AU9607698A - Immobilized mismatch binding protein for detection or purification of mutations or polymorphisms - Google Patents

Description

I

AUSTRALIA

Patents Act 1990 VALIGENE

CORPORATION

ORIGINAL

S S *SSd C S S S C C I

CC

C

C.

C

COMPLETE SPE CIFI

CATION

STANDARD

PATENT

Invention~ Title: Immnobilized inismnatch binding protein for detection or purificatio n of mnutations or poly mforphismls The following statement is a full description of this invention including the best method of performing it known to us: W3 Cs C St ICC C *56

-IA-

IMMOBILIZED MISMATCH BINDING PROTEIN FOR DETEICON OR PURIFICATION OF MTATIONS OR POLYMORPHISMS BACKGROUND OF THE INVENTION Field of the Invention The invention in the fields of molecular biology E and medicine relates to a method for detecting mutations involving as little as one base change or a single base 10 addition to, or deletion from, the wild-type DNA sequence, as well as methods for removing mismatch-containing DNA from batches of amplified DNA.

Descrintion of the Background Art Progress in human molecular and medical genetics 15 depends on the efficient and accurate detection of S. mutations and sequence polymorphisms, the vast majority of which results from single base substitutions and small Sadditionsor deletions. Assays capable of detecting the presence of a particular mutation or mutant nucleic acid 20 sequence in a sample are therefore of substantial importance in the prediction and diagnosis of disease, Sforensic medicine, epidemiology and public health. Such S .assays can be used, for example, to detect the presence of a mutant gene in an individual, allowing determination of 25 the probability that the individual will suffer from a genetic disease. The ability to detect a mutation has "taken on increasing importance in early detection of cancer or discovery of susceptibility to cancer with the discovery that discrete mutations in cellular oncogenes can result in activation of that oncogene leading to the transformation of that cell into a cancer cell (Nishimura, S. et al., 'I 1 I1-~ Blochem. j. 243:313-327 ?19871!; 20, CaMC6-.r Res.

4.9:4682-4689 n T-he desire toZ :ncrease t.-e u ic and applicabili tY Of such assays is often frus raced by assay sers:!- as we! 1 a s czom pl exitY and cost. :ernce, as el would be hi;ghly, desir-able o develon More sensitiveaswl as simple and relatively inexpens4ive assays For detectionl of alterations in DNA.

Nucleic acid detection assays can ne based on any of a number of characteristics of a nucleic acid molecule, such as its size, sequence, Susceptibility to digestion by restriction endonucleases, etc. The sensit ivity of such assays may be increased by altering the manner in which detection is reported or signaled to the observer. Thus, for example, assay sensitivity can be increased through the use of dezeccably labeled reagents, wherein the labels may be enzymes (KourilskY U.S. Patent 4,581,3 33), radioisotopes (Falkow et al., U.S. Patent 4,356,535; Berninger, U.S. Patent 4,446,237) fluorescent labels (Albarella ec al., EP 144914) chemical labels (Sheldon

ITI

o ec U.S. Patent 4,582,789; Albarella et al.,

U.S.

Patent 4,563,417), modified bases (Miyoshi e: al.,

EP

119448) and the like.

Most methods devised to attempt to detect genetic alterations consisting of one or a few bases involve hybridization between a standard nucleic acid (DNA or RNA) and a test DNA such that the mutation is revealed as a mispaired or unpaired base in a heteroduplex molecule.

Detection of these mispaired or unpaired bases has been accomplished by a variety of methods. Mismatches have been detected by means of enzymes (R±~aseA, MUtY) which cut one or both strands of the duplex at the site of a mismatch {Myers, et al., Cold Spring Harbor Syrfp. Quanzt. Biol.

51:275-284 (1986); Gibbs, R. et al., Science 236:303-305 35 Lu, A.S. et 1992, Genomics 14:249-255 (1992)). Duplexes. without mismatches are not cut. By Using radioactively labeled nucleic acid fragments to anneal to a test DNA, it is possible to use these enzymes to generate specific size fragments when a mutation is 3 present in the test DNA. The fragments are distinguished from uncut fragments by means of polyacrylamide gel .iectrophoresis. The major problems with these methods are that they rerquire the use of RNA (RNase method) or have the ability to detect only a limited number of mismatches (Muty method).

Mismatch-containing DNA duplexes have also been distinguished from perfectly matched duplexes by means of denaturing gel electrophoresis. In this system, duplexes are run on a polyacrylamide gel in a denaturing gradient under conditions where mismatch-containing DNA denatures more readily than the identical duplex lacking a mismatch, such that the two kinds of duplexes migrate differently.

This method, while sensitive and accurate, is extremely laborious and requires a high level of technical sophistication.

Two other methods of mutation detection depend on the failure to extend or join fragments of DNA when mismatches are present. Both require the use of standard S 20 DNA oligonucleotides that end precisely at the site of the mutation in question such that, when annealed to test DNA, it is the last base of the oligonucleotide which is mismatched. Mismatch detection depends either on the inability of DNA polymerase to extend an oligonucleotide 25 with a mismatched terminal base or the inability of DNA ligase to join two oligonucleotides when there is a mismatch at the joint between them. Fragment length is determined by gel electrophoresis. Presence of longer fragments than the input oligonucleotides indicates that a 30 mismatch, mutation, was not present in the test DNA.

These methods are also somewhat laborious, require that the Iexact location of the mutation be known and are difficult to interpret when the sample DNA is heterozygous for the mutation in question. Therefore, they are not practical for use in screening for polymorphisms.

SA chemical method for cleavage of mismatched DNA (Cotton, R.G. et al., Proc. Natl. Acad. Sci. USA 85:4397-4401 (1988); Cotton, Nuc. Acids Res l ea~Psm~is~fB;ii~k;~fupi~ru~ 17:4223-4233 (1989)) is based on chemical cleavage at a mismatch site in a DNA-DNA heteroduplex, using a number of agents, in particular osmium tetroxide and hydroxylamine.

In this method DNA probes are prepared by restriction enzyme cleavage of DNA of interesc. Plasmid DNA containing thi sequence of interest is hybridized to labeled probe DNA (either end-labeled or internally labeled with 32 p) Hydroxylamine chemically modifies mismatched cytosines; osmium tetroxide modifies mismatched thymines. Piperidine is then used to cleave the DNA at the modified sites, followed by polyacrylamide gel electrophoresis (PAGE) and autoradiography to identify the cleavage products. This method is said to have the advantage of detecting all possible single base pair mismatches because, the method also results in cleavage at a matched base pair in the vicinity of a mismatch.

Publications from Caskey's laboratory (Caskey, C.T. et al., European Patent Publication 333,465 (9/20/89); Grompe, M. et al., Proc. Natl. Acad. Sci. USA 86:588-5892 20 (1989)) disclose a method for localizing a mutation which utilizes PCR-amplified cDNA as a source of template for the mismatch cleavage reaction. This technique was successfully applied in studying ornithine transcarbamoylase (OTCase) deficiency patients to map mutations.

Kung et al., U.S. Patent 4,963,658, discloses detection of single stranded DNA (ssDNA) by binding with a high-affinity ssDNA-binding protein, such as a topoisomerase or a DNA unwinding protein which itself can be bound to a label, such as O-D-galactosidase.

