(12) INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY(PCT)
`
`(19) World Intellectual Property Organization
`International Bureau
`
`(43) International Publication Date
`24 July 2003 (24.07.2003)
`
`
`
`(10) International Publication Number
`WO 03/060084 A2
`
`Delicias, Rancho Santa Ke, CA 92067-7214 (US).
`PARRA-GESSERT, Lilian [CL/US]; 7023 Whipple
`(21) International Application Number:=PCT/US03/01189
`Avenue, San Diego, CA 92122 (US).
`
`(51) International Patent Classification’:
`
`C12N
`
`(22) International Filing Date: 14 January 2003 (14.01.2003)
`
`(74)
`
`(25) Filing Language:
`
`(26) Publication Language:
`
`English
`
`English
`
`(81)
`
`(30) Priority Data:
`60/348,609
`60/348,761
`60/348,764
`10/077,474
`
`14 January 2002 (14.01.2002)
`14 January 2002 (14.01.2002)
`14 January 2002 (14.01.2002)
`14 February 2002 (14.02.2002)
`
`US
`US
`US
`US
`
`(for all designated States except US): DI-
`(71) Applicant
`4955 Directors
`VERSA CORPORATION [US/US];
`Place, San Diego, CA 92121-1609 (US).
`
`(84)
`
`(72)
`(75)
`
`Inventors; and
`FREY, Ger-
`(for US only):
`Inventors/Applicants
`hard [DE/US]; 13768 Via Cima Bells, San Diego, CA
`92129 (US). SHORT, Jay, M.
`[US/US]; 6801 Paseo
`
`Agents: EINHORN,Gregory,P.et al.; Fish & Richardson
`P.C., 4350 La Jolla Village Drive, Suite 500, San Diego, CA
`92122 (US).
`
`Designated States (national): AE, AG, AL, AM, AT, AU,
`AZ, BA, BB, BG, BR, BY, BZ, CA, CII, CN, CO, CR, CU,
`CZ, DE, DK, DM, DZ, EC, EE, ES, FI, GB, GD, GE, GH,
`GM,HR,HU, ID, IL,IN, IS, JP, KE, KG, KP, KR, KZ, LC,
`LK, LR, LS, LT, LU, LV, MA, MD, MG, MK, MN, MW,
`MX, MZ, NO, NZ, OM, PH, PL, PT, RO, RU, SD, SE, SG,
`SK, SL, TJ, TM, TN, TR, TT, TZ, UA, UG, US, UZ, VN,
`YU, ZA, ZM, ZW.
`
`Designated States (regiona/): ARIPO patent (GII, GM,
`KE, LS, MW, MZ, SD, SL, SZ, TZ, UG, ZM, ZW),
`Eurasian patent (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM),
`European patent (AT, BE, BG, CII, CY, CZ, DE, DK, EE,
`ES, FI, FR, GB, GR, HU,IE, IT, LU, MC, NL, PT, SE,SI,
`SK, TR), OAPI patent (BF, BJ, CF, CG, CI, CM, GA, GN,
`GQ, GW, ML, MR, NE, SN, TD, TG).
`
`[Continued on next page]
`
`(54) Title:
`CLEOTIDES
`
`METHODS FOR MAKING POLYNUCLEOTIDES AND PURIFYING DOUBLE-STRANDED POLYNU-
`
`Elongation Cycle
`Load
`11238
`|
`oe:—
`“a oun:
`‘
`
`Wash
`
`}
`
`Wash
`
`[| Qe111222
`Cut
`@— tooo
`|
`@e— 111111222
`
`Wash
`
`Qo113s:
`
`111338333
`4111222
`
`@—121222
`
`(57) Abstract: The invention provides methods for identifying and purifying double-stranded polynucleotides lacking base pair
`mismatches, insertion/deletion loops and/or nucleotide gaps. The invention provides libraries of nucleic acid building blocks and
`methodsfor generating any nucleic acid sequence, including synthetic genes, antisense constructs and polypeptide coding sequences.
`The invention provides chimeric antigen binding molecules and the nucleic acids that encode them.
`
`O03/060084A2
`
`

`

`WO 03/060084 A2
`
`—_[MITININIIINANITTINY TITERAT ITATMM IT Mt
`
`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 and Abbreviations" appearing at the begin-
`ning ofeach regular issue ofthe PCT Gazette.
