High-fidelity restriction endonucleases

Abstract

Compositions and methods for enzymes with altered properties are provided, which include a systematic approach to mutagenesis and a screening assay that allows selection of desired proteins. Embodiments of the method are particularly suitable for altering specific properties of restriction endonucleases, such as star activity. The composition includes a restriction endonuclease with reduced star activity, as defined by an overall fidelity index improvement factor.

Description

[0001] This application is a divisional application of a patent application with application date of July 14, 2008, application number 200880106647.5, and title “High-fidelity restriction endonuclease”. Background Art

[0002] Restriction endonucleases are enzymes that cleave double-stranded DNA in a sequence-specific manner (Roberts, RJ, Proc Natl Acad Sci USA, 102: 5905-5908 (2005); Roberts, et al., Nucleic Acids Res, 31: 1805-1812 (2003); Roberts, et al., Nucleic Acids Res, 33: D230-232 (2005); Alves, et al., Restriction Endonucleases "Protein Engineering of Restriction Enzymes," ed. Pingoud, Springer-Verlag Berlin Heidelberg, New York, 393-407 (2004)). They are ubiquitous in prokaryotes (Raleigh, et al., Bacterial Genomes Physical Structure and Analysis , Ch.8, eds.De Bruijin, et al., Chapman & Hall, New York, 78-92 (1998)), where they form part of a restriction modification system, which is mainly composed of endonucleases and methyltransferases. The cognate methyltransferase methylates a specific sequence identical to the sequence recognized by its paired endonuclease and confers resistance to endonuclease cleavage on the modified DNA so that the host DNA can be properly protected. However, when there is invasion of foreign DNA, especially phage DNA, the foreign DNA will be degraded before it can be completely methylated. The main biological function of the restriction modification system is to protect the host from phage infection (Arber, Science, 205: 361-365 (1979)). Other functions have also been proposed, such as involvement in recombination and transposition (Carlson, et al., Mol Microbiol, 27:671-676 (1998); Heitman, Genet Eng (NY), 15:57-108 (1993); McKane, et al., Genetics, 139:35-43 (1995)).

[0003] The specificity of the approximately 3,000 known restriction endonucleases for their more than 250 different target sequences can be considered their most interesting property. After the discovery of the sequence-specific nature of the first restriction endonucleases (Danna, et al., Proc Natl Acad Sci USA, 68:2913-2917 (1971); Kelly, et al., J Mol Biol, 51:393-409 (1970)), it did not take scientists long to discover that certain restriction endonucleases cleave sequences similar to, but not identical to, their restricted recognition sequences under non-optimal conditions (Polisky, et al., Proc Natl Acad Sci USA, 72:3310-3314 (1975); Nasri, et al., Nucleic Acids Res, 14:811-821 (1986)). This relaxed specificity is called the star activity of the restriction endonuclease. It has been proposed that the water-mediated interaction between the restriction endonuclease and DNA is the main difference between specificity complexes and star complexes (Robinson, et al., J Mol Biol, 234: 302-306 (1993); Robinson, et al., Proc Natl Acad Sci USA, 92: 3444-3448 (1995), Sidorova, et al., Biophys J, 87: 2564-2576 (2004)).

[0004] Star activity is a problem in molecular biology reactions. Star activity introduces unwanted cleavages in cloning vectors or other DNA. In situations such as forensic applications, where certain DNA substrates need to be cleaved by restriction endonucleases to produce unique fingerprints, star activity will alter the characteristics of the cleavage pattern, thereby complicating the analysis. Avoiding star activity is also critical in applications such as strand displacement amplification techniques (Walker, et al., Proc Natl Acad Sci USA, 89:392-396 (1992)) and serial analysis of gene expression (Velculescu, et al., Science, 270:484-487 (1995)). SUMMARY OF THE INVENTION

[0005] In one embodiment of the present invention, a composition is provided, comprising a restriction endonuclease having at least one artificially introduced mutation and an overall fidelity index (FI) improvement factor of at least 2, wherein the restriction endonuclease can cleave a substrate in a predetermined buffer with a cleavage activity at least similar to the cleavage activity of the restriction endonuclease in the absence of the artificially introduced mutation, wherein the artificially introduced mutation is the product of at least one of targeted mutation, saturation mutagenesis, or a mutation introduced by a PCR amplification process.

[0006] In a further embodiment of the invention, at least one of the artificially introduced mutations is a targeted mutation resulting from replacement of a naturally occurring residue with an oppositely charged residue. Alanine or phenylalanine may replace a naturally occurring residue at the target site.

[0007] In a further embodiment of the present invention, a composition of the type described above comprises a restriction enzyme selected from the group consisting of BamHI, EcoRI, ScaI, SalI, SphI, PstI, NcoI, NheI, SspI, NotI, SacI, PvuII, MfeI, HindIII, SbfI, EagI, EcoRV, AvrII, BstXI, PciI, HpaI, AgeI, BsmBI, BspQI, SapI, KpnI and BsaI, in which no artificially introduced mutations are present.

[0008] Further embodiments of the present invention include the components listed in Table 4.

[0009] In a further embodiment of the present invention, there is provided a DNA encoding any one of the enzymes listed in Table 4, a vector comprising the DNA, and a host cell expressing a protein from the vector.

[0010] In one embodiment of the present invention, a method having the following steps is provided: (a) identifying which amino acid residues in the amino acid sequence of a restriction endonuclease having star activity are charged amino acids; (b) mutating one or more codons encoding one or more charged residues in a gene sequence encoding the restriction endonuclease; (c) generating a library of gene sequences having one or more different codon mutations at different charged residues; (d) obtaining a set of proteins expressed by the mutated gene sequence; and (e) determining the FI and cleavage activity of each expressed protein in a predetermined buffer.

[0011] One embodiment of the method comprises the step of determining an overall FI improvement factor for proteins belonging to the protein group in a defined buffer set, wherein, for example, the buffer set comprises NEB1, NEB2, NEB3 and NEB4 buffers.

[0012] One embodiment of the method comprises the steps described above and additionally, in the same step or a subsequent step in which the codon for the charged amino acid is mutated, the codon encoding the hydroxylated amino acid or the amide amino acid is mutated.

[0013] In one embodiment of the present invention described above, codons other than tyrosine are mutated to alanine, and tyrosine is mutated to phenylalanine.

[0014] In a further embodiment, the overall FI improvement factor is improved using saturation mutagenesis of one or more mutated codons. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] for Figure 1-32 :

[0016] The *marker indicates that the lane to its left contains the lowest concentration of enzyme where asterisk activity is observed.

[0017] The # marker indicates an incompletely cleaved lane, which is adjacent to and to the right of the lane containing enzyme at a concentration sufficient to completely cleave the substrate.

[0018] Grey triangles represent serial decreases in restriction endonuclease concentration.

[0019] "U" stands for enzyme unit.

[0020] exist Figure 1-32 In each reaction described in the Figures, the reaction mixture contained a 3 μl volume of - unless otherwise specified - buffer from New England Biolabs, Inc. (NEB), Ipswich, MA (see Table 1 and NEB catalog), a 3 μl volume of - unless otherwise specified - the indicated restriction endonuclease in diluent from NEB, Ipswich, MA (see Table 1 and NEB catalog), and a variable volume of the indicated substrate (containing 0.6 vg) and water to bring the reaction mixture to a total volume of 30 μl. Reactions were performed at 37°C with an incubation time of 1 hour. Results were analyzed on a 0.8% agarose gel. When the total volume of the reaction mixture, amount of substrate, reaction temperature or incubation time differed from those described above, the values ​​are provided in the legend to the figures.

[0021] exist Figure 1, 5, 8, 11-18, and 20-32, theoretical digestion profiles are provided on the right side of the gel. Those substrates that have only one restriction endonuclease site should digest from the supercoiled form to a linear band.

[0022] Figure 1 Shown is the determination of the FI of wild-type (WT) ScaI by digesting 1.2 μl of λ DNA substrate (0.6 μg) using two-fold serial dilutions of a WT ScaI (1,200 U) preparation in NEB3 buffer and examining the digestion products on an agarose gel. The highest concentration of restriction endonuclease without star activity is shown with a solid arrow; and the minimum concentration that causes complete digestion of the substrate is shown with an open arrow.

[0023] Figure 2A -D shows the results of digestion of 0.5 μl of pUC19 substrate (0.5 μg) for 1 hour with WT BamHI or BamHI (E86P) enzymes that were serially diluted three-fold with diluent A at a starting concentration of 172 U or 512 U. The middle lane is the NEB 1 kb marker (New England Biolabs, Inc. (NEB), Ipswich, MA).

[0024] Figure 2A Results using NEB1 buffer are shown.

[0025] Figure 2B Results using NEB2 buffer are shown.

[0026] Figure 2C Results using NEB3 buffer are shown.

[0027] Figure 2D Results using NEB4 buffer are shown.

[0028] Figure 3A -B shows the comparison of BamHI(E86P) activity during two time periods using 0.6 μl of pBR322 substrate (which contains only 1 BamHI cleavage site) in NEB2 buffer, using an initial concentration of 600 U of the enzyme in 2-fold serial dilutions with diluent A.

[0029] Figure 3A Results for 1 hour are shown.

[0030] Figure 3B Results for 14 hours are shown.

[0031] Figure 4A-B shows on agarose gel the cleavage of 0.6 μl of pBR322 substrate using BamHI-HF (E163A / E167T) serially diluted 2-fold with diluent A after 14 hours of incubation in two different buffers.

[0032] Figure 4A Results using NEB2 buffer and 600 U initial enzyme concentration are shown.

[0033] Figure 4B Results using NEB1 buffer and 2,400 U of initial enzyme concentration are shown.

[0034] Figure 5A -B shows a comparison of the cleavage reaction using BamHI-HF and WT BamHI in NEB4 buffer. The reaction was performed in NEB4 buffer using 1.2 μl of lambda DNA substrate in a 2-fold serial dilution using diluent A.

[0035] Figure 5A A starting concentration of 1,200 U of WT BamHI with FI equal to 4 is shown.

[0036] Figure 5B A starting concentration of 2,400 U of BamHI-HF is shown, with FI ≥ 4000.

[0037] Fig. 6A -D shows a comparison of WT EcoRI and EcoRI (K62A) in NEB 1-4 buffer at a 3-fold serial dilution using NEB diluent C. The reaction mixture contained 2 μl of lambda DNA substrate (1 μg) in NEB 1-4 buffer.

[0038] Fig. 6A Shown are the cleavage results of 120 U WT EcoRI and 240 U EcoRI(K62A) in NEB2 buffer after two-fold serial dilution.

[0039] Figure 6B Shown are the cleavage results of 120 U WT EcoRI and 240 U EcoRI(K62A) in NEB4 buffer after two-fold serial dilution.

[0040] Figure 6C Shown are the cleavage results of 60 U WT EcoRI and 120 U EcoRI(K62A) in NEB1 buffer after two-fold serial dilution.

[0041] Fig.6D Shown are the cleavage results of 120 U WT EcoRI and 60 U EcoRI(K62A) in NEB3 buffer after two-fold serial dilution.

[0042] Fig. 7A -C shows the results of cleavage with two different EcoRI mutants and WT EcoRI. Shown are digestions over 10 hours with 100,000 U enzyme and its 10-fold serial dilutions in diluent C, performed in various buffers using 0.6 μl of Litmus28 substrate. There is only one EcoRI cleavage site in the Litmus28 substrate.

[0043] Fig. 7A : EcoRI mutant K62E in NEB4 buffer.

[0044] Figure 7B : EcoRI mutant K62A in NEB4 buffer.

[0045] Figure 7C : WT EcoRI in EcoRI buffer (see NEB catalog 2007-8).

[0046] Fig. 8A -B shows a comparison of EcoRI-HF and WT EcoRI in NEB4 buffer. The reaction used 1.2 μl of lambda DNA substrate in a 2-fold serial dilution using diluent C.

[0047] Fig. 8A : WT EcoRI with 19,200U starting concentration showed FI=4 in NEB4 buffer.

[0048] Figure 8B : In NEB4 buffer, UEcoRI-HF with a starting concentration of 38,400 showed FI = 16,000.

[0049] Fig. 9A -B shows a comparison of serial dilutions of WT ScaI (4.8U), ScaI (H193A) (9.6U), ScaI (S201F) (19.2U) and ScaI (H193A / S201F) (19.2U). Each sample was initially diluted 1 / 10 and then serially diluted 2-fold in NEB2 buffer with the indicated percentage of glycerol. Each reaction mixture contained 2 μl of λ DNA substrate (1 μg).

[0050] Fig. 9A : 5% Glycerol.

[0051] Fig. 9B : 37% Glycerol.

[0052] Fig. 10A-D shows a comparison of WT ScaI and ScaI-HF (H193A / S201F). The enzymes (unit concentrations as specified) were each diluted 2.5-fold serially with diluent A. The reaction mixture contained 2 μl of lambda DNA substrate and NEB 1-4 buffer.

[0053] Fig. 10A : Cleavage in NEB2 buffer.

[0054] Fig. 10B : Cleaved in NEB4 buffer.

[0055] Fig. 10C : Cleavage in NEB1 buffer.

[0056] Fig. 10D : Cleavage in NEB3 buffer.

[0057] Fig.11A -H shows FI determination for SalI-HF and WT SalI. Both enzymes were diluted in 2-fold serial dilutions using diluent A. The reaction mixture contained 2 μl of HindIII-digested λ DNA substrate.

[0058] Fig.11A , B, C, and D show serial dilutions of 1,200 U, 1,200 U, 300 U, and 1,200 U of SalI-HF, which showed FI ≥ 1,000, FI ≥ 2,000, FI ≥ 500, and FI ≥ 2,000 in NEB 1, 2, 3, and 4 buffers, respectively.

[0059] Fig.11E , F, G, and H show serial dilutions of 19.2 U, 150 U, 9,600 U, and 38.4 U of WT SalI, which showed FI=8, FI=1, FI=32, and FI=1 in NEB1, 2, 3, and 4 buffers, respectively.

[0060] Fig. 12A -B shows the results of a 2-fold serial dilution of SphI-HF (19,200 U) in diluent A and WT SphI (143,600 U) in diluent B reacted with 1.2 μl of λ DNA substrate in NEB 4 buffer.

[0061] Fig. 12A Cleavage by WT SphI is shown.

[0062] Fig. 12B Cleavage of SphI-HF is shown.

[0063] Fig.13A-B shows cleavage of 1.2 μl of λ DNA substrate using 2-fold serial dilutions of PstI-HF (300 U and 150 U) and 2-fold serial dilutions of WT PstI (2,400 U and 4,800 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent C.

[0064] Fig.13A Cleavage of Pstl-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0065] Fig. 13B Cleavage by WT Pstl in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0066] Fig.14A -B shows cleavage of 1.2 μl of lambda DNA substrate using 2-fold serial dilutions of NcoI-HF (4,800 U and 600 U) and 2-fold serial dilutions of WT NcoI (4,800 U and 1,200 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0067] Fig.14A Cleavage of NcoI-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0068] Fig. 14B Cleavage by WT Ncol in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0069] Fig.15A -B shows cleavage of 1.2 μl pXba DNA substrate using 2-fold serial dilutions of NheI-HF (287,200 U and 76.8 U) and 2-fold serial dilutions of WT NheI (9,600 U and 300 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0070] Fig.15A Cleavage of NheI-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0071] Fig. 15B Cleavage by WT NheI in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0072] Fig.16A-B shows cleavage of 1.2 μl of lambda DNA substrate using 2-fold serial dilutions of SspI-HF (9,600 U and 38.4 U) and 2-fold serial dilutions of WT SspI (19,200 U and 19,200 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent C.

[0073] Fig.16A Cleavage by SspI-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0074] Fig. 16B Cleavage by WT SspI in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0075] Fig.17A -B shows cleavage of 1.2 μl of pXba DNA substrate using 2-fold serial dilutions of NotI-HF (287,200 U and 19,200 U) and 2-fold serial dilutions of WT NotI (19,200 U and 76,800 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent C.

[0076] Fig.17A Cleavage of Notl-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0077] Fig. 17B Cleavage by WT Notl in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0078] Fig.18A -B shows cleavage of 1.2 μl of pXba DNA substrate using 2-fold serial dilutions of Sad-HF (4,800 U and 76.8 U) and 2-fold serial dilutions of WT Sad (19,200 U and 1200 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0079] Fig.18A Cleavage of Sad-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0080] Fig.18B Cleavage by WT Sad in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0081] Fig.19A-B shows cleavage of 0.6 μl of pBR322 DNA substrate using 2-fold serial dilutions of PvuII-HF (9,600 U and 19.2 U) and 2-fold serial dilutions of WT PvuII (19,200 U and 300 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0082] Fig.19A Cleavage of PvuII-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0083] Fig.19B Cleavage by WT PvuII in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0084] Fig. 20A -B shows cleavage of 1.2 μl of λ DNA substrate using 2-fold serial dilutions of MfeI-HF (300 U and 19.2 U) and 2-fold serial dilutions of WT MfeI (2,400 U and 38.4 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0085] Fig. 20A Cleavage of MfeI-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0086] Fig. 20B Cleavage by WT Mfel in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0087] Fig.21A -B shows cleavage of 1.2 μl of λ DNA substrate using 2-fold serial dilutions of HindIII-HF (4,800 U and 1,200 U) and 2-fold serial dilutions of WT HindIII (9,600 U and 4,800 U) in NEB3 and NEB4 buffers, respectively. Serial dilutions were performed in diluent A.

[0088] Fig.21A Cleavage by Hindlll-HF in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0089] Fig. 21B Cleavage by WT HindIII in NEB4 buffer (upper panel) and NEB3 buffer (lower panel) is shown.

[0090] Fig.22A-B shows cleavage of 1.2 μl of λ DNA substrate using diluent C in 2-fold serial dilutions of SbfI-HF (starting concentration: 76,800 U) and WT SbfI (starting concentration: 76,800 U) in NEB4 buffer.

[0091] Fig.22A Cleavage by WT SbfI is shown.

[0092] Fig. 22B Cleavage by SbfI-HF is shown.

[0093] Fig.23A -B shows cleavage of 1.2 μl of pXba DNA substrate using 2-fold serial dilutions of EagI-HF (1,200 U and 600 U) and 2-fold serial dilutions of WT EagI (150 U and 38.4 U) in NEB2 and NEB1 buffers, respectively, using diluent C.

[0094] Fig.23A Cleavage of EagI-HF in NEB2 buffer (upper panel) and NEB1 buffer (lower panel) is shown.

[0095] Fig. 23B Cleavage by WT EagI in NEB2 buffer (upper panel) and NEB1 buffer (lower panel) is shown.

[0096] Fig.24A -B shows the cleavage of 1.2 μl of pXba DNA substrate using diluent A in a 2-fold serial dilution of EcoRV-HF (starting concentration: 38,400 U) and WT EcoRV (starting concentration: 2400 U) in NEB4 buffer.