MISMATCH REPAIR SYSTEMS AND MISMATCH BINDING PROTEINS i DNA mismatch repair systems employ a family of proteins including proteins which recognize and bind to mismatch-containing DNA, which are designated mismatch binding proteins (MBPs). For reviews, see Radman, M. et Sal., Annu. Rev. Genet. 20:523-538 (1986); Radman, M. et al., Sci. Amer.. August..1988, pp. 40-46; Modrich, J.

t Biol- Chem. 264:6597-i6oo (1989) The mutS protein was identified as such a component OZt E o:msac ea.-system. See, for examnle La'lu, t Science 245:160-164 (1989); ji7 cr' 0 ta. o Res. 16:7843-73-53 (1988); Su, S.S. LTal, 3i07. Che.

23:82~635(1988); Lahue, R.S. et al., MutaL. Res. 198:37-43 (1988); Dohet, C. et al., Mol. Gen. Genet.

206:181-184 (1987) and Jones, m. eC al. Genetics 115: 605 610 (1987). Analogous proteins are known in other bacterial species including MutS in salmonella typhimurium (Lu, A.L. et al., Genetics 118:593-600 (1988); Haber

L.T.

et al., J. Bacterjol. 170:197-202 (1988); Pang, P.P. et LT. Eacteriol. 163:1007-1015 (1985)) and the hexA protein of Streptococcus pneumoniae (Pr iebe S.D. et al., J. Bacteriol. 170:190-196 (1988), aber et al., supra) Purified MutS protein binds DNA containing rniszaired bases, but does not bind DNA without mismatches or single-stranded DNA. The MutS-DNA interaction. does not result in any'degradation or modification of the DNA. None 20 the above references disclose the Psiiiyo sn MBP or immobilized MBP as part of a mutation detection assay or for purposes of removing mismatched DNA from amplified DNA samples.

0SUMM(ARY OFTH THE INVMTION 225 The Present inventor has conceived of the use of immobilized mismatch binding protein (XP1, for example, the MutS protein of E. coli, for the detection of genetic mutations or genomic polymorphisms, the purification of amplified DNA samples by removing contaminating sequences and sequences containing errors introduced during the amplification processes, and the identification of specific alleles in multi-allelic systems.

The nucleic acid, preferably DNA, being analyzed can be obtained from any source, including blood cells, tumor tissues, cells in culture or any tissue, and can be obtained from any species including humans. The DNA may be labeled by any of a variety of well-kncwn methods, using colorimetric, chemiluminescent or radioactive markers. In fact, it is not necessary to label tes DNA at all For detection of mutations or polymorphism, the assay can be performed with a labeled competing oligonucleotide. For purification of amplified DNA, no label is required. For allele identification, the label must-be in a synthesized single-stranded oligonucleotide probe.

The methods of the present invention depend on the creation of mismatches in the test DNA which are revealed by denaturing the test DNA and allowing it to reanneal. When testing for heterozygosity or for polymorphism within a test DNA sample, the test DNA can simply be self-annealed, resulting in formation of mismatches when the single strands reanneal with a strand I 0- descended from the other parental chromosome. If no S 20 heterozygosity exists, no mismatches will be formed. In Sthis case, the label can be in the primers used for amplification or may be added to the termini of the test DNA if amplification is not required.

A similar procedure and labeling scheme is used 25 to remove sequences containing errors introduced during amplification of DNA or minority sequence species. In these cases, the material which does not bind to the MBP is recovered, and contains only those duplex sequences without l mismatches. These sequences will therefore be greatly S 30 enriched for the majority sequence in the amplified population. When the starting material contains only one sequence, the unbound material will contain those sequences which are identical to the starting material, while those sequences containing errors introduced during amplification will, provided they are relatively rare, have formed I- mismatches which are retained by the immobilized MBP.

STo detect homozygous mutations, it in necessary to anneal the test DNA in the presence of known wild-type 9 *4a5 I sequences. Such sequen~ces can be synthesized artificially' or created during amojifica::zcn adding known wild-type SecieneS tarin maer.2.beforeamifaio.

When annealing is performea in the presence of known w-ildtype seauences, rthe assay will de ec: ot homozvaous and heterozygous mutations.

For allele identification, it is necessary to add a labeled sinigle-stranded probe DNA to the test DNA after amplification. The probe sequence must be identical to the allele of interest such that no mismatches are formed when it anneals to DNA of that allele. The sequence of the probe is selected so that it forms mismatches when annealed to the DNA ofl any other allele. When test DNA is annealed to such a probe (with the test DNA in excess such that the Drobe seqruences all anneal to form a duplex) and expoosed to excess i~mobilized MBP, th'Ie Dresence of unbound label indicates that the allele in question is present in the test DNA sample.

Thus, the present invention is directed to a method fL'or detecting a mutation fram a n-matdsecauence of a taraet roivnucle-oti-_de, ureferably DNA, in a sample, comrxrising: incubating a detectably labeled Polynucleotide or oligonucleotide from the sample with an immobilized mismatch-binding protein under conditions in which mismatch-containing poly-nucleotide molecules bind to the immobilized pro tein; and (b detecting the bindingr of any mismatchcontaining polynucleotide ;from the sample to the mismatch-binding protein, whereby the presence of detectably labeled polynucleotide or oligonucleotide bound to the mismatch-binding protein is *indicative of a mutation in the sequence 'of the target: polynucleotide.

Also provided is a method for detecting a *mutation from a non-mutated sequence of a double stranded target mammalian polynucleotide in a sample, comprising: 9 4* 9** 9..

9. 4 5* 5. 5 *4 o S 4 *5.9 *94 denaturing any double stranded polvnucl eotide in tinIe sample followed by allowi-ng DNA Strands to reannea-1; Un) icubating denatured and reannealei double stranded nucle-:d of' ste'a with,' a mismatchbinding pro:ein imrnonilized on a solid support, either in the Dresence of a det ectably labeled mismatch,-containing oligonucleotide capable of binding to the SP; or (ii) wherein the MBP was preincubated with and allowed to bind a detectably labeled mi smatch- containing oligonucleotide; and detecting the amount of detectably labeled conta ining oligonucleotide bound to the mismatch-binding protein whereby the presence of a mutation in the double stranded mammalian molvnucleotide of the sample results in a decrease in the bindina the detectablv labeled 20 oligonucleotide to the mismatch -binding~ protein., A preferred NMP in the above methods is the coli MutS protein or a functional deia'-ve:e-eoCf.

Preferred solid supports on which the mismatch binding protein is immobilized include, but are not limited to, modified cellulose, polystyrene, polypropylene, polyethylene, dextran, nylon, polyacrvlamide and agarose.

A most preferred solid support is a nitrocellulose membrane.

A pref erred detectable label for the derectably 30 labeled poly- or oligonucleotides in the above methods is biotin.