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`METHODS FOR MAKING POLYNUCLEOTIDES
`AND PURIFYING DOUBLE-STRANDED POLYNUCLEOTIDES
`
`TECHNICAL FIELD
`
`The present invention is generally directed to the fields of genetic and protein
`
`engineering and molecular biology. In particular, the invention provides methods for
`
`identifying and purifying double-stranded polynucleotides lacking base pair mismatches,
`
`insertion/deletion loops and nucleotide gaps.
`
`The present invention is generally directed to the fields of protein and genetic
`
`engineering and molecular biology. In one aspect, the invention is directed to libraries of
`oligonucleotides and methods for generating any nucleic acid sequence, including synthetic
`
`genes, antisense constructs and polypeptide coding sequences. In one aspect, the libraries of
`the invention comprise oligonucleotides comprising restriction endonucleaserestriction sites,
`
`e.g., Type-IIS restriction endonucleaserestriction sites, wherein the restriction endonuclease
`
`cuts at a fixed position outside of the recognition sequence to generate a single stranded
`overhang. The polynucieotide construction methods comprise use oflibraries of pre-made
`multicodon(e.g., dicodon) oligonucleotide building blocks and Type-IIS restriction
`
`10
`
`15
`
`endonucleases.
`
`In one aspect, the invention is directed to methods for generating sets, or
`
`libraries, of nucleic acids encoding chimeric antigen binding molecules, including,e.g.,
`
`20
`
`antibodies and related molecules, such as antigen binding sites and domains and other
`
`antigen binding fragments, including single and double stranded antibodies. This invention
`provides methods for generating new or variant chimeric antigen binding polypeptides,e.g.,
`antigen bindingsites, antibodies and specific domains or fragments of antibodies(e.g., Fab or
`Fc domains) by altering the nucleic acids that encode them by,e.g., saturation mutagenesis,
`an optimized directed evolution system, synthetic ligation reassembly, or a combination
`
`thereof.
`
`The invention also provides libraries of chimeric antigen binding polypeptides
`
`encodedbythe nucleic acid libraries of the invention and generated by the methodsof the
`invention. These antigen binding polypeptides can be analyzed using any liquid orsolid state
`screening method, e.g., phage display, ribosome display, using capillary array platforms, and
`the like. The polypeptides generated by the methods of the invention can be usedin vitro,
`1
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`e.g., to isolateor identify antigensor in vivo, e.g., to treat or diagnose various diseases and
`conditions, to modulate, stimulate or attenuate an immune response. The inventionalsois
`directed to the generation of chimeric immunoglobulins for administering passive immunity
`and nucleic acids encoding these chimeric antigen binding molecules for genetic vaccines.
`
`BACKGROUND
`
`Synthetic oligonucleotides are commonly used to construct nucleic acids,
`including polypeptide coding sequences and gene constructs. However, even the best
`oligonucleotide synthesizer has a 1% to 5% error rate. These errors can result in improper
`base pair sequences, whichcan lead to generation of an erroneousprotein sequences. These
`errors can also result in sequences that cannot be properly transcribed or untranslated,
`including, e.g., premature stop codons. To detect these errors, the oligonucleotides or the
`sequences generated using the oligonucleotides aresequenced. However, sequencing to
`detect errors in nucleic acid synthetic techniques is time consuming and expensive.
`Engineering genes, polypeptide coding sequences and other polynucleotide
`molecules can be impeded bythe needto isolate, synthesize or handle a parental, or template,
`DNAsequence. For example, it may be necessary to alter codon usagefor optimal
`expression in a cell host, requiring manipulation ofthe polynucleotide sequence. Frequently
`is it desirable or necessary to add and/or removerestriction sites to an isolated, cloned or
`amplified polynucleotide to facilitate manipulation ofthe sequence, requiring further
`modification of the molecule. All of these manipulations introduce labor costs and are
`potential sources of sequence and cloningerrors.