[0097] Fig.24A Cleavage by WT EcoRV is shown.

[0098] Fig. 24B Cleavage by EcoRV-HF is shown.

[0099] Fig.25A -B shows cleavage of 1.2 μl T7 DNA substrate using diluent A in 2-fold serial dilutions of AvrII-HF (starting concentration: 1,200 U) and WT AvrII (starting concentration: 1,200 U) in NEB4 buffer.

[0100] Fig.25A Cleavage by WT AvrII is shown.

[0101] Fig.25B Cleavage by AvrII-HF is shown.

[0102] Fig.26A -B shows the cleavage of 1.2 μl of λ DNA substrate in NEB4 buffer by 2-fold serial dilutions of BstXI-HF (starting concentration: 300 U) and WT BstXI (starting concentration: 38.4 U) in diluent A. The reaction was performed at 55°C.

[0103] Fig.26A Cleavage by WT BstXI is shown.

[0104] Fig.26B Cleavage by BstXI-HF is shown.

[0105] Fig.27A -B shows the cleavage of 1.2 μl of pXba DNA substrate using diluent A in 2-fold serial dilutions of PciI-HF (starting concentration: 600 U) and WT PciI (starting concentration: 300 U) in NEB4 buffer.

[0106] Fig.27A Cleavage by WT Pcil is shown.

[0107] Fig.27B Cleavage by PciI-HF is shown.

[0108] Fig.28A -B shows cleavage of 1.2 μl of λ DNA substrate using diluent A in 2-fold serial dilutions of Hpal-HF (starting concentration: 4,800 U) and WT Hpal (starting concentration: 9,600 U) in NEB2 buffer.

[0109] Fig.28A Cleavage by WT HpaI is shown.

[0110] Fig.28B Cleavage by HpaI-HF is shown.

[0111] Fig.29A -B shows the cleavage of 1.2 μl of pXba DNA substrate using diluent C in 2-fold serial dilutions of AgeI-HF (starting concentration: 600 U) and WT AgeI (starting concentration 600 U) in NEB4 buffer.

[0112] Fig.29A Cleavage by WT AgeI is shown.

[0113] Fig.29B Cleavage by AgeI-HF is shown.

[0114] Fig. 30A-B shows the cleavage of 1.2 μl of λ DNA substrate in NEB4 buffer using 2-fold serial dilutions of BsmBI-HF (starting concentration: 300 U) and WT BsmBI (starting concentration 4,800 U) using diluent C. The reaction was performed at 55°C.

[0115] Fig. 30A Cleavage by WT BsmBI is shown.

[0116] Fig. 30B Cleavage by BsmBI-HF is shown.

[0117] Fig.31A -B shows the DNA sequence (SEQ ID NO: 1) and protein sequence (SEQ ID NO: 2) of MluCIM, used for EcoRI and MfeI expression, respectively.

[0118] Fig.32 The DNA sequence of Hpy166IIM (SEQ ID NO: 3) is shown for expression of SalI.

[0119] Fig.33A -B show the DNA sequence (SEQ ID NO: 4) and protein sequence (SEQ ID NO: 5) of MfeI, respectively.

[0120] Fig.34A -B show the DNA sequence (SEQ ID NO: 6) and protein sequence (SEQ ID NO: 7) of BstXI, respectively.

[0121] Fig.35A -B show the DNA sequence (SEQ ID NO: 8) and protein sequence (SEQ ID NO: 9) of M.BstXI, respectively.

[0122] Fig.36A -B show the DNA sequence (SEQ ID NO: 10) and protein sequence (SEQ ID NO: 11) of S.BstXI, respectively.

[0123] Fig.37A -B show the DNA sequence (SEQ ID NO: 12) and protein sequence (SEQ ID NO: 13) of PciI, respectively.

[0124] Fig.38A -B show the DNA sequence (SEQ ID NO: 14) and protein sequence (SEQ ID NO: 15) of M.PciI, respectively.

[0125] Fig.39The DNA sequence encoding MI.EarI (SEQ ID NO: 17) is shown, which recognizes CTCTTC and methylates at N4 cytosine or N6 adenine for cloning SapI and BspQI.

[0126] Fig.40 The DNA sequence encoding M2.EarI (SEQ ID NO: 18) is shown, which recognizes CTCTTC and methylates at N4 cytosine or N6 adenine for cloning SapI and BspQI.

[0127] Fig.41A -B shows an agarose gel measuring the FI of 1 μl WT BspQI and 1 μl mutant (K279P / R388F) BspQI (starting concentrations of 512 U and 1,024 U, respectively) by cleaving 1 μl pUC19 DNA substrate in a 2-fold serial dilution using diluent A and NEB1 buffer plus 10% glycerol. The reaction was performed at 50°C.

[0128] Fig.42 Shown is an agarose gel measuring the FI of 5 μl WT Sapl and 5 μl mutant (K273A) Sapl (starting concentrations 32 U and 16 U respectively) by cleaving pUC19 DNA substrate in 2-fold serial dilutions using diluent A and NEB2 buffer plus 25% DMSO.

[0129] Fig.43A -B shows the catalytic activity and star activity of pXba obtained by 5 μl WT KpnI and 5 μl D16N / E132A / D148E KpnI (initial concentrations of 32 U and 256 U, respectively). 2 μl of pXba DNA substrate (0.5 μg) was digested with 2-fold serial dilutions of the enzyme in NEB2 buffer containing 10 mM Tris-HCl, pH 7.4, 50 mM KCl, 0.1 mM EDTA, 1 mM DTT and 50% glycerol, which was made up to a total volume of 50 μl with water.

[0130] Fig.44A -G shows the amino acid sequences of the enzymes in Table 1, and the DNA sequences of the enzymes in Table 1 that have not been previously published. DETAILED DESCRIPTION OF THE INVENTION

[0131] Embodiments of the present invention provide a general method for selecting a restriction endonuclease with a desired property. The general method relies on appropriate analytical determinations to determine whether the desired restriction endonuclease has been produced. In particular, one embodiment of the general method provides a systematic screening method with a set of steps. The method has been derived by implementing hundreds of reactions using a variety of restriction endonucleases. Most of the examples provided herein relate to identifying restriction endonucleases with reduced star activity but whose cleavage activity is at least similar to that of the WT restriction endonuclease. However, it is expected that the same method can be successfully applied to change (modify) other properties of restriction endonucleases, such as improved cleavage activity in a desired buffer, thermal stability, reaction rate in defined conditions, etc.

[0132] As described above, the endpoint of interest is to convert a restriction endonuclease with star activity into a high-fidelity restriction endonuclease with significantly reduced star activity. Star activity refers to the confusion of a single restriction endonuclease in terms of cutting specificity. The terms "reduced star activity" and "increased fidelity" are used interchangeably herein. Although restriction endonucleases are characterized by their property of cutting DNA at specific sequences, some restriction endonucleases additionally cut DNA inefficiently at a second (secondary) site of the DNA. This secondary cutting can occur consistently or can occur only under certain conditions, such as one of the following conditions: increased concentration, certain buffers, temperature, substrate type, storage, and incubation time.

[0133] It is generally acknowledged that little is known about the complex environment created by the hundreds of amino acids that make up a protein and determine specificity. One prior art approach has been to use crystallography to identify contact points between an enzyme and its substrate. However, crystallography has limitations in freezing structures in time in an unnatural chemical environment.

[0134] Using existing analytical techniques, the rules that determine the contribution of amino acids at any site in a protein and the role played by the structure of the substrate molecule have proven elusive. For example, it is shown here that mutating a single amino acid in a restriction endonuclease can cause all or partial loss of activity.

[0135] In this paper, no structural explanation is proposed to explain why star activity can be increased with high glycerol concentrations (>5% v / v), high enzyme to DNA ratios (typically >100 units of enzyme per μg DNA), low ionic strength (<25 mM salt), high pH (>8.0), the presence of organic solvents (e.g., DMSO, ethanol), and the use of other divalent cations (Mn 2+ 、Co 2+ ) replaces Mg 2+It is acknowledged here that because of the diversity of factors that affect star activity, it will be necessary to compare WT and mutant star activities under the same reaction conditions and in the same predetermined buffer, and to develop standard reaction conditions under which any high-fidelity enzyme must exhibit the described properties - even if these properties are also observed under other reaction conditions.

[0136] Embodiments of the present invention relate to the generation of modified restriction endonucleases with specific improved properties, i.e., with enhanced fidelity of cleavage without a significant reduction in overall cleavage activity or a significant reduction in yield from the host cell producing the protein. The methods that have been developed herein for discovering mutants with improved properties are the result of exhaustive experimentation, and the properties of the resulting enzymes have been defined in the context of specified conditions. The methods described herein can be used to alter the enzymatic properties of any restriction endonuclease under predetermined conditions, but are not limited to the specific defined conditions. Table 1:

[0137]

[0138]

[0139]

[0140] The method is derived from the recognition that the amino acids responsible for homologous activity and star activity are different. The engineering of the high-fidelity restriction endonucleases described herein shows that homologous activity and star activity can be separated, and there are different key amino acid residues that affect these different activities. The position of the amino acid that was found to affect star activity is not necessarily found in the active site of the protein. By developing successful criteria in determining the form of FI (see also Wei et al. Nucleic Acid Res., 36, 9, e50 (2008)) and the overall fidelity index improvement factor, the cutting properties of any restriction endonuclease have been determined for the first time herein.

[0141] "Overall fidelity index improvement factor" refers to the highest FI of the mutant with the greatest cleavage activity in the selected buffer group divided by the highest FI of the corresponding WT endonuclease with the greatest cleavage activity. The selected group can be any size greater than 1, but will actually contain less than 10 different buffers, and more preferably contain 4 buffers. The group can also include less than 4 buffers. Additionally and not just for the buffer group consisting of NEB1, NEB2, NEB3 and NEB4, an overall FI improvement factor of at least 2 should preferably apply to any mutant restriction endonuclease of the claimed invention.

[0142] "Similar cleavage activity" can be measured by reacting the same amount of enzyme with the same amount and type of substrate under the same conditions, and visually comparing the cleavage patterns on the gel after electrophoresis so that the amounts of cleavage products appear to be the same within the standard tolerance of error and where the quantitative similarity is greater than 10%.

[0143] “Artificial” means “man-made”.

[0144] "Standard conditions" refers to the overall FI improvement factor calculated from the results obtained in NEB 1-4 buffers.

[0145] The general method described herein has been exemplified with 27 restriction endonucleases: AgeI, AvrII, BamHI, BsaI, BsmBI, BspQI, BstXI, EagI, EcoRI, EcoRV, HindIII, HpaI, KpnI, MfeI, NcoI, NheI, NotI, PciI, PstI, PvuII, Sad, SalI, SapI, SbfI, ScaI, SphI, and SspI restriction endonucleases. However, as described above, it is contemplated that the method is effective for engineering any restriction endonuclease with significant star activity.

[0146] Embodiments of the method utilize a general approach to generate mutant restriction endonucleases with reduced star activity. For some enzymes, it has been shown that mutations in charged residues that are determined to be conserved between two isojunction enzymes are useful (see, e.g., SapI in Example 25). However, in general, the method includes a first step of identifying all charged and polar residues within the protein sequence of the endonuclease. For example, charged amino acids and polar residues include acidic residues Glu and Asp; basic residues His, Lys, and Arg; amide residues Asn and Gln; aromatic residues Phe, Tyr, and Trp; and nucleophilic residue Cys. Each residue is targeted and mutated to Ala, and the products of these targeted mutations are screened for the desired properties of increased fidelity. If none of the mutants obtained provide satisfactory results, the next step is to target mutations of all hydroxylated amino acids, i.e., Ser, Thr, and Tyr, with preferred mutations being mutations of Ser and Thr to Ala and Tyr to Phe. It is also possible to simultaneously target mutations of two types of residues, as done in Examples 16-23. The mutation to Ala may be replaced by a mutation to Val, Leu or Ile.

[0147] After these analyses, if one or more preferred mutants produced in the above steps still have substandard performance under the selected test, these mutants can be selected and mutated again to each of the other possible 18 amino acids. This is called saturation mutagenesis. Saturation mutagenesis provides preferred high-fidelity mutants of EcoRI (Example 2), part of BamHI (Example 1) and PvuII (Example 12). Depending on the result of saturation mutagenesis, the next step will introduce additional mutations into restriction endonucleases in a targeted or random or both manners. In Example 11, SacI-HF includes random mutations that occur accidentally during inverse PCR. In Example 20, PciI-HF is the result of random mutations, rather than the result of targeted mutations. In Example 26, BspQI-HF contains two mutations that are found to be synergistic in enhancing fidelity.

[0148] The use of different targeted mutagenesis methods such as inverse PCR can include the introduction of non-targeted mutations at a second site in the protein. These second mutations can incidentally provide the desired properties (see Example 20). It is desirable to examine those mutant enzymes with multiple mutations to determine if all mutations are necessary for the observed effect. In Example 11, Q117H had no effect on activity in the double mutant. In Example 20, additional spontaneous mutations appear to be solely responsible for the observed increased fidelity, while in Example 24, the individual mutations act synergistically.

[0149] In some cases, mutations may provide additional advantages beyond improved fidelity (see, e.g., BamHI, where E163A or P173A causes the enzyme to become more thermolabile).

[0150] The high fidelity / reduced star activity properties of the mutants provided in the Examples were selected based on their function in a set of standard buffers. Other mutations may be preferred if a different buffer composition is chosen. However, the same methods can be applied to discover mutants. Table 4 lists the mutations that applied to each restriction endonuclease and provided an overall FI improvement factor in standard buffers.

[0151] Engineering of a high-fidelity restriction endonuclease that provides an overall FI improvement factor of at least 2 comprises one or more of the following steps:

[0152] 1. Assessment of Star Activity of WT Restriction Endonuclease

[0153] In one embodiment of the invention, the extent of star activity of a restriction endonuclease is determined by the following method: for an appropriate substrate, using a high initial concentration of a stock endonuclease and serial dilutions thereof (e.g., 2-fold or 3-fold dilutions), the endonuclease activity is measured. The initial concentration of the restriction endonuclease is not important, as long as it is sufficient to allow star activity to be observed in at least one concentration, such that upon dilution, star activity is no longer detected.

[0154] Suitable substrates contain nucleotide sequences that are cleaved by homologous endonuclease activity and in which star activity can be observed. The substrate can be a vector or a second DNA substrate containing a restriction endonuclease gene. Examples of substrates used in Table 2 are pBC4, pXba, T7, lambda, and pBR322.

[0155] The concentration of the stock restriction endonuclease is initially selected so that star activity can be easily identified and analyzed in WT and mutant restriction endonucleases. According to the guidance of the 2007-08NEB catalog, select appropriate dilution buffers such as NEB diluent A, B or C for serial dilution. The restriction endonucleases of serial dilutions react with a total reaction volume determined by the reaction vessel size with a suitable substrate of a predetermined concentration. For example, it is convenient to carry out multiple reactions in a microtiter plate, in which a 30 μl reaction mixture is an appropriate volume per well. Therefore, examples generally use 0.6 μg of substrate in 30 μl, which is equivalent to 1 μg of substrate in 50 μl. The amount of substrate in the reaction mixture is not critical, but it is preferably constant between reactions. The cleavage reaction is carried out at a predetermined temperature (e.g., 25°C, 30°C, 37°C, 50°C, 55°C or 65°C) for a standard time, such as one hour. The cleavage products can be assayed by any standard technique, such as by 0.8% agarose gel electrophoresis, to determine the fidelity index defined above.

[0156] Not all restriction endonucleases have significant star activity, as determined from their FI. However, if an endonuclease has a maximum FI of no more than about 250 and a minimum FI of less than 100, the restriction endonuclease is classified as having significant star activity. Such endonucleases are selected as targets for enzyme engineering to increase fidelity to a single substrate. In some cases, restriction endonucleases with FI greater than about 500 and FI less than about 100 are also engineered to obtain better cleavage activity.

[0157] Table 2 below lists the FI of some engineered restriction endonucleases before engineering. All samples were analyzed on 0.8% agarose gel.

[0158] Table 2

[0159]

[0160] *Substrates: λ is λ phage DNA; λ(H3) is HindIII-digested λ phage DNA; pXba is pUC19 with XbaI-digested adenovirus fragment; pBC4: a shorter form of pXba; T7: T7 DNA

[0161] **FI-1 to FI-4: Fidelity Index for enzymes in NEB buffers 1, 2, 3, and 4. The numbers in parentheses are values ​​for the relative cleavage activity of a mutant restriction endonuclease in a given buffer of a buffer set compared to the "best" cleavage activity of the same mutant restriction endonuclease in any buffer of that buffer set.

[0162] The composition of NEB buffer is as follows:

[0163] NEB1: 10 mM Bis Tris propane-HCl, 10 mM MgCl 2 , 1 mM (at 25°C, pH 7.0);

[0164] NEB2: 50mM NaCl, 10mM Tris-HCl, 10mM MgCl 2 , 1 mM dithiothreitol (at 25°C, pH 7.9);

[0165] NEB3: 100mM NaCl, 50mM Tris-HCl, 10mM MgCl 2 , 1 mM dithiothreitol (at 25°C, pH 7.9);

[0166] NEB4: 50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM dithiothreitol (pH 7.9 at 25°C).

[0167] ***The composition of the NEB diluent is as follows. (Using diluent instead of water in the dilution will keep the glycerol concentration constant during the reaction.)

[0168] Diluent A: 50 mM KCl, 10 mM Tris-HCl, 0.1 mM EDTA, 1 mM dithiothreitol, 200 mg / ml BSA. 50% glycerol (at 25°C, pH 7.4);

[0169] Diluent B: 300 mM NaCl, 10 mM Tris-HCl, 0.1 mM EDTA, 1 mM dithiothreitol, 500 mg / ml BSA, 50% glycerol (at 25°C, pH 7.4);

[0170] Diluent C: 250 mM NaCl, 10 mM Tris-HCl, 0.1 mM EDTA, 1 mM dithiothreitol, 0.15% Triton X-100, 200 mg / ml BSA, 50% glycerol (pH 7.4 at 25°C).

[0171] 2. Construction of high expression host cell lines

[0172] It is convenient if the host cell can overexpress a mutant restriction endonuclease for which reduced star activity is sought. If the restriction enzyme is highly expressed in E. coli, star activity can be easily detected in crude extracts, which simplifies the screening of high-fidelity restriction endonucleases. However, mutant restriction endonucleases can be expressed in any host cell, provided that the host cell is protected in some way from toxicity resulting from enzymatic cleavage. This may include: the presence of a methylase; production in a cell compartment (e.g., inclusion body or periplasm) that provides a barrier to access the genome; in vitro synthesis; production in emulsion (see U.S. Patent Application Serial No. 12 / 035,872) in the absence of a cleavage site in the host genome; making the enzyme in a building block that undergoes intein-mediated ligation, etc.