Also. provided is a method for the removal from an amplified DNA sample of minority sequences and seauences containing sequence errors introduced during the process ot amplification, which method comprises: (a)subjecting the amplified DNA sample to conditions of denaturation followed by reannealing, such that minority sequence or error-containing 9sequences rommismatch-containing DNA duplexes, zbereby generazinlg a mixture containing mismnatched duzlexes; incubating thxe mixture of step with an z..,m,.cni_zea mismazch-ning oracein so that mismatch-containing duplexes bind to the M13; and removing the immobilized MBP to which mismatchcont aining DNA has bound from the amplified DNA thereby removing the sequences containing sequence errors.

in another embodiment is provided a method for identifying a specific allele in a multi-allelic system in a sample of amplified DNA, comprising: mixing a detectably labeled oli gonucleotide probe which is perfectly complementary to the DNTA sequience of e specific allele wizh an excess of amnc1.1fied test DNA under conditions of aenazuration fol-lowed, by annealing such t~hat.

arter denaturing and annealing, everv copy of the probe will be found in a duplex DNAk; nuai~the mixcure o--steD with an. exc-ess of immobilized 3? such that all misrnatchcontaining DNA is retained on the immobilized

MBP;

removing said immobilized MBP to which has bound any mismatch-containing DNA from the amplified test DNA; and detecting the presence of the detectably labeled urobe in the samnie from which the immobilized MBP has been removed, herein, the presence of labeled DNA in the samole indicates that the probe is perfectly comolementarv to an allele' in the test DNA.

AIn the above methods, the immobilized may be in a form which is removable by centrifugation, Or (b) immnobilized onto a column support material wherein the flow through material is devoid of mismat~ch-containing dun'lexes, C. or immobili1zed on a lfilter suooort such the "iltrate is devoid of" mismatch-containing duplexes.

The -aresent. invenltion is also directed. to a kit useful for detecting a mutation from a non-mutated seqruence D of a target- zolvmucleatide searuence in a sample, the- kit being adapted to receive therein one or more concainers.

the kit comprising: a First container containing an immobilizable mismatch-bhinding protein;.

a second container containing a solid support capable of immobilizing the ME~P; and a third container or a plurali ty of containers containing a reagent or reagents canable- of detecting the binding of a detectably labeled mismatch-containing nucleic acid hybzid to the mismatch-binding protein- Also provided is a kit useful.- f Or detectincr a mutation from a non-mutated sequence of a targetpolynucleotide sequence in- a sample, the kit being adapted to receive 'therein one or mor& containers, the kit comprising: a- first container containing a mismatch-binci-ng 4* 4JJ' nrotein immobiized oa a solid support; and a second container or a plurality of containers 25 containing a reagent or reagents capable of detecting the binding of a detectably labeled mismatch-containing nucleic acid hybrid to the mismatch-binding protein.

In the above kits, the MSP is preferably MutS or 30 a functional derivative thereof. The solid sunort is preferably selected from the group consisting of natural cellulose, modified cellulose, most preferably nitrocellulose, or polystyrene, polypropylene, polyethylene, dextran, nylon, polyacrylamide and agarose.

Also provide is a mismatch binding protein, preferably the MutS protein or a functional derivative thereof, immobilized on a solid support, such that said immobilized mismatch binding protein is capable of binding to mismatch-containing polynucleotide molecules.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows the results of a direct assay of mismatches using nitrocellulose-bound MutS. Increasing amounts of biotinylated mismatch-containing DNA (upper 2 lines) or mismatch-free DNA (lower 2 lines) were added to the reaction mixtures.

SFigure 2 shows the results of a competition assay of mismatched duplexes using nitrocellulose-bound MutS protein. Increasing amounts of unlabeled mismatchcontaining 30-mer (upper 2 lines) or mismatch-free (lower 2 lines) were added to 5 ng biotinylated mismatchcontaining 30-mer. The far right column on the figure represents wells which contained no MutS.

Figure 3 shows the results of binding of mismatch containing DNA to E. coli mutS immobilized on.

20 nitrocellulose. Mismatch-containing DNA duplexes (2157 and Bio-Het+) show darker (or visible bands) at concentrations where homoduplex DNA shows lighter (or no) bands.

Figure 4 shows the nucleotide sequences of synthetic oligonucleotides (30mers) prepared with a single i 25 mismatch at position 15 or 16 or with 1 4 unpaired bases between positions 15 and 16. Mismatched or unpaired bases are shown in boldface. Results of studies detecting these mismatches or unpaired bases are shown in Figure Figure 5 shows the results of binding of DNA 30 duplexes containing the indicated mismatches, or unpaired bases, to E. coli mutS immobilized on nitrocellulose.

I

-i :a: Bi 1 I-4- *o DESCRIPTION OF THE PREFERRED EMBODIMENTS The present inventor conceived of a new, bro applicable and relatively simple method for detecting )adly a a

L-,

;:r

:I:

12 single base change in a DNA sequence or several such base changes. This method is based upon the formation of a mismath-conaining heteroduplex when a strand of mutant DA and a "complementary. strand of wild-type

DNA

The presence of the mismatch is detected in a highly specific manner by first allowing the DNA to bind to an immobilized mismatch-binding protein (MBP), such as the MutS protein of E. col The presence of DNA bound to the MBP is then detected in any of a number of ways, depending on the label used and whether the assay is a direct assay or a competitive assay. This method stands in stark contrast to methods of the prior art which employ mismatch cutting nuclease enzymes capable of breaking DNA at or near a mispaired base pair.

The methods described herein provide a mutation/polymorphism detection system having the advantages of simplicity, accuracy, ic) ability to be used without radioactivity, ability to detect all 20 sngle base substitution mutations and addition or deletion .i mutations of 1-4 bases. etion Standard reference works setting forth the general principles of recombinant DNA technology and cell Sbiology, and describing conditions for isolation and handling of nucleic acids, denaturing and annealing nucleic acids, hybridization assays, and the like, include: Sambrook, J. et al., MOLECULAR CLONING: A LABORATORY I- I'NUAL, 2nd Edition, Cold Spring Harbor Press, Cold Spring SHarbor, NY, 1989; Albers, B. er MOLECULAR BIOLOGY

OF

THE CELL, 2nd Ed., Garland Publishing, Inc. New York, NY, 1989; Watson, et al., MOLECULAR BIOLOGY OF THE GENE, Volumes I and II, Benjamin/Cummings Publishing Co., Inc., Menlo Park, CA, 1987; Darnell J.E

CELL

S BIOLOGY, Scientific American Books, Inc.. New York, NY 1986; Lewin, B.M GENES II, John Wiley Sons, New York, NY, 1985, which references are hereby incorporated by reference in their entirety.

I4 MBPs are proteins of around 100 k~a, have been identified in and isolated from both bacteria and higher organisms and selectively bind DNA containing mismatched bases. MBPs have been found in yeast (Valle G ec al., 199,.

YeaS: 7:981-988; Miret J.J. eC- al., 1993, J. Riol Cher.

268:3507-3513), as well as in humans (Stephenson, C. ec al., 1989, J. Biol. Chem. 264:21177-21782; Karran, P et al., 1990, mutat. Res. 236:269-275; Hughes M.J. et al., 1992, J. Bi~ol. Chem. 267:23876..23882; Reenan, A.G. et al., 1993, Genetics 132:963-973; Reerian, A.G. et al., 1993, Genetics 132:975-985). Mismatch binding proteins from Xenopus and from mouse have been cloned by M. Radman and colleagues.