`The best quality oligonucleotide synthesis systems availablestill contain up to
`1% of (n-1) and (n-2) contaminations leading to a high errorrate in the nucleic acid
`sequences(c.g. genes, gene pathways, or regulatory motifs) built. These errors can manifest
`themselves as frameshifts or as stop codon,resulting in truncated proteinsif the engineered
`gene is expressed. Sometimes, more than 20 clones have to be sequenced anderrors
`corrected (e.g., by site directed mutagenesis) to get the desired nucleotide sequencefor a
`single gene or coding sequence. In the case of chimeric polynucleotide libraries sequencing
`and correcting all errors is not an option and oligo-based sequence errors decrease cloning
`and screening efficiency significantly.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`_. Antigen binding polypeptides, such as antibodies, are increasingly used in a
`variety of therapeutic applications. For example, in immunotherapy, antibodies are used to
`directly kill target cells, such as cancer cells. They can be administered to generate passive
`immunity. Antigen binding polypeptidesare also usedas carriers to deliver cytotoxic or
`imaging reagents. Monoclonal antibodies (mAbs) approved for cancer therapy are now in
`PhaseII and if trials. Certain anti-idiotypic antibodies that bindto the antigen-combining
`sites of antibodies can effectively mimic the three-dimensionalstructures and functions of the
`external antigens and can be used as surrogate antigensfor active specific immunotherapy.
`Bi-specific antibodies combine immunecell activation with tumorcell recognition; thus,
`tumorcells or cells expressing tumorspecific antigens (e.g., tumor vasculature) are killed by
`pre-defined effector cells. Antibodies can be administered to increase or decreasethe levels
`of cytokines or hormonesby direct bindingor by stimulating or inhibiting secretory cells.
`Accordingly, increasing the affinity or avidity of an antibody to a desired antigen, such as a
`cancer-specific antigen, would result in greater specificity of the antibodyto its target,
`resulting in a variety of therapeutic benefits, such as needing to administer less antibody-
`containing pharmaceutical.
`
`SUMMARY
`
`METHODSFOR PURIFYING AND IDENTIFYING DOUBLE-STRANDED NUCLEIC
`ACIDS LACKING BASE PAIR MISMATCHES, INSERTION/DELETION LOOPS OR
`NUCLEOTIDE GAPS
`
`The invention provides methodsfor identifying and purifying double-stranded
`polynucleotides lacking nucleotide gaps, base pair mismatches andinsertion/deletion loops.
`In one aspect, the invention provides methods for purifying double-stranded polynucleotides
`lacking base pair mismatches, insertion/deletion loops and/or nucleotide gaps comprising the
`following steps: (a) providinga plurality of polypeptides that specifically bind to a base pair
`mismatch, an insertion/deletion loop and/or a nucleotide gap or gaps within a double stranded
`polynucleotide; (b) providing a sample comprisirig a plurality of double-stranded
`polynucleotides; (c) contacting the double-stranded polynucleotides of step (b) with the
`polypeptides of step (a) under conditions wherein a polypeptide of step (a) can specifically
`bind to a base pair mismatch,an insertion/deletion loop and/or a nucleotide gap or gaps in a
`double stranded polynucleotide of step (b); and (d) separating the double-stranded
`
`10
`
`15
`
`26
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`polynucleotides lacking a specifically bound polypeptide of step (a) from the double-stranded
`polynucleotides to which a polypeptide ofstep (a) has specifically bound, thereby purifying
`double-stranded polynucleotides lacking base pair mismatches, insertion/deletion loops
`and/or nucleotide gaps. In one aspect, the double-stranded polynucleotide comprises a
`double-stranded oligonucleotide. In one aspect, the double-stranded polynucleotide consists
`of a double-stranded oligonucleotide.
`In alternative aspects, the double-stranded polynucleotide is between about 3
`
`and about 300 basepairs in length; between 10 and about 200 basepairs in length; and,
`
`between 50 and about 150 base‘pairs in length. In alternative aspects, the gaps in the double-
`
`10
`
`stranded polynucleotide are between about 1 and 30, about 2 and 20, about 3 and 15, about 4
`
`and 12 and about 5 and 10 nucleotidesin length.
`
`In alternative aspects, the the base pair mismatch comprises a C:T mismatch,a
`
`G:A mismatch, a C:A mismatch or a G:U/T mismatch.
