(12) INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT)
`
`(19) World Intellectual Property Organization
`International Bureau
`
`(43) International Publication Date
`13 December 2001 (13.12.2001)
`
`
`
`PCT
`
`(10) International Publication Number
`wo 01/94544 A2
`
`(51) International Patent Classification7:
`
`C12N
`
`(21) International Application Number:
`
`PCT/USOI/17804
`
`(22) International Filing Date:
`
`1 June 2001 (01.06.2001)
`
`Rebecca, B. [US/US]; 160 Woodland Meadow, Hamilton,
`MA 01682 (US). SCHILDKRAUT, Ira [US/US]; Post
`
`Office BOX 392, Cem'llose NM 87010-0392 (US)~ WIL-
`SON, Geoffrey, G. [GB/US]; 17 Partridge Lane, Boxford,
`MA 01921 (US).
`
`(25) Filing Language:
`
`(26) Publication Language:
`
`English
`
`English
`
`(74) Agent: WILLIAMS, Gregory, D.; General Counsel, New
`England Biolabs, Inc., 32 Tozer Road, Beverly, MA 01915
`(US).
`
`(30) Priority Data:
`09/586,935
`
`2 June 2000 (02.06.2000)
`
`US
`
`(71) Applicant (for all designated States except US): NEW
`ENGLAND BIOLABS, INC. [US/US]; 32 Tozer Road,
`Beverly, MA 01915 (US).
`
`(72) Inventors; and
`(7S) Inventors/Applicants U’or US only): KONG, Huimin
`[CN/US]; 5 Conrad Circle, Wenham, MA 01984 (US).
`HIGGINS, Lauren, Sears [US/US]; 67 Western Avenue,
`Essex, MA 01929 (US). DALTON, Michael, A. [US/US];
`38 Forest Street, Manchester, MA 01944 (US). KUCERA,
`
`(81) Designated States (national): JP, US.
`
`(84) Designated States (regional): European patent (AT, BE,
`CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC,
`NL, PT, SE, TR).
`
`Published:
`
`without international search report and to be republished
`upon receipt of that report
`
`For two-letter codes and other abbreviations, refer to the ”Guid-
`ance Notes on Codes andAbbreviations ” appearing at the begin-
`ning ofeach regular issue ofthe PCT Gazette.
`
`01/94544A2
`
`(54) Title: CLONING AND PRODUCING THE N.BstNBI NICKING ENDONUCLEASE AND RELATED METHODS FOR US—
`ING NICKING ENDONUCLEASES IN SINGLE—STRANDED DISPLACEMENT AMPLIFICATION
`
`(57) Abstract: The present invention relates to recombinant DNA which encodes a novel nicking endonuclease, N.BstNBI, and the
`O production of N.BstNBI restriction endonuclease from the recombinant DNA utilizing PleI modification methylase. Related expres—
`sion vectors, as well as the application of N.BstNBI and other nicking enzymes in non—modified strand displacement amplification,
`B
`is disclosed also.
`
`ELIX 1006
`
`1
`
`ELIX 1006
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`CLONING AND PRODUCING THE N.BstNBI NICKING ENDONUCLEASE
`AND RELATED METHODS FOR USING NICKING ENDONUCLEASES IN
`SINGLE-STRANDED DISPLACEMENT AMPLIFICATION
`
`BACKGROUND OF THE INVENTION
`
`The present invention relates to the recombinant
`
`DNA which encodes the N.BstNBI nicking endonuclease and
`
`modification methylase, and the production of N.BstNBI
`
`nicking endonuclease from the recombinant DNA. N.BstNBI
`
`nicking endonuclease is originally isolated from
`
`Bacillus stearothermophilus. It recognizes a simple
`
`asymmetric sequence, 5’ GAGTC 3’, and it cleaves only
`
`one DNA strand,
`
`4 bases away from the 3’-end of its
`
`recognition site.
`
`The present invention also relates to the use of
`
`nicking endonucleases in strand—displacement
`
`amplification application (SDA). More particularly, it
`
`relates to liberating such amplification from the
`
`technical limitation of employing modified (particularly
`
`a—thiophosphate substituted) nucleotides.
`
`Restriction endonucleases are enzymes that
`
`recognize and cleave specific DNA sequences. Usually
`
`there is a corresponding DNA methyltransferase that
`
`methylates and therefore protects the endogenous host
`
`DNA from the digestion of a certain restriction
`
`endonuclease. Restriction endonucleases can be
`
`classified into three groups:
`
`type I, II, and III. More
`
`than 3000 restriction endonucleases with over two
`
`hundred different specificities have been isolated from
`
`bacteria (Roberts and Macelis, NUcleic Acids Res.