P referred MBP is characterized by its ability 15 t bin DNA DNAduplexes containing mis-paired or unpaired bases, to the significant exclusion -or- single str1anded uolynuc 1eoc ides or perfec!-iv, matched duplexes. In a preferred embodiment, theinataiv -MutS protein from E. coll is used. However, as used ~20 herein, t-he term "mismatch bindina Drtin r 0p"i i ntended to encombass a functional derivative of t a ntact, native protein. By "functional drvtvl meant a "fragment,' "variant, ',analog," or- C."chemical derivative' of the Protein which retains the 25 ability to bind to a mismatch-containing nucleic acid heteroduplex, which permits its utility in accordance with the present invention.

*A "fragment' of a MBP refers to anyv subset 'th **molecule, that is, a shorter peptide. A 'variant', of the 30 protein refers to a molecule substantially similar to *ei.ther the entire protein or a DNA-hybrid-binding fragment2 therof.A vriat o a ism~chbinding protein, for example, o'6 tuz:S, may be Drenared bv r crbna=, Di A methods well-known in the art.

A pre ferred functional derivative of MutS is a homfolo ue of call Muts in Another species, such as the Mut sprotein of, Salmonella, typhiimuzrum(Lu, ec al., qupra, Haber-L.T. aLii, supra; Pang, P.P et al., supra) S S.

S S 0S

S.

S 4 5 *5 5S5555 41 -14 or the hexA protein of ScrepctocCCUS pneuroniae (Priebe S.D. et: al., suDra; Haber e: al., supra). In addition, ogsible eukaryocic homologues of MutS or HexA can also be used, such as z:hose encoded by the homologous sequ.ences identified in human, mouse, frog or hamster DI"A (Shimada, T. et al., J. Biol. Chem. 264:20171 (19B9) Linton, J. et al., .Molec. Cell. Biol. 7:3058-3072 (1989); Fujii, H. et al., J. Biol. Chem. 264:10057 (1989)).

A "chemical derivative" of the MB3P containls additional chemical moieties not normally a part of the protein, including additionallstretches of amino acids as in a fusion protein. covalent -modifications of the peptide are included within the scope of this invention. Such modifications may be introd'4ced into the molecule by reacting targeted amino acid. residues of the-proteil with an organic derivatizing agent. that is capable of reacting with selected side chains or terthinal residues.

in selecting aeprotein as being a'useful M4EP for the methods of the present invention, assays can be performed by one of ordinary skill in the art using conventional methods. Thus, for example, in~evaluating a sample for the presence -of a MBP useful in the present invention, one can perform a mismatch binding assay, such as that described for MutS by Jiriofly et al. (which reference is hereby incorporated by reference in its entirety) Preferably, a filter binding assay is used.

-To prepare the oligonucledtide heterodu~plex, an oligonucleotide, preferably of' abou t,16 bases is labeled wih 32 p using a kinase reaction and 32 -TPuigakns 30 'such as T 4 -polynucleotide kinase. The oligonucleotide (vihich,can -be stored at -2 0 is then Ia nnealed with-a complementary oligonucleotide having a a sigl bse pair mismatch under standard conditions. The annealed 16, base pair heteroduplex is mixed with an excess.

of the 'Proteinl being, tested And kept on ice for 30 minutes.

the. iixture is. then applied to a nitrocellulose filter w4hich has been prewetted in the assay buffer. Gentle suction is applied for several seconds, and the filter is d

A

4 washed extensively with ice-cold assay buffer. The filter is then dried in air, suspended in scintillation fluid and counted. By virtue of the protein sticking to the filter, any counts on the filter can be attributed to binding to c-te cutative MBP. In the absence of such a protein, the labeled oligonucleotide heteroduplex will pass through the filter. Thus, by using such a simple assay, one can easily detect and select a MBP useful in the methods of the present invention.

As used in the present invention, the MBP is immobilized to a solid support or carrier. By "solid support" or "carrier" is intended any support capable of binding a protein. Well-known supports or carriers include natural cellulose, modified cellulose such as nitrocellulose, polystyrene, polypropylene, polyethylene, dextran, nylon, polyacrylamide, and agarose or Sepharose®.

Also useful are magnetic beads. The support material may have virtually any possible structural configuration so long as the immobilized MBP is capable of binding to the 20 target nucleic acid molecule. Thus, the support configu- 'ration can include microparticles, beads, porous and impermeable strips and membranes, the interior surface of a ".oo reaction vessel such as test tubes and microtiter plates, S: and the like. Preferred supports include nitrocellulose 25 disks or strips. Those skilled in the art will know many other suitable carriers for binding the MBP or will be able to ascertain these by routine experimentation.

Most preferred is a solid support to which the S:o. MBP is attached or fixed by covalent or noncovalent bonds.

'30 Preferably, noncovalent attachment is by adsorption using methods that provide for a suitably stable and strong attachment. The MBP is immobilized using methods wellknown in the art appropriate to the particular solid I: support, providing that the ability of the MBP to bind mismatch-containing DNA is not destroyed.

The immobilized MBP is then easily used to detect heterozygosity (or polymorphism) as well as single base mutations, or to isolate mismatch-containing DNA from a mixture, or to rid a mixture of mismatch-containing

DNA.

In one embodiment, the surface of polystyrene or ther plastic multiwell plates serves as the solid support.

S

n another embodiment, a solid support to which the MB? is Sbound is affixed to the bottom of wells of multiwell Splates.

In a preferred embodiment, the immobilization and DNA binding can be performed in a 96 well blotting apparatus and the resulting sheet of nitrocellulose (or other support) paper can be removed to evaluate reactions.

For example, color development on the nitrocellulose can be used to evaluate binding based on an enzyme as part of the detection system and a chromogenic or chemiluminescent i substrate for the enzyme serving as the precursor of the color reactions.

Following attachment of the MBP to the support, Sthe support is treated ("blocked") to prevent further 20 binding of proteins or nucleic acids, using methods and o reagents well-known in the art.

The immobilized MBP is contacted with and allowed I to.bind (to saturation) small oligonucleotide heteroduplex :molecules. The oligonucleotides preferably have about 25 base pairs. For testing, a DNA duplex containing a mismatch which is well recognized bound) by the MBP is used.

PREPARATION OF OLIGONUCLEOTIDES

CONTAINING

OR LACKING

MISMATCHES

S: 30 Such oligonucleotides are prepared using a nucleotide modified at the 5' end with a detectable label such that they can be quantitatively detected by appropriate detection methods, preferably spectrophotometry or chemiluminescence. In a preferred embodiment, the o ligonucleotide is biotin-modified, and is detectable using a detection system based on avidin or streptavidin which binds with high affinity to biotin. The streptavidin can I K i be conjugated to an enzyne, the presence of which is detected using a chromogenric substrate and measuring the color developed.

Examples of useful enzymes in the methods of the present invention are horseradish peroxidase, alkaline phosphatase, glucose-6-phosphate dehydrogenase, maiate dehydrogenase, staphylococcal nuclease, delta-V-steroid isomerase, yeast alcohol dehydrogenase, a-glycerophosphate dehydrogenase, triose phosphate isomerase, asparaginase, glucose oxidase, 0-galacto-sidase, ribonuclease, urease, catalase, glucoamylase and acetylcholinesterase.