[0173] For production, the overexpression of mutant restriction endonucleases can be achieved using standard cloning techniques, such as using an E. coli host, inserting the endonuclease into a pUC19-derived expression vector, which is a high copy; and using a relatively small plasmid that can continuously express recombinant proteins. The vector may preferably contain a suitable promoter such as the lac promoter and a multicopy insertion site adjacent to the promoter location. Alternatively, an IPTG-induced promoter that requires gene expression may be selected. If the activity in the crude extract is not sufficient, a column purification step of the restriction endonuclease in the crude extract may be performed.

[0174] 3. Restriction endonuclease mutagenesis

[0175] The DNA encoding each charged group or polar group within the restriction endonuclease can be targeted individually, and the mutated DNA cloned and prepared for testing. Multiple mutations can be introduced into a single restriction endonuclease gene. Targeted mutagenesis of restriction endonucleases can be accomplished by methods known in the art. A convenient method used herein is inverse PCR. In this method, a complementary primer pair containing a target codon plus multiple nucleotides (e.g., 18 nt) on the 5' and 3' sides of the codon is synthesized. The selection of appropriate primers can be easily accomplished by evaluating the gene sequence of the target endonuclease around the target amino acid residue. Access to the gene sequence is provided by REBASE and GenBank. The sequences of the endonucleases described herein in the Examples are provided in Figures 31 to 38 and 44. The template for PCR is a plasmid containing the restriction endonuclease gene. Preferably, the polymerase is a high-fidelity polymerase, such as or Deep Vent TM DNA polymerase. By changing the annealing temperature and Mg 2+ Concentration, can realize the successful introduction of most mutations.Then, purify and preferably digest the pcr amplification product by DpnI.In one embodiment of the invention, the product of digestion is transformed into competent host cell (for example intestinal bacteria), and it has used corresponding methylase pre-modification.Bacterial colonies from each mutant are selected and grown under conditions similar to WT growth conditions (for example, using similar growth medium, drug selection and temperature).For the star activity of minimization, the obtained restriction endonuclease is screened.

[0176] 4. Screening for mutant restriction endonucleases with reduced star activity

[0177] Conditions such as buffer composition, temperature and diluent should be defined to determine the star activity in mutant restriction endonucleases. Tables 2 and 3 show the FI of recombinant endonucleases before and after mutation in four different buffers using three different diluents at 37°C. Thus, it is possible to determine which mutants have an overall desired improved fidelity index factor of at least 2, more than 10, at least 50 or more than 500 and select enzymes as preferred high fidelity mutants.

[0178] In one embodiment of the invention, first, the mutant restriction endonuclease activity is screened under normal buffer conditions (no more than 5% glycerol). For those mutants with at least about 10% of the WT restriction endonuclease activity, the activity is also determined under star activity-enhancing conditions that promote star activity, such as high glycerol concentrations and optionally high pH. Preferably, mutants with minimal star activity in normal buffer but with acceptable homologous activity are selected. Then, plasmids can be extracted and sequenced to verify the mutants. In some cases, star activity is not easy to measure - even under high glycerol and high pH conditions. Alternatively, the activities in different buffers are measured and compared, and the mutants with the highest ratio of NEB4 to NEB3 cleavage activity can be further tested for star activity improvement.

[0179] 5. Saturation mutagenesis of a single residue

[0180] As described in the previous section, the first step is to mutate the target amino acid in the restriction endonuclease to Ala. If the result is not satisfactory, saturation mutagenesis is performed. Preferably, this is performed by one of two methods. One method is to change the intended codon to NNN. After mutagenesis, multiple colonies are analyzed under normal conditions and under conditions that promote star activity. Alternatively, different codons can be selected for mutagenesis of each of the following target amino acids, for example: Ala: GCT; Cys: TGC; Asp: GAC; Glu: GAA; His: CAC; Ile: ATC; Lys: AAA; Leu: CTG; Met: ATG; Asn: AAC; Pro: CCG; Gln: CAG; Arg: CGT; Ser: TCC; Thr: ACC; Val: GTT; Trp: TGG and Tyr: TAC.

[0181] 6. Combination

[0182] If a single mutation does not sufficiently reduce star activity, more than one mutation can be introduced into the restriction endonuclease gene. Combinations of mutations and saturation mutagenesis can be performed in any order.

[0183] 7. Mutant Purification and Improvement Evaluation

[0184] High-fidelity mutants can be purified in a variety of ways, including the use of different chromatographic columns. For routine quality assessment, an FPLC heparin column is sufficient to remove DNA and non-specific nucleases from the preparation. Multiple columns including ion exchange columns, hydrophobic columns, size exclusion columns, and affinity columns can be used for further purification.

[0185] The FI of the purified high-fidelity restriction endonuclease was measured in four NEB buffers and compared to the FI of the WT restriction endonuclease. The ratio of the FI of the high-fidelity restriction endonuclease in its optimal buffer to the FI of the WT is the overall improvement factor.

[0186] Table 3 - FI* of exemplary restriction endonucleases

[0187]

[0188] *FI is the ratio of the highest concentration showing no star activity to the lowest concentration that completely digested the substrate.

[0189] **The numbers in parentheses are values ​​for the relative cleavage activity of the mutant restriction endonuclease in a given buffer of the buffer set compared to the maximum cleavage activity of the same mutant restriction endonuclease in any buffer of the buffer set.

[0190] Table 4 - Mutations that provide restriction endonucleases with high fidelity

[0191]

[0192]

[0193] Mutations for each enzyme are separated by semicolons.

[0194] All references cited above and below, as well as U.S. Provisional Application Serial No. 60 / 959,203, are incorporated by reference. Example

[0195] Where amino acids are represented by a single letter code, this is intended to be the standard nomenclature. An explanation of this code is provided, for example, on page 280 of the NEB catalogue 2007 / 2008.

[0196] The plasmid used for cloning and as a substrate has the following sequence:

[0197] pLaczz2 (SEQ ID NO: 102), pSyx20-lacIq (SEQ ID NO: 105), pBC4 (SEQ ID NO: 103), pXba (SEQ ID. NO: 104), and pAGR3 (SEQ ID NO: 106). pACYC is described in GenBank XO 6403, T7 is described in GenBank NC001604, pUC18 is described in GenBank L09136, and pRRS is described in Skoglund et al. Gene, 88: 1-5 (1990). pSX33 was constructed by inserting the lacI gene into pLG339 at the EcoRI site. pLG339 is described in Stoker, et al. Gene 19, 335-341 (1982).

[0198] All buffers referred to herein as NEB buffers are available from New England Biolabs, Inc. (NEB), Ipswich, MA.

[0199] Example 1: Engineering high-fidelity BamHI

[0200] 1. Extract the plasmid containing BamHI methylase and BamHI endonuclease

[0201] Competent E. coli host cells were transformed with pUC18-BamHIR and pACYC184-BamHIM, and BamHIR was extracted by standard Qiagen miniprep procedures (Qiagen, Valencia, CA) using standard miniprep techniques.

[0202] 2. Selection of mutagenesis targets

[0203] BamHI and the related restriction endonuclease OkrAI were cloned and sequenced. OkrAI was found to have significant star activity if the reaction was performed at 37°C in NEB buffers (1, 2 and 4). This analysis tested the hypothesis that the amino acid residue(s) responsible for star activity are similar between BamHI and OkrAI endonucleases.

[0204] The complete protein sequence of BamHI (SEQ ID NO: 19) is:

[0205]

[0206] The complete protein sequence of OkrAI (SEQ ID NO: 20) is:

[0207]

[0208] The "Bestfit" similarity analysis of the protein sequences of BamHI and OkrAI endonucleases by GCG shows the following results, where the upper protein sequence is BamHI and the lower protein sequence is OkrAI:

[0209] bamhir.pep x okrair.pep

[0210]

[0211] Found that similar charged residues (D, E, H, K, R) among BamHI are E28, K30, K52, K61, E77, K84, E86, K88, D94, K97, K106, E111, E113, H121, R122, K126, K146, D154, R155, E161, E163, E170, E182, K193, D196 and R201.In above comparison, these residues are underlined.Known mutants E77K, D94N, E111K and E113K were reported as inactivated (Xu, Shuang-yong et al.J.Bacteriol.266:4425-4429 (1991)) in the past, so exclude them. The initial mutagenesis selection targeted 22 consensus charged amino acid residues, mutating them to alanine: E28A, K30A, K52A, K61A, K84A, E86A, K88A, K97A, K106A, H121A, R122A, K126A, K146A, D154A, R155A, E161A, E163A, E170A, E182A, K193A, D196A, and R201A.

[0212] 3. Mutagenesis of BamHI

[0213] The point mutagenesis of the selected mutations was performed by inverse PCR, and all the corresponding codons were changed to GCA (alanine).

[0214] The following primers were used for mutagenesis:

[0215] E28A

[0216] 5'ATTCAACAAGCATACAAT GCA GTTAAAACATCTATTGT3'

[0217] (SEQ ID NO:23)

[0218] 5'ACAAATAGATGTTTTAAC TGCATTGTATGCTTGTTGAAT3

[0219] (SEQ ID NO:24)

[0220] K30A

[0221] 5'CAAGCATACAATGAAGTT GCA ACATCTATTTGTTCACCT3'

[0222] (SEQ ID NO:25)

[0223] 5'AGGTGAACAAATAGATGT TGC AACTTCATTGTATGCTTG3

[0224] (SEQ ID NO:26)

[0225] K52A

[0226] 5'ACGATTAACACCCGAA GCA AATTGTAACGGTGTAGTA3

[0227] (SEQ ID NO:27)

[0228] 5'TACTACACCGTTACAATT TGC TTCGGTGTTGTTAATCGT3

[0229] (SEQ ID NO:28)

[0230] K61A

[0231] 5'AACGGTGTAGTACCAATT GCA GAACTATGTTACACCTTA3

[0232] (SEQ ID NO:29)

[0233] 5'TAAGGTGTAACATAGTTC TGC AATTGGTACTACACCGTT3

[0234] (SEQ ID NO:30)

[0235] K84A

[0236] 5'AACCCCCTTGATATACTT GCA CTTGAAAAAGAAAAAAGGT3

[0237] (SEQ ID NO:31)

[0238] 5'ACCTTTTTTCTTTTCAAG TGC AAGTATATCAAGGGGTTT3

[0239] (SEQ ID NO:32)

[0240] E86A

[0241] 5'GATATACTTAAACTT GCA AAGAAAAAAGGTGGTCCG3

[0242] (SEQ ID NO:33)

[0243] 5'CGGACCACCTTTTTTCTT TGC AAGTTTAAGTATCAAG3

[0244] (SEQ ID NO:34)

[0245] K88A

[0246] 5'ATACTTAAACTTGAAAAG GCA AAAGGTGGTCCGATTGAT3'

[0247] (SEQ ID NO:35)

[0248] 5'ATCAATCGGACCACCTTT TGC CTTTTTCAAGTTTAAGTAT3

[0249] (SEQ ID NO:36)

[0250] K97A

[0251] 5'GGTCCGATTGATGTTTAT GCA GAGTTCATAGAAAACAGT3

[0252] (SEQ ID NO:37)

[0253] 5'ACTGTTTTCTATGAACTC TGC ATAAACATCAATCGGACC3

[0254] (SEQ ID NO:38)

[0255] K106A

[0256] 5'ATAGAAAAACAGTGAACTT GCA CGTGTAGGTATGGAA3'

[0257] (SEQ ID NO:39)

[0258] 5'AAATTCATACCTACACG TGC AAGTTCACTGTTTTCTAT3'

[0259] (SEQ ID NO:40)

[0260] H121A

[0261] 5'GGAAATATTAGTTCTGCC GCA CGTTCAATGAACAAACTT3'

[0262] (SEQ ID NO:41)

[0263] 5'AAGTTTGTTCATTGAAACG TGC GGCAGAACTAATATTCC3'

[0264] (SEQ ID NO:42)

[0265] R122A

[0266] 5'AATATTAGTTCTGCCCAC GCA TCAATGAACAAACTTCTA3'

[0267] (SEQ ID NO:43)

[0268] 5'TAGAAGTTTGTTCATTGA TGC GTGGGCAGAACTAATATT3'

[0269] (SEQ ID NO:44)

[0270] K126A

[0271] 5'GCCCACCGTTCAATGAAC GCA CTTCTATTAGTTAAACAT3'

[0272] (SEQ ID NO:45)

[0273] 5'ATGTTTTAATCCTAATAGAAG TGC GGTCATTGAACGGTGGGC3'

[0274] (SEQ ID NO:46)

[0275] K146A

[0276] 5'ATTACCTTTATGCCTATT GCA CAATTGGCCTATTATCTT3'

[0277] (SEQ ID NO:47)

[0278] 5'AAGATAATAGGCCAATTG TGC AATAGGCATAAGGATAAT3'

[0279] (SEQ ID NO:48)

[0280] D154A

[0281] 5'TTGGCCTATTATCTTACA GCA CGTGTTACCAATTTCGAG3'

[0282] (SEQ ID NO:49)

[0283] 5'CTCGAAATTGGTAACAG TGC TGTAAGATAATAGGCCAA3'

[0284] (SEQ ID NO:50)

[0285] R155A

[0286] 5'GCCTATTATCTTACAGAT GCA GTTACCAATTTCGAGGAA3'

[0287] (SEQ ID NO:51)

[0288] 5'TTCCTCGAAATTGGTAAC TGC ATCTGTAAGATAATAGGC3'

[0289] (SEQ ID NO:52)

[0290] E161A

[0291] 5'CGTGTTACCAATTTCGAG GCA TTAGAACCTTATTTTGAA3'

[0292] (SEQ ID NO:53)

[0293] 5'TTCAAAATAAGGTTCTAA TGC CTCGAAATTGGTAACAG3'

[0294] (SEQ ID NO:54)

[0295] E163A

[0296] 5'ACCAATTTCGAGGAATTA GCA CCTTATTTTGAACTTACT3'

[0297] (SEQ ID NO:55)

[0298] 5'AGTAAGTTCAAAATAAGG TGC TAATTCCTCGAAATTGGT3'

[0299] (SEQ ID NO:56)

[0300] E170A

[0301] 5'CCTTATTTTGAACTTACT GCA GGACAACCATTTATTTTTTTATT3'

[0302] (SEQ ID NO:57)

[0303] 5'AATAAAAAATAAATGGTTG TGC TGCAGTAAGTTCAAAATAAGG3'

[0304] (SEQ ID NO:58)

[0305] E182A

[0306] 5'TTTATTTTTTATTGGATTTAAT GCT GCAGCTTATAATTCTAATGTC3'

[0307] (SEQ ID NO:59)

[0308] 5'GACATTAGAATTATAAGCTGC AGC ATTAAATCCAATAAAAAATAAA3'

[0309] (SEQ ID NO: 60)

[0310] K193A

[0311] 5'AATGTCCCTTTAATTCCC GCA GGTTCTGACGGTATGTCA3'

[0312] (SEQ ID NO: 61)

[0313] 5'TGACATACCGTCAGAACC TGC GGGAATTAAAGGGACATT3'

[0314] (SEQ ID NO: 62)

[0315] D196A

[0316] 5'TTAATTCCCAAAGGTTCT GCA GGTATGTCAAAACGCTCA3'

[0317] (SEQ ID NO: 63)

[0318] 5'TGAGCGTTTTGACATACC TGC AGAACCTTTGGGAATTAA3'

[0319] (SEQ ID NO: 64)

[0320] R201A

[0321] 5'TCTGACGGTATGTCAAAA GCA TCAATTAAGAAATGGAAA3'

[0322] (SEQ ID NO: 65)

[0323] 5'TTTCCATTTCTTAATTGA TGC TTTTGACATACCGTCAGA3'

[0324] (SEQ ID NO: 66)

[0325] The PCR reaction volume was 100 μl, containing 2 μl of each PCR primer, 1 μl of pUC18-bamhiR, 400 μM dNTPs, 4 units of Deep Vent TM DNA polymerase and 10 μl 10x containing 0, 2 or 6 μl MgSO 4 Add the balance of Thermopol buffer to water.

[0326] PCR reaction conditions were 94°C for 5 min, then 25 cycles of 94°C for 30 sec, 55°C for 30 sec, 72°C for 4 min, and a final extension time of 7 mins at 72°C. PCR products were purified on standard Qiagen spin columns (Qiagen, Valencia, CA). 6 to 16 μl of PCR product was digested with 20 units of DpnI for 1 hour. The digested product was transformed into E. coli (pACYC-bamHIM).

[0327] After 6 PCR reactions, 14 of the 22 engineered mutations were obtained: E28A, K30A, K61A, E86A, K97A, H121A, K126A, K146A, E161A, E163A, E170A, E182A and R201A. Mutant proteins were extracted from cell lysates from overnight cultures and their activities were compared with WT BamHI. Normal enzyme activity was analyzed in NEB2 buffer with or without 5% glycerol, while star activity was determined in NEB2 with 39.2% glycerol, although a lower percentage of glycerol could be used initially. The substrates used for the different reactions were pBR322, pUC19 or lambda DNA. The cleavage reactions were performed at 37°C for 30 minutes or 1 hour. Mutants K97A, H121A, K126A, E161A, E182A, R201A were found to be inactive (less than 1% of WT BamHI activity), while E28A, K146A, E163A, E170A mutants had similar levels of activity as the WT enzyme, including star activity. Three mutants, K30A, E86A and K126A, were found to have significantly reduced star activity compared to WT BamHI. K30A and E86A were also found to have similar total cleavage activity as the WT enzyme, while they showed a significant reduction in star activity. In contrast, K126A had only 25% of the total WT enzyme cleavage activity and a less significant improvement over the star activity observed for K30A and E86A.

[0328] Re-examination of the pUC18-bamHIR plasmid showed that the normal high copy plasmid had mutated into a low copy plasmid. Primer pairs were designed to transfer the bamHIR gene into the high copy plasmid:

[0329] 5'GGTGGT GCATGC GGAGGTAAATAAATGGAAGTAGAAAAAGAGTTTATTACT

[0330] GAT3' (SEQ ID NO: 67)

[0331] 5'GGTGGT GGTACCCTATTTGTTTTCAACTTTATCTTTCCATTTCTTAATTGA3'

[0332] (SEQ ID NO: 68)

[0333] The template was pUC18-bamhIR WT with mutations at K30A, E86A or K126A. The PCR composition contained: 5 μl template, 2 μl of each primer, 400 μM dNTPs, 10 μl 10x Thermopol buffer, 4 units of 2 μl Deep Vent TM polymerase, 72 μl H 2 Q plus 0, 2, 6 μl MgSQ 4 . PCR conditions were 94°C for 5 min, followed by 94°C for 30 sec, 55°C for 30 sec, and 72°C for 40 sec for 25 cycles and a final extension period of 7 minutes. The PCR product was digested with SphI and KpnI and connected to pUC19 with the same enzyme digestion pair. The connected product was transformed into competent E. coli containing pACYC-bamHIM. 26 colonies of pUC19 forms containing BamHIR K30A and 12 of those containing E86A were identified and grown. The activity of BamHI from these cultures was checked. All of them were active. Plasmids from five colonies of each mutation were extracted and 3 BamHIR plasmids from each mutation were sequenced. The identity of plasmids pUC19-BamHI (K30A) and pUC19-BamHI (E86A) was confirmed.