`
`In one aspect, the polypeptide that specifically binds to a base pair mismatch,
`
`15
`
`an insertion/deletion loop and/or a nucleotide gap or gaps in a double stranded
`
`polynucleotide comprises a DNArepair enzyme. In alternative aspects, the DNA repair
`enzymeis.a bacterial DNA repair enzyme, a MutS DNArepair enzyme, a Taq MutS DNA
`repair enzyme, an Fpg DNA repair enzyme, a MutY DNArepair enzyme, a hexA DNA
`mismatch repair enzyme, a Vsr mismatch repair enzyme, a mammalian DNArepair enzyme
`and natural or synthetic variations and isozymes thereof. In one aspect, the DNA repair
`enzyme is a DNA glycosylase that initiates base-excision repair of G:U/T mismatches. The
`DNA glycosylase can comprise a bacterial mismatch-specific uracil-DNA glycosylase
`(MUG) DNArepair enzyme or a eukaryotic thymine-DNA glycosylase (TDG) enzyme.
`In one aspect, the separating of the double-stranded polynucleotides lacking a
`specifically bound polypeptide ofstep (a) from the double-stranded polynucleotides to which
`a polypeptide of step (a) has specifically bound of step (d) comprises use of an
`immunoaffinity column, wherein the column comprises immobilized antibodies capable of
`specifically binding to the specifically bound polypeptide or an epitope boundto the
`specifically bound polypeptide, and the sample is passed through the immunoaffinity column
`under conditions wherein the immobilized antibodies are capable of specifically binding to
`
`20
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`the specifically bound polypeptide or the epitope boundto the specifically bound
`polypeptide.
`
`in one aspect, the separating of the double-stranded polynucleotides lacking a
`specifically bound polypeptide of step (a) from the double-stranded polynucleotides to which
`a polypeptide of step (a) has specifically bound of step (d) comprises use of an antibody,
`wherein the antibody is capable ofspecifically binding to the specifically bound polypeptide
`or an epitope boundto the specifically bound polypeptide and the antibody is contacted with
`the specifically bound polypeptide under conditions wherein the antibodies are capable of
`specifically binding to the specifically bound polypeptide or an epitope boundto the
`specifically bound polypeptide. The antibody can be an immobilized antibody. The
`antibody can be immobilized onto a bead or a magnetized particle or a magnetized bead.
`
`In one aspect, the separating of the double-stranded polynucleotides lacking a
`specifically bound polypeptide of step (a) from the double-stranded polynucleotides to which
`a polypeptide of step (a) has specifically bound of step (d) comprises use of an affinity
`column, wherein the column comprises immobilized binding molecules capable of
`specifically binding to a tag linkedto the specifically bound polypeptide and the sampleis
`passed through the affinity column under conditions wherein the immobilized antibodies are
`
`capable of specifically bindingto the tag linked to the specifically bound polypeptide. The
`immobilized binding molecules can comprise an avidin or a natural or synthetic variation or
`homologuethereof andthe tag linked to the specifically bound polypeptide can comprise a
`biotin or a natural or synthetic variation or homologuethereof.
`
`In one aspect, the separating of the double-stranded polynucleotides lacking a
`specifically bound polypeptide of step (a) from the double-stranded polynucleotides to which
`
`a polypeptide of step (a) has specifically bound of step (d) comprises use of a size exclusion
`
`10
`
`20
`
`25
`
`column, such as a spin column. Alternatively, the separating can comprise use of a size
`
`exclusion gel, such as an agarose gel.
`
`In one aspect, the double-stranded polynucleotide comprises a polypeptide
`coding sequence. The polypeptide coding sequence can comprise a fusion protein coding
`
`30
`
`sequence. The fusion protein can comprise a polypeptide of interest upstream ofan intein,
`wherein the intein comprises a polypeptide. The intein polypeptide can comprise an enzyme,
`such as one used to identify vector or insert positive clones, such as Lac Z. The intein
`
`5
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`10
`
`15
`
`polypeptide can comprise an antibody or a ligand. In one aspect, the intein polypeptide
`comprises a polypeptide selectable marker, such as an antibiotic. The antibiotic can
`comprise a kanamycin,a penicillin or a hygromycin.