The detectable label may also be a radioisotope which can be detected by such means as the use of a gamma counter or a scintillation counter or by autoradiography.

The detectable label may also be a fluorescent compound. When the fluorescently labeled molecule is exposed to light of the proper wave length, its presence can then be detected due to fluorescence using microscopy or fluorometry. Among the most commonly used fluorescent S 20 labelling compounds are fluorescein isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, gphthaldehyde and fluorescamine.

The detectable label may be a fluorescence emitting metal such as 15 Eu, or others of the lanthanide series. These metals can be attached to the oligonucleotide using metal chelating groups such as diethylenetriaminepentaacetic acid or ethylenediaminetetraacetic acid.

We The detectable label may be a chemiluminescent compound. The presence of the chemiluminescent-tagged molecule is then determined by detecting the presence of Sluminescence that arises during the course of a chemical S reaction. Examples of particularly useful chemiluminescent labeling compounds are luminol, isoluminol, theromatic acridinium ester, imidazoie, acridinium salt and oxalate ester.

Likewise, a bioluminescent compound may be used to label the oligonucleotide. Bioluminescence is a type of q chemiluminescence found in biological systems in which a ::atalytic Protei.n increases the efficiency of the chemiluminescent reaction. The presence of a biol~uminescent protein is determined by detecting the presence o 1,minescence. Important bioluminescent compounds for purposes of labeling are luciferin, luciferase and aequorin.

A MEP bound to the DNA-DNA hybrid can be detected either directly or indirectly. For direct detection, the poly- or oligonucleotide duplex is detectably labeled using labels as discussed herein.

For indirect detection, the assay utilizes competition of binding to the MEP of the test DNA with an already bound or a contemporaneously exposed mismatchcontaining duplex. Thus, a labeled mismatch -containing oligonucleotide is pre-bound to the IMBP or is incubated together with the MP and test DNA. The more mismatchcontaining DNA in the cest sample, the less binding of the labeled oligonucleotjde to the ?.BP will occur.

20 The test sample to be assayed can be in any medim ofinterest, and will generally be a sample of medialveterinary, environmental. nutritional, or industrial significance. Human and animal snecimens and body fluids particularly can be assayed by the present 25 method, providing that they contain cells from which nucleic acids can be. prepared. Preferred sources include blood, sperm, other tissue, milk, urine, cerebrospinal *.fluid, sputum, fecal matter, lung aspirates, throat swabs, agenital swabs and exudate, rectal swabs, and nasopharyngeal aspirates.

DETCTIN F ~OZ GOSITY OR POLYORO.'TSM To detect heterozygosity ot polymor-Uism inDNA from a diloid organism, test DNA is preferably prepared by denaturing and annealing PCR-amplified DNA from a diploid organism. The test DNA is prepared with labeled primers, annealed and added to a well or other reaction vessel which contains iummbilized NEP already bound to-mismatched 19 oigonucleocides. Alternatively, test DNA can be mixed with mismatched oligonucleotides and the mixture added to a well or other vessel containing immobilized

MBP

A spectrophotometric reading is made at the wavelength appropriate for quantitative detection of the test DNA. After an incubation period suitable to allow either binding of the test DNA to the immobilized

MBP

or displacement of the mismatched oligonucleotide from the immobilized MBP, the DNA solution is removed, the well washed, and a spectrophotometric reading made at the wavelength appropriate for quantitative detection of the bound mismatched oligonucleotide.

The ratio of the reading for test DNA to the reading for the mismatched oligonucleotide will be vastly different for mismatch-containing and mismatch-free test DNA over a wide range of DNA concentrations. Standard curves are prepared using known quantities of DNA to allow characterization of test DNA as homozygous or heterozygous without having to quantitate the test DNA prior to the assay. Thus, a single DNA sample is sufficient to determine heterozygosity and a single 96 well microplate will allow the testing of at least about 80 different DNA samples.

DETECTION OF HOMOZYGOUS MUTATIONS 25 To detect homozygous mutations, known homozygous wild-type DNA must be combined with the test DNA sample before denaturing and annealing. Only test DNA containing 'I a mutation (homozygous) will form mismatch-containing

DNA

that can compete with the mismatched oligonucleotide for binding to the immobilized MBP.

The method of the present invention can be used with only the mismatched oligonucleotide labeled or (b) only the test DNA labeled. However, both of these methods require that the nucleic acid concentration be determined and that the test be performed at several different test DNA concentrations.

When the mismatched oligonucleotides are labeled, the test is based on competition with several different concentrations of test DNA and comparison of the resulting 7* 77777mm ti curve (with concentration expressed as moles of duplex molecules) with standard curves for mismatched and nonmismatched standards.

When the test DNA is labeled, the test involves S measuring the extent of bininng to the MB? of several concentrations below saturation and comparison of the resulting curve with standard curves for mismatched and non-mismatched standards.

PURIFICATION OF AMPLIFIED DNA SAMPLES One of the most revolutionary and widely used technologies currently employed in modern molecular biology is the process of polymerase chain reaction (PCR), which amolifies DNA sequences from starting amounts so minute as to be nearly undetectable. For reviews of PCR, see: Mullis, 1986, Cold Spring Harbor Symp. Quant. Biol.

51:263-273; Saiki, R.K. et al., 1985, Sio/Techlnolcgy 3:1008-1012; and Mullis, K.B. et al., 1987, Meth. Enzymol.

155:335-350. In addition, because PCR can amplify specific sequences, it allows the purification of specific 20 sequences, basically in a single step, from genomic

DNA.

PCR is an essential component of virtually all studies of the human genome, is a central component of gene identification and cloning, is increasingly used in the diagnosis of genetic and infectious -diseases and is widely used in forensics.

However, for some applications, in particular, gene cloning and mutation detection, PCR suffers from an inherent tendency of the polymerases to make mistakes by inserting incorrect, non-complementary bases during synthesis. Although the fidelity of most replicative polymerases in vivo is such that they insert only one incorrect base for every 1010 bases replicated, polymerases used in PCR can have error rates as high as one incorrect base for every 104 bases replicated. This high an error rate can mean that a significant fraction of the amplified molecules will not be identical in sequence with the starting material.

The present methods are useful for purifying amplified DNA samples using an immobilized MBP to remove

LI

ij S r 1 *e 'ater ones, i4~ U -21 minority sequences and molecules containing sequence alterations introduced by the amplification process. For example, if a DNA segment is amplified through 20 rounds of replication (a common amount of amplification), a significant fraction of the final molecules may contain one or more incorrect bases. In cloning eperiments, this greatly increases the risk of cloning a nucleotide sequence different from the starting sequence.

In mutation detection assays involving denaturation and annealing of a PCR-amplified sample, incorrect bases inserted during PCR may be scored as if they were mutations in the original sample. Thus, for accurate mutation detection it is necessary to eliminate all DNA molecules with sequence alterations introduced by PCR copy errors. The method described here accomplishes this purification in a simple and straight-forward manner.

Immobilized MBP can be used to purify amplified DNA samples. MBP is immobilized by binding to solid phase supports, preferably nitrocellulose filters, sepharose beads or magnetic beads. The filters or beads are treated, if necessary, to prevent the binding of double-stranded .DNA. The amplified DNA sample is denatured, by heating, and allowed to reanneal. Given the random nature of PCR .e mistakes, virtually all incorrect bases will be found in S 25 mismatched base pairs after annealing.