[0334] pUC19-BamHI(K30A) vector was used to repeat mutations that were unsuccessful in pUC18-BamHIR. The PCR mixture contained: 1 μl template and 2 μl of each primer, 400 μM dNTP, 10 μl 10x Thermopol buffer, 4 units of 2 μl Deep Vent TM polymerase, 76 μl H 2 Q plus 0, 2, 6 μl 100 μM MgSO 4 The amplification mixture was prepared by PCR at 94°C for 5 min, followed by 25 cycles of 94°C for 30 sec, 55°C for 30 sec, and 72°C for 3 min 30 sec, and a final extension period of 7 min. The PCR product was digested with DpnI and transformed into competent E. coli transformed with pACYC-BamHIM. Enzyme activity was detected on pUC19 substrate. The reaction composition was: 3 μl cell extract, 3 μl NEB2, 3 μl 50% glycerol, 0.5 μl 0.5 μg pUC19, 20.5 μl H2 O. The reaction was carried out at 37°C for 1 hour. K30A / R122A, K30A / R155A and K30A / K193A are inactive. K30A / K52A and K30A / K88A are about 1 / 10 of the activity of K30A. The normal activities of K30A / K106A, K30A / D154A and K30A / D196A are similar to the activity of K30A BamHI. Comparison of the star activity of these three mutants with K30A at a high concentration of glycerol (39.2%) showed that K30A / D196A had a similar star activity to K30A, K30A / D154A even had more star activity than K30A, and K30A / K106A had less star activity than K30A. Attempts to isolate the K106A mutation of BamHI in the pUC19 vector failed due to cytotoxicity.

[0335] Using inverse PCR, mutations at K30, E86 and K106 were combined: K30A / E86A, E86A / K106A, K30A / K106A and K30A / E86A / K106A. K30A / E86A appeared to be the preferred mutant. After purification, the FI of the BamHI mutant was found to be 25% higher in all NEB buffers.

[0336] Further random mutagenesis was performed at the K30 and E86 sites:

[0337] For K30:

[0338] 5'CAAGCATACAATGAAGTTNNNACATCTATTTGTTCACCT3'

[0339] (SEQ ID NO: 69)

[0340] 5'AGGTGAACAAATAGATGTNNNAAC TTCATTGTATGCTTG3'

[0341] (SEQ ID NO: 70)

[0342] For E86:

[0343] 5'GATATACTTAAACTTNNNAAGAAAAAAAGGTGGTCCG3'

[0344] (SEQ ID NO: 71)

[0345] 5'CGGACCACCTTTTTTCTNNNNAAGTTTAAGTATATCAAG3'

[0346] (SEQ ID NO: 72)

[0347] The PCR composition was: 1 μl template (pUC19-BamHIR (K30A) or pUC19-BamHIR (E86A)) and the amplification mixture described above was used. PCR was performed: 94°C for 5 min, then 94°C for 30 sec, 55°C for 30 sec and 72°C for 3 min 30 sec for 25 cycles and a final extension period of 7 min. The PCR product was digested by DpnI and transformed into E. coli (pACYC-BamHIM).

[0348] A total of 155 colonies were picked for random mutations at K30 and 158 colonies were picked for the E86 locus. Colonies were grown overnight and cell extracts were prepared. 0.5 μg of pUC19 was digested with 1 μl of cell extract in NEB2 buffer with 42.5% glycerol at 37°C for 1 hour. Cell extracts with significantly less star activity were re-analyzed at 37°C for 30 minutes at 1, 4, and 16-fold dilutions against 0.5 μg of pUC19 in NEB2 buffer with 39.2% glycerol. For those mutants observed to have reduced star activity, the corresponding plasmids were extracted and sequenced to confirm the mutation. A total of 3 clones (#12, #66 and #82) contained K30 mutations, and a total of 33 clones (#5, #15, #16, #19, #29, #47, #51, #55, #56, #58, #61, #69, #71, #73, #76, #82, #86, #88, #93, #94, #97, #98, #100, #104, #107, #113, #117, #118, #129, #132, #136, #139 and #151) were sequenced. After sequencing, it was found that #12 and #66 contained K30G mutations, and #82 contained K30N mutations. Surprisingly, all 33 mutations were E86P mutations, just in different codons (CCA, CCT, CCC, CCG). Among these codons, CCG occurred at the highest frequency in E. coli (clones #98, #136, and #139).

[0349] Cell extracts corresponding to K30G, K30N and K30A were serially diluted 1, 2, 4, 8, 16 and 32 times, while E86P and E86A were serially diluted 1, 2, 4, 8, 16, 32, 64, 128 and 256 times. The serially diluted extracts were reacted with 0.5 μg of pUC19 in NEB2 with 39.2% glycerol at 37°C for 30 minutes. Under extreme conditions, E86P showed superiority over other mutants. There was no obvious star activity band at up to 32 times digestion. The difference between E86P and K30 mutants (K30G, K30N and K30A) was so great that it was not necessary to combine any of these mutations in the E86P mutant.

[0350] BamHI(E86P) activity was determined on 1 μg of λ DNA, substrate (also used for WT BamHI activity assay). The assay was performed in NEB1 buffer at 37°C for 1 hour.

[0351] 4. Detailed comparison of BamHI(E86P) and WT BamHI

[0352] A. Activity of BamHI(E86P) in different NEB buffers

[0353] The activity of purified BamHI(E86P) was measured in NEB1, NEB2, NEB3, NEB4 and NEB BamHI buffer using lambda DNA substrate for 1 hour at 37° C. BamHI(E86P) was most active in NEB1 buffer and NEB2, and had activity levels of 50%, 50% and 25% in NEB3, NEB4 and BamHI buffer.

[0354] B. Comparison of the cleavage activity of BamHI(E86P) and WT BamHI on pUC19

[0355] There is one GGATCC site (BamHI site) and six AGATCC sites (BamHI star active sites) in pUC19, so pUC19 was chosen as the preferred substrate for comparison between BamHI(E86P) and WT BamHI.

[0356] In different buffers, 0.5 mg of pUC19 was digested with WT BamHI and BamHI (E86P) diluted 1, 3, 9, 27, 81, 243, 729, 2181, 6561 and 19683 times in NEB dilution buffer A. WT BamHI showed star activity in each NEB normal buffer, while BamHI (E86P) did not show a star activity band at all (Figures 2-5). This proves that BamHI (E86P) has greatly reduced star activity while maintaining homologous cleavage activity.

[0357] C. Comparison of the cleavage activities of BamHI(E86P) and WT BamHI on λDNA substrate

[0358] To calculate the fidelity index, the restriction enzyme was diluted with dilution buffer and the glycerol concentration was kept constant at 5%. In the standard reaction conditions used herein, the λDNA substrate concentration was 1 μg and the total reaction volume was 50 μl. In order to keep the enzyme volume at 10%, the enzyme was added in a volume of 5 μl. This is equivalent to 0.6 μg of substrate digested by 3 μl restriction enzymes in a total volume of 30 μl. In NEB1, NEB2, NEB3, NEB4 and NEB BamHI buffer, 3 μl of 1:2 serial dilutions from 1 to 32768 were used to digest 0.6 mg λDNA for 1 hour at 37°C.

[0359] Table 5: Fidelity index of WT and mutant BamHI in different buffers

[0360]

[0361] 5. Further Improvement of BamHI for High-Fidelity Mutants

[0362] At the one hour level, BamHI(E86P) behaved as a good high fidelity BamHI mutant. However, although the star activity of E86P was not detected at one hour, when the reaction time was extended (e.g. overnight or 14 hours), the star activity band appeared. (Figure 3). The study of improving high fidelity BamHI is ongoing.

[0363] 6. Mutation of other charged and polar residues

[0364] Other charged residues (Arg, Lys, His, Asp, Glu) were mutated to Ala at positions 2, 4, 5, 6, 10, 11, 13, 14, 18, 19, 20, 43, 51, 62, 69, 70, 76, 77, 78, 81, 87, 89, 94, 98, 101, 104, 107, 111, 113, 132, 133, 135, 137, 160, 167, 200, 204, 205, 207, 208, 209, 211, 213 in SEQ ID NO: 19. Mutagenesis was performed on the template of pUC19-BamHI(K30A).

[0365] At positions 9, 17, 26, 32, 36, 41, 42, 44, 46, 50, 65, 66, 71, 72, 75, 96, 103, 114, 118, 119, 123, 150, 151, 153, 157, 165, 169, 184, 186, 195, 199, 202 of SEQ ID NO: 19, other polar residues (Ser, Thr and Tyr) were mutated to Ala, while Tyr was mutated to Phe.

[0366] By using similar mutation and screening methods, the following mutations were found to have reduced star activity: K30A / K87A, E86P / K87E, E86A / Y165F, and K30A / E167A. E86P / K87E was identified as a mutant with improved properties in the presence of additional DMSO. However, the activity of this mutant in normal reaction buffer was much lower than that of WT BamHI.

[0367] The following mutation combinations were performed: E86P / Y165F, E86P / E167A, E86P / Y165F / E167A, K30A / Y165F / E167A, K30G / Y165F / E167A, K30A / Y165F / E167A, E86A / Y165F / E167A. All of them had low activity.

[0368] So far, it was found that E167A and Y165F had strong effects, K87A had a moderate effect, and K30A and E86A had weak effects on BamHI star activity. E86P was a specific mutation that reduced star activity at the 1 hour level but not overnight.

[0369] 7. Mutate E167 and Y165 to all other residues

[0370] In pUC19-BamHI, E167 was mutated to all other residues by changing the codons to GCA-Ala, TGC-Cys, GAC-Asp, TTC-Phe, GGT-Gly, CAC-His, ATC-Ile, AAA-Lys, CTG-Leu, ATG-Met, AAC-Asn, CCG-Pro, CAG-Gln, CGT-Arg, TCC-Ser, ACC-Thr, GTT-Val, TGG-Trp and TAC-Tyr.

[0371] After comparing all mutants, the E167T mutation was preferred, while the E167R, E167K, E167L and E167I mutations showed improvements in reducing star activity compared to E167A.

[0372] Y165 was also mutated to all other amino acid residues by changing the corresponding codons to GCT-Ala, TGC-Cys, GAC-Asp, GAA-Glu, GGT-Gly, CAC-His, ATC-Ile, AAA-Lys, CIG-Leu, ATG-Met, AAC-Asn, CCG-Pro, CAG-Gln, CGT-Arg, TCC-Ser, ACC-Thr, GTT-Val, TGG-Trp.

[0373] After comparing all mutants, the presence of Y165F resulted in significant cleavage activity, while the other mutations listed just above showed low or no cleavage activity.

[0374] 8. Further mutation of BamHI(E167T)

[0375] In the template of puc19-BamHI(E167T), all charged and polar residues were mutated to Ala as above.

[0376] As a preferred mutation, E163A / E167T was identified as BamHI-HF.

[0377] 9. Comparison of BamHI-HF and WT BamHI

[0378] Introduction of mutations at E163 resulted in a BamHI mutant with reduced thermostability, as did the addition of mutation P173A to the other mutations responsible for reduced star activity.

[0379] Unlike BamHI(E86P), BamHI-HF had no significant star activity in overnight reactions in NEB1-4 buffer. Figure 4 shows the results in NEB1 and NEB2. Therefore, BamHI(E163A / E167T) was selected as the preferred high-fidelity BamHI.

[0380] The fidelity index of BamHI-HF was measured on lambda DNA substrate with diluent A at 37°C in all four NEB buffers and compared to the WT enzyme.

[0381] Table 6: Comparison of BamHI-HF and WT BamHI

[0382]

[0383] In NEB1, BamHI-HF had the highest activity with a fidelity index of ≥8000, and in NEB2 and NEB3, WTBamHI had the highest activity with the highest FI of 32. The overall FI improvement factor, which is the ratio of the FI of each mutant and the WT enzyme in the best buffer, was ≥8000 / 32 = 250-fold.

[0384] 10. Additional Mutations of BamHI

[0385] E163A / E167T / P173A is predicted to have a preferred reduction in star activity and is additionally thermolabile.

[0386] Testing (E86P / K87S / K88P / E163S / E170T / P173A). This mutant showed a 10-fold reduction in specific activity with a compensating increased protein yield from the host cells.

[0387] Other BamHI mutants that share reduced thermostability, reduced star activity, and acceptable specific activity include:

[0388] E86P / K87R / K88G / E163S / E170T / P173A

[0389] E86P / K87P / K88R / E163S / E170T / P173A / E211K

[0390] E86P / K87T / K88R / E163S / E170T / P173A / N158S

[0391] E86P / K87S / K88P / E163S / E170T / P173A

[0392] E86P / K87G / K88S / E163S / E170T / P173A

[0393] E86P / K87R / K88Q / E163S / E170T / P173A

[0394] Example 2: Preparation of High-Fidelity EcoRI

[0395] 1. Expression of EcoRI

[0396] The following primers were used for PCR with EcoRI:

[0397] GGTGGT GCATGC GGAGGTAAATAAATGTCTAATAAAAAACAGTCAAATAGGCTA (SEQ ID NO: 73)

[0398] GGTGGT GGTACC TCACTTAGATCTAAGCTGTTCAAACAA

[0399] (SEQ ID NO: 74)

[0400] The PCR product was then digested with a second pair of restriction endonucleases, SphI and Acc65I, and ligated into pUC19 digested with the same second pair of restriction endonucleases. The ligated plasmid was then transformed into competent E. coli pre-modified with pACYC-MlucIM.

[0401] 2. Mutagenesis with EcoRI

[0402] The initial selection of target amino acid residues was the result of comparison of EcoRI with its isoschizine RsrI, whose star activity is also known.

[0403] EcoRI vs RsrI

[0404]

[0405] Except for D91, E111 and K113, which are known active center residues, 42 charged residues are identical or similar in both endonucleases. The charged residues are as follows:

[0406] K4, R9, K15, K29, H31, D32, E37, E49, R56, R58, K63, E68, K71, D74, K89, E96, K98, K99, R105, H114, D118, K130, D1 33. D135, E144, R145, H147, K148, E152, E160, H162, E170, E177, R183, D185, R200, D202, R203, E253, R264, D269.

[0407] All of these charged residues were mutated to Ala (codons GCA, GCT, GCC or GCG), and the mutated genes were amplified and cloned as follows:

[0408] The amplification mixture was the same as that used in Example 1 (2 μl of each PCR primer, 400 mM dNTPs, 4 units of Deep Vent DNA polymerase, with additional 0, 2, and 6 μl of MgSO 4 10 μl 10x Thermopol buffer was added to the total reaction volume of 100 μl) and added to 1 μl pUC19-EcoRI.

[0409] The PCR reaction conditions were 94°C for 5 min, then 25 cycles of 94°C for 30 sec, 55°C for 30 sec, 72°C for 3 min 30 sec, and a final extension time of 7 min at 72°C. After PCR, the product was purified by a standard Qiagen spin column (Qiagen, Valencia, CA). 16 μl of the PCR product was digested with 20 units of DpnI for 1 hour. The digested product was transformed into a methylase-protected competent E. coli preparation.

[0410] 3. Screening for EcoRI high-fidelity mutants

[0411] Three colonies were picked for each mutation and grown overnight in LB with ampicillin and chloramphenicol. Activity assays were performed on pBR322 and λDNA to ensure that the mutants had at least similar activity to WT EcoRI. These mutants were then tested using 3 μl of 2-fold serial dilutions of cell extract, 12 μl 50% glycerol, 3 μl NEB1 buffer, 0.5 μl pBR322 and 11.5 μl water, which were reacted at 37°C for 1 hour. However, no mutation improved the performance of star activity.

[0412] From this result, it was deduced that effective mutations between isoschisinase enzymes are not always recognized as homologous residues.

[0413] 4. Repeat mutagenesis for the remaining 32 charged residues

[0414] All remaining 32 charged residues were mutated to Ala by targeting amino acid residues 5, 12, 14, 26, 40, 43, 44, 59, 62, 65, 72, 76, 100, 103, 117, 123, 131, 192, 221, 225, 226, 227, 228, 242, 244, 245, 247, 249, 257, 268, 272, and 277 as described in step 2.

[0415] The upper numbers correspond to the amino acid positions in the EcoRI protein sequence (SEQ ID NO: 83).

[0416] 5. Repeat selection

[0417] Four colonies were picked from each sample containing different types of mutations and grown in 4 ml LB with CAM. After sonication, cell extracts were tested on λDNA substrate under normal glycerol conditions in NEB1 buffer. Those extracts with similar activity were tested again on pUC19 substrate by adding 3 μl of cell extracts diluted 2-fold serially to 0.5 μl of pUC19 and 23.5 μl of 50% glycerol in 3 μl NEB2 buffer to provide a final concentration of 39.2% glycerol in the reaction mixture.

[0418] Among all these mutants, K62A was found to be the mutation with the least star activity and high FI. R9A, K15A, R123A, K130A, R131A, R183A mutants all showed a partial reduction in star activity. Interestingly, one clone containing the targeted mutation K5A showed a partial improvement. Additionally, after sequencing, a second mutation S2Y was found. Separation of these two mutations showed that the effective mutation of the isolate was S2Y. D135A and R187A EcoRI also had much less star activity. However, the cleavage activity of these mutants was not optimal.

[0419] 6. Comparison of EcoRI(K62A) and WT EcoRI

[0420] A side-by-side comparison was performed by digesting 0.6 μg of lambda DNA in four different NEB buffers and performing a 3-fold serial dilution using NEB Dilution Buffer C ( FIG. 6 ). EcoRI(K62A) has substantially less star activity than WT EcoRI.

[0421] A more quantitative comparison was made by determining the fidelity index measure for EcoRI(K62A) and WT EcoRI. The conditions for the fidelity index measurement were the same as in Table 2, using λDNA as substrate and using dilution buffer C. The reaction was incubated at 37°C for 1 hour, and the digestion products were analyzed on a 0.8% agarose gel.

[0422] Table 7: Fidelity index of EcoRI(K62A) and WT EcoRI

[0423]

[0424] 7. Further Mutation of EcoRI

[0425] Although EcoRI(K62A) did not clearly have star activity with the lambda DNA substrate, star activity was observed after a 10-hour digestion with the Litmus28 substrate. EcoRI(K62A) in NEB4 had significantly reduced star activity compared to WT EcoRI in EcoRI buffer (Figure 7).