`
`26:338—350 (1998)). Type II and type IIs restriction
`
`enzymes cleave DNA at a Specific position, and therefore
`
`are useful in genetic engineering and molecular cloning.
`
`2
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`Most restriction endonucleases catalyze double—
`
`stranded cleavage of DNA substrates via hydrolysis of
`
`two phosphodiester bonds on two DNA strands (Heitman,
`
`Genetic Engineering 15:57—107 (1993)). For example,
`
`type
`
`II enzymes, such as EboRI and EcoRV, recognize
`
`palindromic sequences and cleave both strands
`
`symmetrically within the recognition sequence. Type IIS
`
`endonucleases recognize asymmetric DNA sequences and
`
`cleave both DNA strands outside of the recognition
`
`sequence.
`
`There are some proteins in the literature which
`
`break only one DNA strand and therefore introduce a nick
`
`into the DNA molecule. Most of those proteins are
`
`involved in DNA replication, DNA repair, and other DNA—
`
`related metabolisms (Kornberg and Baker, DNA
`
`replication. 2nd edit. W.H. Freeman and Company, New
`
`York,
`
`(1992)). For example, gpII protein of
`
`bacteriophage fI recognizes and binds a very complicated
`
`sequence at the replication origin. It introduces a nick
`
`in the plus strand, which initiates rolling circle
`
`replication, and it is also involved in circularizing
`
`the plus strand to generate single—stranded circular
`
`phage DNA.
`
`(Geider et al., J. Biol. Chem. 257:6488—6493
`
`(1982); Higashitani et al.,
`
`J1 Mbl. Biol. 237:388—400
`
`(1994)). Another example is the MutH protein, which is
`
`involved in DNA mismatch repair in E. coli. NutH binds
`
`at dam methylation sites (GATC), where it forms a
`
`protein complex with nearby MutS which binds to a
`
`mismatch. The MUtL protein facilitates this interaction
`
`and this triggers single—stranded cleavage by MhtH at
`
`the 5’ end of the unmethylated GATC site. The nick is
`
`then translated by an exonuclease to remove the
`
`3
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`mismatched nucleotide (Modrich,
`
`J1 Biol. Chem. 264:6597—
`
`6600 (1989)).
`
`The nicking enzymes mentioned above are not very
`
`useful in the laboratory for manipulating DNA due to the
`
`fact that they usually recognize long, complicated
`
`sequences and usually associate with other proteins to
`
`form protein complexes which are difficult to
`
`manufacture. Thus none of these nicking proteins are
`
`commercially available. Recently, we have found a
`
`nicking protein, N.BstNBI,
`
`from the thermophilic
`
`bacterium Bacillus stearothermophilus, which is an
`
`isoschizomer of N.BstSEI
`
`(Abdurashitov et al., Mbl.
`
`Biol.
`
`(Mosk) 30:1261—1267 (1996)). Unlike gpII and NutH,
`
`N.BstNBI behaves like a restriction endonuclease. It
`
`recognizes a simple asymmetric sequence, 5’ GAGTC 3’,
`
`and it cleaves only one DNA strand,
`
`4 bases away from
`
`the 3’—end of its recognition site (Fig. 1A).
`
`Because N.BstNBI acts more like a restriction
`
`endonuclease, it should be useful in DNA engineering.
`
`For example, it can be used to generate a DNA substrate
`
`containing a nick at a specific position. N.BstNBI can
`
`also be used to generate DNA with gaps,
`
`long overhangs,
`
`or other structures. DNA templates containing a nick or
`
`gap are useful substrates for researchers in studying
`
`DNA replication, DNA repair and other DNA related
`
`subjects (Kornberg and Baker, DNA replication. 2nd edit.
`
`W.H. Freeman and Company, New York,
`
`(1992)). A potential
`
`application of the nicking endonuclease is its use in
`
`strand displacement amplification (SDA), which is an
`
`isothermal DNA amplification technology. SDA provides an
`
`alternative to polymerase chain reaction (PCR), and it
`
`can reach lOG—fold amplification in 30 minutes without
`
`thermo—cycling (Walker et al., Proc. Natl. Acad. Sci.
`
`4
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`USA 89:392—396 (1992)). SDA uses a restriction enzyme to
`
`nick the DNA and a DNA polymerase to extend the 3'—OH
`
`end of the nick and displace the downstream DNA strand
`
`(walker et al.,
`
`(1992)). The SDA assay provides a simple
`
`(no temperature cycling, only incubation at 60°C) and
`
`very rapid (as short as 15 minutes) detection method and
`
`can be used to detect viral or bacterial DNA. SDA is
`
`being introduced as a diagnostic method to detect
`
`infectious agents, such as Mycobacterium tuberculosis
`
`and Chlamydia trachomatis (Walker and Linn, Clin. Chem.