`The invention provides a method for assembling double-stranded
`oligonucleotides to generate a polynucleotide lacking base pair mismatches,
`insertion/deletion loops and/or nucleotide gaps comprising the following steps: (a)
`providing a plurality ofpolypeptides that specifically bind to a base pair mismatch, an
`insertion/deletion loop and/or a nucleotide gap or gaps in a double stranded polynucleotide;
`(b) providing a sample comprisinga plurality of double-strandedoligonucleotides; (c)
`contacting the double-stranded oligonucleotides of step (b) with the polypeptides of step (a)
`under conditions wherein a polypeptide of step (a) can specifically bindto a base pair
`mismatch, an insertion/deletion loop and/or a nucleotide gap or gaps in a double stranded
`oligonucleotide of step (b); (d) separating the double-stranded oligonucleotides lacking a
`specifically bound polypeptide of step (a) from the double-stranded oligonucleotides to
`which a polypeptide of step (a) has specifically bound, thereby purifying double-stranded
`oligonucleotides lacking base pair mismatches, insertion/deletion loops and/or a nucleotide
`gap or gaps; and (e) joining together the purified double-stranded oligonucleotides lacking
`base pair mismatches andinsertion/deletion loops, thereby generating a polynucleotide
`lacking base pair mismatches, insertion/deletion loops and/or nucleotide gaps.
`In one aspect, the double-stranded oligonucleotides comprise libraries of
`oligonucleotides, e.g., the libraries ofthe invention comprising oligonucleotides comprising
`multicodons. For example, the double-stranded oligonucleotides can comprise libraries of
`oligonucleotides comprising multicodon,e.g., dicodon, building blocks. In one aspect, the
`library comprises a plurality of double-stranded oligonucleotide members, wherein each
`oligonucleotide member comprises two or more codonsin tandem (e.g., a dicodon) and a
`Type-IIS restriction endonuclease recognition sequence flanking the 5’ and the 3’ end ofthe
`multicodon(e.g., dicodon, tricodon, tetracodon, and the like).
`The invention provides a method for generating a polynucleotide lacking base
`pair mismatches, insertion/deletion loops and/or nucleotide gaps comprising the following
`30
`steps: (a) providingaplurality ofpolypeptides that specifically bind to a base pair
`mismatch, an insertion/deletion loop and/or a nucleotide gap or gaps in a double stranded
`
`20
`
`25
`
`6
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`polynucleotide; .(b) providing a sample comprising a plurality of double-stranded
`
`oligonucleotides; (c) joining together the double-stranded oligonucleotides of step (b) to
`
`generate a double-stranded polynucleotide; (d) contacting the double-stranded
`
`polynucleotide of step (c) with the polypeptides of step (a) under conditions wherein a
`polypeptide of step (a) can specifically bind to a base pair mismatch, an insertion/deletion
`
`loop and/or a nucleotide gap or gaps in a double stranded polynucleotide of step (c); and (e)
`separating the double-stranded polynucleotides lacking a specifically bound polypeptide of
`
`step (a) from the double-stranded polynucleotides to which a polypeptide ofstep (a) has
`specifically bound, thereby purifying double-stranded polynucleotides lacking base pair
`
`10
`
`mismatches, insertion/deletion loops and/or nucleotide gaps. In one aspect, the double-
`
`stranded oligonucleotides comprise a library of oligonucleotides multicodon building blocks,
`
`the library comprising a plurality of double-stranded oligonucleotide members, wherein each
`oligonucleotide member comprisesat least two codons in tandem and a Type-IIS restriction
`endonuclease recognition sequence flanking the 5’ and the 3” end of the multicodon.
`.
`In one aspect, the method further comprises providing a set of 61 immobilized
`
`starter oligonucleotides, one oligonucleotide for each possible amino acid codingtriplet,
`
`wherein the oligonucleotides are immobilized on a substrate and havea single-stranded
`overhang corresponding to a single-stranded overhang generated by a Type-IISrestriction
`
`endonuclease, or, the oligonucleotides comprise a Type-IIS restriction endonuclease
`
`recognition site distal to the substrate and a single-stranded overhang is generated by
`digestion with a Type-IIS restriction endonuclease; digesting a second oligonucleotide
`memberfrom the library of step (a) with a Type-IIS restriction endonuclease to generate a
`
`15
`
`20
`
`single-stranded overhang; and contacting the digested second oligonucleotide memberto the
`immobilized first oligonucleotide member under conditions wherein complementary single-
`
`25
`
`stranded base overhangsofthe first and the second oligonucleotides can pair, and,ligating
`
`the second oligonucleotideto the first oligonucleotide, thereby generating a double-stranded
`
`polynucleotide.