[0426] Further improvements were investigated. EcoRI (K62) was mutated to all other amino acid residues by changing K to the corresponding codon as in Example 1. K62S and K62L are similar to K62A. When compared to EcoRI (K62A), EcoRI (K62E) had a ≥ 100-fold overall fidelity index improvement factor, as shown in Figure 6. EcoRI (K62E) was designated EcoRI-HF.

[0427] 8. Comparison of EcoRI-HF and WT EcoRI

[0428] A quantitative comparison was performed by FI measurement on EcoRI-HF and WT EcoRIA in diluent C. The conditions for FI measurement were the same as in Table 2, using λDNA as substrate. The reaction conditions were 37°C for 1 hour, and the results were analyzed on 0.8% agarose gel (Figure 8).

[0429] Table 8: Comparison of EcoRI-HF and WT EcoRI

[0430]

[0431] An overall fidelity index improvement factor of 64-fold was found (16000 for EcoRI-HF in NEB4 vs. 250 for WTEcoRI in NEB3).

[0432] Example 3: Engineering high-fidelity ScaI

[0433] 1. Expression of ScaI

[0434] The sequences of the ScaI restriction endonuclease and methylase are described in REBASE and GenBank and are provided in Figure 44 (SEQ ID NO: 97). The genes expressing these enzymes are inserted into plasmids to produce pRRS-ScaI and pACYC184-ScaIM, respectively. Then, pACYC184-ScaIM is transformed into competent E. coli host cells. The pRRS vector is derived from pUC19 and differs only in the presence of a multiple cloning site. pRRS-ScaIR is transformed into E. coli (pACYC-ScaIM) to make an expression strain. Plating and cell culture are performed at 30°C.

[0435] 2. Mutagenesis of ScaI

[0436] ScaI has two sequenced isoschizins: LlaDI and NmeSI. However, there is no known information about the star activity of LlaDI or NmeSI, nor is there any information about the active sites of these enzymes. Therefore, all 58 charged residues at positions 4, 8, 11, 12, 14, 18, 25, 27, 30, 37, 39, 40, 43, 46, 51, 57, 61, 68, 72, 74, 80, 86, 97, 103, 108, 112, 114, 119, 120, 121, 127, 128, 129, 133, 135, 139, 140, 141, 147, 152, 156, 158, 159, 161, 162, 171, 172, 175, 179, 182, 184, 187, 172, 175, 192, 193, 195, 200, 222, 227 of the protein were initially selected for targeted mutation.

[0437] The upper numbers correspond to the amino acid positions in the ScaI protein sequence (SEQ ID NO: 97).

[0438] The primer design method and PCR method are similar to those described for BamHI in Example 1 and EcoRI in Example 2. Mutagenesis is achieved by changing the annealing temperature and DNA polymerase. The PCR product is digested with DpnI and transformed into competent E. coli (pACYC184-ScaIM).

[0439] 3. Selection of ScaI high-fidelity mutants

[0440] Four colonies from each mutant of the ScaI mutant were picked and grown overnight at 30°C in 4 ml LB with 100 μg / ml Amp and 33 μg / ml Cam. Each cell culture was sonicated and tested for activity against lambda DNA in NEB2 buffer. Those that were clearly active were retested at 10, 100, and 1000-fold dilutions. Because ScaI has very significant star activity, the star activity bands are easily comparable for the mutants versus the WT restriction endonuclease. Those with reduced star activity were retested with a 2-fold serial dilution using NEB dilution buffer A. The FI was measured for each mutant. For WT ScaI in NEB 2 buffer, the FI was 1 / 8. Four mutants with similar activity levels were found to have significantly reduced star activity compared to WT ScaI. Mutant #6-3 ScaI had 2-fold activity compared to WT ScaI and a FI of 4 or 32-fold. #26-2 ScaI has 2-fold activity compared to WT and FI is 8 or 64-fold; #28-2 ScaI has 2-fold activity compared to WT and FI is 120 or 1000-fold; #54-3 has similar activity to WT and FI is 250 or 2000-fold that of WT.

[0441] Four mutants: #6-3, #26-2, #28-2 and #54-3 ScaI were further tested digesting a lambda DNA substrate in the presence of 36.7% glycerol. #54-3 showed a greater improvement in reducing star activity than the other three mutants.

[0442] After plasmid extraction, #6-3 was sequenced and found to have a mutation of R18A. #26-2 was sequenced and found to have a mutation of R112A. #28-2 was sequenced and found to have a mutation of E119A. These mutations were predicted. However, #54-3 was found to have a double mutant, H193A / S201F. S201F was a spontaneous second mutation that occurred during PCR and was located outside the primer region of the H193A mutation.

[0443] To understand which residue was primarily responsible for the reduction in star activity, a single mutation (S201F) was introduced into ScaI using the following primers:

[0444] 5'-GATTGGGTGGCGCAGAAATTTCAAACGGGCCAGCAGTCG-3'

[0445] (SEQ ID NO: 77)

[0446] 5'-CGACTGCTGGCCCGTTTGAAATTTCTGCGCCACCCAATC-3'

[0447] (SEQ ID NO: 78).

[0448] The sequences of ScaI (H193A), ScaI (S201F), and ScaI (H193A / S201F) were confirmed. The three mutants were compared to WT ScaI at 5% and 37% glycerol levels (Figure 9). In contrast to H193A, which only weakly contributed to FI, S201F significantly contributed to FI. However, the two mutations appeared additive in improving FI. In 5% glycerol, S201F showed no star activity, but showed some star activity in 37% glycerol. H193A had some star activity in 5% glycerol and significant star activity in 37% glycerol. However, for the combination of these two mutations, no star activity was detected in either 5% or 37% glycerol. This finding not only shows that amino acids with hydroxyl groups can be the main active residues for star activity, but also that the correct combination of mutations can drive fidelity to very high levels. If mutation of charged residues failed to improve star activity, it was observed here that mutations to Ser, Thr and Tyr could successfully improve the fidelity index. ScaI (H193A / S201F) was labeled ScaI-HF.

[0449] 4. Comparison of ScaI-HF and WT ScaI

[0450] ScaI-HF and WT ScaI were compared on 1 μg of λ DNA in 4 different NEB dilutions at 37°C in different NEB buffers at 2.5-fold serial dilutions with NEB dilution buffer A for 1 hour ( FIG. 10 ).

[0451] Table 9: Comparison of the fidelity index of ScaI-HF and WT ScaI

[0452]

[0453]

[0454] In NEB2 and NEB4 buffers, ScaI-HF performed best, with the best FI being 250, and in NEB2 buffer, WT ScaI performed best, with an FI of 1 / 8. The overall FI improvement factor was 250 / (1 / 8)=4000.

[0455] Example 4: Engineering high-fidelity SalI

[0456] 1. Expression of SalI

[0457] SalI was expressed in E. coli transformed with placzz1-SalIR and pACYC-Hpy166IIM, where placzzI is a pUC19 plasmid that uses the lac promoter to express a restriction endonuclease gene inserted into an adjacent multicopy site. Hpy166IIM protects the outer 4 bases of SalI.

[0458] 2.SalI mutagenesis

[0459] Using a PCR method similar to that in the previous example, 86 charged residues of SalI were mutated to Ala: 5, 6, 8, 9, 12, 13, 19, 27, 31, 34, 35, 37, 42, 43, 45, 50, 60, 63, 65, 67, 73, 82, 83, 84, 90, 93, 97, 100, 101, 103, 107, 109, 111, 114, 116, 119, 126, 129, 131, 134, 140, 143 , 145, 147, 148, 156, 157, 164, 168, 172, 173, 174, 180, 181, 186, 190, 191, 193, 210, 218, 226, 232, 235, 237, 238, 244, 246, 250, 256, 257, 258, 259, 260, 261, 264, 266, 271, 275, 297, 300, 304, 305, 306, 308, 309, 311.

[0460] The upper numbers correspond to the amino acid positions in the SalI protein sequence (SEQ ID NO: 94).

[0461] Mutants were grown overnight at 30°C in LB with Amp and Cam.

[0462] 3. Selection of SalI-HF

[0463] Selection of SalI-HF was performed as described in the previous examples. The main difference is that the star activity of SalI cannot be easily analyzed in crude extracts - either in 5% glycerol or higher glycerol concentrations. Glycerol not only promotes the star activity of SalI, but also greatly inhibits the homologous activity.

[0464] Active mutants were analyzed on HindIII digested lambda DNA in 5% glycerol and 37% glycerol. Mutants #22, #26, #29, #31, #43, and #51 were tested for cleavage activity in all 4 NEB buffers. After several rounds of comparisons in different conditions and substrates, #31, SalI(R107A), was found to be the preferred mutant, which retained high cleavage activity but showed much reduced star activity. SalI(R107A) was labeled SalI-HF.

[0465] 4. Comparison of SalI-HF and WT SalI

[0466] The FI of SalI-HF and WT SalI was determined ( FIG. 11 ). The results are shown in Table 10 (below):

[0467] Table 10: Comparison of SalI-HF and WT SalI

[0468]

[0469] In NEB 2 and NEB 4 buffers, SalI-HF performed best, with FIs ≥ 2000, and WT SalI performed best in NEB 3 buffer, with an FI of 4. The overall FI improvement factor was ≥ 2000 / 4 = ≥ 500.

[0470] Example 5: Engineering high-fidelity SphI

[0471] 1. Expression of SphI

[0472] SphI was expressed in E. coli (placzz1-SphIR, pACYC184-CviAIIM). CviAIIM protects the internal four bases of SphI. The transformed cells were grown overnight at 37°C in LB with Amp and Cam.

[0473] 2. Mutagenesis of SphI

[0474] Using the methods described in Examples 1 to 4, all charged residues of SphI were mutated to Ala. A total of 71 mutations were made: 3, 5, 12, 18, 21, 24, 25, 30, 31, 35, 43, 46, 51, 54, 57, 58, 60, 61, 72, 75, 77, 78, 87, 90, 91, 95, 100, 104, 107, 108, 110, 113, 120, 123, 124, 125, 129, 130 , 131, 139, 140, 142, 146, 147, 155, 157, 159, 164, 170, 172, 173, 175, 178, 184, 186, 190, 194, 196, 197, 198, 206, 207, 209, 212, 215, 221, 227, 230, 231, 232, 235.

[0475] The upper numbers correspond to the amino acid positions in the SphI protein sequence (SEQ ID NO: 98).

[0476] 3. Selection of SphI-HF

[0477] Four colonies of each mutation were grown overnight at 37°C in LB with Amp and Cam. Activity selection was primarily against pBR322 in 5% glycerol and 30% glycerol in NEB2. Using the experience of the previous examples, selection of high-fidelity SphI was simple. It was found that SphI mutants D91A, K100A, D139A, and D164A significantly reduced the star activity in SphI. Among them, K100A was the preferred mutation with minimal star activity. SphI (K100A) was named SphI-HF.

[0478] 4. Comparison of SphI-HF and WT SphI

[0479] Comparison of SphI-HF and WT SphI was performed in parallel in their respective preferred buffers. SphI-HF was serially diluted 2-fold using NEB Dilution Buffer A and reacted in NEB4 with WT SphI serially diluted 2-fold using NEB Dilution Buffer B. Digestion of lambda DNA is compared in FIG12 .

[0480] Table 11: Comparison of FI of SphI-HF and WT SphI

[0481]

[0482] In NEB4, SphI-HF performed best, with FI ≥ 2000; WT SphI performed best in NEB1 or NEB2, with the preferred FI being 64. The overall FI improvement factor was ≥ 32.

[0483] Example 6: Engineering high-fidelity PstI

[0484] 1. Expression of PstI

[0485] PstI was expressed from E. coli (pACYC-HpyCH4VM, pPR594-PstIR). HpyCH4VM protected the internal four bases of PstI. pPR594 is an expression vector with Amp resistance and ptac promoter. Cells were grown at 30°C in LB with Amp and Cam, and the culture was then induced overnight by IPTG.

[0486] 2. Mutagenesis of PstI

[0487] Using the methods described in the previous examples, 92 charged residues were mutated to Ala. These are: 8, 10, 11, 14, 25, 26, 38, 40, 41, 44, 45, 47, 58, 61, 63, 66, 67, 69, 73, 74, 77, 78, 82, 85, 88, 91, 92, 94, 95, 99, 104, 105, 116, 119, 127, 128, 136, 142, 145, 146, 150, 151, 152, 156, 159, 169, 170, 174, 176, 177 79, 180, 184, 188, 191, 197, 202, 204, 207, 212, 214, 217, 218, 226, 227, 228, 231, 236, 237, 238, 239, 240, 246, 251, 257, 258, 261, 263, 273, 282, 284, 286, 287, 295, 297, 302, 305, 306, 309, 313, 314, 319 and 320.

[0488] The upper numbers correspond to the amino acid positions in the PstI protein sequence (SEQ ID NO: 91).

[0489] After digestion of the PCR product with DpnI, the sample was transformed into competent E. coli (pACYC-HpyCH4VM) and grown on LB plates with Amp and Cam.

[0490] 3. Selection of PstI-HF

[0491] Selection of PstI-HF was similar to the previous samples. Normal active enzyme activity was detected on λDNA at 5% glycerol, and star activity was detected on pBR322 substrate under the conditions of NEB4 buffer and 20% DMSO. DMSO significantly enhanced star activity more than the same concentration of glycerol. During the selection period, #26, #56 and #65 had reduced star activity compared to WT. When each was sequenced, the mutations were found to be D91A, E204G and K228A / A289V. The mutant #26PstI (D91A) was labeled PstI-HF.

[0492] 4. Comparison of PstI-HF and WT PstI

[0493] The FI of PstI-HF and WT PstI were measured against lambda DNA substrate in NEB 1-4 buffers, respectively. The dilution buffer was NEB dilution buffer C. The comparison is shown in Figure 13 and the results are listed in Table 12 (below).

[0494] Table 12: Comparison of PstI-HF and WT PstI

[0495]

[0496] In NEB2 and NEB4 buffers, PstI-HF performed best, with a preferred FI of ≥2000; WT PstI performed best in NEB 3 buffer, with a FI of 120. The overall FI improvement factor was ≥2000 / 120 = 16-fold.

[0497] Example 7: Engineering high fidelity NcoI

[0498] 1. Expression of NcoI

[0499] Expression of NcoI was achieved in E. coli (pSYX20-NcoIM, pRRS-NcoIR). pRRS is a pUC19 derivative plasmid and pSYX20 is a low copy number plasmid compatible with the pRRS vector. Cells were grown overnight at 30°C in LB with Amp and kanamycin (Kan).

[0500] 2. Mutagenesis of NcoI

[0501] All 66 charged residues in NcoI were mutated to Ala. These residues are: 7, 8, 19, 22, 27, 30, 31, 32, 33, 37, 39, 42, 46, 55, 56, 61, 62, 64, 68, 69, 75, 84, 88, 89, 92, 93, 95, 97, 100, 116, 136, 144, 146, 162, 166, 170, 178, 183, 185, 187, 188, 189, 196, 199, 202, 204, 209, 211, 212, 213, 216, 219, 227, 229, 237, 241, 244, 250, 251, 257, 259, 261, 268, 279, 282, 285.

[0502] The upper numbers correspond to the amino acid positions in the NcoI protein sequence (SEQ ID NO: 88).

[0503] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pSYX20-NcoIM).

[0504] 3. Selection of NcoI-HF

[0505] Selection of NcoI-HF was similar to that of PstI-HF. Activity was detected as described above with lambda DNA as substrate at 5% glycerol. Star activity was determined in 19% DMSO using pBR322 or lambda. The following mutations were found to improve star activity: A2T / R31A, D56A, H143A, E166A, R212A, and D268A. Among these mutants, NcoI (A2T / R31A) was selected as NcoI-HF.

[0506] 4. Comparison of NcoI-HF and WT NcoI

[0507] The FI of NcoI-HF and WT NcoI were determined separately on lambda DNA in NEB1-4 buffer. The comparison is shown in Figure 14 and the results are listed in Table 13 (below).

[0508] Table 13: Comparison of NcoI-HF and WT NcoI

[0509]

[0510] In NEB4, Ncol-HF showed the greatest reduction in star activity, with a preferred FI of ≥16000; WT Ncol performed best in NEB1, NEB2, and NEB4, with a preferred FI of 120. The overall FI improvement factor was ≥16000 / 120=125.

[0511] Example 8: Engineering high fidelity NheI

[0512] 1. Expression of NheI

[0513] NheI was expressed in E. coli transformed with pACYC-NheIM and placzz1-NheIR. placzz1 is a pUC19 derived plasmid. Cells were grown overnight at 30°C in LB with Amp and Cam.

[0514] 2. Mutagenesis of NheI

[0515] All 92 charged residues in NheI were mutated to Ala, which are the following residues: 5, 6, 7, 14, 17, 19, 22, 25, 28, 31, 38, 39, 42, 47, 49, 52, 56, 58, 59, 60, 64, 74, 75, 76, 77, 80, 91, 93, 104, 105, 110, 112, 116, 117, 123, 126, 130, 131, 133, 135, 137, 147, 149, 152, 159, 160 , 165, 167, 170, 171, 174, 179, 183, 195, 202, 205, 207, 209, 210, 211, 214, 216, 218, 221, 225, 231, 241, 243, 244, 250, 252, 256, 257, 259, 264, 266, 267, 281, 285, 287, 288, 289, 291, 297, 300, 307, 313, 315, 318, 321, 324, 325.

[0516] The upper numbers correspond to the amino acid positions in the NheI protein sequence (SEQ ID NO: 89).

[0517] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-NheIM).

[0518] 3. Selection of NheI-HF

[0519] Selection of NheI-HF was performed according to the previous example. Standard and star activity assays contained NEB4 buffer and pBR322 in 5% glycerol and 39% glycerol, respectively, as substrate. Only one mutation was found to significantly improve NheI. This was E77A. NheI (E77A) was selected as NheI-HF.

[0520] 4. Comparison of NheI-HF and WT NheI

[0521] The FI of NheI-HF and WT NheI were determined separately on pXba, a plasmid substrate containing an XbaI digested fragment from adenovirus, in each of NEB1-4 buffers. The comparison is shown in Figure 15 and the results are listed in Table 14 (below).

[0522] Table 14: Comparison of NheI-HF and WT NheI

[0523]

[0524] In NEB1 buffer, NheI-HF showed the best activity, where its FI was > 128,000. WT NheI had the greatest activity in NEB1 and NEB4 buffers, where its best FI was 32. Therefore, the overall FI improvement factor was > 128,000 / 32 = > 4000.

[0525] Example 9: Engineering high fidelity SspI

[0526] 1. Expression of SspI

[0527] SspI was expressed from E. coli transformed with pACYC-SspIM and placzz1-SspIR. placzz1 is a pUC19 derived plasmid. Cells were grown overnight at 30°C in LB with Amp and Cam.