The immobilized MBP is added to the sample and the solution mixed by gentle shaking. The immobilized MBP, and any bound mismatch-containing DNA, is removed, for ,o example, by removing the filter, by allowing the beads to S" 30 settle out of solution or by removing the beads magnetically, depending on the nature of the solid support Sused. This leaves behind precisely matched DNA duplexes.

SIn addition to purifying amplified DNA samples by removing molecules containing errors introduced during t 35 amplification, purification using an immobilized MBP is used to enrich for majority sequences when samples of diverged, repeated DNA sequences, such as of immunoglobulin genes, are being examined...

STo remove completely a minority species from a sample amplified from a mixed population of DNA (with 22 respect to sequence), it may be necessary to perform more than one round of purification as described herein and possibly more than one round of amplification.

Note that the present method can be used to S purify sequences from both homozygous and heterozygous amplified sequences, since half of the parental sequences in a heterozygous sample will anneal to the complementary strand of the same parental heritage and thus form a Smolecule without mismatches. In other words, when the starting material is heterozygous, half of the annealed molecules will be removed from the sample because they contain a mismatch due to differences in the starting sequences. However, half the annealed molecules will not contain such mismatches and so will be removed from the sample only if they contain mismatches which were created as a result of errors during amplification. In any event, mutation detection assays will require a second round of denaturing and annealing.

ALLELE IDENTIFICATION IN MULTI-ALLELIC

SYSTEMS

20 As more alleles of disease-causing genes are identified, and in the quest to develop a polymorphism map of the human genome, it is becoming increasingly important to be able to identify particular alleles of a given gene.

I Immobilized MBPs provide a simple and straightforward means of allele identification.

Unique, labeled oligonucleotide probes are prepared for each allele of a given.gene, such that the S. robe is perfectly complementary to only one allele, i.e., the probe will form one or more mismatches when paired with S 30 an incorrect allele. The probe is mixed with an excess of Si amplified test DNA such that, after denaturing and S f annealing, every copy of the probe will be found in duplex.

The process is repeated with probes for every allele in question. The annealed DNA mixture is then either: I 35 mixed with immobilized MBP on a support that can be removed from the suspension by centrifugation; passed through a micro-column of immobilized MBP

I:

on an appropriate column support; or R -m, 23 passed through a filter support containing immobilized MBP.

In any case, the immobilized MBP must be in excess such that all mismatch-containing DNA is retained. The S supernatant, column flow through or filtrate is analyzed for the presence of label. Only in those cases where the probe is- perfectly complementary to an allele in the test DNA will label be detected.

In order to be certain that no single-stranded probe sequences are present, it may be necessary, or at least desirable, to include some single-stranded

DNA

binding component on the support for the immobilized MBP.

This system works equally well for homozygous or heterozygous conditions.

KITS

The present invention is also directed to a kit or reagent system useful for practicing the methods described herein. Such a kit will contain a reagent combination comprising the essential elements required to 20 conduct an assay according to the methods disclosed herein.

The reagent system is presented in a commercially packaged form, as a composition or admixture where the compatibility of the reagents will allow, in a test device configuration, or more typically as a test kit, a packaged 25 combination of one or more containers, devices, or the like holding the necessary reagents, and usually including written instructions for the performance of assays. The kit of the present invention may include any configurations and compositions for performing the various assay formats S 30 described herein.

In all cases, the reagent system will comprise an immobilizable or immobilized MBP or functional derivative, preferably mutS, and additional reagents useful in carrying out the assay. The kit may optionally contain labeled mismatch-containing oligonucleotides. For detecting a particular mutation, a kit may also contain labelled primers for carrying out a PCR. A kit according Sto the present invention can additionally include ancillary 24 chemicals such as the components of the solution in which binding of duplexes to the immobilized MBP takes place.

Having now generally described the invention, the same will be more readily understood through reference to the following examples which are provided by way of illustration, and are not intended to be limiting of the present invention, unless specified.

EXAMPLE I BINDING OF MISMATCH-CONTAINING DNA BY IMMOBILIZED MISMATCH BINDING PROTEIN A. Materials and Methods 1. Preparation of immobilized MBP A nitrocellulose sheet (045Am, Schleicher Schull) was wet with reaction buffer (20mM Tris pH 7.6, 0.01 mM EDTA, 5 mM MgC1 2 0.1 mM DTT) and placed in a dot blot apparatus (Bio-Rad).

Purified MBP, E. coli MutS, at a concentration of 0.5 gg/10 il reaction buffer was spotted on the nitrocellulose paper in each well. The wells were 20 incubated at room temperature, and the remaining liquid was pulled through with vacuum. Each well was washed twice with 100 ll reaction buffer by adding the solution to the well and then pouring it out. After the second wash, the remaining solution was pulled through by vacuum.

2. Blocking The nitrocellulose filter was blocked with bovine serum albumin (BSA) to prevent the binding of other proteins or nucleic acids. Reaction buffer (200 4l) M, "containing 1% BSA was added to each well. After 1 hour at room temperature, the solution was poured out and Seach well washed with 2 x 100il reaction buffer by adding the solution to the well and then pouring it out. After the second wash, the remaining liquid was pulled through with vacuum.

I;~asll a~ Ti--,ri

S.

25 3. Oligonucleotides The sequence of the oligonuclectides used in these studies was taken from the 30 base region surrounding the site of the sickle cell mutation in the human a-globin gene. The mismatch is at the site of the sickle cell mutation, although the mutant sequence used to form the mismatch is not the sickle cell mutation (the actual sickle cell mutation is an A:T T:A transversion). Biotinylated oligonucleotides were biotinylated on the 5' end of the mutant strand. Biotinylation is accomplished during synthesis by adding a biot'.n-modified nucleotide to the end of the oligonucleotides.

G:T Mismatch: Mutant GCACCTGACT CCTGGGGAGA AGTCTGCCGT [SEQ ID NO:1] Wild-type CGTGGACTGA GGACTCCTCT TCAGACGGCA (SEQ ID NO:21 No Mismatch: Mutant GCACCTGACT CCTGGGGAGA AGTCTGCCGT [SEQ ID NO:1] Mutant CGTGGACTGA GGACCCCTCT TCAGACGGCA [SEQ ID NO:31 20 4. Binding DNA Biotinylated oligonucleotides, in 20 l reaction buffer containing 1% BSA, were added to each well. After 30 minutes at room temperature, remaining liquid was poured out. Each well was washed with 5 x 100 l reaction buffer by adding the solution to the well and then pouring it'out. After the fifth wash, the remaining solution was pulled through by vacuum.

5. Binding Streptavidin-conjugated Horse Radish Peroxidase (HRP) The presence of biotin was detected by its binding of Streptavidin. A 100 Al volume of Streptavidin-conjugated HRP (Pierce Chemicals) at a concentration of 50 mg/ml in reaction buffer BSA was added to each well. After 2 hours at room 35 temperature,, the solution was poured out and each well washed with 5 x 100 pl reaction buffer by adding the -solution and then pouring it out. After the fifth wash, the remaining solution was pulled through -with vacuumi.

