CA2203627C

CA2203627C - Rapid detection and identification of nucleic acid variants and pathogens

Info

Publication number: CA2203627C
Application number: CA 2203627
Authority: CA
Inventors: James E. Dahlberg; Victor I. Lyamichev; Mary Ann D. Brow; Mary C. Oldenburg; Laura M. Heisler; Lance Fors; David Michael Olive
Original assignee: Third Wave Technologies Inc
Current assignee: Third Wave Technologies Inc
Priority date: 1994-11-09
Filing date: 1995-11-09
Publication date: 2000-06-06
Anticipated expiration: 2015-11-09
Also published as: JPH10509322A; AU4234796A; EP0788557A1; EP0788557A4; WO1996015267A1; CA2203627A1; AU4444699A

Abstract

The present invention relates to means for cleaving a nucleic acid cleavage structure in a site-specific manner. Enzymes, including 5' nucleases and 3' exonucleases, are used to screen for known and unknown mutations, including single base changes, in nucleic acids. Methods are provided which allow for the identification of genetic mutations in human gene sequences, including the human p53 gene, in a sample. Methods are provided which allow for the detection and identification of bacterial and viral pathogens and species in a sample.

Description

DEMANDES OU BREVETS VOLUMINEUX
LA PRESENTE PARTIE DE CETTE DEMANDE OU CE BREVET
COMPREND PLUS D'UN TOME.
CECI EST LE TOME ~ DE ~2 NOTE: Pour les tomes additionels, veuiilez contacter le Bureau canadien des brevets ~z~ 3~~~7 THIS SECT10N OF THE APPLlCATIONIPATENT CONTAINS MORE
THAN ONE VOLUME
THIS IS VOLUME OF -NOTE: For additional volumes-phase contact the Canadian Patent Ofific~

RAPID DETECTION AND IDENTIFICATION OF
NUCLEIC ACID VARIANTS AND PATHOGENS
FIELD OF THE INVENTION
The present invention relates to methods and compositions for treating nucleic acid, and in particular, methods and compositions for detection and characterization of nucleic acid sequences and sequence changes.
BACKGROUND OF THE INVENTION
The detection and characterization of specific nucleic acid sequences and sequence changes have been utilized to detect the presence of viral or bacterial nucleic acid sequences indicative of an infection, the presence of variants or alleles of mammalian genes associated with disease and cancers, and the identification of the source of nucleic acids found in forensic samples, as well as in paternity determinations.
I S Various methods are known in the art which may be used to detect and characterize specific -nucleic acid sequences and sequence changes. Nonetheless, as nucleic acid sequence data of the human genome, as well as the genomes of pathogenic organisms accumulates, the demand for fast, reliable, cost-effective and user-friendly tests for specific sequences continues to grow. Importantly, these tests must be able to create a detectable signal from a very low copy number of the sequence of interest. The following discussion examines three levels of nucleic acid detection currently in use: I. Signal Amplification Technology for detection of rare sequences; II. Direct Detection Technology for detection of higher copy number sequences; and III. Detection of Unknown Sequence Changes for rapid screening of sequence changes anywhere within a defined DNA fragment, I. Signal Amplification Technology Methods For Amplification The "Polymerase Chain Reaction" (PCR) comprises the first generation of methods for nucleic acid amplification. However, several other methods have been developed that employ the same basis of specificity, but create signal by different amplification mechanisms. These methods include the "Ligase Chain Reaction" (LCR), "Self Sustained Synthetic Reaction"
(3SR/NASBA), and "Q(3-Replicase" (Q(3).

Polymerise Chain Reaction (PCR) The polymerise chain reaction (PCR), as described in U.S. Patent Nos. 4,683,19 and 4,683,202 to Mullis and Mullis et al., describe a method for increasing the concentration of a segment of target sequence in a mixture of genomic DNA without cloning or purification.
This technology provides one approach to the problems of low target sequence concentration.
PCR can be used to directly increase the concentration of the target to an easily detectable level. This process for amplifying the target sequence involves introducing a molar excess of two oligonucleotide primers which are complementary to their respective strands of the double-stranded target sequence to the DNA mixture containing the desired target sequence.
The mixture is denatured and then allowed to hybridize. Following hybridization, the primers are extended with polymerise so as to form complementary strands. The steps of denaturation, hybridization, and polymerise extension can be repeated as often as needed, in order to obtain relatively high concentrations of a segment of the desired target sequence.
The length of the segment of the desired target sequence is determined by the relative positions of the primers with respect to each other, and, therefore, this length is a controllable parameter. Because the desired segments of the target sequence become the dominant sequences (in terms of concentration) in the mixture, they are said to be "PCR-amplified."
Lipase Chain Reaction (LCR or LAR) The lipase chain reaction (LCR; sometimes referred to as "Lipase Amplification Reaction" (LAR) described by Barany, Proc. Natl. Acid. Sci., 88:189 (1991);
Barany, PCR
Methods and Applic., 1:5 ( 1991 ); and Wu and Wallace, Genomics 4:560 ( 1989) has developed into a well-recognized alternative method for amplifying nucleic acids. In LCR, four oligonucleotides, two adjacent oligonucleotides which uniquely hybridize to one strand of target DNA, and a complementary set of adjacent oligonucleotides, which hybridize to the opposite strand are mixed and DNA lipase is added to the mixture. Provided that there is complete complementarity at the junction, lipase will covalently link each set of hybridized molecules. Importantly, in LCR, two probes are ligated together only when they base-pair with sequences in the target sample, without gaps or mismatches. Repeated cycles of denaturation, hybridization and ligation amplify a short segment of DNA. LCR
has also been used in combination with PCR to achieve enhanced detection of single-base changes. Segev:
PCT Public. No. W090Q1069 A1 (1990). However, because the four oligonucleotides used in this assay can pair to form two short ligatable fragments, there is the potential for the generation of target-independent background signal. The use of LCR for mutant screening is limited to the examination of specific nucleic acid positions.
__ Self Sustained Synthetic Reaction (3SR/NASBA) The self sustained sequence replication reaction (3SR) (Guatelli et al., Proc.
Natl.
Acid. Sci., 87:1874-1878 [1990], with an erratum at Proc. Natl. Acid. Sci., 87:7797 [1990]) is a transcription-based in vitro amplification system (Kwok et al., Proc.
Natl. Acid. Sci., 86:1173-1177 [1989]) that can exponentially amplify RNA sequences at a uniform temperature. The amplified RNA can then be utilized for mutation detection (Fahy et al., PCR Meth. Appl., 1:25-33 [1991]). In this method, an oligonucleotide primer is used to add a phage RNA polymerise promoter to the 5" end of the sequence of interest. In a cocktail of enzymes and substrates that includes a second primer, reverse transcriptase, RNase H, RNA
polymerise and ribo-and deoxyribonucleoside triphosphates, the target sequence undergoes repeated rounds of transcription, cDNA synthesis and second-strand synthesis to amplify the area of interest. The use of 3SR to detect mutations is kinetically limited to screening small segments of DNA (e.g., 200-300 base pairs).
Q-Beta (Q~) Replicase In this method, a probe which recognizes the sequence of interest is attached to the replicatable RNA template for Q(3 replicase. A previously identified major problem with false positives resulting from the replication of unhybridized probes has been addressed through use of a sequence-specific' ligation step. However, available ther.mostable DNA
ligases are not effective on this RNA substrate, so the ligation must be performed by T4 DNA
lipase at low temperatures (37°C). This prevents the use of high temperature as a means of achieving specificity as in the LCR, the ligation event can be used to detect a mutation at the junction site, but not elsewhere.
Table 1 below, lists some of the features desirable for systems useful in sensitive nucleic acid diagnostics, and summarizes the abilities of each of the major amplification methods (See also, Landgren, Trends in Genetics 9:199 [1993]).
A successful diagnostic method must be very specific. A straight-forward method of ' controlling the specificity of nucleic acid hybridization is by controlling the temperature of the reaction. While the 3SRlNASBA, and Q[3 systems are all able to generate a large quantity of signal, one or more of the enzymes involved in each cannot be used at high temperature ( i. e. , - J -WO 96/15267 PG"TlUS95/14673 >55°C). Therefore the reaction temperatures cannot be raised to prevent non-specific hybridization of the probes. If probes are shortened in order to make them melt more easily at low temperatures, the likelihood of having more than one perfect match in a complex genome increases. For these reasons, PCR and LCR currently dominate the research field in detection technologies. -METHOD:

FEATURE PCR &

PCR LCR
LCR
NASBA
Qa Amplifies Target + + + +

Recognition of Independent+ + + + +
Sequences Required Performed at High Temp.+ +

Operates at Fixed Temp. + +

Exponential Amplification+ + + + +

Generic Signal Generation +

Easily Automatable The basis of the amplification procedure in the PCR and LCR is the fact that the products of one cycle become usable templates in all subsequent cycles, consequently doubling the population with each cycle. The final yield of any such doubling system can be expressed as: (1+X)° = y, where "X" is the mean efficiency (percent copied in each cycle), "n" is the number of cycles, and "y" is the overall efficiency, or yield of the reaction (Mullis, PCR Methods Applic., 1:1 [19910. If every copy of a target DNA is utilized as a template in every cycle of a polymerise chain reaction, then the mean efficiency is 100%.
If 20 cycles of PCR are performed, then the yield will be 2'-°, or 1,048.576 copies of the starting material. If the reaction conditions reduce the mean efficiency to 85%, then the yield in those 20 cycles will be only 1.85'-°, or 220,513 copies of the starting material. In other words, a PCR running a _q._ at 85% efficiency will yield only 21% as much final product,, compared to a reaction running at 100% efficiency. A reaction that is reduced to 50% mean efficiency will yield less than 1 % of the possible product.
In practice, routine polymerase chain reactions rarely achieve the theoretical maximum yield, and PCRs are usually run for more than 20 cycles to compensate for the lower yield.
At 50% mean efficiency, it would take 34 cycles to achieve ~.he million-fold amplification theoretically possible in 20, and at lower efficiencies, the number of cycles required becomes prohibitive. In addition, any background products that amplify with a better mean efficiency than the intended target will become the dominant products.
Also, many variables can influence the mean efficiency of PCR, including target DNA
length and secondary structure, primer length and design, primer and dNTP
concentrations, and buffer composition, to name but a few. Contamination of the reaction with exogenous DNA (e.g., DNA spilled onto lab surfaces) or cross-contamination is also a major consideration. Reaction conditions must be carefully optimized for each different primer pair and target sequence, and the process can take days, even for an experienced investigator. The laboriousness of this process, including numerous technical considerations and other factors, presents a significant drawback to using PCR in the clinical setting. Indeed, PCR has yet to penetrate the clinical market in a significant way. The same concerns arise with LCR, as LCR must also be optimized to use different oligonucleotide sequences for each target sequence. In addition, both methods require expensive equipment, capable of precise temperature cycling.
Many applications of nucleic acid detection technologies, such as in studies of allelic variation, involve not only detection of a specific sequence in a complex background, but also the discrimination between sequences with few, or single, nucleotide differences. One method for the detection of allele-specific variants by PCR is based upon the fact that it is difficult for Tack polymerase to synthesize a DNA strand when there is a mismatch between the template strand and the 3' end of the primer. An allele-specific variant may be detected by the use of a primer that is perfectly matched with only one of the possible alleles; the mismatch to the other allele acts to prevent the extension of the primer, thereby preventing the amplification of that sequence. This method has a substantial limitation in that the base composition of the mismatch influences the ability to prevent extension across the mismatch, and certain mismatches do not prevent extension or have only a minimal effect (Kwol: et cal., Nucl. Acids Res., 18:999 [1990]).) A similar 3'-mismatch strategy is used with greater effect to prevent ligation in the LCR (Barany, PCR Meth. Applic., 1:5 [1991]). Any mismatch effectively blocla the action of the thermostable ligase, but LCR still has the drawback of target-independent background ligation products initiating the amplification. Moreover, the combination of PCR with A
subsequent LCR to identify the nucleotides at individual positions is also a clearly cumbersome proposition for the clinical laboratory.
II. Direct Detection Technology When a sufficient amount of a nucleic acid to be detected is available, there are advantages to detecting that sequence directly, instead of making more copies of that target, (e.~T., as in PCR and LCR). Most notably, a method that does not amplify the signal exponentially is more amenable to quantitative analysis. Even if the signal is enhanced by attaching multiple dyes to a single oligonucleotide, the correlation between the final signal intensity and amount of target is direct. Such a system has an additional advantage that the products of the reaction will not themselves promote further reaction, so contamination of lab surfaces by the products is not as much of a concern. Traditional methods of direct detection including Northern and Southern blotting and RNase protection assays usually require the use of radioactivity and are not amenable to automation. Recently devised techniques have sought to eliminate the use of radioactivity and/or improve the sensitivity in automatable formats.
Two examples are the "Cycling Probe Reaction" (CPR), and "Branched DNA"
(bDNA).
The cycling probe reaction (CPR) (Duck et al., BioTech., 9:142 [1990]), uses a long chimeric oligonucleotide in which a central portion is made of RNA while the two termini are made of DNA. Hybridization of the probe to a target DNA and exposure to a thermostable RNase H causes the RNA portion to be digested. This destabilizes the remaining DNA
portions of the duplex, releasing the remainder of the probe from the target DNA and allowing another probe molecule to repeat the process. The signal, in the form of cleaved probe molecules, accumulates at a linear rate. While the repeating process increases the signal, the RNA portion of the oligonucleotide is vulnerable to RNases that may carried through sample preparation.
Branched DNA (bDNA), described by Urdea et al., Gene 61:253-264 (1987), involves oligonucleotides with branched structures that allow each individual oligonucleotide to carry ' 3~ to 40 labels (e.g., alkaline phosphatase enzymes). While this enhances the signal from a hybridization event, signal from non-specific binding is similarly increased.

III. Detection Of Unknown Sequence Changes The demand for tests which allow the detection of spe;ciflc nucleic acid sequences and sequence changes is growing rapidly in clinical diagnostics. As nucleic acid sequence data for genes from humans and pathogenic organisms accumulates, the demand for fast, cost-s effective, and easy-to-use tests for as yet unknown mutations within specific sequences is rapidly increasing.
A handful of methods have been devised to scan nucleic acid segments for mutations.
One option is to determine the entire gene sequence of each lest sample (e.g., a bacterial isolate). For sequences under approximately 600 nucleotides, this may be accomplished using amplified material (e.g., PCR reaction products). This avoids the time and expense associated with cloning the segment of interest. However, specialized equipment and highly trained personnel are required, and the method is too labor-intense and expensive to be practical and effective in the clinical setting.
In view of the difficulties associated with sequencing, a given segment of nucleic acid may be characterized on several other levels. At the lowest resolution, the size of the molecule can be determined by electrophoresis by comparison to a known standard run on the same gel. A more detailed picture of the molecule may be achieved by cleavage with combinations of restriction enzymes prior to electrophoresis, to allow construction of an ordered map. The presence of specific sequences within the fragment can be detected by hybridization of a labeled probe, or the precise nucleotide sequence can be determined by partial chemical degradation or by primer extension in the presence of chain-terminating nucleotide analogs.
For detection of single-base differences between like sequences, the requirements of the analysis are often at the highest level of resolution. For cases in which the position of the nucleotide in question is known in advance, several methods have been developed for examining single base changes without direct sequencing. For example, if a mutation of interest happens to fall within a restriction recognition sequence, a change in the pattern of digestion can be used as a diagnostic tool (e.g., restriction fragment length polymorphism [RFLP] analysis).
Single point mutations have been also detected by the creation or destruction of ' RFLPs. Mutations are detected and localized by the presence and size of the RNA fragments generated by cleavage at the mismatches. Single nucleotide :mismatches in DNA
heteroduplexes are also recognized and cleaved by some chemicals, providing an alternative WO 96/15267 PC"T/US95/14673 strategy to detect single base substitutions, generically named the "Mismatch Chemical Cleavage" (MCC) (Gogos et al. , Nucl. Acids Res., 18:6807-6817 [ 1990]).
However, this method requires the use of osmium tetroxide and piperidine, two highly noxious chemicals which are not suited for use in a clinical laboratory.
RFLP analysis suffers from low sensitivity and requires a large amount of sample.
F
When RFLP analysis is used for the detection of point mutations, it is, by its nature, limited to the detection of only those single base changes which fall within a restriction sequence of a known restriction endonuclease. Moreover, the majority of the available enzymes have 4 to 6 base-pair recognition sequences, and cleave too frequently for many large-scale DNA
manipulations (Eckstein and Lilley (eds.), Nucleic Acids and Molecular°
Biolo~~, vol. 2, Springer-Verlag, Heidelberg [1988]). Thus, it is applicable only in a small fraction of cases, as most mutations do not fall within such sites.
A handful of rare-cutting restriction enzymes with 8 base-pair specificities have been isolated and these are widely used in genetic mapping, but these enzymes are few in number, are limited to the recognition of G+C-rich sequences, and cleave at sites that tend to be highly clustered (Barlow and Lehrach, Trends Genet., 3:167 [1987]). Recently, endonucleases encoded by group I introns have been discovered that might have greater than 12 base-pair specificity (Penman and Butow, Science 246:1106 [1989]), but again, these are few in number.
If the change is not in a recognition sequence, then allele-specific oligonucleotides (ASOs), can be designed to hybridize in proximity to the unknown nucleotide, such that a primer extension or ligation event can be used as the indicator of a match or a mis-match.
Hybridization with radioactively labeled allelic specific oligonucleotides (ASO) also has been applied to the detection of specific point mutations (Conner et al., Proc.
Natl. Acad. Sci., 80:278-282 [1983]). The method is based on the differences in the melting temperature of short DNA fragments differing by a single nucleotide. Stringent hybridization and washing conditions can differentiate between mutant and wild-type alleles. The ASO
approach applied to PCR products also has been extensively utilized by various researchers to detect and characterize point mutations in ras genes (Vogelstein et crl., N. Eng. J.
Med., 319:52-~3?
[1988]; and Farr et al., Proc. Natl. Acad. Sci., 85:1629-1633 [1988]), and gaplgip oncogenes (Lyons et al., Science 249:655-659 [1990]). Because of the presence of various nucleotide changes in multiple positions, the ASO -method requires the use of many oligonucleotides to cover all possible oncogenic mutations.
_g_ With either of the techniques described above (i.e., RF LP and ASO), the precise location of the suspected mutation must be known in advance: of the test. That is to say, they are inapplicable when one needs to detect the presence of a mutation of an unknown character and position within a gene or sequence of interest.
Two other methods rely on detecting changes in electrophoretic mobility in response to r minor sequence changes. One of these methods, termed "Denaturing Gradient Gel Electrophoresis" (DGGE) is based on the observation that slightly different sequences will display different patterns of local melting when electrophoretically resolved on a gradient gel.
In this manner, variants can be distinguished, as differences in melting properties of homoduplexes versus heteroduplexes differing in a single nucleotide can detect the presence of mutations in the target sequences because of the corresponding changes in their electrophoretic mobilities. The fragments to be analyzed, usually PCR
products, are "clamped" at one end by a long stretch of G-C base pairs (30-80) to allow complete denaturation of the sequence of interest without complete dissociation of the strands. The attachment of a GC "clamp" to the DNA fragments increases the fraction of mutations that can be recognized by DGGE (Abrams et al., Genomics 7:463-475 [1990]).
Attaching a GC
clamp to one primer is critical to ensure that the amplified sequence has a low dissociation temperature (Sheffield et al., Proc. Natl. Acad. Sci., 86:232-236 [1989]; and Lerman and Silverstein, Meth. Enzymol., 155:482-SO1 [1987]). Modifications of the technique have been developed, using temperature gradients (Wartell et al., Nucl. Acids Res., 18:2699-2701 [1990]), and the method can be also applied to RNA:RNA duplexes (Smith et al., Genomics 3:217-223 [1988]).
Limitations on the utility of DGGE include the requirement that the denaturing conditions must be optimized for each type of DNA to be tested. Furthermore, the method requires specialized equipment to prepare the gels and maintain the needed high temperatures during electrophoresis. The expense associated with the synthesis of the clamping tail on one oligonucleotide for each sequence to be tested is also a major consideration.
In addition. long running times are required for DGGE. The long running time of DGGE was shortened in a modification of DGGE called constant denaturant gel electrophoresis (CDGE) (Borrensen et ul., Proc. Natl. Acad. Sci. USA 88:8405 [1991]). CDGE requires that gels be performed under different denaturant conditions in order to reach high efficiency for the detection of unknown mutations.

An technique analogous to DGGE, termed temperature gradient gel electrophoresis (TGGE), uses a-thermal gradient rather than a chemical denaturant gradient (Scholz; et al., Hum. Mol. Genet. 2:2155 [1993]). TGGE requires the use of specialized equipment which can generate a temperature gradient perpendicularly oriented relative to the electrical field.
TGGE can detect mutations in relatively small fragments of DNA therefore scanning of large gene segments requires the use of multiple PCR products prior to running the gel.
Another common method, called "Single-Strand Conformation Polymorphism" (SSCP) was developed by Hayashi, Sekya and colleagues (reviewed by Hayashi, PCR Meth.
Appl..
1:34-38, [1991]) and is based on the observation that single strands of nucleic acid can take on characteristic conformations in non-denaturing conditions, and these conformations influence electrophoretic mobility. The complementary strands assume sufficiently different structures that one strand may be resolved from the other. Changes in sequences within the fragment will also change the conformation, consequently altering the mobility and allowing this to be used as an assay for sequence variations (Orita, et al., Genomics 5:874-879, [1989]).
The SSCP process involves denaturing a DNA segment (e.g., a PCR product) that is labelled on both strands, followed by slow electrophoretic separation on a non-denaturing polyacrylamide gel, so that intra-molecular interactions can form and not be disturbed during the run. This technique is extremely sensitive to variations in gel composition and temperature. A serious limitation of this method is the relative difficulty encountered in comparing data generated in different laboratories, under apparently similar conditions.
The dideoxy fingerprinting (ddF) is another technique developed to scan genes for the presence of unknown mutations (Liu and Sommer, PCR Methods Appli., 4:97 [1994]). The ddF technique combines components of Sanger dideoxy sequencing with SSCP. A
dideoxy sequencing reaction is performed using one dideoxy terminator and then the reaction products are electrophoresised on nondenaturing polyacrylamide gels to detect alterations in mobility of the termination segments as in SSCP analysis. While ddF is an improvement over SSCP in terms of increased sensitivity, ddF requires the use of expensive dideoxynucleotides and this technique is still limited to the analysis of fragments of the size suitable for SSCP (i.c., fragments of 200-300 bases for optimal detection of mutations).
In addition to the above limitations, all of these methods are limited as to the size of the nucleic acid fragment that can be analyzed. For the direct-sequencing approach.
sequences of greater than 600 base pairs require cloning, with the consequent delays and expense of either deletion sub-cloning or primer walking, in order to cover the entire fragment. SSCP and DGGE have even more severe size limitations. Because of reduced sensitivity to sequence changes, these methods are not considered suitable for larger fragments. Although SSCP is reportedly able to detect 90% of single-base substitutions y 5 within a 200 base-pair fragment, the detection drops to less than 50% for 400 base pair fragments. Similarly, the sensitivity of DGGE decreases as the length of the fragment reaches 500 base-pairs. The ddF technique, as a combination of direct sequencing and SSCP, is also limited by the relatively small size of the DNA that can be screened.
Clearly, there remains a need for a method that is less sensitive to size so that entire genes, rather than gene fragments, may be analyzed. Such a. tool must also be robust, so that data from different labs, generated by researchers of diverse backgrounds and skills will be comparable. Ideally, such a method would be compatible with "multiplexing,"
(i.c., the simultaneous analysis of several molecules or genes in a single reaction or gel lane, usually resolved from each other by differential labelling or probing j. Such an analytical procedure would facilitate the use of internal standards for subsequent analysis and data comparison, and increase the productivity of personnel and equipment. The ideal method would also be easily automatable.
SUMMARY OF THE INVENTION
The present invention relates to methods and compositions for treating nucleic acid, and in particular, methods and compositions for detection and characterization of nucleic acid sequences and sequence changes in human gene sequences and in microbial gene sequences.
The present invention provides means for cleaving a nucleic acid cleavage structure in a site-specific manner. In one embodiment, the means for cleaving is an enzyme capable of cleaving cleavage structures on a nucleic acid substrate, forming the basis of a novel method of detection of specific nucleic acid sequences. The present invention contemplates use of the novel detection method for, among other uses, clinical diagnostic purposes, including but not limited to detection and identification of 1 ) mutations in human gene sequences and 2 ) pathogenic organisms.
In one embodiment, the present invention contemplates a DNA sequence encoding a DNA polymerase altered in sequence (i.e., a "mutant" DNA polymerase) relative to the native sequence such that it exhibits altered DNA synthetic activity from that of the native (i.c..
"wild type") DNA polymerase. With regard to the polymera.se, a complete absence of synthesis is not required; it is desired that cleavage reactions occur in the absence of polymerise activity at a level where it interferes with the method. It is preferred that the encoded DNA polymerise is altered such that it exhibits reduced synthetic activity from that of the native DNA polymerise. In this manner, the enzymes of the invention are nucleases and are capable of cleaving nucleic acids in a structure-specific manner.
Importantly, the nucleases of the present invention are capable of cleaving cleavage structures to create discrete cleavage products.
The present invention contemplates nucleases from a variety of sources, including nucleases that are thermostable. Thermostable nucleases are contemplated as particularly useful, as they are capable of operating at temperatures where nucleic acid hybridization is extremely specific, allowing for allele-specific detection (including single-base mismatches).
In one embodiment, the thermostable 5' nucleases are selected from the group consisting of altered polymerises derived from the native polymerises of various Thermus species, including, but not limited to Thermus aquaticzrs, Thermus.flavus and Thermus thermophilu.s.
The present invention is not limited to the use of thermostable nucleases. As demonstrated herein nucleases from mesophilic organisms may also be employed in the methods of the invention (e.g., E. coli Exo III, Saccharomyce.s cerevisiae Radi/RadlO
complex).
The present invention utilizes nucleases in methods for detection and characterization of nucleic acid sequences and sequence changes. The present invention relates to means for cleaving a nucleic acid cleavage structure in a site-specific manner. Nuclease activity is used to screen for known and unknown mutations, including single base changes, in nucleic acids.
In one embodiment, the present invention contemplates a method for treating nucleic acid, comprising: a) providing: i) a cleavage means and ii) nucleic acid substrate; b) treating the nucleic acid substrate under conditions such that the substrate forms one or more cleavage structures; and c) reacting the cleavage means with the cleavage structures so that one or more cleavage products are produced.
In one embodiment, the cleavage means is an enzyme. In a preferred embodiment, the cleavage means is a nuclease. In an alternative preferred embodiment, the nuclease is selected from the group consisting of the CleavaseTM BN enzyme, Thermus aquaticus DNA
polymerise, Thermus thermophilZrs DNA polymerise, Escherichia coli Exo III.
and the Saccharomyces cerevisiae Radl/RadlO complex.

It is contemplated that the nucleic acid substrate comprise a nucleotide analog.
including but not limited to the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.
In one embodiment, the nucleic acid of step is substantially single-stranded.
It is not intended that the nucleic acid substrate be limited to any particular form, indeed, it is contemplated that the nucleic acid substrate is single stranded or double stranded DNA or RNA.
In one embodiment, when a double stranded nucleic acid substrate is employed, the treating step (b) comprises rendering the double-stranded nucleic acid substantially single-stranded and exposing the single-stranded nucleic acid to conditions such that the single-stranded nucleic acid assumes a secondary or characteristic folded structure.
In one preferred embodiment, the double stranded nucleic acid is rendered substantially single-stranded by increased temperature.
In an alternative embodiment, the method of the present invention further comprises the step of detecting said one or more cleavage products.
In a preferred embodiment, the nucleic acid substrate comprises an oligonucleotide containing human p53 gene sequences. In an alternative embodiment, the nucleic acid substrate comprises an oligonucleotide containing microbial gene sequences.
The present invention contemplates further a method for treating nucleic acid, comprising: a) providing: i) a cleavage means in a solution comprising manganese and ii) a nucleic acid substrate; b) treating the nucleic acid substrate with increased temperature; c) reducing the temperature under conditions such that the substrate forms one or more cleavage structures; d) reacting the cleavage means with the cleavage structures so that one or more cleavage products are 'produced: and e) detecting the cleavage products.
Again, the cleavage means may be an enzyme. As noted above, the cleavage means may be a nuclease.
In an alternative preferred embodiment. the nuclease is selected from the group consisting of the CleavaseTM BN enzyme, Thermus aguaticus DNA polymerase, Thermus thermophilz~s DNA
polymerase, Escherichia coli Exo III, and the Saccharornyces cerevisiae Radl/RadlO complex.
It is contemplated that the nucleic acid substrate comprise a nucleotide analog, including, but not limitedto, the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.
In one embodiment, the nucleic acid of step is substantially single-stranded.
It is not intended that the nucleic acid substrate be limited to any particular form, indeed, it is contemplated that the nucleic acid substrate is single stranded or double stranded DNA or RNA.

In a preferred embodiment, the nucleic acid substrate comprises an oligonucleotide containing human p53 gene sequences. In an alternative embodiment, the nucleic acid substrate comprises an oligonucleotide containing microbial gene sequences.
The present invention contemplates further, a method for detecting mutation in the human p53 gene, comprising: a) providing: i) a cleavage means and ii) a nucleic acid substrate containing human p53 gene sequences; b) treating the nucleic acid substrate under conditions such that the substrate forms one or more cleavage structures; c) reacting the cleavage means with the cleavage structures so that one or more cleavage products are produced; and d) comparing said cleavage products to the cleavage products produced by cleavage of a reference p53 gene sequence.
In a preferred embodiment, the cleavage products produced by cleavage of a reference p53 gene sequence are generated by the cleavage of a nucleic acid substrate containing the human p53 gene sequences selected from the group consisting of SEQ ID NOS:79-81, 84-89 and 94-97. Additional p53 mutant sequences are provided herein; SEQ ID N0:79 lists the I ~ sequence of the wild-type p53 cDNA. Table 2 below provides the identity and location of numerous known p53 mutations. Combination of the information in Table 2 with the sequence of the wild-type p53 cDNA in SEQ ID N0:79 allows the generation of the complete nucleotide sequence for cDNAs corresponding to the numerous p53 mutations described in Table 2. In addition, as described fully herein, the method of the invention permits the screening of or "scanning" for heretofore uncharacterized mutations within human gene sequences, such as the human p53 gene.
The present invention also contemplates a process for creating a record reference library of genetic fingerprints characteristic (i.e., diagnostic) of one or more alleles of the human p53 gene comprising: a) providing: i) a cleavage means, and ii) nucleic acid substrate derived from human p53 gene sequences; b) contacting said nucleic acid substrate with a cleavage means under conditions such that said extracted nucleic acid forms one or more secondary structures and said cleavage means cleaves said secondary structures resulting in the generation of multiple cleavage products; c) separating said multiple cleavage products; and d) maintaining a testable record reference of said separated cleavage products. .
By the term "genetic fingerprint" it is meant that changes in the sequence of the nucleic acid (e.g., a deletion, insertion or a single point substitution) alter the structures formed, thus changing the banding pattern (i.e., the "fingerprint" or "bar code") to reflect the difference in the sequence, allowing rapid detection and identification of variants.

The present invention also contemplates a process for creating a record reference library of genetic fingerprints characteristic (i.e., diagnostic) of one or more alleles of one or more genes from a eukaryotic organism (e.g., mammals) comprising: a) providing: i) a cleavage means; and ii) nucleic acid substrate derived from one or more alleles of a gene derived from a eukaryotic organism; b) contacting said nucleic acid substrate with a cleavage means under conditions such that said extracted nucleic acid forms one or more secondary structures and said cleavage means cleaves said secondary structures resulting in the generation of multiple cleavage products; c) separating said multiple cleavage products; and d) maintaining a testable record reference of said separated cleavage products.
The present invention also contemplates a method for identifying strains of microorganisms comprising: a) providing i) a cleavage means; and ii) a nucleic acid substrate containing sequences derived from one or more microorgani sm; b) treating said nucleic acid substrate under conditions such that said substrate forms one or more cleavage structures; and c) reacting said cleavage means with said cleavage structure s so that one or more cleavage I S products are produced.
The preferred cleavage means is an enzyme, such as a nuclease. Examples of enzymes that can be used with success with the method of the present invention include (but are not limited to) the CleavaseTM BN enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA polymerase, Escherichia coli Exo III, and the Saccharomvcc~s cef°evi.siae Rad 1 /Rad 10 complex.
It is contemplated that the nucleic acid substrate comprise a nucleotide analog, including but not limited to the group comprising 7-deaza-d,ATP, 7-deaza-dGTP
and dUTP.
In one embodiment, the nucleic acid of step is substantially single-stranded.
It is not intended that the nucleic acid substrate be limited to any particular form, indeed, it is contemplated that the nucleic acid substrate is single stranded or double stranded DNA or RNA.
In one embodiment, when a double stranded nucleic acid substrate is employed, the treating step (b) comprises rendering the double-stranded nucleic acid substantially single-stranded and exposing the single-stranded nucleic acid to conditions such that the single-stranded nucleic acid assumes a secondary or characteristic folded structure.
In one preferred embodiment, the double stranded nucleic acid is rendered substantially single-stranded by increased temperature.
In an alternative embodiment, the method of the present invention further comprises the step of detecting said one or more cleavage products.

It is contemplated that the microorganisms) of the present invention be selected from a variety of microorganisms; it is not intended that the present invention be limited to any particular type of microorganism. Rather, it is intended that the present invention will be used with organisms including, but not limited to, bacteria, fungi, protozoa, ciliates, and r viruses. It is not intended that the microorganisms be limited to a particular genus, species, strain, or serotype. Indeed, it is contemplated that the bacteria be selected from the group comprising, but not limited to members of the genera Campylobacter, Escherichia, Mvcobacterium, Salmonella, Shigella,and Staphylococczrs. In one preferred embodiment, the microorganisms) comprise strains of mufti-drug resistant Mycobacterium tuberculosis. It is also contemplated that the present invention be used with viruses, including but not limited to hepatitis C virus and simian immunodeficiency virus.
Another embodiment of the present invention contemplates a method for detecting and identifying strains of microorganisms, comprising the steps of extracting nucleic acid from a sample suspected of containing one or more microorganisms and contacting the extracted nucleic acid with a cleavage means under conditions such that the extracted nucleic acid forms one or more secondary structures, and the cleavage means cleaves the secondary structures to produce one or more cleavage products.
In one embodiment, the method further comprises the step of separating said cleavage products. In yet another embodiment, the method further comprises the step of detecting said cleavage products.
In one preferred embodiment, the present invention further comprises comparing said detected cleavage products generated from cleavage of the extracted nucleic acid isolated from the sample with separated cleavage products generated by cleavage of nucleic acids derived from one or more reference microorganisms. In such a case, the sequence of the nucleic acids from one or more reference microorganisms may be related but different (e.g., a wild-type control for a mutant sequence or a known or previously characterized mutant sequence).
In an alternative preferred embodiment, the present invention further comprises the step of isolating a polymorphic locus from said extracted nucleic acid after the extraction step, so as to generate a nucleic acid substrate, wherein the substrate is_ contacted with the cleavage means. In one embodiment, the isolation of a polymorphic locus is accomplished by nucleic acid amplification. The invention is limited by the method of nucleic acid amplification employed. One method of achieving nucleic acid amplification is the polymerase chain reaction. In an alternate embodiment, the nucleic acid amplification is conducted in the WO 96/15267 PC"T/US95/14673 presence of a nucleotide analog, including but not limited to the group comprising 7-deaza-dATP, 7-deaza-dGTP and dUTP. It is contemplated that the nucleic acid amplification (e.g., PCR) will employ oligonucleotide primers which either 1 ) match consensus gene sequences derived from the polymorphic locus (i. e., the primers comprise the same sequence found on a strand of nucleic acid derived from the polymorphic locus) o:r 2) are complementary to consensus gene sequences derived from said polymorphic locus (i. e., they are the complement to a strand of nucleic acid derived from the polymorphic locus). In one embodiment, the polymorphic locus comprises a ribosomal RNA gene. In a particularly preferred embodiment, the ribosomal RNA gene is a 16S ribosomal RNA gene.
In one embodiment of this method, the cleavage means is an enzyme, such as a nuclease. In a particularly preferred embodiment, the nuclease is selected from the group including, but not limited to CleavaseTM BN, Thermus aquaticus DNA polymerase, Thermus thc~rmophilus DNA polymerase, Escherichia coli Exo III, and. the Saccharomyces cerevisicre Rad 1 /Rad I 0 complex. It is also contemplated that the enzyme may have a portion of its amino acid sequence that is homologous to a portion of the amino acid sequence of a thermostable DNA polymerase derived from a eubacterial the:rmophile, the latter being selected from the group consisting of Thermus aquaticus, Thermus , flavus and Thermus thermophilus.
It is not intended that the nucleic acid substrate be limited to any particular form, indeed, it is contemplated that the nucleic acid substrate is single stranded or double-stranded RNA or DNA. When a double stranded nucleic acid substrate is employed, the treating step of the method may comprise rendering double-stranded nucleic acid substantially single-stranded, and exposing the single-stranded nucleic acid to conditions such that the single-stranded nucleic acid has secondary structure. In one preferred embodiment, double-stranded nucleic acid is rendered substantially single-stranded by increased temperature.
It is contemplated that the microorganisms) of the present invention be selected from a variety of microorganisms; it is not intended that the present invention be limited to any particular type of microorganism. Rather, it is intended that the present invention will be used with organisms including, but not limited to, bacteria, fungi, protozoa, ciliates, and viruses. It is not intended that the microorganisms be limited to a particular genus, species, strain, or serotype. Indeed, it is contemplated that the bacteria be selected from the group comprising, but not limited to members of the genera Campylobacter°, Escher-ichicr, Mycobacterium. Salmonella, Shigella, and Staphylococcus. In one preferred embodiment, the microorganisms) comprise strains of mufti-drug resistant Mycobacterium tuberculosis. It is also contemplated that the present invention be used with viruses, including but not .limited to hepatitis C virus and simian immunodeficiency virus.
The present invention also contemplates a process for creating a record reference library of genetic fingerprints characteristic (i.e., diagnostic) of one or more alleles of the , various microorganisms, comprising the steps of providing a cleavage means and nucleic acid substrate derived from microbial gene sequences; contacting the nucleic acid substrate with a cleavage means under conditions such that the extracted nucleic acid forms one or more secondary structures and the cleavage means cleaves the secondary structures, resulting in the generation of multiple cleavage products; separating the multiple cleavage products; and maintaining a testable record reference of the separated cleavage products.
It is not intended that the present invention be limited by the nature of the microorganism. The detection and identification is application to all organisms, including viruses and bacteria. -The present invention also contemplates a process for creating a record reference (e.g., library) of genetic fingerprints characteristic (i.e., diagnostic) of pathogenic microorganisms comprising: a) providing: i) a cleavage means; and ii) a nucleic acid substrate characteristic of (e.~T., derived from a polymorphic locus) isolated from a known pathogenic microorganism; b) contacting said nucleic acid substrate with a cleavage means under conditions such that said extracted nucleic acid forms one or more secondary structures and said cleavage means cleaves said secondary structures resulting in the generation of multiple cleavage products; c) separating said multiple cleavage products; and d) maintaining a record reference of said separated cleavage products.
The present invention also contemplates a nucleic acid treatment kit,-comprising: a) an enzyme capable of reacting with cleavage structures so as to generate cleavage products, and b) a solution comprising manganese. The enzyme of the kit may be a nuclease.
In a preferred embodiment the nuclease is elected from the group including, but not limited to CleavaseTM BN, Thermus aquaticus DNA polymerase, Thernzus--thermophilus DNA
polymerase, Escherichia coli Exo III, and the Saccharomyces cerevisiae Radl/RadlO complex.
The present invention contemplates other reagents useful for the treatment of nucleic acid.
For example, the kit may include reagents for detecting said cleavage products. Furthermore, the kit may include reagents for the cleavage reaction including salt solutions (e.~.. IiCI and NaCl solutions), manganese chloride solutions, buffer solutions and solutions which terminate the cleavage reaction.
The methods of the present invention allow for simultaneous analysis of both strands (e. g., the sense and antisense strands) and are ideal for high-level multiplexing. The products produced are amenable to qualitative, quantitative and positional analysis. The methods may be automated and may be practiced in solution or in the solid phase (e. g., on a solid support). The methods are powerful in that they allow for analysis of longer fragments of nucleic acid than current methodologies.
More specifically, the present invention provides a method for treating nucleic acid, comprising: (a) providing:
i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid substrate; (b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure; and (c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced.
The present invention also provides A method for treating nucleic acid, comprising: (a) providing: i) an enzymatic cleavage means comprising a nuclease, in a solution comprising manganese; and ii) a nucleic acid substrate; (b) treating said nucleic acid substrate with increased temperature;
(c) reducing said temperature under conditions such that said substrate forms at least one cleavage structure; (d) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (e) detecting said at least one cleavage product.
The present invention also provides a method for detecting mutation in the human p53 gene, comprising:(a) providing:
i) an enzymatic cleavage means wherein comprising a nuclease;
and ii) a nucleic acid substrate containing human p53 gene sequences; (b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure; (c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (d) comparing said cleavage product to the cleavage products produced by cleavage of a reference p53 gene sequence.
The present invention also provides a method for identifying strains of microorganisms comprising: (a) providing:
i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid substrate containing sequences derived from at least one microorganism; (b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure; (c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (d) comparing said cleavage product to the cleavage products produced by cleavage of a reference sequence derived from a microorganism.
The present invention also provides a method comprising:
(a) extracting nucleic acid from a sample suspected of containing at least one microorganism; and (b) contacting said extracted nucleic acid with an enzymatic cleavage means comprising a nuclease, under conditions such that said extracted nucleic acid forms one or more secondary structures, and said cleavage means cleaves said secondary structures to produce at least one cleavage product.
The present invention also provides a method comprising:
(a) providing: i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said cleavage means and said substrate under conditions such that said substrate forms at least one secondary structure and said cleavage means cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.
The present invention also provides a method comprising:
(a) providing: i) an enzymatic cleavage means comprising a 19a nuclease; and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said cleavage means and said substrate at an elevated temperature and under conditions such that said substrate forms at least one secondary structure and said cleavage means cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.
The present invention also provides a method comprising:
(a) providing: i) a thermostable DNA polymerase altered in amino acid sequence such that it exhibits reduced DNA synthetic activity from that of the wild-type DNA polymerase but retains substantially the same 5' nuclease activity of the wild-type DNA
polymerase; and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said polymerase and said substrate under conditions such that said substrate forms at least one secondary structure and said polymerase cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.
DESCRIPTION OF THE DRAWINGS
Figure 1 is a comparison of the nucleotide structure of the DNAP genes isolated from Thermus aquaticus (SEQ ID NO:1), Thermus flavus (SEQ ID N0:2) and Thermus thermophilus (SEQ ID
N0:3); the consensus sequence (SEQ ID N0:7) is shown at the top of each row.
Figure 2 is a comparison of the amino acid sequence of the DNAP isolated from Thermus aquaticus (SEQ ID N0:4), Thermus flavus (SEQ ID N0:5) and Thermus thermophilus (SEQ ID N0:6); the consensus sequence (SEQ ID N0:8) is shown at the top of each row.
19b Figure 3 is a schematic showing the CFLPT"" method of generating a characteristic fingerprint from a nucleic acid substrate.
Figure 4 depicts the organization of the human p53 gene;
exons are represented by the solid black boxes and are labelled 1-11. Five hot spot regions are shown as a blow-up of the region spanning exons 5-8; the hot spot regions are labelled A, A', B, C, and D.
Figure 5 provides a schematic showing the use of a first 2-step PCR technique for the generation DNA fragments containing p53 mutations.
Figure 6 provides a schematic showing the use of a second 2-step PCR technique for the generation DNA fragments containing p53 mutations.
Figure 7 depicts a structure which cannot be amplified using DNAPTaq.
Figure 8 is an ethidium bromide-stained gel demonstrating attempts to amplify a bifurcated duplex using either DNAPTaq or DNAPStf(Stoffel).
Figure 9 is an autoradiogram of a gel analyzing the cleavage of a bifurcated duplex by DNAPTaq and lack of cleavage by DNAPStf.
19c WO 96/15267 PC"TlUS95/14673 Figures 10 A-B are a set of autoradiograms of gels analyzing cleavage or lack of cleavage upon addition of different reaction components and change of incubation temperature during attempts to cleave a bifurcated duplex with DNAPTaq.
Figures 11 A-B are an autoradiogram displaying timed cleavage reactions. with and without primer.
Figures 12 A-B are a set of autoradiograms of gels demonstrating attempts to cleave a bifurcated duplex (with and without primer) with various DNAPs.
Figures 13A shows the substrates and oligonucleotides used to test the specific cleavage of substrate DNAs targeted by pilot oligonucleotides.
Figure 13B shows an autoradiogram of a gel showing the results of cleavage reactions using the substrates and oligonucleotides shown Fig. 13A.
Figure 14A shows the substrate and oligonucleotide used to test the specific cleavage of a substrate RNA targeted by a pilot oligonucleotide.
Figure 14B shows an autoradiogram of a gel showing the results of a cleavage reaction using the substrate and oligonucleotide shown in Fig. 14A. -Figure 15 is a diagram of vector pTTQlB.
Figure 16A-G are a set of diagrams of wild-type and synthesis-deficient DNAPTuq genes. _.
Figure 17 is a diagram of vector pET-3c.
Figure 18A depicts the wild-type Thermus flavus polymerase gene.
Figure 18B depicts a synthesis-deficient Thermus.flavus polymerase gene.
Figures 19A-E depict a set of molecules which are suitable substrates for cleavage by the 5' nuclease activity of DNAPs.
Figure 20 is an autoradiogram of a gel showing the results of a cleavage reaction run with synthesis-deficient DNAPs.
Figure 21 is an autoradiogram of a PEI chromatogram resolving the products of an assay for synthetic activity in synthesis-deficient DNAPTaq clones.
Figure 22A depicts the substrate molecule used to test the ability of synthesis-deficient DNAPs to cleave short hairpin structures.
Figure 22B shows an autoradiogram of a gel resolving the products of a cleavage reaction run using the substrate shown in Fig. 22A.
Figure 23 provides the complete 206-mer duplex sequence employed as a substrate for the 5' nucleases of the present invention Figures 24A and B show the cleavage of linear nucleic acid substrates (based on the 206-mer of Figure 23) by wild type DNAPs and 5' nucleases isolated from Thermus~
ayuaticus and Thermus,flavus.
Figure 25A shows the "nibbling" phenomenon detected with the DNAPs of the present invention.
Figure 25B shows that the "nibbling" of Figure 25A is 5' nucleolytic cleavage ald not phosphatase cleavage.
Figure 26 demonstrates that the "nibbling" phenomenon is duplex dependent.
Figure 27 shows an autoradiograph of a gel resolving the products of cleavage reactions run in the presence of either MgCh or MnCI,.
Figure 28 shows an autoradiograph of a gel resolving the products of cleavage reactions run on four similarly sized DNA substrates.
Figure 29 shows an autoradiograph of a gel resolving the products of cleavage reactions run using a wild-type and two mutant tyrosinase gene substrates.
Figure 30 shows an autoradiograph of a gel resolving the products of cleavage reactions run using either a wild-type or mutant tyrosinase substrate varying in length from 157 nucleotides to 1.587 kb.
Figure 31 shows an autoradiograph of a gel resolving the products of cleavage reactions run in various concentrations of MnCl2.
Figure 32 shows an autoradiograph of a gel resolving the products of cleavage reactions run in various concentrations of KCI.
Figure 33 shows an autoradiograph of a gel resolving the products of cleavage reactions run for different lengths of time.
Figure 34 shows an autoradiograph of a gel resolving the products of cleavage reactions run at different temperatures.
Figure 35 shows an autoradiograph of a gel resolving the products of cleavage reactions run using different amounts of the enzyme CleavaseTM BN.
Figure 36 shows an autoradiograph of a gel resolving the products of cleavage reactions run using four different preparations of the DNA substrate.
Figure 37 shows an autoradiograph of a gel resolving the products of cleavage reactions run on either the sense or antisense strand of four different tyrosinase gene substrates.

Figure 38 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a wild-type (3-globin substrate in two different concentrations of KCl and at four different temperatures.
Figure 39 shows an autoradiograph of a gel resolving the products of cleavage reactions run on two different mutant (3-globin substrates in five different concentrations of K
KCI.
Figure 40 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a wild-type and three mutant (3-globin substrates.
Figure 41 shows an autoradiograph of a gel resolving the products of cleavage reactions run on an RNA substrate.
Figure 42 shows an autoradiograph of a gel resolving the products of cleavage reactions run using either the enzyme CleavaseTM BN or Taq DNA polymerise as the 5"
nuclease.
Figure 43 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a double-stranded DNA substrate to demonstrate multiplexing of the cleavage reaction.
Figure 44 shows an autoradiograph of a gel resolving the products of cleavage reactions run on double-stranded DNA substrates consisting of the 419 and 422 mutant alleles derived from exon 4 of the human tyrosinase gene in the presence of various concentrations of MnCI,.
Figure 45 displays two traces representing two channel signals (JOE and FAM
fluorescent dyes) for cleavage fragments derived from a cleavage reaction containing two differently labelled substrates (the wild-type and 422 mutant substrates derived from exon 4 of the tyrosinase gene). The thin lines represent the JOE-labelled wild-type substrate and the thick lines represent the FAM-labelled 422 mutant substrate. Above the tracing is an autoradiograph of a gel resolving the products of cleavage reactions run on double-stranded DNA substrates consisting of the wild-type and 422 mutant alleles derived from exon 4 of the tyrosinase gene.
Figure 46 depicts the nucleotide sequence of six SIV LTR clones corresponding to SEQ ID NOS:63-68. -Figure 47 shows an autoradiograph of a gel resolving the products of cleavage reactions run on six different double-stranded SIV LTR substrates which contained a biotic label on the 5' end of the (-) strand.

Figure 48 shows an autoradiograph of a gel resolving the products of cleavage reactions run on six different double-stranded SIV -LTR substrates which contained a biotin label on the 5' end of the (+) strand.
Figure 49 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in various concentrations of NaCI.
f Figure 50 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in various concentrations of (NH4)~SO4.
Figure 51 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in increasing concentrations of KCI.
Figure 52 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in two concentrations of KCl for various periods of time.
Figure 53 shows an autoradiograph of a gel resolving the products of cleavage reactions run on either the single-stranded or double-stranded form of the same substrate.
Figure 54 shows an autoradiograph of a gel resolving the products of double-stranded I S cleavage reactions run in various concentrations of KCI.
Figure 55 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run in various concentrations of NaCI.
Figure 56 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run in various concentrations of (NH4)~504.
Figure 57 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run for various lengths of time.
Figure 58 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run using various amounts of CleavaseTM BN enzyme for either ~ seconds or 1 minute.
Figure 59 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run at various temperatures.
Figure 60 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run using various amounts of CleavaseTM BN enzyme.
Figure 61 A shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in buffers having various pHs.
Figure 61 B shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in buffers having a pH of either 7.5 or 7.8.
_ 23 -Figure 62A shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run in buffers having a pH of either 8.2 or 7.2.
Figure 62B shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run in buffers having a pH of either 7.5 or 7.8.
Figure 63 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run in the presence of various amounts of human genomic DNA.
Figure 64 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run using the Tfl DNA polymerase in two different concentrations of KCI.
Figure 65 shows an autoradiograph of a-gel- resolving the products of single-stranded cleavage reactions run using the Tth DNA polymerase in two different concentrations of KCI.
Figure 66 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run using the E. coli Exo III enzyme in two different concentrations of KCI.
Figure 67 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run on three different tyrosinase gene substrates (SEQ ID
NOS:34, 41 and 42) using either the Tth DNA polymerase, the E. coli Exo III enzyme or CleavaseTM BN.
Figure 68 is a schematic drawing depicting the location of the ~' and 3"
cleavage sites on a cleavage structure.
Figure 69 shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run on three different tyrosinase gene substrates-(SEQ ID
NOS:34, 41 and 42) using either CleavaseTM BN or the Radl/RadlO complex.
Figure 70 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions run on- a wild-type and two mutant (3-globin substrates.
Figure 71A shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run on a wild-type and three mutant (3-globin substrates.
Figure 71B shows an autoradiograph of a gel resolving the products of single-stranded cleavage reactions run on five mutant (3-globin substrates. -Figure 72 shows an autoradiograph of a gel resolving the products of double-stranded cleavage reactions which varied the order of addition of the reaction components.
Figure 73 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a wild-type and two mutant p53 substrates.
Figure 74 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a wild-type and three mutant p53 substrates.

Figure 75 shows an autoradiograph of a gel resolving the products of cleavage reactions run on a wild-type and a mutant p53 substrate where the mutant and wild-type substrates are present in various concentrations relative to one another.
Figure 76 provides an alignment of HCV clones 1.1 (SEQ ID N0:108), HCV2.1 (SEQ
ID N0:109), HCV3.1 (SEQ ID NO:110), HCV4.2 (SEQ ID NO:111), HCV6.1 (SEQ ID
N0:112) and HCV7.1 (SEQ ID N0:113).
Figure 77 shows a fluoroimager scan of a gel resolving the products of cleavage reactions run on six double-stranded HCV substrates labeled on either the sense or anti-sense strand.
Figure 78 shows an autoradiogram of a gel resolving i:he products of cleavage reactions run on a wild-type and two mutant M. tuberculosis rpoB substrates.
Figure 79A shows a fluoroimager scan of a gel resolving the products of cleavage reactions run on a wild-type and two mutant M. tuberculosis rpoB substrates prepared using either dTTP or dUTP.
Figure 79B shows a fluoroimager scan of the gel shown in Figure 85A following a longer period of electrophoresis.
Figure 80 shows an autoradiogram of a gel resolving 'the products of cleavage reactions run on a wild-type and three mutant M. tuberculosis katG substrates labeled on the sense strand.
Figure 81 shows a fluoroimager scan of a gel resolving the products of cleavage reactions run on a wild-type and three mutant M. tuberculosis katG substrates labeled on the anti-sense strand.
Figure 82 shows the location of primers along the sequence of the E. coli rrsE
gene (SEQ ID N0:145).
Figure 83 provides an alignment of the E. coli rrsE (SEQ ID N0:145), Ccrm.
jcejuni~
(SEQ ID N0:146), and Stp.aureus (SEQ ID N0:147) rRNA genes with the location of consensus PCR rRNA primers indicated .in bold type.
Figure 84 shows a fluoroimager scan of a gel resolving the products of cleavage reactions run on four bacterial 16S rRNA substrates.
Figure 85A shows a fluoroimager scan of a gel resolving the products of cleavage reactions run on five bacterial 16S rRNA substrates.
Figure 85B shows bacterial a fluoroimager scan of a gel resolving the products of cleavage reactions run on five bacterial 16S rRNA substrate s.

Figure 86 shows bacterial a fluoroimager scan of a gel resolving the products of cleavage reactions run on various bacterial 16S i-RNA substrates.-Figure 87 shows bacterial a fluoroimager scan of a gel resolving the products of cleavage reactions run on eight bacterial 16S rRNA substrates.
Figure 88 shows an autoradiogram of a gel resolving the products of cleavage reactions run on a wild-type and mutant tyrosinase gene substrates prepared using naturally occurring deoxynucleotides or deoxynucleotide analogs.
DEFINITIONS
To facilitate understanding of the invention, a number of terms are defined below.
The term "gene" refers to a DNA sequence that comprises control and coding sequences necessary for the production of a polypeptide or precursor. The polypeptide can be encoded by a full length coding sequence or by any portion of the coding sequence so long as the desired enzymatic activity is retained.
The term "wild-type" refers to a gene or gene product which has the characteristics of that gene or gene product when isolated from a naturally occurring source. A
wild-type gene is that which is most frequently observed in a population and is thus arbitrarily designed the "normal" or "wild-type" form of the gene. In contrast, the term "modified" or "mutant" refers to a gene or gene product which displays modifications in sequence and or functional properties (i.c., altered characteristics) when compared to the wild-type gene or gene product.
It is noted that naturally-occurring mutants can be isolated; these are identified by the fact that they have altered characteristics when compared to the wild-type gene or gene product.
The term "recombinant DNA vector" as used herein refers to DNA sequences containing a desired coding sequence and appropriate DNA sequences necessary for the expression of the operably linked coding sequence in a particular host organism. DNA
sequences necessary for expression in procaryotes include a promoter, optionally an operator sequence, a ribosome binding site and possibly other sequences. Eukaryotic cells are known to utilize promoters, polyadenlyation signals and enhancers.
The term "LTR" as used herein refers to the long terminal repeat found at each end of a provirus (i.c., the integrated form of a retrovirus). The LTR contains numerous regulatory signals including transcriptional control elements,-polyadenylation signals and sequences needed for replication and integration of the viral genome. The viral LTR is divided into three regions called U3, R and U5.

The U3 region contains the enhancer and promoter elements. The US region contains the polyadenylation signals. The R (repeat) region separates the U3 and US
regions and transcribed sequences of the R region appear at both the 5' and 3' ends of the viral RNA.
i The term "oligonucleotide" as used herein is defined as a molecule comprised of two or more deoxyribonucleotides or ribonucleotides, preferably more than three, and usually more than ten. The exact size will depend on many factors, which in turn depends on the ultimate function or use of the oligonucleotide. The oligonucleotide may be generated 11 any manner, including chemical synthesis, DNA replication, reverse transcription, or a combination thereof.
Because mononucleotides are reacted to make oligonucleotides in a manner such that the 5' phosphate of one mononucleotide pentose ring is attached to the 3' oxygen of its neighbor in one direction via a phosphodiester linkage, an end of an oligonucleotide is referred to as the "5' end" if its 5' phosphate is not linked to the 3' oxygen of a mononucleotide pentose ring and as the "3' end" if its 3' oxygen is not linked to a 5' phosphate of a subsequent mononucleotide pentose ring. As used herein, a nucleic acid sequence, even if internal to a larger oligonucleotide, also may be said to have 5' and 3' ends.
When two different, non-overlapping oligonucleotides anneal to different regions of the same linear complementary nucleic acid sequence, and the 3' end of one oligonucleotide points towards the 5' end of the other, the former may be called the "upstream"
oligonucleotide and the latter the "downstream" oligonucleotide.
The term "primer" refers to an oligonucleotide which is capable of acting as a point of initiation of synthesis when placed under conditions in which primer extension is initiated.
An oligonucleotide "primer" may .occur naturally, as in a purified restriction digest or may be produced synthetically.
A primer is selected to be "substantially" complementary to a strand of specific sequence of the template. A primer must be sufficiently complementary to hybridize with a template strand for primer elongation to oc-cur. A primer sequence need not reflect the exact sequence of the template. For example, a non-complementary nucleotide fragment may be attached to the 5' end of the primer, with the remainder of the primer sequence being substantially complementary to the strand. Non-complementary bases or longer sequences can be interspersed into the primer, provided that the primer sequence has sufficient complementarity with the sequence of the template to hybridize and thereby form a template primer complex for synthesis of the extension product of the primer.

"Hybridization" methods involve the annealing of a complementary sequence to the target nucleic acid (the sequence to be detected). The ability of two polymers of nucleic acid containing complementary sequences to find each other and anneal through base pairing interaction is a well-recognized phenomenon. The initial observations of the "hybridization"
process by Marmur and Lane, Proc. Natl. Acad. Sci. USA 46:453 (1960) and Doty et al., Proc. Natl. Acad. Sci. USA 46:461 ( 1960) have been followed by the refinement of this process into an essential tool of modern biology. Nonetheless, a number of problems have prevented the wide scale use of hybridization as a tool in human diagnostics.
Among the more formidable problems are: 1 ) the inefficiency of hybridization; 2) the low concentration of specific target sequences in a mixture of genomic DNA; and 3) the hybridization of only partially complementary probes and targets.
With regard to efficiency, it is experimentally observed that only a fraction of the possible number of probe-target complexes are formed in a hybridization reaction. This is particularly true with short oligonucleotide probes (less than 100 bases in length). There are three fundamental causes: a) hybridization cannot occur because of secondary and tertiary structure interactions; b) strands of DNA containing the target sequence have rehybridized (reannealed) to their complementary strand; and c) some target molecules are prevented from hybridization when they are used in hybridization formats that immobilize the target nucleic acids to a solid surface.
Even where the sequence of a probe is completely complementary to the sequence of the target, i.c~., the target's primary structure, the target sequence must be made accessible to the probe via rearrangements of higher-order structure. These higher-order structural rearrangements may concern either the secondary structure or tertiary structure of the molecule. Secondary structure is determined by intramolecular bonding. In the case of DNA
or RNA targets this consists of hybridization within a single, continuous strand of bases (as opposed to hybridization between two different -strands). Depending on the extent and position of intramolecular bonding, the probe can be displaced from the target sequence preventing hybridization.
Solution hybridization of oligonucleotide probes to denatured double-stranded DNA is further complicated by the fact that the longer complementary target strands can renature or reanneal. Again, hybridized probe is displaced by this process. This results in a low yield of hybridization (low "coverage") relative to the starting concentrations of probe and target.

With regard to low target sequence concentration, the DNA fragment containing the target sequence is usually in relatively low abundance in genomic DNA. This presents great technical difficulties; most conventional methods that use oligonucleotide probes lack the sensitivity necessary to detect hybridization at such low levels.
r 5 One attempt at a solution to the target sequence concentration problem is the amplification of the detection signal. Most often this entails placing one or more labels on an oligonucleotide probe. In the case of non-radioactive labels, even the highest affinity reagents have been found to be unsuitable for the detection of single copy genes in genomic DNA with oligonucleotide probes. See Wallace et al., Biochimie 67:755 (1985). In the case of radioactive oligonucleotide probes, only extremely high specific activities are found to show satisfactory results. See Studencki and Wallace, DNA 3:1 (1984) and Studenclci et al., Human Genetics 37:42 (1985).
With regard to complementarity, it is important for some diagnostic applications to determine whether the hybridization represents complete or partial complementarity. For example, where it is desired to detect simply the presence or absence of pathogen DNA (such as from a virus, bacterium, fungi, mycoplasma, protozoan) it is only important that the hybridization method ensures hybridization when the relevant sequence is present; conditions can be selected where both partially complementary probes and completely complementary probes will hybridize. Other diagnostic applications, however, may require that the hybridization method distinguish between partial and complete complementarity.
It may be of interest to detect genetic polymorphisms. For example, human hemoglobin is composed, in part, of four polypeptide chains. Two- of these chains are identical chains of 141 amino acids (alpha chains) and two of these chains are identical chains of 146 amino acids (beta chains).
The gene encoding the beta chain is known to exhibit polymorphism. The normal allele 2~ encodes a beta chain having glutamic acid at the sixth position. The mutant allele encodes a beta chain having valine at the sixth position. This difference in amino acids has a profound (most profound when the individual is homozygous for the mutant allele) physiological impact known clinically as sickle cell anemia. It is well known that the genetic basis of the amino acid change involves a single base difference between the normal allele DNA
sequence and the mutant allele DNA sequence.

WO 96/15267 PG"f/LTS95/14673 Unless combined with other techniques (such as restriction enzyme analysis), methods that allow for the same level of hybridization in the case of both partial as well as complete complementarity are typically unsuited for such applications: the probe will hybridize to both the normal and variant target sequence. Hybridization, regardless of the method used, requires some degree of complementarity between the sequence being assayed (the target sequence) and the fragment of DNA used to perform the test (the probe). (Of course, one can obtain binding without any complementarity but this binding is nonspecific and to be avoided.) The complement of a nucleic acid sequence as used herein refers to an oligonucleotide which, when aligned with the nucleic acid sequence such that the 5' end of one sequence is paired with the 3' end of the other, is in "antiparallel association." Certain bases not commonly found in natural nucleic acids may be included in the nucleic acids of the present invention and include, for example, inosine and 7-deazaguanine.
Complementarity need not be perfect; stable duplexes may contain mismatched base pairs or unmatched bases. Those 1 S skilled in the art of nucleic acid technology can determine duplex stability empirically considering a number of variables including, for example, the length of the oligonucleotide, base composition and sequence of the oligonucleotide, ionic strength and incidence of mismatched base pairs.
Stability of a nucleic acid duplex is measured by the melting temperature, or "Tn,."
The Tm of a particular nucleic acid duplex under specified conditions is the temperature at which on average half of the base pairs have disassociated.
The term "probe" as used herein refers to a labeled oligonucleotide which forms a duplex structure with a sequence in another nucleic acid, due to complementarily of at least one sequence in the probe with a sequence in the other nucleic acid.
The term "label" as used herein refers to any atom or molecule which can be used to provide a detectable (preferably quantifiable) signal, and which can be attached to a nucleic acid or protein. Labels may provide signals detectable by fluorescence, radioactivity, colorimetry, gravimetry, X-ray diffraction or absorption, magnetism, enzymatic activity, and the like. -The term "cleavage structure" as used herein. refers to a region of a single-stranded nucleic acid substrate containing secondary structure, said region being cleavable by a cleavage means, including but not limited to an enzyme. The cleavage structure is a substrate for specific cleavage by said cleavage means in contrast to a nucleic acid molecule which is a substrate for non-specific cleavage by agents such as phosphodiesterases which cleave nucleic acid molecules without regard to secondary structure (i.e., no folding of the substrate is required).
The term "cleavage means" as used herein refers to any means which is capable of S cleaving a cleavage structure, including but not limited to enzymes. The cleavage means may include native DNAPs having 5' nuclease activity (e.g., Taq DNA polymerise, E.
coli DNA
polymerise I) and, more specifically, modified DNAPs having 5' nuclease but lacking synthetic activity. The ability of 5' nucleases to cleave naturally occurring structures in nucleic acid templates (structure-specific cleavage) is useful to detect internal sequence differences in nucleic acids without prior knowledge of the specific sequence of the nucleic acid. In this manner, they are structure-specific enzymes. Structure-specific enzymes are enzymes which recognize specific secondary structures in a nucleic molecule and cleave these structures. The site of cleavage may be on either the 5' or 3' side of the cleavage structure;
alternatively the site of cleavage may be between the 5' and 3' side (i. e. , within or internal to) of the cleavage structure. The cleavage means of the invention cleave a nucleic acid molecule in response to the formation of cleavage structures; it is not necessary that the cleavage means cleave the cleavage structure at any particular location within the cleavage structure.
The cleavage means is not restricted to enzymes having 5' nuclease activity.
The cleavage means may include nuclease activity provided from a variety of sources including the enzyme CleavaseTM, Taq DNA polymerise, E. coli DNA polymerise I and eukaryotic structure-specific endonucleases, marine FEN-1 endonucleases [Harrington and Liener, ( 1994) Genes and Develop. 8:1344] and calf thymus 5' to 3' exonuclease [Murante, R.S., et al.
(1994) J. Biol. Chem. 269:1191]). In addition, enzymes having 3' nuclease activity such as members of the family of DNA repair endonucleases (e.g., the Rrpl enzyme from Drosophila melcrnogaster, the yeast RAD1/RAD10 complex and E. coli Exo III), are also suitable cleavage means for the practice of the methods of the invemion.
The term "cleavage products" as used herein, refers to products generated by the reaction of a cleavage means with a cleavage structure (i.e., the treatment of a cleavage structure with a cleavage means).
The terms "nucleic acid substrate" and nucleic acid template" are used herein interchangeably and refer to a nucleic acid molecule which 'when denatured and allowed to renature (i. e. , to fold upon itself by the formation of intra-strand hydrogen bonds), forms at WO 96/15267 PG"T/US95114673 least one cleavage structure. The nucleic acid substrate may comprise single-or double-stranded DNA or RNA.
The term "substantially single-stranded" when used in reference to a nucleic~acid substrate means that the substrate molecule exists primarily as a single strand of nucleic acid in contrast to a double-stranded substrate which exists as two strands of nucleic acid which are held together by inter-strand base pairing interactions.
Nucleic acids form secondary structures which depend on base-pairing for stability.
When single strands of nucleic acids (single-stranded DNA, denatured double-stranded DNA
or RNA) with different sequences, even closely related ones, are allowed to fold on themselves, they assume characteristic secondary structures. At "elevated temperatures" the duplex regions of the structures are brought to the brink of instability, so that the effects of small changes in sequence are maximized, and revealed as alterations in the cleavage pattern.
In other words, "an elevated temperature" is a temperature at which a given duplex region of the folded substrate molecule is near the temperature at which that duplex melts. An alteration in the sequence of the substrate will then be likely to cause the destruction of a duplex regions) thereby generating a different cleavage pattern when a cleavage agent which is dependent upon the recognition of structure is utilized in the reaction.
While not being limited to any particular theory, it is thought that individual molecules in the target (i.c., the substrate) population may each assume only one or a few of the potential cleavage structures (i.e., duplexed regions), but when the sample is. analyzed as a whole, a composite pattern representing all cleavage sites is detected. Many of the structures recognized as active cleavage sites are likely to be only a few base-pairs long and would appear to be unstable when elevated temperatures used in the cleavage reaction. Nevertheless;
transient formation of these structures allows recognition and cleavage of these structures by said cleavage means.
The formation or disruption of these structures in response to small sequence changes results in changes in the patterns of cleavage. Temperatures in the range of 40-85°C. with the range of 55-85°C being particularly preferred, are suitable elevated temperatures for the practice of the method of the invention.
The term "sequence variation" as used herein refers to differences in nucleic,acid sequence between two nucleic acid templates. For example, a wild-type structural gene and a mutant form of this wild-type structural gene may vary in sequence by the presence of single base substitutions and/or deletions or insertions of one or more nucleotides.
These two forms of the structural gene are said to vary in sequence from one another. A second mutant form of the structural gene may exits. This second mutant form is said to vary in sequence from both the wild-type gene and the first mutant form of the gene. It is noted, however, that the invention does not require that a comparison be made between one or more forms of a gene to detect sequence variations. Because the method of the invention generates .a characteristic and reproducible pattern of cleavage products for a given nucleic acid substrate. a characteristic "fingerprint" may be obtained from any nucleic substrate without reference to a wild-type or other control. The invention contemplates the use of the method for both "fingerprinting" nucleic acids without reference to a control and identification of mutant forms of a substrate nucleic acid by comparison of the mutant form of the substrate with a wild-type or known mutant control.
The term "liberating" as used herein refers to the release of a nucleic acid fragment from a larger nucleic acid fragment, such as an oligonucleotide, by the action of a 5" nuclease such that the released fragment is no longer covalently attached to the remainder of the oligonucleotide.
The term "substrate strand" as used herein, means that strand of nucleic acid in a cleavage structure in which the cleavage mediated by the 5' nuclease activity occurs.
The term "template strand" as used herein, means that strand of nucleic acid in a cleavage structure which is at least partially complementary to the substrate strand and which anneals to the substrate strand to form the cleavage structure.
The term "K"," as used herein refers to the Michaelis-Menten constant for an enzyme and is defined as the concentration of the specific substrate ai: which a given enzyme yields one-half its maximum velocity in an enzyme catalyzed reaction.
The term "nucleotide analog" as used herein refers to modified or non-naturally occurring nucleotides such as 7-deaza purines (i. e., 7-deaza-dATP and 7-deaza-dGTP).
Nucleotide analogs include base analogs and comprise modified forms of deoxyribonucleotides as well as ribonucleotides. As used herein the term "nucleotide analog"
when used in reference to substrates present in a nucleic acid amplification mixture (e.g., a PCR mixture) refers to the use of nucleotides other than dAT'P, dGTP, dCTP and dTTP; thus, the use of dUTP (a naturally occurring dNTP) in a PCR would comprise the use of a nucleotide analog in the PCR. A PCR product generated using dUTP. 7-deaza-dATP, 7-deaza-dGTP or any other nucleotide analog in the reaction mixture is said to contain nucleotide analogs.
_»_ "Oligonucleotide primers matching or complementary to a gene sequence" refers to oligonucleotide primers capable of facilitating the template-dependent synthesis of single or double-stranded nucleic acids. Oligonucleotide primers matching or complementary to a gene sequence may be used in PCRs, RT-PCRs and the like.
A "consensus gene sequence" refers to a gene sequence which is derived by comparison of two or more gene sequences arid which describes the nucleotides most often present in a given segment of the genes; the consensus sequence is the canonical sequence.
The term "polymorphic locus" is a locus present in a population which shows variation between members of the population (i. e., the most common allele has a frequency of less than 0.95). In contrast, a "monomorphic locus" is a genetic locus at little or no variations seen between members of the population (generally taken to be a locus at which the most common allele exceeds a frequency of 0.9~ in the gene pool of the population).
The term "microorganism" as used herein means an organism too small to be observed with the unaided eye and includes, but is not limited to bacteria, virus, protozoans, fungi, and ciliates.
The term "microbial gene sequences" refers to gene sequences derived from a microorganism. -The term "sequences derived from one or more microorganisms" refers to nucleic acid sequences extracted from one or a mixture of more than one microorganism. The extracted sequences may be subjected to further treatment, such as nucleic acid amplification (e.g., polymerase chain reaction) prior to treatment to form and subsequently cleave cleavage structures comprising the microbial nucleic acid sequences.
The term "bacteria" refers to any bacterial species including eubacterial and archaebacterial species.
The term "virus" refers to obligate, ultramicroscopic, intracellular parasites incapable of autonomous replication (i.e., replication requires the use of the host cell's machinery).
The term "mufti-drug resistant" or multiple-drug resistant" refers to a microor~~anism which is resistant to more than one of the antibiotics or antimicrobial agents used in the treatment of said microorganism.
The term "CFLPTM (CleavaseTM Fragment Length Polymorphism) analysis" as used herein refers to analysis, often by electrophoresis, of the products of a reaction in which strands of nucleic acid are i) denatured; ii) cooled, or otherwise allowed to form intra-strand secondary structures and iii) cleaved with a cleavage agent which recognizes and cleaves the WO 96!15267 PCT/US95/14673 nucleic acid having intra-strand secondary structure in response to said structures, thereby creating a collection of fragments (i. e., cleavage products) that are characteristic of the nucleic acid substrate. Nucleic acid substrates which differ in sequence from a control or reference nucleic acid substrate will, when analyzed by this method, show an altered representation of the fragments within the pool of cleavage products, indicating the presence of said differences , in sequence between the substrates.
DESCRIPTION OF THE INVENTION
The present invention relates to methods and compositions for treating nucleic acid, and in particular, methods and compositions for detection and characterization of nucleic acid sequences and sequence changes.
The present invention relates to means for cleaving a nucleic acid cleavage structure in a site-specific manner. In particular, the present invention relates to a cleaving enzyme having 5' nuclease activity without interfering nucleic acid synthetic ability.
1 ~ This invention provides 5' nucleases derived from thermostable DNA
polymerises which exhibit altered DNA synthetic activity from that of native thermostable DNA
polymerises. The 5' nuclease activity of the polymerise is retained while the synthetic activity is reduced or absent. Such 5' nucleases are capable of catalyzing the structure-specific cleavage of nucleic acids in the absence of interfering synthetic activity. The lack of synthetic activity during a cleavage reaction results in nucleic; acid cleavage products of uniform size.
The novel properties of the polymerises of the invention form the basis of a method of detecting specific nucleic acid sequences. This method relies upon the amplification of the detection molecule rather than upon the amplification of the target sequence itself as do existing methods of detecting specific target sequences.
DNA polymerises (DNAPs), such as those isolated from E. coli or from thermophilic bacteria of the genus Thermus, are enzymes that synthesize n.ew DNA strands.
Several of the known DNAPs contain associated nuclease activities in addition to the synthetic activity of the enzyme.
Some DNAPs are known to remove nucleotides from the 5' and 3' ends of DNA
chains jKornberg, DNA Replication, W.H. Freeman and Co., San Francisco, pp.