`
`42:1604—1608 (1996); Spears et al., Anal. Biochem.
`
`247:130—137 (1997)).
`
`For SDA to work, a nick has to be introduced into
`
`the DNA template by a restriction enzyme. Most
`
`restriction endonucleases make double—stranded
`
`cleavages. Therefore, modified a—thio deoxynucleotides
`
`(dNTPaS) have to be incorporated into the DNA,
`
`so that
`
`the endonuclease only cleaves the unmodified strand
`
`which is within the primer region (Walker et al., 1992).
`
`The a—thio deoxynucleotides are eight times more
`
`expensive than regular dNTPs
`
`(Pharmacia), and are not
`
`incorporated well by the Bst DNA polymerase as compared
`
`to regular deoxynucleotides (J. Aliotta, L. Higgins, and
`
`H. Kong, unpublished observation).
`
`Alternatively,
`
`in accordance with the present
`
`invention, it has been found that if a nicking
`
`endonuclease is used in SDA, it will introduce a nick
`
`into the DNA template naturally. Thus the dNTPaS is no
`
`longer needed for the SDA reaction when a nicking
`
`endonuclease is being used. This idea has been tested,
`
`and the result agreed with our speculation. The target
`
`DNA can, for example, be amplified in the presence of
`
`the nicking endonuclease N.BstNBI, dNTPs, and Bst DNA
`
`5
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`polymerase. Other nicking endonucleases can also be
`
`used.
`
`It is even possible to employ a restriction
`
`endonuclease in which the two strands are cleaved
`
`sequentially, such that nicked intermediates accumulate.
`
`With the advent of genetic engineering technology,
`
`it is now possible to clone genes and to produce the
`
`proteins that they encode in greater quantities than are
`
`obtainable by conventional purification techniques. Type
`
`II restriction—modification systems are being cloned
`
`with increasing frequency. The first cloned systems used
`
`bacteriophage infection as a means of identifying or
`
`selecting restriction endonuclease clones (EcoRII:
`
`Kosykh et al., Mblec. Gen. Genet 178:717—719 (1980);
`
`HhaII: Mann et al., Gene 3:97—112 (1978); PstI: walder
`
`et al., Proc. Nat. Acad. Sci. 78:1503—1507 (1981)).
`
`Since the presence of restrictionemodification systems
`
`in bacteria enable them to resist infection by
`
`bacteriophages, cells that carry cloned restriction—
`
`modification genes can,
`
`in principle, be selectively
`
`isolated as survivors from libraries that have been
`
`exposed to phage. This method has been found, however,
`
`to have only limited value. Specifically, it has been
`
`found that cloned restriction—modification genes do not
`
`always manifest sufficient phage resistance to confer
`
`selective survival.
`
`Another cloning approach involves transferring
`
`systems initially characterized as plasmid—borne into E.
`
`coli cloning plasmids (EcoRV: Bougueleret et al., NUCl.
`
`Acids Res. 12:3659—3676 (1984); PaeR7: Gingeras and
`
`Brooks, Proc. Natl. Acad. Sci. USA 80:402—406 (1983);
`
`Theriault and Roy, Gene 19:355—359 (1982); PvuII:
`
`Blumenthal et al.,
`
`J1 Bacteriol. 164:501—509 (1985)).
`
`6
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`A further approach which is being used to clone a
`
`growing number of systems involves selection for an
`
`active methylase gene (refer to U.S. Patent No.
`
`5,200,333 and BsuRI: Kiss et al., NUcl. Acids Res.
`
`13:6403—6421 (1985)). Since restriction and modification
`
`genes are often closely linked, both genes can often be
`
`cloned simultaneously. This selection does not always
`
`yield a complete restriction system however, but instead
`
`yields only the methylase gene (BspRI: Szomolanyi et
`
`al., Gene 10:219-225 (1980); BcnI: Janulaitis et a1,
`
`Gene 20:197—204 (1982); BsuRI: Kiss and Baldauf, Gene
`
`21:111—119 (1983); and MSpI: Walder et al., J. Biol.
`
`Chem. 258:1235—1241 (1983)).
