraph 5, isolated nucleic acid in which the heterogeneous functional domain is a transcription silencer or a transcription inhibitory domain. Claim 9 In paragraph 8, isolated nucleic acid in which the transcription inhibitory domain is a Krüpel-associated box (KRAB) domain, an ERF inhibitory factor domain (ERD), or an mSin3A interaction domain (SID). Claim 10 In paragraph 8, the isolated nucleic acid in which the transcription silencer is heterochromatin protein 1 (HP1). Claim 11 In paragraph 5, isolated nucleic acid in which a heterogeneous functional domain is an enzyme that modifies the methylation state of DNA. Claim 12 In paragraph 11, isolated nucleic acid in which the enzyme modifying the methylation state of DNA is a DNA methyltransferase (DNMT) or 10-11-translocation (TET) protein. Claim 13 In paragraph 12, the isolated nucleic acid in which the TET protein is TET1. Claim 14 In paragraph 5, isolated nucleic acid in which the heterogeneous functional domain is an enzyme that modifies histone subunits. Claim 15 In paragraph 14, isolated nucleic acid in which the enzyme modifying the histone subunit is histone acetyltransferase (HAT), histone deacetylase (HDAC), histone methyltransferase (HMT), or histone demethylase. Claim 16 In paragraph 5, isolated nucleic acid in which the heterogeneous functional domain is a biological tether. Claim 17 In paragraph 16, isolated nucleic acid in which the biological tether is MS2, Csy4, or lambda N protein. Claim 18 In paragraph 5, the isolated nucleic acid in which the heterogeneous functional domain is FokI. Claim 19 A composition comprising: a nucleic acid encoding a protein comprising a variant SaCas9 protein having an amino acid sequence having at least 95% sequence identity with respect to the amino acid sequence of SEQ ID NO:2, wherein in the variant SaCas9 protein, the amino acid corresponding to E782 of SEQ ID NO:2 is mutated to K and / or the amino acid corresponding to N968 of SEQ ID NO:2 is mutated to K and / or the amino acid corresponding to R1015 of SEQ ID NO:2 is mutated to Q or H, wherein the variant SaCas9 protein binds to a guide RNA and a target DNA, wherein any additional difference between the amino acid sequence of the variant SaCas9 protein and the amino acid sequence of SEQ ID NO:2 arises from a conservative substitution of the amino acid sequence of SEQ ID NO:2 or a substitution with alanine; and a nucleic acid encoding a guide RNA that directs the SaCas9 protein to a target genome sequence. Claim 20 A composition according to claim 19, wherein the protein encoded by the isolated nucleic acid further comprises a heterofunctional domain fused to a variant SaCas9 protein with an optional intervening linker, wherein the linker does not interfere with the activity of the fused protein. Claim 21 A composition according to claim 20, wherein the heterogeneous functional domain is a transcriptional activation domain. Claim 22 A composition according to claim 21, wherein the transcriptional activation domain is from VP64 or NF-κB p65. Claim 23 A composition according to claim 20, wherein the heterogeneous functional domain is a transcription silencer or a transcription inhibitory domain. Claim 24 A composition according to claim 23, wherein the transcription inhibitory domain is a Krüpel-associated box (KRAB) domain, an ERF inhibitory factor domain (ERD), or an mSin3A interaction domain (SID). Claim 25 A composition according to claim 23 in which the transcription silencer is heterochromatin protein 1 (HP1). Claim 26 A composition according to claim 20, wherein the heterogeneous functional domain is an enzyme that modifies the methylation state of DNA. Claim 27 A composition according to claim 26, wherein the enzyme that modifies the methylation state of DNA is a DNA methyltransferase (DNMT) or 10-11-translocation (TET) protein. Claim 28 In claim 27, a composition in which the TET protein is TET1. Claim 29 A composition according to claim 20, wherein the heterogeneous functional domain is an enzyme that modifies histone subunits. Claim 30 A composition according to claim 29, wherein the enzyme modifying the histone subunit is histone acetyltransferase (HAT), histone deacetylase (HDAC), histone methyltransferase (HMT), or histone demethylase. Claim 31 A composition in which the heterogeneous functional domain is a biological tether in claim 20. Claim 32 A composition according to claim 31 in which the biological tether is MS2, Csy4, or lambda N protein. Claim 33 A composition according to claim 20, wherein the heterogeneous functional domain is FokI. Claim 34 The isolated nucleic acid according to claim 1, further comprising at least one of the following: the amino acid corresponding to E735 of sequence identification number: 2 in the variant SaCas9 protein is mutated to K; the amino acid corresponding to K929 of sequence identification number: 2 is mutated to R; the amino acid corresponding to A1021 of sequence identification number: 2 is mutated to T; and the amino acid corresponding to K1044 of sequence identification number: 2 is mutated to N. Claim 35 The isolated nucleic acid according to claim 1, wherein the variant SaCas9 protein further comprises at least one of the following mutations: a mutation at the amino acid corresponding to D10 of sequence identification number: 2, a mutation at the amino acid corresponding to D556 of sequence identification number: 2, a mutation at the amino acid corresponding to H557 of sequence identification number: 2, and a mutation at the amino acid corresponding to N580 of sequence identification number:

2. Claim 36 In paragraph 35, isolated nucleic acid in which the amino acid corresponding to D10 in the variant SaCas9 protein is mutated to A and / or the amino acid corresponding to D556 is mutated to A and / or the amino acid corresponding to H557 is mutated to A and / or the amino acid corresponding to N580 is mutated to A. Claim 37 In paragraph 35, an isolated nucleic acid in which (i) the amino acid corresponding to D10 is mutated to A and the amino acid corresponding to H557 is mutated to A, or (ii) the amino acid corresponding to D10 is mutated to A, the amino acid corresponding to D556 is mutated to A, the amino acid corresponding to H557 is mutated to A, and the amino acid corresponding to N580 is mutated to A. Claim 38 A composition according to claim 19, further comprising at least one of the following: a variant SaCas9 protein in which the amino acid corresponding to E735 of sequence identification number: 2 is mutated to K, the amino acid corresponding to K929 of sequence identification number: 2 is mutated to R, the amino acid corresponding to A1021 of sequence identification number: 2 is mutated to T, and the amino acid corresponding to K1044 of sequence identification number: 2 is mutated to N. Claim 39 A composition according to claim 19, further comprising at least one of a mutation in the amino acid corresponding to D10 of sequence identification number: 2, a mutation in the amino acid corresponding to D556 of sequence identification number: 2, a mutation in the amino acid corresponding to H557 of sequence identification number: 2, and a mutation in the amino acid corresponding to N580 of sequence identification number: 2 in the variant SaCas9 protein. Claim 40 A composition according to claim 39, wherein in the variant SaCas9 protein, the amino acid corresponding to D10 is mutated to A and / or the amino acid corresponding to D556 is mutated to A and / or the amino acid corresponding to H557 is mutated to A and / or the amino acid corresponding to N580 is mutated to A. Claim 41 A composition according to claim 39, wherein in the variant SaCas9 protein, (i) the amino acid corresponding to D10 is mutated to A and the amino acid corresponding to H557 is mutated to A, or (ii) the amino acid corresponding to D10 is mutated to A, the amino acid corresponding to D556 is mutated to A, the amino acid corresponding to H557 is mutated to A, and the amino acid corresponding to N580 is mutated to A. Claim 42 delete Claim 43 delete

Images