J i i~-i ;1 d iti

I:I

9 d ;1 ii s :U f

_L

~I

-ij l ;a .i; i -s '4 26

B

99*

B

9. 9 9*99

B..

'ID

6. Enhanced ChemiLuminescence" (ECL) Development The nitrocellulose sheet was removed from dot blot apparatus and washed 3 times in a petri dish with 10 ml reaction buffer. Five ml ECL development solution (Amersham) was poured over the nitrocellulose. The substrate for HRP in this reagent is a chemiluminescent compound. After 1 minute, the solution was removed. The nitrocellulose was blotted dry and placed between 2 clear plastic sheets. The nitrocellulose thus protected was exposed to X-ray film in the dark for varying periods of time. In the experiments reported here, the exposure time was 1 minute.

7. Competition In competition studies, the DNA binding was as described above, except that a constant amount of biotinylated mismatch-containing oligonucleotide was mixed with varying amounts of unlabeled DNA, with or without a mismatch, and added to the wells.

20 B. RESULTS 1. Specific binding of mismatch-containing

DNA

by immobilized MBP Figure 1 shows the results (in duplicate) of adding increasing amounts of biotinylated mismatchcontaining DNA (upper 2 lines) or mismatch-free DNA (lower 2 lines).

The immobilized MBP can detect as little as 0.2 ng of mismatch-containing 30-mer, whereas no detectable binding of mismatch-free 30-mer was observed even with 200 ng of DNA. (The lines around the lower spots were artifacts of incomplete washing.) S 2. Competition Assay i Figure 2 shows the results (in duplicate) of .adding increasing amounts of unlabeled, mismatch- S- containing 30-mer (upper 2 lines) or mismatch-free mer (lower 2 ines) to 5 ng biotinylated mismatchcontaining 30-mer:. Although competition was clearly i

I

I:.

i r :-i .j t d it

I

d: g=

I~

d~ sC~aa~a~Par~s~paRp~sr~- I i:1: i

II

i b ii

R

r!

P

27 visible with so ng of mismatch-containing DNA, the mismatch-free DNA did not compete until 500 ng, if at all. Tefar right column on the figure contains no MB P.

4

C..

*4 C C

C

C.

*J

4tCC *4

V

S The results indicate that, at least with the .J3iners used above, immobilized MBP discriminates between mismatch-containing and perfectly paired DNA with an efficiency of at least three orders of magnitude. Similar results have been obtained using 54mers with a sequence derived from the V3 loop of HIV. Therefore, even if the discrimination decreases as the amount of perfectly paired duplex increases, the discrimination efficiency when using 300iners, considered to be the maximum useful length for polymorphism studies of the human genome, should be on the order of a factor of 100.

EXAMPLE 11 ETEROZYGOTE DETECTION IN HUMAN GENOMIC DNA The methods described above were used to 20 detect heterozygosity at a specific-location-in human' genomic DNA. PCR amplification was performed on a .portion of exon 3 of the human glucokinase gene extending from codon 98 (encoding glutamine),to the stop codon (Soffel et al., Proc. Natl. Acad. Sci. USA 89:7698-7702 (1992)). The wild-type double stranded sequence of 100 bases (SEQ ID NO:4 and SEQ ID corresponding-to exon 3,human glucokinase is as follows: j 30 3' gcactaacttcagggtgatgctggtgaaggtgggagaagg cgtgattgaagtcccactacgaccact tccaccctcttcc tgaggaggjgg~gtggagcgtgaagaccaaacaccagatg .actcctcccc~acctcgcacttctggtttgtggtctac tactccatccccgagga -cgcc 3' (SEQ ID NO:4) atga 9gtaggggctcctgcgg 5' (SEQ ID In .the. heterozygous DWA (see below) The t-of the underlined -CAG'6odonl' was. mutated- to T_ Th NA ;11 -i 28a sequences tested were PCR amplified from genomic DNA obtained from: a known heterozygote at that position, a known homozygote at that position and a presumed homozygoce at that position.

S Test DNAs were denatured, by heating, allowed to reanneal and tested for the presence of mismatches heterozygotes) by testing their binding in an immobilized mismatch binding protein assay according to the present invention, utilizing E. coli MutS.

A. Materials and Methods: 1. PCR Amplification The following templates were used: 2157 heterozygote at glucokinase gene exon 3 DGK-101 Human genomic DNA (homozygote at glucokinase gene exon 3) Human genomic DNA, male .(presumed homozygote at glucokinase gene exon 3) designated Sigma DNA (commercially obtained from Sigma Chemical Co.) 20 The primers (obtained from Operon) were HPLC purified and had the following sequences, .corresponding to the 5' termini of the two DNA strands SEQ ID NO:4 and SEQ ID Primer 5' (biotin).-GCACTAACTTCAGGGTGATG Primer 5' -GCGTCCTCGGGGATGGAGTA PCR Primer- #1 contained biotin bound at the S* end to allow detection in the ECL detection system described below. Primer #2 was radioactively labeled with a 5-' 32 P-phosphate to allow quantitation of the 30 different amplification products after removal of unused primers. Primer #2 was 2 p labeled using a -kinase reaction prior to use in amplification. The kinase reaction mix contained: 70mM TrisHC, pH 7.6; mM mgC1 2 SmM DTT; 20pCi 32P-ATP; 30 units T4 polynucleotide kinase; and 500ng DNA (primer The kinase reaction was performed in a 20pl reaction volume at 37*C for. 30 min. Kinase was inactivated by heating at 70oC for 10 min. DNA was stored at r

I

j ii

I

.j -23 The pCR reaction included: IOMM1 Tris HCI pH1 8.;So MM KC; I .5MM MIC 2 0 -001 gelatin O.05Ml4 dATP; 0.05MM dTrP; 0.05mm dCGTp; 0.05i1 dCTP; 0.19H primer 0.075pM primer 0O.O25pM, 32 crinier f2; 0 0ng template DNA; and 2.5 units AmpliTaq

DNA

Polymerase (Perkin-El~mer) -Reaction volume was lO0jpl.

Amplification was carried out for 30 cycles in a Perkin-.Elmer thermocycler by denaturing at 90 0 C for

I

min., annealing at SS 0 c for 1 min. and extension at 72'C 2 Min. Unused primers were removed by centrifugal dialysis using Centricon-3 0 microconcentrators (Amicon) according to.

manufacturer' a protocol. PCR -products were quantitated by running equal amounits (measured as cpm of radioactivity) on non-denaturing 8kr polyacrylanide gels, staining with ethidium bromide and comparing hento standard

DNA.