`
`Another method for cloning methylase and
`
`endonuclease genes is based on a colorimetric assay for
`
`DNA damage (see U.S. Patent No. 5,492,823). When
`
`screening for a methylase,
`
`the plasmid library is
`
`transformed into the host E. coli strain such as AP1—
`
`200. The expression of a methylase will induce the SOS
`
`response in an E. coli strain which is McrA+, MchC+, or
`
`Mrr+. The AP1—200 strain is temperature sensitive for
`
`the Mcr and Mrr systems and includes a lac—Z gene fused
`
`to the damage inducible dinD locus of E. coli. The
`
`detection of recombinant plasmids encoding a methylase
`
`or endonuclease gene is based on induction at the
`
`restrictive temperature of the lacZ gene. Transformants
`
`encoding methylase genes are detected on LB agar plates
`
`containing X—gal as blue colonies.
`
`(Piekarowicz et al.,
`
`NUCleic Acids Res. 19:1831—1835 (1991) and Piekarowicz
`
`et al.,
`
`J1 Bacteriology 173:150—155 (1991)). Likewise,
`
`the E. coli strain ER1992 contains a dinDl—Lacz fusion
`
`but is lacking the methylation dependent restriction
`
`systems McrA, MchC and Mrr. In this system (called the
`
`"endo-blue" method),
`
`the endonuclease gene can be
`
`7
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`detected in the absence of its cognate methylase when
`
`the endonuclease damages the host cell DNA,
`
`inducing the
`
`SOS response. The SOS—induced cells form deep blue
`
`colonies on LB agar plates supplemented with X—gal.
`
`(Fomenkov et al., NUcleic Acids Res. 22:2399—2403
`
`(1994)).
`
`Sometimes the straight—forward methylase selection
`
`method fails to yield a methylase (and/or endonuclease)
`
`clone due to various obstacles (see, e.g., Lunnen et
`
`al., Gene 74(1):25—32 (1988)). One potential obstacle to
`cloning restriction—modification genes lies in trying to
`
`introduce the endonuclease gene into a host not already
`
`protected by modification. If the methylase gene and
`
`endonuclease gene are introduced together as a single
`
`clone,
`
`the methylase must protectively modify the host
`
`DNA before the endonuclease has the opportunity to
`
`cleave it. On occasion,
`
`therefore, it might only be
`
`possible to clone the genes sequentially, methylase
`
`first then endonuclease (see U.S. Patent No. 5,320,957).
`
`Another obstacle to cloning restriction—
`
`modification systems lies in the discovery that some
`
`strains of E. coli react adversely to cytosine or
`
`adenine modification;
`
`they possess systems that destroy
`
`DNA containing methylated cytosine (Raleigh and Wilson,
`
`Proc. Natl. Acad. Sci. USA 83:9070—9074 (1986)) or
`
`methylated adenine (Heitman and Model,
`
`J1 Bacteriology
`
`196:3243—3250 (1987); Raleigh et al., Genetics 122:279—
`
`296 (1989); Waite—Rees et al., J. Bacteriology 173:5207—
`
`5219 (1991)). Cytosine—specific or adenine—specific
`
`methylase genes cannot be cloned easily into these
`
`strains, either on their own, or together with their
`
`corresponding endonuclease genes. To avoid this problem
`
`it is necessary to use mutant strains of E. coli
`
`(McrA—
`
`8
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`and Mch— and Mrr—)
`
`in which these systems are
`
`defective.
`
`An additional potential difficulty is that some
`
`restriction endonuclease and methylase genes may not
`
`express in E. coli due to differences in the
`
`transcription machinery of the source organism and E.
`
`coli, such as differences in promoter and ribosome
`
`binding sites. The methylase selection technique
`
`requires that the methylase express well enough in E.
`
`coli to fully protect at least some of the plasmids
`
`carrying the gene.
`
`Because purified restriction endonucleases, and to
`
`a lesser extent modification methylases, are useful
`
`tools for characterizing genes in the laboratory,
`
`there
`
`is a commercial incentive to obtain bacterial strains
`
`through recombinant DNA techniques that synthesize these
`
`enzymes in abundance. Such strains would be useful
`
`because they would simplify the task of purification as
`
`well as provide the means for production in commercially
`
`useful amounts.
`
`SUMMARY OF THE INVENTION
`
`A unique combination of methods was used to
`
`directly clone the N.BstNBI endonuclease gene and
`
`express the gene in an E. coli strain premodified by
`
`PleI methylase. To clone the N.BstNBI endonuclease gene
`
`directly, both the N—terminal amino acid sequence and a
`
`stretch of internal amino acid sequence of highly
`
`purified native N.BstNBI restriction endonuclease were
`
`determined. Degenerate primers were designed based on
`
`the amino acid sequences, and PCR techniques were used
`
`to amplify a segment of the DNA gene that encodes the
`
`9
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`N.BstNBI endonuclease protein. The PCR product was
`
`sequenced, and the information was used to design
`
`primers for inverse PCR reactions. By chromosome walking
`
`via inverse PCR,
`
`the endonuclease open reading fram ,
`
`n.bstNBIR, was deduced. Continuing with inverse PCR, an
`
`open reading frame was found adjacent to the
`
`endonuclease gene. Blast analysis suggested that this
`
`gene encoded an adenine methylase (n.bstNBIM).