( 1980)]. These nuclease activities are usually referred to as .5' exonuclease and 3' exonuclease activities, respectively. For example, the 5' exonuclease activity located in the 5_ N-terminal domain of several DNAPs participates in the removal of RNA primers during lagging strand synthesis during DNA replication and the removal of damaged nucleotides during repair. Some DNAPs, such as the E. coli DNA polymerase (DNAPEc 1 ).
also have a 3' exonuclease activity responsible for proof reading during DNA synthesis (Kornberg, supra).
A DNAP isolated from Thermus aquaticus, termed Taq DNA polymerase (DNAPTaq), has a 5' exonuclease activity, but lacks a functional 3' exonucleolytic domain [Tindall and Kunkell, Biochem. 27:6008 (1988)]. Derivatives of DNAPEcI and DNAPTaq, respectively called the Klenow and Stoffel fragments, lack 5' exonuclease domains as a result of enzymatic or genetic manipulations [Brutlag et al., Biocherrz. Biophys. Re.s~.
Commun. 37:982 ( 1969); Erlich et al., Science 252:1643 ( 1991 ); Setlow and Kornberg, J.
Biol. ChenZ. 247:232 (1972)].
The 5" exonuclease activity of DNAPTaq was reported to require concurrent synthesis [Gelfand; PCR Technology - Principles and Applications,for- DNA Amplifrcatior7 (H.A. Erlich, Ed.), Stockton Press, New York, p. 19 (1989)]. Although mononucleotides predominate among the digestion products of the 5' exonucleases of DNAPTaq and DNAPEc 1, short oligonucleotides (_< 12 nucleotides) can also be observed implying that these so-called 5' exonucleases can function endonucleolytically [Setlow, .supoa; Holland et al., Pf~oc. Natl.
Acac~ Sci. USA 88:7276 (1991)].
In WO 92/06200, Gelfand et al. show that the preferred. substrate of the,5' exonuclease activity of the thermostable DNA polymerases is displaced single-stranded DNA.
Hydrolysis of the phosphodiester bond occurs between the displaced single-stranded DNA and the double-helical DNA with the preferred exonuclease cleavage site being a phosphodiester bond in the double helical region. Thus, the 5' exonuclease activity usually associated with DNAPs is a structure-dependent single-stranded endonuclease and is more properly referred to as a 5' nuclease. Exonucleases are enzymes which cleave nucleotide molecules from the ends of the nucleic acid molecule. Endonucleases, on the other hand, are enzymes which cleave the nucleic acid molecule at internal rather than terminal sites.- The nuclease activity associated with some thermostable DNA polymerases cleaves endonucleolytically but this cleavage requires contact with the 5' end of the molecule being cleaved.
Therefore, these nucleases are referred to as 5' nucleases.
When a 5' nuclease activity is associated with a eubacterial Type A DNA
polymerase.
it is found in the one-third N-terminal region of the protein as an independent functional WO 96/15267 PC"T/US95/14673 domain. The C-terminal two-thirds of the molecule constitute the polymerization domain which is responsible for the synthesis of DNA. Some Type A DNA polymerases also have a 3' exonuclease activity associated with the two-third. C-terminal region of the molecule.
The 5' exonuclease activity and the polymerization activity of DNAPs have been separated by proteolytic cleavage or genetic manipulation of the polymerase molecule. To r date thermostable DNAPs have been modified to remove or reduce the amount of 5' nuclease activity while leaving the polymerase activity intact.
The Klenow or large proteolytic cleavage fragment of DNAPEc 1 contains the polymerase and 3' exonuclease activity but lacks the 5' nuclease activity. The Stoffel fragment of DNAPTaq (DNAPStf) lacks the 5' nuclease activity due to a genetic manipulation which deleted the N-terminal 289 amino acids of the polymerase molecule [Erlich et al., Science 252:1643 ( 1991 )]. WO 92/06200 describes a thermostable DNAP with an altered level of 5' to 3' exonuclease. U.S. Patent No. 5,108,892 describes a Thermu.s czquaticus DNAP without a 5' to 3' exonuclease. However, the art of molecular biology lacks a thermostable DNA polymerase with a lessened amount of synthetic activity.
The present invention provides 5' nucleases derived from thermostable Type A
DNA
polymerases that retain 5' nuclease activity but have reduced or absent synthetic activity. The ability to uncouple the synthetic activity of the enzyme from the 5' nuclease activity proves that the 5' nuclease activity does not require concurrent DNA synthesis as was previously reported (Gelfand, PCR Technology, supra).
The description of the invention is divided into: I. Generation of 5' Nucleases Derived From Thermostable DNA Polymerases; II. CleavaseTM Fragment Length Polymorphism for the Detection of Secondary Structure; III. Detection of Mutations in the p53 Tumor Suppressor Gene Using the CFLPTM Method andTV.~etection and_Tdentificat,'_on of Pathogens Using the CFLPTM Method.
I. Generation Of 5' Nucleases From Thermostable DNA Polymerases The methods of the present invention employ 5' nucleases for the detection of specific nucleic acid sequences. The 5' nuclease may be derived from a thermostable DNA
polymerase; however, the methods of the invention are not limited to the use of a 5" nuclease, any cleavage agent capable of generating a unique (i. e.. characteristic) pattern of cleavage products from a substrate nucleic acid may be employed. When a 5' nuclease is to be employed, the 5' nuclease may be derived from a thermostable DNA polymerase as described below.
The genes encoding Type A DNA polymerases share about 85°l°
homology to each other on the DNA sequence level. Preferred examples of thermostable polymerases include those isolated from Thermus aquaticus, Thermus flavus, and Thermus thermophilus. However,-other thermostable Type A polymerases which have 5' nuclease activity are also suitable.
Figures 1 and 2 compare the nucleotide and amino acid sequences of the three above mentioned polymerases. In Figures 1 and 2, the consensus or majority sequence derived from a comparison of the nucleotide (Fig. 1 ) or amino acid (Fig. 2) sequence of the three thermostable DNA polymerases is shown on the top line. A dot appears in the sequences of each of these three polymerases whenever an amino acid residue in a given sequence is identical to that contained in the consensus amino acid sequence. Dashes are used to introduce gaps in order to maximize alignment between the displayed sequences.
When no consensus nucleotide or amino acid is present at a given position, an "X" is placed in the consensus sequence. SEQ ID NOS:1-3 display the nucleotide sequences and SEQ ID
NOS:4-6 display the amino acid sequences of the three wild-type polymerases. SEQ ID
NO:l corresponds to the nucleic acid sequence of the wild type Thermus aquatica~s DNA
polymerase gene isolated from the YT-1 strain [Lawyer et al., .I. Bivl. Chem.
264:6427 (1989)]. SEQ ID N0:2 corresponds to the nucleic acid sequence of the wild type Thel"n22fs .flavus DNA polymerase gene [Akhmetzjanov and Vahhitov, Nucl. Acids Res.
20:5839 (1992)].
SEQ ID N0:3 corresponds to the nucleic acid sequence of the wild type Thermu.s thermophilus DNA polymerase gene jGelfand et al.. WO 91/09950 (1991)]. SEQ ID
NOS:7-8 depict the consensus nucleotide and amino acid sequences, respectively for the above three DNAPs (also shown on the top row in Figs. 1 and 2).
The 5' nucleases o f the invention derived from thermostable polymerases have reduced synthetic ability, but retain substantially the same 5' exonuclease activity as the native DNA
polymerase. The term "substantially the same 5' nuclease activity" as used herein means that the 5' nuclease activity of the modified enzyme retains the ability to function as a structure-dependent single-stranded endonuclease but not necessarily at the same rate of cleavage as - compared to the unmodified enzyme. Type A DNA polymerases may also be modified so as to produce an enzyme which has increases 5' nuclease activity while having a reduced level of synthetic activity. Modified enzymes having reduced synthetic activity and increased 5' nuclease activity are also envisioned by the present invention.

By the term "reduced synthetic activity" as used herein it is meant that the modified enzyme has less than the level of synthetic activity found in the unmodified or "native"
enzyme. The modified enzyme may have no synthetic activity remaining or may have that level of synthetic activity that will not interfere with the use of the modified enzyme in the detection assay described below. The 5' nucleases of the present invention are advantageous in situations where the cleavage activity of the polymerise is desired, but the synthetic ability is not (such as in the detection assay of the invention).
As noted above, it is not intended that the invention be limited by the nature of the alteration necessary to render the polymerise synthesis deficient. The present invention contemplates a variety of methods, including but not limited to: 1) proteolysis; '?) recombinant constructs (including mutants); and 3) physical and/or chemical modification and/or inhibition.
1. Proteolysis Thermostable DNA polymerises having a reduced level of synthetic activity are produced by physically cleaving the unmodified enzyme with proteolytic enzymes to produce fragments of the enzyme that are deficient in synthetic activity but retain 5' nuclease activity.
Following proteolytic digestion, the resulting fragments are separated by standard chromatographic techniques and assayed for the ability to synthesize DNA and to act as a 5' nuclease. The assays to determine synthetic activity and 5' nuclease activity are described below.
2. Recombinant Constructs The examples below describe a preferred method for creating a construct encoding a 5' nuclease derived from a thermostable DNA polymerise. As the Type A DNA
polymerises are similar in DNA sequence; the cloning strategies employed for the Thermus aquaticu.s and .flavus polymerises are applicable to other thermostable Type A polymerises.
In general. a thermostable DNA polymerise is cloned by isolating genomic DNA using molecular biological methods from a bacteria containing a thermostable Type A DNA
polymerise. This genomic DNA is exposed to primers which are capable of amplifying the polymerise gene bS~
PCR. __ This amplified polymerise sequence is then subjected to standard deletion processes to delete the polymerise portion of the gene. Suitable deletion processes are described below in the examples.
. The example below discusses the strategy used to determine which portions of the DNAPTaq polymerise domain could be removed without eliminating the 5' nuclease activity.
Deletion of amino acids from the protein can be done either by deletion of the encoding genetic material, or by introduction of a translational stop codon by mutation or frame shift.
In addition, proteolytic treatment of the protein molecule can be performed to remove segments of the protein.
In the examples below, specific alterations of the Taq gene were: a deletion between nucleotides 1601 and 2502 (the end of the coding region), a 4 nucleotide insertion at position 2043, and deletions between nucleotides 1614 and 1848 and between nucleotides 875 and 1778 (numbering is as in SEQ ID NO:1 ). These modified sequences are described below in the examples and at SEQ ID NOS:9-12.
Those skilled in the art understand that single base pair changes can be innocuous in terms of enzyme structure -and function. Similarly, small additions and deletions can be present without substantially changing the exonuclease or polymerise function of these enzymes.
Other deletions are also suitable to create the 5' nucleases of the present invention. It is preferable that the deletion decrease the polymerise activity of the 5' nucleases to a level at which synthetic activity will not interfere with the use of the 5' nuclease in the detection assay of the invention. Most preferably, the synthetic ability is absent.
Modified polymerises are tested for the presence of synthetic and 5' nuclease activity as in assays described below.
Thoughtful consideration of these assays allows for the screening of candidate enzymes whose structure is heretofore as- yet unknown. In other words, construct "X" can be evaluated according to the protocol described below to determine whether it is a member of the genus of 5' nucleases of the present invention as defined functionally, rather than structurally.
In the example below, the PCR product of the amplified Thermtrs aquaticus genomic DNA did not have the identical nucleotide structure of the native genomic DNA
and did not have the same synthetic ability of the original clone. Base pair changes which result due to the infidelity of DNAPTaq during PCR amplification of a polymerise gene are also a method by which the synthetic ability of a polymerise gene may be inactivated. The examples below and Figs. 4A and SA indicate regions in the native Thermos ayuaticZrs and.flcnurs DNA

polymerises likely to be important for synthetic ability. There are other base pair changes and substitutions that will likely also inactivate the polymerise.
It is not necessary, however, that one start out the process of producing a 5' nuclease from a DNA polymerise with such a mutated amplified product. This is the .method by which the examples below were performed to generate the synthesis-deficient DNAPTay mutants, but it is understood by those skilled in the art that a wild-type DNA
polymerise sequence may be used as the starting material for the introduction of deletions, insertion and substitutions to produce a 5' nuclease. For example, to generate the synthesis-deficient DNAPTfI mutant, the primers listed in SEQ ID NOS:13-14 were used to amplify the wild type DNA polymerise gene from Thermus flavus strain AT-62. The amplified polymerise gene was then subjected to restriction enzyme digestion to delete a large portion of the domain encoding the synthetic activity.
The present invention contemplates that the nucleic acid construct of the present invention be capable of expression in a suitable host. Those in the art know methods for attaching various promoters and 3' sequences to a gene structure to achieve efficient expression. The examples below disclose two suitable vectors and six suitable vector constructs. Of course, there are other promoter/vector combinations that would be suitable. It is not necessary that a host organism be used for the expression of the nucleic acid constructs of the invention. For example, expression of the protein encoded by a nucleic acid construct may be achieved through the use of a cell-free in vitro trap scription/translation system. An example of such a cell-free system is the commercially available TnTTM Coupled Reticulocyte Lysate System (Prorriega Corporation, Madison, WI).
Once a suitable nucleic acid construct has been made°_, the 5' nuclease may be produced from the construct. The examples below and standard molecular biological teachings enable one to manipulate the construct by different suitable methods.
Once the 5' nuclease has been expressed, the polymerise is tested for both synthetic and nuclease activity as described below.
3. Physical And/Or Chemical Modification And/Or Inhibition The synthetic activity of a thermostable DNA polymerise may be reduced by chemical and/or physical means. In one embodiment, the cleavage reaction catalyzed by the 5' nuclease-activity of the polymerise is run under conditions which preferentially inhibit the synthetic activity of the polymerise. The level of synthetic activity need only be reduced to that level of activity which does not interfere with cleavage reactions requiring no significant synthetic activity.
As shown in the examples below. concentrations of M~_-- greater than ~ mNl inhibit the polymerization activity of the native DNAPTa~I. The ability of the ~' nuclease to function under conditions where synthetic activity is inhibited is tested by running the assays for synthetic and 5' nuclease activity. described below, in the presence of a range of M~~-concentrations (~ to 10 mM). The effect of a given concentration of M~_~~ is determined by quantitation of the amount of synthesis and cleavage in the test reaction as compared to the standard reaction for each assay.
The inhibitory effect of other ions, polyamines. denaturants. such as urea.
formamide.
dimethvlsulfoxide. glycerol and non-ionic detergents (Triton X-100*and Tween-?0). nucleic acid binding chemicals such as. actinomvcin D, ethidium bromide and psoralens.
are tested b~~
their addition to the standard reaction buffers for the synthesis and ~' nuclease assays. Those compounds havin~_ a preferential inhibitory effect on the synthetic activity of a thermostable I ~ polymerase are then used to create reaction conditions under which ~' nuclease activim (cleava~_e) is retained while synthetic activity is reduced or eliminated.
Physical means may be used to preferentially inhibit the synthetic actiyim of a polymerase. For example. the synthetic activity of thermostable polvmerases is destrwed by etposure of the polymerase to extreme heat (typically 96 to 100°C) for extended periods of time ~~reater than or equal to ?0 minutes). While these are minor differences with respect to the specific heat tolerance for each of the enzymes. these are readily determined. Polymerases are treated with heat for various periods of time and the effect of the heat treatment upon the synthetic and ~' nuclease activities is determined.
II. CleavaseT"' Fragment Length Polymorphism For The Detection Of Secondan~ Structure Nucleic acids assume secondary structures which depend on base-pairin~~ for stabilim.
When single strands of nucleic acids (single-stranded DNA. denatured DNA or R'vA) with different sequences. even closely related ones. are allowed to fold on themselves. they assume characteristic secondary structures. These differences in structures account for the ability of sin'le strand conformation polymorphism (SSCP) analysis to distinguish between DNA
fragments having closely related sequences.
* Trade-mark _ :1~ _ The 5' nuclease domains of certain DNA polymerases are specific endonucleases that recognize and cleave nucleic acids at specific structures rather than in a sequence-specific manner (as do restriction endonucleases). The isolated nuclease domain of DNAPTaq described herein (termed the enzyme CleavaseTM) recognizes the end of a duplex that has non-base paired strands at the ends. The strand with the ~' end is cleaved at the junction between the single strand and the duplex.
Figure 3 depicts a wild-type substrate and a mutant substrate wherein the mutant substrate differs from the wild-type by a single base change (A to G as indicated). According to the method of the present invention, substrate structures form when nucleic acids are denatured and allowed to fold on themselves (See Figure 3, steps 1 and 2). The step of denaturation may be achieved by treating the nucleic acid with heat, low (<3) or high pH
(>10), the use of low salt concentrations, the absence of cations, chemicals (e.g., urea, formamide) or proteins (e.g., helicases). Folding or renaturation of the nucleic acid is achieved by lowering of the temperature, addition of salt, neutralization of the pH, withdrawal of the chemicals or proteins.
The manner in which the substrate folds is dependent. upon the sequence of the substrate. The 5' nucleases of the invention cleave the structures (See Figure 3, step 3). The end points of the resulting fragments reflect the locations of the cleavage sites. The cleavage itself is dependent upon the formation of a particular structure, not upon a particular sequence -at the cleavage site.
When the 5' nucleases of the invention cleave a nucleic acid substrate, a collection of cleavage products or fragments is generated. These fragments constitute a characteristic fingerprint of the nucleic acid which can be detected (e.g., by electrophoresis on a gel (see-step 4)J. Changes in the sequence of a nucleic acid (e.g., single point mutation between a wild-type and mutant gene) alter the pattern of cleavage structures formed.
When the 5' nucleases of the invention cleave the structures formed by a wild-type and an altered or mutant form of the substrate, the distribution of the cleavage fragments generated will differ between the two substrates reflecting the difference in the sequence of the two substrates (See Figure 3, step 5): -The CleavaseTM enzyme generates a unique pattern of cleavage products for a substrate nucleic acid. Digestion with the CleavaseTM enzyme can be used to detect single base changes in DNA molecules of great length (e.g., 1.6 kb in length) to produce a characteristic pattern of cleavage products. The method of the invention is termed "CleavaseTM Fragment Length Polymorphism" (CFLPTM). However. it is noted that the invention is not limited to the use of the enzyme CleavaseT'': suitable enzymatic cleavage activity may be provided from a variety of sources including the CleavaseT"' enzyme. Taq DNA polymerase. F.
coli DNA
polymerase 1 and eukaryotic structure-specific endonucleases (~~.~.. the yeast R4D? protein and RAD 1 /R.AD 10 complex [Harrington. J.J. and Liener ( 1994) Genes and Develop. 8:1 s44].
murine FEN-1 endonucleases (Harrington and Liener. .supra) and calf thymus ~' to 3' exonuclease [Murante, R.S.. et al. (1994) J. Biol. Chem. 269:1191]). Indeed actual experimental data is provided herein which demonstrates that numerous enzymes may be used to generate a unique pattern of cleavage products for a substrate nucleic acid. Enzymes which are shown herein to be suitable for use in the CFLPT"' method include the CleayaseT"' BN
enzyme. Tuy DNA polymerase. T~h DNA polvmerase. Tn DNA polvmerase. E. roll Exo Ill.
and the yeast Radl/RadlO complex.
The invention demonstrates that numerous enzymes may be suitable for use in the CFLPT" method includine enzymes which have been characterized in the literature a bein~_ ~' 1 ~ exonucleases. In order to test whether an enzyme is suitable for use as a cleava~_e means in the CFLPT'" method (i.c~.. capable of generating a unique pattern of cleavage products for a substrate nucleic acid). the following steps are taken. Careful consideration of the steps described below allows the evaluation of any enzyme ("enzyme X") for use in the CFLPT~' method.
?0 .An initial CFLPTM reaction is prepared using a previously characterized substrate nucleic acid [for example the 1~7 nucleotide fragment of exon 4 of the human tyrosinase ~=ene (SEQ ID N0:34)]. The substrate nucleic acid (approximately 100 fmoles: the nucleic acid template may contain a ~' end or other label to permit easy detection of the cleaya~Te products) is placed into a thin wall microcentrifu~e tube in a solution which comprises '_'s reaction conditions reported to be optimal for the characterized activity of the enzyme ( i. ~-. .
enzyme X). For example. if the enzyme X is a DNA polymerase. the initial reaction conditions would utilize a buffer which has been reported to be optimal for the polymerization activity of the polymerase. If enzyme X is not a polymerase. or if no specific components are reported to be needed for activity. the initial reaction may be assembled by placin'= the s0 substrate nucleic acid in a solution comprising IX CFLPTM buffer (10 mM
MOPS. 0.0~°a Tween-?0. 0.0~% Nonidet P-40~. pH 7.2 to 8.2. 0.2 to 1.0 mM MnCI,.
The substrate nucleic acid is denatured by heating the sample tube to 9~°C for seconds and then the reaction is cooled to a temperature suitable for the enzyme being tested *Trade-mark _4:~_ (e.g., if a thermostable polymerase is being tested the cleavage reaction may proceed at elevated temperatures such as 72°C, if a mesophilic enzyme is being tested the tube is cooled to 37°C for the cleavage reaction). Following denaturation and cooling to the target temperature, the cleavage reaction is initiated by the addition of a solution comprising 1 to 200 units of the enzyme to be tested (i. e. , enzyme X; the enzyme may be diluted into 1 X
CFLPTM buffer, pH 8.2 if desired).
Following the addition of the enzyme X solution, thc: cleavage reaction is allowed to proceed at the target temperature for 2 to 5 minutes. The cleavage reaction is then terminated [this may be accomplished by the addition of a stop solution (95% formamide, 10 mM
EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol)] and the cleavage products are resolved and detected using any suitable method (e.g., electrophoresis on a denaturing polyacrylamide gel followed by transfer to a solid support and nonisotopic detection). The cleavage pattern generated is examined by the criteria described below for the CFLPTM
optimization test.
1 S An enzyme is suitable for use in the CFLPTM method if it is capable of generating a unique (Z.L'., characteristic) pattern of cleavage products from a substrate nucleic acid; this cleavage must be shown to be dependent upon the presence of the enzyme.
Additionally, an enzyme must be able to reproducibly generate the same cleavage pattern when a given substrate is cleaved under the same reaction conditions. To test for reproducibility, the enzyme to be evaluated is used in at least two separate cleavage reactions run on different occasions using the same reaction conditions. If substantially the same cleavage pattern is obtained on both occasions, the enzyme is capable of reproducibly generating a cleavage pattern and is therefore suitable for use in the CFLPTM method.
When enzymes derived from mesophilic organisms are to be tested in the CFLPTM
reaction they may be initially tested at 37°C. However it may be desirable to use theses enzymes at higher temperatures in the cleavage reaction. The ability to cleave nucleic acid substrates over a range of temperatures is desirable when the cleavage reaction is being used to detect sequence variation (i.e., mutation) between different substrates.
Strong secondary structures that may dominate the cleavage pattern are less likely to be destabilized by single-base changes and may therefore interfere with mutation detf:ction. Elevated temperatures can then be used to bring these persistent structures to the brink of instability, so that the effects of small changes in sequence are maximized and revealed as alterations in the cleavage pattern. Mesophilic enzymes may be used at temperatures greater than 37°C under certain - 4~ -conditions known to the art. These conditions include the use of high (i.e., 10-30%) concentrations of glycerol in the reaction conditions. Furthermore. it is noted that while an enzyme may be isolated from a mesophilic organism this fact alone does not mean that the enzyme may not demonstrate thermostability; therefore when testing the suitability of a mesophilic enzyme in the CFLPTM reaction, the reaction should be run at 37°C and at higher temperatures. Alternatively, mild denaturants can be used to destabilize the nucleic acid substrate at a lower temperature (e.g., 1-10% formamide, 1-10% DMSO and 1-10%
glycerol have been used in enzymatic reactions to mimic thermal destabilization).
Nucleic acid substrates that may be analyzed using a cleavage means, such as a 5' nuclease, include many types of both RNA and DNA. Such nucleic acid substrates may all be obtained using standard molecular biological techniques. For example, substrates may be isolated from a tissue sample, tissue culture cells, bacteria or viruses, may be transcribed in vitro from a DNA template, or may be chemically synthesized. Furthermore, substrates may be isolated from an organism, either as genomic material or as a plasmid or similar extrachromosomal DNA, or it may be a fragment of such material generated by treatment with a restriction endonuclease or other cleavage agents or it may be synthetic.
Substrates may also be produced by amplification using the PCR. When the substrate is to be a single-stranded substrate molecule, the substrate may be produced using the PCR
with preferential amplification of one strand (asymmetric PCR). Single-stranded substrates may also be conveniently generated in other ways. For example, a double-stranded molecule containing a biotin label at the end of one of the two strands may be bound to a solid support (e.g., a magnetic bead) linked to a streptavidin moiety. The biotin-labeled strand is selectively captured by binding to the streptavidin-bead complex. It is noted that the subsequent cleavage reaction may be performed using substrate attached to the solid support, as the enzyme CleavaseTM can cleave the substrate while it is bound to the bead. A single-stranded substrate may also be produced from a double-stranded molecule by digestion of one strand with exonuclease.
The nucleic acids of interest may contain a label to aid in their detection following the cleavage reaction. The label may be a radioisotope (e.g., a ''P or ''"S-labeled nucleotide) placed at either the 5' or -3' end of the nucleic acid or alternatively the label may be distributed throughout the nucleic acid (i.e., an internally labeled substrate). The label may be a nonisotopic detectable moiety, such as a fluorophore which can be detected directly. or a reactive group which permits specific recognition by a secondary agent. For example.

biotinylated nucleic acids may be detected by probing with a streptavidin molecule which is coupled to an indicator (e.g., alkaline phosphatase or a fluorophore), or a hapten such as digoxigenin may be detected using a specific antibody coupled to a similar indicator.
Alternatively, unlabeled nucleic acid may be cleaved and vi sualized by staining (e.g., ethidium r 5 bromide staining or silver staining) or by hybridization using a labeled probe. In a preferred embodiment, the substrate nucleic acid is labeled at the 5' end with a biotin molecule and is detected using avidin or streptavidin coupled to alkaline phosphatase. In another preferred embodiment the substrate nucleic acid is labeled at the 5' end with a fluorescein molecule and is detected using an anti-fluorescein antibody-alkaline phosphatase conjugate.
The cleavage patterns are essentially partial digests of the substrate in the reaction.
When the substrate is labelled at one end (e.g., with biotin), all detectable fragments share a common end. Many of the structures recognized as active cleavage sites are likely to be only a few base-pairs long and would appear to be unstable at the elevated temperatures used in the CleavaseTM reaction. The formation or disruption of these structures in response to small 1 ~ sequence changes results in changes in the patterns of cleavage.
The products of the cleavage reaction are a collection of fragments generated by structure specific cleavage of the input nucleic acid. Nucleic acids which differ in size may be analyzed and resolved by a number of methods including electrophoresis, chromatography, fluorescence polarization, mass spectrometry and chip hybridization. The invention is illustrated using electrophoretic separation. However, it is rkoted that the resolution of the cleavage products is not limited to electrophoresis. Electrophoresis is chosen to illustrate the method of the invention because electrophoresis is widely practiced in the art and is easily accessible to the average practitioner.
If abundant quantities of DNA are available for the analysis, it may be advantageous to use direct fluorescence to detect the cleavage fragments, raising the possibility of analyzing several samples in the same tube and on the same gel. This "multiplexing"
would permit automated comparisons of closely related substrates such as wild-type and mutant forms of a gene. _ _ The CFLPTM reaction is useful to rapidly screen for differences between similar nucleic acid molecules. To optimize the CFLPTM reaction for any desired nucleic acid system (e.g.. a wild-type nucleic acid and one or more mutant forms of the wild-type nucleic acid). it is most convenient to use a single substrate from the test system (for example, the wild-type substrate) to determine the best CFLPTM reaction conditions. A single suitable condition is WO 96/15267 PC"T/US95/14673 chosen for doing the comparison CFLPTM reactions on the other molecules of interest. For example, a cleavage reaction may be optimized for a wild-type sequence and mutant sequences may subsequently be cleaved under the same conditions for comparison with the wild-type pattern. The objective of the CFLPTM optimization test is the identification of a set of conditions which allow the test molecule to form an assortment (i.e., a population) of intra-strand structures that are sufficiently stable such that treatment with a structure-specific cleavage agent such as the CleavaseTM enzyme or DNAPTaq will yield a signature array of cleavage products, yet are sufficiently unstable that minor or single-base changes within the test molecule are likely to result in a noticeable change in the array of cleavage products.
The following discussion illustrates the optimization of the CFLPTM method for use with a single-stranded substrate.
A panel of reaction conditions with varying salt concentration and temperature is first performed to identify an optimal set of conditions for the single-stranded CFLPTM. "Optimal CFLPTM" is defined for this test case as the set of conditions that yields the most widely spaced set of bands after electrophoretic separation, with the most even signal intensity between the bands.
Two elements of the cleavage reaction that significantly affect the stability of the nucleic acid structures are the temperature at which the cleavage reaction is performed and the concentration of salt in the reaction solution. Likewise, other factors affecting nucleic acid structures, such as, formamide, urea or extremes in pH may be used. The initial test typically will comprise reactions performed at four temperatures (50°C, ~5°C, 60°C and 65°C) in three different salt concentrations (0 mM, 25 mM and 50 mM) for a total of twelve individual reactions. It is not intended that the present invention be limited by the salt utilized. The salt utilized may be chosen from potassium chloride, sodium chloride, etc. with potassium chloride being a preferred salt.
For each salt concentration to be tested, 30 pl of a master mix containing a DNA
substrate, buffer and salt is prepared. When the substrate is DNA, suitable buffers include 3-jN-Morpholino]propanesulfonic acid (MOPS), pH 6.5 to 9.0, with pH 7.2 to 8.4 being particularly preferred and other "Good" biological buffers such as tris[Hydroxymethyl]aminomethane (Tris) or N,N-bis[2-Hydroxyethyl]glycine (Bicine). pH G.5 to 9.0, with pH 7.5 to 8.4 being particularly preferred. When the nucleic acid substrate is RNA. the pH of the buffer is reduced to the range of 6.0 to 8.5, with pH 6.0 to 7.0 being particularly preferred. When manganese is to used as the divalent cation in the reaction, the use of Tris buffers is not preferred. Manganese tends to precipitate as manganous oxide in Tris if the divalent cation is exposed to the buffer for prolonged periods (such as in incubations of greater than 5 minutes or in the storage of a stock buffer).
When manganese is to be used as the divalent cation, a preferred buffer is the MOPS buffer.
For reactions containing no salt (the "0 mM KCI" mix), the mix includes enough detectable DNA for 5 digests (e.g., approximately 500 fmolf;s of 5' biotinylated DNA or approximately 100 fmoles of 3'-P-S' end labeled DNA) in 30 p,l of 1 X CFLPTM
buffer ( 10 mM
MOPS, pH 7.2 to 8.2) with 1.7 mM MnCh or MgClz (the final concentration of the divalent cation will be 1 mM). Other concentrations of the divalent cation may be used if appropriate for the cleavage agent chosen (e.g., E. coli DNA polymerase I is commonly used in a buffer containing 5 mM MgCI,). The "25 mM KCl" mix includes 41.5 mM KCI in addition to the above components; the "50 mM KCl" mix includes 83.3 mM KCl in addition to the above components.
The mixes are distributed into labeled reaction tubes (0.2 ml, 0.5 ml or 1.5 ml "Eppendorf' style microcentrifuge tubes) in 6 ~l aliquots, overlaid with light mineral oil or a similar barrier, and stored on ice until use. Sixty microliters of an enzyme dilution cocktail is assembled, comprising a 5' nuclease at a suitable concentration in 1X CFLPTM
buffer without MnCI,. Preferred 5' nucleases and concentrations are 25 to 100 ng of the CleavaseTMBN
enzyme. with 25 ng being particularly preferred or 5 units of Taq DNA
polymerase (or another eubacterial Pol A-type DNA polymerase). Suitable amounts of a similar structure-specific cleavage agent in 1X CFLPTM buffer without MnCI., may also be utilized.
If a strong (i.e., stable) secondary structure is formed by the substrates, a single nucleotide change is unlikely to significantly alter that structure, or the cleavage pattern it produces. Elevated temperatures can be used to bring structures to the brink of instability. so that the effects of small changes in sequence are maximized, and revealed as alterations in the cleavage pattern within the target substrate, thus allowing the cleavage reaction to occur at that point. Consequently, it is often desirable to run the reaction at an elevated temperature (i.e., above 50°C).
Preferably, reactions are performed at 50°C, 55°C, 60°C
and 65°C. For each temperature to be tested, a trio of tubes at each of the three KCI
concentrations are brought to 95°C for 5 seconds. then cooled to the selected temperature. The reactions are then started immediately by the addition of 4 ~l of the enzyme cocktail. A duplicate trio of tubes may be included (these tubes receiving 4 p,l of 1X CFLPTM buffer without enzyme or MnCh), to WO 96/15267 PG"T/US95/14673 assess the nucleic acid stability in these reaction conditions. All reactions proceed for 5 minutes, and are stopped by the addition of 8 ~l of 95% formamide with 20 mM
EDTA and 0.05% xylene cyanol and 0.05% bromophenol blue.. Reactions may be assembled and stored on ice if necessary. Completed reactions are stored on ice until all reactions _in the series have been performed.
x Samples are heated to 72°C for 2 minutes and 3 to 7 p.l of each reaction is resolved by electrophoresis through a suitable gel, such as 6 to 10% polyacrylamide ( 19:1 cross-link), with 7M urea, in a buffer of 45 mM Tris-Borate, pH 8.3, 1.4 mM EDTA for nucleic acids up to approximately 1.5 kb, or native or denaturing agarose gels for larger molecules. The nucleic acids may be visualized as described above, by staining, autoradiography (for radioisotopes) or by transfer to a nylon or other membrane support with subsequent hybridization and/or nonisotopic detection. The patterns generated are examined by the criteria described above and a reaction condition is chosen for the performance of the variant comparison CFLPsTM.
A "no enzyme" control allows the assessment of the stability of the nucleic acid substrate under particular reaction conditions. In this instance, the substrate is placed in a tube containing all reaction components except the enzyme and treated the same as the enzyme-containing reactions. Other control reactions may be run. A wild-type substrate may be cleaved each time a new mutant substrate is tested. Alternatively, a previously characterized mutant may be run in parallel with a substrate suspected of containing a different mutation. Previously characterized substrates allow for the comparison of the cleavage pattern produced by the new test substrate with a known cleavage pattern. In this manner, alterations in the new test substrate may be identified.
When the CFLPTM pattern generated by cleavage of a single-stranded substrate contains an overly strong (i.e., intense) band, this indicates the presence of a very stable structure. The preferred method for redistributing the signal is to alter the reaction conditions to increase structure stability (e.g., lower the temperature of the cleavage reaction, raise the monovalent salt concentration); this allows other less stable structures to compete more effectively for cleavage.
When the CFLPTM reaction is to optimized for the cleavage a double-stranded substrate the following steps are taken. The cleavage of double-stranded DNA substrates up to 2.000 base pairs may be optimized in this manner.
The double-stranded substrate is prepared such that it contains a single end-label using any of the methods known to the art. The molar amount of DNA used in the optimization reactions is the same as that use for the optimization of reactions utilizing single-stranded substrates. The most notable differences between the optimization of the CFLPTM reaction for single- versus double-stranded substrates is that the double-stranded substrate is denatured in distilled water without buffer, the concentration of MnCI, in the reaction is reduced to 0.2 mM, the KCl (or other monovalent salt) is omitted, and the enzyme concentration is reduced t to 10 to 25 ng per reaction. In contrast to the optimization of the single-stranded CFLPTM
reaction (described above) where the variation of the monovalent salt (e.g., KCl) concentration is a critical controlling factor, in the optimization of the double-stranded CFLPTM reaction the range of temperature is the more critical controlling factor for optimization of the reaction. When optimizing the double-stranded CFLPTM
reaction a reaction tube containing the substrate and other components described below is set up to allow performance of the reaction at each of the following temperatures:
40°C, 45°C, 50°C, 55°C, 60°C,-65°C, 70°C, and 75°C.
For each temperature to be tested, a mixture comprising the single end labelled double-stranded DNA substrate and distilled water in a volume of 15 ~,l is prepared and placed into a thin walled microcentrifuge tube. This mixture may be overlaid with light mineral oil or liquid wax (this overlay is not generally required but may provide more consistent results with some double-stranded DNA substrates).
A 2 mM solution of MnCh is prepared. For each CFLPTM reaction, 5 ~l of a diluted enzyme solution is prepared comprising 2 ~.1 of 1 OX CFLPTM buffer ( 100 mM
MOPS, pH 7.2 to 8.2, 0.5% Tween-20, 0.5% Nonidet P-40), 2 Pl of 2 mM MnCI., and 25 ng of CleavaseTM
BN enzyme and distilled water to yield a final volume of 5 ~.1.
The DNA mixture is heated to 95°C for 10 to 30 seconds and then individual tubes are cooled to the reaction temperatures to be tested (e.g.; 40°C, 45°C, 50°C, 55°C, 60°C, 65°C, 70°C, and 75°C). The cleavage reaction is started by adding 5 ql of the dilute enzyme solution to each tube at the target reaction temperature. The reaction is incubated at the target temperature for ~ minutes and the reaction is terminated (e.g., by the addition of 16 q.l of stop solution comprising 95% formamide with 10 mM EDTA and 0.05% xylene cyanol and 0.05%
bromophenol blue).
Samples are heated to 72°C for 1 to 2 minutes and 3 to 7 pl of each reaction is resolved by electrophoresis through a suitable gel, such as 6 to 10%
polyacrylamide ( 19:1 cross-lil~l:), with 7M urea, in a buffer of 4~ mM Tris-Borate, pH 8.3. 1.4 mM
EDTA for nucleic acids up to approximately 1.5 kb, or native or denaturing agarose gels for larger molecules. The nucleic acids may be visualized as described above, by staining.
autoradiography (for radioisotopes) or by transfer to a nylon or other membrane support with subsequent hybridization and/or nonisotopic detection. The patterns generated are examined by the criteria described above and a reaction condition is chosen for the performance of the double-stranded CFLPTM. Control reactions may be run as described above to assess nucleic acid stability or to create patterns fox reference.
When performing double-stranded CFLPTM reactions the MnCI, concentration preferably will not exceed 0.25 mM. If the end label on the double-stranded DNA substrate disappears (i.e., loses its 5' end label as judged by a loss of signal upon detection of the cleavage products), the concentration of MnCI, may be reduced to 0.1 mM. Any EDTA
present in the DNA storage buffer will reduce the amount of free Mn" in the reaction, so double-stranded DNA should be dissolved in water or Tris-HCI with a EDTA
concentration of 0.1 mM or less.
When the nucleic acid substrate is labelled at one end (e.g., with biotin or ''-P) all detectable fragments share a common end. For short DNA substrates (less than 2~0 nucleotides) the concentration of the enzyme (e.g., CleavaseTM BN) and the length of the incubation have minimal influence on the distribution of signal intensity, indicating that the cleavage patterns are not partial digests of a single structure assumed by the nucleic acid substrate, but rather are relatively complete digests of a collection of stable structures formed by the substrate. With longer DNA substrates (greater than 250 nucleotides) there is a greater chance of having multiple cleavage sites on each structure, giving apparent overdigestion as indicated by the absence of any residual full-length materials. For these DNA
substrates, the enzyme concentration may be lowered in the cleavage reaction (for example, if ~0 ng of the CleavaseTM BN enzyme were used initially and overdigestion was apparent, the concentration of enzyme may be reduced to 25, 10 or 1 ng per reaction). Alternatively, a combination of Mn'-~ and Mg'+ can be used in CFLPTM buffer, to attenuate the rate of cleavage. When 0.2 mM MnCh is used in a CFLPTM reaction, as described above (with either a single-or double stranded nucleic acid substrate), the use of 1 mM Mg'~' in addition to the Mn'~ slows down the rate of cleavage, in the case of the 1059 by amplicon seen in Figure 30, the rate of cleavage is reduced approximately three-fold (in the Mn'-T/Mg'~ mixture as compared to Mn'--alone). If overdigestion is observed when the substrate is incubated at the reaction temperature for 2 to 5 minutes in the presence of 0.2 to 1.0 mM Mn~~, the 0.2 mM
Mn'~'/1mM Mg'- mixture may be used in conjunction with a reaction time of ~ to 20 minutes.

Cleavage products produced by cleavage of either single-or double-stranded substrates which contain a biotin label may be detected using the following nonisotopic detection method. The following description is exemplary only: the art knows alternative methods for the detection of biotin-labelled products. After electrophoresis of the reaction products, the gel plates are separated allowing the gel to remain flat on one plate. A
positively charged nylon membrane (preferred membranes include Nytran~Plus, 0.2 or 0.45 mm-pore size, Schleicher and Schuell, Keene, NH), cut to size and pre-wetted in O.SX TBE (45 mM tris-Borate, pH 8.3, 1.4 mM EDTA), is laid on top of the exposc;d gel. All air bubbles trapped between the gel and the membrane are removed (e.g., by rolling a 10 ml pipet firmly across the membrane). Two pieces of 3MM filter paper (Whatman) are then placed on top of the membrane, the other glass plate is replaced, and the sandwich is clamped with binder clips or pressed with books or weights. The transfer is allowed to proceed 2 hours to overnight (the signal increases with longer transfer).
After transfer, the membrane is carefully peeled from the gel and allowed to air dry.
Distilled water from a squeeze bottle can be used to loosen any gel that sticks to the membrane. After complete drying, the membrane is agitated for 30 minutes in 1.2X
Sequenase Images Blocking Buffer (United States Biochemical, Cleveland, OH;
avoid any precipitates in the blocking buffer by decanting or filtering); 0.3 ml of the buffer is used per cm'- of membrane (e.g., 30 mls for a lOcm x lOcm blot). A streptavidin-alkaline phosphatase conjugate (SAAP, United Stated Biochemical) is added at a 1:4000 dilution directly to the blocking solution (avoid spotting directly on membrane), and agitated for 15 minutes. The membrane is rinsed briefly with dH~O and then washed 3 times (5 minutes of shaking per/wash) in 1X SAAP buffer (100 mM Tris-HCI, pH 10; 50 mM NaCI) with 0.1%
sodium dodecyl sulfate (SDS), using 0.5 ml buffer/cm' of membrane, with brief water rinses between each wash. The membrane is then washed twice in 1X SAAP buffer (no SDS) with 1 mM
MgCI,, drained thoroughly, and placed in a plastic heat-sealable bag. Using a sterile pipet tip, 0.05 ml/cm' of CDP-StarTM (Tropix, Bedford, MA) is added to the bag and distributed over the entire membrane for 5 minutes. The bag is drained of all excess liquid and air bubbles, sealed, and the membrane is exposed to X-ray film (e.g., Kadak XRP) for 30 W
mutes.
Exposure times are adjusted as necessary for resolution and clarity.
To date, every nucleic acid substrate tested in the CFLPTM system has produced a reproducible pattern of fragments. The sensitivity and specificity of the cleavage reaction make this method of analysis very suitable for the rapid screening of mutations in cancer diagnostics, tissue typing, genetic identity, bacterial and- viral typing, polymorphism analysis, structure analysis, mutant screening in genetic crosses, etc. It could also be applied to enhanced RNA analysis, high level multiplexing and .extension to longer fragments. One distinct benefit of using the CleavaseTM reaction to characterize nucleic acids is that the pattern of cleavage products constitutes a characteristic fingerprint, so a potential mutant can be compared to previously characterized mutants without sequencing. Also, the place in the fragment pattern where a change is observed gives a good indication of the position of the mutation. But it is noted that the mutation need not be at the precise site of cleavage, but only in an area that affects the stability of the structure.
III. Detection of Mutations in the p53 Tumor Suppressor Gene Using the CFLPTM Method Tumor -suppressor genes control cellular proliferation and a variety of other processes important for tissue homeostasis. One of the most extensively studied of these. the p53 gene, 1 ~ encodes a regulator of the cell cycle machinery that can suppress the growth of cancer cells as well as inhibit cell transformation (Levine, Annu. Rev. Biochem. 62:623 [1993)). Tumor suppressor mutations that alter or obliterate normal p53 function are common.
Mutations in the p53 tumor suppressor gene are found in about half of all cases of human cancer making alterations in the p53 gene the most common cancer-related genetic 20 change known at the gene level. In the wild-type or non-mutated form, the p53 gene encodes a 53-kD nuclear phosphoprotein, comprising 393 amino acids, which is involved in the control of cellular proliferation. Mutations in the p53 gene are generally (greater than 90%) missense mutations which cause a change in the identity of an amino acid rather than nonsense mutations which cause inactivation of the protein. It has been postulated that the 2~ high frequency of p53 mutation seen in human tumors is due to the fact that the missense mutations cause both a loss of tumor suppressor function and a gain of oncogenic function jLane, D.P. and Benchimol, S., Genes Dev. 4:1 (1990)).
The gene encoding the p53 protein is large, .spanning 20,000 base pairs, and is divided into 11 exons (see Figure 4). The ability to scan the large p~3 gene for the presence 30 of mutations has important clinical applications. In several major human cancers the presence of a tumor p53 mutation is associated with a poor prognosis. p53 mutation has been shown to be an independent marker of reduced survival in lymph node-negative breast cancers. a finding that may assist clinicians in reaching decisions regarding more aggressive therapeutic treatment. Also, Lowe and co-workers have demonstrated tUat the vulnerability of tumor cells to radiation or chemotherapy is greatly reduced by mutations which abolish p~3-dependent apoptosis [Lowe et al., Cell 74:957 (1995)].
Regions of the p53 gene from approximately 10,000- tumors have been sequenced in the last 4 to 5 years, resulting in characterization of over 3,700 mutations of which approximately 1,200 represent independent p53 mutations (i.e., point mutations, insertion or deletions). A database has been compiled and deposited wish the European Molecular Biology Laboratory (EMBL) Data Library and is available in electronic form [Hollstein, M. et al. (1994) Nucleic Acids Res. 22:3551 and Cariello, N.F. et al. (1994) Nucleic Acids Res.
22:3549]. Iri addition, an IBM PC compatible software package to analyze the information in the database has been developed. [Cariello et al., Nucl. Acids Res. 22:3551 (1994)]. The point mutations in the database were identified by DNA sequencing of PCR-amplified products. In most cases, preliminary screening for mutations by SSCP or DGGE
was performed.
Analysis of the p53 mutations shows that the p53 gene contains 5 hot spot regions (HSR) most frequently mutated in human tumors that show a tight correlation between domains of the protein that are evolutionary highly conserved (ECDs) and seem to be specifically involved in the transformation process (see Figure 4; the height of the bar represent the relative percentage of total mutations associated with the five HSRs). The five HSRs are confined to exons 5 to 8 and account for over 85~% of the mutations detected.
However, because these studies generally confined their analysis to PCR
amplifications and sequencing of regions located between exons 5 to 8, it should be kept in mind that mutations outside this region are underrepresented. As 10% to 15% of the mutations lie outside this region, a clinically effective p53 gene DNA diagnostic should be able to cost-effectively scan for life-threatening mutations scattered across the entire gene.
The following table lists a number of the known p53 mutations.

WO 96!15267 PC"TIUS95114673 Human p53 Gene Mutations CODON NO. WILD-TYPE MUTANT EVENT TUMOR TYPE

36 CCG - CCA GOAT Lung ' 49 GAT CAT GC~CG CML

53 TGG TGT GC~TA CML

110 CGT TGT GC-SAT Hepatoca 113 TTC TGT Double M NSCLC

128 CCT CCG T-~G Breast 128 TCT C-~T Breast 129 GCC GAC GC-ETA Neurofibrosa 130 CTC CTG GC-aCG MDS

132 AAG AAC GC-~CG Colorectal ca 132 CAG AT~CG Breast ca 132 AAT GC~TA Lung (NSCLC) ca 132 CAG AT~CG Pancreatic ca 132 AGG AT-~GC CML

133 ATG TTG AT~TA Colorectal ca 133 AAG AT~TA Burkitt lymphoma 134 TTT TTA AT-ETA Lung (SCLC) ca 135 TGC TAC GOAT Colorectal ca 135 TCC GC-~CG AML

135 TAC GC-SAT Lung (NSCLC) ca 135 TGG GC-~CG MDS

136 CAA GAG Double M Breast ca 138 GCC GTC GC-SAT Rhabdomyosa 138 GGC . GC-~CG _ Lung (SCLC) _ ca 140 ACC TAC AT~TA CML

141 TGC TAC GC-SAT Colorectal ca 141 TAC GOAT Bladder ca 143 GTG GCG AT~GC Colorectal ca 143 TTG GC-ETA Lung (NSCLC) ca 144 CAG TAG GOAT Esophageal ca WO 96/15267 PG"T/US95/14673 144 CCG AT-~CG Burkitt lymphoma 151 CCC CAT Double M Leiomyosa 151 CAC ~ GC-ETA Lung (SCLC) ca 151 TCC GC-SAT Glioblastoma 151 TCC GC-SAT Lung (NSCLC) ca 152 CCG- CTG GC--SAT Leiomvosa 152 TCG GC-SAT Breast ca 154 GGC GTC GC~TA Esophageal ca 154 GTC GC-ETA. LunQ
(NSCLC) ca 154 GTC GC~TA Lung (NSCLC) ca 154 GTC GC~TA Lung (NSCLC) ca 15G CGC CCC GC-~CG Rhabdomyosa I SG CCC GC~CG Osteosa 156 CGT GC-SAT Lung (NSCLC) ca 15G CCC GC-~CCi Lun (NSCLC) ca I 57 GTC TTC GC-ETA Hepatoca 157 TTC GC-ETA Lung (SCLC) ca 157 TTC GC-ETA Lung (NSCLC) ca 157 TTC GC-ETA Breast ca 157 TTC GC~TA Lung (SCLC) ca 157 TTC GC~TA Bladder ca 158 CGC CGT GC-SAT Neurofibrosa 158 CAC GC-~A'l.' Burkitt lymphoma 159 GCC GTC GC-SAT Lung (NSCLC) ca 159 CCC GC->CG Lung (NSCLC) ca 163 TAC TGC AT~GC Breast ca 163 CAC AT~GC Burkitt lymphoma 164 AAG CAG AT-~CG Breast ca 171 GAG TAG GC~TA Lung (SCLC) ca 172 GTT TTT ~ GC~TA Burkitt lymphoma 173 GTG TTG GC->TA Lung (NSCLC) ca' 173 TTG GC-ETA Lung (NSCLC) ca 17 3 GGG AT~CG Burkitt lymphoma 173 GTA GOAT Gastric ca-3J 175 CGC CAC GC-SAT Colorectal ad 175 CAC GC-SAT Colorectal ad 175 ~ [ CAC I GC-SAT Colorectal ad 175 CAC GC-SAT Colorectal ca 175 CAC GC-aAT Colorectal ca 175 CAC GC-SAT Brain tumor 175 CAC GC-SAT Colorectal ca 175 CAC GOAT Colorectal ca 175 CAC GC-SAT Leiomyosa 175 CAC GOAT Esophageal ca 175 CAC GOAT Glioblastoma 175 CAC GOAT Colorectal ca 175 CAC GC-SAT Breast ca 175 CTC GC->TA Breast ca 175 AGC GC~TA Hepatoca 175 CAC GOAT Burkitt lymphoma 175 CAC GC-SAT Burkitt lymphoma 175 CAC GOAT Burkitt lymphoma 175 CAC GOAT Burkitt lymphoma 175 CAC GC-SAT Gastric ca 176 TGC TTC GC~TA Lung (NSCLC) ca 176 TTC GC-ETA Esophageal ca 176 TTC GC-ETA Lung (NSCLC) ca 176 TAC GC-SAT Burkitt lymphoma 177 CCC CGC GC-~CG PTLC

179 CAT TAT GOAT Neurofibrosa 179 CAG AT->CG Lung (SCLC) ca 179 CTT AT-ETA Esophageal ca 179 GAT ~ GC~CG Breast ca 179 CTT AT-ETA Cholangiosa 179 CTT AT~TA Cholanaiosa 181 CGC CAC GOAT Li-Fraumeni sdm 187 GGT TGT GC~TA Breast ca 192 CAG TAG GOAT Esophageal ca 193 CAT CGT AT~GC Lung (SCLC) ca 193 TAT GOAT Esophageal ca 193 CGT AT->GC AML

194 CTT TTT GOAT Breast ca 194 CGT AT~CG Lung (SCLC) ca 194 CGT AT~CG Esophageal ca 194 CGT AT->CG Esophageal r ca 194 CGT AT~CG B-CLL

196 CGA TGA GC-SAT Colorectal ca 196 TGA GOAT T-cell lymphoma 196 TGA GOAT Lung (SCLC) ca 196 TGA GC-SAT Bladder ca 198 GAA TAA GC-ETA Lung (SCLC) ca 198 TAA GC~TA Lung (SCLC) ca 202 CGT CTT GC~TA CML

204 GAG GGG AT~GC CML

205 TAT TGT AT->GC B-ALL

205 TGT AT-~GC B-CLL

205 TTT AT-ETA Gastric ca 211 ACT GCT AT->GC Colorectal ca 213 CGA TGA GC-SAT Colorectal ca 213 CAA GOAT B-cell lymphoma 213 CAA GOAT Burkitt lymphoma 213 CGG AT->GC Lung (SCLC) ca 213 CGG AT-~GC Esophageal ca 213 TGA GC~A'r Lung (NSCLC) ca 213 CGG AT-~GC Lung (NSCLC) ca 213 TGA GC-SAT Burkitt lymphoma 213 TGA GC-->A'T Burkitt lymphoma 215 AGT GGT AT~GC Colorectal ca 21G GTG ATG ~ GC-SAT Brain tumor 216 GAG AT-ETA Burkitt lymphoma 216 TTG GC~TA Gastric ca 216 ATG GC->AT Ovarian ca 220 TAT TGT AT-~GC Colorectal ca 229 TGT TGA AT~TA Lung (SCLC) ca 232 ATC AGC AT-~CG B-CLL

234 TAC CAC AT~GC B-cell lymphoma 234 CAC AT-~GC Burkitt lymphoma 234 TGC AT--~GC Burkitt lymphoma 236 TAC TGC AT-~GC Burkitt lymphoma 237 ATG AGG AT~CG T-ALL

237 ATA GOAT Lung (SCLC) ca 237 ATA GC-SAT Breast ca 237 ATA GC->AT Burkitt lymphoma 237 ATA GOAT Richter's sdm 238 TGT TTT GC-ETA Larynx ca 23g TAT GOAT Burkitt lymphoma 23 g TAT GOAT C M L

239 AAC AGC AT~GC Colorectal ca 239 AGC AT~GC Colorectal ca 239 AGC AT-~GC Burkitt lymphoma 239 AGC AT~GC CML

239 AGC AT-~GC CML

239 AGC AT~GC B-CLL

241 TCC TTC GOAT Colorectal ca 2U 241 TGC GC-~CG Colorectal ca 241 TGC GC-~CG Bladder ca 242 TGC TCC GC~CG Lung (SCLC) ca 242 TTC GC-ETA Breast ca 242 TCC GC-~CG MDS

242 TAC GC-SAT Ependymoma 244 TGC GC->TA Esphageal ca 244 TGC GC-ETA Lung (SCLC) ca 244 AGC GC-SAT Hepatoca 245 GGC GTC ~ GC-ETA Esophageal ca 245 TGC GC~TA Li-Fraumeni sdm 245 AGC GC-SAT Leyomyosa 245 GAC GOAT Li-Fraumeni sdm 245 AGC GOAT Esophageal ca 245 GCC GC-~CG - Bladder ca 245 GAC GOAT Breast ca 245 GAC GOAT Li-Fraumeni sdm 24~ GGC TGC GC-ETA Li-Fraumeni sdm 245 GTC GC~TA Cervical ca 246 ATG GTG AT~GC AML

246 ATC GC-~CG Lung (NSCLC) ca 246 GTG AT~GC Hepatoca 246 GTG AT~GC Bladder ca 247 AAC ATC AT-ETA Lung (NSCLC) ca 248 CGG TGG GC-SAT Colorectal ad 248 TGG GC-SAT Colorectal ca 248 CAG GOAT Colorectal ca 248 CAG GC->AT Colorectal ca 248 CAG GC-SAT Esophageal ca 248 TGG GC-~A'T Li-Fraumeni sdm 248 TGG GC~A'i' Li-Fraumeni sdm 248 TGG GC~A'1' Colorectal ca 248 TGG GC-~A'r Colorectal ca 248 TGG GC-SAT Rhabdomyosa 248 CTG GC~TA Esophageal ca 248 TGG GOAT Lun (NSCLC) ca 248 CAG GC-~A'T Lung (SCLC) ca 248 CTG GC-ETA Lung (SCLC) ca 248 CAG GC-~A'T T-ALL

248 TGG GC-SAT Lun 5 (NSCLC) ca 248 CTG GC-ETA Lung (SCLC) ca 248 TGG GC-SAT Colorectal ca 248 CAG GC->A'T Bladder ca 248 TGG GC-SAT Burkitt lymphoma 248 CAG ~ GOAT Breast ca 248 CAG GC-aAT Burkitt lymphoma 248 TGG GC-SAT Burkitt lymphoma 248 CAG GC->AT Burkitt lymphoma 248 TGG GC-SAT Burkitt lymphoma 248 CAG GOAT Gastric ca 248 TGG GC-SAT Lung (SCLC) ca 248 CAG GOAT Breast ca 248 TGG GC-SAT Li-Fraumeni sdm 248 CAG GC-SAT Li-Fraumeni sdm S 248 TGG GG~AT Colorectal ca 249 AGG AGT GC-ETA Hepatoca 249 AGT GC~TA Hepatoca 249 AGT GC~TA Hepatoca 249 AGC GC-~CG Hepatoca 249 AGT GC~TA Hepatoca 249 AGT GC-ETA Hepatoca 249 AGT GC~TA Hepatoca 249 AGT GC~TA Hepatoca 249 AGT GC-ETA Hepatoca 249 AGT GC->TA Hepatoca 249 AGT GC-ETA Hepatoca 249 AGT GC~TA Esophageal ca 249 AGC GC-~CG Breast ca 249 AGT GC-ETA Lung (NSCLC) ca 249 AGT GC~TA Hepatoca 250 CCC CTC GOAT Burkitt lymphoma 251 ATC AGC AT~CG Gastric ca 252 CTC CCC AT-~GC Li-Fraumeni sdm 252 CTC CCC AT-~GC Li-Fraumeni sdm 254 ATC GAC Double M Burkitt lymphoma 254 AAC AT-ETA Breast ca 256 ACA GCA AT~GC T-ALL

258 GAA AAA GOAT Li-Fraumeni sdm 25g AAA GOAT Burkitt lymphoma 258 AAA ~ GC-SAT Li-Fraumeni sdm 259 GAC GGC AT~GC T-ALL

260 TCC GCC AT~CG T-ALL

266 GGA GTA GC iTA Lung (NSCLC) ca 26G GTA GC-aTA Lun~~ (NSCLC) ca 266 GTA GC-ETA Breast ca 267 CGG CCG GC-~CG Luna (SCLC) ca 270 TTT TGT AT~CG Esophageal ca 270 TGT AT-~CG T-ALL

272 GTG ATG GOAT' Brain tumor 272 CTG ~ GC-~CG Lung (SCLC) ca i 272 ATG GOAT Hepatoca 272 ATG GC->AT AML

273 CGT TGT GC-SAT' Colorectal ad 273 TGT GOAT' Brain tumor 273 CAT GC-SAT Breast ca 273 CAT GC-SAT Colorectal ca 273 TGT GOAT' Lung (NSCLC) ca 273 CTT GC-ETA Lung (SCLC) ca 273 CAT GC-SAT Colorectal ca 273 CAT GOAT Colorectal ca 273 CAT GC-SAT Colorectal ca 273 CAT GC-SAT Lun b (NSCLC) ca 273 CCT GC~CG Lung (NSCLC) ca 273 CTT GC~TA Lung (NSCLC) ca 273 CTT GC-ETA Lun (NSCLC) ca 273 CAT GC-aAT Thyroid ca 273 CAT GOAT Lung (SCLC) ca 273 TGT GC-SAT B-cell lymphoma 273 TGT GC-SAT Burkitt lymphoma 273 TGT GOAT Burkitt lymphoma 273 CAT GC-SAT Li-Fraumeni sdm 273 TGT GOAT Cervical ca 273 CAT GOAT B~CLL

274 GTT GAT ~ AT-ETA Erythroleukemia 276 GCC CCC GC-~CG B-ALL

276 GAC GC~TA Hepatoca 277 TGT TTT GC~TA Lung (SCLC) ca 278 CCT TCT GOAT Esophageal ca 278 CTT GOAT Esophageal ca 278 GCT GC-~CG Breast ca 278 TCT GC-SAT LunQ (SCLC) ca WO 96/15267 PC"TlUS95114673 278 CGT GC-~CG Ovarian ca 280 AGA AAA GOAT Esophageal ca 280 AAA ~ GOAT Breast ca 281 GAC GGC AT~GC Colorectal ca 281 GGC AT~GC Breast ca 281 GAC GAG GC-~CG Richter's sdm 281 TAC GC~TA B-CLL

282 CGG TGG GOAT Colorectal ad 282 TGG GC-SAT Colorectal ca 282 CGG TGG GC-SAT Rhabdomyosa 2g2 GGG GC~CG Lung (NSCLC) ca 282 CCG GC-~CG Breast ca 2g2 TGG GC-SAT Bladder ca 2g2 TGG GC-SAT AML

282 CTG GC~TA Breast ca 2g2 TGG GOAT B-ALL

2g2 TGG GC-SAT Burkitt lymphoma 282 TGG GOAT Richter's sdm 2g2 TGG GC-SAT Ovarian ca 282 TGG GC-SAT Li-Fraumeni sdm 283 CGC TGC GC-SAT Colorectal ca 283 CCC GC-~CG Lung (NSCLC) ca 285 GAG AAG GC->AT Breast ca 28G GAA AAA GC-SAT Colorectal ca 28G GGA AT-~GC Lung (SCLC) ca 28G - GCA AT-~CG Li-Fraumeni sdm 287 GAG TAG GC-ETA Burkitt lymphoma 293 GGG TGG GC-ETA Glioblastoma 298 GAG TAG GC~TA Bladder ca 302 GGG GGT ~ GC~TA Lung (SCLC) ca 305 AAG TAG AT~TA Esophageal ca 305 TAG AT~TA Esophageal ca 307 GCA ACA GC-SAT Breast ca 309 CCC TCC GOAT Colorectal ca 334 GG GTG GC~TA Lung (SCLC) G ca 342 CGA TGA GC-SAT Luna (SCLC) ca DELETIONS/INSERTIONS
CODON EVENT TUMOR TYPE

137 del 7 Gastric ca 143 del 1 Gastric ca 152 del 13 Colorectal ad 167 del 1 Breast ca 168 del 31 He atoca 175 del 18 Breast ca 190 del 3 nu 1 ALL

201 del 1 Breast ca 206 del 1 Burkitt I m homa 206 del 1 Burkitt 1 m homa 214 del I B-ALL

236 del 27 Bladder ca 239 del 1 Lung (NSCLC) ca 262 del 1 Astroc toms 262 del 24 Gastric ca 262 del 24 Lun~ (NSCLC) ca 263 del 1 Eso haaeal ca 264 del 1 AML

286 del 8 He atoca 293 del 1 Lun~ (NSCLC) ca 307 del 1 Li-Fraumeni sdm 381 del 1 He atoca Exon 5 del 15 B-ALL

152 ins 1 B-CLL

239 ins 1 Waldenstrom sdm 252 ins 4 Gastric ca 256 ins 1 AML

275 ins I B-CLL

301 ins 1 MDS

307 ins 1 ~ Glioblastoma Exon 8 ins 25 HCL

WO 96115267 PC"T/US95/14673 SPLICE MUTATIONS
INTRON SITE EVENT TUMOR TYPE

Intron 3 Accept GC->CG Lung (SCLC) ca Intron 4 Donor GC-ETA Lung (SCLC) ca Intron 4 Donor GOAT T-BALL

Intron 5 Donor GOAT CML

Intron 6 Donor AT~CG Lung (SCLC) ca Intron 6 Accept AT->TA Lung (SCLC) ca Intron 6 Accept AT-ETA Lung (NSCLC) ca Intron 7 Donor GC~TA Lung (NSCLC) ca Intron 7 Accept GC-~CG Lung (SCLC) ca Intron 7 Accept CG~AT AML

Intron 7 Donor GC~TA Lung (SCLC) ca 1 J Intron 9 Donor GC-ETA Lung (SCLC) ca A. CFLPTM Analysis of p53 Mutations in Clinical Samples To permit the identification of mutations in the p53 gene from clinical samples, nucleic acid comprising p53 gene sequences are prepared. The nucleic acid may comprise genomic DNA, RNA or cDNA forms of the p53 gene. Nucleic acid may be extracted from a variety of clinical samples [fresh or frozen tissue, suspensions of cells (c.g.. blood), cerebral spinal fluid, sputum, etc.] using a variety of standard techniques or commercially available kits. For example, kits which allow the isolation of RNA or DNA from tissue samples are available from Qiagen, Inc. (Chatsworth, CA) and Stratagene (LaJolla, CA), respectively.
Total RNA may be isolated from tissues and tumors by a number of methods known to those skilled in the art and commercial kits are available to facilitate the isolation. For example, the RNeasy~ kit (Qiagen Inc., Chatsworth, CA) provides protocol, reagents and plasticware to permit the isolation of total RNA from tissues, cultured cells or bacteria, with no modification to the manufacturer's instructions, in approximately 2U minutes. Should it be desirable. in the case of eukaryotic RNA isolates, to further enrich for messenger RNAs, the polyadenylated RNAs in the mixture may be specifically isolated by binding to an oligo-deoxythymidine matrix, through the use of a kit such as the Oligotex~ kit (Qiagen).
Comparable isolation , kits for both of these steps are available through a number of commercial suppliers.
In addition, RNA may be extracted from samples, including biopsy specimens. , conveniently by lysing the homogenized tissue in a buffer containing 0.22 M
NaCI, 0.7~ 111M

MgCI,, 0.1 M Tris-HCI, pH 8.0, 12.5 mM EDTA, 0.25% NP40, 1% SDS, 0.5 mM DTT, u/ml placental RNAse inhibitor and 200 p.g/ml Proteinase K. Following incubation at 37°C
for 30 min, the RNA is extracted with phenol:chloraform (1:1) and the RNA is recovered by ethanol precipitation.
Since the majority of p53 mutations are found within exons 5-8, it is convenient as a first analysis to examine a PCR fragment spanning this region. PCR fragments spanning exons 5- 8 may be amplified from clinical samples using the technique of RT-PCR (reverse transcription-PCR); kits which permit the user to start with tissue and produce a PCR product are available from Perkin Elmer (Norwalk, CT) and Stratagene (LaJolla, CA).
The RT-PCR
technique generates a single-stranded cDNA corresponding to a chosen segment of the coding region of a gene by using reverse transcription of RNA; the single-stranded cDNA is then used as template in the PCR. In the case of the p53 gene, an approximately 600 by fragment spanning exons 5-8 is generated using primers located in the coding region immediately adjacent to exons 5 and 8 in the RT-PCR. The PCR amplified segment is then subjected to the CFLP reaction and the reaction products are analyzed as described above in section VIII.
Fragments suitable for CFLP analysis may also be generated by PCR
amplification of genomic DNA. DNA is extracted from a sample and primers corresponding to sequences present in introns 4 and 8 are used to amplify a segment of the p53 gene spanning exons 5-8 which includes introns 5-7 (an approximately 2 kb fragment). If it is desirable to use smaller fragments of DNA in the CFLP reaction, primers may be chosen to amplify smaller ( 1 lcb or less) segments of genomic DNA or alternatively a large PCR. fragment may be divided into two or more smaller fragments using restriction enzymes.
In order to facilitate the identification of p53 mutations in the clinical setting, a library containing the CFLP pattern produced by previously characterized mutations may be provided. Comparison of the pattern generated using nucleic acid derived from a clinical sample with the patterns produced by cleavage of known and. characterized p53 mutations will allow the rapid identification of the specific p53 mutation present in the patient's tissue. The comparison of CFLP patterns from clinical samples .to the patterns present in the library may be accomplished by a variety of means. The simplest and least expensive comparison involves visual comparison. Given the large number of unique mutations known at the p53 locus, visual (i.e., manual) comparison may be too time-consuming. especially when large numbers of clinical isolates are to be screened. Therefore the CFLP patterns or "bar codes" may be provided in an electronic format for ease and efficiency in camparison.
Electronic entry may comprise storage of scans of gels containing the CFLP products of the reference p53 mutations (using for example, the GeneReader and Gel Doctor Fluorescence Gel documentation system (BioRad. Hercules, CA) or the ImageMaster (Pharmacia Biotech, Piscataway, NJ). Alternatively, as the detection of cleavage patterns may be automated using DNA sequencing instrumentation (see Example 18), the banding pattern may be stored as the signal collected from the appropriate channels during an automated run [examples of instrumentation suitable for such analysis and data collection include fluorescence-based gel imagers such as fluoroimagers produced by Molecular Dynamics and Hitachi or by real-time electrophoresis detection systems such as the ABI 377 or Pharmacia ALF DNA
Sequencer].
B. Generation of a Library of Characterized p53 Mutations The generation of a library ofcharacterized mutations will enable clinical samples to be rapidly and directly screened for the presence of the most common p53 mutations.
Comparison of CFLP patterns generated from clinical samples to the p53 bar code library will establish both the presence of a mutation in the p53 gene and its precise identity without the necessity of costly and time consuming DNA sequence analysis.
The p53 bar code library is generated using reverse genetics. Engineering of p53 mutations ensures the identity and purity of each of the mutations as each engineered mutation is confirmed by DNA sequencing. The individual p53 mutations in p53 bar code library are generated using the 2-step "recombinant PCR" technique [Higuchi, R. (1991) In Ehrlich, H.A.
(Ed.). PCR Technology: Principles and Applications for DNA Amplification, Stoclcton Press, New Yorlc, pp. 61-70 and Nelson, R.M. and Long, G.L. (1989) Analytical Biochem.
180:147]. Figure 5 provides a schematic representation of one method of a 2-step recombinant PCR technique that may be used for the generation of p53 mutations.
The template for the PCR amplifications is the entire human p53 cDNA gene. In the first of the two PCRs (designated "PCR 1" in Fig. 5), an oligonucleotide containing the engineered mutation ("oligo A" in Fig. 5) and an oligonucleotide containing a 5' arm of approximately 20 non-complementary bases ("oligo B") are used.to amplify a relatively small region of the target DNA (100-200 bp). The resulting amplification product will contain the mutation at its extreme 5' end and a foreign sequence at its 3' end. The 3' sequence is designed to include a unique restriction site (e.g., EcoRI) to aid in the directional cloning of the final amplification fragment (important for purposes of sequencing and archiving the DNA
containing the mutation). The product generated in the upstream or first PCR
may be gel purified if desired prior to the use of this first PCR product in the second PCR; however gel purification is not required once it is established that this fragment is the only species amplified in the PCR.
The small PCR fragment containing the engineered mutation is then used to direct a second round of PCR (PCR 2). In PCR 2, the target DNA is a larger fragment (approximately 1 kb) of the same subcloned region of the p53 cDNA. Because the sequence at the 3' end of the small PCR fragment is not complementary to any of the sequences present in the target DNA, only that strand in which the mismatch is at the extreme 5' end is amplified in PCR 2 (a 3' non-templated arm cannot be extended in PCR).
Amplification is accomplished by the addition of a primer complementary to a region of the target DNA
upstream of the locus of the engineered mutation ("oligo C") and by the addition of a primer complementary to the 5' noncomplementary sequence of the small product of PCR
1 ("oligo D"). By directing amplification from the noncomplementary sequence, this procedure results in the specific amplification of only those sequences containing the mutation.
In order to facilitate cloning of these PCR products into a standard vector, a second unique restriction site can be engineered into oligo C (e.g., HindIII).
The use of this 2-step PCR approach requires that only one primer be synthesized for each mutant to be generated after the initial set-up of the system (i.e., oligo A). Oligos B, C
and D can be used for all mutations generated within a given region. Because oligos C and D
are designed to include different and unique restriction sites, subsequent directional cloning of these PCR products into plasmid vectors (such as pUC 19) is greatly simplified. Selective amplification of only 'those sequences that include the desired mutational change simplifies identification of mutation-containing clones as only verification of the sequence of insert containing plasmids is required. Once the sequence of the insert has been verified, each mutation-containing clone may be maintained indefinitely as a bacterial master stoclc. In addition, DNA stocks of each mutant can be maintained in the form of large scale PCR
preparations. This permits distribution of either bacteria harboring plasmids containing a given mutation or a PCR preparation to be distributed as individual controls in kits containing reagents for the scanning of p53 mutations in clinical samples or as part of a supplemental master p53 mutation library control kit.
An alternative 2-step recombinant PCR is diagrammed in Figure 6, and described in Example 30. In this method two mutagenic oligonucleotides, one for each strand, are synthesized. These oligonucleotides are substantially complementary to each other but are WO 96/15267 PCTlI1595/14673 opposite in orientation.. That is, one is positioned to allow amplification of an "upstream"
region of the DNA, with the mutation incorporated into the 3' proximal region of the,.upper, or sense strand, while the other is positioned to allow amplification of a "downstream"
segment with the intended mutation incorporated into the 5' proximal region of the upper, or sense strand. These two double stranded products share the sequence provided by these mutagenic oligonucleotides. When purified, combined, denatured and annealed, those strands that anneal with recessed 3' ends can be extended or filled in by the action of DNA
polymerase, thus recreating a full length molecules with the mutation in the central region.
This recombinant can be amplified by the use of the "outer" primer pair,those used to make the 5' end of the "upstream" and the 3' end of the "downstream" intermediate fragments.
While extra care must be taken with this method (in comparison with the method described above) because the outer primers can amplify both the recombinant and the un-modified sequence, this method does allow rapid recombinant PCR to be performed using existing end primers, and without the introduction of foreign sequences. In summary, this I S method is often used if only a few recombinations are to be performed.
When large volumes of mutagenic PCRs are to be performed, the first described method is preferable as the first method requires a single oligo be synthesized for each mutagenesis and only recombinants are amplified. -An important feature of kits designed for the identification of p53 mutations in clinical samples is the inclusion of the specific primers to be used for generating PCR
fragments to be analyzed for CFLP. While DNA fragments from 100 to over 1500 by can be reproducibly and accurately analyzed for the presence of sequence polymorphisms by this technique, the precise patterns generated from different length fragments of the same input DNA sequence will of course vary. Not only are patterns shifted relative to one another depending on the length of the input DNA, but in some cases, more long range interactions between distant regions of long DNA fragments may result in the generation of additional cleavage products not seen with shorter input DNA products. For this reason, exact matches with the bar code library will be assured through the use of primers designed to amplify the same size fragment from the clinical samples as was used to generate a given version of the p53 bar code library.
_70_ WO 96/15267 PC"T/US95/14673 C. Detection of Unique CFLPTM Patterns for p53 Mutations The simplest and most direct method of analyzing the DNA fragments produced in the CFLPTM reaction is by gel electrophoresis. Because electrophoresis is widely practiced and easily accessible, initial efforts have been aimed at generating a database in this familiar format. It should, however, be noted that resolution of DNA fragments generated by CFLPTM
analysis is not limited to electrophoretic methods. Mass spectrometry, chromatography, fluorescence polarization, and chip hybridization are all approaches that are currently being refined and developed in a number of research laboratories. Once generated, the CFLPTM
database is easily adapted to analysis by any of these methods.
There are several possible alternatives available for detection of CFLP
patterns. A
critical user benefit of CFLP analysis is that the results are not dependent on the chosen method of DNA detection. DNA fragments may be labeled with a radioisotope (e.g., a ''-P or 3sS_labeled nucleotide) placed at either the S' or 3' end of the nucleic acid or alternatively the label may be distributed throughout the nucleic acid (i.e., an internally labeled substrate). The label may be a nonisotopic detectable moiety, such as a fluorophore which can be detected directly, or a reactive group which permits specific recognition by a secondary agent. CFLP
patterns have been detected by immunostaining, biotin-avidin interactions, autoradiography and direct fluorescence imaging. Since radiation use is in rapid decline in clinical settings and since both immunostaining and biotin-avidin based detection schemes require time-consuming transfer of DNA onto an expensive membrane support, fluorescence-based detection methods may be preferred. It is important to note, however, that any of the above methods may be used to generate CFLP bar codes to be input into the database.
In addition to their being a direct, non-isotopic means of detecting CFLP
patterns.
fluorescence-based schemes offer a noteworthy additional advantage in clinical applications.
CFLP allows the analysis of several samples in the same tube and in the same lane on a gel.
This "multiplexing" permits rapid and automated comparison of a large number of samples in a fraction of the time and for a lower cost than can be realized through individual analysis of each sample. This approach opens the door to several alternative applications.
A researcher could decide to double, triple or quadruple (up to 4 dyes have been demonstrated to be detectable and compatible in a single lane in commercially available DNA
sequencing instrumentation such as the ABI 373/377) the number of samples run on a given gel.
Alternatively, the analyst may include a normal p53 gene sample in each tube.
and each gel lane, along with a differentially labeled size standard, as a internal standard to verify both the presence and the exact locations) of a pattern differences) between the normal p53 gene and putative mutants.
VI. Detection and Identification of Pathogens Using the CFLPTM Method A. Detection and Identification of Hepatitis C Virus Hepatitis C virus (HCV) infection is the predominant cause of post-transfusion non-A, non-B (NANB) .hepatitis around the world. In addition, HCV is the major etiologic agent of hepatocellular carcinoma (HCC) and chronic liver disease world wide. HCV
infection is transmitted primarily to blood transfusion recipients and intravenous drug users although maternal transmission to offspring and transmission to recipients of organ transplants have been reported.
The genome of the positive-stranded RNA hepatitis C virus comprises several regions including 5' and 3' noncoding regions (i. e., 5' and 3' untranslated regions) and a polyprotein coding region which encodes the core protein (C), two envelope glycoproteins (E1 and E2/NS 1 ) and six nonstructural glycoproteins (NS2-NSSb). Molecular biological analysis of the small (9.4 kb) RNA genome has showed that some regions of the genome are very highly conserved between isolates, while other regions are fairly rapidly changeable.
The 5' noncoding region (NCR) is the most highly conserved region in the HCV. These analyses have allowed these viruses to be divided into six basic genotype groups, and then further classified into over a dozen sub-types [the nomenclature and division of HCV
genotypes is evolving; see Altamirano et al., J. Infect. Dis. 171:1034 (1995) for a recent classification scheme]. These viral groups are associated with different geographical areas, and accurate identification of the agent in outbreaks is important in monitoring the disease. While only Group 1 HCV has been observed in the United States, multiple HCV genotypes have been observed in both Europe arid Japan.
The ability to determine the genotype of viral isolates also allows comparisons of the clinical outcomes from infection by the different types of HCV, and from infection by multiple types in a single individual. HCV type has.also been associated with differential efficacy of treatment with interferon, with Group 1 infected individuals-showing little response [Kanai et al., Lancet 339:1543 (1992) and Yoshioka et al., Hepatologv 16:293 (1992)]. Pre-screening of infected individuals for the viral type will allow the clinician to make a more accurate diagnosis, and to avoid costly but fruitless drug treatment.