`The invention provides a method for generating a base pair mismatch-free,
`insertion/ deletion loop-free and/or gap-free double-stranded polypeptide coding sequence
`comprising the following steps: (a) providing a plurality of polypeptides that specifically
`bind to a base pair mismatch, an insertion/deletion loop and/or a nucleotide gap or gaps
`
`30
`
`7
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`within a double stranded polynucleotide; (b) providing a sample comprising a plurality of
`double-stranded polynucleotides encoding a fusion protein, wherein the fusion protein coding
`
`sequence comprises a coding sequence for a polypeptide of interest upstream of and in frame
`
`with a coding sequence for a markeror a selection polypeptide; (c) contacting the double-
`
`stranded polynucleotides of step (b) with the polypeptides of step (a) under conditions
`wherein a polypeptide of step (a) can specifically bind to a base pair mismatch, an insertion/
`
`deletion loop and/or a nucleotide gap or gaps in a double stranded polynucleotide of step (b);
`(d) separating the double-stranded polynucleotides lacking a specifically bound polypeptide
`
`of step (a) from the double-stranded polynucleotides to which a polypeptide of step (a) has
`specifically bound, thereby purifying double-stranded polynucleotides lacking base pair
`
`mismatches, insertion/deletion loops and/or a nucleotide gap or gaps; (e) expressing the
`
`purified double-stranded polynucleotides and selecting the polynucleotides expressing the
`
`selection marker polypeptide, thereby generating a base pair mismatch-free, insertion/
`
`deletion loop-free and/or gap-free double-stranded polypeptide coding sequence.
`
`In oneaspect, the marker or selection polypeptide comprisesa self-splicing
`intein, and the method further comprises the self-splicing out ofthe intein marker or selection
`polypeptide from the upstream polypeptide of interest. The markeror selection polypeptide
`can comprise an enzyme, such as a enzymeusedto identity insert or vector-positive clones,
`
`such as a LacZ enzyme. The marker or selection polypeptide can also comprise an antibiotic,
`
`10
`
`15
`
`20
`
`such as 4 kanamycin, a penicillin or a hygromycin.
`
`In alternative aspects of the invention, the methods generate a sample or
`
`“batch” of purified oligonucleotides and/or polynucleotides that are 90%, 95%, 96%, 97%,
`
`98%, 99%, 99.5% and 100% or completely free of base pair mismatches, insertion/deletion
`
`loops and/or a nucleotide gap or gaps.
`The nucleic acids manipulated or altered by any means, including random or
`
`stochastic methods, or, non-stochastic, or “directed evolution,” can be “purified” or
`“processed” by the methodsofthe invention,e.g., the methods of the invention can be used
`to generate a sample or “batch” of double-stranded oligonucleotides and/or polynucleotides
`that are 90%, 95%, 96%, 97%, 98%, 99%, 99.5% and 100% or completely free of base pair
`
`mismatches, insertion/deletion loops and/or a nucleotide gap or gaps, wherein the nucleic
`acids (e.g., oligos, polynucleotides, genes, and the like) have been manipulated by stochastic
`
`25
`
`30
`
`8
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`methods, or, non-stochastic, or “directed evolution.” For example, the methods of the
`invention can be used to “purify” or “process” nucleic acids manipulated by saturation
`mutagenesis, an optimized directed evolution system, synthetic ligation reassembly, or a
`combination thereof, as described herein. The methodsof the invention can be used to
`“purify” or “process”nucleic acids manipulated by a method comprising gene site saturated
`mutagenesis (GSSM). The methods of the invention can be used to “purify” or “process”
`nucleic acids manipulated by genesite saturated mutagenesis (GSSM), step-wise nucleic acid
`reassembly, error-prone PCR,shuffling, oligonucleotide-directed mutagenesis, assembly
`PCR, sexual PCR mutagenesis, in vivo mutagenesis, cassette mutagenesis, recursive
`ensemble mutagenesis, exponential ensemble mutagenesis,site-specific mutagenesis, gene
`reassembly, synthetic ligation reassembly (SLR) or a combination thereof. The methods of
`the invention can be used to “purify” or “process” nucleic acids manipulated by
`recombination, recursive sequence recombination, phosphothioate-modified DNA
`mutagenesis, uracil-containing template mutagenesis, gapped duplex mutagenesis, point
`mismatch repair mutagenesis, repair-deficient host strain mutagenesis, chemical mutagenesis,
`radiogenic mutagenesis, deletion mutagenesis, restriction-selection mutagenesis, restriction-
`purification mutagenesis,artificial gene synthesis, ensemble mutagenesis, chimeric nucleic
`acid multimer creation or a combination thereof.