`
`The N.BstNBI endonuclease gene was cloned into a
`
`low copy—number T7 expression vector, pHKT7, and
`
`transformed into an E. coli host which had been
`
`premodified by a pHKUVS—PleI methylase clone. This
`
`recombinant E. coli strain (NEB#1239) produces about 4 X
`
`107 units N.BstNBI endonuclease per gram cell.
`
`The present invention also relates to a novel
`
`method of DNA amplification. The method of using nicking
`
`endonuclease such as N.BstNBI in the absence of modified
`
`nucleotides such as a—thio dNTPs in strand displacement
`
`amplification is disclosed.
`
`Additional examples of non—modified strand
`
`displacement amplification mediated by four additional
`
`enzymes generated by engineering of other nucleases is
`
`also disclosed. An example of non—modified strand
`
`displacement amplification mediated by a restriction
`
`endonuclease with a nicked intermediate is disclosed.
`
`Finally, approaches for constructing such nicking
`
`endonucleases are disclosed.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`Figure 1A shows the recognition sequence (SEQ ID
`
`NO:1) and site of cleavage of N.BstNBI nicking
`
`10
`
`10
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`-10-
`
`endonuclease. N.BstNBI recognizes a simple asymmetric
`
`sequence, 5' GAGTC 3', and it cleaves only one DNA
`
`strand, 4 bases away from the 3'—end of its recognition
`
`site,
`
`indicated by the arrow head.
`
`Figure 1B shows the gene organization of N.BstNBI
`
`restriction—modification system where n.bstNBIR (R)
`
`is
`
`the N.BstNBI restriction endonuclease gene and n.bstNBIM
`
`(M) is the N.BstNBI modification methyltransferase gene.
`
`Figure 2 shows the DNA sequence of n.bstNBIR gene
`
`and its encoded amino acid sequence (SEQ ID NO:2 AND SEQ
`
`ID NO:3).
`
`Figure 3 shows the DNA sequence of n.bstNBIM'gene
`
`and its encoded amino acid sequence (SEQ ID NO:4 and SEQ
`
`ID NO:5).
`
`Figure 4 shows the DNA sequence of pleIM'gene and
`
`its encoded amino acid sequence (SEQ ID NO:6 and SEQ ID
`
`NQ:7).
`
`Figure 5 shows the cloning vectors of pHKUVS (SEQ
`
`ID NO:8) .
`
`Figure 6 shows the cloning vectors of pHKT7
`
`(SEQ ID
`
`NO:9).
`
`Figure 7 shows the result of non—modified strand
`
`displacement amplification using nicking enzyme
`
`N.BstNBI.
`
`Lane 1 shows the molecular weight standards
`
`and Lane 2 shows the 160—bp DNA fragment produced from
`
`SDA by N.BstNBI, which is indicated by the arrow head.
`
`11
`
`11
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`-11-
`
`Figure 8 shows the result of non—modified Strand
`
`displacement amplification using five nicking enzymes,
`
`with duplicate samples run. Lanes 1 and 12 are the
`
`molecular weight marker lanes (100 bp ladder).
`
`Lanes 2
`
`and 3, N.BstNBl;
`
`lanes 4 and 5, N.AlwI;
`
`lanes 6 and 7
`
`N.MlyI;
`
`lanes 8 and 9, N.BvaI—l—35;
`
`lanes 10 and 11,
`
`BvaI—2—l2. Arrow indicates the position of
`
`the
`
`expected 100—120 bp product bands.
`
`Figure 9 shows the result of non—modified strand
`
`displacement amplification using BerI, an enzyme that
`
`cleaves in two steps. Panel A, SDA reactions as
`
`described in Example 6 with:
`
`lane 1, no DNA substrate,
`
`no product appearing;
`
`lane 2, no BerI, no product
`
`appearing;
`
`lane 3, complete reaction, 150 bp amplicon
`
`appearing. M: size standard markers HaeIII digest of
`
`¢Xl74; Panel B, SDA reactions as described in Example 6
`
`but with different DNA substrates leading to different
`
`sized amplicons: Lane 1, 150 bp product;
`
`lane 2 — 190 bp
`
`product;
`
`lane 3 — 330 bp product;
`
`lane 4 — 430 bp
`
`product;
`
`lane 5 — 500 bp product. M: size standard
`
`markers HaeIII digest of ¢Xl74
`
`DETAILED DESCRIPTION OF THE INVENTION
`
`In accordance with one embodiment of this
`
`invention, procedures to identify and create site—
`
`specific nicking enzymes are described, and suitability
`
`of their application to SDA in the absence of modified
`
`nucleotides such as a—thio nucleotides is demonstrated.