Existing methods for determining the genotype of HC:V isolates include PCR
amplification of segments of the HCV genome coupled with either DNA sequencing or hybridization to HCV-specific probes, RFLP analysis of PCF, amplified HCV DNA
anything else?. All of these methods suffer from the limitations discussed above (i.
e., DNA sequencing is too labor-intensive and expensive to be practical in clinical laboratory settings; RFLP
analysis suffers from low sensitivity).
Universal and genotype specific primers have been dfaigned for the amplification of HCV sequences from RNA extracted from plasma or serum [Okamoto et al. J. Gen.
Virol.
73:673 (1992);Yoshioka et al., Hepatology 16:293 (1992) and Altamirano c~t al.. supra].
These primers can be used to generate PCR products which serve as substrates in the CFLPTM
assay of the present invention. As shown herein CFLPTM analysis provides a rapid and accurate method of typing HCV isolates. CFLPTM analysis of HCV substrates allows a distinction to be made between the major genotypes and subtypes of HCV thus providing improved methods for the genotyping of HCV isolates.
B. Detection and Identification of Multi-Drug Resistant M. tuberculosis In the past decade there has been a tremendous resurgence in the incidence of tuberculosis in this country and throughout the world. In thc~ United States, the incidence of tuberculosis has risen steadily during past decade, accounting for 2000 deaths annually, with as many as 10 million Americans infected with the disease. The situation is critical in New York City, where the incidence has more than doubled in the past decade, accounting for 14%
of all new cases in the United States in 1990 [Frieden et al., New Engl. J.
Med. 328:521 (1993)].
The crisis in New York City is particularly dire because a significant proportion (as many as one-third) of the recent cases are resistant to one or more antituberculosis drugs [Frieden et al, supra and Hughes, Scrip Magazine May (1994)]. Multi-drug resistant tuberculosis (MDR-TB) is an iatrogenic disease that arises from incomplete treatment of a primary infection [Jacobs, Jr., Clin. Infect. Dis. 19:1 (1994)]. MDR-TB
appears to pose an especially serious risk to the immunocompromised, who are more likely to be infected with MDR-TB strains than are otherwise healthy individuals [Jacobs, Jr., supra].
The mortality rate of MDR-TB in immunocompromised individuals is alarmingly high, often exceeding 90%, compared to a mortality rate of <50% in otherwise uncompromised individuals [Donnabella et al.. Am. J. Respir. Dis. 11:639 (1994)].

WO 96!15267 PCT/US95t14673 From a clinical standpoint, tuberculosis has always been difficult to diagnose because of the extremely long generation time of Mycobacterium tuberculosis as well as the environmental prevalence of other, faster growing mycobacterial species. The doubling. time of M. tuberculosis is 20-24 hours, and growth by conventional methods typically requires 4 to ' 6 weeks to positively identify M. tuberculosis [Jacobs, Jr. et al., Science 260:819 (1993) and Shinnick and Jones in Tuberculosis: Pathogenesis, Protection and Control, Bloom, ed., American Society of Microbiology, Washington, D.C. (1994), pp. 517-530]. It can take an additional 3 to 6 weeks to diagnose the drug susceptibility of a given strain [Shinnick and Jones, supra]. Needless to say, the health risks to the infected individual, as well as to the public, during a protracted period in which the patient may or may not be symptomatic, but is almost certainly contagious, are considerable. Once a drug resistance profile has been elucidated and a diagnosis made, treatment of a single patient can cost up to $250,000 and require 24 months.
The recent explosion int he incidence of the disease, together with the dire risks posed by MDR strains, have combined to spur a burst of research activity and commercial development of procedures and products aimed at accelerating the detection of M. tubes°culosis as well the elucidation of drug resistance profiles of M. tuberculosis clinical isolates. A
number of these methods are devoted primarily to the task of determining whether a given strain is M. tuberculosis or a mycobacterial species other than tuberculosis.
Both culture based methods and nucleic-acid based methods have been developed that allow M.
tuberculosis to be positively identified more rapidly than by classical methods: detection times have been reduced from greater than 6 weeks to as little as two weeks (culture-based 111ethOdS) or two days (nucleic acid-based methods). While culture-based methods are currently in wide-spread use in clinical laboratories, a number of rapid nucleic acid-based methods that can be applied directly to clinical samples are under development. For all of the techniques described below, it is necessary to first "decontaminate" the clinical samples, such as sputum (usually done by pretreatment with N-acetyl L-cysteine and NaOH) to reduce contamination by~ non-mycobacterial species [Shinnick and Jones, supra.]
The polymerise chain reaction (PCR) has been applied to the detection of M.
tzcherculosis and can be used to detect its presence directly from clinical specimens within one to two days. The more sensitive techniques rely on a two-step procedure: the first step is the PCR amplification itself, the second is an analytical step such as hybridization of the _~4_ WO 96!15267 PCT/US95/14673 amplicori to a M. tuberculosis-specific oligonucleotide probe, or analysis by RFLP or DNA
sequencing [Shinnick and Jones, supra].
The Amplified M. tuberculosis Direct Test (AMTDT; Gen-Probe) relies on Transcription Mediated Amplification jTMA; essentially a self sustained sequence reaction (3SR) amplification] to amplify target rRNA sequences directly from clinical specimens.
Once the rRNA has been amplified, it is then detected by a dye-labeled assay such as the PACE2. This assay is highly subject to inhibition by substances present in clinical samples.
The Cycling Probe Reaction (CPR; ID Biomedical). This technique, which is under development as a diagnostic tool for detecting the presence of M.
tuberculosis, measures the accumulation of signal probe molecules. The signal amplification is accomplished by hybridizing tripartite DNA-RNA-DNA probes to target nucleic acids, such as M.
tuberculosis-specific sequences. Upon the addition of RNAse H, the RNA portion of the chimeric probe is degraded, releasing the DNA portions, which accumulate linearly over time to indicate that the target sequence is present [Yule, Bio/Technology 12:1335 (1994)]. The need to use of RNA probes is a drawback, particularly for use in crude clinical samples, where RNase contamination is often rampant.
The above nucleic acid-based detection and differentiation methods offer a clear time savings over the more traditional, culture-based methods. While they are beginning to enter the clinical setting, their usefulness in the routine diagnosis of M.
tuberculosis is still in question, in large part because of problems with associated with cross-contamination and low-sensitivity relative to culture-based methods. In addition, many of these procedures are limited to analysis of respiratory specimens [Yule, Bio/Technology 12:1335 (1994)].
ii) Determination of the antibiotic resistance profile of M. tuberculosis a) Culture-based methods: Once a positi~re identification of M.
tuberczclosis has been made, it is necessary to characterize the extent and nature of the strain's resistance to antibiotics. The traditional method used to determine antibiotic resistance is the direct proportion agar dilution method, in which dilutions of culture are plated on media containing antibiotics and on control media without antibiotics. This method typically adds an additional 2-6 weeks to the time required for diagnosis and characterization of an unknown clinical sample [Jacobs, Jr., supra].
The Luciferase Reporter Mycobacteriophage (LRM) assay was first described in [Jacobs, Jr. et al., Science 260:819 (1993)]. In this assay, a mycobacteriophage containing a cloned copy of the luciferase gene is used to infect mycobacterial cultures.
In the presence of _7j_ WO 96115267 PG"TlUS95/14673 luciferin and ATP, the expressed luciferase produces photons, easily distinguishable by eye or by a luminometer, allowing a precise determination of the extent of mycobacterial growth in the presence of antibiotics. Once sufficient culture has been obtained (usually 10-14 days post-inoculation), the assay can be completed in 2 days. This method suffers from the fact that the LRM are not specific for M. tuberculosis: they also infect M.
smegmatis and M.
bovis (e.g., BCG), thereby complicating the interpretation of positive results. Discrimination between the two species must be accomplished by growth on specialized media which does not support the growth of M. tuberculosis (e.g., NAP media). This confirmation requires another 2 to 4 days.
The above culture-based methods for determining antibiotic resistance will continue to play a role in assessing the effectiveness of putative new anti-mycobacterial agents and those drugs for which a genetic target has not yet been identified. However, recent success in elucidating the molecular basis for resistance to a number of anti-mycobacterial agents, including many of the front-line drugs, has made possible the use of much faster, more accurate and more informative DNA polymorphism-based assays-b) DNA-based methods: Genetic loci involved in resistance to isoniazid, rifampin, streptomycin, fluoroquinolones, and ethionamide have been identified [Jacobs, Jr., supra; Heym et al., Lancet 344:293 (1994) and Morris et al., J. Infect. Dis.
171:954 (1995)].
A combination of isoniazid (inh) and rifampin (rif) along with pyrazinamide and ethambutol or streptomycin, is routinely used as the first line of attack against confirmed cases of M.
tuberculosis [Banerjee et al., Science 263:227 (1994)]. Consequently, resistance to one or more of these drugs can have disastrous implications for short course chemotherapy treatment.
The increasing incidence of such resistant strains necessitates the development- of rapid assays to detect them and thereby reduce the expense and community health hazards of pursuing ineffective, and possibly detrimental, treatments. The identification of some of the genetic loci involved in drug resistance has facilitated the adoption of mutation detection technologies for rapid screening of nucleotide changes that result. in drug resistance. The availability of amplification procedures such as PCR and SDA, which have been successful in replicating large amounts of target DNA directly from clinical specimens, makes DNA-based approaches to antibiotic profiling far more rapid than conventional, culture-based methods.
The most widely employed techniques in the genetic identification of mutations leading to drug resistance are DNA sequencing, Restriction Fragment Length Polymorphism (RFLP). PCR-Single Stranded Conformational Polymorphism (PCR-SSCP), and WO 96/15267 PCT/iJS95/14673 PCR-dideoxyfingerprinting (PCR-ddF). All of these techniques have drawbacks as discussed above. None of them offers a rapid, reproducible means of precisely and uniquely identifyinn individual alleles.
In contrast the CFLPTM method of the present invention provides an approach that relies on structure specific cleavage to generate distinct collections of DNA
fragments. This i method is highly sensitive (>98%) in its ability to detect sequence polymorphisms, and requires a fraction of the time, skill and expense of the techniques described above.
The application of the CFLPTM method to the detection of MDR-TB is illustrated herein using segments of DNA amplified from the rpoB and katG genes. Other genes associated with MDR-TB, including but not limited to those involved in conferring resistance to isoniazid (inhA), streptomycin (rpsL and rrs), and fluoroquinoline (gvrA), are equally well suited to the CFLPTM assay.
C. Detection and Identification of Bacterial Pathogens Identification and typing of bacterial pathogens is critical in the clinical management of infectious diseases. Precise identity of a microbe is used not only to differentiate a disease state from a healthy state, but is also fundamental to determining whether and which antibiotics or other antimicrobial therapies are most suitable for treatment.
Traditional methods of pathogen typing have used a variety of phenotypic features, including growth characteristics, color, cell or colony morphology, antibiotic susceptibility, staining, smell and reactivity with specific antibodies to identify bacteria. All of these methods require culture of the suspected pathogen, which suffers from a number of serious shortcomings, including high material and labor costs, danger of worker exposure, false positives due to mishandling and false negatives due to low numbers of viable cells or due to the fastidious culture requirements of many pathogens. In addition, culture methods require a relatively long time to achieve diagnosis, and because of the potentially life-threatening nature of such infections, antimicrobial therapy is often started before the results can be obtained. In many cases the pathogens are very similar to the organisms that make up the normal flora, and may be indistinguishable from the innocuous strains by the methods cited above. In these cases, determination of the presence of the pathogenic strain may require the higher resolution afforded by more recently developed molecular typing methods.
A number of methods of examining the genetic material from organisms of interest have been developed. One way of performing this type of analysis is by hybridization of species-specific nucleic acid probes to the DNA or RNA from the organism to be tested. This _77_ may be done by immobilizing the denatured nucleic acid to be tested on a membrane support, and probing with labeled nucleic acids that will bind only in the presence of the DNA or RNA from the pathogen. In this way, pathogens can be identified. Organisms can be further differentiated by using the RFLP method described above, in which the genomic DNA is digested with one or more restriction enzymes before electrophoretic separation and transfer to a nitrocellulose or nylon membrane support. Probing with the species-specific nucleic acid probes will reveal a banding pattern that, if it shows variation between isolates, can be used as a reproducible way of discriminating between strains. However, these methods are susceptible to the drawbacks outlined above: hybridization-based assays are time-consuming and may give false or misleading results if the stringency of the hybridization is not well controlled, and RFLP identification is dependent on the presence of suitable restriction sites in the DNA to be analyzed.
To address these concerns about hybridization and RFLP as diagnostic tools, several methods of molecular analysis based on polymerase chain reaction (PCR) amplification have gained popularity. In one well-accepted method, called PCR fingerprinting, the size of a fragment generated by PCR is used as an identifier. In this type of assay. the primers are targeted to regions containing variable numbers of tandem repeated sequences (referred to as VNTRs an eukaryotes). The number of repeats, and thus the length of the PCR
alnplicon, can be characteristic of a given pathogen, and co-amplification of several of these loci in a single reaction can create specific and reproducible fingerprints, allowing discrimination between closely related species. -In some cases where organisms are very closely related, however, the target of the amplification does not display a size difference, and the amplified segment must be further probed to achieve more precise identification. This may be done on a solid support, in a fashion analogous to the whole-genome hybridization described above,- but this has the same problem with variable stringency as that assay. Alternatively, the interior of the PCR
fragment may be used as a template for a sequence-specific ligation event. As outlined above for the LCR, in this method, single stranded probes to be Iigated are positioned along the sequence of interest on either side of an identifying polymorphism, so that the success or failure of the ligation will indicate the presence or absence of a specific nucleotide sequence at that site. With either hybridization or ligation methods of PCR product analysis, knowledge of the precise sequence in the area of probe binding must be obtained in advance, _78_ and differences outside the probe binding area are not detected. These methods are poorly suited to the examination and typing of new isolates that have not been fully characterized.
In the methods of the present invention, primers that recognize conserved regions of bacterial ribosomal RNA genes allow amplification of segments of these genes that include sites of variation. The variations in ribosomal gene sequences have become an accepted method not only of differentiating between similar organisms on a DNA sequence level, but their consistent rate of change allows these sequences to be used to evaluate the evolutionary relatedness of organisms. That is to say, the more similar the nucleic acid is at the sequence level, the more closely related the organisms in discussion are considered to be. [Woese, Bacterial Evolution. Microbiological Reviews, vol 51, No. 2. 1987]. The present invention allows the amplification products derived from these sequences to be used to create highly individual barcodes (i. e., cleavage patterns), allowing the detection of sequence polymorphisms without prior knowledge of the site, character or even the presence of said polymorphisms. With appropriate selection of primers, amplification can be made to be either all-inclusive (e.g., using the most highly conserved ribosomal sequences) to allow comparison of distantly related organisms, or the primers can be chosen to be very specific for a given genus, to allow examination at the species and subspecies level.
While the examination of ribosomal genes is extremely useful in these characterizations, the use of the CFLPTM method in bacterial typing is not limited to these genes. Other genes, including but not limited to those associated with specific growth characteristics, (e.g., carbon source preference, antibiotic resistance, resistance to methycillin or antigen production), or with particular cell morphologies (such as pilus formation) are equally well suited to the CFLPTM
assay.
D. Extraction of Nucleic Acids From Clinical Samples To provide nucleic acid substrates for use in the detection and identification of microorganisms in clinical samples using the CFLPTM assay, nucleic acid is extracted from the sample. The nucleic acid may be extracted from a variety of clinical samples [fresh or frozen tissue, suspensions of cells (e.g., blood), cerebral spinal fluid, sputum, urine, etc.] using a variety of standard techniques or commercially available kits. For example, kits which allow the isolation of RNA or DNA from tissue samples are available from Qiagen, Inc.
(Chatsworth, CA) and Stratagene (LaJolla, CA). For example, the QIAamp Blood kits permit the isolation of DNA from blood (fresh, frozen or dried) as well as bone marrow, body fluids WO 96115267 PC"T/US95/14673 or cell suspensions. QIAamp tissue kits permit the isolation of DNA from tissues such as muscles, organs and tumors. , _ _ .
It has been found that crude extracts from relatively homogenous specimens (such as blood, bacterial colonies, viral plaques, or cerebral spinal fluid) are better suited to severing as templates for the amplification of unique PCR products than are more composite specimens (such as urine, sputum or feces;) [Shibata in PCR: The Polymerase Chaifz Reaction, Mullis et al., eds., Birkhauser, Boston (1994), pp. 47-54]. Samples which contain relatively few copies of the material to be amplified (i. e., the target nucleic acid), such as cerebral spinal fluid, can be added directly to a PCR. Blood samples have posed a special problem in PCRs due to the inhibitory properties of red blood cells. The red blood cells must be removed prior to the use of blood in a PCR; there are both classical and commercially available methods for this purpose [e.g., QIAamp Blood kits, passage through a Chelex 100 column (BioRad), etc.].
Extraction of nucleic acid from sputum, the specimen of choice for the direct detection of M.
tuberculosis, requires prior decontamination to kill or inhibit the growth of other bacterial species. This decontamination is typically accomplished by treatment of the sample with N-acetyl L-cysteine and NaOH (Shinnick and Jones, supra). This decontamination process is necessary only when the sputum specimen is to be cultured prior to analysis.

EXPERIMENTAL
The following examples serve to illustrate certain preferred embodiments and aspects of the present invention and are not to be construed as limiting the scope thereof.
In the disclosure which follows, the following abbreviations apply:°C
(degrees Centigrade); g (gravitational field); vol (volume); w/v (weight to volume);
v/v (volume to volume); BSA (bovine serum albumin); CTAB (cetyltrimethylammonium bromide);
HPLC
(high pressure liquid chromatography); DNA (deoxyribonucleic acid); IVS
(intervening sequence); p (plasmid); p,l (microliters); ml (milliliters); ~.g (micrograms);
pmoles (picomoles); mg (milligrams); MOPS (3-jN-Morpholino]propanesulfonic acid); M
(molar);
mM (milliMolar); ~M (microMolar); nm (nanometers); kdal (kilodaltons); OD
(optical density); EDTA (ethylene diamine tetra-acetic acid); FITC (fluorescein isothiocyanate); SDS
(sodium dodecyl sulfate); NaP04 (sodium phosphate); Tris (tris(hydroxymethyl)-aminomethane); PMSF (phenylmethylsulfonylfluoride); TBE (Tris-Borate-EDTA, i.c., Tris buffer tifrated with boric acid rather than HCl and containing EDTA) ; PBS
(phosphate buffered saline); PPBS (phosphate buffered saline containing 1 mM PMSF); PAGE
(polyacrylamide gel electrophoresis); Tween (polyoxyethylene-sorbitan);
Boehringer Mannheim (Boehringer Mannheim, Indianapolis, IN); Dynal (Dynal A.S., Oslo, Norway);
Epicentre (Epicentre Technologies, Madison, WI); National Biosciences (National Biosciences, Plymouth, MN); New England Biolabs (New England Biolabs, Beverly, MA); Novagen (Novagen, Inc., Madison, WI); Perkin Elmer (Perkin Elmer, Norwalk, CT);
Promega Corp.
(Promega Corp., Madison, WI); RJ Research (RJ Research, Inc., Watertown, MA);
Stratagene (Stratagene Cloning Systems, La Jolla, CA); USB (U.S. Biochemical, Cleveland, OH).

Characteristics Of Native Thermostable DNA Polvmerases A. 5' Nuclease Activity Of DNAPTaq During the polymerase chain reaction (PCR) [Saiki et al., Science 239:487 (1988);
Mullis and Faloona, Methods in Enzymology 155:335 (1987)], DNAPTaq is able to amplify many, but not all, DNA sequences. One sequence that cannot be amplified using DNAPTay is shown in Figure 7 (Hairpin structure is SEQ ID NO:15, PRIMERS are SEQ ID
NOS:16-WO 96115267 PCT/US95ii4673 17.) This DNA sequence has the distinguishing characteristic of being able to fold on itself to form a hairpin with two single-stranded arms, which correspond to the primers used in PCR.
To test whether this failure to amplify is due .to the 5' nuclease activity of the enzyme, we compared the abilities of DNAPTaq and DNAPStf to amplify this DNA sequence during 30 cycles of PCR. Synthetic oligonucleotides were obtained from The Biotechnology Center at the University of Wisconsin-Madison. The DNAPTaq and DNAPStf were from Perkin Elmer (i.e., AmpliTaq DNA polymerase and the Stoffel fragment of Amplitaq DNA
poly-merase). The substrate DNA comprised the hairpin structure shown in Figure 7 cloned in a double-stranded form into pUCl9. The primers used in the amplification are listed as SEQ
ID NOS:16-17.Primer SEQ ID N0:17 is shown annealed to the 3' arm of the hairpin struc-ture in Fig. 7. Primer SEQ ID N0:16 is shown as the first 20 nucleotides in bold on the 5' arm of the hairpin in Fig. 7.
Polymerase chain reactions comprised 1 ng of supercoiled plasmid target DNA, 5 pmoles of each primer, 40 ~,M each dNTP, and 2.5 units of DNAPTaq or DNAPStf, in a 50 ~.l solution of 10 mM Tris Cl pH 8.3. The DNAPTaq reactions included 50 mM KCl and 1.5 mM MgCI,. The temperature profile-was 95°C for 30 sec., 55°C for 1-min. and 72°C for 1 min., through 30 cycles. Ten percent of each reaction was analyzed by gel electrophoresis through 6% polyacrylamide (cross-linked 29:1) in a buffer of 45 mM Tris Borate, pH 8.3, 1.4 mM EDTA (O:SX-TBE).
The results are shown in Figure 8. The expected product was made by DNAPStf (indicated simply as "S") but not by DNAPTaq (indicated as "T"). We conclude that the 5' nuclease-activity of DNAPTaq is responsible for the lack of amplification of this DNA se-quence. - - - . _ . _ _ To test whether the 5' unpaired nucleotides in the substrate region of this structured DNA are removed by DNAPTaq, the fate of the end-labeled 5' arm during four cycles of PCR was compared using the same two polymerases (Figure 9). The hairpin templates, such as the one described in Figure 6, were made using DNAPStf and a 3'P-5"-end-labeled primer.
The 5'-end of the DNA was released as a few large fragments by DNAPTarI but not by DNAPSt~ The sizes of these fragments (based on their mobilities) show that they contain most or all of the unpaired 5' arm of the DNA. Thus, cleavage occurs at or near the base of the bifurcated duplex. These released fragments terminate with 3' OH groups.
as evidenced by direct sequence analysis, and the abilities of the fragments to be extended by terminal deoxynucleotidyl transferase.
_g2_ WO 96115267 PC"TlUS95/14673 Figures 10-12 show the results of experiments designed to characterize the cleavage reaction catalyzed by DNAPTag. Unless otherwise specified, the cleavage reactions com-prised 0.01 pmoles of heat-denatured, end-labeled hairpin DNA (with the unlabeled comple-mentary strand also present), 1 pmole primer (complementary to- the 3' arm) and 0.5 units of DNAPTaq (estimated to be 0.026 pmoles) in a total volume of 10 pl of 10 mM
Tris-Cl, pH

8.5, 50 mM KCl and 1.5 mM MgCI,. As indicated, some reactions had different concentra-tions of KCI, and the precise times and temperatures used in each experiment are indicated in the individual figures. The reactions that included a primer used the one shown in Figure 6 (SEQ ID N0:17). In some instances, the primer was extended to the junction site by providing polymerase and selected nucleotides.
Reactions were initiated at the final reaction temperature by the addition of either the MgCI, or enzyme. Reactions were stopped at their incubation temperatures by the addition of 8 p.l of 95% formamide containing 20 mM EDTA and 0.05% marker dyes (stop solution).
The T", calculations listed were made using the OligoTM primer analysis software from National Biosciences, Inc. These were determined using 0.2 i ~M as the DNA
concentration, at either 1 ~ or 65 mM total salt (the 1.5 mM MgCI, in all reactions was given the value of 15 mM salt for these calculations).
Figure 10 is an autoradiogram containing the results of a set of experiments and conditions on the cleavage site. Figure l0A is a determination of reaction components that enable cleavage. Incubation of 5'-end-labeled hairpin DNA was for 30 minutes at 55°C, with the indicated components. The products were resolved by denaturing polyacrylamide gel electrophoresis and the lengths of the products, in nucleotide s, are indicated. Figure 1 OB .
describes the effect of temperature on the site of cleavage in the absence of added primer.
Reactions were incubated in the absence of KCl for 10 minul:es at the indicated temperatures.
The lengths of the products, in nucleotides, are indicated.
Surprisingly, cleavage by DNAPTag requires neither a primer nor dNTPs (.see Fig.
l0A). Thus, the 5' nuclease activity can be uncoupled from polymerization.
Nuclease activity requires magnesium ions, though manganese ions can be substituted.
albeit with potential changes in specificity and activity. Neither zinc nor calcium ions support the cleavage reaction. The reaction occurs over a broad temperature range. from 2~°C to 85°C.
with the rate of cleavage increasing at higher temperatures.
Still referring to Figure 10, the primer is not elongated in the absence of added dNTPs. However, the primer influences both the site and th~° rate of cleavage of the hairpin.
- g3 -WO 96/15267 PG"T/U595I14673 The change in the site of cleavage (Fig. l0A) apparently results from disruption of a short duplex formed between the arms of the DNA substrate. In the absence of primer, the sequences indicated by underlining in Figure 7 could. pair, forming an extended duplex.
Cleavage at the end of the extended duplex would release the 11 nucleotide fragment seen on the Fig. 1OA lanes with no added primer. Addition of excess primer (Fig. 1OA, lanes 3 and 4) or incubation at an elevated temperature (Fig. lOB) disrupts the short extension of the duplex and results in a longer 5' arm and, hence, longer cleavage products.
The location of the 3' end of the primer can influence the precise site of cleavage.
Electrophoretic analysis revealed that in the absence of primer (Fig. lOB), cleavage occurs at the end of the substrate duplex (either the extended or shortened form, depending on the temperature) between the first and second base pairs. When the primer extends up to the base of the duplex, cleavage also occurs one nucleotide into the duplex. However, when a gap of four or six nucleotides exists between the 3' end of the primer and the substrate duplex, the cleavage site is shifted four to six nucleotides in the 5' direction.
Fig. 11 describes the kinetics of cleavage in the presence (Fig. 11 A) or absence (Fig.
I lB) of a primer oligonucleotide. The reactions were run at 55°C with either 50 mM KCl (Fig. 11A) or 20 mM KCl-(Fig. 11B). The reaction products were resolved by denaturing polyacrylamide gel electrophoresis and the lengths of the products, in nucleotides, are indicated. "M", indicating a marker, is a 5' end-labeled 19-nt oligonucleotide. Under these salt conditions, Figs. 11A and 11B indicate that the reaction appears to be about twenty times faster in the presence of primer than in the absence of primer. This effect on the efficiency may be attributable to' proper alignment and stabilization of the enzyme on the substrate.
The relative influence of-primer on cleavage rates becomes much greater when both reactions are run in 50 mM KCI. In the presence of primer, the rate of cleavage increases with KCl concentration, up to about 50 mM. However, inhibition of this reaction in the presence of primer is apparent at 100 mM and is complete at 150 mM KCI. In contrast, in the absence of primer the rate is enhanced by concentration of KCl up to 20 mM, but it is reduced at concentrations above 30 mM. At 50 mM KCI, the reaction is almost completely inhibited. The inhibition of cleavage by KCl in the absence of primer is affected by temperature, being more pronounced at lower temperatures.
Recognition of the 5' end of the arm to be cut appears to be an important feature of ' substrate recognition. Substrates that lack a free 5' end, such as circular M
13 DNA, cannot be cleaved under any conditions tested. Even with substrates having defined 5' arms. the rate of cleavage by DNAPTaq is influenced by the length of the arm. In the presence of primer and 50 mM KCI, cleavage of a 5' extension that is 27 nucleotides long is essentially complete within 2 minutes at 55°C. In contrast, cleavages of molecules with 5' arms of 84 and 188 nucleotides are only about 90% and 40% complete after 20 minutes. Incubation at higher temperatures reduces the inhibitory effects of long extensions indicating that secondary structure in the 5' arm or a heat-labile structure in the enzyme may inhibit the reaction. A
mixing experiment, run under conditions of substrate excess, shows that the molecules with long arms do not preferentially tie up the available enzyme in non-productive complexes.
These results may indicate that the 5' nuclease domain gains access to the cleavage site at the end of the bifurcated duplex by moving down the 5' arm fram one end to the other. Longer 5' arms would be expected to have more adventitious secondary structures (particularly when KCl concentrations are high), which would be likely to impede this movement.
Cleavage does not appear to be inhibited by long 3' arms of either the substrate strand target molecule or pilot nucleic acid, at least up to 2 kilobases. At the other extreme, 3' arms of the pilot nucleic acid as short as one nucleotide can support cleavage in a primer-independent reaction, albeit inefficiently. Fully paired oligonucleotides do not elicit cleavage of DNA templates during primer extension.
The ability of DNAPTaq to cleave molecules even when the complementary strand contains only one unpaired 3' nucleotide may be useful in optimizing allele-specific PCR.
PCR primers that have unpaired 3' ends could act as pilot oligonucleotides to direct selective cleavage of unwanted templates during preincubation of potential template-primer complexes with DNAPTaq in the absence of nucleoside triphosphates.
S. 5' Nuclease Activities Of Other DNAPs To determine whether other 5' nucleases in other DNAPs would be suitable for the present invention, an array of enzymes, several of which were reported in the literature to be free of apparent 5' nuclease activity, were examined. The ability of these other enzymes to cleave nucleic acids in a structure-specific manner was tested using the hairpin substrate shown in Fig. 7 under conditions reported to be optimal for synthesis by each enzyme.
DNAPEcI and DNAP Klenow were obtained from Promega Corporation; the DNAP of Pyrococcus _ furious ["Pfu", Bargseid et al., Strategies 4:34 ( 1991 )] was from Stratagene; the DNAP of Thermococcus litoralis ["Tli", Vent(exo-), Perler et al., Proc. Natl.
Acad. Sci. USA
89:5577 (1992)] was from New England Biolabs; the DNAF of ThermZrs,flavus ["Tfl", Kaledin et al., Biokhimiya 46:1576 (1981)] was from Epicentre Technologies;
and the DNAP
of Thermus thermophilus ["Tth", Carballeira et crl., Biotechniques 9:276 (1990): Myers et al..
Biochem. 30:7661 (1991)] was from U.S. Biochemicals.
0.5 units of each DNA polymerase was assayed in a 20 ~.l reaction, using either the buffers supplied by the manufacturers for the primer-dependent reactions, or 10 mM Tris Cl, pH 8.5, 1.5 mM MgCl2, and 20 mM KCI. Reaction mixtures were at held 72°C before the addition of enzyme.
Figure 12 is an autoradiogram recording the results of these tests. Figure 12A
demonstrates reactions of endonucleases of DNAPs of several thermophilic bacteria. The reactions were incubated at 55°C for 10 minutes in the presence of primer or at 72°C for 30 minutes in the absence of primer, and the products were resolved by denaturing polyacrylamide gel electrophoresis. The lengths of the products, in nucleotides, are indicated.
Figure 12B demonstrates endonucleolytic cleavage by the 5' nuclease of DNAPEcI. The DNAPEcl and DNAP Klenow reactions were incubated for 5 minutes at 37°C.
Note the light 1 ~ band of cleavage products of 25 and 11 nucleotides in the DNAPEcI lanes (made in the presence and absence of primer, respectively). Figure 12B also demonstrates DNAPTaq reactions in the presence (+) or absence (-) of primer. These reactions were run in 50 mM
and 20 mM KCI, respectively, and were incubated at 55°C for 10 minutes.
Referring to Figure 12A, DNAPs from the eubacteria Thermus -thermophilus and Thermzrs , flavus cleave the substrate at the same place as DNAPTaq, both in the presence and absence of primer. In contrast, DNAPs from the archaebacteria Pyrococcus , furioszts and Thermococcus litoralis are unable to cleave the substrates endonucleolytically. The DNAPs from Pyrococcus,furious and Thermococcus litoralis share little sequence homology with eubacterial enzymes (Ito et al. , Nucl. Acids Res. 19:4045 ( 1991 ); Mathur et al. , Nucl. Acids.
Res. 19:6952 (1991); see also Perler et al.). Referring to Figure 12B, DNAPEcI
also cleaves the substrate, but the resulting cleavage products are difficult to detect unless the 3' exonuclease is inhibited. The amino acid sequences of the 5' nuclease domains of DNAPEcI
and DNAPTaq are about 38% homologous (Gelfand, supra).
The 5' nuclease domain of DNAPTaq also shares about 19% homology with the ~~
exonuclease encoded by gene 6 of bacteriophage T7 [Dune et al., J. Mol. Biol.
166:477 (1983)]. This nuclease, which is not covalently attached to a DNAP
polymerization domain.
is also able to cleave DNA endonucleolytically, at a site similar or identical to the site that is cut by the 5' nucleases described above, in the absence of added primers C. Transcleavage The ability of a 5' nuclease to bedirected to cleave efficiently at any specific sequence was demonstrated in the following experiment. A partially complementary oligonucleotide termed a "pilot oligonucleotide" was hybridized to sequences at the desired point of cleavage.
The non-complementary part of the pilot oligonucleotide provided a structure analogous to the 3' arm of the template (see Figure 7), whereas the 5' region of the substrate strand became the 5' arm. A primer was provided by designing the 3' region of the pilot so that it would fold on itself creating a short hairpin with a stabilizing tetra-loop [Antao et al., Nucl. Acids Res. 19:5901 (1991)]. Two pilot oligonucleotides are shown in Figure 13A.
Oligonucleotides 19-12 (SEQ ID N0:18) and 30-12 (SEQ ID N0:19) are 31 or 42 or nucleotides long, respectively. However, oligonucleotides 19-12 ($EQ ID NO:18) and 34-19 (SEQ ID
N0:19) have only 19 and 30 nucleotides, respectively, that are complementary to different sequences in the substrate strand. The pilot oligonucleotides are calculated to melt off their complements at about 50°C (19-12) and about 75°C (30-12). Both pilots have 12 nucleotides at their 3' ends, which act as 3' arms with base-paired primers attached.
To demonstrate that cleavage could be directed by a pilot oligonucleotide, we incubated a single-stranded target DNA with DNAPTaq in the presence of two potential pilot oligonucleotides. The transcleavage reactions, where the target and pilot nucleic acids are not covalently linked, includes 0.01 pmoles of single end-labeled substrate DNA, 1 unit of DNAPTaq and 5 pmoles of pilot oligonucleotide in a volume of 20 p.l of the same buffers.
These components were combined during a one minute incubation at 95°C, to denature the PCR-generated double-stranded substrate DNA, and the temperatures of the reactions were then reduced to their final incubation temperatures. Oligonucleotides 30-12 and 19-12 can hybridize to regions of the substrate DNAs that are 85 and 27 nucleotides from the 5' end of the targeted strand.
Figure 23 shows the complete 206-mer sequence (SEQ ID N0:26). The 206-mer was generated by PCR . The M13/pUC 24-mer reverse sequencing (-48) primer and the M13/pUC
sequencing (-47) primer from New England Biolabs (catalogue nos. 1233 and 1224 respectively) were used (50 pmoles each) with the pGEM3z(f+) plasmid vector (Promega Corp.) as template (10 ng) containing the target sequences. The conditions for PCR were as ' follows: 50 ~.M of each dNTP and 2.5 units of Taq DNA p~olymerase in 100 ~l of 1X PCR
Buffer (20 mM Tris-Cl, pH 8.3, 1.5 mM MgCI,, 50 mM K~CI with 0.05% Tween-20 and 0.05% NP-40). Reactions were cycled 35 times through 95°C for 45 seconds, 63°C for 4~
_87_ WO 96/15267 PC'T/US95/14673 seconds, then 72°C for 75 seconds. After cycling, reactions were finished off with an incubation at 72°C for 5 minutes. The resulting fragment was purified by electrophoresis through a 6% polyacrylamide gel (29:1 cross link) in a buffer of O.SX TBE (45 mM Tris-Borate, pH 8.3, 1.4 mM EDTA), visualized by ethidium bromide staining or autoradiography, excised from the gel, eluted by passive diffusion, and concentrated by ethanol precipitation.
Cleavage of the substrate DNA occurred in the presence of the pilot oligonucleotide I9-12 at 50°C (Figure 13B, lanes 1 and 7) but not at 75°C (lanes 4 and 10). In the presence of oligonucleotide 30-12 cleavage was observed at both temperatures. Cleavage did not occur in the absence of added oligonucleotides (lanes 3, 6 and 12) or at about 80°C even though at 50°C adventitious structures in the substrate allowed primer-independent cleavage in the absence of KCl (Figure 13B, lane 9). A non-specific oligonucleotide with no complementarity to the substrate DNA did not direct cleavage at SO°C, either in the absence or presence of 50 mM KCl (lanes 13 and 14). Thus, the specificity of the cleavage reactions can be controlled by the extent of complementarity to the substrate and by the conditions of incubation.
D. Cleavage Of RNA -An shortened RNA version of the sequence used in the transcleavage experiments discussed above was tested for its ability to serve as a substrate in the reaction. The RNA is cleaved at the expected place, in a reaction that is dependent upon the presence of the pilot oligonucleotide. The RNA substrate, made- by T7 RNA- polymerase in the presence of [a.-3'P]UTP, corresponds to a truncated version of the DNA substrate used in Figure I3B.
Reaction conditions were similar to those in used for the DNA substrates described above, with 50 mM KCI; incubation was for 40 minutes at 55°C. The pilot oligonucleotide used is termed 30-0 (SEQ ID N0:20) and is shown in Figure 14A.
The results of the cleavage reaction is shown in Figure 14B. The reaction was run either in the presence or absence of DNAPTaq or pilot oligonucleotide as indicated in Figure 14B.
Strikingly, in the case of RNA cleavage, a 3' arm is not required for the pilot oligonucleotide. It is very unlikely that this cleavage is due to previously described RNaseH, -which would be expected to cut the RNA in several places along the 30 base-pair long RNA-DNA duplex. The 5' nuclease of DNAPTaq is a'structure-specific RNaseH that cleaves the RNA at a single site near the 5' end of the heteroduplexed region.
_88_ It is surprising that an oligonucleotide lacking a 3' arm is able to act as a pilot in directing efficient cleavage of an RNA target because such ol.igonucleotides are unable to direct efficient cleavage of DNA targets using native,DNAPs. However, some 5' nucleases of the present invention (for example, clones E, F and G shown in Figure 16) can cleave DNA
in the absence of a 3' arm. In other words, a non-extendable cleavage structure is not required for specific cleavage with some 5' nucleases of the present invention derived from thermostable DNA polymerases.
We tested whether cleavage of an RNA template by L>NAPTczq in the presence of a fully complementary primer could help explain why DNAPTag is unable to extend a DNA
oligonucleotide on an RNA template, in a reaction resembling; that of reverse transcriptase.
Another thermophilic DNAP, DNAPTth, is able to use RNA as a template, but only in the presence of Mn++, so we predicted that this enzyme would not cleave RNA in the presence of this cation. Accordingly, we incubated an RNA molecule with an appropriate pilot oligonucleotide in the presence of DNAPTaq or DNAPTth, in buffer containing either MgT' or Mn'*. As expected, both enzymes cleaved the RNA in the presence of Mg++.
However, DNAPTczq, but not DNAPTth, degraded the RNA in the presence of Mn++. We conclude that the 5" nuclease activities of many DNAPs may contribute to their inability to use RNA as templates:

Generation Of 5' Nucleases From Thermostable DNA Polymerases Thermostable DNA polymerases were generated which have reduced synthetic activity, an activity that is an undesirable side-reaction during DNA cleavage in the detection assay of the invention, yet have maintained thermostable nuclease activity. The result is a thermostable polymerase which cleaves nucleic acids DNA with extreme specificity.
Type A DNA polymerases from eubacteria of the genus Then-mus share extensive protein sequence identity (90% in the polymerization domain, using the Lipman-Pearson method in the DNA analysis software from DNAStar, WI) and behave similarly in both polymerization and nuclease assays. Therefore, we have used the genes for t:he DNA polymerase of Tlzez°m2zs "' aquaticus (DNAPTaq) and Thermus flavus (DNAPTfl) as representatives of this class.
Polymerase genes from other eubacterial organisms, such as ThermZCS
thermophilus, Thez°mzzs~
sp.. Thermotoga maritima, Thermosipho af>"icanzc.s and Bacillzzs stearothermophilus are equally WO 96/15267 PC"T/US95t14673 suitable. The DNA polymerases from these thermophilic organisms are capable of surviving and performing at elevated _ temperatures, and can thus be used in reactions in which temperature is used as a selection against non-specific hybridization of nucleic acid strands.
The restriction sites used for deletion mutagenesis, described below, were chosen for convenience. Different sites situated with similar convenience are available in the Thermiss thermophilus gene and can be used to make similar constructs with other Type A
polymerase genes from related organisms.
A. Creation Of 5' Nuclease Constructs 1. Modified DNAPTaq Genes The first step was to place a modified gene for the Taq DNA polymerise on a plasmid under control of an inducible promoter. The modified Taq polymerise gene was isolated as follows: The Taq DNA polymerise gene was amplified by polymerise chain reaction from genomic DNA from Thermus aquaticus, strain YT-1 (Lawyer et al., supra), using as primers the oligonucleotides described in SEQ ID NOS:13-14. The resulting fragment of DNA has a recognition sequence for the restriction endonuclease EcoRI at the 5' end of the coding sequence and a BgIII sequence at the 3' end. Cleavage with BgIII leaves a 5' overhang or "sticlcy end" that is compatible with the end generated by BamHI. The PCR-amplified DNA
was digested with EcoRI and BamHI. The 2512 by fragment containing the coding region for the polymerise gene was gel purified and then ligated into a plasmid which contains an inducible promoter.
In one embodiment of the invention, the pTTQ 18 vector, which contains the hybrid trp-lac (tic) promoter, was used jM.J.R. Stark, Gene 5:255 (1987) and shown in Figure 15.
The tic promoter is under the control of the E. coli lac repressor. Repression allows the synthesis of the gene product to be suppressed until the desired level of bacterial growth has been achieved, at which point repression is removed by addition of a specific inducer, isopropyl-b-D-thiogalactopyranoside (IPTG). Such a system allows the expression of foreign proteins that may slow or prevent growth of transformants.
Bacterial promoters, such as tic, may not be adequately suppressed when they are present on a multiple copy plasmid. If a highly toxic protein is placed under control of such a promoter, the small amount of expression leaking through can be harmful to the bacteria.
In another embodiment of the invention, another option for repressing synthesis of a cloned gene product was used. The non-bacterial promoter, from bacteriophage T7, found in the plasmid vector series pET-3 was used to express the cloned mutant Taq polymerise genes [Figure 15; Studier and Moffatt, J. Mol. Biol. 189:113 (1986)].- This promoter initiates transcription only by T7 RNA polymerise. In a suitable strain, such as BL21 (DE3)pLYS. the gene for this RNA polymerise is carried on the bacterial genome under control of the lac operator. This arrangement has the advantage that expression of the multiple copy gene (on the plasmid) is completely dependent on the expression of T7 RNA polymerise, which is easily suppressed because it is present in a single copy.
For ligation into the pTTQl8 vector (Figure 15), the PCR product DNA
containing the Taq polymerise coding region (mutTaq, clone 4B, SEQ ID NO:21 ) was digested with EcoRI
and BgIII and this fragment was ligated under standard "sticky end" conditions [Sambrook et al. Molecular Cloning, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp. 1.63-1.69 ( 1989)] into the EcoRI and BamHI sites of the plasmid vector pTTQ 18.
Expression of this construct yields a translational fusion product in which the first two residues of the native protein (Met-Arg) are replaced by three from the vector (Met-Asn-Ser), but the remainder of the natural protein would not change. The construct was transformed into the JM109 strain of E. coli and the transformants were plated under incompletely repressing conditions that do not permit growth of bacteria expressing the native protein. These plating conditions allow the isolation of genes containing pre-existing mutations, such as those that result from the infidelity of Taq polymerise during the amplification process.
Using this amplification/selection protocol, we isolated a clone (depicted in Figure 16B) containing a mutated Taq polymerise gene (mutTirq, clone 4B). The mutant was first detected by its phenotype, in which temperature-stable 5' nuclease activity in a crude cell extract was normal, but polymerization activity was almost absent (approximately less than 1 % of wild type Taq polymerise activity).
DNA sequence analysis of the recombinant gene showed that it had changes in the polymerise domain resulting in two amino acid substitutions: an A to G change at nucleotide position 1394 causes a Glu to Gly change at amino acid position 465 (numbered according to the natural nucleic and amino acid sequences, SEQ Il7 NOS:1 and 4) and another A to G
change at nucleotide position 2260 causes a Gln to Arg change at amino acid position 7~4.
Because the Gln to Gly mutation is at a nonconserved position and because the Glu to Ark mutation alters an amino acid that is conserved in virtually all of the known Type A
polymerises, this latter mutation is most likely the one responsible for curtailing the synthesis activity of this protein. The nucleotide sequence for the clone 4B construct (Figure 16B) is given in SEQ ID N0:21. The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ ID N0:21 is listed in SEQ ID N0:72. , Subsequent derivatives of DNAPTaq constructs were made from the mutTcrg gene, thus, they all bear these amino acid substitutions in addition to their other alterations, unless these particular regions were deleted. These mutated sites are indicated by black boxes at these locations in the diagrams in Figure 16. In Figure 16, the designation "3' Exo" is used to indicate the location of the 3' exonuclease activity associated with Type A
polymerases which is not present in DNAPTaq. All constructs except the genes shown in Figures 16E, F
and G were made in the pTTQ 18 vector.
The cloning vector used for the genes in Figures 16E and F was from the commercially available pET-3 series, described above. Though this vector series has only a BamHI site for cloning downstream of the T7 promoter, the series contains variants that allow cloning into any of the three reading frames. For cloning of the PCR product described above, the variant called pET-3c was used (Figure 17). The vector was digested with BcrrnHl, dephosphorylated with calf intestinal phosphatase, and the sticky ends were filled in using the Klenow fragment of DNAPEc 1 and dNTPs. The gene for the mutant Tack DNAP shown in Figure 16B (mutTaq, clone 4B) was released from pTTQlB by digestion with EcoRI
and .ScrlI, and the "sticky ends" were filled in as was done with the vector. The fragment was ligated to the vector under standard blunt-end conditions (Sambrook et crl., Moleculcrr~
Cloning, supra), the construct was transformed into the BL21(DE3)pLYS strain of E. coli, and isolates were screened to identify those that were ligated with the gene in the proper orientation relative to the promoter. This construction yields another translational fusion product, in which the first two amino acids of DNAPTaq (Met-Arg) are replaced by 13 from the vector plus two from the PCR primer (Met-Ala-Ser-Met-Thr-Gly-Gly-Gln-Gln-Met-Gly-Arg-Ile-Asn-Ser) (SEQ ID
N0:23).
Our goal was to generate enzymes that lacked the ability to synthesize DNA.
but retained the ability to cleave nucleic acids with a 5' nuclease activity. The act of primed, templated synthesis of DNA is actually a coordinated series of events, so it is possible to disable DNA synthesis by disrupting one event while not affecting the others.
These steps include, but are not limited to, primer recognition and binding, dNTP binding and catalysis of the inter-nucleotide phosphodiester bond. Some of the amino acids in the polymerization ' domain of DNAPEcI have been linked to these functions, but the precise mechanisms are as yet poorly defined.

One way of destroying the polymerizing ability of a DMA polymerise is to delete all or part of the gene segment that encodes that domain for the protein, or to otherwise render the gene incapable of making a complete polymerization domain. Individual mutant enzymes may differ from each other in stability and solubility both inside and outside cells. For instance, in contrast to the 5' nuclease domain of DNAPEcI, which can be released in an active form from the polymerization domain by gentle proteolysis [Setlow and Kornberg, J.
Biol. ChenZ. 247:232 (1972)], the Thermus nuclease domain, when treated similarly, becomes less soluble and the cleavage activity is often lost.
Using the mutant gene shown in Figure 16B as starting material, several deletion constructs were created. All cloning technologies were standard (Sambrook et al., sup~cr) and are summarized briefly, as follows:
Figure 16C: The mutTaq construct-was digested with .PstI, which cuts once within the polymerise coding region, as indicated, and cuts immediately downstream of the gene in the multiple cloning site of the vector. After release of the fragment between these two sites, the vector was re-ligated, creating an 894-nucleotide deletion, and bringing into frame a stop codon 40 nucleotides downstream of the junction. The nucleotide sequence of this 5' nuclease (clone 4C) is given in SEQ ID N0:9. The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ ID N0:9 is listed in SEQ ID N0:73.
Figure 16D: The mutTaq construct was digested with .NheI, which cuts once in the gene at position 2047. The resulting four-nucleotide 5' overhanging ends were filled in, as described above, and the blunt ends were re-ligated. The resulting four-nucleotide insertion changes the reading frame and causes termination of translation ten amino acids downstream of the mutation. The nucleotide sequence of this 5' nuclease (clone 4D) is given in SEQ ID
NO:IO. The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ
ID NO:10 is listed in SEQ ID N0:74.
Figure 16E: The entire mutTaq gene was cut from pTTQl8 using EcoRI and SaII
and cloned into pET-3c, as described above. This clone was digested with BstXI and XcmI, at unique sites that are situated as shown in Figure 16E. The DNA was treated with the Klenow fragment of DNAPEcl and dNTPs, which resulted in the 3' overhangs of both sites being trimmed to blunt ends. These blunt ends were ligated together, resulting in an out-of frame deletion of 1540 nucleotides. An in-frame termination codon occurs 18 triplets past the junction site. The nucleotide sequence of this 5' nuclease (clone 4E) is given in SEQ ID
NO:11 [The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ

WO 96!15267 PG"T/US95114673 ID NO:l 1 is listed in SEQ ID N0:75]., with the appropriate leader sequence given in SEQ ID
N0:24 (The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ
ID N0:24 is listed in SEQ ID N0:76. It is also referred to as the CleavaseTM
BX enzyme.
Figure 16F: The entire mutTaq gene was cut from pTTQ 18 using EcoRI and SaII
and cloned into pET-3c, as described above. This clone was digested with BstXI and BamHI, at unique sites that are situated as shown in the diagram. The DNA was treated with the Klenow fragment of DNAPEcl and dNTPs, which resulted in the 3' overhang of the BstXI
site being trimmed to a blunt end, while the 5' overhang of the BamHI site was filled in to make a blunt end. These ends were ligated together, resulting in an in-frame deletion of 903 nucleotides. The nucleotide sequence of the 5' nuclease (clone 4F) is given in SEQ ID
N0:12. It is also referred to as the CleavaseTM BB enzyme. The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ ID NO:1? is listed in SEQ
ID N0:77.
Figure 16G: This polymerase is a variant of that shown in Figure 16E. It was cloned in the plasmid vector pET=21 (Novagen). The non-bacterial promoter from bacteriophage T7, found in this vector, initiates transcription only by T7 RNA polymerase. Sec Studier and Moffatt, sups°a. In a suitable strain, such as (DES)pLYS, the gene for this RNA polymerase is carried on the bacterial genome under control of the lac operator. This arrangement has the advantage that expression of the multiple copy gene (on the plasmid) is completely dependent on the expression of T7 RNA polymerase, which is easily suppressed because it is present in a single copy. Because the expression of these mutant genes is under this tightly controlled promoter, potential problems of toxicity of the expressed proteins to the host cells are less of a concern.
The pET-21 vector also features a "His-Tag", a stretch of six consecutive histidine residues that are added on the carboxy terminus of the expressed proteins. The resulting proteins can then be purified in a single step by metal chelation chromatography, using a commercially available (Novagen) column resin with immobilized NiT' ions. The 2.5 ml columns are reusable, and can bind up to 20 mg of the target protein under native or denaturing (guanidine-HCl or urea) conditions. -E. coli (DES)pLYS cells are transformed with the constructs described above using standard transformation techniques, and used to inoculate a standard growth medium (e.g., Luria-Bertani broth). Production of T7 RNA polymerase is induced during log phase growth ' by addition of IPTG and incubated for a further 12 to 17 hours. Aliquots of culture are removed both before and after induction and the proteins are examined by SDS-PAGE.

Staining with Coomassie Blue allows visualization of the foreign proteins if they account for about 3-5% of the cellular protein and do not co-migrate with any of the major host protein bands. Proteins that co-migrate with major host proteins must be expressed as more than 10°~0 of the total protein to be seen at this stage of analysis.
Some mutant proteins are sequestered by the cells into inclusion bodies.
.These are granules that form in the cytoplasm when bacteria are made to express high levels of a .
forei'n protein. and they can be purified from a crude lysate. and analyzed by SDS-PAGE to determine their protein content. If the cloned protein is found in the inclusion bodies. it must be released to assay the cleavage and polymerise activities. Different methods of solubilization may be appropriate for different proteins, and a variety of methods are known.
.See e.g.. Builder & Ogez. U.S. Patent No. 4,~ 11.502 ( 1980: Olson. U.S.
Patent No.
4.~ 18.26 ( I 98~): Olson & Pai. U.S. Patent No. 4.511.03 ( 1985): Jones of ul.. U.S. Patent No. 4.~ l 2.922 ( 198 ) .
The solubilized protein is then purified on the Ni~ column as described above.
I ~ followin~_ the manufacturers instructions (Novagen). The washed proteins are eluted from the column by a combination of imidazole competitor ( 1 Ml and high salt (0.~ M
NaCI ). and dialyzed to exchange the buffer and to allow denatured proteins to refold.
Typical recoveries result in approximately 20 ~g of specific protein per ml of startin~~ culture.
The DNAP
mutant is rei~erred to as the CleavaseTM BN enzyme and the sequence is given in SEQ ID
2() NO:'_'~. The corresponding amino acid sequence encoded by the nucleotide sequence of SEQ
ID N0:2~ is listed in SEQ ID N0:78.
2. Modified DNAPTfl Gene The DNA polymerise gene of Thermus.navu.s was isolated from rlTO "T
.flcn~tr.v" AT-G'_' strain obtained from the American Type Tissue Collection (ATCC 33923 ). This strain has a different restriction map then does the T. flavus strain used to generate the sequence published by Akhmetzianov and Vakhitov. supra. The published sequence is listed as SEQ
ID N0:2.
No sequence data has been published for the DNA polymerise gene from the AT-62 strain of T. llantr.v.
Genomic DNA from T. ,flavus was amplified using the same primers used to amplify 30 the T. aquaticu.s DNA polymerise gene (SEQ ID NOS:13-14). The approximately 200 base pair PCR fragment was digested with EcoR1 and BamHI. The over-hanging ends were made blunt with the Klenow fragment of DNAPEcl and dNTPs. The resulting approximately 1800 base pair fragment containing the coding region for the N-terminus was ligated into pET-3c.

-9~-as described above. This construct, clone SB, is depicted in Figure 18B. The wild type T.
flavus DNA polymerase gene is depicted in Figure 18A. In Figure 18, the designation " 3' Exo" is used to indicate the location of the 3' exonuclease activity associated with Type A
polymerases which is not present in DNAPTfI. The SB clone has the same leader amino acids ' as do the DNAPTaq clones 4E and F which were cloned into pET-3c; it is not known precisely where translation termination occurs, but the vector has a strong transcription termination signal immediately downstream of the cloning site.
S. Growth And Induction Of Transformed Cells Bacterial cells were transformed with the constructs described above using standard transformation techniques and used to inoculate 2 mls of a standard growth medium (e.g., Luria-Bertani broth). The resulting cultures were incubated as appropriate for the particular strain used, and induced if required for a particular expression system. For all of the constructs depicted in Figures 16 and 18, the cultures were grown to an optical density (at 600nm wavelength) of 0.5 OD.
To induce expression of the cloned genes, the cultures were brought to a final concentration of 0.4 mM IPTG and the incubations were continued for 12 to 17 hours. 50 q l aliquots of each culture were removed both before and after induction and were combined with 20 p,l of a standard gel loading buffer for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Subsequent staining with Coomassie Blue (Sambrook et al., supra) allows visualization of the foreign proteins if they account for about 3-5% of the cellular protein and do not co-migrate with any of the major E. coli protein bands. Proteins that do co-migrate with a major host protein must be expressed as- more than 10% of the total protein to be seen at this stage of analysis.
C. Heat Lysis And Fractionation Expressed thermostable proteins, i.e., the 5' nucleases, were isolated by heating crude bacterial cell extracts to cause denaturation and precipitation of the less stable E. coli proteins.
The precipitated E. coli proteins were then, along with other cell debris, removed by centrifugation. 1.7 mls of the culture were pelleted by microcentrifugation at 12.000 to 14,000 rpm for 30 to 60 seconds. After removal of the supernatant, the cells were resuspended in 400 ~l of buffer A (50 mM Tris-HC1, pH 7.9, 50 mM dextrose, 1 mM
EDTA), re-centrifuged, then resuspended in 80 ql of buffer A with 4 mg/ml lysozyme. The ' cells were incubated at room temperature for 15 minutes, then combined with 80 ~1 of buffer B (10 mM Tris-HC1, pH 7.9, 50 mM KCI, 1 mM EDTA, 1 mM PMSF. 0.5% Tween-20, 0.5% Nonidet-P40).
This mixture was incubated at 75°C for 1 hour to denature and precipitate the~host proteins. This cell extract was centrifuged at 14,000 rpm for 15 minutes at 4°C, and the supernatant was transferred to a fresh tube. An aliquot of 0.5 to 1 ~.l of this supernatant was used directly in each test reaction, and the protein content of the extract was determined by subjecting 7 ~,l to electrophoretic analysis, as above. The native recombinant Taq DNA
polymerase [Englke, Anal. Biochem 191:396 (1990)], and the double point mutation protein shown in Figure 16B are both soluble and active at this point.
The foreign protein may not be detected after the heat treatments due to sequestration of the foreign protein by the cells into inclusion bodies. These are granules that form in the cytoplasm when bacteria are made to express high levels of a foreign protein, and they can be purified from a crude lysate, and analyzed SDS PAGE to determine their protein content.
Many methods have been described in the literature, and one approach is described below.
D. Isolation And Solubilization Of Inclusion Bodies A small culture was grown and induced as described above. A 1.7 ml aliquot was pelleted by brief centrifugation, and the bacterial cells were resuspended in 100 ~.l of Lysis buffer (50 mM Tris-HCl, pH 8.0, 1 mM EDTA, 100 mM NaCI). 2.5 ~l of 20 mM PMSF
were added for a final concentration of 0.5 mM, and lysozyme was added to a concentration of 1.0 mg/ml. The cells were incubated at room temperature for 20 minutes, deoxycholic acid was added to 1 mg/ml ( 1 ~.1 of 100 mg/ml solution), and the mixture was further incubated at 37°C for about 15 minutes or until viscous. DNAse I was added to 10 ~.g/ml and the mixture was incubated at room temperature for about 30 minutes or until it was no longer viscous.
From this mixture the inclusion bodies were collected by centrifugation at 14,000 rpm 2~ for 1 ~ minutes at 4°C, and the supernatant was discarded. The pellet was resuspended in 100 ~.1 of lysis buffer with lOmM EDTA (pH 8.0) and 0.5% Triton X-100. After ~
minutes at room temperature, the inclusion bodies were pelleted as before, and the supernatant was saved for later analysis. The inclusion bodies were resuspended in 50 ~.1 of distilled water, and ~ ~l was combined with SDS gel loading buffer (which dissolves the inclusion bodies) and analyzed electrophoretically, along with an aliquot of the supernatant.
' If the cloned protein is found in the inclusion bodies, it may be released to assay the cleavage and polymerase activities and the method of solubilization must be compatible with the particular activity. Different methods of solubilization may be appropriate for different proteins, and a variety of methods are discussed in Molecula~° Cloning (Sambrook et al., supra). The following is an adaptation we have used for several of our isolates.
20 p,l of the inclusion body-water suspension were pelleted by centrifugation at 14,000 rpm for 4 minutes at room temperature, and the supernatant was discarded. To further wash ' the inclusion bodies, the pellet was resuspended in 20 p.l of lysis buffer with 2M urea, and , incubated at room temperature for one hour. The washed inclusion bodies were then resuspended in 2 p.l of lysis buffer with 8M urea; the solution clarified visibly as the inclusion bodies dissolved. Undissolved debris was removed by centrifugation at 14,000 rpm for 4 minutes at room temperature, and the extract supernatant was transferred to a fresh tube.
I0 To reduce the urea concentration, the extract was diluted into KH,P04. A
fresh tube was prepared containing 180 ~1 of 50 mM KH~P04, pH 9.5, 1 mM EDTA and 50 mM
NaCI.
A 2 q.l aliquot of the extract was added and vortexed briefly to mix. This step was repeated until all of the extract had been added for a total of 10 additions. The mixture was allowed to sit at room temperature for 15 minutes, during which time some precipitate often forms.
Precipitates were removed by centrifugation at 14,000 rpm, for 15 minutes at room temperature, and the supernatant was transferred to a fresh tube. To the 200 ql of protein in the KH,P04 solution, 140-200 ~l of saturated (NH4),SO4 were added, so that the resulting mixture was about 41 % to 50% saturated (NH4)~504. The mixture was chilled on ice for 30 minutes to allow the protein to precipitate, and the protein was then collected by centrifugation at 14,000 rpln, for 4 minutes at room temperature. The supernatant was discarded, and the pellet was dissolved in 20 p.l Buffer C (20 mM HEPES, pH
7.9, 1 mM
EDTA, 0.5% PMSF, 25 mM KCl and 0.5 % each of Tween-20 and Nonidet P 40). The protein solution was centrifuged again for 4 minutes to pellet insoluble materials, and the supernatant was removed to a fresh tube. The protein contents of extracts prepared in this manner were visualized by resolving 1-4 pl by SDS-PAGE; 0.5 to 1 pl of extract was tested in the cleavage and polymerization assays as described.
E. Protein Analysis For Presence Of Nuclease And Synthetic Activity The 5' nucleases described above and shown.in Figures 16 and 18 were analyzed by the following methods.
3p 1. Structure Specific Nuclease Assay A candidate modified polymerase is tested for S' nuclease activity by examining its ability to catalyze structure-specific cleavages. By the term "cleavage structure" as used WO 96/15267 PC"T/US95/14673 herein, is meant a nucleic acid structure which-is a substrate for cleavage by the 5' nuclease activity of a DNAP.
The polymerase is exposed to test complexes that have the structures shown in Figure 19. Testing for 5' nuclease activity involves three reactions: 1 ) a primer-directed cleavage (Figure 19B) is performed because it is relatively insensitive i:o variations in the salt concentration of the reaction and can, therefore, be performed in whatever solute conditions the modified enzyme requires for activity; this is generally the same conditions preferred by unmodified polymerases; 2) a similar primer-directed cleavage is performed in a buffer which permits primer-independent cleavage, i.e., a low salt buffer, to demonstrate that the enzyme is viable under these conditions; and 3) a primer-independent cleavage (Figure 19A) is performed in the same low salt buffer.
The bifurcated duplex is formed between a substrate strand and a template strand as shown in Figure 19. By the term "substrate strand" as used herein, is meant that strand of nucleic acid in which the cleavage mediated by the 5' nuclease activity occurs. The substrate 1 ~ strand is always depicted as the top strand in the bifurcated complex which serves as a substrate for 5' nuclease cleavage (Figure 19). By the term "template strand"
as used herein.
is meant the strand of nucleic acid which is at least partially complementary to the substrate strand and which anneals to the substrate strand to form the cleavage structure. The template strand is always depicted as the bottom strand of the bifurcated cleavage structure (Figure 19).
If a primer (a short oligonucleotide of 19 to 30 nucleotides in. length) is added to the complex. as when primer-dependent cleavage is to be tested, it is designed to anneal to the 3' arm of the template strand (Figure 19B). Such a primer would be extended along the template strand if the polymerase used in the reaction has synthetic activity.
The cleavage structure may be made as a single hairpin molecule, with the 3' end of the target and the 5' end of the pilot joined as a loop as shown in Figure 19E. A primer oligonucleotide complementary to the 3' arm is also required for these tests so that the enzyme's sensitivity to the presence of a primer may be tested.
Nucleic acids to be used to form test cleavage structures can be chemically synthesized, or can be generated by standard recombinant DNA techniques. By the latter method, the hairpin portion of the molecule can be created by inserting into a cloning vector duplicate copies of a short DNA segment, adjacent to each other but in opposing orientation.
The double-stranded fragment encompassing this inverted repeat, and including enough flanking sequence to give short (about 20 nucleotides) unpaired 5' and 3' arms, can then be WO 96/15267 PC"TIUS95/14673 released from the vector by restriction enzyme digestion, or by PCR performed with an enzyme lacking a 5' exonuclease (e.g., the Stoffel fragment of AmplitaqTM DNA
polymerase, VentTM DNA polymerase). .
The test DNA can be labeled on either end, or internally, with either a radioisotope, or ' with a non-isotopic tag. Whether the hairpin DNA is a synthetic single strand or a cloned double strand, the DNA is heated prior to use to melt all duplexes. When cooled on ice, the structure depicted in Figure 19E is formed, and is stable for sufficient time to perform these assays.
To test for primer-directed cleavage (Reaction 1 ), a detectable quantity of the test molecule (typically 1-100 fmol of 3'P-labeled hairpin molecule) and a 10 to 100-fold molar excess of primer are placed in a buffer known to be compatible with the test enzyme. For Reaction 2, where primer-directed cleavage is performed under condition which allow primer-independent cleavage, the same quantities of molecules are placed in a solution that is the same as the buffer used in Reaction 1 regarding pH, enzyme stabilizers (e.g., bovine serum albumin, nonionic detergents, gelatin) and reducing agents (e.g., dithiothreitol, 2-mercaptoethanol) but that replaces any monovalent cation salt with 20 mM KCI;
20 mM hCl is the demonstrated optimum for primer-independent cleavage. Buffers for enzymes, such as DNAPEc 1, that usually operate in the absence of salt are not supplemented to achieve this concentration. To test for primer-independent cleavage (Reaction 3) the same quantity of the test molecule, but no primer, are combined under the same buffer conditions used for Reaction 2.
All three test reactions are then exposed to enough of the enzyme that the molar ratio of enzyme to test complex is approximately 1:1. The reactions are incubated at a range of temperatures up to, but not exceeding, the temperature allowed by either the enzyme stability or the complex stability, whichever is lower, up to 80°C fox enzymes from thermophiles, for a time sufficient to allow cleavage ( 10 to 60 minutes). The products of Reactions 1, 2 and 3 are resolved by denaturing polyacrylamide gel electrophoresis, and visualized by autoradiography or by a comparable method appropriate to the labeling system used.
Additional labeling systems include chemiluminescence detection, silver or other stains, blotting and probing and the like. The presence of cleavage products is indicated by the presence of molecules which migrate at a lower molecular weight than does the uncleaved test "
structure. These cleavage products indicate that the candidate polymerase has structure-specific 5' nuclease activity.