`
`In one aspect, method of the invention comprises purifying a double-stranded
`nucleic acid comprising a synthetic, a naturally isolated, or a recombinantly generated
`nucleic acid (a polynucleotide or an oligonucleotide). The synthetic polynucleotide can be
`identical to a parental or a natural sequence. In one aspect, the polynucleotide comprises a
`gene, a chromosome. In oneaspect, the gene further comprises a pathway. In one aspect, the
`gene comprises a regulatory sequence. In one aspect, the polynucleotide comprises a
`promoter or an enhanceror a polypeptide coding sequence. The polypeptide can be an
`enzyme, an antibody, a receptor, a neuropeptide, a chemokine, a hormone,a signal sequence,
`or a structural gene. In one aspect, the polynucleotide comprises non-coding sequence.
`In one aspect, a polynucleotide purified by a method of the invention
`comprises a DNA(e.g., a gene or coding sequence), an RNA(e.g., an iRNA, an rRNA, a
`tRNA or an mRNA)or a combination thereof. For example, the methodsof the invention
`can be used to generate a sample or “batch” of double-stranded DNA or RNAthat are 90%,
`
`9
`
`10
`
`15
`
`20
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`95%, 96%, 97%, 98%, 99%, 99.5% and 100% or completely free ofbase pair mismatches,
`insertion/deletion loops and/or a nucleotide gap or gaps. In one aspect, the double-stranded
`polynucleotide comprises an iRNA. The double-stranded polynucleotide can comprise a
`DNA,e.g., a gene. In one aspect, the DNA comprises a chromosome.
`
`COMPOSITIONS AND METHODS FOR MAKING POLYNUCLEOTIDES BY
`ASSEMBLY OF CODON BUILDING BLOCKS
`The invention provides methods and compositions for making nucleic acids
`by iterative assembly ofoligonucleotide building blocks. In one aspect, the invention
`provideslibraries of oligonucleotides comprising multicodon(e.g., dicodon,tricodon)
`building blocks. In one aspect, the library comprises a plurality of double-stranded
`oligonucleotide members, wherein each oligonucleotide member comprises two or more
`codonsin tandem (e.g., a dicodon) and a Type-IIS restriction endonuclease recognition
`sequence flanking the 5’ and the 3’ end of the multicodon(e.g., dicodon, tricodon,
`tetracodon, and the like).
`In different aspects, this invention provides that the building blocks can be X-
`mers (where can be any integer from 3 to one billion). In other aspects, six-mers can be used
`that are not dicodonsprior to assembly with other building blocks (because they are frame-
`shifted), but that can become codonsafter assembly with other building blocks. In other
`aspects, the intended product is not a coding sequence (but may be, e.g. a promoter, an
`enhancer, or any other regulatory motif), so the building blocks do not need to function as
`codonseither before or after assembly with other building blocks. In other aspects,the
`assembly product can be, e.g., operons, gene pathways, chromosomes, or genomes. Thus,
`the term “codon”includesall nucleic acid sequences, including sequencesthat code for “non-
`coding” sequences such as regulatory motifs (e.g., promoters, enhancers), operons, structural
`sequences(e.g., telomeres) and thelike.
`In one aspect,the library comprises oligonucleotide members comprising all
`possible codon combinations, e.g., all possible dimer (dicodon) combinations, tricodon
`combinations, tetracodon combinations, and thelike. In one aspect, the library of the
`invention can comprise oligonucleotide members comprising 4096 different possible codon
`dimer (dicodon) combinations(proteins are synthesized according to base triplets (codons) in
`a given DNA sequence; there are 61 different triplets coding for 20 different amino acids).