`
`Those skilled in the art will appreciate that for
`
`use in SDA, a nicking enzyme must have sequence—
`
`specificity in that activity, so that a single nick can
`
`be introduced at the location of the desired priming
`
`12
`
`12
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`_12_
`
`site.
`
`In SDA as conventionally applied,
`
`the sequence—
`
`specific nicking activity derives from two factors:
`
`the
`
`sequence—specificity of the restriction endonuclease
`
`employed and the strand—specificity enforced by the
`
`employment of modified (e.g. a—thiophosphate
`
`substituted, boron—substituted (a—boronated) dNTPs or
`
`cytosine—5 dNTP) nucleotides. This procedure increases
`
`the cost
`
`(due to the expense of the modified
`
`nucleotides) and reduces the length of the amplicon that
`
`can be synthesized (due to poor incorporation by the
`
`polymerase).
`
`In the present invention, it is demonstrated that
`
`appropriate cleavage specificity can be enabled in other
`
`general ways. Five examples of such enzymes are
`
`disclosed in the present invention, obtained in four
`
`different ways.
`
`In one preferred embodiment, both sequence
`
`specificity and strand specificity are obtained in an
`
`enzyme as found in the original host, exemplified by
`
`N.BstNBI.
`
`The cloning of the N.BstNBI restriction
`
`endonuclease gene from Bacillus stearothermophilus 33M
`
`(NEB #928, New England Biolabs, Inc., Beverly, MA)
`
`proved to be challenging. A methylase selection strategy
`
`was tried and one methylase expression clone was
`
`isolated. However,
`
`the flanking ORFs did not encode the
`
`N.BstNBI nicking enzyme. This turned out to be an
`
`orphan methylase, i.e., a methylase not associated with
`
`the cognate endonuclease gene.
`
`The method by which the
`
`N.BstNBI nicking endonuclease was preferably cloned and
`
`expressed in E. coli is described herein:
`
`13
`
`13
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`_13_
`
`1.
`
`Purification of the N.BstNBI restriction
`
`endonuclease to near homogeneity and N—terminal and
`internal amino acid sequence determination.
`
`Nine chromatography columns were used to purify
`
`the N.BstNBI endonuclease protein. They included an XK
`
`50/14 fast flow P—cell column, an HR 16/10 Sourcefl415Q,
`
`five HR 16/10 Heparin—TSK—Guardgel columns, an HR 10/10
`
`SourceTM 15Q coliuma and a ResourceTM 15S. The
`
`purification yielded one protein band at approximately
`
`72 kDa on an SDS—PAGE protein gel following Coomassie
`
`blue staining. The N—terminal 31 amino acid residues
`
`were determined by sequential degradation of the
`
`purified protein on an automated sequencer. To determine
`
`its internal protein sequence, a 6—kDa polypeptide
`
`fragment was obtained following cyanogen bromide
`
`digestion of the 72—kDa N.BstNBI protein. The first 13
`
`amino acid residues of this 6—kDa were determined. This
`
`13—amino acid sequence differs from the sequence of the
`
`N—terminal 31 amino acid residues, suggesting it was
`
`internal N.BstNBI protein sequence.
`
`Amplification of a segment of the N.BstNBI
`2.
`endonuclease gene and subsequent cloning.
`
`Degenerate primers were designed based on both the
`
`N—terminal and internal amino acid sequences. These
`
`primers were used to PCR amplify the 5’ end of the
`
`endonuclease gene. PCR products were cloned into plasmid
`
`pCAB16 and sequenced. The approximately 1.4 kb PCR
`
`fragment was then identified by comparing the amino acid
`
`sequences deduced from.the cloned DNA with the N—
`
`terminal amino acid sequence of the N.BstNBI
`
`endonuclease protein.
`
`14
`
`14
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`_14_
`
`Chromosome walking via inverse PCR to isolate
`3.
`the N.BstNBI endonuclease and methylase gene.
`
`To clone the entire N.BstNBI endonuclease gene as
`
`well as its corresponding DNA methylase gene,
`
`inverse
`
`PCR techniques were adopted to amplify DNA adjacent to
`
`the original 1.4 kb endonuclease gene fragment
`
`(Ochman
`
`et al., Genetics 120:621 (1988); Triglia et al., NUcl.