WO 96/15267 PCTlUS95/14673 . To determine whether a modified DNA polymerise has substantially the same 5' nuclease activity as that of the native DNA polymerise, the results of the above-described tests axe compared with the results obtained from these tests performed with the native DNA
polymerise. By "substantially the same 5' nuclease activity" we mean that the modified y 5 polymerise and the native polymerise will both cleave test molecules in the same manner. It is not necessary that the modified polymerise cleave at the same rate as the native DNA
polymerise.
Some enzymes or enzyme preparations may have other associated or contaminating activities that may be functional under the cleavage conditions described above and that may interfere with 5' nuclease detection. Reaction conditions can be modified in consideration of these other activities, to avoid destruction of the substrate, or other masking of the 5' nuclease cleavage and its products. For example, the DNA polymerise I of E. coli (Pol I), in addition to its polymerise and 5' nuclease activities, has a 3' exonuclease that can degrade DNA in a 3' to 5' direction. Consequently, when the molecule in Figure 19E is exposed to this polymerise under the conditions described above, the 3' exonuclease quickly removes the unpaired 3' arm, destroying the bifurcated structure required of a substrate for the 5' exonuclease cleavage and no cleavage is detected. The true ability of Pol I to cleave the structure can be revealed if the 3' exonuclease is inhibited by a change of conditions (e.g., pH), mutation, or by addition of a competitor for the activity. Addition of 500 pmoles of a single-stranded competitor oligonucleotide, unrelated to the Figure 19E
structure, to the cleavage reaction with Pol I effectively inhibits the digestion of the 3' arm of the Figure 19E
structure without interfering with the 5' exonuclease release of the 5' arm.
The concentration of the competitor is not critical, but should be high enough to occupy the 3' exonuclease for the duration of the reaction.
Similar destruction of the test molecule may be caused by contaminants in the candidate polymerise preparation. Several sets of the structure specific nuclease reactions may be performed to determine the purity of the candidate nuclease and to find the window between under and over exposure of the test molecule to the polymerise preparation being investigated.
The above described modified polymerises were tested for 5' nuclease activity as follows: Reaction 1 was performed in a buffer of 10 mM Tr:is-Cl, pH 8.5 at 20°C, l.~ mM
MgCI, and 50 mM KCl and in Reaction 2, the KCl concentration was reduced to 20 mM. In Reactions 1 and 2, 10 fmoles of the test substrate molecule shown in Figure 16E were combined with 1 pmole of the indicated primer and 0.5 to 1.0 pl of extract containing the modified polymerise (prepared as described above). This mixture was then incubated for 10 minutes at 55°C. For all of the mutant polymerises tested these conditions were sufficient to give complete cleavage. When the molecule shown in Figure 19E was labeled at the S' end, the released 5' fragment, 25 nucleotides long, was conveniently resolved on a 20%
polyacrylamide gel ( 19:1 cross-linked) with 7 M urea in a buffer of O.SX TBE.
Clones 4C-F
and SB exhibited structure-specific cleavage comparable to that of the unmodified DNA
polymerise. Additionally, clones 4E, 4F and 4G have the added ability to cleave DNA in the absence of a 3' arm as discussed above. Representative cleavage reactions are shown in Figure 20.
For the reactions shown in Figure 20, the mutant polymerise clones 4E (Taq mutant) and SB (Tfl mutant) were examined for their ability to cleave the hairpin substrate molecule shown in Figure 19E. The substrate molecule was labeled at the 5' terminus with 3'P. Ten fmoles of heat-denatured, end-labeled substrate DNA and 0.5 units of DNAPTuq (lane 1 ) or 1 ~ 0.5 p.l of 4e or Sb extract (Figure 20, lanes 2-7, extract was prepared as described above) were mixed together in a buffer containing 10 mM Tris-CI, pH 8.5, 50 mM KCl and l.~ mM
MgCI,. The final reaction volume was 10 p.l~ Reactions shown in lanes 4 and 7 contain in addition 50 p.M of each dNTP. Reactions shown in lanes 3, 4, 6 and 7 contain 0.2 pM of the primer oligonucleotide (complementary to the 3' arm of the substrate and shown in Figure 19E). Reactions were incubated at 55° C for 4 minutes. Reactions were stopped by the addition of 8 ~.1 of stop solution per 10 ~l reaction volume. Samples were then applied to 12% denaturing acrylamide gels. Following electrophoresis, the gels were autoradiographed.
Figure 20 shows that clones 4E and SB exhibit cleavage activity similar to that of the native DNAPTaq. Note that some cleavage occurs in these reactions in the absence of the primer.
When long hairpin structure, such as the one used here (Figure 19E), are used in cleavage reactions performed in buffers containing 50 mM KCl a low level of primer-independent cleavage is seen. Higher concentrations of KCl suppress, -but do not eliminate, this primer-independent cleavage under these conditions.
2. Assay For Synthetic Activity The ability of the modified enzyme or proteolytic fragments is assayed by adding the modified enzyme to an assay system in which a primer is annealed to a template and DNA ' synthesis is catalyzed by the added enzyme. Many standard laboratory techniques employ such an assay. For example, nick translation and enzymatic sequencing involve extension of a primer along a DNA template by a polymerase molecule.
In a preferred assay for determining the synthetic activity of a modified enzyme an oligonucleotide primer is annealed to a single-stranded DNA template, e.g., bacteriophage M13 DNA, and the primer/template duplex is incubated in the presence of the modified polymerase in question, deoxynucleoside triphosphates (dNTPs) and the buffer and salts known to be appropriate for the unmodified or native enzyme. Detection of either primer extension (by denaturing gel electrophoresis) or dNTP incorporation (by acid precipitation or chromatography) is indicative of an active polymerase. A label, either isotopic or non-isotopic, is preferably included on either the primer or as a dNTP to facilitate detection of polymerization products. Synthetic activity is quantified as the amount of free nucleotide incorporated into the growing DNA chain and is expressed as amount incorporated per unit of time under specific reaction conditions.
Representative results of an assay for synthetic activity is shown in Figure 21. The synthetic activity of the mutant DNAPTaq clones 4B-F was tested as follows: A
master mixture of the following buffer was made: 1.2X PCR buffer, 50 p.M each of dGTP, dATP
and dTTP, 5 ~M dCTP and 0.125 ~M a.-3'-P-dCTP at 600 Ci/mmol. Before adjusting this mixture to its final volume, it was divided into two equal aliquots. One received distilled water up to a volume of 50 p.l to give the concentrations above. The other received 5 yg of single-stranded Ml3mpl8 DNA (approximately 2.5 pmol or 0.05 ~M final concentration) and 250 pmol of M13 sequencing primer (5 pM final concentration) and distilled water to a final volume of 50 pl. Each cocktail was warmed to 75°C for 5 minutes and then cooled to room temperature. This allowed the primers to anneal to the DNA in the DNA-containing mixtures.
For each assay, 4 ~,l of the cocktail with the DNA wa s combined with 1 p.l of the mutant polymerise, prepared as described, or 1 unit of DNAPTaq (Perkin Elmer) in 1 EGl of dH~O. A "no DNA" control was done in the presence of the DNAPTaq (Figure 21, lane 1 ), and a "no enzyme" control was done using water in place of the enzyme (lane 2). Each reaction was mixed, then incubated at room temperature (approx. 22°C) for 5 minutes, then at 55°C for 2 minutes, then at 72°C for 2 minutes. This step incubation was done to detect polymerization in any mutants that might have optimal temperatures lower than 72°C. After the final incubation, the tubes were spun briefly to collect an5~ condensation and were placed on ice. One ~.l of each reaction was spotted at an origin 1.5 cm from the bottom edge of a polyethyleneimine (PEI) cellulose thin layer chromatography plate and allowed to dry. The chromatography plate was run in 0.75 M NaH,P04, pH 3.5, until the buffer front had run approximately 9 cm from the origin. The plate was dried, wrapped in plastic wrap. marked with luminescent ink, and exposed to X-ray film. Incorporation was detected as counts that stuck where originally spotted, while the unincorporated nucleotides were carried by the salt solution from the origin.
Comparison of the locations of the counts with the two control lanes confirmed the lack of polymerization activity in the mutant preparations. Among the modified DNAPTac~
clones, only clone 4B retains any residual synthetic activity as shown in Figure 21.
- E~MPLE 3 5' Nucleases Derived From Thermostable DNA
Polymerases Can Cleave Short Hairpin Structures With Specificity The ability of the 5' nucleases to cleave hairpin structures to generate a cleaved hairpin structure suitable as a detection molecule was examined. The structure and sequence of the hairpin test molecule is shown in Figure 22A (SEQ ID NO:15). The oligonucleotide (the primer in Figure 22A, SEQ ID N0:22) is- shown annealed to its complementary sequence on the 3' arm of the hairpin test molecule. The hairpin test molecule was single-end labeled with 3'P using a labeled T7 promoter primer in a polymerase chain reaction.
The label is present on the 5' arm of the hairpin test molecule and is represented by the star in Figure 22A.
The cleavage reaction was performed by adding 10 fmoles of heat-denatured. end-labeled hairpin test molecule, 0.2 p.M of the primer oligonucleotide (complementary to the 3' arm of the hairpin), 50 ~M of each dNTP and 0.5 units of DNAPTaq (Perkin Elmer) or 0.5 pl of extract containing a 5'nuclease (prepared as described above) in a total volume of 10 ~l in a buffer containing 10 W M Tris-Cl, pH 8.5, 50 mM KCl and 1.5 mM MgCI,.
Reactions shown in lanes 3, 5 and 7 were run in the absence of dNTPs.
Reactions were incubated at 55° C for 4 minutes. Reactions were stopped at 55° C by the addition of 8 ~1 of stop solution per 10 ~l reaction volume. Samples were not heated before loading onto denaturing polyacrylamide gels (10% polyacrylamide, 19:1 crosslinking.
and 7 M urea, in a buffer of 1X TBE [89 mM Tris-borate, pH 8.3, 2.8 mM EDTA]).
The samples were not heated to allow for the resolution of single-stranded and re-duplexed uncleaved hairpin molecules.

Figure 22B shows that altered polymerises lacking any detectable synthetic activity cleave a hairpin structure when an oligonucleotide is annealed to the single-stranded 3' arm of the hairpin to yield a single species of cleaved product (Figure 22B, lanes 3 and 4). 5' - nucleases, such as clone 4D, shown in lanes 3 and 4, produce a single cleaved product even in the presence of dNTPs. 5' nucleases which retain a residual amount of synthetic activity (less y than 1 % of wild type activity) produce multiple cleavage products as the polymerise can extend the oligonucleotide annealed to the 3' arm of the hairpin thereby moving the site of cleavage (clone 4B, lanes 5 and 6). Native DNATaq produces even more species of cleavage products than do mutant polymerises retaining residual synthetic activity and additionally converts the hairpin structure to a double-stranded form in the presence of dNTPs due to the high level of synthetic activity in the native polymerise (Figure 22B, lane 8) Cleavage Of Linear Nucleic Acid Substrates From the above, it should be clear that native (i.e., "wild type") thermostable DNA
polymerises are capable of cleaving hairpin structures in a specific manner and that this discovery can be applied with success to a detection assay. In this example, the mutant DNAPs of the present invention are tested against three different cleavage structures shown in Figure 24A. Structure 1 in Figure 24A is simply single stranded 206-mer (the preparation and sequence information for which was discussed above). Structures 2 and 3 are duplexes;
structure 2 is the same hairpin structure as shown in Figure 13A (bottom), while structure 3 has the hairpin portion of structure 2 removed.
The cleavage reactions comprised 0.01 pmoles of the resulting substrate DNA, and 1 pmole of pilot oligonucleotide in a total volume of 10 p.l of 10 mM Tris-Cl, pH 8.3, 100 mM
KCI, 1 mM MgCh. Reactions were incubated for 30 minutes at 55°C, and stopped by the addition of 8 q.l of stop solution. Samples were heated to 75°C for 2 minutes immediately before electrophoresis through a 10% polyacrylamide gel (19:1 cross lil~l:), with 7M urea. in a buffer of O.SX TBE.
The results were visualized by autoradiography and are shown in Figure 24B
with the ° enzymes indicated as follows: I is native Taq DNAP; II is native Tfl DNAP; III is the CleavaseTM BX enzyme shown in Figure 16E; IV is the Cle<~vaseTM BB enzyme shown in Figure 16F; V is the mutant shown in Figure 18B; and VI is the CleavaseTM BN
enzyme shown in Figure 16G. Structure 2 was used to "normalize" the comparison. For example. it was found that it took 50 ng of Taq DNAP and 300 ng of the CleavaseTM BN
enzyme to give similar amounts of cleavage of Structure 2 in thirty (30) minutes. .-Under these conditions native TayDNAP is unable to cleave Structure 3 to any significant degree.
Native Tfl DNAP ' cleaves Structure 3 in a manner that creates multiple products.
By contrast, all of the mutants tested cleave the linear duplex of Structure 3. This finding indicates that this characteristic of the mutant DNA polymerises is consistent of thermostable polymerises across thermophilic species.

5' Exonucleolvtic Cleavage ("Nibbling"1 By Thermostable DNAPs It has been found that thermostable DNAPs, including those of the present invention, have a true 5' exonuclease capable of nibbling the 5' end of a linear duplex nucleic acid structures. In this example, the 206 base pair DNA duplex substrate is again employed (see above). In this case, it was produced by the use of one 3'P-labeled primer and one unlabeled primer in a polymerise chain reaction. The cleavage reactions comprised 0.01 pmoles of heat-denatured, end-labeled substrate DNA (with the unlabeled strand also present), 5 pmoles of pilot oligonucleotide (see pilot oligos in Figure 13A) and 0.5 units of DNAPTaq or 0.5 ~.1 of the CleavaseTM BB enzyme in the E. coli extract (see above), in a total volume of 10 ~.l of 10 mM Tris~Cl, pH 8.5, 50 mM KCI, 1.5 mM MgCl2.
Reactions were initiated at 65°C by the addition of pre-warmed enzyme, then shifted to the final incubation temperature for 30 minutes. The results are shown in Figure 25A.
Samples in lanes 1-4 are the results with native Taq DNAP, while lanes 5-8 shown the results with the CleavaseTM BB enzyme. The reactions for lanes 1, 2, 5, and 6 were performed at 65°C and reactions for lanes 3, 4, 7, and 8 were performed at 50°C and all were stopped at temperature by the addition of 8 ~1 of 95% formamide with 20 mM EDTA and 0.05%
marker dyes. Samples were heated to 75°C for 2 minutes immediately before electrophoresis through a 10% acrylamide gel (19:1 cross-linked), with 7 M urea, in a buffer of 0.5X
TBE. The expected product in reactions l, 2, 5, and 6 is 85 nucleotides long; in reactions 3 and 7, the expected product is 27 nucleotides long. Reactions 4 and 8 were performed without pilot. and °
should remain at 206 nucleotides. The faint band seen at 24 nucleotides is residual end-labeled primer from the PCR.
d The surprising result is that the CleavaseTM BB enzyme under these conditions causes all of the label to appear in a very small species, suggesting the possibility that the enzyme completely hydrolyzed the substrate. To determine the composition of the fastest-migrating band seen in lanes 5-8 (reactions performed with the deletion mutant), samples of the 206 ~ base pair duplex were treated with either T7 gene 6 exonucle;~se (USB) or with calf intestine alkaline phosphatase (Promega), according to manufacturers" instructions, to produce either labeled mononucleotide (lane a of Figure 25B) or free ''-P-labeled inorganic phosphate (lane b of Figure 25B), respectively. These products, along with the products seen in lane 7 of panel A were resolved by brief electrophoresis through a 20% acrylamide gel (19:1 cross-link), with 7 M urea, in a buffer of O.SX TBE. The CleavaseTM BB enzyme is thus capable of converting the substrate to mononucleotides.

_ Nibblin I~ s Duplex Dependent The nibbling by the enzyme CleavaseTM BB is duplex dependent. In this example, internally labeled, single strands of the 206-mer were produced by 15 cycles of primer extension incorporating a.-3'-P labeled dCTP combined with all four unlabeled dNTPs, using an unlabeled 206-by fragment as a template. Single and double stranded products were resolved by electrophoresis through a non-denaturing 6% polyacrylamide gel (29:1 cross-link) in a buffer of O.SX TBE, visualized by autoradiography, excised from the gel, eluted by passive diffusion, and concentrated by ethanol precipitation.
The cleavage reactions comprised 0.04 pmoles of substrate DNA, and 2 p.l of the enzyme Cleavase BB (in an E. coli extract as described above) in a total volume of 40 l:~l of 10 mM Tris~Cl, pH 8.5, 50 mM KCI, 1.5 mM MgCI,. Reactions were initiated by the addition of pre-warmed enzyme; 10 ~l aliquots were removed at 5, 10, 20, and 30 minutes, and transferred to prepared tubes containing 8 ~,1 of 95% forrnamide with 30 mM EDTA and 0.05% marker dyes. Samples were heated to 75°C for 2 minutes immediately before electrophoresis through a 10% acrylamide gel (19:1 cross-linked), with 7 M
urea, in a buffer of O.SX TBE. Results were visualized by autoradiography as shown in Figure 26.
Clearly, the cleavage by the CleavaseTM BB enzyme depends on a duplex structure; no cleavage of the single strand structure is detected whereas cleavage of the 206-mer duplex is complete.

Purification Of CleavaseTM Enzymes As noted above, expressed thermostable proteins, i.c., the 5' nucleases. were isolated by crude bacterial cell extracts. The precipitated E. coli proteins were then, along with other cell debris, removed by centrifugation. In this example, cells expressing the CleavaseTM BN
clone were cultured and collected (500 grams). For each gram (wet weight) of E. coli, 3 ml of lysis buffer (SO mM Tris-HCI, pH 8.0, 1 mM EDTA, 100 pM NaCI) was added.
The cells were lysed with 200 ~,g/ml lysozyme at room temperature for 20 minutes.
Thereafter deoxycholic acid was added to make a 0.2% final concentration and the mixture was incubated 15 minutes at room temperature.
The lysate was sonicated for approximately 6-8 minutes at 0°C. The precipitate was removed by centrifugation (39,OOOg for 20 minutes). Polyethyleneimine was added (0.5%) to the supernatant and the mixture was incubated on ice for 15 minutes.
The mixture was centrifuged (S,OOOg for 15 minutes) and the supernatant was retained.
This was heated for 30 minutes at 60°C and then centrifuged again (S,OOOg for 15 minutes) and the supernatant was again retained.
The supernatant was precipitated with 35% ammonium sulfate at 4°C for 15 minutes.
The mixture was then centrifuged (S,OOOg for 15 minutes) and the supernatant was removed.
The precipitate was then dissolved in 0.25 M KCI, 20 mM Tris, pH 7.6, 0.2%
Tween and 0.1 EDTA) and then dialyzed against Binding Buffer (8X Binding Buffer comprises:
40 mM
imidazole, 4 M NaCI, 160 mM Tris-HCI, pH 7.9).
The solubilized protein was then purified on the Ni++ column (Novagen). The Binding Buffer was allowed to drain to the top of the column bed and the column was then loaded with the prepared extract. A flow rate of about 10 column volumes per hour is optimal for efficient purification. If the flow rate is too fast, more impurities will contaminate the eluted fraction.
The column was washed with 25 ml ( 10 volumes) of 1 X Binding Buffer and then washed with I S ml (6 volumes) of 1X Wash Buffer (8X Wash Buffer comprises:
480mM
imidazole, 4 M NaCI, 160 mM Tris-HCI, pH 7.9). The bound protein was eluted with 1 ~ ml (6 volumes) of 1X Elute Buffer (4X Elute Buffer comprises: 4 mM imidazole, 2 M
NaCI, 80 mM Tris-HCI, pH 7.9). Protein is then reprecipitated with 35% Ammonium Sulfate as above. The precipitate was then dissolved and dialyzed against: 20 mM Tris, 100 mM KC 1, 1 mM EDTA). The solution was brought up to 0.1 % each of Tween 20 and NP-40 and stored at 4°C.

5' Nucleases Cut Nucleic Acid Substrates At Naturally Occurring Areas Of Secondary Structure The ability of a 5' nuclease to recognize and cleave nucleic acid substrates at naturally occurring areas of secondary structure in the absence of a pilot oligonucleotide (i.e., primer independent cleavage) was shown in Example 1 C (Figure 13, lane 9). When DNAPTaq was incubated at 50°C in the presence of a 206 by DNA substrate (single end labeled, double stranded template) in a buffer containing 10 mM Tris-HCI, pH 8.5 and 1.5 mM
MgCI,, adventitious (i.e., naturally occurring) structures in the DNA substrate were cleaved by the 5' nuclease activity of the enzyme. This cleavage generated three prominent fragments (Figure 13, lane 9); this cleavage pattern provides a "fingerprint" of t:he DNA
template.
The ability of 5' nucleases to cleave naturally occurring structures in nucleic acid templates (structure-specific cleavage) is useful to detect internal sequence differences in nucleic acids without prior knowledge of the specific sequence of the nucleic acid. To develop a general method to scan nucleic acids for mutations [e.g., single base changes (point mutations), small insertions or deletions, etc.] using 5' nucleases, the following series of experiments were performed.
A. The Substitution Of MnCl2 For MgCl2 In The Cleavage Reaction Produces Enhanced Cleavage Patterns The effect of substituting of Mn'+ in place of Mg'- upon the cleavage pattern created by 5' nuclease activity on a double-stranded DNA substrate was examined. A 157 by fragment derived from exon 4 of either the wild-type (SEQ ID N0:27) or the mutant G419R
(SEQ ID N0:28) tyrosinase gene was prepared by PCR as follows.
The primer pair 5' biotin-CACCGTCCTCTT~AAGAAG 3' (SEQ ID N0:29) and 5' fluorescein-CTGAATCTTGTAGATAGCTA 3' (SEQ ID N0:30) was used to prime the PCRs. Tlie synthetic primers were obtained from Promega; the primers were labeled on the 5' end with biotin or fluorescein during synthesis.
The target DNA for the generation of the 157 by fragment of mutant G419R
(Ding, R.A., et al., (1991) Mol. Biol. Med. 8:19; here after referred to as the 419 mutant) was a 339 by PCR product (SEQ ID N0:31) generated using genomic DNA homozygous for the mutation. Genomic DNA was isolated using standard techniques from peripheral blood leukocytes isolated from patients. This 339 by PCR product was prepared as follows. ' The symmetric PCR reaction comprised 10 ng .of genomic DNA from the 419 mutant, 100 pmoles of the primer 5' biotin-GCCTTATTTTACTTTAAAAAT-3' (SEQ ID N0:32), ' 100 pmoles of the primer 5' fluorescein-TAAAGTTTTGTGTTATCTCA-3' (SEQ ID
N0:33), , and 50 p.M of each dNTP in 1X PCR buffer. The primers of SEQ ID NOS:32 and 33 were obtained from Integrated DNA Technologies, Coralville, IA. A tube containing 45 yl of the ' above mixture was overlaid with two drops of light mineral oil and the tube was heated to 95°C for 1 min. Tay polymerase was then added as 1.25 units of enzyme in 5 p,l of 1X PCR
buffer. The tube was heated-to 94°C -for 40 sec, cooled to 55°C
for 50 sec, heated to 72°C
for 70 sec for 29 repetitions with a 5 min incubation at 72°C after the last repetition.
The PCR products were gel purified as follows. The products were resolved by electrophoresis through a 6% polyacrylamide gel (29:1 cross-link) in a buffer containing O.SX
TBE. The DNA was visualized by ethidium bromide staining and the 339 by fragment was excised from the gel. The DNA was eluted from the gel slice by passive diffusion overnight into a solution containing 0.5 M NH40Ac, 0.1% SDS and 0.1 M EDTA. The DNA was then precipitated with ethanol in the presence of 4 ~.g of glycogen carrier. The DNA was pelleted and resuspended in 40 ~,l of TE (10 mM Tris-Cl, pH 8.0, 0.1 mM EDTA).
To generate the 157 by fragment from the 419 mutant, the purified 339 by 419 PCR
fragment was used as the target in an asymmetric PCR. The asymmetric PCR
comprised 100 pmoles of the biotinylated primer of SEQ ID N0:32, 1 pmole of the fluoresceinated primer of SEQ ID N0:33, 50 p.M of each dNTP, in 1X PCR buffer. A tube containing 45 ~.t of the above mixture was overlaid with two drops of light mineral oil and the tube was heated to 95°C for 5 sec and then cooled to 70°C. Taq polymerase was then added as 1.25 units of enzyme in 5 p.l of 1X PCR buffer. The tube was heated to 95°C for 45 sec. cooled to 50°C
for 45 sec, heated to 72°C -for 1 min 15 sec for 30 repetitions with a 5 min incubation at 72°C -after the last repetition.
The asymmetric PCR products were gel purified as follows. The products were resolved by electrophoresis through a 6% polyacrylamide gel (29:1 cross-link) in a buffer containing O.SX TBE. The DNA was visualized by ethidium bromide staining; the double-stranded DNA was differentiated from the single-stranded DNA due to the mobility shift commonly seen with single-stranded DNA produced from asymmetric PCR (in an asymmetric PCR both single-stranded and double-stranded products are produced; -typically the single-WO 96!15267 PCT/US95/14673 stranded product will have a slower speed of migration through the gel and will appear closer to the origin than will the double-stranded product). The double-stranded 1~7 by substrate corresponding to the 419 mutant (SEQ ID N0:28) was excised from the gel.
The 157 by wild-type fragment was generated by asymmetric PCR as described above for the 419 mutant with the exception that the target DNA was 10 ng of supercoiled pcTYR-NlTyr plasmid DNA. The pcTYR-NlTyr plasmid contains the entire wild-type tyrosinase cDNA [Geibel, L.B., et al. (1991) Genomics 9:435].
Following the asymmetric PCRs, the reaction products were resolved on an acrylamide gel and the double-stranded fragments of interest were excisf:d, eluted and precipitated as described above. The precipitated 157 by wild-type (SEQ ID N0:27) and 419 mutant (SEQ
ID N0:28) fragments were resuspended in 40 ~l of TE.
Cleavage reactions comprised 100 fmoles of the resulting double-stranded substrate DNAs (the substrates contain a biotin moiety at the 5' end of the sense strand) in a total volume of 10 ~1 of 10 mM MOPS, pH 8.2; 1 mM divalent ration (either MgCI, or MnCI~) and 1 unit of DNAPTaq. The reactions were overlaid with a drop of light mineral oil.
Reactions were heated to 95°C for 5 seconds to denature the substrate and then the tubes were quickly cooled to 65°C (this step allows the DNA assume its unique secondary structure by allowing the formation of intra-strand hydrogen bonds between complimentary bases). The reaction can be performed in either a thermocycler (MJ Research, Watertown, MA) programmed to heat to 95°C for 5 seconds then drop the temperature immediately to 65°C or alternatively the tubes can be placed manually in a heat block set at 95°C and then transferred to a second heat block set at 65°C.
The reaction was incubated at 65°C for 10 minutes and was stopped by the addition of 8 p.l of stop buffer. Samples were heated to 72°C for 2 minutes and 5 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated allowing the gel to remain flat on one plate. A 0.2 p.m-pore positively-charged nylon membrane (Schleicher and Schuell, Keene, NH), pre-wetted in O.SX TBE, was laid on top of the exposed acrylamide gel. All air bubbles trapped between the gel and the membrane were removed. Two pieces of 3MM filter ' paper (Whatman) were then placed on top of the membrane, the other glass plate was replaced, and the sandwich was clamped with binder clips. Transfer was allowed to proceed overnight. After transfer, the membrane was carefully peeled from the gel and allowed to air WO 96/15267 PG"TIUS95/14673 dry. After complete drying, the membrane was washed in 1.2X Sequenase Images Blocking Buffer (United States Biochemical) fox 30 minutes. Three tenths of a ml of the buffer was used per cm' of membrane. A streptavidin-alkaline pliosphatase conjugate (SAAP, United States Biochemical) was added to a 1:4000 dilution directly to the blocking solution, and ' agitated for 15 minutes. The membrane was rinsed briefly with HBO and then washed 3 times (5 minutes/wash) in 1X SAAP buffer (100 mM Tris-HCL, pH 10; 50 mM NaCI) with 0.1%
sodium dodecyl sulfate (SDS) using 0.5 ml buffer/cm'- of the buffer, with brief HBO rinses between each wash. Similarly, for fluorescein-labeled DNA, anti-fluorescein fragment (Boehringer Mannheim Biochemicals, Indianapolis, IN) at a 1:20.000 final dilution maybe added followed by three washes (5 min/wash) in 1X SAAP buffer containing 0.1%
SDS and 0.025% Tween 20. The membrane was then washed once in 1X SAAP buffer without SDS, drained thoroughly and placed in a plastic heat-sealable bag. Using a sterile pipet tip, 0.05 ml/cm'- of CDP-Star'"' (Tropix, Bedford, MA) was added to-the bag and distributed over the entire membrane for 5 minutes. The bag was drained of all excess liquid and air bubbles.
The membrane was then exposed to X-ray film (Kodax XRP) for an initial 30 minutes.
Exposure times were adjusted as necessary for resolution and clarity. The results are shown in Figure 27. - -In Figure 27, the lane marked "M" contains molecular weight markers. The marker fragments were generated by digestion of pUCl9 with HaeIII followed by the addition of biotinylated dideoxynucleotides (Boehringer Mannheim, Indianapolis, IN) to the cut ends using terminal transferase (Promega). Lanes 1, 3 and 5 contain the reaction products from the incubation of the wild type 157 nucleotide substrate in the absence of the DNAPTaq enzyme (lane 1), in the presence of MgCh and enzyme (lane 3) or in the-presence of MnCh and enzyme (lane 5). Lanes 2, 4 and 6 contains the reaction products from the incubation of the ?5 157 nucleotide substrate derived from the 419 mutant in the absence of enzyme (lane 2), in the presence of MgChand enzyme (lane 4) or in the presence of MnCI, and enzyme (lane 6).
Figure 27 demonstrates that the use of MnCh rather than MgCI, in the cleavage reaction results in the production of an enhanced cleavage pattern. It is desirable that the cleavage products are of different sizes so that the products do not all cluster at one end of the gel. The ability to spread the cleavage products out over the entire length of the gel makes it more likely that alterations in cleavage products between the wild type and mutant ' substrates will be identified. Figure 27 shows that when Mg'~' is used as the divalent canon, the majority of the cleavage products cluster together in the upper portion of the gel. In contrast when Mn'+ is used as the divalent cation, the substrate assumes structures which, when cleaved, generate products of widely differing mobilities. These results show that Mn'-is the preferred divalent cation far the cleavage reaction.
B. 5' Nuclease Cleavage Of Different But Similarly Sized DNAs. Generates Unique Cleavage Fragments The ability of 5' nuclease to generate a cleavage pattern or "fingerprint"
which is unique to a given piece of DNA was shown by incubating four similarly sized DNA
substrates with the Cleavase'~"' BN enzyme. The four DNA substrates used were a 157 nucleotide fragment from the sense (or coding) strand of exon 4 of the wild-type tyrosinase gene (SEQ ID N0:34); a 157 nucleotide fragment from the anti-sense (or non-coding) strand of exon 4 of the wild-type tyrosinase gene (SEQ ID N0:35); a 165 nucleotide DNA fragment derived from pGEM3Zf(+) (SEQ ID N0:36) and a 206 nucleotide DNA fragment derived from pGEM3Zf(+) (SEQ ID N0:37). The DNA substrates contained either a biotin or fluorescein label at their 5' or 3' ends. The substrates were made as follows.
To produce the sense and anti-sense single-stranded substrates corresponding to exon 4 of the wild-type tyrosinase gene, a double-stranded DNA fragment, 157 nucleotides in length (SEQ ID N0:27), was generated using symmetric PCR. The target for the symmetric PCR
was genomic DNA containing the wild-type tyrosinase gene. The symmetric PCR
comprised 50-100 ng of genomic wild-type DNA, 25 pmoles each of primers SEQ ID NOS:42 and 43, 50 pM each dNTP and 1.25 units of Taq polymerase in 50 pl of 1X PCR buffer.
The reaction mixture was overlaid with two drops of light mineral oil and the tube was heated to 94°C for 30 sec, cooled to 50°C for 1 min, heated to 72°C
for 2 min for 30 repetitions. The double-stranded PCR product was gel purified, precipitated and resuspended in 40 p.l of TE
buffer as described above in a).
The single-stranded sense and anti-sense 157 nucleotide DNA fragments were generated using the above 157 by wild-type DNA fragment (SEQ ID N0:27) in two asymmetric PCR reactions. The sense strand fragment was generated using 5 p.l of the above purified 157 by fragment (SEQ ID N0:27) as the target in an asymmetric PCR.
The reaction mixtures for the asymmetric PCR were as above for the syrmnetric PCR with the exception that 100 pmoles of the biotin-labeled sense primer (SEQ ID N0:29) and 1 pmole of the fluorescein-labeled anti-sense primer (SEQ ID N0:30) was used to prime the reaction. The anti-sense fragment was generated using 5 pl of the above purified 157 by fragment as the target in an asymmetric PCR. The reaction conditions for the asymmetric PCR
were as above for the symmetric PCR with the exception that 1 pmole of the sense primer (SEQ
ID N0:29) and 100 pmoles of the anti-sense primer (SEQ ID N0:30) was used to prime the reaction.
The reaction conditions for the asymmetric PCR were 95°C for 45 sec,.50°C for 45 sec, 72°C for 1 min and 15 sec for 30 repetitions with a 5 min incubation at 72°C after the last repetition. The reaction products were visualized. extracted and collected as described above with the single stranded DNA being identified by a shift in mobility when compared to a double stranded DNA control.
The single-stranded 165 nucleotide fragment from pGEM3Zf(+) (SEQ ID N0:36) was generated by asymmetric PCR. The PCR comprised 50 pmoles of 5' biotin-AGCGGATAACAATTTCACACAGGA-3' (SEQ ID N0:38: Promega) and 1 pmole of 5'-CACGGATCCTAATACGACTCACTATAGGG-3' (SEQ ID NO:39; Integrated DNA
Technologies. Coralville, IA), 50 ~M each dNTP, in 1 X PCR buffer. Forty-five microliters of this reaction mixture was overlaid with two drops of light mineral oil and the tube was heated to 95°C for 5 sec and then cooled to 70°C. Taq polymerase was then added at 1.25 units in 5 ~l of 1X PCR buffer. The tubes were heated to 95°C fox 45 sec, cooled to 50°C
for 45 sec, heated to 72°C for 1 min 15 sec for 30 repetitions with a 5 min incubation at 72°C after the last repetition. The reaction products were visualized, extracted and collected as described above with the 164 nucleotide DNA fragment being identified by a shift in mobility when compared to a double stranded DNA control.
The 206 nucleotide DNA fragment (SEQ ID N0:37) was prepared by asymmetric PCR, performed as described above, using 1 pmole of a double-stranded 206 by PCR product (generated as described in Example 1C), and 50 pmoles of the primer 5'-CGCCAGGGTTTTCCCAGTCACGAC-3' (SEQ ID N0:40). The tubes were heated to 95°C
for 45 sec, cooled to 63°C for 45 sec, heated to 72°C for I min I S sec for I S repetitions with a 5 min incubation at 72°C after the last repetition. The reaction products were visualized.
extracted and collected as described above with the 206_ nucleotide DNA
fragment being identified by a shift in mobility when compared to a double stranded DNA
control. The precipitated DNA was resuspended in 70 p.l of TE buffer. _ Twenty-five microliters of the above product was biotinylated on the 3' end using 10-20 units of terminal deoxynucleotidyl transferase (TdT) (Promega) in a 50 ~l reaction. The reaction comprised 0.5 nmoles of biotin-16-ddUTP(Boehringer Mannheim) and 1X
TdT
buffer (500 mM -cacoodylate buffer, pH 6.8, 5 mM CoCI,, 0.5 mM DTT and 500 ~.g/ml °

BSA). The tubes were incubated at 37°C for 15 min followed by ethanol precipitation in the presence of 4 ~g of glycogen. The DNA was ethanol precipitated a second time and then resuspended in 25 q.l of TE. - . ' The cleavage reactions were carried out in a final volume of 10 q,l of 10 mM
MOPS, pH 8.2, with 1 mM MnCh using approximately 100 fmoles of substrate DNA and 250 ng of the enzyme CleavaseTT'' BN. Parallel reactions lacking the enzyme CleavaseT"' BN (no enzyme control) were set up as above with the exception that one third as much DNA
template was used (approximately 33 fmoles of each template) to balance the signal on the autoradiograph.
Each substrate DNA was placed in a 200 ~l thin wall microcentrifuge tube (BioRad, Hercules, CA) in 5 ~1 of 10 mM MOPS, pH 8.2, with 2 mM MnCh. The solution was overlaid with one drop of light mineral oil. Tubes were brought to 95°C
for ~ seconds to denature the substrates and then the tubes were quickly cooled to 65°C.
Cleavage reactions were started immediately by the addition of a diluted enzyme mixture comprising 1 pl of the enzyme CleavaseTM BN [250 ng/q.l in 1X dilution buffer (0.5% NP40, 0.5% Tween20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 mg/ml BSA)] in ~l of 10 mM MOPS, pH 8.2, without MnCh. The enzyme solution was at room temperature before addition to the cleavage reaction. After 5 minutes at 65°C, the reactions were stopped by the addition of 8 p.l of stop buffer. Samples were heated t:o 72°C
for 2 minutes and 5 p.l of each reaction were resolved by electrophoresis through a I 0%
polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a 0.45 q.m-pore positively charged nylon membrane (United States Biochemical). The DNA was transferred to the membrane and the membrane was dried, washed in 1.2X Sequenase Images Blocking Buffer, treated with 1 X SAAP buffer as described above. The signal was developed using Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega) in place of the CDP-Star~'~"''; the membrane was then exposed to X-ray film as described above. The resulting autoradiograph is shown in Figure 28.
Figure 28 shows the results of incubation of the four substrates described above in the presence or absence of the CleavaseT"'' BN enzyme. Four sets of reactions are shown. Set one contains the reaction products from the incubation of the 157 nucleotide sense strand fragment of the tyrosinase gene (SEQ ID N0:34) in the absence or presence of the CleavaseTM BN enzyme. Set two contains the reaction products from the incubation of the - 11~ -157 nucleotide anti-sense strand fragment of the tyrosinase gene (SEQ ID
N0:35) in the absence or presence of the CleavaseT"~' BN enzyme. Set three contains the reaction products from the incubation of the 165 base bottom strand fragment of the plasmid pGEM3Zf(+) (SEQ ID N0:36) in the absence or presence of the CleavaseT"' BN enzyme. Set four contains ' the reaction products from the incubation of the 206 base top strand fragment of the plasmid pGEM3Zf(+) (SEQ ID N0:37) in the absence or presence of the CleavaseTM BN
enzyme.
Lanes marked "M" contain biotin-labeled molecular weight markers prepared as described above; the sizes of the marker fragments are indicated in Figure 28. In the absence of the CleavaseT"'' BN enzyme, no cleavage of the substrates is observed. In the presence of the CleavaseTM BN enzyme, each substrate is cleaved generating a unique set of cleavage products. When these cleavage products are resolved on a polyacrylamide gel, a unique pattern or fingerprint is seen for each substrate DNA. Thus, although the four substrates are similar in size ( 157 to 206 bases), the CleavaseT"' BN enzyme generates a unique collection of cleavage products from each substrate. These unique cleavage patterns result from the 1 ~ characteristic conformation each substrate DNA assumes.
The present invention contemplates the ability to generate a unique cleavage pattern for two or more DNA substrates of the same size as part of a method for the detection of genetic mutations. This method compares a normal (or wild type or non-mutated) substrate with a substrate from a patient suspected of having a mutation in that substrate. The two substrates would be of the same length and the cleavage reaction would be used to probe the patient DNA substrate for conformational changes relative to the pattern seen in the wild type control substrate. -Cleavage Directed By The CleavaseT"' BN Enzyme Can Detect Single Base Changes In DNA Substrates The ability of the CleavaseT"'' BN enzyme to cleave DNA substrates of the same size but which contain single base changes between the substrates is herein demonstrated. The human tyrosinase gene was chosen as a model system because numerous single point.
mutations have been identified in exon 4 of this gene [Spritz, R.A. (1994) Human Molecular Genetics 3:1469]. Mutation of the tyrosinase gene leads to oculocutaneous albinism in humans.

Three single-stranded substrate DNAs were prepared; the substrates contain a biotin label at their 5' end. The wild type substrate comprises the 157 nucleotide fragment from the sense strand of the human tyrosinase gene [(SEQ ID N0:34); Geibel, L.B., et al. ( 1991 ) Genomics 9:435]. Two mutation-containing substrates were used. The 419 substrate (SEQ
ID N0:41 ) is derived from the tyrosinase mutant G419R which contains a glycine (GGA) to arginine (AGA) substitution; this mutant differs from the wild-type exon 4 fragment by a single base change at nucleotide 2675 [King, R.A., et al. ( 1991 ) Mol. Biol.
Med. 8:19]. The 422 substrate (SEQ ID N0:42) is derived from the tyrosinasc~ mutant R422Q
which contains an arginine (CGG) to glutamine (CAG) substitution; this mutant differs from the wild type exon 4 fragment by a single base change at nucleotide 2685 [Giebel, L.B., et al. (1991) J.
Clin. Invest. 87:1119].
Single-stranded DNA containing a biotin label at the 5" end was generated for each substrate using asymmetric PCR as described in Example 8a with the exception that the single-stranded PCR products were recovered from the gel rather than the double-stranded products.
The following primer pair was used to amplify each DNA (the 419 and 422 mutations are located internally to the exon 4 fragment amplified by the primer pair thus the same primer pair can be used to amplify the wild type and two mutant templates).
The primer listed as SEQ ID N0:29 sense primer) contains a biotin label at the 5' end and was used in a 100-fold excess over the anti-sense primer of SEQ ID N0:30.
To generate-the single stranded substrates the following templates were used.
Ten ng of supercoiled plasmid DNA was used as the target to generate the wild-type (plasmid pcTYR-NlTyr) or 422 mutant (plasmid pcTYR-A422) 157 nucleotide fragments. Five microliters of the gel purified 339 by PCR fragment (SEQ ID N0:31 ) derived from genomic DNA homozygous for the 419 mutation (described in Example 8a) was used as the target to generate the 157 nucleotide 419 mutant fragment (SEQ ID N0:41).
For each target DNA, the asymmetric PCR comprised 100 pmoles of SEQ ID N0:29 and I pmole of SEQ ID N0:30, and 50 ~.M each dNTP in 1X PCR buffer. The reaction mixture (45 p.l) was overlaid with two drops of light mineral oil and the tubes were heated to 95°C for 5 sec then cooled to 70°C. Tag polymerise was then added as 1.25 units of enzyme in 5 ~1 of 1X PCR buffer. The tubes were heated to 95°C for 45 sec.
cooled to 50°C for 4~
sec, heated to 72°C for 1 min 15 sec for 30 repetitions with a 5 min incubation at 72°C after WO 96/15267 PC"T/US95/14673 the last repetition. The single stranded PCR products were gel purified, precipitated and resuspended in 40 ~,l of TE buffer as described above. .
Cleavage reactions were performed as descibed in Example 8b. The samples were heated to -72°C for 2 minutes and 7 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE. After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP and CDPStar as described in Example 8a. The resulting autoradiograph is shown in Figure 29.
In Figure 29, lanes marked "M" contain molecular weight markers prepared as described in Example 8. Lanes 1-3 contain the no enzyme control for the wild type (SEQ ID
N0:34), the 419 mutant (SEQ ID N0:41) and the 422 mutant (SEQ ID N0:42) substrates, respectively. Lane 4 contains the cleavage products from the wild type template. Lane 5 contains the cleavage products from the 419 mutant. Lane 6 contains the cleavage products from the 422 mutant.
Figure 29 shows that a similar, but distinctly different, pattern of cleavage products is generated by digestion of the three template DNAs with the CleavaseTM BN
enzyme. Note that in the digest of mutant 419, the bands below about 40 nucleotides are absent. when compared to wild-type, while in the digest of mutant 422 several new bands appear in the 53 nucleotide range.
Although the three template DNAs differed in only one of the 157 nucleotides, a unique pattern of cleavage-fragments was generated for each. Thus a single base change in a 157 nucleotide fragment gives rise to different secondary structures which are recognized by the CleavaseTM enzyme.
-__ EXAMPLE 10 Single Base Changes In Large DNA
Fragments Are Detected By The Enzyme CleavaseTM BN
The previous example demonstrated that the 5' nuclease activity of the CleavaseTM BN
enzyme could be used to detect single point mutations within a 157 nucleotide DNA
fragment. The ability of the CleavaseT"' BN enzyme to detect single point mutations within , larger DNA fragments is herein demonstrated.
Increasingly larger fragments derived from the 422 tyrosinase mutant was compared to the same size fragments derived from the wild-type tyrosinase gene. Four sets of single-stranded substrates were utilized: 1 ) a 157 nucleotide template derived from the sense strand of exon 4 from the wild-type (SEQ ID N0:34) and 422 mutant (SEQ ID N0:42), 2) a 378 nucleotide fragment containing exons 4 and 5 from the wild-type (SEQ ID N0:43) and 422 mutant (SEQ ID N0:44), 3) a 1.059 kb fragment containing exons 1-4 from the wild-type (SEQ ID N0:45) and 422 mutant (SEQ ID N0:46) and 4) a 1.587 kb fragment containing exons 1-5 from the wild-type (SEQ ID N0:47) and 422 mutant (SEQ ID N0:48). The only difference between the wild type and 422 mutant templates is the G to A change in exon 4 regardless of the length of the template used. The G to A point mutation is located 27, 27, 929 and 1237 nucleotides from the labeled ends of the 157 base, 378 base, 1.059 kb and 1.6 lcb substrate DNAs, respectively.
A) Preparation Of The Substrate DNA
A cDNA clone containing either the wild-type [pcTYR-NlTyr, Bouchard, B., et crl.
( 1989) J. Exp. Med. 169:2029] or 422 mutant [pcTYR-A422, Giebel, L.B., et al.
( 1991 ) 87:1119] tyrosinase gene was utilized as the target DNA in PCRs to generate the above substrate DNAs. The primer pair consisting of SEQ ID NOS:42 and 43 were used to generate a double stranded 157 by DNA fragment from either the mutant of wild-type cDNA
clone.
The primer pair consisting of SEQ ID N0:29 and SEQ ID N0:49 was used to generate a double stranded 378 by DNA fragment from either the wild-type or mutant cDNA
clone. The primer pair consisting of SEQ ID NO:50 and SEQ ID N0:3G was used to generate a double stranded 1.059 kbp DNA fragment from either the wild-type or mutant cDNA
clone. The primer pair consisting of SEQ ID NO:51 and SEQ ID N0:49 was used to generate a double stranded 1.587 kbp DNA fragment from either the wild-type or mutant cDNA
clone. In each case the sense strand primer contained a biotin label at the 5' end.
The PCR reactions were carried out as follows. One to two ng of plasmid DNA
from the wild-type or 422 mutant was used as the target DNA in a 100 p.l reaction containing 50 ~,M of each dNTP, 1 p,M of each primer in a given primer pair, in 1 X PCR
buffer. Tubes containing the above mixture were overlaid with three drops of light mineral oil and the tubes were heated to 94°C for 1 min, then cooled to 70°C.. Taq polymerase was then added as 2.~
units of enzyme in 5 p,l of 1X PCR buffer. The tube was heated to 93°C
for 45 sec, cooled to 52°C for 2 min, heated to 72°C for 1 min 45 sec for 35 repetitions, with a 5 min incubation at 72°C after the last repetition.
Following the PCR, excess primers were removed using a QIA Quick-Spin PCR
Purification kit (Qiagen, Inc. Chatsworth, CA) following the manufacturer's instructions; the WO 96/15267 PG"TlUS95/14673 DNA was eluted in 50 p.l of TE. The sense strand of each of the double-stranded fragments from the wild-type and 422 mutant gene were isolated as follows. Streptavidin-coated paramagnetic beads (Dynal M280 beads) [0.5 mg in 50 p.l; pre-washed in 2X bind and wash (B&W) buffer (2 M NaCI, 10 mM Tris-HCI, pH 7.5, 1 mM EDTA, 0.1 % Tween 20)) were added to each purified PCR product. The samples were incubated at room temperature for 15 , minutes with occasional shaking. The beads were removed from the supernatant by exposing the tube to a magnetic plate and the supernatant was discarded. The bead-DNA
complexes were washed twice in 2X B&W buffer. One hundred microliters of 0.1 M NaOH were added to the beads and the samples were incubated at room temperature for 15 minutes (for the 157, 378 by DNAs); for DNA fragments larger than 1 kb, the beads were incubated at 47°C for 30 minutes. After incubation, the beads were washed twice with 2X B&W buffer.
Finally, the bead-ssDNA complexes were resuspended in 50 p.l 2X B&W buffer and stored at 4°C.
B) Cleavage Reaction Conditions The cleavage reactions were performed directly on the single-stranded DNA-bead complexes. 5 to 10 ~,l of DNA-bead complex (about 100 fmoles of DNA) were placed in a 200 pl microcentrifuge tube and washed once with 10 p.l of sterile HBO. 7.5 microliters of 10-mM -MOPS, pH 8.2, with 1.3 mM MnCI, (to yield a final concentration of 1 mM) was then added to each tube. The reaction tubes were prewarmed to 65°C for 2 minutes and cleavage was initiated by the addition of 2.5 ~.1 of the enzyme Cleavase-'~"'' BN ( 10-50 ng in 1 X dilution buffer). The reaction was carried out at 65°C for 5 min.
Immediately after this 5 min incubation, the beads were allowed to settle to the bottom of the tube and the supernatant was removed and discarded. Ten to forty microliters of stop buffer was then added to the beads and the sample was incubated at 90°C
for 5-10 minutes.
The formamide/EDTA solution releases the biotinylated DNA from the beads. The beads were allowed to settle to the bottom of the tube. The supernatant containing the cleavage products was collected. Two to eight microliters of the supernatant solution loaded onto 6%
polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP and CDPStar as described in Example 8a. The resulting autoradiograph is shown in Figure 30.
In Figure 30, lanes marked "M" contain molecular weight- markers prepared as ' described in Example 8. Lanes l, 3, ~ and 7 contain cleavage products using the 157, 378.
1056 or 1587 nucleotide sense strand fragment from the wild-type tyrosinase gene.

WO 96115267 ~ PCT/US95/14673 respectively. Lanes 2. 4, 6 and 8 contain cleavage products using the 157, 378, 1056 or 1587 nucleotide sense strand fragment from the 422 mutant tyrosinase gene, respectively.
As shown in Figure 30, the clear pattern of cleavages seen between the wild type and ' 422 mutant was not obscured when the single base change was located in longer DNA
fragments. Thus, thecleavage reaction of the invention can be used to scan large fragments of DNA for mutations. Fragments greater than about 500 by in length cannot be scanned using existing methodologies such as SSCP or DGGE analysis.

The Cleavase~'~'' Reaction Is Insensitive To Large Changes In Reaction Conditions The results shown above demonstrated that the CleavaseTM BN enzyme can be used to probe DNA templates in a structure-specific but sequence independent manner.
These results demonstrated that the Cleavase~'~'' BN enzyme could be used as an efficient way to recognize conformational changes in nucleic acids caused by sequence variations. This suggested that the 5' nuclease activity of the CleavaseT"' BN enzyme could be used to develop a method to scan nucleic acid templates for sequence alterations relative to a wild-type template. The experiments below showed that this was the case. Furthermore it is demonstrated below that the method of the invention is relatively insensitive to large changes in conditions thereby making the method suitable for practice in clinical laboratories.
First, the effect of varying the concentration of MnCh on the cleavage reaction was determined. Second, the effect of different amounts of salt (KCl) on the cleavage pattern was examined. Third, a time course was performed to investigate when complete cleavage was obtained. Fourth, a temperature titration was performed to determine the effect of temperature variations on the cleavage pattern. Next, the enzyme was titrated to determine the effect of a 50-fold variation in enzyme concentration on the cleavage reaction. The results of these experiments showed that the Cleavase'~"' reaction is remarkably robust to large changes in conditions.
. 30 A) MnClz Titration To determine the sensitivity of the cleavage reaction 1:o fluctuations in the concentration of MnCI,. a single template was incubated in the presence of a fixed amount of the CleavaseTM BN enzyme (250 ng) in a buffer containing 10 mM MOPS, pH 8.2, and various amount of MnCI,. The cleavage reaction was performed as follows. One hundred WO 96/15267 PG"T/US95I14673 fmoles of the 157 nucleotide sense strand- fragment of the tyrosinase gene (SEQ ID N0:42;
prepared by asymmetric PCR .as described in Example 9) was placed i11 a 200 ul thin wall microcentrifuge tube (BioRad) in 5 ~.l of 10 mM MOPS, pH 8.2, with 0, 2, 4. 8, 12 or 20 mM MnCh (to yield a final concentration of either 0, 1, 2, 4, 6, 8 or 10 -mM
MnCh). A tube containing 100 fmoles -template DNA in 5 q.l of 10 mM MOPS, pH 8.2 with 10 MnCI, was y prepared and served as the no enzyme (or uncut) control. Each reaction mixture was overlaid with a drop of light mineral oil. The tubes were heated to 95°C for 5 sec and then cooled to GS°C. .. _ Cleavage reactions were initiated and terminated as descibed in Example 8b.
After the addition of stop solution, the samples were heated to 72°C for 2 minutes and 8 ~.1 of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link).
with 7 M urea, in a buffer containing O.SX TBE. After electrophoresis, the DNA
was transferred to a nylon membrane and processed with SAAP and CDPStar as described in Example 8a. The resulting autoradiograph is shown in Figure 31.
In Figure 31, lanes marked "M" contain molecular.weight markers. Lane I
contains the no enzyme control and shows the migration of the uncleaved template DNA.
Lanes 2 through 8 contain reaction products-incubated in the presence of the enzyme CleavaseT"'~ BN
in a buffer containing 10, 8, 6, 4, 2, 1, or 0 mM MnCI,, respectively.
Figure 31 shows that no cleavage occurs in the absence of divalent canons (lane 8, 0 mM MnCI,). Efficient production of cleavage fragments was promoted by the inclusion of MnCI,. The most distinct pattern of cleavage seen at 1 mM MnCh (lane 7), but little change in the pattern was seen when the Mn'-T concentration varied from 1 to 4 mM;
High concentrations of MnCh tend to suppress the cleavage reaction (concentrations above 6 mMj.
These results show that the cleavage reaction requires a divalent cation but that changes in the amount of divalent cation present have little effect upon the cleavage pattern.
B) Effect Of Salt Concentration On The Cleavage Reaction To determine the effect of salt concentration upon the cleavage reaction, a single template was incubated in the presence of a fixed amount of the CleavaseTM BN
enzyme (250 ng) -in a buffer containing 10 mM MOPS, pH 8.2, 1mM MnCI, and various amounts of IhCI.
.One hundred fmoles of the 157 base fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34; prepared as described in Example 8a) was placed in a 200 yl thin wall microcentrifuge tube (BioRad) in a buffer containing 10 mM
MOPS, pH 8.?

and I mM MnCh. KCl was added to give a final concentration of either 0, 10, 20, 30, 40, or 50 mM KCI; the final reaction volume was 10 pl.
A tube containing 10 mM MOPS, pH 8.2, I mM MnCh, 33 fmoles template DNA and 50 mM KCl was prepared and served as the no enzyme (or uncut) control. Each reaction S mixture was overlaid with a drop of light mineral oil. The tubes were heated to 95°C for 5 seconds and then cooled to 65°C.
Cleavage reactions were initiated and terminated as descibed in Example 8b.
After the addition of stop solution, the samples were heated to 72°C for 2 minutes and 8 ~.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE. After electrophoresis, the DNA
was transferred to a nylon membrane and processed with SAAP and Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega Corp., Madison WI) and exposed to X-ray film as described in Example 8b. The resulting autoradiograph is shown in Figure 32.
I S In Figure 32 , lanes marked "M" contain molecular weight markers. Lane I
contains the no enzyme control and shows the migration of the uncleaved template DNA.
Lanes 2 through 7 contain reaction products incubated in the presence of the CleavaseT"' BN enzyme in a buffer containing 50, 40, 30, 20, 10 or 0 mM KCI, respectively.
The results shown in Figure 32 show that the CleavaseT"' reaction is relatively insensitive to variations in salt concentration. The same cleavage pattern was obtained when the 157 nucleotide tyrosinase DNA template (SEQ ID N0:34) was incubated with the CleavaseT"' enzyme regardless of whether the KCI concentration varied from 0 to 50 mM.
C) Time Course Of The Cleavage Reaction To determine how quickly the cleavage reaction is completed, a single template was incubated in the presence of a fixed amount of the CleavaseT"'' BN enzyme for various lengths of time. A master mix comprising 20 ~.l of a solution containing 10 mM MOPS, pH 8.2, ~
mM MnCI,, and 400 fmoles of the 157 base fragment derived from the sense strand of exon 4 of the tyrosinase gene [(SEQ ID N0:34); prepared as described in Example 8b]
was made.
Five microliter aliquots were placed in 200 p.l thin wall microcentrifuge tube (BioRad) for each time -point examined. A no enzyme control tube was run; this reaction contained 33 fmoles of the template DNA in 10 mM MOPS, pH 8.2 with 1 mM MnCI, (in a final reaction volume of 10 pl). The solutions were overlaid with one drop of light mineral oil. The tubes were brought to 95°C for 5 seconds to denature the templates and then the tubes were cooled to 65°C.
Cleavage reactions were started by the addition of a diluted enzyme mixture as described in Example 8b. At the indicated time points, the reactions were stopped immediately by the addition of 8 p.l of stop solution. The no enzyme control was incubated at 65°C, for 10 minutes and treated in the same manner as the other reactions by the addition of 8 ~.1 of stop buffer. Samples were heated to 72°C for 2 minutes and 5 ~.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE. After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP, then reacted with and Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega Corp., Madison WI) and exposed to X-ray film as described in Example 8b. The resulting autoradiograph is shown in Figure 33.
In Figure 33, lanes marked "M" contain molecular weight markers prepared as described in Example 8. Lane 1 contains the no enzyme control incubated for 10 minutes.
Lanes 2-5 contain the cleavage products from reactions incubated for 0.1, 1, 5 or 10 minutes at 65°C. Figure 36 shows that the cleavage reaction mediated by the Cleavase~'~'' BN enzyme is very rapid. Cleavage is already apparent at less than 6 seconds (<0.1 min) and is complete within one minute. These results also show that the same pattern of cleavage is produced whether the reaction is run for I or 10 minutes.
D) Temperature Titration Of The Cleavase Reaction To determine the effect of temperature variation on the cleavage pattern, the 157 base fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ
ID N0:34) was incubated in the presence of a fixed amount of the CleavaseTM BN enzyme for 5 minutes at various temperatures. One hundred fmoles of substrate DNA (prepared as described in Example 8b) was placed in a 200 pl thin wall microcentrifuge tube (BioRad) in 5 ~.l of 10 mM _MOPS, pH 8.2 with 2 mM MnCI,. .Two "no enzyme" test control tubes were set-up as above with the exception that these reactions contained 33 fmoles of substrate DNA in 10 yl of the above buffer with 1 mM MnCh. The solution was overlaid with one drop of light mineral oil. Tubes were brought to 95°C for 5 seconds to denature the templates and then the tubes were cooled to the desired temperature.
Cleavage reactions were started immediately by the addition of a diluted enzyme mixture as described in Example 8b. The tubes placed at either 55°, 60°, 65°, 70°, 7~° or 80°C. After 5 minutes at a given temperature, the reactions were stopped by the addition of 8 yl of stop buffer.
Samples were heated to 72°C for 2 minutes and 5 ql of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M
urea, in a s 5 buffer containing O.SX TBE. After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP, then reacted with and Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega Corp., Madison WI) and exposed to X-ray film as described in Example 8b. The resulting autoradiograph is shown in Figure 34.
In Figure 34, the lanes marker "M" contain molecular weight markers prepared as described in Example 8. Lanes 1 and 2 contain no enzyme controls incubated at 55°C and 80°C, respectively. Lanes 3-8 contain the cleavage products from the CleavaseT"1 enzyme-containing reactions incubated at 55°C, 60°C, 65°C, 70°C, 75°C or 80°C, respectively.
Figure 34 shows that the CleavaseTM reaction can be performed over a wide range of temperatures. The pattern of cleavages changed progressively in response to the temperature of incubation, in the range of 55°C to 75°C. Some bands were evident only upon incubation at higher temperatures. Presumably some structures responsible for cleavage at the intermediate temperatures were not favored at the lower temperatures. As expected, cleavages became progressively less abundant in the high end of the temperature range tested as structures were melted out. At 80°C cleavage was inhibited completely presumably due to complete denaturation of the template.
These results show that the cleavage reaction can be performed over a wide range of temperatures. The ability to run the cleavage reaction at elevated temperatures is important. If a strong (i.e., stable) secondary structure is assumed by the templates, a single nucleotide change is unlikely to significantly alter that structure, or the cleavage pattern it produces.
Elevated temperatures can be used to bring structures to the brink of instability, so that the effects of small changes in sequence are maximized, and revealed as alterations in the cleavage pattern.within the target template, thus allowing the cleavage reaction to occur at that point.
E) Titration Of The CleavaseT"' BN Enzvme The effect of varying the concentration of the CleavaseTM BN enzyme in the cleavage reaction was examined. One hundred fmoles of the 157 base fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34) was placed in 4 microcentrifuge _ 12$ _ tubes in 5 q.l of 10 mM MOPS, pH 8.2 with 2 mM MnCh. A no enzyme control tube was run; this reaction contained 33 fmoles of substrate DNA in 10 q.l of 10 mM
MOPS, pH 8.2 containing 1 mM MnCI,. The solutions were overlaid with one drop of light mineral oil.
The tubes were brought to 95°C for 5 seconds to denature the templates and then the tubes were cooled to 65°C.
Cleavage reactions were started immediately by the addition of a diluted enzyme mixture comprising 1 ~.l of the CleavaseTT'' BN enzyme in 1X dilution buffer such that 10, 50, 100 or 250 ng of enzyme was in the tubes in 5 p,l of 10 mM MOPS, pH 8.2 without MnCh.
After 5 minutes at 65°C, the reactions were stopped by the addition of 8 ql of stop buffer.
The samples were heated to 72°C for 2 minutes and 7 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M
urea, in a buffer containing O.SX TBE.
After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP, then reacted with and Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega Corp., Madison WI) and exposed to X-ray film as described in Example 8b. The resulting autoradiograph is shown in Figure 3~.
The lanes marked "M" in Figure 35 contain molecular weight markers. Lane 1 contains the no enzyme control and shows the migration of the uncut substrate.
Lanes 2-~
contain reaction products from reactions containing 10, 50, 100 or 250 ng of the CleavaseT"'' BN enzyme , respectively.
These results show that the same cleavage pattern was obtained using the 157 nucleotide tyrosinase DNA substrate regardless of whether the amount of enzyme used in the reaction varied over a 25-fold range. Thus, the method is ideally suited for practice in clinical laboratories where reactions conditions are not as controlled as in research laboratories.
F) Consistent Cleavage Patterns Are Obtained Using Different DNA
Preparations To demonstrate that the same cleavage pattern is consistently obtain from a given substrate, several different preparations of the 157 base fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34) were generated. The substrate. was generated as described in Example 8b. Three independent PCR reactions performed on ' separate days were conducted. One of these PCR samples was split into two and one aliquot was gel-purified on the day of generation while the other aliquot was stored at 4°C overnight and then gel-purified the next dav.
Cleavage reactions were performed as described in Example 8b. Samples were run on ' an acrylamide gel and processed as described in Example 8b. The resulting autoradiograph is shown in Figure 36.
In Figure 36, the lanes marked "M" contain biotinylated molecular weight markers.
Set 1 contains the products from a cleavage reaction performed in the absence (-) or presence (+) of enzyme on preparation no. 1. Set 2 contains the products from a cleavage reaction performed in the absence (-) or presence (+) of enzyme on I>reparation no. 2.
Set 3 contains the products from a cleavage reaction performed in the absence (-) or presence (+) of enzyme on preparation no. 3. Preparation no. 3 was derived from preparation 2 and is identical except that preparation no. 3 was gel-purified one day after preparation no. 2. Set 4 contains the products from a cleavage reaction performed in the absence (-) or presence (+) of enzyme on preparation no. 4. The same pattern of cleavage products is generated from these independently prepared substrate samples.
These results show that independently produced prep~~rations of the 157 nucleotide DNA fragment gave identical cleavage patterns. Thus, the CleavaseTM reaction is not effected by minor differences present between substrate preparations.

Point Mutations Are Detected Using Either The Sense Or Anti-Sense Strand Of The Tyrosinase Gene The ability of the Cleavase'~"'~ enzyme to create a unique pattern of cleavage products (i.e., a fingerprint) using either the sense (coding) or anti-sense (non-coding) strand of a gene fragment was examined.
Single stranded DNA substrates corresponding to either the sense (SEQ ID
N0:34) or anti-sense strand (SEQ ID N0:35) of the 157 nucleptide fragment derived from the wild-type tyrosinase gene were prepared using asymmetric PCR as described in Example 8a.
The sense strand wild-type substrate contains a biotin label at the 5' end; the anti-sense strand contains a fluorescein label at the 5' end.
A single stranded DNA substrate corresponding to thf: sense strand of the 157 nucleotide fragment derived from the 419 mutant tyrosinase gene (SEQ ID N0:41 ) was _ 127 _ prepared using asymmetric PCR as described in Example 9. The, sense strand 419 mutant substrate contains a biotin label at the 5' end.
A single stranded DNA substrate corresponding to the anti-sense strand of the nucleotide fragment derived from the 419 mutant tyrosinase gene (SEQ ID N0:52) was prepared using asymmetric PCR as described in Example 9, with the exception that 100 pmoles of the fluorescein-labeled anti-sense primer (SEQ ID N0:30) and 1 pmole of the biotin-labelled sense primer (SEQ ID N0:29) were used. The resulting anti-sense strand 419 mutant substrate contains a fluorescein label at the 5' end.
A single stranded DNA substrate corresponding to the sense strand of the 157 nucleotide fragment derived from the 422 mutant tyrosinase gene (SEQ ID N0:42) was prepared using asymmetric PCR as described in Example 9. The sense strand 422 mutant substrate contains a biotin label at the 5' end.
A single stranded DNA substrate corresponding to the anti-sense strand of the nucleotide fragment derived from the 422 mutant tyrosinase gene (SEQ ID N0:53) was prepared using asymmetric PCR as described in Example 9 with the exception that 100 pmoles of the fluorescein-labeled anti-sense primer (SEQ ID N0:30) and 1 pmole of the biotin-labelled sense primer (SEQ ID N0:29) were used. The resulting anti-sense strand 422 mutant substrate contains a fluorescein label at the 5' end.
Following asymmetric PCR, the single stranded PCR products were gel purified, precipitated and resuspended in 40 p,l of TE buffer as described in Example 8.
Cleavage reactions were performed as described in Example 9, and were resolved by electrophoresis as described in Example 8a. After electrophoresis, the DNA was transferred to a nylon membrane and processed with SAAP conjugate and antifluorescein antibody (Fab)-allcaline phosphatase conjugate, and visualized using CDPStar as described in Example 8a.
The resulting autoradiograph is shown in Figure 37.
In Figure 37, lanes marked "M" contain biotinylated molecular weight markers prepared as described in Example 8. Lanes 1-6 contain biotinylated sense strand substrates from the wild-type, 419 and 422 mutant 157 nucleotide fragments. Lanes 1-3 contain no enzyme controls for the wild-type, 419 and 422 mutant fragments, respectively.
Lanes 4-6 contain the reactionproducts from the incubation of the sense strand of the wild-type, 419 and 422 mutant fragments with theCleavase~'T'' BN enzyme , respectively. Lanes 7-12 contain ' fluoresceinated anti-sense strand substrates from the wild-type, 419 and 422 mutant 157 z nucleotide fragments. Lanes 1-3 contain "no enzyme" controls for the wild-type, 419 and 422 mutant fragments, respectively. Lanes 4-6 contain the reaction products from the incubation of the anti-sense strand of the wild-type, 419 and 422 mutant fragments with the CleavaseTh' BN , respectively. ' As expected, distinct but unique patterns of cleavage products are generated for the wild-type, 419 and 422 mutant fragments when either the sense or anti-sense fragment is V
utilized. The ability to use either the sense or anti-sense strand of a gene as the substrate is advantageous because under a given set of reaction conditions one of the two strands may produce a more desirable banding pattern (i. e., the cleavage products are spread out over the length of the gel rather than clustering at either end), or may have a mutation more favorably placed to create a significant structural shift. This could be more important in the analysis of long DNA substrates which contain mutations closer to one end or the other.
Additionally, detection on both strands serves as a confirmation of a sequence change.