[0528] 2. Mutagenesis of SspI

[0529] All 81 charged residues in SspI were mutated to Ala; these are: 3, 8, 12, 13, 18, 19, 20, 35, 40, 42, 44, 47, 52, 60, 62, 65, 68, 69, 72, 74, 76, 77, 78, 79, 83, 85, 88, 89, 90, 96, 100, 108, 109, 118, 119, 127, 128, 129, 131, 132, 137, 144, 153, 154, 155, 156, 158, 165, 168, 170, 172, 177, 178, 179, 181, 185, 186, 187, 191, 194, 195, 197, 202, 204, 215, 222, 229, 237, 240, 246, 250, 256, 257, 259, 260, 264, 265, 267, 268, 269, 274.

[0530] The upper numbers correspond to the amino acid positions in the SspI protein sequence (SEQ ID NO: 99).

[0531] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-SspIM).

[0532] 3. Selection of SspI high-fidelity mutants

[0533] Standard homology and star activity assays of NheI were performed using ΦX174 substrate in NEB4 buffer and 5% glycerol and 39% glycerol, respectively. Mutants #16 (H65A), #20 (K74A), #23 (E78A), #26 (E85A), #28 (E89A), #33 (K109A), #34 (E 118A), #52 (R177A), #62 (K197A), #67 (D229A) all showed reduced star activity. K109A showed the greatest reduction in star activity. It was decided to seek further improvements in star activity.

[0534] 4. Further mutation

[0535] All residues originally identified as Tyr were mutated to Phe, while the other residues Cys, Phe, Met, Asn, Gln, Ser, Thr, and Trp were mutated to Ala. The panel included 95 residue mutations at the following positions: 2, 6, 7, 9, 10, 13, 22, 25, 26, 27, 29, 30, 32, 33, 34, 39, 41, 51, 53, 55, 56, 57, 58, 59, 61, 63, 71, 75, 81, 84, 87, 91, 94, 98, 104, 106, 107, 110, 111, 113, 114, 123, 125, 134, 136, 139, 140, 141, 142, 143 , 146, 152, 157, 159, 160, 164, 173, 175, 180, 183, 190, 192, 193, 196, 198, 199, 201, 205, 207, 211, 214, 218, 219, 220, 221, 223, 225, 226, 227, 228, 230, 232, 233, 235, 238, 239, 241, 249, 254, 255, 272, 275, 276, 277, 280.

[0536] The upper numbers correspond to the amino acid positions in the SspI protein sequence (SEQ ID NO: 113).

[0537] By the same method as above, PCR and selection were performed. Among these mutants, Y98F was found to have the least star activity and was better than SspI(K109A) in this regard. SspI(Y98F) was labeled as SspI-HF and stored as a production strain.

[0538] 5. Comparison of SspI-HF and WT SspI

[0539] The FI of SspI-HF and WT SspI were determined separately using lambda DNA substrate in NEB 1-4 buffer. The diluent was NEBC. The comparison is shown in Figure 16 and the results are listed in Table 15 (below).

[0540] Table 15: Comparison of SspI-HF and WT SspI

[0541]

[0542] In NEB4, SspI-HF performed best, with a preferred FI of 500, and WT SspI performed best in NEB1, NEB2, and NEB4, with a preferred FI of 64. The overall FI improvement factor was 500 / 64=8.

[0543] Example 10: Engineering High Fidelity NotI

[0544] 1. Expression of NotI

[0545] NotI has significant star activity in NEB4 buffer and less star activity in NEB3 buffer. NotI was engineered to reduce star activity in any NEB buffer. NotI was expressed in competent E. coli transformed with pACYC184-EagIM and placzz2-NotIR. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0546] 2. Mutagenesis of NotI

[0547] All 97 charged residues in NotI were mutated to Ala, such as the following residues: 2, 4, 8, 10, 17, 21, 22, 26, 31, 34, 35, 36, 49, 52, 57, 59, 62, 72, 74, 75, 77, 84, 87, 96, 97, 105, 117, 121, 122, 125, 126, 129, 130, 133, 140, 141, 145, 150, 152, 156, 160, 165, 167, 174, 176, 177, 182, 187, 189 , 193, 194, 200, 205, 208, 210, 219, 224, 225, 227, 236, 237, 245, 251, 253, 267, 271, 272, 280, 283, 290, 292, 294, 296, 304, 306, 308, 310, 314, 319, 321, 323, 327, 331, 335, 336, 339, 353, 354, 356, 358, 361, 365, 367, 368, 369, 370, 378, 382.

[0548] The upper numbers correspond to the amino acid positions in the NotI protein sequence (SEQ ID NO: 90).

[0549] The method for introducing the mutant into the enzyme was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli containing pACYC-EagIM.

[0550] 3. Selection of NotI-HF

[0551] Selection for NotI-HF was performed as described in the previous example. Standard cognate and star activity assays used pXba substrate in NEB4 buffer with 5% glycerol and NEB ExoI buffer (67 mM glycine-KOH, pH 9.5, 6.7 mM MgCl2 , 10 mM 2-mercaptoethanol) and 37% glycerol. #37 (K150A), #44 (K176A), #45 (R177A), #63 (R253A) all showed reduced star activity. K150A is the preferred mutation that reduces star activity. NotI (K150A) was selected as NotI-HF.

[0552] 4. Comparison of NotI-HF and WT NotI

[0553] The FI of NotI-HF and WT NotI were determined separately using pXba substrate in NEB1-4 buffer. The comparison is shown in Figure 17 and the results are listed in Table 16 (below).

[0554] Table 16: Comparison of NotI-HF and WT NotI

[0555]

[0556] ND: Not detectable, because both FIs are out of range.

[0557] In NEB2, NotI-HF performed best, with a preferred FI of ≥128,000; WT NheI performed best in NEB3, with a preferred FI of 4000. The overall fidelity index improvement factor was ≥128,000 / 4000 = ≥32. Engineering NotI not only further improved the FI of NotI, but also changed the optimal buffer.

[0558] Example 11: Engineering high-fidelity SacI

[0559] 1. Expression of SacI

[0560] SacI was expressed in E. coli transformed with pLG-SacIM and pRRS-SacIR. pRRS is a pUC19 derivative plasmid and pLG is a low copy compatible plasmid. Cells were grown overnight at 30°C in LB with Amp and Kan.

[0561] 2. Mutagenesis of SacI

[0562] All 101 charged residues in SacI were mutated to Ala, such as the following residues: 6, 7, 11, 15, 16, 19, 24, 25, 29, 30, 39, 40, 42, 45, 58, 61, 62, 63, 65, 67, 70, 71, 72, 74, 75, 76, 81, 85, 94, 98, 104, 105, 114, 116, 120, 123, 127, 129, 133, 134, 141, 143, 144, 145, 146, 150, 151, 154, 169, 170, 172, 181, 187, 196, 197, 200, 201, 211, 216, 220, 221, 224, 227, 228, 232, 238, 240, 246, 248, 250, 258, 270, 271, 277, 281, 288, 289, 295, 296, 297, 299, 303, 306, 313, 314, 321, 322, 324, 332, 336, 337, 340, 342, 344, 345, 347, 349, 350, 353, 357.

[0563] The upper numbers correspond to the amino acid positions in the Sad protein sequence (SEQ ID NO: 93).

[0564] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pLG-SacIM).

[0565] 3. Selection of SacI-HF

[0566] Using a method similar to the previous example, selection of SacI-HF was completed. Standard activity checks used pUC19 in NEB4 with 5% glycerol, and star activity checks were performed on pUC 19 at 39% glycerol in NEB4 buffer. #52 SacI (Q117H / R154A / L284P) and #60 SacI (Q117H / R200A) both had reduced star activity, and SacIQ117H / R200A proved to be the preferred mutation. Q117H is a carryover mutation from the template that does not affect the activity of SacI. SacI (Q117H / R200A) was selected as SacI-HF.

[0567] 4. Comparison of SacI-HF and WT SacI

[0568] The FI of Sad-HF and WT Sad were determined separately on pXba substrate in NEB 1-4 buffer. The comparison is shown in Figure 18 and the results are listed in Table 17 (below).

[0569] Table 17: Comparison of SacI-HF and WT SacI

[0570]

[0571] In NEB4, SacI-HF performed best, with an FI of 4000; WT SacI performed best in NEB1 and NEB4, with a preferred FI of 120. The overall FI improvement factor was 4000 / 120=32.

[0572] Example 12: Engineering high-fidelity PvuII

[0573] 1. Expression of PvuII

[0574] PvuII was expressed in E. coli transformed with pACYC-PvuIIM and placzz2-PvuIIR. Placzz2 is a pUC19 derivative plasmid; pACYC is a low copy compatible plasmid. Cells were grown overnight at 30°C in LB with Amp and Cam.

[0575] 2. Mutagenesis of PvuII

[0576] All 47 charged residues in PvuII were mutated to Ala, such as the following residues: 3, 5, 8, 11, 15, 18, 21, 25, 26, 30, 34, 38, 54, 55, 58, 61, 66, 68, 70, 75, 78, 83, 84, 85, 93, 95, 105, 110, 114, 116, 118, 119, 121, 125, 126, 128, 129, 130, 134, 136, 137, 138, 143, 147, 151, 152, and 155.

[0577] The upper numbers correspond to the amino acid positions in the PvuII protein sequence (SEQ ID NO: 92).

[0578] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-PvuIIM).

[0579] 3. Selection of PvuII High-Fidelity Mutants

[0580] Selection of PvuII-HF was similar to the previous example. Standard activity assays were performed on pBR322 using lambda DNA substrate in NEB4 with 5% glycerol, and star activity assays were performed at 39% glycerol in NEB4 buffer. No mutants qualified as high fidelity PvuII.

[0581] 4. Additional mutagenesis steps

[0582] An additional mutagenesis step was to mutate all Ser, Thr to Ala and Tyr to Phe in PvuII. The positions mutated were: 2, 19, 46, 49, 67, 71, 77, 81, 82, 94, 104, 113, 123, 124, 132, 133, 148, 154 and 157.

[0583] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-PvuIIM).

[0584] PvuII(T46A) appeared to have less star activity than WT PvuII, however, further improvement is desired.

[0585] T46 was mutated to all other amino acid residues by changing the codon to the corresponding amino acid. Among all these mutations, T46H, T46K, T46Y, T46G were better than T46A. T46G was selected as PvuII-HF.

[0586] 5. Comparison of PvuII-HF and WT PvuII

[0587] The FI of PvuII-HF and WT PvuII on pBR322 were determined separately in NEB 1-4 buffer using diluent A. The comparison is shown in Figure 19 and the results are listed in Table 18 (below).

[0588] Table 18: Comparison of PvuII-HF and WT PvuII

[0589]

[0590] In NEB4, PvuII-HF performed best, with an FI of 500, and WT PvuII performed best in NEB1 and NEB4, with a preferred FI of 250. The overall FI improvement factor was 500 / 250 = 2. Although the overall FI improvement factor was not high for PvuII, the FI was improved 2000-fold in NEB4.

[0591] Example 13: Engineering high-fidelity MfeI

[0592] 1. Expression of MfeI

[0593] MfeI was expressed in E. coli transformed with pACYC-MluCIM and pRRS-MfeIR. pRRS is a pUC19 derivative plasmid; pACYC is a low copy compatible plasmid. MluCIM methylates AATT, which is an internal four nucleic acid sequence of MfeI. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0594] 2. Mutagenesis of MfeI

[0595] The mutagenesis of MfeI was performed in three batches. The first batch was all charged residues, which were mutated to Ala, such as the following amino acid positions: 5, 9, 19, 24, 36, 39, 44, 45, 47, 48, 50, 60, 61, 64, 65, 72, 83, 87, 90, 92, 93, 98, 100, 101, 103, 107, 109, 110, 115, 119, 120, 121, 124, 132, 135, 14 2, 143, 144, 153, 155, 158, 159, 161, 162, 164, 165, 171, 172, 175, 181, 184, 187, 188, 192, 195, 196, 198, 199, 200; the second group is all residues with hydroxyl groups: Ser, Thr and Tyr, wherein Ser and Thr are changed to Ala, and Tyr is changed to Phe. The residues are: 4, 7, 21, 28, 38, 40, 43, 53, 74, 75, 76, 81, 89, 91, 112, 122, 127, 134, 136, 157, 167, 170, 173, 177, 185 and 200. The third batch is residues Cys, Phe, Met, Asn, Gln, Trp, all changed to Ala, the residues are: 10, 12, 13, 25, 26, 29, 31, 32, 35, 51, 55, 67, 68, 77, 78, 84, 88, 96, 102, 105, 117, 123, 126, 141, 148, 149, 152, 168, 169, 174, 176, 178, 179, 180, 183, 191, 193, 194.

[0596] The upper numbers correspond to the amino acid positions in the MfeI protein sequence (SEQ ID NO: 5).

[0597] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-MluCIM).

[0598] 3. Selection of MfeI-HF

[0599] Selection of MfeI-HF was accomplished using methods similar to the previous examples. In NEB4, cleavage activity was determined using ΦX174 substrate with 5% glycerol, and in NEB4, star activity was determined using ΦX174 substrate with 39% glycerol. A significant difficulty with this enzyme is that many mutations improve the cleavage activity of the enzyme and reduce star activity, but require higher glycerol concentrations than the WT enzyme. MfeI(K50A) is an example, which has reduced star activity and high cleavage activity in high glycerol concentrations, while the activity is low in lower glycerol concentrations. MfeI(Y173A) also reduces star activity. The preferred mutation is Q13A / F35Y. The mutation of F35Y comes from the template, and Q13A is a targeted mutation. MfeI(Q13A / F35Y) is labeled MfeI-HF.

[0600] 4. Comparison of MfeI-HF and WT MfeI

[0601] The FI of MfeI-HF and WT MfeI were determined separately on a lambda DNA substrate in NEB 1-4 buffers, with dilutions performed in NEB Diluent A. The comparison is shown in Figure 20 and the results are listed in Table 19 (below).

[0602] Table 19: Comparison of MfeI-HF and WT MfeI

[0603]

[0604] In NEB1 and NEB4, MfeI-HF performed best, with a preferred FI of ≥1000; in NEB1 and NEB4, WTMfeI performed best, with a preferred FI of 32. The overall FI improvement factor was ≥1000 / 32=32 times.

[0605] Example 14: Engineering high fidelity HindIII

[0606] 1. Expression of HindIII

[0607] HindIII was expressed in E. coli transformed with pUC19-HindIIIRM, which contains the HindIII endonuclease and methylase genes. Cells were grown overnight at 30°C in LB with Amp.

[0608] 2. HindIII mutagenesis

[0609] 88 charged residues in HindIII were mutated to Ala. These are: 2, 3, 7, 8, 14, 20, 22, 34, 37, 39, 42, 45, 52, 55, 61, 62, 66, 69, 74, 84, 87, 89, 94, 100, 101, 109, 111, 114, 117, 120, 123, 124, 126, 128, 132, 134, 135, 136, 137, 138, 153, 158, 162, 163, 171, 172, 180 , 182, 183, 190, 197, 198, 201, 202, 207, 209, 214, 215, 218, 222, 225, 227, 228, 229, 237, 238, 243, 244, 245, 249, 250, 251, 254, 255, 261, 265, 266, 267, 270, 274, 275, 281, 283, 286, 290, 293, 296, 297.

[0610] in positions 4, 11, 15, 17, 18, 19, 21, 23, 26, 27, 30, 31, 36, 38, 46, 57, 58, 59, 60, 63, 64, 76, 77, 80, 82, 83, 88, 91, 99, 102, 103, 104, 112, 113, 116, 118, 121, 122, 125, 131, 133, 139, 143, 146, 147, 148, 149, 151, 152, 154, 155, 157, 159, 160, 164, 168, 169, 170, 178, 184, 185, 187, 188, 1 89, 191, 193, 194, 195, 199, 200, 203, 204, 206, 210, 211, 212, 213, 216, 217, 219, 220, 221, 224, 230, 232, 233, 236, 240, 241, 246, 252, 253, 256, 258, 262, 263, 264, 277, 278, 279, 280, 284, 287, 288, 294, 295, 299, all residues Cys, Met, Asn, Gln, Ser, Thr, Trp were changed to Ala, and Tyr was changed to Phe.

[0611] The upper numbers correspond to the amino acid positions in the HindIII protein sequence (SEQ ID NO: 85).

[0612] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli strain ER3081.

[0613] 3. Selection of HindIII-HF

[0614] Selection of HindIII-HF was accomplished using a method similar to the previous example. Standard activity checks used λDNA in NEB4 with 5% glycerol, and star activity was measured using λDNA substrate in NEB4 buffer with 39% glycerol. Two mutants of HindIII were found to have reduced star activity. These are HindIII(K198A) and S188P / E190A. HindIII(K198A) was labeled HindIII-HF.

[0615] 4. Comparison of HindIII-HF and WT HindIII

[0616] The FI of HindIII-HF and WT HindIII were determined separately using lambda DNA substrate in each of NEB 1-4 buffers with diluent B. The comparison is shown in Figure 21 and the results are listed in Table 20 (below).

[0617] Table 20: Comparison of HindIII-HF and WT HindIII

[0618]

[0619] In NEB2, HindIII-HF performed best, with a preferred FI of ≥64000; WT HindIII performed best in NEB2, with a preferred FI of 250. The overall FI improvement factor was 4000 / 120=32.

[0620] Example 15: Engineering high-fidelity SbfI

[0621] 1. Expression of SbfI

[0622] SbfI was expressed in E. coli transformed with pUC19-SbfIRM. Cells were grown overnight at 30°C in LB with Amp.

[0623] 2. Mutagenesis of SbfI

[0624] 78 charged residues in SbfI were mutated to Ala. These residues are: 5, 8, 15, 18, 23, 27, 30, 34, 46, 49, 50, 53, 58, 63, 66, 70, 71, 74, 81, 82, 83, 85, 86, 87, 90, 94, 103, 115, 120, 121, 127, 132, 135, 136, 143, 144, 147, 150, 152, 154, 164, 169, 170, 183, 184, 187, 1 88, 192, 196, 204, 206, 208, 213, 214, 215, 218, 219, 226, 228, 230, 233, 237, 238, 239, 241, 248, 251, 253, 257, 258, 259, 260, 262, 266, 282, 284, 285, 288, 293, 297, 299, 304, 305, 307, 311, 316 and 322.

[0625] The Ser and Thr residues in SbfI were also mutated to Ala. Tyr was mutated to Phe. The following positions were targeted: 3, 4, 5, 10, 13, 16, 31, 35, 38, 54, 55, 56, 68, 76, 78, 80, 88, 109, 111, 116, 119, 129, 131, 137, 146, 162, 174, 197, 198, 201, 205, 210, 224, 252, 263, 270, 272, 286, 298, 315, 321.