2. immobilized Msac 2nigProcej- U Assays: Elmer thermocycler) accordin tth olow4~ schedule: 1000C for- mmi; 50*C for 1 hour; 73OC. for 4 min; 500C for 30 min, foloebycligc ol temperatureloe ycoln oro Anitrocellulose sheet (0.45Mr, Schleicher and Schuell.,. BA85) was wet by f loating. in .reaction buf fer Tris Md1, pH 7.6; 5MM MgCI 2 0.1MM DTT; 0.01LmM EDTA) and placed in a slot blot, apparatus (Ioefer 9 AI ~Scientific Instruments) over 3 'sheets of'botn 130- paper (Schleicher and Schuell GBOQ2). 10 0gl oz reaction buffer was added to each well After 5 min.

agebb at, room temperature, any'remaining buffer was pulled thog by vacuum. M tS (50O ng inl 204.l reaction buf fer) was added to each well.; The saMe volume of reaction- buffe -Was added to

TM

40 MutS".wells. The apparatus, was ket troom tmeaue o 0m b~efr poeedin to the next step.

Thnfitrc lo, wa bocked b adding:'200AL: of HRP-free B8A toeach well. After 1 hr at room '4 r 30 cemperature, any remaining solution was pulled through by vacuum.

DNA preparations were added in 20l reaction buffer containing 3% HRP-free BSA. After 30 min at room temperature, wells were washed 5 times with 100gl reaction buffer by adding the solution to the wells and decanting it.

The presence of biotin-labeled DNA bound on the nitrocellulose sheet was detected by visualizing the binding of streptavidin to biotin. 100 Al streptavidin-HRP in reaction buffer containing 3% HRPfree BSA was added to each well. After 20 min at room temperature, any remaining solution was decanted.

Wells were washed 5 times with 10041 reaction buffer by adding solution to each well and then decanting.

Any solution remaining after the fifth wash was removed by vacuum.

3. Enhanced Chemiluminescence

(ECL)

development The nitrocellulose sheet was removed from the apparatus and washed 4 times for 1 min with 50 ml Sreaction buffer in a small tray. Nitrocellulose was A blotted dry and immersed in 10 ml of ECL development :solution (Amersham).. After 1 min, the nitrocellulose S 25 sheet was removed, blotted dry and placed between two clear plastic sheets. Thenitrocellulose thus protected was exposed to X-ray film in the dark for sec.

B. B. RESULTS 30 The results are shown in Figure 3. Bio-Het Srefers to a synthetic 30mer duplex of SEQ ID NO:1 and SEQ ID NO:3. This duplex contains a G:T mismatch at S:position 15. Bio-Homo refers to a synthetic duplex of SEQ ID NO:1 and SEQ ID NO:2, which lacks the mismatch at position 15. No binding of Bio-Homo was detected even at 1ang of DNA whereas Bio-Het was detected at amounts as low as o.lng. No binding was I .observed in the absence of Muts no MutS" columns) V-1-; ,1.

31 indicating that all DNA binding observed is MutS dependent.

Binding of heterozygote nucleic acid (2157) to mutS was clearly visible at 0.6ng. Homozygote binding (independent of the source of the DNA) was faintly detectable at 1.25ng and was clearly detected at Thus, heterozygote DNA was detected at least 2-4 times better than homozygote DNA in this assay.

The binding of homozygote DNA at higher concentrations was considered to be the result of errors introduced during amplification by the Taq polymerase. Such a high error rate in nucleotide incorporation by this polymerase is a well-known phenomenon. Thus, such inappropriate binding still represented mismatch binding by the immobilized MutS.

The use of immobilized mismatch binding protein assays for mutation, heterozygosity or polymorphism detection are limited only by the ability to orovide substrates free of random mismatches, such 20 as those produced by polymerase errors during PCR amplification.

EXAMPLE III DETECTION OF SPECIFIC MISMATCHES AND UNPAIRED BASES BY IMMOBILIZED MISMATCH BINDING PROTEIN ,i :11 ,'1 i ro r sc r r r.ir rr r.

r ik

I

L .35 Studies were performed to detect single base mismatches and one to four unpaired bases in heteroduplex DNA. These studies utilized synthetic oligonucleotides (30mers, which include SEQ ID NO:1, SEQ ID NO:2 and SEQ ID NO:3) prepared with a single mismatch at position'15 or 16 or with 1 4 unpaired bases between positions 15 and 16. The sequences of these oligonucleotides are shown in Figure 4. As described above, these sequences were taken from, or are closely related to, the human 0-globin gene in the region of the mutation responsible for sickle-cell anemia. The oligonucleotides were prepared with a biotin label (to allow detection by the ECL detection Al 4i i

I

r'

S.

S

a. 4. 6 32 system) and were annealed to unlabeled oligonucleotides such that only one strand of each duplex was labeled. The duplexes were used in a mismatch detection assay using immobilized mismatch binding protein (MutS), essentially as described above in the previous Examples.

Oligonucleotides were diluted in TNE buffer Tris HC1, pH 8.0; 0.01M NaCl, ImM EDTA). Biotin labeled oligonucleotides were diluted to 10 mg/pl and unlabeled oligonucleotides to 100 ng/pl. Equal volumes of diluted oligonucleotides were mixed and annealed at 70 0 C for 10 min, room temperature for min, followed by quenching on ice and were stored at The unlabeled oligonucleotides were in excess to assure that all biotin-labeled strands were in duplexes.

The results are presented in Figure 5. All mismatches except C:C and G:A were detected in these assays. T:C and C:A mismatches were not teseted in 20 this experiment. Not all detected mismatches are detected equally well. The order of detection sensitivity appeared to be G:T G:G C:T A:C T:T A:A A:G C:C.

The fact that A:G mismatches were better detected than G:A mismatches suggested that the sequence of the individual strands may influence the extent of mismatch detection at least in relatively short oligonucleotides such as those used here.

However, G:T and T:G mismatches were equally welldetected, suggesting that well-detected mismatches are detected independent of strain orientation.

Reteroduplexes with 1 4 unpaired bases were also detected. A heteroduplex with two unpaired bases was more easily detected than a heteroduplex with one unpaired base. Heteroduplexes with three unpaired bases were less well detected than any of the i. detectable mismatches, and heteroduplexes with four unpaired bases were even less well detected.

hese results indicate that immobilize misniatch binding protein assays will detect all r *I C

I

r:: a I i 33 mutations due to a single base change. Thus, although C:C went undetected, the corresponding G:G mismatch, which would of necessity occur in any mutant:wild type pairing that had a C:C mismatch, was easily detected.

In addition, the detection system can easily detect mutations arising from the addition of I 3 bases.

The references cited above are all incorporated by reference herein, whether specifically incorporated or not.

1 0 Having now fully described this invention, it will be appreciated by those skilled in the art that the same can be performed within a wide range of equivalent parameters, concentrations, and conditions without departing from the spirit and scope of the invention and without undue experimentation.

While this invention has been described in S: connection with specific embodiments thereof, it will be understood that it is capable of further modifications. This application is intended to cover 20 any variations, uses, or adaptations of the inventions following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as 25 may be applied to the essential features hereinbefore .set forth as follows in the scope of the appended claims.

Z 9 i

Claims (3)

1. A mismatch-binding protein immobilized on a solid support, such that said immobilized mismatch binding protein is capable of binding to mismatch-containing polynucleotide molecules.

2. An immobilized mismatch-binding protein according to claim 1 which is the MutS protein or a functional derivative thereof.

3. An immobilized mismatch-binding protein according to claim 1 or claim 2 substantially as hereinbefore described with reference to any one of the Examples or Figures. Dated this fourth day of December 1998 VALIGENE CORPORATION SPatent Attorneys for the Applicant: o F B RICE CO see. e 0 S 0 a i s i T