Detection Of Mutations In The Human Beta-Globin Gene Using The Enzyme CleavaseT"' The results shown in Examples 8-12 showed that the CleavaseTM reaction could be used to detect single base changes in fragments of the tyrosi:nase gene ranging from 157 nucleotides to 1.6 kb. To demonstrate that the CleavaseT"' reaction is generally applicable for the detection of mutations, a second model system was examined.
The human (3-globin gene is known to be mutated in a number of hemoglobinopathies such as sickle cell anemia and [3-thalassemia. These disorders generally involve small ( 1 to 4) nucleotide changes in the DNA sequence of the wild type (3-globin gene [Orkin, S.H. and Kazazian, H.H., Jr. (1984) Annu. Rev. Genet. 18:131 and Collins, F.S. and Weissman, S.M.
( 1984) Prog. Nucleic Acid Res. Mol. Biol. 31:315]. At least 47 different mutations in the [3-globin gene have been identified which give rise to a [i-thala.ssemia:
Three [3-globin mutants were compared to the wild type (3-globin gene [Lawn, R.M., et al. ( 198-0) Cell 21:647] using the CleavaseT"' reaction. Mutant 1 contains a nonsense mutation in codon 39; the wild-type sequence at codon 39 is CAG; the mutant 1 sequence at this codon is TAG [Orkin, S.H. and Goff, S.C. (1981) J. Biol. Chem. 256:9782]. Mutant 2 contains a T
to A substitution in codon 24 which results in improper splicing of the primary transcript [Goldsmith, M.E., et al. (1983) Proc. Natl. Acad. Sci. USA 80:2318]. Mutant 3 contains a deletion of two A residues in codon 8 which results in a shift in the reading frame; mutant 3 also contains a silent C to T substitution in codon 9 [Orkin. S.H. and Goff.
S.C. ( 1981 ) ,1.
Biol. Chem. 26:9782].
A) Preparation Of Wiid Type And Mutant -Globin Gene Substrates Single stranded substrate DNA was prepared from the above wild type and mutant [3-globin genes as follows. Bacteria harboring the appropriate plasmids were streaked onto antibiotic plates and grown overnight at 37°C (bacteria with the wild-type plasmid and the plasmid containing the mutant 3, were grown on tetracycline plates: bacteria with the plasmids containin= the mutant 1 and mutant 2 sequences were grown on ampicillin plates). Colonies from the plates were then used to isolate plasmid DNAs using the Wizard Minipreps DNA
Purification System (Promega Corp., Madison, WI). The colonies were resuspended in ?00 pl of "Cell Resuspension Buffer" from the kit. The DNA was extracted according to the manufacturers protocol. Final yields of approximately ?.3 pg of each plasmid were obtained.
A 5 36 (wild-type. mutants 1 and ?) or X34 (mutant 3) nucleotide fragment was amplified from each of the above plasmids in polvmerase chain reactions comprisin~T ~ n~~ ~i~
1 ~ plasmid DNA. ?3 pmoles each of 3~-biotinylated KM?9 primer (SEQ ID NO:>.~) and ~'-fiuorescein labeled RS4? primer (SEQ ID NO:>j). 50 ~M each dNTP and 1.'_'~
units of lcrc~
DNA Polvmerase in sU ul of 1X PCR buffer. The reactions were overlaid with '_' drops of~
light mineral oiI and were heated to 93°C for 30 seconds. cooled to »°C for s0 seconds.
heated to 7'?°C for 60 seconds, for 3~ repetitions in a thermocvcler (MJ Research. Vvatert~wn.
?0 MA). The products of these reactions were purified from the residual dNTPs and primers by use of a Vl~'izard'~P~R Cleanup kit (Promega Corp.. Madison. WI). leaving the duplex D'~.a in 50 l.tl of TE.
To generate single stranded copies of these DNAs. the PCRs described above were repeated using= 1 pl of the duplex PCR DNA as template. and omitting the RS4'_' primer. The products of this asymmetric PCR were loaded directly on a 6% polyacrylamide gel (?9:1 cross-link) in a buffer of O.~X TBE. alongside an aliquot of the original PCR
DNA to identify the location of the double-strand DNA product. After electrophoretic separation. the DN.as were visualized by staining with ethidium bromide and the single stranded DNAs. identified by altered mobility when compared to the duplex DNAs. were excised and eluted from the «el 30 slices by passive diffusion overnight into a solution comprising 0.5 M
NH,OAc. U.1 °r SDS ' and 0.? mM EDTA. The products were collected by ethanol precipitation and dissolved in -IO
pl of TE.
* Trade-mark - 1 s0 -WO 96/15267 PG"T/US951i4673 The sequence of the 536 nucleotide fragment from the wild-type (3-globin gene is listed in SEQ ID N0:56. The sequence of the 534 nucleotide fragment from mutant 3 is listed in SEQ ID N0:57. The sequence of the 536 nucleotide fragment from mutant 1 is listed in SEQ ID N0:58. The sequence of the 536 nucleotide fragment from mutant 2 is listed in SEQ ID N0:59.
B) Optimization Of The Cleavage Reaction Using The Wild-Type Beta-Globin Substrate The optimal conditions (salt concentration, temperature) which produce an array of cleavage products having widely differing mobilities from the (3-globin substrate were determined. Conditions which produce a cleavage pattern having the broadest spread array with the most uniform intensity between the bands were determined (the production of such an array of bands aids in the detection of differences seen between a wild-type and mutant substrate). This experiment involved running the cleavage reaction on the wild type ~3-globin substrate (SEQ ID N0:56) at several different temperatures in the presence of either no KC1 or 50 mM KCI.
For each KCL concentration to be tested, 30 q.l of a master mix containing DNA, CFLPT"1 buffer and salts was prepared. For the "0 mM KCl" reactions, the mix included approximately 500 fmoles of single-stranded, 5' biotinylated 536-mer PCR DNA
from plasmid pHBG 1 in 30 p.l of 10 mM MOPS, pH 8.2, with 1.7 mM MnC 1 ~ (for 1 mM
in the final reaction); the "50 mM KCl" mix included 83.3 mM KC 1 in addition to the above components. The mixes were distributed into labeled reaction tubes in 6 q,l aliquots, and stored on ice until use. An enzyme dilution cocktail included 450 ng of the enzyme CleavaseT"' BN in 10 mM MOPS, pH 8.2 without MnCl.,.
Cleavage reactions were performed at 60°C, 65°C, 70°C
and 75°C. For each temperature to be tested, a pair of tubes with and without KC 1 were brought to 95 ° C for 5 seconds, then cooled to the selected temperature. The reactions were then started immediately by the addition of 4 p.l of the enzyme cocktail. In the 75°C test, a duplicate pair of tubes was included, and these tubes received 4 ~.1 of 10 mM MOPS, pH 8.2 without MnCl~
in place of the enzyme addition. No oil overlay was used. All reactions proceeded for 5 minutes. and were stopped by the addition of 8 q.l of stop buffer. Completed and yet-to-be-started reactions were stored on ice until all reactions had been performed. Samples were heated to 72°C for 2 minutes and 5 ~1 of each reaction was resolved by electrophoresis through a 6%
polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer of O.SX TBE.
After electrophoresis. the gel plates were separated allowing the gel to remain flat on one plate. A
0.? mm-pore positively-charged nylon membrane (NYTRAIv':~Schleicher and Schuell. Keene.
NH). pre-wetted in H,O. was laid on top of the exposed gel. All air bubbles were removed.
Two pieces of 3MM filter paper (Whatman) were then placed on top of the membrane. the other glass plate was replaced. and the sandwich was clamped with binder clips. Transfer was allowed to proceed overnight. After transfer. the membrane was carefully peeled from the gel and allowed to air dry. After complete drying the membrane was washed in l.'?X
Sequenase Images Blocking Buffer (United States Biochemical) using 0.3 ml of buffer/cmv of membrane.
The wash was performed for 30 minutes. A streptavidin-alkaline phosphatase conjugate (SAAP. United States Biochemical) was added to a 1:4000 dilution directly to the blockin«
solution. and agitated for 1 ~ minutes. The membrane was rinsed briefly with H,O and then washed three times for ~ minutes per wash using 0.~ ml/cm= of IX SA.4P buffer ( 100 m~~1 Tris-HCI. pH 10. ~0 mM NaCI) with O.l% sodium dodecvl sulfate (SDS). The membrane was rinsed brief7v with H,0 bemveen each wash. The membrane was then washed once in 1 X
1~ SA.~P''1 mi\~t MgCI= without SDS. drained thoroughl~ and placed in a plastic heat-sealable ba~_. Using a sterile pipet. ~ mls of either CSPDT''' or CDP-StarT"' (Tropix.
Bedford. '\-La i chemiluminescent substrates for alkaline phosphatase were added to the bag and distributed over the entire membrane for ?-3 minutes. The CSPDT"'-treated membranes were incubated at 37°C for 30 minutes before an initial exposure to XRP X-ray filth (Kodak) for 60 minutes.
?0 CDP-StarT"'-treated membranes did not require preincubation. and initial exposures were fur 10 minutes. Exposure times were adjusted as necessary for resolution and clarity. The results are shown in Figure 38.
In Fib=ure 38 he lane marked "M" contains molecular weight markers. Lanes 1-contain reaction products from reactions run in the absence of KCI. Lane 1 contains the a reaction run without enzyme at 7~°C. Lanes 2-~ contain reaction products from reactions run at 60°C. 6~'C. 70°C and 75°C. respectively. Lanes 6-10 contain reaction products from reactions run in the presence of ~0 mM KC1. Lane 6 contains the a reaction run without enzyme at 7~°C. Lanes 7-10 contain reaction products from reactions run at 60°C. 6~'C.
70°C and 7>°C. respectively.
O) In general. a preferred pattern of cleavage products was produced when the reaction included ~0 mM KCI. As seen in Lanes 7-10, the reaction products are more widely spaced in the ~0 m:~i KCL-containing reactions at every temperature tested as compared to the reactions run in the absence of KCL (lanes ?-~: more of the cleava~~e products are found *Trade-mark - I;~ _ clustered at the top of the gel near the uncut substrate). As seen in Lane 7 of Figure 41, cleavage reactions performed in 50 mM KCl at 60.°C produced a pattern of cleavage products in which the products are maximally spread out, particularly in the upper portion of the gel, when compared to other reaction condition patterns.
From the results obtained in this experiment, the optimal cleavage conditions for the 536 nucleotide sense strand fragment derived from the wild-type (3-globin gene (SEQ ID
N0:56) were determined to be 10 mM MOPS, pH 8.2 containing 1 mM MnCh and 50 mM
KCl at 60°C.
C) Optimization Of The Cleavage Reaction Using Two Mutant Beta-Globin Substrates From the results obtained above in a) and b), 60°C was chosen as the optimum temperature for the cleavage reaction when a (3-globin substrate was to be used. When the wild-type substrate was utilized, running the cleavage reaction in the presence of 50 mM KCl generate the optimal pattern of cleavage products. The effect of varying the concentration of KCl upon the cleavage pattern generated when both wild-type and mutant (3-globin substrates were utilized was next examined to determine the optimal salt concentration to allow a comparison between the wild-type and mutant (3-globin substrates.
Single stranded substrates, 536 nucleotides in length, corresponding to mutant I (SEQ
ID N0:58) and mutant 2 (SEQ ID N0:59) mutations were prepared as described above in a).
These two mutants each differ from the wild-type sequence by 1 nucleotide;
they differ from each other by 2 nucleotides.
For each substrate tested, 39 p.l of a master mix containing DNA, CFLPT°~ buffer and MnCI, was prepared. These mixes each included approximately 500_ fmoles of single-stranded, 5' biotinylated 536 nucleotide substrate DNA, 39 p.l of 10 mM MOPS.
pH 8.2 2~ containing 1.54 mM MnCI? (giving a final concentration of 1 mM MnCI,). The mixes were distributed into labeled reaction tubes in 6.5 p.l aliquots. Each aliquot then received 0.5 ~.1 of 200 mM KCl for each 10 mIvl final KCl concentration (e.g., 2.0 p.l added to the 40 mM
reaction tube) and all volumes were brought to 9 ~.l,with dH~O. No oil overlay was used.
The reactions were brought to 95°C for 5 seconds, then cooled to 65°C. The reactions were then started immediately by the addition of 50 ng of the enzyme CleavaseTM BN
in 1 pl of enzyme dilution buffer (20 mM Tris-HCI, pH 8.0, 50 mM KCI, 0.5% NP40, 0.5%
Tween 20, 10 mg/~1 BSA). All reactions proceeded for 5 minutes, and were stopped by the addition of 8 ~l of stop buffer. Samples were heated to 72°C for 2 minutes and 5 ~1 of each reaction was resolved by electrophoresis through a 6% polyacrylamide gel ( I 9:1 cross-link), with 7 M
urea, in a buffer of O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described above. The DNA was transferred to the membrane and the membrane was treated as described above in b) and then exposed to X-ray-film. The resulting autoradiograph is shown in Figure 39 In Figure 39, the lane marked "M" contains molecular weight markers. Lanes 1, 3, 5, 7, 9 and 11 contain reaction products from cleavage reactions using the mutant 1 substrate in the presence of 0,- 10, 20, 30, 40 or 50 mM KCI, respectively. Lanes 2, 4, 6, 8, 10 and 12 contain reaction products from cleavage reactions using the mutant 2 substrate in the presence of 0, 10, 20, 30, 40 or 50 mM KCI, respectively.
Figure 39 shows that while the pattern of cleavage products- generated from each mutant changes as the KCl concentration is increased, distinct patterns are generated from each mutant and differences in banding patterns are seen between the two mutants at every 1~ concentration of KCl tested. In the mid-salt ranges (10 to 20 mM KCl), the larger cleavage bands disappear and smaller molecular weight bands appear (lanes 3-6). At higher salt concentrations (30 to 50 mM KCl), the larger molecular weight cleavage bands reappear with the cominant loss of the low molecular weight bands (lanes 7-12). Reaction conditions comprising the use of 50 mM KCl were chosen as optimal- from the results shown in Figure 39. Clear differences in the intensities of a band appearing about 200 nucleotides (see arrow in Figure 39) is seen between the two mutant substrates under these reaction conditions.
D) The Enzyme CleavaseTnt Generates Unique Cleavage Products From Wild-Type And Mutant Beta-Globin Substrates From the experiments performed above, the optimal reaction conditions when the wild-type or mutant (3-globin substrates were determined to be the use of 50 mM KCl and a temperature of 60°C. These conditions were then used to allow the comparison-of the cleavage patterns generated when the wild-type substrate (SEQ ID N0:56) was compared to the mutant 1 (SEQ ID N0:58), mutant 2 (SEQ ID N0:59) and mutant 3 (SEQ ID
N0:57) substrates.
Single-stranded substrate DNA, 534 or 536 nucleotides in length,- was prepared for the wild-type, mutant 1, mutant 2 and mutant 3 [3-globin genes as described above in a). ' Cleavage reactions were performed as follows. Reaction tubes were assembled which contained approximately 100 fmoles of each DNA substrate in 9 ~.1 of 1.10 mM
MOPS, pH ' WO 96115267 PG"T/US95/14673 8.2 ( 1 X final concentration) with 1.1 mM MnCh ( 1 mM final concentration) and 55.6 mM
KCl (50 mM final concentration). A "no enzyme" or uncut control was set up for each substrate DNA. The uncut controls contained one third as much DNA ( approximately 33 fmoles) as did the enzyme-containing reactions to balance thf; signal seen on the autoradiograph.
The tubes were heated to 95°C for 5 sec, cooled to 60°C and the reactions were started immediately by the addition of 1 ~l of the enzyme CleavaseT"'' BN (50 ng per ~l in 1X
dilution buffer). The uncut controls received 1 ~,l of 1 X dilution buffer.
Reactions proceeded for 5 min and then were stopped by the addition of 8 ~.l of stop buffer. The samples were heated to 72°C for 2 min and 5 ~.l of each reaction was resolved by electrophoresis through a 6% polyacrylamide gel (19:1 cross-link), with 7 M
urea, in a buffer of 0.5X TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described above. The DNA was transferred to the membrane and the membrane was treated as described above in b) and then exposed to X-ray film. The resulting autoradiograph is shown in Figure 40.
In Figure 40, two panels are shown. The first panel shows the reaction products from the above cleavage reactions; the uncut controls are shown in lanes 1-4 and the cleavage products are shown in lanes 5-6. The second panel is a magnification of lanes 5-8 to better shown the different banding patterns seen between the substrate DNAs in the upper portion of the gel.
In Figure 40, the lanes marked "M" contain biotinylal:ed molecular weight markers prepared as described in Example 8. Lanes 1-4 contain the uncut controls for the wild-type, mutant l, mutant 2 and mutant 3 (3-globin substrates, respectively. Lanes 5-8 contain the cleavage products from the wild-type, mutant 1, mutant 2 anal mutant 3 substrates, respectively.
From the results shown in Figure 40, the CleavaseT"'' BN enzyme generates a unidue pattern of cleavage products from each (3-globin substrate tested. Differences in banding patterns are seen between the wild-type and each mutant; dii:ferent banding patterns are seen for each mutant allowing not only a discrimination of the mutant from the wild-type but also a discrimination of each mutant from the others.

WO 96115267 PCTlUS95/14673 The results shown here for the (3-globin gene and above for the tyrosinase gene demonstrate that the Cleavase~ reaction provides a powerful ne~~ tool for the detection of mutated genes.

Treatment Of-RNA Substrates Generates Unique Cleavage Patterns The present invention contemplates 5' nuclease cleavage of single- or double-stranded DNA substrates to generate a unique pattern of bands characteristic of a given substrate. The ability of the 5' nuclease activity of the enzyme CleavaseTT' BN to utilize RNA as the substrate nucleic acid was next demonstrated. This experiment showed that RNA
can be utilized as a substrate for the generation of a cleavage pattern using appropriate conditions (Lowering of the pH to 6.5 from 8.2 to reduce manganese-mediated degradation of the RNA
substrate). The experiment was performed as follows.
An RNA transcript internally labelled with biotin was produced to serve as the substrate. The plasmid pGEM3Zf (Promega) was digested with EcoRI. EcoRI cuts the plasmid once generating a linear template. An RNA transcript 64 nucleotides in length (SEQ
ID NO:GO) was generated by SP6 transcription of the linearized template using a Riboprobe Gemini System kit from Promega, Corp.; the manufacturer's directions were followed with the exception that 25% of the UTP in the reaction was replaced with biotin-UTP
(Boehringer Mannheim) to produce an internally labelled transcript. Following the transcription reaction (1 hour at 37°C), the DNA template was removed by treatment with RQ1 RNase-free DNAse (from the Riboprobe kit and used according to the manufacturer's instructions) and the RNA
was collected and purified by precipitating the sample twice in the presence of 2 M NH40Ac and ethanol. The resulting RNA pellet was rinsed with 70% ethanol, air dried and resuspended in 40 ~.1 of 10 mM Tris-HCI, pH 8.0 and 1 mM EDTA.
Cleavage reactions contained 1 ~,1 of the above RNA substrate and 50 ng of the enzyme Cleavase~'~'"' BN in 10 ~.l of 1X RNA-CFLP~ buffer (10 mM MOPSz pH G.3) and 1 mM of either MgCI, or MnCh. The reactions were assembled with, all the components except, the enzyme and were warmed to 45°C for 30 sec. Reactions were started by the addition of 50 ng of the enzyme CleavaseT"' BN in 1 ~l of dilution buffer (20 mM Tris-HCI, pH
8.0, ~0 mM ' ICCI, 0.5% NP40, 0.5% Tween 20, 10 ~.g/ml BSA). Reactions proceeded for 10 min and were stopped by the addition of 8 ~.1 of stop buffer. The samples were heated to 72°C for ? ' minutes and 5 ~.l of each reaction were resolved by electrophoresis through a 10%
polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.~X TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1.2X Sequenase Images Blacking Buffer, treated with 1X
SAAP buffer and reacted with Lumiphos-530 (United States Biochemical) or Quantum Yield Chemiluminescent Substrate (Promega) and exposed to X-ray film as described in Example 8b. The resulting autoradiograph is shown in Figure 41.
In Figure 41 , lanes marked "M" contain molecular weight markers. Lane 1 contains the no enzyme control and shows the migration of the uncut substrate. Lanes 2 and 3 contain reaction products from the incubation of the RNA substrate in a buffer containing MgCI, in the presence or absence of the Cleavase~'~"' BN enzyme , respectively. Lanes 4~ and 5 contain reaction products from the incubation of the RNA substrate in a buffer containing MnCI, in the presence or absence of the CleavaseT'''' BN enzyme , respectively. A
pattern of cleavage I S products is seen when the enzyme is incubated with the RNA substrate in the presence of MnCI,, (lane 5).
These results show that the Cleavase~'~"'' enzyme can be used to probe RNA
substrates for changes in sequence (i. e., point mutations, deletions, substitutions).
This capability enables the examination of genes which have very large introns (e.g., greater than 10 kb) interrupting the coding sequences. The spliced RNA transcript represents a simpler target for the scanning for mutations. In addition, the structural (i, e., folding) information gained by cleavage of the RNA would be useful in targeting of hybridization or ribozyme probes to unstructured regions of RNAs. Furthermore, because the cleavage reaction occurs so quickly, the Cleavase'~"'' enzyme can be used to study various types of RNA folding and the kinetics with which this folding occurs.

The 5' Nuclease Activity From Both the CleavaseT"' 13N Enzyme Taq Polymerase Generates Unickue Cleavage Patterns Using Double-Stranded DNA Substrates The ability of both the enzyme CleavaseT"' BN and Taq polymerase to generate cleavage patterns on single-stranded DNA templates was examined. The substrates utilized in this experiment were the 378 nucleotide fragment from either the wild-type (SEQ ID N0:43) or 422 mutant (SEQ ID N0:44) tyrosinase gene. These single-stranded substrates were generated as described in Example 10a.
Cleavage reactions were performed as described in Example lOb with the exception that half of the reactions were cut with the enzyme CleavaseT"'' BN as described and a parallel set of reaction was cut with Taq polymerase. The Taq polymerase reactions contained 1.25 ;
units of Taq polymerase in 10 mM MOPS, pH 8.2. The reaction products were resolved by electrophoresis and the autoradiograph was generated as described in Example lOb. The autoradiograph is shown in Figure 42.
In Figure 42, lanes marked "M" contain biotinylated molecular weight markers.
Lanes 1 and 2 contain the wild-type or 422 mutant substrate cleaved with the CleavaseT"'' BN
enzyme, respectively. Lanes 3 and 4 contain the wild-type or 422 mutant substrate cleaved with Taq polymerase, respectively.
Figure 42 shows that both the Cleavase~'~"'' BNenzyme and Tag polymerase generate a characteristic set of cleavage bands for each substrate allowing the differentiation of the wild-type and 422 mutant substrates. The two enzyme produce similar but distinct arrays of bands for each template.
These results show that the 5' nuclease of both the CleavaseT"'' BN enzyme and Taq polymerase are useful for practicing the cleavage reaction of the invention.
Cleavage with Taq polymerase would find application when substrates are generated using the PCR and no intervening purification step is employed other than the removal of excess nucleotides using alkaline phosphatase Multiplex Cleavage Reactions The above Examples show that the cleavage reaction can be used to generate a characteristic set of cleavage products from single-stranded DNA and RNA
substrates. The ability of the cleavage reaction to utilize double-stranded DNA templates was examined. For many applications, it would be ideal to run the cleavage reaction directly upon a double-stranded PCR product without the need to isolate a single-stranded substrate from the initial PCR. Additionally it would be advantageous to be able to analyze multiple substrates in the same reaction tube ("multiplex" reactions).

Cleavage reactions were performed using a double-stranded template which was carried a 5' biotin label on the sense-strand and a 5' fluorescein label on the anti-sense strand. The double-stranded substrate was denatured prior to cleavage. 7.'he double-stranded substrate was cleaved using Taq polymerase. Taq polymerase was used in this experiment because it has a w 5 weaker duplex-dependent 5' to 3' exonuclease activity than does the enzyme CleavaseT"' BN
and thus Taq polymerase does not remove the 5' end label from the re-natured DNA duplexes as efficiently as does the enzyme Cleavase~'~"'' BN; therefore less signal is lost in the reaction.
The substrate utilized was a 157 by fragment derived from either the wild-type (SEQ
ID N0:34), 419 mutant (SEQ ID N0:41) or 422 mutant (SEQ ID N0:42) of the tyrosinase gene. The wild-type fragment was generated as described in Example 8a, the 419 mutant fragment was generated as described in Example 8a and the 422 mutant fragment was generated as described in Example 9 using PCR. The sense strand primer (SEQ ID
N0:29) contains a 5' biotin label and the anti-sense primer (SEQ ID N0:30) contains a 5' fluorescein label resulting in the generation of a double-stranded PCR product label on each strand with a I S different label. The PCR products were gel purified as described in Example 8a.
The cleavage reactions were performed as follows. Reaction tubes were assembled with approximately 100 fmoles of the double-stranded DNA substrates in 5 ~.l of 10 mM
MOPS, pH 8.2, 1 mM MnCI,. The solutions were overlaid with a drop of mineral oil. The tubes were heated to 95°C for 30 sec and 1 unit of Taq polymerase (Promega) was added.
Uncut controls contained 33 fmoles of double-stranded DNA substrates in 5 pl of 10 mM
MOPS, pH 8.2, 1 mM MnCl2. The reactions were cooled to 65°C and incubated at this temperature for 15 minutes. The reactions were stopped by the addition of 8 ~,l of stop buffer. The samples were heated to 72°C for 2 min and 5 yl of reaction were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M
urea in a buffer containing O.SX TBE. The entire set of reactions was loaded in duplicate on the gel such that duplicate nylon membranes containing the full set of reactions were created.
After transfer to a nylon membrane (performed as described in Example 8a), the membrane was cut in half:
one half was probed using a streptavidin-alkaline phosphatase conjugate to visualize the biotinylated sense-strand products (as described in Example 8a). The other half of the membrane was probed with an anti-fluorescein antibody-alkaline phosphatase conjugate to visualize the fluorescein-labelled anti-sense strand products (as described in Example ~). The blots were visualized using the chemiluminescent procedures described in Examples 8a and ~

for biotin-labeled or fluorescein-labeled DNA, respectively. The autoradiographs are shown side-by side in Figure 43.
In Figure 43, the lane labeled "Ml" contains biotinylated molecular weight markers prepared as described in Example 8a. The lane labeled "M2" contains molecular weight ' markers generated by digestion of pUCl9 with MspI, followed by Klenow treatment to fill-in the ends. The blunt ends were then labeled with fluoresceinated dideoxynucleotides (Boehringer Mannheim) using terminal transferase (Promega). Lanes Ml and 1-6 were developed using the protocol for biotinylated DNA. Lanes 7-12 and M2 were developed using the protocol for fluorescein-labeled DNA. Note that in all lanes both strands of the substrate are present; only one strand is visualized in a given development protocol.
In Figure 43, lanes 1-3 and 7-9 contain the "no enzyme" or uncut controls using the wild-type, 419 or 422 mutant substrates, respectively. Lanes 4-6 and 10-12 contain cleavage products from the wild-type, 419 or 422 mutant substrates, respectively.
Unique patterns of cleavage products are seen for each strand of each of the three substrates examined. Thus, a single reaction allowed the -generation of a unique fingerprint from either the sense or anti-sense strand of each of the three tyrosinase fragments tested.
The results shown in Figure 43 demonstrate that a cleavage pattern can be generated from a double-stranded DNA fragment by denaturing the fragment before performing the cleavage reaction. Note that in Figure 43 the cleavage patterns were generated in the course of a single round of heating to denature and cooling to cleave and that much of the substrate remains in an uncut form. This reaction would be amenable to performing multiple cycles of denaturation and cleavage in a thermocycler. Such cycling conditions would increase the signal intensity seen for the cleavage products. Substrates generated by the PCR performed in the standard PCR buffer (50 mM KCI, 10 mM Tris-Cl, pH 8.3, 1.5 mM MgCI,, 0.01 gelatin) can be treated to remove remaining dNTPs (e.g., addition of alkaline phosphatase) and to provide Mn'+. Under these conditions the cleavase reaction can be performed on both strands of one or more products generated in that PCR. Such a protocol reduces sample preparation to a minimum resulting in a savings of both time and expense.
The above example also demonstrates that two distinct substrates can be analyzed in a single reaction thereby allowing the "multiplexing" of the cleavage reaction.

WO 96/15267 PG"T/US95/14673 Optimization Of Manganese Ion Concentration For Cleavage Of Double Stranded DNA Substrates As discussed above, it may be desirable to run the cleavage reaction on double-stranded DNA substrates such restriction fragments or segments generated by balanced or symmetric PCR. The effect of varying the concentration of Mn'-+ in cleavage reactions using double-stranded DNA substrates was investigated. The results shown below demonstrate that the optimal concentration of Mn'-+ is lower when a double-stranded DNA
substrate is employed in the cleavage reaction as compared to single-stranded DNA
substrates.
Two double-stranded (ds) DNA substrates, 157 by in length, derived from the tyrosinase mutants 419 (SEQ ID N0:27) and 422 (SEQ ID 1'J0:71 ) were prepared by PCR
amplification of the exon 4 region of human tyrosinase gene as described above in Example 16. The sense strand of the 419 and 422 tyrosinase mutant substrates contained a biotin-1 ~ labeled at the 5' end following the PCR. The PCR products were gel purified as described in Example 8a.
The cleavage reactions were performed as follows. Reaction tubes were assembled with approximately 40 fmoles of the ds DNA substrates in 10 ~,l of water. The tubes were brought to 95°C for 10 seconds in a PTC-100TT' Programmable Thermal Controller (M,T
Research, Inc.) to denature the DNA. The cleavage reactions were started by the addition of 10 ~l of 2X CFLPT"' buffer (pH 8.2) containing 1 p.l of the enzyme Cleavase~'~"' BN (25 ng in 1 X dilution buffer) and different concentrations of MnCh such that the final concentration of MnCh in reaction mixture (20 ~,1 final volume) was either 0,.5 mM, 0.2~ mM, 0.15 mM, 0.1 mM, 0.05 mM and 0 mM. After mixing, the samples were immediately cooled to 65°C and incubated at this temperature for 5 minutes. The reactions were terminated by placing the samples on ice and adding 10 ~1 of stop buffer. The samplers were heated to 85°C for 30 sec and 10 p.l of each reaction were resolved by electrophoresis through a 10%
polyacrylamide gel ( 19:1 cross-link), with 7 M urea in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described in Example 8a. The DNA was transferred to the membrane and the membrane was dried, washed in 1X Sequenase Images Blocking Buffer (USB), treated with 1 X SAAP buffer and reacted with CDP-Star'"'' (Tropix) and exposed to X-ray film as described in Example 8a. The resulting autoradiograph is shown in Figure 44.

In Figure 44, the lane marked "M" contains molecular weight markers prepared as described in Example 8. Lanes 1-6 contain the cleavage products generated by cleavage of the 419 mutant and lanes 7-12 contain the cleavage products generated by cleavage ~of the 422 mutant. The reaction products generated by cleavage of the ds DNA substrates in 10 mM ' MOPS, pH 8.2 containing 0.5 mM (lanes 1,7); 0.25 mM (lanes 2,8); 0.15 mM
(lanes 3,9); 0.1 mM (lanes 4,10); 0.05 mM (lanes S,11 ) and 0 mM MnCh (lanes 6,12) are shown.
The results shown in Figure 44 show no cleavage is seen in the absence of divalent cations as is also the case for cleavage of ss DNA substrates [see Example 11 (a) and Figure 31 ]. Optimal cleavage (i. e., production of the most distinct pattern of cleavage fragments) of ds DNA substrates was seen in the presence of 0.25 mM MnCh. This optimum is considerably lower than that obtained using ss DNA substrates [Example 11 and Figure 31 show that cleavage of ss DNA substrates was optimal in 1 mM MnCh.]. Figure 44 shows that the efficiency of cleavage of ds DNA substrates decreases as the concentration of MnCh is lowered; this effect is likely due to the lower efficiency of the enzyme in decreasing concentrations of MnCh.
Figure 44 shows that the cleavage pattern for dsDNA substrates apparently disappears when high concentrations of MnCh (0.5 mM, lanes 1 and 7) are-employed in the cleavage reaction. This result is in contrast to the results obtained when cleavage reactions are performed on single-stranded DNA (ssDNA) substrates. Example 11 (a) showed that efficient cleavage of ss DNA substrates were obtained in 1 mM MnCI, and little change in the cleavage pattern was seen when the Mn'+ concentration varied from 1 to 4 mM.
The loss of the signal seen when ds DNA substrates are cleaved in buffers containing high concentrations of MnCh may be explained as follows. The presence of high concentrations of divalent ions promotes the reannealling of the DNA strands of the ds substrate during the course of the cleavage reaction. The enzyme CleavaseT"' BN can nibble ds DNA substrates from the 5' end (i.e., the enzyme removes short DNA
fragments from the 5' end in an exonucleolytic manner; see Example 5). This nibbling results in the apparent removal of the label from the substrate DNA (as the DNA contains a 5' end label). Very short DNA fragments which contain the 5' end label are not visualized as they run out of the gel or are not efficiently transferred to the membrane.

WO 96115267 PC"T/US95/14673 Detection of Cleavage Patterns Can Be Automated The ability to detect the characteristic genetic fingerprint of a nucleic acid substrate generated by the cleavage reaction using fluorescently labelled substrates in conjunction with automated DNA sequencing instrumentation would facilitate the use of the CFLPT"' method in both clinical and research applications. This example demonstrates that differently labelled isolates (two different dyes) can be cleaved in a single reaction tube and can be detected and analyzed using automated DNA sequencing instrumentation.
Double-stranded DNA substrates, which contained either the N-hydroxy succinimidyl ester JOE-NHS (JOE) or FAM-NHS (FAM) on the sense-strand, were generated using the PCR and primers labelled with fluorescent dyes. The anti-sense strand contained a biotin label. The substrates utilized in this experiment were the 157 by fragments from the wild-type (SEQ ID N0:27) and 422 mutant (SEQ ID N0:71 ) of exon 4 of the tyrosinase gene.
The wild-type and 422 mutant tyrosinase gene substrates were amplified from cDNA
plasmid clones containing either the wild-type [pcTYR-N 1 Tyr, Bouchard, B., et al. ( 1989) J.
Exp. Med. 169:2029] or the 422 mutant [pcTYR-A422, Giebel, L.B., et al. ( 1991 ) 87:1119]
forms of the tyrosinase gene. Each double-stranded substrate was amplified and the 5' ends labelled with either a biotin moiety or a fluorescent dye by using the following primer pairs in the PCR. The anti-sense primer of SEQ ID N0:30 containing a 5'-biotin moiety was obtained from Integrated DNA Technologies, Inc. (IDT, Coralville, IA). The biotinylated anti-sense primer was used to prime the synthesis of both the wild-type and 422 mutant substrates. The sense primer of SEQ ID N0:29 labelled with JOE was used to prime synthesis of the wild-type tyrosinase gene. The sense primer of SEQ ID N0:29 labelled with FAM was used to prime synthesis of the 422 mutant tyrosinase gene. The JOE and FAM-labelled primers were obtained from Genset (Paris, France).
The PCR reactions were carried out as follows. One to two nanograms of plasmid DNA from the wild-type or 422 mutant were used as, the target DNA in a 100 p.l reaction containing 50 q.M of each dNTP, and 1 ~.M of each primer in a given primer pair, in 1 X PCR
buffer. Tubes containing the above mixture were overlaid with 70 ~1 Chill Out 14TM liquid wax (MJ Research, Watertown, MA). The tubes were heated to 95°C for 1 min and then cooled to 70°C. Taq DNA polymerase (Perkin-Elmer) was then added as 2.5 units of enzyme in 5 q,l of 1X PCR buffer. The tubes were heated to 95°C for 45 sec, cooled to 50°C for 45 WO 96115267 PC"T/US95/14673 sec, heated to 72°C for 1 min and 15 sec for 35 repetitions. Following the last repetition, the tubes were incubated at 72°C for 5 min.
The PCR products were gel purified as follows. The products were resolved by electrophoresis through a 6% polyacrylamide gel (29:1 cross-link) in a buffer containing O.SX
TBE. The DNA was visualized by ethidium bromide staining and the 157 by fragments were , excised from the gel. The DNA was eluted from the gel slices by passive diffusion overnight into a solution containing 0.5 M NH40Ac, 0.1 % SDS and 0.1 M EDTA. The DNA was then precipitated with ethanol in the presence of 4 p.g of glycogen carrier. The DNA was pelleted and resuspended in 30 p.l of H,O.
The cleavage reactions were performed as follows. Approximately 100 fmoles of each double-stranded DNA substrate (1-3 p.l of each gel purified DNA) in a total volume of 6 yl in HBO was placed in a 500 ~,l thin wall microcentrifuge tube (Perkin-Elrner) _ The tube was heated to 95°C for 10 seconds to denature the substrates and then the tube was quickly cooled to 50°C (this step allows the DNA to assume its unique secondary structure by allowing the formation of intra-strand hydrogen bonds between complimentary bases). The cleavage reaction was started by adding 2 p,l of 50 mM MOPS (pH 7.2), 1 p.l of 1 mM
MnCh and 1 p.l of CleavaseT"' BN (50 ng/p.l). The cleavage reaction was performed in a thermocycler (Perkin-Elmer DNA Thermal Cycler 480, Norwalk, CT) programmed to heat to 95°C for 10 seconds and then cooled immediately to 50°C. The reaction was then incubated at 50°C for 5 minutes and stopped by the addition of 1 pl of 1.0 mM EDTA.
Following the cleavage reaction, the sample was resolved by gel electrophoresis using an ABI 373A DNA Sequericer (Foster City, CA). Prior to loading, the sample was denatured by adding 5 ~l of a solution containing 95% formamide and 10 mM EDTA and heating to 90°C for 2 minutes. Fivemicroliters of the sample was resolved by electrophoresis through a 6% polyacrylamide gel ( 19:1 cross-link), with 6 M urea, in 1 X TBE buffer (89 mM Tris-Borate, pH 8.3, 2 mM EDTA). The gel was run at 30 watts for 14 hours. Signals from four wavelength channels were collected using the Applied Biosystem Data Collection program on a Macintosh computer. The raw data was analyzed with the BaseFinder program [Giddings.
M., et al. (1993) Nucl. Acids Res. 21:4530 which corrects for the fluorescent spectrum overlap in the four channel signals and mobility shifts caused by the use of different dye labels.
The results are shown in Figure 45. Figure 45 shows two traces representing the two channel signals for the wild-type and mutant samples. The wild-type sample.
which was labeled with JOE dye, is shown by the thin lines. The mutant sample (R422Q).
which was labeled with FAM dye, is shown by the thick lines. For comparison, a photograph of a high resolution polyacrylamide gel ( 10% gel with 19:1 crosslink) containing the resolved cleavage - products is shown above the traces (the top lane contains cleavage fragments produced by r 5 clevage of the wild-type substrate; the bottom lane contains cleavage fragments produced by clevage of the R422Q mutant substrate). The cleavage products shown in the gel, which contain biotin labels at the 5' end of the sense strand, were l;enerated, transferred to a nylon membrane and visualized as described in Example 8a. Arrows point from selected bands seen upon cleavage of the 422 mutant substrate_to the corresponding peaks in the trace generated by the automated DNA sequencer (the arrows are labelled 1 through 7 beginning with the left-hand side of Figure 45).
Comparison of the two traces shows that differences i.n the cleavage patterns generated from the cleavage of the wild-type and 422 mutant substrates in the same reaction are detected using automated DNA sequencing instrumentation. For example, cleavage of the 422 substrate generates a cleavage product of approximately 56 nucleotides which is not seen when the wild-type substrate is cleaved. This 56 nucleotide product is seen as the peak depicted by arrow 6 in Figure 45. Figure 45 shows that not only are new cleavage products generated by cleavage of the mutant substrate, but that the cleavage of certain structures is enhanced in the mutant substrate as compared to the wild-type substrate (compare the intensity of the peaks corresponding to arrows 2-5 in the wild-type and mutant traces). In addition, certain cleavage products are shared between the two substrates and serve as reference markers (see arrows l and 7).
The above results show that automated DNA sequencing instrumentation can be used to detect the characteristic genetic fingerprint of a nucleic acid substrate generated by the cleavage reaction. The results also demonstrate that the cleavage reaction can be run as a multiplex reaction. In this experiment both the wild-type and the mutant ds DNA substrates were cleaved in the same reaction (i. e., a multiplex reaction) and then were resolved on the same electrophoretic run using an automated DNA sequencer.

Identification of Viral Strains Using the CleavaseT"'' Reaction The above examples demonstrate that the CleavaseT"'' reaction could be used to detect single base changes in fragments of varying size from the human ~3-globin and tyrosinase genes. These examples showed the utility of the Cleavase~ reaction for the detection and characterization of mutations in the human population. The ability of the Cleavase~"' reaction to detect sequence variations characteristic of different strains of a virus was next examined.
The simian immunodeficiency virus (SIV) infection of monkeys is a widely used animal model for the transmission of human immunodeficiency virus type-1 (HIV) in humans.
Biological isolates of SIV contain multiple virus strains. When a monkey is infected with a biological isolate of SIV, unique subsets of the virus stock are recovered from the infected animals (specific strains are also able to infect tissue culture cells).
Different genotypes of the virus are isolated from infected animals depending on the route of infection [Trivedi, P. c~t crl.
Journal of Virology 68:7649 (1994)]. The SIV long terminal repeat (LTR) contains sequences which vary between the different viral strains and can be used as a marker for the identification of the viral genotype.
In order to develop a rapid method for the identification of viral strains) in a sample (c.g., a clinical isolate), the CleavaseTM reaction was used to characterized SIV genotypes isolated after infection of cultured cells in vitro or after infection of rhesus monkeys by either intravenous or intrarectal inoculation with uncloned biological SIV stocla .
Six clones generated from viral DNA isolated following in vitro infection of the CEMx174 cell line (L.CEM/251/12 clone), after intravenous inoculation of monkeys (L100.8-1 clone), after intrarectal low-dose inoculation of monkeys (L46.16-10 and L46.16-12 clones) and after intrarectal high-dose inoculation of monkeys (L19.16-3 and L36.8-3 clones) were obtained from C. David Pauza (Wisconsin Primate Research Center, Madison, WI). These clones were generated as described by Trivedi, P. et al. Journal of Virology 68:7649 (1994). These plasmid clones contained viral LTR sequences and were utilized to generate double-stranded DNA (ds DNA) substrates for the cleavage reaction.
A) Preparation Of The Substrate DNA
The six SIV plasmids were utilized as templates in PCRs in order to generate dsDNA
substrates for the cleavage reaction. The primer pair utilized spans the U3-R
boundary in the SIV LTR and amplifies an approximately 350 by fragment. This portion of the SIV LTR
contains recognition sequences for transcription factors (including Sp 1 and NF-kB) as well as the TATA box for transcription initiation and is polymorphic in different viral strains - [Trivedi, P. et al., supra].
The primer pair consisting of SEQ ID NOS:74 and 75 was used to amplify the SIV
LTR clones in the PCR. SEQ ID N0:61 primes synthesis of the (+) strand of the SIV LTR
and comprises 5'-GGCTGACAAGAAGGAAACTC-3'. SEQ ID N0:62 primes synthesis of the (-) strand of the SIV LTR and comprises 5'-CCAGGCGGCG GCTAGGAGAGATGGG-3'. To visualize the cleavage pattern generated by cleavage of the (+) strand of the LTR, the PCR was performed using the primer consisting of SEQ ID N0:61 containing a biotin label at 5' end and unlabeled primer consisting of SEQ ID N0:62. To visualize the cleavage pattern generated by clevage of the (-) strand of the viral LTR, the PCR was performed using the primer pair consisting of SEQ ID N0:62 containing a biotin label at the 5' end and unlabeled primer SEQ ID N0:61.
The PCR reactions were carried out as follows. Ten to twenty nanograms of plasmid DNA from each of the above 6 SIV LTR clones was used as the target DNA in separate 100 ~,1 reactions containing 60 pM of each dNTP, 0.2 p.M of each primer in a given primer pair, 10 mM Tris-Cl, pH 9.0 (at 25°C), 2 mM MgCh, 50 mM K<:1, with 0.1%
Triton X-100.
Tubes containing the above mixture were overlaid with two drops of light mineral oil and the tubes were heated to 96°C for 3 min and Taq DNA polymerase (Perkin-Elmer) was then added as 2.5 units of enzyme in 0.5 ~.l of 20 mM Tris-HCI, pH 8.0, 100 mM KCI, 0.1 mM
EDTA, 1 mM DTT, 50% glycerol and 0.5% Tween 20 and 0.5% Nonidet P-4~0. The tubes were heated to 96°C for 45 seconds, cooled to 60°C for 45 seconds, heated to 72°C for 1 minute for 35 repetitions. Following the PCR, the reaction mixture was separated from the mineral oil and 5 p,l of SM NaCI, 4 pl of 10 mg/ml glycogen and 250 ~,l of 100% ethanol were added to the aqueous PCR samples. After incubation at -20°C for 1 hour, the DNA was pelleted by centrifugation in a Marathon Micro A centrifuge (Fisher Scientific) at maximum speed for 5 minutes and resuspended in 40 q,l of 10 mM Tris-HCI, pH 8.0, 0.1 mM EDTA
TE.
The PCR products were gel purified as follows. The DNA was mixed with 0.5 ' volume of loading buffer (95% formamide, ~mM EDTA, 0.02% bromphenol blue, 0.02%
xylene cyanol) and heated to 75°C for 2 minutes. The products were resolved by r electrophoresis through a 6% polyacrylamide denaturing gel. ( 19:1 cross-link) in a buffer containing 7M urea, O.SX TBE. The DNA was visualized by ethidium bromide staining and the product bands were excised from the gel. The DNA was eluted from the gel slices by passive diffusion overnight into a solution containing 0.5 M NH40Ac, 0.1 % SDS
and 0.1 M
EDTA. The DNA was then precipitated with ethanol in the presence of 4 p.g of tRNA
carrier. The DNA was pelleted and resuspended in 50 ~.l of 0.2 M NaCI, 10 mM
Tris-HCI, pH8.0, 0.1 mM EDTA. The DNA was precipitated with ethanol and resuspended in 50 E~l of TE. The final DNA concentration was estimated to be 40 fmole/~1 for each double-stranded SIV LTR PCR product.
B) DNA Sequence Analysis Of The SIV LTR PCR Products The DNA sequence of the six PCR fragments generated in section a) above was determined using the fmol~'~"'' DNA Sequencing System (Promega) according to the manufacturer's instructions. For each set of the sequencing reactions 0.2 pmoles of the PCR
product and 2 pmoles of one of the two 5'-biotinylated PCR primers SEQ ID
NOS:74 and 75 was used (i.e., both strands of the PCR fragments were sequenced). Following the sequencing reactions, the sequencing products were resolved by electrophoresis. After electrophoresis, the DNA bands were visualized following transfer to a nylon membrane as described in Example 17 with the following modification. A solution containing 0.2% Blocking reagent (Boehringer-Mannheim) and 0.2% SDS in TBS buffer ( 100-mM Tris-HCI, pH7.4; 68 mM
NaCI) was used in place of the 1X Sequenase Images Blocking Buffer (USB).
The sequence of the 351 by fragment derived from the L100.8-1 LTR clone is listed in SEQ ID N0:63. The sequence of the 340 by fragment from the L46.16-10 LTR clone is listed in SEQ ID N0:64. The sequence of the 340 by fragment derived from the L46.16-12 LTR clone is listed in SEQ ID N0:65. The sequence of the 351 by fragment from the L19.16-3 LTR clone is listed in SEQ ID N0:66. The sequence of the 351 by fragment derived from the LCEM/251/12 LTR clone is listed in SEQ ID N0:67. The sequence of the 351 by fragment derived from the L36.8-3 LTR clone is listed in SEQ-ID N0:68.
Analysis of sequenced LTR fragments shows that they have multiple substitutions and a deletion relative to the L100.8-1 LTR sequence (SEQ ID N0:63); the L100.8-1 LTR
sequence was chosen as a reference to permit comparisons between the six LTR
clones. For the ease of discussion, the first or 5'-terminal nucleotide of the (+) strand of L 100.8-1 LTR
sequence (SEQ ID N0:63) is defined as number 1 and the last or 3'-terminal nucleotide of this sequence is defined as number 351.

Figure 46 displays the nucleotide sequence of the six LTR clones. The reference clone, L.100.8-1 (SEQ ID N0:63), is shown on the top line. Sequences appearing in bold type represent sequence changes relative to the sequence of clone L.100.8-1 (SEQ ID N0:63).
The sequences outlined by the brackets in Figure 46 represent palindromic sequences which ,; 5 can form a very stable hairpin structure having a stem of 14 base pairs (12/14 bases in the stem are complementary) and a loop of 7 nucleotides in the reference clone L.100.8-1 (SEQ
ID N0:63). This hairpin structure is present in all six LTR clones although the sequence of the stem and loop structures varies between the clones.
In comparison with L100.8-1 sequence (SEQ ID N0:63), the L46.16-10 sequence (SEQ ID N0:64) has seven substitutions and one 11 nucleotide deletion corresponding to nucleotides 65-75 of SEQ ID N0:63. The substitutions are: C to T in position 28 (C28T), C57T, G90A, C97T, G238A, G242A and G313A. The L46.16-12 sequence (SEQ ID
N0:65) has seven substitutions and one 11 nucleotide deletion corresponding to nucleotides 65-75 of SEQ ID N0:63. The substitutions are: C28T, C57T, G90A, C97T, A103G, G242A and G313A. L 19.16-3 sequence (SEQ ID N0:66) has two substitutions: A94C and A317T.
LCEM/251/12 sequence (SEQ ID N0:67) has seven substitutions: G26A, G72A, C97T, G258A, A281C, G313A and C316T. L36.8-3 sequence (SEQ ID N0:68) has six substitutions: G60A, G172A, G207A, G221A, T256C and C316T.
C) Cleavage Reaction Conditions And CFLPT"° Analysis Of The (-) Strand Of The SIV LTR
Double-stranded substrates corresponding to the SIV LTR sequences listed in SEQ ID
NOS:62-6$ were labelled on the (-) strand using the PCR and the primer pair corresponding to SEQ ID NOS:61 and 62. The primer of SEQ ID N0:62 [the (-) strand primer]contained a biotin label at the 5' end. The PCR was performed and the reaction products were isolated as described in section a).
The cleavage reactions were performed as follows. Reaction tubes were assembled with approximately 60 fmoles of the ds DNA substrates in 6 q.l of water. The following reagents were added to the DNA: 2 ql of SX CFLPT"' buffer (pH 7.2) containing 150 mM
KCl (to yield a final concentration of 30 mM KCl) and 1 ql of the CleavaseT"' BN enzyme (25 ng in 1X dilution buffer). A reaction tube containing the above components with the exception that 1 p.l of HBO was added in place of the Cleava.seTM BN enzyme was prepared and run as the uncut or no enzyme control. The tubes were brought to95°C for 10 seconds in a PTC-100T"'' Programmable Thermal Controller (MJ Rese;arch, Inc.) to denature the DNA.

Following the denaturation step, the tubes were immediately cooled to 40°C. The cleavage reaction was immediately started by the addition of 1 ~.l of 2 mM MnCI, (to achieve a final concentration of 0.2 mM). The tubes were incubated at 40°C for 5 minutes. The reactions were terminated by adding 6 ~.1 of stop buffer. The samples were-heated to 85°C for 30 sec and 5 ~.l of each reaction were resolved by electrophoresis through a 12%
polyacrylamide gel ( 19:1 cross-link), with 7 M urea in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described in Example 8a. The DNA was transferred to the membrane and the membrane was dried, washed in 0.2% Blocking reagent (Boehringer Mannheim);
0.2°I° SDS
in 100 mM Tris-HCI, pH 7.43 68 mM NaCI, treated with 1X SAAP buffer and reacted with CDP-StarTn'' (Tropix) and exposed to X-ray film as described in Example 8a.
The resulting autoradiograph is shown in Figure 47.
Figure 47 shows the cleavage patterns which correspond to the cleavage of the (-) strand of the double-stranded LTR substrates. In Figure 47, the lane marked "M" contains molecular weight markers (prepared as described in Example 8). Lanes 1-6 contain the cleavage products generated by cleavage of the L100.8-l, L46.16-10, L46.16-12, L19.16-3, LCEM/251/12 and L36.8-3 LTR PCR fragments, respectively. Lanes 7-12 contain the uncut controls of each of the 6 LTR substrates in the same order as that described for Lanes 1-6.
The results shown in Figure 47 show that the cleavage or CFLP~'~'~'' pattern for each LTR substrate contains multiple bands which range in size from approximately nucleotides (the uncut substrate) to less than 24 nucleotides. The bands located below about 100 nucleotides in length show differences between the six clones which reflect nucleotide changes characteristic of the-different- SIV LTR isolates. Examination of the CFLPT''~ patterns revealed that the reaction detected five unique cleavage patterns among the six SIV LTR
isolates. From the DNA sequence data, it was known that all six LTR clones were unique.
However, the CFLPT"'' pattern appeared to be identical for the clones shown in lanes 2 and 3.
The CFLP~ patterns generated by cleavage of the (-) strand from all six substrates contain a strong band which corresponds to a fragment of approximately 100 nucleotides in length. This band corresponds to cleavage of all six LTR substrates at the long palindromic sequence located 97 nucleotides from the 5' end of the (-) strand (see the bracketed region in Figure 46). This palindromic sequence- forms a very stable hairpin structure in single-stranded DNA and provides an optimal substrate for the Cleavase~ BN enzyme. Cleavage of this hairpin structure is predicted to generate a fragment of approximately 100 nucleotides.

The LTR substrates, L46.16-10 (SEQ ID N0:64) and. L46.16-12 (SEQ ID N0:65), shown in lanes 2 and 3 were generated from the same animal using the same route of infection [Trivedi, P. et al., supra]. These substrates. have identical sequences in the region corresponding to the detectable cleavage sites (i. e., below 100 nucleotides) with the exception S of a single base; the L46.16-10 clone (SEQ ID N0:64) contains a G to A
change at position 239 (G239A) relative to the reference sequence listed in SEQ ID N0:63 .
Examination of the DNA sequence of these two clones reveals that this substitution is located in the loop region of a strong hairpin structure (see the palindromic region bracketed in Figure 46). Because the single base difference between these two sequences is located in the loop region of the hairpin structure, it may not change DNA secondary structure of the two substrates sufficiently to generate different CFLP~'~"'' patterns under the conditions utilized here. It may be possible to detect this single base difference between these two clones by varying the reaction conditions in a way that destabilizes the strong hairpin structure.
The results shown in Figure 47 demonstrate that the CFLP-'~"'' reaction can be used to detect the majority (5/6 or 83%) of the sequence variations present in the six SIV LTR clones studied. In addition, Figure 47 demonstrates that the CFLPTM reaction is a sensitive means for probing the secondary structure of single strands of nucleic acids.
D) Cleavage Reaction Conditions And CFLPT"' Analysis Of The (+) Strand Of The SIV LTR
Double-stranded substrates corresponding to the SIV :LTR sequences listed in SEQ ID
NOS:76-81 were labelled on the (+) strand using the PCR and the primer pair corresponding to SEQ ID NO: 74 and 75. The primer of SEQ ID N0:61 [the (+) strand primer]contained a biotin label at the 5' end. The PCR was performed and the reaction products were isolated as described in section a). The cleavage reactions, electrophoresis and DNA
visualization were performed as described above in section c). The resulting autoradiograph is shown in Figure 48.
Figure 48 shows the resulting pattern corresponding to the cleavage products of the (+) strand of six SIV LTR fragments. The lane marked."M" contains molecular weight markers (prepared as described in Example 8). Lanes 1-6 contain the cleavage products generated by cleavage of the L100.8-1, L46.16-10, L46.16-12, L19.16-3, LCEM/251/12 and L36.8-3 LTR
PCR fragments, respectively. Lanes 7-12 contain the uncut controls of each of the 6 LTR
substrates in the same order as that described for Lanes 1-6.

WO 96/15267 PC"TlUS95l14673 As was shown for cleavage of the (-) strand of the LTR clones, the CFLP~'T'' pattern for each (+) strand of the SIV LTR substrates contains unique features that characterize the majority of specific nucleotide substitutions. For example, deletion of 11 nucleotides can be easiiy detected for L46.16-10 (SEQ ID N0:64) and L46.16-12 (SEQ ID NO:65) (Figure 48, ' lanes 2 and 3). This deletion removes one of the three SpI binding sites and is a change , characteristic of the genotype of SIV which predominates in animals which are infected using low-doses of virus stock via intrarectal inoculation [Trivedi, P. et al., supra]. The CFLP-'~"'' pattern generated by cleavage of the (+) strand of the substrates derived from clones L46.16-and L46.16-12 again were identical under these reaction conditions.
10 The results shown above demonstrate that the CFLP'~"'' reaction can be used as a means to rapidly identify different strains (i. e., genotypes) of virus. The ability to rapidly identify the particular strain of virus or other pathogenic organism in a sample is of clinical importance. The above results show that the CFLP~ reaction can be used to provide a fast method of strain or species identification.

The Effects Of Alterations In Salt Conditions In Cleavage Reactions Using A Single-Stranded DNA Substrate : _ _ - r In Example 11 it was shown that the Cleavase~'~"'' reaction is insensitive to large changes in reactions conditions when a single-stranded DNA is-employed as the substrate.
Example 11 showed that the cleavage reaction can be performed using a range of salt concentrations (0 to 50 mM KCl) in conjunction with single-stranded substrates. In this example, the effect of substituting other salts in place of KCI was examined in cleavage reactions using single-stranded DNA substrates.
A) Effect Of Substituting NaCI For KCl In Cleavage Reactions Using A
Single-Stranded Template To determine the effect of substituting NaCI in place of KCI upon the cleavage pattern created by 5' nuclease activity on a single-stranded DNA substrate, the following experiment was performed. A single template was incubated in the presence of a fixed amount of the Cleavase~'~"' BN enzyme (50 ng) in a buffer containing 10 mM MOPS, pH 8.2, 1mM
MnCh and various amounts of NaCI.
- 1~2 -Approximately 100 fmoles of the 157 nucleotide fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID NO 47; prepared as described in Example 8b) were placed in a 500 p.l thin wall microcentrifuge tubes (Perkin Elmer, Norwalk, CT) in 1 X CFLP-'~"' buffer, pH 8.2 and 1.33 mM MnCI, (to yield a final concentration of 1 mM
MnCI,) in a volume of 15 p,l. NaCI was added to yield a final concentration of 0, 10, 20, 30, 40, 50, 75 or 100 mM. The final reaction volume was 20 yl.
A tube containing 10 mM MOPS, pH 8.2, 1 mM MnCI, and 100 fmoles substrate DNA was prepared and served as the no salt, no enzyme control (sterile distilled water was substituted for Cleavase~'~"'' BN and all reaction components were added prior to denaturation at 95°C).
The tubes were heated to 95°C for 20 seconds and then rapidly cooled to 65°C. The cleavage reaction was started immediately by the addition of 5 p,l of a diluted enzyme mixture comprising 1 pl of CleavaseT"'' BN [50 ng/~1 in 1 X dilution buffer (0.5%
NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 ~g/ml BSA)J in 10 mM MOPS, pH 8.2, without MnCI,.
After 5 minutes at 65°C, reactions were stopped by the addition of 16 pl of stop buffer (95% formamide, 10 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol).
Samples were heated to 72°C for 2 minutes and 7 ~1 of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7M urea, in a buffer containing O.SX TBE, as described in Example 8a.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical), washed, reacted with CDP-StarT'"' (Tropix, Bedford, MA) as described in Example 8a with the exception that 0.01 ml CDP-Star"' was added per cm'- of membrane. The membrane was exposed to X-ray film as described in Example 8a. The results are shown in Figure 49.
In Figure 49, the lane marked "M" contains molecular weight markers as described in Example 8a. Lane 1 contains the no salt, no enzyme control and shows the mobility of the uncleaved template DNA. Lanes 2 through 9 contain reaction products incubated in the presence of CleavaseT"' BN enzyme in a buffer containing 0, 10, 20, _ 30, 40, ~ 0 75 or I 00 mM NaCI, respectively.

WO 96/15267 PG"T/US95114673 The results shown in Figure 49 demonstrate that the substitution of NaCI in place of KCl has little or no effect upon the cleavage pattern generated using the 157 nucleotide tyrosinase DNA substrate (SEQ ID N0:34). Essentially the same dependence of the cleavage pattern on salt concentration was observed using this single-stranded DNA
substrate when either KCl (See example 13b, Figure 32) or NaCI (Figure 49) was employed in the cleavage reaction.
B) Effect Of Substituting (NIi4)ZS04 For KCl In Cleavage Reactions Using A
Single-Stranded Template _ In an approach similar to that described in above in section a), the effect of substituting (NH4)~504 in place of KCl upon the-cleavage pattern created by 5' nuclease activity on a single-stranded DNA substrate was tested. Cleavage reactions were set up exactly as described in section a) with the exception that variable amounts of (NH4)~SOa were used in place of the NaCI.
Approximately 100 fmoles of the 157 nucleotide fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID NO 47; prepared as described in Example 8a) were placed in 500 p.l thin wall microcentrifuge tubes (Perkin Elmer.
Norwalk, CT) in 1 X CFLPT"'' buffer, pH 8.2 and 1.33 mM MnCI, (to yield a final-concentration of 1 mM) in a volume of 15 p,l. (NH4),504 was added to yield a final concentration of 0, 10, 20, 30, 40, 50, 75 or 100 mM. The final reaction volume was 20 pl.
A tube containing 10 mM MOPS, pH 8.2, 1 mM MnCl2 and 100 fmoles substrate DNA was prepared and served as the no salt, no enzyme control (sterile distilled water was substituted for Cleavase~ BN and all reaction components were added prior to denaturation at 95°C). _ The tubes were heated to 95°C for 20 seconds and then rapidly cooled to 65°C. The cleavage reaction was started immediately by the addition of 5 pl of a diluted enzyme mixture comprising 1 p.l of CleavaseT'~' BN [50 ng/ml in 1 X dilution buffer (0.5%
NP40, 0.5%
Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 mg/pl BSA)~ in 10 mM MOPS, pH
8.2, without MnCh.
After 5 minutes at 65°C, reactions were stopped by the addition of 16 pl of stop buffer. Samples were heated to 72°C for 2 minutes and 7 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7M urea, in a buffer containing O.SX TBE, as described in,Example 8a.

. After electrophoresis, the DNA was transferred to a membrane and the detected as described in section a) above. The resulting autoradiograph is shown in Figure 50.
In Figure 50, the lane marked "M" contains molecular weight markers as described in example 10a. Lane 1 contains the no enzyme control and shows the mobility of the uncleaved template DNA. Lanes 2 through 9 contain reaction products incubated in the presence of Cleavase~'~"'' BN enzyme in a buffer containing 0, 10, 20, 30, 40, 50 75 or 100 mM (NH4)~SO4, respectively.
The results shown in Figure 50 demonstrate that the cleavage reaction is severely inhibited by the presence of (NH4)~SO4. The reaction is completely inhibited by as little as 20 mM (NH4)~504; the extent of the cleavage reaction in 10 mM (NH4)~504 is comparable to that obtained in 50 mM KCl or NaCI and is significantly reduced relative that obtained at 0 mM (NH4),SO4. The pattern of cleavage obtained at 10 mM (NH4)~SO4, however, is identical to that observed when the 157 nucleotide template (SEQ ID N0:34) incubated in the absence of (NH4)~504 or in KCl or NaCI. This indicates that the choice of salt included in the cleavase reaction has no effect on the nature of the sites recognized as substrates by the CleavaseT"'' BN enzyme (i.e., the inhibitory effect seen is duc: the effect of (NH4)~504 upon enzyme activity not upon the formation of the cleavage structures).
C) Effect of Increasing KCl Concentration on the Cleavage of Single-Stranded Substrates The effect of increasing the concentration of KCl in cleavage reactions using a single-stranded DNA substrate was examined by performing the cleavage reaction in concentrations of KCl which varied from 0 to 100 mM. The cleavage reactions were performed as described in section a) with the exception that KCl was added to yield final concentrations of 0, 2~. 50, 75 or 100 mM and 200 fmoles of the substrate were used in the reaction;
additionally the substrate DNA was denatured by incubation at 95°C for 5 seconds.
Following the cleavage reaction, the samples were ele:ctrophoresed, transferred to a membrane and detected as described in section a) above. The resulting autoradiograph is shown in Figure 51.
In Figure 51, the lanes marked "M" contains molecular weight markers as described in Example 8a. Lane 1 is the no enzyme control; Lanes 2-7 contain reactions carried out in the presence of 0, 25, 50, 75, 100 or 100 mM KCl (the 100 mlVl sample was repeated twice).
respectively.

WO 96!15267 PCT/US95/14673 The results shown in Figure 51 demonstrate that the extent of cleavage in the cleavage reaction decreased as a function of increasing KCl concentration (although residual cleavage was detectable at 100 mM KCl). Furthermore, the pattern of fragments generated by cleavage of single-stranded substrates with Cleavase~ BN is unaffected by the concentration of KC1 present in the reactions.
D) Effect Of High KCl Concentrations On Cleavage Reactions Using A Single-Stranded Substrate The ability of the Cleavase~ reaction to be carried out at relatively high concentrations of KCl was tested by performing the cleavage reaction in the presence of variable concentrations of KCl in excess of 100 mM. The reactions were performed using the 157 nucleotide fragment from exon 4 of the tyrosinase gene (SEQ ID N0:34) as described above in section c), with the exception that KCl was added to yield a final concentration of 0;
100, 150, 200. 250 or 300 mM.
Following the cleavage reaction, the samples were electrophoresed, transferred to a membrane and detected as described in section a) above. The results (data not shown) indicated that the cleavage reaction was severely inhibited by KCl-concentrations in excess of 100 mM. Some residual cleavage did, however, persist at these elevated salt concentrations, up to and including 300 mM KCI.
E) Effect Of KCl Concentration On The Stability Of The Cleavage Pattern During Extended Incubation Periods The results presented above demonstrate that the Cleavase~ reaction is inhibited by elevated concentrations (i.e., above 50 mM) of either KCl or NaCI. To determine whether this iWibition would effectively result in the stabilization of the cleavage pattern after extended reaction times (i.e., due to inhibition of enzyme activity), reactions were examined at varying extended time points at both 0 and 50 mM KCI.
Approximately 100 fmoles of the 157 nucleotide fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID NO 47; prepared as described in example l0a) were placed in 200 p,l thin wall microcentrifuge tubes (BioRad, Richmond, CA) in 1 X
CFLP~'~''' buffer, pH 8.2, 1.33 mM MnCh (to yield a final concentration of I
mM) and KC1 to yield a final concentration of 0 or 50 mM KCI. The final reaction volume was 20 p.l.
Control reactions which lacked enzyme were set up in parallel for each time point examined; these no enzyme controls were prepared as described above with the exception that sterile distilled water was substituted for Cleavase~'~"'' BN and all reaction components were added prior to denaturation at 95°C.
The tubes were heated to 95°C for 20 seconds and then rapidly cooled to 65°C. The cleavage reaction was started immediately by the addition of 5 ~l of a diluted, enzyme mixture comprising 1 ~1 of Cleavase~ BN [50 ng/ml in 1 X dilution buffer (0.5% NP40, 0.5%
Tween 20,-20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 ~g/ml BSA)] in 10 mM MOPS, pH
8.2, without MnCl2. Twenty microliters of Chill Out 14T"'' (MJ Research, Watertown, MA) were added to each tube after the addition of the enzyme. 'the reactions were allowed to proceed at 65°C for 5 min, 30 min, 1 hour, 2 hours, 4 hours and 17 hours.
At the desired time point, the reactions were stopped by the addition of 16 p.l of stop buffer. Samples were heated to 72°C for 2 minutes and 7 ~1 of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7M urea, in a buffer containing 0.5X TBE, as described in Example 8a.
After electrophoresis, the DNA was transferred to a membrane and the detected as described in section a) above. The resulting autoradiograph is shown in Figure 52.
In Figure 52, the lane marked "M" contains molecular weight markers as described in example 10a. Lanes 1-10 contain products from reactions carried out in the absence of KCI;
lanes 11-20 contain products from reactions carried out in the presence of 50 mM KCI.
Lanes l, 3, 5, 7, and 9 contain no enzyme controls incubated for 5 minutes, 30 minutes, 1 hour, 2 hours, 4 hours and 17 hours, respectively. Lanes 2, 4, 6, 8, ald 10 contain the reaction products from reactions incubated at 65°C for 5 minutes, 30 minutes, 1 hour, 2 hours, 4 hours and 17 hours, respectively. Lanes 11, 13, 15, 17, and 19 contain no enzyme controls incubated in 50 mM KCl for 5 minutes, 30 minutes, 1 hour, 2 hours, 4 hours and 17 hours, respectively. Lanes 12, 14, 16, 18 and 20 contain reaction products from CFLP'~"'' reactions incubated in 50 mM KCl at 65°C for 5 minutes, 30 minutes, 1 hour, 2 hours, 4 hours and 17 hours, respectively.
The results indicated that cleavage was retarded in the presence of 50 mM KCl which resulted in a significant stabilization of the cleavage pattern (i.e., the cleavage pattern remained the same over time because the rate of cleavage was dramatically slowed and thus the larger cleavage fragments are not further cleaved to produce smaller fragments). Note that at the extended incubation times, the reactions carried out in the absence of KCl were significantly overdigested; after 1 hour at 65°C, essentially no large fragments remain, and there is substantial accumulation of small cleavage products. In contrast, the reactions carried out at 50 mM KCI were essentially static between 30 minutes and 4 hours;
overdigestion was only apparent at the longest time point and was not as extensive as that observed in the absence of KCI. .

Comparison Of The Patterns Of Cleavage Generated By Cleava e-Of Sin le-Stranded And Double-Stranded DNA Substrates In CleavaseT"'' BN-mediated primer-independent cleavage of double-stranded DNA
substrates, the two strands of DNA are separated in a denaturation step prior to the addition of enzyme. Therefore, the patterns generated by cleaving double-stranded templates should be identical to those generated by cleaving single-stranded template. This assumption was verified by the experiment described below.
The single-stranded substrate comprising the 157 nucleotide fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34) was prepared as described in example lOb with the following modification. After gel purification and precipitation in the presence of glycogen carrier, the PCR products were resuspended in TE (lOmM
Tris-CI. pH
8.0, 1 mM EDTA) and then reprecipitated with 2 M NH40Ac and 2.5 volumes of ethanol.
The DNA was then resuspended in 400 ~l of TE.
Approximately 50 or 100 fmoles of the single-stranded 157 nucleotide fragment (SEQ
ID NO: 47) were placed in a 200 pl centrifuge tube (BioRad, Richmond, CA) in 1 X CFLP~
buffer, pH 8.2 and 1.33 mM MnClz (final concentration was 1 mM MnCh) in a volume of 15 ~1. The final reaction volume was 20 ~.1. A 20 ~l no salt, no enzyme control was set up in parallel; this reaction contained sterile distilled water in place of the Cleavase~ BN enzyme and all reaction components were added prior to denaturation at 95°C.
The reaction tubes were heated to 95°C for 5 seconds and then rapidly cooled to 65°C.
The cleavage reactions were started immediately by the addition of 5 p.l of a diluted enzyme mixture comprising 1 ~,l of Cleavase'~ BN [50 ng/~.l in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 ~.g/ml BSA)] in 10 mM
MOPS, pH 8.2, without MnCh. After 5 minutes at 65°C, reactions were stopped by the addition of , 16 ~.I of stop buffer.
A double stranded form of the 157 nucleotide substrate was cleaved with CleavaseT"'' -BN in the same experiment. This double-stranded substrate (SEQ ID N0:27) was generated as described in Example 8b with the following modification s. After gel purification and precipitation in the presence of glycogen carrier, the PCR products were resuspended in TE
( 1 OmM Tris-Cl, pH 8.0, 1 mM EDTA) and then reprecipitated with 2 M NH40Ac and 2.5 ' volumes of ethanol. The DNA was then resuspended in 400 ~.1 of TE.
w 5 Approximately 33 or 66 fmoles of the double-stranded 157 by fragment (SEQ
ID
N0:27) were placed in a 200 p,l thin walled microcentrifuge tube (BioRad, Richmond, CA).
Sterile distilled water was added to a volume of 15 pl.
The reaction tubes were heated to 95°C for 5 seconds and then rapidly cooled to 65°C.
The cleavage reactions were started immediately by the addition of 5 p.l of a diluted enzyme mixture comprising 10 mM MOPS, pH 8.2, 0.8 mIVI MnCh (to yield a final concentration of 10 mM MOPS, pH 8.2 and 0.2 mM MnCI,) and 0.5 ~.l of CleavaseTM BN [50 ng/p.l in 1 X
dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, ~,g/ml BSA)]. A 20 pl no salt, no enzyme double-stranded substrate control was set up in parallel; this reaction contained sterile distilled water in place of the CleavaseTM BN enzyme.
After 5 minutes at 65°C, the reactions were stopped by the addition of 16 pl of stop buffer. The samples were then heated to 72°C for 2 minutes and the reaction products were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7M
urea, in a buffer containing O.SX TBE, as described in Example 8a.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in example 10a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical), washed, reacted with CDP-Star (Tropix, Bedford, MA), and exposed to X-ray film as described in Example 20a.
The resulting autoradiograph is shown in Figure 53.
In Figure 53, lanes 1-3 contain reaction products derived from reactions containing the single-stranded substrate; lanes 4-7 contain reaction products derived from reactions containing the double-stranded substrate. Lanes 1 and 3 contain 7.0 p.l of the reaction products derived from the cleavage reactions which contained either 50 or 100 fmoles of the single-stranded substrate, respectively. Lane 2 contains 7.0 yl of the uncut single-stranded substrate control reaction. Lanes 4 and 6 contain 7.0 ~.l of the uncut double-stranded control reactions which contained either 33 or 66 fmoles of the substrate, respectively. Lanes 5 and 7 contain 7.0 p.l of the reaction products derived from cleavage reactions which contained either 33 or 66 fmoles of the double-stranded substrate, respectively. Note that the uncut double-stranded WO 96/15267 PG"TILTS95/14673 control shows a doublet underneath the prominent band containing the 157 by substrate; it is believed that this doublet represents alternative structures which migrate with an altered mobility rather than degradation products. This doublet does not appear in experiments performed using double-stranded DNA purified from a denaturing gel (See Example 22) Comparison of the cleavage patterns generated by cleavage of either the single-stranded or double-stranded substrate shows that identical patterns are generated.

The Cleavase"~'' Reaction Using A Double Stranded DNA
Template Is Sensitive to Large Changes In Reaction Conditions The results presented in Example 11 demonstrated that the Cleavase~l reaction is relatively insensitive to significant changes in numerous reaction conditions including, the concentration of MnCh and KCI, temperature, the incubation period, the amount of CleavaseT"' BN enzyme added and DNA preparation. The results shown in Example demonstrated that when the Cleavase~ reaction is performed using a single-stranded substrate, the reaction is remarkably robust to large changes in conditions.
The experiments shown below show that the cleavage of double-stranded substrates is restricted to a somewhat narrower range of reaction conditions.
A) Generation Of The Double-Stranded 157 by Fragment Of Exon 4 Of The Tyrosinase Gene The following experiments examine the effect of changes in reaction conditions when double-stranded DNA templates are used in the Cleavase~'~"'' reaction. The double-stranded substrate utilized was the157 by fragment of the wild type tyrosinase gene (SEQ ID N0:27).
This 157 by fragment was generated using symmetric PCR as described in Example 8b.
Briefly, approximately 75 fmoles of double-stranded substrate DNA were incubated with ~0 pmoles of the primer 5' biotin-GCCTTATTTTACTTTAAAAAT-3' (SEQ ID N0:32), 50 pmoles of the primer 5' fluorescein-TAAAGTTTTGTGTTATCTCA-3' (SEQ ID N0:33).
and 50 mM of each dNTP in 1X PCR buffer. Tubes containing 95 p.l of the above mixture were heated to 95°G for 5 seconds and cooled to 70°C. Five microliters of enzyme mix containin;~
1.25 units of Taq DNA polymerase in 1X PCR buffer were then added. Each tube was overlaid with 50 ~l of Chill Out 14~ (MJ Research, Watertown, MA).