[0626] An additional 55 Cys, Phe, Met, Asn, Gln, Trp residues were also mutated to Ala at the following positions: 2, 24, 26, 29, 32, 51, 62, 65, 67, 72, 84, 91, 92, 95, 97, 101, 104, 106, 110, 112, 114, 117, 124, 134, 140, 157, 160, 171, 178, 179, 185, 189, 193, 212, 217, 225, 231, 243, 245, 247, 256, 265, 268, 277, 279, 280, 281, 283, 287, 289, 290, 296, 301, 313, and 317.

[0627] The upper numbers correspond to the amino acid positions in the SbfI protein sequence (SEQ ID NO: 96).

[0628] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The mutated product was then transformed into E. coli strain ER2984.

[0629] 3. Selection of SbfI-HF

[0630] Selection for SbfI-HF was done as described in the previous examples. Standard activity assays used lambda DNA in NEB4 with 5% glycerol and star activity assays were performed against lambda DNA in Exonuclease I buffer. SbfI (K251A) was labeled SbfI-HF.

[0631] 4. Comparison of SbfI-HF and WT SbfI

[0632] The FI of SbfI-HF and WT SbfI were determined separately on lambda DNA in NEB 1-4 buffer using diluent C. The comparison is shown in Figure 22 and the results are listed in Table 21 (below).

[0633] Table 21: Comparison of SbfI-HF and WT SbfI

[0634]

[0635] In NEB1 and NEB4, SbfI-HF performed best, where the preferred FI was 1000; WT SbfI performed best in NEB1, where the preferred FI was 8. The overall FI improvement factor was 1000 / 8 = 125 times.

[0636] Example 16: Engineering high-fidelity EagI

[0637] 1. Expression of EagI

[0638] EagI was expressed in E. coli transformed with pBR322-EagIRM. Cells were grown overnight at 30°C in LB with 20 μg / ml tetracycline.

[0639] 2. Mutagenesis of EagI

[0640] Asp, Glu, His, Lys, Arg, Ser, Thr, Asn, and Gln residues were mutated to Ala. Tyr was mutated to Phe. These are the following residues: 2, 3, 4, 5, 6, 9, 13, 14, 17, 19, 21, 23, 27, 35, 36, 37, 40, 42, 43, 44, 45, 46, 49, 51, 53, 55, 56, 58, 60, 66, 67, 69, 71, 72, 73, 74, 75, 77, 78, 80, 82, 86, 87, 92, 93, 94, 95, 98, 99, 100, 102, 103, 104, 105, 112, 113, 114, 116, 117, 119, 122, 125, 127, 132, 134, 135, 137, 139, 140, 141, 145, 147, 148, 150, 152, 154, 155, 156, 157, 160, 162, 163 , 164, 166, 169, 172, 173, 176, 177, 178, 179, 182, 185, 187, 188, 189, 193, 196, 197, 201, 202, 203, 204, 205, 206, 208, 209, 212, 217, 220, 221, 222, 224, 225, 230, 235, 236 , 237, 238, 239, 240, 241, 243, 245, 246, 247, 248, 251, 255, 257, 258, 259, 260, 263, 264, 265, 266, 270, 272, 273, 275, 276, 277, 279, 280, 283, 286, 288, 289, 291, 295, 296.

[0641] The upper numbers correspond to the amino acid positions in the EagI protein sequence (SEQ ID NO: 82).

[0642] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli strain ER 3081 and grown on LB agar plates with tetracycline.

[0643] 3. Selection of EagI-HF

[0644] Selection of EagI-HF was accomplished using a different approach than in the previous examples, which used high concentrations of glycerol, high concentrations of DMSO, or high pH. In crude extracts, purification of each mutant to check for star activity would have been very tedious because expression was too low to show star activity. Based on the previous examples, it was inferred that the HF endonuclease tended to have increased cleavage activity in NEB4 compared to NEB3. Therefore, the activity of EagI in crude extracts was measured in both NEB3 and NEB4; the one with the highest NEB4 / NEB3 ratio was selected. EagI (H43A) was labeled EagI-HF.

[0645] 4. Comparison of EagI-HF and WT EagI

[0646] The FI of EagI-HF and WT EagI were determined separately on pXba substrate in each of NEB1-4 buffers. The comparison is shown in Figure 23 and the results are listed in Table 22 (below).

[0647] Table 22: Comparison of EagI-HF and WT EagI

[0648]

[0649] In NEB2 and NEB4, EagI-HF performed best, where the preferred FI was 500; WT EagI performed best in NEB3 and NEB4, where the preferred FI was 250. The overall FI improvement factor was 500 / 250=2.

[0650] Example 17: Engineering High-Fidelity EcoRV

[0651] 1. Expression of EcoRV

[0652] EcoRV was expressed in E. coli transformed with pACYC-EcoRVM and placzz1-EcoRV. Placzz1 is a pUC19 derivative plasmid and pACYC is a low copy compatible plasmid. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0653] 2. Mutagenesis of EcoRV

[0654] Change Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser, Thr, and Trp residues to Ala. Change Tyr to Phe. These are: 2, 4, 5, 6, 9, 12, 13, 14, 15, 16, 17, 18, 19, 21, 25, 27, 29, 31, 35, 36, 37, 38, 41, 42, 44, 45, 47, 48, 49, 53, 54, 57, 58, 59, 61, 64, 65, 67, 68, 69, 70, 71, 72, 74, 75, 76, 78, 79, 81, 82, 84, 85, 86, 90, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 104, 105, 106, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 123, 125, 126, 127, 128, 131, 132, 136, 138, 139 , 140, 143, 144, 145, 146, 147, 149, 150, 151, 152, 154, 155, 157, 158, 161, 163, 164, 167, 169, 171, 172, 173, 174, 179, 183, 185, 186, 187, 188, 191, 193, 195, 196, 197, 198, 199, 2 01, 203, 206, 207, 208, 209, 210, 211, 212, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 226, 227, 228, 229, 230, 231, 232, 234, 235, 236, 237, 238, 239, 241, 242, 244 and 245.

[0655] The upper numbers correspond to the amino acid positions in the EcoRV protein sequence (SEQ ID NO: 84).

[0656] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-EcoRVM).

[0657] 3. EcoRV-HF selection

[0658] Selection of EcoRV-HF was accomplished using a method similar to the previous example. Standard activity checks were performed using pXba in NEB4 using 5% glycerol, and star activity checks were performed against pXba using 39% glycerol in exonuclease I buffer. EcoRV (D19A / E27A) was found to have reduced star activity compared to WT EcoRV. This mutant was labeled EcoRV-HF. For this mutant, E27A was a targeted mutation and D19A was a spontaneous mutation. The double mutant had a greater reduction in star activity than the D19A and E27A single mutants.

[0659] 4. Comparison between EcoRV-HF and WT EcoRV

[0660] The FI of EcoRV-HF and WT EcoRV were determined separately for pXba substrate in each of the NEB1-4 buffers. The comparison is shown in Figure 24 and the results are listed in Table 23 (below).

[0661] Table 23: Comparison of EcoRV-HF and WT EcoRV

[0662]

[0663] EcoRV-HF performed best in NEB2 and NEB4, where the preferred FI was ≥64000, and WT EcoRV performed best in NEB3, where the preferred FI was 1000. The overall FI improvement factor was ≥64000 / 1000=64.

[0664] Example 18: Engineering high-fidelity AvrII

[0665] 1. Expression of AvrII

[0666] AvrII was expressed in E. coli transformed with pUC19-AvrIIRM. Cells were grown overnight at 30°C in LB with Amp.

[0667] 2. Mutagenesis of AvrII

[0668] Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser, Thr, and Trp residues were mutated to Ala. Tyr was changed to Phe. These are: 2, 3, 4, 6, 8, 9, 10, 12, 15, 17, 19, 20, 22, 23, 27, 29, 30, 31, 32, 34, 36, 40, 41, 42, 43, 44, 46, 47, 48, 50, 51, 53, 55, 56, 57, 58, 59, 60, 65, 68, 70, 72, 74, 75, 76, 77, 79, 80, 82, 83, 84, 86, 87, 88, 94, 95, 96, 97, 100, 104, 105, 106, 107, 108 , 110, 112, 113, 116, 117, 119, 120, 121, 122, 123, 124, 126, 127, 129, 130, 131, 132, 134, 136, 139, 142, 143, 144, 145, 150, 151, 152, 153, 154, 156, 157, 158, 161, 163, 164, 165, 166, 168, 169, 173, 174, 177, 178, 181, 182, 184, 186, 187, 188 8, 189, 190, 191, 192, 195, 198, 200, 202, 206, 207, 211, 215, 216, 220, 223, 224, 226, 229, 230, 231, 232, 233, 234, 235, 236, 237, 239, 243, 244, 245, 246, 248, 249, 253, 255, 256, 260, 262, 264, 265, 266, 267, 268, 269, 270, 272, 274, 276, 2 77, 278, 279, 280, 281, 284, 285, 286, 288, 289, 290, 291, 299, 302, 303, 304, 305, 306, 308, 310, 312, 314, 315, 316, 318, 321, 322, 324, 325, 328, 331, 333, 335, 337, 338, 339, 340, 342, 343, 346, 347, 348, 350, 351, 353, 354, 355, 356, 358.

[0669] The upper numbers correspond to the amino acid positions in the AvrII protein sequence (SEQ ID NO: 80).

[0670] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The processed product was then transformed into E. coli strain ER2984.

[0671] 3. Selection of AvrII-HF

[0672] Selection of AvrII-HF was accomplished using methods similar to the previous examples. Cleavage activity was determined using pBC4 in NEB4 with 5% glycerol, and star activity was determined using pBC4 in ExoI buffer with 39% glycerol. Mutants #16 (M29A), #57 (E96A), #60 (Y104F), #62 (K106A), #154 (S127A), #170 (F142A) all showed improvement. AvrII (Y104F) was labeled AvrII-HF.

[0673] 4. Comparison of AvrII-HF and WT AvrII

[0674] The FI of Avrll-HF and WT Avrll were determined separately on T7 DNA substrate in each of NEB 1-4 buffers using diluent B. The comparison is shown in Figure 25 and the results are listed in Table 24 (below).

[0675] Table 24: Comparison of AvrII-HF and WT AvrII

[0676]

[0677] AvrII-HF performed best in NEB1 and NEB4, where the preferred FI was 1000; WT AvrII performed best in NEB1 and NEB4, where the preferred FI was 64. The overall FI improvement factor was 1000 / 64=16.

[0678] Example 19: Engineering High-Fidelity BstXI

[0679] 1. Expression of BstXI

[0680] BstXI was expressed in E. coli transformed with pACYCBstXIMS and pUC19-BstXIR. pACYC is a low copy compatible plasmid. BstXI must have the methylase gene and the specificity gene to have methylase function. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0681] 2. Mutagenesis of BstXI

[0682] As follows, 237 amino acid mutations were performed in BstXI: Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser, Thr, Trp were mutated to Ala, and Try was mutated to Phe. These are: 4, 6, 7, 9, 11, 12, 14, 15, 17, 18, 20, 21, 22, 23, 24, 26, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 42, 43, 46, 48, 50, 53, 54, 57, 58, 59, 60, 62, 63, 64, 65, 66, 71, 72, 73, 75, 76, 78, 80, 81, 82, 83, 84, 86, 89, 91, 93, 94, 95, 96, 97, 98, 103, 105, 106, 108, 110, 111, 112, 14, 117, 118, 120, 123, 124, 125, 126, 127, 128, 129, 130, 131, 137, 138, 139, 141, 142, 144, 145, 146, 148, 151, 152, 153, 154, 155, 156, 159, 162, 163, 166, 168, 169, 171, 172, 173, 174, 175, 176, 177, 178, 182, 185, 188, 189, 191, 193, 194, 195, 196, 198, 199, 201, 204, 205 8, 209, 210, 211, 212, 214, 215, 216, 217, 218, 219, 220, 221, 223, 228, 229, 230, 233, 235, 236, 238, 239, 240, 244, 245, 248, 249, 250, 253, 254, 255, 258, 259, 260, 261, 263, 264, 265, 267, 268, 269, 272, 276, 277, 278, 279, 280, 282, 285, 286, 287, 288, 289, 291, 293, 294 , 295, 300, 301, 302, 304, 305, 306, 308, 309, 312, 314, 317, 318, 319, 320, 323, 324, 325, 326, 330, 331, 333, 334, 335, 337, 343, 344, 345, 346, 347, 349, 353, 355, 356, 357, 358, 359, 360, 362, 363, 364, 365, 367, 369, 371, 373, 374, 376, 377, 378, 379, 380, 381, 382, ​​and 383.

[0683] The upper numbers correspond to the amino acid positions in the BstXI protein sequence (SEQ ID NO: 7).

[0684] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-BstXIMS).

[0685] 3. Selection of BstXI-HF

[0686] Selection of BstXI-HF was accomplished using a method similar to the previous example. Cleavage activity was determined using pBC4 in NEB4 using 5% glycerol, and star activity was determined using pBC4 DNA substrate in NEB4 buffer with 39% glycerol. Mutants #36 (Y57F), #44 (N65A), #48 (E75A), #49 (N76A), and #124 (K199A) all had reduced star activity. BstXI (N65A) was labeled as BstXI-HF.

[0687] 4. Comparison of BstXI-HF and WT BstXI

[0688] The FI of BstXI-HF and WT BstXI were determined separately on the lambda DNA substrate in each of the NEB 1-4 buffers using diluent A. The comparison is shown in Figure 26 and the results are listed in Table 25 (below).

[0689] Table 25: Comparison of BstXI-HF and WT BstXI

[0690]

[0691] In NEB2 and NEB4, BstXI-HF performed best, where the preferred FI was ≥250; WT BstXI performed best in NEB2, NEB3, and NEB4, where the preferred FI was 32. The overall FI improvement factor was ≥250 / 32=8.

[0692] Example 20: Engineering high-fidelity PciI

[0693] 1. Expression of PciI

[0694] PciI was expressed in E. coli transformed with pACYC-PciIM and placzz1-PciIR. placzz1 is a pUC19 derivative plasmid and pACYC is a low copy compatible plasmid. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0695] 2. Mutagenesis of PciI

[0696] 151 amino acid residues in PciI were mutated using Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser, Thr. Trp was changed to Ala, and Tyr was changed to Phe. These are: 2, 3, 4, 6, 8, 9, 10, 11, 12, 14, 17, 18, 19, 21, 24, 25, 26, 28, 29, 30, 31, 33, 34, 35, 36, 38, 39, 41, 44, 46, 47, 49, 50, 51, 54, 55, 56, 58, 59, 60, 63, 67, 68, 69, 71, 74 ,75,78,80,81,82,85,86,91,92,95,97,98,101,103,104,107,109,113,114,115,118,119,120,121,122,124,126,127,129,130,131,132,133,135,136,137,138 8, 143, 145, 146, 147, 148, 149, 151, 152, 153, 154, 155, 157, 158, 159, 161, 164, 165, 167, 172, 175, 178, 179, 180, 182, 184, 185, 186, 190, 192, 193, 196, 197, 198, 1 99, 200, 202, 203, 206, 207, 209, 210, 215, 218, 221, 222, 228, 229, 230, 231, 232, 233, 234, 235, 237, 238, 239, 241, 242, 243, 244, 246, 247, 248, 253, 254, 255, 256.

[0697] The numbers above correspond to the amino acid positions in the PciI protein sequence (SEQ ID NO: 15).

[0698] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-PciIM).

[0699] 3. Selection of PciI-HF

[0700] Using a method similar to the previous example, selection of PciI-HF was completed. Cleavage activity was determined using SalI-cleaved pBR322 in NEB4 using 5% glycerol, and star activity was determined using SalI-cleaved pBR322 in ExoI buffer with 39% glycerol. The double mutant PciI (E78A / S133A) has reduced star activity and strong cleavage activity. This mutant is not one of the targeted mutations described above, but rather a random event.

[0701] 4. Comparison of PciI-HF and WT PciI

[0702] The FI of Pcil-HF and WTPcil were determined separately for pXba substrate in each of NEB1-4 buffers using diluent A. The comparison is shown in Figure 27 and the results are listed in Table 26 (below).

[0703] Table 26: Comparison of PciI-HF and WT PciI

[0704]

[0705] In NEB2, NEB3, and NEB4, PciI-HF performed best, where the preferred FI was ≥2000; WT PciI performed best in NEB3, where the preferred FI was 120. The overall FI improvement factor was ≥2000 / 120=16.

[0706] Example 21: Engineering high fidelity HpaI

[0707] 1. Expression of HpaI

[0708] HpaI was expressed in E. coli transformed with pACYC-MseIM and placzz1-HpaIR. placzz1 is a pUC19 derivative plasmid and pACYC is a low copy compatible plasmid. Cells were grown overnight at 37°C in LB with Amp and Cam.

[0709] 2. Mutagenesis of HpaI

[0710] 156 amino acid residues in HpaI were mutated using Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser and Thr. Trp was changed to Ala, and Tyr was changed to Phe. These are: 7, 8, 9, 13, 14, 16, 17, 19, 20, 21, 22, 23, 26, 27, 29, 30, 33, 34, 35, 36, 37, 38, 40, 41, 42, 46, 47, 48, 50, 51, 56, 57, 59, 60, 65, 67, 69, 71, 72, 74, 75, 78, 79, 80, 81, 82, 83 ,84,85,86,89,91,93,94,95,99,100,104,105,106,108,109,110,113,115,117,119,121,122,123,124,127,128,130,131,133,135,136,137,138,139,141,142,146,1 47, 149, 150, 152, 156, 158, 159, 160, 162, 164, 165, 166, 167, 168, 169, 170, 172, 173, 176, 177, 180, 181, 182, 184, 185, 187, 188, 190, 191, 192, 193, 195, 196, 197, 202, 208 4, 206, 208, 209, 211, 212, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 228, 230, 231, 233, 234, 235, 236, 237, 238, 240, 241, 242, 243, 244, 245, 247, 248, 249.

[0711] The upper numbers correspond to the amino acid positions in the HpaI protein sequence (SEQ ID NO: 86).

[0712] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-MseIM).

[0713] 3. Selection of HpaI-HF

[0714] Selection of HpaI-HF was accomplished using a different approach than in the previous examples. Cleavage and star activity were determined in NEB2 buffer using a lambda DNA substrate. In NEB2, the HpaI had much more star activity than in NEB4 and was clearly observed in 5% glycerol.

[0715] Hpal(Y29F) and Hpal(E56A) are both preferred mutations with reduced star activity. Hpal(E56A) is labeled Hpal-HF.

[0716] 4. Comparison of HpaI-HF and WT HpaI

[0717] The FI of Hpal-HF and WT Hpal were determined separately on lambda DNA substrate in each of NEB 1-4 buffers using diluent A. The comparison is shown in Figure 28 and the results are listed in Table 27 (below).

[0718] Table 27: Comparison of HpaI-HF and WT HpaI

[0719]

[0720] In NEB2, HpaI-HF performed best, where the preferred FI was ≥2000; WT PciI performed best in NEB4, where the preferred FI was 16. The overall FI improvement factor was ≥2000 / 16=120.