`
`Acids Res. 16:8186 (1988) and Silver and Keerikatte, J.
`
`Cell. Biochem.
`
`(Suppl.)
`
`l3Ez306, Abstract No. WH239
`
`(1989)). In total,
`
`two rounds of inverse PCR were
`
`performed. At that point,
`
`the endonuclease and the
`
`methylase Open reading frames (ORF) were identified
`
`(Figure 1B).
`
`The endonuclease gene (n.bstNBIR)
`
`turned out to be
`
`a l8lS—bp ORF that codes for a 604—amino acid protein
`
`with a deduced molecular weight of 70,368 Daltons
`
`(Figure 2). This agreed with the observed molecular mass
`
`of the N.BstNBI endonuclease that was purified from
`
`native Bacillus Stearothermophilus 33M. Close to the
`
`endonuclease gene a 906—bp ORF, n.bstNBIM, was found. It
`
`was oriented in a convergent manner relative to the
`
`endonuclease (Figure 1B). The protein sequence deduced
`
`from the n.bstNBIM‘gene shares significant sequence
`
`similarity with other adenine methylases (Figure 3).
`
`Expression of N.BstNBI endonuclease gene
`4.
`using pHKUVS and pHKT7 plasmids.
`
`The two—step method for cloning restriction—
`
`modification systems is described in U.S. Patent No.
`
`5,320,957. The first step is protection of the host cell
`
`from corresponding endonuclease digestion by pre—
`
`modification of recognition sequences. This is
`
`accomplished by introducing the methylase gene into a
`
`host cell and expressing the gene therein. The second
`
`15
`
`15
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`-15_
`
`step includes introduction of the endonuclease gene into
`
`the pre—modified host cell and subsequent endonuclease
`
`production.
`
`The pleIM'gene (Figure 4) was cloned into plasmid
`
`pHKUVS
`
`(Figure 5) and transformed into E. coli cells. As
`
`a result,
`
`the E. coli cells were modified by the pHKUVS—
`
`pleIM. In this case,
`
`the PleI methylase (pleIM) was used
`
`for pre—modification of the host cells because PleI and
`
`N.BstNBI share the same recognition sequence.
`
`The endonuclease gene, n.bstNBIR, was cloned into
`
`pHKT7
`
`(Figure 6), and then introduced into E. coli
`
`ER2566 containing pHKUVS—pleIM1 The culture was grown to
`
`middle log and then induced by the addition of IPTG to a
`
`final concentration of 0.4 mM. The yield of recombinant
`
`N.BstNBI endonuclease is 4 X 107 units per gram cells.
`
`In other embodiments, appropriate cleavage
`
`specificity for SDA is enabled by mutational alteration
`
`of enzymes having double—stranded cleavage activity.
`
`In
`
`a preferred embodiment,
`
`the sequence specificity is
`
`conferred by the specificity of a restriction enzyme, as
`
`in conventional SDA, but the strand specificity is
`
`engineered into it by mutation, so that a single
`
`purified enzyme recognizes a specific sequence and
`
`specifically nicks only one strand. Three distinct
`
`approaches to obtaining strand—specificity (nicking
`
`activity) have been devised and exemplified.
`
`Each
`
`enables performance of SDA in the absence of a—thio
`
`nucleotides. These approaches are described
`
`hereinbelow.
`
`16
`
`16
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`—l6—
`
`Identification of suitable target enZymes for
`1.
`engineering into nicking enzymes
`
`Sequence—specific restriction endonucleases can be
`
`identified by methods well known in the art, and many
`
`approaches to cloning these have been devised, as
`
`described above.
`
`For the present invention,
`
`two
`
`subclasses of restriction endonucleases can be
`
`identified that are preferred starting materials for
`
`creation of sequence—specific nicking endonucleases.
`
`These will be referred to below as subclass A and
`
`subclass B.
`
`For one of these classes,
`
`the approach to
`
`obtaining mutants that nick specifically is divided into
`
`two subsets,
`
`to be referred to as subclass Al and
`
`subclass A2. Isolation and characterization of mutants
`
`as described in subclass A is disclosed in detail in
`
`U.S. Application Serial No.
`
`filed concurrently
`
`herewith and will be summarized here. Isolation and
`
`characterization of mutants of subclass B enzymes will
`
`be described in detail here.
`
`Both classes of enzymes are found among those
`
`listed in REBASE (http://rebase.neb.com/rebase.charts.
`
`html “Type Iis enzymes” link; Roberts and Marcelis,
`
`NUCleic Acids Res. 29:368—269 (2001))as Type IIS
`
`endonucleases. These can be identified among
`
`restriction endonucleases as those in which the
`
`recognition site is asymmetric.