The tubes were heated to 95°C for 4~ seconds, cooled to 50°C for 45 seconds, heated to 72°C for 75 seconds for 30 repetitions with a ~ minute incubation at 72°C after the last repetition. The reactions were then ethanol precipitated to rf:duce the volume to be gel purified. NaCI was added to a final concentration of 400 mM and glycogen (in distilled water) was added to a final concentration of 200 p.g/ml. Two and one-half volumes of 100%
ethanol were added to each tube, and the tubes were chilled to -70°C
for two and one-half hours. The DNA was pelleted and resuspended in one-fifth the original volume of sterile distilled water.
The PCR products were gel purified as follows. An equal volume of stop buffer was added to each tube and the tubes were heated to 72°C for 2 minutes. The products were resolved by electrophoresis through a 6 % denaturing polyacrylamide gel ( I
9:1 cross-link) and 7 M urea in a buffer containing 45 mM Tris-Borate, pH 8.3 and 1.4 mM EDTA. The DNA
was visualized by ethidium bromide staining and the 157 by fragment was excised from the gel. The DNA was eluted from the gel slice by passive diffusion as described in Example 8a with the exception that diffusion was allowed to occur over 'two days at room temperature.
The DNA was then precipitated with ethanol in the presence of 200 mM NaCI and no added carrier molecules. The DNA was pelleted and resuspended in 150 pl TE.
S) Effect Of KCl Concentration On The Double-Stranded Cleavage Reaction To determine the effect of salt concentration upon thf: cleavage reaction when a double-stranded substrate was utilized, a single substrate was incubated in the presence of a fixed amount of the enzyme Cleavase'~"' BN (25 ng) in a buffer containing 10 mM MOPS, pH
7.5, 0.2 mM MnCI, and varying concentrations of KCl from 0 to 100 mM.
Approximately 100 fmoles of the 157 by fragment derived from the exon 4 of the tyrosinase gene (SEQ ID N0:27; prepared as described above in section a) were placed in 200 q.l thin wall microcentrifuge tubes (BioRad, Richmond, CA) in sterile distilled water in a volume of 6.25 ~l (the final reaction volume was 10 pl). The tubes were heated to 95°C for 15 seconds and then rapidly cooled to 45°C. The cleavage reactions were started by the addition of 3.75 p.l of an enzyme mix containing 2.7 X CFLP~ buffer, pH 7.5 (to yield a final concentration of 1 X), 0.531 mM MnCh (to yield a final concentration of 0.2 mM), 0.5 yl CleavaseT"'' BN [50 ng/p,l in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM
Tris-HCI, pH 8.0, 50 mM KCI, 10 p,g/ml BSA)], and KCl to yield a final concentration of 0, 2.5, 5. 10, 15, 20, 25, 30, 50 or 100 mM. The final reaction volume was 10 ~1.
The enzyme solution was brought to room temperature before addition to the cleavage reaction. No enzyme (i. e., uncut) controls were set up in parallel at either 0 or 100 mM
KCI, with the difference that sterile distilled water was substituted for the CleavaseTM BN.
After 5 minutes at 45°C, the reactions were stopped by the addition of 8 ~.l of stop buffer. Samples were heated to 72°C for 2 minutes and 4 ~.I of each reaction were resolved "
by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried. washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 54.
In Figure 54, the lane marked "M" contains molecular weight markers. Lane 1 contains the uncut control in 0 mM KCl and shows the mobility of the uncleaved template DNA. Lanes 2 through 11 contain reaction products generated by incubation of the substrate in the presence of CleavaseT"'' BN enzyme in a buffer containing 0, 2.5, 5, 10, 15, 20, 25, 30, 50, or 100 mM KCI, respectively. Lane 12 contains the uncut control incubated in a buffer containing 100 mM KCI.
The results shown in Figure 54 demonstrate that the Cleavase-"~' reaction carried out on double-stranded DNA template was sensitive to variations in salt concentration. Essentially no cleavage was detected in reactions containing- greater than 30 mM KCI. The same cleavage pattern was obtained when the 157 by tyrosinase DNA substrate (SEQ ID
N0:27) was incubated with the CleavaseT"'' BN enzyme regardless of whether the concentration of KCl was varied from 0 to 30 mM.
C) Effect Of NaCI On The Double-Stranded Cleavage Reaction The effect of substituting NaCI in place of KCI upon the cleavage pattern created by 5' nuclease activity on a double-stranded DNA substrate was examined.
Approximately 100 fmoles of the 157 by fragment derived from exon 4 of the tyrosinase gene (SEQ
ID NO 40;
prepared as described in Example 22a) were placed .in 200 ~.l thin wall microcentrifuge tubes (BioRad, Richmond, CA) in sterile distilled water in a volume of 6.25 ~1 and were heated to 95°C for 15 seconds. The tubes were cooled to 45°C. The cleavage-reactionyvas started by the addition of 3.75 ~.l of an enzyme mix containing 2.7 X CFLPT"' buffer, pH
7.5 (to yield a ' final concentration of 1 X), 0.53 mM MnCI, (to yield a final concentration of 0.2 mM). 0.5 ~.1 Cleavase'~"'' BN [50 ng/~I in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20. 20 mM ' Tris-HCI, pH 8.0, 50 mM KCI, 10 ~.g/ml BSA)], and NaCI to yield a final concentration of 0, 2.5, 5, 10, 15, 20, 25, 30, 50 or 100 mM. The final reaction volume was 10 ~1.
No enzyme (i. e., uncut) controls were set up in parallel at either.0 or 100 mM NaCI, with the difference that sterile distilled water was substituted for the CleavaseT"' BN.
The reactions were incubated at 45°C for 5 minutes and were stopped by the addition of 8 ~l of stop buffer. Samples were heated to 72°C for 2 minutes and 4 ~1 of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a with the exception that the distilled water -washes were omitted. The resulting autoradiograph is shown in Figure 55.
In Figure 55 , the lane marked "M" contains molecular weight markers. Lane 1 contains the no enzyme control incubated in a buffer containing 0 mM NaCI and shows the mobility of the uncleaved template DNA. Lanes 2 through 11 contain reaction products generated by cleavage of the 157 by substrate (SEQ ID N0:27) with the CleavaseTM BN
enzyme in a buffer containing 0, 2.5, 5, 10, 15, 20, 25, 30, 50, or 100 mM
NaCI, respectively. Lane 12 contains the no enzyme control incubated in a buffer containing 100 mM NaCI.
The results shown in Figure 55 demonstrate that the CleavaseT"'' reaction carried out on a double-stranded DNA template was sensitive to variations in NaCI
concentration.
Essentially no cleavage was detected above 20 mM NaCI. The same cleavage pattern was obtained when the 157 by tyrosinase DNA template (SEQ ID N0:27) was incubated with the CleavaseTM BN enzyme regardless of whether the NaCI concentration was varied from 0 to 20 mM.
D) Effect Of Substituting (NH4)ZSO4 For KCI In Cleavage Of Double-Stranded Template In an approach similar to that described in Example 20b, the ability of (NH4),SO~, to substitute for KCl in the cleavage reaction when double-stranded substrates were utilized was tested. Cleavage reactions were set up exactly as described in Examples 22b and c with the exception that variable amounts of (NH4)~504 were substituted for the KCl or NaCI.

Approximately 100 fmoles of the 157 by fragment derived exon 4 of the tyrosinase gene (SEQ ID NO 40; prepared as described above in section a) were placed in 200 ~l thin wall microcentrifuge tubes (BioRad, Richmond, CA) , in sterile distilled water in a volume of 6.25 p.l and were heated to 95°C for 15 seconds. The tubes were cooled to 45°C.
Cleavage reactions were started by the addition of 3.75 p.l of an enzyme mix , containing 2.7 X CFLPT"'' buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM
MnCI, (to yield a final concentration of 0.2 mM MnCI,), 0.5 ~,l CleavaseT"" BN
[50 ng/p,l in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM
KCI, 10 Elg/ml BSA)], and (NH4)~SO4 to yield a final concentration of 0, 2.5, 5, 10, 15, 20, 25, 30, 50 or 100 mM. The final reaction volume was 10 p.l. No enzyme (i. e., uncut) controls were set up in parallel at either 0 or 100 mM (NH4).,SO4, with the difference that sterile distilled water was substituted for the Cleavase'M BN.
The reactions were incubated at 45°C for 5 minutes and were stopped by the addition of 8 yl of stop buffer. Samples were heated to 72°C for 2 minutes and 4 ~l of each reaction 1 S were resolved by electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 56.
In Figure 56, the lane marked "M" contains molecular weight markers. Lane 1 contains the no enzyme control incubated in a buffer containing 0 mM
(NH~,)~S04 and shows the migration of the uncleaved substrate DNA. Lanes 2 through 11 contain reaction products generated by incubation of the substrate in the presence of Cleavase~'~"' BN
enzyme in a buffer containing 0, 2.5, 5, 10, 15, 20, 25, 30, 50, or 100 mM (NH4)~504, respectively. Lane 12 contains the no enzyme control incubated in a buffer containing 100 mM
(NH4),SO~.
The results shown in Figure 56 demonstrate that the Cleavase'~''' reaction was severely inhibited by the presence of (NH4)~SO~. The reaction was completely inhibited by as little as 15 mM -(NH4),504; the extent of the cleavage reaction in 5 mM (NH4),SO~ v~ras comparable to that ~ obtained in 20 mM KCl and was significantly reduced relative to that obtained in 0 mM
(NH~),S04. The pattern of cleavage- obtained using 5 mM (NHQ),504, however, was identical to that observed when the 157 by substrate was incubated in the absence of (NH~),SO~ or in KCl or NaCI, indicating that the choice of salt included in the CleavaseT"' reaction has no effect on the nature of the sites recognized by the enzyme.
E) Time Course Of The Double-Stranded Cleavage Reaction . To determine how quickly the double-stranded cleavage reaction is completed, a single substrate was incubated in the presence of a fixed amount of CleavaseTM BN
enzyme for various lengths of time. Approximately 100 fmoles of the double-stranded 157 by fragment of exon 4 of the tyrosinase gene (SEQ ID NO 40; prepared as described above in Example 22a) were placed in sterile distilled water in 200 ~.l thin walled centrifuge tubes (BioRad, Richmond, CA) in a volume of 6.25 p.l. The tubes were heated to 95°C
for 15 seconds, as described in section b), and cooled to 45°C.
Cleavage reactions were started by the addition of 3.'75 ~.1 of an enzyme mix containing 2.7 X CFLPTM buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM
MnCI, (to yield a final concentration of 0.2 mM MnCI,), 0.5~,l CleavaseT"' BN
[50 ng/yl in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM
KC1. 10 ~.g/ml BSA)]. The final reaction volume was 10 ~l. No enzyme (i.e., uncut]
controls were set up in parallel and stopped after either 5 minutes or 120 minutes, with the difference that sterile distilled water was substituted for the CleavaseT"' BN enzyme.
The cleavage reactions were stopped by the addition of 8 ~,l of stop buffer at the following times: 5 seconds, l, 2, 5, 10, 15, 20, 30, 60 or 120 minutes.
Samples were heated to 72°C for 2 minutes and 4 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in. a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a with the exception that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 57.
In Figure 57, lane 1 contains the no enzyme control after a ~ minute incubation at 45°C and shows the mobility of the uncleaved template DNA. Lanes 2-10 contain cleavage fragments derived from reactions incubated in the presence of the CleavaseT"' BN enzyme for 305 sec, l, 2, 5, 10, 15, 20, 30, 60 (i hr), or 120 minutes (2 hr), respectively. Lane 11 contains the no enzyme control after a 120 minute incubation at 45°C.
Figure 57 shows that the cleavage of a double-stranded DNA template mediated by the CleavaseT"'' BN enzyme was rapid. A full cleavage pattern was apparent and essentially complete within one minute. Unlike the example of cleavage of a single-stranded DNA
template (Example 11 c), very little cleavage is detectable after 5 seconds.
This reaction contained one-tenth the amount of enzyme used in the reaction described in Example 11 c. In addition, whereas incubation of single-stranded cleavage reactions for extended periods "
generated a pattern of increasingly truncated fragments (Example 20e), extended incubation of , the double-stranded cleavage reaction resulted in a complete loss of signal (Figure 57, lane 10); this result is probably due to nibbling by the enzyme of the 5' biotin moiety from the reannealed strands. It is important to note that these results show that the same pattern of cleavage was produced for cleavage of double-stranded DNA, as for single-stranded, whether the reaction is run for 1 or 30 minutes. That is, the full representation of the cleavage products (i.e., bands) is seen over a 30-fold difference in time of incubation; thus the double-stranded CFLP~ reaction need not be strictly controlled in terms of incubation time.
The results shown in Figure 58 contain short time courses of cleavage reactions performed at a variety of enzyme concentrations. Approximately 100 fmoles of the double stranded 157 by fragment of exon 4 of the tyrosinase gene (SEQ ID N0:27) were placed in sterile distilled water in 200 p.l thin walled centrifuge tubes (BioRad, Richmond, CA) in a volume of 6.25 p.l. The tubes were heated to 95°C for 15 seconds, as described in Example 22b, and cooled to 45°C. Cleavage reactions were started by the addition of 3.75 yl of an enzyme mix containing 2.7X CFLP~ buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM MnCI, (to yield a final concentration of 0:2 mM MnCh), 0.5 pl CleavaseT"'' BN [at either 50, 100, 200, or S00-rig/p.l in 1 X dilution buffer (0.5% NP40, 0.5%
Tween 20, 20 mM
Tris-HCI, pH 8Ø 50 mM KCh 10 p.g/ml BSA) to yield a final amount of enzyme of 25, 50, 100, or 250 ng]. The final- reaction volume was 10 p,l. A no enzyme control was set up in parallel, with the difference that sterile distilled water was substituted for the CleavaseT"' BN
enzyme, and stopped after 1 minute at 45°C.
The cleavage reactions were stopped by the addition of 8 ~1 of stop buffer after either 5 seconds or 1 minute. Samples were heated to 72°C for 2 minutes and 4 ~l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the ' membrane was dried, washed in 1 X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 58.
In Figure 58, lane "M" contains molecular weight markers as described in Example 8a.
Lane 1 contains the no enzyme control. Lanes 2 and 3 each contain reaction products S generated by incubation of the substrate in the presence of 2 5 ng of the CleavaseT"' BN
enzyme; the reaction in lane 2 was stopped after 5 seconds, that in lane 3, after 1 minute.
Lanes 4 and 5 contain reaction products generated by cleavage of the substrate in the presence of 50 ng of the CleavaseT"'' BN enzyme; the reaction in lane 4 was stopped after 5 seconds, that in lane 5, after 1 minute. Lanes 6 and 7 contain reactian products generated by cleavage of the substrate in the presence of 100 ng of Cleavase~'~"'' BN enzyme; the reaction in lane 6 was stopped after 5 seconds, that in lane 7, after 1 minute. The reactions shown in lanes 8 and 9 each contain 250 ng of the Cleavase~'~'' BN enzyme; that in lane 8 was stopped after 5 seconds, that in lane 9, after 1 minute.
The results presented in Figure 58 indicate that the rate of cleavage of double-stranded DNA increased with increasing enzyme concentration. Note that as the concentration of enzyme was increased, there was a corresponding reduction in the amount of uncut DNA that remained after 1 minute. As was demonstrated below, in Figure 60, the concentration of enzyme included in the cleavage reaction had no effect on the cleavage pattern generated.
Comparison of the 250 ng reaction (shown in Figure 58, lanes 8 and 9) to the short time point digestion described in Example 11 c, indicates that the amount of enzyme rather than the double-stranded or single-stranded nature of the substrate controls the extent of cleavage in the very early time points.
F) Temperature Titration Of The Double-Stranded Cleavage Reaction To determine the effect of temperature variation on the cleavage pattern, the 157 by fragment of exon 4 of the tyrosinase gene (SEQ ID N0:27) was incubated in the presence of a fixed amount of CleavaseT"'' BN enzyme for 5 minutes at various temperatures.
Approximately 100 fmoles of substrate DNA (prepared as described in Example 22a) were placed in sterile distilled water in 200 p.l thin walled centrifuge tubes (BioRad, Richmond.
CA) in a volume of 6.25 p.l. The tubes were heated to 95°C for 15 seconds and cooled to either 37, 40, 45, 50, 55, 60, 65, 60, 75, or 80°C. _ Cleavage reactions were started by the addition of 3.75 p.l of an enzyme mix containing 2.7 X CFLPT"' buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM IVInCh (to yield a final concentration of 0.2 mM
MnCI,), 0.5 p.l Cleavase~ BN [50 ng/p.l in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, WO 96/15267 PG"T/US95/14673 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 ~.g/ml BSA)]. The enzyme mix was kept on ice throughout the duration of the experiment, but individual aliquots of the enzyme mix were allowed to come to room temperature before being added to the reactions. A
second reaction was run at 37°C at the end of the experiment to control for any loss of enzyme activity that may have occurred during the course of the experiment. No enzyme controls were set up in a parallel and incubated at either 37°C or 80°C, with the difference that sterile distilled water was substituted for the Cleavase'~'' BN. The reactions were stopped by the addition of 8 ~1 of stop buffer.
Samples were heated to 72°C for 2 minutes and 5 pl of each reaction were resolved by I 0 electrophoresis through a 10% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 59.
In Figure 59, the lane marked "M" contains molecular weight markers, prepared as described in Example 8a. Lane 1 contains the no enzyme control after a 5 minute incubation at 37°C. Lanes 2 and 3 contain reactions incubated at 37°C, run at the beginning and end of the experiment, respectively. Lanes 4-13 contain reactions incubated at 40, 45, 50, 55, 60, 65, 70, 75, or 80°C [there are two 80°C samples; the first was not covered with Chill Out 14T"'' (MJ Research, Watertown, MA), the second was overlaid with 20 ~,l Chill Out 14~''' after the addition of the enzyme mix], respectively. Lane 14 contains a no enzyme control incubated at 80°C for 5 minutes.
Figure 59 shows that cleavage of double-stranded DNA substrates proceeded most effectively at lower temperatures. The distribution of signal and pattern of cleavage changed smoothly in response to the temperature of incubation over the range of 37°C to 60°C. Some cleavage products were evident only upon incubation at higher temperatures, whereas others were far more predominant at lower temperatures. Presumably, certain structures that are substrates for the CleavaseT"'' BN enzyme at one end of the temperature range are not favored , at the other. As expected, the production of cleavage fragments became progressively less ' abundant in the high end of the temperature range as the cleavage structures were melted out.
Above 70°C, the cleavage products were restricted to small fragments, presumably due to extensive denaturation of the substrate. When longer DNAs (350 to 1000 nucleotides) are used, it has been found that useful patterns of cleavage are l;enerated up to 75°C.
These results show that the cleavage reaction can be performed over a fairly .wide range of temperatures using a double-stranded DNA substrate. As in the case of the single-r 5 stranded cleavage reaction, the ability to cleave double-stranded DNA over a range of temperatures is important. Strong secondary structures that may dominate the cleavage pattern are not likely to be destabilized by single-base changes and may therefore interfere with mutation detection. Elevated temperatures can then be used to bring these persistent structures to the brink of instability, so that the effects of small changes in sequence are maximized and revealed as alterations in the cleavage pattern. This also demonstrates that within the useful temperature range, small changes in the reaction temperature due to heating block drift or similar device variations will not cause radical changes in the cleavage pattern.
g) Titration Of The CleavaseT"' BN Enzyme In Double-Stranded Cleavage Reactions The effect of varying the concentration of the CleavaseTM BN enzyme in the double-stranded cleavage reaction was examined. Approximately 100 fmoles of the 157 by fragment of exon 4 of the tyrosinase gene (SEQ ID N0:27; prepared as described in Example 22a) were placed in sterile distilled water in 200 p.l thin walled centrifuge tubes (BioRad, Richmond, CA) in a total volume of 6.25 q.l. These tubes were heated to 95°C for 15 seconds and then rapidly cooled to 45°C.
Cleavage reactions were started immediately by the addition of 3.75 ~1 of a diluted enzyme mix containing 2.7 X CFLPT"'' buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM MnCI, (to yield a final concentration of 0.2 mM MnCh), 0.5 pl CleavaseT"' BN [2, 10, 20, 50, 100, 200, 500 ng/~l in 1 X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM
Tris-HCI, pH 8Ø 50 mM KCI, 10 ~,g/ml BSA) such that 1, 5, 10, 25, 50, 100 or 250 ng of enzyme was added to the reactions]. No enzyme controls were set up in parallel, with the difference that 1X dilution buffer was substituted for the CleavaseT"' BN.
After 5 minutes at 45 ° C, the reactions were stopped by the addition of 8 q.l of stop buffer. The samples were heated to 72°C for 2 minutes and 4 ~l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7 M urea, in a buffer containing 0.5X TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane as described in Example 8b. The DNA was tran sferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiograph is shown in Figure 60.
The lane marked "M" in Figure 60 contains molecular weight markers. Lane 1 contains the no enzyme control and shows the migration of the uncut substrate.
Lanes 2-8 contain cleavage products derived from reactions containing 1, 5, 10, 25, 50, 100 or 250 ng of the Cleavase~'' BN enzyme, respectively.
These results show that the same cleavage pattern was obtained using the 157 by tyrosinase DNA substrate (SEQ ID N0:27) regardless of whether the amount of enzyme used in the reaction varied over a 50-fold range. Thus, the double-stranded cleavage reaction is ideally suited for practice in clinical laboratories where reaction conditions are not as controlled as in research laboratories. Note, however, that there is a distinct optimum for cleavage at intermediate enzyme concentrations for a double-stranded template, in marked contrast to what was observed on single-stranded substrates (Example 11 e).
The progressive loss of signal in the double-stranded reactions at increasing concentrations of CleavaseTM BN
is likely due to the nibbling of the 5' biotin label off the end of the reannealed double-stranded template. -E~MPLE 23 Determination Of The pH Optimum For Single Stranded And Double-Stranded Cleavage Reactions In order to establish optimal pH conditions for the two types of primer-independent cleavage reactions (i. e., single-stranded and double-stranded cleavage reactions), the CleavaseT"'' reaction buffer was prepared at a range of different pHs.
A) Establishment Of A pH Optimum For The Single-Stranded Cleavage Reaction The effect of varying the pH of the CleavaseT"'' reaction (i.e., CFLPTM) buffer upon the cleavage of single-stranded substrates was examined. Several 10 X buffer solutions were made with 0.5 M MOPS at pH 6.3, 7.2, 7.5, 7.8, 8.0 and 8.2 by titrating a 1 M
solution of MOPS at pH 6.3 with 6 N NaOH. The volume was then adjusted to yield a 0.5 M
solution at each pH.

Approximately 100 fmoles of a single-stranded substrate prepared from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34; prepared as described in Example 8a), were placed in 200 ~l thin walled centrifuge tubes (BioRad, Richmond, CA) in 15 pl of 1 X CFLPT"' buffer, at varying pH, and 1.33 mM MnCh (to yield a final concentration of 1 mM). The final reaction volume was 20 ~.1. The reaction mixes were heated to 95°C for 5 seconds and rapidly cooled to 65°C. The reactions were started by the addition of 5 ~,1 of diluted enzyme mix containing 1 pl of Cleavase"''' BN [50 ng/~.l in 1 X
dilution buffer (0.5%
NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 p.g/ml BSA)] in 1 X
CFLPT"' buffer (without MnCh), again at the appropriate pH. A 20 p.l no salt, no enzyme control was set up in parallel and incubated at 65°C for each of the indicated pHs, with the difference that sterile distilled water was substituted for CleavaseT"'' BN
and all reaction components were added prior to denaturation. Reactions were stopped by the addition of 16 p.l of stop buffer after 5 minutes.
Samples were heated to 72°C for 2 minutes and 7 ~,l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7M
urea, in a buffer containing O.SX TBE, as described in Example 8a.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical), washed, reacted with CDP-StarT"'' (Tropix, Bedford, MA), and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The results are presented in Figure 61.
In Figure 61, panels A and B contain reactions which used single-stranded DNA
substrates. In panel A, 5 pairs of reactions are presented. In each case, the first lane of the pair is the no enzyme control and the second is the single-stranded cleavage reaction. Lanes 1 and 2 depict reaction products obtained using a reaction buffer at pH 6.3;
lanes 3 and 4, at pH
7.2; lanes 5 and 6, pH 7.8; lanes 7 and 8, pH 8.0; lanes 9 and 10, at pH 8.2.
Panel B
contains the results of a separate experiment comparing cleavage reactions performed using a reaction buffer at pH 7.5 (lanes 1 and 2, uncut and cut, respectively) and at pH 7.8 (lanes 3 and 4, uncut and cut, respectively).
°
The results shown in Figure 61, panels A and B, indicate that the cleavage of the single-stranded DNA template was sensitive to relatively small changes in pH.
There v~~as a pH optimum for the reaction between pH 7.5 and 8Ø Because the pK~ of MOPS is 7.2. the WO 96/15267 PC"T/US95/14673 pH closest to that value which supported cleavage, pH 7.5, was determined to be optimal for the single-stranded cleavage-reaction.
B) Establishment Of A pH Optimum For The Double-Stranded Cleavage Reaction The effect of varying the pH of the CleavaseT"'' reaction (i. e., CFLPT"'') buffer upon the , cleavage of double-stranded substrates was examined. Several 10 X buffer solutions were made with 0.5 M MOPS at pH 7.2, 7.5, 7.8, and 8.0, as described above in section a).
Approximately 100 fmoles of the double-stranded 157 by fragment of exon 4 of the tyrosinase gene (SEQ ID N0:27; prepared as described in Example 8) were placed in 200 pl thin walled centrifuge tubes (BioRad, Richmond, CA) in a total volume of 6.25 p.l. The tubes were heated to 95°C for 15 seconds and cooled to 45°C. The clevage reactions were started by the addition of 3.75 ~I of diluted enzyme mix containing 2.7 X CFLP~'~"'' buffer, pH 7.5 (to yield a final concentration of 1 X), 0.53 mM MnCh (to yield a final concentration of 0.2 mM
MnCI,), 0.5 ~1 of CleavaseT"'' BN [50 ng/p.l in 1 X dilution buffer (0.5%
NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 p.g/ml BSA)].
The cleavage reactions were incubated for 5 minutes and then were terminated by the addition of 8 p,l of stop buffer.
Samples were heated to 72°C for 2 minutes and 4 p.l of each reaction were resolved by electrophoresis through a 10% polyacrylamide gel (19:1 cross-link), with 7M
urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8b. The DNA was transferred to the membrane and the membrane was dried, washed in 1X I-Block Blocking Buffer, washed and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The resulting autoradiographs are shown in Figure 62, panels A and B.
In Figure 62, panel A, lanes 1 and 2 contain cleavage products from reactions run in a buffer at pH 8.2 (lane 1 contains the cleavage reaction; lane 2 is the uncut control). Lanes 3 and 4 contain cleavage products from reactions run in a buffer at pH 7.2 (lane 3 contains the cleavage reaction; lane 4 is the uncut control). In panel B, lanes 1 and 2 contain cleavage products from reactions run in a buffer at pH 7.5 (lane 1 is the uncut control; lane 2 contains the cleavage reaction). Lanes 3 and 4 contain cleavage products from reactions run in a buffer at pH 7.8 (lane 3 contains the uncut control; lane 4 contains the cleavage reaction).
-172_ . The results in Figure 62, panels A and B, demonstrate that the cleavage of double-stranded DNA was not sensitive to changes in pH over the range of buffer conditions tested.
Because the cleavage of single-stranded DNA, however, was sensitive to changes in~ pH, the buffer conditions that were determined to be optimal for the single-stranded cleavage reaction a 5 were chosen for subsequent double-stranded cleavage experiments.

The Presence Of Competitor DNA Does Not Alter The Cleavage Pattern The effect of the presence of competitor (i. e., non-labelled substrate) DNA
upon the cleavage reaction was examined. The cleavage reaction was run using the 157 nucleotide fragment from the sense strand of the human tyrosinase gent: (SEQ ID N0:34) and human genomic DNA. The results shown below demonstrate that the presence of non-substrate DNA
has no effect on the CFLPT"'' pattern obtained in the cleavage reaction.
A) Preparation Of The Substrate DNA And The Cleavage Reactions The 157 nucleotide single-stranded wild type tyrosinase substrate (SEQ ID
N0:34) containing a biotin label on the 5' end was prepared as described in Example 9. Human genomic DNA (Promega) present at 235 p.g/ml in Tris-HCI, pH 8.0; I mM EDTA was ethanol precipitated and resuspended in Tris-HCI, pH 8.0; 0.1 mM EDTA to final concentration 400 ~g/ml. This DNA was used as a competitor in standard CFLPTM
single-stranded reactions (described in Example 9). Tyrosinase DNA substrate (SEQ ID
N0:34) and human genomic DNA were mixed in H,O in final volume of 6 p,l. Samples were heated at 95°C for 10 seconds to denature the DNA, cooled to the target temperature of 65°C, and mixture of 2 ~,I SX CFLPT"' buffer, pH 7.5, 1 ~1 10 mM MnCh and 1 p,l (2~ ng) the enzyme Cleavase~'~"'' BN in dilution buffer was added. After 5 minutes at 65°C, 6 ~.1 of stop buffer was added to terminate reaction and 5 pl of each sample was separated on a 10%
denaturing polyacrylamide gel. Membrane transfer and DNA visualization were performed as described in Example 19.
B) The Presence Of Genomic DNA Does Not Alter The CFLPT"' Pattern Figure 63 shows the resulting pattern corresponding to the cleavage products of the sense strand of the wild type tyrosinase substrate (SEQ ID N0:34) in the presence of 0 yg/ml (lane 2), 20 pg/ml (lane 3), 40 p.g/ml (lane 4), 80 ~g/ml (lane 5), 120 ~,g/ml (lane 6) and 200 q.g/ml (lane 7) unlabeled human genomic DNA. Lane 1 shows an uncut control in the absence of the enzyme Cleavase~ BN and lane marked "M" contains the molecular weight markers prepared as described in Example 8.
Figure 63 shows that the presence of genomic DNA in the cleavage reaction did not change either the position or the relative intensity of the product bands produced. Increasing the amount of nonspecific DNA in the reaction did, however, decrease the efficiency of the cleavage reaction and reduced the overall intensity of the pattern. These results can be explained by the binding of the Cleavase~'~"' BN enzyme to the nonspecific DNA
which has the effect of decreasing the effective enzyme concentration in the reaction. This effect became significant when the concentration of genomic DNA in the reaction was equal to or greater than 120 ~g/ml jFigure 63 , lanes 6 (120 p.g/ml) and 7 (200 ~.g/ml)]. Under these conditions, the genomic DNA was present at more than a 20,000-fold-excess relative to the specific substrate DNA; nonetheless the CFLP~ pattern could still be recognized under these conditions. The observed stability of the CFLPT"' pattern in the presence of genomic DNA
1 S ruled out the possibility that nonspecific DNA could significantly change the structure of the substrate DNA or alter the interaction of the Cleavase~'~"' BN enzyme with the substrate.

The CFLP~ Reaction Can Be Practiced Using A Variety of Enzymes The above Examples demonstrated the ability of the CleavaseTM BN enzyme, a 5' nuclease derived from Taq DNA polymerase, to generate a characteristic set of cleavage fragments from a nucleic acid substrate. The following experiments demonstrate that a number of other enzymes can be used to generate a set of cleavage products which are -characteristic of a given nucleic acid. These enzymes are not limited to the class of enzymes characterized as 5' nucleases.
A) Cleavage Patterns Generated by Other DNA Polymerases From The Genus Tl:ermus To determine whether ~' nuclease activity associated with DNA polymerases (DNAPs) ' other than Tad DNAP could generate a distinct cleavage pattern from_ nucleic acid substrates, DNAPs from two species-of Thermus were examined. The DNAP of Thermus,flavzr.s ["Tfl", WO 96/15267 PCT/US95t14673 Kaledin et al., Biokhimiya 46:1576 (1981); obtained from Promega Corp., Madison, WI] and the DNAP of Thermus thermophilus ["Tth", Carballeira et al., Biotechniques 9:276 (1990);
Myers et al., Biochem. 30:7661 (1991); obtained from U.S. Biochemicals, Cleveland, OH]
were examined for their ability to generate suitable cleavage patterns (i.e., patterns which can . 5 be used to characterize a given nucleic acid substrate).
The ability of these other enzymes to cleave nucleic acids in a structure-specific manner was tested using the single-stranded 157 nucleotide fragment of the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34) under conditions reported to be optimal for the synthesis of DNA by each enzyme.
Approximately 100 fmoles of the 157 nucleotide fragment derived from the sense strand of exon 4 of the tyrosinase gene (SEQ ID NO 47; prepared as described in example l0a) were placed in 200 p,l thin wall microcentrifuge tubes (BioRad, Richmond, CA) in 1 X
CFLPT"' buffer, pH 8.2 and 1.33 mM MnCh (to yield a final concentration of 1 mM) and KCl to yield a final concentration of either 0 or 50 mM. Final reaction volumes were 20 ~.1.
Samples were heated to 95°C for 5 seconds and then cooled to 65°C. A 20 p.l no salt, no enzyme control was set up in parallel, with the differences that sterile distilled water was substituted for the CleavaseT"'' BN enzyme and all reaction components were added prior to denaturation at 95 ° C.
The cleavage reactions were started by the addition of 5 pl of a diluted enzyme mix containing either 1.25 units or 5 units of the indicated enzyme (see below) in buffer, pH 8.2. After 5 minutes, reactions were stopped by the addition of 16 p,l of stop buffer.
Samples were heated to 72°C for 2 minutes and 7 ~l (in the case of the samples digested with Tfl) or 5 p.l (in the case of the samples digested with Tth) were electrophoresed through a 10% polyacrylamide gel (19:1 cross-link), with 71VI urea, in a buffer containing O.SX TBE, as described in Example 8a.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in Example 8a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical, Cleveland, OH), washed. reacted with CDP-StarTT'1 (Tropix, Bedford, MA), and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The results are presented in Figures 64 and 65.
-17~-WO 96/15267 PG"T/US95/14673 In Figure 64, lane 1 contains the no enzyme control and indicates the migration of the uncut DNA. Lanes 2-5 contain cleavage products derived from reactions incubated with Tfl DNAP: The reactions represented in lane 2 and 3 each contained 5 units of Tfl DNAP; the sample in lane 2 was incubated in a reaction buffer containing 0 mM KCI, while the sample in lane 3 was incubated in a reaction buffer containing 50 mM KCI. The reactions in lanes 4 and 5 each contained 1.25 units of T,fl DNAP; the sample in lane 4 was incubated in a reaction buffer containing 0 mM KCI; that in lane 5 was incubated in a reaction buffer containing 50 mM KCI.
In Figure 65, lanes 1 and 2 each contain cleavage products derived from reactions incubated with 1.25 units of Tth DNAP. The sample in lane 1 was incubated in a reaction buffer containing 0 mM KCI; that in lane 2 was incubated in a reaction buffer containing 50 mM KCI. Lanes 3 and 4 contain cleavage products derived from reactions incubated with S
units of Ttla DNAP. The sample shown in lane 3 was incubated in a reaction buffer containing 0 mM KCI; that in lane 4 was incubated in a reaction buffer containing 50 mM
KCI.
Figures 64 and 65 demonstrates that both Tth DNAP and Tfl DNAP display structure specific endonuclease activity similar in nature to that seen in the Cleavase~"' BN enzyme. A
comparison of the results shown in Figures 64 and 65 showed that the Tth DNAP
was more efficient at generating a cleavage pattern under the reaction conditions tested. Comparison of the cleavage patterns generated by Tth DNAP with those generated by the Cleavase'~'~' BN
enzyme the indicates that essentially the same structures are recognized by these two enzymes [compare Figure 66, lane 2 (Cleavase"~'' BN) with Figure 65 (Tth DNAP)].
B) Enzymes Characterized As 3' Nucleases Can be Used To Generate Distinct Cleavage Patterns To determine whether enzymes possessing 3' nucleolytic activity could also generate a distinct cleavage pattern, enzymes other than DNAPs (which possess 5' nuclease activity) were tested in the cleavage reaction. Exonuclease III from Escher-ichia coli (E. coli Exo III) was tested in a cleavage reaction using the 157 nucleotide fragment prepared from the sense strand of exon 4 of the tyrosinase gen (SEQ ID N0:34). As a comparison, a reaction containing this substrate (SEQ ID N0:34) and the CleavaseT"' BN enzyme was also prepared.
Approximately 100 fmoles of the 157 nucleotide fragment prepared from the sense strand of exon 4 of the tyrosinase gene (SEQ ID N0:34; prepared as described in Example 8a) were placed in 200 p.l thin wall microcentrifuge tubes (BioRad, Richmond, CA) in 1 X

CFLP~'~'' buffer, pH 8.2 and 1.33 mM MnCl2 (to yield a final concentration of 1 mM) and KCl to yield a final concentration of either 0 or 50 mM in a volume of 15 ul.
Final reaction volumes were 20 ~.1.
The samples were heated to 95°C for 5 seconds and then rapidly cooled to 37°C. A 20 p.l no salt, no enzyme control was set up in parallel, with the differences that sterile distilled water was substituted for the Cleavase'~"'' BN enzyme and all reaction components were added prior to denaturation at 95°C.
A reaction tube containing 100 fmoles of the 157 nucleotide fragment (SEQ ID
N0:34) and 50 ng of the Cleavase~'~"'' BN enzyme in a buffer containing 0 mM
KCl was prepared and treated as described in Example 21 (i.e., denatured by incubation at 95°C for 5 seconds followed by cooling to 65°C and the addition of the; enzyme and incubation at 65°C
for 5 minutes).
The cleavage reactions were started by the addition of 5 ~l of a diluted enzyme mix containing either 1.25 units or 200 units of Exo III (United States Biochemical. Cleveland, OH) in 1 X CFLP~'~"'' buffer, pH 8.2 (without MnCl2) were added to the 15 ~l reactions, and the reactions were incubated for 5 minutes. After 5 minutes at 37°C, the reactions were stopped by the addition of 16 ~,l of stop buffer.
The samples were heated to 72°C for 2 minutes and 5 ~.I were electrophoresed through a 10% polyacrylamide gel (19:1 cross-link), with 7M urea, in a buffer containing 0.5X TBE, as described in Example 8a.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in example 10a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical, Cleveland, OH), washed, reacted with CDP-Star'"'' (Tropix, Bedford, MA), and exposed to X-ray film as described in Example 20a, except that the distilled water washes were omitted. The results are presented in Figure Lane 1 in Figure 66 contains the no enzyme control and indicates the mobility of the uncut DNA. Lane 2 contains cleavage fragments generated by incubation of the substrate with the CleavaseT"'' BN enzyme and provides a comparison of the patterns generated by the two different enzymes. Lanes 3-6 contain cleavage fragments generated by incubation of the substrate with Exo III. Lanes 3 and 4 each contain reaction products generated in reactions which contained 200 units of Exo III; the reaction in lane 3 was run in a buffer containing 0 mM KCI, that in lane 4 was run in a buffer containing 50 mM KCI. Lanes 5 and 6 each contain reaction products generated in reactions which contained 1-.25 units of Exo III: the reaction in lane 5 was run in a buffer containing 0 mM KCI, that in lane 6 was run in a buffer containing 50 mM KCI.
The results presented in Figure 66 demonstrate that Exo III generated a distinct cleavage pattern when incubated with a single-stranded DNA substrate. The pattern generated by Exo III was entirely distinct from that generated by the CleavaseT"'' BN
enzyme. The results shown in Figure 66 also show that significant differences in the cleavage pattern generated by Exo III were observed depending on the concentrations of both the enzyme and KCl included in the reactions. -C) Ability Of Alternative Enzymes To Identify Single Base Changes In sections and a) and b) above it was shown that enzymes other than the Cleavase~'~"' BN enzyme could generate a distinct pattern of cleavage fragments when incubated in the presence of a nucleic acid substrate. Because both Tth DNAP and E. coli Exo III generated distinct cleavage patterns on single-stranded DNA, the ability of these enzymes to detect single base changes present in DNA substrates of the same size was examined.
As in Example 9, the human tyrosinase gene was chosen as a model system because numerous single point mutations have been identified in exon 4 of this gene.
Three single-stranded substrate DNAs were prepared; all three substrates contained a biotin label at their 5' end. The wild type substrate comprises the 157 nucleotide fragment from the sense strand of the human tyrosinase gene (SEQ ID N0:34). Two mutation-containing substrates were used. The 419 substrate (SEQ ID N0:41 ) and the 422 substrate (SEQ ID N0:42), both of which are described in Example 9. Single-stranded DNA
containing a biotin label at the 5' end was generated for each substrate using asymmetric PCR
as described in Example 8a with the exception that the single-stranded PCR
products were recovered from the gel rather than the double-stranded products.
Cleavage reactions were performed as follows. Each substrate DNA
(approximately 100 fmoles) was placed in a 200 ~.l thin wall microcentrifuge tubes (BioRad.
Richmond, CA) in 5 ~1 of 10 mM MOPS, pH 8.2, with 1.33 mM MnCI, (to yield a final concentration of 1 mM). A no enzyme control was set up with the wild type DNA fragment in parallel and incubated at 65°C for each of the indicated time points, with the differences that sterile distilled water was substituted for the CleavaseT"'' BN enzyme and all reaction components were added prior to denaturation at 95°C. The reaction tubes were brought to 95°C for ~

seconds to denature the substrates and then the tubes were ~u~Ckl~ cZSoZed fo ~5°C~or the reactions containing Tth DNAP and 37°C for the reactions containing Exo III.
Cleavage reactions were started immediately by the addition of a diluted enzyme mixture containing 1.25 units of the enzyme either Tth DNAP or Exo III in 5 p,l of 10 mM
MOPS, pZI 8.2 without MnCh. The enzyme solution was brought to room temperature before addition to the cleavage reaction. After 5 minutes at 65°C, the reactions were stopped by the addition of 8 p.l of stop buffer. The samples were heated to 72°C for 2 minutes and 7 ~l of each reaction were resolved by electrophoresis through a 10~% polyacrylamide gel ( 19:1 cross-link), with 7 M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated and overlaid with a nylon membrane, as described in example 10a. The DNA was transferred to the membrane and the membrane was dried, blocked in 1 X I-Block (Tropix, Bedford, MA), conjugated with streptavidin-alkaline phosphatase (United States Biochemical), washed, reacted with CDP-Star'~"' (Tropix, Bedford, MA), and exposed to X-ray film as described in Example 20a with the exception that the distilled water washes were omitted. 'The results are presented in Figure 67. _ In Figure 67, lanes 1-3 contain cleavage fragments generated by incubation of either the wild-type, mutant 419 and mutant 422 alleles of the tyrosinase gene, respectively, with Tth DNAP. Lanes 4-6 contain cleavage fragments generated by incubation of either the wild type, mutant 419 and mutant 422 substrates, respectively, with Exo III in a buffer containing 0 mM KCI. Lanes 7-9 contain cleavage fragments generates( by incubation of either the wild type, mutant 419 and mutant 422 substrates, respectively, in<;ubated with Exo III in a buffer containing SO mM KCI. Lane 10 contains cleavage fragments generated by incubation of the wild type DNA substrate with the Cleavase~'~"'' BN enzyme in a buffer containing 0 mM KCI:
this reaction provides a comparison of the patterns generated by the three different enzymes (i. e.. the CleavaseT"' BN enzyme, -Tth DNAP and Exo III). Lane 11 contains the no enzyme control with the wild type DNA substrate incubated in the presence of 50 mM
KC1.
The results shown in Figure 67 demonstrate that both Tth DNAP and Exo III were able to detect single base changes in a single-stranded DNA substrate relative to a wild-type DNA substrate. The patterns generated by Tth DNAP were comparable to those generated by the CleavaseT~"'' BN enzyme for all three DNA substrates (See Figure 29 for a comparison of the pattern generated by the CleavaseT"' BN enzyme).

WO 96115267 PCTlUS95114673 The patterns generated by Exo III v~~ere entirely distinct from those generated by enzymes derived from the genus Thermus (i.e., the CleavaseT~"' BN enzyme and Tth DNAP).
Furthermore, the pattern produced by cleavage of the DNA substrates by Exo III
were distinct depending on which concentration of KCl was employed in the reaction (Figure 67). A -distinct pattern change was evident for the 419 mutant at both KCI
concentrations. As shown in Figure 67, at 0 mM KCI, a band appears in_the 40 nucleotide range in the 419 mutant (lane 5); at 50 mM KCI, the 419 mutant contains an additional band in the 70 nucleotide range (lane 8). Pattern changes were not discernable for the 422 mutant (relative to the wild-type) in the Exo III digestions; this difference in the ability of the E. coli Exo III enzyme to detect single base changes could relate to the relative positions- of the changes with respect to secondary structures that act as substrates for the structure specific cleavage reaction, and the position of the label (5' or 3' end) relative to the preferred cleavage site (5' or 3'), Figure 68.
D) The Drosophila RrpI Enzyme Can Be Used to Generate Cleavage Patterns Another protein in the Exo III family of DNA repair endonucleases, RrpI from Drosophila melanogaster (Nugent, M, Huang, S.-M., and Sander, M.
Biochemistyy~, 199 3: 32, pp. 11445-11452), was tested for its ability to generate a distinct cleavage pattern on a single-stranded DNA template. Because its characteristics in the cleavage assay were unknown, this enzyme was tested under a variety of buffer conditions. Varying amounts of this enzyme (1 ng or 30 ng) were incubated with approximately 100 fmoles of the 157 nucleotide fragment of the sense strand of exon 4 of the tyrosinase gene (SEQ ID NO: 47) in either 1 mM MnC 1 or 5 mM MgC1? and either 1 X CFLPT"'' buffer, pH 8.2 or 1 X CFLP~'~"' buffer, pH 7.8, with 10 mM NaCI. Samples were heated to 95°C and begun by the addition of a diluted enzyme mix containing either 1 or 30 ng of RrpI in 1 X CFLPT"'' buffer. Reactions were carried out at 30°C for either 5 or 30 minutes. The results (data not shown) indicated that this enzyme 2~ generates a weak, but distinct cleavage pattern on a single-stranded DNA
template.
E) The Radl/RadlO Complex Can Be Used To Generate Cleavage Patterns The Radl-RadlO endonuclease (Radl/10) from S. cerevisiae is a specific 3"
endonuclease which participates in nucleotide excision repair in yeast. This enzyme is a heterodimer consisting of two proteins, Radl and RadlO. Radl and RadlO alone do not have enzymatic activity. Radl/10 recognizes structures comprising a bifurcated DNA
duplex and cleaves the single-stranded 3' arm at the end of the duplex [Bardwell, A.J et al. ( 1994) Science 265:2082]. In this sense Radl/10 shares the same substrate specificity as does the CleavaseT"'' BN enzyme. However, the cleavage products produced by Radl/10 and the CleavaseT"' BN enzyme differ as the Radl/10 cleaves on the 3' single-stranded arm of the duplex while the CleavaseT"'' BN enzyme cuts on the 5' single-stranded arm.
Figure 68 provides a schematic drawing depicting the site of cleavage by these two enzymes on a bifurcated DNA duplex (formed by the hairpin structure shown). In Figure 68, . 5 the hairpin structure at the top shows the site of cleavage by a 5' nuclease (e.g., the enzyme Cleavase'~"'' BN enzyme). The hairpin structure shown at the bottom of Figure 68 shows the site of cleavage by an enzyme which cleaves at the 3' single-stranded arm (e.g., Radl/10).
Enzymes which cleave on the 5' single-stranded arm are referred to as CleavaseTM 5' enzymes; enzymes which cleave on the 3' single-stranded arm are referred to as CleavaseTM 3' enzymes.
In order to determine whether the Radl/10 protein is able to detect single base changes in DNA substrates, the cleavage patterns created by cleavage of DNA substrates by the Radl/10 and CleavaseT"'' BN enzymes were compared. In this comparison the following substrates were used. The 157 nucleotide fragment from the wild type (SEQ ID
N0:34). the 419 mutant (SEQ ID N0:41 ) and the 422 mutant (SEQ ID N0:42) .alleles derived from the sense strand of exon 4 of the human tyrosinase gene was generated containing a biotin label at the 5' end as described in Example 9.
The Radl and RadlO proteins were generously provided by Dr. Errol C. Friedberg (The University of Texas Southwestern Medical Center, Dallas). The Radl/10 complex was prepared by mixing Radl and RadlO proteins in 1X dilution buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-HCI, pH 8.0, 50 mM KCI, 10 pg/ml BSA) to achieve a final concentration of 0.1 mM of each protein.
Cleavage reactions using the Radl/10 endonuclease were performed as follows.
The substrate DNA and 15 ng (0.1 pmole) of Radl/10 complex in 1 ~.1 of 1X dilution buffer were mixed on ice in 10 ~.l of 10 mM MOPS, pH 7.8, 1 mM MnCI,. The reaction was then incubated at 37°C for 5 minutes. The cleavage reaction was stopped by addition of G ~.1 of stop buffer.
Cleavage reactions using the CleavaseT"'' BN enzyme were done exactly as described above for the Radl/10 cleavages with the exception that 10 ng of the CleavaseT"' BN enzyme was added and the incubation at 37°C was performed for 3 minutes. Uncut or no enzyme controls were run for each substrate DNA and were prepared as described for the reactions containing enzyme with the exception that sterile water was added in place of the enzyme (data not shown).

WO 96/15267 PG"TlUS95/14673 The cleavage products (3 ~.1 each) were separated by electrophoresis through a 10%
denaturing polyacrylamide gel, transferred to a membrane and visualized as described in Example 19. The resulting autoradiograph is shown in Figure 69.
Figure 69 shows the resulting patterns corresponding to the cleavage products of the sense strand of the wild type tyrosinase substrate (SEQ ID N0:34) (lanes 1 and 4), the 419 mutant (SEQ ID N0:41) (lanes 2 and 5) and the 422 mutant (SEQ ID N0:42) (lanes 3 and 6). Lanes 1-3 show the cleavage pattern created by incubation of the three substrate DNAs with the Cleavase~ BN enzyme and lanes 4-6 show cleavage patterns created by incubation of the three substrate DNAs with the Radl/10 enzyme. Lanes marked "M" contain molecular weight markers prepared as described in Example 8.
The results shown in Figure 69 demonstrate that the Radl/10 enzyme was able to produce distinctive cleavage patterns from the substrate DNAs (lanes 4-6); the average product length produced by cleavage of the substrate was longer than that produced by the CleavaseT'~' BN enzyme. Importantly, the results shown in Figure 69 demonstrate that the single base substitutions found in the mutant tyrosinase substrates resulted in the production of specific changes in the otherwise similar cleavage patterns of tyrosinase substrates (compare lanes 5 and 6 with lane 4). Note that in the digestion of the mutant 419 substrate with Radl/10, the bands below about 40 nucleotides have lower intensity and one band is absent, when compared to wild-type, while in the digest of the mutant 422 substrate several new bands appear in the range of 42-80 nucleotides. . Since both enzymes were tested using the same reaction conditions, these results show that Radl/10 was able to detect the same differences in DNA secondary structure that were recognized by the CleavaseT"' BN enzyme.
Radl/10 generates a different cleavage pattern relative to that produced by the CleavaseT"' BN
enzyme, since cleavage takes place at the 3' end of DNA hairpins producing inherently longer fragments when the substrate contains- a 5' end label. _ .

Detection Of Mutations In The Human (3-globin Gene Using Double-Stranded DNA Substrates The results shown in Example 13 demonstrated that single base changes in fragments of the (3-globin gene can be detected by cleavage of single-stranded DNA
substrates with the .

WO 96/15267 PC"T/US95/14673 CleavaseT"' BN enzyme. In this example it is shown that rrautations in the (3-globin gene can be detected by cleavage of double-stranded DNA substrates using the CleavaseT"' BN enzyme.
Double-stranded substrate DNA comprising 536 by fragments derived from the wild-' type (3-globin gene (SEQ ID N0:56), mutant 1 (SEQ ID N0:58) and mutant .2 (SEQ ID
N0:59) were generated containing a 5' biotin label on the sense strand using the PCR. PCR
amplification of these substrates was done as described in Example 13a. Gel purification and isolation of double-stranded fragments was performed as described in Example 19a.
The cleavage reactions were performed as described in Example 19c. Briefly, 2 ~l of stock DNA (80 ng) in TE was mixed with 3 ~,l H,O and denatured at 95°C
for 20 seconds.
The denatured DNA was cooled to 70°C and a mixture consisting of 2 ~.1 of SX CFLP~'~"'' buffer pH 7.5, 2 ~.l of 2 mM MnCI, and 1 pl (25 ng) of the enzyme CleavaseT"' BN in dilution buffer was added to start the cleavage reaction. The cleavage reactions were stopped after 1 minute by the addition of 6 ~,l of stop buffer. Control uncut reactions were performed as described above with the exception that of 1 ~.l of HBO vras used in place of 1 ~l of the 1 S CleavaseT"'' BN enzyme. The cleavage products (5 ~.1 each) were separated by electrophoresis through a 6% denaturing polyacrylamide gel, transferred to a membrane and visualized as described in Example 19. The resulting autoradiograph is shown in Figure 70.
Figure 70 shows the cleavage patterns which correspond to the cleavage of the sense strand of the wild type (3-globin 536 by fragment (lane 4), mutant 1 fragment (lane 5) and mutant 2 fragment (lane 6). Lanes 1-3 show the uncut controls for wild-type, mutant 1 and mutant 2 substrates, respectively. The lane marked "M" contains biotinylated molecular weight markers prepared as described in Example 8.
As shown in Figure 70, the base substitution present in mutant 1 results in a reduction in the intensity of a band which migrates close to the uncut DNA (lane 5), when compared to wild-type cleavage pattern. The base substitution present in mutant 2 results in the disappearance of the band present in the region just above major product band (approximately 174 nucleotides), when compared to the wild-type cleavage pattern.
For the double-stranded cleavage reactions described above, different reaction conditions were used than those employed for the cleavage of the single-stranded (3-globin DNA substrates described in Example 13. The conditions employed for the cleavage of the double-stranded substrates used a lower MnCI, concentration, no KCl was added, a higher temperature and shorter time course relative to the conditions used in Example 13. Although the cleavage patterns generated by cleavage of the double-stranded and single-stranded ~3-globin DNA were slightly different, the positions of the pattern changes for mutants 1 and are similar to those demonstrated in Example 13, and it was possible to detect the base substitutions in both double-stranded cases. These results show that the subtle changes in DNA secondary structure caused by single base substitutions in larger DNA
substrates can be ' detected by the Cleavase.'~'' BN enzyme whether a single- or double-stranded form of the t DNA substrate is employed.

Identification_Of Mutations In The Human (3-globin Gene CFLPT"'' Patterns Of Unknowns By Comparison To An Existing Library of Patterns The results shown in Examples 13 demonstrated that the CleavaseT"'' BN enzyme generates a unique pattern of cleavage products from each (3-globin substrate tested.
Differences in banding patterns were seen between the wild-type and each mutant; different banding patterns were seen for each mutant allowing not only a discrimination of the mutants from the wild-type but also a discrimination of each mutant from the others.
To demonstrate that the products of the Cleavase~ reaction can be compared to previously characterized mutants for purposes of identification and classification, a second set of (3-globin mutants were characterized and the CFLP~'~"' patterns, by comparison to the set, analyzed in Example 13, were used to determine if the mutants in the second set were the same as any in the first set, or were unique to the second set. Although these isolates have all been described previously (specific references are cited for of these isolates at the end of this example). the experiment was performed "blind", with the samples identified only by a number.
Five (3-globin mutants were compared to the CFLPT"'' patterns from the first set: the wild type [3-globin gene (SEQ ID N0:56) or mutant 1 (SEQ ID N0:58), mutant 2 (SEQ ID
N0:59)or mutant 3 (SEQ ID N0:57). Plasmids for containing these 5 new isolates were grown and purified, and single-stranded substrate DNA, 534 or 536 nucleotides in length. was prepared for each of the 5 (3-globin genes as described above in Example 13a.
Cleavage reactions were performed and reaction products were resolved as described in Example 13; the resulting autoradiograph is shown in Figure 71.
In Figure 71, two panels are shown. Panel A shows the reaction products from the ~-globin isolates described in Example 13 (and as seen in Figure 40). Panel B
shows the reaction products of the five additional isolates, numbered 4, 5, 6, 7 and 8.
The lanes marked "M" contain biotinylated molecular weight markers prepared as described in Example 8.
By comparison to the CFLP~ patterns shown in Panel A, the isolates shown in Panel B can be characterized. It can be seen that the banding pattern of isolate 4 (Panel B, lanel) is the same as was seen for the wild-type (3-globin substrate shown in Panel A
(lane 1 ); isolate 8 (Panel B, lane 5) is comparable to the previously characterized mutant 3 (Panel A, lane 4);
isolate number 6 (Panel B, lane3) has changes in two areas of the pattern and appears to have features of both isolates 2 (Panel A, lane 3)and 3 (Panel A, lane 4); isolates 5 and 7 (Panel B, lanes 2 and 4, respectively) appear to be identical, and they show a pattern not seen in panel A.
To confirm the relationships between the different isolates, the identities of the mutations were then determined by primer extension sequencing using the,finoleT"' DNA
Sequencing System (Promega Corp., Madison, WI) using the PCR primers [5'-biotinylated KM29 primer (SEQ ID N0:54) and 5'-biotinylated RS42 primer (SEQ ID NO:55)], according to the manufacturer's protocol. The sequencing reactions were visualized by the same procedures used for the (3-globin CFLPT"'' reactions, as described in Example 13b.
The two isolates that matched members of the original set by CFLPTM pattern analysis matched by sequence also. Isolate 4 is identical to the wild type sequence (SEQ ID N0:56);
isolate 8 is a duplicate of mutant 3 (SEQ ID N0:57).
Isolate 6 appears by CFLP-'~"'' pattern to have changes similar to both mutant 2 and mutant 3 of the original set. The sequence of mutant 6 (SEQ ID N0:69) reveals that it shares a one base change with mutant 3, a silent C to T substitution in codon 3.
Mutant 6 also has a G-to-A substitution in codon 26, only 4 bases downstream of that found in mutant 2 (SEQ ID
N0:59). This mutation has been shown to enhance a cryptic splice site causing a fraction of the mRNA to encode a nonfunctional protein [Orkin, S.H., et al. (1982) Nature, 300:768]. It is worthy of note that while mutant 6 and mutant 2 both showed alteration in the band that migrates at about 200 nucleotides (e.g., the band is missing or weak in mutant 2 but appears to be split into 3 weak bands in mutant 6) these changes are not of identical appearance.
These CFLPT"' changes, caused by mutations four nucleotides apart, are distinguishable from each other.
The last two isolates, 5 and 7, had the same sequence (SEQ ID N0:70), and revealed a single base substitution within the first intron, at IVS position 110. This mutation is associated with abnormal splicing leading to premature termination of translation of the (3-- 18~ -globin protein [R.A. Spritz et al. (1981) Proc. Natl. Aead. Sci. USA, 78:2455]. It is worthy of note that the band that disappears in the CFLP~ patterns for these mutants (at , approximately 260 nucleotides, as compared to the size markers) is between the indicative bands in the mutant 1 (at approximately 400 nucleotides) and mutant 2 (at approximately 200 nucleotides) CFLPT"'' patterns, and the actual mutation (at nucleotide 334 from the labeled 5' end) is between those of mutants l and 2, at nucleotides 380 and 207, respectively. Thus, the CFLPT"' analysis not only indicated the presence of a change, but also gave positional information as well.
From the results shown in Figure 71, the unique pattern of cleavage products generated by the CleavaseT"'' BN enzyme from each of the first four (wild type plus three variants) (3-globin substrates tested was used as reference to characterize additional (3-globin isolates. The banding patterns show an overall "familial" similarity, with subtle differences (c~.g., missing or shifted bands) associated with each particular variant. Differences in banding patterns were seen between the wild-type and each mutant; different banding patterns were seen for each mutant allowing not only a discrimination of the mutant from the wild-type but also a discrimination of each mutant from the others.

Effect Of The Order Of Addition Of The Reaction Components On The Double-Stranded Cleavage Pattern The cleavage reaction using a double-stranded DNA substrate can be considered a two-step process. The first step is the denaturation of the DNA substrate and the second step is the initiation of the cleavage reaction at the target temperature. As it is possible that the resulting cleavage pattern may differ depending on the conditions present during denaturation (e.g., whether the DNA is denatured in water or in a buffer) as well as on the conditions of reaction initiation (e.g.. whether the cleavage reaction is started by the addition of enzyme or MnCh) the following experiment was performed.
To study the effect of the addition of the reaction components on the resulting cleavage pattern, all possible mixing combinations for 4 reaction components (i.e.. DNA, ' CFLP~' buffer, MnCh and the Cleavase~"~'' BN enzyme) were varied. A single DNA
substrate was used which comprised the 536 by fragment derived from the wild-type (3-globin gene (SEQ ID N0:56). The substrate DNA contained a biotin label at the 5' end of the sense strand and was prepared as described in Example 26.
The substrate was cut in 8 different cleavage reactions which employed different - combinations for the addition of the reaction components at the denaturing and initiation steps. These reactions are described below.
Figure 72 shows the resulting patterns generated by c;leavage of the sense strand of the wild-type (3-globin 536-by substrate (SEQ ID NO:56). In lane 1, the substrate DNA (40 fmoles of DNA in 1 ~1 of TE mixed with 5 ~.1 Hz0) was denatured at 95°C
for 10 seconds, cooled to 55°C and the reaction was started by the addition of a mixture containing 2 yl of SX CFLPT"' buffer with 150 mM KCI, 1 ~,l of 2 mM MnCl7 and 1 1.~I (50 ng) of the CleavaseT"' BN enzyme. In lane 2, the DNA was denatured in the presence of 2 p.l of SX
CFLPTM buffer and reaction was started at 55°C by the addition of 1 ~.I
MnCI, and I ~.l (50 ng) of the CleavaseT"' BN enzyme. In lane 3, the DNA was. denatured in the presence of MnCh and the reaction was started with addition of the buffer and the enzyme.
In lane 4, the denaturation mixture included the substrate DNA and the enzyme and the reaction was started with addition of the buffer and MnCI,. In lane 5, the substrate DNA was denatured in the presence of CFLPTM buffer and MnCh and then the enzyme was added at 55°C. In lane 6, the substrate DNA was denatured in the presence of CFLPT"' buffer and the enzyme and then MnCI= was added at 55°C. Lane 7 shows the uncut control. In lane 8, the DNA was denatured in the presence of the enzyme and MnCh and then the buffer was added at ~5°C.
In lane 9, the substrate DNA was denatured in the presence of the enzyme, MnCh and the CFLPT"' buffer and then the mixture was incubated at 55°C :for 5 minutes. The lane marked "M" contains biotinylated molecular weight markers prepared as described in Example 8.
In all cases reaction was stopped by addition of 6 ~l of stop buffer. The reaction products (5 ~1 each) were resolved by electrophoresis through a 10% denaturing polyacrylamide gel and the DNA was transferred to a membrane and visualized as described in Example 19. The resulting autoradiograph is shown in Figure 72.
The results shown in Figure 72 demonstrate that most of denaturation-initiation protocols employed generated identical cleavage patterns with the exception of the reaction shown in lane 3. In the reaction shown in lane 3, the DNA was denatured in the presence of MnCh and in the absence of CFLP~''' buffer. In the cases where the enzyme and MnCI, were added before the denaturation step (lanes 8,9) no labeled material was detected. In these cases the label was released in a form of short DNA fragments which were produced as a result of nibbling (i.e., the exonucleolytic removal) of the label from the 5' end of the double-stranded DNA template.
The results shown in Figure 72 demonstrate that the order of addition of the reaction components has little effect upon the cleavage pattern produced with the exception that 1 ) the DNA should not be denatured in the presence of MnCI., but in the absence of any buffering solution and 2) the CleavaseT'~'' BN enzyme and MnCI, should not be added together to the DNA prior to the denaturation step. Under these two exceptional conditions, the 5' label was removed from the 5' end of the substrate by the enzyme resulting in a loss of the signal.

Detection Of Mutations In Human p53 Gene By CleavaseT"' Fragment Length Polymorphism (CFLPTMI Analysis The results shown in preceding examples demonstrated that the CFLPT~' reaction could detect single base changes in fragments of varying size from the human (3-globin and tyrosinase genes and that the CFLPT"'' -reaction could be used to identify different strains of virus. The ability of the CleavaseT"' reaction to-detect single base changes in the human tumor suppressor gene p53 was next examined. Mutation of the human p53 gene is the most common cancer-related genetic change; mutations in the p53 gene are found in about half of all cases of human cancer.
The ability of the Cleavase~"~' BN enzyme to cleave DNA fragments derived from the human p53 gene and to detect single base changes in fragments-of the same size was examined. Plamsids containing cDNA clones containing either wild type or mutant p53 sequences were used to generate templates for analysis in the CFLPT"~
reaction. The p53 gene is quite large, spanning 20,000 base pairs and is divided into 11 exons. The use of a template derived from a cDNA allows for maximization of the amount of protein-encoding sequence that can be examined in a DNA fragment of a given size.
The nucleotide sequence of the coding region of the wild type human p~3 cDNA
gene _ is listed in SEQ ID N0:79. The nucleotide sequence of the_coding region of the mutant 14 3 human p53 cDNA gene is listed in SEQ ID N0:80. The nucleotide sequence of the coding region of the -mutant 249 (silent) human p53 cDNA gene is listed in SEQ ID
N0:81. A 601 nucleotide fragment spanning exons-5 through 8 was generated from each of these three p53 cDNAs as follows.

A) Preparation Of The Substrate DNA
Six double stranded substrate DNAs were prepared for analysis in the CFLP~'T'~
reaction. The substrates contained a biotin label at either their 5' or 3' end. The wild type substrate comprises a 601 nucleotide fragment spanning exons 5 through 8 of the cDNA
sequence of the human p53 gene (SEQ ID N0:79) [Baker, S. J. et al., Science ( 1990) 249:912]. Two mutation containing substrates were used. The mutant 143 substrate (SEQ
ID:93) is derived from a p53 mutant V 143A which contains a valine (GTG) to alanine (GCG) substitution; this mutation differs from the wild type p53 ea;on 5-8 fragment by a single nucleotide change [Baker, S. J. et al., Science (1990) 249:912]. The mutant 249 (silent) substrate is derived from a p53 mutant which contains a single base change at amino acid 249. from AGG to AGA (SEQ ID N0:81 ). This single base change does not result in a corresponding amino acid change and is therefore referred to as a silent mutation.
The 601 by double stranded PCR fragments were generated as follows. The primer pair 5'-TCTGGGCTTCTTGCATTCTG (SEQ ID N0:82) and 5'-GTTGGGCAGTGCTCGCTTAG (SEQ ID N0:83) were used to prime the PCRs. The synthetic primers were obtained from Integrated DNA Technologies (Coralville, IA). The primers were biotinylated on their 5' ends with the Oligonucleotide Biotinylation Kit purchased from USB-Amersham (Cleveland, OH) according to the manufacturers' protocols.
When the sense strand was to be analyzed in the CFLPT"' reaction, the primer listed in SEQ
ID N0:82 was labeled at the 5' end with the biotin. When the anti-sense strand was to be analyzed in the CFLP~ reaction, the primer listed in SEQ ID N0:83 was labeled at the 5' end with the biotin.
The target DNA used in the PCR for the generation of the 601 by fragment derived from the wild type p~3 cDNA was the plasmid CMV-p53-SN3 [Baker, S. J. et crl., .supr°a];
this plasmid contains the coding region listed in SEQ ID N0:79. The target for the generation of the 601 by fragment derived from the mutant 143 was the plasmid CMV-p53-SCX3 [Baker, S. J. et al., supra]; this plasmid contains the coding region listed in SEQ ID
N0:80. REF). The target for the generation of the 601 by fragment derived from mutant 249 (silent) was the plasmid LTR 273 His jChen, P.-L. et al., Science (1990) 250:1576]; this plasmid contains the coding region listed in SEQ ID N0:81. DNA was prepared from i bacteria harboring each plasmid (plasmid DNA was isolated using standard techniques). The 601 by PCR products were prepared as follows.

The symmetric PCR reactions contained 50 ng of plasmid DNA, 50 pmoles primer 5"-TCTGGGCTTCTTGCATTCTG (SEQ ID:95), 50 pmoles of primer 5'-GTTGGGCAGTGCTCGCTTAG (SEQ ID:96), and .50 pM each dNTP in 95 p.l of 1X PCR
buffer. The reaction mixtures were overlaid with 50 ~.l ChillOutT"'' (MJ
Research, Watertown, MA) and the tubes were heated to 95°C for 2.5 min. Tag DNA polymerase (Promega Corp., Madison, WI) was then added as 1.25 units of enzyme in 5 ~.l of 1X PCR buffer.
The tubes were then heated to 95°C for 45 seconds, cooled to 55°C for 45 seconds and heated to 72°C
for 75 seconds for 34 cycles with a 5 min incubation at 72°C after the last cycle.
The PCR products were gel purified as follows. The products were precipitated by the addition of NaCI to a final concentration of 0.4M, 20 p.g glycogen carrier and 500 pl ethanol.
The DNA was pelleted by centrifugation and the PCR products were resuspended in 25 or 50 q.l sterile distilled water to which was added an equal volume of a solution containing 95%
formamide, 20 mM EDTA and 0.05% each xylene cyanol and bromophenol blue. The tubes were then heated to 85°C for 2 min and the reaction products were resolved by electrophoresis through a 6% polyacrylimide gel ( 19:1 cross-link) containing 7 M urea in a buffer containing 0.5X TBE. The DNA was visualized by ethidium bromide staining and the 601 by fragments were excised from the gel slices by passive diffusion overnight into a solution containing 0.5 M NH40Ac, 0.1 % SDS and 0.1 % EDTA. The DNA was then precipitated with ethanol in the presence of 4 ~.g of glycogen carrier. The DNA was pelleted, resuspended in sterile distilled water and reprecipitated by the addition of NaCI to a final aqueous concentration of 0.2 M and 80% ethanol. After the second precipitation, the DNA
was pelleted and resuspended in 30 ~.l sterile distilled water or TE.
The nucleotide sequence of these 601 by templates are listed in SEQ ID NOS:84-89.
The sense strand of the 601 nucleotide wild type fragment is listed in SEQ ID
N0:84. The anti-sense strand of the 601 nucleotide wild type fragment is listed in SEQ ID
N0:85. The sense strand of the 601 nucleotide mutant 143 fragment is listed in SEQ ID
N0:86. The anti-sense strand of the 601 nucleotide mutant 143 fragment is listed in SEQ ID
N0:87. The sense strand of the 601 nucleotide mutant 249 (silent) fragment is listed in SEQ ID N0:88.
The anti-sense strand of the 601 nucleotide mutant 249 (silent) fragment is listed in SEQ ID
N0:89.