[0721] Example 22: Engineering high fidelity AgeI

[0722] 1. Expression of AgeI

[0723] AgeI was expressed in E. coli transformed with pRRS-AgeIRM and psyx20-lacIq. pRRS is a pUC19 derivative plasmid and psyx20-lacIq is a low copy compatible plasmid carrying lacI expressed under the lacIq promoter. Cells were grown to 200 Klett units at 37°C in LB with Amp and Kan and then induced overnight with 0.5 mM IPTG at 25°C. Expression of AgeI is extremely difficult to achieve because it is unstable.

[0724] 2. Mutagenesis of AgeI

[0725] 149 amino acid residues in AgeI were mutated using Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser and Thr. Trp was changed to Ala, and Tyr was changed to Phe. These are: 2, 4, 6, 7, 9, 14, 16, 18, 19, 21, 22, 23, 24, 25, 26, 27, 29, 30, 31, 32, 37, 38, 40, 42, 43, 44, 45, 49, 51, 53, 55, 56, 58, 60, 62, 64, 65, 67, 68, 69, 72, 73, 75, 77, 78 ,79,82,83,85,86,87,88,90,91,92,94,96,97,102,103,104,105,110,111,114,116,119,120,122,123,128,129,130,134,135,138,139,140,142,144,146,1 47, 148, 152, 153, 155, 157, 159, 166, 168, 170, 173, 174, 176, 177, 178, 182, 183, 185, 186, 188, 192, 195, 198, 200, 201, 206, 211, 212, 214, 217, 219, 220, 222, 223 3, 224, 225, 226, 227, 229, 231, 233, 234, 235, 237, 238, 239, 240, 241, 243, 245, 247, 248, 250, 251, 253, 255, 256, 258, 260, 262, 265, 266, 267, 268, 269, 271, 272.

[0726] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (psyx20-lacIq).

[0727] The upper numbers correspond to the amino acid positions in the AgeI protein sequence (SEQ ID NO: 79).

[0728] 3. Selection of AgeI-HF

[0729] Selections for AgeI-HF were completed using methods similar to the previous examples. Standard activity checks were performed with pXba in NEB4 using 5% glycerol, and star activity checks were performed on pXba in NEB4 buffer with 39% glycerol. Because of the difficulty of the expression system, the selection was repeated eight times before meaningful mutants were obtained. Two mutants, S201A and R139A, had reduced star activity, and R139A was labeled AgeI-HF.

[0730] 4. Comparison of AgeI-HF and WT AgeI

[0731] The FI of AgeI-HF and WTAgel were determined separately for pXba substrate in each of NEB1-4 buffers using diluent A. The comparison is shown in Figure 29 and the results are listed in Table 28 (below).

[0732] Table 28: Comparison of AgeI-HF and WT AgeI

[0733]

[0734] AgeI-HF performed best in NEB1 and NEB4, where the preferred FI was ≥500; WT AgeI performed best in NEB3, where the preferred FI was 16. The overall FI improvement factor was ≥500 / 16=32.

[0735] Example 23: Engineering high-fidelity BsmBI

[0736] 1. Expression of BsmBI

[0737] BsmBI was expressed in E. coli transformed with pACYC-BsmAIM, ptaczz2-BsmBIR and psyx20-lacIq. Ptaczz2 is a pUC19 derived plasmid carrying the inducible ptac promoter and pACYC is a low copy compatible plasmid. BsmAIM (GTCTC) covers BsmBI (CGTCTC) specificity. psyx20-lacIq is a low copy vector with strong lacI expression. Cells were grown at 37°C and then induced in LB with Amp, Cam and Kan.

[0738] 2. Mutagenesis of BsmBI

[0739] 358 amino acid residues in BsmBI were mutated using Cys, Asp, Glu, Phe, His, Lys, Met, Asn, Gln, Arg, Ser and Thr. Trp was changed to Ala, and Tyr was changed to Phe. These are: 8, 9, 12, 13, 14, 15, 17, 18, 19, 20, 21, 24, 25, 26, 27, 28, 29, 31, 33, 37, 38, 40, 42, 43, 44, 46, 47, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 60, 61, 62, 63, 64, 65, 66, 67, 69, 70, 72, 76, 78, 79, 80, 81, 82, 83, 84, 88, 91, 93, 95, 96, 98, 99, 101, 103, 104, 105, 106, 109, 110, 111, 113, 114、115、117、118、119、120、121、122、124、126、127、128、130、131、132、133、134、135、138、141、143、144、145、147、149、150、154、155、157、158、160、162、163、164、165、166、167、168、169、172、174、175、176、177、179、180、181、182、184、185、186、188、189、191、194、195、197、 200、201、203、205、206、207、208、211、212、213、214、215、216、217、220、221、222、223、224、225、226、228、229、230、231、232、233、235、236、237、238、239、240、242、243、247、250、251、252、257、258、260、262、263、264、265、266、268、269、271、273、274、279、280、282、283、284、 287、288、289、292、294、295、296、297、299、300、301、302、303、304、305、306、307、309、310、313、314、315、316、317、318、320、321、324、325、326、328、331、332、333、334、335、336、339、340、341、342、343、344、345、348、349、351、352、353、355、356、357、358、360、361、363、364、366, 367, 368, 370, 372, 375, 376, 377, 379, 381, 382, ​​384, 385, 386, 388, 389, 390, 393, 394, 395, 396, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 418, 421, 424, 425, 426, 428, 430, 431, 432, 433, 434, 436, 437, 438, 439, 442, 443, 444, 445, 446 6、446、448、449、450、451、452、453、454、457、458、460、461、462、464、466、467、468、470、471、472、473、474、477、478、479、480、482、483、484、485、486、487、488、489、491、492、495、496、497、498、499、500、502、503、504、505、506、507、510、511、515、516、517、518、519、522、523.

[0740] The upper numbers correspond to the amino acid positions in the BsmBI protein sequence (SEQ ID NO: 81).

[0741] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-BsmAIM, psyx20-lacIq).

[0742] 3. Selection of BsmBI-HF

[0743] Selection for BsmBI-HF was accomplished using a method similar to the previous example. Cleavage was determined using lambda DNA in NEB4 using 5% glycerol, and star activity was determined using Litmus28i in NEB4 buffer with 39% glycerol. Preferred mutants include H230A, D231A, and N185Y / R232A. N185Y / R232A is labeled BsmBI-HF.

[0744] 4. Comparison of BsmBI-HF and WT BsmBI

[0745] The FI of BsmBI-HF and WT BsmBI were determined separately on the lambda DNA substrate in each of the NEB 1-4 buffers using diluent A. The comparison is shown in Figure 30 and the results are listed in Table 29 (below).

[0746] Table 29: Comparison of BsmBI-HF and WT BsmBI

[0747]

[0748] BsmBI-HF performed best in NEB1, NEB2, and NEB4, where the preferred FI was ≥500; WT BsmBI performed best in NEB3, where the preferred FI was 120. The overall FI improvement factor was ≥500 / 120=4.

[0749] Example 24: Engineering BspQI variants with reduced star activity

[0750] 1. Site-directed mutagenesis of BspQI restriction endonuclease

[0751] Use Km R and Amp R E. coli was transformed with pSX33-EarIM1M2 and pZZlacI-PspQ. M. EarI (CTCTTC) also modifies the BspQI site (GCTCTTC), and thus pSX33-earIM1M2 (Figures 17 and 18) was used to co-transform and modify the E. coli chromosome. The WT amino acid sequence is shown in Figure 16.

[0752] The 122 charged or uncharged amino acid residues (Arg, Lys, His, Glu, Asp, Gln, Asn, Cys) in BspQI were changed to Ala by site-directed mutagenesis. PCR was performed under the following conditions: DNA denaturation, 98°C for 30 sec, 1 cycle; DNA denaturation / primer annealing / extension, 98°C for 10 sec, 55°C to 65°C for 30 sec, 72°C for 2 min, 18 PCR cycles; 72°C for 15 min, 1 cycle. In 100 μl reaction, 2 units of Phusion TM DNA polymerase (NEB, Ipswich, MA), 1 mM dNTPs, 10 ng to 100 ng template DNA, 20 μl 5x reaction buffer, 0.04 μM primers, sterile water to 100 μl total volume.

[0753] The PCR DNA was digested with DpnI to destroy the template DNA (Dam methylated) and co-transformed with pSX33-earIM1M2 into E. coli. Each transformant was cultured overnight (5 ml LB, 50 μg / ml Km R and 100 μg / ml Amp R) and divided into two portions. One portion (1.5 ml) was harvested by centrifugation and lysed by sonication in sonication buffer (20 mM Tris-HCl, pH 7.5, 0.1 mM DTT, 50 mM NaCl, 10% glycerol). The cell extract was heated at 50°C for 1 hour and denatured E. coli proteins were removed by centrifugation. The clarified lysate was analyzed for restriction activity and star activity on pUC19 DNA.

[0754] BspQI star activity assay conditions: 1 μg pUC19 DNA, 5 μl 10x NEB buffer 1, 25% DMSO, 2.5 μl clarified cell extract, sterile deionized water to 50 μl total volume, and incubate for 1 hour at 50° C. The digested DNA was resolved by electrophoresis in 0.8 to 1% agarose gel.

[0755] A second cell culture (uninduced) was harvested and plasmid DNA was prepared by the Qiagen spin column purification method (Qiagen, Valencia, CA). The bspQIR allele was sequenced by the Big-dye dideoxy-terminator sequencing method to confirm the desired mutation. After identification of reduced star activity mutants, new transformants were obtained and IPTG-induced cultures were made. Restriction activity and star activity were analyzed again to confirm reduced star activity compared to the WT enzyme in all four buffers.

[0756] Among the 122 BspQI mutants constructed by site-directed mutagenesis, two BspQI variants, R388A and K279A, showed reduced star activity. In buffer 1 and 10% glycerol, the star activity of R388A was reduced by approximately 16-fold. However, R388A still showed star activity at high enzyme concentrations. BspQI variant K279A also showed reduced star activity (>8-fold improvement in reduced star activity).

[0757] To further reduce star activity at high enzyme concentrations, R388 and K279 were substituted with Phe, Pro, Tyr, Glu, Asp or Leu. IPTG-induced cell extracts of various R388X and K279X mutants were analyzed for restriction activity and star activity. It was found that R388F or K279P showed minimal star activity in cell extracts or purified enzyme. The amino acid substitutions did not affect specific activity.

[0758] To further reduce BspQI star activity, two amino acid substitutions were combined into one mutant enzyme (double mutant, K279P / R388F) by site-directed mutagenesis. This double mutant lacked star activity in both buffer 1 and buffer 2 with 10% glycerol ( Fig.41B ).

[0759] Table 30: BspQI-HF vs. WT BspQI

[0760]

[0761] Example 25: Engineering of SapI variants with reduced star activity

[0762] The conserved amino acid residues K279 and R388 were found in BspQI, where the corresponding positions in SapI are K273 and R380. A 6x His-tagged SapI expression clone was first constructed in pUC19. The SapI expression strain was constructed using Km R and Amp R E. coli transformed with pSX33-earIM1M2 and pUC-SapI. The amino acid substitutions of Lys273 to Pro (K273P) and Arg380 to Phe (R380F) were introduced into SapI by site-directed mutagenesis. A SapI single mutant R380A was also constructed. When subjected to restriction activity and star activity reactions, both SapI variants R380A and K273P / R380F showed reduced star activity ( Fig.42 ).

[0763] PCR, transformation, plasmid DNA preparation and enzyme activity analysis were performed as described for BspQI, except that Sapl activity was assayed at 37° C. The 6× His-tagged Sapl variant K273P / R380F was purified by Ni-NTA column chromatography and shown to display reduced star activity in the presence of 25% DMSO or 5% glycerol.

[0764] Example 26: Engineering KpnI high-fidelity mutants

[0765] KpnI, which contains two activities, has been altered to a mutant with lower star activity (International Publication No. WO 07 / 027464). The following examples describe novel mutants with improved star activity and cleavage activity similar to the wild type.

[0766] Charged amino acid residues (Asp, Glu, Arg, Lys, and His) or polar amino acids (Ser, Thr, Tyr, Asn, Gln, Phe, Trp, Cys, and Met) except the catalytic residues were individually mutated to alanine.

[0767] Mutagenesis was performed by inverse PCR using primers with the desired mutation. In general, inverse PCR was performed using 0.4 mM of each 4 dNTP, 1 x ThermoPol buffer (NEB), 20 ng of template DNA, 40 μmol of each primer and 4 U of Vent DNA polymerase (NEB) in a final volume of 100 μL.

[0768] Plasmids pUC19 or pAGR3 containing KpnIR under the control of Plac or Ptac promoter, respectively, were used as templates. PCR was performed using the following temperature protocol: 94°C for 4 minutes, followed by 25 cycles of 94°C for 30 seconds for denaturation, 55°C for 30 seconds for annealing, and 72°C for 5 minutes for extension. After cycling, incubate at 72°C for 7 minutes, then treat the reaction with 20U of DpnI (NEB) at 37°C for 1 hour to degrade the template DNA. After inactivating DpnI at 80°C for 20 minutes, 2 μl of the reaction was used to transform 50 μL of chemically competent NEB5alpha (NEB) pre-transformed with pSYX20-KpnIM. The transformed bacteria were plated on LB plates containing 100 μg / ml ampicillin and 30 μg / ml kanamycin and incubated at 37°C for 12 to 15 hours. 3 to 4 colonies of every kind of construction were in 1ml of LB containing 100 μg / ml ampicillin and 30 μg / ml kanamycin, and were cultured for 12 to 15 hours at 37°C with shaking at 200rpm. The culture was centrifuged (spin down), and resuspended in 0.2ml of ultrasonic treatment buffer (20mM Tris-HCl, pH 8.3, 50mM NaCl, 1mM EDTA, 1mM PMSF). The resuspended cells were ultrasonically treated for 20 seconds, then centrifuged at 4°C for 5 minutes at 13,000rpm. The supernatant was diluted, and 5 μL of KpnI cleavage activity was analyzed.

[0769] For screening mutants, activity assay reactions were performed using 5 μL of 10- or 100-fold diluted lysate supernatant, 0.5 μg of pXba DNA (NEB) and 1× NEBuffer 4 in a total volume of 50 μL. After incubation for 1 hour at 37°C, the assay reactions were stopped by adding 10 μl of 6× DNA loading dye and analyzed by 0.8% agarose gel electrophoresis in 1×TBE. The assay showed reduced star activity in a buffer containing 25% DMSO for mutants with increased total cleavage activity compared to the parental enzyme (KpnI D148E).

[0770] The assay was performed using 5 μL of the enzyme dilution incubated with 1 μg of pXba DNA (NEB) in the presence of 5% glycerol and 0.2 mg / ml BSA in 1x NEBuffer 4 (total volume = 50 μL). After incubation at 37°C for 1 hour, the reaction was treated with 20 μg of protein kinase K (NEB) for 15 minutes at 37°C and then analyzed by 0.8% agarose gel electrophoresis in 1x TBE. 50 U of enzyme and 1 μg of pXba DNA were used in an ELISA containing 20 mM Tris HCl, 50 mM NaCl, pH 7.9, 1 mM DTT and increasing concentrations of MgSO 4 / CaCl 2 / MnCl 2 The divalent metal dependence analysis was performed in 50 μL buffer at 37°C for 1 hour followed by electrophoresis on a 0.8% agarose gel in 1xTBE.

[0771] The result of random mutagenesis is the preferred mutant KpnI D148E, which shows a lower star activity than the wild-type enzyme and the previously described KpnI D163I / K165A mutant (International Publication No. WO 07 / 027464). However, KpnI D148E shows star activity at high enzyme concentrations. In the D148E background, double or triple mutants are constructed to reduce the observed star activity. It is found that KpnI (D16N / E132A / D148E) has a lower star activity and a higher specific activity than mutant D148E. By site-directed mutagenesis, amino acid substitutions D148 to E and E132 to A are introduced. The mutation of D16 to N is introduced by PCR. In standard reaction conditions, the purified enzyme is incubated with substrate DNA pXba (containing 6 KpnI sites) at 37°C for 1 hour to produce the reduced star activity shown in Table 31.

[0772] Table 31: Fidelity index of KpnI mutants

[0773]

[0774] No star activity was observed for pXba for mutant D16N / E132A / D148E up to 4000 U. The star activity observed for the pBR322 substrate, which does not contain a KpnI site, was also reduced when cut with KpnI D148E and D16N / E132A / D148E.

[0775] Example 27: Engineering high-fidelity BsaI

[0776] 1. Expression of BsaI

[0777] BsaI was expressed in E. coli transformed with pACYC-BsmAIM, pUC 19-BsaIR and psyx20-lacIq. pACYC is a low copy compatible plasmid. BsmAIM (GTCTC) masks BsmBI specificity (CGTCTC). psyx20-lacIq is a low copy vector with strong lacI expression. Cells were grown at 37°C and then induced in LB with Amp, Cam and Kan.

[0778] 2. Mutagenesis of BsaI

[0779] The amino acid of BsaI is similar to that of BsmBI. Eleven amino acids around and at the corresponding aforementioned effective sites are mutated to R229A, S230A, Y231F, T232A, T233A, D234A, R235A, R236A, F238A, E239A, and Y240F.

[0780] The method was the same as in the previous example, using inverse PCR followed by DpnI digestion. The treated product was then transformed into E. coli (pACYC-BsmAIM, psyx20-lacIq).

[0781] 3. Selection of BsaI-HF

[0782] Selection for BsaI-HF was accomplished using a method similar to that of the previous example. Standard activity assays were performed using λDNA in NEB4 using 5% glycerol, and star activity assays were performed on litmus28i in NEB4 buffer with 39% glycerol. One mutant, Y231F, among the 11 designed mutants reduced star activity and was labeled as BsaI-HF.

[0783] 4. Comparison of BsaI-HF and WT BsaI

[0784] The FI of BsaI-HF and WT BsaI were determined separately on lambda DNA in NEB 1-4 buffer using diluent A. The results are listed in Table 32 (below).

[0785] Table 32: Comparison of BsaI-HF and WT BsaI

[0786]

[0787] In NEB2, NEB3, and NEB4, BsaI-HF performed best, with the best FI being ≥8000; WT BsaI performed best in NEB2 and NEB4, with the best FI being 120. Therefore, the overall FI improvement factor was ≥8000 / 120=≥64.

Claims

1. A composition comprising a mutant restriction endonuclease AgeI restriction endonuclease, wherein the mutant restriction endonuclease consists of SEQ ID No. 79 and has an artificially introduced mutation corresponding to R139A or S201A in SEQ ID No.

79.

2. A DNA molecule encoding the restriction endonuclease according to claim 1. A vector comprising the DNA molecule according to claim 2.

4. A host cell comprising the DNA molecule of claim 2 or the vector of claim 3 for expressing the restriction endonuclease.

Images