`
`10
`
`10
`
`20
`
`25
`
`30
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`Thelibrary can be of any size and can include anywhere from one to 4096 different
`members, e.g., the library can comprise about 50, 100, 150, 200, 250, 300, 350, 400, 450,
`500, 600, 700, 800, 900, 1000, 2000, 3000, 4000 or more different members.
`In one aspect,
`
`noneof the codonsare stop codons.
`In one aspect, the Type-IIS restriction endonuclease recognition sequenceat
`the 5’ end of the dicodon differs from the Type-IIS restriction endonuclease recognition
`sequenceat the 3’ end of the dicodon. The Type-IIS restriction endonuclease recognition
`sequence can bespecific for a restriction endonuclease that, upon digestion of the
`oligonucleotide library member, generates a base overhang,including a onebasesingle-
`stranded overhang,a two base single-stranded overhang, a three base single-stranded
`overhang, a four base single-stranded overhang, and thelike. The restriction endonuclease
`can comprise a SapIrestriction endonuclease or an isochizomerthereof, or, an Earl
`restriction endonuclease or an isochizomerthereof. In one aspect, the Type-IIS restriction
`
`endonuclease recognition sequenceis specific for a restriction endonuclease that, upon
`digestion of the oligonucleotide library member, generates a two base single-stranded
`overhang. Therestriction endonuclease can be a BseRI, a BsgI or a Bpmlrestriction
`endonuclease. In one aspect, the Type-IIS restriction endonuclease recognition sequenceis
`specific for a restriction endonuclease that, upon digestion ofthe oligonucleotide library
`member, generates a one base single-stranded overhang. The restriction endonuclease can be
`an N.AlwIor an N.BstNBIrestriction endonuclease.
`
`In one aspect, the Type-IIS restriction endonuclease recognition sequenceis
`specific for a restriction endonucleasethat, upon digestion of the oligonucleotide library
`member, cuts on both sides of the Type-IIS restriction endonuclease recognition sequence.
`Therestriction endonuclease can be a Begl, a BsaXI or a BspCNIrestriction endonuclease.
`In one aspect, each oligonucleotide library memberconsists essentially of two
`codonsin tandem (a dicodon) and a Type-IIS restriction endonuclease recognition sequence
`flanking the 5’ and the 3’ end of the dicodon.
`In alternative aspects, the oligonucleotide library members are between about
`20 and 400 basepairs in length, between about 40 and 200 basepairsin length or between
`about 100 and 150 basepairs in length.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`11
`
`

`

`WO 03/060084
`
`PCT/US03/01189
`
`_. The oligonucleotide library member can comprise a (complementary base
`paired) sequence (NNN)(NNN) AGAAGAGC (SEQ ID NO:1) and (NNN)(NNN)
`TCTTCTCG (SEQ ID NO:2), wherein (NNN)is a codon and N is A, C, T or G or an
`
`equivalent thereof.
`
`The oligonucleotide library member can comprise a (complementary base
`
`paired) sequence (NNN)(NNN) TGAAGAGAG(SEQ ID NO:3) and (NNN)(NNN)
`
`ACTTCTCTC(SEQ ID NO:4), wherein (NNN) is a codon and N is A, C, T or G or an
`
`equivalent thereof.
`
`10
`
`The oligonucleotide library member can comprise a (complementary base
`paired) sequence (NNN)(NNN) TGAAGAGAG CT GCTACTAACT GCA (SEQ ID NO:5)
`and (NNN)(NNN) ACTTCTCTC GA CGATGATTG (SEQ ID NO:6), wherein (NNN) is a
`codon and N is A, C, T or G or an equivalent thereof.
`
`The oligonucleotide library member can comprise a (complementary base
`
`paired) sequence CTCTCTTCA NNN NNN AGAAGAGC (SEQ ID NO:7) and
`GAGAGAAGT NNN NNN TCTTCTCG (SEQ ID NO:8), wherein (NNN) is a codon and N
`
`is A, C, T or G or an equivalent thereof.
`
`The oligonucleotide library member can comprise a (complementary base
`
`paired) sequence CTCTCTTCA NNN NNN AGAAGAGC GGG