B) Cleavage Reaction Conditions Cleavage reactions comprised approximately 100 fmoles of the resulting double stranded substrate DNAs (the substrates contained a. biotin moiety at the ~' end of the sense or antisense strand) in a total volume of 5 ~,l of sterile distilled water.
The reactions were heated to 95°C for 15 seconds to denature the substrates and then quickly cooled to 50°C (this step allows the DNA to assume its unique secondary structure by allowing the formation of intra-strand hydrogen bonds between complimentary bases).
The reactions were performed in either a thermocycler (MJ Research. Watertown, MA) programmed to heat to 95°C for 15 seconds and then cooled immediately to 50° C.
Once the tubes were cooled to the reaction temperatwe of 50°C, the following components were added: 5 ~.1 of a diluted enzyme mix containing 0.2 pl of CleavaseT"'' BN
[SOng/p.l 1 X CleavaseT"'t Dilution Buffer (0.5% NP40, 0.5°ro Tween 20, 20 mM Tris-Cl, pH
8.0, 50 mM KCl, 10 ~.g/ml BSA)]; 1 p,l of 10 X CFLP~'~"'' reaction buffer (100 mM MOPS, pH 7.~. 0.5% NP 40, 0.5% Tween 20), and 1 pl of 2mM IVInCI,.
A no enzyme control ( 10 ~.1) was set up in parallel for each PCR fragment examined;
this control differed from the above reaction mixture only in that sterile distilled water was substituted for Cleavase'~"~ BN enzyme. Reactions were stopped after 3 minutes by the addition of 8 p.l of stop buffer.
The samples were then heated to 85°C for 2 minutes and 4 p.l of each reaction mixture were resolved by electrophoresis through a 6% polyacrylimide gel (19:1 cross-link), with 7M
urea, in-a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated allowing the gel to remain flat on one plate. -A 0.2 ~.m-pore positively charged nylon membrane (Schleicher and Schuell.
Keene. NH), pre-wetted with O.SX TBE, was laid on top of the exposed acrylamide gel. All air bubbles trapped between the gel and the membrane were removed. Two pieces of 3MM
filter paper (Whatman) were then placed on top of the membrane, the other glass plate was replaced, and the sandwich was clamped with binder clips. Transfer was allowed to proceed overnight. After transfer, the membrane was carefully peelE;d from the gel and washed in 1 X
Sequencase Images Blocking Buffer (United States Biocherr~ical) for two 15 minute intervals with gentle agitation. Three tenths of a ml of the buffer was used per cm' of membrane. A
streptavidin-alkaline phosphatase conjugate (SAAP, United States Biochemical, Cleveland, OH) was added to a 1:3000 dilution directly to the blocking solution. and agitated for 15 minutes. The membrane was washed 3 times (~ min/wash) in 1 X SAAP buffer ( 1 OOmM

Tris-HCI, pH 10; 50 mM NaCI) with 0.1% SDS, using 0.5 mls/cm'' of membrane.
The membrane was then washed twice in 1 X SAAP buffer without SDS, but containing 1 mM
MgCh, drained thoroughly and placed in a heat sealable bag. Using a sterile pipet ,tip, 0.0~
ml/cm'- of CDP-StarT"'' (Tropix. Bedford, MA) was added to the bag and distributed over the membrane for 5 minutes. The bag was drained of all excess liquid and air bubbles. The membrane was then exposed to X-ray film (Kodak XRP) for an initial 30 minute exposure.
Exposure times were adjusted as necessary for resolution and clarity. The result autoradiograph is shown in Figure 73.
In Figure 73, the lane marked M contains biotinylated molecular weight markers. The marker fragments were purchased from Amersham (Arlington Heights, IL). - Lanes contain the reaction products from the incubation of double stranded DNA
substrates in the absence of the CleavaseT"' BN enzyme (i.e., uncut controls). Lane I contains the wild type fragment labeled on the sense strand of the 601 by PCR fragment. Lane 2 contains the mutant 143 fragment labeled on the sense strand of the 601 by PCR fragment.
Lane 3 contains the wild type fragment labeled on the antisense strand of the PCR
product. Lane 4 contains the fragment encoding the silent mutation at amino acid 249 labeled on the antisense strand of the PCR product. Lanes 5-8 contain the reaction products from the incubation of the 601 by double stranded substrates with the CleavaseT"'' BN enzyme. Lane 5 contains products generated using the wild type fragment labeled on the sense strand;
lane 6 contains products generated using the mutant 143 labeled on the sense strand. Lanes 7 and 8 contain products generated using the wild type and mutant 249 (silent) substrates, respectively, labeled on the anti-sense strand.
The results shown in Figure 73 demonstrate that a similar, but distinctly different, pattern of cleavage products was generated by the digestion of wild type and mutant-containing templates by the CleavaseT"'' BN enzyme. Comparison of lanes 5 and 6 reveals a difference in the band pattern in the 100 nucleotide range. Specifically, the strong band present in the wild type (at around 100 nucleotides) was missing in the V 143A
mutant while two bands immediately below this strong band were prominent in the mutant and not evident in the wild type. In the 200 nucleotide range, a pronounced doublet seen in the wild type is missing from the mutant, which instead contained a strong single band migrating slightly a faster than the wild type doublet. Similarly, comparison of lanes 7 and 8 revealed differences between the pattern generated from cleavage of the anti-sense strand of the wild type fragment and the mutant 249 (silent) fragment. In the 100 nucleotide range, the wild type -fragment exhibited a strong doublet whereas tl lupper band of this doublet was missing in the mutant 249 (silent) fragment. In addition, two prominent bands present in the wild type pattern in the 150-180 by range were completely absent from the mutant 249 (silent) cleavage products. -~ 5 Although each mutant fragment analyzed in Figure 73 differs from the wild type by only one of the 601 nucleotides, a unique pattern of cleavage: fragments was generated for each. Furthermore, at least one pattern change occurred in each mutant in the immediate vicinity (i. e., within 10-20 nucleotides) of the DNA sequence change. This experiment demonstrates that CFLPT"'' is capable of distinguishing the presence of single base changes in PCR fragments containing exons 5 through 8 of the p53 gene.

Detection Of Genetically Engineered Mutations In PCR Fragments Of The Human p53 Gene The ability of the Cleavase~'~'' BN enzyme to detect single base changes genetically engineered into PCR fragments containing exons 5 through 8 of the human p53 gene was analyzed. The single base changes introduced were 1 ) a change from arginine (AGG) to serine (AGT) at amino acid 249 (termed the R249S mutation) and 2) a change from arginine (CGT) to histidine (CAT) at amino acid 273 (termed the R273H mutation). Both of these mutations have been found in human tumors and have been identified as mutational hot spots [Hollstein et al., Science 253:49 (1991)]. The R249S mutation is strongly correlated with exposure to aflatoxin B and/or infection with hepatitis B virus [Caron de Fromental and Soussi, Genes, Chromosomes and Cancer (1992) pp. 1-15]. The R273H mutation arises as a result of a transition at a CpG dinucleotide. Such transitions account for approximately one-third of the known p53 mutations and are characteristic of a variety of tumor types [Caron de Fromental and Soussi, supra; Hollstein et al., supra].
Plasmids containing the R249S and R273H mutations were engineered according to a variation of a protocol described by R. Higuchi [in PCR Technolog3e Principles and ~ Applications,for DNA Amplification, H. A. Ehrlich, Ed.(1989) Stockton Press, NY, pp. 61-70]. This methodology allows the generation of collection of plasmids containing DNA
sequences corresponding to known p53 mutations. The availability of this collection allows the generation of p53 "bar code" library which contains the CFLPT"'' patterns generated by cleavage of the p53 mutants- using the CleavaseTM enzymes. , A) Construction of a 601 by PCR fragment Containing the R249S Mutation To generate a 601 by fragment containing the R249S mutation, a 2-step recombinant PCR was performed (see Figure 6 for a schematic representation of the 2-step recombinant PCR). In the first or "upstream" PCR, oligonucleotides 5'-TCTGGGCTTCTTGCATTCTG
(SEQ ID N0:82) and 5'-GAGGATGGGACTCCGGTTCATG (SEQ ID N0:90) were used to amplify a 427 by fragment containing the G to T base change resulting in the mutation; the sequence of the 427 by fragment is listed in SEQ ID N0:98. In the second or "downstream" PCR, oligonucleotide 5'-CATGAACCGGAGTCCCATCCTCAC (SEQ ID
N0:91 ) and 5'-GTTGGGCAGTGCTCGCTTAG (SEQ ID N0:83) were used to amplify a 196 by fragment containing the same base change on the complementary strand; the sequence of the 196 by fragment is listed in SEQ ID N0:99. __ For each PCR, 10 ng of a cDNA clone encoding the wild type p53 gene (coding region listed in SEQ ID N0:79) were used as the template in a 50 p.l PCR
reaction. In the case of the upstream fragment, 10 ng of template were added to a tube containing 5 picomoles of the oligonucleotide 5'-TCTGGGCTTCTTGCATT CTG (SEQ ID N0:82), 5 pmoles of the oligonucleotide 5'-GAGGATGGGACTCC GGTTCATG (SEQ ID N0:90), and 50 q.M each dNTP, in 45 p.l of 1X PCR buffer. For the downstream fragment, 10 ng of the wild type template, plasmid CMV-p53-SN3 (Example 29) were added to 5 picomoles of the oligonucleotide 5'-CATGAACCGGAGTCCCATCCTCAC (SEQ ID N0:91 ) and 5 picomoles of the oligonucleotide 5'-GTTGGGCAGTGCTCGCTTAG (SEQ-ID N0:83), and 50 ~.M each dNTP in 1 X PCR buffer.
Tubes containing 45 p.l of the above mixtures for each template to be amplified were overlaid with 50 p.l ChillOut~' (MJ Research, Watertown, MA) and the tubes were heated to 95°C for 2.5 min and then cooled to 70°C. Taq DNA polymerase (Promega) was then added as 1.23 units of enzyme in 5 ~.l of 1X PCR buffer. The tubes were then heated to 95°C for 45 seconds, cooled to 55°C for 45 seconds and heated to 72°C for 75 seconds for 24 cycles with a 5 min incubation at 72°C after the last cycle.
The PCR products were gel purified as follows. Ten microliters of each PCR
product were mixed with 10 p.l of stop buffer. The tubes were then heated to 85°C for 2 min and the ' reaction products were resolved by electrophoresis through a 6% polyacrylimide gel ( 19:1 cross-link) containing 7 M urea in a buffer containing O.SX TBE (the-polyacrylimide solutions ' used were freshly prepared). The DNA was visualized by eahidium bromide staining and the fragment was excised from the gel slice by passive diffusion overnight into a solution containing 0.5 M NH40Ac, 0.1 % SDS and 0.1 % EDTA at 37° C.
' Ten microliters of each eluted PCR product were combined to serve as the recombinant template to prime a second round of PCR. To this template, 10 picomoles of 5'-' biotin exon 8 primer (SEQ ID N0:83), 10 pmoles of 5'-exon 5 primer (SEQ ID
N0:82). and 50 p.M each dNTP in 1X PCR buffer were added. Tubes containing 90 pl of the above mixtures for each template to be amplified were overlaid with 50 q.l ChillOutT"' (MJ Research, Watertown, MA) and the tubes were heated to 95°C for 2.5 min and then cooled to 70°C.
Tuy DNA polymerise (Promega) was then added as 2.5 units of enzyme in 5 ~,1 of buffer. The tubes were then heated to 95°C for 45 seconds, cooled to 47°C to allow the two template molecules to anneal, then heated to 72° C to allow extension of the primers by Tcrg DNA polymerise. Following this initial cycle of denaturation, annealing and extension, 25 cycles in which the reactions were heated to 95°C for 45 seconds, cooled to 55°C for 45 l~ seconds, and then heated to 72°C for 1 minute were carried out, followed by a ~ min extension at 72° C. The fragments were then ethanol precipitated and gel purified as described in Example 29.
B) Construction of a 601 by PCR Fragment Containing the R273H Mutation To generate a 601 by fragment containing the R273H mutation, a 2-step recombinant PCR was performed using the procedure described in section a) was used to simultaneously amplify PCR fragments encoding a single base change from arginine (CGT) to histidine (CAT) at amino acid 273. In the first or "upstream" PCR, oligonucleotide 5'-TCTGGGC
TTCTTGCATTCTG-3' (SEQ ID N0:82) and 5'-GCACAAACATGCACCTCAAAGCT-3' (SEQ ID N0:92) were used to generate the 498 by fragment whose sequence is listed in SEQ
ID NO:100. In the second or "downstream" PCR, oligonucleotide 5'-CAGCTTTG
AGGTGCATGTTTGT-3' (SEQ ID N0:93) was paired with oligonucleotide 5'-GTTGGG
CAGTGCTCGCTTAG-3' (SEQ ID N0:83) to generate a 127 nucleotide fragment whose sequence is listed in SEQ ID NO:101. The DNA fragments were electrophoresed.
eluted, combined and used to prime a second round of PCR as described in section a) to generate a 601 by PCR product containing the R273H mutation.
..
C) Sequence Analysis of the 601 Nucleotide PCR Fragments The recombinant 601-by PCR products generated through this two step PCR
procedure were gel purified as described in Example 29. The PCR products were sequenced using the -19~-fmol~ DNA Sequencing System (Promega) in conjunction with oligonucleotide 5'-biotin-GTTGGGCAGTGCTCGCTTAG (SEQ ID N0:83) according to manufacturers' standard protocols to verify the presence of the engineered mutations.
The nucleotide sequence corresponding to the sense strand of the 601 nucleotide R249S mutant fragment is listed in SEQ ID N0:94. The anti-sense strand of the nucleotide R249S mutant fragment is listed in SEQ ID N0:95. The sense -strand of the 601 nucleotide R273H mutant fragment is listed in SEQ ID N0:96. The anti-sense strand of the 601 nucleotide R273H mutant fragment is listed in SEQ ID N0:97.
D) Cleavage Reactions In order to generate ample quantities-of DNA for subsequent CFLP~'~"'' analysis, the 601 by fragments containing either the R249S or the R273H mutation were used as templates in an additional round of PCR. Approximately 2 fmoles of each 601 by fragment were added to pmoles of the primers corresponding to SEQ ID NOS:82 and 83 (SEQ ID N0:83 contained a biotin on the 5' end), 50 ~.M each dNTP, 20 mM Tris-HCI, pH 8.3.
1.5 mM
15 MgCI,, 50 mM KCI, 0.05% Tween 20 and 0.05% NP40. Tubes containing 90 pl of the above mixture were assembled for each template to be amplified; the tubes were overlaid with 50 p.l ChillOutT''~ (MJ Research, Watertown, MA) and the tubes were heated to 95°C for 2.5 min and then cooled to 70°C. Taq DNA polymerise (Promega) was then added as 2.5 units of enzyme in 5 p.l of 1X PCR buffer. The tubes were then heated to 95°C
for 45 seconds, 20 cooled to 47°C to allow the two template molecules to anneal, then heated to 72° C to allow extension of the primers by Taq DNA polymerise. Following this initial cycle of denaturation, annealing and extension, 25 cycles in which the reactions were heated to 95°C
for 45 seconds, cooled to 55°C for 45 seconds, and then heated to 72°C for 1 minute were carried out. followed by a 5 min extension at 72° C. The fragments were then ethanol precipitated and gel purified as described in Example 29. The gel purified fragments were then used in CFLP~' reactions as follows.
Cleavage reactions comprised approximately 100 fmoles of the resulting double stranded substrate DNAs (the substrates contained a biotin moiety at either the 5" end of the sense or anti-sense strand) in a total volume of 5 p.l (sterile distilled water was used to bring the volume to 5 ~1). The reactions were heated to 95°C for 15 seconds to denature the substrates and then quickly cooled to 50°C (this step allows the DNA to assume its unique secondary structure by allowing the formation of intra-strand hydrogen bonds between complimentary bases). The reaction were performed in either a thermocycler (MJ
Research, Watertown, MA) programmed to heat to 95°C for 15 seconds and then cool immediately to 50° C or the tubes were placed manually in a heat block set at 95°C and then transferred to a second heat block set at 50°C.
' Once the tubes were cooled to the reaction temperature of 50°C, 5 ~1 of a diluted enzyme mix containing 0.2 ~,l of Cleavase'~"" BN enzyme [50 ng/p.l 1 X
CleavaseTM Dilution Buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-Cl, pH 8.0, 50 mM KCI, 10 ~.g/ml BSA)], 1 ~l of 10 X CFLP~ reaction buffer (100 mM MOPS, pH 7.5, 0.5% NP 40, 0.5%
Tween 20), and 1 ~1 of 2 mM MnCh. A 10 ~l no enzyme control was set up in parallel for each PCR fragment examined in which sterile distilled water was substituted for the Cleavase rM BN
enzyme. After 2 minutes at 50°C, the reactions were stopped by the addition of 8 ~1 of stop buffer . -The samples were heated to 85°C for 2 minutes and 7 ~1 of each reaction were resolved by electrophoresis through a 10% polyacrylimide gel ( 19:1 cross-link), with 7M urea, in a buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated allowing the gel to remain flat on one plate. A 0.2 ~.m-pore positively charged nylon membrane (Schleicher and Schuell, Keene, NH), pre-wetted with O.SX TBE, was laid on top of the exposed acrylamide gel. All air bubbles trapped between the gel and the membrane were removed. Two pieces of 3MM
filter paper (Whatman) were then placed on top of the membrane, the other glass plate was replaced, and the sandwich was clamped with binder clips. Transfer was allowed to proceed overnight. After transfer, the membrane was carefully peeled from the gel and washed in 1 X
Sequencase Images Blocking Buffer (United States Biochemical) for two 1 ~
minute intervals with gentle agitation. Three tenths of a ml of the buffer was used per cm? of membrane. A
streptavidin-alkaline phosphatase conjugate (SAAP, United States Biochemical, Cleveland, OH) was added to a 1:3000 dilution directly to the blocking solution, and agitated for 15 minutes. The membrane was washed 3 times (5 min/wash) in 1 X SAAP buffer ( 100 mM
Tris-HCI, pH 10; 50 mM NaCI) with 0.1% SDS, using 0.5 .mls/cm'- of membrane.
The membrane was then washed twice in 1 X SAAP buffer witlxout SDS, but containing 1mM
MgCI,, drained thoroughly and placed in a heat sealable bag. Using a sterile pipet tip, 0.0~
. 30 ml/cm' of CDP-StarT"' (Tropix, Bedford, MA) was added to the bag and distribu~:.d over the membrane for S minutes. The bag was drained of all excess liquid and air bubbles. The membrane was then exposed to X-ray film (Kodak XRP) far an initial 30 minute exposure.

Exposure times were adjusted as necessary for resolution and clarity. The results are shown in Figure 74.
In Figure 74, the lane marked M contains biotinylated molecular weight markers.
The marker fragments were purchased from Amersham (Arlington Heights, IL) and include ' bands corresponding to lengths of 50, 100, 200, 300, 400, 500, 700, and 1000 nucleotides. , Lanes 1-4 contain the reaction products from the incubation of double stranded DNA
substrates labeled on the antisense strand in the absence of the Cleavase~'~'' BN enzyme. Lane 1 contains the reaction products from the wild type fragment (SEQ ID N0:85):
lane 2 contains the reaction products from the engineered R249S mutation (SEQ ID
N0:95); lane 3 contains the reaction products from the 249 (silent) mutation (SEQ ID N0:89);
lane 4 contains the reaction products from the engineered R273H mutation (SEQ ID
N0:97).
Lanes 5-8 contain the cleavage products generated from the sense strand each of these templates when incubated in the presence of the CleavaseT"' BN enzyme. Lane ~
contains the cleavage products from the wild type fragment (SEQ ID N0:84);-lane 6 contains the cleavage products from the R249S fragment (SEQ ID N0:94); lane 7 contains the cleavage products from the 249 (silent) mutant fragment (SEQ ID N0:88); lane 8 contains the cleavage products from the 8273 S fragment (SEQ ID N0:96).
The results shown in Figure 74 demonstrate that similar, but distinctly different, patterns of cleavage were generated from each of these templates containing single-base changes. Lane 6 shows the attenuation of bands in the 150-180 nucleotide range, as well as, the loss of a band in the 100 nucleotide range when compared to the wild-type pattern shown in lane 5. In addition, lane 6 shows a new band appearing in the 140 nucleotide range, and increased intensity in the top band of a doublet at about 120 nucleotides.
Examination of the silent 249 mutant (lane 7) which differs from wild-type at the same nucleotide position as R249S (lane 6), revealed pattern differences relative to both the wild type (lane 5) as well as to the R249S (lane 6) mutation. Specifically, comparison to lane 5 shows an attenuation of bands in the 150-180 nucleotide range as well as the loss of a band in the 100 nucleotide range, as was seen in lane 6. However, the sample, in lane 7 does not exhibit the additional band in the 140 nucleotide range, nor the increased intensity in the top band of the doublet in the 120 nucleotide range seen in lane 6. This result demonstrates that the CFLPT"1 technique is capable of distinguishing between changes to a different base at the same nucleotide "
position.

Examination of the reaction products in lane 8 reveals the loss of a band in the 100 nucleotide range in the R273S fragment when compared to the wild-type pattern in lane 5.
This CFLPT"' pattern is distinct from those in lanes 6 and 7, however, in that it does not show attenuation of bands in the 150-180 nucleotide range; in thi:> region of the gel this pattern is essentially indistinguishable from that generated from the wild type fragment.
The above results demonstrate that CFLP~ can be used to detect clinically significant mutations in the human p53. Further, these results indicate that the CFLPTM
technique is sufficiently sensitive to distinguish different base changes at the same position from one another, as well as from wild type. In addition these results show that the 2-PCR technique can be used to generate a collection of PCR fragments containing known p53 mutations; such a collection allows the generation of a p53 bar code library containing the CFLPTM patterns generated by different p53 mutations.

Detection Of The Presence Of Wild Type And Mutant Sequences In Mixed Samples The ability of the CFLP~ reaction to detect the presence of different alleles of the same sized PCR fragments in a mixed sample, such as might be found in heterozygous or otherwise heterogenous tissue, samples was examined.
PCR products_containing a biotin label on the sense strand were produced and purified as described in Example 29 for the wild type p53 (SEQ ID N0:84) and mutant 143 (SEQ ID
N0:86) 601-by fragments. Aliquots of these samples were diluted to a final concentration of approximately 12.5 fmols/~,1 and mixed in different proportions to give a spectrum of ratios of wild type to mutant DNA. Four microliters of the diluted DNA samples, for an approximate total of 50 fmols of DNA in each sample, mixed in various combinations, were placed in microfuge tubes and heated to 95 °C for 15 seconds. The tubes were rapidly cooled to 50°C
and 6 q.l of a diluted enzyme mix containing 0.2 ~.l of the Cleavase'~'' BN
enzyme [SOn~~/q l 1 X CleavaseT"' Dilution Buffer (0.5% NP40, 0.5% Tween 20, 20 mM Tris-Cl, pH
8.0, 50 mM
KCI, 10 q.g/ml BSA)~ , 1 ~l of 10 X CFLPT"'' reaction buffer (100mM MOPS, pH
7.~, 0.5%
NP 40, 0.5% Tween 20), and 1 q,l of 2mM MnCI,. A 10 yl no enzyme control was set up in parallel for each PCR fragment examined, with the difference that sterile distilled water was substituted for the CleavaseT'~' BN enzyme. After 1.5 minutes at 50°C, the reactions were stopped by the addition of 8 ~l of stop buffer. In addition, 4 ~I of wild type only as well as 4 ~.l of V 143A only were analyzed by the same method for comparison to the mixed samples.
Samples were heated to 85°C for 2 minutes and 7 p.l of each reaction were resolved .
by electrophoresis through a 10% polyacrylimide gel ( 19:1 cross-link), with 7M urea, in a ' buffer containing O.SX TBE.
After electrophoresis, the gel plates were separated allowing the gel to remain flat on one plate. A 0.2 ~.m-pore positively charged nylon membrane (Schleicher and Schuell, Keene, NH), pre-wetted with O.SX TBE, was laid on top of the exposed acrylamide gel. All air bubbles trapped between the gel and the membrane were removed. Two pieces of 3MM
filter paper (Whatman) were then placed on top of the membrane, the other glass plate was replaced, and the sandwich was clamped with binder clips. Transfer was allowed to proceed overnight. After transfer, the membrane was carefully peeled from the gel and washed in 1 X
Sequencase Images Blocking Buffer (United States Biochemical) for two I S
minute intervals with gentle agitation. Three tenths of a ml of the buffer was used per cm' of membrane. A
streptavidin-alkaline phosphatase conjugate (SAAP, United States Biochemical, Cleveland, OH) was added to a 1:300 dilution directly to the blocking solution, and agitated for 15 minutes. The membrane was washed 3 times (5 min/wash) in 1 X SAAP buffer ( 1 OOmM
Tris-HCI, pH 10; 50 mM NaCI) with 0.1 % SDS, using 0.5 mls/cm' of membrane.
The membrane was then washed twice in 1 X SAAP buffer without SDS, but containing 1mM
MgCI,, drained thoroughly and placed in a heat sealable bag.- Using a sterile pipet tip, 0.05 ml/cm' of CDP-Star (Tropix, Bedford, MA) was added to the bag and distributed over the membrane for 5 minutes. = The bag was drained of all excess liquid and air bubbles. The membrane was then exposed to X-ray film (Kodak XRP) for an initial 30 minute exposure.
Exposure times were adjusted as necessary for resolution and clarity. The resulting autoradiograph is shown in Figure 75.
In Figure 75, the lane marked .M contains biotinylated molecular weight markers obtained from Amersham (Arlington Heights, IL) and include bands corresponding to lengths of 50, 100, 200, 300, 400s 500, 700, and 1000 nucleotides. Lanes 1 and 2 contain the reaction products from the no enzyme controls for the wild type and V 143A
mutant fragments, respectively. Lane 3 contains cleavage products from the sample containing the wild type fragment only. Lane 4 contains cleavage products from wild type and mutant fragmentsmixed in a 1:1 ratio. Lane ~ contains cleavage products from a reaction containing a 1: 2 ratio of wild type to mutant fragment. Lane 6 contains reaction products present in a ' ratio of wild type to mutant of 1:9. Lane 7 contains cleavage products from a sample containing V 143A mutant DNA only. Lane 8 contains cleavage products mixed at a ratio of _, wild type to mutant of 2:1. Lane 9 contains cleavage products mixed at a ratio of wild type to mutant 4:1. Lane 10 contains cleavage products mixed a ratio of wild type to mutant to 9:1.
t The results shown in Figure 75 demonstrate that the presence of different alleles can be detected in a mixed sample. Comparison of lanes 4-6 anal lanes 8-10 with either lane 3 or lane 7 demonstrates that the lanes containing mixed reactions exhibit distinct differences from either sample alone. Specifically, in the 100 nucleotide region, there is a doublet in the wild type sample that shifts in the mutant (see discussion of Figure 74 in Example 29). All three of these bands are present in the mixed samples (lanes 4-6 and lanes 8-10) whereas only one or the other pair is detectable in lanes 3 and 7.

Detection and Identification of Hepatitis C Virus Genotypes By Cleavase.'~"'' Fragment Length Polymorphism Analysis Hepatitis C virus (HCV) infection is the predominant cause of post-transfusion non-A, non-B (NANB) hepatitis around the world. In addition, HCV is the major etiologic agent of hepatocellular carcinoma (HCC) and chronic liver disease world wide. Molecular biological analysis of the small (9.4 kb) RNA genome has showed that some regions of the genome are very highly conserved between isolates, while other regions are subject to fairly rapid mutation. These analyses have allowed these viruses to be divided into six basic genotype groups, and then further classified into several sub-types [Altamirano et al., J. Infect. Dis.
171:1034 (1995)]. These viral groups are associated with different geographical areas, and and accurate identification of the agent in outbreaks is important in monitoring the disease.
While only genotype 1 HCV has been observed in the United States, multiple HCV
genotypes have been observed in both Europe and Japan. HCV genotype has also been associated with differential efficacy of treatment with interferon, with Group 1 infected individuals showing little response. The ability to identify the genotype of HCV present in an infected individual allows comparisons of the clinical outcomes from infection by the different types of HCV, and from infection by multiple types in a single individual. Pre-screening of infected WO 96115267 PCTlUS95/14673 individuals for the viral type will allow the clinician to make a more accurate diagnosis, and to avoid costly but fruitless drug treatment.
In order to develop a rapid and accurate method of typing HCV present in infected individuals, the ability of the Cleavase~ reaction to detect and distinguish between the major ' genotypes and subtypes of HCV was examined. Plasmids containing DNA derived from the conserved 5' untranslated region of six different HCV RNA isolates were used to generate templates for analysis in the CFLPTT' reaction. The HCV sequences contained within these six plasmids represent genotypes 1 (four sub-types represented; 1 a, 1 b, I c and Ol c), 2 and 3.
The nomenclature of the HCV genotypes used is that of Simmonds et al. [as described in Altamirano et al., supra].
A) Generation of Plasmids Containing HCV Sequences Six DNA fragments derived from HCV were generated by RT-PCR using RNA
extracted from serum samples of blood donors; these PCR fragments were a gift of Dr. M.
Altamirano (University of British Columbia, Vancouver). These PCR fragments represent HCV sequences derived from HCV genotypes 1 a, 1 b, 1 c, D 1 c, 2c and 3 a.
The RNA extraction, reverse transcription and PCR were performed using standard techniques [Altamirano et al., J. Infect. Dis. 171:1034 (1995)]. Briefly, RNA
was extracted from 100 p,l of serum using guanidine isothiocyanate, sodium lauryl sarkosate and phenol-chloroform [Inchauspe et al., Hepatology 14:595 (1991)]. Reverse transcription was performed according to the manufacturer's instructions using a GeneAmp rTh reverse transcriptase RNA PCR kit (Perkin-Elmer) in the presence of an external antisense primer.
HCV342. The sequence of the HCV342 primer is 5'-GGTTTTTCTTTGAGGTTTAG-3' (SEQ ID N0:102). Following termination of the RT reaction, the sense primer HCV7 [5'-GCGACACTCCACCATAGAT-3' (SEQ ID N0:103)] and magnesium were added and a first PCR was performed. Aliquots of the first PCR products were used in the second (nested) PCR in the presence of primers HCV46 [5'-CTGTCTTCACGCAGAAAGC-3' (SEQ ID
N0:104)] and HCV308 [5'-GCACGGTCTACGAGACCTC-3' (SEQ ID NO:105)]. The PCRs produced a 281 by product which corresponds to a.conserved 5' noncoding region (NCR) region of HCV between-positions -284 and -4 of the HCV genome [Altramirano eI
crl., .1.
Infect. Dis. 171:1034 (1995)].
.
The six 281 by PCR fragments were used directly for cloning or they were subjected ' to an additional amplification step using a 50 ~,1 PCR comprising approximately 100 finoles of DNA, the HCV46 and HCV308 primers at 0.1 ~.M, 100 pM of all four dNTPs and 2.~ ' °-units of Taq polymerise in a buffer containing 10 mM Tris-HCI, pH"~.~, 50 mM
KCI, 1.5 mM MgCI, and 0.1% Tween 20. The PCRS were cycled 25~ times at 96°C for 45 sec., 55°C
for 45 sec. and 72°C for 1 min. Two microliters,of.either the,original DNA samples or the = reamplified PCR products were used for cloning in the linear pT7Blue T-vector (Novagen, w 5 Madison,WI) according to manufacturer protocol. After the PCR products were ligated to the pT7Blue T-vector; the ligation reaction mixture was used to transform competent JM 109 cells (Promega). Clones containing the pT7Blue T-vector with an insert were selected by the presence of colonies having a white color on LB plates containing 40 ~.g/ml X-Gal, 40 l~g/ml IPTG and 50 ~g/ml ampicillin. Four colonies for each PCR sample were picked and grown overnight in 2 ml LB media containing 50 ~.g/ml carbenicillin. Plasmid DNA was isolated using the following alkaline miniprep protocol. Cells from 1.5 ml of the overnight culture were collected by centrifugation for 2 min. in a microcentrifuge ( 14K rpm), the supernatant was discarded and the cell pellet was resuspended in 50 ~,l TE buffer with 10 ~.g/ml RNAse A (Pharmacia). One hundred microliters of a solution containing 0.2N NaOH, 1%
SDS was added and the cells were lysed for 2 min. The lysate was gently mixed with 100 ~,1 of 1.32 M potassium acetate, pH 4.8, and the mixture was centifugated for 4 min. in a microcentrifuge ( 14K rpm); the pellet comprising cell debris was discarded.
Plasmid DNA
was precipitated from the supernatant with 200 ~.l ethanol and pelleted by centrifugation a microcentrifuge (14K rpm). The DNA pellet was air dried forl5 min. and was then redissolved in 50 ~.l TE.
To analyze the cloned HCV inserts, 1 ~.1 of plasmid DNA (approximately 10 to ng) reamplified in a 50 p.l PCR using the HCV46 and HCV308 primers as described above with the exception that 30 cycles of amplification were employed. The PCR
products were separated by electrophoresis on a 6% non-denaturing acrylamide gel (29:1 cross linked) in O.SX TBE buffer; clones that gave rise to a 281 by PCR product were selected for further analysis.
For sequencing purposes, plasmid DNA from selected clones was PEG purified as follows. To 50 ~1 of plasmid DNA in TE buffer (approximately 10-100 ng/yl), 2~
~l of SM
NaCI and 10 ~I 20% PEG (M.W.8,000; Fisher) was added, mixed well, and the mixture was incubated on ice for 1 hour. The mixture was then centrifuged for 5 min in a table-top microcentrifuge (at 14K rpm), the pellet was removed and an additional 15 ~.l of 20% PEG
was added to the supernatant. After incubation for 1 hour on ice, a second pellet was collected by centrifugation, the supernatant was discarded, and the pellet was redissolved in 20 p,l H,O. Two microliters of PEG-purified plasmid DNA (approximately 100 ng) was used in cycle-sequencing reactions using the_fmol~' DNA Sequencing System (Promega.
Madison.
WI) according to manufacturer protocol, in conjunction with the HCV46 or HCV308 primers.
The HCV46 or HCV308 primers were biotinylated at the 5' end using Oligonucleotide Biotin ' Labeling kit (Amersham, Arlington Heights, IL) prior to use in the sequencing reactions.
Sequencing reactions were separated on 10% denaturing acrylamide gel.
transferred on nylon membrane and visualized as described in Example 19.
Alternatively, DNA sequencing was done using either the Blue-Tl [5'-GATCTAC
TAGTCATATGGAT-3' (SEQ ID N0:106)] and Blue-T2 [5'-TCGGTACCCG
GGGATCCGAT-3' (SEQ ID N0:107)] primers labeled at the 5' end with tetra chloro fluorescein (TET) dye (Integrated DNA Technologies). In this case, the sequencing reactions were separated on a 10% denaturing acrylamide gel and the products were visualized using a FMBIO-100 Image Analyzer (Hitachi). The six HCV clones were termed HCV1.1, HCV2.1, HCV3.1, HCV4.2, HCV6.1and HCV7.1; the double-stranded DNA sequence of these clones are listed in SEQ ID NOS:108-113, respectively. The sequence of the sense strand for each of the six HCV clones is shown as the top line in SEQ ID NOS:108-113. The sequence of the anti-sense strand for HCV clones HCVl.I, HCV2.1, HCV3.1, HCV4.2, HCV6.1 and HCV7.1 is listed in SEQ ID NOS:114-118, respectively.
The DNA sequences of each of the six HCV clones are aligned in Figure 76. In Figure 76, nucleotides which represent variations between the six HCV clones are-indicated by bold type and underlining; dashes are used to indicate gaps introduced to maximize alignment between the sequences (necessary due to the insertion found in clone HCV4.2).
This alignment shows that these six HCV clones represent six different HCV
genotypes.
HCVI.I represents a genotype lc HCV; HCV2.1 represents a genotype la HCV;
HCV3.1 represents a genotype lb HCV; HCV4.2 represents a genotype lc HCV; HCV6.1 represents a genotype 2c HCV and HCV7.1 represents a genotype 3a HCV. For one sample, HCV4.2. an insertion of an "G" nucleotide was found at position 146 (relative to the protypical HCV;
Altamirano et al., supra), since no insertion or deletions in the HCV NCR have been previously reported, a second independent clone derived from the PCR products corresponding to HCV4 was sequenced. This second HCV4 clone was found to have the same sequence as that shown for HCV4.2 in Figure 76. ' B) Preparation of HCV Substrates Six double stranded substrate DNA were prepared for analysis in the CFLP~'~'1 reaction.
The substrates were labelled at the 5' end of either the sense or the anti-sense strand by the use of labeled primers in the PCR to permit CFLP~'~"'' analysis of each strand of the HCV
a 5 DNA substrates.
r To prepare PCR products for CFLP~'~"'' analysis, the HCV46 and HCV308 primers were 5' end labeled with TMR dye using the ONLYT"' BODIPY"~ TMR Oligonucleotide Phosphate Labeling Kit (Molecular Probes, Inc., Eugene, OR) according to manufacturer protocol. All six HCV 281 by NCR sequences were PCR amplified using 10 ng of template and 30 cycles of amplification as described above in section a).
For sense strand analysis, the PCR was conducted using the HCV46 primer (SEQ
ID
N0:104) labeled with TMR and unlabeled HCV308 primer (SEQ ID NO:105). For antisense analysis, the PCR was conducted using unlabeled HCV46 primer (SEQ ID N0:104) and HCV308 primer (SEQ ID NO:105) labeled with TMR. The: PCR products were purified by I S electrophoresis on a 6% denaturing acrylamide gel and eluted overnight as described above in Example 19. The gel-purified DNA substrates were redissolved in 20 pl HBO at an approximate concentration of 100 fmoles/~1.
C) Cleavage Reaction Conditions Cleavage reactions comprised 1 ~,1 of TMR-labeled I'CR products (approximately fmoles of the double-stranded substrates) in a total volume of 10 ~.I 10 mM
MOPS, pH 7.5;
with 0.5% each Tween 20 and NP-40 and 10 ng Cleavase~ BN enzyme. All components except the MnCh were assembled in a volume of 8 ~1. The: reactions were heated to 95°C for 15 seconds to denature the substrates and then quickly cooled to 55°C.
The reaction were performed in either a thermocycler (MJ Research, Waterto~~n, MA) programmed to heat to 95°C for 15 seconds and then cool immediately to 55° C or the tubes were placed manually in a heat block set at 95°C and then transferred to a second heat block set at 55°C.
Once the tubes were cooled to the reaction temperature of 55°C, the cleavage reaction was started by the addition of 2 p.l of 1 mM MnCI,. After 2 minutes at 55°C, the reactions were stopped by the addition of 5 ~I of a solution containing 95% formamide.
10 mM EDTA
and 0.02% methyl violet.
i Five microliters of each reaction mixture were heated at 85°C for 2 min, and where than resolved by electrophoresis through a 12% denaturing polyacrylamide gel ( 19:1 cross link) with 7M urea in a buffer of O.SX TBE. The gels were run at 33 watts for 1.5 hours.

The labeled reaction products were visualized using the FMBIO-100 Image Analyzer (Hitachi), with the resulting fluoroimager scan shown in Figure 77.
In Figure 77, the CFLPTMpatterns produced by cleavage of the six HCV samples labeled on the sense strand are shown in lanes 1-G; the CFLPT~'' patterns produced by cleavage ' of the six HCV samples labeled on the anti-sense strand are shown in lanes 7-12. The position of molecular weight markers is indicated on the left-hand side of the fluoroimager scan by the large arrowheads; the size of the markers is indicated in nucleotides.
The experiment presented in Figure 77 demonstrates the ability of CFLPT"' to differentiate six distinct hepatitis C viral subtypes. The six samples in the left hand side of the panel (lanes 1-G) were labeled on the 5' end of the sense strand; the six on the right (lanes 7-12), on the 5' end of the antisense strand. The first four samples in each set all contain samples amplified from HCV type 1. Subtypes a, b, and c are represented, as is a single base deletion of type 1 c (i. e., Q 1 c). Analysis of either strand points out numerous similarities as well as several distinctive differences between the subtypes. Most notable among the similarities on the sense strand are prominent bands marked A, B and C.
Specifically, whereas bands B and C are evident in the patterns generated from both subtypes I a and 1 b (and are, in fact, more prominent in subtype 1 b than in 1 a), they are barely visible in subtype 1 c. Band A, though present in all 4 of these samples, is more prominent in the patterns generated from subtypes 1 c and 1 a. Differences between subtypes 2c and 3 a vs. all of the subtype 1 samples, are evident in the region between 50 and 100 nt (compare bands D and E) on the sense strand and between the 80 and 150 nt on the antisense strand (compare bands F-J). Viral type 2 gives rise to the most significantly altered CFLPTM pattern, while type 3 appears to be similar to type 1; these relationships appear to be consistent with the relative number of sequence differences between the different isolates.
The results shown in Figure 77 demonstrate that the CFLPT"'' method provides a simple and rapid method to determine the genotype of HCV strains. This method will facilitate the diagnosis of HCV infection, permit appropriate treatment of HCV-infected patients, and aid in the monitoring of HCV outbreaks.
P

Detection of Mutations Associated With Antiobiotic Resistance in Mvcobacterium tuberculosis ' In the past decade there has been a tremendous resurgence in the incidence of ' tuberculosis in this country and throughout the world. Worldwide, the number of new cases reported annually is forecast to increase from 7.5 million in 1990 to 10.2 million by the year 2000. An alarming feature of this resurgence in tuberculosis is the increasing numbers of patients presenting with strains of M. tuberculosis which are resistant to one or more antituberculosis drugs [i.e., mufti-drug resistant tuberculosis (MDR-TB)].
Resistance to either or both of the antibiotics rifampin (rift and isoniazid (inh) is the standard by which M. tuberculosis strains are judged to be mufti-drug resistant. Both because of their potent bactericidal activities and because acquisition. of primary resistance to these drugs is rare (the spontaneous mutation rate of resistance to rifampin is approximately 10-h and to isoniazid, 10-8 to 10-9), until very recently, these two antibiotics were among the most powerful front-line drugs used to combat the advance and spread of tuberculosis. However surveys of tuberculosis patients in the U.S. reveal that as many as one-third were infected with strains resistant to one or more antituberculosis drugs; greater than 25%
of the M.
tuberculosis cultures isolated were resistant to isoniazid and 19% were resistant to both isoniazid and rifampin [Frieden et al., New Eng. J. Med. 328:521 (1993)].
As discussed above (Description of the Invention), resistance to rifampin is associated with mutation of the rpoB gene in M. tuberculosis. While the exact mechanism of resistance to isoniazid is not clear, the majority (as many as 80%) of inh' mutations occur in the katG
and inhA genes of M. tuberculosis. To investigate whether CFLP'~"'' could be used to detect mutations in the genes involved in MDR-TB, DNA fragments were amplified from the yoB
and katG genes of M. tuberculosis. DNA fragments derived from wild-type (i.
e., antibiotic-sensitive) or mutant (i.e., antibiotic-resistant) strains of M. tuberculosis were subjected to CFLPTM analysis.
A) CFLPTn' Analysis of Mutations in the RpoB Gene of M. tuberculosis i) Generation of Plasmids Containing RpoB Geue Sequences Genomic DNA isolated from wild-type M. tuberculosis or M. tuberculosis strains containing mutations in the rpoB gene associated with rifarnpin resistance were obtained from Dr. T. Schinnick (Centers for Disease Control and Prevention. Atlanta, GA).
The rifampin resistant strain #13 (91-3083) contains a tyrosine residue at codon 451 of the rpoB gene in place of the histidine residue found in the wild-type strain (i. e., H451 Y);
this mutation is is present in 28% of rifampin resistant TB isolates. The H451 Y mutation is hereinafter referred to as mutant 1. The rifampin resistant strain #56 (91-2763) contains a Ieucine residue at codon 456 of the rpoB gene in place of the serine residue found in_the wild-type strain (i.e., ' S456L); this mutation is present in 52% of rifampin resistant TB isolates. The mutation is hereinafter referred to as mutant 2.
A 620 by region of the TB rpoB gene was amplified using the PCR from DNA
derived from the wild-type and mutant 1 and mutant 2 strains. The primers used to amplify the rpoB gene sequences v~ere PoIB-SA [S'-ATCAACATCCGGCCGGTGGT-3' (SEQ ID
N0:120] and PoIB-SB [5'-GGGGCCTCGCTACGGACCAG-3' (SEQ ID N0:121 )]; these PCR primers amplify a 620 by region of the rpoB gene which spans both the H451 Y and S456L mutations [Miller et al., Antimicrob. Agents Chemother., 38:805 (1994)].
The PCRs were conducted in a final reaction volume of 50 q.I containing the PoIB-SA and PoIB-SB
primers at 1 ~.M, 1X PCR buffer and 60 ~M of all four dNTPs. The reaction mixture was heated at 95°C for 3 min.- Amplification was started by the addition of 2.5 units of Tcrcf polymerase and was continued for 35 cycles at 95°C for 1 min, 60°C for 1 min and 72°C for 2 min.
To clone the PCR-amplified fragments, 1 ~.I of each PCR product was used for ligation in the linear pT7Blue T-vector (Novagen, Madison,WI). The ligation products were used to transform competent JM109 cells and clones containing pT7Blue T-vector with an insert were selected by white color on LB plates containing 40 ~.g/ml X-Gal, 40 ~g/ml IPTG
and 50 ~.g/ml ampicillin. For each PCR sample (i.e., wild-type and mutants 1 and 2), five independent colonies were picked and grown overnight in 2 ml of LB media containing 50 qg/ml carbenicillin. Plasmid DNA was isolated using the alkaline miniprep protocol described above in Example 32.
To analyze the cloned fragments, 1 ~.l of plasmid DNA from each clone was amplified by PCR using 50 ~.1 reaction containing. the PoIB-SA and PoIB-SB primers at 1 q.M. 1X PCR
buffer, 60 q.M of all 4 dNTPs and 2.5 units of-Taq.polymerase. The PCRs were cycled 35 times at 95°C forl min, 60°C for 1 min and 72°C for 2 min. The PCR products were separated by electrophoresis on a 6% native polyacrylamide gel in O.SX TBE
buffer and clones that gave rise to a 620 by fragment were selected for further analysis.
For sequencing purposes, plasmid DNA from selected clones was PEG-purified as described in Example 32. Two microliters of PEG-purified plasmid DNA
(approximately 100 ' ng) was used for cycle-sequencing with fmol~ kit (Promega, Madison, WI) in conjunction with the PoIB-SA and PoIB-SB primers containing a biotin moiety at the ~' end.
Biotinylation of the primers was performed using an Oligonucleotide Biotin Labeling kit (Amersham). Sequencing reactions were separated in a 8% denaturing polyacrylamide gel, transferred to a nylon membrane and visualized as described above in Example 19. The DNA
sequences of the 620 by rpoB gene fragment derived from l:he wild-type, mutant 1 and mutant 2 strains are listed in SEQ ID NOS:122-124. The sequence of the sense strand for each of the three TB strains is shown as the top line in SEQ ID NOS:122-124. The sequence of the anti-sense strand for the wild-type, mutant 1 and mutant 2 TB strains is listed in SEQ ID
NOS:125-127, respectively.
ii) Preparation of M. tuberculosis rpoB Gene Substrates In order to generate substrates for use in CFLPT"'' reactions, the cloned 620 by fragment derived from the wild type and mutants 1 and 2 rpoB gene were amplified using the PCR. The PCRs were conducted using one primer of the primer pair labeled at the 5' end so that the resulting PCR product would permit the analysis of either the sense or anti-sense strand of the rpoB gene fragments. In order to generate substrates labelled on the anti-sense strand, ten nanograms of plasmid DNA from the sequenced clones was used as the template in 50 Ld reactions containing 1 ~.M of each the PoIB-SA primer (unlabelled) and PoIB-SB primer biotinylated at the S' end using Oligonucleotide Biotin Labeling kit (Amersham), 1X PCR
buffer, 60 q.M of all 4 dNTPs and 2.5 units of Taq polymerase. The reactions were cycled 35 times at 95°C for 1 min, 60°C for 1 min and 72°C for 2 min. The resulting 620 by PCR
products containing a biotin-labeled antisense strand were gel-purified as described in Example 19. The purified fragments were dissolved in 20 pl HBO.
To generate substrates labelled on the sense strand of the 620 by fragment of ~poB
gene fragments (wild-type and mutants 1 and 2), the PCRs were conducted using 1 ~.M each PoIB-SA primer 5' end labeled with TMR dye using ONLY~'~'~' BODIPY~" TMR
Oligonucleotide Phosphate Labeling Kit (Molecular Probes, Inc., Eugene, OR) and unlabeled ° PoIB-SB primer. The PCR reactions also contained 1X PCR buffer, 60 q.M of all 4 dNTPs, 5 units -of Taq polymerase and 10 ng of plasmid DNA from the sequenced clones as a template " 30 in a final volume of 100 ~,1. The reactions were cycled for a total of 35 cycles comprising 95°C for 1 min, 60°C for 1 min and 72°C for 2 min.
In addition to the above PCR conditions, the PCR reactions were also conducted using dUTP in place of dTTP to generate uridine-containing PCR fragments. Uridine-containing PCR fragments have become the standard type of PCR fragment analyzed in clinical laboratories. In order to demonstrate that uridine-containing PCRfragments can be used to produce distinct CFLP~'' patterns from substrates which vary by a single base pair change within a 620 by fragment, rpoB gene fragments containing a 5' TMR label on the sense strand and uridine in place of thymidine were generated as follows. Uridine-containing 620 by fragments (wild-type and mutants l and 2) were amplified according to the PCR protocol described above for the generation of fragments labelled at the 5' end of the sense strand with TMR -with the exception that 2.5 mM MgCI, was used in place- of 1.5 mM MgCh and 100 p.M dATP, 100 p.M dCTP, 100 p.M dGTP and 200 p.M dUTP were used in place of the mixture containing 60 p.M each of all 4 dNTPs (i.e., dATP, dCTP, dGTP and dTTP).
The 620 by PCR products containing a TMR-labeled sense strand (either uridine-or thymidine-containing) were purified in 6% denaturing gel as described above, eluted overnight, precipitated with ethanol and redissolved in 20 p.l HBO as described in Example 19, for a concentration of approximately 15 fmoles/~l.
iii) Cleavage Reaction Conditions -Cleavage reaction conditions for analysis of the 620 by rpoB fragments containing a biotin-labelled antisense strand were as follows. Six microliters of biotin labeled PCR product were combined with 1 p,l of lOX CFLP~'~"'' buffer (100 mM MOPS, pH 7.5, 0.5%
each Tween and NP-40) and 25 ng of the Cleavase~ BN enzyme. Prior to the initiation of the 20 cleavage reaction, the DNA mixtures were denatured by incubation at 95°C for 10 sec. The reactions were then cooled to 60°C and reaction was started by the addition of 1 p,l of 2 mM
MnCI,. The cleavage reactions were conducted at 60°C for 2 min.
Cleavage reactions were stopped after 2 min. by adding 5 p.l of stop buffer. Six microliters of each sample were resolved by electrophoresis on a 6% denaturing polyacrylamide gel and labeled fragments were visualized as described in Example 19. The resulting autoradiogram is shown in Figure 78.
In Figure 78, the lane marked "M" contains biotinylated molecular weight markers obtained from Amersham (Arlington Heights, IL) and include bands corresponding to lengths of 200, -300, 400, 500 nucleotides. The size of the markers and of the uncleaved yoB
substrates (620) is indicated on the left-hand side of the autoradiograph using large , arrowheads. Lanes 1-3 contain the reaction products generated by the cleavage of the mutant l, wild-type and mutant 2 substrates labelled on the anti-sense strand, respectively. The distance of the point mutation (relative to the wild-type sequence) from the 5' end label was 511 nucleotides for the mutant 1 substrate and 49~ nucleotides for the mutant 2 substrate.
- The results shown in Figure 78 demonstrate that similar, but distinctly different patterns of cleavage were generated from the each of the rpoB substrates labelled on the anti-sense strand. In comparison with the cleavage pattern generated by the wild-type substrate, ' the pattern generated by cleavage of the mutant 1 substrate shows a disappearance of Band A.
A comparison of the pattern generated by cleavage of the wild-type and mutant 2 substrates shows that the mutant 2 substrate has a significant reduction of intensity of Band B. Thus, the two mutants can be distinguished from the wild-type and from each other.
Cleavage reaction conditions for analysis of the 620 by rpoB fragments containing a TMR-labelled sense strand were as follows. Four microliters of TMR-labeled PCR
product were cleaved as described above. Cleavage reactions were stopped after 2 min.
by adding 5 p.l 95% formamide, 10 mM EDTA and 0.02% of methyl violet (Sigma).
The reactions were heated to 85°C for 2 min. and five microliters of each reaction mixture were resolved by electrophoresis through a 12% denaturing polyacrylamide gel ( 19:1 cross link) with 7M urea in a buffer containing O.SX TBE. The gel was run at 33W (watts) for 1.5 hours. The labeled reaction products were visualized using the FMBIO-100 Image Analyzer (Hitachi) with the resulting fluoroimager scan shown in Figure 79, Panel A. the gel was then electrophoresed for another 1 hour, and the second scan is shown in Panel B.
In Figure 79, two panels, A and B, are shown. Panel B represents a scan of the same gel shown in Panel A following a longer period of electrophoresis than that shown in Panel A. Thus, Panel B serves to spread out the banding pattern seen in the upper portion of Panel A (lines connecting Panels A and B show the region of expansion). In Figure 79, Panels A
and B, lanes 1-4 contain the reaction products produced by cleavage of thymidine-containing substrates having a TMR-label on the sense strand derived from the mutant 1, wild-type.
mutant 2 and a mixture of the wild-type and mutant 2 substrates, respectively.
Lane 5 of Panels A and B contains the 157 by fragment derived from exon 4 of the tyrosinase gene (SEQ ID N0:27) labeled with TET as a marker. Lanes 6-9 of Panels A and B
contain the reaction products produced by cleavage of uridine-containing substrates having a TMR-label on the sense strand derived from the mutant 1, wild-type, mutant 2 and a mixture of the wild-type and mutant 2 substrates. respectively. Mixtures of the wild-type and mutant 2 substrates (lanes 4 and 9) were generated by mixing together 5 ~.l of each substrate after the cleavage reaction; 6 ~1 of the mixture was then loaded on the gel. The distance of the point mutation WO 96/15267 pC"T~JS95/14673 (relative to the wild-type sequence) from the 5' end label was 100 nucleotides for the mutant 1 substrate and 116 nucleotides for the mutant 2 substrate.
The results shown in Figure 79 demonstrate that similar, but distinctly different patterns of cleavage were generated from the each of the rpoB substrates labelled on the sense ' strand. The left hand set of each panel contains CFLPTM patterns generated from PCR .
products containing dNTPs, while the right hand side contains CFLPTM patterns generated from PCR products in which dUTP was substituted for dTTP. Comparison of the CFLPTM
patterns generated from dNTP-containing amplicons of mutant 1 and wild-type reveals a marked reduction in intensity of a band approximately 80 nt from the labeled ~' end (band A), in the vicinity of the sequence change in this mutant (100 by from the labeled 5' end). In addition, a band migrating at approximately 200-250 nt from the labeled 5' end (band B) is missing in mutant 1. In contrast, comparison of the patterns generated from wild-type and mutant 2 reveals the loss of a band 120 nt from the labeled 5' end (band C).
Furthermore, examination of the region of the gel corresponding to 120 nt shows, particularly in Panel B, that band D is shifted downward in mutant 2 relative to wild-type. In Panel B, another band, migrating just above band D (labeled band D') also appears to be shifted downward in mutant 2 relative to wild-type. Lane 4 of each panel, in which aliquots from the wild-type and mutant CFLP reactions were mixed prior to electrophoresis demonstrates that this shift (in band D') in mutant 2 is real and not due to an electrophoresis artifact.
Examination of the CFLPTM patterns generated from the dUTP-containing amplicons demonstrates that the ability to distinguish these mutants from one another, as well as from the wt, is not adversely affected by substitution of dUTP for dTTP and may, in fact, be enhanced. In this example, both mutants 1 and 2 are more readily distinguished from the w~t when the patterns are generated from amplicons containing dUTP than dTTP. In the right-hand portion of panel A, comparison of the lanes containing mutant 1 and wt reveals several distinctive differences between the two amplicons, while others are new and unanticipated.
Specifically, band A is reduced in intensity in the mutant, as compared to the wt, in much the same way that it is in the left-hand portion of this panel. A band migrating at approximately -110 nt (band E) appears to be missing from the mutant, as does a band at approximately 250 nt (compare to band B in the left-hand portion of the gel). In addition, the strong band labeled F, while not noticeably different in the three samples containing dTTP, is much ' stronger in the wt pattern generated from dUTP-containing amplicons than it is in the mutants. Comparison of the patterns- generated from wt and mutant 2 also reveals a number ' PGT/US95/146_73 of pronounced differences. Most notably, a band migrating at approximately 60 nt appears in mutant 2 (band G), as does a complex of 2 new bands migrating at approximately 150 nt (band H). Interestingly, while some of the elements that make each of these patterns distinct from one another are different if dUTP is substituted for dTTP in the PCR
amplification, the vast majority of the cleavage fragments are identical in the two experiments.
This result suggests that substitution of dUTP results in subtle alterations in the single-stranded DNA
substrate which may be the result of altered stability of secondary structures or an altered affinity of the CleavaseT"' enzyme for secondary structures containing modified nucleotides.
These differences in CleavaseTM-based recognition of secondary structures in DNA fragments containing dUTP provides an unexpected benefit of using; this nucleotide substitution.
B) CFLPT"' Analysis of Mutations in the KrntG Gene of M. tuberculosis Generation of Plasmids Containing Kate Gene Sequences Genomic DNA isolated from wild-type M. tuberculosis or M. tuberculosis strains containing mutations in the katG gene associated with isoniazid resistance were obtained from Dr. J. Uhl (Mayo Clinic, Rochester, MN). These four strains are termed wild-type, S315T, R463L and S315T;R463L [Cockerill, III et al, J. Infect. Dis. 171:240 (1995).
Strain S315T
contains a G to C mutation in codon 315 of the wild-type katG gene. Strain R463L contains a G to T mutation in codon 463 of the wild-type gene and strain S315T;R463L
contains both the G to C mutation in codon 315 and the G to T mutation in codon 463.
A 620 by region of the M. tuberculosis katG gene was amplified using the PCR
from DNA derived from the above four strains. The primers used to amplify the katG
gene sequences were Kat~04 [5'-AGCTCGTATGGCACCGGAAC-3' (SEQ ID N0:128) and Kate 1523 [5'-TTGACCTCCCACCCGACTTG-3' (SEQ ID N0:129)]; these primers amplify a 620 by region of katG gene which spans both the S315T and R463L mutations.
The PCRs were conducted in a final reaction volume of 100 ~,l and contained the KatG904 and KatG1523 primers at 0.5 ~.M, 1X PCR buffer, 60 pM of all 4 dNTPs. The reaction mixtures were heated at 95°C for 3 min, then amplification was started with addition of ~ units of Taq polymerase and continued for 35 cycles at 95°C for 1 min, 60°C
for 1 min and 72°C for 2 min.
To clone the PCR-amplified katG fragments, 1 p.l of each PCR product was used for ligation into the linear pT7Blue T-vector (Novagen, Madison,WI). The ligation products were used to transform competent JM109 cells and clones containing pT7Blue T-vector with an insert were selected by white color on LB plates containing 40 p,g/ml X-Gal.
40 ~g/ml IPTG

and 50 p.g/ml ampicillin. For each of the four PCR samples, four colonies were picked and grown overnight in 2 ml LB media containing 50 p,g/ml carbenicillin. Plasmid DNA was isolated using the alkaline miniprep protocol described in Example 32.
To analyze the cloned katv fragments, 1 p.l of plasmid DNA from each clone was S amplified by PCR using 100 p.l reactions containing the KatG904 and Kate 1523 primers at , 0.5 p.M IX PCR buffer, 60 pM of all 4 dNTPs and 5 units of Tag polymerase. The PCRs were cycled 35 times at 95°C for 1 min, 60°C for 1 min and 72°C for 2 min. PCR products were separated by electrophoresis on a 6% native polyacrylamide gel in O.SX
TBE buffer and clones that gave rise to a 620 by fragment were selected for further analysis.
For sequencing purposes, plasmid DNA from selected clones was PEG-purified according to the protocol described in Example 32. Two microliters of plasmid DNA
(approximately 100ng) was used for cycle-sequencing with fmolR kit (Promega, Madison. WI) in conjunction with the KatG904 and KatG1523 primers containing a biotin moiety at the 5' end. Biotinylation of the primers was performed using an Oligonucleotide Biotin Labeling kit I S (Amersham). Sequencing reactions were separated in a 8% denaturing polyacrylamide gel.
transferred to a nylon membrane and visualized as described above in Example 19. The DNA
sequences of the 620 by katG gene fragments from the wild-type and mutant strains S315T.
R463L and S315T;R463L are listed in SEQ ID NOS:130-133, respectively. The sequence of the sense strand for each of the four katG gene fragments is shown as the top line in SEQ ID
NOS:130-133, respectively. The sequence of the anti-sense strand of the 620 by katG gene fragments from the wild-type and mutant strains S315T, R463L and S315T;R463L
is listed in SEQ ID NOS:134-137, respectively.
ii) Preparation of M. tuberculosis Kate Gene Substrates In order to generate substrates for use in CFLPT"'' reactions, the cloned 620 by fragments derived from the wild-type and S315T, R463L and S315T;R463L M.
tuherculosi.s strains were amplified using the PCR. The PCRs were conducted in a final reaction volume of 100 p.l and contained 0.5 p.M each KatG904 and Kate 1523 primers, I X PCR
buffer. 60 mM of all 4 dNTPs, 5 units of Taq polymerase and 10 ng of plasmid DNA from the sequenced clones as a template. The reactions were cycled 35 times at 95°C for 1 min. 60'C
for 1 min and 72°C for 2 min. ' To obtain 620 by PCR fragments of the katG gene having a biotin label on the sense strand. and unlabeled KatG1523 primer (SEQ ID N0:129) and S"-biotinylated KatG904 primer (SEQ ID N0:128) was used in the PCR; biotinylation was achieved using the Oligonucleotide Biotin Labeling kit (Amersham). To produce the same fragments having the TMR label on the antisense strand, unlabeled KatG904 (SEQ ID N0:128) and TMR-labeled KatG1523 (SEQ ID N0:129) primers were used in the PCR. Amplified PCR products were purified on a 6% denaturing gel, eluted overnight, precipitated with ethanol and redissolved in s.
50 p,l H,O as described in Example 19.
iii) Cleavage Reaction Conditions The cleavage reaction conditions for analysis of Iu~tG substrates labelled on the sense strand were as follows. Five microliters of biotin labeled PCR product were combined with 1 p,l of lOX CFLPTM buffer (100 mM MOPS, pH 7.5, 0.5°ro each Tween 20 and NP-40) and 25 ng Cleavase~'~"'' BN enzyme. Prior to the initiation of the cleavage reaction, the DNA
mixtures were denatured by incubation at 95°C for 10 sec. The reactions were then cooled to 50°C and the reaction was started by the addition of 1 p,l of 2 mM
MnCh. The cleavage reactions were incubated for 2 min. at 50°C and were stopped by adding 5 p.l of stop buffer.
Four and one-half microliters of each sample were run on a 10% denaturing polyacrylamide 1 S gel and labeled fragments were visualized following transfer to a nylon membrane as described in Example I9. The resulting autoradiogram is shown in Figure 80.
In Figure 80, lanes marked "M" contain biotinylated molecular weight markers obtained from Amersham (Arlington Heights, IL) and include bands corresponding to lengths of S0, I00, 200, 300. 400, 500, 700, and 1000 nucleotides; the size of the markers is indicated by the use of large arrowheads. Lanes 1-4 contain the reaction products obtained by incubating the R463L. R463L;S315T, S315T and wild-type katG substrates in the presence of Cleavase'~"'' BN enzyme, respectively. The mutation distance from the 5' end label is 485 nucleotides for the R463L mutation and 41 nucleotides for the S315T mutation when the label is present on the sense strand.
The results shown in Figure 80 demonstrate that similar, but distinctly different patterns of cleavage were generated from the wild-type and S315L mutant (seen in both the S3I ST and S3 I5;R463L substrates) katG substrates labelled on the sense strand. Comparison ' of the CFLP'~"'' pattern for wild-type fragment (lane 4) shows that the S315T mutation ( seen in both mutants R463L:S315T and S315T; lanes 2 and 3) results in disappearance of Band B
which is located around 40 nucleotides from the end label i.n the wild-type substrate. The disappearance of Band B correlates very well with the distance of S315T
mutation from the 5' end (41 nucleotides from the 5' end label on the sense strand). Subsequent experiments ' have demonstrated that the R463L mutant can be distinguished from wild-type by a mobiliti~
- 21~ -shift in a band migrating at approximately 500 nt from the S' end label on the sense strand (the band shifts downward in the R463L mutant), but is difficult to resolve in many gels systems.
The cleavage reaction conditions for analysis of katG substrates labeled on the anti-s sense strand were as described for the sense strand. Four and one-half microliters of each sample were run on a 10% denaturing polyacrylamide gel and labeled fragments were visualized using the Hitachi FMBIO-100 fluoroimager as described in Example 33(a)(iii).
The resulting scan is shown in Figure 81.
In Figure 81, lanes marked "M" contain plasmid pUCl9 DNA digested with MspI
and 3' end labeled with fluorescein ddUTP using terminal deoxynucleotidyl transferase as described in Example 8. This marker includes bands corresponding to lengths off 10/111, 147, 190, 242, 331, 404, 489 and 501 bp. Additional marker bands of 26, 34.
and 67 by are not visible in this figure; the size of the markers is indicated by the use of large arrowheads.
Lanes 1-4 contain the reaction products obtained by incubating the R436L.
S31~T:R463L.
S315T, and wild-type katG substrates in the presence of CleavaseT"'' BN
enzyme, respectively.
The location of the single base mutation from the 5' end label is 136 nucleotides for the R463L mutation and 580 nucleotides for the S315T mutation when the label is present on the anti-sense strand.
The results shown in Figure 81 demonstrate that wild-type can be distinguished from mutants containing the R463L substitution on the anti-sense strand. Comparison of the lanes containing the S315T;R463L double mutant or the R463L mutant by itself demonstrates that the R463L mutation is associated with the presence of a strong band migrating at approximately 130 nt (band A). This result, taken with that presented in Figure 80.
demonstrates that all three of these mutants can be distinguished from one another. as well as from wild type, by CFLP~'~"'' analysis.
The CFLP~'~"'' technology offers cost benefits by reducing gel electrophoresis processing time from 12-18 hours down to 5 to 10 minutes. Adapting the readout to mufti-lane Fluorescence Image Detectors allows for an expanded volume of work by allowing simultaneous processing of up to 48 reactions. The consequent decrease in turnaround time in performing the analyses reduces the turnaround time of reporting patient results from days to hours, or, as in the case of MDR-TB patients, from weeks to hours. Early detection of MDR-TB can save thousands of dollars per patient by reducing the expense of extended stays in isolation wards, spent while testing various antibiotic treatments for efficacy.

Rapid Identification of Bacterial Strains by(''Fr pT" ~awsis ' The results shown above demonstrated that CFLPT"~ analysis can be used to detect the Y Y
presence of wild-type and drug-resistant mutations of M. tuberculosis by examining portions of gene associated with drug resistance (e.g., rpoB and katG). In order to examine whether the CFLP~'~"' analysis could be used as a method of detecting and identifying a wide variety of -.
microorganisms, CFLP'~"' analysis was conducted using substrates derived from bacterial 16S
rRNA genes.
Bacterial 16S rRNA genes vary throughout the phylogenetic tree; these genes do contain segments which are conserved at the species, genus or kingdom level.
These features have been exploited to generate primers containing consensus sequences which flank regions of variability. These primers have been used to amplify segments of bacterial 16S rRNA
genes which are then characterized by either Southern blot hybridization [Greisen et al.. .I.
Clip. Microbiol. 32:335 (1994)] or SSCP analysis [Widjojoatmondjo et al., J.
Clin. Microhiol.
32:3002 (1994)]. These types of analysis, while faster than traditional culturing methods, are at best limited to the differentiation of species within a particular genus and higher bacterial taxons. However, it is often desirable to differentiate between different strains of the same species. For example, a given species may contain subspecies which comprise harmless as well as pathogenic organisms. In order to develop a technique which would allow the differentiation between species and/or subspecies, CFLPT"' analysis was applied to segments derived from bacterial 16S rRNA genes.
A) Bacterial Strains Table 3 below lists the bacterial strains used in this study. These strains were derived from the ATCC strains listed below with the exception of .Desulfurococcus amylolyticus Strain Z-533 which was derived from a deposit obtained from the Deutsche Sammlung von Mikroorganismen (DSM).
f ORGANISM STRAIN NO. CHARACTERISTICS

E. coli ATCC 11303 Strain B -E. coli ATCC 14948 Derived from E. coli strain E. coli Serotype 01~7:H7ATCC 43895 Produces Shiga-like toxins , I and II

Campylobacter jejuni ATCC 33291 Isolated from human stool subsp. jejuni Shigella dysenteriae ATCC 29027 Isolated from human stool Serotype 2 Salmonella choleraesui.sATCC 6539 Used for germicide testing subsp. choleraesuis Serotype typhi Staphylococcus aureusMethicillin-subsp. aureus ATCC resistant S. aureus subsp. aureusATCC 33592 Gentamicin- and methicillin-resistant S. aureus subsp. aureusATCC 13565 Produces enterotoxin A and large amounts of beta-hemolysin Staphylococcus hominisATCC 29885 Methicillin control for MIC
testing Staphylococcus warneriATCC 17917 Used for soap germicide testing Desulfurococcus STRAIN 3822 hermophilic archaebacterium T

amylolyticus The strains listed in Table 3 represent pathogenic microorganisms with the exception of E. coli strains B and K-12 and Desulfurococcus amylolyticus.
Desulfurococcus amylolyticus was included in this study to determine whether the consensus primers, whose design was based upon known rRNA gene sequences, could also be used to amplify rRNA
gene fragments sequences from archeabacterial species whose rRNA gene sequences have not been reported. The strains listed in Table 3 were selected to provide representatives from several different genera (e.~, Escherichia, Shigella, Salmonella, Campylobacter. etc.) as well as to provide several representatives of different species (or subspecies) within a given genus.
For example, three different strains of E. coli were chosen so that the consistency (or lack thereof) of the CFLP~ banding pattern generated by cleavage of an rRNA gene substrate could be examined between species within a given genus. In addition, E. coli Serotype 0157:H7 was examined as this strain has been implicated in hemorrhagic colitis outbreaks. It "_ was of interest to examine whether the CFLP~ pattern observed from clevage of a rRNA
gene substrate from E. coli strains B or K-12 differed from that produced by cleavage of a _ rRNA gene substrate from E. coli Serotype 0157:H7.

Table 4 below describes the phylogenic relationship between the strains used in this example.
.. TABLE 4 Phylogenetic Position of Strains from Prokaryotic Small SubUnit rRNA Taxonomic List' 1.2 CRENARCHAEOTA ' 1.2.1 CRENARCHAEOTA-GROUP-I

Desul furococcus amylolyticus 2.13 PURPLE-BACTERIA

2.13.3 GAMMA-SUBDIVISION

2.13.3.15 ENTERICS AND RELATIVES

2.13.3.15.2ESCHERICHIA-SALMONELL A ASSEMBLAGE

Escherichia coli Strain B

Escherichia coli Strain K-12-derived Escherichia coli Serotype 0157: H7 Shigella dysenteriae Seroty.pe 2 Salmonella choleraesuis subsp. choleraesuis Serotype typhi 2.13.5 EPSILON-SUBDIVISION

2.13.5.2 CAMPYLOBACTER AND RELATIVES

Campylobacter.jejuni subsp..jejuni 2.15 GRAM-POSITIVE PHYLUM

2.15.5 BACILLUS-LACTOBACILLUS-STREPTOCOCCUS

SUBDIVISION

2.15.5.10 STAPHYLOCOCCUS GROUP

2.15.5.10.2STAPHYLOCOCCUS SUBGROUP

Staphylococcus aureus subsp. aureus ATCC

Staphylococcus aureus subsp. aureus ATCC

Staphylococcus aureus subsp. aureus ATCC

Staphylococcus hominis Staphylococcus warneri ' Data derived from the Ribosomal Database Project; available on the Internet at http://rdp.life.uiuc.edu/index.html; Maidak et al. Nucleic Acids Res.. 22:348 ( 1994).
r WO 96/15267 PCTlUS95/14673 B) Growth of Microorganisms In order to minimize handling of the pathogenic strains, the microorganisms were grown on slant cultures or on plates rather than in liquid culture.
i) Growth of Escherichia, Shigella, Salmonella, and Staphvlococcus species All strains were derived from the ATCC strains listed above in Table 3 as follows. A ~, loopful of a culture previously frozen in Trypticase Soy Broth and 15%
glycerol (Remel Corp., Lenexa, KS. Cat. 06-5024) was subcultured onto a trypticase soy agar slant (Remel, Cat. 06-4860). The cultures were incubated overnight at 37°C.
ii) Growth of Campylobacter species A loopful of a culture previously frozen in Trypticase Soy Broth and 15%
glycerol (Remel Corp., Lenexa, KS. Cat. 06-5024) was subcultured onto Campylobacter Agar supplemented with 10% sheep blood, amphotericin B, cephalothin, trimethoprim.
vancomycin, and polymyxin B (BBL, Cat. 21727). Inoculated plates were sealed in Campy microaerophilic pouches (BBL, Cat. 4360656) and incubated at 42°C for 3 days.
C) Extraction of Genomic DNA from Microorganisms For each bacterial sample, 300 p.l of TE buffer and 300 ~I
phenol:chloroform:isoamyl alcohol (25:24:1 ) were placed in a 1.5 ml microfuge tube. This combination is referred to as the extraction buffer. A loopful (approximately 0.1 ml) of the desired bacterial strain was removed from a slant culture or plate and combined with the extraction buffer in a 1.5 ml microfuge tube and the contents were vortexed for two minutes. The extracted DNA present in the aqueous phase was processed for further purification as described below.
Samples of E. coli and C. jejuni strains were ethanol precipitated and dissolved in 50 ~.l TE buffer. The samples were then treated with 0.5 p.g RNase A at 37°C for 30 min.
DNA was precipitated with ethanol, collected by centrifugation and dissolved in 200 p.l 10 mM Tris-HCl (pH 8.0 at 25°C).
Samples of Shigella, Salmonella, and Staphylococcus strains were concentrated using a Microcon T"' 30 filter (Amicon) to 50 p.l and then transferred to TE buffer using MicrospinT"' v S-200 HR gel filtration columns (Pharmacia Biotech). The samples were then treated with 0.5 p.g RNase A at 37°C for 80 min. DNA was precipitated with ethanol, collected by centrifugation and dissolved in 200 pl 10 mM Tris-HCl (pH 8.0 at 25°C).
Genomic DNA of E. coli Strain B (ATCC 11303) was obtained from Pharmacia Biotech (Piscataway, NJ; Cat. 27-4566-O1, Lot 411456601 ). The DNA was dissolved in 10 mM Tris-HCl (pH 8.0 at 25°C).