`
`However, specifically those enzymes belonging to
`
`subclass A are frequently referred to as 'Type IIS'
`
`endonucleases (Szybalski, Gene 100:13—26 (1991)). These
`
`enzymes recognize asymmetric sequences and cleave the
`
`DNA outside of, and to one side of,
`
`the recognition
`
`sequence. The examples that have been studied each
`
`comprise an N—terminal sequence—specific DNA binding
`
`17
`
`17
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`_ l '7 _.
`
`moiety,
`
`joined with a C—terminal sequence—non—specific
`
`cleavage moiety by zero or more amino acids.
`
`Enzymes belonging to subclass B are often referred
`
`to as 'Type IIT' endonucleases (Kessler, et al., Gene
`
`47:1—153 (1986); Stankevicius, et al. NUcleic Acids Res.
`
`26:1084—1091 (1998)), or alternately as 'Type IIQ'
`
`endonucleases (Degtyarev, et al., NUcleic Acids Res.
`
`18:5807—5810 (1990); Degtyarev, et al., NUcleic Acids
`
`Res. 28:e56 (2000)). These enzymes also recognize
`
`asymmetric sequences but they cleave the DNA within the
`
`recognition sequence.
`
`Methods for identifying and characterizing the
`
`recognition site of a restriction endonuclease are well—
`
`known in the art. In addition, a list of the known
`
`enzymes belonging to these, and other, groups may be
`
`obtained from REBASE at http://rebase.neb.com.
`
`2.
`
`Creation of nicking mutants from subclass A
`
`The subclass A enzymes studied were Pka, MlyI,
`
`PleI, and Ale. Enzymes of this subclass are thought to
`
`act symmetrically with respect to strand—cleavage. The
`
`C—terminal domains of two identical protein molecules
`
`are believed to interact transiently during DNA cleavage
`
`to form a homodimer.
`
`Two of the enzymes disclosed in the present
`
`invention were derived from subclass A enzymes in one of
`
`two ways.
`
`In one preferred embodiment
`
`(method A1)
`
`cleavage of one of the two DNA strands was suppressed by
`
`mutating, within the endonuclease gene,
`
`the region
`
`coding for the dimerization interface that is needed for
`
`double—strand cleavage, such that only one cleavage
`
`occurs. This mutation may comprise alteration of
`
`18
`
`18
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`—l8—
`
`particular residues required for dimerization
`
`individually or together.
`
`In the other preferred embodiment
`
`(method A2),
`
`cleavage of one of the two strands was suppressed by
`
`substitution of the region of the endonuclease
`
`containing the dimerization interface with a
`
`corresponding region from an endonuclease known to be
`
`dimerization—defective. This region may be obtained
`
`from a portion of a gene such as the gene encoding
`
`N.BstNBI,
`
`the endonuclease of the present invention
`
`described above, or may be obtained from other
`
`naturally—occurring or from engineered genes containing
`
`this dimerization function.
`
`3.
`
`Creation of nicking mutants from subclass B.
`
`The fourth and fifth nicking endonucleases
`
`disclosed in the present invention were derived from the
`
`enzyme BvaI, a member of subclass B. Enzymes of
`
`subclass B are thought to act asymmetrically with
`
`respect to strand—cleavage. They are envisaged to be
`
`functionally heterodimeric,
`
`that is to say to comprise
`
`two different subunits, or domains, each with its own
`
`catalytic site.
`
`In the active enzyme,
`
`the two subunits,
`
`or domains,
`
`interact to achieve DNA recognition
`
`together, and to catalyze double—strand cleavage. Of
`
`four subclass B enzymes studied—AciI, BsrBI, BssSI, and
`
`BvaI—only BvaI comprised two different protein
`
`subunits. The other three enzymes were single proteins
`
`each of which, we presume, comprises two different
`
`domains.
`
`In principle, nicking mutants can be made from
`
`either kind of enzyme, although doing so is more
`
`straightforward using enzymes that,
`
`like BvaI, comprise
`
`separate, rather than joined, subunits.
`
`19
`
`19
`
`

`

`WO 01/94544
`
`PCT/US01/17804
`
`-19_
`
`Identification of heterodimeric enzymes of
`A.
`subclass B.
`
`Heterodimeric members of the subclass may be
`
`recognized in two ways: by analysis of endonuclease
`
`purified from the original organism or from a
`
`recombinant host containing the cloned restriction
`
`system, or by sequence analysis of the cloned
`
`restriction system.
`
`In the former case,
`
`the purified
`
`endonuclease may be characterized by electrophoresis on
`
`SDS—PAGE, which will usually reveal the presence of two
`
`protein components migrating at different