Genomic DNA from Desulfurococcus amylolyticus Strain Z-533 (DSM 3822) was isolated and purified using the standard technique of cesium chloride centrifugation. [Bonch-Osmolovskaya, et al., Microbiology (Engl. Transl. of Milcrobiologiya) 57: 78 (1988)].
The concentration of the genomic DNA preparations was determined by measuring the 0 5 OD~~~, of the preparations.
D) Design of Primer for the Amplification of 16s rRNA Genes of Bacterial Species _ Primers and probes have been reported which allow the amplification or detection of 16S rRNA sequences from a wide variety of bacterial strains. These oligonucleotide primers or probes represent consensus sequences derived from a comparison of the 16s rRNA ene g sequences from a variety of eubacterial species. For example, oligonucleotide primers suitable for either PCR amplification or dot blot hybridization of bacterial rRNA gene sequences have been reported [e.g., PCT Publication WO 90/15157; Widjojoatmodjo et al., J.
Clin. Microbiol.
32:3002 (1994)]. Typically the conserved primer sequences are designed to flank nonconserved regions of the 16s rRNA gene with species-:specific sequences.
A number of previously published consensus primers derived from 16S rRNA gene sequences were examined for the ability to produce substrates for use in CFLPT"' reactions.
Primers 1638, 1659 and 1743 were described in PCT Publication WO 90/15157.
Primer ER10 was described in Widjojoatmodjo et al., supra. Primers SB-1, SB-3 and SB-4 represent new primers (i. e., not previously published). The primers used in this example are listed in Table 5 below.

Primers for PCR Amplification of 16S rRNA Genes PRIMER SEQ ID NO: SEQUENCE

1638 138 5'-AGAGTTTGATCCTGGCTCAG-3' ER10 139 5'-GGCGGACGGGTGAGTAA-3' 1659 140 5'-CTGCTGCCTCCCGTAGGAGT-3' SB-4 141 5'-ATGACGTCAAGTCATCATGGCCCTTACGA-3' 1743 142 5'-GTACAAGGCCCGGGAACGTATTCACCG-3' SB-I 143 5'-GCAACGAGCGCAACCC-3' SB-3 144 5'-ATGACGTCAAGTCATCATGGCCCTTA -3' The oligonucleotide primers were obtained from Integrated DNA Technologies, Inc.
The oligonucleotides were dissolved in 10 mM Tris-HCl (pH 8 at 25°C) at a concentration of 20 ~.M. Two sets of primers were synthesized; one set having an OH group at the 5' end (i.c., unlabelled primers) and the other set having the fluorescent dye TET
(tetrachlorinated analog of 6-carboxyfluorescein, Applied Biosystems) at the 5' end (i. c., TET-labelled primers). TET-labelled primers are indicated by the use of "TET" as a suffix to the primer name (for example, TET-1638 indicates the 1638 primer having a 5' TET label).
The location of each of the primers listed in Table 5 is shown along the sequence of the E. coli rrsE gene (encodes a 16S rRNA) in Figure 82. In Figure 82 the primer sequences are shown in bold type and underlining is used to indicate complete identity between primer sequences and E. coli rrsE gene sequences. The sequence of the E. coli rrsE
gene is listed in SEQ ID N0:145. As shown in Figure 82, the 1638, ER10, SB-1, SB-3, SB-4 primers 2~ correspond to sequences present on the sense strand of the 16S rRNA gene.
The 1659. 174 3 primers correspond to sequences present on the anti-sense strand of the 16S
rRNA gene. .
Figure 83 provides an alignment of the E. coli rrsE gene (SEQ ID N0:145). the Cam.jejun5 gene (a rRNA gene from C. jejuni) (SEQ ID N0:146) and the Stp.aureus gene (a rRNA gene from S. aureus) (SEQ ID N0:147). The location of the 1638. ER10, 169 ' (shown as the complement of 1659), SB-1, SB-3, SB-4 and 1743 (shown as the complement of 1743) primers is indicated by the bold type. Gaps (dashes) are introduced to maximize ' alignment between the rRNA genes.

PGT/US95/14673 _ In procaryotes the ribosomal RNA genes are present in 2 to 10 copies, with an average of 7 copies in Escherichia strains. Any PCR amplification produces a mixed population of these genes and is in essence a "multiplex" PCR from that strain. The CFLP
represents a ~Y
composite pattern from the slightly varied rRNA genes within that organism so no one S particular rRNA sequence is directly responsible for the entire "bar code."
In some cases these minor variations (between rRNA genes; see, for example, minor variations between the E. cnli rRNA gens in Figure 82) cause shifts in the minor (lower signal) bands in the CFLPTM~~
pattern, allowing discrimination between very closely related organisms. More dramatic sequence variations, found in most or all copies of these genes, are seen when more distantly related organisms are compared (see, for example, the extensive variations between the E.
coli, C. jejuni and S. aureus. rRNA genes in Figure 83) and these larger differences are reflected in the CFLP patterns as more dramatic pattern changes. Despite the variable nature of these genes, the amplification by PCR can be performed between conserved regions of the rRNA genes, so prior knowledge of the entire collection of rRNA sequences for any microbe I S of interest is not required.
Three primers (TET-1638, TET-ER10, and TET-SB-4) were used for making the ~' end fluorescently labeled fragments of the sense strand of 16S rRNA genes; two other primers (TET-1659 and TET-1743) were used for making labeled fragments of the antisense strands.
The predicted size of PCR products produced by amplification of 16s rRNA gene sequences from a variety of bacterial genera using the indicated primer pairs is shown in Table 6. In Table 6, the size of the predicted PCR product is based upon the known sequence of the 16S rRNA gene in the indicated species. The following abbreviations are used in Table 6: Dco (Desulfurocnccus); E.co (E. coli), Cam (Carr~pylobacter) and Stp (Staphylococus). The location of the PCR product relative to the sequence of the E. coli r ~ sE
gene (see Figure 82) is given in the last column.
r - 273 _ Combinations of Primers for PCR Amplification of 1 GS rRNA Sequences w Anti-Primer Sense Sense Labeled Size Position Pair Primer Primer Strand (bp) Dco E.co Cam Stp (E.co) A TET-1638 1659 sense 350 348 347 8-357 B TET-1638 1743 sense 1388 1365 1397 8-1395 C TET-ERIO 1659 sense 254 254 263 104-357 D TET-ER10 1743 sense 1278 1292 1271 1303 104-1395 E 1638 TET-1659 antisense 350 348 347 8-357 F ERIO TET-1659 antisense 254 254 263 104-357 G TET-SB-4 1743 sense 208 208 1188-H TET-1743 1638 antisense 1388 1365 1397 8-1395 1 TET-1743 ER10 antisense1278 1292 1271 1303 104-1395 J SB-4 TET-1743 antisense 208 208 1188-K SB-I TET-1743 antisense305 297 29G 296 1099-L SB-3 TET-1743 antisense 208 208 208 1188-E) PCR Amplification of 16S rRNA Gene Sequences The ability of each primer pair listed in Table 6 to amplify 1 GS rRNA gene sequences from each bacterial strain listed in Table 3 was examined. It is well known that commercial preparations of recombinant Taq DNA polymerase contain various amount of E.
coli I GS
rRNA gene sequences. In order to minimize amplification of contaminating E.
coli 16S
rRNA sequences during the amplification of bacterial DNA samples. AmpliTaq DNA
polymerase, LD (Low DNA) (Perkin Elmer) was used in the PCRs. This preparation of Tuq DNA polymerase is tested by the manufacturer to verify that less than or equal to 10 copies of bacterial 1GS ribosomal RNA gene sequences are present in a standard 2.5 unit aliquot of enzyme.

wo 9snsis7 PCT'/US95/14673 Each primer pair (Table 6) was tested in PCRs. The PCR reactions contained 10 mM
Tris-HCl (pH 8.3 at 25°C), 50 mM KCI, 1.5 mM MgCh, 0.001% w/v gelatin.
60 pM each of dGTP, dATP, dTTP, and dCTP, 1 pM each of one S'-TET labeled and one unlabeled ' primers, 2.5 units AmpliTaq DNA polymerise, LD. The reactions were conducted in a final volume of 50 p,l using AmpliTaq DNA polymerise, LD, Lot E0332, or 100 pl volume using AmpliTaq DNA polymerise, LD, Lot D0008. The amount of genomic DNA added varied from 6 to 900 ng per PCR. Control reactions which contained no input bacterial genomic DNA were also run to examine the amount of 16S rRNA product produced due to contaminants in the AmpliTaq DNA polymerise, LD preparations.
PCR reactions were performed on PTC-100'"'' Programmable Thermal Controller (MJ
Research, Inc.). Two sets of cycling conditions were utilized. The first set of conditions comprised 30 cycles of 95°C for 30 sec; 60°C for 1 min;
72°C for 30 sec; after the last cycle the tubes were cooled to 4°C. The second set of conditions comprised 30 cycles of of 95° for 30 sec; 60°C for 1 min; 72°C for 90 sec; after the last cycle the tubes were cooled to 4°C.
Thus, the difference between the two cycling conditions i s the length of time the reactions are held at the elongation temperature (72°C). These two elongation times were tested because the predicted size of the 16S rRNA targets varied from 208 to 1388 by depending on the primer pair used in the amplification.
As a rule of thumb, when the target to be amplified is less than 500 by in length, a 30 sec elongation step is used; when the target is about 500-1000 by in length, an elongation step of 30 to 60 sec is used; when the target is greater than 1 kb in length, the elongation is conducted for approximately 1 min per 1 kb length. While the first set of PCR
conditions (30 sec elongation step) worked with the longer amplicons, the yield was lower than that obtained when the second set of PCR conditions (90 sec elongation) was used.
Following the thermal cycling, 400 ~I of formamidf: containing 1 mM EDTA was added to each sample and the samples were concentrated to a volume of 40 pl in a Microcon 30. The samples (40 p,l) were loaded on a denaturing 6% polyacrylamide gel (7 M urea, O.SX TBE running buffer), that was prewarmed to 50-55°C prior to the loading of the samples. The samples were run at 20 W for 20 min (200-350 by fragments) or 40 min (more than 1 kb fragments). The gels were scanned using a Fluorescent Method Bio Image Analyzer Model 100 (FMBIO-100, Hitachi) with a 585 or 505 nm filter.
The results of these PCRs showed that each primer pair (Table 6) tested successfully amplified a fragment of the expected size. Thus the primer pairs shown in Table 6 are - 225 _ suitable for the amplification of end labeled DNA fragments using genomic DNA
from variety of prokaryotes including archaea, gram-positive and gram-negative bacteria, different species of the same genus and different strains of the same species. These PCRs also demonstrated that, although the amount of genomic DNA present in the PCR
varied from strain to strain, the yield of the amplified product was always many-fold higher than the trace yield of product from the E. coli genomic DNA present in AmpliTaq DNA
polymerase, LD, seen in the reactions which contained no input bacterial genomic DNA.
F) Preparation of 16S rRNA Gene Substrates To generate labelled PCR products corresponding to bacterial 16S rRNA
sequences for use in CFLPT"'' reactions, the following primer pairs were used in PCRs.
1. The SB-1/TET-1743 pair was used to amplify an approximately 297 by fragment from genomic DNA derived from Desulfurococcus amylolyticus (DSM
3822), E.
coli Strain K-12 (ATCC 14948), S. aureus subsp. aureus (ATCC 33591) and S.
aureus subsp.
ccacreus (ATCC 33592). The resulting PCR product contains a 5' TET-label on the antisense strand.
2. The TET-SB-4/1743 pair was used to amplify an approximately 208 by fragment from genomic DNA derived from E. coli Stain B (ATCC 11303), E. coli Strain K-12 (ATCC 14948), E. coli Serotype 0157: H7 (ATCC 43890, Shigella dysenteriac~
Serotype 2 (ATCC 29027), and Salmonella choleraesuis subsp. choleraesuis Serotype typhi (ATCC
6539). The resulting PCR product contains a 5' TET-label on the sense strand.
3. The 1638/TET-1659 pair was used to amplify an approximately 350 by fragment from genomic DNA derived from E. coli Stain B (ATCC 11303), E. coli Strain K-12 (ATCC 14948), E. coli Serotype 0157: H7 (ATCC 43895), Shigella dusenteriae Serotype 2 (ATCC 29027), and Salmonella choleraesuis subsp. choleraesui.s Serotype typhi (ATCC
6539). The resulting PCR product contains a 5' TET-label on the antisense strand.
4. The TET-ER10/1743 pair was used to amplify an approximately 1292 by fragment from genomic DNA derived from E. coli Strain K-12 (ATCC 14948) and Campylnbacter jejuni subsp. jejuni (ATCC 33291). The resulting PCR product contains a ~' TET-label on the sense strand.
5. The 16381T'ET-1659 pair was used to amplify an approximately 350 by o fragment from genomic DNA derived from E. coliSerotype 0157: H7 (ATCC 43895).
Salmonella choleraesuis subsp. clzoleraesuis Serotype typhi (ATCC 6539), Shigella dysenteriae Serotype 2 (ATCC 29027), S. aureus subsp. aureus (ATCC 33591 ). S.
crzrrcu.s PCTlUS95/14673 subsp. aureus (ATCC 33592), S. aureus subsp. aureus (ATCC 13565), S. hominis (ATCC
29885), and S. warneri (ATCC 17917).
The PCRs were conducted as described in section (e) above. Two separate PCR
sr reactions were performed using 0.2 pg of genomic DNA derived from Camylobacter.jejurri subsp. jejuni (ATCC 33291) and the TET-ER10/1743 primer pair. One reaction was ' conducted in a final volume of 50 ~.l and used an extension step of 30 sec at 72°C during thermal cycling. The second reaction was conducted in a final volume of 100 pl and used an extension step of 90 sec at 72°C. The yield of PCR product produced in the second reaction was 76% higher (as compared to first reaction). Following the amplification reaction, the samples were processed for electrophoresis on denaturing polyacrylamide gels as described in section (e) above. After electrophoresis, the desired bands were cut from the gel and eluted by placing the gel slice into 0.4 ml of a solution containing 0.5 M ammonium acetate, 0.1 mM EDTA and 0.1 % SDS. The mixture was then incubated at 55°C for 2 h and then at 37°C for 12 h. The samples were concentrated to 25 ~l using a Microcon 30 (Amicon) and transferred into water using S-200 microspin columns (Pharmacia).
G) Cleavage Reaction Conditions Cleavage reactions were conducted in a final volume of 10 ~.1 volume containing approximately 0.2 to 1 pmole (as indicated below) 5' TET'-labeled DNA
substrate, 10 ng CleavaseT'''' BN enzyme (Third Wave Technologies), 1X CFLP buffer and 0.2 mM
MnCI,.
The reactions were first assembled as a 9 p,l mixture lacking MnCh; this mixture was heated to 95°C for 10 sec and then cooled down to the desired incubation temperature (45°C, 50°C or 65°C). Optimal reaction temperature for each substrate was chosen based on even distribution of bands, and the presence of some undigested material to indicate representation of molecules all the way up to full length. Selected optimal temperatures for each substrate are indicated in the description of Figures 84-87 below.
The cleavage reaction was started by the addition of 1 ~.1 of 2 mM MnCI,.
Following incubation at the desired temperature for 2 min, the reaction was stopped by the addition of 10 ~,l of a solution containing 95% formamide, 5 mM EDTA, 5%
glycerol and 0.02% methyl violet. Uncut or "no enzyme" controls were set up for each substrate as described above with the exception that H,O was used in place of the CleavaseT"' BN enzyme.
Samples (approximately 4 to 8 ~l) were run on 6 to 12% denaturing polyacrylamide gels ( 19:1 cross link) with 7 M urea in a buffer containing 45 mM Tris Borate, pH
8.3, 1.4 mM
EDTA at I S to 20 W for 9 minutes (specific gel percentages are indicated below in the descriptions of Figures 84-87). The gels were then scanned using a FMBIO-100 (Hitachi) with the 585 nm filter.
The resulting fluoroimager scans are shown in Figures 84-87. In Figure 84. the cleavage products generated by cleavage of an approximately 297 by 16S rRNA
substrate generated using the SB-1/TET-1743 pair and genomic DNA derived from Desulfurococcus amylolyticus (DSM 3822), E. coli Strain K-12 (ATCC 14948), S. aureus subsp.'aureus (ATCC 33591) and S. aureus subsp. aureus (ATCC 33592) is shown. Lanes 1-4 contain the products generated by incubation of the substrate derived from Desu~rococczrs amylolyticus (DSM 3822), E. coli Strain K-12 (ATCC 14948), S. aureus subsp. aureus (ATCC
33591 ) and S. aureus subsp. aureus (ATCC 33592) in the absence of Cleavase~'~"'' BN
enzyme, respectively. Lanes 5-8 contain the products generated by incubation of the substrate derived from Desulfurococcus amylolyticus (DSM 3822), E. coli Strain K-12 (ATCC
14948), S.
aureus subsp. aureus (ATCC 33591) and S. aureus subsp. aureus (ATCC 33592) in the presence of CleavaseT"'' BN enzyme, respectively. The CFLPT"'' reactions were performed using approximately 1 pmole of each PCR product and the cleavage reactions were incubated at 50°C for 2 min. The cleavage products were resolved by electrophoresis on an 8%
polyacrylamide gel, as described above.
The results shown in Figure 84 demonstrate that distinct CFLPT"'' patterns are obtained using the Desulfurococcus amylolyticus (DSM 3822), E.. coli Strain K-12 (ATCC
14948) and S. aureus subsp. aureus substrates. The same CFLPT"'' pattern was generated by cleavage of the two S. aureus subsp. aureus substrates (lanes 7 and 8); these two S.
aurezrs subsp. azrreus strains (ATCC 33591 and 33592) are considered different subspecies based upon differences in sensitivities to the antibiotics methicillin and gentamicin. Resistant or sensitivity to these antibiotics is not associated with mutation in the 16S rRNA gene; therefore it was not expected that different CFLPT"'' patterns would be observed using a 16S rRNA
substrate.
The results shown in Figure 84 show that the SB-1/TET-1743 pair can be used to generate substrates for CFLPT"' analysis which allow the identification and discrimination of Desulfurococcus amylolyticus (DSM 3822), E. coli Strain K-12 (ATCC 14948) and .S: aureus -subsp. aureus. -In Figure 85, Panel A shows the reaction products generated by cleavage of an approximately 208 by 16S rRNA substrate generated using the TET-SB-4/1743 pair and genomic DNA derived from E. c_oli Stain B (ATCC 11303). E. coli Strain K-12 (ATCC
14948), E. coli Serotype 0157: H7 (ATCC 43890, Shigella dyserrteriae Serotype 2 (ATCC

PCT'/US95l14673 29027), and Salmonella choleraesuis subsp. choleraesuis Serotype typhi (ATCC
6539). The TET-SB-4/1743 pair amplifies a portion of the 165 rRNA gene located in the 3' region of the gene (see Figure 82).
h_ The CFLPTM reactions shown in Figure 85, Panel A were performed using s 5 approximately 0.7 pmole of each PCR product and the cleavage reactions were incubated at 50°C for 2 min. The cleavage products were resolved by electrophoresis on an 8% denaturing polyacrylamide gel, as described for Figure 84.
In Figure 85, Panel B shows the reaction products generated by cleavage of an approximately 350 by 165 rRNA substrate generated using; the 1638/TET-1659 pair and genomic DNA derived from E. coli Stain B (ATCC 11303), E coli Strain K-12 (ATCC
14948), E. coli Serotype 0157: H7 (ATCC 43895), Shigella dysenteriae Serotype 2 (ATCC
29027), and Salmonella choleraesuis subsp. choleraesuis Serotype typhi (ATCC
6539). The 1638/TET-1659 pair amplifies a portion of the 165 rRNA gene located in the 5' region of the gene (see Figure 82).
The CFLPT"' reactions shown in Figure 85, Panel B were performed using approximately 1 pmole of each PCR product and the cleavage reactions were incubated at 45°C. The cleavage products were resolved by electrophoresis on an 8%
polyacrylamide gel.
The lanes marked "M" in Figure 85, Panels A and B contain plasmid pUC 19 DNA
digested with MspI and 3' end labeled with fluorescein ddLJTP using terminal deoxynucleotidyl transferase as described in Example 8. Tllis marker includes bands corresponding to lengths of 26, 34, 67, 110/111, 147, 190, 242 and 331 bp.
Additional marker bands of 404, 489 and 501 by are not visible in this figure. In Panel A. lanes 1-5 contain the uncut (i.e., no enzyme) controls and lanes 6-10 contain the cleavage products generated by the incubation of substrates derived from E. coli Stain B (ATCC
11303), E coli 2~ Strain K-12 (ATCC 14948), E coli Serotype 0157: H7 (ATCC 43895), Shigella dysenteriae Serotype 2 (ATCC 2902'7), and Salmonella choleraesuis subsp. choleraesuis Serotype typhi (ATCC 6539), respectively. In Panel B, lanes 1-5 contain the uncut (i.e., no enzyme) controls '' and lanes 6-10 contain the cleavage products generated by the incubation of substrates derived from E. coli Stain K-12 (ATCC 14948), E. coli Strain B (ATCC 11303), E coli Serotype 0157: H7 (ATCC 43895), Shigella dysenteriae Serotype 2 (.ATCC 29027), and Salmonella choleraesuis subsp. choleraesuis Serotype typhi (ATCC 6539), respectively.
The lower molecular weight materials seen in the "uncut" lanes has been found to be due to degradation of the gel-purified material after storage for several days in dH20. This WO 96/15267 PCT/US9St14673 degradation may be due to environmental nucleases that are active when EDTA is not present in the storage solution (i. e., the necessary metal ions may be present in trace amounts). This degradation is effectively suppressed by inclusion of tRNA in the storage solution (see Example 19). The degradation seen in these uncut controls (Pane B, lanes 1-5) does not effect the CFLPTM results.
The results shown in Figure 85 demonstrate that some regions of the 16S rRNA
genes are more variable than others, and that analysis of these regions are particularly useful when comparing very closely related organisms. For example, substrates generated by the 1638/TET-1659 pair (which amplifies a portion of the 16S rRNA gene located in the 5' region of the gene) can be used to generate CFLPT"' patterns which distinguish not only between the DNA derived from the genera of Escherichia, Shigella, and Salmonella (Panel B, lanes 6-10), but which also creates distinct cleavage patterns from the DNA
derived from the three strains of E. coli tested (i.e., strains B, K-12 and 0157: H7) (Panel B
lanes 6-8).
In contrast, no substantial difference in CFLPT"'' patterns was observed between the strains of the Escherichia-Salmonella assemblage for DNA fragments produced using the TET-SB-4/1743 pair which generates an approximately 208 by fragment located near the 3' end of 16S rRNA genes (Panel A, lanes 6-10). This contrast in the degree of variation between the 5' and 3' regions of the 16S rRNA genes is consistent with the results reported by Widjojoatmondjo et al., supra, in which the comparisons between strains of the Escherichia-Salmonella assemblage were made by SSCP analysis.
Since each organism has multiple copies of the 16S rRNA gene, and these co-amplify in each PCR, it was important to show that the products of different amplifications from the same organism produced the same cleavage pattern. In Figure 86, the cleavage products generated by cleavage of an approximately 1292 by 16S rRNA substrate generated using the TET-ER10/1743 pair in two separate PCR reactions from Campylobacter jejuni subsp. .)ejtll~l (ATCC 33291 ) are shown in lanes 2 and 3. For comparison, the same region amplified from E. coli Strain K-12 (ATCC 14948) is shown in lane 1. The CFLP~ reactions were performed using approximately 60 fmole of each PCR product and the cleavage reactions were incubated at 50°C for 2 min. Reactions were stopped by the addition of 95%
formamide, 5 mM EDTA, 5% glycerol and 0.02% methyl violet. The cleavage products were =
resolved by electrophoresis on a 6% denaturing polyacrylamide gel as described above.
The results shown in Figure 86 demonstrate that very different CFLPT~'' patterns were generated using substrates from Gamma (Escherichia. lane 1 ) and Epsilon (Campvlobacter.

lanes 2 and 3) subdivisions of Purple bacteria, but that the same CFLPTM
pattern was observed between the products of separate PCR reactions on the same genomic DNA (lanes 2 and 3).
In Figure 87, the cleavage products generated by cleavage of an approximately 350 by 16S rRNA substrate generated using the 1638/TET-1659 pair and genomic DNA
derived from 3 5 E. coli Serotype 0157: H7 (ATCC 43895), S. choleraesuis subsp.
choleraesuis Serotype typhi ' (ATCC 6539), Shigella dysenteriae Serotype 2 (ATCC 29027), S. aureus subsp.
aureus (ATCC 33591 ), S. aureus subsp. aureus (ATCC 33592), ..f. aureus subsp. aureus (ATCC
13565), S. hominis (ATCC 29885), and S. warneri (ATCC'. 17917) are shown in lanes I-8, respectively. The CFLPT"' reactions were performed as described above, using approximately IO 200 fmol of each PCR product; the cleavage reactions were incubated at 65°C for 2 min. The cleavage products were resolved by electrophoresis on a 10% denaturing polyacrylamide gel as described above.
The results shown in Figure 87 demonstrate that very different CFLPTM patterns were produced using DNA derived from strains representing Purple bacteria (lanes I-3) and the I S Gram-positive phylum (lanes 4-8). A substantial difference between CFLPT"' patterns was detected between the genera Escherichia (lane 1), Salmonella (lane 2), and Shigella (lane 3).
Additionally, a substantial difference between the C:FLPT"' patterns was detected between species of Staphylococcus aureus (lanes 4-6), hominis (lane 7), and warneri (lane 8).
No substantial difference between CFLPTM patterns was observed between the three strains of 20 Staphylococcus aureus subsp. aureus ATCC 33591 (lane 4), ATCC 33592 (lane 5), and ATCC 13565 (lane 6). These S. aureus isolates differ in reported antibiotic resistance, but are so closely related that the rRNA genes do not yet show divergence by CFLPT"' analysis.
The above results demonstrate that CFLPTM analysis can be used to discriminate between bacterial genera as well as between different species and subspecies (depending on 25 the region of the 16S rRNA gene used as the substrate). A comparison of the CFLPTM
patterns generated within the same or similar genera (e.g., S'almonella, Shigella and E. coli) shows an overall similarity in the banding pattern with differences revealed as changes in a '' small subset of the bands. When the comparison is made across different genera (e.g., between E coli and S. aureus) a more striking change in barcode pattern is evident indicatin~~
30 that CFLPT"' patterns may not only be used to detect differences between organisms. but the degree to which the patterns change may be used to assess the degree of evolutionary divergence between organisms.

WO 96/15267 PC"T/US95/14673 Substrates for CFLPT"'' analysis were produced by PCR amplification using different sets of primers. Some primer pairs (sets) are reported to be universal for all procaryotic organisms; other primer pairs have been observed to be specific for representatives of lower , taxons (See, PCT Publication WO 90/15157). Except for the primer sequences, no knowledge of the DNA sequence of the rRNA gene from any specific organisms) is required d for amplification and CFLP~ analysis of bacterial 16S rRNA genes.
Distinct CFLP'~"'' patterns were observed between representatives of archeaea and eubacteria, different phyla of eubacteria, different phyla within eubacteria, different subdivisions of the same phylum, different genera of the same assemblage, different species of the same genus and different strains of the same species. Distinct signatures in CFLP~'~"'' patterns were found that allowed discrimination of pathogenic isolates, including those associated with food poisoning, from innocuous members of the normal flora.
While the PCR products generated using genomic DNA from different organisms with the same set of primers are indistinguishable by their mobility during gel electrophoresis (on non-gradient polyacrylamide gels), the CleavaseT"'' BN enzyme cleaves these PCR products into shorter fragments thereby generating a characteristic set of cleavage products (i.e., a distinct CFLP~ signature). The pattern of cleavage products generated is reproducible; DNA
substrates generated in independent PCRs from the same organism using a given primer pair yield the same pattern of cleavage products.
CFLP"'' patterns can be generated using large DNA fragments (e.g., at least about 1.6 kb) and thus could cover the entire length of the bacterial 16S rRNA gene.
CFLP~'~"'' can also be used in conjunction with shorter DNA fragments (about 200 bp) which are located at different positions throughout the 16S rRNA gene.

CFLP'~"'' Analysis of Substrates Containing Nucleotide Analogs The effect of using various nucleotide analogs to generate substrates for CFLPT"' reactions was examined. As discussed below, nucleotide analogs are used in PCRs for several reasons; therefore, the ability to analyze the modified products of PCRs (i.e., nucleotide analog-containing PCR products) by CFLP~ analysis was investigated. The 7-deaza purine analogs (7-deaza-dATP and 7-deaza-dGTP) serve to destabilize regions of secondary structure by weakening the intrastrand stacking of multiple adjacent purines. This effect can allow _ '73'7 _ amplification of nucleic acids that, with the use of natural dNTPs, are resistant to amplification because of strong secondary structure [McConlogue et al..
Nucleic Acids Res.
16:20 (1988)].
' Similarly, the analog dUTP is often used to replace dTTP, but for different reasons.
z 5 dUTP-containing DNA (this nomenclature is shorthand for PCR products generated using ' dUTP; the actual PCR product will contain dUMP) can be destroyed by the enzymatic activity of uracil DNA glycosylase (UDG) while dTTP-containing DNA is untouched. When PCR
products are produced containing dUMP in place of dTMP, UDG can be used in all subsequent reactions to eliminate false positive results due to carry-over from the earlier PCRs, without preventing amplification from the normal I)NA of interest. This method is widely used in clinical laboratories for performing PCR and thus this method would be used by most clinical laboratories using PCR in conjunction with CFLPT"' for pathogen typing.
Thus, the ability of the CFLP"~' reaction to suitably cleave dUTP-containing DNA fragments (i.e., produce strong reproducible band patterns) was examined.
For these comparisons, substrates corresponding to a 157 by fragment derived from exon of of the wild-type and R422Q mutant of the human tryosinase gene were generated by PCR amplification using either 1) the standard mixture of dNTPs (i.e., dATP, dCTP, dGTP
and dTTP); 2) dUTP in place of dTTP; 3) 7-deaza-dGTP (d'GTP) in place of dGTP;
and 4) 7-deaza-dATP (d'ATP) in place of dATP. These substrates were then incubated with Cleavase~'~"'' BN enzyme and the effect the presence of the various nucleotide analogs on the cleavage pattern was examined.
B) Preparation of Substrates Containing Nucleotide Analogs A 157 by fragment of the human tyrosinase gene (exon 4) was amplified in PCRs using the following pair: 5' CACCGTCCTCTTCAAGAAG 3' (SEQ ID N0:29) and 5"
biotin-CTGAATCTTGTAGATAGCTA 3' (SEQ ID N0:30). Plasmids containing cDNA
derived from the wild-type or R422Q mutant of the tyrosinase gene were used as template (see Example 8 for a description of these plasmids). The resulting double-stranded PCR
products contain the 5' biotin label on the anti-sense strand such that sequence detected in the CFLPT~'' reaction is SEQ ID N0:35 (wild-type anti-sense strand) or SEQ ID
N0:53 (R422Q
mutant anti-sense strand). All PCRs were conducted in a final volume of 100 q.l. dATP.
h dCTP, dGTP, dTTP and dUTP were obtained from Perkin Elmer; d'ATP and d'GTP
were obtained from Pharmacia. Taq DNA polymerase was obtained from Promega. The PCR
' mixtures were assembled as shown below in Table 7.
- 7jj WO 96/15267 PC'~'/US95114673 Reaction Components [Stock]Aliquot [Final]

Plasmid cDNA 4 ng/~.l1 ~.l 40 pg PCR Buffer' 1OX 10 ~l 1X , Unlabelled primer 100 0.25 ~,l 25 pmole pM

Labeled primer 100 0.25 ~.1 25 pmole ~,M

dATP 10 mM 1 ~.l 100 ~.m dCTP 10 -mM 1 ~.l I 00 ~.m dGTP 10 mM 1 ~.l 100 ~.m dTTP 10 mM 1 g.l 100 ~,m d'ATP'- 5 mM 2 ~l 100 ~,m d'GTP 3 5 mM 2 ~I 100 ~m dUTP 4 20 mM 4 ~.l 800 ~m Taq polymerise 5 units0.5 ~.l 2.5 units /~l dH20 to 100 ~,l 1X concentration contains 20 mM Tris-HCI, pH 8.5; 1.5 mM MgCh; 50 mM KCI; 0.5%
Tween 20; and 0.5% NP-40.
'- d'ATP completely substituted for dATP in the PCR.
3 d'GTP completely substituted for dGTP in the PCR.
4 dUTP completely substituted for dTTP in the PCR. Other nucleotides were present at a final concentration of 200 Vim. In this reaction, the PCR buffer used was the lOX
buffer (500 mM
KCI, 100 mM Tris-Cl, pH 9.0, 1.0% Triton X-100) provided by Promega. 25 mM
MgCh was added separately to a final concentration of 2.5 mM.
Wild-type and the mutant R422Q substrates were amplified using the natural and substituted nucleotide analogs listed above. For reactions containing the natural dNTPs, d'ATP and d'GTP, all reaction components were added together. Reactions containing dUTP
were initially assembled without the polymerise (see below).
The assembled reactions were placed in a thermocycler (MJ Research, Watertown, MA) that was preheated to 95°C. The tubes were allowed to incubate for one minute at 95°C

before amplification. The program was then set at 94°C for 30 minutes.
50°C for one minute.
72°C degrees for two minutes for 34 cycles with a final 72°C
incubation for 5 minutes.
Reactions containing dUTP were performed with a "hot start." All components except the polymerase were mixed, heated to 95°C for 1 minute, then cooled to 72°C. Taq z 5 polymerase (2.5 units) was then added in 10 pl of 1X PCR buffer for a final volume of 100 " l.~.l.
At the end of the amplification, the PCR products were made 0.3M Na04Ac, with the exception of reactions containing dUTP; the dUTP-containing reactions were brought to 2M
NH40Ac; all were then precipitated by the addition of 2.5 volumes (total aqueous volumes) of absolute ethanol. The DNA pellets were collected by centrifugation and then dried under vacuum. The pellets were resuspended in 10 p,l of TE and 10 pl of STOP buffer (20 p.l TIOE0.1 and 16 p.l of STOP for the dUTP-containing reactions). The tubes were then heated to 85°C for 2 minutes and the mixtures were resolved by electrophoresis through 10% (6%
for dUTP) denaturing acrylamide gel (19:1 cross link) with 7M urea in a buffer of O.SX TBE.
I S The PCR products corresponding to the 157 by substrate derived from the wild-type and R422Q~ mutant were gel purified as described in Example 19. The gel-purified DNAs were resuspended in T10E0.1 buffer using the following volumes: 40 pl for fragments containing only dNTPs; 40 p.l for fragments containing d'ATP; 25 p,l for fragments containing d'GTP and 25 p,l for fragments containing dUTP.
B) Cleavage Reaction Conditions The gel purified 157 by tyrosinase substrates containing natural deoxynucleotides and nucleotide analogs were analyzed in cleavage reactions as i:ollows. Final reaction mixtures comprised 1 ~1 of the resuspended gel-purified DNA [see section (a) above] and 25 ng CleavaseT"'' BN in 10 mM MOPS, pH 7.5 with 0.2 mM MllCl,, and 0.05% each Tween and NP-40 in a volume of 20 p.l. No enzyme controls were assembled in which distilled water replaced the CleavaseT"'' BN enzyme. The substrate DNAs were distributed into reaction tubes and brought to a volume of 15 p.l with H,O. The remaining reaction r components were mixed in a volume of 5 pl (i. e., at a 4X concentration).
The DNAs were heated for 15 sec. at 95°C to denature the DNA. The cleavage reactions were initiated by the addition of 5 p.l of the enzyme/buffer mixture (the 4X concentrate). The cleavage reactions were incubated at 45°C for three minutes, and the reactions were terminated by the addition of 16 ~1 of Stop solution (described in section a). Seven microliters of each sample was heated to 85°C for two minutes prior to loading onto a 10% denaturing acrylamide gel ( 19:1 _ 23j _ cross link), with 7M urea in a buffer of 45 mM Tris Borate pH 8.3, 1.4 mM
EDTA.. The gel was run at a constant 800 V until the bromophenol blue had migrated the length of the gel.
Following electrophoresis, the biotinylated fragments were detected as described in Example 8 with the exception that 4 p,l of the SAAP conjugate was added to 100 p.l of USB
s blocking buffer (1:25,000 dilution). After washing, 5 p,ls of CDP-Star~'~'' was used as the chemiluminescent substrate. The resulting autoradiogram is shown in Figure 88.
In Figure 88, the lanes marked "M" contain biotinylated molecular weight markers -obtained from Amersham (Arlington Heights, IL) and include bands corresponding to lengths of 50, 100 and 200 nucleotides (size indicated by use of numbers and large arrowheads).
Lanes 1-8 contain reaction products obtained by incubation of the substrates in the absence of CleavaseT"' BN enzyme (i.e., no enzyme or uncut controls). Lanes 9-16 contain reaction products obtained by incubation of the substrates in the presence of CleavaseT"'' BN enzyme.
Lanes 1, 3, 5, 7, 9, 11, 13 and 15 contain the wild-type substrate; lanes 2, 4, G, 8, 10, I?, 14 and I 6 contain the R422Q mutant substrate. The products shown in lanes 1, 2.
9 and 10 were generated from substrates generated using dNTPs in the PCRs. The products shown in lanes 3, 4, 11 and 12 were generated from substrates generated using dUTP in place of dTTP in the PCRs. The products shown in lanes S, 6, 13 and 14 were generated from substrates generated using d'GTP in place of dGTP in the PCRs. The products shown in lanes 7, 8, I
S and 16 were generated from substrates generated using d'ATP in place of dATP in the PCRs. It can be seen from this example that modified DNA fragments are suitable for cleavage in CFLP
reactions. Though the banding pattern is substantially different with these substitutions, the wild-type and R422Q mutant DNAs are readily distinguishable in all cases.
While not limiting the invention to any particular theory, the changes in banding patterns observed when nucleotide analogs are utilized can be attributed to two sources. In all cases, but particularly in reference to the 7-deaza purines, the use of nucleotide analogs may substantially change the nature and stability of the intrastrand folded structures formed during the cleavage reaction. As a consequence, the locations of the cleavage sites would naturall~~
shift. In addition, the substitution of the modified nucleotides may change the affinity of the cleavage enzyme for the folded cleavage structure, either strengthening or weakening cleavage at a particular site. -Examination of the variations seen between the wild-type and R422Q mutant when _ different analogs are used also shows that the use of these substituants'can enhance the contrast between the variants. For example, with regard to the cleavage products of the two substrate DNAs (generated using dUTP or dTTP) in the region just above the 50 by marker:
one significant band that reduces in intensity between the wild-type and the mutant is more dramatically reduced in the dU-containing samples.
The results shown in Figure 88 demonstrate that nucleotide analogs may be used for the generation of CFLP~'~"'' substrates. The substrates derived from the wild-type or R422Q
' mutant of the tyrosinase gene which contain nucleotide analogs produce distinct cleavage patterns which allow the discrimination and identification of the mutant and wild-type alleles.
This example demonstrates that even with 100% substitution with either 7-deaza-GTP
for dGTP or 7-deaza-ATP for dATP, robust CFLP patterns are generated, although the precise sites of clevage are different in the dNTP-containing and i'-deaza-dNTP
containing substrates.
The above results also demonstrated that single base changes present within DNA fragments containing nucleotide analogs still influence the folded structure sufficiently to cause cleavage pattern changes similar to those seen when DNA fragments lacking nucleotide analogs are analyzed using the CFLPT"'' assay.
From the above it is clear that the invention provides reagents and methods to permit the rapid screening of nucleic acid sequences for variations. These methods allow the identification of viral and bacterial pathogens as well as permit the detection of mutations associated with gene sequences (e.g., mutations associated with multiple drug resistance in M.
tuberculosis or mutations associated with human disease). These methods provide improved means for the identification and characterization of pathogens.

SEQUENCE LISTING
(1) GENERAL
INFORMATION:

(i) APPLICANT: DAHLBERG, JAMES E.

LYAMICHEV, VICTOR I.
S

BROW, MARY ANN D.

OLDENBURG, MARY C.

HEISLER, LAURA M. , FORS, LANCE

OLIVE, DAVID M. ' IO (ii) TITLE OF INVENTION: RAPID DETECTION AND IDENTIFICATION OF

NUCLEIC ACID VARIANCE AND PATHOGENS

(iii) NUMBER OF SEQUENCES: 147 IS (iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: MEDLEN & CARROLL

(B) STREET: 220 MONTGOMERY STREET, SUITE 2200 (C) CITY: SAN FRANCISCO

(D) STATE: CALIFORNIA

ZO (E) COUNTRY: UNITED STATES OF AMERICA

(F) ZIP: 94104 (v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible ,~S (C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: PatentIn Release #1.0, Version #1.25 (vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US

(B) FILING DATE:

(C) CLASSIFICATION:

3O (vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/520,946 (B) FILING DATE: 30-AUG-1995 (vii) PRIOR APPLICATION DATA:

3S (A) APPLICATION NUMBER: US 08/484,956 (B) FILING DATE: 07-JUN-1995 (vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/402 , (B) FILING DATE: 09-MAR-1995 _ (vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 08/337,164 (B) FILING DATE: 09-NOV-1994 (vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 08/254,359 (B) FILING DATE: 06-JUN-1994 4S (vii) PRIOR APPLICATION DATA: , (A) APPLICATION NUMBER: US 08/073,384 (B) FILING DATE: 04-JUN-1993 (vii) PRIOR APPLICATION DATA:
SO (A) APPLICATION NUMBER: US 07/986,330 ;
(B) FILING DATE: 12-DEC-1992 (viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: CARROLL, PETER G.
(B) REGISTRATION NUMBER: 32,837 ., (C) REFERENCE/DOCKET NUMBER: FORS-02000 -(ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (415) 705-8410 (B) TELEFAX: (415) 397-8338 (2) INFORMATION FOR SEQ ID NO:1:

S (ij SEQUENCE CHARACTERISTICS:
(A) LENGTH: 2506 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear 10- ( i t )- ;v~LEGuLE TYPE : DNA ( genomi c ) (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

ATGAGGGGGA TGCTGCCCCT CTTTGAGCCC AAGGGCCGGG TCC7.'CCTGGT 60 IS CACCTGGCCT ACCGCACCTT CCACGCCCTG AAGGGCCTCA CCAC'CAGCCG 180 GCGGTGATCG TGGTCTTTGA CGCCAAGGCC CCCTCCTTCC GCCA.CGAGGC 420 GTCCTGGCCA GCCTGGCCAA GAAGGCGGAA AAGGAGGGCT ACGAGGTCCG 7gp 3O GCCGACAAAG ~CCTI'T~C'rCA GCTCCTTfiC~ ~ACCcicATCC ACGTCCTCCA900 CTGGACCGGC TGAAGCCCGC CATCCGGGAG AAGATCCTGG CCCACATGGA
CGATCTGAAG
CTCTCCTGGG ACCTGGCCAA GGTGCGCACC GACCTGCCCC TGGAG~GTGGA
CTTCGCCAAA
AGGCGGGAGC CCGACCGGGA GAGGCTTAGG GCCTTTCTGG AGAGGCTTGA
GTTTGGCAGC
CTCCTCCACG AGTTCGGCCT TCTGGAAAGC CCCAAGGCCC TGGAGGAGGC
CCCCTGGCCC
CCGCCGGAAG GGGCCTTCGT GGGCTTTGTG CTTTCCCGCA AGGAGCCCAT
GTGGGCCGAT
CTTCTGGCCC TGGCCGCCGC CAGGGGGGGC CGGGTCCACC GGGCCCCCGA
GCCTTATAAA
GCCCTCAGGG ACCTGAAGGA GGCGCGGGGG CTTCTCGCCA AAGACCTGAG
CGTTCTGGCC
CTGAGGGAAG GCCTTGGCCT CCCGCCCGGC GACGACCCCA TGCTCCTCGC
CTACCTCCTG
GACCCTTCCA ACACCACCCC CGAGGGGGTG GCCCGGCGCT ACGGCGGGGA
GTGGACGGAG
GAGGCGGGGG AGCGGGCCGC CCTTTCCGAG AGGCTCTTCG CCAACCTGTG
GGGGAGGCTT
GAGGGGGAGG AGAGGCTCCT TTGGCTTTAC CGGGAGGTGG AGAGGCCCCT
TTCCGCTGTC

CTGGAGGTGG CCGAGGAGAT CGCCCGCCTC GAGGCCGAGG TCTTCCGCCT
GGCCGGCCAC

GCCCTCCGCG AGGCCCACCC CATCGTGGAG AAGATCCTGC AGTACC'.GGGA GCTCACCAAG 1620 _ ~3g ACCCCAGGAC GGGCCGCCTC

CGATCCCAAC

CTCCAGAACA TCCCCGTCCGCACCCCGCTT GGGCAGAGGA 1800 ,, TCCGCCGGGC CTTCATCGCC

TAGAGCTCAG GGTGCTGGCC

S CACCTCTCCG GCGACGAGAACCTGATCCGG GTCTTCCAGGAGGGGCGGGA 1920 ' CATCCACACG

(2) INFORMATION
FOR SEQ
ID N0:2:

(i) SEQUENCE
CHARACTERISTICS:

(A) LENGTH: 2496 base pairs (B) ~'~E:
nucleic acid 20 (C) STRANDEDNESS:
double (D) TOPOLOGY:
linear (ii) MOLECULE
TYPE: DNA
(genomic) (xi) SEQUENCE
DESCRIPTION:
SEQ ID
N0:2:

GACCGCGACC TCTACCAGCTCCTTTCGGAG CGCATCGCCATCCTCCACCCTGAGGGGTAC480 '.

~J GACCAGGTGA AGCCCTCCTTGCGGGAGAAG CTCCAGGCGGGCATGGAGGCCCTGGCCCTT720 CTCCACGAGT TCGGCCTCCT GGAGGGGCCG AAGGCGGCAG AGGAGGCCCC
'" CTGGCCCCCT 900 CCGGAAGGGG CTTTTTTGGG CTTTTCCTTT TCCCGTCCCG AGCCCATGTG

CTGGCCCTGG CTGGGGCGTG GGAGGGGCGC CTCCATCGGG CACAAGACCC

CTGAGGGACC TTAAGGGGGT GCGGGGAATC CTGGCCAAGG ACCTGGCGGT
lO TTTGGCCCTG 1080 CGGGAGGGCC TGGACCTCTT CCCAGAGGAC GACCCCATGC TCCTGGCCTA

CCCTCCAACA CCACCCCTGA GGGGGTGGCC CGGCGTTACG GGGGGGAGTG

GCGGGGGAGA GGGCCCTCCT GGCCGAGCGC CTCTTCCAGA CCC7.'AAAGGA

GGAGAAGAAC GCCTGCTTTG GCTTTACGAG GAGGTGGAGA AGCC'GCTTTC

GAGGTGGAGG CGGAGGTGCG CCAGCTGGAG GAGGAGGTCT TCCGCCTGGC

TTCAACCTCA ACTCCCGCGA CCAGCTGGAG CGGGTGCTCT TTGACGAGCT

GCCATCGGCA AGACGGAGAA GACGGGGAAA CGCTCCACCA GCGCTGCCGT

CTGCGAGAGG CCCACCCCAT CGTGGACCGC ATCCTGCAGT ACCGGGAGCT

AAGAACACCT ACATAGACCC CCTGCCCGCC CTGGTCCACC CCAAGACCGG

ACCCGCTTCA ACCAGACGGC CACCGCCACG GGCAGGCTTT CCAGCTCCGA

CAGAACATCC CCGTGCGCAC CCCTCTGGGC CAGCGCATCC GCCGAGCCTT

GAGGGCTGGG TGCTGGTGGT CTTGGACTAC AGCCAGATTG AGCT7.'CGGGT

CTCTCCGGGG ACGAGAACCT GATCCGGGTC TTTCAGGAGG GGAGGGACAT

ACCGCCAGCT GGATGTTCGG CGTTTCCCCC GAAGGGGTAG ACCCTCTGAT

GCCAAGACCA TCAACTTCGG GGTGCTCTAC GGCATGTCCG CCCACCGCCT

CTTTCCATCC CCTACGAGGA GGCGGTGGCC TTCATTGAGC GCTACTTCCA

AAGGTGCGGG CCTGGATTGA GGGGACCCTC GAGGAGGGCC GCCGGCGGGG

ACCCTCTTCG GCCGCCGGCG CTATGTGCCC GACCTCAACG CCCGGGTGAA

GAGGCGGCGG AGCGCATGGC CTTCAACATG CCGGTCCAGG GCACCGCCGC

AAGCTGGCCA TGGTGCGGCT TTTCCCCCGG CTTCAGGAAC TGGGGGCGAG

CAGGTGCACG ACGAGCTGGT CCTCGAGGCC CCCAAGGACC GGGCGGAGAG

TTGGCCAAGG AGGTCATGGA GGGGGTCTGG CCCCTGCAGG TGCCCCTGGA

GGCCTGGGGG AGGACTGGCT CTCCGCCAAG GAGTAG

(2) INFORMATION FOR SEQ ID N0:3:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 2504 base pairs - (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear ~S (ii) MOLECULE TYPE: DNA (genomic) (xi) S EQUENCE CRIPTION:EQ ID
DES S N0:3:

AAAGGCCGGG
TCCTCCTGGT

CCACGAGCCG r GCTTCCCCAA GGTGCGGGCC TGGATAGAAA AGACCCTGGA GGA.GGGGAGGAAGCGGGGCT 2160 '" GCGTCAGGGA GGCCGCGGAG CGCATGGCCT TCAACATGCCCGTCCAGGGCACCGCCGCCG 2280 TGGCGGCTTT GGCCAAGGAG GCCATGGAGA AGGCCTATCC CCTCGCCGTGCCCCTGGAGG 2460 ' (2) INFORMATION FOR SEQ ID N0:4:

IO (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 832 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear IS (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID N0:4:

Met Arg Gly Met Leu Pro Leu Phe Glu Pro Ly~~ Gly 1 5 10 Arg Val Leu Leu Val Asp Gly His His Leu Ala Tyr Arg Thr Phe His ZO 20 25 Ala Leu Lys Gly Leu Thr Thr Ser Arg Gly Glu Pro Val Gln Ala. Val 35 40 Tyr Gly Phe Ala Lys Ser Leu Leu Lys Ala Leu Lys Glu Asp Gly Asp 50 55 Ala Val Ile Val ZS Val Phe Asp Ala Lys Ala Pro Ser Phe Arg His Glu Tyr Gly 65 70 Ala Gly Tyr Lys Ala Gly Arg Ala Pro Thr Pro Glu Asp Phe Arg Gln 85 90 Pro Leu Ala Leu Ile Lys Glu Leu Val Asp Leu Leu Gly Leu Arg Leu 100 105 Ala Glu Val Pro Gly Tyr Glu Ala Asp Asp Val Leu Ala Ser Ala Lys 115 120 Leu Lys Ala Glu Lys Glu Gly Tyr Glu Val Arg Ile Leu Thr Asp Lys 130 135 Ala Asp 3S Leu Tyr Gln Leu Leu Ser Asp Arg Ile His Val Leu Pro Glu 145 150 His Gly _ 155 Tyr Leu Ile Thr Pro Ala Trp Leu Trp Glu Lys Tyr Leu Arg 165 170 Gly Pro Asp Gln Trp Ala Asp Tyr Arg Ala Leu Thr Gly Asp Ser Asp 4'O 180 185 Glu Asn Leu Pro Gly Val Lys Gly Ile Gly Glu Lys Thr Ala Lys Leu 195 200 Arg Leu Glu Glu Trp Gly Ser Leu Glu Ala Leu Leu Lys Asn Asp Arg 210 215 Leu Leu Lys Pro Ala Ile Arg Glu Lys Ile Leu Ala His Met Asp Asp Leu Lys Leu Ser Trp Asp Leu Ala Lys Val Arg Thr Asp Leu Pro Leu Glu Val Asp Phe Ala Lys Arg Arg Glu Pro Asp Arg Glu Arg Leu Arg Ala Phe Leu Glu Arg Leu Glu Phe Gly Ser Leu Leu His Glu Phe Gly Leu Leu Glu Ser Pro Lys Ala Leu Glu Glu Ala Pro Trp Pro Pro Pro Glu Gly Ala Phe Val Gly Phe Val Leu Ser Arg Lys Glu Pro Met Trp Ala Asp Leu Leu Ala Leu Ala Ala Ala Arg Gly Gly Arg Val His Arg Ala Pro 15 Glu Pro Tyr Lys Ala Leu Arg Asp Leu Lys Glu Ala Arg Gly Leu Leu Ala Lys Asp Leu Ser Val Leu Ala Leu Arg Glu Gly Leu Gly Leu Pro Pro Gly AspAspProMet LeuLeu AlaTyrLeu Leu Asn 2~ 370 375 Asp Pro Ser Thr Thr ProGluGlyVal AlaArg ArgTyrGly GlyGluTrp Glu 385 390 395 Thr Glu Ala GlyGluArgAla AlaLeu SerGluArg LeuPheAlaAsn Leu 2$ Trp Gly ArgLeuGluGly GluGlu ArgLeuLeu TrpLeuTyrArg Glu Val Glu ArgProLeuSer AlaVal LeuAlaHis MetGluAlaThr Gly Val Arg LeuAspValAla TyrLeu ArgAlaLeu SerLeuGluVal Ala Glu Glu IleAlaArgLeu GluAla GluValPhe ArgLeuAlaGly His Pro Phe AsnLeuAsnSer ArgAsp GlnLeuGlu ArgValLeuPhe Asp Glu Leu GlyLeuProAla IleGly LysThrGlu LysThrGlyLys Arg Ser Thr SerAlaAlaVal LeuGlu AlaLeuArg GluAlaHisPro Ile Val Glu LysIleLeuGln TyrArg GluLeuThr LysLeuLysSer Thr Tyr Ile AspProLeuPro AspLeu IleHisPro ArgThrGlyArg Leu His Thr ArgPheAsnGln ThrAla ThrAlaThr GlyArgLeuSer Ser 4S Ser Asp Pro LeuGln Ile ProVal ThrProLeuGly Asn Asn 585Arg 590Gln Arg Ile Arg Glu Gl Arg Tr Ala L
Phe Ile Ala Glu y 595 600 p eu Leu Val Ala Leu Asp Tyr Ser Gln Ile Glu Leu Arg Val Leu Ala Hi L

s 610 615 eu Ser Gly .. ~ Asp Glu Asn Phe Leu Gln Ile Glu Arg Gly Val Ar As Il g 625 630 p e His Thr ' Glu Thr AlaSerTrpMet Gly Pro Ar Phe Val Gl Al g s 645 u a Val Asp Pro 1~ Leu Me't ArgArgAlaAla Thr Asn Phe: Gl Lys Ile Val L

y 660 eu Tyr Gly Met Ser AlaHisArgLeu Gln Leu Ala Ser Glu Il .
675 e Pro Tyr Glu Glu Ala 69n AlaPheIleGlu Tyr Gln Ser Ph Arg Phe P

0 e 695 ro Lys Val Arg 15 Ala Trp IleGluLysThr Glu Gly Ar Leu Glu Ar A

g 705 710 g rg Gly Tyr Val Glu Thr LeuPheGlyArg Arg Val Pro As Arg Tyr Leu Gl p 725 u Ala Arg Val Lys SerValArgGlu Ala Arg Met Ala Ph Ala Glu e Asn Met Pro Val Gln GlyThrA1aAla Leu Lys Leu Al Asp Met M

a 755 et Val Lys Leu Phe Pro Arg LeuGluGlu Gly Arg Met Le Met Ala L

u 770 775 eu Gln Val His 25 Asp Glu Leu ValLeuGlu Pro Glu Ar Ala Lys Ala Gl g 785 790 u Ala Val Ala Arg Leu Ala LysGluVal Val Tyr Pro Leu Al Met Glu Gly a Val Pro Leu Glu Val Glu Gly Asp Trp Leu Ser Al Val Ile L
Gly Glu a - 820 825 ys Glu (2) INFORMATION FOR SEQ ID N0:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 831 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID N0:5:
Met Ala Met Leu Pro Leu Phe Glu Pro Lys Gly Arg Val Leu Leu Val Asp Gly His His Leu Ala Tyr Arg Thr Phe Phe Ala Leu Lys Gly Leu Thr Thr Ser Arg Gly Glu Pro Val Gln Ala Val Tyr Gly Phe Ala Lys 4S Ser Leu Leu Lys Ala Leu Lys Glu Asp Gly Asp Val Val Val Val Val - 24~ -Phe Asp Ala Lys Ala Pro Ser Phe Arg His Glu Ala Tyr Glu Ala Tyr Lys Ala Gly Arg Ala Pro Thr Pro Glu Asp Phe Pro Arg Gln Leu Ala $ Leu Ile Lys Glu Leu Val Asp Leu Leu Gly Leu ValArg Leu Glu Val Pro Gly Phe G1-a Ala Asp Asp Val Leu Ala Thr Leu Ala Lys Arg Ala Glu Lys Glu Gly Tyr Glu Val Arg Ile Leu Thr Ala Asp Arg Asp Leu Tyr Gln Leu Leu Ser Glu Arg Ile Ala Ile Leu His Pro Glu Gly Tyr Leu Ile Thr Pro Ala Trp Leu Tyr Glu Lys Tyr Gly Leu Arg Pro Glu l$ Gln Trp Val Asp Tyr Arg Ala Leu Ala Gly Asp Pro Ser Asp Asn Ile Pro Gly Val Lys Gly Ile Gly Glu Lys Thr Ala Gln Arg Leu Ile Arg Glu TrpGly SerLeu GluAsnLeu PheGlnHisLeu AspGlnVal L

ys Pro SerLeu ArgGlu LysLeuGln AlaGlyMetGlu AlaLeuAla Leu Ser ArgLys LeuSer GlnValHis ThrAspLeuPro LeuGluVal Asp 2$ Phe GlyArg ArgArg ThrProAsn LeuGluGlyLeu ArgAlaPhe Leu Glu ArgLeu GluPhe GlySerLeu LeuHisGluPhe GlyLeuLeu Glu Gly ProLys AlaAla GluGluAla ProTrpProPro ProGluGly Ala Phe Leu Gly Phe Ser Phe Ser Arg Pro Glu Pro Met Trp Ala Glu Leu Le~u Ala Leu Ala Gly Ala Trp Glu Gly Arg Leu His Arg Ala Gln Asp 3$ Pro Leu Arg Gly-Leu Arg Asp Leu Lys Gly Val Arg Gly Ile Leu Ala Lys Asp Leu Ala Val Leu Ala Leu Arg Glu Gly Leu Asp Leu Phe Pro Glu Asp Asp Pro Met Leu Leu Ala Tyr Leu Leu AspJPro~Ser Asn Thr Thr Pro Glu Gly Val Ala Arg Arg Tyr Gly Gly Glu Trp Thr Glu Asp Ala Gly Glu Arg Ala Leu Leu Ala Glu Arg Leu Phe Gln Thr Leu Lys 4$ Glu Arg Leu Lys Gly Glu Glu Arg Leu Leu Trp Leu Tyr Glu Glu Val WO 96/15267 _PCTlUS95/14673 Glu Lys Pro Leu Ser Arg Val Leu Ala Arg Met Glu ~~'~

Arg Leu Asp Val Ala Tyr Leu Gln Ala Leu Ser Leu Glu Val Glu Ala '_ S Glu Val Arg Gln Leu Glu Glu Glu Val Phe Arg Leu Ala Gly His Pro 465 470 47-'°.

Phe Asn Leu Asn Ser Arg Asp Gln Leu Glu Arg Val Leu Phe Asp Glu Leu Gly Leu Pro Ala Ile Gly Lys Thr Glu Lys Thr Gly Lys Arg Ser Thr Ser Ala Ala Val Leu Glu Ala Leu Arg Glu Ala His Pro Ile Val Asp Arg Ile Leu Gln Tyr Arg Glu Leu Thr Lys Leu Lys Asn Thr Tyr 15 Ile Asp Pro Leu Pro Ala Leu Val His Pro Lys Thr Gly Arg Leu His Thr Arg Phe Asn Gln Thr Ala Thr Ala Thr Gly Arg Leu Ser Ser Ser Asp ProAsn LeuGln AsnIlePro ValArgThr ProLeuGly GlnArg Ile ArgArg AlaPhe ValAlaGlu GluGlyTrp ValLeuVal ValLeu Asp TyrSer GlnIle GluLeuArg ValLeuAla HisLeuSer GlyAsp 25 Glu AsnLeu IleArg ValPheGln GluGlyArg AspIleHis ThrGln Thr AlaSer TrpMet PheGlyVal SerProGlu GlyValAsp ProLeu Met ArgArg AlaAla LysThrIle AsnPheGly ValLeuTyr GlyMet Ser AlaHis ArgLeu SerGlyGlu LeuSerIle ProTyrGlu GluAla Val AlaPhe IleGlu ArgTyrPhe GlnSerTyr ProLysVal ArgAla 35 Trp IleGlu GlyThr LeuGluGlu GlyArgArg ArgGlyTyr ValGlu Thr Leu Phe Gly Arg Arg Arg Tyr Val Pro Asp Leu Asn Ala Arg Val Lys Ser Val Arg Glu Ala Ala Glu Arg Met Ala Phe Asn Met Pro Val Gln Gly Thr Ala Ala Asp Leu Met Lys Leu Ala Met Val Arg Leu Phe Pro Arg Leu Gln Glu Leu Gly Ala Arg Met Leu Leu Gln Val His Asp Glu Leu Val Leu Glu Ala Pro Lys Asp Arg Ala Glu Arg Val Ala Ala DEMANDES OU BREVETS VOLUMlNEUX

COMPREND PLUS D'UN TOME_ CEC1 EST LE TOME ~ DE
NOTE: Pour les tomes additionels, veuiilez contacter le Bureau canadien des brevets 3 ~ ~.'7 THiS SECTION OF THE APPLICAT10N/PATENT CONTAINS MORE
'THAN ONE VOLUME ~ , THIS 1S VOLUME y~_ OF
' NOTE: For additional volumes-please contact the Canadian Patent Office .

Claims

CLAIMS:

1. A method for treating nucleic acid, comprising:
(a) providing:
i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid substrate;
(b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure;
and (c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced.

2. The method of claim 1, wherein said nuclease is selected from the group consisting of Cleavase TM BN enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

3. The method of claim 1 or 2, wherein said nucleic acid substrate comprises a nucleotide analog.

4. The method of claim 3, wherein said nucleotide analog is selected from the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.

5. The method of claim 1 or 2, wherein said nucleic acid substrate of step (a) is substantially single-stranded.

6. The method of claim 1 or 2, wherein said nucleic acid substrate is RNA.

7. The method of claim 1 or 2, wherein said nucleic acid substrate is DNA.

8. The method of claim 1 or 2, wherein said nucleic acid substrate of step (a) is double-stranded.

9. The method of claim 8, wherein said treating of step (b) comprises:
i) rendering said double-stranded nucleic acid substantially single-stranded; and ii) exposing said single-stranded nucleic acid to conditions such that said single-stranded nucleic acid has secondary structure.

10. The method of claim 9, wherein said double-stranded nucleic acid is rendered substantially single-stranded by increased temperature.

11. The method of claim 1 or 2, wherein said nucleic acid substrate comprises an oligonucleotide containing human p53 gene sequence.

12. The method of any one of claims 1 to 11 further comprising the step of detecting said at least one cleavage product.

13. A method for treating nucleic acid, comprising:
(a) providing:
i) an enzymatic cleavage means comprising a nuclease, in a solution comprising manganese; and ii) a nucleic acid substrate;
(b) treating said nucleic acid substrate with increased temperature;
(c) reducing said temperature under conditions such that said substrate forms at least one cleavage structure;
(d) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (e) detecting said at least one cleavage product.

14. The method of claim 13, wherein said nuclease is selected from the group consisting of Cleavase TM BN enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

15. The method of claim 13 or 14, wherein said nucleic acid substrate comprises a nucleotide analog.

16. The method of claim 15, wherein said nucleotide analog is selected from the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.

17. The method of claim 13 or 14, wherein said nucleic acid substrate is RNA.

18. The method of claim 13 or 14, wherein said nucleic acid substrate is DNA.

19. The method of claim 13 or 14, wherein said nucleic acid substrate of step (a) is double-stranded.

20. The method of claim 13 or 14, wherein said nucleic acid substrate of step (a) is single-stranded.

21. The method of claim 13 or 14, wherein said nucleic acid substrate comprises an oligonucleotide containing human p53 gene sequence.

22. The method of claim 13 or 14, wherein said nucleic acid substrate comprises an oligonucleotide containing microbial gene sequences.

23. A method for detecting mutation in the human p53 gene, comprising:

(a) providing:
i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid substrate containing human p53 gene sequences;
(b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure;
(c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (d) comparing said cleavage product to the cleavage products produced by cleavage of a reference p53 gene sequence.

24. The method of claim 23, wherein said cleavage products produced by cleavage of a reference p53 gene sequence are generated by the cleavage of a nucleic acid substrate containing the human p53 gene sequences selected from the group consisting of SEQ ID NOS:79-81, 84-89 and 94-97.

25. A method for identifying strains of microorganisms comprising:
(a) providing:
i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid substrate containing sequences derived from at least one microorganism;
(b) treating said nucleic acid substrate under conditions such that said substrate forms at least one cleavage structure;
(c) reacting said cleavage means with said cleavage structure so that at least one cleavage product is produced; and (d) comparing said cleavage product to the cleavage products produced by cleavage of a reference sequence derived from a microorganism.

26. The method of claim 25, wherein said nuclease is selected from the group consisting of Cleavase TM BN enzyme, Thermos aquaticus DNA polymerase, Thermos thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

27. The method of claim 26, wherein said nucleic acid substrate comprises a nucleotide analog.

28. The method of claim 27, wherein said nucleotide analog is selected from the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.

29. The method of claim 23 or 24, wherein said nucleic acid substrate of step (a) is substantially single-stranded.

30. The method of claim 23 or 24, wherein said nucleic acid substrate is RNA.

31. The method of claim 23 or 24, wherein said nucleic acid substrate is DNA.

32. The method of claim 23 or 24, wherein said nucleic acid substrate of step (a) is double-stranded.

33. The method of claim 32, wherein said treating of step (b) comprises:
i) rendering said double-stranded nucleic acid substantially single-stranded; and ii) exposing said single-stranded nucleic acid to conditions such that said single-stranded nucleic acid has secondary structure.

34. The method of claim 33, wherein said double-stranded nucleic acid is rendered substantially single-stranded by increased temperature.

35. The method of any one of claims 25 to 34, wherein said microorganism comprises bacteria.

36. The method of claim 35, wherein said bacteria are selected from the group comprising members of the genera Campylobacter, Escherichia, Mycobacterium, Salmonella, Shigella and Staphylococcus.

37. The method of claim 36, wherein said members of the genus Mycobacterium comprise strains of multi-drug resistant Mycobacterium tuberculosis.

38. The method of any one of claims 25 to 34, wherein said microorganism comprises virus.

39. The method of claim 38, wherein said virus is selected from the group comprising hepatitis C virus and simian immunodeficiency virus.

40. The method of any one of claims 23 to 39, further comprising the step of detecting said at least one cleavage product.

41. A method comprising:
(a) extracting nucleic acid from a sample suspected of containing at least one microorganism; and (b) contacting said extracted nucleic acid with an enzymatic cleavage means comprising a nuclease, under conditions such that said extracted nucleic acid forms one or more secondary structures, and said cleavage means cleaves said secondary structures to produce at least one cleavage product.

42. The method of claim 41, further comprising the step of isolating said cleavage product.

43. The method of claim 41, further comprising the step of detecting said cleavage product.

44. The method of claim 43, further comprising comparing said detected cleavage product generated from cleavage of said extracted nucleic acid isolated from said sample with separated cleavage products generated by cleavage of nucleic acids derived from at least one reference microorganism.

45. The method of any one of claims 41 to 44, further comprising the step of isolating a polymorphic locus from said extracted nucleic acid after the extraction of step a), to generate a nucleic acid substrate comprising sequences derived from said polymorphic locus wherein said substrate is contacted with the cleavage means of step b).

46. The method of claim 45, wherein said polymorphic locus is isolated by nucleic acid amplification.

47. The method of claim 46, wherein said amplification is conducted in the presence of a nucleotide analog.

48. The method of claim 47, wherein said nucleotide analog is selected from the group comprising 7-deaza-dATP, 7-deaza-dGTP
and dUTP.

49. The method of claim 46, wherein said amplification employs oligonucleotide primers which match consensus gene sequences derived from said polymorphic locus.

50. The method of claim 46, wherein said amplification employs oligonucleotide primers which are complementary to consensus gene sequences derived from said polymorphic locus.

51. The method of claim 45, wherein said polymorphic locus comprises a ribosomal RNA gene.

52. The method of claim 51, wherein said ribosomal RNA
gene is a 16S ribosomal RNA gene.

53. The method of any one of claims 41 to 52, wherein said nuclease is selected from the group consisting of Cleavase TM BN
enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

54. The method of claim 41, wherein said nucleic acid of step (a) is substantially single-stranded.

55. The method of claim 41, wherein said nucleic acid is RNA.

56. The method of claim 41, wherein said nucleic acid is DNA.

57. The method of claim 41, wherein said nucleic acid of step (a) is double-stranded.

58. The method of claim 57, wherein said contacting of step (b) comprises i) rendering said double-stranded nucleic acid substantially single-stranded; and ii) exposing said single-stranded nucleic acid to conditions such that said single-stranded nucleic acid has secondary structure.

59. The method of claim 58, wherein said double-stranded nucleic acid is rendered substantially single-stranded by increased temperature.

60. The method of any one of claims 41 to 59, wherein said microorganism comprises bacteria.

61. The method of claim 60, wherein said bacteria are selected from the group comprising members of the genera Campylobacter, Escherichia, Mycobacterium, Salmonella, Shigella and Staphylococcus.

62. The method of claim 61, wherein said members of the genus Mycobacterium comprise strains of multi-drug resistant Mycobacterium tuberculosis.

63. The method of any one of claims 41 to 59, wherein said microorganism comprises virus.

64. The method of claim 63, wherein said virus is selected from the group comprising hepatitis C virus and simian immunodeficiency virus.

65. A nucleic acid treatment kit, comprising:
(a) a nuclease capable of reacting with cleavage structures so as to generate cleavage products; and (b) a solution comprising manganese.

66. The kit of claim 65, further comprising reagents for detecting said cleavage products.

67. The kit of claim 65 or 66, wherein said nuclease is selected from the group consisting of the Cleavase TM BN enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

68. A method comprising:
(a) providing:

i) an enzymatic cleavage means comprising a nuclease; and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said cleavage means and said substrate under conditions such that said substrate forms at least one secondary structure and said cleavage means cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.

69. The method of claim 68, further comprising (d) comparing said separated cleavage products from said target substrate with a reference control.

70. The method of claim 68 or 69, wherein said nuclease is selected from the group consisting of Cleavase TM BN enzyme, Thermos aquaticus DNA polymerase, Thermos thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Radl/Rad10 complex.

71. The method of claim 68 or 69, wherein said cleavage means comprises a thermostable 5' nuclease.

72. The method of claim 71, wherein a portion of the amino acid sequence of said nuclease is homologous to a portion of the amino acid sequence of a thermostable DNA polymerase derived from a eubacterial thermophile.

73. The method of claim 72, wherein said thermophile is selected from the group consisting of Thermos aquaticus, Thermos flavus and Thermos thermophilus.

74. The method of claim 73, wherein said nuclease comprises an amino acid sequence selected from the group consisting of SEQ ID NOS:72-78.

75. A method comprising:
(a) providing:
i) an enzymatic cleavage means comprising a nuclease;
and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said cleavage means and said substrate at an elevated temperature and under conditions such that said substrate forms at least one secondary structure and said cleavage means cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.

76. The method of claim 75, further comprising (d) comparing said separated cleavage products from said target nucleic acid with a reference control.

77. The method of claim 75 or 76, wherein said nuclease is selected from the group consisting of Cleavase TM BN enzyme, Thermus aquaticus DNA polymerase, Thermus thermophilus DNA
polymerase, Escherichia coli Exo III, and Saccharomyces cerevisiae Rad1/Rad10 complex.

78. The method of claim 75 or 76, wherein said cleavage means comprises a thermostable 5' nuclease.

79. The method of claim 75, wherein said nucleic acid target comprises single-stranded DNA.

80. The method of claim 75, wherein said nucleic acid target comprises double-stranded DNA.

81. The method of claim 75, wherein said nucleic acid target comprises RNA.

82. A method comprising:
(a) providing:
i) a thermostable DNA polymerase altered in amino acid sequence such that it exhibits reduced DNA synthetic activity from that of the wild-type DNA polymerase but retains substantially the same 5' nuclease activity of the wild-type DNA
polymerase; and ii) a nucleic acid target substrate suspected of containing sequence variation relative to a reference control;
(b) mixing said polymerase and said substrate under conditions such that said substrate forms at least one secondary structure and said polymerase cleaves said secondary structure resulting in the generation of multiple cleavage products; and (c) separating said multiple cleavage products so as to detect said sequence variation.

83. The method of claim 82, further comprising (d) comparing said separated cleavage products from said target nucleic acid with a reference control.

84. The method of claim 82 or 83, wherein said thermostable DNA polymerase is selected from the group consisting of polymerases from the thermophiles Thermus aquaticus, Thermus flavus and Thermus thermophilus.

85. The method of claim 82, 83 or 84, wherein the alteration to said wild-type sequence of said thermostable polymerase comprises a deletion.

86. The method of claim 82 wherein said thermostable polymerase comprises an amino acid sequence selected from the group consisting of SEQ ID NOS:72-78.

87. The method of claim 82, wherein said nucleic acid target comprises single-stranded DNA.

88. The method of any one of claims 82 to 87, wherein said nucleic acid target contains a fluorescent label.

89. The method of claim 88 wherein said detection of step c) comprises detection of fluorescently labelled fragments.